diff options
Diffstat (limited to 'drivers/gpu/drm/amd/amdkfd/kfd_int_process_v10.c')
| -rw-r--r-- | drivers/gpu/drm/amd/amdkfd/kfd_int_process_v10.c | 35 |
1 files changed, 27 insertions, 8 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 9a06c6fb6605..8e0d0356e810 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v10.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v10.c @@ -134,6 +134,7 @@ static void event_interrupt_poison_consumption(struct kfd_node *dev, { enum amdgpu_ras_block block = 0; int old_poison, ret = -EINVAL; + uint32_t reset = 0; struct kfd_process *p = kfd_lookup_process_by_pasid(pasid); if (!p) @@ -153,6 +154,8 @@ static void event_interrupt_poison_consumption(struct kfd_node *dev, 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; break; case SOC15_IH_CLIENTID_SDMA0: case SOC15_IH_CLIENTID_SDMA1: @@ -160,6 +163,7 @@ static void event_interrupt_poison_consumption(struct kfd_node *dev, case SOC15_IH_CLIENTID_SDMA3: case SOC15_IH_CLIENTID_SDMA4: block = AMDGPU_RAS_BLOCK__SDMA; + reset = AMDGPU_RAS_GPU_RESET_MODE2_RESET; break; default: break; @@ -170,17 +174,16 @@ static void event_interrupt_poison_consumption(struct kfd_node *dev, /* resetting queue passes, do page retirement without gpu reset * resetting queue fails, fallback to gpu reset solution */ - if (!ret) { + if (!ret) 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, block, false); - } else { + 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, block, true); - } + + amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, block, reset); } static bool event_interrupt_isr_v10(struct kfd_node *dev, @@ -339,7 +342,8 @@ static void event_interrupt_wq_v10(struct kfd_node *dev, break; } kfd_signal_event_interrupt(pasid, context_id0 & 0x7fffff, 23); - } else if (source_id == SOC15_INTSRC_CP_BAD_OPCODE) { + } else if (source_id == SOC15_INTSRC_CP_BAD_OPCODE && + KFD_DBG_EC_TYPE_IS_PACKET(KFD_DEBUG_CP_BAD_OP_ECODE(context_id0))) { kfd_set_dbg_ev_from_interrupt(dev, pasid, KFD_DEBUG_DOORBELL_ID(context_id0), KFD_EC_MASK(KFD_DEBUG_CP_BAD_OP_ECODE(context_id0)), @@ -367,10 +371,25 @@ static void event_interrupt_wq_v10(struct kfd_node *dev, client_id == SOC15_IH_CLIENTID_UTCL2) { struct kfd_vm_fault_info info = {0}; uint16_t ring_id = SOC15_RING_ID_FROM_IH_ENTRY(ih_ring_entry); + uint32_t node_id = SOC15_NODEID_FROM_IH_ENTRY(ih_ring_entry); + uint32_t vmid_type = SOC15_VMID_TYPE_FROM_IH_ENTRY(ih_ring_entry); + int hub_inst = 0; struct kfd_hsa_memory_exception_data exception_data; - if (client_id == SOC15_IH_CLIENTID_UTCL2 && - amdgpu_amdkfd_ras_query_utcl2_poison_status(dev->adev)) { + /* gfxhub */ + if (!vmid_type && dev->adev->gfx.funcs->ih_node_to_logical_xcc) { + hub_inst = dev->adev->gfx.funcs->ih_node_to_logical_xcc(dev->adev, + node_id); + if (hub_inst < 0) + hub_inst = 0; + } + + /* mmhub */ + if (vmid_type && client_id == SOC15_IH_CLIENTID_VMC) + hub_inst = node_id / 4; + + if (amdgpu_amdkfd_ras_query_utcl2_poison_status(dev->adev, + hub_inst, vmid_type)) { event_interrupt_poison_consumption(dev, pasid, client_id); return; } |