[PATCH 1/2] drm/i915/selftests: Use a full emulation of a user ppgtt context

[PATCH 1/2] drm/i915/selftests: Use a full emulation of a user ppgtt context

[Chris Wilson]

To test eviction from a ppgtt, we just want a ppgtt i.e. something other
than the Global GTT which is shared and used by the kernel for HW
features like fencing and scanout. However, we also need it to pass
!i915_is_ggtt() and the simplest way is to emulate a full user context
rather than the internal kernel context that is used for the GGTT.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
---
 drivers/gpu/drm/i915/selftests/intel_hangcheck.c | 16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/selftests/intel_hangcheck.c b/drivers/gpu/drm/i915/selftests/intel_hangcheck.c
index 65d66cdedd26..b2d6d15f025a 100644
--- a/drivers/gpu/drm/i915/selftests/intel_hangcheck.c
+++ b/drivers/gpu/drm/i915/selftests/intel_hangcheck.c
@@ -1144,19 +1144,27 @@ static int igt_reset_evict_ppgtt(void *arg)
 {
 	struct drm_i915_private *i915 = arg;
 	struct i915_gem_context *ctx;
+	struct drm_file *file;
 	int err;
 
+	file = mock_file(i915);
+	if (IS_ERR(file))
+		return PTR_ERR(file);
+
 	mutex_lock(&i915->drm.struct_mutex);
-	ctx = kernel_context(i915);
+	ctx = live_context(i915, file);
 	mutex_unlock(&i915->drm.struct_mutex);
-	if (IS_ERR(ctx))
-		return PTR_ERR(ctx);
+	if (IS_ERR(ctx)) {
+		err = PTR_ERR(ctx);
+		goto out;
+	}
 
 	err = 0;
 	if (ctx->ppgtt) /* aliasing == global gtt locking, covered above */
 		err = __igt_reset_evict_vma(i915, &ctx->ppgtt->vm);
 
-	kernel_context_close(ctx);
+out:
+	mock_file_free(i915, file);
 	return err;
 }
 
-- 
2.18.0

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

[PATCH 2/2] drm/i915/selftests: Exercise resetting in the middle of a wait-on-fence

[Chris Wilson]

On older HW, gen2/3, fence registers are used for detiling GPU commands
and as such changing those registers requires serialisation with the
requests on the GPU. Anything running on the GPU is subject to a hang,
and so we must be able to recover cleanly in the middle of a stuck wait
on a fence register.

We can simulate using the fence on the GPU simply by marking the fence
as active on the request for this vma, the interface being common to all
gen, thus broadening the test.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
---
 .../gpu/drm/i915/selftests/intel_hangcheck.c  | 85 +++++++++++++++++--
 1 file changed, 77 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/i915/selftests/intel_hangcheck.c b/drivers/gpu/drm/i915/selftests/intel_hangcheck.c
index b2d6d15f025a..db378226ac10 100644
--- a/drivers/gpu/drm/i915/selftests/intel_hangcheck.c
+++ b/drivers/gpu/drm/i915/selftests/intel_hangcheck.c
@@ -1018,8 +1018,41 @@ static int evict_vma(void *data)
 	return err;
 }
 
+static int evict_fence(void *data)
+{
+	struct evict_vma *arg = data;
+	struct drm_i915_private *i915 = arg->vma->vm->i915;
+	int err;
+
+	complete(&arg->completion);
+
+	mutex_lock(&i915->drm.struct_mutex);
+
+	/* Mark the fence register as dirty to force the mmio update. */
+	err = i915_gem_object_set_tiling(arg->vma->obj, I915_TILING_Y, 512);
+	if (err) {
+		pr_err("Invalid Y-tiling settings; err:%d\n", err);
+		goto out_unlock;
+	}
+
+	err = i915_vma_pin_fence(arg->vma);
+	if (err) {
+		pr_err("Unable to pin Y-tiled fence; err:%d\n", err);
+		goto out_unlock;
+	}
+
+	i915_vma_unpin_fence(arg->vma);
+
+out_unlock:
+	mutex_unlock(&i915->drm.struct_mutex);
+
+	return err;
+}
+
 static int __igt_reset_evict_vma(struct drm_i915_private *i915,
-				 struct i915_address_space *vm)
+				 struct i915_address_space *vm,
+				 int (*fn)(void *),
+				 unsigned int flags)
 {
 	struct drm_i915_gem_object *obj;
 	struct task_struct *tsk = NULL;
@@ -1040,12 +1073,20 @@ static int __igt_reset_evict_vma(struct drm_i915_private *i915,
 	if (err)
 		goto unlock;
 
