git/linux-cryptodev-2.6
1
0
Fork 0
mirror of https://git.kernel.org/pub/scm/linux/kernel/git/herbert/cryptodev-2.6.git synced 2026-04-27 03:49:57 -04:00
Code Issues Packages Projects Releases Wiki Activity

drm/amdgpu: Update usage for bad page threshold

Browse Source 
The driver's behavior varies based on
the configuration of amdgpu_bad_page_threshold setting

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This commit is contained in:
Hawking Zhang
2025-01-22 19:34:33 +08:00
committed by Alex Deucher
parent c003b5ccaf
commit 16b85a0942
4 changed files with 45 additions and 40 deletions
Download Patch File Download Diff File Expand all files Collapse all files

40
drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
View File

@@ -3080,31 +3080,29 @@ static void amdgpu_ras_validate_threshold(struct amdgpu_device *adev,
struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
/*
* Justification of value bad_page_cnt_threshold in ras structure
*
* Generally, 0 <= amdgpu_bad_page_threshold <= max record length
* in eeprom or amdgpu_bad_page_threshold == -2, introduce two
* scenarios accordingly.
*
* Bad page retirement enablement:
* - If amdgpu_bad_page_threshold = -2,
* bad_page_cnt_threshold = typical value by formula.
*
* - When the value from user is 0 < amdgpu_bad_page_threshold <
* max record length in eeprom, use it directly.
*
* Bad page retirement disablement:
* - If amdgpu_bad_page_threshold = 0, bad page retirement
* functionality is disabled, and bad_page_cnt_threshold will
* take no effect.
* amdgpu_bad_page_threshold is used to config
* the threshold for the number of bad pages.
* -1: Threshold is set to default value
* Driver will issue a warning message when threshold is reached
* and continue runtime services.
* 0: Disable bad page retirement
* Driver will not retire bad pages
* which is intended for debugging purpose.
* -2: Threshold is determined by a formula
* that assumes 1 bad page per 100M of local memory.
* Driver will continue runtime services when threhold is reached.
* 0 < threshold < max number of bad page records in EEPROM,
* A user-defined threshold is set
* Driver will halt runtime services when this custom threshold is reached.
*/
if (amdgpu_bad_page_threshold < 0) {
if (amdgpu_bad_page_threshold == -2) {
u64 val = adev->gmc.mc_vram_size;
do_div(val, RAS_BAD_PAGE_COVER);
con->bad_page_cnt_threshold = min(lower_32_bits(val),
max_count);
} else if (amdgpu_bad_page_threshold == -1) {
con->bad_page_cnt_threshold = ((con->reserved_pages_in_bytes) >> 21) << 4;
} else {
con->bad_page_cnt_threshold = min_t(int, max_count,
amdgpu_bad_page_threshold);
@@ -3848,8 +3846,10 @@ static void amdgpu_ras_init_reserved_vram_size(struct amdgpu_device *adev)
case IP_VERSION(13, 0, 2):
case IP_VERSION(13, 0, 6):
case IP_VERSION(13, 0, 12):
con->reserved_pages_in_bytes = AMDGPU_RAS_RESERVED_VRAM_SIZE_DEFAULT;
break;
case IP_VERSION(13, 0, 14):
con->reserved_pages_in_bytes = AMDGPU_RAS_RESERVED_VRAM_SIZE;
con->reserved_pages_in_bytes = (AMDGPU_RAS_RESERVED_VRAM_SIZE_DEFAULT << 1);
break;
default:
break;