mirrors/linux
1
0
Fork 0
mirror of https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git synced 2026-05-01 03:44:27 -04:00
Code Issues Packages Projects Releases Wiki Activity

drm/amdgpu: store only one RAS bad page record for all pages in one row

Browse Source 
So eeprom space can be saved, compatible with legacy way.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This commit is contained in:
Tao Zhou
2024-10-18 18:58:54 +08:00
committed by Alex Deucher
parent e1ee2111ca
commit c3d4acf0c3
1 changed files with 27 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

35
drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
View File

@@ -2849,7 +2849,7 @@ int amdgpu_ras_save_bad_pages(struct amdgpu_device *adev,
struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
struct ras_err_handler_data *data;
struct amdgpu_ras_eeprom_control *control;
int save_count;
int save_count, unit_num, bad_page_num, i;
if (!con || !con->eh_data) {
if (new_cnt)
@@ -2861,19 +2861,38 @@ int amdgpu_ras_save_bad_pages(struct amdgpu_device *adev,
mutex_lock(&con->recovery_lock);
control = &con->eeprom_control;
data = con->eh_data;
save_count = data->count - control->ras_num_recs;
bad_page_num = control->ras_num_recs;
/* one record on eeprom stands for all pages in one memory row
* in this mode
*/
if (control->rec_type == AMDGPU_RAS_EEPROM_REC_MCA)
bad_page_num = control->ras_num_recs * adev->umc.retire_unit;
save_count = data->count - bad_page_num;
mutex_unlock(&con->recovery_lock);
unit_num = save_count / adev->umc.retire_unit;
if (new_cnt)
*new_cnt = save_count / adev->umc.retire_unit;
*new_cnt = unit_num;
/* only new entries are saved */
if (save_count > 0) {
if (amdgpu_ras_eeprom_append(control,
&data->bps[control->ras_num_recs],
save_count)) {
dev_err(adev->dev, "Failed to save EEPROM table data!");
return -EIO;
if (control->rec_type == AMDGPU_RAS_EEPROM_REC_PA) {
if (amdgpu_ras_eeprom_append(control,
&data->bps[control->ras_num_recs],
save_count)) {
dev_err(adev->dev, "Failed to save EEPROM table data!");
return -EIO;
}
} else {
for (i = 0; i < unit_num; i++) {
if (amdgpu_ras_eeprom_append(control,
&data->bps[bad_page_num + i * adev->umc.retire_unit],
1)) {
dev_err(adev->dev, "Failed to save EEPROM table data!");
return -EIO;
}
}
}
dev_info(adev->dev, "Saved %d pages to EEPROM table.\n", save_count);