-	obj = i915_gem_object_create_internal(i915, PAGE_SIZE);
+	obj = i915_gem_object_create_internal(i915, SZ_1M);
 	if (IS_ERR(obj)) {
 		err = PTR_ERR(obj);
 		goto fini;
 	}
 
+	if (flags & EXEC_OBJECT_NEEDS_FENCE) {
+		err = i915_gem_object_set_tiling(obj, I915_TILING_X, 512);
+		if (err) {
+			pr_err("Invalid X-tiling settings; err:%d\n", err);
+			goto out_obj;
+		}
+	}
+
 	arg.vma = i915_vma_instance(obj, vm, NULL);
 	if (IS_ERR(arg.vma)) {
 		err = PTR_ERR(arg.vma);
@@ -1059,11 +1100,28 @@ static int __igt_reset_evict_vma(struct drm_i915_private *i915,
 	}
 
 	err = i915_vma_pin(arg.vma, 0, 0,
-			   i915_vma_is_ggtt(arg.vma) ? PIN_GLOBAL : PIN_USER);
-	if (err)
+			   i915_vma_is_ggtt(arg.vma) ?
+			   PIN_GLOBAL | PIN_MAPPABLE :
+			   PIN_USER);
+	if (err) {
+		i915_request_add(rq);
 		goto out_obj;
+	}
+
+	if (flags & EXEC_OBJECT_NEEDS_FENCE) {
+		err = i915_vma_pin_fence(arg.vma);
+		if (err) {
+			pr_err("Unable to pin X-tiled fence; err:%d\n", err);
+			i915_vma_unpin(arg.vma);
+			i915_request_add(rq);
+			goto out_obj;
+		}
+	}
 
-	err = i915_vma_move_to_active(arg.vma, rq, EXEC_OBJECT_WRITE);
+	err = i915_vma_move_to_active(arg.vma, rq, flags);
+
+	if (flags & EXEC_OBJECT_NEEDS_FENCE)
+		i915_vma_unpin_fence(arg.vma);
 	i915_vma_unpin(arg.vma);
 
 	i915_request_get(rq);
@@ -1086,7 +1144,7 @@ static int __igt_reset_evict_vma(struct drm_i915_private *i915,
 
 	init_completion(&arg.completion);
 
-	tsk = kthread_run(evict_vma, &arg, "igt/evict_vma");
+	tsk = kthread_run(fn, &arg, "igt/evict_vma");
 	if (IS_ERR(tsk)) {
 		err = PTR_ERR(tsk);
 		tsk = NULL;
@@ -1137,7 +1195,8 @@ static int igt_reset_evict_ggtt(void *arg)
 {
 	struct drm_i915_private *i915 = arg;
 
-	return __igt_reset_evict_vma(i915, &i915->ggtt.vm);
+	return __igt_reset_evict_vma(i915, &i915->ggtt.vm,
+				     evict_vma, EXEC_OBJECT_WRITE);
 }
 
 static int igt_reset_evict_ppgtt(void *arg)
@@ -1161,13 +1220,22 @@ static int igt_reset_evict_ppgtt(void *arg)
 
 	err = 0;
 	if (ctx->ppgtt) /* aliasing == global gtt locking, covered above */
-		err = __igt_reset_evict_vma(i915, &ctx->ppgtt->vm);
+		err = __igt_reset_evict_vma(i915, &ctx->ppgtt->vm,
+					    evict_vma, EXEC_OBJECT_WRITE);
 
