drm/amdgpu: replace VM fault error by info logs
Subsystems (2)
PATCH v2 1/1 drm/amdgpu: replace VM fault error by info logs
This is not a kernel error. These logs are caused by VM faults that could not be handled. Typically, generated by user mode applications.
--- drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c | 14 +++++++------- drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c | 14 +++++++------- drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c | 4 ++-- drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c | 8 ++++---- drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c | 8 ++++---- drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c | 8 ++++---- drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 20 ++++++++++---------- drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c | 14 +++++++------- drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c | 14 +++++++------- 9 files changed, 52 insertions(+), 52 deletions(-)
drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c +7 -7
@@ -79,25 +79,25 @@ gfxhub_v2_0_print_l2_protection_fault_status(struct amdgpu_device *adev,u32 cid =REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,CID ); -dev_err (adev->dev, +dev_info (adev->dev,"GCVM_L2_PROTECTION_FAULT_STATUS:0x%08X\n" , status); -dev_err (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , +dev_info (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , cid >=ARRAY_SIZE (gfxhub_client_ids) ?"unknown" : gfxhub_client_ids[cid], cid); -dev_err (adev->dev,"\t MORE_FAULTS: 0x%lx\n" , +dev_info (adev->dev,"\t MORE_FAULTS: 0x%lx\n" ,REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,MORE_FAULTS )); -dev_err (adev->dev,"\t WALKER_ERROR: 0x%lx\n" , +dev_info (adev->dev,"\t WALKER_ERROR: 0x%lx\n" ,REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,WALKER_ERROR )); -dev_err (adev->dev,"\t PERMISSION_FAULTS: 0x%lx\n" , +dev_info (adev->dev,"\t PERMISSION_FAULTS: 0x%lx\n" ,REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,PERMISSION_FAULTS )); -dev_err (adev->dev,"\t MAPPING_ERROR: 0x%lx\n" , +dev_info (adev->dev,"\t MAPPING_ERROR: 0x%lx\n" ,REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,MAPPING_ERROR )); -dev_err (adev->dev,"\t RW: 0x%lx\n" , +dev_info (adev->dev,"\t RW: 0x%lx\n" ,REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,RW )); }
drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c +7 -7
@@ -82,25 +82,25 @@ gfxhub_v2_1_print_l2_protection_fault_status(struct amdgpu_device *adev,u32 cid =REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,CID ); -dev_err (adev->dev, +dev_info (adev->dev,"GCVM_L2_PROTECTION_FAULT_STATUS:0x%08X\n" , status); -dev_err (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , +dev_info (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , cid >=ARRAY_SIZE (gfxhub_client_ids) ?"unknown" : gfxhub_client_ids[cid], cid); -dev_err (adev->dev,"\t MORE_FAULTS: 0x%lx\n" , +dev_info (adev->dev,"\t MORE_FAULTS: 0x%lx\n" ,REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,MORE_FAULTS )); -dev_err (adev->dev,"\t WALKER_ERROR: 0x%lx\n" , +dev_info (adev->dev,"\t WALKER_ERROR: 0x%lx\n" ,REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,WALKER_ERROR )); -dev_err (adev->dev,"\t PERMISSION_FAULTS: 0x%lx\n" , +dev_info (adev->dev,"\t PERMISSION_FAULTS: 0x%lx\n" ,REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,PERMISSION_FAULTS )); -dev_err (adev->dev,"\t MAPPING_ERROR: 0x%lx\n" , +dev_info (adev->dev,"\t MAPPING_ERROR: 0x%lx\n" ,REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,MAPPING_ERROR )); -dev_err (adev->dev,"\t RW: 0x%lx\n" , +dev_info (adev->dev,"\t RW: 0x%lx\n" ,REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,RW )); }
drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c +2 -2
@@ -148,14 +148,14 @@ static int gmc_v10_0_process_interrupt(struct amdgpu_device *adev,memset (&task_info,0 ,sizeof (struct amdgpu_task_info ));amdgpu_vm_get_task_info (adev, entry->pasid, &task_info); -dev_err (adev->dev, +dev_info (adev->dev,"[%s] page fault (src_id:%u ring:%u vmid:%u pasid:%u, " "for process %s pid %d thread %s pid %d)\n" , entry->vmid_src ?"mmhub" :"gfxhub" , entry->src_id, entry->ring_id, entry->vmid, entry->pasid, task_info.process_name, task_info.tgid, task_info.task_name, task_info.pid); -dev_err (adev->dev," in page starting at address 0x%016llx from client 0x%x (%s)\n" , +dev_info (adev->dev," in page starting at address 0x%016llx from client 0x%x (%s)\n" , addr, entry->client_id, soc15_ih_clientid_name[entry->client_id]);
drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c +4 -4
@@ -620,7 +620,7 @@ static void gmc_v6_0_vm_decode_fault(struct amdgpu_device *adev, mc_id =REG_GET_FIELD (status,VM_CONTEXT1_PROTECTION_FAULT_STATUS ,MEMORY_CLIENT_ID ); - dev_err(adev->dev, "VM fault (0x%02x, vmid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", + dev_info(adev->dev, "VM fault (0x%02x, vmid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", protections, vmid, addr, REG_GET_FIELD(status, VM_CONTEXT1_PROTECTION_FAULT_STATUS, MEMORY_CLIENT_RW) ?
@@ -1083,11 +1083,11 @@ static int gmc_v6_0_process_interrupt(struct amdgpu_device *adev,gmc_v6_0_set_fault_enable_default (adev, false);if (printk_ratelimit ()) { -dev_err (adev->dev,"GPU fault detected: %d 0x%08x\n" , +dev_info (adev->dev,"GPU fault detected: %d 0x%08x\n" , entry->src_id, entry->src_data[0 ]); -dev_err (adev->dev," VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n" , +dev_info (adev->dev," VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n" , addr); -dev_err (adev->dev," VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n" , +dev_info (adev->dev," VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n" , status);gmc_v6_0_vm_decode_fault (adev, status, addr,0 ); }
drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c +4 -4
@@ -781,7 +781,7 @@ static void gmc_v7_0_vm_decode_fault(struct amdgpu_device *adev, u32 status, mc_id =REG_GET_FIELD (status,VM_CONTEXT1_PROTECTION_FAULT_STATUS ,MEMORY_CLIENT_ID ); - dev_err(adev->dev, "VM fault (0x%02x, vmid %d, pasid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", + dev_info(adev->dev, "VM fault (0x%02x, vmid %d, pasid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", protections, vmid, pasid, addr, REG_GET_FIELD(status, VM_CONTEXT1_PROTECTION_FAULT_STATUS, MEMORY_CLIENT_RW) ?
@@ -1286,11 +1286,11 @@ static int gmc_v7_0_process_interrupt(struct amdgpu_device *adev,gmc_v7_0_set_fault_enable_default (adev, false);if (printk_ratelimit ()) { -dev_err (adev->dev,"GPU fault detected: %d 0x%08x\n" , +dev_info (adev->dev,"GPU fault detected: %d 0x%08x\n" , entry->src_id, entry->src_data[0 ]); -dev_err (adev->dev," VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n" , +dev_info (adev->dev," VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n" , addr); -dev_err (adev->dev," VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n" , +dev_info (adev->dev," VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n" , status);gmc_v7_0_vm_decode_fault (adev, status, addr, mc_client, entry->pasid);
drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +4 -4
@@ -1021,7 +1021,7 @@ static void gmc_v8_0_vm_decode_fault(struct amdgpu_device *adev, u32 status, mc_id =REG_GET_FIELD (status,VM_CONTEXT1_PROTECTION_FAULT_STATUS ,MEMORY_CLIENT_ID ); - dev_err(adev->dev, "VM fault (0x%02x, vmid %d, pasid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", + dev_info(adev->dev, "VM fault (0x%02x, vmid %d, pasid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", protections, vmid, pasid, addr, REG_GET_FIELD(status, VM_CONTEXT1_PROTECTION_FAULT_STATUS, MEMORY_CLIENT_RW) ?
