since gpu_scheduler source domain cannot access amdgpu variable
so need create the hang_limit membewr for sched, and it can
refer it for the upcoming GPU RESET patches
v2:
make hang_limit a parameter of sched_init()
Change-Id: I977ae2717e55a8b87c59e58a288bffc3b458b653
Signed-off-by: Monk Liu
Change-Id: I44019f6475b1eaaba55633cf5f8bb84284f19a2c
Signed-off-by: Monk Liu
Reviewed-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
this member will be used later, it will points to
the real var inside of context and CS_SUBMIT & gpu schdduler
can decide if skip a job depends on context->guilty or *entity->guilty
Change-Id: I411e117a01c54286db0765fd2f6bf9d3bda01a3b
Signed-off-by: Monk Liu
---
merge the setting guilty on context into this function
to avoid implement extra routine.
v2:
go through entity list and compare the fence_ctx
before operate on the entity, otherwise the entity
may be just a wild pointer
Change-Id: I7a0063464fdc85d5ac9080046380e745565ff540
Signed-off-by: Monk Liu
this way the real interested guilty is connected to entity->guilty
pointer, and we can use entity->pointer later in gpu recovery procedure
Change-Id: I09b392ebfc59254795e4fbd5816abd3d94a95853
Signed-off-by: Monk Liu
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h | 1 +
drivers/gpu/drm/amd/amdgpu//psp_v3_1.c:389:13:
warning: format ‘%x’ expects argument of type ‘unsigned int’, but
argument 4 has type ‘struct psp_gfx_rb_frame *’ [-Wformat=]
DRM_ERROR("ring_buffer_start = %x; ring_buffer_end = %x; write_frame
= %x\n",
^
./include/drm/drmP.h:178:36:
Sorry, this patch is invalide, please ignore this one, will send out v2 for
review
-Original Message-
From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf Of Monk
Liu
Sent: 2017年10月23日 13:34
To: amd-gfx@lists.freedesktop.org
Cc: Liu, Monk
Subject:
this way the real interested guilty is connected to entity->guilty
pointer, and we can use entity->pointer later in gpu recovery procedure
Change-Id: I09b392ebfc59254795e4fbd5816abd3d94a95853
Signed-off-by: Monk Liu
---
drivers/gpu/drm/amd/amdgpu/amdgpu.h | 1 +
this member will be used later, it will points to
the real var inside of context and CS_SUBMIT & gpu schdduler
can decide if skip a job depends on context->guilty or *entity->guilty
Change-Id: I411e117a01c54286db0765fd2f6bf9d3bda01a3b
Signed-off-by: Monk Liu
---
Change-Id: I44019f6475b1eaaba55633cf5f8bb84284f19a2c
Signed-off-by: Monk Liu
Reviewed-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
merge the setting guilty on context into this function
to avoid implement extra routine.
Change-Id: I7a0063464fdc85d5ac9080046380e745565ff540
Signed-off-by: Monk Liu
Reviewed-by: Christian König
---
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c| 4
since gpu_scheduler source domain cannot access amdgpu variable
so need create the hang_limit membewr for sched, and it can
refer it for the upcoming GPU RESET patches
v2:
make hang_limit a parameter of sched_init()
Change-Id: I977ae2717e55a8b87c59e58a288bffc3b458b653
Signed-off-by: Monk Liu
If the deadlock issue could be solved I don't see why we give up kfifo and
switch to SPSC ..
-Original Message-
From: amd-gfx [mailto:amd-gfx-boun...@lists.freedesktop.org] On Behalf Of
Andrey Grodzovsky
Sent: 2017年10月20日 21:32
To: amd-gfx@lists.freedesktop.org
Cc: Grodzovsky,
Why not use a more simple way ?
Like moving ttm_eu_fence_buffer_objects() to before amd_sched_entity_push_job()
?
That could solve the deadlock from your description
And the push order is already guaranteed by context->mutex (which is also a
patch from you)
BR Monk
-Original
Change-Id: I6d06b81b8b894775e55e495e96f3c999b83cf2be
Signed-off-by: Monk Liu
---
drivers/gpu/drm/ttm/ttm_bo.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
index 97125cb..3c9c0d9 100644
---
Am 22.10.2017 23:48, schrieb Dieter Nützel:
Am 21.10.2017 23:22, schrieb Alex Deucher:
Hi Dave,
Last batch of new stuff for DC. Highlights:
- Fix some memory leaks
- S3 fixes
- Hotplug fixes
- Fix some CX multi-display issues
- MST fixes
- DML updates from the hw team
- Various code cleanups
-
Am 17.10.2017 16:59, schrieb Michel Dänzer:
On 17/10/17 04:53 PM, Harry Wentland wrote:
On 2017-10-17 10:47 AM, Michel Dänzer wrote:
On 13/10/17 09:22 PM, Harry Wentland wrote:
On 2017-10-12 08:22 PM, Dieter Nützel wrote:
next (regression) compilation error:
Hi Harsha,
[auto build test ERROR on v4.14-rc3]
[cannot apply to drm/drm-next next-20171018]
[if your patch is applied to the wrong git tree, please drop us a note to help
improve the system]
url:
18 matches
Mail list logo