[igt-dev] [PATCH i-g-t 00/23] Add DG2 OA test

[igt-dev] [PATCH i-g-t 00/23] Add DG2 OA test

[Umesh Nerlige Ramappa]

Add tests for changes in perf OA for DG2.

v2:
- Rebase
- Reposting since some patches did no land in the ML

Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>

Lionel Landwerlin (4):
  lib/i915/perf: implement report accumulation for new format
  lib/i915/perf: fixup conversion script for XEHPSDV
  lib/i915/perf: make warning message more helpful
  lib/i915/perf: Add DG2 metrics

Nerlige Ramappa, Umesh (1):
  i915/perf: Limit sseu-config tests for gen11

Umesh Nerlige Ramappa (18):
  i915/perf: Check regularly if we are done reading reports
  i915/perf: Fix OA short_reads test
  i915/perf: Check return value from getparam
  i915/perf: Bump timestamp tolerance for DG1
  i915/perf: Account for OA sampling interval in polling test
  i915/perf: Add support for 64-bit counters
  i915/perf: Define OA report types
  i915/perf: Use ARRAY_SIZE consistently for num_properties
  i915/perf: Use gt in perf tests and lib
  i915/perf: Explicitly state rendercopy needs for a test
  i915/perf: Skip tests that use rendercopy
  i915/perf: Add OA formats for DG2
  i915/perf: Add a test for non-power-of-2 oa reports
  i915/perf: Fix CS timestamp vs OA timstamp mismatch
  i915/perf: Treat ticks as 64 bit
  i915/perf: Treat timestamp as 64 bit value
  i915/perf: Fix DG2 A0 report header
  i915/perf: Wait longer for rc6 residency in DG2

 include/drm-uapi/i915_drm.h                   |   12 +
 lib/i915/perf-configs/guids.xml               |   24 +
 lib/i915/perf-configs/mdapi-xml-convert.py    |   70 +-
 lib/i915/perf-configs/oa-dg2.xml              | 9920 +++++++++++++++++
 .../perf-configs/perf-metricset-codegen.py    |   13 +
 lib/i915/perf-configs/update-guids.py         |    2 +-
 lib/i915/perf.c                               |  104 +-
 lib/meson.build                               |    2 +-
 tests/i915/perf.c                             |  623 +-
 9 files changed, 10589 insertions(+), 181 deletions(-)
 create mode 100644 lib/i915/perf-configs/oa-dg2.xml

-- 
2.25.1

[igt-dev] [PATCH i-g-t 01/23] i915/perf: Check regularly if we are done reading reports

[Umesh Nerlige Ramappa]

If we pass a user buffer as large as the OA buffer, the OA unit will
return all available reports. In the second part of the
buffer-fill/enable-disable tests, we are collecting all reports within a
time window, but we only check the time window in the outer while loop.
We may have read well past that window if OA returns more reports. Fix
this rare failure by checking for the window within the inner for loop.

Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
---
 tests/i915/perf.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/tests/i915/perf.c b/tests/i915/perf.c
index 5502a3fb..51e2dc87 100644
--- a/tests/i915/perf.c
+++ b/tests/i915/perf.c
@@ -2522,6 +2522,9 @@ test_buffer_fill(void)
 						first_timestamp = report[1];
 					last_timestamp = report[1];
 
+					if (((last_timestamp - first_timestamp) * oa_period) < (fill_duration / 2))
+						break;
+
 					if (oa_report_is_periodic(oa_exponent, report)) {
 						memcpy(last_periodic_report, report,
 						       sizeof(last_periodic_report));
@@ -2537,6 +2540,8 @@ test_buffer_fill(void)
 
 		do_ioctl(stream_fd, I915_PERF_IOCTL_DISABLE, 0);
 
+		igt_debug("first ts = %u, last ts = %u\n", first_timestamp, last_timestamp);
+
 		igt_debug("%f < %zu < %f\n",
 			  report_size * n_full_oa_reports * 0.45,
 			  n_periodic_reports * report_size,
@@ -2732,6 +2737,9 @@ test_enable_disable(void)
 						  oa_report_is_periodic(oa_exponent, report),
 						  oa_report_get_ctx_id(report));
 
+					if (((last_timestamp - first_timestamp) * oa_period) < (fill_duration / 2))
+						break;
+
 					if (oa_report_is_periodic(oa_exponent, report)) {
 						memcpy(last_periodic_report, report,
 						       sizeof(last_periodic_report));
@@ -2755,6 +2763,8 @@ test_enable_disable(void)
 
 		do_ioctl(stream_fd, I915_PERF_IOCTL_DISABLE, 0);
 
+		igt_debug("first ts = %u, last ts = %u\n", first_timestamp, last_timestamp);
+
 		igt_debug("%f < %zu < %f\n",
 			  report_size * n_full_oa_reports * 0.45,
 			  n_periodic_reports * report_size,
-- 
2.25.1

Re: [igt-dev] [PATCH i-g-t 01/23] i915/perf: Check regularly if we are done reading reports

[Lionel Landwerlin]

On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
> If we pass a user buffer as large as the OA buffer, the OA unit will
> return all available reports. In the second part of the
> buffer-fill/enable-disable tests, we are collecting all reports within a
> time window, but we only check the time window in the outer while loop.
> We may have read well past that window if OA returns more reports. Fix
> this rare failure by checking for the window within the inner for loop.
>
> Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
> ---
>   tests/i915/perf.c | 10 ++++++++++
>   1 file changed, 10 insertions(+)
>
> diff --git a/tests/i915/perf.c b/tests/i915/perf.c
> index 5502a3fb..51e2dc87 100644
> --- a/tests/i915/perf.c
> +++ b/tests/i915/perf.c
> @@ -2522,6 +2522,9 @@ test_buffer_fill(void)
>   						first_timestamp = report[1];
>   					last_timestamp = report[1];
>   
> +					if (((last_timestamp - first_timestamp) * oa_period) < (fill_duration / 2))
> +						break;
> +
>   					if (oa_report_is_periodic(oa_exponent, report)) {
>   						memcpy(last_periodic_report, report,
>   						       sizeof(last_periodic_report));
> @@ -2537,6 +2540,8 @@ test_buffer_fill(void)
>   
>   		do_ioctl(stream_fd, I915_PERF_IOCTL_DISABLE, 0);
>   
> +		igt_debug("first ts = %u, last ts = %u\n", first_timestamp, last_timestamp);
> +
>   		igt_debug("%f < %zu < %f\n",
>   			  report_size * n_full_oa_reports * 0.45,
>   			  n_periodic_reports * report_size,
> @@ -2732,6 +2737,9 @@ test_enable_disable(void)
>   						  oa_report_is_periodic(oa_exponent, report),
>   						  oa_report_get_ctx_id(report));
>   
> +					if (((last_timestamp - first_timestamp) * oa_period) < (fill_duration / 2))
> +						break;
> +
>   					if (oa_report_is_periodic(oa_exponent, report)) {
>   						memcpy(last_periodic_report, report,
>   						       sizeof(last_periodic_report));
> @@ -2755,6 +2763,8 @@ test_enable_disable(void)
>   
>   		do_ioctl(stream_fd, I915_PERF_IOCTL_DISABLE, 0);
>   
> +		igt_debug("first ts = %u, last ts = %u\n", first_timestamp, last_timestamp);
> +
>   		igt_debug("%f < %zu < %f\n",
>   			  report_size * n_full_oa_reports * 0.45,
>   			  n_periodic_reports * report_size,

[igt-dev] [PATCH i-g-t 02/23] i915/perf: Fix OA short_reads test

[Umesh Nerlige Ramappa]

With 5000 ns oa exponent, report lost errors can occur. A report lost
header can be read by the user even when there are no valid OA report
samples, so we get an 8 byte read. This defeats the EFAULT check.

When checking for EFAULT on a sampled report, discard any report lost
errors.

Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
---
 tests/i915/perf.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/tests/i915/perf.c b/tests/i915/perf.c
index 51e2dc87..0b8f6ac8 100644
--- a/tests/i915/perf.c
+++ b/tests/i915/perf.c
@@ -2853,8 +2853,13 @@ test_short_reads(void)
 
 	/* A read that can't return a single record because it would result
 	 * in a fault on buffer overrun should result in an EFAULT error...
+	 *
+	 * Make sure to weed out all report lost errors before verifying EFAULT.
 	 */
-	ret = read(stream_fd, pages + page_size - 16, page_size);
+	header = (void *)(pages + page_size - 16);
+	do {
+		ret = read(stream_fd, header, page_size);
+	} while (ret > 0 && header->type == DRM_I915_PERF_RECORD_OA_REPORT_LOST);
 	igt_assert_eq(ret, -1);
 	igt_assert_eq(errno, EFAULT);
 
-- 
2.25.1

Re: [igt-dev] [PATCH i-g-t 02/23] i915/perf: Fix OA short_reads test

[Lionel Landwerlin]

On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
> With 5000 ns oa exponent, report lost errors can occur. A report lost
> header can be read by the user even when there are no valid OA report
> samples, so we get an 8 byte read. This defeats the EFAULT check.
>
> When checking for EFAULT on a sampled report, discard any report lost
> errors.
>
> Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
> ---
>   tests/i915/perf.c | 7 ++++++-
>   1 file changed, 6 insertions(+), 1 deletion(-)
>
> diff --git a/tests/i915/perf.c b/tests/i915/perf.c
> index 51e2dc87..0b8f6ac8 100644
> --- a/tests/i915/perf.c
> +++ b/tests/i915/perf.c
> @@ -2853,8 +2853,13 @@ test_short_reads(void)
>   
>   	/* A read that can't return a single record because it would result
>   	 * in a fault on buffer overrun should result in an EFAULT error...
> +	 *
> +	 * Make sure to weed out all report lost errors before verifying EFAULT.
>   	 */
> -	ret = read(stream_fd, pages + page_size - 16, page_size);
> +	header = (void *)(pages + page_size - 16);
> +	do {
> +		ret = read(stream_fd, header, page_size);
> +	} while (ret > 0 && header->type == DRM_I915_PERF_RECORD_OA_REPORT_LOST);
>   	igt_assert_eq(ret, -1);
>   	igt_assert_eq(errno, EFAULT);
>   

[igt-dev] [PATCH i-g-t 03/23] i915/perf: Check return value from getparam

[Umesh Nerlige Ramappa]

In some cases getparam could fail. Check return from getparam and fail
early on.

Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
---
 lib/i915/perf.c | 28 +++++++++++++++-------------
 1 file changed, 15 insertions(+), 13 deletions(-)

diff --git a/lib/i915/perf.c b/lib/i915/perf.c
index d88835ff..7349a460 100644
--- a/lib/i915/perf.c
+++ b/lib/i915/perf.c
@@ -287,19 +287,16 @@ intel_perf_for_devinfo(uint32_t device_id,
 	return perf;
 }
 
-static uint32_t
-getparam(int drm_fd, uint32_t param)
+static int
+getparam(int drm_fd, uint32_t param, uint32_t *val)
 {
-        struct drm_i915_getparam gp;
-        int val = -1;
-
-        memset(&gp, 0, sizeof(gp));
-        gp.param = param;
-        gp.value = &val;
+	struct drm_i915_getparam gp;
 
-	perf_ioctl(drm_fd, DRM_IOCTL_I915_GETPARAM, &gp);
+	memset(&gp, 0, sizeof(gp));
+	gp.param = param;
+	gp.value = (int *)val;
 
-        return val;
+	return perf_ioctl(drm_fd, DRM_IOCTL_I915_GETPARAM, &gp);
 }
 
 static bool
@@ -415,9 +412,9 @@ open_master_sysfs_dir(int drm_fd)
 struct intel_perf *
 intel_perf_for_fd(int drm_fd)
 {
-	uint32_t device_id = getparam(drm_fd, I915_PARAM_CHIPSET_ID);
-	uint32_t device_revision = getparam(drm_fd, I915_PARAM_REVISION);
-	uint32_t timestamp_frequency = getparam(drm_fd, I915_PARAM_CS_TIMESTAMP_FREQUENCY);
+	uint32_t device_id;
+	uint32_t device_revision;
+	uint32_t timestamp_frequency;
 	uint64_t gt_min_freq;
 	uint64_t gt_max_freq;
 	struct drm_i915_query_topology_info *topology;
@@ -434,6 +431,11 @@ intel_perf_for_fd(int drm_fd)
 	}
 	close(sysfs_dir_fd);
 
+	if (getparam(drm_fd, I915_PARAM_CHIPSET_ID, &device_id) ||
+	    getparam(drm_fd, I915_PARAM_REVISION, &device_revision) ||
+	    getparam(drm_fd, I915_PARAM_CS_TIMESTAMP_FREQUENCY, &timestamp_frequency))
+		return NULL;
+
 	topology = query_topology(drm_fd);
 	if (!topology)
 		return NULL;
-- 
2.25.1

Re: [igt-dev] [PATCH i-g-t 03/23] i915/perf: Check return value from getparam

[Lionel Landwerlin]

On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
> In some cases getparam could fail. Check return from getparam and fail
> early on.
>
> Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>


Something is odd with the indentation but otherwise :


Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>


> ---
>   lib/i915/perf.c | 28 +++++++++++++++-------------
>   1 file changed, 15 insertions(+), 13 deletions(-)
>
> diff --git a/lib/i915/perf.c b/lib/i915/perf.c
> index d88835ff..7349a460 100644
> --- a/lib/i915/perf.c
> +++ b/lib/i915/perf.c
> @@ -287,19 +287,16 @@ intel_perf_for_devinfo(uint32_t device_id,
>   	return perf;
>   }
>   
> -static uint32_t
> -getparam(int drm_fd, uint32_t param)
> +static int
> +getparam(int drm_fd, uint32_t param, uint32_t *val)
>   {
> -        struct drm_i915_getparam gp;
> -        int val = -1;
> -
> -        memset(&gp, 0, sizeof(gp));
> -        gp.param = param;
> -        gp.value = &val;
> +	struct drm_i915_getparam gp;
>   
> -	perf_ioctl(drm_fd, DRM_IOCTL_I915_GETPARAM, &gp);
> +	memset(&gp, 0, sizeof(gp));
> +	gp.param = param;
> +	gp.value = (int *)val;
>   
> -        return val;
> +	return perf_ioctl(drm_fd, DRM_IOCTL_I915_GETPARAM, &gp);
>   }
>   
>   static bool
> @@ -415,9 +412,9 @@ open_master_sysfs_dir(int drm_fd)
>   struct intel_perf *
>   intel_perf_for_fd(int drm_fd)
>   {
> -	uint32_t device_id = getparam(drm_fd, I915_PARAM_CHIPSET_ID);
> -	uint32_t device_revision = getparam(drm_fd, I915_PARAM_REVISION);
> -	uint32_t timestamp_frequency = getparam(drm_fd, I915_PARAM_CS_TIMESTAMP_FREQUENCY);
> +	uint32_t device_id;
> +	uint32_t device_revision;
> +	uint32_t timestamp_frequency;
>   	uint64_t gt_min_freq;
>   	uint64_t gt_max_freq;
>   	struct drm_i915_query_topology_info *topology;
> @@ -434,6 +431,11 @@ intel_perf_for_fd(int drm_fd)
>   	}
>   	close(sysfs_dir_fd);
>   
> +	if (getparam(drm_fd, I915_PARAM_CHIPSET_ID, &device_id) ||
> +	    getparam(drm_fd, I915_PARAM_REVISION, &device_revision) ||
> +	    getparam(drm_fd, I915_PARAM_CS_TIMESTAMP_FREQUENCY, &timestamp_frequency))
> +		return NULL;
> +
>   	topology = query_topology(drm_fd);
>   	if (!topology)
>   		return NULL;

[igt-dev] [PATCH i-g-t 04/23] i915/perf: Limit sseu-config tests for gen11

[Umesh Nerlige Ramappa]

From: "Nerlige Ramappa, Umesh" <umesh.nerlige.ramappa@intel.com>

On Gen11, the Media driver will need to limit dispatch to only big
subslices (usually half the EUs depending on EU fusing). If media,
render and OA are running in parallel, then OA will lose the NOA config
because half of the EUs will be turned off by media. To avoid this with
concurrent use cases, an sseu config to keep all EUs powered on is
passed to OA in the i915_perf_open_ioctl params. This configuration is
made part of the render contexts, so that it's applied when 3d contexts
are running. Since this is an issue on gen11, limit sseu-config tests to
gen11.

Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
---
 tests/i915/perf.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tests/i915/perf.c b/tests/i915/perf.c
index 0b8f6ac8..ca3ccc17 100644
--- a/tests/i915/perf.c
+++ b/tests/i915/perf.c
@@ -5112,6 +5112,7 @@ igt_main
 	igt_subtest_group {
 		igt_fixture {
 			igt_require(i915_perf_revision(drm_fd) >= 4);
+			igt_require(intel_gen(devid) == 11);
 		}
 
 		igt_describe("Verify invalid SSEU opening parameters");
-- 
2.25.1

Re: [igt-dev] [PATCH i-g-t 04/23] i915/perf: Limit sseu-config tests for gen11

[Lionel Landwerlin]

On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
> From: "Nerlige Ramappa, Umesh" <umesh.nerlige.ramappa@intel.com>
>
> On Gen11, the Media driver will need to limit dispatch to only big
> subslices (usually half the EUs depending on EU fusing). If media,
> render and OA are running in parallel, then OA will lose the NOA config
> because half of the EUs will be turned off by media. To avoid this with
> concurrent use cases, an sseu config to keep all EUs powered on is
> passed to OA in the i915_perf_open_ioctl params. This configuration is
> made part of the render contexts, so that it's applied when 3d contexts
> are running. Since this is an issue on gen11, limit sseu-config tests to
> gen11.
>
> Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
> ---
>   tests/i915/perf.c | 1 +
>   1 file changed, 1 insertion(+)
>
> diff --git a/tests/i915/perf.c b/tests/i915/perf.c
> index 0b8f6ac8..ca3ccc17 100644
> --- a/tests/i915/perf.c
> +++ b/tests/i915/perf.c
> @@ -5112,6 +5112,7 @@ igt_main
>   	igt_subtest_group {
>   		igt_fixture {
>   			igt_require(i915_perf_revision(drm_fd) >= 4);
> +			igt_require(intel_gen(devid) == 11);
>   		}
>   
>   		igt_describe("Verify invalid SSEU opening parameters");

[igt-dev] [PATCH i-g-t 05/23] i915/perf: Bump timestamp tolerance for DG1

[Umesh Nerlige Ramappa]

Timestamp deltas obtained from the batch vs the delta obtained from oa
reports have a delta that's between 4 to 5 us in DG1. Adjust the test
accordingly.

Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
---
 tests/i915/perf.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/tests/i915/perf.c b/tests/i915/perf.c
index ca3ccc17..95600562 100644
--- a/tests/i915/perf.c
+++ b/tests/i915/perf.c
@@ -4084,8 +4084,9 @@ static void gen12_single_ctx_helper(void)
 	delta_delta = delta_ts64_ns > delta_oa32_ns ?
 		      (delta_ts64_ns - delta_oa32_ns) :
 		      (delta_oa32_ns - delta_ts64_ns);
-	if (delta_delta > 500) {
-		igt_debug("delta_delta exceeds margin, skipping..\n");
+	if (delta_delta > 5000) {
+		igt_debug("delta_delta = %d. exceeds margin, skipping..\n",
+			  delta_delta);
 		exit(EAGAIN);
 	}
 
-- 
2.25.1

Re: [igt-dev] [PATCH i-g-t 05/23] i915/perf: Bump timestamp tolerance for DG1

[Lionel Landwerlin]

On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
> Timestamp deltas obtained from the batch vs the delta obtained from oa
> reports have a delta that's between 4 to 5 us in DG1. Adjust the test
> accordingly.
>
> Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>

It's odd...

4/5us between 2 commands that follow each other exactly in a command stream?

Even if the PIPE_CONTROL stalls, it's supposed to write the timestamp 
after it's done stalling.

And the MI_REPORT_PERF_COUNT command is just after.


If you're sure there is no other explanation : Reviewed-by: Lionel 
Landwerlin <lionel.g.landwerlin@intel.com>


> ---
>   tests/i915/perf.c | 5 +++--
>   1 file changed, 3 insertions(+), 2 deletions(-)
>
> diff --git a/tests/i915/perf.c b/tests/i915/perf.c
> index ca3ccc17..95600562 100644
> --- a/tests/i915/perf.c
> +++ b/tests/i915/perf.c
> @@ -4084,8 +4084,9 @@ static void gen12_single_ctx_helper(void)
>   	delta_delta = delta_ts64_ns > delta_oa32_ns ?
>   		      (delta_ts64_ns - delta_oa32_ns) :
>   		      (delta_oa32_ns - delta_ts64_ns);
> -	if (delta_delta > 500) {
> -		igt_debug("delta_delta exceeds margin, skipping..\n");
> +	if (delta_delta > 5000) {
> +		igt_debug("delta_delta = %d. exceeds margin, skipping..\n",
> +			  delta_delta);
>   		exit(EAGAIN);
>   	}
>   

Re: [igt-dev] [PATCH i-g-t 05/23] i915/perf: Bump timestamp tolerance for DG1

[Umesh Nerlige Ramappa]

On Tue, Sep 06, 2022 at 04:06:37PM +0300, Lionel Landwerlin wrote:
>On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
>>Timestamp deltas obtained from the batch vs the delta obtained from oa
>>reports have a delta that's between 4 to 5 us in DG1. Adjust the test
>>accordingly.
>>
>>Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
>
>It's odd...
>
>4/5us between 2 commands that follow each other exactly in a command stream?
>
>Even if the PIPE_CONTROL stalls, it's supposed to write the timestamp 
>after it's done stalling.
>
>And the MI_REPORT_PERF_COUNT command is just after.
>
>
>If you're sure there is no other explanation : Reviewed-by: Lionel 
>Landwerlin <lionel.g.landwerlin@intel.com>

Hmm, I debugged this a while ago. I will give it a quick retry and see 
if it's still the same issue.

>
>
>>---
>>  tests/i915/perf.c | 5 +++--
>>  1 file changed, 3 insertions(+), 2 deletions(-)
>>
>>diff --git a/tests/i915/perf.c b/tests/i915/perf.c
>>index ca3ccc17..95600562 100644
>>--- a/tests/i915/perf.c
>>+++ b/tests/i915/perf.c
>>@@ -4084,8 +4084,9 @@ static void gen12_single_ctx_helper(void)
>>  	delta_delta = delta_ts64_ns > delta_oa32_ns ?
>>  		      (delta_ts64_ns - delta_oa32_ns) :
>>  		      (delta_oa32_ns - delta_ts64_ns);
>>-	if (delta_delta > 500) {
>>-		igt_debug("delta_delta exceeds margin, skipping..\n");
>>+	if (delta_delta > 5000) {
>>+		igt_debug("delta_delta = %d. exceeds margin, skipping..\n",
>>+			  delta_delta);
>>  		exit(EAGAIN);
>>  	}
>
>

[igt-dev] [PATCH i-g-t 06/23] i915/perf: Account for OA sampling interval in polling test

[Umesh Nerlige Ramappa]

The polling test reads all the reports and then expects EAGAIN on a
subsequent read since it has consumed all available data. This
works well for OA sampling periods in the order of 10s of ms. For
smaller sampling periods like 500 us, we end up reading valid data and
the test fails. Fix the check to account for the OA sampling period.

Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
---
 tests/i915/perf.c | 16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

diff --git a/tests/i915/perf.c b/tests/i915/perf.c
index 95600562..5fd143e9 100644
--- a/tests/i915/perf.c
+++ b/tests/i915/perf.c
@@ -2220,14 +2220,22 @@ test_polling(uint64_t requested_oa_period, bool set_kernel_hrtimer, uint64_t ker
 			n_extra_iterations++;
 
 		/* At this point, after consuming pending reports (and hoping
-		 * the scheduler hasn't stopped us for too long we now
-		 * expect EAGAIN on read.
+		 * the scheduler hasn't stopped us for too long) we now expect
+		 * EAGAIN on read. While this works most of the times, there are
+		 * some rare failures when the OA period passed to this test is
+		 * very small (say 500 us) and that results in some valid
+		 * reports here. To weed out those rare occurences we assert
+		 * only if the OA period is >= 40 ms because 40 ms has withstood
+		 * the test of time on most platforms (ref: subtest: polling).
 		 */
 		while ((ret = read(stream_fd, buf, sizeof(buf))) < 0 &&
 		       errno == EINTR)
 			;
-		igt_assert_eq(ret, -1);
-		igt_assert_eq(errno, EAGAIN);
+
+		if (requested_oa_period >= 40000000) {
+			igt_assert_eq(ret, -1);
+			igt_assert_eq(errno, EAGAIN);
+		}
 
 		n++;
 	}
-- 
2.25.1

[igt-dev] [PATCH i-g-t 07/23] i915/perf: Add support for 64-bit counters

[Umesh Nerlige Ramappa]

Add support for 64-bit counters.

Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
---
 tests/i915/perf.c | 70 ++++++++++++++++++++++++++++++++++++++++++-----
 1 file changed, 63 insertions(+), 7 deletions(-)

diff --git a/tests/i915/perf.c b/tests/i915/perf.c
index 5fd143e9..19c33324 100644
--- a/tests/i915/perf.c
+++ b/tests/i915/perf.c
@@ -101,6 +101,8 @@ struct oa_format {
 	int a40_high_off; /* bytes */
 	int a40_low_off;
 	int n_a40;
+	int a64_off;
+	int n_a64;
 	int a_off;
 	int n_a;
 	int first_a;
@@ -455,7 +457,7 @@ max_oa_exponent_for_freq_gt(uint64_t frequency)
 static uint64_t
 oa_exponent_to_ns(int exponent)
 {
-       return 1000000000ULL * (2ULL << exponent) / intel_perf->devinfo.timestamp_frequency;
+	return 1000000000ULL * (2ULL << exponent) / intel_perf->devinfo.timestamp_frequency;
 }
 
 static bool
@@ -658,6 +660,15 @@ gen8_read_40bit_a_counter(const uint32_t *report,
 	return a40_low[a_id] | high;
 }
 
+static uint64_t
+gen12_read_64bit_a_counter(const uint32_t *report, enum drm_i915_oa_format fmt, int a_id)
+{
+	struct oa_format format = get_oa_format(fmt);
+	uint64_t *a64 = (uint64_t *)(((uint8_t *)report) + format.a64_off);
+
+	return a64[a_id];
+}
+
 static uint64_t
 gen8_40bit_a_delta(uint64_t value0, uint64_t value1)
 {
@@ -670,8 +681,8 @@ gen8_40bit_a_delta(uint64_t value0, uint64_t value1)
 static void
 accumulate_uint32(size_t offset,
 		  uint32_t *report0,
-                  uint32_t *report1,
-                  uint64_t *delta)
+		  uint32_t *report1,
+		  uint64_t *delta)
 {
 	uint32_t value0 = *(uint32_t *)(((uint8_t *)report0) + offset);
 	uint32_t value1 = *(uint32_t *)(((uint8_t *)report1) + offset);
@@ -681,10 +692,10 @@ accumulate_uint32(size_t offset,
 
 static void
 accumulate_uint40(int a_index,
-                  uint32_t *report0,
-                  uint32_t *report1,
+		  uint32_t *report0,
+		  uint32_t *report1,
 		  enum drm_i915_oa_format format,
-                  uint64_t *delta)
+		  uint64_t *delta)
 {
 	uint64_t value0 = gen8_read_40bit_a_counter(report0, format, a_index),
 		 value1 = gen8_read_40bit_a_counter(report1, format, a_index);
@@ -692,6 +703,19 @@ accumulate_uint40(int a_index,
 	*delta += gen8_40bit_a_delta(value0, value1);
 }
 
+static void
+accumulate_uint64(int a_index,
+		  const uint32_t *report0,
+		  const uint32_t *report1,
+		  enum drm_i915_oa_format format,
+		  uint64_t *delta)
+{
+	uint64_t value0 = gen12_read_64bit_a_counter(report0, format, a_index),
+		 value1 = gen12_read_64bit_a_counter(report1, format, a_index);
+
+	*delta += (value1 - value0);
+}
+
 static void
 accumulate_reports(struct accumulator *accumulator,
 		   uint32_t *start,
@@ -717,6 +741,11 @@ accumulate_reports(struct accumulator *accumulator,
 				  deltas + idx++);
 	}
 
+	for (int i = 0; i < format.n_a64; i++) {
+		accumulate_uint64(i, start, end, accumulator->format,
+				  deltas + idx++);
+	}
+
 	for (int i = 0; i < format.n_a; i++) {
 		accumulate_uint32(format.a_off + 4 * i,
 				  start, end, deltas + idx++);
@@ -747,6 +776,9 @@ accumulator_print(struct accumulator *accumulator, const char *title)
 
 		for (int i = 0; i < format.n_a40; i++)
 			igt_debug("\tA%u = %"PRIu64"\n", i, deltas[idx++]);
+
+		for (int i = 0; i < format.n_a64; i++)
+			igt_debug("\tA64_%u = %"PRIu64"\n", i, deltas[idx++]);
 	} else {
 		igt_debug("\ttime delta = %"PRIu64"\n", deltas[idx++]);
 	}
@@ -803,7 +835,19 @@ gen8_sanity_check_test_oa_reports(const uint32_t *oa_report0,
 		if (undefined_a_counters[j])
 			continue;
 
-		igt_debug("A%d: delta = %"PRIu64"\n", j, delta);
+		igt_debug("A40_%d: delta = %"PRIu64"\n", j, delta);
+		igt_assert(delta <= max_delta);
+	}
+
+	for (int j = 0; j < format.n_a64; j++) {
+		uint64_t delta = 0;
+
+		accumulate_uint64(j, oa_report0, oa_report1, fmt, &delta);
+
+		if (undefined_a_counters[j])
+			continue;
+
+		igt_debug("A64_%d: delta = %"PRIu64"\n", format.first_a + j, delta);
 		igt_assert(delta <= max_delta);
 	}
 
@@ -1368,6 +1412,18 @@ print_reports(uint32_t *oa_report0, uint32_t *oa_report1, int fmt)
 			  j, value0, value1, delta);
 	}
 
+	for (int j = 0; j < format.n_a64; j++) {
+		uint64_t value0 = gen12_read_64bit_a_counter(oa_report0, fmt, j);
+		uint64_t value1 = gen12_read_64bit_a_counter(oa_report1, fmt, j);
+		uint64_t delta = value1 - value0;
+
+		if (undefined_a_counters[j])
+			continue;
+
+		igt_debug("A_64%d: 1st = %"PRIu64", 2nd = %"PRIu64", delta = %"PRIu64"\n",
+			  format.first_a + j, value0, value1, delta);
+	}
+
 	for (int j = 0; j < format.n_a; j++) {
 		uint32_t *a0 = (uint32_t *)(((uint8_t *)oa_report0) +
 					    format.a_off);
-- 
2.25.1

Re: [igt-dev] [PATCH i-g-t 07/23] i915/perf: Add support for 64-bit counters

[Lionel Landwerlin]

On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
> Add support for 64-bit counters.
>
> Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>

With the unnecessary changes removed :

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>

> ---
>   tests/i915/perf.c | 70 ++++++++++++++++++++++++++++++++++++++++++-----
>   1 file changed, 63 insertions(+), 7 deletions(-)
>
> diff --git a/tests/i915/perf.c b/tests/i915/perf.c
> index 5fd143e9..19c33324 100644
> --- a/tests/i915/perf.c
> +++ b/tests/i915/perf.c
> @@ -101,6 +101,8 @@ struct oa_format {
>   	int a40_high_off; /* bytes */
>   	int a40_low_off;
>   	int n_a40;
> +	int a64_off;
> +	int n_a64;
>   	int a_off;
>   	int n_a;
>   	int first_a;
> @@ -455,7 +457,7 @@ max_oa_exponent_for_freq_gt(uint64_t frequency)
>   static uint64_t
>   oa_exponent_to_ns(int exponent)
>   {
> -       return 1000000000ULL * (2ULL << exponent) / intel_perf->devinfo.timestamp_frequency;
> +	return 1000000000ULL * (2ULL << exponent) / intel_perf->devinfo.timestamp_frequency;
Can't see any change on the line above.
>   }
>   
>   static bool
> @@ -658,6 +660,15 @@ gen8_read_40bit_a_counter(const uint32_t *report,
>   	return a40_low[a_id] | high;
>   }
>   
> +static uint64_t
> +gen12_read_64bit_a_counter(const uint32_t *report, enum drm_i915_oa_format fmt, int a_id)
> +{
> +	struct oa_format format = get_oa_format(fmt);
> +	uint64_t *a64 = (uint64_t *)(((uint8_t *)report) + format.a64_off);
> +
> +	return a64[a_id];
> +}
> +
>   static uint64_t
>   gen8_40bit_a_delta(uint64_t value0, uint64_t value1)
>   {
> @@ -670,8 +681,8 @@ gen8_40bit_a_delta(uint64_t value0, uint64_t value1)
>   static void
>   accumulate_uint32(size_t offset,
>   		  uint32_t *report0,
> -                  uint32_t *report1,
> -                  uint64_t *delta)
> +		  uint32_t *report1,
> +		  uint64_t *delta)
This hunk appears to just be noise.
>   {
>   	uint32_t value0 = *(uint32_t *)(((uint8_t *)report0) + offset);
>   	uint32_t value1 = *(uint32_t *)(((uint8_t *)report1) + offset);
> @@ -681,10 +692,10 @@ accumulate_uint32(size_t offset,
>   
>   static void
>   accumulate_uint40(int a_index,
> -                  uint32_t *report0,
> -                  uint32_t *report1,
> +		  uint32_t *report0,
> +		  uint32_t *report1,
>   		  enum drm_i915_oa_format format,
> -                  uint64_t *delta)
> +		  uint64_t *delta)

Here too

>   {
>   	uint64_t value0 = gen8_read_40bit_a_counter(report0, format, a_index),
>   		 value1 = gen8_read_40bit_a_counter(report1, format, a_index);
> @@ -692,6 +703,19 @@ accumulate_uint40(int a_index,
>   	*delta += gen8_40bit_a_delta(value0, value1);
>   }
>   
> +static void
> +accumulate_uint64(int a_index,
> +		  const uint32_t *report0,
> +		  const uint32_t *report1,
> +		  enum drm_i915_oa_format format,
> +		  uint64_t *delta)
> +{
> +	uint64_t value0 = gen12_read_64bit_a_counter(report0, format, a_index),
> +		 value1 = gen12_read_64bit_a_counter(report1, format, a_index);
> +
> +	*delta += (value1 - value0);
> +}
> +
>   static void
>   accumulate_reports(struct accumulator *accumulator,
>   		   uint32_t *start,
> @@ -717,6 +741,11 @@ accumulate_reports(struct accumulator *accumulator,
>   				  deltas + idx++);
>   	}
>   
> +	for (int i = 0; i < format.n_a64; i++) {
> +		accumulate_uint64(i, start, end, accumulator->format,
> +				  deltas + idx++);
> +	}
> +
>   	for (int i = 0; i < format.n_a; i++) {
>   		accumulate_uint32(format.a_off + 4 * i,
>   				  start, end, deltas + idx++);
> @@ -747,6 +776,9 @@ accumulator_print(struct accumulator *accumulator, const char *title)
>   
>   		for (int i = 0; i < format.n_a40; i++)
>   			igt_debug("\tA%u = %"PRIu64"\n", i, deltas[idx++]);
> +
> +		for (int i = 0; i < format.n_a64; i++)
> +			igt_debug("\tA64_%u = %"PRIu64"\n", i, deltas[idx++]);
>   	} else {
>   		igt_debug("\ttime delta = %"PRIu64"\n", deltas[idx++]);
>   	}
> @@ -803,7 +835,19 @@ gen8_sanity_check_test_oa_reports(const uint32_t *oa_report0,
>   		if (undefined_a_counters[j])
>   			continue;
>   
> -		igt_debug("A%d: delta = %"PRIu64"\n", j, delta);
> +		igt_debug("A40_%d: delta = %"PRIu64"\n", j, delta);
> +		igt_assert(delta <= max_delta);
> +	}
> +
> +	for (int j = 0; j < format.n_a64; j++) {
> +		uint64_t delta = 0;
> +
> +		accumulate_uint64(j, oa_report0, oa_report1, fmt, &delta);
> +
> +		if (undefined_a_counters[j])
> +			continue;
> +
> +		igt_debug("A64_%d: delta = %"PRIu64"\n", format.first_a + j, delta);
>   		igt_assert(delta <= max_delta);
>   	}
>   
> @@ -1368,6 +1412,18 @@ print_reports(uint32_t *oa_report0, uint32_t *oa_report1, int fmt)
>   			  j, value0, value1, delta);
>   	}
>   
> +	for (int j = 0; j < format.n_a64; j++) {
> +		uint64_t value0 = gen12_read_64bit_a_counter(oa_report0, fmt, j);
> +		uint64_t value1 = gen12_read_64bit_a_counter(oa_report1, fmt, j);
> +		uint64_t delta = value1 - value0;
> +
> +		if (undefined_a_counters[j])
> +			continue;
> +
> +		igt_debug("A_64%d: 1st = %"PRIu64", 2nd = %"PRIu64", delta = %"PRIu64"\n",
> +			  format.first_a + j, value0, value1, delta);
> +	}
> +
>   	for (int j = 0; j < format.n_a; j++) {
>   		uint32_t *a0 = (uint32_t *)(((uint8_t *)oa_report0) +
>   					    format.a_off);

Re: [igt-dev] [PATCH i-g-t 07/23] i915/perf: Add support for 64-bit counters

[Umesh Nerlige Ramappa]

On Tue, Sep 06, 2022 at 04:05:59PM +0300, Lionel Landwerlin wrote:
>On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
>>Add support for 64-bit counters.
>>
>>Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
>
>With the unnecessary changes removed :

The noise is from minor checkpatch corrections - spaces to tabs in the 
beginning of the line. I agree, it's not relevant to this patch.

Do you still want me to remove them? I am okay either ways.

Thanks,
Umesh

>
>Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
>
>>---
>>  tests/i915/perf.c | 70 ++++++++++++++++++++++++++++++++++++++++++-----
>>  1 file changed, 63 insertions(+), 7 deletions(-)
>>
>>diff --git a/tests/i915/perf.c b/tests/i915/perf.c
>>index 5fd143e9..19c33324 100644
>>--- a/tests/i915/perf.c
>>+++ b/tests/i915/perf.c
>>@@ -101,6 +101,8 @@ struct oa_format {
>>  	int a40_high_off; /* bytes */
>>  	int a40_low_off;
>>  	int n_a40;
>>+	int a64_off;
>>+	int n_a64;
>>  	int a_off;
>>  	int n_a;
>>  	int first_a;
>>@@ -455,7 +457,7 @@ max_oa_exponent_for_freq_gt(uint64_t frequency)
>>  static uint64_t
>>  oa_exponent_to_ns(int exponent)
>>  {
>>-       return 1000000000ULL * (2ULL << exponent) / intel_perf->devinfo.timestamp_frequency;
>>+	return 1000000000ULL * (2ULL << exponent) / intel_perf->devinfo.timestamp_frequency;
>Can't see any change on the line above.
>>  }
>>  static bool
>>@@ -658,6 +660,15 @@ gen8_read_40bit_a_counter(const uint32_t *report,
>>  	return a40_low[a_id] | high;
>>  }
>>+static uint64_t
>>+gen12_read_64bit_a_counter(const uint32_t *report, enum drm_i915_oa_format fmt, int a_id)
>>+{
>>+	struct oa_format format = get_oa_format(fmt);
>>+	uint64_t *a64 = (uint64_t *)(((uint8_t *)report) + format.a64_off);
>>+
>>+	return a64[a_id];
>>+}
>>+
>>  static uint64_t
>>  gen8_40bit_a_delta(uint64_t value0, uint64_t value1)
>>  {
>>@@ -670,8 +681,8 @@ gen8_40bit_a_delta(uint64_t value0, uint64_t value1)
>>  static void
>>  accumulate_uint32(size_t offset,
>>  		  uint32_t *report0,
>>-                  uint32_t *report1,
>>-                  uint64_t *delta)
>>+		  uint32_t *report1,
>>+		  uint64_t *delta)
>This hunk appears to just be noise.
>>  {
>>  	uint32_t value0 = *(uint32_t *)(((uint8_t *)report0) + offset);
>>  	uint32_t value1 = *(uint32_t *)(((uint8_t *)report1) + offset);
>>@@ -681,10 +692,10 @@ accumulate_uint32(size_t offset,
>>  static void
>>  accumulate_uint40(int a_index,
>>-                  uint32_t *report0,
>>-                  uint32_t *report1,
>>+		  uint32_t *report0,
>>+		  uint32_t *report1,
>>  		  enum drm_i915_oa_format format,
>>-                  uint64_t *delta)
>>+		  uint64_t *delta)
>
>Here too
>
>>  {
>>  	uint64_t value0 = gen8_read_40bit_a_counter(report0, format, a_index),
>>  		 value1 = gen8_read_40bit_a_counter(report1, format, a_index);
>>@@ -692,6 +703,19 @@ accumulate_uint40(int a_index,
>>  	*delta += gen8_40bit_a_delta(value0, value1);
>>  }
>>+static void
>>+accumulate_uint64(int a_index,
>>+		  const uint32_t *report0,
>>+		  const uint32_t *report1,
>>+		  enum drm_i915_oa_format format,
>>+		  uint64_t *delta)
>>+{
>>+	uint64_t value0 = gen12_read_64bit_a_counter(report0, format, a_index),
>>+		 value1 = gen12_read_64bit_a_counter(report1, format, a_index);
>>+
>>+	*delta += (value1 - value0);
>>+}
>>+
>>  static void
>>  accumulate_reports(struct accumulator *accumulator,
>>  		   uint32_t *start,
>>@@ -717,6 +741,11 @@ accumulate_reports(struct accumulator *accumulator,
>>  				  deltas + idx++);
>>  	}
>>+	for (int i = 0; i < format.n_a64; i++) {
>>+		accumulate_uint64(i, start, end, accumulator->format,
>>+				  deltas + idx++);
>>+	}
>>+
>>  	for (int i = 0; i < format.n_a; i++) {
>>  		accumulate_uint32(format.a_off + 4 * i,
>>  				  start, end, deltas + idx++);
>>@@ -747,6 +776,9 @@ accumulator_print(struct accumulator *accumulator, const char *title)
>>  		for (int i = 0; i < format.n_a40; i++)
>>  			igt_debug("\tA%u = %"PRIu64"\n", i, deltas[idx++]);
>>+
>>+		for (int i = 0; i < format.n_a64; i++)
>>+			igt_debug("\tA64_%u = %"PRIu64"\n", i, deltas[idx++]);
>>  	} else {
>>  		igt_debug("\ttime delta = %"PRIu64"\n", deltas[idx++]);
>>  	}
>>@@ -803,7 +835,19 @@ gen8_sanity_check_test_oa_reports(const uint32_t *oa_report0,
>>  		if (undefined_a_counters[j])
>>  			continue;
>>-		igt_debug("A%d: delta = %"PRIu64"\n", j, delta);
>>+		igt_debug("A40_%d: delta = %"PRIu64"\n", j, delta);
>>+		igt_assert(delta <= max_delta);
>>+	}
>>+
>>+	for (int j = 0; j < format.n_a64; j++) {
>>+		uint64_t delta = 0;
>>+
>>+		accumulate_uint64(j, oa_report0, oa_report1, fmt, &delta);
>>+
>>+		if (undefined_a_counters[j])
>>+			continue;
>>+
>>+		igt_debug("A64_%d: delta = %"PRIu64"\n", format.first_a + j, delta);
>>  		igt_assert(delta <= max_delta);
>>  	}
>>@@ -1368,6 +1412,18 @@ print_reports(uint32_t *oa_report0, uint32_t *oa_report1, int fmt)
>>  			  j, value0, value1, delta);
>>  	}
>>+	for (int j = 0; j < format.n_a64; j++) {
>>+		uint64_t value0 = gen12_read_64bit_a_counter(oa_report0, fmt, j);
>>+		uint64_t value1 = gen12_read_64bit_a_counter(oa_report1, fmt, j);
>>+		uint64_t delta = value1 - value0;
>>+
>>+		if (undefined_a_counters[j])
>>+			continue;
>>+
>>+		igt_debug("A_64%d: 1st = %"PRIu64", 2nd = %"PRIu64", delta = %"PRIu64"\n",
>>+			  format.first_a + j, value0, value1, delta);
>>+	}
>>+
>>  	for (int j = 0; j < format.n_a; j++) {
>>  		uint32_t *a0 = (uint32_t *)(((uint8_t *)oa_report0) +
>>  					    format.a_off);
>
>

Re: [igt-dev] [PATCH i-g-t 07/23] i915/perf: Add support for 64-bit counters

[Lionel Landwerlin]

On 06/09/2022 22:37, Umesh Nerlige Ramappa wrote:
> On Tue, Sep 06, 2022 at 04:05:59PM +0300, Lionel Landwerlin wrote:
>> On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
>>> Add support for 64-bit counters.
>>>
>>> Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
>>
>> With the unnecessary changes removed :
>
> The noise is from minor checkpatch corrections - spaces to tabs in the 
> beginning of the line. I agree, it's not relevant to this patch.
>
> Do you still want me to remove them? I am okay either ways.
>
> Thanks,
> Umesh


I'm not familiar with checkpatch but I'm struggling to see why it would 
insert random changes like that.

If you ran it on previous patches it would have produced the same 
isolated changes no?

It would be best to have single commit doing coding style changes at the 
beginning or end of the series.


Thanks,


-Lionel


>
>>
>> Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
>>
>>> ---
>>>  tests/i915/perf.c | 70 ++++++++++++++++++++++++++++++++++++++++++-----
>>>  1 file changed, 63 insertions(+), 7 deletions(-)
>>>
>>> diff --git a/tests/i915/perf.c b/tests/i915/perf.c
>>> index 5fd143e9..19c33324 100644
>>> --- a/tests/i915/perf.c
>>> +++ b/tests/i915/perf.c
>>> @@ -101,6 +101,8 @@ struct oa_format {
>>>      int a40_high_off; /* bytes */
>>>      int a40_low_off;
>>>      int n_a40;
>>> +    int a64_off;
>>> +    int n_a64;
>>>      int a_off;
>>>      int n_a;
>>>      int first_a;
>>> @@ -455,7 +457,7 @@ max_oa_exponent_for_freq_gt(uint64_t frequency)
>>>  static uint64_t
>>>  oa_exponent_to_ns(int exponent)
>>>  {
>>> -       return 1000000000ULL * (2ULL << exponent) / 
>>> intel_perf->devinfo.timestamp_frequency;
>>> +    return 1000000000ULL * (2ULL << exponent) / 
>>> intel_perf->devinfo.timestamp_frequency;
>> Can't see any change on the line above.
>>>  }
>>>  static bool
>>> @@ -658,6 +660,15 @@ gen8_read_40bit_a_counter(const uint32_t *report,
>>>      return a40_low[a_id] | high;
>>>  }
>>> +static uint64_t
>>> +gen12_read_64bit_a_counter(const uint32_t *report, enum 
>>> drm_i915_oa_format fmt, int a_id)
>>> +{
>>> +    struct oa_format format = get_oa_format(fmt);
>>> +    uint64_t *a64 = (uint64_t *)(((uint8_t *)report) + 
>>> format.a64_off);
>>> +
>>> +    return a64[a_id];
>>> +}
>>> +
>>>  static uint64_t
>>>  gen8_40bit_a_delta(uint64_t value0, uint64_t value1)
>>>  {
>>> @@ -670,8 +681,8 @@ gen8_40bit_a_delta(uint64_t value0, uint64_t 
>>> value1)
>>>  static void
>>>  accumulate_uint32(size_t offset,
>>>            uint32_t *report0,
>>> -                  uint32_t *report1,
>>> -                  uint64_t *delta)
>>> +          uint32_t *report1,
>>> +          uint64_t *delta)
>> This hunk appears to just be noise.
>>>  {
>>>      uint32_t value0 = *(uint32_t *)(((uint8_t *)report0) + offset);
>>>      uint32_t value1 = *(uint32_t *)(((uint8_t *)report1) + offset);
>>> @@ -681,10 +692,10 @@ accumulate_uint32(size_t offset,
>>>  static void
>>>  accumulate_uint40(int a_index,
>>> -                  uint32_t *report0,
>>> -                  uint32_t *report1,
>>> +          uint32_t *report0,
>>> +          uint32_t *report1,
>>>            enum drm_i915_oa_format format,
>>> -                  uint64_t *delta)
>>> +          uint64_t *delta)
>>
>> Here too
>>
>>>  {
>>>      uint64_t value0 = gen8_read_40bit_a_counter(report0, format, 
>>> a_index),
>>>           value1 = gen8_read_40bit_a_counter(report1, format, a_index);
>>> @@ -692,6 +703,19 @@ accumulate_uint40(int a_index,
>>>      *delta += gen8_40bit_a_delta(value0, value1);
>>>  }
>>> +static void
>>> +accumulate_uint64(int a_index,
>>> +          const uint32_t *report0,
>>> +          const uint32_t *report1,
>>> +          enum drm_i915_oa_format format,
>>> +          uint64_t *delta)
>>> +{
>>> +    uint64_t value0 = gen12_read_64bit_a_counter(report0, format, 
>>> a_index),
>>> +         value1 = gen12_read_64bit_a_counter(report1, format, 
>>> a_index);
>>> +
>>> +    *delta += (value1 - value0);
>>> +}
>>> +
>>>  static void
>>>  accumulate_reports(struct accumulator *accumulator,
>>>             uint32_t *start,
>>> @@ -717,6 +741,11 @@ accumulate_reports(struct accumulator 
>>> *accumulator,
>>>                    deltas + idx++);
>>>      }
>>> +    for (int i = 0; i < format.n_a64; i++) {
>>> +        accumulate_uint64(i, start, end, accumulator->format,
>>> +                  deltas + idx++);
>>> +    }
>>> +
>>>      for (int i = 0; i < format.n_a; i++) {
>>>          accumulate_uint32(format.a_off + 4 * i,
>>>                    start, end, deltas + idx++);
>>> @@ -747,6 +776,9 @@ accumulator_print(struct accumulator 
>>> *accumulator, const char *title)
>>>          for (int i = 0; i < format.n_a40; i++)
>>>              igt_debug("\tA%u = %"PRIu64"\n", i, deltas[idx++]);
>>> +
>>> +        for (int i = 0; i < format.n_a64; i++)
>>> +            igt_debug("\tA64_%u = %"PRIu64"\n", i, deltas[idx++]);
>>>      } else {
>>>          igt_debug("\ttime delta = %"PRIu64"\n", deltas[idx++]);
>>>      }
>>> @@ -803,7 +835,19 @@ gen8_sanity_check_test_oa_reports(const 
>>> uint32_t *oa_report0,
>>>          if (undefined_a_counters[j])
>>>              continue;
>>> -        igt_debug("A%d: delta = %"PRIu64"\n", j, delta);
>>> +        igt_debug("A40_%d: delta = %"PRIu64"\n", j, delta);
>>> +        igt_assert(delta <= max_delta);
>>> +    }
>>> +
>>> +    for (int j = 0; j < format.n_a64; j++) {
>>> +        uint64_t delta = 0;
>>> +
>>> +        accumulate_uint64(j, oa_report0, oa_report1, fmt, &delta);
>>> +
>>> +        if (undefined_a_counters[j])
>>> +            continue;
>>> +
>>> +        igt_debug("A64_%d: delta = %"PRIu64"\n", format.first_a + 
>>> j, delta);
>>>          igt_assert(delta <= max_delta);
>>>      }
>>> @@ -1368,6 +1412,18 @@ print_reports(uint32_t *oa_report0, uint32_t 
>>> *oa_report1, int fmt)
>>>                j, value0, value1, delta);
>>>      }
>>> +    for (int j = 0; j < format.n_a64; j++) {
>>> +        uint64_t value0 = gen12_read_64bit_a_counter(oa_report0, 
>>> fmt, j);
>>> +        uint64_t value1 = gen12_read_64bit_a_counter(oa_report1, 
>>> fmt, j);
>>> +        uint64_t delta = value1 - value0;
>>> +
>>> +        if (undefined_a_counters[j])
>>> +            continue;
>>> +
>>> +        igt_debug("A_64%d: 1st = %"PRIu64", 2nd = %"PRIu64", delta 
>>> = %"PRIu64"\n",
>>> +              format.first_a + j, value0, value1, delta);
>>> +    }
>>> +
>>>      for (int j = 0; j < format.n_a; j++) {
>>>          uint32_t *a0 = (uint32_t *)(((uint8_t *)oa_report0) +
>>>                          format.a_off);
>>
>>

Re: [igt-dev] [PATCH i-g-t 07/23] i915/perf: Add support for 64-bit counters

[Umesh Nerlige Ramappa]

On Tue, Sep 06, 2022 at 11:02:49PM +0300, Lionel Landwerlin wrote:
>On 06/09/2022 22:37, Umesh Nerlige Ramappa wrote:
>>On Tue, Sep 06, 2022 at 04:05:59PM +0300, Lionel Landwerlin wrote:
>>>On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
>>>>Add support for 64-bit counters.
>>>>
>>>>Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
>>>
>>>With the unnecessary changes removed :
>>
>>The noise is from minor checkpatch corrections - spaces to tabs in 
>>the beginning of the line. I agree, it's not relevant to this patch.
>>
>>Do you still want me to remove them? I am okay either ways.
>>
>>Thanks,
>>Umesh
>
>
>I'm not familiar with checkpatch but I'm struggling to see why it 
>would insert random changes like that.
>
>If you ran it on previous patches it would have produced the same 
>isolated changes no?
>
>It would be best to have single commit doing coding style changes at 
>the beginning or end of the series.

I am going to drop the coding style changes here and follow that up as a 
separate series since it's not relevant here.

Thanks,
Umesh
>
>
>Thanks,
>
>
>-Lionel
>
>
>>
>>>
>>>Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
>>>
>>>>---
>>>> tests/i915/perf.c | 70 ++++++++++++++++++++++++++++++++++++++++++-----
>>>> 1 file changed, 63 insertions(+), 7 deletions(-)
>>>>
>>>>diff --git a/tests/i915/perf.c b/tests/i915/perf.c
>>>>index 5fd143e9..19c33324 100644
>>>>--- a/tests/i915/perf.c
>>>>+++ b/tests/i915/perf.c
>>>>@@ -101,6 +101,8 @@ struct oa_format {
>>>>     int a40_high_off; /* bytes */
>>>>     int a40_low_off;
>>>>     int n_a40;
>>>>+    int a64_off;
>>>>+    int n_a64;
>>>>     int a_off;
>>>>     int n_a;
>>>>     int first_a;
>>>>@@ -455,7 +457,7 @@ max_oa_exponent_for_freq_gt(uint64_t frequency)
>>>> static uint64_t
>>>> oa_exponent_to_ns(int exponent)
>>>> {
>>>>-       return 1000000000ULL * (2ULL << exponent) / 
>>>>intel_perf->devinfo.timestamp_frequency;
>>>>+    return 1000000000ULL * (2ULL << exponent) / 
>>>>intel_perf->devinfo.timestamp_frequency;
>>>Can't see any change on the line above.
>>>> }
>>>> static bool
>>>>@@ -658,6 +660,15 @@ gen8_read_40bit_a_counter(const uint32_t *report,
>>>>     return a40_low[a_id] | high;
>>>> }
>>>>+static uint64_t
>>>>+gen12_read_64bit_a_counter(const uint32_t *report, enum 
>>>>drm_i915_oa_format fmt, int a_id)
>>>>+{
>>>>+    struct oa_format format = get_oa_format(fmt);
>>>>+    uint64_t *a64 = (uint64_t *)(((uint8_t *)report) + 
>>>>format.a64_off);
>>>>+
>>>>+    return a64[a_id];
>>>>+}
>>>>+
>>>> static uint64_t
>>>> gen8_40bit_a_delta(uint64_t value0, uint64_t value1)
>>>> {
>>>>@@ -670,8 +681,8 @@ gen8_40bit_a_delta(uint64_t value0, uint64_t 
>>>>value1)
>>>> static void
>>>> accumulate_uint32(size_t offset,
>>>>           uint32_t *report0,
>>>>-                  uint32_t *report1,
>>>>-                  uint64_t *delta)
>>>>+          uint32_t *report1,
>>>>+          uint64_t *delta)
>>>This hunk appears to just be noise.
>>>> {
>>>>     uint32_t value0 = *(uint32_t *)(((uint8_t *)report0) + offset);
>>>>     uint32_t value1 = *(uint32_t *)(((uint8_t *)report1) + offset);
>>>>@@ -681,10 +692,10 @@ accumulate_uint32(size_t offset,
>>>> static void
>>>> accumulate_uint40(int a_index,
>>>>-                  uint32_t *report0,
>>>>-                  uint32_t *report1,
>>>>+          uint32_t *report0,
>>>>+          uint32_t *report1,
>>>>           enum drm_i915_oa_format format,
>>>>-                  uint64_t *delta)
>>>>+          uint64_t *delta)
>>>
>>>Here too
>>>
>>>> {
>>>>     uint64_t value0 = gen8_read_40bit_a_counter(report0, 
>>>>format, a_index),
>>>>          value1 = gen8_read_40bit_a_counter(report1, format, a_index);
>>>>@@ -692,6 +703,19 @@ accumulate_uint40(int a_index,
>>>>     *delta += gen8_40bit_a_delta(value0, value1);
>>>> }
>>>>+static void
>>>>+accumulate_uint64(int a_index,
>>>>+          const uint32_t *report0,
>>>>+          const uint32_t *report1,
>>>>+          enum drm_i915_oa_format format,
>>>>+          uint64_t *delta)
>>>>+{
>>>>+    uint64_t value0 = gen12_read_64bit_a_counter(report0, 
>>>>format, a_index),
>>>>+         value1 = gen12_read_64bit_a_counter(report1, format, 
>>>>a_index);
>>>>+
>>>>+    *delta += (value1 - value0);
>>>>+}
>>>>+
>>>> static void
>>>> accumulate_reports(struct accumulator *accumulator,
>>>>            uint32_t *start,
>>>>@@ -717,6 +741,11 @@ accumulate_reports(struct accumulator 
>>>>*accumulator,
>>>>                   deltas + idx++);
>>>>     }
>>>>+    for (int i = 0; i < format.n_a64; i++) {
>>>>+        accumulate_uint64(i, start, end, accumulator->format,
>>>>+                  deltas + idx++);
>>>>+    }
>>>>+
>>>>     for (int i = 0; i < format.n_a; i++) {
>>>>         accumulate_uint32(format.a_off + 4 * i,
>>>>                   start, end, deltas + idx++);
>>>>@@ -747,6 +776,9 @@ accumulator_print(struct accumulator 
>>>>*accumulator, const char *title)
>>>>         for (int i = 0; i < format.n_a40; i++)
>>>>             igt_debug("\tA%u = %"PRIu64"\n", i, deltas[idx++]);
>>>>+
>>>>+        for (int i = 0; i < format.n_a64; i++)
>>>>+            igt_debug("\tA64_%u = %"PRIu64"\n", i, deltas[idx++]);
>>>>     } else {
>>>>         igt_debug("\ttime delta = %"PRIu64"\n", deltas[idx++]);
>>>>     }
>>>>@@ -803,7 +835,19 @@ gen8_sanity_check_test_oa_reports(const 
>>>>uint32_t *oa_report0,
>>>>         if (undefined_a_counters[j])
>>>>             continue;
>>>>-        igt_debug("A%d: delta = %"PRIu64"\n", j, delta);
>>>>+        igt_debug("A40_%d: delta = %"PRIu64"\n", j, delta);
>>>>+        igt_assert(delta <= max_delta);
>>>>+    }
>>>>+
>>>>+    for (int j = 0; j < format.n_a64; j++) {
>>>>+        uint64_t delta = 0;
>>>>+
>>>>+        accumulate_uint64(j, oa_report0, oa_report1, fmt, &delta);
>>>>+
>>>>+        if (undefined_a_counters[j])
>>>>+            continue;
>>>>+
>>>>+        igt_debug("A64_%d: delta = %"PRIu64"\n", format.first_a 
>>>>+ j, delta);
>>>>         igt_assert(delta <= max_delta);
>>>>     }
>>>>@@ -1368,6 +1412,18 @@ print_reports(uint32_t *oa_report0, 
>>>>uint32_t *oa_report1, int fmt)
>>>>               j, value0, value1, delta);
>>>>     }
>>>>+    for (int j = 0; j < format.n_a64; j++) {
>>>>+        uint64_t value0 = 
>>>>gen12_read_64bit_a_counter(oa_report0, fmt, j);
>>>>+        uint64_t value1 = 
>>>>gen12_read_64bit_a_counter(oa_report1, fmt, j);
>>>>+        uint64_t delta = value1 - value0;
>>>>+
>>>>+        if (undefined_a_counters[j])
>>>>+            continue;
>>>>+
>>>>+        igt_debug("A_64%d: 1st = %"PRIu64", 2nd = %"PRIu64", 
>>>>delta = %"PRIu64"\n",
>>>>+              format.first_a + j, value0, value1, delta);
>>>>+    }
>>>>+
>>>>     for (int j = 0; j < format.n_a; j++) {
>>>>         uint32_t *a0 = (uint32_t *)(((uint8_t *)oa_report0) +
>>>>                         format.a_off);
>>>
>>>
>

[igt-dev] [PATCH i-g-t 08/23] i915/perf: Define OA report types

[Umesh Nerlige Ramappa]

Differentiate between OAG and OAR format types for tests like
oa-formats.

Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
---
 tests/i915/perf.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/tests/i915/perf.c b/tests/i915/perf.c
index 19c33324..b61dd6e2 100644
--- a/tests/i915/perf.c
+++ b/tests/i915/perf.c
@@ -95,6 +95,13 @@ struct accumulator {
 	uint64_t deltas[MAX_RAW_OA_COUNTERS];
 };
 
+enum {
+	OAG,
+	OAR,
+
+	MAX_OA_TYPE,
+};
+
 struct oa_format {
 	const char *name;
 	size_t size;
@@ -110,6 +117,7 @@ struct oa_format {
 	int n_b;
 	int c_off;
 	int n_c;
+	int oa_type;
 };
 
 static struct oa_format hsw_oa_formats[I915_OA_FORMAT_MAX] = {
@@ -1539,6 +1547,9 @@ test_oa_formats(void)
 		if (!format.name) /* sparse, indexed by ID */
 			continue;
 
+		if (format.oa_type != OAG) /* sparse, indexed by ID */
+			continue;
+
 		igt_debug("Checking OA format %s\n", format.name);
 
 		open_and_read_2_oa_reports(i,
-- 
2.25.1

Re: [igt-dev] [PATCH i-g-t 08/23] i915/perf: Define OA report types

[Lionel Landwerlin]

On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
> Differentiate between OAG and OAR format types for tests like
> oa-formats.
>
> Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
> ---
>   tests/i915/perf.c | 11 +++++++++++
>   1 file changed, 11 insertions(+)
>
> diff --git a/tests/i915/perf.c b/tests/i915/perf.c
> index 19c33324..b61dd6e2 100644
> --- a/tests/i915/perf.c
> +++ b/tests/i915/perf.c
> @@ -95,6 +95,13 @@ struct accumulator {
>   	uint64_t deltas[MAX_RAW_OA_COUNTERS];
>   };
>   
> +enum {
> +	OAG,
> +	OAR,
> +
> +	MAX_OA_TYPE,
> +};
> +
>   struct oa_format {
>   	const char *name;
>   	size_t size;
> @@ -110,6 +117,7 @@ struct oa_format {
>   	int n_b;
>   	int c_off;
>   	int n_c;
> +	int oa_type;
>   };
>   
>   static struct oa_format hsw_oa_formats[I915_OA_FORMAT_MAX] = {
> @@ -1539,6 +1547,9 @@ test_oa_formats(void)
>   		if (!format.name) /* sparse, indexed by ID */
>   			continue;
>   
> +		if (format.oa_type != OAG) /* sparse, indexed by ID */
> +			continue;
> +

I can't make sense of this comment.

Also you're not going to test OAR reports?


>   		igt_debug("Checking OA format %s\n", format.name);
>   
>   		open_and_read_2_oa_reports(i,

Re: [igt-dev] [PATCH i-g-t 08/23] i915/perf: Define OA report types

[Umesh Nerlige Ramappa]

On Tue, Sep 06, 2022 at 04:08:15PM +0300, Lionel Landwerlin wrote:
>On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
>>Differentiate between OAG and OAR format types for tests like
>>oa-formats.
>>
>>Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
>>---
>>  tests/i915/perf.c | 11 +++++++++++
>>  1 file changed, 11 insertions(+)
>>
>>diff --git a/tests/i915/perf.c b/tests/i915/perf.c
>>index 19c33324..b61dd6e2 100644
>>--- a/tests/i915/perf.c
>>+++ b/tests/i915/perf.c
>>@@ -95,6 +95,13 @@ struct accumulator {
>>  	uint64_t deltas[MAX_RAW_OA_COUNTERS];
>>  };
>>+enum {
>>+	OAG,
>>+	OAR,
>>+
>>+	MAX_OA_TYPE,
>>+};
>>+
>>  struct oa_format {
>>  	const char *name;
>>  	size_t size;
>>@@ -110,6 +117,7 @@ struct oa_format {
>>  	int n_b;
>>  	int c_off;
>>  	int n_c;
>>+	int oa_type;
>>  };
>>  static struct oa_format hsw_oa_formats[I915_OA_FORMAT_MAX] = {
>>@@ -1539,6 +1547,9 @@ test_oa_formats(void)
>>  		if (!format.name) /* sparse, indexed by ID */
>>  			continue;
>>+		if (format.oa_type != OAG) /* sparse, indexed by ID */
>>+			continue;
>>+
>
>I can't make sense of this comment.

sorry, now that I read the commit msg, it lacks details. The idea is to 
only test OAG reports in oa-formats test. OAR reports must be tested 
with the MI RPC command. This was not an issue up until TGL. In later 
platforms the OAR and OAG formats have subtle differences that would 
break the sanity check.

>
>Also you're not going to test OAR reports?
>

MI RPC should ideally loop over all OAR formats. Currently it uses only 
one OAR report. Test needs to be updated.

Thanks,
Umesh

>
>>  		igt_debug("Checking OA format %s\n", format.name);
>>  		open_and_read_2_oa_reports(i,
>
>

[igt-dev] [PATCH i-g-t 09/23] i915/perf: Use ARRAY_SIZE consistently for num_properties

[Umesh Nerlige Ramappa]

Replace all variations of calculating the num_properties with
ARRAY_SIZE.

Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
---
 tests/i915/perf.c | 62 +++++++++++++++++++++++------------------------
 1 file changed, 30 insertions(+), 32 deletions(-)

diff --git a/tests/i915/perf.c b/tests/i915/perf.c
index b61dd6e2..c120c898 100644
--- a/tests/i915/perf.c
+++ b/tests/i915/perf.c
@@ -86,8 +86,6 @@ IGT_TEST_DESCRIPTION("Test the i915 perf metrics streaming interface");
 
 #define MAX_OA_BUF_SIZE (16 * 1024 * 1024)
 
-#define NUM_PROPERTIES(p) (sizeof(p) / (2 * sizeof(uint64_t)))
-
 struct accumulator {
 #define MAX_RAW_OA_COUNTERS 62
 	enum drm_i915_oa_format format;
@@ -1069,7 +1067,7 @@ test_system_wide_paranoid(void)
 		struct drm_i915_perf_open_param param = {
 			.flags = I915_PERF_FLAG_FD_CLOEXEC |
 				I915_PERF_FLAG_FD_NONBLOCK,
-			.num_properties = sizeof(properties) / 16,
+			.num_properties = ARRAY_SIZE(properties) / 2,
 			.properties_ptr = to_user_pointer(properties),
 		};
 
@@ -1095,7 +1093,7 @@ test_system_wide_paranoid(void)
 		struct drm_i915_perf_open_param param = {
 			.flags = I915_PERF_FLAG_FD_CLOEXEC |
 				I915_PERF_FLAG_FD_NONBLOCK,
-			.num_properties = sizeof(properties) / 16,
+			.num_properties = ARRAY_SIZE(properties) / 2,
 			.properties_ptr = to_user_pointer(properties),
 		};
 		write_u64_file("/proc/sys/dev/i915/perf_stream_paranoid", 0);
@@ -1126,7 +1124,7 @@ test_invalid_open_flags(void)
 	};
 	struct drm_i915_perf_open_param param = {
 		.flags = ~0, /* Undefined flag bits set! */
-		.num_properties = sizeof(properties) / 16,
+		.num_properties = ARRAY_SIZE(properties) / 2,
 		.properties_ptr = to_user_pointer(properties),
 	};
 
@@ -1148,7 +1146,7 @@ test_invalid_oa_metric_set_id(void)
 	struct drm_i915_perf_open_param param = {
 		.flags = I915_PERF_FLAG_FD_CLOEXEC |
 			I915_PERF_FLAG_FD_NONBLOCK,
-		.num_properties = sizeof(properties) / 16,
+		.num_properties = ARRAY_SIZE(properties) / 2,
 		.properties_ptr = to_user_pointer(properties),
 	};
 
@@ -1182,7 +1180,7 @@ test_invalid_oa_format_id(void)
 	struct drm_i915_perf_open_param param = {
 		.flags = I915_PERF_FLAG_FD_CLOEXEC |
 			I915_PERF_FLAG_FD_NONBLOCK,
-		.num_properties = sizeof(properties) / 16,
+		.num_properties = ARRAY_SIZE(properties) / 2,
 		.properties_ptr = to_user_pointer(properties),
 	};
 
@@ -1214,7 +1212,7 @@ test_missing_sample_flags(void)
 	};
 	struct drm_i915_perf_open_param param = {
 		.flags = I915_PERF_FLAG_FD_CLOEXEC,
-		.num_properties = sizeof(properties) / 16,
+		.num_properties = ARRAY_SIZE(properties) / 2,
 		.properties_ptr = to_user_pointer(properties),
 	};
 
@@ -1354,7 +1352,7 @@ open_and_read_2_oa_reports(int format_id,
 	};
 	struct drm_i915_perf_open_param param = {
 		.flags = I915_PERF_FLAG_FD_CLOEXEC,
-		.num_properties = sizeof(properties) / 16,
+		.num_properties = ARRAY_SIZE(properties) / 2,
 		.properties_ptr = to_user_pointer(properties),
 	};
 
@@ -1833,7 +1831,7 @@ test_invalid_oa_exponent(void)
 	};
 	struct drm_i915_perf_open_param param = {
 		.flags = I915_PERF_FLAG_FD_CLOEXEC,
-		.num_properties = sizeof(properties) / 16,
+		.num_properties = ARRAY_SIZE(properties) / 2,
 		.properties_ptr = to_user_pointer(properties),
 	};
 
@@ -1868,7 +1866,7 @@ test_low_oa_exponent_permissions(void)
 	};
 	struct drm_i915_perf_open_param param = {
 		.flags = I915_PERF_FLAG_FD_CLOEXEC,
-		.num_properties = sizeof(properties) / 16,
+		.num_properties = ARRAY_SIZE(properties) / 2,
 		.properties_ptr = to_user_pointer(properties),
 	};
 	uint64_t oa_period, oa_freq;
@@ -1931,7 +1929,7 @@ test_per_context_mode_unprivileged(void)
 	};
 	struct drm_i915_perf_open_param param = {
 		.flags = I915_PERF_FLAG_FD_CLOEXEC,
-		.num_properties = sizeof(properties) / 16,
+		.num_properties = ARRAY_SIZE(properties) / 2,
 		.properties_ptr = to_user_pointer(properties),
 	};
 
@@ -2007,8 +2005,8 @@ test_blocking(uint64_t requested_oa_period, bool set_kernel_hrtimer, uint64_t ke
 		.flags = I915_PERF_FLAG_FD_CLOEXEC |
 			I915_PERF_FLAG_DISABLED,
 		.num_properties = set_kernel_hrtimer ?
-				  NUM_PROPERTIES(properties) :
-				  NUM_PROPERTIES(properties) - 1,
+				  ARRAY_SIZE(properties) / 2 :
+				  (ARRAY_SIZE(properties) / 2) - 1,
 		.properties_ptr = to_user_pointer(properties),
 	};
 	uint8_t buf[1024 * 1024];
@@ -2167,8 +2165,8 @@ test_polling(uint64_t requested_oa_period, bool set_kernel_hrtimer, uint64_t ker
 			I915_PERF_FLAG_DISABLED |
 			I915_PERF_FLAG_FD_NONBLOCK,
 		.num_properties = set_kernel_hrtimer ?
-				  NUM_PROPERTIES(properties) :
-				  NUM_PROPERTIES(properties) - 1,
+				  ARRAY_SIZE(properties) / 2 :
+				  (ARRAY_SIZE(properties) / 2) - 1,
 		.properties_ptr = to_user_pointer(properties),
 	};
 	uint8_t buf[1024 * 1024];
@@ -2358,7 +2356,7 @@ static void test_polling_small_buf(void)
 		.flags = I915_PERF_FLAG_FD_CLOEXEC |
 			I915_PERF_FLAG_DISABLED |
 			I915_PERF_FLAG_FD_NONBLOCK,
-		.num_properties = NUM_PROPERTIES(properties),
+		.num_properties = ARRAY_SIZE(properties) / 2,
 		.properties_ptr = to_user_pointer(properties),
 	};
 	uint32_t test_duration = 80 * 1000 * 1000;
@@ -2458,7 +2456,7 @@ gen12_test_oa_tlb_invalidate(void)
 	struct drm_i915_perf_open_param param = {
 		.flags = I915_PERF_FLAG_FD_CLOEXEC |
 			I915_PERF_FLAG_DISABLED,
-		.num_properties = sizeof(properties) / 16,
+		.num_properties = ARRAY_SIZE(properties) / 2,
 		.properties_ptr = to_user_pointer(properties),
 	};
 	int num_reports1, num_reports2, num_expected_reports;
@@ -2649,7 +2647,7 @@ test_non_zero_reason(void)
 	};
 	struct drm_i915_perf_open_param param = {
 		.flags = I915_PERF_FLAG_FD_CLOEXEC,
-		.num_properties = sizeof(properties) / 16,
+		.num_properties = ARRAY_SIZE(properties) / 2,
 		.properties_ptr = to_user_pointer(properties),
 	};
 	struct drm_i915_perf_record_header *header;
@@ -2734,7 +2732,7 @@ test_enable_disable(void)
 	struct drm_i915_perf_open_param param = {
 		.flags = I915_PERF_FLAG_FD_CLOEXEC |
 			 I915_PERF_FLAG_DISABLED, /* Verify we start disabled */
-		.num_properties = sizeof(properties) / 16,
+		.num_properties = ARRAY_SIZE(properties) / 2,
 		.properties_ptr = to_user_pointer(properties),
 	};
 	int buf_size = 65536 * (256 + sizeof(struct drm_i915_perf_record_header));
@@ -2883,7 +2881,7 @@ test_short_reads(void)
 	};
 	struct drm_i915_perf_open_param param = {
 		.flags = I915_PERF_FLAG_FD_CLOEXEC,
-		.num_properties = sizeof(properties) / 16,
+		.num_properties = ARRAY_SIZE(properties) / 2,
 		.properties_ptr = to_user_pointer(properties),
 	};
 	size_t record_size = 256 + sizeof(struct drm_i915_perf_record_header);
@@ -2975,7 +2973,7 @@ test_non_sampling_read_error(void)
 	};
 	struct drm_i915_perf_open_param param = {
 		.flags = I915_PERF_FLAG_FD_CLOEXEC,
-		.num_properties = sizeof(properties) / 16,
+		.num_properties = ARRAY_SIZE(properties) / 2,
 		.properties_ptr = to_user_pointer(properties),
 	};
 	int ret;
@@ -3011,7 +3009,7 @@ test_disabled_read_error(void)
 	struct drm_i915_perf_open_param param = {
 		.flags = I915_PERF_FLAG_FD_CLOEXEC |
 			 I915_PERF_FLAG_DISABLED, /* XXX: open disabled */
-		.num_properties = sizeof(properties) / 16,
+		.num_properties = ARRAY_SIZE(properties) / 2,
 		.properties_ptr = to_user_pointer(properties),
 	};
 	uint32_t oa_report0[64];
@@ -3168,7 +3166,7 @@ test_mi_rpc(void)
 	};
 	struct drm_i915_perf_open_param param = {
 		.flags = I915_PERF_FLAG_FD_CLOEXEC,
-		.num_properties = sizeof(properties) / 16,
+		.num_properties = ARRAY_SIZE(properties) / 2,
 		.properties_ptr = to_user_pointer(properties),
 	};
 	struct buf_ops *bops = buf_ops_create(drm_fd);
@@ -3271,7 +3269,7 @@ hsw_test_single_ctx_counters(void)
 	};
 	struct drm_i915_perf_open_param param = {
 		.flags = I915_PERF_FLAG_FD_CLOEXEC,
-		.num_properties = sizeof(properties) / 16,
+		.num_properties = ARRAY_SIZE(properties) / 2,
 		.properties_ptr = to_user_pointer(properties),
 	};
 
@@ -4253,7 +4251,7 @@ test_rc6_disable(void)
 	};
 	struct drm_i915_perf_open_param param = {
 		.flags = I915_PERF_FLAG_FD_CLOEXEC,
-		.num_properties = sizeof(properties) / 16,
+		.num_properties = ARRAY_SIZE(properties) / 2,
 		.properties_ptr = to_user_pointer(properties),
 	};
 	unsigned long rc6_start, rc6_end, rc6_enabled;
@@ -4309,7 +4307,7 @@ test_stress_open_close(void)
 		struct drm_i915_perf_open_param param = {
 			.flags = I915_PERF_FLAG_FD_CLOEXEC |
 			         I915_PERF_FLAG_DISABLED, /* XXX: open disabled */
-			.num_properties = NUM_PROPERTIES(properties),
+			.num_properties = ARRAY_SIZE(properties) / 2,
 			.properties_ptr = to_user_pointer(properties),
 		};
 
@@ -4405,8 +4403,8 @@ test_global_sseu_config_invalid(void)
 	struct drm_i915_perf_open_param param = {
 		.flags = I915_PERF_FLAG_FD_CLOEXEC |
 		I915_PERF_FLAG_DISABLED, /* XXX: open disabled */
-		.num_properties = NUM_PROPERTIES(properties),
-			.properties_ptr = to_user_pointer(properties),
+		.num_properties = ARRAY_SIZE(properties) / 2,
+		.properties_ptr = to_user_pointer(properties),
 	};
 
 	memset(&default_sseu, 0, sizeof(default_sseu));
@@ -4480,8 +4478,8 @@ test_global_sseu_config(void)
 	struct drm_i915_perf_open_param param = {
 		.flags = I915_PERF_FLAG_FD_CLOEXEC |
 		I915_PERF_FLAG_DISABLED, /* XXX: open disabled */
-		.num_properties = NUM_PROPERTIES(properties),
-			.properties_ptr = to_user_pointer(properties),
+		.num_properties = ARRAY_SIZE(properties) / 2,
+		.properties_ptr = to_user_pointer(properties),
 	};
 
 	memset(&default_sseu, 0, sizeof(default_sseu));
@@ -4913,7 +4911,7 @@ test_i915_ref_count(void)
 	};
 	struct drm_i915_perf_open_param param = {
 		.flags = I915_PERF_FLAG_FD_CLOEXEC,
-		.num_properties = sizeof(properties) / 16,
+		.num_properties = ARRAY_SIZE(properties) / 2,
 		.properties_ptr = to_user_pointer(properties),
 	};
 	unsigned baseline, ref_count0, ref_count1;
-- 
2.25.1

Re: [igt-dev] [PATCH i-g-t 09/23] i915/perf: Use ARRAY_SIZE consistently for num_properties

[Lionel Landwerlin]

On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
> Replace all variations of calculating the num_properties with
> ARRAY_SIZE.
>
> Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
> ---
>   tests/i915/perf.c | 62 +++++++++++++++++++++++------------------------
>   1 file changed, 30 insertions(+), 32 deletions(-)
>
> diff --git a/tests/i915/perf.c b/tests/i915/perf.c
> index b61dd6e2..c120c898 100644
> --- a/tests/i915/perf.c
> +++ b/tests/i915/perf.c
> @@ -86,8 +86,6 @@ IGT_TEST_DESCRIPTION("Test the i915 perf metrics streaming interface");
>   
>   #define MAX_OA_BUF_SIZE (16 * 1024 * 1024)
>   
> -#define NUM_PROPERTIES(p) (sizeof(p) / (2 * sizeof(uint64_t)))
> -
>   struct accumulator {
>   #define MAX_RAW_OA_COUNTERS 62
>   	enum drm_i915_oa_format format;
> @@ -1069,7 +1067,7 @@ test_system_wide_paranoid(void)
>   		struct drm_i915_perf_open_param param = {
>   			.flags = I915_PERF_FLAG_FD_CLOEXEC |
>   				I915_PERF_FLAG_FD_NONBLOCK,
> -			.num_properties = sizeof(properties) / 16,
> +			.num_properties = ARRAY_SIZE(properties) / 2,
>   			.properties_ptr = to_user_pointer(properties),
>   		};
>   
> @@ -1095,7 +1093,7 @@ test_system_wide_paranoid(void)
>   		struct drm_i915_perf_open_param param = {
>   			.flags = I915_PERF_FLAG_FD_CLOEXEC |
>   				I915_PERF_FLAG_FD_NONBLOCK,
> -			.num_properties = sizeof(properties) / 16,
> +			.num_properties = ARRAY_SIZE(properties) / 2,
>   			.properties_ptr = to_user_pointer(properties),
>   		};
>   		write_u64_file("/proc/sys/dev/i915/perf_stream_paranoid", 0);
> @@ -1126,7 +1124,7 @@ test_invalid_open_flags(void)
>   	};
>   	struct drm_i915_perf_open_param param = {
>   		.flags = ~0, /* Undefined flag bits set! */
> -		.num_properties = sizeof(properties) / 16,
> +		.num_properties = ARRAY_SIZE(properties) / 2,
>   		.properties_ptr = to_user_pointer(properties),
>   	};
>   
> @@ -1148,7 +1146,7 @@ test_invalid_oa_metric_set_id(void)
>   	struct drm_i915_perf_open_param param = {
>   		.flags = I915_PERF_FLAG_FD_CLOEXEC |
>   			I915_PERF_FLAG_FD_NONBLOCK,
> -		.num_properties = sizeof(properties) / 16,
> +		.num_properties = ARRAY_SIZE(properties) / 2,
>   		.properties_ptr = to_user_pointer(properties),
>   	};
>   
> @@ -1182,7 +1180,7 @@ test_invalid_oa_format_id(void)
>   	struct drm_i915_perf_open_param param = {
>   		.flags = I915_PERF_FLAG_FD_CLOEXEC |
>   			I915_PERF_FLAG_FD_NONBLOCK,
> -		.num_properties = sizeof(properties) / 16,
> +		.num_properties = ARRAY_SIZE(properties) / 2,
>   		.properties_ptr = to_user_pointer(properties),
>   	};
>   
> @@ -1214,7 +1212,7 @@ test_missing_sample_flags(void)
>   	};
>   	struct drm_i915_perf_open_param param = {
>   		.flags = I915_PERF_FLAG_FD_CLOEXEC,
> -		.num_properties = sizeof(properties) / 16,
> +		.num_properties = ARRAY_SIZE(properties) / 2,
>   		.properties_ptr = to_user_pointer(properties),
>   	};
>   
> @@ -1354,7 +1352,7 @@ open_and_read_2_oa_reports(int format_id,
>   	};
>   	struct drm_i915_perf_open_param param = {
>   		.flags = I915_PERF_FLAG_FD_CLOEXEC,
> -		.num_properties = sizeof(properties) / 16,
> +		.num_properties = ARRAY_SIZE(properties) / 2,
>   		.properties_ptr = to_user_pointer(properties),
>   	};
>   
> @@ -1833,7 +1831,7 @@ test_invalid_oa_exponent(void)
>   	};
>   	struct drm_i915_perf_open_param param = {
>   		.flags = I915_PERF_FLAG_FD_CLOEXEC,
> -		.num_properties = sizeof(properties) / 16,
> +		.num_properties = ARRAY_SIZE(properties) / 2,
>   		.properties_ptr = to_user_pointer(properties),
>   	};
>   
> @@ -1868,7 +1866,7 @@ test_low_oa_exponent_permissions(void)
>   	};
>   	struct drm_i915_perf_open_param param = {
>   		.flags = I915_PERF_FLAG_FD_CLOEXEC,
> -		.num_properties = sizeof(properties) / 16,
> +		.num_properties = ARRAY_SIZE(properties) / 2,
>   		.properties_ptr = to_user_pointer(properties),
>   	};
>   	uint64_t oa_period, oa_freq;
> @@ -1931,7 +1929,7 @@ test_per_context_mode_unprivileged(void)
>   	};
>   	struct drm_i915_perf_open_param param = {
>   		.flags = I915_PERF_FLAG_FD_CLOEXEC,
> -		.num_properties = sizeof(properties) / 16,
> +		.num_properties = ARRAY_SIZE(properties) / 2,
>   		.properties_ptr = to_user_pointer(properties),
>   	};
>   
> @@ -2007,8 +2005,8 @@ test_blocking(uint64_t requested_oa_period, bool set_kernel_hrtimer, uint64_t ke
>   		.flags = I915_PERF_FLAG_FD_CLOEXEC |
>   			I915_PERF_FLAG_DISABLED,
>   		.num_properties = set_kernel_hrtimer ?
> -				  NUM_PROPERTIES(properties) :
> -				  NUM_PROPERTIES(properties) - 1,
> +				  ARRAY_SIZE(properties) / 2 :
> +				  (ARRAY_SIZE(properties) / 2) - 1,
>   		.properties_ptr = to_user_pointer(properties),
>   	};
>   	uint8_t buf[1024 * 1024];
> @@ -2167,8 +2165,8 @@ test_polling(uint64_t requested_oa_period, bool set_kernel_hrtimer, uint64_t ker
>   			I915_PERF_FLAG_DISABLED |
>   			I915_PERF_FLAG_FD_NONBLOCK,
>   		.num_properties = set_kernel_hrtimer ?
> -				  NUM_PROPERTIES(properties) :
> -				  NUM_PROPERTIES(properties) - 1,
> +				  ARRAY_SIZE(properties) / 2 :
> +				  (ARRAY_SIZE(properties) / 2) - 1,
>   		.properties_ptr = to_user_pointer(properties),
>   	};
>   	uint8_t buf[1024 * 1024];
> @@ -2358,7 +2356,7 @@ static void test_polling_small_buf(void)
>   		.flags = I915_PERF_FLAG_FD_CLOEXEC |
>   			I915_PERF_FLAG_DISABLED |
>   			I915_PERF_FLAG_FD_NONBLOCK,
> -		.num_properties = NUM_PROPERTIES(properties),
> +		.num_properties = ARRAY_SIZE(properties) / 2,
>   		.properties_ptr = to_user_pointer(properties),
>   	};
>   	uint32_t test_duration = 80 * 1000 * 1000;
> @@ -2458,7 +2456,7 @@ gen12_test_oa_tlb_invalidate(void)
>   	struct drm_i915_perf_open_param param = {
>   		.flags = I915_PERF_FLAG_FD_CLOEXEC |
>   			I915_PERF_FLAG_DISABLED,
> -		.num_properties = sizeof(properties) / 16,
> +		.num_properties = ARRAY_SIZE(properties) / 2,
>   		.properties_ptr = to_user_pointer(properties),
>   	};
>   	int num_reports1, num_reports2, num_expected_reports;
> @@ -2649,7 +2647,7 @@ test_non_zero_reason(void)
>   	};
>   	struct drm_i915_perf_open_param param = {
>   		.flags = I915_PERF_FLAG_FD_CLOEXEC,
> -		.num_properties = sizeof(properties) / 16,
> +		.num_properties = ARRAY_SIZE(properties) / 2,
>   		.properties_ptr = to_user_pointer(properties),
>   	};
>   	struct drm_i915_perf_record_header *header;
> @@ -2734,7 +2732,7 @@ test_enable_disable(void)
>   	struct drm_i915_perf_open_param param = {
>   		.flags = I915_PERF_FLAG_FD_CLOEXEC |
>   			 I915_PERF_FLAG_DISABLED, /* Verify we start disabled */
> -		.num_properties = sizeof(properties) / 16,
> +		.num_properties = ARRAY_SIZE(properties) / 2,
>   		.properties_ptr = to_user_pointer(properties),
>   	};
>   	int buf_size = 65536 * (256 + sizeof(struct drm_i915_perf_record_header));
> @@ -2883,7 +2881,7 @@ test_short_reads(void)
>   	};
>   	struct drm_i915_perf_open_param param = {
>   		.flags = I915_PERF_FLAG_FD_CLOEXEC,
> -		.num_properties = sizeof(properties) / 16,
> +		.num_properties = ARRAY_SIZE(properties) / 2,
>   		.properties_ptr = to_user_pointer(properties),
>   	};
>   	size_t record_size = 256 + sizeof(struct drm_i915_perf_record_header);
> @@ -2975,7 +2973,7 @@ test_non_sampling_read_error(void)
>   	};
>   	struct drm_i915_perf_open_param param = {
>   		.flags = I915_PERF_FLAG_FD_CLOEXEC,
> -		.num_properties = sizeof(properties) / 16,
> +		.num_properties = ARRAY_SIZE(properties) / 2,
>   		.properties_ptr = to_user_pointer(properties),
>   	};
>   	int ret;
> @@ -3011,7 +3009,7 @@ test_disabled_read_error(void)
>   	struct drm_i915_perf_open_param param = {
>   		.flags = I915_PERF_FLAG_FD_CLOEXEC |
>   			 I915_PERF_FLAG_DISABLED, /* XXX: open disabled */
> -		.num_properties = sizeof(properties) / 16,
> +		.num_properties = ARRAY_SIZE(properties) / 2,
>   		.properties_ptr = to_user_pointer(properties),
>   	};
>   	uint32_t oa_report0[64];
> @@ -3168,7 +3166,7 @@ test_mi_rpc(void)
>   	};
>   	struct drm_i915_perf_open_param param = {
>   		.flags = I915_PERF_FLAG_FD_CLOEXEC,
> -		.num_properties = sizeof(properties) / 16,
> +		.num_properties = ARRAY_SIZE(properties) / 2,
>   		.properties_ptr = to_user_pointer(properties),
>   	};
>   	struct buf_ops *bops = buf_ops_create(drm_fd);
> @@ -3271,7 +3269,7 @@ hsw_test_single_ctx_counters(void)
>   	};
>   	struct drm_i915_perf_open_param param = {
>   		.flags = I915_PERF_FLAG_FD_CLOEXEC,
> -		.num_properties = sizeof(properties) / 16,
> +		.num_properties = ARRAY_SIZE(properties) / 2,
>   		.properties_ptr = to_user_pointer(properties),
>   	};
>   
> @@ -4253,7 +4251,7 @@ test_rc6_disable(void)
>   	};
>   	struct drm_i915_perf_open_param param = {
>   		.flags = I915_PERF_FLAG_FD_CLOEXEC,
> -		.num_properties = sizeof(properties) / 16,
> +		.num_properties = ARRAY_SIZE(properties) / 2,
>   		.properties_ptr = to_user_pointer(properties),
>   	};
>   	unsigned long rc6_start, rc6_end, rc6_enabled;
> @@ -4309,7 +4307,7 @@ test_stress_open_close(void)
>   		struct drm_i915_perf_open_param param = {
>   			.flags = I915_PERF_FLAG_FD_CLOEXEC |
>   			         I915_PERF_FLAG_DISABLED, /* XXX: open disabled */
> -			.num_properties = NUM_PROPERTIES(properties),
> +			.num_properties = ARRAY_SIZE(properties) / 2,
>   			.properties_ptr = to_user_pointer(properties),
>   		};
>   
> @@ -4405,8 +4403,8 @@ test_global_sseu_config_invalid(void)
>   	struct drm_i915_perf_open_param param = {
>   		.flags = I915_PERF_FLAG_FD_CLOEXEC |
>   		I915_PERF_FLAG_DISABLED, /* XXX: open disabled */
> -		.num_properties = NUM_PROPERTIES(properties),
> -			.properties_ptr = to_user_pointer(properties),
> +		.num_properties = ARRAY_SIZE(properties) / 2,
> +		.properties_ptr = to_user_pointer(properties),
>   	};
>   
>   	memset(&default_sseu, 0, sizeof(default_sseu));
> @@ -4480,8 +4478,8 @@ test_global_sseu_config(void)
>   	struct drm_i915_perf_open_param param = {
>   		.flags = I915_PERF_FLAG_FD_CLOEXEC |
>   		I915_PERF_FLAG_DISABLED, /* XXX: open disabled */
> -		.num_properties = NUM_PROPERTIES(properties),
> -			.properties_ptr = to_user_pointer(properties),
> +		.num_properties = ARRAY_SIZE(properties) / 2,
> +		.properties_ptr = to_user_pointer(properties),
>   	};
>   
>   	memset(&default_sseu, 0, sizeof(default_sseu));
> @@ -4913,7 +4911,7 @@ test_i915_ref_count(void)
>   	};
>   	struct drm_i915_perf_open_param param = {
>   		.flags = I915_PERF_FLAG_FD_CLOEXEC,
> -		.num_properties = sizeof(properties) / 16,
> +		.num_properties = ARRAY_SIZE(properties) / 2,
>   		.properties_ptr = to_user_pointer(properties),
>   	};
>   	unsigned baseline, ref_count0, ref_count1;

[igt-dev] [PATCH i-g-t 10/23] i915/perf: Use gt in perf tests and lib

[Umesh Nerlige Ramappa]

Enable perf tests to use gt information for sysfs attributes.

Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
---
 lib/i915/perf.c   | 36 ++++++++++++++++++++++++++++++++++--
 tests/i915/perf.c | 27 ++++++++++++++++++++-------
 2 files changed, 54 insertions(+), 9 deletions(-)

diff --git a/lib/i915/perf.c b/lib/i915/perf.c
index 7349a460..3f131aaa 100644
--- a/lib/i915/perf.c
+++ b/lib/i915/perf.c
@@ -409,6 +409,35 @@ open_master_sysfs_dir(int drm_fd)
 	return sysfs;
 }
 
+typedef enum {
+	RPS_MIN_FREQ_MHZ,
+	RPS_MAX_FREQ_MHZ,
+
+	RPS_MAX_ATTR,
+} intel_sysfs_attr_id;
+
+static const char *intel_sysfs_attr_name[2][RPS_MAX_ATTR] =
+{
+	{
+		"gt_min_freq_mhz",
+		"gt_max_freq_mhz",
+	},
+	{
+		"gt/gt0/rps_min_freq_mhz",
+		"gt/gt0/rps_max_freq_mhz",
+	},
+};
+
+static const char *
+intel_sysfs_attr_id_to_name(int sysfs_dirfd, intel_sysfs_attr_id id)
+{
+	assert(id < RPS_MAX_ATTR);
+
+	return !faccessat(sysfs_dirfd, "gt", O_RDONLY, 0) ?
+		intel_sysfs_attr_name[1][id] :
+		intel_sysfs_attr_name[0][id];
+}
+
 struct intel_perf *
 intel_perf_for_fd(int drm_fd)
 {
@@ -424,8 +453,11 @@ intel_perf_for_fd(int drm_fd)
 	if (sysfs_dir_fd < 0)
 		return NULL;
 
-	if (!read_sysfs(sysfs_dir_fd, "gt_min_freq_mhz", &gt_min_freq) ||
-	    !read_sysfs(sysfs_dir_fd, "gt_max_freq_mhz", &gt_max_freq)) {
+#define read_sysfs_rps(fd, id, value) \
+	read_sysfs(fd, intel_sysfs_attr_id_to_name(fd, id), value)
+
+	if (!read_sysfs_rps(sysfs_dir_fd, RPS_MIN_FREQ_MHZ, &gt_min_freq) ||
+	    !read_sysfs_rps(sysfs_dir_fd, RPS_MAX_FREQ_MHZ, &gt_max_freq)) {
 		close(sysfs_dir_fd);
 		return NULL;
 	}
diff --git a/tests/i915/perf.c b/tests/i915/perf.c
index c120c898..89aa4931 100644
--- a/tests/i915/perf.c
+++ b/tests/i915/perf.c
@@ -347,11 +347,11 @@ try_sysfs_read_u64(const char *path, uint64_t *val)
 }
 
 static unsigned long
-sysfs_read(const char *path)
+sysfs_read(enum i915_attr_id id)
 {
 	unsigned long value;
 
-	igt_assert(igt_sysfs_scanf(sysfs, path, "%lu", &value) == 1);
+	igt_assert(igt_sysfs_rps_scanf(sysfs, id, "%lu", &value) == 1);
 
 	return value;
 }
@@ -4234,7 +4234,7 @@ gen12_test_single_ctx_render_target_writes_a_counter(void)
 
 static unsigned long rc6_residency_ms(void)
 {
-	return sysfs_read("power/rc6_residency_ms");
+	return sysfs_read(RC6_RESIDENCY_MS);
 }
 
 static void
@@ -4257,7 +4257,7 @@ test_rc6_disable(void)
 	unsigned long rc6_start, rc6_end, rc6_enabled;
 
 	rc6_enabled = 0;
-	igt_sysfs_scanf(sysfs, "power/rc6_enable", "%lu", &rc6_enabled);
+	igt_sysfs_rps_scanf(sysfs, RC6_ENABLE, "%lu", &rc6_enabled);
 	igt_require(rc6_enabled);
 
 	/* Verify rc6 is functional by measuring residency while idle */
@@ -4894,6 +4894,19 @@ done:
 	return ref_count;
 }
 
+static int perf_sysfs_open(int i915)
+{
+	int dirfd, gt;
+
+	/* use the first available sysfs interface */
+	for_each_sysfs_gt_dirfd(i915, dirfd, gt)
+		break;
+
+	igt_assert(dirfd != -1);
+
+	return dirfd;
+}
+
 /* check that an open i915 perf stream holds a reference on the drm i915 module
  * including in the corner case where the original drm fd has been closed.
  */
@@ -4929,7 +4942,7 @@ test_i915_ref_count(void)
 	drm_fd = __drm_open_driver(DRIVER_INTEL);
 	igt_require_intel(drm_fd);
 	devid = intel_get_drm_devid(drm_fd);
-	sysfs = igt_sysfs_open(drm_fd);
+	sysfs = perf_sysfs_open(drm_fd);
 
 	/* Note: these global variables are only initialized after calling
 	 * init_sys_info()...
@@ -5031,14 +5044,14 @@ igt_main
 		igt_require_gem(drm_fd);
 
 		devid = intel_get_drm_devid(drm_fd);
-		sysfs = igt_sysfs_open(drm_fd);
+		sysfs = perf_sysfs_open(drm_fd);
 
 		igt_require(init_sys_info());
 
 		write_u64_file("/proc/sys/dev/i915/perf_stream_paranoid", 1);
 		write_u64_file("/proc/sys/dev/i915/oa_max_sample_rate", 100000);
 
-		gt_max_freq_mhz = sysfs_read("gt_boost_freq_mhz");
+		gt_max_freq_mhz = sysfs_read(RPS_RP0_FREQ_MHZ);
 
 		render_copy = igt_get_render_copyfunc(devid);
 		igt_require_f(render_copy, "no render-copy function\n");
-- 
2.25.1

Re: [igt-dev] [PATCH i-g-t 10/23] i915/perf: Use gt in perf tests and lib

[Lionel Landwerlin]

On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
> Enable perf tests to use gt information for sysfs attributes.
>
> Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
> ---
>   lib/i915/perf.c   | 36 ++++++++++++++++++++++++++++++++++--
>   tests/i915/perf.c | 27 ++++++++++++++++++++-------
>   2 files changed, 54 insertions(+), 9 deletions(-)
>
> diff --git a/lib/i915/perf.c b/lib/i915/perf.c
> index 7349a460..3f131aaa 100644
> --- a/lib/i915/perf.c
> +++ b/lib/i915/perf.c
> @@ -409,6 +409,35 @@ open_master_sysfs_dir(int drm_fd)
>   	return sysfs;
>   }
>   
> +typedef enum {
> +	RPS_MIN_FREQ_MHZ,
> +	RPS_MAX_FREQ_MHZ,
> +
> +	RPS_MAX_ATTR,
> +} intel_sysfs_attr_id;
> +
> +static const char *intel_sysfs_attr_name[2][RPS_MAX_ATTR] =
> +{
> +	{
> +		"gt_min_freq_mhz",
> +		"gt_max_freq_mhz",
> +	},
> +	{
> +		"gt/gt0/rps_min_freq_mhz",
> +		"gt/gt0/rps_max_freq_mhz",
> +	},
> +};
> +
> +static const char *
> +intel_sysfs_attr_id_to_name(int sysfs_dirfd, intel_sysfs_attr_id id)
> +{
> +	assert(id < RPS_MAX_ATTR);
> +
> +	return !faccessat(sysfs_dirfd, "gt", O_RDONLY, 0) ?
> +		intel_sysfs_attr_name[1][id] :
> +		intel_sysfs_attr_name[0][id];
> +}
> +
>   struct intel_perf *
>   intel_perf_for_fd(int drm_fd)
>   {
> @@ -424,8 +453,11 @@ intel_perf_for_fd(int drm_fd)
>   	if (sysfs_dir_fd < 0)
>   		return NULL;
>   
> -	if (!read_sysfs(sysfs_dir_fd, "gt_min_freq_mhz", &gt_min_freq) ||
> -	    !read_sysfs(sysfs_dir_fd, "gt_max_freq_mhz", &gt_max_freq)) {
> +#define read_sysfs_rps(fd, id, value) \
> +	read_sysfs(fd, intel_sysfs_attr_id_to_name(fd, id), value)
> +
> +	if (!read_sysfs_rps(sysfs_dir_fd, RPS_MIN_FREQ_MHZ, &gt_min_freq) ||
> +	    !read_sysfs_rps(sysfs_dir_fd, RPS_MAX_FREQ_MHZ, &gt_max_freq)) {
>   		close(sysfs_dir_fd);
>   		return NULL;
>   	}
> diff --git a/tests/i915/perf.c b/tests/i915/perf.c
> index c120c898..89aa4931 100644
> --- a/tests/i915/perf.c
> +++ b/tests/i915/perf.c
> @@ -347,11 +347,11 @@ try_sysfs_read_u64(const char *path, uint64_t *val)
>   }
>   
>   static unsigned long
> -sysfs_read(const char *path)
> +sysfs_read(enum i915_attr_id id)
>   {
>   	unsigned long value;
>   
> -	igt_assert(igt_sysfs_scanf(sysfs, path, "%lu", &value) == 1);
> +	igt_assert(igt_sysfs_rps_scanf(sysfs, id, "%lu", &value) == 1);
>   
>   	return value;
>   }
> @@ -4234,7 +4234,7 @@ gen12_test_single_ctx_render_target_writes_a_counter(void)
>   
>   static unsigned long rc6_residency_ms(void)
>   {
> -	return sysfs_read("power/rc6_residency_ms");
> +	return sysfs_read(RC6_RESIDENCY_MS);
>   }
>   
>   static void
> @@ -4257,7 +4257,7 @@ test_rc6_disable(void)
>   	unsigned long rc6_start, rc6_end, rc6_enabled;
>   
>   	rc6_enabled = 0;
> -	igt_sysfs_scanf(sysfs, "power/rc6_enable", "%lu", &rc6_enabled);
> +	igt_sysfs_rps_scanf(sysfs, RC6_ENABLE, "%lu", &rc6_enabled);
>   	igt_require(rc6_enabled);
>   
>   	/* Verify rc6 is functional by measuring residency while idle */
> @@ -4894,6 +4894,19 @@ done:
>   	return ref_count;
>   }
>   
> +static int perf_sysfs_open(int i915)
> +{
> +	int dirfd, gt;
> +
> +	/* use the first available sysfs interface */
> +	for_each_sysfs_gt_dirfd(i915, dirfd, gt)
> +		break;
> +
> +	igt_assert(dirfd != -1);
> +
> +	return dirfd;
> +}
> +
>   /* check that an open i915 perf stream holds a reference on the drm i915 module
>    * including in the corner case where the original drm fd has been closed.
>    */
> @@ -4929,7 +4942,7 @@ test_i915_ref_count(void)
>   	drm_fd = __drm_open_driver(DRIVER_INTEL);
>   	igt_require_intel(drm_fd);
>   	devid = intel_get_drm_devid(drm_fd);
> -	sysfs = igt_sysfs_open(drm_fd);
> +	sysfs = perf_sysfs_open(drm_fd);
>   
>   	/* Note: these global variables are only initialized after calling
>   	 * init_sys_info()...
> @@ -5031,14 +5044,14 @@ igt_main
>   		igt_require_gem(drm_fd);
>   
>   		devid = intel_get_drm_devid(drm_fd);
> -		sysfs = igt_sysfs_open(drm_fd);
> +		sysfs = perf_sysfs_open(drm_fd);
>   
>   		igt_require(init_sys_info());
>   
>   		write_u64_file("/proc/sys/dev/i915/perf_stream_paranoid", 1);
>   		write_u64_file("/proc/sys/dev/i915/oa_max_sample_rate", 100000);
>   
> -		gt_max_freq_mhz = sysfs_read("gt_boost_freq_mhz");
> +		gt_max_freq_mhz = sysfs_read(RPS_RP0_FREQ_MHZ);
>   
>   		render_copy = igt_get_render_copyfunc(devid);
>   		igt_require_f(render_copy, "no render-copy function\n");

[igt-dev] [PATCH i-g-t 11/23] i915/perf: Explicitly state rendercopy needs for a test

[Umesh Nerlige Ramappa]

Let tests explicitly require rendercopy before running. This helps in
temporarily blocking these tests until rendercopy is available on new
platforms.

Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
---
 tests/i915/perf.c | 25 ++++++++++++++++++++++---
 1 file changed, 22 insertions(+), 3 deletions(-)

diff --git a/tests/i915/perf.c b/tests/i915/perf.c
index 89aa4931..ba8c3792 100644
--- a/tests/i915/perf.c
+++ b/tests/i915/perf.c
@@ -1601,6 +1601,9 @@ static void load_helper_set_load(enum load load)
 
 static void load_helper_run(enum load load)
 {
+	if (!render_copy)
+		return;
+
 	/*
 	 * FIXME fork helpers won't get cleaned up when started from within a
 	 * subtest, so handle the case where it sticks around a bit too long.
@@ -1633,12 +1636,20 @@ static void load_helper_run(enum load load)
 
 static void load_helper_stop(void)
 {
+	if (!render_copy)
+		return;
+
 	kill(lh.igt_proc.pid, SIGUSR1);
 	igt_assert(igt_wait_helper(&lh.igt_proc) == 0);
 }
 
 static void load_helper_init(void)
 {
+	if (!render_copy) {
+		igt_info("Running test without render_copy\n");
+		return;
+	}
+
 	lh.devid = intel_get_drm_devid(drm_fd);
 
 	/* MI_STORE_DATA can only use GTT address on gen4+/g33 and needs
@@ -1659,7 +1670,12 @@ static void load_helper_init(void)
 
 static void load_helper_fini(void)
 {
-	int i915 = buf_ops_get_fd(lh.bops);
+	int i915;
+
+	if (!render_copy)
+		return;
+
+	i915 = buf_ops_get_fd(lh.bops);
 
 	if (lh.igt_proc.running)
 		load_helper_stop();
@@ -5054,7 +5070,6 @@ igt_main
 		gt_max_freq_mhz = sysfs_read(RPS_RP0_FREQ_MHZ);
 
 		render_copy = igt_get_render_copyfunc(devid);
-		igt_require_f(render_copy, "no render-copy function\n");
 	}
 
 	igt_subtest("non-system-wide-paranoid")
@@ -5157,6 +5172,7 @@ igt_main
 
 	igt_subtest("unprivileged-single-ctx-counters") {
 		igt_require(IS_HASWELL(devid));
+		igt_require_f(render_copy, "no render-copy function\n");
 		hsw_test_single_ctx_counters();
 	}
 
@@ -5170,6 +5186,7 @@ igt_main
 		 * For gen12 implement a separate test that uses only OAR
 		 */
 		igt_require(intel_gen(devid) >= 8 && intel_gen(devid) < 12);
+		igt_require_f(render_copy, "no render-copy function\n");
 		gen8_test_single_ctx_render_target_writes_a_counter();
 	}
 
@@ -5185,8 +5202,10 @@ igt_main
 			gen12_test_oa_tlb_invalidate();
 
 		igt_describe("Measure performance for a specific context using OAR in Gen 12");
-		igt_subtest("gen12-unprivileged-single-ctx-counters")
+		igt_subtest("gen12-unprivileged-single-ctx-counters") {
+			igt_require_f(render_copy, "no render-copy function\n");
 			gen12_test_single_ctx_render_target_writes_a_counter();
+		}
 	}
 
 	igt_subtest("rc6-disable")
-- 
2.25.1

Re: [igt-dev] [PATCH i-g-t 11/23] i915/perf: Explicitly state rendercopy needs for a test

[Lionel Landwerlin]

On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
> Let tests explicitly require rendercopy before running. This helps in
> temporarily blocking these tests until rendercopy is available on new
> platforms.
>
> Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
> ---
>   tests/i915/perf.c | 25 ++++++++++++++++++++++---
>   1 file changed, 22 insertions(+), 3 deletions(-)
>
> diff --git a/tests/i915/perf.c b/tests/i915/perf.c
> index 89aa4931..ba8c3792 100644
> --- a/tests/i915/perf.c
> +++ b/tests/i915/perf.c
> @@ -1601,6 +1601,9 @@ static void load_helper_set_load(enum load load)
>   
>   static void load_helper_run(enum load load)
>   {
> +	if (!render_copy)
> +		return;
> +
>   	/*
>   	 * FIXME fork helpers won't get cleaned up when started from within a
>   	 * subtest, so handle the case where it sticks around a bit too long.
> @@ -1633,12 +1636,20 @@ static void load_helper_run(enum load load)
>   
>   static void load_helper_stop(void)
>   {
> +	if (!render_copy)
> +		return;
> +
>   	kill(lh.igt_proc.pid, SIGUSR1);
>   	igt_assert(igt_wait_helper(&lh.igt_proc) == 0);
>   }
>   
>   static void load_helper_init(void)
>   {
> +	if (!render_copy) {
> +		igt_info("Running test without render_copy\n");
> +		return;
> +	}
> +
>   	lh.devid = intel_get_drm_devid(drm_fd);
>   
>   	/* MI_STORE_DATA can only use GTT address on gen4+/g33 and needs
> @@ -1659,7 +1670,12 @@ static void load_helper_init(void)
>   
>   static void load_helper_fini(void)
>   {
> -	int i915 = buf_ops_get_fd(lh.bops);
> +	int i915;
> +
> +	if (!render_copy)
> +		return;
> +
> +	i915 = buf_ops_get_fd(lh.bops);
>   
>   	if (lh.igt_proc.running)
>   		load_helper_stop();
> @@ -5054,7 +5070,6 @@ igt_main
>   		gt_max_freq_mhz = sysfs_read(RPS_RP0_FREQ_MHZ);
>   
>   		render_copy = igt_get_render_copyfunc(devid);
> -		igt_require_f(render_copy, "no render-copy function\n");
>   	}
>   
>   	igt_subtest("non-system-wide-paranoid")
> @@ -5157,6 +5172,7 @@ igt_main
>   
>   	igt_subtest("unprivileged-single-ctx-counters") {
>   		igt_require(IS_HASWELL(devid));
> +		igt_require_f(render_copy, "no render-copy function\n");
>   		hsw_test_single_ctx_counters();
>   	}
>   
> @@ -5170,6 +5186,7 @@ igt_main
>   		 * For gen12 implement a separate test that uses only OAR
>   		 */
>   		igt_require(intel_gen(devid) >= 8 && intel_gen(devid) < 12);
> +		igt_require_f(render_copy, "no render-copy function\n");
>   		gen8_test_single_ctx_render_target_writes_a_counter();
>   	}
>   
> @@ -5185,8 +5202,10 @@ igt_main
>   			gen12_test_oa_tlb_invalidate();
>   
>   		igt_describe("Measure performance for a specific context using OAR in Gen 12");
> -		igt_subtest("gen12-unprivileged-single-ctx-counters")
> +		igt_subtest("gen12-unprivileged-single-ctx-counters") {
> +			igt_require_f(render_copy, "no render-copy function\n");
>   			gen12_test_single_ctx_render_target_writes_a_counter();
> +		}
>   	}
>   
>   	igt_subtest("rc6-disable")

[igt-dev] [PATCH i-g-t 12/23] i915/perf: Skip tests that use rendercopy

[Umesh Nerlige Ramappa]

Running tests that require rendercopy sometimes leaves the DUT in an
unsusable state. Further tests fail to run and DUT requires reboot.
Since it affects CI efficiency, skip tests is rendercopy is missing.

Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
---
 tests/i915/perf.c | 17 ++++++++++++++++-
 1 file changed, 16 insertions(+), 1 deletion(-)

diff --git a/tests/i915/perf.c b/tests/i915/perf.c
index ba8c3792..85f3c5de 100644
--- a/tests/i915/perf.c
+++ b/tests/i915/perf.c
@@ -41,6 +41,7 @@
 #include "i915/gem.h"
 #include "i915/perf.h"
 #include "igt.h"
+#include "igt_perf.h"
 #include "igt_sysfs.h"
 #include "drm.h"
 
@@ -5025,6 +5026,19 @@ static int i915_perf_revision(int fd)
 	return value;
 }
 
+static bool has_class_instance(int i915, uint16_t class, uint16_t instance)
+{
+	int fd;
+
+	fd = perf_i915_open(i915, I915_PMU_ENGINE_BUSY(class, instance));
+	if (fd >= 0) {
+		close(fd);
+		return true;
+	}
+
+	return false;
+}
+
 igt_main
 {
 	igt_fixture {
@@ -5069,7 +5083,8 @@ igt_main
 
 		gt_max_freq_mhz = sysfs_read(RPS_RP0_FREQ_MHZ);
 
-		render_copy = igt_get_render_copyfunc(devid);
+		if (has_class_instance(drm_fd, I915_ENGINE_CLASS_RENDER, 0))
+			render_copy = igt_get_render_copyfunc(devid);
 	}
 
 	igt_subtest("non-system-wide-paranoid")
-- 
2.25.1

Re: [igt-dev] [PATCH i-g-t 12/23] i915/perf: Skip tests that use rendercopy

[Lionel Landwerlin]

On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
> Running tests that require rendercopy sometimes leaves the DUT in an
> unsusable state. Further tests fail to run and DUT requires reboot.
> Since it affects CI efficiency, skip tests is rendercopy is missing.
>
> Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>

Kind of awkward that igt_get_render_copyfunc() returns something when it 
doesn't seem to be have an engine able to run the copy function...


Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>

> ---
>   tests/i915/perf.c | 17 ++++++++++++++++-
>   1 file changed, 16 insertions(+), 1 deletion(-)
>
> diff --git a/tests/i915/perf.c b/tests/i915/perf.c
> index ba8c3792..85f3c5de 100644
> --- a/tests/i915/perf.c
> +++ b/tests/i915/perf.c
> @@ -41,6 +41,7 @@
>   #include "i915/gem.h"
>   #include "i915/perf.h"
>   #include "igt.h"
> +#include "igt_perf.h"
>   #include "igt_sysfs.h"
>   #include "drm.h"
>   
> @@ -5025,6 +5026,19 @@ static int i915_perf_revision(int fd)
>   	return value;
>   }
>   
> +static bool has_class_instance(int i915, uint16_t class, uint16_t instance)
> +{
> +	int fd;
> +
> +	fd = perf_i915_open(i915, I915_PMU_ENGINE_BUSY(class, instance));
> +	if (fd >= 0) {
> +		close(fd);
> +		return true;
> +	}
> +
> +	return false;
> +}
> +
>   igt_main
>   {
>   	igt_fixture {
> @@ -5069,7 +5083,8 @@ igt_main
>   
>   		gt_max_freq_mhz = sysfs_read(RPS_RP0_FREQ_MHZ);
>   
> -		render_copy = igt_get_render_copyfunc(devid);
> +		if (has_class_instance(drm_fd, I915_ENGINE_CLASS_RENDER, 0))
> +			render_copy = igt_get_render_copyfunc(devid);
>   	}
>   
>   	igt_subtest("non-system-wide-paranoid")

[igt-dev] [PATCH i-g-t 13/23] i915/perf: Add OA formats for DG2

[Umesh Nerlige Ramappa]

Add OA formats for DG2.

Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
---
 include/drm-uapi/i915_drm.h |  6 ++++++
 tests/i915/perf.c           | 38 ++++++++++++++++++++++++++++++++++---
 2 files changed, 41 insertions(+), 3 deletions(-)

diff --git a/include/drm-uapi/i915_drm.h b/include/drm-uapi/i915_drm.h
index b4efc96c..c7587a0f 100644
--- a/include/drm-uapi/i915_drm.h
+++ b/include/drm-uapi/i915_drm.h
@@ -2535,6 +2535,12 @@ enum drm_i915_oa_format {
 	I915_OA_FORMAT_A12_B8_C8,
 	I915_OA_FORMAT_A32u40_A4u32_B8_C8,
 
+	/* DG2 */
+	I915_OAR_FORMAT_A32u40_A4u32_B8_C8,
+	I915_OA_FORMAT_A24u40_A14u32_B8_C8,
+	I915_OAR_FORMAT_A36u64_B8_C8,
+	I915_OA_FORMAT_A38u64_R2u64_B8_C8,
+
 	I915_OA_FORMAT_MAX	    /* non-ABI */
 };
 
diff --git a/tests/i915/perf.c b/tests/i915/perf.c
index 85f3c5de..d601eb36 100644
--- a/tests/i915/perf.c
+++ b/tests/i915/perf.c
@@ -181,6 +181,36 @@ static struct oa_format gen12_oa_formats[I915_OA_FORMAT_MAX] = {
 		.c_off = 224, .n_c = 8, },
 };
 
+static struct oa_format dg2_oa_formats[I915_OA_FORMAT_MAX] = {
+	[I915_OAR_FORMAT_A32u40_A4u32_B8_C8] = {
+		"A32u40_A4u32_B8_C8", .size = 256,
+		.a40_high_off = 160, .a40_low_off = 16, .n_a40 = 32,
+		.a_off = 144, .n_a = 4, .first_a = 32,
+		.b_off = 192, .n_b = 8,
+		.c_off = 224, .n_c = 8, .oa_type = OAR, },
+	/* This format has A36 and A37 interleaved with high bytes of some A
+	 * counters, so we will accumulate only subset of counters.
+	 */
+	[I915_OA_FORMAT_A24u40_A14u32_B8_C8] = {
+		"A24u40_A14u32_B8_C8", .size = 256,
+		/* u40: A4 - A23 */
+		.a40_high_off = 164, .a40_low_off = 32, .n_a40 = 20,
+		/* u32: A24 - A27 */
+		.a_off = 144, .n_a = 4, .first_a = 24,
+		.b_off = 192, .n_b = 8,
+		.c_off = 224, .n_c = 8, .oa_type = OAG, },
+	[I915_OA_FORMAT_A38u64_R2u64_B8_C8] = {
+		"A38u64_R2u64_B8_C8", .size = 448,
+		.a64_off = 32, .n_a64 = 38,
+		.b_off = 352, .n_b = 8,
+		.c_off = 384, .n_c = 8, .oa_type = OAG, },
+	[I915_OAR_FORMAT_A36u64_B8_C8] = {
+		"A36u64_B8_C8", .size = 384,
+		.a64_off = 32, .n_a64 = 36,
+		.b_off = 320, .n_b = 8,
+		.c_off = 352, .n_c = 8, .oa_type = OAR, },
+};
+
 static bool hsw_undefined_a_counters[45] = {
 	[4] = true,
 	[6] = true,
@@ -239,6 +269,8 @@ get_oa_format(enum drm_i915_oa_format format)
 {
 	if (IS_HASWELL(devid))
 		return hsw_oa_formats[format];
+	else if (IS_DG2(devid))
+		return dg2_oa_formats[format];
 	else if (IS_GEN12(devid))
 		return gen12_oa_formats[format];
 	else
@@ -1304,7 +1336,7 @@ read_2_oa_reports(int format_id,
 			igt_assert_eq(header->size, sample_size);
 
 			report = (const void *)(header + 1);
-			dump_report(report, 64, "oa-formats");
+			dump_report(report, format_size / 4, "oa-formats");
 
 			igt_debug("read report: reason = %x, timestamp = %x, exponent mask=%x\n",
 				  report[0], report[1], exponent_mask);
@@ -1540,8 +1572,8 @@ test_oa_formats(void)
 {
 	for (int i = 0; i < I915_OA_FORMAT_MAX; i++) {
 		struct oa_format format = get_oa_format(i);
-		uint32_t oa_report0[64];
-		uint32_t oa_report1[64];
+		uint32_t oa_report0[format.size / 4];
+		uint32_t oa_report1[format.size / 4];
 
 		if (!format.name) /* sparse, indexed by ID */
 			continue;
-- 
2.25.1

Re: [igt-dev] [PATCH i-g-t 13/23] i915/perf: Add OA formats for DG2

[Lionel Landwerlin]

On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
> Add OA formats for DG2.
>
> Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
> ---
>   include/drm-uapi/i915_drm.h |  6 ++++++
>   tests/i915/perf.c           | 38 ++++++++++++++++++++++++++++++++++---
>   2 files changed, 41 insertions(+), 3 deletions(-)
>
> diff --git a/include/drm-uapi/i915_drm.h b/include/drm-uapi/i915_drm.h
> index b4efc96c..c7587a0f 100644
> --- a/include/drm-uapi/i915_drm.h
> +++ b/include/drm-uapi/i915_drm.h
> @@ -2535,6 +2535,12 @@ enum drm_i915_oa_format {
>   	I915_OA_FORMAT_A12_B8_C8,
>   	I915_OA_FORMAT_A32u40_A4u32_B8_C8,
>   
> +	/* DG2 */
> +	I915_OAR_FORMAT_A32u40_A4u32_B8_C8,
> +	I915_OA_FORMAT_A24u40_A14u32_B8_C8,
> +	I915_OAR_FORMAT_A36u64_B8_C8,
> +	I915_OA_FORMAT_A38u64_R2u64_B8_C8,
> +
>   	I915_OA_FORMAT_MAX	    /* non-ABI */
>   };
>   
> diff --git a/tests/i915/perf.c b/tests/i915/perf.c
> index 85f3c5de..d601eb36 100644
> --- a/tests/i915/perf.c
> +++ b/tests/i915/perf.c
> @@ -181,6 +181,36 @@ static struct oa_format gen12_oa_formats[I915_OA_FORMAT_MAX] = {
>   		.c_off = 224, .n_c = 8, },
>   };
>   
> +static struct oa_format dg2_oa_formats[I915_OA_FORMAT_MAX] = {
> +	[I915_OAR_FORMAT_A32u40_A4u32_B8_C8] = {
> +		"A32u40_A4u32_B8_C8", .size = 256,
> +		.a40_high_off = 160, .a40_low_off = 16, .n_a40 = 32,
> +		.a_off = 144, .n_a = 4, .first_a = 32,
> +		.b_off = 192, .n_b = 8,
> +		.c_off = 224, .n_c = 8, .oa_type = OAR, },
> +	/* This format has A36 and A37 interleaved with high bytes of some A
> +	 * counters, so we will accumulate only subset of counters.
> +	 */
> +	[I915_OA_FORMAT_A24u40_A14u32_B8_C8] = {
> +		"A24u40_A14u32_B8_C8", .size = 256,
> +		/* u40: A4 - A23 */
> +		.a40_high_off = 164, .a40_low_off = 32, .n_a40 = 20,
> +		/* u32: A24 - A27 */
> +		.a_off = 144, .n_a = 4, .first_a = 24,
> +		.b_off = 192, .n_b = 8,
> +		.c_off = 224, .n_c = 8, .oa_type = OAG, },
> +	[I915_OA_FORMAT_A38u64_R2u64_B8_C8] = {
> +		"A38u64_R2u64_B8_C8", .size = 448,
> +		.a64_off = 32, .n_a64 = 38,
> +		.b_off = 352, .n_b = 8,
> +		.c_off = 384, .n_c = 8, .oa_type = OAG, },
> +	[I915_OAR_FORMAT_A36u64_B8_C8] = {
> +		"A36u64_B8_C8", .size = 384,
> +		.a64_off = 32, .n_a64 = 36,
> +		.b_off = 320, .n_b = 8,
> +		.c_off = 352, .n_c = 8, .oa_type = OAR, },
> +};
> +
>   static bool hsw_undefined_a_counters[45] = {
>   	[4] = true,
>   	[6] = true,
> @@ -239,6 +269,8 @@ get_oa_format(enum drm_i915_oa_format format)
>   {
>   	if (IS_HASWELL(devid))
>   		return hsw_oa_formats[format];
> +	else if (IS_DG2(devid))
> +		return dg2_oa_formats[format];
>   	else if (IS_GEN12(devid))
>   		return gen12_oa_formats[format];
>   	else
> @@ -1304,7 +1336,7 @@ read_2_oa_reports(int format_id,
>   			igt_assert_eq(header->size, sample_size);
>   
>   			report = (const void *)(header + 1);
> -			dump_report(report, 64, "oa-formats");
> +			dump_report(report, format_size / 4, "oa-formats");
>   
>   			igt_debug("read report: reason = %x, timestamp = %x, exponent mask=%x\n",
>   				  report[0], report[1], exponent_mask);
> @@ -1540,8 +1572,8 @@ test_oa_formats(void)
>   {
>   	for (int i = 0; i < I915_OA_FORMAT_MAX; i++) {
>   		struct oa_format format = get_oa_format(i);
> -		uint32_t oa_report0[64];
> -		uint32_t oa_report1[64];
> +		uint32_t oa_report0[format.size / 4];
> +		uint32_t oa_report1[format.size / 4];
>   
>   		if (!format.name) /* sparse, indexed by ID */
>   			continue;

[igt-dev] [PATCH i-g-t 14/23] i915/perf: Add a test for non-power-of-2 oa reports

[Umesh Nerlige Ramappa]

Some OA report sizes are no longer a power of 2. Verify that such
reports can be correctly read across the OA buffer boundaries.

Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
---
 tests/i915/perf.c | 34 +++++++++++++++++++++++++---------
 1 file changed, 25 insertions(+), 9 deletions(-)

diff --git a/tests/i915/perf.c b/tests/i915/perf.c
index d601eb36..3caf7cac 100644
--- a/tests/i915/perf.c
+++ b/tests/i915/perf.c
@@ -2681,7 +2681,7 @@ test_buffer_fill(void)
 }
 
 static void
-test_non_zero_reason(void)
+test_multi_buffer_fills(uint64_t format, uint64_t oa_buf_size, bool load)
 {
 	/* ~20 micro second period */
 	int oa_exponent = max_oa_exponent_for_period_lte(20000);
@@ -2691,7 +2691,7 @@ test_non_zero_reason(void)
 
 		/* OA unit configuration */
 		DRM_I915_PERF_PROP_OA_METRICS_SET, test_set->perf_oa_metrics_set,
-		DRM_I915_PERF_PROP_OA_FORMAT, test_set->perf_oa_format,
+		DRM_I915_PERF_PROP_OA_FORMAT, format,
 		DRM_I915_PERF_PROP_OA_EXPONENT, oa_exponent,
 	};
 	struct drm_i915_perf_open_param param = {
@@ -2700,7 +2700,9 @@ test_non_zero_reason(void)
 		.properties_ptr = to_user_pointer(properties),
 	};
 	struct drm_i915_perf_record_header *header;
-	uint32_t buf_size = 3 * 65536 * (256 + sizeof(struct drm_i915_perf_record_header));
+	size_t report_size = get_oa_format(format).size;
+	uint32_t buf_size = 3 * (oa_buf_size / report_size) *
+		(report_size + sizeof(struct drm_i915_perf_record_header));
 	uint8_t *buf = malloc(buf_size);
 	uint32_t total_len = 0, reports_lost;
 	const uint32_t *last_report;
@@ -2710,8 +2712,10 @@ test_non_zero_reason(void)
 
 	igt_debug("Ready to read about %u bytes\n", buf_size);
 
-	load_helper_init();
-	load_helper_run(HIGH);
+	if (load) {
+		load_helper_init();
+		load_helper_run(HIGH);
+	}
 
 	stream_fd = __perf_open(drm_fd, &param, true /* prevent_pm */);
 
@@ -2724,8 +2728,10 @@ test_non_zero_reason(void)
 
 	__perf_close(stream_fd);
 
-	load_helper_stop();
-	load_helper_fini();
+	if (load) {
+		load_helper_stop();
+		load_helper_fini();
+	}
 
 	igt_debug("Got %u bytes\n", total_len);
 
@@ -2747,7 +2753,7 @@ test_non_zero_reason(void)
 
 			if (last_report) {
 				sanity_check_reports(last_report, report,
-						     test_set->perf_oa_format);
+						     format);
 			}
 			last_report = report;
 			break;
@@ -5137,6 +5143,14 @@ igt_main
 	igt_subtest("oa-formats")
 		test_oa_formats();
 
+	igt_describe("Test OA buffer for non power of 2 report sizes");
+	igt_subtest("non-power-of-2-oa-formats") {
+		igt_require(IS_DG2(devid));
+		test_multi_buffer_fills(I915_OA_FORMAT_A38u64_R2u64_B8_C8,
+					MAX_OA_BUF_SIZE,
+					false);
+	}
+
 	igt_subtest("invalid-oa-exponent")
 		test_invalid_oa_exponent();
 	igt_subtest("low-oa-exponent-permissions")
@@ -5156,7 +5170,9 @@ igt_main
 	igt_subtest("non-zero-reason") {
 		/* Reason field is only available on Gen8+ */
 		igt_require(intel_gen(devid) >= 8);
-		test_non_zero_reason();
+		test_multi_buffer_fills(test_set->perf_oa_format,
+					MAX_OA_BUF_SIZE,
+					true);
 	}
 
 	igt_subtest("disabled-read-error")
-- 
2.25.1

Re: [igt-dev] [PATCH i-g-t 14/23] i915/perf: Add a test for non-power-of-2 oa reports

[Lionel Landwerlin]

On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
> Some OA report sizes are no longer a power of 2. Verify that such
> reports can be correctly read across the OA buffer boundaries.
>
> Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>


It seems you've disabled the load helper to avoid checking the content 
of the report?

I think the A counters should still be defined same as before and we 
could still check their value.


