[Intel-gfx] [PATCH 1/2] drm/i915/pmu: Correct the rc6 offset upon enabling

[Intel-gfx] [PATCH 1/2] drm/i915/pmu: Correct the rc6 offset upon enabling

[Chris Wilson]

The rc6 residency starts ticking from 0 from BIOS POST, but the kernel
starts measuring the time from its boot. If we start measuruing
I915_PMU_RC6_RESIDENCY while the GT is idle, we start our sampling from
0 and then upon first activity (park/unpark) add in all the rc6
residency since boot. After the first park with the sampler engaged, the
sleep/active counters are aligned.

v2: With a wakeref to be sure

Fixes: df6a42053513 ("drm/i915/pmu: Ensure monotonic rc6")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
---
 drivers/gpu/drm/i915/i915_pmu.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/drivers/gpu/drm/i915/i915_pmu.c b/drivers/gpu/drm/i915/i915_pmu.c
index 28a82c849bac..ec0299490dd4 100644
--- a/drivers/gpu/drm/i915/i915_pmu.c
+++ b/drivers/gpu/drm/i915/i915_pmu.c
@@ -637,8 +637,10 @@ static void i915_pmu_enable(struct perf_event *event)
 		container_of(event->pmu, typeof(*i915), pmu.base);
 	unsigned int bit = event_enabled_bit(event);
 	struct i915_pmu *pmu = &i915->pmu;
+	intel_wakeref_t wakeref;
 	unsigned long flags;
 
+	wakeref = intel_runtime_pm_get(&i915->runtime_pm);
 	spin_lock_irqsave(&pmu->lock, flags);
 
 	/*
@@ -648,6 +650,14 @@ static void i915_pmu_enable(struct perf_event *event)
 	BUILD_BUG_ON(ARRAY_SIZE(pmu->enable_count) != I915_PMU_MASK_BITS);
 	GEM_BUG_ON(bit >= ARRAY_SIZE(pmu->enable_count));
 	GEM_BUG_ON(pmu->enable_count[bit] == ~0);
+
+	if (pmu->enable_count[bit] == 0 &&
+	    config_enabled_mask(I915_PMU_RC6_RESIDENCY) & BIT_ULL(bit)) {
+		pmu->sample[__I915_SAMPLE_RC6_LAST_REPORTED].cur = 0;
+		pmu->sample[__I915_SAMPLE_RC6].cur = __get_rc6(&i915->gt);
+		pmu->sleep_last = ktime_get();
+	}
+
 	pmu->enable |= BIT_ULL(bit);
 	pmu->enable_count[bit]++;
 
@@ -688,6 +698,8 @@ static void i915_pmu_enable(struct perf_event *event)
 	 * an existing non-zero value.
 	 */
 	local64_set(&event->hw.prev_count, __i915_pmu_event_read(event));
+
+	intel_runtime_pm_put(&i915->runtime_pm, wakeref);
 }
 
 static void i915_pmu_disable(struct perf_event *event)
-- 
2.25.0.rc2

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

[Intel-gfx] [PATCH 2/2] drm/i915/gt: Clear rc6 residency trackers across suspend

[Chris Wilson]

On suspend, the rc6 residency counters (stored in HW registers) will be
lost and cleared. However, we keep track of the rc6 residency to provide
a continuous 64b sampling, and if we see the HW value go backwards, we
assume it overflowed and add on 32b/40b -- an interesting artifact when
sampling across suspend.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
---
 drivers/gpu/drm/i915/gt/intel_rc6.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/i915/gt/intel_rc6.c b/drivers/gpu/drm/i915/gt/intel_rc6.c
index 9e303c29d6e3..04eeb7740e53 100644
--- a/drivers/gpu/drm/i915/gt/intel_rc6.c
+++ b/drivers/gpu/drm/i915/gt/intel_rc6.c
@@ -542,6 +542,8 @@ void intel_rc6_init(struct intel_rc6 *rc6)
 
 void intel_rc6_sanitize(struct intel_rc6 *rc6)
 {
+	memset(rc6->prev_hw_residency, 0, sizeof(rc6->prev_hw_residency));
+
 	if (rc6->enabled) { /* unbalanced suspend/resume */
 		rpm_get(rc6);
 		rc6->enabled = false;
-- 
2.25.0.rc2

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [Intel-gfx] [PATCH 1/2] drm/i915/pmu: Correct the rc6 offset upon enabling

[Chris Wilson]

Quoting Chris Wilson (2020-01-14 10:56:47)
> The rc6 residency starts ticking from 0 from BIOS POST, but the kernel
> starts measuring the time from its boot. If we start measuruing
> I915_PMU_RC6_RESIDENCY while the GT is idle, we start our sampling from
> 0 and then upon first activity (park/unpark) add in all the rc6
> residency since boot. After the first park with the sampler engaged, the
> sleep/active counters are aligned.
> 
> v2: With a wakeref to be sure
> 
> Fixes: df6a42053513 ("drm/i915/pmu: Ensure monotonic rc6")
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
> ---
>  drivers/gpu/drm/i915/i915_pmu.c | 12 ++++++++++++
>  1 file changed, 12 insertions(+)
> 
> diff --git a/drivers/gpu/drm/i915/i915_pmu.c b/drivers/gpu/drm/i915/i915_pmu.c
> index 28a82c849bac..ec0299490dd4 100644
> --- a/drivers/gpu/drm/i915/i915_pmu.c
> +++ b/drivers/gpu/drm/i915/i915_pmu.c
> @@ -637,8 +637,10 @@ static void i915_pmu_enable(struct perf_event *event)
>                 container_of(event->pmu, typeof(*i915), pmu.base);
>         unsigned int bit = event_enabled_bit(event);
>         struct i915_pmu *pmu = &i915->pmu;
> +       intel_wakeref_t wakeref;
>         unsigned long flags;
>  
> +       wakeref = intel_runtime_pm_get(&i915->runtime_pm);
>         spin_lock_irqsave(&pmu->lock, flags);
>  
>         /*
> @@ -648,6 +650,14 @@ static void i915_pmu_enable(struct perf_event *event)
>         BUILD_BUG_ON(ARRAY_SIZE(pmu->enable_count) != I915_PMU_MASK_BITS);
>         GEM_BUG_ON(bit >= ARRAY_SIZE(pmu->enable_count));
>         GEM_BUG_ON(pmu->enable_count[bit] == ~0);
> +
> +       if (pmu->enable_count[bit] == 0 &&
> +           config_enabled_mask(I915_PMU_RC6_RESIDENCY) & BIT_ULL(bit)) {
> +               pmu->sample[__I915_SAMPLE_RC6_LAST_REPORTED].cur = 0;

