[igt-dev] [PATCH i-g-t v3 00/17] Extend intel_blt to work on Xe

[igt-dev] [PATCH i-g-t v3 00/17] Extend intel_blt to work on Xe

[Zbigniew Kempczyński]

Blitter library currently supports block-copy, ctrl-surf-copy
and fast-copy on i915. Lets extend this to xe as most of the
code is driver independent.

v2: Rewrite tracking allocator calls alloc()/free() to handle
    multiprocess/multithreaded scenarios. api_intel_allocator
    now supports both drivers (i915 and xe).
v3: Address review comments (Karolina)

Cc: Kamil Konieczny <kamil.konieczny@linux.intel.com>
Cc: Karolina Stolarek <karolina.stolarek@intel.com>
Cc: Juha-Pekka Heikkila <juhapekka.heikkila@gmail.com>
Cc: Bhanuprakash Modem <bhanuprakash.modem@intel.com>

Zbigniew Kempczyński (17):
  tests/api_intel_allocator: Don't use allocator ahnd aliasing api
  lib/intel_allocator: Drop aliasing allocator handle api
  lib/intel_allocator: Remove extensive debugging
  lib/xe_ioctl: Export non-assert xe_exec function
  lib/xe_query: Use vramN when returning string region name
  lib/xe_query: Add xe_region_class() helper
  lib/drmtest: Add get_intel_driver() helper
  lib/xe_util: Return dynamic subtest name for Xe
  lib/xe_util: Add vm bind/unbind helper for Xe
  lib/intel_allocator: Add intel_allocator_bind()
  lib/intel_ctx: Add xe context information
  lib/intel_blt: Introduce blt_copy_init() helper to cache driver
  lib/intel_blt: Extend blitter library to support xe driver
  tests/xe_ccs: Check if flatccs is working with block-copy for Xe
  tests/xe_exercise_blt: Check blitter library fast-copy for Xe
  tests/api-intel-allocator: Adopt to exercise allocator to Xe
  tests/gem_ccs: Avoid writing png twice

 lib/drmtest.c                    |  10 +
 lib/drmtest.h                    |   1 +
 lib/igt_core.c                   |   5 +
 lib/igt_fb.c                     |   2 +-
 lib/intel_allocator.c            | 341 ++++++++++----
 lib/intel_allocator.h            |   4 +-
 lib/intel_allocator_msgchannel.h |   8 +-
 lib/intel_blt.c                  | 271 +++++++----
 lib/intel_blt.h                  |  10 +-
 lib/intel_ctx.c                  |  98 +++-
 lib/intel_ctx.h                  |  14 +
 lib/meson.build                  |   3 +-
 lib/xe/xe_ioctl.c                |   2 +-
 lib/xe/xe_ioctl.h                |   1 +
 lib/xe/xe_query.c                |  20 +-
 lib/xe/xe_query.h                |   1 +
 lib/xe/xe_util.c                 | 229 ++++++++++
 lib/xe/xe_util.h                 |  47 ++
 tests/i915/api_intel_allocator.c |  46 +-
 tests/i915/gem_ccs.c             |  35 +-
 tests/i915/gem_exercise_blt.c    |  22 +-
 tests/i915/gem_lmem_swapping.c   |   4 +-
 tests/meson.build                |   2 +
 tests/xe/xe_ccs.c                | 757 +++++++++++++++++++++++++++++++
 tests/xe/xe_exercise_blt.c       | 377 +++++++++++++++
 25 files changed, 2074 insertions(+), 236 deletions(-)
 create mode 100644 lib/xe/xe_util.c
 create mode 100644 lib/xe/xe_util.h
 create mode 100644 tests/xe/xe_ccs.c
 create mode 100644 tests/xe/xe_exercise_blt.c

-- 
2.34.1

[igt-dev] [PATCH i-g-t v3 01/17] tests/api_intel_allocator: Don't use allocator ahnd aliasing api

[Zbigniew Kempczyński]

There's no tests (apart this one) which are using aliasing ahnd
- intel_allocator_open_vm_as(). Additionally it is problematic
on adopting allocator to xe where I need to track allocations
to support easy vm binding. Let's adopt "open-vm" to not to use
this api.

Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
Reviewed-by: Karolina Stolarek <karolina.stolarek@intel.com>
---
 tests/i915/api_intel_allocator.c | 21 ++++++++-------------
 1 file changed, 8 insertions(+), 13 deletions(-)

diff --git a/tests/i915/api_intel_allocator.c b/tests/i915/api_intel_allocator.c
index b7e3efb87f..77c2d25e83 100644
--- a/tests/i915/api_intel_allocator.c
+++ b/tests/i915/api_intel_allocator.c
@@ -612,32 +612,27 @@ static void reopen_fork(int fd)
 
 static void open_vm(int fd)
 {
-	uint64_t ahnd[4], offset[4], size = 0x1000;
+	uint64_t ahnd[3], offset[3], size = 0x1000;
 	int i, n = ARRAY_SIZE(ahnd);
 
 	ahnd[0] = intel_allocator_open_vm(fd, 1, INTEL_ALLOCATOR_SIMPLE);
 	ahnd[1] = intel_allocator_open_vm(fd, 1, INTEL_ALLOCATOR_SIMPLE);
-	ahnd[2] = intel_allocator_open_vm_as(ahnd[1], 2);
-	ahnd[3] = intel_allocator_open(fd, 3, INTEL_ALLOCATOR_SIMPLE);
+	ahnd[2] = intel_allocator_open(fd, 2, INTEL_ALLOCATOR_SIMPLE);
 
 	offset[0] = intel_allocator_alloc(ahnd[0], 1, size, 0);
 	offset[1] = intel_allocator_alloc(ahnd[1], 2, size, 0);
 	igt_assert(offset[0] != offset[1]);
 
-	offset[2] = intel_allocator_alloc(ahnd[2], 3, size, 0);
-	igt_assert(offset[0] != offset[2] && offset[1] != offset[2]);
-
-	offset[3] = intel_allocator_alloc(ahnd[3], 1, size, 0);
-	igt_assert(offset[0] == offset[3]);
+	offset[2] = intel_allocator_alloc(ahnd[2], 1, size, 0);
+	igt_assert(offset[0] == offset[2]);
 
 	/*
-	 * As ahnd[0-2] lead to same allocator check can we free all handles
+	 * As ahnd[0-1] lead to same allocator check we can free all handles
 	 * using selected ahnd.
 	 */
-	intel_allocator_free(ahnd[0], 1);
-	intel_allocator_free(ahnd[0], 2);
-	intel_allocator_free(ahnd[0], 3);
-	intel_allocator_free(ahnd[3], 1);
+	igt_assert_eq(intel_allocator_free(ahnd[0], 1), true);
+	igt_assert_eq(intel_allocator_free(ahnd[1], 2), true);
+	igt_assert_eq(intel_allocator_free(ahnd[2], 1), true);
 
 	for (i = 0; i < n - 1; i++)
 		igt_assert_eq(intel_allocator_close(ahnd[i]), (i == n - 2));
-- 
2.34.1

[igt-dev] [PATCH i-g-t v3 02/17] lib/intel_allocator: Drop aliasing allocator handle api

[Zbigniew Kempczyński]

There's no real user of this api, lets drop it.

Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
--
v3: Remove also ipc message passing for open-as (Karolina)
---
 lib/intel_allocator.c            | 73 --------------------------------
 lib/intel_allocator.h            |  1 -
 lib/intel_allocator_msgchannel.h |  8 +---
 3 files changed, 1 insertion(+), 81 deletions(-)

diff --git a/lib/intel_allocator.c b/lib/intel_allocator.c
index 8161221dbf..09f665fbb9 100644
--- a/lib/intel_allocator.c
+++ b/lib/intel_allocator.c
@@ -25,7 +25,6 @@
 static const char *reqtype_str[] = {
 	[REQ_STOP]		= "stop",
 	[REQ_OPEN]		= "open",
-	[REQ_OPEN_AS]		= "open as",
 	[REQ_CLOSE]		= "close",
 	[REQ_ADDRESS_RANGE]	= "address range",
 	[REQ_ALLOC]		= "alloc",
@@ -376,19 +375,6 @@ static struct allocator *allocator_open(int fd, uint32_t ctx, uint32_t vm,
 	return al;
 }
 
-static struct allocator *allocator_open_as(struct allocator *base,
-					   uint32_t new_vm, uint64_t *ahndp)
-{
-	struct allocator *al;
-
-	igt_assert(ahndp);
-	al = __allocator_create(base->fd, base->ctx, new_vm, base->ial);
-	__allocator_get(al);
-	*ahndp = __handle_create(al);
-
-	return al;
-}
-
 static bool allocator_close(uint64_t ahnd)
 {
 	struct allocator *al;
@@ -512,47 +498,6 @@ static int handle_request(struct alloc_req *req, struct alloc_resp *resp)
 				   refcnt - 1, refcnt, ret - 1, ret);
 			break;
 
-		case REQ_OPEN_AS:
-			/* lock first to avoid concurrent close */
-			pthread_mutex_lock(&map_mutex);
-
-			al = __allocator_find_by_handle(req->allocator_handle);
-			resp->response_type = RESP_OPEN_AS;
-
-			if (!al) {
-				alloc_info("<open as> [tid: %ld] ahnd: %" PRIx64
-					   " -> no handle\n",
-					   (long) req->tid, req->allocator_handle);
-				pthread_mutex_unlock(&map_mutex);
-				break;
-			}
-
-			if (!al->vm) {
-				alloc_info("<open as> [tid: %ld] ahnd: %" PRIx64
-					   " -> only open as for <fd, vm> is possible\n",
-					   (long) req->tid, req->allocator_handle);
-				pthread_mutex_unlock(&map_mutex);
-				break;
-			}
-
-
-			al = allocator_open_as(al, req->open_as.new_vm, &ahnd);
-			refcnt = atomic_load(&al->refcount);
-			ret = atomic_load(&al->ial->refcount);
-			pthread_mutex_unlock(&map_mutex);
-
-			resp->response_type = RESP_OPEN_AS;
-			resp->open.allocator_handle = ahnd;
-
-			alloc_info("<open as> [tid: %ld] fd: %d, ahnd: %" PRIx64
-				   ", ctx: %u, vm: %u"
-				   ", alloc_type: %u, al->refcnt: %ld->%ld"
-				   ", refcnt: %d->%d\n",
-				   (long) req->tid, al->fd, ahnd,
-				   al->ctx, al->vm, al->ial->type,
-				   refcnt - 1, refcnt, ret - 1, ret);
-			break;
-
 		case REQ_CLOSE:
 			pthread_mutex_lock(&map_mutex);
 			al = __allocator_find_by_handle(req->allocator_handle);
@@ -1037,24 +982,6 @@ uint64_t intel_allocator_open_vm(int fd, uint32_t vm, uint8_t allocator_type)
 					    ALLOC_STRATEGY_HIGH_TO_LOW, 0);
 }
 
-uint64_t intel_allocator_open_vm_as(uint64_t allocator_handle, uint32_t new_vm)
-{
-	struct alloc_req req = { .request_type = REQ_OPEN_AS,
-				 .allocator_handle = allocator_handle,
-				 .open_as.new_vm = new_vm };
-	struct alloc_resp resp;
-
-	/* Get child_tid only once at open() */
-	if (child_tid == -1)
-		child_tid = gettid();
-
-	igt_assert(handle_request(&req, &resp) == 0);
-	igt_assert(resp.open_as.allocator_handle);
-	igt_assert(resp.response_type == RESP_OPEN_AS);
-
-	return resp.open.allocator_handle;
-}
-
 /**
  * intel_allocator_close:
  * @allocator_handle: handle to the allocator that will be closed
diff --git a/lib/intel_allocator.h b/lib/intel_allocator.h
index a6bf573e9d..3ec74f6191 100644
--- a/lib/intel_allocator.h
+++ b/lib/intel_allocator.h
@@ -182,7 +182,6 @@ uint64_t intel_allocator_open_vm_full(int fd, uint32_t vm,
 				      enum allocator_strategy strategy,
 				      uint64_t default_alignment);
 
-uint64_t intel_allocator_open_vm_as(uint64_t allocator_handle, uint32_t new_vm);
 bool intel_allocator_close(uint64_t allocator_handle);
 void intel_allocator_get_address_range(uint64_t allocator_handle,
 				       uint64_t *startp, uint64_t *endp);
diff --git a/lib/intel_allocator_msgchannel.h b/lib/intel_allocator_msgchannel.h
index cefbf4ce33..ba38530fd1 100644
--- a/lib/intel_allocator_msgchannel.h
+++ b/lib/intel_allocator_msgchannel.h
@@ -13,7 +13,6 @@
 enum reqtype {
 	REQ_STOP,
 	REQ_OPEN,
-	REQ_OPEN_AS,
 	REQ_CLOSE,
 	REQ_ADDRESS_RANGE,
 	REQ_ALLOC,
@@ -27,7 +26,6 @@ enum reqtype {
 
 enum resptype {
 	RESP_OPEN,
-	RESP_OPEN_AS,
 	RESP_CLOSE,
 	RESP_ADDRESS_RANGE,
 	RESP_ALLOC,
@@ -58,10 +56,6 @@ struct alloc_req {
 			uint64_t default_alignment;
 		} open;
 
-		struct {
-			uint32_t new_vm;
-		} open_as;
-
 		struct {
 			uint32_t handle;
 			uint64_t size;
@@ -100,7 +94,7 @@ struct alloc_resp {
 	union {
 		struct {
 			uint64_t allocator_handle;
-		} open, open_as;
+		} open;
 
 		struct {
 			bool is_empty;
-- 
2.34.1

[igt-dev] [PATCH i-g-t v3 03/17] lib/intel_allocator: Remove extensive debugging

[Zbigniew Kempczyński]

Debugging map keys comparison leads to produce extensive logging what
in turn obscures analysis. Remove it to get clearer logs.

Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
Reviewed-by: Karolina Stolarek <karolina.stolarek@intel.com>
---
 lib/intel_allocator.c | 9 ---------
 1 file changed, 9 deletions(-)

diff --git a/lib/intel_allocator.c b/lib/intel_allocator.c
index 09f665fbb9..cf964ccc41 100644
--- a/lib/intel_allocator.c
+++ b/lib/intel_allocator.c
@@ -1330,9 +1330,6 @@ static int equal_handles(const void *key1, const void *key2)
 {
 	const struct handle_entry *h1 = key1, *h2 = key2;
 
-	alloc_debug("h1: %llx, h2: %llx\n",
-		   (long long) h1->handle, (long long) h2->handle);
-
 	return h1->handle == h2->handle;
 }
 
@@ -1340,9 +1337,6 @@ static int equal_ctx(const void *key1, const void *key2)
 {
 	const struct allocator *a1 = key1, *a2 = key2;
 
-	alloc_debug("a1: <fd: %d, ctx: %u>, a2 <fd: %d, ctx: %u>\n",
-		   a1->fd, a1->ctx, a2->fd, a2->ctx);
-
 	return a1->fd == a2->fd && a1->ctx == a2->ctx;
 }
 
@@ -1350,9 +1344,6 @@ static int equal_vm(const void *key1, const void *key2)
 {
 	const struct allocator *a1 = key1, *a2 = key2;
 
-	alloc_debug("a1: <fd: %d, vm: %u>, a2 <fd: %d, vm: %u>\n",
-		   a1->fd, a1->vm, a2->fd, a2->vm);
-
 	return a1->fd == a2->fd && a1->vm == a2->vm;
 }
 
-- 
2.34.1

[igt-dev] [PATCH i-g-t v3 04/17] lib/xe_ioctl: Export non-assert xe_exec function

[Zbigniew Kempczyński]

Export __xe_exec() helper to allow caller to handle the error
individually.

Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
---
 lib/xe/xe_ioctl.c | 2 +-
 lib/xe/xe_ioctl.h | 1 +
 2 files changed, 2 insertions(+), 1 deletion(-)

diff --git a/lib/xe/xe_ioctl.c b/lib/xe/xe_ioctl.c
index 9ead7307c4..0b5c7a4fce 100644
--- a/lib/xe/xe_ioctl.c
+++ b/lib/xe/xe_ioctl.c
@@ -357,7 +357,7 @@ void *xe_bo_mmap_ext(int fd, uint32_t bo, size_t size, int prot)
 	return __xe_bo_map(fd, bo, size, prot);
 }
 
-static int __xe_exec(int fd, struct drm_xe_exec *exec)
+int __xe_exec(int fd, struct drm_xe_exec *exec)
 {
 	int err = 0;
 
diff --git a/lib/xe/xe_ioctl.h b/lib/xe/xe_ioctl.h
index e014ad66c9..320e0f9f68 100644
--- a/lib/xe/xe_ioctl.h
+++ b/lib/xe/xe_ioctl.h
@@ -75,6 +75,7 @@ void xe_engine_destroy(int fd, uint32_t engine);
 uint64_t xe_bo_mmap_offset(int fd, uint32_t bo);
 void *xe_bo_map(int fd, uint32_t bo, size_t size);
 void *xe_bo_mmap_ext(int fd, uint32_t bo, size_t size, int prot);
+int __xe_exec(int fd, struct drm_xe_exec *exec);
 void xe_exec(int fd, struct drm_xe_exec *exec);
 void xe_exec_sync(int fd, uint32_t engine, uint64_t addr,
 		  struct drm_xe_sync *sync, uint32_t num_syncs);
-- 
2.34.1

[igt-dev] [PATCH i-g-t v3 05/17] lib/xe_query: Use vramN when returning string region name

[Zbigniew Kempczyński]

For tests which are mixing regions (like xe_ccs) name is confusing.
As an example might be example "subtest-name-vram-1-vram-1". It's
more readable when it will be renamed to "subtest-name-vram1-vram1".

Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
Reviewed-by: Karolina Stolarek <karolina.stolarek@intel.com>
---
 lib/xe/xe_query.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/lib/xe/xe_query.c b/lib/xe/xe_query.c
index 48fb5afba7..830b7e401d 100644
--- a/lib/xe/xe_query.c
+++ b/lib/xe/xe_query.c
@@ -445,7 +445,7 @@ struct drm_xe_query_mem_region *xe_mem_region(int fd, uint64_t region)
  * xe_region_name:
  * @region: region mask
  *
- * Returns region string like "system" or "vram-n" where n=0...62.
+ * Returns region string like "system" or "vramN" where N=0...62.
  */
 const char *xe_region_name(uint64_t region)
 {
@@ -457,7 +457,7 @@ const char *xe_region_name(uint64_t region)
 		vrams = calloc(64, sizeof(char *));
 		for (int i = 0; i < 64; i++) {
 			if (i != 0)
-				asprintf(&vrams[i], "vram-%d", i - 1);
+				asprintf(&vrams[i], "vram%d", i - 1);
 			else
 				asprintf(&vrams[i], "system");
 			igt_assert(vrams[i]);
-- 
2.34.1

[igt-dev] [PATCH i-g-t v3 06/17] lib/xe_query: Add xe_region_class() helper

[Zbigniew Kempczyński]

In the common code we often need to be aware of region class.
Add helper which returns it from region id.

Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
Reviewed-by: Karolina Stolarek <karolina.stolarek@intel.com>
---
 lib/xe/xe_query.c | 16 ++++++++++++++++
 lib/xe/xe_query.h |  1 +
 2 files changed, 17 insertions(+)

diff --git a/lib/xe/xe_query.c b/lib/xe/xe_query.c
index 830b7e401d..f535ad8534 100644
--- a/lib/xe/xe_query.c
+++ b/lib/xe/xe_query.c
@@ -467,6 +467,22 @@ const char *xe_region_name(uint64_t region)
 	return vrams[region_idx];
 }
 
+/**
+ * xe_region_class:
+ * @fd: xe device fd
+ * @region: region mask
+ *
+ * Returns class of memory region structure for @region mask.
+ */
+uint16_t xe_region_class(int fd, uint64_t region)
+{
+	struct drm_xe_query_mem_region *memreg;
+
+	memreg = xe_mem_region(fd, region);
+
+	return memreg->mem_class;
+}
+
 /**
  * xe_min_page_size:
  * @fd: xe device fd
diff --git a/lib/xe/xe_query.h b/lib/xe/xe_query.h
index ff328ab942..68ca5a680c 100644
--- a/lib/xe/xe_query.h
+++ b/lib/xe/xe_query.h
@@ -85,6 +85,7 @@ struct drm_xe_engine_class_instance *xe_hw_engines(int fd);
 struct drm_xe_engine_class_instance *xe_hw_engine(int fd, int idx);
 struct drm_xe_query_mem_region *xe_mem_region(int fd, uint64_t region);
 const char *xe_region_name(uint64_t region);
+uint16_t xe_region_class(int fd, uint64_t region);
 uint32_t xe_min_page_size(int fd, uint64_t region);
 struct drm_xe_query_config *xe_config(int fd);
 unsigned int xe_number_hw_engines(int fd);
-- 
2.34.1

[igt-dev] [PATCH i-g-t v3 07/17] lib/drmtest: Add get_intel_driver() helper

[Zbigniew Kempczyński]

In libraries with i915 and Xe code divergence we might use
is_xe_device() or is_i915_device() to distinct code paths.
But to avoid additional open and string compare we can cache
this information in data structures.

Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
Reviewed-by: Karolina Stolarek <karolina.stolarek@intel.com>
---
 lib/drmtest.c | 10 ++++++++++
 lib/drmtest.h |  1 +
 2 files changed, 11 insertions(+)

diff --git a/lib/drmtest.c b/lib/drmtest.c
index 5cdb0196d3..e1da66c877 100644
--- a/lib/drmtest.c
+++ b/lib/drmtest.c
@@ -151,6 +151,16 @@ bool is_intel_device(int fd)
 	return is_i915_device(fd) || is_xe_device(fd);
 }
 
+enum intel_driver get_intel_driver(int fd)
+{
+	if (is_xe_device(fd))
+		return INTEL_DRIVER_XE;
+	else if (is_i915_device(fd))
+		return INTEL_DRIVER_I915;
+
+	igt_assert_f(0, "Device is not handled by Intel driver\n");
+}
+
 static char _forced_driver[16] = "";
 
 /**
diff --git a/lib/drmtest.h b/lib/drmtest.h
index 9c3ea5d14c..97ab6e759e 100644
--- a/lib/drmtest.h
+++ b/lib/drmtest.h
@@ -124,6 +124,7 @@ bool is_nouveau_device(int fd);
 bool is_vc4_device(int fd);
 bool is_xe_device(int fd);
 bool is_intel_device(int fd);
+enum intel_driver get_intel_driver(int fd);
 
 /**
  * do_or_die:
-- 
2.34.1

[igt-dev] [PATCH i-g-t v3 08/17] lib/xe_util: Return dynamic subtest name for Xe

[Zbigniew Kempczyński]

For tests which are working on more than one region using name suffix
like "vram01-system" etc. is common thing. Instead handcrafting this
naming add xe_memregion_dynamic_subtest_name() function which is
similar to memregion_dynamic_subtest_name() for i915.

Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
Reviewed-by: Karolina Stolarek <karolina.stolarek@intel.com>
---
 lib/meson.build  |   3 +-
 lib/xe/xe_util.c | 104 +++++++++++++++++++++++++++++++++++++++++++++++
 lib/xe/xe_util.h |  30 ++++++++++++++
 3 files changed, 136 insertions(+), 1 deletion(-)
 create mode 100644 lib/xe/xe_util.c
 create mode 100644 lib/xe/xe_util.h

diff --git a/lib/meson.build b/lib/meson.build
index 5523b4450e..ce11c0715f 100644
--- a/lib/meson.build
+++ b/lib/meson.build
@@ -105,7 +105,8 @@ lib_sources = [
 	'xe/xe_compute_square_kernels.c',
 	'xe/xe_ioctl.c',
 	'xe/xe_query.c',
-	'xe/xe_spin.c'
+	'xe/xe_spin.c',
+	'xe/xe_util.c',
 ]
 
 lib_deps = [
diff --git a/lib/xe/xe_util.c b/lib/xe/xe_util.c
new file mode 100644
index 0000000000..5f0f69a8c6
--- /dev/null
+++ b/lib/xe/xe_util.c
@@ -0,0 +1,104 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+
+#include "igt.h"
+#include "igt_syncobj.h"
+#include "xe/xe_ioctl.h"
+#include "xe/xe_query.h"
+#include "xe/xe_util.h"
+
+static bool __region_belongs_to_regions_type(struct drm_xe_query_mem_region *region,
+					     uint32_t *mem_regions_type,
+					     int num_regions)
+{
+	for (int i = 0; i < num_regions; i++)
+		if (mem_regions_type[i] == region->mem_class)
+			return true;
+	return false;
+}
+
+struct igt_collection *
+__xe_get_memory_region_set(int xe, uint32_t *mem_regions_type, int num_regions)
+{
+	struct drm_xe_query_mem_region *memregion;
+	struct igt_collection *set = NULL;
+	uint64_t memreg = all_memory_regions(xe), region;
+	int count = 0, pos = 0;
+
+	xe_for_each_mem_region(xe, memreg, region) {
+		memregion = xe_mem_region(xe, region);
+		if (__region_belongs_to_regions_type(memregion,
+						     mem_regions_type,
+						     num_regions))
+			count++;
+	}
+
+	set = igt_collection_create(count);
+
+	xe_for_each_mem_region(xe, memreg, region) {
+		memregion = xe_mem_region(xe, region);
+		igt_assert(region < (1ull << 31));
+		if (__region_belongs_to_regions_type(memregion,
+						     mem_regions_type,
+						     num_regions)) {
+			igt_collection_set_value(set, pos++, (int)region);
+		}
+	}
+
+	igt_assert(count == pos);
+
+	return set;
+}
+
+/**
+ * xe_memregion_dynamic_subtest_name:
+ * @xe: drm fd of Xe device
+ * @igt_collection: memory region collection
+ *
+ * Function iterates over all memory regions inside the collection (keeped
+ * in the value field) and generates the name which can be used during dynamic
+ * subtest creation.
+ *
+ * Returns: newly allocated string, has to be freed by caller. Asserts if
+ * caller tries to create a name using empty collection.
+ */
+char *xe_memregion_dynamic_subtest_name(int xe, struct igt_collection *set)
+{
+	struct igt_collection_data *data;
+	char *name, *p;
+	uint32_t region, len;
+
+	igt_assert(set && set->size);
+	/* enough for "name%d-" * n */
+	len = set->size * 8;
+	p = name = malloc(len);
+	igt_assert(name);
+
+	for_each_collection_data(data, set) {
+		struct drm_xe_query_mem_region *memreg;
+		int r;
+
+		region = data->value;
+		memreg = xe_mem_region(xe, region);
+
+		if (XE_IS_CLASS_VRAM(memreg))
+			r = snprintf(p, len, "%s%d-",
+				     xe_region_name(region),
+				     memreg->instance);
+		else
+			r = snprintf(p, len, "%s-",
+				     xe_region_name(region));
+
+		igt_assert(r > 0);
+		p += r;
+		len -= r;
+	}
+
+	/* remove last '-' */
+	*(p - 1) = 0;
+
+	return name;
+}
+
diff --git a/lib/xe/xe_util.h b/lib/xe/xe_util.h
new file mode 100644
index 0000000000..9f56fa9898
--- /dev/null
+++ b/lib/xe/xe_util.h
@@ -0,0 +1,30 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2023 Intel Corporation
+ *
+ */
+
+#ifndef XE_UTIL_H
+#define XE_UTIL_H
+
+#include <stdbool.h>
+#include <stddef.h>
+#include <stdint.h>
+#include <xe_drm.h>
+
+#define XE_IS_SYSMEM_MEMORY_REGION(fd, region) \
+	(xe_region_class(fd, region) == XE_MEM_REGION_CLASS_SYSMEM)
+#define XE_IS_VRAM_MEMORY_REGION(fd, region) \
+	(xe_region_class(fd, region) == XE_MEM_REGION_CLASS_VRAM)
+
+struct igt_collection *
+__xe_get_memory_region_set(int xe, uint32_t *mem_regions_type, int num_regions);
+
+#define xe_get_memory_region_set(regions, mem_region_types...) ({ \
+	unsigned int arr__[] = { mem_region_types }; \
+	__xe_get_memory_region_set(regions, arr__, ARRAY_SIZE(arr__)); \
+})
+
+char *xe_memregion_dynamic_subtest_name(int xe, struct igt_collection *set);
+
+#endif /* XE_UTIL_H */
-- 
2.34.1

[igt-dev] [PATCH i-g-t v3 09/17] lib/xe_util: Add vm bind/unbind helper for Xe

[Zbigniew Kempczyński]

Before calling exec we need to prepare vm to contain valid entries.
Bind/unbind in xe expects single bind_op or vector of bind_ops what
makes preparation of it a little bit inconvinient. Add function
which iterates over list of xe_object (auxiliary structure which
describes bind information for object) and performs the bind/unbind
in one step. It also supports passing syncobj in/out to work in
pipelined executions.

Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
--
v3: - Drop priv field in the structure
---
 lib/xe/xe_util.c | 125 +++++++++++++++++++++++++++++++++++++++++++++++
 lib/xe/xe_util.h |  17 +++++++
 2 files changed, 142 insertions(+)

diff --git a/lib/xe/xe_util.c b/lib/xe/xe_util.c
index 5f0f69a8c6..7d78065a13 100644
--- a/lib/xe/xe_util.c
+++ b/lib/xe/xe_util.c
@@ -102,3 +102,128 @@ char *xe_memregion_dynamic_subtest_name(int xe, struct igt_collection *set)
 	return name;
 }
 
+#ifdef XEBINDDBG
+#define bind_info igt_info
+#define bind_debug igt_debug
+#else
+#define bind_info(...) {}
+#define bind_debug(...) {}
+#endif
+
+static struct drm_xe_vm_bind_op *xe_alloc_bind_ops(struct igt_list_head *obj_list,
+						   uint32_t *num_ops)
+{
+	struct drm_xe_vm_bind_op *bind_ops, *ops;
+	struct xe_object *obj;
+	uint32_t num_objects = 0, i = 0, op;
+
+	igt_list_for_each_entry(obj, obj_list, link)
+		num_objects++;
+
+	*num_ops = num_objects;
+	if (!num_objects) {
+		bind_info(" [nothing to bind]\n");
+		return NULL;
+	}
+
+	bind_ops = calloc(num_objects, sizeof(*bind_ops));
+	igt_assert(bind_ops);
+
+	igt_list_for_each_entry(obj, obj_list, link) {
+		ops = &bind_ops[i];
+
+		if (obj->bind_op == XE_OBJECT_BIND) {
+			op = XE_VM_BIND_OP_MAP | XE_VM_BIND_FLAG_ASYNC;
+			ops->obj = obj->handle;
+		} else {
+			op = XE_VM_BIND_OP_UNMAP | XE_VM_BIND_FLAG_ASYNC;
+		}
+
+		ops->op = op;
+		ops->obj_offset = 0;
+		ops->addr = obj->offset;
+		ops->range = obj->size;
+		ops->region = 0;
+
+		bind_info("  [%d]: [%6s] handle: %u, offset: %llx, size: %llx\n",
+			  i, obj->bind_op == XE_OBJECT_BIND ? "BIND" : "UNBIND",
+			  ops->obj, (long long)ops->addr, (long long)ops->range);
+		i++;
+	}
+
+	return bind_ops;
+}
+
+/**
+ * xe_bind_unbind_async:
+ * @xe: drm fd of Xe device
+ * @vm: vm to bind/unbind objects to/from
+ * @bind_engine: bind engine, 0 if default
+ * @obj_list: list of xe_object
+ * @sync_in: sync object (fence-in), 0 if there's no input dependency
+ * @sync_out: sync object (fence-out) to signal on bind/unbind completion,
+ *            if 0 wait for bind/unbind completion.
+ *
+ * Function iterates over xe_object @obj_list, prepares binding operation
+ * and does bind/unbind in one step. Providing sync_in / sync_out allows
+ * working in pipelined mode. With sync_in and sync_out set to 0 function
+ * waits until binding operation is complete.
+ */
+void xe_bind_unbind_async(int xe, uint32_t vm, uint32_t bind_engine,
+			  struct igt_list_head *obj_list,
+			  uint32_t sync_in, uint32_t sync_out)
+{
+	struct drm_xe_vm_bind_op *bind_ops;
+	struct drm_xe_sync tabsyncs[2] = {
+		{ .flags = DRM_XE_SYNC_SYNCOBJ, .handle = sync_in },
+		{ .flags = DRM_XE_SYNC_SYNCOBJ | DRM_XE_SYNC_SIGNAL, .handle = sync_out },
+	};
+	struct drm_xe_sync *syncs;
+	uint32_t num_binds = 0;
+	int num_syncs;
+
+	bind_info("[Binding to vm: %u]\n", vm);
+	bind_ops = xe_alloc_bind_ops(obj_list, &num_binds);
+
+	if (!num_binds) {
+		if (sync_out)
+			syncobj_signal(xe, &sync_out, 1);
+		return;
+	}
+
+	if (sync_in) {
+		syncs = tabsyncs;
+		num_syncs = 2;
+	} else {
+		syncs = &tabsyncs[1];
+		num_syncs = 1;
+	}
+
+	/* User didn't pass sync out, create it and wait for completion */
+	if (!sync_out)
+		tabsyncs[1].handle = syncobj_create(xe, 0);
+
+	bind_info("[Binding syncobjs: (in: %u, out: %u)]\n",
+		  tabsyncs[0].handle, tabsyncs[1].handle);
+
+	if (num_binds == 1) {
+		if ((bind_ops[0].op & 0xffff) == XE_VM_BIND_OP_MAP)
+			xe_vm_bind_async(xe, vm, bind_engine, bind_ops[0].obj, 0,
+					bind_ops[0].addr, bind_ops[0].range,
+					syncs, num_syncs);
+		else
+			xe_vm_unbind_async(xe, vm, bind_engine, 0,
+					   bind_ops[0].addr, bind_ops[0].range,
+					   syncs, num_syncs);
+	} else {
+		xe_vm_bind_array(xe, vm, bind_engine, bind_ops,
+				 num_binds, syncs, num_syncs);
+	}
+
+	if (!sync_out) {
+		igt_assert_eq(syncobj_wait_err(xe, &tabsyncs[1].handle, 1, INT64_MAX, 0), 0);
+		syncobj_destroy(xe, tabsyncs[1].handle);
+	}
+
+	free(bind_ops);
+}
diff --git a/lib/xe/xe_util.h b/lib/xe/xe_util.h
index 9f56fa9898..61511dd97f 100644
--- a/lib/xe/xe_util.h
+++ b/lib/xe/xe_util.h
@@ -27,4 +27,21 @@ __xe_get_memory_region_set(int xe, uint32_t *mem_regions_type, int num_regions);
 
 char *xe_memregion_dynamic_subtest_name(int xe, struct igt_collection *set);
 
+enum xe_bind_op {
+	XE_OBJECT_BIND,
+	XE_OBJECT_UNBIND,
+};
+
+struct xe_object {
+	uint32_t handle;
+	uint64_t offset;
+	uint64_t size;
+	enum xe_bind_op bind_op;
+	struct igt_list_head link;
+};
+
+void xe_bind_unbind_async(int fd, uint32_t vm, uint32_t bind_engine,
+			  struct igt_list_head *obj_list,
+			  uint32_t sync_in, uint32_t sync_out);
+
 #endif /* XE_UTIL_H */
-- 
2.34.1

[igt-dev] [PATCH i-g-t v3 10/17] lib/intel_allocator: Add intel_allocator_bind()

[Zbigniew Kempczyński]

Synchronize allocator state to vm.

This change allows xe user to execute vm-bind/unbind for allocator
alloc()/free() operations which occurred since last binding/unbinding.
Before doing exec user should call intel_allocator_bind() to ensure
all vma's are in place.

Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
---
v2: Rewrite tracking mechanism: previous code uses bind map embedded
    in allocator structure. Unfortunately this wasn't good idea
    - for xe binding everything was fine, but it regress multiprocess/
    multithreaded allocations. Main reason of this was children
    processes couldn't get its reference as this memory was allocated
    on allocator thread (separate process). Currently each child
    contains its own separate tracking maps for ahnd and for each
    ahnd bind map.
v3: - Don't use priv field (we may clean bind_map apriori as failing
      asserts and outdated bind_map is not a problem anymore).
    - Use only vm on tracking ahnd (Karolina)
    - Add braces to avoid checkpatch complaining (Karolina)
    - Use igt_assert_f() instead conditional block (Karolina)
---
 lib/igt_core.c        |   5 +
 lib/intel_allocator.c | 259 +++++++++++++++++++++++++++++++++++++++++-
 lib/intel_allocator.h |   3 +
 3 files changed, 265 insertions(+), 2 deletions(-)

diff --git a/lib/igt_core.c b/lib/igt_core.c
index 2ae2cb6883..041e4b3288 100644
--- a/lib/igt_core.c
+++ b/lib/igt_core.c
@@ -74,6 +74,7 @@
 #include "igt_sysrq.h"
 #include "igt_rc.h"
 #include "igt_list.h"
+#include "igt_map.h"
 #include "igt_device_scan.h"
 #include "igt_thread.h"
 #include "runnercomms.h"
@@ -320,6 +321,8 @@ bool test_multi_fork_child;
 /* For allocator purposes */
 pid_t child_pid  = -1;
 __thread pid_t child_tid  = -1;
+struct igt_map *ahnd_map;
+pthread_mutex_t ahnd_map_mutex;
 
 enum {
 	/*
@@ -2523,6 +2526,8 @@ bool __igt_fork(void)
 	case 0:
 		test_child = true;
 		pthread_mutex_init(&print_mutex, NULL);
+		pthread_mutex_init(&ahnd_map_mutex, NULL);
+		ahnd_map = igt_map_create(igt_map_hash_64, igt_map_equal_64);
 		child_pid = getpid();
 		child_tid = -1;
 		exit_handler_count = 0;
diff --git a/lib/intel_allocator.c b/lib/intel_allocator.c
index cf964ccc41..f0a9b7fb53 100644
--- a/lib/intel_allocator.c
+++ b/lib/intel_allocator.c
@@ -17,8 +17,8 @@
 #include "intel_allocator.h"
 #include "intel_allocator_msgchannel.h"
 #include "xe/xe_query.h"
+#include "xe/xe_util.h"
 
-//#define ALLOCDBG
 #ifdef ALLOCDBG
 #define alloc_info igt_info
 #define alloc_debug igt_debug
@@ -45,6 +45,14 @@ static inline const char *reqstr(enum reqtype request_type)
 #define alloc_debug(...) {}
 #endif
 
+#ifdef ALLOCBINDDBG
+#define bind_info igt_info
+#define bind_debug igt_debug
+#else
+#define bind_info(...) {}
+#define bind_debug(...) {}
+#endif
+
 /*
  * We limit allocator space to avoid hang when batch would be
  * pinned in the last page.
@@ -64,6 +72,30 @@ struct handle_entry {
 	struct allocator *al;
 };
 
+/* For tracking alloc()/free() for Xe */
+struct ahnd_info {
+	int fd;
+	uint64_t ahnd;
+	uint32_t vm;
+	enum intel_driver driver;
+	struct igt_map *bind_map;
+	pthread_mutex_t bind_map_mutex;
+};
+
+enum allocator_bind_op {
+	BOUND,
+	TO_BIND,
+	TO_UNBIND,
+};
+
+struct allocator_object {
+	uint32_t handle;
+	uint64_t offset;
+	uint64_t size;
+
+	enum allocator_bind_op bind_op;
+};
+
 struct intel_allocator *
 intel_allocator_reloc_create(int fd, uint64_t start, uint64_t end);
 struct intel_allocator *
@@ -122,6 +154,13 @@ static pid_t allocator_pid = -1;
 extern pid_t child_pid;
 extern __thread pid_t child_tid;
 
+/*
+ * Track alloc()/free() requires storing in local process which has
+ * an access to real drm fd it can work on.
+ */
+extern struct igt_map *ahnd_map;
+extern pthread_mutex_t ahnd_map_mutex;
+
 /*
  * - for parent process we have child_pid == -1
  * - for child which calls intel_allocator_init() allocator_pid == child_pid
@@ -837,6 +876,45 @@ void intel_allocator_multiprocess_stop(void)
 	}
 }
 
+static void track_ahnd(int fd, uint64_t ahnd, uint32_t vm)
+{
+	struct ahnd_info *ainfo;
+
+	pthread_mutex_lock(&ahnd_map_mutex);
+	ainfo = igt_map_search(ahnd_map, &ahnd);
+	if (!ainfo) {
+		ainfo = malloc(sizeof(*ainfo));
+		ainfo->fd = fd;
+		ainfo->ahnd = ahnd;
+		ainfo->vm = vm;
+		ainfo->driver = get_intel_driver(fd);
+		ainfo->bind_map = igt_map_create(igt_map_hash_32, igt_map_equal_32);
+		pthread_mutex_init(&ainfo->bind_map_mutex, NULL);
+		bind_debug("[TRACK AHND] pid: %d, tid: %d, create <fd: %d, "
+			   "ahnd: %llx, vm: %u, driver: %d, ahnd_map: %p, bind_map: %p>\n",
+			   getpid(), gettid(), ainfo->fd,
+			   (long long)ainfo->ahnd, ainfo->vm,
+			   ainfo->driver, ahnd_map, ainfo->bind_map);
+		igt_map_insert(ahnd_map, &ainfo->ahnd, ainfo);
+	}
+
+	pthread_mutex_unlock(&ahnd_map_mutex);
+}
+
+static void untrack_ahnd(uint64_t ahnd)
+{
+	struct ahnd_info *ainfo;
+
+	pthread_mutex_lock(&ahnd_map_mutex);
+	ainfo = igt_map_search(ahnd_map, &ahnd);
+	if (ainfo) {
+		bind_debug("[UNTRACK AHND]: pid: %d, tid: %d, removing ahnd: %llx\n",
+			   getpid(), gettid(), (long long)ahnd);
+		igt_map_remove(ahnd_map, &ahnd, map_entry_free_func);
+	}
+	pthread_mutex_unlock(&ahnd_map_mutex);
+}
+
 static uint64_t __intel_allocator_open_full(int fd, uint32_t ctx,
 					    uint32_t vm,
 					    uint64_t start, uint64_t end,
@@ -895,6 +973,12 @@ static uint64_t __intel_allocator_open_full(int fd, uint32_t ctx,
 	igt_assert(resp.open.allocator_handle);
 	igt_assert(resp.response_type == RESP_OPEN);
 
+	/*
+	 * Igts mostly uses ctx as id when opening the allocator (i915 legacy).
+	 * If ctx is passed let's use it as an vm id, otherwise use vm.
+	 */
+	track_ahnd(fd, resp.open.allocator_handle, ctx ?: vm);
+
 	return resp.open.allocator_handle;
 }
 
@@ -1001,6 +1085,8 @@ bool intel_allocator_close(uint64_t allocator_handle)
 	igt_assert(handle_request(&req, &resp) == 0);
 	igt_assert(resp.response_type == RESP_CLOSE);
 
+	untrack_ahnd(allocator_handle);
+
 	return resp.close.is_empty;
 }
 
@@ -1034,6 +1120,74 @@ void intel_allocator_get_address_range(uint64_t allocator_handle,
 		*endp = resp.address_range.end;
 }
 
+static bool is_same(struct allocator_object *obj,
+		    uint32_t handle, uint64_t offset, uint64_t size,
+		    enum allocator_bind_op bind_op)
+{
+	return obj->handle == handle &&	obj->offset == offset && obj->size == size &&
+	       (obj->bind_op == bind_op || obj->bind_op == BOUND);
+}
+
+static void track_object(uint64_t allocator_handle, uint32_t handle,
+			 uint64_t offset, uint64_t size,
+			 enum allocator_bind_op bind_op)
+{
+	struct ahnd_info *ainfo;
+	struct allocator_object *obj;
+
+	bind_debug("[TRACK OBJECT]: [%s] pid: %d, tid: %d, ahnd: %llx, handle: %u, offset: %llx, size: %llx\n",
+		   bind_op == TO_BIND ? "BIND" : "UNBIND",
+		   getpid(), gettid(),
+		   (long long)allocator_handle,
+		   handle, (long long)offset, (long long)size);
+
+	if (offset == ALLOC_INVALID_ADDRESS) {
+		bind_debug("[TRACK OBJECT] => invalid address %llx, skipping tracking\n",
+			   (long long)offset);
+		return;
+	}
+
+	pthread_mutex_lock(&ahnd_map_mutex);
+	ainfo = igt_map_search(ahnd_map, &allocator_handle);
+	pthread_mutex_unlock(&ahnd_map_mutex);
+	igt_assert_f(ainfo, "[TRACK OBJECT] => MISSING ahnd %llx <=\n",
+		     (long long)allocator_handle);
+
+	if (ainfo->driver == INTEL_DRIVER_I915)
+		return; /* no-op for i915, at least for now */
+
+	pthread_mutex_lock(&ainfo->bind_map_mutex);
+	obj = igt_map_search(ainfo->bind_map, &handle);
+	if (obj) {
+		/*
+		 * User may call alloc() couple of times, check object is the
+		 * same. For free() there's simple case, just remove from
+		 * bind_map.
+		 */
+		if (bind_op == TO_BIND) {
+			igt_assert_eq(is_same(obj, handle, offset, size, bind_op), true);
+		} else if (bind_op == TO_UNBIND) {
+			if (obj->bind_op == TO_BIND)
+				igt_map_remove(ainfo->bind_map, &obj->handle, map_entry_free_func);
+			else if (obj->bind_op == BOUND)
+				obj->bind_op = bind_op;
+		}
+	} else {
+		/* Ignore to unbind bo which wasn't previously inserted */
+		if (bind_op == TO_UNBIND)
+			goto out;
+
+		obj = calloc(1, sizeof(*obj));
+		obj->handle = handle;
+		obj->offset = offset;
+		obj->size = size;
+		obj->bind_op = bind_op;
+		igt_map_insert(ainfo->bind_map, &obj->handle, obj);
+	}
+out:
+	pthread_mutex_unlock(&ainfo->bind_map_mutex);
+}
+
 /**
  * __intel_allocator_alloc:
  * @allocator_handle: handle to an allocator
@@ -1065,6 +1219,8 @@ uint64_t __intel_allocator_alloc(uint64_t allocator_handle, uint32_t handle,
 	igt_assert(handle_request(&req, &resp) == 0);
 	igt_assert(resp.response_type == RESP_ALLOC);
 
+	track_object(allocator_handle, handle, resp.alloc.offset, size, TO_BIND);
+
 	return resp.alloc.offset;
 }
 
@@ -1142,6 +1298,8 @@ bool intel_allocator_free(uint64_t allocator_handle, uint32_t handle)
 	igt_assert(handle_request(&req, &resp) == 0);
 	igt_assert(resp.response_type == RESP_FREE);
 
+	track_object(allocator_handle, handle, 0, 0, TO_UNBIND);
+
 	return resp.free.freed;
 }
 
@@ -1326,6 +1484,84 @@ void intel_allocator_print(uint64_t allocator_handle)
 	}
 }
 
+static void __xe_op_bind(struct ahnd_info *ainfo, uint32_t sync_in, uint32_t sync_out)
+{
+	struct allocator_object *obj;
+	struct igt_map_entry *pos;
+	struct igt_list_head obj_list;
+	struct xe_object *entry, *tmp;
+
+	IGT_INIT_LIST_HEAD(&obj_list);
+
+	pthread_mutex_lock(&ainfo->bind_map_mutex);
+	igt_map_foreach(ainfo->bind_map, pos) {
+		obj = pos->data;
+
+		if (obj->bind_op == BOUND)
+			continue;
+
+		bind_info("= [vm: %u] %s => %u %lx %lx\n",
+			  ainfo->vm,
+			  obj->bind_op == TO_BIND ? "TO BIND" : "TO UNBIND",
+			  obj->handle, obj->offset,
+			  obj->size);
+
+		entry = malloc(sizeof(*entry));
+		entry->handle = obj->handle;
+		entry->offset = obj->offset;
+		entry->size = obj->size;
+		entry->bind_op = obj->bind_op == TO_BIND ? XE_OBJECT_BIND :
+							   XE_OBJECT_UNBIND;
+		igt_list_add(&entry->link, &obj_list);
+
+		/*
+		 * We clean bind_map even before calling bind/unbind
+		 * as all binding operations asserts in case of error.
+		 */
+		if (obj->bind_op == TO_BIND)
+			obj->bind_op = BOUND;
+		else
+			igt_map_remove(ainfo->bind_map, &obj->handle,
+				       map_entry_free_func);
+	}
+	pthread_mutex_unlock(&ainfo->bind_map_mutex);
+
+	xe_bind_unbind_async(ainfo->fd, ainfo->vm, 0, &obj_list, sync_in, sync_out);
+
+	igt_list_for_each_entry_safe(entry, tmp, &obj_list, link) {
+		igt_list_del(&entry->link);
+		free(entry);
+	}
+}
+
+/**
+ * intel_allocator_bind:
+ * @allocator_handle: handle to an allocator
+ * @sync_in: syncobj (fence-in)
+ * @sync_out: syncobj (fence-out)
+ *
+ * Function binds and unbinds all objects added to the allocator which weren't
+ * previously binded/unbinded.
+ *
+ **/
+void intel_allocator_bind(uint64_t allocator_handle,
+			  uint32_t sync_in, uint32_t sync_out)
+{
+	struct ahnd_info *ainfo;
+
+	pthread_mutex_lock(&ahnd_map_mutex);
+	ainfo = igt_map_search(ahnd_map, &allocator_handle);
+	pthread_mutex_unlock(&ahnd_map_mutex);
+	igt_assert(ainfo);
+
+	/*
+	 * We collect bind/unbind operations on alloc()/free() to do group
+	 * operation getting @sync_in as syncobj handle (fence-in). If user
+	 * passes 0 as @sync_out we bind/unbind synchronously.
+	 */
+	__xe_op_bind(ainfo, sync_in, sync_out);
+}
+
 static int equal_handles(const void *key1, const void *key2)
 {
 	const struct handle_entry *h1 = key1, *h2 = key2;
@@ -1383,6 +1619,22 @@ static void __free_maps(struct igt_map *map, bool close_allocators)
 	igt_map_destroy(map, map_entry_free_func);
 }
 
+static void __free_ahnd_map(void)
+{
+	struct igt_map_entry *pos;
+	struct ahnd_info *ainfo;
+
+	if (!ahnd_map)
+		return;
+
+	igt_map_foreach(ahnd_map, pos) {
+		ainfo = pos->data;
+		igt_map_destroy(ainfo->bind_map, map_entry_free_func);
+	}
+
+	igt_map_destroy(ahnd_map, map_entry_free_func);
+}
+
 /**
  * intel_allocator_init:
  *
@@ -1400,12 +1652,15 @@ void intel_allocator_init(void)
 	__free_maps(handles, true);
 	__free_maps(ctx_map, false);
 	__free_maps(vm_map, false);
+	__free_ahnd_map();
 
 	atomic_init(&next_handle, 1);
 	handles = igt_map_create(hash_handles, equal_handles);
 	ctx_map = igt_map_create(hash_instance, equal_ctx);
 	vm_map = igt_map_create(hash_instance, equal_vm);
-	igt_assert(handles && ctx_map && vm_map);
+	pthread_mutex_init(&ahnd_map_mutex, NULL);
+	ahnd_map = igt_map_create(igt_map_hash_64, igt_map_equal_64);
+	igt_assert(handles && ctx_map && vm_map && ahnd_map);
 
 	channel = intel_allocator_get_msgchannel(CHANNEL_SYSVIPC_MSGQUEUE);
 }
diff --git a/lib/intel_allocator.h b/lib/intel_allocator.h
index 3ec74f6191..f9ff7f1cc9 100644
--- a/lib/intel_allocator.h
+++ b/lib/intel_allocator.h
@@ -210,6 +210,9 @@ bool intel_allocator_reserve_if_not_allocated(uint64_t allocator_handle,
 
 void intel_allocator_print(uint64_t allocator_handle);
 
+void intel_allocator_bind(uint64_t allocator_handle,
+			  uint32_t sync_in, uint32_t sync_out);
+
 #define ALLOC_INVALID_ADDRESS (-1ull)
 #define INTEL_ALLOCATOR_NONE   0
 #define INTEL_ALLOCATOR_RELOC  1
-- 
2.34.1

[igt-dev] [PATCH i-g-t v3 11/17] lib/intel_ctx: Add xe context information

[Zbigniew Kempczyński]

Most complicated part in adopting i915_blt to intel_blt - which should
handle both drivers - is how to achieve pipelined execution. In term
pipelined execution I mean all gpu workloads are executed without
stalls.

Comparing to i915 relocations and softpinning xe architecture migrates
binding (this means also unbind operation) responsibility from the
kernel to user via vm_bind ioctl(). To avoid stalls user has to
provide in/out fences (syncobjs) between consecutive bindings/execs.
Of course for many igt tests we don't need pipelined execution,
just synchronous bind, then exec. But exercising the driver should
also cover pipelining to verify it is possible to work without stalls.

I decided to extend intel_ctx_t with all necessary for xe objects
(vm, engine, syncobjs) to get flexibility in deciding how to bind,
execute and wait for (synchronize) those operations. Context object
along with i915 engine is already passed to blitter library so adding
xe required fields doesn't break i915 but will allow xe path to get
all necessary data to execute.

Using intel_ctx with xe requires some code patterns caused by usage
of an allocator. For xe the allocator started tracking alloc()/free()
operations to do bind/unbind in one call just before execution.
I've added two helpers in intel_ctx which - intel_ctx_xe_exec()
and intel_ctx_xe_sync(). Depending how intel_ctx was created
(with 0 or real syncobj handles as in/bind/out fences) bind and exec
in intel_ctx_xe_exec() are pipelined but synchronize last operation
(exec). For real syncobjs they are used to join bind + exec calls
but there's no wait for exec (sync-out) completion. This allows
building more cascaded bind + exec operations without stalls.

To wait for a sync-out fence caller may use intel_ctx_xe_sync()
which is synchronous wait on syncobj. It allows user to reset
fences to prepare for next operation.

Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
---
v3: drop local __xe_exec() (Karolina)
---
 lib/intel_ctx.c | 98 ++++++++++++++++++++++++++++++++++++++++++++++++-
 lib/intel_ctx.h | 14 +++++++
 2 files changed, 111 insertions(+), 1 deletion(-)

diff --git a/lib/intel_ctx.c b/lib/intel_ctx.c
index ded9c0f1e4..0c1ce6d10b 100644
--- a/lib/intel_ctx.c
+++ b/lib/intel_ctx.c
@@ -5,9 +5,12 @@
 
 #include <stddef.h>
 
+#include "i915/gem_engine_topology.h"
+#include "igt_syncobj.h"
+#include "intel_allocator.h"
 #include "intel_ctx.h"
 #include "ioctl_wrappers.h"
-#include "i915/gem_engine_topology.h"
+#include "xe/xe_ioctl.h"
 
 /**
  * SECTION:intel_ctx
@@ -390,3 +393,96 @@ unsigned int intel_ctx_engine_class(const intel_ctx_t *ctx, unsigned int engine)
 {
 	return intel_ctx_cfg_engine_class(&ctx->cfg, engine);
 }
+
+/**
+ * intel_ctx_xe:
+ * @fd: open i915 drm file descriptor
+ * @vm: vm
+ * @engine: engine
+ *
+ * Returns an intel_ctx_t representing the xe context.
+ */
+intel_ctx_t *intel_ctx_xe(int fd, uint32_t vm, uint32_t engine,
+			  uint32_t sync_in, uint32_t sync_bind, uint32_t sync_out)
+{
+	intel_ctx_t *ctx;
+
+	ctx = calloc(1, sizeof(*ctx));
+	igt_assert(ctx);
+
+	ctx->fd = fd;
+	ctx->vm = vm;
+	ctx->engine = engine;
+	ctx->sync_in = sync_in;
+	ctx->sync_bind = sync_bind;
+	ctx->sync_out = sync_out;
+
+	return ctx;
+}
+
+int __intel_ctx_xe_exec(const intel_ctx_t *ctx, uint64_t ahnd, uint64_t bb_offset)
+{
+	struct drm_xe_sync syncs[2] = {
+		{ .flags = DRM_XE_SYNC_SYNCOBJ, },
+		{ .flags = DRM_XE_SYNC_SYNCOBJ | DRM_XE_SYNC_SIGNAL, },
+	};
+	struct drm_xe_exec exec = {
+		.engine_id = ctx->engine,
+		.syncs = (uintptr_t)syncs,
+		.num_syncs = 2,
+		.address = bb_offset,
+		.num_batch_buffer = 1,
+	};
+	uint32_t sync_in = ctx->sync_in;
+	uint32_t sync_bind = ctx->sync_bind ?: syncobj_create(ctx->fd, 0);
+	uint32_t sync_out = ctx->sync_out ?: syncobj_create(ctx->fd, 0);
+	int ret;
+
+	/* Synchronize allocator state -> vm */
+	intel_allocator_bind(ahnd, sync_in, sync_bind);
+
+	/* Pipelined exec */
+	syncs[0].handle = sync_bind;
+	syncs[1].handle = sync_out;
+
+	ret = __xe_exec(ctx->fd, &exec);
+	if (ret)
+		goto err;
+
+	if (!ctx->sync_bind || !ctx->sync_out)
+		syncobj_wait_err(ctx->fd, &sync_out, 1, INT64_MAX, 0);
+
+err:
+	if (!ctx->sync_bind)
+		syncobj_destroy(ctx->fd, sync_bind);
+
+	if (!ctx->sync_out)
+		syncobj_destroy(ctx->fd, sync_out);
+
+	return ret;
+}
+
+void intel_ctx_xe_exec(const intel_ctx_t *ctx, uint64_t ahnd, uint64_t bb_offset)
+{
+	igt_assert_eq(__intel_ctx_xe_exec(ctx, ahnd, bb_offset), 0);
+}
+
+#define RESET_SYNCOBJ(__fd, __sync) do { \
+	if (__sync) \
+		syncobj_reset((__fd), &(__sync), 1); \
+} while (0)
+
+int intel_ctx_xe_sync(intel_ctx_t *ctx, bool reset_syncs)
+{
+	int ret;
+
+	ret = syncobj_wait_err(ctx->fd, &ctx->sync_out, 1, INT64_MAX, 0);
+
+	if (reset_syncs) {
+		RESET_SYNCOBJ(ctx->fd, ctx->sync_in);
+		RESET_SYNCOBJ(ctx->fd, ctx->sync_bind);
+		RESET_SYNCOBJ(ctx->fd, ctx->sync_out);
+	}
+
+	return ret;
+}
diff --git a/lib/intel_ctx.h b/lib/intel_ctx.h
index 3cfeaae81e..59d0360ada 100644
--- a/lib/intel_ctx.h
+++ b/lib/intel_ctx.h
@@ -67,6 +67,14 @@ int intel_ctx_cfg_engine_class(const intel_ctx_cfg_t *cfg, unsigned int engine);
 typedef struct intel_ctx {
 	uint32_t id;
 	intel_ctx_cfg_t cfg;
+
+	/* Xe */
+	int fd;
+	uint32_t vm;
+	uint32_t engine;
+	uint32_t sync_in;
+	uint32_t sync_bind;
+	uint32_t sync_out;
 } intel_ctx_t;
 
 int __intel_ctx_create(int fd, const intel_ctx_cfg_t *cfg,
@@ -81,4 +89,10 @@ void intel_ctx_destroy(int fd, const intel_ctx_t *ctx);
 
 unsigned int intel_ctx_engine_class(const intel_ctx_t *ctx, unsigned int engine);
 
+intel_ctx_t *intel_ctx_xe(int fd, uint32_t vm, uint32_t engine,
+			  uint32_t sync_in, uint32_t sync_bind, uint32_t sync_out);
+int __intel_ctx_xe_exec(const intel_ctx_t *ctx, uint64_t ahnd, uint64_t bb_offset);
+void intel_ctx_xe_exec(const intel_ctx_t *ctx, uint64_t ahnd, uint64_t bb_offset);
+int intel_ctx_xe_sync(intel_ctx_t *ctx, bool reset_syncs);
+
 #endif
-- 
2.34.1

[igt-dev] [PATCH i-g-t v3 12/17] lib/intel_blt: Introduce blt_copy_init() helper to cache driver

[Zbigniew Kempczyński]

Instead of calling is_xe_device() and is_i915_device() multiple
times in code which distincs xe and i915 paths add driver field
to structures used in blitter library.

Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
Reviewed-by: Karolina Stolarek <karolina.stolarek@intel.com>
---
 lib/igt_fb.c                   |  2 +-
 lib/intel_blt.c                | 40 +++++++++++++++++++++++++++++++---
 lib/intel_blt.h                |  8 ++++++-
 tests/i915/gem_ccs.c           | 34 ++++++++++++++++-------------
 tests/i915/gem_exercise_blt.c  | 22 ++++++++++---------
 tests/i915/gem_lmem_swapping.c |  4 ++--
 6 files changed, 78 insertions(+), 32 deletions(-)

diff --git a/lib/igt_fb.c b/lib/igt_fb.c
index a8988274f2..1814e8db11 100644
--- a/lib/igt_fb.c
+++ b/lib/igt_fb.c
@@ -2900,7 +2900,7 @@ static void blitcopy(const struct igt_fb *dst_fb,
 			src = blt_fb_init(src_fb, i, mem_region);
 			dst = blt_fb_init(dst_fb, i, mem_region);
 
-			memset(&blt, 0, sizeof(blt));
+			blt_copy_init(src_fb->fd, &blt);
 			blt.color_depth = blt_get_bpp(src_fb);
 			blt_set_copy_object(&blt.src, src);
 			blt_set_copy_object(&blt.dst, dst);
diff --git a/lib/intel_blt.c b/lib/intel_blt.c
index bc28f15e8d..f2f86e4947 100644
--- a/lib/intel_blt.c
+++ b/lib/intel_blt.c
@@ -692,6 +692,22 @@ static void dump_bb_ext(struct gen12_block_copy_data_ext *data)
 		 data->dw21.src_array_index);
 }
 
+/**
+ * blt_copy_init:
+ * @fd: drm fd
+ * @blt: structure for initialization
+ *
+ * Function is zeroing @blt and sets fd and driver fields (INTEL_DRIVER_I915 or
+ * INTEL_DRIVER_XE).
+ */
+void blt_copy_init(int fd, struct blt_copy_data *blt)
+{
+	memset(blt, 0, sizeof(*blt));
+
+	blt->fd = fd;
+	blt->driver = get_intel_driver(fd);
+}
+
 /**
  * emit_blt_block_copy:
  * @fd: drm fd
@@ -889,6 +905,22 @@ static void dump_bb_surf_ctrl_cmd(const struct gen12_ctrl_surf_copy_data *data)
 		 cmd[4], data->dw04.dst_address_hi, data->dw04.dst_mocs);
 }
 
+/**
+ * blt_ctrl_surf_copy_init:
+ * @fd: drm fd
+ * @surf: structure for initialization
+ *
+ * Function is zeroing @surf and sets fd and driver fields (INTEL_DRIVER_I915 or
+ * INTEL_DRIVER_XE).
+ */
+void blt_ctrl_surf_copy_init(int fd, struct blt_ctrl_surf_copy_data *surf)
+{
+	memset(surf, 0, sizeof(*surf));
+
+	surf->fd = fd;
+	surf->driver = get_intel_driver(fd);
+}
+
 /**
  * emit_blt_ctrl_surf_copy:
  * @fd: drm fd
@@ -1317,7 +1349,7 @@ void blt_set_batch(struct blt_copy_batch *batch,
 }
 
 struct blt_copy_object *
-blt_create_object(int fd, uint32_t region,
+blt_create_object(const struct blt_copy_data *blt, uint32_t region,
 		  uint32_t width, uint32_t height, uint32_t bpp, uint8_t mocs,
 		  enum blt_tiling_type tiling,
 		  enum blt_compression compression,
@@ -1329,10 +1361,12 @@ blt_create_object(int fd, uint32_t region,
 	uint32_t stride = tiling == T_LINEAR ? width * 4 : width;
 	uint32_t handle;
 
+	igt_assert_f(blt->driver, "Driver isn't set, have you called blt_copy_init()?\n");
+
 	obj = calloc(1, sizeof(*obj));
 
 	obj->size = size;
-	igt_assert(__gem_create_in_memory_regions(fd, &handle,
+	igt_assert(__gem_create_in_memory_regions(blt->fd, &handle,
 						  &size, region) == 0);
 
 	blt_set_object(obj, handle, size, region, mocs, tiling,
@@ -1340,7 +1374,7 @@ blt_create_object(int fd, uint32_t region,
 	blt_set_geom(obj, stride, 0, 0, width, height, 0, 0);
 
 	if (create_mapping)
-		obj->ptr = gem_mmap__device_coherent(fd, handle, 0, size,
+		obj->ptr = gem_mmap__device_coherent(blt->fd, handle, 0, size,
 						     PROT_READ | PROT_WRITE);
 
 	return obj;
diff --git a/lib/intel_blt.h b/lib/intel_blt.h
index 9c4ddc7a89..7516ce8ac7 100644
--- a/lib/intel_blt.h
+++ b/lib/intel_blt.h
@@ -102,6 +102,7 @@ struct blt_copy_batch {
 /* Common for block-copy and fast-copy */
 struct blt_copy_data {
 	int fd;
+	enum intel_driver driver;
 	struct blt_copy_object src;
 	struct blt_copy_object dst;
 	struct blt_copy_batch bb;
@@ -155,6 +156,7 @@ struct blt_ctrl_surf_copy_object {
 
 struct blt_ctrl_surf_copy_data {
 	int fd;
+	enum intel_driver driver;
 	struct blt_ctrl_surf_copy_object src;
 	struct blt_ctrl_surf_copy_object dst;
 	struct blt_copy_batch bb;
@@ -185,6 +187,8 @@ bool blt_uses_extended_block_copy(int fd);
 
 const char *blt_tiling_name(enum blt_tiling_type tiling);
 
+void blt_copy_init(int fd, struct blt_copy_data *blt);
+
 uint64_t emit_blt_block_copy(int fd,
 			     uint64_t ahnd,
 			     const struct blt_copy_data *blt,
@@ -205,6 +209,8 @@ uint64_t emit_blt_ctrl_surf_copy(int fd,
 				 uint64_t bb_pos,
 				 bool emit_bbe);
 
+void blt_ctrl_surf_copy_init(int fd, struct blt_ctrl_surf_copy_data *surf);
+
 int blt_ctrl_surf_copy(int fd,
 		       const intel_ctx_t *ctx,
 		       const struct intel_execution_engine2 *e,
@@ -230,7 +236,7 @@ void blt_set_batch(struct blt_copy_batch *batch,
 		   uint32_t handle, uint64_t size, uint32_t region);
 
 struct blt_copy_object *
-blt_create_object(int fd, uint32_t region,
+blt_create_object(const struct blt_copy_data *blt, uint32_t region,
 		  uint32_t width, uint32_t height, uint32_t bpp, uint8_t mocs,
 		  enum blt_tiling_type tiling,
 		  enum blt_compression compression,
diff --git a/tests/i915/gem_ccs.c b/tests/i915/gem_ccs.c
index f9ad9267df..d9d785ed9b 100644
--- a/tests/i915/gem_ccs.c
+++ b/tests/i915/gem_ccs.c
@@ -167,7 +167,7 @@ static void surf_copy(int i915,
 	ccs = gem_create(i915, ccssize);
 	ccs2 = gem_create(i915, ccssize);
 
-	surf.fd = i915;
+	blt_ctrl_surf_copy_init(i915, &surf);
 	surf.print_bb = param.print_bb;
 	set_surf_object(&surf.src, mid->handle, mid->region, mid->size,
 			uc_mocs, BLT_INDIRECT_ACCESS);
@@ -219,7 +219,7 @@ static void surf_copy(int i915,
 			uc_mocs, INDIRECT_ACCESS);
 	blt_ctrl_surf_copy(i915, ctx, e, ahnd, &surf);
 
-	memset(&blt, 0, sizeof(blt));
+	blt_copy_init(i915, &blt);
 	blt.color_depth = CD_32bit;
 	blt.print_bb = param.print_bb;
 	blt_set_copy_object(&blt.src, mid);
@@ -310,7 +310,7 @@ static int blt_block_copy3(int i915,
 	bb_offset = get_offset(ahnd, blt3->bb.handle, blt3->bb.size, alignment);
 
 	/* First blit src -> mid */
-	memset(&blt0, 0, sizeof(blt0));
+	blt_copy_init(i915, &blt0);
 	blt0.src = blt3->src;
 	blt0.dst = blt3->mid;
 	blt0.bb = blt3->bb;
@@ -321,7 +321,7 @@ static int blt_block_copy3(int i915,
 	bb_pos = emit_blt_block_copy(i915, ahnd, &blt0, &ext0, bb_pos, false);
 
 	/* Second blit mid -> dst */
-	memset(&blt0, 0, sizeof(blt0));
+	blt_copy_init(i915, &blt0);
 	blt0.src = blt3->mid;
 	blt0.dst = blt3->dst;
 	blt0.bb = blt3->bb;
@@ -332,7 +332,7 @@ static int blt_block_copy3(int i915,
 	bb_pos = emit_blt_block_copy(i915, ahnd, &blt0, &ext0, bb_pos, false);
 
 	/* Third blit dst -> final */
-	memset(&blt0, 0, sizeof(blt0));
+	blt_copy_init(i915, &blt0);
 	blt0.src = blt3->dst;
 	blt0.dst = blt3->final;
 	blt0.bb = blt3->bb;
@@ -390,11 +390,13 @@ static void block_copy(int i915,
 	if (!blt_uses_extended_block_copy(i915))
 		pext = NULL;
 
-	src = blt_create_object(i915, region1, width, height, bpp, uc_mocs,
+	blt_copy_init(i915, &blt);
+
+	src = blt_create_object(&blt, region1, width, height, bpp, uc_mocs,
 				T_LINEAR, COMPRESSION_DISABLED, comp_type, true);
-	mid = blt_create_object(i915, mid_region, width, height, bpp, uc_mocs,
+	mid = blt_create_object(&blt, mid_region, width, height, bpp, uc_mocs,
 				mid_tiling, mid_compression, comp_type, true);
-	dst = blt_create_object(i915, region1, width, height, bpp, uc_mocs,
+	dst = blt_create_object(&blt, region1, width, height, bpp, uc_mocs,
 				T_LINEAR, COMPRESSION_DISABLED, comp_type, true);
 	igt_assert(src->size == dst->size);
 	PRINT_SURFACE_INFO("src", src);
@@ -404,7 +406,6 @@ static void block_copy(int i915,
 	blt_surface_fill_rect(i915, src, width, height);
 	WRITE_PNG(i915, run_id, "src", src, width, height);
 
-	memset(&blt, 0, sizeof(blt));
 	blt.color_depth = CD_32bit;
 	blt.print_bb = param.print_bb;
 	blt_set_copy_object(&blt.src, src);
@@ -449,7 +450,7 @@ static void block_copy(int i915,
 		}
 	}
 
-	memset(&blt, 0, sizeof(blt));
+	blt_copy_init(i915, &blt);
 	blt.color_depth = CD_32bit;
 	blt.print_bb = param.print_bb;
 	blt_set_copy_object(&blt.src, mid);
@@ -486,6 +487,7 @@ static void block_multicopy(int i915,
 			    const struct test_config *config)
 {
 	struct blt_copy3_data blt3 = {};
+	struct blt_copy_data blt = {};
 	struct blt_block_copy3_data_ext ext3 = {}, *pext3 = &ext3;
 	struct blt_copy_object *src, *mid, *dst, *final;
 	const uint32_t bpp = 32;
@@ -505,13 +507,16 @@ static void block_multicopy(int i915,
 	if (!blt_uses_extended_block_copy(i915))
 		pext3 = NULL;
 
-	src = blt_create_object(i915, region1, width, height, bpp, uc_mocs,
+	/* For object creation */
+	blt_copy_init(i915, &blt);
+
+	src = blt_create_object(&blt, region1, width, height, bpp, uc_mocs,
 				T_LINEAR, COMPRESSION_DISABLED, comp_type, true);
-	mid = blt_create_object(i915, mid_region, width, height, bpp, uc_mocs,
+	mid = blt_create_object(&blt, mid_region, width, height, bpp, uc_mocs,
 				mid_tiling, mid_compression, comp_type, true);
-	dst = blt_create_object(i915, region1, width, height, bpp, uc_mocs,
+	dst = blt_create_object(&blt, region1, width, height, bpp, uc_mocs,
 				mid_tiling, COMPRESSION_DISABLED, comp_type, true);
-	final = blt_create_object(i915, region1, width, height, bpp, uc_mocs,
+	final = blt_create_object(&blt, region1, width, height, bpp, uc_mocs,
 				  T_LINEAR, COMPRESSION_DISABLED, comp_type, true);
 	igt_assert(src->size == dst->size);
 	PRINT_SURFACE_INFO("src", src);
@@ -521,7 +526,6 @@ static void block_multicopy(int i915,
 
 	blt_surface_fill_rect(i915, src, width, height);
 
-	memset(&blt3, 0, sizeof(blt3));
 	blt3.color_depth = CD_32bit;
 	blt3.print_bb = param.print_bb;
 	blt_set_copy_object(&blt3.src, src);
diff --git a/tests/i915/gem_exercise_blt.c b/tests/i915/gem_exercise_blt.c
index 0cd1820430..7355eabbe9 100644
--- a/tests/i915/gem_exercise_blt.c
+++ b/tests/i915/gem_exercise_blt.c
@@ -89,7 +89,7 @@ static int fast_copy_one_bb(int i915,
 	bb_offset = get_offset(ahnd, blt->bb.handle, blt->bb.size, alignment);
 
 	/* First blit */
-	memset(&blt_tmp, 0, sizeof(blt_tmp));
+	blt_copy_init(i915, &blt_tmp);
 	blt_tmp.src = blt->src;
 	blt_tmp.dst = blt->mid;
 	blt_tmp.bb = blt->bb;
@@ -98,7 +98,7 @@ static int fast_copy_one_bb(int i915,
 	bb_pos = emit_blt_fast_copy(i915, ahnd, &blt_tmp, bb_pos, false);
 
 	/* Second blit */
-	memset(&blt_tmp, 0, sizeof(blt_tmp));
+	blt_copy_init(i915, &blt_tmp);
 	blt_tmp.src = blt->mid;
 	blt_tmp.dst = blt->dst;
 	blt_tmp.bb = blt->bb;
@@ -140,6 +140,7 @@ static void fast_copy_emit(int i915, const intel_ctx_t *ctx,
 			   uint32_t region1, uint32_t region2,
 			   enum blt_tiling_type mid_tiling)
 {
+	struct blt_copy_data bltinit = {};
 	struct blt_fast_copy_data blt = {};
 	struct blt_copy_object *src, *mid, *dst;
 	const uint32_t bpp = 32;
@@ -152,11 +153,12 @@ static void fast_copy_emit(int i915, const intel_ctx_t *ctx,
 
 	igt_assert(__gem_create_in_memory_regions(i915, &bb, &bb_size, region1) == 0);
 
-	src = blt_create_object(i915, region1, width, height, bpp, 0,
+	blt_copy_init(i915, &bltinit);
+	src = blt_create_object(&bltinit, region1, width, height, bpp, 0,
 				T_LINEAR, COMPRESSION_DISABLED, 0, true);
-	mid = blt_create_object(i915, region2, width, height, bpp, 0,
+	mid = blt_create_object(&bltinit, region2, width, height, bpp, 0,
 				mid_tiling, COMPRESSION_DISABLED, 0, true);
-	dst = blt_create_object(i915, region1, width, height, bpp, 0,
+	dst = blt_create_object(&bltinit, region1, width, height, bpp, 0,
 				T_LINEAR, COMPRESSION_DISABLED, 0, true);
 	igt_assert(src->size == dst->size);
 
@@ -212,17 +214,17 @@ static void fast_copy(int i915, const intel_ctx_t *ctx,
 
 	igt_assert(__gem_create_in_memory_regions(i915, &bb, &bb_size, region1) == 0);
 
-	src = blt_create_object(i915, region1, width, height, bpp, 0,
+	blt_copy_init(i915, &blt);
+	src = blt_create_object(&blt, region1, width, height, bpp, 0,
 				T_LINEAR, COMPRESSION_DISABLED, 0, true);
-	mid = blt_create_object(i915, region2, width, height, bpp, 0,
+	mid = blt_create_object(&blt, region2, width, height, bpp, 0,
 				mid_tiling, COMPRESSION_DISABLED, 0, true);
-	dst = blt_create_object(i915, region1, width, height, bpp, 0,
+	dst = blt_create_object(&blt, region1, width, height, bpp, 0,
 				T_LINEAR, COMPRESSION_DISABLED, 0, true);
 	igt_assert(src->size == dst->size);
 
 	blt_surface_fill_rect(i915, src, width, height);
 
-	memset(&blt, 0, sizeof(blt));
 	blt.color_depth = CD_32bit;
 	blt.print_bb = param.print_bb;
 	blt_set_copy_object(&blt.src, src);
@@ -235,7 +237,7 @@ static void fast_copy(int i915, const intel_ctx_t *ctx,
 	WRITE_PNG(i915, mid_tiling, "src", &blt.src, width, height);
 	WRITE_PNG(i915, mid_tiling, "mid", &blt.dst, width, height);
 
-	memset(&blt, 0, sizeof(blt));
+	blt_copy_init(i915, &blt);
 	blt.color_depth = CD_32bit;
 	blt.print_bb = param.print_bb;
 	blt_set_copy_object(&blt.src, mid);
diff --git a/tests/i915/gem_lmem_swapping.c b/tests/i915/gem_lmem_swapping.c
index 83dbebec83..2921de8f9f 100644
--- a/tests/i915/gem_lmem_swapping.c
+++ b/tests/i915/gem_lmem_swapping.c
@@ -308,7 +308,7 @@ init_object_ccs(int i915, struct object *obj, struct blt_copy_object *tmp,
 		buf[j] = seed++;
 	munmap(buf, obj->size);
 
-	memset(&blt, 0, sizeof(blt));
+	blt_copy_init(i915, &blt);
 	blt.color_depth = CD_32bit;
 
 	memcpy(&blt.src, tmp, sizeof(blt.src));
@@ -366,7 +366,7 @@ verify_object_ccs(int i915, const struct object *obj,
 	cmd->handle = gem_create_from_pool(i915, &size, region);
 	blt_set_batch(cmd, cmd->handle, size, region);
 
-	memset(&blt, 0, sizeof(blt));
+	blt_copy_init(i915, &blt);
 	blt.color_depth = CD_32bit;
 
 	memcpy(&blt.src, obj->blt_obj, sizeof(blt.src));
-- 
2.34.1

[igt-dev] [PATCH i-g-t v3 13/17] lib/intel_blt: Extend blitter library to support xe driver

[Zbigniew Kempczyński]

Use already written for i915 blitter library in xe development.
Add appropriate code paths which are unique for those drivers.

Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
---
v3: Extract common helpers for getting mapping and alignment (Karolina)
---
 lib/intel_blt.c | 235 ++++++++++++++++++++++++++++++------------------
 lib/intel_blt.h |   2 +-
 2 files changed, 150 insertions(+), 87 deletions(-)

diff --git a/lib/intel_blt.c b/lib/intel_blt.c
index f2f86e4947..5387a60ee3 100644
--- a/lib/intel_blt.c
+++ b/lib/intel_blt.c
@@ -9,9 +9,13 @@
 #include <malloc.h>
 #include <cairo.h>
 #include "drm.h"
-#include "igt.h"
 #include "i915/gem_create.h"
+#include "igt.h"
+#include "igt_syncobj.h"
 #include "intel_blt.h"
+#include "xe/xe_ioctl.h"
+#include "xe/xe_query.h"
+#include "xe/xe_util.h"
 
 #define BITRANGE(start, end) (end - start + 1)
 #define GET_CMDS_INFO(__fd) intel_get_cmds_info(intel_get_drm_devid(__fd))
@@ -468,24 +472,40 @@ static int __special_mode(const struct blt_copy_data *blt)
 	return SM_NONE;
 }
 
-static int __memory_type(uint32_t region)
+static int __memory_type(int fd, enum intel_driver driver, uint32_t region)
 {
-	igt_assert_f(IS_DEVICE_MEMORY_REGION(region) ||
-		     IS_SYSTEM_MEMORY_REGION(region),
-		     "Invalid region: %x\n", region);
+	if (driver == INTEL_DRIVER_I915) {
+		igt_assert_f(IS_DEVICE_MEMORY_REGION(region) ||
+			     IS_SYSTEM_MEMORY_REGION(region),
+			     "Invalid region: %x\n", region);
+	} else {
+		igt_assert_f(XE_IS_VRAM_MEMORY_REGION(fd, region) ||
+			     XE_IS_SYSMEM_MEMORY_REGION(fd, region),
+			     "Invalid region: %x\n", region);
+	}
 
-	if (IS_DEVICE_MEMORY_REGION(region))
+	if (driver == INTEL_DRIVER_I915 && IS_DEVICE_MEMORY_REGION(region))
 		return TM_LOCAL_MEM;
+	else if (driver == INTEL_DRIVER_XE && XE_IS_VRAM_MEMORY_REGION(fd, region))
+		return TM_LOCAL_MEM;
+
 	return TM_SYSTEM_MEM;
 }
 
-static enum blt_aux_mode __aux_mode(const struct blt_copy_object *obj)
+static enum blt_aux_mode __aux_mode(int fd,
+				    enum intel_driver driver,
+				    const struct blt_copy_object *obj)
 {
-	if (obj->compression == COMPRESSION_ENABLED) {
+	if (driver == INTEL_DRIVER_I915 && obj->compression == COMPRESSION_ENABLED) {
 		igt_assert_f(IS_DEVICE_MEMORY_REGION(obj->region),
 			     "XY_BLOCK_COPY_BLT supports compression "
 			     "on device memory only\n");
 		return AM_AUX_CCS_E;
+	} else if (driver == INTEL_DRIVER_XE && obj->compression == COMPRESSION_ENABLED) {
+		igt_assert_f(XE_IS_VRAM_MEMORY_REGION(fd, obj->region),
+			     "XY_BLOCK_COPY_BLT supports compression "
+			     "on device memory only\n");
+		return AM_AUX_CCS_E;
 	}
 
 	return AM_AUX_NONE;
@@ -508,9 +528,9 @@ static void fill_data(struct gen12_block_copy_data *data,
 	data->dw00.length = extended_command ? 20 : 10;
 
 	if (__special_mode(blt) == SM_FULL_RESOLVE)
-		data->dw01.dst_aux_mode = __aux_mode(&blt->src);
+		data->dw01.dst_aux_mode = __aux_mode(blt->fd, blt->driver, &blt->src);
 	else
-		data->dw01.dst_aux_mode = __aux_mode(&blt->dst);
+		data->dw01.dst_aux_mode = __aux_mode(blt->fd, blt->driver, &blt->dst);
 	data->dw01.dst_pitch = blt->dst.pitch - 1;
 
 	data->dw01.dst_mocs = blt->dst.mocs;
@@ -531,13 +551,13 @@ static void fill_data(struct gen12_block_copy_data *data,
 
 	data->dw06.dst_x_offset = blt->dst.x_offset;
 	data->dw06.dst_y_offset = blt->dst.y_offset;
-	data->dw06.dst_target_memory = __memory_type(blt->dst.region);
+	data->dw06.dst_target_memory = __memory_type(blt->fd, blt->driver, blt->dst.region);
 
 	data->dw07.src_x1 = blt->src.x1;
 	data->dw07.src_y1 = blt->src.y1;
 
 	data->dw08.src_pitch = blt->src.pitch - 1;
-	data->dw08.src_aux_mode = __aux_mode(&blt->src);
+	data->dw08.src_aux_mode = __aux_mode(blt->fd, blt->driver, &blt->src);
 	data->dw08.src_mocs = blt->src.mocs;
 	data->dw08.src_compression = blt->src.compression;
 	data->dw08.src_tiling = __block_tiling(blt->src.tiling);
@@ -550,7 +570,7 @@ static void fill_data(struct gen12_block_copy_data *data,
 
 	data->dw11.src_x_offset = blt->src.x_offset;
 	data->dw11.src_y_offset = blt->src.y_offset;
-	data->dw11.src_target_memory = __memory_type(blt->src.region);
+	data->dw11.src_target_memory = __memory_type(blt->fd, blt->driver, blt->src.region);
 }
 
 static void fill_data_ext(struct gen12_block_copy_data_ext *dext,
@@ -692,6 +712,24 @@ static void dump_bb_ext(struct gen12_block_copy_data_ext *data)
 		 data->dw21.src_array_index);
 }
 
+static uint64_t get_default_alignment(int fd, enum intel_driver driver)
+{
+	if (driver == INTEL_DRIVER_XE)
+		return xe_get_default_alignment(fd);
+
+	return gem_detect_safe_alignment(fd);
+}
+
+static void *bo_map(int fd, uint32_t handle, uint64_t size,
+		    enum intel_driver driver)
+{
+	if (driver == INTEL_DRIVER_XE)
+		return xe_bo_map(fd, handle, size);
+
+	return gem_mmap__device_coherent(fd, handle, 0, size,
+					 PROT_READ | PROT_WRITE);
+}
+
 /**
  * blt_copy_init:
  * @fd: drm fd
@@ -739,7 +777,7 @@ uint64_t emit_blt_block_copy(int fd,
 	igt_assert_f(ahnd, "block-copy supports softpin only\n");
 	igt_assert_f(blt, "block-copy requires data to do blit\n");
 
-	alignment = gem_detect_safe_alignment(fd);
+	alignment = get_default_alignment(fd, blt->driver);
 	src_offset = get_offset(ahnd, blt->src.handle, blt->src.size, alignment)
 		     + blt->src.plane_offset;
 	dst_offset = get_offset(ahnd, blt->dst.handle, blt->dst.size, alignment)
@@ -748,8 +786,7 @@ uint64_t emit_blt_block_copy(int fd,
 
 	fill_data(&data, blt, src_offset, dst_offset, ext);
 
-	bb = gem_mmap__device_coherent(fd, blt->bb.handle, 0, blt->bb.size,
-				       PROT_READ | PROT_WRITE);
+	bb = bo_map(fd, blt->bb.handle, blt->bb.size, blt->driver);
 
 	igt_assert(bb_pos + sizeof(data) < blt->bb.size);
 	memcpy(bb + bb_pos, &data, sizeof(data));
@@ -812,29 +849,35 @@ int blt_block_copy(int fd,
 
 	igt_assert_f(ahnd, "block-copy supports softpin only\n");
 	igt_assert_f(blt, "block-copy requires data to do blit\n");
+	igt_assert_neq(blt->driver, 0);
 
-	alignment = gem_detect_safe_alignment(fd);
+	alignment = get_default_alignment(fd, blt->driver);
 	src_offset = get_offset(ahnd, blt->src.handle, blt->src.size, alignment);
 	dst_offset = get_offset(ahnd, blt->dst.handle, blt->dst.size, alignment);
 	bb_offset = get_offset(ahnd, blt->bb.handle, blt->bb.size, alignment);
 
 	emit_blt_block_copy(fd, ahnd, blt, ext, 0, true);
 
-	obj[0].offset = CANONICAL(dst_offset);
-	obj[1].offset = CANONICAL(src_offset);
-	obj[2].offset = CANONICAL(bb_offset);
-	obj[0].handle = blt->dst.handle;
-	obj[1].handle = blt->src.handle;
-	obj[2].handle = blt->bb.handle;
-	obj[0].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_WRITE |
-		       EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
-	obj[1].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
-	obj[2].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
-	execbuf.buffer_count = 3;
-	execbuf.buffers_ptr = to_user_pointer(obj);
-	execbuf.rsvd1 = ctx ? ctx->id : 0;
-	execbuf.flags = e ? e->flags : I915_EXEC_BLT;
-	ret = __gem_execbuf(fd, &execbuf);
+	if (blt->driver == INTEL_DRIVER_XE) {
+		intel_ctx_xe_exec(ctx, ahnd, CANONICAL(bb_offset));
+	} else {
+		obj[0].offset = CANONICAL(dst_offset);
+		obj[1].offset = CANONICAL(src_offset);
+		obj[2].offset = CANONICAL(bb_offset);
+		obj[0].handle = blt->dst.handle;
+		obj[1].handle = blt->src.handle;
+		obj[2].handle = blt->bb.handle;
+		obj[0].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_WRITE |
+				EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
+		obj[1].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
+		obj[2].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
+		execbuf.buffer_count = 3;
+		execbuf.buffers_ptr = to_user_pointer(obj);
+		execbuf.rsvd1 = ctx ? ctx->id : 0;
+		execbuf.flags = e ? e->flags : I915_EXEC_BLT;
+
+		ret = __gem_execbuf(fd, &execbuf);
+	}
 
 	return ret;
 }
@@ -950,7 +993,7 @@ uint64_t emit_blt_ctrl_surf_copy(int fd,
 	igt_assert_f(ahnd, "ctrl-surf-copy supports softpin only\n");
 	igt_assert_f(surf, "ctrl-surf-copy requires data to do ctrl-surf-copy blit\n");
 
-	alignment = max_t(uint64_t, gem_detect_safe_alignment(fd), 1ull << 16);
+	alignment = max_t(uint64_t, get_default_alignment(fd, surf->driver), 1ull << 16);
 
 	data.dw00.client = 0x2;
 	data.dw00.opcode = 0x48;
@@ -973,8 +1016,7 @@ uint64_t emit_blt_ctrl_surf_copy(int fd,
 	data.dw04.dst_address_hi = dst_offset >> 32;
 	data.dw04.dst_mocs = surf->dst.mocs;
 
-	bb = gem_mmap__device_coherent(fd, surf->bb.handle, 0, surf->bb.size,
-				       PROT_READ | PROT_WRITE);
+	bb = bo_map(fd, surf->bb.handle, surf->bb.size, surf->driver);
 
 	igt_assert(bb_pos + sizeof(data) < surf->bb.size);
 	memcpy(bb + bb_pos, &data, sizeof(data));
@@ -1002,7 +1044,7 @@ uint64_t emit_blt_ctrl_surf_copy(int fd,
 
 /**
  * blt_ctrl_surf_copy:
- * @fd: drm fd
+ * @blt: bldrm fd
  * @ctx: intel_ctx_t context
  * @e: blitter engine for @ctx
  * @ahnd: allocator handle
@@ -1026,32 +1068,37 @@ int blt_ctrl_surf_copy(int fd,
 
 	igt_assert_f(ahnd, "ctrl-surf-copy supports softpin only\n");
 	igt_assert_f(surf, "ctrl-surf-copy requires data to do ctrl-surf-copy blit\n");
+	igt_assert_neq(surf->driver, 0);
 
-	alignment = max_t(uint64_t, gem_detect_safe_alignment(fd), 1ull << 16);
+	alignment = max_t(uint64_t, get_default_alignment(fd, surf->driver), 1ull << 16);
 	src_offset = get_offset(ahnd, surf->src.handle, surf->src.size, alignment);
 	dst_offset = get_offset(ahnd, surf->dst.handle, surf->dst.size, alignment);
 	bb_offset = get_offset(ahnd, surf->bb.handle, surf->bb.size, alignment);
 
 	emit_blt_ctrl_surf_copy(fd, ahnd, surf, 0, true);
 
-	obj[0].offset = CANONICAL(dst_offset);
-	obj[1].offset = CANONICAL(src_offset);
-	obj[2].offset = CANONICAL(bb_offset);
-	obj[0].handle = surf->dst.handle;
-	obj[1].handle = surf->src.handle;
-	obj[2].handle = surf->bb.handle;
-	obj[0].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_WRITE |
-		       EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
-	obj[1].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
-	obj[2].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
-	execbuf.buffer_count = 3;
-	execbuf.buffers_ptr = to_user_pointer(obj);
-	execbuf.flags = e ? e->flags : I915_EXEC_BLT;
-	execbuf.rsvd1 = ctx ? ctx->id : 0;
-	gem_execbuf(fd, &execbuf);
-	put_offset(ahnd, surf->dst.handle);
-	put_offset(ahnd, surf->src.handle);
-	put_offset(ahnd, surf->bb.handle);
+	if (surf->driver == INTEL_DRIVER_XE) {
+		intel_ctx_xe_exec(ctx, ahnd, CANONICAL(bb_offset));
+	} else {
+		obj[0].offset = CANONICAL(dst_offset);
+		obj[1].offset = CANONICAL(src_offset);
+		obj[2].offset = CANONICAL(bb_offset);
+		obj[0].handle = surf->dst.handle;
+		obj[1].handle = surf->src.handle;
+		obj[2].handle = surf->bb.handle;
+		obj[0].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_WRITE |
+				EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
+		obj[1].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
+		obj[2].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
+		execbuf.buffer_count = 3;
+		execbuf.buffers_ptr = to_user_pointer(obj);
+		execbuf.flags = e ? e->flags : I915_EXEC_BLT;
+		execbuf.rsvd1 = ctx ? ctx->id : 0;
+		gem_execbuf(fd, &execbuf);
+		put_offset(ahnd, surf->dst.handle);
+		put_offset(ahnd, surf->src.handle);
+		put_offset(ahnd, surf->bb.handle);
+	}
 
 	return 0;
 }
@@ -1208,7 +1255,7 @@ uint64_t emit_blt_fast_copy(int fd,
 	uint32_t bbe = MI_BATCH_BUFFER_END;
 	uint32_t *bb;
 
-	alignment = gem_detect_safe_alignment(fd);
+	alignment = get_default_alignment(fd, blt->driver);
 
 	data.dw00.client = 0x2;
 	data.dw00.opcode = 0x42;
@@ -1218,8 +1265,8 @@ uint64_t emit_blt_fast_copy(int fd,
 
 	data.dw01.dst_pitch = blt->dst.pitch;
 	data.dw01.color_depth = __fast_color_depth(blt->color_depth);
-	data.dw01.dst_memory = __memory_type(blt->dst.region);
-	data.dw01.src_memory = __memory_type(blt->src.region);
+	data.dw01.dst_memory = __memory_type(blt->fd, blt->driver, blt->dst.region);
+	data.dw01.src_memory = __memory_type(blt->fd, blt->driver, blt->src.region);
 	data.dw01.dst_type_y = __new_tile_y_type(blt->dst.tiling) ? 1 : 0;
 	data.dw01.src_type_y = __new_tile_y_type(blt->src.tiling) ? 1 : 0;
 
@@ -1246,8 +1293,7 @@ uint64_t emit_blt_fast_copy(int fd,
 	data.dw08.src_address_lo = src_offset;
 	data.dw09.src_address_hi = src_offset >> 32;
 
-	bb = gem_mmap__device_coherent(fd, blt->bb.handle, 0, blt->bb.size,
-				       PROT_READ | PROT_WRITE);
+	bb = bo_map(fd, blt->bb.handle, blt->bb.size, blt->driver);
 
 	igt_assert(bb_pos + sizeof(data) < blt->bb.size);
 	memcpy(bb + bb_pos, &data, sizeof(data));
@@ -1297,32 +1343,39 @@ int blt_fast_copy(int fd,
 	uint64_t dst_offset, src_offset, bb_offset, alignment;
 	int ret;
 
-	alignment = gem_detect_safe_alignment(fd);
+	igt_assert_f(ahnd, "fast-copy supports softpin only\n");
+	igt_assert_f(blt, "fast-copy requires data to do fast-copy blit\n");
+	igt_assert_neq(blt->driver, 0);
 
+	alignment = get_default_alignment(fd, blt->driver);
 	src_offset = get_offset(ahnd, blt->src.handle, blt->src.size, alignment);
 	dst_offset = get_offset(ahnd, blt->dst.handle, blt->dst.size, alignment);
 	bb_offset = get_offset(ahnd, blt->bb.handle, blt->bb.size, alignment);
 
 	emit_blt_fast_copy(fd, ahnd, blt, 0, true);
 
-	obj[0].offset = CANONICAL(dst_offset);
-	obj[1].offset = CANONICAL(src_offset);
-	obj[2].offset = CANONICAL(bb_offset);
-	obj[0].handle = blt->dst.handle;
-	obj[1].handle = blt->src.handle;
-	obj[2].handle = blt->bb.handle;
-	obj[0].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_WRITE |
-		       EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
-	obj[1].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
-	obj[2].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
-	execbuf.buffer_count = 3;
-	execbuf.buffers_ptr = to_user_pointer(obj);
-	execbuf.rsvd1 = ctx ? ctx->id : 0;
-	execbuf.flags = e ? e->flags : I915_EXEC_BLT;
-	ret = __gem_execbuf(fd, &execbuf);
-	put_offset(ahnd, blt->dst.handle);
-	put_offset(ahnd, blt->src.handle);
-	put_offset(ahnd, blt->bb.handle);
+	if (blt->driver == INTEL_DRIVER_XE) {
+		intel_ctx_xe_exec(ctx, ahnd, CANONICAL(bb_offset));
+	} else {
+		obj[0].offset = CANONICAL(dst_offset);
+		obj[1].offset = CANONICAL(src_offset);
+		obj[2].offset = CANONICAL(bb_offset);
+		obj[0].handle = blt->dst.handle;
+		obj[1].handle = blt->src.handle;
+		obj[2].handle = blt->bb.handle;
+		obj[0].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_WRITE |
+				EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
+		obj[1].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
+		obj[2].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
+		execbuf.buffer_count = 3;
+		execbuf.buffers_ptr = to_user_pointer(obj);
+		execbuf.rsvd1 = ctx ? ctx->id : 0;
+		execbuf.flags = e ? e->flags : I915_EXEC_BLT;
+		ret = __gem_execbuf(fd, &execbuf);
+		put_offset(ahnd, blt->dst.handle);
+		put_offset(ahnd, blt->src.handle);
+		put_offset(ahnd, blt->bb.handle);
+	}
 
 	return ret;
 }
@@ -1366,16 +1419,21 @@ blt_create_object(const struct blt_copy_data *blt, uint32_t region,
 	obj = calloc(1, sizeof(*obj));
 
 	obj->size = size;
-	igt_assert(__gem_create_in_memory_regions(blt->fd, &handle,
-						  &size, region) == 0);
+
+	if (blt->driver == INTEL_DRIVER_XE) {
+		size = ALIGN(size, xe_get_default_alignment(blt->fd));
+		handle = xe_bo_create_flags(blt->fd, 0, size, region);
+	} else {
+		igt_assert(__gem_create_in_memory_regions(blt->fd, &handle,
+							  &size, region) == 0);
+	}
 
 	blt_set_object(obj, handle, size, region, mocs, tiling,
 		       compression, compression_type);
 	blt_set_geom(obj, stride, 0, 0, width, height, 0, 0);
 
 	if (create_mapping)
-		obj->ptr = gem_mmap__device_coherent(blt->fd, handle, 0, size,
-						     PROT_READ | PROT_WRITE);
+		obj->ptr = bo_map(blt->fd, handle, size, blt->driver);
 
 	return obj;
 }
@@ -1518,14 +1576,19 @@ void blt_surface_to_png(int fd, uint32_t run_id, const char *fileid,
 	int format;
 	int stride = obj->tiling ? obj->pitch * 4 : obj->pitch;
 	char filename[FILENAME_MAX];
+	bool is_xe = is_xe_device(fd);
 
 	snprintf(filename, FILENAME_MAX-1, "%d-%s-%s-%ux%u-%s.png",
 		 run_id, fileid, blt_tiling_name(obj->tiling), width, height,
 		 obj->compression ? "compressed" : "uncompressed");
 
-	if (!map)
-		map = gem_mmap__device_coherent(fd, obj->handle, 0,
-						obj->size, PROT_READ);
+	if (!map) {
+		if (is_xe)
+			map = xe_bo_map(fd, obj->handle, obj->size);
+		else
+			map = gem_mmap__device_coherent(fd, obj->handle, 0,
+							obj->size, PROT_READ);
+	}
 	format = CAIRO_FORMAT_RGB24;
 	surface = cairo_image_surface_create_for_data(map,
 						      format, width, height,
diff --git a/lib/intel_blt.h b/lib/intel_blt.h
index 7516ce8ac7..944e2b4ae7 100644
--- a/lib/intel_blt.h
+++ b/lib/intel_blt.h
@@ -8,7 +8,7 @@
 
 /**
  * SECTION:intel_blt
- * @short_description: i915 blitter library
+ * @short_description: i915/xe blitter library
  * @title: Blitter library
  * @include: intel_blt.h
  *
-- 
2.34.1

[igt-dev] [PATCH i-g-t v3 14/17] tests/xe_ccs: Check if flatccs is working with block-copy for Xe

[Zbigniew Kempczyński]

This is ported to xe copy of i915 gem_ccs test. Ported means all driver
dependent calls - like working on regions, binding and execution were
replaced by xe counterparts. I wondered to add conditionals for xe
in gem_ccs but this would decrease test readability so I dropped
this idea.

Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
---
v3: Drop unused macro, use system_memory() instead REGION_SMEM
    in xe test (Karolina)
---
 tests/meson.build |   1 +
 tests/xe/xe_ccs.c | 757 ++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 758 insertions(+)
 create mode 100644 tests/xe/xe_ccs.c

diff --git a/tests/meson.build b/tests/meson.build
index d56e4c89da..da8653fabc 100644
--- a/tests/meson.build
+++ b/tests/meson.build
@@ -263,6 +263,7 @@ i915_progs = [
 ]
 
 xe_progs = [
+	'xe_ccs',
 	'xe_create',
 	'xe_compute',
 	'xe_dma_buf_sync',
diff --git a/tests/xe/xe_ccs.c b/tests/xe/xe_ccs.c
new file mode 100644
index 0000000000..ee653790bf
--- /dev/null
+++ b/tests/xe/xe_ccs.c
@@ -0,0 +1,757 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+
+#include <errno.h>
+#include <glib.h>
+#include <sys/ioctl.h>
+#include <sys/time.h>
+#include <malloc.h>
+#include "drm.h"
+#include "igt.h"
+#include "igt_syncobj.h"
+#include "intel_blt.h"
+#include "intel_mocs.h"
+#include "xe/xe_ioctl.h"
+#include "xe/xe_query.h"
+#include "xe/xe_util.h"
+/**
+ * TEST: xe ccs
+ * Description: Exercise gen12 blitter with and without flatccs compression on Xe
+ * Run type: FULL
+ *
+ * SUBTEST: block-copy-compressed
+ * Description: Check block-copy flatccs compressed blit
+ *
+ * SUBTEST: block-copy-uncompressed
+ * Description: Check block-copy uncompressed blit
+ *
+ * SUBTEST: block-multicopy-compressed
+ * Description: Check block-multicopy flatccs compressed blit
+ *
+ * SUBTEST: block-multicopy-inplace
+ * Description: Check block-multicopy flatccs inplace decompression blit
+ *
+ * SUBTEST: ctrl-surf-copy
+ * Description: Check flatccs data can be copied from/to surface
+ *
+ * SUBTEST: ctrl-surf-copy-new-ctx
+ * Description: Check flatccs data are physically tagged and visible in vm
+ *
+ * SUBTEST: suspend-resume
+ * Description: Check flatccs data persists after suspend / resume (S0)
+ */
+
+IGT_TEST_DESCRIPTION("Exercise gen12 blitter with and without flatccs compression on Xe");
+
+static struct param {
+	int compression_format;
+	int tiling;
+	bool write_png;
+	bool print_bb;
+	bool print_surface_info;
+	int width;
+	int height;
+} param = {
+	.compression_format = 0,
+	.tiling = -1,
+	.write_png = false,
+	.print_bb = false,
+	.print_surface_info = false,
+	.width = 512,
+	.height = 512,
+};
+
+struct test_config {
+	bool compression;
+	bool inplace;
+	bool surfcopy;
+	bool new_ctx;
+	bool suspend_resume;
+};
+
+static void set_surf_object(struct blt_ctrl_surf_copy_object *obj,
+			    uint32_t handle, uint32_t region, uint64_t size,
+			    uint8_t mocs, enum blt_access_type access_type)
+{
+	obj->handle = handle;
+	obj->region = region;
+	obj->size = size;
+	obj->mocs = mocs;
+	obj->access_type = access_type;
+}
+
+#define PRINT_SURFACE_INFO(name, obj) do { \
+	if (param.print_surface_info) \
+		blt_surface_info((name), (obj)); } while (0)
+
+#define WRITE_PNG(fd, id, name, obj, w, h) do { \
+	if (param.write_png) \
+		blt_surface_to_png((fd), (id), (name), (obj), (w), (h)); } while (0)
+
+static int compare_nxn(const struct blt_copy_object *surf1,
+		       const struct blt_copy_object *surf2,
+		       int xsize, int ysize, int bx, int by)
+{
+	int x, y, corrupted;
+	uint32_t pos, px1, px2;
+
+	corrupted = 0;
+	for (y = 0; y < ysize; y++) {
+		for (x = 0; x < xsize; x++) {
+			pos = bx * xsize + by * ysize * surf1->pitch / 4;
+			pos += x + y * surf1->pitch / 4;
+			px1 = surf1->ptr[pos];
+			px2 = surf2->ptr[pos];
+			if (px1 != px2)
+				corrupted++;
+		}
+	}
+
+	return corrupted;
+}
+
+static void dump_corruption_info(const struct blt_copy_object *surf1,
+				 const struct blt_copy_object *surf2)
+{
+	const int xsize = 8, ysize = 8;
+	int w, h, bx, by, corrupted;
+
+	igt_assert(surf1->x1 == surf2->x1 && surf1->x2 == surf2->x2);
+	igt_assert(surf1->y1 == surf2->y1 && surf1->y2 == surf2->y2);
+	w = surf1->x2;
+	h = surf1->y2;
+
+	igt_info("dump corruption - width: %d, height: %d, sizex: %x, sizey: %x\n",
+		 surf1->x2, surf1->y2, xsize, ysize);
+
+	for (by = 0; by < h / ysize; by++) {
+		for (bx = 0; bx < w / xsize; bx++) {
+			corrupted = compare_nxn(surf1, surf2, xsize, ysize, bx, by);
+			if (corrupted == 0)
+				igt_info(".");
+			else
+				igt_info("%c", '0' + corrupted);
+		}
+		igt_info("\n");
+	}
+}
+
+static void surf_copy(int xe,
+		      intel_ctx_t *ctx,
+		      uint64_t ahnd,
+		      const struct blt_copy_object *src,
+		      const struct blt_copy_object *mid,
+		      const struct blt_copy_object *dst,
+		      int run_id, bool suspend_resume)
+{
+	struct blt_copy_data blt = {};
+	struct blt_block_copy_data_ext ext = {};
+	struct blt_ctrl_surf_copy_data surf = {};
+	uint32_t bb1, bb2, ccs, ccs2, *ccsmap, *ccsmap2;
+	uint64_t bb_size, ccssize = mid->size / CCS_RATIO;
+	uint32_t *ccscopy;
+	uint8_t uc_mocs = intel_get_uc_mocs(xe);
+	uint32_t sysmem = system_memory(xe);
+	int result;
+
+	igt_assert(mid->compression);
+	ccscopy = (uint32_t *) malloc(ccssize);
+	ccs = xe_bo_create_flags(xe, 0, ccssize, sysmem);
+	ccs2 = xe_bo_create_flags(xe, 0, ccssize, sysmem);
+
+	blt_ctrl_surf_copy_init(xe, &surf);
+	surf.print_bb = param.print_bb;
+	set_surf_object(&surf.src, mid->handle, mid->region, mid->size,
+			uc_mocs, BLT_INDIRECT_ACCESS);
+	set_surf_object(&surf.dst, ccs, sysmem, ccssize, uc_mocs, DIRECT_ACCESS);
+	bb_size = xe_get_default_alignment(xe);
+	bb1 = xe_bo_create_flags(xe, 0, bb_size, sysmem);
+	blt_set_batch(&surf.bb, bb1, bb_size, sysmem);
+	blt_ctrl_surf_copy(xe, ctx, NULL, ahnd, &surf);
+	intel_ctx_xe_sync(ctx, true);
+
+	ccsmap = xe_bo_map(xe, ccs, surf.dst.size);
+	memcpy(ccscopy, ccsmap, ccssize);
+
+	if (suspend_resume) {
+		char *orig, *orig2, *newsum, *newsum2;
+
+		orig = g_compute_checksum_for_data(G_CHECKSUM_SHA1,
+						   (void *)ccsmap, surf.dst.size);
+		orig2 = g_compute_checksum_for_data(G_CHECKSUM_SHA1,
+						    (void *)mid->ptr, mid->size);
+
+		igt_system_suspend_autoresume(SUSPEND_STATE_FREEZE, SUSPEND_TEST_NONE);
+
+		set_surf_object(&surf.dst, ccs2, system_memory(xe), ccssize,
+				0, DIRECT_ACCESS);
+		blt_ctrl_surf_copy(xe, ctx, NULL, ahnd, &surf);
+		intel_ctx_xe_sync(ctx, true);
+
+		ccsmap2 = xe_bo_map(xe, ccs2, surf.dst.size);
+		newsum = g_compute_checksum_for_data(G_CHECKSUM_SHA1,
+						     (void *)ccsmap2, surf.dst.size);
+		newsum2 = g_compute_checksum_for_data(G_CHECKSUM_SHA1,
+						      (void *)mid->ptr, mid->size);
+
+		munmap(ccsmap2, ccssize);
+		igt_assert(!strcmp(orig, newsum));
+		igt_assert(!strcmp(orig2, newsum2));
+		g_free(orig);
+		g_free(orig2);
+		g_free(newsum);
+		g_free(newsum2);
+	}
+
+	/* corrupt ccs */
+	for (int i = 0; i < surf.dst.size / sizeof(uint32_t); i++)
+		ccsmap[i] = i;
+	set_surf_object(&surf.src, ccs, sysmem, ccssize,
+			uc_mocs, DIRECT_ACCESS);
+	set_surf_object(&surf.dst, mid->handle, mid->region, mid->size,
+			uc_mocs, INDIRECT_ACCESS);
+	blt_ctrl_surf_copy(xe, ctx, NULL, ahnd, &surf);
+	intel_ctx_xe_sync(ctx, true);
+
+	blt_copy_init(xe, &blt);
+	blt.color_depth = CD_32bit;
+	blt.print_bb = param.print_bb;
+	blt_set_copy_object(&blt.src, mid);
+	blt_set_copy_object(&blt.dst, dst);
+	blt_set_object_ext(&ext.src, mid->compression_type, mid->x2, mid->y2, SURFACE_TYPE_2D);
+	blt_set_object_ext(&ext.dst, 0, dst->x2, dst->y2, SURFACE_TYPE_2D);
+	bb2 = xe_bo_create_flags(xe, 0, bb_size, sysmem);
+	blt_set_batch(&blt.bb, bb2, bb_size, sysmem);
+	blt_block_copy(xe, ctx, NULL, ahnd, &blt, &ext);
+	intel_ctx_xe_sync(ctx, true);
+	WRITE_PNG(xe, run_id, "corrupted", &blt.dst, dst->x2, dst->y2);
+	result = memcmp(src->ptr, dst->ptr, src->size);
+	igt_assert(result != 0);
+
+	/* retrieve back ccs */
+	memcpy(ccsmap, ccscopy, ccssize);
+	blt_ctrl_surf_copy(xe, ctx, NULL, ahnd, &surf);
+
+	blt_block_copy(xe, ctx, NULL, ahnd, &blt, &ext);
+	intel_ctx_xe_sync(ctx, true);
+	WRITE_PNG(xe, run_id, "corrected", &blt.dst, dst->x2, dst->y2);
+	result = memcmp(src->ptr, dst->ptr, src->size);
+	if (result)
+		dump_corruption_info(src, dst);
+
+	munmap(ccsmap, ccssize);
+	gem_close(xe, ccs);
+	gem_close(xe, ccs2);
+	gem_close(xe, bb1);
+	gem_close(xe, bb2);
+
+	igt_assert_f(result == 0,
+		     "Source and destination surfaces are different after "
+		     "restoring source ccs data\n");
+}
+
+struct blt_copy3_data {
+	int xe;
+	struct blt_copy_object src;
+	struct blt_copy_object mid;
+	struct blt_copy_object dst;
+	struct blt_copy_object final;
+	struct blt_copy_batch bb;
+	enum blt_color_depth color_depth;
+
+	/* debug stuff */
+	bool print_bb;
+};
+
+struct blt_block_copy3_data_ext {
+	struct blt_block_copy_object_ext src;
+	struct blt_block_copy_object_ext mid;
+	struct blt_block_copy_object_ext dst;
+	struct blt_block_copy_object_ext final;
+};
+
+static int blt_block_copy3(int xe,
+			   const intel_ctx_t *ctx,
+			   uint64_t ahnd,
+			   const struct blt_copy3_data *blt3,
+			   const struct blt_block_copy3_data_ext *ext3)
+{
+	struct blt_copy_data blt0;
+	struct blt_block_copy_data_ext ext0;
+	uint64_t bb_offset, alignment;
+	uint64_t bb_pos = 0;
+	int ret;
+
+	igt_assert_f(ahnd, "block-copy3 supports softpin only\n");
+	igt_assert_f(blt3, "block-copy3 requires data to do blit\n");
+
+	alignment = xe_get_default_alignment(xe);
+	get_offset(ahnd, blt3->src.handle, blt3->src.size, alignment);
+	get_offset(ahnd, blt3->mid.handle, blt3->mid.size, alignment);
+	get_offset(ahnd, blt3->dst.handle, blt3->dst.size, alignment);
+	get_offset(ahnd, blt3->final.handle, blt3->final.size, alignment);
+	bb_offset = get_offset(ahnd, blt3->bb.handle, blt3->bb.size, alignment);
+
+	/* First blit src -> mid */
+	blt_copy_init(xe, &blt0);
+	blt0.src = blt3->src;
+	blt0.dst = blt3->mid;
+	blt0.bb = blt3->bb;
+	blt0.color_depth = blt3->color_depth;
+	blt0.print_bb = blt3->print_bb;
+	ext0.src = ext3->src;
+	ext0.dst = ext3->mid;
+	bb_pos = emit_blt_block_copy(xe, ahnd, &blt0, &ext0, bb_pos, false);
+
+	/* Second blit mid -> dst */
+	blt_copy_init(xe, &blt0);
+	blt0.src = blt3->mid;
+	blt0.dst = blt3->dst;
+	blt0.bb = blt3->bb;
+	blt0.color_depth = blt3->color_depth;
+	blt0.print_bb = blt3->print_bb;
+	ext0.src = ext3->mid;
+	ext0.dst = ext3->dst;
+	bb_pos = emit_blt_block_copy(xe, ahnd, &blt0, &ext0, bb_pos, false);
+
+	/* Third blit dst -> final */
+	blt_copy_init(xe, &blt0);
+	blt0.src = blt3->dst;
+	blt0.dst = blt3->final;
+	blt0.bb = blt3->bb;
+	blt0.color_depth = blt3->color_depth;
+	blt0.print_bb = blt3->print_bb;
+	ext0.src = ext3->dst;
+	ext0.dst = ext3->final;
+	bb_pos = emit_blt_block_copy(xe, ahnd, &blt0, &ext0, bb_pos, true);
+
+	intel_ctx_xe_exec(ctx, ahnd, bb_offset);
+
+	return ret;
+}
+
+static void block_copy(int xe,
+		       intel_ctx_t *ctx,
+		       uint32_t region1, uint32_t region2,
+		       enum blt_tiling_type mid_tiling,
+		       const struct test_config *config)
+{
+	struct blt_copy_data blt = {};
+	struct blt_block_copy_data_ext ext = {}, *pext = &ext;
+	struct blt_copy_object *src, *mid, *dst;
+	const uint32_t bpp = 32;
+	uint64_t bb_size = xe_get_default_alignment(xe);
+	uint64_t ahnd = intel_allocator_open(xe, ctx->vm, INTEL_ALLOCATOR_RELOC);
+	uint32_t run_id = mid_tiling;
+	uint32_t mid_region = region2, bb;
+	uint32_t width = param.width, height = param.height;
+	enum blt_compression mid_compression = config->compression;
+	int mid_compression_format = param.compression_format;
+	enum blt_compression_type comp_type = COMPRESSION_TYPE_3D;
+	uint8_t uc_mocs = intel_get_uc_mocs(xe);
+	int result;
+
+	bb = xe_bo_create_flags(xe, 0, bb_size, region1);
+
+	if (!blt_uses_extended_block_copy(xe))
+		pext = NULL;
+
+	blt_copy_init(xe, &blt);
+
+	src = blt_create_object(&blt, region1, width, height, bpp, uc_mocs,
+				T_LINEAR, COMPRESSION_DISABLED, comp_type, true);
+	mid = blt_create_object(&blt, mid_region, width, height, bpp, uc_mocs,
+				mid_tiling, mid_compression, comp_type, true);
+	dst = blt_create_object(&blt, region1, width, height, bpp, uc_mocs,
+				T_LINEAR, COMPRESSION_DISABLED, comp_type, true);
+	igt_assert(src->size == dst->size);
+	PRINT_SURFACE_INFO("src", src);
+	PRINT_SURFACE_INFO("mid", mid);
+	PRINT_SURFACE_INFO("dst", dst);
+
+	blt_surface_fill_rect(xe, src, width, height);
+	WRITE_PNG(xe, run_id, "src", src, width, height);
+
+	blt.color_depth = CD_32bit;
+	blt.print_bb = param.print_bb;
+	blt_set_copy_object(&blt.src, src);
+	blt_set_copy_object(&blt.dst, mid);
+	blt_set_object_ext(&ext.src, 0, width, height, SURFACE_TYPE_2D);
+	blt_set_object_ext(&ext.dst, mid_compression_format, width, height, SURFACE_TYPE_2D);
+	blt_set_batch(&blt.bb, bb, bb_size, region1);
+	blt_block_copy(xe, ctx, NULL, ahnd, &blt, pext);
+	intel_ctx_xe_sync(ctx, true);
+
+	/* We expect mid != src if there's compression */
+	if (mid->compression)
+		igt_assert(memcmp(src->ptr, mid->ptr, src->size) != 0);
+
+	WRITE_PNG(xe, run_id, "mid", &blt.dst, width, height);
+
+	if (config->surfcopy && pext) {
+		struct drm_xe_engine_class_instance inst = {
+			.engine_class = DRM_XE_ENGINE_CLASS_COPY,
+		};
+		intel_ctx_t *surf_ctx = ctx;
+		uint64_t surf_ahnd = ahnd;
+		uint32_t vm, engine;
+
+		if (config->new_ctx) {
+			vm = xe_vm_create(xe, DRM_XE_VM_CREATE_ASYNC_BIND_OPS, 0);
+			engine = xe_engine_create(xe, vm, &inst, 0);
+			surf_ctx = intel_ctx_xe(xe, vm, engine, 0, 0, 0);
+			surf_ahnd = intel_allocator_open(xe, surf_ctx->vm,
+							 INTEL_ALLOCATOR_RELOC);
+		}
+		surf_copy(xe, surf_ctx, surf_ahnd, src, mid, dst, run_id,
+			  config->suspend_resume);
+
+		if (surf_ctx != ctx) {
+			xe_engine_destroy(xe, engine);
+			xe_vm_destroy(xe, vm);
+			free(surf_ctx);
+			put_ahnd(surf_ahnd);
+		}
+	}
+
+	blt_copy_init(xe, &blt);
+	blt.color_depth = CD_32bit;
+	blt.print_bb = param.print_bb;
+	blt_set_copy_object(&blt.src, mid);
+	blt_set_copy_object(&blt.dst, dst);
+	blt_set_object_ext(&ext.src, mid_compression_format, width, height, SURFACE_TYPE_2D);
+	blt_set_object_ext(&ext.dst, 0, width, height, SURFACE_TYPE_2D);
+	if (config->inplace) {
+		blt_set_object(&blt.dst, mid->handle, dst->size, mid->region, 0,
+			       T_LINEAR, COMPRESSION_DISABLED, comp_type);
+		blt.dst.ptr = mid->ptr;
+	}
+
+	blt_set_batch(&blt.bb, bb, bb_size, region1);
+	blt_block_copy(xe, ctx, NULL, ahnd, &blt, pext);
+	intel_ctx_xe_sync(ctx, true);
+
+	WRITE_PNG(xe, run_id, "dst", &blt.dst, width, height);
+
+	result = memcmp(src->ptr, blt.dst.ptr, src->size);
+
+	/* Politely clean vm */
+	put_offset(ahnd, src->handle);
+	put_offset(ahnd, mid->handle);
+	put_offset(ahnd, dst->handle);
+	put_offset(ahnd, bb);
+	intel_allocator_bind(ahnd, 0, 0);
+	blt_destroy_object(xe, src);
+	blt_destroy_object(xe, mid);
+	blt_destroy_object(xe, dst);
+	gem_close(xe, bb);
+	put_ahnd(ahnd);
+
+	igt_assert_f(!result, "source and destination surfaces differs!\n");
+}
+
+static void block_multicopy(int xe,
+			    intel_ctx_t *ctx,
+			    uint32_t region1, uint32_t region2,
+			    enum blt_tiling_type mid_tiling,
+			    const struct test_config *config)
+{
+	struct blt_copy3_data blt3 = {};
+	struct blt_copy_data blt = {};
+	struct blt_block_copy3_data_ext ext3 = {}, *pext3 = &ext3;
+	struct blt_copy_object *src, *mid, *dst, *final;
+	const uint32_t bpp = 32;
+	uint64_t bb_size = xe_get_default_alignment(xe);
+	uint64_t ahnd = intel_allocator_open(xe, ctx->vm, INTEL_ALLOCATOR_RELOC);
+	uint32_t run_id = mid_tiling;
+	uint32_t mid_region = region2, bb;
+	uint32_t width = param.width, height = param.height;
+	enum blt_compression mid_compression = config->compression;
+	int mid_compression_format = param.compression_format;
+	enum blt_compression_type comp_type = COMPRESSION_TYPE_3D;
+	uint8_t uc_mocs = intel_get_uc_mocs(xe);
+	int result;
+
+	bb = xe_bo_create_flags(xe, 0, bb_size, region1);
+
+	if (!blt_uses_extended_block_copy(xe))
+		pext3 = NULL;
+
+	blt_copy_init(xe, &blt);
+
+	src = blt_create_object(&blt, region1, width, height, bpp, uc_mocs,
+				T_LINEAR, COMPRESSION_DISABLED, comp_type, true);
+	mid = blt_create_object(&blt, mid_region, width, height, bpp, uc_mocs,
+				mid_tiling, mid_compression, comp_type, true);
+	dst = blt_create_object(&blt, region1, width, height, bpp, uc_mocs,
+				mid_tiling, COMPRESSION_DISABLED, comp_type, true);
+	final = blt_create_object(&blt, region1, width, height, bpp, uc_mocs,
+				  T_LINEAR, COMPRESSION_DISABLED, comp_type, true);
+	igt_assert(src->size == dst->size);
+	PRINT_SURFACE_INFO("src", src);
+	PRINT_SURFACE_INFO("mid", mid);
+	PRINT_SURFACE_INFO("dst", dst);
+	PRINT_SURFACE_INFO("final", final);
+
+	blt_surface_fill_rect(xe, src, width, height);
+
+	blt3.color_depth = CD_32bit;
+	blt3.print_bb = param.print_bb;
+	blt_set_copy_object(&blt3.src, src);
+	blt_set_copy_object(&blt3.mid, mid);
+	blt_set_copy_object(&blt3.dst, dst);
+	blt_set_copy_object(&blt3.final, final);
+
+	if (config->inplace) {
+		blt_set_object(&blt3.dst, mid->handle, dst->size, mid->region,
+			       mid->mocs, mid_tiling, COMPRESSION_DISABLED,
+			       comp_type);
+		blt3.dst.ptr = mid->ptr;
+	}
+
+	blt_set_object_ext(&ext3.src, 0, width, height, SURFACE_TYPE_2D);
+	blt_set_object_ext(&ext3.mid, mid_compression_format, width, height, SURFACE_TYPE_2D);
+	blt_set_object_ext(&ext3.dst, 0, width, height, SURFACE_TYPE_2D);
+	blt_set_object_ext(&ext3.final, 0, width, height, SURFACE_TYPE_2D);
+	blt_set_batch(&blt3.bb, bb, bb_size, region1);
+
+	blt_block_copy3(xe, ctx, ahnd, &blt3, pext3);
+	intel_ctx_xe_sync(ctx, true);
+
+	WRITE_PNG(xe, run_id, "src", &blt3.src, width, height);
+	if (!config->inplace)
+		WRITE_PNG(xe, run_id, "mid", &blt3.mid, width, height);
+	WRITE_PNG(xe, run_id, "dst", &blt3.dst, width, height);
+	WRITE_PNG(xe, run_id, "final", &blt3.final, width, height);
+
+	result = memcmp(src->ptr, blt3.final.ptr, src->size);
+
+	put_offset(ahnd, src->handle);
+	put_offset(ahnd, mid->handle);
+	put_offset(ahnd, dst->handle);
+	put_offset(ahnd, final->handle);
+	put_offset(ahnd, bb);
+	intel_allocator_bind(ahnd, 0, 0);
+	blt_destroy_object(xe, src);
+	blt_destroy_object(xe, mid);
+	blt_destroy_object(xe, dst);
+	blt_destroy_object(xe, final);
+	gem_close(xe, bb);
+	put_ahnd(ahnd);
+
+	igt_assert_f(!result, "source and destination surfaces differs!\n");
+}
+
+enum copy_func {
+	BLOCK_COPY,
+	BLOCK_MULTICOPY,
+};
+
+static const struct {
+	const char *suffix;
+	void (*copyfn)(int fd,
+		       intel_ctx_t *ctx,
+		       uint32_t region1, uint32_t region2,
+		       enum blt_tiling_type btype,
+		       const struct test_config *config);
+} copyfns[] = {
+	[BLOCK_COPY] = { "", block_copy },
+	[BLOCK_MULTICOPY] = { "-multicopy", block_multicopy },
+};
+
+static void block_copy_test(int xe,
+			    const struct test_config *config,
+			    struct igt_collection *set,
+			    enum copy_func copy_function)
+{
+	struct drm_xe_engine_class_instance inst = {
+		.engine_class = DRM_XE_ENGINE_CLASS_COPY,
+	};
+	intel_ctx_t *ctx;
+	struct igt_collection *regions;
+	uint32_t vm, engine;
+	int tiling;
+
+	if (config->compression && !blt_block_copy_supports_compression(xe))
+		return;
+
+	if (config->inplace && !config->compression)
+		return;
+
+	for_each_tiling(tiling) {
+		if (!blt_block_copy_supports_tiling(xe, tiling) ||
+		    (param.tiling >= 0 && param.tiling != tiling))
+			continue;
+
+		for_each_variation_r(regions, 2, set) {
+			uint32_t region1, region2;
+			char *regtxt;
+
+			region1 = igt_collection_get_value(regions, 0);
+			region2 = igt_collection_get_value(regions, 1);
+
+			/* Compressed surface must be in device memory */
+			if (config->compression && !XE_IS_VRAM_MEMORY_REGION(xe, region2))
+				continue;
+
+			regtxt = xe_memregion_dynamic_subtest_name(xe, regions);
+
+			igt_dynamic_f("%s-%s-compfmt%d-%s%s",
+				      blt_tiling_name(tiling),
+				      config->compression ?
+					      "compressed" : "uncompressed",
+				      param.compression_format, regtxt,
+				      copyfns[copy_function].suffix) {
+				uint32_t sync_bind, sync_out;
+
+				vm = xe_vm_create(xe, DRM_XE_VM_CREATE_ASYNC_BIND_OPS, 0);
+				engine = xe_engine_create(xe, vm, &inst, 0);
+				sync_bind = syncobj_create(xe, 0);
+				sync_out = syncobj_create(xe, 0);
+				ctx = intel_ctx_xe(xe, vm, engine,
+						   0, sync_bind, sync_out);
+
+				copyfns[copy_function].copyfn(xe, ctx,
+							      region1, region2,
+							      tiling, config);
+
+				xe_engine_destroy(xe, engine);
+				xe_vm_destroy(xe, vm);
+				syncobj_destroy(xe, sync_bind);
+				syncobj_destroy(xe, sync_out);
+				free(ctx);
+			}
+
+			free(regtxt);
+		}
+	}
+}
+
+static int opt_handler(int opt, int opt_index, void *data)
+{
+	switch (opt) {
+	case 'b':
+		param.print_bb = true;
+		igt_debug("Print bb: %d\n", param.print_bb);
+		break;
+	case 'f':
+		param.compression_format = atoi(optarg);
+		igt_debug("Compression format: %d\n", param.compression_format);
+		igt_assert((param.compression_format & ~0x1f) == 0);
+		break;
+	case 'p':
+		param.write_png = true;
+		igt_debug("Write png: %d\n", param.write_png);
+		break;
+	case 's':
+		param.print_surface_info = true;
+		igt_debug("Print surface info: %d\n", param.print_surface_info);
+		break;
+	case 't':
+		param.tiling = atoi(optarg);
+		igt_debug("Tiling: %d\n", param.tiling);
+		break;
+	case 'W':
+		param.width = atoi(optarg);
+		igt_debug("Width: %d\n", param.width);
+		break;
+	case 'H':
+		param.height = atoi(optarg);
+		igt_debug("Height: %d\n", param.height);
+		break;
+	default:
+		return IGT_OPT_HANDLER_ERROR;
+	}
+
+	return IGT_OPT_HANDLER_SUCCESS;
+}
+
+const char *help_str =
+	"  -b\tPrint bb\n"
+	"  -f\tCompression format (0-31)\n"
+	"  -p\tWrite PNG\n"
+	"  -s\tPrint surface info\n"
+	"  -t\tTiling format (0 - linear, 1 - XMAJOR, 2 - YMAJOR, 3 - TILE4, 4 - TILE64)\n"
+	"  -W\tWidth (default 512)\n"
+	"  -H\tHeight (default 512)"
+	;
+
+igt_main_args("bf:pst:W:H:", NULL, help_str, opt_handler, NULL)
+{
+	struct igt_collection *set;
+	int xe;
+
+	igt_fixture {
+		xe = drm_open_driver(DRIVER_XE);
+		igt_require(blt_has_block_copy(xe));
+
+		xe_device_get(xe);
+
+		set = xe_get_memory_region_set(xe,
+					       XE_MEM_REGION_CLASS_SYSMEM,
+					       XE_MEM_REGION_CLASS_VRAM);
+	}
+
+	igt_describe("Check block-copy uncompressed blit");
+	igt_subtest_with_dynamic("block-copy-uncompressed") {
+		struct test_config config = {};
+
+		block_copy_test(xe, &config, set, BLOCK_COPY);
+	}
+
+	igt_describe("Check block-copy flatccs compressed blit");
+	igt_subtest_with_dynamic("block-copy-compressed") {
+		struct test_config config = { .compression = true };
+
+		block_copy_test(xe, &config, set, BLOCK_COPY);
+	}
+
+	igt_describe("Check block-multicopy flatccs compressed blit");
+	igt_subtest_with_dynamic("block-multicopy-compressed") {
+		struct test_config config = { .compression = true };
+
+		block_copy_test(xe, &config, set, BLOCK_MULTICOPY);
+	}
+
+	igt_describe("Check block-multicopy flatccs inplace decompression blit");
+	igt_subtest_with_dynamic("block-multicopy-inplace") {
+		struct test_config config = { .compression = true,
+					      .inplace = true };
+
+		block_copy_test(xe, &config, set, BLOCK_MULTICOPY);
+	}
+
+	igt_describe("Check flatccs data can be copied from/to surface");
+	igt_subtest_with_dynamic("ctrl-surf-copy") {
+		struct test_config config = { .compression = true,
+					      .surfcopy = true };
+
+		block_copy_test(xe, &config, set, BLOCK_COPY);
+	}
+
+	igt_describe("Check flatccs data are physically tagged and visible"
+		     " in different contexts");
+	igt_subtest_with_dynamic("ctrl-surf-copy-new-ctx") {
+		struct test_config config = { .compression = true,
+					      .surfcopy = true,
+					      .new_ctx = true };
+
+		block_copy_test(xe, &config, set, BLOCK_COPY);
+	}
+
+	igt_describe("Check flatccs data persists after suspend / resume (S0)");
+	igt_subtest_with_dynamic("suspend-resume") {
+		struct test_config config = { .compression = true,
+					      .surfcopy = true,
+					      .suspend_resume = true };
+
+		block_copy_test(xe, &config, set, BLOCK_COPY);
+	}
+
+	igt_fixture {
+		xe_device_put(xe);
+		close(xe);
+	}
+}
-- 
2.34.1

[igt-dev] [PATCH i-g-t v3 15/17] tests/xe_exercise_blt: Check blitter library fast-copy for Xe

[Zbigniew Kempczyński]

Port this test to work on xe. Instead of adding conditional code for
xe code which would decrease readability this is new test for xe.

Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
---
v3: - Explicitely release offsets and unbind (Karolina)
    - Use has-fast-copy instead of copied has-block-copy check (Karolina)
---
 tests/meson.build          |   1 +
 tests/xe/xe_exercise_blt.c | 377 +++++++++++++++++++++++++++++++++++++
 2 files changed, 378 insertions(+)
 create mode 100644 tests/xe/xe_exercise_blt.c

diff --git a/tests/meson.build b/tests/meson.build
index da8653fabc..3eddb2fb4e 100644
--- a/tests/meson.build
+++ b/tests/meson.build
@@ -276,6 +276,7 @@ xe_progs = [
 	'xe_exec_reset',
 	'xe_exec_store',
 	'xe_exec_threads',
+	'xe_exercise_blt',
 	'xe_gpgpu_fill',
 	'xe_guc_pc',
 	'xe_huc_copy',
diff --git a/tests/xe/xe_exercise_blt.c b/tests/xe/xe_exercise_blt.c
new file mode 100644
index 0000000000..693e80d0b7
--- /dev/null
+++ b/tests/xe/xe_exercise_blt.c
@@ -0,0 +1,377 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+
+#include "igt.h"
+#include "drm.h"
+#include "lib/intel_chipset.h"
+#include "intel_blt.h"
+#include "intel_mocs.h"
+#include "xe/xe_ioctl.h"
+#include "xe/xe_query.h"
+#include "xe/xe_util.h"
+
+/**
+ * TEST: xe exercise blt
+ * Description: Exercise blitter commands on Xe
+ * Feature: blitter
+ * Run type: FULL
+ * Test category: GEM_Legacy
+ *
+ * SUBTEST: fast-copy
+ * Description:
+ *   Check fast-copy blit
+ *   blitter
+ *
+ * SUBTEST: fast-copy-emit
+ * Description:
+ *   Check multiple fast-copy in one batch
+ *   blitter
+ */
+
+IGT_TEST_DESCRIPTION("Exercise blitter commands on Xe");
+
+static struct param {
+	int tiling;
+	bool write_png;
+	bool print_bb;
+	bool print_surface_info;
+	int width;
+	int height;
+} param = {
+	.tiling = -1,
+	.write_png = false,
+	.print_bb = false,
+	.print_surface_info = false,
+	.width = 512,
+	.height = 512,
+};
+
+#define PRINT_SURFACE_INFO(name, obj) do { \
+	if (param.print_surface_info) \
+		blt_surface_info((name), (obj)); } while (0)
+
+#define WRITE_PNG(fd, id, name, obj, w, h) do { \
+	if (param.write_png) \
+		blt_surface_to_png((fd), (id), (name), (obj), (w), (h)); } while (0)
+
+struct blt_fast_copy_data {
+	int xe;
+	struct blt_copy_object src;
+	struct blt_copy_object mid;
+	struct blt_copy_object dst;
+
+	struct blt_copy_batch bb;
+	enum blt_color_depth color_depth;
+
+	/* debug stuff */
+	bool print_bb;
+};
+
+static int fast_copy_one_bb(int xe,
+			    const intel_ctx_t *ctx,
+			    uint64_t ahnd,
+			    const struct blt_fast_copy_data *blt)
+{
+	struct blt_copy_data blt_tmp;
+	uint64_t bb_offset, alignment;
+	uint64_t bb_pos = 0;
+	int ret = 0;
+
+	alignment = xe_get_default_alignment(xe);
+
+	get_offset(ahnd, blt->src.handle, blt->src.size, alignment);
+	get_offset(ahnd, blt->mid.handle, blt->mid.size, alignment);
+	get_offset(ahnd, blt->dst.handle, blt->dst.size, alignment);
+	bb_offset = get_offset(ahnd, blt->bb.handle, blt->bb.size, alignment);
+
+	/* First blit */
+	blt_copy_init(xe, &blt_tmp);
+	blt_tmp.src = blt->src;
+	blt_tmp.dst = blt->mid;
+	blt_tmp.bb = blt->bb;
+	blt_tmp.color_depth = blt->color_depth;
+	blt_tmp.print_bb = blt->print_bb;
+	bb_pos = emit_blt_fast_copy(xe, ahnd, &blt_tmp, bb_pos, false);
+
+	/* Second blit */
+	blt_copy_init(xe, &blt_tmp);
+	blt_tmp.src = blt->mid;
+	blt_tmp.dst = blt->dst;
+	blt_tmp.bb = blt->bb;
+	blt_tmp.color_depth = blt->color_depth;
+	blt_tmp.print_bb = blt->print_bb;
+	bb_pos = emit_blt_fast_copy(xe, ahnd, &blt_tmp, bb_pos, true);
+
+	intel_ctx_xe_exec(ctx, ahnd, bb_offset);
+
+	return ret;
+}
+
+static void fast_copy_emit(int xe, const intel_ctx_t *ctx,
+			   uint32_t region1, uint32_t region2,
+			   enum blt_tiling_type mid_tiling)
+{
+	struct blt_copy_data bltinit = {};
+	struct blt_fast_copy_data blt = {};
+	struct blt_copy_object *src, *mid, *dst;
+	const uint32_t bpp = 32;
+	uint64_t bb_size = xe_get_default_alignment(xe);
+	uint64_t ahnd = intel_allocator_open_full(xe, ctx->vm, 0, 0,
+						  INTEL_ALLOCATOR_SIMPLE,
+						  ALLOC_STRATEGY_LOW_TO_HIGH, 0);
+	uint32_t bb, width = param.width, height = param.height;
+	int result;
+
+	bb = xe_bo_create_flags(xe, 0, bb_size, region1);
+
+	blt_copy_init(xe, &bltinit);
+	src = blt_create_object(&bltinit, region1, width, height, bpp, 0,
+				T_LINEAR, COMPRESSION_DISABLED, 0, true);
+	mid = blt_create_object(&bltinit, region2, width, height, bpp, 0,
+				mid_tiling, COMPRESSION_DISABLED, 0, true);
+	dst = blt_create_object(&bltinit, region1, width, height, bpp, 0,
+				T_LINEAR, COMPRESSION_DISABLED, 0, true);
+	igt_assert(src->size == dst->size);
+
+	PRINT_SURFACE_INFO("src", src);
+	PRINT_SURFACE_INFO("mid", mid);
+	PRINT_SURFACE_INFO("dst", dst);
+
+	blt_surface_fill_rect(xe, src, width, height);
+	WRITE_PNG(xe, mid_tiling, "src", src, width, height);
+
+	memset(&blt, 0, sizeof(blt));
+	blt.color_depth = CD_32bit;
+	blt.print_bb = param.print_bb;
+	blt_set_copy_object(&blt.src, src);
+	blt_set_copy_object(&blt.mid, mid);
+	blt_set_copy_object(&blt.dst, dst);
+	blt_set_batch(&blt.bb, bb, bb_size, region1);
+
+	fast_copy_one_bb(xe, ctx, ahnd, &blt);
+
+	WRITE_PNG(xe, mid_tiling, "mid", &blt.mid, width, height);
+	WRITE_PNG(xe, mid_tiling, "dst", &blt.dst, width, height);
+
+	result = memcmp(src->ptr, blt.dst.ptr, src->size);
+
+	blt_destroy_object(xe, src);
+	blt_destroy_object(xe, mid);
+	blt_destroy_object(xe, dst);
+	gem_close(xe, bb);
+	put_ahnd(ahnd);
+
+	munmap(&bb, bb_size);
+
+	igt_assert_f(!result, "source and destination surfaces differs!\n");
+}
+
+static void fast_copy(int xe, const intel_ctx_t *ctx,
+		      uint32_t region1, uint32_t region2,
+		      enum blt_tiling_type mid_tiling)
+{
+	struct blt_copy_data blt = {};
+	struct blt_copy_object *src, *mid, *dst;
+	const uint32_t bpp = 32;
+	uint64_t bb_size = xe_get_default_alignment(xe);
+	uint64_t ahnd = intel_allocator_open_full(xe, ctx->vm, 0, 0,
+						  INTEL_ALLOCATOR_SIMPLE,
+						  ALLOC_STRATEGY_LOW_TO_HIGH, 0);
+	uint32_t bb;
+	uint32_t width = param.width, height = param.height;
+	int result;
+
+	bb = xe_bo_create_flags(xe, 0, bb_size, region1);
+
+	blt_copy_init(xe, &blt);
+	src = blt_create_object(&blt, region1, width, height, bpp, 0,
+				T_LINEAR, COMPRESSION_DISABLED, 0, true);
+	mid = blt_create_object(&blt, region2, width, height, bpp, 0,
+				mid_tiling, COMPRESSION_DISABLED, 0, true);
+	dst = blt_create_object(&blt, region1, width, height, bpp, 0,
+				T_LINEAR, COMPRESSION_DISABLED, 0, true);
+	igt_assert(src->size == dst->size);
+
+	blt_surface_fill_rect(xe, src, width, height);
+
+	blt.color_depth = CD_32bit;
+	blt.print_bb = param.print_bb;
+	blt_set_copy_object(&blt.src, src);
+	blt_set_copy_object(&blt.dst, mid);
+	blt_set_batch(&blt.bb, bb, bb_size, region1);
+
+	blt_fast_copy(xe, ctx, NULL, ahnd, &blt);
+
+	WRITE_PNG(xe, mid_tiling, "src", &blt.src, width, height);
+	WRITE_PNG(xe, mid_tiling, "mid", &blt.dst, width, height);
+
+	blt_copy_init(xe, &blt);
+	blt.color_depth = CD_32bit;
+	blt.print_bb = param.print_bb;
+	blt_set_copy_object(&blt.src, mid);
+	blt_set_copy_object(&blt.dst, dst);
+	blt_set_batch(&blt.bb, bb, bb_size, region1);
+
+	blt_fast_copy(xe, ctx, NULL, ahnd, &blt);
+
+	WRITE_PNG(xe, mid_tiling, "dst", &blt.dst, width, height);
+
+	result = memcmp(src->ptr, blt.dst.ptr, src->size);
+
+	put_offset(ahnd, src->handle);
+	put_offset(ahnd, mid->handle);
+	put_offset(ahnd, dst->handle);
+	put_offset(ahnd, bb);
+	intel_allocator_bind(ahnd, 0, 0);
+	blt_destroy_object(xe, src);
+	blt_destroy_object(xe, mid);
+	blt_destroy_object(xe, dst);
+	gem_close(xe, bb);
+	put_ahnd(ahnd);
+
+	igt_assert_f(!result, "source and destination surfaces differs!\n");
+}
+
+enum fast_copy_func {
+	FAST_COPY,
+	FAST_COPY_EMIT
+};
+
+static char
+	*full_subtest_str(char *regtxt, enum blt_tiling_type tiling,
+			  enum fast_copy_func func)
+{
+	char *name;
+	uint32_t len;
+
+	len = asprintf(&name, "%s-%s%s", blt_tiling_name(tiling), regtxt,
+		       func == FAST_COPY_EMIT ? "-emit" : "");
+
+	igt_assert_f(len >= 0, "asprintf failed!\n");
+
+	return name;
+}
+
+static void fast_copy_test(int xe,
+			   struct igt_collection *set,
+			   enum fast_copy_func func)
+{
+	struct drm_xe_engine_class_instance inst = {
+		.engine_class = DRM_XE_ENGINE_CLASS_COPY,
+	};
+	struct igt_collection *regions;
+	void (*copy_func)(int xe, const intel_ctx_t *ctx,
+			  uint32_t r1, uint32_t r2, enum blt_tiling_type tiling);
+	intel_ctx_t *ctx;
+	int tiling;
+
+	for_each_tiling(tiling) {
+		if (!blt_fast_copy_supports_tiling(xe, tiling))
+			continue;
+
+		for_each_variation_r(regions, 2, set) {
+			uint32_t region1, region2;
+			uint32_t vm, engine;
+			char *regtxt, *test_name;
+
+			region1 = igt_collection_get_value(regions, 0);
+			region2 = igt_collection_get_value(regions, 1);
+
+			vm = xe_vm_create(xe, DRM_XE_VM_CREATE_ASYNC_BIND_OPS, 0);
+			engine = xe_engine_create(xe, vm, &inst, 0);
+			ctx = intel_ctx_xe(xe, vm, engine, 0, 0, 0);
+
+			copy_func = (func == FAST_COPY) ? fast_copy : fast_copy_emit;
+			regtxt = xe_memregion_dynamic_subtest_name(xe, regions);
+			test_name = full_subtest_str(regtxt, tiling, func);
+
+			igt_dynamic_f("%s", test_name) {
+				copy_func(xe, ctx,
+					  region1, region2,
+					  tiling);
+			}
+
+			free(regtxt);
+			free(test_name);
+			xe_engine_destroy(xe, engine);
+			xe_vm_destroy(xe, vm);
+			free(ctx);
+		}
+	}
+}
+
+static int opt_handler(int opt, int opt_index, void *data)
+{
+	switch (opt) {
+	case 'b':
+		param.print_bb = true;
+		igt_debug("Print bb: %d\n", param.print_bb);
+		break;
+	case 'p':
+		param.write_png = true;
+		igt_debug("Write png: %d\n", param.write_png);
+		break;
+	case 's':
+		param.print_surface_info = true;
+		igt_debug("Print surface info: %d\n", param.print_surface_info);
+		break;
+	case 't':
+		param.tiling = atoi(optarg);
+		igt_debug("Tiling: %d\n", param.tiling);
+		break;
+	case 'W':
+		param.width = atoi(optarg);
+		igt_debug("Width: %d\n", param.width);
+		break;
+	case 'H':
+		param.height = atoi(optarg);
+		igt_debug("Height: %d\n", param.height);
+		break;
+	default:
+		return IGT_OPT_HANDLER_ERROR;
+	}
+
+	return IGT_OPT_HANDLER_SUCCESS;
+}
+
+const char *help_str =
+	"  -b\tPrint bb\n"
+	"  -p\tWrite PNG\n"
+	"  -s\tPrint surface info\n"
+	"  -t\tTiling format (0 - linear, 1 - XMAJOR, 2 - YMAJOR, 3 - TILE4, 4 - TILE64, 5 - YFMAJOR)\n"
+	"  -W\tWidth (default 512)\n"
+	"  -H\tHeight (default 512)"
+	;
+
+igt_main_args("b:pst:W:H:", NULL, help_str, opt_handler, NULL)
+{
+	struct igt_collection *set;
+	int xe;
+
+	igt_fixture {
+		xe = drm_open_driver(DRIVER_XE);
+		igt_require(blt_has_fast_copy(xe));
+
+		xe_device_get(xe);
+
+		set = xe_get_memory_region_set(xe,
+					       XE_MEM_REGION_CLASS_SYSMEM,
+					       XE_MEM_REGION_CLASS_VRAM);
+	}
+
+	igt_describe("Check fast-copy blit");
+	igt_subtest_with_dynamic("fast-copy") {
+		fast_copy_test(xe, set, FAST_COPY);
+	}
+
+	igt_describe("Check multiple fast-copy in one batch");
+	igt_subtest_with_dynamic("fast-copy-emit") {
+		fast_copy_test(xe, set, FAST_COPY_EMIT);
+	}
+
+	igt_fixture {
+		drm_close_driver(xe);
+	}
+}
-- 
2.34.1

[igt-dev] [PATCH i-g-t v3 16/17] tests/api-intel-allocator: Adopt to exercise allocator to Xe

[Zbigniew Kempczyński]

Xe vm binding requires some cooperation from allocator side (tracking
alloc()/free()) operations. This diverges the path internally inside
the allocator so it is necessary to check if allocator supports
properly both drivers.

Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
Reviewed-by: Karolina Stolarek <karolina.stolarek@intel.com>
---
 tests/i915/api_intel_allocator.c | 25 +++++++++++++++++--------
 1 file changed, 17 insertions(+), 8 deletions(-)

diff --git a/tests/i915/api_intel_allocator.c b/tests/i915/api_intel_allocator.c
index 77c2d25e83..f3fcf8a341 100644
--- a/tests/i915/api_intel_allocator.c
+++ b/tests/i915/api_intel_allocator.c
@@ -9,6 +9,9 @@
 #include "igt.h"
 #include "igt_aux.h"
 #include "intel_allocator.h"
+#include "xe/xe_ioctl.h"
+#include "xe/xe_query.h"
+
 /**
  * TEST: api intel allocator
  * Category: Infrastructure
@@ -454,6 +457,7 @@ static void __simple_allocs(int fd)
 	uint32_t handles[SIMPLE_GROUP_ALLOCS];
 	uint64_t ahnd;
 	uint32_t ctx;
+	bool is_xe = is_xe_device(fd);
 	int i;
 
 	ctx = rand() % 2;
@@ -463,7 +467,12 @@ static void __simple_allocs(int fd)
 		uint32_t size;
 
 		size = (rand() % 4 + 1) * 0x1000;
-		handles[i] = gem_create(fd, size);
+		if (is_xe)
+			handles[i] = xe_bo_create_flags(fd, 0, size,
+							system_memory(fd));
+		else
+			handles[i] = gem_create(fd, size);
+
 		intel_allocator_alloc(ahnd, handles[i], size, 0x1000);
 	}
 
@@ -573,8 +582,6 @@ static void reopen(int fd)
 {
 	int fd2;
 
-	igt_require_gem(fd);
-
 	fd2 = drm_reopen_driver(fd);
 
 	__reopen_allocs(fd, fd2, true);
@@ -587,8 +594,6 @@ static void reopen_fork(int fd)
 {
 	int fd2;
 
-	igt_require_gem(fd);
-
 	intel_allocator_multiprocess_start();
 
 	fd2 = drm_reopen_driver(fd);
@@ -838,7 +843,7 @@ igt_main
 	struct allocators *a;
 
 	igt_fixture {
-		fd = drm_open_driver(DRIVER_INTEL);
+		fd = drm_open_driver(DRIVER_INTEL | DRIVER_XE);
 		atomic_init(&next_handle, 1);
 		srandom(0xdeadbeef);
 	}
@@ -911,12 +916,16 @@ igt_main
 	igt_subtest_f("open-vm")
 		open_vm(fd);
 
-	igt_subtest_f("execbuf-with-allocator")
+	igt_subtest_f("execbuf-with-allocator") {
+		igt_require(is_i915_device(fd));
 		execbuf_with_allocator(fd);
+	}
 
 	igt_describe("Verifies creating and executing bb from gem pool");
-	igt_subtest_f("gem-pool")
+	igt_subtest_f("gem-pool") {
+		igt_require(is_i915_device(fd));
 		gem_pool(fd);
+	}
 
 	igt_fixture
 		drm_close_driver(fd);
-- 
2.34.1

[igt-dev] [PATCH i-g-t v3 17/17] tests/gem_ccs: Avoid writing png twice

[Zbigniew Kempczyński]

As Karolina noticed we write source png twice what's just takes
time as source surface is not touched. Drop one png write.

Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
Cc: Karolina Stolarek <karolina.stolarek@intel.com>
---
 tests/i915/gem_ccs.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/tests/i915/gem_ccs.c b/tests/i915/gem_ccs.c
index d9d785ed9b..a1302e0a4d 100644
--- a/tests/i915/gem_ccs.c
+++ b/tests/i915/gem_ccs.c
@@ -421,7 +421,6 @@ static void block_copy(int i915,
 	if (mid->compression)
 		igt_assert(memcmp(src->ptr, mid->ptr, src->size) != 0);
 
-	WRITE_PNG(i915, run_id, "src", &blt.src, width, height);
 	WRITE_PNG(i915, run_id, "mid", &blt.dst, width, height);
 
 	if (config->surfcopy && pext) {
-- 
2.34.1

[igt-dev] ✗ GitLab.Pipeline: warning for Extend intel_blt to work on Xe (rev3)

[Patchwork]

== Series Details ==

Series: Extend intel_blt to work on Xe (rev3)
URL   : https://patchwork.freedesktop.org/series/120162/
State : warning

== Summary ==

Pipeline status: FAILED.

see https://gitlab.freedesktop.org/gfx-ci/igt-ci-tags/-/pipelines/932186 for the overview.

build-containers:build-debian has failed (https://gitlab.freedesktop.org/gfx-ci/igt-ci-tags/-/jobs/45263081):
  time="2023-07-11T11:38:24Z" level=fatal msg="Invalid status code returned when fetching blob 500 (Internal Server Error)" 
  Building!
  STEP 1: FROM debian:buster
  Getting image source signatures
  Copying blob sha256:8eb6dba554cffc072c7a6c696c8a23fc311e543399d84ab3ebc55c07ab54414f
  Copying config sha256:1de12428f6f470169eba22c18e53addae00d60e92ebe96b2c3bdd99c17c34010
  Writing manifest to image destination
  Storing signatures
  STEP 2: RUN apt-get update
  error running container: error creating container for [/bin/sh -c apt-get update]: time="2023-07-11T11:38:32Z" level=warning msg="signal: killed"
  time="2023-07-11T11:38:32Z" level=error msg="container_linux.go:346: starting container process caused \"process_linux.go:297: applying cgroup configuration for process caused \\\"mountpoint for cgroup not found\\\"\"\n"
  container_linux.go:346: starting container process caused "process_linux.go:297: applying cgroup configuration for process caused \"mountpoint for cgroup not found\""
  : exit status 1
  Error: error building at STEP "RUN apt-get update": error while running runtime: exit status 1
  section_end:1689075513:step_script
  section_start:1689075513:cleanup_file_variables
  Cleaning up project directory and file based variables
  section_end:1689075516:cleanup_file_variables
  ERROR: Job failed: exit code 1
  

build-containers:build-debian-arm64 has failed (https://gitlab.freedesktop.org/gfx-ci/igt-ci-tags/-/jobs/45263083):
      "DockerVersion": "",
      "Labels": {},
      "Architecture": "amd64",
      "Os": "linux",
      "Layers": [
          "sha256:a1c4b97398ec8bc7b17c52a01ef6941dc58811cb43f945b093db7a131586297a"
      ]
  }
  Skipping, already built
  Getting image source signatures
  Copying blob sha256:a1c4b97398ec8bc7b17c52a01ef6941dc58811cb43f945b093db7a131586297a
  Copying config sha256:7360075a71dacfc66f0b49b3271b9a459904dbe51c5760efac48fe52da27946c
  Writing manifest to image destination
  time="2023-07-11T11:38:23Z" level=fatal msg="Error writing manifest: Error uploading manifest commit-72ee64f2d536b357e190a6a85a90cdcebc7ed1e8 to registry.freedesktop.org/gfx-ci/igt-ci-tags/build-debian-arm64: received unexpected HTTP status: 500 Internal Server Error" 
  section_end:1689075504:step_script
  section_start:1689075504:cleanup_file_variables
  Cleaning up project directory and file based variables
  section_end:1689075504:cleanup_file_variables
  ERROR: Job failed: exit code 1
  

build-containers:build-debian-armhf has failed (https://gitlab.freedesktop.org/gfx-ci/igt-ci-tags/-/jobs/45263082):
          "intel-IGT_5164"
      ],
      "Created": "2019-11-20T12:54:52.073706506Z",
      "DockerVersion": "",
      "Labels": {},
      "Architecture": "amd64",
      "Os": "linux",
      "Layers": [
          "sha256:45252bb38c9c74dfe76c4b3808269db61d3b2aebe5e26bf7fcd85e48b93e16f9"
      ]
  }
  Skipping, already built
  Getting image source signatures
  time="2023-07-11T11:38:22Z" level=fatal msg="Error trying to reuse blob sha256:45252bb38c9c74dfe76c4b3808269db61d3b2aebe5e26bf7fcd85e48b93e16f9 at destination: failed to read from destination repository gfx-ci/igt-ci-tags/build-debian-armhf: 500 (Internal Server Error)" 
  section_end:1689075503:step_script
  section_start:1689075503:cleanup_file_variables
  Cleaning up project directory and file based variables
  section_end:1689075504:cleanup_file_variables
  ERROR: Job failed: exit code 1
  

build-containers:build-debian-mips has failed (https://gitlab.freedesktop.org/gfx-ci/igt-ci-tags/-/jobs/45263084):
      "DockerVersion": "",
      "Labels": {},
      "Architecture": "amd64",
      "Os": "linux",
      "Layers": [
          "sha256:04854e5e3796b945522a085073b5a49cadad51bd1c5414c91654fa3d63d8db4d"
      ]
  }
  Skipping, already built
  Getting image source signatures
  Copying blob sha256:04854e5e3796b945522a085073b5a49cadad51bd1c5414c91654fa3d63d8db4d
  Copying config sha256:cc55efdc667be826910d414a562c76ce1130a9c15255a0dd115431bc42f83448
  Writing manifest to image destination
  time="2023-07-11T11:38:23Z" level=fatal msg="Error writing manifest: Error uploading manifest commit-72ee64f2d536b357e190a6a85a90cdcebc7ed1e8 to registry.freedesktop.org/gfx-ci/igt-ci-tags/build-debian-mips: received unexpected HTTP status: 500 Internal Server Error" 
  section_end:1689075504:step_script
  section_start:1689075504:cleanup_file_variables
  Cleaning up project directory and file based variables
  section_end:1689075504:cleanup_file_variables
  ERROR: Job failed: exit code 1
  

build-containers:build-fedora has failed (https://gitlab.freedesktop.org/gfx-ci/igt-ci-tags/-/jobs/45263085):
  time="2023-07-11T11:38:20Z" level=fatal msg="Error reading manifest dockerfile-d55a9096f48b128e63002d049c1e9ea16a6c5bee in registry.freedesktop.org/gfx-ci/igt-ci-tags/build-fedora: received unexpected HTTP status: 500 Internal Server Error" 
  Building!
  STEP 1: FROM fedora:31
  Getting image source signatures
  Copying blob sha256:854946d575a439a894349addd141568875d7c1e673d3286b08250f3dde002e6a
  Copying config sha256:7e94ed77b448a8d2ff08b92d3ca743e4e862c744892d6886c73487581eb5863a
  Writing manifest to image destination
  Storing signatures
  STEP 2: RUN dnf install -y 	gcc flex bison libatomic meson ninja-build xdotool 	'pkgconfig(libdrm)' 	'pkgconfig(pciaccess)' 	'pkgconfig(libkmod)' 	'pkgconfig(libprocps)' 	'pkgconfig(libunwind)' 	'pkgconfig(libdw)' 	'pkgconfig(pixman-1)' 	'pkgconfig(valgrind)' 	'pkgconfig(cairo)' 	'pkgconfig(libudev)' 	'pkgconfig(glib-2.0)' 	'pkgconfig(gsl)' 	'pkgconfig(alsa)' 	'pkgconfig(xmlrpc)' 	'pkgconfig(xmlrpc_util)' 	'pkgconfig(xmlrpc_client)' 	'pkgconfig(json-c)' 	'pkgconfig(gtk-doc)' 	'pkgconfig(xv)' 	'pkgconfig(xrandr)' 	python3-docutils
  error running container: error creating container for [/bin/sh -c dnf install -y 	gcc flex bison libatomic meson ninja-build xdotool 	'pkgconfig(libdrm)' 	'pkgconfig(pciaccess)' 	'pkgconfig(libkmod)' 	'pkgconfig(libprocps)' 	'pkgconfig(libunwind)' 	'pkgconfig(libdw)' 	'pkgconfig(pixman-1)' 	'pkgconfig(valgrind)' 	'pkgconfig(cairo)' 	'pkgconfig(libudev)' 	'pkgconfig(glib-2.0)' 	'pkgconfig(gsl)' 	'pkgconfig(alsa)' 	'pkgconfig(xmlrpc)' 	'pkgconfig(xmlrpc_util)' 	'pkgconfig(xmlrpc_client)' 	'pkgconfig(json-c)' 	'pkgconfig(gtk-doc)' 	'pkgconfig(xv)' 	'pkgconfig(xrandr)' 	python3-docutils]: time="2023-07-11T11:38:26Z" level=warning msg="signal: killed"
  time="2023-07-11T11:38:26Z" level=error msg="container_linux.go:346: starting container process caused \"process_linux.go:297: applying cgroup configuration for process caused \\\"mountpoint for cgroup not found\\\"\"\n"
  container_linux.go:346: starting container process caused "process_linux.go:297: applying cgroup configuration for process caused \"mountpoint for cgroup not found\""
  : exit status 1
  Error: error building at STEP "RUN dnf install -y 	gcc flex bison libatomic meson ninja-build xdotool 	'pkgconfig(libdrm)' 	'pkgconfig(pciaccess)' 	'pkgconfig(libkmod)' 	'pkgconfig(libprocps)' 	'pkgconfig(libunwind)' 	'pkgconfig(libdw)' 	'pkgconfig(pixman-1)' 	'pkgconfig(valgrind)' 	'pkgconfig(cairo)' 	'pkgconfig(libudev)' 	'pkgconfig(glib-2.0)' 	'pkgconfig(gsl)' 	'pkgconfig(alsa)' 	'pkgconfig(xmlrpc)' 	'pkgconfig(xmlrpc_util)' 	'pkgconfig(xmlrpc_client)' 	'pkgconfig(json-c)' 	'pkgconfig(gtk-doc)' 	'pkgconfig(xv)' 	'pkgconfig(xrandr)' 	python3-docutils": error while running runtime: exit status 1
  section_end:1689075506:step_script
  section_start:1689075506:cleanup_file_variables
  Cleaning up project directory and file based variables
  section_end:1689075507:cleanup_file_variables
  ERROR: Job failed: exit code 1

== Logs ==

For more details see: https://gitlab.freedesktop.org/gfx-ci/igt-ci-tags/-/pipelines/932186

[igt-dev] ○ CI.xeBAT: info for Extend intel_blt to work on Xe (rev3)

[Patchwork]

[-- Attachment #1: Type: text/plain, Size: 325 bytes --]

== Series Details ==

Series: Extend intel_blt to work on Xe (rev3)
URL   : https://patchwork.freedesktop.org/series/120162/
State : info

== Summary ==

Participating hosts:
bat-atsm-2
bat-dg2-oem2
bat-adlp-7
Missing hosts results[0]:
Results: [IGTPW_9383](https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_9383/index.html)



[-- Attachment #2: Type: text/html, Size: 835 bytes --]

[igt-dev] ✓ Fi.CI.BAT: success for Extend intel_blt to work on Xe (rev3)

[Patchwork]

[-- Attachment #1: Type: text/plain, Size: 9427 bytes --]

== Series Details ==

Series: Extend intel_blt to work on Xe (rev3)
URL   : https://patchwork.freedesktop.org/series/120162/
State : success

== Summary ==

CI Bug Log - changes from CI_DRM_13369 -> IGTPW_9383
====================================================

Summary
-------

  **SUCCESS**

  No regressions found.