@@ -1466,12 +1466,12 @@ static int gmc_v8_0_process_interrupt(struct amdgpu_device *adev,memset (&task_info,0 ,sizeof (struct amdgpu_task_info ));amdgpu_vm_get_task_info (adev, entry->pasid, &task_info); -dev_err (adev->dev,"GPU fault detected: %d 0x%08x for process %s pid %d thread %s pid %d\n" , +dev_info (adev->dev,"GPU fault detected: %d 0x%08x for process %s pid %d thread %s pid %d\n" , entry->src_id, entry->src_data[0 ], task_info.process_name, task_info.tgid, task_info.task_name, task_info.pid); -dev_err (adev->dev," VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n" , +dev_info (adev->dev," VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n" , addr); -dev_err (adev->dev," VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n" , +dev_info (adev->dev," VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n" , status);gmc_v8_0_vm_decode_fault (adev, status, addr, mc_client, entry->pasid);
drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +10 -10
@@ -582,14 +582,14 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,memset (&task_info,0 ,sizeof (struct amdgpu_task_info ));amdgpu_vm_get_task_info (adev, entry->pasid, &task_info); -dev_err (adev->dev, +dev_info (adev->dev,"[%s] %s page fault (src_id:%u ring:%u vmid:%u " "pasid:%u, for process %s pid %d thread %s pid %d)\n" , hub_name, retry_fault ?"retry" :"no-retry" , entry->src_id, entry->ring_id, entry->vmid, entry->pasid, task_info.process_name, task_info.tgid, task_info.task_name, task_info.pid); -dev_err (adev->dev," in page starting at address 0x%016llx from IH client 0x%x (%s)\n" , +dev_info (adev->dev," in page starting at address 0x%016llx from IH client 0x%x (%s)\n" , addr, entry->client_id, soc15_ih_clientid_name[entry->client_id]);
@@ -611,11 +611,11 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,WREG32_P (hub->vm_l2_pro_fault_cntl,1 , ~1 ); -dev_err (adev->dev, +dev_info (adev->dev,"VM_L2_PROTECTION_FAULT_STATUS:0x%08X\n" , status);if (hub == &adev->vmhub[AMDGPU_GFXHUB_0 ]) { -dev_err (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , +dev_info (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , cid >=ARRAY_SIZE (gfxhub_client_ids) ?"unknown" : gfxhub_client_ids[cid], cid);
@@ -648,22 +648,22 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev, mmhub_cid =NULL ;break ; } -dev_err (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , +dev_info (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , mmhub_cid ? mmhub_cid :"unknown" , cid); } -dev_err (adev->dev,"\t MORE_FAULTS: 0x%lx\n" , +dev_info (adev->dev,"\t MORE_FAULTS: 0x%lx\n" ,REG_GET_FIELD (status,VM_L2_PROTECTION_FAULT_STATUS ,MORE_FAULTS )); -dev_err (adev->dev,"\t WALKER_ERROR: 0x%lx\n" , +dev_info (adev->dev,"\t WALKER_ERROR: 0x%lx\n" ,REG_GET_FIELD (status,VM_L2_PROTECTION_FAULT_STATUS ,WALKER_ERROR )); -dev_err (adev->dev,"\t PERMISSION_FAULTS: 0x%lx\n" , +dev_info (adev->dev,"\t PERMISSION_FAULTS: 0x%lx\n" ,REG_GET_FIELD (status,VM_L2_PROTECTION_FAULT_STATUS ,PERMISSION_FAULTS )); -dev_err (adev->dev,"\t MAPPING_ERROR: 0x%lx\n" , +dev_info (adev->dev,"\t MAPPING_ERROR: 0x%lx\n" ,REG_GET_FIELD (status,VM_L2_PROTECTION_FAULT_STATUS ,MAPPING_ERROR )); -dev_err (adev->dev,"\t RW: 0x%x\n" , rw); +dev_info (adev->dev,"\t RW: 0x%x\n" , rw);return 0 ; }
drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c +7 -7
@@ -150,7 +150,7 @@ mmhub_v2_0_print_l2_protection_fault_status(struct amdgpu_device *adev, rw =REG_GET_FIELD (status,MMVM_L2_PROTECTION_FAULT_STATUS ,RW ); -dev_err (adev->dev, +dev_info (adev->dev,"MMVM_L2_PROTECTION_FAULT_STATUS:0x%08X\n" , status);switch (adev->ip_versions[MMHUB_HWIP ][0 ]) {
@@ -169,21 +169,21 @@ mmhub_v2_0_print_l2_protection_fault_status(struct amdgpu_device *adev, mmhub_cid =NULL ;break ; } -dev_err (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , +dev_info (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , mmhub_cid ? mmhub_cid :"unknown" , cid); -dev_err (adev->dev,"\t MORE_FAULTS: 0x%lx\n" , +dev_info (adev->dev,"\t MORE_FAULTS: 0x%lx\n" ,REG_GET_FIELD (status,MMVM_L2_PROTECTION_FAULT_STATUS ,MORE_FAULTS )); -dev_err (adev->dev,"\t WALKER_ERROR: 0x%lx\n" , +dev_info (adev->dev,"\t WALKER_ERROR: 0x%lx\n" ,REG_GET_FIELD (status,MMVM_L2_PROTECTION_FAULT_STATUS ,WALKER_ERROR )); -dev_err (adev->dev,"\t PERMISSION_FAULTS: 0x%lx\n" , +dev_info (adev->dev,"\t PERMISSION_FAULTS: 0x%lx\n" ,REG_GET_FIELD (status,MMVM_L2_PROTECTION_FAULT_STATUS ,PERMISSION_FAULTS )); -dev_err (adev->dev,"\t MAPPING_ERROR: 0x%lx\n" , +dev_info (adev->dev,"\t MAPPING_ERROR: 0x%lx\n" ,REG_GET_FIELD (status,MMVM_L2_PROTECTION_FAULT_STATUS ,MAPPING_ERROR )); -dev_err (adev->dev,"\t RW: 0x%x\n" , rw); +dev_info (adev->dev,"\t RW: 0x%x\n" , rw); } static void mmhub_v2_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c +7 -7
@@ -87,7 +87,7 @@ mmhub_v2_3_print_l2_protection_fault_status(struct amdgpu_device *adev, rw =REG_GET_FIELD (status,MMVM_L2_PROTECTION_FAULT_STATUS ,RW ); -dev_err (adev->dev, +dev_info (adev->dev,"MMVM_L2_PROTECTION_FAULT_STATUS:0x%08X\n" , status);switch (adev->ip_versions[MMHUB_HWIP ][0 ]) {
@@ -100,21 +100,21 @@ mmhub_v2_3_print_l2_protection_fault_status(struct amdgpu_device *adev, mmhub_cid =NULL ;break ; } -dev_err (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , +dev_info (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , mmhub_cid ? mmhub_cid :"unknown" , cid); -dev_err (adev->dev,"\t MORE_FAULTS: 0x%lx\n" , +dev_info (adev->dev,"\t MORE_FAULTS: 0x%lx\n" ,REG_GET_FIELD (status,MMVM_L2_PROTECTION_FAULT_STATUS ,MORE_FAULTS )); -dev_err (adev->dev,"\t WALKER_ERROR: 0x%lx\n" , +dev_info (adev->dev,"\t WALKER_ERROR: 0x%lx\n" ,REG_GET_FIELD (status,MMVM_L2_PROTECTION_FAULT_STATUS ,WALKER_ERROR )); -dev_err (adev->dev,"\t PERMISSION_FAULTS: 0x%lx\n" , +dev_info (adev->dev,"\t PERMISSION_FAULTS: 0x%lx\n" ,REG_GET_FIELD (status,MMVM_L2_PROTECTION_FAULT_STATUS ,PERMISSION_FAULTS )); -dev_err (adev->dev,"\t MAPPING_ERROR: 0x%lx\n" , +dev_info (adev->dev,"\t MAPPING_ERROR: 0x%lx\n" ,REG_GET_FIELD (status,MMVM_L2_PROTECTION_FAULT_STATUS ,MAPPING_ERROR )); -dev_err (adev->dev,"\t RW: 0x%x\n" , rw); +dev_info (adev->dev,"\t RW: 0x%x\n" , rw); } static void mmhub_v2_3_setup_vm_pt_regs(struct amdgpu_device *adev,
Signature · 1 line
2.32.0
This v2 patch extends to all GMC versions.
Regards,
Alex
Quoted 427 lines
-----Original Message-----
From: Sierra Guiza, Alejandro (Alex) <[email protected]>
Sent: Friday, April 22, 2022 9:25 PM
To: [email protected]
Cc: Sierra Guiza, Alejandro (Alex) <[email protected]>
Subject: [PATCH v2] drm/amdgpu: replace VM fault error by info logsThis is not a kernel error. These logs are caused by VM faults that could not be handled. Typically, generated by user mode applications.
Signed-off-by Alex Sierra <[email protected]>--- drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c | 14 +++++++------- drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c | 14 +++++++------- drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c | 4 ++-- drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c | 8 ++++---- drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c | 8 ++++---- drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c | 8 ++++---- drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 20 ++++++++++---------- drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c | 14 +++++++------- drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c | 14 +++++++------- 9 files changed, 52 insertions(+), 52 deletions(-)drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c +6
-6@@ -79,25 +79,25 @@ gfxhub_v2_0_print_l2_protection_fault_status(structamdgpu_device *adev, u32 cid = REG_GET_FIELD(status,GCVM_L2_PROTECTION_FAULT_STATUS ,CID ); -dev_err (adev->dev, +dev_info (adev->dev,"GCVM_L2_PROTECTION_FAULT_STATUS:0x%08X\n" , status); -dev_err (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , +dev_info (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , cid >=ARRAY_SIZE (gfxhub_client_ids) ?"unknown" : gfxhub_client_ids[cid], cid); -dev_err (adev->dev,"\t MORE_FAULTS: 0x%lx\n" , +dev_info (adev->dev,"\t MORE_FAULTS: 0x%lx\n" ,REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,MORE_FAULTS )); -dev_err (adev->dev,"\t WALKER_ERROR: 0x%lx\n" , +dev_info (adev->dev,"\t WALKER_ERROR: 0x%lx\n" ,REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,WALKER_ERROR )); -dev_err (adev->dev,"\t PERMISSION_FAULTS: 0x%lx\n" , +dev_info (adev->dev,"\t PERMISSION_FAULTS: 0x%lx\n" ,REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,PERMISSION_FAULTS )); - dev_err(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", + dev_info(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", REG_GET_FIELD(status, GCVM_L2_PROTECTION_FAULT_STATUS,
MAPPING_ERROR));drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c +1
-1-dev_err (adev->dev,"\t RW: 0x%lx\n" , +dev_info (adev->dev,"\t RW: 0x%lx\n" ,REG_GET_FIELD (status, GCVM_L2_PROTECTION_FAULT_STATUS, RW)); } diff --git
a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.cb/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c
indexff738e9725ee..fdcca1477592100644drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c +6
-6@@ -82,25 +82,25 @@ gfxhub_v2_1_print_l2_protection_fault_status(structamdgpu_device *adev, u32 cid = REG_GET_FIELD(status,GCVM_L2_PROTECTION_FAULT_STATUS ,CID ); -dev_err (adev->dev, +dev_info (adev->dev,"GCVM_L2_PROTECTION_FAULT_STATUS:0x%08X\n" , status); -dev_err (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , +dev_info (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , cid >=ARRAY_SIZE (gfxhub_client_ids) ?"unknown" : gfxhub_client_ids[cid], cid); -dev_err (adev->dev,"\t MORE_FAULTS: 0x%lx\n" , +dev_info (adev->dev,"\t MORE_FAULTS: 0x%lx\n" ,REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,MORE_FAULTS )); -dev_err (adev->dev,"\t WALKER_ERROR: 0x%lx\n" , +dev_info (adev->dev,"\t WALKER_ERROR: 0x%lx\n" ,REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,WALKER_ERROR )); -dev_err (adev->dev,"\t PERMISSION_FAULTS: 0x%lx\n" , +dev_info (adev->dev,"\t PERMISSION_FAULTS: 0x%lx\n" ,REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,PERMISSION_FAULTS )); - dev_err(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", + dev_info(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", REG_GET_FIELD(status, GCVM_L2_PROTECTION_FAULT_STATUS,
MAPPING_ERROR));drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c +1
-1-dev_err (adev->dev,"\t RW: 0x%lx\n" , +dev_info (adev->dev,"\t RW: 0x%lx\n" ,REG_GET_FIELD (status, GCVM_L2_PROTECTION_FAULT_STATUS, RW)); } diff --git
a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.cb/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
indexa455e59f41f4..864fcc0edb90100644drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c +3
-2@@ -148,14 +148,14 @@ static int gmc_v10_0_process_interrupt(struct amdgpu_device *adev,memset (&task_info,0 ,sizeof (struct amdgpu_task_info ));amdgpu_vm_get_task_info (adev, entry->pasid, &task_info); -dev_err (adev->dev, +dev_info (adev->dev,"[%s] page fault (src_id:%u ring:%u vmid:%u pasid:%u, " "for process %s pid %d thread %s pid %d)\n" , entry->vmid_src ?"mmhub" :"gfxhub" , entry->src_id, entry->ring_id, entry->vmid, entry->pasid, task_info.process_name, task_info.tgid, task_info.task_name, task_info.pid); - dev_err(adev->dev, " in page starting at address 0x%016llx from client 0x%x (%s)\n", + dev_info(adev->dev, " in page starting at address 0x%016llx from +client 0x%x (%s)\n", addr, entry->client_id, soc15_ih_clientid_name[entry->client_id]);drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c +1