> ---
>   tests/i915/perf.c | 34 +++++++++++++++++++++++++---------
>   1 file changed, 25 insertions(+), 9 deletions(-)
>
> diff --git a/tests/i915/perf.c b/tests/i915/perf.c
> index d601eb36..3caf7cac 100644
> --- a/tests/i915/perf.c
> +++ b/tests/i915/perf.c
> @@ -2681,7 +2681,7 @@ test_buffer_fill(void)
>   }
>   
>   static void
> -test_non_zero_reason(void)
> +test_multi_buffer_fills(uint64_t format, uint64_t oa_buf_size, bool load)
>   {
>   	/* ~20 micro second period */
>   	int oa_exponent = max_oa_exponent_for_period_lte(20000);
> @@ -2691,7 +2691,7 @@ test_non_zero_reason(void)
>   
>   		/* OA unit configuration */
>   		DRM_I915_PERF_PROP_OA_METRICS_SET, test_set->perf_oa_metrics_set,
> -		DRM_I915_PERF_PROP_OA_FORMAT, test_set->perf_oa_format,
> +		DRM_I915_PERF_PROP_OA_FORMAT, format,
>   		DRM_I915_PERF_PROP_OA_EXPONENT, oa_exponent,
>   	};
>   	struct drm_i915_perf_open_param param = {
> @@ -2700,7 +2700,9 @@ test_non_zero_reason(void)
>   		.properties_ptr = to_user_pointer(properties),
>   	};
>   	struct drm_i915_perf_record_header *header;
> -	uint32_t buf_size = 3 * 65536 * (256 + sizeof(struct drm_i915_perf_record_header));
> +	size_t report_size = get_oa_format(format).size;
> +	uint32_t buf_size = 3 * (oa_buf_size / report_size) *
> +		(report_size + sizeof(struct drm_i915_perf_record_header));
>   	uint8_t *buf = malloc(buf_size);
>   	uint32_t total_len = 0, reports_lost;
>   	const uint32_t *last_report;
> @@ -2710,8 +2712,10 @@ test_non_zero_reason(void)
>   
>   	igt_debug("Ready to read about %u bytes\n", buf_size);
>   
> -	load_helper_init();
> -	load_helper_run(HIGH);
> +	if (load) {
> +		load_helper_init();
> +		load_helper_run(HIGH);
> +	}
>   
>   	stream_fd = __perf_open(drm_fd, &param, true /* prevent_pm */);
>   
> @@ -2724,8 +2728,10 @@ test_non_zero_reason(void)
>   
>   	__perf_close(stream_fd);
>   
> -	load_helper_stop();
> -	load_helper_fini();
> +	if (load) {
> +		load_helper_stop();
> +		load_helper_fini();
> +	}
>   
>   	igt_debug("Got %u bytes\n", total_len);
>   
> @@ -2747,7 +2753,7 @@ test_non_zero_reason(void)
>   
>   			if (last_report) {
>   				sanity_check_reports(last_report, report,
> -						     test_set->perf_oa_format);
> +						     format);
>   			}
>   			last_report = report;
>   			break;
> @@ -5137,6 +5143,14 @@ igt_main
>   	igt_subtest("oa-formats")
>   		test_oa_formats();
>   
> +	igt_describe("Test OA buffer for non power of 2 report sizes");
> +	igt_subtest("non-power-of-2-oa-formats") {
> +		igt_require(IS_DG2(devid));
> +		test_multi_buffer_fills(I915_OA_FORMAT_A38u64_R2u64_B8_C8,
> +					MAX_OA_BUF_SIZE,
> +					false);
> +	}


We could actually iterator over non power of 2 report sizes :


for (unsigned r = 0; r < ARRAY_SIZE(gen12_oa_formats); r++) {

     if (__builtin_popcount(gen12_oa_formats[r].size) == 1)

         continue;

     igt_describe("Test OA buffer for %s format", 
gen12_oa_formats[r].name);

     igt_subtest_f("oa-format-%s", gen12_oa_formats[r].name) {

         igt_require(intel_gen(devid) >= 12);

         test_multi_buffer_fills(r, MAX_OA_BUF_SIZE, false);

     }

}


> +
>   	igt_subtest("invalid-oa-exponent")
>   		test_invalid_oa_exponent();
>   	igt_subtest("low-oa-exponent-permissions")
> @@ -5156,7 +5170,9 @@ igt_main
>   	igt_subtest("non-zero-reason") {
>   		/* Reason field is only available on Gen8+ */
>   		igt_require(intel_gen(devid) >= 8);
> -		test_non_zero_reason();
> +		test_multi_buffer_fills(test_set->perf_oa_format,
> +					MAX_OA_BUF_SIZE,
> +					true);
>   	}
>   
>   	igt_subtest("disabled-read-error")

Re: [igt-dev] [PATCH i-g-t 14/23] i915/perf: Add a test for non-power-of-2 oa reports

[Umesh Nerlige Ramappa]

On Tue, Sep 06, 2022 at 05:08:34PM +0300, Lionel Landwerlin wrote:
>On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
>>Some OA report sizes are no longer a power of 2. Verify that such
>>reports can be correctly read across the OA buffer boundaries.
>>
>>Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
>
>
>It seems you've disabled the load helper to avoid checking the content 
>of the report?

yes, I think at the time this was added, there was only a media oa 
format with an odd size and I thought we don't want to run render copy 
with that. I think now we can enable the load since there are other 
formats that are applicable here.

>
>I think the A counters should still be defined same as before and we 
>could still check their value.
>
>
>>---
>>  tests/i915/perf.c | 34 +++++++++++++++++++++++++---------
>>  1 file changed, 25 insertions(+), 9 deletions(-)
>>
>>diff --git a/tests/i915/perf.c b/tests/i915/perf.c
>>index d601eb36..3caf7cac 100644
>>--- a/tests/i915/perf.c
>>+++ b/tests/i915/perf.c
>>@@ -2681,7 +2681,7 @@ test_buffer_fill(void)
>>  }
>>  static void
>>-test_non_zero_reason(void)
>>+test_multi_buffer_fills(uint64_t format, uint64_t oa_buf_size, bool load)
>>  {
>>  	/* ~20 micro second period */
>>  	int oa_exponent = max_oa_exponent_for_period_lte(20000);
>>@@ -2691,7 +2691,7 @@ test_non_zero_reason(void)
>>  		/* OA unit configuration */
>>  		DRM_I915_PERF_PROP_OA_METRICS_SET, test_set->perf_oa_metrics_set,
>>-		DRM_I915_PERF_PROP_OA_FORMAT, test_set->perf_oa_format,
>>+		DRM_I915_PERF_PROP_OA_FORMAT, format,
>>  		DRM_I915_PERF_PROP_OA_EXPONENT, oa_exponent,
>>  	};
>>  	struct drm_i915_perf_open_param param = {
>>@@ -2700,7 +2700,9 @@ test_non_zero_reason(void)
>>  		.properties_ptr = to_user_pointer(properties),
>>  	};
>>  	struct drm_i915_perf_record_header *header;
>>-	uint32_t buf_size = 3 * 65536 * (256 + sizeof(struct drm_i915_perf_record_header));
>>+	size_t report_size = get_oa_format(format).size;
>>+	uint32_t buf_size = 3 * (oa_buf_size / report_size) *
>>+		(report_size + sizeof(struct drm_i915_perf_record_header));
>>  	uint8_t *buf = malloc(buf_size);
>>  	uint32_t total_len = 0, reports_lost;
>>  	const uint32_t *last_report;
>>@@ -2710,8 +2712,10 @@ test_non_zero_reason(void)
>>  	igt_debug("Ready to read about %u bytes\n", buf_size);
>>-	load_helper_init();
>>-	load_helper_run(HIGH);
>>+	if (load) {
>>+		load_helper_init();
>>+		load_helper_run(HIGH);
>>+	}
>>  	stream_fd = __perf_open(drm_fd, &param, true /* prevent_pm */);
>>@@ -2724,8 +2728,10 @@ test_non_zero_reason(void)
>>  	__perf_close(stream_fd);
>>-	load_helper_stop();
>>-	load_helper_fini();
>>+	if (load) {
>>+		load_helper_stop();
>>+		load_helper_fini();
>>+	}
>>  	igt_debug("Got %u bytes\n", total_len);
>>@@ -2747,7 +2753,7 @@ test_non_zero_reason(void)
>>  			if (last_report) {
>>  				sanity_check_reports(last_report, report,
>>-						     test_set->perf_oa_format);
>>+						     format);
>>  			}
>>  			last_report = report;
>>  			break;
>>@@ -5137,6 +5143,14 @@ igt_main
>>  	igt_subtest("oa-formats")
>>  		test_oa_formats();
>>+	igt_describe("Test OA buffer for non power of 2 report sizes");
>>+	igt_subtest("non-power-of-2-oa-formats") {
>>+		igt_require(IS_DG2(devid));
>>+		test_multi_buffer_fills(I915_OA_FORMAT_A38u64_R2u64_B8_C8,
>>+					MAX_OA_BUF_SIZE,
>>+					false);
>>+	}
>
>
>We could actually iterator over non power of 2 report sizes :
>
>
>for (unsigned r = 0; r < ARRAY_SIZE(gen12_oa_formats); r++) {
>
>    if (__builtin_popcount(gen12_oa_formats[r].size) == 1)
>
>        continue;
>
>    igt_describe("Test OA buffer for %s format", 
>gen12_oa_formats[r].name);
>
>    igt_subtest_f("oa-format-%s", gen12_oa_formats[r].name) {
>
>        igt_require(intel_gen(devid) >= 12);
>
>        test_multi_buffer_fills(r, MAX_OA_BUF_SIZE, false);
>
>    }

will add,

Thanks,
Umesh
>
>}
>
>
>>+
>>  	igt_subtest("invalid-oa-exponent")
>>  		test_invalid_oa_exponent();
>>  	igt_subtest("low-oa-exponent-permissions")
>>@@ -5156,7 +5170,9 @@ igt_main
>>  	igt_subtest("non-zero-reason") {
>>  		/* Reason field is only available on Gen8+ */
>>  		igt_require(intel_gen(devid) >= 8);
>>-		test_non_zero_reason();
>>+		test_multi_buffer_fills(test_set->perf_oa_format,
>>+					MAX_OA_BUF_SIZE,
>>+					true);
>>  	}
>>  	igt_subtest("disabled-read-error")
>
>

[igt-dev] [PATCH i-g-t 15/23] i915/perf: Fix CS timestamp vs OA timstamp mismatch

[Umesh Nerlige Ramappa]

CS timestamp can have an additional configuration that shifts the
timestamp value by a specified number of bits. On DG2, this additional
configuration is ignored by OA, so OA timestamps are missing the shift.

Add newly defined interface to get the exact OA timestamp frequency so
that the deltas can be accurately calculated.

Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
---
 include/drm-uapi/i915_drm.h |  6 ++++++
 lib/i915/perf.c             |  6 +++++-
 tests/i915/perf.c           | 37 +++++++++++++++++++++++++++++++++++--
 3 files changed, 46 insertions(+), 3 deletions(-)

diff --git a/include/drm-uapi/i915_drm.h b/include/drm-uapi/i915_drm.h
index c7587a0f..11c40811 100644
--- a/include/drm-uapi/i915_drm.h
+++ b/include/drm-uapi/i915_drm.h
@@ -749,6 +749,12 @@ typedef struct drm_i915_irq_wait {
 /* Query if the kernel supports the I915_USERPTR_PROBE flag. */
 #define I915_PARAM_HAS_USERPTR_PROBE 56
 
+/*
+ * Frequency of the timestamps in OA reports. This used to be the same as the CS
+ * timestamp frequency, but differs on some platforms.
+ */
+#define I915_PARAM_OA_TIMESTAMP_FREQUENCY 57
+
 /* Must be kept compact -- no holes and well documented */
 
 typedef struct drm_i915_getparam {
diff --git a/lib/i915/perf.c b/lib/i915/perf.c
index 3f131aaa..2f24cbd8 100644
--- a/lib/i915/perf.c
+++ b/lib/i915/perf.c
@@ -464,7 +464,11 @@ intel_perf_for_fd(int drm_fd)
 	close(sysfs_dir_fd);
 
 	if (getparam(drm_fd, I915_PARAM_CHIPSET_ID, &device_id) ||
-	    getparam(drm_fd, I915_PARAM_REVISION, &device_revision) ||
+	    getparam(drm_fd, I915_PARAM_REVISION, &device_revision))
+		return NULL;
+
+	/* if OA_TIMESTAMP_FREQUENCY is not supported, fall back to CS_TIMESTAMP_FREQUENCY */
+	if (getparam(drm_fd, I915_PARAM_OA_TIMESTAMP_FREQUENCY, &timestamp_frequency) &&
 	    getparam(drm_fd, I915_PARAM_CS_TIMESTAMP_FREQUENCY, &timestamp_frequency))
 		return NULL;
 
diff --git a/tests/i915/perf.c b/tests/i915/perf.c
index 3caf7cac..3eb31f3b 100644
--- a/tests/i915/perf.c
+++ b/tests/i915/perf.c
@@ -453,6 +453,29 @@ gen8_read_report_reason(const uint32_t *report)
 		return "unknown";
 }
 
+static uint32_t
+cs_timestamp_frequency(int fd)
+{
+	struct drm_i915_getparam gp = {};
+	static uint32_t value = 0;
+
+	if (value)
+		return value;
+
+	gp.param = I915_PARAM_CS_TIMESTAMP_FREQUENCY;
+	gp.value = (int *)(&value);
+
+	igt_assert_eq(igt_ioctl(fd, DRM_IOCTL_I915_GETPARAM, &gp), 0);
+
+	return value;
+}
+
+static uint64_t
+cs_timebase_scale(uint32_t u32_delta)
+{
+	return ((uint64_t)u32_delta * NSEC_PER_SEC) / cs_timestamp_frequency(drm_fd);
+}
+
 static uint64_t
 timebase_scale(uint32_t u32_delta)
 {
@@ -543,6 +566,15 @@ oa_report_get_ctx_id(uint32_t *report)
 	return report[2];
 }
 
+static int
+oar_unit_default_format(void)
+{
+	if (IS_DG2(devid))
+		return I915_OAR_FORMAT_A32u40_A4u32_B8_C8;
+
+	return test_set->perf_oa_format;
+}
+
 /*
  * Temporary wrapper to distinguish mappings on !llc platforms,
  * where it seems cache over GEM_MMAP_OFFSET is not flushed before execution.
@@ -3959,6 +3991,7 @@ again:
 
 static void gen12_single_ctx_helper(void)
 {
+	uint64_t fmt = oar_unit_default_format();
 	uint64_t properties[] = {
 		/* Have a random value here for the context id, but initialize
 		 * it once you figure out the context ID for the work to be
@@ -3974,7 +4007,7 @@ static void gen12_single_ctx_helper(void)
 		 * values.
 		 */
 		DRM_I915_PERF_PROP_OA_METRICS_SET, test_set->perf_oa_metrics_set,
-		DRM_I915_PERF_PROP_OA_FORMAT, test_set->perf_oa_format,
+		DRM_I915_PERF_PROP_OA_FORMAT, fmt,
 	};
 	struct drm_i915_perf_open_param param = {
 		.flags = I915_PERF_FLAG_FD_CLOEXEC,
@@ -4202,7 +4235,7 @@ static void gen12_single_ctx_helper(void)
 	/* Sanity check that we can pass the delta to timebase_scale */
 	igt_assert(delta_ts64 < UINT32_MAX);
 	delta_oa32_ns = timebase_scale(delta_oa32);
-	delta_ts64_ns = timebase_scale(delta_ts64);
+	delta_ts64_ns = cs_timebase_scale(delta_ts64);
 
 	igt_debug("oa32 delta = %u, = %uns\n",
 			delta_oa32, (unsigned)delta_oa32_ns);
-- 
2.25.1

Re: [igt-dev] [PATCH i-g-t 15/23] i915/perf: Fix CS timestamp vs OA timstamp mismatch

[Lionel Landwerlin]

On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
> CS timestamp can have an additional configuration that shifts the
> timestamp value by a specified number of bits. On DG2, this additional
> configuration is ignored by OA, so OA timestamps are missing the shift.
>
> Add newly defined interface to get the exact OA timestamp frequency so
> that the deltas can be accurately calculated.
>
> Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
> ---
>   include/drm-uapi/i915_drm.h |  6 ++++++
>   lib/i915/perf.c             |  6 +++++-
>   tests/i915/perf.c           | 37 +++++++++++++++++++++++++++++++++++--
>   3 files changed, 46 insertions(+), 3 deletions(-)
>
> diff --git a/include/drm-uapi/i915_drm.h b/include/drm-uapi/i915_drm.h
> index c7587a0f..11c40811 100644
> --- a/include/drm-uapi/i915_drm.h
> +++ b/include/drm-uapi/i915_drm.h
> @@ -749,6 +749,12 @@ typedef struct drm_i915_irq_wait {
>   /* Query if the kernel supports the I915_USERPTR_PROBE flag. */
>   #define I915_PARAM_HAS_USERPTR_PROBE 56
>   
> +/*
> + * Frequency of the timestamps in OA reports. This used to be the same as the CS
> + * timestamp frequency, but differs on some platforms.
> + */
> +#define I915_PARAM_OA_TIMESTAMP_FREQUENCY 57
> +
>   /* Must be kept compact -- no holes and well documented */
>   
>   typedef struct drm_i915_getparam {
> diff --git a/lib/i915/perf.c b/lib/i915/perf.c
> index 3f131aaa..2f24cbd8 100644
> --- a/lib/i915/perf.c
> +++ b/lib/i915/perf.c
> @@ -464,7 +464,11 @@ intel_perf_for_fd(int drm_fd)
>   	close(sysfs_dir_fd);
>   
>   	if (getparam(drm_fd, I915_PARAM_CHIPSET_ID, &device_id) ||
> -	    getparam(drm_fd, I915_PARAM_REVISION, &device_revision) ||
> +	    getparam(drm_fd, I915_PARAM_REVISION, &device_revision))
> +		return NULL;
> +
> +	/* if OA_TIMESTAMP_FREQUENCY is not supported, fall back to CS_TIMESTAMP_FREQUENCY */
> +	if (getparam(drm_fd, I915_PARAM_OA_TIMESTAMP_FREQUENCY, &timestamp_frequency) &&
>   	    getparam(drm_fd, I915_PARAM_CS_TIMESTAMP_FREQUENCY, &timestamp_frequency))
>   		return NULL;
>   
> diff --git a/tests/i915/perf.c b/tests/i915/perf.c
> index 3caf7cac..3eb31f3b 100644
> --- a/tests/i915/perf.c
> +++ b/tests/i915/perf.c
> @@ -453,6 +453,29 @@ gen8_read_report_reason(const uint32_t *report)
>   		return "unknown";
>   }
>   
> +static uint32_t
> +cs_timestamp_frequency(int fd)
> +{
> +	struct drm_i915_getparam gp = {};
> +	static uint32_t value = 0;
> +
> +	if (value)
> +		return value;
> +
> +	gp.param = I915_PARAM_CS_TIMESTAMP_FREQUENCY;
> +	gp.value = (int *)(&value);
> +
> +	igt_assert_eq(igt_ioctl(fd, DRM_IOCTL_I915_GETPARAM, &gp), 0);
> +
> +	return value;
> +}
> +
> +static uint64_t
> +cs_timebase_scale(uint32_t u32_delta)
> +{
> +	return ((uint64_t)u32_delta * NSEC_PER_SEC) / cs_timestamp_frequency(drm_fd);
> +}
> +
>   static uint64_t
>   timebase_scale(uint32_t u32_delta)
>   {
> @@ -543,6 +566,15 @@ oa_report_get_ctx_id(uint32_t *report)
>   	return report[2];
>   }
>   
> +static int
> +oar_unit_default_format(void)
> +{
> +	if (IS_DG2(devid))
> +		return I915_OAR_FORMAT_A32u40_A4u32_B8_C8;
> +
> +	return test_set->perf_oa_format;
> +}
> +
>   /*
>    * Temporary wrapper to distinguish mappings on !llc platforms,
>    * where it seems cache over GEM_MMAP_OFFSET is not flushed before execution.
> @@ -3959,6 +3991,7 @@ again:
>   
>   static void gen12_single_ctx_helper(void)
>   {
> +	uint64_t fmt = oar_unit_default_format();
>   	uint64_t properties[] = {
>   		/* Have a random value here for the context id, but initialize
>   		 * it once you figure out the context ID for the work to be
> @@ -3974,7 +4007,7 @@ static void gen12_single_ctx_helper(void)
>   		 * values.
>   		 */
>   		DRM_I915_PERF_PROP_OA_METRICS_SET, test_set->perf_oa_metrics_set,
> -		DRM_I915_PERF_PROP_OA_FORMAT, test_set->perf_oa_format,
> +		DRM_I915_PERF_PROP_OA_FORMAT, fmt,
>   	};
>   	struct drm_i915_perf_open_param param = {
>   		.flags = I915_PERF_FLAG_FD_CLOEXEC,
> @@ -4202,7 +4235,7 @@ static void gen12_single_ctx_helper(void)
>   	/* Sanity check that we can pass the delta to timebase_scale */
>   	igt_assert(delta_ts64 < UINT32_MAX);
>   	delta_oa32_ns = timebase_scale(delta_oa32);
> -	delta_ts64_ns = timebase_scale(delta_ts64);
> +	delta_ts64_ns = cs_timebase_scale(delta_ts64);
>   
>   	igt_debug("oa32 delta = %u, = %uns\n",
>   			delta_oa32, (unsigned)delta_oa32_ns);

[igt-dev] [PATCH i-g-t 16/23] i915/perf: Treat ticks as 64 bit

[Umesh Nerlige Ramappa]

DG2 A0 has a buggy header in the 64 bit OAG format. Except the ticks,
all other fields are 32 bit. To resolve this, we will store ticks in 64
bit type.

Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
---
 tests/i915/perf.c | 82 ++++++++++++++++++++++++++++++++---------------
 1 file changed, 56 insertions(+), 26 deletions(-)

diff --git a/tests/i915/perf.c b/tests/i915/perf.c
index 3eb31f3b..8836da66 100644
--- a/tests/i915/perf.c
+++ b/tests/i915/perf.c
@@ -117,6 +117,7 @@ struct oa_format {
 	int c_off;
 	int n_c;
 	int oa_type;
+	bool report_hdr_64bit;
 };
 
 static struct oa_format hsw_oa_formats[I915_OA_FORMAT_MAX] = {
@@ -203,7 +204,8 @@ static struct oa_format dg2_oa_formats[I915_OA_FORMAT_MAX] = {
 		"A38u64_R2u64_B8_C8", .size = 448,
 		.a64_off = 32, .n_a64 = 38,
 		.b_off = 352, .n_b = 8,
-		.c_off = 384, .n_c = 8, .oa_type = OAG, },
+		.c_off = 384, .n_c = 8, .oa_type = OAG,
+		.report_hdr_64bit = true, },
 	[I915_OAR_FORMAT_A36u64_B8_C8] = {
 		"A36u64_B8_C8", .size = 384,
 		.a64_off = 32, .n_a64 = 36,
@@ -245,7 +247,7 @@ static bool *undefined_a_counters;
 static uint64_t oa_exp_1_millisec;
 
 static igt_render_copyfunc_t render_copy = NULL;
-static uint32_t (*read_report_ticks)(const uint32_t *report,
+static uint64_t (*read_report_ticks)(const uint32_t *report,
 				     enum drm_i915_oa_format format);
 static void (*sanity_check_reports)(const uint32_t *oa_report0,
 				    const uint32_t *oa_report1,
@@ -395,7 +397,7 @@ sysfs_read(enum i915_attr_id id)
  * C2 corresponds to a clock counter for the Haswell render basic metric set
  * but it's not included in all of the formats.
  */
-static uint32_t
+static uint64_t
 hsw_read_report_ticks(const uint32_t *report, enum drm_i915_oa_format format)
 {
 	uint32_t *c = (uint32_t *)(((uint8_t *)report) + get_oa_format(format).c_off);
@@ -405,10 +407,41 @@ hsw_read_report_ticks(const uint32_t *report, enum drm_i915_oa_format format)
 	return c[2];
 }
 
-static uint32_t
+static uint64_t
 gen8_read_report_ticks(const uint32_t *report, enum drm_i915_oa_format format)
 {
-	return report[3];
+
+	struct oa_format fmt = get_oa_format(format);
+
+	return fmt.report_hdr_64bit ? *(uint64_t *)&report[6] : report[3];
+}
+
+/*
+ * t0 is a value sampled before t1. width is number of bits used to represent
+ * t0/t1. Normally t1 is greater than t0. In cases where t1 < t0 use this
+ * helper. Since the size of t1/t0 is already 64 bits, no special handling is
+ * needed for width = 64.
+ */
+static uint64_t
+__elapsed_delta(uint64_t t1, uint64_t t0, uint32_t width)
+{
+	uint32_t max_bits = sizeof(t1) * 8;
+
+	igt_assert(width <= max_bits); 
+
+	if (t1 < t0 && width != max_bits)
+		return ((1ULL << width) - t0) + t1;
+
+	return t1 - t0;
+}
+
+static uint64_t
+oa_tick_delta(const uint32_t *report1,
+	      const uint32_t *report0,
+	      enum drm_i915_oa_format format)
+{
+	return __elapsed_delta(read_report_ticks(report1, format),
+			       read_report_ticks(report0, format), 32);
 }
 
 static void
@@ -638,8 +671,8 @@ hsw_sanity_check_render_basic_reports(const uint32_t *oa_report0,
 				      enum drm_i915_oa_format fmt)
 {
 	uint32_t time_delta = timebase_scale(oa_report1[1] - oa_report0[1]);
-	uint32_t clock_delta;
-	uint32_t max_delta;
+	uint64_t clock_delta;
+	uint64_t max_delta;
 	struct oa_format format = get_oa_format(fmt);
 
 	igt_assert_neq(time_delta, 0);
@@ -654,21 +687,19 @@ hsw_sanity_check_render_basic_reports(const uint32_t *oa_report0,
 		clock_delta = (gt_max_freq_mhz *
 			       (uint64_t)time_delta) / 1000;
 	} else {
-		uint32_t ticks0 = read_report_ticks(oa_report0, fmt);
-		uint32_t ticks1 = read_report_ticks(oa_report1, fmt);
 		uint64_t freq;
 
-		clock_delta = ticks1 - ticks0;
+		clock_delta = oa_tick_delta(oa_report1, oa_report0, fmt);
 
-		igt_assert_neq(clock_delta, 0);
+		igt_assert_neq_u64(clock_delta, 0);
 
-		freq = ((uint64_t)clock_delta * 1000) / time_delta;
+		freq = (clock_delta * 1000) / time_delta;
 		igt_debug("freq = %"PRIu64"\n", freq);
 
 		igt_assert(freq <= gt_max_freq_mhz);
 	}
 
-	igt_debug("clock delta = %"PRIu32"\n", clock_delta);
+	igt_debug("clock delta = %"PRIu64"\n", clock_delta);
 
 	/* The maximum rate for any HSW counter =
 	 *   clock_delta * N EUs
@@ -801,7 +832,8 @@ accumulate_reports(struct accumulator *accumulator,
 		accumulate_uint32(4, start, end, deltas + idx++);
 
 		/* clock cycles */
-		accumulate_uint32(12, start, end, deltas + idx++);
+		deltas[idx] += oa_tick_delta(end, start, accumulator->format);
+		idx++;
 	} else {
 		/* timestamp */
 		accumulate_uint32(4, start, end, deltas + idx++);
@@ -874,10 +906,8 @@ gen8_sanity_check_test_oa_reports(const uint32_t *oa_report0,
 {
 	struct oa_format format = get_oa_format(fmt);
 	uint32_t time_delta = timebase_scale(oa_report1[1] - oa_report0[1]);
-	uint32_t ticks0 = read_report_ticks(oa_report0, fmt);
-	uint32_t ticks1 = read_report_ticks(oa_report1, fmt);
-	uint32_t clock_delta = ticks1 - ticks0;
-	uint32_t max_delta;
+	uint64_t clock_delta = oa_tick_delta(oa_report1, oa_report0, fmt);
+	uint64_t max_delta;
 	uint64_t freq;
 	uint32_t *rpt0_b = (uint32_t *)(((uint8_t *)oa_report0) +
 					format.b_off);
@@ -890,10 +920,10 @@ gen8_sanity_check_test_oa_reports(const uint32_t *oa_report0,
 		  gen8_read_report_reason(oa_report0),
 		  gen8_read_report_reason(oa_report1));
 
-	freq = time_delta ? ((uint64_t)clock_delta * 1000) / time_delta : 0;
+	freq = time_delta ? (clock_delta * 1000) / time_delta : 0;
 	igt_debug("freq = %"PRIu64"\n", freq);
 
-	igt_debug("clock delta = %"PRIu32"\n", clock_delta);
+	igt_debug("clock delta = %"PRIu64"\n", clock_delta);
 
 	max_delta = clock_delta * intel_perf->devinfo.n_eus;
 
@@ -994,7 +1024,7 @@ gen8_sanity_check_test_oa_reports(const uint32_t *oa_report0,
 					    format.c_off);
 		uint32_t delta = c1[j] - c0[j];
 
-		igt_debug("C%d: delta = %"PRIu32", max_delta=%"PRIu32"\n",
+		igt_debug("C%d: delta = %"PRIu32", max_delta=%"PRIu64"\n",
 			  j, delta, max_delta);
 		igt_assert(delta <= max_delta);
 	}
@@ -1440,10 +1470,10 @@ print_reports(uint32_t *oa_report0, uint32_t *oa_report1, int fmt)
 	if (IS_HASWELL(devid) && format.n_c == 0) {
 		igt_debug("CLOCK = N/A\n");
 	} else {
-		uint32_t clock0 = read_report_ticks(oa_report0, fmt);
-		uint32_t clock1 = read_report_ticks(oa_report1, fmt);
+		uint64_t clock0 = read_report_ticks(oa_report0, fmt);
+		uint64_t clock1 = read_report_ticks(oa_report1, fmt);
 
-		igt_debug("CLOCK: 1st = %"PRIu32", 2nd = %"PRIu32", delta = %"PRIu32"\n",
+		igt_debug("CLOCK: 1st = %"PRIu64", 2nd = %"PRIu64", delta = %"PRIu64"\n",
 			  clock0, clock1, clock1 - clock0);
 	}
 
@@ -1545,9 +1575,9 @@ print_report(uint32_t *report, int fmt)
 	if (IS_HASWELL(devid) && format.n_c == 0) {
 		igt_debug("CLOCK = N/A\n");
 	} else {
-		uint32_t clock = read_report_ticks(report, fmt);
+		uint64_t clock = read_report_ticks(report, fmt);
 
-		igt_debug("CLOCK: %"PRIu32"\n", clock);
+		igt_debug("CLOCK: %"PRIu64"\n", clock);
 	}
 
 	if (intel_gen(devid) >= 8) {
-- 
2.25.1

Re: [igt-dev] [PATCH i-g-t 16/23] i915/perf: Treat ticks as 64 bit

[Lionel Landwerlin]

On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
> DG2 A0 has a buggy header in the 64 bit OAG format. Except the ticks,
> all other fields are 32 bit. To resolve this, we will store ticks in 64
> bit type.
>
> Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
> ---
>   tests/i915/perf.c | 82 ++++++++++++++++++++++++++++++++---------------
>   1 file changed, 56 insertions(+), 26 deletions(-)
>
> diff --git a/tests/i915/perf.c b/tests/i915/perf.c
> index 3eb31f3b..8836da66 100644
> --- a/tests/i915/perf.c
> +++ b/tests/i915/perf.c
> @@ -117,6 +117,7 @@ struct oa_format {
>   	int c_off;
>   	int n_c;
>   	int oa_type;
> +	bool report_hdr_64bit;
>   };
>   
>   static struct oa_format hsw_oa_formats[I915_OA_FORMAT_MAX] = {
> @@ -203,7 +204,8 @@ static struct oa_format dg2_oa_formats[I915_OA_FORMAT_MAX] = {
>   		"A38u64_R2u64_B8_C8", .size = 448,
>   		.a64_off = 32, .n_a64 = 38,
>   		.b_off = 352, .n_b = 8,
> -		.c_off = 384, .n_c = 8, .oa_type = OAG, },
> +		.c_off = 384, .n_c = 8, .oa_type = OAG,
> +		.report_hdr_64bit = true, },
>   	[I915_OAR_FORMAT_A36u64_B8_C8] = {
>   		"A36u64_B8_C8", .size = 384,
>   		.a64_off = 32, .n_a64 = 36,
> @@ -245,7 +247,7 @@ static bool *undefined_a_counters;
>   static uint64_t oa_exp_1_millisec;
>   
>   static igt_render_copyfunc_t render_copy = NULL;
> -static uint32_t (*read_report_ticks)(const uint32_t *report,
> +static uint64_t (*read_report_ticks)(const uint32_t *report,
>   				     enum drm_i915_oa_format format);
>   static void (*sanity_check_reports)(const uint32_t *oa_report0,
>   				    const uint32_t *oa_report1,
> @@ -395,7 +397,7 @@ sysfs_read(enum i915_attr_id id)
>    * C2 corresponds to a clock counter for the Haswell render basic metric set
>    * but it's not included in all of the formats.
>    */
> -static uint32_t
> +static uint64_t
>   hsw_read_report_ticks(const uint32_t *report, enum drm_i915_oa_format format)
>   {
>   	uint32_t *c = (uint32_t *)(((uint8_t *)report) + get_oa_format(format).c_off);
> @@ -405,10 +407,41 @@ hsw_read_report_ticks(const uint32_t *report, enum drm_i915_oa_format format)
>   	return c[2];
>   }
>   
> -static uint32_t
> +static uint64_t
>   gen8_read_report_ticks(const uint32_t *report, enum drm_i915_oa_format format)
>   {
> -	return report[3];
> +
> +	struct oa_format fmt = get_oa_format(format);
> +
> +	return fmt.report_hdr_64bit ? *(uint64_t *)&report[6] : report[3];
> +}
> +
> +/*
> + * t0 is a value sampled before t1. width is number of bits used to represent
> + * t0/t1. Normally t1 is greater than t0. In cases where t1 < t0 use this
> + * helper. Since the size of t1/t0 is already 64 bits, no special handling is
> + * needed for width = 64.
> + */
> +static uint64_t
> +__elapsed_delta(uint64_t t1, uint64_t t0, uint32_t width)
> +{
> +	uint32_t max_bits = sizeof(t1) * 8;
> +
> +	igt_assert(width <= max_bits);
> +
> +	if (t1 < t0 && width != max_bits)
> +		return ((1ULL << width) - t0) + t1;


Should this be : ((1ULL << width) + t1) - t0 ?


Since we know the t1 wrapped around the counter's max bit size, the 
actual difference is the max value + whatever is in the counter minus t0.


> +
> +	return t1 - t0;
> +}
> +
> +static uint64_t
> +oa_tick_delta(const uint32_t *report1,
> +	      const uint32_t *report0,
> +	      enum drm_i915_oa_format format)
> +{
> +	return __elapsed_delta(read_report_ticks(report1, format),
> +			       read_report_ticks(report0, format), 32);
>   }
>   
>   static void
> @@ -638,8 +671,8 @@ hsw_sanity_check_render_basic_reports(const uint32_t *oa_report0,
>   				      enum drm_i915_oa_format fmt)
>   {
>   	uint32_t time_delta = timebase_scale(oa_report1[1] - oa_report0[1]);
> -	uint32_t clock_delta;
> -	uint32_t max_delta;
> +	uint64_t clock_delta;
> +	uint64_t max_delta;
>   	struct oa_format format = get_oa_format(fmt);
>   
>   	igt_assert_neq(time_delta, 0);
> @@ -654,21 +687,19 @@ hsw_sanity_check_render_basic_reports(const uint32_t *oa_report0,
>   		clock_delta = (gt_max_freq_mhz *
>   			       (uint64_t)time_delta) / 1000;
>   	} else {
> -		uint32_t ticks0 = read_report_ticks(oa_report0, fmt);
> -		uint32_t ticks1 = read_report_ticks(oa_report1, fmt);
>   		uint64_t freq;
>   
> -		clock_delta = ticks1 - ticks0;
> +		clock_delta = oa_tick_delta(oa_report1, oa_report0, fmt);
>   
> -		igt_assert_neq(clock_delta, 0);
> +		igt_assert_neq_u64(clock_delta, 0);
>   
> -		freq = ((uint64_t)clock_delta * 1000) / time_delta;
> +		freq = (clock_delta * 1000) / time_delta;
>   		igt_debug("freq = %"PRIu64"\n", freq);
>   
>   		igt_assert(freq <= gt_max_freq_mhz);
>   	}
>   
> -	igt_debug("clock delta = %"PRIu32"\n", clock_delta);
> +	igt_debug("clock delta = %"PRIu64"\n", clock_delta);
>   
>   	/* The maximum rate for any HSW counter =
>   	 *   clock_delta * N EUs
> @@ -801,7 +832,8 @@ accumulate_reports(struct accumulator *accumulator,
>   		accumulate_uint32(4, start, end, deltas + idx++);
>   
>   		/* clock cycles */
> -		accumulate_uint32(12, start, end, deltas + idx++);
> +		deltas[idx] += oa_tick_delta(end, start, accumulator->format);
> +		idx++;
>   	} else {
>   		/* timestamp */
>   		accumulate_uint32(4, start, end, deltas + idx++);
> @@ -874,10 +906,8 @@ gen8_sanity_check_test_oa_reports(const uint32_t *oa_report0,
>   {
>   	struct oa_format format = get_oa_format(fmt);
>   	uint32_t time_delta = timebase_scale(oa_report1[1] - oa_report0[1]);
> -	uint32_t ticks0 = read_report_ticks(oa_report0, fmt);
> -	uint32_t ticks1 = read_report_ticks(oa_report1, fmt);
> -	uint32_t clock_delta = ticks1 - ticks0;
> -	uint32_t max_delta;
> +	uint64_t clock_delta = oa_tick_delta(oa_report1, oa_report0, fmt);
> +	uint64_t max_delta;
>   	uint64_t freq;
>   	uint32_t *rpt0_b = (uint32_t *)(((uint8_t *)oa_report0) +
>   					format.b_off);
> @@ -890,10 +920,10 @@ gen8_sanity_check_test_oa_reports(const uint32_t *oa_report0,
>   		  gen8_read_report_reason(oa_report0),
>   		  gen8_read_report_reason(oa_report1));
>   
> -	freq = time_delta ? ((uint64_t)clock_delta * 1000) / time_delta : 0;
> +	freq = time_delta ? (clock_delta * 1000) / time_delta : 0;
>   	igt_debug("freq = %"PRIu64"\n", freq);
>   
> -	igt_debug("clock delta = %"PRIu32"\n", clock_delta);
> +	igt_debug("clock delta = %"PRIu64"\n", clock_delta);
>   
>   	max_delta = clock_delta * intel_perf->devinfo.n_eus;
>   
> @@ -994,7 +1024,7 @@ gen8_sanity_check_test_oa_reports(const uint32_t *oa_report0,
>   					    format.c_off);
>   		uint32_t delta = c1[j] - c0[j];
>   
> -		igt_debug("C%d: delta = %"PRIu32", max_delta=%"PRIu32"\n",
> +		igt_debug("C%d: delta = %"PRIu32", max_delta=%"PRIu64"\n",
>   			  j, delta, max_delta);
>   		igt_assert(delta <= max_delta);
>   	}
> @@ -1440,10 +1470,10 @@ print_reports(uint32_t *oa_report0, uint32_t *oa_report1, int fmt)
>   	if (IS_HASWELL(devid) && format.n_c == 0) {
>   		igt_debug("CLOCK = N/A\n");
>   	} else {
> -		uint32_t clock0 = read_report_ticks(oa_report0, fmt);
> -		uint32_t clock1 = read_report_ticks(oa_report1, fmt);
> +		uint64_t clock0 = read_report_ticks(oa_report0, fmt);
> +		uint64_t clock1 = read_report_ticks(oa_report1, fmt);
>   
> -		igt_debug("CLOCK: 1st = %"PRIu32", 2nd = %"PRIu32", delta = %"PRIu32"\n",
> +		igt_debug("CLOCK: 1st = %"PRIu64", 2nd = %"PRIu64", delta = %"PRIu64"\n",
>   			  clock0, clock1, clock1 - clock0);
>   	}
>   
> @@ -1545,9 +1575,9 @@ print_report(uint32_t *report, int fmt)
>   	if (IS_HASWELL(devid) && format.n_c == 0) {
>   		igt_debug("CLOCK = N/A\n");
>   	} else {
> -		uint32_t clock = read_report_ticks(report, fmt);
> +		uint64_t clock = read_report_ticks(report, fmt);
>   
> -		igt_debug("CLOCK: %"PRIu32"\n", clock);
> +		igt_debug("CLOCK: %"PRIu64"\n", clock);
>   	}
>   
>   	if (intel_gen(devid) >= 8) {

Re: [igt-dev] [PATCH i-g-t 16/23] i915/perf: Treat ticks as 64 bit

[Umesh Nerlige Ramappa]

On Tue, Sep 06, 2022 at 05:08:50PM +0300, Lionel Landwerlin wrote:
>On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
>>DG2 A0 has a buggy header in the 64 bit OAG format. Except the ticks,
>>all other fields are 32 bit. To resolve this, we will store ticks in 64
>>bit type.
>>
>>Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
>>---
>>  tests/i915/perf.c | 82 ++++++++++++++++++++++++++++++++---------------
>>  1 file changed, 56 insertions(+), 26 deletions(-)
>>
>>diff --git a/tests/i915/perf.c b/tests/i915/perf.c
>>index 3eb31f3b..8836da66 100644
>>--- a/tests/i915/perf.c
>>+++ b/tests/i915/perf.c
>>@@ -117,6 +117,7 @@ struct oa_format {
>>  	int c_off;
>>  	int n_c;
>>  	int oa_type;
>>+	bool report_hdr_64bit;
>>  };
>>  static struct oa_format hsw_oa_formats[I915_OA_FORMAT_MAX] = {
>>@@ -203,7 +204,8 @@ static struct oa_format dg2_oa_formats[I915_OA_FORMAT_MAX] = {
>>  		"A38u64_R2u64_B8_C8", .size = 448,
>>  		.a64_off = 32, .n_a64 = 38,
>>  		.b_off = 352, .n_b = 8,
>>-		.c_off = 384, .n_c = 8, .oa_type = OAG, },
>>+		.c_off = 384, .n_c = 8, .oa_type = OAG,
>>+		.report_hdr_64bit = true, },
>>  	[I915_OAR_FORMAT_A36u64_B8_C8] = {
>>  		"A36u64_B8_C8", .size = 384,
>>  		.a64_off = 32, .n_a64 = 36,
>>@@ -245,7 +247,7 @@ static bool *undefined_a_counters;
>>  static uint64_t oa_exp_1_millisec;
>>  static igt_render_copyfunc_t render_copy = NULL;
>>-static uint32_t (*read_report_ticks)(const uint32_t *report,
>>+static uint64_t (*read_report_ticks)(const uint32_t *report,
>>  				     enum drm_i915_oa_format format);
>>  static void (*sanity_check_reports)(const uint32_t *oa_report0,
>>  				    const uint32_t *oa_report1,
>>@@ -395,7 +397,7 @@ sysfs_read(enum i915_attr_id id)
>>   * C2 corresponds to a clock counter for the Haswell render basic metric set
>>   * but it's not included in all of the formats.
>>   */
>>-static uint32_t
>>+static uint64_t
>>  hsw_read_report_ticks(const uint32_t *report, enum drm_i915_oa_format format)
>>  {
>>  	uint32_t *c = (uint32_t *)(((uint8_t *)report) + get_oa_format(format).c_off);
>>@@ -405,10 +407,41 @@ hsw_read_report_ticks(const uint32_t *report, enum drm_i915_oa_format format)
>>  	return c[2];
>>  }
>>-static uint32_t
>>+static uint64_t
>>  gen8_read_report_ticks(const uint32_t *report, enum drm_i915_oa_format format)
>>  {
>>-	return report[3];
>>+
>>+	struct oa_format fmt = get_oa_format(format);
>>+
>>+	return fmt.report_hdr_64bit ? *(uint64_t *)&report[6] : report[3];
>>+}
>>+
>>+/*
>>+ * t0 is a value sampled before t1. width is number of bits used to represent
>>+ * t0/t1. Normally t1 is greater than t0. In cases where t1 < t0 use this
>>+ * helper. Since the size of t1/t0 is already 64 bits, no special handling is
>>+ * needed for width = 64.
>>+ */
>>+static uint64_t
>>+__elapsed_delta(uint64_t t1, uint64_t t0, uint32_t width)
>>+{
>>+	uint32_t max_bits = sizeof(t1) * 8;
>>+
>>+	igt_assert(width <= max_bits);
>>+
>>+	if (t1 < t0 && width != max_bits)
>>+		return ((1ULL << width) - t0) + t1;
>
>
>Should this be : ((1ULL << width) + t1) - t0 ?
>

Both ways should yield the same result.

>
>Since we know the t1 wrapped around the counter's max bit size, the 
>actual difference is the max value + whatever is in the counter minus 
>t0.
>
>
>>+
>>+	return t1 - t0;
>>+}
>>+
>>+static uint64_t
>>+oa_tick_delta(const uint32_t *report1,
>>+	      const uint32_t *report0,
>>+	      enum drm_i915_oa_format format)
>>+{
>>+	return __elapsed_delta(read_report_ticks(report1, format),
>>+			       read_report_ticks(report0, format), 32);
>>  }
>>  static void
>>@@ -638,8 +671,8 @@ hsw_sanity_check_render_basic_reports(const uint32_t *oa_report0,
>>  				      enum drm_i915_oa_format fmt)
>>  {
>>  	uint32_t time_delta = timebase_scale(oa_report1[1] - oa_report0[1]);
>>-	uint32_t clock_delta;
>>-	uint32_t max_delta;
>>+	uint64_t clock_delta;
>>+	uint64_t max_delta;
>>  	struct oa_format format = get_oa_format(fmt);
>>  	igt_assert_neq(time_delta, 0);
>>@@ -654,21 +687,19 @@ hsw_sanity_check_render_basic_reports(const uint32_t *oa_report0,
>>  		clock_delta = (gt_max_freq_mhz *
>>  			       (uint64_t)time_delta) / 1000;
>>  	} else {
>>-		uint32_t ticks0 = read_report_ticks(oa_report0, fmt);
>>-		uint32_t ticks1 = read_report_ticks(oa_report1, fmt);
>>  		uint64_t freq;
>>-		clock_delta = ticks1 - ticks0;
>>+		clock_delta = oa_tick_delta(oa_report1, oa_report0, fmt);
>>-		igt_assert_neq(clock_delta, 0);
>>+		igt_assert_neq_u64(clock_delta, 0);
>>-		freq = ((uint64_t)clock_delta * 1000) / time_delta;
>>+		freq = (clock_delta * 1000) / time_delta;
>>  		igt_debug("freq = %"PRIu64"\n", freq);
>>  		igt_assert(freq <= gt_max_freq_mhz);
>>  	}
>>-	igt_debug("clock delta = %"PRIu32"\n", clock_delta);
>>+	igt_debug("clock delta = %"PRIu64"\n", clock_delta);
>>  	/* The maximum rate for any HSW counter =
>>  	 *   clock_delta * N EUs
>>@@ -801,7 +832,8 @@ accumulate_reports(struct accumulator *accumulator,
>>  		accumulate_uint32(4, start, end, deltas + idx++);
>>  		/* clock cycles */
>>-		accumulate_uint32(12, start, end, deltas + idx++);
>>+		deltas[idx] += oa_tick_delta(end, start, accumulator->format);
>>+		idx++;
>>  	} else {
>>  		/* timestamp */
>>  		accumulate_uint32(4, start, end, deltas + idx++);
>>@@ -874,10 +906,8 @@ gen8_sanity_check_test_oa_reports(const uint32_t *oa_report0,
>>  {
>>  	struct oa_format format = get_oa_format(fmt);
>>  	uint32_t time_delta = timebase_scale(oa_report1[1] - oa_report0[1]);
>>-	uint32_t ticks0 = read_report_ticks(oa_report0, fmt);
>>-	uint32_t ticks1 = read_report_ticks(oa_report1, fmt);
>>-	uint32_t clock_delta = ticks1 - ticks0;
>>-	uint32_t max_delta;
>>+	uint64_t clock_delta = oa_tick_delta(oa_report1, oa_report0, fmt);
>>+	uint64_t max_delta;
>>  	uint64_t freq;
>>  	uint32_t *rpt0_b = (uint32_t *)(((uint8_t *)oa_report0) +
>>  					format.b_off);
>>@@ -890,10 +920,10 @@ gen8_sanity_check_test_oa_reports(const uint32_t *oa_report0,
>>  		  gen8_read_report_reason(oa_report0),
>>  		  gen8_read_report_reason(oa_report1));
>>-	freq = time_delta ? ((uint64_t)clock_delta * 1000) / time_delta : 0;
>>+	freq = time_delta ? (clock_delta * 1000) / time_delta : 0;
>>  	igt_debug("freq = %"PRIu64"\n", freq);
>>-	igt_debug("clock delta = %"PRIu32"\n", clock_delta);
>>+	igt_debug("clock delta = %"PRIu64"\n", clock_delta);
>>  	max_delta = clock_delta * intel_perf->devinfo.n_eus;
>>@@ -994,7 +1024,7 @@ gen8_sanity_check_test_oa_reports(const uint32_t *oa_report0,
>>  					    format.c_off);
>>  		uint32_t delta = c1[j] - c0[j];
>>-		igt_debug("C%d: delta = %"PRIu32", max_delta=%"PRIu32"\n",
>>+		igt_debug("C%d: delta = %"PRIu32", max_delta=%"PRIu64"\n",
>>  			  j, delta, max_delta);
>>  		igt_assert(delta <= max_delta);
>>  	}
>>@@ -1440,10 +1470,10 @@ print_reports(uint32_t *oa_report0, uint32_t *oa_report1, int fmt)
>>  	if (IS_HASWELL(devid) && format.n_c == 0) {
>>  		igt_debug("CLOCK = N/A\n");
>>  	} else {
>>-		uint32_t clock0 = read_report_ticks(oa_report0, fmt);
>>-		uint32_t clock1 = read_report_ticks(oa_report1, fmt);
>>+		uint64_t clock0 = read_report_ticks(oa_report0, fmt);
>>+		uint64_t clock1 = read_report_ticks(oa_report1, fmt);
>>-		igt_debug("CLOCK: 1st = %"PRIu32", 2nd = %"PRIu32", delta = %"PRIu32"\n",
>>+		igt_debug("CLOCK: 1st = %"PRIu64", 2nd = %"PRIu64", delta = %"PRIu64"\n",
>>  			  clock0, clock1, clock1 - clock0);
>>  	}
>>@@ -1545,9 +1575,9 @@ print_report(uint32_t *report, int fmt)
>>  	if (IS_HASWELL(devid) && format.n_c == 0) {
>>  		igt_debug("CLOCK = N/A\n");
>>  	} else {
>>-		uint32_t clock = read_report_ticks(report, fmt);
>>+		uint64_t clock = read_report_ticks(report, fmt);
>>-		igt_debug("CLOCK: %"PRIu32"\n", clock);
>>+		igt_debug("CLOCK: %"PRIu64"\n", clock);
>>  	}
>>  	if (intel_gen(devid) >= 8) {
>
>

Re: [igt-dev] [PATCH i-g-t 16/23] i915/perf: Treat ticks as 64 bit

[Lionel Landwerlin]

On 06/09/2022 21:56, Umesh Nerlige Ramappa wrote:
> On Tue, Sep 06, 2022 at 05:08:50PM +0300, Lionel Landwerlin wrote:
>> On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
>>> DG2 A0 has a buggy header in the 64 bit OAG format. Except the ticks,
>>> all other fields are 32 bit. To resolve this, we will store ticks in 64
>>> bit type.
>>>
>>> Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
>>> ---
>>>  tests/i915/perf.c | 82 ++++++++++++++++++++++++++++++++---------------
>>>  1 file changed, 56 insertions(+), 26 deletions(-)
>>>
>>> diff --git a/tests/i915/perf.c b/tests/i915/perf.c
>>> index 3eb31f3b..8836da66 100644
>>> --- a/tests/i915/perf.c
>>> +++ b/tests/i915/perf.c
>>> @@ -117,6 +117,7 @@ struct oa_format {
>>>      int c_off;
>>>      int n_c;
>>>      int oa_type;
>>> +    bool report_hdr_64bit;
>>>  };
>>>  static struct oa_format hsw_oa_formats[I915_OA_FORMAT_MAX] = {
>>> @@ -203,7 +204,8 @@ static struct oa_format 
>>> dg2_oa_formats[I915_OA_FORMAT_MAX] = {
>>>          "A38u64_R2u64_B8_C8", .size = 448,
>>>          .a64_off = 32, .n_a64 = 38,
>>>          .b_off = 352, .n_b = 8,
>>> -        .c_off = 384, .n_c = 8, .oa_type = OAG, },
>>> +        .c_off = 384, .n_c = 8, .oa_type = OAG,
>>> +        .report_hdr_64bit = true, },
>>>      [I915_OAR_FORMAT_A36u64_B8_C8] = {
>>>          "A36u64_B8_C8", .size = 384,
>>>          .a64_off = 32, .n_a64 = 36,
>>> @@ -245,7 +247,7 @@ static bool *undefined_a_counters;
>>>  static uint64_t oa_exp_1_millisec;
>>>  static igt_render_copyfunc_t render_copy = NULL;
>>> -static uint32_t (*read_report_ticks)(const uint32_t *report,
>>> +static uint64_t (*read_report_ticks)(const uint32_t *report,
>>>                       enum drm_i915_oa_format format);
>>>  static void (*sanity_check_reports)(const uint32_t *oa_report0,
>>>                      const uint32_t *oa_report1,
>>> @@ -395,7 +397,7 @@ sysfs_read(enum i915_attr_id id)
>>>   * C2 corresponds to a clock counter for the Haswell render basic 
>>> metric set
>>>   * but it's not included in all of the formats.
>>>   */
>>> -static uint32_t
>>> +static uint64_t
>>>  hsw_read_report_ticks(const uint32_t *report, enum 
>>> drm_i915_oa_format format)
>>>  {
>>>      uint32_t *c = (uint32_t *)(((uint8_t *)report) + 
>>> get_oa_format(format).c_off);
>>> @@ -405,10 +407,41 @@ hsw_read_report_ticks(const uint32_t *report, 
>>> enum drm_i915_oa_format format)
>>>      return c[2];
>>>  }
>>> -static uint32_t
>>> +static uint64_t
>>>  gen8_read_report_ticks(const uint32_t *report, enum 
>>> drm_i915_oa_format format)
>>>  {
>>> -    return report[3];
>>> +
>>> +    struct oa_format fmt = get_oa_format(format);
>>> +
>>> +    return fmt.report_hdr_64bit ? *(uint64_t *)&report[6] : report[3];
>>> +}
>>> +
>>> +/*
>>> + * t0 is a value sampled before t1. width is number of bits used to 
>>> represent
>>> + * t0/t1. Normally t1 is greater than t0. In cases where t1 < t0 
>>> use this
>>> + * helper. Since the size of t1/t0 is already 64 bits, no special 
>>> handling is
>>> + * needed for width = 64.
>>> + */
>>> +static uint64_t
>>> +__elapsed_delta(uint64_t t1, uint64_t t0, uint32_t width)
>>> +{
>>> +    uint32_t max_bits = sizeof(t1) * 8;
>>> +
>>> +    igt_assert(width <= max_bits);
>>> +
>>> +    if (t1 < t0 && width != max_bits)
>>> +        return ((1ULL << width) - t0) + t1;
>>
>>
>> Should this be : ((1ULL << width) + t1) - t0 ?
>>
>
> Both ways should yield the same result.


My bad :


Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>


>
>>
>> Since we know the t1 wrapped around the counter's max bit size, the 
>> actual difference is the max value + whatever is in the counter minus 
>> t0.
>>
>>
>>> +
>>> +    return t1 - t0;
>>> +}
>>> +
>>> +static uint64_t
>>> +oa_tick_delta(const uint32_t *report1,
>>> +          const uint32_t *report0,
>>> +          enum drm_i915_oa_format format)
>>> +{
>>> +    return __elapsed_delta(read_report_ticks(report1, format),
>>> +                   read_report_ticks(report0, format), 32);
>>>  }
>>>  static void
>>> @@ -638,8 +671,8 @@ hsw_sanity_check_render_basic_reports(const 
>>> uint32_t *oa_report0,
>>>                        enum drm_i915_oa_format fmt)
>>>  {
>>>      uint32_t time_delta = timebase_scale(oa_report1[1] - 
>>> oa_report0[1]);
>>> -    uint32_t clock_delta;
>>> -    uint32_t max_delta;
>>> +    uint64_t clock_delta;
>>> +    uint64_t max_delta;
>>>      struct oa_format format = get_oa_format(fmt);
>>>      igt_assert_neq(time_delta, 0);
>>> @@ -654,21 +687,19 @@ hsw_sanity_check_render_basic_reports(const 
>>> uint32_t *oa_report0,
>>>          clock_delta = (gt_max_freq_mhz *
>>>                     (uint64_t)time_delta) / 1000;
>>>      } else {
>>> -        uint32_t ticks0 = read_report_ticks(oa_report0, fmt);
>>> -        uint32_t ticks1 = read_report_ticks(oa_report1, fmt);
>>>          uint64_t freq;
>>> -        clock_delta = ticks1 - ticks0;
>>> +        clock_delta = oa_tick_delta(oa_report1, oa_report0, fmt);
>>> -        igt_assert_neq(clock_delta, 0);
>>> +        igt_assert_neq_u64(clock_delta, 0);
>>> -        freq = ((uint64_t)clock_delta * 1000) / time_delta;
>>> +        freq = (clock_delta * 1000) / time_delta;
>>>          igt_debug("freq = %"PRIu64"\n", freq);
>>>          igt_assert(freq <= gt_max_freq_mhz);
>>>      }
>>> -    igt_debug("clock delta = %"PRIu32"\n", clock_delta);
>>> +    igt_debug("clock delta = %"PRIu64"\n", clock_delta);
>>>      /* The maximum rate for any HSW counter =
>>>       *   clock_delta * N EUs
>>> @@ -801,7 +832,8 @@ accumulate_reports(struct accumulator *accumulator,
>>>          accumulate_uint32(4, start, end, deltas + idx++);
>>>          /* clock cycles */
>>> -        accumulate_uint32(12, start, end, deltas + idx++);
>>> +        deltas[idx] += oa_tick_delta(end, start, accumulator->format);
>>> +        idx++;
>>>      } else {
>>>          /* timestamp */
>>>          accumulate_uint32(4, start, end, deltas + idx++);
>>> @@ -874,10 +906,8 @@ gen8_sanity_check_test_oa_reports(const 
>>> uint32_t *oa_report0,
>>>  {
>>>      struct oa_format format = get_oa_format(fmt);
>>>      uint32_t time_delta = timebase_scale(oa_report1[1] - 
>>> oa_report0[1]);
>>> -    uint32_t ticks0 = read_report_ticks(oa_report0, fmt);
>>> -    uint32_t ticks1 = read_report_ticks(oa_report1, fmt);
>>> -    uint32_t clock_delta = ticks1 - ticks0;
>>> -    uint32_t max_delta;
>>> +    uint64_t clock_delta = oa_tick_delta(oa_report1, oa_report0, fmt);
>>> +    uint64_t max_delta;
>>>      uint64_t freq;
>>>      uint32_t *rpt0_b = (uint32_t *)(((uint8_t *)oa_report0) +
>>>                      format.b_off);
>>> @@ -890,10 +920,10 @@ gen8_sanity_check_test_oa_reports(const 
>>> uint32_t *oa_report0,
>>>            gen8_read_report_reason(oa_report0),
>>>            gen8_read_report_reason(oa_report1));
>>> -    freq = time_delta ? ((uint64_t)clock_delta * 1000) / time_delta 
>>> : 0;
>>> +    freq = time_delta ? (clock_delta * 1000) / time_delta : 0;
>>>      igt_debug("freq = %"PRIu64"\n", freq);
>>> -    igt_debug("clock delta = %"PRIu32"\n", clock_delta);
>>> +    igt_debug("clock delta = %"PRIu64"\n", clock_delta);
>>>      max_delta = clock_delta * intel_perf->devinfo.n_eus;
>>> @@ -994,7 +1024,7 @@ gen8_sanity_check_test_oa_reports(const 
>>> uint32_t *oa_report0,
>>>                          format.c_off);
>>>          uint32_t delta = c1[j] - c0[j];
>>> -        igt_debug("C%d: delta = %"PRIu32", max_delta=%"PRIu32"\n",
>>> +        igt_debug("C%d: delta = %"PRIu32", max_delta=%"PRIu64"\n",
>>>                j, delta, max_delta);
>>>          igt_assert(delta <= max_delta);
>>>      }
>>> @@ -1440,10 +1470,10 @@ print_reports(uint32_t *oa_report0, uint32_t 
>>> *oa_report1, int fmt)
>>>      if (IS_HASWELL(devid) && format.n_c == 0) {
>>>          igt_debug("CLOCK = N/A\n");
>>>      } else {
>>> -        uint32_t clock0 = read_report_ticks(oa_report0, fmt);
>>> -        uint32_t clock1 = read_report_ticks(oa_report1, fmt);
>>> +        uint64_t clock0 = read_report_ticks(oa_report0, fmt);
>>> +        uint64_t clock1 = read_report_ticks(oa_report1, fmt);
>>> -        igt_debug("CLOCK: 1st = %"PRIu32", 2nd = %"PRIu32", delta = 
>>> %"PRIu32"\n",
>>> +        igt_debug("CLOCK: 1st = %"PRIu64", 2nd = %"PRIu64", delta = 
>>> %"PRIu64"\n",
>>>                clock0, clock1, clock1 - clock0);
>>>      }
>>> @@ -1545,9 +1575,9 @@ print_report(uint32_t *report, int fmt)
>>>      if (IS_HASWELL(devid) && format.n_c == 0) {
>>>          igt_debug("CLOCK = N/A\n");
>>>      } else {
>>> -        uint32_t clock = read_report_ticks(report, fmt);
>>> +        uint64_t clock = read_report_ticks(report, fmt);
>>> -        igt_debug("CLOCK: %"PRIu32"\n", clock);
>>> +        igt_debug("CLOCK: %"PRIu64"\n", clock);
>>>      }
>>>      if (intel_gen(devid) >= 8) {
>>
>>

[igt-dev] [PATCH i-g-t 17/23] i915/perf: Treat timestamp as 64 bit value

[Umesh Nerlige Ramappa]

New OA formats in DG2 have 64 bit OA report header. Update the timestamp
in the tests to be dafault 64 bit.

Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
---
 tests/i915/perf.c | 133 +++++++++++++++++++++++++++-------------------
 1 file changed, 79 insertions(+), 54 deletions(-)

diff --git a/tests/i915/perf.c b/tests/i915/perf.c
index 8836da66..52de5d38 100644
--- a/tests/i915/perf.c
+++ b/tests/i915/perf.c
@@ -510,9 +510,28 @@ cs_timebase_scale(uint32_t u32_delta)
 }
 
 static uint64_t
-timebase_scale(uint32_t u32_delta)
+oa_timestamp(const uint32_t *report, enum drm_i915_oa_format format)
 {
-	return ((uint64_t)u32_delta * NSEC_PER_SEC) / intel_perf->devinfo.timestamp_frequency;
+	struct oa_format fmt = get_oa_format(format);
+
+	return fmt.report_hdr_64bit ? *(uint64_t *)&report[2] : report[1];
+}
+
+static uint64_t
+oa_timestamp_delta(const uint32_t *report1,
+		   const uint32_t *report0,
+		   enum drm_i915_oa_format format)
+{
+	uint32_t width = intel_graphics_ver(devid) >= IP_VER(12, 55) ? 56 : 32;
+
+	return __elapsed_delta(oa_timestamp(report1, format),
+			       oa_timestamp(report0, format), width);
+}
+
+static uint64_t
+timebase_scale(uint64_t delta)
+{
+	return (delta * NSEC_PER_SEC) / intel_perf->devinfo.timestamp_frequency;
 }
 
 /* Returns: the largest OA exponent that will still result in a sampling period
@@ -521,7 +540,7 @@ timebase_scale(uint32_t u32_delta)
 static int
 max_oa_exponent_for_period_lte(uint64_t period)
 {
-	/* NB: timebase_scale() takes a uint32_t and an exponent of 30
+	/* NB: timebase_scale() takes a uint64_t and an exponent of 30
 	 * would already represent a period of ~3 minutes so there's
 	 * really no need to consider higher exponents.
 	 */
@@ -670,12 +689,14 @@ hsw_sanity_check_render_basic_reports(const uint32_t *oa_report0,
 				      const uint32_t *oa_report1,
 				      enum drm_i915_oa_format fmt)
 {
-	uint32_t time_delta = timebase_scale(oa_report1[1] - oa_report0[1]);
+	uint64_t time_delta = timebase_scale(oa_timestamp_delta(oa_report1,
+								oa_report0,
+								fmt));
 	uint64_t clock_delta;
 	uint64_t max_delta;
 	struct oa_format format = get_oa_format(fmt);
 
-	igt_assert_neq(time_delta, 0);
+	igt_assert_neq_u64(time_delta, 0);
 
 	/* As a special case we have to consider that on Haswell we
 	 * can't explicitly derive a clock delta for all OA report
@@ -684,8 +705,7 @@ hsw_sanity_check_render_basic_reports(const uint32_t *oa_report0,
 	if (format.n_c == 0) {
 		/* Assume running at max freq for sake of
 		 * below sanity check on counters... */
-		clock_delta = (gt_max_freq_mhz *
-			       (uint64_t)time_delta) / 1000;
+		clock_delta = (gt_max_freq_mhz * time_delta) / 1000;
 	} else {
 		uint64_t freq;
 
@@ -829,7 +849,8 @@ accumulate_reports(struct accumulator *accumulator,
 
 	if (intel_gen(devid) >= 8) {
 		/* timestamp */
-		accumulate_uint32(4, start, end, deltas + idx++);
+		deltas[idx] += oa_timestamp_delta(end, start, accumulator->format);
+		idx++;
 
 		/* clock cycles */
 		deltas[idx] += oa_tick_delta(end, start, accumulator->format);
@@ -905,7 +926,9 @@ gen8_sanity_check_test_oa_reports(const uint32_t *oa_report0,
 				  enum drm_i915_oa_format fmt)
 {
 	struct oa_format format = get_oa_format(fmt);
-	uint32_t time_delta = timebase_scale(oa_report1[1] - oa_report0[1]);
+	uint64_t time_delta = timebase_scale(oa_timestamp_delta(oa_report1,
+								oa_report0,
+								fmt));
 	uint64_t clock_delta = oa_tick_delta(oa_report1, oa_report0, fmt);
 	uint64_t max_delta;
 	uint64_t freq;
@@ -1093,11 +1116,11 @@ static int
 i915_read_reports_until_timestamp(enum drm_i915_oa_format oa_format,
 				  uint8_t *buf,
 				  uint32_t max_size,
-				  uint32_t start_timestamp,
-				  uint32_t end_timestamp)
+				  uint64_t start_timestamp,
+				  uint64_t end_timestamp)
 {
 	size_t format_size = get_oa_format(oa_format).size;
-	uint32_t last_seen_timestamp = start_timestamp;
+	uint64_t last_seen_timestamp = start_timestamp;
 	int total_len = 0;
 
 	while (last_seen_timestamp < end_timestamp) {
@@ -1106,7 +1129,7 @@ i915_read_reports_until_timestamp(enum drm_i915_oa_format oa_format,
 		/* Running out of space. */
 		if ((max_size - total_len) < format_size) {
 			igt_warn("run out of space before reaching "
-				 "end timestamp (%u/%u)\n",
+				 "end timestamp (%"PRIu64"/%"PRIu64")\n",
 				 last_seen_timestamp, end_timestamp);
 			return -1;
 		}
@@ -1135,7 +1158,7 @@ i915_read_reports_until_timestamp(enum drm_i915_oa_format oa_format,
 			uint32_t *report = (uint32_t *) (header + 1);
 
 			if (header->type == DRM_I915_PERF_RECORD_SAMPLE)
-				last_seen_timestamp = report[1];
+				last_seen_timestamp = oa_timestamp(report, oa_format);
 
 			offset += header->size;
 		}
@@ -1400,11 +1423,11 @@ read_2_oa_reports(int format_id,
 			report = (const void *)(header + 1);
 			dump_report(report, format_size / 4, "oa-formats");
 
-			igt_debug("read report: reason = %x, timestamp = %x, exponent mask=%x\n",
-				  report[0], report[1], exponent_mask);
+			igt_debug("read report: reason = %x, timestamp = %"PRIx64", exponent mask=%x\n",
+				  report[0], oa_timestamp(report, format_id), exponent_mask);
 
 			/* Don't expect zero for timestamps */
-			igt_assert_neq(report[1], 0);
+			igt_assert_neq_u64(oa_timestamp(report, format_id), 0);
 
 			if (timer_only) {
 				if (!oa_report_is_periodic(exponent, report)) {
@@ -1463,9 +1486,11 @@ static void
 print_reports(uint32_t *oa_report0, uint32_t *oa_report1, int fmt)
 {
 	struct oa_format format = get_oa_format(fmt);
+	uint64_t ts0 = oa_timestamp(oa_report0, fmt);
+	uint64_t ts1 = oa_timestamp(oa_report1, fmt);
 
-	igt_debug("TIMESTAMP: 1st = %"PRIu32", 2nd = %"PRIu32", delta = %"PRIu32"\n",
-		  oa_report0[1], oa_report1[1], oa_report1[1] - oa_report0[1]);
+	igt_debug("TIMESTAMP: 1st = %"PRIu64", 2nd = %"PRIu64", delta = %"PRIu64"\n",
+		  ts0, ts1, ts1 - ts0);
 
 	if (IS_HASWELL(devid) && format.n_c == 0) {
 		igt_debug("CLOCK = N/A\n");
@@ -1570,7 +1595,7 @@ print_report(uint32_t *report, int fmt)
 {
 	struct oa_format format = get_oa_format(fmt);
 
-	igt_debug("TIMESTAMP: %"PRIu32"\n", report[1]);
+	igt_debug("TIMESTAMP: %"PRIu64"\n", oa_timestamp(report, fmt));
 
 	if (IS_HASWELL(devid) && format.n_c == 0) {
 		igt_debug("CLOCK = N/A\n");
@@ -1885,15 +1910,16 @@ test_oa_exponents(void)
 
 		__perf_close(stream_fd);
 
-		igt_debug("report%04i ts=%08x hw_id=0x%08x\n", 0,
-			  timer_reports[0].report[1],
+		igt_debug("report%04i ts=%"PRIx64" hw_id=0x%08x\n", 0,
+			  oa_timestamp(timer_reports[0].report, test_set->perf_oa_format),
 			  oa_report_get_ctx_id(timer_reports[0].report));
 		for (int i = 1; i < n_timer_reports; i++) {
-			uint32_t delta =
-				timer_reports[i].report[1] - timer_reports[i - 1].report[1];
+			uint64_t delta = oa_timestamp_delta(timer_reports[i].report,
+							    timer_reports[i - 1].report,
+							    test_set->perf_oa_format);
 
-			igt_debug("report%04i ts=%08x hw_id=0x%08x delta=%u %s\n", i,
-				  timer_reports[i].report[1],
+			igt_debug("report%04i ts=%"PRIx64" hw_id=0x%08x delta=%"PRIu64" %s\n", i,
+				  oa_timestamp(timer_reports[i].report, test_set->perf_oa_format),
 				  oa_report_get_ctx_id(timer_reports[i].report),
 				  delta, expected_report_timing_delta(delta,
 								      expected_timestamp_delta) ? "" : "******");
@@ -2694,17 +2720,17 @@ test_buffer_fill(void)
 					igt_debug("report loss, trying again\n");
 					break;
 				case DRM_I915_PERF_RECORD_SAMPLE:
-					igt_debug(" > report ts=%u"
-						  " ts_delta_last_periodic=%8u is_timer=%i ctx_id=%8x nb_periodic=%u\n",
-						  report[1],
-						  n_periodic_reports > 0 ? report[1] - last_periodic_report[1] : 0,
+					igt_debug(" > report ts=%"PRIu64""
+						  " ts_delta_last_periodic=%"PRIu64" is_timer=%i ctx_id=%8x nb_periodic=%u\n",
+						  oa_timestamp(report, test_set->perf_oa_format),
+						  n_periodic_reports > 0 ? oa_timestamp_delta(report, last_periodic_report, test_set->perf_oa_format) : 0,
 						  oa_report_is_periodic(oa_exponent, report),
 						  oa_report_get_ctx_id(report),
 						  n_periodic_reports);
 
 					if (first_timestamp == 0)
-						first_timestamp = report[1];
-					last_timestamp = report[1];
+						first_timestamp = oa_timestamp(report, test_set->perf_oa_format);
+					last_timestamp = oa_timestamp(report, test_set->perf_oa_format);
 
 					if (((last_timestamp - first_timestamp) * oa_period) < (fill_duration / 2))
 						break;
@@ -2918,12 +2944,12 @@ test_enable_disable(void)
 						first_timestamp = report[1];
 					last_timestamp = report[1];
 
-					igt_debug(" > report ts=%8x"
-						  " ts_delta_last_periodic=%s%8u"
+					igt_debug(" > report ts=%"PRIx64""
+						  " ts_delta_last_periodic=%s%"PRIu64""
 						  " is_timer=%i ctx_id=0x%8x\n",
-						  report[1],
+						  oa_timestamp(report, test_set->perf_oa_format),
 						  oa_report_is_periodic(oa_exponent, report) ? " " : "*",
-						  n_periodic_reports > 0 ? (report[1] - last_periodic_report[1]) : 0,
+						  n_periodic_reports > 0 ?  oa_timestamp_delta(report, last_periodic_report, test_set->perf_oa_format) : 0,
 						  oa_report_is_periodic(oa_exponent, report),
 						  oa_report_get_ctx_id(report));
 
@@ -3254,7 +3280,7 @@ gen12_test_mi_rpc(void)
 	 * size amount of data was written.
 	 */
 	igt_assert_eq(report32[0], REPORT_ID);
-	igt_assert_neq(report32[1], 0);
+	igt_assert(oa_timestamp(report32, test_set->perf_oa_format));
 	igt_assert_neq(report32[format.b_off >> 2], 0x80808080);
 	igt_assert_eq(report32[format_size_32], 0x80808080);
 
@@ -3315,7 +3341,7 @@ test_mi_rpc(void)
 	report32 = buf->ptr;
 	dump_report(report32, 64, "mi-rpc");
 	igt_assert_eq(report32[0], 0xdeadbeef); /* report ID */
-	igt_assert_neq(report32[1], 0); /* timestamp */
+	igt_assert(oa_timestamp(report32, test_set->perf_oa_format)); /* timestamp */
 
 	igt_assert_neq(report32[63], 0x80808080); /* end of report */
 	igt_assert_eq(report32[64], 0x80808080); /* after 256 byte report */
@@ -4050,9 +4076,9 @@ static void gen12_single_ctx_helper(void)
 	uint32_t context0_id, context1_id;
 	uint32_t *report0_32, *report1_32, *report2_32, *report3_32;
 	uint64_t timestamp0_64, timestamp1_64;
-	uint32_t delta_ts64, delta_oa32;
+	uint64_t delta_ts64, delta_oa32;
 	uint64_t delta_ts64_ns, delta_oa32_ns;
-	uint32_t delta_delta;
+	uint64_t delta_delta;
 	int width = 800;
 	int height = 600;
 #define INVALID_CTX_ID 0xffffffff
@@ -4060,7 +4086,7 @@ static void gen12_single_ctx_helper(void)
 	uint32_t ctx1_id = INVALID_CTX_ID;
 	int ret;
 	struct accumulator accumulator = {
-		.format = test_set->perf_oa_format
+		.format = fmt
 	};
 
 	bops = buf_ops_create(drm_fd);
@@ -4207,14 +4233,14 @@ static void gen12_single_ctx_helper(void)
 	 */
 	report0_32 = dst_buf->ptr;
 	igt_assert_eq(report0_32[0], 0xdeadbeef);
-	igt_assert_neq(report0_32[1], 0);
+	igt_assert(oa_timestamp(report0_32, fmt));
 	ctx0_id = report0_32[2];
 	igt_debug("MI_RPC(start) CTX ID: %u\n", ctx0_id);
 	dump_report(report0_32, 64, "report0_32");
 
 	report1_32 = report0_32 + 64;
 	igt_assert_eq(report1_32[0], 0xbeefbeef);
-	igt_assert_neq(report1_32[1], 0);
+	igt_assert(oa_timestamp(report1_32, fmt));
 	ctx1_id = report1_32[2];
 	igt_debug("CTX ID1: %u\n", ctx1_id);
 	dump_report(report1_32, 64, "report1_32");
@@ -4222,7 +4248,7 @@ static void gen12_single_ctx_helper(void)
 	/* Verify that counters in context1 are all zeroes */
 	report2_32 = report0_32 + 128;
 	igt_assert_eq(report2_32[0], 0x00c0ffee);
-	igt_assert_neq(report2_32[1], 0);
+	igt_assert(oa_timestamp(report2_32, fmt));
 	dump_report(report2_32, 64, "report2_32");
 	igt_assert_eq(0, memcmp(&report2_32[4],
 				(uint8_t *) dst_buf->ptr + 2048,
@@ -4230,7 +4256,7 @@ static void gen12_single_ctx_helper(void)
 
 	report3_32 = report0_32 + 192;
 	igt_assert_eq(report3_32[0], 0x01c0ffee);
-	igt_assert_neq(report3_32[1], 0);
+	igt_assert(oa_timestamp(report3_32, fmt));
 	dump_report(report3_32, 64, "report3_32");
 	igt_assert_eq(0, memcmp(&report3_32[4],
 				(uint8_t *) dst_buf->ptr + 2048,
@@ -4244,8 +4270,8 @@ static void gen12_single_ctx_helper(void)
 			accumulator.deltas[2 + 21],
 			accumulator.deltas[2 + 26]);
 
-	igt_debug("oa_timestamp32 0 = %u\n", report0_32[1]);
-	igt_debug("oa_timestamp32 1 = %u\n", report1_32[1]);
+	igt_debug("oa_timestamp32 0 = %"PRIu64"\n", oa_timestamp(report0_32, fmt));
+	igt_debug("oa_timestamp32 1 = %"PRIu64"\n", oa_timestamp(report1_32, fmt));
 	igt_debug("ctx_id 0 = %u\n", report0_32[2]);
 	igt_debug("ctx_id 1 = %u\n", report1_32[2]);
 
@@ -4260,23 +4286,22 @@ static void gen12_single_ctx_helper(void)
 	igt_debug("ts_timestamp64 1 = %"PRIu64"\n", timestamp1_64);
 
 	delta_ts64 = timestamp1_64 - timestamp0_64;
-	delta_oa32 = report1_32[1] - report0_32[1];
+	delta_oa32 = oa_timestamp_delta(report1_32, report0_32, fmt);
 
 	/* Sanity check that we can pass the delta to timebase_scale */
-	igt_assert(delta_ts64 < UINT32_MAX);
 	delta_oa32_ns = timebase_scale(delta_oa32);
 	delta_ts64_ns = cs_timebase_scale(delta_ts64);
 
-	igt_debug("oa32 delta = %u, = %uns\n",
-			delta_oa32, (unsigned)delta_oa32_ns);
-	igt_debug("ts64 delta = %u, = %uns\n",
-			delta_ts64, (unsigned)delta_ts64_ns);
+	igt_debug("oa32 delta = %"PRIu64", = %"PRIu64"ns\n",
+			delta_oa32, delta_oa32_ns);
+	igt_debug("ts64 delta = %"PRIu64", = %"PRIu64"ns\n",
+			delta_ts64, delta_ts64_ns);
 
 	delta_delta = delta_ts64_ns > delta_oa32_ns ?
 		      (delta_ts64_ns - delta_oa32_ns) :
 		      (delta_oa32_ns - delta_ts64_ns);
 	if (delta_delta > 5000) {
-		igt_debug("delta_delta = %d. exceeds margin, skipping..\n",
+		igt_debug("delta_delta = %"PRIu64". exceeds margin, skipping..\n",
 			  delta_delta);
 		exit(EAGAIN);
 	}
-- 
2.25.1

Re: [igt-dev] [PATCH i-g-t 17/23] i915/perf: Treat timestamp as 64 bit value

[Lionel Landwerlin]

On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
> New OA formats in DG2 have 64 bit OA report header. Update the timestamp
> in the tests to be dafault 64 bit.
>
> Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
> ---
>   tests/i915/perf.c | 133 +++++++++++++++++++++++++++-------------------
>   1 file changed, 79 insertions(+), 54 deletions(-)
>
> diff --git a/tests/i915/perf.c b/tests/i915/perf.c
> index 8836da66..52de5d38 100644
> --- a/tests/i915/perf.c
> +++ b/tests/i915/perf.c
> @@ -510,9 +510,28 @@ cs_timebase_scale(uint32_t u32_delta)
>   }
>   
>   static uint64_t
> -timebase_scale(uint32_t u32_delta)
> +oa_timestamp(const uint32_t *report, enum drm_i915_oa_format format)
>   {
> -	return ((uint64_t)u32_delta * NSEC_PER_SEC) / intel_perf->devinfo.timestamp_frequency;
> +	struct oa_format fmt = get_oa_format(format);
> +
> +	return fmt.report_hdr_64bit ? *(uint64_t *)&report[2] : report[1];
> +}
> +
> +static uint64_t
> +oa_timestamp_delta(const uint32_t *report1,
> +		   const uint32_t *report0,
> +		   enum drm_i915_oa_format format)
> +{
> +	uint32_t width = intel_graphics_ver(devid) >= IP_VER(12, 55) ? 56 : 32;
> +
> +	return __elapsed_delta(oa_timestamp(report1, format),
> +			       oa_timestamp(report0, format), width);
> +}
> +
> +static uint64_t
> +timebase_scale(uint64_t delta)
> +{
> +	return (delta * NSEC_PER_SEC) / intel_perf->devinfo.timestamp_frequency;
>   }
>   
>   /* Returns: the largest OA exponent that will still result in a sampling period
> @@ -521,7 +540,7 @@ timebase_scale(uint32_t u32_delta)
>   static int
>   max_oa_exponent_for_period_lte(uint64_t period)
>   {
> -	/* NB: timebase_scale() takes a uint32_t and an exponent of 30
> +	/* NB: timebase_scale() takes a uint64_t and an exponent of 30
>   	 * would already represent a period of ~3 minutes so there's
>   	 * really no need to consider higher exponents.
>   	 */
> @@ -670,12 +689,14 @@ hsw_sanity_check_render_basic_reports(const uint32_t *oa_report0,
>   				      const uint32_t *oa_report1,
>   				      enum drm_i915_oa_format fmt)
>   {
> -	uint32_t time_delta = timebase_scale(oa_report1[1] - oa_report0[1]);
> +	uint64_t time_delta = timebase_scale(oa_timestamp_delta(oa_report1,
> +								oa_report0,
> +								fmt));
>   	uint64_t clock_delta;
>   	uint64_t max_delta;
>   	struct oa_format format = get_oa_format(fmt);
>   
> -	igt_assert_neq(time_delta, 0);
> +	igt_assert_neq_u64(time_delta, 0);
>   
>   	/* As a special case we have to consider that on Haswell we
>   	 * can't explicitly derive a clock delta for all OA report
> @@ -684,8 +705,7 @@ hsw_sanity_check_render_basic_reports(const uint32_t *oa_report0,
>   	if (format.n_c == 0) {
>   		/* Assume running at max freq for sake of
>   		 * below sanity check on counters... */
> -		clock_delta = (gt_max_freq_mhz *
> -			       (uint64_t)time_delta) / 1000;
> +		clock_delta = (gt_max_freq_mhz * time_delta) / 1000;
>   	} else {
>   		uint64_t freq;
>   
> @@ -829,7 +849,8 @@ accumulate_reports(struct accumulator *accumulator,
>   
>   	if (intel_gen(devid) >= 8) {
>   		/* timestamp */
> -		accumulate_uint32(4, start, end, deltas + idx++);
> +		deltas[idx] += oa_timestamp_delta(end, start, accumulator->format);
> +		idx++;
>   
>   		/* clock cycles */
>   		deltas[idx] += oa_tick_delta(end, start, accumulator->format);
> @@ -905,7 +926,9 @@ gen8_sanity_check_test_oa_reports(const uint32_t *oa_report0,
>   				  enum drm_i915_oa_format fmt)
>   {
>   	struct oa_format format = get_oa_format(fmt);
> -	uint32_t time_delta = timebase_scale(oa_report1[1] - oa_report0[1]);
> +	uint64_t time_delta = timebase_scale(oa_timestamp_delta(oa_report1,
> +								oa_report0,
> +								fmt));
>   	uint64_t clock_delta = oa_tick_delta(oa_report1, oa_report0, fmt);
>   	uint64_t max_delta;
>   	uint64_t freq;
> @@ -1093,11 +1116,11 @@ static int
>   i915_read_reports_until_timestamp(enum drm_i915_oa_format oa_format,
>   				  uint8_t *buf,
>   				  uint32_t max_size,
> -				  uint32_t start_timestamp,
> -				  uint32_t end_timestamp)
> +				  uint64_t start_timestamp,
> +				  uint64_t end_timestamp)
>   {
>   	size_t format_size = get_oa_format(oa_format).size;
> -	uint32_t last_seen_timestamp = start_timestamp;
> +	uint64_t last_seen_timestamp = start_timestamp;
>   	int total_len = 0;
>   
>   	while (last_seen_timestamp < end_timestamp) {
> @@ -1106,7 +1129,7 @@ i915_read_reports_until_timestamp(enum drm_i915_oa_format oa_format,
>   		/* Running out of space. */
>   		if ((max_size - total_len) < format_size) {
>   			igt_warn("run out of space before reaching "
> -				 "end timestamp (%u/%u)\n",
> +				 "end timestamp (%"PRIu64"/%"PRIu64")\n",
>   				 last_seen_timestamp, end_timestamp);
>   			return -1;
>   		}
> @@ -1135,7 +1158,7 @@ i915_read_reports_until_timestamp(enum drm_i915_oa_format oa_format,
>   			uint32_t *report = (uint32_t *) (header + 1);
>   
>   			if (header->type == DRM_I915_PERF_RECORD_SAMPLE)
> -				last_seen_timestamp = report[1];
> +				last_seen_timestamp = oa_timestamp(report, oa_format);
>   
>   			offset += header->size;
>   		}
> @@ -1400,11 +1423,11 @@ read_2_oa_reports(int format_id,
>   			report = (const void *)(header + 1);
>   			dump_report(report, format_size / 4, "oa-formats");
>   
> -			igt_debug("read report: reason = %x, timestamp = %x, exponent mask=%x\n",
> -				  report[0], report[1], exponent_mask);
> +			igt_debug("read report: reason = %x, timestamp = %"PRIx64", exponent mask=%x\n",
> +				  report[0], oa_timestamp(report, format_id), exponent_mask);
>   
>   			/* Don't expect zero for timestamps */
> -			igt_assert_neq(report[1], 0);
> +			igt_assert_neq_u64(oa_timestamp(report, format_id), 0);
>   
>   			if (timer_only) {
>   				if (!oa_report_is_periodic(exponent, report)) {
> @@ -1463,9 +1486,11 @@ static void
>   print_reports(uint32_t *oa_report0, uint32_t *oa_report1, int fmt)
>   {
>   	struct oa_format format = get_oa_format(fmt);
> +	uint64_t ts0 = oa_timestamp(oa_report0, fmt);
> +	uint64_t ts1 = oa_timestamp(oa_report1, fmt);
>   
> -	igt_debug("TIMESTAMP: 1st = %"PRIu32", 2nd = %"PRIu32", delta = %"PRIu32"\n",
> -		  oa_report0[1], oa_report1[1], oa_report1[1] - oa_report0[1]);
> +	igt_debug("TIMESTAMP: 1st = %"PRIu64", 2nd = %"PRIu64", delta = %"PRIu64"\n",
> +		  ts0, ts1, ts1 - ts0);
>   
>   	if (IS_HASWELL(devid) && format.n_c == 0) {
>   		igt_debug("CLOCK = N/A\n");
> @@ -1570,7 +1595,7 @@ print_report(uint32_t *report, int fmt)
>   {
>   	struct oa_format format = get_oa_format(fmt);
>   
> -	igt_debug("TIMESTAMP: %"PRIu32"\n", report[1]);
> +	igt_debug("TIMESTAMP: %"PRIu64"\n", oa_timestamp(report, fmt));
>   
>   	if (IS_HASWELL(devid) && format.n_c == 0) {
>   		igt_debug("CLOCK = N/A\n");
> @@ -1885,15 +1910,16 @@ test_oa_exponents(void)
>   
>   		__perf_close(stream_fd);
>   
> -		igt_debug("report%04i ts=%08x hw_id=0x%08x\n", 0,
> -			  timer_reports[0].report[1],
> +		igt_debug("report%04i ts=%"PRIx64" hw_id=0x%08x\n", 0,
> +			  oa_timestamp(timer_reports[0].report, test_set->perf_oa_format),
>   			  oa_report_get_ctx_id(timer_reports[0].report));
>   		for (int i = 1; i < n_timer_reports; i++) {
> -			uint32_t delta =
> -				timer_reports[i].report[1] - timer_reports[i - 1].report[1];
> +			uint64_t delta = oa_timestamp_delta(timer_reports[i].report,
> +							    timer_reports[i - 1].report,
> +							    test_set->perf_oa_format);
>   
> -			igt_debug("report%04i ts=%08x hw_id=0x%08x delta=%u %s\n", i,
> -				  timer_reports[i].report[1],
> +			igt_debug("report%04i ts=%"PRIx64" hw_id=0x%08x delta=%"PRIu64" %s\n", i,
> +				  oa_timestamp(timer_reports[i].report, test_set->perf_oa_format),
>   				  oa_report_get_ctx_id(timer_reports[i].report),
>   				  delta, expected_report_timing_delta(delta,
>   								      expected_timestamp_delta) ? "" : "******");
> @@ -2694,17 +2720,17 @@ test_buffer_fill(void)
>   					igt_debug("report loss, trying again\n");
>   					break;
>   				case DRM_I915_PERF_RECORD_SAMPLE:
> -					igt_debug(" > report ts=%u"
> -						  " ts_delta_last_periodic=%8u is_timer=%i ctx_id=%8x nb_periodic=%u\n",
> -						  report[1],
> -						  n_periodic_reports > 0 ? report[1] - last_periodic_report[1] : 0,
> +					igt_debug(" > report ts=%"PRIu64""
> +						  " ts_delta_last_periodic=%"PRIu64" is_timer=%i ctx_id=%8x nb_periodic=%u\n",
> +						  oa_timestamp(report, test_set->perf_oa_format),
> +						  n_periodic_reports > 0 ? oa_timestamp_delta(report, last_periodic_report, test_set->perf_oa_format) : 0,
>   						  oa_report_is_periodic(oa_exponent, report),
>   						  oa_report_get_ctx_id(report),
>   						  n_periodic_reports);
>   
>   					if (first_timestamp == 0)
> -						first_timestamp = report[1];
> -					last_timestamp = report[1];
> +						first_timestamp = oa_timestamp(report, test_set->perf_oa_format);
> +					last_timestamp = oa_timestamp(report, test_set->perf_oa_format);
>   
>   					if (((last_timestamp - first_timestamp) * oa_period) < (fill_duration / 2))
>   						break;
> @@ -2918,12 +2944,12 @@ test_enable_disable(void)
>   						first_timestamp = report[1];
>   					last_timestamp = report[1];
>   
> -					igt_debug(" > report ts=%8x"
> -						  " ts_delta_last_periodic=%s%8u"
> +					igt_debug(" > report ts=%"PRIx64""
> +						  " ts_delta_last_periodic=%s%"PRIu64""
>   						  " is_timer=%i ctx_id=0x%8x\n",
> -						  report[1],
> +						  oa_timestamp(report, test_set->perf_oa_format),
>   						  oa_report_is_periodic(oa_exponent, report) ? " " : "*",
> -						  n_periodic_reports > 0 ? (report[1] - last_periodic_report[1]) : 0,
> +						  n_periodic_reports > 0 ?  oa_timestamp_delta(report, last_periodic_report, test_set->perf_oa_format) : 0,
>   						  oa_report_is_periodic(oa_exponent, report),
>   						  oa_report_get_ctx_id(report));
>   
> @@ -3254,7 +3280,7 @@ gen12_test_mi_rpc(void)
>   	 * size amount of data was written.
>   	 */
>   	igt_assert_eq(report32[0], REPORT_ID);
> -	igt_assert_neq(report32[1], 0);
> +	igt_assert(oa_timestamp(report32, test_set->perf_oa_format));
>   	igt_assert_neq(report32[format.b_off >> 2], 0x80808080);
>   	igt_assert_eq(report32[format_size_32], 0x80808080);
>   
> @@ -3315,7 +3341,7 @@ test_mi_rpc(void)
>   	report32 = buf->ptr;
>   	dump_report(report32, 64, "mi-rpc");
>   	igt_assert_eq(report32[0], 0xdeadbeef); /* report ID */
> -	igt_assert_neq(report32[1], 0); /* timestamp */
> +	igt_assert(oa_timestamp(report32, test_set->perf_oa_format)); /* timestamp */
>   
>   	igt_assert_neq(report32[63], 0x80808080); /* end of report */
>   	igt_assert_eq(report32[64], 0x80808080); /* after 256 byte report */
> @@ -4050,9 +4076,9 @@ static void gen12_single_ctx_helper(void)
>   	uint32_t context0_id, context1_id;
>   	uint32_t *report0_32, *report1_32, *report2_32, *report3_32;
>   	uint64_t timestamp0_64, timestamp1_64;
> -	uint32_t delta_ts64, delta_oa32;
> +	uint64_t delta_ts64, delta_oa32;
>   	uint64_t delta_ts64_ns, delta_oa32_ns;
> -	uint32_t delta_delta;
> +	uint64_t delta_delta;
>   	int width = 800;
>   	int height = 600;
>   #define INVALID_CTX_ID 0xffffffff
> @@ -4060,7 +4086,7 @@ static void gen12_single_ctx_helper(void)
>   	uint32_t ctx1_id = INVALID_CTX_ID;
>   	int ret;
>   	struct accumulator accumulator = {
> -		.format = test_set->perf_oa_format
> +		.format = fmt
>   	};
>   
>   	bops = buf_ops_create(drm_fd);
> @@ -4207,14 +4233,14 @@ static void gen12_single_ctx_helper(void)
>   	 */
>   	report0_32 = dst_buf->ptr;
>   	igt_assert_eq(report0_32[0], 0xdeadbeef);
> -	igt_assert_neq(report0_32[1], 0);
> +	igt_assert(oa_timestamp(report0_32, fmt));
>   	ctx0_id = report0_32[2];
>   	igt_debug("MI_RPC(start) CTX ID: %u\n", ctx0_id);
>   	dump_report(report0_32, 64, "report0_32");
>   
>   	report1_32 = report0_32 + 64;
>   	igt_assert_eq(report1_32[0], 0xbeefbeef);
> -	igt_assert_neq(report1_32[1], 0);
> +	igt_assert(oa_timestamp(report1_32, fmt));
>   	ctx1_id = report1_32[2];
>   	igt_debug("CTX ID1: %u\n", ctx1_id);
>   	dump_report(report1_32, 64, "report1_32");
> @@ -4222,7 +4248,7 @@ static void gen12_single_ctx_helper(void)
>   	/* Verify that counters in context1 are all zeroes */
>   	report2_32 = report0_32 + 128;
>   	igt_assert_eq(report2_32[0], 0x00c0ffee);
> -	igt_assert_neq(report2_32[1], 0);
> +	igt_assert(oa_timestamp(report2_32, fmt));
>   	dump_report(report2_32, 64, "report2_32");
>   	igt_assert_eq(0, memcmp(&report2_32[4],
>   				(uint8_t *) dst_buf->ptr + 2048,
> @@ -4230,7 +4256,7 @@ static void gen12_single_ctx_helper(void)
>   
>   	report3_32 = report0_32 + 192;
>   	igt_assert_eq(report3_32[0], 0x01c0ffee);
> -	igt_assert_neq(report3_32[1], 0);
> +	igt_assert(oa_timestamp(report3_32, fmt));
>   	dump_report(report3_32, 64, "report3_32");
>   	igt_assert_eq(0, memcmp(&report3_32[4],
>   				(uint8_t *) dst_buf->ptr + 2048,
> @@ -4244,8 +4270,8 @@ static void gen12_single_ctx_helper(void)
>   			accumulator.deltas[2 + 21],
>   			accumulator.deltas[2 + 26]);
>   
> -	igt_debug("oa_timestamp32 0 = %u\n", report0_32[1]);
> -	igt_debug("oa_timestamp32 1 = %u\n", report1_32[1]);
> +	igt_debug("oa_timestamp32 0 = %"PRIu64"\n", oa_timestamp(report0_32, fmt));
> +	igt_debug("oa_timestamp32 1 = %"PRIu64"\n", oa_timestamp(report1_32, fmt));
>   	igt_debug("ctx_id 0 = %u\n", report0_32[2]);
>   	igt_debug("ctx_id 1 = %u\n", report1_32[2]);
>   
> @@ -4260,23 +4286,22 @@ static void gen12_single_ctx_helper(void)
>   	igt_debug("ts_timestamp64 1 = %"PRIu64"\n", timestamp1_64);
>   
>   	delta_ts64 = timestamp1_64 - timestamp0_64;
> -	delta_oa32 = report1_32[1] - report0_32[1];
> +	delta_oa32 = oa_timestamp_delta(report1_32, report0_32, fmt);
>   
>   	/* Sanity check that we can pass the delta to timebase_scale */
> -	igt_assert(delta_ts64 < UINT32_MAX);
>   	delta_oa32_ns = timebase_scale(delta_oa32);
>   	delta_ts64_ns = cs_timebase_scale(delta_ts64);
>   
> -	igt_debug("oa32 delta = %u, = %uns\n",
> -			delta_oa32, (unsigned)delta_oa32_ns);
> -	igt_debug("ts64 delta = %u, = %uns\n",
> -			delta_ts64, (unsigned)delta_ts64_ns);
> +	igt_debug("oa32 delta = %"PRIu64", = %"PRIu64"ns\n",
> +			delta_oa32, delta_oa32_ns);
> +	igt_debug("ts64 delta = %"PRIu64", = %"PRIu64"ns\n",
> +			delta_ts64, delta_ts64_ns);
>   
>   	delta_delta = delta_ts64_ns > delta_oa32_ns ?
>   		      (delta_ts64_ns - delta_oa32_ns) :
>   		      (delta_oa32_ns - delta_ts64_ns);
>   	if (delta_delta > 5000) {
> -		igt_debug("delta_delta = %d. exceeds margin, skipping..\n",
> +		igt_debug("delta_delta = %"PRIu64". exceeds margin, skipping..\n",
>   			  delta_delta);
>   		exit(EAGAIN);
>   	}

[igt-dev] [PATCH i-g-t 18/23] i915/perf: Fix DG2 A0 report header

[Umesh Nerlige Ramappa]

Timestamp is still 32 bits in DG2 A0. Fix that. Specify variants of DG2
that have the OA format header bug

Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
---
 tests/i915/perf.c | 45 ++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 44 insertions(+), 1 deletion(-)

diff --git a/tests/i915/perf.c b/tests/i915/perf.c
index 52de5d38..6951735b 100644
--- a/tests/i915/perf.c
+++ b/tests/i915/perf.c
@@ -40,6 +40,7 @@
 
 #include "i915/gem.h"
 #include "i915/perf.h"
+#include "i915_pciids.h"
 #include "igt.h"
 #include "igt_perf.h"
 #include "igt_sysfs.h"
@@ -503,6 +504,47 @@ cs_timestamp_frequency(int fd)
 	return value;
 }
 
+/*
+ * Hacky ways until IGT is fully aware of steppings:
+ *
+ * WA is permanent for G11. For other DG2s, bug is fixed from B0 onwards.
+ * Steps A0 and A1 correspond to 0 and 1 revid values.
+ */
+struct perf_dev_info {
+	bool is_dg2_g11;
+};
+
+static const struct perf_dev_info dg2_g11_info = {
+	.is_dg2_g11 = true,
+};
+
+static const struct pci_id_match perf_dev_match[] = {
+	INTEL_DG2_G11_IDS(&dg2_g11_info),
+};
+
+static const struct perf_dev_info *__dev_info(uint32_t device_id)
+{
+	static const struct perf_dev_info *cache;
+	static int _iter;
+
+	if (_iter)
+		goto out;
+
+	for (_iter = 0; _iter < ARRAY_SIZE(perf_dev_match); _iter++)
+		if (device_id == perf_dev_match[_iter].device_id)
+			cache = (void *) perf_dev_match[_iter].match_data;
+
+out:
+	return cache;
+}
+
+#define IS_DG2_G11(__d) (__dev_info(__d) ? __dev_info(__d)->is_dg2_g11 : false)
+#define IS_DG2_A0(__d) (IS_DG2(__d) && (intel_perf->devinfo.revision == 0))
+#define IS_DG2_A1(__d) (IS_DG2(__d) && (intel_perf->devinfo.revision == 1))
+
+#define HAS_OA_FORMAT_64BIT_HDR_BUG(__d) \
+	(IS_DG2_G11(__d) || IS_DG2_A0(__d) || IS_DG2_A1(__d))
+
 static uint64_t
 cs_timebase_scale(uint32_t u32_delta)
 {
@@ -514,7 +556,8 @@ oa_timestamp(const uint32_t *report, enum drm_i915_oa_format format)
 {
 	struct oa_format fmt = get_oa_format(format);
 
-	return fmt.report_hdr_64bit ? *(uint64_t *)&report[2] : report[1];
+	return fmt.report_hdr_64bit && !HAS_OA_FORMAT_64BIT_HDR_BUG(devid) ?
+		*(uint64_t *)&report[2] : report[1];
 }
 
 static uint64_t
-- 
2.25.1

Re: [igt-dev] [PATCH i-g-t 18/23] i915/perf: Fix DG2 A0 report header

[Lionel Landwerlin]

On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
> Timestamp is still 32 bits in DG2 A0. Fix that. Specify variants of DG2
> that have the OA format header bug
>
> Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
> ---
>   tests/i915/perf.c | 45 ++++++++++++++++++++++++++++++++++++++++++++-
>   1 file changed, 44 insertions(+), 1 deletion(-)
>
> diff --git a/tests/i915/perf.c b/tests/i915/perf.c
> index 52de5d38..6951735b 100644
> --- a/tests/i915/perf.c
> +++ b/tests/i915/perf.c
> @@ -40,6 +40,7 @@
>   
>   #include "i915/gem.h"
>   #include "i915/perf.h"
> +#include "i915_pciids.h"
>   #include "igt.h"
>   #include "igt_perf.h"
>   #include "igt_sysfs.h"
> @@ -503,6 +504,47 @@ cs_timestamp_frequency(int fd)
>   	return value;
>   }
>   
> +/*
> + * Hacky ways until IGT is fully aware of steppings:
> + *
> + * WA is permanent for G11. For other DG2s, bug is fixed from B0 onwards.
> + * Steps A0 and A1 correspond to 0 and 1 revid values.


Do you have the Wa number for reference?


> + */
> +struct perf_dev_info {
> +	bool is_dg2_g11;
> +};
> +
> +static const struct perf_dev_info dg2_g11_info = {
> +	.is_dg2_g11 = true,
> +};
> +
> +static const struct pci_id_match perf_dev_match[] = {
> +	INTEL_DG2_G11_IDS(&dg2_g11_info),
> +};
> +
> +static const struct perf_dev_info *__dev_info(uint32_t device_id)
> +{
> +	static const struct perf_dev_info *cache;
> +	static int _iter;
> +
> +	if (_iter)
> +		goto out;
> +
> +	for (_iter = 0; _iter < ARRAY_SIZE(perf_dev_match); _iter++)
> +		if (device_id == perf_dev_match[_iter].device_id)
> +			cache = (void *) perf_dev_match[_iter].match_data;
> +
> +out:
> +	return cache;
> +}
> +
> +#define IS_DG2_G11(__d) (__dev_info(__d) ? __dev_info(__d)->is_dg2_g11 : false)
> +#define IS_DG2_A0(__d) (IS_DG2(__d) && (intel_perf->devinfo.revision == 0))
> +#define IS_DG2_A1(__d) (IS_DG2(__d) && (intel_perf->devinfo.revision == 1))
> +
> +#define HAS_OA_FORMAT_64BIT_HDR_BUG(__d) \
> +	(IS_DG2_G11(__d) || IS_DG2_A0(__d) || IS_DG2_A1(__d))
> +
>   static uint64_t
>   cs_timebase_scale(uint32_t u32_delta)
>   {
> @@ -514,7 +556,8 @@ oa_timestamp(const uint32_t *report, enum drm_i915_oa_format format)
>   {
>   	struct oa_format fmt = get_oa_format(format);
>   
> -	return fmt.report_hdr_64bit ? *(uint64_t *)&report[2] : report[1];
> +	return fmt.report_hdr_64bit && !HAS_OA_FORMAT_64BIT_HDR_BUG(devid) ?
> +		*(uint64_t *)&report[2] : report[1];
>   }
>   
>   static uint64_t

Re: [igt-dev] [PATCH i-g-t 18/23] i915/perf: Fix DG2 A0 report header

[Umesh Nerlige Ramappa]

On Tue, Sep 06, 2022 at 05:17:14PM +0300, Lionel Landwerlin wrote:
>On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
>>Timestamp is still 32 bits in DG2 A0. Fix that. Specify variants of DG2
>>that have the OA format header bug
>>
>>Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
>>---
>>  tests/i915/perf.c | 45 ++++++++++++++++++++++++++++++++++++++++++++-
>>  1 file changed, 44 insertions(+), 1 deletion(-)
>>
>>diff --git a/tests/i915/perf.c b/tests/i915/perf.c
>>index 52de5d38..6951735b 100644
>>--- a/tests/i915/perf.c
>>+++ b/tests/i915/perf.c
>>@@ -40,6 +40,7 @@
>>  #include "i915/gem.h"
>>  #include "i915/perf.h"
>>+#include "i915_pciids.h"
>>  #include "igt.h"
>>  #include "igt_perf.h"
>>  #include "igt_sysfs.h"
>>@@ -503,6 +504,47 @@ cs_timestamp_frequency(int fd)
>>  	return value;
>>  }
>>+/*
>>+ * Hacky ways until IGT is fully aware of steppings:
>>+ *
>>+ * WA is permanent for G11. For other DG2s, bug is fixed from B0 onwards.
>>+ * Steps A0 and A1 correspond to 0 and 1 revid values.
>
>
>Do you have the Wa number for reference?

Should be - 1608133521, the corresponding kernel patch is 

https://patchwork.freedesktop.org/patch/498940/?series=107584&rev=2

Thanks,
Umesh
>
>
>>+ */
>>+struct perf_dev_info {
>>+	bool is_dg2_g11;
>>+};
>>+
>>+static const struct perf_dev_info dg2_g11_info = {
>>+	.is_dg2_g11 = true,
>>+};
>>+
>>+static const struct pci_id_match perf_dev_match[] = {
>>+	INTEL_DG2_G11_IDS(&dg2_g11_info),
>>+};
>>+
>>+static const struct perf_dev_info *__dev_info(uint32_t device_id)
>>+{
>>+	static const struct perf_dev_info *cache;
>>+	static int _iter;
>>+
>>+	if (_iter)
>>+		goto out;
>>+
>>+	for (_iter = 0; _iter < ARRAY_SIZE(perf_dev_match); _iter++)
>>+		if (device_id == perf_dev_match[_iter].device_id)
>>+			cache = (void *) perf_dev_match[_iter].match_data;
>>+
>>+out:
>>+	return cache;
>>+}
>>+
>>+#define IS_DG2_G11(__d) (__dev_info(__d) ? __dev_info(__d)->is_dg2_g11 : false)
>>+#define IS_DG2_A0(__d) (IS_DG2(__d) && (intel_perf->devinfo.revision == 0))
>>+#define IS_DG2_A1(__d) (IS_DG2(__d) && (intel_perf->devinfo.revision == 1))
>>+
>>+#define HAS_OA_FORMAT_64BIT_HDR_BUG(__d) \
>>+	(IS_DG2_G11(__d) || IS_DG2_A0(__d) || IS_DG2_A1(__d))
>>+
>>  static uint64_t
>>  cs_timebase_scale(uint32_t u32_delta)
>>  {
>>@@ -514,7 +556,8 @@ oa_timestamp(const uint32_t *report, enum drm_i915_oa_format format)
>>  {
>>  	struct oa_format fmt = get_oa_format(format);
>>-	return fmt.report_hdr_64bit ? *(uint64_t *)&report[2] : report[1];
>>+	return fmt.report_hdr_64bit && !HAS_OA_FORMAT_64BIT_HDR_BUG(devid) ?
>>+		*(uint64_t *)&report[2] : report[1];
>>  }
>>  static uint64_t
>
>

Re: [igt-dev] [PATCH i-g-t 18/23] i915/perf: Fix DG2 A0 report header

[Lionel Landwerlin]

On 06/09/2022 21:37, Umesh Nerlige Ramappa wrote:
> On Tue, Sep 06, 2022 at 05:17:14PM +0300, Lionel Landwerlin wrote:
>> On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
>>> Timestamp is still 32 bits in DG2 A0. Fix that. Specify variants of DG2
>>> that have the OA format header bug
>>>
>>> Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
>>> ---
>>>  tests/i915/perf.c | 45 ++++++++++++++++++++++++++++++++++++++++++++-
>>>  1 file changed, 44 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/tests/i915/perf.c b/tests/i915/perf.c
>>> index 52de5d38..6951735b 100644
>>> --- a/tests/i915/perf.c
>>> +++ b/tests/i915/perf.c
>>> @@ -40,6 +40,7 @@
>>>  #include "i915/gem.h"
>>>  #include "i915/perf.h"
>>> +#include "i915_pciids.h"
>>>  #include "igt.h"
>>>  #include "igt_perf.h"
>>>  #include "igt_sysfs.h"
>>> @@ -503,6 +504,47 @@ cs_timestamp_frequency(int fd)
>>>      return value;
>>>  }
>>> +/*
>>> + * Hacky ways until IGT is fully aware of steppings:
>>> + *
>>> + * WA is permanent for G11. For other DG2s, bug is fixed from B0 
>>> onwards.
>>> + * Steps A0 and A1 correspond to 0 and 1 revid values.
>>
>>
>> Do you have the Wa number for reference?
>
> Should be - 1608133521, the corresponding kernel patch is
> https://patchwork.freedesktop.org/patch/498940/?series=107584&rev=2
>
> Thanks,
> Umesh


If you can put it in the comment :


Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>


Thanks!


>>
>>
>>> + */
>>> +struct perf_dev_info {
>>> +    bool is_dg2_g11;
>>> +};
>>> +
>>> +static const struct perf_dev_info dg2_g11_info = {
>>> +    .is_dg2_g11 = true,
>>> +};
>>> +
>>> +static const struct pci_id_match perf_dev_match[] = {
>>> +    INTEL_DG2_G11_IDS(&dg2_g11_info),
>>> +};
>>> +
>>> +static const struct perf_dev_info *__dev_info(uint32_t device_id)
>>> +{
>>> +    static const struct perf_dev_info *cache;
>>> +    static int _iter;
>>> +
>>> +    if (_iter)
>>> +        goto out;
>>> +
>>> +    for (_iter = 0; _iter < ARRAY_SIZE(perf_dev_match); _iter++)
>>> +        if (device_id == perf_dev_match[_iter].device_id)
>>> +            cache = (void *) perf_dev_match[_iter].match_data;
>>> +
>>> +out:
>>> +    return cache;
>>> +}
>>> +
>>> +#define IS_DG2_G11(__d) (__dev_info(__d) ? 
>>> __dev_info(__d)->is_dg2_g11 : false)
>>> +#define IS_DG2_A0(__d) (IS_DG2(__d) && 
>>> (intel_perf->devinfo.revision == 0))
>>> +#define IS_DG2_A1(__d) (IS_DG2(__d) && 
>>> (intel_perf->devinfo.revision == 1))
>>> +
>>> +#define HAS_OA_FORMAT_64BIT_HDR_BUG(__d) \
>>> +    (IS_DG2_G11(__d) || IS_DG2_A0(__d) || IS_DG2_A1(__d))
>>> +
>>>  static uint64_t
>>>  cs_timebase_scale(uint32_t u32_delta)
>>>  {
>>> @@ -514,7 +556,8 @@ oa_timestamp(const uint32_t *report, enum 
>>> drm_i915_oa_format format)
>>>  {
>>>      struct oa_format fmt = get_oa_format(format);
>>> -    return fmt.report_hdr_64bit ? *(uint64_t *)&report[2] : report[1];
>>> +    return fmt.report_hdr_64bit && 
>>> !HAS_OA_FORMAT_64BIT_HDR_BUG(devid) ?
>>> +        *(uint64_t *)&report[2] : report[1];
>>>  }
>>>  static uint64_t
>>
>>

[igt-dev] [PATCH i-g-t 19/23] i915/perf: Wait longer for rc6 residency in DG2

[Umesh Nerlige Ramappa]

igt@perf@rc6-disable waits for 50ms to check if we entered rc6. One some
failures this seems to take longer to enter rc6, about 100ms. Tweak the
test for DG2 to use a longer wait time.

Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
---
 tests/i915/perf.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/tests/i915/perf.c b/tests/i915/perf.c
index 6951735b..3ddfe587 100644
--- a/tests/i915/perf.c
+++ b/tests/i915/perf.c
@@ -4441,7 +4441,10 @@ test_rc6_disable(void)
 		.properties_ptr = to_user_pointer(properties),
 	};
 	unsigned long rc6_start, rc6_end, rc6_enabled;
+	uint32_t rc6_duration;
 
+	/* On DG2, we wait longer to enter rc6 */
+	rc6_duration = IS_DG2(devid) ? 100000 : 50000;
 	rc6_enabled = 0;
 	igt_sysfs_rps_scanf(sysfs, RC6_ENABLE, "%lu", &rc6_enabled);
 	igt_require(rc6_enabled);
@@ -4449,7 +4452,7 @@ test_rc6_disable(void)
 	/* Verify rc6 is functional by measuring residency while idle */
 	gem_quiescent_gpu(drm_fd);
 	rc6_start = rc6_residency_ms();
-	usleep(50000);
+	usleep(rc6_duration);
 	rc6_end = rc6_residency_ms();
 	igt_require(rc6_end != rc6_start);
 
@@ -4457,7 +4460,7 @@ test_rc6_disable(void)
 	stream_fd = __perf_open(drm_fd, &param, false);
 
 	rc6_start = rc6_residency_ms();
-	usleep(50000);
+	usleep(rc6_duration);
 	rc6_end = rc6_residency_ms();
 	igt_assert_eq(rc6_end - rc6_start, 0);
 
@@ -4466,7 +4469,7 @@ test_rc6_disable(void)
 
 	/* But once OA is closed, we expect the device to sleep again */
 	rc6_start = rc6_residency_ms();
-	usleep(50000);
+	usleep(rc6_duration);
 	rc6_end = rc6_residency_ms();
 	igt_assert_neq(rc6_end - rc6_start, 0);
 }
-- 
2.25.1

Re: [igt-dev] [PATCH i-g-t 19/23] i915/perf: Wait longer for rc6 residency in DG2

[Lionel Landwerlin]

On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
> igt@perf@rc6-disable waits for 50ms to check if we entered rc6. One some
> failures this seems to take longer to enter rc6, about 100ms. Tweak the
> test for DG2 to use a longer wait time.
>
> Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>

Acked-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>

> ---
>   tests/i915/perf.c | 9 ++++++---
>   1 file changed, 6 insertions(+), 3 deletions(-)
>
> diff --git a/tests/i915/perf.c b/tests/i915/perf.c
> index 6951735b..3ddfe587 100644
> --- a/tests/i915/perf.c
> +++ b/tests/i915/perf.c
> @@ -4441,7 +4441,10 @@ test_rc6_disable(void)
>   		.properties_ptr = to_user_pointer(properties),
>   	};
>   	unsigned long rc6_start, rc6_end, rc6_enabled;
> +	uint32_t rc6_duration;
>   
> +	/* On DG2, we wait longer to enter rc6 */
> +	rc6_duration = IS_DG2(devid) ? 100000 : 50000;
>   	rc6_enabled = 0;
>   	igt_sysfs_rps_scanf(sysfs, RC6_ENABLE, "%lu", &rc6_enabled);
>   	igt_require(rc6_enabled);
> @@ -4449,7 +4452,7 @@ test_rc6_disable(void)
>   	/* Verify rc6 is functional by measuring residency while idle */
>   	gem_quiescent_gpu(drm_fd);
>   	rc6_start = rc6_residency_ms();
> -	usleep(50000);
> +	usleep(rc6_duration);
>   	rc6_end = rc6_residency_ms();
>   	igt_require(rc6_end != rc6_start);
>   
> @@ -4457,7 +4460,7 @@ test_rc6_disable(void)
>   	stream_fd = __perf_open(drm_fd, &param, false);
>   
>   	rc6_start = rc6_residency_ms();
> -	usleep(50000);
> +	usleep(rc6_duration);
>   	rc6_end = rc6_residency_ms();
>   	igt_assert_eq(rc6_end - rc6_start, 0);
>   
> @@ -4466,7 +4469,7 @@ test_rc6_disable(void)
>   
>   	/* But once OA is closed, we expect the device to sleep again */
>   	rc6_start = rc6_residency_ms();
> -	usleep(50000);
> +	usleep(rc6_duration);
>   	rc6_end = rc6_residency_ms();
>   	igt_assert_neq(rc6_end - rc6_start, 0);
>   }

[igt-dev] [PATCH i-g-t 20/23] lib/i915/perf: implement report accumulation for new format

[Umesh Nerlige Ramappa]

From: Lionel Landwerlin <lionel.g.landwerlin@intel.com>

Available on XEHPSDV+.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
---
 lib/i915/perf.c | 33 +++++++++++++++++++++++++++++++++
 1 file changed, 33 insertions(+)

diff --git a/lib/i915/perf.c b/lib/i915/perf.c
index 2f24cbd8..d004313f 100644
--- a/lib/i915/perf.c
+++ b/lib/i915/perf.c
@@ -676,6 +676,39 @@ void intel_perf_accumulate_reports(struct intel_perf_accumulator *acc,
 	memset(acc, 0, sizeof(*acc));
 
 	switch (oa_format) {
+	case I915_OA_FORMAT_A24u40_A14u32_B8_C8:
+		accumulate_uint32(start + 1, end + 1, deltas + idx++); /* timestamp */
+		accumulate_uint32(start + 3, end + 3, deltas + idx++); /* clock */
+
+		/* 4x 32bit A0-3 counters... */
+		for (i = 0; i < 4; i++)
+			accumulate_uint32(start + 4 + i, end + 4 + i, deltas + idx++);
+
+		/* 20x 40bit A4-23 counters... */
+		for (i = 0; i < 20; i++)
+			accumulate_uint40(i + 4, start, end, deltas + idx++);
+
+		/* 4x 32bit A24-27 counters... */
+		for (i = 0; i < 4; i++)
+			accumulate_uint32(start + 28 + i, end + 28 + i, deltas + idx++);
+
+		/* 4x 40bit A28-31 counters... */
+		for (i = 0; i < 4; i++)
+			accumulate_uint40(i + 28, start, end, deltas + idx++);
+
+		/* 5x 32bit A32-36 counters... */
+		for (i = 0; i < 5; i++)
+			accumulate_uint32(start + 36 + i, end + 36 + i, deltas + idx++);
+
+		/* 1x 32bit A37 counter... */
+		accumulate_uint32(start + 46, end + 46, deltas + idx++);
+
+		/* 8x 32bit B counters + 8x 32bit C counters... */
+		for (i = 0; i < 16; i++)
+			accumulate_uint32(start + 48 + i, end + 48 + i, deltas + idx++);
+		break;
+
+	case I915_OAR_FORMAT_A32u40_A4u32_B8_C8:
 	case I915_OA_FORMAT_A32u40_A4u32_B8_C8:
 		accumulate_uint32(start + 1, end + 1, deltas + idx++); /* timestamp */
 		accumulate_uint32(start + 3, end + 3, deltas + idx++); /* clock */
-- 
2.25.1

Re: [igt-dev] [PATCH i-g-t 20/23] lib/i915/perf: implement report accumulation for new format

[Lionel Landwerlin]

On 23/08/2022 21:30, Umesh Nerlige Ramappa wrote:
> From: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
>
> Available on XEHPSDV+.
>
> Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
> ---
>   lib/i915/perf.c | 33 +++++++++++++++++++++++++++++++++
>   1 file changed, 33 insertions(+)
>
> diff --git a/lib/i915/perf.c b/lib/i915/perf.c
> index 2f24cbd8..d004313f 100644
> --- a/lib/i915/perf.c
> +++ b/lib/i915/perf.c
> @@ -676,6 +676,39 @@ void intel_perf_accumulate_reports(struct intel_perf_accumulator *acc,
>   	memset(acc, 0, sizeof(*acc));
>   
>   	switch (oa_format) {
> +	case I915_OA_FORMAT_A24u40_A14u32_B8_C8:
> +		accumulate_uint32(start + 1, end + 1, deltas + idx++); /* timestamp */
> +		accumulate_uint32(start + 3, end + 3, deltas + idx++); /* clock */
> +
> +		/* 4x 32bit A0-3 counters... */
> +		for (i = 0; i < 4; i++)
> +			accumulate_uint32(start + 4 + i, end + 4 + i, deltas + idx++);
> +
> +		/* 20x 40bit A4-23 counters... */
> +		for (i = 0; i < 20; i++)
> +			accumulate_uint40(i + 4, start, end, deltas + idx++);
> +
> +		/* 4x 32bit A24-27 counters... */
> +		for (i = 0; i < 4; i++)
> +			accumulate_uint32(start + 28 + i, end + 28 + i, deltas + idx++);
> +
> +		/* 4x 40bit A28-31 counters... */
> +		for (i = 0; i < 4; i++)
> +			accumulate_uint40(i + 28, start, end, deltas + idx++);
> +
> +		/* 5x 32bit A32-36 counters... */
> +		for (i = 0; i < 5; i++)
> +			accumulate_uint32(start + 36 + i, end + 36 + i, deltas + idx++);
> +
> +		/* 1x 32bit A37 counter... */
> +		accumulate_uint32(start + 46, end + 46, deltas + idx++);
> +
> +		/* 8x 32bit B counters + 8x 32bit C counters... */
> +		for (i = 0; i < 16; i++)
> +			accumulate_uint32(start + 48 + i, end + 48 + i, deltas + idx++);
> +		break;
> +
> +	case I915_OAR_FORMAT_A32u40_A4u32_B8_C8:
>   	case I915_OA_FORMAT_A32u40_A4u32_B8_C8:
>   		accumulate_uint32(start + 1, end + 1, deltas + idx++); /* timestamp */
>   		accumulate_uint32(start + 3, end + 3, deltas + idx++); /* clock */

Re: [igt-dev] [PATCH i-g-t 20/23] lib/i915/perf: implement report accumulation for new format

[Umesh Nerlige Ramappa]

On Tue, Aug 23, 2022 at 06:30:33PM +0000, Umesh Nerlige Ramappa wrote:
>From: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
>
>Available on XEHPSDV+.
>
>Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>

Thanks,
Umesh
>---
> lib/i915/perf.c | 33 +++++++++++++++++++++++++++++++++
> 1 file changed, 33 insertions(+)
>
>diff --git a/lib/i915/perf.c b/lib/i915/perf.c
>index 2f24cbd8..d004313f 100644
>--- a/lib/i915/perf.c
>+++ b/lib/i915/perf.c
>@@ -676,6 +676,39 @@ void intel_perf_accumulate_reports(struct intel_perf_accumulator *acc,
> 	memset(acc, 0, sizeof(*acc));
>
> 	switch (oa_format) {
>+	case I915_OA_FORMAT_A24u40_A14u32_B8_C8:
>+		accumulate_uint32(start + 1, end + 1, deltas + idx++); /* timestamp */
>+		accumulate_uint32(start + 3, end + 3, deltas + idx++); /* clock */
>+
>+		/* 4x 32bit A0-3 counters... */
>+		for (i = 0; i < 4; i++)
>+			accumulate_uint32(start + 4 + i, end + 4 + i, deltas + idx++);
>+
>+		/* 20x 40bit A4-23 counters... */
>+		for (i = 0; i < 20; i++)
>+			accumulate_uint40(i + 4, start, end, deltas + idx++);
>+
>+		/* 4x 32bit A24-27 counters... */
>+		for (i = 0; i < 4; i++)
>+			accumulate_uint32(start + 28 + i, end + 28 + i, deltas + idx++);
>+
>+		/* 4x 40bit A28-31 counters... */
>+		for (i = 0; i < 4; i++)
>+			accumulate_uint40(i + 28, start, end, deltas + idx++);
>+
>+		/* 5x 32bit A32-36 counters... */
>+		for (i = 0; i < 5; i++)
>+			accumulate_uint32(start + 36 + i, end + 36 + i, deltas + idx++);
>+
>+		/* 1x 32bit A37 counter... */
>+		accumulate_uint32(start + 46, end + 46, deltas + idx++);
>+
>+		/* 8x 32bit B counters + 8x 32bit C counters... */
>+		for (i = 0; i < 16; i++)
>+			accumulate_uint32(start + 48 + i, end + 48 + i, deltas + idx++);
>+		break;
>+
>+	case I915_OAR_FORMAT_A32u40_A4u32_B8_C8:
> 	case I915_OA_FORMAT_A32u40_A4u32_B8_C8:
> 		accumulate_uint32(start + 1, end + 1, deltas + idx++); /* timestamp */
> 		accumulate_uint32(start + 3, end + 3, deltas + idx++); /* clock */
>-- 
>2.25.1
>

[igt-dev] [PATCH i-g-t 21/23] lib/i915/perf: fixup conversion script for XEHPSDV

[Umesh Nerlige Ramappa]

From: Lionel Landwerlin <lionel.g.landwerlin@intel.com>

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
---
 lib/i915/perf-configs/mdapi-xml-convert.py | 56 ++++++++++++++++++----
 1 file changed, 48 insertions(+), 8 deletions(-)

diff --git a/lib/i915/perf-configs/mdapi-xml-convert.py b/lib/i915/perf-configs/mdapi-xml-convert.py
index d26ff623..f27d6657 100755
--- a/lib/i915/perf-configs/mdapi-xml-convert.py
+++ b/lib/i915/perf-configs/mdapi-xml-convert.py
@@ -62,6 +62,24 @@ gen8_11_chipset_params = {
     'config_reg_blacklist': {
         0x2364, # OACTXID
     },
+    'register_offsets': {
+        0x1f0: 'PERFCNT 0',
+        0x1f8: 'PERFCNT 1',
+    }
+}
+
+xehpsdv_chipset_params = {
+    'a_offset': 16,
+    'b_offset': 192,
+    'c_offset': 224,
+    'oa_report_size': 256,
+    'config_reg_blacklist': {
+        0x2364, # OACTXID
+    },
+    'register_offsets': {
+        0x1b0: 'PERFCNT 0',
+        0x1b8: 'PERFCNT 1',
+    }
 }
 
 chipsets = {
@@ -161,14 +179,9 @@ def underscore(name):
 def print_err(*args):
     sys.stderr.write(' '.join(map(str,args)) + '\n')
 
-read_register_offsets = {
-    0x1f0: 'PERFCNT 0',
-    0x1f8: 'PERFCNT 1',
-}
-
 def read_value(chipset, offset):
-    if offset in read_register_offsets:
-        return read_register_offsets[offset]
+    if offset in chipsets[chipset]['register_offsets']:
+        return chipsets[chipset]['register_offsets'][offset]
     print_err("Unknown offset register at offset {0}".format(offset))
     assert 0
 
@@ -185,6 +198,7 @@ def read_token_to_rpn_read(chipset, token, raw_offsets):
     offset = int(offset_str, 16)
 
     if raw_offsets:
+        # Location in the HW reports
         a_offset = chipsets[chipset]['a_offset']
         b_offset = chipsets[chipset]['b_offset']
         c_offset = chipsets[chipset]['c_offset']
@@ -199,7 +213,15 @@ def read_token_to_rpn_read(chipset, token, raw_offsets):
             else:
                 assert 0
         elif offset < b_offset:
-            return "A " + str(int((offset - a_offset) / 4)) + " READ"
+            a_cnt_offset = int((offset - a_offset) / 4)
+            if chipset == "XEHPSDV":
+                # Most A counters are in a contiguous array, except
+                # this A37.
+                if a_cnt_offset == 42:
+                    return "A 37 READ"
+                return "A " + str(a_cnt_offset) + " READ"
+            else:
+                return "A " + str(a_cnt_offset) + " READ"
         elif offset < c_offset:
             return "B " + str(int((offset - b_offset) / 4)) + " READ"
         elif offset < report_size:
@@ -207,6 +229,7 @@ def read_token_to_rpn_read(chipset, token, raw_offsets):
         else:
             return "{0} READ".format(read_value(chipset, offset))
     else:
+        # Location in the accumulated deltas
         idx = int(offset / 8)
         if chipset == "HSW":
             # On Haswell accumulated counters are assumed to start
@@ -222,6 +245,23 @@ def read_token_to_rpn_read(chipset, token, raw_offsets):
                 return "C " + str(idx - 54) + " READ"
             else:
                 return "{0} READ".format(read_value(chipset, offset))
+        elif chipset == "XEHPSDV":
+            # For XEHPSDV the array of accumulated counters is
+            # assumed to start with a GPU_TIME then GPU_CLOCK,
+            # then 38 A counters, then 8 B counters and finally
+            # 8 C counters.
+            if idx == 0:
+                return "GPU_TIME 0 READ"
+            elif idx == 1:
+                return "GPU_CLOCK 0 READ"
+            elif idx < 40:
+                return "A " + str(idx - 2) + " READ"
+            elif idx < 48:
+                return "B " + str(idx - 40) + " READ"
+            elif idx < 56:
+                return "C " + str(idx - 48) + " READ"
+            else:
+                return "{0} READ".format(read_value(chipset, offset))
         else:
             # For Gen8+ the array of accumulated counters is
             # assumed to start with a GPU_TIME then GPU_CLOCK,
-- 
2.25.1

Re: [igt-dev] [PATCH i-g-t 21/23] lib/i915/perf: fixup conversion script for XEHPSDV

[Umesh Nerlige Ramappa]

On Tue, Aug 23, 2022 at 06:30:34PM +0000, Umesh Nerlige Ramappa wrote:
>From: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
>
>Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>

Umesh
>---
> lib/i915/perf-configs/mdapi-xml-convert.py | 56 ++++++++++++++++++----
> 1 file changed, 48 insertions(+), 8 deletions(-)
>
>diff --git a/lib/i915/perf-configs/mdapi-xml-convert.py b/lib/i915/perf-configs/mdapi-xml-convert.py
>index d26ff623..f27d6657 100755
>--- a/lib/i915/perf-configs/mdapi-xml-convert.py
>+++ b/lib/i915/perf-configs/mdapi-xml-convert.py
>@@ -62,6 +62,24 @@ gen8_11_chipset_params = {
>     'config_reg_blacklist': {
>         0x2364, # OACTXID
>     },
>+    'register_offsets': {
>+        0x1f0: 'PERFCNT 0',
>+        0x1f8: 'PERFCNT 1',
>+    }
>+}
>+
>+xehpsdv_chipset_params = {
>+    'a_offset': 16,
>+    'b_offset': 192,
>+    'c_offset': 224,
>+    'oa_report_size': 256,
>+    'config_reg_blacklist': {
>+        0x2364, # OACTXID
>+    },
>+    'register_offsets': {
>+        0x1b0: 'PERFCNT 0',
>+        0x1b8: 'PERFCNT 1',
>+    }
> }
>
> chipsets = {
>@@ -161,14 +179,9 @@ def underscore(name):
> def print_err(*args):
>     sys.stderr.write(' '.join(map(str,args)) + '\n')
>
>-read_register_offsets = {
>-    0x1f0: 'PERFCNT 0',
>-    0x1f8: 'PERFCNT 1',
>-}
>-
> def read_value(chipset, offset):
>-    if offset in read_register_offsets:
>-        return read_register_offsets[offset]
>+    if offset in chipsets[chipset]['register_offsets']:
>+        return chipsets[chipset]['register_offsets'][offset]
>     print_err("Unknown offset register at offset {0}".format(offset))
>     assert 0
>
>@@ -185,6 +198,7 @@ def read_token_to_rpn_read(chipset, token, raw_offsets):
>     offset = int(offset_str, 16)
>
>     if raw_offsets:
>+        # Location in the HW reports
>         a_offset = chipsets[chipset]['a_offset']
>         b_offset = chipsets[chipset]['b_offset']
>         c_offset = chipsets[chipset]['c_offset']
>@@ -199,7 +213,15 @@ def read_token_to_rpn_read(chipset, token, raw_offsets):
>             else:
>                 assert 0
>         elif offset < b_offset:
>-            return "A " + str(int((offset - a_offset) / 4)) + " READ"
>+            a_cnt_offset = int((offset - a_offset) / 4)
>+            if chipset == "XEHPSDV":
>+                # Most A counters are in a contiguous array, except
>+                # this A37.
>+                if a_cnt_offset == 42:
>+                    return "A 37 READ"
>+                return "A " + str(a_cnt_offset) + " READ"
>+            else:
>+                return "A " + str(a_cnt_offset) + " READ"
>         elif offset < c_offset:
>             return "B " + str(int((offset - b_offset) / 4)) + " READ"
>         elif offset < report_size:
>@@ -207,6 +229,7 @@ def read_token_to_rpn_read(chipset, token, raw_offsets):
>         else:
>             return "{0} READ".format(read_value(chipset, offset))
>     else:
>+        # Location in the accumulated deltas
>         idx = int(offset / 8)
>         if chipset == "HSW":
>             # On Haswell accumulated counters are assumed to start
>@@ -222,6 +245,23 @@ def read_token_to_rpn_read(chipset, token, raw_offsets):
>                 return "C " + str(idx - 54) + " READ"
>             else:
>                 return "{0} READ".format(read_value(chipset, offset))
>+        elif chipset == "XEHPSDV":
>+            # For XEHPSDV the array of accumulated counters is
>+            # assumed to start with a GPU_TIME then GPU_CLOCK,
>+            # then 38 A counters, then 8 B counters and finally
>+            # 8 C counters.
>+            if idx == 0:
>+                return "GPU_TIME 0 READ"
>+            elif idx == 1:
>+                return "GPU_CLOCK 0 READ"
>+            elif idx < 40:
>+                return "A " + str(idx - 2) + " READ"
>+            elif idx < 48:
>+                return "B " + str(idx - 40) + " READ"
>+            elif idx < 56:
>+                return "C " + str(idx - 48) + " READ"
>+            else:
>+                return "{0} READ".format(read_value(chipset, offset))
>         else:
>             # For Gen8+ the array of accumulated counters is
>             # assumed to start with a GPU_TIME then GPU_CLOCK,
>-- 
>2.25.1
>

[igt-dev] [PATCH i-g-t 22/23] lib/i915/perf: make warning message more helpful

[Umesh Nerlige Ramappa]

From: Lionel Landwerlin <lionel.g.landwerlin@intel.com>

This helped debugging some XEHPSDV stuff but is also more readable.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
---
 lib/i915/perf-configs/mdapi-xml-convert.py | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

diff --git a/lib/i915/perf-configs/mdapi-xml-convert.py b/lib/i915/perf-configs/mdapi-xml-convert.py
index f27d6657..1d161674 100755
--- a/lib/i915/perf-configs/mdapi-xml-convert.py
+++ b/lib/i915/perf-configs/mdapi-xml-convert.py
@@ -884,10 +884,15 @@ for arg in args.xml:
             # delta equations are expected to be identical so warn if that's
             # not true...
             if bool(raw_read_eq) ^ bool(delta_read_eq) or raw_read_eq != delta_read_eq:
-                print_err("WARNING: Inconsistent raw and delta report equations for " + \
-                          mdapi_set.get('ShortName') + " :: " + mdapi_counter.get('SymbolName') + \
-                          "(" + mdapi_counter.get('ShortName') + ")" + ": raw=\"" + str(raw_read_eq) + \
-                          "\" delta=\"" + str(delta_read_eq) + "\" (SKIPPING)")
+                print_err(("WARNING: Inconsistent raw and delta report equations for {0} :: {1} ({2}): " +
+                           "raw=\"{3}\" / \"{4}\" delta=\"{5}\" / \"{6}\" (SKIPPING)")
+                          .format(mdapi_set.get('ShortName'),
+                                  mdapi_counter.get('SymbolName'),
+                                  mdapi_counter.get('ShortName'),
+                                  str(raw_read_eq),
+                                  mdapi_counter.get('SnapshotReportReadEquation'),
+                                  str(delta_read_eq),
+                                  mdapi_counter.get('DeltaReportReadEquation')))
                 set.remove(counter)
                 continue
 
-- 
2.25.1

Re: [igt-dev] [PATCH i-g-t 22/23] lib/i915/perf: make warning message more helpful

[Umesh Nerlige Ramappa]

On Tue, Aug 23, 2022 at 06:30:35PM +0000, Umesh Nerlige Ramappa wrote:
>From: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
>
>This helped debugging some XEHPSDV stuff but is also more readable.
>
>Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>

Umesh
>---
> lib/i915/perf-configs/mdapi-xml-convert.py | 13 +++++++++----
> 1 file changed, 9 insertions(+), 4 deletions(-)
>
>diff --git a/lib/i915/perf-configs/mdapi-xml-convert.py b/lib/i915/perf-configs/mdapi-xml-convert.py
>index f27d6657..1d161674 100755
>--- a/lib/i915/perf-configs/mdapi-xml-convert.py
>+++ b/lib/i915/perf-configs/mdapi-xml-convert.py
>@@ -884,10 +884,15 @@ for arg in args.xml:
>             # delta equations are expected to be identical so warn if that's
>             # not true...
>             if bool(raw_read_eq) ^ bool(delta_read_eq) or raw_read_eq != delta_read_eq:
>-                print_err("WARNING: Inconsistent raw and delta report equations for " + \
>-                          mdapi_set.get('ShortName') + " :: " + mdapi_counter.get('SymbolName') + \
>-                          "(" + mdapi_counter.get('ShortName') + ")" + ": raw=\"" + str(raw_read_eq) + \
>-                          "\" delta=\"" + str(delta_read_eq) + "\" (SKIPPING)")
>+                print_err(("WARNING: Inconsistent raw and delta report equations for {0} :: {1} ({2}): " +
>+                           "raw=\"{3}\" / \"{4}\" delta=\"{5}\" / \"{6}\" (SKIPPING)")
>+                          .format(mdapi_set.get('ShortName'),
>+                                  mdapi_counter.get('SymbolName'),
>+                                  mdapi_counter.get('ShortName'),
>+                                  str(raw_read_eq),
>+                                  mdapi_counter.get('SnapshotReportReadEquation'),
>+                                  str(delta_read_eq),
>+                                  mdapi_counter.get('DeltaReportReadEquation')))
>                 set.remove(counter)
>                 continue
>
>-- 
>2.25.1
>

[igt-dev] [PATCH i-g-t 23/23] lib/i915/perf: Add DG2 metrics

[Umesh Nerlige Ramappa]

From: Lionel Landwerlin <lionel.g.landwerlin@intel.com>

Add DG2 metrics XML files for perf OA tests.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
---
 lib/i915/perf-configs/guids.xml               |   24 +
 lib/i915/perf-configs/mdapi-xml-convert.py    |    5 +-
 lib/i915/perf-configs/oa-dg2.xml              | 9920 +++++++++++++++++
 .../perf-configs/perf-metricset-codegen.py    |   13 +
 lib/i915/perf-configs/update-guids.py         |    2 +-
 lib/i915/perf.c                               |    3 +
 lib/meson.build                               |    2 +-
 7 files changed, 9965 insertions(+), 4 deletions(-)
 create mode 100644 lib/i915/perf-configs/oa-dg2.xml

diff --git a/lib/i915/perf-configs/guids.xml b/lib/i915/perf-configs/guids.xml
index 95c9d717..61a11542 100644
--- a/lib/i915/perf-configs/guids.xml
+++ b/lib/i915/perf-configs/guids.xml
@@ -431,4 +431,28 @@
     <guid config_hash="b420d7b956fb1b097a7bb812b6c87dc0" mdapi_config_hash="d81f15a77d18a099782c18acaa3132f7" id="c3ea2a23-f1c7-4a19-9da2-b569226fb6de" chipset="adl" name="EuActivity7" />
     <guid config_hash="5face026dac6a0549f8a28cf226dfbc8" mdapi_config_hash="de7326838bcc8abf024bac9c771e22e4" id="a5e2f79b-cecb-4eff-8f29-cda8e2a58749" chipset="adl" name="EuActivity8" />
     <guid config_hash="1affe7abeb3739d2c845cc6fa9aacf88" mdapi_config_hash="d071f4933574945979595246be61d3f0" id="89173c19-fcfe-48da-ac26-fb64425f141b" chipset="adl" name="TestOa" />
+    <guid config_hash="9091b7d8f1627a3dda5d6d996eadd033" mdapi_config_hash="4a3cf9efdf4fd581b69308368b986500" id="8b20a2e7-04db-4196-a5ca-26b53cbf9a12" chipset="dg2" name="RenderBasic" />
+    <guid config_hash="a0afa17cc8f442e6eabb3b1782cbca44" mdapi_config_hash="2a23a2f9f1bcf91a9b038b2cc6c9c911" id="f66d7204-54f5-4663-9d68-5213f70165ba" chipset="dg2" name="ComputeBasic" />
+    <guid config_hash="34875baa1a12dda6dcf6d40e33a8d5ef" mdapi_config_hash="31dfc8c6ef9a79789bbe0b991c94112f" id="3b06d2f4-f594-49b5-b399-98550fa2d7b1" chipset="dg2" name="RenderPipeProfile" />
+    <guid config_hash="5f0f4b7c1e4cae4d04ab60a66f1d98a8" mdapi_config_hash="58ec2ac3713bd8e21112f8fddbe28c48" id="1afb0edc-585c-4be9-8c0c-bc72325106b0" chipset="dg2" name="HDCAndSF1" />
+    <guid config_hash="69dbeb5aba00131465b68f20c910421b" mdapi_config_hash="cf8e3757a118ffffb76db95917cae040" id="a3b35902-f293-41fe-97d0-3d895bc5fab3" chipset="dg2" name="HDCAndSF2" />
+    <guid config_hash="e563f102f8b3046e8deaea0eb81b8784" mdapi_config_hash="81d17bd3ab0a18f4491c2d0a11037c7a" id="b04e9460-630f-4682-b0ed-ebb0e74a6c9d" chipset="dg2" name="RasterizerAndPixelBackend1" />
+    <guid config_hash="b574d106f1f71eae1e8a0908cad2c010" mdapi_config_hash="d9c0a07f470d2a9bf7bbf20f79488abf" id="57a0abf4-8d81-4856-ab86-01f15f92c81c" chipset="dg2" name="RasterizerAndPixelBackend2" />
+    <guid config_hash="3316babac928063b799008fc324812b1" mdapi_config_hash="c9cd0a76400e340b0ffdb129fff205c7" id="2314eaa3-164f-4004-9684-3f1d16e875f9" chipset="dg2" name="Sampler_1" />
+    <guid config_hash="31c5ffa0c3508fb166ace213356edbaf" mdapi_config_hash="e2bce131c17fa27ab826a05ed65cad5a" id="c60dea82-db13-4882-8d31-a02bce151ef7" chipset="dg2" name="Sampler_2" />
+    <guid config_hash="b3460c2d0d4e7bcbcc46a3f1c9df7e8f" mdapi_config_hash="ee6821f6954a10562b6314c5ef394b50" id="7f5cb8ec-38bb-48c9-9ad2-60075eba2833" chipset="dg2" name="TDL_1" />
+    <guid config_hash="9ff61984b39a3d19a2abf3a0fd4c2285" mdapi_config_hash="013fa392a5c8b3bb93a776146d31bc95" id="141ce507-de6c-428c-95b2-076194d650db" chipset="dg2" name="TDL_2" />
+    <guid config_hash="11b056437f6c491c2a44e4ca029dd57b" mdapi_config_hash="c4a4c4cec33f45576af76022e6f634ae" id="54a85b07-9ff2-4a12-bfb6-6bcdd9d4019c" chipset="dg2" name="TDL_3" />
+    <guid config_hash="5cf268036a3ae548ddd918fdbd67a64b" mdapi_config_hash="4744760a7ff39c509ecab43f3f715790" id="651c40e5-195c-49da-b232-a9a0f753d899" chipset="dg2" name="TDL_4" />
+    <guid config_hash="f43033715b8153ccb51147452de5bf99" mdapi_config_hash="e77513bc5aa341b061701d6d973b1b4c" id="9a1a14ad-ab92-4204-8a91-c4159a54f712" chipset="dg2" name="L3_1" />
+    <guid config_hash="1c4da3e296779bdbef2262460f3f0fcb" mdapi_config_hash="27c652b5bcd6e800c23a3cacc103c2cb" id="85de4c29-76b8-42a6-b6ef-8a2bed008fcf" chipset="dg2" name="GpuBusyness" />
+    <guid config_hash="a1c8aac6e58018ff3bf78169e5d7cf00" mdapi_config_hash="3d1142aa73a474eccef2bd23027e71b9" id="8dc7142a-df10-41f2-acaa-9faea547ce73" chipset="dg2" name="EuActivity1" />
+    <guid config_hash="880a408e8bc0c0a82df014cbbe0a1a9e" mdapi_config_hash="d998d2f096753140dc92ed9421414ae1" id="d3c05cce-4ae7-4291-82ab-b0bcade6cf7a" chipset="dg2" name="EuActivity2" />
+    <guid config_hash="6687d4c4e45e62d0d28792b04882edae" mdapi_config_hash="c5b1271909b4d72ad1503d5d8a50d30a" id="4fd009c5-6eae-4ecb-a2d1-41edf211cd06" chipset="dg2" name="EuActivity3" />
+    <guid config_hash="f770062c29026a186485a2190b7ce3fb" mdapi_config_hash="7a83985438139dcfdf7d653be1e665b0" id="5226090d-b636-4845-9f76-545cb5db7867" chipset="dg2" name="EuActivity4" />
+    <guid config_hash="4e13b511b35757cc97602aae22b7d2d3" mdapi_config_hash="5405e0eff1e1b0af86f45e832c604fcd" id="c8d33b8f-3152-4c81-99bf-d3cb45d7e1ac" chipset="dg2" name="EuActivity5" />
+    <guid config_hash="243a5047b39033bca9a16db6bf2aca3f" mdapi_config_hash="2bc02385284ccdbbd0d233a428861e59" id="708d01ed-bc9b-420e-8048-9060abcbb1f0" chipset="dg2" name="EuActivity6" />
+    <guid config_hash="d0f2a1c661860e4aa94e25e9f4a6f8eb" mdapi_config_hash="75694dd6a99c38e9446d7bf68f2269a3" id="72262d5b-a9e6-4742-820b-2a7bef9a7e61" chipset="dg2" name="EuActivity7" />
+    <guid config_hash="5e7354b4cbe7b9b03acb504ef00aa8ce" mdapi_config_hash="299c68337c46a2483c1254fdead4cdcc" id="11c685de-1445-41b8-acb5-8e79c75c4fd8" chipset="dg2" name="EuActivity8" />
+    <guid config_hash="24977eb13d34b0da58c0b33f6d7f841a" mdapi_config_hash="48715fb10d2ada8c23f7108303b7c343" id="2db1e1d4-8c48-4bb7-b248-0f40b70433ff" chipset="dg2" name="TestOa" />
 </guids>
diff --git a/lib/i915/perf-configs/mdapi-xml-convert.py b/lib/i915/perf-configs/mdapi-xml-convert.py
index 1d161674..72b52bfd 100755
--- a/lib/i915/perf-configs/mdapi-xml-convert.py
+++ b/lib/i915/perf-configs/mdapi-xml-convert.py
@@ -141,6 +141,7 @@ chipsets = {
     'RKL': gen8_11_chipset_params,
     'DG1': gen8_11_chipset_params,
     'ADL': gen8_11_chipset_params,
+    'DG2': xehpsdv_chipset_params,
 }
 
 register_types = { 'OA', 'NOA', 'FLEX', 'PM' }
@@ -214,7 +215,7 @@ def read_token_to_rpn_read(chipset, token, raw_offsets):
                 assert 0
         elif offset < b_offset:
             a_cnt_offset = int((offset - a_offset) / 4)
-            if chipset == "XEHPSDV":
+            if chipset == "XEHPSDV" or chipset == "DG2":
                 # Most A counters are in a contiguous array, except
                 # this A37.
                 if a_cnt_offset == 42:
@@ -245,7 +246,7 @@ def read_token_to_rpn_read(chipset, token, raw_offsets):
                 return "C " + str(idx - 54) + " READ"
             else:
                 return "{0} READ".format(read_value(chipset, offset))
-        elif chipset == "XEHPSDV":
+        elif chipset == "XEHPSDV" or chipset == "DG2":
             # For XEHPSDV the array of accumulated counters is
             # assumed to start with a GPU_TIME then GPU_CLOCK,
             # then 38 A counters, then 8 B counters and finally
diff --git a/lib/i915/perf-configs/oa-dg2.xml b/lib/i915/perf-configs/oa-dg2.xml
new file mode 100644
index 00000000..9d3e2541
--- /dev/null
+++ b/lib/i915/perf-configs/oa-dg2.xml
@@ -0,0 +1,9920 @@
+<?xml version="1.0"?>
+<metrics version="1602592998" merge_md5="">
+  <set name="Render Basic Metrics"
+       chipset="DG2"
+       symbol_name="RenderBasic"
+       underscore_name="render_basic"
+       mdapi_supported_apis="VK OGL OCL IO"
+       hw_config_guid="8b20a2e7-04db-4196-a5ca-26b53cbf9a12"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Rasterized Pixels"
+             symbol_name="RasterizedPixels"
+             underscore_name="rasterized_pixels"
+             description="The total number of rasterized pixels."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 21 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Early Hi-Depth Test Fails"
+             symbol_name="HiDepthTestFails"
+             underscore_name="hi_depth_test_fails"
+             description="The total number of pixels dropped on early hierarchical depth test."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 22 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer/Hi-Depth Test"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Early Depth Test Fails"
+             symbol_name="EarlyDepthTestFails"
+             underscore_name="early_depth_test_fails"
+             description="The total number of pixels dropped on early depth test."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 23 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer/Early Depth Test"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Killed in FS"
+             symbol_name="SamplesKilledInPs"
+             underscore_name="samples_killed_in_ps"
+             description="The total number of samples or pixels dropped in fragment shaders."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 24 READ 4 UMUL"
+             mdapi_group="3D Pipe/Fragment Shader"
+             mdapi_usage_flags="Tier4 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Pixels Failing Tests"
+             symbol_name="PixelsFailingPostPsTests"
+             underscore_name="pixels_failing_post_ps_tests"
+             description="The total number of pixels dropped on post-FS alpha, stencil, or depth tests."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 25 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Written"
+             symbol_name="SamplesWritten"
+             underscore_name="samples_written"
+             description="The total number of samples or pixels written to all render targets."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 26 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Blended"
+             symbol_name="SamplesBlended"
+             underscore_name="samples_blended"
+             description="The total number of blended samples or pixels written to all render targets."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 27 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Texels"
+             symbol_name="SamplerTexels"
+             underscore_name="sampler_texels"
+             description="The total number of texels seen on input (with 2x2 accuracy) in all sampler units."
+             data_type="uint64"
+             units="texels"
+             semantic_type="event"
+             equation="A 28 READ 4 UMUL"
+             mdapi_group="Sampler/Sampler Input"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Texels Misses"
+             symbol_name="SamplerTexelMisses"
+             underscore_name="sampler_texel_misses"
+             description="The total number of texels lookups (with 2x2 accuracy) that missed L1 sampler cache."
+             data_type="uint64"
+             units="texels"
+             semantic_type="event"
+             equation="A 29 READ 4 UMUL"
+             mdapi_group="Sampler/Sampler Cache"
+             mdapi_usage_flags="Tier3 Batch Frame Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SLM Bytes Read"
+             symbol_name="SlmBytesRead"
+             underscore_name="slm_bytes_read"
+             description="The total number of GPU memory bytes read from shared local memory."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="A 30 READ 64 UMUL"
+             mdapi_group="L3/Data Port/SLM"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SLM Bytes Written"
+             symbol_name="SlmBytesWritten"
+             underscore_name="slm_bytes_written"
+             description="The total number of GPU memory bytes written into shared local memory."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="A 31 READ 64 UMUL"
+             mdapi_group="L3/Data Port/SLM"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Memory Accesses"
+             symbol_name="ShaderMemoryAccesses"
+             underscore_name="shader_memory_accesses"
+             description="The total number of shader memory accesses to L3."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 32 READ"
+             mdapi_group="L3/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Atomic Memory Accesses"
+             symbol_name="ShaderAtomics"
+             underscore_name="shader_atomics"
+             description="The total number of shader atomic memory accesses."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 34 READ"
+             mdapi_group="L3/Data Port/Atomics"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="L3 Shader Throughput"
+             symbol_name="L3ShaderThroughput"
+             underscore_name="l3_shader_throughput"
+             description="The total number of GPU memory bytes transferred between shaders and L3 caches w/o URB."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="$ShaderMemoryAccesses 64 UMUL"
+             mdapi_group="L3/Data Port"
+             mdapi_usage_flags="Tier2 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Barrier Messages"
+             symbol_name="ShaderBarriers"
+             underscore_name="shader_barriers"
+             description="The total number of shader barrier messages."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 35 READ"
+             mdapi_group="EU Array/Barrier"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GTI Read Throughput"
+             symbol_name="GtiReadThroughput"
+             underscore_name="gti_read_throughput"
+             description="The total number of GPU memory bytes read from GTI."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 512 UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="128  A 36 READ UMUL"
+             mdapi_group="GTI"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GTI Write Throughput"
+             symbol_name="GtiWriteThroughput"
+             underscore_name="gti_write_throughput"
+             description="The total number of GPU memory bytes written to GTI."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 512 UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="C 5 READ C 4 READ UADD C 3 READ UADD C 2 READ UADD C 1 READ UADD C 0 READ UADD B 7 READ UADD B 6 READ UADD 128 UMUL"
+             mdapi_group="GTI"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Slice0 Dualsubslice0 is busy"
+             symbol_name="Sampler00Busy"
+             underscore_name="sampler00_busy"
+             description="The percentage of time when sampler slice0 dualsubslice0 is busy"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 1 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Samplers Busy"
+             symbol_name="SamplersBusy"
+             underscore_name="samplers_busy"
+             description="The percentage of time in which samplers or L1 cache are not idle."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="$Sampler00Busy"
+             availability="$DualSubsliceMask 1 AND"
+             mdapi_group="Sampler"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Sampler00 Bottleneck"
+             symbol_name="Sampler00Bottleneck"
+             underscore_name="sampler00_bottleneck"
+             description="The percentage of time in which Slice0 Sampler0 has been slowing down the pipe when processing EU requests."
+             low_watermark="5"
+             high_watermark="15"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 1 AND"
+             mdapi_group="Sampler"
+             mdapi_usage_flags="Tier3 Indicate Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Samplers Bottleneck"
+             symbol_name="SamplerBottleneck"
+             underscore_name="sampler_bottleneck"
+             description="The percentage of time in which samplers have been slowing down the pipe when processing EU requests."
+             low_watermark="5"
+             high_watermark="15"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="$Sampler00Bottleneck"
+             availability="$DualSubsliceMask 1 AND"
+             mdapi_group="Sampler"
+             mdapi_usage_flags="Tier3 Overview Indicate System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x0000000C" />
+        <register type="NOA" address="0x00009888" value="0x1C840000" />
+        <register type="NOA" address="0x00009888" value="0x1E842800" />
+        <register type="NOA" address="0x00009888" value="0x02844000" />
+        <register type="NOA" address="0x00009888" value="0x04840010" />
+        <register type="NOA" address="0x00009888" value="0x12840800" />
+        <register type="NOA" address="0x00009888" value="0x14840037" />
+        <register type="NOA" address="0x00009888" value="0x00840000" />
+        <register type="NOA" address="0x00009888" value="0x02865000" />
+        <register type="NOA" address="0x00009888" value="0x0208A000" />
+        <register type="NOA" address="0x00009888" value="0x020A5000" />
+        <register type="NOA" address="0x00009884" value="0x00000001" />
+        <register type="NOA" address="0x00009888" value="0x18013800" />
+        <register type="NOA" address="0x00009888" value="0x18213800" />
+        <register type="NOA" address="0x00009888" value="0x18413800" />
+        <register type="NOA" address="0x00009888" value="0x18613800" />
+        <register type="NOA" address="0x00009888" value="0x18813800" />
+        <register type="NOA" address="0x00009888" value="0x18A13800" />
+        <register type="NOA" address="0x00009888" value="0x18C13800" />
+        <register type="NOA" address="0x00009888" value="0x18E13800" />
+        <register type="NOA" address="0x00009888" value="0x02010006" />
+        <register type="NOA" address="0x00009888" value="0x14010004" />
+        <register type="NOA" address="0x00009888" value="0x00010000" />
+        <register type="NOA" address="0x00009888" value="0x02210006" />
+        <register type="NOA" address="0x00009888" value="0x12210100" />
+        <register type="NOA" address="0x00009888" value="0x00210000" />
+        <register type="NOA" address="0x00009888" value="0x02410006" />
+        <register type="NOA" address="0x00009888" value="0x12410004" />
+        <register type="NOA" address="0x00009888" value="0x00410000" />
+        <register type="NOA" address="0x00009888" value="0x02610006" />
+        <register type="NOA" address="0x00009888" value="0x10610100" />
+        <register type="NOA" address="0x00009888" value="0x00610000" />
+        <register type="NOA" address="0x00009888" value="0x02810006" />
+        <register type="NOA" address="0x00009888" value="0x10810004" />
+        <register type="NOA" address="0x00009888" value="0x00810000" />
+        <register type="NOA" address="0x00009888" value="0x02A10006" />
+        <register type="NOA" address="0x00009888" value="0x0EA10100" />
+        <register type="NOA" address="0x00009888" value="0x00A10000" />
+        <register type="NOA" address="0x00009888" value="0x02C10006" />
+        <register type="NOA" address="0x00009888" value="0x0EC10004" />
+        <register type="NOA" address="0x00009888" value="0x00C10000" />
+        <register type="NOA" address="0x00009888" value="0x02E10006" />
+        <register type="NOA" address="0x00009888" value="0x0CE10100" />
+        <register type="NOA" address="0x00009888" value="0x00E10000" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x0208A000" />
+        <register type="NOA" address="0x00009888" value="0x02034920" />
+        <register type="NOA" address="0x00009888" value="0x04032924" />
+        <register type="NOA" address="0x00009888" value="0x06030002" />
+        <register type="NOA" address="0x00009888" value="0x020914E0" />
+        <register type="NOA" address="0x00009888" value="0x020A1248" />
+        <register type="NOA" address="0x00009888" value="0x040A0492" />
+        <register type="NOA" address="0x00009888" value="0x040B029C" />
+        <register type="NOA" address="0x00009888" value="0x46030860" />
+        <register type="NOA" address="0x00009888" value="0x48030643" />
+        <register type="NOA" address="0x00009888" value="0x4A030664" />
+        <register type="NOA" address="0x00009888" value="0x16033000" />
+        <register type="NOA" address="0x00009888" value="0x3E030840" />
+        <register type="NOA" address="0x00009888" value="0x18033000" />
+        <register type="NOA" address="0x00009888" value="0x1A033000" />
+        <register type="NOA" address="0x00009888" value="0x40030042" />
+        <register type="NOA" address="0x00009888" value="0x1C033000" />
+        <register type="NOA" address="0x00009888" value="0x1E030000" />
+        <register type="NOA" address="0x00009888" value="0x20030000" />
+        <register type="NOA" address="0x00009888" value="0x42030000" />
+        <register type="NOA" address="0x00009884" value="0x00000008" />
+        <register type="NOA" address="0x00009888" value="0x0001C000" />
+        <register type="NOA" address="0x00009888" value="0x02025000" />
+        <register type="NOA" address="0x00009888" value="0x02045000" />
+        <register type="NOA" address="0x00009888" value="0x02055000" />
+        <register type="NOA" address="0x00009888" value="0x0208A000" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x00800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00000000" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="Compute Metrics Basic"
+       chipset="DG2"
+       symbol_name="ComputeBasic"
+       underscore_name="compute_basic"
+       mdapi_supported_apis="VK OGL OCL IO"
+       hw_config_guid="f66d7204-54f5-4663-9d68-5213f70165ba"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Rasterized Pixels"
+             symbol_name="RasterizedPixels"
+             underscore_name="rasterized_pixels"
+             description="The total number of rasterized pixels."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 21 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Early Hi-Depth Test Fails"
+             symbol_name="HiDepthTestFails"
+             underscore_name="hi_depth_test_fails"
+             description="The total number of pixels dropped on early hierarchical depth test."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 22 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer/Hi-Depth Test"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Early Depth Test Fails"
+             symbol_name="EarlyDepthTestFails"
+             underscore_name="early_depth_test_fails"
+             description="The total number of pixels dropped on early depth test."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 23 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer/Early Depth Test"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Killed in FS"
+             symbol_name="SamplesKilledInPs"
+             underscore_name="samples_killed_in_ps"
+             description="The total number of samples or pixels dropped in fragment shaders."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 24 READ 4 UMUL"
+             mdapi_group="3D Pipe/Fragment Shader"
+             mdapi_usage_flags="Tier4 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Pixels Failing Tests"
+             symbol_name="PixelsFailingPostPsTests"
+             underscore_name="pixels_failing_post_ps_tests"
+             description="The total number of pixels dropped on post-FS alpha, stencil, or depth tests."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 25 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Written"
+             symbol_name="SamplesWritten"
+             underscore_name="samples_written"
+             description="The total number of samples or pixels written to all render targets."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 26 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Blended"
+             symbol_name="SamplesBlended"
+             underscore_name="samples_blended"
+             description="The total number of blended samples or pixels written to all render targets."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 27 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Texels"
+             symbol_name="SamplerTexels"
+             underscore_name="sampler_texels"
+             description="The total number of texels seen on input (with 2x2 accuracy) in all sampler units."
+             data_type="uint64"
+             units="texels"
+             semantic_type="event"
+             equation="A 28 READ 4 UMUL"
+             mdapi_group="Sampler/Sampler Input"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Texels Misses"
+             symbol_name="SamplerTexelMisses"
+             underscore_name="sampler_texel_misses"
+             description="The total number of texels lookups (with 2x2 accuracy) that missed L1 sampler cache."
+             data_type="uint64"
+             units="texels"
+             semantic_type="event"
+             equation="A 29 READ 4 UMUL"
+             mdapi_group="Sampler/Sampler Cache"
+             mdapi_usage_flags="Tier3 Batch Frame Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SLM Bytes Read"
+             symbol_name="SlmBytesRead"
+             underscore_name="slm_bytes_read"
+             description="The total number of GPU memory bytes read from shared local memory."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="A 30 READ 64 UMUL"
+             mdapi_group="L3/Data Port/SLM"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SLM Bytes Written"
+             symbol_name="SlmBytesWritten"
+             underscore_name="slm_bytes_written"
+             description="The total number of GPU memory bytes written into shared local memory."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="A 31 READ 64 UMUL"
+             mdapi_group="L3/Data Port/SLM"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Memory Accesses"
+             symbol_name="ShaderMemoryAccesses"
+             underscore_name="shader_memory_accesses"
+             description="The total number of shader memory accesses to L3."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 32 READ"
+             mdapi_group="L3/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Atomic Memory Accesses"
+             symbol_name="ShaderAtomics"
+             underscore_name="shader_atomics"
+             description="The total number of shader atomic memory accesses."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 34 READ"
+             mdapi_group="L3/Data Port/Atomics"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="L3 Shader Throughput"
+             symbol_name="L3ShaderThroughput"
+             underscore_name="l3_shader_throughput"
+             description="The total number of GPU memory bytes transferred between shaders and L3 caches w/o URB."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="$ShaderMemoryAccesses 64 UMUL"
+             mdapi_group="L3/Data Port"
+             mdapi_usage_flags="Tier2 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Barrier Messages"
+             symbol_name="ShaderBarriers"
+             underscore_name="shader_barriers"
+             description="The total number of shader barrier messages."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 35 READ"
+             mdapi_group="EU Array/Barrier"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GTI Read Throughput"
+             symbol_name="GtiReadThroughput"
+             underscore_name="gti_read_throughput"
+             description="The total number of GPU memory bytes read from GTI."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 512 UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="128  A 36 READ UMUL"
+             mdapi_group="GTI"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GTI Write Throughput"
+             symbol_name="GtiWriteThroughput"
+             underscore_name="gti_write_throughput"
+             description="The total number of GPU memory bytes written to GTI."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 512 UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="C 7 READ C 6 READ UADD C 5 READ UADD C 4 READ UADD C 3 READ UADD C 2 READ UADD C 1 READ UADD C 0 READ UADD 128 UMUL"
+             mdapi_group="GTI"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x00000001" />
+        <register type="NOA" address="0x00009888" value="0x18013800" />
+        <register type="NOA" address="0x00009888" value="0x18213800" />
+        <register type="NOA" address="0x00009888" value="0x18413800" />
+        <register type="NOA" address="0x00009888" value="0x18613800" />
+        <register type="NOA" address="0x00009888" value="0x18813800" />
+        <register type="NOA" address="0x00009888" value="0x18A13800" />
+        <register type="NOA" address="0x00009888" value="0x18C13800" />
+        <register type="NOA" address="0x00009888" value="0x18E13800" />
+        <register type="NOA" address="0x00009888" value="0x02010006" />
+        <register type="NOA" address="0x00009888" value="0x16010004" />
+        <register type="NOA" address="0x00009888" value="0x00010000" />
+        <register type="NOA" address="0x00009888" value="0x02210006" />
+        <register type="NOA" address="0x00009888" value="0x14210100" />
+        <register type="NOA" address="0x00009888" value="0x00210000" />
+        <register type="NOA" address="0x00009888" value="0x02410006" />
+        <register type="NOA" address="0x00009888" value="0x14410004" />
+        <register type="NOA" address="0x00009888" value="0x00410000" />
+        <register type="NOA" address="0x00009888" value="0x02610006" />
+        <register type="NOA" address="0x00009888" value="0x12610100" />
+        <register type="NOA" address="0x00009888" value="0x00610000" />
+        <register type="NOA" address="0x00009888" value="0x02810006" />
+        <register type="NOA" address="0x00009888" value="0x12810004" />
+        <register type="NOA" address="0x00009888" value="0x00810000" />
+        <register type="NOA" address="0x00009888" value="0x02A10006" />
+        <register type="NOA" address="0x00009888" value="0x10A10100" />
+        <register type="NOA" address="0x00009888" value="0x00A10000" />
+        <register type="NOA" address="0x00009888" value="0x02C10006" />
+        <register type="NOA" address="0x00009888" value="0x10C10004" />
+        <register type="NOA" address="0x00009888" value="0x00C10000" />
+        <register type="NOA" address="0x00009888" value="0x02E10006" />
+        <register type="NOA" address="0x00009888" value="0x0EE10100" />
+        <register type="NOA" address="0x00009888" value="0x00E10000" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x02034800" />
+        <register type="NOA" address="0x00009888" value="0x04034924" />
+        <register type="NOA" address="0x00009888" value="0x06030004" />
+        <register type="NOA" address="0x00009888" value="0x02093800" />
+        <register type="NOA" address="0x00009888" value="0x0409000A" />
+        <register type="NOA" address="0x00009888" value="0x020A1200" />
+        <register type="NOA" address="0x00009888" value="0x040A2489" />
+        <register type="NOA" address="0x00009888" value="0x060A0002" />
+        <register type="NOA" address="0x00009888" value="0x040B2700" />
+        <register type="NOA" address="0x00009888" value="0x060B0001" />
+        <register type="NOA" address="0x00009888" value="0x48030000" />
+        <register type="NOA" address="0x00009888" value="0x4A030402" />
+        <register type="NOA" address="0x00009888" value="0x16030000" />
+        <register type="NOA" address="0x00009888" value="0x3E030440" />
+        <register type="NOA" address="0x00009888" value="0x18030000" />
+        <register type="NOA" address="0x00009888" value="0x1A030000" />
+        <register type="NOA" address="0x00009888" value="0x40030000" />
+        <register type="NOA" address="0x00009888" value="0x1C030000" />
+        <register type="NOA" address="0x00009888" value="0x1E030000" />
+        <register type="NOA" address="0x00009888" value="0x20030000" />
+        <register type="NOA" address="0x00009888" value="0x42030000" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x00800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00000000" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="Render Metrics for 3D Pipeline Profile"
+       chipset="DG2"
+       symbol_name="RenderPipeProfile"
+       underscore_name="render_pipe_profile"
+       mdapi_supported_apis="VK OGL OCL IO"
+       hw_config_guid="3b06d2f4-f594-49b5-b399-98550fa2d7b1"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Rasterized Pixels"
+             symbol_name="RasterizedPixels"
+             underscore_name="rasterized_pixels"
+             description="The total number of rasterized pixels."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 21 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Early Hi-Depth Test Fails"
+             symbol_name="HiDepthTestFails"
+             underscore_name="hi_depth_test_fails"
+             description="The total number of pixels dropped on early hierarchical depth test."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 22 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer/Hi-Depth Test"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Early Depth Test Fails"
+             symbol_name="EarlyDepthTestFails"
+             underscore_name="early_depth_test_fails"
+             description="The total number of pixels dropped on early depth test."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 23 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer/Early Depth Test"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Killed in FS"
+             symbol_name="SamplesKilledInPs"
+             underscore_name="samples_killed_in_ps"
+             description="The total number of samples or pixels dropped in fragment shaders."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 24 READ 4 UMUL"
+             mdapi_group="3D Pipe/Fragment Shader"
+             mdapi_usage_flags="Tier4 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Pixels Failing Tests"
+             symbol_name="PixelsFailingPostPsTests"
+             underscore_name="pixels_failing_post_ps_tests"
+             description="The total number of pixels dropped on post-FS alpha, stencil, or depth tests."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 25 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Written"
+             symbol_name="SamplesWritten"
+             underscore_name="samples_written"
+             description="The total number of samples or pixels written to all render targets."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 26 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Blended"
+             symbol_name="SamplesBlended"
+             underscore_name="samples_blended"
+             description="The total number of blended samples or pixels written to all render targets."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 27 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Texels"
+             symbol_name="SamplerTexels"
+             underscore_name="sampler_texels"
+             description="The total number of texels seen on input (with 2x2 accuracy) in all sampler units."
+             data_type="uint64"
+             units="texels"
+             semantic_type="event"
+             equation="A 28 READ 4 UMUL"
+             mdapi_group="Sampler/Sampler Input"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Texels Misses"
+             symbol_name="SamplerTexelMisses"
+             underscore_name="sampler_texel_misses"
+             description="The total number of texels lookups (with 2x2 accuracy) that missed L1 sampler cache."
+             data_type="uint64"
+             units="texels"
+             semantic_type="event"
+             equation="A 29 READ 4 UMUL"
+             mdapi_group="Sampler/Sampler Cache"
+             mdapi_usage_flags="Tier3 Batch Frame Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SLM Bytes Read"
+             symbol_name="SlmBytesRead"
+             underscore_name="slm_bytes_read"
+             description="The total number of GPU memory bytes read from shared local memory."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="A 30 READ 64 UMUL"
+             mdapi_group="L3/Data Port/SLM"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SLM Bytes Written"
+             symbol_name="SlmBytesWritten"
+             underscore_name="slm_bytes_written"
+             description="The total number of GPU memory bytes written into shared local memory."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="A 31 READ 64 UMUL"
+             mdapi_group="L3/Data Port/SLM"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Memory Accesses"
+             symbol_name="ShaderMemoryAccesses"
+             underscore_name="shader_memory_accesses"
+             description="The total number of shader memory accesses to L3."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 32 READ"
+             mdapi_group="L3/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Atomic Memory Accesses"
+             symbol_name="ShaderAtomics"
+             underscore_name="shader_atomics"
+             description="The total number of shader atomic memory accesses."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 34 READ"
+             mdapi_group="L3/Data Port/Atomics"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="L3 Shader Throughput"
+             symbol_name="L3ShaderThroughput"
+             underscore_name="l3_shader_throughput"
+             description="The total number of GPU memory bytes transferred between shaders and L3 caches w/o URB."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="$ShaderMemoryAccesses 64 UMUL"
+             mdapi_group="L3/Data Port"
+             mdapi_usage_flags="Tier2 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Barrier Messages"
+             symbol_name="ShaderBarriers"
+             underscore_name="shader_barriers"
+             description="The total number of shader barrier messages."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 35 READ"
+             mdapi_group="EU Array/Barrier"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VF0 Bottleneck"
+             symbol_name="Vf0Bottleneck"
+             underscore_name="vf0_bottleneck"
+             description="The percentage of time in which vertex fetch pipeline stage was slowing down the 3D pipeline."
+             low_watermark="5"
+             high_watermark="15"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Input Assembler"
+             mdapi_usage_flags="Tier3 Indicate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS0 Bottleneck"
+             symbol_name="Vs0Bottleneck"
+             underscore_name="vs0_bottleneck"
+             description="The percentage of time in which VS0 pipeline stage was slowing down the 3D pipeline."
+             low_watermark="10"
+             high_watermark="30"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Vertex Shader"
+             mdapi_usage_flags="Tier3 Indicate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Clipper0 Bottleneck"
+             symbol_name="Cl0Bottleneck"
+             underscore_name="cl0_bottleneck"
+             description="The percentage of time in which clipper pipeline stage was slowing down the 3D pipeline."
+             low_watermark="10"
+             high_watermark="30"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Clipper"
+             mdapi_usage_flags="Tier3 Indicate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS0 Bottleneck"
+             symbol_name="Hs0Bottleneck"
+             underscore_name="hs0_bottleneck"
+             description="The percentage of time in which hull shader pipeline stage was slowing down the 3D pipeline."
+             low_watermark="3"
+             high_watermark="9"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Hull Shader"
+             mdapi_usage_flags="Tier3 Indicate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS0 Bottleneck"
+             symbol_name="Ds0Bottleneck"
+             underscore_name="ds0_bottleneck"
+             description="The percentage of time in which domain shader pipeline stage was slowing down the 3D pipeline."
+             low_watermark="5"
+             high_watermark="15"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Domain Shader"
+             mdapi_usage_flags="Tier3 Indicate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS0 Bottleneck"
+             symbol_name="Gs0Bottleneck"
+             underscore_name="gs0_bottleneck"
+             description="The percentage of time in which geometry shader pipeline stage was slowing down the 3D pipeline."
+             low_watermark="5"
+             high_watermark="15"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Geometry Shader"
+             mdapi_usage_flags="Tier3 Indicate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SF0 Bottleneck"
+             symbol_name="Sf0Bottleneck"
+             underscore_name="sf0_bottleneck"
+             description="The percentage of time in which strip-fans pipeline stage was slowing down the 3D pipeline."
+             low_watermark="5"
+             high_watermark="10"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Rasterizer/Strip-Fans"
+             mdapi_usage_flags="Tier3 Indicate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SO0 Bottleneck"
+             symbol_name="So0Bottleneck"
+             underscore_name="so0_bottleneck"
+             description="The percentage of time in which stream output pipeline stage was slowing down the 3D pipeline."
+             low_watermark="5"
+             high_watermark="15"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Stream Output"
+             mdapi_usage_flags="Tier3 Indicate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CL0 Stall"
+             symbol_name="Cl0Stall"
+             underscore_name="cl0_stall"
+             description="The percentage of time in which clipper pipeline stage was stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Clipper"
+             mdapi_usage_flags="Tier4 Correlate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS0 Stall"
+             symbol_name="Hs0Stall"
+             underscore_name="hs0_stall"
+             description="The percentage of time in which hull stall pipeline stage was stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Hull Shader"
+             mdapi_usage_flags="Tier4 Correlate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS0 Stall"
+             symbol_name="Ds0Stall"
+             underscore_name="ds0_stall"
+             description="The percentage of time in which domain shader pipeline stage was stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Domain Shader"
+             mdapi_usage_flags="Tier4 Correlate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SF0 Stall"
+             symbol_name="Sf0Stall"
+             underscore_name="sf0_stall"
+             description="The percentage of time in which strip-fans pipeline stage was stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Rasterizer/Strip-Fans"
+             mdapi_usage_flags="Tier4 Correlate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SO0 Stall"
+             symbol_name="So0Stall"
+             underscore_name="so0_stall"
+             description="The percentage of time in which stream-output pipeline stage was stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Stream Output"
+             mdapi_usage_flags="Tier4 Correlate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Hi-Depth Bottleneck"
+             symbol_name="HiDepthBottleneck"
+             underscore_name="hi_depth_bottleneck"
+             description="The percentage of time in which early hierarchical depth test pipeline stage was slowing down the 3D pipeline."
+             low_watermark="5"
+             high_watermark="15"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Rasterizer/Hi-Depth Test"
+             mdapi_usage_flags="Tier3 Indicate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="BC Bottleneck"
+             symbol_name="BcBottleneck"
+             underscore_name="bc_bottleneck"
+             description="The percentage of time in which barycentric coordinates calculation pipeline stage was slowing down the 3D pipeline."
+             low_watermark="5"
+             high_watermark="15"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Rasterizer/Barycentric Calc"
+             mdapi_usage_flags="Tier3 Indicate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x0000000C" />
+        <register type="NOA" address="0x00009888" value="0x168607C0" />
+        <register type="NOA" address="0x00009888" value="0x1603001F" />
+        <register type="NOA" address="0x00009888" value="0x0C030000" />
+        <register type="NOA" address="0x00009888" value="0x12040640" />
+        <register type="NOA" address="0x00009888" value="0x02040000" />
+        <register type="NOA" address="0x00009888" value="0x1C070000" />
+        <register type="NOA" address="0x00009888" value="0x16090000" />
+        <register type="NOA" address="0x00009888" value="0x1C0903C0" />
+        <register type="NOA" address="0x00009888" value="0x160A01F0" />
+        <register type="NOA" address="0x00009888" value="0x160B7C00" />
+        <register type="NOA" address="0x00009888" value="0x04860000" />
+        <register type="NOA" address="0x00009888" value="0x06860400" />
+        <register type="NOA" address="0x00009888" value="0x00860000" />
+        <register type="NOA" address="0x00009888" value="0x02030002" />
+        <register type="NOA" address="0x00009888" value="0x04030010" />
+        <register type="NOA" address="0x00009888" value="0x0A0301C0" />
+        <register type="NOA" address="0x00009888" value="0x100300C0" />
+        <register type="NOA" address="0x00009888" value="0x12030C80" />
+        <register type="NOA" address="0x00009888" value="0x00030000" />
+        <register type="NOA" address="0x00009888" value="0x00044000" />
+        <register type="NOA" address="0x00009888" value="0x08040021" />
+        <register type="NOA" address="0x00009888" value="0x0C040080" />
+        <register type="NOA" address="0x00009888" value="0x0E040022" />
+        <register type="NOA" address="0x00009888" value="0x04074000" />
+        <register type="NOA" address="0x00009888" value="0x0A070021" />
+        <register type="NOA" address="0x00009888" value="0x0E0708C0" />
+        <register type="NOA" address="0x00009888" value="0x00070800" />
+        <register type="NOA" address="0x00009888" value="0x020700A0" />
+        <register type="NOA" address="0x00009888" value="0x00083032" />
+        <register type="NOA" address="0x00009888" value="0x02083300" />
+        <register type="NOA" address="0x00009888" value="0x02094003" />
+        <register type="NOA" address="0x00009888" value="0x08090880" />
+        <register type="NOA" address="0x00009888" value="0x0C090001" />
+        <register type="NOA" address="0x00009888" value="0x0E090023" />
+        <register type="NOA" address="0x00009888" value="0x12090003" />
+        <register type="NOA" address="0x00009888" value="0x00090D28" />
+        <register type="NOA" address="0x00009888" value="0x040A0200" />
+        <register type="NOA" address="0x00009888" value="0x140A0940" />
+        <register type="NOA" address="0x00009888" value="0x000A9A91" />
+        <register type="NOA" address="0x00009888" value="0x020A19AA" />
+        <register type="NOA" address="0x00009888" value="0x00015044" />
+        <register type="NOA" address="0x00009888" value="0x040B0004" />
+        <register type="NOA" address="0x00009888" value="0x080B0000" />
+        <register type="NOA" address="0x00009888" value="0x000B2000" />
+        <register type="NOA" address="0x00009888" value="0x020B2200" />
+        <register type="NOA" address="0x00009888" value="0x00020C20" />
+        <register type="NOA" address="0x00009888" value="0x000C0800" />
+        <register type="NOA" address="0x00009888" value="0x020C00A0" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x0008AAA2" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009888" value="0x00032482" />
+        <register type="NOA" address="0x00009888" value="0x02032492" />
+        <register type="NOA" address="0x00009888" value="0x04032492" />
+        <register type="NOA" address="0x00009888" value="0x06030002" />
+        <register type="NOA" address="0x00009888" value="0x14030000" />
+        <register type="NOA" address="0x00009888" value="0x3E030007" />
+        <register type="NOA" address="0x00009888" value="0x2C030000" />
+        <register type="NOA" address="0x00009888" value="0x4203180A" />
+        <register type="NOA" address="0x00009888" value="0x32031000" />
+        <register type="NOA" address="0x00009888" value="0x440300C2" />
+        <register type="NOA" address="0x00009888" value="0x38031000" />
+        <register type="NOA" address="0x00009888" value="0x46030200" />
+        <register type="NOA" address="0x00009888" value="0x48030261" />
+        <register type="NOA" address="0x00009888" value="0x4A030C22" />
+        <register type="NOA" address="0x00009888" value="0x16030000" />
+        <register type="NOA" address="0x00009888" value="0x18030000" />
+        <register type="NOA" address="0x00009888" value="0x1A030000" />
+        <register type="NOA" address="0x00009888" value="0x400304C1" />
+        <register type="NOA" address="0x00009888" value="0x1C031000" />
+        <register type="NOA" address="0x00009888" value="0x1E030000" />
+        <register type="NOA" address="0x00009888" value="0x20031000" />
+        <register type="NOA" address="0x00009884" value="0x00000008" />
+        <register type="NOA" address="0x00009888" value="0x0001FFFD" />
+        <register type="NOA" address="0x00009888" value="0x00025551" />
+        <register type="NOA" address="0x00009888" value="0x02025555" />
+        <register type="NOA" address="0x00009888" value="0x00045551" />
+        <register type="NOA" address="0x00009888" value="0x02045555" />
+        <register type="NOA" address="0x00009888" value="0x00055551" />
+        <register type="NOA" address="0x00009888" value="0x02055555" />
+        <register type="NOA" address="0x00009888" value="0x0008AAA2" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x00800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00000000" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="Metric set HDCAndSF1"
+       chipset="DG2"
+       symbol_name="HDCAndSF1"
+       underscore_name="hdc_and_sf1"
+       mdapi_supported_apis="VK OGL OCL IO"
+       hw_config_guid="1afb0edc-585c-4be9-8c0c-bc72325106b0"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Rasterized Pixels"
+             symbol_name="RasterizedPixels"
+             underscore_name="rasterized_pixels"
+             description="The total number of rasterized pixels."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 21 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Early Hi-Depth Test Fails"
+             symbol_name="HiDepthTestFails"
+             underscore_name="hi_depth_test_fails"
+             description="The total number of pixels dropped on early hierarchical depth test."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 22 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer/Hi-Depth Test"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Early Depth Test Fails"
+             symbol_name="EarlyDepthTestFails"
+             underscore_name="early_depth_test_fails"
+             description="The total number of pixels dropped on early depth test."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 23 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer/Early Depth Test"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Killed in FS"
+             symbol_name="SamplesKilledInPs"
+             underscore_name="samples_killed_in_ps"
+             description="The total number of samples or pixels dropped in fragment shaders."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 24 READ 4 UMUL"
+             mdapi_group="3D Pipe/Fragment Shader"
+             mdapi_usage_flags="Tier4 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Pixels Failing Tests"
+             symbol_name="PixelsFailingPostPsTests"
+             underscore_name="pixels_failing_post_ps_tests"
+             description="The total number of pixels dropped on post-FS alpha, stencil, or depth tests."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 25 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Written"
+             symbol_name="SamplesWritten"
+             underscore_name="samples_written"
+             description="The total number of samples or pixels written to all render targets."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 26 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Blended"
+             symbol_name="SamplesBlended"
+             underscore_name="samples_blended"
+             description="The total number of blended samples or pixels written to all render targets."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 27 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Texels"
+             symbol_name="SamplerTexels"
+             underscore_name="sampler_texels"
+             description="The total number of texels seen on input (with 2x2 accuracy) in all sampler units."
+             data_type="uint64"
+             units="texels"
+             semantic_type="event"
+             equation="A 28 READ 4 UMUL"
+             mdapi_group="Sampler/Sampler Input"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Texels Misses"
+             symbol_name="SamplerTexelMisses"
+             underscore_name="sampler_texel_misses"
+             description="The total number of texels lookups (with 2x2 accuracy) that missed L1 sampler cache."
+             data_type="uint64"
+             units="texels"
+             semantic_type="event"
+             equation="A 29 READ 4 UMUL"
+             mdapi_group="Sampler/Sampler Cache"
+             mdapi_usage_flags="Tier3 Batch Frame Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SLM Bytes Read"
+             symbol_name="SlmBytesRead"
+             underscore_name="slm_bytes_read"
+             description="The total number of GPU memory bytes read from shared local memory."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="A 30 READ 64 UMUL"
+             mdapi_group="L3/Data Port/SLM"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SLM Bytes Written"
+             symbol_name="SlmBytesWritten"
+             underscore_name="slm_bytes_written"
+             description="The total number of GPU memory bytes written into shared local memory."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="A 31 READ 64 UMUL"
+             mdapi_group="L3/Data Port/SLM"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Memory Accesses"
+             symbol_name="ShaderMemoryAccesses"
+             underscore_name="shader_memory_accesses"
+             description="The total number of shader memory accesses to L3."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 32 READ"
+             mdapi_group="L3/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Atomic Memory Accesses"
+             symbol_name="ShaderAtomics"
+             underscore_name="shader_atomics"
+             description="The total number of shader atomic memory accesses."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 34 READ"
+             mdapi_group="L3/Data Port/Atomics"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="L3 Shader Throughput"
+             symbol_name="L3ShaderThroughput"
+             underscore_name="l3_shader_throughput"
+             description="The total number of GPU memory bytes transferred between shaders and L3 caches w/o URB."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="$ShaderMemoryAccesses 64 UMUL"
+             mdapi_group="L3/Data Port"
+             mdapi_usage_flags="Tier2 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Barrier Messages"
+             symbol_name="ShaderBarriers"
+             underscore_name="shader_barriers"
+             description="The total number of shader barrier messages."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 35 READ"
+             mdapi_group="EU Array/Barrier"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Slice0 Dualsubslice0 Non-sampler Shader Access Stalled On L3"
+             symbol_name="NonSamplerShader00AccessStalledOnL3"
+             underscore_name="non_sampler_shader00_access_stalled_on_l3"
+             description="Percentage of time when HDC has messages to L3, but it's stalled due to lack of credits (Slice0 Dualsubslice0)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 1 AND"
+             mdapi_group="GPU/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Slice0 Dualsubslice1 Non-sampler Shader Access Stalled On L3"
+             symbol_name="NonSamplerShader01AccessStalledOnL3"
+             underscore_name="non_sampler_shader01_access_stalled_on_l3"
+             description="Percentage of time when HDC has messages to L3, but it's stalled due to lack of credits (Slice0 Dualsubslice1)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 2 AND"
+             mdapi_group="GPU/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Slice0 Dualsubslice2 Non-sampler Shader Access Stalled On L3"
+             symbol_name="NonSamplerShader02AccessStalledOnL3"
+             underscore_name="non_sampler_shader02_access_stalled_on_l3"
+             description="Percentage of time when HDC has messages to L3, but it's stalled due to lack of credits (Slice0 Dualsubslice2)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 4 AND"
+             mdapi_group="GPU/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Slice0 Dualsubslice3 Non-sampler Shader Access Stalled On L3"
+             symbol_name="NonSamplerShader03AccessStalledOnL3"
+             underscore_name="non_sampler_shader03_access_stalled_on_l3"
+             description="Percentage of time when HDC has messages to L3, but it's stalled due to lack of credits (Slice0 Dualsubslice3)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 8 AND"
+             mdapi_group="GPU/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Slice1 Dualsubslice0 Non-sampler Shader Access Stalled On L3"
+             symbol_name="NonSamplerShader10AccessStalledOnL3"
+             underscore_name="non_sampler_shader10_access_stalled_on_l3"
+             description="Percentage of time when HDC has messages to L3, but it's stalled due to lack of credits (Slice1 Dualsubslice0)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 64 AND"
+             mdapi_group="GPU/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Slice1 Dualsubslice1 Non-sampler Shader Access Stalled On L3"
+             symbol_name="NonSamplerShader11AccessStalledOnL3"
+             underscore_name="non_sampler_shader11_access_stalled_on_l3"
+             description="Percentage of time when HDC has messages to L3, but it's stalled due to lack of credits (Slice1 Dualsubslice1)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 128 AND"
+             mdapi_group="GPU/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Slice1 Dualsubslice2 Non-sampler Shader Access Stalled On L3"
+             symbol_name="NonSamplerShader12AccessStalledOnL3"
+             underscore_name="non_sampler_shader12_access_stalled_on_l3"
+             description="Percentage of time when HDC has messages to L3, but it's stalled due to lack of credits (Slice1 Dualsubslice2)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 256 AND"
+             mdapi_group="GPU/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Slice1 Dualsubslice3 Non-sampler Shader Access Stalled On L3"
+             symbol_name="NonSamplerShader13AccessStalledOnL3"
+             underscore_name="non_sampler_shader13_access_stalled_on_l3"
+             description="Percentage of time when HDC has messages to L3, but it's stalled due to lack of credits (Slice1 Dualsubslice3)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 512 AND"
+             mdapi_group="GPU/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="SQ00 is full"
+             symbol_name="GTRequestQueue00Full"
+             underscore_name="gt_request_queue00_full"
+             description="The percentage of time when IDI0 SQ00 is filled above a threshold (usually 48 entries)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SQ10 is full"
+             symbol_name="GTRequestQueue10Full"
+             underscore_name="gt_request_queue10_full"
+             description="The percentage of time when IDI0 SQ10 is filled above a threshold (usually 48 entries)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SQ20 is full"
+             symbol_name="GTRequestQueue20Full"
+             underscore_name="gt_request_queue20_full"
+             description="The percentage of time when IDI0 SQ20 is filled above a threshold (usually 48 entries)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SQ30 is full"
+             symbol_name="GTRequestQueue30Full"
+             underscore_name="gt_request_queue30_full"
+             description="The percentage of time when IDI0 SQ30 is filled above a threshold (usually 48 entries)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SQ40 is full"
+             symbol_name="GTRequestQueue40Full"
+             underscore_name="gt_request_queue40_full"
+             description="The percentage of time when IDI0 SQ40 is filled above a threshold (usually 48 entries)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SQ50 is full"
+             symbol_name="GTRequestQueue50Full"
+             underscore_name="gt_request_queue50_full"
+             description="The percentage of time when IDI0 SQ50 is filled above a threshold (usually 48 entries)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SQ60 is full"
+             symbol_name="GTRequestQueue60Full"
+             underscore_name="gt_request_queue60_full"
+             description="The percentage of time when IDI0 SQ60 is filled above a threshold (usually 48 entries)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SQ70 is full"
+             symbol_name="GTRequestQueue70Full"
+             underscore_name="gt_request_queue70_full"
+             description="The percentage of time when IDI0 SQ70 is filled above a threshold (usually 48 entries)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x0000000C" />
+        <register type="NOA" address="0x00009888" value="0x1E810200" />
+        <register type="NOA" address="0x00009888" value="0x1EA10200" />
+        <register type="NOA" address="0x00009888" value="0x1EC10200" />
+        <register type="NOA" address="0x00009888" value="0x1EE10200" />
+        <register type="NOA" address="0x00009888" value="0x1F810200" />
+        <register type="NOA" address="0x00009888" value="0x1FA10200" />
+        <register type="NOA" address="0x00009888" value="0x1FC10200" />
+        <register type="NOA" address="0x00009888" value="0x1FE10200" />
+        <register type="NOA" address="0x00009888" value="0x04810400" />
+        <register type="NOA" address="0x00009888" value="0x08810015" />
+        <register type="NOA" address="0x00009888" value="0x00810000" />
+        <register type="NOA" address="0x00009888" value="0x00830002" />
+        <register type="NOA" address="0x00009888" value="0x00840001" />
+        <register type="NOA" address="0x00009888" value="0x00860001" />
+        <register type="NOA" address="0x00009888" value="0x00870001" />
+        <register type="NOA" address="0x00009888" value="0x04A10400" />
+        <register type="NOA" address="0x00009888" value="0x08A10540" />
+        <register type="NOA" address="0x00009888" value="0x00A10000" />
+        <register type="NOA" address="0x00009888" value="0x00A30008" />
+        <register type="NOA" address="0x00009888" value="0x00A40002" />
+        <register type="NOA" address="0x00009888" value="0x00A60004" />
+        <register type="NOA" address="0x00009888" value="0x00A70004" />
+        <register type="NOA" address="0x00009888" value="0x04C10400" />
+        <register type="NOA" address="0x00009888" value="0x0AC10015" />
+        <register type="NOA" address="0x00009888" value="0x08C10000" />
+        <register type="NOA" address="0x00009888" value="0x00C10000" />
+        <register type="NOA" address="0x00009888" value="0x00C30020" />
+        <register type="NOA" address="0x00009888" value="0x00C40004" />
+        <register type="NOA" address="0x00009888" value="0x00C60010" />
+        <register type="NOA" address="0x00009888" value="0x00C70010" />
+        <register type="NOA" address="0x00009888" value="0x04E10400" />
+        <register type="NOA" address="0x00009888" value="0x0AE10540" />
+        <register type="NOA" address="0x00009888" value="0x08E10000" />
+        <register type="NOA" address="0x00009888" value="0x00E10000" />
+        <register type="NOA" address="0x00009888" value="0x00E30080" />
+        <register type="NOA" address="0x00009888" value="0x00E40008" />
+        <register type="NOA" address="0x00009888" value="0x00E60040" />
+        <register type="NOA" address="0x00009888" value="0x00E70040" />
+        <register type="NOA" address="0x00009888" value="0x05810400" />
+        <register type="NOA" address="0x00009888" value="0x0D810015" />
+        <register type="NOA" address="0x00009888" value="0x09810000" />
+        <register type="NOA" address="0x00009888" value="0x01810000" />
+        <register type="NOA" address="0x00009888" value="0x01830200" />
+        <register type="NOA" address="0x00009888" value="0x01840010" />
+        <register type="NOA" address="0x00009888" value="0x01860100" />
+        <register type="NOA" address="0x00009888" value="0x01870100" />
+        <register type="NOA" address="0x00009888" value="0x05A10400" />
+        <register type="NOA" address="0x00009888" value="0x0DA10540" />
+        <register type="NOA" address="0x00009888" value="0x09A10000" />
+        <register type="NOA" address="0x00009888" value="0x01A10000" />
+        <register type="NOA" address="0x00009888" value="0x01A30800" />
+        <register type="NOA" address="0x00009888" value="0x01A40020" />
+        <register type="NOA" address="0x00009888" value="0x01A60400" />
+        <register type="NOA" address="0x00009888" value="0x01A70400" />
+        <register type="NOA" address="0x00009888" value="0x05C10400" />
+        <register type="NOA" address="0x00009888" value="0x0FC10015" />
+        <register type="NOA" address="0x00009888" value="0x09C10000" />
+        <register type="NOA" address="0x00009888" value="0x01C10000" />
+        <register type="NOA" address="0x00009888" value="0x01C32000" />
+        <register type="NOA" address="0x00009888" value="0x01C40040" />
+        <register type="NOA" address="0x00009888" value="0x01C61000" />
+        <register type="NOA" address="0x00009888" value="0x01C71000" />
+        <register type="NOA" address="0x00009888" value="0x05E10400" />
+        <register type="NOA" address="0x00009888" value="0x0FE10540" />
+        <register type="NOA" address="0x00009888" value="0x09E10000" />
+        <register type="NOA" address="0x00009888" value="0x01E10000" />
+        <register type="NOA" address="0x00009888" value="0x01E38000" />
+        <register type="NOA" address="0x00009888" value="0x01E40080" />
+        <register type="NOA" address="0x00009888" value="0x01E64000" />
+        <register type="NOA" address="0x00009888" value="0x01E74000" />
+        <register type="NOA" address="0x00009888" value="0x000300F0" />
+        <register type="NOA" address="0x00009888" value="0x00070060" />
+        <register type="NOA" address="0x00009888" value="0x0008FF0E" />
+        <register type="NOA" address="0x00009888" value="0x0009000C" />
+        <register type="NOA" address="0x00009888" value="0x000A55A5" />
+        <register type="NOA" address="0x00009888" value="0x000100F2" />
+        <register type="NOA" address="0x00009888" value="0x000BAA04" />
+        <register type="NOA" address="0x00009888" value="0x00020004" />
+        <register type="NOA" address="0x00009888" value="0x000C0010" />
+        <register type="NOA" address="0x00009888" value="0x01076000" />
+        <register type="NOA" address="0x00009888" value="0x01080E00" />
+        <register type="NOA" address="0x00009888" value="0x010900C0" />
+        <register type="NOA" address="0x00009888" value="0x010AA500" />
+        <register type="NOA" address="0x00009888" value="0x01010020" />
+        <register type="NOA" address="0x00009888" value="0x010B0400" />
+        <register type="NOA" address="0x00009888" value="0x01020040" />
+        <register type="NOA" address="0x00009888" value="0x010C1000" />
+        <register type="NOA" address="0x00009884" value="0x00000001" />
+        <register type="NOA" address="0x00009888" value="0x0C017000" />
+        <register type="NOA" address="0x00009888" value="0x0C217000" />
+        <register type="NOA" address="0x00009888" value="0x0C417000" />
+        <register type="NOA" address="0x00009888" value="0x0C617000" />
+        <register type="NOA" address="0x00009888" value="0x0C817000" />
+        <register type="NOA" address="0x00009888" value="0x0CA17000" />
+        <register type="NOA" address="0x00009888" value="0x0CC17000" />
+        <register type="NOA" address="0x00009888" value="0x0CE17000" />
+        <register type="NOA" address="0x00009888" value="0x02010010" />
+        <register type="NOA" address="0x00009888" value="0x16010006" />
+        <register type="NOA" address="0x00009888" value="0x00010000" />
+        <register type="NOA" address="0x00009888" value="0x02210010" />
+        <register type="NOA" address="0x00009888" value="0x14210180" />
+        <register type="NOA" address="0x00009888" value="0x00210000" />
+        <register type="NOA" address="0x00009888" value="0x02410010" />
+        <register type="NOA" address="0x00009888" value="0x14410006" />
+        <register type="NOA" address="0x00009888" value="0x00410000" />
+        <register type="NOA" address="0x00009888" value="0x02610010" />
+        <register type="NOA" address="0x00009888" value="0x12610180" />
+        <register type="NOA" address="0x00009888" value="0x00610000" />
+        <register type="NOA" address="0x00009888" value="0x02810010" />
+        <register type="NOA" address="0x00009888" value="0x12810006" />
+        <register type="NOA" address="0x00009888" value="0x00810000" />
+        <register type="NOA" address="0x00009888" value="0x02A10010" />
+        <register type="NOA" address="0x00009888" value="0x10A10180" />
+        <register type="NOA" address="0x00009888" value="0x00A10000" />
+        <register type="NOA" address="0x00009888" value="0x02C10010" />
+        <register type="NOA" address="0x00009888" value="0x10C10006" />
+        <register type="NOA" address="0x00009888" value="0x00C10000" />
+        <register type="NOA" address="0x00009888" value="0x02E10010" />
+        <register type="NOA" address="0x00009888" value="0x0EE10180" />
+        <register type="NOA" address="0x00009888" value="0x00E10000" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
+        <register type="NOA" address="0x00009888" value="0x00032492" />
+        <register type="NOA" address="0x00009888" value="0x02034892" />
+        <register type="NOA" address="0x00009888" value="0x04034924" />
+        <register type="NOA" address="0x00009888" value="0x06030004" />
+        <register type="NOA" address="0x00009888" value="0x02093800" />
+        <register type="NOA" address="0x00009888" value="0x0409000A" />
+        <register type="NOA" address="0x00009888" value="0x020A1200" />
+        <register type="NOA" address="0x00009888" value="0x040A2489" />
+        <register type="NOA" address="0x00009888" value="0x060A0002" />
+        <register type="NOA" address="0x00009888" value="0x040B2700" />
+        <register type="NOA" address="0x00009888" value="0x060B0001" />
+        <register type="NOA" address="0x00009888" value="0x14037000" />
+        <register type="NOA" address="0x00009888" value="0x3E0314CA" />
+        <register type="NOA" address="0x00009888" value="0x26037000" />
+        <register type="NOA" address="0x00009888" value="0x42032004" />
+        <register type="NOA" address="0x00009888" value="0x2C033000" />
+        <register type="NOA" address="0x00009888" value="0x32037000" />
+        <register type="NOA" address="0x00009888" value="0x44030047" />
+        <register type="NOA" address="0x00009888" value="0x38031000" />
+        <register type="NOA" address="0x00009888" value="0x46030003" />
+        <register type="NOA" address="0x00009888" value="0x48031EA0" />
+        <register type="NOA" address="0x00009888" value="0x4A030DE6" />
+        <register type="NOA" address="0x00009888" value="0x1603F000" />
+        <register type="NOA" address="0x00009888" value="0x1803F000" />
+        <register type="NOA" address="0x00009888" value="0x1A03F000" />
+        <register type="NOA" address="0x00009888" value="0x40031084" />
+        <register type="NOA" address="0x00009888" value="0x1C03F000" />
+        <register type="NOA" address="0x00009888" value="0x1E03F000" />
+        <register type="NOA" address="0x00009888" value="0x2003F000" />
+        <register type="NOA" address="0x00009884" value="0x00000008" />
+        <register type="NOA" address="0x00009888" value="0x000100FF" />
+        <register type="NOA" address="0x00009888" value="0x00025555" />
+        <register type="NOA" address="0x00009888" value="0x00045555" />
+        <register type="NOA" address="0x00009888" value="0x00055555" />
+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0xF0800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0xF0800000" />
+        <register type="OA" address="0x0000DC40" value="0x00FF0000" />
+        <register type="OA" address="0x0000D940" value="0x00000002" />
+        <register type="OA" address="0x0000D944" value="0x0000FFFE" />
+        <register type="OA" address="0x0000DC00" value="0x00000002" />
+        <register type="OA" address="0x0000DC04" value="0x0000FFFE" />
+        <register type="OA" address="0x0000D948" value="0x00000002" />
+        <register type="OA" address="0x0000D94C" value="0x0000FFFD" />
+        <register type="OA" address="0x0000DC08" value="0x00000002" />
+        <register type="OA" address="0x0000DC0C" value="0x0000FFFD" />
+        <register type="OA" address="0x0000D950" value="0x00000002" />
+        <register type="OA" address="0x0000D954" value="0x0000FFFB" />
+        <register type="OA" address="0x0000DC10" value="0x00000002" />
+        <register type="OA" address="0x0000DC14" value="0x0000FFFB" />
+        <register type="OA" address="0x0000D958" value="0x00000002" />
+        <register type="OA" address="0x0000D95C" value="0x0000FFF7" />
+        <register type="OA" address="0x0000DC18" value="0x00000002" />
+        <register type="OA" address="0x0000DC1C" value="0x0000FFF7" />
+        <register type="OA" address="0x0000D960" value="0x00000002" />
+        <register type="OA" address="0x0000D964" value="0x0000FFEF" />
+        <register type="OA" address="0x0000DC20" value="0x00000002" />
+        <register type="OA" address="0x0000DC24" value="0x0000FFEF" />
+        <register type="OA" address="0x0000D968" value="0x00000002" />
+        <register type="OA" address="0x0000D96C" value="0x0000FFDF" />
+        <register type="OA" address="0x0000DC28" value="0x00000002" />
+        <register type="OA" address="0x0000DC2C" value="0x0000FFDF" />
+        <register type="OA" address="0x0000D970" value="0x00000002" />
+        <register type="OA" address="0x0000D974" value="0x0000FFBF" />
+        <register type="OA" address="0x0000DC30" value="0x00000002" />
+        <register type="OA" address="0x0000DC34" value="0x0000FFBF" />
+        <register type="OA" address="0x0000D978" value="0x00000002" />
+        <register type="OA" address="0x0000D97C" value="0x0000FF7F" />
+        <register type="OA" address="0x0000DC38" value="0x00000002" />
+        <register type="OA" address="0x0000DC3C" value="0x0000FF7F" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="Metric set HDCAndSF2"
+       chipset="DG2"
+       symbol_name="HDCAndSF2"
+       underscore_name="hdc_and_sf2"
+       mdapi_supported_apis="VK OGL OCL IO"
+       hw_config_guid="a3b35902-f293-41fe-97d0-3d895bc5fab3"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Rasterized Pixels"
+             symbol_name="RasterizedPixels"
+             underscore_name="rasterized_pixels"
+             description="The total number of rasterized pixels."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 21 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Early Hi-Depth Test Fails"
+             symbol_name="HiDepthTestFails"
+             underscore_name="hi_depth_test_fails"
+             description="The total number of pixels dropped on early hierarchical depth test."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 22 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer/Hi-Depth Test"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Early Depth Test Fails"
+             symbol_name="EarlyDepthTestFails"
+             underscore_name="early_depth_test_fails"
+             description="The total number of pixels dropped on early depth test."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 23 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer/Early Depth Test"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Killed in FS"
+             symbol_name="SamplesKilledInPs"
+             underscore_name="samples_killed_in_ps"
+             description="The total number of samples or pixels dropped in fragment shaders."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 24 READ 4 UMUL"
+             mdapi_group="3D Pipe/Fragment Shader"
+             mdapi_usage_flags="Tier4 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Pixels Failing Tests"
+             symbol_name="PixelsFailingPostPsTests"
+             underscore_name="pixels_failing_post_ps_tests"
+             description="The total number of pixels dropped on post-FS alpha, stencil, or depth tests."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 25 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Written"
+             symbol_name="SamplesWritten"
+             underscore_name="samples_written"
+             description="The total number of samples or pixels written to all render targets."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 26 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Blended"
+             symbol_name="SamplesBlended"
+             underscore_name="samples_blended"
+             description="The total number of blended samples or pixels written to all render targets."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 27 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Texels"
+             symbol_name="SamplerTexels"
+             underscore_name="sampler_texels"
+             description="The total number of texels seen on input (with 2x2 accuracy) in all sampler units."
+             data_type="uint64"
+             units="texels"
+             semantic_type="event"
+             equation="A 28 READ 4 UMUL"
+             mdapi_group="Sampler/Sampler Input"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Texels Misses"
+             symbol_name="SamplerTexelMisses"
+             underscore_name="sampler_texel_misses"
+             description="The total number of texels lookups (with 2x2 accuracy) that missed L1 sampler cache."
+             data_type="uint64"
+             units="texels"
+             semantic_type="event"
+             equation="A 29 READ 4 UMUL"
+             mdapi_group="Sampler/Sampler Cache"
+             mdapi_usage_flags="Tier3 Batch Frame Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SLM Bytes Read"
+             symbol_name="SlmBytesRead"
+             underscore_name="slm_bytes_read"
+             description="The total number of GPU memory bytes read from shared local memory."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="A 30 READ 64 UMUL"
+             mdapi_group="L3/Data Port/SLM"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SLM Bytes Written"
+             symbol_name="SlmBytesWritten"
+             underscore_name="slm_bytes_written"
+             description="The total number of GPU memory bytes written into shared local memory."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="A 31 READ 64 UMUL"
+             mdapi_group="L3/Data Port/SLM"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Memory Accesses"
+             symbol_name="ShaderMemoryAccesses"
+             underscore_name="shader_memory_accesses"
+             description="The total number of shader memory accesses to L3."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 32 READ"
+             mdapi_group="L3/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Atomic Memory Accesses"
+             symbol_name="ShaderAtomics"
+             underscore_name="shader_atomics"
+             description="The total number of shader atomic memory accesses."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 34 READ"
+             mdapi_group="L3/Data Port/Atomics"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="L3 Shader Throughput"
+             symbol_name="L3ShaderThroughput"
+             underscore_name="l3_shader_throughput"
+             description="The total number of GPU memory bytes transferred between shaders and L3 caches w/o URB."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="$ShaderMemoryAccesses 64 UMUL"
+             mdapi_group="L3/Data Port"
+             mdapi_usage_flags="Tier2 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Barrier Messages"
+             symbol_name="ShaderBarriers"
+             underscore_name="shader_barriers"
+             description="The total number of shader barrier messages."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 35 READ"
+             mdapi_group="EU Array/Barrier"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Polygon (Slice0) Data Ready"
+             symbol_name="Poly0DataReady"
+             underscore_name="poly0_data_ready"
+             description="The percentage of time in which geometry pipeline output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU/3D Pipe/Strip-Fans"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Polygon (Slice1) Data Ready"
+             symbol_name="Poly1DataReady"
+             underscore_name="poly1_data_ready"
+             description="The percentage of time in which geometry pipeline output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU/3D Pipe/Strip-Fans"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x0000000C" />
+        <register type="NOA" address="0x00009888" value="0x12056400" />
+        <register type="NOA" address="0x00009888" value="0x13056400" />
+        <register type="NOA" address="0x00009888" value="0x00030001" />
+        <register type="NOA" address="0x00009888" value="0x00050002" />
+        <register type="NOA" address="0x00009888" value="0x02050700" />
+        <register type="NOA" address="0x00009888" value="0x00060002" />
+        <register type="NOA" address="0x00009888" value="0x00080007" />
+        <register type="NOA" address="0x00009888" value="0x000A0005" />
+        <register type="NOA" address="0x00009888" value="0x00010001" />
+        <register type="NOA" address="0x00009888" value="0x000B0002" />
+        <register type="NOA" address="0x00009888" value="0x01050002" />
+        <register type="NOA" address="0x00009888" value="0x0305001C" />
+        <register type="NOA" address="0x00009888" value="0x01060001" />
+        <register type="NOA" address="0x00009888" value="0x01080001" />
+        <register type="NOA" address="0x00009888" value="0x010A0001" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x0008000A" />
+        <register type="NOA" address="0x00009888" value="0x00030012" />
+        <register type="NOA" address="0x00009888" value="0x14030000" />
+        <register type="NOA" address="0x00009888" value="0x3E030000" />
+        <register type="NOA" address="0x00009888" value="0x26033000" />
+        <register type="NOA" address="0x00009888" value="0x42030100" />
+        <register type="NOA" address="0x00009888" value="0x02030000" />
+        <register type="NOA" address="0x00009884" value="0x00000008" />
+        <register type="NOA" address="0x00009888" value="0x00010003" />
+        <register type="NOA" address="0x00009888" value="0x00020005" />
+        <register type="NOA" address="0x00009888" value="0x00040005" />
+        <register type="NOA" address="0x00009888" value="0x00050005" />
+        <register type="NOA" address="0x00009888" value="0x0008000A" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x30800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00030000" />
+        <register type="OA" address="0x0000D940" value="0x00000002" />
+        <register type="OA" address="0x0000D944" value="0x0000FFFE" />
+        <register type="OA" address="0x0000DC00" value="0x00000002" />
+        <register type="OA" address="0x0000DC04" value="0x0000FFFE" />
+        <register type="OA" address="0x0000D948" value="0x00000002" />
+        <register type="OA" address="0x0000D94C" value="0x0000FFFD" />
+        <register type="OA" address="0x0000DC08" value="0x00000002" />
+        <register type="OA" address="0x0000DC0C" value="0x0000FFFD" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="Metric set RasterizerAndPixelBackend1"
+       chipset="DG2"
+       symbol_name="RasterizerAndPixelBackend1"
+       underscore_name="rasterizer_and_pixel_backend1"
+       mdapi_supported_apis="VK OGL OCL IO"
+       hw_config_guid="b04e9460-630f-4682-b0ed-ebb0e74a6c9d"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Rasterized Pixels"
+             symbol_name="RasterizedPixels"
+             underscore_name="rasterized_pixels"
+             description="The total number of rasterized pixels."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 21 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Early Hi-Depth Test Fails"
+             symbol_name="HiDepthTestFails"
+             underscore_name="hi_depth_test_fails"
+             description="The total number of pixels dropped on early hierarchical depth test."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 22 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer/Hi-Depth Test"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Early Depth Test Fails"
+             symbol_name="EarlyDepthTestFails"
+             underscore_name="early_depth_test_fails"
+             description="The total number of pixels dropped on early depth test."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 23 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer/Early Depth Test"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Killed in FS"
+             symbol_name="SamplesKilledInPs"
+             underscore_name="samples_killed_in_ps"
+             description="The total number of samples or pixels dropped in fragment shaders."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 24 READ 4 UMUL"
+             mdapi_group="3D Pipe/Fragment Shader"
+             mdapi_usage_flags="Tier4 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Pixels Failing Tests"
+             symbol_name="PixelsFailingPostPsTests"
+             underscore_name="pixels_failing_post_ps_tests"
+             description="The total number of pixels dropped on post-FS alpha, stencil, or depth tests."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 25 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Written"
+             symbol_name="SamplesWritten"
+             underscore_name="samples_written"
+             description="The total number of samples or pixels written to all render targets."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 26 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Blended"
+             symbol_name="SamplesBlended"
+             underscore_name="samples_blended"
+             description="The total number of blended samples or pixels written to all render targets."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 27 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Texels"
+             symbol_name="SamplerTexels"
+             underscore_name="sampler_texels"
+             description="The total number of texels seen on input (with 2x2 accuracy) in all sampler units."
+             data_type="uint64"
+             units="texels"
+             semantic_type="event"
+             equation="A 28 READ 4 UMUL"
+             mdapi_group="Sampler/Sampler Input"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Texels Misses"
+             symbol_name="SamplerTexelMisses"
+             underscore_name="sampler_texel_misses"
+             description="The total number of texels lookups (with 2x2 accuracy) that missed L1 sampler cache."
+             data_type="uint64"
+             units="texels"
+             semantic_type="event"
+             equation="A 29 READ 4 UMUL"
+             mdapi_group="Sampler/Sampler Cache"
+             mdapi_usage_flags="Tier3 Batch Frame Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SLM Bytes Read"
+             symbol_name="SlmBytesRead"
+             underscore_name="slm_bytes_read"
+             description="The total number of GPU memory bytes read from shared local memory."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="A 30 READ 64 UMUL"
+             mdapi_group="L3/Data Port/SLM"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SLM Bytes Written"
+             symbol_name="SlmBytesWritten"
+             underscore_name="slm_bytes_written"
+             description="The total number of GPU memory bytes written into shared local memory."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="A 31 READ 64 UMUL"
+             mdapi_group="L3/Data Port/SLM"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Memory Accesses"
+             symbol_name="ShaderMemoryAccesses"
+             underscore_name="shader_memory_accesses"
+             description="The total number of shader memory accesses to L3."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 32 READ"
+             mdapi_group="L3/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Atomic Memory Accesses"
+             symbol_name="ShaderAtomics"
+             underscore_name="shader_atomics"
+             description="The total number of shader atomic memory accesses."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 34 READ"
+             mdapi_group="L3/Data Port/Atomics"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="L3 Shader Throughput"
+             symbol_name="L3ShaderThroughput"
+             underscore_name="l3_shader_throughput"
+             description="The total number of GPU memory bytes transferred between shaders and L3 caches w/o URB."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="$ShaderMemoryAccesses 64 UMUL"
+             mdapi_group="L3/Data Port"
+             mdapi_usage_flags="Tier2 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Barrier Messages"
+             symbol_name="ShaderBarriers"
+             underscore_name="shader_barriers"
+             description="The total number of shader barrier messages."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 35 READ"
+             mdapi_group="EU Array/Barrier"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Slice0 Rasterizer Input Available"
+             symbol_name="Rasterizer0InputAvailable"
+             underscore_name="rasterizer0_input_available"
+             description="The percentage of time in which slice0 rasterizer input is available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$SliceMask 1 AND"
+             mdapi_group="GPU/Rasterizer"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Slice1 Rasterizer Output Ready"
+             symbol_name="Rasterizer1OutputReady"
+             underscore_name="rasterizer1_output_ready"
+             description="The percentage of time in which slice1 rasterizer output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$SliceMask 2 AND"
+             mdapi_group="GPU/Rasterizer"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Slice0 Pipe0 Post-EarlyZ Pixel Data Ready"
+             symbol_name="PixelData00Ready"
+             underscore_name="pixel_data00_ready"
+             description="The percentage of time in which slice0  pipe0 post-EarlyZ pixel data is ready (after early Z tests have been applied)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU/Rasterizer/Early Depth Test"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Slice0 Pipe1 Post-EarlyZ Pixel Data Ready"
+             symbol_name="PixelData01Ready"
+             underscore_name="pixel_data01_ready"
+             description="The percentage of time in which slice0  pipe1 post-EarlyZ pixel data is ready (after early Z tests have been applied)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU/Rasterizer/Early Depth Test"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x0000000C" />
+        <register type="NOA" address="0x00009888" value="0x1A056000" />
+        <register type="NOA" address="0x00009888" value="0x1C050000" />
+        <register type="NOA" address="0x00009888" value="0x0C0B7000" />
+        <register type="NOA" address="0x00009888" value="0x0E0B6000" />
+        <register type="NOA" address="0x00009888" value="0x0C0C701F" />
+        <register type="NOA" address="0x00009888" value="0x0E0C6000" />
+        <register type="NOA" address="0x00009888" value="0x1B057000" />
+        <register type="NOA" address="0x00009888" value="0x00038000" />
+        <register type="NOA" address="0x00009888" value="0x00050000" />
+        <register type="NOA" address="0x00009888" value="0x020535D2" />
+        <register type="NOA" address="0x00009888" value="0x0405003E" />
+        <register type="NOA" address="0x00009888" value="0x00060007" />
+        <register type="NOA" address="0x00009888" value="0x00072800" />
+        <register type="NOA" address="0x00009888" value="0x000803D5" />
+        <register type="NOA" address="0x00009888" value="0x0208C000" />
+        <register type="NOA" address="0x00009888" value="0x00090060" />
+        <register type="NOA" address="0x00009888" value="0x000A2955" />
+        <register type="NOA" address="0x00009888" value="0x020A4000" />
+        <register type="NOA" address="0x00009888" value="0x00018018" />
+        <register type="NOA" address="0x00009888" value="0x040B0200" />
+        <register type="NOA" address="0x00009888" value="0x080B0400" />
+        <register type="NOA" address="0x00009888" value="0x0A0B001F" />
+        <register type="NOA" address="0x00009888" value="0x000B0000" />
+        <register type="NOA" address="0x00009888" value="0x020B8000" />
+        <register type="NOA" address="0x00009888" value="0x00020060" />
+        <register type="NOA" address="0x00009888" value="0x040C0200" />
+        <register type="NOA" address="0x00009888" value="0x0A0C0400" />
+        <register type="NOA" address="0x00009888" value="0x080C0000" />
+        <register type="NOA" address="0x00009888" value="0x000C0000" />
+        <register type="NOA" address="0x00009888" value="0x01050000" />
+        <register type="NOA" address="0x00009888" value="0x11050DC0" />
+        <register type="NOA" address="0x00009888" value="0x03050000" />
+        <register type="NOA" address="0x00009888" value="0x01068000" />
+        <register type="NOA" address="0x00009888" value="0x03084000" />
+        <register type="NOA" address="0x00009888" value="0x030A4000" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x00082AAA" />
+        <register type="NOA" address="0x00009888" value="0x02088000" />
+        <register type="NOA" address="0x00009888" value="0x00032492" />
+        <register type="NOA" address="0x00009888" value="0x02030012" />
+        <register type="NOA" address="0x00009888" value="0x06030002" />
+        <register type="NOA" address="0x00009888" value="0x14033000" />
+        <register type="NOA" address="0x00009888" value="0x3E030008" />
+        <register type="NOA" address="0x00009888" value="0x26033000" />
+        <register type="NOA" address="0x00009888" value="0x42032100" />
+        <register type="NOA" address="0x00009888" value="0x2C033000" />
+        <register type="NOA" address="0x00009888" value="0x32033000" />
+        <register type="NOA" address="0x00009888" value="0x44030CA5" />
+        <register type="NOA" address="0x00009888" value="0x38033000" />
+        <register type="NOA" address="0x00009888" value="0x46030462" />
+        <register type="NOA" address="0x00009888" value="0x20030000" />
+        <register type="NOA" address="0x00009884" value="0x00000008" />
+        <register type="NOA" address="0x00009888" value="0x0001807F" />
+        <register type="NOA" address="0x00009888" value="0x00021555" />
+        <register type="NOA" address="0x00009888" value="0x02024000" />
+        <register type="NOA" address="0x00009888" value="0x00041555" />
+        <register type="NOA" address="0x00009888" value="0x02044000" />
+        <register type="NOA" address="0x00009888" value="0x00051555" />
+        <register type="NOA" address="0x00009888" value="0x02054000" />
+        <register type="NOA" address="0x00009888" value="0x00082AAA" />
+        <register type="NOA" address="0x00009888" value="0x02088000" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x70800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00070000" />
+        <register type="OA" address="0x0000D940" value="0x00000038" />
+        <register type="OA" address="0x0000D944" value="0x0000FFF8" />
+        <register type="OA" address="0x0000DC00" value="0x00000038" />
+        <register type="OA" address="0x0000DC04" value="0x0000FFF8" />
+        <register type="OA" address="0x0000D948" value="0x000000C0" />
+        <register type="OA" address="0x0000D94C" value="0x0000FFE7" />
+        <register type="OA" address="0x0000DC08" value="0x000000C0" />
+        <register type="OA" address="0x0000DC0C" value="0x0000FFE7" />
+        <register type="OA" address="0x0000D950" value="0x00000300" />
+        <register type="OA" address="0x0000D954" value="0x0000FF9F" />
+        <register type="OA" address="0x0000DC10" value="0x00000300" />
+        <register type="OA" address="0x0000DC14" value="0x0000FF9F" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="Metric set RasterizerAndPixelBackend2"
+       chipset="DG2"
+       symbol_name="RasterizerAndPixelBackend2"
+       underscore_name="rasterizer_and_pixel_backend2"
+       mdapi_supported_apis="VK OGL OCL IO"
+       hw_config_guid="57a0abf4-8d81-4856-ab86-01f15f92c81c"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Rasterized Pixels"
+             symbol_name="RasterizedPixels"
+             underscore_name="rasterized_pixels"
+             description="The total number of rasterized pixels."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 21 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Early Hi-Depth Test Fails"
+             symbol_name="HiDepthTestFails"
+             underscore_name="hi_depth_test_fails"
+             description="The total number of pixels dropped on early hierarchical depth test."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 22 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer/Hi-Depth Test"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Early Depth Test Fails"
+             symbol_name="EarlyDepthTestFails"
+             underscore_name="early_depth_test_fails"
+             description="The total number of pixels dropped on early depth test."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 23 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer/Early Depth Test"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Killed in FS"
+             symbol_name="SamplesKilledInPs"
+             underscore_name="samples_killed_in_ps"
+             description="The total number of samples or pixels dropped in fragment shaders."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 24 READ 4 UMUL"
+             mdapi_group="3D Pipe/Fragment Shader"
+             mdapi_usage_flags="Tier4 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Pixels Failing Tests"
+             symbol_name="PixelsFailingPostPsTests"
+             underscore_name="pixels_failing_post_ps_tests"
+             description="The total number of pixels dropped on post-FS alpha, stencil, or depth tests."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 25 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Written"
+             symbol_name="SamplesWritten"
+             underscore_name="samples_written"
+             description="The total number of samples or pixels written to all render targets."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 26 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Blended"
+             symbol_name="SamplesBlended"
+             underscore_name="samples_blended"
+             description="The total number of blended samples or pixels written to all render targets."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 27 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Texels"
+             symbol_name="SamplerTexels"
+             underscore_name="sampler_texels"
+             description="The total number of texels seen on input (with 2x2 accuracy) in all sampler units."
+             data_type="uint64"
+             units="texels"
+             semantic_type="event"
+             equation="A 28 READ 4 UMUL"
+             mdapi_group="Sampler/Sampler Input"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Texels Misses"
+             symbol_name="SamplerTexelMisses"
+             underscore_name="sampler_texel_misses"
+             description="The total number of texels lookups (with 2x2 accuracy) that missed L1 sampler cache."
+             data_type="uint64"
+             units="texels"
+             semantic_type="event"
+             equation="A 29 READ 4 UMUL"
+             mdapi_group="Sampler/Sampler Cache"
+             mdapi_usage_flags="Tier3 Batch Frame Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SLM Bytes Read"
+             symbol_name="SlmBytesRead"
+             underscore_name="slm_bytes_read"
+             description="The total number of GPU memory bytes read from shared local memory."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="A 30 READ 64 UMUL"
+             mdapi_group="L3/Data Port/SLM"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SLM Bytes Written"
+             symbol_name="SlmBytesWritten"
+             underscore_name="slm_bytes_written"
+             description="The total number of GPU memory bytes written into shared local memory."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="A 31 READ 64 UMUL"
+             mdapi_group="L3/Data Port/SLM"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Memory Accesses"
+             symbol_name="ShaderMemoryAccesses"
+             underscore_name="shader_memory_accesses"
+             description="The total number of shader memory accesses to L3."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 32 READ"
+             mdapi_group="L3/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Atomic Memory Accesses"
+             symbol_name="ShaderAtomics"
+             underscore_name="shader_atomics"
+             description="The total number of shader atomic memory accesses."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 34 READ"
+             mdapi_group="L3/Data Port/Atomics"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="L3 Shader Throughput"
+             symbol_name="L3ShaderThroughput"
+             underscore_name="l3_shader_throughput"
+             description="The total number of GPU memory bytes transferred between shaders and L3 caches w/o URB."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="$ShaderMemoryAccesses 64 UMUL"
+             mdapi_group="L3/Data Port"
+             mdapi_usage_flags="Tier2 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Barrier Messages"
+             symbol_name="ShaderBarriers"
+             underscore_name="shader_barriers"
+             description="The total number of shader barrier messages."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 35 READ"
+             mdapi_group="EU Array/Barrier"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Slice1 Rasterizer Input Available"
+             symbol_name="Rasterizer1InputAvailable"
+             underscore_name="rasterizer1_input_available"
+             description="The percentage of time in which slice1 rasterizer input is available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$SliceMask 2 AND"
+             mdapi_group="GPU/Rasterizer"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Slice0 Rasterizer Output Ready"
+             symbol_name="Rasterizer0OutputReady"
+             underscore_name="rasterizer0_output_ready"
+             description="The percentage of time in which slice0 rasterizer output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$SliceMask 1 AND"
+             mdapi_group="GPU/Rasterizer"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Slice1 Pipe0 Post-EarlyZ Pixel Data Ready"
+             symbol_name="PixelData10Ready"
+             underscore_name="pixel_data10_ready"
+             description="The percentage of time in which slice1  pipe0 post-EarlyZ pixel data is ready (after early Z tests have been applied)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU/Rasterizer/Early Depth Test"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Slice1 Pipe1 Post-EarlyZ Pixel Data Ready"
+             symbol_name="PixelData11Ready"
+             underscore_name="pixel_data11_ready"
+             description="The percentage of time in which slice1  pipe1 post-EarlyZ pixel data is ready (after early Z tests have been applied)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU/Rasterizer/Early Depth Test"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x0000000C" />
+        <register type="NOA" address="0x00009888" value="0x1A057000" />
+        <register type="NOA" address="0x00009888" value="0x1B056000" />
+        <register type="NOA" address="0x00009888" value="0x1D050000" />
+        <register type="NOA" address="0x00009888" value="0x0D0B701F" />
+        <register type="NOA" address="0x00009888" value="0x0F0B6000" />
+        <register type="NOA" address="0x00009888" value="0x0D0C7000" />
+        <register type="NOA" address="0x00009888" value="0x0F0C6000" />
+        <register type="NOA" address="0x00009888" value="0x0003007F" />
+        <register type="NOA" address="0x00009888" value="0x00050000" />
+        <register type="NOA" address="0x00009888" value="0x10050DC0" />
+        <register type="NOA" address="0x00009888" value="0x02050000" />
+        <register type="NOA" address="0x00009888" value="0x00068000" />
+        <register type="NOA" address="0x00009888" value="0x00083FFF" />
+        <register type="NOA" address="0x00009888" value="0x02084000" />
+        <register type="NOA" address="0x00009888" value="0x000A1555" />
+        <register type="NOA" address="0x00009888" value="0x020A4000" />
+        <register type="NOA" address="0x00009888" value="0x0001007F" />
+        <register type="NOA" address="0x00009888" value="0x000B2AAA" />
+        <register type="NOA" address="0x00009888" value="0x01050000" />
+        <register type="NOA" address="0x00009888" value="0x030535D2" />
+        <register type="NOA" address="0x00009888" value="0x0505003E" />
+        <register type="NOA" address="0x00009888" value="0x01060007" />
+        <register type="NOA" address="0x00009888" value="0x01070280" />
+        <register type="NOA" address="0x00009888" value="0x01083C15" />
+        <register type="NOA" address="0x00009888" value="0x01090018" />
+        <register type="NOA" address="0x00009888" value="0x010A1695" />
+        <register type="NOA" address="0x00009888" value="0x01010060" />
+        <register type="NOA" address="0x00009888" value="0x050B0200" />
+        <register type="NOA" address="0x00009888" value="0x0B0B0400" />
+        <register type="NOA" address="0x00009888" value="0x090B0000" />
+        <register type="NOA" address="0x00009888" value="0x010B0000" />
+        <register type="NOA" address="0x00009888" value="0x01020018" />
+        <register type="NOA" address="0x00009888" value="0x050C0200" />
+        <register type="NOA" address="0x00009888" value="0x090C0400" />
+        <register type="NOA" address="0x00009888" value="0x0B0C001F" />
+        <register type="NOA" address="0x00009888" value="0x010C0000" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x00082AAA" />
+        <register type="NOA" address="0x00009888" value="0x02088000" />
+        <register type="NOA" address="0x00009888" value="0x00032492" />
+        <register type="NOA" address="0x00009888" value="0x02030012" />
+        <register type="NOA" address="0x00009888" value="0x06030002" />
+        <register type="NOA" address="0x00009888" value="0x14030000" />
+        <register type="NOA" address="0x00009888" value="0x3E030006" />
+        <register type="NOA" address="0x00009888" value="0x26030000" />
+        <register type="NOA" address="0x00009888" value="0x420318C3" />
+        <register type="NOA" address="0x00009888" value="0x2C030000" />
+        <register type="NOA" address="0x00009888" value="0x32030000" />
+        <register type="NOA" address="0x00009888" value="0x44030000" />
+        <register type="NOA" address="0x00009888" value="0x38030000" />
+        <register type="NOA" address="0x00009888" value="0x46030603" />
+        <register type="NOA" address="0x00009888" value="0x20037000" />
+        <register type="NOA" address="0x00009884" value="0x00000008" />
+        <register type="NOA" address="0x00009888" value="0x0001807F" />
+        <register type="NOA" address="0x00009888" value="0x00021555" />
+        <register type="NOA" address="0x00009888" value="0x02024000" />
+        <register type="NOA" address="0x00009888" value="0x00041555" />
+        <register type="NOA" address="0x00009888" value="0x02044000" />
+        <register type="NOA" address="0x00009888" value="0x00051555" />
+        <register type="NOA" address="0x00009888" value="0x02054000" />
+        <register type="NOA" address="0x00009888" value="0x00082AAA" />
+        <register type="NOA" address="0x00009888" value="0x02088000" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x70800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00070000" />
+        <register type="OA" address="0x0000D940" value="0x00000038" />
+        <register type="OA" address="0x0000D944" value="0x0000FFF8" />
+        <register type="OA" address="0x0000DC00" value="0x00000038" />
+        <register type="OA" address="0x0000DC04" value="0x0000FFF8" />
+        <register type="OA" address="0x0000D948" value="0x000000C0" />
+        <register type="OA" address="0x0000D94C" value="0x0000FFE7" />
+        <register type="OA" address="0x0000DC08" value="0x000000C0" />
+        <register type="OA" address="0x0000DC0C" value="0x0000FFE7" />
+        <register type="OA" address="0x0000D950" value="0x00000300" />
+        <register type="OA" address="0x0000D954" value="0x0000FF9F" />
+        <register type="OA" address="0x0000DC10" value="0x00000300" />
+        <register type="OA" address="0x0000DC14" value="0x0000FF9F" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="Sampler_1"
+       chipset="DG2"
+       symbol_name="Sampler_1"
+       underscore_name="sampler_1"
+       mdapi_supported_apis="VK OGL OCL MEDIA IO"
+       hw_config_guid="2314eaa3-164f-4004-9684-3f1d16e875f9"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Slice0 DualSubslice0 Input Available"
+             symbol_name="Sampler00InputAvailable"
+             underscore_name="sampler00_input_available"
+             description="The percentage of time in which slice0 dualsubslice0 sampler input is available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 1 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Slice0 DualSubslice1 Input Available"
+             symbol_name="Sampler01InputAvailable"
+             underscore_name="sampler01_input_available"
+             description="The percentage of time in which slice0 dualsubslice1 sampler input is available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 2 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Slice0 DualSubslice2 Input Available"
+             symbol_name="Sampler02InputAvailable"
+             underscore_name="sampler02_input_available"
+             description="The percentage of time in which slice0 dualsubslice2 sampler input is available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 4 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Slice0 DualSubslice3 Input Available"
+             symbol_name="Sampler03InputAvailable"
+             underscore_name="sampler03_input_available"
+             description="The percentage of time in which slice0 dualsubslice3 sampler input is available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 8 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Slice0 DualSubslice0 Sampler Output Ready"
+             symbol_name="Sampler00OutputReady"
+             underscore_name="sampler00_output_ready"
+             description="The percentage of time in which slice0 dualsubslice0 sampler output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 1 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Slice0 DualSubslice1 Sampler Output Ready"
+             symbol_name="Sampler01OutputReady"
+             underscore_name="sampler01_output_ready"
+             description="The percentage of time in which slice0 dualsubslice1 sampler output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 2 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Slice0 DualSubslice2 Sampler Output Ready"
+             symbol_name="Sampler02OutputReady"
+             underscore_name="sampler02_output_ready"
+             description="The percentage of time in which slice0 dualsubslice2 sampler output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 4 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Slice0 DualSubslice3 Sampler Output Ready"
+             symbol_name="Sampler03OutputReady"
+             underscore_name="sampler03_output_ready"
+             description="The percentage of time in which slice0 dualsubslice3 sampler output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 8 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x0000000C" />
+        <register type="NOA" address="0x00009888" value="0x1C824800" />
+        <register type="NOA" address="0x00009888" value="0x14832C00" />
+        <register type="NOA" address="0x00009888" value="0x1CA24800" />
+        <register type="NOA" address="0x00009888" value="0x14A32C00" />
+        <register type="NOA" address="0x00009888" value="0x1CC24800" />
+        <register type="NOA" address="0x00009888" value="0x14C32C00" />
+        <register type="NOA" address="0x00009888" value="0x1CE24800" />
+        <register type="NOA" address="0x00009888" value="0x14E32C00" />
+        <register type="NOA" address="0x00009888" value="0x04824000" />
+        <register type="NOA" address="0x00009888" value="0x16820027" />
+        <register type="NOA" address="0x00009888" value="0x02820000" />
+        <register type="NOA" address="0x00009888" value="0x0E830580" />
+        <register type="NOA" address="0x00009888" value="0x04830000" />
+        <register type="NOA" address="0x00009888" value="0x02834000" />
+        <register type="NOA" address="0x00009888" value="0x00848800" />
+        <register type="NOA" address="0x00009888" value="0x02864040" />
+        <register type="NOA" address="0x00009888" value="0x04A24000" />
+        <register type="NOA" address="0x00009888" value="0x14A209C0" />
+        <register type="NOA" address="0x00009888" value="0x02A20000" />
+        <register type="NOA" address="0x00009888" value="0x0EA30016" />
+        <register type="NOA" address="0x00009888" value="0x04A30000" />
+        <register type="NOA" address="0x00009888" value="0x02A31000" />
+        <register type="NOA" address="0x00009888" value="0x00A44400" />
+        <register type="NOA" address="0x00009888" value="0x02A61010" />
+        <register type="NOA" address="0x00009888" value="0x04C24000" />
+        <register type="NOA" address="0x00009888" value="0x14C20027" />
+        <register type="NOA" address="0x00009888" value="0x02C20000" />
+        <register type="NOA" address="0x00009888" value="0x0CC30580" />
+        <register type="NOA" address="0x00009888" value="0x04C30000" />
+        <register type="NOA" address="0x00009888" value="0x02C30400" />
+        <register type="NOA" address="0x00009888" value="0x00C42200" />
+        <register type="NOA" address="0x00009888" value="0x02C60404" />
+        <register type="NOA" address="0x00009888" value="0x04E24000" />
+        <register type="NOA" address="0x00009888" value="0x12E209C0" />
+        <register type="NOA" address="0x00009888" value="0x02E20000" />
+        <register type="NOA" address="0x00009888" value="0x0CE30016" />
+        <register type="NOA" address="0x00009888" value="0x04E30000" />
+        <register type="NOA" address="0x00009888" value="0x02E30100" />
+        <register type="NOA" address="0x00009888" value="0x00E41100" />
+        <register type="NOA" address="0x00009888" value="0x02E60101" />
+        <register type="NOA" address="0x00009888" value="0x02070909" />
+        <register type="NOA" address="0x00009888" value="0x0208B0B0" />
+        <register type="NOA" address="0x00009888" value="0x00093300" />
+        <register type="NOA" address="0x00009888" value="0x020A5A5A" />
+        <register type="NOA" address="0x00009888" value="0x00014400" />
+        <register type="NOA" address="0x00009888" value="0x020B1010" />
+        <register type="NOA" address="0x00009888" value="0x00022200" />
+        <register type="NOA" address="0x00009888" value="0x020C0404" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009888" value="0x02032400" />
+        <register type="NOA" address="0x00009888" value="0x04032492" />
+        <register type="NOA" address="0x00009888" value="0x06030002" />
+        <register type="NOA" address="0x00009888" value="0x48030200" />
+        <register type="NOA" address="0x00009888" value="0x4A030803" />
+        <register type="NOA" address="0x00009888" value="0x16030000" />
+        <register type="NOA" address="0x00009888" value="0x3E0318E0" />
+        <register type="NOA" address="0x00009888" value="0x18031000" />
+        <register type="NOA" address="0x00009888" value="0x1A030000" />
+        <register type="NOA" address="0x00009888" value="0x40030C0A" />
+        <register type="NOA" address="0x00009888" value="0x1C030000" />
+        <register type="NOA" address="0x00009888" value="0x1E030000" />
+        <register type="NOA" address="0x00009888" value="0x20031000" />
+        <register type="NOA" address="0x00009888" value="0x42030002" />
+        <register type="NOA" address="0x00009884" value="0x00000008" />
+        <register type="NOA" address="0x00009888" value="0x0001FF00" />
+        <register type="NOA" address="0x00009888" value="0x02025555" />
+        <register type="NOA" address="0x00009888" value="0x02045555" />
+        <register type="NOA" address="0x00009888" value="0x02055555" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x00800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00000000" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="Sampler_2"
+       chipset="DG2"
+       symbol_name="Sampler_2"
+       underscore_name="sampler_2"
+       mdapi_supported_apis="VK OGL OCL MEDIA IO"
+       hw_config_guid="c60dea82-db13-4882-8d31-a02bce151ef7"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Slice1 DualSubslice0 Input Available"
+             symbol_name="Sampler10InputAvailable"
+             underscore_name="sampler10_input_available"
+             description="The percentage of time in which slice1 dualsubslice0 sampler input is available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 64 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Slice1 DualSubslice1 Input Available"
+             symbol_name="Sampler11InputAvailable"
+             underscore_name="sampler11_input_available"
+             description="The percentage of time in which slice1 dualsubslice1 sampler input is available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 128 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Slice1 DualSubslice2 Input Available"
+             symbol_name="Sampler12InputAvailable"
+             underscore_name="sampler12_input_available"
+             description="The percentage of time in which slice1 dualsubslice2 sampler input is available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 256 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Slice1 DualSubslice3 Input Available"
+             symbol_name="Sampler13InputAvailable"
+             underscore_name="sampler13_input_available"
+             description="The percentage of time in which slice1 dualsubslice3 sampler input is available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 512 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Slice1 DualSubslice0 Sampler Output Ready"
+             symbol_name="Sampler10OutputReady"
+             underscore_name="sampler10_output_ready"
+             description="The percentage of time in which slice1 dualsubslice0 sampler output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 64 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Slice1 DualSubslice1 Sampler Output Ready"
+             symbol_name="Sampler11OutputReady"
+             underscore_name="sampler11_output_ready"
+             description="The percentage of time in which slice1 dualsubslice1 sampler output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 128 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Slice1 DualSubslice2 Sampler Output Ready"
+             symbol_name="Sampler12OutputReady"
+             underscore_name="sampler12_output_ready"
+             description="The percentage of time in which slice1 dualsubslice2 sampler output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 256 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Slice1 DualSubslice3 Sampler Output Ready"
+             symbol_name="Sampler13OutputReady"
+             underscore_name="sampler13_output_ready"
+             description="The percentage of time in which slice1 dualsubslice3 sampler output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 512 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x0000000C" />
+        <register type="NOA" address="0x00009888" value="0x1D824800" />
+        <register type="NOA" address="0x00009888" value="0x15832C00" />
+        <register type="NOA" address="0x00009888" value="0x1DA24800" />
+        <register type="NOA" address="0x00009888" value="0x15A32C00" />
+        <register type="NOA" address="0x00009888" value="0x1DC24800" />
+        <register type="NOA" address="0x00009888" value="0x15C32C00" />
+        <register type="NOA" address="0x00009888" value="0x1DE24800" />
+        <register type="NOA" address="0x00009888" value="0x15E32C00" />
+        <register type="NOA" address="0x00009888" value="0x05824000" />
+        <register type="NOA" address="0x00009888" value="0x17820027" />
+        <register type="NOA" address="0x00009888" value="0x03820000" />
+        <register type="NOA" address="0x00009888" value="0x0F830580" />
+        <register type="NOA" address="0x00009888" value="0x05830000" />
+        <register type="NOA" address="0x00009888" value="0x03834000" />
+        <register type="NOA" address="0x00009888" value="0x01848800" />
+        <register type="NOA" address="0x00009888" value="0x03864040" />
+        <register type="NOA" address="0x00009888" value="0x05A24000" />
+        <register type="NOA" address="0x00009888" value="0x15A209C0" />
+        <register type="NOA" address="0x00009888" value="0x03A20000" />
+        <register type="NOA" address="0x00009888" value="0x0FA30016" />
+        <register type="NOA" address="0x00009888" value="0x05A30000" />
+        <register type="NOA" address="0x00009888" value="0x03A31000" />
+        <register type="NOA" address="0x00009888" value="0x01A44400" />
+        <register type="NOA" address="0x00009888" value="0x03A61010" />
+        <register type="NOA" address="0x00009888" value="0x05C24000" />
+        <register type="NOA" address="0x00009888" value="0x15C20027" />
+        <register type="NOA" address="0x00009888" value="0x03C20000" />
+        <register type="NOA" address="0x00009888" value="0x0DC30580" />
+        <register type="NOA" address="0x00009888" value="0x05C30000" />
+        <register type="NOA" address="0x00009888" value="0x03C30400" />
+        <register type="NOA" address="0x00009888" value="0x01C42200" />
+        <register type="NOA" address="0x00009888" value="0x03C60404" />
+        <register type="NOA" address="0x00009888" value="0x05E24000" />
+        <register type="NOA" address="0x00009888" value="0x13E209C0" />
+        <register type="NOA" address="0x00009888" value="0x03E20000" />
+        <register type="NOA" address="0x00009888" value="0x0DE30016" />
+        <register type="NOA" address="0x00009888" value="0x05E30000" />
+        <register type="NOA" address="0x00009888" value="0x03E30100" />
+        <register type="NOA" address="0x00009888" value="0x01E41100" />
+        <register type="NOA" address="0x00009888" value="0x03E60101" />
+        <register type="NOA" address="0x00009888" value="0x0003FF00" />
+        <register type="NOA" address="0x00009888" value="0x0208FFFF" />
+        <register type="NOA" address="0x00009888" value="0x020A5555" />
+        <register type="NOA" address="0x00009888" value="0x0001FF00" />
+        <register type="NOA" address="0x00009888" value="0x020BAAAA" />
+        <register type="NOA" address="0x00009888" value="0x03070909" />
+        <register type="NOA" address="0x00009888" value="0x0308B0B0" />
+        <register type="NOA" address="0x00009888" value="0x01093300" />
+        <register type="NOA" address="0x00009888" value="0x030A5A5A" />
+        <register type="NOA" address="0x00009888" value="0x01014400" />
+        <register type="NOA" address="0x00009888" value="0x030B1010" />
+        <register type="NOA" address="0x00009888" value="0x01022200" />
+        <register type="NOA" address="0x00009888" value="0x030C0404" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009888" value="0x02032400" />
+        <register type="NOA" address="0x00009888" value="0x04032492" />
+        <register type="NOA" address="0x00009888" value="0x06030002" />
+        <register type="NOA" address="0x00009888" value="0x48030200" />
+        <register type="NOA" address="0x00009888" value="0x4A030803" />
+        <register type="NOA" address="0x00009888" value="0x16030000" />
+        <register type="NOA" address="0x00009888" value="0x3E0318E0" />
+        <register type="NOA" address="0x00009888" value="0x18031000" />
+        <register type="NOA" address="0x00009888" value="0x1A030000" />
+        <register type="NOA" address="0x00009888" value="0x40030C0A" />
+        <register type="NOA" address="0x00009888" value="0x1C030000" />
+        <register type="NOA" address="0x00009888" value="0x1E030000" />
+        <register type="NOA" address="0x00009888" value="0x20031000" />
+        <register type="NOA" address="0x00009888" value="0x42030002" />
+        <register type="NOA" address="0x00009884" value="0x00000008" />
+        <register type="NOA" address="0x00009888" value="0x0001FF00" />
+        <register type="NOA" address="0x00009888" value="0x02025555" />
+        <register type="NOA" address="0x00009888" value="0x02045555" />
+        <register type="NOA" address="0x00009888" value="0x02055555" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x00800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00000000" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="TDL_1"
+       chipset="DG2"
+       symbol_name="TDL_1"
+       underscore_name="tdl_1"
+       mdapi_supported_apis="VK OGL OCL MEDIA IO"
+       hw_config_guid="7f5cb8ec-38bb-48c9-9ad2-60075eba2833"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Non-PS Thread Ready For Dispatch on Slice0 DualSubslice0 Thread Dispatcher"
+             symbol_name="NonPSThread00ReadyForDispatch"
+             underscore_name="non_ps_thread00_ready_for_dispatch"
+             description="The percentage of time in which non-PS thread is ready for dispatch on slice0 dualsubslice0 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 1 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Non-PS Thread Ready For Dispatch on Slice0 DualSubslice1 Thread Dispatcher"
+             symbol_name="NonPSThread01ReadyForDispatch"
+             underscore_name="non_ps_thread01_ready_for_dispatch"
+             description="The percentage of time in which non-PS thread is ready for dispatch on slice0 dualsubslice1 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 2 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Non-PS Thread Ready For Dispatch on Slice0 DualSubslice2 Thread Dispatcher"
+             symbol_name="NonPSThread02ReadyForDispatch"
+             underscore_name="non_ps_thread02_ready_for_dispatch"
+             description="The percentage of time in which non-PS thread is ready for dispatch on slice0 dualsubslice2 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 4 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Non-PS Thread Ready For Dispatch on Slice0 DualSubslice3 Thread Dispatcher"
+             symbol_name="NonPSThread03ReadyForDispatch"
+             underscore_name="non_ps_thread03_ready_for_dispatch"
+             description="The percentage of time in which non-PS thread is ready for dispatch on slice0 dualsubslice3 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 8 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Non-PS Thread Ready For Dispatch on Slice1 DualSubslice0 Thread Dispatcher"
+             symbol_name="NonPSThread10ReadyForDispatch"
+             underscore_name="non_ps_thread10_ready_for_dispatch"
+             description="The percentage of time in which non-PS thread is ready for dispatch on slice1 dualsubslice0 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 64 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Non-PS Thread Ready For Dispatch on Slice1 DualSubslice1 Thread Dispatcher"
+             symbol_name="NonPSThread11ReadyForDispatch"
+             underscore_name="non_ps_thread11_ready_for_dispatch"
+             description="The percentage of time in which non-PS thread is ready for dispatch on slice1 dualsubslice1 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 128 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Non-PS Thread Ready For Dispatch on Slice1 DualSubslice2 Thread Dispatcher"
+             symbol_name="NonPSThread12ReadyForDispatch"
+             underscore_name="non_ps_thread12_ready_for_dispatch"
+             description="The percentage of time in which non-PS thread is ready for dispatch on slice1 dualsubslice2 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 256 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Non-PS Thread Ready For Dispatch on Slice1 DualSubslice3 Thread Dispatcher"
+             symbol_name="NonPSThread13ReadyForDispatch"
+             underscore_name="non_ps_thread13_ready_for_dispatch"
+             description="The percentage of time in which non-PS thread is ready for dispatch on slice1 dualsubslice3 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 512 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice3 Thread Dispatcher Port 0"
+             symbol_name="ThreadHeader03ReadyPort0"
+             underscore_name="thread_header03_ready_port0"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice3 thread dispatcher port 0"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 8 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice3 Thread Dispatcher Port 1"
+             symbol_name="ThreadHeader03ReadyPort1"
+             underscore_name="thread_header03_ready_port1"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice3 thread dispatcher port 1"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 8 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice3 Thread Dispatcher Port 2"
+             symbol_name="ThreadHeader03ReadyPort2"
+             underscore_name="thread_header03_ready_port2"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice3 thread dispatcher port 2"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 8 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice3 Thread Dispatcher Port 3"
+             symbol_name="ThreadHeader03ReadyPort3"
+             underscore_name="thread_header03_ready_port3"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice3 thread dispatcher port 3"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 8 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice3 Thread Dispatcher Port 0"
+             symbol_name="ThreadHeader13ReadyPort0"
+             underscore_name="thread_header13_ready_port0"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice3 thread dispatcher port 0"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 512 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice3 Thread Dispatcher Port 1"
+             symbol_name="ThreadHeader13ReadyPort1"
+             underscore_name="thread_header13_ready_port1"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice3 thread dispatcher port 1"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 512 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice3 Thread Dispatcher Port 2"
+             symbol_name="ThreadHeader13ReadyPort2"
+             underscore_name="thread_header13_ready_port2"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice3 thread dispatcher port 2"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 512 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice3 Thread Dispatcher Port 3"
+             symbol_name="ThreadHeader13ReadyPort3"
+             underscore_name="thread_header13_ready_port3"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice3 thread dispatcher port 3"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 512 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x0000000C" />
+        <register type="NOA" address="0x00009888" value="0x2A81001C" />
+        <register type="NOA" address="0x00009888" value="0x2AA1001C" />
+        <register type="NOA" address="0x00009888" value="0x2AC1001C" />
+        <register type="NOA" address="0x00009888" value="0x2AE1001C" />
+        <register type="NOA" address="0x00009888" value="0x2B81001C" />
+        <register type="NOA" address="0x00009888" value="0x2BA1001C" />
+        <register type="NOA" address="0x00009888" value="0x2BC1001C" />
+        <register type="NOA" address="0x00009888" value="0x2BE1001C" />
+        <register type="NOA" address="0x00009888" value="0x06810300" />
+        <register type="NOA" address="0x00009888" value="0x08810030" />
+        <register type="NOA" address="0x00009888" value="0x00810000" />
+        <register type="NOA" address="0x00009888" value="0x00830002" />
+        <register type="NOA" address="0x00009888" value="0x00840001" />
+        <register type="NOA" address="0x00009888" value="0x00860001" />
+        <register type="NOA" address="0x00009888" value="0x00870001" />
+        <register type="NOA" address="0x00009888" value="0x06A10300" />
+        <register type="NOA" address="0x00009888" value="0x08A10C00" />
+        <register type="NOA" address="0x00009888" value="0x00A10000" />
+        <register type="NOA" address="0x00009888" value="0x00A30008" />
+        <register type="NOA" address="0x00009888" value="0x00A40002" />
+        <register type="NOA" address="0x00009888" value="0x00A60004" />
+        <register type="NOA" address="0x00009888" value="0x00A70004" />
+        <register type="NOA" address="0x00009888" value="0x06C10300" />
+        <register type="NOA" address="0x00009888" value="0x0AC10030" />
+        <register type="NOA" address="0x00009888" value="0x08C10000" />
+        <register type="NOA" address="0x00009888" value="0x00C10000" />
+        <register type="NOA" address="0x00009888" value="0x00C30020" />
+        <register type="NOA" address="0x00009888" value="0x00C40004" />
+        <register type="NOA" address="0x00009888" value="0x00C60010" />
+        <register type="NOA" address="0x00009888" value="0x00C70010" />
+        <register type="NOA" address="0x00009888" value="0x06E10300" />
+        <register type="NOA" address="0x00009888" value="0x0AE10C00" />
+        <register type="NOA" address="0x00009888" value="0x14E10CF4" />
+        <register type="NOA" address="0x00009888" value="0x16E10C72" />
+        <register type="NOA" address="0x00009888" value="0x08E10000" />
+        <register type="NOA" address="0x00009888" value="0x00E10000" />
+        <register type="NOA" address="0x00009888" value="0x02E10000" />
+        <register type="NOA" address="0x00009888" value="0x00E30080" />
+        <register type="NOA" address="0x00009888" value="0x02E3AA00" />
+        <register type="NOA" address="0x00009888" value="0x00E4F008" />
+        <register type="NOA" address="0x00009888" value="0x00E60040" />
+        <register type="NOA" address="0x00009888" value="0x02E65500" />
+        <register type="NOA" address="0x00009888" value="0x00E70040" />
+        <register type="NOA" address="0x00009888" value="0x02E75500" />
+        <register type="NOA" address="0x00009888" value="0x07810300" />
+        <register type="NOA" address="0x00009888" value="0x0D810030" />
+        <register type="NOA" address="0x00009888" value="0x09810000" />
+        <register type="NOA" address="0x00009888" value="0x01810000" />
+        <register type="NOA" address="0x00009888" value="0x01830200" />
+        <register type="NOA" address="0x00009888" value="0x01840010" />
+        <register type="NOA" address="0x00009888" value="0x01860100" />
+        <register type="NOA" address="0x00009888" value="0x01870100" />
+        <register type="NOA" address="0x00009888" value="0x07A10300" />
+        <register type="NOA" address="0x00009888" value="0x0DA10C00" />
+        <register type="NOA" address="0x00009888" value="0x09A10000" />
+        <register type="NOA" address="0x00009888" value="0x01A10000" />
+        <register type="NOA" address="0x00009888" value="0x01A30800" />
+        <register type="NOA" address="0x00009888" value="0x01A40020" />
+        <register type="NOA" address="0x00009888" value="0x01A60400" />
+        <register type="NOA" address="0x00009888" value="0x01A70400" />
+        <register type="NOA" address="0x00009888" value="0x07C10300" />
+        <register type="NOA" address="0x00009888" value="0x0FC10030" />
+        <register type="NOA" address="0x00009888" value="0x09C10000" />
+        <register type="NOA" address="0x00009888" value="0x01C10000" />
+        <register type="NOA" address="0x00009888" value="0x01C32000" />
+        <register type="NOA" address="0x00009888" value="0x01C40040" />
+        <register type="NOA" address="0x00009888" value="0x01C61000" />
+        <register type="NOA" address="0x00009888" value="0x01C71000" />
+        <register type="NOA" address="0x00009888" value="0x07E10300" />
+        <register type="NOA" address="0x00009888" value="0x0FE10C00" />
+        <register type="NOA" address="0x00009888" value="0x11E10CF4" />
+        <register type="NOA" address="0x00009888" value="0x13E10C72" />
+        <register type="NOA" address="0x00009888" value="0x09E10000" />
+        <register type="NOA" address="0x00009888" value="0x01E10000" />
+        <register type="NOA" address="0x00009888" value="0x03E10000" />
+        <register type="NOA" address="0x00009888" value="0x01E38000" />
+        <register type="NOA" address="0x00009888" value="0x03E300AA" />
+        <register type="NOA" address="0x00009888" value="0x01E40F80" />
+        <register type="NOA" address="0x00009888" value="0x01E64000" />
+        <register type="NOA" address="0x00009888" value="0x03E60055" />
+        <register type="NOA" address="0x00009888" value="0x01E74000" />
+        <register type="NOA" address="0x00009888" value="0x03E70055" />
+        <register type="NOA" address="0x00009888" value="0x00030FF0" />
+        <register type="NOA" address="0x00009888" value="0x00070060" />
+        <register type="NOA" address="0x00009888" value="0x02075500" />
+        <register type="NOA" address="0x00009888" value="0x0008FF0E" />
+        <register type="NOA" address="0x00009888" value="0x020800FF" />
+        <register type="NOA" address="0x00009888" value="0x0009F00C" />
+        <register type="NOA" address="0x00009888" value="0x000A55A5" />
+        <register type="NOA" address="0x00009888" value="0x020AAA55" />
+        <register type="NOA" address="0x00009888" value="0x00010FF2" />
+        <register type="NOA" address="0x00009888" value="0x000BAA04" />
+        <register type="NOA" address="0x00009888" value="0x020B00AA" />
+        <register type="NOA" address="0x00009888" value="0x00020004" />
+        <register type="NOA" address="0x00009888" value="0x000C0010" />
+        <register type="NOA" address="0x00009888" value="0x01076000" />
+        <register type="NOA" address="0x00009888" value="0x03070055" />
+        <register type="NOA" address="0x00009888" value="0x01080E00" />
+        <register type="NOA" address="0x00009888" value="0x01090FC0" />
+        <register type="NOA" address="0x00009888" value="0x010AA500" />
+        <register type="NOA" address="0x00009888" value="0x030A00AA" />
+        <register type="NOA" address="0x00009888" value="0x01010020" />
+        <register type="NOA" address="0x00009888" value="0x010B0400" />
+        <register type="NOA" address="0x00009888" value="0x01020040" />
+        <register type="NOA" address="0x00009888" value="0x010C1000" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009888" value="0x00032492" />
+        <register type="NOA" address="0x00009888" value="0x02032492" />
+        <register type="NOA" address="0x00009888" value="0x04032492" />
+        <register type="NOA" address="0x00009888" value="0x06030002" />
+        <register type="NOA" address="0x00009888" value="0x14037000" />
+        <register type="NOA" address="0x00009888" value="0x3E03294A" />
+        <register type="NOA" address="0x00009888" value="0x26037000" />
+        <register type="NOA" address="0x00009888" value="0x42032007" />
+        <register type="NOA" address="0x00009888" value="0x2C033000" />
+        <register type="NOA" address="0x00009888" value="0x32037000" />
+        <register type="NOA" address="0x00009888" value="0x44030047" />
+        <register type="NOA" address="0x00009888" value="0x38031000" />
+        <register type="NOA" address="0x00009888" value="0x46030003" />
+        <register type="NOA" address="0x00009888" value="0x480300A0" />
+        <register type="NOA" address="0x00009888" value="0x4A03140A" />
+        <register type="NOA" address="0x00009888" value="0x16030000" />
+        <register type="NOA" address="0x00009888" value="0x18030000" />
+        <register type="NOA" address="0x00009888" value="0x1A037000" />
+        <register type="NOA" address="0x00009888" value="0x40031CE7" />
+        <register type="NOA" address="0x00009888" value="0x1C037000" />
+        <register type="NOA" address="0x00009888" value="0x1E037000" />
+        <register type="NOA" address="0x00009888" value="0x20037000" />
+        <register type="NOA" address="0x00009884" value="0x00000008" />
+        <register type="NOA" address="0x00009888" value="0x0001FFFF" />
+        <register type="NOA" address="0x00009888" value="0x00025555" />
+        <register type="NOA" address="0x00009888" value="0x02025555" />
+        <register type="NOA" address="0x00009888" value="0x00045555" />
+        <register type="NOA" address="0x00009888" value="0x02045555" />
+        <register type="NOA" address="0x00009888" value="0x00055555" />
+        <register type="NOA" address="0x00009888" value="0x02055555" />
+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0xF0800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0xF0800000" />
+        <register type="OA" address="0x0000DC40" value="0x00FF0000" />
+        <register type="OA" address="0x0000D940" value="0x00000002" />
+        <register type="OA" address="0x0000D944" value="0x0000FFFE" />
+        <register type="OA" address="0x0000DC00" value="0x00000002" />
+        <register type="OA" address="0x0000DC04" value="0x0000FFFE" />
+        <register type="OA" address="0x0000D948" value="0x00000002" />
+        <register type="OA" address="0x0000D94C" value="0x0000FFFD" />
+        <register type="OA" address="0x0000DC08" value="0x00000002" />
+        <register type="OA" address="0x0000DC0C" value="0x0000FFFD" />
+        <register type="OA" address="0x0000D950" value="0x00000002" />
+        <register type="OA" address="0x0000D954" value="0x0000FFFB" />
+        <register type="OA" address="0x0000DC10" value="0x00000002" />
+        <register type="OA" address="0x0000DC14" value="0x0000FFFB" />
+        <register type="OA" address="0x0000D958" value="0x00000002" />
+        <register type="OA" address="0x0000D95C" value="0x0000FFF7" />
+        <register type="OA" address="0x0000DC18" value="0x00000002" />
+        <register type="OA" address="0x0000DC1C" value="0x0000FFF7" />
+        <register type="OA" address="0x0000D960" value="0x00000002" />
+        <register type="OA" address="0x0000D964" value="0x0000FFEF" />
+        <register type="OA" address="0x0000DC20" value="0x00000002" />
+        <register type="OA" address="0x0000DC24" value="0x0000FFEF" />
+        <register type="OA" address="0x0000D968" value="0x00000002" />
+        <register type="OA" address="0x0000D96C" value="0x0000FFDF" />
+        <register type="OA" address="0x0000DC28" value="0x00000002" />
+        <register type="OA" address="0x0000DC2C" value="0x0000FFDF" />
+        <register type="OA" address="0x0000D970" value="0x00000002" />
+        <register type="OA" address="0x0000D974" value="0x0000FFBF" />
+        <register type="OA" address="0x0000DC30" value="0x00000002" />
+        <register type="OA" address="0x0000DC34" value="0x0000FFBF" />
+        <register type="OA" address="0x0000D978" value="0x00000002" />
+        <register type="OA" address="0x0000D97C" value="0x0000FF7F" />
+        <register type="OA" address="0x0000DC38" value="0x00000002" />
+        <register type="OA" address="0x0000DC3C" value="0x0000FF7F" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="TDL_2"
+       chipset="DG2"
+       symbol_name="TDL_2"
+       underscore_name="tdl_2"
+       mdapi_supported_apis="VK OGL OCL MEDIA IO"
+       hw_config_guid="141ce507-de6c-428c-95b2-076194d650db"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="PS Thread Ready For Dispatch on Slice0 Dualsubslice0 Thread Dispatcher"
+             symbol_name="PSThread00ReadyForDispatch"
+             underscore_name="ps_thread00_ready_for_dispatch"
+             description="The percentage of time in which PS thread is ready for dispatch on slice0 dualsubslice0 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 1 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="PS Thread Ready For Dispatch on Slice0 Dualsubslice1 Thread Dispatcher"
+             symbol_name="PSThread01ReadyForDispatch"
+             underscore_name="ps_thread01_ready_for_dispatch"
+             description="The percentage of time in which PS thread is ready for dispatch on slice0 dualsubslice1 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 2 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="PS Thread Ready For Dispatch on Slice0 Dualsubslice2 Thread Dispatcher"
+             symbol_name="PSThread02ReadyForDispatch"
+             underscore_name="ps_thread02_ready_for_dispatch"
+             description="The percentage of time in which PS thread is ready for dispatch on slice0 dualsubslice2 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 4 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="PS Thread Ready For Dispatch on Slice0 Dualsubslice3 Thread Dispatcher"
+             symbol_name="PSThread03ReadyForDispatch"
+             underscore_name="ps_thread03_ready_for_dispatch"
+             description="The percentage of time in which PS thread is ready for dispatch on slice0 dualsubslice3 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 8 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="PS Thread Ready For Dispatch on Slice1 Dualsubslice0 Thread Dispatcher"
+             symbol_name="PSThread10ReadyForDispatch"
+             underscore_name="ps_thread10_ready_for_dispatch"
+             description="The percentage of time in which PS thread is ready for dispatch on slice1 dualsubslice0 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 64 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="PS Thread Ready For Dispatch on Slice1 Dualsubslice1 Thread Dispatcher"
+             symbol_name="PSThread11ReadyForDispatch"
+             underscore_name="ps_thread11_ready_for_dispatch"
+             description="The percentage of time in which PS thread is ready for dispatch on slice1 dualsubslice1 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 128 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="PS Thread Ready For Dispatch on Slice1 Dualsubslice2 Thread Dispatcher"
+             symbol_name="PSThread12ReadyForDispatch"
+             underscore_name="ps_thread12_ready_for_dispatch"
+             description="The percentage of time in which PS thread is ready for dispatch on slice1 dualsubslice2 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 256 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="PS Thread Ready For Dispatch on Slice1 Dualsubslice3 Thread Dispatcher"
+             symbol_name="PSThread13ReadyForDispatch"
+             underscore_name="ps_thread13_ready_for_dispatch"
+             description="The percentage of time in which PS thread is ready for dispatch on slice1 dualsubslice3 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 512 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x0000000C" />
+        <register type="NOA" address="0x00009888" value="0x28810340" />
+        <register type="NOA" address="0x00009888" value="0x28A10340" />
+        <register type="NOA" address="0x00009888" value="0x28C10340" />
+        <register type="NOA" address="0x00009888" value="0x28E10340" />
+        <register type="NOA" address="0x00009888" value="0x29810340" />
+        <register type="NOA" address="0x00009888" value="0x29A10340" />
+        <register type="NOA" address="0x00009888" value="0x29C10340" />
+        <register type="NOA" address="0x00009888" value="0x29E10340" />
+        <register type="NOA" address="0x00009888" value="0x06810100" />
+        <register type="NOA" address="0x00009888" value="0x08810C31" />
+        <register type="NOA" address="0x00009888" value="0x00810000" />
+        <register type="NOA" address="0x00009888" value="0x0083000A" />
+        <register type="NOA" address="0x00009888" value="0x00840003" />
+        <register type="NOA" address="0x00009888" value="0x00860005" />
+        <register type="NOA" address="0x00009888" value="0x00870005" />
+        <register type="NOA" address="0x00009888" value="0x06A10100" />
+        <register type="NOA" address="0x00009888" value="0x0AA10C31" />
+        <register type="NOA" address="0x00009888" value="0x08A10000" />
+        <register type="NOA" address="0x00009888" value="0x00A10000" />
+        <register type="NOA" address="0x00009888" value="0x00A300A0" />
+        <register type="NOA" address="0x00009888" value="0x00A4000C" />
+        <register type="NOA" address="0x00009888" value="0x00A60050" />
+        <register type="NOA" address="0x00009888" value="0x00A70050" />
+        <register type="NOA" address="0x00009888" value="0x06C10100" />
+        <register type="NOA" address="0x00009888" value="0x0CC10C31" />
+        <register type="NOA" address="0x00009888" value="0x08C10000" />
+        <register type="NOA" address="0x00009888" value="0x00C10000" />
+        <register type="NOA" address="0x00009888" value="0x00C30A00" />
+        <register type="NOA" address="0x00009888" value="0x00C40030" />
+        <register type="NOA" address="0x00009888" value="0x00C60500" />
+        <register type="NOA" address="0x00009888" value="0x00C70500" />
+        <register type="NOA" address="0x00009888" value="0x06E10100" />
+        <register type="NOA" address="0x00009888" value="0x0EE10C31" />
+        <register type="NOA" address="0x00009888" value="0x08E10000" />
+        <register type="NOA" address="0x00009888" value="0x00E10000" />
+        <register type="NOA" address="0x00009888" value="0x00E3A000" />
+        <register type="NOA" address="0x00009888" value="0x00E400C0" />
+        <register type="NOA" address="0x00009888" value="0x00E65000" />
+        <register type="NOA" address="0x00009888" value="0x00E75000" />
+        <register type="NOA" address="0x00009888" value="0x07810100" />
+        <register type="NOA" address="0x00009888" value="0x11810C31" />
+        <register type="NOA" address="0x00009888" value="0x09810000" />
+        <register type="NOA" address="0x00009888" value="0x03810000" />
+        <register type="NOA" address="0x00009888" value="0x0383000A" />
+        <register type="NOA" address="0x00009888" value="0x01840300" />
+        <register type="NOA" address="0x00009888" value="0x03860005" />
+        <register type="NOA" address="0x00009888" value="0x03870005" />
+        <register type="NOA" address="0x00009888" value="0x07A10100" />
+        <register type="NOA" address="0x00009888" value="0x13A10C31" />
+        <register type="NOA" address="0x00009888" value="0x09A10000" />
+        <register type="NOA" address="0x00009888" value="0x03A10000" />
+        <register type="NOA" address="0x00009888" value="0x03A300A0" />
+        <register type="NOA" address="0x00009888" value="0x01A40C00" />
+        <register type="NOA" address="0x00009888" value="0x03A60050" />
+        <register type="NOA" address="0x00009888" value="0x03A70050" />
+        <register type="NOA" address="0x00009888" value="0x07C10100" />
+        <register type="NOA" address="0x00009888" value="0x15C10C31" />
+        <register type="NOA" address="0x00009888" value="0x09C10000" />
+        <register type="NOA" address="0x00009888" value="0x03C10000" />
+        <register type="NOA" address="0x00009888" value="0x03C30A00" />
+        <register type="NOA" address="0x00009888" value="0x01C43000" />
+        <register type="NOA" address="0x00009888" value="0x03C60500" />
+        <register type="NOA" address="0x00009888" value="0x03C70500" />
+        <register type="NOA" address="0x00009888" value="0x07E10100" />
+        <register type="NOA" address="0x00009888" value="0x17E10C31" />
+        <register type="NOA" address="0x00009888" value="0x09E10000" />
+        <register type="NOA" address="0x00009888" value="0x03E10000" />
+        <register type="NOA" address="0x00009888" value="0x03E3A000" />
+        <register type="NOA" address="0x00009888" value="0x01E4C000" />
+        <register type="NOA" address="0x00009888" value="0x03E65000" />
+        <register type="NOA" address="0x00009888" value="0x03E75000" />
+        <register type="NOA" address="0x00009888" value="0x0003FF00" />
+        <register type="NOA" address="0x00009888" value="0x00075A00" />
+        <register type="NOA" address="0x00009888" value="0x000800FA" />
+        <register type="NOA" address="0x00009888" value="0x0208FFFF" />
+        <register type="NOA" address="0x00009888" value="0x000900F0" />
+        <register type="NOA" address="0x00009888" value="0x000AAA55" />
+        <register type="NOA" address="0x00009888" value="0x020A5555" />
+        <register type="NOA" address="0x00009888" value="0x0001FF0C" />
+        <register type="NOA" address="0x00009888" value="0x000B0050" />
+        <register type="NOA" address="0x00009888" value="0x020BAAAA" />
+        <register type="NOA" address="0x00009888" value="0x00020030" />
+        <register type="NOA" address="0x00009888" value="0x000C0500" />
+        <register type="NOA" address="0x00009888" value="0x03075A00" />
+        <register type="NOA" address="0x00009888" value="0x030800FA" />
+        <register type="NOA" address="0x00009888" value="0x0109F000" />
+        <register type="NOA" address="0x00009888" value="0x030AAA55" />
+        <register type="NOA" address="0x00009888" value="0x01010C00" />
+        <register type="NOA" address="0x00009888" value="0x030B0050" />
+        <register type="NOA" address="0x00009888" value="0x01023000" />
+        <register type="NOA" address="0x00009888" value="0x030C0500" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009888" value="0x00032492" />
+        <register type="NOA" address="0x00009888" value="0x02032492" />
+        <register type="NOA" address="0x00009888" value="0x04032492" />
+        <register type="NOA" address="0x00009888" value="0x06030002" />
+        <register type="NOA" address="0x00009888" value="0x14037000" />
+        <register type="NOA" address="0x00009888" value="0x3E030C6A" />
+        <register type="NOA" address="0x00009888" value="0x26037000" />
+        <register type="NOA" address="0x00009888" value="0x4203014A" />
+        <register type="NOA" address="0x00009888" value="0x2C037000" />
+        <register type="NOA" address="0x00009888" value="0x32037000" />
+        <register type="NOA" address="0x00009888" value="0x44030D00" />
+        <register type="NOA" address="0x00009888" value="0x38033000" />
+        <register type="NOA" address="0x00009888" value="0x46030EE8" />
+        <register type="NOA" address="0x00009888" value="0x48030277" />
+        <register type="NOA" address="0x00009888" value="0x4A030422" />
+        <register type="NOA" address="0x00009888" value="0x16030000" />
+        <register type="NOA" address="0x00009888" value="0x18030000" />
+        <register type="NOA" address="0x00009888" value="0x1A030000" />
+        <register type="NOA" address="0x00009888" value="0x40032800" />
+        <register type="NOA" address="0x00009888" value="0x1C030000" />
+        <register type="NOA" address="0x00009888" value="0x1E030000" />
+        <register type="NOA" address="0x00009888" value="0x20030000" />
+        <register type="NOA" address="0x00009884" value="0x00000008" />
+        <register type="NOA" address="0x00009888" value="0x0001FFFF" />
+        <register type="NOA" address="0x00009888" value="0x00025555" />
+        <register type="NOA" address="0x00009888" value="0x02025555" />
+        <register type="NOA" address="0x00009888" value="0x00045555" />
+        <register type="NOA" address="0x00009888" value="0x02045555" />
+        <register type="NOA" address="0x00009888" value="0x00055555" />
+        <register type="NOA" address="0x00009888" value="0x02055555" />
+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0xF0800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0xF0800000" />
+        <register type="OA" address="0x0000DC40" value="0x00FF0000" />
+        <register type="OA" address="0x0000D940" value="0x00000000" />
+        <register type="OA" address="0x0000D944" value="0x0000FFFC" />
+        <register type="OA" address="0x0000DC00" value="0x00000000" />
+        <register type="OA" address="0x0000DC04" value="0x0000FFFC" />
+        <register type="OA" address="0x0000D948" value="0x00000000" />
+        <register type="OA" address="0x0000D94C" value="0x0000FFF3" />
+        <register type="OA" address="0x0000DC08" value="0x00000000" />
+        <register type="OA" address="0x0000DC0C" value="0x0000FFF3" />
+        <register type="OA" address="0x0000D950" value="0x00000000" />
+        <register type="OA" address="0x0000D954" value="0x0000FFCF" />
+        <register type="OA" address="0x0000DC10" value="0x00000000" />
+        <register type="OA" address="0x0000DC14" value="0x0000FFCF" />
+        <register type="OA" address="0x0000D958" value="0x00000000" />
+        <register type="OA" address="0x0000D95C" value="0x0000FF3F" />
+        <register type="OA" address="0x0000DC18" value="0x00000000" />
+        <register type="OA" address="0x0000DC1C" value="0x0000FF3F" />
+        <register type="OA" address="0x0000D960" value="0x00000000" />
+        <register type="OA" address="0x0000D964" value="0x0000FCFF" />
+        <register type="OA" address="0x0000DC20" value="0x00000000" />
+        <register type="OA" address="0x0000DC24" value="0x0000FCFF" />
+        <register type="OA" address="0x0000D968" value="0x00000000" />
+        <register type="OA" address="0x0000D96C" value="0x0000F3FF" />
+        <register type="OA" address="0x0000DC28" value="0x00000000" />
+        <register type="OA" address="0x0000DC2C" value="0x0000F3FF" />
+        <register type="OA" address="0x0000D970" value="0x00000000" />
+        <register type="OA" address="0x0000D974" value="0x0000CFFF" />
+        <register type="OA" address="0x0000DC30" value="0x00000000" />
+        <register type="OA" address="0x0000DC34" value="0x0000CFFF" />
+        <register type="OA" address="0x0000D978" value="0x00000000" />
+        <register type="OA" address="0x0000D97C" value="0x00003FFF" />
+        <register type="OA" address="0x0000DC38" value="0x00000000" />
+        <register type="OA" address="0x0000DC3C" value="0x00003FFF" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="TDL_3"
+       chipset="DG2"
+       symbol_name="TDL_3"
+       underscore_name="tdl_3"
+       mdapi_supported_apis="VK OGL OCL MEDIA IO"
+       hw_config_guid="54a85b07-9ff2-4a12-bfb6-6bcdd9d4019c"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice0 Thread Dispatcher"
+             symbol_name="ThreadHeader00Ready"
+             underscore_name="thread_header00_ready"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice0 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 1 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice0 Thread Dispatcher Port 0"
+             symbol_name="ThreadHeader00ReadyPort0"
+             underscore_name="thread_header00_ready_port0"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice0 thread dispatcher port 0"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 1 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice0 Thread Dispatcher Port 1"
+             symbol_name="ThreadHeader00ReadyPort1"
+             underscore_name="thread_header00_ready_port1"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice0 thread dispatcher port 1"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 1 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice0 Thread Dispatcher Port 2"
+             symbol_name="ThreadHeader00ReadyPort2"
+             underscore_name="thread_header00_ready_port2"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice0 thread dispatcher port 2"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 1 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice0 Thread Dispatcher Port 3"
+             symbol_name="ThreadHeader00ReadyPort3"
+             underscore_name="thread_header00_ready_port3"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice0 thread dispatcher port 3"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 1 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice1 Thread Dispatcher"
+             symbol_name="ThreadHeader01Ready"
+             underscore_name="thread_header01_ready"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice1 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 2 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice1 Thread Dispatcher Port 0"
+             symbol_name="ThreadHeader01ReadyPort0"
+             underscore_name="thread_header01_ready_port0"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice1 thread dispatcher port 0"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 2 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice1 Thread Dispatcher Port 1"
+             symbol_name="ThreadHeader01ReadyPort1"
+             underscore_name="thread_header01_ready_port1"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice1 thread dispatcher port 1"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 2 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice1 Thread Dispatcher Port 2"
+             symbol_name="ThreadHeader01ReadyPort2"
+             underscore_name="thread_header01_ready_port2"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice1 thread dispatcher port 2"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 2 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice1 Thread Dispatcher Port 3"
+             symbol_name="ThreadHeader01ReadyPort3"
+             underscore_name="thread_header01_ready_port3"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice1 thread dispatcher port 3"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 2 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice2 Thread Dispatcher"
+             symbol_name="ThreadHeader02Ready"
+             underscore_name="thread_header02_ready"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice2 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 4 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice2 Thread Dispatcher Port 0"
+             symbol_name="ThreadHeader02ReadyPort0"
+             underscore_name="thread_header02_ready_port0"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice2 thread dispatcher port 0"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 4 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice2 Thread Dispatcher Port 1"
+             symbol_name="ThreadHeader02ReadyPort1"
+             underscore_name="thread_header02_ready_port1"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice2 thread dispatcher port 1"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 4 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice2 Thread Dispatcher Port 2"
+             symbol_name="ThreadHeader02ReadyPort2"
+             underscore_name="thread_header02_ready_port2"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice2 thread dispatcher port 2"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 4 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice2 Thread Dispatcher Port 3"
+             symbol_name="ThreadHeader02ReadyPort3"
+             underscore_name="thread_header02_ready_port3"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice2 thread dispatcher port 3"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 4 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice3 Thread Dispatcher"
+             symbol_name="ThreadHeader03Ready"
+             underscore_name="thread_header03_ready"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice3 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 8 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x0000000C" />
+        <register type="NOA" address="0x00009888" value="0x2A81001C" />
+        <register type="NOA" address="0x00009888" value="0x2AA1001C" />
+        <register type="NOA" address="0x00009888" value="0x2AC1001C" />
+        <register type="NOA" address="0x00009888" value="0x2AE1001C" />
+        <register type="NOA" address="0x00009888" value="0x06810300" />
+        <register type="NOA" address="0x00009888" value="0x14810CF4" />
+        <register type="NOA" address="0x00009888" value="0x16810C72" />
+        <register type="NOA" address="0x00009888" value="0x08810000" />
+        <register type="NOA" address="0x00009888" value="0x02810000" />
+        <register type="NOA" address="0x00009888" value="0x0283AA00" />
+        <register type="NOA" address="0x00009888" value="0x0084F000" />
+        <register type="NOA" address="0x00009888" value="0x02865500" />
+        <register type="NOA" address="0x00009888" value="0x02875500" />
+        <register type="NOA" address="0x00009888" value="0x06A10300" />
+        <register type="NOA" address="0x00009888" value="0x10A10CF4" />
+        <register type="NOA" address="0x00009888" value="0x12A10C72" />
+        <register type="NOA" address="0x00009888" value="0x08A10000" />
+        <register type="NOA" address="0x00009888" value="0x02A10000" />
+        <register type="NOA" address="0x00009888" value="0x02A300AA" />
+        <register type="NOA" address="0x00009888" value="0x00A40F00" />
+        <register type="NOA" address="0x00009888" value="0x02A60055" />
+        <register type="NOA" address="0x00009888" value="0x02A70055" />
+        <register type="NOA" address="0x00009888" value="0x06C10300" />
+        <register type="NOA" address="0x00009888" value="0x0CC10CF4" />
+        <register type="NOA" address="0x00009888" value="0x0EC10C72" />
+        <register type="NOA" address="0x00009888" value="0x08C10000" />
+        <register type="NOA" address="0x00009888" value="0x00C10000" />
+        <register type="NOA" address="0x00009888" value="0x00C3AA00" />
+        <register type="NOA" address="0x00009888" value="0x00C400F0" />
+        <register type="NOA" address="0x00009888" value="0x00C65500" />
+        <register type="NOA" address="0x00009888" value="0x00C75500" />
+        <register type="NOA" address="0x00009888" value="0x06E10300" />
+        <register type="NOA" address="0x00009888" value="0x08E10CB1" />
+        <register type="NOA" address="0x00009888" value="0x0AE10D33" />
+        <register type="NOA" address="0x00009888" value="0x00E10000" />
+        <register type="NOA" address="0x00009888" value="0x00E300AA" />
+        <register type="NOA" address="0x00009888" value="0x00E4000F" />
+        <register type="NOA" address="0x00009888" value="0x00E60055" />
+        <register type="NOA" address="0x00009888" value="0x00E70055" />
+        <register type="NOA" address="0x00009888" value="0x0007AA55" />
+        <register type="NOA" address="0x00009888" value="0x0208AAFF" />
+        <register type="NOA" address="0x00009888" value="0x000900FF" />
+        <register type="NOA" address="0x00009888" value="0x000AAAAA" />
+        <register type="NOA" address="0x00009888" value="0x020A5555" />
+        <register type="NOA" address="0x00009888" value="0x00010F00" />
+        <register type="NOA" address="0x00009888" value="0x020B0055" />
+        <register type="NOA" address="0x00009888" value="0x000200F0" />
+        <register type="NOA" address="0x00009888" value="0x000C5500" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009888" value="0x00032492" />
+        <register type="NOA" address="0x00009888" value="0x02032492" />
+        <register type="NOA" address="0x00009888" value="0x04032492" />
+        <register type="NOA" address="0x00009888" value="0x06030002" />
+        <register type="NOA" address="0x00009888" value="0x14030000" />
+        <register type="NOA" address="0x00009888" value="0x3E030C6A" />
+        <register type="NOA" address="0x00009888" value="0x26030000" />
+        <register type="NOA" address="0x00009888" value="0x42032942" />
+        <register type="NOA" address="0x00009888" value="0x2C030000" />
+        <register type="NOA" address="0x00009888" value="0x32030000" />
+        <register type="NOA" address="0x00009888" value="0x4403000A" />
+        <register type="NOA" address="0x00009888" value="0x38030000" />
+        <register type="NOA" address="0x00009888" value="0x46030000" />
+        <register type="NOA" address="0x00009888" value="0x48030000" />
+        <register type="NOA" address="0x00009888" value="0x4A030603" />
+        <register type="NOA" address="0x00009888" value="0x16030000" />
+        <register type="NOA" address="0x00009888" value="0x18030000" />
+        <register type="NOA" address="0x00009888" value="0x1A031000" />
+        <register type="NOA" address="0x00009888" value="0x40030842" />
+        <register type="NOA" address="0x00009888" value="0x1C031000" />
+        <register type="NOA" address="0x00009888" value="0x1E031000" />
+        <register type="NOA" address="0x00009888" value="0x20031000" />
+        <register type="NOA" address="0x00009884" value="0x00000008" />
+        <register type="NOA" address="0x00009888" value="0x0001FFFF" />
+        <register type="NOA" address="0x00009888" value="0x00025555" />
+        <register type="NOA" address="0x00009888" value="0x02025555" />
+        <register type="NOA" address="0x00009888" value="0x00045555" />
+        <register type="NOA" address="0x00009888" value="0x02045555" />
+        <register type="NOA" address="0x00009888" value="0x00055555" />
+        <register type="NOA" address="0x00009888" value="0x02055555" />
+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0xF0800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x000F0000" />
+        <register type="OA" address="0x0000D940" value="0x00078000" />
+        <register type="OA" address="0x0000D944" value="0x00000FFF" />
+        <register type="OA" address="0x0000DC00" value="0x00078000" />
+        <register type="OA" address="0x0000DC04" value="0x00000FFF" />
+        <register type="OA" address="0x0000D948" value="0x00007800" />
+        <register type="OA" address="0x0000D94C" value="0x0000F0FF" />
+        <register type="OA" address="0x0000DC08" value="0x00007800" />
+        <register type="OA" address="0x0000DC0C" value="0x0000F0FF" />
+        <register type="OA" address="0x0000D950" value="0x00000780" />
+        <register type="OA" address="0x0000D954" value="0x0000FF0F" />
+        <register type="OA" address="0x0000DC10" value="0x00000780" />
+        <register type="OA" address="0x0000DC14" value="0x0000FF0F" />
+        <register type="OA" address="0x0000D958" value="0x00000078" />
+        <register type="OA" address="0x0000D95C" value="0x0000FFF0" />
+        <register type="OA" address="0x0000DC18" value="0x00000078" />
+        <register type="OA" address="0x0000DC1C" value="0x0000FFF0" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="TDL_4"
+       chipset="DG2"
+       symbol_name="TDL_4"
+       underscore_name="tdl_4"
+       mdapi_supported_apis="VK OGL OCL MEDIA IO"
+       hw_config_guid="651c40e5-195c-49da-b232-a9a0f753d899"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice0 Thread Dispatcher"
+             symbol_name="ThreadHeader10Ready"
+             underscore_name="thread_header10_ready"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice0 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 64 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice0 Thread Dispatcher Port 0"
+             symbol_name="ThreadHeader10ReadyPort0"
+             underscore_name="thread_header10_ready_port0"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice0 thread dispatcher port 0"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 64 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice0 Thread Dispatcher Port 1"
+             symbol_name="ThreadHeader10ReadyPort1"
+             underscore_name="thread_header10_ready_port1"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice0 thread dispatcher port 1"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 64 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice0 Thread Dispatcher Port 2"
+             symbol_name="ThreadHeader10ReadyPort2"
+             underscore_name="thread_header10_ready_port2"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice0 thread dispatcher port 2"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 64 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice0 Thread Dispatcher Port 3"
+             symbol_name="ThreadHeader10ReadyPort3"
+             underscore_name="thread_header10_ready_port3"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice0 thread dispatcher port 3"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 64 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice1 Thread Dispatcher"
+             symbol_name="ThreadHeader11Ready"
+             underscore_name="thread_header11_ready"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice1 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 128 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice1 Thread Dispatcher Port 0"
+             symbol_name="ThreadHeader11ReadyPort0"
+             underscore_name="thread_header11_ready_port0"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice1 thread dispatcher port 0"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 128 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice1 Thread Dispatcher Port 1"
+             symbol_name="ThreadHeader11ReadyPort1"
+             underscore_name="thread_header11_ready_port1"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice1 thread dispatcher port 1"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 128 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice1 Thread Dispatcher Port 2"
+             symbol_name="ThreadHeader11ReadyPort2"
+             underscore_name="thread_header11_ready_port2"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice1 thread dispatcher port 2"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 128 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice1 Thread Dispatcher Port 3"
+             symbol_name="ThreadHeader11ReadyPort3"
+             underscore_name="thread_header11_ready_port3"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice1 thread dispatcher port 3"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 128 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice2 Thread Dispatcher"
+             symbol_name="ThreadHeader12Ready"
+             underscore_name="thread_header12_ready"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice2 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 256 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice2 Thread Dispatcher Port 0"
+             symbol_name="ThreadHeader12ReadyPort0"
+             underscore_name="thread_header12_ready_port0"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice2 thread dispatcher port 0"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 256 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice2 Thread Dispatcher Port 1"
+             symbol_name="ThreadHeader12ReadyPort1"
+             underscore_name="thread_header12_ready_port1"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice2 thread dispatcher port 1"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 256 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice2 Thread Dispatcher Port 2"
+             symbol_name="ThreadHeader12ReadyPort2"
+             underscore_name="thread_header12_ready_port2"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice2 thread dispatcher port 2"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 256 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice2 Thread Dispatcher Port 3"
+             symbol_name="ThreadHeader12ReadyPort3"
+             underscore_name="thread_header12_ready_port3"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice2 thread dispatcher port 3"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 256 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice3 Thread Dispatcher"
+             symbol_name="ThreadHeader13Ready"
+             underscore_name="thread_header13_ready"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice3 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 512 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x0000000C" />
+        <register type="NOA" address="0x00009888" value="0x2B81001C" />
+        <register type="NOA" address="0x00009888" value="0x2BA1001C" />
+        <register type="NOA" address="0x00009888" value="0x2BC1001C" />
+        <register type="NOA" address="0x00009888" value="0x2BE1001C" />
+        <register type="NOA" address="0x00009888" value="0x07810300" />
+        <register type="NOA" address="0x00009888" value="0x15810CF4" />
+        <register type="NOA" address="0x00009888" value="0x17810C72" />
+        <register type="NOA" address="0x00009888" value="0x09810000" />
+        <register type="NOA" address="0x00009888" value="0x03810000" />
+        <register type="NOA" address="0x00009888" value="0x0383AA00" />
+        <register type="NOA" address="0x00009888" value="0x0184F000" />
+        <register type="NOA" address="0x00009888" value="0x03865500" />
+        <register type="NOA" address="0x00009888" value="0x03875500" />
+        <register type="NOA" address="0x00009888" value="0x07A10300" />
+        <register type="NOA" address="0x00009888" value="0x11A10CF4" />
+        <register type="NOA" address="0x00009888" value="0x13A10C72" />
+        <register type="NOA" address="0x00009888" value="0x09A10000" />
+        <register type="NOA" address="0x00009888" value="0x03A10000" />
+        <register type="NOA" address="0x00009888" value="0x03A300AA" />
+        <register type="NOA" address="0x00009888" value="0x01A40F00" />
+        <register type="NOA" address="0x00009888" value="0x03A60055" />
+        <register type="NOA" address="0x00009888" value="0x03A70055" />
+        <register type="NOA" address="0x00009888" value="0x07C10300" />
+        <register type="NOA" address="0x00009888" value="0x0DC10CF4" />
+        <register type="NOA" address="0x00009888" value="0x0FC10C72" />
+        <register type="NOA" address="0x00009888" value="0x09C10000" />
+        <register type="NOA" address="0x00009888" value="0x01C10000" />
+        <register type="NOA" address="0x00009888" value="0x01C3AA00" />
+        <register type="NOA" address="0x00009888" value="0x01C400F0" />
+        <register type="NOA" address="0x00009888" value="0x01C65500" />
+        <register type="NOA" address="0x00009888" value="0x01C75500" />
+        <register type="NOA" address="0x00009888" value="0x07E10300" />
+        <register type="NOA" address="0x00009888" value="0x09E10CB1" />
+        <register type="NOA" address="0x00009888" value="0x0BE10D33" />
+        <register type="NOA" address="0x00009888" value="0x01E10000" />
+        <register type="NOA" address="0x00009888" value="0x01E300AA" />
+        <register type="NOA" address="0x00009888" value="0x01E4000F" />
+        <register type="NOA" address="0x00009888" value="0x01E60055" />
+        <register type="NOA" address="0x00009888" value="0x01E70055" />
+        <register type="NOA" address="0x00009888" value="0x0003FFFF" />
+        <register type="NOA" address="0x00009888" value="0x0008FFFF" />
+        <register type="NOA" address="0x00009888" value="0x0208FFFF" />
+        <register type="NOA" address="0x00009888" value="0x000A5555" />
+        <register type="NOA" address="0x00009888" value="0x020A5555" />
+        <register type="NOA" address="0x00009888" value="0x0001FFFF" />
+        <register type="NOA" address="0x00009888" value="0x000BAAAA" />
+        <register type="NOA" address="0x00009888" value="0x020BAAAA" />
+        <register type="NOA" address="0x00009888" value="0x0107AA55" />
+        <register type="NOA" address="0x00009888" value="0x0308AAFF" />
+        <register type="NOA" address="0x00009888" value="0x010900FF" />
+        <register type="NOA" address="0x00009888" value="0x010AAAAA" />
+        <register type="NOA" address="0x00009888" value="0x030A5555" />
+        <register type="NOA" address="0x00009888" value="0x01010F00" />
+        <register type="NOA" address="0x00009888" value="0x030B0055" />
+        <register type="NOA" address="0x00009888" value="0x010200F0" />
+        <register type="NOA" address="0x00009888" value="0x010C5500" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009888" value="0x00032492" />
+        <register type="NOA" address="0x00009888" value="0x02032492" />
+        <register type="NOA" address="0x00009888" value="0x04032492" />
+        <register type="NOA" address="0x00009888" value="0x06030002" />
+        <register type="NOA" address="0x00009888" value="0x14030000" />
+        <register type="NOA" address="0x00009888" value="0x3E030C6A" />
+        <register type="NOA" address="0x00009888" value="0x26030000" />
+        <register type="NOA" address="0x00009888" value="0x42032942" />
+        <register type="NOA" address="0x00009888" value="0x2C030000" />
+        <register type="NOA" address="0x00009888" value="0x32030000" />
+        <register type="NOA" address="0x00009888" value="0x4403000A" />
+        <register type="NOA" address="0x00009888" value="0x38030000" />
+        <register type="NOA" address="0x00009888" value="0x46030000" />
+        <register type="NOA" address="0x00009888" value="0x48030000" />
+        <register type="NOA" address="0x00009888" value="0x4A030603" />
+        <register type="NOA" address="0x00009888" value="0x16030000" />
+        <register type="NOA" address="0x00009888" value="0x18030000" />
+        <register type="NOA" address="0x00009888" value="0x1A031000" />
+        <register type="NOA" address="0x00009888" value="0x40030842" />
+        <register type="NOA" address="0x00009888" value="0x1C031000" />
+        <register type="NOA" address="0x00009888" value="0x1E031000" />
+        <register type="NOA" address="0x00009888" value="0x20031000" />
+        <register type="NOA" address="0x00009884" value="0x00000008" />
+        <register type="NOA" address="0x00009888" value="0x0001FFFF" />
+        <register type="NOA" address="0x00009888" value="0x00025555" />
+        <register type="NOA" address="0x00009888" value="0x02025555" />
+        <register type="NOA" address="0x00009888" value="0x00045555" />
+        <register type="NOA" address="0x00009888" value="0x02045555" />
+        <register type="NOA" address="0x00009888" value="0x00055555" />
+        <register type="NOA" address="0x00009888" value="0x02055555" />
+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0xF0800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x000F0000" />
+        <register type="OA" address="0x0000D940" value="0x00078000" />
+        <register type="OA" address="0x0000D944" value="0x00000FFF" />
+        <register type="OA" address="0x0000DC00" value="0x00078000" />
+        <register type="OA" address="0x0000DC04" value="0x00000FFF" />
+        <register type="OA" address="0x0000D948" value="0x00007800" />
+        <register type="OA" address="0x0000D94C" value="0x0000F0FF" />
+        <register type="OA" address="0x0000DC08" value="0x00007800" />
+        <register type="OA" address="0x0000DC0C" value="0x0000F0FF" />
+        <register type="OA" address="0x0000D950" value="0x00000780" />
+        <register type="OA" address="0x0000D954" value="0x0000FF0F" />
+        <register type="OA" address="0x0000DC10" value="0x00000780" />
+        <register type="OA" address="0x0000DC14" value="0x0000FF0F" />
+        <register type="OA" address="0x0000D958" value="0x00000078" />
+        <register type="OA" address="0x0000D95C" value="0x0000FFF0" />
+        <register type="OA" address="0x0000DC18" value="0x00000078" />
+        <register type="OA" address="0x0000DC1C" value="0x0000FFF0" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="L3_1"
+       chipset="DG2"
+       symbol_name="L3_1"
+       underscore_name="l3_1"
+       mdapi_supported_apis="VK OGL OCL MEDIA IO"
+       hw_config_guid="9a1a14ad-ab92-4204-8a91-c4159a54f712"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Quad0 L3 Bank0 Input Available"
+             symbol_name="L30Bank0InputAvailable"
+             underscore_name="l30_bank0_input_available"
+             description="The percentage of time in which Quad0 L3 bank0 has input available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank1 Input Available"
+             symbol_name="L30Bank1InputAvailable"
+             underscore_name="l30_bank1_input_available"
+             description="The percentage of time in which Quad0 L3 bank1 has input available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank2 Input Available"
+             symbol_name="L30Bank2InputAvailable"
+             underscore_name="l30_bank2_input_available"
+             description="The percentage of time in which Quad0 L3 bank2 has input available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank3 Input Available"
+             symbol_name="L30Bank3InputAvailable"
+             underscore_name="l30_bank3_input_available"
+             description="The percentage of time in which Quad0 L3 bank3 has input available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank4 Input Available"
+             symbol_name="L30Bank4InputAvailable"
+             underscore_name="l30_bank4_input_available"
+             description="The percentage of time in which Quad0 L3 bank4 has input available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank5 Input Available"
+             symbol_name="L30Bank5InputAvailable"
+             underscore_name="l30_bank5_input_available"
+             description="The percentage of time in which Quad0 L3 bank5 has input available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank6 Input Available"
+             symbol_name="L30Bank6InputAvailable"
+             underscore_name="l30_bank6_input_available"
+             description="The percentage of time in which Quad0 L3 bank6 has input available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank7 Input Available"
+             symbol_name="L30Bank7InputAvailable"
+             underscore_name="l30_bank7_input_available"
+             description="The percentage of time in which Quad0 L3 bank7 has input available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank0 Output Ready"
+             symbol_name="L30Bank0OutputReady"
+             underscore_name="l30_bank0_output_ready"
+             description="The percentage of time in which quad0 L3 bank0 output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank1 Output Ready"
+             symbol_name="L30Bank1OutputReady"
+             underscore_name="l30_bank1_output_ready"
+             description="The percentage of time in which quad0 L3 bank1 output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank2 Output Ready"
+             symbol_name="L30Bank2OutputReady"
+             underscore_name="l30_bank2_output_ready"
+             description="The percentage of time in which quad0 L3 bank2 output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank3 Output Ready"
+             symbol_name="L30Bank3OutputReady"
+             underscore_name="l30_bank3_output_ready"
+             description="The percentage of time in which quad0 L3 bank3 output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank4 Output Ready"
+             symbol_name="L30Bank4OutputReady"
+             underscore_name="l30_bank4_output_ready"
+             description="The percentage of time in which quad0 L3 bank4 output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank5 Output Ready"
+             symbol_name="L30Bank5OutputReady"
+             underscore_name="l30_bank5_output_ready"
+             description="The percentage of time in which quad0 L3 bank5 output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank6 Output Ready"
+             symbol_name="L30Bank6OutputReady"
+             underscore_name="l30_bank6_output_ready"
+             description="The percentage of time in which quad0 L3 bank6 output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank7 Output Ready"
+             symbol_name="L30Bank7OutputReady"
+             underscore_name="l30_bank7_output_ready"
+             description="The percentage of time in which quad0 L3 bank7 output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x00000008" />
+        <register type="NOA" address="0x00009888" value="0x090B1000" />
+        <register type="NOA" address="0x00009888" value="0x092B1006" />
+        <register type="NOA" address="0x00009888" value="0x094B1000" />
+        <register type="NOA" address="0x00009888" value="0x096B1000" />
+        <register type="NOA" address="0x00009888" value="0x090C1000" />
+        <register type="NOA" address="0x00009888" value="0x092C1180" />
+        <register type="NOA" address="0x00009888" value="0x094C1000" />
+        <register type="NOA" address="0x00009888" value="0x096C1000" />
+        <register type="NOA" address="0x00009888" value="0x010B0002" />
+        <register type="NOA" address="0x00009888" value="0x0B0B0006" />
+        <register type="NOA" address="0x00009888" value="0x130B0007" />
+        <register type="NOA" address="0x00009888" value="0x030B0000" />
+        <register type="NOA" address="0x00009888" value="0x012B0002" />
+        <register type="NOA" address="0x00009888" value="0x112B0007" />
+        <register type="NOA" address="0x00009888" value="0x032B0000" />
+        <register type="NOA" address="0x00009888" value="0x014B0002" />
+        <register type="NOA" address="0x00009888" value="0x074B0006" />
+        <register type="NOA" address="0x00009888" value="0x0F4B0007" />
+        <register type="NOA" address="0x00009888" value="0x034B0000" />
+        <register type="NOA" address="0x00009888" value="0x016B0002" />
+        <register type="NOA" address="0x00009888" value="0x056B0006" />
+        <register type="NOA" address="0x00009888" value="0x0D6B0007" />
+        <register type="NOA" address="0x00009888" value="0x036B0000" />
+        <register type="NOA" address="0x00009888" value="0x010C0002" />
+        <register type="NOA" address="0x00009888" value="0x0B0C0180" />
+        <register type="NOA" address="0x00009888" value="0x130C01C0" />
+        <register type="NOA" address="0x00009888" value="0x030C0000" />
+        <register type="NOA" address="0x00009888" value="0x012C0002" />
+        <register type="NOA" address="0x00009888" value="0x112C01C0" />
+        <register type="NOA" address="0x00009888" value="0x032C0000" />
+        <register type="NOA" address="0x00009888" value="0x014C0002" />
+        <register type="NOA" address="0x00009888" value="0x074C0180" />
+        <register type="NOA" address="0x00009888" value="0x0F4C01C0" />
+        <register type="NOA" address="0x00009888" value="0x034C0000" />
+        <register type="NOA" address="0x00009888" value="0x016C0002" />
+        <register type="NOA" address="0x00009888" value="0x056C0180" />
+        <register type="NOA" address="0x00009888" value="0x0D6C01C0" />
+        <register type="NOA" address="0x00009888" value="0x036C0000" />
+        <register type="NOA" address="0x00009888" value="0x00020030" />
+        <register type="NOA" address="0x00009888" value="0x02020030" />
+        <register type="NOA" address="0x00009888" value="0x00030048" />
+        <register type="NOA" address="0x00009888" value="0x02030048" />
+        <register type="NOA" address="0x00009888" value="0x0004009B" />
+        <register type="NOA" address="0x00009888" value="0x0204009B" />
+        <register type="NOA" address="0x00009888" value="0x00052055" />
+        <register type="NOA" address="0x00009888" value="0x02052055" />
+        <register type="NOA" address="0x00009888" value="0x00064800" />
+        <register type="NOA" address="0x00009888" value="0x02064800" />
+        <register type="NOA" address="0x00009888" value="0x0007A0A0" />
+        <register type="NOA" address="0x00009888" value="0x000867AA" />
+        <register type="NOA" address="0x00009888" value="0x020867AA" />
+        <register type="NOA" address="0x00009888" value="0x0009A0A0" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009888" value="0x00032492" />
+        <register type="NOA" address="0x00009888" value="0x02032492" />
+        <register type="NOA" address="0x00009888" value="0x04032492" />
+        <register type="NOA" address="0x00009888" value="0x06030002" />
+        <register type="NOA" address="0x00009888" value="0x14030000" />
+        <register type="NOA" address="0x00009888" value="0x3E030024" />
+        <register type="NOA" address="0x00009888" value="0x26030000" />
+        <register type="NOA" address="0x00009888" value="0x42030401" />
+        <register type="NOA" address="0x00009888" value="0x2C030000" />
+        <register type="NOA" address="0x00009888" value="0x32030000" />
+        <register type="NOA" address="0x00009888" value="0x44030140" />
+        <register type="NOA" address="0x00009888" value="0x38030000" />
+        <register type="NOA" address="0x00009888" value="0x46030E01" />
+        <register type="NOA" address="0x00009888" value="0x48030010" />
+        <register type="NOA" address="0x00009888" value="0x4A030004" />
+        <register type="NOA" address="0x00009888" value="0x16030000" />
+        <register type="NOA" address="0x00009888" value="0x18030000" />
+        <register type="NOA" address="0x00009888" value="0x1A030000" />
+        <register type="NOA" address="0x00009888" value="0x40031C2A" />
+        <register type="NOA" address="0x00009888" value="0x1C030000" />
+        <register type="NOA" address="0x00009888" value="0x1E030000" />
+        <register type="NOA" address="0x00009888" value="0x20030000" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x00800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00000000" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="GpuBusyness"
+       chipset="DG2"
+       symbol_name="GpuBusyness"
+       underscore_name="gpu_busyness"
+       mdapi_supported_apis="VK OGL OCL MEDIA IO"
+       hw_config_guid="85de4c29-76b8-42a6-b6ef-8a2bed008fcf"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Render Ring Busy"
+             symbol_name="RenderBusy"
+             underscore_name="render_busy"
+             description="The percentage of time when render command streamer was busy."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="System Frame Batch"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="ComputeEngine0 Ring Busy"
+             symbol_name="ComputeEngine0Busy"
+             underscore_name="compute_engine0_busy"
+             description="The percentage of time when compute command streamer 0 was busy."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="System Frame Batch"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="ComputeEngine1 Ring Busy"
+             symbol_name="ComputeEngine1Busy"
+             underscore_name="compute_engine1_busy"
+             description="The percentage of time when compute command streamer 1 was busy."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="System Frame Batch"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="ComputeEngine2 Ring Busy"
+             symbol_name="ComputeEngine2Busy"
+             underscore_name="compute_engine2_busy"
+             description="The percentage of time when compute command streamer 2 was busy."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="System Frame Batch"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="ComputeEngine3 Ring Busy"
+             symbol_name="ComputeEngine3Busy"
+             underscore_name="compute_engine3_busy"
+             description="The percentage of time when compute command streamer 3 was busy."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="System Frame Batch"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Posh Ring Busy"
+             symbol_name="PoshEngineBusy"
+             underscore_name="posh_engine_busy"
+             description="The percentage of time when posh command streamer was busy."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="System Frame Batch"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Blitter Ring Busy"
+             symbol_name="BlitterBusy"
+             underscore_name="blitter_busy"
+             description="The percentage of time when blitter command streamer was busy."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="System Frame Batch"
+             mdapi_supported_apis="IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Vdbox00 Ring Busy"
+             symbol_name="Vdbox00Busy"
+             underscore_name="vdbox00_busy"
+             description="The percentage of time when vdbox00 command streamer was busy."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="System Frame Batch"
+             mdapi_supported_apis="IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Vebox0 Ring Busy"
+             symbol_name="Vebox0Busy"
+             underscore_name="vebox0_busy"
+             description="The percentage of time when vebox0 command streamer was busy."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="System Frame Batch"
+             mdapi_supported_apis="IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Vdbox10 Ring Busy"
+             symbol_name="Vdbox10Busy"
+             underscore_name="vdbox10_busy"
+             description="The percentage of time when vdbox10 command streamer was busy."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="System Frame Batch"
+             mdapi_supported_apis="IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Vebox1 Ring Busy"
+             symbol_name="Vebox1Busy"
+             underscore_name="vebox1_busy"
+             description="The percentage of time when vebox1 command streamer was busy."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="System Frame Batch"
+             mdapi_supported_apis="IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Any compute engine busy"
+             symbol_name="AnyComputeEngineBusy"
+             underscore_name="any_compute_engine_busy"
+             description="The percentage of time when any compute engine is busy"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Render and any compute engines are simultaneously busy"
+             symbol_name="RenderAndAnyComputeEngineBusy"
+             underscore_name="render_and_any_compute_engine_busy"
+             description="The percentage of time when render and any compute engines are simultaneously busy"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x00000004" />
+        <register type="NOA" address="0x00009888" value="0x420A2400" />
+        <register type="NOA" address="0x00009888" value="0x1A131200" />
+        <register type="NOA" address="0x00009888" value="0x040A0022" />
+        <register type="NOA" address="0x00009888" value="0x120A0014" />
+        <register type="NOA" address="0x00009888" value="0x020A0000" />
+        <register type="NOA" address="0x00009888" value="0x000A0000" />
+        <register type="NOA" address="0x00009888" value="0x00104000" />
+        <register type="NOA" address="0x00009888" value="0x02100002" />
+        <register type="NOA" address="0x00009888" value="0x00124000" />
+        <register type="NOA" address="0x00009888" value="0x00134000" />
+        <register type="NOA" address="0x00009888" value="0x0A130024" />
+        <register type="NOA" address="0x00009884" value="0x00000005" />
+        <register type="NOA" address="0x00009888" value="0x420A2400" />
+        <register type="NOA" address="0x00009888" value="0x1A131200" />
+        <register type="NOA" address="0x00009888" value="0x040A0022" />
+        <register type="NOA" address="0x00009888" value="0x100A0014" />
+        <register type="NOA" address="0x00009888" value="0x020A0000" />
+        <register type="NOA" address="0x00009888" value="0x000A0000" />
+        <register type="NOA" address="0x00009888" value="0x00102400" />
+        <register type="NOA" address="0x00009888" value="0x00120400" />
+        <register type="NOA" address="0x00009888" value="0x00134000" />
+        <register type="NOA" address="0x00009888" value="0x08130024" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x18012400" />
+        <register type="NOA" address="0x00009888" value="0x18042400" />
+        <register type="NOA" address="0x00009888" value="0x20042400" />
+        <register type="NOA" address="0x00009888" value="0x28042400" />
+        <register type="NOA" address="0x00009888" value="0x32040009" />
+        <register type="NOA" address="0x00009888" value="0x3C041200" />
+        <register type="NOA" address="0x00009888" value="0x4C041200" />
+        <register type="NOA" address="0x00009888" value="0x02010006" />
+        <register type="NOA" address="0x00009888" value="0x10010004" />
+        <register type="NOA" address="0x00009888" value="0x0001FC00" />
+        <register type="NOA" address="0x00009888" value="0x00042EE9" />
+        <register type="NOA" address="0x00009888" value="0x02040239" />
+        <register type="NOA" address="0x00009888" value="0x0404000F" />
+        <register type="NOA" address="0x00009888" value="0x0E040D00" />
+        <register type="NOA" address="0x00009888" value="0x10040304" />
+        <register type="NOA" address="0x00009888" value="0x12040714" />
+        <register type="NOA" address="0x00009888" value="0x14040024" />
+        <register type="NOA" address="0x00009888" value="0x02085554" />
+        <register type="NOA" address="0x00009888" value="0x00056800" />
+        <register type="NOA" address="0x00009888" value="0x02050001" />
+        <register type="NOA" address="0x00009888" value="0x02032B6D" />
+        <register type="NOA" address="0x00009888" value="0x04032492" />
+        <register type="NOA" address="0x00009888" value="0x06030002" />
+        <register type="NOA" address="0x00009888" value="0x44030000" />
+        <register type="NOA" address="0x00009888" value="0x46030400" />
+        <register type="NOA" address="0x00009888" value="0x48030000" />
+        <register type="NOA" address="0x00009888" value="0x4A030C62" />
+        <register type="NOA" address="0x00009888" value="0x16033000" />
+        <register type="NOA" address="0x00009888" value="0x3E031080" />
+        <register type="NOA" address="0x00009888" value="0x18033000" />
+        <register type="NOA" address="0x00009888" value="0x1A033000" />
+        <register type="NOA" address="0x00009888" value="0x40031084" />
+        <register type="NOA" address="0x00009888" value="0x1C033000" />
+        <register type="NOA" address="0x00009888" value="0x1E033000" />
+        <register type="NOA" address="0x00009888" value="0x20033000" />
+        <register type="NOA" address="0x00009888" value="0x42030004" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x30800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00030000" />
+        <register type="OA" address="0x0000D940" value="0x00000000" />
+        <register type="OA" address="0x0000D944" value="0x000087FF" />
+        <register type="OA" address="0x0000DC00" value="0x00000000" />
+        <register type="OA" address="0x0000DC04" value="0x000087FF" />
+        <register type="OA" address="0x0000D948" value="0x00040001" />
+        <register type="OA" address="0x0000D94C" value="0x000007FF" />
+        <register type="OA" address="0x0000DC08" value="0x00040001" />
+        <register type="OA" address="0x0000DC0C" value="0x000007FF" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="EuActivity1"
+       chipset="DG2"
+       symbol_name="EuActivity1"
+       underscore_name="eu_activity1"
+       mdapi_supported_apis="OGL OGL4 OCL MEDIA IO"
+       hw_config_guid="8dc7142a-df10-41f2-acaa-9faea547ce73"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS FPU Pipe Active"
+             symbol_name="VsFpuActive"
+             underscore_name="vs_fpu_active"
+             description="The percentage of time in which EU FPU INT64 pipeline was actively processing a vertex shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS FPU Pipe Active"
+             symbol_name="PsFpuActive"
+             underscore_name="ps_fpu_active"
+             description="The percentage of time in which EU FPU INT64 pipeline was actively processing a fragment shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Send Pipe Active"
+             symbol_name="SendActive"
+             underscore_name="send_active"
+             description="The percentage of time in which EU send pipeline was actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Pipes"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x00800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00000000" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00810710" />
+        <register type="FLEX" address="0x0000E558" value="0x00A10910" />
+        <register type="FLEX" address="0x0000E658" value="0x00850750" />
+        <register type="FLEX" address="0x0000E758" value="0x00A50950" />
+        <register type="FLEX" address="0x0000E45C" value="0x00802702" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A02902" />
+    </register_config>
+  </set>
+
+  <set name="EuActivity2"
+       chipset="DG2"
+       symbol_name="EuActivity2"
+       underscore_name="eu_activity2"
+       mdapi_supported_apis="OGL OGL4 OCL MEDIA IO"
+       hw_config_guid="d3c05cce-4ae7-4291-82ab-b0bcade6cf7a"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS EM Pipe Active"
+             symbol_name="CsEmActive"
+             underscore_name="cs_em_active"
+             description="The percentage of time in which EU EM INT pipeline was actively processing a compute shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS FPU Pipe Active"
+             symbol_name="CsFpuActive"
+             underscore_name="cs_fpu_active"
+             description="The percentage of time in which EU FPU INT64 pipeline was actively processing a compute shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Send Pipe Active"
+             symbol_name="CsSendActive"
+             underscore_name="cs_send_active"
+             description="The percentage of time in which EU send pipeline was actively processing a compute shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x00800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00000000" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00861761" />
+        <register type="FLEX" address="0x0000E558" value="0x00A61961" />
+        <register type="FLEX" address="0x0000E658" value="0x00860760" />
+        <register type="FLEX" address="0x0000E758" value="0x00A60960" />
+        <register type="FLEX" address="0x0000E45C" value="0x00862762" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A62962" />
+    </register_config>
+  </set>
+
+  <set name="EuActivity3"
+       chipset="DG2"
+       symbol_name="EuActivity3"
+       underscore_name="eu_activity3"
+       mdapi_supported_apis="OGL OGL4 OCL MEDIA IO"
+       hw_config_guid="4fd009c5-6eae-4ecb-a2d1-41edf211cd06"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS EM Pipe Active"
+             symbol_name="VsEmActive"
+             underscore_name="vs_em_active"
+             description="The percentage of time in which EU EM INT pipeline was actively processing a vertex shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS EM Pipe Active"
+             symbol_name="PsEmActive"
+             underscore_name="ps_em_active"
+             description="The percentage of time in which EU EM INT pipeline was actively processing a fragment shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Send Pipe Active"
+             symbol_name="PsSendActive"
+             underscore_name="ps_send_active"
+             description="The percentage of time in which EU send pipeline was actively processing a fragment shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x00800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00000000" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00811711" />
+        <register type="FLEX" address="0x0000E558" value="0x00A11911" />
+        <register type="FLEX" address="0x0000E658" value="0x00851751" />
+        <register type="FLEX" address="0x0000E758" value="0x00A51951" />
+        <register type="FLEX" address="0x0000E45C" value="0x00852752" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A52952" />
+    </register_config>
+  </set>
+
+  <set name="EuActivity4"
+       chipset="DG2"
+       symbol_name="EuActivity4"
+       underscore_name="eu_activity4"
+       mdapi_supported_apis="OGL OGL4 OCL MEDIA IO"
+       hw_config_guid="5226090d-b636-4845-9f76-545cb5db7867"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS FPU Pipe Active"
+             symbol_name="HsFpuActive"
+             underscore_name="hs_fpu_active"
+             description="The percentage of time in which EU FPU INT64 pipeline was actively processing a hull shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS FPU Pipe Active"
+             symbol_name="DsFpuActive"
+             underscore_name="ds_fpu_active"
+             description="The percentage of time in which EU FPU INT64 pipeline was actively processing a domain shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Send Pipe Active"
+             symbol_name="VsSendActive"
+             underscore_name="vs_send_active"
+             description="The percentage of time in which EU send pipeline was actively processing a vertex shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x00800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00000000" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00820720" />
+        <register type="FLEX" address="0x0000E558" value="0x00A20920" />
+        <register type="FLEX" address="0x0000E658" value="0x00830730" />
+        <register type="FLEX" address="0x0000E758" value="0x00A30930" />
+        <register type="FLEX" address="0x0000E45C" value="0x00812712" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A12912" />
+    </register_config>
+  </set>
+
+  <set name="EuActivity5"
+       chipset="DG2"
+       symbol_name="EuActivity5"
+       underscore_name="eu_activity5"
+       mdapi_supported_apis="OGL OGL4 OCL MEDIA IO"
+       hw_config_guid="c8d33b8f-3152-4c81-99bf-d3cb45d7e1ac"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS EM Pipe Active"
+             symbol_name="HsEmActive"
+             underscore_name="hs_em_active"
+             description="The percentage of time in which EU EM INT pipeline was actively processing a hull shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS EM Pipe Active"
+             symbol_name="DsEmActive"
+             underscore_name="ds_em_active"
+             description="The percentage of time in which EU EM INT pipeline was actively processing a domain shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Send Pipe Active"
+             symbol_name="HsSendActive"
+             underscore_name="hs_send_active"
+             description="The percentage of time in which EU send pipeline was actively processing a hull shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x00800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00000000" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00821721" />
+        <register type="FLEX" address="0x0000E558" value="0x00A21921" />
+        <register type="FLEX" address="0x0000E658" value="0x00831731" />
+        <register type="FLEX" address="0x0000E758" value="0x00A31931" />
+        <register type="FLEX" address="0x0000E45C" value="0x00822722" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A22922" />
+    </register_config>
+  </set>
+
+  <set name="EuActivity6"
+       chipset="DG2"
+       symbol_name="EuActivity6"
+       underscore_name="eu_activity6"
+       mdapi_supported_apis="OGL OGL4 OCL MEDIA IO"
+       hw_config_guid="708d01ed-bc9b-420e-8048-9060abcbb1f0"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS FPU Pipe Active"
+             symbol_name="GsFpuActive"
+             underscore_name="gs_fpu_active"
+             description="The percentage of time in which EU FPU INT64 pipeline was actively processing a geometry shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS EM Pipe Active"
+             symbol_name="GsEmActive"
+             underscore_name="gs_em_active"
+             description="The percentage of time in which EU EM INT pipeline was actively processing a geometry shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Send Pipe Active"
+             symbol_name="GsSendActive"
+             underscore_name="gs_send_active"
+             description="The percentage of time in which EU send pipeline was actively processing a geometry shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x00800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00000000" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00840740" />
+        <register type="FLEX" address="0x0000E558" value="0x00A40940" />
+        <register type="FLEX" address="0x0000E658" value="0x00841741" />
+        <register type="FLEX" address="0x0000E758" value="0x00A41941" />
+        <register type="FLEX" address="0x0000E45C" value="0x00842742" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A42942" />
+    </register_config>
+  </set>
+
+  <set name="EuActivity7"
+       chipset="DG2"
+       symbol_name="EuActivity7"
+       underscore_name="eu_activity7"
+       mdapi_supported_apis="OGL OGL4 OCL MEDIA IO"
+       hw_config_guid="72262d5b-a9e6-4742-820b-2a7bef9a7e61"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU FPU Pipe Active"
+             symbol_name="FpuActive"
+             underscore_name="fpu_active"
+             description="The percentage of time in which EU FPU INT64 pipeline was actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Pipes"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EM Pipe Active"
+             symbol_name="EmActive"
+             underscore_name="em_active"
+             description="The percentage of time in which EU EM INT pipeline was actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Pipes"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EM And FPU Pipes Active"
+             symbol_name="EmFpuActive"
+             underscore_name="em_fpu_active"
+             description="The percentage of time in which EU EM INT and FPU INT64 pipelines were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Pipes"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU AVG IPC Rate"
+             symbol_name="EuAvgIpcRate"
+             underscore_name="eu_avg_ipc_rate"
+             description="The average rate of IPC calculated for 2 FPU pipelines."
+             data_type="float"
+             max_equation="2"
+             units="number"
+             semantic_type="ratio"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD  A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD  A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FADD  A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FSUB FDIV 1 FADD"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier4 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x00800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00000000" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00803703" />
+        <register type="FLEX" address="0x0000E558" value="0x00A03903" />
+        <register type="FLEX" address="0x0000E658" value="0x00800700" />
+        <register type="FLEX" address="0x0000E758" value="0x00A00900" />
+        <register type="FLEX" address="0x0000E45C" value="0x00801701" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A01901" />
+    </register_config>
+  </set>
+
+  <set name="EuActivity8"
+       chipset="DG2"
+       symbol_name="EuActivity8"
+       underscore_name="eu_activity8"
+       mdapi_supported_apis="OGL OGL4 OCL MEDIA IO"
+       hw_config_guid="11c685de-1445-41b8-acb5-8e79c75c4fd8"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Send Pipe Active"
+             symbol_name="DsSendActive"
+             underscore_name="ds_send_active"
+             description="The percentage of time in which EU send pipeline was actively processing a domain shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SYS And FPU Pipes Active"
+             symbol_name="SysFpuActive"
+             underscore_name="sys_fpu_active"
+             description="The percentage of time in which EU Systolic and FPU INT64 pipelines were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Pipes"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SYS Pipe Active"
+             symbol_name="SysActive"
+             underscore_name="sys_active"
+             description="The percentage of time in which EU SYS pipeline was actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Pipes"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x00800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00000000" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00832732" />
+        <register type="FLEX" address="0x0000E558" value="0x00A32932" />
+        <register type="FLEX" address="0x0000E658" value="0x00809709" />
+        <register type="FLEX" address="0x0000E758" value="0x00A09909" />
+        <register type="FLEX" address="0x0000E45C" value="0x0080A70A" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A0A90A" />
+    </register_config>
+  </set>
+
+  <set name="Metric set TestOa"
+       chipset="DG2"
+       symbol_name="TestOa"
+       underscore_name="test_oa"
+       mdapi_supported_apis="VK OGL OCL MEDIA IO"
+       hw_config_guid="2db1e1d4-8c48-4bb7-b248-0f40b70433ff"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="TestCounter0"
+             symbol_name="Counter0"
+             underscore_name="counter0"
+             description="HW test counter 0. Factor: 0.0"
+             data_type="uint64"
+             units="events"
+             semantic_type="event"
+             equation="B 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="TestCounter1"
+             symbol_name="Counter1"
+             underscore_name="counter1"
+             description="HW test counter 1. Factor: 1.0"
+             data_type="uint64"
+             units="events"
+             semantic_type="event"
+             equation="B 1 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="TestCounter2"
+             symbol_name="Counter2"
+             underscore_name="counter2"
+             description="HW test counter 2. Factor: 1.0"
+             data_type="uint64"
+             units="events"
+             semantic_type="event"
+             equation="B 2 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="TestCounter3"
+             symbol_name="Counter3"
+             underscore_name="counter3"
+             description="HW test counter 3. Factor: 0.5"
+             data_type="uint64"
+             units="events"
+             semantic_type="event"
+             equation="B 3 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="TestCounter4"
+             symbol_name="Counter4"
+             underscore_name="counter4"
+             description="HW test counter 4. Factor: 0.3333"
+             data_type="uint64"
+             units="events"
+             semantic_type="event"
+             equation="B 4 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="TestCounter5"
+             symbol_name="Counter5"
+             underscore_name="counter5"
+             description="HW test counter 5. Factor: 0.3333"
+             data_type="uint64"
+             units="events"
+             semantic_type="event"
+             equation="B 5 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="TestCounter6"
+             symbol_name="Counter6"
+             underscore_name="counter6"
+             description="HW test counter 6. Factor: 0.16666"
+             data_type="uint64"
+             units="events"
+             semantic_type="event"
+             equation="B 6 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="TestCounter7"
+             symbol_name="Counter7"
+             underscore_name="counter7"
+             description="HW test counter 7. Factor: 0.6666"
+             data_type="uint64"
+             units="events"
+             semantic_type="event"
+             equation="B 7 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="TestCounter8"
+             symbol_name="Counter8"
+             underscore_name="counter8"
+             description="HW test counter 8. Should be equal to 1 in IOStream or in OAG query mode"
+             data_type="uint64"
+             units="events"
+             semantic_type="event"
+             equation="C 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="TestCounter9 - OAR enable"
+             symbol_name="Counter9"
+             underscore_name="counter9"
+             description="HW test counter 9. Should be equal to 1 in query."
+             data_type="uint64"
+             units="events"
+             semantic_type="event"
+             equation="C 1 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x2E080000" />
+        <register type="NOA" address="0x00009888" value="0x34080000" />
+        <register type="NOA" address="0x00009888" value="0x06088000" />
+        <register type="NOA" address="0x00009888" value="0x10080FF3" />
+        <register type="NOA" address="0x00009888" value="0x08080000" />
+        <register type="NOA" address="0x00009888" value="0x02080000" />
+        <register type="NOA" address="0x00009888" value="0x02032400" />
+        <register type="NOA" address="0x00009888" value="0x48030000" />
+        <register type="NOA" address="0x00009888" value="0x4A030000" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0xF0800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0xF0800000" />
+        <register type="OA" address="0x0000DC40" value="0x00FF0000" />
+        <register type="OA" address="0x0000D940" value="0x00000004" />
+        <register type="OA" address="0x0000D944" value="0x0000FFFF" />
+        <register type="OA" address="0x0000DC00" value="0x00000004" />
+        <register type="OA" address="0x0000DC04" value="0x0000FFFF" />
+        <register type="OA" address="0x0000D948" value="0x00000003" />
+        <register type="OA" address="0x0000D94C" value="0x0000FFFF" />
+        <register type="OA" address="0x0000DC08" value="0x00000003" />
+        <register type="OA" address="0x0000DC0C" value="0x0000FFFF" />
+        <register type="OA" address="0x0000D950" value="0x00000007" />
+        <register type="OA" address="0x0000D954" value="0x0000FFFF" />
+        <register type="OA" address="0x0000DC10" value="0x00000007" />
+        <register type="OA" address="0x0000DC14" value="0x0000FFFF" />
+        <register type="OA" address="0x0000D958" value="0x00100002" />
+        <register type="OA" address="0x0000D95C" value="0x0000FFF7" />
+        <register type="OA" address="0x0000DC18" value="0x00100002" />
+        <register type="OA" address="0x0000DC1C" value="0x0000FFF7" />
+        <register type="OA" address="0x0000D960" value="0x00100002" />
+        <register type="OA" address="0x0000D964" value="0x0000FFCF" />
+        <register type="OA" address="0x0000DC20" value="0x00100002" />
+        <register type="OA" address="0x0000DC24" value="0x0000FFCF" />
+        <register type="OA" address="0x0000D968" value="0x00100082" />
+        <register type="OA" address="0x0000D96C" value="0x0000FFEF" />
+        <register type="OA" address="0x0000DC28" value="0x00100082" />
+        <register type="OA" address="0x0000DC2C" value="0x0000FFEF" />
+        <register type="OA" address="0x0000D970" value="0x001000C2" />
+        <register type="OA" address="0x0000D974" value="0x0000FFE7" />
+        <register type="OA" address="0x0000DC30" value="0x001000C2" />
+        <register type="OA" address="0x0000DC34" value="0x0000FFE7" />
+        <register type="OA" address="0x0000D978" value="0x00100001" />
+        <register type="OA" address="0x0000D97C" value="0x0000FFE7" />
+        <register type="OA" address="0x0000DC38" value="0x00100001" />
+        <register type="OA" address="0x0000DC3C" value="0x0000FFE7" />
+    </register_config>
+  </set>
+
+</metrics>
diff --git a/lib/i915/perf-configs/perf-metricset-codegen.py b/lib/i915/perf-configs/perf-metricset-codegen.py
index 00528c31..58237c27 100644
--- a/lib/i915/perf-configs/perf-metricset-codegen.py
+++ b/lib/i915/perf-configs/perf-metricset-codegen.py
@@ -126,6 +126,19 @@ def generate_metric_sets(args, gen):
                 metric_set->c_offset = metric_set->b_offset + 8;
                 metric_set->perfcnt_offset = metric_set->c_offset + 8;
 