I can't decide if it's better to have discrete sampling appear
monotonic, or to reset just in case we drifted far off.
-Chris
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [Intel-gfx] [PATCH 2/2] drm/i915/gt: Clear rc6 residency trackers across suspend

[Tvrtko Ursulin]

On 14/01/2020 10:56, Chris Wilson wrote:
> On suspend, the rc6 residency counters (stored in HW registers) will be
> lost and cleared. However, we keep track of the rc6 residency to provide
> a continuous 64b sampling, and if we see the HW value go backwards, we
> assume it overflowed and add on 32b/40b -- an interesting artifact when
> sampling across suspend.
> 
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
> ---
>   drivers/gpu/drm/i915/gt/intel_rc6.c | 2 ++
>   1 file changed, 2 insertions(+)
> 
> diff --git a/drivers/gpu/drm/i915/gt/intel_rc6.c b/drivers/gpu/drm/i915/gt/intel_rc6.c
> index 9e303c29d6e3..04eeb7740e53 100644
> --- a/drivers/gpu/drm/i915/gt/intel_rc6.c
> +++ b/drivers/gpu/drm/i915/gt/intel_rc6.c
> @@ -542,6 +542,8 @@ void intel_rc6_init(struct intel_rc6 *rc6)
>   
>   void intel_rc6_sanitize(struct intel_rc6 *rc6)
>   {
> +	memset(rc6->prev_hw_residency, 0, sizeof(rc6->prev_hw_residency));
> +
>   	if (rc6->enabled) { /* unbalanced suspend/resume */
>   		rpm_get(rc6);
>   		rc6->enabled = false;
> 

Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>

Regards,

Tvrtko
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [Intel-gfx] [PATCH 1/2] drm/i915/pmu: Correct the rc6 offset upon enabling

[Tvrtko Ursulin]

On 14/01/2020 11:06, Chris Wilson wrote:
> Quoting Chris Wilson (2020-01-14 10:56:47)
>> The rc6 residency starts ticking from 0 from BIOS POST, but the kernel
>> starts measuring the time from its boot. If we start measuruing
>> I915_PMU_RC6_RESIDENCY while the GT is idle, we start our sampling from
>> 0 and then upon first activity (park/unpark) add in all the rc6
>> residency since boot. After the first park with the sampler engaged, the
>> sleep/active counters are aligned.
>>
>> v2: With a wakeref to be sure
>>
>> Fixes: df6a42053513 ("drm/i915/pmu: Ensure monotonic rc6")
>> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
>> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>> ---
>>   drivers/gpu/drm/i915/i915_pmu.c | 12 ++++++++++++
>>   1 file changed, 12 insertions(+)
>>
>> diff --git a/drivers/gpu/drm/i915/i915_pmu.c b/drivers/gpu/drm/i915/i915_pmu.c
>> index 28a82c849bac..ec0299490dd4 100644
>> --- a/drivers/gpu/drm/i915/i915_pmu.c
>> +++ b/drivers/gpu/drm/i915/i915_pmu.c
>> @@ -637,8 +637,10 @@ static void i915_pmu_enable(struct perf_event *event)
>>                  container_of(event->pmu, typeof(*i915), pmu.base);
>>          unsigned int bit = event_enabled_bit(event);
>>          struct i915_pmu *pmu = &i915->pmu;
>> +       intel_wakeref_t wakeref;
>>          unsigned long flags;
>>   
>> +       wakeref = intel_runtime_pm_get(&i915->runtime_pm);

I think it would be nicer to use with_intel_runtime_pm directly at the 
__get_rc6 call site. That would show/localise where it is actually needed.

>>          spin_lock_irqsave(&pmu->lock, flags);
>>   
>>          /*
>> @@ -648,6 +650,14 @@ static void i915_pmu_enable(struct perf_event *event)
>>          BUILD_BUG_ON(ARRAY_SIZE(pmu->enable_count) != I915_PMU_MASK_BITS);
>>          GEM_BUG_ON(bit >= ARRAY_SIZE(pmu->enable_count));
>>          GEM_BUG_ON(pmu->enable_count[bit] == ~0);
>> +
>> +       if (pmu->enable_count[bit] == 0 &&
>> +           config_enabled_mask(I915_PMU_RC6_RESIDENCY) & BIT_ULL(bit)) {
>> +               pmu->sample[__I915_SAMPLE_RC6_LAST_REPORTED].cur = 0;
> 
> I can't decide if it's better to have discrete sampling appear
> monotonic, or to reset just in case we drifted far off.

What do you mean?

This looks correct to me as you implemented it. On enable it samples the 
real RC6 and updates pmu->sleep_last. So regardless if the next even 
read comes with device awake or suspended it will report monotonic and 
without adding up any time outside the enabled window.

Drift can normally come when we overestimate because hw RC6 can be less 
than our time between park and unpark. I don't see how to reset that and 
stay monotonic. Or you are thinking it doesn't need to be monotonic?

Regards,

Tvrtko
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

[Intel-gfx] ✓ Fi.CI.BAT: success for series starting with [1/2] drm/i915/pmu: Correct the rc6 offset upon enabling

[Patchwork]

== Series Details ==

Series: series starting with [1/2] drm/i915/pmu: Correct the rc6 offset upon enabling
URL   : https://patchwork.freedesktop.org/series/72001/
State : success

== Summary ==

CI Bug Log - changes from CI_DRM_7737 -> Patchwork_16090
====================================================

Summary
-------

  **SUCCESS**

  No regressions found.

  External URL: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/index.html

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

  Here are the changes found in Patchwork_16090 that come from known issues:

### IGT changes ###

#### Issues hit ####

  * igt@i915_module_load@reload-with-fault-injection:
    - fi-icl-guc:         [PASS][1] -> [DMESG-WARN][2] ([i915#109])
   [1]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/fi-icl-guc/igt@i915_module_load@reload-with-fault-injection.html
   [2]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/fi-icl-guc/igt@i915_module_load@reload-with-fault-injection.html

  * igt@i915_selftest@live_coherency:
    - fi-cfl-guc:         [PASS][3] -> [DMESG-FAIL][4] ([i915#889]) +7 similar issues
   [3]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/fi-cfl-guc/igt@i915_selftest@live_coherency.html
   [4]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/fi-cfl-guc/igt@i915_selftest@live_coherency.html

  * igt@i915_selftest@live_gt_timelines:
    - fi-cfl-guc:         [PASS][5] -> [DMESG-WARN][6] ([i915#889]) +23 similar issues
   [5]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/fi-cfl-guc/igt@i915_selftest@live_gt_timelines.html
   [6]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/fi-cfl-guc/igt@i915_selftest@live_gt_timelines.html