-1@@ -620,7 +620,7 @@ static void gmc_v6_0_vm_decode_fault(structamdgpu_device *adev, mc_id =REG_GET_FIELD (status,VM_CONTEXT1_PROTECTION_FAULT_STATUS ,MEMORY_CLIENT_ID ); - dev_err(adev->dev, "VM fault (0x%02x, vmid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", + dev_info(adev->dev, "VM fault (0x%02x, vmid %d) at page %u, %sfrom
+'%s' (0x%08x) (%d)\n",drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c +0
-0protections, vmid, addr, REG_GET_FIELD(status,
VM_CONTEXT1_PROTECTION_FAULT_STATUS,drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c +2
-3MEMORY_CLIENT_RW) ?@@ -1083,11 +1083,11 @@ static int gmc_v6_0_process_interrupt(structamdgpu_device *adev,gmc_v6_0_set_fault_enable_default (adev ,false );if (printk_ratelimit ()) { -dev_err (adev->dev,"GPU fault detected: %d 0x%08x\n" , +dev_info (adev->dev,"GPU fault detected: %d 0x%08x\n" , entry->src_id, entry->src_data[0 ]); - dev_err(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n", + dev_info(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n",addr ); - dev_err(adev->dev, "
VM_CONTEXT1_PROTECTION_FAULT_STATUS0x%08X\n",drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c +1
-0+ dev_info(adev->dev, "
VM_CONTEXT1_PROTECTION_FAULT_STATUS0x%08X\n",drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c +0
-0status);gmc_v6_0_vm_decode_fault (adev, status, addr,0 ); }drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c +1
-1@@ -781,7 +781,7 @@ static void gmc_v7_0_vm_decode_fault(struct amdgpu_device *adev, u32 status, mc_id =REG_GET_FIELD (status,VM_CONTEXT1_PROTECTION_FAULT_STATUS ,MEMORY_CLIENT_ID ); - dev_err(adev->dev, "VM fault (0x%02x, vmid %d, pasid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", + dev_info(adev->dev, "VM fault (0x%02x, vmid %d, pasid %d) at page%u,
+%s from '%s' (0x%08x) (%d)\n",drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c +0
-0protections, vmid, pasid, addr, REG_GET_FIELD(status,
VM_CONTEXT1_PROTECTION_FAULT_STATUS,drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c +2
-3MEMORY_CLIENT_RW) ?@@ -1286,11 +1286,11 @@ static int gmc_v7_0_process_interrupt(structamdgpu_device *adev,gmc_v7_0_set_fault_enable_default (adev ,false );if (printk_ratelimit ()) { -dev_err (adev->dev,"GPU fault detected: %d 0x%08x\n" , +dev_info (adev->dev,"GPU fault detected: %d 0x%08x\n" , entry->src_id, entry->src_data[0 ]); - dev_err(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n", + dev_info(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n",addr ); - dev_err(adev->dev, "
VM_CONTEXT1_PROTECTION_FAULT_STATUS0x%08X\n",drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c +1
-0+ dev_info(adev->dev, "
VM_CONTEXT1_PROTECTION_FAULT_STATUS0x%08X\n",drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c +0
-0status);gmc_v7_0_vm_decode_fault (adev, status, addr, mc_client, entry->pasid);drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +1
-1@@ -1021,7 +1021,7 @@ static void gmc_v8_0_vm_decode_fault(struct amdgpu_device *adev, u32 status, mc_id =REG_GET_FIELD (status,VM_CONTEXT1_PROTECTION_FAULT_STATUS ,MEMORY_CLIENT_ID ); - dev_err(adev->dev, "VM fault (0x%02x, vmid %d, pasid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", + dev_info(adev->dev, "VM fault (0x%02x, vmid %d, pasid %d) at page%u,
+%s from '%s' (0x%08x) (%d)\n",drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +0
-0protections, vmid, pasid, addr, REG_GET_FIELD(status,
VM_CONTEXT1_PROTECTION_FAULT_STATUS,drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +1
-1MEMORY_CLIENT_RW) ?@@ -1466,12 +1466,12 @@ static int gmc_v8_0_process_interrupt(struct amdgpu_device *adev,memset (&task_info,0 ,sizeof (struct amdgpu_task_info ));amdgpu_vm_get_task_info (adev, entry->pasid, &task_info); - dev_err(adev->dev, "GPU fault detected: %d 0x%08x for process %s pid %d thread %s pid %d\n", + dev_info(adev->dev, "GPU fault detected: %d 0x%08x forprocess %s pid
+%d thread %s pid %d\n",
entry->src_id, entry->src_data[0],
task_info.process_name,drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +0
-1task_info.tgid, task_info.task_name, task_info.pid); - dev_err(adev->dev, "
VM_CONTEXT1_PROTECTION_FAULT_ADDR0x%08X\n",drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +1
-0+ dev_info(adev->dev, "
VM_CONTEXT1_PROTECTION_FAULT_ADDR0x%08X\n",drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +0
-1addr); - dev_err(adev->dev, "
VM_CONTEXT1_PROTECTION_FAULT_STATUS0x%08X\n",drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +1
-0+ dev_info(adev->dev, "
VM_CONTEXT1_PROTECTION_FAULT_STATUS0x%08X\n",drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +0
-0status);gmc_v8_0_vm_decode_fault (adev, status, addr, mc_client, entry->pasid);drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +5
-4@@ -582,14 +582,14 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,memset (&task_info,0 ,sizeof (struct amdgpu_task_info ));amdgpu_vm_get_task_info (adev, entry->pasid, &task_info); -dev_err (adev->dev, +dev_info (adev->dev,"[%s] %s page fault (src_id:%u ring:%u vmid:%u " "pasid:%u, for process %s pid %d thread %s pid %d)\n" , hub_name, retry_fault ?"retry" :"no-retry" , entry->src_id, entry->ring_id, entry->vmid, entry->pasid, task_info.process_name, task_info.tgid, task_info.task_name, task_info.pid); - dev_err(adev->dev, " in page starting at address 0x%016llx from IH client 0x%x (%s)\n", + dev_info(adev->dev, " in page starting at address 0x%016llx from IH +client 0x%x (%s)\n", addr, entry->client_id, soc15_ih_clientid_name[entry->client_id]);@@ -611,11 +611,11 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,WREG32_P (hub->vm_l2_pro_fault_cntl,1 , ~1 ); -dev_err (adev->dev, +dev_info (adev->dev,"VM_L2_PROTECTION_FAULT_STATUS:0x%08X\n" , status);if (hub == &adev->vmhub[AMDGPU_GFXHUB_0 ]) { - dev_err(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n", + dev_info(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n", cid >= ARRAY_SIZE(gfxhub_client_ids) ? "unknown" :drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +6
-6gfxhub_client_ids[cid], cid);@@ -648,22 +648,22 @@ static int gmc_v9_0_process_interrupt(structamdgpu_device *adev, mmhub_cid =NULL ;break ; } -dev_err (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , + dev_info(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n", mmhub_cid ? mmhub_cid :"unknown" , cid); } -dev_err (adev->dev,"\t MORE_FAULTS: 0x%lx\n" , +dev_info (adev->dev,"\t MORE_FAULTS: 0x%lx\n" ,REG_GET_FIELD (status,VM_L2_PROTECTION_FAULT_STATUS ,MORE_FAULTS )); -dev_err (adev->dev,"\t WALKER_ERROR: 0x%lx\n" , +dev_info (adev->dev,"\t WALKER_ERROR: 0x%lx\n" ,REG_GET_FIELD (status,VM_L2_PROTECTION_FAULT_STATUS ,WALKER_ERROR )); -dev_err (adev->dev,"\t PERMISSION_FAULTS: 0x%lx\n" , +dev_info (adev->dev,"\t PERMISSION_FAULTS: 0x%lx\n" ,REG_GET_FIELD (status,VM_L2_PROTECTION_FAULT_STATUS ,PERMISSION_FAULTS )); -dev_err (adev->dev,"\t MAPPING_ERROR: 0x%lx\n" , +dev_info (adev->dev,"\t MAPPING_ERROR: 0x%lx\n" ,REG_GET_FIELD (status,VM_L2_PROTECTION_FAULT_STATUS ,MAPPING_ERROR )); -dev_err (adev->dev,"\t RW: 0x%x\n" , rw); +dev_info (adev->dev,"\t RW: 0x%x\n" , rw);drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +0
-0return 0 ; }drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c +0
-0@@ -150,7 +150,7 @@mmhub_v2_0_print_l2_protection_fault_status(struct
amdgpu_device*adev,drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c +1
-1rw =REG_GET_FIELD (status,MMVM_L2_PROTECTION_FAULT_STATUS ,RW ); -dev_err (adev->dev, +dev_info (adev->dev,"MMVM_L2_PROTECTION_FAULT_STATUS:0x%08X\n" , status); switch (adev->ip_versions[MMHUB_HWIP][0]) { @@ -169,21 +169,21@@ mmhub_v2_0_print_l2_protection_fault_status(struct
amdgpu_device*adev,drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c +3
-3mmhub_cid =NULL ;break ; } -dev_err (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , +dev_info (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , mmhub_cid ? mmhub_cid :"unknown" , cid); -dev_err (adev->dev,"\t MORE_FAULTS: 0x%lx\n" , +dev_info (adev->dev,"\t MORE_FAULTS: 0x%lx\n" ,REG_GET_FIELD (status,MMVM_L2_PROTECTION_FAULT_STATUS ,MORE_FAULTS )); - dev_err(adev->dev, "\t WALKER_ERROR: 0x%lx\n", + dev_info(adev->dev, "\t WALKER_ERROR: 0x%lx\n", REG_GET_FIELD(status, MMVM_L2_PROTECTION_FAULT_STATUS,
WALKER_ERROR));drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c +1
-1- dev_err(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n", + dev_info(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n", REG_GET_FIELD(status, MMVM_L2_PROTECTION_FAULT_STATUS,
PERMISSION_FAULTS));drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c +1
-1- dev_err(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", + dev_info(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", REG_GET_FIELD(status, MMVM_L2_PROTECTION_FAULT_STATUS,
MAPPING_ERROR));drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c +1
-1-dev_err (adev->dev,"\t RW: 0x%x\n" , rw); +dev_info (adev->dev,"\t RW: 0x%x\n" , rw); } static void mmhub_v2_0_setup_vm_pt_regs(struct amdgpu_device *adev,
uint32_tvmid, diff --gita/drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.cb/drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c
indexff44c5364a8c..72dda850e7d3100644drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c +1
-1@@ -87,7 +87,7 @@ mmhub_v2_3_print_l2_protection_fault_status(structamdgpu_device *adev, rw =REG_GET_FIELD (status,MMVM_L2_PROTECTION_FAULT_STATUS ,RW ); -dev_err (adev->dev, +dev_info (adev->dev,"MMVM_L2_PROTECTION_FAULT_STATUS:0x%08X\n" , status);switch (adev->ip_versions[MMHUB_HWIP][0]) { @@ -100,21 +100,21
@@ mmhub_v2_3_print_l2_protection_fault_status(structamdgpu_device*adev,drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c +3
-3mmhub_cid =NULL ;break ; } -dev_err (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , +dev_info (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , mmhub_cid ? mmhub_cid :"unknown" , cid); -dev_err (adev->dev,"\t MORE_FAULTS: 0x%lx\n" , +dev_info (adev->dev,"\t MORE_FAULTS: 0x%lx\n" ,REG_GET_FIELD (status,MMVM_L2_PROTECTION_FAULT_STATUS ,MORE_FAULTS )); - dev_err(adev->dev, "\t WALKER_ERROR: 0x%lx\n", + dev_info(adev->dev, "\t WALKER_ERROR: 0x%lx\n", REG_GET_FIELD(status, MMVM_L2_PROTECTION_FAULT_STATUS,
WALKER_ERROR));drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c +1
-1- dev_err(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n", + dev_info(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n", REG_GET_FIELD(status, MMVM_L2_PROTECTION_FAULT_STATUS,
PERMISSION_FAULTS));drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c +1
-1- dev_err(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", + dev_info(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", REG_GET_FIELD(status, MMVM_L2_PROTECTION_FAULT_STATUS,
MAPPING_ERROR));drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c +1
-2-dev_err (adev->dev,"\t RW: 0x%x\n" , rw); +dev_info (adev->dev,"\t RW: 0x%x\n" , rw); } static void mmhub_v2_3_setup_vm_pt_regs(struct amdgpu_device *adev, --2.32.0
Am 23.04.22 um 04:24 schrieb Alex Sierra:
Quoted 2 lines
This is not a kernel error. These logs are caused by VM faults that could not be handled. Typically, generated by user mode applications.
