[Intel-gfx] [PATCH 1/2] drm/i915/execlists: Pull tasklet interrupt-bh local to direct submission

[Intel-gfx] [PATCH 1/2] drm/i915/execlists: Pull tasklet interrupt-bh local to direct submission

[Chris Wilson]

We dropped calling process_csb prior to handling direct submission in
order to avoid the nesting of spinlocks and lift process_csb() and the
majority of the tasklet out of irq-off. However, we do want to avoid
ksoftirqd latency in the fast path, so try and pull the interrupt-bh
local to direct submission if we can acquire the tasklet's lock.

v2: Document the read of pending[0] from outside the tasklet with
READ_ONCE.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
---
 drivers/gpu/drm/i915/gt/intel_lrc.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c
index f88d3b95c4e1..d49baade0986 100644
--- a/drivers/gpu/drm/i915/gt/intel_lrc.c
+++ b/drivers/gpu/drm/i915/gt/intel_lrc.c
@@ -2891,6 +2891,13 @@ static void __submit_queue_imm(struct intel_engine_cs *engine)
 	if (reset_in_progress(execlists))
 		return; /* defer until we restart the engine following reset */
 
+	/* Hopefully we clear execlists->pending[] to let us through */
+	if (READ_ONCE(execlists->pending[0]) &&
+	    tasklet_trylock(&execlists->tasklet)) {
+		process_csb(engine);
+		tasklet_unlock(&execlists->tasklet);
+	}
+
 	__execlists_submission_tasklet(engine);
 }
 
-- 
2.20.1

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

[Intel-gfx] [PATCH 2/2] drm/i915: Immediately execute the fenced work

[Chris Wilson]

If the caller allows and we do not have to wait for any signals,
immediately execute the work within the caller's process. By doing so we
avoid the overhead of scheduling a new task, and the latency in
executing it, at the cost of pulling that work back into the immediate
context. (Sometimes we still prefer to offload the task to another cpu,
especially if we plan on executing many such tasks in parallel for this
client.)

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
---
 .../gpu/drm/i915/gem/i915_gem_execbuffer.c    |  2 +-
 drivers/gpu/drm/i915/i915_sw_fence_work.c     |  5 +++-
 drivers/gpu/drm/i915/i915_sw_fence_work.h     | 23 +++++++++++++++++++
 drivers/gpu/drm/i915/i915_vma.c               |  2 +-
 4 files changed, 29 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
index 6b3013d20851..c643eec4dca0 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
@@ -1822,7 +1822,7 @@ static int eb_parse_pipeline(struct i915_execbuffer *eb,
 	dma_resv_add_excl_fence(shadow->resv, &pw->base.dma);
 	dma_resv_unlock(shadow->resv);
 
-	dma_fence_work_commit(&pw->base);
+	dma_fence_work_commit_imm(&pw->base);
 	return 0;
 
 err_batch_unlock:
diff --git a/drivers/gpu/drm/i915/i915_sw_fence_work.c b/drivers/gpu/drm/i915/i915_sw_fence_work.c
index 997b2998f1f2..a3a81bb8f2c3 100644
--- a/drivers/gpu/drm/i915/i915_sw_fence_work.c
+++ b/drivers/gpu/drm/i915/i915_sw_fence_work.c
@@ -38,7 +38,10 @@ fence_notify(struct i915_sw_fence *fence, enum i915_sw_fence_notify state)
 
 		if (!f->dma.error) {
 			dma_fence_get(&f->dma);
-			queue_work(system_unbound_wq, &f->work);
+			if (test_bit(DMA_FENCE_WORK_IMM, &f->dma.flags))
+				fence_work(&f->work);
+			else
+				queue_work(system_unbound_wq, &f->work);
 		} else {
 			fence_complete(f);
 		}
diff --git a/drivers/gpu/drm/i915/i915_sw_fence_work.h b/drivers/gpu/drm/i915/i915_sw_fence_work.h
index 3a22b287e201..2c409f11c5c5 100644
--- a/drivers/gpu/drm/i915/i915_sw_fence_work.h
+++ b/drivers/gpu/drm/i915/i915_sw_fence_work.h
@@ -32,6 +32,10 @@ struct dma_fence_work {
 	const struct dma_fence_work_ops *ops;
 };
 
