mirrors/linux-stable
1
0
Fork 0
mirror of https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git synced 2024-10-02 15:18:19 +00:00
Code Issues Releases Wiki Activity

drm/amdgpu: Fix SDMA RAS error reporting on Aldebaran

Browse source 
Fix the following issues with SDMA RAS error reporting:
1. Read the EDC_COUNTER2 register also to fetch error counts
   for all sub-blocks in SDMA.
2. SDMA RAS on Aldebaran suports single-bit uncorrectable errors
   only. So, report error count in UE count instead of CE count.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-By: John Clements <John.Clements@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This commit is contained in:
Mukul Joshi 2021-03-24 11:51:35 -04:00 committed by Alex Deucher
parent 1f0d8e3781
commit ceb47e0d84
1 changed files with 28 additions and 7 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

35
drivers/gpu/drm/amd/amdgpu/sdma_v4_4.c
View file

@ -160,6 +160,7 @@ static const struct soc15_ras_field_entry sdma_v4_4_ras_fields[] = {
};
static void sdma_v4_4_get_ras_error_count(struct amdgpu_device *adev,
uint32_t reg_offset,
uint32_t value,
uint32_t instance,
uint32_t *sec_count)
@ -169,6 +170,9 @@ static void sdma_v4_4_get_ras_error_count(struct amdgpu_device *adev,
/* double bits error (multiple bits) error detection is not supported */
for (i = 0; i < ARRAY_SIZE(sdma_v4_4_ras_fields); i++) {
if (sdma_v4_4_ras_fields[i].reg_offset != reg_offset)
continue;
/* the SDMA_EDC_COUNTER register in each sdma instance
* shares the same sed shift_mask
* */
@ -197,13 +201,30 @@ static int sdma_v4_4_query_ras_error_count(struct amdgpu_device *adev,
reg_value = RREG32(reg_offset);
/* double bit error is not supported */
if (reg_value)
sdma_v4_4_get_ras_error_count(adev, reg_value, instance, &sec_count);
/* err_data->ce_count should be initialized to 0
* before calling into this function */
err_data->ce_count += sec_count;
/* double bit error is not supported
* set ue count to 0 */
err_data->ue_count = 0;
sdma_v4_4_get_ras_error_count(adev, regSDMA0_EDC_COUNTER, reg_value,
instance, &sec_count);
reg_offset = sdma_v4_4_get_reg_offset(adev, instance, regSDMA0_EDC_COUNTER2);
reg_value = RREG32(reg_offset);
/* double bit error is not supported */
if (reg_value)
sdma_v4_4_get_ras_error_count(adev, regSDMA0_EDC_COUNTER2, reg_value,
instance, &sec_count);
/*
* err_data->ue_count should be initialized to 0
* before calling into this function
*
* SDMA RAS supports single bit uncorrectable error detection.
* So, increment uncorrectable error count.
*/
err_data->ue_count += sec_count;
/*
* SDMA RAS does not support correctable errors.
* Set ce count to 0.
*/
err_data->ce_count = 0;
return 0;
};