Well it's still a hardware fault which should be logged as an error.
So I'm absolutely not keen about reducing it to just an information.
Regards,
Christian.
Quoted 340 lines
Signed-off-by Alex Sierra <[email protected]>--- drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c | 14 +++++++------- drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c | 14 +++++++------- drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c | 4 ++-- drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c | 8 ++++---- drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c | 8 ++++---- drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c | 8 ++++---- drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 20 ++++++++++---------- drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c | 14 +++++++------- drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c | 14 +++++++------- 9 files changed, 52 insertions(+), 52 deletions(-)drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c +7
-7@@ -79,25 +79,25 @@ gfxhub_v2_0_print_l2_protection_fault_status(struct amdgpu_device *adev,u32 cid =REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,CID ); -dev_err (adev->dev, +dev_info (adev->dev,"GCVM_L2_PROTECTION_FAULT_STATUS:0x%08X\n" , status); -dev_err (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , +dev_info (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , cid >=ARRAY_SIZE (gfxhub_client_ids) ?"unknown" : gfxhub_client_ids[cid], cid); -dev_err (adev->dev,"\t MORE_FAULTS: 0x%lx\n" , +dev_info (adev->dev,"\t MORE_FAULTS: 0x%lx\n" ,REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,MORE_FAULTS )); -dev_err (adev->dev,"\t WALKER_ERROR: 0x%lx\n" , +dev_info (adev->dev,"\t WALKER_ERROR: 0x%lx\n" ,REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,WALKER_ERROR )); -dev_err (adev->dev,"\t PERMISSION_FAULTS: 0x%lx\n" , +dev_info (adev->dev,"\t PERMISSION_FAULTS: 0x%lx\n" ,REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,PERMISSION_FAULTS )); -dev_err (adev->dev,"\t MAPPING_ERROR: 0x%lx\n" , +dev_info (adev->dev,"\t MAPPING_ERROR: 0x%lx\n" ,REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,MAPPING_ERROR )); -dev_err (adev->dev,"\t RW: 0x%lx\n" , +dev_info (adev->dev,"\t RW: 0x%lx\n" ,REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,RW )); }drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c +7
-7@@ -82,25 +82,25 @@ gfxhub_v2_1_print_l2_protection_fault_status(struct amdgpu_device *adev,u32 cid =REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,CID ); -dev_err (adev->dev, +dev_info (adev->dev,"GCVM_L2_PROTECTION_FAULT_STATUS:0x%08X\n" , status); -dev_err (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , +dev_info (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , cid >=ARRAY_SIZE (gfxhub_client_ids) ?"unknown" : gfxhub_client_ids[cid], cid); -dev_err (adev->dev,"\t MORE_FAULTS: 0x%lx\n" , +dev_info (adev->dev,"\t MORE_FAULTS: 0x%lx\n" ,REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,MORE_FAULTS )); -dev_err (adev->dev,"\t WALKER_ERROR: 0x%lx\n" , +dev_info (adev->dev,"\t WALKER_ERROR: 0x%lx\n" ,REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,WALKER_ERROR )); -dev_err (adev->dev,"\t PERMISSION_FAULTS: 0x%lx\n" , +dev_info (adev->dev,"\t PERMISSION_FAULTS: 0x%lx\n" ,REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,PERMISSION_FAULTS )); -dev_err (adev->dev,"\t MAPPING_ERROR: 0x%lx\n" , +dev_info (adev->dev,"\t MAPPING_ERROR: 0x%lx\n" ,REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,MAPPING_ERROR )); -dev_err (adev->dev,"\t RW: 0x%lx\n" , +dev_info (adev->dev,"\t RW: 0x%lx\n" ,REG_GET_FIELD (status,GCVM_L2_PROTECTION_FAULT_STATUS ,RW )); }drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c +2
-2@@ -148,14 +148,14 @@ static int gmc_v10_0_process_interrupt(struct amdgpu_device *adev,memset (&task_info,0 ,sizeof (struct amdgpu_task_info ));amdgpu_vm_get_task_info (adev, entry->pasid, &task_info); -dev_err (adev->dev, +dev_info (adev->dev,"[%s] page fault (src_id:%u ring:%u vmid:%u pasid:%u, " "for process %s pid %d thread %s pid %d)\n" , entry->vmid_src ?"mmhub" :"gfxhub" , entry->src_id, entry->ring_id, entry->vmid, entry->pasid, task_info.process_name, task_info.tgid, task_info.task_name, task_info.pid); -dev_err (adev->dev," in page starting at address 0x%016llx from client 0x%x (%s)\n" , +dev_info (adev->dev," in page starting at address 0x%016llx from client 0x%x (%s)\n" , addr, entry->client_id, soc15_ih_clientid_name[entry->client_id]);drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c +4
-4@@ -620,7 +620,7 @@ static void gmc_v6_0_vm_decode_fault(struct amdgpu_device *adev, mc_id =REG_GET_FIELD (status,VM_CONTEXT1_PROTECTION_FAULT_STATUS ,MEMORY_CLIENT_ID ); - dev_err(adev->dev, "VM fault (0x%02x, vmid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", + dev_info(adev->dev, "VM fault (0x%02x, vmid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", protections, vmid, addr, REG_GET_FIELD(status, VM_CONTEXT1_PROTECTION_FAULT_STATUS, MEMORY_CLIENT_RW) ?@@ -1083,11 +1083,11 @@ static int gmc_v6_0_process_interrupt(struct amdgpu_device *adev,gmc_v6_0_set_fault_enable_default (adev, false);if (printk_ratelimit ()) { -dev_err (adev->dev,"GPU fault detected: %d 0x%08x\n" , +dev_info (adev->dev,"GPU fault detected: %d 0x%08x\n" , entry->src_id, entry->src_data[0 ]); -dev_err (adev->dev," VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n" , +dev_info (adev->dev," VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n" , addr); -dev_err (adev->dev," VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n" , +dev_info (adev->dev," VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n" , status);gmc_v6_0_vm_decode_fault (adev, status, addr,0 ); }drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c +4
-4@@ -781,7 +781,7 @@ static void gmc_v7_0_vm_decode_fault(struct amdgpu_device *adev, u32 status, mc_id =REG_GET_FIELD (status,VM_CONTEXT1_PROTECTION_FAULT_STATUS ,MEMORY_CLIENT_ID ); - dev_err(adev->dev, "VM fault (0x%02x, vmid %d, pasid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", + dev_info(adev->dev, "VM fault (0x%02x, vmid %d, pasid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", protections, vmid, pasid, addr, REG_GET_FIELD(status, VM_CONTEXT1_PROTECTION_FAULT_STATUS, MEMORY_CLIENT_RW) ?@@ -1286,11 +1286,11 @@ static int gmc_v7_0_process_interrupt(struct amdgpu_device *adev,gmc_v7_0_set_fault_enable_default (adev, false);if (printk_ratelimit ()) { -dev_err (adev->dev,"GPU fault detected: %d 0x%08x\n" , +dev_info (adev->dev,"GPU fault detected: %d 0x%08x\n" , entry->src_id, entry->src_data[0 ]); -dev_err (adev->dev," VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n" , +dev_info (adev->dev," VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n" , addr); -dev_err (adev->dev," VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n" , +dev_info (adev->dev," VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n" , status);gmc_v7_0_vm_decode_fault (adev, status, addr, mc_client, entry->pasid);drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +4