+enum {
+	DMA_FENCE_WORK_IMM = DMA_FENCE_FLAG_USER_BITS,
+};
+
 void dma_fence_work_init(struct dma_fence_work *f,
 			 const struct dma_fence_work_ops *ops);
 int dma_fence_work_chain(struct dma_fence_work *f, struct dma_fence *signal);
@@ -41,4 +45,23 @@ static inline void dma_fence_work_commit(struct dma_fence_work *f)
 	i915_sw_fence_commit(&f->chain);
 }
 
+/**
+ * dma_fence_work_commit_imm: Commit the fence, and if possible execute locally.
+ * @f: the fenced worker
+ *
+ * Instead of always scheduling a worker to execute the callback (see
+ * dma_fence_work_commit()), we try to execute the callback immediately in
+ * the local context. It is required that the fence be committed before it
+ * is published, and that no other threads try to tamper with the number
+ * of asynchronous waits on the fence (or else the callback will be
+ * executed in the wrong context, i.e. not the callers).
+ */
+static inline void dma_fence_work_commit_imm(struct dma_fence_work *f)
+{
+	if (atomic_read(&f->chain.pending) <= 1)
+		__set_bit(DMA_FENCE_WORK_IMM, &f->dma.flags);
+
+	dma_fence_work_commit(f);
+}
+
 #endif /* I915_SW_FENCE_WORK_H */
diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c
index 08699fa069aa..191577a98390 100644
--- a/drivers/gpu/drm/i915/i915_vma.c
+++ b/drivers/gpu/drm/i915/i915_vma.c
@@ -980,7 +980,7 @@ int i915_vma_pin(struct i915_vma *vma, u64 size, u64 alignment, u64 flags)
 	mutex_unlock(&vma->vm->mutex);
 err_fence:
 	if (work)
-		dma_fence_work_commit(&work->base);
+		dma_fence_work_commit_imm(&work->base);
 	if (wakeref)
 		intel_runtime_pm_put(&vma->vm->i915->runtime_pm, wakeref);
 err_pages:
-- 
2.20.1

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

[Intel-gfx] ✓ Fi.CI.BAT: success for series starting with [1/2] drm/i915/execlists: Pull tasklet interrupt-bh local to direct submission

[Patchwork]

== Series Details ==

Series: series starting with [1/2] drm/i915/execlists: Pull tasklet interrupt-bh local to direct submission
URL   : https://patchwork.freedesktop.org/series/75068/
State : success

== Summary ==

CI Bug Log - changes from CI_DRM_8186 -> Patchwork_17083
====================================================

Summary
-------

  **SUCCESS**

  No regressions found.

  External URL: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_17083/index.html

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

  Here are the changes found in Patchwork_17083 that come from known issues:

### IGT changes ###

#### Possible fixes ####

  * igt@i915_selftest@live@gem_contexts:
    - fi-cfl-guc:         [DMESG-FAIL][1] ([i915#730] / [i915#933]) -> [PASS][2]
   [1]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_8186/fi-cfl-guc/igt@i915_selftest@live@gem_contexts.html
   [2]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_17083/fi-cfl-guc/igt@i915_selftest@live@gem_contexts.html

  
  [i915#730]: https://gitlab.freedesktop.org/drm/intel/issues/730
  [i915#933]: https://gitlab.freedesktop.org/drm/intel/issues/933


Participating hosts (44 -> 39)
------------------------------

  Additional (4): fi-skl-lmem fi-bwr-2160 fi-skl-6600u fi-snb-2600 
  Missing    (9): fi-bdw-5557u fi-hsw-4200u fi-byt-j1900 fi-byt-squawks fi-bsw-cyan fi-ctg-p8600 fi-elk-e7500 fi-blb-e6850 fi-bdw-samus 


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

  * CI: CI-20190529 -> None
  * Linux: CI_DRM_8186 -> Patchwork_17083

  CI-20190529: 20190529
  CI_DRM_8186: 1abdd8c4027177dd054471cf58a5e9ec3f2df46d @ git://anongit.freedesktop.org/gfx-ci/linux
  IGT_5537: 190245120758e754813d76b2c6c613413a0dba29 @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
  Patchwork_17083: 0eba3cb69652551e1a54e32cad9c830e932d7683 @ git://anongit.freedesktop.org/gfx-ci/linux


