drm/amdgpu: support reset flag set for gpu reset

Move reset_context out of gpu recover function to make it configurable
for different reset purpose.
For the reset way of call gpu_recovery sysfs, force to use full reset
method. Otherwise, try soft reset by default if the related ASIC
supportted, if soft reset failed, will use full reset.

Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This commit is contained in:
Likun Gao
2022-07-08 11:14:05 +08:00
committed by Alex Deucher
parent 58e969b60d
commit f1549c09c5
9 changed files with 74 additions and 27 deletions

View File

@@ -29,6 +29,7 @@
#include "amdgpu.h"
#include "amdgpu_trace.h"
#include "amdgpu_reset.h"
static enum drm_gpu_sched_stat amdgpu_job_timedout(struct drm_sched_job *s_job)
{
@@ -64,7 +65,14 @@ static enum drm_gpu_sched_stat amdgpu_job_timedout(struct drm_sched_job *s_job)
ti.process_name, ti.tgid, ti.task_name, ti.pid);
if (amdgpu_device_should_recover_gpu(ring->adev)) {
r = amdgpu_device_gpu_recover(ring->adev, job);
struct amdgpu_reset_context reset_context;
memset(&reset_context, 0, sizeof(reset_context));
reset_context.method = AMD_RESET_METHOD_NONE;
reset_context.reset_req_dev = adev;
clear_bit(AMDGPU_NEED_FULL_RESET, &reset_context.flags);
r = amdgpu_device_gpu_recover(ring->adev, job, &reset_context);
if (r)
DRM_ERROR("GPU Recovery Failed: %d\n", r);
} else {