-4@@ -1021,7 +1021,7 @@ static void gmc_v8_0_vm_decode_fault(struct amdgpu_device *adev, u32 status, mc_id =REG_GET_FIELD (status,VM_CONTEXT1_PROTECTION_FAULT_STATUS ,MEMORY_CLIENT_ID ); - dev_err(adev->dev, "VM fault (0x%02x, vmid %d, pasid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", + dev_info(adev->dev, "VM fault (0x%02x, vmid %d, pasid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", protections, vmid, pasid, addr, REG_GET_FIELD(status, VM_CONTEXT1_PROTECTION_FAULT_STATUS, MEMORY_CLIENT_RW) ?@@ -1466,12 +1466,12 @@ static int gmc_v8_0_process_interrupt(struct amdgpu_device *adev,memset (&task_info,0 ,sizeof (struct amdgpu_task_info ));amdgpu_vm_get_task_info (adev, entry->pasid, &task_info); -dev_err (adev->dev,"GPU fault detected: %d 0x%08x for process %s pid %d thread %s pid %d\n" , +dev_info (adev->dev,"GPU fault detected: %d 0x%08x for process %s pid %d thread %s pid %d\n" , entry->src_id, entry->src_data[0 ], task_info.process_name, task_info.tgid, task_info.task_name, task_info.pid); -dev_err (adev->dev," VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n" , +dev_info (adev->dev," VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n" , addr); -dev_err (adev->dev," VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n" , +dev_info (adev->dev," VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n" , status);gmc_v8_0_vm_decode_fault (adev, status, addr, mc_client, entry->pasid);drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +10
-10@@ -582,14 +582,14 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,memset (&task_info,0 ,sizeof (struct amdgpu_task_info ));amdgpu_vm_get_task_info (adev, entry->pasid, &task_info); -dev_err (adev->dev, +dev_info (adev->dev,"[%s] %s page fault (src_id:%u ring:%u vmid:%u " "pasid:%u, for process %s pid %d thread %s pid %d)\n" , hub_name, retry_fault ?"retry" :"no-retry" , entry->src_id, entry->ring_id, entry->vmid, entry->pasid, task_info.process_name, task_info.tgid, task_info.task_name, task_info.pid); -dev_err (adev->dev," in page starting at address 0x%016llx from IH client 0x%x (%s)\n" , +dev_info (adev->dev," in page starting at address 0x%016llx from IH client 0x%x (%s)\n" , addr, entry->client_id, soc15_ih_clientid_name[entry->client_id]);@@ -611,11 +611,11 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,WREG32_P (hub->vm_l2_pro_fault_cntl,1 , ~1 ); -dev_err (adev->dev, +dev_info (adev->dev,"VM_L2_PROTECTION_FAULT_STATUS:0x%08X\n" , status);if (hub == &adev->vmhub[AMDGPU_GFXHUB_0 ]) { -dev_err (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , +dev_info (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , cid >=ARRAY_SIZE (gfxhub_client_ids) ?"unknown" : gfxhub_client_ids[cid], cid);@@ -648,22 +648,22 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev, mmhub_cid =NULL ;break ; } -dev_err (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , +dev_info (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , mmhub_cid ? mmhub_cid :"unknown" , cid); } -dev_err (adev->dev,"\t MORE_FAULTS: 0x%lx\n" , +dev_info (adev->dev,"\t MORE_FAULTS: 0x%lx\n" ,REG_GET_FIELD (status,VM_L2_PROTECTION_FAULT_STATUS ,MORE_FAULTS )); -dev_err (adev->dev,"\t WALKER_ERROR: 0x%lx\n" , +dev_info (adev->dev,"\t WALKER_ERROR: 0x%lx\n" ,REG_GET_FIELD (status,VM_L2_PROTECTION_FAULT_STATUS ,WALKER_ERROR )); -dev_err (adev->dev,"\t PERMISSION_FAULTS: 0x%lx\n" , +dev_info (adev->dev,"\t PERMISSION_FAULTS: 0x%lx\n" ,REG_GET_FIELD (status,VM_L2_PROTECTION_FAULT_STATUS ,PERMISSION_FAULTS )); -dev_err (adev->dev,"\t MAPPING_ERROR: 0x%lx\n" , +dev_info (adev->dev,"\t MAPPING_ERROR: 0x%lx\n" ,REG_GET_FIELD (status,VM_L2_PROTECTION_FAULT_STATUS ,MAPPING_ERROR )); -dev_err (adev->dev,"\t RW: 0x%x\n" , rw); +dev_info (adev->dev,"\t RW: 0x%x\n" , rw);return 0 ; }drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c +7
-7@@ -150,7 +150,7 @@ mmhub_v2_0_print_l2_protection_fault_status(struct amdgpu_device *adev, rw =REG_GET_FIELD (status,MMVM_L2_PROTECTION_FAULT_STATUS ,RW ); -dev_err (adev->dev, +dev_info (adev->dev,"MMVM_L2_PROTECTION_FAULT_STATUS:0x%08X\n" , status);switch (adev->ip_versions[MMHUB_HWIP ][0 ]) {@@ -169,21 +169,21 @@ mmhub_v2_0_print_l2_protection_fault_status(struct amdgpu_device *adev, mmhub_cid =NULL ;break ; } -dev_err (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , +dev_info (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , mmhub_cid ? mmhub_cid :"unknown" , cid); -dev_err (adev->dev,"\t MORE_FAULTS: 0x%lx\n" , +dev_info (adev->dev,"\t MORE_FAULTS: 0x%lx\n" ,REG_GET_FIELD (status,MMVM_L2_PROTECTION_FAULT_STATUS ,MORE_FAULTS )); -dev_err (adev->dev,"\t WALKER_ERROR: 0x%lx\n" , +dev_info (adev->dev,"\t WALKER_ERROR: 0x%lx\n" ,REG_GET_FIELD (status,MMVM_L2_PROTECTION_FAULT_STATUS ,WALKER_ERROR )); -dev_err (adev->dev,"\t PERMISSION_FAULTS: 0x%lx\n" , +dev_info (adev->dev,"\t PERMISSION_FAULTS: 0x%lx\n" ,REG_GET_FIELD (status,MMVM_L2_PROTECTION_FAULT_STATUS ,PERMISSION_FAULTS )); -dev_err (adev->dev,"\t MAPPING_ERROR: 0x%lx\n" , +dev_info (adev->dev,"\t MAPPING_ERROR: 0x%lx\n" ,REG_GET_FIELD (status,MMVM_L2_PROTECTION_FAULT_STATUS ,MAPPING_ERROR )); -dev_err (adev->dev,"\t RW: 0x%x\n" , rw); +dev_info (adev->dev,"\t RW: 0x%x\n" , rw); } static void mmhub_v2_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c +7
-7@@ -87,7 +87,7 @@ mmhub_v2_3_print_l2_protection_fault_status(struct amdgpu_device *adev, rw =REG_GET_FIELD (status,MMVM_L2_PROTECTION_FAULT_STATUS ,RW ); -dev_err (adev->dev, +dev_info (adev->dev,"MMVM_L2_PROTECTION_FAULT_STATUS:0x%08X\n" , status);switch (adev->ip_versions[MMHUB_HWIP ][0 ]) {@@ -100,21 +100,21 @@ mmhub_v2_3_print_l2_protection_fault_status(struct amdgpu_device *adev, mmhub_cid =NULL ;break ; } -dev_err (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , +dev_info (adev->dev,"\t Faulty UTCL2 client ID: %s (0x%x)\n" , mmhub_cid ? mmhub_cid :"unknown" , cid); -dev_err (adev->dev,"\t MORE_FAULTS: 0x%lx\n" , +dev_info (adev->dev,"\t MORE_FAULTS: 0x%lx\n" ,REG_GET_FIELD (status,MMVM_L2_PROTECTION_FAULT_STATUS ,MORE_FAULTS )); -dev_err (adev->dev,"\t WALKER_ERROR: 0x%lx\n" , +dev_info (adev->dev,"\t WALKER_ERROR: 0x%lx\n" ,REG_GET_FIELD (status,MMVM_L2_PROTECTION_FAULT_STATUS ,WALKER_ERROR )); -dev_err (adev->dev,"\t PERMISSION_FAULTS: 0x%lx\n" , +dev_info (adev->dev,"\t PERMISSION_FAULTS: 0x%lx\n" ,REG_GET_FIELD (status,MMVM_L2_PROTECTION_FAULT_STATUS ,PERMISSION_FAULTS )); -dev_err (adev->dev,"\t MAPPING_ERROR: 0x%lx\n" , +dev_info (adev->dev,"\t MAPPING_ERROR: 0x%lx\n" ,REG_GET_FIELD (status,MMVM_L2_PROTECTION_FAULT_STATUS ,MAPPING_ERROR )); -dev_err (adev->dev,"\t RW: 0x%x\n" , rw); +dev_info (adev->dev,"\t RW: 0x%x\n" , rw); } static void mmhub_v2_3_setup_vm_pt_regs(struct amdgpu_device *adev,
Am 2022-04-23 um 13:54 schrieb Christian König:
Quoted 5 lines
Am 23.04.22 um 04:24 schrieb Alex Sierra:
Quoted 2 lines
This is not a kernel error. These logs are caused by VM faults that could not be handled. Typically, generated by user mode applications.
Well it's still a hardware fault which should be logged as an error.
At least in ROCm compute applications, a VM fault does not take down the hardware. It only leads to termination of the process that caused the fault. It's very similar to a segfault in an application, which is not considered a HW fault either. Other processes are completely unaffected. The cause of the error is typically in user mode. I think the general policy is, that user mode errors should not spam the kernel logs with error messages. The messages are useful for debugging application issues, so it's good to have them. But IMHO they should not be error messages. Such messages often lead to spurious bug reports against the kernel for things that aren't really kernel issues.
Regards,
Felix
Quoted 363 lines
So I'm absolutely not keen about reducing it to just an information.