== Linux commits ==

0eba3cb69652 drm/i915: Immediately execute the fenced work
7a0b5667f8c9 drm/i915/execlists: Pull tasklet interrupt-bh local to direct submission

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_17083/index.html
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [Intel-gfx] [PATCH 1/2] drm/i915/execlists: Pull tasklet interrupt-bh local to direct submission

[Tvrtko Ursulin]

On 25/03/2020 12:02, Chris Wilson wrote:
> We dropped calling process_csb prior to handling direct submission in
> order to avoid the nesting of spinlocks and lift process_csb() and the
> majority of the tasklet out of irq-off. However, we do want to avoid
> ksoftirqd latency in the fast path, so try and pull the interrupt-bh
> local to direct submission if we can acquire the tasklet's lock.
> 
> v2: Document the read of pending[0] from outside the tasklet with
> READ_ONCE.
> 
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
> ---
>   drivers/gpu/drm/i915/gt/intel_lrc.c | 7 +++++++
>   1 file changed, 7 insertions(+)
> 
> diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c
> index f88d3b95c4e1..d49baade0986 100644
> --- a/drivers/gpu/drm/i915/gt/intel_lrc.c
> +++ b/drivers/gpu/drm/i915/gt/intel_lrc.c
> @@ -2891,6 +2891,13 @@ static void __submit_queue_imm(struct intel_engine_cs *engine)
>   	if (reset_in_progress(execlists))
>   		return; /* defer until we restart the engine following reset */
>   
> +	/* Hopefully we clear execlists->pending[] to let us through */
> +	if (READ_ONCE(execlists->pending[0]) &&
> +	    tasklet_trylock(&execlists->tasklet)) {
> +		process_csb(engine);
> +		tasklet_unlock(&execlists->tasklet);
> +	}
> +
>   	__execlists_submission_tasklet(engine);
>   }
>   
> 

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

Regards,

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

Re: [Intel-gfx] [PATCH 2/2] drm/i915: Immediately execute the fenced work

[Tvrtko Ursulin]

