github-mirror/linux
1
0
Fork 0
mirror of https://github.com/torvalds/linux.git synced 2026-05-26 16:12:59 +02:00
Code Issues Actions Packages Projects Releases Wiki Activity Graph

drm/amdgpu: Unregister mce notifier

Browse Source 
Unregister mce notifier on unload.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This commit is contained in:
Lijo Lazar 2025-11-13 17:07:42 +05:30 committed by Alex Deucher
parent b8ecd07f3d
commit c58b5d8269
1 changed files with 27 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

28
drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
View File

@ -150,6 +150,8 @@ static void amdgpu_ras_critical_region_fini(struct amdgpu_device *adev);
#ifdef CONFIG_X86_MCE_AMD
static void amdgpu_register_bad_pages_mca_notifier(struct amdgpu_device *adev);
static void
amdgpu_unregister_bad_pages_mca_notifier(struct amdgpu_device *adev);
struct mce_notifier_adev_list {
struct amdgpu_device *devs[MAX_GPU_INSTANCE];
int num_gpu;
@ -3954,7 +3956,9 @@ static int amdgpu_ras_recovery_fini(struct amdgpu_device *adev)
mutex_unlock(&con->recovery_lock);
amdgpu_ras_critical_region_init(adev);
#ifdef CONFIG_X86_MCE_AMD
amdgpu_unregister_bad_pages_mca_notifier(adev);
#endif
return 0;
}
/* recovery end */
@ -4988,6 +4992,28 @@ static void amdgpu_register_bad_pages_mca_notifier(struct amdgpu_device *adev)
notifier_registered = true;
}
}
static void amdgpu_unregister_bad_pages_mca_notifier(struct amdgpu_device *adev)
{
int i, j;
if (!notifier_registered && !mce_adev_list.num_gpu)
return;
for (i = 0, j = 0; i < mce_adev_list.num_gpu; i++) {
if (mce_adev_list.devs[i] == adev)
mce_adev_list.devs[i] = NULL;
if (!mce_adev_list.devs[i])
++j;
}
if (j == mce_adev_list.num_gpu) {
mce_adev_list.num_gpu = 0;
/* Unregister x86 notifier with MCE subsystem. */
if (notifier_registered) {
mce_unregister_decode_chain(&amdgpu_bad_page_nb);
notifier_registered = false;
}
}
}
#endif
struct amdgpu_ras *amdgpu_ras_get_context(struct amdgpu_device *adev)