Regards,
Christian.Quoted 356 lines
Signed-off-by Alex Sierra <[email protected]>---
drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c| 14 +++++++-------drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c| 14 +++++++------- drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c | 4 ++--drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c | 8 ++++---- drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c | 8 ++++---- drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c | 8 ++++---- drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 20 ++++++++++---------- drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c | 14 +++++++------- drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c | 14 +++++++------- 9 files changed, 52 insertions(+), 52 deletions(-)drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c +7
-6@@ -79,25 +79,25 @@ gfxhub_v2_0_print_l2_protection_fault_status(struct amdgpu_device *adev, u32 cid = REG_GET_FIELD(status, GCVM_L2_PROTECTION_FAULT_STATUS, CID); - dev_err(adev->dev, + dev_info(adev->dev, "GCVM_L2_PROTECTION_FAULT_STATUS:0x%08X\n", status); - dev_err(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n", + dev_info(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n", cid >= ARRAY_SIZE(gfxhub_client_ids) ? "unknown" : gfxhub_client_ids[cid], cid); - dev_err(adev->dev, "\t MORE_FAULTS: 0x%lx\n", + dev_info(adev->dev, "\t MORE_FAULTS: 0x%lx\n", REG_GET_FIELD(status, GCVM_L2_PROTECTION_FAULT_STATUS, MORE_FAULTS)); - dev_err(adev->dev, "\t WALKER_ERROR: 0x%lx\n", + dev_info(adev->dev, "\t WALKER_ERROR: 0x%lx\n", REG_GET_FIELD(status, GCVM_L2_PROTECTION_FAULT_STATUS, WALKER_ERROR)); - dev_err(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n", + dev_info(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n", REG_GET_FIELD(status, GCVM_L2_PROTECTION_FAULT_STATUS, PERMISSION_FAULTS)); - dev_err(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", + dev_info(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", REG_GET_FIELD(status, GCVM_L2_PROTECTION_FAULT_STATUS, MAPPING_ERROR)); - dev_err(adev->dev, "\t RW: 0x%lx\n", + dev_info(adev->dev, "\t RW: 0x%lx\n", REG_GET_FIELD(status, GCVM_L2_PROTECTION_FAULT_STATUS, RW));}
drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c +7
-6@@ -82,25 +82,25 @@ gfxhub_v2_1_print_l2_protection_fault_status(struct amdgpu_device *adev, u32 cid = REG_GET_FIELD(status, GCVM_L2_PROTECTION_FAULT_STATUS, CID); - dev_err(adev->dev, + dev_info(adev->dev, "GCVM_L2_PROTECTION_FAULT_STATUS:0x%08X\n", status); - dev_err(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n", + dev_info(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n", cid >= ARRAY_SIZE(gfxhub_client_ids) ? "unknown" : gfxhub_client_ids[cid], cid); - dev_err(adev->dev, "\t MORE_FAULTS: 0x%lx\n", + dev_info(adev->dev, "\t MORE_FAULTS: 0x%lx\n", REG_GET_FIELD(status, GCVM_L2_PROTECTION_FAULT_STATUS, MORE_FAULTS)); - dev_err(adev->dev, "\t WALKER_ERROR: 0x%lx\n", + dev_info(adev->dev, "\t WALKER_ERROR: 0x%lx\n", REG_GET_FIELD(status, GCVM_L2_PROTECTION_FAULT_STATUS, WALKER_ERROR)); - dev_err(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n", + dev_info(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n", REG_GET_FIELD(status, GCVM_L2_PROTECTION_FAULT_STATUS, PERMISSION_FAULTS)); - dev_err(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", + dev_info(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", REG_GET_FIELD(status, GCVM_L2_PROTECTION_FAULT_STATUS, MAPPING_ERROR)); - dev_err(adev->dev, "\t RW: 0x%lx\n", + dev_info(adev->dev, "\t RW: 0x%lx\n", REG_GET_FIELD(status, GCVM_L2_PROTECTION_FAULT_STATUS, RW));}
drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c +1
-0@@ -148,14 +148,14 @@ static int gmc_v10_0_process_interrupt(structamdgpu_device *adev, memset(&task_info, 0, sizeof(struct amdgpu_task_info)); amdgpu_vm_get_task_info(adev, entry->pasid, &task_info); - dev_err(adev->dev, + dev_info(adev->dev,"[%s] page fault (src_id:%u ring:%u vmid:%u pasid:%u, "
"for process %s pid %d thread %s pid %d)\n",
entry->vmid_src? "mmhub" : "gfxhub",
entry->src_id,entry->ring_id,entry->vmid,
entry->pasid, task_info.process_name, task_info.tgid, task_info.task_name, task_info.pid);
- dev_err(adev->dev, " in page starting at address 0x%016llx from
client 0x%x (%s)\n",
+ dev_info(adev->dev, " in page starting at address 0x%016llx
from client 0x%x (%s)\n",
addr,entry->client_id,
soc15_ih_clientid_name[entry->client_id]);
diff --gita/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.cb/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.cindexec291d28edff..3d830fd7706b100644drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c +1
-0@@ -620,7 +620,7 @@ static void gmc_v6_0_vm_decode_fault(structamdgpu_device *adev, mc_id = REG_GET_FIELD(status, VM_CONTEXT1_PROTECTION_FAULT_STATUS, MEMORY_CLIENT_ID); - dev_err(adev->dev, "VM fault (0x%02x, vmid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", + dev_info(adev->dev, "VM fault (0x%02x, vmid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", protections, vmid, addr,
REG_GET_FIELD(status, VM_CONTEXT1_PROTECTION_FAULT_STATUS, MEMORY_CLIENT_RW)?drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c +3
-3@@ -1083,11 +1083,11 @@ static int gmc_v6_0_process_interrupt(structamdgpu_device *adev, gmc_v6_0_set_fault_enable_default(adev, false); if (printk_ratelimit()) { - dev_err(adev->dev, "GPU fault detected: %d 0x%08x\n", + dev_info(adev->dev, "GPU fault detected: %d 0x%08x\n", entry->src_id, entry->src_data[0]); - dev_err(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n", + dev_info(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n", addr); - dev_err(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n", + dev_info(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_STATUS0x%08X\n",
status);gmc_v6_0_vm_decode_fault(adev, status, addr, 0); }drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c +1
-0@@ -781,7 +781,7 @@ static void gmc_v7_0_vm_decode_fault(struct amdgpu_device *adev, u32 status, mc_id = REG_GET_FIELD(status, VM_CONTEXT1_PROTECTION_FAULT_STATUS, MEMORY_CLIENT_ID); - dev_err(adev->dev, "VM fault (0x%02x, vmid %d, pasid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", + dev_info(adev->dev, "VM fault (0x%02x, vmid %d, pasid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", protections, vmid, pasid, addr,
REG_GET_FIELD(status, VM_CONTEXT1_PROTECTION_FAULT_STATUS, MEMORY_CLIENT_RW)?drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c +3
-3@@ -1286,11 +1286,11 @@ static int gmc_v7_0_process_interrupt(structamdgpu_device *adev, gmc_v7_0_set_fault_enable_default(adev, false); if (printk_ratelimit()) { - dev_err(adev->dev, "GPU fault detected: %d 0x%08x\n", + dev_info(adev->dev, "GPU fault detected: %d 0x%08x\n", entry->src_id, entry->src_data[0]); - dev_err(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n", + dev_info(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n", addr); - dev_err(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n", + dev_info(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_STATUS0x%08X\n",
status);
gmc_v7_0_vm_decode_fault(adev, status, addr, mc_client, entry->pasid);drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +1
-0@@ -1021,7 +1021,7 @@ static void gmc_v8_0_vm_decode_fault(struct amdgpu_device *adev, u32 status, mc_id = REG_GET_FIELD(status, VM_CONTEXT1_PROTECTION_FAULT_STATUS, MEMORY_CLIENT_ID); - dev_err(adev->dev, "VM fault (0x%02x, vmid %d, pasid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", + dev_info(adev->dev, "VM fault (0x%02x, vmid %d, pasid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", protections, vmid, pasid, addr,
REG_GET_FIELD(status, VM_CONTEXT1_PROTECTION_FAULT_STATUS, MEMORY_CLIENT_RW)?drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +2
-2@@ -1466,12 +1466,12 @@ static int gmc_v8_0_process_interrupt(structamdgpu_device *adev, memset(&task_info, 0, sizeof(struct amdgpu_task_info)); amdgpu_vm_get_task_info(adev, entry->pasid, &task_info); - dev_err(adev->dev, "GPU fault detected: %d 0x%08x for process %s pid %d thread %s pid %d\n", + dev_info(adev->dev, "GPU fault detected: %d 0x%08x for process %s pid %d thread %s pid %d\n", entry->src_id, entry->src_data[0], task_info.process_name, task_info.tgid, task_info.task_name, task_info.pid); - dev_err(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n", + dev_info(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n", addr); - dev_err(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_STATUS0x%08X\n",
+ dev_info(adev->dev, "VM_CONTEXT1_PROTECTION_FAULT_STATUS0x%08X\n",
status);
gmc_v8_0_vm_decode_fault(adev, status, addr, mc_client, entry->pasid);drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +1
-0@@ -582,14 +582,14 @@ static int gmc_v9_0_process_interrupt(structamdgpu_device *adev, memset(&task_info, 0, sizeof(struct amdgpu_task_info)); amdgpu_vm_get_task_info(adev, entry->pasid, &task_info); - dev_err(adev->dev, + dev_info(adev->dev,"[%s] %s page fault (src_id:%u ring:%u vmid:%u "
"pasid:%u, for process %s pid %d thread %s pid %d)\n",
hub_name,retry_fault? "retry" : "no-retry",
entry->src_id,entry->ring_id,entry->vmid,
entry->pasid, task_info.process_name, task_info.tgid, task_info.task_name, task_info.pid);
- dev_err(adev->dev, " in page starting at address 0x%016llx from
IH client 0x%x (%s)\n",
+ dev_info(adev->dev, " in page starting at address 0x%016llx
from IH client 0x%x (%s)\n",
addr,entry->client_id, soc15_ih_clientid_name[entry->client_id]);
@@ -611,11 +611,11 @@ static int gmc_v9_0_process_interrupt(structamdgpu_device*adev,WREG32_P(hub->vm_l2_pro_fault_cntl, 1, ~1);
- dev_err(adev->dev,
+ dev_info(adev->dev,
"VM_L2_PROTECTION_FAULT_STATUS:0x%08X\n",
status);
if (hub == &adev->vmhub[AMDGPU_GFXHUB_0]) {
- dev_err(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n",
+ dev_info(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n",
cid >=ARRAY_SIZE(gfxhub_client_ids)? "unknown" :
gfxhub_client_ids[cid],
cid);drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +6
