Skip to contentSkip to conversation

drm/amdgpu: Use driver mode reset for data poison handling

PATCH In review Hawking Zhang 2 participants 3 messages Apr 16, 2024

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 drm/amdgpu: Use driver mode reset for data poison handling

mode-2 reset is the only reliable method that can get GC/SDMA back when poison is consumed. mmhub requires mode-1 reset.

---
 drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)
drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c +2 -6
index c368c70df3f4a..b6caf6eda8a0c 100644
@@ -163,17 +163,13 @@ static void event_interrupt_poison_consumption_v9(struct kfd_node *dev,
 	case SOC15_IH_CLIENTID_SE2SH:
 	case SOC15_IH_CLIENTID_SE3SH:
 	case SOC15_IH_CLIENTID_UTCL2:
-		ret = kfd_dqm_evict_pasid(dev->dqm, pasid);
 		block = AMDGPU_RAS_BLOCK__GFX;
-		if (ret)
-			reset = AMDGPU_RAS_GPU_RESET_MODE2_RESET;
+		reset = AMDGPU_RAS_GPU_RESET_MODE2_RESET;
 		break;
 	case SOC15_IH_CLIENTID_VMC:
 	case SOC15_IH_CLIENTID_VMC1:
-		ret = kfd_dqm_evict_pasid(dev->dqm, pasid);
 		block = AMDGPU_RAS_BLOCK__MMHUB;
-		if (ret)
-			reset = AMDGPU_RAS_GPU_RESET_MODE1_RESET;
+		reset = AMDGPU_RAS_GPU_RESET_MODE1_RESET;
 		break;
 	case SOC15_IH_CLIENTID_SDMA0:
 	case SOC15_IH_CLIENTID_SDMA1:
Signature · 1 line
2.17.1
Reviewed-by on v1 1/1
#1 lore raw

[AMD Official Use Only - General]

Quoted 44 lines

-----Original Message-----
From: Hawking Zhang <[email protected]>
Sent: Tuesday, April 16, 2024 12:34 PM
To: [email protected]; Zhou1, Tao <[email protected]>
Cc: Zhang, Hawking <[email protected]>
Subject: [PATCH] drm/amdgpu: Use driver mode reset for data poison handling

mode-2 reset is the only reliable method that can get GC/SDMA back when poison is consumed. mmhub requires
mode-1 reset.

---
 drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)
drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c +2 -6
index c368c70df3f4a..b6caf6eda8a0c 100644
@@ -163,17 +163,13 @@ static void
 event_interrupt_poison_consumption_v9(struct kfd_node *dev,
      case SOC15_IH_CLIENTID_SE2SH:
      case SOC15_IH_CLIENTID_SE3SH:
      case SOC15_IH_CLIENTID_UTCL2:
-             ret = kfd_dqm_evict_pasid(dev->dqm, pasid);
              block = AMDGPU_RAS_BLOCK__GFX;
-             if (ret)
-                     reset = AMDGPU_RAS_GPU_RESET_MODE2_RESET;
+             reset = AMDGPU_RAS_GPU_RESET_MODE2_RESET;
              break;
      case SOC15_IH_CLIENTID_VMC:
      case SOC15_IH_CLIENTID_VMC1:
-             ret = kfd_dqm_evict_pasid(dev->dqm, pasid);
              block = AMDGPU_RAS_BLOCK__MMHUB;
-             if (ret)
-                     reset = AMDGPU_RAS_GPU_RESET_MODE1_RESET;
+             reset = AMDGPU_RAS_GPU_RESET_MODE1_RESET;
              break;
      case SOC15_IH_CLIENTID_SDMA0:

case SOC15_IH_CLIENTID_SDMA1:
--
2.17.1

author Reviewed-by Tao Zhou
#2 lore raw

[AMD Official Use Only - General]

Please ignore this one, will send out a new one

-----Original Message-----
From: Zhou1, Tao <[email protected]>
Sent: Tuesday, April 16, 2024 01:08
To: Zhang, Hawking <[email protected]>; [email protected]
Cc: Zhang, Hawking <[email protected]>
Subject: RE: [PATCH] drm/amdgpu: Use driver mode reset for data poison handling

[AMD Official Use Only - General]

Quoted 45 lines

-----Original Message-----
From: Hawking Zhang <[email protected]>
Sent: Tuesday, April 16, 2024 12:34 PM
To: [email protected]; Zhou1, Tao <[email protected]>
Cc: Zhang, Hawking <[email protected]>
Subject: [PATCH] drm/amdgpu: Use driver mode reset for data poison handling

mode-2 reset is the only reliable method that can get GC/SDMA back when poison is consumed. mmhub requires
mode-1 reset.

---
 drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)
drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c +2 -6
index c368c70df3f4a..b6caf6eda8a0c 100644
@@ -163,17 +163,13 @@ static void
 event_interrupt_poison_consumption_v9(struct kfd_node *dev,
      case SOC15_IH_CLIENTID_SE2SH:
      case SOC15_IH_CLIENTID_SE3SH:
      case SOC15_IH_CLIENTID_UTCL2:
-             ret = kfd_dqm_evict_pasid(dev->dqm, pasid);
              block = AMDGPU_RAS_BLOCK__GFX;
-             if (ret)
-                     reset = AMDGPU_RAS_GPU_RESET_MODE2_RESET;
+             reset = AMDGPU_RAS_GPU_RESET_MODE2_RESET;
              break;
      case SOC15_IH_CLIENTID_VMC:
      case SOC15_IH_CLIENTID_VMC1:
-             ret = kfd_dqm_evict_pasid(dev->dqm, pasid);
              block = AMDGPU_RAS_BLOCK__MMHUB;
-             if (ret)
-                     reset = AMDGPU_RAS_GPU_RESET_MODE1_RESET;
+             reset = AMDGPU_RAS_GPU_RESET_MODE1_RESET;
              break;
      case SOC15_IH_CLIENTID_SDMA0:

case SOC15_IH_CLIENTID_SDMA1:
--
2.17.1