On 10/25/2024 12:49 PM, Yang Wang wrote:
> - skip to print CE ACA log.
> - optimize ACA log print for MCA.
>
> Signed-off-by: Yang Wang <kevinyang.wang@xxxxxxx>
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c | 21 ++++++++++++++++++++-
> 1 file changed, 20 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c
> index 18ee60378727..3ca03b5e0f91 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c
> @@ -348,6 +348,24 @@ static bool amdgpu_mca_bank_should_update(struct amdgpu_device *adev, enum amdgp
> return ret;
> }
>
> +static bool amdgpu_mca_bank_should_dump(struct amdgpu_device *adev, enum amdgpu_mca_error_type type,
> + struct mca_bank_entry *entry)
> +{
> + bool ret;
> +
> + switch (type) {
> + case AMDGPU_MCA_ERROR_TYPE_CE:
> + ret = amdgpu_mca_is_deferred_error(adev, entry->regs[MCA_REG_IDX_STATUS]);
AFAIK, deferred errors are not correctable. Shouldn't it be checked
against AMDGPU_MCA_ERROR_TYPE_DE?
Thanks,
Lijo
> + break;
> + case AMDGPU_MCA_ERROR_TYPE_UE:
> + default:
> + ret = true;
> + break;
> + }
> +
> + return ret;
> +}
> +
> static int amdgpu_mca_smu_get_mca_set(struct amdgpu_device *adev, enum amdgpu_mca_error_type type, struct mca_bank_set *mca_set,
> struct ras_query_context *qctx)
> {
> @@ -373,7 +391,8 @@ static int amdgpu_mca_smu_get_mca_set(struct amdgpu_device *adev, enum amdgpu_mc
>
> amdgpu_mca_bank_set_add_entry(mca_set, &entry);
>
> - amdgpu_mca_smu_mca_bank_dump(adev, i, &entry, qctx);
> + if (amdgpu_mca_bank_should_dump(adev, type, &entry))
> + amdgpu_mca_smu_mca_bank_dump(adev, i, &entry, qctx);
> }
>
> return 0;