+            """))
+        elif gen.chipset == "dg2":
+            c(textwrap.dedent("""\
+                metric_set->perf_oa_format = I915_OA_FORMAT_A24u40_A14u32_B8_C8;
+
+                metric_set->perf_raw_size = 256;
+                metric_set->gpu_time_offset = 0;
+                metric_set->gpu_clock_offset = 1;
+                metric_set->a_offset = 2;
+                metric_set->b_offset = metric_set->a_offset + 38;
+                metric_set->c_offset = metric_set->b_offset + 8;
+                metric_set->perfcnt_offset = metric_set->c_offset + 8;
+
             """))
         else:
             c(textwrap.dedent("""\
diff --git a/lib/i915/perf-configs/update-guids.py b/lib/i915/perf-configs/update-guids.py
index d38a7a95..3cd06f13 100755
--- a/lib/i915/perf-configs/update-guids.py
+++ b/lib/i915/perf-configs/update-guids.py
@@ -174,7 +174,7 @@ chipsets = [ 'hsw',
              'bxt', 'glk',
              'cnl',
              'icl', 'ehl',
-             'tglgt1', 'tglgt2', 'rkl', 'dg1', 'adl' ]
+             'tglgt1', 'tglgt2', 'rkl', 'dg1', 'adl', 'dg2', ]
 
 for chipset in chipsets:
     filename = 'oa-' + chipset + '.xml'
diff --git a/lib/i915/perf.c b/lib/i915/perf.c
index d004313f..39179546 100644
--- a/lib/i915/perf.c
+++ b/lib/i915/perf.c
@@ -58,6 +58,7 @@
 #include "i915_perf_metrics_rkl.h"
 #include "i915_perf_metrics_dg1.h"
 #include "i915_perf_metrics_adl.h"
+#include "i915_perf_metrics_dg2.h"
 
 static int
 perf_ioctl(int fd, unsigned long request, void *arg)
@@ -280,6 +281,8 @@ intel_perf_for_devinfo(uint32_t device_id,
 	} else if (devinfo->is_alderlake_s || devinfo->is_alderlake_p ||
 		   devinfo->is_raptorlake_s || devinfo->is_alderlake_n) {
 		intel_perf_load_metrics_adl(perf);
+	} else if (devinfo->is_dg2) {
+		intel_perf_load_metrics_dg2(perf);
 	} else {
 		return unsupported_i915_perf_platform(perf);
 	}
diff --git a/lib/meson.build b/lib/meson.build
index 98c2803b..8dabd64d 100644
--- a/lib/meson.build
+++ b/lib/meson.build
@@ -254,7 +254,7 @@ i915_perf_hardware = [
   'bxt', 'glk',
   'cnl',
   'icl', 'ehl',
-  'tglgt1', 'tglgt2', 'rkl', 'dg1', 'adl',
+  'tglgt1', 'tglgt2', 'rkl', 'dg1', 'adl', 'dg2',
 ]
 
 i915_xml_files = []
-- 
2.25.1

Re: [igt-dev] [PATCH i-g-t 23/23] lib/i915/perf: Add DG2 metrics

[Umesh Nerlige Ramappa]

On Tue, Aug 23, 2022 at 06:30:36PM +0000, Umesh Nerlige Ramappa wrote:
>From: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
>
>Add DG2 metrics XML files for perf OA tests.
>
>Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>