  * igt@kms_busy@basic-flip-pipe-a:
    - fi-icl-u2:          [PASS][7] -> [INCOMPLETE][8] ([i915#140])
   [7]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/fi-icl-u2/igt@kms_busy@basic-flip-pipe-a.html
   [8]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/fi-icl-u2/igt@kms_busy@basic-flip-pipe-a.html

  
#### Possible fixes ####

  * igt@gem_exec_fence@basic-wait-default:
    - {fi-ehl-1}:         [INCOMPLETE][9] ([i915#937]) -> [PASS][10]
   [9]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/fi-ehl-1/igt@gem_exec_fence@basic-wait-default.html
   [10]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/fi-ehl-1/igt@gem_exec_fence@basic-wait-default.html

  * igt@i915_selftest@live_blt:
    - fi-hsw-4770:        [DMESG-FAIL][11] ([i915#553] / [i915#725]) -> [PASS][12]
   [11]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/fi-hsw-4770/igt@i915_selftest@live_blt.html
   [12]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/fi-hsw-4770/igt@i915_selftest@live_blt.html

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

  [i915#109]: https://gitlab.freedesktop.org/drm/intel/issues/109
  [i915#140]: https://gitlab.freedesktop.org/drm/intel/issues/140
  [i915#553]: https://gitlab.freedesktop.org/drm/intel/issues/553
  [i915#725]: https://gitlab.freedesktop.org/drm/intel/issues/725
  [i915#889]: https://gitlab.freedesktop.org/drm/intel/issues/889
  [i915#937]: https://gitlab.freedesktop.org/drm/intel/issues/937


Participating hosts (40 -> 42)
------------------------------

  Additional (12): fi-hsw-4770r fi-hsw-peppy fi-skl-6770hq fi-bwr-2160 fi-snb-2520m fi-ilk-650 fi-kbl-7500u fi-gdg-551 fi-ivb-3770 fi-blb-e6850 fi-tgl-y fi-skl-6600u 
  Missing    (10): fi-bdw-5557u fi-hsw-4200u fi-byt-squawks fi-bsw-cyan fi-ctg-p8600 fi-skl-lmem fi-byt-n2820 fi-byt-clapper fi-bsw-nick fi-snb-2600 


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

  * CI: CI-20190529 -> None
  * Linux: CI_DRM_7737 -> Patchwork_16090

  CI-20190529: 20190529
  CI_DRM_7737: 2a331333791d2e499ac843e1dc25cd8ea5bdc81f @ git://anongit.freedesktop.org/gfx-ci/linux
  IGT_5364: b7cb6ffdb65cbd233f5ddee2f2dabf97b34fa640 @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
  Patchwork_16090: 126fa2e67360c681eea5ba6af79582bc95227e52 @ git://anongit.freedesktop.org/gfx-ci/linux


== Kernel 32bit build ==

Warning: Kernel 32bit buildtest failed:
https://intel-gfx-ci.01.org/Patchwork_16090/build_32bit.log

  CALL    scripts/checksyscalls.sh
  CALL    scripts/atomic/check-atomics.sh
  CHK     include/generated/compile.h
Kernel: arch/x86/boot/bzImage is ready  (#1)
  Building modules, stage 2.
  MODPOST 122 modules
ERROR: "__udivdi3" [drivers/gpu/drm/amd/amdgpu/amdgpu.ko] undefined!
scripts/Makefile.modpost:93: recipe for target '__modpost' failed
make[1]: *** [__modpost] Error 1
Makefile:1282: recipe for target 'modules' failed
make: *** [modules] Error 2


== Linux commits ==

126fa2e67360 drm/i915/gt: Clear rc6 residency trackers across suspend
c8f4c2b3a1c8 drm/i915/pmu: Correct the rc6 offset upon enabling

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/index.html
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

[Intel-gfx] ✗ Fi.CI.BUILD: warning for series starting with [1/2] drm/i915/pmu: Correct the rc6 offset upon enabling

[Patchwork]

== Series Details ==

Series: series starting with [1/2] drm/i915/pmu: Correct the rc6 offset upon enabling
URL   : https://patchwork.freedesktop.org/series/72001/
State : warning

== Summary ==

CALL    scripts/checksyscalls.sh
  CALL    scripts/atomic/check-atomics.sh
  CHK     include/generated/compile.h
Kernel: arch/x86/boot/bzImage is ready  (#1)
  Building modules, stage 2.
  MODPOST 122 modules
ERROR: "__udivdi3" [drivers/gpu/drm/amd/amdgpu/amdgpu.ko] undefined!
scripts/Makefile.modpost:93: recipe for target '__modpost' failed
make[1]: *** [__modpost] Error 1
Makefile:1282: recipe for target 'modules' failed
make: *** [modules] Error 2

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/build_32bit.log
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [Intel-gfx] [PATCH 1/2] drm/i915/pmu: Correct the rc6 offset upon enabling

[Tvrtko Ursulin]

On 14/01/2020 11:37, Tvrtko Ursulin wrote:
> 
> On 14/01/2020 11:06, Chris Wilson wrote:
>> Quoting Chris Wilson (2020-01-14 10:56:47)
>>> The rc6 residency starts ticking from 0 from BIOS POST, but the kernel
>>> starts measuring the time from its boot. If we start measuruing
>>> I915_PMU_RC6_RESIDENCY while the GT is idle, we start our sampling from
>>> 0 and then upon first activity (park/unpark) add in all the rc6
>>> residency since boot. After the first park with the sampler engaged, the
>>> sleep/active counters are aligned.
>>>
>>> v2: With a wakeref to be sure
>>>
>>> Fixes: df6a42053513 ("drm/i915/pmu: Ensure monotonic rc6")
>>> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
>>> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>>> ---
>>>   drivers/gpu/drm/i915/i915_pmu.c | 12 ++++++++++++
>>>   1 file changed, 12 insertions(+)
>>>
>>> diff --git a/drivers/gpu/drm/i915/i915_pmu.c 
>>> b/drivers/gpu/drm/i915/i915_pmu.c
>>> index 28a82c849bac..ec0299490dd4 100644
>>> --- a/drivers/gpu/drm/i915/i915_pmu.c
>>> +++ b/drivers/gpu/drm/i915/i915_pmu.c
>>> @@ -637,8 +637,10 @@ static void i915_pmu_enable(struct perf_event 
>>> *event)
>>>                  container_of(event->pmu, typeof(*i915), pmu.base);
>>>          unsigned int bit = event_enabled_bit(event);
>>>          struct i915_pmu *pmu = &i915->pmu;
>>> +       intel_wakeref_t wakeref;
>>>          unsigned long flags;
>>> +       wakeref = intel_runtime_pm_get(&i915->runtime_pm);
> 
> I think it would be nicer to use with_intel_runtime_pm directly at the 
> __get_rc6 call site. That would show/localise where it is actually needed.

Spinlock though, my bad.

