drm/amdgpu: Support passing poison consumption ras block to SRIOV

Support passing poison consumption ras blocks
to SRIOV.

Signed-off-by: YiPeng Chai <YiPeng.Chai@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

3 files changed, 15 insertions, 6 deletions

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v10.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v10.c
index a7697ec8188e..9a06c6fb6605 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v10.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v10.c
@@ -132,6 +132,7 @@ enum SQ_INTERRUPT_ERROR_TYPE {
 static void event_interrupt_poison_consumption(struct kfd_node *dev,
 				uint16_t pasid, uint16_t client_id)
 {
+	enum amdgpu_ras_block block = 0;
 	int old_poison, ret = -EINVAL;
 	struct kfd_process *p = kfd_lookup_process_by_pasid(pasid);
 
@@ -151,12 +152,14 @@ static void event_interrupt_poison_consumption(struct kfd_node *dev,
 	case SOC15_IH_CLIENTID_SE3SH:
 	case SOC15_IH_CLIENTID_UTCL2:
 		ret = kfd_dqm_evict_pasid(dev->dqm, pasid);
+		block = AMDGPU_RAS_BLOCK__GFX;
 		break;
 	case SOC15_IH_CLIENTID_SDMA0:
 	case SOC15_IH_CLIENTID_SDMA1:
 	case SOC15_IH_CLIENTID_SDMA2:
 	case SOC15_IH_CLIENTID_SDMA3:
 	case SOC15_IH_CLIENTID_SDMA4:
+		block = AMDGPU_RAS_BLOCK__SDMA;
 		break;
 	default:
 		break;
@@ -171,12 +174,12 @@ static void event_interrupt_poison_consumption(struct kfd_node *dev,
 		dev_warn(dev->adev->dev,
 			"RAS poison consumption, unmap queue flow succeeded: client id %d\n",
 			client_id);
-		amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, false);
+		amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, block, false);
 	} else {
 		dev_warn(dev->adev->dev,
 			"RAS poison consumption, fall back to gpu reset flow: client id %d\n",
 			client_id);
-		amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, true);
+		amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, block, true);
 	}
 }
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c
index 2a65792fd116..7e2859736a55 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c
@@ -191,6 +191,7 @@ static void print_sq_intr_info_error(uint32_t context_id0, uint32_t context_id1)
 static void event_interrupt_poison_consumption_v11(struct kfd_node *dev,
 				uint16_t pasid, uint16_t source_id)
 {
+	enum amdgpu_ras_block block = 0;
 	int ret = -EINVAL;
 	struct kfd_process *p = kfd_lookup_process_by_pasid(pasid);
 
@@ -210,9 +211,11 @@ static void event_interrupt_poison_consumption_v11(struct kfd_node *dev,
 	case SOC15_INTSRC_SQ_INTERRUPT_MSG:
 		if (dev->dqm->ops.reset_queues)
 			ret = dev->dqm->ops.reset_queues(dev->dqm, pasid);
+		block = AMDGPU_RAS_BLOCK__GFX;
 		break;
 	case SOC21_INTSRC_SDMA_ECC:
 	default:
+		block = AMDGPU_RAS_BLOCK__GFX;
 		break;
 	}
 
@@ -221,9 +224,9 @@ static void event_interrupt_poison_consumption_v11(struct kfd_node *dev,
 	/* resetting queue passes, do page retirement without gpu reset
 	   resetting queue fails, fallback to gpu reset solution */
 	if (!ret)
-		amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, false);
+		amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, block, false);
 	else
-		amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, true);
+		amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, block, true);
 }
 
 static bool event_interrupt_isr_v11(struct kfd_node *dev,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
index 27cdaea40501..91dd5e045b51 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c
@@ -143,6 +143,7 @@ enum SQ_INTERRUPT_ERROR_TYPE {
 static void event_interrupt_poison_consumption_v9(struct kfd_node *dev,
 				uint16_t pasid, uint16_t client_id)
 {
+	enum amdgpu_ras_block block = 0;
 	int old_poison, ret = -EINVAL;
 	struct kfd_process *p = kfd_lookup_process_by_pasid(pasid);
 
@@ -162,12 +163,14 @@ static void event_interrupt_poison_consumption_v9(struct kfd_node *dev,
 	case SOC15_IH_CLIENTID_SE3SH:
 	case SOC15_IH_CLIENTID_UTCL2:
 		ret = kfd_dqm_evict_pasid(dev->dqm, pasid);
+		block = AMDGPU_RAS_BLOCK__GFX;
 		break;
 	case SOC15_IH_CLIENTID_SDMA0:
 	case SOC15_IH_CLIENTID_SDMA1:
 	case SOC15_IH_CLIENTID_SDMA2:
 	case SOC15_IH_CLIENTID_SDMA3:
 	case SOC15_IH_CLIENTID_SDMA4:
+		block = AMDGPU_RAS_BLOCK__SDMA;
 		break;
 	default:
 		break;
@@ -182,12 +185,12 @@ static void event_interrupt_poison_consumption_v9(struct kfd_node *dev,
 		dev_warn(dev->adev->dev,
 			"RAS poison consumption, unmap queue flow succeeded: client id %d\n",
 			client_id);
-		amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, false);
+		amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, block, false);
 	} else {
 		dev_warn(dev->adev->dev,
 			"RAS poison consumption, fall back to gpu reset flow: client id %d\n",
 			client_id);
-		amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, true);
+		amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, block, true);
 	}
 }