  External URL: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/index.html

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

  Additional (1): fi-pnv-d510 
  Missing    (1): fi-snb-2520m 

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

  Here are the changes found in IGTPW_9383 that come from known issues:

### IGT changes ###

#### Issues hit ####

  * igt@i915_pm_rpm@module-reload:
    - fi-rkl-11600:       [PASS][1] -> [FAIL][2] ([i915#7940])
   [1]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/fi-rkl-11600/igt@i915_pm_rpm@module-reload.html
   [2]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/fi-rkl-11600/igt@i915_pm_rpm@module-reload.html
    - fi-tgl-1115g4:      [PASS][3] -> [FAIL][4] ([i915#7940])
   [3]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/fi-tgl-1115g4/igt@i915_pm_rpm@module-reload.html
   [4]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/fi-tgl-1115g4/igt@i915_pm_rpm@module-reload.html

  * igt@i915_selftest@live@gem_contexts:
    - bat-mtlp-8:         [PASS][5] -> [ABORT][6] ([i915#8671] / [i915#8704])
   [5]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/bat-mtlp-8/igt@i915_selftest@live@gem_contexts.html
   [6]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/bat-mtlp-8/igt@i915_selftest@live@gem_contexts.html

  * igt@i915_selftest@live@requests:
    - bat-mtlp-8:         [PASS][7] -> [DMESG-FAIL][8] ([i915#8497])
   [7]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/bat-mtlp-8/igt@i915_selftest@live@requests.html
   [8]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/bat-mtlp-8/igt@i915_selftest@live@requests.html
    - bat-mtlp-6:         [PASS][9] -> [DMESG-FAIL][10] ([i915#8497])
   [9]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/bat-mtlp-6/igt@i915_selftest@live@requests.html
   [10]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/bat-mtlp-6/igt@i915_selftest@live@requests.html

  * igt@i915_selftest@live@slpc:
    - bat-mtlp-6:         [PASS][11] -> [DMESG-WARN][12] ([i915#6367])
   [11]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/bat-mtlp-6/igt@i915_selftest@live@slpc.html
   [12]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/bat-mtlp-6/igt@i915_selftest@live@slpc.html
    - bat-rpls-1:         [PASS][13] -> [DMESG-WARN][14] ([i915#6367])
   [13]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/bat-rpls-1/igt@i915_selftest@live@slpc.html
   [14]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/bat-rpls-1/igt@i915_selftest@live@slpc.html

  * igt@i915_suspend@basic-s3-without-i915:
    - fi-pnv-d510:        NOTRUN -> [ABORT][15] ([i915#8844])
   [15]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/fi-pnv-d510/igt@i915_suspend@basic-s3-without-i915.html
    - bat-rpls-2:         [PASS][16] -> [ABORT][17] ([i915#6687] / [i915#7978] / [i915#8668])
   [16]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/bat-rpls-2/igt@i915_suspend@basic-s3-without-i915.html
   [17]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/bat-rpls-2/igt@i915_suspend@basic-s3-without-i915.html

  * igt@kms_pipe_crc_basic@read-crc-frame-sequence@pipe-d-edp-1:
    - bat-rplp-1:         [PASS][18] -> [ABORT][19] ([i915#8442] / [i915#8668])
   [18]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/bat-rplp-1/igt@kms_pipe_crc_basic@read-crc-frame-sequence@pipe-d-edp-1.html
   [19]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/bat-rplp-1/igt@kms_pipe_crc_basic@read-crc-frame-sequence@pipe-d-edp-1.html

  * igt@kms_psr@primary_page_flip:
    - fi-pnv-d510:        NOTRUN -> [SKIP][20] ([fdo#109271]) +37 similar issues
   [20]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/fi-pnv-d510/igt@kms_psr@primary_page_flip.html

  
#### Possible fixes ####

  * igt@gem_exec_suspend@basic-s0@smem:
    - bat-dg1-7:          [FAIL][21] ([fdo#103375] / [i915#6121] / [i915#8011]) -> [PASS][22] +1 similar issue
   [21]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/bat-dg1-7/igt@gem_exec_suspend@basic-s0@smem.html
   [22]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/bat-dg1-7/igt@gem_exec_suspend@basic-s0@smem.html

  * igt@gem_exec_suspend@basic-s3@smem:
    - bat-dg1-7:          [FAIL][23] ([fdo#103375] / [i915#6121]) -> [PASS][24] +1 similar issue
   [23]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/bat-dg1-7/igt@gem_exec_suspend@basic-s3@smem.html
   [24]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/bat-dg1-7/igt@gem_exec_suspend@basic-s3@smem.html

  * igt@i915_selftest@live@gt_heartbeat:
    - fi-apl-guc:         [DMESG-FAIL][25] ([i915#5334]) -> [PASS][26]
   [25]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/fi-apl-guc/igt@i915_selftest@live@gt_heartbeat.html
   [26]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/fi-apl-guc/igt@i915_selftest@live@gt_heartbeat.html

  * igt@i915_selftest@live@gt_mocs:
    - bat-mtlp-8:         [DMESG-FAIL][27] ([i915#7059]) -> [PASS][28]
   [27]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/bat-mtlp-8/igt@i915_selftest@live@gt_mocs.html
   [28]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/bat-mtlp-8/igt@i915_selftest@live@gt_mocs.html

  * igt@i915_selftest@live@hangcheck:
    - fi-skl-guc:         [DMESG-FAIL][29] ([i915#8723]) -> [PASS][30]
   [29]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/fi-skl-guc/igt@i915_selftest@live@hangcheck.html
   [30]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/fi-skl-guc/igt@i915_selftest@live@hangcheck.html

  * igt@i915_selftest@live@migrate:
    - bat-dg2-11:         [DMESG-WARN][31] ([i915#7699]) -> [PASS][32]
   [31]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/bat-dg2-11/igt@i915_selftest@live@migrate.html
   [32]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/bat-dg2-11/igt@i915_selftest@live@migrate.html

  * igt@i915_selftest@live@slpc:
    - bat-rpls-2:         [DMESG-WARN][33] ([i915#6367]) -> [PASS][34]
   [33]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/bat-rpls-2/igt@i915_selftest@live@slpc.html
   [34]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/bat-rpls-2/igt@i915_selftest@live@slpc.html

  
#### Warnings ####

  * igt@core_auth@basic-auth:
    - bat-adlp-11:        [ABORT][35] ([i915#4423] / [i915#8011]) -> [ABORT][36] ([i915#8011])
   [35]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/bat-adlp-11/igt@core_auth@basic-auth.html
   [36]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/bat-adlp-11/igt@core_auth@basic-auth.html

  * igt@i915_pm_rpm@basic-pci-d3-state:
    - fi-kbl-guc:         [SKIP][37] ([fdo#109271]) -> [FAIL][38] ([i915#7940])
   [37]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/fi-kbl-guc/igt@i915_pm_rpm@basic-pci-d3-state.html
   [38]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/fi-kbl-guc/igt@i915_pm_rpm@basic-pci-d3-state.html

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

  [fdo#103375]: https://bugs.freedesktop.org/show_bug.cgi?id=103375
  [fdo#109271]: https://bugs.freedesktop.org/show_bug.cgi?id=109271
  [i915#4423]: https://gitlab.freedesktop.org/drm/intel/issues/4423
  [i915#5334]: https://gitlab.freedesktop.org/drm/intel/issues/5334
  [i915#6121]: https://gitlab.freedesktop.org/drm/intel/issues/6121
  [i915#6367]: https://gitlab.freedesktop.org/drm/intel/issues/6367
  [i915#6687]: https://gitlab.freedesktop.org/drm/intel/issues/6687
  [i915#7059]: https://gitlab.freedesktop.org/drm/intel/issues/7059
  [i915#7699]: https://gitlab.freedesktop.org/drm/intel/issues/7699
  [i915#7940]: https://gitlab.freedesktop.org/drm/intel/issues/7940
  [i915#7978]: https://gitlab.freedesktop.org/drm/intel/issues/7978
  [i915#8011]: https://gitlab.freedesktop.org/drm/intel/issues/8011
  [i915#8442]: https://gitlab.freedesktop.org/drm/intel/issues/8442
  [i915#8497]: https://gitlab.freedesktop.org/drm/intel/issues/8497
  [i915#8668]: https://gitlab.freedesktop.org/drm/intel/issues/8668
  [i915#8671]: https://gitlab.freedesktop.org/drm/intel/issues/8671
  [i915#8704]: https://gitlab.freedesktop.org/drm/intel/issues/8704
  [i915#8723]: https://gitlab.freedesktop.org/drm/intel/issues/8723
  [i915#8844]: https://gitlab.freedesktop.org/drm/intel/issues/8844


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

  * CI: CI-20190529 -> None
  * IGT: IGT_7380 -> IGTPW_9383

  CI-20190529: 20190529
  CI_DRM_13369: 0595153b5c237d83f9990f3ce44d072a262dcb0e @ git://anongit.freedesktop.org/gfx-ci/linux
  IGTPW_9383: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/index.html
  IGT_7380: 8e65f12de2fd52c05dc48fdbcb8cfe86f6de1a75 @ https://gitlab.freedesktop.org/drm/igt-gpu-tools.git


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

+igt@xe_ccs@block-copy-compressed
+igt@xe_ccs@block-copy-uncompressed
+igt@xe_ccs@block-multicopy-compressed
+igt@xe_ccs@block-multicopy-inplace
+igt@xe_ccs@ctrl-surf-copy
+igt@xe_ccs@ctrl-surf-copy-new-ctx
+igt@xe_ccs@suspend-resume
+igt@xe_exercise_blt@fast-copy
+igt@xe_exercise_blt@fast-copy-emit

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/index.html

[-- Attachment #2: Type: text/html, Size: 11275 bytes --]

[igt-dev] ✗ Fi.CI.IGT: failure for Extend intel_blt to work on Xe (rev3)

[Patchwork]

[-- Attachment #1: Type: text/plain, Size: 78616 bytes --]

== Series Details ==

Series: Extend intel_blt to work on Xe (rev3)
URL   : https://patchwork.freedesktop.org/series/120162/
State : failure

== Summary ==

CI Bug Log - changes from CI_DRM_13369_full -> IGTPW_9383_full
====================================================

Summary
-------

  **FAILURE**

  Serious unknown changes coming with IGTPW_9383_full absolutely need to be
  verified manually.
  
  If you think the reported changes have nothing to do with the changes
  introduced in IGTPW_9383_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_9383/index.html

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

  No changes in participating hosts

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

  Here are the unknown changes that may have been introduced in IGTPW_9383_full:

### IGT changes ###

#### Possible regressions ####

  * igt@prime_vgem@fence-wait@ccs0:
    - shard-mtlp:         [PASS][1] -> [DMESG-WARN][2]
   [1]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-3/igt@prime_vgem@fence-wait@ccs0.html
   [2]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-4/igt@prime_vgem@fence-wait@ccs0.html

  * igt@prime_vgem@fence-wait@vecs0:
    - shard-mtlp:         [PASS][3] -> [ABORT][4]
   [3]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-3/igt@prime_vgem@fence-wait@vecs0.html
   [4]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-4/igt@prime_vgem@fence-wait@vecs0.html

  * igt@vgem_basic@sysfs:
    - shard-snb:          [PASS][5] -> [ABORT][6]
   [5]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-snb7/igt@vgem_basic@sysfs.html
   [6]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-snb2/igt@vgem_basic@sysfs.html

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

  Here are the changes found in IGTPW_9383_full that come from known issues:

### IGT changes ###

#### Issues hit ####

  * igt@api_intel_bb@crc32:
    - shard-tglu:         NOTRUN -> [SKIP][7] ([i915#6230])
   [7]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-8/igt@api_intel_bb@crc32.html

  * igt@device_reset@unbind-cold-reset-rebind:
    - shard-rkl:          NOTRUN -> [SKIP][8] ([i915#7701])
   [8]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@device_reset@unbind-cold-reset-rebind.html
    - shard-dg2:          NOTRUN -> [SKIP][9] ([i915#7701])
   [9]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-6/igt@device_reset@unbind-cold-reset-rebind.html

  * igt@drm_fdinfo@busy-hang@bcs0:
    - shard-dg2:          NOTRUN -> [SKIP][10] ([i915#8414]) +19 similar issues
   [10]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-3/igt@drm_fdinfo@busy-hang@bcs0.html

  * igt@drm_fdinfo@most-busy-idle-check-all@rcs0:
    - shard-rkl:          [PASS][11] -> [FAIL][12] ([i915#7742])
   [11]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-rkl-2/igt@drm_fdinfo@most-busy-idle-check-all@rcs0.html
   [12]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-4/igt@drm_fdinfo@most-busy-idle-check-all@rcs0.html

  * igt@drm_fdinfo@virtual-busy-all:
    - shard-mtlp:         NOTRUN -> [SKIP][13] ([i915#8414])
   [13]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-6/igt@drm_fdinfo@virtual-busy-all.html

  * igt@gem_bad_reloc@negative-reloc-bltcopy:
    - shard-mtlp:         NOTRUN -> [SKIP][14] ([i915#3281]) +5 similar issues
   [14]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-3/igt@gem_bad_reloc@negative-reloc-bltcopy.html

  * igt@gem_basic@multigpu-create-close:
    - shard-rkl:          NOTRUN -> [SKIP][15] ([i915#7697]) +1 similar issue
   [15]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@gem_basic@multigpu-create-close.html

  * igt@gem_ccs@block-copy-compressed:
    - shard-tglu:         NOTRUN -> [SKIP][16] ([i915#3555] / [i915#5325])
   [16]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-2/igt@gem_ccs@block-copy-compressed.html

  * igt@gem_ccs@ctrl-surf-copy-new-ctx:
    - shard-mtlp:         NOTRUN -> [SKIP][17] ([i915#5325])
   [17]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-4/igt@gem_ccs@ctrl-surf-copy-new-ctx.html

  * igt@gem_close_race@multigpu-basic-process:
    - shard-dg2:          NOTRUN -> [SKIP][18] ([i915#7697]) +1 similar issue
   [18]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-7/igt@gem_close_race@multigpu-basic-process.html

  * igt@gem_create@create-ext-cpu-access-big:
    - shard-tglu:         NOTRUN -> [SKIP][19] ([i915#6335])
   [19]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-4/igt@gem_create@create-ext-cpu-access-big.html

  * igt@gem_ctx_exec@basic-nohangcheck:
    - shard-tglu:         [PASS][20] -> [FAIL][21] ([i915#6268])
   [20]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-tglu-3/igt@gem_ctx_exec@basic-nohangcheck.html
   [21]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-5/igt@gem_ctx_exec@basic-nohangcheck.html

  * igt@gem_ctx_param@set-priority-not-supported:
    - shard-dg2:          NOTRUN -> [SKIP][22] ([fdo#109314])
   [22]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-11/igt@gem_ctx_param@set-priority-not-supported.html
    - shard-rkl:          NOTRUN -> [SKIP][23] ([fdo#109314])
   [23]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@gem_ctx_param@set-priority-not-supported.html

  * igt@gem_ctx_persistence@saturated-hostile@vecs0:
    - shard-mtlp:         [PASS][24] -> [FAIL][25] ([i915#7816]) +2 similar issues
   [24]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-5/igt@gem_ctx_persistence@saturated-hostile@vecs0.html
   [25]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-2/igt@gem_ctx_persistence@saturated-hostile@vecs0.html

  * igt@gem_eio@in-flight-contexts-1us:
    - shard-mtlp:         [PASS][26] -> [ABORT][27] ([i915#8503])
   [26]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-3/igt@gem_eio@in-flight-contexts-1us.html
   [27]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-8/igt@gem_eio@in-flight-contexts-1us.html

  * igt@gem_eio@kms:
    - shard-dg2:          [PASS][28] -> [INCOMPLETE][29] ([i915#7892])
   [28]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg2-8/igt@gem_eio@kms.html
   [29]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-3/igt@gem_eio@kms.html

  * igt@gem_exec_balancer@bonded-true-hang:
    - shard-dg2:          NOTRUN -> [SKIP][30] ([i915#4812]) +2 similar issues
   [30]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-12/igt@gem_exec_balancer@bonded-true-hang.html

  * igt@gem_exec_fair@basic-none@bcs0:
    - shard-tglu:         NOTRUN -> [FAIL][31] ([i915#2842]) +4 similar issues
   [31]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-10/igt@gem_exec_fair@basic-none@bcs0.html

  * igt@gem_exec_fair@basic-pace@vcs0:
    - shard-glk:          [PASS][32] -> [FAIL][33] ([i915#2842]) +1 similar issue
   [32]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-glk7/igt@gem_exec_fair@basic-pace@vcs0.html
   [33]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-glk1/igt@gem_exec_fair@basic-pace@vcs0.html

  * igt@gem_exec_fair@basic-throttle:
    - shard-dg2:          NOTRUN -> [SKIP][34] ([i915#3539])
   [34]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-2/igt@gem_exec_fair@basic-throttle.html

  * igt@gem_exec_fence@submit67:
    - shard-mtlp:         NOTRUN -> [SKIP][35] ([i915#4812])
   [35]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-5/igt@gem_exec_fence@submit67.html

  * igt@gem_exec_params@secure-non-root:
    - shard-tglu:         NOTRUN -> [SKIP][36] ([fdo#112283])
   [36]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-4/igt@gem_exec_params@secure-non-root.html

  * igt@gem_exec_reloc@basic-write-gtt-active:
    - shard-dg2:          NOTRUN -> [SKIP][37] ([i915#3281]) +2 similar issues
   [37]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-10/igt@gem_exec_reloc@basic-write-gtt-active.html
    - shard-rkl:          NOTRUN -> [SKIP][38] ([i915#3281]) +1 similar issue
   [38]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-1/igt@gem_exec_reloc@basic-write-gtt-active.html

  * igt@gem_exec_whisper@basic-forked:
    - shard-mtlp:         NOTRUN -> [FAIL][39] ([i915#6363])
   [39]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-2/igt@gem_exec_whisper@basic-forked.html

  * igt@gem_exec_whisper@basic-forked-all:
    - shard-mtlp:         [PASS][40] -> [FAIL][41] ([i915#6363])
   [40]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-3/igt@gem_exec_whisper@basic-forked-all.html
   [41]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-3/igt@gem_exec_whisper@basic-forked-all.html

  * igt@gem_lmem_swapping@heavy-verify-random-ccs:
    - shard-tglu:         NOTRUN -> [SKIP][42] ([i915#4613]) +2 similar issues
   [42]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-10/igt@gem_lmem_swapping@heavy-verify-random-ccs.html

  * igt@gem_lmem_swapping@parallel-multi:
    - shard-apl:          NOTRUN -> [SKIP][43] ([fdo#109271] / [i915#4613])
   [43]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-apl1/igt@gem_lmem_swapping@parallel-multi.html

  * igt@gem_lmem_swapping@parallel-random:
    - shard-mtlp:         NOTRUN -> [SKIP][44] ([i915#4613])
   [44]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-3/igt@gem_lmem_swapping@parallel-random.html

  * igt@gem_lmem_swapping@smem-oom@lmem0:
    - shard-dg2:          [PASS][45] -> [TIMEOUT][46] ([i915#5493])
   [45]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg2-11/igt@gem_lmem_swapping@smem-oom@lmem0.html
   [46]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-10/igt@gem_lmem_swapping@smem-oom@lmem0.html

  * igt@gem_media_vme:
    - shard-tglu:         NOTRUN -> [SKIP][47] ([i915#284])
   [47]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-4/igt@gem_media_vme.html

  * igt@gem_mmap_wc@write:
    - shard-mtlp:         NOTRUN -> [SKIP][48] ([i915#4083]) +1 similar issue
   [48]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-6/igt@gem_mmap_wc@write.html
    - shard-dg2:          NOTRUN -> [SKIP][49] ([i915#4083]) +3 similar issues
   [49]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-5/igt@gem_mmap_wc@write.html

  * igt@gem_partial_pwrite_pread@reads:
    - shard-mtlp:         NOTRUN -> [SKIP][50] ([i915#3282]) +3 similar issues
   [50]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-5/igt@gem_partial_pwrite_pread@reads.html

  * igt@gem_partial_pwrite_pread@writes-after-reads:
    - shard-rkl:          NOTRUN -> [SKIP][51] ([i915#3282]) +2 similar issues
   [51]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-1/igt@gem_partial_pwrite_pread@writes-after-reads.html

  * igt@gem_pxp@create-regular-buffer:
    - shard-mtlp:         NOTRUN -> [SKIP][52] ([i915#4270])
   [52]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-6/igt@gem_pxp@create-regular-buffer.html

  * igt@gem_pxp@fail-invalid-protected-context:
    - shard-tglu:         NOTRUN -> [SKIP][53] ([i915#4270]) +1 similar issue
   [53]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-10/igt@gem_pxp@fail-invalid-protected-context.html

  * igt@gem_readwrite@beyond-eob:
    - shard-dg2:          NOTRUN -> [SKIP][54] ([i915#3282]) +3 similar issues
   [54]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-5/igt@gem_readwrite@beyond-eob.html

  * igt@gem_render_copy@y-tiled-ccs-to-yf-tiled-ccs:
    - shard-mtlp:         NOTRUN -> [SKIP][55] ([i915#8428]) +3 similar issues
   [55]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-4/igt@gem_render_copy@y-tiled-ccs-to-yf-tiled-ccs.html

  * igt@gem_set_tiling_vs_pwrite:
    - shard-dg2:          NOTRUN -> [SKIP][56] ([i915#4079])
   [56]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-7/igt@gem_set_tiling_vs_pwrite.html

  * igt@gem_tiled_fence_blits@basic:
    - shard-dg2:          NOTRUN -> [SKIP][57] ([i915#4077]) +3 similar issues
   [57]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-6/igt@gem_tiled_fence_blits@basic.html

  * igt@gem_tiled_pread_basic:
    - shard-mtlp:         NOTRUN -> [SKIP][58] ([i915#4079])
   [58]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-2/igt@gem_tiled_pread_basic.html

  * igt@gem_userptr_blits@coherency-unsync:
    - shard-mtlp:         NOTRUN -> [SKIP][59] ([i915#3297])
   [59]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-5/igt@gem_userptr_blits@coherency-unsync.html

  * igt@gem_userptr_blits@invalid-mmap-offset-unsync:
    - shard-dg2:          NOTRUN -> [SKIP][60] ([i915#3297])
   [60]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-3/igt@gem_userptr_blits@invalid-mmap-offset-unsync.html
    - shard-rkl:          NOTRUN -> [SKIP][61] ([i915#3297])
   [61]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@gem_userptr_blits@invalid-mmap-offset-unsync.html

  * igt@gem_userptr_blits@map-fixed-invalidate-overlap-busy:
    - shard-dg2:          NOTRUN -> [SKIP][62] ([i915#3297] / [i915#4880])
   [62]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-7/igt@gem_userptr_blits@map-fixed-invalidate-overlap-busy.html

  * igt@gem_userptr_blits@nohangcheck:
    - shard-mtlp:         [PASS][63] -> [FAIL][64] ([i915#7916])
   [63]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-7/igt@gem_userptr_blits@nohangcheck.html
   [64]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-8/igt@gem_userptr_blits@nohangcheck.html

  * igt@gen7_exec_parse@load-register-reg:
    - shard-dg2:          NOTRUN -> [SKIP][65] ([fdo#109289])
   [65]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-6/igt@gen7_exec_parse@load-register-reg.html

  * igt@gen7_exec_parse@oacontrol-tracking:
    - shard-mtlp:         NOTRUN -> [SKIP][66] ([fdo#109289])
   [66]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-8/igt@gen7_exec_parse@oacontrol-tracking.html

  * igt@gen9_exec_parse@bb-large:
    - shard-dg2:          NOTRUN -> [SKIP][67] ([i915#2856])
   [67]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-10/igt@gen9_exec_parse@bb-large.html
    - shard-rkl:          NOTRUN -> [SKIP][68] ([i915#2527])
   [68]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-1/igt@gen9_exec_parse@bb-large.html

  * igt@i915_hangman@gt-engine-error@vcs0:
    - shard-mtlp:         [PASS][69] -> [FAIL][70] ([i915#7069])
   [69]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-5/igt@i915_hangman@gt-engine-error@vcs0.html
   [70]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-4/igt@i915_hangman@gt-engine-error@vcs0.html

  * igt@i915_module_load@load:
    - shard-rkl:          NOTRUN -> [SKIP][71] ([i915#6227])
   [71]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@i915_module_load@load.html

  * igt@i915_pm_backlight@bad-brightness:
    - shard-dg2:          NOTRUN -> [SKIP][72] ([i915#5354] / [i915#7561])
   [72]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-1/igt@i915_pm_backlight@bad-brightness.html

  * igt@i915_pm_dc@dc5-psr:
    - shard-tglu:         NOTRUN -> [SKIP][73] ([i915#658])
   [73]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-3/igt@i915_pm_dc@dc5-psr.html

  * igt@i915_pm_dc@dc9-dpms:
    - shard-apl:          [PASS][74] -> [SKIP][75] ([fdo#109271])
   [74]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-apl3/igt@i915_pm_dc@dc9-dpms.html
   [75]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-apl7/igt@i915_pm_dc@dc9-dpms.html

  * igt@i915_pm_lpsp@screens-disabled:
    - shard-mtlp:         NOTRUN -> [SKIP][76] ([i915#8430])
   [76]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-3/igt@i915_pm_lpsp@screens-disabled.html

  * igt@i915_pm_rpm@modeset-non-lpsp-stress:
    - shard-dg2:          [PASS][77] -> [SKIP][78] ([i915#1397]) +1 similar issue
   [77]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg2-7/igt@i915_pm_rpm@modeset-non-lpsp-stress.html
   [78]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-12/igt@i915_pm_rpm@modeset-non-lpsp-stress.html

  * igt@i915_pm_rpm@pc8-residency:
    - shard-mtlp:         NOTRUN -> [SKIP][79] ([fdo#109293])
   [79]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-3/igt@i915_pm_rpm@pc8-residency.html

  * igt@i915_pm_rpm@system-suspend-execbuf:
    - shard-tglu:         NOTRUN -> [FAIL][80] ([i915#7940])
   [80]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-2/igt@i915_pm_rpm@system-suspend-execbuf.html

  * igt@i915_suspend@basic-s3-without-i915:
    - shard-dg2:          [PASS][81] -> [FAIL][82] ([fdo#103375] / [i915#6121]) +2 similar issues
   [81]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg2-2/igt@i915_suspend@basic-s3-without-i915.html
   [82]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-5/igt@i915_suspend@basic-s3-without-i915.html
    - shard-tglu:         NOTRUN -> [INCOMPLETE][83] ([i915#7443] / [i915#8102])
   [83]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-3/igt@i915_suspend@basic-s3-without-i915.html

  * igt@i915_suspend@debugfs-reader:
    - shard-apl:          [PASS][84] -> [ABORT][85] ([i915#180] / [i915#8213])
   [84]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-apl4/igt@i915_suspend@debugfs-reader.html
   [85]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-apl7/igt@i915_suspend@debugfs-reader.html

  * igt@i915_suspend@fence-restore-tiled2untiled:
    - shard-snb:          NOTRUN -> [DMESG-WARN][86] ([i915#8841]) +4 similar issues
   [86]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-snb4/igt@i915_suspend@fence-restore-tiled2untiled.html

  * igt@kms_addfb_basic@addfb25-x-tiled-legacy:
    - shard-mtlp:         NOTRUN -> [SKIP][87] ([i915#4212])
   [87]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-5/igt@kms_addfb_basic@addfb25-x-tiled-legacy.html

  * igt@kms_addfb_basic@basic-y-tiled-legacy:
    - shard-dg2:          NOTRUN -> [SKIP][88] ([i915#4215] / [i915#5190])
   [88]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-10/igt@kms_addfb_basic@basic-y-tiled-legacy.html

  * igt@kms_async_flips@crc@pipe-a-hdmi-a-3:
    - shard-dg2:          NOTRUN -> [FAIL][89] ([i915#8247]) +3 similar issues
   [89]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-5/igt@kms_async_flips@crc@pipe-a-hdmi-a-3.html

  * igt@kms_big_fb@4-tiled-16bpp-rotate-90:
    - shard-mtlp:         NOTRUN -> [SKIP][90] ([fdo#111614])
   [90]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-6/igt@kms_big_fb@4-tiled-16bpp-rotate-90.html

  * igt@kms_big_fb@4-tiled-8bpp-rotate-270:
    - shard-dg2:          NOTRUN -> [SKIP][91] ([fdo#111614]) +1 similar issue
   [91]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-6/igt@kms_big_fb@4-tiled-8bpp-rotate-270.html

  * igt@kms_big_fb@4-tiled-max-hw-stride-32bpp-rotate-180-async-flip:
    - shard-mtlp:         [PASS][92] -> [FAIL][93] ([i915#3743])
   [92]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-8/igt@kms_big_fb@4-tiled-max-hw-stride-32bpp-rotate-180-async-flip.html
   [93]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-1/igt@kms_big_fb@4-tiled-max-hw-stride-32bpp-rotate-180-async-flip.html

  * igt@kms_big_fb@4-tiled-max-hw-stride-32bpp-rotate-180-hflip:
    - shard-tglu:         NOTRUN -> [SKIP][94] ([fdo#111615] / [i915#5286])
   [94]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-9/igt@kms_big_fb@4-tiled-max-hw-stride-32bpp-rotate-180-hflip.html

  * igt@kms_big_fb@4-tiled-max-hw-stride-64bpp-rotate-0-hflip:
    - shard-rkl:          NOTRUN -> [SKIP][95] ([i915#5286]) +1 similar issue
   [95]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-2/igt@kms_big_fb@4-tiled-max-hw-stride-64bpp-rotate-0-hflip.html
    - shard-mtlp:         [PASS][96] -> [FAIL][97] ([i915#5138])
   [96]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-7/igt@kms_big_fb@4-tiled-max-hw-stride-64bpp-rotate-0-hflip.html
   [97]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-5/igt@kms_big_fb@4-tiled-max-hw-stride-64bpp-rotate-0-hflip.html

  * igt@kms_big_fb@x-tiled-32bpp-rotate-90:
    - shard-rkl:          NOTRUN -> [SKIP][98] ([fdo#111614] / [i915#3638])
   [98]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@kms_big_fb@x-tiled-32bpp-rotate-90.html

  * igt@kms_big_fb@x-tiled-max-hw-stride-64bpp-rotate-0-hflip-async-flip:
    - shard-mtlp:         NOTRUN -> [FAIL][99] ([i915#3743])
   [99]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-8/igt@kms_big_fb@x-tiled-max-hw-stride-64bpp-rotate-0-hflip-async-flip.html

  * igt@kms_big_fb@y-tiled-8bpp-rotate-90:
    - shard-tglu:         NOTRUN -> [SKIP][100] ([fdo#111614])
   [100]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-2/igt@kms_big_fb@y-tiled-8bpp-rotate-90.html

  * igt@kms_big_fb@yf-tiled-64bpp-rotate-90:
    - shard-rkl:          NOTRUN -> [SKIP][101] ([fdo#110723]) +2 similar issues
   [101]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-2/igt@kms_big_fb@yf-tiled-64bpp-rotate-90.html

  * igt@kms_big_fb@yf-tiled-addfb-size-overflow:
    - shard-dg2:          NOTRUN -> [SKIP][102] ([i915#5190]) +5 similar issues
   [102]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-7/igt@kms_big_fb@yf-tiled-addfb-size-overflow.html
    - shard-tglu:         NOTRUN -> [SKIP][103] ([fdo#111615]) +1 similar issue
   [103]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-10/igt@kms_big_fb@yf-tiled-addfb-size-overflow.html
    - shard-mtlp:         NOTRUN -> [SKIP][104] ([i915#6187])
   [104]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-8/igt@kms_big_fb@yf-tiled-addfb-size-overflow.html

  * igt@kms_big_fb@yf-tiled-max-hw-stride-32bpp-rotate-180-async-flip:
    - shard-mtlp:         NOTRUN -> [SKIP][105] ([fdo#111615]) +4 similar issues
   [105]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-4/igt@kms_big_fb@yf-tiled-max-hw-stride-32bpp-rotate-180-async-flip.html

  * igt@kms_big_fb@yf-tiled-max-hw-stride-64bpp-rotate-0:
    - shard-dg2:          NOTRUN -> [SKIP][106] ([i915#4538] / [i915#5190]) +3 similar issues
   [106]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-1/igt@kms_big_fb@yf-tiled-max-hw-stride-64bpp-rotate-0.html

  * igt@kms_ccs@pipe-a-bad-pixel-format-y_tiled_gen12_mc_ccs:
    - shard-mtlp:         NOTRUN -> [SKIP][107] ([i915#3886] / [i915#6095]) +5 similar issues
   [107]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-3/igt@kms_ccs@pipe-a-bad-pixel-format-y_tiled_gen12_mc_ccs.html

  * igt@kms_ccs@pipe-a-ccs-on-another-bo-4_tiled_mtl_rc_ccs_cc:
    - shard-rkl:          NOTRUN -> [SKIP][108] ([i915#5354] / [i915#6095]) +4 similar issues
   [108]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@kms_ccs@pipe-a-ccs-on-another-bo-4_tiled_mtl_rc_ccs_cc.html

  * igt@kms_ccs@pipe-a-crc-sprite-planes-basic-y_tiled_ccs:
    - shard-dg2:          NOTRUN -> [SKIP][109] ([i915#3689] / [i915#5354]) +12 similar issues
   [109]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-12/igt@kms_ccs@pipe-a-crc-sprite-planes-basic-y_tiled_ccs.html

  * igt@kms_ccs@pipe-a-missing-ccs-buffer-4_tiled_mtl_rc_ccs_cc:
    - shard-tglu:         NOTRUN -> [SKIP][110] ([i915#5354] / [i915#6095]) +8 similar issues
   [110]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-3/igt@kms_ccs@pipe-a-missing-ccs-buffer-4_tiled_mtl_rc_ccs_cc.html

  * igt@kms_ccs@pipe-b-bad-aux-stride-y_tiled_gen12_mc_ccs:
    - shard-rkl:          NOTRUN -> [SKIP][111] ([i915#3886] / [i915#5354] / [i915#6095])
   [111]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-2/igt@kms_ccs@pipe-b-bad-aux-stride-y_tiled_gen12_mc_ccs.html

  * igt@kms_ccs@pipe-b-ccs-on-another-bo-y_tiled_gen12_mc_ccs:
    - shard-tglu:         NOTRUN -> [SKIP][112] ([i915#3689] / [i915#3886] / [i915#5354] / [i915#6095]) +2 similar issues
   [112]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-7/igt@kms_ccs@pipe-b-ccs-on-another-bo-y_tiled_gen12_mc_ccs.html

  * igt@kms_ccs@pipe-b-crc-primary-basic-yf_tiled_ccs:
    - shard-rkl:          NOTRUN -> [SKIP][113] ([i915#3734] / [i915#5354] / [i915#6095]) +1 similar issue
   [113]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-2/igt@kms_ccs@pipe-b-crc-primary-basic-yf_tiled_ccs.html

  * igt@kms_ccs@pipe-c-bad-aux-stride-yf_tiled_ccs:
    - shard-tglu:         NOTRUN -> [SKIP][114] ([fdo#111615] / [i915#3689] / [i915#5354] / [i915#6095]) +1 similar issue
   [114]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-5/igt@kms_ccs@pipe-c-bad-aux-stride-yf_tiled_ccs.html

  * igt@kms_ccs@pipe-c-bad-rotation-90-4_tiled_mtl_mc_ccs:
    - shard-dg2:          NOTRUN -> [SKIP][115] ([i915#5354]) +18 similar issues
   [115]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-7/igt@kms_ccs@pipe-c-bad-rotation-90-4_tiled_mtl_mc_ccs.html
    - shard-rkl:          NOTRUN -> [SKIP][116] ([i915#5354]) +8 similar issues
   [116]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-1/igt@kms_ccs@pipe-c-bad-rotation-90-4_tiled_mtl_mc_ccs.html

  * igt@kms_ccs@pipe-c-bad-rotation-90-y_tiled_gen12_rc_ccs_cc:
    - shard-dg2:          NOTRUN -> [SKIP][117] ([i915#3689] / [i915#3886] / [i915#5354]) +3 similar issues
   [117]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-5/igt@kms_ccs@pipe-c-bad-rotation-90-y_tiled_gen12_rc_ccs_cc.html

  * igt@kms_ccs@pipe-d-bad-rotation-90-4_tiled_dg2_rc_ccs_cc:
    - shard-tglu:         NOTRUN -> [SKIP][118] ([i915#3689] / [i915#5354] / [i915#6095]) +9 similar issues
   [118]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-5/igt@kms_ccs@pipe-d-bad-rotation-90-4_tiled_dg2_rc_ccs_cc.html

  * igt@kms_ccs@pipe-d-crc-primary-rotation-180-yf_tiled_ccs:
    - shard-mtlp:         NOTRUN -> [SKIP][119] ([i915#6095]) +16 similar issues
   [119]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-5/igt@kms_ccs@pipe-d-crc-primary-rotation-180-yf_tiled_ccs.html

  * igt@kms_chamelium_audio@dp-audio:
    - shard-tglu:         NOTRUN -> [SKIP][120] ([i915#7828]) +3 similar issues
   [120]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-10/igt@kms_chamelium_audio@dp-audio.html
    - shard-mtlp:         NOTRUN -> [SKIP][121] ([i915#7828])
   [121]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-8/igt@kms_chamelium_audio@dp-audio.html

  * igt@kms_chamelium_color@ctm-0-50:
    - shard-tglu:         NOTRUN -> [SKIP][122] ([fdo#111827]) +1 similar issue
   [122]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-6/igt@kms_chamelium_color@ctm-0-50.html

  * igt@kms_chamelium_color@ctm-0-75:
    - shard-mtlp:         NOTRUN -> [SKIP][123] ([fdo#111827])
   [123]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-2/igt@kms_chamelium_color@ctm-0-75.html

  * igt@kms_chamelium_color@ctm-blue-to-red:
    - shard-dg2:          NOTRUN -> [SKIP][124] ([fdo#111827])
   [124]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-1/igt@kms_chamelium_color@ctm-blue-to-red.html

  * igt@kms_chamelium_edid@dp-edid-stress-resolution-non-4k:
    - shard-rkl:          NOTRUN -> [SKIP][125] ([i915#7828])
   [125]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@kms_chamelium_edid@dp-edid-stress-resolution-non-4k.html

  * igt@kms_chamelium_frames@hdmi-cmp-planar-formats:
    - shard-dg2:          NOTRUN -> [SKIP][126] ([i915#7828]) +1 similar issue
   [126]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-5/igt@kms_chamelium_frames@hdmi-cmp-planar-formats.html

  * igt@kms_content_protection@legacy:
    - shard-dg2:          NOTRUN -> [SKIP][127] ([i915#7118]) +2 similar issues
   [127]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-2/igt@kms_content_protection@legacy.html

  * igt@kms_content_protection@mei_interface:
    - shard-mtlp:         NOTRUN -> [SKIP][128] ([i915#8063])
   [128]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-6/igt@kms_content_protection@mei_interface.html

  * igt@kms_content_protection@uevent:
    - shard-rkl:          NOTRUN -> [SKIP][129] ([i915#7118])
   [129]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-2/igt@kms_content_protection@uevent.html

  * igt@kms_cursor_crc@cursor-onscreen-512x170:
    - shard-rkl:          NOTRUN -> [SKIP][130] ([fdo#109279] / [i915#3359])
   [130]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-1/igt@kms_cursor_crc@cursor-onscreen-512x170.html

  * igt@kms_cursor_crc@cursor-onscreen-512x512:
    - shard-tglu:         NOTRUN -> [SKIP][131] ([i915#3359]) +1 similar issue
   [131]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-10/igt@kms_cursor_crc@cursor-onscreen-512x512.html

  * igt@kms_cursor_crc@cursor-random-32x10:
    - shard-mtlp:         NOTRUN -> [SKIP][132] ([i915#8814])
   [132]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-1/igt@kms_cursor_crc@cursor-random-32x10.html

  * igt@kms_cursor_crc@cursor-random-512x512:
    - shard-mtlp:         NOTRUN -> [SKIP][133] ([i915#3359])
   [133]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-6/igt@kms_cursor_crc@cursor-random-512x512.html

  * igt@kms_cursor_crc@cursor-sliding-512x512:
    - shard-dg2:          NOTRUN -> [SKIP][134] ([i915#3359]) +2 similar issues
   [134]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-5/igt@kms_cursor_crc@cursor-sliding-512x512.html
    - shard-rkl:          NOTRUN -> [SKIP][135] ([i915#3359]) +1 similar issue
   [135]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@kms_cursor_crc@cursor-sliding-512x512.html

  * igt@kms_cursor_legacy@2x-long-flip-vs-cursor-legacy:
    - shard-mtlp:         NOTRUN -> [SKIP][136] ([i915#3546])
   [136]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-5/igt@kms_cursor_legacy@2x-long-flip-vs-cursor-legacy.html

  * igt@kms_cursor_legacy@cursora-vs-flipb-legacy:
    - shard-tglu:         NOTRUN -> [SKIP][137] ([fdo#109274]) +3 similar issues
   [137]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-9/igt@kms_cursor_legacy@cursora-vs-flipb-legacy.html

  * igt@kms_cursor_legacy@cursorb-vs-flipa-toggle:
    - shard-dg2:          NOTRUN -> [SKIP][138] ([fdo#109274] / [i915#5354]) +1 similar issue
   [138]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-3/igt@kms_cursor_legacy@cursorb-vs-flipa-toggle.html
    - shard-rkl:          NOTRUN -> [SKIP][139] ([fdo#111825]) +1 similar issue
   [139]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@kms_cursor_legacy@cursorb-vs-flipa-toggle.html

  * igt@kms_cursor_legacy@cursorb-vs-flipb-toggle:
    - shard-dg2:          NOTRUN -> [SKIP][140] ([fdo#109274] / [fdo#111767] / [i915#5354])
   [140]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-3/igt@kms_cursor_legacy@cursorb-vs-flipb-toggle.html
    - shard-rkl:          NOTRUN -> [SKIP][141] ([fdo#111767] / [fdo#111825])
   [141]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-4/igt@kms_cursor_legacy@cursorb-vs-flipb-toggle.html

  * igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size:
    - shard-apl:          [PASS][142] -> [FAIL][143] ([i915#2346])
   [142]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-apl6/igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size.html
   [143]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-apl3/igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size.html
    - shard-glk:          [PASS][144] -> [FAIL][145] ([i915#2346])
   [144]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-glk8/igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size.html
   [145]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-glk3/igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size.html

  * igt@kms_dither@fb-8bpc-vs-panel-6bpc@pipe-a-hdmi-a-1:
    - shard-rkl:          NOTRUN -> [SKIP][146] ([i915#3804])
   [146]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@kms_dither@fb-8bpc-vs-panel-6bpc@pipe-a-hdmi-a-1.html

  * igt@kms_dp_aux_dev:
    - shard-tglu:         NOTRUN -> [SKIP][147] ([i915#1257])
   [147]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-2/igt@kms_dp_aux_dev.html

  * igt@kms_dsc@dsc-with-bpc-formats:
    - shard-tglu:         NOTRUN -> [SKIP][148] ([i915#3555] / [i915#3840])
   [148]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-9/igt@kms_dsc@dsc-with-bpc-formats.html

  * igt@kms_dsc@dsc-with-output-formats:
    - shard-mtlp:         NOTRUN -> [SKIP][149] ([i915#3840])
   [149]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-1/igt@kms_dsc@dsc-with-output-formats.html

  * igt@kms_fbcon_fbt@psr:
    - shard-dg2:          NOTRUN -> [SKIP][150] ([i915#3469])
   [150]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-7/igt@kms_fbcon_fbt@psr.html
    - shard-rkl:          NOTRUN -> [SKIP][151] ([fdo#110189] / [i915#3955])
   [151]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-1/igt@kms_fbcon_fbt@psr.html

  * igt@kms_flip@2x-blocking-absolute-wf_vblank:
    - shard-tglu:         NOTRUN -> [SKIP][152] ([fdo#109274] / [i915#3637]) +2 similar issues
   [152]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-4/igt@kms_flip@2x-blocking-absolute-wf_vblank.html

  * igt@kms_flip@2x-flip-vs-blocking-wf-vblank:
    - shard-snb:          NOTRUN -> [SKIP][153] ([fdo#109271] / [fdo#111767])
   [153]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-snb5/igt@kms_flip@2x-flip-vs-blocking-wf-vblank.html

  * igt@kms_flip@2x-flip-vs-fences-interruptible:
    - shard-mtlp:         NOTRUN -> [SKIP][154] ([i915#8381])
   [154]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-3/igt@kms_flip@2x-flip-vs-fences-interruptible.html

  * igt@kms_flip@2x-plain-flip-fb-recreate:
    - shard-apl:          NOTRUN -> [SKIP][155] ([fdo#109271]) +24 similar issues
   [155]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-apl6/igt@kms_flip@2x-plain-flip-fb-recreate.html

  * igt@kms_flip@2x-plain-flip-interruptible:
    - shard-mtlp:         NOTRUN -> [SKIP][156] ([i915#3637])
   [156]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-1/igt@kms_flip@2x-plain-flip-interruptible.html

  * igt@kms_flip@2x-wf_vblank-ts-check-interruptible:
    - shard-dg2:          NOTRUN -> [SKIP][157] ([fdo#109274])
   [157]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-8/igt@kms_flip@2x-wf_vblank-ts-check-interruptible.html

  * igt@kms_flip_scaled_crc@flip-32bpp-4tile-to-32bpp-4tiledg2rcccs-downscaling@pipe-a-valid-mode:
    - shard-rkl:          NOTRUN -> [SKIP][158] ([i915#2672])
   [158]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-2/igt@kms_flip_scaled_crc@flip-32bpp-4tile-to-32bpp-4tiledg2rcccs-downscaling@pipe-a-valid-mode.html

  * igt@kms_flip_scaled_crc@flip-32bpp-ytile-to-32bpp-ytilegen12rcccs-upscaling@pipe-a-default-mode:
    - shard-mtlp:         NOTRUN -> [SKIP][159] ([i915#2672]) +1 similar issue
   [159]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-6/igt@kms_flip_scaled_crc@flip-32bpp-ytile-to-32bpp-ytilegen12rcccs-upscaling@pipe-a-default-mode.html

  * igt@kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-downscaling@pipe-a-default-mode:
    - shard-mtlp:         NOTRUN -> [SKIP][160] ([i915#8810])
   [160]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-2/igt@kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-downscaling@pipe-a-default-mode.html

  * igt@kms_force_connector_basic@prune-stale-modes:
    - shard-mtlp:         NOTRUN -> [SKIP][161] ([i915#5274])
   [161]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-3/igt@kms_force_connector_basic@prune-stale-modes.html
    - shard-dg2:          NOTRUN -> [SKIP][162] ([i915#5274])
   [162]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-5/igt@kms_force_connector_basic@prune-stale-modes.html

  * igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-draw-mmap-gtt:
    - shard-mtlp:         NOTRUN -> [SKIP][163] ([i915#8708]) +1 similar issue
   [163]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-3/igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-draw-mmap-gtt.html

  * igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-fullscreen:
    - shard-dg2:          [PASS][164] -> [FAIL][165] ([i915#6880])
   [164]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg2-8/igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-fullscreen.html
   [165]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-11/igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-fullscreen.html

  * igt@kms_frontbuffer_tracking@fbcpsr-1p-primscrn-pri-shrfb-draw-mmap-gtt:
    - shard-rkl:          NOTRUN -> [SKIP][166] ([i915#3023]) +4 similar issues
   [166]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-4/igt@kms_frontbuffer_tracking@fbcpsr-1p-primscrn-pri-shrfb-draw-mmap-gtt.html

  * igt@kms_frontbuffer_tracking@fbcpsr-2p-primscrn-pri-shrfb-draw-mmap-wc:
    - shard-dg2:          NOTRUN -> [SKIP][167] ([i915#8708]) +5 similar issues
   [167]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-5/igt@kms_frontbuffer_tracking@fbcpsr-2p-primscrn-pri-shrfb-draw-mmap-wc.html
    - shard-rkl:          NOTRUN -> [SKIP][168] ([fdo#111825] / [i915#1825]) +8 similar issues
   [168]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-4/igt@kms_frontbuffer_tracking@fbcpsr-2p-primscrn-pri-shrfb-draw-mmap-wc.html

  * igt@kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-pri-shrfb-draw-pwrite:
    - shard-tglu:         NOTRUN -> [SKIP][169] ([fdo#109280]) +14 similar issues
   [169]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-4/igt@kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-pri-shrfb-draw-pwrite.html

  * igt@kms_frontbuffer_tracking@fbcpsr-tiling-4:
    - shard-tglu:         NOTRUN -> [SKIP][170] ([i915#5439])
   [170]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-6/igt@kms_frontbuffer_tracking@fbcpsr-tiling-4.html

  * igt@kms_frontbuffer_tracking@psr-1p-offscren-pri-indfb-draw-mmap-cpu:
    - shard-tglu:         NOTRUN -> [SKIP][171] ([fdo#110189]) +12 similar issues
   [171]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-10/igt@kms_frontbuffer_tracking@psr-1p-offscren-pri-indfb-draw-mmap-cpu.html

  * igt@kms_frontbuffer_tracking@psr-1p-offscren-pri-shrfb-draw-render:
    - shard-dg2:          NOTRUN -> [SKIP][172] ([i915#3458]) +3 similar issues
   [172]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-1/igt@kms_frontbuffer_tracking@psr-1p-offscren-pri-shrfb-draw-render.html

  * igt@kms_frontbuffer_tracking@psr-2p-scndscrn-pri-shrfb-draw-mmap-cpu:
    - shard-mtlp:         NOTRUN -> [SKIP][173] ([i915#1825]) +10 similar issues
   [173]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-1/igt@kms_frontbuffer_tracking@psr-2p-scndscrn-pri-shrfb-draw-mmap-cpu.html

  * igt@kms_hdr@bpc-switch:
    - shard-tglu:         NOTRUN -> [SKIP][174] ([i915#3555] / [i915#8228])
   [174]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-7/igt@kms_hdr@bpc-switch.html

  * igt@kms_hdr@bpc-switch-suspend:
    - shard-rkl:          NOTRUN -> [SKIP][175] ([i915#3555] / [i915#8228])
   [175]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@kms_hdr@bpc-switch-suspend.html

  * igt@kms_hdr@static-toggle:
    - shard-dg2:          NOTRUN -> [SKIP][176] ([i915#3555] / [i915#8228]) +1 similar issue
   [176]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-5/igt@kms_hdr@static-toggle.html

  * igt@kms_pipe_b_c_ivb@pipe-b-dpms-off-modeset-pipe-c:
    - shard-tglu:         NOTRUN -> [SKIP][177] ([fdo#109289]) +1 similar issue
   [177]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-7/igt@kms_pipe_b_c_ivb@pipe-b-dpms-off-modeset-pipe-c.html

  * igt@kms_plane_scaling@intel-max-src-size:
    - shard-dg2:          NOTRUN -> [SKIP][178] ([i915#6953])
   [178]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-6/igt@kms_plane_scaling@intel-max-src-size.html

  * igt@kms_plane_scaling@plane-scaler-with-clipping-clamping-rotation@pipe-a-vga-1:
    - shard-snb:          NOTRUN -> [SKIP][179] ([fdo#109271]) +101 similar issues
   [179]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-snb2/igt@kms_plane_scaling@plane-scaler-with-clipping-clamping-rotation@pipe-a-vga-1.html

  * igt@kms_plane_scaling@plane-upscale-with-rotation-factor-0-25@pipe-a-hdmi-a-1:
    - shard-tglu:         NOTRUN -> [SKIP][180] ([i915#5176]) +15 similar issues
   [180]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-10/igt@kms_plane_scaling@plane-upscale-with-rotation-factor-0-25@pipe-a-hdmi-a-1.html

  * igt@kms_plane_scaling@plane-upscale-with-rotation-factor-0-25@pipe-a-hdmi-a-2:
    - shard-rkl:          NOTRUN -> [SKIP][181] ([i915#5176]) +7 similar issues
   [181]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-1/igt@kms_plane_scaling@plane-upscale-with-rotation-factor-0-25@pipe-a-hdmi-a-2.html

  * igt@kms_plane_scaling@planes-downscale-factor-0-25@pipe-d-hdmi-a-3:
    - shard-dg2:          NOTRUN -> [SKIP][182] ([i915#5235]) +15 similar issues
   [182]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-6/igt@kms_plane_scaling@planes-downscale-factor-0-25@pipe-d-hdmi-a-3.html

  * igt@kms_plane_scaling@planes-upscale-20x20-downscale-factor-0-25@pipe-b-edp-1:
    - shard-mtlp:         NOTRUN -> [SKIP][183] ([i915#5235]) +3 similar issues
   [183]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-7/igt@kms_plane_scaling@planes-upscale-20x20-downscale-factor-0-25@pipe-b-edp-1.html

  * igt@kms_plane_scaling@planes-upscale-factor-0-25-downscale-factor-0-25@pipe-b-hdmi-a-2:
    - shard-rkl:          NOTRUN -> [SKIP][184] ([i915#5235]) +3 similar issues
   [184]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-1/igt@kms_plane_scaling@planes-upscale-factor-0-25-downscale-factor-0-25@pipe-b-hdmi-a-2.html

  * igt@kms_psr2_sf@cursor-plane-move-continuous-sf:
    - shard-apl:          NOTRUN -> [SKIP][185] ([fdo#109271] / [i915#658])
   [185]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-apl6/igt@kms_psr2_sf@cursor-plane-move-continuous-sf.html

  * igt@kms_psr2_su@frontbuffer-xrgb8888:
    - shard-dg2:          NOTRUN -> [SKIP][186] ([i915#658]) +1 similar issue
   [186]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-5/igt@kms_psr2_su@frontbuffer-xrgb8888.html
    - shard-mtlp:         NOTRUN -> [SKIP][187] ([i915#4348])
   [187]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-1/igt@kms_psr2_su@frontbuffer-xrgb8888.html

  * igt@kms_psr2_su@page_flip-xrgb8888:
    - shard-tglu:         NOTRUN -> [SKIP][188] ([fdo#109642] / [fdo#111068] / [i915#658]) +1 similar issue
   [188]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-10/igt@kms_psr2_su@page_flip-xrgb8888.html

  * igt@kms_psr@primary_mmap_cpu:
    - shard-dg2:          NOTRUN -> [SKIP][189] ([i915#1072]) +4 similar issues
   [189]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-3/igt@kms_psr@primary_mmap_cpu.html
    - shard-rkl:          NOTRUN -> [SKIP][190] ([i915#1072]) +2 similar issues
   [190]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@kms_psr@primary_mmap_cpu.html

  * igt@kms_psr@psr2_cursor_blt:
    - shard-mtlp:         [PASS][191] -> [FAIL][192] ([i915#8726])
   [191]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-8/igt@kms_psr@psr2_cursor_blt.html
   [192]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-7/igt@kms_psr@psr2_cursor_blt.html

  * igt@kms_psr@psr2_cursor_mmap_gtt:
    - shard-mtlp:         NOTRUN -> [SKIP][193] ([i915#4077]) +2 similar issues
   [193]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-5/igt@kms_psr@psr2_cursor_mmap_gtt.html

  * igt@kms_rmfb@close-fd@pipe-a-edp-1:
    - shard-mtlp:         [PASS][194] -> [DMESG-WARN][195] ([i915#1982])
   [194]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-1/igt@kms_rmfb@close-fd@pipe-a-edp-1.html
   [195]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-6/igt@kms_rmfb@close-fd@pipe-a-edp-1.html

  * igt@kms_rotation_crc@primary-y-tiled-reflect-x-270:
    - shard-mtlp:         NOTRUN -> [SKIP][196] ([i915#4235])
   [196]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-5/igt@kms_rotation_crc@primary-y-tiled-reflect-x-270.html

  * igt@kms_scaling_modes@scaling-mode-full:
    - shard-tglu:         NOTRUN -> [SKIP][197] ([i915#3555]) +2 similar issues
   [197]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-4/igt@kms_scaling_modes@scaling-mode-full.html

  * igt@kms_tiled_display@basic-test-pattern:
    - shard-tglu:         NOTRUN -> [SKIP][198] ([i915#8623])
   [198]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-4/igt@kms_tiled_display@basic-test-pattern.html

  * igt@kms_tiled_display@basic-test-pattern-with-chamelium:
    - shard-mtlp:         NOTRUN -> [SKIP][199] ([i915#8623])
   [199]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-1/igt@kms_tiled_display@basic-test-pattern-with-chamelium.html

  * igt@kms_vblank@pipe-c-query-forked-busy:
    - shard-rkl:          NOTRUN -> [SKIP][200] ([i915#4070] / [i915#6768])
   [200]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-2/igt@kms_vblank@pipe-c-query-forked-busy.html

  * igt@kms_vblank@pipe-d-wait-idle:
    - shard-rkl:          NOTRUN -> [SKIP][201] ([i915#4070] / [i915#533] / [i915#6768])
   [201]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@kms_vblank@pipe-d-wait-idle.html

  * igt@kms_writeback@writeback-pixel-formats:
    - shard-tglu:         NOTRUN -> [SKIP][202] ([i915#2437])
   [202]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-6/igt@kms_writeback@writeback-pixel-formats.html

  * igt@perf@oa-exponents@0-rcs0:
    - shard-glk:          [PASS][203] -> [ABORT][204] ([i915#5213] / [i915#7941])
   [203]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-glk8/igt@perf@oa-exponents@0-rcs0.html
   [204]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-glk6/igt@perf@oa-exponents@0-rcs0.html

  * igt@perf_pmu@event-wait@rcs0:
    - shard-mtlp:         NOTRUN -> [SKIP][205] ([i915#8807])
   [205]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-1/igt@perf_pmu@event-wait@rcs0.html

  * igt@perf_pmu@rc6@other-idle-gt0:
    - shard-dg2:          NOTRUN -> [SKIP][206] ([i915#8516])
   [206]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-8/igt@perf_pmu@rc6@other-idle-gt0.html
    - shard-tglu:         NOTRUN -> [SKIP][207] ([i915#8516])
   [207]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-2/igt@perf_pmu@rc6@other-idle-gt0.html

  * igt@perf_pmu@render-node-busy-idle@ccs0:
    - shard-mtlp:         [PASS][208] -> [FAIL][209] ([i915#4349]) +7 similar issues
   [208]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-7/igt@perf_pmu@render-node-busy-idle@ccs0.html
   [209]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-2/igt@perf_pmu@render-node-busy-idle@ccs0.html

  * igt@prime_vgem@basic-gtt:
    - shard-mtlp:         NOTRUN -> [SKIP][210] ([i915#3708] / [i915#4077])
   [210]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-3/igt@prime_vgem@basic-gtt.html

  * igt@sysfs_heartbeat_interval@nopreempt@bcs0:
    - shard-mtlp:         [PASS][211] -> [FAIL][212] ([i915#6015]) +1 similar issue
   [211]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-6/igt@sysfs_heartbeat_interval@nopreempt@bcs0.html
   [212]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-6/igt@sysfs_heartbeat_interval@nopreempt@bcs0.html

  * igt@sysfs_preempt_timeout@timeout@vecs0:
    - shard-mtlp:         [PASS][213] -> [ABORT][214] ([i915#8521])
   [213]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-5/igt@sysfs_preempt_timeout@timeout@vecs0.html
   [214]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-3/igt@sysfs_preempt_timeout@timeout@vecs0.html

  * igt@v3d/v3d_perfmon@create-perfmon-exceed:
    - shard-mtlp:         NOTRUN -> [SKIP][215] ([i915#2575]) +5 similar issues
   [215]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-2/igt@v3d/v3d_perfmon@create-perfmon-exceed.html

  * igt@v3d/v3d_submit_cl@valid-submission:
    - shard-tglu:         NOTRUN -> [SKIP][216] ([fdo#109315] / [i915#2575]) +4 similar issues
   [216]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-2/igt@v3d/v3d_submit_cl@valid-submission.html

  * igt@v3d/v3d_submit_csd@bad-multisync-in-sync:
    - shard-dg2:          NOTRUN -> [SKIP][217] ([i915#2575]) +2 similar issues
   [217]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-7/igt@v3d/v3d_submit_csd@bad-multisync-in-sync.html
    - shard-rkl:          NOTRUN -> [SKIP][218] ([fdo#109315]) +2 similar issues
   [218]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-2/igt@v3d/v3d_submit_csd@bad-multisync-in-sync.html

  * igt@vc4/vc4_mmap@mmap-bo:
    - shard-dg2:          NOTRUN -> [SKIP][219] ([i915#7711]) +3 similar issues
   [219]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-1/igt@vc4/vc4_mmap@mmap-bo.html
    - shard-rkl:          NOTRUN -> [SKIP][220] ([i915#7711]) +2 similar issues
   [220]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-2/igt@vc4/vc4_mmap@mmap-bo.html

  * igt@vc4/vc4_perfmon@create-perfmon-exceed:
    - shard-mtlp:         NOTRUN -> [SKIP][221] ([i915#7711]) +3 similar issues
   [221]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-4/igt@vc4/vc4_perfmon@create-perfmon-exceed.html

  * igt@vc4/vc4_purgeable_bo@mark-unpurgeable-purged:
    - shard-tglu:         NOTRUN -> [SKIP][222] ([i915#2575]) +4 similar issues
   [222]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-8/igt@vc4/vc4_purgeable_bo@mark-unpurgeable-purged.html

  * igt@vgem_basic@bad-pad:
    - shard-mtlp:         [PASS][223] -> [DMESG-WARN][224] ([i915#2017])
   [223]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-3/igt@vgem_basic@bad-pad.html
   [224]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-6/igt@vgem_basic@bad-pad.html

  
#### Possible fixes ####

  * igt@gem_create@hog-create@smem0:
    - shard-dg2:          [FAIL][225] ([i915#5892] / [i915#8758]) -> [PASS][226]
   [225]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg2-11/igt@gem_create@hog-create@smem0.html
   [226]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-1/igt@gem_create@hog-create@smem0.html

  * igt@gem_eio@hibernate:
    - {shard-dg1}:        [ABORT][227] ([i915#4391] / [i915#7975] / [i915#8213]) -> [PASS][228]
   [227]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg1-14/igt@gem_eio@hibernate.html
   [228]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg1-16/igt@gem_eio@hibernate.html

  * igt@gem_exec_fair@basic-pace-share@rcs0:
    - shard-glk:          [FAIL][229] ([i915#2842]) -> [PASS][230]
   [229]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-glk8/igt@gem_exec_fair@basic-pace-share@rcs0.html
   [230]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-glk1/igt@gem_exec_fair@basic-pace-share@rcs0.html

  * igt@gem_exec_fair@basic-throttle@rcs0:
    - shard-rkl:          [FAIL][231] ([i915#2842]) -> [PASS][232] +3 similar issues
   [231]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-rkl-4/igt@gem_exec_fair@basic-throttle@rcs0.html
   [232]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-2/igt@gem_exec_fair@basic-throttle@rcs0.html

  * igt@gem_exec_whisper@basic-contexts-forked-all:
    - shard-mtlp:         [ABORT][233] ([i915#8131]) -> [PASS][234]
   [233]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-1/igt@gem_exec_whisper@basic-contexts-forked-all.html
   [234]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-5/igt@gem_exec_whisper@basic-contexts-forked-all.html

  * igt@gem_exec_whisper@basic-fds-priority-all:
    - shard-mtlp:         [FAIL][235] ([i915#6363]) -> [PASS][236]
   [235]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-5/igt@gem_exec_whisper@basic-fds-priority-all.html
   [236]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-1/igt@gem_exec_whisper@basic-fds-priority-all.html

  * igt@gem_lmem_swapping@smem-oom@lmem0:
    - {shard-dg1}:        [DMESG-WARN][237] ([i915#4936] / [i915#5493]) -> [PASS][238]
   [237]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg1-13/igt@gem_lmem_swapping@smem-oom@lmem0.html
   [238]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg1-17/igt@gem_lmem_swapping@smem-oom@lmem0.html

  * igt@i915_hangman@engine-engine-hang@vcs0:
    - shard-mtlp:         [FAIL][239] ([i915#7069]) -> [PASS][240]
   [239]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-8/igt@i915_hangman@engine-engine-hang@vcs0.html
   [240]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-5/igt@i915_hangman@engine-engine-hang@vcs0.html

  * igt@i915_module_load@reload-with-fault-injection:
    - shard-mtlp:         [ABORT][241] ([i915#8489] / [i915#8668]) -> [PASS][242]
   [241]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-4/igt@i915_module_load@reload-with-fault-injection.html
   [242]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-3/igt@i915_module_load@reload-with-fault-injection.html

  * igt@i915_pm_dc@dc9-dpms:
    - shard-tglu:         [SKIP][243] ([i915#4281]) -> [PASS][244]
   [243]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-tglu-6/igt@i915_pm_dc@dc9-dpms.html
   [244]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-4/igt@i915_pm_dc@dc9-dpms.html

  * igt@i915_pm_rc6_residency@rc6-idle@rcs0:
    - {shard-dg1}:        [FAIL][245] ([i915#3591]) -> [PASS][246] +1 similar issue
   [245]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg1-19/igt@i915_pm_rc6_residency@rc6-idle@rcs0.html
   [246]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg1-18/igt@i915_pm_rc6_residency@rc6-idle@rcs0.html

  * igt@i915_pm_rpm@cursor-dpms:
    - shard-tglu:         [FAIL][247] ([i915#7940]) -> [PASS][248] +1 similar issue
   [247]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-tglu-4/igt@i915_pm_rpm@cursor-dpms.html
   [248]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-10/igt@i915_pm_rpm@cursor-dpms.html

  * igt@i915_pm_rpm@dpms-mode-unset-lpsp:
    - {shard-dg1}:        [SKIP][249] ([i915#1397]) -> [PASS][250]
   [249]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg1-15/igt@i915_pm_rpm@dpms-mode-unset-lpsp.html
   [250]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg1-19/igt@i915_pm_rpm@dpms-mode-unset-lpsp.html

  * igt@i915_pm_rpm@dpms-non-lpsp:
    - shard-rkl:          [SKIP][251] ([i915#1397]) -> [PASS][252] +1 similar issue
   [251]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-rkl-7/igt@i915_pm_rpm@dpms-non-lpsp.html
   [252]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-2/igt@i915_pm_rpm@dpms-non-lpsp.html

  * igt@i915_pm_rpm@modeset-non-lpsp-stress-no-wait:
    - shard-dg2:          [SKIP][253] ([i915#1397]) -> [PASS][254] +2 similar issues
   [253]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg2-12/igt@i915_pm_rpm@modeset-non-lpsp-stress-no-wait.html
   [254]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-7/igt@i915_pm_rpm@modeset-non-lpsp-stress-no-wait.html

  * igt@i915_selftest@live@gt_mocs:
    - shard-mtlp:         [DMESG-FAIL][255] ([i915#7059]) -> [PASS][256]
   [255]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-8/igt@i915_selftest@live@gt_mocs.html
   [256]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-7/igt@i915_selftest@live@gt_mocs.html

  * igt@i915_selftest@live@hangcheck:
    - shard-dg2:          [ABORT][257] ([i915#7913]) -> [PASS][258]
   [257]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg2-7/igt@i915_selftest@live@hangcheck.html
   [258]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-12/igt@i915_selftest@live@hangcheck.html

  * igt@i915_selftest@live@workarounds:
    - shard-mtlp:         [DMESG-FAIL][259] ([i915#6763]) -> [PASS][260]
   [259]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-8/igt@i915_selftest@live@workarounds.html
   [260]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-7/igt@i915_selftest@live@workarounds.html

  * igt@kms_big_fb@x-tiled-max-hw-stride-64bpp-rotate-180-hflip-async-flip:
    - shard-mtlp:         [FAIL][261] ([i915#3743]) -> [PASS][262]
   [261]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-1/igt@kms_big_fb@x-tiled-max-hw-stride-64bpp-rotate-180-hflip-async-flip.html
   [262]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-2/igt@kms_big_fb@x-tiled-max-hw-stride-64bpp-rotate-180-hflip-async-flip.html

  * igt@kms_cursor_legacy@2x-long-flip-vs-cursor-atomic:
    - shard-glk:          [FAIL][263] ([i915#72]) -> [PASS][264]
   [263]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-glk8/igt@kms_cursor_legacy@2x-long-flip-vs-cursor-atomic.html
   [264]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-glk2/igt@kms_cursor_legacy@2x-long-flip-vs-cursor-atomic.html

  * igt@kms_cursor_legacy@cursor-vs-flip-toggle:
    - shard-mtlp:         [FAIL][265] ([i915#8248]) -> [PASS][266]
   [265]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-7/igt@kms_cursor_legacy@cursor-vs-flip-toggle.html
   [266]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-4/igt@kms_cursor_legacy@cursor-vs-flip-toggle.html

  * igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions:
    - shard-glk:          [FAIL][267] ([i915#2346]) -> [PASS][268]
   [267]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-glk8/igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions.html
   [268]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-glk4/igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions.html

  * igt@kms_flip@flip-vs-expired-vblank-interruptible@a-hdmi-a4:
    - {shard-dg1}:        [FAIL][269] ([i915#79]) -> [PASS][270]
   [269]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg1-17/igt@kms_flip@flip-vs-expired-vblank-interruptible@a-hdmi-a4.html
   [270]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg1-17/igt@kms_flip@flip-vs-expired-vblank-interruptible@a-hdmi-a4.html

  * igt@kms_flip@flip-vs-suspend-interruptible@a-dp1:
    - shard-apl:          [ABORT][271] ([i915#180]) -> [PASS][272]
   [271]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-apl2/igt@kms_flip@flip-vs-suspend-interruptible@a-dp1.html
   [272]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-apl6/igt@kms_flip@flip-vs-suspend-interruptible@a-dp1.html

  * igt@kms_frontbuffer_tracking@fbc-1p-primscrn-indfb-plflip-blt:
    - shard-dg2:          [FAIL][273] ([i915#6880]) -> [PASS][274] +1 similar issue
   [273]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg2-11/igt@kms_frontbuffer_tracking@fbc-1p-primscrn-indfb-plflip-blt.html
   [274]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-12/igt@kms_frontbuffer_tracking@fbc-1p-primscrn-indfb-plflip-blt.html

  * igt@kms_plane_scaling@intel-max-src-size@pipe-a-hdmi-a-2:
    - shard-rkl:          [FAIL][275] ([i915#8292]) -> [PASS][276]
   [275]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-rkl-4/igt@kms_plane_scaling@intel-max-src-size@pipe-a-hdmi-a-2.html
   [276]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-4/igt@kms_plane_scaling@intel-max-src-size@pipe-a-hdmi-a-2.html

  * igt@kms_sysfs_edid_timing:
    - shard-dg2:          [FAIL][277] ([IGT#2]) -> [PASS][278]
   [277]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg2-5/igt@kms_sysfs_edid_timing.html
   [278]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-11/igt@kms_sysfs_edid_timing.html

  * igt@kms_vblank@pipe-a-ts-continuation-suspend:
    - shard-dg2:          [FAIL][279] ([fdo#103375] / [i915#6121]) -> [PASS][280]
   [279]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg2-5/igt@kms_vblank@pipe-a-ts-continuation-suspend.html
   [280]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-6/igt@kms_vblank@pipe-a-ts-continuation-suspend.html

  * igt@perf_pmu@busy-double-start@ccs0:
    - shard-mtlp:         [FAIL][281] ([i915#4349]) -> [PASS][282]
   [281]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-4/igt@perf_pmu@busy-double-start@ccs0.html
   [282]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-7/igt@perf_pmu@busy-double-start@ccs0.html

  * igt@perf_pmu@most-busy-idle-check-all@bcs0:
    - shard-mtlp:         [FAIL][283] ([i915#5234]) -> [PASS][284] +1 similar issue
   [283]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-3/igt@perf_pmu@most-busy-idle-check-all@bcs0.html
   [284]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-2/igt@perf_pmu@most-busy-idle-check-all@bcs0.html

  * igt@perf_pmu@most-busy-idle-check-all@rcs0:
    - shard-dg2:          [FAIL][285] ([i915#5234]) -> [PASS][286]
   [285]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg2-5/igt@perf_pmu@most-busy-idle-check-all@rcs0.html
   [286]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-2/igt@perf_pmu@most-busy-idle-check-all@rcs0.html
    - {shard-dg1}:        [FAIL][287] ([i915#5234]) -> [PASS][288]
   [287]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg1-13/igt@perf_pmu@most-busy-idle-check-all@rcs0.html
   [288]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg1-18/igt@perf_pmu@most-busy-idle-check-all@rcs0.html

  * igt@sysfs_timeslice_duration@timeout@vecs0:
    - shard-mtlp:         [ABORT][289] ([i915#8521]) -> [PASS][290]
   [289]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-2/igt@sysfs_timeslice_duration@timeout@vecs0.html
   [290]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-5/igt@sysfs_timeslice_duration@timeout@vecs0.html

  
#### Warnings ####

  * igt@gem_exec_suspend@basic-s0@smem:
    - shard-snb:          [DMESG-FAIL][291] ([fdo#103375]) -> [DMESG-WARN][292] ([i915#8841])
   [291]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-snb7/igt@gem_exec_suspend@basic-s0@smem.html
   [292]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-snb4/igt@gem_exec_suspend@basic-s0@smem.html

  * igt@i915_pm_rc6_residency@rc6-idle@bcs0:
    - shard-tglu:         [FAIL][293] ([i915#2681] / [i915#3591]) -> [WARN][294] ([i915#2681])
   [293]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-tglu-5/igt@i915_pm_rc6_residency@rc6-idle@bcs0.html
   [294]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-8/igt@i915_pm_rc6_residency@rc6-idle@bcs0.html

  * igt@i915_pm_rc6_residency@rc6-idle@rcs0:
    - shard-tglu:         [WARN][295] ([i915#2681]) -> [FAIL][296] ([i915#2681] / [i915#3591])
   [295]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-tglu-5/igt@i915_pm_rc6_residency@rc6-idle@rcs0.html
   [296]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-8/igt@i915_pm_rc6_residency@rc6-idle@rcs0.html

  * igt@i915_suspend@basic-s2idle-without-i915:
    - shard-snb:          [DMESG-WARN][297] ([i915#8841]) -> [ABORT][298] ([i915#4528] / [i915#8213])
   [297]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-snb7/igt@i915_suspend@basic-s2idle-without-i915.html
   [298]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-snb4/igt@i915_suspend@basic-s2idle-without-i915.html

  * igt@kms_content_protection@mei_interface:
    - shard-dg2:          [SKIP][299] ([i915#7118] / [i915#7162]) -> [SKIP][300] ([i915#7118])
   [299]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg2-11/igt@kms_content_protection@mei_interface.html
   [300]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-1/igt@kms_content_protection@mei_interface.html
    - shard-rkl:          [SKIP][301] ([fdo#109300]) -> [SKIP][302] ([i915#7118])
   [301]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-rkl-1/igt@kms_content_protection@mei_interface.html
   [302]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-6/igt@kms_content_protection@mei_interface.html
    - shard-tglu:         [SKIP][303] ([fdo#109300]) -> [SKIP][304] ([i915#6944] / [i915#7116] / [i915#7118])
   [303]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-tglu-7/igt@kms_content_protection@mei_interface.html
   [304]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-2/igt@kms_content_protection@mei_interface.html

  * igt@kms_force_connector_basic@force-load-detect:
    - shard-rkl:          [SKIP][305] ([fdo#109285] / [i915#4098]) -> [SKIP][306] ([fdo#109285])
   [305]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-rkl-4/igt@kms_force_connector_basic@force-load-detect.html
   [306]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@kms_force_connector_basic@force-load-detect.html

  * igt@prime_mmap@test_aperture_limit@test_aperture_limit-smem:
    - shard-dg2:          [CRASH][307] ([i915#7331]) -> [INCOMPLETE][308] ([i915#5493])
   [307]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg2-10/igt@prime_mmap@test_aperture_limit@test_aperture_limit-smem.html
   [308]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-5/igt@prime_mmap@test_aperture_limit@test_aperture_limit-smem.html