Regards,

Tvrtko
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [Intel-gfx] [PATCH 1/2] drm/i915/pmu: Correct the rc6 offset upon enabling

[Chris Wilson]

Quoting Tvrtko Ursulin (2020-01-14 11:37:09)
> 
> On 14/01/2020 11:06, Chris Wilson wrote:
> > Quoting Chris Wilson (2020-01-14 10:56:47)
> >> The rc6 residency starts ticking from 0 from BIOS POST, but the kernel
> >> starts measuring the time from its boot. If we start measuruing
> >> I915_PMU_RC6_RESIDENCY while the GT is idle, we start our sampling from
> >> 0 and then upon first activity (park/unpark) add in all the rc6
> >> residency since boot. After the first park with the sampler engaged, the
> >> sleep/active counters are aligned.
> >>
> >> v2: With a wakeref to be sure
> >>
> >> Fixes: df6a42053513 ("drm/i915/pmu: Ensure monotonic rc6")
> >> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> >> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
> >> ---
> >>   drivers/gpu/drm/i915/i915_pmu.c | 12 ++++++++++++
> >>   1 file changed, 12 insertions(+)
> >>
> >> diff --git a/drivers/gpu/drm/i915/i915_pmu.c b/drivers/gpu/drm/i915/i915_pmu.c
> >> index 28a82c849bac..ec0299490dd4 100644
> >> --- a/drivers/gpu/drm/i915/i915_pmu.c
> >> +++ b/drivers/gpu/drm/i915/i915_pmu.c
> >> @@ -637,8 +637,10 @@ static void i915_pmu_enable(struct perf_event *event)
> >>                  container_of(event->pmu, typeof(*i915), pmu.base);
> >>          unsigned int bit = event_enabled_bit(event);
> >>          struct i915_pmu *pmu = &i915->pmu;
> >> +       intel_wakeref_t wakeref;
> >>          unsigned long flags;
> >>   
> >> +       wakeref = intel_runtime_pm_get(&i915->runtime_pm);
> 
> I think it would be nicer to use with_intel_runtime_pm directly at the 
> __get_rc6 call site. That would show/localise where it is actually needed.

We can't, as it gets called under the spinlock :(

And I don't see a way around that, as we require the fixup to be applied
while idle and so require the wakeref.

> >>          spin_lock_irqsave(&pmu->lock, flags);
> >>   
> >>          /*
> >> @@ -648,6 +650,14 @@ static void i915_pmu_enable(struct perf_event *event)
> >>          BUILD_BUG_ON(ARRAY_SIZE(pmu->enable_count) != I915_PMU_MASK_BITS);
> >>          GEM_BUG_ON(bit >= ARRAY_SIZE(pmu->enable_count));
> >>          GEM_BUG_ON(pmu->enable_count[bit] == ~0);
> >> +
> >> +       if (pmu->enable_count[bit] == 0 &&
> >> +           config_enabled_mask(I915_PMU_RC6_RESIDENCY) & BIT_ULL(bit)) {
> >> +               pmu->sample[__I915_SAMPLE_RC6_LAST_REPORTED].cur = 0;
> > 
> > I can't decide if it's better to have discrete sampling appear
> > monotonic, or to reset just in case we drifted far off.
> 
> What do you mean?
> 
> This looks correct to me as you implemented it. On enable it samples the 
> real RC6 and updates pmu->sleep_last. So regardless if the next even 
> read comes with device awake or suspended it will report monotonic and 
> without adding up any time outside the enabled window.

u64 sample[2];

fd = perf_open(RC6);
sample[0] = read(fd);
close(fd);

fd = perf_open(RC6);
sample[1] = read(fd);
close(fd);

/* assume idle system */
assert(sample[1] > sample[0]);

Do we want that? I don't think that's required by the perf API, as the
counters are only valid while the event is enabled (iirc).

> Drift can normally come when we overestimate because hw RC6 can be less 
> than our time between park and unpark. I don't see how to reset that and 
> stay monotonic. Or you are thinking it doesn't need to be monotonic?

I was mostly thinking of bugs, e.g. across suspend. We also have a problem
if we wait longer than 2*counter_wrap between perf_event_reads, and we
probably should install a very lax timer for rc6 to ensure the
__get_rc6() calls remain monotonic. There was a fdo bug for that :)
-Chris
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [Intel-gfx] [PATCH 1/2] drm/i915/pmu: Correct the rc6 offset upon enabling

[Tvrtko Ursulin]

On 14/01/2020 11:49, Chris Wilson wrote:
> Quoting Tvrtko Ursulin (2020-01-14 11:37:09)
>> On 14/01/2020 11:06, Chris Wilson wrote:
>>> Quoting Chris Wilson (2020-01-14 10:56:47)
>>>> The rc6 residency starts ticking from 0 from BIOS POST, but the kernel
>>>> starts measuring the time from its boot. If we start measuruing
>>>> I915_PMU_RC6_RESIDENCY while the GT is idle, we start our sampling from
>>>> 0 and then upon first activity (park/unpark) add in all the rc6
>>>> residency since boot. After the first park with the sampler engaged, the
>>>> sleep/active counters are aligned.
>>>>
>>>> v2: With a wakeref to be sure
>>>>
>>>> Fixes: df6a42053513 ("drm/i915/pmu: Ensure monotonic rc6")
>>>> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
>>>> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>>>> ---
>>>>    drivers/gpu/drm/i915/i915_pmu.c | 12 ++++++++++++
>>>>    1 file changed, 12 insertions(+)
>>>>
>>>> diff --git a/drivers/gpu/drm/i915/i915_pmu.c b/drivers/gpu/drm/i915/i915_pmu.c
>>>> index 28a82c849bac..ec0299490dd4 100644
>>>> --- a/drivers/gpu/drm/i915/i915_pmu.c
>>>> +++ b/drivers/gpu/drm/i915/i915_pmu.c
>>>> @@ -637,8 +637,10 @@ static void i915_pmu_enable(struct perf_event *event)
>>>>                   container_of(event->pmu, typeof(*i915), pmu.base);
>>>>           unsigned int bit = event_enabled_bit(event);
>>>>           struct i915_pmu *pmu = &i915->pmu;
>>>> +       intel_wakeref_t wakeref;
>>>>           unsigned long flags;
>>>>    
>>>> +       wakeref = intel_runtime_pm_get(&i915->runtime_pm);
>>
>> I think it would be nicer to use with_intel_runtime_pm directly at the
>> __get_rc6 call site. That would show/localise where it is actually needed.
> 
> We can't, as it gets called under the spinlock :(
> 
> And I don't see a way around that, as we require the fixup to be applied
> while idle and so require the wakeref.
> 
>>>>           spin_lock_irqsave(&pmu->lock, flags);
>>>>    
>>>>           /*
>>>> @@ -648,6 +650,14 @@ static void i915_pmu_enable(struct perf_event *event)
>>>>           BUILD_BUG_ON(ARRAY_SIZE(pmu->enable_count) != I915_PMU_MASK_BITS);
>>>>           GEM_BUG_ON(bit >= ARRAY_SIZE(pmu->enable_count));
>>>>           GEM_BUG_ON(pmu->enable_count[bit] == ~0);
>>>> +
>>>> +       if (pmu->enable_count[bit] == 0 &&
>>>> +           config_enabled_mask(I915_PMU_RC6_RESIDENCY) & BIT_ULL(bit)) {
>>>> +               pmu->sample[__I915_SAMPLE_RC6_LAST_REPORTED].cur = 0;
>>>
>>> I can't decide if it's better to have discrete sampling appear
>>> monotonic, or to reset just in case we drifted far off.
>>
>> What do you mean?
>>
>> This looks correct to me as you implemented it. On enable it samples the
>> real RC6 and updates pmu->sleep_last. So regardless if the next even
>> read comes with device awake or suspended it will report monotonic and
>> without adding up any time outside the enabled window.
> 
> u64 sample[2];
> 
> fd = perf_open(RC6);
> sample[0] = read(fd);
> close(fd);
> 
> fd = perf_open(RC6);
> sample[1] = read(fd);
> close(fd);
> 
> /* assume idle system */
> assert(sample[1] > sample[0]);
> 
> Do we want that? I don't think that's required by the perf API, as the
> counters are only valid while the event is enabled (iirc).