 out:
 	mock_file_free(i915, file);
 	return err;
 }
 
+static int igt_reset_evict_fence(void *arg)
+{
+	struct drm_i915_private *i915 = arg;
+
+	return __igt_reset_evict_vma(i915, &i915->ggtt.vm,
+				     evict_fence, EXEC_OBJECT_NEEDS_FENCE);
+}
+
 static int wait_for_others(struct drm_i915_private *i915,
 			   struct intel_engine_cs *exclude)
 {
@@ -1417,6 +1485,7 @@ int intel_hangcheck_live_selftests(struct drm_i915_private *i915)
 		SUBTEST(igt_reset_wait),
 		SUBTEST(igt_reset_evict_ggtt),
 		SUBTEST(igt_reset_evict_ppgtt),
+		SUBTEST(igt_reset_evict_fence),
 		SUBTEST(igt_handle_error),
 	};
 	bool saved_hangcheck;
-- 
2.18.0

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

✗ Fi.CI.BAT: failure for series starting with [1/2] drm/i915/selftests: Use a full emulation of a user ppgtt context

[Patchwork]

== Series Details ==

Series: series starting with [1/2] drm/i915/selftests: Use a full emulation of a user ppgtt context
URL   : https://patchwork.freedesktop.org/series/46890/
State : failure

== Summary ==

= CI Bug Log - changes from CI_DRM_4515 -> Patchwork_9721 =

== Summary - FAILURE ==

  Serious unknown changes coming with Patchwork_9721 absolutely need to be
  verified manually.
  
  If you think the reported changes have nothing to do with the changes
  introduced in Patchwork_9721, please notify your bug team to allow them
  to document this new failure mode, which will reduce false positives in CI.

  External URL: https://patchwork.freedesktop.org/api/1.0/series/46890/revisions/1/mbox/

== Possible new issues ==

  Here are the unknown changes that may have been introduced in Patchwork_9721:

  === IGT changes ===

    ==== Possible regressions ====

    igt@drv_selftest@live_hangcheck:
      fi-cfl-guc:         PASS -> DMESG-FAIL

    
    ==== Warnings ====

    igt@drv_selftest@live_execlists:
      {fi-cfl-8109u}:     SKIP -> PASS +1

    
== Known issues ==

  Here are the changes found in Patchwork_9721 that come from known issues:

  === IGT changes ===

    ==== Issues hit ====

    igt@drv_selftest@live_hangcheck:
      fi-skl-guc:         PASS -> DMESG-FAIL (fdo#107174)
      {fi-skl-iommu}:     PASS -> DMESG-FAIL (fdo#106560, fdo#107174)

    
    ==== Possible fixes ====

    igt@drv_selftest@live_hangcheck:
      {fi-cfl-8109u}:     DMESG-FAIL -> PASS

    igt@gem_exec_suspend@basic-s4-devices:
      fi-kbl-7500u:       DMESG-WARN (fdo#105128, fdo#107139) -> PASS

    igt@kms_flip@basic-flip-vs-dpms:
      fi-skl-6700hq:      DMESG-WARN (fdo#105998) -> PASS

    
    ==== Warnings ====

    igt@gem_exec_suspend@basic-s4-devices:
      {fi-kbl-8809g}:     DMESG-WARN (fdo#107139) -> INCOMPLETE (fdo#103665, fdo#107139)

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

  fdo#103665 https://bugs.freedesktop.org/show_bug.cgi?id=103665
  fdo#105128 https://bugs.freedesktop.org/show_bug.cgi?id=105128
  fdo#105998 https://bugs.freedesktop.org/show_bug.cgi?id=105998
  fdo#106560 https://bugs.freedesktop.org/show_bug.cgi?id=106560
  fdo#107139 https://bugs.freedesktop.org/show_bug.cgi?id=107139
  fdo#107174 https://bugs.freedesktop.org/show_bug.cgi?id=107174


