[PATCH 1/2] drm/amd/amdgpu: customized the reset to skip soft recovery

[Feng, Kenneth]

[AMD Official Use Only - General]

+ at Zhang, GuoQing (Sam)

-----Original Message-----
From: Kenneth Feng <kenneth.feng at amd.com>
Sent: Monday, April 29, 2024 3:32 PM
To: amd-gfx at lists.freedesktop.org
Cc: Zhang, Owen(SRDC) <Owen.Zhang2 at amd.com>; Aldabagh, Maad <Maad.Aldabagh at amd.com>; Ma, Qing (Mark) <Qing.Ma at amd.com>; Feng, Kenneth <Kenneth.Feng at amd.com>
Subject: [PATCH 1/2] drm/amd/amdgpu: customized the reset to skip soft recovery

customized the reset to skip soft recovery

Signed-off-by: Kenneth Feng <kenneth.feng at amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 2 +-  drivers/gpu/drm/amd/amdgpu/amdgpu_job.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index ea14f1c8f430..65c3a387fec7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -548,7 +548,7 @@ module_param_named(compute_multipipe, amdgpu_compute_multipipe, int, 0444);
  * DOC: gpu_recovery (int)
  * Set to enable GPU recovery mechanism (1 = enable, 0 = disable). The default is -1 (auto, disabled except SRIOV).
  */
-MODULE_PARM_DESC(gpu_recovery, "Enable GPU recovery mechanism, (1 = enable, 0 = disable, -1 = auto)");
+MODULE_PARM_DESC(gpu_recovery, "Enable GPU recovery mechanism, (1 =
+enable, 0 = disable, -1 = auto, 2 = customized)");
 module_param_named(gpu_recovery, amdgpu_gpu_recovery, int, 0444);

 /**
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
index e4742b65032d..5c75993698a5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
@@ -51,7 +51,7 @@ static enum drm_gpu_sched_stat amdgpu_job_timedout(struct drm_sched_job *s_job)

        adev->job_hang = true;

-       if (amdgpu_gpu_recovery &&
+       if (amdgpu_gpu_recovery && amdgpu_gpu_recovery != 2 &&
            amdgpu_ring_soft_recovery(ring, job->vmid, s_job->s_fence->parent)) {
                DRM_ERROR("ring %s timeout, but soft recovered\n",
                          s_job->sched->name);
--
2.34.1