Yeah I'd say this is not interesting.

> 
>> Drift can normally come when we overestimate because hw RC6 can be less
>> than our time between park and unpark. I don't see how to reset that and
>> stay monotonic. Or you are thinking it doesn't need to be monotonic?
> 
> I was mostly thinking of bugs, e.g. across suspend. We also have a problem
> if we wait longer than 2*counter_wrap between perf_event_reads, and we
> probably should install a very lax timer for rc6 to ensure the
> __get_rc6() calls remain monotonic. There was a fdo bug for that :)

Yes I remember now. With 1.28us tick hw overflow is ~5500 seconds so 
perf_event_read every ~3hrs is enough to keep outside overflow. It's 
probably enough. :)

For the patch:

Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>

Regards,

Tvrtko



_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

[Intel-gfx] ✓ Fi.CI.IGT: success for series starting with [1/2] drm/i915/pmu: Correct the rc6 offset upon enabling

[Patchwork]

== Series Details ==

Series: series starting with [1/2] drm/i915/pmu: Correct the rc6 offset upon enabling
URL   : https://patchwork.freedesktop.org/series/72001/
State : success

== Summary ==

CI Bug Log - changes from CI_DRM_7737_full -> Patchwork_16090_full
====================================================

Summary
-------

  **SUCCESS**

  No regressions found.

  

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

  Here are the changes found in Patchwork_16090_full that come from known issues:

### IGT changes ###

#### Issues hit ####

  * igt@gem_ctx_persistence@bcs0-mixed-process:
    - shard-glk:          [PASS][1] -> [FAIL][2] ([i915#679])
   [1]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-glk9/igt@gem_ctx_persistence@bcs0-mixed-process.html
   [2]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-glk4/igt@gem_ctx_persistence@bcs0-mixed-process.html

  * igt@gem_ctx_persistence@vcs1-mixed-process:
    - shard-iclb:         [PASS][3] -> [SKIP][4] ([fdo#109276] / [fdo#112080]) +2 similar issues
   [3]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-iclb4/igt@gem_ctx_persistence@vcs1-mixed-process.html
   [4]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-iclb3/igt@gem_ctx_persistence@vcs1-mixed-process.html

  * igt@gem_ctx_shared@exec-single-timeline-bsd:
    - shard-iclb:         [PASS][5] -> [SKIP][6] ([fdo#110841])
   [5]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-iclb6/igt@gem_ctx_shared@exec-single-timeline-bsd.html
   [6]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-iclb1/igt@gem_ctx_shared@exec-single-timeline-bsd.html

  * igt@gem_exec_gttfill@basic:
    - shard-tglb:         [PASS][7] -> [INCOMPLETE][8] ([fdo#111593] / [i915#472])
   [7]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-tglb4/igt@gem_exec_gttfill@basic.html
   [8]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-tglb7/igt@gem_exec_gttfill@basic.html

  * igt@gem_exec_parallel@vcs1-fds:
    - shard-iclb:         [PASS][9] -> [SKIP][10] ([fdo#112080]) +6 similar issues
   [9]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-iclb1/igt@gem_exec_parallel@vcs1-fds.html
   [10]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-iclb5/igt@gem_exec_parallel@vcs1-fds.html

  * igt@gem_exec_schedule@pi-userfault-bsd:
    - shard-iclb:         [PASS][11] -> [SKIP][12] ([i915#677])
   [11]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-iclb5/igt@gem_exec_schedule@pi-userfault-bsd.html
   [12]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-iclb2/igt@gem_exec_schedule@pi-userfault-bsd.html

  * igt@gem_exec_schedule@reorder-wide-bsd:
    - shard-iclb:         [PASS][13] -> [SKIP][14] ([fdo#112146]) +8 similar issues
   [13]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-iclb5/igt@gem_exec_schedule@reorder-wide-bsd.html
   [14]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-iclb2/igt@gem_exec_schedule@reorder-wide-bsd.html

  * igt@gem_exec_schedule@smoketest-bsd2:
    - shard-tglb:         [PASS][15] -> [INCOMPLETE][16] ([i915#472] / [i915#707])
   [15]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-tglb8/igt@gem_exec_schedule@smoketest-bsd2.html
   [16]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-tglb6/igt@gem_exec_schedule@smoketest-bsd2.html

  * igt@gem_ppgtt@flink-and-close-vma-leak:
    - shard-glk:          [PASS][17] -> [FAIL][18] ([i915#644])
   [17]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-glk7/igt@gem_ppgtt@flink-and-close-vma-leak.html
   [18]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-glk2/igt@gem_ppgtt@flink-and-close-vma-leak.html

  * igt@i915_pm_rps@min-max-config-loaded:
    - shard-apl:          [PASS][19] -> [FAIL][20] ([i915#39])
   [19]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-apl2/igt@i915_pm_rps@min-max-config-loaded.html
   [20]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-apl8/igt@i915_pm_rps@min-max-config-loaded.html

  * igt@i915_suspend@forcewake:
    - shard-snb:          [PASS][21] -> [DMESG-WARN][22] ([i915#42])
   [21]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-snb2/igt@i915_suspend@forcewake.html
   [22]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-snb4/igt@i915_suspend@forcewake.html

