[PATCH 4/4] drm/amdgpu: add page retirement handling for CPU RAS

Zhang, Hawking Hawking.Zhang at amd.com
Wed Sep 28 07:56:20 UTC 2022


[AMD Official Use Only - General]

Series is

Reviewed-by: Hawking Zhang <Hawking.Zhang at amd.com>

Regards,
Hawking
-----Original Message-----
From: Zhou1, Tao <Tao.Zhou1 at amd.com> 
Sent: Friday, September 23, 2022 17:21
To: amd-gfx at lists.freedesktop.org; Zhang, Hawking <Hawking.Zhang at amd.com>; Yang, Stanley <Stanley.Yang at amd.com>
Cc: Zhou1, Tao <Tao.Zhou1 at amd.com>
Subject: [PATCH 4/4] drm/amdgpu: add page retirement handling for CPU RAS

Do RAS page retirement in poison consumption handler unconditional.

Signed-off-by: Tao Zhou <tao.zhou1 at amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index c7b1a2dfde13..9e98f3866edc 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -756,11 +756,7 @@ void amdgpu_amdkfd_ras_poison_consumption_handler(struct amdgpu_device *adev, bo  {
 	struct ras_err_data err_data = {0, 0, 0, NULL};
 
-	/* CPU MCA will handle page retirement if connected_to_cpu is 1 */
-	if (!adev->gmc.xgmi.connected_to_cpu)
-		amdgpu_umc_poison_handler(adev, &err_data, reset);
-	else if (reset)
-		amdgpu_amdkfd_gpu_reset(adev);
+	amdgpu_umc_poison_handler(adev, &err_data, reset);
 }
 
 bool amdgpu_amdkfd_ras_query_utcl2_poison_status(struct amdgpu_device *adev)
--
2.35.1


More information about the amd-gfx mailing list