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

  [IGT#2]: https://gitlab.freedesktop.org/drm/igt-gpu-tools/issues/2
  [fdo#103375]: https://bugs.freedesktop.org/show_bug.cgi?id=103375
  [fdo#109271]: https://bugs.freedesktop.org/show_bug.cgi?id=109271
  [fdo#109274]: https://bugs.freedesktop.org/show_bug.cgi?id=109274
  [fdo#109279]: https://bugs.freedesktop.org/show_bug.cgi?id=109279
  [fdo#109280]: https://bugs.freedesktop.org/show_bug.cgi?id=109280
  [fdo#109285]: https://bugs.freedesktop.org/show_bug.cgi?id=109285
  [fdo#109289]: https://bugs.freedesktop.org/show_bug.cgi?id=109289
  [fdo#109293]: https://bugs.freedesktop.org/show_bug.cgi?id=109293
  [fdo#109300]: https://bugs.freedesktop.org/show_bug.cgi?id=109300
  [fdo#109314]: https://bugs.freedesktop.org/show_bug.cgi?id=109314
  [fdo#109315]: https://bugs.freedesktop.org/show_bug.cgi?id=109315
  [fdo#109642]: https://bugs.freedesktop.org/show_bug.cgi?id=109642
  [fdo#110189]: https://bugs.freedesktop.org/show_bug.cgi?id=110189
  [fdo#110723]: https://bugs.freedesktop.org/show_bug.cgi?id=110723
  [fdo#111068]: https://bugs.freedesktop.org/show_bug.cgi?id=111068
  [fdo#111614]: https://bugs.freedesktop.org/show_bug.cgi?id=111614
  [fdo#111615]: https://bugs.freedesktop.org/show_bug.cgi?id=111615
  [fdo#111767]: https://bugs.freedesktop.org/show_bug.cgi?id=111767
  [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#1072]: https://gitlab.freedesktop.org/drm/intel/issues/1072
  [i915#1257]: https://gitlab.freedesktop.org/drm/intel/issues/1257
  [i915#1397]: https://gitlab.freedesktop.org/drm/intel/issues/1397
  [i915#180]: https://gitlab.freedesktop.org/drm/intel/issues/180
  [i915#1825]: https://gitlab.freedesktop.org/drm/intel/issues/1825
  [i915#1982]: https://gitlab.freedesktop.org/drm/intel/issues/1982
  [i915#2017]: https://gitlab.freedesktop.org/drm/intel/issues/2017
  [i915#2346]: https://gitlab.freedesktop.org/drm/intel/issues/2346
  [i915#2437]: https://gitlab.freedesktop.org/drm/intel/issues/2437
  [i915#2527]: https://gitlab.freedesktop.org/drm/intel/issues/2527
  [i915#2575]: https://gitlab.freedesktop.org/drm/intel/issues/2575
  [i915#2587]: https://gitlab.freedesktop.org/drm/intel/issues/2587
  [i915#2672]: https://gitlab.freedesktop.org/drm/intel/issues/2672
  [i915#2681]: https://gitlab.freedesktop.org/drm/intel/issues/2681
  [i915#284]: https://gitlab.freedesktop.org/drm/intel/issues/284
  [i915#2842]: https://gitlab.freedesktop.org/drm/intel/issues/2842
  [i915#2856]: https://gitlab.freedesktop.org/drm/intel/issues/2856
  [i915#3023]: https://gitlab.freedesktop.org/drm/intel/issues/3023
  [i915#3281]: https://gitlab.freedesktop.org/drm/intel/issues/3281
  [i915#3282]: https://gitlab.freedesktop.org/drm/intel/issues/3282
  [i915#3297]: https://gitlab.freedesktop.org/drm/intel/issues/3297
  [i915#3359]: https://gitlab.freedesktop.org/drm/intel/issues/3359
  [i915#3458]: https://gitlab.freedesktop.org/drm/intel/issues/3458
  [i915#3469]: https://gitlab.freedesktop.org/drm/intel/issues/3469
  [i915#3539]: https://gitlab.freedesktop.org/drm/intel/issues/3539
  [i915#3546]: https://gitlab.freedesktop.org/drm/intel/issues/3546
  [i915#3555]: https://gitlab.freedesktop.org/drm/intel/issues/3555
  [i915#3591]: https://gitlab.freedesktop.org/drm/intel/issues/3591
  [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#3743]: https://gitlab.freedesktop.org/drm/intel/issues/3743
  [i915#3804]: https://gitlab.freedesktop.org/drm/intel/issues/3804
  [i915#3840]: https://gitlab.freedesktop.org/drm/intel/issues/3840
  [i915#3886]: https://gitlab.freedesktop.org/drm/intel/issues/3886
  [i915#3955]: https://gitlab.freedesktop.org/drm/intel/issues/3955
  [i915#4070]: https://gitlab.freedesktop.org/drm/intel/issues/4070
  [i915#4077]: https://gitlab.freedesktop.org/drm/intel/issues/4077
  [i915#4078]: https://gitlab.freedesktop.org/drm/intel/issues/4078
  [i915#4079]: https://gitlab.freedesktop.org/drm/intel/issues/4079
  [i915#4083]: https://gitlab.freedesktop.org/drm/intel/issues/4083
  [i915#4098]: https://gitlab.freedesktop.org/drm/intel/issues/4098
  [i915#4212]: https://gitlab.freedesktop.org/drm/intel/issues/4212
  [i915#4215]: https://gitlab.freedesktop.org/drm/intel/issues/4215
  [i915#4235]: https://gitlab.freedesktop.org/drm/intel/issues/4235
  [i915#4270]: https://gitlab.freedesktop.org/drm/intel/issues/4270
  [i915#4281]: https://gitlab.freedesktop.org/drm/intel/issues/4281
  [i915#4348]: https://gitlab.freedesktop.org/drm/intel/issues/4348
  [i915#4349]: https://gitlab.freedesktop.org/drm/intel/issues/4349
  [i915#4391]: https://gitlab.freedesktop.org/drm/intel/issues/4391
  [i915#4528]: https://gitlab.freedesktop.org/drm/intel/issues/4528
  [i915#4538]: https://gitlab.freedesktop.org/drm/intel/issues/4538
  [i915#4613]: https://gitlab.freedesktop.org/drm/intel/issues/4613
  [i915#4812]: https://gitlab.freedesktop.org/drm/intel/issues/4812
  [i915#4880]: https://gitlab.freedesktop.org/drm/intel/issues/4880
  [i915#4936]: https://gitlab.freedesktop.org/drm/intel/issues/4936
  [i915#5138]: https://gitlab.freedesktop.org/drm/intel/issues/5138
  [i915#5176]: https://gitlab.freedesktop.org/drm/intel/issues/5176
  [i915#5190]: https://gitlab.freedesktop.org/drm/intel/issues/5190
  [i915#5213]: https://gitlab.freedesktop.org/drm/intel/issues/5213
  [i915#5234]: https://gitlab.freedesktop.org/drm/intel/issues/5234
  [i915#5235]: https://gitlab.freedesktop.org/drm/intel/issues/5235
  [i915#5274]: https://gitlab.freedesktop.org/drm/intel/issues/5274
  [i915#5286]: https://gitlab.freedesktop.org/drm/intel/issues/5286
  [i915#5325]: https://gitlab.freedesktop.org/drm/intel/issues/5325
  [i915#533]: https://gitlab.freedesktop.org/drm/intel/issues/533
  [i915#5354]: https://gitlab.freedesktop.org/drm/intel/issues/5354
  [i915#5439]: https://gitlab.freedesktop.org/drm/intel/issues/5439
  [i915#5493]: https://gitlab.freedesktop.org/drm/intel/issues/5493
  [i915#5784]: https://gitlab.freedesktop.org/drm/intel/issues/5784
  [i915#5892]: https://gitlab.freedesktop.org/drm/intel/issues/5892
  [i915#6015]: https://gitlab.freedesktop.org/drm/intel/issues/6015
  [i915#6095]: https://gitlab.freedesktop.org/drm/intel/issues/6095
  [i915#6121]: https://gitlab.freedesktop.org/drm/intel/issues/6121
  [i915#6187]: https://gitlab.freedesktop.org/drm/intel/issues/6187
  [i915#6227]: https://gitlab.freedesktop.org/drm/intel/issues/6227
  [i915#6230]: https://gitlab.freedesktop.org/drm/intel/issues/6230
  [i915#6268]: https://gitlab.freedesktop.org/drm/intel/issues/6268
  [i915#6335]: https://gitlab.freedesktop.org/drm/intel/issues/6335
  [i915#6363]: https://gitlab.freedesktop.org/drm/intel/issues/6363
  [i915#658]: https://gitlab.freedesktop.org/drm/intel/issues/658
  [i915#6763]: https://gitlab.freedesktop.org/drm/intel/issues/6763
  [i915#6768]: https://gitlab.freedesktop.org/drm/intel/issues/6768
  [i915#6880]: https://gitlab.freedesktop.org/drm/intel/issues/6880
  [i915#6944]: https://gitlab.freedesktop.org/drm/intel/issues/6944
  [i915#6953]: https://gitlab.freedesktop.org/drm/intel/issues/6953
  [i915#7059]: https://gitlab.freedesktop.org/drm/intel/issues/7059
  [i915#7069]: https://gitlab.freedesktop.org/drm/intel/issues/7069
  [i915#7116]: https://gitlab.freedesktop.org/drm/intel/issues/7116
  [i915#7118]: https://gitlab.freedesktop.org/drm/intel/issues/7118
  [i915#7162]: https://gitlab.freedesktop.org/drm/intel/issues/7162
  [i915#72]: https://gitlab.freedesktop.org/drm/intel/issues/72
  [i915#7331]: https://gitlab.freedesktop.org/drm/intel/issues/7331
  [i915#7443]: https://gitlab.freedesktop.org/drm/intel/issues/7443
  [i915#7561]: https://gitlab.freedesktop.org/drm/intel/issues/7561
  [i915#7697]: https://gitlab.freedesktop.org/drm/intel/issues/7697
  [i915#7701]: https://gitlab.freedesktop.org/drm/intel/issues/7701
  [i915#7711]: https://gitlab.freedesktop.org/drm/intel/issues/7711
  [i915#7742]: https://gitlab.freedesktop.org/drm/intel/issues/7742
  [i915#7816]: https://gitlab.freedesktop.org/drm/intel/issues/7816
  [i915#7828]: https://gitlab.freedesktop.org/drm/intel/issues/7828
  [i915#7892]: https://gitlab.freedesktop.org/drm/intel/issues/7892
  [i915#79]: https://gitlab.freedesktop.org/drm/intel/issues/79
  [i915#7913]: https://gitlab.freedesktop.org/drm/intel/issues/7913
  [i915#7916]: https://gitlab.freedesktop.org/drm/intel/issues/7916
  [i915#7940]: https://gitlab.freedesktop.org/drm/intel/issues/7940
  [i915#7941]: https://gitlab.freedesktop.org/drm/intel/issues/7941
  [i915#7975]: https://gitlab.freedesktop.org/drm/intel/issues/7975
  [i915#8063]: https://gitlab.freedesktop.org/drm/intel/issues/8063
  [i915#8102]: https://gitlab.freedesktop.org/drm/intel/issues/8102
  [i915#8131]: https://gitlab.freedesktop.org/drm/intel/issues/8131
  [i915#8213]: https://gitlab.freedesktop.org/drm/intel/issues/8213
  [i915#8228]: https://gitlab.freedesktop.org/drm/intel/issues/8228
  [i915#8247]: https://gitlab.freedesktop.org/drm/intel/issues/8247
  [i915#8248]: https://gitlab.freedesktop.org/drm/intel/issues/8248
  [i915#8292]: https://gitlab.freedesktop.org/drm/intel/issues/8292
  [i915#8381]: https://gitlab.freedesktop.org/drm/intel/issues/8381
  [i915#8414]: https://gitlab.freedesktop.org/drm/intel/issues/8414
  [i915#8428]: https://gitlab.freedesktop.org/drm/intel/issues/8428
  [i915#8430]: https://gitlab.freedesktop.org/drm/intel/issues/8430
  [i915#8489]: https://gitlab.freedesktop.org/drm/intel/issues/8489
  [i915#8502]: https://gitlab.freedesktop.org/drm/intel/issues/8502
  [i915#8503]: https://gitlab.freedesktop.org/drm/intel/issues/8503
  [i915#8516]: https://gitlab.freedesktop.org/drm/intel/issues/8516
  [i915#8521]: https://gitlab.freedesktop.org/drm/intel/issues/8521
  [i915#8623]: https://gitlab.freedesktop.org/drm/intel/issues/8623
  [i915#8661]: https://gitlab.freedesktop.org/drm/intel/issues/8661
  [i915#8668]: https://gitlab.freedesktop.org/drm/intel/issues/8668
  [i915#8708]: https://gitlab.freedesktop.org/drm/intel/issues/8708
  [i915#8709]: https://gitlab.freedesktop.org/drm/intel/issues/8709
  [i915#8726]: https://gitlab.freedesktop.org/drm/intel/issues/8726
  [i915#8758]: https://gitlab.freedesktop.org/drm/intel/issues/8758
  [i915#8807]: https://gitlab.freedesktop.org/drm/intel/issues/8807
  [i915#8810]: https://gitlab.freedesktop.org/drm/intel/issues/8810
  [i915#8814]: https://gitlab.freedesktop.org/drm/intel/issues/8814
  [i915#8841]: https://gitlab.freedesktop.org/drm/intel/issues/8841


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

  * CI: CI-20190529 -> None
  * IGT: IGT_7380 -> IGTPW_9383
  * Piglit: piglit_4509 -> None

  CI-20190529: 20190529
  CI_DRM_13369: 0595153b5c237d83f9990f3ce44d072a262dcb0e @ git://anongit.freedesktop.org/gfx-ci/linux
  IGTPW_9383: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/index.html
  IGT_7380: 8e65f12de2fd52c05dc48fdbcb8cfe86f6de1a75 @ 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_9383/index.html

[-- Attachment #2: Type: text/html, Size: 94490 bytes --]

Re: [igt-dev] [PATCH i-g-t v3 02/17] lib/intel_allocator: Drop aliasing allocator handle api

[Karolina Stolarek]

On 11.07.2023 13:19, Zbigniew Kempczyński wrote:
> There's no real user of this api, lets drop it.
> 
> Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
> --
> v3: Remove also ipc message passing for open-as (Karolina)
> ---
>   lib/intel_allocator.c            | 73 --------------------------------
>   lib/intel_allocator.h            |  1 -
>   lib/intel_allocator_msgchannel.h |  8 +---
>   3 files changed, 1 insertion(+), 81 deletions(-)
> 

Reviewed-by: Karolina Stolarek <karolina.stolarek@intel.com>

> diff --git a/lib/intel_allocator.c b/lib/intel_allocator.c
> index 8161221dbf..09f665fbb9 100644
> --- a/lib/intel_allocator.c
> +++ b/lib/intel_allocator.c
> @@ -25,7 +25,6 @@
>   static const char *reqtype_str[] = {
>   	[REQ_STOP]		= "stop",
>   	[REQ_OPEN]		= "open",
> -	[REQ_OPEN_AS]		= "open as",
>   	[REQ_CLOSE]		= "close",
>   	[REQ_ADDRESS_RANGE]	= "address range",
>   	[REQ_ALLOC]		= "alloc",
> @@ -376,19 +375,6 @@ static struct allocator *allocator_open(int fd, uint32_t ctx, uint32_t vm,
>   	return al;
>   }
>   
> -static struct allocator *allocator_open_as(struct allocator *base,
> -					   uint32_t new_vm, uint64_t *ahndp)
> -{
> -	struct allocator *al;
> -
> -	igt_assert(ahndp);
> -	al = __allocator_create(base->fd, base->ctx, new_vm, base->ial);
> -	__allocator_get(al);
> -	*ahndp = __handle_create(al);
> -
> -	return al;
> -}
> -
>   static bool allocator_close(uint64_t ahnd)
>   {
>   	struct allocator *al;
> @@ -512,47 +498,6 @@ static int handle_request(struct alloc_req *req, struct alloc_resp *resp)
>   				   refcnt - 1, refcnt, ret - 1, ret);
>   			break;
>   
> -		case REQ_OPEN_AS:
> -			/* lock first to avoid concurrent close */
> -			pthread_mutex_lock(&map_mutex);
> -
> -			al = __allocator_find_by_handle(req->allocator_handle);
> -			resp->response_type = RESP_OPEN_AS;
> -
> -			if (!al) {
> -				alloc_info("<open as> [tid: %ld] ahnd: %" PRIx64
> -					   " -> no handle\n",
> -					   (long) req->tid, req->allocator_handle);
> -				pthread_mutex_unlock(&map_mutex);
> -				break;
> -			}
> -
> -			if (!al->vm) {
> -				alloc_info("<open as> [tid: %ld] ahnd: %" PRIx64
> -					   " -> only open as for <fd, vm> is possible\n",
> -					   (long) req->tid, req->allocator_handle);
> -				pthread_mutex_unlock(&map_mutex);
> -				break;
> -			}
> -
> -
> -			al = allocator_open_as(al, req->open_as.new_vm, &ahnd);
> -			refcnt = atomic_load(&al->refcount);
> -			ret = atomic_load(&al->ial->refcount);
> -			pthread_mutex_unlock(&map_mutex);
> -
> -			resp->response_type = RESP_OPEN_AS;
> -			resp->open.allocator_handle = ahnd;
> -
> -			alloc_info("<open as> [tid: %ld] fd: %d, ahnd: %" PRIx64
> -				   ", ctx: %u, vm: %u"
> -				   ", alloc_type: %u, al->refcnt: %ld->%ld"
> -				   ", refcnt: %d->%d\n",
> -				   (long) req->tid, al->fd, ahnd,
> -				   al->ctx, al->vm, al->ial->type,
> -				   refcnt - 1, refcnt, ret - 1, ret);
> -			break;
> -
>   		case REQ_CLOSE:
>   			pthread_mutex_lock(&map_mutex);
>   			al = __allocator_find_by_handle(req->allocator_handle);
> @@ -1037,24 +982,6 @@ uint64_t intel_allocator_open_vm(int fd, uint32_t vm, uint8_t allocator_type)
>   					    ALLOC_STRATEGY_HIGH_TO_LOW, 0);
>   }
>   
> -uint64_t intel_allocator_open_vm_as(uint64_t allocator_handle, uint32_t new_vm)
> -{
> -	struct alloc_req req = { .request_type = REQ_OPEN_AS,
> -				 .allocator_handle = allocator_handle,
> -				 .open_as.new_vm = new_vm };
> -	struct alloc_resp resp;
> -
> -	/* Get child_tid only once at open() */
> -	if (child_tid == -1)
> -		child_tid = gettid();
> -
> -	igt_assert(handle_request(&req, &resp) == 0);
> -	igt_assert(resp.open_as.allocator_handle);
> -	igt_assert(resp.response_type == RESP_OPEN_AS);
> -
> -	return resp.open.allocator_handle;
> -}
> -
>   /**
>    * intel_allocator_close:
>    * @allocator_handle: handle to the allocator that will be closed
> diff --git a/lib/intel_allocator.h b/lib/intel_allocator.h
> index a6bf573e9d..3ec74f6191 100644
> --- a/lib/intel_allocator.h
> +++ b/lib/intel_allocator.h
> @@ -182,7 +182,6 @@ uint64_t intel_allocator_open_vm_full(int fd, uint32_t vm,
>   				      enum allocator_strategy strategy,
>   				      uint64_t default_alignment);
>   
> -uint64_t intel_allocator_open_vm_as(uint64_t allocator_handle, uint32_t new_vm);
>   bool intel_allocator_close(uint64_t allocator_handle);
>   void intel_allocator_get_address_range(uint64_t allocator_handle,
>   				       uint64_t *startp, uint64_t *endp);
> diff --git a/lib/intel_allocator_msgchannel.h b/lib/intel_allocator_msgchannel.h
> index cefbf4ce33..ba38530fd1 100644
> --- a/lib/intel_allocator_msgchannel.h
> +++ b/lib/intel_allocator_msgchannel.h
> @@ -13,7 +13,6 @@
>   enum reqtype {
>   	REQ_STOP,
>   	REQ_OPEN,
> -	REQ_OPEN_AS,
>   	REQ_CLOSE,
>   	REQ_ADDRESS_RANGE,
>   	REQ_ALLOC,
> @@ -27,7 +26,6 @@ enum reqtype {
>   
>   enum resptype {
>   	RESP_OPEN,
> -	RESP_OPEN_AS,
>   	RESP_CLOSE,
>   	RESP_ADDRESS_RANGE,
>   	RESP_ALLOC,
> @@ -58,10 +56,6 @@ struct alloc_req {
>   			uint64_t default_alignment;
>   		} open;
>   
> -		struct {
> -			uint32_t new_vm;
> -		} open_as;
> -
>   		struct {
>   			uint32_t handle;
>   			uint64_t size;
> @@ -100,7 +94,7 @@ struct alloc_resp {
>   	union {
>   		struct {
>   			uint64_t allocator_handle;
> -		} open, open_as;
> +		} open;
>   
>   		struct {
>   			bool is_empty;

Re: [igt-dev] [PATCH i-g-t v3 04/17] lib/xe_ioctl: Export non-assert xe_exec function

[Karolina Stolarek]

On 11.07.2023 13:19, Zbigniew Kempczyński wrote:
> Export __xe_exec() helper to allow caller to handle the error
> individually.
> 
> Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
> ---
>   lib/xe/xe_ioctl.c | 2 +-
>   lib/xe/xe_ioctl.h | 1 +
>   2 files changed, 2 insertions(+), 1 deletion(-)

That change is fine by me, but if you're really keen, you can check it 
with others.

But for now:
Reviewed-by: Karolina Stolarek <karolina.stolarek@intel.com>

> 
> diff --git a/lib/xe/xe_ioctl.c b/lib/xe/xe_ioctl.c
> index 9ead7307c4..0b5c7a4fce 100644
> --- a/lib/xe/xe_ioctl.c
> +++ b/lib/xe/xe_ioctl.c
> @@ -357,7 +357,7 @@ void *xe_bo_mmap_ext(int fd, uint32_t bo, size_t size, int prot)
>   	return __xe_bo_map(fd, bo, size, prot);
>   }
>   
> -static int __xe_exec(int fd, struct drm_xe_exec *exec)
> +int __xe_exec(int fd, struct drm_xe_exec *exec)
>   {
>   	int err = 0;
>   
> diff --git a/lib/xe/xe_ioctl.h b/lib/xe/xe_ioctl.h
> index e014ad66c9..320e0f9f68 100644
> --- a/lib/xe/xe_ioctl.h
> +++ b/lib/xe/xe_ioctl.h
> @@ -75,6 +75,7 @@ void xe_engine_destroy(int fd, uint32_t engine);
>   uint64_t xe_bo_mmap_offset(int fd, uint32_t bo);
>   void *xe_bo_map(int fd, uint32_t bo, size_t size);
>   void *xe_bo_mmap_ext(int fd, uint32_t bo, size_t size, int prot);
> +int __xe_exec(int fd, struct drm_xe_exec *exec);
>   void xe_exec(int fd, struct drm_xe_exec *exec);
>   void xe_exec_sync(int fd, uint32_t engine, uint64_t addr,
>   		  struct drm_xe_sync *sync, uint32_t num_syncs);

Re: [igt-dev] [PATCH i-g-t v3 09/17] lib/xe_util: Add vm bind/unbind helper for Xe

[Karolina Stolarek]

On 11.07.2023 13:20, Zbigniew Kempczyński wrote:
> Before calling exec we need to prepare vm to contain valid entries.
> Bind/unbind in xe expects single bind_op or vector of bind_ops what
> makes preparation of it a little bit inconvinient. Add function
> which iterates over list of xe_object (auxiliary structure which
> describes bind information for object) and performs the bind/unbind
> in one step. It also supports passing syncobj in/out to work in
> pipelined executions.
> 
> Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>

I'll see how the changes are used in 10/17, but the helpers themself 
look good to me:

Reviewed-by: Karolina Stolarek <karolina.stolarek@intel.com>

> --
> v3: - Drop priv field in the structure
> ---
>   lib/xe/xe_util.c | 125 +++++++++++++++++++++++++++++++++++++++++++++++
>   lib/xe/xe_util.h |  17 +++++++
>   2 files changed, 142 insertions(+)
> 
> diff --git a/lib/xe/xe_util.c b/lib/xe/xe_util.c
> index 5f0f69a8c6..7d78065a13 100644
> --- a/lib/xe/xe_util.c
> +++ b/lib/xe/xe_util.c
> @@ -102,3 +102,128 @@ char *xe_memregion_dynamic_subtest_name(int xe, struct igt_collection *set)
>   	return name;
>   }
>   
> +#ifdef XEBINDDBG
> +#define bind_info igt_info
> +#define bind_debug igt_debug
> +#else
> +#define bind_info(...) {}
> +#define bind_debug(...) {}
> +#endif
> +
> +static struct drm_xe_vm_bind_op *xe_alloc_bind_ops(struct igt_list_head *obj_list,
> +						   uint32_t *num_ops)
> +{
> +	struct drm_xe_vm_bind_op *bind_ops, *ops;
> +	struct xe_object *obj;
> +	uint32_t num_objects = 0, i = 0, op;
> +
> +	igt_list_for_each_entry(obj, obj_list, link)
> +		num_objects++;
> +
> +	*num_ops = num_objects;
> +	if (!num_objects) {
> +		bind_info(" [nothing to bind]\n");
> +		return NULL;
> +	}
> +
> +	bind_ops = calloc(num_objects, sizeof(*bind_ops));
> +	igt_assert(bind_ops);
> +
> +	igt_list_for_each_entry(obj, obj_list, link) {
> +		ops = &bind_ops[i];
> +
> +		if (obj->bind_op == XE_OBJECT_BIND) {
> +			op = XE_VM_BIND_OP_MAP | XE_VM_BIND_FLAG_ASYNC;
> +			ops->obj = obj->handle;
> +		} else {
> +			op = XE_VM_BIND_OP_UNMAP | XE_VM_BIND_FLAG_ASYNC;
> +		}
> +
> +		ops->op = op;
> +		ops->obj_offset = 0;
> +		ops->addr = obj->offset;
> +		ops->range = obj->size;
> +		ops->region = 0;
> +
> +		bind_info("  [%d]: [%6s] handle: %u, offset: %llx, size: %llx\n",
> +			  i, obj->bind_op == XE_OBJECT_BIND ? "BIND" : "UNBIND",
> +			  ops->obj, (long long)ops->addr, (long long)ops->range);
> +		i++;
> +	}
> +
> +	return bind_ops;
> +}
> +
> +/**
> + * xe_bind_unbind_async:
> + * @xe: drm fd of Xe device
> + * @vm: vm to bind/unbind objects to/from
> + * @bind_engine: bind engine, 0 if default
> + * @obj_list: list of xe_object
> + * @sync_in: sync object (fence-in), 0 if there's no input dependency
> + * @sync_out: sync object (fence-out) to signal on bind/unbind completion,
> + *            if 0 wait for bind/unbind completion.
> + *
> + * Function iterates over xe_object @obj_list, prepares binding operation
> + * and does bind/unbind in one step. Providing sync_in / sync_out allows
> + * working in pipelined mode. With sync_in and sync_out set to 0 function
> + * waits until binding operation is complete.
> + */
> +void xe_bind_unbind_async(int xe, uint32_t vm, uint32_t bind_engine,
> +			  struct igt_list_head *obj_list,
> +			  uint32_t sync_in, uint32_t sync_out)
> +{
> +	struct drm_xe_vm_bind_op *bind_ops;
> +	struct drm_xe_sync tabsyncs[2] = {
> +		{ .flags = DRM_XE_SYNC_SYNCOBJ, .handle = sync_in },
> +		{ .flags = DRM_XE_SYNC_SYNCOBJ | DRM_XE_SYNC_SIGNAL, .handle = sync_out },
> +	};
> +	struct drm_xe_sync *syncs;
> +	uint32_t num_binds = 0;
> +	int num_syncs;
> +
> +	bind_info("[Binding to vm: %u]\n", vm);
> +	bind_ops = xe_alloc_bind_ops(obj_list, &num_binds);
> +
> +	if (!num_binds) {
> +		if (sync_out)
> +			syncobj_signal(xe, &sync_out, 1);
> +		return;
> +	}
> +
> +	if (sync_in) {
> +		syncs = tabsyncs;
> +		num_syncs = 2;
> +	} else {
> +		syncs = &tabsyncs[1];
> +		num_syncs = 1;
> +	}
> +
> +	/* User didn't pass sync out, create it and wait for completion */
> +	if (!sync_out)
> +		tabsyncs[1].handle = syncobj_create(xe, 0);
> +
> +	bind_info("[Binding syncobjs: (in: %u, out: %u)]\n",
> +		  tabsyncs[0].handle, tabsyncs[1].handle);
> +
> +	if (num_binds == 1) {
> +		if ((bind_ops[0].op & 0xffff) == XE_VM_BIND_OP_MAP)
> +			xe_vm_bind_async(xe, vm, bind_engine, bind_ops[0].obj, 0,
> +					bind_ops[0].addr, bind_ops[0].range,
> +					syncs, num_syncs);
> +		else
> +			xe_vm_unbind_async(xe, vm, bind_engine, 0,
> +					   bind_ops[0].addr, bind_ops[0].range,
> +					   syncs, num_syncs);
> +	} else {
> +		xe_vm_bind_array(xe, vm, bind_engine, bind_ops,
> +				 num_binds, syncs, num_syncs);
> +	}
> +
> +	if (!sync_out) {
> +		igt_assert_eq(syncobj_wait_err(xe, &tabsyncs[1].handle, 1, INT64_MAX, 0), 0);
> +		syncobj_destroy(xe, tabsyncs[1].handle);
> +	}
> +
> +	free(bind_ops);
> +}
> diff --git a/lib/xe/xe_util.h b/lib/xe/xe_util.h
> index 9f56fa9898..61511dd97f 100644
> --- a/lib/xe/xe_util.h
> +++ b/lib/xe/xe_util.h
> @@ -27,4 +27,21 @@ __xe_get_memory_region_set(int xe, uint32_t *mem_regions_type, int num_regions);
>   
>   char *xe_memregion_dynamic_subtest_name(int xe, struct igt_collection *set);
>   
> +enum xe_bind_op {
> +	XE_OBJECT_BIND,
> +	XE_OBJECT_UNBIND,
> +};
> +
> +struct xe_object {
> +	uint32_t handle;
> +	uint64_t offset;
> +	uint64_t size;
> +	enum xe_bind_op bind_op;
> +	struct igt_list_head link;
> +};
> +
> +void xe_bind_unbind_async(int fd, uint32_t vm, uint32_t bind_engine,
> +			  struct igt_list_head *obj_list,
> +			  uint32_t sync_in, uint32_t sync_out);
> +
>   #endif /* XE_UTIL_H */

Re: [igt-dev] ✗ Fi.CI.IGT: failure for Extend intel_blt to work on Xe (rev3)

[Zbigniew Kempczyński]

On Tue, Jul 11, 2023 at 02:52:12PM +0000, Patchwork wrote:
>    Patch Details
> 
>    Series:  Extend intel_blt to work on Xe (rev3)                          
>    URL:     https://patchwork.freedesktop.org/series/120162/               
>    State:   failure                                                        
>    Details: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/index.html 
> 
>          CI Bug Log - changes from CI_DRM_13369_full -> IGTPW_9383_full
> 
> Summary
> 
>    FAILURE
> 
>    Serious unknown changes coming with IGTPW_9383_full absolutely need to be
>    verified manually.
> 
>    If you think the reported changes have nothing to do with the changes
>    introduced in IGTPW_9383_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_9383/index.html
> 
> Participating hosts (9 -> 9)
> 
>    No changes in participating hosts
> 
> Possible new issues
> 
>    Here are the unknown changes that may have been introduced in
>    IGTPW_9383_full:
> 
>   IGT changes
> 
>     Possible regressions
> 
>      * igt@prime_vgem@fence-wait@ccs0:
> 
>           * shard-mtlp: PASS -> DMESG-WARN
>      * igt@prime_vgem@fence-wait@vecs0:
> 
>           * shard-mtlp: PASS -> ABORT
>      * igt@vgem_basic@sysfs:
> 
>           * shard-snb: PASS -> ABORT

Unrelated to the change.

--
Zbigniew

> 
> Known issues
> 
>    Here are the changes found in IGTPW_9383_full that come from known issues:
> 
>   IGT changes
> 
>     Issues hit
> 
>      * igt@api_intel_bb@crc32:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#6230)
>      * igt@device_reset@unbind-cold-reset-rebind:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#7701)
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#7701)
> 
>      * igt@drm_fdinfo@busy-hang@bcs0:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#8414) +19 similar issues
>      * igt@drm_fdinfo@most-busy-idle-check-all@rcs0:
> 
>           * shard-rkl: PASS -> FAIL (i915#7742)
>      * igt@drm_fdinfo@virtual-busy-all:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#8414)
>      * igt@gem_bad_reloc@negative-reloc-bltcopy:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#3281) +5 similar issues
>      * igt@gem_basic@multigpu-create-close:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#7697) +1 similar issue
>      * igt@gem_ccs@block-copy-compressed:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#3555 / i915#5325)
>      * igt@gem_ccs@ctrl-surf-copy-new-ctx:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#5325)
>      * igt@gem_close_race@multigpu-basic-process:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#7697) +1 similar issue
>      * igt@gem_create@create-ext-cpu-access-big:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#6335)
>      * igt@gem_ctx_exec@basic-nohangcheck:
> 
>           * shard-tglu: PASS -> FAIL (i915#6268)
>      * igt@gem_ctx_param@set-priority-not-supported:
> 
>           * shard-dg2: NOTRUN -> SKIP (fdo#109314)
> 
>           * shard-rkl: NOTRUN -> SKIP (fdo#109314)
> 
>      * igt@gem_ctx_persistence@saturated-hostile@vecs0:
> 
>           * shard-mtlp: PASS -> FAIL (i915#7816) +2 similar issues
>      * igt@gem_eio@in-flight-contexts-1us:
> 
>           * shard-mtlp: PASS -> ABORT (i915#8503)
>      * igt@gem_eio@kms:
> 
>           * shard-dg2: PASS -> INCOMPLETE (i915#7892)
>      * igt@gem_exec_balancer@bonded-true-hang:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#4812) +2 similar issues
>      * igt@gem_exec_fair@basic-none@bcs0:
> 
>           * shard-tglu: NOTRUN -> FAIL (i915#2842) +4 similar issues
>      * igt@gem_exec_fair@basic-pace@vcs0:
> 
>           * shard-glk: PASS -> FAIL (i915#2842) +1 similar issue
>      * igt@gem_exec_fair@basic-throttle:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#3539)
>      * igt@gem_exec_fence@submit67:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#4812)
>      * igt@gem_exec_params@secure-non-root:
> 
>           * shard-tglu: NOTRUN -> SKIP (fdo#112283)
>      * igt@gem_exec_reloc@basic-write-gtt-active:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#3281) +2 similar issues
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#3281) +1 similar issue
> 
>      * igt@gem_exec_whisper@basic-forked:
> 
>           * shard-mtlp: NOTRUN -> FAIL (i915#6363)
>      * igt@gem_exec_whisper@basic-forked-all:
> 
>           * shard-mtlp: PASS -> FAIL (i915#6363)
>      * igt@gem_lmem_swapping@heavy-verify-random-ccs:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#4613) +2 similar issues
>      * igt@gem_lmem_swapping@parallel-multi:
> 
>           * shard-apl: NOTRUN -> SKIP (fdo#109271 / i915#4613)
>      * igt@gem_lmem_swapping@parallel-random:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#4613)
>      * igt@gem_lmem_swapping@smem-oom@lmem0:
> 
>           * shard-dg2: PASS -> TIMEOUT (i915#5493)
>      * igt@gem_media_vme:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#284)
>      * igt@gem_mmap_wc@write:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#4083) +1 similar issue
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#4083) +3 similar issues
> 
>      * igt@gem_partial_pwrite_pread@reads:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#3282) +3 similar issues
>      * igt@gem_partial_pwrite_pread@writes-after-reads:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#3282) +2 similar issues
>      * igt@gem_pxp@create-regular-buffer:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#4270)
>      * igt@gem_pxp@fail-invalid-protected-context:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#4270) +1 similar issue
>      * igt@gem_readwrite@beyond-eob:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#3282) +3 similar issues
>      * igt@gem_render_copy@y-tiled-ccs-to-yf-tiled-ccs:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#8428) +3 similar issues
>      * igt@gem_set_tiling_vs_pwrite:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#4079)
>      * igt@gem_tiled_fence_blits@basic:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#4077) +3 similar issues
>      * igt@gem_tiled_pread_basic:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#4079)
>      * igt@gem_userptr_blits@coherency-unsync:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#3297)
>      * igt@gem_userptr_blits@invalid-mmap-offset-unsync:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#3297)
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#3297)
> 
>      * igt@gem_userptr_blits@map-fixed-invalidate-overlap-busy:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#3297 / i915#4880)
>      * igt@gem_userptr_blits@nohangcheck:
> 
>           * shard-mtlp: PASS -> FAIL (i915#7916)
>      * igt@gen7_exec_parse@load-register-reg:
> 
>           * shard-dg2: NOTRUN -> SKIP (fdo#109289)
>      * igt@gen7_exec_parse@oacontrol-tracking:
> 
>           * shard-mtlp: NOTRUN -> SKIP (fdo#109289)
>      * igt@gen9_exec_parse@bb-large:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#2856)
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#2527)
> 
>      * igt@i915_hangman@gt-engine-error@vcs0:
> 
>           * shard-mtlp: PASS -> FAIL (i915#7069)
>      * igt@i915_module_load@load:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#6227)
>      * igt@i915_pm_backlight@bad-brightness:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#5354 / i915#7561)
>      * igt@i915_pm_dc@dc5-psr:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#658)
>      * igt@i915_pm_dc@dc9-dpms:
> 
>           * shard-apl: PASS -> SKIP (fdo#109271)
>      * igt@i915_pm_lpsp@screens-disabled:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#8430)
>      * igt@i915_pm_rpm@modeset-non-lpsp-stress:
> 
>           * shard-dg2: PASS -> SKIP (i915#1397) +1 similar issue
>      * igt@i915_pm_rpm@pc8-residency:
> 
>           * shard-mtlp: NOTRUN -> SKIP (fdo#109293)
>      * igt@i915_pm_rpm@system-suspend-execbuf:
> 
>           * shard-tglu: NOTRUN -> FAIL (i915#7940)
>      * igt@i915_suspend@basic-s3-without-i915:
> 
>           * shard-dg2: PASS -> FAIL (fdo#103375 / i915#6121) +2 similar
>             issues
> 
>           * shard-tglu: NOTRUN -> INCOMPLETE (i915#7443 / i915#8102)
> 
>      * igt@i915_suspend@debugfs-reader:
> 
>           * shard-apl: PASS -> ABORT (i915#180 / i915#8213)
>      * igt@i915_suspend@fence-restore-tiled2untiled:
> 
>           * shard-snb: NOTRUN -> DMESG-WARN (i915#8841) +4 similar issues
>      * igt@kms_addfb_basic@addfb25-x-tiled-legacy:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#4212)
>      * igt@kms_addfb_basic@basic-y-tiled-legacy:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#4215 / i915#5190)
>      * igt@kms_async_flips@crc@pipe-a-hdmi-a-3:
> 
>           * shard-dg2: NOTRUN -> FAIL (i915#8247) +3 similar issues
>      * igt@kms_big_fb@4-tiled-16bpp-rotate-90:
> 
>           * shard-mtlp: NOTRUN -> SKIP (fdo#111614)
>      * igt@kms_big_fb@4-tiled-8bpp-rotate-270:
> 
>           * shard-dg2: NOTRUN -> SKIP (fdo#111614) +1 similar issue
>      * igt@kms_big_fb@4-tiled-max-hw-stride-32bpp-rotate-180-async-flip:
> 
>           * shard-mtlp: PASS -> FAIL (i915#3743)
>      * igt@kms_big_fb@4-tiled-max-hw-stride-32bpp-rotate-180-hflip:
> 
>           * shard-tglu: NOTRUN -> SKIP (fdo#111615 / i915#5286)
>      * igt@kms_big_fb@4-tiled-max-hw-stride-64bpp-rotate-0-hflip:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#5286) +1 similar issue
> 
>           * shard-mtlp: PASS -> FAIL (i915#5138)
> 
>      * igt@kms_big_fb@x-tiled-32bpp-rotate-90:
> 
>           * shard-rkl: NOTRUN -> SKIP (fdo#111614 / i915#3638)
>      * igt@kms_big_fb@x-tiled-max-hw-stride-64bpp-rotate-0-hflip-async-flip:
> 
>           * shard-mtlp: NOTRUN -> FAIL (i915#3743)
>      * igt@kms_big_fb@y-tiled-8bpp-rotate-90:
> 
>           * shard-tglu: NOTRUN -> SKIP (fdo#111614)
>      * igt@kms_big_fb@yf-tiled-64bpp-rotate-90:
> 
>           * shard-rkl: NOTRUN -> SKIP (fdo#110723) +2 similar issues
>      * igt@kms_big_fb@yf-tiled-addfb-size-overflow:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#5190) +5 similar issues
> 
>           * shard-tglu: NOTRUN -> SKIP (fdo#111615) +1 similar issue
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#6187)
> 
>      * igt@kms_big_fb@yf-tiled-max-hw-stride-32bpp-rotate-180-async-flip:
> 
>           * shard-mtlp: NOTRUN -> SKIP (fdo#111615) +4 similar issues
>      * igt@kms_big_fb@yf-tiled-max-hw-stride-64bpp-rotate-0:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#4538 / i915#5190) +3 similar
>             issues
>      * igt@kms_ccs@pipe-a-bad-pixel-format-y_tiled_gen12_mc_ccs:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#3886 / i915#6095) +5 similar
>             issues
>      * igt@kms_ccs@pipe-a-ccs-on-another-bo-4_tiled_mtl_rc_ccs_cc:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#5354 / i915#6095) +4 similar
>             issues
>      * igt@kms_ccs@pipe-a-crc-sprite-planes-basic-y_tiled_ccs:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#3689 / i915#5354) +12 similar
>             issues
>      * igt@kms_ccs@pipe-a-missing-ccs-buffer-4_tiled_mtl_rc_ccs_cc:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#5354 / i915#6095) +8 similar
>             issues
>      * igt@kms_ccs@pipe-b-bad-aux-stride-y_tiled_gen12_mc_ccs:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#3886 / i915#5354 / i915#6095)
>      * igt@kms_ccs@pipe-b-ccs-on-another-bo-y_tiled_gen12_mc_ccs:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#3689 / i915#3886 / i915#5354 /
>             i915#6095) +2 similar issues
>      * igt@kms_ccs@pipe-b-crc-primary-basic-yf_tiled_ccs:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#3734 / i915#5354 / i915#6095) +1
>             similar issue
>      * igt@kms_ccs@pipe-c-bad-aux-stride-yf_tiled_ccs:
> 
>           * shard-tglu: NOTRUN -> SKIP (fdo#111615 / i915#3689 / i915#5354 /
>             i915#6095) +1 similar issue
>      * igt@kms_ccs@pipe-c-bad-rotation-90-4_tiled_mtl_mc_ccs:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#5354) +18 similar issues
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#5354) +8 similar issues
> 
>      * igt@kms_ccs@pipe-c-bad-rotation-90-y_tiled_gen12_rc_ccs_cc:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#3689 / i915#3886 / i915#5354) +3
>             similar issues
>      * igt@kms_ccs@pipe-d-bad-rotation-90-4_tiled_dg2_rc_ccs_cc:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#3689 / i915#5354 / i915#6095) +9
>             similar issues
>      * igt@kms_ccs@pipe-d-crc-primary-rotation-180-yf_tiled_ccs:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#6095) +16 similar issues
>      * igt@kms_chamelium_audio@dp-audio:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#7828) +3 similar issues
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#7828)
> 
>      * igt@kms_chamelium_color@ctm-0-50:
> 
>           * shard-tglu: NOTRUN -> SKIP (fdo#111827) +1 similar issue
>      * igt@kms_chamelium_color@ctm-0-75:
> 
>           * shard-mtlp: NOTRUN -> SKIP (fdo#111827)
>      * igt@kms_chamelium_color@ctm-blue-to-red:
> 
>           * shard-dg2: NOTRUN -> SKIP (fdo#111827)
>      * igt@kms_chamelium_edid@dp-edid-stress-resolution-non-4k:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#7828)
>      * igt@kms_chamelium_frames@hdmi-cmp-planar-formats:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#7828) +1 similar issue
>      * igt@kms_content_protection@legacy:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#7118) +2 similar issues
>      * igt@kms_content_protection@mei_interface:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#8063)
>      * igt@kms_content_protection@uevent:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#7118)
>      * igt@kms_cursor_crc@cursor-onscreen-512x170:
> 
>           * shard-rkl: NOTRUN -> SKIP (fdo#109279 / i915#3359)
>      * igt@kms_cursor_crc@cursor-onscreen-512x512:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#3359) +1 similar issue
>      * igt@kms_cursor_crc@cursor-random-32x10:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#8814)
>      * igt@kms_cursor_crc@cursor-random-512x512:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#3359)
>      * igt@kms_cursor_crc@cursor-sliding-512x512:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#3359) +2 similar issues
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#3359) +1 similar issue
> 
>      * igt@kms_cursor_legacy@2x-long-flip-vs-cursor-legacy:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#3546)
>      * igt@kms_cursor_legacy@cursora-vs-flipb-legacy:
> 
>           * shard-tglu: NOTRUN -> SKIP (fdo#109274) +3 similar issues
>      * igt@kms_cursor_legacy@cursorb-vs-flipa-toggle:
> 
>           * shard-dg2: NOTRUN -> SKIP (fdo#109274 / i915#5354) +1 similar
>             issue
> 
>           * shard-rkl: NOTRUN -> SKIP (fdo#111825) +1 similar issue
> 
>      * igt@kms_cursor_legacy@cursorb-vs-flipb-toggle:
> 
>           * shard-dg2: NOTRUN -> SKIP (fdo#109274 / fdo#111767 / i915#5354)
> 
>           * shard-rkl: NOTRUN -> SKIP (fdo#111767 / fdo#111825)
> 
>      * igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size:
> 
>           * shard-apl: PASS -> FAIL (i915#2346)
> 
>           * shard-glk: PASS -> FAIL (i915#2346)
> 
>      * igt@kms_dither@fb-8bpc-vs-panel-6bpc@pipe-a-hdmi-a-1:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#3804)
>      * igt@kms_dp_aux_dev:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#1257)
>      * igt@kms_dsc@dsc-with-bpc-formats:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#3555 / i915#3840)
>      * igt@kms_dsc@dsc-with-output-formats:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#3840)
>      * igt@kms_fbcon_fbt@psr:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#3469)
> 
>           * shard-rkl: NOTRUN -> SKIP (fdo#110189 / i915#3955)
> 
>      * igt@kms_flip@2x-blocking-absolute-wf_vblank:
> 
>           * shard-tglu: NOTRUN -> SKIP (fdo#109274 / i915#3637) +2 similar
>             issues
>      * igt@kms_flip@2x-flip-vs-blocking-wf-vblank:
> 
>           * shard-snb: NOTRUN -> SKIP (fdo#109271 / fdo#111767)
>      * igt@kms_flip@2x-flip-vs-fences-interruptible:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#8381)
>      * igt@kms_flip@2x-plain-flip-fb-recreate:
> 
>           * shard-apl: NOTRUN -> SKIP (fdo#109271) +24 similar issues
>      * igt@kms_flip@2x-plain-flip-interruptible:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#3637)
>      * igt@kms_flip@2x-wf_vblank-ts-check-interruptible:
> 
>           * shard-dg2: NOTRUN -> SKIP (fdo#109274)
>      * igt@kms_flip_scaled_crc@flip-32bpp-4tile-to-32bpp-4tiledg2rcccs-downscaling@pipe-a-valid-mode:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#2672)
>      * igt@kms_flip_scaled_crc@flip-32bpp-ytile-to-32bpp-ytilegen12rcccs-upscaling@pipe-a-default-mode:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#2672) +1 similar issue
>      * igt@kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-downscaling@pipe-a-default-mode:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#8810)
>      * igt@kms_force_connector_basic@prune-stale-modes:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#5274)
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#5274)
> 
>      * igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-draw-mmap-gtt:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#8708) +1 similar issue
>      * igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-fullscreen:
> 
>           * shard-dg2: PASS -> FAIL (i915#6880)
>      * igt@kms_frontbuffer_tracking@fbcpsr-1p-primscrn-pri-shrfb-draw-mmap-gtt:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#3023) +4 similar issues
>      * igt@kms_frontbuffer_tracking@fbcpsr-2p-primscrn-pri-shrfb-draw-mmap-wc:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#8708) +5 similar issues
> 
>           * shard-rkl: NOTRUN -> SKIP (fdo#111825 / i915#1825) +8 similar
>             issues
> 
>      * igt@kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-pri-shrfb-draw-pwrite:
> 
>           * shard-tglu: NOTRUN -> SKIP (fdo#109280) +14 similar issues
>      * igt@kms_frontbuffer_tracking@fbcpsr-tiling-4:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#5439)
>      * igt@kms_frontbuffer_tracking@psr-1p-offscren-pri-indfb-draw-mmap-cpu:
> 
>           * shard-tglu: NOTRUN -> SKIP (fdo#110189) +12 similar issues
>      * igt@kms_frontbuffer_tracking@psr-1p-offscren-pri-shrfb-draw-render:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#3458) +3 similar issues
>      * igt@kms_frontbuffer_tracking@psr-2p-scndscrn-pri-shrfb-draw-mmap-cpu:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#1825) +10 similar issues
>      * igt@kms_hdr@bpc-switch:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#3555 / i915#8228)
>      * igt@kms_hdr@bpc-switch-suspend:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#3555 / i915#8228)
>      * igt@kms_hdr@static-toggle:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#3555 / i915#8228) +1 similar
>             issue
>      * igt@kms_pipe_b_c_ivb@pipe-b-dpms-off-modeset-pipe-c:
> 
>           * shard-tglu: NOTRUN -> SKIP (fdo#109289) +1 similar issue
>      * igt@kms_plane_scaling@intel-max-src-size:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#6953)
>      * igt@kms_plane_scaling@plane-scaler-with-clipping-clamping-rotation@pipe-a-vga-1:
> 
>           * shard-snb: NOTRUN -> SKIP (fdo#109271) +101 similar issues
>      * igt@kms_plane_scaling@plane-upscale-with-rotation-factor-0-25@pipe-a-hdmi-a-1:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#5176) +15 similar issues
>      * igt@kms_plane_scaling@plane-upscale-with-rotation-factor-0-25@pipe-a-hdmi-a-2:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#5176) +7 similar issues
>      * igt@kms_plane_scaling@planes-downscale-factor-0-25@pipe-d-hdmi-a-3:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#5235) +15 similar issues
>      * igt@kms_plane_scaling@planes-upscale-20x20-downscale-factor-0-25@pipe-b-edp-1:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#5235) +3 similar issues
>      * igt@kms_plane_scaling@planes-upscale-factor-0-25-downscale-factor-0-25@pipe-b-hdmi-a-2:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#5235) +3 similar issues
>      * igt@kms_psr2_sf@cursor-plane-move-continuous-sf:
> 
>           * shard-apl: NOTRUN -> SKIP (fdo#109271 / i915#658)
>      * igt@kms_psr2_su@frontbuffer-xrgb8888:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#658) +1 similar issue
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#4348)
> 
>      * igt@kms_psr2_su@page_flip-xrgb8888:
> 
>           * shard-tglu: NOTRUN -> SKIP (fdo#109642 / fdo#111068 / i915#658)
>             +1 similar issue
>      * igt@kms_psr@primary_mmap_cpu:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#1072) +4 similar issues
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#1072) +2 similar issues
> 
>      * igt@kms_psr@psr2_cursor_blt:
> 
>           * shard-mtlp: PASS -> FAIL (i915#8726)
>      * igt@kms_psr@psr2_cursor_mmap_gtt:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#4077) +2 similar issues
>      * igt@kms_rmfb@close-fd@pipe-a-edp-1:
> 
>           * shard-mtlp: PASS -> DMESG-WARN (i915#1982)
>      * igt@kms_rotation_crc@primary-y-tiled-reflect-x-270:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#4235)
>      * igt@kms_scaling_modes@scaling-mode-full:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#3555) +2 similar issues
>      * igt@kms_tiled_display@basic-test-pattern:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#8623)
>      * igt@kms_tiled_display@basic-test-pattern-with-chamelium:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#8623)
>      * igt@kms_vblank@pipe-c-query-forked-busy:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#4070 / i915#6768)
>      * igt@kms_vblank@pipe-d-wait-idle:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#4070 / i915#533 / i915#6768)
>      * igt@kms_writeback@writeback-pixel-formats:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#2437)
>      * igt@perf@oa-exponents@0-rcs0:
> 
>           * shard-glk: PASS -> ABORT (i915#5213 / i915#7941)
>      * igt@perf_pmu@event-wait@rcs0:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#8807)
>      * igt@perf_pmu@rc6@other-idle-gt0:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#8516)
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#8516)
> 
>      * igt@perf_pmu@render-node-busy-idle@ccs0:
> 
>           * shard-mtlp: PASS -> FAIL (i915#4349) +7 similar issues
>      * igt@prime_vgem@basic-gtt:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#3708 / i915#4077)
>      * igt@sysfs_heartbeat_interval@nopreempt@bcs0:
> 
>           * shard-mtlp: PASS -> FAIL (i915#6015) +1 similar issue
>      * igt@sysfs_preempt_timeout@timeout@vecs0:
> 
>           * shard-mtlp: PASS -> ABORT (i915#8521)
>      * igt@v3d/v3d_perfmon@create-perfmon-exceed:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#2575) +5 similar issues
>      * igt@v3d/v3d_submit_cl@valid-submission:
> 
>           * shard-tglu: NOTRUN -> SKIP (fdo#109315 / i915#2575) +4 similar
>             issues
>      * igt@v3d/v3d_submit_csd@bad-multisync-in-sync:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#2575) +2 similar issues
> 
>           * shard-rkl: NOTRUN -> SKIP (fdo#109315) +2 similar issues
> 
>      * igt@vc4/vc4_mmap@mmap-bo:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#7711) +3 similar issues
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#7711) +2 similar issues
> 
>      * igt@vc4/vc4_perfmon@create-perfmon-exceed:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#7711) +3 similar issues
>      * igt@vc4/vc4_purgeable_bo@mark-unpurgeable-purged:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#2575) +4 similar issues
>      * igt@vgem_basic@bad-pad:
> 
>           * shard-mtlp: PASS -> DMESG-WARN (i915#2017)
> 
>     Possible fixes
> 
>      * igt@gem_create@hog-create@smem0:
> 
>           * shard-dg2: FAIL (i915#5892 / i915#8758) -> PASS
>      * igt@gem_eio@hibernate:
> 
>           * {shard-dg1}: ABORT (i915#4391 / i915#7975 / i915#8213) -> PASS
>      * igt@gem_exec_fair@basic-pace-share@rcs0:
> 
>           * shard-glk: FAIL (i915#2842) -> PASS
>      * igt@gem_exec_fair@basic-throttle@rcs0:
> 
>           * shard-rkl: FAIL (i915#2842) -> PASS +3 similar issues
>      * igt@gem_exec_whisper@basic-contexts-forked-all:
> 
>           * shard-mtlp: ABORT (i915#8131) -> PASS
>      * igt@gem_exec_whisper@basic-fds-priority-all:
> 
>           * shard-mtlp: FAIL (i915#6363) -> PASS
>      * igt@gem_lmem_swapping@smem-oom@lmem0:
> 
>           * {shard-dg1}: DMESG-WARN (i915#4936 / i915#5493) -> PASS
>      * igt@i915_hangman@engine-engine-hang@vcs0:
> 
>           * shard-mtlp: FAIL (i915#7069) -> PASS
>      * igt@i915_module_load@reload-with-fault-injection:
> 
>           * shard-mtlp: ABORT (i915#8489 / i915#8668) -> PASS
>      * igt@i915_pm_dc@dc9-dpms:
> 
>           * shard-tglu: SKIP (i915#4281) -> PASS
>      * igt@i915_pm_rc6_residency@rc6-idle@rcs0:
> 
>           * {shard-dg1}: FAIL (i915#3591) -> PASS +1 similar issue
>      * igt@i915_pm_rpm@cursor-dpms:
> 
>           * shard-tglu: FAIL (i915#7940) -> PASS +1 similar issue
>      * igt@i915_pm_rpm@dpms-mode-unset-lpsp:
> 
>           * {shard-dg1}: SKIP (i915#1397) -> PASS
>      * igt@i915_pm_rpm@dpms-non-lpsp:
> 
>           * shard-rkl: SKIP (i915#1397) -> PASS +1 similar issue
>      * igt@i915_pm_rpm@modeset-non-lpsp-stress-no-wait:
> 
>           * shard-dg2: SKIP (i915#1397) -> PASS +2 similar issues
>      * igt@i915_selftest@live@gt_mocs:
> 
>           * shard-mtlp: DMESG-FAIL (i915#7059) -> PASS
>      * igt@i915_selftest@live@hangcheck:
> 
>           * shard-dg2: ABORT (i915#7913) -> PASS
>      * igt@i915_selftest@live@workarounds:
> 
>           * shard-mtlp: DMESG-FAIL (i915#6763) -> PASS
>      * igt@kms_big_fb@x-tiled-max-hw-stride-64bpp-rotate-180-hflip-async-flip:
> 
>           * shard-mtlp: FAIL (i915#3743) -> PASS
>      * igt@kms_cursor_legacy@2x-long-flip-vs-cursor-atomic:
> 
>           * shard-glk: FAIL (i915#72) -> PASS
>      * igt@kms_cursor_legacy@cursor-vs-flip-toggle:
> 
>           * shard-mtlp: FAIL (i915#8248) -> PASS
>      * igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions:
> 
>           * shard-glk: FAIL (i915#2346) -> PASS
>      * igt@kms_flip@flip-vs-expired-vblank-interruptible@a-hdmi-a4:
> 
>           * {shard-dg1}: FAIL (i915#79) -> PASS
>      * igt@kms_flip@flip-vs-suspend-interruptible@a-dp1:
> 
>           * shard-apl: ABORT (i915#180) -> PASS
>      * igt@kms_frontbuffer_tracking@fbc-1p-primscrn-indfb-plflip-blt:
> 
>           * shard-dg2: FAIL (i915#6880) -> PASS +1 similar issue
>      * igt@kms_plane_scaling@intel-max-src-size@pipe-a-hdmi-a-2:
> 
>           * shard-rkl: FAIL (i915#8292) -> PASS
>      * igt@kms_sysfs_edid_timing:
> 
>           * shard-dg2: FAIL (IGT#2) -> PASS
>      * igt@kms_vblank@pipe-a-ts-continuation-suspend:
> 
>           * shard-dg2: FAIL (fdo#103375 / i915#6121) -> PASS
>      * igt@perf_pmu@busy-double-start@ccs0:
> 
>           * shard-mtlp: FAIL (i915#4349) -> PASS
>      * igt@perf_pmu@most-busy-idle-check-all@bcs0:
> 
>           * shard-mtlp: FAIL (i915#5234) -> PASS +1 similar issue
>      * igt@perf_pmu@most-busy-idle-check-all@rcs0:
> 
>           * shard-dg2: FAIL (i915#5234) -> PASS
> 
>           * {shard-dg1}: FAIL (i915#5234) -> PASS
> 
>      * igt@sysfs_timeslice_duration@timeout@vecs0:
> 
>           * shard-mtlp: ABORT (i915#8521) -> PASS
> 
>     Warnings
> 
>      * igt@gem_exec_suspend@basic-s0@smem:
> 
>           * shard-snb: DMESG-FAIL (fdo#103375) -> DMESG-WARN (i915#8841)
>      * igt@i915_pm_rc6_residency@rc6-idle@bcs0:
> 
>           * shard-tglu: FAIL (i915#2681 / i915#3591) -> WARN (i915#2681)
>      * igt@i915_pm_rc6_residency@rc6-idle@rcs0:
> 
>           * shard-tglu: WARN (i915#2681) -> FAIL (i915#2681 / i915#3591)
>      * igt@i915_suspend@basic-s2idle-without-i915:
> 
>           * shard-snb: DMESG-WARN (i915#8841) -> ABORT (i915#4528 /
>             i915#8213)
>      * igt@kms_content_protection@mei_interface:
> 
>           * shard-dg2: SKIP (i915#7118 / i915#7162) -> SKIP (i915#7118)
> 
>           * shard-rkl: SKIP (fdo#109300) -> SKIP (i915#7118)
> 
>           * shard-tglu: SKIP (fdo#109300) -> SKIP (i915#6944 / i915#7116 /
>             i915#7118)
> 
>      * igt@kms_force_connector_basic@force-load-detect:
> 
>           * shard-rkl: SKIP (fdo#109285 / i915#4098) -> SKIP (fdo#109285)
>      * igt@prime_mmap@test_aperture_limit@test_aperture_limit-smem:
> 
>           * shard-dg2: CRASH (i915#7331) -> INCOMPLETE (i915#5493)
> 
>    {name}: This element is suppressed. This means it is ignored when
>    computing
>    the status of the difference (SUCCESS, WARNING, or FAILURE).
> 
> Build changes
> 
>      * CI: CI-20190529 -> None
>      * IGT: IGT_7380 -> IGTPW_9383
>      * Piglit: piglit_4509 -> None
> 
>    CI-20190529: 20190529
>    CI_DRM_13369: 0595153b5c237d83f9990f3ce44d072a262dcb0e @
>    git://anongit.freedesktop.org/gfx-ci/linux
>    IGTPW_9383: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/index.html
>    IGT_7380: 8e65f12de2fd52c05dc48fdbcb8cfe86f6de1a75 @
>    https://gitlab.freedesktop.org/drm/igt-gpu-tools.git
>    piglit_4509: fdc5a4ca11124ab8413c7988896eec4c97336694 @
>    git://anongit.freedesktop.org/piglit

Re: [igt-dev] [PATCH i-g-t v3 10/17] lib/intel_allocator: Add intel_allocator_bind()

[Karolina Stolarek]

On 11.07.2023 13:20, Zbigniew Kempczyński wrote:
> Synchronize allocator state to vm.
> 
> This change allows xe user to execute vm-bind/unbind for allocator
> alloc()/free() operations which occurred since last binding/unbinding.
> Before doing exec user should call intel_allocator_bind() to ensure
> all vma's are in place.
> 
> Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
> ---
> v2: Rewrite tracking mechanism: previous code uses bind map embedded
>      in allocator structure. Unfortunately this wasn't good idea
>      - for xe binding everything was fine, but it regress multiprocess/
>      multithreaded allocations. Main reason of this was children
>      processes couldn't get its reference as this memory was allocated
>      on allocator thread (separate process). Currently each child
>      contains its own separate tracking maps for ahnd and for each
>      ahnd bind map.
> v3: - Don't use priv field (we may clean bind_map apriori as failing
>        asserts and outdated bind_map is not a problem anymore).
>      - Use only vm on tracking ahnd (Karolina)
>      - Add braces to avoid checkpatch complaining (Karolina)
>      - Use igt_assert_f() instead conditional block (Karolina)
> ---
>   lib/igt_core.c        |   5 +
>   lib/intel_allocator.c | 259 +++++++++++++++++++++++++++++++++++++++++-
>   lib/intel_allocator.h |   3 +
>   3 files changed, 265 insertions(+), 2 deletions(-)
> 
> diff --git a/lib/igt_core.c b/lib/igt_core.c
> index 2ae2cb6883..041e4b3288 100644
> --- a/lib/igt_core.c
> +++ b/lib/igt_core.c
> @@ -74,6 +74,7 @@
>   #include "igt_sysrq.h"
>   #include "igt_rc.h"
>   #include "igt_list.h"
> +#include "igt_map.h"
>   #include "igt_device_scan.h"
>   #include "igt_thread.h"
>   #include "runnercomms.h"
> @@ -320,6 +321,8 @@ bool test_multi_fork_child;
>   /* For allocator purposes */
>   pid_t child_pid  = -1;
>   __thread pid_t child_tid  = -1;
> +struct igt_map *ahnd_map;
> +pthread_mutex_t ahnd_map_mutex;
>   
>   enum {
>   	/*
> @@ -2523,6 +2526,8 @@ bool __igt_fork(void)
>   	case 0:
>   		test_child = true;
>   		pthread_mutex_init(&print_mutex, NULL);
> +		pthread_mutex_init(&ahnd_map_mutex, NULL);
> +		ahnd_map = igt_map_create(igt_map_hash_64, igt_map_equal_64);
>   		child_pid = getpid();
>   		child_tid = -1;
>   		exit_handler_count = 0;
> diff --git a/lib/intel_allocator.c b/lib/intel_allocator.c
> index cf964ccc41..f0a9b7fb53 100644
> --- a/lib/intel_allocator.c
> +++ b/lib/intel_allocator.c
> @@ -17,8 +17,8 @@
>   #include "intel_allocator.h"
>   #include "intel_allocator_msgchannel.h"
>   #include "xe/xe_query.h"
> +#include "xe/xe_util.h"
>   
> -//#define ALLOCDBG
>   #ifdef ALLOCDBG
>   #define alloc_info igt_info
>   #define alloc_debug igt_debug
> @@ -45,6 +45,14 @@ static inline const char *reqstr(enum reqtype request_type)
>   #define alloc_debug(...) {}
>   #endif
>   
> +#ifdef ALLOCBINDDBG
> +#define bind_info igt_info
> +#define bind_debug igt_debug
> +#else
> +#define bind_info(...) {}
> +#define bind_debug(...) {}
> +#endif
> +
>   /*
>    * We limit allocator space to avoid hang when batch would be
>    * pinned in the last page.
> @@ -64,6 +72,30 @@ struct handle_entry {
>   	struct allocator *al;
>   };
>   
> +/* For tracking alloc()/free() for Xe */
> +struct ahnd_info {
> +	int fd;
> +	uint64_t ahnd;
> +	uint32_t vm;
> +	enum intel_driver driver;
> +	struct igt_map *bind_map;
> +	pthread_mutex_t bind_map_mutex;
> +};
> +
> +enum allocator_bind_op {
> +	BOUND,
> +	TO_BIND,
> +	TO_UNBIND,
> +};
> +
> +struct allocator_object {
> +	uint32_t handle;
> +	uint64_t offset;
> +	uint64_t size;
> +
> +	enum allocator_bind_op bind_op;
> +};
> +
>   struct intel_allocator *
>   intel_allocator_reloc_create(int fd, uint64_t start, uint64_t end);
>   struct intel_allocator *
> @@ -122,6 +154,13 @@ static pid_t allocator_pid = -1;
>   extern pid_t child_pid;
>   extern __thread pid_t child_tid;
>   
> +/*
> + * Track alloc()/free() requires storing in local process which has
> + * an access to real drm fd it can work on.
> + */
> +extern struct igt_map *ahnd_map;
> +extern pthread_mutex_t ahnd_map_mutex;
> +
>   /*
>    * - for parent process we have child_pid == -1
>    * - for child which calls intel_allocator_init() allocator_pid == child_pid
> @@ -837,6 +876,45 @@ void intel_allocator_multiprocess_stop(void)
>   	}
>   }
>   
> +static void track_ahnd(int fd, uint64_t ahnd, uint32_t vm)
> +{
> +	struct ahnd_info *ainfo;
> +
> +	pthread_mutex_lock(&ahnd_map_mutex);
> +	ainfo = igt_map_search(ahnd_map, &ahnd);
> +	if (!ainfo) {
> +		ainfo = malloc(sizeof(*ainfo));
> +		ainfo->fd = fd;
> +		ainfo->ahnd = ahnd;
> +		ainfo->vm = vm;
> +		ainfo->driver = get_intel_driver(fd);
> +		ainfo->bind_map = igt_map_create(igt_map_hash_32, igt_map_equal_32);
> +		pthread_mutex_init(&ainfo->bind_map_mutex, NULL);
> +		bind_debug("[TRACK AHND] pid: %d, tid: %d, create <fd: %d, "
> +			   "ahnd: %llx, vm: %u, driver: %d, ahnd_map: %p, bind_map: %p>\n",
> +			   getpid(), gettid(), ainfo->fd,
> +			   (long long)ainfo->ahnd, ainfo->vm,
> +			   ainfo->driver, ahnd_map, ainfo->bind_map);
> +		igt_map_insert(ahnd_map, &ainfo->ahnd, ainfo);
> +	}
> +
> +	pthread_mutex_unlock(&ahnd_map_mutex);
> +}
> +
> +static void untrack_ahnd(uint64_t ahnd)
> +{
> +	struct ahnd_info *ainfo;
> +
> +	pthread_mutex_lock(&ahnd_map_mutex);
> +	ainfo = igt_map_search(ahnd_map, &ahnd);
> +	if (ainfo) {
> +		bind_debug("[UNTRACK AHND]: pid: %d, tid: %d, removing ahnd: %llx\n",
> +			   getpid(), gettid(), (long long)ahnd);
> +		igt_map_remove(ahnd_map, &ahnd, map_entry_free_func);
> +	}
> +	pthread_mutex_unlock(&ahnd_map_mutex);
> +}
> +
>   static uint64_t __intel_allocator_open_full(int fd, uint32_t ctx,
>   					    uint32_t vm,
>   					    uint64_t start, uint64_t end,
> @@ -895,6 +973,12 @@ static uint64_t __intel_allocator_open_full(int fd, uint32_t ctx,
>   	igt_assert(resp.open.allocator_handle);
>   	igt_assert(resp.response_type == RESP_OPEN);
>   
> +	/*
> +	 * Igts mostly uses ctx as id when opening the allocator (i915 legacy).
> +	 * If ctx is passed let's use it as an vm id, otherwise use vm.
> +	 */
> +	track_ahnd(fd, resp.open.allocator_handle, ctx ?: vm);
> +
>   	return resp.open.allocator_handle;
>   }
>   
> @@ -1001,6 +1085,8 @@ bool intel_allocator_close(uint64_t allocator_handle)
>   	igt_assert(handle_request(&req, &resp) == 0);
>   	igt_assert(resp.response_type == RESP_CLOSE);
>   
> +	untrack_ahnd(allocator_handle);
> +
>   	return resp.close.is_empty;
>   }
>   
> @@ -1034,6 +1120,74 @@ void intel_allocator_get_address_range(uint64_t allocator_handle,
>   		*endp = resp.address_range.end;
>   }
>   
> +static bool is_same(struct allocator_object *obj,
> +		    uint32_t handle, uint64_t offset, uint64_t size,
> +		    enum allocator_bind_op bind_op)
> +{
> +	return obj->handle == handle &&	obj->offset == offset && obj->size == size &&
> +	       (obj->bind_op == bind_op || obj->bind_op == BOUND);
> +}
> +
> +static void track_object(uint64_t allocator_handle, uint32_t handle,
> +			 uint64_t offset, uint64_t size,
> +			 enum allocator_bind_op bind_op)
> +{
> +	struct ahnd_info *ainfo;
> +	struct allocator_object *obj;
> +
> +	bind_debug("[TRACK OBJECT]: [%s] pid: %d, tid: %d, ahnd: %llx, handle: %u, offset: %llx, size: %llx\n",
> +		   bind_op == TO_BIND ? "BIND" : "UNBIND",
> +		   getpid(), gettid(),
> +		   (long long)allocator_handle,
> +		   handle, (long long)offset, (long long)size);
> +
> +	if (offset == ALLOC_INVALID_ADDRESS) {
> +		bind_debug("[TRACK OBJECT] => invalid address %llx, skipping tracking\n",
> +			   (long long)offset);
> +		return;
> +	}
> +
> +	pthread_mutex_lock(&ahnd_map_mutex);
> +	ainfo = igt_map_search(ahnd_map, &allocator_handle);
> +	pthread_mutex_unlock(&ahnd_map_mutex);
> +	igt_assert_f(ainfo, "[TRACK OBJECT] => MISSING ahnd %llx <=\n",
> +		     (long long)allocator_handle);
> +
> +	if (ainfo->driver == INTEL_DRIVER_I915)
> +		return; /* no-op for i915, at least for now */
> +
> +	pthread_mutex_lock(&ainfo->bind_map_mutex);
> +	obj = igt_map_search(ainfo->bind_map, &handle);
> +	if (obj) {
> +		/*
> +		 * User may call alloc() couple of times, check object is the
> +		 * same. For free() there's simple case, just remove from
> +		 * bind_map.
> +		 */
> +		if (bind_op == TO_BIND) {
> +			igt_assert_eq(is_same(obj, handle, offset, size, bind_op), true);
> +		} else if (bind_op == TO_UNBIND) {
> +			if (obj->bind_op == TO_BIND)
> +				igt_map_remove(ainfo->bind_map, &obj->handle, map_entry_free_func);
> +			else if (obj->bind_op == BOUND)
> +				obj->bind_op = bind_op;
> +		}
> +	} else {
> +		/* Ignore to unbind bo which wasn't previously inserted */
> +		if (bind_op == TO_UNBIND)
> +			goto out;
> +
> +		obj = calloc(1, sizeof(*obj));
> +		obj->handle = handle;
> +		obj->offset = offset;
> +		obj->size = size;
> +		obj->bind_op = bind_op;
> +		igt_map_insert(ainfo->bind_map, &obj->handle, obj);
> +	}
> +out:
> +	pthread_mutex_unlock(&ainfo->bind_map_mutex);
> +}
> +
>   /**
>    * __intel_allocator_alloc:
>    * @allocator_handle: handle to an allocator
> @@ -1065,6 +1219,8 @@ uint64_t __intel_allocator_alloc(uint64_t allocator_handle, uint32_t handle,
>   	igt_assert(handle_request(&req, &resp) == 0);
>   	igt_assert(resp.response_type == RESP_ALLOC);
>   
> +	track_object(allocator_handle, handle, resp.alloc.offset, size, TO_BIND);
> +
>   	return resp.alloc.offset;
>   }
>   
> @@ -1142,6 +1298,8 @@ bool intel_allocator_free(uint64_t allocator_handle, uint32_t handle)
>   	igt_assert(handle_request(&req, &resp) == 0);
>   	igt_assert(resp.response_type == RESP_FREE);
>   
> +	track_object(allocator_handle, handle, 0, 0, TO_UNBIND);
> +
>   	return resp.free.freed;
>   }
>   
> @@ -1326,6 +1484,84 @@ void intel_allocator_print(uint64_t allocator_handle)
>   	}
>   }
>   
> +static void __xe_op_bind(struct ahnd_info *ainfo, uint32_t sync_in, uint32_t sync_out)
> +{
> +	struct allocator_object *obj;
> +	struct igt_map_entry *pos;
> +	struct igt_list_head obj_list;
> +	struct xe_object *entry, *tmp;
> +
> +	IGT_INIT_LIST_HEAD(&obj_list);
> +
> +	pthread_mutex_lock(&ainfo->bind_map_mutex);
> +	igt_map_foreach(ainfo->bind_map, pos) {
> +		obj = pos->data;
> +
> +		if (obj->bind_op == BOUND)
> +			continue;
> +
> +		bind_info("= [vm: %u] %s => %u %lx %lx\n",
> +			  ainfo->vm,
> +			  obj->bind_op == TO_BIND ? "TO BIND" : "TO UNBIND",
> +			  obj->handle, obj->offset,
> +			  obj->size);
> +
> +		entry = malloc(sizeof(*entry));
> +		entry->handle = obj->handle;
> +		entry->offset = obj->offset;
> +		entry->size = obj->size;
> +		entry->bind_op = obj->bind_op == TO_BIND ? XE_OBJECT_BIND :
> +							   XE_OBJECT_UNBIND;
> +		igt_list_add(&entry->link, &obj_list);
> +
> +		/*
> +		 * We clean bind_map even before calling bind/unbind
> +		 * as all binding operations asserts in case of error.
> +		 */
> +		if (obj->bind_op == TO_BIND)
> +			obj->bind_op = BOUND;
> +		else
> +			igt_map_remove(ainfo->bind_map, &obj->handle,
> +				       map_entry_free_func);

I don't quite understand why we can clean up the map even before the 
actual bind/unbind happens. If an assert fails, it means that 
vm_(un)bind failed, so the actual operation didn't succeed, and that 
state change is incorrect. I believe that there's a reason we can do it 
this way, I just don't understand it.

Apart from that one thing, the patch looks good to me.