  * igt@kms_color@pipe-a-ctm-negative:
    - shard-skl:          [PASS][23] -> [DMESG-WARN][24] ([i915#109]) +1 similar issue
   [23]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-skl2/igt@kms_color@pipe-a-ctm-negative.html
   [24]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-skl6/igt@kms_color@pipe-a-ctm-negative.html

  * igt@kms_cursor_crc@pipe-b-cursor-128x128-rapid-movement:
    - shard-snb:          [PASS][25] -> [SKIP][26] ([fdo#109271]) +3 similar issues
   [25]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-snb4/igt@kms_cursor_crc@pipe-b-cursor-128x128-rapid-movement.html
   [26]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-snb6/igt@kms_cursor_crc@pipe-b-cursor-128x128-rapid-movement.html

  * igt@kms_flip@flip-vs-suspend-interruptible:
    - shard-apl:          [PASS][27] -> [DMESG-WARN][28] ([i915#180]) +3 similar issues
   [27]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-apl3/igt@kms_flip@flip-vs-suspend-interruptible.html
   [28]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-apl1/igt@kms_flip@flip-vs-suspend-interruptible.html

  * igt@kms_frontbuffer_tracking@fbcpsr-1p-primscrn-pri-shrfb-draw-pwrite:
    - shard-tglb:         [PASS][29] -> [FAIL][30] ([i915#49]) +1 similar issue
   [29]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-tglb4/igt@kms_frontbuffer_tracking@fbcpsr-1p-primscrn-pri-shrfb-draw-pwrite.html
   [30]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-tglb8/igt@kms_frontbuffer_tracking@fbcpsr-1p-primscrn-pri-shrfb-draw-pwrite.html

  * igt@kms_plane@plane-panning-bottom-right-suspend-pipe-c-planes:
    - shard-kbl:          [PASS][31] -> [DMESG-WARN][32] ([i915#180]) +1 similar issue
   [31]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-kbl2/igt@kms_plane@plane-panning-bottom-right-suspend-pipe-c-planes.html
   [32]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-kbl6/igt@kms_plane@plane-panning-bottom-right-suspend-pipe-c-planes.html

  * igt@kms_psr2_su@frontbuffer:
    - shard-iclb:         [PASS][33] -> [SKIP][34] ([fdo#109642] / [fdo#111068])
   [33]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-iclb2/igt@kms_psr2_su@frontbuffer.html
   [34]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-iclb4/igt@kms_psr2_su@frontbuffer.html

  * igt@kms_psr@psr2_primary_mmap_cpu:
    - shard-iclb:         [PASS][35] -> [SKIP][36] ([fdo#109441])
   [35]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-iclb2/igt@kms_psr@psr2_primary_mmap_cpu.html
   [36]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-iclb1/igt@kms_psr@psr2_primary_mmap_cpu.html

  * igt@kms_setmode@basic:
    - shard-hsw:          [PASS][37] -> [FAIL][38] ([i915#31])
   [37]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-hsw2/igt@kms_setmode@basic.html
   [38]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-hsw7/igt@kms_setmode@basic.html

  * igt@prime_vgem@fence-wait-bsd2:
    - shard-iclb:         [PASS][39] -> [SKIP][40] ([fdo#109276]) +23 similar issues
   [39]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-iclb1/igt@prime_vgem@fence-wait-bsd2.html
   [40]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-iclb5/igt@prime_vgem@fence-wait-bsd2.html

  
#### Possible fixes ####

  * igt@gem_busy@busy-vcs1:
    - shard-iclb:         [SKIP][41] ([fdo#112080]) -> [PASS][42] +16 similar issues
   [41]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-iclb8/igt@gem_busy@busy-vcs1.html
   [42]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-iclb4/igt@gem_busy@busy-vcs1.html

  * igt@gem_ctx_isolation@bcs0-s3:
    - shard-apl:          [DMESG-WARN][43] ([i915#180]) -> [PASS][44] +3 similar issues
   [43]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-apl1/igt@gem_ctx_isolation@bcs0-s3.html
   [44]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-apl4/igt@gem_ctx_isolation@bcs0-s3.html

  * igt@gem_ctx_isolation@vecs0-s3:
    - shard-iclb:         [DMESG-WARN][45] ([fdo#111764]) -> [PASS][46]
   [45]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-iclb8/igt@gem_ctx_isolation@vecs0-s3.html
   [46]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-iclb4/igt@gem_ctx_isolation@vecs0-s3.html

  * igt@gem_ctx_persistence@vcs1-queued:
    - shard-iclb:         [SKIP][47] ([fdo#109276] / [fdo#112080]) -> [PASS][48] +4 similar issues
   [47]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-iclb8/igt@gem_ctx_persistence@vcs1-queued.html
   [48]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-iclb4/igt@gem_ctx_persistence@vcs1-queued.html

  * igt@gem_exec_nop@basic-sequential:
    - shard-tglb:         [INCOMPLETE][49] ([i915#472]) -> [PASS][50]
   [49]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-tglb5/igt@gem_exec_nop@basic-sequential.html
   [50]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-tglb1/igt@gem_exec_nop@basic-sequential.html

  * igt@gem_exec_schedule@pi-distinct-iova-bsd:
    - shard-iclb:         [SKIP][51] ([i915#677]) -> [PASS][52] +1 similar issue
   [51]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-iclb2/igt@gem_exec_schedule@pi-distinct-iova-bsd.html
   [52]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-iclb8/igt@gem_exec_schedule@pi-distinct-iova-bsd.html

  * igt@gem_exec_schedule@preempt-queue-contexts-chain-bsd:
    - shard-iclb:         [SKIP][53] ([fdo#112146]) -> [PASS][54] +3 similar issues
   [53]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-iclb1/igt@gem_exec_schedule@preempt-queue-contexts-chain-bsd.html
   [54]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-iclb5/igt@gem_exec_schedule@preempt-queue-contexts-chain-bsd.html

  * igt@gem_exec_suspend@basic-s3:
    - shard-tglb:         [INCOMPLETE][55] ([fdo#111736] / [i915#460] / [i915#472]) -> [PASS][56]
   [55]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-tglb3/igt@gem_exec_suspend@basic-s3.html
   [56]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-tglb8/igt@gem_exec_suspend@basic-s3.html

  * igt@gem_persistent_relocs@forked-interruptible-faulting-reloc-thrash-inactive:
    - shard-glk:          [TIMEOUT][57] ([fdo#112271] / [i915#530]) -> [PASS][58]
   [57]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-glk9/igt@gem_persistent_relocs@forked-interruptible-faulting-reloc-thrash-inactive.html
   [58]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-glk8/igt@gem_persistent_relocs@forked-interruptible-faulting-reloc-thrash-inactive.html