Umesh
>---
> lib/i915/perf-configs/guids.xml               |   24 +
> lib/i915/perf-configs/mdapi-xml-convert.py    |    5 +-
> lib/i915/perf-configs/oa-dg2.xml              | 9920 +++++++++++++++++
> .../perf-configs/perf-metricset-codegen.py    |   13 +
> lib/i915/perf-configs/update-guids.py         |    2 +-
> lib/i915/perf.c                               |    3 +
> lib/meson.build                               |    2 +-
> 7 files changed, 9965 insertions(+), 4 deletions(-)
> create mode 100644 lib/i915/perf-configs/oa-dg2.xml
>
>diff --git a/lib/i915/perf-configs/guids.xml b/lib/i915/perf-configs/guids.xml
>index 95c9d717..61a11542 100644
>--- a/lib/i915/perf-configs/guids.xml
>+++ b/lib/i915/perf-configs/guids.xml
>@@ -431,4 +431,28 @@
>     <guid config_hash="b420d7b956fb1b097a7bb812b6c87dc0" mdapi_config_hash="d81f15a77d18a099782c18acaa3132f7" id="c3ea2a23-f1c7-4a19-9da2-b569226fb6de" chipset="adl" name="EuActivity7" />
>     <guid config_hash="5face026dac6a0549f8a28cf226dfbc8" mdapi_config_hash="de7326838bcc8abf024bac9c771e22e4" id="a5e2f79b-cecb-4eff-8f29-cda8e2a58749" chipset="adl" name="EuActivity8" />
>     <guid config_hash="1affe7abeb3739d2c845cc6fa9aacf88" mdapi_config_hash="d071f4933574945979595246be61d3f0" id="89173c19-fcfe-48da-ac26-fb64425f141b" chipset="adl" name="TestOa" />
>+    <guid config_hash="9091b7d8f1627a3dda5d6d996eadd033" mdapi_config_hash="4a3cf9efdf4fd581b69308368b986500" id="8b20a2e7-04db-4196-a5ca-26b53cbf9a12" chipset="dg2" name="RenderBasic" />
>+    <guid config_hash="a0afa17cc8f442e6eabb3b1782cbca44" mdapi_config_hash="2a23a2f9f1bcf91a9b038b2cc6c9c911" id="f66d7204-54f5-4663-9d68-5213f70165ba" chipset="dg2" name="ComputeBasic" />
>+    <guid config_hash="34875baa1a12dda6dcf6d40e33a8d5ef" mdapi_config_hash="31dfc8c6ef9a79789bbe0b991c94112f" id="3b06d2f4-f594-49b5-b399-98550fa2d7b1" chipset="dg2" name="RenderPipeProfile" />
>+    <guid config_hash="5f0f4b7c1e4cae4d04ab60a66f1d98a8" mdapi_config_hash="58ec2ac3713bd8e21112f8fddbe28c48" id="1afb0edc-585c-4be9-8c0c-bc72325106b0" chipset="dg2" name="HDCAndSF1" />
>+    <guid config_hash="69dbeb5aba00131465b68f20c910421b" mdapi_config_hash="cf8e3757a118ffffb76db95917cae040" id="a3b35902-f293-41fe-97d0-3d895bc5fab3" chipset="dg2" name="HDCAndSF2" />
>+    <guid config_hash="e563f102f8b3046e8deaea0eb81b8784" mdapi_config_hash="81d17bd3ab0a18f4491c2d0a11037c7a" id="b04e9460-630f-4682-b0ed-ebb0e74a6c9d" chipset="dg2" name="RasterizerAndPixelBackend1" />
>+    <guid config_hash="b574d106f1f71eae1e8a0908cad2c010" mdapi_config_hash="d9c0a07f470d2a9bf7bbf20f79488abf" id="57a0abf4-8d81-4856-ab86-01f15f92c81c" chipset="dg2" name="RasterizerAndPixelBackend2" />
>+    <guid config_hash="3316babac928063b799008fc324812b1" mdapi_config_hash="c9cd0a76400e340b0ffdb129fff205c7" id="2314eaa3-164f-4004-9684-3f1d16e875f9" chipset="dg2" name="Sampler_1" />
>+    <guid config_hash="31c5ffa0c3508fb166ace213356edbaf" mdapi_config_hash="e2bce131c17fa27ab826a05ed65cad5a" id="c60dea82-db13-4882-8d31-a02bce151ef7" chipset="dg2" name="Sampler_2" />
>+    <guid config_hash="b3460c2d0d4e7bcbcc46a3f1c9df7e8f" mdapi_config_hash="ee6821f6954a10562b6314c5ef394b50" id="7f5cb8ec-38bb-48c9-9ad2-60075eba2833" chipset="dg2" name="TDL_1" />
>+    <guid config_hash="9ff61984b39a3d19a2abf3a0fd4c2285" mdapi_config_hash="013fa392a5c8b3bb93a776146d31bc95" id="141ce507-de6c-428c-95b2-076194d650db" chipset="dg2" name="TDL_2" />
>+    <guid config_hash="11b056437f6c491c2a44e4ca029dd57b" mdapi_config_hash="c4a4c4cec33f45576af76022e6f634ae" id="54a85b07-9ff2-4a12-bfb6-6bcdd9d4019c" chipset="dg2" name="TDL_3" />
>+    <guid config_hash="5cf268036a3ae548ddd918fdbd67a64b" mdapi_config_hash="4744760a7ff39c509ecab43f3f715790" id="651c40e5-195c-49da-b232-a9a0f753d899" chipset="dg2" name="TDL_4" />
>+    <guid config_hash="f43033715b8153ccb51147452de5bf99" mdapi_config_hash="e77513bc5aa341b061701d6d973b1b4c" id="9a1a14ad-ab92-4204-8a91-c4159a54f712" chipset="dg2" name="L3_1" />
>+    <guid config_hash="1c4da3e296779bdbef2262460f3f0fcb" mdapi_config_hash="27c652b5bcd6e800c23a3cacc103c2cb" id="85de4c29-76b8-42a6-b6ef-8a2bed008fcf" chipset="dg2" name="GpuBusyness" />
>+    <guid config_hash="a1c8aac6e58018ff3bf78169e5d7cf00" mdapi_config_hash="3d1142aa73a474eccef2bd23027e71b9" id="8dc7142a-df10-41f2-acaa-9faea547ce73" chipset="dg2" name="EuActivity1" />
>+    <guid config_hash="880a408e8bc0c0a82df014cbbe0a1a9e" mdapi_config_hash="d998d2f096753140dc92ed9421414ae1" id="d3c05cce-4ae7-4291-82ab-b0bcade6cf7a" chipset="dg2" name="EuActivity2" />
>+    <guid config_hash="6687d4c4e45e62d0d28792b04882edae" mdapi_config_hash="c5b1271909b4d72ad1503d5d8a50d30a" id="4fd009c5-6eae-4ecb-a2d1-41edf211cd06" chipset="dg2" name="EuActivity3" />
>+    <guid config_hash="f770062c29026a186485a2190b7ce3fb" mdapi_config_hash="7a83985438139dcfdf7d653be1e665b0" id="5226090d-b636-4845-9f76-545cb5db7867" chipset="dg2" name="EuActivity4" />
>+    <guid config_hash="4e13b511b35757cc97602aae22b7d2d3" mdapi_config_hash="5405e0eff1e1b0af86f45e832c604fcd" id="c8d33b8f-3152-4c81-99bf-d3cb45d7e1ac" chipset="dg2" name="EuActivity5" />
>+    <guid config_hash="243a5047b39033bca9a16db6bf2aca3f" mdapi_config_hash="2bc02385284ccdbbd0d233a428861e59" id="708d01ed-bc9b-420e-8048-9060abcbb1f0" chipset="dg2" name="EuActivity6" />
>+    <guid config_hash="d0f2a1c661860e4aa94e25e9f4a6f8eb" mdapi_config_hash="75694dd6a99c38e9446d7bf68f2269a3" id="72262d5b-a9e6-4742-820b-2a7bef9a7e61" chipset="dg2" name="EuActivity7" />
>+    <guid config_hash="5e7354b4cbe7b9b03acb504ef00aa8ce" mdapi_config_hash="299c68337c46a2483c1254fdead4cdcc" id="11c685de-1445-41b8-acb5-8e79c75c4fd8" chipset="dg2" name="EuActivity8" />
>+    <guid config_hash="24977eb13d34b0da58c0b33f6d7f841a" mdapi_config_hash="48715fb10d2ada8c23f7108303b7c343" id="2db1e1d4-8c48-4bb7-b248-0f40b70433ff" chipset="dg2" name="TestOa" />
> </guids>
>diff --git a/lib/i915/perf-configs/mdapi-xml-convert.py b/lib/i915/perf-configs/mdapi-xml-convert.py
>index 1d161674..72b52bfd 100755
>--- a/lib/i915/perf-configs/mdapi-xml-convert.py
>+++ b/lib/i915/perf-configs/mdapi-xml-convert.py
>@@ -141,6 +141,7 @@ chipsets = {
>     'RKL': gen8_11_chipset_params,
>     'DG1': gen8_11_chipset_params,
>     'ADL': gen8_11_chipset_params,
>+    'DG2': xehpsdv_chipset_params,
> }
>
> register_types = { 'OA', 'NOA', 'FLEX', 'PM' }
>@@ -214,7 +215,7 @@ def read_token_to_rpn_read(chipset, token, raw_offsets):
>                 assert 0
>         elif offset < b_offset:
>             a_cnt_offset = int((offset - a_offset) / 4)
>-            if chipset == "XEHPSDV":
>+            if chipset == "XEHPSDV" or chipset == "DG2":
>                 # Most A counters are in a contiguous array, except
>                 # this A37.
>                 if a_cnt_offset == 42:
>@@ -245,7 +246,7 @@ def read_token_to_rpn_read(chipset, token, raw_offsets):
>                 return "C " + str(idx - 54) + " READ"
>             else:
>                 return "{0} READ".format(read_value(chipset, offset))
>-        elif chipset == "XEHPSDV":
>+        elif chipset == "XEHPSDV" or chipset == "DG2":
>             # For XEHPSDV the array of accumulated counters is
>             # assumed to start with a GPU_TIME then GPU_CLOCK,
>             # then 38 A counters, then 8 B counters and finally
>diff --git a/lib/i915/perf-configs/oa-dg2.xml b/lib/i915/perf-configs/oa-dg2.xml
>new file mode 100644
>index 00000000..9d3e2541
>--- /dev/null
>+++ b/lib/i915/perf-configs/oa-dg2.xml
>@@ -0,0 +1,9920 @@
>+<?xml version="1.0"?>
>+<metrics version="1602592998" merge_md5="">
>+  <set name="Render Basic Metrics"
>+       chipset="DG2"
>+       symbol_name="RenderBasic"
>+       underscore_name="render_basic"
>+       mdapi_supported_apis="VK OGL OCL IO"
>+       hw_config_guid="8b20a2e7-04db-4196-a5ca-26b53cbf9a12"
>+       >
>+    <counter name="GPU Time Elapsed"
>+             symbol_name="GpuTime"
>+             underscore_name="gpu_time"
>+             description="Time elapsed on the GPU during the measurement."
>+             data_type="uint64"
>+             units="ns"
>+             semantic_type="duration"
>+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Core Clocks"
>+             symbol_name="GpuCoreClocks"
>+             underscore_name="gpu_core_clocks"
>+             description="The total number of GPU core clocks elapsed during the measurement."
>+             data_type="uint64"
>+             units="cycles"
>+             semantic_type="event"
>+             equation="GPU_CLOCK 0 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="AVG GPU Core Frequency"
>+             symbol_name="AvgGpuCoreFrequency"
>+             underscore_name="avg_gpu_core_frequency"
>+             description="Average GPU Core Frequency in the measurement."
>+             data_type="uint64"
>+             max_equation="$GpuMaxFrequency"
>+             units="hz"
>+             semantic_type="event"
>+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="VS Threads Dispatched"
>+             symbol_name="VsThreads"
>+             underscore_name="vs_threads"
>+             description="The total number of vertex shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 1 READ"
>+             mdapi_group="EU Array/Vertex Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="HS Threads Dispatched"
>+             symbol_name="HsThreads"
>+             underscore_name="hs_threads"
>+             description="The total number of hull shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 2 READ"
>+             mdapi_group="EU Array/Hull Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="DS Threads Dispatched"
>+             symbol_name="DsThreads"
>+             underscore_name="ds_threads"
>+             description="The total number of domain shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 3 READ"
>+             mdapi_group="EU Array/Domain Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GS Threads Dispatched"
>+             symbol_name="GsThreads"
>+             underscore_name="gs_threads"
>+             description="The total number of geometry shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 5 READ"
>+             mdapi_group="EU Array/Geometry Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="FS Threads Dispatched"
>+             symbol_name="PsThreads"
>+             underscore_name="ps_threads"
>+             description="The total number of fragment shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 6 READ"
>+             mdapi_group="EU Array/Fragment Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="CS Threads Dispatched"
>+             symbol_name="CsThreads"
>+             underscore_name="cs_threads"
>+             description="The total number of compute shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 4 READ"
>+             mdapi_group="EU Array/Compute Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Busy"
>+             symbol_name="GpuBusy"
>+             underscore_name="gpu_busy"
>+             description="The percentage of time in which the GPU has been processing GPU commands."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Active"
>+             symbol_name="EuActive"
>+             underscore_name="eu_active"
>+             description="The percentage of time in which the Execution Units were actively processing."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Stall"
>+             symbol_name="EuStall"
>+             underscore_name="eu_stall"
>+             description="The percentage of time in which the Execution Units were stalled."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Thread Occupancy"
>+             symbol_name="EuThreadOccupancy"
>+             underscore_name="eu_thread_occupancy"
>+             description="The percentage of time in which hardware threads occupied EUs."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Rasterized Pixels"
>+             symbol_name="RasterizedPixels"
>+             underscore_name="rasterized_pixels"
>+             description="The total number of rasterized pixels."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 21 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Rasterizer"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Early Hi-Depth Test Fails"
>+             symbol_name="HiDepthTestFails"
>+             underscore_name="hi_depth_test_fails"
>+             description="The total number of pixels dropped on early hierarchical depth test."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 22 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Rasterizer/Hi-Depth Test"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Early Depth Test Fails"
>+             symbol_name="EarlyDepthTestFails"
>+             underscore_name="early_depth_test_fails"
>+             description="The total number of pixels dropped on early depth test."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 23 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Rasterizer/Early Depth Test"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Samples Killed in FS"
>+             symbol_name="SamplesKilledInPs"
>+             underscore_name="samples_killed_in_ps"
>+             description="The total number of samples or pixels dropped in fragment shaders."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 24 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Fragment Shader"
>+             mdapi_usage_flags="Tier4 Overview Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Pixels Failing Tests"
>+             symbol_name="PixelsFailingPostPsTests"
>+             underscore_name="pixels_failing_post_ps_tests"
>+             description="The total number of pixels dropped on post-FS alpha, stencil, or depth tests."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 25 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Output Merger"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Samples Written"
>+             symbol_name="SamplesWritten"
>+             underscore_name="samples_written"
>+             description="The total number of samples or pixels written to all render targets."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 26 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Output Merger"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Samples Blended"
>+             symbol_name="SamplesBlended"
>+             underscore_name="samples_blended"
>+             description="The total number of blended samples or pixels written to all render targets."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 27 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Output Merger"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Sampler Texels"
>+             symbol_name="SamplerTexels"
>+             underscore_name="sampler_texels"
>+             description="The total number of texels seen on input (with 2x2 accuracy) in all sampler units."
>+             data_type="uint64"
>+             units="texels"
>+             semantic_type="event"
>+             equation="A 28 READ 4 UMUL"
>+             mdapi_group="Sampler/Sampler Input"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Sampler Texels Misses"
>+             symbol_name="SamplerTexelMisses"
>+             underscore_name="sampler_texel_misses"
>+             description="The total number of texels lookups (with 2x2 accuracy) that missed L1 sampler cache."
>+             data_type="uint64"
>+             units="texels"
>+             semantic_type="event"
>+             equation="A 29 READ 4 UMUL"
>+             mdapi_group="Sampler/Sampler Cache"
>+             mdapi_usage_flags="Tier3 Batch Frame Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="SLM Bytes Read"
>+             symbol_name="SlmBytesRead"
>+             underscore_name="slm_bytes_read"
>+             description="The total number of GPU memory bytes read from shared local memory."
>+             data_type="uint64"
>+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
>+             units="bytes"
>+             semantic_type="throughput"
>+             equation="A 30 READ 64 UMUL"
>+             mdapi_group="L3/Data Port/SLM"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="SLM Bytes Written"
>+             symbol_name="SlmBytesWritten"
>+             underscore_name="slm_bytes_written"
>+             description="The total number of GPU memory bytes written into shared local memory."
>+             data_type="uint64"
>+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
>+             units="bytes"
>+             semantic_type="throughput"
>+             equation="A 31 READ 64 UMUL"
>+             mdapi_group="L3/Data Port/SLM"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Shader Memory Accesses"
>+             symbol_name="ShaderMemoryAccesses"
>+             underscore_name="shader_memory_accesses"
>+             description="The total number of shader memory accesses to L3."
>+             data_type="uint64"
>+             units="messages"
>+             semantic_type="event"
>+             equation="A 32 READ"
>+             mdapi_group="L3/Data Port"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Shader Atomic Memory Accesses"
>+             symbol_name="ShaderAtomics"
>+             underscore_name="shader_atomics"
>+             description="The total number of shader atomic memory accesses."
>+             data_type="uint64"
>+             units="messages"
>+             semantic_type="event"
>+             equation="A 34 READ"
>+             mdapi_group="L3/Data Port/Atomics"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="L3 Shader Throughput"
>+             symbol_name="L3ShaderThroughput"
>+             underscore_name="l3_shader_throughput"
>+             description="The total number of GPU memory bytes transferred between shaders and L3 caches w/o URB."
>+             data_type="uint64"
>+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
>+             units="bytes"
>+             semantic_type="throughput"
>+             equation="$ShaderMemoryAccesses 64 UMUL"
>+             mdapi_group="L3/Data Port"
>+             mdapi_usage_flags="Tier2 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Shader Barrier Messages"
>+             symbol_name="ShaderBarriers"
>+             underscore_name="shader_barriers"
>+             description="The total number of shader barrier messages."
>+             data_type="uint64"
>+             units="messages"
>+             semantic_type="event"
>+             equation="A 35 READ"
>+             mdapi_group="EU Array/Barrier"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GTI Read Throughput"
>+             symbol_name="GtiReadThroughput"
>+             underscore_name="gti_read_throughput"
>+             description="The total number of GPU memory bytes read from GTI."
>+             data_type="uint64"
>+             max_equation="$GpuCoreClocks 512 UMUL"
>+             units="bytes"
>+             semantic_type="throughput"
>+             equation="128  A 36 READ UMUL"
>+             mdapi_group="GTI"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GTI Write Throughput"
>+             symbol_name="GtiWriteThroughput"
>+             underscore_name="gti_write_throughput"
>+             description="The total number of GPU memory bytes written to GTI."
>+             data_type="uint64"
>+             max_equation="$GpuCoreClocks 512 UMUL"
>+             units="bytes"
>+             semantic_type="throughput"
>+             equation="C 5 READ C 4 READ UADD C 3 READ UADD C 2 READ UADD C 1 READ UADD C 0 READ UADD B 7 READ UADD B 6 READ UADD 128 UMUL"
>+             mdapi_group="GTI"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Sampler Slice0 Dualsubslice0 is busy"
>+             symbol_name="Sampler00Busy"
>+             underscore_name="sampler00_busy"
>+             description="The percentage of time when sampler slice0 dualsubslice0 is busy"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 1 AND"
>+             mdapi_group="GPU/Sampler"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="subslice"
>+             />
>+    <counter name="Samplers Busy"
>+             symbol_name="SamplersBusy"
>+             underscore_name="samplers_busy"
>+             description="The percentage of time in which samplers or L1 cache are not idle."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="$Sampler00Busy"
>+             availability="$DualSubsliceMask 1 AND"
>+             mdapi_group="Sampler"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="slice"
>+             />
>+    <counter name="Sampler00 Bottleneck"
>+             symbol_name="Sampler00Bottleneck"
>+             underscore_name="sampler00_bottleneck"
>+             description="The percentage of time in which Slice0 Sampler0 has been slowing down the pipe when processing EU requests."
>+             low_watermark="5"
>+             high_watermark="15"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 1 AND"
>+             mdapi_group="Sampler"
>+             mdapi_usage_flags="Tier3 Indicate Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="subslice"
>+             />
>+    <counter name="Samplers Bottleneck"
>+             symbol_name="SamplerBottleneck"
>+             underscore_name="sampler_bottleneck"
>+             description="The percentage of time in which samplers have been slowing down the pipe when processing EU requests."
>+             low_watermark="5"
>+             high_watermark="15"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="$Sampler00Bottleneck"
>+             availability="$DualSubsliceMask 1 AND"
>+             mdapi_group="Sampler"
>+             mdapi_usage_flags="Tier3 Overview Indicate System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="slice"
>+             />
>+    <register_config type="NOA">
>+        <register type="NOA" address="0x00000D04" value="0x00000200" />
>+        <register type="NOA" address="0x00009884" value="0x0000000C" />
>+        <register type="NOA" address="0x00009888" value="0x1C840000" />
>+        <register type="NOA" address="0x00009888" value="0x1E842800" />
>+        <register type="NOA" address="0x00009888" value="0x02844000" />
>+        <register type="NOA" address="0x00009888" value="0x04840010" />
>+        <register type="NOA" address="0x00009888" value="0x12840800" />
>+        <register type="NOA" address="0x00009888" value="0x14840037" />
>+        <register type="NOA" address="0x00009888" value="0x00840000" />
>+        <register type="NOA" address="0x00009888" value="0x02865000" />
>+        <register type="NOA" address="0x00009888" value="0x0208A000" />
>+        <register type="NOA" address="0x00009888" value="0x020A5000" />
>+        <register type="NOA" address="0x00009884" value="0x00000001" />
>+        <register type="NOA" address="0x00009888" value="0x18013800" />
>+        <register type="NOA" address="0x00009888" value="0x18213800" />
>+        <register type="NOA" address="0x00009888" value="0x18413800" />
>+        <register type="NOA" address="0x00009888" value="0x18613800" />
>+        <register type="NOA" address="0x00009888" value="0x18813800" />
>+        <register type="NOA" address="0x00009888" value="0x18A13800" />
>+        <register type="NOA" address="0x00009888" value="0x18C13800" />
>+        <register type="NOA" address="0x00009888" value="0x18E13800" />
>+        <register type="NOA" address="0x00009888" value="0x02010006" />
>+        <register type="NOA" address="0x00009888" value="0x14010004" />
>+        <register type="NOA" address="0x00009888" value="0x00010000" />
>+        <register type="NOA" address="0x00009888" value="0x02210006" />
>+        <register type="NOA" address="0x00009888" value="0x12210100" />
>+        <register type="NOA" address="0x00009888" value="0x00210000" />
>+        <register type="NOA" address="0x00009888" value="0x02410006" />
>+        <register type="NOA" address="0x00009888" value="0x12410004" />
>+        <register type="NOA" address="0x00009888" value="0x00410000" />
>+        <register type="NOA" address="0x00009888" value="0x02610006" />
>+        <register type="NOA" address="0x00009888" value="0x10610100" />
>+        <register type="NOA" address="0x00009888" value="0x00610000" />
>+        <register type="NOA" address="0x00009888" value="0x02810006" />
>+        <register type="NOA" address="0x00009888" value="0x10810004" />
>+        <register type="NOA" address="0x00009888" value="0x00810000" />
>+        <register type="NOA" address="0x00009888" value="0x02A10006" />
>+        <register type="NOA" address="0x00009888" value="0x0EA10100" />
>+        <register type="NOA" address="0x00009888" value="0x00A10000" />
>+        <register type="NOA" address="0x00009888" value="0x02C10006" />
>+        <register type="NOA" address="0x00009888" value="0x0EC10004" />
>+        <register type="NOA" address="0x00009888" value="0x00C10000" />
>+        <register type="NOA" address="0x00009888" value="0x02E10006" />
>+        <register type="NOA" address="0x00009888" value="0x0CE10100" />
>+        <register type="NOA" address="0x00009888" value="0x00E10000" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x0208A000" />
>+        <register type="NOA" address="0x00009888" value="0x02034920" />
>+        <register type="NOA" address="0x00009888" value="0x04032924" />
>+        <register type="NOA" address="0x00009888" value="0x06030002" />
>+        <register type="NOA" address="0x00009888" value="0x020914E0" />
>+        <register type="NOA" address="0x00009888" value="0x020A1248" />
>+        <register type="NOA" address="0x00009888" value="0x040A0492" />
>+        <register type="NOA" address="0x00009888" value="0x040B029C" />
>+        <register type="NOA" address="0x00009888" value="0x46030860" />
>+        <register type="NOA" address="0x00009888" value="0x48030643" />
>+        <register type="NOA" address="0x00009888" value="0x4A030664" />
>+        <register type="NOA" address="0x00009888" value="0x16033000" />
>+        <register type="NOA" address="0x00009888" value="0x3E030840" />
>+        <register type="NOA" address="0x00009888" value="0x18033000" />
>+        <register type="NOA" address="0x00009888" value="0x1A033000" />
>+        <register type="NOA" address="0x00009888" value="0x40030042" />
>+        <register type="NOA" address="0x00009888" value="0x1C033000" />
>+        <register type="NOA" address="0x00009888" value="0x1E030000" />
>+        <register type="NOA" address="0x00009888" value="0x20030000" />
>+        <register type="NOA" address="0x00009888" value="0x42030000" />
>+        <register type="NOA" address="0x00009884" value="0x00000008" />
>+        <register type="NOA" address="0x00009888" value="0x0001C000" />
>+        <register type="NOA" address="0x00009888" value="0x02025000" />
>+        <register type="NOA" address="0x00009888" value="0x02045000" />
>+        <register type="NOA" address="0x00009888" value="0x02055000" />
>+        <register type="NOA" address="0x00009888" value="0x0208A000" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x5C030002" />
>+    </register_config>
>+    <register_config type="OA">
>+        <register type="OA" address="0x0000D920" value="0x00000000" />
>+        <register type="OA" address="0x0000D900" value="0x00000000" />
>+        <register type="OA" address="0x0000D904" value="0x00800000" />
>+        <register type="OA" address="0x0000D910" value="0x00000000" />
>+        <register type="OA" address="0x0000D914" value="0x00800000" />
>+        <register type="OA" address="0x0000DC40" value="0x00000000" />
>+    </register_config>
>+    <register_config type="FLEX">
>+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
>+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
>+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
>+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
>+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
>+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
>+    </register_config>
>+  </set>
>+
>+  <set name="Compute Metrics Basic"
>+       chipset="DG2"
>+       symbol_name="ComputeBasic"
>+       underscore_name="compute_basic"
>+       mdapi_supported_apis="VK OGL OCL IO"
>+       hw_config_guid="f66d7204-54f5-4663-9d68-5213f70165ba"
>+       >
>+    <counter name="GPU Time Elapsed"
>+             symbol_name="GpuTime"
>+             underscore_name="gpu_time"
>+             description="Time elapsed on the GPU during the measurement."
>+             data_type="uint64"
>+             units="ns"
>+             semantic_type="duration"
>+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Core Clocks"
>+             symbol_name="GpuCoreClocks"
>+             underscore_name="gpu_core_clocks"
>+             description="The total number of GPU core clocks elapsed during the measurement."
>+             data_type="uint64"
>+             units="cycles"
>+             semantic_type="event"
>+             equation="GPU_CLOCK 0 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="AVG GPU Core Frequency"
>+             symbol_name="AvgGpuCoreFrequency"
>+             underscore_name="avg_gpu_core_frequency"
>+             description="Average GPU Core Frequency in the measurement."
>+             data_type="uint64"
>+             max_equation="$GpuMaxFrequency"
>+             units="hz"
>+             semantic_type="event"
>+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Busy"
>+             symbol_name="GpuBusy"
>+             underscore_name="gpu_busy"
>+             description="The percentage of time in which the GPU has been processing GPU commands."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="VS Threads Dispatched"
>+             symbol_name="VsThreads"
>+             underscore_name="vs_threads"
>+             description="The total number of vertex shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 1 READ"
>+             mdapi_group="EU Array/Vertex Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="HS Threads Dispatched"
>+             symbol_name="HsThreads"
>+             underscore_name="hs_threads"
>+             description="The total number of hull shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 2 READ"
>+             mdapi_group="EU Array/Hull Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="DS Threads Dispatched"
>+             symbol_name="DsThreads"
>+             underscore_name="ds_threads"
>+             description="The total number of domain shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 3 READ"
>+             mdapi_group="EU Array/Domain Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GS Threads Dispatched"
>+             symbol_name="GsThreads"
>+             underscore_name="gs_threads"
>+             description="The total number of geometry shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 5 READ"
>+             mdapi_group="EU Array/Geometry Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="FS Threads Dispatched"
>+             symbol_name="PsThreads"
>+             underscore_name="ps_threads"
>+             description="The total number of fragment shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 6 READ"
>+             mdapi_group="EU Array/Fragment Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="CS Threads Dispatched"
>+             symbol_name="CsThreads"
>+             underscore_name="cs_threads"
>+             description="The total number of compute shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 4 READ"
>+             mdapi_group="EU Array/Compute Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Active"
>+             symbol_name="EuActive"
>+             underscore_name="eu_active"
>+             description="The percentage of time in which the Execution Units were actively processing."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Stall"
>+             symbol_name="EuStall"
>+             underscore_name="eu_stall"
>+             description="The percentage of time in which the Execution Units were stalled."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Thread Occupancy"
>+             symbol_name="EuThreadOccupancy"
>+             underscore_name="eu_thread_occupancy"
>+             description="The percentage of time in which hardware threads occupied EUs."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Rasterized Pixels"
>+             symbol_name="RasterizedPixels"
>+             underscore_name="rasterized_pixels"
>+             description="The total number of rasterized pixels."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 21 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Rasterizer"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Early Hi-Depth Test Fails"
>+             symbol_name="HiDepthTestFails"
>+             underscore_name="hi_depth_test_fails"
>+             description="The total number of pixels dropped on early hierarchical depth test."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 22 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Rasterizer/Hi-Depth Test"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Early Depth Test Fails"
>+             symbol_name="EarlyDepthTestFails"
>+             underscore_name="early_depth_test_fails"
>+             description="The total number of pixels dropped on early depth test."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 23 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Rasterizer/Early Depth Test"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Samples Killed in FS"
>+             symbol_name="SamplesKilledInPs"
>+             underscore_name="samples_killed_in_ps"
>+             description="The total number of samples or pixels dropped in fragment shaders."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 24 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Fragment Shader"
>+             mdapi_usage_flags="Tier4 Overview Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Pixels Failing Tests"
>+             symbol_name="PixelsFailingPostPsTests"
>+             underscore_name="pixels_failing_post_ps_tests"
>+             description="The total number of pixels dropped on post-FS alpha, stencil, or depth tests."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 25 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Output Merger"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Samples Written"
>+             symbol_name="SamplesWritten"
>+             underscore_name="samples_written"
>+             description="The total number of samples or pixels written to all render targets."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 26 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Output Merger"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Samples Blended"
>+             symbol_name="SamplesBlended"
>+             underscore_name="samples_blended"
>+             description="The total number of blended samples or pixels written to all render targets."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 27 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Output Merger"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Sampler Texels"
>+             symbol_name="SamplerTexels"
>+             underscore_name="sampler_texels"
>+             description="The total number of texels seen on input (with 2x2 accuracy) in all sampler units."
>+             data_type="uint64"
>+             units="texels"
>+             semantic_type="event"
>+             equation="A 28 READ 4 UMUL"
>+             mdapi_group="Sampler/Sampler Input"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Sampler Texels Misses"
>+             symbol_name="SamplerTexelMisses"
>+             underscore_name="sampler_texel_misses"
>+             description="The total number of texels lookups (with 2x2 accuracy) that missed L1 sampler cache."
>+             data_type="uint64"
>+             units="texels"
>+             semantic_type="event"
>+             equation="A 29 READ 4 UMUL"
>+             mdapi_group="Sampler/Sampler Cache"
>+             mdapi_usage_flags="Tier3 Batch Frame Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="SLM Bytes Read"
>+             symbol_name="SlmBytesRead"
>+             underscore_name="slm_bytes_read"
>+             description="The total number of GPU memory bytes read from shared local memory."
>+             data_type="uint64"
>+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
>+             units="bytes"
>+             semantic_type="throughput"
>+             equation="A 30 READ 64 UMUL"
>+             mdapi_group="L3/Data Port/SLM"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="SLM Bytes Written"
>+             symbol_name="SlmBytesWritten"
>+             underscore_name="slm_bytes_written"
>+             description="The total number of GPU memory bytes written into shared local memory."
>+             data_type="uint64"
>+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
>+             units="bytes"
>+             semantic_type="throughput"
>+             equation="A 31 READ 64 UMUL"
>+             mdapi_group="L3/Data Port/SLM"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Shader Memory Accesses"
>+             symbol_name="ShaderMemoryAccesses"
>+             underscore_name="shader_memory_accesses"
>+             description="The total number of shader memory accesses to L3."
>+             data_type="uint64"
>+             units="messages"
>+             semantic_type="event"
>+             equation="A 32 READ"
>+             mdapi_group="L3/Data Port"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Shader Atomic Memory Accesses"
>+             symbol_name="ShaderAtomics"
>+             underscore_name="shader_atomics"
>+             description="The total number of shader atomic memory accesses."
>+             data_type="uint64"
>+             units="messages"
>+             semantic_type="event"
>+             equation="A 34 READ"
>+             mdapi_group="L3/Data Port/Atomics"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="L3 Shader Throughput"
>+             symbol_name="L3ShaderThroughput"
>+             underscore_name="l3_shader_throughput"
>+             description="The total number of GPU memory bytes transferred between shaders and L3 caches w/o URB."
>+             data_type="uint64"
>+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
>+             units="bytes"
>+             semantic_type="throughput"
>+             equation="$ShaderMemoryAccesses 64 UMUL"
>+             mdapi_group="L3/Data Port"
>+             mdapi_usage_flags="Tier2 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Shader Barrier Messages"
>+             symbol_name="ShaderBarriers"
>+             underscore_name="shader_barriers"
>+             description="The total number of shader barrier messages."
>+             data_type="uint64"
>+             units="messages"
>+             semantic_type="event"
>+             equation="A 35 READ"
>+             mdapi_group="EU Array/Barrier"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GTI Read Throughput"
>+             symbol_name="GtiReadThroughput"
>+             underscore_name="gti_read_throughput"
>+             description="The total number of GPU memory bytes read from GTI."
>+             data_type="uint64"
>+             max_equation="$GpuCoreClocks 512 UMUL"
>+             units="bytes"
>+             semantic_type="throughput"
>+             equation="128  A 36 READ UMUL"
>+             mdapi_group="GTI"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GTI Write Throughput"
>+             symbol_name="GtiWriteThroughput"
>+             underscore_name="gti_write_throughput"
>+             description="The total number of GPU memory bytes written to GTI."
>+             data_type="uint64"
>+             max_equation="$GpuCoreClocks 512 UMUL"
>+             units="bytes"
>+             semantic_type="throughput"
>+             equation="C 7 READ C 6 READ UADD C 5 READ UADD C 4 READ UADD C 3 READ UADD C 2 READ UADD C 1 READ UADD C 0 READ UADD 128 UMUL"
>+             mdapi_group="GTI"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <register_config type="NOA">
>+        <register type="NOA" address="0x00000D04" value="0x00000200" />
>+        <register type="NOA" address="0x00009884" value="0x00000001" />
>+        <register type="NOA" address="0x00009888" value="0x18013800" />
>+        <register type="NOA" address="0x00009888" value="0x18213800" />
>+        <register type="NOA" address="0x00009888" value="0x18413800" />
>+        <register type="NOA" address="0x00009888" value="0x18613800" />
>+        <register type="NOA" address="0x00009888" value="0x18813800" />
>+        <register type="NOA" address="0x00009888" value="0x18A13800" />
>+        <register type="NOA" address="0x00009888" value="0x18C13800" />
>+        <register type="NOA" address="0x00009888" value="0x18E13800" />
>+        <register type="NOA" address="0x00009888" value="0x02010006" />
>+        <register type="NOA" address="0x00009888" value="0x16010004" />
>+        <register type="NOA" address="0x00009888" value="0x00010000" />
>+        <register type="NOA" address="0x00009888" value="0x02210006" />
>+        <register type="NOA" address="0x00009888" value="0x14210100" />
>+        <register type="NOA" address="0x00009888" value="0x00210000" />
>+        <register type="NOA" address="0x00009888" value="0x02410006" />
>+        <register type="NOA" address="0x00009888" value="0x14410004" />
>+        <register type="NOA" address="0x00009888" value="0x00410000" />
>+        <register type="NOA" address="0x00009888" value="0x02610006" />
>+        <register type="NOA" address="0x00009888" value="0x12610100" />
>+        <register type="NOA" address="0x00009888" value="0x00610000" />
>+        <register type="NOA" address="0x00009888" value="0x02810006" />
>+        <register type="NOA" address="0x00009888" value="0x12810004" />
>+        <register type="NOA" address="0x00009888" value="0x00810000" />
>+        <register type="NOA" address="0x00009888" value="0x02A10006" />
>+        <register type="NOA" address="0x00009888" value="0x10A10100" />
>+        <register type="NOA" address="0x00009888" value="0x00A10000" />
>+        <register type="NOA" address="0x00009888" value="0x02C10006" />
>+        <register type="NOA" address="0x00009888" value="0x10C10004" />
>+        <register type="NOA" address="0x00009888" value="0x00C10000" />
>+        <register type="NOA" address="0x00009888" value="0x02E10006" />
>+        <register type="NOA" address="0x00009888" value="0x0EE10100" />
>+        <register type="NOA" address="0x00009888" value="0x00E10000" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x02034800" />
>+        <register type="NOA" address="0x00009888" value="0x04034924" />
>+        <register type="NOA" address="0x00009888" value="0x06030004" />
>+        <register type="NOA" address="0x00009888" value="0x02093800" />
>+        <register type="NOA" address="0x00009888" value="0x0409000A" />
>+        <register type="NOA" address="0x00009888" value="0x020A1200" />
>+        <register type="NOA" address="0x00009888" value="0x040A2489" />
>+        <register type="NOA" address="0x00009888" value="0x060A0002" />
>+        <register type="NOA" address="0x00009888" value="0x040B2700" />
>+        <register type="NOA" address="0x00009888" value="0x060B0001" />
>+        <register type="NOA" address="0x00009888" value="0x48030000" />
>+        <register type="NOA" address="0x00009888" value="0x4A030402" />
>+        <register type="NOA" address="0x00009888" value="0x16030000" />
>+        <register type="NOA" address="0x00009888" value="0x3E030440" />
>+        <register type="NOA" address="0x00009888" value="0x18030000" />
>+        <register type="NOA" address="0x00009888" value="0x1A030000" />
>+        <register type="NOA" address="0x00009888" value="0x40030000" />
>+        <register type="NOA" address="0x00009888" value="0x1C030000" />
>+        <register type="NOA" address="0x00009888" value="0x1E030000" />
>+        <register type="NOA" address="0x00009888" value="0x20030000" />
>+        <register type="NOA" address="0x00009888" value="0x42030000" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x5C030002" />
>+    </register_config>
>+    <register_config type="OA">
>+        <register type="OA" address="0x0000D920" value="0x00000000" />
>+        <register type="OA" address="0x0000D900" value="0x00000000" />
>+        <register type="OA" address="0x0000D904" value="0x00800000" />
>+        <register type="OA" address="0x0000D910" value="0x00000000" />
>+        <register type="OA" address="0x0000D914" value="0x00800000" />
>+        <register type="OA" address="0x0000DC40" value="0x00000000" />
>+    </register_config>
>+    <register_config type="FLEX">
>+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
>+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
>+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
>+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
>+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
>+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
>+    </register_config>
>+  </set>
>+
>+  <set name="Render Metrics for 3D Pipeline Profile"
>+       chipset="DG2"
>+       symbol_name="RenderPipeProfile"
>+       underscore_name="render_pipe_profile"
>+       mdapi_supported_apis="VK OGL OCL IO"
>+       hw_config_guid="3b06d2f4-f594-49b5-b399-98550fa2d7b1"
>+       >
>+    <counter name="GPU Time Elapsed"
>+             symbol_name="GpuTime"
>+             underscore_name="gpu_time"
>+             description="Time elapsed on the GPU during the measurement."
>+             data_type="uint64"
>+             units="ns"
>+             semantic_type="duration"
>+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Core Clocks"
>+             symbol_name="GpuCoreClocks"
>+             underscore_name="gpu_core_clocks"
>+             description="The total number of GPU core clocks elapsed during the measurement."
>+             data_type="uint64"
>+             units="cycles"
>+             semantic_type="event"
>+             equation="GPU_CLOCK 0 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="AVG GPU Core Frequency"
>+             symbol_name="AvgGpuCoreFrequency"
>+             underscore_name="avg_gpu_core_frequency"
>+             description="Average GPU Core Frequency in the measurement."
>+             data_type="uint64"
>+             max_equation="$GpuMaxFrequency"
>+             units="hz"
>+             semantic_type="event"
>+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Busy"
>+             symbol_name="GpuBusy"
>+             underscore_name="gpu_busy"
>+             description="The percentage of time in which the GPU has been processing GPU commands."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="VS Threads Dispatched"
>+             symbol_name="VsThreads"
>+             underscore_name="vs_threads"
>+             description="The total number of vertex shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 1 READ"
>+             mdapi_group="EU Array/Vertex Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="HS Threads Dispatched"
>+             symbol_name="HsThreads"
>+             underscore_name="hs_threads"
>+             description="The total number of hull shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 2 READ"
>+             mdapi_group="EU Array/Hull Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="DS Threads Dispatched"
>+             symbol_name="DsThreads"
>+             underscore_name="ds_threads"
>+             description="The total number of domain shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 3 READ"
>+             mdapi_group="EU Array/Domain Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GS Threads Dispatched"
>+             symbol_name="GsThreads"
>+             underscore_name="gs_threads"
>+             description="The total number of geometry shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 5 READ"
>+             mdapi_group="EU Array/Geometry Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="FS Threads Dispatched"
>+             symbol_name="PsThreads"
>+             underscore_name="ps_threads"
>+             description="The total number of fragment shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 6 READ"
>+             mdapi_group="EU Array/Fragment Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="CS Threads Dispatched"
>+             symbol_name="CsThreads"
>+             underscore_name="cs_threads"
>+             description="The total number of compute shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 4 READ"
>+             mdapi_group="EU Array/Compute Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Active"
>+             symbol_name="EuActive"
>+             underscore_name="eu_active"
>+             description="The percentage of time in which the Execution Units were actively processing."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Stall"
>+             symbol_name="EuStall"
>+             underscore_name="eu_stall"
>+             description="The percentage of time in which the Execution Units were stalled."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Thread Occupancy"
>+             symbol_name="EuThreadOccupancy"
>+             underscore_name="eu_thread_occupancy"
>+             description="The percentage of time in which hardware threads occupied EUs."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Rasterized Pixels"
>+             symbol_name="RasterizedPixels"
>+             underscore_name="rasterized_pixels"
>+             description="The total number of rasterized pixels."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 21 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Rasterizer"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Early Hi-Depth Test Fails"
>+             symbol_name="HiDepthTestFails"
>+             underscore_name="hi_depth_test_fails"
>+             description="The total number of pixels dropped on early hierarchical depth test."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 22 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Rasterizer/Hi-Depth Test"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Early Depth Test Fails"
>+             symbol_name="EarlyDepthTestFails"
>+             underscore_name="early_depth_test_fails"
>+             description="The total number of pixels dropped on early depth test."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 23 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Rasterizer/Early Depth Test"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Samples Killed in FS"
>+             symbol_name="SamplesKilledInPs"
>+             underscore_name="samples_killed_in_ps"
>+             description="The total number of samples or pixels dropped in fragment shaders."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 24 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Fragment Shader"
>+             mdapi_usage_flags="Tier4 Overview Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Pixels Failing Tests"
>+             symbol_name="PixelsFailingPostPsTests"
>+             underscore_name="pixels_failing_post_ps_tests"
>+             description="The total number of pixels dropped on post-FS alpha, stencil, or depth tests."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 25 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Output Merger"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Samples Written"
>+             symbol_name="SamplesWritten"
>+             underscore_name="samples_written"
>+             description="The total number of samples or pixels written to all render targets."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 26 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Output Merger"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Samples Blended"
>+             symbol_name="SamplesBlended"
>+             underscore_name="samples_blended"
>+             description="The total number of blended samples or pixels written to all render targets."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 27 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Output Merger"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Sampler Texels"
>+             symbol_name="SamplerTexels"
>+             underscore_name="sampler_texels"
>+             description="The total number of texels seen on input (with 2x2 accuracy) in all sampler units."
>+             data_type="uint64"
>+             units="texels"
>+             semantic_type="event"
>+             equation="A 28 READ 4 UMUL"
>+             mdapi_group="Sampler/Sampler Input"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Sampler Texels Misses"
>+             symbol_name="SamplerTexelMisses"
>+             underscore_name="sampler_texel_misses"
>+             description="The total number of texels lookups (with 2x2 accuracy) that missed L1 sampler cache."
>+             data_type="uint64"
>+             units="texels"
>+             semantic_type="event"
>+             equation="A 29 READ 4 UMUL"
>+             mdapi_group="Sampler/Sampler Cache"
>+             mdapi_usage_flags="Tier3 Batch Frame Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="SLM Bytes Read"
>+             symbol_name="SlmBytesRead"
>+             underscore_name="slm_bytes_read"
>+             description="The total number of GPU memory bytes read from shared local memory."
>+             data_type="uint64"
>+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
>+             units="bytes"
>+             semantic_type="throughput"
>+             equation="A 30 READ 64 UMUL"
>+             mdapi_group="L3/Data Port/SLM"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="SLM Bytes Written"
>+             symbol_name="SlmBytesWritten"
>+             underscore_name="slm_bytes_written"
>+             description="The total number of GPU memory bytes written into shared local memory."
>+             data_type="uint64"
>+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
>+             units="bytes"
>+             semantic_type="throughput"
>+             equation="A 31 READ 64 UMUL"
>+             mdapi_group="L3/Data Port/SLM"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Shader Memory Accesses"
>+             symbol_name="ShaderMemoryAccesses"
>+             underscore_name="shader_memory_accesses"
>+             description="The total number of shader memory accesses to L3."
>+             data_type="uint64"
>+             units="messages"
>+             semantic_type="event"
>+             equation="A 32 READ"
>+             mdapi_group="L3/Data Port"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Shader Atomic Memory Accesses"
>+             symbol_name="ShaderAtomics"
>+             underscore_name="shader_atomics"
>+             description="The total number of shader atomic memory accesses."
>+             data_type="uint64"
>+             units="messages"
>+             semantic_type="event"
>+             equation="A 34 READ"
>+             mdapi_group="L3/Data Port/Atomics"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="L3 Shader Throughput"
>+             symbol_name="L3ShaderThroughput"
>+             underscore_name="l3_shader_throughput"
>+             description="The total number of GPU memory bytes transferred between shaders and L3 caches w/o URB."
>+             data_type="uint64"
>+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
>+             units="bytes"
>+             semantic_type="throughput"
>+             equation="$ShaderMemoryAccesses 64 UMUL"
>+             mdapi_group="L3/Data Port"
>+             mdapi_usage_flags="Tier2 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Shader Barrier Messages"
>+             symbol_name="ShaderBarriers"
>+             underscore_name="shader_barriers"
>+             description="The total number of shader barrier messages."
>+             data_type="uint64"
>+             units="messages"
>+             semantic_type="event"
>+             equation="A 35 READ"
>+             mdapi_group="EU Array/Barrier"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="VF0 Bottleneck"
>+             symbol_name="Vf0Bottleneck"
>+             underscore_name="vf0_bottleneck"
>+             description="The percentage of time in which vertex fetch pipeline stage was slowing down the 3D pipeline."
>+             low_watermark="5"
>+             high_watermark="15"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="3D Pipe/Input Assembler"
>+             mdapi_usage_flags="Tier3 Indicate Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="VS0 Bottleneck"
>+             symbol_name="Vs0Bottleneck"
>+             underscore_name="vs0_bottleneck"
>+             description="The percentage of time in which VS0 pipeline stage was slowing down the 3D pipeline."
>+             low_watermark="10"
>+             high_watermark="30"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="3D Pipe/Vertex Shader"
>+             mdapi_usage_flags="Tier3 Indicate Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Clipper0 Bottleneck"
>+             symbol_name="Cl0Bottleneck"
>+             underscore_name="cl0_bottleneck"
>+             description="The percentage of time in which clipper pipeline stage was slowing down the 3D pipeline."
>+             low_watermark="10"
>+             high_watermark="30"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 5 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="3D Pipe/Clipper"
>+             mdapi_usage_flags="Tier3 Indicate Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="HS0 Bottleneck"
>+             symbol_name="Hs0Bottleneck"
>+             underscore_name="hs0_bottleneck"
>+             description="The percentage of time in which hull shader pipeline stage was slowing down the 3D pipeline."
>+             low_watermark="3"
>+             high_watermark="9"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 4 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="3D Pipe/Hull Shader"
>+             mdapi_usage_flags="Tier3 Indicate Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="DS0 Bottleneck"
>+             symbol_name="Ds0Bottleneck"
>+             underscore_name="ds0_bottleneck"
>+             description="The percentage of time in which domain shader pipeline stage was slowing down the 3D pipeline."
>+             low_watermark="5"
>+             high_watermark="15"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 3 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="3D Pipe/Domain Shader"
>+             mdapi_usage_flags="Tier3 Indicate Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GS0 Bottleneck"
>+             symbol_name="Gs0Bottleneck"
>+             underscore_name="gs0_bottleneck"
>+             description="The percentage of time in which geometry shader pipeline stage was slowing down the 3D pipeline."
>+             low_watermark="5"
>+             high_watermark="15"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 2 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="3D Pipe/Geometry Shader"
>+             mdapi_usage_flags="Tier3 Indicate Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="SF0 Bottleneck"
>+             symbol_name="Sf0Bottleneck"
>+             underscore_name="sf0_bottleneck"
>+             description="The percentage of time in which strip-fans pipeline stage was slowing down the 3D pipeline."
>+             low_watermark="5"
>+             high_watermark="10"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 1 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="3D Pipe/Rasterizer/Strip-Fans"
>+             mdapi_usage_flags="Tier3 Indicate Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="SO0 Bottleneck"
>+             symbol_name="So0Bottleneck"
>+             underscore_name="so0_bottleneck"
>+             description="The percentage of time in which stream output pipeline stage was slowing down the 3D pipeline."
>+             low_watermark="5"
>+             high_watermark="15"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="3D Pipe/Stream Output"
>+             mdapi_usage_flags="Tier3 Indicate Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="CL0 Stall"
>+             symbol_name="Cl0Stall"
>+             underscore_name="cl0_stall"
>+             description="The percentage of time in which clipper pipeline stage was stalled."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 7 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="3D Pipe/Clipper"
>+             mdapi_usage_flags="Tier4 Correlate Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="HS0 Stall"
>+             symbol_name="Hs0Stall"
>+             underscore_name="hs0_stall"
>+             description="The percentage of time in which hull stall pipeline stage was stalled."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 6 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="3D Pipe/Hull Shader"
>+             mdapi_usage_flags="Tier4 Correlate Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="DS0 Stall"
>+             symbol_name="Ds0Stall"
>+             underscore_name="ds0_stall"
>+             description="The percentage of time in which domain shader pipeline stage was stalled."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 5 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="3D Pipe/Domain Shader"
>+             mdapi_usage_flags="Tier4 Correlate Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="SF0 Stall"
>+             symbol_name="Sf0Stall"
>+             underscore_name="sf0_stall"
>+             description="The percentage of time in which strip-fans pipeline stage was stalled."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 4 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="3D Pipe/Rasterizer/Strip-Fans"
>+             mdapi_usage_flags="Tier4 Correlate Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="SO0 Stall"
>+             symbol_name="So0Stall"
>+             underscore_name="so0_stall"
>+             description="The percentage of time in which stream-output pipeline stage was stalled."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 3 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="3D Pipe/Stream Output"
>+             mdapi_usage_flags="Tier4 Correlate Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Hi-Depth Bottleneck"
>+             symbol_name="HiDepthBottleneck"
>+             underscore_name="hi_depth_bottleneck"
>+             description="The percentage of time in which early hierarchical depth test pipeline stage was slowing down the 3D pipeline."
>+             low_watermark="5"
>+             high_watermark="15"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 2 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="3D Pipe/Rasterizer/Hi-Depth Test"
>+             mdapi_usage_flags="Tier3 Indicate Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="BC Bottleneck"
>+             symbol_name="BcBottleneck"
>+             underscore_name="bc_bottleneck"
>+             description="The percentage of time in which barycentric coordinates calculation pipeline stage was slowing down the 3D pipeline."
>+             low_watermark="5"
>+             high_watermark="15"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="3D Pipe/Rasterizer/Barycentric Calc"
>+             mdapi_usage_flags="Tier3 Indicate Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <register_config type="NOA">
>+        <register type="NOA" address="0x00000D04" value="0x00000200" />
>+        <register type="NOA" address="0x00009884" value="0x0000000C" />
>+        <register type="NOA" address="0x00009888" value="0x168607C0" />
>+        <register type="NOA" address="0x00009888" value="0x1603001F" />
>+        <register type="NOA" address="0x00009888" value="0x0C030000" />
>+        <register type="NOA" address="0x00009888" value="0x12040640" />
>+        <register type="NOA" address="0x00009888" value="0x02040000" />
>+        <register type="NOA" address="0x00009888" value="0x1C070000" />
>+        <register type="NOA" address="0x00009888" value="0x16090000" />
>+        <register type="NOA" address="0x00009888" value="0x1C0903C0" />
>+        <register type="NOA" address="0x00009888" value="0x160A01F0" />
>+        <register type="NOA" address="0x00009888" value="0x160B7C00" />
>+        <register type="NOA" address="0x00009888" value="0x04860000" />
>+        <register type="NOA" address="0x00009888" value="0x06860400" />
>+        <register type="NOA" address="0x00009888" value="0x00860000" />
>+        <register type="NOA" address="0x00009888" value="0x02030002" />
>+        <register type="NOA" address="0x00009888" value="0x04030010" />
>+        <register type="NOA" address="0x00009888" value="0x0A0301C0" />
>+        <register type="NOA" address="0x00009888" value="0x100300C0" />
>+        <register type="NOA" address="0x00009888" value="0x12030C80" />
>+        <register type="NOA" address="0x00009888" value="0x00030000" />
>+        <register type="NOA" address="0x00009888" value="0x00044000" />
>+        <register type="NOA" address="0x00009888" value="0x08040021" />
>+        <register type="NOA" address="0x00009888" value="0x0C040080" />
>+        <register type="NOA" address="0x00009888" value="0x0E040022" />
>+        <register type="NOA" address="0x00009888" value="0x04074000" />
>+        <register type="NOA" address="0x00009888" value="0x0A070021" />
>+        <register type="NOA" address="0x00009888" value="0x0E0708C0" />
>+        <register type="NOA" address="0x00009888" value="0x00070800" />
>+        <register type="NOA" address="0x00009888" value="0x020700A0" />
>+        <register type="NOA" address="0x00009888" value="0x00083032" />
>+        <register type="NOA" address="0x00009888" value="0x02083300" />
>+        <register type="NOA" address="0x00009888" value="0x02094003" />
>+        <register type="NOA" address="0x00009888" value="0x08090880" />
>+        <register type="NOA" address="0x00009888" value="0x0C090001" />
>+        <register type="NOA" address="0x00009888" value="0x0E090023" />
>+        <register type="NOA" address="0x00009888" value="0x12090003" />
>+        <register type="NOA" address="0x00009888" value="0x00090D28" />
>+        <register type="NOA" address="0x00009888" value="0x040A0200" />
>+        <register type="NOA" address="0x00009888" value="0x140A0940" />
>+        <register type="NOA" address="0x00009888" value="0x000A9A91" />
>+        <register type="NOA" address="0x00009888" value="0x020A19AA" />
>+        <register type="NOA" address="0x00009888" value="0x00015044" />
>+        <register type="NOA" address="0x00009888" value="0x040B0004" />
>+        <register type="NOA" address="0x00009888" value="0x080B0000" />
>+        <register type="NOA" address="0x00009888" value="0x000B2000" />
>+        <register type="NOA" address="0x00009888" value="0x020B2200" />
>+        <register type="NOA" address="0x00009888" value="0x00020C20" />
>+        <register type="NOA" address="0x00009888" value="0x000C0800" />
>+        <register type="NOA" address="0x00009888" value="0x020C00A0" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x0008AAA2" />
>+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
>+        <register type="NOA" address="0x00009888" value="0x00032482" />
>+        <register type="NOA" address="0x00009888" value="0x02032492" />
>+        <register type="NOA" address="0x00009888" value="0x04032492" />
>+        <register type="NOA" address="0x00009888" value="0x06030002" />
>+        <register type="NOA" address="0x00009888" value="0x14030000" />
>+        <register type="NOA" address="0x00009888" value="0x3E030007" />
>+        <register type="NOA" address="0x00009888" value="0x2C030000" />
>+        <register type="NOA" address="0x00009888" value="0x4203180A" />
>+        <register type="NOA" address="0x00009888" value="0x32031000" />
>+        <register type="NOA" address="0x00009888" value="0x440300C2" />
>+        <register type="NOA" address="0x00009888" value="0x38031000" />
>+        <register type="NOA" address="0x00009888" value="0x46030200" />
>+        <register type="NOA" address="0x00009888" value="0x48030261" />
>+        <register type="NOA" address="0x00009888" value="0x4A030C22" />
>+        <register type="NOA" address="0x00009888" value="0x16030000" />
>+        <register type="NOA" address="0x00009888" value="0x18030000" />
>+        <register type="NOA" address="0x00009888" value="0x1A030000" />
>+        <register type="NOA" address="0x00009888" value="0x400304C1" />
>+        <register type="NOA" address="0x00009888" value="0x1C031000" />
>+        <register type="NOA" address="0x00009888" value="0x1E030000" />
>+        <register type="NOA" address="0x00009888" value="0x20031000" />
>+        <register type="NOA" address="0x00009884" value="0x00000008" />
>+        <register type="NOA" address="0x00009888" value="0x0001FFFD" />
>+        <register type="NOA" address="0x00009888" value="0x00025551" />
>+        <register type="NOA" address="0x00009888" value="0x02025555" />
>+        <register type="NOA" address="0x00009888" value="0x00045551" />
>+        <register type="NOA" address="0x00009888" value="0x02045555" />
>+        <register type="NOA" address="0x00009888" value="0x00055551" />
>+        <register type="NOA" address="0x00009888" value="0x02055555" />
>+        <register type="NOA" address="0x00009888" value="0x0008AAA2" />
>+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x5C030002" />
>+    </register_config>
>+    <register_config type="OA">
>+        <register type="OA" address="0x0000D920" value="0x00000000" />
>+        <register type="OA" address="0x0000D900" value="0x00000000" />
>+        <register type="OA" address="0x0000D904" value="0x00800000" />
>+        <register type="OA" address="0x0000D910" value="0x00000000" />
>+        <register type="OA" address="0x0000D914" value="0x00800000" />
>+        <register type="OA" address="0x0000DC40" value="0x00000000" />
>+    </register_config>
>+    <register_config type="FLEX">
>+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
>+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
>+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
>+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
>+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
>+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
>+    </register_config>
>+  </set>
>+
>+  <set name="Metric set HDCAndSF1"
>+       chipset="DG2"
>+       symbol_name="HDCAndSF1"
>+       underscore_name="hdc_and_sf1"
>+       mdapi_supported_apis="VK OGL OCL IO"
>+       hw_config_guid="1afb0edc-585c-4be9-8c0c-bc72325106b0"
>+       >
>+    <counter name="GPU Time Elapsed"
>+             symbol_name="GpuTime"
>+             underscore_name="gpu_time"
>+             description="Time elapsed on the GPU during the measurement."
>+             data_type="uint64"
>+             units="ns"
>+             semantic_type="duration"
>+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Core Clocks"
>+             symbol_name="GpuCoreClocks"
>+             underscore_name="gpu_core_clocks"
>+             description="The total number of GPU core clocks elapsed during the measurement."
>+             data_type="uint64"
>+             units="cycles"
>+             semantic_type="event"
>+             equation="GPU_CLOCK 0 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="AVG GPU Core Frequency"
>+             symbol_name="AvgGpuCoreFrequency"
>+             underscore_name="avg_gpu_core_frequency"
>+             description="Average GPU Core Frequency in the measurement."
>+             data_type="uint64"
>+             max_equation="$GpuMaxFrequency"
>+             units="hz"
>+             semantic_type="event"
>+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Busy"
>+             symbol_name="GpuBusy"
>+             underscore_name="gpu_busy"
>+             description="The percentage of time in which the GPU has been processing GPU commands."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="VS Threads Dispatched"
>+             symbol_name="VsThreads"
>+             underscore_name="vs_threads"
>+             description="The total number of vertex shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 1 READ"
>+             mdapi_group="EU Array/Vertex Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="HS Threads Dispatched"
>+             symbol_name="HsThreads"
>+             underscore_name="hs_threads"
>+             description="The total number of hull shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 2 READ"
>+             mdapi_group="EU Array/Hull Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="DS Threads Dispatched"
>+             symbol_name="DsThreads"
>+             underscore_name="ds_threads"
>+             description="The total number of domain shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 3 READ"
>+             mdapi_group="EU Array/Domain Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GS Threads Dispatched"
>+             symbol_name="GsThreads"
>+             underscore_name="gs_threads"
>+             description="The total number of geometry shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 5 READ"
>+             mdapi_group="EU Array/Geometry Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="FS Threads Dispatched"
>+             symbol_name="PsThreads"
>+             underscore_name="ps_threads"
>+             description="The total number of fragment shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 6 READ"
>+             mdapi_group="EU Array/Fragment Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="CS Threads Dispatched"
>+             symbol_name="CsThreads"
>+             underscore_name="cs_threads"
>+             description="The total number of compute shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 4 READ"
>+             mdapi_group="EU Array/Compute Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Active"
>+             symbol_name="EuActive"
>+             underscore_name="eu_active"
>+             description="The percentage of time in which the Execution Units were actively processing."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Stall"
>+             symbol_name="EuStall"
>+             underscore_name="eu_stall"
>+             description="The percentage of time in which the Execution Units were stalled."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Thread Occupancy"
>+             symbol_name="EuThreadOccupancy"
>+             underscore_name="eu_thread_occupancy"
>+             description="The percentage of time in which hardware threads occupied EUs."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Rasterized Pixels"
>+             symbol_name="RasterizedPixels"
>+             underscore_name="rasterized_pixels"
>+             description="The total number of rasterized pixels."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 21 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Rasterizer"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Early Hi-Depth Test Fails"
>+             symbol_name="HiDepthTestFails"
>+             underscore_name="hi_depth_test_fails"
>+             description="The total number of pixels dropped on early hierarchical depth test."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 22 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Rasterizer/Hi-Depth Test"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Early Depth Test Fails"
>+             symbol_name="EarlyDepthTestFails"
>+             underscore_name="early_depth_test_fails"
>+             description="The total number of pixels dropped on early depth test."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 23 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Rasterizer/Early Depth Test"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Samples Killed in FS"
>+             symbol_name="SamplesKilledInPs"
>+             underscore_name="samples_killed_in_ps"
>+             description="The total number of samples or pixels dropped in fragment shaders."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 24 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Fragment Shader"
>+             mdapi_usage_flags="Tier4 Overview Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Pixels Failing Tests"
>+             symbol_name="PixelsFailingPostPsTests"
>+             underscore_name="pixels_failing_post_ps_tests"
>+             description="The total number of pixels dropped on post-FS alpha, stencil, or depth tests."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 25 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Output Merger"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Samples Written"
>+             symbol_name="SamplesWritten"
>+             underscore_name="samples_written"
>+             description="The total number of samples or pixels written to all render targets."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 26 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Output Merger"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Samples Blended"
>+             symbol_name="SamplesBlended"
>+             underscore_name="samples_blended"
>+             description="The total number of blended samples or pixels written to all render targets."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 27 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Output Merger"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Sampler Texels"
>+             symbol_name="SamplerTexels"
>+             underscore_name="sampler_texels"
>+             description="The total number of texels seen on input (with 2x2 accuracy) in all sampler units."
>+             data_type="uint64"
>+             units="texels"
>+             semantic_type="event"
>+             equation="A 28 READ 4 UMUL"
>+             mdapi_group="Sampler/Sampler Input"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Sampler Texels Misses"
>+             symbol_name="SamplerTexelMisses"
>+             underscore_name="sampler_texel_misses"
>+             description="The total number of texels lookups (with 2x2 accuracy) that missed L1 sampler cache."
>+             data_type="uint64"
>+             units="texels"
>+             semantic_type="event"
>+             equation="A 29 READ 4 UMUL"
>+             mdapi_group="Sampler/Sampler Cache"
>+             mdapi_usage_flags="Tier3 Batch Frame Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="SLM Bytes Read"
>+             symbol_name="SlmBytesRead"
>+             underscore_name="slm_bytes_read"
>+             description="The total number of GPU memory bytes read from shared local memory."
>+             data_type="uint64"
>+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
>+             units="bytes"
>+             semantic_type="throughput"
>+             equation="A 30 READ 64 UMUL"
>+             mdapi_group="L3/Data Port/SLM"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="SLM Bytes Written"
>+             symbol_name="SlmBytesWritten"
>+             underscore_name="slm_bytes_written"
>+             description="The total number of GPU memory bytes written into shared local memory."
>+             data_type="uint64"
>+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
>+             units="bytes"
>+             semantic_type="throughput"
>+             equation="A 31 READ 64 UMUL"
>+             mdapi_group="L3/Data Port/SLM"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Shader Memory Accesses"
>+             symbol_name="ShaderMemoryAccesses"
>+             underscore_name="shader_memory_accesses"
>+             description="The total number of shader memory accesses to L3."
>+             data_type="uint64"
>+             units="messages"
>+             semantic_type="event"
>+             equation="A 32 READ"
>+             mdapi_group="L3/Data Port"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Shader Atomic Memory Accesses"
>+             symbol_name="ShaderAtomics"
>+             underscore_name="shader_atomics"
>+             description="The total number of shader atomic memory accesses."
>+             data_type="uint64"
>+             units="messages"
>+             semantic_type="event"
>+             equation="A 34 READ"
>+             mdapi_group="L3/Data Port/Atomics"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="L3 Shader Throughput"
>+             symbol_name="L3ShaderThroughput"
>+             underscore_name="l3_shader_throughput"
>+             description="The total number of GPU memory bytes transferred between shaders and L3 caches w/o URB."
>+             data_type="uint64"
>+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
>+             units="bytes"
>+             semantic_type="throughput"
>+             equation="$ShaderMemoryAccesses 64 UMUL"
>+             mdapi_group="L3/Data Port"
>+             mdapi_usage_flags="Tier2 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Shader Barrier Messages"
>+             symbol_name="ShaderBarriers"
>+             underscore_name="shader_barriers"
>+             description="The total number of shader barrier messages."
>+             data_type="uint64"
>+             units="messages"
>+             semantic_type="event"
>+             equation="A 35 READ"
>+             mdapi_group="EU Array/Barrier"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Slice0 Dualsubslice0 Non-sampler Shader Access Stalled On L3"
>+             symbol_name="NonSamplerShader00AccessStalledOnL3"
>+             underscore_name="non_sampler_shader00_access_stalled_on_l3"
>+             description="Percentage of time when HDC has messages to L3, but it's stalled due to lack of credits (Slice0 Dualsubslice0)"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 1 AND"
>+             mdapi_group="GPU/Data Port"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="subslice"
>+             />
>+    <counter name="Slice0 Dualsubslice1 Non-sampler Shader Access Stalled On L3"
>+             symbol_name="NonSamplerShader01AccessStalledOnL3"
>+             underscore_name="non_sampler_shader01_access_stalled_on_l3"
>+             description="Percentage of time when HDC has messages to L3, but it's stalled due to lack of credits (Slice0 Dualsubslice1)"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 2 AND"
>+             mdapi_group="GPU/Data Port"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="subslice"
>+             />
>+    <counter name="Slice0 Dualsubslice2 Non-sampler Shader Access Stalled On L3"
>+             symbol_name="NonSamplerShader02AccessStalledOnL3"
>+             underscore_name="non_sampler_shader02_access_stalled_on_l3"
>+             description="Percentage of time when HDC has messages to L3, but it's stalled due to lack of credits (Slice0 Dualsubslice2)"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 2 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 4 AND"
>+             mdapi_group="GPU/Data Port"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="subslice"
>+             />
>+    <counter name="Slice0 Dualsubslice3 Non-sampler Shader Access Stalled On L3"
>+             symbol_name="NonSamplerShader03AccessStalledOnL3"
>+             underscore_name="non_sampler_shader03_access_stalled_on_l3"
>+             description="Percentage of time when HDC has messages to L3, but it's stalled due to lack of credits (Slice0 Dualsubslice3)"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 3 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 8 AND"
>+             mdapi_group="GPU/Data Port"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="subslice"
>+             />
>+    <counter name="Slice1 Dualsubslice0 Non-sampler Shader Access Stalled On L3"
>+             symbol_name="NonSamplerShader10AccessStalledOnL3"
>+             underscore_name="non_sampler_shader10_access_stalled_on_l3"
>+             description="Percentage of time when HDC has messages to L3, but it's stalled due to lack of credits (Slice1 Dualsubslice0)"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 4 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 64 AND"
>+             mdapi_group="GPU/Data Port"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="subslice"
>+             />
>+    <counter name="Slice1 Dualsubslice1 Non-sampler Shader Access Stalled On L3"
>+             symbol_name="NonSamplerShader11AccessStalledOnL3"
>+             underscore_name="non_sampler_shader11_access_stalled_on_l3"
>+             description="Percentage of time when HDC has messages to L3, but it's stalled due to lack of credits (Slice1 Dualsubslice1)"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 5 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 128 AND"
>+             mdapi_group="GPU/Data Port"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="subslice"
>+             />
>+    <counter name="Slice1 Dualsubslice2 Non-sampler Shader Access Stalled On L3"
>+             symbol_name="NonSamplerShader12AccessStalledOnL3"
>+             underscore_name="non_sampler_shader12_access_stalled_on_l3"
>+             description="Percentage of time when HDC has messages to L3, but it's stalled due to lack of credits (Slice1 Dualsubslice2)"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 6 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 256 AND"
>+             mdapi_group="GPU/Data Port"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="subslice"
>+             />
>+    <counter name="Slice1 Dualsubslice3 Non-sampler Shader Access Stalled On L3"
>+             symbol_name="NonSamplerShader13AccessStalledOnL3"
>+             underscore_name="non_sampler_shader13_access_stalled_on_l3"
>+             description="Percentage of time when HDC has messages to L3, but it's stalled due to lack of credits (Slice1 Dualsubslice3)"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 7 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 512 AND"
>+             mdapi_group="GPU/Data Port"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="subslice"
>+             />
>+    <counter name="SQ00 is full"
>+             symbol_name="GTRequestQueue00Full"
>+             underscore_name="gt_request_queue00_full"
>+             description="The percentage of time when IDI0 SQ00 is filled above a threshold (usually 48 entries)"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GTI"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="SQ10 is full"
>+             symbol_name="GTRequestQueue10Full"
>+             underscore_name="gt_request_queue10_full"
>+             description="The percentage of time when IDI0 SQ10 is filled above a threshold (usually 48 entries)"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GTI"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="SQ20 is full"
>+             symbol_name="GTRequestQueue20Full"
>+             underscore_name="gt_request_queue20_full"
>+             description="The percentage of time when IDI0 SQ20 is filled above a threshold (usually 48 entries)"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 5 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GTI"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="SQ30 is full"
>+             symbol_name="GTRequestQueue30Full"
>+             underscore_name="gt_request_queue30_full"
>+             description="The percentage of time when IDI0 SQ30 is filled above a threshold (usually 48 entries)"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 4 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GTI"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="SQ40 is full"
>+             symbol_name="GTRequestQueue40Full"
>+             underscore_name="gt_request_queue40_full"
>+             description="The percentage of time when IDI0 SQ40 is filled above a threshold (usually 48 entries)"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 3 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GTI"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="SQ50 is full"
>+             symbol_name="GTRequestQueue50Full"
>+             underscore_name="gt_request_queue50_full"
>+             description="The percentage of time when IDI0 SQ50 is filled above a threshold (usually 48 entries)"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 2 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GTI"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="SQ60 is full"
>+             symbol_name="GTRequestQueue60Full"
>+             underscore_name="gt_request_queue60_full"
>+             description="The percentage of time when IDI0 SQ60 is filled above a threshold (usually 48 entries)"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 1 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GTI"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="SQ70 is full"
>+             symbol_name="GTRequestQueue70Full"
>+             underscore_name="gt_request_queue70_full"
>+             description="The percentage of time when IDI0 SQ70 is filled above a threshold (usually 48 entries)"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GTI"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <register_config type="NOA">
>+        <register type="NOA" address="0x00000D04" value="0x00000200" />
>+        <register type="NOA" address="0x00009884" value="0x0000000C" />
>+        <register type="NOA" address="0x00009888" value="0x1E810200" />
>+        <register type="NOA" address="0x00009888" value="0x1EA10200" />
>+        <register type="NOA" address="0x00009888" value="0x1EC10200" />
>+        <register type="NOA" address="0x00009888" value="0x1EE10200" />
>+        <register type="NOA" address="0x00009888" value="0x1F810200" />
>+        <register type="NOA" address="0x00009888" value="0x1FA10200" />
>+        <register type="NOA" address="0x00009888" value="0x1FC10200" />
>+        <register type="NOA" address="0x00009888" value="0x1FE10200" />
>+        <register type="NOA" address="0x00009888" value="0x04810400" />
>+        <register type="NOA" address="0x00009888" value="0x08810015" />
>+        <register type="NOA" address="0x00009888" value="0x00810000" />
>+        <register type="NOA" address="0x00009888" value="0x00830002" />
>+        <register type="NOA" address="0x00009888" value="0x00840001" />
>+        <register type="NOA" address="0x00009888" value="0x00860001" />
>+        <register type="NOA" address="0x00009888" value="0x00870001" />
>+        <register type="NOA" address="0x00009888" value="0x04A10400" />
>+        <register type="NOA" address="0x00009888" value="0x08A10540" />
>+        <register type="NOA" address="0x00009888" value="0x00A10000" />
>+        <register type="NOA" address="0x00009888" value="0x00A30008" />
>+        <register type="NOA" address="0x00009888" value="0x00A40002" />
>+        <register type="NOA" address="0x00009888" value="0x00A60004" />
>+        <register type="NOA" address="0x00009888" value="0x00A70004" />
>+        <register type="NOA" address="0x00009888" value="0x04C10400" />
>+        <register type="NOA" address="0x00009888" value="0x0AC10015" />
>+        <register type="NOA" address="0x00009888" value="0x08C10000" />
>+        <register type="NOA" address="0x00009888" value="0x00C10000" />
>+        <register type="NOA" address="0x00009888" value="0x00C30020" />
>+        <register type="NOA" address="0x00009888" value="0x00C40004" />
>+        <register type="NOA" address="0x00009888" value="0x00C60010" />
>+        <register type="NOA" address="0x00009888" value="0x00C70010" />
>+        <register type="NOA" address="0x00009888" value="0x04E10400" />
>+        <register type="NOA" address="0x00009888" value="0x0AE10540" />
>+        <register type="NOA" address="0x00009888" value="0x08E10000" />
>+        <register type="NOA" address="0x00009888" value="0x00E10000" />
>+        <register type="NOA" address="0x00009888" value="0x00E30080" />
>+        <register type="NOA" address="0x00009888" value="0x00E40008" />
>+        <register type="NOA" address="0x00009888" value="0x00E60040" />
>+        <register type="NOA" address="0x00009888" value="0x00E70040" />
>+        <register type="NOA" address="0x00009888" value="0x05810400" />
>+        <register type="NOA" address="0x00009888" value="0x0D810015" />
>+        <register type="NOA" address="0x00009888" value="0x09810000" />
>+        <register type="NOA" address="0x00009888" value="0x01810000" />
>+        <register type="NOA" address="0x00009888" value="0x01830200" />
>+        <register type="NOA" address="0x00009888" value="0x01840010" />
>+        <register type="NOA" address="0x00009888" value="0x01860100" />
>+        <register type="NOA" address="0x00009888" value="0x01870100" />
>+        <register type="NOA" address="0x00009888" value="0x05A10400" />
>+        <register type="NOA" address="0x00009888" value="0x0DA10540" />
>+        <register type="NOA" address="0x00009888" value="0x09A10000" />
>+        <register type="NOA" address="0x00009888" value="0x01A10000" />
>+        <register type="NOA" address="0x00009888" value="0x01A30800" />
>+        <register type="NOA" address="0x00009888" value="0x01A40020" />
>+        <register type="NOA" address="0x00009888" value="0x01A60400" />
>+        <register type="NOA" address="0x00009888" value="0x01A70400" />
>+        <register type="NOA" address="0x00009888" value="0x05C10400" />
>+        <register type="NOA" address="0x00009888" value="0x0FC10015" />
>+        <register type="NOA" address="0x00009888" value="0x09C10000" />
>+        <register type="NOA" address="0x00009888" value="0x01C10000" />
>+        <register type="NOA" address="0x00009888" value="0x01C32000" />
>+        <register type="NOA" address="0x00009888" value="0x01C40040" />
>+        <register type="NOA" address="0x00009888" value="0x01C61000" />
>+        <register type="NOA" address="0x00009888" value="0x01C71000" />
>+        <register type="NOA" address="0x00009888" value="0x05E10400" />
>+        <register type="NOA" address="0x00009888" value="0x0FE10540" />
>+        <register type="NOA" address="0x00009888" value="0x09E10000" />
>+        <register type="NOA" address="0x00009888" value="0x01E10000" />
>+        <register type="NOA" address="0x00009888" value="0x01E38000" />
>+        <register type="NOA" address="0x00009888" value="0x01E40080" />
>+        <register type="NOA" address="0x00009888" value="0x01E64000" />
>+        <register type="NOA" address="0x00009888" value="0x01E74000" />
>+        <register type="NOA" address="0x00009888" value="0x000300F0" />
>+        <register type="NOA" address="0x00009888" value="0x00070060" />
>+        <register type="NOA" address="0x00009888" value="0x0008FF0E" />
>+        <register type="NOA" address="0x00009888" value="0x0009000C" />
>+        <register type="NOA" address="0x00009888" value="0x000A55A5" />
>+        <register type="NOA" address="0x00009888" value="0x000100F2" />
>+        <register type="NOA" address="0x00009888" value="0x000BAA04" />
>+        <register type="NOA" address="0x00009888" value="0x00020004" />
>+        <register type="NOA" address="0x00009888" value="0x000C0010" />
>+        <register type="NOA" address="0x00009888" value="0x01076000" />
>+        <register type="NOA" address="0x00009888" value="0x01080E00" />
>+        <register type="NOA" address="0x00009888" value="0x010900C0" />
>+        <register type="NOA" address="0x00009888" value="0x010AA500" />
>+        <register type="NOA" address="0x00009888" value="0x01010020" />
>+        <register type="NOA" address="0x00009888" value="0x010B0400" />
>+        <register type="NOA" address="0x00009888" value="0x01020040" />
>+        <register type="NOA" address="0x00009888" value="0x010C1000" />
>+        <register type="NOA" address="0x00009884" value="0x00000001" />
>+        <register type="NOA" address="0x00009888" value="0x0C017000" />
>+        <register type="NOA" address="0x00009888" value="0x0C217000" />
>+        <register type="NOA" address="0x00009888" value="0x0C417000" />
>+        <register type="NOA" address="0x00009888" value="0x0C617000" />
>+        <register type="NOA" address="0x00009888" value="0x0C817000" />
>+        <register type="NOA" address="0x00009888" value="0x0CA17000" />
>+        <register type="NOA" address="0x00009888" value="0x0CC17000" />
>+        <register type="NOA" address="0x00009888" value="0x0CE17000" />
>+        <register type="NOA" address="0x00009888" value="0x02010010" />
>+        <register type="NOA" address="0x00009888" value="0x16010006" />
>+        <register type="NOA" address="0x00009888" value="0x00010000" />
>+        <register type="NOA" address="0x00009888" value="0x02210010" />
>+        <register type="NOA" address="0x00009888" value="0x14210180" />
>+        <register type="NOA" address="0x00009888" value="0x00210000" />
>+        <register type="NOA" address="0x00009888" value="0x02410010" />
>+        <register type="NOA" address="0x00009888" value="0x14410006" />
>+        <register type="NOA" address="0x00009888" value="0x00410000" />
>+        <register type="NOA" address="0x00009888" value="0x02610010" />
>+        <register type="NOA" address="0x00009888" value="0x12610180" />
>+        <register type="NOA" address="0x00009888" value="0x00610000" />
>+        <register type="NOA" address="0x00009888" value="0x02810010" />
>+        <register type="NOA" address="0x00009888" value="0x12810006" />
>+        <register type="NOA" address="0x00009888" value="0x00810000" />
>+        <register type="NOA" address="0x00009888" value="0x02A10010" />
>+        <register type="NOA" address="0x00009888" value="0x10A10180" />
>+        <register type="NOA" address="0x00009888" value="0x00A10000" />
>+        <register type="NOA" address="0x00009888" value="0x02C10010" />
>+        <register type="NOA" address="0x00009888" value="0x10C10006" />
>+        <register type="NOA" address="0x00009888" value="0x00C10000" />
>+        <register type="NOA" address="0x00009888" value="0x02E10010" />
>+        <register type="NOA" address="0x00009888" value="0x0EE10180" />
>+        <register type="NOA" address="0x00009888" value="0x00E10000" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
>+        <register type="NOA" address="0x00009888" value="0x00032492" />
>+        <register type="NOA" address="0x00009888" value="0x02034892" />
>+        <register type="NOA" address="0x00009888" value="0x04034924" />
>+        <register type="NOA" address="0x00009888" value="0x06030004" />
>+        <register type="NOA" address="0x00009888" value="0x02093800" />
>+        <register type="NOA" address="0x00009888" value="0x0409000A" />
>+        <register type="NOA" address="0x00009888" value="0x020A1200" />
>+        <register type="NOA" address="0x00009888" value="0x040A2489" />
>+        <register type="NOA" address="0x00009888" value="0x060A0002" />
>+        <register type="NOA" address="0x00009888" value="0x040B2700" />
>+        <register type="NOA" address="0x00009888" value="0x060B0001" />
>+        <register type="NOA" address="0x00009888" value="0x14037000" />
>+        <register type="NOA" address="0x00009888" value="0x3E0314CA" />
>+        <register type="NOA" address="0x00009888" value="0x26037000" />
>+        <register type="NOA" address="0x00009888" value="0x42032004" />
>+        <register type="NOA" address="0x00009888" value="0x2C033000" />
>+        <register type="NOA" address="0x00009888" value="0x32037000" />
>+        <register type="NOA" address="0x00009888" value="0x44030047" />
>+        <register type="NOA" address="0x00009888" value="0x38031000" />
>+        <register type="NOA" address="0x00009888" value="0x46030003" />
>+        <register type="NOA" address="0x00009888" value="0x48031EA0" />
>+        <register type="NOA" address="0x00009888" value="0x4A030DE6" />
>+        <register type="NOA" address="0x00009888" value="0x1603F000" />
>+        <register type="NOA" address="0x00009888" value="0x1803F000" />
>+        <register type="NOA" address="0x00009888" value="0x1A03F000" />
>+        <register type="NOA" address="0x00009888" value="0x40031084" />
>+        <register type="NOA" address="0x00009888" value="0x1C03F000" />
>+        <register type="NOA" address="0x00009888" value="0x1E03F000" />
>+        <register type="NOA" address="0x00009888" value="0x2003F000" />
>+        <register type="NOA" address="0x00009884" value="0x00000008" />
>+        <register type="NOA" address="0x00009888" value="0x000100FF" />
>+        <register type="NOA" address="0x00009888" value="0x00025555" />
>+        <register type="NOA" address="0x00009888" value="0x00045555" />
>+        <register type="NOA" address="0x00009888" value="0x00055555" />
>+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x5C030002" />
>+    </register_config>
>+    <register_config type="OA">
>+        <register type="OA" address="0x0000D920" value="0x00000000" />
>+        <register type="OA" address="0x0000D900" value="0x00000000" />
>+        <register type="OA" address="0x0000D904" value="0xF0800000" />
>+        <register type="OA" address="0x0000D910" value="0x00000000" />
>+        <register type="OA" address="0x0000D914" value="0xF0800000" />
>+        <register type="OA" address="0x0000DC40" value="0x00FF0000" />
>+        <register type="OA" address="0x0000D940" value="0x00000002" />
>+        <register type="OA" address="0x0000D944" value="0x0000FFFE" />
>+        <register type="OA" address="0x0000DC00" value="0x00000002" />
>+        <register type="OA" address="0x0000DC04" value="0x0000FFFE" />
>+        <register type="OA" address="0x0000D948" value="0x00000002" />
>+        <register type="OA" address="0x0000D94C" value="0x0000FFFD" />
>+        <register type="OA" address="0x0000DC08" value="0x00000002" />
>+        <register type="OA" address="0x0000DC0C" value="0x0000FFFD" />
>+        <register type="OA" address="0x0000D950" value="0x00000002" />
>+        <register type="OA" address="0x0000D954" value="0x0000FFFB" />
>+        <register type="OA" address="0x0000DC10" value="0x00000002" />
>+        <register type="OA" address="0x0000DC14" value="0x0000FFFB" />
>+        <register type="OA" address="0x0000D958" value="0x00000002" />
>+        <register type="OA" address="0x0000D95C" value="0x0000FFF7" />
>+        <register type="OA" address="0x0000DC18" value="0x00000002" />
>+        <register type="OA" address="0x0000DC1C" value="0x0000FFF7" />
>+        <register type="OA" address="0x0000D960" value="0x00000002" />
>+        <register type="OA" address="0x0000D964" value="0x0000FFEF" />
>+        <register type="OA" address="0x0000DC20" value="0x00000002" />
>+        <register type="OA" address="0x0000DC24" value="0x0000FFEF" />
>+        <register type="OA" address="0x0000D968" value="0x00000002" />
>+        <register type="OA" address="0x0000D96C" value="0x0000FFDF" />
>+        <register type="OA" address="0x0000DC28" value="0x00000002" />
>+        <register type="OA" address="0x0000DC2C" value="0x0000FFDF" />
>+        <register type="OA" address="0x0000D970" value="0x00000002" />
>+        <register type="OA" address="0x0000D974" value="0x0000FFBF" />
>+        <register type="OA" address="0x0000DC30" value="0x00000002" />
>+        <register type="OA" address="0x0000DC34" value="0x0000FFBF" />
>+        <register type="OA" address="0x0000D978" value="0x00000002" />
>+        <register type="OA" address="0x0000D97C" value="0x0000FF7F" />
>+        <register type="OA" address="0x0000DC38" value="0x00000002" />
>+        <register type="OA" address="0x0000DC3C" value="0x0000FF7F" />
>+    </register_config>
>+    <register_config type="FLEX">
>+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
>+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
>+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
>+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
>+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
>+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
>+    </register_config>
>+  </set>
>+
>+  <set name="Metric set HDCAndSF2"
>+       chipset="DG2"
>+       symbol_name="HDCAndSF2"
>+       underscore_name="hdc_and_sf2"
>+       mdapi_supported_apis="VK OGL OCL IO"
>+       hw_config_guid="a3b35902-f293-41fe-97d0-3d895bc5fab3"
>+       >
>+    <counter name="GPU Time Elapsed"
>+             symbol_name="GpuTime"
>+             underscore_name="gpu_time"
>+             description="Time elapsed on the GPU during the measurement."
>+             data_type="uint64"
>+             units="ns"
>+             semantic_type="duration"
>+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Core Clocks"
>+             symbol_name="GpuCoreClocks"
>+             underscore_name="gpu_core_clocks"
>+             description="The total number of GPU core clocks elapsed during the measurement."
>+             data_type="uint64"
>+             units="cycles"
>+             semantic_type="event"
>+             equation="GPU_CLOCK 0 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="AVG GPU Core Frequency"
>+             symbol_name="AvgGpuCoreFrequency"
>+             underscore_name="avg_gpu_core_frequency"
>+             description="Average GPU Core Frequency in the measurement."
>+             data_type="uint64"
>+             max_equation="$GpuMaxFrequency"
>+             units="hz"
>+             semantic_type="event"
>+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Busy"
>+             symbol_name="GpuBusy"
>+             underscore_name="gpu_busy"
>+             description="The percentage of time in which the GPU has been processing GPU commands."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="VS Threads Dispatched"
>+             symbol_name="VsThreads"
>+             underscore_name="vs_threads"
>+             description="The total number of vertex shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 1 READ"
>+             mdapi_group="EU Array/Vertex Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="HS Threads Dispatched"
>+             symbol_name="HsThreads"
>+             underscore_name="hs_threads"
>+             description="The total number of hull shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 2 READ"
>+             mdapi_group="EU Array/Hull Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="DS Threads Dispatched"
>+             symbol_name="DsThreads"
>+             underscore_name="ds_threads"
>+             description="The total number of domain shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 3 READ"
>+             mdapi_group="EU Array/Domain Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GS Threads Dispatched"
>+             symbol_name="GsThreads"
>+             underscore_name="gs_threads"
>+             description="The total number of geometry shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 5 READ"
>+             mdapi_group="EU Array/Geometry Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="FS Threads Dispatched"
>+             symbol_name="PsThreads"
>+             underscore_name="ps_threads"
>+             description="The total number of fragment shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 6 READ"
>+             mdapi_group="EU Array/Fragment Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="CS Threads Dispatched"
>+             symbol_name="CsThreads"
>+             underscore_name="cs_threads"
>+             description="The total number of compute shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 4 READ"
>+             mdapi_group="EU Array/Compute Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Active"
>+             symbol_name="EuActive"
>+             underscore_name="eu_active"
>+             description="The percentage of time in which the Execution Units were actively processing."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Stall"
>+             symbol_name="EuStall"
>+             underscore_name="eu_stall"
>+             description="The percentage of time in which the Execution Units were stalled."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Thread Occupancy"
>+             symbol_name="EuThreadOccupancy"
>+             underscore_name="eu_thread_occupancy"
>+             description="The percentage of time in which hardware threads occupied EUs."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Rasterized Pixels"
>+             symbol_name="RasterizedPixels"
>+             underscore_name="rasterized_pixels"
>+             description="The total number of rasterized pixels."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 21 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Rasterizer"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Early Hi-Depth Test Fails"
>+             symbol_name="HiDepthTestFails"
>+             underscore_name="hi_depth_test_fails"
>+             description="The total number of pixels dropped on early hierarchical depth test."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 22 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Rasterizer/Hi-Depth Test"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Early Depth Test Fails"
>+             symbol_name="EarlyDepthTestFails"
>+             underscore_name="early_depth_test_fails"
>+             description="The total number of pixels dropped on early depth test."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 23 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Rasterizer/Early Depth Test"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Samples Killed in FS"
>+             symbol_name="SamplesKilledInPs"
>+             underscore_name="samples_killed_in_ps"
>+             description="The total number of samples or pixels dropped in fragment shaders."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 24 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Fragment Shader"
>+             mdapi_usage_flags="Tier4 Overview Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Pixels Failing Tests"
>+             symbol_name="PixelsFailingPostPsTests"
>+             underscore_name="pixels_failing_post_ps_tests"
>+             description="The total number of pixels dropped on post-FS alpha, stencil, or depth tests."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 25 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Output Merger"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Samples Written"
>+             symbol_name="SamplesWritten"
>+             underscore_name="samples_written"
>+             description="The total number of samples or pixels written to all render targets."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 26 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Output Merger"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Samples Blended"
>+             symbol_name="SamplesBlended"
>+             underscore_name="samples_blended"
>+             description="The total number of blended samples or pixels written to all render targets."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 27 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Output Merger"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Sampler Texels"
>+             symbol_name="SamplerTexels"
>+             underscore_name="sampler_texels"
>+             description="The total number of texels seen on input (with 2x2 accuracy) in all sampler units."
>+             data_type="uint64"
>+             units="texels"
>+             semantic_type="event"
>+             equation="A 28 READ 4 UMUL"
>+             mdapi_group="Sampler/Sampler Input"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Sampler Texels Misses"
>+             symbol_name="SamplerTexelMisses"
>+             underscore_name="sampler_texel_misses"
>+             description="The total number of texels lookups (with 2x2 accuracy) that missed L1 sampler cache."
>+             data_type="uint64"
>+             units="texels"
>+             semantic_type="event"
>+             equation="A 29 READ 4 UMUL"
>+             mdapi_group="Sampler/Sampler Cache"
>+             mdapi_usage_flags="Tier3 Batch Frame Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="SLM Bytes Read"
>+             symbol_name="SlmBytesRead"
>+             underscore_name="slm_bytes_read"
>+             description="The total number of GPU memory bytes read from shared local memory."
>+             data_type="uint64"
>+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
>+             units="bytes"
>+             semantic_type="throughput"
>+             equation="A 30 READ 64 UMUL"
>+             mdapi_group="L3/Data Port/SLM"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="SLM Bytes Written"
>+             symbol_name="SlmBytesWritten"
>+             underscore_name="slm_bytes_written"
>+             description="The total number of GPU memory bytes written into shared local memory."
>+             data_type="uint64"
>+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
>+             units="bytes"
>+             semantic_type="throughput"
>+             equation="A 31 READ 64 UMUL"
>+             mdapi_group="L3/Data Port/SLM"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Shader Memory Accesses"
>+             symbol_name="ShaderMemoryAccesses"
>+             underscore_name="shader_memory_accesses"
>+             description="The total number of shader memory accesses to L3."
>+             data_type="uint64"
>+             units="messages"
>+             semantic_type="event"
>+             equation="A 32 READ"
>+             mdapi_group="L3/Data Port"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Shader Atomic Memory Accesses"
>+             symbol_name="ShaderAtomics"
>+             underscore_name="shader_atomics"
>+             description="The total number of shader atomic memory accesses."
>+             data_type="uint64"
>+             units="messages"
>+             semantic_type="event"
>+             equation="A 34 READ"
>+             mdapi_group="L3/Data Port/Atomics"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="L3 Shader Throughput"
>+             symbol_name="L3ShaderThroughput"
>+             underscore_name="l3_shader_throughput"
>+             description="The total number of GPU memory bytes transferred between shaders and L3 caches w/o URB."
>+             data_type="uint64"
>+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
>+             units="bytes"
>+             semantic_type="throughput"
>+             equation="$ShaderMemoryAccesses 64 UMUL"
>+             mdapi_group="L3/Data Port"
>+             mdapi_usage_flags="Tier2 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Shader Barrier Messages"
>+             symbol_name="ShaderBarriers"
>+             underscore_name="shader_barriers"
>+             description="The total number of shader barrier messages."
>+             data_type="uint64"
>+             units="messages"
>+             semantic_type="event"
>+             equation="A 35 READ"
>+             mdapi_group="EU Array/Barrier"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Polygon (Slice0) Data Ready"
>+             symbol_name="Poly0DataReady"
>+             underscore_name="poly0_data_ready"
>+             description="The percentage of time in which geometry pipeline output is ready"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU/3D Pipe/Strip-Fans"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Polygon (Slice1) Data Ready"
>+             symbol_name="Poly1DataReady"
>+             underscore_name="poly1_data_ready"
>+             description="The percentage of time in which geometry pipeline output is ready"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU/3D Pipe/Strip-Fans"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <register_config type="NOA">
>+        <register type="NOA" address="0x00000D04" value="0x00000200" />
>+        <register type="NOA" address="0x00009884" value="0x0000000C" />
>+        <register type="NOA" address="0x00009888" value="0x12056400" />
>+        <register type="NOA" address="0x00009888" value="0x13056400" />
>+        <register type="NOA" address="0x00009888" value="0x00030001" />
>+        <register type="NOA" address="0x00009888" value="0x00050002" />
>+        <register type="NOA" address="0x00009888" value="0x02050700" />
>+        <register type="NOA" address="0x00009888" value="0x00060002" />
>+        <register type="NOA" address="0x00009888" value="0x00080007" />
>+        <register type="NOA" address="0x00009888" value="0x000A0005" />
>+        <register type="NOA" address="0x00009888" value="0x00010001" />
>+        <register type="NOA" address="0x00009888" value="0x000B0002" />
>+        <register type="NOA" address="0x00009888" value="0x01050002" />
>+        <register type="NOA" address="0x00009888" value="0x0305001C" />
>+        <register type="NOA" address="0x00009888" value="0x01060001" />
>+        <register type="NOA" address="0x00009888" value="0x01080001" />
>+        <register type="NOA" address="0x00009888" value="0x010A0001" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x0008000A" />
>+        <register type="NOA" address="0x00009888" value="0x00030012" />
>+        <register type="NOA" address="0x00009888" value="0x14030000" />
>+        <register type="NOA" address="0x00009888" value="0x3E030000" />
>+        <register type="NOA" address="0x00009888" value="0x26033000" />
>+        <register type="NOA" address="0x00009888" value="0x42030100" />
>+        <register type="NOA" address="0x00009888" value="0x02030000" />
>+        <register type="NOA" address="0x00009884" value="0x00000008" />
>+        <register type="NOA" address="0x00009888" value="0x00010003" />
>+        <register type="NOA" address="0x00009888" value="0x00020005" />
>+        <register type="NOA" address="0x00009888" value="0x00040005" />
>+        <register type="NOA" address="0x00009888" value="0x00050005" />
>+        <register type="NOA" address="0x00009888" value="0x0008000A" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x5C030002" />
>+    </register_config>
>+    <register_config type="OA">
>+        <register type="OA" address="0x0000D920" value="0x00000000" />
>+        <register type="OA" address="0x0000D900" value="0x00000000" />
>+        <register type="OA" address="0x0000D904" value="0x30800000" />
>+        <register type="OA" address="0x0000D910" value="0x00000000" />
>+        <register type="OA" address="0x0000D914" value="0x00800000" />
>+        <register type="OA" address="0x0000DC40" value="0x00030000" />
>+        <register type="OA" address="0x0000D940" value="0x00000002" />
>+        <register type="OA" address="0x0000D944" value="0x0000FFFE" />
>+        <register type="OA" address="0x0000DC00" value="0x00000002" />
>+        <register type="OA" address="0x0000DC04" value="0x0000FFFE" />
>+        <register type="OA" address="0x0000D948" value="0x00000002" />
>+        <register type="OA" address="0x0000D94C" value="0x0000FFFD" />
>+        <register type="OA" address="0x0000DC08" value="0x00000002" />
>+        <register type="OA" address="0x0000DC0C" value="0x0000FFFD" />
>+    </register_config>
>+    <register_config type="FLEX">
>+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
>+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
>+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
>+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
>+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
>+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
>+    </register_config>
>+  </set>
>+
>+  <set name="Metric set RasterizerAndPixelBackend1"
>+       chipset="DG2"
>+       symbol_name="RasterizerAndPixelBackend1"
>+       underscore_name="rasterizer_and_pixel_backend1"
>+       mdapi_supported_apis="VK OGL OCL IO"
>+       hw_config_guid="b04e9460-630f-4682-b0ed-ebb0e74a6c9d"
>+       >
>+    <counter name="GPU Time Elapsed"
>+             symbol_name="GpuTime"
>+             underscore_name="gpu_time"
>+             description="Time elapsed on the GPU during the measurement."
>+             data_type="uint64"
>+             units="ns"
>+             semantic_type="duration"
>+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Core Clocks"
>+             symbol_name="GpuCoreClocks"
>+             underscore_name="gpu_core_clocks"
>+             description="The total number of GPU core clocks elapsed during the measurement."
>+             data_type="uint64"
>+             units="cycles"
>+             semantic_type="event"
>+             equation="GPU_CLOCK 0 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="AVG GPU Core Frequency"
>+             symbol_name="AvgGpuCoreFrequency"
>+             underscore_name="avg_gpu_core_frequency"
>+             description="Average GPU Core Frequency in the measurement."
>+             data_type="uint64"
>+             max_equation="$GpuMaxFrequency"
>+             units="hz"
>+             semantic_type="event"
>+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Busy"
>+             symbol_name="GpuBusy"
>+             underscore_name="gpu_busy"
>+             description="The percentage of time in which the GPU has been processing GPU commands."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="VS Threads Dispatched"
>+             symbol_name="VsThreads"
>+             underscore_name="vs_threads"
>+             description="The total number of vertex shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 1 READ"
>+             mdapi_group="EU Array/Vertex Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="HS Threads Dispatched"
>+             symbol_name="HsThreads"
>+             underscore_name="hs_threads"
>+             description="The total number of hull shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 2 READ"
>+             mdapi_group="EU Array/Hull Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="DS Threads Dispatched"
>+             symbol_name="DsThreads"
>+             underscore_name="ds_threads"
>+             description="The total number of domain shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 3 READ"
>+             mdapi_group="EU Array/Domain Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GS Threads Dispatched"
>+             symbol_name="GsThreads"
>+             underscore_name="gs_threads"
>+             description="The total number of geometry shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 5 READ"
>+             mdapi_group="EU Array/Geometry Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="FS Threads Dispatched"
>+             symbol_name="PsThreads"
>+             underscore_name="ps_threads"
>+             description="The total number of fragment shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 6 READ"
>+             mdapi_group="EU Array/Fragment Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="CS Threads Dispatched"
>+             symbol_name="CsThreads"
>+             underscore_name="cs_threads"
>+             description="The total number of compute shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 4 READ"
>+             mdapi_group="EU Array/Compute Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Active"
>+             symbol_name="EuActive"
>+             underscore_name="eu_active"
>+             description="The percentage of time in which the Execution Units were actively processing."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Stall"
>+             symbol_name="EuStall"
>+             underscore_name="eu_stall"
>+             description="The percentage of time in which the Execution Units were stalled."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Thread Occupancy"
>+             symbol_name="EuThreadOccupancy"
>+             underscore_name="eu_thread_occupancy"
>+             description="The percentage of time in which hardware threads occupied EUs."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Rasterized Pixels"
>+             symbol_name="RasterizedPixels"
>+             underscore_name="rasterized_pixels"
>+             description="The total number of rasterized pixels."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 21 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Rasterizer"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Early Hi-Depth Test Fails"
>+             symbol_name="HiDepthTestFails"
>+             underscore_name="hi_depth_test_fails"
>+             description="The total number of pixels dropped on early hierarchical depth test."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 22 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Rasterizer/Hi-Depth Test"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Early Depth Test Fails"
>+             symbol_name="EarlyDepthTestFails"
>+             underscore_name="early_depth_test_fails"
>+             description="The total number of pixels dropped on early depth test."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 23 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Rasterizer/Early Depth Test"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Samples Killed in FS"
>+             symbol_name="SamplesKilledInPs"
>+             underscore_name="samples_killed_in_ps"
>+             description="The total number of samples or pixels dropped in fragment shaders."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 24 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Fragment Shader"
>+             mdapi_usage_flags="Tier4 Overview Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Pixels Failing Tests"
>+             symbol_name="PixelsFailingPostPsTests"
>+             underscore_name="pixels_failing_post_ps_tests"
>+             description="The total number of pixels dropped on post-FS alpha, stencil, or depth tests."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 25 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Output Merger"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Samples Written"
>+             symbol_name="SamplesWritten"
>+             underscore_name="samples_written"
>+             description="The total number of samples or pixels written to all render targets."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 26 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Output Merger"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Samples Blended"
>+             symbol_name="SamplesBlended"
>+             underscore_name="samples_blended"
>+             description="The total number of blended samples or pixels written to all render targets."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 27 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Output Merger"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Sampler Texels"
>+             symbol_name="SamplerTexels"
>+             underscore_name="sampler_texels"
>+             description="The total number of texels seen on input (with 2x2 accuracy) in all sampler units."
>+             data_type="uint64"
>+             units="texels"
>+             semantic_type="event"
>+             equation="A 28 READ 4 UMUL"
>+             mdapi_group="Sampler/Sampler Input"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Sampler Texels Misses"
>+             symbol_name="SamplerTexelMisses"
>+             underscore_name="sampler_texel_misses"
>+             description="The total number of texels lookups (with 2x2 accuracy) that missed L1 sampler cache."
>+             data_type="uint64"
>+             units="texels"
>+             semantic_type="event"
>+             equation="A 29 READ 4 UMUL"
>+             mdapi_group="Sampler/Sampler Cache"
>+             mdapi_usage_flags="Tier3 Batch Frame Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="SLM Bytes Read"
>+             symbol_name="SlmBytesRead"
>+             underscore_name="slm_bytes_read"
>+             description="The total number of GPU memory bytes read from shared local memory."
>+             data_type="uint64"
>+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
>+             units="bytes"
>+             semantic_type="throughput"
>+             equation="A 30 READ 64 UMUL"
>+             mdapi_group="L3/Data Port/SLM"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="SLM Bytes Written"
>+             symbol_name="SlmBytesWritten"
>+             underscore_name="slm_bytes_written"
>+             description="The total number of GPU memory bytes written into shared local memory."
>+             data_type="uint64"
>+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
>+             units="bytes"
>+             semantic_type="throughput"
>+             equation="A 31 READ 64 UMUL"
>+             mdapi_group="L3/Data Port/SLM"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Shader Memory Accesses"
>+             symbol_name="ShaderMemoryAccesses"
>+             underscore_name="shader_memory_accesses"
>+             description="The total number of shader memory accesses to L3."
>+             data_type="uint64"
>+             units="messages"
>+             semantic_type="event"
>+             equation="A 32 READ"
>+             mdapi_group="L3/Data Port"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Shader Atomic Memory Accesses"
>+             symbol_name="ShaderAtomics"
>+             underscore_name="shader_atomics"
>+             description="The total number of shader atomic memory accesses."
>+             data_type="uint64"
>+             units="messages"
>+             semantic_type="event"
>+             equation="A 34 READ"
>+             mdapi_group="L3/Data Port/Atomics"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="L3 Shader Throughput"
>+             symbol_name="L3ShaderThroughput"
>+             underscore_name="l3_shader_throughput"
>+             description="The total number of GPU memory bytes transferred between shaders and L3 caches w/o URB."
>+             data_type="uint64"
>+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
>+             units="bytes"
>+             semantic_type="throughput"
>+             equation="$ShaderMemoryAccesses 64 UMUL"
>+             mdapi_group="L3/Data Port"
>+             mdapi_usage_flags="Tier2 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Shader Barrier Messages"
>+             symbol_name="ShaderBarriers"
>+             underscore_name="shader_barriers"
>+             description="The total number of shader barrier messages."
>+             data_type="uint64"
>+             units="messages"
>+             semantic_type="event"
>+             equation="A 35 READ"
>+             mdapi_group="EU Array/Barrier"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Slice0 Rasterizer Input Available"
>+             symbol_name="Rasterizer0InputAvailable"
>+             underscore_name="rasterizer0_input_available"
>+             description="The percentage of time in which slice0 rasterizer input is available"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$SliceMask 1 AND"
>+             mdapi_group="GPU/Rasterizer"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="slice"
>+             />
>+    <counter name="Slice1 Rasterizer Output Ready"
>+             symbol_name="Rasterizer1OutputReady"
>+             underscore_name="rasterizer1_output_ready"
>+             description="The percentage of time in which slice1 rasterizer output is ready"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$SliceMask 2 AND"
>+             mdapi_group="GPU/Rasterizer"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="slice"
>+             />
>+    <counter name="Slice0 Pipe0 Post-EarlyZ Pixel Data Ready"
>+             symbol_name="PixelData00Ready"
>+             underscore_name="pixel_data00_ready"
>+             description="The percentage of time in which slice0  pipe0 post-EarlyZ pixel data is ready (after early Z tests have been applied)"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU/Rasterizer/Early Depth Test"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="slice"
>+             />
>+    <counter name="Slice0 Pipe1 Post-EarlyZ Pixel Data Ready"
>+             symbol_name="PixelData01Ready"
>+             underscore_name="pixel_data01_ready"
>+             description="The percentage of time in which slice0  pipe1 post-EarlyZ pixel data is ready (after early Z tests have been applied)"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 2 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU/Rasterizer/Early Depth Test"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="slice"
>+             />
>+    <register_config type="NOA">
>+        <register type="NOA" address="0x00000D04" value="0x00000200" />
>+        <register type="NOA" address="0x00009884" value="0x0000000C" />
>+        <register type="NOA" address="0x00009888" value="0x1A056000" />
>+        <register type="NOA" address="0x00009888" value="0x1C050000" />
>+        <register type="NOA" address="0x00009888" value="0x0C0B7000" />
>+        <register type="NOA" address="0x00009888" value="0x0E0B6000" />
>+        <register type="NOA" address="0x00009888" value="0x0C0C701F" />
>+        <register type="NOA" address="0x00009888" value="0x0E0C6000" />
>+        <register type="NOA" address="0x00009888" value="0x1B057000" />
>+        <register type="NOA" address="0x00009888" value="0x00038000" />
>+        <register type="NOA" address="0x00009888" value="0x00050000" />
>+        <register type="NOA" address="0x00009888" value="0x020535D2" />
>+        <register type="NOA" address="0x00009888" value="0x0405003E" />
>+        <register type="NOA" address="0x00009888" value="0x00060007" />
>+        <register type="NOA" address="0x00009888" value="0x00072800" />
>+        <register type="NOA" address="0x00009888" value="0x000803D5" />
>+        <register type="NOA" address="0x00009888" value="0x0208C000" />
>+        <register type="NOA" address="0x00009888" value="0x00090060" />
>+        <register type="NOA" address="0x00009888" value="0x000A2955" />
>+        <register type="NOA" address="0x00009888" value="0x020A4000" />
>+        <register type="NOA" address="0x00009888" value="0x00018018" />
>+        <register type="NOA" address="0x00009888" value="0x040B0200" />
>+        <register type="NOA" address="0x00009888" value="0x080B0400" />
>+        <register type="NOA" address="0x00009888" value="0x0A0B001F" />
>+        <register type="NOA" address="0x00009888" value="0x000B0000" />
>+        <register type="NOA" address="0x00009888" value="0x020B8000" />
>+        <register type="NOA" address="0x00009888" value="0x00020060" />
>+        <register type="NOA" address="0x00009888" value="0x040C0200" />
>+        <register type="NOA" address="0x00009888" value="0x0A0C0400" />
>+        <register type="NOA" address="0x00009888" value="0x080C0000" />
>+        <register type="NOA" address="0x00009888" value="0x000C0000" />
>+        <register type="NOA" address="0x00009888" value="0x01050000" />
>+        <register type="NOA" address="0x00009888" value="0x11050DC0" />
>+        <register type="NOA" address="0x00009888" value="0x03050000" />
>+        <register type="NOA" address="0x00009888" value="0x01068000" />
>+        <register type="NOA" address="0x00009888" value="0x03084000" />
>+        <register type="NOA" address="0x00009888" value="0x030A4000" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x00082AAA" />
>+        <register type="NOA" address="0x00009888" value="0x02088000" />
>+        <register type="NOA" address="0x00009888" value="0x00032492" />
>+        <register type="NOA" address="0x00009888" value="0x02030012" />
>+        <register type="NOA" address="0x00009888" value="0x06030002" />
>+        <register type="NOA" address="0x00009888" value="0x14033000" />
>+        <register type="NOA" address="0x00009888" value="0x3E030008" />
>+        <register type="NOA" address="0x00009888" value="0x26033000" />
>+        <register type="NOA" address="0x00009888" value="0x42032100" />
>+        <register type="NOA" address="0x00009888" value="0x2C033000" />
>+        <register type="NOA" address="0x00009888" value="0x32033000" />
>+        <register type="NOA" address="0x00009888" value="0x44030CA5" />
>+        <register type="NOA" address="0x00009888" value="0x38033000" />
>+        <register type="NOA" address="0x00009888" value="0x46030462" />
>+        <register type="NOA" address="0x00009888" value="0x20030000" />
>+        <register type="NOA" address="0x00009884" value="0x00000008" />
>+        <register type="NOA" address="0x00009888" value="0x0001807F" />
>+        <register type="NOA" address="0x00009888" value="0x00021555" />
>+        <register type="NOA" address="0x00009888" value="0x02024000" />
>+        <register type="NOA" address="0x00009888" value="0x00041555" />
>+        <register type="NOA" address="0x00009888" value="0x02044000" />
>+        <register type="NOA" address="0x00009888" value="0x00051555" />
>+        <register type="NOA" address="0x00009888" value="0x02054000" />
>+        <register type="NOA" address="0x00009888" value="0x00082AAA" />
>+        <register type="NOA" address="0x00009888" value="0x02088000" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x5C030002" />
>+    </register_config>
>+    <register_config type="OA">
>+        <register type="OA" address="0x0000D920" value="0x00000000" />
>+        <register type="OA" address="0x0000D900" value="0x00000000" />
>+        <register type="OA" address="0x0000D904" value="0x70800000" />
>+        <register type="OA" address="0x0000D910" value="0x00000000" />
>+        <register type="OA" address="0x0000D914" value="0x00800000" />
>+        <register type="OA" address="0x0000DC40" value="0x00070000" />
>+        <register type="OA" address="0x0000D940" value="0x00000038" />
>+        <register type="OA" address="0x0000D944" value="0x0000FFF8" />
>+        <register type="OA" address="0x0000DC00" value="0x00000038" />
>+        <register type="OA" address="0x0000DC04" value="0x0000FFF8" />
>+        <register type="OA" address="0x0000D948" value="0x000000C0" />
>+        <register type="OA" address="0x0000D94C" value="0x0000FFE7" />
>+        <register type="OA" address="0x0000DC08" value="0x000000C0" />
>+        <register type="OA" address="0x0000DC0C" value="0x0000FFE7" />
>+        <register type="OA" address="0x0000D950" value="0x00000300" />
>+        <register type="OA" address="0x0000D954" value="0x0000FF9F" />
>+        <register type="OA" address="0x0000DC10" value="0x00000300" />
>+        <register type="OA" address="0x0000DC14" value="0x0000FF9F" />
>+    </register_config>
>+    <register_config type="FLEX">
>+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
>+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
>+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
>+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
>+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
>+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
>+    </register_config>
>+  </set>
>+
>+  <set name="Metric set RasterizerAndPixelBackend2"
>+       chipset="DG2"
>+       symbol_name="RasterizerAndPixelBackend2"
>+       underscore_name="rasterizer_and_pixel_backend2"
>+       mdapi_supported_apis="VK OGL OCL IO"
>+       hw_config_guid="57a0abf4-8d81-4856-ab86-01f15f92c81c"
>+       >
>+    <counter name="GPU Time Elapsed"
>+             symbol_name="GpuTime"
>+             underscore_name="gpu_time"
>+             description="Time elapsed on the GPU during the measurement."
>+             data_type="uint64"
>+             units="ns"
>+             semantic_type="duration"
>+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Core Clocks"
>+             symbol_name="GpuCoreClocks"
>+             underscore_name="gpu_core_clocks"
>+             description="The total number of GPU core clocks elapsed during the measurement."
>+             data_type="uint64"
>+             units="cycles"
>+             semantic_type="event"
>+             equation="GPU_CLOCK 0 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="AVG GPU Core Frequency"
>+             symbol_name="AvgGpuCoreFrequency"
>+             underscore_name="avg_gpu_core_frequency"
>+             description="Average GPU Core Frequency in the measurement."
>+             data_type="uint64"
>+             max_equation="$GpuMaxFrequency"
>+             units="hz"
>+             semantic_type="event"
>+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Busy"
>+             symbol_name="GpuBusy"
>+             underscore_name="gpu_busy"
>+             description="The percentage of time in which the GPU has been processing GPU commands."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="VS Threads Dispatched"
>+             symbol_name="VsThreads"
>+             underscore_name="vs_threads"
>+             description="The total number of vertex shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 1 READ"
>+             mdapi_group="EU Array/Vertex Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="HS Threads Dispatched"
>+             symbol_name="HsThreads"
>+             underscore_name="hs_threads"
>+             description="The total number of hull shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 2 READ"
>+             mdapi_group="EU Array/Hull Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="DS Threads Dispatched"
>+             symbol_name="DsThreads"
>+             underscore_name="ds_threads"
>+             description="The total number of domain shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 3 READ"
>+             mdapi_group="EU Array/Domain Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GS Threads Dispatched"
>+             symbol_name="GsThreads"
>+             underscore_name="gs_threads"
>+             description="The total number of geometry shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 5 READ"
>+             mdapi_group="EU Array/Geometry Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="FS Threads Dispatched"
>+             symbol_name="PsThreads"
>+             underscore_name="ps_threads"
>+             description="The total number of fragment shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 6 READ"
>+             mdapi_group="EU Array/Fragment Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="CS Threads Dispatched"
>+             symbol_name="CsThreads"
>+             underscore_name="cs_threads"
>+             description="The total number of compute shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 4 READ"
>+             mdapi_group="EU Array/Compute Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Active"
>+             symbol_name="EuActive"
>+             underscore_name="eu_active"
>+             description="The percentage of time in which the Execution Units were actively processing."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Stall"
>+             symbol_name="EuStall"
>+             underscore_name="eu_stall"
>+             description="The percentage of time in which the Execution Units were stalled."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Thread Occupancy"
>+             symbol_name="EuThreadOccupancy"
>+             underscore_name="eu_thread_occupancy"
>+             description="The percentage of time in which hardware threads occupied EUs."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Rasterized Pixels"
>+             symbol_name="RasterizedPixels"
>+             underscore_name="rasterized_pixels"
>+             description="The total number of rasterized pixels."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 21 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Rasterizer"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Early Hi-Depth Test Fails"
>+             symbol_name="HiDepthTestFails"
>+             underscore_name="hi_depth_test_fails"
>+             description="The total number of pixels dropped on early hierarchical depth test."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 22 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Rasterizer/Hi-Depth Test"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Early Depth Test Fails"
>+             symbol_name="EarlyDepthTestFails"
>+             underscore_name="early_depth_test_fails"
>+             description="The total number of pixels dropped on early depth test."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 23 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Rasterizer/Early Depth Test"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Samples Killed in FS"
>+             symbol_name="SamplesKilledInPs"
>+             underscore_name="samples_killed_in_ps"
>+             description="The total number of samples or pixels dropped in fragment shaders."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 24 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Fragment Shader"
>+             mdapi_usage_flags="Tier4 Overview Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Pixels Failing Tests"
>+             symbol_name="PixelsFailingPostPsTests"
>+             underscore_name="pixels_failing_post_ps_tests"
>+             description="The total number of pixels dropped on post-FS alpha, stencil, or depth tests."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 25 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Output Merger"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Samples Written"
>+             symbol_name="SamplesWritten"
>+             underscore_name="samples_written"
>+             description="The total number of samples or pixels written to all render targets."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 26 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Output Merger"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Samples Blended"
>+             symbol_name="SamplesBlended"
>+             underscore_name="samples_blended"
>+             description="The total number of blended samples or pixels written to all render targets."
>+             data_type="uint64"
>+             units="pixels"
>+             semantic_type="event"
>+             equation="A 27 READ 4 UMUL"
>+             mdapi_group="3D Pipe/Output Merger"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Sampler Texels"
>+             symbol_name="SamplerTexels"
>+             underscore_name="sampler_texels"
>+             description="The total number of texels seen on input (with 2x2 accuracy) in all sampler units."
>+             data_type="uint64"
>+             units="texels"
>+             semantic_type="event"
>+             equation="A 28 READ 4 UMUL"
>+             mdapi_group="Sampler/Sampler Input"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Sampler Texels Misses"
>+             symbol_name="SamplerTexelMisses"
>+             underscore_name="sampler_texel_misses"
>+             description="The total number of texels lookups (with 2x2 accuracy) that missed L1 sampler cache."
>+             data_type="uint64"
>+             units="texels"
>+             semantic_type="event"
>+             equation="A 29 READ 4 UMUL"
>+             mdapi_group="Sampler/Sampler Cache"
>+             mdapi_usage_flags="Tier3 Batch Frame Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="SLM Bytes Read"
>+             symbol_name="SlmBytesRead"
>+             underscore_name="slm_bytes_read"
>+             description="The total number of GPU memory bytes read from shared local memory."
>+             data_type="uint64"
>+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
>+             units="bytes"
>+             semantic_type="throughput"
>+             equation="A 30 READ 64 UMUL"
>+             mdapi_group="L3/Data Port/SLM"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="SLM Bytes Written"
>+             symbol_name="SlmBytesWritten"
>+             underscore_name="slm_bytes_written"
>+             description="The total number of GPU memory bytes written into shared local memory."
>+             data_type="uint64"
>+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
>+             units="bytes"
>+             semantic_type="throughput"
>+             equation="A 31 READ 64 UMUL"
>+             mdapi_group="L3/Data Port/SLM"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Shader Memory Accesses"
>+             symbol_name="ShaderMemoryAccesses"
>+             underscore_name="shader_memory_accesses"
>+             description="The total number of shader memory accesses to L3."
>+             data_type="uint64"
>+             units="messages"
>+             semantic_type="event"
>+             equation="A 32 READ"
>+             mdapi_group="L3/Data Port"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Shader Atomic Memory Accesses"
>+             symbol_name="ShaderAtomics"
>+             underscore_name="shader_atomics"
>+             description="The total number of shader atomic memory accesses."
>+             data_type="uint64"
>+             units="messages"
>+             semantic_type="event"
>+             equation="A 34 READ"
>+             mdapi_group="L3/Data Port/Atomics"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="L3 Shader Throughput"
>+             symbol_name="L3ShaderThroughput"
>+             underscore_name="l3_shader_throughput"
>+             description="The total number of GPU memory bytes transferred between shaders and L3 caches w/o URB."
>+             data_type="uint64"
>+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
>+             units="bytes"
>+             semantic_type="throughput"
>+             equation="$ShaderMemoryAccesses 64 UMUL"
>+             mdapi_group="L3/Data Port"
>+             mdapi_usage_flags="Tier2 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Shader Barrier Messages"
>+             symbol_name="ShaderBarriers"
>+             underscore_name="shader_barriers"
>+             description="The total number of shader barrier messages."
>+             data_type="uint64"
>+             units="messages"
>+             semantic_type="event"
>+             equation="A 35 READ"
>+             mdapi_group="EU Array/Barrier"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Slice1 Rasterizer Input Available"
>+             symbol_name="Rasterizer1InputAvailable"
>+             underscore_name="rasterizer1_input_available"
>+             description="The percentage of time in which slice1 rasterizer input is available"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$SliceMask 2 AND"
>+             mdapi_group="GPU/Rasterizer"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="slice"
>+             />
>+    <counter name="Slice0 Rasterizer Output Ready"
>+             symbol_name="Rasterizer0OutputReady"
>+             underscore_name="rasterizer0_output_ready"
>+             description="The percentage of time in which slice0 rasterizer output is ready"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$SliceMask 1 AND"
>+             mdapi_group="GPU/Rasterizer"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="slice"
>+             />
>+    <counter name="Slice1 Pipe0 Post-EarlyZ Pixel Data Ready"
>+             symbol_name="PixelData10Ready"
>+             underscore_name="pixel_data10_ready"
>+             description="The percentage of time in which slice1  pipe0 post-EarlyZ pixel data is ready (after early Z tests have been applied)"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 2 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU/Rasterizer/Early Depth Test"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="slice"
>+             />
>+    <counter name="Slice1 Pipe1 Post-EarlyZ Pixel Data Ready"
>+             symbol_name="PixelData11Ready"
>+             underscore_name="pixel_data11_ready"
>+             description="The percentage of time in which slice1  pipe1 post-EarlyZ pixel data is ready (after early Z tests have been applied)"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU/Rasterizer/Early Depth Test"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="slice"
>+             />
>+    <register_config type="NOA">
>+        <register type="NOA" address="0x00000D04" value="0x00000200" />
>+        <register type="NOA" address="0x00009884" value="0x0000000C" />
>+        <register type="NOA" address="0x00009888" value="0x1A057000" />
>+        <register type="NOA" address="0x00009888" value="0x1B056000" />
>+        <register type="NOA" address="0x00009888" value="0x1D050000" />
>+        <register type="NOA" address="0x00009888" value="0x0D0B701F" />
>+        <register type="NOA" address="0x00009888" value="0x0F0B6000" />
>+        <register type="NOA" address="0x00009888" value="0x0D0C7000" />
>+        <register type="NOA" address="0x00009888" value="0x0F0C6000" />
>+        <register type="NOA" address="0x00009888" value="0x0003007F" />
>+        <register type="NOA" address="0x00009888" value="0x00050000" />
>+        <register type="NOA" address="0x00009888" value="0x10050DC0" />
>+        <register type="NOA" address="0x00009888" value="0x02050000" />
>+        <register type="NOA" address="0x00009888" value="0x00068000" />
>+        <register type="NOA" address="0x00009888" value="0x00083FFF" />
>+        <register type="NOA" address="0x00009888" value="0x02084000" />
>+        <register type="NOA" address="0x00009888" value="0x000A1555" />
>+        <register type="NOA" address="0x00009888" value="0x020A4000" />
>+        <register type="NOA" address="0x00009888" value="0x0001007F" />
>+        <register type="NOA" address="0x00009888" value="0x000B2AAA" />
>+        <register type="NOA" address="0x00009888" value="0x01050000" />
>+        <register type="NOA" address="0x00009888" value="0x030535D2" />
>+        <register type="NOA" address="0x00009888" value="0x0505003E" />
>+        <register type="NOA" address="0x00009888" value="0x01060007" />
>+        <register type="NOA" address="0x00009888" value="0x01070280" />
>+        <register type="NOA" address="0x00009888" value="0x01083C15" />
>+        <register type="NOA" address="0x00009888" value="0x01090018" />
>+        <register type="NOA" address="0x00009888" value="0x010A1695" />
>+        <register type="NOA" address="0x00009888" value="0x01010060" />
>+        <register type="NOA" address="0x00009888" value="0x050B0200" />
>+        <register type="NOA" address="0x00009888" value="0x0B0B0400" />
>+        <register type="NOA" address="0x00009888" value="0x090B0000" />
>+        <register type="NOA" address="0x00009888" value="0x010B0000" />
>+        <register type="NOA" address="0x00009888" value="0x01020018" />
>+        <register type="NOA" address="0x00009888" value="0x050C0200" />
>+        <register type="NOA" address="0x00009888" value="0x090C0400" />
>+        <register type="NOA" address="0x00009888" value="0x0B0C001F" />
>+        <register type="NOA" address="0x00009888" value="0x010C0000" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x00082AAA" />
>+        <register type="NOA" address="0x00009888" value="0x02088000" />
>+        <register type="NOA" address="0x00009888" value="0x00032492" />
>+        <register type="NOA" address="0x00009888" value="0x02030012" />
>+        <register type="NOA" address="0x00009888" value="0x06030002" />
>+        <register type="NOA" address="0x00009888" value="0x14030000" />
>+        <register type="NOA" address="0x00009888" value="0x3E030006" />
>+        <register type="NOA" address="0x00009888" value="0x26030000" />
>+        <register type="NOA" address="0x00009888" value="0x420318C3" />
>+        <register type="NOA" address="0x00009888" value="0x2C030000" />
>+        <register type="NOA" address="0x00009888" value="0x32030000" />
>+        <register type="NOA" address="0x00009888" value="0x44030000" />
>+        <register type="NOA" address="0x00009888" value="0x38030000" />
>+        <register type="NOA" address="0x00009888" value="0x46030603" />
>+        <register type="NOA" address="0x00009888" value="0x20037000" />
>+        <register type="NOA" address="0x00009884" value="0x00000008" />
>+        <register type="NOA" address="0x00009888" value="0x0001807F" />
>+        <register type="NOA" address="0x00009888" value="0x00021555" />
>+        <register type="NOA" address="0x00009888" value="0x02024000" />
>+        <register type="NOA" address="0x00009888" value="0x00041555" />
>+        <register type="NOA" address="0x00009888" value="0x02044000" />
>+        <register type="NOA" address="0x00009888" value="0x00051555" />
>+        <register type="NOA" address="0x00009888" value="0x02054000" />
>+        <register type="NOA" address="0x00009888" value="0x00082AAA" />
>+        <register type="NOA" address="0x00009888" value="0x02088000" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x5C030002" />
>+    </register_config>
>+    <register_config type="OA">
>+        <register type="OA" address="0x0000D920" value="0x00000000" />
>+        <register type="OA" address="0x0000D900" value="0x00000000" />
>+        <register type="OA" address="0x0000D904" value="0x70800000" />
>+        <register type="OA" address="0x0000D910" value="0x00000000" />
>+        <register type="OA" address="0x0000D914" value="0x00800000" />
>+        <register type="OA" address="0x0000DC40" value="0x00070000" />
>+        <register type="OA" address="0x0000D940" value="0x00000038" />
>+        <register type="OA" address="0x0000D944" value="0x0000FFF8" />
>+        <register type="OA" address="0x0000DC00" value="0x00000038" />
>+        <register type="OA" address="0x0000DC04" value="0x0000FFF8" />
>+        <register type="OA" address="0x0000D948" value="0x000000C0" />
>+        <register type="OA" address="0x0000D94C" value="0x0000FFE7" />
>+        <register type="OA" address="0x0000DC08" value="0x000000C0" />
>+        <register type="OA" address="0x0000DC0C" value="0x0000FFE7" />
>+        <register type="OA" address="0x0000D950" value="0x00000300" />
>+        <register type="OA" address="0x0000D954" value="0x0000FF9F" />
>+        <register type="OA" address="0x0000DC10" value="0x00000300" />
>+        <register type="OA" address="0x0000DC14" value="0x0000FF9F" />
>+    </register_config>
>+    <register_config type="FLEX">
>+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
>+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
>+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
>+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
>+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
>+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
>+    </register_config>
>+  </set>
>+
>+  <set name="Sampler_1"
>+       chipset="DG2"
>+       symbol_name="Sampler_1"
>+       underscore_name="sampler_1"
>+       mdapi_supported_apis="VK OGL OCL MEDIA IO"
>+       hw_config_guid="2314eaa3-164f-4004-9684-3f1d16e875f9"
>+       >
>+    <counter name="GPU Time Elapsed"
>+             symbol_name="GpuTime"
>+             underscore_name="gpu_time"
>+             description="Time elapsed on the GPU during the measurement."
>+             data_type="uint64"
>+             units="ns"
>+             semantic_type="duration"
>+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Core Clocks"
>+             symbol_name="GpuCoreClocks"
>+             underscore_name="gpu_core_clocks"
>+             description="The total number of GPU core clocks elapsed during the measurement."
>+             data_type="uint64"
>+             units="cycles"
>+             semantic_type="event"
>+             equation="GPU_CLOCK 0 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="AVG GPU Core Frequency"
>+             symbol_name="AvgGpuCoreFrequency"
>+             underscore_name="avg_gpu_core_frequency"
>+             description="Average GPU Core Frequency in the measurement."
>+             data_type="uint64"
>+             max_equation="$GpuMaxFrequency"
>+             units="hz"
>+             semantic_type="event"
>+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Busy"
>+             symbol_name="GpuBusy"
>+             underscore_name="gpu_busy"
>+             description="The percentage of time in which the GPU has been processing GPU commands."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="VS Threads Dispatched"
>+             symbol_name="VsThreads"
>+             underscore_name="vs_threads"
>+             description="The total number of vertex shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 1 READ"
>+             mdapi_group="EU Array/Vertex Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="HS Threads Dispatched"
>+             symbol_name="HsThreads"
>+             underscore_name="hs_threads"
>+             description="The total number of hull shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 2 READ"
>+             mdapi_group="EU Array/Hull Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="DS Threads Dispatched"
>+             symbol_name="DsThreads"
>+             underscore_name="ds_threads"
>+             description="The total number of domain shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 3 READ"
>+             mdapi_group="EU Array/Domain Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GS Threads Dispatched"
>+             symbol_name="GsThreads"
>+             underscore_name="gs_threads"
>+             description="The total number of geometry shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 5 READ"
>+             mdapi_group="EU Array/Geometry Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="FS Threads Dispatched"
>+             symbol_name="PsThreads"
>+             underscore_name="ps_threads"
>+             description="The total number of fragment shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 6 READ"
>+             mdapi_group="EU Array/Fragment Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="CS Threads Dispatched"
>+             symbol_name="CsThreads"
>+             underscore_name="cs_threads"
>+             description="The total number of compute shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 4 READ"
>+             mdapi_group="EU Array/Compute Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Active"
>+             symbol_name="EuActive"
>+             underscore_name="eu_active"
>+             description="The percentage of time in which the Execution Units were actively processing."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Stall"
>+             symbol_name="EuStall"
>+             underscore_name="eu_stall"
>+             description="The percentage of time in which the Execution Units were stalled."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Thread Occupancy"
>+             symbol_name="EuThreadOccupancy"
>+             underscore_name="eu_thread_occupancy"
>+             description="The percentage of time in which hardware threads occupied EUs."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Slice0 DualSubslice0 Input Available"
>+             symbol_name="Sampler00InputAvailable"
>+             underscore_name="sampler00_input_available"
>+             description="The percentage of time in which slice0 dualsubslice0 sampler input is available"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 1 AND"
>+             mdapi_group="GPU/Sampler"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Slice0 DualSubslice1 Input Available"
>+             symbol_name="Sampler01InputAvailable"
>+             underscore_name="sampler01_input_available"
>+             description="The percentage of time in which slice0 dualsubslice1 sampler input is available"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 2 AND"
>+             mdapi_group="GPU/Sampler"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Slice0 DualSubslice2 Input Available"
>+             symbol_name="Sampler02InputAvailable"
>+             underscore_name="sampler02_input_available"
>+             description="The percentage of time in which slice0 dualsubslice2 sampler input is available"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 5 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 4 AND"
>+             mdapi_group="GPU/Sampler"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Slice0 DualSubslice3 Input Available"
>+             symbol_name="Sampler03InputAvailable"
>+             underscore_name="sampler03_input_available"
>+             description="The percentage of time in which slice0 dualsubslice3 sampler input is available"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 4 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 8 AND"
>+             mdapi_group="GPU/Sampler"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Slice0 DualSubslice0 Sampler Output Ready"
>+             symbol_name="Sampler00OutputReady"
>+             underscore_name="sampler00_output_ready"
>+             description="The percentage of time in which slice0 dualsubslice0 sampler output is ready"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 3 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 1 AND"
>+             mdapi_group="GPU/Sampler"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="subslice"
>+             />
>+    <counter name="Slice0 DualSubslice1 Sampler Output Ready"
>+             symbol_name="Sampler01OutputReady"
>+             underscore_name="sampler01_output_ready"
>+             description="The percentage of time in which slice0 dualsubslice1 sampler output is ready"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 2 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 2 AND"
>+             mdapi_group="GPU/Sampler"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="subslice"
>+             />
>+    <counter name="Slice0 DualSubslice2 Sampler Output Ready"
>+             symbol_name="Sampler02OutputReady"
>+             underscore_name="sampler02_output_ready"
>+             description="The percentage of time in which slice0 dualsubslice2 sampler output is ready"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 1 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 4 AND"
>+             mdapi_group="GPU/Sampler"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="subslice"
>+             />
>+    <counter name="Slice0 DualSubslice3 Sampler Output Ready"
>+             symbol_name="Sampler03OutputReady"
>+             underscore_name="sampler03_output_ready"
>+             description="The percentage of time in which slice0 dualsubslice3 sampler output is ready"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 8 AND"
>+             mdapi_group="GPU/Sampler"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="subslice"
>+             />
>+    <register_config type="NOA">
>+        <register type="NOA" address="0x00000D04" value="0x00000200" />
>+        <register type="NOA" address="0x00009884" value="0x0000000C" />
>+        <register type="NOA" address="0x00009888" value="0x1C824800" />
>+        <register type="NOA" address="0x00009888" value="0x14832C00" />
>+        <register type="NOA" address="0x00009888" value="0x1CA24800" />
>+        <register type="NOA" address="0x00009888" value="0x14A32C00" />
>+        <register type="NOA" address="0x00009888" value="0x1CC24800" />
>+        <register type="NOA" address="0x00009888" value="0x14C32C00" />
>+        <register type="NOA" address="0x00009888" value="0x1CE24800" />
>+        <register type="NOA" address="0x00009888" value="0x14E32C00" />
>+        <register type="NOA" address="0x00009888" value="0x04824000" />
>+        <register type="NOA" address="0x00009888" value="0x16820027" />
>+        <register type="NOA" address="0x00009888" value="0x02820000" />
>+        <register type="NOA" address="0x00009888" value="0x0E830580" />
>+        <register type="NOA" address="0x00009888" value="0x04830000" />
>+        <register type="NOA" address="0x00009888" value="0x02834000" />
>+        <register type="NOA" address="0x00009888" value="0x00848800" />
>+        <register type="NOA" address="0x00009888" value="0x02864040" />
>+        <register type="NOA" address="0x00009888" value="0x04A24000" />
>+        <register type="NOA" address="0x00009888" value="0x14A209C0" />
>+        <register type="NOA" address="0x00009888" value="0x02A20000" />
>+        <register type="NOA" address="0x00009888" value="0x0EA30016" />
>+        <register type="NOA" address="0x00009888" value="0x04A30000" />
>+        <register type="NOA" address="0x00009888" value="0x02A31000" />
>+        <register type="NOA" address="0x00009888" value="0x00A44400" />
>+        <register type="NOA" address="0x00009888" value="0x02A61010" />
>+        <register type="NOA" address="0x00009888" value="0x04C24000" />
>+        <register type="NOA" address="0x00009888" value="0x14C20027" />
>+        <register type="NOA" address="0x00009888" value="0x02C20000" />
>+        <register type="NOA" address="0x00009888" value="0x0CC30580" />
>+        <register type="NOA" address="0x00009888" value="0x04C30000" />
>+        <register type="NOA" address="0x00009888" value="0x02C30400" />
>+        <register type="NOA" address="0x00009888" value="0x00C42200" />
>+        <register type="NOA" address="0x00009888" value="0x02C60404" />
>+        <register type="NOA" address="0x00009888" value="0x04E24000" />
>+        <register type="NOA" address="0x00009888" value="0x12E209C0" />
>+        <register type="NOA" address="0x00009888" value="0x02E20000" />
>+        <register type="NOA" address="0x00009888" value="0x0CE30016" />
>+        <register type="NOA" address="0x00009888" value="0x04E30000" />
>+        <register type="NOA" address="0x00009888" value="0x02E30100" />
>+        <register type="NOA" address="0x00009888" value="0x00E41100" />
>+        <register type="NOA" address="0x00009888" value="0x02E60101" />
>+        <register type="NOA" address="0x00009888" value="0x02070909" />
>+        <register type="NOA" address="0x00009888" value="0x0208B0B0" />
>+        <register type="NOA" address="0x00009888" value="0x00093300" />
>+        <register type="NOA" address="0x00009888" value="0x020A5A5A" />
>+        <register type="NOA" address="0x00009888" value="0x00014400" />
>+        <register type="NOA" address="0x00009888" value="0x020B1010" />
>+        <register type="NOA" address="0x00009888" value="0x00022200" />
>+        <register type="NOA" address="0x00009888" value="0x020C0404" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
>+        <register type="NOA" address="0x00009888" value="0x02032400" />
>+        <register type="NOA" address="0x00009888" value="0x04032492" />
>+        <register type="NOA" address="0x00009888" value="0x06030002" />
>+        <register type="NOA" address="0x00009888" value="0x48030200" />
>+        <register type="NOA" address="0x00009888" value="0x4A030803" />
>+        <register type="NOA" address="0x00009888" value="0x16030000" />
>+        <register type="NOA" address="0x00009888" value="0x3E0318E0" />
>+        <register type="NOA" address="0x00009888" value="0x18031000" />
>+        <register type="NOA" address="0x00009888" value="0x1A030000" />
>+        <register type="NOA" address="0x00009888" value="0x40030C0A" />
>+        <register type="NOA" address="0x00009888" value="0x1C030000" />
>+        <register type="NOA" address="0x00009888" value="0x1E030000" />
>+        <register type="NOA" address="0x00009888" value="0x20031000" />
>+        <register type="NOA" address="0x00009888" value="0x42030002" />
>+        <register type="NOA" address="0x00009884" value="0x00000008" />
>+        <register type="NOA" address="0x00009888" value="0x0001FF00" />
>+        <register type="NOA" address="0x00009888" value="0x02025555" />
>+        <register type="NOA" address="0x00009888" value="0x02045555" />
>+        <register type="NOA" address="0x00009888" value="0x02055555" />
>+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x5C030002" />
>+    </register_config>
>+    <register_config type="OA">
>+        <register type="OA" address="0x0000D920" value="0x00000000" />
>+        <register type="OA" address="0x0000D900" value="0x00000000" />
>+        <register type="OA" address="0x0000D904" value="0x00800000" />
>+        <register type="OA" address="0x0000D910" value="0x00000000" />
>+        <register type="OA" address="0x0000D914" value="0x00800000" />
>+        <register type="OA" address="0x0000DC40" value="0x00000000" />
>+    </register_config>
>+    <register_config type="FLEX">
>+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
>+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
>+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
>+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
>+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
>+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
>+    </register_config>
>+  </set>
>+
>+  <set name="Sampler_2"
>+       chipset="DG2"
>+       symbol_name="Sampler_2"
>+       underscore_name="sampler_2"
>+       mdapi_supported_apis="VK OGL OCL MEDIA IO"
>+       hw_config_guid="c60dea82-db13-4882-8d31-a02bce151ef7"
>+       >
>+    <counter name="GPU Time Elapsed"
>+             symbol_name="GpuTime"
>+             underscore_name="gpu_time"
>+             description="Time elapsed on the GPU during the measurement."
>+             data_type="uint64"
>+             units="ns"
>+             semantic_type="duration"
>+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Core Clocks"
>+             symbol_name="GpuCoreClocks"
>+             underscore_name="gpu_core_clocks"
>+             description="The total number of GPU core clocks elapsed during the measurement."
>+             data_type="uint64"
>+             units="cycles"
>+             semantic_type="event"
>+             equation="GPU_CLOCK 0 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="AVG GPU Core Frequency"
>+             symbol_name="AvgGpuCoreFrequency"
>+             underscore_name="avg_gpu_core_frequency"
>+             description="Average GPU Core Frequency in the measurement."
>+             data_type="uint64"
>+             max_equation="$GpuMaxFrequency"
>+             units="hz"
>+             semantic_type="event"
>+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Busy"
>+             symbol_name="GpuBusy"
>+             underscore_name="gpu_busy"
>+             description="The percentage of time in which the GPU has been processing GPU commands."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="VS Threads Dispatched"
>+             symbol_name="VsThreads"
>+             underscore_name="vs_threads"
>+             description="The total number of vertex shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 1 READ"
>+             mdapi_group="EU Array/Vertex Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="HS Threads Dispatched"
>+             symbol_name="HsThreads"
>+             underscore_name="hs_threads"
>+             description="The total number of hull shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 2 READ"
>+             mdapi_group="EU Array/Hull Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="DS Threads Dispatched"
>+             symbol_name="DsThreads"
>+             underscore_name="ds_threads"
>+             description="The total number of domain shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 3 READ"
>+             mdapi_group="EU Array/Domain Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GS Threads Dispatched"
>+             symbol_name="GsThreads"
>+             underscore_name="gs_threads"
>+             description="The total number of geometry shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 5 READ"
>+             mdapi_group="EU Array/Geometry Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="FS Threads Dispatched"
>+             symbol_name="PsThreads"
>+             underscore_name="ps_threads"
>+             description="The total number of fragment shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 6 READ"
>+             mdapi_group="EU Array/Fragment Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="CS Threads Dispatched"
>+             symbol_name="CsThreads"
>+             underscore_name="cs_threads"
>+             description="The total number of compute shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 4 READ"
>+             mdapi_group="EU Array/Compute Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Active"
>+             symbol_name="EuActive"
>+             underscore_name="eu_active"
>+             description="The percentage of time in which the Execution Units were actively processing."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Stall"
>+             symbol_name="EuStall"
>+             underscore_name="eu_stall"
>+             description="The percentage of time in which the Execution Units were stalled."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Thread Occupancy"
>+             symbol_name="EuThreadOccupancy"
>+             underscore_name="eu_thread_occupancy"
>+             description="The percentage of time in which hardware threads occupied EUs."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Slice1 DualSubslice0 Input Available"
>+             symbol_name="Sampler10InputAvailable"
>+             underscore_name="sampler10_input_available"
>+             description="The percentage of time in which slice1 dualsubslice0 sampler input is available"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 64 AND"
>+             mdapi_group="GPU/Sampler"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Slice1 DualSubslice1 Input Available"
>+             symbol_name="Sampler11InputAvailable"
>+             underscore_name="sampler11_input_available"
>+             description="The percentage of time in which slice1 dualsubslice1 sampler input is available"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 128 AND"
>+             mdapi_group="GPU/Sampler"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Slice1 DualSubslice2 Input Available"
>+             symbol_name="Sampler12InputAvailable"
>+             underscore_name="sampler12_input_available"
>+             description="The percentage of time in which slice1 dualsubslice2 sampler input is available"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 5 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 256 AND"
>+             mdapi_group="GPU/Sampler"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Slice1 DualSubslice3 Input Available"
>+             symbol_name="Sampler13InputAvailable"
>+             underscore_name="sampler13_input_available"
>+             description="The percentage of time in which slice1 dualsubslice3 sampler input is available"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 4 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 512 AND"
>+             mdapi_group="GPU/Sampler"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Slice1 DualSubslice0 Sampler Output Ready"
>+             symbol_name="Sampler10OutputReady"
>+             underscore_name="sampler10_output_ready"
>+             description="The percentage of time in which slice1 dualsubslice0 sampler output is ready"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 3 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 64 AND"
>+             mdapi_group="GPU/Sampler"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="subslice"
>+             />
>+    <counter name="Slice1 DualSubslice1 Sampler Output Ready"
>+             symbol_name="Sampler11OutputReady"
>+             underscore_name="sampler11_output_ready"
>+             description="The percentage of time in which slice1 dualsubslice1 sampler output is ready"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 2 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 128 AND"
>+             mdapi_group="GPU/Sampler"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="subslice"
>+             />
>+    <counter name="Slice1 DualSubslice2 Sampler Output Ready"
>+             symbol_name="Sampler12OutputReady"
>+             underscore_name="sampler12_output_ready"
>+             description="The percentage of time in which slice1 dualsubslice2 sampler output is ready"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 1 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 256 AND"
>+             mdapi_group="GPU/Sampler"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="subslice"
>+             />
>+    <counter name="Slice1 DualSubslice3 Sampler Output Ready"
>+             symbol_name="Sampler13OutputReady"
>+             underscore_name="sampler13_output_ready"
>+             description="The percentage of time in which slice1 dualsubslice3 sampler output is ready"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 512 AND"
>+             mdapi_group="GPU/Sampler"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="subslice"
>+             />
>+    <register_config type="NOA">
>+        <register type="NOA" address="0x00000D04" value="0x00000200" />
>+        <register type="NOA" address="0x00009884" value="0x0000000C" />
>+        <register type="NOA" address="0x00009888" value="0x1D824800" />
>+        <register type="NOA" address="0x00009888" value="0x15832C00" />
>+        <register type="NOA" address="0x00009888" value="0x1DA24800" />
>+        <register type="NOA" address="0x00009888" value="0x15A32C00" />
>+        <register type="NOA" address="0x00009888" value="0x1DC24800" />
>+        <register type="NOA" address="0x00009888" value="0x15C32C00" />
>+        <register type="NOA" address="0x00009888" value="0x1DE24800" />
>+        <register type="NOA" address="0x00009888" value="0x15E32C00" />
>+        <register type="NOA" address="0x00009888" value="0x05824000" />
>+        <register type="NOA" address="0x00009888" value="0x17820027" />
>+        <register type="NOA" address="0x00009888" value="0x03820000" />
>+        <register type="NOA" address="0x00009888" value="0x0F830580" />
>+        <register type="NOA" address="0x00009888" value="0x05830000" />
>+        <register type="NOA" address="0x00009888" value="0x03834000" />
>+        <register type="NOA" address="0x00009888" value="0x01848800" />
>+        <register type="NOA" address="0x00009888" value="0x03864040" />
>+        <register type="NOA" address="0x00009888" value="0x05A24000" />
>+        <register type="NOA" address="0x00009888" value="0x15A209C0" />
>+        <register type="NOA" address="0x00009888" value="0x03A20000" />
>+        <register type="NOA" address="0x00009888" value="0x0FA30016" />
>+        <register type="NOA" address="0x00009888" value="0x05A30000" />
>+        <register type="NOA" address="0x00009888" value="0x03A31000" />
>+        <register type="NOA" address="0x00009888" value="0x01A44400" />
>+        <register type="NOA" address="0x00009888" value="0x03A61010" />
>+        <register type="NOA" address="0x00009888" value="0x05C24000" />
>+        <register type="NOA" address="0x00009888" value="0x15C20027" />
>+        <register type="NOA" address="0x00009888" value="0x03C20000" />
>+        <register type="NOA" address="0x00009888" value="0x0DC30580" />
>+        <register type="NOA" address="0x00009888" value="0x05C30000" />
>+        <register type="NOA" address="0x00009888" value="0x03C30400" />
>+        <register type="NOA" address="0x00009888" value="0x01C42200" />
>+        <register type="NOA" address="0x00009888" value="0x03C60404" />
>+        <register type="NOA" address="0x00009888" value="0x05E24000" />
>+        <register type="NOA" address="0x00009888" value="0x13E209C0" />
>+        <register type="NOA" address="0x00009888" value="0x03E20000" />
>+        <register type="NOA" address="0x00009888" value="0x0DE30016" />
>+        <register type="NOA" address="0x00009888" value="0x05E30000" />
>+        <register type="NOA" address="0x00009888" value="0x03E30100" />
>+        <register type="NOA" address="0x00009888" value="0x01E41100" />
>+        <register type="NOA" address="0x00009888" value="0x03E60101" />
>+        <register type="NOA" address="0x00009888" value="0x0003FF00" />
>+        <register type="NOA" address="0x00009888" value="0x0208FFFF" />
>+        <register type="NOA" address="0x00009888" value="0x020A5555" />
>+        <register type="NOA" address="0x00009888" value="0x0001FF00" />
>+        <register type="NOA" address="0x00009888" value="0x020BAAAA" />
>+        <register type="NOA" address="0x00009888" value="0x03070909" />
>+        <register type="NOA" address="0x00009888" value="0x0308B0B0" />
>+        <register type="NOA" address="0x00009888" value="0x01093300" />
>+        <register type="NOA" address="0x00009888" value="0x030A5A5A" />
>+        <register type="NOA" address="0x00009888" value="0x01014400" />
>+        <register type="NOA" address="0x00009888" value="0x030B1010" />
>+        <register type="NOA" address="0x00009888" value="0x01022200" />
>+        <register type="NOA" address="0x00009888" value="0x030C0404" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
>+        <register type="NOA" address="0x00009888" value="0x02032400" />
>+        <register type="NOA" address="0x00009888" value="0x04032492" />
>+        <register type="NOA" address="0x00009888" value="0x06030002" />
>+        <register type="NOA" address="0x00009888" value="0x48030200" />
>+        <register type="NOA" address="0x00009888" value="0x4A030803" />
>+        <register type="NOA" address="0x00009888" value="0x16030000" />
>+        <register type="NOA" address="0x00009888" value="0x3E0318E0" />
>+        <register type="NOA" address="0x00009888" value="0x18031000" />
>+        <register type="NOA" address="0x00009888" value="0x1A030000" />
>+        <register type="NOA" address="0x00009888" value="0x40030C0A" />
>+        <register type="NOA" address="0x00009888" value="0x1C030000" />
>+        <register type="NOA" address="0x00009888" value="0x1E030000" />
>+        <register type="NOA" address="0x00009888" value="0x20031000" />
>+        <register type="NOA" address="0x00009888" value="0x42030002" />
>+        <register type="NOA" address="0x00009884" value="0x00000008" />
>+        <register type="NOA" address="0x00009888" value="0x0001FF00" />
>+        <register type="NOA" address="0x00009888" value="0x02025555" />
>+        <register type="NOA" address="0x00009888" value="0x02045555" />
>+        <register type="NOA" address="0x00009888" value="0x02055555" />
>+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x5C030002" />
>+    </register_config>
>+    <register_config type="OA">
>+        <register type="OA" address="0x0000D920" value="0x00000000" />
>+        <register type="OA" address="0x0000D900" value="0x00000000" />
>+        <register type="OA" address="0x0000D904" value="0x00800000" />
>+        <register type="OA" address="0x0000D910" value="0x00000000" />
>+        <register type="OA" address="0x0000D914" value="0x00800000" />
>+        <register type="OA" address="0x0000DC40" value="0x00000000" />
>+    </register_config>
>+    <register_config type="FLEX">
>+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
>+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
>+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
>+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
>+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
>+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
>+    </register_config>
>+  </set>
>+
>+  <set name="TDL_1"
>+       chipset="DG2"
>+       symbol_name="TDL_1"
>+       underscore_name="tdl_1"
>+       mdapi_supported_apis="VK OGL OCL MEDIA IO"
>+       hw_config_guid="7f5cb8ec-38bb-48c9-9ad2-60075eba2833"
>+       >
>+    <counter name="GPU Time Elapsed"
>+             symbol_name="GpuTime"
>+             underscore_name="gpu_time"
>+             description="Time elapsed on the GPU during the measurement."
>+             data_type="uint64"
>+             units="ns"
>+             semantic_type="duration"
>+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Core Clocks"
>+             symbol_name="GpuCoreClocks"
>+             underscore_name="gpu_core_clocks"
>+             description="The total number of GPU core clocks elapsed during the measurement."
>+             data_type="uint64"
>+             units="cycles"
>+             semantic_type="event"
>+             equation="GPU_CLOCK 0 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="AVG GPU Core Frequency"
>+             symbol_name="AvgGpuCoreFrequency"
>+             underscore_name="avg_gpu_core_frequency"
>+             description="Average GPU Core Frequency in the measurement."
>+             data_type="uint64"
>+             max_equation="$GpuMaxFrequency"
>+             units="hz"
>+             semantic_type="event"
>+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Busy"
>+             symbol_name="GpuBusy"
>+             underscore_name="gpu_busy"
>+             description="The percentage of time in which the GPU has been processing GPU commands."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="VS Threads Dispatched"
>+             symbol_name="VsThreads"
>+             underscore_name="vs_threads"
>+             description="The total number of vertex shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 1 READ"
>+             mdapi_group="EU Array/Vertex Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="HS Threads Dispatched"
>+             symbol_name="HsThreads"
>+             underscore_name="hs_threads"
>+             description="The total number of hull shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 2 READ"
>+             mdapi_group="EU Array/Hull Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="DS Threads Dispatched"
>+             symbol_name="DsThreads"
>+             underscore_name="ds_threads"
>+             description="The total number of domain shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 3 READ"
>+             mdapi_group="EU Array/Domain Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GS Threads Dispatched"
>+             symbol_name="GsThreads"
>+             underscore_name="gs_threads"
>+             description="The total number of geometry shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 5 READ"
>+             mdapi_group="EU Array/Geometry Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="FS Threads Dispatched"
>+             symbol_name="PsThreads"
>+             underscore_name="ps_threads"
>+             description="The total number of fragment shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 6 READ"
>+             mdapi_group="EU Array/Fragment Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="CS Threads Dispatched"
>+             symbol_name="CsThreads"
>+             underscore_name="cs_threads"
>+             description="The total number of compute shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 4 READ"
>+             mdapi_group="EU Array/Compute Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Active"
>+             symbol_name="EuActive"
>+             underscore_name="eu_active"
>+             description="The percentage of time in which the Execution Units were actively processing."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Stall"
>+             symbol_name="EuStall"
>+             underscore_name="eu_stall"
>+             description="The percentage of time in which the Execution Units were stalled."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Thread Occupancy"
>+             symbol_name="EuThreadOccupancy"
>+             underscore_name="eu_thread_occupancy"
>+             description="The percentage of time in which hardware threads occupied EUs."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Non-PS Thread Ready For Dispatch on Slice0 DualSubslice0 Thread Dispatcher"
>+             symbol_name="NonPSThread00ReadyForDispatch"
>+             underscore_name="non_ps_thread00_ready_for_dispatch"
>+             description="The percentage of time in which non-PS thread is ready for dispatch on slice0 dualsubslice0 thread dispatcher"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 1 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Non-PS Thread Ready For Dispatch on Slice0 DualSubslice1 Thread Dispatcher"
>+             symbol_name="NonPSThread01ReadyForDispatch"
>+             underscore_name="non_ps_thread01_ready_for_dispatch"
>+             description="The percentage of time in which non-PS thread is ready for dispatch on slice0 dualsubslice1 thread dispatcher"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 2 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Non-PS Thread Ready For Dispatch on Slice0 DualSubslice2 Thread Dispatcher"
>+             symbol_name="NonPSThread02ReadyForDispatch"
>+             underscore_name="non_ps_thread02_ready_for_dispatch"
>+             description="The percentage of time in which non-PS thread is ready for dispatch on slice0 dualsubslice2 thread dispatcher"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 2 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 4 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Non-PS Thread Ready For Dispatch on Slice0 DualSubslice3 Thread Dispatcher"
>+             symbol_name="NonPSThread03ReadyForDispatch"
>+             underscore_name="non_ps_thread03_ready_for_dispatch"
>+             description="The percentage of time in which non-PS thread is ready for dispatch on slice0 dualsubslice3 thread dispatcher"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 3 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 8 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Non-PS Thread Ready For Dispatch on Slice1 DualSubslice0 Thread Dispatcher"
>+             symbol_name="NonPSThread10ReadyForDispatch"
>+             underscore_name="non_ps_thread10_ready_for_dispatch"
>+             description="The percentage of time in which non-PS thread is ready for dispatch on slice1 dualsubslice0 thread dispatcher"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 4 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 64 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Non-PS Thread Ready For Dispatch on Slice1 DualSubslice1 Thread Dispatcher"
>+             symbol_name="NonPSThread11ReadyForDispatch"
>+             underscore_name="non_ps_thread11_ready_for_dispatch"
>+             description="The percentage of time in which non-PS thread is ready for dispatch on slice1 dualsubslice1 thread dispatcher"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 5 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 128 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Non-PS Thread Ready For Dispatch on Slice1 DualSubslice2 Thread Dispatcher"
>+             symbol_name="NonPSThread12ReadyForDispatch"
>+             underscore_name="non_ps_thread12_ready_for_dispatch"
>+             description="The percentage of time in which non-PS thread is ready for dispatch on slice1 dualsubslice2 thread dispatcher"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 6 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 256 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Non-PS Thread Ready For Dispatch on Slice1 DualSubslice3 Thread Dispatcher"
>+             symbol_name="NonPSThread13ReadyForDispatch"
>+             underscore_name="non_ps_thread13_ready_for_dispatch"
>+             description="The percentage of time in which non-PS thread is ready for dispatch on slice1 dualsubslice3 thread dispatcher"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 7 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 512 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice0 DualSubslice3 Thread Dispatcher Port 0"
>+             symbol_name="ThreadHeader03ReadyPort0"
>+             underscore_name="thread_header03_ready_port0"
>+             description="The percentage of time in which thread header is ready on slice0 dualsubslice3 thread dispatcher port 0"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 8 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice0 DualSubslice3 Thread Dispatcher Port 1"
>+             symbol_name="ThreadHeader03ReadyPort1"
>+             underscore_name="thread_header03_ready_port1"
>+             description="The percentage of time in which thread header is ready on slice0 dualsubslice3 thread dispatcher port 1"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 8 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice0 DualSubslice3 Thread Dispatcher Port 2"
>+             symbol_name="ThreadHeader03ReadyPort2"
>+             underscore_name="thread_header03_ready_port2"
>+             description="The percentage of time in which thread header is ready on slice0 dualsubslice3 thread dispatcher port 2"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 5 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 8 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice0 DualSubslice3 Thread Dispatcher Port 3"
>+             symbol_name="ThreadHeader03ReadyPort3"
>+             underscore_name="thread_header03_ready_port3"
>+             description="The percentage of time in which thread header is ready on slice0 dualsubslice3 thread dispatcher port 3"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 4 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 8 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice1 DualSubslice3 Thread Dispatcher Port 0"
>+             symbol_name="ThreadHeader13ReadyPort0"
>+             underscore_name="thread_header13_ready_port0"
>+             description="The percentage of time in which thread header is ready on slice1 dualsubslice3 thread dispatcher port 0"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 3 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 512 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice1 DualSubslice3 Thread Dispatcher Port 1"
>+             symbol_name="ThreadHeader13ReadyPort1"
>+             underscore_name="thread_header13_ready_port1"
>+             description="The percentage of time in which thread header is ready on slice1 dualsubslice3 thread dispatcher port 1"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 2 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 512 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice1 DualSubslice3 Thread Dispatcher Port 2"
>+             symbol_name="ThreadHeader13ReadyPort2"
>+             underscore_name="thread_header13_ready_port2"
>+             description="The percentage of time in which thread header is ready on slice1 dualsubslice3 thread dispatcher port 2"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 1 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 512 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice1 DualSubslice3 Thread Dispatcher Port 3"
>+             symbol_name="ThreadHeader13ReadyPort3"
>+             underscore_name="thread_header13_ready_port3"
>+             description="The percentage of time in which thread header is ready on slice1 dualsubslice3 thread dispatcher port 3"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 512 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <register_config type="NOA">
>+        <register type="NOA" address="0x00000D04" value="0x00000200" />
>+        <register type="NOA" address="0x00009884" value="0x0000000C" />
>+        <register type="NOA" address="0x00009888" value="0x2A81001C" />
>+        <register type="NOA" address="0x00009888" value="0x2AA1001C" />
>+        <register type="NOA" address="0x00009888" value="0x2AC1001C" />
>+        <register type="NOA" address="0x00009888" value="0x2AE1001C" />
>+        <register type="NOA" address="0x00009888" value="0x2B81001C" />
>+        <register type="NOA" address="0x00009888" value="0x2BA1001C" />
>+        <register type="NOA" address="0x00009888" value="0x2BC1001C" />
>+        <register type="NOA" address="0x00009888" value="0x2BE1001C" />
>+        <register type="NOA" address="0x00009888" value="0x06810300" />
>+        <register type="NOA" address="0x00009888" value="0x08810030" />
>+        <register type="NOA" address="0x00009888" value="0x00810000" />
>+        <register type="NOA" address="0x00009888" value="0x00830002" />
>+        <register type="NOA" address="0x00009888" value="0x00840001" />
>+        <register type="NOA" address="0x00009888" value="0x00860001" />
>+        <register type="NOA" address="0x00009888" value="0x00870001" />
>+        <register type="NOA" address="0x00009888" value="0x06A10300" />
>+        <register type="NOA" address="0x00009888" value="0x08A10C00" />
>+        <register type="NOA" address="0x00009888" value="0x00A10000" />
>+        <register type="NOA" address="0x00009888" value="0x00A30008" />
>+        <register type="NOA" address="0x00009888" value="0x00A40002" />
>+        <register type="NOA" address="0x00009888" value="0x00A60004" />
>+        <register type="NOA" address="0x00009888" value="0x00A70004" />
>+        <register type="NOA" address="0x00009888" value="0x06C10300" />
>+        <register type="NOA" address="0x00009888" value="0x0AC10030" />
>+        <register type="NOA" address="0x00009888" value="0x08C10000" />
>+        <register type="NOA" address="0x00009888" value="0x00C10000" />
>+        <register type="NOA" address="0x00009888" value="0x00C30020" />
>+        <register type="NOA" address="0x00009888" value="0x00C40004" />
>+        <register type="NOA" address="0x00009888" value="0x00C60010" />
>+        <register type="NOA" address="0x00009888" value="0x00C70010" />
>+        <register type="NOA" address="0x00009888" value="0x06E10300" />
>+        <register type="NOA" address="0x00009888" value="0x0AE10C00" />
>+        <register type="NOA" address="0x00009888" value="0x14E10CF4" />
>+        <register type="NOA" address="0x00009888" value="0x16E10C72" />
>+        <register type="NOA" address="0x00009888" value="0x08E10000" />
>+        <register type="NOA" address="0x00009888" value="0x00E10000" />
>+        <register type="NOA" address="0x00009888" value="0x02E10000" />
>+        <register type="NOA" address="0x00009888" value="0x00E30080" />
>+        <register type="NOA" address="0x00009888" value="0x02E3AA00" />
>+        <register type="NOA" address="0x00009888" value="0x00E4F008" />
>+        <register type="NOA" address="0x00009888" value="0x00E60040" />
>+        <register type="NOA" address="0x00009888" value="0x02E65500" />
>+        <register type="NOA" address="0x00009888" value="0x00E70040" />
>+        <register type="NOA" address="0x00009888" value="0x02E75500" />
>+        <register type="NOA" address="0x00009888" value="0x07810300" />
>+        <register type="NOA" address="0x00009888" value="0x0D810030" />
>+        <register type="NOA" address="0x00009888" value="0x09810000" />
>+        <register type="NOA" address="0x00009888" value="0x01810000" />
>+        <register type="NOA" address="0x00009888" value="0x01830200" />
>+        <register type="NOA" address="0x00009888" value="0x01840010" />
>+        <register type="NOA" address="0x00009888" value="0x01860100" />
>+        <register type="NOA" address="0x00009888" value="0x01870100" />
>+        <register type="NOA" address="0x00009888" value="0x07A10300" />
>+        <register type="NOA" address="0x00009888" value="0x0DA10C00" />
>+        <register type="NOA" address="0x00009888" value="0x09A10000" />
>+        <register type="NOA" address="0x00009888" value="0x01A10000" />
>+        <register type="NOA" address="0x00009888" value="0x01A30800" />
>+        <register type="NOA" address="0x00009888" value="0x01A40020" />
>+        <register type="NOA" address="0x00009888" value="0x01A60400" />
>+        <register type="NOA" address="0x00009888" value="0x01A70400" />
>+        <register type="NOA" address="0x00009888" value="0x07C10300" />
>+        <register type="NOA" address="0x00009888" value="0x0FC10030" />
>+        <register type="NOA" address="0x00009888" value="0x09C10000" />
>+        <register type="NOA" address="0x00009888" value="0x01C10000" />
>+        <register type="NOA" address="0x00009888" value="0x01C32000" />
>+        <register type="NOA" address="0x00009888" value="0x01C40040" />
>+        <register type="NOA" address="0x00009888" value="0x01C61000" />
>+        <register type="NOA" address="0x00009888" value="0x01C71000" />
>+        <register type="NOA" address="0x00009888" value="0x07E10300" />
>+        <register type="NOA" address="0x00009888" value="0x0FE10C00" />
>+        <register type="NOA" address="0x00009888" value="0x11E10CF4" />
>+        <register type="NOA" address="0x00009888" value="0x13E10C72" />
>+        <register type="NOA" address="0x00009888" value="0x09E10000" />
>+        <register type="NOA" address="0x00009888" value="0x01E10000" />
>+        <register type="NOA" address="0x00009888" value="0x03E10000" />
>+        <register type="NOA" address="0x00009888" value="0x01E38000" />
>+        <register type="NOA" address="0x00009888" value="0x03E300AA" />
>+        <register type="NOA" address="0x00009888" value="0x01E40F80" />
>+        <register type="NOA" address="0x00009888" value="0x01E64000" />
>+        <register type="NOA" address="0x00009888" value="0x03E60055" />
>+        <register type="NOA" address="0x00009888" value="0x01E74000" />
>+        <register type="NOA" address="0x00009888" value="0x03E70055" />
>+        <register type="NOA" address="0x00009888" value="0x00030FF0" />
>+        <register type="NOA" address="0x00009888" value="0x00070060" />
>+        <register type="NOA" address="0x00009888" value="0x02075500" />
>+        <register type="NOA" address="0x00009888" value="0x0008FF0E" />
>+        <register type="NOA" address="0x00009888" value="0x020800FF" />
>+        <register type="NOA" address="0x00009888" value="0x0009F00C" />
>+        <register type="NOA" address="0x00009888" value="0x000A55A5" />
>+        <register type="NOA" address="0x00009888" value="0x020AAA55" />
>+        <register type="NOA" address="0x00009888" value="0x00010FF2" />
>+        <register type="NOA" address="0x00009888" value="0x000BAA04" />
>+        <register type="NOA" address="0x00009888" value="0x020B00AA" />
>+        <register type="NOA" address="0x00009888" value="0x00020004" />
>+        <register type="NOA" address="0x00009888" value="0x000C0010" />
>+        <register type="NOA" address="0x00009888" value="0x01076000" />
>+        <register type="NOA" address="0x00009888" value="0x03070055" />
>+        <register type="NOA" address="0x00009888" value="0x01080E00" />
>+        <register type="NOA" address="0x00009888" value="0x01090FC0" />
>+        <register type="NOA" address="0x00009888" value="0x010AA500" />
>+        <register type="NOA" address="0x00009888" value="0x030A00AA" />
>+        <register type="NOA" address="0x00009888" value="0x01010020" />
>+        <register type="NOA" address="0x00009888" value="0x010B0400" />
>+        <register type="NOA" address="0x00009888" value="0x01020040" />
>+        <register type="NOA" address="0x00009888" value="0x010C1000" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
>+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
>+        <register type="NOA" address="0x00009888" value="0x00032492" />
>+        <register type="NOA" address="0x00009888" value="0x02032492" />
>+        <register type="NOA" address="0x00009888" value="0x04032492" />
>+        <register type="NOA" address="0x00009888" value="0x06030002" />
>+        <register type="NOA" address="0x00009888" value="0x14037000" />
>+        <register type="NOA" address="0x00009888" value="0x3E03294A" />
>+        <register type="NOA" address="0x00009888" value="0x26037000" />
>+        <register type="NOA" address="0x00009888" value="0x42032007" />
>+        <register type="NOA" address="0x00009888" value="0x2C033000" />
>+        <register type="NOA" address="0x00009888" value="0x32037000" />
>+        <register type="NOA" address="0x00009888" value="0x44030047" />
>+        <register type="NOA" address="0x00009888" value="0x38031000" />
>+        <register type="NOA" address="0x00009888" value="0x46030003" />
>+        <register type="NOA" address="0x00009888" value="0x480300A0" />
>+        <register type="NOA" address="0x00009888" value="0x4A03140A" />
>+        <register type="NOA" address="0x00009888" value="0x16030000" />
>+        <register type="NOA" address="0x00009888" value="0x18030000" />
>+        <register type="NOA" address="0x00009888" value="0x1A037000" />
>+        <register type="NOA" address="0x00009888" value="0x40031CE7" />
>+        <register type="NOA" address="0x00009888" value="0x1C037000" />
>+        <register type="NOA" address="0x00009888" value="0x1E037000" />
>+        <register type="NOA" address="0x00009888" value="0x20037000" />
>+        <register type="NOA" address="0x00009884" value="0x00000008" />
>+        <register type="NOA" address="0x00009888" value="0x0001FFFF" />
>+        <register type="NOA" address="0x00009888" value="0x00025555" />
>+        <register type="NOA" address="0x00009888" value="0x02025555" />
>+        <register type="NOA" address="0x00009888" value="0x00045555" />
>+        <register type="NOA" address="0x00009888" value="0x02045555" />
>+        <register type="NOA" address="0x00009888" value="0x00055555" />
>+        <register type="NOA" address="0x00009888" value="0x02055555" />
>+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
>+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x5C030002" />
>+    </register_config>
>+    <register_config type="OA">
>+        <register type="OA" address="0x0000D920" value="0x00000000" />
>+        <register type="OA" address="0x0000D900" value="0x00000000" />
>+        <register type="OA" address="0x0000D904" value="0xF0800000" />
>+        <register type="OA" address="0x0000D910" value="0x00000000" />
>+        <register type="OA" address="0x0000D914" value="0xF0800000" />
>+        <register type="OA" address="0x0000DC40" value="0x00FF0000" />
>+        <register type="OA" address="0x0000D940" value="0x00000002" />
>+        <register type="OA" address="0x0000D944" value="0x0000FFFE" />
>+        <register type="OA" address="0x0000DC00" value="0x00000002" />
>+        <register type="OA" address="0x0000DC04" value="0x0000FFFE" />
>+        <register type="OA" address="0x0000D948" value="0x00000002" />
>+        <register type="OA" address="0x0000D94C" value="0x0000FFFD" />
>+        <register type="OA" address="0x0000DC08" value="0x00000002" />
>+        <register type="OA" address="0x0000DC0C" value="0x0000FFFD" />
>+        <register type="OA" address="0x0000D950" value="0x00000002" />
>+        <register type="OA" address="0x0000D954" value="0x0000FFFB" />
>+        <register type="OA" address="0x0000DC10" value="0x00000002" />
>+        <register type="OA" address="0x0000DC14" value="0x0000FFFB" />
>+        <register type="OA" address="0x0000D958" value="0x00000002" />
>+        <register type="OA" address="0x0000D95C" value="0x0000FFF7" />
>+        <register type="OA" address="0x0000DC18" value="0x00000002" />
>+        <register type="OA" address="0x0000DC1C" value="0x0000FFF7" />
>+        <register type="OA" address="0x0000D960" value="0x00000002" />
>+        <register type="OA" address="0x0000D964" value="0x0000FFEF" />
>+        <register type="OA" address="0x0000DC20" value="0x00000002" />
>+        <register type="OA" address="0x0000DC24" value="0x0000FFEF" />
>+        <register type="OA" address="0x0000D968" value="0x00000002" />
>+        <register type="OA" address="0x0000D96C" value="0x0000FFDF" />
>+        <register type="OA" address="0x0000DC28" value="0x00000002" />
>+        <register type="OA" address="0x0000DC2C" value="0x0000FFDF" />
>+        <register type="OA" address="0x0000D970" value="0x00000002" />
>+        <register type="OA" address="0x0000D974" value="0x0000FFBF" />
>+        <register type="OA" address="0x0000DC30" value="0x00000002" />
>+        <register type="OA" address="0x0000DC34" value="0x0000FFBF" />
>+        <register type="OA" address="0x0000D978" value="0x00000002" />
>+        <register type="OA" address="0x0000D97C" value="0x0000FF7F" />
>+        <register type="OA" address="0x0000DC38" value="0x00000002" />
>+        <register type="OA" address="0x0000DC3C" value="0x0000FF7F" />
>+    </register_config>
>+    <register_config type="FLEX">
>+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
>+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
>+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
>+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
>+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
>+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
>+    </register_config>
>+  </set>
>+
>+  <set name="TDL_2"
>+       chipset="DG2"
>+       symbol_name="TDL_2"
>+       underscore_name="tdl_2"
>+       mdapi_supported_apis="VK OGL OCL MEDIA IO"
>+       hw_config_guid="141ce507-de6c-428c-95b2-076194d650db"
>+       >
>+    <counter name="GPU Time Elapsed"
>+             symbol_name="GpuTime"
>+             underscore_name="gpu_time"
>+             description="Time elapsed on the GPU during the measurement."
>+             data_type="uint64"
>+             units="ns"
>+             semantic_type="duration"
>+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Core Clocks"
>+             symbol_name="GpuCoreClocks"
>+             underscore_name="gpu_core_clocks"
>+             description="The total number of GPU core clocks elapsed during the measurement."
>+             data_type="uint64"
>+             units="cycles"
>+             semantic_type="event"
>+             equation="GPU_CLOCK 0 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="AVG GPU Core Frequency"
>+             symbol_name="AvgGpuCoreFrequency"
>+             underscore_name="avg_gpu_core_frequency"
>+             description="Average GPU Core Frequency in the measurement."
>+             data_type="uint64"
>+             max_equation="$GpuMaxFrequency"
>+             units="hz"
>+             semantic_type="event"
>+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Busy"
>+             symbol_name="GpuBusy"
>+             underscore_name="gpu_busy"
>+             description="The percentage of time in which the GPU has been processing GPU commands."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="VS Threads Dispatched"
>+             symbol_name="VsThreads"
>+             underscore_name="vs_threads"
>+             description="The total number of vertex shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 1 READ"
>+             mdapi_group="EU Array/Vertex Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="HS Threads Dispatched"
>+             symbol_name="HsThreads"
>+             underscore_name="hs_threads"
>+             description="The total number of hull shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 2 READ"
>+             mdapi_group="EU Array/Hull Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="DS Threads Dispatched"
>+             symbol_name="DsThreads"
>+             underscore_name="ds_threads"
>+             description="The total number of domain shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 3 READ"
>+             mdapi_group="EU Array/Domain Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GS Threads Dispatched"
>+             symbol_name="GsThreads"
>+             underscore_name="gs_threads"
>+             description="The total number of geometry shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 5 READ"
>+             mdapi_group="EU Array/Geometry Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="FS Threads Dispatched"
>+             symbol_name="PsThreads"
>+             underscore_name="ps_threads"
>+             description="The total number of fragment shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 6 READ"
>+             mdapi_group="EU Array/Fragment Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="CS Threads Dispatched"
>+             symbol_name="CsThreads"
>+             underscore_name="cs_threads"
>+             description="The total number of compute shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 4 READ"
>+             mdapi_group="EU Array/Compute Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Active"
>+             symbol_name="EuActive"
>+             underscore_name="eu_active"
>+             description="The percentage of time in which the Execution Units were actively processing."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Stall"
>+             symbol_name="EuStall"
>+             underscore_name="eu_stall"
>+             description="The percentage of time in which the Execution Units were stalled."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Thread Occupancy"
>+             symbol_name="EuThreadOccupancy"
>+             underscore_name="eu_thread_occupancy"
>+             description="The percentage of time in which hardware threads occupied EUs."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="PS Thread Ready For Dispatch on Slice0 Dualsubslice0 Thread Dispatcher"
>+             symbol_name="PSThread00ReadyForDispatch"
>+             underscore_name="ps_thread00_ready_for_dispatch"
>+             description="The percentage of time in which PS thread is ready for dispatch on slice0 dualsubslice0 thread dispatcher"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 1 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="PS Thread Ready For Dispatch on Slice0 Dualsubslice1 Thread Dispatcher"
>+             symbol_name="PSThread01ReadyForDispatch"
>+             underscore_name="ps_thread01_ready_for_dispatch"
>+             description="The percentage of time in which PS thread is ready for dispatch on slice0 dualsubslice1 thread dispatcher"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 2 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="PS Thread Ready For Dispatch on Slice0 Dualsubslice2 Thread Dispatcher"
>+             symbol_name="PSThread02ReadyForDispatch"
>+             underscore_name="ps_thread02_ready_for_dispatch"
>+             description="The percentage of time in which PS thread is ready for dispatch on slice0 dualsubslice2 thread dispatcher"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 2 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 4 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="PS Thread Ready For Dispatch on Slice0 Dualsubslice3 Thread Dispatcher"
>+             symbol_name="PSThread03ReadyForDispatch"
>+             underscore_name="ps_thread03_ready_for_dispatch"
>+             description="The percentage of time in which PS thread is ready for dispatch on slice0 dualsubslice3 thread dispatcher"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 3 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 8 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="PS Thread Ready For Dispatch on Slice1 Dualsubslice0 Thread Dispatcher"
>+             symbol_name="PSThread10ReadyForDispatch"
>+             underscore_name="ps_thread10_ready_for_dispatch"
>+             description="The percentage of time in which PS thread is ready for dispatch on slice1 dualsubslice0 thread dispatcher"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 4 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 64 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="PS Thread Ready For Dispatch on Slice1 Dualsubslice1 Thread Dispatcher"
>+             symbol_name="PSThread11ReadyForDispatch"
>+             underscore_name="ps_thread11_ready_for_dispatch"
>+             description="The percentage of time in which PS thread is ready for dispatch on slice1 dualsubslice1 thread dispatcher"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 5 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 128 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="PS Thread Ready For Dispatch on Slice1 Dualsubslice2 Thread Dispatcher"
>+             symbol_name="PSThread12ReadyForDispatch"
>+             underscore_name="ps_thread12_ready_for_dispatch"
>+             description="The percentage of time in which PS thread is ready for dispatch on slice1 dualsubslice2 thread dispatcher"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 6 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 256 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="PS Thread Ready For Dispatch on Slice1 Dualsubslice3 Thread Dispatcher"
>+             symbol_name="PSThread13ReadyForDispatch"
>+             underscore_name="ps_thread13_ready_for_dispatch"
>+             description="The percentage of time in which PS thread is ready for dispatch on slice1 dualsubslice3 thread dispatcher"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 7 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 512 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <register_config type="NOA">
>+        <register type="NOA" address="0x00000D04" value="0x00000200" />
>+        <register type="NOA" address="0x00009884" value="0x0000000C" />
>+        <register type="NOA" address="0x00009888" value="0x28810340" />
>+        <register type="NOA" address="0x00009888" value="0x28A10340" />
>+        <register type="NOA" address="0x00009888" value="0x28C10340" />
>+        <register type="NOA" address="0x00009888" value="0x28E10340" />
>+        <register type="NOA" address="0x00009888" value="0x29810340" />
>+        <register type="NOA" address="0x00009888" value="0x29A10340" />
>+        <register type="NOA" address="0x00009888" value="0x29C10340" />
>+        <register type="NOA" address="0x00009888" value="0x29E10340" />
>+        <register type="NOA" address="0x00009888" value="0x06810100" />
>+        <register type="NOA" address="0x00009888" value="0x08810C31" />
>+        <register type="NOA" address="0x00009888" value="0x00810000" />
>+        <register type="NOA" address="0x00009888" value="0x0083000A" />
>+        <register type="NOA" address="0x00009888" value="0x00840003" />
>+        <register type="NOA" address="0x00009888" value="0x00860005" />
>+        <register type="NOA" address="0x00009888" value="0x00870005" />
>+        <register type="NOA" address="0x00009888" value="0x06A10100" />
>+        <register type="NOA" address="0x00009888" value="0x0AA10C31" />
>+        <register type="NOA" address="0x00009888" value="0x08A10000" />
>+        <register type="NOA" address="0x00009888" value="0x00A10000" />
>+        <register type="NOA" address="0x00009888" value="0x00A300A0" />
>+        <register type="NOA" address="0x00009888" value="0x00A4000C" />
>+        <register type="NOA" address="0x00009888" value="0x00A60050" />
>+        <register type="NOA" address="0x00009888" value="0x00A70050" />
>+        <register type="NOA" address="0x00009888" value="0x06C10100" />
>+        <register type="NOA" address="0x00009888" value="0x0CC10C31" />
>+        <register type="NOA" address="0x00009888" value="0x08C10000" />
>+        <register type="NOA" address="0x00009888" value="0x00C10000" />
>+        <register type="NOA" address="0x00009888" value="0x00C30A00" />
>+        <register type="NOA" address="0x00009888" value="0x00C40030" />
>+        <register type="NOA" address="0x00009888" value="0x00C60500" />
>+        <register type="NOA" address="0x00009888" value="0x00C70500" />
>+        <register type="NOA" address="0x00009888" value="0x06E10100" />
>+        <register type="NOA" address="0x00009888" value="0x0EE10C31" />
>+        <register type="NOA" address="0x00009888" value="0x08E10000" />
>+        <register type="NOA" address="0x00009888" value="0x00E10000" />
>+        <register type="NOA" address="0x00009888" value="0x00E3A000" />
>+        <register type="NOA" address="0x00009888" value="0x00E400C0" />
>+        <register type="NOA" address="0x00009888" value="0x00E65000" />
>+        <register type="NOA" address="0x00009888" value="0x00E75000" />
>+        <register type="NOA" address="0x00009888" value="0x07810100" />
>+        <register type="NOA" address="0x00009888" value="0x11810C31" />
>+        <register type="NOA" address="0x00009888" value="0x09810000" />
>+        <register type="NOA" address="0x00009888" value="0x03810000" />
>+        <register type="NOA" address="0x00009888" value="0x0383000A" />
>+        <register type="NOA" address="0x00009888" value="0x01840300" />
>+        <register type="NOA" address="0x00009888" value="0x03860005" />
>+        <register type="NOA" address="0x00009888" value="0x03870005" />
>+        <register type="NOA" address="0x00009888" value="0x07A10100" />
>+        <register type="NOA" address="0x00009888" value="0x13A10C31" />
>+        <register type="NOA" address="0x00009888" value="0x09A10000" />
>+        <register type="NOA" address="0x00009888" value="0x03A10000" />
>+        <register type="NOA" address="0x00009888" value="0x03A300A0" />
>+        <register type="NOA" address="0x00009888" value="0x01A40C00" />
>+        <register type="NOA" address="0x00009888" value="0x03A60050" />
>+        <register type="NOA" address="0x00009888" value="0x03A70050" />
>+        <register type="NOA" address="0x00009888" value="0x07C10100" />
>+        <register type="NOA" address="0x00009888" value="0x15C10C31" />
>+        <register type="NOA" address="0x00009888" value="0x09C10000" />
>+        <register type="NOA" address="0x00009888" value="0x03C10000" />
>+        <register type="NOA" address="0x00009888" value="0x03C30A00" />
>+        <register type="NOA" address="0x00009888" value="0x01C43000" />
>+        <register type="NOA" address="0x00009888" value="0x03C60500" />
>+        <register type="NOA" address="0x00009888" value="0x03C70500" />
>+        <register type="NOA" address="0x00009888" value="0x07E10100" />
>+        <register type="NOA" address="0x00009888" value="0x17E10C31" />
>+        <register type="NOA" address="0x00009888" value="0x09E10000" />
>+        <register type="NOA" address="0x00009888" value="0x03E10000" />
>+        <register type="NOA" address="0x00009888" value="0x03E3A000" />
>+        <register type="NOA" address="0x00009888" value="0x01E4C000" />
>+        <register type="NOA" address="0x00009888" value="0x03E65000" />
>+        <register type="NOA" address="0x00009888" value="0x03E75000" />
>+        <register type="NOA" address="0x00009888" value="0x0003FF00" />
>+        <register type="NOA" address="0x00009888" value="0x00075A00" />
>+        <register type="NOA" address="0x00009888" value="0x000800FA" />
>+        <register type="NOA" address="0x00009888" value="0x0208FFFF" />
>+        <register type="NOA" address="0x00009888" value="0x000900F0" />
>+        <register type="NOA" address="0x00009888" value="0x000AAA55" />
>+        <register type="NOA" address="0x00009888" value="0x020A5555" />
>+        <register type="NOA" address="0x00009888" value="0x0001FF0C" />
>+        <register type="NOA" address="0x00009888" value="0x000B0050" />
>+        <register type="NOA" address="0x00009888" value="0x020BAAAA" />
>+        <register type="NOA" address="0x00009888" value="0x00020030" />
>+        <register type="NOA" address="0x00009888" value="0x000C0500" />
>+        <register type="NOA" address="0x00009888" value="0x03075A00" />
>+        <register type="NOA" address="0x00009888" value="0x030800FA" />
>+        <register type="NOA" address="0x00009888" value="0x0109F000" />
>+        <register type="NOA" address="0x00009888" value="0x030AAA55" />
>+        <register type="NOA" address="0x00009888" value="0x01010C00" />
>+        <register type="NOA" address="0x00009888" value="0x030B0050" />
>+        <register type="NOA" address="0x00009888" value="0x01023000" />
>+        <register type="NOA" address="0x00009888" value="0x030C0500" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
>+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
>+        <register type="NOA" address="0x00009888" value="0x00032492" />
>+        <register type="NOA" address="0x00009888" value="0x02032492" />
>+        <register type="NOA" address="0x00009888" value="0x04032492" />
>+        <register type="NOA" address="0x00009888" value="0x06030002" />
>+        <register type="NOA" address="0x00009888" value="0x14037000" />
>+        <register type="NOA" address="0x00009888" value="0x3E030C6A" />
>+        <register type="NOA" address="0x00009888" value="0x26037000" />
>+        <register type="NOA" address="0x00009888" value="0x4203014A" />
>+        <register type="NOA" address="0x00009888" value="0x2C037000" />
>+        <register type="NOA" address="0x00009888" value="0x32037000" />
>+        <register type="NOA" address="0x00009888" value="0x44030D00" />
>+        <register type="NOA" address="0x00009888" value="0x38033000" />
>+        <register type="NOA" address="0x00009888" value="0x46030EE8" />
>+        <register type="NOA" address="0x00009888" value="0x48030277" />
>+        <register type="NOA" address="0x00009888" value="0x4A030422" />
>+        <register type="NOA" address="0x00009888" value="0x16030000" />
>+        <register type="NOA" address="0x00009888" value="0x18030000" />
>+        <register type="NOA" address="0x00009888" value="0x1A030000" />
>+        <register type="NOA" address="0x00009888" value="0x40032800" />
>+        <register type="NOA" address="0x00009888" value="0x1C030000" />
>+        <register type="NOA" address="0x00009888" value="0x1E030000" />
>+        <register type="NOA" address="0x00009888" value="0x20030000" />
>+        <register type="NOA" address="0x00009884" value="0x00000008" />
>+        <register type="NOA" address="0x00009888" value="0x0001FFFF" />
>+        <register type="NOA" address="0x00009888" value="0x00025555" />
>+        <register type="NOA" address="0x00009888" value="0x02025555" />
>+        <register type="NOA" address="0x00009888" value="0x00045555" />
>+        <register type="NOA" address="0x00009888" value="0x02045555" />
>+        <register type="NOA" address="0x00009888" value="0x00055555" />
>+        <register type="NOA" address="0x00009888" value="0x02055555" />
>+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
>+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x5C030002" />
>+    </register_config>
>+    <register_config type="OA">
>+        <register type="OA" address="0x0000D920" value="0x00000000" />
>+        <register type="OA" address="0x0000D900" value="0x00000000" />
>+        <register type="OA" address="0x0000D904" value="0xF0800000" />
>+        <register type="OA" address="0x0000D910" value="0x00000000" />
>+        <register type="OA" address="0x0000D914" value="0xF0800000" />
>+        <register type="OA" address="0x0000DC40" value="0x00FF0000" />
>+        <register type="OA" address="0x0000D940" value="0x00000000" />
>+        <register type="OA" address="0x0000D944" value="0x0000FFFC" />
>+        <register type="OA" address="0x0000DC00" value="0x00000000" />
>+        <register type="OA" address="0x0000DC04" value="0x0000FFFC" />
>+        <register type="OA" address="0x0000D948" value="0x00000000" />
>+        <register type="OA" address="0x0000D94C" value="0x0000FFF3" />
>+        <register type="OA" address="0x0000DC08" value="0x00000000" />
>+        <register type="OA" address="0x0000DC0C" value="0x0000FFF3" />
>+        <register type="OA" address="0x0000D950" value="0x00000000" />
>+        <register type="OA" address="0x0000D954" value="0x0000FFCF" />
>+        <register type="OA" address="0x0000DC10" value="0x00000000" />
>+        <register type="OA" address="0x0000DC14" value="0x0000FFCF" />
>+        <register type="OA" address="0x0000D958" value="0x00000000" />
>+        <register type="OA" address="0x0000D95C" value="0x0000FF3F" />
>+        <register type="OA" address="0x0000DC18" value="0x00000000" />
>+        <register type="OA" address="0x0000DC1C" value="0x0000FF3F" />
>+        <register type="OA" address="0x0000D960" value="0x00000000" />
>+        <register type="OA" address="0x0000D964" value="0x0000FCFF" />
>+        <register type="OA" address="0x0000DC20" value="0x00000000" />
>+        <register type="OA" address="0x0000DC24" value="0x0000FCFF" />
>+        <register type="OA" address="0x0000D968" value="0x00000000" />
>+        <register type="OA" address="0x0000D96C" value="0x0000F3FF" />
>+        <register type="OA" address="0x0000DC28" value="0x00000000" />
>+        <register type="OA" address="0x0000DC2C" value="0x0000F3FF" />
>+        <register type="OA" address="0x0000D970" value="0x00000000" />
>+        <register type="OA" address="0x0000D974" value="0x0000CFFF" />
>+        <register type="OA" address="0x0000DC30" value="0x00000000" />
>+        <register type="OA" address="0x0000DC34" value="0x0000CFFF" />
>+        <register type="OA" address="0x0000D978" value="0x00000000" />
>+        <register type="OA" address="0x0000D97C" value="0x00003FFF" />
>+        <register type="OA" address="0x0000DC38" value="0x00000000" />
>+        <register type="OA" address="0x0000DC3C" value="0x00003FFF" />
>+    </register_config>
>+    <register_config type="FLEX">
>+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
>+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
>+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
>+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
>+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
>+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
>+    </register_config>
>+  </set>
>+
>+  <set name="TDL_3"
>+       chipset="DG2"
>+       symbol_name="TDL_3"
>+       underscore_name="tdl_3"
>+       mdapi_supported_apis="VK OGL OCL MEDIA IO"
>+       hw_config_guid="54a85b07-9ff2-4a12-bfb6-6bcdd9d4019c"
>+       >
>+    <counter name="GPU Time Elapsed"
>+             symbol_name="GpuTime"
>+             underscore_name="gpu_time"
>+             description="Time elapsed on the GPU during the measurement."
>+             data_type="uint64"
>+             units="ns"
>+             semantic_type="duration"
>+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Core Clocks"
>+             symbol_name="GpuCoreClocks"
>+             underscore_name="gpu_core_clocks"
>+             description="The total number of GPU core clocks elapsed during the measurement."
>+             data_type="uint64"
>+             units="cycles"
>+             semantic_type="event"
>+             equation="GPU_CLOCK 0 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="AVG GPU Core Frequency"
>+             symbol_name="AvgGpuCoreFrequency"
>+             underscore_name="avg_gpu_core_frequency"
>+             description="Average GPU Core Frequency in the measurement."
>+             data_type="uint64"
>+             max_equation="$GpuMaxFrequency"
>+             units="hz"
>+             semantic_type="event"
>+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Busy"
>+             symbol_name="GpuBusy"
>+             underscore_name="gpu_busy"
>+             description="The percentage of time in which the GPU has been processing GPU commands."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="VS Threads Dispatched"
>+             symbol_name="VsThreads"
>+             underscore_name="vs_threads"
>+             description="The total number of vertex shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 1 READ"
>+             mdapi_group="EU Array/Vertex Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="HS Threads Dispatched"
>+             symbol_name="HsThreads"
>+             underscore_name="hs_threads"
>+             description="The total number of hull shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 2 READ"
>+             mdapi_group="EU Array/Hull Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="DS Threads Dispatched"
>+             symbol_name="DsThreads"
>+             underscore_name="ds_threads"
>+             description="The total number of domain shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 3 READ"
>+             mdapi_group="EU Array/Domain Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GS Threads Dispatched"
>+             symbol_name="GsThreads"
>+             underscore_name="gs_threads"
>+             description="The total number of geometry shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 5 READ"
>+             mdapi_group="EU Array/Geometry Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="FS Threads Dispatched"
>+             symbol_name="PsThreads"
>+             underscore_name="ps_threads"
>+             description="The total number of fragment shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 6 READ"
>+             mdapi_group="EU Array/Fragment Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="CS Threads Dispatched"
>+             symbol_name="CsThreads"
>+             underscore_name="cs_threads"
>+             description="The total number of compute shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 4 READ"
>+             mdapi_group="EU Array/Compute Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Active"
>+             symbol_name="EuActive"
>+             underscore_name="eu_active"
>+             description="The percentage of time in which the Execution Units were actively processing."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Stall"
>+             symbol_name="EuStall"
>+             underscore_name="eu_stall"
>+             description="The percentage of time in which the Execution Units were stalled."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Thread Occupancy"
>+             symbol_name="EuThreadOccupancy"
>+             underscore_name="eu_thread_occupancy"
>+             description="The percentage of time in which hardware threads occupied EUs."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Thread Header Ready on Slice0 DualSubslice0 Thread Dispatcher"
>+             symbol_name="ThreadHeader00Ready"
>+             underscore_name="thread_header00_ready"
>+             description="The percentage of time in which thread header is ready on slice0 dualsubslice0 thread dispatcher"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 1 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice0 DualSubslice0 Thread Dispatcher Port 0"
>+             symbol_name="ThreadHeader00ReadyPort0"
>+             underscore_name="thread_header00_ready_port0"
>+             description="The percentage of time in which thread header is ready on slice0 dualsubslice0 thread dispatcher port 0"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 1 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice0 DualSubslice0 Thread Dispatcher Port 1"
>+             symbol_name="ThreadHeader00ReadyPort1"
>+             underscore_name="thread_header00_ready_port1"
>+             description="The percentage of time in which thread header is ready on slice0 dualsubslice0 thread dispatcher port 1"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 1 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice0 DualSubslice0 Thread Dispatcher Port 2"
>+             symbol_name="ThreadHeader00ReadyPort2"
>+             underscore_name="thread_header00_ready_port2"
>+             description="The percentage of time in which thread header is ready on slice0 dualsubslice0 thread dispatcher port 2"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 5 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 1 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice0 DualSubslice0 Thread Dispatcher Port 3"
>+             symbol_name="ThreadHeader00ReadyPort3"
>+             underscore_name="thread_header00_ready_port3"
>+             description="The percentage of time in which thread header is ready on slice0 dualsubslice0 thread dispatcher port 3"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 4 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 1 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice0 DualSubslice1 Thread Dispatcher"
>+             symbol_name="ThreadHeader01Ready"
>+             underscore_name="thread_header01_ready"
>+             description="The percentage of time in which thread header is ready on slice0 dualsubslice1 thread dispatcher"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 2 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice0 DualSubslice1 Thread Dispatcher Port 0"
>+             symbol_name="ThreadHeader01ReadyPort0"
>+             underscore_name="thread_header01_ready_port0"
>+             description="The percentage of time in which thread header is ready on slice0 dualsubslice1 thread dispatcher port 0"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 3 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 2 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice0 DualSubslice1 Thread Dispatcher Port 1"
>+             symbol_name="ThreadHeader01ReadyPort1"
>+             underscore_name="thread_header01_ready_port1"
>+             description="The percentage of time in which thread header is ready on slice0 dualsubslice1 thread dispatcher port 1"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 2 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 2 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice0 DualSubslice1 Thread Dispatcher Port 2"
>+             symbol_name="ThreadHeader01ReadyPort2"
>+             underscore_name="thread_header01_ready_port2"
>+             description="The percentage of time in which thread header is ready on slice0 dualsubslice1 thread dispatcher port 2"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 1 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 2 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice0 DualSubslice1 Thread Dispatcher Port 3"
>+             symbol_name="ThreadHeader01ReadyPort3"
>+             underscore_name="thread_header01_ready_port3"
>+             description="The percentage of time in which thread header is ready on slice0 dualsubslice1 thread dispatcher port 3"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 2 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice0 DualSubslice2 Thread Dispatcher"
>+             symbol_name="ThreadHeader02Ready"
>+             underscore_name="thread_header02_ready"
>+             description="The percentage of time in which thread header is ready on slice0 dualsubslice2 thread dispatcher"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 2 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 4 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice0 DualSubslice2 Thread Dispatcher Port 0"
>+             symbol_name="ThreadHeader02ReadyPort0"
>+             underscore_name="thread_header02_ready_port0"
>+             description="The percentage of time in which thread header is ready on slice0 dualsubslice2 thread dispatcher port 0"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 7 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 4 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice0 DualSubslice2 Thread Dispatcher Port 1"
>+             symbol_name="ThreadHeader02ReadyPort1"
>+             underscore_name="thread_header02_ready_port1"
>+             description="The percentage of time in which thread header is ready on slice0 dualsubslice2 thread dispatcher port 1"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 6 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 4 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice0 DualSubslice2 Thread Dispatcher Port 2"
>+             symbol_name="ThreadHeader02ReadyPort2"
>+             underscore_name="thread_header02_ready_port2"
>+             description="The percentage of time in which thread header is ready on slice0 dualsubslice2 thread dispatcher port 2"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 5 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 4 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice0 DualSubslice2 Thread Dispatcher Port 3"
>+             symbol_name="ThreadHeader02ReadyPort3"
>+             underscore_name="thread_header02_ready_port3"
>+             description="The percentage of time in which thread header is ready on slice0 dualsubslice2 thread dispatcher port 3"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 4 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 4 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice0 DualSubslice3 Thread Dispatcher"
>+             symbol_name="ThreadHeader03Ready"
>+             underscore_name="thread_header03_ready"
>+             description="The percentage of time in which thread header is ready on slice0 dualsubslice3 thread dispatcher"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 3 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 8 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <register_config type="NOA">
>+        <register type="NOA" address="0x00000D04" value="0x00000200" />
>+        <register type="NOA" address="0x00009884" value="0x0000000C" />
>+        <register type="NOA" address="0x00009888" value="0x2A81001C" />
>+        <register type="NOA" address="0x00009888" value="0x2AA1001C" />
>+        <register type="NOA" address="0x00009888" value="0x2AC1001C" />
>+        <register type="NOA" address="0x00009888" value="0x2AE1001C" />
>+        <register type="NOA" address="0x00009888" value="0x06810300" />
>+        <register type="NOA" address="0x00009888" value="0x14810CF4" />
>+        <register type="NOA" address="0x00009888" value="0x16810C72" />
>+        <register type="NOA" address="0x00009888" value="0x08810000" />
>+        <register type="NOA" address="0x00009888" value="0x02810000" />
>+        <register type="NOA" address="0x00009888" value="0x0283AA00" />
>+        <register type="NOA" address="0x00009888" value="0x0084F000" />
>+        <register type="NOA" address="0x00009888" value="0x02865500" />
>+        <register type="NOA" address="0x00009888" value="0x02875500" />
>+        <register type="NOA" address="0x00009888" value="0x06A10300" />
>+        <register type="NOA" address="0x00009888" value="0x10A10CF4" />
>+        <register type="NOA" address="0x00009888" value="0x12A10C72" />
>+        <register type="NOA" address="0x00009888" value="0x08A10000" />
>+        <register type="NOA" address="0x00009888" value="0x02A10000" />
>+        <register type="NOA" address="0x00009888" value="0x02A300AA" />
>+        <register type="NOA" address="0x00009888" value="0x00A40F00" />
>+        <register type="NOA" address="0x00009888" value="0x02A60055" />
>+        <register type="NOA" address="0x00009888" value="0x02A70055" />
>+        <register type="NOA" address="0x00009888" value="0x06C10300" />
>+        <register type="NOA" address="0x00009888" value="0x0CC10CF4" />
>+        <register type="NOA" address="0x00009888" value="0x0EC10C72" />
>+        <register type="NOA" address="0x00009888" value="0x08C10000" />
>+        <register type="NOA" address="0x00009888" value="0x00C10000" />
>+        <register type="NOA" address="0x00009888" value="0x00C3AA00" />
>+        <register type="NOA" address="0x00009888" value="0x00C400F0" />
>+        <register type="NOA" address="0x00009888" value="0x00C65500" />
>+        <register type="NOA" address="0x00009888" value="0x00C75500" />
>+        <register type="NOA" address="0x00009888" value="0x06E10300" />
>+        <register type="NOA" address="0x00009888" value="0x08E10CB1" />
>+        <register type="NOA" address="0x00009888" value="0x0AE10D33" />
>+        <register type="NOA" address="0x00009888" value="0x00E10000" />
>+        <register type="NOA" address="0x00009888" value="0x00E300AA" />
>+        <register type="NOA" address="0x00009888" value="0x00E4000F" />
>+        <register type="NOA" address="0x00009888" value="0x00E60055" />
>+        <register type="NOA" address="0x00009888" value="0x00E70055" />
>+        <register type="NOA" address="0x00009888" value="0x0007AA55" />
>+        <register type="NOA" address="0x00009888" value="0x0208AAFF" />
>+        <register type="NOA" address="0x00009888" value="0x000900FF" />
>+        <register type="NOA" address="0x00009888" value="0x000AAAAA" />
>+        <register type="NOA" address="0x00009888" value="0x020A5555" />
>+        <register type="NOA" address="0x00009888" value="0x00010F00" />
>+        <register type="NOA" address="0x00009888" value="0x020B0055" />
>+        <register type="NOA" address="0x00009888" value="0x000200F0" />
>+        <register type="NOA" address="0x00009888" value="0x000C5500" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
>+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
>+        <register type="NOA" address="0x00009888" value="0x00032492" />
>+        <register type="NOA" address="0x00009888" value="0x02032492" />
>+        <register type="NOA" address="0x00009888" value="0x04032492" />
>+        <register type="NOA" address="0x00009888" value="0x06030002" />
>+        <register type="NOA" address="0x00009888" value="0x14030000" />
>+        <register type="NOA" address="0x00009888" value="0x3E030C6A" />
>+        <register type="NOA" address="0x00009888" value="0x26030000" />
>+        <register type="NOA" address="0x00009888" value="0x42032942" />
>+        <register type="NOA" address="0x00009888" value="0x2C030000" />
>+        <register type="NOA" address="0x00009888" value="0x32030000" />
>+        <register type="NOA" address="0x00009888" value="0x4403000A" />
>+        <register type="NOA" address="0x00009888" value="0x38030000" />
>+        <register type="NOA" address="0x00009888" value="0x46030000" />
>+        <register type="NOA" address="0x00009888" value="0x48030000" />
>+        <register type="NOA" address="0x00009888" value="0x4A030603" />
>+        <register type="NOA" address="0x00009888" value="0x16030000" />
>+        <register type="NOA" address="0x00009888" value="0x18030000" />
>+        <register type="NOA" address="0x00009888" value="0x1A031000" />
>+        <register type="NOA" address="0x00009888" value="0x40030842" />
>+        <register type="NOA" address="0x00009888" value="0x1C031000" />
>+        <register type="NOA" address="0x00009888" value="0x1E031000" />
>+        <register type="NOA" address="0x00009888" value="0x20031000" />
>+        <register type="NOA" address="0x00009884" value="0x00000008" />
>+        <register type="NOA" address="0x00009888" value="0x0001FFFF" />
>+        <register type="NOA" address="0x00009888" value="0x00025555" />
>+        <register type="NOA" address="0x00009888" value="0x02025555" />
>+        <register type="NOA" address="0x00009888" value="0x00045555" />
>+        <register type="NOA" address="0x00009888" value="0x02045555" />
>+        <register type="NOA" address="0x00009888" value="0x00055555" />
>+        <register type="NOA" address="0x00009888" value="0x02055555" />
>+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
>+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x5C030002" />
>+    </register_config>
>+    <register_config type="OA">
>+        <register type="OA" address="0x0000D920" value="0x00000000" />
>+        <register type="OA" address="0x0000D900" value="0x00000000" />
>+        <register type="OA" address="0x0000D904" value="0xF0800000" />
>+        <register type="OA" address="0x0000D910" value="0x00000000" />
>+        <register type="OA" address="0x0000D914" value="0x00800000" />
>+        <register type="OA" address="0x0000DC40" value="0x000F0000" />
>+        <register type="OA" address="0x0000D940" value="0x00078000" />
>+        <register type="OA" address="0x0000D944" value="0x00000FFF" />
>+        <register type="OA" address="0x0000DC00" value="0x00078000" />
>+        <register type="OA" address="0x0000DC04" value="0x00000FFF" />
>+        <register type="OA" address="0x0000D948" value="0x00007800" />
>+        <register type="OA" address="0x0000D94C" value="0x0000F0FF" />
>+        <register type="OA" address="0x0000DC08" value="0x00007800" />
>+        <register type="OA" address="0x0000DC0C" value="0x0000F0FF" />
>+        <register type="OA" address="0x0000D950" value="0x00000780" />
>+        <register type="OA" address="0x0000D954" value="0x0000FF0F" />
>+        <register type="OA" address="0x0000DC10" value="0x00000780" />
>+        <register type="OA" address="0x0000DC14" value="0x0000FF0F" />
>+        <register type="OA" address="0x0000D958" value="0x00000078" />
>+        <register type="OA" address="0x0000D95C" value="0x0000FFF0" />
>+        <register type="OA" address="0x0000DC18" value="0x00000078" />
>+        <register type="OA" address="0x0000DC1C" value="0x0000FFF0" />
>+    </register_config>
>+    <register_config type="FLEX">
>+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
>+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
>+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
>+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
>+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
>+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
>+    </register_config>
>+  </set>
>+
>+  <set name="TDL_4"
>+       chipset="DG2"
>+       symbol_name="TDL_4"
>+       underscore_name="tdl_4"
>+       mdapi_supported_apis="VK OGL OCL MEDIA IO"
>+       hw_config_guid="651c40e5-195c-49da-b232-a9a0f753d899"
>+       >
>+    <counter name="GPU Time Elapsed"
>+             symbol_name="GpuTime"
>+             underscore_name="gpu_time"
>+             description="Time elapsed on the GPU during the measurement."
>+             data_type="uint64"
>+             units="ns"
>+             semantic_type="duration"
>+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Core Clocks"
>+             symbol_name="GpuCoreClocks"
>+             underscore_name="gpu_core_clocks"
>+             description="The total number of GPU core clocks elapsed during the measurement."
>+             data_type="uint64"
>+             units="cycles"
>+             semantic_type="event"
>+             equation="GPU_CLOCK 0 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="AVG GPU Core Frequency"
>+             symbol_name="AvgGpuCoreFrequency"
>+             underscore_name="avg_gpu_core_frequency"
>+             description="Average GPU Core Frequency in the measurement."
>+             data_type="uint64"
>+             max_equation="$GpuMaxFrequency"
>+             units="hz"
>+             semantic_type="event"
>+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Busy"
>+             symbol_name="GpuBusy"
>+             underscore_name="gpu_busy"
>+             description="The percentage of time in which the GPU has been processing GPU commands."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="VS Threads Dispatched"
>+             symbol_name="VsThreads"
>+             underscore_name="vs_threads"
>+             description="The total number of vertex shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 1 READ"
>+             mdapi_group="EU Array/Vertex Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="HS Threads Dispatched"
>+             symbol_name="HsThreads"
>+             underscore_name="hs_threads"
>+             description="The total number of hull shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 2 READ"
>+             mdapi_group="EU Array/Hull Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="DS Threads Dispatched"
>+             symbol_name="DsThreads"
>+             underscore_name="ds_threads"
>+             description="The total number of domain shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 3 READ"
>+             mdapi_group="EU Array/Domain Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GS Threads Dispatched"
>+             symbol_name="GsThreads"
>+             underscore_name="gs_threads"
>+             description="The total number of geometry shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 5 READ"
>+             mdapi_group="EU Array/Geometry Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="FS Threads Dispatched"
>+             symbol_name="PsThreads"
>+             underscore_name="ps_threads"
>+             description="The total number of fragment shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 6 READ"
>+             mdapi_group="EU Array/Fragment Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="CS Threads Dispatched"
>+             symbol_name="CsThreads"
>+             underscore_name="cs_threads"
>+             description="The total number of compute shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 4 READ"
>+             mdapi_group="EU Array/Compute Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Active"
>+             symbol_name="EuActive"
>+             underscore_name="eu_active"
>+             description="The percentage of time in which the Execution Units were actively processing."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Stall"
>+             symbol_name="EuStall"
>+             underscore_name="eu_stall"
>+             description="The percentage of time in which the Execution Units were stalled."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Thread Occupancy"
>+             symbol_name="EuThreadOccupancy"
>+             underscore_name="eu_thread_occupancy"
>+             description="The percentage of time in which hardware threads occupied EUs."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Thread Header Ready on Slice1 DualSubslice0 Thread Dispatcher"
>+             symbol_name="ThreadHeader10Ready"
>+             underscore_name="thread_header10_ready"
>+             description="The percentage of time in which thread header is ready on slice1 dualsubslice0 thread dispatcher"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 64 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice1 DualSubslice0 Thread Dispatcher Port 0"
>+             symbol_name="ThreadHeader10ReadyPort0"
>+             underscore_name="thread_header10_ready_port0"
>+             description="The percentage of time in which thread header is ready on slice1 dualsubslice0 thread dispatcher port 0"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 64 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice1 DualSubslice0 Thread Dispatcher Port 1"
>+             symbol_name="ThreadHeader10ReadyPort1"
>+             underscore_name="thread_header10_ready_port1"
>+             description="The percentage of time in which thread header is ready on slice1 dualsubslice0 thread dispatcher port 1"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 64 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice1 DualSubslice0 Thread Dispatcher Port 2"
>+             symbol_name="ThreadHeader10ReadyPort2"
>+             underscore_name="thread_header10_ready_port2"
>+             description="The percentage of time in which thread header is ready on slice1 dualsubslice0 thread dispatcher port 2"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 5 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 64 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice1 DualSubslice0 Thread Dispatcher Port 3"
>+             symbol_name="ThreadHeader10ReadyPort3"
>+             underscore_name="thread_header10_ready_port3"
>+             description="The percentage of time in which thread header is ready on slice1 dualsubslice0 thread dispatcher port 3"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 4 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 64 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice1 DualSubslice1 Thread Dispatcher"
>+             symbol_name="ThreadHeader11Ready"
>+             underscore_name="thread_header11_ready"
>+             description="The percentage of time in which thread header is ready on slice1 dualsubslice1 thread dispatcher"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 128 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice1 DualSubslice1 Thread Dispatcher Port 0"
>+             symbol_name="ThreadHeader11ReadyPort0"
>+             underscore_name="thread_header11_ready_port0"
>+             description="The percentage of time in which thread header is ready on slice1 dualsubslice1 thread dispatcher port 0"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 3 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 128 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice1 DualSubslice1 Thread Dispatcher Port 1"
>+             symbol_name="ThreadHeader11ReadyPort1"
>+             underscore_name="thread_header11_ready_port1"
>+             description="The percentage of time in which thread header is ready on slice1 dualsubslice1 thread dispatcher port 1"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 2 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 128 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice1 DualSubslice1 Thread Dispatcher Port 2"
>+             symbol_name="ThreadHeader11ReadyPort2"
>+             underscore_name="thread_header11_ready_port2"
>+             description="The percentage of time in which thread header is ready on slice1 dualsubslice1 thread dispatcher port 2"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 1 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 128 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice1 DualSubslice1 Thread Dispatcher Port 3"
>+             symbol_name="ThreadHeader11ReadyPort3"
>+             underscore_name="thread_header11_ready_port3"
>+             description="The percentage of time in which thread header is ready on slice1 dualsubslice1 thread dispatcher port 3"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 128 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice1 DualSubslice2 Thread Dispatcher"
>+             symbol_name="ThreadHeader12Ready"
>+             underscore_name="thread_header12_ready"
>+             description="The percentage of time in which thread header is ready on slice1 dualsubslice2 thread dispatcher"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 2 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 256 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice1 DualSubslice2 Thread Dispatcher Port 0"
>+             symbol_name="ThreadHeader12ReadyPort0"
>+             underscore_name="thread_header12_ready_port0"
>+             description="The percentage of time in which thread header is ready on slice1 dualsubslice2 thread dispatcher port 0"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 7 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 256 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice1 DualSubslice2 Thread Dispatcher Port 1"
>+             symbol_name="ThreadHeader12ReadyPort1"
>+             underscore_name="thread_header12_ready_port1"
>+             description="The percentage of time in which thread header is ready on slice1 dualsubslice2 thread dispatcher port 1"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 6 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 256 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice1 DualSubslice2 Thread Dispatcher Port 2"
>+             symbol_name="ThreadHeader12ReadyPort2"
>+             underscore_name="thread_header12_ready_port2"
>+             description="The percentage of time in which thread header is ready on slice1 dualsubslice2 thread dispatcher port 2"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 5 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 256 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice1 DualSubslice2 Thread Dispatcher Port 3"
>+             symbol_name="ThreadHeader12ReadyPort3"
>+             underscore_name="thread_header12_ready_port3"
>+             description="The percentage of time in which thread header is ready on slice1 dualsubslice2 thread dispatcher port 3"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 4 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 256 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <counter name="Thread Header Ready on Slice1 DualSubslice3 Thread Dispatcher"
>+             symbol_name="ThreadHeader13Ready"
>+             underscore_name="thread_header13_ready"
>+             description="The percentage of time in which thread header is ready on slice1 dualsubslice3 thread dispatcher"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 3 READ 100 UMUL $GpuCoreClocks FDIV"
>+             availability="$DualSubsliceMask 512 AND"
>+             mdapi_group="GPU/Thread Dispatcher"
>+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="dualsubslice"
>+             />
>+    <register_config type="NOA">
>+        <register type="NOA" address="0x00000D04" value="0x00000200" />
>+        <register type="NOA" address="0x00009884" value="0x0000000C" />
>+        <register type="NOA" address="0x00009888" value="0x2B81001C" />
>+        <register type="NOA" address="0x00009888" value="0x2BA1001C" />
>+        <register type="NOA" address="0x00009888" value="0x2BC1001C" />
>+        <register type="NOA" address="0x00009888" value="0x2BE1001C" />
>+        <register type="NOA" address="0x00009888" value="0x07810300" />
>+        <register type="NOA" address="0x00009888" value="0x15810CF4" />
>+        <register type="NOA" address="0x00009888" value="0x17810C72" />
>+        <register type="NOA" address="0x00009888" value="0x09810000" />
>+        <register type="NOA" address="0x00009888" value="0x03810000" />
>+        <register type="NOA" address="0x00009888" value="0x0383AA00" />
>+        <register type="NOA" address="0x00009888" value="0x0184F000" />
>+        <register type="NOA" address="0x00009888" value="0x03865500" />
>+        <register type="NOA" address="0x00009888" value="0x03875500" />
>+        <register type="NOA" address="0x00009888" value="0x07A10300" />
>+        <register type="NOA" address="0x00009888" value="0x11A10CF4" />
>+        <register type="NOA" address="0x00009888" value="0x13A10C72" />
>+        <register type="NOA" address="0x00009888" value="0x09A10000" />
>+        <register type="NOA" address="0x00009888" value="0x03A10000" />
>+        <register type="NOA" address="0x00009888" value="0x03A300AA" />
>+        <register type="NOA" address="0x00009888" value="0x01A40F00" />
>+        <register type="NOA" address="0x00009888" value="0x03A60055" />
>+        <register type="NOA" address="0x00009888" value="0x03A70055" />
>+        <register type="NOA" address="0x00009888" value="0x07C10300" />
>+        <register type="NOA" address="0x00009888" value="0x0DC10CF4" />
>+        <register type="NOA" address="0x00009888" value="0x0FC10C72" />
>+        <register type="NOA" address="0x00009888" value="0x09C10000" />
>+        <register type="NOA" address="0x00009888" value="0x01C10000" />
>+        <register type="NOA" address="0x00009888" value="0x01C3AA00" />
>+        <register type="NOA" address="0x00009888" value="0x01C400F0" />
>+        <register type="NOA" address="0x00009888" value="0x01C65500" />
>+        <register type="NOA" address="0x00009888" value="0x01C75500" />
>+        <register type="NOA" address="0x00009888" value="0x07E10300" />
>+        <register type="NOA" address="0x00009888" value="0x09E10CB1" />
>+        <register type="NOA" address="0x00009888" value="0x0BE10D33" />
>+        <register type="NOA" address="0x00009888" value="0x01E10000" />
>+        <register type="NOA" address="0x00009888" value="0x01E300AA" />
>+        <register type="NOA" address="0x00009888" value="0x01E4000F" />
>+        <register type="NOA" address="0x00009888" value="0x01E60055" />
>+        <register type="NOA" address="0x00009888" value="0x01E70055" />
>+        <register type="NOA" address="0x00009888" value="0x0003FFFF" />
>+        <register type="NOA" address="0x00009888" value="0x0008FFFF" />
>+        <register type="NOA" address="0x00009888" value="0x0208FFFF" />
>+        <register type="NOA" address="0x00009888" value="0x000A5555" />
>+        <register type="NOA" address="0x00009888" value="0x020A5555" />
>+        <register type="NOA" address="0x00009888" value="0x0001FFFF" />
>+        <register type="NOA" address="0x00009888" value="0x000BAAAA" />
>+        <register type="NOA" address="0x00009888" value="0x020BAAAA" />
>+        <register type="NOA" address="0x00009888" value="0x0107AA55" />
>+        <register type="NOA" address="0x00009888" value="0x0308AAFF" />
>+        <register type="NOA" address="0x00009888" value="0x010900FF" />
>+        <register type="NOA" address="0x00009888" value="0x010AAAAA" />
>+        <register type="NOA" address="0x00009888" value="0x030A5555" />
>+        <register type="NOA" address="0x00009888" value="0x01010F00" />
>+        <register type="NOA" address="0x00009888" value="0x030B0055" />
>+        <register type="NOA" address="0x00009888" value="0x010200F0" />
>+        <register type="NOA" address="0x00009888" value="0x010C5500" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
>+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
>+        <register type="NOA" address="0x00009888" value="0x00032492" />
>+        <register type="NOA" address="0x00009888" value="0x02032492" />
>+        <register type="NOA" address="0x00009888" value="0x04032492" />
>+        <register type="NOA" address="0x00009888" value="0x06030002" />
>+        <register type="NOA" address="0x00009888" value="0x14030000" />
>+        <register type="NOA" address="0x00009888" value="0x3E030C6A" />
>+        <register type="NOA" address="0x00009888" value="0x26030000" />
>+        <register type="NOA" address="0x00009888" value="0x42032942" />
>+        <register type="NOA" address="0x00009888" value="0x2C030000" />
>+        <register type="NOA" address="0x00009888" value="0x32030000" />
>+        <register type="NOA" address="0x00009888" value="0x4403000A" />
>+        <register type="NOA" address="0x00009888" value="0x38030000" />
>+        <register type="NOA" address="0x00009888" value="0x46030000" />
>+        <register type="NOA" address="0x00009888" value="0x48030000" />
>+        <register type="NOA" address="0x00009888" value="0x4A030603" />
>+        <register type="NOA" address="0x00009888" value="0x16030000" />
>+        <register type="NOA" address="0x00009888" value="0x18030000" />
>+        <register type="NOA" address="0x00009888" value="0x1A031000" />
>+        <register type="NOA" address="0x00009888" value="0x40030842" />
>+        <register type="NOA" address="0x00009888" value="0x1C031000" />
>+        <register type="NOA" address="0x00009888" value="0x1E031000" />
>+        <register type="NOA" address="0x00009888" value="0x20031000" />
>+        <register type="NOA" address="0x00009884" value="0x00000008" />
>+        <register type="NOA" address="0x00009888" value="0x0001FFFF" />
>+        <register type="NOA" address="0x00009888" value="0x00025555" />
>+        <register type="NOA" address="0x00009888" value="0x02025555" />
>+        <register type="NOA" address="0x00009888" value="0x00045555" />
>+        <register type="NOA" address="0x00009888" value="0x02045555" />
>+        <register type="NOA" address="0x00009888" value="0x00055555" />
>+        <register type="NOA" address="0x00009888" value="0x02055555" />
>+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
>+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x5C030002" />
>+    </register_config>
>+    <register_config type="OA">
>+        <register type="OA" address="0x0000D920" value="0x00000000" />
>+        <register type="OA" address="0x0000D900" value="0x00000000" />
>+        <register type="OA" address="0x0000D904" value="0xF0800000" />
>+        <register type="OA" address="0x0000D910" value="0x00000000" />
>+        <register type="OA" address="0x0000D914" value="0x00800000" />
>+        <register type="OA" address="0x0000DC40" value="0x000F0000" />
>+        <register type="OA" address="0x0000D940" value="0x00078000" />
>+        <register type="OA" address="0x0000D944" value="0x00000FFF" />
>+        <register type="OA" address="0x0000DC00" value="0x00078000" />
>+        <register type="OA" address="0x0000DC04" value="0x00000FFF" />
>+        <register type="OA" address="0x0000D948" value="0x00007800" />
>+        <register type="OA" address="0x0000D94C" value="0x0000F0FF" />
>+        <register type="OA" address="0x0000DC08" value="0x00007800" />
>+        <register type="OA" address="0x0000DC0C" value="0x0000F0FF" />
>+        <register type="OA" address="0x0000D950" value="0x00000780" />
>+        <register type="OA" address="0x0000D954" value="0x0000FF0F" />
>+        <register type="OA" address="0x0000DC10" value="0x00000780" />
>+        <register type="OA" address="0x0000DC14" value="0x0000FF0F" />
>+        <register type="OA" address="0x0000D958" value="0x00000078" />
>+        <register type="OA" address="0x0000D95C" value="0x0000FFF0" />
>+        <register type="OA" address="0x0000DC18" value="0x00000078" />
>+        <register type="OA" address="0x0000DC1C" value="0x0000FFF0" />
>+    </register_config>
>+    <register_config type="FLEX">
>+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
>+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
>+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
>+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
>+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
>+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
>+    </register_config>
>+  </set>
>+
>+  <set name="L3_1"
>+       chipset="DG2"
>+       symbol_name="L3_1"
>+       underscore_name="l3_1"
>+       mdapi_supported_apis="VK OGL OCL MEDIA IO"
>+       hw_config_guid="9a1a14ad-ab92-4204-8a91-c4159a54f712"
>+       >
>+    <counter name="GPU Time Elapsed"
>+             symbol_name="GpuTime"
>+             underscore_name="gpu_time"
>+             description="Time elapsed on the GPU during the measurement."
>+             data_type="uint64"
>+             units="ns"
>+             semantic_type="duration"
>+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Core Clocks"
>+             symbol_name="GpuCoreClocks"
>+             underscore_name="gpu_core_clocks"
>+             description="The total number of GPU core clocks elapsed during the measurement."
>+             data_type="uint64"
>+             units="cycles"
>+             semantic_type="event"
>+             equation="GPU_CLOCK 0 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="AVG GPU Core Frequency"
>+             symbol_name="AvgGpuCoreFrequency"
>+             underscore_name="avg_gpu_core_frequency"
>+             description="Average GPU Core Frequency in the measurement."
>+             data_type="uint64"
>+             max_equation="$GpuMaxFrequency"
>+             units="hz"
>+             semantic_type="event"
>+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Busy"
>+             symbol_name="GpuBusy"
>+             underscore_name="gpu_busy"
>+             description="The percentage of time in which the GPU has been processing GPU commands."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Active"
>+             symbol_name="EuActive"
>+             underscore_name="eu_active"
>+             description="The percentage of time in which the Execution Units were actively processing."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Stall"
>+             symbol_name="EuStall"
>+             underscore_name="eu_stall"
>+             description="The percentage of time in which the Execution Units were stalled."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Thread Occupancy"
>+             symbol_name="EuThreadOccupancy"
>+             underscore_name="eu_thread_occupancy"
>+             description="The percentage of time in which hardware threads occupied EUs."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="VS Threads Dispatched"
>+             symbol_name="VsThreads"
>+             underscore_name="vs_threads"
>+             description="The total number of vertex shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 1 READ"
>+             mdapi_group="EU Array/Vertex Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="HS Threads Dispatched"
>+             symbol_name="HsThreads"
>+             underscore_name="hs_threads"
>+             description="The total number of hull shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 2 READ"
>+             mdapi_group="EU Array/Hull Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="DS Threads Dispatched"
>+             symbol_name="DsThreads"
>+             underscore_name="ds_threads"
>+             description="The total number of domain shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 3 READ"
>+             mdapi_group="EU Array/Domain Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GS Threads Dispatched"
>+             symbol_name="GsThreads"
>+             underscore_name="gs_threads"
>+             description="The total number of geometry shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 5 READ"
>+             mdapi_group="EU Array/Geometry Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="FS Threads Dispatched"
>+             symbol_name="PsThreads"
>+             underscore_name="ps_threads"
>+             description="The total number of fragment shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 6 READ"
>+             mdapi_group="EU Array/Fragment Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="CS Threads Dispatched"
>+             symbol_name="CsThreads"
>+             underscore_name="cs_threads"
>+             description="The total number of compute shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 4 READ"
>+             mdapi_group="EU Array/Compute Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Quad0 L3 Bank0 Input Available"
>+             symbol_name="L30Bank0InputAvailable"
>+             underscore_name="l30_bank0_input_available"
>+             description="The percentage of time in which Quad0 L3 bank0 has input available"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GTI/L3"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="slice"
>+             />
>+    <counter name="Quad0 L3 Bank1 Input Available"
>+             symbol_name="L30Bank1InputAvailable"
>+             underscore_name="l30_bank1_input_available"
>+             description="The percentage of time in which Quad0 L3 bank1 has input available"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GTI/L3"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="slice"
>+             />
>+    <counter name="Quad0 L3 Bank2 Input Available"
>+             symbol_name="L30Bank2InputAvailable"
>+             underscore_name="l30_bank2_input_available"
>+             description="The percentage of time in which Quad0 L3 bank2 has input available"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 5 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GTI/L3"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="slice"
>+             />
>+    <counter name="Quad0 L3 Bank3 Input Available"
>+             symbol_name="L30Bank3InputAvailable"
>+             underscore_name="l30_bank3_input_available"
>+             description="The percentage of time in which Quad0 L3 bank3 has input available"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 4 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GTI/L3"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="slice"
>+             />
>+    <counter name="Quad0 L3 Bank4 Input Available"
>+             symbol_name="L30Bank4InputAvailable"
>+             underscore_name="l30_bank4_input_available"
>+             description="The percentage of time in which Quad0 L3 bank4 has input available"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 3 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GTI/L3"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="slice"
>+             />
>+    <counter name="Quad0 L3 Bank5 Input Available"
>+             symbol_name="L30Bank5InputAvailable"
>+             underscore_name="l30_bank5_input_available"
>+             description="The percentage of time in which Quad0 L3 bank5 has input available"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 2 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GTI/L3"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="slice"
>+             />
>+    <counter name="Quad0 L3 Bank6 Input Available"
>+             symbol_name="L30Bank6InputAvailable"
>+             underscore_name="l30_bank6_input_available"
>+             description="The percentage of time in which Quad0 L3 bank6 has input available"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 1 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GTI/L3"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="slice"
>+             />
>+    <counter name="Quad0 L3 Bank7 Input Available"
>+             symbol_name="L30Bank7InputAvailable"
>+             underscore_name="l30_bank7_input_available"
>+             description="The percentage of time in which Quad0 L3 bank7 has input available"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GTI/L3"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="slice"
>+             />
>+    <counter name="Quad0 L3 Bank0 Output Ready"
>+             symbol_name="L30Bank0OutputReady"
>+             underscore_name="l30_bank0_output_ready"
>+             description="The percentage of time in which quad0 L3 bank0 output is ready"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 7 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GTI/L3"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="slice"
>+             />
>+    <counter name="Quad0 L3 Bank1 Output Ready"
>+             symbol_name="L30Bank1OutputReady"
>+             underscore_name="l30_bank1_output_ready"
>+             description="The percentage of time in which quad0 L3 bank1 output is ready"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 6 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GTI/L3"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="slice"
>+             />
>+    <counter name="Quad0 L3 Bank2 Output Ready"
>+             symbol_name="L30Bank2OutputReady"
>+             underscore_name="l30_bank2_output_ready"
>+             description="The percentage of time in which quad0 L3 bank2 output is ready"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 5 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GTI/L3"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="slice"
>+             />
>+    <counter name="Quad0 L3 Bank3 Output Ready"
>+             symbol_name="L30Bank3OutputReady"
>+             underscore_name="l30_bank3_output_ready"
>+             description="The percentage of time in which quad0 L3 bank3 output is ready"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 4 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GTI/L3"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="slice"
>+             />
>+    <counter name="Quad0 L3 Bank4 Output Ready"
>+             symbol_name="L30Bank4OutputReady"
>+             underscore_name="l30_bank4_output_ready"
>+             description="The percentage of time in which quad0 L3 bank4 output is ready"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 3 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GTI/L3"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="slice"
>+             />
>+    <counter name="Quad0 L3 Bank5 Output Ready"
>+             symbol_name="L30Bank5OutputReady"
>+             underscore_name="l30_bank5_output_ready"
>+             description="The percentage of time in which quad0 L3 bank5 output is ready"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 2 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GTI/L3"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="slice"
>+             />
>+    <counter name="Quad0 L3 Bank6 Output Ready"
>+             symbol_name="L30Bank6OutputReady"
>+             underscore_name="l30_bank6_output_ready"
>+             description="The percentage of time in which quad0 L3 bank6 output is ready"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GTI/L3"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="slice"
>+             />
>+    <counter name="Quad0 L3 Bank7 Output Ready"
>+             symbol_name="L30Bank7OutputReady"
>+             underscore_name="l30_bank7_output_ready"
>+             description="The percentage of time in which quad0 L3 bank7 output is ready"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GTI/L3"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="slice"
>+             />
>+    <register_config type="NOA">
>+        <register type="NOA" address="0x00000D04" value="0x00000200" />
>+        <register type="NOA" address="0x00009884" value="0x00000008" />
>+        <register type="NOA" address="0x00009888" value="0x090B1000" />
>+        <register type="NOA" address="0x00009888" value="0x092B1006" />
>+        <register type="NOA" address="0x00009888" value="0x094B1000" />
>+        <register type="NOA" address="0x00009888" value="0x096B1000" />
>+        <register type="NOA" address="0x00009888" value="0x090C1000" />
>+        <register type="NOA" address="0x00009888" value="0x092C1180" />
>+        <register type="NOA" address="0x00009888" value="0x094C1000" />
>+        <register type="NOA" address="0x00009888" value="0x096C1000" />
>+        <register type="NOA" address="0x00009888" value="0x010B0002" />
>+        <register type="NOA" address="0x00009888" value="0x0B0B0006" />
>+        <register type="NOA" address="0x00009888" value="0x130B0007" />
>+        <register type="NOA" address="0x00009888" value="0x030B0000" />
>+        <register type="NOA" address="0x00009888" value="0x012B0002" />
>+        <register type="NOA" address="0x00009888" value="0x112B0007" />
>+        <register type="NOA" address="0x00009888" value="0x032B0000" />
>+        <register type="NOA" address="0x00009888" value="0x014B0002" />
>+        <register type="NOA" address="0x00009888" value="0x074B0006" />
>+        <register type="NOA" address="0x00009888" value="0x0F4B0007" />
>+        <register type="NOA" address="0x00009888" value="0x034B0000" />
>+        <register type="NOA" address="0x00009888" value="0x016B0002" />
>+        <register type="NOA" address="0x00009888" value="0x056B0006" />
>+        <register type="NOA" address="0x00009888" value="0x0D6B0007" />
>+        <register type="NOA" address="0x00009888" value="0x036B0000" />
>+        <register type="NOA" address="0x00009888" value="0x010C0002" />
>+        <register type="NOA" address="0x00009888" value="0x0B0C0180" />
>+        <register type="NOA" address="0x00009888" value="0x130C01C0" />
>+        <register type="NOA" address="0x00009888" value="0x030C0000" />
>+        <register type="NOA" address="0x00009888" value="0x012C0002" />
>+        <register type="NOA" address="0x00009888" value="0x112C01C0" />
>+        <register type="NOA" address="0x00009888" value="0x032C0000" />
>+        <register type="NOA" address="0x00009888" value="0x014C0002" />
>+        <register type="NOA" address="0x00009888" value="0x074C0180" />
>+        <register type="NOA" address="0x00009888" value="0x0F4C01C0" />
>+        <register type="NOA" address="0x00009888" value="0x034C0000" />
>+        <register type="NOA" address="0x00009888" value="0x016C0002" />
>+        <register type="NOA" address="0x00009888" value="0x056C0180" />
>+        <register type="NOA" address="0x00009888" value="0x0D6C01C0" />
>+        <register type="NOA" address="0x00009888" value="0x036C0000" />
>+        <register type="NOA" address="0x00009888" value="0x00020030" />
>+        <register type="NOA" address="0x00009888" value="0x02020030" />
>+        <register type="NOA" address="0x00009888" value="0x00030048" />
>+        <register type="NOA" address="0x00009888" value="0x02030048" />
>+        <register type="NOA" address="0x00009888" value="0x0004009B" />
>+        <register type="NOA" address="0x00009888" value="0x0204009B" />
>+        <register type="NOA" address="0x00009888" value="0x00052055" />
>+        <register type="NOA" address="0x00009888" value="0x02052055" />
>+        <register type="NOA" address="0x00009888" value="0x00064800" />
>+        <register type="NOA" address="0x00009888" value="0x02064800" />
>+        <register type="NOA" address="0x00009888" value="0x0007A0A0" />
>+        <register type="NOA" address="0x00009888" value="0x000867AA" />
>+        <register type="NOA" address="0x00009888" value="0x020867AA" />
>+        <register type="NOA" address="0x00009888" value="0x0009A0A0" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
>+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
>+        <register type="NOA" address="0x00009888" value="0x00032492" />
>+        <register type="NOA" address="0x00009888" value="0x02032492" />
>+        <register type="NOA" address="0x00009888" value="0x04032492" />
>+        <register type="NOA" address="0x00009888" value="0x06030002" />
>+        <register type="NOA" address="0x00009888" value="0x14030000" />
>+        <register type="NOA" address="0x00009888" value="0x3E030024" />
>+        <register type="NOA" address="0x00009888" value="0x26030000" />
>+        <register type="NOA" address="0x00009888" value="0x42030401" />
>+        <register type="NOA" address="0x00009888" value="0x2C030000" />
>+        <register type="NOA" address="0x00009888" value="0x32030000" />
>+        <register type="NOA" address="0x00009888" value="0x44030140" />
>+        <register type="NOA" address="0x00009888" value="0x38030000" />
>+        <register type="NOA" address="0x00009888" value="0x46030E01" />
>+        <register type="NOA" address="0x00009888" value="0x48030010" />
>+        <register type="NOA" address="0x00009888" value="0x4A030004" />
>+        <register type="NOA" address="0x00009888" value="0x16030000" />
>+        <register type="NOA" address="0x00009888" value="0x18030000" />
>+        <register type="NOA" address="0x00009888" value="0x1A030000" />
>+        <register type="NOA" address="0x00009888" value="0x40031C2A" />
>+        <register type="NOA" address="0x00009888" value="0x1C030000" />
>+        <register type="NOA" address="0x00009888" value="0x1E030000" />
>+        <register type="NOA" address="0x00009888" value="0x20030000" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x5C030002" />
>+    </register_config>
>+    <register_config type="OA">
>+        <register type="OA" address="0x0000D920" value="0x00000000" />
>+        <register type="OA" address="0x0000D900" value="0x00000000" />
>+        <register type="OA" address="0x0000D904" value="0x00800000" />
>+        <register type="OA" address="0x0000D910" value="0x00000000" />
>+        <register type="OA" address="0x0000D914" value="0x00800000" />
>+        <register type="OA" address="0x0000DC40" value="0x00000000" />
>+    </register_config>
>+    <register_config type="FLEX">
>+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
>+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
>+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
>+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
>+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
>+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
>+    </register_config>
>+  </set>
>+
>+  <set name="GpuBusyness"
>+       chipset="DG2"
>+       symbol_name="GpuBusyness"
>+       underscore_name="gpu_busyness"
>+       mdapi_supported_apis="VK OGL OCL MEDIA IO"
>+       hw_config_guid="85de4c29-76b8-42a6-b6ef-8a2bed008fcf"
>+       >
>+    <counter name="GPU Time Elapsed"
>+             symbol_name="GpuTime"
>+             underscore_name="gpu_time"
>+             description="Time elapsed on the GPU during the measurement."
>+             data_type="uint64"
>+             units="ns"
>+             semantic_type="duration"
>+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Core Clocks"
>+             symbol_name="GpuCoreClocks"
>+             underscore_name="gpu_core_clocks"
>+             description="The total number of GPU core clocks elapsed during the measurement."
>+             data_type="uint64"
>+             units="cycles"
>+             semantic_type="event"
>+             equation="GPU_CLOCK 0 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="AVG GPU Core Frequency"
>+             symbol_name="AvgGpuCoreFrequency"
>+             underscore_name="avg_gpu_core_frequency"
>+             description="Average GPU Core Frequency in the measurement."
>+             data_type="uint64"
>+             max_equation="$GpuMaxFrequency"
>+             units="hz"
>+             semantic_type="event"
>+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Busy"
>+             symbol_name="GpuBusy"
>+             underscore_name="gpu_busy"
>+             description="The percentage of time in which the GPU has been processing GPU commands."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Active"
>+             symbol_name="EuActive"
>+             underscore_name="eu_active"
>+             description="The percentage of time in which the Execution Units were actively processing."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Stall"
>+             symbol_name="EuStall"
>+             underscore_name="eu_stall"
>+             description="The percentage of time in which the Execution Units were stalled."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU Thread Occupancy"
>+             symbol_name="EuThreadOccupancy"
>+             underscore_name="eu_thread_occupancy"
>+             description="The percentage of time in which hardware threads occupied EUs."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="VS Threads Dispatched"
>+             symbol_name="VsThreads"
>+             underscore_name="vs_threads"
>+             description="The total number of vertex shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 1 READ"
>+             mdapi_group="EU Array/Vertex Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="HS Threads Dispatched"
>+             symbol_name="HsThreads"
>+             underscore_name="hs_threads"
>+             description="The total number of hull shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 2 READ"
>+             mdapi_group="EU Array/Hull Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="DS Threads Dispatched"
>+             symbol_name="DsThreads"
>+             underscore_name="ds_threads"
>+             description="The total number of domain shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 3 READ"
>+             mdapi_group="EU Array/Domain Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GS Threads Dispatched"
>+             symbol_name="GsThreads"
>+             underscore_name="gs_threads"
>+             description="The total number of geometry shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 5 READ"
>+             mdapi_group="EU Array/Geometry Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="FS Threads Dispatched"
>+             symbol_name="PsThreads"
>+             underscore_name="ps_threads"
>+             description="The total number of fragment shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 6 READ"
>+             mdapi_group="EU Array/Fragment Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="CS Threads Dispatched"
>+             symbol_name="CsThreads"
>+             underscore_name="cs_threads"
>+             description="The total number of compute shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 4 READ"
>+             mdapi_group="EU Array/Compute Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Render Ring Busy"
>+             symbol_name="RenderBusy"
>+             underscore_name="render_busy"
>+             description="The percentage of time when render command streamer was busy."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="System Frame Batch"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="ComputeEngine0 Ring Busy"
>+             symbol_name="ComputeEngine0Busy"
>+             underscore_name="compute_engine0_busy"
>+             description="The percentage of time when compute command streamer 0 was busy."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 3 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="System Frame Batch"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="ComputeEngine1 Ring Busy"
>+             symbol_name="ComputeEngine1Busy"
>+             underscore_name="compute_engine1_busy"
>+             description="The percentage of time when compute command streamer 1 was busy."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 4 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="System Frame Batch"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="ComputeEngine2 Ring Busy"
>+             symbol_name="ComputeEngine2Busy"
>+             underscore_name="compute_engine2_busy"
>+             description="The percentage of time when compute command streamer 2 was busy."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 5 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="System Frame Batch"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="ComputeEngine3 Ring Busy"
>+             symbol_name="ComputeEngine3Busy"
>+             underscore_name="compute_engine3_busy"
>+             description="The percentage of time when compute command streamer 3 was busy."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="System Frame Batch"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Posh Ring Busy"
>+             symbol_name="PoshEngineBusy"
>+             underscore_name="posh_engine_busy"
>+             description="The percentage of time when posh command streamer was busy."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 2 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="System Frame Batch"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Blitter Ring Busy"
>+             symbol_name="BlitterBusy"
>+             underscore_name="blitter_busy"
>+             description="The percentage of time when blitter command streamer was busy."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 1 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="System Frame Batch"
>+             mdapi_supported_apis="IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Vdbox00 Ring Busy"
>+             symbol_name="Vdbox00Busy"
>+             underscore_name="vdbox00_busy"
>+             description="The percentage of time when vdbox00 command streamer was busy."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="C 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="System Frame Batch"
>+             mdapi_supported_apis="IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Vebox0 Ring Busy"
>+             symbol_name="Vebox0Busy"
>+             underscore_name="vebox0_busy"
>+             description="The percentage of time when vebox0 command streamer was busy."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 7 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="System Frame Batch"
>+             mdapi_supported_apis="IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Vdbox10 Ring Busy"
>+             symbol_name="Vdbox10Busy"
>+             underscore_name="vdbox10_busy"
>+             description="The percentage of time when vdbox10 command streamer was busy."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 6 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="System Frame Batch"
>+             mdapi_supported_apis="IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Vebox1 Ring Busy"
>+             symbol_name="Vebox1Busy"
>+             underscore_name="vebox1_busy"
>+             description="The percentage of time when vebox1 command streamer was busy."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 5 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="System Frame Batch"
>+             mdapi_supported_apis="IO"
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Any compute engine busy"
>+             symbol_name="AnyComputeEngineBusy"
>+             underscore_name="any_compute_engine_busy"
>+             description="The percentage of time when any compute engine is busy"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Render and any compute engines are simultaneously busy"
>+             symbol_name="RenderAndAnyComputeEngineBusy"
>+             underscore_name="render_and_any_compute_engine_busy"
>+             description="The percentage of time when render and any compute engines are simultaneously busy"
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <register_config type="NOA">
>+        <register type="NOA" address="0x00000D04" value="0x00000200" />
>+        <register type="NOA" address="0x00009884" value="0x00000004" />
>+        <register type="NOA" address="0x00009888" value="0x420A2400" />
>+        <register type="NOA" address="0x00009888" value="0x1A131200" />
>+        <register type="NOA" address="0x00009888" value="0x040A0022" />
>+        <register type="NOA" address="0x00009888" value="0x120A0014" />
>+        <register type="NOA" address="0x00009888" value="0x020A0000" />
>+        <register type="NOA" address="0x00009888" value="0x000A0000" />
>+        <register type="NOA" address="0x00009888" value="0x00104000" />
>+        <register type="NOA" address="0x00009888" value="0x02100002" />
>+        <register type="NOA" address="0x00009888" value="0x00124000" />
>+        <register type="NOA" address="0x00009888" value="0x00134000" />
>+        <register type="NOA" address="0x00009888" value="0x0A130024" />
>+        <register type="NOA" address="0x00009884" value="0x00000005" />
>+        <register type="NOA" address="0x00009888" value="0x420A2400" />
>+        <register type="NOA" address="0x00009888" value="0x1A131200" />
>+        <register type="NOA" address="0x00009888" value="0x040A0022" />
>+        <register type="NOA" address="0x00009888" value="0x100A0014" />
>+        <register type="NOA" address="0x00009888" value="0x020A0000" />
>+        <register type="NOA" address="0x00009888" value="0x000A0000" />
>+        <register type="NOA" address="0x00009888" value="0x00102400" />
>+        <register type="NOA" address="0x00009888" value="0x00120400" />
>+        <register type="NOA" address="0x00009888" value="0x00134000" />
>+        <register type="NOA" address="0x00009888" value="0x08130024" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x18012400" />
>+        <register type="NOA" address="0x00009888" value="0x18042400" />
>+        <register type="NOA" address="0x00009888" value="0x20042400" />
>+        <register type="NOA" address="0x00009888" value="0x28042400" />
>+        <register type="NOA" address="0x00009888" value="0x32040009" />
>+        <register type="NOA" address="0x00009888" value="0x3C041200" />
>+        <register type="NOA" address="0x00009888" value="0x4C041200" />
>+        <register type="NOA" address="0x00009888" value="0x02010006" />
>+        <register type="NOA" address="0x00009888" value="0x10010004" />
>+        <register type="NOA" address="0x00009888" value="0x0001FC00" />
>+        <register type="NOA" address="0x00009888" value="0x00042EE9" />
>+        <register type="NOA" address="0x00009888" value="0x02040239" />
>+        <register type="NOA" address="0x00009888" value="0x0404000F" />
>+        <register type="NOA" address="0x00009888" value="0x0E040D00" />
>+        <register type="NOA" address="0x00009888" value="0x10040304" />
>+        <register type="NOA" address="0x00009888" value="0x12040714" />
>+        <register type="NOA" address="0x00009888" value="0x14040024" />
>+        <register type="NOA" address="0x00009888" value="0x02085554" />
>+        <register type="NOA" address="0x00009888" value="0x00056800" />
>+        <register type="NOA" address="0x00009888" value="0x02050001" />
>+        <register type="NOA" address="0x00009888" value="0x02032B6D" />
>+        <register type="NOA" address="0x00009888" value="0x04032492" />
>+        <register type="NOA" address="0x00009888" value="0x06030002" />
>+        <register type="NOA" address="0x00009888" value="0x44030000" />
>+        <register type="NOA" address="0x00009888" value="0x46030400" />
>+        <register type="NOA" address="0x00009888" value="0x48030000" />
>+        <register type="NOA" address="0x00009888" value="0x4A030C62" />
>+        <register type="NOA" address="0x00009888" value="0x16033000" />
>+        <register type="NOA" address="0x00009888" value="0x3E031080" />
>+        <register type="NOA" address="0x00009888" value="0x18033000" />
>+        <register type="NOA" address="0x00009888" value="0x1A033000" />
>+        <register type="NOA" address="0x00009888" value="0x40031084" />
>+        <register type="NOA" address="0x00009888" value="0x1C033000" />
>+        <register type="NOA" address="0x00009888" value="0x1E033000" />
>+        <register type="NOA" address="0x00009888" value="0x20033000" />
>+        <register type="NOA" address="0x00009888" value="0x42030004" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x5C030002" />
>+    </register_config>
>+    <register_config type="OA">
>+        <register type="OA" address="0x0000D920" value="0x00000000" />
>+        <register type="OA" address="0x0000D900" value="0x00000000" />
>+        <register type="OA" address="0x0000D904" value="0x30800000" />
>+        <register type="OA" address="0x0000D910" value="0x00000000" />
>+        <register type="OA" address="0x0000D914" value="0x00800000" />
>+        <register type="OA" address="0x0000DC40" value="0x00030000" />
>+        <register type="OA" address="0x0000D940" value="0x00000000" />
>+        <register type="OA" address="0x0000D944" value="0x000087FF" />
>+        <register type="OA" address="0x0000DC00" value="0x00000000" />
>+        <register type="OA" address="0x0000DC04" value="0x000087FF" />
>+        <register type="OA" address="0x0000D948" value="0x00040001" />
>+        <register type="OA" address="0x0000D94C" value="0x000007FF" />
>+        <register type="OA" address="0x0000DC08" value="0x00040001" />
>+        <register type="OA" address="0x0000DC0C" value="0x000007FF" />
>+    </register_config>
>+    <register_config type="FLEX">
>+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
>+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
>+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
>+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
>+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
>+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
>+    </register_config>
>+  </set>
>+
>+  <set name="EuActivity1"
>+       chipset="DG2"
>+       symbol_name="EuActivity1"
>+       underscore_name="eu_activity1"
>+       mdapi_supported_apis="OGL OGL4 OCL MEDIA IO"
>+       hw_config_guid="8dc7142a-df10-41f2-acaa-9faea547ce73"
>+       >
>+    <counter name="GPU Time Elapsed"
>+             symbol_name="GpuTime"
>+             underscore_name="gpu_time"
>+             description="Time elapsed on the GPU during the measurement."
>+             data_type="uint64"
>+             units="ns"
>+             semantic_type="duration"
>+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Core Clocks"
>+             symbol_name="GpuCoreClocks"
>+             underscore_name="gpu_core_clocks"
>+             description="The total number of GPU core clocks elapsed during the measurement."
>+             data_type="uint64"
>+             units="cycles"
>+             semantic_type="event"
>+             equation="GPU_CLOCK 0 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="AVG GPU Core Frequency"
>+             symbol_name="AvgGpuCoreFrequency"
>+             underscore_name="avg_gpu_core_frequency"
>+             description="Average GPU Core Frequency in the measurement."
>+             data_type="uint64"
>+             max_equation="$GpuMaxFrequency"
>+             units="hz"
>+             semantic_type="event"
>+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Busy"
>+             symbol_name="GpuBusy"
>+             underscore_name="gpu_busy"
>+             description="The percentage of time in which the GPU has been processing GPU commands."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="VS Threads Dispatched"
>+             symbol_name="VsThreads"
>+             underscore_name="vs_threads"
>+             description="The total number of vertex shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 1 READ"
>+             mdapi_group="EU Array/Vertex Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="HS Threads Dispatched"
>+             symbol_name="HsThreads"
>+             underscore_name="hs_threads"
>+             description="The total number of hull shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 2 READ"
>+             mdapi_group="EU Array/Hull Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="DS Threads Dispatched"
>+             symbol_name="DsThreads"
>+             underscore_name="ds_threads"
>+             description="The total number of domain shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 3 READ"
>+             mdapi_group="EU Array/Domain Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GS Threads Dispatched"
>+             symbol_name="GsThreads"
>+             underscore_name="gs_threads"
>+             description="The total number of geometry shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 5 READ"
>+             mdapi_group="EU Array/Geometry Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="FS Threads Dispatched"
>+             symbol_name="PsThreads"
>+             underscore_name="ps_threads"
>+             description="The total number of fragment shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 6 READ"
>+             mdapi_group="EU Array/Fragment Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="CS Threads Dispatched"
>+             symbol_name="CsThreads"
>+             underscore_name="cs_threads"
>+             description="The total number of compute shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 4 READ"
>+             mdapi_group="EU Array/Compute Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="VS FPU Pipe Active"
>+             symbol_name="VsFpuActive"
>+             underscore_name="vs_fpu_active"
>+             description="The percentage of time in which EU FPU INT64 pipeline was actively processing a vertex shader instruction."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array/Vertex Shader"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="FS FPU Pipe Active"
>+             symbol_name="PsFpuActive"
>+             underscore_name="ps_fpu_active"
>+             description="The percentage of time in which EU FPU INT64 pipeline was actively processing a fragment shader instruction."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array/Fragment Shader"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="Send Pipe Active"
>+             symbol_name="SendActive"
>+             underscore_name="send_active"
>+             description="The percentage of time in which EU send pipeline was actively processing."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array/Pipes"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <register_config type="OA">
>+        <register type="OA" address="0x0000D920" value="0x00000000" />
>+        <register type="OA" address="0x0000D900" value="0x00000000" />
>+        <register type="OA" address="0x0000D904" value="0x00800000" />
>+        <register type="OA" address="0x0000D910" value="0x00000000" />
>+        <register type="OA" address="0x0000D914" value="0x00800000" />
>+        <register type="OA" address="0x0000DC40" value="0x00000000" />
>+    </register_config>
>+    <register_config type="FLEX">
>+        <register type="FLEX" address="0x0000E458" value="0x00810710" />
>+        <register type="FLEX" address="0x0000E558" value="0x00A10910" />
>+        <register type="FLEX" address="0x0000E658" value="0x00850750" />
>+        <register type="FLEX" address="0x0000E758" value="0x00A50950" />
>+        <register type="FLEX" address="0x0000E45C" value="0x00802702" />
>+        <register type="FLEX" address="0x0000E55C" value="0x00A02902" />
>+    </register_config>
>+  </set>
>+
>+  <set name="EuActivity2"
>+       chipset="DG2"
>+       symbol_name="EuActivity2"
>+       underscore_name="eu_activity2"
>+       mdapi_supported_apis="OGL OGL4 OCL MEDIA IO"
>+       hw_config_guid="d3c05cce-4ae7-4291-82ab-b0bcade6cf7a"
>+       >
>+    <counter name="GPU Time Elapsed"
>+             symbol_name="GpuTime"
>+             underscore_name="gpu_time"
>+             description="Time elapsed on the GPU during the measurement."
>+             data_type="uint64"
>+             units="ns"
>+             semantic_type="duration"
>+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Core Clocks"
>+             symbol_name="GpuCoreClocks"
>+             underscore_name="gpu_core_clocks"
>+             description="The total number of GPU core clocks elapsed during the measurement."
>+             data_type="uint64"
>+             units="cycles"
>+             semantic_type="event"
>+             equation="GPU_CLOCK 0 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="AVG GPU Core Frequency"
>+             symbol_name="AvgGpuCoreFrequency"
>+             underscore_name="avg_gpu_core_frequency"
>+             description="Average GPU Core Frequency in the measurement."
>+             data_type="uint64"
>+             max_equation="$GpuMaxFrequency"
>+             units="hz"
>+             semantic_type="event"
>+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Busy"
>+             symbol_name="GpuBusy"
>+             underscore_name="gpu_busy"
>+             description="The percentage of time in which the GPU has been processing GPU commands."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="VS Threads Dispatched"
>+             symbol_name="VsThreads"
>+             underscore_name="vs_threads"
>+             description="The total number of vertex shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 1 READ"
>+             mdapi_group="EU Array/Vertex Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="HS Threads Dispatched"
>+             symbol_name="HsThreads"
>+             underscore_name="hs_threads"
>+             description="The total number of hull shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 2 READ"
>+             mdapi_group="EU Array/Hull Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="DS Threads Dispatched"
>+             symbol_name="DsThreads"
>+             underscore_name="ds_threads"
>+             description="The total number of domain shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 3 READ"
>+             mdapi_group="EU Array/Domain Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GS Threads Dispatched"
>+             symbol_name="GsThreads"
>+             underscore_name="gs_threads"
>+             description="The total number of geometry shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 5 READ"
>+             mdapi_group="EU Array/Geometry Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="FS Threads Dispatched"
>+             symbol_name="PsThreads"
>+             underscore_name="ps_threads"
>+             description="The total number of fragment shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 6 READ"
>+             mdapi_group="EU Array/Fragment Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="CS Threads Dispatched"
>+             symbol_name="CsThreads"
>+             underscore_name="cs_threads"
>+             description="The total number of compute shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 4 READ"
>+             mdapi_group="EU Array/Compute Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="CS EM Pipe Active"
>+             symbol_name="CsEmActive"
>+             underscore_name="cs_em_active"
>+             description="The percentage of time in which EU EM INT pipeline was actively processing a compute shader instruction."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array/Compute Shader"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="CS FPU Pipe Active"
>+             symbol_name="CsFpuActive"
>+             underscore_name="cs_fpu_active"
>+             description="The percentage of time in which EU FPU INT64 pipeline was actively processing a compute shader instruction."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array/Compute Shader"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="CS Send Pipe Active"
>+             symbol_name="CsSendActive"
>+             underscore_name="cs_send_active"
>+             description="The percentage of time in which EU send pipeline was actively processing a compute shader instruction."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array/Compute Shader"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <register_config type="OA">
>+        <register type="OA" address="0x0000D920" value="0x00000000" />
>+        <register type="OA" address="0x0000D900" value="0x00000000" />
>+        <register type="OA" address="0x0000D904" value="0x00800000" />
>+        <register type="OA" address="0x0000D910" value="0x00000000" />
>+        <register type="OA" address="0x0000D914" value="0x00800000" />
>+        <register type="OA" address="0x0000DC40" value="0x00000000" />
>+    </register_config>
>+    <register_config type="FLEX">
>+        <register type="FLEX" address="0x0000E458" value="0x00861761" />
>+        <register type="FLEX" address="0x0000E558" value="0x00A61961" />
>+        <register type="FLEX" address="0x0000E658" value="0x00860760" />
>+        <register type="FLEX" address="0x0000E758" value="0x00A60960" />
>+        <register type="FLEX" address="0x0000E45C" value="0x00862762" />
>+        <register type="FLEX" address="0x0000E55C" value="0x00A62962" />
>+    </register_config>
>+  </set>
>+
>+  <set name="EuActivity3"
>+       chipset="DG2"
>+       symbol_name="EuActivity3"
>+       underscore_name="eu_activity3"
>+       mdapi_supported_apis="OGL OGL4 OCL MEDIA IO"
>+       hw_config_guid="4fd009c5-6eae-4ecb-a2d1-41edf211cd06"
>+       >
>+    <counter name="GPU Time Elapsed"
>+             symbol_name="GpuTime"
>+             underscore_name="gpu_time"
>+             description="Time elapsed on the GPU during the measurement."
>+             data_type="uint64"
>+             units="ns"
>+             semantic_type="duration"
>+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Core Clocks"
>+             symbol_name="GpuCoreClocks"
>+             underscore_name="gpu_core_clocks"
>+             description="The total number of GPU core clocks elapsed during the measurement."
>+             data_type="uint64"
>+             units="cycles"
>+             semantic_type="event"
>+             equation="GPU_CLOCK 0 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="AVG GPU Core Frequency"
>+             symbol_name="AvgGpuCoreFrequency"
>+             underscore_name="avg_gpu_core_frequency"
>+             description="Average GPU Core Frequency in the measurement."
>+             data_type="uint64"
>+             max_equation="$GpuMaxFrequency"
>+             units="hz"
>+             semantic_type="event"
>+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Busy"
>+             symbol_name="GpuBusy"
>+             underscore_name="gpu_busy"
>+             description="The percentage of time in which the GPU has been processing GPU commands."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="VS Threads Dispatched"
>+             symbol_name="VsThreads"
>+             underscore_name="vs_threads"
>+             description="The total number of vertex shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 1 READ"
>+             mdapi_group="EU Array/Vertex Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="HS Threads Dispatched"
>+             symbol_name="HsThreads"
>+             underscore_name="hs_threads"
>+             description="The total number of hull shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 2 READ"
>+             mdapi_group="EU Array/Hull Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="DS Threads Dispatched"
>+             symbol_name="DsThreads"
>+             underscore_name="ds_threads"
>+             description="The total number of domain shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 3 READ"
>+             mdapi_group="EU Array/Domain Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GS Threads Dispatched"
>+             symbol_name="GsThreads"
>+             underscore_name="gs_threads"
>+             description="The total number of geometry shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 5 READ"
>+             mdapi_group="EU Array/Geometry Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="FS Threads Dispatched"
>+             symbol_name="PsThreads"
>+             underscore_name="ps_threads"
>+             description="The total number of fragment shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 6 READ"
>+             mdapi_group="EU Array/Fragment Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="CS Threads Dispatched"
>+             symbol_name="CsThreads"
>+             underscore_name="cs_threads"
>+             description="The total number of compute shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 4 READ"
>+             mdapi_group="EU Array/Compute Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="VS EM Pipe Active"
>+             symbol_name="VsEmActive"
>+             underscore_name="vs_em_active"
>+             description="The percentage of time in which EU EM INT pipeline was actively processing a vertex shader instruction."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array/Vertex Shader"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="FS EM Pipe Active"
>+             symbol_name="PsEmActive"
>+             underscore_name="ps_em_active"
>+             description="The percentage of time in which EU EM INT pipeline was actively processing a fragment shader instruction."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array/Fragment Shader"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="FS Send Pipe Active"
>+             symbol_name="PsSendActive"
>+             underscore_name="ps_send_active"
>+             description="The percentage of time in which EU send pipeline was actively processing a fragment shader instruction."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array/Fragment Shader"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <register_config type="OA">
>+        <register type="OA" address="0x0000D920" value="0x00000000" />
>+        <register type="OA" address="0x0000D900" value="0x00000000" />
>+        <register type="OA" address="0x0000D904" value="0x00800000" />
>+        <register type="OA" address="0x0000D910" value="0x00000000" />
>+        <register type="OA" address="0x0000D914" value="0x00800000" />
>+        <register type="OA" address="0x0000DC40" value="0x00000000" />
>+    </register_config>
>+    <register_config type="FLEX">
>+        <register type="FLEX" address="0x0000E458" value="0x00811711" />
>+        <register type="FLEX" address="0x0000E558" value="0x00A11911" />
>+        <register type="FLEX" address="0x0000E658" value="0x00851751" />
>+        <register type="FLEX" address="0x0000E758" value="0x00A51951" />
>+        <register type="FLEX" address="0x0000E45C" value="0x00852752" />
>+        <register type="FLEX" address="0x0000E55C" value="0x00A52952" />
>+    </register_config>
>+  </set>
>+
>+  <set name="EuActivity4"
>+       chipset="DG2"
>+       symbol_name="EuActivity4"
>+       underscore_name="eu_activity4"
>+       mdapi_supported_apis="OGL OGL4 OCL MEDIA IO"
>+       hw_config_guid="5226090d-b636-4845-9f76-545cb5db7867"
>+       >
>+    <counter name="GPU Time Elapsed"
>+             symbol_name="GpuTime"
>+             underscore_name="gpu_time"
>+             description="Time elapsed on the GPU during the measurement."
>+             data_type="uint64"
>+             units="ns"
>+             semantic_type="duration"
>+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Core Clocks"
>+             symbol_name="GpuCoreClocks"
>+             underscore_name="gpu_core_clocks"
>+             description="The total number of GPU core clocks elapsed during the measurement."
>+             data_type="uint64"
>+             units="cycles"
>+             semantic_type="event"
>+             equation="GPU_CLOCK 0 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="AVG GPU Core Frequency"
>+             symbol_name="AvgGpuCoreFrequency"
>+             underscore_name="avg_gpu_core_frequency"
>+             description="Average GPU Core Frequency in the measurement."
>+             data_type="uint64"
>+             max_equation="$GpuMaxFrequency"
>+             units="hz"
>+             semantic_type="event"
>+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Busy"
>+             symbol_name="GpuBusy"
>+             underscore_name="gpu_busy"
>+             description="The percentage of time in which the GPU has been processing GPU commands."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="VS Threads Dispatched"
>+             symbol_name="VsThreads"
>+             underscore_name="vs_threads"
>+             description="The total number of vertex shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 1 READ"
>+             mdapi_group="EU Array/Vertex Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="HS Threads Dispatched"
>+             symbol_name="HsThreads"
>+             underscore_name="hs_threads"
>+             description="The total number of hull shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 2 READ"
>+             mdapi_group="EU Array/Hull Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="DS Threads Dispatched"
>+             symbol_name="DsThreads"
>+             underscore_name="ds_threads"
>+             description="The total number of domain shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 3 READ"
>+             mdapi_group="EU Array/Domain Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GS Threads Dispatched"
>+             symbol_name="GsThreads"
>+             underscore_name="gs_threads"
>+             description="The total number of geometry shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 5 READ"
>+             mdapi_group="EU Array/Geometry Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="FS Threads Dispatched"
>+             symbol_name="PsThreads"
>+             underscore_name="ps_threads"
>+             description="The total number of fragment shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 6 READ"
>+             mdapi_group="EU Array/Fragment Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="CS Threads Dispatched"
>+             symbol_name="CsThreads"
>+             underscore_name="cs_threads"
>+             description="The total number of compute shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 4 READ"
>+             mdapi_group="EU Array/Compute Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="HS FPU Pipe Active"
>+             symbol_name="HsFpuActive"
>+             underscore_name="hs_fpu_active"
>+             description="The percentage of time in which EU FPU INT64 pipeline was actively processing a hull shader instruction."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array/Hull Shader"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="DS FPU Pipe Active"
>+             symbol_name="DsFpuActive"
>+             underscore_name="ds_fpu_active"
>+             description="The percentage of time in which EU FPU INT64 pipeline was actively processing a domain shader instruction."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array/Domain Shader"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="VS Send Pipe Active"
>+             symbol_name="VsSendActive"
>+             underscore_name="vs_send_active"
>+             description="The percentage of time in which EU send pipeline was actively processing a vertex shader instruction."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array/Vertex Shader"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <register_config type="OA">
>+        <register type="OA" address="0x0000D920" value="0x00000000" />
>+        <register type="OA" address="0x0000D900" value="0x00000000" />
>+        <register type="OA" address="0x0000D904" value="0x00800000" />
>+        <register type="OA" address="0x0000D910" value="0x00000000" />
>+        <register type="OA" address="0x0000D914" value="0x00800000" />
>+        <register type="OA" address="0x0000DC40" value="0x00000000" />
>+    </register_config>
>+    <register_config type="FLEX">
>+        <register type="FLEX" address="0x0000E458" value="0x00820720" />
>+        <register type="FLEX" address="0x0000E558" value="0x00A20920" />
>+        <register type="FLEX" address="0x0000E658" value="0x00830730" />
>+        <register type="FLEX" address="0x0000E758" value="0x00A30930" />
>+        <register type="FLEX" address="0x0000E45C" value="0x00812712" />
>+        <register type="FLEX" address="0x0000E55C" value="0x00A12912" />
>+    </register_config>
>+  </set>
>+
>+  <set name="EuActivity5"
>+       chipset="DG2"
>+       symbol_name="EuActivity5"
>+       underscore_name="eu_activity5"
>+       mdapi_supported_apis="OGL OGL4 OCL MEDIA IO"
>+       hw_config_guid="c8d33b8f-3152-4c81-99bf-d3cb45d7e1ac"
>+       >
>+    <counter name="GPU Time Elapsed"
>+             symbol_name="GpuTime"
>+             underscore_name="gpu_time"
>+             description="Time elapsed on the GPU during the measurement."
>+             data_type="uint64"
>+             units="ns"
>+             semantic_type="duration"
>+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Core Clocks"
>+             symbol_name="GpuCoreClocks"
>+             underscore_name="gpu_core_clocks"
>+             description="The total number of GPU core clocks elapsed during the measurement."
>+             data_type="uint64"
>+             units="cycles"
>+             semantic_type="event"
>+             equation="GPU_CLOCK 0 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="AVG GPU Core Frequency"
>+             symbol_name="AvgGpuCoreFrequency"
>+             underscore_name="avg_gpu_core_frequency"
>+             description="Average GPU Core Frequency in the measurement."
>+             data_type="uint64"
>+             max_equation="$GpuMaxFrequency"
>+             units="hz"
>+             semantic_type="event"
>+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Busy"
>+             symbol_name="GpuBusy"
>+             underscore_name="gpu_busy"
>+             description="The percentage of time in which the GPU has been processing GPU commands."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="VS Threads Dispatched"
>+             symbol_name="VsThreads"
>+             underscore_name="vs_threads"
>+             description="The total number of vertex shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 1 READ"
>+             mdapi_group="EU Array/Vertex Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="HS Threads Dispatched"
>+             symbol_name="HsThreads"
>+             underscore_name="hs_threads"
>+             description="The total number of hull shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 2 READ"
>+             mdapi_group="EU Array/Hull Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="DS Threads Dispatched"
>+             symbol_name="DsThreads"
>+             underscore_name="ds_threads"
>+             description="The total number of domain shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 3 READ"
>+             mdapi_group="EU Array/Domain Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GS Threads Dispatched"
>+             symbol_name="GsThreads"
>+             underscore_name="gs_threads"
>+             description="The total number of geometry shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 5 READ"
>+             mdapi_group="EU Array/Geometry Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="FS Threads Dispatched"
>+             symbol_name="PsThreads"
>+             underscore_name="ps_threads"
>+             description="The total number of fragment shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 6 READ"
>+             mdapi_group="EU Array/Fragment Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="CS Threads Dispatched"
>+             symbol_name="CsThreads"
>+             underscore_name="cs_threads"
>+             description="The total number of compute shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 4 READ"
>+             mdapi_group="EU Array/Compute Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="HS EM Pipe Active"
>+             symbol_name="HsEmActive"
>+             underscore_name="hs_em_active"
>+             description="The percentage of time in which EU EM INT pipeline was actively processing a hull shader instruction."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array/Hull Shader"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="DS EM Pipe Active"
>+             symbol_name="DsEmActive"
>+             underscore_name="ds_em_active"
>+             description="The percentage of time in which EU EM INT pipeline was actively processing a domain shader instruction."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array/Domain Shader"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="HS Send Pipe Active"
>+             symbol_name="HsSendActive"
>+             underscore_name="hs_send_active"
>+             description="The percentage of time in which EU send pipeline was actively processing a hull shader instruction."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array/Hull Shader"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <register_config type="OA">
>+        <register type="OA" address="0x0000D920" value="0x00000000" />
>+        <register type="OA" address="0x0000D900" value="0x00000000" />
>+        <register type="OA" address="0x0000D904" value="0x00800000" />
>+        <register type="OA" address="0x0000D910" value="0x00000000" />
>+        <register type="OA" address="0x0000D914" value="0x00800000" />
>+        <register type="OA" address="0x0000DC40" value="0x00000000" />
>+    </register_config>
>+    <register_config type="FLEX">
>+        <register type="FLEX" address="0x0000E458" value="0x00821721" />
>+        <register type="FLEX" address="0x0000E558" value="0x00A21921" />
>+        <register type="FLEX" address="0x0000E658" value="0x00831731" />
>+        <register type="FLEX" address="0x0000E758" value="0x00A31931" />
>+        <register type="FLEX" address="0x0000E45C" value="0x00822722" />
>+        <register type="FLEX" address="0x0000E55C" value="0x00A22922" />
>+    </register_config>
>+  </set>
>+
>+  <set name="EuActivity6"
>+       chipset="DG2"
>+       symbol_name="EuActivity6"
>+       underscore_name="eu_activity6"
>+       mdapi_supported_apis="OGL OGL4 OCL MEDIA IO"
>+       hw_config_guid="708d01ed-bc9b-420e-8048-9060abcbb1f0"
>+       >
>+    <counter name="GPU Time Elapsed"
>+             symbol_name="GpuTime"
>+             underscore_name="gpu_time"
>+             description="Time elapsed on the GPU during the measurement."
>+             data_type="uint64"
>+             units="ns"
>+             semantic_type="duration"
>+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Core Clocks"
>+             symbol_name="GpuCoreClocks"
>+             underscore_name="gpu_core_clocks"
>+             description="The total number of GPU core clocks elapsed during the measurement."
>+             data_type="uint64"
>+             units="cycles"
>+             semantic_type="event"
>+             equation="GPU_CLOCK 0 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="AVG GPU Core Frequency"
>+             symbol_name="AvgGpuCoreFrequency"
>+             underscore_name="avg_gpu_core_frequency"
>+             description="Average GPU Core Frequency in the measurement."
>+             data_type="uint64"
>+             max_equation="$GpuMaxFrequency"
>+             units="hz"
>+             semantic_type="event"
>+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Busy"
>+             symbol_name="GpuBusy"
>+             underscore_name="gpu_busy"
>+             description="The percentage of time in which the GPU has been processing GPU commands."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="VS Threads Dispatched"
>+             symbol_name="VsThreads"
>+             underscore_name="vs_threads"
>+             description="The total number of vertex shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 1 READ"
>+             mdapi_group="EU Array/Vertex Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="HS Threads Dispatched"
>+             symbol_name="HsThreads"
>+             underscore_name="hs_threads"
>+             description="The total number of hull shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 2 READ"
>+             mdapi_group="EU Array/Hull Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="DS Threads Dispatched"
>+             symbol_name="DsThreads"
>+             underscore_name="ds_threads"
>+             description="The total number of domain shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 3 READ"
>+             mdapi_group="EU Array/Domain Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GS Threads Dispatched"
>+             symbol_name="GsThreads"
>+             underscore_name="gs_threads"
>+             description="The total number of geometry shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 5 READ"
>+             mdapi_group="EU Array/Geometry Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="FS Threads Dispatched"
>+             symbol_name="PsThreads"
>+             underscore_name="ps_threads"
>+             description="The total number of fragment shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 6 READ"
>+             mdapi_group="EU Array/Fragment Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="CS Threads Dispatched"
>+             symbol_name="CsThreads"
>+             underscore_name="cs_threads"
>+             description="The total number of compute shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 4 READ"
>+             mdapi_group="EU Array/Compute Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GS FPU Pipe Active"
>+             symbol_name="GsFpuActive"
>+             underscore_name="gs_fpu_active"
>+             description="The percentage of time in which EU FPU INT64 pipeline was actively processing a geometry shader instruction."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array/Geometry Shader"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GS EM Pipe Active"
>+             symbol_name="GsEmActive"
>+             underscore_name="gs_em_active"
>+             description="The percentage of time in which EU EM INT pipeline was actively processing a geometry shader instruction."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array/Geometry Shader"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GS Send Pipe Active"
>+             symbol_name="GsSendActive"
>+             underscore_name="gs_send_active"
>+             description="The percentage of time in which EU send pipeline was actively processing a geometry shader instruction."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array/Geometry Shader"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <register_config type="OA">
>+        <register type="OA" address="0x0000D920" value="0x00000000" />
>+        <register type="OA" address="0x0000D900" value="0x00000000" />
>+        <register type="OA" address="0x0000D904" value="0x00800000" />
>+        <register type="OA" address="0x0000D910" value="0x00000000" />
>+        <register type="OA" address="0x0000D914" value="0x00800000" />
>+        <register type="OA" address="0x0000DC40" value="0x00000000" />
>+    </register_config>
>+    <register_config type="FLEX">
>+        <register type="FLEX" address="0x0000E458" value="0x00840740" />
>+        <register type="FLEX" address="0x0000E558" value="0x00A40940" />
>+        <register type="FLEX" address="0x0000E658" value="0x00841741" />
>+        <register type="FLEX" address="0x0000E758" value="0x00A41941" />
>+        <register type="FLEX" address="0x0000E45C" value="0x00842742" />
>+        <register type="FLEX" address="0x0000E55C" value="0x00A42942" />
>+    </register_config>
>+  </set>
>+
>+  <set name="EuActivity7"
>+       chipset="DG2"
>+       symbol_name="EuActivity7"
>+       underscore_name="eu_activity7"
>+       mdapi_supported_apis="OGL OGL4 OCL MEDIA IO"
>+       hw_config_guid="72262d5b-a9e6-4742-820b-2a7bef9a7e61"
>+       >
>+    <counter name="GPU Time Elapsed"
>+             symbol_name="GpuTime"
>+             underscore_name="gpu_time"
>+             description="Time elapsed on the GPU during the measurement."
>+             data_type="uint64"
>+             units="ns"
>+             semantic_type="duration"
>+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Core Clocks"
>+             symbol_name="GpuCoreClocks"
>+             underscore_name="gpu_core_clocks"
>+             description="The total number of GPU core clocks elapsed during the measurement."
>+             data_type="uint64"
>+             units="cycles"
>+             semantic_type="event"
>+             equation="GPU_CLOCK 0 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="AVG GPU Core Frequency"
>+             symbol_name="AvgGpuCoreFrequency"
>+             underscore_name="avg_gpu_core_frequency"
>+             description="Average GPU Core Frequency in the measurement."
>+             data_type="uint64"
>+             max_equation="$GpuMaxFrequency"
>+             units="hz"
>+             semantic_type="event"
>+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Busy"
>+             symbol_name="GpuBusy"
>+             underscore_name="gpu_busy"
>+             description="The percentage of time in which the GPU has been processing GPU commands."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="VS Threads Dispatched"
>+             symbol_name="VsThreads"
>+             underscore_name="vs_threads"
>+             description="The total number of vertex shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 1 READ"
>+             mdapi_group="EU Array/Vertex Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="HS Threads Dispatched"
>+             symbol_name="HsThreads"
>+             underscore_name="hs_threads"
>+             description="The total number of hull shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 2 READ"
>+             mdapi_group="EU Array/Hull Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="DS Threads Dispatched"
>+             symbol_name="DsThreads"
>+             underscore_name="ds_threads"
>+             description="The total number of domain shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 3 READ"
>+             mdapi_group="EU Array/Domain Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GS Threads Dispatched"
>+             symbol_name="GsThreads"
>+             underscore_name="gs_threads"
>+             description="The total number of geometry shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 5 READ"
>+             mdapi_group="EU Array/Geometry Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="FS Threads Dispatched"
>+             symbol_name="PsThreads"
>+             underscore_name="ps_threads"
>+             description="The total number of fragment shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 6 READ"
>+             mdapi_group="EU Array/Fragment Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="CS Threads Dispatched"
>+             symbol_name="CsThreads"
>+             underscore_name="cs_threads"
>+             description="The total number of compute shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 4 READ"
>+             mdapi_group="EU Array/Compute Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU FPU Pipe Active"
>+             symbol_name="FpuActive"
>+             underscore_name="fpu_active"
>+             description="The percentage of time in which EU FPU INT64 pipeline was actively processing."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array/Pipes"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EM Pipe Active"
>+             symbol_name="EmActive"
>+             underscore_name="em_active"
>+             description="The percentage of time in which EU EM INT pipeline was actively processing."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array/Pipes"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EM And FPU Pipes Active"
>+             symbol_name="EmFpuActive"
>+             underscore_name="em_fpu_active"
>+             description="The percentage of time in which EU EM INT and FPU INT64 pipelines were actively processing."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array/Pipes"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="EU AVG IPC Rate"
>+             symbol_name="EuAvgIpcRate"
>+             underscore_name="eu_avg_ipc_rate"
>+             description="The average rate of IPC calculated for 2 FPU pipelines."
>+             data_type="float"
>+             max_equation="2"
>+             units="number"
>+             semantic_type="ratio"
>+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD  A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD  A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FADD  A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FSUB FDIV 1 FADD"
>+             mdapi_group="EU Array"
>+             mdapi_usage_flags="Tier4 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <register_config type="OA">
>+        <register type="OA" address="0x0000D920" value="0x00000000" />
>+        <register type="OA" address="0x0000D900" value="0x00000000" />
>+        <register type="OA" address="0x0000D904" value="0x00800000" />
>+        <register type="OA" address="0x0000D910" value="0x00000000" />
>+        <register type="OA" address="0x0000D914" value="0x00800000" />
>+        <register type="OA" address="0x0000DC40" value="0x00000000" />
>+    </register_config>
>+    <register_config type="FLEX">
>+        <register type="FLEX" address="0x0000E458" value="0x00803703" />
>+        <register type="FLEX" address="0x0000E558" value="0x00A03903" />
>+        <register type="FLEX" address="0x0000E658" value="0x00800700" />
>+        <register type="FLEX" address="0x0000E758" value="0x00A00900" />
>+        <register type="FLEX" address="0x0000E45C" value="0x00801701" />
>+        <register type="FLEX" address="0x0000E55C" value="0x00A01901" />
>+    </register_config>
>+  </set>
>+
>+  <set name="EuActivity8"
>+       chipset="DG2"
>+       symbol_name="EuActivity8"
>+       underscore_name="eu_activity8"
>+       mdapi_supported_apis="OGL OGL4 OCL MEDIA IO"
>+       hw_config_guid="11c685de-1445-41b8-acb5-8e79c75c4fd8"
>+       >
>+    <counter name="GPU Time Elapsed"
>+             symbol_name="GpuTime"
>+             underscore_name="gpu_time"
>+             description="Time elapsed on the GPU during the measurement."
>+             data_type="uint64"
>+             units="ns"
>+             semantic_type="duration"
>+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Core Clocks"
>+             symbol_name="GpuCoreClocks"
>+             underscore_name="gpu_core_clocks"
>+             description="The total number of GPU core clocks elapsed during the measurement."
>+             data_type="uint64"
>+             units="cycles"
>+             semantic_type="event"
>+             equation="GPU_CLOCK 0 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="AVG GPU Core Frequency"
>+             symbol_name="AvgGpuCoreFrequency"
>+             underscore_name="avg_gpu_core_frequency"
>+             description="Average GPU Core Frequency in the measurement."
>+             data_type="uint64"
>+             max_equation="$GpuMaxFrequency"
>+             units="hz"
>+             semantic_type="event"
>+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Busy"
>+             symbol_name="GpuBusy"
>+             underscore_name="gpu_busy"
>+             description="The percentage of time in which the GPU has been processing GPU commands."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="VS Threads Dispatched"
>+             symbol_name="VsThreads"
>+             underscore_name="vs_threads"
>+             description="The total number of vertex shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 1 READ"
>+             mdapi_group="EU Array/Vertex Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="HS Threads Dispatched"
>+             symbol_name="HsThreads"
>+             underscore_name="hs_threads"
>+             description="The total number of hull shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 2 READ"
>+             mdapi_group="EU Array/Hull Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="DS Threads Dispatched"
>+             symbol_name="DsThreads"
>+             underscore_name="ds_threads"
>+             description="The total number of domain shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 3 READ"
>+             mdapi_group="EU Array/Domain Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GS Threads Dispatched"
>+             symbol_name="GsThreads"
>+             underscore_name="gs_threads"
>+             description="The total number of geometry shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 5 READ"
>+             mdapi_group="EU Array/Geometry Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="FS Threads Dispatched"
>+             symbol_name="PsThreads"
>+             underscore_name="ps_threads"
>+             description="The total number of fragment shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 6 READ"
>+             mdapi_group="EU Array/Fragment Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="CS Threads Dispatched"
>+             symbol_name="CsThreads"
>+             underscore_name="cs_threads"
>+             description="The total number of compute shader hardware threads dispatched."
>+             data_type="uint64"
>+             units="threads"
>+             semantic_type="event"
>+             equation="A 4 READ"
>+             mdapi_group="EU Array/Compute Shader"
>+             mdapi_usage_flags="Tier3 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="DS Send Pipe Active"
>+             symbol_name="DsSendActive"
>+             underscore_name="ds_send_active"
>+             description="The percentage of time in which EU send pipeline was actively processing a domain shader instruction."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array/Domain Shader"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="SYS And FPU Pipes Active"
>+             symbol_name="SysFpuActive"
>+             underscore_name="sys_fpu_active"
>+             description="The percentage of time in which EU Systolic and FPU INT64 pipelines were actively processing."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array/Pipes"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="SYS Pipe Active"
>+             symbol_name="SysActive"
>+             underscore_name="sys_active"
>+             description="The percentage of time in which EU SYS pipeline was actively processing."
>+             data_type="float"
>+             max_equation="100"
>+             units="percent"
>+             semantic_type="duration"
>+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
>+             mdapi_group="EU Array/Pipes"
>+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <register_config type="OA">
>+        <register type="OA" address="0x0000D920" value="0x00000000" />
>+        <register type="OA" address="0x0000D900" value="0x00000000" />
>+        <register type="OA" address="0x0000D904" value="0x00800000" />
>+        <register type="OA" address="0x0000D910" value="0x00000000" />
>+        <register type="OA" address="0x0000D914" value="0x00800000" />
>+        <register type="OA" address="0x0000DC40" value="0x00000000" />
>+    </register_config>
>+    <register_config type="FLEX">
>+        <register type="FLEX" address="0x0000E458" value="0x00832732" />
>+        <register type="FLEX" address="0x0000E558" value="0x00A32932" />
>+        <register type="FLEX" address="0x0000E658" value="0x00809709" />
>+        <register type="FLEX" address="0x0000E758" value="0x00A09909" />
>+        <register type="FLEX" address="0x0000E45C" value="0x0080A70A" />
>+        <register type="FLEX" address="0x0000E55C" value="0x00A0A90A" />
>+    </register_config>
>+  </set>
>+
>+  <set name="Metric set TestOa"
>+       chipset="DG2"
>+       symbol_name="TestOa"
>+       underscore_name="test_oa"
>+       mdapi_supported_apis="VK OGL OCL MEDIA IO"
>+       hw_config_guid="2db1e1d4-8c48-4bb7-b248-0f40b70433ff"
>+       >
>+    <counter name="GPU Time Elapsed"
>+             symbol_name="GpuTime"
>+             underscore_name="gpu_time"
>+             description="Time elapsed on the GPU during the measurement."
>+             data_type="uint64"
>+             units="ns"
>+             semantic_type="duration"
>+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="GPU Core Clocks"
>+             symbol_name="GpuCoreClocks"
>+             underscore_name="gpu_core_clocks"
>+             description="The total number of GPU core clocks elapsed during the measurement."
>+             data_type="uint64"
>+             units="cycles"
>+             semantic_type="event"
>+             equation="GPU_CLOCK 0 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="AVG GPU Core Frequency"
>+             symbol_name="AvgGpuCoreFrequency"
>+             underscore_name="avg_gpu_core_frequency"
>+             description="Average GPU Core Frequency in the measurement."
>+             data_type="uint64"
>+             max_equation="$GpuMaxFrequency"
>+             units="hz"
>+             semantic_type="event"
>+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="TestCounter0"
>+             symbol_name="Counter0"
>+             underscore_name="counter0"
>+             description="HW test counter 0. Factor: 0.0"
>+             data_type="uint64"
>+             units="events"
>+             semantic_type="event"
>+             equation="B 0 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="TestCounter1"
>+             symbol_name="Counter1"
>+             underscore_name="counter1"
>+             description="HW test counter 1. Factor: 1.0"
>+             data_type="uint64"
>+             units="events"
>+             semantic_type="event"
>+             equation="B 1 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="TestCounter2"
>+             symbol_name="Counter2"
>+             underscore_name="counter2"
>+             description="HW test counter 2. Factor: 1.0"
>+             data_type="uint64"
>+             units="events"
>+             semantic_type="event"
>+             equation="B 2 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="TestCounter3"
>+             symbol_name="Counter3"
>+             underscore_name="counter3"
>+             description="HW test counter 3. Factor: 0.5"
>+             data_type="uint64"
>+             units="events"
>+             semantic_type="event"
>+             equation="B 3 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="TestCounter4"
>+             symbol_name="Counter4"
>+             underscore_name="counter4"
>+             description="HW test counter 4. Factor: 0.3333"
>+             data_type="uint64"
>+             units="events"
>+             semantic_type="event"
>+             equation="B 4 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="TestCounter5"
>+             symbol_name="Counter5"
>+             underscore_name="counter5"
>+             description="HW test counter 5. Factor: 0.3333"
>+             data_type="uint64"
>+             units="events"
>+             semantic_type="event"
>+             equation="B 5 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="TestCounter6"
>+             symbol_name="Counter6"
>+             underscore_name="counter6"
>+             description="HW test counter 6. Factor: 0.16666"
>+             data_type="uint64"
>+             units="events"
>+             semantic_type="event"
>+             equation="B 6 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="TestCounter7"
>+             symbol_name="Counter7"
>+             underscore_name="counter7"
>+             description="HW test counter 7. Factor: 0.6666"
>+             data_type="uint64"
>+             units="events"
>+             semantic_type="event"
>+             equation="B 7 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="TestCounter8"
>+             symbol_name="Counter8"
>+             underscore_name="counter8"
>+             description="HW test counter 8. Should be equal to 1 in IOStream or in OAG query mode"
>+             data_type="uint64"
>+             units="events"
>+             semantic_type="event"
>+             equation="C 0 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <counter name="TestCounter9 - OAR enable"
>+             symbol_name="Counter9"
>+             underscore_name="counter9"
>+             description="HW test counter 9. Should be equal to 1 in query."
>+             data_type="uint64"
>+             units="events"
>+             semantic_type="event"
>+             equation="C 1 READ"
>+             mdapi_group="GPU"
>+             mdapi_usage_flags="Frame Batch Draw"
>+             mdapi_supported_apis=""
>+             mdapi_hw_unit_type="gpu"
>+             />
>+    <register_config type="NOA">
>+        <register type="NOA" address="0x00000D04" value="0x00000200" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x2E080000" />
>+        <register type="NOA" address="0x00009888" value="0x34080000" />
>+        <register type="NOA" address="0x00009888" value="0x06088000" />
>+        <register type="NOA" address="0x00009888" value="0x10080FF3" />
>+        <register type="NOA" address="0x00009888" value="0x08080000" />
>+        <register type="NOA" address="0x00009888" value="0x02080000" />
>+        <register type="NOA" address="0x00009888" value="0x02032400" />
>+        <register type="NOA" address="0x00009888" value="0x48030000" />
>+        <register type="NOA" address="0x00009888" value="0x4A030000" />
>+        <register type="NOA" address="0x00009884" value="0x00000000" />
>+        <register type="NOA" address="0x00009888" value="0x5C030002" />
>+    </register_config>
>+    <register_config type="OA">
>+        <register type="OA" address="0x0000D920" value="0x00000000" />
>+        <register type="OA" address="0x0000D900" value="0x00000000" />
>+        <register type="OA" address="0x0000D904" value="0xF0800000" />
>+        <register type="OA" address="0x0000D910" value="0x00000000" />
>+        <register type="OA" address="0x0000D914" value="0xF0800000" />
>+        <register type="OA" address="0x0000DC40" value="0x00FF0000" />
>+        <register type="OA" address="0x0000D940" value="0x00000004" />
>+        <register type="OA" address="0x0000D944" value="0x0000FFFF" />
>+        <register type="OA" address="0x0000DC00" value="0x00000004" />
>+        <register type="OA" address="0x0000DC04" value="0x0000FFFF" />
>+        <register type="OA" address="0x0000D948" value="0x00000003" />
>+        <register type="OA" address="0x0000D94C" value="0x0000FFFF" />
>+        <register type="OA" address="0x0000DC08" value="0x00000003" />
>+        <register type="OA" address="0x0000DC0C" value="0x0000FFFF" />
>+        <register type="OA" address="0x0000D950" value="0x00000007" />
>+        <register type="OA" address="0x0000D954" value="0x0000FFFF" />
>+        <register type="OA" address="0x0000DC10" value="0x00000007" />
>+        <register type="OA" address="0x0000DC14" value="0x0000FFFF" />
>+        <register type="OA" address="0x0000D958" value="0x00100002" />
>+        <register type="OA" address="0x0000D95C" value="0x0000FFF7" />
>+        <register type="OA" address="0x0000DC18" value="0x00100002" />
>+        <register type="OA" address="0x0000DC1C" value="0x0000FFF7" />
>+        <register type="OA" address="0x0000D960" value="0x00100002" />
>+        <register type="OA" address="0x0000D964" value="0x0000FFCF" />
>+        <register type="OA" address="0x0000DC20" value="0x00100002" />
>+        <register type="OA" address="0x0000DC24" value="0x0000FFCF" />
>+        <register type="OA" address="0x0000D968" value="0x00100082" />
>+        <register type="OA" address="0x0000D96C" value="0x0000FFEF" />
>+        <register type="OA" address="0x0000DC28" value="0x00100082" />
>+        <register type="OA" address="0x0000DC2C" value="0x0000FFEF" />
>+        <register type="OA" address="0x0000D970" value="0x001000C2" />
>+        <register type="OA" address="0x0000D974" value="0x0000FFE7" />
>+        <register type="OA" address="0x0000DC30" value="0x001000C2" />
>+        <register type="OA" address="0x0000DC34" value="0x0000FFE7" />
>+        <register type="OA" address="0x0000D978" value="0x00100001" />
>+        <register type="OA" address="0x0000D97C" value="0x0000FFE7" />
>+        <register type="OA" address="0x0000DC38" value="0x00100001" />
>+        <register type="OA" address="0x0000DC3C" value="0x0000FFE7" />
>+    </register_config>
>+  </set>
>+
>+</metrics>
>diff --git a/lib/i915/perf-configs/perf-metricset-codegen.py b/lib/i915/perf-configs/perf-metricset-codegen.py
>index 00528c31..58237c27 100644
>--- a/lib/i915/perf-configs/perf-metricset-codegen.py
>+++ b/lib/i915/perf-configs/perf-metricset-codegen.py
>@@ -126,6 +126,19 @@ def generate_metric_sets(args, gen):
>                 metric_set->c_offset = metric_set->b_offset + 8;
>                 metric_set->perfcnt_offset = metric_set->c_offset + 8;
>
>+            """))
>+        elif gen.chipset == "dg2":
>+            c(textwrap.dedent("""\
>+                metric_set->perf_oa_format = I915_OA_FORMAT_A24u40_A14u32_B8_C8;
>+
>+                metric_set->perf_raw_size = 256;
>+                metric_set->gpu_time_offset = 0;
>+                metric_set->gpu_clock_offset = 1;
>+                metric_set->a_offset = 2;
>+                metric_set->b_offset = metric_set->a_offset + 38;
>+                metric_set->c_offset = metric_set->b_offset + 8;
>+                metric_set->perfcnt_offset = metric_set->c_offset + 8;
>+
>             """))
>         else:
>             c(textwrap.dedent("""\
>diff --git a/lib/i915/perf-configs/update-guids.py b/lib/i915/perf-configs/update-guids.py
>index d38a7a95..3cd06f13 100755
>--- a/lib/i915/perf-configs/update-guids.py
>+++ b/lib/i915/perf-configs/update-guids.py
>@@ -174,7 +174,7 @@ chipsets = [ 'hsw',
>              'bxt', 'glk',
>              'cnl',
>              'icl', 'ehl',
>-             'tglgt1', 'tglgt2', 'rkl', 'dg1', 'adl' ]
>+             'tglgt1', 'tglgt2', 'rkl', 'dg1', 'adl', 'dg2', ]
>
> for chipset in chipsets:
>     filename = 'oa-' + chipset + '.xml'
>diff --git a/lib/i915/perf.c b/lib/i915/perf.c
>index d004313f..39179546 100644
>--- a/lib/i915/perf.c
>+++ b/lib/i915/perf.c
>@@ -58,6 +58,7 @@
> #include "i915_perf_metrics_rkl.h"
> #include "i915_perf_metrics_dg1.h"
> #include "i915_perf_metrics_adl.h"
>+#include "i915_perf_metrics_dg2.h"
>
> static int
> perf_ioctl(int fd, unsigned long request, void *arg)
>@@ -280,6 +281,8 @@ intel_perf_for_devinfo(uint32_t device_id,
> 	} else if (devinfo->is_alderlake_s || devinfo->is_alderlake_p ||
> 		   devinfo->is_raptorlake_s || devinfo->is_alderlake_n) {
> 		intel_perf_load_metrics_adl(perf);
>+	} else if (devinfo->is_dg2) {
>+		intel_perf_load_metrics_dg2(perf);
> 	} else {
> 		return unsupported_i915_perf_platform(perf);
> 	}
>diff --git a/lib/meson.build b/lib/meson.build
>index 98c2803b..8dabd64d 100644
>--- a/lib/meson.build
>+++ b/lib/meson.build
>@@ -254,7 +254,7 @@ i915_perf_hardware = [
>   'bxt', 'glk',
>   'cnl',
>   'icl', 'ehl',
>-  'tglgt1', 'tglgt2', 'rkl', 'dg1', 'adl',
>+  'tglgt1', 'tglgt2', 'rkl', 'dg1', 'adl', 'dg2',
> ]
>
> i915_xml_files = []
>-- 
>2.25.1
>