On 25/03/2020 12:02, Chris Wilson wrote:
> If the caller allows and we do not have to wait for any signals,
> immediately execute the work within the caller's process. By doing so we
> avoid the overhead of scheduling a new task, and the latency in
> executing it, at the cost of pulling that work back into the immediate
> context. (Sometimes we still prefer to offload the task to another cpu,
> especially if we plan on executing many such tasks in parallel for this
> client.)
> 
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> ---
>   .../gpu/drm/i915/gem/i915_gem_execbuffer.c    |  2 +-
>   drivers/gpu/drm/i915/i915_sw_fence_work.c     |  5 +++-
>   drivers/gpu/drm/i915/i915_sw_fence_work.h     | 23 +++++++++++++++++++
>   drivers/gpu/drm/i915/i915_vma.c               |  2 +-
>   4 files changed, 29 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
> index 6b3013d20851..c643eec4dca0 100644
> --- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
> +++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
> @@ -1822,7 +1822,7 @@ static int eb_parse_pipeline(struct i915_execbuffer *eb,
>   	dma_resv_add_excl_fence(shadow->resv, &pw->base.dma);
>   	dma_resv_unlock(shadow->resv);
>   
> -	dma_fence_work_commit(&pw->base);
> +	dma_fence_work_commit_imm(&pw->base);
>   	return 0;
>   
>   err_batch_unlock:
> diff --git a/drivers/gpu/drm/i915/i915_sw_fence_work.c b/drivers/gpu/drm/i915/i915_sw_fence_work.c
> index 997b2998f1f2..a3a81bb8f2c3 100644
> --- a/drivers/gpu/drm/i915/i915_sw_fence_work.c
> +++ b/drivers/gpu/drm/i915/i915_sw_fence_work.c
> @@ -38,7 +38,10 @@ fence_notify(struct i915_sw_fence *fence, enum i915_sw_fence_notify state)
>   
>   		if (!f->dma.error) {
>   			dma_fence_get(&f->dma);
> -			queue_work(system_unbound_wq, &f->work);
> +			if (test_bit(DMA_FENCE_WORK_IMM, &f->dma.flags))
> +				fence_work(&f->work);
> +			else
> +				queue_work(system_unbound_wq, &f->work);
>   		} else {
>   			fence_complete(f);
>   		}
> diff --git a/drivers/gpu/drm/i915/i915_sw_fence_work.h b/drivers/gpu/drm/i915/i915_sw_fence_work.h
> index 3a22b287e201..2c409f11c5c5 100644
> --- a/drivers/gpu/drm/i915/i915_sw_fence_work.h
> +++ b/drivers/gpu/drm/i915/i915_sw_fence_work.h
> @@ -32,6 +32,10 @@ struct dma_fence_work {
>   	const struct dma_fence_work_ops *ops;
>   };
>   
> +enum {
> +	DMA_FENCE_WORK_IMM = DMA_FENCE_FLAG_USER_BITS,
> +};
> +
>   void dma_fence_work_init(struct dma_fence_work *f,
>   			 const struct dma_fence_work_ops *ops);
>   int dma_fence_work_chain(struct dma_fence_work *f, struct dma_fence *signal);
> @@ -41,4 +45,23 @@ static inline void dma_fence_work_commit(struct dma_fence_work *f)
>   	i915_sw_fence_commit(&f->chain);
>   }
>   
> +/**
> + * dma_fence_work_commit_imm: Commit the fence, and if possible execute locally.
> + * @f: the fenced worker
> + *
> + * Instead of always scheduling a worker to execute the callback (see
> + * dma_fence_work_commit()), we try to execute the callback immediately in
> + * the local context. It is required that the fence be committed before it
> + * is published, and that no other threads try to tamper with the number
> + * of asynchronous waits on the fence (or else the callback will be
> + * executed in the wrong context, i.e. not the callers).
> + */
> +static inline void dma_fence_work_commit_imm(struct dma_fence_work *f)
> +{
> +	if (atomic_read(&f->chain.pending) <= 1)
> +		__set_bit(DMA_FENCE_WORK_IMM, &f->dma.flags);
> +
> +	dma_fence_work_commit(f);
> +}
> +
>   #endif /* I915_SW_FENCE_WORK_H */
> diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c
> index 08699fa069aa..191577a98390 100644
> --- a/drivers/gpu/drm/i915/i915_vma.c
> +++ b/drivers/gpu/drm/i915/i915_vma.c
> @@ -980,7 +980,7 @@ int i915_vma_pin(struct i915_vma *vma, u64 size, u64 alignment, u64 flags)
>   	mutex_unlock(&vma->vm->mutex);
>   err_fence:
>   	if (work)
> -		dma_fence_work_commit(&work->base);
> +		dma_fence_work_commit_imm(&work->base);
>   	if (wakeref)
>   		intel_runtime_pm_put(&vma->vm->i915->runtime_pm, wakeref);
>   err_pages:
> 

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

Regards,

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

[Intel-gfx] ✓ Fi.CI.BAT: success for series starting with [1/2] drm/i915/execlists: Pull tasklet interrupt-bh local to direct submission

[Patchwork]

== Series Details ==

Series: series starting with [1/2] drm/i915/execlists: Pull tasklet interrupt-bh local to direct submission
URL   : https://patchwork.freedesktop.org/series/75008/
State : success

== Summary ==

CI Bug Log - changes from CI_DRM_8182 -> Patchwork_17065
====================================================

Summary
-------

  **SUCCESS**

  No regressions found.

  External URL: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_17065/index.html

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

  Here are the changes found in Patchwork_17065 that come from known issues:

### IGT changes ###

#### Issues hit ####

  * igt@i915_selftest@live@execlists:
    - fi-cml-u2:          [PASS][1] -> [INCOMPLETE][2] ([i915#283] / [i915#656])
   [1]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_8182/fi-cml-u2/igt@i915_selftest@live@execlists.html
   [2]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_17065/fi-cml-u2/igt@i915_selftest@live@execlists.html

  * igt@i915_selftest@live@hangcheck:
    - fi-icl-guc:         [PASS][3] -> [INCOMPLETE][4] ([fdo#108569])
   [3]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_8182/fi-icl-guc/igt@i915_selftest@live@hangcheck.html
   [4]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_17065/fi-icl-guc/igt@i915_selftest@live@hangcheck.html