  * igt@gem_pipe_control_store_loop@reused-buffer:
    - shard-tglb:         [INCOMPLETE][59] ([i915#707] / [i915#796]) -> [PASS][60]
   [59]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-tglb8/igt@gem_pipe_control_store_loop@reused-buffer.html
   [60]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-tglb1/igt@gem_pipe_control_store_loop@reused-buffer.html

  * igt@gem_sync@basic-many-each:
    - shard-tglb:         [INCOMPLETE][61] ([i915#472] / [i915#707]) -> [PASS][62]
   [61]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-tglb3/igt@gem_sync@basic-many-each.html
   [62]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-tglb9/igt@gem_sync@basic-many-each.html

  * igt@gem_tiled_blits@normal:
    - shard-tglb:         [INCOMPLETE][63] -> [PASS][64]
   [63]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-tglb2/igt@gem_tiled_blits@normal.html
   [64]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-tglb4/igt@gem_tiled_blits@normal.html

  * igt@i915_pm_dc@dc5-dpms:
    - shard-iclb:         [FAIL][65] ([i915#447]) -> [PASS][66]
   [65]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-iclb3/igt@i915_pm_dc@dc5-dpms.html
   [66]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-iclb6/igt@i915_pm_dc@dc5-dpms.html

  * {igt@i915_pm_rc6_residency@rc6-idle}:
    - shard-skl:          [FAIL][67] ([i915#973]) -> [PASS][68]
   [67]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-skl8/igt@i915_pm_rc6_residency@rc6-idle.html
   [68]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-skl5/igt@i915_pm_rc6_residency@rc6-idle.html
    - shard-tglb:         [FAIL][69] ([i915#973]) -> [PASS][70]
   [69]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-tglb5/igt@i915_pm_rc6_residency@rc6-idle.html
   [70]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-tglb7/igt@i915_pm_rc6_residency@rc6-idle.html
    - shard-apl:          [FAIL][71] ([i915#973]) -> [PASS][72]
   [71]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-apl7/igt@i915_pm_rc6_residency@rc6-idle.html
   [72]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-apl3/igt@i915_pm_rc6_residency@rc6-idle.html
    - shard-glk:          [FAIL][73] ([i915#973]) -> [PASS][74]
   [73]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-glk9/igt@i915_pm_rc6_residency@rc6-idle.html
   [74]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-glk8/igt@i915_pm_rc6_residency@rc6-idle.html
    - shard-kbl:          [FAIL][75] ([i915#973]) -> [PASS][76]
   [75]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-kbl2/igt@i915_pm_rc6_residency@rc6-idle.html
   [76]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-kbl6/igt@i915_pm_rc6_residency@rc6-idle.html
    - shard-iclb:         [FAIL][77] ([i915#973]) -> [PASS][78]
   [77]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-iclb8/igt@i915_pm_rc6_residency@rc6-idle.html
   [78]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-iclb4/igt@i915_pm_rc6_residency@rc6-idle.html

  * igt@i915_pm_rps@reset:
    - shard-iclb:         [FAIL][79] ([i915#413]) -> [PASS][80] +1 similar issue
   [79]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-iclb6/igt@i915_pm_rps@reset.html
   [80]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-iclb1/igt@i915_pm_rps@reset.html

  * igt@kms_color@pipe-a-ctm-green-to-red:
    - shard-skl:          [DMESG-WARN][81] ([i915#109]) -> [PASS][82]
   [81]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-skl8/igt@kms_color@pipe-a-ctm-green-to-red.html
   [82]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-skl7/igt@kms_color@pipe-a-ctm-green-to-red.html

  * igt@kms_flip@flip-vs-expired-vblank:
    - shard-skl:          [FAIL][83] ([i915#79]) -> [PASS][84]
   [83]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-skl1/igt@kms_flip@flip-vs-expired-vblank.html
   [84]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-skl10/igt@kms_flip@flip-vs-expired-vblank.html

  * igt@kms_frontbuffer_tracking@fbcpsr-rgb565-draw-pwrite:
    - shard-tglb:         [FAIL][85] ([i915#49]) -> [PASS][86] +2 similar issues
   [85]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-tglb8/igt@kms_frontbuffer_tracking@fbcpsr-rgb565-draw-pwrite.html
   [86]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-tglb8/igt@kms_frontbuffer_tracking@fbcpsr-rgb565-draw-pwrite.html

  * igt@kms_pipe_crc_basic@nonblocking-crc-pipe-b:
    - shard-snb:          [SKIP][87] ([fdo#109271]) -> [PASS][88] +3 similar issues
   [87]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-snb2/igt@kms_pipe_crc_basic@nonblocking-crc-pipe-b.html
   [88]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-snb2/igt@kms_pipe_crc_basic@nonblocking-crc-pipe-b.html

  * igt@kms_psr@psr2_sprite_plane_move:
    - shard-iclb:         [SKIP][89] ([fdo#109441]) -> [PASS][90] +4 similar issues
   [89]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-iclb4/igt@kms_psr@psr2_sprite_plane_move.html
   [90]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-iclb2/igt@kms_psr@psr2_sprite_plane_move.html

  * igt@kms_setmode@basic:
    - shard-tglb:         [FAIL][91] ([i915#31]) -> [PASS][92]
   [91]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-tglb1/igt@kms_setmode@basic.html
   [92]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-tglb3/igt@kms_setmode@basic.html

  * igt@kms_vblank@pipe-a-ts-continuation-suspend:
    - shard-kbl:          [DMESG-WARN][93] ([i915#180]) -> [PASS][94] +6 similar issues
   [93]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-kbl1/igt@kms_vblank@pipe-a-ts-continuation-suspend.html
   [94]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-kbl1/igt@kms_vblank@pipe-a-ts-continuation-suspend.html

  * igt@prime_busy@hang-bsd2:
    - shard-iclb:         [SKIP][95] ([fdo#109276]) -> [PASS][96] +17 similar issues
   [95]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-iclb6/igt@prime_busy@hang-bsd2.html
   [96]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-iclb4/igt@prime_busy@hang-bsd2.html

  
#### Warnings ####

  * igt@gem_ctx_isolation@vcs1-nonpriv:
    - shard-iclb:         [FAIL][97] ([IGT#28]) -> [SKIP][98] ([fdo#109276] / [fdo#112080])
   [97]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-iclb1/igt@gem_ctx_isolation@vcs1-nonpriv.html
   [98]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-iclb3/igt@gem_ctx_isolation@vcs1-nonpriv.html

  * igt@gem_ctx_isolation@vcs2-dirty-create:
    - shard-tglb:         [SKIP][99] ([fdo#111912] / [fdo#112080]) -> [SKIP][100] ([fdo#112080])
   [99]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-tglb7/igt@gem_ctx_isolation@vcs2-dirty-create.html
   [100]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-tglb9/igt@gem_ctx_isolation@vcs2-dirty-create.html