-6@@ -648,22 +648,22 @@ static int gmc_v9_0_process_interrupt(structamdgpu_device *adev, mmhub_cid = NULL; break; } - dev_err(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n", + dev_info(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n", mmhub_cid ? mmhub_cid : "unknown", cid); } - dev_err(adev->dev, "\t MORE_FAULTS: 0x%lx\n", + dev_info(adev->dev, "\t MORE_FAULTS: 0x%lx\n", REG_GET_FIELD(status, VM_L2_PROTECTION_FAULT_STATUS, MORE_FAULTS)); - dev_err(adev->dev, "\t WALKER_ERROR: 0x%lx\n", + dev_info(adev->dev, "\t WALKER_ERROR: 0x%lx\n", REG_GET_FIELD(status, VM_L2_PROTECTION_FAULT_STATUS, WALKER_ERROR)); - dev_err(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n", + dev_info(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n", REG_GET_FIELD(status, VM_L2_PROTECTION_FAULT_STATUS, PERMISSION_FAULTS)); - dev_err(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", + dev_info(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", REG_GET_FIELD(status, VM_L2_PROTECTION_FAULT_STATUS, MAPPING_ERROR)); - dev_err(adev->dev, "\t RW: 0x%x\n", rw); + dev_info(adev->dev, "\t RW: 0x%x\n", rw); return 0; }diff --git
a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.cb/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.cindex636abd855686..ec8c8b2cab36100644drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c +1
-0@@ -150,7 +150,7 @@ mmhub_v2_0_print_l2_protection_fault_status(struct amdgpu_device *adev, rw = REG_GET_FIELD(status, MMVM_L2_PROTECTION_FAULT_STATUS, RW); - dev_err(adev->dev, + dev_info(adev->dev, "MMVM_L2_PROTECTION_FAULT_STATUS:0x%08X\n", status);switch (adev->ip_versions[MMHUB_HWIP][0]) {
drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c +6
-6@@ -169,21 +169,21 @@ mmhub_v2_0_print_l2_protection_fault_status(struct amdgpu_device *adev, mmhub_cid = NULL; break; } - dev_err(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n", + dev_info(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n", mmhub_cid ? mmhub_cid : "unknown", cid); - dev_err(adev->dev, "\t MORE_FAULTS: 0x%lx\n", + dev_info(adev->dev, "\t MORE_FAULTS: 0x%lx\n", REG_GET_FIELD(status, MMVM_L2_PROTECTION_FAULT_STATUS, MORE_FAULTS)); - dev_err(adev->dev, "\t WALKER_ERROR: 0x%lx\n", + dev_info(adev->dev, "\t WALKER_ERROR: 0x%lx\n", REG_GET_FIELD(status, MMVM_L2_PROTECTION_FAULT_STATUS, WALKER_ERROR)); - dev_err(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n", + dev_info(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n", REG_GET_FIELD(status, MMVM_L2_PROTECTION_FAULT_STATUS, PERMISSION_FAULTS)); - dev_err(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", + dev_info(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", REG_GET_FIELD(status, MMVM_L2_PROTECTION_FAULT_STATUS, MAPPING_ERROR)); - dev_err(adev->dev, "\t RW: 0x%x\n", rw); + dev_info(adev->dev, "\t RW: 0x%x\n", rw); } static void mmhub_v2_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c +1
-0@@ -87,7 +87,7 @@ mmhub_v2_3_print_l2_protection_fault_status(structamdgpu_device *adev, rw = REG_GET_FIELD(status, MMVM_L2_PROTECTION_FAULT_STATUS, RW); - dev_err(adev->dev, + dev_info(adev->dev, "MMVM_L2_PROTECTION_FAULT_STATUS:0x%08X\n", status);switch (adev->ip_versions[MMHUB_HWIP][0]) {
drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c +6
-6@@ -100,21 +100,21 @@ mmhub_v2_3_print_l2_protection_fault_status(struct amdgpu_device *adev, mmhub_cid = NULL; break; } - dev_err(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n", + dev_info(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n", mmhub_cid ? mmhub_cid : "unknown", cid); - dev_err(adev->dev, "\t MORE_FAULTS: 0x%lx\n", + dev_info(adev->dev, "\t MORE_FAULTS: 0x%lx\n", REG_GET_FIELD(status, MMVM_L2_PROTECTION_FAULT_STATUS, MORE_FAULTS)); - dev_err(adev->dev, "\t WALKER_ERROR: 0x%lx\n", + dev_info(adev->dev, "\t WALKER_ERROR: 0x%lx\n", REG_GET_FIELD(status, MMVM_L2_PROTECTION_FAULT_STATUS, WALKER_ERROR)); - dev_err(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n", + dev_info(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n", REG_GET_FIELD(status, MMVM_L2_PROTECTION_FAULT_STATUS, PERMISSION_FAULTS)); - dev_err(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", + dev_info(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", REG_GET_FIELD(status, MMVM_L2_PROTECTION_FAULT_STATUS, MAPPING_ERROR)); - dev_err(adev->dev, "\t RW: 0x%x\n", rw); + dev_info(adev->dev, "\t RW: 0x%x\n", rw); } static void mmhub_v2_3_setup_vm_pt_regs(struct amdgpu_device *adev,
Am 2022-04-25 um 11:29 schrieb Felix Kuehling:
Quoted 8 lines
Am 2022-04-23 um 13:54 schrieb Christian König:
Quoted 5 lines
Am 23.04.22 um 04:24 schrieb Alex Sierra:
Quoted 2 lines
This is not a kernel error. These logs are caused by VM faults that could not be handled. Typically, generated by user mode applications.
Well it's still a hardware fault which should be logged as an error.
At least in ROCm compute applications, a VM fault does not take down the hardware. It only leads to termination of the process that caused the fault. It's very similar to a segfault in an application, which is not considered a HW fault either.
Turns out, a segfault also prints an error message in dmesg. So I guess the VM fault can remain an error message as well.
In a test that triggers VM faults on purpose, we can add some message to let users know to expect kernel error messages from the test.
Regards,
Felix
Quoted 369 lines
Other processes are completely unaffected. The cause of the error is typically in user mode. I think the general policy is, that user mode errors should not spam the kernel logs with error messages. The messages are useful for debugging application issues, so it's good to have them. But IMHO they should not be error messages. Such messages often lead to spurious bug reports against the kernel for things that aren't really kernel issues.
Regards,
FelixQuoted 363 lines
So I'm absolutely not keen about reducing it to just an information.
Regards,
Christian.Quoted 356 lines
Signed-off-by Alex Sierra <[email protected]>---
drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c| 14 +++++++-------drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c| 14 +++++++------- drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c | 4 ++--drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c | 8 ++++---- drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c | 8 ++++---- drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c | 8 ++++---- drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 20 ++++++++++---------- drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c | 14 +++++++------- drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c | 14 +++++++------- 9 files changed, 52 insertions(+), 52 deletions(-)drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c +7
-6@@ -79,25 +79,25 @@ gfxhub_v2_0_print_l2_protection_fault_status(struct amdgpu_device *adev, u32 cid = REG_GET_FIELD(status, GCVM_L2_PROTECTION_FAULT_STATUS, CID); - dev_err(adev->dev, + dev_info(adev->dev, "GCVM_L2_PROTECTION_FAULT_STATUS:0x%08X\n", status); - dev_err(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n", + dev_info(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n", cid >= ARRAY_SIZE(gfxhub_client_ids) ? "unknown" : gfxhub_client_ids[cid], cid); - dev_err(adev->dev, "\t MORE_FAULTS: 0x%lx\n", + dev_info(adev->dev, "\t MORE_FAULTS: 0x%lx\n", REG_GET_FIELD(status, GCVM_L2_PROTECTION_FAULT_STATUS, MORE_FAULTS)); - dev_err(adev->dev, "\t WALKER_ERROR: 0x%lx\n", + dev_info(adev->dev, "\t WALKER_ERROR: 0x%lx\n", REG_GET_FIELD(status, GCVM_L2_PROTECTION_FAULT_STATUS, WALKER_ERROR)); - dev_err(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n", + dev_info(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n", REG_GET_FIELD(status, GCVM_L2_PROTECTION_FAULT_STATUS, PERMISSION_FAULTS)); - dev_err(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", + dev_info(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", REG_GET_FIELD(status, GCVM_L2_PROTECTION_FAULT_STATUS, MAPPING_ERROR)); - dev_err(adev->dev, "\t RW: 0x%lx\n", + dev_info(adev->dev, "\t RW: 0x%lx\n", REG_GET_FIELD(status, GCVM_L2_PROTECTION_FAULT_STATUS, RW));}
drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c +7
-6@@ -82,25 +82,25 @@ gfxhub_v2_1_print_l2_protection_fault_status(struct amdgpu_device *adev, u32 cid = REG_GET_FIELD(status, GCVM_L2_PROTECTION_FAULT_STATUS, CID); - dev_err(adev->dev, + dev_info(adev->dev, "GCVM_L2_PROTECTION_FAULT_STATUS:0x%08X\n", status); - dev_err(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n", + dev_info(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n", cid >= ARRAY_SIZE(gfxhub_client_ids) ? "unknown" : gfxhub_client_ids[cid], cid); - dev_err(adev->dev, "\t MORE_FAULTS: 0x%lx\n", + dev_info(adev->dev, "\t MORE_FAULTS: 0x%lx\n", REG_GET_FIELD(status, GCVM_L2_PROTECTION_FAULT_STATUS, MORE_FAULTS)); - dev_err(adev->dev, "\t WALKER_ERROR: 0x%lx\n", + dev_info(adev->dev, "\t WALKER_ERROR: 0x%lx\n", REG_GET_FIELD(status, GCVM_L2_PROTECTION_FAULT_STATUS, WALKER_ERROR)); - dev_err(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n", + dev_info(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n", REG_GET_FIELD(status, GCVM_L2_PROTECTION_FAULT_STATUS, PERMISSION_FAULTS)); - dev_err(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", + dev_info(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", REG_GET_FIELD(status, GCVM_L2_PROTECTION_FAULT_STATUS, MAPPING_ERROR)); - dev_err(adev->dev, "\t RW: 0x%lx\n", + dev_info(adev->dev, "\t RW: 0x%lx\n", REG_GET_FIELD(status, GCVM_L2_PROTECTION_FAULT_STATUS, RW));}
drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c +1
-0@@ -148,14 +148,14 @@ static int gmc_v10_0_process_interrupt(structamdgpu_device *adev, memset(&task_info, 0, sizeof(struct amdgpu_task_info)); amdgpu_vm_get_task_info(adev, entry->pasid, &task_info); - dev_err(adev->dev, + dev_info(adev->dev,"[%s] page fault (src_id:%u ring:%u vmid:%u pasid:%u, "
"for process %s pid %d thread %s pid %d)\n",
entry->vmid_src? "mmhub" : "gfxhub",
entry->src_id,entry->ring_id,entry->vmid,
entry->pasid, task_info.process_name, task_info.tgid, task_info.task_name, task_info.pid);
- dev_err(adev->dev, " in page starting at address 0x%016llx
from client 0x%x (%s)\n",
