linux.git/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c, branch v4.18-rc2

drm/amdgpu/vg20:Restruct uvd.inst to support multiple instances

2018-05-18T21:08:12+00:00

Vega20 has dual-UVD. Need add multiple instances support for uvd.
Restruct uvd.inst, using uvd.inst[0] to replace uvd.inst->.
Repurpose amdgpu_ring::me for instance index, and initialize to 0.
There are no any logical changes here.

Signed-off-by: James Zhu 
Reviewed-by: Alex Deucher 
Signed-off-by: Alex Deucher

drm/amdgpu/vg20:Restruct uvd to support multiple uvds

2018-05-18T21:08:12+00:00

Vega20 has dual-UVD. Need Restruct amdgpu_device::uvd to support
multiple uvds. There are no any logical changes here.

Signed-off-by: James Zhu 
Reviewed-by: Alex Deucher 
Signed-off-by: Alex Deucher

drm/amdgpu: optionally do a writeback but don't invalidate TC for IB fences

2018-05-15T18:43:32+00:00

There is a new IB flag that enables this new behavior.
Full invalidation is unnecessary for RELEASE_MEM and doesn't make sense
when draw calls from two adjacent gfx IBs run in parallel. This will be
the new default for Mesa.

v2: bump the version

Signed-off-by: Marek Olšák 
Reviewed-by: Christian König 
Signed-off-by: Alex Deucher

drm/amdgpu: drop compute ring timeout setting for non-sriov only (v2)

2018-04-03T17:52:56+00:00

Sriov still wants these error messags on timeout. So, for sriov
use case, the timeout setting on compute rings is kept.

-v2: clean the code

Signed-off-by: Evan Quan 
Reviewed-by: Christian König 
Reviewed-by: Monk Liu 
Signed-off-by: Alex Deucher

drm/amdgpu: no job timeout setting on compute queues

2018-03-21T19:36:57+00:00

Under some heavy computing environment(e.g. dgemm test), it
takes the asic over 10+ seconds to finish the dispatched job
which will trigger the timeout.

It's quite confusing although it does not seem to bring any
real problems. As a quick workround, we choose to not enfoce
the timeout setting on compute queues.

Signed-off-by: Evan Quan 
Acked-by: Alex Deucher 
Reviewed-by: Christian König 
Signed-off-by: Alex Deucher

drm/amdgpu: rename amdgpu_gpu_recover

2017-12-18T15:59:58+00:00

add device to the name for consistency.

Acked-by: Christian König 
Signed-off-by: Alex Deucher

drm/amdgpu: Simplify amdgpu_lockup_timeout usage.

2017-12-15T22:15:00+00:00

With introduction of amdgpu_gpu_recovery we don't need any more
to rely on amdgpu_lockup_timeout == 0 for disabling GPU reset.

Signed-off-by: Andrey Grodzovsky 
Reviewed-by: Christian König 
Signed-off-by: Alex Deucher

drm/amdgpu: Add gpu_recovery parameter

2017-12-15T22:14:50+00:00

Add new parameter to control GPU recovery procedure.

v2:
Add auto logic where reset is disabled for bare metal and enabled
for SR-IOV.
Allow forced reset from debugfs.

Signed-off-by: Andrey Grodzovsky 
Reviewed-by: Christian König 
Signed-off-by: Alex Deucher

drm/amdgpu: no need with INT for fence polling

2017-12-12T19:50:00+00:00

We are polling so no need for INT.

Signed-off-by: Monk Liu 
Reviewed-by: Christian König 
Signed-off-by: Alex Deucher

drm: move amd_gpu_scheduler into common location

2017-12-07T16:51:56+00:00

This moves and renames the AMDGPU scheduler to a common location in DRM
in order to facilitate re-use by other drivers. This is mostly a straight
forward rename with no code changes.

One notable exception is the function to_drm_sched_fence(), which is no
longer a inline header function to avoid the need to export the
drm_sched_fence_ops_scheduled and drm_sched_fence_ops_finished structures.

Reviewed-by: Chunming Zhou 
Tested-by: Dieter Nützel 
Acked-by: Alex Deucher 
Signed-off-by: Lucas Stach 
Signed-off-by: Alex Deucher