  * igt@gem_eio@kms:
    - shard-snb:          [DMESG-WARN][101] ([i915#444]) -> [INCOMPLETE][102] ([i915#82])
   [101]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-snb5/igt@gem_eio@kms.html
   [102]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-snb2/igt@gem_eio@kms.html

  * igt@gem_tiled_blits@interruptible:
    - shard-hsw:          [FAIL][103] ([i915#818]) -> [FAIL][104] ([i915#694])
   [103]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-hsw7/igt@gem_tiled_blits@interruptible.html
   [104]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-hsw7/igt@gem_tiled_blits@interruptible.html

  * igt@i915_pm_dc@dc6-psr:
    - shard-tglb:         [FAIL][105] ([i915#454]) -> [SKIP][106] ([i915#468])
   [105]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-tglb7/igt@i915_pm_dc@dc6-psr.html
   [106]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-tglb2/igt@i915_pm_dc@dc6-psr.html

  * igt@kms_atomic_transition@3x-modeset-transitions-nonblocking:
    - shard-snb:          [SKIP][107] ([fdo#109271]) -> [SKIP][108] ([fdo#109271] / [i915#439])
   [107]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-snb2/igt@kms_atomic_transition@3x-modeset-transitions-nonblocking.html
   [108]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-snb2/igt@kms_atomic_transition@3x-modeset-transitions-nonblocking.html

  * igt@kms_ccs@pipe-d-crc-primary-rotation-180:
    - shard-tglb:         [DMESG-FAIL][109] ([i915#402]) -> [DMESG-WARN][110] ([i915#402])
   [109]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-tglb5/igt@kms_ccs@pipe-d-crc-primary-rotation-180.html
   [110]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-tglb7/igt@kms_ccs@pipe-d-crc-primary-rotation-180.html

  * igt@runner@aborted:
    - shard-snb:          ([FAIL][111], [FAIL][112]) ([i915#436] / [i915#974]) -> [FAIL][113] ([i915#974])
   [111]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-snb1/igt@runner@aborted.html
   [112]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7737/shard-snb5/igt@runner@aborted.html
   [113]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/shard-snb5/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).

  [IGT#28]: https://gitlab.freedesktop.org/drm/igt-gpu-tools/issues/28
  [fdo#109271]: https://bugs.freedesktop.org/show_bug.cgi?id=109271
  [fdo#109276]: https://bugs.freedesktop.org/show_bug.cgi?id=109276
  [fdo#109441]: https://bugs.freedesktop.org/show_bug.cgi?id=109441
  [fdo#109642]: https://bugs.freedesktop.org/show_bug.cgi?id=109642
  [fdo#110841]: https://bugs.freedesktop.org/show_bug.cgi?id=110841
  [fdo#111068]: https://bugs.freedesktop.org/show_bug.cgi?id=111068
  [fdo#111593]: https://bugs.freedesktop.org/show_bug.cgi?id=111593
  [fdo#111736]: https://bugs.freedesktop.org/show_bug.cgi?id=111736
  [fdo#111764]: https://bugs.freedesktop.org/show_bug.cgi?id=111764
  [fdo#111912]: https://bugs.freedesktop.org/show_bug.cgi?id=111912
  [fdo#112080]: https://bugs.freedesktop.org/show_bug.cgi?id=112080
  [fdo#112146]: https://bugs.freedesktop.org/show_bug.cgi?id=112146
  [fdo#112271]: https://bugs.freedesktop.org/show_bug.cgi?id=112271
  [i915#109]: https://gitlab.freedesktop.org/drm/intel/issues/109
  [i915#180]: https://gitlab.freedesktop.org/drm/intel/issues/180
  [i915#31]: https://gitlab.freedesktop.org/drm/intel/issues/31
  [i915#39]: https://gitlab.freedesktop.org/drm/intel/issues/39
  [i915#402]: https://gitlab.freedesktop.org/drm/intel/issues/402
  [i915#413]: https://gitlab.freedesktop.org/drm/intel/issues/413
  [i915#42]: https://gitlab.freedesktop.org/drm/intel/issues/42
  [i915#436]: https://gitlab.freedesktop.org/drm/intel/issues/436
  [i915#439]: https://gitlab.freedesktop.org/drm/intel/issues/439
  [i915#444]: https://gitlab.freedesktop.org/drm/intel/issues/444
  [i915#447]: https://gitlab.freedesktop.org/drm/intel/issues/447
  [i915#454]: https://gitlab.freedesktop.org/drm/intel/issues/454
  [i915#460]: https://gitlab.freedesktop.org/drm/intel/issues/460
  [i915#468]: https://gitlab.freedesktop.org/drm/intel/issues/468
  [i915#472]: https://gitlab.freedesktop.org/drm/intel/issues/472
  [i915#49]: https://gitlab.freedesktop.org/drm/intel/issues/49
  [i915#530]: https://gitlab.freedesktop.org/drm/intel/issues/530
  [i915#644]: https://gitlab.freedesktop.org/drm/intel/issues/644
  [i915#677]: https://gitlab.freedesktop.org/drm/intel/issues/677
  [i915#679]: https://gitlab.freedesktop.org/drm/intel/issues/679
  [i915#694]: https://gitlab.freedesktop.org/drm/intel/issues/694
  [i915#707]: https://gitlab.freedesktop.org/drm/intel/issues/707
  [i915#79]: https://gitlab.freedesktop.org/drm/intel/issues/79
  [i915#796]: https://gitlab.freedesktop.org/drm/intel/issues/796
  [i915#818]: https://gitlab.freedesktop.org/drm/intel/issues/818
  [i915#82]: https://gitlab.freedesktop.org/drm/intel/issues/82
  [i915#973]: https://gitlab.freedesktop.org/drm/intel/issues/973
  [i915#974]: https://gitlab.freedesktop.org/drm/intel/issues/974


Participating hosts (11 -> 10)
------------------------------

  Missing    (1): pig-hsw-4770r 


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

  * CI: CI-20190529 -> None
  * Linux: CI_DRM_7737 -> Patchwork_16090

  CI-20190529: 20190529
  CI_DRM_7737: 2a331333791d2e499ac843e1dc25cd8ea5bdc81f @ git://anongit.freedesktop.org/gfx-ci/linux
  IGT_5364: b7cb6ffdb65cbd233f5ddee2f2dabf97b34fa640 @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
  Patchwork_16090: 126fa2e67360c681eea5ba6af79582bc95227e52 @ git://anongit.freedesktop.org/gfx-ci/linux
  piglit_4509: fdc5a4ca11124ab8413c7988896eec4c97336694 @ git://anongit.freedesktop.org/piglit

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16090/index.html
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx