[PATCH 1/3] drm/amdgpu: add variable to record gpu reset reason
YiPeng Chai
YiPeng.Chai at amd.com
Thu Apr 27 08:02:17 UTC 2023
Add variable to record gpu reset reason.
Signed-off-by: YiPeng Chai <YiPeng.Chai at amd.com>
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h | 3 +++
drivers/gpu/drm/amd/amdgpu/gfx_v11_0_3.c | 6 +++++-
2 files changed, 8 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
index ef38f4c93df0..dd9e1ff360c5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
@@ -385,6 +385,9 @@ struct amdgpu_ras {
/* Indicates smu whether need update bad channel info */
bool update_channel_flag;
+
+ /* gpu reset caused by gfx poison */
+ bool reset_by_gfx_poison;
};
struct ras_fs_data {
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0_3.c
index 068b9586a223..e2faaf6b0a59 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0_3.c
@@ -84,8 +84,12 @@ static int gfx_v11_0_3_poison_consumption_handler(struct amdgpu_device *adev,
/* Workaround: when vmid and pasid are both zero, trigger gpu reset in KGD. */
if (entry && (entry->client_id == SOC21_IH_CLIENTID_GFX) &&
(entry->src_id == GFX_11_0_0__SRCID__RLC_GC_FED_INTERRUPT) &&
- !entry->vmid && !entry->pasid)
+ !entry->vmid && !entry->pasid) {
+ struct amdgpu_ras *ras = amdgpu_ras_get_context(adev);
+
+ ras->reset_by_gfx_poison = true;
amdgpu_ras_reset_gpu(adev);
+ }
return 0;
}
--
2.34.1
More information about the amd-gfx
mailing list