== Participating hosts (47 -> 42) ==

  Missing    (5): fi-ctg-p8600 fi-ilk-m540 fi-byt-squawks fi-bsw-cyan fi-hsw-4200u 


== Build changes ==

    * Linux: CI_DRM_4515 -> Patchwork_9721

  CI_DRM_4515: 7d965fa499d11f7547066a827b3ae420f9e26f39 @ git://anongit.freedesktop.org/gfx-ci/linux
  IGT_4568: 86f7b724ef18986bc58d35558d22e1ed3f8df4f9 @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
  Patchwork_9721: dd1dfc0b200245d46cf302f57a196deb5510a926 @ git://anongit.freedesktop.org/gfx-ci/linux


== Linux commits ==

dd1dfc0b2002 drm/i915/selftests: Exercise resetting in the middle of a wait-on-fence
02691ccd8fc9 drm/i915/selftests: Use a full emulation of a user ppgtt context

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_9721/issues.html
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

✓ Fi.CI.BAT: success for series starting with [1/2] drm/i915/selftests: Use a full emulation of a user ppgtt context

[Patchwork]

== Series Details ==

Series: series starting with [1/2] drm/i915/selftests: Use a full emulation of a user ppgtt context
URL   : https://patchwork.freedesktop.org/series/46890/
State : success

== Summary ==

= CI Bug Log - changes from CI_DRM_4516 -> Patchwork_9722 =

== Summary - SUCCESS ==

  No regressions found.

  External URL: https://patchwork.freedesktop.org/api/1.0/series/46890/revisions/1/mbox/

== Known issues ==

  Here are the changes found in Patchwork_9722 that come from known issues:

  === IGT changes ===

    ==== Issues hit ====

    igt@drv_selftest@live_workarounds:
      {fi-cfl-8109u}:     NOTRUN -> DMESG-FAIL (fdo#107292)

    igt@kms_pipe_crc_basic@suspend-read-crc-pipe-c:
      fi-bxt-dsi:         PASS -> INCOMPLETE (fdo#103927)

    
    ==== Possible fixes ====

    igt@debugfs_test@read_all_entries:
      fi-snb-2520m:       INCOMPLETE (fdo#103713) -> PASS

    igt@kms_pipe_crc_basic@suspend-read-crc-pipe-b:
      {fi-cfl-8109u}:     INCOMPLETE (fdo#106070) -> PASS

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

  fdo#103713 https://bugs.freedesktop.org/show_bug.cgi?id=103713
  fdo#103927 https://bugs.freedesktop.org/show_bug.cgi?id=103927
  fdo#106070 https://bugs.freedesktop.org/show_bug.cgi?id=106070
  fdo#107292 https://bugs.freedesktop.org/show_bug.cgi?id=107292


== Participating hosts (47 -> 42) ==

  Missing    (5): fi-ctg-p8600 fi-ilk-m540 fi-byt-squawks fi-bsw-cyan fi-hsw-4200u 


== Build changes ==

    * Linux: CI_DRM_4516 -> Patchwork_9722

  CI_DRM_4516: c8b4bb2d3541b7780a5125d09a21610f3440baef @ git://anongit.freedesktop.org/gfx-ci/linux
  IGT_4568: 86f7b724ef18986bc58d35558d22e1ed3f8df4f9 @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
  Patchwork_9722: 64c38cb15e4096f095021c901d8e0c036d7c52fa @ git://anongit.freedesktop.org/gfx-ci/linux


== Linux commits ==

64c38cb15e40 drm/i915/selftests: Exercise resetting in the middle of a wait-on-fence
4d8e5baa6a03 drm/i915/selftests: Use a full emulation of a user ppgtt context

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_9722/issues.html
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

✓ Fi.CI.IGT: success for series starting with [1/2] drm/i915/selftests: Use a full emulation of a user ppgtt context

[Patchwork]

== Series Details ==

Series: series starting with [1/2] drm/i915/selftests: Use a full emulation of a user ppgtt context
URL   : https://patchwork.freedesktop.org/series/46890/
State : success

== Summary ==

= CI Bug Log - changes from CI_DRM_4515_full -> Patchwork_9721_full =

== Summary - WARNING ==

  Minor unknown changes coming with Patchwork_9721_full need to be verified
  manually.
  