All the best,
Karolina

> +	}
> +	pthread_mutex_unlock(&ainfo->bind_map_mutex);
> +
> +	xe_bind_unbind_async(ainfo->fd, ainfo->vm, 0, &obj_list, sync_in, sync_out);
> +
> +	igt_list_for_each_entry_safe(entry, tmp, &obj_list, link) {
> +		igt_list_del(&entry->link);
> +		free(entry);
> +	}
> +}
> +
> +/**
> + * intel_allocator_bind:
> + * @allocator_handle: handle to an allocator
> + * @sync_in: syncobj (fence-in)
> + * @sync_out: syncobj (fence-out)
> + *
> + * Function binds and unbinds all objects added to the allocator which weren't
> + * previously binded/unbinded.
> + *
> + **/
> +void intel_allocator_bind(uint64_t allocator_handle,
> +			  uint32_t sync_in, uint32_t sync_out)
> +{
> +	struct ahnd_info *ainfo;
> +
> +	pthread_mutex_lock(&ahnd_map_mutex);
> +	ainfo = igt_map_search(ahnd_map, &allocator_handle);
> +	pthread_mutex_unlock(&ahnd_map_mutex);
> +	igt_assert(ainfo);
> +
> +	/*
> +	 * We collect bind/unbind operations on alloc()/free() to do group
> +	 * operation getting @sync_in as syncobj handle (fence-in). If user
> +	 * passes 0 as @sync_out we bind/unbind synchronously.
> +	 */
> +	__xe_op_bind(ainfo, sync_in, sync_out);
> +}
> +
>   static int equal_handles(const void *key1, const void *key2)
>   {
>   	const struct handle_entry *h1 = key1, *h2 = key2;
> @@ -1383,6 +1619,22 @@ static void __free_maps(struct igt_map *map, bool close_allocators)
>   	igt_map_destroy(map, map_entry_free_func);
>   }
>   
> +static void __free_ahnd_map(void)
> +{
> +	struct igt_map_entry *pos;
> +	struct ahnd_info *ainfo;
> +
> +	if (!ahnd_map)
> +		return;
> +
> +	igt_map_foreach(ahnd_map, pos) {
> +		ainfo = pos->data;
> +		igt_map_destroy(ainfo->bind_map, map_entry_free_func);
> +	}
> +
> +	igt_map_destroy(ahnd_map, map_entry_free_func);
> +}
> +
>   /**
>    * intel_allocator_init:
>    *
> @@ -1400,12 +1652,15 @@ void intel_allocator_init(void)
>   	__free_maps(handles, true);
>   	__free_maps(ctx_map, false);
>   	__free_maps(vm_map, false);
> +	__free_ahnd_map();
>   
>   	atomic_init(&next_handle, 1);
>   	handles = igt_map_create(hash_handles, equal_handles);
>   	ctx_map = igt_map_create(hash_instance, equal_ctx);
>   	vm_map = igt_map_create(hash_instance, equal_vm);
> -	igt_assert(handles && ctx_map && vm_map);
> +	pthread_mutex_init(&ahnd_map_mutex, NULL);
> +	ahnd_map = igt_map_create(igt_map_hash_64, igt_map_equal_64);
> +	igt_assert(handles && ctx_map && vm_map && ahnd_map);
>   
>   	channel = intel_allocator_get_msgchannel(CHANNEL_SYSVIPC_MSGQUEUE);
>   }
> diff --git a/lib/intel_allocator.h b/lib/intel_allocator.h
> index 3ec74f6191..f9ff7f1cc9 100644
> --- a/lib/intel_allocator.h
> +++ b/lib/intel_allocator.h
> @@ -210,6 +210,9 @@ bool intel_allocator_reserve_if_not_allocated(uint64_t allocator_handle,
>   
>   void intel_allocator_print(uint64_t allocator_handle);
>   
> +void intel_allocator_bind(uint64_t allocator_handle,
> +			  uint32_t sync_in, uint32_t sync_out);
> +
>   #define ALLOC_INVALID_ADDRESS (-1ull)
>   #define INTEL_ALLOCATOR_NONE   0
>   #define INTEL_ALLOCATOR_RELOC  1

Re: [igt-dev] [PATCH i-g-t v3 11/17] lib/intel_ctx: Add xe context information

[Karolina Stolarek]

On 11.07.2023 13:20, Zbigniew Kempczyński wrote:
> Most complicated part in adopting i915_blt to intel_blt - which should
> handle both drivers - is how to achieve pipelined execution. In term
> pipelined execution I mean all gpu workloads are executed without
> stalls.
> 
> Comparing to i915 relocations and softpinning xe architecture migrates
> binding (this means also unbind operation) responsibility from the
> kernel to user via vm_bind ioctl(). To avoid stalls user has to
> provide in/out fences (syncobjs) between consecutive bindings/execs.
> Of course for many igt tests we don't need pipelined execution,
> just synchronous bind, then exec. But exercising the driver should
> also cover pipelining to verify it is possible to work without stalls.
> 
> I decided to extend intel_ctx_t with all necessary for xe objects
> (vm, engine, syncobjs) to get flexibility in deciding how to bind,
> execute and wait for (synchronize) those operations. Context object
> along with i915 engine is already passed to blitter library so adding
> xe required fields doesn't break i915 but will allow xe path to get
> all necessary data to execute.
> 
> Using intel_ctx with xe requires some code patterns caused by usage
> of an allocator. For xe the allocator started tracking alloc()/free()
> operations to do bind/unbind in one call just before execution.
> I've added two helpers in intel_ctx which - intel_ctx_xe_exec()
> and intel_ctx_xe_sync(). Depending how intel_ctx was created
> (with 0 or real syncobj handles as in/bind/out fences) bind and exec
> in intel_ctx_xe_exec() are pipelined but synchronize last operation
> (exec). For real syncobjs they are used to join bind + exec calls
> but there's no wait for exec (sync-out) completion. This allows
> building more cascaded bind + exec operations without stalls.
> 
> To wait for a sync-out fence caller may use intel_ctx_xe_sync()
> which is synchronous wait on syncobj. It allows user to reset
> fences to prepare for next operation.
> 
> Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>

Reviewed-by: Karolina Stolarek <karolina.stolarek@intel.com>

> ---
> v3: drop local __xe_exec() (Karolina)
> ---
>   lib/intel_ctx.c | 98 ++++++++++++++++++++++++++++++++++++++++++++++++-
>   lib/intel_ctx.h | 14 +++++++
>   2 files changed, 111 insertions(+), 1 deletion(-)
> 
> diff --git a/lib/intel_ctx.c b/lib/intel_ctx.c
> index ded9c0f1e4..0c1ce6d10b 100644
> --- a/lib/intel_ctx.c
> +++ b/lib/intel_ctx.c
> @@ -5,9 +5,12 @@
>   
>   #include <stddef.h>
>   
> +#include "i915/gem_engine_topology.h"
> +#include "igt_syncobj.h"
> +#include "intel_allocator.h"
>   #include "intel_ctx.h"
>   #include "ioctl_wrappers.h"
> -#include "i915/gem_engine_topology.h"
> +#include "xe/xe_ioctl.h"
>   
>   /**
>    * SECTION:intel_ctx
> @@ -390,3 +393,96 @@ unsigned int intel_ctx_engine_class(const intel_ctx_t *ctx, unsigned int engine)
>   {
>   	return intel_ctx_cfg_engine_class(&ctx->cfg, engine);
>   }
> +
> +/**
> + * intel_ctx_xe:
> + * @fd: open i915 drm file descriptor
> + * @vm: vm
> + * @engine: engine
> + *
> + * Returns an intel_ctx_t representing the xe context.
> + */
> +intel_ctx_t *intel_ctx_xe(int fd, uint32_t vm, uint32_t engine,
> +			  uint32_t sync_in, uint32_t sync_bind, uint32_t sync_out)
> +{
> +	intel_ctx_t *ctx;
> +
> +	ctx = calloc(1, sizeof(*ctx));
> +	igt_assert(ctx);
> +
> +	ctx->fd = fd;
> +	ctx->vm = vm;
> +	ctx->engine = engine;
> +	ctx->sync_in = sync_in;
> +	ctx->sync_bind = sync_bind;
> +	ctx->sync_out = sync_out;
> +
> +	return ctx;
> +}
> +
> +int __intel_ctx_xe_exec(const intel_ctx_t *ctx, uint64_t ahnd, uint64_t bb_offset)
> +{
> +	struct drm_xe_sync syncs[2] = {
> +		{ .flags = DRM_XE_SYNC_SYNCOBJ, },
> +		{ .flags = DRM_XE_SYNC_SYNCOBJ | DRM_XE_SYNC_SIGNAL, },
> +	};
> +	struct drm_xe_exec exec = {
> +		.engine_id = ctx->engine,
> +		.syncs = (uintptr_t)syncs,
> +		.num_syncs = 2,
> +		.address = bb_offset,
> +		.num_batch_buffer = 1,
> +	};
> +	uint32_t sync_in = ctx->sync_in;
> +	uint32_t sync_bind = ctx->sync_bind ?: syncobj_create(ctx->fd, 0);
> +	uint32_t sync_out = ctx->sync_out ?: syncobj_create(ctx->fd, 0);
> +	int ret;
> +
> +	/* Synchronize allocator state -> vm */
> +	intel_allocator_bind(ahnd, sync_in, sync_bind);
> +
> +	/* Pipelined exec */
> +	syncs[0].handle = sync_bind;
> +	syncs[1].handle = sync_out;
> +
> +	ret = __xe_exec(ctx->fd, &exec);
> +	if (ret)
> +		goto err;
> +
> +	if (!ctx->sync_bind || !ctx->sync_out)
> +		syncobj_wait_err(ctx->fd, &sync_out, 1, INT64_MAX, 0);
> +
> +err:
> +	if (!ctx->sync_bind)
> +		syncobj_destroy(ctx->fd, sync_bind);
> +
> +	if (!ctx->sync_out)
> +		syncobj_destroy(ctx->fd, sync_out);
> +
> +	return ret;
> +}
> +
> +void intel_ctx_xe_exec(const intel_ctx_t *ctx, uint64_t ahnd, uint64_t bb_offset)
> +{
> +	igt_assert_eq(__intel_ctx_xe_exec(ctx, ahnd, bb_offset), 0);
> +}
> +
> +#define RESET_SYNCOBJ(__fd, __sync) do { \
> +	if (__sync) \
> +		syncobj_reset((__fd), &(__sync), 1); \
> +} while (0)
> +
> +int intel_ctx_xe_sync(intel_ctx_t *ctx, bool reset_syncs)
> +{
> +	int ret;
> +
> +	ret = syncobj_wait_err(ctx->fd, &ctx->sync_out, 1, INT64_MAX, 0);
> +
> +	if (reset_syncs) {
> +		RESET_SYNCOBJ(ctx->fd, ctx->sync_in);
> +		RESET_SYNCOBJ(ctx->fd, ctx->sync_bind);
> +		RESET_SYNCOBJ(ctx->fd, ctx->sync_out);
> +	}
> +
> +	return ret;
> +}
> diff --git a/lib/intel_ctx.h b/lib/intel_ctx.h
> index 3cfeaae81e..59d0360ada 100644
> --- a/lib/intel_ctx.h
> +++ b/lib/intel_ctx.h
> @@ -67,6 +67,14 @@ int intel_ctx_cfg_engine_class(const intel_ctx_cfg_t *cfg, unsigned int engine);
>   typedef struct intel_ctx {
>   	uint32_t id;
>   	intel_ctx_cfg_t cfg;
> +
> +	/* Xe */
> +	int fd;
> +	uint32_t vm;
> +	uint32_t engine;
> +	uint32_t sync_in;
> +	uint32_t sync_bind;
> +	uint32_t sync_out;
>   } intel_ctx_t;
>   
>   int __intel_ctx_create(int fd, const intel_ctx_cfg_t *cfg,
> @@ -81,4 +89,10 @@ void intel_ctx_destroy(int fd, const intel_ctx_t *ctx);
>   
>   unsigned int intel_ctx_engine_class(const intel_ctx_t *ctx, unsigned int engine);
>   
> +intel_ctx_t *intel_ctx_xe(int fd, uint32_t vm, uint32_t engine,
> +			  uint32_t sync_in, uint32_t sync_bind, uint32_t sync_out);
> +int __intel_ctx_xe_exec(const intel_ctx_t *ctx, uint64_t ahnd, uint64_t bb_offset);
> +void intel_ctx_xe_exec(const intel_ctx_t *ctx, uint64_t ahnd, uint64_t bb_offset);
> +int intel_ctx_xe_sync(intel_ctx_t *ctx, bool reset_syncs);
> +
>   #endif

Re: [igt-dev] [PATCH i-g-t v3 13/17] lib/intel_blt: Extend blitter library to support xe driver

[Karolina Stolarek]

On 11.07.2023 13:20, Zbigniew Kempczyński wrote:
> Use already written for i915 blitter library in xe development.
> Add appropriate code paths which are unique for those drivers.
> 
> Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
> ---
> v3: Extract common helpers for getting mapping and alignment (Karolina)
> ---
>   lib/intel_blt.c | 235 ++++++++++++++++++++++++++++++------------------
>   lib/intel_blt.h |   2 +-
>   2 files changed, 150 insertions(+), 87 deletions(-)
> 
> diff --git a/lib/intel_blt.c b/lib/intel_blt.c
> index f2f86e4947..5387a60ee3 100644
> --- a/lib/intel_blt.c
> +++ b/lib/intel_blt.c
> @@ -9,9 +9,13 @@
>   #include <malloc.h>
>   #include <cairo.h>
>   #include "drm.h"
> -#include "igt.h"
>   #include "i915/gem_create.h"
> +#include "igt.h"
> +#include "igt_syncobj.h"
>   #include "intel_blt.h"
> +#include "xe/xe_ioctl.h"
> +#include "xe/xe_query.h"
> +#include "xe/xe_util.h"
>   
>   #define BITRANGE(start, end) (end - start + 1)
>   #define GET_CMDS_INFO(__fd) intel_get_cmds_info(intel_get_drm_devid(__fd))
> @@ -468,24 +472,40 @@ static int __special_mode(const struct blt_copy_data *blt)
>   	return SM_NONE;
>   }
>   
> -static int __memory_type(uint32_t region)
> +static int __memory_type(int fd, enum intel_driver driver, uint32_t region)
>   {
> -	igt_assert_f(IS_DEVICE_MEMORY_REGION(region) ||
> -		     IS_SYSTEM_MEMORY_REGION(region),
> -		     "Invalid region: %x\n", region);
> +	if (driver == INTEL_DRIVER_I915) {
> +		igt_assert_f(IS_DEVICE_MEMORY_REGION(region) ||
> +			     IS_SYSTEM_MEMORY_REGION(region),
> +			     "Invalid region: %x\n", region);
> +	} else {
> +		igt_assert_f(XE_IS_VRAM_MEMORY_REGION(fd, region) ||
> +			     XE_IS_SYSMEM_MEMORY_REGION(fd, region),
> +			     "Invalid region: %x\n", region);
> +	}
>   
> -	if (IS_DEVICE_MEMORY_REGION(region))
> +	if (driver == INTEL_DRIVER_I915 && IS_DEVICE_MEMORY_REGION(region))
>   		return TM_LOCAL_MEM;
> +	else if (driver == INTEL_DRIVER_XE && XE_IS_VRAM_MEMORY_REGION(fd, region))
> +		return TM_LOCAL_MEM;
> +
>   	return TM_SYSTEM_MEM;
>   }
>   
> -static enum blt_aux_mode __aux_mode(const struct blt_copy_object *obj)
> +static enum blt_aux_mode __aux_mode(int fd,
> +				    enum intel_driver driver,
> +				    const struct blt_copy_object *obj)
>   {
> -	if (obj->compression == COMPRESSION_ENABLED) {
> +	if (driver == INTEL_DRIVER_I915 && obj->compression == COMPRESSION_ENABLED) {
>   		igt_assert_f(IS_DEVICE_MEMORY_REGION(obj->region),
>   			     "XY_BLOCK_COPY_BLT supports compression "
>   			     "on device memory only\n");
>   		return AM_AUX_CCS_E;
> +	} else if (driver == INTEL_DRIVER_XE && obj->compression == COMPRESSION_ENABLED) {
> +		igt_assert_f(XE_IS_VRAM_MEMORY_REGION(fd, obj->region),
> +			     "XY_BLOCK_COPY_BLT supports compression "
> +			     "on device memory only\n");
> +		return AM_AUX_CCS_E;
>   	}
>   
>   	return AM_AUX_NONE;
> @@ -508,9 +528,9 @@ static void fill_data(struct gen12_block_copy_data *data,
>   	data->dw00.length = extended_command ? 20 : 10;
>   
>   	if (__special_mode(blt) == SM_FULL_RESOLVE)
> -		data->dw01.dst_aux_mode = __aux_mode(&blt->src);
> +		data->dw01.dst_aux_mode = __aux_mode(blt->fd, blt->driver, &blt->src);
>   	else
> -		data->dw01.dst_aux_mode = __aux_mode(&blt->dst);
> +		data->dw01.dst_aux_mode = __aux_mode(blt->fd, blt->driver, &blt->dst);
>   	data->dw01.dst_pitch = blt->dst.pitch - 1;
>   
>   	data->dw01.dst_mocs = blt->dst.mocs;
> @@ -531,13 +551,13 @@ static void fill_data(struct gen12_block_copy_data *data,
>   
>   	data->dw06.dst_x_offset = blt->dst.x_offset;
>   	data->dw06.dst_y_offset = blt->dst.y_offset;
> -	data->dw06.dst_target_memory = __memory_type(blt->dst.region);
> +	data->dw06.dst_target_memory = __memory_type(blt->fd, blt->driver, blt->dst.region);
>   
>   	data->dw07.src_x1 = blt->src.x1;
>   	data->dw07.src_y1 = blt->src.y1;
>   
>   	data->dw08.src_pitch = blt->src.pitch - 1;
> -	data->dw08.src_aux_mode = __aux_mode(&blt->src);
> +	data->dw08.src_aux_mode = __aux_mode(blt->fd, blt->driver, &blt->src);
>   	data->dw08.src_mocs = blt->src.mocs;
>   	data->dw08.src_compression = blt->src.compression;
>   	data->dw08.src_tiling = __block_tiling(blt->src.tiling);
> @@ -550,7 +570,7 @@ static void fill_data(struct gen12_block_copy_data *data,
>   
>   	data->dw11.src_x_offset = blt->src.x_offset;
>   	data->dw11.src_y_offset = blt->src.y_offset;
> -	data->dw11.src_target_memory = __memory_type(blt->src.region);
> +	data->dw11.src_target_memory = __memory_type(blt->fd, blt->driver, blt->src.region);
>   }
>   
>   static void fill_data_ext(struct gen12_block_copy_data_ext *dext,
> @@ -692,6 +712,24 @@ static void dump_bb_ext(struct gen12_block_copy_data_ext *data)
>   		 data->dw21.src_array_index);
>   }
>   
> +static uint64_t get_default_alignment(int fd, enum intel_driver driver)
> +{
> +	if (driver == INTEL_DRIVER_XE)
> +		return xe_get_default_alignment(fd);
> +
> +	return gem_detect_safe_alignment(fd);
> +}
> +
> +static void *bo_map(int fd, uint32_t handle, uint64_t size,
> +		    enum intel_driver driver)
> +{
> +	if (driver == INTEL_DRIVER_XE)
> +		return xe_bo_map(fd, handle, size);
> +
> +	return gem_mmap__device_coherent(fd, handle, 0, size,
> +					 PROT_READ | PROT_WRITE);
> +}
> +
>   /**
>    * blt_copy_init:
>    * @fd: drm fd
> @@ -739,7 +777,7 @@ uint64_t emit_blt_block_copy(int fd,
>   	igt_assert_f(ahnd, "block-copy supports softpin only\n");
>   	igt_assert_f(blt, "block-copy requires data to do blit\n");
>   
> -	alignment = gem_detect_safe_alignment(fd);
> +	alignment = get_default_alignment(fd, blt->driver);
>   	src_offset = get_offset(ahnd, blt->src.handle, blt->src.size, alignment)
>   		     + blt->src.plane_offset;
>   	dst_offset = get_offset(ahnd, blt->dst.handle, blt->dst.size, alignment)
> @@ -748,8 +786,7 @@ uint64_t emit_blt_block_copy(int fd,
>   
>   	fill_data(&data, blt, src_offset, dst_offset, ext);
>   
> -	bb = gem_mmap__device_coherent(fd, blt->bb.handle, 0, blt->bb.size,
> -				       PROT_READ | PROT_WRITE);
> +	bb = bo_map(fd, blt->bb.handle, blt->bb.size, blt->driver);
>   
>   	igt_assert(bb_pos + sizeof(data) < blt->bb.size);
>   	memcpy(bb + bb_pos, &data, sizeof(data));
> @@ -812,29 +849,35 @@ int blt_block_copy(int fd,
>   
>   	igt_assert_f(ahnd, "block-copy supports softpin only\n");
>   	igt_assert_f(blt, "block-copy requires data to do blit\n");
> +	igt_assert_neq(blt->driver, 0);
>   
> -	alignment = gem_detect_safe_alignment(fd);
> +	alignment = get_default_alignment(fd, blt->driver);
>   	src_offset = get_offset(ahnd, blt->src.handle, blt->src.size, alignment);
>   	dst_offset = get_offset(ahnd, blt->dst.handle, blt->dst.size, alignment);
>   	bb_offset = get_offset(ahnd, blt->bb.handle, blt->bb.size, alignment);
>   
>   	emit_blt_block_copy(fd, ahnd, blt, ext, 0, true);
>   
> -	obj[0].offset = CANONICAL(dst_offset);
> -	obj[1].offset = CANONICAL(src_offset);
> -	obj[2].offset = CANONICAL(bb_offset);
> -	obj[0].handle = blt->dst.handle;
> -	obj[1].handle = blt->src.handle;
> -	obj[2].handle = blt->bb.handle;
> -	obj[0].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_WRITE |
> -		       EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> -	obj[1].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> -	obj[2].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> -	execbuf.buffer_count = 3;
> -	execbuf.buffers_ptr = to_user_pointer(obj);
> -	execbuf.rsvd1 = ctx ? ctx->id : 0;
> -	execbuf.flags = e ? e->flags : I915_EXEC_BLT;
> -	ret = __gem_execbuf(fd, &execbuf);
> +	if (blt->driver == INTEL_DRIVER_XE) {
> +		intel_ctx_xe_exec(ctx, ahnd, CANONICAL(bb_offset));
> +	} else {
> +		obj[0].offset = CANONICAL(dst_offset);
> +		obj[1].offset = CANONICAL(src_offset);
> +		obj[2].offset = CANONICAL(bb_offset);
> +		obj[0].handle = blt->dst.handle;
> +		obj[1].handle = blt->src.handle;
> +		obj[2].handle = blt->bb.handle;
> +		obj[0].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_WRITE |
> +				EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> +		obj[1].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> +		obj[2].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> +		execbuf.buffer_count = 3;
> +		execbuf.buffers_ptr = to_user_pointer(obj);
> +		execbuf.rsvd1 = ctx ? ctx->id : 0;
> +		execbuf.flags = e ? e->flags : I915_EXEC_BLT;
> +
> +		ret = __gem_execbuf(fd, &execbuf);
> +	}
>   
>   	return ret;
>   }
> @@ -950,7 +993,7 @@ uint64_t emit_blt_ctrl_surf_copy(int fd,
>   	igt_assert_f(ahnd, "ctrl-surf-copy supports softpin only\n");
>   	igt_assert_f(surf, "ctrl-surf-copy requires data to do ctrl-surf-copy blit\n");
>   
> -	alignment = max_t(uint64_t, gem_detect_safe_alignment(fd), 1ull << 16);
> +	alignment = max_t(uint64_t, get_default_alignment(fd, surf->driver), 1ull << 16);
>   
>   	data.dw00.client = 0x2;
>   	data.dw00.opcode = 0x48;
> @@ -973,8 +1016,7 @@ uint64_t emit_blt_ctrl_surf_copy(int fd,
>   	data.dw04.dst_address_hi = dst_offset >> 32;
>   	data.dw04.dst_mocs = surf->dst.mocs;
>   
> -	bb = gem_mmap__device_coherent(fd, surf->bb.handle, 0, surf->bb.size,
> -				       PROT_READ | PROT_WRITE);
> +	bb = bo_map(fd, surf->bb.handle, surf->bb.size, surf->driver);
>   
>   	igt_assert(bb_pos + sizeof(data) < surf->bb.size);
>   	memcpy(bb + bb_pos, &data, sizeof(data));
> @@ -1002,7 +1044,7 @@ uint64_t emit_blt_ctrl_surf_copy(int fd,
>   
>   /**
>    * blt_ctrl_surf_copy:
> - * @fd: drm fd
> + * @blt: bldrm fd

Emm, the argument seem to still be fd, and there's no blt. A typo?

Thanks for adding the helpers, the code looks nice and tidy (although 
I'm scared of __special_mode and the like, but there's no better way to 
do it).

All the best,
Karolina

>    * @ctx: intel_ctx_t context
>    * @e: blitter engine for @ctx
>    * @ahnd: allocator handle
> @@ -1026,32 +1068,37 @@ int blt_ctrl_surf_copy(int fd,
>   
>   	igt_assert_f(ahnd, "ctrl-surf-copy supports softpin only\n");
>   	igt_assert_f(surf, "ctrl-surf-copy requires data to do ctrl-surf-copy blit\n");
> +	igt_assert_neq(surf->driver, 0);
>   
> -	alignment = max_t(uint64_t, gem_detect_safe_alignment(fd), 1ull << 16);
> +	alignment = max_t(uint64_t, get_default_alignment(fd, surf->driver), 1ull << 16);
>   	src_offset = get_offset(ahnd, surf->src.handle, surf->src.size, alignment);
>   	dst_offset = get_offset(ahnd, surf->dst.handle, surf->dst.size, alignment);
>   	bb_offset = get_offset(ahnd, surf->bb.handle, surf->bb.size, alignment);
>   
>   	emit_blt_ctrl_surf_copy(fd, ahnd, surf, 0, true);
>   
> -	obj[0].offset = CANONICAL(dst_offset);
> -	obj[1].offset = CANONICAL(src_offset);
> -	obj[2].offset = CANONICAL(bb_offset);
> -	obj[0].handle = surf->dst.handle;
> -	obj[1].handle = surf->src.handle;
> -	obj[2].handle = surf->bb.handle;
> -	obj[0].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_WRITE |
> -		       EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> -	obj[1].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> -	obj[2].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> -	execbuf.buffer_count = 3;
> -	execbuf.buffers_ptr = to_user_pointer(obj);
> -	execbuf.flags = e ? e->flags : I915_EXEC_BLT;
> -	execbuf.rsvd1 = ctx ? ctx->id : 0;
> -	gem_execbuf(fd, &execbuf);
> -	put_offset(ahnd, surf->dst.handle);
> -	put_offset(ahnd, surf->src.handle);
> -	put_offset(ahnd, surf->bb.handle);
> +	if (surf->driver == INTEL_DRIVER_XE) {
> +		intel_ctx_xe_exec(ctx, ahnd, CANONICAL(bb_offset));
> +	} else {
> +		obj[0].offset = CANONICAL(dst_offset);
> +		obj[1].offset = CANONICAL(src_offset);
> +		obj[2].offset = CANONICAL(bb_offset);
> +		obj[0].handle = surf->dst.handle;
> +		obj[1].handle = surf->src.handle;
> +		obj[2].handle = surf->bb.handle;
> +		obj[0].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_WRITE |
> +				EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> +		obj[1].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> +		obj[2].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> +		execbuf.buffer_count = 3;
> +		execbuf.buffers_ptr = to_user_pointer(obj);
> +		execbuf.flags = e ? e->flags : I915_EXEC_BLT;
> +		execbuf.rsvd1 = ctx ? ctx->id : 0;
> +		gem_execbuf(fd, &execbuf);
> +		put_offset(ahnd, surf->dst.handle);
> +		put_offset(ahnd, surf->src.handle);
> +		put_offset(ahnd, surf->bb.handle);
> +	}
>   
>   	return 0;
>   }
> @@ -1208,7 +1255,7 @@ uint64_t emit_blt_fast_copy(int fd,
>   	uint32_t bbe = MI_BATCH_BUFFER_END;
>   	uint32_t *bb;
>   
> -	alignment = gem_detect_safe_alignment(fd);
> +	alignment = get_default_alignment(fd, blt->driver);
>   
>   	data.dw00.client = 0x2;
>   	data.dw00.opcode = 0x42;
> @@ -1218,8 +1265,8 @@ uint64_t emit_blt_fast_copy(int fd,
>   
>   	data.dw01.dst_pitch = blt->dst.pitch;
>   	data.dw01.color_depth = __fast_color_depth(blt->color_depth);
> -	data.dw01.dst_memory = __memory_type(blt->dst.region);
> -	data.dw01.src_memory = __memory_type(blt->src.region);
> +	data.dw01.dst_memory = __memory_type(blt->fd, blt->driver, blt->dst.region);
> +	data.dw01.src_memory = __memory_type(blt->fd, blt->driver, blt->src.region);
>   	data.dw01.dst_type_y = __new_tile_y_type(blt->dst.tiling) ? 1 : 0;
>   	data.dw01.src_type_y = __new_tile_y_type(blt->src.tiling) ? 1 : 0;
>   
> @@ -1246,8 +1293,7 @@ uint64_t emit_blt_fast_copy(int fd,
>   	data.dw08.src_address_lo = src_offset;
>   	data.dw09.src_address_hi = src_offset >> 32;
>   
> -	bb = gem_mmap__device_coherent(fd, blt->bb.handle, 0, blt->bb.size,
> -				       PROT_READ | PROT_WRITE);
> +	bb = bo_map(fd, blt->bb.handle, blt->bb.size, blt->driver);
>   
>   	igt_assert(bb_pos + sizeof(data) < blt->bb.size);
>   	memcpy(bb + bb_pos, &data, sizeof(data));
> @@ -1297,32 +1343,39 @@ int blt_fast_copy(int fd,
>   	uint64_t dst_offset, src_offset, bb_offset, alignment;
>   	int ret;
>   
> -	alignment = gem_detect_safe_alignment(fd);
> +	igt_assert_f(ahnd, "fast-copy supports softpin only\n");
> +	igt_assert_f(blt, "fast-copy requires data to do fast-copy blit\n");
> +	igt_assert_neq(blt->driver, 0);
>   
> +	alignment = get_default_alignment(fd, blt->driver);
>   	src_offset = get_offset(ahnd, blt->src.handle, blt->src.size, alignment);
>   	dst_offset = get_offset(ahnd, blt->dst.handle, blt->dst.size, alignment);
>   	bb_offset = get_offset(ahnd, blt->bb.handle, blt->bb.size, alignment);
>   
>   	emit_blt_fast_copy(fd, ahnd, blt, 0, true);
>   
> -	obj[0].offset = CANONICAL(dst_offset);
> -	obj[1].offset = CANONICAL(src_offset);
> -	obj[2].offset = CANONICAL(bb_offset);
> -	obj[0].handle = blt->dst.handle;
> -	obj[1].handle = blt->src.handle;
> -	obj[2].handle = blt->bb.handle;
> -	obj[0].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_WRITE |
> -		       EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> -	obj[1].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> -	obj[2].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> -	execbuf.buffer_count = 3;
> -	execbuf.buffers_ptr = to_user_pointer(obj);
> -	execbuf.rsvd1 = ctx ? ctx->id : 0;
> -	execbuf.flags = e ? e->flags : I915_EXEC_BLT;
> -	ret = __gem_execbuf(fd, &execbuf);
> -	put_offset(ahnd, blt->dst.handle);
> -	put_offset(ahnd, blt->src.handle);
> -	put_offset(ahnd, blt->bb.handle);
> +	if (blt->driver == INTEL_DRIVER_XE) {
> +		intel_ctx_xe_exec(ctx, ahnd, CANONICAL(bb_offset));
> +	} else {
> +		obj[0].offset = CANONICAL(dst_offset);
> +		obj[1].offset = CANONICAL(src_offset);
> +		obj[2].offset = CANONICAL(bb_offset);
> +		obj[0].handle = blt->dst.handle;
> +		obj[1].handle = blt->src.handle;
> +		obj[2].handle = blt->bb.handle;
> +		obj[0].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_WRITE |
> +				EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> +		obj[1].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> +		obj[2].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> +		execbuf.buffer_count = 3;
> +		execbuf.buffers_ptr = to_user_pointer(obj);
> +		execbuf.rsvd1 = ctx ? ctx->id : 0;
> +		execbuf.flags = e ? e->flags : I915_EXEC_BLT;
> +		ret = __gem_execbuf(fd, &execbuf);
> +		put_offset(ahnd, blt->dst.handle);
> +		put_offset(ahnd, blt->src.handle);
> +		put_offset(ahnd, blt->bb.handle);
> +	}
>   
>   	return ret;
>   }
> @@ -1366,16 +1419,21 @@ blt_create_object(const struct blt_copy_data *blt, uint32_t region,
>   	obj = calloc(1, sizeof(*obj));
>   
>   	obj->size = size;
> -	igt_assert(__gem_create_in_memory_regions(blt->fd, &handle,
> -						  &size, region) == 0);
> +
> +	if (blt->driver == INTEL_DRIVER_XE) {
> +		size = ALIGN(size, xe_get_default_alignment(blt->fd));
> +		handle = xe_bo_create_flags(blt->fd, 0, size, region);
> +	} else {
> +		igt_assert(__gem_create_in_memory_regions(blt->fd, &handle,
> +							  &size, region) == 0);
> +	}
>   
>   	blt_set_object(obj, handle, size, region, mocs, tiling,
>   		       compression, compression_type);
>   	blt_set_geom(obj, stride, 0, 0, width, height, 0, 0);
>   
>   	if (create_mapping)
> -		obj->ptr = gem_mmap__device_coherent(blt->fd, handle, 0, size,
> -						     PROT_READ | PROT_WRITE);
> +		obj->ptr = bo_map(blt->fd, handle, size, blt->driver);
>   
>   	return obj;
>   }
> @@ -1518,14 +1576,19 @@ void blt_surface_to_png(int fd, uint32_t run_id, const char *fileid,
>   	int format;
>   	int stride = obj->tiling ? obj->pitch * 4 : obj->pitch;
>   	char filename[FILENAME_MAX];
> +	bool is_xe = is_xe_device(fd);
>   
>   	snprintf(filename, FILENAME_MAX-1, "%d-%s-%s-%ux%u-%s.png",
>   		 run_id, fileid, blt_tiling_name(obj->tiling), width, height,
>   		 obj->compression ? "compressed" : "uncompressed");
>   
> -	if (!map)
> -		map = gem_mmap__device_coherent(fd, obj->handle, 0,
> -						obj->size, PROT_READ);
> +	if (!map) {
> +		if (is_xe)
> +			map = xe_bo_map(fd, obj->handle, obj->size);
> +		else
> +			map = gem_mmap__device_coherent(fd, obj->handle, 0,
> +							obj->size, PROT_READ);
> +	}
>   	format = CAIRO_FORMAT_RGB24;
>   	surface = cairo_image_surface_create_for_data(map,
>   						      format, width, height,
> diff --git a/lib/intel_blt.h b/lib/intel_blt.h
> index 7516ce8ac7..944e2b4ae7 100644
> --- a/lib/intel_blt.h
> +++ b/lib/intel_blt.h
> @@ -8,7 +8,7 @@
>   
>   /**
>    * SECTION:intel_blt
> - * @short_description: i915 blitter library
> + * @short_description: i915/xe blitter library
>    * @title: Blitter library
>    * @include: intel_blt.h
>    *

Re: [igt-dev] [PATCH i-g-t v3 17/17] tests/gem_ccs: Avoid writing png twice

[Karolina Stolarek]

On 11.07.2023 13:20, Zbigniew Kempczyński wrote:
> As Karolina noticed we write source png twice what's just takes
> time as source surface is not touched. Drop one png write.

It was outside of the scope, but you did it! Thanks for writing this patch:

Reviewed-by: Karolina Stolarek <karolina.stolarek@intel.com>

> 
> Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
> Cc: Karolina Stolarek <karolina.stolarek@intel.com>
> ---
>   tests/i915/gem_ccs.c | 1 -
>   1 file changed, 1 deletion(-)
> 
> diff --git a/tests/i915/gem_ccs.c b/tests/i915/gem_ccs.c
> index d9d785ed9b..a1302e0a4d 100644
> --- a/tests/i915/gem_ccs.c
> +++ b/tests/i915/gem_ccs.c
> @@ -421,7 +421,6 @@ static void block_copy(int i915,
>   	if (mid->compression)
>   		igt_assert(memcmp(src->ptr, mid->ptr, src->size) != 0);
>   
> -	WRITE_PNG(i915, run_id, "src", &blt.src, width, height);
>   	WRITE_PNG(i915, run_id, "mid", &blt.dst, width, height);
>   
>   	if (config->surfcopy && pext) {

Re: [igt-dev] [PATCH i-g-t v3 10/17] lib/intel_allocator: Add intel_allocator_bind()

[Zbigniew Kempczyński]

On Wed, Jul 12, 2023 at 11:13:24AM +0200, Karolina Stolarek wrote:
> On 11.07.2023 13:20, Zbigniew Kempczyński wrote:
> > Synchronize allocator state to vm.
> > 
> > This change allows xe user to execute vm-bind/unbind for allocator
> > alloc()/free() operations which occurred since last binding/unbinding.
> > Before doing exec user should call intel_allocator_bind() to ensure
> > all vma's are in place.
> > 
> > Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
> > ---
> > v2: Rewrite tracking mechanism: previous code uses bind map embedded
> >      in allocator structure. Unfortunately this wasn't good idea
> >      - for xe binding everything was fine, but it regress multiprocess/
> >      multithreaded allocations. Main reason of this was children
> >      processes couldn't get its reference as this memory was allocated
> >      on allocator thread (separate process). Currently each child
> >      contains its own separate tracking maps for ahnd and for each
> >      ahnd bind map.
> > v3: - Don't use priv field (we may clean bind_map apriori as failing
> >        asserts and outdated bind_map is not a problem anymore).
> >      - Use only vm on tracking ahnd (Karolina)
> >      - Add braces to avoid checkpatch complaining (Karolina)
> >      - Use igt_assert_f() instead conditional block (Karolina)
> > ---
> >   lib/igt_core.c        |   5 +
> >   lib/intel_allocator.c | 259 +++++++++++++++++++++++++++++++++++++++++-
> >   lib/intel_allocator.h |   3 +
> >   3 files changed, 265 insertions(+), 2 deletions(-)
> > 
> > diff --git a/lib/igt_core.c b/lib/igt_core.c
> > index 2ae2cb6883..041e4b3288 100644
> > --- a/lib/igt_core.c
> > +++ b/lib/igt_core.c
> > @@ -74,6 +74,7 @@
> >   #include "igt_sysrq.h"
> >   #include "igt_rc.h"
> >   #include "igt_list.h"
> > +#include "igt_map.h"
> >   #include "igt_device_scan.h"
> >   #include "igt_thread.h"
> >   #include "runnercomms.h"
> > @@ -320,6 +321,8 @@ bool test_multi_fork_child;
> >   /* For allocator purposes */
> >   pid_t child_pid  = -1;
> >   __thread pid_t child_tid  = -1;
> > +struct igt_map *ahnd_map;
> > +pthread_mutex_t ahnd_map_mutex;
> >   enum {
> >   	/*
> > @@ -2523,6 +2526,8 @@ bool __igt_fork(void)
> >   	case 0:
> >   		test_child = true;
> >   		pthread_mutex_init(&print_mutex, NULL);
> > +		pthread_mutex_init(&ahnd_map_mutex, NULL);
> > +		ahnd_map = igt_map_create(igt_map_hash_64, igt_map_equal_64);
> >   		child_pid = getpid();
> >   		child_tid = -1;
> >   		exit_handler_count = 0;
> > diff --git a/lib/intel_allocator.c b/lib/intel_allocator.c
> > index cf964ccc41..f0a9b7fb53 100644
> > --- a/lib/intel_allocator.c
> > +++ b/lib/intel_allocator.c
> > @@ -17,8 +17,8 @@
> >   #include "intel_allocator.h"
> >   #include "intel_allocator_msgchannel.h"
> >   #include "xe/xe_query.h"
> > +#include "xe/xe_util.h"
> > -//#define ALLOCDBG
> >   #ifdef ALLOCDBG
> >   #define alloc_info igt_info
> >   #define alloc_debug igt_debug
> > @@ -45,6 +45,14 @@ static inline const char *reqstr(enum reqtype request_type)
> >   #define alloc_debug(...) {}
> >   #endif
> > +#ifdef ALLOCBINDDBG
> > +#define bind_info igt_info
> > +#define bind_debug igt_debug
> > +#else
> > +#define bind_info(...) {}
> > +#define bind_debug(...) {}
> > +#endif
> > +
> >   /*
> >    * We limit allocator space to avoid hang when batch would be
> >    * pinned in the last page.
> > @@ -64,6 +72,30 @@ struct handle_entry {
> >   	struct allocator *al;
> >   };
> > +/* For tracking alloc()/free() for Xe */
> > +struct ahnd_info {
> > +	int fd;
> > +	uint64_t ahnd;
> > +	uint32_t vm;
> > +	enum intel_driver driver;
> > +	struct igt_map *bind_map;
> > +	pthread_mutex_t bind_map_mutex;
> > +};
> > +
> > +enum allocator_bind_op {
> > +	BOUND,
> > +	TO_BIND,
> > +	TO_UNBIND,
> > +};
> > +
> > +struct allocator_object {
> > +	uint32_t handle;
> > +	uint64_t offset;
> > +	uint64_t size;
> > +
> > +	enum allocator_bind_op bind_op;
> > +};
> > +
> >   struct intel_allocator *
> >   intel_allocator_reloc_create(int fd, uint64_t start, uint64_t end);
> >   struct intel_allocator *
> > @@ -122,6 +154,13 @@ static pid_t allocator_pid = -1;
> >   extern pid_t child_pid;
> >   extern __thread pid_t child_tid;
> > +/*
> > + * Track alloc()/free() requires storing in local process which has
> > + * an access to real drm fd it can work on.
> > + */
> > +extern struct igt_map *ahnd_map;
> > +extern pthread_mutex_t ahnd_map_mutex;
> > +
> >   /*
> >    * - for parent process we have child_pid == -1
> >    * - for child which calls intel_allocator_init() allocator_pid == child_pid
> > @@ -837,6 +876,45 @@ void intel_allocator_multiprocess_stop(void)
> >   	}
> >   }
> > +static void track_ahnd(int fd, uint64_t ahnd, uint32_t vm)
> > +{
> > +	struct ahnd_info *ainfo;
> > +
> > +	pthread_mutex_lock(&ahnd_map_mutex);
> > +	ainfo = igt_map_search(ahnd_map, &ahnd);
> > +	if (!ainfo) {
> > +		ainfo = malloc(sizeof(*ainfo));
> > +		ainfo->fd = fd;
> > +		ainfo->ahnd = ahnd;
> > +		ainfo->vm = vm;
> > +		ainfo->driver = get_intel_driver(fd);
> > +		ainfo->bind_map = igt_map_create(igt_map_hash_32, igt_map_equal_32);
> > +		pthread_mutex_init(&ainfo->bind_map_mutex, NULL);
> > +		bind_debug("[TRACK AHND] pid: %d, tid: %d, create <fd: %d, "
> > +			   "ahnd: %llx, vm: %u, driver: %d, ahnd_map: %p, bind_map: %p>\n",
> > +			   getpid(), gettid(), ainfo->fd,
> > +			   (long long)ainfo->ahnd, ainfo->vm,
> > +			   ainfo->driver, ahnd_map, ainfo->bind_map);
> > +		igt_map_insert(ahnd_map, &ainfo->ahnd, ainfo);
> > +	}
> > +
> > +	pthread_mutex_unlock(&ahnd_map_mutex);
> > +}
> > +
> > +static void untrack_ahnd(uint64_t ahnd)
> > +{
> > +	struct ahnd_info *ainfo;
> > +
> > +	pthread_mutex_lock(&ahnd_map_mutex);
> > +	ainfo = igt_map_search(ahnd_map, &ahnd);
> > +	if (ainfo) {
> > +		bind_debug("[UNTRACK AHND]: pid: %d, tid: %d, removing ahnd: %llx\n",
> > +			   getpid(), gettid(), (long long)ahnd);
> > +		igt_map_remove(ahnd_map, &ahnd, map_entry_free_func);
> > +	}
> > +	pthread_mutex_unlock(&ahnd_map_mutex);
> > +}
> > +
> >   static uint64_t __intel_allocator_open_full(int fd, uint32_t ctx,
> >   					    uint32_t vm,
> >   					    uint64_t start, uint64_t end,
> > @@ -895,6 +973,12 @@ static uint64_t __intel_allocator_open_full(int fd, uint32_t ctx,
> >   	igt_assert(resp.open.allocator_handle);
> >   	igt_assert(resp.response_type == RESP_OPEN);
> > +	/*
> > +	 * Igts mostly uses ctx as id when opening the allocator (i915 legacy).
> > +	 * If ctx is passed let's use it as an vm id, otherwise use vm.
> > +	 */
> > +	track_ahnd(fd, resp.open.allocator_handle, ctx ?: vm);
> > +
> >   	return resp.open.allocator_handle;
> >   }
> > @@ -1001,6 +1085,8 @@ bool intel_allocator_close(uint64_t allocator_handle)
> >   	igt_assert(handle_request(&req, &resp) == 0);
> >   	igt_assert(resp.response_type == RESP_CLOSE);
> > +	untrack_ahnd(allocator_handle);
> > +
> >   	return resp.close.is_empty;
> >   }
> > @@ -1034,6 +1120,74 @@ void intel_allocator_get_address_range(uint64_t allocator_handle,
> >   		*endp = resp.address_range.end;
> >   }
> > +static bool is_same(struct allocator_object *obj,
> > +		    uint32_t handle, uint64_t offset, uint64_t size,
> > +		    enum allocator_bind_op bind_op)
> > +{
> > +	return obj->handle == handle &&	obj->offset == offset && obj->size == size &&
> > +	       (obj->bind_op == bind_op || obj->bind_op == BOUND);
> > +}
> > +
> > +static void track_object(uint64_t allocator_handle, uint32_t handle,
> > +			 uint64_t offset, uint64_t size,
> > +			 enum allocator_bind_op bind_op)
> > +{
> > +	struct ahnd_info *ainfo;
> > +	struct allocator_object *obj;
> > +
> > +	bind_debug("[TRACK OBJECT]: [%s] pid: %d, tid: %d, ahnd: %llx, handle: %u, offset: %llx, size: %llx\n",
> > +		   bind_op == TO_BIND ? "BIND" : "UNBIND",
> > +		   getpid(), gettid(),
> > +		   (long long)allocator_handle,
> > +		   handle, (long long)offset, (long long)size);
> > +
> > +	if (offset == ALLOC_INVALID_ADDRESS) {
> > +		bind_debug("[TRACK OBJECT] => invalid address %llx, skipping tracking\n",
> > +			   (long long)offset);
> > +		return;
> > +	}
> > +
> > +	pthread_mutex_lock(&ahnd_map_mutex);
> > +	ainfo = igt_map_search(ahnd_map, &allocator_handle);
> > +	pthread_mutex_unlock(&ahnd_map_mutex);
> > +	igt_assert_f(ainfo, "[TRACK OBJECT] => MISSING ahnd %llx <=\n",
> > +		     (long long)allocator_handle);
> > +
> > +	if (ainfo->driver == INTEL_DRIVER_I915)
> > +		return; /* no-op for i915, at least for now */
> > +
> > +	pthread_mutex_lock(&ainfo->bind_map_mutex);
> > +	obj = igt_map_search(ainfo->bind_map, &handle);
> > +	if (obj) {
> > +		/*
> > +		 * User may call alloc() couple of times, check object is the
> > +		 * same. For free() there's simple case, just remove from
> > +		 * bind_map.
> > +		 */
> > +		if (bind_op == TO_BIND) {
> > +			igt_assert_eq(is_same(obj, handle, offset, size, bind_op), true);
> > +		} else if (bind_op == TO_UNBIND) {
> > +			if (obj->bind_op == TO_BIND)
> > +				igt_map_remove(ainfo->bind_map, &obj->handle, map_entry_free_func);
> > +			else if (obj->bind_op == BOUND)
> > +				obj->bind_op = bind_op;
> > +		}
> > +	} else {
> > +		/* Ignore to unbind bo which wasn't previously inserted */
> > +		if (bind_op == TO_UNBIND)
> > +			goto out;
> > +
> > +		obj = calloc(1, sizeof(*obj));
> > +		obj->handle = handle;
> > +		obj->offset = offset;
> > +		obj->size = size;
> > +		obj->bind_op = bind_op;
> > +		igt_map_insert(ainfo->bind_map, &obj->handle, obj);
> > +	}
> > +out:
> > +	pthread_mutex_unlock(&ainfo->bind_map_mutex);
> > +}
> > +
> >   /**
> >    * __intel_allocator_alloc:
> >    * @allocator_handle: handle to an allocator
> > @@ -1065,6 +1219,8 @@ uint64_t __intel_allocator_alloc(uint64_t allocator_handle, uint32_t handle,
> >   	igt_assert(handle_request(&req, &resp) == 0);
> >   	igt_assert(resp.response_type == RESP_ALLOC);
> > +	track_object(allocator_handle, handle, resp.alloc.offset, size, TO_BIND);
> > +
> >   	return resp.alloc.offset;
> >   }
> > @@ -1142,6 +1298,8 @@ bool intel_allocator_free(uint64_t allocator_handle, uint32_t handle)
> >   	igt_assert(handle_request(&req, &resp) == 0);
> >   	igt_assert(resp.response_type == RESP_FREE);
> > +	track_object(allocator_handle, handle, 0, 0, TO_UNBIND);
> > +
> >   	return resp.free.freed;
> >   }
> > @@ -1326,6 +1484,84 @@ void intel_allocator_print(uint64_t allocator_handle)
> >   	}
> >   }
> > +static void __xe_op_bind(struct ahnd_info *ainfo, uint32_t sync_in, uint32_t sync_out)
> > +{
> > +	struct allocator_object *obj;
> > +	struct igt_map_entry *pos;
> > +	struct igt_list_head obj_list;
> > +	struct xe_object *entry, *tmp;
> > +
> > +	IGT_INIT_LIST_HEAD(&obj_list);
> > +
> > +	pthread_mutex_lock(&ainfo->bind_map_mutex);
> > +	igt_map_foreach(ainfo->bind_map, pos) {
> > +		obj = pos->data;
> > +
> > +		if (obj->bind_op == BOUND)
> > +			continue;
> > +
> > +		bind_info("= [vm: %u] %s => %u %lx %lx\n",
> > +			  ainfo->vm,
> > +			  obj->bind_op == TO_BIND ? "TO BIND" : "TO UNBIND",
> > +			  obj->handle, obj->offset,
> > +			  obj->size);
> > +
> > +		entry = malloc(sizeof(*entry));
> > +		entry->handle = obj->handle;
> > +		entry->offset = obj->offset;
> > +		entry->size = obj->size;
> > +		entry->bind_op = obj->bind_op == TO_BIND ? XE_OBJECT_BIND :
> > +							   XE_OBJECT_UNBIND;
> > +		igt_list_add(&entry->link, &obj_list);
> > +
> > +		/*
> > +		 * We clean bind_map even before calling bind/unbind
> > +		 * as all binding operations asserts in case of error.
> > +		 */
> > +		if (obj->bind_op == TO_BIND)
> > +			obj->bind_op = BOUND;
> > +		else
> > +			igt_map_remove(ainfo->bind_map, &obj->handle,
> > +				       map_entry_free_func);
> 
> I don't quite understand why we can clean up the map even before the actual
> bind/unbind happens. If an assert fails, it means that vm_(un)bind failed,
> so the actual operation didn't succeed, and that state change is incorrect.
> I believe that there's a reason we can do it this way, I just don't
> understand it.

I allocate separate list of xe_objects which is an argument to
xe_bind_unbind_async(). As list is detached (separate) from the bind_map
I may apriori do cleanup in the map. If bind/unbind is unsuccessful
outdated bind_map is the smallest problem here. On assert path igt_fail()
calls intel_allocator_init() which just frees contents on ahnd_map[]->bind_map.
In children there's simpler because they just dies. Real problem is we
still have dangling vm (no xe_vm_destroy()).
So I decided to update the bind_map before calling bind/unbind as on failure
doesn't matter what it will contain.

--
Zbigniew

> 
> Apart from that one thing, the patch looks good to me.
> 
> All the best,
> Karolina
> 
> > +	}
> > +	pthread_mutex_unlock(&ainfo->bind_map_mutex);
> > +
> > +	xe_bind_unbind_async(ainfo->fd, ainfo->vm, 0, &obj_list, sync_in, sync_out);
> > +
> > +	igt_list_for_each_entry_safe(entry, tmp, &obj_list, link) {
> > +		igt_list_del(&entry->link);
> > +		free(entry);
> > +	}
> > +}
> > +
> > +/**
> > + * intel_allocator_bind:
> > + * @allocator_handle: handle to an allocator
> > + * @sync_in: syncobj (fence-in)
> > + * @sync_out: syncobj (fence-out)
> > + *
> > + * Function binds and unbinds all objects added to the allocator which weren't
> > + * previously binded/unbinded.
> > + *
> > + **/
> > +void intel_allocator_bind(uint64_t allocator_handle,
> > +			  uint32_t sync_in, uint32_t sync_out)
> > +{
> > +	struct ahnd_info *ainfo;
> > +
> > +	pthread_mutex_lock(&ahnd_map_mutex);
> > +	ainfo = igt_map_search(ahnd_map, &allocator_handle);
> > +	pthread_mutex_unlock(&ahnd_map_mutex);
> > +	igt_assert(ainfo);
> > +
> > +	/*
> > +	 * We collect bind/unbind operations on alloc()/free() to do group
> > +	 * operation getting @sync_in as syncobj handle (fence-in). If user
> > +	 * passes 0 as @sync_out we bind/unbind synchronously.
> > +	 */
> > +	__xe_op_bind(ainfo, sync_in, sync_out);
> > +}
> > +
> >   static int equal_handles(const void *key1, const void *key2)
> >   {
> >   	const struct handle_entry *h1 = key1, *h2 = key2;
> > @@ -1383,6 +1619,22 @@ static void __free_maps(struct igt_map *map, bool close_allocators)
> >   	igt_map_destroy(map, map_entry_free_func);
> >   }
> > +static void __free_ahnd_map(void)
> > +{
> > +	struct igt_map_entry *pos;
> > +	struct ahnd_info *ainfo;
> > +
> > +	if (!ahnd_map)
> > +		return;
> > +
> > +	igt_map_foreach(ahnd_map, pos) {
> > +		ainfo = pos->data;
> > +		igt_map_destroy(ainfo->bind_map, map_entry_free_func);
> > +	}
> > +
> > +	igt_map_destroy(ahnd_map, map_entry_free_func);
> > +}
> > +
> >   /**
> >    * intel_allocator_init:
> >    *
> > @@ -1400,12 +1652,15 @@ void intel_allocator_init(void)
> >   	__free_maps(handles, true);
> >   	__free_maps(ctx_map, false);
> >   	__free_maps(vm_map, false);
> > +	__free_ahnd_map();
> >   	atomic_init(&next_handle, 1);
> >   	handles = igt_map_create(hash_handles, equal_handles);
> >   	ctx_map = igt_map_create(hash_instance, equal_ctx);
> >   	vm_map = igt_map_create(hash_instance, equal_vm);
> > -	igt_assert(handles && ctx_map && vm_map);
> > +	pthread_mutex_init(&ahnd_map_mutex, NULL);
> > +	ahnd_map = igt_map_create(igt_map_hash_64, igt_map_equal_64);
> > +	igt_assert(handles && ctx_map && vm_map && ahnd_map);
> >   	channel = intel_allocator_get_msgchannel(CHANNEL_SYSVIPC_MSGQUEUE);
> >   }
> > diff --git a/lib/intel_allocator.h b/lib/intel_allocator.h
> > index 3ec74f6191..f9ff7f1cc9 100644
> > --- a/lib/intel_allocator.h
> > +++ b/lib/intel_allocator.h
> > @@ -210,6 +210,9 @@ bool intel_allocator_reserve_if_not_allocated(uint64_t allocator_handle,
> >   void intel_allocator_print(uint64_t allocator_handle);
> > +void intel_allocator_bind(uint64_t allocator_handle,
> > +			  uint32_t sync_in, uint32_t sync_out);
> > +
> >   #define ALLOC_INVALID_ADDRESS (-1ull)
> >   #define INTEL_ALLOCATOR_NONE   0
> >   #define INTEL_ALLOCATOR_RELOC  1

Re: [igt-dev] [PATCH i-g-t v3 13/17] lib/intel_blt: Extend blitter library to support xe driver

[Zbigniew Kempczyński]

On Wed, Jul 12, 2023 at 11:22:00AM +0200, Karolina Stolarek wrote:
> On 11.07.2023 13:20, Zbigniew Kempczyński wrote:
> > Use already written for i915 blitter library in xe development.
> > Add appropriate code paths which are unique for those drivers.
> > 
> > Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
> > ---
> > v3: Extract common helpers for getting mapping and alignment (Karolina)
> > ---
> >   lib/intel_blt.c | 235 ++++++++++++++++++++++++++++++------------------
> >   lib/intel_blt.h |   2 +-
> >   2 files changed, 150 insertions(+), 87 deletions(-)
> > 
> > diff --git a/lib/intel_blt.c b/lib/intel_blt.c
> > index f2f86e4947..5387a60ee3 100644
> > --- a/lib/intel_blt.c
> > +++ b/lib/intel_blt.c
> > @@ -9,9 +9,13 @@
> >   #include <malloc.h>
> >   #include <cairo.h>
> >   #include "drm.h"
> > -#include "igt.h"
> >   #include "i915/gem_create.h"
> > +#include "igt.h"
> > +#include "igt_syncobj.h"
> >   #include "intel_blt.h"
> > +#include "xe/xe_ioctl.h"
> > +#include "xe/xe_query.h"
> > +#include "xe/xe_util.h"
> >   #define BITRANGE(start, end) (end - start + 1)
> >   #define GET_CMDS_INFO(__fd) intel_get_cmds_info(intel_get_drm_devid(__fd))
> > @@ -468,24 +472,40 @@ static int __special_mode(const struct blt_copy_data *blt)
> >   	return SM_NONE;
> >   }
> > -static int __memory_type(uint32_t region)
> > +static int __memory_type(int fd, enum intel_driver driver, uint32_t region)
> >   {
> > -	igt_assert_f(IS_DEVICE_MEMORY_REGION(region) ||
> > -		     IS_SYSTEM_MEMORY_REGION(region),
> > -		     "Invalid region: %x\n", region);
> > +	if (driver == INTEL_DRIVER_I915) {
> > +		igt_assert_f(IS_DEVICE_MEMORY_REGION(region) ||
> > +			     IS_SYSTEM_MEMORY_REGION(region),
> > +			     "Invalid region: %x\n", region);
> > +	} else {
> > +		igt_assert_f(XE_IS_VRAM_MEMORY_REGION(fd, region) ||
> > +			     XE_IS_SYSMEM_MEMORY_REGION(fd, region),
> > +			     "Invalid region: %x\n", region);
> > +	}
> > -	if (IS_DEVICE_MEMORY_REGION(region))
> > +	if (driver == INTEL_DRIVER_I915 && IS_DEVICE_MEMORY_REGION(region))
> >   		return TM_LOCAL_MEM;
> > +	else if (driver == INTEL_DRIVER_XE && XE_IS_VRAM_MEMORY_REGION(fd, region))
> > +		return TM_LOCAL_MEM;
> > +
> >   	return TM_SYSTEM_MEM;
> >   }
> > -static enum blt_aux_mode __aux_mode(const struct blt_copy_object *obj)
> > +static enum blt_aux_mode __aux_mode(int fd,
> > +				    enum intel_driver driver,
> > +				    const struct blt_copy_object *obj)
> >   {
> > -	if (obj->compression == COMPRESSION_ENABLED) {
> > +	if (driver == INTEL_DRIVER_I915 && obj->compression == COMPRESSION_ENABLED) {
> >   		igt_assert_f(IS_DEVICE_MEMORY_REGION(obj->region),
> >   			     "XY_BLOCK_COPY_BLT supports compression "
> >   			     "on device memory only\n");
> >   		return AM_AUX_CCS_E;
> > +	} else if (driver == INTEL_DRIVER_XE && obj->compression == COMPRESSION_ENABLED) {
> > +		igt_assert_f(XE_IS_VRAM_MEMORY_REGION(fd, obj->region),
> > +			     "XY_BLOCK_COPY_BLT supports compression "
> > +			     "on device memory only\n");
> > +		return AM_AUX_CCS_E;
> >   	}
> >   	return AM_AUX_NONE;
> > @@ -508,9 +528,9 @@ static void fill_data(struct gen12_block_copy_data *data,
> >   	data->dw00.length = extended_command ? 20 : 10;
> >   	if (__special_mode(blt) == SM_FULL_RESOLVE)
> > -		data->dw01.dst_aux_mode = __aux_mode(&blt->src);
> > +		data->dw01.dst_aux_mode = __aux_mode(blt->fd, blt->driver, &blt->src);
> >   	else
> > -		data->dw01.dst_aux_mode = __aux_mode(&blt->dst);
> > +		data->dw01.dst_aux_mode = __aux_mode(blt->fd, blt->driver, &blt->dst);
> >   	data->dw01.dst_pitch = blt->dst.pitch - 1;
> >   	data->dw01.dst_mocs = blt->dst.mocs;
> > @@ -531,13 +551,13 @@ static void fill_data(struct gen12_block_copy_data *data,
> >   	data->dw06.dst_x_offset = blt->dst.x_offset;
> >   	data->dw06.dst_y_offset = blt->dst.y_offset;
> > -	data->dw06.dst_target_memory = __memory_type(blt->dst.region);
> > +	data->dw06.dst_target_memory = __memory_type(blt->fd, blt->driver, blt->dst.region);
> >   	data->dw07.src_x1 = blt->src.x1;
> >   	data->dw07.src_y1 = blt->src.y1;
> >   	data->dw08.src_pitch = blt->src.pitch - 1;
> > -	data->dw08.src_aux_mode = __aux_mode(&blt->src);
> > +	data->dw08.src_aux_mode = __aux_mode(blt->fd, blt->driver, &blt->src);
> >   	data->dw08.src_mocs = blt->src.mocs;
> >   	data->dw08.src_compression = blt->src.compression;
> >   	data->dw08.src_tiling = __block_tiling(blt->src.tiling);
> > @@ -550,7 +570,7 @@ static void fill_data(struct gen12_block_copy_data *data,
> >   	data->dw11.src_x_offset = blt->src.x_offset;
> >   	data->dw11.src_y_offset = blt->src.y_offset;
> > -	data->dw11.src_target_memory = __memory_type(blt->src.region);
> > +	data->dw11.src_target_memory = __memory_type(blt->fd, blt->driver, blt->src.region);
> >   }
> >   static void fill_data_ext(struct gen12_block_copy_data_ext *dext,
> > @@ -692,6 +712,24 @@ static void dump_bb_ext(struct gen12_block_copy_data_ext *data)
> >   		 data->dw21.src_array_index);
> >   }
> > +static uint64_t get_default_alignment(int fd, enum intel_driver driver)
> > +{
> > +	if (driver == INTEL_DRIVER_XE)
> > +		return xe_get_default_alignment(fd);
> > +
> > +	return gem_detect_safe_alignment(fd);
> > +}
> > +
> > +static void *bo_map(int fd, uint32_t handle, uint64_t size,
> > +		    enum intel_driver driver)
> > +{
> > +	if (driver == INTEL_DRIVER_XE)
> > +		return xe_bo_map(fd, handle, size);
> > +
> > +	return gem_mmap__device_coherent(fd, handle, 0, size,
> > +					 PROT_READ | PROT_WRITE);
> > +}
> > +
> >   /**
> >    * blt_copy_init:
> >    * @fd: drm fd
> > @@ -739,7 +777,7 @@ uint64_t emit_blt_block_copy(int fd,
> >   	igt_assert_f(ahnd, "block-copy supports softpin only\n");
> >   	igt_assert_f(blt, "block-copy requires data to do blit\n");
> > -	alignment = gem_detect_safe_alignment(fd);
> > +	alignment = get_default_alignment(fd, blt->driver);
> >   	src_offset = get_offset(ahnd, blt->src.handle, blt->src.size, alignment)
> >   		     + blt->src.plane_offset;
> >   	dst_offset = get_offset(ahnd, blt->dst.handle, blt->dst.size, alignment)
> > @@ -748,8 +786,7 @@ uint64_t emit_blt_block_copy(int fd,
> >   	fill_data(&data, blt, src_offset, dst_offset, ext);
> > -	bb = gem_mmap__device_coherent(fd, blt->bb.handle, 0, blt->bb.size,
> > -				       PROT_READ | PROT_WRITE);
> > +	bb = bo_map(fd, blt->bb.handle, blt->bb.size, blt->driver);
> >   	igt_assert(bb_pos + sizeof(data) < blt->bb.size);
> >   	memcpy(bb + bb_pos, &data, sizeof(data));
> > @@ -812,29 +849,35 @@ int blt_block_copy(int fd,
> >   	igt_assert_f(ahnd, "block-copy supports softpin only\n");
> >   	igt_assert_f(blt, "block-copy requires data to do blit\n");
> > +	igt_assert_neq(blt->driver, 0);
> > -	alignment = gem_detect_safe_alignment(fd);
> > +	alignment = get_default_alignment(fd, blt->driver);
> >   	src_offset = get_offset(ahnd, blt->src.handle, blt->src.size, alignment);
> >   	dst_offset = get_offset(ahnd, blt->dst.handle, blt->dst.size, alignment);
> >   	bb_offset = get_offset(ahnd, blt->bb.handle, blt->bb.size, alignment);
> >   	emit_blt_block_copy(fd, ahnd, blt, ext, 0, true);
> > -	obj[0].offset = CANONICAL(dst_offset);
> > -	obj[1].offset = CANONICAL(src_offset);
> > -	obj[2].offset = CANONICAL(bb_offset);
> > -	obj[0].handle = blt->dst.handle;
> > -	obj[1].handle = blt->src.handle;
> > -	obj[2].handle = blt->bb.handle;
> > -	obj[0].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_WRITE |
> > -		       EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> > -	obj[1].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> > -	obj[2].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> > -	execbuf.buffer_count = 3;
> > -	execbuf.buffers_ptr = to_user_pointer(obj);
> > -	execbuf.rsvd1 = ctx ? ctx->id : 0;
> > -	execbuf.flags = e ? e->flags : I915_EXEC_BLT;
> > -	ret = __gem_execbuf(fd, &execbuf);
> > +	if (blt->driver == INTEL_DRIVER_XE) {
> > +		intel_ctx_xe_exec(ctx, ahnd, CANONICAL(bb_offset));
> > +	} else {
> > +		obj[0].offset = CANONICAL(dst_offset);
> > +		obj[1].offset = CANONICAL(src_offset);
> > +		obj[2].offset = CANONICAL(bb_offset);
> > +		obj[0].handle = blt->dst.handle;
> > +		obj[1].handle = blt->src.handle;
> > +		obj[2].handle = blt->bb.handle;
> > +		obj[0].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_WRITE |
> > +				EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> > +		obj[1].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> > +		obj[2].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> > +		execbuf.buffer_count = 3;
> > +		execbuf.buffers_ptr = to_user_pointer(obj);
> > +		execbuf.rsvd1 = ctx ? ctx->id : 0;
> > +		execbuf.flags = e ? e->flags : I915_EXEC_BLT;
> > +
> > +		ret = __gem_execbuf(fd, &execbuf);
> > +	}
> >   	return ret;
> >   }
> > @@ -950,7 +993,7 @@ uint64_t emit_blt_ctrl_surf_copy(int fd,
> >   	igt_assert_f(ahnd, "ctrl-surf-copy supports softpin only\n");
> >   	igt_assert_f(surf, "ctrl-surf-copy requires data to do ctrl-surf-copy blit\n");
> > -	alignment = max_t(uint64_t, gem_detect_safe_alignment(fd), 1ull << 16);
> > +	alignment = max_t(uint64_t, get_default_alignment(fd, surf->driver), 1ull << 16);
> >   	data.dw00.client = 0x2;
> >   	data.dw00.opcode = 0x48;
> > @@ -973,8 +1016,7 @@ uint64_t emit_blt_ctrl_surf_copy(int fd,
> >   	data.dw04.dst_address_hi = dst_offset >> 32;
> >   	data.dw04.dst_mocs = surf->dst.mocs;
> > -	bb = gem_mmap__device_coherent(fd, surf->bb.handle, 0, surf->bb.size,
> > -				       PROT_READ | PROT_WRITE);
> > +	bb = bo_map(fd, surf->bb.handle, surf->bb.size, surf->driver);
> >   	igt_assert(bb_pos + sizeof(data) < surf->bb.size);
> >   	memcpy(bb + bb_pos, &data, sizeof(data));
> > @@ -1002,7 +1044,7 @@ uint64_t emit_blt_ctrl_surf_copy(int fd,
> >   /**
> >    * blt_ctrl_surf_copy:
> > - * @fd: drm fd
> > + * @blt: bldrm fd
> 
> Emm, the argument seem to still be fd, and there's no blt. A typo?

I'm not sure how this appeared here. Will fix and send in v4.

> 
> Thanks for adding the helpers, the code looks nice and tidy (although I'm
> scared of __special_mode and the like, but there's no better way to do it).

Yes, I also don't see better way to handle two different drivers here.
But I believe conditional is readable and easily understandable.

--
Zbigniew

> 
> All the best,
> Karolina
> 
> >    * @ctx: intel_ctx_t context
> >    * @e: blitter engine for @ctx
> >    * @ahnd: allocator handle
> > @@ -1026,32 +1068,37 @@ int blt_ctrl_surf_copy(int fd,
> >   	igt_assert_f(ahnd, "ctrl-surf-copy supports softpin only\n");
> >   	igt_assert_f(surf, "ctrl-surf-copy requires data to do ctrl-surf-copy blit\n");
> > +	igt_assert_neq(surf->driver, 0);
> > -	alignment = max_t(uint64_t, gem_detect_safe_alignment(fd), 1ull << 16);
> > +	alignment = max_t(uint64_t, get_default_alignment(fd, surf->driver), 1ull << 16);
> >   	src_offset = get_offset(ahnd, surf->src.handle, surf->src.size, alignment);
> >   	dst_offset = get_offset(ahnd, surf->dst.handle, surf->dst.size, alignment);
> >   	bb_offset = get_offset(ahnd, surf->bb.handle, surf->bb.size, alignment);
> >   	emit_blt_ctrl_surf_copy(fd, ahnd, surf, 0, true);
> > -	obj[0].offset = CANONICAL(dst_offset);
> > -	obj[1].offset = CANONICAL(src_offset);
> > -	obj[2].offset = CANONICAL(bb_offset);
> > -	obj[0].handle = surf->dst.handle;
> > -	obj[1].handle = surf->src.handle;
> > -	obj[2].handle = surf->bb.handle;
> > -	obj[0].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_WRITE |
> > -		       EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> > -	obj[1].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> > -	obj[2].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> > -	execbuf.buffer_count = 3;
> > -	execbuf.buffers_ptr = to_user_pointer(obj);
> > -	execbuf.flags = e ? e->flags : I915_EXEC_BLT;
> > -	execbuf.rsvd1 = ctx ? ctx->id : 0;
> > -	gem_execbuf(fd, &execbuf);
> > -	put_offset(ahnd, surf->dst.handle);
> > -	put_offset(ahnd, surf->src.handle);
> > -	put_offset(ahnd, surf->bb.handle);
> > +	if (surf->driver == INTEL_DRIVER_XE) {
> > +		intel_ctx_xe_exec(ctx, ahnd, CANONICAL(bb_offset));
> > +	} else {
> > +		obj[0].offset = CANONICAL(dst_offset);
> > +		obj[1].offset = CANONICAL(src_offset);
> > +		obj[2].offset = CANONICAL(bb_offset);
> > +		obj[0].handle = surf->dst.handle;
> > +		obj[1].handle = surf->src.handle;
> > +		obj[2].handle = surf->bb.handle;
> > +		obj[0].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_WRITE |
> > +				EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> > +		obj[1].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> > +		obj[2].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> > +		execbuf.buffer_count = 3;
> > +		execbuf.buffers_ptr = to_user_pointer(obj);
> > +		execbuf.flags = e ? e->flags : I915_EXEC_BLT;
> > +		execbuf.rsvd1 = ctx ? ctx->id : 0;
> > +		gem_execbuf(fd, &execbuf);
> > +		put_offset(ahnd, surf->dst.handle);
> > +		put_offset(ahnd, surf->src.handle);
> > +		put_offset(ahnd, surf->bb.handle);
> > +	}
> >   	return 0;
> >   }
> > @@ -1208,7 +1255,7 @@ uint64_t emit_blt_fast_copy(int fd,
> >   	uint32_t bbe = MI_BATCH_BUFFER_END;
> >   	uint32_t *bb;
> > -	alignment = gem_detect_safe_alignment(fd);
> > +	alignment = get_default_alignment(fd, blt->driver);
> >   	data.dw00.client = 0x2;
> >   	data.dw00.opcode = 0x42;
> > @@ -1218,8 +1265,8 @@ uint64_t emit_blt_fast_copy(int fd,
> >   	data.dw01.dst_pitch = blt->dst.pitch;
> >   	data.dw01.color_depth = __fast_color_depth(blt->color_depth);
> > -	data.dw01.dst_memory = __memory_type(blt->dst.region);
> > -	data.dw01.src_memory = __memory_type(blt->src.region);
> > +	data.dw01.dst_memory = __memory_type(blt->fd, blt->driver, blt->dst.region);
> > +	data.dw01.src_memory = __memory_type(blt->fd, blt->driver, blt->src.region);
> >   	data.dw01.dst_type_y = __new_tile_y_type(blt->dst.tiling) ? 1 : 0;
> >   	data.dw01.src_type_y = __new_tile_y_type(blt->src.tiling) ? 1 : 0;
> > @@ -1246,8 +1293,7 @@ uint64_t emit_blt_fast_copy(int fd,
> >   	data.dw08.src_address_lo = src_offset;
> >   	data.dw09.src_address_hi = src_offset >> 32;
> > -	bb = gem_mmap__device_coherent(fd, blt->bb.handle, 0, blt->bb.size,
> > -				       PROT_READ | PROT_WRITE);
> > +	bb = bo_map(fd, blt->bb.handle, blt->bb.size, blt->driver);
> >   	igt_assert(bb_pos + sizeof(data) < blt->bb.size);
> >   	memcpy(bb + bb_pos, &data, sizeof(data));
> > @@ -1297,32 +1343,39 @@ int blt_fast_copy(int fd,
> >   	uint64_t dst_offset, src_offset, bb_offset, alignment;
> >   	int ret;
> > -	alignment = gem_detect_safe_alignment(fd);
> > +	igt_assert_f(ahnd, "fast-copy supports softpin only\n");
> > +	igt_assert_f(blt, "fast-copy requires data to do fast-copy blit\n");
> > +	igt_assert_neq(blt->driver, 0);
> > +	alignment = get_default_alignment(fd, blt->driver);
> >   	src_offset = get_offset(ahnd, blt->src.handle, blt->src.size, alignment);
> >   	dst_offset = get_offset(ahnd, blt->dst.handle, blt->dst.size, alignment);
> >   	bb_offset = get_offset(ahnd, blt->bb.handle, blt->bb.size, alignment);
> >   	emit_blt_fast_copy(fd, ahnd, blt, 0, true);
> > -	obj[0].offset = CANONICAL(dst_offset);
> > -	obj[1].offset = CANONICAL(src_offset);
> > -	obj[2].offset = CANONICAL(bb_offset);
> > -	obj[0].handle = blt->dst.handle;
> > -	obj[1].handle = blt->src.handle;
> > -	obj[2].handle = blt->bb.handle;
> > -	obj[0].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_WRITE |
> > -		       EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> > -	obj[1].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> > -	obj[2].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> > -	execbuf.buffer_count = 3;
> > -	execbuf.buffers_ptr = to_user_pointer(obj);
> > -	execbuf.rsvd1 = ctx ? ctx->id : 0;
> > -	execbuf.flags = e ? e->flags : I915_EXEC_BLT;
> > -	ret = __gem_execbuf(fd, &execbuf);
> > -	put_offset(ahnd, blt->dst.handle);
> > -	put_offset(ahnd, blt->src.handle);
> > -	put_offset(ahnd, blt->bb.handle);
> > +	if (blt->driver == INTEL_DRIVER_XE) {
> > +		intel_ctx_xe_exec(ctx, ahnd, CANONICAL(bb_offset));
> > +	} else {
> > +		obj[0].offset = CANONICAL(dst_offset);
> > +		obj[1].offset = CANONICAL(src_offset);
> > +		obj[2].offset = CANONICAL(bb_offset);
> > +		obj[0].handle = blt->dst.handle;
> > +		obj[1].handle = blt->src.handle;
> > +		obj[2].handle = blt->bb.handle;
> > +		obj[0].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_WRITE |
> > +				EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> > +		obj[1].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> > +		obj[2].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
> > +		execbuf.buffer_count = 3;
> > +		execbuf.buffers_ptr = to_user_pointer(obj);
> > +		execbuf.rsvd1 = ctx ? ctx->id : 0;
> > +		execbuf.flags = e ? e->flags : I915_EXEC_BLT;
> > +		ret = __gem_execbuf(fd, &execbuf);
> > +		put_offset(ahnd, blt->dst.handle);
> > +		put_offset(ahnd, blt->src.handle);
> > +		put_offset(ahnd, blt->bb.handle);
> > +	}
> >   	return ret;
> >   }
> > @@ -1366,16 +1419,21 @@ blt_create_object(const struct blt_copy_data *blt, uint32_t region,
> >   	obj = calloc(1, sizeof(*obj));
> >   	obj->size = size;
> > -	igt_assert(__gem_create_in_memory_regions(blt->fd, &handle,
> > -						  &size, region) == 0);
> > +
> > +	if (blt->driver == INTEL_DRIVER_XE) {
> > +		size = ALIGN(size, xe_get_default_alignment(blt->fd));
> > +		handle = xe_bo_create_flags(blt->fd, 0, size, region);
> > +	} else {
> > +		igt_assert(__gem_create_in_memory_regions(blt->fd, &handle,
> > +							  &size, region) == 0);
> > +	}
> >   	blt_set_object(obj, handle, size, region, mocs, tiling,
> >   		       compression, compression_type);
> >   	blt_set_geom(obj, stride, 0, 0, width, height, 0, 0);
> >   	if (create_mapping)
> > -		obj->ptr = gem_mmap__device_coherent(blt->fd, handle, 0, size,
> > -						     PROT_READ | PROT_WRITE);
> > +		obj->ptr = bo_map(blt->fd, handle, size, blt->driver);
> >   	return obj;
> >   }
> > @@ -1518,14 +1576,19 @@ void blt_surface_to_png(int fd, uint32_t run_id, const char *fileid,
> >   	int format;
> >   	int stride = obj->tiling ? obj->pitch * 4 : obj->pitch;
> >   	char filename[FILENAME_MAX];
> > +	bool is_xe = is_xe_device(fd);
> >   	snprintf(filename, FILENAME_MAX-1, "%d-%s-%s-%ux%u-%s.png",
> >   		 run_id, fileid, blt_tiling_name(obj->tiling), width, height,
> >   		 obj->compression ? "compressed" : "uncompressed");
> > -	if (!map)
> > -		map = gem_mmap__device_coherent(fd, obj->handle, 0,
> > -						obj->size, PROT_READ);
> > +	if (!map) {
> > +		if (is_xe)
> > +			map = xe_bo_map(fd, obj->handle, obj->size);
> > +		else
> > +			map = gem_mmap__device_coherent(fd, obj->handle, 0,
> > +							obj->size, PROT_READ);
> > +	}
> >   	format = CAIRO_FORMAT_RGB24;
> >   	surface = cairo_image_surface_create_for_data(map,
> >   						      format, width, height,
> > diff --git a/lib/intel_blt.h b/lib/intel_blt.h
> > index 7516ce8ac7..944e2b4ae7 100644
> > --- a/lib/intel_blt.h
> > +++ b/lib/intel_blt.h
> > @@ -8,7 +8,7 @@
> >   /**
> >    * SECTION:intel_blt
> > - * @short_description: i915 blitter library
> > + * @short_description: i915/xe blitter library
> >    * @title: Blitter library
> >    * @include: intel_blt.h
> >    *

Re: [igt-dev] [PATCH i-g-t v3 14/17] tests/xe_ccs: Check if flatccs is working with block-copy for Xe

[Karolina Stolarek]

On 11.07.2023 13:20, Zbigniew Kempczyński wrote:
> This is ported to xe copy of i915 gem_ccs test. Ported means all driver
> dependent calls - like working on regions, binding and execution were
> replaced by xe counterparts. I wondered to add conditionals for xe
> in gem_ccs but this would decrease test readability so I dropped
> this idea.
> 
> Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
> ---
> v3: Drop unused macro, use system_memory() instead REGION_SMEM
>      in xe test (Karolina)
> ---
>   tests/meson.build |   1 +
>   tests/xe/xe_ccs.c | 757 ++++++++++++++++++++++++++++++++++++++++++++++
>   2 files changed, 758 insertions(+)
>   create mode 100644 tests/xe/xe_ccs.c
> 
> diff --git a/tests/meson.build b/tests/meson.build
> index d56e4c89da..da8653fabc 100644
> --- a/tests/meson.build
> +++ b/tests/meson.build
> @@ -263,6 +263,7 @@ i915_progs = [
>   ]
>   
>   xe_progs = [
> +	'xe_ccs',
>   	'xe_create',
>   	'xe_compute',
>   	'xe_dma_buf_sync',
> diff --git a/tests/xe/xe_ccs.c b/tests/xe/xe_ccs.c
> new file mode 100644
> index 0000000000..ee653790bf
> --- /dev/null
> +++ b/tests/xe/xe_ccs.c
> @@ -0,0 +1,757 @@
> +// SPDX-License-Identifier: MIT
> +/*
> + * Copyright © 2023 Intel Corporation
> + */
> +
> +#include <errno.h>
> +#include <glib.h>
> +#include <sys/ioctl.h>
> +#include <sys/time.h>
> +#include <malloc.h>
> +#include "drm.h"
> +#include "igt.h"
> +#include "igt_syncobj.h"
> +#include "intel_blt.h"
> +#include "intel_mocs.h"
> +#include "xe/xe_ioctl.h"
> +#include "xe/xe_query.h"
> +#include "xe/xe_util.h"
> +/**
> + * TEST: xe ccs
> + * Description: Exercise gen12 blitter with and without flatccs compression on Xe
> + * Run type: FULL
> + *
> + * SUBTEST: block-copy-compressed
> + * Description: Check block-copy flatccs compressed blit
> + *
> + * SUBTEST: block-copy-uncompressed
> + * Description: Check block-copy uncompressed blit
> + *
> + * SUBTEST: block-multicopy-compressed
> + * Description: Check block-multicopy flatccs compressed blit
> + *
> + * SUBTEST: block-multicopy-inplace
> + * Description: Check block-multicopy flatccs inplace decompression blit
> + *
> + * SUBTEST: ctrl-surf-copy
> + * Description: Check flatccs data can be copied from/to surface
> + *
> + * SUBTEST: ctrl-surf-copy-new-ctx
> + * Description: Check flatccs data are physically tagged and visible in vm
> + *
> + * SUBTEST: suspend-resume
> + * Description: Check flatccs data persists after suspend / resume (S0)
> + */
> +
> +IGT_TEST_DESCRIPTION("Exercise gen12 blitter with and without flatccs compression on Xe");
> +
> +static struct param {
> +	int compression_format;
> +	int tiling;
> +	bool write_png;
> +	bool print_bb;
> +	bool print_surface_info;
> +	int width;
> +	int height;
> +} param = {
> +	.compression_format = 0,
> +	.tiling = -1,
> +	.write_png = false,
> +	.print_bb = false,
> +	.print_surface_info = false,
> +	.width = 512,
> +	.height = 512,
> +};
> +
> +struct test_config {
> +	bool compression;
> +	bool inplace;
> +	bool surfcopy;
> +	bool new_ctx;
> +	bool suspend_resume;
> +};
> +
> +static void set_surf_object(struct blt_ctrl_surf_copy_object *obj,
> +			    uint32_t handle, uint32_t region, uint64_t size,
> +			    uint8_t mocs, enum blt_access_type access_type)
> +{
> +	obj->handle = handle;
> +	obj->region = region;
> +	obj->size = size;
> +	obj->mocs = mocs;
> +	obj->access_type = access_type;
> +}

My comment about extracting helpers was about this function, 
compare_nxn() and dump_corruption_info(). Do you think it would be 
possible to share them across xe_css and gem_ccs?

Other changes look good to me.