[igt-dev] [PATCH i-g-t 18/23] i915/perf: Fix DG2 A0 report header

[Umesh Nerlige Ramappa]

Timestamp is still 32 bits in DG2 A0. Fix that. Specify variants of DG2
that have the OA format header bug

Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
---
 tests/i915/perf.c | 45 ++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 44 insertions(+), 1 deletion(-)

diff --git a/tests/i915/perf.c b/tests/i915/perf.c
index 52de5d38..6951735b 100644
--- a/tests/i915/perf.c
+++ b/tests/i915/perf.c
@@ -40,6 +40,7 @@
 
 #include "i915/gem.h"
 #include "i915/perf.h"
+#include "i915_pciids.h"
 #include "igt.h"
 #include "igt_perf.h"
 #include "igt_sysfs.h"
@@ -503,6 +504,47 @@ cs_timestamp_frequency(int fd)
 	return value;
 }
 
+/*
+ * Hacky ways until IGT is fully aware of steppings:
+ *
+ * WA is permanent for G11. For other DG2s, bug is fixed from B0 onwards.
+ * Steps A0 and A1 correspond to 0 and 1 revid values.
+ */
+struct perf_dev_info {
+	bool is_dg2_g11;
+};
+
+static const struct perf_dev_info dg2_g11_info = {
+	.is_dg2_g11 = true,
+};
+
+static const struct pci_id_match perf_dev_match[] = {
+	INTEL_DG2_G11_IDS(&dg2_g11_info),
+};
+
+static const struct perf_dev_info *__dev_info(uint32_t device_id)
+{
+	static const struct perf_dev_info *cache;
+	static int _iter;
+
+	if (_iter)
+		goto out;
+
+	for (_iter = 0; _iter < ARRAY_SIZE(perf_dev_match); _iter++)
+		if (device_id == perf_dev_match[_iter].device_id)
+			cache = (void *) perf_dev_match[_iter].match_data;
+
+out:
+	return cache;
+}
+
+#define IS_DG2_G11(__d) (__dev_info(__d) ? __dev_info(__d)->is_dg2_g11 : false)
+#define IS_DG2_A0(__d) (IS_DG2(__d) && (intel_perf->devinfo.revision == 0))
+#define IS_DG2_A1(__d) (IS_DG2(__d) && (intel_perf->devinfo.revision == 1))
+
+#define HAS_OA_FORMAT_64BIT_HDR_BUG(__d) \
+	(IS_DG2_G11(__d) || IS_DG2_A0(__d) || IS_DG2_A1(__d))
+
 static uint64_t
 cs_timebase_scale(uint32_t u32_delta)
 {
@@ -514,7 +556,8 @@ oa_timestamp(const uint32_t *report, enum drm_i915_oa_format format)
 {
 	struct oa_format fmt = get_oa_format(format);
 
-	return fmt.report_hdr_64bit ? *(uint64_t *)&report[2] : report[1];
+	return fmt.report_hdr_64bit && !HAS_OA_FORMAT_64BIT_HDR_BUG(devid) ?
+		*(uint64_t *)&report[2] : report[1];
 }
 
 static uint64_t
-- 
2.25.1

Re: [igt-dev] [PATCH i-g-t 18/23] i915/perf: Fix DG2 A0 report header

[Lionel Landwerlin]

On 23/08/2022 23:30, Umesh Nerlige Ramappa wrote:
> Timestamp is still 32 bits in DG2 A0. Fix that. Specify variants of DG2
> that have the OA format header bug
>
> Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>


Are A0 chipsets available outside of Intel?

Maybe it's not worth bothering with this workaround?


Either way :


Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>


> ---
>   tests/i915/perf.c | 45 ++++++++++++++++++++++++++++++++++++++++++++-
>   1 file changed, 44 insertions(+), 1 deletion(-)
>
> diff --git a/tests/i915/perf.c b/tests/i915/perf.c
> index 52de5d38..6951735b 100644
> --- a/tests/i915/perf.c
> +++ b/tests/i915/perf.c
> @@ -40,6 +40,7 @@
>   
>   #include "i915/gem.h"
>   #include "i915/perf.h"
> +#include "i915_pciids.h"
>   #include "igt.h"
>   #include "igt_perf.h"
>   #include "igt_sysfs.h"
> @@ -503,6 +504,47 @@ cs_timestamp_frequency(int fd)
>   	return value;
>   }
>   
> +/*
> + * Hacky ways until IGT is fully aware of steppings:
> + *
> + * WA is permanent for G11. For other DG2s, bug is fixed from B0 onwards.
> + * Steps A0 and A1 correspond to 0 and 1 revid values.
> + */
> +struct perf_dev_info {
> +	bool is_dg2_g11;
> +};
> +
> +static const struct perf_dev_info dg2_g11_info = {
> +	.is_dg2_g11 = true,
> +};
> +
> +static const struct pci_id_match perf_dev_match[] = {
> +	INTEL_DG2_G11_IDS(&dg2_g11_info),
> +};
> +
> +static const struct perf_dev_info *__dev_info(uint32_t device_id)
> +{
> +	static const struct perf_dev_info *cache;
> +	static int _iter;
> +
> +	if (_iter)
> +		goto out;
> +
> +	for (_iter = 0; _iter < ARRAY_SIZE(perf_dev_match); _iter++)
> +		if (device_id == perf_dev_match[_iter].device_id)
> +			cache = (void *) perf_dev_match[_iter].match_data;
> +
> +out:
> +	return cache;
> +}
> +
> +#define IS_DG2_G11(__d) (__dev_info(__d) ? __dev_info(__d)->is_dg2_g11 : false)
> +#define IS_DG2_A0(__d) (IS_DG2(__d) && (intel_perf->devinfo.revision == 0))
> +#define IS_DG2_A1(__d) (IS_DG2(__d) && (intel_perf->devinfo.revision == 1))
> +
> +#define HAS_OA_FORMAT_64BIT_HDR_BUG(__d) \
> +	(IS_DG2_G11(__d) || IS_DG2_A0(__d) || IS_DG2_A1(__d))
> +
>   static uint64_t
>   cs_timebase_scale(uint32_t u32_delta)
>   {
> @@ -514,7 +556,8 @@ oa_timestamp(const uint32_t *report, enum drm_i915_oa_format format)
>   {
>   	struct oa_format fmt = get_oa_format(format);
>   
> -	return fmt.report_hdr_64bit ? *(uint64_t *)&report[2] : report[1];
> +	return fmt.report_hdr_64bit && !HAS_OA_FORMAT_64BIT_HDR_BUG(devid) ?
> +		*(uint64_t *)&report[2] : report[1];
>   }
>   
>   static uint64_t

Re: [igt-dev] [PATCH i-g-t 18/23] i915/perf: Fix DG2 A0 report header

[Umesh Nerlige Ramappa]

On Tue, Sep 06, 2022 at 05:23:54PM +0300, Lionel Landwerlin wrote:
>On 23/08/2022 23:30, Umesh Nerlige Ramappa wrote:
>>Timestamp is still 32 bits in DG2 A0. Fix that. Specify variants of DG2
>>that have the OA format header bug
>>
>>Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
>
>
>Are A0 chipsets available outside of Intel?

No. I plan to remove A0 specific WAs in the follow up revisions.

Thanks,
Umesh
>
>Maybe it's not worth bothering with this workaround?
>
>
>Either way :
>
>
>Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
>
>
>>---
>>  tests/i915/perf.c | 45 ++++++++++++++++++++++++++++++++++++++++++++-
>>  1 file changed, 44 insertions(+), 1 deletion(-)
>>
>>diff --git a/tests/i915/perf.c b/tests/i915/perf.c
>>index 52de5d38..6951735b 100644
>>--- a/tests/i915/perf.c
>>+++ b/tests/i915/perf.c
>>@@ -40,6 +40,7 @@
>>  #include "i915/gem.h"
>>  #include "i915/perf.h"
>>+#include "i915_pciids.h"
>>  #include "igt.h"
>>  #include "igt_perf.h"
>>  #include "igt_sysfs.h"
>>@@ -503,6 +504,47 @@ cs_timestamp_frequency(int fd)
>>  	return value;
>>  }
>>+/*
>>+ * Hacky ways until IGT is fully aware of steppings:
>>+ *
>>+ * WA is permanent for G11. For other DG2s, bug is fixed from B0 onwards.
>>+ * Steps A0 and A1 correspond to 0 and 1 revid values.
>>+ */
>>+struct perf_dev_info {
>>+	bool is_dg2_g11;
>>+};
>>+
>>+static const struct perf_dev_info dg2_g11_info = {
>>+	.is_dg2_g11 = true,
>>+};
>>+
>>+static const struct pci_id_match perf_dev_match[] = {
>>+	INTEL_DG2_G11_IDS(&dg2_g11_info),
>>+};
>>+
>>+static const struct perf_dev_info *__dev_info(uint32_t device_id)
>>+{
>>+	static const struct perf_dev_info *cache;
>>+	static int _iter;
>>+
>>+	if (_iter)
>>+		goto out;
>>+
>>+	for (_iter = 0; _iter < ARRAY_SIZE(perf_dev_match); _iter++)
>>+		if (device_id == perf_dev_match[_iter].device_id)
>>+			cache = (void *) perf_dev_match[_iter].match_data;
>>+
>>+out:
>>+	return cache;
>>+}
>>+
>>+#define IS_DG2_G11(__d) (__dev_info(__d) ? __dev_info(__d)->is_dg2_g11 : false)
>>+#define IS_DG2_A0(__d) (IS_DG2(__d) && (intel_perf->devinfo.revision == 0))
>>+#define IS_DG2_A1(__d) (IS_DG2(__d) && (intel_perf->devinfo.revision == 1))
>>+
>>+#define HAS_OA_FORMAT_64BIT_HDR_BUG(__d) \
>>+	(IS_DG2_G11(__d) || IS_DG2_A0(__d) || IS_DG2_A1(__d))
>>+
>>  static uint64_t
>>  cs_timebase_scale(uint32_t u32_delta)
>>  {
>>@@ -514,7 +556,8 @@ oa_timestamp(const uint32_t *report, enum drm_i915_oa_format format)
>>  {
>>  	struct oa_format fmt = get_oa_format(format);
>>-	return fmt.report_hdr_64bit ? *(uint64_t *)&report[2] : report[1];
>>+	return fmt.report_hdr_64bit && !HAS_OA_FORMAT_64BIT_HDR_BUG(devid) ?
>>+		*(uint64_t *)&report[2] : report[1];
>>  }
>>  static uint64_t
>
>

[igt-dev] ✗ Fi.CI.BUILD: failure for Add DG2 OA test

[Patchwork]

== Series Details ==

Series: Add DG2 OA test
URL   : https://patchwork.freedesktop.org/series/107633/
State : failure

== Summary ==

IGT patchset build failed on latest successful build
e01fe99f00692864b709253638c809231d1fb333 i915/guc: Disable i915_pm_rps when SLPC is enabled

[igt-dev] ✗ Fi.CI.BUILD: failure for Add DG2 OA test (rev2)

[Patchwork]

== Series Details ==

Series: Add DG2 OA test (rev2)
URL   : https://patchwork.freedesktop.org/series/107633/
State : failure

== Summary ==

IGT patchset build failed on latest successful build
e01fe99f00692864b709253638c809231d1fb333 i915/guc: Disable i915_pm_rps when SLPC is enabled

Re: [igt-dev] ✗ Fi.CI.BUILD: failure for Add DG2 OA test (rev2)

[Petri Latvala]

On Wed, Aug 24, 2022 at 09:33:36AM +0000, Patchwork wrote:
> == Series Details ==
> 
> Series: Add DG2 OA test (rev2)
> URL   : https://patchwork.freedesktop.org/series/107633/
> State : failure
> 
> == Summary ==
> 
> IGT patchset build failed on latest successful build
> e01fe99f00692864b709253638c809231d1fb333 i915/guc: Disable i915_pm_rps when SLPC is enabled
> 
> 

Sorry for this, libdrm building changed and everything failed to build
after that. I'll keep poking this and other affected series through
the pipeline.


-- 
Petri Latvala

[igt-dev] ✓ Fi.CI.BAT: success for Add DG2 OA test (rev3)

[Patchwork]

[-- Attachment #1: Type: text/plain, Size: 1913 bytes --]

== Series Details ==

Series: Add DG2 OA test (rev3)
URL   : https://patchwork.freedesktop.org/series/107633/
State : success

== Summary ==

CI Bug Log - changes from CI_DRM_12020 -> IGTPW_7680
====================================================

Summary
-------

  **SUCCESS**

  No regressions found.

  External URL: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/index.html

Participating hosts (38 -> 37)
------------------------------

  Missing    (1): fi-pnv-d510 

Known issues
------------

  Here are the changes found in IGTPW_7680 that come from known issues:

### IGT changes ###

#### Warnings ####

  * igt@i915_selftest@live@hangcheck:
    - bat-dg1-5:          [DMESG-FAIL][1] ([i915#4957]) -> [DMESG-FAIL][2] ([i915#4494] / [i915#4957])
   [1]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/bat-dg1-5/igt@i915_selftest@live@hangcheck.html
   [2]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/bat-dg1-5/igt@i915_selftest@live@hangcheck.html

  
  {name}: This element is suppressed. This means it is ignored when computing
          the status of the difference (SUCCESS, WARNING, or FAILURE).

  [i915#4494]: https://gitlab.freedesktop.org/drm/intel/issues/4494
  [i915#4957]: https://gitlab.freedesktop.org/drm/intel/issues/4957
  [i915#6530]: https://gitlab.freedesktop.org/drm/intel/issues/6530


Build changes
-------------

  * CI: CI-20190529 -> None
  * IGT: IGT_6636 -> IGTPW_7680

  CI-20190529: 20190529
  CI_DRM_12020: 9e188e66f942a9b7db1c41d56ac9ff1433473d8e @ git://anongit.freedesktop.org/gfx-ci/linux
  IGTPW_7680: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/index.html
  IGT_6636: 1298b5f0e1b3e010657ffba41d2e775fab028e08 @ https://gitlab.freedesktop.org/drm/igt-gpu-tools.git


Testlist changes
----------------

+igt@perf@non-power-of-2-oa-formats

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/index.html

[-- Attachment #2: Type: text/html, Size: 2499 bytes --]

[igt-dev] ✗ Fi.CI.IGT: failure for Add DG2 OA test (rev3)

[Patchwork]

[-- Attachment #1: Type: text/plain, Size: 65231 bytes --]

== Series Details ==

Series: Add DG2 OA test (rev3)
URL   : https://patchwork.freedesktop.org/series/107633/
State : failure

== Summary ==

CI Bug Log - changes from CI_DRM_12020_full -> IGTPW_7680_full
====================================================

Summary
-------

  **FAILURE**

  Serious unknown changes coming with IGTPW_7680_full absolutely need to be
  verified manually.
  
  If you think the reported changes have nothing to do with the changes
  introduced in IGTPW_7680_full, please notify your bug team to allow them
  to document this new failure mode, which will reduce false positives in CI.

  External URL: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/index.html

Participating hosts (12 -> 9)
------------------------------

  Missing    (3): pig-skl-6260u pig-kbl-iris pig-glk-j5005 

Possible new issues
-------------------

  Here are the unknown changes that may have been introduced in IGTPW_7680_full:

### IGT changes ###

#### Possible regressions ####

  * igt@gem_exec_suspend@basic-s0@smem:
    - shard-glk:          NOTRUN -> [INCOMPLETE][1]
   [1]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-glk1/igt@gem_exec_suspend@basic-s0@smem.html

  * igt@i915_suspend@basic-s2idle-without-i915:
    - shard-snb:          [PASS][2] -> [INCOMPLETE][3]
   [2]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-snb2/igt@i915_suspend@basic-s2idle-without-i915.html
   [3]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-snb2/igt@i915_suspend@basic-s2idle-without-i915.html
    - shard-apl:          [PASS][4] -> [INCOMPLETE][5]
   [4]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-apl2/igt@i915_suspend@basic-s2idle-without-i915.html
   [5]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-apl8/igt@i915_suspend@basic-s2idle-without-i915.html
    - shard-iclb:         [PASS][6] -> [INCOMPLETE][7]
   [6]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-iclb1/igt@i915_suspend@basic-s2idle-without-i915.html
   [7]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb5/igt@i915_suspend@basic-s2idle-without-i915.html

  * igt@perf@global-sseu-config-invalid:
    - shard-tglb:         [PASS][8] -> [SKIP][9] +1 similar issue
   [8]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-tglb7/igt@perf@global-sseu-config-invalid.html
   [9]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-tglb7/igt@perf@global-sseu-config-invalid.html

  * {igt@perf@non-power-of-2-oa-formats} (NEW):
    - {shard-rkl}:        NOTRUN -> [SKIP][10]
   [10]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-rkl-1/igt@perf@non-power-of-2-oa-formats.html
    - shard-tglb:         NOTRUN -> [SKIP][11]
   [11]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-tglb5/igt@perf@non-power-of-2-oa-formats.html
    - {shard-tglu}:       NOTRUN -> [SKIP][12]
   [12]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-tglu-2/igt@perf@non-power-of-2-oa-formats.html
    - shard-iclb:         NOTRUN -> [SKIP][13]
   [13]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb2/igt@perf@non-power-of-2-oa-formats.html

  
#### Suppressed ####

  The following results come from untrusted machines, tests, or statuses.
  They do not affect the overall result.