  If you think the reported changes have nothing to do with the changes
  introduced in Patchwork_9721_full, please notify your bug team to allow them
  to document this new failure mode, which will reduce false positives in CI.

  

== Possible new issues ==

  Here are the unknown changes that may have been introduced in Patchwork_9721_full:

  === IGT changes ===

    ==== Warnings ====

    igt@gem_exec_schedule@deep-bsd1:
      shard-kbl:          PASS -> SKIP +1

    
== Known issues ==

  Here are the changes found in Patchwork_9721_full that come from known issues:

  === IGT changes ===

    ==== Issues hit ====

    igt@kms_cursor_legacy@cursor-vs-flip-atomic-transitions-varying-size:
      shard-hsw:          PASS -> FAIL (fdo#103355)

    igt@kms_flip@plain-flip-fb-recreate:
      shard-glk:          PASS -> FAIL (fdo#100368)

    igt@kms_pipe_crc_basic@suspend-read-crc-pipe-b:
      shard-snb:          PASS -> INCOMPLETE (fdo#105411)

    
    ==== Possible fixes ====

    igt@drv_suspend@shrink:
      shard-apl:          FAIL (fdo#106886) -> PASS

    igt@kms_rotation_crc@sprite-rotation-180:
      shard-hsw:          FAIL (fdo#103925) -> PASS

    
  fdo#100368 https://bugs.freedesktop.org/show_bug.cgi?id=100368
  fdo#103355 https://bugs.freedesktop.org/show_bug.cgi?id=103355
  fdo#103925 https://bugs.freedesktop.org/show_bug.cgi?id=103925
  fdo#105411 https://bugs.freedesktop.org/show_bug.cgi?id=105411
  fdo#106886 https://bugs.freedesktop.org/show_bug.cgi?id=106886


== Participating hosts (5 -> 5) ==

  No changes in participating hosts


== Build changes ==

    * Linux: CI_DRM_4515 -> Patchwork_9721

  CI_DRM_4515: 7d965fa499d11f7547066a827b3ae420f9e26f39 @ git://anongit.freedesktop.org/gfx-ci/linux
  IGT_4568: 86f7b724ef18986bc58d35558d22e1ed3f8df4f9 @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
  Patchwork_9721: dd1dfc0b200245d46cf302f57a196deb5510a926 @ git://anongit.freedesktop.org/gfx-ci/linux
  piglit_4509: fdc5a4ca11124ab8413c7988896eec4c97336694 @ git://anongit.freedesktop.org/piglit

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_9721/shards.html
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

✓ Fi.CI.IGT: success for series starting with [1/2] drm/i915/selftests: Use a full emulation of a user ppgtt context

[Patchwork]

== Series Details ==

Series: series starting with [1/2] drm/i915/selftests: Use a full emulation of a user ppgtt context
URL   : https://patchwork.freedesktop.org/series/46890/
State : success

== Summary ==

= CI Bug Log - changes from CI_DRM_4516_full -> Patchwork_9722_full =

== Summary - SUCCESS ==

  No regressions found.

  

== Known issues ==

  Here are the changes found in Patchwork_9722_full that come from known issues:

  === IGT changes ===

    ==== Issues hit ====

    igt@gem_exec_schedule@pi-ringfull-vebox:
      shard-apl:          NOTRUN -> FAIL (fdo#103158)

    igt@kms_flip@2x-dpms-vs-vblank-race-interruptible:
      shard-glk:          PASS -> FAIL (fdo#103060)

    igt@kms_setmode@basic:
      shard-apl:          PASS -> FAIL (fdo#99912)