+ dev_info(adev->dev, " in page starting at address 0x%016llx
from client 0x%x (%s)\n",
addr,entry->client_id,
soc15_ih_clientid_name[entry->client_id]);
diff --gita/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.cb/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.cindexec291d28edff..3d830fd7706b100644drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c +1
-0@@ -620,7 +620,7 @@ static void gmc_v6_0_vm_decode_fault(structamdgpu_device *adev, mc_id = REG_GET_FIELD(status, VM_CONTEXT1_PROTECTION_FAULT_STATUS, MEMORY_CLIENT_ID); - dev_err(adev->dev, "VM fault (0x%02x, vmid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", + dev_info(adev->dev, "VM fault (0x%02x, vmid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", protections, vmid, addr,
REG_GET_FIELD(status, VM_CONTEXT1_PROTECTION_FAULT_STATUS, MEMORY_CLIENT_RW)?drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c +3
-3@@ -1083,11 +1083,11 @@ static int gmc_v6_0_process_interrupt(structamdgpu_device *adev, gmc_v6_0_set_fault_enable_default(adev, false); if (printk_ratelimit()) { - dev_err(adev->dev, "GPU fault detected: %d 0x%08x\n", + dev_info(adev->dev, "GPU fault detected: %d 0x%08x\n", entry->src_id, entry->src_data[0]); - dev_err(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n", + dev_info(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n", addr); - dev_err(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n", + dev_info(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_STATUS0x%08X\n",
status);gmc_v6_0_vm_decode_fault(adev, status, addr, 0); }drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c +1
-0@@ -781,7 +781,7 @@ static void gmc_v7_0_vm_decode_fault(struct amdgpu_device *adev, u32 status, mc_id = REG_GET_FIELD(status, VM_CONTEXT1_PROTECTION_FAULT_STATUS, MEMORY_CLIENT_ID); - dev_err(adev->dev, "VM fault (0x%02x, vmid %d, pasid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", + dev_info(adev->dev, "VM fault (0x%02x, vmid %d, pasid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", protections, vmid, pasid, addr,
REG_GET_FIELD(status, VM_CONTEXT1_PROTECTION_FAULT_STATUS, MEMORY_CLIENT_RW)?drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c +3
-3@@ -1286,11 +1286,11 @@ static int gmc_v7_0_process_interrupt(structamdgpu_device *adev, gmc_v7_0_set_fault_enable_default(adev, false); if (printk_ratelimit()) { - dev_err(adev->dev, "GPU fault detected: %d 0x%08x\n", + dev_info(adev->dev, "GPU fault detected: %d 0x%08x\n", entry->src_id, entry->src_data[0]); - dev_err(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n", + dev_info(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n", addr); - dev_err(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n", + dev_info(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_STATUS0x%08X\n",
status);
gmc_v7_0_vm_decode_fault(adev, status, addr, mc_client, entry->pasid);drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +1
-0@@ -1021,7 +1021,7 @@ static void gmc_v8_0_vm_decode_fault(struct amdgpu_device *adev, u32 status, mc_id = REG_GET_FIELD(status, VM_CONTEXT1_PROTECTION_FAULT_STATUS, MEMORY_CLIENT_ID); - dev_err(adev->dev, "VM fault (0x%02x, vmid %d, pasid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", + dev_info(adev->dev, "VM fault (0x%02x, vmid %d, pasid %d) at page %u, %s from '%s' (0x%08x) (%d)\n", protections, vmid, pasid, addr,
REG_GET_FIELD(status, VM_CONTEXT1_PROTECTION_FAULT_STATUS, MEMORY_CLIENT_RW)?drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +2
-2@@ -1466,12 +1466,12 @@ static int gmc_v8_0_process_interrupt(structamdgpu_device *adev, memset(&task_info, 0, sizeof(struct amdgpu_task_info)); amdgpu_vm_get_task_info(adev, entry->pasid, &task_info); - dev_err(adev->dev, "GPU fault detected: %d 0x%08x for process %s pid %d thread %s pid %d\n", + dev_info(adev->dev, "GPU fault detected: %d 0x%08x for process %s pid %d thread %s pid %d\n", entry->src_id, entry->src_data[0], task_info.process_name, task_info.tgid, task_info.task_name, task_info.pid); - dev_err(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n", + dev_info(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n", addr); - dev_err(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_STATUS0x%08X\n",
+ dev_info(adev->dev, "VM_CONTEXT1_PROTECTION_FAULT_STATUS0x%08X\n",
status);
gmc_v8_0_vm_decode_fault(adev, status, addr, mc_client, entry->pasid);drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +1
-0@@ -582,14 +582,14 @@ static int gmc_v9_0_process_interrupt(structamdgpu_device *adev, memset(&task_info, 0, sizeof(struct amdgpu_task_info)); amdgpu_vm_get_task_info(adev, entry->pasid, &task_info); - dev_err(adev->dev, + dev_info(adev->dev,"[%s] %s page fault (src_id:%u ring:%u vmid:%u "
"pasid:%u, for process %s pid %d thread %s pid %d)\n",
hub_name,retry_fault? "retry" : "no-retry",
entry->src_id,entry->ring_id,entry->vmid,
entry->pasid, task_info.process_name, task_info.tgid, task_info.task_name, task_info.pid);
- dev_err(adev->dev, " in page starting at address 0x%016llx
from IH client 0x%x (%s)\n",
+ dev_info(adev->dev, " in page starting at address 0x%016llx
from IH client 0x%x (%s)\n",
addr,entry->client_id, soc15_ih_clientid_name[entry->client_id]);
@@ -611,11 +611,11 @@ static int gmc_v9_0_process_interrupt(structamdgpu_device*adev,WREG32_P(hub->vm_l2_pro_fault_cntl, 1, ~1);
- dev_err(adev->dev,
+ dev_info(adev->dev,
"VM_L2_PROTECTION_FAULT_STATUS:0x%08X\n",
status);
if (hub == &adev->vmhub[AMDGPU_GFXHUB_0]) {
- dev_err(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n",
+ dev_info(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n",
cid >=ARRAY_SIZE(gfxhub_client_ids)? "unknown" :
gfxhub_client_ids[cid],
cid);drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +6
-6@@ -648,22 +648,22 @@ static int gmc_v9_0_process_interrupt(structamdgpu_device *adev, mmhub_cid = NULL; break; } - dev_err(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n", + dev_info(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n", mmhub_cid ? mmhub_cid : "unknown", cid); } - dev_err(adev->dev, "\t MORE_FAULTS: 0x%lx\n", + dev_info(adev->dev, "\t MORE_FAULTS: 0x%lx\n", REG_GET_FIELD(status, VM_L2_PROTECTION_FAULT_STATUS, MORE_FAULTS)); - dev_err(adev->dev, "\t WALKER_ERROR: 0x%lx\n", + dev_info(adev->dev, "\t WALKER_ERROR: 0x%lx\n", REG_GET_FIELD(status, VM_L2_PROTECTION_FAULT_STATUS, WALKER_ERROR)); - dev_err(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n", + dev_info(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n", REG_GET_FIELD(status, VM_L2_PROTECTION_FAULT_STATUS, PERMISSION_FAULTS)); - dev_err(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", + dev_info(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", REG_GET_FIELD(status, VM_L2_PROTECTION_FAULT_STATUS, MAPPING_ERROR)); - dev_err(adev->dev, "\t RW: 0x%x\n", rw); + dev_info(adev->dev, "\t RW: 0x%x\n", rw); return 0; }diff --git
a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.cb/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.cindex636abd855686..ec8c8b2cab36100644drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c +1
-0@@ -150,7 +150,7 @@ mmhub_v2_0_print_l2_protection_fault_status(struct amdgpu_device *adev, rw = REG_GET_FIELD(status, MMVM_L2_PROTECTION_FAULT_STATUS, RW); - dev_err(adev->dev, + dev_info(adev->dev, "MMVM_L2_PROTECTION_FAULT_STATUS:0x%08X\n", status);switch (adev->ip_versions[MMHUB_HWIP][0]) {
drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c +6
-6@@ -169,21 +169,21 @@ mmhub_v2_0_print_l2_protection_fault_status(struct amdgpu_device *adev, mmhub_cid = NULL; break; } - dev_err(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n", + dev_info(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n", mmhub_cid ? mmhub_cid : "unknown", cid); - dev_err(adev->dev, "\t MORE_FAULTS: 0x%lx\n", + dev_info(adev->dev, "\t MORE_FAULTS: 0x%lx\n", REG_GET_FIELD(status, MMVM_L2_PROTECTION_FAULT_STATUS, MORE_FAULTS)); - dev_err(adev->dev, "\t WALKER_ERROR: 0x%lx\n", + dev_info(adev->dev, "\t WALKER_ERROR: 0x%lx\n", REG_GET_FIELD(status, MMVM_L2_PROTECTION_FAULT_STATUS, WALKER_ERROR)); - dev_err(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n", + dev_info(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n", REG_GET_FIELD(status, MMVM_L2_PROTECTION_FAULT_STATUS, PERMISSION_FAULTS)); - dev_err(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", + dev_info(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", REG_GET_FIELD(status, MMVM_L2_PROTECTION_FAULT_STATUS, MAPPING_ERROR)); - dev_err(adev->dev, "\t RW: 0x%x\n", rw); + dev_info(adev->dev, "\t RW: 0x%x\n", rw); } static void mmhub_v2_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c +1
-0@@ -87,7 +87,7 @@ mmhub_v2_3_print_l2_protection_fault_status(structamdgpu_device *adev, rw = REG_GET_FIELD(status, MMVM_L2_PROTECTION_FAULT_STATUS, RW); - dev_err(adev->dev, + dev_info(adev->dev, "MMVM_L2_PROTECTION_FAULT_STATUS:0x%08X\n", status);switch (adev->ip_versions[MMHUB_HWIP][0]) {
drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c +6
-6@@ -100,21 +100,21 @@ mmhub_v2_3_print_l2_protection_fault_status(struct amdgpu_device *adev, mmhub_cid = NULL; break; } - dev_err(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n", + dev_info(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n", mmhub_cid ? mmhub_cid : "unknown", cid); - dev_err(adev->dev, "\t MORE_FAULTS: 0x%lx\n", + dev_info(adev->dev, "\t MORE_FAULTS: 0x%lx\n", REG_GET_FIELD(status, MMVM_L2_PROTECTION_FAULT_STATUS, MORE_FAULTS)); - dev_err(adev->dev, "\t WALKER_ERROR: 0x%lx\n", + dev_info(adev->dev, "\t WALKER_ERROR: 0x%lx\n", REG_GET_FIELD(status, MMVM_L2_PROTECTION_FAULT_STATUS, WALKER_ERROR)); - dev_err(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n", + dev_info(adev->dev, "\t PERMISSION_FAULTS: 0x%lx\n", REG_GET_FIELD(status, MMVM_L2_PROTECTION_FAULT_STATUS, PERMISSION_FAULTS)); - dev_err(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", + dev_info(adev->dev, "\t MAPPING_ERROR: 0x%lx\n", REG_GET_FIELD(status, MMVM_L2_PROTECTION_FAULT_STATUS, MAPPING_ERROR)); - dev_err(adev->dev, "\t RW: 0x%x\n", rw); + dev_info(adev->dev, "\t RW: 0x%x\n", rw); } static void mmhub_v2_3_setup_vm_pt_regs(struct amdgpu_device *adev,