  * igt@perf@global-sseu-config-invalid:
    - {shard-rkl}:        NOTRUN -> [SKIP][14]
   [14]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-rkl-1/igt@perf@global-sseu-config-invalid.html

  
New tests
---------

  New tests have been introduced between CI_DRM_12020_full and IGTPW_7680_full:

### New IGT tests (1) ###

  * igt@perf@non-power-of-2-oa-formats:
    - Statuses : 8 skip(s)
    - Exec time: [0.0] s

  

Known issues
------------

  Here are the changes found in IGTPW_7680_full that come from known issues:

### IGT changes ###

#### Issues hit ####

  * igt@drm_buddy@all:
    - shard-iclb:         NOTRUN -> [SKIP][15] ([i915#6433])
   [15]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb5/igt@drm_buddy@all.html

  * igt@gem_create@create-ext-cpu-access-sanity-check:
    - shard-iclb:         NOTRUN -> [SKIP][16] ([i915#6335])
   [16]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb5/igt@gem_create@create-ext-cpu-access-sanity-check.html

  * igt@gem_create@create-massive:
    - shard-snb:          NOTRUN -> [DMESG-WARN][17] ([i915#4991])
   [17]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-snb4/igt@gem_create@create-massive.html
    - shard-apl:          NOTRUN -> [DMESG-WARN][18] ([i915#4991])
   [18]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-apl3/igt@gem_create@create-massive.html
    - shard-glk:          NOTRUN -> [DMESG-WARN][19] ([i915#4991])
   [19]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-glk8/igt@gem_create@create-massive.html

  * igt@gem_ctx_persistence@hostile:
    - shard-snb:          NOTRUN -> [SKIP][20] ([fdo#109271] / [i915#1099]) +1 similar issue
   [20]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-snb4/igt@gem_ctx_persistence@hostile.html

  * igt@gem_eio@kms:
    - shard-tglb:         [PASS][21] -> [FAIL][22] ([i915#5784])
   [21]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-tglb7/igt@gem_eio@kms.html
   [22]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-tglb3/igt@gem_eio@kms.html

  * igt@gem_exec_balancer@parallel-keep-in-fence:
    - shard-iclb:         [PASS][23] -> [SKIP][24] ([i915#4525]) +1 similar issue
   [23]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-iclb1/igt@gem_exec_balancer@parallel-keep-in-fence.html
   [24]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb6/igt@gem_exec_balancer@parallel-keep-in-fence.html

  * igt@gem_exec_capture@capture-recoverable:
    - shard-iclb:         NOTRUN -> [SKIP][25] ([i915#6344])
   [25]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb2/igt@gem_exec_capture@capture-recoverable.html

  * igt@gem_exec_fair@basic-deadline:
    - shard-kbl:          [PASS][26] -> [FAIL][27] ([i915#2846])
   [26]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-kbl7/igt@gem_exec_fair@basic-deadline.html
   [27]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl1/igt@gem_exec_fair@basic-deadline.html

  * igt@gem_exec_fair@basic-none-rrul@rcs0:
    - shard-glk:          NOTRUN -> [FAIL][28] ([i915#2842]) +1 similar issue
   [28]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-glk6/igt@gem_exec_fair@basic-none-rrul@rcs0.html
    - shard-kbl:          [PASS][29] -> [FAIL][30] ([i915#2842])
   [29]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-kbl1/igt@gem_exec_fair@basic-none-rrul@rcs0.html
   [30]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl7/igt@gem_exec_fair@basic-none-rrul@rcs0.html

  * igt@gem_exec_fair@basic-none-solo@rcs0:
    - shard-apl:          [PASS][31] -> [FAIL][32] ([i915#2842])
   [31]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-apl3/igt@gem_exec_fair@basic-none-solo@rcs0.html
   [32]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-apl3/igt@gem_exec_fair@basic-none-solo@rcs0.html

  * igt@gem_exec_fair@basic-none@vcs0:
    - shard-glk:          [PASS][33] -> [FAIL][34] ([i915#2842])
   [33]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-glk2/igt@gem_exec_fair@basic-none@vcs0.html
   [34]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-glk6/igt@gem_exec_fair@basic-none@vcs0.html

  * igt@gem_exec_fair@basic-none@vcs1:
    - shard-iclb:         NOTRUN -> [FAIL][35] ([i915#2842]) +6 similar issues
   [35]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb1/igt@gem_exec_fair@basic-none@vcs1.html

  * igt@gem_exec_params@no-vebox:
    - shard-iclb:         NOTRUN -> [SKIP][36] ([fdo#109283])
   [36]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb1/igt@gem_exec_params@no-vebox.html

  * igt@gem_exec_params@secure-non-root:
    - shard-iclb:         NOTRUN -> [SKIP][37] ([fdo#112283]) +1 similar issue
   [37]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb2/igt@gem_exec_params@secure-non-root.html

  * igt@gem_huc_copy@huc-copy:
    - shard-kbl:          NOTRUN -> [SKIP][38] ([fdo#109271] / [i915#2190])
   [38]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl7/igt@gem_huc_copy@huc-copy.html
    - shard-apl:          NOTRUN -> [SKIP][39] ([fdo#109271] / [i915#2190])
   [39]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-apl1/igt@gem_huc_copy@huc-copy.html
    - shard-glk:          NOTRUN -> [SKIP][40] ([fdo#109271] / [i915#2190])
   [40]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-glk9/igt@gem_huc_copy@huc-copy.html
    - shard-iclb:         NOTRUN -> [SKIP][41] ([i915#2190])
   [41]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb1/igt@gem_huc_copy@huc-copy.html

  * igt@gem_lmem_swapping@heavy-multi:
    - shard-kbl:          NOTRUN -> [SKIP][42] ([fdo#109271] / [i915#4613]) +1 similar issue
   [42]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl4/igt@gem_lmem_swapping@heavy-multi.html

  * igt@gem_lmem_swapping@parallel-random:
    - shard-iclb:         NOTRUN -> [SKIP][43] ([i915#4613]) +4 similar issues
   [43]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb1/igt@gem_lmem_swapping@parallel-random.html

  * igt@gem_lmem_swapping@verify-random-ccs:
    - shard-apl:          NOTRUN -> [SKIP][44] ([fdo#109271] / [i915#4613]) +4 similar issues
   [44]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-apl7/igt@gem_lmem_swapping@verify-random-ccs.html
    - shard-glk:          NOTRUN -> [SKIP][45] ([fdo#109271] / [i915#4613]) +4 similar issues
   [45]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-glk7/igt@gem_lmem_swapping@verify-random-ccs.html

  * igt@gem_pxp@dmabuf-shared-protected-dst-is-context-refcounted:
    - shard-iclb:         NOTRUN -> [SKIP][46] ([i915#4270]) +2 similar issues
   [46]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb2/igt@gem_pxp@dmabuf-shared-protected-dst-is-context-refcounted.html

  * igt@gem_render_copy@yf-tiled-to-vebox-x-tiled:
    - shard-iclb:         NOTRUN -> [SKIP][47] ([i915#768]) +4 similar issues
   [47]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb7/igt@gem_render_copy@yf-tiled-to-vebox-x-tiled.html

  * igt@gem_softpin@evict-single-offset:
    - shard-kbl:          NOTRUN -> [FAIL][48] ([i915#4171])
   [48]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl7/igt@gem_softpin@evict-single-offset.html

  * igt@gem_userptr_blits@dmabuf-sync:
    - shard-kbl:          NOTRUN -> [SKIP][49] ([fdo#109271] / [i915#3323])
   [49]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl4/igt@gem_userptr_blits@dmabuf-sync.html

  * igt@gem_userptr_blits@input-checking:
    - shard-kbl:          NOTRUN -> [DMESG-WARN][50] ([i915#4991])
   [50]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl1/igt@gem_userptr_blits@input-checking.html

  * igt@gem_userptr_blits@readonly-unsync:
    - shard-iclb:         NOTRUN -> [SKIP][51] ([i915#3297])
   [51]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb2/igt@gem_userptr_blits@readonly-unsync.html

  * igt@gem_vm_create@invalid-create:
    - shard-snb:          NOTRUN -> [SKIP][52] ([fdo#109271]) +309 similar issues
   [52]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-snb4/igt@gem_vm_create@invalid-create.html

  * igt@gen7_exec_parse@basic-offset:
    - shard-iclb:         NOTRUN -> [SKIP][53] ([fdo#109289]) +3 similar issues
   [53]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb5/igt@gen7_exec_parse@basic-offset.html

  * igt@gen9_exec_parse@allowed-all:
    - shard-iclb:         NOTRUN -> [SKIP][54] ([i915#2856]) +4 similar issues
   [54]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb7/igt@gen9_exec_parse@allowed-all.html

  * igt@gen9_exec_parse@batch-without-end:
    - shard-tglb:         NOTRUN -> [SKIP][55] ([i915#2527] / [i915#2856])
   [55]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-tglb7/igt@gen9_exec_parse@batch-without-end.html

  * igt@i915_pm_dc@dc3co-vpb-simulation:
    - shard-kbl:          NOTRUN -> [SKIP][56] ([fdo#109271] / [i915#658]) +2 similar issues
   [56]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl4/igt@i915_pm_dc@dc3co-vpb-simulation.html

  * igt@i915_pm_dc@dc6-dpms:
    - shard-kbl:          NOTRUN -> [FAIL][57] ([i915#454])
   [57]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl7/igt@i915_pm_dc@dc6-dpms.html

  * igt@i915_pm_rc6_residency@rc6-idle@vcs0:
    - shard-iclb:         NOTRUN -> [WARN][58] ([i915#2684]) +3 similar issues
   [58]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb7/igt@i915_pm_rc6_residency@rc6-idle@vcs0.html

  * igt@i915_pm_rpm@system-suspend:
    - shard-apl:          NOTRUN -> [INCOMPLETE][59] ([i915#4939] / [i915#6598])
   [59]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-apl2/igt@i915_pm_rpm@system-suspend.html

  * igt@i915_query@query-topology-unsupported:
    - shard-iclb:         NOTRUN -> [SKIP][60] ([fdo#109302])
   [60]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb5/igt@i915_query@query-topology-unsupported.html

  * igt@i915_query@test-query-geometry-subslices:
    - shard-iclb:         NOTRUN -> [SKIP][61] ([i915#5723])
   [61]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb7/igt@i915_query@test-query-geometry-subslices.html

  * igt@i915_suspend@forcewake:
    - shard-glk:          NOTRUN -> [INCOMPLETE][62] ([i915#6598]) +6 similar issues
   [62]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-glk1/igt@i915_suspend@forcewake.html

  * igt@kms_async_flips@alternate-sync-async-flip@pipe-a-hdmi-a-1:
    - shard-glk:          [PASS][63] -> [FAIL][64] ([i915#2521])
   [63]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-glk5/igt@kms_async_flips@alternate-sync-async-flip@pipe-a-hdmi-a-1.html
   [64]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-glk1/igt@kms_async_flips@alternate-sync-async-flip@pipe-a-hdmi-a-1.html

  * igt@kms_atomic@atomic_plane_damage:
    - shard-iclb:         NOTRUN -> [SKIP][65] ([i915#4765])
   [65]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb1/igt@kms_atomic@atomic_plane_damage.html

  * igt@kms_big_fb@4-tiled-16bpp-rotate-90:
    - shard-iclb:         NOTRUN -> [SKIP][66] ([i915#5286]) +8 similar issues
   [66]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb2/igt@kms_big_fb@4-tiled-16bpp-rotate-90.html

  * igt@kms_big_fb@4-tiled-max-hw-stride-32bpp-rotate-180:
    - shard-tglb:         NOTRUN -> [SKIP][67] ([i915#5286])
   [67]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-tglb7/igt@kms_big_fb@4-tiled-max-hw-stride-32bpp-rotate-180.html

  * igt@kms_big_fb@linear-64bpp-rotate-270:
    - shard-iclb:         NOTRUN -> [SKIP][68] ([fdo#110725] / [fdo#111614]) +6 similar issues
   [68]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb7/igt@kms_big_fb@linear-64bpp-rotate-270.html

  * igt@kms_big_fb@yf-tiled-max-hw-stride-64bpp-rotate-0:
    - shard-iclb:         NOTRUN -> [SKIP][69] ([fdo#110723]) +3 similar issues
   [69]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb5/igt@kms_big_fb@yf-tiled-max-hw-stride-64bpp-rotate-0.html

  * igt@kms_ccs@pipe-a-crc-primary-rotation-180-y_tiled_gen12_rc_ccs_cc:
    - shard-apl:          NOTRUN -> [SKIP][70] ([fdo#109271] / [i915#3886]) +12 similar issues
   [70]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-apl6/igt@kms_ccs@pipe-a-crc-primary-rotation-180-y_tiled_gen12_rc_ccs_cc.html

  * igt@kms_ccs@pipe-a-crc-sprite-planes-basic-y_tiled_gen12_rc_ccs_cc:
    - shard-kbl:          NOTRUN -> [SKIP][71] ([fdo#109271] / [i915#3886]) +2 similar issues
   [71]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl7/igt@kms_ccs@pipe-a-crc-sprite-planes-basic-y_tiled_gen12_rc_ccs_cc.html

  * igt@kms_ccs@pipe-b-crc-primary-rotation-180-y_tiled_gen12_rc_ccs_cc:
    - shard-iclb:         NOTRUN -> [SKIP][72] ([fdo#109278] / [i915#3886]) +12 similar issues
   [72]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb7/igt@kms_ccs@pipe-b-crc-primary-rotation-180-y_tiled_gen12_rc_ccs_cc.html

  * igt@kms_ccs@pipe-c-crc-sprite-planes-basic-y_tiled_gen12_rc_ccs_cc:
    - shard-glk:          NOTRUN -> [SKIP][73] ([fdo#109271] / [i915#3886]) +11 similar issues
   [73]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-glk5/igt@kms_ccs@pipe-c-crc-sprite-planes-basic-y_tiled_gen12_rc_ccs_cc.html

  * igt@kms_ccs@pipe-d-crc-primary-basic-4_tiled_dg2_rc_ccs:
    - shard-iclb:         NOTRUN -> [SKIP][74] ([fdo#109278]) +32 similar issues
   [74]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb5/igt@kms_ccs@pipe-d-crc-primary-basic-4_tiled_dg2_rc_ccs.html

  * igt@kms_cdclk@mode-transition:
    - shard-iclb:         NOTRUN -> [SKIP][75] ([i915#3742])
   [75]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb2/igt@kms_cdclk@mode-transition.html

  * igt@kms_chamelium@dp-frame-dump:
    - shard-apl:          NOTRUN -> [SKIP][76] ([fdo#109271] / [fdo#111827]) +7 similar issues
   [76]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-apl3/igt@kms_chamelium@dp-frame-dump.html

  * igt@kms_chamelium@hdmi-crc-multiple:
    - shard-glk:          NOTRUN -> [SKIP][77] ([fdo#109271] / [fdo#111827]) +7 similar issues
   [77]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-glk9/igt@kms_chamelium@hdmi-crc-multiple.html

  * igt@kms_chamelium@hdmi-crc-single:
    - shard-iclb:         NOTRUN -> [SKIP][78] ([fdo#109284] / [fdo#111827]) +9 similar issues
   [78]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb8/igt@kms_chamelium@hdmi-crc-single.html

  * igt@kms_color_chamelium@ctm-green-to-red:
    - shard-snb:          NOTRUN -> [SKIP][79] ([fdo#109271] / [fdo#111827]) +8 similar issues
   [79]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-snb4/igt@kms_color_chamelium@ctm-green-to-red.html

  * igt@kms_color_chamelium@gamma:
    - shard-kbl:          NOTRUN -> [SKIP][80] ([fdo#109271] / [fdo#111827]) +4 similar issues
   [80]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl7/igt@kms_color_chamelium@gamma.html

  * igt@kms_content_protection@atomic:
    - shard-apl:          NOTRUN -> [TIMEOUT][81] ([i915#1319]) +2 similar issues
   [81]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-apl3/igt@kms_content_protection@atomic.html

  * igt@kms_content_protection@dp-mst-lic-type-1:
    - shard-iclb:         NOTRUN -> [SKIP][82] ([i915#3116])
   [82]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb5/igt@kms_content_protection@dp-mst-lic-type-1.html

  * igt@kms_content_protection@lic:
    - shard-iclb:         NOTRUN -> [SKIP][83] ([fdo#109300] / [fdo#111066]) +3 similar issues
   [83]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb7/igt@kms_content_protection@lic.html
    - shard-kbl:          NOTRUN -> [TIMEOUT][84] ([i915#1319] / [i915#6637]) +1 similar issue
   [84]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl7/igt@kms_content_protection@lic.html
    - shard-tglb:         NOTRUN -> [SKIP][85] ([i915#1063])
   [85]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-tglb3/igt@kms_content_protection@lic.html

  * igt@kms_cursor_legacy@flip-vs-cursor@atomic-transitions-varying-size:
    - shard-glk:          [PASS][86] -> [FAIL][87] ([i915#2346])
   [86]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-glk1/igt@kms_cursor_legacy@flip-vs-cursor@atomic-transitions-varying-size.html
   [87]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-glk3/igt@kms_cursor_legacy@flip-vs-cursor@atomic-transitions-varying-size.html

  * igt@kms_dp_tiled_display@basic-test-pattern-with-chamelium:
    - shard-iclb:         NOTRUN -> [SKIP][88] ([i915#3528])
   [88]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb5/igt@kms_dp_tiled_display@basic-test-pattern-with-chamelium.html

  * igt@kms_draw_crc@draw-method-rgb565-mmap-gtt-4tiled:
    - shard-iclb:         NOTRUN -> [SKIP][89] ([i915#5287]) +3 similar issues
   [89]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb6/igt@kms_draw_crc@draw-method-rgb565-mmap-gtt-4tiled.html

  * igt@kms_dsc@dsc-with-bpp:
    - shard-iclb:         NOTRUN -> [SKIP][90] ([i915#3828])
   [90]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb8/igt@kms_dsc@dsc-with-bpp.html

  * igt@kms_flip@2x-flip-vs-panning-interruptible:
    - shard-iclb:         NOTRUN -> [SKIP][91] ([fdo#109274]) +6 similar issues
   [91]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb6/igt@kms_flip@2x-flip-vs-panning-interruptible.html

  * igt@kms_flip@flip-vs-absolute-wf_vblank-interruptible@c-hdmi-a1:
    - shard-glk:          [PASS][92] -> [FAIL][93] ([i915#2122])
   [92]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-glk9/igt@kms_flip@flip-vs-absolute-wf_vblank-interruptible@c-hdmi-a1.html
   [93]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-glk6/igt@kms_flip@flip-vs-absolute-wf_vblank-interruptible@c-hdmi-a1.html

  * igt@kms_flip@flip-vs-suspend@a-dp1:
    - shard-kbl:          [PASS][94] -> [DMESG-WARN][95] ([i915#180]) +7 similar issues
   [94]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-kbl7/igt@kms_flip@flip-vs-suspend@a-dp1.html
   [95]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl4/igt@kms_flip@flip-vs-suspend@a-dp1.html

  * igt@kms_flip_scaled_crc@flip-32bpp-yftileccs-to-64bpp-yftile-upscaling@pipe-a-valid-mode:
    - shard-iclb:         NOTRUN -> [SKIP][96] ([i915#2672]) +7 similar issues
   [96]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb8/igt@kms_flip_scaled_crc@flip-32bpp-yftileccs-to-64bpp-yftile-upscaling@pipe-a-valid-mode.html

  * igt@kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-downscaling@pipe-a-default-mode:
    - shard-iclb:         NOTRUN -> [SKIP][97] ([i915#3555]) +2 similar issues
   [97]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb2/igt@kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-downscaling@pipe-a-default-mode.html

  * igt@kms_flip_scaled_crc@flip-64bpp-yftile-to-16bpp-yftile-upscaling@pipe-a-valid-mode:
    - shard-tglb:         NOTRUN -> [SKIP][98] ([i915#2672])
   [98]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-tglb5/igt@kms_flip_scaled_crc@flip-64bpp-yftile-to-16bpp-yftile-upscaling@pipe-a-valid-mode.html

  * igt@kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilegen12rcccs-upscaling@pipe-a-default-mode:
    - shard-iclb:         NOTRUN -> [SKIP][99] ([i915#2672] / [i915#3555])
   [99]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb2/igt@kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilegen12rcccs-upscaling@pipe-a-default-mode.html

  * igt@kms_force_connector_basic@force-load-detect:
    - shard-iclb:         NOTRUN -> [SKIP][100] ([fdo#109285])
   [100]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb2/igt@kms_force_connector_basic@force-load-detect.html

  * igt@kms_frontbuffer_tracking@fbc-2p-primscrn-cur-indfb-draw-mmap-cpu:
    - shard-tglb:         NOTRUN -> [SKIP][101] ([fdo#109280] / [fdo#111825]) +1 similar issue
   [101]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-tglb3/igt@kms_frontbuffer_tracking@fbc-2p-primscrn-cur-indfb-draw-mmap-cpu.html

  * igt@kms_frontbuffer_tracking@fbc-2p-primscrn-cur-indfb-draw-render:
    - shard-kbl:          NOTRUN -> [SKIP][102] ([fdo#109271]) +141 similar issues
   [102]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl4/igt@kms_frontbuffer_tracking@fbc-2p-primscrn-cur-indfb-draw-render.html

  * igt@kms_frontbuffer_tracking@fbc-2p-scndscrn-pri-shrfb-draw-pwrite:
    - shard-glk:          NOTRUN -> [FAIL][103] ([i915#1888] / [i915#2546])
   [103]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-glk9/igt@kms_frontbuffer_tracking@fbc-2p-scndscrn-pri-shrfb-draw-pwrite.html

  * igt@kms_frontbuffer_tracking@fbc-suspend:
    - shard-iclb:         NOTRUN -> [INCOMPLETE][104] ([i915#6598]) +4 similar issues
   [104]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb1/igt@kms_frontbuffer_tracking@fbc-suspend.html
    - shard-snb:          NOTRUN -> [INCOMPLETE][105] ([i915#6598] / [i915#6625])
   [105]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-snb6/igt@kms_frontbuffer_tracking@fbc-suspend.html

  * igt@kms_frontbuffer_tracking@fbcpsr-1p-primscrn-cur-indfb-draw-mmap-gtt:
    - shard-tglb:         NOTRUN -> [SKIP][106] ([i915#6497])
   [106]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-tglb3/igt@kms_frontbuffer_tracking@fbcpsr-1p-primscrn-cur-indfb-draw-mmap-gtt.html

  * igt@kms_frontbuffer_tracking@fbcpsr-tiling-4:
    - shard-iclb:         NOTRUN -> [SKIP][107] ([i915#5438])
   [107]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb2/igt@kms_frontbuffer_tracking@fbcpsr-tiling-4.html

  * igt@kms_frontbuffer_tracking@psr-2p-primscrn-pri-indfb-draw-mmap-wc:
    - shard-iclb:         NOTRUN -> [SKIP][108] ([fdo#109280]) +47 similar issues
   [108]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb2/igt@kms_frontbuffer_tracking@psr-2p-primscrn-pri-indfb-draw-mmap-wc.html

  * igt@kms_hdr@bpc-switch@pipe-a-dp-1:
    - shard-kbl:          [PASS][109] -> [FAIL][110] ([i915#1188])
   [109]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-kbl4/igt@kms_hdr@bpc-switch@pipe-a-dp-1.html
   [110]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl4/igt@kms_hdr@bpc-switch@pipe-a-dp-1.html

  * igt@kms_pipe_crc_basic@suspend-read-crc@pipe-b-dp-1:
    - shard-apl:          NOTRUN -> [INCOMPLETE][111] ([i915#6598]) +2 similar issues
   [111]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-apl2/igt@kms_pipe_crc_basic@suspend-read-crc@pipe-b-dp-1.html

  * igt@kms_pipe_crc_basic@suspend-read-crc@pipe-b-vga-1:
    - shard-snb:          NOTRUN -> [INCOMPLETE][112] ([i915#6520] / [i915#6598])
   [112]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-snb5/igt@kms_pipe_crc_basic@suspend-read-crc@pipe-b-vga-1.html

  * igt@kms_plane_alpha_blend@pipe-a-alpha-opaque-fb:
    - shard-glk:          NOTRUN -> [FAIL][113] ([fdo#108145] / [i915#265]) +2 similar issues
   [113]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-glk7/igt@kms_plane_alpha_blend@pipe-a-alpha-opaque-fb.html
    - shard-kbl:          NOTRUN -> [FAIL][114] ([fdo#108145] / [i915#265]) +1 similar issue
   [114]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl4/igt@kms_plane_alpha_blend@pipe-a-alpha-opaque-fb.html

  * igt@kms_plane_alpha_blend@pipe-a-alpha-transparent-fb:
    - shard-apl:          NOTRUN -> [FAIL][115] ([i915#265])
   [115]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-apl7/igt@kms_plane_alpha_blend@pipe-a-alpha-transparent-fb.html
    - shard-glk:          NOTRUN -> [FAIL][116] ([i915#265])
   [116]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-glk1/igt@kms_plane_alpha_blend@pipe-a-alpha-transparent-fb.html

  * igt@kms_plane_alpha_blend@pipe-b-alpha-opaque-fb:
    - shard-apl:          NOTRUN -> [FAIL][117] ([fdo#108145] / [i915#265]) +2 similar issues
   [117]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-apl6/igt@kms_plane_alpha_blend@pipe-b-alpha-opaque-fb.html

  * igt@kms_plane_scaling@plane-scaler-with-clipping-clamping-rotation@pipe-c-edp-1:
    - shard-iclb:         NOTRUN -> [SKIP][118] ([i915#5176]) +5 similar issues
   [118]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb7/igt@kms_plane_scaling@plane-scaler-with-clipping-clamping-rotation@pipe-c-edp-1.html

  * igt@kms_plane_scaling@planes-unity-scaling-downscale-factor-0-25@pipe-b-edp-1:
    - shard-iclb:         NOTRUN -> [SKIP][119] ([i915#5235]) +2 similar issues
   [119]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb6/igt@kms_plane_scaling@planes-unity-scaling-downscale-factor-0-25@pipe-b-edp-1.html

  * igt@kms_psr2_sf@cursor-plane-update-sf:
    - shard-iclb:         NOTRUN -> [SKIP][120] ([fdo#111068] / [i915#658])
   [120]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb1/igt@kms_psr2_sf@cursor-plane-update-sf.html

  * igt@kms_psr2_sf@overlay-plane-move-continuous-sf:
    - shard-apl:          NOTRUN -> [SKIP][121] ([fdo#109271] / [i915#658]) +4 similar issues
   [121]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-apl3/igt@kms_psr2_sf@overlay-plane-move-continuous-sf.html

  * igt@kms_psr2_sf@overlay-plane-update-continuous-sf:
    - shard-iclb:         NOTRUN -> [SKIP][122] ([i915#2920])
   [122]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb2/igt@kms_psr2_sf@overlay-plane-update-continuous-sf.html

  * igt@kms_psr2_su@frontbuffer-xrgb8888:
    - shard-glk:          NOTRUN -> [SKIP][123] ([fdo#109271] / [i915#658]) +3 similar issues
   [123]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-glk2/igt@kms_psr2_su@frontbuffer-xrgb8888.html
    - shard-iclb:         NOTRUN -> [SKIP][124] ([fdo#109642] / [fdo#111068] / [i915#658])
   [124]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb3/igt@kms_psr2_su@frontbuffer-xrgb8888.html

  * igt@kms_psr@psr2_no_drrs:
    - shard-iclb:         NOTRUN -> [SKIP][125] ([fdo#109441]) +2 similar issues
   [125]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb1/igt@kms_psr@psr2_no_drrs.html

  * igt@kms_psr@psr2_sprite_plane_move:
    - shard-iclb:         [PASS][126] -> [SKIP][127] ([fdo#109441]) +1 similar issue
   [126]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-iclb2/igt@kms_psr@psr2_sprite_plane_move.html
   [127]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb3/igt@kms_psr@psr2_sprite_plane_move.html

  * igt@kms_psr@psr2_sprite_plane_onoff:
    - shard-apl:          NOTRUN -> [SKIP][128] ([fdo#109271]) +231 similar issues
   [128]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-apl7/igt@kms_psr@psr2_sprite_plane_onoff.html

  * igt@kms_rotation_crc@primary-4-tiled-reflect-x-0:
    - shard-iclb:         NOTRUN -> [SKIP][129] ([i915#5289])
   [129]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb8/igt@kms_rotation_crc@primary-4-tiled-reflect-x-0.html

  * igt@kms_scaling_modes@scaling-mode-none@edp-1-pipe-b:
    - shard-iclb:         NOTRUN -> [SKIP][130] ([i915#5030]) +2 similar issues
   [130]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb7/igt@kms_scaling_modes@scaling-mode-none@edp-1-pipe-b.html

  * igt@kms_tv_load_detect@load-detect:
    - shard-iclb:         NOTRUN -> [SKIP][131] ([fdo#109309])
   [131]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb1/igt@kms_tv_load_detect@load-detect.html

  * igt@kms_vblank@pipe-c-ts-continuation-dpms-suspend:
    - shard-iclb:         [PASS][132] -> [INCOMPLETE][133] ([i915#6598])
   [132]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-iclb5/igt@kms_vblank@pipe-c-ts-continuation-dpms-suspend.html
   [133]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb7/igt@kms_vblank@pipe-c-ts-continuation-dpms-suspend.html
    - shard-apl:          [PASS][134] -> [INCOMPLETE][135] ([i915#4939] / [i915#6598])
   [134]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-apl2/igt@kms_vblank@pipe-c-ts-continuation-dpms-suspend.html
   [135]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-apl3/igt@kms_vblank@pipe-c-ts-continuation-dpms-suspend.html

  * igt@kms_vblank@pipe-d-wait-busy-hang:
    - shard-glk:          NOTRUN -> [SKIP][136] ([fdo#109271]) +192 similar issues
   [136]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-glk1/igt@kms_vblank@pipe-d-wait-busy-hang.html

  * igt@kms_vblank@pipe-d-wait-idle:
    - shard-apl:          NOTRUN -> [SKIP][137] ([fdo#109271] / [i915#533])
   [137]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-apl6/igt@kms_vblank@pipe-d-wait-idle.html
    - shard-glk:          NOTRUN -> [SKIP][138] ([fdo#109271] / [i915#533])
   [138]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-glk8/igt@kms_vblank@pipe-d-wait-idle.html

  * igt@kms_writeback@writeback-invalid-parameters:
    - shard-apl:          NOTRUN -> [SKIP][139] ([fdo#109271] / [i915#2437]) +1 similar issue
   [139]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-apl8/igt@kms_writeback@writeback-invalid-parameters.html
    - shard-glk:          NOTRUN -> [SKIP][140] ([fdo#109271] / [i915#2437])
   [140]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-glk5/igt@kms_writeback@writeback-invalid-parameters.html
    - shard-iclb:         NOTRUN -> [SKIP][141] ([i915#2437])
   [141]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb1/igt@kms_writeback@writeback-invalid-parameters.html

  * igt@nouveau_crc@pipe-a-ctx-flip-skip-current-frame:
    - shard-iclb:         NOTRUN -> [SKIP][142] ([i915#2530]) +1 similar issue
   [142]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb5/igt@nouveau_crc@pipe-a-ctx-flip-skip-current-frame.html

  * igt@perf@global-sseu-config-invalid:
    - shard-glk:          [PASS][143] -> [SKIP][144] ([fdo#109271])
   [143]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-glk7/igt@perf@global-sseu-config-invalid.html
   [144]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-glk8/igt@perf@global-sseu-config-invalid.html
    - shard-kbl:          [PASS][145] -> [SKIP][146] ([fdo#109271]) +1 similar issue
   [145]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-kbl7/igt@perf@global-sseu-config-invalid.html
   [146]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl1/igt@perf@global-sseu-config-invalid.html
    - shard-apl:          [PASS][147] -> [SKIP][148] ([fdo#109271])
   [147]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-apl2/igt@perf@global-sseu-config-invalid.html
   [148]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-apl8/igt@perf@global-sseu-config-invalid.html

  * igt@perf_pmu@rc6-suspend:
    - shard-apl:          [PASS][149] -> [DMESG-WARN][150] ([i915#180])
   [149]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-apl2/igt@perf_pmu@rc6-suspend.html
   [150]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-apl1/igt@perf_pmu@rc6-suspend.html

  * igt@prime_nv_pcopy@test3_2:
    - shard-iclb:         NOTRUN -> [SKIP][151] ([fdo#109291]) +6 similar issues
   [151]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb5/igt@prime_nv_pcopy@test3_2.html

  * igt@prime_vgem@coherency-gtt:
    - shard-iclb:         NOTRUN -> [SKIP][152] ([fdo#109292] / [fdo#109295])
   [152]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb6/igt@prime_vgem@coherency-gtt.html

  * igt@sysfs_clients@busy:
    - shard-iclb:         NOTRUN -> [SKIP][153] ([i915#2994]) +1 similar issue
   [153]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb6/igt@sysfs_clients@busy.html

  * igt@sysfs_clients@sema-25:
    - shard-kbl:          NOTRUN -> [SKIP][154] ([fdo#109271] / [i915#2994])
   [154]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl7/igt@sysfs_clients@sema-25.html
    - shard-apl:          NOTRUN -> [SKIP][155] ([fdo#109271] / [i915#2994]) +1 similar issue
   [155]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-apl3/igt@sysfs_clients@sema-25.html
    - shard-glk:          NOTRUN -> [SKIP][156] ([fdo#109271] / [i915#2994])
   [156]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-glk9/igt@sysfs_clients@sema-25.html

  
#### Possible fixes ####

  * igt@gem_ctx_exec@basic-nohangcheck:
    - shard-tglb:         [FAIL][157] ([i915#6268]) -> [PASS][158]
   [157]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-tglb7/igt@gem_ctx_exec@basic-nohangcheck.html
   [158]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-tglb7/igt@gem_ctx_exec@basic-nohangcheck.html
    - {shard-tglu}:       [FAIL][159] ([i915#6268]) -> [PASS][160]
   [159]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-tglu-5/igt@gem_ctx_exec@basic-nohangcheck.html
   [160]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-tglu-1/igt@gem_ctx_exec@basic-nohangcheck.html

  * igt@gem_ctx_persistence@many-contexts:
    - {shard-rkl}:        [FAIL][161] ([i915#2410]) -> [PASS][162]
   [161]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-rkl-1/igt@gem_ctx_persistence@many-contexts.html
   [162]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-rkl-5/igt@gem_ctx_persistence@many-contexts.html

  * igt@gem_eio@in-flight-contexts-10ms:
    - {shard-tglu}:       [TIMEOUT][163] ([i915#3063]) -> [PASS][164]
   [163]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-tglu-2/igt@gem_eio@in-flight-contexts-10ms.html
   [164]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-tglu-5/igt@gem_eio@in-flight-contexts-10ms.html

  * igt@gem_exec_balancer@parallel-bb-first:
    - shard-iclb:         [SKIP][165] ([i915#4525]) -> [PASS][166]
   [165]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-iclb6/igt@gem_exec_balancer@parallel-bb-first.html
   [166]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb2/igt@gem_exec_balancer@parallel-bb-first.html

  * igt@gem_exec_fair@basic-none-share@rcs0:
    - shard-iclb:         [FAIL][167] ([i915#2842]) -> [PASS][168]
   [167]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-iclb2/igt@gem_exec_fair@basic-none-share@rcs0.html
   [168]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb8/igt@gem_exec_fair@basic-none-share@rcs0.html

  * igt@gem_exec_fair@basic-pace-solo@rcs0:
    - shard-apl:          [FAIL][169] ([i915#2842]) -> [PASS][170]
   [169]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-apl7/igt@gem_exec_fair@basic-pace-solo@rcs0.html
   [170]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-apl2/igt@gem_exec_fair@basic-pace-solo@rcs0.html

  * igt@gem_exec_fair@basic-pace@rcs0:
    - shard-kbl:          [FAIL][171] ([i915#2842]) -> [PASS][172] +4 similar issues
   [171]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-kbl7/igt@gem_exec_fair@basic-pace@rcs0.html
   [172]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl4/igt@gem_exec_fair@basic-pace@rcs0.html

  * igt@gem_exec_reloc@basic-wc-active:
    - {shard-rkl}:        [SKIP][173] ([i915#3281]) -> [PASS][174] +3 similar issues
   [173]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-rkl-2/igt@gem_exec_reloc@basic-wc-active.html
   [174]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-rkl-5/igt@gem_exec_reloc@basic-wc-active.html

  * igt@gem_exec_suspend@basic-s0@smem:
    - shard-snb:          [INCOMPLETE][175] ([i915#6598]) -> [PASS][176]
   [175]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-snb4/igt@gem_exec_suspend@basic-s0@smem.html
   [176]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-snb2/igt@gem_exec_suspend@basic-s0@smem.html

  * igt@gem_huc_copy@huc-copy:
    - shard-tglb:         [SKIP][177] ([i915#2190]) -> [PASS][178]
   [177]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-tglb7/igt@gem_huc_copy@huc-copy.html
   [178]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-tglb3/igt@gem_huc_copy@huc-copy.html

  * igt@gem_mmap_gtt@coherency:
    - {shard-rkl}:        [SKIP][179] ([fdo#111656]) -> [PASS][180]
   [179]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-rkl-1/igt@gem_mmap_gtt@coherency.html
   [180]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-rkl-5/igt@gem_mmap_gtt@coherency.html

  * igt@gem_userptr_blits@forbidden-operations:
    - {shard-rkl}:        [SKIP][181] ([i915#3282]) -> [PASS][182] +1 similar issue
   [181]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-rkl-2/igt@gem_userptr_blits@forbidden-operations.html
   [182]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-rkl-5/igt@gem_userptr_blits@forbidden-operations.html

  * igt@gem_workarounds@suspend-resume-fd:
    - shard-snb:          [SKIP][183] ([fdo#109271]) -> [PASS][184]
   [183]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-snb5/igt@gem_workarounds@suspend-resume-fd.html
   [184]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-snb2/igt@gem_workarounds@suspend-resume-fd.html

  * igt@i915_pm_rpm@system-suspend-execbuf:
    - shard-apl:          [INCOMPLETE][185] ([i915#4939]) -> [PASS][186]
   [185]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-apl8/igt@i915_pm_rpm@system-suspend-execbuf.html
   [186]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-apl6/igt@i915_pm_rpm@system-suspend-execbuf.html
    - shard-glk:          [INCOMPLETE][187] -> [PASS][188]
   [187]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-glk3/igt@i915_pm_rpm@system-suspend-execbuf.html
   [188]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-glk1/igt@i915_pm_rpm@system-suspend-execbuf.html

  * igt@i915_suspend@fence-restore-tiled2untiled:
    - shard-kbl:          [DMESG-WARN][189] ([i915#180]) -> [PASS][190] +2 similar issues
   [189]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-kbl4/igt@i915_suspend@fence-restore-tiled2untiled.html
   [190]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl1/igt@i915_suspend@fence-restore-tiled2untiled.html

  * igt@i915_suspend@sysfs-reader:
    - shard-apl:          [DMESG-WARN][191] ([i915#180]) -> [PASS][192]
   [191]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-apl1/igt@i915_suspend@sysfs-reader.html
   [192]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-apl3/igt@i915_suspend@sysfs-reader.html

  * igt@kms_big_fb@x-tiled-max-hw-stride-32bpp-rotate-180-hflip-async-flip:
    - {shard-rkl}:        [SKIP][193] ([i915#1845] / [i915#4098]) -> [PASS][194] +9 similar issues
   [193]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-rkl-1/igt@kms_big_fb@x-tiled-max-hw-stride-32bpp-rotate-180-hflip-async-flip.html
   [194]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-rkl-6/igt@kms_big_fb@x-tiled-max-hw-stride-32bpp-rotate-180-hflip-async-flip.html

  * igt@kms_big_fb@y-tiled-8bpp-rotate-0:
    - shard-glk:          [FAIL][195] ([i915#1888] / [i915#5138]) -> [PASS][196]
   [195]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-glk3/igt@kms_big_fb@y-tiled-8bpp-rotate-0.html
   [196]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-glk3/igt@kms_big_fb@y-tiled-8bpp-rotate-0.html

  * igt@kms_draw_crc@draw-method-xrgb2101010-blt-untiled:
    - {shard-rkl}:        [SKIP][197] ([fdo#111314] / [i915#4098] / [i915#4369]) -> [PASS][198]
   [197]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-rkl-1/igt@kms_draw_crc@draw-method-xrgb2101010-blt-untiled.html
   [198]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-rkl-6/igt@kms_draw_crc@draw-method-xrgb2101010-blt-untiled.html

  * igt@kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-upscaling@pipe-a-valid-mode:
    - shard-glk:          [FAIL][199] ([i915#1888]) -> [PASS][200]
   [199]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-glk3/igt@kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-upscaling@pipe-a-valid-mode.html
   [200]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-glk3/igt@kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-upscaling@pipe-a-valid-mode.html

  * igt@kms_frontbuffer_tracking@fbc-1p-primscrn-pri-indfb-draw-pwrite:
    - {shard-rkl}:        [SKIP][201] ([i915#1849] / [i915#4098]) -> [PASS][202] +4 similar issues
   [201]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-rkl-1/igt@kms_frontbuffer_tracking@fbc-1p-primscrn-pri-indfb-draw-pwrite.html
   [202]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-rkl-6/igt@kms_frontbuffer_tracking@fbc-1p-primscrn-pri-indfb-draw-pwrite.html

  * igt@kms_plane@plane-position-hole-dpms@pipe-b-planes:
    - {shard-rkl}:        [SKIP][203] ([i915#1849] / [i915#3558]) -> [PASS][204] +1 similar issue
   [203]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-rkl-2/igt@kms_plane@plane-position-hole-dpms@pipe-b-planes.html
   [204]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-rkl-6/igt@kms_plane@plane-position-hole-dpms@pipe-b-planes.html

  * igt@kms_plane_alpha_blend@pipe-b-coverage-7efc:
    - {shard-rkl}:        [SKIP][205] ([i915#1849] / [i915#3546] / [i915#4070] / [i915#4098]) -> [PASS][206]
   [205]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-rkl-2/igt@kms_plane_alpha_blend@pipe-b-coverage-7efc.html
   [206]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-rkl-6/igt@kms_plane_alpha_blend@pipe-b-coverage-7efc.html

  * igt@kms_plane_multiple@atomic-pipe-b-tiling-x:
    - {shard-rkl}:        [SKIP][207] ([i915#1849] / [i915#3558] / [i915#4070]) -> [PASS][208]
   [207]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-rkl-2/igt@kms_plane_multiple@atomic-pipe-b-tiling-x.html
   [208]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-rkl-6/igt@kms_plane_multiple@atomic-pipe-b-tiling-x.html

  * igt@kms_plane_scaling@planes-downscale-factor-0-5@pipe-a-edp-1:
    - shard-iclb:         [SKIP][209] ([i915#5235]) -> [PASS][210] +2 similar issues
   [209]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-iclb2/igt@kms_plane_scaling@planes-downscale-factor-0-5@pipe-a-edp-1.html
   [210]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb6/igt@kms_plane_scaling@planes-downscale-factor-0-5@pipe-a-edp-1.html

  * igt@kms_psr@psr2_cursor_plane_move:
    - shard-iclb:         [SKIP][211] ([fdo#109441]) -> [PASS][212]
   [211]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-iclb3/igt@kms_psr@psr2_cursor_plane_move.html
   [212]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb2/igt@kms_psr@psr2_cursor_plane_move.html

  * igt@kms_psr@suspend:
    - shard-iclb:         [INCOMPLETE][213] -> [PASS][214] +1 similar issue
   [213]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-iclb1/igt@kms_psr@suspend.html
   [214]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb2/igt@kms_psr@suspend.html

  * igt@kms_psr_stress_test@flip-primary-invalidate-overlay:
    - shard-iclb:         [SKIP][215] ([i915#5519]) -> [PASS][216]
   [215]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-iclb7/igt@kms_psr_stress_test@flip-primary-invalidate-overlay.html
   [216]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb7/igt@kms_psr_stress_test@flip-primary-invalidate-overlay.html

  * igt@kms_vblank@pipe-a-ts-continuation-suspend:
    - shard-snb:          [INCOMPLETE][217] ([i915#4939] / [i915#6598]) -> [PASS][218]
   [217]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-snb2/igt@kms_vblank@pipe-a-ts-continuation-suspend.html
   [218]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-snb2/igt@kms_vblank@pipe-a-ts-continuation-suspend.html
    - shard-apl:          [INCOMPLETE][219] ([i915#4939] / [i915#6598]) -> [PASS][220]
   [219]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-apl8/igt@kms_vblank@pipe-a-ts-continuation-suspend.html
   [220]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-apl7/igt@kms_vblank@pipe-a-ts-continuation-suspend.html

  
#### Warnings ####

  * igt@gem_eio@unwedge-stress:
    - shard-tglb:         [FAIL][221] ([i915#5784]) -> [TIMEOUT][222] ([i915#3063])
   [221]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-tglb5/igt@gem_eio@unwedge-stress.html
   [222]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-tglb7/igt@gem_eio@unwedge-stress.html

  * igt@i915_pm_dc@dc3co-vpb-simulation:
    - shard-iclb:         [SKIP][223] ([i915#658]) -> [SKIP][224] ([i915#588])
   [223]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-iclb3/igt@i915_pm_dc@dc3co-vpb-simulation.html
   [224]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb2/igt@i915_pm_dc@dc3co-vpb-simulation.html

  * igt@kms_fbcon_fbt@fbc-suspend:
    - shard-kbl:          [INCOMPLETE][225] ([i915#180] / [i915#4939] / [i915#6598]) -> [FAIL][226] ([i915#4767])
   [225]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-kbl4/igt@kms_fbcon_fbt@fbc-suspend.html
   [226]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl7/igt@kms_fbcon_fbt@fbc-suspend.html

  * igt@kms_flip@flip-vs-suspend-interruptible@a-dp1:
    - shard-apl:          [INCOMPLETE][227] ([i915#1982] / [i915#6598]) -> [INCOMPLETE][228] ([i915#6598])
   [227]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-apl7/igt@kms_flip@flip-vs-suspend-interruptible@a-dp1.html
   [228]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-apl8/igt@kms_flip@flip-vs-suspend-interruptible@a-dp1.html

  * igt@kms_flip@flip-vs-suspend-interruptible@a-hdmi-a1:
    - shard-glk:          [INCOMPLETE][229] ([i915#1982] / [i915#6598]) -> [INCOMPLETE][230] ([i915#6598])
   [229]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-glk6/igt@kms_flip@flip-vs-suspend-interruptible@a-hdmi-a1.html
   [230]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-glk5/igt@kms_flip@flip-vs-suspend-interruptible@a-hdmi-a1.html

  * igt@kms_flip@flip-vs-suspend@a-edp1:
    - shard-iclb:         [INCOMPLETE][231] ([i915#6598]) -> [INCOMPLETE][232] ([i915#6598] / [i915#6604])
   [231]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-iclb6/igt@kms_flip@flip-vs-suspend@a-edp1.html
   [232]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb7/igt@kms_flip@flip-vs-suspend@a-edp1.html

  * igt@kms_psr2_sf@cursor-plane-move-continuous-exceed-fully-sf:
    - shard-iclb:         [SKIP][233] ([i915#2920]) -> [SKIP][234] ([i915#658])
   [233]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-iclb2/igt@kms_psr2_sf@cursor-plane-move-continuous-exceed-fully-sf.html
   [234]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb5/igt@kms_psr2_sf@cursor-plane-move-continuous-exceed-fully-sf.html

  * igt@kms_psr2_su@page_flip-nv12:
    - shard-iclb:         [FAIL][235] ([i915#5939]) -> [SKIP][236] ([fdo#109642] / [fdo#111068] / [i915#658])
   [235]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-iclb2/igt@kms_psr2_su@page_flip-nv12.html
   [236]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-iclb8/igt@kms_psr2_su@page_flip-nv12.html

  * igt@runner@aborted:
    - shard-apl:          ([FAIL][237], [FAIL][238], [FAIL][239]) ([fdo#109271] / [i915#180] / [i915#3002] / [i915#4312] / [i915#5257] / [i915#6599]) -> ([FAIL][240], [FAIL][241], [FAIL][242]) ([i915#3002] / [i915#4312] / [i915#5257] / [i915#6599])
   [237]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-apl7/igt@runner@aborted.html
   [238]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-apl3/igt@runner@aborted.html
   [239]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-apl1/igt@runner@aborted.html
   [240]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-apl3/igt@runner@aborted.html
   [241]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-apl1/igt@runner@aborted.html
   [242]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-apl1/igt@runner@aborted.html
    - shard-kbl:          ([FAIL][243], [FAIL][244], [FAIL][245], [FAIL][246], [FAIL][247], [FAIL][248], [FAIL][249]) ([i915#180] / [i915#3002] / [i915#4312] / [i915#5257] / [i915#92]) -> ([FAIL][250], [FAIL][251], [FAIL][252], [FAIL][253], [FAIL][254], [FAIL][255], [FAIL][256], [FAIL][257], [FAIL][258]) ([fdo#109271] / [i915#180] / [i915#3002] / [i915#4312] / [i915#5257])
   [243]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-kbl4/igt@runner@aborted.html
   [244]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-kbl4/igt@runner@aborted.html
   [245]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-kbl4/igt@runner@aborted.html
   [246]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-kbl4/igt@runner@aborted.html
   [247]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-kbl4/igt@runner@aborted.html
   [248]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-kbl4/igt@runner@aborted.html
   [249]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_12020/shard-kbl4/igt@runner@aborted.html
   [250]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl7/igt@runner@aborted.html
   [251]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl4/igt@runner@aborted.html
   [252]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl4/igt@runner@aborted.html
   [253]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl1/igt@runner@aborted.html
   [254]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl4/igt@runner@aborted.html
   [255]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl4/igt@runner@aborted.html
   [256]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl4/igt@runner@aborted.html
   [257]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl4/igt@runner@aborted.html
   [258]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/shard-kbl4/igt@runner@aborted.html

  
  {name}: This element is suppressed. This means it is ignored when computing
          the status of the difference (SUCCESS, WARNING, or FAILURE).

  [fdo#108145]: https://bugs.freedesktop.org/show_bug.cgi?id=108145
  [fdo#109271]: https://bugs.freedesktop.org/show_bug.cgi?id=109271
  [fdo#109274]: https://bugs.freedesktop.org/show_bug.cgi?id=109274
  [fdo#109278]: https://bugs.freedesktop.org/show_bug.cgi?id=109278
  [fdo#109280]: https://bugs.freedesktop.org/show_bug.cgi?id=109280
  [fdo#109283]: https://bugs.freedesktop.org/show_bug.cgi?id=109283
  [fdo#109284]: https://bugs.freedesktop.org/show_bug.cgi?id=109284
  [fdo#109285]: https://bugs.freedesktop.org/show_bug.cgi?id=109285
  [fdo#109289]: https://bugs.freedesktop.org/show_bug.cgi?id=109289
  [fdo#109291]: https://bugs.freedesktop.org/show_bug.cgi?id=109291
  [fdo#109292]: https://bugs.freedesktop.org/show_bug.cgi?id=109292
  [fdo#109295]: https://bugs.freedesktop.org/show_bug.cgi?id=109295
  [fdo#109300]: https://bugs.freedesktop.org/show_bug.cgi?id=109300
  [fdo#109302]: https://bugs.freedesktop.org/show_bug.cgi?id=109302
  [fdo#109307]: https://bugs.freedesktop.org/show_bug.cgi?id=109307
  [fdo#109308]: https://bugs.freedesktop.org/show_bug.cgi?id=109308
  [fdo#109309]: https://bugs.freedesktop.org/show_bug.cgi?id=109309
  [fdo#109313]: https://bugs.freedesktop.org/show_bug.cgi?id=109313
  [fdo#109441]: https://bugs.freedesktop.org/show_bug.cgi?id=109441
  [fdo#109506]: https://bugs.freedesktop.org/show_bug.cgi?id=109506
  [fdo#109642]: https://bugs.freedesktop.org/show_bug.cgi?id=109642
  [fdo#110189]: https://bugs.freedesktop.org/show_bug.cgi?id=110189
  [fdo#110254]: https://bugs.freedesktop.org/show_bug.cgi?id=110254
  [fdo#110542]: https://bugs.freedesktop.org/show_bug.cgi?id=110542
  [fdo#110723]: https://bugs.freedesktop.org/show_bug.cgi?id=110723
  [fdo#110725]: https://bugs.freedesktop.org/show_bug.cgi?id=110725
  [fdo#111066]: https://bugs.freedesktop.org/show_bug.cgi?id=111066
  [fdo#111068]: https://bugs.freedesktop.org/show_bug.cgi?id=111068
  [fdo#111314]: https://bugs.freedesktop.org/show_bug.cgi?id=111314
  [fdo#111614]: https://bugs.freedesktop.org/show_bug.cgi?id=111614
  [fdo#111615]: https://bugs.freedesktop.org/show_bug.cgi?id=111615
  [fdo#111656]: https://bugs.freedesktop.org/show_bug.cgi?id=111656
  [fdo#111825]: https://bugs.freedesktop.org/show_bug.cgi?id=111825
  [fdo#111827]: https://bugs.freedesktop.org/show_bug.cgi?id=111827
  [fdo#112283]: https://bugs.freedesktop.org/show_bug.cgi?id=112283
  [i915#1063]: https://gitlab.freedesktop.org/drm/intel/issues/1063
  [i915#1072]: https://gitlab.freedesktop.org/drm/intel/issues/1072
  [i915#1099]: https://gitlab.freedesktop.org/drm/intel/issues/1099
  [i915#1155]: https://gitlab.freedesktop.org/drm/intel/issues/1155
  [i915#1188]: https://gitlab.freedesktop.org/drm/intel/issues/1188
  [i915#1257]: https://gitlab.freedesktop.org/drm/intel/issues/1257
  [i915#1319]: https://gitlab.freedesktop.org/drm/intel/issues/1319
  [i915#132]: https://gitlab.freedesktop.org/drm/intel/issues/132
  [i915#1397]: https://gitlab.freedesktop.org/drm/intel/issues/1397
  [i915#1769]: https://gitlab.freedesktop.org/drm/intel/issues/1769
  [i915#180]: https://gitlab.freedesktop.org/drm/intel/issues/180
  [i915#1825]: https://gitlab.freedesktop.org/drm/intel/issues/1825
  [i915#1839]: https://gitlab.freedesktop.org/drm/intel/issues/1839
  [i915#1845]: https://gitlab.freedesktop.org/drm/intel/issues/1845
  [i915#1849]: https://gitlab.freedesktop.org/drm/intel/issues/1849
  [i915#1888]: https://gitlab.freedesktop.org/drm/intel/issues/1888
  [i915#1911]: https://gitlab.freedesktop.org/drm/intel/issues/1911
  [i915#1982]: https://gitlab.freedesktop.org/drm/intel/issues/1982
  [i915#2122]: https://gitlab.freedesktop.org/drm/intel/issues/2122
  [i915#2190]: https://gitlab.freedesktop.org/drm/intel/issues/2190
  [i915#2346]: https://gitlab.freedesktop.org/drm/intel/issues/2346
  [i915#2410]: https://gitlab.freedesktop.org/drm/intel/issues/2410
  [i915#2434]: https://gitlab.freedesktop.org/drm/intel/issues/2434
  [i915#2437]: https://gitlab.freedesktop.org/drm/intel/issues/2437
  [i915#2521]: https://gitlab.freedesktop.org/drm/intel/issues/2521
  [i915#2527]: https://gitlab.freedesktop.org/drm/intel/issues/2527
  [i915#2530]: https://gitlab.freedesktop.org/drm/intel/issues/2530
  [i915#2546]: https://gitlab.freedesktop.org/drm/intel/issues/2546
  [i915#265]: https://gitlab.freedesktop.org/drm/intel/issues/265
  [i915#2658]: https://gitlab.freedesktop.org/drm/intel/issues/2658
  [i915#2672]: https://gitlab.freedesktop.org/drm/intel/issues/2672
  [i915#2681]: https://gitlab.freedesktop.org/drm/intel/issues/2681
  [i915#2684]: https://gitlab.freedesktop.org/drm/intel/issues/2684
  [i915#2705]: https://gitlab.freedesktop.org/drm/intel/issues/2705
  [i915#280]: https://gitlab.freedesktop.org/drm/intel/issues/280
  [i915#284]: https://gitlab.freedesktop.org/drm/intel/issues/284
  [i915#2842]: https://gitlab.freedesktop.org/drm/intel/issues/2842
  [i915#2846]: https://gitlab.freedesktop.org/drm/intel/issues/2846
  [i915#2856]: https://gitlab.freedesktop.org/drm/intel/issues/2856
  [i915#2920]: https://gitlab.freedesktop.org/drm/intel/issues/2920
  [i915#2994]: https://gitlab.freedesktop.org/drm/intel/issues/2994
  [i915#3002]: https://gitlab.freedesktop.org/drm/intel/issues/3002
  [i915#3063]: https://gitlab.freedesktop.org/drm/intel/issues/3063
  [i915#3116]: https://gitlab.freedesktop.org/drm/intel/issues/3116
  [i915#3281]: https://gitlab.freedesktop.org/drm/intel/issues/3281
  [i915#3282]: https://gitlab.freedesktop.org/drm/intel/issues/3282
  [i915#3291]: https://gitlab.freedesktop.org/drm/intel/issues/3291
  [i915#3297]: https://gitlab.freedesktop.org/drm/intel/issues/3297
  [i915#3299]: https://gitlab.freedesktop.org/drm/intel/issues/3299
  [i915#3318]: https://gitlab.freedesktop.org/drm/intel/issues/3318
  [i915#3323]: https://gitlab.freedesktop.org/drm/intel/issues/3323
  [i915#3361]: https://gitlab.freedesktop.org/drm/intel/issues/3361
  [i915#3376]: https://gitlab.freedesktop.org/drm/intel/issues/3376
  [i915#3528]: https://gitlab.freedesktop.org/drm/intel/issues/3528
  [i915#3546]: https://gitlab.freedesktop.org/drm/intel/issues/3546
  [i915#3555]: https://gitlab.freedesktop.org/drm/intel/issues/3555
  [i915#3558]: https://gitlab.freedesktop.org/drm/intel/issues/3558
  [i915#3637]: https://gitlab.freedesktop.org/drm/intel/issues/3637
  [i915#3638]: https://gitlab.freedesktop.org/drm/intel/issues/3638
  [i915#3689]: https://gitlab.freedesktop.org/drm/intel/issues/3689
  [i915#3708]: https://gitlab.freedesktop.org/drm/intel/issues/3708
  [i915#3734]: https://gitlab.freedesktop.org/drm/intel/issues/3734
  [i915#3742]: https://gitlab.freedesktop.org/drm/intel/issues/3742
  [i915#3826]: https://gitlab.freedesktop.org/drm/intel/issues/3826
  [i915#3828]: https://gitlab.freedesktop.org/drm/intel/issues/3828
  [i915#3886]: https://gitlab.freedesktop.org/drm/intel/issues/3886
  [i915#3966]: https://gitlab.freedesktop.org/drm/intel/issues/3966
  [i915#4070]: https://gitlab.freedesktop.org/drm/intel/issues/4070
  [i915#4098]: https://gitlab.freedesktop.org/drm/intel/issues/4098
  [i915#4103]: https://gitlab.freedesktop.org/drm/intel/issues/4103
  [i915#4171]: https://gitlab.freedesktop.org/drm/intel/issues/4171
  [i915#4270]: https://gitlab.freedesktop.org/drm/intel/issues/4270
  [i915#4281]: https://gitlab.freedesktop.org/drm/intel/issues/4281
  [i915#4312]: https://gitlab.freedesktop.org/drm/intel/issues/4312
  [i915#4369]: https://gitlab.freedesktop.org/drm/intel/issues/4369
  [i915#4525]: https://gitlab.freedesktop.org/drm/intel/issues/4525
  [i915#454]: https://gitlab.freedesktop.org/drm/intel/issues/454
  [i915#4613]: https://gitlab.freedesktop.org/drm/intel/issues/4613
  [i915#4765]: https://gitlab.freedesktop.org/drm/intel/issues/4765
  [i915#4767]: https://gitlab.freedesktop.org/drm/intel/issues/4767
  [i915#4893]: https://gitlab.freedesktop.org/drm/intel/issues/4893
  [i915#4939]: https://gitlab.freedesktop.org/drm/intel/issues/4939
  [i915#4941]: https://gitlab.freedesktop.org/drm/intel/issues/4941
  [i915#4991]: https://gitlab.freedesktop.org/drm/intel/issues/4991
  [i915#5030]: https://gitlab.freedesktop.org/drm/intel/issues/5030
  [i915#5138]: https://gitlab.freedesktop.org/drm/intel/issues/5138
  [i915#5176]: https://gitlab.freedesktop.org/drm/intel/issues/5176
  [i915#5182]: https://gitlab.freedesktop.org/drm/intel/issues/5182
  [i915#5235]: https://gitlab.freedesktop.org/drm/intel/issues/5235
  [i915#5257]: https://gitlab.freedesktop.org/drm/intel/issues/5257
  [i915#5286]: https://gitlab.freedesktop.org/drm/intel/issues/5286
  [i915#5287]: https://gitlab.freedesktop.org/drm/intel/issues/5287
  [i915#5288]: https://gitlab.freedesktop.org/drm/intel/issues/5288
  [i915#5289]: https://gitlab.freedesktop.org/drm/intel/issues/5289
  [i915#5325]: https://gitlab.freedesktop.org/drm/intel/issues/5325
  [i915#533]: https://gitlab.freedesktop.org/drm/intel/issues/533
  [i915#5438]: https://gitlab.freedesktop.org/drm/intel/issues/5438
  [i915#5519]: https://gitlab.freedesktop.org/drm/intel/issues/5519
  [i915#5723]: https://gitlab.freedesktop.org/drm/intel/issues/5723
  [i915#5784]: https://gitlab.freedesktop.org/drm/intel/issues/5784
  [i915#588]: https://gitlab.freedesktop.org/drm/intel/issues/588
  [i915#5939]: https://gitlab.freedesktop.org/drm/intel/issues/5939
  [i915#5982]: https://gitlab.freedesktop.org/drm/intel/issues/5982
  [i915#6095]: https://gitlab.freedesktop.org/drm/intel/issues/6095
  [i915#6230]: https://gitlab.freedesktop.org/drm/intel/issues/6230
  [i915#6245]: https://gitlab.freedesktop.org/drm/intel/issues/6245
  [i915#6248]: https://gitlab.freedesktop.org/drm/intel/issues/6248
  [i915#6268]: https://gitlab.freedesktop.org/drm/intel/issues/6268
  [i915#6334]: https://gitlab.freedesktop.org/drm/intel/issues/6334
  [i915#6335]: https://gitlab.freedesktop.org/drm/intel/issues/6335
  [i915#6344]: https://gitlab.freedesktop.org/drm/intel/issues/6344
  [i915#6412]: https://gitlab.freedesktop.org/drm/intel/issues/6412
  [i915#6433]: https://gitlab.freedesktop.org/drm/intel/issues/6433
  [i915#6497]: https://gitlab.freedesktop.org/drm/intel/issues/6497
  [i915#6520]: https://gitlab.freedesktop.org/drm/intel/issues/6520
  [i915#6524]: https://gitlab.freedesktop.org/drm/intel/issues/6524
  [i915#658]: https://gitlab.freedesktop.org/drm/intel/issues/658
  [i915#6590]: https://gitlab.freedesktop.org/drm/intel/issues/6590
  [i915#6598]: https://gitlab.freedesktop.org/drm/intel/issues/6598
  [i915#6599]: https://gitlab.freedesktop.org/drm/intel/issues/6599
  [i915#6604]: https://gitlab.freedesktop.org/drm/intel/issues/6604
  [i915#6625]: https://gitlab.freedesktop.org/drm/intel/issues/6625
  [i915#6637]: https://gitlab.freedesktop.org/drm/intel/issues/6637
  [i915#768]: https://gitlab.freedesktop.org/drm/intel/issues/768
  [i915#92]: https://gitlab.freedesktop.org/drm/intel/issues/92


Build changes
-------------

  * CI: CI-20190529 -> None
  * IGT: IGT_6636 -> IGTPW_7680
  * Piglit: piglit_4509 -> None

  CI-20190529: 20190529
  CI_DRM_12020: 9e188e66f942a9b7db1c41d56ac9ff1433473d8e @ git://anongit.freedesktop.org/gfx-ci/linux
  IGTPW_7680: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/index.html
  IGT_6636: 1298b5f0e1b3e010657ffba41d2e775fab028e08 @ https://gitlab.freedesktop.org/drm/igt-gpu-tools.git
  piglit_4509: fdc5a4ca11124ab8413c7988896eec4c97336694 @ git://anongit.freedesktop.org/piglit

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_7680/index.html

[-- Attachment #2: Type: text/html, Size: 77045 bytes --]

[igt-dev] [PATCH i-g-t 00/23] Add DG2 OA test

[Umesh Nerlige Ramappa]

Add tests for changes in perf OA for DG2.

Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>

Lionel Landwerlin (4):
  lib/i915/perf: implement report accumulation for new format
  lib/i915/perf: fixup conversion script for XEHPSDV
  lib/i915/perf: make warning message more helpful
  lib/i915/perf: Add DG2 metrics

Nerlige Ramappa, Umesh (1):
  i915/perf: Limit sseu-config tests for gen11

Umesh Nerlige Ramappa (18):
  i915/perf: Check regularly if we are done reading reports
  i915/perf: Fix OA short_reads test
  i915/perf: Check return value from getparam
  i915/perf: Bump timestamp tolerance for DG1
  i915/perf: Account for OA sampling interval in polling test
  i915/perf: Add support for 64-bit counters
  i915/perf: Define OA report types
  i915/perf: Use ARRAY_SIZE consistently for num_properties
  i915/perf: Use gt in perf tests and lib
  i915/perf: Explicitly state rendercopy needs for a test
  i915/perf: Skip tests that use rendercopy
  i915/perf: Add OA formats for DG2
  i915/perf: Add a test for non-power-of-2 oa reports
  i915/perf: Fix CS timestamp vs OA timstamp mismatch
  i915/perf: Treat ticks as 64 bit
  i915/perf: Treat timestamp as 64 bit value
  i915/perf: Fix DG2 A0 report header
  i915/perf: Wait longer for rc6 residency in DG2

 include/drm-uapi/i915_drm.h                   |   12 +
 lib/i915/perf-configs/guids.xml               |   24 +
 lib/i915/perf-configs/mdapi-xml-convert.py    |   70 +-
 lib/i915/perf-configs/oa-dg2.xml              | 9920 +++++++++++++++++
 .../perf-configs/perf-metricset-codegen.py    |   13 +
 lib/i915/perf-configs/update-guids.py         |    2 +-
 lib/i915/perf.c                               |  104 +-
 lib/meson.build                               |    2 +-
 tests/i915/perf.c                             |  623 +-
 9 files changed, 10589 insertions(+), 181 deletions(-)
 create mode 100644 lib/i915/perf-configs/oa-dg2.xml

-- 
2.25.1

[igt-dev] [PATCH i-g-t 23/23] lib/i915/perf: Add DG2 metrics

[Umesh Nerlige Ramappa]

From: Lionel Landwerlin <lionel.g.landwerlin@intel.com>

Add DG2 metrics XML files for perf OA tests.

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
---
 lib/i915/perf-configs/guids.xml               |   24 +
 lib/i915/perf-configs/mdapi-xml-convert.py    |    5 +-
 lib/i915/perf-configs/oa-dg2.xml              | 9920 +++++++++++++++++
 .../perf-configs/perf-metricset-codegen.py    |   13 +
 lib/i915/perf-configs/update-guids.py         |    2 +-
 lib/i915/perf.c                               |    3 +
 lib/meson.build                               |    2 +-
 7 files changed, 9965 insertions(+), 4 deletions(-)
 create mode 100644 lib/i915/perf-configs/oa-dg2.xml

diff --git a/lib/i915/perf-configs/guids.xml b/lib/i915/perf-configs/guids.xml
index 95c9d717..61a11542 100644
--- a/lib/i915/perf-configs/guids.xml
+++ b/lib/i915/perf-configs/guids.xml
@@ -431,4 +431,28 @@
     <guid config_hash="b420d7b956fb1b097a7bb812b6c87dc0" mdapi_config_hash="d81f15a77d18a099782c18acaa3132f7" id="c3ea2a23-f1c7-4a19-9da2-b569226fb6de" chipset="adl" name="EuActivity7" />
     <guid config_hash="5face026dac6a0549f8a28cf226dfbc8" mdapi_config_hash="de7326838bcc8abf024bac9c771e22e4" id="a5e2f79b-cecb-4eff-8f29-cda8e2a58749" chipset="adl" name="EuActivity8" />
     <guid config_hash="1affe7abeb3739d2c845cc6fa9aacf88" mdapi_config_hash="d071f4933574945979595246be61d3f0" id="89173c19-fcfe-48da-ac26-fb64425f141b" chipset="adl" name="TestOa" />
+    <guid config_hash="9091b7d8f1627a3dda5d6d996eadd033" mdapi_config_hash="4a3cf9efdf4fd581b69308368b986500" id="8b20a2e7-04db-4196-a5ca-26b53cbf9a12" chipset="dg2" name="RenderBasic" />
+    <guid config_hash="a0afa17cc8f442e6eabb3b1782cbca44" mdapi_config_hash="2a23a2f9f1bcf91a9b038b2cc6c9c911" id="f66d7204-54f5-4663-9d68-5213f70165ba" chipset="dg2" name="ComputeBasic" />
+    <guid config_hash="34875baa1a12dda6dcf6d40e33a8d5ef" mdapi_config_hash="31dfc8c6ef9a79789bbe0b991c94112f" id="3b06d2f4-f594-49b5-b399-98550fa2d7b1" chipset="dg2" name="RenderPipeProfile" />
+    <guid config_hash="5f0f4b7c1e4cae4d04ab60a66f1d98a8" mdapi_config_hash="58ec2ac3713bd8e21112f8fddbe28c48" id="1afb0edc-585c-4be9-8c0c-bc72325106b0" chipset="dg2" name="HDCAndSF1" />
+    <guid config_hash="69dbeb5aba00131465b68f20c910421b" mdapi_config_hash="cf8e3757a118ffffb76db95917cae040" id="a3b35902-f293-41fe-97d0-3d895bc5fab3" chipset="dg2" name="HDCAndSF2" />
+    <guid config_hash="e563f102f8b3046e8deaea0eb81b8784" mdapi_config_hash="81d17bd3ab0a18f4491c2d0a11037c7a" id="b04e9460-630f-4682-b0ed-ebb0e74a6c9d" chipset="dg2" name="RasterizerAndPixelBackend1" />
+    <guid config_hash="b574d106f1f71eae1e8a0908cad2c010" mdapi_config_hash="d9c0a07f470d2a9bf7bbf20f79488abf" id="57a0abf4-8d81-4856-ab86-01f15f92c81c" chipset="dg2" name="RasterizerAndPixelBackend2" />
+    <guid config_hash="3316babac928063b799008fc324812b1" mdapi_config_hash="c9cd0a76400e340b0ffdb129fff205c7" id="2314eaa3-164f-4004-9684-3f1d16e875f9" chipset="dg2" name="Sampler_1" />
+    <guid config_hash="31c5ffa0c3508fb166ace213356edbaf" mdapi_config_hash="e2bce131c17fa27ab826a05ed65cad5a" id="c60dea82-db13-4882-8d31-a02bce151ef7" chipset="dg2" name="Sampler_2" />
+    <guid config_hash="b3460c2d0d4e7bcbcc46a3f1c9df7e8f" mdapi_config_hash="ee6821f6954a10562b6314c5ef394b50" id="7f5cb8ec-38bb-48c9-9ad2-60075eba2833" chipset="dg2" name="TDL_1" />
+    <guid config_hash="9ff61984b39a3d19a2abf3a0fd4c2285" mdapi_config_hash="013fa392a5c8b3bb93a776146d31bc95" id="141ce507-de6c-428c-95b2-076194d650db" chipset="dg2" name="TDL_2" />
+    <guid config_hash="11b056437f6c491c2a44e4ca029dd57b" mdapi_config_hash="c4a4c4cec33f45576af76022e6f634ae" id="54a85b07-9ff2-4a12-bfb6-6bcdd9d4019c" chipset="dg2" name="TDL_3" />
+    <guid config_hash="5cf268036a3ae548ddd918fdbd67a64b" mdapi_config_hash="4744760a7ff39c509ecab43f3f715790" id="651c40e5-195c-49da-b232-a9a0f753d899" chipset="dg2" name="TDL_4" />
+    <guid config_hash="f43033715b8153ccb51147452de5bf99" mdapi_config_hash="e77513bc5aa341b061701d6d973b1b4c" id="9a1a14ad-ab92-4204-8a91-c4159a54f712" chipset="dg2" name="L3_1" />
+    <guid config_hash="1c4da3e296779bdbef2262460f3f0fcb" mdapi_config_hash="27c652b5bcd6e800c23a3cacc103c2cb" id="85de4c29-76b8-42a6-b6ef-8a2bed008fcf" chipset="dg2" name="GpuBusyness" />
+    <guid config_hash="a1c8aac6e58018ff3bf78169e5d7cf00" mdapi_config_hash="3d1142aa73a474eccef2bd23027e71b9" id="8dc7142a-df10-41f2-acaa-9faea547ce73" chipset="dg2" name="EuActivity1" />
+    <guid config_hash="880a408e8bc0c0a82df014cbbe0a1a9e" mdapi_config_hash="d998d2f096753140dc92ed9421414ae1" id="d3c05cce-4ae7-4291-82ab-b0bcade6cf7a" chipset="dg2" name="EuActivity2" />
+    <guid config_hash="6687d4c4e45e62d0d28792b04882edae" mdapi_config_hash="c5b1271909b4d72ad1503d5d8a50d30a" id="4fd009c5-6eae-4ecb-a2d1-41edf211cd06" chipset="dg2" name="EuActivity3" />
+    <guid config_hash="f770062c29026a186485a2190b7ce3fb" mdapi_config_hash="7a83985438139dcfdf7d653be1e665b0" id="5226090d-b636-4845-9f76-545cb5db7867" chipset="dg2" name="EuActivity4" />
+    <guid config_hash="4e13b511b35757cc97602aae22b7d2d3" mdapi_config_hash="5405e0eff1e1b0af86f45e832c604fcd" id="c8d33b8f-3152-4c81-99bf-d3cb45d7e1ac" chipset="dg2" name="EuActivity5" />
+    <guid config_hash="243a5047b39033bca9a16db6bf2aca3f" mdapi_config_hash="2bc02385284ccdbbd0d233a428861e59" id="708d01ed-bc9b-420e-8048-9060abcbb1f0" chipset="dg2" name="EuActivity6" />
+    <guid config_hash="d0f2a1c661860e4aa94e25e9f4a6f8eb" mdapi_config_hash="75694dd6a99c38e9446d7bf68f2269a3" id="72262d5b-a9e6-4742-820b-2a7bef9a7e61" chipset="dg2" name="EuActivity7" />
+    <guid config_hash="5e7354b4cbe7b9b03acb504ef00aa8ce" mdapi_config_hash="299c68337c46a2483c1254fdead4cdcc" id="11c685de-1445-41b8-acb5-8e79c75c4fd8" chipset="dg2" name="EuActivity8" />
+    <guid config_hash="24977eb13d34b0da58c0b33f6d7f841a" mdapi_config_hash="48715fb10d2ada8c23f7108303b7c343" id="2db1e1d4-8c48-4bb7-b248-0f40b70433ff" chipset="dg2" name="TestOa" />
 </guids>
diff --git a/lib/i915/perf-configs/mdapi-xml-convert.py b/lib/i915/perf-configs/mdapi-xml-convert.py
index 1d161674..72b52bfd 100755
--- a/lib/i915/perf-configs/mdapi-xml-convert.py
+++ b/lib/i915/perf-configs/mdapi-xml-convert.py
@@ -141,6 +141,7 @@ chipsets = {
     'RKL': gen8_11_chipset_params,
     'DG1': gen8_11_chipset_params,
     'ADL': gen8_11_chipset_params,
+    'DG2': xehpsdv_chipset_params,
 }
 
