Merge tag 'amd-drm-next-6.5-2023-06-16' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

amd-drm-next-6.5-2023-06-16:

amdgpu:
- Misc display fixes
- W=1 fixes
- Improve scheduler naming
- DCN 3.1.4 fixes
- kdoc fixes
- Enable W=1
- VCN 4.0 fix
- xgmi fixes
- TOPDOWN fix for large BAR systems
- eDP fix
- PSR fixes
- SubVP fixes
- Freesync fix
- DPIA fix
- SMU 13.0.5 fixes
- vblflash fix
- RAS fixes
- SDMA 4 fix
- BO locking fix
- BO backing store fix
- NBIO 7.9 fixes
- GC 9.4.3 fixes
- GPU reset recovery fixes
- HMM fix

amdkfd:
- Fix NULL check
- Trap fixes
- Queue count fix
- Add event age tracking

radeon:
- fbdev client fix

scheduler:
- Avoid an infinite loop

UAPI:
- Add KFD event age tracking:
  Proposed ROCT-Thunk-Interface:
  https://github.com/RadeonOpenCompute/ROCT-Thunk-Interface/commit/efdbf6cfbc026bd68ac3c35d00dacf84370eb81e
  https://github.com/RadeonOpenCompute/ROCT-Thunk-Interface/commit/1820ae0a2db85b6f584611dc0cde1a00e7c22915
  Proposed ROCR-Runtime:
  https://github.com/RadeonOpenCompute/ROCR-Runtime/compare/master...zhums:ROCR-Runtime:new_event_wait_review
  https://github.com/RadeonOpenCompute/ROCR-Runtime/commit/e1f5bdb88eb882ac798aeca2c00ea3fbb2dba459
  https://github.com/RadeonOpenCompute/ROCR-Runtime/commit/7d26afd14107b5c2a754c1a3f415d89f3aabb503

drm:
- DP MST fix

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230616163548.7706-1-alexander.deucher@amd.com

1 files changed, 19 insertions, 15 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c
index 1edf8e6aeb16..db0d94ca4ffc 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c
@@ -169,27 +169,31 @@ int amdgpu_umc_poison_handler(struct amdgpu_device *adev, bool reset)
 {
 	int ret = AMDGPU_RAS_SUCCESS;
 
-	if (!amdgpu_sriov_vf(adev)) {
-		if (!adev->gmc.xgmi.connected_to_cpu) {
-			struct ras_err_data err_data = {0, 0, 0, NULL};
-			struct ras_common_if head = {
-				.block = AMDGPU_RAS_BLOCK__UMC,
-			};
-			struct ras_manager *obj = amdgpu_ras_find_obj(adev, &head);
-
-			ret = amdgpu_umc_do_page_retirement(adev, &err_data, NULL, reset);
-
-			if (ret == AMDGPU_RAS_SUCCESS && obj) {
-				obj->err_data.ue_count += err_data.ue_count;
-				obj->err_data.ce_count += err_data.ce_count;
-			}
-		} else if (reset) {
+	if (adev->gmc.xgmi.connected_to_cpu ||
+		adev->gmc.is_app_apu) {
+		if (reset) {
 			/* MCA poison handler is only responsible for GPU reset,
 			 * let MCA notifier do page retirement.
 			 */
 			kgd2kfd_set_sram_ecc_flag(adev->kfd.dev);
 			amdgpu_ras_reset_gpu(adev);
 		}
+		return ret;
+	}
+
+	if (!amdgpu_sriov_vf(adev)) {
+		struct ras_err_data err_data = {0, 0, 0, NULL};
+		struct ras_common_if head = {
+			.block = AMDGPU_RAS_BLOCK__UMC,
+		};
+		struct ras_manager *obj = amdgpu_ras_find_obj(adev, &head);
+
+		ret = amdgpu_umc_do_page_retirement(adev, &err_data, NULL, reset);
+
+		if (ret == AMDGPU_RAS_SUCCESS && obj) {
+			obj->err_data.ue_count += err_data.ue_count;
+			obj->err_data.ce_count += err_data.ce_count;
+		}
 	} else {
 		if (adev->virt.ops && adev->virt.ops->ras_poison_handler)
 			adev->virt.ops->ras_poison_handler(adev);