    
    ==== Possible fixes ====

    igt@drv_selftest@mock_sanitycheck:
      shard-snb:          INCOMPLETE (fdo#105411) -> PASS

    igt@drv_suspend@shrink:
      shard-apl:          INCOMPLETE (fdo#103927, fdo#106886) -> PASS

    igt@kms_flip@flip-vs-expired-vblank:
      shard-glk:          FAIL (fdo#102887, fdo#105363) -> PASS

    igt@kms_flip@wf_vblank-ts-check-interruptible:
      shard-glk:          FAIL (fdo#100368) -> PASS

    igt@kms_rotation_crc@cursor-rotation-180:
      shard-hsw:          FAIL (fdo#103925) -> PASS

    
  fdo#100368 https://bugs.freedesktop.org/show_bug.cgi?id=100368
  fdo#102887 https://bugs.freedesktop.org/show_bug.cgi?id=102887
  fdo#103060 https://bugs.freedesktop.org/show_bug.cgi?id=103060
  fdo#103158 https://bugs.freedesktop.org/show_bug.cgi?id=103158
  fdo#103925 https://bugs.freedesktop.org/show_bug.cgi?id=103925
  fdo#103927 https://bugs.freedesktop.org/show_bug.cgi?id=103927
  fdo#105363 https://bugs.freedesktop.org/show_bug.cgi?id=105363
  fdo#105411 https://bugs.freedesktop.org/show_bug.cgi?id=105411
  fdo#106886 https://bugs.freedesktop.org/show_bug.cgi?id=106886
  fdo#99912 https://bugs.freedesktop.org/show_bug.cgi?id=99912


== Participating hosts (5 -> 5) ==

  No changes in participating hosts


== Build changes ==

    * Linux: CI_DRM_4516 -> Patchwork_9722

  CI_DRM_4516: c8b4bb2d3541b7780a5125d09a21610f3440baef @ git://anongit.freedesktop.org/gfx-ci/linux
  IGT_4568: 86f7b724ef18986bc58d35558d22e1ed3f8df4f9 @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
  Patchwork_9722: 64c38cb15e4096f095021c901d8e0c036d7c52fa @ git://anongit.freedesktop.org/gfx-ci/linux
  piglit_4509: fdc5a4ca11124ab8413c7988896eec4c97336694 @ git://anongit.freedesktop.org/piglit

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_9722/shards.html
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 1/2] drm/i915/selftests: Use a full emulation of a user ppgtt context

[Matthew Auld]

On 19 July 2018 at 20:47, Chris Wilson <chris@chris-wilson.co.uk> wrote:
> To test eviction from a ppgtt, we just want a ppgtt i.e. something other
> than the Global GTT which is shared and used by the kernel for HW
> features like fencing and scanout. However, we also need it to pass
> !i915_is_ggtt() and the simplest way is to emulate a full user context
> rather than the internal kernel context that is used for the GGTT.
>
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 2/2] drm/i915/selftests: Exercise resetting in the middle of a wait-on-fence

[Matthew Auld]

On 19 July 2018 at 20:47, Chris Wilson <chris@chris-wilson.co.uk> wrote:
> On older HW, gen2/3, fence registers are used for detiling GPU commands
> and as such changing those registers requires serialisation with the
> requests on the GPU. Anything running on the GPU is subject to a hang,
> and so we must be able to recover cleanly in the middle of a stuck wait
> on a fence register.
>
> We can simulate using the fence on the GPU simply by marking the fence
> as active on the request for this vma, the interface being common to all
> gen, thus broadening the test.
>
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 2/2] drm/i915/selftests: Exercise resetting in the middle of a wait-on-fence

[Chris Wilson]

Quoting Matthew Auld (2018-07-26 13:26:08)
> On 19 July 2018 at 20:47, Chris Wilson <chris@chris-wilson.co.uk> wrote:
> > On older HW, gen2/3, fence registers are used for detiling GPU commands
> > and as such changing those registers requires serialisation with the
> > requests on the GPU. Anything running on the GPU is subject to a hang,
> > and so we must be able to recover cleanly in the middle of a stuck wait
> > on a fence register.
> >
> > We can simulate using the fence on the GPU simply by marking the fence
> > as active on the request for this vma, the interface being common to all
> > gen, thus broadening the test.
> >
> > Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> > Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
> Reviewed-by: Matthew Auld <matthew.auld@intel.com>

Ta, this test has proven itself quite useful at picking out some of the
nasty deadlocks in trying to make reset BKLless.
-Chris
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx