[PATCH 1/5] drm/amdgpu: define ras_reset_error_count function
Tao Zhou
tao.zhou1 at amd.com
Thu Oct 12 09:00:30 UTC 2023
Make the code architecture more simple.
Signed-off-by: Tao Zhou <tao.zhou1 at amd.com>
---
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 8 ++------
drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 17 +++++++++++++++++
drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h | 2 ++
drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c | 4 ++--
4 files changed, 23 insertions(+), 8 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 31f8c3ead161..04cfd67a37a0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -3594,9 +3594,7 @@ static void amdgpu_device_xgmi_reset_func(struct work_struct *__work)
if (adev->asic_reset_res)
goto fail;
- if (adev->mmhub.ras && adev->mmhub.ras->ras_block.hw_ops &&
- adev->mmhub.ras->ras_block.hw_ops->reset_ras_error_count)
- adev->mmhub.ras->ras_block.hw_ops->reset_ras_error_count(adev);
+ amdgpu_ras_reset_error_count(adev, AMDGPU_RAS_BLOCK__MMHUB);
} else {
task_barrier_full(&hive->tb);
@@ -5242,9 +5240,7 @@ int amdgpu_do_asic_reset(struct list_head *device_list_handle,
if (!r && amdgpu_ras_intr_triggered()) {
list_for_each_entry(tmp_adev, device_list_handle, reset_list) {
- if (tmp_adev->mmhub.ras && tmp_adev->mmhub.ras->ras_block.hw_ops &&
- tmp_adev->mmhub.ras->ras_block.hw_ops->reset_ras_error_count)
- tmp_adev->mmhub.ras->ras_block.hw_ops->reset_ras_error_count(tmp_adev);
+ amdgpu_ras_reset_error_count(tmp_adev, AMDGPU_RAS_BLOCK__MMHUB);
}
amdgpu_ras_intr_cleared();
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index dacce5f2bfa3..82ee6f3d12da 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -1101,6 +1101,23 @@ int amdgpu_ras_query_error_status(struct amdgpu_device *adev,
return 0;
}
+int amdgpu_ras_reset_error_count(struct amdgpu_device *adev,
+ enum amdgpu_ras_block block)
+{
+ struct amdgpu_ras_block_object *block_obj = amdgpu_ras_get_ras_block(adev, block, 0);
+
+ if (!block_obj || !block_obj->hw_ops)
+ return 0;
+
+ if (!amdgpu_ras_is_supported(adev, block))
+ return 0;
+
+ if (block_obj->hw_ops->reset_ras_error_count)
+ block_obj->hw_ops->reset_ras_error_count(adev);
+
+ return 0;
+}
+
int amdgpu_ras_reset_error_status(struct amdgpu_device *adev,
enum amdgpu_ras_block block)
{
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
index 728f98c6fc1c..deea10b6c184 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
@@ -694,6 +694,8 @@ void amdgpu_ras_debugfs_create_all(struct amdgpu_device *adev);
int amdgpu_ras_query_error_status(struct amdgpu_device *adev,
struct ras_query_if *info);
+int amdgpu_ras_reset_error_count(struct amdgpu_device *adev,
+ enum amdgpu_ras_block block);
int amdgpu_ras_reset_error_status(struct amdgpu_device *adev,
enum amdgpu_ras_block block);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
index 70e38b013309..2b7dc490ba6b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
@@ -914,7 +914,7 @@ static int amdgpu_xgmi_ras_late_init(struct amdgpu_device *adev, struct ras_comm
adev->gmc.xgmi.num_physical_nodes == 0)
return 0;
- adev->gmc.xgmi.ras->ras_block.hw_ops->reset_ras_error_count(adev);
+ amdgpu_ras_reset_error_count(adev, AMDGPU_RAS_BLOCK__XGMI_WAFL);
return amdgpu_ras_block_late_init(adev, ras_block);
}
@@ -1081,7 +1081,7 @@ static void amdgpu_xgmi_query_ras_error_count(struct amdgpu_device *adev,
break;
}
- adev->gmc.xgmi.ras->ras_block.hw_ops->reset_ras_error_count(adev);
+ amdgpu_ras_reset_error_count(adev, AMDGPU_RAS_BLOCK__XGMI_WAFL);
err_data->ue_count += ue_cnt;
err_data->ce_count += ce_cnt;
--
2.35.1
More information about the amd-gfx
mailing list