 register_types = { 'OA', 'NOA', 'FLEX', 'PM' }
@@ -214,7 +215,7 @@ def read_token_to_rpn_read(chipset, token, raw_offsets):
                 assert 0
         elif offset < b_offset:
             a_cnt_offset = int((offset - a_offset) / 4)
-            if chipset == "XEHPSDV":
+            if chipset == "XEHPSDV" or chipset == "DG2":
                 # Most A counters are in a contiguous array, except
                 # this A37.
                 if a_cnt_offset == 42:
@@ -245,7 +246,7 @@ def read_token_to_rpn_read(chipset, token, raw_offsets):
                 return "C " + str(idx - 54) + " READ"
             else:
                 return "{0} READ".format(read_value(chipset, offset))
-        elif chipset == "XEHPSDV":
+        elif chipset == "XEHPSDV" or chipset == "DG2":
             # For XEHPSDV the array of accumulated counters is
             # assumed to start with a GPU_TIME then GPU_CLOCK,
             # then 38 A counters, then 8 B counters and finally
diff --git a/lib/i915/perf-configs/oa-dg2.xml b/lib/i915/perf-configs/oa-dg2.xml
new file mode 100644
index 00000000..9d3e2541
--- /dev/null
+++ b/lib/i915/perf-configs/oa-dg2.xml
@@ -0,0 +1,9920 @@
+<?xml version="1.0"?>
+<metrics version="1602592998" merge_md5="">
+  <set name="Render Basic Metrics"
+       chipset="DG2"
+       symbol_name="RenderBasic"
+       underscore_name="render_basic"
+       mdapi_supported_apis="VK OGL OCL IO"
+       hw_config_guid="8b20a2e7-04db-4196-a5ca-26b53cbf9a12"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Rasterized Pixels"
+             symbol_name="RasterizedPixels"
+             underscore_name="rasterized_pixels"
+             description="The total number of rasterized pixels."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 21 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Early Hi-Depth Test Fails"
+             symbol_name="HiDepthTestFails"
+             underscore_name="hi_depth_test_fails"
+             description="The total number of pixels dropped on early hierarchical depth test."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 22 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer/Hi-Depth Test"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Early Depth Test Fails"
+             symbol_name="EarlyDepthTestFails"
+             underscore_name="early_depth_test_fails"
+             description="The total number of pixels dropped on early depth test."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 23 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer/Early Depth Test"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Killed in FS"
+             symbol_name="SamplesKilledInPs"
+             underscore_name="samples_killed_in_ps"
+             description="The total number of samples or pixels dropped in fragment shaders."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 24 READ 4 UMUL"
+             mdapi_group="3D Pipe/Fragment Shader"
+             mdapi_usage_flags="Tier4 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Pixels Failing Tests"
+             symbol_name="PixelsFailingPostPsTests"
+             underscore_name="pixels_failing_post_ps_tests"
+             description="The total number of pixels dropped on post-FS alpha, stencil, or depth tests."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 25 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Written"
+             symbol_name="SamplesWritten"
+             underscore_name="samples_written"
+             description="The total number of samples or pixels written to all render targets."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 26 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Blended"
+             symbol_name="SamplesBlended"
+             underscore_name="samples_blended"
+             description="The total number of blended samples or pixels written to all render targets."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 27 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Texels"
+             symbol_name="SamplerTexels"
+             underscore_name="sampler_texels"
+             description="The total number of texels seen on input (with 2x2 accuracy) in all sampler units."
+             data_type="uint64"
+             units="texels"
+             semantic_type="event"
+             equation="A 28 READ 4 UMUL"
+             mdapi_group="Sampler/Sampler Input"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Texels Misses"
+             symbol_name="SamplerTexelMisses"
+             underscore_name="sampler_texel_misses"
+             description="The total number of texels lookups (with 2x2 accuracy) that missed L1 sampler cache."
+             data_type="uint64"
+             units="texels"
+             semantic_type="event"
+             equation="A 29 READ 4 UMUL"
+             mdapi_group="Sampler/Sampler Cache"
+             mdapi_usage_flags="Tier3 Batch Frame Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SLM Bytes Read"
+             symbol_name="SlmBytesRead"
+             underscore_name="slm_bytes_read"
+             description="The total number of GPU memory bytes read from shared local memory."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="A 30 READ 64 UMUL"
+             mdapi_group="L3/Data Port/SLM"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SLM Bytes Written"
+             symbol_name="SlmBytesWritten"
+             underscore_name="slm_bytes_written"
+             description="The total number of GPU memory bytes written into shared local memory."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="A 31 READ 64 UMUL"
+             mdapi_group="L3/Data Port/SLM"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Memory Accesses"
+             symbol_name="ShaderMemoryAccesses"
+             underscore_name="shader_memory_accesses"
+             description="The total number of shader memory accesses to L3."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 32 READ"
+             mdapi_group="L3/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Atomic Memory Accesses"
+             symbol_name="ShaderAtomics"
+             underscore_name="shader_atomics"
+             description="The total number of shader atomic memory accesses."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 34 READ"
+             mdapi_group="L3/Data Port/Atomics"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="L3 Shader Throughput"
+             symbol_name="L3ShaderThroughput"
+             underscore_name="l3_shader_throughput"
+             description="The total number of GPU memory bytes transferred between shaders and L3 caches w/o URB."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="$ShaderMemoryAccesses 64 UMUL"
+             mdapi_group="L3/Data Port"
+             mdapi_usage_flags="Tier2 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Barrier Messages"
+             symbol_name="ShaderBarriers"
+             underscore_name="shader_barriers"
+             description="The total number of shader barrier messages."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 35 READ"
+             mdapi_group="EU Array/Barrier"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GTI Read Throughput"
+             symbol_name="GtiReadThroughput"
+             underscore_name="gti_read_throughput"
+             description="The total number of GPU memory bytes read from GTI."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 512 UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="128  A 36 READ UMUL"
+             mdapi_group="GTI"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GTI Write Throughput"
+             symbol_name="GtiWriteThroughput"
+             underscore_name="gti_write_throughput"
+             description="The total number of GPU memory bytes written to GTI."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 512 UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="C 5 READ C 4 READ UADD C 3 READ UADD C 2 READ UADD C 1 READ UADD C 0 READ UADD B 7 READ UADD B 6 READ UADD 128 UMUL"
+             mdapi_group="GTI"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Slice0 Dualsubslice0 is busy"
+             symbol_name="Sampler00Busy"
+             underscore_name="sampler00_busy"
+             description="The percentage of time when sampler slice0 dualsubslice0 is busy"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 1 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Samplers Busy"
+             symbol_name="SamplersBusy"
+             underscore_name="samplers_busy"
+             description="The percentage of time in which samplers or L1 cache are not idle."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="$Sampler00Busy"
+             availability="$DualSubsliceMask 1 AND"
+             mdapi_group="Sampler"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Sampler00 Bottleneck"
+             symbol_name="Sampler00Bottleneck"
+             underscore_name="sampler00_bottleneck"
+             description="The percentage of time in which Slice0 Sampler0 has been slowing down the pipe when processing EU requests."
+             low_watermark="5"
+             high_watermark="15"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 1 AND"
+             mdapi_group="Sampler"
+             mdapi_usage_flags="Tier3 Indicate Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Samplers Bottleneck"
+             symbol_name="SamplerBottleneck"
+             underscore_name="sampler_bottleneck"
+             description="The percentage of time in which samplers have been slowing down the pipe when processing EU requests."
+             low_watermark="5"
+             high_watermark="15"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="$Sampler00Bottleneck"
+             availability="$DualSubsliceMask 1 AND"
+             mdapi_group="Sampler"
+             mdapi_usage_flags="Tier3 Overview Indicate System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x0000000C" />
+        <register type="NOA" address="0x00009888" value="0x1C840000" />
+        <register type="NOA" address="0x00009888" value="0x1E842800" />
+        <register type="NOA" address="0x00009888" value="0x02844000" />
+        <register type="NOA" address="0x00009888" value="0x04840010" />
+        <register type="NOA" address="0x00009888" value="0x12840800" />
+        <register type="NOA" address="0x00009888" value="0x14840037" />
+        <register type="NOA" address="0x00009888" value="0x00840000" />
+        <register type="NOA" address="0x00009888" value="0x02865000" />
+        <register type="NOA" address="0x00009888" value="0x0208A000" />
+        <register type="NOA" address="0x00009888" value="0x020A5000" />
+        <register type="NOA" address="0x00009884" value="0x00000001" />
+        <register type="NOA" address="0x00009888" value="0x18013800" />
+        <register type="NOA" address="0x00009888" value="0x18213800" />
+        <register type="NOA" address="0x00009888" value="0x18413800" />
+        <register type="NOA" address="0x00009888" value="0x18613800" />
+        <register type="NOA" address="0x00009888" value="0x18813800" />
+        <register type="NOA" address="0x00009888" value="0x18A13800" />
+        <register type="NOA" address="0x00009888" value="0x18C13800" />
+        <register type="NOA" address="0x00009888" value="0x18E13800" />
+        <register type="NOA" address="0x00009888" value="0x02010006" />
+        <register type="NOA" address="0x00009888" value="0x14010004" />
+        <register type="NOA" address="0x00009888" value="0x00010000" />
+        <register type="NOA" address="0x00009888" value="0x02210006" />
+        <register type="NOA" address="0x00009888" value="0x12210100" />
+        <register type="NOA" address="0x00009888" value="0x00210000" />
+        <register type="NOA" address="0x00009888" value="0x02410006" />
+        <register type="NOA" address="0x00009888" value="0x12410004" />
+        <register type="NOA" address="0x00009888" value="0x00410000" />
+        <register type="NOA" address="0x00009888" value="0x02610006" />
+        <register type="NOA" address="0x00009888" value="0x10610100" />
+        <register type="NOA" address="0x00009888" value="0x00610000" />
+        <register type="NOA" address="0x00009888" value="0x02810006" />
+        <register type="NOA" address="0x00009888" value="0x10810004" />
+        <register type="NOA" address="0x00009888" value="0x00810000" />
+        <register type="NOA" address="0x00009888" value="0x02A10006" />
+        <register type="NOA" address="0x00009888" value="0x0EA10100" />
+        <register type="NOA" address="0x00009888" value="0x00A10000" />
+        <register type="NOA" address="0x00009888" value="0x02C10006" />
+        <register type="NOA" address="0x00009888" value="0x0EC10004" />
+        <register type="NOA" address="0x00009888" value="0x00C10000" />
+        <register type="NOA" address="0x00009888" value="0x02E10006" />
+        <register type="NOA" address="0x00009888" value="0x0CE10100" />
+        <register type="NOA" address="0x00009888" value="0x00E10000" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x0208A000" />
+        <register type="NOA" address="0x00009888" value="0x02034920" />
+        <register type="NOA" address="0x00009888" value="0x04032924" />
+        <register type="NOA" address="0x00009888" value="0x06030002" />
+        <register type="NOA" address="0x00009888" value="0x020914E0" />
+        <register type="NOA" address="0x00009888" value="0x020A1248" />
+        <register type="NOA" address="0x00009888" value="0x040A0492" />
+        <register type="NOA" address="0x00009888" value="0x040B029C" />
+        <register type="NOA" address="0x00009888" value="0x46030860" />
+        <register type="NOA" address="0x00009888" value="0x48030643" />
+        <register type="NOA" address="0x00009888" value="0x4A030664" />
+        <register type="NOA" address="0x00009888" value="0x16033000" />
+        <register type="NOA" address="0x00009888" value="0x3E030840" />
+        <register type="NOA" address="0x00009888" value="0x18033000" />
+        <register type="NOA" address="0x00009888" value="0x1A033000" />
+        <register type="NOA" address="0x00009888" value="0x40030042" />
+        <register type="NOA" address="0x00009888" value="0x1C033000" />
+        <register type="NOA" address="0x00009888" value="0x1E030000" />
+        <register type="NOA" address="0x00009888" value="0x20030000" />
+        <register type="NOA" address="0x00009888" value="0x42030000" />
+        <register type="NOA" address="0x00009884" value="0x00000008" />
+        <register type="NOA" address="0x00009888" value="0x0001C000" />
+        <register type="NOA" address="0x00009888" value="0x02025000" />
+        <register type="NOA" address="0x00009888" value="0x02045000" />
+        <register type="NOA" address="0x00009888" value="0x02055000" />
+        <register type="NOA" address="0x00009888" value="0x0208A000" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x00800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00000000" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="Compute Metrics Basic"
+       chipset="DG2"
+       symbol_name="ComputeBasic"
+       underscore_name="compute_basic"
+       mdapi_supported_apis="VK OGL OCL IO"
+       hw_config_guid="f66d7204-54f5-4663-9d68-5213f70165ba"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Rasterized Pixels"
+             symbol_name="RasterizedPixels"
+             underscore_name="rasterized_pixels"
+             description="The total number of rasterized pixels."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 21 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Early Hi-Depth Test Fails"
+             symbol_name="HiDepthTestFails"
+             underscore_name="hi_depth_test_fails"
+             description="The total number of pixels dropped on early hierarchical depth test."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 22 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer/Hi-Depth Test"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Early Depth Test Fails"
+             symbol_name="EarlyDepthTestFails"
+             underscore_name="early_depth_test_fails"
+             description="The total number of pixels dropped on early depth test."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 23 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer/Early Depth Test"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Killed in FS"
+             symbol_name="SamplesKilledInPs"
+             underscore_name="samples_killed_in_ps"
+             description="The total number of samples or pixels dropped in fragment shaders."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 24 READ 4 UMUL"
+             mdapi_group="3D Pipe/Fragment Shader"
+             mdapi_usage_flags="Tier4 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Pixels Failing Tests"
+             symbol_name="PixelsFailingPostPsTests"
+             underscore_name="pixels_failing_post_ps_tests"
+             description="The total number of pixels dropped on post-FS alpha, stencil, or depth tests."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 25 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Written"
+             symbol_name="SamplesWritten"
+             underscore_name="samples_written"
+             description="The total number of samples or pixels written to all render targets."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 26 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Blended"
+             symbol_name="SamplesBlended"
+             underscore_name="samples_blended"
+             description="The total number of blended samples or pixels written to all render targets."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 27 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Texels"
+             symbol_name="SamplerTexels"
+             underscore_name="sampler_texels"
+             description="The total number of texels seen on input (with 2x2 accuracy) in all sampler units."
+             data_type="uint64"
+             units="texels"
+             semantic_type="event"
+             equation="A 28 READ 4 UMUL"
+             mdapi_group="Sampler/Sampler Input"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Texels Misses"
+             symbol_name="SamplerTexelMisses"
+             underscore_name="sampler_texel_misses"
+             description="The total number of texels lookups (with 2x2 accuracy) that missed L1 sampler cache."
+             data_type="uint64"
+             units="texels"
+             semantic_type="event"
+             equation="A 29 READ 4 UMUL"
+             mdapi_group="Sampler/Sampler Cache"
+             mdapi_usage_flags="Tier3 Batch Frame Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SLM Bytes Read"
+             symbol_name="SlmBytesRead"
+             underscore_name="slm_bytes_read"
+             description="The total number of GPU memory bytes read from shared local memory."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="A 30 READ 64 UMUL"
+             mdapi_group="L3/Data Port/SLM"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SLM Bytes Written"
+             symbol_name="SlmBytesWritten"
+             underscore_name="slm_bytes_written"
+             description="The total number of GPU memory bytes written into shared local memory."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="A 31 READ 64 UMUL"
+             mdapi_group="L3/Data Port/SLM"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Memory Accesses"
+             symbol_name="ShaderMemoryAccesses"
+             underscore_name="shader_memory_accesses"
+             description="The total number of shader memory accesses to L3."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 32 READ"
+             mdapi_group="L3/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Atomic Memory Accesses"
+             symbol_name="ShaderAtomics"
+             underscore_name="shader_atomics"
+             description="The total number of shader atomic memory accesses."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 34 READ"
+             mdapi_group="L3/Data Port/Atomics"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="L3 Shader Throughput"
+             symbol_name="L3ShaderThroughput"
+             underscore_name="l3_shader_throughput"
+             description="The total number of GPU memory bytes transferred between shaders and L3 caches w/o URB."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="$ShaderMemoryAccesses 64 UMUL"
+             mdapi_group="L3/Data Port"
+             mdapi_usage_flags="Tier2 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Barrier Messages"
+             symbol_name="ShaderBarriers"
+             underscore_name="shader_barriers"
+             description="The total number of shader barrier messages."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 35 READ"
+             mdapi_group="EU Array/Barrier"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GTI Read Throughput"
+             symbol_name="GtiReadThroughput"
+             underscore_name="gti_read_throughput"
+             description="The total number of GPU memory bytes read from GTI."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 512 UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="128  A 36 READ UMUL"
+             mdapi_group="GTI"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GTI Write Throughput"
+             symbol_name="GtiWriteThroughput"
+             underscore_name="gti_write_throughput"
+             description="The total number of GPU memory bytes written to GTI."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 512 UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="C 7 READ C 6 READ UADD C 5 READ UADD C 4 READ UADD C 3 READ UADD C 2 READ UADD C 1 READ UADD C 0 READ UADD 128 UMUL"
+             mdapi_group="GTI"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x00000001" />
+        <register type="NOA" address="0x00009888" value="0x18013800" />
+        <register type="NOA" address="0x00009888" value="0x18213800" />
+        <register type="NOA" address="0x00009888" value="0x18413800" />
+        <register type="NOA" address="0x00009888" value="0x18613800" />
+        <register type="NOA" address="0x00009888" value="0x18813800" />
+        <register type="NOA" address="0x00009888" value="0x18A13800" />
+        <register type="NOA" address="0x00009888" value="0x18C13800" />
+        <register type="NOA" address="0x00009888" value="0x18E13800" />
+        <register type="NOA" address="0x00009888" value="0x02010006" />
+        <register type="NOA" address="0x00009888" value="0x16010004" />
+        <register type="NOA" address="0x00009888" value="0x00010000" />
+        <register type="NOA" address="0x00009888" value="0x02210006" />
+        <register type="NOA" address="0x00009888" value="0x14210100" />
+        <register type="NOA" address="0x00009888" value="0x00210000" />
+        <register type="NOA" address="0x00009888" value="0x02410006" />
+        <register type="NOA" address="0x00009888" value="0x14410004" />
+        <register type="NOA" address="0x00009888" value="0x00410000" />
+        <register type="NOA" address="0x00009888" value="0x02610006" />
+        <register type="NOA" address="0x00009888" value="0x12610100" />
+        <register type="NOA" address="0x00009888" value="0x00610000" />
+        <register type="NOA" address="0x00009888" value="0x02810006" />
+        <register type="NOA" address="0x00009888" value="0x12810004" />
+        <register type="NOA" address="0x00009888" value="0x00810000" />
+        <register type="NOA" address="0x00009888" value="0x02A10006" />
+        <register type="NOA" address="0x00009888" value="0x10A10100" />
+        <register type="NOA" address="0x00009888" value="0x00A10000" />
+        <register type="NOA" address="0x00009888" value="0x02C10006" />
+        <register type="NOA" address="0x00009888" value="0x10C10004" />
+        <register type="NOA" address="0x00009888" value="0x00C10000" />
+        <register type="NOA" address="0x00009888" value="0x02E10006" />
+        <register type="NOA" address="0x00009888" value="0x0EE10100" />
+        <register type="NOA" address="0x00009888" value="0x00E10000" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x02034800" />
+        <register type="NOA" address="0x00009888" value="0x04034924" />
+        <register type="NOA" address="0x00009888" value="0x06030004" />
+        <register type="NOA" address="0x00009888" value="0x02093800" />
+        <register type="NOA" address="0x00009888" value="0x0409000A" />
+        <register type="NOA" address="0x00009888" value="0x020A1200" />
+        <register type="NOA" address="0x00009888" value="0x040A2489" />
+        <register type="NOA" address="0x00009888" value="0x060A0002" />
+        <register type="NOA" address="0x00009888" value="0x040B2700" />
+        <register type="NOA" address="0x00009888" value="0x060B0001" />
+        <register type="NOA" address="0x00009888" value="0x48030000" />
+        <register type="NOA" address="0x00009888" value="0x4A030402" />
+        <register type="NOA" address="0x00009888" value="0x16030000" />
+        <register type="NOA" address="0x00009888" value="0x3E030440" />
+        <register type="NOA" address="0x00009888" value="0x18030000" />
+        <register type="NOA" address="0x00009888" value="0x1A030000" />
+        <register type="NOA" address="0x00009888" value="0x40030000" />
+        <register type="NOA" address="0x00009888" value="0x1C030000" />
+        <register type="NOA" address="0x00009888" value="0x1E030000" />
+        <register type="NOA" address="0x00009888" value="0x20030000" />
+        <register type="NOA" address="0x00009888" value="0x42030000" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x00800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00000000" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="Render Metrics for 3D Pipeline Profile"
+       chipset="DG2"
+       symbol_name="RenderPipeProfile"
+       underscore_name="render_pipe_profile"
+       mdapi_supported_apis="VK OGL OCL IO"
+       hw_config_guid="3b06d2f4-f594-49b5-b399-98550fa2d7b1"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Rasterized Pixels"
+             symbol_name="RasterizedPixels"
+             underscore_name="rasterized_pixels"
+             description="The total number of rasterized pixels."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 21 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Early Hi-Depth Test Fails"
+             symbol_name="HiDepthTestFails"
+             underscore_name="hi_depth_test_fails"
+             description="The total number of pixels dropped on early hierarchical depth test."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 22 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer/Hi-Depth Test"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Early Depth Test Fails"
+             symbol_name="EarlyDepthTestFails"
+             underscore_name="early_depth_test_fails"
+             description="The total number of pixels dropped on early depth test."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 23 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer/Early Depth Test"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Killed in FS"
+             symbol_name="SamplesKilledInPs"
+             underscore_name="samples_killed_in_ps"
+             description="The total number of samples or pixels dropped in fragment shaders."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 24 READ 4 UMUL"
+             mdapi_group="3D Pipe/Fragment Shader"
+             mdapi_usage_flags="Tier4 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Pixels Failing Tests"
+             symbol_name="PixelsFailingPostPsTests"
+             underscore_name="pixels_failing_post_ps_tests"
+             description="The total number of pixels dropped on post-FS alpha, stencil, or depth tests."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 25 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Written"
+             symbol_name="SamplesWritten"
+             underscore_name="samples_written"
+             description="The total number of samples or pixels written to all render targets."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 26 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Blended"
+             symbol_name="SamplesBlended"
+             underscore_name="samples_blended"
+             description="The total number of blended samples or pixels written to all render targets."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 27 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Texels"
+             symbol_name="SamplerTexels"
+             underscore_name="sampler_texels"
+             description="The total number of texels seen on input (with 2x2 accuracy) in all sampler units."
+             data_type="uint64"
+             units="texels"
+             semantic_type="event"
+             equation="A 28 READ 4 UMUL"
+             mdapi_group="Sampler/Sampler Input"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Texels Misses"
+             symbol_name="SamplerTexelMisses"
+             underscore_name="sampler_texel_misses"
+             description="The total number of texels lookups (with 2x2 accuracy) that missed L1 sampler cache."
+             data_type="uint64"
+             units="texels"
+             semantic_type="event"
+             equation="A 29 READ 4 UMUL"
+             mdapi_group="Sampler/Sampler Cache"
+             mdapi_usage_flags="Tier3 Batch Frame Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SLM Bytes Read"
+             symbol_name="SlmBytesRead"
+             underscore_name="slm_bytes_read"
+             description="The total number of GPU memory bytes read from shared local memory."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="A 30 READ 64 UMUL"
+             mdapi_group="L3/Data Port/SLM"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SLM Bytes Written"
+             symbol_name="SlmBytesWritten"
+             underscore_name="slm_bytes_written"
+             description="The total number of GPU memory bytes written into shared local memory."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="A 31 READ 64 UMUL"
+             mdapi_group="L3/Data Port/SLM"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Memory Accesses"
+             symbol_name="ShaderMemoryAccesses"
+             underscore_name="shader_memory_accesses"
+             description="The total number of shader memory accesses to L3."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 32 READ"
+             mdapi_group="L3/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Atomic Memory Accesses"
+             symbol_name="ShaderAtomics"
+             underscore_name="shader_atomics"
+             description="The total number of shader atomic memory accesses."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 34 READ"
+             mdapi_group="L3/Data Port/Atomics"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="L3 Shader Throughput"
+             symbol_name="L3ShaderThroughput"
+             underscore_name="l3_shader_throughput"
+             description="The total number of GPU memory bytes transferred between shaders and L3 caches w/o URB."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="$ShaderMemoryAccesses 64 UMUL"
+             mdapi_group="L3/Data Port"
+             mdapi_usage_flags="Tier2 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Barrier Messages"
+             symbol_name="ShaderBarriers"
+             underscore_name="shader_barriers"
+             description="The total number of shader barrier messages."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 35 READ"
+             mdapi_group="EU Array/Barrier"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VF0 Bottleneck"
+             symbol_name="Vf0Bottleneck"
+             underscore_name="vf0_bottleneck"
+             description="The percentage of time in which vertex fetch pipeline stage was slowing down the 3D pipeline."
+             low_watermark="5"
+             high_watermark="15"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Input Assembler"
+             mdapi_usage_flags="Tier3 Indicate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS0 Bottleneck"
+             symbol_name="Vs0Bottleneck"
+             underscore_name="vs0_bottleneck"
+             description="The percentage of time in which VS0 pipeline stage was slowing down the 3D pipeline."
+             low_watermark="10"
+             high_watermark="30"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Vertex Shader"
+             mdapi_usage_flags="Tier3 Indicate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Clipper0 Bottleneck"
+             symbol_name="Cl0Bottleneck"
+             underscore_name="cl0_bottleneck"
+             description="The percentage of time in which clipper pipeline stage was slowing down the 3D pipeline."
+             low_watermark="10"
+             high_watermark="30"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Clipper"
+             mdapi_usage_flags="Tier3 Indicate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS0 Bottleneck"
+             symbol_name="Hs0Bottleneck"
+             underscore_name="hs0_bottleneck"
+             description="The percentage of time in which hull shader pipeline stage was slowing down the 3D pipeline."
+             low_watermark="3"
+             high_watermark="9"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Hull Shader"
+             mdapi_usage_flags="Tier3 Indicate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS0 Bottleneck"
+             symbol_name="Ds0Bottleneck"
+             underscore_name="ds0_bottleneck"
+             description="The percentage of time in which domain shader pipeline stage was slowing down the 3D pipeline."
+             low_watermark="5"
+             high_watermark="15"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Domain Shader"
+             mdapi_usage_flags="Tier3 Indicate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS0 Bottleneck"
+             symbol_name="Gs0Bottleneck"
+             underscore_name="gs0_bottleneck"
+             description="The percentage of time in which geometry shader pipeline stage was slowing down the 3D pipeline."
+             low_watermark="5"
+             high_watermark="15"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Geometry Shader"
+             mdapi_usage_flags="Tier3 Indicate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SF0 Bottleneck"
+             symbol_name="Sf0Bottleneck"
+             underscore_name="sf0_bottleneck"
+             description="The percentage of time in which strip-fans pipeline stage was slowing down the 3D pipeline."
+             low_watermark="5"
+             high_watermark="10"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Rasterizer/Strip-Fans"
+             mdapi_usage_flags="Tier3 Indicate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SO0 Bottleneck"
+             symbol_name="So0Bottleneck"
+             underscore_name="so0_bottleneck"
+             description="The percentage of time in which stream output pipeline stage was slowing down the 3D pipeline."
+             low_watermark="5"
+             high_watermark="15"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Stream Output"
+             mdapi_usage_flags="Tier3 Indicate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CL0 Stall"
+             symbol_name="Cl0Stall"
+             underscore_name="cl0_stall"
+             description="The percentage of time in which clipper pipeline stage was stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Clipper"
+             mdapi_usage_flags="Tier4 Correlate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS0 Stall"
+             symbol_name="Hs0Stall"
+             underscore_name="hs0_stall"
+             description="The percentage of time in which hull stall pipeline stage was stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Hull Shader"
+             mdapi_usage_flags="Tier4 Correlate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS0 Stall"
+             symbol_name="Ds0Stall"
+             underscore_name="ds0_stall"
+             description="The percentage of time in which domain shader pipeline stage was stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Domain Shader"
+             mdapi_usage_flags="Tier4 Correlate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SF0 Stall"
+             symbol_name="Sf0Stall"
+             underscore_name="sf0_stall"
+             description="The percentage of time in which strip-fans pipeline stage was stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Rasterizer/Strip-Fans"
+             mdapi_usage_flags="Tier4 Correlate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SO0 Stall"
+             symbol_name="So0Stall"
+             underscore_name="so0_stall"
+             description="The percentage of time in which stream-output pipeline stage was stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Stream Output"
+             mdapi_usage_flags="Tier4 Correlate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Hi-Depth Bottleneck"
+             symbol_name="HiDepthBottleneck"
+             underscore_name="hi_depth_bottleneck"
+             description="The percentage of time in which early hierarchical depth test pipeline stage was slowing down the 3D pipeline."
+             low_watermark="5"
+             high_watermark="15"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Rasterizer/Hi-Depth Test"
+             mdapi_usage_flags="Tier3 Indicate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="BC Bottleneck"
+             symbol_name="BcBottleneck"
+             underscore_name="bc_bottleneck"
+             description="The percentage of time in which barycentric coordinates calculation pipeline stage was slowing down the 3D pipeline."
+             low_watermark="5"
+             high_watermark="15"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="3D Pipe/Rasterizer/Barycentric Calc"
+             mdapi_usage_flags="Tier3 Indicate Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x0000000C" />
+        <register type="NOA" address="0x00009888" value="0x168607C0" />
+        <register type="NOA" address="0x00009888" value="0x1603001F" />
+        <register type="NOA" address="0x00009888" value="0x0C030000" />
+        <register type="NOA" address="0x00009888" value="0x12040640" />
+        <register type="NOA" address="0x00009888" value="0x02040000" />
+        <register type="NOA" address="0x00009888" value="0x1C070000" />
+        <register type="NOA" address="0x00009888" value="0x16090000" />
+        <register type="NOA" address="0x00009888" value="0x1C0903C0" />
+        <register type="NOA" address="0x00009888" value="0x160A01F0" />
+        <register type="NOA" address="0x00009888" value="0x160B7C00" />
+        <register type="NOA" address="0x00009888" value="0x04860000" />
+        <register type="NOA" address="0x00009888" value="0x06860400" />
+        <register type="NOA" address="0x00009888" value="0x00860000" />
+        <register type="NOA" address="0x00009888" value="0x02030002" />
+        <register type="NOA" address="0x00009888" value="0x04030010" />
+        <register type="NOA" address="0x00009888" value="0x0A0301C0" />
+        <register type="NOA" address="0x00009888" value="0x100300C0" />
+        <register type="NOA" address="0x00009888" value="0x12030C80" />
+        <register type="NOA" address="0x00009888" value="0x00030000" />
+        <register type="NOA" address="0x00009888" value="0x00044000" />
+        <register type="NOA" address="0x00009888" value="0x08040021" />
+        <register type="NOA" address="0x00009888" value="0x0C040080" />
+        <register type="NOA" address="0x00009888" value="0x0E040022" />
+        <register type="NOA" address="0x00009888" value="0x04074000" />
+        <register type="NOA" address="0x00009888" value="0x0A070021" />
+        <register type="NOA" address="0x00009888" value="0x0E0708C0" />
+        <register type="NOA" address="0x00009888" value="0x00070800" />
+        <register type="NOA" address="0x00009888" value="0x020700A0" />
+        <register type="NOA" address="0x00009888" value="0x00083032" />
+        <register type="NOA" address="0x00009888" value="0x02083300" />
+        <register type="NOA" address="0x00009888" value="0x02094003" />
+        <register type="NOA" address="0x00009888" value="0x08090880" />
+        <register type="NOA" address="0x00009888" value="0x0C090001" />
+        <register type="NOA" address="0x00009888" value="0x0E090023" />
+        <register type="NOA" address="0x00009888" value="0x12090003" />
+        <register type="NOA" address="0x00009888" value="0x00090D28" />
+        <register type="NOA" address="0x00009888" value="0x040A0200" />
+        <register type="NOA" address="0x00009888" value="0x140A0940" />
+        <register type="NOA" address="0x00009888" value="0x000A9A91" />
+        <register type="NOA" address="0x00009888" value="0x020A19AA" />
+        <register type="NOA" address="0x00009888" value="0x00015044" />
+        <register type="NOA" address="0x00009888" value="0x040B0004" />
+        <register type="NOA" address="0x00009888" value="0x080B0000" />
+        <register type="NOA" address="0x00009888" value="0x000B2000" />
+        <register type="NOA" address="0x00009888" value="0x020B2200" />
+        <register type="NOA" address="0x00009888" value="0x00020C20" />
+        <register type="NOA" address="0x00009888" value="0x000C0800" />
+        <register type="NOA" address="0x00009888" value="0x020C00A0" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x0008AAA2" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009888" value="0x00032482" />
+        <register type="NOA" address="0x00009888" value="0x02032492" />
+        <register type="NOA" address="0x00009888" value="0x04032492" />
+        <register type="NOA" address="0x00009888" value="0x06030002" />
+        <register type="NOA" address="0x00009888" value="0x14030000" />
+        <register type="NOA" address="0x00009888" value="0x3E030007" />
+        <register type="NOA" address="0x00009888" value="0x2C030000" />
+        <register type="NOA" address="0x00009888" value="0x4203180A" />
+        <register type="NOA" address="0x00009888" value="0x32031000" />
+        <register type="NOA" address="0x00009888" value="0x440300C2" />
+        <register type="NOA" address="0x00009888" value="0x38031000" />
+        <register type="NOA" address="0x00009888" value="0x46030200" />
+        <register type="NOA" address="0x00009888" value="0x48030261" />
+        <register type="NOA" address="0x00009888" value="0x4A030C22" />
+        <register type="NOA" address="0x00009888" value="0x16030000" />
+        <register type="NOA" address="0x00009888" value="0x18030000" />
+        <register type="NOA" address="0x00009888" value="0x1A030000" />
+        <register type="NOA" address="0x00009888" value="0x400304C1" />
+        <register type="NOA" address="0x00009888" value="0x1C031000" />
+        <register type="NOA" address="0x00009888" value="0x1E030000" />
+        <register type="NOA" address="0x00009888" value="0x20031000" />
+        <register type="NOA" address="0x00009884" value="0x00000008" />
+        <register type="NOA" address="0x00009888" value="0x0001FFFD" />
+        <register type="NOA" address="0x00009888" value="0x00025551" />
+        <register type="NOA" address="0x00009888" value="0x02025555" />
+        <register type="NOA" address="0x00009888" value="0x00045551" />
+        <register type="NOA" address="0x00009888" value="0x02045555" />
+        <register type="NOA" address="0x00009888" value="0x00055551" />
+        <register type="NOA" address="0x00009888" value="0x02055555" />
+        <register type="NOA" address="0x00009888" value="0x0008AAA2" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x00800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00000000" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="Metric set HDCAndSF1"
+       chipset="DG2"
+       symbol_name="HDCAndSF1"
+       underscore_name="hdc_and_sf1"
+       mdapi_supported_apis="VK OGL OCL IO"
+       hw_config_guid="1afb0edc-585c-4be9-8c0c-bc72325106b0"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Rasterized Pixels"
+             symbol_name="RasterizedPixels"
+             underscore_name="rasterized_pixels"
+             description="The total number of rasterized pixels."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 21 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Early Hi-Depth Test Fails"
+             symbol_name="HiDepthTestFails"
+             underscore_name="hi_depth_test_fails"
+             description="The total number of pixels dropped on early hierarchical depth test."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 22 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer/Hi-Depth Test"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Early Depth Test Fails"
+             symbol_name="EarlyDepthTestFails"
+             underscore_name="early_depth_test_fails"
+             description="The total number of pixels dropped on early depth test."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 23 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer/Early Depth Test"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Killed in FS"
+             symbol_name="SamplesKilledInPs"
+             underscore_name="samples_killed_in_ps"
+             description="The total number of samples or pixels dropped in fragment shaders."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 24 READ 4 UMUL"
+             mdapi_group="3D Pipe/Fragment Shader"
+             mdapi_usage_flags="Tier4 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Pixels Failing Tests"
+             symbol_name="PixelsFailingPostPsTests"
+             underscore_name="pixels_failing_post_ps_tests"
+             description="The total number of pixels dropped on post-FS alpha, stencil, or depth tests."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 25 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Written"
+             symbol_name="SamplesWritten"
+             underscore_name="samples_written"
+             description="The total number of samples or pixels written to all render targets."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 26 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Blended"
+             symbol_name="SamplesBlended"
+             underscore_name="samples_blended"
+             description="The total number of blended samples or pixels written to all render targets."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 27 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Texels"
+             symbol_name="SamplerTexels"
+             underscore_name="sampler_texels"
+             description="The total number of texels seen on input (with 2x2 accuracy) in all sampler units."
+             data_type="uint64"
+             units="texels"
+             semantic_type="event"
+             equation="A 28 READ 4 UMUL"
+             mdapi_group="Sampler/Sampler Input"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Texels Misses"
+             symbol_name="SamplerTexelMisses"
+             underscore_name="sampler_texel_misses"
+             description="The total number of texels lookups (with 2x2 accuracy) that missed L1 sampler cache."
+             data_type="uint64"
+             units="texels"
+             semantic_type="event"
+             equation="A 29 READ 4 UMUL"
+             mdapi_group="Sampler/Sampler Cache"
+             mdapi_usage_flags="Tier3 Batch Frame Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SLM Bytes Read"
+             symbol_name="SlmBytesRead"
+             underscore_name="slm_bytes_read"
+             description="The total number of GPU memory bytes read from shared local memory."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="A 30 READ 64 UMUL"
+             mdapi_group="L3/Data Port/SLM"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SLM Bytes Written"
+             symbol_name="SlmBytesWritten"
+             underscore_name="slm_bytes_written"
+             description="The total number of GPU memory bytes written into shared local memory."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="A 31 READ 64 UMUL"
+             mdapi_group="L3/Data Port/SLM"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Memory Accesses"
+             symbol_name="ShaderMemoryAccesses"
+             underscore_name="shader_memory_accesses"
+             description="The total number of shader memory accesses to L3."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 32 READ"
+             mdapi_group="L3/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Atomic Memory Accesses"
+             symbol_name="ShaderAtomics"
+             underscore_name="shader_atomics"
+             description="The total number of shader atomic memory accesses."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 34 READ"
+             mdapi_group="L3/Data Port/Atomics"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="L3 Shader Throughput"
+             symbol_name="L3ShaderThroughput"
+             underscore_name="l3_shader_throughput"
+             description="The total number of GPU memory bytes transferred between shaders and L3 caches w/o URB."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="$ShaderMemoryAccesses 64 UMUL"
+             mdapi_group="L3/Data Port"
+             mdapi_usage_flags="Tier2 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Barrier Messages"
+             symbol_name="ShaderBarriers"
+             underscore_name="shader_barriers"
+             description="The total number of shader barrier messages."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 35 READ"
+             mdapi_group="EU Array/Barrier"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Slice0 Dualsubslice0 Non-sampler Shader Access Stalled On L3"
+             symbol_name="NonSamplerShader00AccessStalledOnL3"
+             underscore_name="non_sampler_shader00_access_stalled_on_l3"
+             description="Percentage of time when HDC has messages to L3, but it's stalled due to lack of credits (Slice0 Dualsubslice0)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 1 AND"
+             mdapi_group="GPU/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Slice0 Dualsubslice1 Non-sampler Shader Access Stalled On L3"
+             symbol_name="NonSamplerShader01AccessStalledOnL3"
+             underscore_name="non_sampler_shader01_access_stalled_on_l3"
+             description="Percentage of time when HDC has messages to L3, but it's stalled due to lack of credits (Slice0 Dualsubslice1)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 2 AND"
+             mdapi_group="GPU/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Slice0 Dualsubslice2 Non-sampler Shader Access Stalled On L3"
+             symbol_name="NonSamplerShader02AccessStalledOnL3"
+             underscore_name="non_sampler_shader02_access_stalled_on_l3"
+             description="Percentage of time when HDC has messages to L3, but it's stalled due to lack of credits (Slice0 Dualsubslice2)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 4 AND"
+             mdapi_group="GPU/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Slice0 Dualsubslice3 Non-sampler Shader Access Stalled On L3"
+             symbol_name="NonSamplerShader03AccessStalledOnL3"
+             underscore_name="non_sampler_shader03_access_stalled_on_l3"
+             description="Percentage of time when HDC has messages to L3, but it's stalled due to lack of credits (Slice0 Dualsubslice3)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 8 AND"
+             mdapi_group="GPU/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Slice1 Dualsubslice0 Non-sampler Shader Access Stalled On L3"
+             symbol_name="NonSamplerShader10AccessStalledOnL3"
+             underscore_name="non_sampler_shader10_access_stalled_on_l3"
+             description="Percentage of time when HDC has messages to L3, but it's stalled due to lack of credits (Slice1 Dualsubslice0)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 64 AND"
+             mdapi_group="GPU/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Slice1 Dualsubslice1 Non-sampler Shader Access Stalled On L3"
+             symbol_name="NonSamplerShader11AccessStalledOnL3"
+             underscore_name="non_sampler_shader11_access_stalled_on_l3"
+             description="Percentage of time when HDC has messages to L3, but it's stalled due to lack of credits (Slice1 Dualsubslice1)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 128 AND"
+             mdapi_group="GPU/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Slice1 Dualsubslice2 Non-sampler Shader Access Stalled On L3"
+             symbol_name="NonSamplerShader12AccessStalledOnL3"
+             underscore_name="non_sampler_shader12_access_stalled_on_l3"
+             description="Percentage of time when HDC has messages to L3, but it's stalled due to lack of credits (Slice1 Dualsubslice2)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 256 AND"
+             mdapi_group="GPU/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Slice1 Dualsubslice3 Non-sampler Shader Access Stalled On L3"
+             symbol_name="NonSamplerShader13AccessStalledOnL3"
+             underscore_name="non_sampler_shader13_access_stalled_on_l3"
+             description="Percentage of time when HDC has messages to L3, but it's stalled due to lack of credits (Slice1 Dualsubslice3)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 512 AND"
+             mdapi_group="GPU/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="SQ00 is full"
+             symbol_name="GTRequestQueue00Full"
+             underscore_name="gt_request_queue00_full"
+             description="The percentage of time when IDI0 SQ00 is filled above a threshold (usually 48 entries)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SQ10 is full"
+             symbol_name="GTRequestQueue10Full"
+             underscore_name="gt_request_queue10_full"
+             description="The percentage of time when IDI0 SQ10 is filled above a threshold (usually 48 entries)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SQ20 is full"
+             symbol_name="GTRequestQueue20Full"
+             underscore_name="gt_request_queue20_full"
+             description="The percentage of time when IDI0 SQ20 is filled above a threshold (usually 48 entries)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SQ30 is full"
+             symbol_name="GTRequestQueue30Full"
+             underscore_name="gt_request_queue30_full"
+             description="The percentage of time when IDI0 SQ30 is filled above a threshold (usually 48 entries)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SQ40 is full"
+             symbol_name="GTRequestQueue40Full"
+             underscore_name="gt_request_queue40_full"
+             description="The percentage of time when IDI0 SQ40 is filled above a threshold (usually 48 entries)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SQ50 is full"
+             symbol_name="GTRequestQueue50Full"
+             underscore_name="gt_request_queue50_full"
+             description="The percentage of time when IDI0 SQ50 is filled above a threshold (usually 48 entries)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SQ60 is full"
+             symbol_name="GTRequestQueue60Full"
+             underscore_name="gt_request_queue60_full"
+             description="The percentage of time when IDI0 SQ60 is filled above a threshold (usually 48 entries)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SQ70 is full"
+             symbol_name="GTRequestQueue70Full"
+             underscore_name="gt_request_queue70_full"
+             description="The percentage of time when IDI0 SQ70 is filled above a threshold (usually 48 entries)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x0000000C" />
+        <register type="NOA" address="0x00009888" value="0x1E810200" />
+        <register type="NOA" address="0x00009888" value="0x1EA10200" />
+        <register type="NOA" address="0x00009888" value="0x1EC10200" />
+        <register type="NOA" address="0x00009888" value="0x1EE10200" />
+        <register type="NOA" address="0x00009888" value="0x1F810200" />
+        <register type="NOA" address="0x00009888" value="0x1FA10200" />
+        <register type="NOA" address="0x00009888" value="0x1FC10200" />
+        <register type="NOA" address="0x00009888" value="0x1FE10200" />
+        <register type="NOA" address="0x00009888" value="0x04810400" />
+        <register type="NOA" address="0x00009888" value="0x08810015" />
+        <register type="NOA" address="0x00009888" value="0x00810000" />
+        <register type="NOA" address="0x00009888" value="0x00830002" />
+        <register type="NOA" address="0x00009888" value="0x00840001" />
+        <register type="NOA" address="0x00009888" value="0x00860001" />
+        <register type="NOA" address="0x00009888" value="0x00870001" />
+        <register type="NOA" address="0x00009888" value="0x04A10400" />
+        <register type="NOA" address="0x00009888" value="0x08A10540" />
+        <register type="NOA" address="0x00009888" value="0x00A10000" />
+        <register type="NOA" address="0x00009888" value="0x00A30008" />
+        <register type="NOA" address="0x00009888" value="0x00A40002" />
+        <register type="NOA" address="0x00009888" value="0x00A60004" />
+        <register type="NOA" address="0x00009888" value="0x00A70004" />
+        <register type="NOA" address="0x00009888" value="0x04C10400" />
+        <register type="NOA" address="0x00009888" value="0x0AC10015" />
+        <register type="NOA" address="0x00009888" value="0x08C10000" />
+        <register type="NOA" address="0x00009888" value="0x00C10000" />
+        <register type="NOA" address="0x00009888" value="0x00C30020" />
+        <register type="NOA" address="0x00009888" value="0x00C40004" />
+        <register type="NOA" address="0x00009888" value="0x00C60010" />
+        <register type="NOA" address="0x00009888" value="0x00C70010" />
+        <register type="NOA" address="0x00009888" value="0x04E10400" />
+        <register type="NOA" address="0x00009888" value="0x0AE10540" />
+        <register type="NOA" address="0x00009888" value="0x08E10000" />
+        <register type="NOA" address="0x00009888" value="0x00E10000" />
+        <register type="NOA" address="0x00009888" value="0x00E30080" />
+        <register type="NOA" address="0x00009888" value="0x00E40008" />
+        <register type="NOA" address="0x00009888" value="0x00E60040" />
+        <register type="NOA" address="0x00009888" value="0x00E70040" />
+        <register type="NOA" address="0x00009888" value="0x05810400" />
+        <register type="NOA" address="0x00009888" value="0x0D810015" />
+        <register type="NOA" address="0x00009888" value="0x09810000" />
+        <register type="NOA" address="0x00009888" value="0x01810000" />
+        <register type="NOA" address="0x00009888" value="0x01830200" />
+        <register type="NOA" address="0x00009888" value="0x01840010" />
+        <register type="NOA" address="0x00009888" value="0x01860100" />
+        <register type="NOA" address="0x00009888" value="0x01870100" />
+        <register type="NOA" address="0x00009888" value="0x05A10400" />
+        <register type="NOA" address="0x00009888" value="0x0DA10540" />
+        <register type="NOA" address="0x00009888" value="0x09A10000" />
+        <register type="NOA" address="0x00009888" value="0x01A10000" />
+        <register type="NOA" address="0x00009888" value="0x01A30800" />
+        <register type="NOA" address="0x00009888" value="0x01A40020" />
+        <register type="NOA" address="0x00009888" value="0x01A60400" />
+        <register type="NOA" address="0x00009888" value="0x01A70400" />
+        <register type="NOA" address="0x00009888" value="0x05C10400" />
+        <register type="NOA" address="0x00009888" value="0x0FC10015" />
+        <register type="NOA" address="0x00009888" value="0x09C10000" />
+        <register type="NOA" address="0x00009888" value="0x01C10000" />
+        <register type="NOA" address="0x00009888" value="0x01C32000" />
+        <register type="NOA" address="0x00009888" value="0x01C40040" />
+        <register type="NOA" address="0x00009888" value="0x01C61000" />
+        <register type="NOA" address="0x00009888" value="0x01C71000" />
+        <register type="NOA" address="0x00009888" value="0x05E10400" />
+        <register type="NOA" address="0x00009888" value="0x0FE10540" />
+        <register type="NOA" address="0x00009888" value="0x09E10000" />
+        <register type="NOA" address="0x00009888" value="0x01E10000" />
+        <register type="NOA" address="0x00009888" value="0x01E38000" />
+        <register type="NOA" address="0x00009888" value="0x01E40080" />
+        <register type="NOA" address="0x00009888" value="0x01E64000" />
+        <register type="NOA" address="0x00009888" value="0x01E74000" />
+        <register type="NOA" address="0x00009888" value="0x000300F0" />
+        <register type="NOA" address="0x00009888" value="0x00070060" />
+        <register type="NOA" address="0x00009888" value="0x0008FF0E" />
+        <register type="NOA" address="0x00009888" value="0x0009000C" />
+        <register type="NOA" address="0x00009888" value="0x000A55A5" />
+        <register type="NOA" address="0x00009888" value="0x000100F2" />
+        <register type="NOA" address="0x00009888" value="0x000BAA04" />
+        <register type="NOA" address="0x00009888" value="0x00020004" />
+        <register type="NOA" address="0x00009888" value="0x000C0010" />
+        <register type="NOA" address="0x00009888" value="0x01076000" />
+        <register type="NOA" address="0x00009888" value="0x01080E00" />
+        <register type="NOA" address="0x00009888" value="0x010900C0" />
+        <register type="NOA" address="0x00009888" value="0x010AA500" />
+        <register type="NOA" address="0x00009888" value="0x01010020" />
+        <register type="NOA" address="0x00009888" value="0x010B0400" />
+        <register type="NOA" address="0x00009888" value="0x01020040" />
+        <register type="NOA" address="0x00009888" value="0x010C1000" />
+        <register type="NOA" address="0x00009884" value="0x00000001" />
+        <register type="NOA" address="0x00009888" value="0x0C017000" />
+        <register type="NOA" address="0x00009888" value="0x0C217000" />
+        <register type="NOA" address="0x00009888" value="0x0C417000" />
+        <register type="NOA" address="0x00009888" value="0x0C617000" />
+        <register type="NOA" address="0x00009888" value="0x0C817000" />
+        <register type="NOA" address="0x00009888" value="0x0CA17000" />
+        <register type="NOA" address="0x00009888" value="0x0CC17000" />
+        <register type="NOA" address="0x00009888" value="0x0CE17000" />
+        <register type="NOA" address="0x00009888" value="0x02010010" />
+        <register type="NOA" address="0x00009888" value="0x16010006" />
+        <register type="NOA" address="0x00009888" value="0x00010000" />
+        <register type="NOA" address="0x00009888" value="0x02210010" />
+        <register type="NOA" address="0x00009888" value="0x14210180" />
+        <register type="NOA" address="0x00009888" value="0x00210000" />
+        <register type="NOA" address="0x00009888" value="0x02410010" />
+        <register type="NOA" address="0x00009888" value="0x14410006" />
+        <register type="NOA" address="0x00009888" value="0x00410000" />
+        <register type="NOA" address="0x00009888" value="0x02610010" />
+        <register type="NOA" address="0x00009888" value="0x12610180" />
+        <register type="NOA" address="0x00009888" value="0x00610000" />
+        <register type="NOA" address="0x00009888" value="0x02810010" />
+        <register type="NOA" address="0x00009888" value="0x12810006" />
+        <register type="NOA" address="0x00009888" value="0x00810000" />
+        <register type="NOA" address="0x00009888" value="0x02A10010" />
+        <register type="NOA" address="0x00009888" value="0x10A10180" />
+        <register type="NOA" address="0x00009888" value="0x00A10000" />
+        <register type="NOA" address="0x00009888" value="0x02C10010" />
+        <register type="NOA" address="0x00009888" value="0x10C10006" />
+        <register type="NOA" address="0x00009888" value="0x00C10000" />
+        <register type="NOA" address="0x00009888" value="0x02E10010" />
+        <register type="NOA" address="0x00009888" value="0x0EE10180" />
+        <register type="NOA" address="0x00009888" value="0x00E10000" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
+        <register type="NOA" address="0x00009888" value="0x00032492" />
+        <register type="NOA" address="0x00009888" value="0x02034892" />
+        <register type="NOA" address="0x00009888" value="0x04034924" />
+        <register type="NOA" address="0x00009888" value="0x06030004" />
+        <register type="NOA" address="0x00009888" value="0x02093800" />
+        <register type="NOA" address="0x00009888" value="0x0409000A" />
+        <register type="NOA" address="0x00009888" value="0x020A1200" />
+        <register type="NOA" address="0x00009888" value="0x040A2489" />
+        <register type="NOA" address="0x00009888" value="0x060A0002" />
+        <register type="NOA" address="0x00009888" value="0x040B2700" />
+        <register type="NOA" address="0x00009888" value="0x060B0001" />
+        <register type="NOA" address="0x00009888" value="0x14037000" />
+        <register type="NOA" address="0x00009888" value="0x3E0314CA" />
+        <register type="NOA" address="0x00009888" value="0x26037000" />
+        <register type="NOA" address="0x00009888" value="0x42032004" />
+        <register type="NOA" address="0x00009888" value="0x2C033000" />
+        <register type="NOA" address="0x00009888" value="0x32037000" />
+        <register type="NOA" address="0x00009888" value="0x44030047" />
+        <register type="NOA" address="0x00009888" value="0x38031000" />
+        <register type="NOA" address="0x00009888" value="0x46030003" />
+        <register type="NOA" address="0x00009888" value="0x48031EA0" />
+        <register type="NOA" address="0x00009888" value="0x4A030DE6" />
+        <register type="NOA" address="0x00009888" value="0x1603F000" />
+        <register type="NOA" address="0x00009888" value="0x1803F000" />
+        <register type="NOA" address="0x00009888" value="0x1A03F000" />
+        <register type="NOA" address="0x00009888" value="0x40031084" />
+        <register type="NOA" address="0x00009888" value="0x1C03F000" />
+        <register type="NOA" address="0x00009888" value="0x1E03F000" />
+        <register type="NOA" address="0x00009888" value="0x2003F000" />
+        <register type="NOA" address="0x00009884" value="0x00000008" />
+        <register type="NOA" address="0x00009888" value="0x000100FF" />
+        <register type="NOA" address="0x00009888" value="0x00025555" />
+        <register type="NOA" address="0x00009888" value="0x00045555" />
+        <register type="NOA" address="0x00009888" value="0x00055555" />
+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0xF0800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0xF0800000" />
+        <register type="OA" address="0x0000DC40" value="0x00FF0000" />
+        <register type="OA" address="0x0000D940" value="0x00000002" />
+        <register type="OA" address="0x0000D944" value="0x0000FFFE" />
+        <register type="OA" address="0x0000DC00" value="0x00000002" />
+        <register type="OA" address="0x0000DC04" value="0x0000FFFE" />
+        <register type="OA" address="0x0000D948" value="0x00000002" />
+        <register type="OA" address="0x0000D94C" value="0x0000FFFD" />
+        <register type="OA" address="0x0000DC08" value="0x00000002" />
+        <register type="OA" address="0x0000DC0C" value="0x0000FFFD" />
+        <register type="OA" address="0x0000D950" value="0x00000002" />
+        <register type="OA" address="0x0000D954" value="0x0000FFFB" />
+        <register type="OA" address="0x0000DC10" value="0x00000002" />
+        <register type="OA" address="0x0000DC14" value="0x0000FFFB" />
+        <register type="OA" address="0x0000D958" value="0x00000002" />
+        <register type="OA" address="0x0000D95C" value="0x0000FFF7" />
+        <register type="OA" address="0x0000DC18" value="0x00000002" />
+        <register type="OA" address="0x0000DC1C" value="0x0000FFF7" />
+        <register type="OA" address="0x0000D960" value="0x00000002" />
+        <register type="OA" address="0x0000D964" value="0x0000FFEF" />
+        <register type="OA" address="0x0000DC20" value="0x00000002" />
+        <register type="OA" address="0x0000DC24" value="0x0000FFEF" />
+        <register type="OA" address="0x0000D968" value="0x00000002" />
+        <register type="OA" address="0x0000D96C" value="0x0000FFDF" />
+        <register type="OA" address="0x0000DC28" value="0x00000002" />
+        <register type="OA" address="0x0000DC2C" value="0x0000FFDF" />
+        <register type="OA" address="0x0000D970" value="0x00000002" />
+        <register type="OA" address="0x0000D974" value="0x0000FFBF" />
+        <register type="OA" address="0x0000DC30" value="0x00000002" />
+        <register type="OA" address="0x0000DC34" value="0x0000FFBF" />
+        <register type="OA" address="0x0000D978" value="0x00000002" />
+        <register type="OA" address="0x0000D97C" value="0x0000FF7F" />
+        <register type="OA" address="0x0000DC38" value="0x00000002" />
+        <register type="OA" address="0x0000DC3C" value="0x0000FF7F" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="Metric set HDCAndSF2"
+       chipset="DG2"
+       symbol_name="HDCAndSF2"
+       underscore_name="hdc_and_sf2"
+       mdapi_supported_apis="VK OGL OCL IO"
+       hw_config_guid="a3b35902-f293-41fe-97d0-3d895bc5fab3"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Rasterized Pixels"
+             symbol_name="RasterizedPixels"
+             underscore_name="rasterized_pixels"
+             description="The total number of rasterized pixels."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 21 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Early Hi-Depth Test Fails"
+             symbol_name="HiDepthTestFails"
+             underscore_name="hi_depth_test_fails"
+             description="The total number of pixels dropped on early hierarchical depth test."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 22 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer/Hi-Depth Test"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Early Depth Test Fails"
+             symbol_name="EarlyDepthTestFails"
+             underscore_name="early_depth_test_fails"
+             description="The total number of pixels dropped on early depth test."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 23 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer/Early Depth Test"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Killed in FS"
+             symbol_name="SamplesKilledInPs"
+             underscore_name="samples_killed_in_ps"
+             description="The total number of samples or pixels dropped in fragment shaders."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 24 READ 4 UMUL"
+             mdapi_group="3D Pipe/Fragment Shader"
+             mdapi_usage_flags="Tier4 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Pixels Failing Tests"
+             symbol_name="PixelsFailingPostPsTests"
+             underscore_name="pixels_failing_post_ps_tests"
+             description="The total number of pixels dropped on post-FS alpha, stencil, or depth tests."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 25 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Written"
+             symbol_name="SamplesWritten"
+             underscore_name="samples_written"
+             description="The total number of samples or pixels written to all render targets."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 26 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Blended"
+             symbol_name="SamplesBlended"
+             underscore_name="samples_blended"
+             description="The total number of blended samples or pixels written to all render targets."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 27 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Texels"
+             symbol_name="SamplerTexels"
+             underscore_name="sampler_texels"
+             description="The total number of texels seen on input (with 2x2 accuracy) in all sampler units."
+             data_type="uint64"
+             units="texels"
+             semantic_type="event"
+             equation="A 28 READ 4 UMUL"
+             mdapi_group="Sampler/Sampler Input"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Texels Misses"
+             symbol_name="SamplerTexelMisses"
+             underscore_name="sampler_texel_misses"
+             description="The total number of texels lookups (with 2x2 accuracy) that missed L1 sampler cache."
+             data_type="uint64"
+             units="texels"
+             semantic_type="event"
+             equation="A 29 READ 4 UMUL"
+             mdapi_group="Sampler/Sampler Cache"
+             mdapi_usage_flags="Tier3 Batch Frame Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SLM Bytes Read"
+             symbol_name="SlmBytesRead"
+             underscore_name="slm_bytes_read"
+             description="The total number of GPU memory bytes read from shared local memory."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="A 30 READ 64 UMUL"
+             mdapi_group="L3/Data Port/SLM"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SLM Bytes Written"
+             symbol_name="SlmBytesWritten"
+             underscore_name="slm_bytes_written"
+             description="The total number of GPU memory bytes written into shared local memory."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="A 31 READ 64 UMUL"
+             mdapi_group="L3/Data Port/SLM"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Memory Accesses"
+             symbol_name="ShaderMemoryAccesses"
+             underscore_name="shader_memory_accesses"
+             description="The total number of shader memory accesses to L3."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 32 READ"
+             mdapi_group="L3/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Atomic Memory Accesses"
+             symbol_name="ShaderAtomics"
+             underscore_name="shader_atomics"
+             description="The total number of shader atomic memory accesses."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 34 READ"
+             mdapi_group="L3/Data Port/Atomics"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="L3 Shader Throughput"
+             symbol_name="L3ShaderThroughput"
+             underscore_name="l3_shader_throughput"
+             description="The total number of GPU memory bytes transferred between shaders and L3 caches w/o URB."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="$ShaderMemoryAccesses 64 UMUL"
+             mdapi_group="L3/Data Port"
+             mdapi_usage_flags="Tier2 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Barrier Messages"
+             symbol_name="ShaderBarriers"
+             underscore_name="shader_barriers"
+             description="The total number of shader barrier messages."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 35 READ"
+             mdapi_group="EU Array/Barrier"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Polygon (Slice0) Data Ready"
+             symbol_name="Poly0DataReady"
+             underscore_name="poly0_data_ready"
+             description="The percentage of time in which geometry pipeline output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU/3D Pipe/Strip-Fans"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Polygon (Slice1) Data Ready"
+             symbol_name="Poly1DataReady"
+             underscore_name="poly1_data_ready"
+             description="The percentage of time in which geometry pipeline output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU/3D Pipe/Strip-Fans"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x0000000C" />
+        <register type="NOA" address="0x00009888" value="0x12056400" />
+        <register type="NOA" address="0x00009888" value="0x13056400" />
+        <register type="NOA" address="0x00009888" value="0x00030001" />
+        <register type="NOA" address="0x00009888" value="0x00050002" />
+        <register type="NOA" address="0x00009888" value="0x02050700" />
+        <register type="NOA" address="0x00009888" value="0x00060002" />
+        <register type="NOA" address="0x00009888" value="0x00080007" />
+        <register type="NOA" address="0x00009888" value="0x000A0005" />
+        <register type="NOA" address="0x00009888" value="0x00010001" />
+        <register type="NOA" address="0x00009888" value="0x000B0002" />
+        <register type="NOA" address="0x00009888" value="0x01050002" />
+        <register type="NOA" address="0x00009888" value="0x0305001C" />
+        <register type="NOA" address="0x00009888" value="0x01060001" />
+        <register type="NOA" address="0x00009888" value="0x01080001" />
+        <register type="NOA" address="0x00009888" value="0x010A0001" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x0008000A" />
+        <register type="NOA" address="0x00009888" value="0x00030012" />
+        <register type="NOA" address="0x00009888" value="0x14030000" />
+        <register type="NOA" address="0x00009888" value="0x3E030000" />
+        <register type="NOA" address="0x00009888" value="0x26033000" />
+        <register type="NOA" address="0x00009888" value="0x42030100" />
+        <register type="NOA" address="0x00009888" value="0x02030000" />
+        <register type="NOA" address="0x00009884" value="0x00000008" />
+        <register type="NOA" address="0x00009888" value="0x00010003" />
+        <register type="NOA" address="0x00009888" value="0x00020005" />
+        <register type="NOA" address="0x00009888" value="0x00040005" />
+        <register type="NOA" address="0x00009888" value="0x00050005" />
+        <register type="NOA" address="0x00009888" value="0x0008000A" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x30800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00030000" />
+        <register type="OA" address="0x0000D940" value="0x00000002" />
+        <register type="OA" address="0x0000D944" value="0x0000FFFE" />
+        <register type="OA" address="0x0000DC00" value="0x00000002" />
+        <register type="OA" address="0x0000DC04" value="0x0000FFFE" />
+        <register type="OA" address="0x0000D948" value="0x00000002" />
+        <register type="OA" address="0x0000D94C" value="0x0000FFFD" />
+        <register type="OA" address="0x0000DC08" value="0x00000002" />
+        <register type="OA" address="0x0000DC0C" value="0x0000FFFD" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="Metric set RasterizerAndPixelBackend1"
+       chipset="DG2"
+       symbol_name="RasterizerAndPixelBackend1"
+       underscore_name="rasterizer_and_pixel_backend1"
+       mdapi_supported_apis="VK OGL OCL IO"
+       hw_config_guid="b04e9460-630f-4682-b0ed-ebb0e74a6c9d"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Rasterized Pixels"
+             symbol_name="RasterizedPixels"
+             underscore_name="rasterized_pixels"
+             description="The total number of rasterized pixels."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 21 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Early Hi-Depth Test Fails"
+             symbol_name="HiDepthTestFails"
+             underscore_name="hi_depth_test_fails"
+             description="The total number of pixels dropped on early hierarchical depth test."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 22 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer/Hi-Depth Test"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Early Depth Test Fails"
+             symbol_name="EarlyDepthTestFails"
+             underscore_name="early_depth_test_fails"
+             description="The total number of pixels dropped on early depth test."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 23 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer/Early Depth Test"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Killed in FS"
+             symbol_name="SamplesKilledInPs"
+             underscore_name="samples_killed_in_ps"
+             description="The total number of samples or pixels dropped in fragment shaders."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 24 READ 4 UMUL"
+             mdapi_group="3D Pipe/Fragment Shader"
+             mdapi_usage_flags="Tier4 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Pixels Failing Tests"
+             symbol_name="PixelsFailingPostPsTests"
+             underscore_name="pixels_failing_post_ps_tests"
+             description="The total number of pixels dropped on post-FS alpha, stencil, or depth tests."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 25 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Written"
+             symbol_name="SamplesWritten"
+             underscore_name="samples_written"
+             description="The total number of samples or pixels written to all render targets."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 26 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Blended"
+             symbol_name="SamplesBlended"
+             underscore_name="samples_blended"
+             description="The total number of blended samples or pixels written to all render targets."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 27 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Texels"
+             symbol_name="SamplerTexels"
+             underscore_name="sampler_texels"
+             description="The total number of texels seen on input (with 2x2 accuracy) in all sampler units."
+             data_type="uint64"
+             units="texels"
+             semantic_type="event"
+             equation="A 28 READ 4 UMUL"
+             mdapi_group="Sampler/Sampler Input"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Texels Misses"
+             symbol_name="SamplerTexelMisses"
+             underscore_name="sampler_texel_misses"
+             description="The total number of texels lookups (with 2x2 accuracy) that missed L1 sampler cache."
+             data_type="uint64"
+             units="texels"
+             semantic_type="event"
+             equation="A 29 READ 4 UMUL"
+             mdapi_group="Sampler/Sampler Cache"
+             mdapi_usage_flags="Tier3 Batch Frame Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SLM Bytes Read"
+             symbol_name="SlmBytesRead"
+             underscore_name="slm_bytes_read"
+             description="The total number of GPU memory bytes read from shared local memory."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="A 30 READ 64 UMUL"
+             mdapi_group="L3/Data Port/SLM"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SLM Bytes Written"
+             symbol_name="SlmBytesWritten"
+             underscore_name="slm_bytes_written"
+             description="The total number of GPU memory bytes written into shared local memory."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="A 31 READ 64 UMUL"
+             mdapi_group="L3/Data Port/SLM"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Memory Accesses"
+             symbol_name="ShaderMemoryAccesses"
+             underscore_name="shader_memory_accesses"
+             description="The total number of shader memory accesses to L3."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 32 READ"
+             mdapi_group="L3/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Atomic Memory Accesses"
+             symbol_name="ShaderAtomics"
+             underscore_name="shader_atomics"
+             description="The total number of shader atomic memory accesses."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 34 READ"
+             mdapi_group="L3/Data Port/Atomics"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="L3 Shader Throughput"
+             symbol_name="L3ShaderThroughput"
+             underscore_name="l3_shader_throughput"
+             description="The total number of GPU memory bytes transferred between shaders and L3 caches w/o URB."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="$ShaderMemoryAccesses 64 UMUL"
+             mdapi_group="L3/Data Port"
+             mdapi_usage_flags="Tier2 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Barrier Messages"
+             symbol_name="ShaderBarriers"
+             underscore_name="shader_barriers"
+             description="The total number of shader barrier messages."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 35 READ"
+             mdapi_group="EU Array/Barrier"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Slice0 Rasterizer Input Available"
+             symbol_name="Rasterizer0InputAvailable"
+             underscore_name="rasterizer0_input_available"
+             description="The percentage of time in which slice0 rasterizer input is available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$SliceMask 1 AND"
+             mdapi_group="GPU/Rasterizer"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Slice1 Rasterizer Output Ready"
+             symbol_name="Rasterizer1OutputReady"
+             underscore_name="rasterizer1_output_ready"
+             description="The percentage of time in which slice1 rasterizer output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$SliceMask 2 AND"
+             mdapi_group="GPU/Rasterizer"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Slice0 Pipe0 Post-EarlyZ Pixel Data Ready"
+             symbol_name="PixelData00Ready"
+             underscore_name="pixel_data00_ready"
+             description="The percentage of time in which slice0  pipe0 post-EarlyZ pixel data is ready (after early Z tests have been applied)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU/Rasterizer/Early Depth Test"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Slice0 Pipe1 Post-EarlyZ Pixel Data Ready"
+             symbol_name="PixelData01Ready"
+             underscore_name="pixel_data01_ready"
+             description="The percentage of time in which slice0  pipe1 post-EarlyZ pixel data is ready (after early Z tests have been applied)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU/Rasterizer/Early Depth Test"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x0000000C" />
+        <register type="NOA" address="0x00009888" value="0x1A056000" />
+        <register type="NOA" address="0x00009888" value="0x1C050000" />
+        <register type="NOA" address="0x00009888" value="0x0C0B7000" />
+        <register type="NOA" address="0x00009888" value="0x0E0B6000" />
+        <register type="NOA" address="0x00009888" value="0x0C0C701F" />
+        <register type="NOA" address="0x00009888" value="0x0E0C6000" />
+        <register type="NOA" address="0x00009888" value="0x1B057000" />
+        <register type="NOA" address="0x00009888" value="0x00038000" />
+        <register type="NOA" address="0x00009888" value="0x00050000" />
+        <register type="NOA" address="0x00009888" value="0x020535D2" />
+        <register type="NOA" address="0x00009888" value="0x0405003E" />
+        <register type="NOA" address="0x00009888" value="0x00060007" />
+        <register type="NOA" address="0x00009888" value="0x00072800" />
+        <register type="NOA" address="0x00009888" value="0x000803D5" />
+        <register type="NOA" address="0x00009888" value="0x0208C000" />
+        <register type="NOA" address="0x00009888" value="0x00090060" />
+        <register type="NOA" address="0x00009888" value="0x000A2955" />
+        <register type="NOA" address="0x00009888" value="0x020A4000" />
+        <register type="NOA" address="0x00009888" value="0x00018018" />
+        <register type="NOA" address="0x00009888" value="0x040B0200" />
+        <register type="NOA" address="0x00009888" value="0x080B0400" />
+        <register type="NOA" address="0x00009888" value="0x0A0B001F" />
+        <register type="NOA" address="0x00009888" value="0x000B0000" />
+        <register type="NOA" address="0x00009888" value="0x020B8000" />
+        <register type="NOA" address="0x00009888" value="0x00020060" />
+        <register type="NOA" address="0x00009888" value="0x040C0200" />
+        <register type="NOA" address="0x00009888" value="0x0A0C0400" />
+        <register type="NOA" address="0x00009888" value="0x080C0000" />
+        <register type="NOA" address="0x00009888" value="0x000C0000" />
+        <register type="NOA" address="0x00009888" value="0x01050000" />
+        <register type="NOA" address="0x00009888" value="0x11050DC0" />
+        <register type="NOA" address="0x00009888" value="0x03050000" />
+        <register type="NOA" address="0x00009888" value="0x01068000" />
+        <register type="NOA" address="0x00009888" value="0x03084000" />
+        <register type="NOA" address="0x00009888" value="0x030A4000" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x00082AAA" />
+        <register type="NOA" address="0x00009888" value="0x02088000" />
+        <register type="NOA" address="0x00009888" value="0x00032492" />
+        <register type="NOA" address="0x00009888" value="0x02030012" />
+        <register type="NOA" address="0x00009888" value="0x06030002" />
+        <register type="NOA" address="0x00009888" value="0x14033000" />
+        <register type="NOA" address="0x00009888" value="0x3E030008" />
+        <register type="NOA" address="0x00009888" value="0x26033000" />
+        <register type="NOA" address="0x00009888" value="0x42032100" />
+        <register type="NOA" address="0x00009888" value="0x2C033000" />
+        <register type="NOA" address="0x00009888" value="0x32033000" />
+        <register type="NOA" address="0x00009888" value="0x44030CA5" />
+        <register type="NOA" address="0x00009888" value="0x38033000" />
+        <register type="NOA" address="0x00009888" value="0x46030462" />
+        <register type="NOA" address="0x00009888" value="0x20030000" />
+        <register type="NOA" address="0x00009884" value="0x00000008" />
+        <register type="NOA" address="0x00009888" value="0x0001807F" />
+        <register type="NOA" address="0x00009888" value="0x00021555" />
+        <register type="NOA" address="0x00009888" value="0x02024000" />
+        <register type="NOA" address="0x00009888" value="0x00041555" />
+        <register type="NOA" address="0x00009888" value="0x02044000" />
+        <register type="NOA" address="0x00009888" value="0x00051555" />
+        <register type="NOA" address="0x00009888" value="0x02054000" />
+        <register type="NOA" address="0x00009888" value="0x00082AAA" />
+        <register type="NOA" address="0x00009888" value="0x02088000" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x70800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00070000" />
+        <register type="OA" address="0x0000D940" value="0x00000038" />
+        <register type="OA" address="0x0000D944" value="0x0000FFF8" />
+        <register type="OA" address="0x0000DC00" value="0x00000038" />
+        <register type="OA" address="0x0000DC04" value="0x0000FFF8" />
+        <register type="OA" address="0x0000D948" value="0x000000C0" />
+        <register type="OA" address="0x0000D94C" value="0x0000FFE7" />
+        <register type="OA" address="0x0000DC08" value="0x000000C0" />
+        <register type="OA" address="0x0000DC0C" value="0x0000FFE7" />
+        <register type="OA" address="0x0000D950" value="0x00000300" />
+        <register type="OA" address="0x0000D954" value="0x0000FF9F" />
+        <register type="OA" address="0x0000DC10" value="0x00000300" />
+        <register type="OA" address="0x0000DC14" value="0x0000FF9F" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="Metric set RasterizerAndPixelBackend2"
+       chipset="DG2"
+       symbol_name="RasterizerAndPixelBackend2"
+       underscore_name="rasterizer_and_pixel_backend2"
+       mdapi_supported_apis="VK OGL OCL IO"
+       hw_config_guid="57a0abf4-8d81-4856-ab86-01f15f92c81c"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Rasterized Pixels"
+             symbol_name="RasterizedPixels"
+             underscore_name="rasterized_pixels"
+             description="The total number of rasterized pixels."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 21 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Early Hi-Depth Test Fails"
+             symbol_name="HiDepthTestFails"
+             underscore_name="hi_depth_test_fails"
+             description="The total number of pixels dropped on early hierarchical depth test."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 22 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer/Hi-Depth Test"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Early Depth Test Fails"
+             symbol_name="EarlyDepthTestFails"
+             underscore_name="early_depth_test_fails"
+             description="The total number of pixels dropped on early depth test."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 23 READ 4 UMUL"
+             mdapi_group="3D Pipe/Rasterizer/Early Depth Test"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Killed in FS"
+             symbol_name="SamplesKilledInPs"
+             underscore_name="samples_killed_in_ps"
+             description="The total number of samples or pixels dropped in fragment shaders."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 24 READ 4 UMUL"
+             mdapi_group="3D Pipe/Fragment Shader"
+             mdapi_usage_flags="Tier4 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Pixels Failing Tests"
+             symbol_name="PixelsFailingPostPsTests"
+             underscore_name="pixels_failing_post_ps_tests"
+             description="The total number of pixels dropped on post-FS alpha, stencil, or depth tests."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 25 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Written"
+             symbol_name="SamplesWritten"
+             underscore_name="samples_written"
+             description="The total number of samples or pixels written to all render targets."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 26 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Samples Blended"
+             symbol_name="SamplesBlended"
+             underscore_name="samples_blended"
+             description="The total number of blended samples or pixels written to all render targets."
+             data_type="uint64"
+             units="pixels"
+             semantic_type="event"
+             equation="A 27 READ 4 UMUL"
+             mdapi_group="3D Pipe/Output Merger"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Texels"
+             symbol_name="SamplerTexels"
+             underscore_name="sampler_texels"
+             description="The total number of texels seen on input (with 2x2 accuracy) in all sampler units."
+             data_type="uint64"
+             units="texels"
+             semantic_type="event"
+             equation="A 28 READ 4 UMUL"
+             mdapi_group="Sampler/Sampler Input"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Sampler Texels Misses"
+             symbol_name="SamplerTexelMisses"
+             underscore_name="sampler_texel_misses"
+             description="The total number of texels lookups (with 2x2 accuracy) that missed L1 sampler cache."
+             data_type="uint64"
+             units="texels"
+             semantic_type="event"
+             equation="A 29 READ 4 UMUL"
+             mdapi_group="Sampler/Sampler Cache"
+             mdapi_usage_flags="Tier3 Batch Frame Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SLM Bytes Read"
+             symbol_name="SlmBytesRead"
+             underscore_name="slm_bytes_read"
+             description="The total number of GPU memory bytes read from shared local memory."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="A 30 READ 64 UMUL"
+             mdapi_group="L3/Data Port/SLM"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SLM Bytes Written"
+             symbol_name="SlmBytesWritten"
+             underscore_name="slm_bytes_written"
+             description="The total number of GPU memory bytes written into shared local memory."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="A 31 READ 64 UMUL"
+             mdapi_group="L3/Data Port/SLM"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Memory Accesses"
+             symbol_name="ShaderMemoryAccesses"
+             underscore_name="shader_memory_accesses"
+             description="The total number of shader memory accesses to L3."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 32 READ"
+             mdapi_group="L3/Data Port"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Atomic Memory Accesses"
+             symbol_name="ShaderAtomics"
+             underscore_name="shader_atomics"
+             description="The total number of shader atomic memory accesses."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 34 READ"
+             mdapi_group="L3/Data Port/Atomics"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="L3 Shader Throughput"
+             symbol_name="L3ShaderThroughput"
+             underscore_name="l3_shader_throughput"
+             description="The total number of GPU memory bytes transferred between shaders and L3 caches w/o URB."
+             data_type="uint64"
+             max_equation="$GpuCoreClocks 64 UMUL $EuSubslicesTotalCount UMUL"
+             units="bytes"
+             semantic_type="throughput"
+             equation="$ShaderMemoryAccesses 64 UMUL"
+             mdapi_group="L3/Data Port"
+             mdapi_usage_flags="Tier2 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Shader Barrier Messages"
+             symbol_name="ShaderBarriers"
+             underscore_name="shader_barriers"
+             description="The total number of shader barrier messages."
+             data_type="uint64"
+             units="messages"
+             semantic_type="event"
+             equation="A 35 READ"
+             mdapi_group="EU Array/Barrier"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Slice1 Rasterizer Input Available"
+             symbol_name="Rasterizer1InputAvailable"
+             underscore_name="rasterizer1_input_available"
+             description="The percentage of time in which slice1 rasterizer input is available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$SliceMask 2 AND"
+             mdapi_group="GPU/Rasterizer"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Slice0 Rasterizer Output Ready"
+             symbol_name="Rasterizer0OutputReady"
+             underscore_name="rasterizer0_output_ready"
+             description="The percentage of time in which slice0 rasterizer output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$SliceMask 1 AND"
+             mdapi_group="GPU/Rasterizer"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Slice1 Pipe0 Post-EarlyZ Pixel Data Ready"
+             symbol_name="PixelData10Ready"
+             underscore_name="pixel_data10_ready"
+             description="The percentage of time in which slice1  pipe0 post-EarlyZ pixel data is ready (after early Z tests have been applied)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU/Rasterizer/Early Depth Test"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Slice1 Pipe1 Post-EarlyZ Pixel Data Ready"
+             symbol_name="PixelData11Ready"
+             underscore_name="pixel_data11_ready"
+             description="The percentage of time in which slice1  pipe1 post-EarlyZ pixel data is ready (after early Z tests have been applied)"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU/Rasterizer/Early Depth Test"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x0000000C" />
+        <register type="NOA" address="0x00009888" value="0x1A057000" />
+        <register type="NOA" address="0x00009888" value="0x1B056000" />
+        <register type="NOA" address="0x00009888" value="0x1D050000" />
+        <register type="NOA" address="0x00009888" value="0x0D0B701F" />
+        <register type="NOA" address="0x00009888" value="0x0F0B6000" />
+        <register type="NOA" address="0x00009888" value="0x0D0C7000" />
+        <register type="NOA" address="0x00009888" value="0x0F0C6000" />
+        <register type="NOA" address="0x00009888" value="0x0003007F" />
+        <register type="NOA" address="0x00009888" value="0x00050000" />
+        <register type="NOA" address="0x00009888" value="0x10050DC0" />
+        <register type="NOA" address="0x00009888" value="0x02050000" />
+        <register type="NOA" address="0x00009888" value="0x00068000" />
+        <register type="NOA" address="0x00009888" value="0x00083FFF" />
+        <register type="NOA" address="0x00009888" value="0x02084000" />
+        <register type="NOA" address="0x00009888" value="0x000A1555" />
+        <register type="NOA" address="0x00009888" value="0x020A4000" />
+        <register type="NOA" address="0x00009888" value="0x0001007F" />
+        <register type="NOA" address="0x00009888" value="0x000B2AAA" />
+        <register type="NOA" address="0x00009888" value="0x01050000" />
+        <register type="NOA" address="0x00009888" value="0x030535D2" />
+        <register type="NOA" address="0x00009888" value="0x0505003E" />
+        <register type="NOA" address="0x00009888" value="0x01060007" />
+        <register type="NOA" address="0x00009888" value="0x01070280" />
+        <register type="NOA" address="0x00009888" value="0x01083C15" />
+        <register type="NOA" address="0x00009888" value="0x01090018" />
+        <register type="NOA" address="0x00009888" value="0x010A1695" />
+        <register type="NOA" address="0x00009888" value="0x01010060" />
+        <register type="NOA" address="0x00009888" value="0x050B0200" />
+        <register type="NOA" address="0x00009888" value="0x0B0B0400" />
+        <register type="NOA" address="0x00009888" value="0x090B0000" />
+        <register type="NOA" address="0x00009888" value="0x010B0000" />
+        <register type="NOA" address="0x00009888" value="0x01020018" />
+        <register type="NOA" address="0x00009888" value="0x050C0200" />
+        <register type="NOA" address="0x00009888" value="0x090C0400" />
+        <register type="NOA" address="0x00009888" value="0x0B0C001F" />
+        <register type="NOA" address="0x00009888" value="0x010C0000" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x00082AAA" />
+        <register type="NOA" address="0x00009888" value="0x02088000" />
+        <register type="NOA" address="0x00009888" value="0x00032492" />
+        <register type="NOA" address="0x00009888" value="0x02030012" />
+        <register type="NOA" address="0x00009888" value="0x06030002" />
+        <register type="NOA" address="0x00009888" value="0x14030000" />
+        <register type="NOA" address="0x00009888" value="0x3E030006" />
+        <register type="NOA" address="0x00009888" value="0x26030000" />
+        <register type="NOA" address="0x00009888" value="0x420318C3" />
+        <register type="NOA" address="0x00009888" value="0x2C030000" />
+        <register type="NOA" address="0x00009888" value="0x32030000" />
+        <register type="NOA" address="0x00009888" value="0x44030000" />
+        <register type="NOA" address="0x00009888" value="0x38030000" />
+        <register type="NOA" address="0x00009888" value="0x46030603" />
+        <register type="NOA" address="0x00009888" value="0x20037000" />
+        <register type="NOA" address="0x00009884" value="0x00000008" />
+        <register type="NOA" address="0x00009888" value="0x0001807F" />
+        <register type="NOA" address="0x00009888" value="0x00021555" />
+        <register type="NOA" address="0x00009888" value="0x02024000" />
+        <register type="NOA" address="0x00009888" value="0x00041555" />
+        <register type="NOA" address="0x00009888" value="0x02044000" />
+        <register type="NOA" address="0x00009888" value="0x00051555" />
+        <register type="NOA" address="0x00009888" value="0x02054000" />
+        <register type="NOA" address="0x00009888" value="0x00082AAA" />
+        <register type="NOA" address="0x00009888" value="0x02088000" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x70800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00070000" />
+        <register type="OA" address="0x0000D940" value="0x00000038" />
+        <register type="OA" address="0x0000D944" value="0x0000FFF8" />
+        <register type="OA" address="0x0000DC00" value="0x00000038" />
+        <register type="OA" address="0x0000DC04" value="0x0000FFF8" />
+        <register type="OA" address="0x0000D948" value="0x000000C0" />
+        <register type="OA" address="0x0000D94C" value="0x0000FFE7" />
+        <register type="OA" address="0x0000DC08" value="0x000000C0" />
+        <register type="OA" address="0x0000DC0C" value="0x0000FFE7" />
+        <register type="OA" address="0x0000D950" value="0x00000300" />
+        <register type="OA" address="0x0000D954" value="0x0000FF9F" />
+        <register type="OA" address="0x0000DC10" value="0x00000300" />
+        <register type="OA" address="0x0000DC14" value="0x0000FF9F" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="Sampler_1"
+       chipset="DG2"
+       symbol_name="Sampler_1"
+       underscore_name="sampler_1"
+       mdapi_supported_apis="VK OGL OCL MEDIA IO"
+       hw_config_guid="2314eaa3-164f-4004-9684-3f1d16e875f9"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Slice0 DualSubslice0 Input Available"
+             symbol_name="Sampler00InputAvailable"
+             underscore_name="sampler00_input_available"
+             description="The percentage of time in which slice0 dualsubslice0 sampler input is available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 1 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Slice0 DualSubslice1 Input Available"
+             symbol_name="Sampler01InputAvailable"
+             underscore_name="sampler01_input_available"
+             description="The percentage of time in which slice0 dualsubslice1 sampler input is available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 2 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Slice0 DualSubslice2 Input Available"
+             symbol_name="Sampler02InputAvailable"
+             underscore_name="sampler02_input_available"
+             description="The percentage of time in which slice0 dualsubslice2 sampler input is available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 4 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Slice0 DualSubslice3 Input Available"
+             symbol_name="Sampler03InputAvailable"
+             underscore_name="sampler03_input_available"
+             description="The percentage of time in which slice0 dualsubslice3 sampler input is available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 8 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Slice0 DualSubslice0 Sampler Output Ready"
+             symbol_name="Sampler00OutputReady"
+             underscore_name="sampler00_output_ready"
+             description="The percentage of time in which slice0 dualsubslice0 sampler output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 1 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Slice0 DualSubslice1 Sampler Output Ready"
+             symbol_name="Sampler01OutputReady"
+             underscore_name="sampler01_output_ready"
+             description="The percentage of time in which slice0 dualsubslice1 sampler output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 2 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Slice0 DualSubslice2 Sampler Output Ready"
+             symbol_name="Sampler02OutputReady"
+             underscore_name="sampler02_output_ready"
+             description="The percentage of time in which slice0 dualsubslice2 sampler output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 4 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Slice0 DualSubslice3 Sampler Output Ready"
+             symbol_name="Sampler03OutputReady"
+             underscore_name="sampler03_output_ready"
+             description="The percentage of time in which slice0 dualsubslice3 sampler output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 8 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x0000000C" />
+        <register type="NOA" address="0x00009888" value="0x1C824800" />
+        <register type="NOA" address="0x00009888" value="0x14832C00" />
+        <register type="NOA" address="0x00009888" value="0x1CA24800" />
+        <register type="NOA" address="0x00009888" value="0x14A32C00" />
+        <register type="NOA" address="0x00009888" value="0x1CC24800" />
+        <register type="NOA" address="0x00009888" value="0x14C32C00" />
+        <register type="NOA" address="0x00009888" value="0x1CE24800" />
+        <register type="NOA" address="0x00009888" value="0x14E32C00" />
+        <register type="NOA" address="0x00009888" value="0x04824000" />
+        <register type="NOA" address="0x00009888" value="0x16820027" />
+        <register type="NOA" address="0x00009888" value="0x02820000" />
+        <register type="NOA" address="0x00009888" value="0x0E830580" />
+        <register type="NOA" address="0x00009888" value="0x04830000" />
+        <register type="NOA" address="0x00009888" value="0x02834000" />
+        <register type="NOA" address="0x00009888" value="0x00848800" />
+        <register type="NOA" address="0x00009888" value="0x02864040" />
+        <register type="NOA" address="0x00009888" value="0x04A24000" />
+        <register type="NOA" address="0x00009888" value="0x14A209C0" />
+        <register type="NOA" address="0x00009888" value="0x02A20000" />
+        <register type="NOA" address="0x00009888" value="0x0EA30016" />
+        <register type="NOA" address="0x00009888" value="0x04A30000" />
+        <register type="NOA" address="0x00009888" value="0x02A31000" />
+        <register type="NOA" address="0x00009888" value="0x00A44400" />
+        <register type="NOA" address="0x00009888" value="0x02A61010" />
+        <register type="NOA" address="0x00009888" value="0x04C24000" />
+        <register type="NOA" address="0x00009888" value="0x14C20027" />
+        <register type="NOA" address="0x00009888" value="0x02C20000" />
+        <register type="NOA" address="0x00009888" value="0x0CC30580" />
+        <register type="NOA" address="0x00009888" value="0x04C30000" />
+        <register type="NOA" address="0x00009888" value="0x02C30400" />
+        <register type="NOA" address="0x00009888" value="0x00C42200" />
+        <register type="NOA" address="0x00009888" value="0x02C60404" />
+        <register type="NOA" address="0x00009888" value="0x04E24000" />
+        <register type="NOA" address="0x00009888" value="0x12E209C0" />
+        <register type="NOA" address="0x00009888" value="0x02E20000" />
+        <register type="NOA" address="0x00009888" value="0x0CE30016" />
+        <register type="NOA" address="0x00009888" value="0x04E30000" />
+        <register type="NOA" address="0x00009888" value="0x02E30100" />
+        <register type="NOA" address="0x00009888" value="0x00E41100" />
+        <register type="NOA" address="0x00009888" value="0x02E60101" />
+        <register type="NOA" address="0x00009888" value="0x02070909" />
+        <register type="NOA" address="0x00009888" value="0x0208B0B0" />
+        <register type="NOA" address="0x00009888" value="0x00093300" />
+        <register type="NOA" address="0x00009888" value="0x020A5A5A" />
+        <register type="NOA" address="0x00009888" value="0x00014400" />
+        <register type="NOA" address="0x00009888" value="0x020B1010" />
+        <register type="NOA" address="0x00009888" value="0x00022200" />
+        <register type="NOA" address="0x00009888" value="0x020C0404" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009888" value="0x02032400" />
+        <register type="NOA" address="0x00009888" value="0x04032492" />
+        <register type="NOA" address="0x00009888" value="0x06030002" />
+        <register type="NOA" address="0x00009888" value="0x48030200" />
+        <register type="NOA" address="0x00009888" value="0x4A030803" />
+        <register type="NOA" address="0x00009888" value="0x16030000" />
+        <register type="NOA" address="0x00009888" value="0x3E0318E0" />
+        <register type="NOA" address="0x00009888" value="0x18031000" />
+        <register type="NOA" address="0x00009888" value="0x1A030000" />
+        <register type="NOA" address="0x00009888" value="0x40030C0A" />
+        <register type="NOA" address="0x00009888" value="0x1C030000" />
+        <register type="NOA" address="0x00009888" value="0x1E030000" />
+        <register type="NOA" address="0x00009888" value="0x20031000" />
+        <register type="NOA" address="0x00009888" value="0x42030002" />
+        <register type="NOA" address="0x00009884" value="0x00000008" />
+        <register type="NOA" address="0x00009888" value="0x0001FF00" />
+        <register type="NOA" address="0x00009888" value="0x02025555" />
+        <register type="NOA" address="0x00009888" value="0x02045555" />
+        <register type="NOA" address="0x00009888" value="0x02055555" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x00800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00000000" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="Sampler_2"
+       chipset="DG2"
+       symbol_name="Sampler_2"
+       underscore_name="sampler_2"
+       mdapi_supported_apis="VK OGL OCL MEDIA IO"
+       hw_config_guid="c60dea82-db13-4882-8d31-a02bce151ef7"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Slice1 DualSubslice0 Input Available"
+             symbol_name="Sampler10InputAvailable"
+             underscore_name="sampler10_input_available"
+             description="The percentage of time in which slice1 dualsubslice0 sampler input is available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 64 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Slice1 DualSubslice1 Input Available"
+             symbol_name="Sampler11InputAvailable"
+             underscore_name="sampler11_input_available"
+             description="The percentage of time in which slice1 dualsubslice1 sampler input is available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 128 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Slice1 DualSubslice2 Input Available"
+             symbol_name="Sampler12InputAvailable"
+             underscore_name="sampler12_input_available"
+             description="The percentage of time in which slice1 dualsubslice2 sampler input is available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 256 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Slice1 DualSubslice3 Input Available"
+             symbol_name="Sampler13InputAvailable"
+             underscore_name="sampler13_input_available"
+             description="The percentage of time in which slice1 dualsubslice3 sampler input is available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 512 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Slice1 DualSubslice0 Sampler Output Ready"
+             symbol_name="Sampler10OutputReady"
+             underscore_name="sampler10_output_ready"
+             description="The percentage of time in which slice1 dualsubslice0 sampler output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 64 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Slice1 DualSubslice1 Sampler Output Ready"
+             symbol_name="Sampler11OutputReady"
+             underscore_name="sampler11_output_ready"
+             description="The percentage of time in which slice1 dualsubslice1 sampler output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 128 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Slice1 DualSubslice2 Sampler Output Ready"
+             symbol_name="Sampler12OutputReady"
+             underscore_name="sampler12_output_ready"
+             description="The percentage of time in which slice1 dualsubslice2 sampler output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 256 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <counter name="Slice1 DualSubslice3 Sampler Output Ready"
+             symbol_name="Sampler13OutputReady"
+             underscore_name="sampler13_output_ready"
+             description="The percentage of time in which slice1 dualsubslice3 sampler output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 512 AND"
+             mdapi_group="GPU/Sampler"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="subslice"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x0000000C" />
+        <register type="NOA" address="0x00009888" value="0x1D824800" />
+        <register type="NOA" address="0x00009888" value="0x15832C00" />
+        <register type="NOA" address="0x00009888" value="0x1DA24800" />
+        <register type="NOA" address="0x00009888" value="0x15A32C00" />
+        <register type="NOA" address="0x00009888" value="0x1DC24800" />
+        <register type="NOA" address="0x00009888" value="0x15C32C00" />
+        <register type="NOA" address="0x00009888" value="0x1DE24800" />
+        <register type="NOA" address="0x00009888" value="0x15E32C00" />
+        <register type="NOA" address="0x00009888" value="0x05824000" />
+        <register type="NOA" address="0x00009888" value="0x17820027" />
+        <register type="NOA" address="0x00009888" value="0x03820000" />
+        <register type="NOA" address="0x00009888" value="0x0F830580" />
+        <register type="NOA" address="0x00009888" value="0x05830000" />
+        <register type="NOA" address="0x00009888" value="0x03834000" />
+        <register type="NOA" address="0x00009888" value="0x01848800" />
+        <register type="NOA" address="0x00009888" value="0x03864040" />
+        <register type="NOA" address="0x00009888" value="0x05A24000" />
+        <register type="NOA" address="0x00009888" value="0x15A209C0" />
+        <register type="NOA" address="0x00009888" value="0x03A20000" />
+        <register type="NOA" address="0x00009888" value="0x0FA30016" />
+        <register type="NOA" address="0x00009888" value="0x05A30000" />
+        <register type="NOA" address="0x00009888" value="0x03A31000" />
+        <register type="NOA" address="0x00009888" value="0x01A44400" />
+        <register type="NOA" address="0x00009888" value="0x03A61010" />
+        <register type="NOA" address="0x00009888" value="0x05C24000" />
+        <register type="NOA" address="0x00009888" value="0x15C20027" />
+        <register type="NOA" address="0x00009888" value="0x03C20000" />
+        <register type="NOA" address="0x00009888" value="0x0DC30580" />
+        <register type="NOA" address="0x00009888" value="0x05C30000" />
+        <register type="NOA" address="0x00009888" value="0x03C30400" />
+        <register type="NOA" address="0x00009888" value="0x01C42200" />
+        <register type="NOA" address="0x00009888" value="0x03C60404" />
+        <register type="NOA" address="0x00009888" value="0x05E24000" />
+        <register type="NOA" address="0x00009888" value="0x13E209C0" />
+        <register type="NOA" address="0x00009888" value="0x03E20000" />
+        <register type="NOA" address="0x00009888" value="0x0DE30016" />
+        <register type="NOA" address="0x00009888" value="0x05E30000" />
+        <register type="NOA" address="0x00009888" value="0x03E30100" />
+        <register type="NOA" address="0x00009888" value="0x01E41100" />
+        <register type="NOA" address="0x00009888" value="0x03E60101" />
+        <register type="NOA" address="0x00009888" value="0x0003FF00" />
+        <register type="NOA" address="0x00009888" value="0x0208FFFF" />
+        <register type="NOA" address="0x00009888" value="0x020A5555" />
+        <register type="NOA" address="0x00009888" value="0x0001FF00" />
+        <register type="NOA" address="0x00009888" value="0x020BAAAA" />
+        <register type="NOA" address="0x00009888" value="0x03070909" />
+        <register type="NOA" address="0x00009888" value="0x0308B0B0" />
+        <register type="NOA" address="0x00009888" value="0x01093300" />
+        <register type="NOA" address="0x00009888" value="0x030A5A5A" />
+        <register type="NOA" address="0x00009888" value="0x01014400" />
+        <register type="NOA" address="0x00009888" value="0x030B1010" />
+        <register type="NOA" address="0x00009888" value="0x01022200" />
+        <register type="NOA" address="0x00009888" value="0x030C0404" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009888" value="0x02032400" />
+        <register type="NOA" address="0x00009888" value="0x04032492" />
+        <register type="NOA" address="0x00009888" value="0x06030002" />
+        <register type="NOA" address="0x00009888" value="0x48030200" />
+        <register type="NOA" address="0x00009888" value="0x4A030803" />
+        <register type="NOA" address="0x00009888" value="0x16030000" />
+        <register type="NOA" address="0x00009888" value="0x3E0318E0" />
+        <register type="NOA" address="0x00009888" value="0x18031000" />
+        <register type="NOA" address="0x00009888" value="0x1A030000" />
+        <register type="NOA" address="0x00009888" value="0x40030C0A" />
+        <register type="NOA" address="0x00009888" value="0x1C030000" />
+        <register type="NOA" address="0x00009888" value="0x1E030000" />
+        <register type="NOA" address="0x00009888" value="0x20031000" />
+        <register type="NOA" address="0x00009888" value="0x42030002" />
+        <register type="NOA" address="0x00009884" value="0x00000008" />
+        <register type="NOA" address="0x00009888" value="0x0001FF00" />
+        <register type="NOA" address="0x00009888" value="0x02025555" />
+        <register type="NOA" address="0x00009888" value="0x02045555" />
+        <register type="NOA" address="0x00009888" value="0x02055555" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x00800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00000000" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="TDL_1"
+       chipset="DG2"
+       symbol_name="TDL_1"
+       underscore_name="tdl_1"
+       mdapi_supported_apis="VK OGL OCL MEDIA IO"
+       hw_config_guid="7f5cb8ec-38bb-48c9-9ad2-60075eba2833"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Non-PS Thread Ready For Dispatch on Slice0 DualSubslice0 Thread Dispatcher"
+             symbol_name="NonPSThread00ReadyForDispatch"
+             underscore_name="non_ps_thread00_ready_for_dispatch"
+             description="The percentage of time in which non-PS thread is ready for dispatch on slice0 dualsubslice0 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 1 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Non-PS Thread Ready For Dispatch on Slice0 DualSubslice1 Thread Dispatcher"
+             symbol_name="NonPSThread01ReadyForDispatch"
+             underscore_name="non_ps_thread01_ready_for_dispatch"
+             description="The percentage of time in which non-PS thread is ready for dispatch on slice0 dualsubslice1 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 2 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Non-PS Thread Ready For Dispatch on Slice0 DualSubslice2 Thread Dispatcher"
+             symbol_name="NonPSThread02ReadyForDispatch"
+             underscore_name="non_ps_thread02_ready_for_dispatch"
+             description="The percentage of time in which non-PS thread is ready for dispatch on slice0 dualsubslice2 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 4 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Non-PS Thread Ready For Dispatch on Slice0 DualSubslice3 Thread Dispatcher"
+             symbol_name="NonPSThread03ReadyForDispatch"
+             underscore_name="non_ps_thread03_ready_for_dispatch"
+             description="The percentage of time in which non-PS thread is ready for dispatch on slice0 dualsubslice3 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 8 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Non-PS Thread Ready For Dispatch on Slice1 DualSubslice0 Thread Dispatcher"
+             symbol_name="NonPSThread10ReadyForDispatch"
+             underscore_name="non_ps_thread10_ready_for_dispatch"
+             description="The percentage of time in which non-PS thread is ready for dispatch on slice1 dualsubslice0 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 64 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Non-PS Thread Ready For Dispatch on Slice1 DualSubslice1 Thread Dispatcher"
+             symbol_name="NonPSThread11ReadyForDispatch"
+             underscore_name="non_ps_thread11_ready_for_dispatch"
+             description="The percentage of time in which non-PS thread is ready for dispatch on slice1 dualsubslice1 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 128 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Non-PS Thread Ready For Dispatch on Slice1 DualSubslice2 Thread Dispatcher"
+             symbol_name="NonPSThread12ReadyForDispatch"
+             underscore_name="non_ps_thread12_ready_for_dispatch"
+             description="The percentage of time in which non-PS thread is ready for dispatch on slice1 dualsubslice2 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 256 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Non-PS Thread Ready For Dispatch on Slice1 DualSubslice3 Thread Dispatcher"
+             symbol_name="NonPSThread13ReadyForDispatch"
+             underscore_name="non_ps_thread13_ready_for_dispatch"
+             description="The percentage of time in which non-PS thread is ready for dispatch on slice1 dualsubslice3 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 512 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice3 Thread Dispatcher Port 0"
+             symbol_name="ThreadHeader03ReadyPort0"
+             underscore_name="thread_header03_ready_port0"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice3 thread dispatcher port 0"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 8 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice3 Thread Dispatcher Port 1"
+             symbol_name="ThreadHeader03ReadyPort1"
+             underscore_name="thread_header03_ready_port1"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice3 thread dispatcher port 1"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 8 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice3 Thread Dispatcher Port 2"
+             symbol_name="ThreadHeader03ReadyPort2"
+             underscore_name="thread_header03_ready_port2"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice3 thread dispatcher port 2"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 8 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice3 Thread Dispatcher Port 3"
+             symbol_name="ThreadHeader03ReadyPort3"
+             underscore_name="thread_header03_ready_port3"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice3 thread dispatcher port 3"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 8 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice3 Thread Dispatcher Port 0"
+             symbol_name="ThreadHeader13ReadyPort0"
+             underscore_name="thread_header13_ready_port0"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice3 thread dispatcher port 0"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 512 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice3 Thread Dispatcher Port 1"
+             symbol_name="ThreadHeader13ReadyPort1"
+             underscore_name="thread_header13_ready_port1"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice3 thread dispatcher port 1"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 512 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice3 Thread Dispatcher Port 2"
+             symbol_name="ThreadHeader13ReadyPort2"
+             underscore_name="thread_header13_ready_port2"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice3 thread dispatcher port 2"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 512 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice3 Thread Dispatcher Port 3"
+             symbol_name="ThreadHeader13ReadyPort3"
+             underscore_name="thread_header13_ready_port3"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice3 thread dispatcher port 3"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 512 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x0000000C" />
+        <register type="NOA" address="0x00009888" value="0x2A81001C" />
+        <register type="NOA" address="0x00009888" value="0x2AA1001C" />
+        <register type="NOA" address="0x00009888" value="0x2AC1001C" />
+        <register type="NOA" address="0x00009888" value="0x2AE1001C" />
+        <register type="NOA" address="0x00009888" value="0x2B81001C" />
+        <register type="NOA" address="0x00009888" value="0x2BA1001C" />
+        <register type="NOA" address="0x00009888" value="0x2BC1001C" />
+        <register type="NOA" address="0x00009888" value="0x2BE1001C" />
+        <register type="NOA" address="0x00009888" value="0x06810300" />
+        <register type="NOA" address="0x00009888" value="0x08810030" />
+        <register type="NOA" address="0x00009888" value="0x00810000" />
+        <register type="NOA" address="0x00009888" value="0x00830002" />
+        <register type="NOA" address="0x00009888" value="0x00840001" />
+        <register type="NOA" address="0x00009888" value="0x00860001" />
+        <register type="NOA" address="0x00009888" value="0x00870001" />
+        <register type="NOA" address="0x00009888" value="0x06A10300" />
+        <register type="NOA" address="0x00009888" value="0x08A10C00" />
+        <register type="NOA" address="0x00009888" value="0x00A10000" />
+        <register type="NOA" address="0x00009888" value="0x00A30008" />
+        <register type="NOA" address="0x00009888" value="0x00A40002" />
+        <register type="NOA" address="0x00009888" value="0x00A60004" />
+        <register type="NOA" address="0x00009888" value="0x00A70004" />
+        <register type="NOA" address="0x00009888" value="0x06C10300" />
+        <register type="NOA" address="0x00009888" value="0x0AC10030" />
+        <register type="NOA" address="0x00009888" value="0x08C10000" />
+        <register type="NOA" address="0x00009888" value="0x00C10000" />
+        <register type="NOA" address="0x00009888" value="0x00C30020" />
+        <register type="NOA" address="0x00009888" value="0x00C40004" />
+        <register type="NOA" address="0x00009888" value="0x00C60010" />
+        <register type="NOA" address="0x00009888" value="0x00C70010" />
+        <register type="NOA" address="0x00009888" value="0x06E10300" />
+        <register type="NOA" address="0x00009888" value="0x0AE10C00" />
+        <register type="NOA" address="0x00009888" value="0x14E10CF4" />
+        <register type="NOA" address="0x00009888" value="0x16E10C72" />
+        <register type="NOA" address="0x00009888" value="0x08E10000" />
+        <register type="NOA" address="0x00009888" value="0x00E10000" />
+        <register type="NOA" address="0x00009888" value="0x02E10000" />
+        <register type="NOA" address="0x00009888" value="0x00E30080" />
+        <register type="NOA" address="0x00009888" value="0x02E3AA00" />
+        <register type="NOA" address="0x00009888" value="0x00E4F008" />
+        <register type="NOA" address="0x00009888" value="0x00E60040" />
+        <register type="NOA" address="0x00009888" value="0x02E65500" />
+        <register type="NOA" address="0x00009888" value="0x00E70040" />
+        <register type="NOA" address="0x00009888" value="0x02E75500" />
+        <register type="NOA" address="0x00009888" value="0x07810300" />
+        <register type="NOA" address="0x00009888" value="0x0D810030" />
+        <register type="NOA" address="0x00009888" value="0x09810000" />
+        <register type="NOA" address="0x00009888" value="0x01810000" />
+        <register type="NOA" address="0x00009888" value="0x01830200" />
+        <register type="NOA" address="0x00009888" value="0x01840010" />
+        <register type="NOA" address="0x00009888" value="0x01860100" />
+        <register type="NOA" address="0x00009888" value="0x01870100" />
+        <register type="NOA" address="0x00009888" value="0x07A10300" />
+        <register type="NOA" address="0x00009888" value="0x0DA10C00" />
+        <register type="NOA" address="0x00009888" value="0x09A10000" />
+        <register type="NOA" address="0x00009888" value="0x01A10000" />
+        <register type="NOA" address="0x00009888" value="0x01A30800" />
+        <register type="NOA" address="0x00009888" value="0x01A40020" />
+        <register type="NOA" address="0x00009888" value="0x01A60400" />
+        <register type="NOA" address="0x00009888" value="0x01A70400" />
+        <register type="NOA" address="0x00009888" value="0x07C10300" />
+        <register type="NOA" address="0x00009888" value="0x0FC10030" />
+        <register type="NOA" address="0x00009888" value="0x09C10000" />
+        <register type="NOA" address="0x00009888" value="0x01C10000" />
+        <register type="NOA" address="0x00009888" value="0x01C32000" />
+        <register type="NOA" address="0x00009888" value="0x01C40040" />
+        <register type="NOA" address="0x00009888" value="0x01C61000" />
+        <register type="NOA" address="0x00009888" value="0x01C71000" />
+        <register type="NOA" address="0x00009888" value="0x07E10300" />
+        <register type="NOA" address="0x00009888" value="0x0FE10C00" />
+        <register type="NOA" address="0x00009888" value="0x11E10CF4" />
+        <register type="NOA" address="0x00009888" value="0x13E10C72" />
+        <register type="NOA" address="0x00009888" value="0x09E10000" />
+        <register type="NOA" address="0x00009888" value="0x01E10000" />
+        <register type="NOA" address="0x00009888" value="0x03E10000" />
+        <register type="NOA" address="0x00009888" value="0x01E38000" />
+        <register type="NOA" address="0x00009888" value="0x03E300AA" />
+        <register type="NOA" address="0x00009888" value="0x01E40F80" />
+        <register type="NOA" address="0x00009888" value="0x01E64000" />
+        <register type="NOA" address="0x00009888" value="0x03E60055" />
+        <register type="NOA" address="0x00009888" value="0x01E74000" />
+        <register type="NOA" address="0x00009888" value="0x03E70055" />
+        <register type="NOA" address="0x00009888" value="0x00030FF0" />
+        <register type="NOA" address="0x00009888" value="0x00070060" />
+        <register type="NOA" address="0x00009888" value="0x02075500" />
+        <register type="NOA" address="0x00009888" value="0x0008FF0E" />
+        <register type="NOA" address="0x00009888" value="0x020800FF" />
+        <register type="NOA" address="0x00009888" value="0x0009F00C" />
+        <register type="NOA" address="0x00009888" value="0x000A55A5" />
+        <register type="NOA" address="0x00009888" value="0x020AAA55" />
+        <register type="NOA" address="0x00009888" value="0x00010FF2" />
+        <register type="NOA" address="0x00009888" value="0x000BAA04" />
+        <register type="NOA" address="0x00009888" value="0x020B00AA" />
+        <register type="NOA" address="0x00009888" value="0x00020004" />
+        <register type="NOA" address="0x00009888" value="0x000C0010" />
+        <register type="NOA" address="0x00009888" value="0x01076000" />
+        <register type="NOA" address="0x00009888" value="0x03070055" />
+        <register type="NOA" address="0x00009888" value="0x01080E00" />
+        <register type="NOA" address="0x00009888" value="0x01090FC0" />
+        <register type="NOA" address="0x00009888" value="0x010AA500" />
+        <register type="NOA" address="0x00009888" value="0x030A00AA" />
+        <register type="NOA" address="0x00009888" value="0x01010020" />
+        <register type="NOA" address="0x00009888" value="0x010B0400" />
+        <register type="NOA" address="0x00009888" value="0x01020040" />
+        <register type="NOA" address="0x00009888" value="0x010C1000" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009888" value="0x00032492" />
+        <register type="NOA" address="0x00009888" value="0x02032492" />
+        <register type="NOA" address="0x00009888" value="0x04032492" />
+        <register type="NOA" address="0x00009888" value="0x06030002" />
+        <register type="NOA" address="0x00009888" value="0x14037000" />
+        <register type="NOA" address="0x00009888" value="0x3E03294A" />
+        <register type="NOA" address="0x00009888" value="0x26037000" />
+        <register type="NOA" address="0x00009888" value="0x42032007" />
+        <register type="NOA" address="0x00009888" value="0x2C033000" />
+        <register type="NOA" address="0x00009888" value="0x32037000" />
+        <register type="NOA" address="0x00009888" value="0x44030047" />
+        <register type="NOA" address="0x00009888" value="0x38031000" />
+        <register type="NOA" address="0x00009888" value="0x46030003" />
+        <register type="NOA" address="0x00009888" value="0x480300A0" />
+        <register type="NOA" address="0x00009888" value="0x4A03140A" />
+        <register type="NOA" address="0x00009888" value="0x16030000" />
+        <register type="NOA" address="0x00009888" value="0x18030000" />
+        <register type="NOA" address="0x00009888" value="0x1A037000" />
+        <register type="NOA" address="0x00009888" value="0x40031CE7" />
+        <register type="NOA" address="0x00009888" value="0x1C037000" />
+        <register type="NOA" address="0x00009888" value="0x1E037000" />
+        <register type="NOA" address="0x00009888" value="0x20037000" />
+        <register type="NOA" address="0x00009884" value="0x00000008" />
+        <register type="NOA" address="0x00009888" value="0x0001FFFF" />
+        <register type="NOA" address="0x00009888" value="0x00025555" />
+        <register type="NOA" address="0x00009888" value="0x02025555" />
+        <register type="NOA" address="0x00009888" value="0x00045555" />
+        <register type="NOA" address="0x00009888" value="0x02045555" />
+        <register type="NOA" address="0x00009888" value="0x00055555" />
+        <register type="NOA" address="0x00009888" value="0x02055555" />
+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0xF0800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0xF0800000" />
+        <register type="OA" address="0x0000DC40" value="0x00FF0000" />
+        <register type="OA" address="0x0000D940" value="0x00000002" />
+        <register type="OA" address="0x0000D944" value="0x0000FFFE" />
+        <register type="OA" address="0x0000DC00" value="0x00000002" />
+        <register type="OA" address="0x0000DC04" value="0x0000FFFE" />
+        <register type="OA" address="0x0000D948" value="0x00000002" />
+        <register type="OA" address="0x0000D94C" value="0x0000FFFD" />
+        <register type="OA" address="0x0000DC08" value="0x00000002" />
+        <register type="OA" address="0x0000DC0C" value="0x0000FFFD" />
+        <register type="OA" address="0x0000D950" value="0x00000002" />
+        <register type="OA" address="0x0000D954" value="0x0000FFFB" />
+        <register type="OA" address="0x0000DC10" value="0x00000002" />
+        <register type="OA" address="0x0000DC14" value="0x0000FFFB" />
+        <register type="OA" address="0x0000D958" value="0x00000002" />
+        <register type="OA" address="0x0000D95C" value="0x0000FFF7" />
+        <register type="OA" address="0x0000DC18" value="0x00000002" />
+        <register type="OA" address="0x0000DC1C" value="0x0000FFF7" />
+        <register type="OA" address="0x0000D960" value="0x00000002" />
+        <register type="OA" address="0x0000D964" value="0x0000FFEF" />
+        <register type="OA" address="0x0000DC20" value="0x00000002" />
+        <register type="OA" address="0x0000DC24" value="0x0000FFEF" />
+        <register type="OA" address="0x0000D968" value="0x00000002" />
+        <register type="OA" address="0x0000D96C" value="0x0000FFDF" />
+        <register type="OA" address="0x0000DC28" value="0x00000002" />
+        <register type="OA" address="0x0000DC2C" value="0x0000FFDF" />
+        <register type="OA" address="0x0000D970" value="0x00000002" />
+        <register type="OA" address="0x0000D974" value="0x0000FFBF" />
+        <register type="OA" address="0x0000DC30" value="0x00000002" />
+        <register type="OA" address="0x0000DC34" value="0x0000FFBF" />
+        <register type="OA" address="0x0000D978" value="0x00000002" />
+        <register type="OA" address="0x0000D97C" value="0x0000FF7F" />
+        <register type="OA" address="0x0000DC38" value="0x00000002" />
+        <register type="OA" address="0x0000DC3C" value="0x0000FF7F" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="TDL_2"
+       chipset="DG2"
+       symbol_name="TDL_2"
+       underscore_name="tdl_2"
+       mdapi_supported_apis="VK OGL OCL MEDIA IO"
+       hw_config_guid="141ce507-de6c-428c-95b2-076194d650db"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="PS Thread Ready For Dispatch on Slice0 Dualsubslice0 Thread Dispatcher"
+             symbol_name="PSThread00ReadyForDispatch"
+             underscore_name="ps_thread00_ready_for_dispatch"
+             description="The percentage of time in which PS thread is ready for dispatch on slice0 dualsubslice0 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 1 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="PS Thread Ready For Dispatch on Slice0 Dualsubslice1 Thread Dispatcher"
+             symbol_name="PSThread01ReadyForDispatch"
+             underscore_name="ps_thread01_ready_for_dispatch"
+             description="The percentage of time in which PS thread is ready for dispatch on slice0 dualsubslice1 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 2 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="PS Thread Ready For Dispatch on Slice0 Dualsubslice2 Thread Dispatcher"
+             symbol_name="PSThread02ReadyForDispatch"
+             underscore_name="ps_thread02_ready_for_dispatch"
+             description="The percentage of time in which PS thread is ready for dispatch on slice0 dualsubslice2 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 4 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="PS Thread Ready For Dispatch on Slice0 Dualsubslice3 Thread Dispatcher"
+             symbol_name="PSThread03ReadyForDispatch"
+             underscore_name="ps_thread03_ready_for_dispatch"
+             description="The percentage of time in which PS thread is ready for dispatch on slice0 dualsubslice3 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 8 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="PS Thread Ready For Dispatch on Slice1 Dualsubslice0 Thread Dispatcher"
+             symbol_name="PSThread10ReadyForDispatch"
+             underscore_name="ps_thread10_ready_for_dispatch"
+             description="The percentage of time in which PS thread is ready for dispatch on slice1 dualsubslice0 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 64 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="PS Thread Ready For Dispatch on Slice1 Dualsubslice1 Thread Dispatcher"
+             symbol_name="PSThread11ReadyForDispatch"
+             underscore_name="ps_thread11_ready_for_dispatch"
+             description="The percentage of time in which PS thread is ready for dispatch on slice1 dualsubslice1 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 128 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="PS Thread Ready For Dispatch on Slice1 Dualsubslice2 Thread Dispatcher"
+             symbol_name="PSThread12ReadyForDispatch"
+             underscore_name="ps_thread12_ready_for_dispatch"
+             description="The percentage of time in which PS thread is ready for dispatch on slice1 dualsubslice2 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 256 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="PS Thread Ready For Dispatch on Slice1 Dualsubslice3 Thread Dispatcher"
+             symbol_name="PSThread13ReadyForDispatch"
+             underscore_name="ps_thread13_ready_for_dispatch"
+             description="The percentage of time in which PS thread is ready for dispatch on slice1 dualsubslice3 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 512 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x0000000C" />
+        <register type="NOA" address="0x00009888" value="0x28810340" />
+        <register type="NOA" address="0x00009888" value="0x28A10340" />
+        <register type="NOA" address="0x00009888" value="0x28C10340" />
+        <register type="NOA" address="0x00009888" value="0x28E10340" />
+        <register type="NOA" address="0x00009888" value="0x29810340" />
+        <register type="NOA" address="0x00009888" value="0x29A10340" />
+        <register type="NOA" address="0x00009888" value="0x29C10340" />
+        <register type="NOA" address="0x00009888" value="0x29E10340" />
+        <register type="NOA" address="0x00009888" value="0x06810100" />
+        <register type="NOA" address="0x00009888" value="0x08810C31" />
+        <register type="NOA" address="0x00009888" value="0x00810000" />
+        <register type="NOA" address="0x00009888" value="0x0083000A" />
+        <register type="NOA" address="0x00009888" value="0x00840003" />
+        <register type="NOA" address="0x00009888" value="0x00860005" />
+        <register type="NOA" address="0x00009888" value="0x00870005" />
+        <register type="NOA" address="0x00009888" value="0x06A10100" />
+        <register type="NOA" address="0x00009888" value="0x0AA10C31" />
+        <register type="NOA" address="0x00009888" value="0x08A10000" />
+        <register type="NOA" address="0x00009888" value="0x00A10000" />
+        <register type="NOA" address="0x00009888" value="0x00A300A0" />
+        <register type="NOA" address="0x00009888" value="0x00A4000C" />
+        <register type="NOA" address="0x00009888" value="0x00A60050" />
+        <register type="NOA" address="0x00009888" value="0x00A70050" />
+        <register type="NOA" address="0x00009888" value="0x06C10100" />
+        <register type="NOA" address="0x00009888" value="0x0CC10C31" />
+        <register type="NOA" address="0x00009888" value="0x08C10000" />
+        <register type="NOA" address="0x00009888" value="0x00C10000" />
+        <register type="NOA" address="0x00009888" value="0x00C30A00" />
+        <register type="NOA" address="0x00009888" value="0x00C40030" />
+        <register type="NOA" address="0x00009888" value="0x00C60500" />
+        <register type="NOA" address="0x00009888" value="0x00C70500" />
+        <register type="NOA" address="0x00009888" value="0x06E10100" />
+        <register type="NOA" address="0x00009888" value="0x0EE10C31" />
+        <register type="NOA" address="0x00009888" value="0x08E10000" />
+        <register type="NOA" address="0x00009888" value="0x00E10000" />
+        <register type="NOA" address="0x00009888" value="0x00E3A000" />
+        <register type="NOA" address="0x00009888" value="0x00E400C0" />
+        <register type="NOA" address="0x00009888" value="0x00E65000" />
+        <register type="NOA" address="0x00009888" value="0x00E75000" />
+        <register type="NOA" address="0x00009888" value="0x07810100" />
+        <register type="NOA" address="0x00009888" value="0x11810C31" />
+        <register type="NOA" address="0x00009888" value="0x09810000" />
+        <register type="NOA" address="0x00009888" value="0x03810000" />
+        <register type="NOA" address="0x00009888" value="0x0383000A" />
+        <register type="NOA" address="0x00009888" value="0x01840300" />
+        <register type="NOA" address="0x00009888" value="0x03860005" />
+        <register type="NOA" address="0x00009888" value="0x03870005" />
+        <register type="NOA" address="0x00009888" value="0x07A10100" />
+        <register type="NOA" address="0x00009888" value="0x13A10C31" />
+        <register type="NOA" address="0x00009888" value="0x09A10000" />
+        <register type="NOA" address="0x00009888" value="0x03A10000" />
+        <register type="NOA" address="0x00009888" value="0x03A300A0" />
+        <register type="NOA" address="0x00009888" value="0x01A40C00" />
+        <register type="NOA" address="0x00009888" value="0x03A60050" />
+        <register type="NOA" address="0x00009888" value="0x03A70050" />
+        <register type="NOA" address="0x00009888" value="0x07C10100" />
+        <register type="NOA" address="0x00009888" value="0x15C10C31" />
+        <register type="NOA" address="0x00009888" value="0x09C10000" />
+        <register type="NOA" address="0x00009888" value="0x03C10000" />
+        <register type="NOA" address="0x00009888" value="0x03C30A00" />
+        <register type="NOA" address="0x00009888" value="0x01C43000" />
+        <register type="NOA" address="0x00009888" value="0x03C60500" />
+        <register type="NOA" address="0x00009888" value="0x03C70500" />
+        <register type="NOA" address="0x00009888" value="0x07E10100" />
+        <register type="NOA" address="0x00009888" value="0x17E10C31" />
+        <register type="NOA" address="0x00009888" value="0x09E10000" />
+        <register type="NOA" address="0x00009888" value="0x03E10000" />
+        <register type="NOA" address="0x00009888" value="0x03E3A000" />
+        <register type="NOA" address="0x00009888" value="0x01E4C000" />
+        <register type="NOA" address="0x00009888" value="0x03E65000" />
+        <register type="NOA" address="0x00009888" value="0x03E75000" />
+        <register type="NOA" address="0x00009888" value="0x0003FF00" />
+        <register type="NOA" address="0x00009888" value="0x00075A00" />
+        <register type="NOA" address="0x00009888" value="0x000800FA" />
+        <register type="NOA" address="0x00009888" value="0x0208FFFF" />
+        <register type="NOA" address="0x00009888" value="0x000900F0" />
+        <register type="NOA" address="0x00009888" value="0x000AAA55" />
+        <register type="NOA" address="0x00009888" value="0x020A5555" />
+        <register type="NOA" address="0x00009888" value="0x0001FF0C" />
+        <register type="NOA" address="0x00009888" value="0x000B0050" />
+        <register type="NOA" address="0x00009888" value="0x020BAAAA" />
+        <register type="NOA" address="0x00009888" value="0x00020030" />
+        <register type="NOA" address="0x00009888" value="0x000C0500" />
+        <register type="NOA" address="0x00009888" value="0x03075A00" />
+        <register type="NOA" address="0x00009888" value="0x030800FA" />
+        <register type="NOA" address="0x00009888" value="0x0109F000" />
+        <register type="NOA" address="0x00009888" value="0x030AAA55" />
+        <register type="NOA" address="0x00009888" value="0x01010C00" />
+        <register type="NOA" address="0x00009888" value="0x030B0050" />
+        <register type="NOA" address="0x00009888" value="0x01023000" />
+        <register type="NOA" address="0x00009888" value="0x030C0500" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009888" value="0x00032492" />
+        <register type="NOA" address="0x00009888" value="0x02032492" />
+        <register type="NOA" address="0x00009888" value="0x04032492" />
+        <register type="NOA" address="0x00009888" value="0x06030002" />
+        <register type="NOA" address="0x00009888" value="0x14037000" />
+        <register type="NOA" address="0x00009888" value="0x3E030C6A" />
+        <register type="NOA" address="0x00009888" value="0x26037000" />
+        <register type="NOA" address="0x00009888" value="0x4203014A" />
+        <register type="NOA" address="0x00009888" value="0x2C037000" />
+        <register type="NOA" address="0x00009888" value="0x32037000" />
+        <register type="NOA" address="0x00009888" value="0x44030D00" />
+        <register type="NOA" address="0x00009888" value="0x38033000" />
+        <register type="NOA" address="0x00009888" value="0x46030EE8" />
+        <register type="NOA" address="0x00009888" value="0x48030277" />
+        <register type="NOA" address="0x00009888" value="0x4A030422" />
+        <register type="NOA" address="0x00009888" value="0x16030000" />
+        <register type="NOA" address="0x00009888" value="0x18030000" />
+        <register type="NOA" address="0x00009888" value="0x1A030000" />
+        <register type="NOA" address="0x00009888" value="0x40032800" />
+        <register type="NOA" address="0x00009888" value="0x1C030000" />
+        <register type="NOA" address="0x00009888" value="0x1E030000" />
+        <register type="NOA" address="0x00009888" value="0x20030000" />
+        <register type="NOA" address="0x00009884" value="0x00000008" />
+        <register type="NOA" address="0x00009888" value="0x0001FFFF" />
+        <register type="NOA" address="0x00009888" value="0x00025555" />
+        <register type="NOA" address="0x00009888" value="0x02025555" />
+        <register type="NOA" address="0x00009888" value="0x00045555" />
+        <register type="NOA" address="0x00009888" value="0x02045555" />
+        <register type="NOA" address="0x00009888" value="0x00055555" />
+        <register type="NOA" address="0x00009888" value="0x02055555" />
+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0xF0800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0xF0800000" />
+        <register type="OA" address="0x0000DC40" value="0x00FF0000" />
+        <register type="OA" address="0x0000D940" value="0x00000000" />
+        <register type="OA" address="0x0000D944" value="0x0000FFFC" />
+        <register type="OA" address="0x0000DC00" value="0x00000000" />
+        <register type="OA" address="0x0000DC04" value="0x0000FFFC" />
+        <register type="OA" address="0x0000D948" value="0x00000000" />
+        <register type="OA" address="0x0000D94C" value="0x0000FFF3" />
+        <register type="OA" address="0x0000DC08" value="0x00000000" />
+        <register type="OA" address="0x0000DC0C" value="0x0000FFF3" />
+        <register type="OA" address="0x0000D950" value="0x00000000" />
+        <register type="OA" address="0x0000D954" value="0x0000FFCF" />
+        <register type="OA" address="0x0000DC10" value="0x00000000" />
+        <register type="OA" address="0x0000DC14" value="0x0000FFCF" />
+        <register type="OA" address="0x0000D958" value="0x00000000" />
+        <register type="OA" address="0x0000D95C" value="0x0000FF3F" />
+        <register type="OA" address="0x0000DC18" value="0x00000000" />
+        <register type="OA" address="0x0000DC1C" value="0x0000FF3F" />
+        <register type="OA" address="0x0000D960" value="0x00000000" />
+        <register type="OA" address="0x0000D964" value="0x0000FCFF" />
+        <register type="OA" address="0x0000DC20" value="0x00000000" />
+        <register type="OA" address="0x0000DC24" value="0x0000FCFF" />
+        <register type="OA" address="0x0000D968" value="0x00000000" />
+        <register type="OA" address="0x0000D96C" value="0x0000F3FF" />
+        <register type="OA" address="0x0000DC28" value="0x00000000" />
+        <register type="OA" address="0x0000DC2C" value="0x0000F3FF" />
+        <register type="OA" address="0x0000D970" value="0x00000000" />
+        <register type="OA" address="0x0000D974" value="0x0000CFFF" />
+        <register type="OA" address="0x0000DC30" value="0x00000000" />
+        <register type="OA" address="0x0000DC34" value="0x0000CFFF" />
+        <register type="OA" address="0x0000D978" value="0x00000000" />
+        <register type="OA" address="0x0000D97C" value="0x00003FFF" />
+        <register type="OA" address="0x0000DC38" value="0x00000000" />
+        <register type="OA" address="0x0000DC3C" value="0x00003FFF" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="TDL_3"
+       chipset="DG2"
+       symbol_name="TDL_3"
+       underscore_name="tdl_3"
+       mdapi_supported_apis="VK OGL OCL MEDIA IO"
+       hw_config_guid="54a85b07-9ff2-4a12-bfb6-6bcdd9d4019c"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice0 Thread Dispatcher"
+             symbol_name="ThreadHeader00Ready"
+             underscore_name="thread_header00_ready"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice0 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 1 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice0 Thread Dispatcher Port 0"
+             symbol_name="ThreadHeader00ReadyPort0"
+             underscore_name="thread_header00_ready_port0"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice0 thread dispatcher port 0"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 1 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice0 Thread Dispatcher Port 1"
+             symbol_name="ThreadHeader00ReadyPort1"
+             underscore_name="thread_header00_ready_port1"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice0 thread dispatcher port 1"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 1 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice0 Thread Dispatcher Port 2"
+             symbol_name="ThreadHeader00ReadyPort2"
+             underscore_name="thread_header00_ready_port2"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice0 thread dispatcher port 2"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 1 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice0 Thread Dispatcher Port 3"
+             symbol_name="ThreadHeader00ReadyPort3"
+             underscore_name="thread_header00_ready_port3"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice0 thread dispatcher port 3"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 1 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice1 Thread Dispatcher"
+             symbol_name="ThreadHeader01Ready"
+             underscore_name="thread_header01_ready"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice1 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 2 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice1 Thread Dispatcher Port 0"
+             symbol_name="ThreadHeader01ReadyPort0"
+             underscore_name="thread_header01_ready_port0"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice1 thread dispatcher port 0"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 2 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice1 Thread Dispatcher Port 1"
+             symbol_name="ThreadHeader01ReadyPort1"
+             underscore_name="thread_header01_ready_port1"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice1 thread dispatcher port 1"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 2 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice1 Thread Dispatcher Port 2"
+             symbol_name="ThreadHeader01ReadyPort2"
+             underscore_name="thread_header01_ready_port2"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice1 thread dispatcher port 2"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 2 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice1 Thread Dispatcher Port 3"
+             symbol_name="ThreadHeader01ReadyPort3"
+             underscore_name="thread_header01_ready_port3"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice1 thread dispatcher port 3"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 2 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice2 Thread Dispatcher"
+             symbol_name="ThreadHeader02Ready"
+             underscore_name="thread_header02_ready"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice2 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 4 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice2 Thread Dispatcher Port 0"
+             symbol_name="ThreadHeader02ReadyPort0"
+             underscore_name="thread_header02_ready_port0"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice2 thread dispatcher port 0"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 4 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice2 Thread Dispatcher Port 1"
+             symbol_name="ThreadHeader02ReadyPort1"
+             underscore_name="thread_header02_ready_port1"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice2 thread dispatcher port 1"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 4 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice2 Thread Dispatcher Port 2"
+             symbol_name="ThreadHeader02ReadyPort2"
+             underscore_name="thread_header02_ready_port2"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice2 thread dispatcher port 2"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 4 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice2 Thread Dispatcher Port 3"
+             symbol_name="ThreadHeader02ReadyPort3"
+             underscore_name="thread_header02_ready_port3"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice2 thread dispatcher port 3"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 4 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice0 DualSubslice3 Thread Dispatcher"
+             symbol_name="ThreadHeader03Ready"
+             underscore_name="thread_header03_ready"
+             description="The percentage of time in which thread header is ready on slice0 dualsubslice3 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 8 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x0000000C" />
+        <register type="NOA" address="0x00009888" value="0x2A81001C" />
+        <register type="NOA" address="0x00009888" value="0x2AA1001C" />
+        <register type="NOA" address="0x00009888" value="0x2AC1001C" />
+        <register type="NOA" address="0x00009888" value="0x2AE1001C" />
+        <register type="NOA" address="0x00009888" value="0x06810300" />
+        <register type="NOA" address="0x00009888" value="0x14810CF4" />
+        <register type="NOA" address="0x00009888" value="0x16810C72" />
+        <register type="NOA" address="0x00009888" value="0x08810000" />
+        <register type="NOA" address="0x00009888" value="0x02810000" />
+        <register type="NOA" address="0x00009888" value="0x0283AA00" />
+        <register type="NOA" address="0x00009888" value="0x0084F000" />
+        <register type="NOA" address="0x00009888" value="0x02865500" />
+        <register type="NOA" address="0x00009888" value="0x02875500" />
+        <register type="NOA" address="0x00009888" value="0x06A10300" />
+        <register type="NOA" address="0x00009888" value="0x10A10CF4" />
+        <register type="NOA" address="0x00009888" value="0x12A10C72" />
+        <register type="NOA" address="0x00009888" value="0x08A10000" />
+        <register type="NOA" address="0x00009888" value="0x02A10000" />
+        <register type="NOA" address="0x00009888" value="0x02A300AA" />
+        <register type="NOA" address="0x00009888" value="0x00A40F00" />
+        <register type="NOA" address="0x00009888" value="0x02A60055" />
+        <register type="NOA" address="0x00009888" value="0x02A70055" />
+        <register type="NOA" address="0x00009888" value="0x06C10300" />
+        <register type="NOA" address="0x00009888" value="0x0CC10CF4" />
+        <register type="NOA" address="0x00009888" value="0x0EC10C72" />
+        <register type="NOA" address="0x00009888" value="0x08C10000" />
+        <register type="NOA" address="0x00009888" value="0x00C10000" />
+        <register type="NOA" address="0x00009888" value="0x00C3AA00" />
+        <register type="NOA" address="0x00009888" value="0x00C400F0" />
+        <register type="NOA" address="0x00009888" value="0x00C65500" />
+        <register type="NOA" address="0x00009888" value="0x00C75500" />
+        <register type="NOA" address="0x00009888" value="0x06E10300" />
+        <register type="NOA" address="0x00009888" value="0x08E10CB1" />
+        <register type="NOA" address="0x00009888" value="0x0AE10D33" />
+        <register type="NOA" address="0x00009888" value="0x00E10000" />
+        <register type="NOA" address="0x00009888" value="0x00E300AA" />
+        <register type="NOA" address="0x00009888" value="0x00E4000F" />
+        <register type="NOA" address="0x00009888" value="0x00E60055" />
+        <register type="NOA" address="0x00009888" value="0x00E70055" />
+        <register type="NOA" address="0x00009888" value="0x0007AA55" />
+        <register type="NOA" address="0x00009888" value="0x0208AAFF" />
+        <register type="NOA" address="0x00009888" value="0x000900FF" />
+        <register type="NOA" address="0x00009888" value="0x000AAAAA" />
+        <register type="NOA" address="0x00009888" value="0x020A5555" />
+        <register type="NOA" address="0x00009888" value="0x00010F00" />
+        <register type="NOA" address="0x00009888" value="0x020B0055" />
+        <register type="NOA" address="0x00009888" value="0x000200F0" />
+        <register type="NOA" address="0x00009888" value="0x000C5500" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009888" value="0x00032492" />
+        <register type="NOA" address="0x00009888" value="0x02032492" />
+        <register type="NOA" address="0x00009888" value="0x04032492" />
+        <register type="NOA" address="0x00009888" value="0x06030002" />
+        <register type="NOA" address="0x00009888" value="0x14030000" />
+        <register type="NOA" address="0x00009888" value="0x3E030C6A" />
+        <register type="NOA" address="0x00009888" value="0x26030000" />
+        <register type="NOA" address="0x00009888" value="0x42032942" />
+        <register type="NOA" address="0x00009888" value="0x2C030000" />
+        <register type="NOA" address="0x00009888" value="0x32030000" />
+        <register type="NOA" address="0x00009888" value="0x4403000A" />
+        <register type="NOA" address="0x00009888" value="0x38030000" />
+        <register type="NOA" address="0x00009888" value="0x46030000" />
+        <register type="NOA" address="0x00009888" value="0x48030000" />
+        <register type="NOA" address="0x00009888" value="0x4A030603" />
+        <register type="NOA" address="0x00009888" value="0x16030000" />
+        <register type="NOA" address="0x00009888" value="0x18030000" />
+        <register type="NOA" address="0x00009888" value="0x1A031000" />
+        <register type="NOA" address="0x00009888" value="0x40030842" />
+        <register type="NOA" address="0x00009888" value="0x1C031000" />
+        <register type="NOA" address="0x00009888" value="0x1E031000" />
+        <register type="NOA" address="0x00009888" value="0x20031000" />
+        <register type="NOA" address="0x00009884" value="0x00000008" />
+        <register type="NOA" address="0x00009888" value="0x0001FFFF" />
+        <register type="NOA" address="0x00009888" value="0x00025555" />
+        <register type="NOA" address="0x00009888" value="0x02025555" />
+        <register type="NOA" address="0x00009888" value="0x00045555" />
+        <register type="NOA" address="0x00009888" value="0x02045555" />
+        <register type="NOA" address="0x00009888" value="0x00055555" />
+        <register type="NOA" address="0x00009888" value="0x02055555" />
+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0xF0800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x000F0000" />
+        <register type="OA" address="0x0000D940" value="0x00078000" />
+        <register type="OA" address="0x0000D944" value="0x00000FFF" />
+        <register type="OA" address="0x0000DC00" value="0x00078000" />
+        <register type="OA" address="0x0000DC04" value="0x00000FFF" />
+        <register type="OA" address="0x0000D948" value="0x00007800" />
+        <register type="OA" address="0x0000D94C" value="0x0000F0FF" />
+        <register type="OA" address="0x0000DC08" value="0x00007800" />
+        <register type="OA" address="0x0000DC0C" value="0x0000F0FF" />
+        <register type="OA" address="0x0000D950" value="0x00000780" />
+        <register type="OA" address="0x0000D954" value="0x0000FF0F" />
+        <register type="OA" address="0x0000DC10" value="0x00000780" />
+        <register type="OA" address="0x0000DC14" value="0x0000FF0F" />
+        <register type="OA" address="0x0000D958" value="0x00000078" />
+        <register type="OA" address="0x0000D95C" value="0x0000FFF0" />
+        <register type="OA" address="0x0000DC18" value="0x00000078" />
+        <register type="OA" address="0x0000DC1C" value="0x0000FFF0" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="TDL_4"
+       chipset="DG2"
+       symbol_name="TDL_4"
+       underscore_name="tdl_4"
+       mdapi_supported_apis="VK OGL OCL MEDIA IO"
+       hw_config_guid="651c40e5-195c-49da-b232-a9a0f753d899"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice0 Thread Dispatcher"
+             symbol_name="ThreadHeader10Ready"
+             underscore_name="thread_header10_ready"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice0 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 64 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice0 Thread Dispatcher Port 0"
+             symbol_name="ThreadHeader10ReadyPort0"
+             underscore_name="thread_header10_ready_port0"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice0 thread dispatcher port 0"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 64 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice0 Thread Dispatcher Port 1"
+             symbol_name="ThreadHeader10ReadyPort1"
+             underscore_name="thread_header10_ready_port1"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice0 thread dispatcher port 1"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 64 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice0 Thread Dispatcher Port 2"
+             symbol_name="ThreadHeader10ReadyPort2"
+             underscore_name="thread_header10_ready_port2"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice0 thread dispatcher port 2"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 64 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice0 Thread Dispatcher Port 3"
+             symbol_name="ThreadHeader10ReadyPort3"
+             underscore_name="thread_header10_ready_port3"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice0 thread dispatcher port 3"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 64 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice1 Thread Dispatcher"
+             symbol_name="ThreadHeader11Ready"
+             underscore_name="thread_header11_ready"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice1 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 128 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice1 Thread Dispatcher Port 0"
+             symbol_name="ThreadHeader11ReadyPort0"
+             underscore_name="thread_header11_ready_port0"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice1 thread dispatcher port 0"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 128 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice1 Thread Dispatcher Port 1"
+             symbol_name="ThreadHeader11ReadyPort1"
+             underscore_name="thread_header11_ready_port1"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice1 thread dispatcher port 1"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 128 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice1 Thread Dispatcher Port 2"
+             symbol_name="ThreadHeader11ReadyPort2"
+             underscore_name="thread_header11_ready_port2"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice1 thread dispatcher port 2"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 128 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice1 Thread Dispatcher Port 3"
+             symbol_name="ThreadHeader11ReadyPort3"
+             underscore_name="thread_header11_ready_port3"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice1 thread dispatcher port 3"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 128 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice2 Thread Dispatcher"
+             symbol_name="ThreadHeader12Ready"
+             underscore_name="thread_header12_ready"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice2 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 256 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice2 Thread Dispatcher Port 0"
+             symbol_name="ThreadHeader12ReadyPort0"
+             underscore_name="thread_header12_ready_port0"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice2 thread dispatcher port 0"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 256 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice2 Thread Dispatcher Port 1"
+             symbol_name="ThreadHeader12ReadyPort1"
+             underscore_name="thread_header12_ready_port1"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice2 thread dispatcher port 1"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 256 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice2 Thread Dispatcher Port 2"
+             symbol_name="ThreadHeader12ReadyPort2"
+             underscore_name="thread_header12_ready_port2"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice2 thread dispatcher port 2"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 256 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice2 Thread Dispatcher Port 3"
+             symbol_name="ThreadHeader12ReadyPort3"
+             underscore_name="thread_header12_ready_port3"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice2 thread dispatcher port 3"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 256 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <counter name="Thread Header Ready on Slice1 DualSubslice3 Thread Dispatcher"
+             symbol_name="ThreadHeader13Ready"
+             underscore_name="thread_header13_ready"
+             description="The percentage of time in which thread header is ready on slice1 dualsubslice3 thread dispatcher"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             availability="$DualSubsliceMask 512 AND"
+             mdapi_group="GPU/Thread Dispatcher"
+             mdapi_usage_flags="Tier3 Overview Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="dualsubslice"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x0000000C" />
+        <register type="NOA" address="0x00009888" value="0x2B81001C" />
+        <register type="NOA" address="0x00009888" value="0x2BA1001C" />
+        <register type="NOA" address="0x00009888" value="0x2BC1001C" />
+        <register type="NOA" address="0x00009888" value="0x2BE1001C" />
+        <register type="NOA" address="0x00009888" value="0x07810300" />
+        <register type="NOA" address="0x00009888" value="0x15810CF4" />
+        <register type="NOA" address="0x00009888" value="0x17810C72" />
+        <register type="NOA" address="0x00009888" value="0x09810000" />
+        <register type="NOA" address="0x00009888" value="0x03810000" />
+        <register type="NOA" address="0x00009888" value="0x0383AA00" />
+        <register type="NOA" address="0x00009888" value="0x0184F000" />
+        <register type="NOA" address="0x00009888" value="0x03865500" />
+        <register type="NOA" address="0x00009888" value="0x03875500" />
+        <register type="NOA" address="0x00009888" value="0x07A10300" />
+        <register type="NOA" address="0x00009888" value="0x11A10CF4" />
+        <register type="NOA" address="0x00009888" value="0x13A10C72" />
+        <register type="NOA" address="0x00009888" value="0x09A10000" />
+        <register type="NOA" address="0x00009888" value="0x03A10000" />
+        <register type="NOA" address="0x00009888" value="0x03A300AA" />
+        <register type="NOA" address="0x00009888" value="0x01A40F00" />
+        <register type="NOA" address="0x00009888" value="0x03A60055" />
+        <register type="NOA" address="0x00009888" value="0x03A70055" />
+        <register type="NOA" address="0x00009888" value="0x07C10300" />
+        <register type="NOA" address="0x00009888" value="0x0DC10CF4" />
+        <register type="NOA" address="0x00009888" value="0x0FC10C72" />
+        <register type="NOA" address="0x00009888" value="0x09C10000" />
+        <register type="NOA" address="0x00009888" value="0x01C10000" />
+        <register type="NOA" address="0x00009888" value="0x01C3AA00" />
+        <register type="NOA" address="0x00009888" value="0x01C400F0" />
+        <register type="NOA" address="0x00009888" value="0x01C65500" />
+        <register type="NOA" address="0x00009888" value="0x01C75500" />
+        <register type="NOA" address="0x00009888" value="0x07E10300" />
+        <register type="NOA" address="0x00009888" value="0x09E10CB1" />
+        <register type="NOA" address="0x00009888" value="0x0BE10D33" />
+        <register type="NOA" address="0x00009888" value="0x01E10000" />
+        <register type="NOA" address="0x00009888" value="0x01E300AA" />
+        <register type="NOA" address="0x00009888" value="0x01E4000F" />
+        <register type="NOA" address="0x00009888" value="0x01E60055" />
+        <register type="NOA" address="0x00009888" value="0x01E70055" />
+        <register type="NOA" address="0x00009888" value="0x0003FFFF" />
+        <register type="NOA" address="0x00009888" value="0x0008FFFF" />
+        <register type="NOA" address="0x00009888" value="0x0208FFFF" />
+        <register type="NOA" address="0x00009888" value="0x000A5555" />
+        <register type="NOA" address="0x00009888" value="0x020A5555" />
+        <register type="NOA" address="0x00009888" value="0x0001FFFF" />
+        <register type="NOA" address="0x00009888" value="0x000BAAAA" />
+        <register type="NOA" address="0x00009888" value="0x020BAAAA" />
+        <register type="NOA" address="0x00009888" value="0x0107AA55" />
+        <register type="NOA" address="0x00009888" value="0x0308AAFF" />
+        <register type="NOA" address="0x00009888" value="0x010900FF" />
+        <register type="NOA" address="0x00009888" value="0x010AAAAA" />
+        <register type="NOA" address="0x00009888" value="0x030A5555" />
+        <register type="NOA" address="0x00009888" value="0x01010F00" />
+        <register type="NOA" address="0x00009888" value="0x030B0055" />
+        <register type="NOA" address="0x00009888" value="0x010200F0" />
+        <register type="NOA" address="0x00009888" value="0x010C5500" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009888" value="0x00032492" />
+        <register type="NOA" address="0x00009888" value="0x02032492" />
+        <register type="NOA" address="0x00009888" value="0x04032492" />
+        <register type="NOA" address="0x00009888" value="0x06030002" />
+        <register type="NOA" address="0x00009888" value="0x14030000" />
+        <register type="NOA" address="0x00009888" value="0x3E030C6A" />
+        <register type="NOA" address="0x00009888" value="0x26030000" />
+        <register type="NOA" address="0x00009888" value="0x42032942" />
+        <register type="NOA" address="0x00009888" value="0x2C030000" />
+        <register type="NOA" address="0x00009888" value="0x32030000" />
+        <register type="NOA" address="0x00009888" value="0x4403000A" />
+        <register type="NOA" address="0x00009888" value="0x38030000" />
+        <register type="NOA" address="0x00009888" value="0x46030000" />
+        <register type="NOA" address="0x00009888" value="0x48030000" />
+        <register type="NOA" address="0x00009888" value="0x4A030603" />
+        <register type="NOA" address="0x00009888" value="0x16030000" />
+        <register type="NOA" address="0x00009888" value="0x18030000" />
+        <register type="NOA" address="0x00009888" value="0x1A031000" />
+        <register type="NOA" address="0x00009888" value="0x40030842" />
+        <register type="NOA" address="0x00009888" value="0x1C031000" />
+        <register type="NOA" address="0x00009888" value="0x1E031000" />
+        <register type="NOA" address="0x00009888" value="0x20031000" />
+        <register type="NOA" address="0x00009884" value="0x00000008" />
+        <register type="NOA" address="0x00009888" value="0x0001FFFF" />
+        <register type="NOA" address="0x00009888" value="0x00025555" />
+        <register type="NOA" address="0x00009888" value="0x02025555" />
+        <register type="NOA" address="0x00009888" value="0x00045555" />
+        <register type="NOA" address="0x00009888" value="0x02045555" />
+        <register type="NOA" address="0x00009888" value="0x00055555" />
+        <register type="NOA" address="0x00009888" value="0x02055555" />
+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0xF0800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x000F0000" />
+        <register type="OA" address="0x0000D940" value="0x00078000" />
+        <register type="OA" address="0x0000D944" value="0x00000FFF" />
+        <register type="OA" address="0x0000DC00" value="0x00078000" />
+        <register type="OA" address="0x0000DC04" value="0x00000FFF" />
+        <register type="OA" address="0x0000D948" value="0x00007800" />
+        <register type="OA" address="0x0000D94C" value="0x0000F0FF" />
+        <register type="OA" address="0x0000DC08" value="0x00007800" />
+        <register type="OA" address="0x0000DC0C" value="0x0000F0FF" />
+        <register type="OA" address="0x0000D950" value="0x00000780" />
+        <register type="OA" address="0x0000D954" value="0x0000FF0F" />
+        <register type="OA" address="0x0000DC10" value="0x00000780" />
+        <register type="OA" address="0x0000DC14" value="0x0000FF0F" />
+        <register type="OA" address="0x0000D958" value="0x00000078" />
+        <register type="OA" address="0x0000D95C" value="0x0000FFF0" />
+        <register type="OA" address="0x0000DC18" value="0x00000078" />
+        <register type="OA" address="0x0000DC1C" value="0x0000FFF0" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="L3_1"
+       chipset="DG2"
+       symbol_name="L3_1"
+       underscore_name="l3_1"
+       mdapi_supported_apis="VK OGL OCL MEDIA IO"
+       hw_config_guid="9a1a14ad-ab92-4204-8a91-c4159a54f712"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Quad0 L3 Bank0 Input Available"
+             symbol_name="L30Bank0InputAvailable"
+             underscore_name="l30_bank0_input_available"
+             description="The percentage of time in which Quad0 L3 bank0 has input available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank1 Input Available"
+             symbol_name="L30Bank1InputAvailable"
+             underscore_name="l30_bank1_input_available"
+             description="The percentage of time in which Quad0 L3 bank1 has input available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank2 Input Available"
+             symbol_name="L30Bank2InputAvailable"
+             underscore_name="l30_bank2_input_available"
+             description="The percentage of time in which Quad0 L3 bank2 has input available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank3 Input Available"
+             symbol_name="L30Bank3InputAvailable"
+             underscore_name="l30_bank3_input_available"
+             description="The percentage of time in which Quad0 L3 bank3 has input available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank4 Input Available"
+             symbol_name="L30Bank4InputAvailable"
+             underscore_name="l30_bank4_input_available"
+             description="The percentage of time in which Quad0 L3 bank4 has input available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank5 Input Available"
+             symbol_name="L30Bank5InputAvailable"
+             underscore_name="l30_bank5_input_available"
+             description="The percentage of time in which Quad0 L3 bank5 has input available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank6 Input Available"
+             symbol_name="L30Bank6InputAvailable"
+             underscore_name="l30_bank6_input_available"
+             description="The percentage of time in which Quad0 L3 bank6 has input available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank7 Input Available"
+             symbol_name="L30Bank7InputAvailable"
+             underscore_name="l30_bank7_input_available"
+             description="The percentage of time in which Quad0 L3 bank7 has input available"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank0 Output Ready"
+             symbol_name="L30Bank0OutputReady"
+             underscore_name="l30_bank0_output_ready"
+             description="The percentage of time in which quad0 L3 bank0 output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank1 Output Ready"
+             symbol_name="L30Bank1OutputReady"
+             underscore_name="l30_bank1_output_ready"
+             description="The percentage of time in which quad0 L3 bank1 output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank2 Output Ready"
+             symbol_name="L30Bank2OutputReady"
+             underscore_name="l30_bank2_output_ready"
+             description="The percentage of time in which quad0 L3 bank2 output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank3 Output Ready"
+             symbol_name="L30Bank3OutputReady"
+             underscore_name="l30_bank3_output_ready"
+             description="The percentage of time in which quad0 L3 bank3 output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank4 Output Ready"
+             symbol_name="L30Bank4OutputReady"
+             underscore_name="l30_bank4_output_ready"
+             description="The percentage of time in which quad0 L3 bank4 output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank5 Output Ready"
+             symbol_name="L30Bank5OutputReady"
+             underscore_name="l30_bank5_output_ready"
+             description="The percentage of time in which quad0 L3 bank5 output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank6 Output Ready"
+             symbol_name="L30Bank6OutputReady"
+             underscore_name="l30_bank6_output_ready"
+             description="The percentage of time in which quad0 L3 bank6 output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <counter name="Quad0 L3 Bank7 Output Ready"
+             symbol_name="L30Bank7OutputReady"
+             underscore_name="l30_bank7_output_ready"
+             description="The percentage of time in which quad0 L3 bank7 output is ready"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GTI/L3"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="slice"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x00000008" />
+        <register type="NOA" address="0x00009888" value="0x090B1000" />
+        <register type="NOA" address="0x00009888" value="0x092B1006" />
+        <register type="NOA" address="0x00009888" value="0x094B1000" />
+        <register type="NOA" address="0x00009888" value="0x096B1000" />
+        <register type="NOA" address="0x00009888" value="0x090C1000" />
+        <register type="NOA" address="0x00009888" value="0x092C1180" />
+        <register type="NOA" address="0x00009888" value="0x094C1000" />
+        <register type="NOA" address="0x00009888" value="0x096C1000" />
+        <register type="NOA" address="0x00009888" value="0x010B0002" />
+        <register type="NOA" address="0x00009888" value="0x0B0B0006" />
+        <register type="NOA" address="0x00009888" value="0x130B0007" />
+        <register type="NOA" address="0x00009888" value="0x030B0000" />
+        <register type="NOA" address="0x00009888" value="0x012B0002" />
+        <register type="NOA" address="0x00009888" value="0x112B0007" />
+        <register type="NOA" address="0x00009888" value="0x032B0000" />
+        <register type="NOA" address="0x00009888" value="0x014B0002" />
+        <register type="NOA" address="0x00009888" value="0x074B0006" />
+        <register type="NOA" address="0x00009888" value="0x0F4B0007" />
+        <register type="NOA" address="0x00009888" value="0x034B0000" />
+        <register type="NOA" address="0x00009888" value="0x016B0002" />
+        <register type="NOA" address="0x00009888" value="0x056B0006" />
+        <register type="NOA" address="0x00009888" value="0x0D6B0007" />
+        <register type="NOA" address="0x00009888" value="0x036B0000" />
+        <register type="NOA" address="0x00009888" value="0x010C0002" />
+        <register type="NOA" address="0x00009888" value="0x0B0C0180" />
+        <register type="NOA" address="0x00009888" value="0x130C01C0" />
+        <register type="NOA" address="0x00009888" value="0x030C0000" />
+        <register type="NOA" address="0x00009888" value="0x012C0002" />
+        <register type="NOA" address="0x00009888" value="0x112C01C0" />
+        <register type="NOA" address="0x00009888" value="0x032C0000" />
+        <register type="NOA" address="0x00009888" value="0x014C0002" />
+        <register type="NOA" address="0x00009888" value="0x074C0180" />
+        <register type="NOA" address="0x00009888" value="0x0F4C01C0" />
+        <register type="NOA" address="0x00009888" value="0x034C0000" />
+        <register type="NOA" address="0x00009888" value="0x016C0002" />
+        <register type="NOA" address="0x00009888" value="0x056C0180" />
+        <register type="NOA" address="0x00009888" value="0x0D6C01C0" />
+        <register type="NOA" address="0x00009888" value="0x036C0000" />
+        <register type="NOA" address="0x00009888" value="0x00020030" />
+        <register type="NOA" address="0x00009888" value="0x02020030" />
+        <register type="NOA" address="0x00009888" value="0x00030048" />
+        <register type="NOA" address="0x00009888" value="0x02030048" />
+        <register type="NOA" address="0x00009888" value="0x0004009B" />
+        <register type="NOA" address="0x00009888" value="0x0204009B" />
+        <register type="NOA" address="0x00009888" value="0x00052055" />
+        <register type="NOA" address="0x00009888" value="0x02052055" />
+        <register type="NOA" address="0x00009888" value="0x00064800" />
+        <register type="NOA" address="0x00009888" value="0x02064800" />
+        <register type="NOA" address="0x00009888" value="0x0007A0A0" />
+        <register type="NOA" address="0x00009888" value="0x000867AA" />
+        <register type="NOA" address="0x00009888" value="0x020867AA" />
+        <register type="NOA" address="0x00009888" value="0x0009A0A0" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x0008AAAA" />
+        <register type="NOA" address="0x00009888" value="0x0208AAAA" />
+        <register type="NOA" address="0x00009888" value="0x00032492" />
+        <register type="NOA" address="0x00009888" value="0x02032492" />
+        <register type="NOA" address="0x00009888" value="0x04032492" />
+        <register type="NOA" address="0x00009888" value="0x06030002" />
+        <register type="NOA" address="0x00009888" value="0x14030000" />
+        <register type="NOA" address="0x00009888" value="0x3E030024" />
+        <register type="NOA" address="0x00009888" value="0x26030000" />
+        <register type="NOA" address="0x00009888" value="0x42030401" />
+        <register type="NOA" address="0x00009888" value="0x2C030000" />
+        <register type="NOA" address="0x00009888" value="0x32030000" />
+        <register type="NOA" address="0x00009888" value="0x44030140" />
+        <register type="NOA" address="0x00009888" value="0x38030000" />
+        <register type="NOA" address="0x00009888" value="0x46030E01" />
+        <register type="NOA" address="0x00009888" value="0x48030010" />
+        <register type="NOA" address="0x00009888" value="0x4A030004" />
+        <register type="NOA" address="0x00009888" value="0x16030000" />
+        <register type="NOA" address="0x00009888" value="0x18030000" />
+        <register type="NOA" address="0x00009888" value="0x1A030000" />
+        <register type="NOA" address="0x00009888" value="0x40031C2A" />
+        <register type="NOA" address="0x00009888" value="0x1C030000" />
+        <register type="NOA" address="0x00009888" value="0x1E030000" />
+        <register type="NOA" address="0x00009888" value="0x20030000" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x00800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00000000" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="GpuBusyness"
+       chipset="DG2"
+       symbol_name="GpuBusyness"
+       underscore_name="gpu_busyness"
+       mdapi_supported_apis="VK OGL OCL MEDIA IO"
+       hw_config_guid="85de4c29-76b8-42a6-b6ef-8a2bed008fcf"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Active"
+             symbol_name="EuActive"
+             underscore_name="eu_active"
+             description="The percentage of time in which the Execution Units were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Stall"
+             symbol_name="EuStall"
+             underscore_name="eu_stall"
+             description="The percentage of time in which the Execution Units were stalled."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU Thread Occupancy"
+             symbol_name="EuThreadOccupancy"
+             underscore_name="eu_thread_occupancy"
+             description="The percentage of time in which hardware threads occupied EUs."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="8 A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FMUL $EuThreadsCount FDIV $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier2 Overview System Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis="VK OGL OCL IO MEDIA"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Render Ring Busy"
+             symbol_name="RenderBusy"
+             underscore_name="render_busy"
+             description="The percentage of time when render command streamer was busy."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="System Frame Batch"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="ComputeEngine0 Ring Busy"
+             symbol_name="ComputeEngine0Busy"
+             underscore_name="compute_engine0_busy"
+             description="The percentage of time when compute command streamer 0 was busy."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 3 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="System Frame Batch"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="ComputeEngine1 Ring Busy"
+             symbol_name="ComputeEngine1Busy"
+             underscore_name="compute_engine1_busy"
+             description="The percentage of time when compute command streamer 1 was busy."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 4 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="System Frame Batch"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="ComputeEngine2 Ring Busy"
+             symbol_name="ComputeEngine2Busy"
+             underscore_name="compute_engine2_busy"
+             description="The percentage of time when compute command streamer 2 was busy."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="System Frame Batch"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="ComputeEngine3 Ring Busy"
+             symbol_name="ComputeEngine3Busy"
+             underscore_name="compute_engine3_busy"
+             description="The percentage of time when compute command streamer 3 was busy."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="System Frame Batch"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Posh Ring Busy"
+             symbol_name="PoshEngineBusy"
+             underscore_name="posh_engine_busy"
+             description="The percentage of time when posh command streamer was busy."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 2 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="System Frame Batch"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Blitter Ring Busy"
+             symbol_name="BlitterBusy"
+             underscore_name="blitter_busy"
+             description="The percentage of time when blitter command streamer was busy."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="System Frame Batch"
+             mdapi_supported_apis="IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Vdbox00 Ring Busy"
+             symbol_name="Vdbox00Busy"
+             underscore_name="vdbox00_busy"
+             description="The percentage of time when vdbox00 command streamer was busy."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="C 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="System Frame Batch"
+             mdapi_supported_apis="IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Vebox0 Ring Busy"
+             symbol_name="Vebox0Busy"
+             underscore_name="vebox0_busy"
+             description="The percentage of time when vebox0 command streamer was busy."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 7 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="System Frame Batch"
+             mdapi_supported_apis="IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Vdbox10 Ring Busy"
+             symbol_name="Vdbox10Busy"
+             underscore_name="vdbox10_busy"
+             description="The percentage of time when vdbox10 command streamer was busy."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 6 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="System Frame Batch"
+             mdapi_supported_apis="IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Vebox1 Ring Busy"
+             symbol_name="Vebox1Busy"
+             underscore_name="vebox1_busy"
+             description="The percentage of time when vebox1 command streamer was busy."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 5 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="System Frame Batch"
+             mdapi_supported_apis="IO"
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Any compute engine busy"
+             symbol_name="AnyComputeEngineBusy"
+             underscore_name="any_compute_engine_busy"
+             description="The percentage of time when any compute engine is busy"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Render and any compute engines are simultaneously busy"
+             symbol_name="RenderAndAnyComputeEngineBusy"
+             underscore_name="render_and_any_compute_engine_busy"
+             description="The percentage of time when render and any compute engines are simultaneously busy"
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="B 1 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x00000004" />
+        <register type="NOA" address="0x00009888" value="0x420A2400" />
+        <register type="NOA" address="0x00009888" value="0x1A131200" />
+        <register type="NOA" address="0x00009888" value="0x040A0022" />
+        <register type="NOA" address="0x00009888" value="0x120A0014" />
+        <register type="NOA" address="0x00009888" value="0x020A0000" />
+        <register type="NOA" address="0x00009888" value="0x000A0000" />
+        <register type="NOA" address="0x00009888" value="0x00104000" />
+        <register type="NOA" address="0x00009888" value="0x02100002" />
+        <register type="NOA" address="0x00009888" value="0x00124000" />
+        <register type="NOA" address="0x00009888" value="0x00134000" />
+        <register type="NOA" address="0x00009888" value="0x0A130024" />
+        <register type="NOA" address="0x00009884" value="0x00000005" />
+        <register type="NOA" address="0x00009888" value="0x420A2400" />
+        <register type="NOA" address="0x00009888" value="0x1A131200" />
+        <register type="NOA" address="0x00009888" value="0x040A0022" />
+        <register type="NOA" address="0x00009888" value="0x100A0014" />
+        <register type="NOA" address="0x00009888" value="0x020A0000" />
+        <register type="NOA" address="0x00009888" value="0x000A0000" />
+        <register type="NOA" address="0x00009888" value="0x00102400" />
+        <register type="NOA" address="0x00009888" value="0x00120400" />
+        <register type="NOA" address="0x00009888" value="0x00134000" />
+        <register type="NOA" address="0x00009888" value="0x08130024" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x18012400" />
+        <register type="NOA" address="0x00009888" value="0x18042400" />
+        <register type="NOA" address="0x00009888" value="0x20042400" />
+        <register type="NOA" address="0x00009888" value="0x28042400" />
+        <register type="NOA" address="0x00009888" value="0x32040009" />
+        <register type="NOA" address="0x00009888" value="0x3C041200" />
+        <register type="NOA" address="0x00009888" value="0x4C041200" />
+        <register type="NOA" address="0x00009888" value="0x02010006" />
+        <register type="NOA" address="0x00009888" value="0x10010004" />
+        <register type="NOA" address="0x00009888" value="0x0001FC00" />
+        <register type="NOA" address="0x00009888" value="0x00042EE9" />
+        <register type="NOA" address="0x00009888" value="0x02040239" />
+        <register type="NOA" address="0x00009888" value="0x0404000F" />
+        <register type="NOA" address="0x00009888" value="0x0E040D00" />
+        <register type="NOA" address="0x00009888" value="0x10040304" />
+        <register type="NOA" address="0x00009888" value="0x12040714" />
+        <register type="NOA" address="0x00009888" value="0x14040024" />
+        <register type="NOA" address="0x00009888" value="0x02085554" />
+        <register type="NOA" address="0x00009888" value="0x00056800" />
+        <register type="NOA" address="0x00009888" value="0x02050001" />
+        <register type="NOA" address="0x00009888" value="0x02032B6D" />
+        <register type="NOA" address="0x00009888" value="0x04032492" />
+        <register type="NOA" address="0x00009888" value="0x06030002" />
+        <register type="NOA" address="0x00009888" value="0x44030000" />
+        <register type="NOA" address="0x00009888" value="0x46030400" />
+        <register type="NOA" address="0x00009888" value="0x48030000" />
+        <register type="NOA" address="0x00009888" value="0x4A030C62" />
+        <register type="NOA" address="0x00009888" value="0x16033000" />
+        <register type="NOA" address="0x00009888" value="0x3E031080" />
+        <register type="NOA" address="0x00009888" value="0x18033000" />
+        <register type="NOA" address="0x00009888" value="0x1A033000" />
+        <register type="NOA" address="0x00009888" value="0x40031084" />
+        <register type="NOA" address="0x00009888" value="0x1C033000" />
+        <register type="NOA" address="0x00009888" value="0x1E033000" />
+        <register type="NOA" address="0x00009888" value="0x20033000" />
+        <register type="NOA" address="0x00009888" value="0x42030004" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x30800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00030000" />
+        <register type="OA" address="0x0000D940" value="0x00000000" />
+        <register type="OA" address="0x0000D944" value="0x000087FF" />
+        <register type="OA" address="0x0000DC00" value="0x00000000" />
+        <register type="OA" address="0x0000DC04" value="0x000087FF" />
+        <register type="OA" address="0x0000D948" value="0x00040001" />
+        <register type="OA" address="0x0000D94C" value="0x000007FF" />
+        <register type="OA" address="0x0000DC08" value="0x00040001" />
+        <register type="OA" address="0x0000DC0C" value="0x000007FF" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00804704" />
+        <register type="FLEX" address="0x0000E558" value="0x00A04904" />
+        <register type="FLEX" address="0x0000E658" value="0x00805705" />
+        <register type="FLEX" address="0x0000E758" value="0x00A05905" />
+        <register type="FLEX" address="0x0000E45C" value="0x00808708" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A08908" />
+    </register_config>
+  </set>
+
+  <set name="EuActivity1"
+       chipset="DG2"
+       symbol_name="EuActivity1"
+       underscore_name="eu_activity1"
+       mdapi_supported_apis="OGL OGL4 OCL MEDIA IO"
+       hw_config_guid="8dc7142a-df10-41f2-acaa-9faea547ce73"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS FPU Pipe Active"
+             symbol_name="VsFpuActive"
+             underscore_name="vs_fpu_active"
+             description="The percentage of time in which EU FPU INT64 pipeline was actively processing a vertex shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS FPU Pipe Active"
+             symbol_name="PsFpuActive"
+             underscore_name="ps_fpu_active"
+             description="The percentage of time in which EU FPU INT64 pipeline was actively processing a fragment shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="Send Pipe Active"
+             symbol_name="SendActive"
+             underscore_name="send_active"
+             description="The percentage of time in which EU send pipeline was actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Pipes"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x00800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00000000" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00810710" />
+        <register type="FLEX" address="0x0000E558" value="0x00A10910" />
+        <register type="FLEX" address="0x0000E658" value="0x00850750" />
+        <register type="FLEX" address="0x0000E758" value="0x00A50950" />
+        <register type="FLEX" address="0x0000E45C" value="0x00802702" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A02902" />
+    </register_config>
+  </set>
+
+  <set name="EuActivity2"
+       chipset="DG2"
+       symbol_name="EuActivity2"
+       underscore_name="eu_activity2"
+       mdapi_supported_apis="OGL OGL4 OCL MEDIA IO"
+       hw_config_guid="d3c05cce-4ae7-4291-82ab-b0bcade6cf7a"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS EM Pipe Active"
+             symbol_name="CsEmActive"
+             underscore_name="cs_em_active"
+             description="The percentage of time in which EU EM INT pipeline was actively processing a compute shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS FPU Pipe Active"
+             symbol_name="CsFpuActive"
+             underscore_name="cs_fpu_active"
+             description="The percentage of time in which EU FPU INT64 pipeline was actively processing a compute shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Send Pipe Active"
+             symbol_name="CsSendActive"
+             underscore_name="cs_send_active"
+             description="The percentage of time in which EU send pipeline was actively processing a compute shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x00800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00000000" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00861761" />
+        <register type="FLEX" address="0x0000E558" value="0x00A61961" />
+        <register type="FLEX" address="0x0000E658" value="0x00860760" />
+        <register type="FLEX" address="0x0000E758" value="0x00A60960" />
+        <register type="FLEX" address="0x0000E45C" value="0x00862762" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A62962" />
+    </register_config>
+  </set>
+
+  <set name="EuActivity3"
+       chipset="DG2"
+       symbol_name="EuActivity3"
+       underscore_name="eu_activity3"
+       mdapi_supported_apis="OGL OGL4 OCL MEDIA IO"
+       hw_config_guid="4fd009c5-6eae-4ecb-a2d1-41edf211cd06"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS EM Pipe Active"
+             symbol_name="VsEmActive"
+             underscore_name="vs_em_active"
+             description="The percentage of time in which EU EM INT pipeline was actively processing a vertex shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS EM Pipe Active"
+             symbol_name="PsEmActive"
+             underscore_name="ps_em_active"
+             description="The percentage of time in which EU EM INT pipeline was actively processing a fragment shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Send Pipe Active"
+             symbol_name="PsSendActive"
+             underscore_name="ps_send_active"
+             description="The percentage of time in which EU send pipeline was actively processing a fragment shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x00800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00000000" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00811711" />
+        <register type="FLEX" address="0x0000E558" value="0x00A11911" />
+        <register type="FLEX" address="0x0000E658" value="0x00851751" />
+        <register type="FLEX" address="0x0000E758" value="0x00A51951" />
+        <register type="FLEX" address="0x0000E45C" value="0x00852752" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A52952" />
+    </register_config>
+  </set>
+
+  <set name="EuActivity4"
+       chipset="DG2"
+       symbol_name="EuActivity4"
+       underscore_name="eu_activity4"
+       mdapi_supported_apis="OGL OGL4 OCL MEDIA IO"
+       hw_config_guid="5226090d-b636-4845-9f76-545cb5db7867"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS FPU Pipe Active"
+             symbol_name="HsFpuActive"
+             underscore_name="hs_fpu_active"
+             description="The percentage of time in which EU FPU INT64 pipeline was actively processing a hull shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS FPU Pipe Active"
+             symbol_name="DsFpuActive"
+             underscore_name="ds_fpu_active"
+             description="The percentage of time in which EU FPU INT64 pipeline was actively processing a domain shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Send Pipe Active"
+             symbol_name="VsSendActive"
+             underscore_name="vs_send_active"
+             description="The percentage of time in which EU send pipeline was actively processing a vertex shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x00800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00000000" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00820720" />
+        <register type="FLEX" address="0x0000E558" value="0x00A20920" />
+        <register type="FLEX" address="0x0000E658" value="0x00830730" />
+        <register type="FLEX" address="0x0000E758" value="0x00A30930" />
+        <register type="FLEX" address="0x0000E45C" value="0x00812712" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A12912" />
+    </register_config>
+  </set>
+
+  <set name="EuActivity5"
+       chipset="DG2"
+       symbol_name="EuActivity5"
+       underscore_name="eu_activity5"
+       mdapi_supported_apis="OGL OGL4 OCL MEDIA IO"
+       hw_config_guid="c8d33b8f-3152-4c81-99bf-d3cb45d7e1ac"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS EM Pipe Active"
+             symbol_name="HsEmActive"
+             underscore_name="hs_em_active"
+             description="The percentage of time in which EU EM INT pipeline was actively processing a hull shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS EM Pipe Active"
+             symbol_name="DsEmActive"
+             underscore_name="ds_em_active"
+             description="The percentage of time in which EU EM INT pipeline was actively processing a domain shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Send Pipe Active"
+             symbol_name="HsSendActive"
+             underscore_name="hs_send_active"
+             description="The percentage of time in which EU send pipeline was actively processing a hull shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x00800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00000000" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00821721" />
+        <register type="FLEX" address="0x0000E558" value="0x00A21921" />
+        <register type="FLEX" address="0x0000E658" value="0x00831731" />
+        <register type="FLEX" address="0x0000E758" value="0x00A31931" />
+        <register type="FLEX" address="0x0000E45C" value="0x00822722" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A22922" />
+    </register_config>
+  </set>
+
+  <set name="EuActivity6"
+       chipset="DG2"
+       symbol_name="EuActivity6"
+       underscore_name="eu_activity6"
+       mdapi_supported_apis="OGL OGL4 OCL MEDIA IO"
+       hw_config_guid="708d01ed-bc9b-420e-8048-9060abcbb1f0"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS FPU Pipe Active"
+             symbol_name="GsFpuActive"
+             underscore_name="gs_fpu_active"
+             description="The percentage of time in which EU FPU INT64 pipeline was actively processing a geometry shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS EM Pipe Active"
+             symbol_name="GsEmActive"
+             underscore_name="gs_em_active"
+             description="The percentage of time in which EU EM INT pipeline was actively processing a geometry shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Send Pipe Active"
+             symbol_name="GsSendActive"
+             underscore_name="gs_send_active"
+             description="The percentage of time in which EU send pipeline was actively processing a geometry shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x00800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00000000" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00840740" />
+        <register type="FLEX" address="0x0000E558" value="0x00A40940" />
+        <register type="FLEX" address="0x0000E658" value="0x00841741" />
+        <register type="FLEX" address="0x0000E758" value="0x00A41941" />
+        <register type="FLEX" address="0x0000E45C" value="0x00842742" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A42942" />
+    </register_config>
+  </set>
+
+  <set name="EuActivity7"
+       chipset="DG2"
+       symbol_name="EuActivity7"
+       underscore_name="eu_activity7"
+       mdapi_supported_apis="OGL OGL4 OCL MEDIA IO"
+       hw_config_guid="72262d5b-a9e6-4742-820b-2a7bef9a7e61"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU FPU Pipe Active"
+             symbol_name="FpuActive"
+             underscore_name="fpu_active"
+             description="The percentage of time in which EU FPU INT64 pipeline was actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Pipes"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EM Pipe Active"
+             symbol_name="EmActive"
+             underscore_name="em_active"
+             description="The percentage of time in which EU EM INT pipeline was actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Pipes"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EM And FPU Pipes Active"
+             symbol_name="EmFpuActive"
+             underscore_name="em_fpu_active"
+             description="The percentage of time in which EU EM INT and FPU INT64 pipelines were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Pipes"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="EU AVG IPC Rate"
+             symbol_name="EuAvgIpcRate"
+             underscore_name="eu_avg_ipc_rate"
+             description="The average rate of IPC calculated for 2 FPU pipelines."
+             data_type="float"
+             max_equation="2"
+             units="number"
+             semantic_type="ratio"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD  A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD  A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FADD  A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD FSUB FDIV 1 FADD"
+             mdapi_group="EU Array"
+             mdapi_usage_flags="Tier4 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x00800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00000000" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00803703" />
+        <register type="FLEX" address="0x0000E558" value="0x00A03903" />
+        <register type="FLEX" address="0x0000E658" value="0x00800700" />
+        <register type="FLEX" address="0x0000E758" value="0x00A00900" />
+        <register type="FLEX" address="0x0000E45C" value="0x00801701" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A01901" />
+    </register_config>
+  </set>
+
+  <set name="EuActivity8"
+       chipset="DG2"
+       symbol_name="EuActivity8"
+       underscore_name="eu_activity8"
+       mdapi_supported_apis="OGL OGL4 OCL MEDIA IO"
+       hw_config_guid="11c685de-1445-41b8-acb5-8e79c75c4fd8"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Busy"
+             symbol_name="GpuBusy"
+             underscore_name="gpu_busy"
+             description="The percentage of time in which the GPU has been processing GPU commands."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 0 READ 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="VS Threads Dispatched"
+             symbol_name="VsThreads"
+             underscore_name="vs_threads"
+             description="The total number of vertex shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 1 READ"
+             mdapi_group="EU Array/Vertex Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="HS Threads Dispatched"
+             symbol_name="HsThreads"
+             underscore_name="hs_threads"
+             description="The total number of hull shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 2 READ"
+             mdapi_group="EU Array/Hull Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Threads Dispatched"
+             symbol_name="DsThreads"
+             underscore_name="ds_threads"
+             description="The total number of domain shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 3 READ"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GS Threads Dispatched"
+             symbol_name="GsThreads"
+             underscore_name="gs_threads"
+             description="The total number of geometry shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 5 READ"
+             mdapi_group="EU Array/Geometry Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="FS Threads Dispatched"
+             symbol_name="PsThreads"
+             underscore_name="ps_threads"
+             description="The total number of fragment shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 6 READ"
+             mdapi_group="EU Array/Fragment Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="CS Threads Dispatched"
+             symbol_name="CsThreads"
+             underscore_name="cs_threads"
+             description="The total number of compute shader hardware threads dispatched."
+             data_type="uint64"
+             units="threads"
+             semantic_type="event"
+             equation="A 4 READ"
+             mdapi_group="EU Array/Compute Shader"
+             mdapi_usage_flags="Tier3 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="DS Send Pipe Active"
+             symbol_name="DsSendActive"
+             underscore_name="ds_send_active"
+             description="The percentage of time in which EU send pipeline was actively processing a domain shader instruction."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 7 READ A 8 READ FADD A 9 READ FADD A 10 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Domain Shader"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SYS And FPU Pipes Active"
+             symbol_name="SysFpuActive"
+             underscore_name="sys_fpu_active"
+             description="The percentage of time in which EU Systolic and FPU INT64 pipelines were actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 15 READ A 16 READ FADD A 17 READ FADD A 18 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Pipes"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="SYS Pipe Active"
+             symbol_name="SysActive"
+             underscore_name="sys_active"
+             description="The percentage of time in which EU SYS pipeline was actively processing."
+             data_type="float"
+             max_equation="100"
+             units="percent"
+             semantic_type="duration"
+             equation="A 11 READ A 12 READ FADD A 13 READ FADD A 14 READ FADD $EuCoresTotalCount UDIV 100 UMUL $GpuCoreClocks FDIV"
+             mdapi_group="EU Array/Pipes"
+             mdapi_usage_flags="Tier3 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0x00800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0x00800000" />
+        <register type="OA" address="0x0000DC40" value="0x00000000" />
+    </register_config>
+    <register_config type="FLEX">
+        <register type="FLEX" address="0x0000E458" value="0x00832732" />
+        <register type="FLEX" address="0x0000E558" value="0x00A32932" />
+        <register type="FLEX" address="0x0000E658" value="0x00809709" />
+        <register type="FLEX" address="0x0000E758" value="0x00A09909" />
+        <register type="FLEX" address="0x0000E45C" value="0x0080A70A" />
+        <register type="FLEX" address="0x0000E55C" value="0x00A0A90A" />
+    </register_config>
+  </set>
+
+  <set name="Metric set TestOa"
+       chipset="DG2"
+       symbol_name="TestOa"
+       underscore_name="test_oa"
+       mdapi_supported_apis="VK OGL OCL MEDIA IO"
+       hw_config_guid="2db1e1d4-8c48-4bb7-b248-0f40b70433ff"
+       >
+    <counter name="GPU Time Elapsed"
+             symbol_name="GpuTime"
+             underscore_name="gpu_time"
+             description="Time elapsed on the GPU during the measurement."
+             data_type="uint64"
+             units="ns"
+             semantic_type="duration"
+             equation="GPU_TIME 0 READ 1000000000 UMUL $GpuTimestampFrequency UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="GPU Core Clocks"
+             symbol_name="GpuCoreClocks"
+             underscore_name="gpu_core_clocks"
+             description="The total number of GPU core clocks elapsed during the measurement."
+             data_type="uint64"
+             units="cycles"
+             semantic_type="event"
+             equation="GPU_CLOCK 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="AVG GPU Core Frequency"
+             symbol_name="AvgGpuCoreFrequency"
+             underscore_name="avg_gpu_core_frequency"
+             description="Average GPU Core Frequency in the measurement."
+             data_type="uint64"
+             max_equation="$GpuMaxFrequency"
+             units="hz"
+             semantic_type="event"
+             equation="$GpuCoreClocks 1000000000 UMUL $GpuTime UDIV"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Tier1 Overview System Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="TestCounter0"
+             symbol_name="Counter0"
+             underscore_name="counter0"
+             description="HW test counter 0. Factor: 0.0"
+             data_type="uint64"
+             units="events"
+             semantic_type="event"
+             equation="B 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="TestCounter1"
+             symbol_name="Counter1"
+             underscore_name="counter1"
+             description="HW test counter 1. Factor: 1.0"
+             data_type="uint64"
+             units="events"
+             semantic_type="event"
+             equation="B 1 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="TestCounter2"
+             symbol_name="Counter2"
+             underscore_name="counter2"
+             description="HW test counter 2. Factor: 1.0"
+             data_type="uint64"
+             units="events"
+             semantic_type="event"
+             equation="B 2 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="TestCounter3"
+             symbol_name="Counter3"
+             underscore_name="counter3"
+             description="HW test counter 3. Factor: 0.5"
+             data_type="uint64"
+             units="events"
+             semantic_type="event"
+             equation="B 3 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="TestCounter4"
+             symbol_name="Counter4"
+             underscore_name="counter4"
+             description="HW test counter 4. Factor: 0.3333"
+             data_type="uint64"
+             units="events"
+             semantic_type="event"
+             equation="B 4 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="TestCounter5"
+             symbol_name="Counter5"
+             underscore_name="counter5"
+             description="HW test counter 5. Factor: 0.3333"
+             data_type="uint64"
+             units="events"
+             semantic_type="event"
+             equation="B 5 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="TestCounter6"
+             symbol_name="Counter6"
+             underscore_name="counter6"
+             description="HW test counter 6. Factor: 0.16666"
+             data_type="uint64"
+             units="events"
+             semantic_type="event"
+             equation="B 6 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="TestCounter7"
+             symbol_name="Counter7"
+             underscore_name="counter7"
+             description="HW test counter 7. Factor: 0.6666"
+             data_type="uint64"
+             units="events"
+             semantic_type="event"
+             equation="B 7 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="TestCounter8"
+             symbol_name="Counter8"
+             underscore_name="counter8"
+             description="HW test counter 8. Should be equal to 1 in IOStream or in OAG query mode"
+             data_type="uint64"
+             units="events"
+             semantic_type="event"
+             equation="C 0 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <counter name="TestCounter9 - OAR enable"
+             symbol_name="Counter9"
+             underscore_name="counter9"
+             description="HW test counter 9. Should be equal to 1 in query."
+             data_type="uint64"
+             units="events"
+             semantic_type="event"
+             equation="C 1 READ"
+             mdapi_group="GPU"
+             mdapi_usage_flags="Frame Batch Draw"
+             mdapi_supported_apis=""
+             mdapi_hw_unit_type="gpu"
+             />
+    <register_config type="NOA">
+        <register type="NOA" address="0x00000D04" value="0x00000200" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x2E080000" />
+        <register type="NOA" address="0x00009888" value="0x34080000" />
+        <register type="NOA" address="0x00009888" value="0x06088000" />
+        <register type="NOA" address="0x00009888" value="0x10080FF3" />
+        <register type="NOA" address="0x00009888" value="0x08080000" />
+        <register type="NOA" address="0x00009888" value="0x02080000" />
+        <register type="NOA" address="0x00009888" value="0x02032400" />
+        <register type="NOA" address="0x00009888" value="0x48030000" />
+        <register type="NOA" address="0x00009888" value="0x4A030000" />
+        <register type="NOA" address="0x00009884" value="0x00000000" />
+        <register type="NOA" address="0x00009888" value="0x5C030002" />
+    </register_config>
+    <register_config type="OA">
+        <register type="OA" address="0x0000D920" value="0x00000000" />
+        <register type="OA" address="0x0000D900" value="0x00000000" />
+        <register type="OA" address="0x0000D904" value="0xF0800000" />
+        <register type="OA" address="0x0000D910" value="0x00000000" />
+        <register type="OA" address="0x0000D914" value="0xF0800000" />
+        <register type="OA" address="0x0000DC40" value="0x00FF0000" />
+        <register type="OA" address="0x0000D940" value="0x00000004" />
+        <register type="OA" address="0x0000D944" value="0x0000FFFF" />
+        <register type="OA" address="0x0000DC00" value="0x00000004" />
+        <register type="OA" address="0x0000DC04" value="0x0000FFFF" />
+        <register type="OA" address="0x0000D948" value="0x00000003" />
+        <register type="OA" address="0x0000D94C" value="0x0000FFFF" />
+        <register type="OA" address="0x0000DC08" value="0x00000003" />
+        <register type="OA" address="0x0000DC0C" value="0x0000FFFF" />
+        <register type="OA" address="0x0000D950" value="0x00000007" />
+        <register type="OA" address="0x0000D954" value="0x0000FFFF" />
+        <register type="OA" address="0x0000DC10" value="0x00000007" />
+        <register type="OA" address="0x0000DC14" value="0x0000FFFF" />
+        <register type="OA" address="0x0000D958" value="0x00100002" />
+        <register type="OA" address="0x0000D95C" value="0x0000FFF7" />
+        <register type="OA" address="0x0000DC18" value="0x00100002" />
+        <register type="OA" address="0x0000DC1C" value="0x0000FFF7" />
+        <register type="OA" address="0x0000D960" value="0x00100002" />
+        <register type="OA" address="0x0000D964" value="0x0000FFCF" />
+        <register type="OA" address="0x0000DC20" value="0x00100002" />
+        <register type="OA" address="0x0000DC24" value="0x0000FFCF" />
+        <register type="OA" address="0x0000D968" value="0x00100082" />
+        <register type="OA" address="0x0000D96C" value="0x0000FFEF" />
+        <register type="OA" address="0x0000DC28" value="0x00100082" />
+        <register type="OA" address="0x0000DC2C" value="0x0000FFEF" />
+        <register type="OA" address="0x0000D970" value="0x001000C2" />
+        <register type="OA" address="0x0000D974" value="0x0000FFE7" />
+        <register type="OA" address="0x0000DC30" value="0x001000C2" />
+        <register type="OA" address="0x0000DC34" value="0x0000FFE7" />
+        <register type="OA" address="0x0000D978" value="0x00100001" />
+        <register type="OA" address="0x0000D97C" value="0x0000FFE7" />
+        <register type="OA" address="0x0000DC38" value="0x00100001" />
+        <register type="OA" address="0x0000DC3C" value="0x0000FFE7" />
+    </register_config>
+  </set>
+
+</metrics>
diff --git a/lib/i915/perf-configs/perf-metricset-codegen.py b/lib/i915/perf-configs/perf-metricset-codegen.py
index 00528c31..58237c27 100644
--- a/lib/i915/perf-configs/perf-metricset-codegen.py
+++ b/lib/i915/perf-configs/perf-metricset-codegen.py
@@ -126,6 +126,19 @@ def generate_metric_sets(args, gen):
                 metric_set->c_offset = metric_set->b_offset + 8;
                 metric_set->perfcnt_offset = metric_set->c_offset + 8;
 
+            """))
+        elif gen.chipset == "dg2":
+            c(textwrap.dedent("""\
+                metric_set->perf_oa_format = I915_OA_FORMAT_A24u40_A14u32_B8_C8;
+
+                metric_set->perf_raw_size = 256;
+                metric_set->gpu_time_offset = 0;
+                metric_set->gpu_clock_offset = 1;
+                metric_set->a_offset = 2;
+                metric_set->b_offset = metric_set->a_offset + 38;
+                metric_set->c_offset = metric_set->b_offset + 8;
+                metric_set->perfcnt_offset = metric_set->c_offset + 8;
+
             """))
         else:
             c(textwrap.dedent("""\
diff --git a/lib/i915/perf-configs/update-guids.py b/lib/i915/perf-configs/update-guids.py
index d38a7a95..3cd06f13 100755
--- a/lib/i915/perf-configs/update-guids.py
+++ b/lib/i915/perf-configs/update-guids.py
@@ -174,7 +174,7 @@ chipsets = [ 'hsw',
              'bxt', 'glk',
              'cnl',
              'icl', 'ehl',
-             'tglgt1', 'tglgt2', 'rkl', 'dg1', 'adl' ]
+             'tglgt1', 'tglgt2', 'rkl', 'dg1', 'adl', 'dg2', ]
 
 for chipset in chipsets:
     filename = 'oa-' + chipset + '.xml'
diff --git a/lib/i915/perf.c b/lib/i915/perf.c
index d004313f..39179546 100644
--- a/lib/i915/perf.c
+++ b/lib/i915/perf.c
@@ -58,6 +58,7 @@
 #include "i915_perf_metrics_rkl.h"
 #include "i915_perf_metrics_dg1.h"
 #include "i915_perf_metrics_adl.h"
+#include "i915_perf_metrics_dg2.h"
 
 static int
 perf_ioctl(int fd, unsigned long request, void *arg)
@@ -280,6 +281,8 @@ intel_perf_for_devinfo(uint32_t device_id,
 	} else if (devinfo->is_alderlake_s || devinfo->is_alderlake_p ||
 		   devinfo->is_raptorlake_s || devinfo->is_alderlake_n) {
 		intel_perf_load_metrics_adl(perf);
+	} else if (devinfo->is_dg2) {
+		intel_perf_load_metrics_dg2(perf);
 	} else {
 		return unsupported_i915_perf_platform(perf);
 	}
diff --git a/lib/meson.build b/lib/meson.build
index 98c2803b..8dabd64d 100644
--- a/lib/meson.build
+++ b/lib/meson.build
@@ -254,7 +254,7 @@ i915_perf_hardware = [
   'bxt', 'glk',
   'cnl',
   'icl', 'ehl',
-  'tglgt1', 'tglgt2', 'rkl', 'dg1', 'adl',
+  'tglgt1', 'tglgt2', 'rkl', 'dg1', 'adl', 'dg2',
 ]
 
 i915_xml_files = []
-- 
2.25.1