All the best,
Karolina

> +
> +#define PRINT_SURFACE_INFO(name, obj) do { \
> +	if (param.print_surface_info) \
> +		blt_surface_info((name), (obj)); } while (0)
> +
> +#define WRITE_PNG(fd, id, name, obj, w, h) do { \
> +	if (param.write_png) \
> +		blt_surface_to_png((fd), (id), (name), (obj), (w), (h)); } while (0)
> +
> +static int compare_nxn(const struct blt_copy_object *surf1,
> +		       const struct blt_copy_object *surf2,
> +		       int xsize, int ysize, int bx, int by)
> +{
> +	int x, y, corrupted;
> +	uint32_t pos, px1, px2;
> +
> +	corrupted = 0;
> +	for (y = 0; y < ysize; y++) {
> +		for (x = 0; x < xsize; x++) {
> +			pos = bx * xsize + by * ysize * surf1->pitch / 4;
> +			pos += x + y * surf1->pitch / 4;
> +			px1 = surf1->ptr[pos];
> +			px2 = surf2->ptr[pos];
> +			if (px1 != px2)
> +				corrupted++;
> +		}
> +	}
> +
> +	return corrupted;
> +}
> +
> +static void dump_corruption_info(const struct blt_copy_object *surf1,
> +				 const struct blt_copy_object *surf2)
> +{
> +	const int xsize = 8, ysize = 8;
> +	int w, h, bx, by, corrupted;
> +
> +	igt_assert(surf1->x1 == surf2->x1 && surf1->x2 == surf2->x2);
> +	igt_assert(surf1->y1 == surf2->y1 && surf1->y2 == surf2->y2);
> +	w = surf1->x2;
> +	h = surf1->y2;
> +
> +	igt_info("dump corruption - width: %d, height: %d, sizex: %x, sizey: %x\n",
> +		 surf1->x2, surf1->y2, xsize, ysize);
> +
> +	for (by = 0; by < h / ysize; by++) {
> +		for (bx = 0; bx < w / xsize; bx++) {
> +			corrupted = compare_nxn(surf1, surf2, xsize, ysize, bx, by);
> +			if (corrupted == 0)
> +				igt_info(".");
> +			else
> +				igt_info("%c", '0' + corrupted);
> +		}
> +		igt_info("\n");
> +	}
> +}
> +
> +static void surf_copy(int xe,
> +		      intel_ctx_t *ctx,
> +		      uint64_t ahnd,
> +		      const struct blt_copy_object *src,
> +		      const struct blt_copy_object *mid,
> +		      const struct blt_copy_object *dst,
> +		      int run_id, bool suspend_resume)
> +{
> +	struct blt_copy_data blt = {};
> +	struct blt_block_copy_data_ext ext = {};
> +	struct blt_ctrl_surf_copy_data surf = {};
> +	uint32_t bb1, bb2, ccs, ccs2, *ccsmap, *ccsmap2;
> +	uint64_t bb_size, ccssize = mid->size / CCS_RATIO;
> +	uint32_t *ccscopy;
> +	uint8_t uc_mocs = intel_get_uc_mocs(xe);
> +	uint32_t sysmem = system_memory(xe);
> +	int result;
> +
> +	igt_assert(mid->compression);
> +	ccscopy = (uint32_t *) malloc(ccssize);
> +	ccs = xe_bo_create_flags(xe, 0, ccssize, sysmem);
> +	ccs2 = xe_bo_create_flags(xe, 0, ccssize, sysmem);
> +
> +	blt_ctrl_surf_copy_init(xe, &surf);
> +	surf.print_bb = param.print_bb;
> +	set_surf_object(&surf.src, mid->handle, mid->region, mid->size,
> +			uc_mocs, BLT_INDIRECT_ACCESS);
> +	set_surf_object(&surf.dst, ccs, sysmem, ccssize, uc_mocs, DIRECT_ACCESS);
> +	bb_size = xe_get_default_alignment(xe);
> +	bb1 = xe_bo_create_flags(xe, 0, bb_size, sysmem);
> +	blt_set_batch(&surf.bb, bb1, bb_size, sysmem);
> +	blt_ctrl_surf_copy(xe, ctx, NULL, ahnd, &surf);
> +	intel_ctx_xe_sync(ctx, true);
> +
> +	ccsmap = xe_bo_map(xe, ccs, surf.dst.size);
> +	memcpy(ccscopy, ccsmap, ccssize);
> +
> +	if (suspend_resume) {
> +		char *orig, *orig2, *newsum, *newsum2;
> +
> +		orig = g_compute_checksum_for_data(G_CHECKSUM_SHA1,
> +						   (void *)ccsmap, surf.dst.size);
> +		orig2 = g_compute_checksum_for_data(G_CHECKSUM_SHA1,
> +						    (void *)mid->ptr, mid->size);
> +
> +		igt_system_suspend_autoresume(SUSPEND_STATE_FREEZE, SUSPEND_TEST_NONE);
> +
> +		set_surf_object(&surf.dst, ccs2, system_memory(xe), ccssize,
> +				0, DIRECT_ACCESS);
> +		blt_ctrl_surf_copy(xe, ctx, NULL, ahnd, &surf);
> +		intel_ctx_xe_sync(ctx, true);
> +
> +		ccsmap2 = xe_bo_map(xe, ccs2, surf.dst.size);
> +		newsum = g_compute_checksum_for_data(G_CHECKSUM_SHA1,
> +						     (void *)ccsmap2, surf.dst.size);
> +		newsum2 = g_compute_checksum_for_data(G_CHECKSUM_SHA1,
> +						      (void *)mid->ptr, mid->size);
> +
> +		munmap(ccsmap2, ccssize);
> +		igt_assert(!strcmp(orig, newsum));
> +		igt_assert(!strcmp(orig2, newsum2));
> +		g_free(orig);
> +		g_free(orig2);
> +		g_free(newsum);
> +		g_free(newsum2);
> +	}
> +
> +	/* corrupt ccs */
> +	for (int i = 0; i < surf.dst.size / sizeof(uint32_t); i++)
> +		ccsmap[i] = i;
> +	set_surf_object(&surf.src, ccs, sysmem, ccssize,
> +			uc_mocs, DIRECT_ACCESS);
> +	set_surf_object(&surf.dst, mid->handle, mid->region, mid->size,
> +			uc_mocs, INDIRECT_ACCESS);
> +	blt_ctrl_surf_copy(xe, ctx, NULL, ahnd, &surf);
> +	intel_ctx_xe_sync(ctx, true);
> +
> +	blt_copy_init(xe, &blt);
> +	blt.color_depth = CD_32bit;
> +	blt.print_bb = param.print_bb;
> +	blt_set_copy_object(&blt.src, mid);
> +	blt_set_copy_object(&blt.dst, dst);
> +	blt_set_object_ext(&ext.src, mid->compression_type, mid->x2, mid->y2, SURFACE_TYPE_2D);
> +	blt_set_object_ext(&ext.dst, 0, dst->x2, dst->y2, SURFACE_TYPE_2D);
> +	bb2 = xe_bo_create_flags(xe, 0, bb_size, sysmem);
> +	blt_set_batch(&blt.bb, bb2, bb_size, sysmem);
> +	blt_block_copy(xe, ctx, NULL, ahnd, &blt, &ext);
> +	intel_ctx_xe_sync(ctx, true);
> +	WRITE_PNG(xe, run_id, "corrupted", &blt.dst, dst->x2, dst->y2);
> +	result = memcmp(src->ptr, dst->ptr, src->size);
> +	igt_assert(result != 0);
> +
> +	/* retrieve back ccs */
> +	memcpy(ccsmap, ccscopy, ccssize);
> +	blt_ctrl_surf_copy(xe, ctx, NULL, ahnd, &surf);
> +
> +	blt_block_copy(xe, ctx, NULL, ahnd, &blt, &ext);
> +	intel_ctx_xe_sync(ctx, true);
> +	WRITE_PNG(xe, run_id, "corrected", &blt.dst, dst->x2, dst->y2);
> +	result = memcmp(src->ptr, dst->ptr, src->size);
> +	if (result)
> +		dump_corruption_info(src, dst);
> +
> +	munmap(ccsmap, ccssize);
> +	gem_close(xe, ccs);
> +	gem_close(xe, ccs2);
> +	gem_close(xe, bb1);
> +	gem_close(xe, bb2);
> +
> +	igt_assert_f(result == 0,
> +		     "Source and destination surfaces are different after "
> +		     "restoring source ccs data\n");
> +}
> +
> +struct blt_copy3_data {
> +	int xe;
> +	struct blt_copy_object src;
> +	struct blt_copy_object mid;
> +	struct blt_copy_object dst;
> +	struct blt_copy_object final;
> +	struct blt_copy_batch bb;
> +	enum blt_color_depth color_depth;
> +
> +	/* debug stuff */
> +	bool print_bb;
> +};
> +
> +struct blt_block_copy3_data_ext {
> +	struct blt_block_copy_object_ext src;
> +	struct blt_block_copy_object_ext mid;
> +	struct blt_block_copy_object_ext dst;
> +	struct blt_block_copy_object_ext final;
> +};
> +
> +static int blt_block_copy3(int xe,
> +			   const intel_ctx_t *ctx,
> +			   uint64_t ahnd,
> +			   const struct blt_copy3_data *blt3,
> +			   const struct blt_block_copy3_data_ext *ext3)
> +{
> +	struct blt_copy_data blt0;
> +	struct blt_block_copy_data_ext ext0;
> +	uint64_t bb_offset, alignment;
> +	uint64_t bb_pos = 0;
> +	int ret;
> +
> +	igt_assert_f(ahnd, "block-copy3 supports softpin only\n");
> +	igt_assert_f(blt3, "block-copy3 requires data to do blit\n");
> +
> +	alignment = xe_get_default_alignment(xe);
> +	get_offset(ahnd, blt3->src.handle, blt3->src.size, alignment);
> +	get_offset(ahnd, blt3->mid.handle, blt3->mid.size, alignment);
> +	get_offset(ahnd, blt3->dst.handle, blt3->dst.size, alignment);
> +	get_offset(ahnd, blt3->final.handle, blt3->final.size, alignment);
> +	bb_offset = get_offset(ahnd, blt3->bb.handle, blt3->bb.size, alignment);
> +
> +	/* First blit src -> mid */
> +	blt_copy_init(xe, &blt0);
> +	blt0.src = blt3->src;
> +	blt0.dst = blt3->mid;
> +	blt0.bb = blt3->bb;
> +	blt0.color_depth = blt3->color_depth;
> +	blt0.print_bb = blt3->print_bb;
> +	ext0.src = ext3->src;
> +	ext0.dst = ext3->mid;
> +	bb_pos = emit_blt_block_copy(xe, ahnd, &blt0, &ext0, bb_pos, false);
> +
> +	/* Second blit mid -> dst */
> +	blt_copy_init(xe, &blt0);
> +	blt0.src = blt3->mid;
> +	blt0.dst = blt3->dst;
> +	blt0.bb = blt3->bb;
> +	blt0.color_depth = blt3->color_depth;
> +	blt0.print_bb = blt3->print_bb;
> +	ext0.src = ext3->mid;
> +	ext0.dst = ext3->dst;
> +	bb_pos = emit_blt_block_copy(xe, ahnd, &blt0, &ext0, bb_pos, false);
> +
> +	/* Third blit dst -> final */
> +	blt_copy_init(xe, &blt0);
> +	blt0.src = blt3->dst;
> +	blt0.dst = blt3->final;
> +	blt0.bb = blt3->bb;
> +	blt0.color_depth = blt3->color_depth;
> +	blt0.print_bb = blt3->print_bb;
> +	ext0.src = ext3->dst;
> +	ext0.dst = ext3->final;
> +	bb_pos = emit_blt_block_copy(xe, ahnd, &blt0, &ext0, bb_pos, true);
> +
> +	intel_ctx_xe_exec(ctx, ahnd, bb_offset);
> +
> +	return ret;
> +}
> +
> +static void block_copy(int xe,
> +		       intel_ctx_t *ctx,
> +		       uint32_t region1, uint32_t region2,
> +		       enum blt_tiling_type mid_tiling,
> +		       const struct test_config *config)
> +{
> +	struct blt_copy_data blt = {};
> +	struct blt_block_copy_data_ext ext = {}, *pext = &ext;
> +	struct blt_copy_object *src, *mid, *dst;
> +	const uint32_t bpp = 32;
> +	uint64_t bb_size = xe_get_default_alignment(xe);
> +	uint64_t ahnd = intel_allocator_open(xe, ctx->vm, INTEL_ALLOCATOR_RELOC);
> +	uint32_t run_id = mid_tiling;
> +	uint32_t mid_region = region2, bb;
> +	uint32_t width = param.width, height = param.height;
> +	enum blt_compression mid_compression = config->compression;
> +	int mid_compression_format = param.compression_format;
> +	enum blt_compression_type comp_type = COMPRESSION_TYPE_3D;
> +	uint8_t uc_mocs = intel_get_uc_mocs(xe);
> +	int result;
> +
> +	bb = xe_bo_create_flags(xe, 0, bb_size, region1);
> +
> +	if (!blt_uses_extended_block_copy(xe))
> +		pext = NULL;
> +
> +	blt_copy_init(xe, &blt);
> +
> +	src = blt_create_object(&blt, region1, width, height, bpp, uc_mocs,
> +				T_LINEAR, COMPRESSION_DISABLED, comp_type, true);
> +	mid = blt_create_object(&blt, mid_region, width, height, bpp, uc_mocs,
> +				mid_tiling, mid_compression, comp_type, true);
> +	dst = blt_create_object(&blt, region1, width, height, bpp, uc_mocs,
> +				T_LINEAR, COMPRESSION_DISABLED, comp_type, true);
> +	igt_assert(src->size == dst->size);
> +	PRINT_SURFACE_INFO("src", src);
> +	PRINT_SURFACE_INFO("mid", mid);
> +	PRINT_SURFACE_INFO("dst", dst);
> +
> +	blt_surface_fill_rect(xe, src, width, height);
> +	WRITE_PNG(xe, run_id, "src", src, width, height);
> +
> +	blt.color_depth = CD_32bit;
> +	blt.print_bb = param.print_bb;
> +	blt_set_copy_object(&blt.src, src);
> +	blt_set_copy_object(&blt.dst, mid);
> +	blt_set_object_ext(&ext.src, 0, width, height, SURFACE_TYPE_2D);
> +	blt_set_object_ext(&ext.dst, mid_compression_format, width, height, SURFACE_TYPE_2D);
> +	blt_set_batch(&blt.bb, bb, bb_size, region1);
> +	blt_block_copy(xe, ctx, NULL, ahnd, &blt, pext);
> +	intel_ctx_xe_sync(ctx, true);
> +
> +	/* We expect mid != src if there's compression */
> +	if (mid->compression)
> +		igt_assert(memcmp(src->ptr, mid->ptr, src->size) != 0);
> +
> +	WRITE_PNG(xe, run_id, "mid", &blt.dst, width, height);
> +
> +	if (config->surfcopy && pext) {
> +		struct drm_xe_engine_class_instance inst = {
> +			.engine_class = DRM_XE_ENGINE_CLASS_COPY,
> +		};
> +		intel_ctx_t *surf_ctx = ctx;
> +		uint64_t surf_ahnd = ahnd;
> +		uint32_t vm, engine;
> +
> +		if (config->new_ctx) {
> +			vm = xe_vm_create(xe, DRM_XE_VM_CREATE_ASYNC_BIND_OPS, 0);
> +			engine = xe_engine_create(xe, vm, &inst, 0);
> +			surf_ctx = intel_ctx_xe(xe, vm, engine, 0, 0, 0);
> +			surf_ahnd = intel_allocator_open(xe, surf_ctx->vm,
> +							 INTEL_ALLOCATOR_RELOC);
> +		}
> +		surf_copy(xe, surf_ctx, surf_ahnd, src, mid, dst, run_id,
> +			  config->suspend_resume);
> +
> +		if (surf_ctx != ctx) {
> +			xe_engine_destroy(xe, engine);
> +			xe_vm_destroy(xe, vm);
> +			free(surf_ctx);
> +			put_ahnd(surf_ahnd);
> +		}
> +	}
> +
> +	blt_copy_init(xe, &blt);
> +	blt.color_depth = CD_32bit;
> +	blt.print_bb = param.print_bb;
> +	blt_set_copy_object(&blt.src, mid);
> +	blt_set_copy_object(&blt.dst, dst);
> +	blt_set_object_ext(&ext.src, mid_compression_format, width, height, SURFACE_TYPE_2D);
> +	blt_set_object_ext(&ext.dst, 0, width, height, SURFACE_TYPE_2D);
> +	if (config->inplace) {
> +		blt_set_object(&blt.dst, mid->handle, dst->size, mid->region, 0,
> +			       T_LINEAR, COMPRESSION_DISABLED, comp_type);
> +		blt.dst.ptr = mid->ptr;
> +	}
> +
> +	blt_set_batch(&blt.bb, bb, bb_size, region1);
> +	blt_block_copy(xe, ctx, NULL, ahnd, &blt, pext);
> +	intel_ctx_xe_sync(ctx, true);
> +
> +	WRITE_PNG(xe, run_id, "dst", &blt.dst, width, height);
> +
> +	result = memcmp(src->ptr, blt.dst.ptr, src->size);
> +
> +	/* Politely clean vm */
> +	put_offset(ahnd, src->handle);
> +	put_offset(ahnd, mid->handle);
> +	put_offset(ahnd, dst->handle);
> +	put_offset(ahnd, bb);
> +	intel_allocator_bind(ahnd, 0, 0);
> +	blt_destroy_object(xe, src);
> +	blt_destroy_object(xe, mid);
> +	blt_destroy_object(xe, dst);
> +	gem_close(xe, bb);
> +	put_ahnd(ahnd);
> +
> +	igt_assert_f(!result, "source and destination surfaces differs!\n");
> +}
> +
> +static void block_multicopy(int xe,
> +			    intel_ctx_t *ctx,
> +			    uint32_t region1, uint32_t region2,
> +			    enum blt_tiling_type mid_tiling,
> +			    const struct test_config *config)
> +{
> +	struct blt_copy3_data blt3 = {};
> +	struct blt_copy_data blt = {};
> +	struct blt_block_copy3_data_ext ext3 = {}, *pext3 = &ext3;
> +	struct blt_copy_object *src, *mid, *dst, *final;
> +	const uint32_t bpp = 32;
> +	uint64_t bb_size = xe_get_default_alignment(xe);
> +	uint64_t ahnd = intel_allocator_open(xe, ctx->vm, INTEL_ALLOCATOR_RELOC);
> +	uint32_t run_id = mid_tiling;
> +	uint32_t mid_region = region2, bb;
> +	uint32_t width = param.width, height = param.height;
> +	enum blt_compression mid_compression = config->compression;
> +	int mid_compression_format = param.compression_format;
> +	enum blt_compression_type comp_type = COMPRESSION_TYPE_3D;
> +	uint8_t uc_mocs = intel_get_uc_mocs(xe);
> +	int result;
> +
> +	bb = xe_bo_create_flags(xe, 0, bb_size, region1);
> +
> +	if (!blt_uses_extended_block_copy(xe))
> +		pext3 = NULL;
> +
> +	blt_copy_init(xe, &blt);
> +
> +	src = blt_create_object(&blt, region1, width, height, bpp, uc_mocs,
> +				T_LINEAR, COMPRESSION_DISABLED, comp_type, true);
> +	mid = blt_create_object(&blt, mid_region, width, height, bpp, uc_mocs,
> +				mid_tiling, mid_compression, comp_type, true);
> +	dst = blt_create_object(&blt, region1, width, height, bpp, uc_mocs,
> +				mid_tiling, COMPRESSION_DISABLED, comp_type, true);
> +	final = blt_create_object(&blt, region1, width, height, bpp, uc_mocs,
> +				  T_LINEAR, COMPRESSION_DISABLED, comp_type, true);
> +	igt_assert(src->size == dst->size);
> +	PRINT_SURFACE_INFO("src", src);
> +	PRINT_SURFACE_INFO("mid", mid);
> +	PRINT_SURFACE_INFO("dst", dst);
> +	PRINT_SURFACE_INFO("final", final);
> +
> +	blt_surface_fill_rect(xe, src, width, height);
> +
> +	blt3.color_depth = CD_32bit;
> +	blt3.print_bb = param.print_bb;
> +	blt_set_copy_object(&blt3.src, src);
> +	blt_set_copy_object(&blt3.mid, mid);
> +	blt_set_copy_object(&blt3.dst, dst);
> +	blt_set_copy_object(&blt3.final, final);
> +
> +	if (config->inplace) {
> +		blt_set_object(&blt3.dst, mid->handle, dst->size, mid->region,
> +			       mid->mocs, mid_tiling, COMPRESSION_DISABLED,
> +			       comp_type);
> +		blt3.dst.ptr = mid->ptr;
> +	}
> +
> +	blt_set_object_ext(&ext3.src, 0, width, height, SURFACE_TYPE_2D);
> +	blt_set_object_ext(&ext3.mid, mid_compression_format, width, height, SURFACE_TYPE_2D);
> +	blt_set_object_ext(&ext3.dst, 0, width, height, SURFACE_TYPE_2D);
> +	blt_set_object_ext(&ext3.final, 0, width, height, SURFACE_TYPE_2D);
> +	blt_set_batch(&blt3.bb, bb, bb_size, region1);
> +
> +	blt_block_copy3(xe, ctx, ahnd, &blt3, pext3);
> +	intel_ctx_xe_sync(ctx, true);
> +
> +	WRITE_PNG(xe, run_id, "src", &blt3.src, width, height);
> +	if (!config->inplace)
> +		WRITE_PNG(xe, run_id, "mid", &blt3.mid, width, height);
> +	WRITE_PNG(xe, run_id, "dst", &blt3.dst, width, height);
> +	WRITE_PNG(xe, run_id, "final", &blt3.final, width, height);
> +
> +	result = memcmp(src->ptr, blt3.final.ptr, src->size);
> +
> +	put_offset(ahnd, src->handle);
> +	put_offset(ahnd, mid->handle);
> +	put_offset(ahnd, dst->handle);
> +	put_offset(ahnd, final->handle);
> +	put_offset(ahnd, bb);
> +	intel_allocator_bind(ahnd, 0, 0);
> +	blt_destroy_object(xe, src);
> +	blt_destroy_object(xe, mid);
> +	blt_destroy_object(xe, dst);
> +	blt_destroy_object(xe, final);
> +	gem_close(xe, bb);
> +	put_ahnd(ahnd);
> +
> +	igt_assert_f(!result, "source and destination surfaces differs!\n");
> +}
> +
> +enum copy_func {
> +	BLOCK_COPY,
> +	BLOCK_MULTICOPY,
> +};
> +
> +static const struct {
> +	const char *suffix;
> +	void (*copyfn)(int fd,
> +		       intel_ctx_t *ctx,
> +		       uint32_t region1, uint32_t region2,
> +		       enum blt_tiling_type btype,
> +		       const struct test_config *config);
> +} copyfns[] = {
> +	[BLOCK_COPY] = { "", block_copy },
> +	[BLOCK_MULTICOPY] = { "-multicopy", block_multicopy },
> +};
> +
> +static void block_copy_test(int xe,
> +			    const struct test_config *config,
> +			    struct igt_collection *set,
> +			    enum copy_func copy_function)
> +{
> +	struct drm_xe_engine_class_instance inst = {
> +		.engine_class = DRM_XE_ENGINE_CLASS_COPY,
> +	};
> +	intel_ctx_t *ctx;
> +	struct igt_collection *regions;
> +	uint32_t vm, engine;
> +	int tiling;
> +
> +	if (config->compression && !blt_block_copy_supports_compression(xe))
> +		return;
> +
> +	if (config->inplace && !config->compression)
> +		return;
> +
> +	for_each_tiling(tiling) {
> +		if (!blt_block_copy_supports_tiling(xe, tiling) ||
> +		    (param.tiling >= 0 && param.tiling != tiling))
> +			continue;
> +
> +		for_each_variation_r(regions, 2, set) {
> +			uint32_t region1, region2;
> +			char *regtxt;
> +
> +			region1 = igt_collection_get_value(regions, 0);
> +			region2 = igt_collection_get_value(regions, 1);
> +
> +			/* Compressed surface must be in device memory */
> +			if (config->compression && !XE_IS_VRAM_MEMORY_REGION(xe, region2))
> +				continue;
> +
> +			regtxt = xe_memregion_dynamic_subtest_name(xe, regions);
> +
> +			igt_dynamic_f("%s-%s-compfmt%d-%s%s",
> +				      blt_tiling_name(tiling),
> +				      config->compression ?
> +					      "compressed" : "uncompressed",
> +				      param.compression_format, regtxt,
> +				      copyfns[copy_function].suffix) {
> +				uint32_t sync_bind, sync_out;
> +
> +				vm = xe_vm_create(xe, DRM_XE_VM_CREATE_ASYNC_BIND_OPS, 0);
> +				engine = xe_engine_create(xe, vm, &inst, 0);
> +				sync_bind = syncobj_create(xe, 0);
> +				sync_out = syncobj_create(xe, 0);
> +				ctx = intel_ctx_xe(xe, vm, engine,
> +						   0, sync_bind, sync_out);
> +
> +				copyfns[copy_function].copyfn(xe, ctx,
> +							      region1, region2,
> +							      tiling, config);
> +
> +				xe_engine_destroy(xe, engine);
> +				xe_vm_destroy(xe, vm);
> +				syncobj_destroy(xe, sync_bind);
> +				syncobj_destroy(xe, sync_out);
> +				free(ctx);
> +			}
> +
> +			free(regtxt);
> +		}
> +	}
> +}
> +
> +static int opt_handler(int opt, int opt_index, void *data)
> +{
> +	switch (opt) {
> +	case 'b':
> +		param.print_bb = true;
> +		igt_debug("Print bb: %d\n", param.print_bb);
> +		break;
> +	case 'f':
> +		param.compression_format = atoi(optarg);
> +		igt_debug("Compression format: %d\n", param.compression_format);
> +		igt_assert((param.compression_format & ~0x1f) == 0);
> +		break;
> +	case 'p':
> +		param.write_png = true;
> +		igt_debug("Write png: %d\n", param.write_png);
> +		break;
> +	case 's':
> +		param.print_surface_info = true;
> +		igt_debug("Print surface info: %d\n", param.print_surface_info);
> +		break;
> +	case 't':
> +		param.tiling = atoi(optarg);
> +		igt_debug("Tiling: %d\n", param.tiling);
> +		break;
> +	case 'W':
> +		param.width = atoi(optarg);
> +		igt_debug("Width: %d\n", param.width);
> +		break;
> +	case 'H':
> +		param.height = atoi(optarg);
> +		igt_debug("Height: %d\n", param.height);
> +		break;
> +	default:
> +		return IGT_OPT_HANDLER_ERROR;
> +	}
> +
> +	return IGT_OPT_HANDLER_SUCCESS;
> +}
> +
> +const char *help_str =
> +	"  -b\tPrint bb\n"
> +	"  -f\tCompression format (0-31)\n"
> +	"  -p\tWrite PNG\n"
> +	"  -s\tPrint surface info\n"
> +	"  -t\tTiling format (0 - linear, 1 - XMAJOR, 2 - YMAJOR, 3 - TILE4, 4 - TILE64)\n"
> +	"  -W\tWidth (default 512)\n"
> +	"  -H\tHeight (default 512)"
> +	;
> +
> +igt_main_args("bf:pst:W:H:", NULL, help_str, opt_handler, NULL)
> +{
> +	struct igt_collection *set;
> +	int xe;
> +
> +	igt_fixture {
> +		xe = drm_open_driver(DRIVER_XE);
> +		igt_require(blt_has_block_copy(xe));
> +
> +		xe_device_get(xe);
> +
> +		set = xe_get_memory_region_set(xe,
> +					       XE_MEM_REGION_CLASS_SYSMEM,
> +					       XE_MEM_REGION_CLASS_VRAM);
> +	}
> +
> +	igt_describe("Check block-copy uncompressed blit");
> +	igt_subtest_with_dynamic("block-copy-uncompressed") {
> +		struct test_config config = {};
> +
> +		block_copy_test(xe, &config, set, BLOCK_COPY);
> +	}
> +
> +	igt_describe("Check block-copy flatccs compressed blit");
> +	igt_subtest_with_dynamic("block-copy-compressed") {
> +		struct test_config config = { .compression = true };
> +
> +		block_copy_test(xe, &config, set, BLOCK_COPY);
> +	}
> +
> +	igt_describe("Check block-multicopy flatccs compressed blit");
> +	igt_subtest_with_dynamic("block-multicopy-compressed") {
> +		struct test_config config = { .compression = true };
> +
> +		block_copy_test(xe, &config, set, BLOCK_MULTICOPY);
> +	}
> +
> +	igt_describe("Check block-multicopy flatccs inplace decompression blit");
> +	igt_subtest_with_dynamic("block-multicopy-inplace") {
> +		struct test_config config = { .compression = true,
> +					      .inplace = true };
> +
> +		block_copy_test(xe, &config, set, BLOCK_MULTICOPY);
> +	}
> +
> +	igt_describe("Check flatccs data can be copied from/to surface");
> +	igt_subtest_with_dynamic("ctrl-surf-copy") {
> +		struct test_config config = { .compression = true,
> +					      .surfcopy = true };
> +
> +		block_copy_test(xe, &config, set, BLOCK_COPY);
> +	}
> +
> +	igt_describe("Check flatccs data are physically tagged and visible"
> +		     " in different contexts");
> +	igt_subtest_with_dynamic("ctrl-surf-copy-new-ctx") {
> +		struct test_config config = { .compression = true,
> +					      .surfcopy = true,
> +					      .new_ctx = true };
> +
> +		block_copy_test(xe, &config, set, BLOCK_COPY);
> +	}
> +
> +	igt_describe("Check flatccs data persists after suspend / resume (S0)");
> +	igt_subtest_with_dynamic("suspend-resume") {
> +		struct test_config config = { .compression = true,
> +					      .surfcopy = true,
> +					      .suspend_resume = true };
> +
> +		block_copy_test(xe, &config, set, BLOCK_COPY);
> +	}
> +
> +	igt_fixture {
> +		xe_device_put(xe);
> +		close(xe);
> +	}
> +}

Re: [igt-dev] [PATCH i-g-t v3 15/17] tests/xe_exercise_blt: Check blitter library fast-copy for Xe

[Karolina Stolarek]

On 11.07.2023 13:20, Zbigniew Kempczyński wrote:
> Port this test to work on xe. Instead of adding conditional code for
> xe code which would decrease readability this is new test for xe.
> 
> Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
> ---
> v3: - Explicitely release offsets and unbind (Karolina)
>      - Use has-fast-copy instead of copied has-block-copy check (Karolina)
> ---
>   tests/meson.build          |   1 +
>   tests/xe/xe_exercise_blt.c | 377 +++++++++++++++++++++++++++++++++++++
>   2 files changed, 378 insertions(+)
>   create mode 100644 tests/xe/xe_exercise_blt.c
> 
> diff --git a/tests/meson.build b/tests/meson.build
> index da8653fabc..3eddb2fb4e 100644
> --- a/tests/meson.build
> +++ b/tests/meson.build
> @@ -276,6 +276,7 @@ xe_progs = [
>   	'xe_exec_reset',
>   	'xe_exec_store',
>   	'xe_exec_threads',
> +	'xe_exercise_blt',
>   	'xe_gpgpu_fill',
>   	'xe_guc_pc',
>   	'xe_huc_copy',
> diff --git a/tests/xe/xe_exercise_blt.c b/tests/xe/xe_exercise_blt.c
> new file mode 100644
> index 0000000000..693e80d0b7
> --- /dev/null
> +++ b/tests/xe/xe_exercise_blt.c
> @@ -0,0 +1,377 @@
> +// SPDX-License-Identifier: MIT
> +/*
> + * Copyright © 2023 Intel Corporation
> + */
> +
> +#include "igt.h"
> +#include "drm.h"
> +#include "lib/intel_chipset.h"
> +#include "intel_blt.h"
> +#include "intel_mocs.h"
> +#include "xe/xe_ioctl.h"
> +#include "xe/xe_query.h"
> +#include "xe/xe_util.h"
> +
> +/**
> + * TEST: xe exercise blt
> + * Description: Exercise blitter commands on Xe
> + * Feature: blitter
> + * Run type: FULL
> + * Test category: GEM_Legacy
> + *
> + * SUBTEST: fast-copy
> + * Description:
> + *   Check fast-copy blit
> + *   blitter
> + *
> + * SUBTEST: fast-copy-emit
> + * Description:
> + *   Check multiple fast-copy in one batch
> + *   blitter
> + */
> +
> +IGT_TEST_DESCRIPTION("Exercise blitter commands on Xe");
> +
> +static struct param {
> +	int tiling;
> +	bool write_png;
> +	bool print_bb;
> +	bool print_surface_info;
> +	int width;
> +	int height;
> +} param = {
> +	.tiling = -1,
> +	.write_png = false,
> +	.print_bb = false,
> +	.print_surface_info = false,
> +	.width = 512,
> +	.height = 512,
> +};
> +
> +#define PRINT_SURFACE_INFO(name, obj) do { \
> +	if (param.print_surface_info) \
> +		blt_surface_info((name), (obj)); } while (0)
> +
> +#define WRITE_PNG(fd, id, name, obj, w, h) do { \
> +	if (param.write_png) \
> +		blt_surface_to_png((fd), (id), (name), (obj), (w), (h)); } while (0)
> +
> +struct blt_fast_copy_data {
> +	int xe;
> +	struct blt_copy_object src;
> +	struct blt_copy_object mid;
> +	struct blt_copy_object dst;
> +
> +	struct blt_copy_batch bb;
> +	enum blt_color_depth color_depth;
> +
> +	/* debug stuff */
> +	bool print_bb;
> +};
> +
> +static int fast_copy_one_bb(int xe,
> +			    const intel_ctx_t *ctx,
> +			    uint64_t ahnd,
> +			    const struct blt_fast_copy_data *blt)
> +{
> +	struct blt_copy_data blt_tmp;
> +	uint64_t bb_offset, alignment;
> +	uint64_t bb_pos = 0;
> +	int ret = 0;
> +
> +	alignment = xe_get_default_alignment(xe);
> +
> +	get_offset(ahnd, blt->src.handle, blt->src.size, alignment);
> +	get_offset(ahnd, blt->mid.handle, blt->mid.size, alignment);
> +	get_offset(ahnd, blt->dst.handle, blt->dst.size, alignment);
> +	bb_offset = get_offset(ahnd, blt->bb.handle, blt->bb.size, alignment);
> +
> +	/* First blit */
> +	blt_copy_init(xe, &blt_tmp);
> +	blt_tmp.src = blt->src;
> +	blt_tmp.dst = blt->mid;
> +	blt_tmp.bb = blt->bb;
> +	blt_tmp.color_depth = blt->color_depth;
> +	blt_tmp.print_bb = blt->print_bb;
> +	bb_pos = emit_blt_fast_copy(xe, ahnd, &blt_tmp, bb_pos, false);
> +
> +	/* Second blit */
> +	blt_copy_init(xe, &blt_tmp);
> +	blt_tmp.src = blt->mid;
> +	blt_tmp.dst = blt->dst;
> +	blt_tmp.bb = blt->bb;
> +	blt_tmp.color_depth = blt->color_depth;
> +	blt_tmp.print_bb = blt->print_bb;
> +	bb_pos = emit_blt_fast_copy(xe, ahnd, &blt_tmp, bb_pos, true);
> +
> +	intel_ctx_xe_exec(ctx, ahnd, bb_offset);
> +
> +	return ret;
> +}
> +
> +static void fast_copy_emit(int xe, const intel_ctx_t *ctx,
> +			   uint32_t region1, uint32_t region2,
> +			   enum blt_tiling_type mid_tiling)
> +{
> +	struct blt_copy_data bltinit = {};
> +	struct blt_fast_copy_data blt = {};
> +	struct blt_copy_object *src, *mid, *dst;
> +	const uint32_t bpp = 32;
> +	uint64_t bb_size = xe_get_default_alignment(xe);
> +	uint64_t ahnd = intel_allocator_open_full(xe, ctx->vm, 0, 0,
> +						  INTEL_ALLOCATOR_SIMPLE,
> +						  ALLOC_STRATEGY_LOW_TO_HIGH, 0);
> +	uint32_t bb, width = param.width, height = param.height;
> +	int result;
> +
> +	bb = xe_bo_create_flags(xe, 0, bb_size, region1);
> +
> +	blt_copy_init(xe, &bltinit);
> +	src = blt_create_object(&bltinit, region1, width, height, bpp, 0,
> +				T_LINEAR, COMPRESSION_DISABLED, 0, true);
> +	mid = blt_create_object(&bltinit, region2, width, height, bpp, 0,
> +				mid_tiling, COMPRESSION_DISABLED, 0, true);
> +	dst = blt_create_object(&bltinit, region1, width, height, bpp, 0,
> +				T_LINEAR, COMPRESSION_DISABLED, 0, true);
> +	igt_assert(src->size == dst->size);
> +
> +	PRINT_SURFACE_INFO("src", src);
> +	PRINT_SURFACE_INFO("mid", mid);
> +	PRINT_SURFACE_INFO("dst", dst);
> +
> +	blt_surface_fill_rect(xe, src, width, height);
> +	WRITE_PNG(xe, mid_tiling, "src", src, width, height);
> +
> +	memset(&blt, 0, sizeof(blt));

Could we use blt_copy_init() here?

Thanks,
Karolina

> +	blt.color_depth = CD_32bit;
> +	blt.print_bb = param.print_bb;
> +	blt_set_copy_object(&blt.src, src);
> +	blt_set_copy_object(&blt.mid, mid);
> +	blt_set_copy_object(&blt.dst, dst);
> +	blt_set_batch(&blt.bb, bb, bb_size, region1);
> +
> +	fast_copy_one_bb(xe, ctx, ahnd, &blt);
> +
> +	WRITE_PNG(xe, mid_tiling, "mid", &blt.mid, width, height);
> +	WRITE_PNG(xe, mid_tiling, "dst", &blt.dst, width, height);
> +
> +	result = memcmp(src->ptr, blt.dst.ptr, src->size);
> +
> +	blt_destroy_object(xe, src);
> +	blt_destroy_object(xe, mid);
> +	blt_destroy_object(xe, dst);
> +	gem_close(xe, bb);
> +	put_ahnd(ahnd);
> +
> +	munmap(&bb, bb_size);
> +
> +	igt_assert_f(!result, "source and destination surfaces differs!\n");
> +}
> +
> +static void fast_copy(int xe, const intel_ctx_t *ctx,
> +		      uint32_t region1, uint32_t region2,
> +		      enum blt_tiling_type mid_tiling)
> +{
> +	struct blt_copy_data blt = {};
> +	struct blt_copy_object *src, *mid, *dst;
> +	const uint32_t bpp = 32;
> +	uint64_t bb_size = xe_get_default_alignment(xe);
> +	uint64_t ahnd = intel_allocator_open_full(xe, ctx->vm, 0, 0,
> +						  INTEL_ALLOCATOR_SIMPLE,
> +						  ALLOC_STRATEGY_LOW_TO_HIGH, 0);
> +	uint32_t bb;
> +	uint32_t width = param.width, height = param.height;
> +	int result;
> +
> +	bb = xe_bo_create_flags(xe, 0, bb_size, region1);
> +
> +	blt_copy_init(xe, &blt);
> +	src = blt_create_object(&blt, region1, width, height, bpp, 0,
> +				T_LINEAR, COMPRESSION_DISABLED, 0, true);
> +	mid = blt_create_object(&blt, region2, width, height, bpp, 0,
> +				mid_tiling, COMPRESSION_DISABLED, 0, true);
> +	dst = blt_create_object(&blt, region1, width, height, bpp, 0,
> +				T_LINEAR, COMPRESSION_DISABLED, 0, true);
> +	igt_assert(src->size == dst->size);
> +
> +	blt_surface_fill_rect(xe, src, width, height);
> +
> +	blt.color_depth = CD_32bit;
> +	blt.print_bb = param.print_bb;
> +	blt_set_copy_object(&blt.src, src);
> +	blt_set_copy_object(&blt.dst, mid);
> +	blt_set_batch(&blt.bb, bb, bb_size, region1);
> +
> +	blt_fast_copy(xe, ctx, NULL, ahnd, &blt);
> +
> +	WRITE_PNG(xe, mid_tiling, "src", &blt.src, width, height);
> +	WRITE_PNG(xe, mid_tiling, "mid", &blt.dst, width, height);
> +
> +	blt_copy_init(xe, &blt);
> +	blt.color_depth = CD_32bit;
> +	blt.print_bb = param.print_bb;
> +	blt_set_copy_object(&blt.src, mid);
> +	blt_set_copy_object(&blt.dst, dst);
> +	blt_set_batch(&blt.bb, bb, bb_size, region1);
> +
> +	blt_fast_copy(xe, ctx, NULL, ahnd, &blt);
> +
> +	WRITE_PNG(xe, mid_tiling, "dst", &blt.dst, width, height);
> +
> +	result = memcmp(src->ptr, blt.dst.ptr, src->size);
> +
> +	put_offset(ahnd, src->handle);
> +	put_offset(ahnd, mid->handle);
> +	put_offset(ahnd, dst->handle);
> +	put_offset(ahnd, bb);
> +	intel_allocator_bind(ahnd, 0, 0);
> +	blt_destroy_object(xe, src);
> +	blt_destroy_object(xe, mid);
> +	blt_destroy_object(xe, dst);
> +	gem_close(xe, bb);
> +	put_ahnd(ahnd);
> +
> +	igt_assert_f(!result, "source and destination surfaces differs!\n");
> +}
> +
> +enum fast_copy_func {
> +	FAST_COPY,
> +	FAST_COPY_EMIT
> +};
> +
> +static char
> +	*full_subtest_str(char *regtxt, enum blt_tiling_type tiling,
> +			  enum fast_copy_func func)
> +{
> +	char *name;
> +	uint32_t len;
> +
> +	len = asprintf(&name, "%s-%s%s", blt_tiling_name(tiling), regtxt,
> +		       func == FAST_COPY_EMIT ? "-emit" : "");
> +
> +	igt_assert_f(len >= 0, "asprintf failed!\n");
> +
> +	return name;
> +}
> +
> +static void fast_copy_test(int xe,
> +			   struct igt_collection *set,
> +			   enum fast_copy_func func)
> +{
> +	struct drm_xe_engine_class_instance inst = {
> +		.engine_class = DRM_XE_ENGINE_CLASS_COPY,
> +	};
> +	struct igt_collection *regions;
> +	void (*copy_func)(int xe, const intel_ctx_t *ctx,
> +			  uint32_t r1, uint32_t r2, enum blt_tiling_type tiling);
> +	intel_ctx_t *ctx;
> +	int tiling;
> +
> +	for_each_tiling(tiling) {
> +		if (!blt_fast_copy_supports_tiling(xe, tiling))
> +			continue;
> +
> +		for_each_variation_r(regions, 2, set) {
> +			uint32_t region1, region2;
> +			uint32_t vm, engine;
> +			char *regtxt, *test_name;
> +
> +			region1 = igt_collection_get_value(regions, 0);
> +			region2 = igt_collection_get_value(regions, 1);
> +
> +			vm = xe_vm_create(xe, DRM_XE_VM_CREATE_ASYNC_BIND_OPS, 0);
> +			engine = xe_engine_create(xe, vm, &inst, 0);
> +			ctx = intel_ctx_xe(xe, vm, engine, 0, 0, 0);
> +
> +			copy_func = (func == FAST_COPY) ? fast_copy : fast_copy_emit;
> +			regtxt = xe_memregion_dynamic_subtest_name(xe, regions);
> +			test_name = full_subtest_str(regtxt, tiling, func);
> +
> +			igt_dynamic_f("%s", test_name) {
> +				copy_func(xe, ctx,
> +					  region1, region2,
> +					  tiling);
> +			}
> +
> +			free(regtxt);
> +			free(test_name);
> +			xe_engine_destroy(xe, engine);
> +			xe_vm_destroy(xe, vm);
> +			free(ctx);
> +		}
> +	}
> +}
> +
> +static int opt_handler(int opt, int opt_index, void *data)
> +{
> +	switch (opt) {
> +	case 'b':
> +		param.print_bb = true;
> +		igt_debug("Print bb: %d\n", param.print_bb);
> +		break;
> +	case 'p':
> +		param.write_png = true;
> +		igt_debug("Write png: %d\n", param.write_png);
> +		break;
> +	case 's':
> +		param.print_surface_info = true;
> +		igt_debug("Print surface info: %d\n", param.print_surface_info);
> +		break;
> +	case 't':
> +		param.tiling = atoi(optarg);
> +		igt_debug("Tiling: %d\n", param.tiling);
> +		break;
> +	case 'W':
> +		param.width = atoi(optarg);
> +		igt_debug("Width: %d\n", param.width);
> +		break;
> +	case 'H':
> +		param.height = atoi(optarg);
> +		igt_debug("Height: %d\n", param.height);
> +		break;
> +	default:
> +		return IGT_OPT_HANDLER_ERROR;
> +	}
> +
> +	return IGT_OPT_HANDLER_SUCCESS;
> +}
> +
> +const char *help_str =
> +	"  -b\tPrint bb\n"
> +	"  -p\tWrite PNG\n"
> +	"  -s\tPrint surface info\n"
> +	"  -t\tTiling format (0 - linear, 1 - XMAJOR, 2 - YMAJOR, 3 - TILE4, 4 - TILE64, 5 - YFMAJOR)\n"
> +	"  -W\tWidth (default 512)\n"
> +	"  -H\tHeight (default 512)"
> +	;
> +
> +igt_main_args("b:pst:W:H:", NULL, help_str, opt_handler, NULL)
> +{
> +	struct igt_collection *set;
> +	int xe;
> +
> +	igt_fixture {
> +		xe = drm_open_driver(DRIVER_XE);
> +		igt_require(blt_has_fast_copy(xe));
> +
> +		xe_device_get(xe);
> +
> +		set = xe_get_memory_region_set(xe,
> +					       XE_MEM_REGION_CLASS_SYSMEM,
> +					       XE_MEM_REGION_CLASS_VRAM);
> +	}
> +
> +	igt_describe("Check fast-copy blit");
> +	igt_subtest_with_dynamic("fast-copy") {
> +		fast_copy_test(xe, set, FAST_COPY);
> +	}
> +
> +	igt_describe("Check multiple fast-copy in one batch");
> +	igt_subtest_with_dynamic("fast-copy-emit") {
> +		fast_copy_test(xe, set, FAST_COPY_EMIT);
> +	}
> +
> +	igt_fixture {
> +		drm_close_driver(xe);
> +	}
> +}

Re: [igt-dev] [PATCH i-g-t v3 13/17] lib/intel_blt: Extend blitter library to support xe driver

[Karolina Stolarek]

On 12.07.2023 11:43, Zbigniew Kempczyński wrote:
> On Wed, Jul 12, 2023 at 11:22:00AM +0200, Karolina Stolarek wrote:
>> On 11.07.2023 13:20, Zbigniew Kempczyński wrote:
>>> Use already written for i915 blitter library in xe development.
>>> Add appropriate code paths which are unique for those drivers.
>>>
>>> Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
>>> ---
>>> v3: Extract common helpers for getting mapping and alignment (Karolina)
>>> ---
>>>    lib/intel_blt.c | 235 ++++++++++++++++++++++++++++++------------------
>>>    lib/intel_blt.h |   2 +-
>>>    2 files changed, 150 insertions(+), 87 deletions(-)
>>>
>>> diff --git a/lib/intel_blt.c b/lib/intel_blt.c
>>> index f2f86e4947..5387a60ee3 100644
>>> --- a/lib/intel_blt.c
>>> +++ b/lib/intel_blt.c
>>> @@ -9,9 +9,13 @@
>>>    #include <malloc.h>
>>>    #include <cairo.h>
>>>    #include "drm.h"
>>> -#include "igt.h"
>>>    #include "i915/gem_create.h"
>>> +#include "igt.h"
>>> +#include "igt_syncobj.h"
>>>    #include "intel_blt.h"
>>> +#include "xe/xe_ioctl.h"
>>> +#include "xe/xe_query.h"
>>> +#include "xe/xe_util.h"
>>>    #define BITRANGE(start, end) (end - start + 1)
>>>    #define GET_CMDS_INFO(__fd) intel_get_cmds_info(intel_get_drm_devid(__fd))
>>> @@ -468,24 +472,40 @@ static int __special_mode(const struct blt_copy_data *blt)
>>>    	return SM_NONE;
>>>    }
>>> -static int __memory_type(uint32_t region)
>>> +static int __memory_type(int fd, enum intel_driver driver, uint32_t region)
>>>    {
>>> -	igt_assert_f(IS_DEVICE_MEMORY_REGION(region) ||
>>> -		     IS_SYSTEM_MEMORY_REGION(region),
>>> -		     "Invalid region: %x\n", region);
>>> +	if (driver == INTEL_DRIVER_I915) {
>>> +		igt_assert_f(IS_DEVICE_MEMORY_REGION(region) ||
>>> +			     IS_SYSTEM_MEMORY_REGION(region),
>>> +			     "Invalid region: %x\n", region);
>>> +	} else {
>>> +		igt_assert_f(XE_IS_VRAM_MEMORY_REGION(fd, region) ||
>>> +			     XE_IS_SYSMEM_MEMORY_REGION(fd, region),
>>> +			     "Invalid region: %x\n", region);
>>> +	}
>>> -	if (IS_DEVICE_MEMORY_REGION(region))
>>> +	if (driver == INTEL_DRIVER_I915 && IS_DEVICE_MEMORY_REGION(region))
>>>    		return TM_LOCAL_MEM;
>>> +	else if (driver == INTEL_DRIVER_XE && XE_IS_VRAM_MEMORY_REGION(fd, region))
>>> +		return TM_LOCAL_MEM;
>>> +
>>>    	return TM_SYSTEM_MEM;
>>>    }
>>> -static enum blt_aux_mode __aux_mode(const struct blt_copy_object *obj)
>>> +static enum blt_aux_mode __aux_mode(int fd,
>>> +				    enum intel_driver driver,
>>> +				    const struct blt_copy_object *obj)
>>>    {
>>> -	if (obj->compression == COMPRESSION_ENABLED) {
>>> +	if (driver == INTEL_DRIVER_I915 && obj->compression == COMPRESSION_ENABLED) {
>>>    		igt_assert_f(IS_DEVICE_MEMORY_REGION(obj->region),
>>>    			     "XY_BLOCK_COPY_BLT supports compression "
>>>    			     "on device memory only\n");
>>>    		return AM_AUX_CCS_E;
>>> +	} else if (driver == INTEL_DRIVER_XE && obj->compression == COMPRESSION_ENABLED) {
>>> +		igt_assert_f(XE_IS_VRAM_MEMORY_REGION(fd, obj->region),
>>> +			     "XY_BLOCK_COPY_BLT supports compression "
>>> +			     "on device memory only\n");
>>> +		return AM_AUX_CCS_E;
>>>    	}
>>>    	return AM_AUX_NONE;
>>> @@ -508,9 +528,9 @@ static void fill_data(struct gen12_block_copy_data *data,
>>>    	data->dw00.length = extended_command ? 20 : 10;
>>>    	if (__special_mode(blt) == SM_FULL_RESOLVE)
>>> -		data->dw01.dst_aux_mode = __aux_mode(&blt->src);
>>> +		data->dw01.dst_aux_mode = __aux_mode(blt->fd, blt->driver, &blt->src);
>>>    	else
>>> -		data->dw01.dst_aux_mode = __aux_mode(&blt->dst);
>>> +		data->dw01.dst_aux_mode = __aux_mode(blt->fd, blt->driver, &blt->dst);
>>>    	data->dw01.dst_pitch = blt->dst.pitch - 1;
>>>    	data->dw01.dst_mocs = blt->dst.mocs;
>>> @@ -531,13 +551,13 @@ static void fill_data(struct gen12_block_copy_data *data,
>>>    	data->dw06.dst_x_offset = blt->dst.x_offset;
>>>    	data->dw06.dst_y_offset = blt->dst.y_offset;
>>> -	data->dw06.dst_target_memory = __memory_type(blt->dst.region);
>>> +	data->dw06.dst_target_memory = __memory_type(blt->fd, blt->driver, blt->dst.region);
>>>    	data->dw07.src_x1 = blt->src.x1;
>>>    	data->dw07.src_y1 = blt->src.y1;
>>>    	data->dw08.src_pitch = blt->src.pitch - 1;
>>> -	data->dw08.src_aux_mode = __aux_mode(&blt->src);
>>> +	data->dw08.src_aux_mode = __aux_mode(blt->fd, blt->driver, &blt->src);
>>>    	data->dw08.src_mocs = blt->src.mocs;
>>>    	data->dw08.src_compression = blt->src.compression;
>>>    	data->dw08.src_tiling = __block_tiling(blt->src.tiling);
>>> @@ -550,7 +570,7 @@ static void fill_data(struct gen12_block_copy_data *data,
>>>    	data->dw11.src_x_offset = blt->src.x_offset;
>>>    	data->dw11.src_y_offset = blt->src.y_offset;
>>> -	data->dw11.src_target_memory = __memory_type(blt->src.region);
>>> +	data->dw11.src_target_memory = __memory_type(blt->fd, blt->driver, blt->src.region);
>>>    }
>>>    static void fill_data_ext(struct gen12_block_copy_data_ext *dext,
>>> @@ -692,6 +712,24 @@ static void dump_bb_ext(struct gen12_block_copy_data_ext *data)
>>>    		 data->dw21.src_array_index);
>>>    }
>>> +static uint64_t get_default_alignment(int fd, enum intel_driver driver)
>>> +{
>>> +	if (driver == INTEL_DRIVER_XE)
>>> +		return xe_get_default_alignment(fd);
>>> +
>>> +	return gem_detect_safe_alignment(fd);
>>> +}
>>> +
>>> +static void *bo_map(int fd, uint32_t handle, uint64_t size,
>>> +		    enum intel_driver driver)
>>> +{
>>> +	if (driver == INTEL_DRIVER_XE)
>>> +		return xe_bo_map(fd, handle, size);
>>> +
>>> +	return gem_mmap__device_coherent(fd, handle, 0, size,
>>> +					 PROT_READ | PROT_WRITE);
>>> +}
>>> +
>>>    /**
>>>     * blt_copy_init:
>>>     * @fd: drm fd
>>> @@ -739,7 +777,7 @@ uint64_t emit_blt_block_copy(int fd,
>>>    	igt_assert_f(ahnd, "block-copy supports softpin only\n");
>>>    	igt_assert_f(blt, "block-copy requires data to do blit\n");
>>> -	alignment = gem_detect_safe_alignment(fd);
>>> +	alignment = get_default_alignment(fd, blt->driver);
>>>    	src_offset = get_offset(ahnd, blt->src.handle, blt->src.size, alignment)
>>>    		     + blt->src.plane_offset;
>>>    	dst_offset = get_offset(ahnd, blt->dst.handle, blt->dst.size, alignment)
>>> @@ -748,8 +786,7 @@ uint64_t emit_blt_block_copy(int fd,
>>>    	fill_data(&data, blt, src_offset, dst_offset, ext);
>>> -	bb = gem_mmap__device_coherent(fd, blt->bb.handle, 0, blt->bb.size,
>>> -				       PROT_READ | PROT_WRITE);
>>> +	bb = bo_map(fd, blt->bb.handle, blt->bb.size, blt->driver);
>>>    	igt_assert(bb_pos + sizeof(data) < blt->bb.size);
>>>    	memcpy(bb + bb_pos, &data, sizeof(data));
>>> @@ -812,29 +849,35 @@ int blt_block_copy(int fd,
>>>    	igt_assert_f(ahnd, "block-copy supports softpin only\n");
>>>    	igt_assert_f(blt, "block-copy requires data to do blit\n");
>>> +	igt_assert_neq(blt->driver, 0);
>>> -	alignment = gem_detect_safe_alignment(fd);
>>> +	alignment = get_default_alignment(fd, blt->driver);
>>>    	src_offset = get_offset(ahnd, blt->src.handle, blt->src.size, alignment);
>>>    	dst_offset = get_offset(ahnd, blt->dst.handle, blt->dst.size, alignment);
>>>    	bb_offset = get_offset(ahnd, blt->bb.handle, blt->bb.size, alignment);
>>>    	emit_blt_block_copy(fd, ahnd, blt, ext, 0, true);
>>> -	obj[0].offset = CANONICAL(dst_offset);
>>> -	obj[1].offset = CANONICAL(src_offset);
>>> -	obj[2].offset = CANONICAL(bb_offset);
>>> -	obj[0].handle = blt->dst.handle;
>>> -	obj[1].handle = blt->src.handle;
>>> -	obj[2].handle = blt->bb.handle;
>>> -	obj[0].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_WRITE |
>>> -		       EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
>>> -	obj[1].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
>>> -	obj[2].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
>>> -	execbuf.buffer_count = 3;
>>> -	execbuf.buffers_ptr = to_user_pointer(obj);
>>> -	execbuf.rsvd1 = ctx ? ctx->id : 0;
>>> -	execbuf.flags = e ? e->flags : I915_EXEC_BLT;
>>> -	ret = __gem_execbuf(fd, &execbuf);
>>> +	if (blt->driver == INTEL_DRIVER_XE) {
>>> +		intel_ctx_xe_exec(ctx, ahnd, CANONICAL(bb_offset));
>>> +	} else {
>>> +		obj[0].offset = CANONICAL(dst_offset);
>>> +		obj[1].offset = CANONICAL(src_offset);
>>> +		obj[2].offset = CANONICAL(bb_offset);
>>> +		obj[0].handle = blt->dst.handle;
>>> +		obj[1].handle = blt->src.handle;
>>> +		obj[2].handle = blt->bb.handle;
>>> +		obj[0].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_WRITE |
>>> +				EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
>>> +		obj[1].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
>>> +		obj[2].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
>>> +		execbuf.buffer_count = 3;
>>> +		execbuf.buffers_ptr = to_user_pointer(obj);
>>> +		execbuf.rsvd1 = ctx ? ctx->id : 0;
>>> +		execbuf.flags = e ? e->flags : I915_EXEC_BLT;
>>> +
>>> +		ret = __gem_execbuf(fd, &execbuf);
>>> +	}
>>>    	return ret;
>>>    }
>>> @@ -950,7 +993,7 @@ uint64_t emit_blt_ctrl_surf_copy(int fd,
>>>    	igt_assert_f(ahnd, "ctrl-surf-copy supports softpin only\n");
>>>    	igt_assert_f(surf, "ctrl-surf-copy requires data to do ctrl-surf-copy blit\n");
>>> -	alignment = max_t(uint64_t, gem_detect_safe_alignment(fd), 1ull << 16);
>>> +	alignment = max_t(uint64_t, get_default_alignment(fd, surf->driver), 1ull << 16);
>>>    	data.dw00.client = 0x2;
>>>    	data.dw00.opcode = 0x48;
>>> @@ -973,8 +1016,7 @@ uint64_t emit_blt_ctrl_surf_copy(int fd,
>>>    	data.dw04.dst_address_hi = dst_offset >> 32;
>>>    	data.dw04.dst_mocs = surf->dst.mocs;
>>> -	bb = gem_mmap__device_coherent(fd, surf->bb.handle, 0, surf->bb.size,
>>> -				       PROT_READ | PROT_WRITE);
>>> +	bb = bo_map(fd, surf->bb.handle, surf->bb.size, surf->driver);
>>>    	igt_assert(bb_pos + sizeof(data) < surf->bb.size);
>>>    	memcpy(bb + bb_pos, &data, sizeof(data));
>>> @@ -1002,7 +1044,7 @@ uint64_t emit_blt_ctrl_surf_copy(int fd,
>>>    /**
>>>     * blt_ctrl_surf_copy:
>>> - * @fd: drm fd
>>> + * @blt: bldrm fd
>>
>> Emm, the argument seem to still be fd, and there's no blt. A typo?
> 
> I'm not sure how this appeared here. Will fix and send in v4.
> 
>>
>> Thanks for adding the helpers, the code looks nice and tidy (although I'm
>> scared of __special_mode and the like, but there's no better way to do it).
> 
> Yes, I also don't see better way to handle two different drivers here.
> But I believe conditional is readable and easily understandable.

It is readable, so I'm fine with leaving it as it is.

All the best,
Karolina

> 
> --
> Zbigniew
> 
>>
>> All the best,
>> Karolina
>>
>>>     * @ctx: intel_ctx_t context
>>>     * @e: blitter engine for @ctx
>>>     * @ahnd: allocator handle
>>> @@ -1026,32 +1068,37 @@ int blt_ctrl_surf_copy(int fd,
>>>    	igt_assert_f(ahnd, "ctrl-surf-copy supports softpin only\n");
>>>    	igt_assert_f(surf, "ctrl-surf-copy requires data to do ctrl-surf-copy blit\n");
>>> +	igt_assert_neq(surf->driver, 0);
>>> -	alignment = max_t(uint64_t, gem_detect_safe_alignment(fd), 1ull << 16);
>>> +	alignment = max_t(uint64_t, get_default_alignment(fd, surf->driver), 1ull << 16);
>>>    	src_offset = get_offset(ahnd, surf->src.handle, surf->src.size, alignment);
>>>    	dst_offset = get_offset(ahnd, surf->dst.handle, surf->dst.size, alignment);
>>>    	bb_offset = get_offset(ahnd, surf->bb.handle, surf->bb.size, alignment);
>>>    	emit_blt_ctrl_surf_copy(fd, ahnd, surf, 0, true);
>>> -	obj[0].offset = CANONICAL(dst_offset);
>>> -	obj[1].offset = CANONICAL(src_offset);
>>> -	obj[2].offset = CANONICAL(bb_offset);
>>> -	obj[0].handle = surf->dst.handle;
>>> -	obj[1].handle = surf->src.handle;
>>> -	obj[2].handle = surf->bb.handle;
>>> -	obj[0].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_WRITE |
>>> -		       EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
>>> -	obj[1].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
>>> -	obj[2].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
>>> -	execbuf.buffer_count = 3;
>>> -	execbuf.buffers_ptr = to_user_pointer(obj);
>>> -	execbuf.flags = e ? e->flags : I915_EXEC_BLT;
>>> -	execbuf.rsvd1 = ctx ? ctx->id : 0;
>>> -	gem_execbuf(fd, &execbuf);
>>> -	put_offset(ahnd, surf->dst.handle);
>>> -	put_offset(ahnd, surf->src.handle);
>>> -	put_offset(ahnd, surf->bb.handle);
>>> +	if (surf->driver == INTEL_DRIVER_XE) {
>>> +		intel_ctx_xe_exec(ctx, ahnd, CANONICAL(bb_offset));
>>> +	} else {
>>> +		obj[0].offset = CANONICAL(dst_offset);
>>> +		obj[1].offset = CANONICAL(src_offset);
>>> +		obj[2].offset = CANONICAL(bb_offset);
>>> +		obj[0].handle = surf->dst.handle;
>>> +		obj[1].handle = surf->src.handle;
>>> +		obj[2].handle = surf->bb.handle;
>>> +		obj[0].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_WRITE |
>>> +				EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
>>> +		obj[1].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
>>> +		obj[2].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
>>> +		execbuf.buffer_count = 3;
>>> +		execbuf.buffers_ptr = to_user_pointer(obj);
>>> +		execbuf.flags = e ? e->flags : I915_EXEC_BLT;
>>> +		execbuf.rsvd1 = ctx ? ctx->id : 0;
>>> +		gem_execbuf(fd, &execbuf);
>>> +		put_offset(ahnd, surf->dst.handle);
>>> +		put_offset(ahnd, surf->src.handle);
>>> +		put_offset(ahnd, surf->bb.handle);
>>> +	}
>>>    	return 0;
>>>    }
>>> @@ -1208,7 +1255,7 @@ uint64_t emit_blt_fast_copy(int fd,
>>>    	uint32_t bbe = MI_BATCH_BUFFER_END;
>>>    	uint32_t *bb;
>>> -	alignment = gem_detect_safe_alignment(fd);
>>> +	alignment = get_default_alignment(fd, blt->driver);
>>>    	data.dw00.client = 0x2;
>>>    	data.dw00.opcode = 0x42;
>>> @@ -1218,8 +1265,8 @@ uint64_t emit_blt_fast_copy(int fd,
>>>    	data.dw01.dst_pitch = blt->dst.pitch;
>>>    	data.dw01.color_depth = __fast_color_depth(blt->color_depth);
>>> -	data.dw01.dst_memory = __memory_type(blt->dst.region);
>>> -	data.dw01.src_memory = __memory_type(blt->src.region);
>>> +	data.dw01.dst_memory = __memory_type(blt->fd, blt->driver, blt->dst.region);
>>> +	data.dw01.src_memory = __memory_type(blt->fd, blt->driver, blt->src.region);
>>>    	data.dw01.dst_type_y = __new_tile_y_type(blt->dst.tiling) ? 1 : 0;
>>>    	data.dw01.src_type_y = __new_tile_y_type(blt->src.tiling) ? 1 : 0;
>>> @@ -1246,8 +1293,7 @@ uint64_t emit_blt_fast_copy(int fd,
>>>    	data.dw08.src_address_lo = src_offset;
>>>    	data.dw09.src_address_hi = src_offset >> 32;
>>> -	bb = gem_mmap__device_coherent(fd, blt->bb.handle, 0, blt->bb.size,
>>> -				       PROT_READ | PROT_WRITE);
>>> +	bb = bo_map(fd, blt->bb.handle, blt->bb.size, blt->driver);
>>>    	igt_assert(bb_pos + sizeof(data) < blt->bb.size);
>>>    	memcpy(bb + bb_pos, &data, sizeof(data));
>>> @@ -1297,32 +1343,39 @@ int blt_fast_copy(int fd,
>>>    	uint64_t dst_offset, src_offset, bb_offset, alignment;
>>>    	int ret;
>>> -	alignment = gem_detect_safe_alignment(fd);
>>> +	igt_assert_f(ahnd, "fast-copy supports softpin only\n");
>>> +	igt_assert_f(blt, "fast-copy requires data to do fast-copy blit\n");
>>> +	igt_assert_neq(blt->driver, 0);
>>> +	alignment = get_default_alignment(fd, blt->driver);
>>>    	src_offset = get_offset(ahnd, blt->src.handle, blt->src.size, alignment);
>>>    	dst_offset = get_offset(ahnd, blt->dst.handle, blt->dst.size, alignment);
>>>    	bb_offset = get_offset(ahnd, blt->bb.handle, blt->bb.size, alignment);
>>>    	emit_blt_fast_copy(fd, ahnd, blt, 0, true);
>>> -	obj[0].offset = CANONICAL(dst_offset);
>>> -	obj[1].offset = CANONICAL(src_offset);
>>> -	obj[2].offset = CANONICAL(bb_offset);
>>> -	obj[0].handle = blt->dst.handle;
>>> -	obj[1].handle = blt->src.handle;
>>> -	obj[2].handle = blt->bb.handle;
>>> -	obj[0].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_WRITE |
>>> -		       EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
>>> -	obj[1].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
>>> -	obj[2].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
>>> -	execbuf.buffer_count = 3;
>>> -	execbuf.buffers_ptr = to_user_pointer(obj);
>>> -	execbuf.rsvd1 = ctx ? ctx->id : 0;
>>> -	execbuf.flags = e ? e->flags : I915_EXEC_BLT;
>>> -	ret = __gem_execbuf(fd, &execbuf);
>>> -	put_offset(ahnd, blt->dst.handle);
>>> -	put_offset(ahnd, blt->src.handle);
>>> -	put_offset(ahnd, blt->bb.handle);
>>> +	if (blt->driver == INTEL_DRIVER_XE) {
>>> +		intel_ctx_xe_exec(ctx, ahnd, CANONICAL(bb_offset));
>>> +	} else {
>>> +		obj[0].offset = CANONICAL(dst_offset);
>>> +		obj[1].offset = CANONICAL(src_offset);
>>> +		obj[2].offset = CANONICAL(bb_offset);
>>> +		obj[0].handle = blt->dst.handle;
>>> +		obj[1].handle = blt->src.handle;
>>> +		obj[2].handle = blt->bb.handle;
>>> +		obj[0].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_WRITE |
>>> +				EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
>>> +		obj[1].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
>>> +		obj[2].flags = EXEC_OBJECT_PINNED | EXEC_OBJECT_SUPPORTS_48B_ADDRESS;
>>> +		execbuf.buffer_count = 3;
>>> +		execbuf.buffers_ptr = to_user_pointer(obj);
>>> +		execbuf.rsvd1 = ctx ? ctx->id : 0;
>>> +		execbuf.flags = e ? e->flags : I915_EXEC_BLT;
>>> +		ret = __gem_execbuf(fd, &execbuf);
>>> +		put_offset(ahnd, blt->dst.handle);
>>> +		put_offset(ahnd, blt->src.handle);
>>> +		put_offset(ahnd, blt->bb.handle);
>>> +	}
>>>    	return ret;
>>>    }
>>> @@ -1366,16 +1419,21 @@ blt_create_object(const struct blt_copy_data *blt, uint32_t region,
>>>    	obj = calloc(1, sizeof(*obj));
>>>    	obj->size = size;
>>> -	igt_assert(__gem_create_in_memory_regions(blt->fd, &handle,
>>> -						  &size, region) == 0);
>>> +
>>> +	if (blt->driver == INTEL_DRIVER_XE) {
>>> +		size = ALIGN(size, xe_get_default_alignment(blt->fd));
>>> +		handle = xe_bo_create_flags(blt->fd, 0, size, region);
>>> +	} else {
>>> +		igt_assert(__gem_create_in_memory_regions(blt->fd, &handle,
>>> +							  &size, region) == 0);
>>> +	}
>>>    	blt_set_object(obj, handle, size, region, mocs, tiling,
>>>    		       compression, compression_type);
>>>    	blt_set_geom(obj, stride, 0, 0, width, height, 0, 0);
>>>    	if (create_mapping)
>>> -		obj->ptr = gem_mmap__device_coherent(blt->fd, handle, 0, size,
>>> -						     PROT_READ | PROT_WRITE);
>>> +		obj->ptr = bo_map(blt->fd, handle, size, blt->driver);
>>>    	return obj;
>>>    }
>>> @@ -1518,14 +1576,19 @@ void blt_surface_to_png(int fd, uint32_t run_id, const char *fileid,
>>>    	int format;
>>>    	int stride = obj->tiling ? obj->pitch * 4 : obj->pitch;
>>>    	char filename[FILENAME_MAX];
>>> +	bool is_xe = is_xe_device(fd);
>>>    	snprintf(filename, FILENAME_MAX-1, "%d-%s-%s-%ux%u-%s.png",
>>>    		 run_id, fileid, blt_tiling_name(obj->tiling), width, height,
>>>    		 obj->compression ? "compressed" : "uncompressed");
>>> -	if (!map)
>>> -		map = gem_mmap__device_coherent(fd, obj->handle, 0,
>>> -						obj->size, PROT_READ);
>>> +	if (!map) {
>>> +		if (is_xe)
>>> +			map = xe_bo_map(fd, obj->handle, obj->size);
>>> +		else
>>> +			map = gem_mmap__device_coherent(fd, obj->handle, 0,
>>> +							obj->size, PROT_READ);
>>> +	}
>>>    	format = CAIRO_FORMAT_RGB24;
>>>    	surface = cairo_image_surface_create_for_data(map,
>>>    						      format, width, height,
>>> diff --git a/lib/intel_blt.h b/lib/intel_blt.h
>>> index 7516ce8ac7..944e2b4ae7 100644
>>> --- a/lib/intel_blt.h
>>> +++ b/lib/intel_blt.h
>>> @@ -8,7 +8,7 @@
>>>    /**
>>>     * SECTION:intel_blt
>>> - * @short_description: i915 blitter library
>>> + * @short_description: i915/xe blitter library
>>>     * @title: Blitter library
>>>     * @include: intel_blt.h
>>>     *

Re: [igt-dev] [PATCH i-g-t v3 10/17] lib/intel_allocator: Add intel_allocator_bind()

[Karolina Stolarek]

On 12.07.2023 11:33, Zbigniew Kempczyński wrote:
> On Wed, Jul 12, 2023 at 11:13:24AM +0200, Karolina Stolarek wrote:
>> On 11.07.2023 13:20, Zbigniew Kempczyński wrote:
>>> Synchronize allocator state to vm.
>>>
>>> This change allows xe user to execute vm-bind/unbind for allocator
>>> alloc()/free() operations which occurred since last binding/unbinding.
>>> Before doing exec user should call intel_allocator_bind() to ensure
>>> all vma's are in place.
>>>
>>> Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
>>> ---
>>> v2: Rewrite tracking mechanism: previous code uses bind map embedded
>>>       in allocator structure. Unfortunately this wasn't good idea
>>>       - for xe binding everything was fine, but it regress multiprocess/
>>>       multithreaded allocations. Main reason of this was children
>>>       processes couldn't get its reference as this memory was allocated
>>>       on allocator thread (separate process). Currently each child
>>>       contains its own separate tracking maps for ahnd and for each
>>>       ahnd bind map.
>>> v3: - Don't use priv field (we may clean bind_map apriori as failing
>>>         asserts and outdated bind_map is not a problem anymore).
>>>       - Use only vm on tracking ahnd (Karolina)
>>>       - Add braces to avoid checkpatch complaining (Karolina)
>>>       - Use igt_assert_f() instead conditional block (Karolina)
>>> ---
>>>    lib/igt_core.c        |   5 +
>>>    lib/intel_allocator.c | 259 +++++++++++++++++++++++++++++++++++++++++-
>>>    lib/intel_allocator.h |   3 +
>>>    3 files changed, 265 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/lib/igt_core.c b/lib/igt_core.c
>>> index 2ae2cb6883..041e4b3288 100644
>>> --- a/lib/igt_core.c
>>> +++ b/lib/igt_core.c
>>> @@ -74,6 +74,7 @@
>>>    #include "igt_sysrq.h"
>>>    #include "igt_rc.h"
>>>    #include "igt_list.h"
>>> +#include "igt_map.h"
>>>    #include "igt_device_scan.h"
>>>    #include "igt_thread.h"
>>>    #include "runnercomms.h"
>>> @@ -320,6 +321,8 @@ bool test_multi_fork_child;
>>>    /* For allocator purposes */
>>>    pid_t child_pid  = -1;
>>>    __thread pid_t child_tid  = -1;
>>> +struct igt_map *ahnd_map;
>>> +pthread_mutex_t ahnd_map_mutex;
>>>    enum {
>>>    	/*
>>> @@ -2523,6 +2526,8 @@ bool __igt_fork(void)
>>>    	case 0:
>>>    		test_child = true;
>>>    		pthread_mutex_init(&print_mutex, NULL);
>>> +		pthread_mutex_init(&ahnd_map_mutex, NULL);
>>> +		ahnd_map = igt_map_create(igt_map_hash_64, igt_map_equal_64);
>>>    		child_pid = getpid();
>>>    		child_tid = -1;
>>>    		exit_handler_count = 0;
>>> diff --git a/lib/intel_allocator.c b/lib/intel_allocator.c
>>> index cf964ccc41..f0a9b7fb53 100644
>>> --- a/lib/intel_allocator.c
>>> +++ b/lib/intel_allocator.c
>>> @@ -17,8 +17,8 @@
>>>    #include "intel_allocator.h"
>>>    #include "intel_allocator_msgchannel.h"
>>>    #include "xe/xe_query.h"
>>> +#include "xe/xe_util.h"
>>> -//#define ALLOCDBG
>>>    #ifdef ALLOCDBG
>>>    #define alloc_info igt_info
>>>    #define alloc_debug igt_debug
>>> @@ -45,6 +45,14 @@ static inline const char *reqstr(enum reqtype request_type)
>>>    #define alloc_debug(...) {}
>>>    #endif
>>> +#ifdef ALLOCBINDDBG
>>> +#define bind_info igt_info
>>> +#define bind_debug igt_debug
>>> +#else
>>> +#define bind_info(...) {}
>>> +#define bind_debug(...) {}
>>> +#endif
>>> +
>>>    /*
>>>     * We limit allocator space to avoid hang when batch would be
>>>     * pinned in the last page.
>>> @@ -64,6 +72,30 @@ struct handle_entry {
>>>    	struct allocator *al;
>>>    };
>>> +/* For tracking alloc()/free() for Xe */
>>> +struct ahnd_info {
>>> +	int fd;
>>> +	uint64_t ahnd;
>>> +	uint32_t vm;
>>> +	enum intel_driver driver;
>>> +	struct igt_map *bind_map;
>>> +	pthread_mutex_t bind_map_mutex;
>>> +};
>>> +
>>> +enum allocator_bind_op {
>>> +	BOUND,
>>> +	TO_BIND,
>>> +	TO_UNBIND,
>>> +};
>>> +
>>> +struct allocator_object {
>>> +	uint32_t handle;
>>> +	uint64_t offset;
>>> +	uint64_t size;
>>> +
>>> +	enum allocator_bind_op bind_op;
>>> +};
>>> +
>>>    struct intel_allocator *
>>>    intel_allocator_reloc_create(int fd, uint64_t start, uint64_t end);
>>>    struct intel_allocator *
>>> @@ -122,6 +154,13 @@ static pid_t allocator_pid = -1;
>>>    extern pid_t child_pid;
>>>    extern __thread pid_t child_tid;
>>> +/*
>>> + * Track alloc()/free() requires storing in local process which has
>>> + * an access to real drm fd it can work on.
>>> + */
>>> +extern struct igt_map *ahnd_map;
>>> +extern pthread_mutex_t ahnd_map_mutex;
>>> +
>>>    /*
>>>     * - for parent process we have child_pid == -1
>>>     * - for child which calls intel_allocator_init() allocator_pid == child_pid
>>> @@ -837,6 +876,45 @@ void intel_allocator_multiprocess_stop(void)
>>>    	}
>>>    }
>>> +static void track_ahnd(int fd, uint64_t ahnd, uint32_t vm)
>>> +{
>>> +	struct ahnd_info *ainfo;
>>> +
>>> +	pthread_mutex_lock(&ahnd_map_mutex);
>>> +	ainfo = igt_map_search(ahnd_map, &ahnd);
>>> +	if (!ainfo) {
>>> +		ainfo = malloc(sizeof(*ainfo));
>>> +		ainfo->fd = fd;
>>> +		ainfo->ahnd = ahnd;
>>> +		ainfo->vm = vm;
>>> +		ainfo->driver = get_intel_driver(fd);
>>> +		ainfo->bind_map = igt_map_create(igt_map_hash_32, igt_map_equal_32);
>>> +		pthread_mutex_init(&ainfo->bind_map_mutex, NULL);
>>> +		bind_debug("[TRACK AHND] pid: %d, tid: %d, create <fd: %d, "
>>> +			   "ahnd: %llx, vm: %u, driver: %d, ahnd_map: %p, bind_map: %p>\n",
>>> +			   getpid(), gettid(), ainfo->fd,
>>> +			   (long long)ainfo->ahnd, ainfo->vm,
>>> +			   ainfo->driver, ahnd_map, ainfo->bind_map);
>>> +		igt_map_insert(ahnd_map, &ainfo->ahnd, ainfo);
>>> +	}
>>> +
>>> +	pthread_mutex_unlock(&ahnd_map_mutex);
>>> +}
>>> +
>>> +static void untrack_ahnd(uint64_t ahnd)
>>> +{
>>> +	struct ahnd_info *ainfo;
>>> +
>>> +	pthread_mutex_lock(&ahnd_map_mutex);
>>> +	ainfo = igt_map_search(ahnd_map, &ahnd);
>>> +	if (ainfo) {
>>> +		bind_debug("[UNTRACK AHND]: pid: %d, tid: %d, removing ahnd: %llx\n",
>>> +			   getpid(), gettid(), (long long)ahnd);
>>> +		igt_map_remove(ahnd_map, &ahnd, map_entry_free_func);
>>> +	}
>>> +	pthread_mutex_unlock(&ahnd_map_mutex);
>>> +}
>>> +
>>>    static uint64_t __intel_allocator_open_full(int fd, uint32_t ctx,
>>>    					    uint32_t vm,
>>>    					    uint64_t start, uint64_t end,
>>> @@ -895,6 +973,12 @@ static uint64_t __intel_allocator_open_full(int fd, uint32_t ctx,
>>>    	igt_assert(resp.open.allocator_handle);
>>>    	igt_assert(resp.response_type == RESP_OPEN);
>>> +	/*
>>> +	 * Igts mostly uses ctx as id when opening the allocator (i915 legacy).
>>> +	 * If ctx is passed let's use it as an vm id, otherwise use vm.
>>> +	 */
>>> +	track_ahnd(fd, resp.open.allocator_handle, ctx ?: vm);
>>> +
>>>    	return resp.open.allocator_handle;
>>>    }
>>> @@ -1001,6 +1085,8 @@ bool intel_allocator_close(uint64_t allocator_handle)
>>>    	igt_assert(handle_request(&req, &resp) == 0);
>>>    	igt_assert(resp.response_type == RESP_CLOSE);
>>> +	untrack_ahnd(allocator_handle);
>>> +
>>>    	return resp.close.is_empty;
>>>    }
>>> @@ -1034,6 +1120,74 @@ void intel_allocator_get_address_range(uint64_t allocator_handle,
>>>    		*endp = resp.address_range.end;
>>>    }
>>> +static bool is_same(struct allocator_object *obj,
>>> +		    uint32_t handle, uint64_t offset, uint64_t size,
>>> +		    enum allocator_bind_op bind_op)
>>> +{
>>> +	return obj->handle == handle &&	obj->offset == offset && obj->size == size &&
>>> +	       (obj->bind_op == bind_op || obj->bind_op == BOUND);
>>> +}
>>> +
>>> +static void track_object(uint64_t allocator_handle, uint32_t handle,
>>> +			 uint64_t offset, uint64_t size,
>>> +			 enum allocator_bind_op bind_op)
>>> +{
>>> +	struct ahnd_info *ainfo;
>>> +	struct allocator_object *obj;
>>> +
>>> +	bind_debug("[TRACK OBJECT]: [%s] pid: %d, tid: %d, ahnd: %llx, handle: %u, offset: %llx, size: %llx\n",
>>> +		   bind_op == TO_BIND ? "BIND" : "UNBIND",
>>> +		   getpid(), gettid(),
>>> +		   (long long)allocator_handle,
>>> +		   handle, (long long)offset, (long long)size);
>>> +
>>> +	if (offset == ALLOC_INVALID_ADDRESS) {
>>> +		bind_debug("[TRACK OBJECT] => invalid address %llx, skipping tracking\n",
>>> +			   (long long)offset);
>>> +		return;
>>> +	}
>>> +
>>> +	pthread_mutex_lock(&ahnd_map_mutex);
>>> +	ainfo = igt_map_search(ahnd_map, &allocator_handle);
>>> +	pthread_mutex_unlock(&ahnd_map_mutex);
>>> +	igt_assert_f(ainfo, "[TRACK OBJECT] => MISSING ahnd %llx <=\n",
>>> +		     (long long)allocator_handle);
>>> +
>>> +	if (ainfo->driver == INTEL_DRIVER_I915)
>>> +		return; /* no-op for i915, at least for now */
>>> +
>>> +	pthread_mutex_lock(&ainfo->bind_map_mutex);
>>> +	obj = igt_map_search(ainfo->bind_map, &handle);
>>> +	if (obj) {
>>> +		/*
>>> +		 * User may call alloc() couple of times, check object is the
>>> +		 * same. For free() there's simple case, just remove from
>>> +		 * bind_map.
>>> +		 */
>>> +		if (bind_op == TO_BIND) {
>>> +			igt_assert_eq(is_same(obj, handle, offset, size, bind_op), true);
>>> +		} else if (bind_op == TO_UNBIND) {
>>> +			if (obj->bind_op == TO_BIND)
>>> +				igt_map_remove(ainfo->bind_map, &obj->handle, map_entry_free_func);
>>> +			else if (obj->bind_op == BOUND)
>>> +				obj->bind_op = bind_op;
>>> +		}
>>> +	} else {
>>> +		/* Ignore to unbind bo which wasn't previously inserted */
>>> +		if (bind_op == TO_UNBIND)
>>> +			goto out;
>>> +
>>> +		obj = calloc(1, sizeof(*obj));
>>> +		obj->handle = handle;
>>> +		obj->offset = offset;
>>> +		obj->size = size;
>>> +		obj->bind_op = bind_op;
>>> +		igt_map_insert(ainfo->bind_map, &obj->handle, obj);
>>> +	}
>>> +out:
>>> +	pthread_mutex_unlock(&ainfo->bind_map_mutex);
>>> +}
>>> +
>>>    /**
>>>     * __intel_allocator_alloc:
>>>     * @allocator_handle: handle to an allocator
>>> @@ -1065,6 +1219,8 @@ uint64_t __intel_allocator_alloc(uint64_t allocator_handle, uint32_t handle,
>>>    	igt_assert(handle_request(&req, &resp) == 0);
>>>    	igt_assert(resp.response_type == RESP_ALLOC);
>>> +	track_object(allocator_handle, handle, resp.alloc.offset, size, TO_BIND);
>>> +
>>>    	return resp.alloc.offset;
>>>    }
>>> @@ -1142,6 +1298,8 @@ bool intel_allocator_free(uint64_t allocator_handle, uint32_t handle)
>>>    	igt_assert(handle_request(&req, &resp) == 0);
>>>    	igt_assert(resp.response_type == RESP_FREE);
>>> +	track_object(allocator_handle, handle, 0, 0, TO_UNBIND);
>>> +
>>>    	return resp.free.freed;
>>>    }
>>> @@ -1326,6 +1484,84 @@ void intel_allocator_print(uint64_t allocator_handle)
>>>    	}
>>>    }
>>> +static void __xe_op_bind(struct ahnd_info *ainfo, uint32_t sync_in, uint32_t sync_out)
>>> +{
>>> +	struct allocator_object *obj;
>>> +	struct igt_map_entry *pos;
>>> +	struct igt_list_head obj_list;
>>> +	struct xe_object *entry, *tmp;
>>> +
>>> +	IGT_INIT_LIST_HEAD(&obj_list);
>>> +
>>> +	pthread_mutex_lock(&ainfo->bind_map_mutex);
>>> +	igt_map_foreach(ainfo->bind_map, pos) {
>>> +		obj = pos->data;
>>> +
>>> +		if (obj->bind_op == BOUND)
>>> +			continue;
>>> +
>>> +		bind_info("= [vm: %u] %s => %u %lx %lx\n",
>>> +			  ainfo->vm,
>>> +			  obj->bind_op == TO_BIND ? "TO BIND" : "TO UNBIND",
>>> +			  obj->handle, obj->offset,
>>> +			  obj->size);
>>> +
>>> +		entry = malloc(sizeof(*entry));
>>> +		entry->handle = obj->handle;
>>> +		entry->offset = obj->offset;
>>> +		entry->size = obj->size;
>>> +		entry->bind_op = obj->bind_op == TO_BIND ? XE_OBJECT_BIND :
>>> +							   XE_OBJECT_UNBIND;
>>> +		igt_list_add(&entry->link, &obj_list);
>>> +
>>> +		/*
>>> +		 * We clean bind_map even before calling bind/unbind
>>> +		 * as all binding operations asserts in case of error.
>>> +		 */
>>> +		if (obj->bind_op == TO_BIND)
>>> +			obj->bind_op = BOUND;
>>> +		else
>>> +			igt_map_remove(ainfo->bind_map, &obj->handle,
>>> +				       map_entry_free_func);
>>
>> I don't quite understand why we can clean up the map even before the actual
>> bind/unbind happens. If an assert fails, it means that vm_(un)bind failed,
>> so the actual operation didn't succeed, and that state change is incorrect.
>> I believe that there's a reason we can do it this way, I just don't
>> understand it.
> 
> I allocate separate list of xe_objects which is an argument to
> xe_bind_unbind_async(). As list is detached (separate) from the bind_map
> I may apriori do cleanup in the map. If bind/unbind is unsuccessful
> outdated bind_map is the smallest problem here.