  * igt@kms_chamelium@dp-crc-fast:
    - fi-cml-u2:          [PASS][5] -> [FAIL][6] ([i915#217])
   [5]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_8182/fi-cml-u2/igt@kms_chamelium@dp-crc-fast.html
   [6]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_17065/fi-cml-u2/igt@kms_chamelium@dp-crc-fast.html

  * igt@kms_flip@basic-flip-vs-wf_vblank:
    - fi-pnv-d510:        [PASS][7] -> [FAIL][8] ([i915#34])
   [7]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_8182/fi-pnv-d510/igt@kms_flip@basic-flip-vs-wf_vblank.html
   [8]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_17065/fi-pnv-d510/igt@kms_flip@basic-flip-vs-wf_vblank.html

  
#### Possible fixes ####

  * igt@i915_selftest@live@gem_contexts:
    - fi-cfl-8700k:       [DMESG-FAIL][9] ([i915#481]) -> [PASS][10]
   [9]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_8182/fi-cfl-8700k/igt@i915_selftest@live@gem_contexts.html
   [10]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_17065/fi-cfl-8700k/igt@i915_selftest@live@gem_contexts.html
    - fi-cml-s:           [DMESG-FAIL][11] ([i915#877]) -> [PASS][12]
   [11]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_8182/fi-cml-s/igt@i915_selftest@live@gem_contexts.html
   [12]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_17065/fi-cml-s/igt@i915_selftest@live@gem_contexts.html
    - fi-cfl-guc:         [DMESG-FAIL][13] ([i915#481]) -> [PASS][14]
   [13]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_8182/fi-cfl-guc/igt@i915_selftest@live@gem_contexts.html
   [14]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_17065/fi-cfl-guc/igt@i915_selftest@live@gem_contexts.html
    - fi-skl-lmem:        [INCOMPLETE][15] ([i915#424]) -> [PASS][16]
   [15]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_8182/fi-skl-lmem/igt@i915_selftest@live@gem_contexts.html
   [16]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_17065/fi-skl-lmem/igt@i915_selftest@live@gem_contexts.html

  
#### Warnings ####

  * igt@kms_chamelium@vga-edid-read:
    - fi-kbl-7500u:       [SKIP][17] ([fdo#109271]) -> [FAIL][18] ([i915#217])
   [17]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_8182/fi-kbl-7500u/igt@kms_chamelium@vga-edid-read.html
   [18]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_17065/fi-kbl-7500u/igt@kms_chamelium@vga-edid-read.html

  
  [fdo#108569]: https://bugs.freedesktop.org/show_bug.cgi?id=108569
  [fdo#109271]: https://bugs.freedesktop.org/show_bug.cgi?id=109271
  [i915#217]: https://gitlab.freedesktop.org/drm/intel/issues/217
  [i915#283]: https://gitlab.freedesktop.org/drm/intel/issues/283
  [i915#34]: https://gitlab.freedesktop.org/drm/intel/issues/34
  [i915#424]: https://gitlab.freedesktop.org/drm/intel/issues/424
  [i915#481]: https://gitlab.freedesktop.org/drm/intel/issues/481
  [i915#656]: https://gitlab.freedesktop.org/drm/intel/issues/656
  [i915#877]: https://gitlab.freedesktop.org/drm/intel/issues/877


Participating hosts (51 -> 40)
------------------------------

  Missing    (11): fi-ilk-m540 fi-hsw-4200u fi-skl-6770hq fi-glk-dsi fi-byt-squawks fi-bwr-2160 fi-bsw-cyan fi-ctg-p8600 fi-elk-e7500 fi-blb-e6850 fi-byt-clapper 


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

  * CI: CI-20190529 -> None
  * Linux: CI_DRM_8182 -> Patchwork_17065

  CI-20190529: 20190529
  CI_DRM_8182: e5245084567cd7f6f93b07baaebf8a2b4d914620 @ git://anongit.freedesktop.org/gfx-ci/linux
  IGT_5531: 79e7382202c104b247a672c61a6186d1f51e4958 @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
  Patchwork_17065: 6cfb39892bf9c0b2cfa57708bb6ae3276f9179fa @ git://anongit.freedesktop.org/gfx-ci/linux


== Linux commits ==

6cfb39892bf9 drm/i915: Immediately execute the fenced work
1b776cbb8e68 drm/i915/execlists: Pull tasklet interrupt-bh local to direct submission

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_17065/index.html
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx