drm/amdgpu: move the VRAM lost counter per context

Instead of per device track the VRAM lost per context and return ECANCELED
instead of ENODEV.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This commit is contained in:
Christian König
2017-10-09 15:18:43 +02:00
committed by Alex Deucher
parent 14e47f93c5
commit e55f2b646d
4 changed files with 8 additions and 24 deletions

View File

@@ -75,6 +75,7 @@ static int amdgpu_ctx_init(struct amdgpu_device *adev,
}
ctx->reset_counter = atomic_read(&adev->gpu_reset_counter);
ctx->vram_lost_counter = atomic_read(&adev->vram_lost_counter);
ctx->init_priority = priority;
ctx->override_priority = AMD_SCHED_PRIORITY_UNSET;