I'm aware they are separate entities, I just wonder what is the benefit 
of doing it here. Are there any dangers of removing that binding before 
it's actually unbound?

All the best,
Karolina

> On assert path igt_fail()
> calls intel_allocator_init() which just frees contents on ahnd_map[]->bind_map.
> In children there's simpler because they just dies. Real problem is we
> still have dangling vm (no xe_vm_destroy()).
> So I decided to update the bind_map before calling bind/unbind as on failure
> doesn't matter what it will contain.
> 
> --
> Zbigniew
> 
>>
>> Apart from that one thing, the patch looks good to me.
>>
>> All the best,
>> Karolina
>>
>>> +	}
>>> +	pthread_mutex_unlock(&ainfo->bind_map_mutex);
>>> +
>>> +	xe_bind_unbind_async(ainfo->fd, ainfo->vm, 0, &obj_list, sync_in, sync_out);
>>> +
>>> +	igt_list_for_each_entry_safe(entry, tmp, &obj_list, link) {
>>> +		igt_list_del(&entry->link);
>>> +		free(entry);
>>> +	}
>>> +}
>>> +
>>> +/**
>>> + * intel_allocator_bind:
>>> + * @allocator_handle: handle to an allocator
>>> + * @sync_in: syncobj (fence-in)
>>> + * @sync_out: syncobj (fence-out)
>>> + *
>>> + * Function binds and unbinds all objects added to the allocator which weren't
>>> + * previously binded/unbinded.
>>> + *
>>> + **/
>>> +void intel_allocator_bind(uint64_t allocator_handle,
>>> +			  uint32_t sync_in, uint32_t sync_out)
>>> +{
>>> +	struct ahnd_info *ainfo;
>>> +
>>> +	pthread_mutex_lock(&ahnd_map_mutex);
>>> +	ainfo = igt_map_search(ahnd_map, &allocator_handle);
>>> +	pthread_mutex_unlock(&ahnd_map_mutex);
>>> +	igt_assert(ainfo);
>>> +
>>> +	/*
>>> +	 * We collect bind/unbind operations on alloc()/free() to do group
>>> +	 * operation getting @sync_in as syncobj handle (fence-in). If user
>>> +	 * passes 0 as @sync_out we bind/unbind synchronously.
>>> +	 */
>>> +	__xe_op_bind(ainfo, sync_in, sync_out);
>>> +}
>>> +
>>>    static int equal_handles(const void *key1, const void *key2)
>>>    {
>>>    	const struct handle_entry *h1 = key1, *h2 = key2;
>>> @@ -1383,6 +1619,22 @@ static void __free_maps(struct igt_map *map, bool close_allocators)
>>>    	igt_map_destroy(map, map_entry_free_func);
>>>    }
>>> +static void __free_ahnd_map(void)
>>> +{
>>> +	struct igt_map_entry *pos;
>>> +	struct ahnd_info *ainfo;
>>> +
>>> +	if (!ahnd_map)
>>> +		return;
>>> +
>>> +	igt_map_foreach(ahnd_map, pos) {
>>> +		ainfo = pos->data;
>>> +		igt_map_destroy(ainfo->bind_map, map_entry_free_func);
>>> +	}
>>> +
>>> +	igt_map_destroy(ahnd_map, map_entry_free_func);
>>> +}
>>> +
>>>    /**
>>>     * intel_allocator_init:
>>>     *
>>> @@ -1400,12 +1652,15 @@ void intel_allocator_init(void)
>>>    	__free_maps(handles, true);
>>>    	__free_maps(ctx_map, false);
>>>    	__free_maps(vm_map, false);
>>> +	__free_ahnd_map();
>>>    	atomic_init(&next_handle, 1);
>>>    	handles = igt_map_create(hash_handles, equal_handles);
>>>    	ctx_map = igt_map_create(hash_instance, equal_ctx);
>>>    	vm_map = igt_map_create(hash_instance, equal_vm);
>>> -	igt_assert(handles && ctx_map && vm_map);
>>> +	pthread_mutex_init(&ahnd_map_mutex, NULL);
>>> +	ahnd_map = igt_map_create(igt_map_hash_64, igt_map_equal_64);
>>> +	igt_assert(handles && ctx_map && vm_map && ahnd_map);
>>>    	channel = intel_allocator_get_msgchannel(CHANNEL_SYSVIPC_MSGQUEUE);
>>>    }
>>> diff --git a/lib/intel_allocator.h b/lib/intel_allocator.h
>>> index 3ec74f6191..f9ff7f1cc9 100644
>>> --- a/lib/intel_allocator.h
>>> +++ b/lib/intel_allocator.h
>>> @@ -210,6 +210,9 @@ bool intel_allocator_reserve_if_not_allocated(uint64_t allocator_handle,
>>>    void intel_allocator_print(uint64_t allocator_handle);
>>> +void intel_allocator_bind(uint64_t allocator_handle,
>>> +			  uint32_t sync_in, uint32_t sync_out);
>>> +
>>>    #define ALLOC_INVALID_ADDRESS (-1ull)
>>>    #define INTEL_ALLOCATOR_NONE   0
>>>    #define INTEL_ALLOCATOR_RELOC  1

Re: [igt-dev] [PATCH i-g-t v3 10/17] lib/intel_allocator: Add intel_allocator_bind()

[Zbigniew Kempczyński]

On Wed, Jul 12, 2023 at 12:39:42PM +0200, Karolina Stolarek wrote:
> On 12.07.2023 11:33, Zbigniew Kempczyński wrote:
> > On Wed, Jul 12, 2023 at 11:13:24AM +0200, Karolina Stolarek wrote:
> > > On 11.07.2023 13:20, Zbigniew Kempczyński wrote:
> > > > Synchronize allocator state to vm.
> > > > 
> > > > This change allows xe user to execute vm-bind/unbind for allocator
> > > > alloc()/free() operations which occurred since last binding/unbinding.
> > > > Before doing exec user should call intel_allocator_bind() to ensure
> > > > all vma's are in place.
> > > > 
> > > > Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
> > > > ---
> > > > v2: Rewrite tracking mechanism: previous code uses bind map embedded
> > > >       in allocator structure. Unfortunately this wasn't good idea
> > > >       - for xe binding everything was fine, but it regress multiprocess/
> > > >       multithreaded allocations. Main reason of this was children
> > > >       processes couldn't get its reference as this memory was allocated
> > > >       on allocator thread (separate process). Currently each child
> > > >       contains its own separate tracking maps for ahnd and for each
> > > >       ahnd bind map.
> > > > v3: - Don't use priv field (we may clean bind_map apriori as failing
> > > >         asserts and outdated bind_map is not a problem anymore).
> > > >       - Use only vm on tracking ahnd (Karolina)
> > > >       - Add braces to avoid checkpatch complaining (Karolina)
> > > >       - Use igt_assert_f() instead conditional block (Karolina)
> > > > ---
> > > >    lib/igt_core.c        |   5 +
> > > >    lib/intel_allocator.c | 259 +++++++++++++++++++++++++++++++++++++++++-
> > > >    lib/intel_allocator.h |   3 +
> > > >    3 files changed, 265 insertions(+), 2 deletions(-)
> > > > 
> > > > diff --git a/lib/igt_core.c b/lib/igt_core.c
> > > > index 2ae2cb6883..041e4b3288 100644
> > > > --- a/lib/igt_core.c
> > > > +++ b/lib/igt_core.c
> > > > @@ -74,6 +74,7 @@
> > > >    #include "igt_sysrq.h"
> > > >    #include "igt_rc.h"
> > > >    #include "igt_list.h"
> > > > +#include "igt_map.h"
> > > >    #include "igt_device_scan.h"
> > > >    #include "igt_thread.h"
> > > >    #include "runnercomms.h"
> > > > @@ -320,6 +321,8 @@ bool test_multi_fork_child;
> > > >    /* For allocator purposes */
> > > >    pid_t child_pid  = -1;
> > > >    __thread pid_t child_tid  = -1;
> > > > +struct igt_map *ahnd_map;
> > > > +pthread_mutex_t ahnd_map_mutex;
> > > >    enum {
> > > >    	/*
> > > > @@ -2523,6 +2526,8 @@ bool __igt_fork(void)
> > > >    	case 0:
> > > >    		test_child = true;
> > > >    		pthread_mutex_init(&print_mutex, NULL);
> > > > +		pthread_mutex_init(&ahnd_map_mutex, NULL);
> > > > +		ahnd_map = igt_map_create(igt_map_hash_64, igt_map_equal_64);
> > > >    		child_pid = getpid();
> > > >    		child_tid = -1;
> > > >    		exit_handler_count = 0;
> > > > diff --git a/lib/intel_allocator.c b/lib/intel_allocator.c
> > > > index cf964ccc41..f0a9b7fb53 100644
> > > > --- a/lib/intel_allocator.c
> > > > +++ b/lib/intel_allocator.c
> > > > @@ -17,8 +17,8 @@
> > > >    #include "intel_allocator.h"
> > > >    #include "intel_allocator_msgchannel.h"
> > > >    #include "xe/xe_query.h"
> > > > +#include "xe/xe_util.h"
> > > > -//#define ALLOCDBG
> > > >    #ifdef ALLOCDBG
> > > >    #define alloc_info igt_info
> > > >    #define alloc_debug igt_debug
> > > > @@ -45,6 +45,14 @@ static inline const char *reqstr(enum reqtype request_type)
> > > >    #define alloc_debug(...) {}
> > > >    #endif
> > > > +#ifdef ALLOCBINDDBG
> > > > +#define bind_info igt_info
> > > > +#define bind_debug igt_debug
> > > > +#else
> > > > +#define bind_info(...) {}
> > > > +#define bind_debug(...) {}
> > > > +#endif
> > > > +
> > > >    /*
> > > >     * We limit allocator space to avoid hang when batch would be
> > > >     * pinned in the last page.
> > > > @@ -64,6 +72,30 @@ struct handle_entry {
> > > >    	struct allocator *al;
> > > >    };
> > > > +/* For tracking alloc()/free() for Xe */
> > > > +struct ahnd_info {
> > > > +	int fd;
> > > > +	uint64_t ahnd;
> > > > +	uint32_t vm;
> > > > +	enum intel_driver driver;
> > > > +	struct igt_map *bind_map;
> > > > +	pthread_mutex_t bind_map_mutex;
> > > > +};
> > > > +
> > > > +enum allocator_bind_op {
> > > > +	BOUND,
> > > > +	TO_BIND,
> > > > +	TO_UNBIND,
> > > > +};
> > > > +
> > > > +struct allocator_object {
> > > > +	uint32_t handle;
> > > > +	uint64_t offset;
> > > > +	uint64_t size;
> > > > +
> > > > +	enum allocator_bind_op bind_op;
> > > > +};
> > > > +
> > > >    struct intel_allocator *
> > > >    intel_allocator_reloc_create(int fd, uint64_t start, uint64_t end);
> > > >    struct intel_allocator *
> > > > @@ -122,6 +154,13 @@ static pid_t allocator_pid = -1;
> > > >    extern pid_t child_pid;
> > > >    extern __thread pid_t child_tid;
> > > > +/*
> > > > + * Track alloc()/free() requires storing in local process which has
> > > > + * an access to real drm fd it can work on.
> > > > + */
> > > > +extern struct igt_map *ahnd_map;
> > > > +extern pthread_mutex_t ahnd_map_mutex;
> > > > +
> > > >    /*
> > > >     * - for parent process we have child_pid == -1
> > > >     * - for child which calls intel_allocator_init() allocator_pid == child_pid
> > > > @@ -837,6 +876,45 @@ void intel_allocator_multiprocess_stop(void)
> > > >    	}
> > > >    }
> > > > +static void track_ahnd(int fd, uint64_t ahnd, uint32_t vm)
> > > > +{
> > > > +	struct ahnd_info *ainfo;
> > > > +
> > > > +	pthread_mutex_lock(&ahnd_map_mutex);
> > > > +	ainfo = igt_map_search(ahnd_map, &ahnd);
> > > > +	if (!ainfo) {
> > > > +		ainfo = malloc(sizeof(*ainfo));
> > > > +		ainfo->fd = fd;
> > > > +		ainfo->ahnd = ahnd;
> > > > +		ainfo->vm = vm;
> > > > +		ainfo->driver = get_intel_driver(fd);
> > > > +		ainfo->bind_map = igt_map_create(igt_map_hash_32, igt_map_equal_32);
> > > > +		pthread_mutex_init(&ainfo->bind_map_mutex, NULL);
> > > > +		bind_debug("[TRACK AHND] pid: %d, tid: %d, create <fd: %d, "
> > > > +			   "ahnd: %llx, vm: %u, driver: %d, ahnd_map: %p, bind_map: %p>\n",
> > > > +			   getpid(), gettid(), ainfo->fd,
> > > > +			   (long long)ainfo->ahnd, ainfo->vm,
> > > > +			   ainfo->driver, ahnd_map, ainfo->bind_map);
> > > > +		igt_map_insert(ahnd_map, &ainfo->ahnd, ainfo);
> > > > +	}
> > > > +
> > > > +	pthread_mutex_unlock(&ahnd_map_mutex);
> > > > +}
> > > > +
> > > > +static void untrack_ahnd(uint64_t ahnd)
> > > > +{
> > > > +	struct ahnd_info *ainfo;
> > > > +
> > > > +	pthread_mutex_lock(&ahnd_map_mutex);
> > > > +	ainfo = igt_map_search(ahnd_map, &ahnd);
> > > > +	if (ainfo) {
> > > > +		bind_debug("[UNTRACK AHND]: pid: %d, tid: %d, removing ahnd: %llx\n",
> > > > +			   getpid(), gettid(), (long long)ahnd);
> > > > +		igt_map_remove(ahnd_map, &ahnd, map_entry_free_func);
> > > > +	}
> > > > +	pthread_mutex_unlock(&ahnd_map_mutex);
> > > > +}
> > > > +
> > > >    static uint64_t __intel_allocator_open_full(int fd, uint32_t ctx,
> > > >    					    uint32_t vm,
> > > >    					    uint64_t start, uint64_t end,
> > > > @@ -895,6 +973,12 @@ static uint64_t __intel_allocator_open_full(int fd, uint32_t ctx,
> > > >    	igt_assert(resp.open.allocator_handle);
> > > >    	igt_assert(resp.response_type == RESP_OPEN);
> > > > +	/*
> > > > +	 * Igts mostly uses ctx as id when opening the allocator (i915 legacy).
> > > > +	 * If ctx is passed let's use it as an vm id, otherwise use vm.
> > > > +	 */
> > > > +	track_ahnd(fd, resp.open.allocator_handle, ctx ?: vm);
> > > > +
> > > >    	return resp.open.allocator_handle;
> > > >    }
> > > > @@ -1001,6 +1085,8 @@ bool intel_allocator_close(uint64_t allocator_handle)
> > > >    	igt_assert(handle_request(&req, &resp) == 0);
> > > >    	igt_assert(resp.response_type == RESP_CLOSE);
> > > > +	untrack_ahnd(allocator_handle);
> > > > +
> > > >    	return resp.close.is_empty;
> > > >    }
> > > > @@ -1034,6 +1120,74 @@ void intel_allocator_get_address_range(uint64_t allocator_handle,
> > > >    		*endp = resp.address_range.end;
> > > >    }
> > > > +static bool is_same(struct allocator_object *obj,
> > > > +		    uint32_t handle, uint64_t offset, uint64_t size,
> > > > +		    enum allocator_bind_op bind_op)
> > > > +{
> > > > +	return obj->handle == handle &&	obj->offset == offset && obj->size == size &&
> > > > +	       (obj->bind_op == bind_op || obj->bind_op == BOUND);
> > > > +}
> > > > +
> > > > +static void track_object(uint64_t allocator_handle, uint32_t handle,
> > > > +			 uint64_t offset, uint64_t size,
> > > > +			 enum allocator_bind_op bind_op)
> > > > +{
> > > > +	struct ahnd_info *ainfo;
> > > > +	struct allocator_object *obj;
> > > > +
> > > > +	bind_debug("[TRACK OBJECT]: [%s] pid: %d, tid: %d, ahnd: %llx, handle: %u, offset: %llx, size: %llx\n",
> > > > +		   bind_op == TO_BIND ? "BIND" : "UNBIND",
> > > > +		   getpid(), gettid(),
> > > > +		   (long long)allocator_handle,
> > > > +		   handle, (long long)offset, (long long)size);
> > > > +
> > > > +	if (offset == ALLOC_INVALID_ADDRESS) {
> > > > +		bind_debug("[TRACK OBJECT] => invalid address %llx, skipping tracking\n",
> > > > +			   (long long)offset);
> > > > +		return;
> > > > +	}
> > > > +
> > > > +	pthread_mutex_lock(&ahnd_map_mutex);
> > > > +	ainfo = igt_map_search(ahnd_map, &allocator_handle);
> > > > +	pthread_mutex_unlock(&ahnd_map_mutex);
> > > > +	igt_assert_f(ainfo, "[TRACK OBJECT] => MISSING ahnd %llx <=\n",
> > > > +		     (long long)allocator_handle);
> > > > +
> > > > +	if (ainfo->driver == INTEL_DRIVER_I915)
> > > > +		return; /* no-op for i915, at least for now */
> > > > +
> > > > +	pthread_mutex_lock(&ainfo->bind_map_mutex);
> > > > +	obj = igt_map_search(ainfo->bind_map, &handle);
> > > > +	if (obj) {
> > > > +		/*
> > > > +		 * User may call alloc() couple of times, check object is the
> > > > +		 * same. For free() there's simple case, just remove from
> > > > +		 * bind_map.
> > > > +		 */
> > > > +		if (bind_op == TO_BIND) {
> > > > +			igt_assert_eq(is_same(obj, handle, offset, size, bind_op), true);
> > > > +		} else if (bind_op == TO_UNBIND) {
> > > > +			if (obj->bind_op == TO_BIND)
> > > > +				igt_map_remove(ainfo->bind_map, &obj->handle, map_entry_free_func);
> > > > +			else if (obj->bind_op == BOUND)
> > > > +				obj->bind_op = bind_op;
> > > > +		}
> > > > +	} else {
> > > > +		/* Ignore to unbind bo which wasn't previously inserted */
> > > > +		if (bind_op == TO_UNBIND)
> > > > +			goto out;
> > > > +
> > > > +		obj = calloc(1, sizeof(*obj));
> > > > +		obj->handle = handle;
> > > > +		obj->offset = offset;
> > > > +		obj->size = size;
> > > > +		obj->bind_op = bind_op;
> > > > +		igt_map_insert(ainfo->bind_map, &obj->handle, obj);
> > > > +	}
> > > > +out:
> > > > +	pthread_mutex_unlock(&ainfo->bind_map_mutex);
> > > > +}
> > > > +
> > > >    /**
> > > >     * __intel_allocator_alloc:
> > > >     * @allocator_handle: handle to an allocator
> > > > @@ -1065,6 +1219,8 @@ uint64_t __intel_allocator_alloc(uint64_t allocator_handle, uint32_t handle,
> > > >    	igt_assert(handle_request(&req, &resp) == 0);
> > > >    	igt_assert(resp.response_type == RESP_ALLOC);
> > > > +	track_object(allocator_handle, handle, resp.alloc.offset, size, TO_BIND);
> > > > +
> > > >    	return resp.alloc.offset;
> > > >    }
> > > > @@ -1142,6 +1298,8 @@ bool intel_allocator_free(uint64_t allocator_handle, uint32_t handle)
> > > >    	igt_assert(handle_request(&req, &resp) == 0);
> > > >    	igt_assert(resp.response_type == RESP_FREE);
> > > > +	track_object(allocator_handle, handle, 0, 0, TO_UNBIND);
> > > > +
> > > >    	return resp.free.freed;
> > > >    }
> > > > @@ -1326,6 +1484,84 @@ void intel_allocator_print(uint64_t allocator_handle)
> > > >    	}
> > > >    }
> > > > +static void __xe_op_bind(struct ahnd_info *ainfo, uint32_t sync_in, uint32_t sync_out)
> > > > +{
> > > > +	struct allocator_object *obj;
> > > > +	struct igt_map_entry *pos;
> > > > +	struct igt_list_head obj_list;
> > > > +	struct xe_object *entry, *tmp;
> > > > +
> > > > +	IGT_INIT_LIST_HEAD(&obj_list);
> > > > +
> > > > +	pthread_mutex_lock(&ainfo->bind_map_mutex);
> > > > +	igt_map_foreach(ainfo->bind_map, pos) {
> > > > +		obj = pos->data;
> > > > +
> > > > +		if (obj->bind_op == BOUND)
> > > > +			continue;
> > > > +
> > > > +		bind_info("= [vm: %u] %s => %u %lx %lx\n",
> > > > +			  ainfo->vm,
> > > > +			  obj->bind_op == TO_BIND ? "TO BIND" : "TO UNBIND",
> > > > +			  obj->handle, obj->offset,
> > > > +			  obj->size);
> > > > +
> > > > +		entry = malloc(sizeof(*entry));
> > > > +		entry->handle = obj->handle;
> > > > +		entry->offset = obj->offset;
> > > > +		entry->size = obj->size;
> > > > +		entry->bind_op = obj->bind_op == TO_BIND ? XE_OBJECT_BIND :
> > > > +							   XE_OBJECT_UNBIND;
> > > > +		igt_list_add(&entry->link, &obj_list);
> > > > +
> > > > +		/*
> > > > +		 * We clean bind_map even before calling bind/unbind
> > > > +		 * as all binding operations asserts in case of error.
> > > > +		 */
> > > > +		if (obj->bind_op == TO_BIND)
> > > > +			obj->bind_op = BOUND;
> > > > +		else
> > > > +			igt_map_remove(ainfo->bind_map, &obj->handle,
> > > > +				       map_entry_free_func);
> > > 
> > > I don't quite understand why we can clean up the map even before the actual
> > > bind/unbind happens. If an assert fails, it means that vm_(un)bind failed,
> > > so the actual operation didn't succeed, and that state change is incorrect.
> > > I believe that there's a reason we can do it this way, I just don't
> > > understand it.
> > 
> > I allocate separate list of xe_objects which is an argument to
> > xe_bind_unbind_async(). As list is detached (separate) from the bind_map
> > I may apriori do cleanup in the map. If bind/unbind is unsuccessful
> > outdated bind_map is the smallest problem here.
> 
> I'm aware they are separate entities, I just wonder what is the benefit of
> doing it here. Are there any dangers of removing that binding before it's
> actually unbound?

Benefit is that I don't need to track xe_object -> allocator_object or keep
backward reference to update the bind_map. Unsuccessful bind/unbind asserts
and what the bind_map contains doesn't matter. We enter failure path so igt
core is cleaning up preparing to run next subtest.

From my perspective if user is properly handling offsets and calls
intel_allocator_bind() before the exec operation no assert can occur. I did
profound test in api_intel_allocator@two-level-inception adding vm for xe
and binding/unbinding calling intel_allocator_bind(). For RELOC stress
test worked properly (there's no risk of reusing offset returned to the
pool by other child/thread). For SIMPLE we may encounter situation where
one offset is returned in some child/thread, unbind didn't happened yet,
another child/thread allocated getting same offset so binding it will fail
(vm still has unbounded offset). As RELOC doesn't reuse same offset (until
wrapping around) no risk of overlapping occurs (almost ;).

Anyway - it is possible to extend the allocator to be more "transactional"
and aware what offsets are still not unbounded but I decided to not add this
(at least) now. Especially for RELOC (mostly used) it doesn't add too much
value. And for SIMPLE it requires deferred releasing after unbind is complete
(what is also problematic if you want to allow user to pass its own syncobjs).

--
Zbigniew

> 
> All the best,
> Karolina
> 
> > On assert path igt_fail()
> > calls intel_allocator_init() which just frees contents on ahnd_map[]->bind_map.
> > In children there's simpler because they just dies. Real problem is we
> > still have dangling vm (no xe_vm_destroy()).
> > So I decided to update the bind_map before calling bind/unbind as on failure
> > doesn't matter what it will contain.
> > 
> > --
> > Zbigniew
> > 
> > > 
> > > Apart from that one thing, the patch looks good to me.
> > > 
> > > All the best,
> > > Karolina
> > > 
> > > > +	}
> > > > +	pthread_mutex_unlock(&ainfo->bind_map_mutex);
> > > > +
> > > > +	xe_bind_unbind_async(ainfo->fd, ainfo->vm, 0, &obj_list, sync_in, sync_out);
> > > > +
> > > > +	igt_list_for_each_entry_safe(entry, tmp, &obj_list, link) {
> > > > +		igt_list_del(&entry->link);
> > > > +		free(entry);
> > > > +	}
> > > > +}
> > > > +
> > > > +/**
> > > > + * intel_allocator_bind:
> > > > + * @allocator_handle: handle to an allocator
> > > > + * @sync_in: syncobj (fence-in)
> > > > + * @sync_out: syncobj (fence-out)
> > > > + *
> > > > + * Function binds and unbinds all objects added to the allocator which weren't
> > > > + * previously binded/unbinded.
> > > > + *
> > > > + **/
> > > > +void intel_allocator_bind(uint64_t allocator_handle,
> > > > +			  uint32_t sync_in, uint32_t sync_out)
> > > > +{
> > > > +	struct ahnd_info *ainfo;
> > > > +
> > > > +	pthread_mutex_lock(&ahnd_map_mutex);
> > > > +	ainfo = igt_map_search(ahnd_map, &allocator_handle);
> > > > +	pthread_mutex_unlock(&ahnd_map_mutex);
> > > > +	igt_assert(ainfo);
> > > > +
> > > > +	/*
> > > > +	 * We collect bind/unbind operations on alloc()/free() to do group
> > > > +	 * operation getting @sync_in as syncobj handle (fence-in). If user
> > > > +	 * passes 0 as @sync_out we bind/unbind synchronously.
> > > > +	 */
> > > > +	__xe_op_bind(ainfo, sync_in, sync_out);
> > > > +}
> > > > +
> > > >    static int equal_handles(const void *key1, const void *key2)
> > > >    {
> > > >    	const struct handle_entry *h1 = key1, *h2 = key2;
> > > > @@ -1383,6 +1619,22 @@ static void __free_maps(struct igt_map *map, bool close_allocators)
> > > >    	igt_map_destroy(map, map_entry_free_func);
> > > >    }
> > > > +static void __free_ahnd_map(void)
> > > > +{
> > > > +	struct igt_map_entry *pos;
> > > > +	struct ahnd_info *ainfo;
> > > > +
> > > > +	if (!ahnd_map)
> > > > +		return;
> > > > +
> > > > +	igt_map_foreach(ahnd_map, pos) {
> > > > +		ainfo = pos->data;
> > > > +		igt_map_destroy(ainfo->bind_map, map_entry_free_func);
> > > > +	}
> > > > +
> > > > +	igt_map_destroy(ahnd_map, map_entry_free_func);
> > > > +}
> > > > +
> > > >    /**
> > > >     * intel_allocator_init:
> > > >     *
> > > > @@ -1400,12 +1652,15 @@ void intel_allocator_init(void)
> > > >    	__free_maps(handles, true);
> > > >    	__free_maps(ctx_map, false);
> > > >    	__free_maps(vm_map, false);
> > > > +	__free_ahnd_map();
> > > >    	atomic_init(&next_handle, 1);
> > > >    	handles = igt_map_create(hash_handles, equal_handles);
> > > >    	ctx_map = igt_map_create(hash_instance, equal_ctx);
> > > >    	vm_map = igt_map_create(hash_instance, equal_vm);
> > > > -	igt_assert(handles && ctx_map && vm_map);
> > > > +	pthread_mutex_init(&ahnd_map_mutex, NULL);
> > > > +	ahnd_map = igt_map_create(igt_map_hash_64, igt_map_equal_64);
> > > > +	igt_assert(handles && ctx_map && vm_map && ahnd_map);
> > > >    	channel = intel_allocator_get_msgchannel(CHANNEL_SYSVIPC_MSGQUEUE);
> > > >    }
> > > > diff --git a/lib/intel_allocator.h b/lib/intel_allocator.h
> > > > index 3ec74f6191..f9ff7f1cc9 100644
> > > > --- a/lib/intel_allocator.h
> > > > +++ b/lib/intel_allocator.h
> > > > @@ -210,6 +210,9 @@ bool intel_allocator_reserve_if_not_allocated(uint64_t allocator_handle,
> > > >    void intel_allocator_print(uint64_t allocator_handle);
> > > > +void intel_allocator_bind(uint64_t allocator_handle,
> > > > +			  uint32_t sync_in, uint32_t sync_out);
> > > > +
> > > >    #define ALLOC_INVALID_ADDRESS (-1ull)
> > > >    #define INTEL_ALLOCATOR_NONE   0
> > > >    #define INTEL_ALLOCATOR_RELOC  1

Re: [igt-dev] [PATCH i-g-t v3 15/17] tests/xe_exercise_blt: Check blitter library fast-copy for Xe

[Zbigniew Kempczyński]

On Wed, Jul 12, 2023 at 12:18:23PM +0200, Karolina Stolarek wrote:
> On 11.07.2023 13:20, Zbigniew Kempczyński wrote:
> > Port this test to work on xe. Instead of adding conditional code for
> > xe code which would decrease readability this is new test for xe.
> > 
> > Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
> > ---
> > v3: - Explicitely release offsets and unbind (Karolina)
> >      - Use has-fast-copy instead of copied has-block-copy check (Karolina)
> > ---
> >   tests/meson.build          |   1 +
> >   tests/xe/xe_exercise_blt.c | 377 +++++++++++++++++++++++++++++++++++++
> >   2 files changed, 378 insertions(+)
> >   create mode 100644 tests/xe/xe_exercise_blt.c
> > 
> > diff --git a/tests/meson.build b/tests/meson.build
> > index da8653fabc..3eddb2fb4e 100644
> > --- a/tests/meson.build
> > +++ b/tests/meson.build
> > @@ -276,6 +276,7 @@ xe_progs = [
> >   	'xe_exec_reset',
> >   	'xe_exec_store',
> >   	'xe_exec_threads',
> > +	'xe_exercise_blt',
> >   	'xe_gpgpu_fill',
> >   	'xe_guc_pc',
> >   	'xe_huc_copy',
> > diff --git a/tests/xe/xe_exercise_blt.c b/tests/xe/xe_exercise_blt.c
> > new file mode 100644
> > index 0000000000..693e80d0b7
> > --- /dev/null
> > +++ b/tests/xe/xe_exercise_blt.c
> > @@ -0,0 +1,377 @@
> > +// SPDX-License-Identifier: MIT
> > +/*
> > + * Copyright © 2023 Intel Corporation
> > + */
> > +
> > +#include "igt.h"
> > +#include "drm.h"
> > +#include "lib/intel_chipset.h"
> > +#include "intel_blt.h"
> > +#include "intel_mocs.h"
> > +#include "xe/xe_ioctl.h"
> > +#include "xe/xe_query.h"
> > +#include "xe/xe_util.h"
> > +
> > +/**
> > + * TEST: xe exercise blt
> > + * Description: Exercise blitter commands on Xe
> > + * Feature: blitter
> > + * Run type: FULL
> > + * Test category: GEM_Legacy
> > + *
> > + * SUBTEST: fast-copy
> > + * Description:
> > + *   Check fast-copy blit
> > + *   blitter
> > + *
> > + * SUBTEST: fast-copy-emit
> > + * Description:
> > + *   Check multiple fast-copy in one batch
> > + *   blitter
> > + */
> > +
> > +IGT_TEST_DESCRIPTION("Exercise blitter commands on Xe");
> > +
> > +static struct param {
> > +	int tiling;
> > +	bool write_png;
> > +	bool print_bb;
> > +	bool print_surface_info;
> > +	int width;
> > +	int height;
> > +} param = {
> > +	.tiling = -1,
> > +	.write_png = false,
> > +	.print_bb = false,
> > +	.print_surface_info = false,
> > +	.width = 512,
> > +	.height = 512,
> > +};
> > +
> > +#define PRINT_SURFACE_INFO(name, obj) do { \
> > +	if (param.print_surface_info) \
> > +		blt_surface_info((name), (obj)); } while (0)
> > +
> > +#define WRITE_PNG(fd, id, name, obj, w, h) do { \
> > +	if (param.write_png) \
> > +		blt_surface_to_png((fd), (id), (name), (obj), (w), (h)); } while (0)
> > +
> > +struct blt_fast_copy_data {
> > +	int xe;
> > +	struct blt_copy_object src;
> > +	struct blt_copy_object mid;
> > +	struct blt_copy_object dst;
> > +
> > +	struct blt_copy_batch bb;
> > +	enum blt_color_depth color_depth;
> > +
> > +	/* debug stuff */
> > +	bool print_bb;
> > +};
> > +
> > +static int fast_copy_one_bb(int xe,
> > +			    const intel_ctx_t *ctx,
> > +			    uint64_t ahnd,
> > +			    const struct blt_fast_copy_data *blt)
> > +{
> > +	struct blt_copy_data blt_tmp;
> > +	uint64_t bb_offset, alignment;
> > +	uint64_t bb_pos = 0;
> > +	int ret = 0;
> > +
> > +	alignment = xe_get_default_alignment(xe);
> > +
> > +	get_offset(ahnd, blt->src.handle, blt->src.size, alignment);
> > +	get_offset(ahnd, blt->mid.handle, blt->mid.size, alignment);
> > +	get_offset(ahnd, blt->dst.handle, blt->dst.size, alignment);
> > +	bb_offset = get_offset(ahnd, blt->bb.handle, blt->bb.size, alignment);
> > +
> > +	/* First blit */
> > +	blt_copy_init(xe, &blt_tmp);
> > +	blt_tmp.src = blt->src;
> > +	blt_tmp.dst = blt->mid;
> > +	blt_tmp.bb = blt->bb;
> > +	blt_tmp.color_depth = blt->color_depth;
> > +	blt_tmp.print_bb = blt->print_bb;
> > +	bb_pos = emit_blt_fast_copy(xe, ahnd, &blt_tmp, bb_pos, false);
> > +
> > +	/* Second blit */
> > +	blt_copy_init(xe, &blt_tmp);
> > +	blt_tmp.src = blt->mid;
> > +	blt_tmp.dst = blt->dst;
> > +	blt_tmp.bb = blt->bb;
> > +	blt_tmp.color_depth = blt->color_depth;
> > +	blt_tmp.print_bb = blt->print_bb;
> > +	bb_pos = emit_blt_fast_copy(xe, ahnd, &blt_tmp, bb_pos, true);
> > +
> > +	intel_ctx_xe_exec(ctx, ahnd, bb_offset);
> > +
> > +	return ret;
> > +}
> > +
> > +static void fast_copy_emit(int xe, const intel_ctx_t *ctx,
> > +			   uint32_t region1, uint32_t region2,
> > +			   enum blt_tiling_type mid_tiling)
> > +{
> > +	struct blt_copy_data bltinit = {};
> > +	struct blt_fast_copy_data blt = {};
> > +	struct blt_copy_object *src, *mid, *dst;
> > +	const uint32_t bpp = 32;
> > +	uint64_t bb_size = xe_get_default_alignment(xe);
> > +	uint64_t ahnd = intel_allocator_open_full(xe, ctx->vm, 0, 0,
> > +						  INTEL_ALLOCATOR_SIMPLE,
> > +						  ALLOC_STRATEGY_LOW_TO_HIGH, 0);
> > +	uint32_t bb, width = param.width, height = param.height;
> > +	int result;
> > +
> > +	bb = xe_bo_create_flags(xe, 0, bb_size, region1);
> > +
> > +	blt_copy_init(xe, &bltinit);
> > +	src = blt_create_object(&bltinit, region1, width, height, bpp, 0,
> > +				T_LINEAR, COMPRESSION_DISABLED, 0, true);
> > +	mid = blt_create_object(&bltinit, region2, width, height, bpp, 0,
> > +				mid_tiling, COMPRESSION_DISABLED, 0, true);
> > +	dst = blt_create_object(&bltinit, region1, width, height, bpp, 0,
> > +				T_LINEAR, COMPRESSION_DISABLED, 0, true);
> > +	igt_assert(src->size == dst->size);
> > +
> > +	PRINT_SURFACE_INFO("src", src);
> > +	PRINT_SURFACE_INFO("mid", mid);
> > +	PRINT_SURFACE_INFO("dst", dst);
> > +
> > +	blt_surface_fill_rect(xe, src, width, height);
> > +	WRITE_PNG(xe, mid_tiling, "src", src, width, height);
> > +
> > +	memset(&blt, 0, sizeof(blt));
> 
> Could we use blt_copy_init() here?
> 

No :) That's you local structure incompatibile with blt_copy_data.

--
Zbigniew

> Thanks,
> Karolina
> 
> > +	blt.color_depth = CD_32bit;
> > +	blt.print_bb = param.print_bb;
> > +	blt_set_copy_object(&blt.src, src);
> > +	blt_set_copy_object(&blt.mid, mid);
> > +	blt_set_copy_object(&blt.dst, dst);
> > +	blt_set_batch(&blt.bb, bb, bb_size, region1);
> > +
> > +	fast_copy_one_bb(xe, ctx, ahnd, &blt);
> > +
> > +	WRITE_PNG(xe, mid_tiling, "mid", &blt.mid, width, height);
> > +	WRITE_PNG(xe, mid_tiling, "dst", &blt.dst, width, height);
> > +
> > +	result = memcmp(src->ptr, blt.dst.ptr, src->size);
> > +
> > +	blt_destroy_object(xe, src);
> > +	blt_destroy_object(xe, mid);
> > +	blt_destroy_object(xe, dst);
> > +	gem_close(xe, bb);
> > +	put_ahnd(ahnd);
> > +
> > +	munmap(&bb, bb_size);
> > +
> > +	igt_assert_f(!result, "source and destination surfaces differs!\n");
> > +}
> > +
> > +static void fast_copy(int xe, const intel_ctx_t *ctx,
> > +		      uint32_t region1, uint32_t region2,
> > +		      enum blt_tiling_type mid_tiling)
> > +{
> > +	struct blt_copy_data blt = {};
> > +	struct blt_copy_object *src, *mid, *dst;
> > +	const uint32_t bpp = 32;
> > +	uint64_t bb_size = xe_get_default_alignment(xe);
> > +	uint64_t ahnd = intel_allocator_open_full(xe, ctx->vm, 0, 0,
> > +						  INTEL_ALLOCATOR_SIMPLE,
> > +						  ALLOC_STRATEGY_LOW_TO_HIGH, 0);
> > +	uint32_t bb;
> > +	uint32_t width = param.width, height = param.height;
> > +	int result;
> > +
> > +	bb = xe_bo_create_flags(xe, 0, bb_size, region1);
> > +
> > +	blt_copy_init(xe, &blt);
> > +	src = blt_create_object(&blt, region1, width, height, bpp, 0,
> > +				T_LINEAR, COMPRESSION_DISABLED, 0, true);
> > +	mid = blt_create_object(&blt, region2, width, height, bpp, 0,
> > +				mid_tiling, COMPRESSION_DISABLED, 0, true);
> > +	dst = blt_create_object(&blt, region1, width, height, bpp, 0,
> > +				T_LINEAR, COMPRESSION_DISABLED, 0, true);
> > +	igt_assert(src->size == dst->size);
> > +
> > +	blt_surface_fill_rect(xe, src, width, height);
> > +
> > +	blt.color_depth = CD_32bit;
> > +	blt.print_bb = param.print_bb;
> > +	blt_set_copy_object(&blt.src, src);
> > +	blt_set_copy_object(&blt.dst, mid);
> > +	blt_set_batch(&blt.bb, bb, bb_size, region1);
> > +
> > +	blt_fast_copy(xe, ctx, NULL, ahnd, &blt);
> > +
> > +	WRITE_PNG(xe, mid_tiling, "src", &blt.src, width, height);
> > +	WRITE_PNG(xe, mid_tiling, "mid", &blt.dst, width, height);
> > +
> > +	blt_copy_init(xe, &blt);
> > +	blt.color_depth = CD_32bit;
> > +	blt.print_bb = param.print_bb;
> > +	blt_set_copy_object(&blt.src, mid);
> > +	blt_set_copy_object(&blt.dst, dst);
> > +	blt_set_batch(&blt.bb, bb, bb_size, region1);
> > +
> > +	blt_fast_copy(xe, ctx, NULL, ahnd, &blt);
> > +
> > +	WRITE_PNG(xe, mid_tiling, "dst", &blt.dst, width, height);
> > +
> > +	result = memcmp(src->ptr, blt.dst.ptr, src->size);
> > +
> > +	put_offset(ahnd, src->handle);
> > +	put_offset(ahnd, mid->handle);
> > +	put_offset(ahnd, dst->handle);
> > +	put_offset(ahnd, bb);
> > +	intel_allocator_bind(ahnd, 0, 0);
> > +	blt_destroy_object(xe, src);
> > +	blt_destroy_object(xe, mid);
> > +	blt_destroy_object(xe, dst);
> > +	gem_close(xe, bb);
> > +	put_ahnd(ahnd);
> > +
> > +	igt_assert_f(!result, "source and destination surfaces differs!\n");
> > +}
> > +
> > +enum fast_copy_func {
> > +	FAST_COPY,
> > +	FAST_COPY_EMIT
> > +};
> > +
> > +static char
> > +	*full_subtest_str(char *regtxt, enum blt_tiling_type tiling,
> > +			  enum fast_copy_func func)
> > +{
> > +	char *name;
> > +	uint32_t len;
> > +
> > +	len = asprintf(&name, "%s-%s%s", blt_tiling_name(tiling), regtxt,
> > +		       func == FAST_COPY_EMIT ? "-emit" : "");
> > +
> > +	igt_assert_f(len >= 0, "asprintf failed!\n");
> > +
> > +	return name;
> > +}
> > +
> > +static void fast_copy_test(int xe,
> > +			   struct igt_collection *set,
> > +			   enum fast_copy_func func)
> > +{
> > +	struct drm_xe_engine_class_instance inst = {
> > +		.engine_class = DRM_XE_ENGINE_CLASS_COPY,
> > +	};
> > +	struct igt_collection *regions;
> > +	void (*copy_func)(int xe, const intel_ctx_t *ctx,
> > +			  uint32_t r1, uint32_t r2, enum blt_tiling_type tiling);
> > +	intel_ctx_t *ctx;
> > +	int tiling;
> > +
> > +	for_each_tiling(tiling) {
> > +		if (!blt_fast_copy_supports_tiling(xe, tiling))
> > +			continue;
> > +
> > +		for_each_variation_r(regions, 2, set) {
> > +			uint32_t region1, region2;
> > +			uint32_t vm, engine;
> > +			char *regtxt, *test_name;
> > +
> > +			region1 = igt_collection_get_value(regions, 0);
> > +			region2 = igt_collection_get_value(regions, 1);
> > +
> > +			vm = xe_vm_create(xe, DRM_XE_VM_CREATE_ASYNC_BIND_OPS, 0);
> > +			engine = xe_engine_create(xe, vm, &inst, 0);
> > +			ctx = intel_ctx_xe(xe, vm, engine, 0, 0, 0);
> > +
> > +			copy_func = (func == FAST_COPY) ? fast_copy : fast_copy_emit;
> > +			regtxt = xe_memregion_dynamic_subtest_name(xe, regions);
> > +			test_name = full_subtest_str(regtxt, tiling, func);
> > +
> > +			igt_dynamic_f("%s", test_name) {
> > +				copy_func(xe, ctx,
> > +					  region1, region2,
> > +					  tiling);
> > +			}
> > +
> > +			free(regtxt);
> > +			free(test_name);
> > +			xe_engine_destroy(xe, engine);
> > +			xe_vm_destroy(xe, vm);
> > +			free(ctx);
> > +		}
> > +	}
> > +}
> > +
> > +static int opt_handler(int opt, int opt_index, void *data)
> > +{
> > +	switch (opt) {
> > +	case 'b':
> > +		param.print_bb = true;
> > +		igt_debug("Print bb: %d\n", param.print_bb);
> > +		break;
> > +	case 'p':
> > +		param.write_png = true;
> > +		igt_debug("Write png: %d\n", param.write_png);
> > +		break;
> > +	case 's':
> > +		param.print_surface_info = true;
> > +		igt_debug("Print surface info: %d\n", param.print_surface_info);
> > +		break;
> > +	case 't':
> > +		param.tiling = atoi(optarg);
> > +		igt_debug("Tiling: %d\n", param.tiling);
> > +		break;
> > +	case 'W':
> > +		param.width = atoi(optarg);
> > +		igt_debug("Width: %d\n", param.width);
> > +		break;
> > +	case 'H':
> > +		param.height = atoi(optarg);
> > +		igt_debug("Height: %d\n", param.height);
> > +		break;
> > +	default:
> > +		return IGT_OPT_HANDLER_ERROR;
> > +	}
> > +
> > +	return IGT_OPT_HANDLER_SUCCESS;
> > +}
> > +
> > +const char *help_str =
> > +	"  -b\tPrint bb\n"
> > +	"  -p\tWrite PNG\n"
> > +	"  -s\tPrint surface info\n"
> > +	"  -t\tTiling format (0 - linear, 1 - XMAJOR, 2 - YMAJOR, 3 - TILE4, 4 - TILE64, 5 - YFMAJOR)\n"
> > +	"  -W\tWidth (default 512)\n"
> > +	"  -H\tHeight (default 512)"
> > +	;
> > +
> > +igt_main_args("b:pst:W:H:", NULL, help_str, opt_handler, NULL)
> > +{
> > +	struct igt_collection *set;
> > +	int xe;
> > +
> > +	igt_fixture {
> > +		xe = drm_open_driver(DRIVER_XE);
> > +		igt_require(blt_has_fast_copy(xe));
> > +
> > +		xe_device_get(xe);
> > +
> > +		set = xe_get_memory_region_set(xe,
> > +					       XE_MEM_REGION_CLASS_SYSMEM,
> > +					       XE_MEM_REGION_CLASS_VRAM);
> > +	}
> > +
> > +	igt_describe("Check fast-copy blit");
> > +	igt_subtest_with_dynamic("fast-copy") {
> > +		fast_copy_test(xe, set, FAST_COPY);
> > +	}
> > +
> > +	igt_describe("Check multiple fast-copy in one batch");
> > +	igt_subtest_with_dynamic("fast-copy-emit") {
> > +		fast_copy_test(xe, set, FAST_COPY_EMIT);
> > +	}
> > +
> > +	igt_fixture {
> > +		drm_close_driver(xe);
> > +	}
> > +}

[igt-dev] ✓ Fi.CI.IGT: success for Extend intel_blt to work on Xe (rev3)

[Patchwork]

[-- Attachment #1: Type: text/plain, Size: 78314 bytes --]

== Series Details ==

Series: Extend intel_blt to work on Xe (rev3)
URL   : https://patchwork.freedesktop.org/series/120162/
State : success

== Summary ==

CI Bug Log - changes from CI_DRM_13369_full -> IGTPW_9383_full
====================================================

Summary
-------

  **SUCCESS**

  No regressions found.

  External URL: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/index.html

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

  No changes in participating hosts

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

  Here are the changes found in IGTPW_9383_full that come from known issues:

### IGT changes ###

#### Issues hit ####

  * igt@api_intel_bb@crc32:
    - shard-tglu:         NOTRUN -> [SKIP][1] ([i915#6230])
   [1]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-8/igt@api_intel_bb@crc32.html

  * igt@device_reset@unbind-cold-reset-rebind:
    - shard-rkl:          NOTRUN -> [SKIP][2] ([i915#7701])
   [2]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@device_reset@unbind-cold-reset-rebind.html
    - shard-dg2:          NOTRUN -> [SKIP][3] ([i915#7701])
   [3]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-6/igt@device_reset@unbind-cold-reset-rebind.html

  * igt@drm_fdinfo@busy-hang@bcs0:
    - shard-dg2:          NOTRUN -> [SKIP][4] ([i915#8414]) +19 similar issues
   [4]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-3/igt@drm_fdinfo@busy-hang@bcs0.html

  * igt@drm_fdinfo@most-busy-idle-check-all@rcs0:
    - shard-rkl:          [PASS][5] -> [FAIL][6] ([i915#7742])
   [5]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-rkl-2/igt@drm_fdinfo@most-busy-idle-check-all@rcs0.html
   [6]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-4/igt@drm_fdinfo@most-busy-idle-check-all@rcs0.html

  * igt@drm_fdinfo@virtual-busy-all:
    - shard-mtlp:         NOTRUN -> [SKIP][7] ([i915#8414])
   [7]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-6/igt@drm_fdinfo@virtual-busy-all.html

  * igt@gem_bad_reloc@negative-reloc-bltcopy:
    - shard-mtlp:         NOTRUN -> [SKIP][8] ([i915#3281]) +5 similar issues
   [8]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-3/igt@gem_bad_reloc@negative-reloc-bltcopy.html

  * igt@gem_basic@multigpu-create-close:
    - shard-rkl:          NOTRUN -> [SKIP][9] ([i915#7697]) +1 similar issue
   [9]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@gem_basic@multigpu-create-close.html

  * igt@gem_ccs@block-copy-compressed:
    - shard-tglu:         NOTRUN -> [SKIP][10] ([i915#3555] / [i915#5325])
   [10]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-2/igt@gem_ccs@block-copy-compressed.html

  * igt@gem_ccs@ctrl-surf-copy-new-ctx:
    - shard-mtlp:         NOTRUN -> [SKIP][11] ([i915#5325])
   [11]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-4/igt@gem_ccs@ctrl-surf-copy-new-ctx.html

  * igt@gem_close_race@multigpu-basic-process:
    - shard-dg2:          NOTRUN -> [SKIP][12] ([i915#7697]) +1 similar issue
   [12]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-7/igt@gem_close_race@multigpu-basic-process.html

  * igt@gem_create@create-ext-cpu-access-big:
    - shard-tglu:         NOTRUN -> [SKIP][13] ([i915#6335])
   [13]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-4/igt@gem_create@create-ext-cpu-access-big.html

  * igt@gem_ctx_exec@basic-nohangcheck:
    - shard-tglu:         [PASS][14] -> [FAIL][15] ([i915#6268])
   [14]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-tglu-3/igt@gem_ctx_exec@basic-nohangcheck.html
   [15]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-5/igt@gem_ctx_exec@basic-nohangcheck.html

  * igt@gem_ctx_param@set-priority-not-supported:
    - shard-dg2:          NOTRUN -> [SKIP][16] ([fdo#109314])
   [16]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-11/igt@gem_ctx_param@set-priority-not-supported.html
    - shard-rkl:          NOTRUN -> [SKIP][17] ([fdo#109314])
   [17]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@gem_ctx_param@set-priority-not-supported.html

  * igt@gem_ctx_persistence@saturated-hostile@vecs0:
    - shard-mtlp:         [PASS][18] -> [FAIL][19] ([i915#7816]) +2 similar issues
   [18]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-5/igt@gem_ctx_persistence@saturated-hostile@vecs0.html
   [19]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-2/igt@gem_ctx_persistence@saturated-hostile@vecs0.html

  * igt@gem_eio@in-flight-contexts-1us:
    - shard-mtlp:         [PASS][20] -> [ABORT][21] ([i915#8503])
   [20]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-3/igt@gem_eio@in-flight-contexts-1us.html
   [21]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-8/igt@gem_eio@in-flight-contexts-1us.html

  * igt@gem_eio@kms:
    - shard-dg2:          [PASS][22] -> [INCOMPLETE][23] ([i915#7892])
   [22]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg2-8/igt@gem_eio@kms.html
   [23]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-3/igt@gem_eio@kms.html

  * igt@gem_exec_balancer@bonded-true-hang:
    - shard-dg2:          NOTRUN -> [SKIP][24] ([i915#4812]) +2 similar issues
   [24]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-12/igt@gem_exec_balancer@bonded-true-hang.html

  * igt@gem_exec_fair@basic-none@bcs0:
    - shard-tglu:         NOTRUN -> [FAIL][25] ([i915#2842]) +4 similar issues
   [25]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-10/igt@gem_exec_fair@basic-none@bcs0.html

  * igt@gem_exec_fair@basic-pace@vcs0:
    - shard-glk:          [PASS][26] -> [FAIL][27] ([i915#2842]) +1 similar issue
   [26]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-glk7/igt@gem_exec_fair@basic-pace@vcs0.html
   [27]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-glk1/igt@gem_exec_fair@basic-pace@vcs0.html

  * igt@gem_exec_fair@basic-throttle:
    - shard-dg2:          NOTRUN -> [SKIP][28] ([i915#3539])
   [28]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-2/igt@gem_exec_fair@basic-throttle.html

  * igt@gem_exec_fence@submit67:
    - shard-mtlp:         NOTRUN -> [SKIP][29] ([i915#4812])
   [29]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-5/igt@gem_exec_fence@submit67.html

  * igt@gem_exec_params@secure-non-root:
    - shard-tglu:         NOTRUN -> [SKIP][30] ([fdo#112283])
   [30]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-4/igt@gem_exec_params@secure-non-root.html

  * igt@gem_exec_reloc@basic-write-gtt-active:
    - shard-dg2:          NOTRUN -> [SKIP][31] ([i915#3281]) +2 similar issues
   [31]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-10/igt@gem_exec_reloc@basic-write-gtt-active.html
    - shard-rkl:          NOTRUN -> [SKIP][32] ([i915#3281]) +1 similar issue
   [32]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-1/igt@gem_exec_reloc@basic-write-gtt-active.html

  * igt@gem_exec_whisper@basic-forked:
    - shard-mtlp:         NOTRUN -> [FAIL][33] ([i915#6363])
   [33]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-2/igt@gem_exec_whisper@basic-forked.html

  * igt@gem_exec_whisper@basic-forked-all:
    - shard-mtlp:         [PASS][34] -> [FAIL][35] ([i915#6363])
   [34]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-3/igt@gem_exec_whisper@basic-forked-all.html
   [35]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-3/igt@gem_exec_whisper@basic-forked-all.html

  * igt@gem_lmem_swapping@heavy-verify-random-ccs:
    - shard-tglu:         NOTRUN -> [SKIP][36] ([i915#4613]) +2 similar issues
   [36]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-10/igt@gem_lmem_swapping@heavy-verify-random-ccs.html

  * igt@gem_lmem_swapping@parallel-multi:
    - shard-apl:          NOTRUN -> [SKIP][37] ([fdo#109271] / [i915#4613])
   [37]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-apl1/igt@gem_lmem_swapping@parallel-multi.html

  * igt@gem_lmem_swapping@parallel-random:
    - shard-mtlp:         NOTRUN -> [SKIP][38] ([i915#4613])
   [38]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-3/igt@gem_lmem_swapping@parallel-random.html

  * igt@gem_lmem_swapping@smem-oom@lmem0:
    - shard-dg2:          [PASS][39] -> [TIMEOUT][40] ([i915#5493])
   [39]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg2-11/igt@gem_lmem_swapping@smem-oom@lmem0.html
   [40]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-10/igt@gem_lmem_swapping@smem-oom@lmem0.html

  * igt@gem_media_vme:
    - shard-tglu:         NOTRUN -> [SKIP][41] ([i915#284])
   [41]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-4/igt@gem_media_vme.html

  * igt@gem_mmap_wc@write:
    - shard-mtlp:         NOTRUN -> [SKIP][42] ([i915#4083]) +1 similar issue
   [42]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-6/igt@gem_mmap_wc@write.html
    - shard-dg2:          NOTRUN -> [SKIP][43] ([i915#4083]) +3 similar issues
   [43]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-5/igt@gem_mmap_wc@write.html

  * igt@gem_partial_pwrite_pread@reads:
    - shard-mtlp:         NOTRUN -> [SKIP][44] ([i915#3282]) +3 similar issues
   [44]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-5/igt@gem_partial_pwrite_pread@reads.html

  * igt@gem_partial_pwrite_pread@writes-after-reads:
    - shard-rkl:          NOTRUN -> [SKIP][45] ([i915#3282]) +2 similar issues
   [45]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-1/igt@gem_partial_pwrite_pread@writes-after-reads.html

  * igt@gem_pxp@create-regular-buffer:
    - shard-mtlp:         NOTRUN -> [SKIP][46] ([i915#4270])
   [46]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-6/igt@gem_pxp@create-regular-buffer.html

  * igt@gem_pxp@fail-invalid-protected-context:
    - shard-tglu:         NOTRUN -> [SKIP][47] ([i915#4270]) +1 similar issue
   [47]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-10/igt@gem_pxp@fail-invalid-protected-context.html

  * igt@gem_readwrite@beyond-eob:
    - shard-dg2:          NOTRUN -> [SKIP][48] ([i915#3282]) +3 similar issues
   [48]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-5/igt@gem_readwrite@beyond-eob.html

  * igt@gem_render_copy@y-tiled-ccs-to-yf-tiled-ccs:
    - shard-mtlp:         NOTRUN -> [SKIP][49] ([i915#8428]) +3 similar issues
   [49]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-4/igt@gem_render_copy@y-tiled-ccs-to-yf-tiled-ccs.html

  * igt@gem_set_tiling_vs_pwrite:
    - shard-dg2:          NOTRUN -> [SKIP][50] ([i915#4079])
   [50]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-7/igt@gem_set_tiling_vs_pwrite.html

  * igt@gem_tiled_fence_blits@basic:
    - shard-dg2:          NOTRUN -> [SKIP][51] ([i915#4077]) +3 similar issues
   [51]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-6/igt@gem_tiled_fence_blits@basic.html

  * igt@gem_tiled_pread_basic:
    - shard-mtlp:         NOTRUN -> [SKIP][52] ([i915#4079])
   [52]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-2/igt@gem_tiled_pread_basic.html

  * igt@gem_userptr_blits@coherency-unsync:
    - shard-mtlp:         NOTRUN -> [SKIP][53] ([i915#3297])
   [53]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-5/igt@gem_userptr_blits@coherency-unsync.html

  * igt@gem_userptr_blits@invalid-mmap-offset-unsync:
    - shard-dg2:          NOTRUN -> [SKIP][54] ([i915#3297])
   [54]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-3/igt@gem_userptr_blits@invalid-mmap-offset-unsync.html
    - shard-rkl:          NOTRUN -> [SKIP][55] ([i915#3297])
   [55]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@gem_userptr_blits@invalid-mmap-offset-unsync.html

  * igt@gem_userptr_blits@map-fixed-invalidate-overlap-busy:
    - shard-dg2:          NOTRUN -> [SKIP][56] ([i915#3297] / [i915#4880])
   [56]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-7/igt@gem_userptr_blits@map-fixed-invalidate-overlap-busy.html

  * igt@gem_userptr_blits@nohangcheck:
    - shard-mtlp:         [PASS][57] -> [FAIL][58] ([i915#7916])
   [57]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-7/igt@gem_userptr_blits@nohangcheck.html
   [58]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-8/igt@gem_userptr_blits@nohangcheck.html

  * igt@gen7_exec_parse@load-register-reg:
    - shard-dg2:          NOTRUN -> [SKIP][59] ([fdo#109289])
   [59]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-6/igt@gen7_exec_parse@load-register-reg.html

  * igt@gen7_exec_parse@oacontrol-tracking:
    - shard-mtlp:         NOTRUN -> [SKIP][60] ([fdo#109289])
   [60]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-8/igt@gen7_exec_parse@oacontrol-tracking.html

  * igt@gen9_exec_parse@bb-large:
    - shard-dg2:          NOTRUN -> [SKIP][61] ([i915#2856])
   [61]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-10/igt@gen9_exec_parse@bb-large.html
    - shard-rkl:          NOTRUN -> [SKIP][62] ([i915#2527])
   [62]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-1/igt@gen9_exec_parse@bb-large.html

  * igt@i915_hangman@gt-engine-error@vcs0:
    - shard-mtlp:         [PASS][63] -> [FAIL][64] ([i915#7069])
   [63]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-5/igt@i915_hangman@gt-engine-error@vcs0.html
   [64]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-4/igt@i915_hangman@gt-engine-error@vcs0.html

  * igt@i915_module_load@load:
    - shard-rkl:          NOTRUN -> [SKIP][65] ([i915#6227])
   [65]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@i915_module_load@load.html

  * igt@i915_pm_backlight@bad-brightness:
    - shard-dg2:          NOTRUN -> [SKIP][66] ([i915#5354] / [i915#7561])
   [66]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-1/igt@i915_pm_backlight@bad-brightness.html

  * igt@i915_pm_dc@dc5-psr:
    - shard-tglu:         NOTRUN -> [SKIP][67] ([i915#658])
   [67]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-3/igt@i915_pm_dc@dc5-psr.html

  * igt@i915_pm_dc@dc9-dpms:
    - shard-apl:          [PASS][68] -> [SKIP][69] ([fdo#109271])
   [68]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-apl3/igt@i915_pm_dc@dc9-dpms.html
   [69]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-apl7/igt@i915_pm_dc@dc9-dpms.html

  * igt@i915_pm_lpsp@screens-disabled:
    - shard-mtlp:         NOTRUN -> [SKIP][70] ([i915#8430])
   [70]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-3/igt@i915_pm_lpsp@screens-disabled.html

  * igt@i915_pm_rpm@modeset-non-lpsp-stress:
    - shard-dg2:          [PASS][71] -> [SKIP][72] ([i915#1397]) +1 similar issue
   [71]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg2-7/igt@i915_pm_rpm@modeset-non-lpsp-stress.html
   [72]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-12/igt@i915_pm_rpm@modeset-non-lpsp-stress.html

  * igt@i915_pm_rpm@pc8-residency:
    - shard-mtlp:         NOTRUN -> [SKIP][73] ([fdo#109293])
   [73]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-3/igt@i915_pm_rpm@pc8-residency.html

  * igt@i915_pm_rpm@system-suspend-execbuf:
    - shard-tglu:         NOTRUN -> [FAIL][74] ([i915#7940])
   [74]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-2/igt@i915_pm_rpm@system-suspend-execbuf.html

  * igt@i915_suspend@basic-s3-without-i915:
    - shard-dg2:          [PASS][75] -> [FAIL][76] ([fdo#103375] / [i915#6121]) +2 similar issues
   [75]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg2-2/igt@i915_suspend@basic-s3-without-i915.html
   [76]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-5/igt@i915_suspend@basic-s3-without-i915.html
    - shard-tglu:         NOTRUN -> [INCOMPLETE][77] ([i915#7443] / [i915#8102])
   [77]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-3/igt@i915_suspend@basic-s3-without-i915.html

  * igt@i915_suspend@debugfs-reader:
    - shard-apl:          [PASS][78] -> [ABORT][79] ([i915#180] / [i915#8213])
   [78]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-apl4/igt@i915_suspend@debugfs-reader.html
   [79]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-apl7/igt@i915_suspend@debugfs-reader.html

  * igt@i915_suspend@fence-restore-tiled2untiled:
    - shard-snb:          NOTRUN -> [DMESG-WARN][80] ([i915#8841]) +4 similar issues
   [80]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-snb4/igt@i915_suspend@fence-restore-tiled2untiled.html

  * igt@kms_addfb_basic@addfb25-x-tiled-legacy:
    - shard-mtlp:         NOTRUN -> [SKIP][81] ([i915#4212])
   [81]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-5/igt@kms_addfb_basic@addfb25-x-tiled-legacy.html

  * igt@kms_addfb_basic@basic-y-tiled-legacy:
    - shard-dg2:          NOTRUN -> [SKIP][82] ([i915#4215] / [i915#5190])
   [82]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-10/igt@kms_addfb_basic@basic-y-tiled-legacy.html

  * igt@kms_async_flips@crc@pipe-a-hdmi-a-3:
    - shard-dg2:          NOTRUN -> [FAIL][83] ([i915#8247]) +3 similar issues
   [83]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-5/igt@kms_async_flips@crc@pipe-a-hdmi-a-3.html

  * igt@kms_big_fb@4-tiled-16bpp-rotate-90:
    - shard-mtlp:         NOTRUN -> [SKIP][84] ([fdo#111614])
   [84]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-6/igt@kms_big_fb@4-tiled-16bpp-rotate-90.html

  * igt@kms_big_fb@4-tiled-8bpp-rotate-270:
    - shard-dg2:          NOTRUN -> [SKIP][85] ([fdo#111614]) +1 similar issue
   [85]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-6/igt@kms_big_fb@4-tiled-8bpp-rotate-270.html

  * igt@kms_big_fb@4-tiled-max-hw-stride-32bpp-rotate-180-async-flip:
    - shard-mtlp:         [PASS][86] -> [FAIL][87] ([i915#3743])
   [86]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-8/igt@kms_big_fb@4-tiled-max-hw-stride-32bpp-rotate-180-async-flip.html
   [87]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-1/igt@kms_big_fb@4-tiled-max-hw-stride-32bpp-rotate-180-async-flip.html

  * igt@kms_big_fb@4-tiled-max-hw-stride-32bpp-rotate-180-hflip:
    - shard-tglu:         NOTRUN -> [SKIP][88] ([fdo#111615] / [i915#5286])
   [88]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-9/igt@kms_big_fb@4-tiled-max-hw-stride-32bpp-rotate-180-hflip.html

  * igt@kms_big_fb@4-tiled-max-hw-stride-64bpp-rotate-0-hflip:
    - shard-rkl:          NOTRUN -> [SKIP][89] ([i915#5286]) +1 similar issue
   [89]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-2/igt@kms_big_fb@4-tiled-max-hw-stride-64bpp-rotate-0-hflip.html
    - shard-mtlp:         [PASS][90] -> [FAIL][91] ([i915#5138])
   [90]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-7/igt@kms_big_fb@4-tiled-max-hw-stride-64bpp-rotate-0-hflip.html
   [91]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-5/igt@kms_big_fb@4-tiled-max-hw-stride-64bpp-rotate-0-hflip.html

  * igt@kms_big_fb@x-tiled-32bpp-rotate-90:
    - shard-rkl:          NOTRUN -> [SKIP][92] ([fdo#111614] / [i915#3638])
   [92]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@kms_big_fb@x-tiled-32bpp-rotate-90.html

  * igt@kms_big_fb@x-tiled-max-hw-stride-64bpp-rotate-0-hflip-async-flip:
    - shard-mtlp:         NOTRUN -> [FAIL][93] ([i915#3743])
   [93]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-8/igt@kms_big_fb@x-tiled-max-hw-stride-64bpp-rotate-0-hflip-async-flip.html

  * igt@kms_big_fb@y-tiled-8bpp-rotate-90:
    - shard-tglu:         NOTRUN -> [SKIP][94] ([fdo#111614])
   [94]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-2/igt@kms_big_fb@y-tiled-8bpp-rotate-90.html

  * igt@kms_big_fb@yf-tiled-64bpp-rotate-90:
    - shard-rkl:          NOTRUN -> [SKIP][95] ([fdo#110723]) +2 similar issues
   [95]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-2/igt@kms_big_fb@yf-tiled-64bpp-rotate-90.html

  * igt@kms_big_fb@yf-tiled-addfb-size-overflow:
    - shard-dg2:          NOTRUN -> [SKIP][96] ([i915#5190]) +5 similar issues
   [96]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-7/igt@kms_big_fb@yf-tiled-addfb-size-overflow.html
    - shard-tglu:         NOTRUN -> [SKIP][97] ([fdo#111615]) +1 similar issue
   [97]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-10/igt@kms_big_fb@yf-tiled-addfb-size-overflow.html
    - shard-mtlp:         NOTRUN -> [SKIP][98] ([i915#6187])
   [98]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-8/igt@kms_big_fb@yf-tiled-addfb-size-overflow.html

  * igt@kms_big_fb@yf-tiled-max-hw-stride-32bpp-rotate-180-async-flip:
    - shard-mtlp:         NOTRUN -> [SKIP][99] ([fdo#111615]) +4 similar issues
   [99]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-4/igt@kms_big_fb@yf-tiled-max-hw-stride-32bpp-rotate-180-async-flip.html

  * igt@kms_big_fb@yf-tiled-max-hw-stride-64bpp-rotate-0:
    - shard-dg2:          NOTRUN -> [SKIP][100] ([i915#4538] / [i915#5190]) +3 similar issues
   [100]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-1/igt@kms_big_fb@yf-tiled-max-hw-stride-64bpp-rotate-0.html

  * igt@kms_ccs@pipe-a-bad-pixel-format-y_tiled_gen12_mc_ccs:
    - shard-mtlp:         NOTRUN -> [SKIP][101] ([i915#3886] / [i915#6095]) +5 similar issues
   [101]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-3/igt@kms_ccs@pipe-a-bad-pixel-format-y_tiled_gen12_mc_ccs.html

  * igt@kms_ccs@pipe-a-ccs-on-another-bo-4_tiled_mtl_rc_ccs_cc:
    - shard-rkl:          NOTRUN -> [SKIP][102] ([i915#5354] / [i915#6095]) +4 similar issues
   [102]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@kms_ccs@pipe-a-ccs-on-another-bo-4_tiled_mtl_rc_ccs_cc.html

  * igt@kms_ccs@pipe-a-crc-sprite-planes-basic-y_tiled_ccs:
    - shard-dg2:          NOTRUN -> [SKIP][103] ([i915#3689] / [i915#5354]) +12 similar issues
   [103]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-12/igt@kms_ccs@pipe-a-crc-sprite-planes-basic-y_tiled_ccs.html

  * igt@kms_ccs@pipe-a-missing-ccs-buffer-4_tiled_mtl_rc_ccs_cc:
    - shard-tglu:         NOTRUN -> [SKIP][104] ([i915#5354] / [i915#6095]) +8 similar issues
   [104]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-3/igt@kms_ccs@pipe-a-missing-ccs-buffer-4_tiled_mtl_rc_ccs_cc.html

  * igt@kms_ccs@pipe-b-bad-aux-stride-y_tiled_gen12_mc_ccs:
    - shard-rkl:          NOTRUN -> [SKIP][105] ([i915#3886] / [i915#5354] / [i915#6095])
   [105]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-2/igt@kms_ccs@pipe-b-bad-aux-stride-y_tiled_gen12_mc_ccs.html

  * igt@kms_ccs@pipe-b-ccs-on-another-bo-y_tiled_gen12_mc_ccs:
    - shard-tglu:         NOTRUN -> [SKIP][106] ([i915#3689] / [i915#3886] / [i915#5354] / [i915#6095]) +2 similar issues
   [106]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-7/igt@kms_ccs@pipe-b-ccs-on-another-bo-y_tiled_gen12_mc_ccs.html

  * igt@kms_ccs@pipe-b-crc-primary-basic-yf_tiled_ccs:
    - shard-rkl:          NOTRUN -> [SKIP][107] ([i915#3734] / [i915#5354] / [i915#6095]) +1 similar issue
   [107]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-2/igt@kms_ccs@pipe-b-crc-primary-basic-yf_tiled_ccs.html

  * igt@kms_ccs@pipe-c-bad-aux-stride-yf_tiled_ccs:
    - shard-tglu:         NOTRUN -> [SKIP][108] ([fdo#111615] / [i915#3689] / [i915#5354] / [i915#6095]) +1 similar issue
   [108]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-5/igt@kms_ccs@pipe-c-bad-aux-stride-yf_tiled_ccs.html

  * igt@kms_ccs@pipe-c-bad-rotation-90-4_tiled_mtl_mc_ccs:
    - shard-dg2:          NOTRUN -> [SKIP][109] ([i915#5354]) +18 similar issues
   [109]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-7/igt@kms_ccs@pipe-c-bad-rotation-90-4_tiled_mtl_mc_ccs.html
    - shard-rkl:          NOTRUN -> [SKIP][110] ([i915#5354]) +8 similar issues
   [110]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-1/igt@kms_ccs@pipe-c-bad-rotation-90-4_tiled_mtl_mc_ccs.html

  * igt@kms_ccs@pipe-c-bad-rotation-90-y_tiled_gen12_rc_ccs_cc:
    - shard-dg2:          NOTRUN -> [SKIP][111] ([i915#3689] / [i915#3886] / [i915#5354]) +3 similar issues
   [111]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-5/igt@kms_ccs@pipe-c-bad-rotation-90-y_tiled_gen12_rc_ccs_cc.html

  * igt@kms_ccs@pipe-d-bad-rotation-90-4_tiled_dg2_rc_ccs_cc:
    - shard-tglu:         NOTRUN -> [SKIP][112] ([i915#3689] / [i915#5354] / [i915#6095]) +9 similar issues
   [112]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-5/igt@kms_ccs@pipe-d-bad-rotation-90-4_tiled_dg2_rc_ccs_cc.html

  * igt@kms_ccs@pipe-d-crc-primary-rotation-180-yf_tiled_ccs:
    - shard-mtlp:         NOTRUN -> [SKIP][113] ([i915#6095]) +16 similar issues
   [113]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-5/igt@kms_ccs@pipe-d-crc-primary-rotation-180-yf_tiled_ccs.html

  * igt@kms_chamelium_audio@dp-audio:
    - shard-tglu:         NOTRUN -> [SKIP][114] ([i915#7828]) +3 similar issues
   [114]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-10/igt@kms_chamelium_audio@dp-audio.html
    - shard-mtlp:         NOTRUN -> [SKIP][115] ([i915#7828])
   [115]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-8/igt@kms_chamelium_audio@dp-audio.html

  * igt@kms_chamelium_color@ctm-0-50:
    - shard-tglu:         NOTRUN -> [SKIP][116] ([fdo#111827]) +1 similar issue
   [116]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-6/igt@kms_chamelium_color@ctm-0-50.html

  * igt@kms_chamelium_color@ctm-0-75:
    - shard-mtlp:         NOTRUN -> [SKIP][117] ([fdo#111827])
   [117]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-2/igt@kms_chamelium_color@ctm-0-75.html

  * igt@kms_chamelium_color@ctm-blue-to-red:
    - shard-dg2:          NOTRUN -> [SKIP][118] ([fdo#111827])
   [118]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-1/igt@kms_chamelium_color@ctm-blue-to-red.html

  * igt@kms_chamelium_edid@dp-edid-stress-resolution-non-4k:
    - shard-rkl:          NOTRUN -> [SKIP][119] ([i915#7828])
   [119]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@kms_chamelium_edid@dp-edid-stress-resolution-non-4k.html

  * igt@kms_chamelium_frames@hdmi-cmp-planar-formats:
    - shard-dg2:          NOTRUN -> [SKIP][120] ([i915#7828]) +1 similar issue
   [120]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-5/igt@kms_chamelium_frames@hdmi-cmp-planar-formats.html

  * igt@kms_content_protection@legacy:
    - shard-dg2:          NOTRUN -> [SKIP][121] ([i915#7118]) +2 similar issues
   [121]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-2/igt@kms_content_protection@legacy.html

  * igt@kms_content_protection@mei_interface:
    - shard-mtlp:         NOTRUN -> [SKIP][122] ([i915#8063])
   [122]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-6/igt@kms_content_protection@mei_interface.html

  * igt@kms_content_protection@uevent:
    - shard-rkl:          NOTRUN -> [SKIP][123] ([i915#7118])
   [123]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-2/igt@kms_content_protection@uevent.html

  * igt@kms_cursor_crc@cursor-onscreen-512x170:
    - shard-rkl:          NOTRUN -> [SKIP][124] ([fdo#109279] / [i915#3359])
   [124]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-1/igt@kms_cursor_crc@cursor-onscreen-512x170.html

  * igt@kms_cursor_crc@cursor-onscreen-512x512:
    - shard-tglu:         NOTRUN -> [SKIP][125] ([i915#3359]) +1 similar issue
   [125]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-10/igt@kms_cursor_crc@cursor-onscreen-512x512.html

  * igt@kms_cursor_crc@cursor-random-32x10:
    - shard-mtlp:         NOTRUN -> [SKIP][126] ([i915#8814])
   [126]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-1/igt@kms_cursor_crc@cursor-random-32x10.html

  * igt@kms_cursor_crc@cursor-random-512x512:
    - shard-mtlp:         NOTRUN -> [SKIP][127] ([i915#3359])
   [127]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-6/igt@kms_cursor_crc@cursor-random-512x512.html

  * igt@kms_cursor_crc@cursor-sliding-512x512:
    - shard-dg2:          NOTRUN -> [SKIP][128] ([i915#3359]) +2 similar issues
   [128]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-5/igt@kms_cursor_crc@cursor-sliding-512x512.html
    - shard-rkl:          NOTRUN -> [SKIP][129] ([i915#3359]) +1 similar issue
   [129]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@kms_cursor_crc@cursor-sliding-512x512.html

  * igt@kms_cursor_legacy@2x-long-flip-vs-cursor-legacy:
    - shard-mtlp:         NOTRUN -> [SKIP][130] ([i915#3546])
   [130]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-5/igt@kms_cursor_legacy@2x-long-flip-vs-cursor-legacy.html

  * igt@kms_cursor_legacy@cursora-vs-flipb-legacy:
    - shard-tglu:         NOTRUN -> [SKIP][131] ([fdo#109274]) +3 similar issues
   [131]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-9/igt@kms_cursor_legacy@cursora-vs-flipb-legacy.html

  * igt@kms_cursor_legacy@cursorb-vs-flipa-toggle:
    - shard-dg2:          NOTRUN -> [SKIP][132] ([fdo#109274] / [i915#5354]) +1 similar issue
   [132]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-3/igt@kms_cursor_legacy@cursorb-vs-flipa-toggle.html
    - shard-rkl:          NOTRUN -> [SKIP][133] ([fdo#111825]) +1 similar issue
   [133]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@kms_cursor_legacy@cursorb-vs-flipa-toggle.html

  * igt@kms_cursor_legacy@cursorb-vs-flipb-toggle:
    - shard-dg2:          NOTRUN -> [SKIP][134] ([fdo#109274] / [fdo#111767] / [i915#5354])
   [134]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-3/igt@kms_cursor_legacy@cursorb-vs-flipb-toggle.html
    - shard-rkl:          NOTRUN -> [SKIP][135] ([fdo#111767] / [fdo#111825])
   [135]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-4/igt@kms_cursor_legacy@cursorb-vs-flipb-toggle.html

  * igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size:
    - shard-apl:          [PASS][136] -> [FAIL][137] ([i915#2346])
   [136]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-apl6/igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size.html
   [137]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-apl3/igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size.html
    - shard-glk:          [PASS][138] -> [FAIL][139] ([i915#2346])
   [138]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-glk8/igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size.html
   [139]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-glk3/igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size.html

  * igt@kms_dither@fb-8bpc-vs-panel-6bpc@pipe-a-hdmi-a-1:
    - shard-rkl:          NOTRUN -> [SKIP][140] ([i915#3804])
   [140]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@kms_dither@fb-8bpc-vs-panel-6bpc@pipe-a-hdmi-a-1.html

  * igt@kms_dp_aux_dev:
    - shard-tglu:         NOTRUN -> [SKIP][141] ([i915#1257])
   [141]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-2/igt@kms_dp_aux_dev.html

  * igt@kms_dsc@dsc-with-bpc-formats:
    - shard-tglu:         NOTRUN -> [SKIP][142] ([i915#3555] / [i915#3840])
   [142]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-9/igt@kms_dsc@dsc-with-bpc-formats.html

  * igt@kms_dsc@dsc-with-output-formats:
    - shard-mtlp:         NOTRUN -> [SKIP][143] ([i915#3840])
   [143]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-1/igt@kms_dsc@dsc-with-output-formats.html

  * igt@kms_fbcon_fbt@psr:
    - shard-dg2:          NOTRUN -> [SKIP][144] ([i915#3469])
   [144]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-7/igt@kms_fbcon_fbt@psr.html
    - shard-rkl:          NOTRUN -> [SKIP][145] ([fdo#110189] / [i915#3955])
   [145]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-1/igt@kms_fbcon_fbt@psr.html

  * igt@kms_flip@2x-blocking-absolute-wf_vblank:
    - shard-tglu:         NOTRUN -> [SKIP][146] ([fdo#109274] / [i915#3637]) +2 similar issues
   [146]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-4/igt@kms_flip@2x-blocking-absolute-wf_vblank.html

  * igt@kms_flip@2x-flip-vs-blocking-wf-vblank:
    - shard-snb:          NOTRUN -> [SKIP][147] ([fdo#109271] / [fdo#111767])
   [147]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-snb5/igt@kms_flip@2x-flip-vs-blocking-wf-vblank.html

  * igt@kms_flip@2x-flip-vs-fences-interruptible:
    - shard-mtlp:         NOTRUN -> [SKIP][148] ([i915#8381])
   [148]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-3/igt@kms_flip@2x-flip-vs-fences-interruptible.html

  * igt@kms_flip@2x-plain-flip-fb-recreate:
    - shard-apl:          NOTRUN -> [SKIP][149] ([fdo#109271]) +24 similar issues
   [149]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-apl6/igt@kms_flip@2x-plain-flip-fb-recreate.html

  * igt@kms_flip@2x-plain-flip-interruptible:
    - shard-mtlp:         NOTRUN -> [SKIP][150] ([i915#3637])
   [150]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-1/igt@kms_flip@2x-plain-flip-interruptible.html

  * igt@kms_flip@2x-wf_vblank-ts-check-interruptible:
    - shard-dg2:          NOTRUN -> [SKIP][151] ([fdo#109274])
   [151]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-8/igt@kms_flip@2x-wf_vblank-ts-check-interruptible.html

  * igt@kms_flip_scaled_crc@flip-32bpp-4tile-to-32bpp-4tiledg2rcccs-downscaling@pipe-a-valid-mode:
    - shard-rkl:          NOTRUN -> [SKIP][152] ([i915#2672])
   [152]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-2/igt@kms_flip_scaled_crc@flip-32bpp-4tile-to-32bpp-4tiledg2rcccs-downscaling@pipe-a-valid-mode.html

  * igt@kms_flip_scaled_crc@flip-32bpp-ytile-to-32bpp-ytilegen12rcccs-upscaling@pipe-a-default-mode:
    - shard-mtlp:         NOTRUN -> [SKIP][153] ([i915#2672]) +1 similar issue
   [153]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-6/igt@kms_flip_scaled_crc@flip-32bpp-ytile-to-32bpp-ytilegen12rcccs-upscaling@pipe-a-default-mode.html

  * igt@kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-downscaling@pipe-a-default-mode:
    - shard-mtlp:         NOTRUN -> [SKIP][154] ([i915#8810])
   [154]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-2/igt@kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-downscaling@pipe-a-default-mode.html

  * igt@kms_force_connector_basic@prune-stale-modes:
    - shard-mtlp:         NOTRUN -> [SKIP][155] ([i915#5274])
   [155]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-3/igt@kms_force_connector_basic@prune-stale-modes.html
    - shard-dg2:          NOTRUN -> [SKIP][156] ([i915#5274])
   [156]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-5/igt@kms_force_connector_basic@prune-stale-modes.html

  * igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-draw-mmap-gtt:
    - shard-mtlp:         NOTRUN -> [SKIP][157] ([i915#8708]) +1 similar issue
   [157]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-3/igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-draw-mmap-gtt.html

  * igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-fullscreen:
    - shard-dg2:          [PASS][158] -> [FAIL][159] ([i915#6880])
   [158]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg2-8/igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-fullscreen.html
   [159]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-11/igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-fullscreen.html

  * igt@kms_frontbuffer_tracking@fbcpsr-1p-primscrn-pri-shrfb-draw-mmap-gtt:
    - shard-rkl:          NOTRUN -> [SKIP][160] ([i915#3023]) +4 similar issues
   [160]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-4/igt@kms_frontbuffer_tracking@fbcpsr-1p-primscrn-pri-shrfb-draw-mmap-gtt.html

  * igt@kms_frontbuffer_tracking@fbcpsr-2p-primscrn-pri-shrfb-draw-mmap-wc:
    - shard-dg2:          NOTRUN -> [SKIP][161] ([i915#8708]) +5 similar issues
   [161]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-5/igt@kms_frontbuffer_tracking@fbcpsr-2p-primscrn-pri-shrfb-draw-mmap-wc.html
    - shard-rkl:          NOTRUN -> [SKIP][162] ([fdo#111825] / [i915#1825]) +8 similar issues
   [162]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-4/igt@kms_frontbuffer_tracking@fbcpsr-2p-primscrn-pri-shrfb-draw-mmap-wc.html

  * igt@kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-pri-shrfb-draw-pwrite:
    - shard-tglu:         NOTRUN -> [SKIP][163] ([fdo#109280]) +14 similar issues
   [163]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-4/igt@kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-pri-shrfb-draw-pwrite.html

  * igt@kms_frontbuffer_tracking@fbcpsr-tiling-4:
    - shard-tglu:         NOTRUN -> [SKIP][164] ([i915#5439])
   [164]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-6/igt@kms_frontbuffer_tracking@fbcpsr-tiling-4.html

  * igt@kms_frontbuffer_tracking@psr-1p-offscren-pri-indfb-draw-mmap-cpu:
    - shard-tglu:         NOTRUN -> [SKIP][165] ([fdo#110189]) +12 similar issues
   [165]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-10/igt@kms_frontbuffer_tracking@psr-1p-offscren-pri-indfb-draw-mmap-cpu.html

  * igt@kms_frontbuffer_tracking@psr-1p-offscren-pri-shrfb-draw-render:
    - shard-dg2:          NOTRUN -> [SKIP][166] ([i915#3458]) +3 similar issues
   [166]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-1/igt@kms_frontbuffer_tracking@psr-1p-offscren-pri-shrfb-draw-render.html

  * igt@kms_frontbuffer_tracking@psr-2p-scndscrn-pri-shrfb-draw-mmap-cpu:
    - shard-mtlp:         NOTRUN -> [SKIP][167] ([i915#1825]) +10 similar issues
   [167]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-1/igt@kms_frontbuffer_tracking@psr-2p-scndscrn-pri-shrfb-draw-mmap-cpu.html

  * igt@kms_hdr@bpc-switch:
    - shard-tglu:         NOTRUN -> [SKIP][168] ([i915#3555] / [i915#8228])
   [168]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-7/igt@kms_hdr@bpc-switch.html

  * igt@kms_hdr@bpc-switch-suspend:
    - shard-rkl:          NOTRUN -> [SKIP][169] ([i915#3555] / [i915#8228])
   [169]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@kms_hdr@bpc-switch-suspend.html

  * igt@kms_hdr@static-toggle:
    - shard-dg2:          NOTRUN -> [SKIP][170] ([i915#3555] / [i915#8228]) +1 similar issue
   [170]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-5/igt@kms_hdr@static-toggle.html

  * igt@kms_pipe_b_c_ivb@pipe-b-dpms-off-modeset-pipe-c:
    - shard-tglu:         NOTRUN -> [SKIP][171] ([fdo#109289]) +1 similar issue
   [171]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-7/igt@kms_pipe_b_c_ivb@pipe-b-dpms-off-modeset-pipe-c.html

  * igt@kms_plane_scaling@intel-max-src-size:
    - shard-dg2:          NOTRUN -> [SKIP][172] ([i915#6953])
   [172]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-6/igt@kms_plane_scaling@intel-max-src-size.html

  * igt@kms_plane_scaling@plane-scaler-with-clipping-clamping-rotation@pipe-a-vga-1:
    - shard-snb:          NOTRUN -> [SKIP][173] ([fdo#109271]) +101 similar issues
   [173]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-snb2/igt@kms_plane_scaling@plane-scaler-with-clipping-clamping-rotation@pipe-a-vga-1.html

  * igt@kms_plane_scaling@plane-upscale-with-rotation-factor-0-25@pipe-a-hdmi-a-1:
    - shard-tglu:         NOTRUN -> [SKIP][174] ([i915#5176]) +15 similar issues
   [174]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-10/igt@kms_plane_scaling@plane-upscale-with-rotation-factor-0-25@pipe-a-hdmi-a-1.html

  * igt@kms_plane_scaling@plane-upscale-with-rotation-factor-0-25@pipe-a-hdmi-a-2:
    - shard-rkl:          NOTRUN -> [SKIP][175] ([i915#5176]) +7 similar issues
   [175]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-1/igt@kms_plane_scaling@plane-upscale-with-rotation-factor-0-25@pipe-a-hdmi-a-2.html

  * igt@kms_plane_scaling@planes-downscale-factor-0-25@pipe-d-hdmi-a-3:
    - shard-dg2:          NOTRUN -> [SKIP][176] ([i915#5235]) +15 similar issues
   [176]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-6/igt@kms_plane_scaling@planes-downscale-factor-0-25@pipe-d-hdmi-a-3.html

  * igt@kms_plane_scaling@planes-upscale-20x20-downscale-factor-0-25@pipe-b-edp-1:
    - shard-mtlp:         NOTRUN -> [SKIP][177] ([i915#5235]) +3 similar issues
   [177]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-7/igt@kms_plane_scaling@planes-upscale-20x20-downscale-factor-0-25@pipe-b-edp-1.html

  * igt@kms_plane_scaling@planes-upscale-factor-0-25-downscale-factor-0-25@pipe-b-hdmi-a-2:
    - shard-rkl:          NOTRUN -> [SKIP][178] ([i915#5235]) +3 similar issues
   [178]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-1/igt@kms_plane_scaling@planes-upscale-factor-0-25-downscale-factor-0-25@pipe-b-hdmi-a-2.html

  * igt@kms_psr2_sf@cursor-plane-move-continuous-sf:
    - shard-apl:          NOTRUN -> [SKIP][179] ([fdo#109271] / [i915#658])
   [179]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-apl6/igt@kms_psr2_sf@cursor-plane-move-continuous-sf.html

  * igt@kms_psr2_su@frontbuffer-xrgb8888:
    - shard-dg2:          NOTRUN -> [SKIP][180] ([i915#658]) +1 similar issue
   [180]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-5/igt@kms_psr2_su@frontbuffer-xrgb8888.html
    - shard-mtlp:         NOTRUN -> [SKIP][181] ([i915#4348])
   [181]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-1/igt@kms_psr2_su@frontbuffer-xrgb8888.html

  * igt@kms_psr2_su@page_flip-xrgb8888:
    - shard-tglu:         NOTRUN -> [SKIP][182] ([fdo#109642] / [fdo#111068] / [i915#658]) +1 similar issue
   [182]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-10/igt@kms_psr2_su@page_flip-xrgb8888.html

  * igt@kms_psr@primary_mmap_cpu:
    - shard-dg2:          NOTRUN -> [SKIP][183] ([i915#1072]) +4 similar issues
   [183]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-3/igt@kms_psr@primary_mmap_cpu.html
    - shard-rkl:          NOTRUN -> [SKIP][184] ([i915#1072]) +2 similar issues
   [184]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@kms_psr@primary_mmap_cpu.html

  * igt@kms_psr@psr2_cursor_blt:
    - shard-mtlp:         [PASS][185] -> [FAIL][186] ([i915#8726])
   [185]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-8/igt@kms_psr@psr2_cursor_blt.html
   [186]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-7/igt@kms_psr@psr2_cursor_blt.html

  * igt@kms_psr@psr2_cursor_mmap_gtt:
    - shard-mtlp:         NOTRUN -> [SKIP][187] ([i915#4077]) +2 similar issues
   [187]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-5/igt@kms_psr@psr2_cursor_mmap_gtt.html

  * igt@kms_rmfb@close-fd@pipe-a-edp-1:
    - shard-mtlp:         [PASS][188] -> [DMESG-WARN][189] ([i915#1982])
   [188]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-1/igt@kms_rmfb@close-fd@pipe-a-edp-1.html
   [189]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-6/igt@kms_rmfb@close-fd@pipe-a-edp-1.html

  * igt@kms_rotation_crc@primary-y-tiled-reflect-x-270:
    - shard-mtlp:         NOTRUN -> [SKIP][190] ([i915#4235])
   [190]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-5/igt@kms_rotation_crc@primary-y-tiled-reflect-x-270.html

  * igt@kms_scaling_modes@scaling-mode-full:
    - shard-tglu:         NOTRUN -> [SKIP][191] ([i915#3555]) +2 similar issues
   [191]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-4/igt@kms_scaling_modes@scaling-mode-full.html

  * igt@kms_tiled_display@basic-test-pattern:
    - shard-tglu:         NOTRUN -> [SKIP][192] ([i915#8623])
   [192]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-4/igt@kms_tiled_display@basic-test-pattern.html

  * igt@kms_tiled_display@basic-test-pattern-with-chamelium:
    - shard-mtlp:         NOTRUN -> [SKIP][193] ([i915#8623])
   [193]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-1/igt@kms_tiled_display@basic-test-pattern-with-chamelium.html

  * igt@kms_vblank@pipe-c-query-forked-busy:
    - shard-rkl:          NOTRUN -> [SKIP][194] ([i915#4070] / [i915#6768])
   [194]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-2/igt@kms_vblank@pipe-c-query-forked-busy.html

  * igt@kms_vblank@pipe-d-wait-idle:
    - shard-rkl:          NOTRUN -> [SKIP][195] ([i915#4070] / [i915#533] / [i915#6768])
   [195]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@kms_vblank@pipe-d-wait-idle.html

  * igt@kms_writeback@writeback-pixel-formats:
    - shard-tglu:         NOTRUN -> [SKIP][196] ([i915#2437])
   [196]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-6/igt@kms_writeback@writeback-pixel-formats.html

  * igt@perf@oa-exponents@0-rcs0:
    - shard-glk:          [PASS][197] -> [ABORT][198] ([i915#5213] / [i915#7941])
   [197]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-glk8/igt@perf@oa-exponents@0-rcs0.html
   [198]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-glk6/igt@perf@oa-exponents@0-rcs0.html

  * igt@perf_pmu@event-wait@rcs0:
    - shard-mtlp:         NOTRUN -> [SKIP][199] ([i915#8807])
   [199]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-1/igt@perf_pmu@event-wait@rcs0.html

  * igt@perf_pmu@rc6@other-idle-gt0:
    - shard-dg2:          NOTRUN -> [SKIP][200] ([i915#8516])
   [200]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-8/igt@perf_pmu@rc6@other-idle-gt0.html
    - shard-tglu:         NOTRUN -> [SKIP][201] ([i915#8516])
   [201]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-2/igt@perf_pmu@rc6@other-idle-gt0.html

  * igt@perf_pmu@render-node-busy-idle@ccs0:
    - shard-mtlp:         [PASS][202] -> [FAIL][203] ([i915#4349]) +7 similar issues
   [202]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-7/igt@perf_pmu@render-node-busy-idle@ccs0.html
   [203]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-2/igt@perf_pmu@render-node-busy-idle@ccs0.html

  * igt@prime_vgem@basic-gtt:
    - shard-mtlp:         NOTRUN -> [SKIP][204] ([i915#3708] / [i915#4077])
   [204]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-3/igt@prime_vgem@basic-gtt.html

  * igt@prime_vgem@fence-wait@ccs0:
    - shard-mtlp:         [PASS][205] -> [DMESG-WARN][206] ([i915#8875])
   [205]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-3/igt@prime_vgem@fence-wait@ccs0.html
   [206]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-4/igt@prime_vgem@fence-wait@ccs0.html

  * igt@prime_vgem@fence-wait@vecs0:
    - shard-mtlp:         [PASS][207] -> [ABORT][208] ([i915#8875])
   [207]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-3/igt@prime_vgem@fence-wait@vecs0.html
   [208]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-4/igt@prime_vgem@fence-wait@vecs0.html

  * igt@sysfs_heartbeat_interval@nopreempt@bcs0:
    - shard-mtlp:         [PASS][209] -> [FAIL][210] ([i915#6015]) +1 similar issue
   [209]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-6/igt@sysfs_heartbeat_interval@nopreempt@bcs0.html
   [210]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-6/igt@sysfs_heartbeat_interval@nopreempt@bcs0.html

  * igt@sysfs_preempt_timeout@timeout@vecs0:
    - shard-mtlp:         [PASS][211] -> [ABORT][212] ([i915#8521])
   [211]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-5/igt@sysfs_preempt_timeout@timeout@vecs0.html
   [212]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-3/igt@sysfs_preempt_timeout@timeout@vecs0.html

  * igt@v3d/v3d_perfmon@create-perfmon-exceed:
    - shard-mtlp:         NOTRUN -> [SKIP][213] ([i915#2575]) +5 similar issues
   [213]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-2/igt@v3d/v3d_perfmon@create-perfmon-exceed.html

  * igt@v3d/v3d_submit_cl@valid-submission:
    - shard-tglu:         NOTRUN -> [SKIP][214] ([fdo#109315] / [i915#2575]) +4 similar issues
   [214]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-2/igt@v3d/v3d_submit_cl@valid-submission.html

  * igt@v3d/v3d_submit_csd@bad-multisync-in-sync:
    - shard-dg2:          NOTRUN -> [SKIP][215] ([i915#2575]) +2 similar issues
   [215]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-7/igt@v3d/v3d_submit_csd@bad-multisync-in-sync.html
    - shard-rkl:          NOTRUN -> [SKIP][216] ([fdo#109315]) +2 similar issues
   [216]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-2/igt@v3d/v3d_submit_csd@bad-multisync-in-sync.html

  * igt@vc4/vc4_mmap@mmap-bo:
    - shard-dg2:          NOTRUN -> [SKIP][217] ([i915#7711]) +3 similar issues
   [217]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-1/igt@vc4/vc4_mmap@mmap-bo.html
    - shard-rkl:          NOTRUN -> [SKIP][218] ([i915#7711]) +2 similar issues
   [218]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-2/igt@vc4/vc4_mmap@mmap-bo.html

  * igt@vc4/vc4_perfmon@create-perfmon-exceed:
    - shard-mtlp:         NOTRUN -> [SKIP][219] ([i915#7711]) +3 similar issues
   [219]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-4/igt@vc4/vc4_perfmon@create-perfmon-exceed.html

  * igt@vc4/vc4_purgeable_bo@mark-unpurgeable-purged:
    - shard-tglu:         NOTRUN -> [SKIP][220] ([i915#2575]) +4 similar issues
   [220]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-8/igt@vc4/vc4_purgeable_bo@mark-unpurgeable-purged.html

  * igt@vgem_basic@bad-pad:
    - shard-mtlp:         [PASS][221] -> [DMESG-WARN][222] ([i915#2017])
   [221]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-3/igt@vgem_basic@bad-pad.html
   [222]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-6/igt@vgem_basic@bad-pad.html

  * igt@vgem_basic@sysfs:
    - shard-snb:          [PASS][223] -> [ABORT][224] ([i915#8878])
   [223]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-snb7/igt@vgem_basic@sysfs.html
   [224]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-snb2/igt@vgem_basic@sysfs.html

  
#### Possible fixes ####

  * igt@gem_create@hog-create@smem0:
    - shard-dg2:          [FAIL][225] ([i915#5892] / [i915#8758]) -> [PASS][226]
   [225]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg2-11/igt@gem_create@hog-create@smem0.html
   [226]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-1/igt@gem_create@hog-create@smem0.html

  * igt@gem_eio@hibernate:
    - {shard-dg1}:        [ABORT][227] ([i915#4391] / [i915#7975] / [i915#8213]) -> [PASS][228]
   [227]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg1-14/igt@gem_eio@hibernate.html
   [228]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg1-16/igt@gem_eio@hibernate.html

  * igt@gem_exec_fair@basic-pace-share@rcs0:
    - shard-glk:          [FAIL][229] ([i915#2842]) -> [PASS][230]
   [229]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-glk8/igt@gem_exec_fair@basic-pace-share@rcs0.html
   [230]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-glk1/igt@gem_exec_fair@basic-pace-share@rcs0.html

  * igt@gem_exec_fair@basic-throttle@rcs0:
    - shard-rkl:          [FAIL][231] ([i915#2842]) -> [PASS][232] +3 similar issues
   [231]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-rkl-4/igt@gem_exec_fair@basic-throttle@rcs0.html
   [232]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-2/igt@gem_exec_fair@basic-throttle@rcs0.html

  * igt@gem_exec_whisper@basic-contexts-forked-all:
    - shard-mtlp:         [ABORT][233] ([i915#8131]) -> [PASS][234]
   [233]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-1/igt@gem_exec_whisper@basic-contexts-forked-all.html
   [234]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-5/igt@gem_exec_whisper@basic-contexts-forked-all.html

  * igt@gem_exec_whisper@basic-fds-priority-all:
    - shard-mtlp:         [FAIL][235] ([i915#6363]) -> [PASS][236]
   [235]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-5/igt@gem_exec_whisper@basic-fds-priority-all.html
   [236]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-1/igt@gem_exec_whisper@basic-fds-priority-all.html

  * igt@gem_lmem_swapping@smem-oom@lmem0:
    - {shard-dg1}:        [DMESG-WARN][237] ([i915#4936] / [i915#5493]) -> [PASS][238]
   [237]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg1-13/igt@gem_lmem_swapping@smem-oom@lmem0.html
   [238]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg1-17/igt@gem_lmem_swapping@smem-oom@lmem0.html

  * igt@i915_hangman@engine-engine-hang@vcs0:
    - shard-mtlp:         [FAIL][239] ([i915#7069]) -> [PASS][240]
   [239]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-8/igt@i915_hangman@engine-engine-hang@vcs0.html
   [240]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-5/igt@i915_hangman@engine-engine-hang@vcs0.html

  * igt@i915_module_load@reload-with-fault-injection:
    - shard-mtlp:         [ABORT][241] ([i915#8489] / [i915#8668]) -> [PASS][242]
   [241]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-4/igt@i915_module_load@reload-with-fault-injection.html
   [242]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-3/igt@i915_module_load@reload-with-fault-injection.html

  * igt@i915_pm_dc@dc9-dpms:
    - shard-tglu:         [SKIP][243] ([i915#4281]) -> [PASS][244]
   [243]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-tglu-6/igt@i915_pm_dc@dc9-dpms.html
   [244]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-4/igt@i915_pm_dc@dc9-dpms.html

  * igt@i915_pm_rc6_residency@rc6-idle@rcs0:
    - {shard-dg1}:        [FAIL][245] ([i915#3591]) -> [PASS][246] +1 similar issue
   [245]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg1-19/igt@i915_pm_rc6_residency@rc6-idle@rcs0.html
   [246]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg1-18/igt@i915_pm_rc6_residency@rc6-idle@rcs0.html

  * igt@i915_pm_rpm@cursor-dpms:
    - shard-tglu:         [FAIL][247] ([i915#7940]) -> [PASS][248] +1 similar issue
   [247]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-tglu-4/igt@i915_pm_rpm@cursor-dpms.html
   [248]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-10/igt@i915_pm_rpm@cursor-dpms.html

  * igt@i915_pm_rpm@dpms-mode-unset-lpsp:
    - {shard-dg1}:        [SKIP][249] ([i915#1397]) -> [PASS][250]
   [249]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg1-15/igt@i915_pm_rpm@dpms-mode-unset-lpsp.html
   [250]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg1-19/igt@i915_pm_rpm@dpms-mode-unset-lpsp.html

  * igt@i915_pm_rpm@dpms-non-lpsp:
    - shard-rkl:          [SKIP][251] ([i915#1397]) -> [PASS][252] +1 similar issue
   [251]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-rkl-7/igt@i915_pm_rpm@dpms-non-lpsp.html
   [252]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-2/igt@i915_pm_rpm@dpms-non-lpsp.html

  * igt@i915_pm_rpm@modeset-non-lpsp-stress-no-wait:
    - shard-dg2:          [SKIP][253] ([i915#1397]) -> [PASS][254] +2 similar issues
   [253]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg2-12/igt@i915_pm_rpm@modeset-non-lpsp-stress-no-wait.html
   [254]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-7/igt@i915_pm_rpm@modeset-non-lpsp-stress-no-wait.html

  * igt@i915_selftest@live@gt_mocs:
    - shard-mtlp:         [DMESG-FAIL][255] ([i915#7059]) -> [PASS][256]
   [255]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-8/igt@i915_selftest@live@gt_mocs.html
   [256]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-7/igt@i915_selftest@live@gt_mocs.html

  * igt@i915_selftest@live@hangcheck:
    - shard-dg2:          [ABORT][257] ([i915#7913]) -> [PASS][258]
   [257]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg2-7/igt@i915_selftest@live@hangcheck.html
   [258]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-12/igt@i915_selftest@live@hangcheck.html

  * igt@i915_selftest@live@workarounds:
    - shard-mtlp:         [DMESG-FAIL][259] ([i915#6763]) -> [PASS][260]
   [259]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-8/igt@i915_selftest@live@workarounds.html
   [260]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-7/igt@i915_selftest@live@workarounds.html

  * igt@kms_big_fb@x-tiled-max-hw-stride-64bpp-rotate-180-hflip-async-flip:
    - shard-mtlp:         [FAIL][261] ([i915#3743]) -> [PASS][262]
   [261]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-1/igt@kms_big_fb@x-tiled-max-hw-stride-64bpp-rotate-180-hflip-async-flip.html
   [262]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-2/igt@kms_big_fb@x-tiled-max-hw-stride-64bpp-rotate-180-hflip-async-flip.html

  * igt@kms_cursor_legacy@2x-long-flip-vs-cursor-atomic:
    - shard-glk:          [FAIL][263] ([i915#72]) -> [PASS][264]
   [263]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-glk8/igt@kms_cursor_legacy@2x-long-flip-vs-cursor-atomic.html
   [264]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-glk2/igt@kms_cursor_legacy@2x-long-flip-vs-cursor-atomic.html

  * igt@kms_cursor_legacy@cursor-vs-flip-toggle:
    - shard-mtlp:         [FAIL][265] ([i915#8248]) -> [PASS][266]
   [265]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-7/igt@kms_cursor_legacy@cursor-vs-flip-toggle.html
   [266]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-4/igt@kms_cursor_legacy@cursor-vs-flip-toggle.html

  * igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions:
    - shard-glk:          [FAIL][267] ([i915#2346]) -> [PASS][268]
   [267]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-glk8/igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions.html
   [268]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-glk4/igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions.html

  * igt@kms_flip@flip-vs-expired-vblank-interruptible@a-hdmi-a4:
    - {shard-dg1}:        [FAIL][269] ([i915#79]) -> [PASS][270]
   [269]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg1-17/igt@kms_flip@flip-vs-expired-vblank-interruptible@a-hdmi-a4.html
   [270]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg1-17/igt@kms_flip@flip-vs-expired-vblank-interruptible@a-hdmi-a4.html

  * igt@kms_flip@flip-vs-suspend-interruptible@a-dp1:
    - shard-apl:          [ABORT][271] ([i915#180]) -> [PASS][272]
   [271]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-apl2/igt@kms_flip@flip-vs-suspend-interruptible@a-dp1.html
   [272]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-apl6/igt@kms_flip@flip-vs-suspend-interruptible@a-dp1.html

  * igt@kms_frontbuffer_tracking@fbc-1p-primscrn-indfb-plflip-blt:
    - shard-dg2:          [FAIL][273] ([i915#6880]) -> [PASS][274] +1 similar issue
   [273]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg2-11/igt@kms_frontbuffer_tracking@fbc-1p-primscrn-indfb-plflip-blt.html
   [274]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-12/igt@kms_frontbuffer_tracking@fbc-1p-primscrn-indfb-plflip-blt.html

  * igt@kms_plane_scaling@intel-max-src-size@pipe-a-hdmi-a-2:
    - shard-rkl:          [FAIL][275] ([i915#8292]) -> [PASS][276]
   [275]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-rkl-4/igt@kms_plane_scaling@intel-max-src-size@pipe-a-hdmi-a-2.html
   [276]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-4/igt@kms_plane_scaling@intel-max-src-size@pipe-a-hdmi-a-2.html

  * igt@kms_sysfs_edid_timing:
    - shard-dg2:          [FAIL][277] ([IGT#2]) -> [PASS][278]
   [277]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg2-5/igt@kms_sysfs_edid_timing.html
   [278]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-11/igt@kms_sysfs_edid_timing.html

  * igt@kms_vblank@pipe-a-ts-continuation-suspend:
    - shard-dg2:          [FAIL][279] ([fdo#103375] / [i915#6121]) -> [PASS][280]
   [279]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg2-5/igt@kms_vblank@pipe-a-ts-continuation-suspend.html
   [280]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-6/igt@kms_vblank@pipe-a-ts-continuation-suspend.html

  * igt@perf_pmu@busy-double-start@ccs0:
    - shard-mtlp:         [FAIL][281] ([i915#4349]) -> [PASS][282]
   [281]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-4/igt@perf_pmu@busy-double-start@ccs0.html
   [282]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-7/igt@perf_pmu@busy-double-start@ccs0.html

  * igt@perf_pmu@most-busy-idle-check-all@bcs0:
    - shard-mtlp:         [FAIL][283] ([i915#5234]) -> [PASS][284] +1 similar issue
   [283]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-3/igt@perf_pmu@most-busy-idle-check-all@bcs0.html
   [284]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-2/igt@perf_pmu@most-busy-idle-check-all@bcs0.html

  * igt@perf_pmu@most-busy-idle-check-all@rcs0:
    - shard-dg2:          [FAIL][285] ([i915#5234]) -> [PASS][286]
   [285]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg2-5/igt@perf_pmu@most-busy-idle-check-all@rcs0.html
   [286]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-2/igt@perf_pmu@most-busy-idle-check-all@rcs0.html
    - {shard-dg1}:        [FAIL][287] ([i915#5234]) -> [PASS][288]
   [287]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg1-13/igt@perf_pmu@most-busy-idle-check-all@rcs0.html
   [288]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg1-18/igt@perf_pmu@most-busy-idle-check-all@rcs0.html

  * igt@sysfs_timeslice_duration@timeout@vecs0:
    - shard-mtlp:         [ABORT][289] ([i915#8521]) -> [PASS][290]
   [289]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-mtlp-2/igt@sysfs_timeslice_duration@timeout@vecs0.html
   [290]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-mtlp-5/igt@sysfs_timeslice_duration@timeout@vecs0.html

  
#### Warnings ####

  * igt@gem_exec_suspend@basic-s0@smem:
    - shard-snb:          [DMESG-FAIL][291] ([fdo#103375]) -> [DMESG-WARN][292] ([i915#8841])
   [291]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-snb7/igt@gem_exec_suspend@basic-s0@smem.html
   [292]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-snb4/igt@gem_exec_suspend@basic-s0@smem.html

  * igt@i915_pm_rc6_residency@rc6-idle@bcs0:
    - shard-tglu:         [FAIL][293] ([i915#2681] / [i915#3591]) -> [WARN][294] ([i915#2681])
   [293]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-tglu-5/igt@i915_pm_rc6_residency@rc6-idle@bcs0.html
   [294]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-8/igt@i915_pm_rc6_residency@rc6-idle@bcs0.html

  * igt@i915_pm_rc6_residency@rc6-idle@rcs0:
    - shard-tglu:         [WARN][295] ([i915#2681]) -> [FAIL][296] ([i915#2681] / [i915#3591])
   [295]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-tglu-5/igt@i915_pm_rc6_residency@rc6-idle@rcs0.html
   [296]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-8/igt@i915_pm_rc6_residency@rc6-idle@rcs0.html

  * igt@i915_suspend@basic-s2idle-without-i915:
    - shard-snb:          [DMESG-WARN][297] ([i915#8841]) -> [ABORT][298] ([i915#4528] / [i915#8213])
   [297]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-snb7/igt@i915_suspend@basic-s2idle-without-i915.html
   [298]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-snb4/igt@i915_suspend@basic-s2idle-without-i915.html

  * igt@kms_content_protection@mei_interface:
    - shard-dg2:          [SKIP][299] ([i915#7118] / [i915#7162]) -> [SKIP][300] ([i915#7118])
   [299]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg2-11/igt@kms_content_protection@mei_interface.html
   [300]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-1/igt@kms_content_protection@mei_interface.html
    - shard-rkl:          [SKIP][301] ([fdo#109300]) -> [SKIP][302] ([i915#7118])
   [301]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-rkl-1/igt@kms_content_protection@mei_interface.html
   [302]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-6/igt@kms_content_protection@mei_interface.html
    - shard-tglu:         [SKIP][303] ([fdo#109300]) -> [SKIP][304] ([i915#6944] / [i915#7116] / [i915#7118])
   [303]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-tglu-7/igt@kms_content_protection@mei_interface.html
   [304]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-tglu-2/igt@kms_content_protection@mei_interface.html

  * igt@kms_force_connector_basic@force-load-detect:
    - shard-rkl:          [SKIP][305] ([fdo#109285] / [i915#4098]) -> [SKIP][306] ([fdo#109285])
   [305]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-rkl-4/igt@kms_force_connector_basic@force-load-detect.html
   [306]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-rkl-7/igt@kms_force_connector_basic@force-load-detect.html

  * igt@prime_mmap@test_aperture_limit@test_aperture_limit-smem:
    - shard-dg2:          [CRASH][307] ([i915#7331]) -> [INCOMPLETE][308] ([i915#5493])
   [307]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_13369/shard-dg2-10/igt@prime_mmap@test_aperture_limit@test_aperture_limit-smem.html
   [308]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/shard-dg2-5/igt@prime_mmap@test_aperture_limit@test_aperture_limit-smem.html

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

  [IGT#2]: https://gitlab.freedesktop.org/drm/igt-gpu-tools/issues/2
  [fdo#103375]: https://bugs.freedesktop.org/show_bug.cgi?id=103375
  [fdo#109271]: https://bugs.freedesktop.org/show_bug.cgi?id=109271
  [fdo#109274]: https://bugs.freedesktop.org/show_bug.cgi?id=109274
  [fdo#109279]: https://bugs.freedesktop.org/show_bug.cgi?id=109279
  [fdo#109280]: https://bugs.freedesktop.org/show_bug.cgi?id=109280
  [fdo#109285]: https://bugs.freedesktop.org/show_bug.cgi?id=109285
  [fdo#109289]: https://bugs.freedesktop.org/show_bug.cgi?id=109289
  [fdo#109293]: https://bugs.freedesktop.org/show_bug.cgi?id=109293
  [fdo#109300]: https://bugs.freedesktop.org/show_bug.cgi?id=109300
  [fdo#109314]: https://bugs.freedesktop.org/show_bug.cgi?id=109314
  [fdo#109315]: https://bugs.freedesktop.org/show_bug.cgi?id=109315
  [fdo#109642]: https://bugs.freedesktop.org/show_bug.cgi?id=109642
  [fdo#110189]: https://bugs.freedesktop.org/show_bug.cgi?id=110189
  [fdo#110723]: https://bugs.freedesktop.org/show_bug.cgi?id=110723
  [fdo#111068]: https://bugs.freedesktop.org/show_bug.cgi?id=111068
  [fdo#111614]: https://bugs.freedesktop.org/show_bug.cgi?id=111614
  [fdo#111615]: https://bugs.freedesktop.org/show_bug.cgi?id=111615
  [fdo#111767]: https://bugs.freedesktop.org/show_bug.cgi?id=111767
  [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#1072]: https://gitlab.freedesktop.org/drm/intel/issues/1072
  [i915#1257]: https://gitlab.freedesktop.org/drm/intel/issues/1257
  [i915#1397]: https://gitlab.freedesktop.org/drm/intel/issues/1397
  [i915#180]: https://gitlab.freedesktop.org/drm/intel/issues/180
  [i915#1825]: https://gitlab.freedesktop.org/drm/intel/issues/1825
  [i915#1982]: https://gitlab.freedesktop.org/drm/intel/issues/1982
  [i915#2017]: https://gitlab.freedesktop.org/drm/intel/issues/2017
  [i915#2346]: https://gitlab.freedesktop.org/drm/intel/issues/2346
  [i915#2437]: https://gitlab.freedesktop.org/drm/intel/issues/2437
  [i915#2527]: https://gitlab.freedesktop.org/drm/intel/issues/2527
  [i915#2575]: https://gitlab.freedesktop.org/drm/intel/issues/2575
  [i915#2587]: https://gitlab.freedesktop.org/drm/intel/issues/2587
  [i915#2672]: https://gitlab.freedesktop.org/drm/intel/issues/2672
  [i915#2681]: https://gitlab.freedesktop.org/drm/intel/issues/2681
  [i915#284]: https://gitlab.freedesktop.org/drm/intel/issues/284
  [i915#2842]: https://gitlab.freedesktop.org/drm/intel/issues/2842
  [i915#2856]: https://gitlab.freedesktop.org/drm/intel/issues/2856
  [i915#3023]: https://gitlab.freedesktop.org/drm/intel/issues/3023
  [i915#3281]: https://gitlab.freedesktop.org/drm/intel/issues/3281
  [i915#3282]: https://gitlab.freedesktop.org/drm/intel/issues/3282
  [i915#3297]: https://gitlab.freedesktop.org/drm/intel/issues/3297
  [i915#3359]: https://gitlab.freedesktop.org/drm/intel/issues/3359
  [i915#3458]: https://gitlab.freedesktop.org/drm/intel/issues/3458
  [i915#3469]: https://gitlab.freedesktop.org/drm/intel/issues/3469
  [i915#3539]: https://gitlab.freedesktop.org/drm/intel/issues/3539
  [i915#3546]: https://gitlab.freedesktop.org/drm/intel/issues/3546
  [i915#3555]: https://gitlab.freedesktop.org/drm/intel/issues/3555
  [i915#3591]: https://gitlab.freedesktop.org/drm/intel/issues/3591
  [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#3743]: https://gitlab.freedesktop.org/drm/intel/issues/3743
  [i915#3804]: https://gitlab.freedesktop.org/drm/intel/issues/3804
  [i915#3840]: https://gitlab.freedesktop.org/drm/intel/issues/3840
  [i915#3886]: https://gitlab.freedesktop.org/drm/intel/issues/3886
  [i915#3955]: https://gitlab.freedesktop.org/drm/intel/issues/3955
  [i915#4070]: https://gitlab.freedesktop.org/drm/intel/issues/4070
  [i915#4077]: https://gitlab.freedesktop.org/drm/intel/issues/4077
  [i915#4078]: https://gitlab.freedesktop.org/drm/intel/issues/4078
  [i915#4079]: https://gitlab.freedesktop.org/drm/intel/issues/4079
  [i915#4083]: https://gitlab.freedesktop.org/drm/intel/issues/4083
  [i915#4098]: https://gitlab.freedesktop.org/drm/intel/issues/4098
  [i915#4212]: https://gitlab.freedesktop.org/drm/intel/issues/4212
  [i915#4215]: https://gitlab.freedesktop.org/drm/intel/issues/4215
  [i915#4235]: https://gitlab.freedesktop.org/drm/intel/issues/4235
  [i915#4270]: https://gitlab.freedesktop.org/drm/intel/issues/4270
  [i915#4281]: https://gitlab.freedesktop.org/drm/intel/issues/4281
  [i915#4348]: https://gitlab.freedesktop.org/drm/intel/issues/4348
  [i915#4349]: https://gitlab.freedesktop.org/drm/intel/issues/4349
  [i915#4391]: https://gitlab.freedesktop.org/drm/intel/issues/4391
  [i915#4528]: https://gitlab.freedesktop.org/drm/intel/issues/4528
  [i915#4538]: https://gitlab.freedesktop.org/drm/intel/issues/4538
  [i915#4613]: https://gitlab.freedesktop.org/drm/intel/issues/4613
  [i915#4812]: https://gitlab.freedesktop.org/drm/intel/issues/4812
  [i915#4880]: https://gitlab.freedesktop.org/drm/intel/issues/4880
  [i915#4936]: https://gitlab.freedesktop.org/drm/intel/issues/4936
  [i915#5138]: https://gitlab.freedesktop.org/drm/intel/issues/5138
  [i915#5176]: https://gitlab.freedesktop.org/drm/intel/issues/5176
  [i915#5190]: https://gitlab.freedesktop.org/drm/intel/issues/5190
  [i915#5213]: https://gitlab.freedesktop.org/drm/intel/issues/5213
  [i915#5234]: https://gitlab.freedesktop.org/drm/intel/issues/5234
  [i915#5235]: https://gitlab.freedesktop.org/drm/intel/issues/5235
  [i915#5274]: https://gitlab.freedesktop.org/drm/intel/issues/5274
  [i915#5286]: https://gitlab.freedesktop.org/drm/intel/issues/5286
  [i915#5325]: https://gitlab.freedesktop.org/drm/intel/issues/5325
  [i915#533]: https://gitlab.freedesktop.org/drm/intel/issues/533
  [i915#5354]: https://gitlab.freedesktop.org/drm/intel/issues/5354
  [i915#5439]: https://gitlab.freedesktop.org/drm/intel/issues/5439
  [i915#5493]: https://gitlab.freedesktop.org/drm/intel/issues/5493
  [i915#5784]: https://gitlab.freedesktop.org/drm/intel/issues/5784
  [i915#5892]: https://gitlab.freedesktop.org/drm/intel/issues/5892
  [i915#6015]: https://gitlab.freedesktop.org/drm/intel/issues/6015
  [i915#6095]: https://gitlab.freedesktop.org/drm/intel/issues/6095
  [i915#6121]: https://gitlab.freedesktop.org/drm/intel/issues/6121
  [i915#6187]: https://gitlab.freedesktop.org/drm/intel/issues/6187
  [i915#6227]: https://gitlab.freedesktop.org/drm/intel/issues/6227
  [i915#6230]: https://gitlab.freedesktop.org/drm/intel/issues/6230
  [i915#6268]: https://gitlab.freedesktop.org/drm/intel/issues/6268
  [i915#6335]: https://gitlab.freedesktop.org/drm/intel/issues/6335
  [i915#6363]: https://gitlab.freedesktop.org/drm/intel/issues/6363
  [i915#658]: https://gitlab.freedesktop.org/drm/intel/issues/658
  [i915#6763]: https://gitlab.freedesktop.org/drm/intel/issues/6763
  [i915#6768]: https://gitlab.freedesktop.org/drm/intel/issues/6768
  [i915#6880]: https://gitlab.freedesktop.org/drm/intel/issues/6880
  [i915#6944]: https://gitlab.freedesktop.org/drm/intel/issues/6944
  [i915#6953]: https://gitlab.freedesktop.org/drm/intel/issues/6953
  [i915#7059]: https://gitlab.freedesktop.org/drm/intel/issues/7059
  [i915#7069]: https://gitlab.freedesktop.org/drm/intel/issues/7069
  [i915#7116]: https://gitlab.freedesktop.org/drm/intel/issues/7116
  [i915#7118]: https://gitlab.freedesktop.org/drm/intel/issues/7118
  [i915#7162]: https://gitlab.freedesktop.org/drm/intel/issues/7162
  [i915#72]: https://gitlab.freedesktop.org/drm/intel/issues/72
  [i915#7331]: https://gitlab.freedesktop.org/drm/intel/issues/7331
  [i915#7443]: https://gitlab.freedesktop.org/drm/intel/issues/7443
  [i915#7561]: https://gitlab.freedesktop.org/drm/intel/issues/7561
  [i915#7697]: https://gitlab.freedesktop.org/drm/intel/issues/7697
  [i915#7701]: https://gitlab.freedesktop.org/drm/intel/issues/7701
  [i915#7711]: https://gitlab.freedesktop.org/drm/intel/issues/7711
  [i915#7742]: https://gitlab.freedesktop.org/drm/intel/issues/7742
  [i915#7816]: https://gitlab.freedesktop.org/drm/intel/issues/7816
  [i915#7828]: https://gitlab.freedesktop.org/drm/intel/issues/7828
  [i915#7892]: https://gitlab.freedesktop.org/drm/intel/issues/7892
  [i915#79]: https://gitlab.freedesktop.org/drm/intel/issues/79
  [i915#7913]: https://gitlab.freedesktop.org/drm/intel/issues/7913
  [i915#7916]: https://gitlab.freedesktop.org/drm/intel/issues/7916
  [i915#7940]: https://gitlab.freedesktop.org/drm/intel/issues/7940
  [i915#7941]: https://gitlab.freedesktop.org/drm/intel/issues/7941
  [i915#7975]: https://gitlab.freedesktop.org/drm/intel/issues/7975
  [i915#8063]: https://gitlab.freedesktop.org/drm/intel/issues/8063
  [i915#8102]: https://gitlab.freedesktop.org/drm/intel/issues/8102
  [i915#8131]: https://gitlab.freedesktop.org/drm/intel/issues/8131
  [i915#8213]: https://gitlab.freedesktop.org/drm/intel/issues/8213
  [i915#8228]: https://gitlab.freedesktop.org/drm/intel/issues/8228
  [i915#8247]: https://gitlab.freedesktop.org/drm/intel/issues/8247
  [i915#8248]: https://gitlab.freedesktop.org/drm/intel/issues/8248
  [i915#8292]: https://gitlab.freedesktop.org/drm/intel/issues/8292
  [i915#8381]: https://gitlab.freedesktop.org/drm/intel/issues/8381
  [i915#8414]: https://gitlab.freedesktop.org/drm/intel/issues/8414
  [i915#8428]: https://gitlab.freedesktop.org/drm/intel/issues/8428
  [i915#8430]: https://gitlab.freedesktop.org/drm/intel/issues/8430
  [i915#8489]: https://gitlab.freedesktop.org/drm/intel/issues/8489
  [i915#8502]: https://gitlab.freedesktop.org/drm/intel/issues/8502
  [i915#8503]: https://gitlab.freedesktop.org/drm/intel/issues/8503
  [i915#8516]: https://gitlab.freedesktop.org/drm/intel/issues/8516
  [i915#8521]: https://gitlab.freedesktop.org/drm/intel/issues/8521
  [i915#8623]: https://gitlab.freedesktop.org/drm/intel/issues/8623
  [i915#8661]: https://gitlab.freedesktop.org/drm/intel/issues/8661
  [i915#8668]: https://gitlab.freedesktop.org/drm/intel/issues/8668
  [i915#8708]: https://gitlab.freedesktop.org/drm/intel/issues/8708
  [i915#8709]: https://gitlab.freedesktop.org/drm/intel/issues/8709
  [i915#8726]: https://gitlab.freedesktop.org/drm/intel/issues/8726
  [i915#8758]: https://gitlab.freedesktop.org/drm/intel/issues/8758
  [i915#8807]: https://gitlab.freedesktop.org/drm/intel/issues/8807
  [i915#8810]: https://gitlab.freedesktop.org/drm/intel/issues/8810
  [i915#8814]: https://gitlab.freedesktop.org/drm/intel/issues/8814
  [i915#8841]: https://gitlab.freedesktop.org/drm/intel/issues/8841
  [i915#8875]: https://gitlab.freedesktop.org/drm/intel/issues/8875
  [i915#8878]: https://gitlab.freedesktop.org/drm/intel/issues/8878


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

  * CI: CI-20190529 -> None
  * IGT: IGT_7380 -> IGTPW_9383
  * Piglit: piglit_4509 -> None

  CI-20190529: 20190529
  CI_DRM_13369: 0595153b5c237d83f9990f3ce44d072a262dcb0e @ git://anongit.freedesktop.org/gfx-ci/linux
  IGTPW_9383: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/index.html
  IGT_7380: 8e65f12de2fd52c05dc48fdbcb8cfe86f6de1a75 @ 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_9383/index.html

[-- Attachment #2: Type: text/html, Size: 94229 bytes --]

Re: [igt-dev] ✗ Fi.CI.IGT: failure for Extend intel_blt to work on Xe (rev3)

[Yedireswarapu, SaiX Nandan]

Hi,

Issue re-reported, https://patchwork.freedesktop.org/series/120162/#rev3

Thanks,
Y Sai Nandan

-----Original Message-----
From: Kempczynski, Zbigniew <zbigniew.kempczynski@intel.com> 
Sent: Wednesday, July 12, 2023 1:44 PM
To: igt-dev@lists.freedesktop.org
Cc: Yedireswarapu, SaiX Nandan <saix.nandan.yedireswarapu@intel.com>
Subject: Re: ✗ Fi.CI.IGT: failure for Extend intel_blt to work on Xe (rev3)

On Tue, Jul 11, 2023 at 02:52:12PM +0000, Patchwork wrote:
>    Patch Details
> 
>    Series:  Extend intel_blt to work on Xe (rev3)                          
>    URL:     https://patchwork.freedesktop.org/series/120162/               
>    State:   failure                                                        
>    Details: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/index.html 
> 
>          CI Bug Log - changes from CI_DRM_13369_full -> IGTPW_9383_full
> 
> Summary
> 
>    FAILURE
> 
>    Serious unknown changes coming with IGTPW_9383_full absolutely need to be
>    verified manually.
> 
>    If you think the reported changes have nothing to do with the changes
>    introduced in IGTPW_9383_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_9383/index.html
> 
> Participating hosts (9 -> 9)
> 
>    No changes in participating hosts
> 
> Possible new issues
> 
>    Here are the unknown changes that may have been introduced in
>    IGTPW_9383_full:
> 
>   IGT changes
> 
>     Possible regressions
> 
>      * igt@prime_vgem@fence-wait@ccs0:
> 
>           * shard-mtlp: PASS -> DMESG-WARN
>      * igt@prime_vgem@fence-wait@vecs0:
> 
>           * shard-mtlp: PASS -> ABORT
>      * igt@vgem_basic@sysfs:
> 
>           * shard-snb: PASS -> ABORT

Unrelated to the change.

--
Zbigniew

> 
> Known issues
> 
>    Here are the changes found in IGTPW_9383_full that come from known issues:
> 
>   IGT changes
> 
>     Issues hit
> 
>      * igt@api_intel_bb@crc32:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#6230)
>      * igt@device_reset@unbind-cold-reset-rebind:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#7701)
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#7701)
> 
>      * igt@drm_fdinfo@busy-hang@bcs0:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#8414) +19 similar issues
>      * igt@drm_fdinfo@most-busy-idle-check-all@rcs0:
> 
>           * shard-rkl: PASS -> FAIL (i915#7742)
>      * igt@drm_fdinfo@virtual-busy-all:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#8414)
>      * igt@gem_bad_reloc@negative-reloc-bltcopy:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#3281) +5 similar issues
>      * igt@gem_basic@multigpu-create-close:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#7697) +1 similar issue
>      * igt@gem_ccs@block-copy-compressed:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#3555 / i915#5325)
>      * igt@gem_ccs@ctrl-surf-copy-new-ctx:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#5325)
>      * igt@gem_close_race@multigpu-basic-process:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#7697) +1 similar issue
>      * igt@gem_create@create-ext-cpu-access-big:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#6335)
>      * igt@gem_ctx_exec@basic-nohangcheck:
> 
>           * shard-tglu: PASS -> FAIL (i915#6268)
>      * igt@gem_ctx_param@set-priority-not-supported:
> 
>           * shard-dg2: NOTRUN -> SKIP (fdo#109314)
> 
>           * shard-rkl: NOTRUN -> SKIP (fdo#109314)
> 
>      * igt@gem_ctx_persistence@saturated-hostile@vecs0:
> 
>           * shard-mtlp: PASS -> FAIL (i915#7816) +2 similar issues
>      * igt@gem_eio@in-flight-contexts-1us:
> 
>           * shard-mtlp: PASS -> ABORT (i915#8503)
>      * igt@gem_eio@kms:
> 
>           * shard-dg2: PASS -> INCOMPLETE (i915#7892)
>      * igt@gem_exec_balancer@bonded-true-hang:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#4812) +2 similar issues
>      * igt@gem_exec_fair@basic-none@bcs0:
> 
>           * shard-tglu: NOTRUN -> FAIL (i915#2842) +4 similar issues
>      * igt@gem_exec_fair@basic-pace@vcs0:
> 
>           * shard-glk: PASS -> FAIL (i915#2842) +1 similar issue
>      * igt@gem_exec_fair@basic-throttle:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#3539)
>      * igt@gem_exec_fence@submit67:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#4812)
>      * igt@gem_exec_params@secure-non-root:
> 
>           * shard-tglu: NOTRUN -> SKIP (fdo#112283)
>      * igt@gem_exec_reloc@basic-write-gtt-active:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#3281) +2 similar issues
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#3281) +1 similar issue
> 
>      * igt@gem_exec_whisper@basic-forked:
> 
>           * shard-mtlp: NOTRUN -> FAIL (i915#6363)
>      * igt@gem_exec_whisper@basic-forked-all:
> 
>           * shard-mtlp: PASS -> FAIL (i915#6363)
>      * igt@gem_lmem_swapping@heavy-verify-random-ccs:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#4613) +2 similar issues
>      * igt@gem_lmem_swapping@parallel-multi:
> 
>           * shard-apl: NOTRUN -> SKIP (fdo#109271 / i915#4613)
>      * igt@gem_lmem_swapping@parallel-random:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#4613)
>      * igt@gem_lmem_swapping@smem-oom@lmem0:
> 
>           * shard-dg2: PASS -> TIMEOUT (i915#5493)
>      * igt@gem_media_vme:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#284)
>      * igt@gem_mmap_wc@write:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#4083) +1 similar issue
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#4083) +3 similar issues
> 
>      * igt@gem_partial_pwrite_pread@reads:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#3282) +3 similar issues
>      * igt@gem_partial_pwrite_pread@writes-after-reads:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#3282) +2 similar issues
>      * igt@gem_pxp@create-regular-buffer:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#4270)
>      * igt@gem_pxp@fail-invalid-protected-context:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#4270) +1 similar issue
>      * igt@gem_readwrite@beyond-eob:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#3282) +3 similar issues
>      * igt@gem_render_copy@y-tiled-ccs-to-yf-tiled-ccs:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#8428) +3 similar issues
>      * igt@gem_set_tiling_vs_pwrite:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#4079)
>      * igt@gem_tiled_fence_blits@basic:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#4077) +3 similar issues
>      * igt@gem_tiled_pread_basic:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#4079)
>      * igt@gem_userptr_blits@coherency-unsync:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#3297)
>      * igt@gem_userptr_blits@invalid-mmap-offset-unsync:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#3297)
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#3297)
> 
>      * igt@gem_userptr_blits@map-fixed-invalidate-overlap-busy:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#3297 / i915#4880)
>      * igt@gem_userptr_blits@nohangcheck:
> 
>           * shard-mtlp: PASS -> FAIL (i915#7916)
>      * igt@gen7_exec_parse@load-register-reg:
> 
>           * shard-dg2: NOTRUN -> SKIP (fdo#109289)
>      * igt@gen7_exec_parse@oacontrol-tracking:
> 
>           * shard-mtlp: NOTRUN -> SKIP (fdo#109289)
>      * igt@gen9_exec_parse@bb-large:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#2856)
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#2527)
> 
>      * igt@i915_hangman@gt-engine-error@vcs0:
> 
>           * shard-mtlp: PASS -> FAIL (i915#7069)
>      * igt@i915_module_load@load:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#6227)
>      * igt@i915_pm_backlight@bad-brightness:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#5354 / i915#7561)
>      * igt@i915_pm_dc@dc5-psr:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#658)
>      * igt@i915_pm_dc@dc9-dpms:
> 
>           * shard-apl: PASS -> SKIP (fdo#109271)
>      * igt@i915_pm_lpsp@screens-disabled:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#8430)
>      * igt@i915_pm_rpm@modeset-non-lpsp-stress:
> 
>           * shard-dg2: PASS -> SKIP (i915#1397) +1 similar issue
>      * igt@i915_pm_rpm@pc8-residency:
> 
>           * shard-mtlp: NOTRUN -> SKIP (fdo#109293)
>      * igt@i915_pm_rpm@system-suspend-execbuf:
> 
>           * shard-tglu: NOTRUN -> FAIL (i915#7940)
>      * igt@i915_suspend@basic-s3-without-i915:
> 
>           * shard-dg2: PASS -> FAIL (fdo#103375 / i915#6121) +2 similar
>             issues
> 
>           * shard-tglu: NOTRUN -> INCOMPLETE (i915#7443 / i915#8102)
> 
>      * igt@i915_suspend@debugfs-reader:
> 
>           * shard-apl: PASS -> ABORT (i915#180 / i915#8213)
>      * igt@i915_suspend@fence-restore-tiled2untiled:
> 
>           * shard-snb: NOTRUN -> DMESG-WARN (i915#8841) +4 similar issues
>      * igt@kms_addfb_basic@addfb25-x-tiled-legacy:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#4212)
>      * igt@kms_addfb_basic@basic-y-tiled-legacy:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#4215 / i915#5190)
>      * igt@kms_async_flips@crc@pipe-a-hdmi-a-3:
> 
>           * shard-dg2: NOTRUN -> FAIL (i915#8247) +3 similar issues
>      * igt@kms_big_fb@4-tiled-16bpp-rotate-90:
> 
>           * shard-mtlp: NOTRUN -> SKIP (fdo#111614)
>      * igt@kms_big_fb@4-tiled-8bpp-rotate-270:
> 
>           * shard-dg2: NOTRUN -> SKIP (fdo#111614) +1 similar issue
>      * igt@kms_big_fb@4-tiled-max-hw-stride-32bpp-rotate-180-async-flip:
> 
>           * shard-mtlp: PASS -> FAIL (i915#3743)
>      * igt@kms_big_fb@4-tiled-max-hw-stride-32bpp-rotate-180-hflip:
> 
>           * shard-tglu: NOTRUN -> SKIP (fdo#111615 / i915#5286)
>      * igt@kms_big_fb@4-tiled-max-hw-stride-64bpp-rotate-0-hflip:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#5286) +1 similar issue
> 
>           * shard-mtlp: PASS -> FAIL (i915#5138)
> 
>      * igt@kms_big_fb@x-tiled-32bpp-rotate-90:
> 
>           * shard-rkl: NOTRUN -> SKIP (fdo#111614 / i915#3638)
>      * igt@kms_big_fb@x-tiled-max-hw-stride-64bpp-rotate-0-hflip-async-flip:
> 
>           * shard-mtlp: NOTRUN -> FAIL (i915#3743)
>      * igt@kms_big_fb@y-tiled-8bpp-rotate-90:
> 
>           * shard-tglu: NOTRUN -> SKIP (fdo#111614)
>      * igt@kms_big_fb@yf-tiled-64bpp-rotate-90:
> 
>           * shard-rkl: NOTRUN -> SKIP (fdo#110723) +2 similar issues
>      * igt@kms_big_fb@yf-tiled-addfb-size-overflow:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#5190) +5 similar issues
> 
>           * shard-tglu: NOTRUN -> SKIP (fdo#111615) +1 similar issue
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#6187)
> 
>      * igt@kms_big_fb@yf-tiled-max-hw-stride-32bpp-rotate-180-async-flip:
> 
>           * shard-mtlp: NOTRUN -> SKIP (fdo#111615) +4 similar issues
>      * igt@kms_big_fb@yf-tiled-max-hw-stride-64bpp-rotate-0:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#4538 / i915#5190) +3 similar
>             issues
>      * igt@kms_ccs@pipe-a-bad-pixel-format-y_tiled_gen12_mc_ccs:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#3886 / i915#6095) +5 similar
>             issues
>      * igt@kms_ccs@pipe-a-ccs-on-another-bo-4_tiled_mtl_rc_ccs_cc:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#5354 / i915#6095) +4 similar
>             issues
>      * igt@kms_ccs@pipe-a-crc-sprite-planes-basic-y_tiled_ccs:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#3689 / i915#5354) +12 similar
>             issues
>      * igt@kms_ccs@pipe-a-missing-ccs-buffer-4_tiled_mtl_rc_ccs_cc:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#5354 / i915#6095) +8 similar
>             issues
>      * igt@kms_ccs@pipe-b-bad-aux-stride-y_tiled_gen12_mc_ccs:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#3886 / i915#5354 / i915#6095)
>      * igt@kms_ccs@pipe-b-ccs-on-another-bo-y_tiled_gen12_mc_ccs:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#3689 / i915#3886 / i915#5354 /
>             i915#6095) +2 similar issues
>      * igt@kms_ccs@pipe-b-crc-primary-basic-yf_tiled_ccs:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#3734 / i915#5354 / i915#6095) +1
>             similar issue
>      * igt@kms_ccs@pipe-c-bad-aux-stride-yf_tiled_ccs:
> 
>           * shard-tglu: NOTRUN -> SKIP (fdo#111615 / i915#3689 / i915#5354 /
>             i915#6095) +1 similar issue
>      * igt@kms_ccs@pipe-c-bad-rotation-90-4_tiled_mtl_mc_ccs:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#5354) +18 similar issues
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#5354) +8 similar issues
> 
>      * igt@kms_ccs@pipe-c-bad-rotation-90-y_tiled_gen12_rc_ccs_cc:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#3689 / i915#3886 / i915#5354) +3
>             similar issues
>      * igt@kms_ccs@pipe-d-bad-rotation-90-4_tiled_dg2_rc_ccs_cc:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#3689 / i915#5354 / i915#6095) +9
>             similar issues
>      * igt@kms_ccs@pipe-d-crc-primary-rotation-180-yf_tiled_ccs:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#6095) +16 similar issues
>      * igt@kms_chamelium_audio@dp-audio:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#7828) +3 similar issues
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#7828)
> 
>      * igt@kms_chamelium_color@ctm-0-50:
> 
>           * shard-tglu: NOTRUN -> SKIP (fdo#111827) +1 similar issue
>      * igt@kms_chamelium_color@ctm-0-75:
> 
>           * shard-mtlp: NOTRUN -> SKIP (fdo#111827)
>      * igt@kms_chamelium_color@ctm-blue-to-red:
> 
>           * shard-dg2: NOTRUN -> SKIP (fdo#111827)
>      * igt@kms_chamelium_edid@dp-edid-stress-resolution-non-4k:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#7828)
>      * igt@kms_chamelium_frames@hdmi-cmp-planar-formats:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#7828) +1 similar issue
>      * igt@kms_content_protection@legacy:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#7118) +2 similar issues
>      * igt@kms_content_protection@mei_interface:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#8063)
>      * igt@kms_content_protection@uevent:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#7118)
>      * igt@kms_cursor_crc@cursor-onscreen-512x170:
> 
>           * shard-rkl: NOTRUN -> SKIP (fdo#109279 / i915#3359)
>      * igt@kms_cursor_crc@cursor-onscreen-512x512:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#3359) +1 similar issue
>      * igt@kms_cursor_crc@cursor-random-32x10:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#8814)
>      * igt@kms_cursor_crc@cursor-random-512x512:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#3359)
>      * igt@kms_cursor_crc@cursor-sliding-512x512:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#3359) +2 similar issues
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#3359) +1 similar issue
> 
>      * igt@kms_cursor_legacy@2x-long-flip-vs-cursor-legacy:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#3546)
>      * igt@kms_cursor_legacy@cursora-vs-flipb-legacy:
> 
>           * shard-tglu: NOTRUN -> SKIP (fdo#109274) +3 similar issues
>      * igt@kms_cursor_legacy@cursorb-vs-flipa-toggle:
> 
>           * shard-dg2: NOTRUN -> SKIP (fdo#109274 / i915#5354) +1 similar
>             issue
> 
>           * shard-rkl: NOTRUN -> SKIP (fdo#111825) +1 similar issue
> 
>      * igt@kms_cursor_legacy@cursorb-vs-flipb-toggle:
> 
>           * shard-dg2: NOTRUN -> SKIP (fdo#109274 / fdo#111767 / i915#5354)
> 
>           * shard-rkl: NOTRUN -> SKIP (fdo#111767 / fdo#111825)
> 
>      * igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size:
> 
>           * shard-apl: PASS -> FAIL (i915#2346)
> 
>           * shard-glk: PASS -> FAIL (i915#2346)
> 
>      * igt@kms_dither@fb-8bpc-vs-panel-6bpc@pipe-a-hdmi-a-1:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#3804)
>      * igt@kms_dp_aux_dev:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#1257)
>      * igt@kms_dsc@dsc-with-bpc-formats:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#3555 / i915#3840)
>      * igt@kms_dsc@dsc-with-output-formats:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#3840)
>      * igt@kms_fbcon_fbt@psr:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#3469)
> 
>           * shard-rkl: NOTRUN -> SKIP (fdo#110189 / i915#3955)
> 
>      * igt@kms_flip@2x-blocking-absolute-wf_vblank:
> 
>           * shard-tglu: NOTRUN -> SKIP (fdo#109274 / i915#3637) +2 similar
>             issues
>      * igt@kms_flip@2x-flip-vs-blocking-wf-vblank:
> 
>           * shard-snb: NOTRUN -> SKIP (fdo#109271 / fdo#111767)
>      * igt@kms_flip@2x-flip-vs-fences-interruptible:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#8381)
>      * igt@kms_flip@2x-plain-flip-fb-recreate:
> 
>           * shard-apl: NOTRUN -> SKIP (fdo#109271) +24 similar issues
>      * igt@kms_flip@2x-plain-flip-interruptible:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#3637)
>      * igt@kms_flip@2x-wf_vblank-ts-check-interruptible:
> 
>           * shard-dg2: NOTRUN -> SKIP (fdo#109274)
>      * igt@kms_flip_scaled_crc@flip-32bpp-4tile-to-32bpp-4tiledg2rcccs-downscaling@pipe-a-valid-mode:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#2672)
>      * igt@kms_flip_scaled_crc@flip-32bpp-ytile-to-32bpp-ytilegen12rcccs-upscaling@pipe-a-default-mode:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#2672) +1 similar issue
>      * igt@kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-downscaling@pipe-a-default-mode:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#8810)
>      * igt@kms_force_connector_basic@prune-stale-modes:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#5274)
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#5274)
> 
>      * igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-draw-mmap-gtt:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#8708) +1 similar issue
>      * igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-fullscreen:
> 
>           * shard-dg2: PASS -> FAIL (i915#6880)
>      * igt@kms_frontbuffer_tracking@fbcpsr-1p-primscrn-pri-shrfb-draw-mmap-gtt:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#3023) +4 similar issues
>      * igt@kms_frontbuffer_tracking@fbcpsr-2p-primscrn-pri-shrfb-draw-mmap-wc:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#8708) +5 similar issues
> 
>           * shard-rkl: NOTRUN -> SKIP (fdo#111825 / i915#1825) +8 similar
>             issues
> 
>      * igt@kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-pri-shrfb-draw-pwrite:
> 
>           * shard-tglu: NOTRUN -> SKIP (fdo#109280) +14 similar issues
>      * igt@kms_frontbuffer_tracking@fbcpsr-tiling-4:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#5439)
>      * igt@kms_frontbuffer_tracking@psr-1p-offscren-pri-indfb-draw-mmap-cpu:
> 
>           * shard-tglu: NOTRUN -> SKIP (fdo#110189) +12 similar issues
>      * igt@kms_frontbuffer_tracking@psr-1p-offscren-pri-shrfb-draw-render:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#3458) +3 similar issues
>      * igt@kms_frontbuffer_tracking@psr-2p-scndscrn-pri-shrfb-draw-mmap-cpu:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#1825) +10 similar issues
>      * igt@kms_hdr@bpc-switch:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#3555 / i915#8228)
>      * igt@kms_hdr@bpc-switch-suspend:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#3555 / i915#8228)
>      * igt@kms_hdr@static-toggle:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#3555 / i915#8228) +1 similar
>             issue
>      * igt@kms_pipe_b_c_ivb@pipe-b-dpms-off-modeset-pipe-c:
> 
>           * shard-tglu: NOTRUN -> SKIP (fdo#109289) +1 similar issue
>      * igt@kms_plane_scaling@intel-max-src-size:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#6953)
>      * igt@kms_plane_scaling@plane-scaler-with-clipping-clamping-rotation@pipe-a-vga-1:
> 
>           * shard-snb: NOTRUN -> SKIP (fdo#109271) +101 similar issues
>      * igt@kms_plane_scaling@plane-upscale-with-rotation-factor-0-25@pipe-a-hdmi-a-1:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#5176) +15 similar issues
>      * igt@kms_plane_scaling@plane-upscale-with-rotation-factor-0-25@pipe-a-hdmi-a-2:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#5176) +7 similar issues
>      * igt@kms_plane_scaling@planes-downscale-factor-0-25@pipe-d-hdmi-a-3:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#5235) +15 similar issues
>      * igt@kms_plane_scaling@planes-upscale-20x20-downscale-factor-0-25@pipe-b-edp-1:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#5235) +3 similar issues
>      * igt@kms_plane_scaling@planes-upscale-factor-0-25-downscale-factor-0-25@pipe-b-hdmi-a-2:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#5235) +3 similar issues
>      * igt@kms_psr2_sf@cursor-plane-move-continuous-sf:
> 
>           * shard-apl: NOTRUN -> SKIP (fdo#109271 / i915#658)
>      * igt@kms_psr2_su@frontbuffer-xrgb8888:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#658) +1 similar issue
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#4348)
> 
>      * igt@kms_psr2_su@page_flip-xrgb8888:
> 
>           * shard-tglu: NOTRUN -> SKIP (fdo#109642 / fdo#111068 / i915#658)
>             +1 similar issue
>      * igt@kms_psr@primary_mmap_cpu:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#1072) +4 similar issues
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#1072) +2 similar issues
> 
>      * igt@kms_psr@psr2_cursor_blt:
> 
>           * shard-mtlp: PASS -> FAIL (i915#8726)
>      * igt@kms_psr@psr2_cursor_mmap_gtt:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#4077) +2 similar issues
>      * igt@kms_rmfb@close-fd@pipe-a-edp-1:
> 
>           * shard-mtlp: PASS -> DMESG-WARN (i915#1982)
>      * igt@kms_rotation_crc@primary-y-tiled-reflect-x-270:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#4235)
>      * igt@kms_scaling_modes@scaling-mode-full:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#3555) +2 similar issues
>      * igt@kms_tiled_display@basic-test-pattern:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#8623)
>      * igt@kms_tiled_display@basic-test-pattern-with-chamelium:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#8623)
>      * igt@kms_vblank@pipe-c-query-forked-busy:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#4070 / i915#6768)
>      * igt@kms_vblank@pipe-d-wait-idle:
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#4070 / i915#533 / i915#6768)
>      * igt@kms_writeback@writeback-pixel-formats:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#2437)
>      * igt@perf@oa-exponents@0-rcs0:
> 
>           * shard-glk: PASS -> ABORT (i915#5213 / i915#7941)
>      * igt@perf_pmu@event-wait@rcs0:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#8807)
>      * igt@perf_pmu@rc6@other-idle-gt0:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#8516)
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#8516)
> 
>      * igt@perf_pmu@render-node-busy-idle@ccs0:
> 
>           * shard-mtlp: PASS -> FAIL (i915#4349) +7 similar issues
>      * igt@prime_vgem@basic-gtt:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#3708 / i915#4077)
>      * igt@sysfs_heartbeat_interval@nopreempt@bcs0:
> 
>           * shard-mtlp: PASS -> FAIL (i915#6015) +1 similar issue
>      * igt@sysfs_preempt_timeout@timeout@vecs0:
> 
>           * shard-mtlp: PASS -> ABORT (i915#8521)
>      * igt@v3d/v3d_perfmon@create-perfmon-exceed:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#2575) +5 similar issues
>      * igt@v3d/v3d_submit_cl@valid-submission:
> 
>           * shard-tglu: NOTRUN -> SKIP (fdo#109315 / i915#2575) +4 similar
>             issues
>      * igt@v3d/v3d_submit_csd@bad-multisync-in-sync:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#2575) +2 similar issues
> 
>           * shard-rkl: NOTRUN -> SKIP (fdo#109315) +2 similar issues
> 
>      * igt@vc4/vc4_mmap@mmap-bo:
> 
>           * shard-dg2: NOTRUN -> SKIP (i915#7711) +3 similar issues
> 
>           * shard-rkl: NOTRUN -> SKIP (i915#7711) +2 similar issues
> 
>      * igt@vc4/vc4_perfmon@create-perfmon-exceed:
> 
>           * shard-mtlp: NOTRUN -> SKIP (i915#7711) +3 similar issues
>      * igt@vc4/vc4_purgeable_bo@mark-unpurgeable-purged:
> 
>           * shard-tglu: NOTRUN -> SKIP (i915#2575) +4 similar issues
>      * igt@vgem_basic@bad-pad:
> 
>           * shard-mtlp: PASS -> DMESG-WARN (i915#2017)
> 
>     Possible fixes
> 
>      * igt@gem_create@hog-create@smem0:
> 
>           * shard-dg2: FAIL (i915#5892 / i915#8758) -> PASS
>      * igt@gem_eio@hibernate:
> 
>           * {shard-dg1}: ABORT (i915#4391 / i915#7975 / i915#8213) -> PASS
>      * igt@gem_exec_fair@basic-pace-share@rcs0:
> 
>           * shard-glk: FAIL (i915#2842) -> PASS
>      * igt@gem_exec_fair@basic-throttle@rcs0:
> 
>           * shard-rkl: FAIL (i915#2842) -> PASS +3 similar issues
>      * igt@gem_exec_whisper@basic-contexts-forked-all:
> 
>           * shard-mtlp: ABORT (i915#8131) -> PASS
>      * igt@gem_exec_whisper@basic-fds-priority-all:
> 
>           * shard-mtlp: FAIL (i915#6363) -> PASS
>      * igt@gem_lmem_swapping@smem-oom@lmem0:
> 
>           * {shard-dg1}: DMESG-WARN (i915#4936 / i915#5493) -> PASS
>      * igt@i915_hangman@engine-engine-hang@vcs0:
> 
>           * shard-mtlp: FAIL (i915#7069) -> PASS
>      * igt@i915_module_load@reload-with-fault-injection:
> 
>           * shard-mtlp: ABORT (i915#8489 / i915#8668) -> PASS
>      * igt@i915_pm_dc@dc9-dpms:
> 
>           * shard-tglu: SKIP (i915#4281) -> PASS
>      * igt@i915_pm_rc6_residency@rc6-idle@rcs0:
> 
>           * {shard-dg1}: FAIL (i915#3591) -> PASS +1 similar issue
>      * igt@i915_pm_rpm@cursor-dpms:
> 
>           * shard-tglu: FAIL (i915#7940) -> PASS +1 similar issue
>      * igt@i915_pm_rpm@dpms-mode-unset-lpsp:
> 
>           * {shard-dg1}: SKIP (i915#1397) -> PASS
>      * igt@i915_pm_rpm@dpms-non-lpsp:
> 
>           * shard-rkl: SKIP (i915#1397) -> PASS +1 similar issue
>      * igt@i915_pm_rpm@modeset-non-lpsp-stress-no-wait:
> 
>           * shard-dg2: SKIP (i915#1397) -> PASS +2 similar issues
>      * igt@i915_selftest@live@gt_mocs:
> 
>           * shard-mtlp: DMESG-FAIL (i915#7059) -> PASS
>      * igt@i915_selftest@live@hangcheck:
> 
>           * shard-dg2: ABORT (i915#7913) -> PASS
>      * igt@i915_selftest@live@workarounds:
> 
>           * shard-mtlp: DMESG-FAIL (i915#6763) -> PASS
>      * igt@kms_big_fb@x-tiled-max-hw-stride-64bpp-rotate-180-hflip-async-flip:
> 
>           * shard-mtlp: FAIL (i915#3743) -> PASS
>      * igt@kms_cursor_legacy@2x-long-flip-vs-cursor-atomic:
> 
>           * shard-glk: FAIL (i915#72) -> PASS
>      * igt@kms_cursor_legacy@cursor-vs-flip-toggle:
> 
>           * shard-mtlp: FAIL (i915#8248) -> PASS
>      * igt@kms_cursor_legacy@flip-vs-cursor-atomic-transitions:
> 
>           * shard-glk: FAIL (i915#2346) -> PASS
>      * igt@kms_flip@flip-vs-expired-vblank-interruptible@a-hdmi-a4:
> 
>           * {shard-dg1}: FAIL (i915#79) -> PASS
>      * igt@kms_flip@flip-vs-suspend-interruptible@a-dp1:
> 
>           * shard-apl: ABORT (i915#180) -> PASS
>      * igt@kms_frontbuffer_tracking@fbc-1p-primscrn-indfb-plflip-blt:
> 
>           * shard-dg2: FAIL (i915#6880) -> PASS +1 similar issue
>      * igt@kms_plane_scaling@intel-max-src-size@pipe-a-hdmi-a-2:
> 
>           * shard-rkl: FAIL (i915#8292) -> PASS
>      * igt@kms_sysfs_edid_timing:
> 
>           * shard-dg2: FAIL (IGT#2) -> PASS
>      * igt@kms_vblank@pipe-a-ts-continuation-suspend:
> 
>           * shard-dg2: FAIL (fdo#103375 / i915#6121) -> PASS
>      * igt@perf_pmu@busy-double-start@ccs0:
> 
>           * shard-mtlp: FAIL (i915#4349) -> PASS
>      * igt@perf_pmu@most-busy-idle-check-all@bcs0:
> 
>           * shard-mtlp: FAIL (i915#5234) -> PASS +1 similar issue
>      * igt@perf_pmu@most-busy-idle-check-all@rcs0:
> 
>           * shard-dg2: FAIL (i915#5234) -> PASS
> 
>           * {shard-dg1}: FAIL (i915#5234) -> PASS
> 
>      * igt@sysfs_timeslice_duration@timeout@vecs0:
> 
>           * shard-mtlp: ABORT (i915#8521) -> PASS
> 
>     Warnings
> 
>      * igt@gem_exec_suspend@basic-s0@smem:
> 
>           * shard-snb: DMESG-FAIL (fdo#103375) -> DMESG-WARN (i915#8841)
>      * igt@i915_pm_rc6_residency@rc6-idle@bcs0:
> 
>           * shard-tglu: FAIL (i915#2681 / i915#3591) -> WARN (i915#2681)
>      * igt@i915_pm_rc6_residency@rc6-idle@rcs0:
> 
>           * shard-tglu: WARN (i915#2681) -> FAIL (i915#2681 / i915#3591)
>      * igt@i915_suspend@basic-s2idle-without-i915:
> 
>           * shard-snb: DMESG-WARN (i915#8841) -> ABORT (i915#4528 /
>             i915#8213)
>      * igt@kms_content_protection@mei_interface:
> 
>           * shard-dg2: SKIP (i915#7118 / i915#7162) -> SKIP (i915#7118)
> 
>           * shard-rkl: SKIP (fdo#109300) -> SKIP (i915#7118)
> 
>           * shard-tglu: SKIP (fdo#109300) -> SKIP (i915#6944 / i915#7116 /
>             i915#7118)
> 
>      * igt@kms_force_connector_basic@force-load-detect:
> 
>           * shard-rkl: SKIP (fdo#109285 / i915#4098) -> SKIP (fdo#109285)
>      * igt@prime_mmap@test_aperture_limit@test_aperture_limit-smem:
> 
>           * shard-dg2: CRASH (i915#7331) -> INCOMPLETE (i915#5493)
> 
>    {name}: This element is suppressed. This means it is ignored when
>    computing
>    the status of the difference (SUCCESS, WARNING, or FAILURE).
> 
> Build changes
> 
>      * CI: CI-20190529 -> None
>      * IGT: IGT_7380 -> IGTPW_9383
>      * Piglit: piglit_4509 -> None
> 
>    CI-20190529: 20190529
>    CI_DRM_13369: 0595153b5c237d83f9990f3ce44d072a262dcb0e @
>    git://anongit.freedesktop.org/gfx-ci/linux
>    IGTPW_9383: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_9383/index.html
>    IGT_7380: 8e65f12de2fd52c05dc48fdbcb8cfe86f6de1a75 @
>    https://gitlab.freedesktop.org/drm/igt-gpu-tools.git
>    piglit_4509: fdc5a4ca11124ab8413c7988896eec4c97336694 @
>    git://anongit.freedesktop.org/piglit