On 20/04/11 11:50, Lionel Landwerlin wrote:
> On 10/04/2020 19:51, Venkata Sandeep Dhanalakota wrote:
> > From: Lionel Landwerlin
> >
> > To allow faster engine to engine synchronization, peel the layer of
> > dma-fence-chain to expose potential i915 fences so that th
simple tests using drm api for timeline semaphore.
Signed-off-by: Venkata Sandeep Dhanalakota
---
drivers/gpu/drm/selftests/Makefile| 2 +-
.../drm/selftests/drm_timeline_selftests.h| 16 +
.../selftests/test-drm_timeline_semaphore.c | 545 ++
3 files
From: Lionel Landwerlin
To allow faster engine to engine synchronization, peel the layer of
dma-fence-chain to expose potential i915 fences so that the
i915-request code can emit HW semaphore wait/signal operations in the
ring which is faster than waking up the host to submit unblocked
workloads
://lists.freedesktop.org/archives/dri-devel/2019-August/229287.html
drop the ordering check from v8 (Lionel)
v10: Set first extension enum item to 1 (Jason)
v11: Add wait on previous sync points in timelines (Sandeep)
Signed-off-by: Lionel Landwerlin
Signed-off-by: Venkata Sandeep Dhanalakota
---
.../gpu/drm
From: Lionel Landwerlin
We're planning to use this for a couple of new feature where we need
to provide additional parameters to execbuf.
v2: Check for invalid flags in execbuffer2 (Lionel)
v3: Rename I915_EXEC_EXT -> I915_EXEC_USE_EXTENSIONS (Chris)
Signed-off-by: Lionel Landwerlin
Reviewed-
id *arg)
> +{
> + struct fence_chains fc;
> + struct task_struct *tsk;
> + int err;
> + int i;
> +
> + err = fence_chains_init(&fc, 64 << 10, seqno_inc);
> + if (err)
> + return err;
> +
>
*pfence = NULL;
> + break;
> + }
> +
Looks good to me.
Tested-by: Venkata Sandeep Dhanalakota
Reviewed-by: Venkata Sandeep Dhanalakota
> if ((*pfence)->context != chain->base.context ||
> to_
On 20/04/08 07:29, Lionel Landwerlin wrote:
> On 06/04/2020 23:07, Venkata Sandeep Dhanalakota wrote:
> > Introduces a new parameters to execbuf so that we can specify syncobj
> > handles as well as timeline points.
> >
> > v2: Reuse i915_user_extension_fn
> >
://lists.freedesktop.org/archives/dri-devel/2019-August/229287.html
drop the ordering check from v8 (Lionel)
v10: Set first extension enum item to 1 (Jason)
v11: Add wait on previous sync points in timelines (Sandeep)
Signed-off-by: Lionel Landwerlin
Signed-off-by: Venkata Sandeep Dhanalakota
---
.../gpu/drm
From: Lionel Landwerlin
We're planning to use this for a couple of new feature where we need
to provide additional parameters to execbuf.
v2: Check for invalid flags in execbuffer2 (Lionel)
v3: Rename I915_EXEC_EXT -> I915_EXEC_USE_EXTENSIONS (Chris)
Signed-off-by: Lionel Landwerlin
Reviewed-
From: Lionel Landwerlin
To allow faster engine to engine synchronization, peel the layer of
dma-fence-chain to expose potential i915 fences so that the
i915-request code can emit HW semaphore wait/signal operations in the
ring which is faster than waking up the host to submit unblocked
workloads
Cc: Jani Nikula
Signed-off-by: Venkata Sandeep Dhanalakota
---
drivers/gpu/drm/i915/gt/intel_gt.h | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/gpu/drm/i915/gt/intel_gt.h
b/drivers/gpu/drm/i915/gt/intel_gt.h
index 9d9e8831daeb..2355cf129e9c 100644
--- a/drive
: Rodrigo Vivi
Cc: Daniel Vetter
Cc: Chris Wilson
Cc: Jani Nikula
Reviewed-by: Chris Wilson
Reviewed-by: Lucas De Marchi
Signed-off-by: Venkata Sandeep Dhanalakota
---
drivers/gpu/drm/i915/i915_pci.c| 9 -
drivers/gpu/drm/i915/i915_perf.c | 17 +
drivers
: Chris Wilson
Signed-off-by: Chris Wilson
Signed-off-by: Venkata Sandeep Dhanalakota
---
drivers/gpu/drm/i915/gem/i915_gem_pm.c| 4 +-
drivers/gpu/drm/i915/gt/intel_context.c | 11 +-
drivers/gpu/drm/i915/gt/intel_context.h | 7 ++
drivers/gpu/drm/i915/gt/intel_engine.h
: Venkata Sandeep Dhanalakota
---
drivers/gpu/drm/i915/gem/i915_gem_pm.c| 4 +-
drivers/gpu/drm/i915/gt/intel_context.c | 9 +-
drivers/gpu/drm/i915/gt/intel_context.h | 7 ++
drivers/gpu/drm/i915/gt/intel_engine.h| 8 ++
drivers/gpu/drm/i915/gt/intel_engine_cs.c
: Venkata Sandeep Dhanalakota
---
drivers/gpu/drm/i915/i915_pci.c| 9 -
drivers/gpu/drm/i915/i915_perf.c | 18 ++
drivers/gpu/drm/i915/i915_perf.h | 2 ++
drivers/gpu/drm/i915/i915_perf_types.h | 1 -
4 files changed, 24 insertions(+), 6 deletions(-)
diff
: Venkata Sandeep Dhanalakota
---
drivers/gpu/drm/i915/i915_pci.c| 9 -
drivers/gpu/drm/i915/i915_perf.c | 18 ++
drivers/gpu/drm/i915/i915_perf.h | 2 ++
drivers/gpu/drm/i915/i915_perf_types.h | 1 -
4 files changed, 24 insertions(+), 6 deletions(-)
diff
New macros ENGINE_TRACE(), CE_TRACE() and RQ_TRACE()
are introduce to tag device name and engine name with
contexts and requests tracing in i915.
Cc: Sudeep Dutt
Cc: Rodrigo Vivi
Cc: Daniel Vetter
Cc: Chris Wilson
Cc: Jani Nikula
Signed-off-by: Chris Wilson
Signed-off-by: Venkata Sandeep
On 19/12/12 01:34, Lucas De Marchi wrote:
> On Wed, Dec 11, 2019 at 11:35:21PM -0800, Venkata Sandeep Dhanalakota wrote:
> > We do not require to register the sysctl paths per instance,
> > so making registration global.
> >
> > v2: make sysctl path register and
t; {
> - if (ips_mchdev == rps_to_i915(rps))
> + if (rcu_access_pointer(ips_mchdev) == rps_to_i915(rps))
> rcu_assign_pointer(ips_mchdev, NULL);
> }
Right way to access rcu pointer.
Reviewed-by: Venkata Sandeep Dhanalakota
>
> --
> 2.24.0
>
&g
On 19/12/11 10:50, Chris Wilson wrote:
> drivers/gpu/drm/i915/gt/intel_rps.c:1726:24: error: incompatible types in
> comparison expression (different address spaces):
> drivers/gpu/drm/i915/gt/intel_rps.c:1726:24:struct drm_i915_private
> [noderef] *
> drivers/gpu/drm/i915/gt/intel_rps.c:172
: Venkata Sandeep Dhanalakota
---
drivers/gpu/drm/i915/i915_pci.c| 6 ++
drivers/gpu/drm/i915/i915_perf.c | 19 ---
drivers/gpu/drm/i915/i915_perf.h | 2 ++
drivers/gpu/drm/i915/i915_perf_types.h | 1 -
4 files changed, 24 insertions(+), 4 deletions(-)
diff
Adding device name to trace makes debugging easier,
when dealing with multiple gpus.
Cc: Sudeep Dutt
Cc: Rodrigo Vivi
Cc: Daniel Vetter
Cc: Chris Wilson
Cc: Jani Nikula
Signed-off-by: Venkata Sandeep Dhanalakota
---
drivers/gpu/drm/i915/gem/i915_gem_pm.c| 4 +-
drivers/gpu/drm
On 19/12/11 04:39, Tvrtko Ursulin wrote:
>
> On 11/12/2019 16:31, Tvrtko Ursulin wrote:
> > On 11/12/2019 16:07, Venkata Sandeep Dhanalakota wrote:
> > > We do not require to register the sysctl paths per instance,
> > > so making registration global.
> >
On 19/12/11 04:07, Chris Wilson wrote:
> Quoting Venkata Sandeep Dhanalakota (2019-12-11 15:59:09)
> > On 19/12/11 03:02, Chris Wilson wrote:
> > > Use the dev_name(i915) to identify the requests for debugging, so we can
> > > tell different device timelines apart.
Make each instance to report the hang only once.
Cc: Sudeep Dutt
Cc: Rodrigo Vivi
Cc: Daniel Vetter
Cc: Chris Wilson
Cc: Jani Nikula
Signed-off-by: Venkata Sandeep Dhanalakota
---
drivers/gpu/drm/i915/i915_drv.h | 2 ++
drivers/gpu/drm/i915/i915_gpu_error.c | 3 +--
2 files changed
We do not require to register the sysctl paths per instance,
so making registration global.
Cc: Sudeep Dutt
Cc: Rodrigo Vivi
Cc: Daniel Vetter
Cc: Chris Wilson
Cc: Jani Nikula
Signed-off-by: Venkata Sandeep Dhanalakota
---
drivers/gpu/drm/i915/i915_perf.c | 10 --
drivers/gpu
On 19/12/11 03:02, Chris Wilson wrote:
> Use the dev_name(i915) to identify the requests for debugging, so we can
> tell different device timelines apart.
>
> Signed-off-by: Chris Wilson
> Cc: Venkata Sandeep Dhanalakota
> ---
> drivers/gpu/drm/i915/i915_request.c | 2 +-
On 19/12/10 03:36, Chris Wilson wrote:
> Check that we own the global pointer before deregistering.
>
> Reported-by: Venkata Sandeep Dhanalakota
> Signed-off-by: Chris Wilson
> Cc: Venkata Sandeep Dhanalakota
> Cc: Andi Shyti
> ---
> drivers/gpu/drm/i915/gt/intel_
29 matches
Mail list logo
