DRA746: [j6-dra74x] GPU Error log

SangGyu Kim

Prodigy 60 points

Part Number: DRA746

Dear Support team,

This is fields issue for the j6-dra74x.

We use GPU for the Around View Monitoring, and after finished boot sequence the below log suddenly printed.

After that, it was repeated every 1 second, so the around view display looked abnormal.

Could you check what is the reason of recovering SGX Hardware?

==================================================================================================================

INFO 2020-09-15 15:00:49.047 [System] PVR_K:(Error): SGXOSTimer() detected SGX lockup (0x1003c tasks) [1828, services4/srvkm/devices/sgx/sgxinit.c]

INFO 2020-09-15 15:00:49.047 [System] PVR_K: HWRecoveryResetSGX: SGX Hardware Recovery triggered by timer

INFO 2020-09-15 15:00:49.047 [System] PVR_K: SGXDumpDebugInfo called from function HWRecoveryResetSGX at line 1688 of services4/srvkm/devices/sgx/sgxinit.c

INFO 2020-09-15 15:00:49.047 [System] PVR_K: SGX debug (SGX_DDK_NTO sgxddk 19 1.9@2291151)

INFO 2020-09-15 15:00:49.047 [System] PVR_K: SGX Register Base Address (Linear): 0x280C0000

INFO 2020-09-15 15:00:49.047 [System] PVR_K: SGX Register Base Address (Physical): 0x56000000

INFO 2020-09-15 15:00:49.047 [System] PVR_K: (P0) EUR_CR_CORE_ID: 01191201

INFO 2020-09-15 15:00:49.047 [System] PVR_K: (P0) EUR_CR_CORE_REVISION: 00010106

INFO 2020-09-15 15:00:49.047 [System] PVR_K: (HYD) EUR_CR_MASTER_BIF_INT_STAT: 00080000

INFO 2020-09-15 15:00:49.047 [System] PVR_K: (HYD) EUR_CR_MASTER_BIF_FAULT: 00000000

INFO 2020-09-15 15:00:49.047 [System] PVR_K: (HYD) EUR_CR_MASTER_CLKGATESTATUS2: 00000005

INFO 2020-09-15 15:00:49.048 [System] PVR_K: (HYD) EUR_CR_MASTER_VDM_PIM_STATUS: 00000110

INFO 2020-09-15 15:00:49.048 [System] PVR_K: (HYD) EUR_CR_MASTER_BIF_BANK_SET: 00000000

INFO 2020-09-15 15:00:49.049 [System] PVR_K: (HYD) EUR_CR_MASTER_EVENT_STATUS: 20000000

INFO 2020-09-15 15:00:49.050 [System] PVR_K: (HYD) EUR_CR_MASTER_EVENT_STATUS2: 00000018

INFO 2020-09-15 15:00:49.050 [System] PVR_K: (HYD) EUR_CR_MASTER_MP_PRIMITIVE: 000003E8

INFO 2020-09-15 15:00:49.051 [System] PVR_K: (HYD) EUR_CR_MASTER_DPM_DPLIST_STATUS: 00000043

INFO 2020-09-15 15:00:49.051 [System] PVR_K: (HYD) EUR_CR_MASTER_DPM_PROACTIVE_PIM_SPEC: 00001FFF

INFO 2020-09-15 15:00:49.051 [System] PVR_K: (HYD) EUR_CR_MASTER_PAGE_MANAGEOP: 00000000

INFO 2020-09-15 15:00:49.051 [System] PVR_K: (HYD) EUR_CR_MASTER_VDM_CONTEXT_STORE_SNAPSHOT: 00000000

INFO 2020-09-15 15:00:49.052 [System] PVR_K: (HYD) EUR_CR_MASTER_VDM_CONTEXT_LOAD_STATUS: 00000000

INFO 2020-09-15 15:00:49.052 [System] PVR_K: (HYD) EUR_CR_MASTER_VDM_CONTEXT_STORE_STREAM: 100A3764

INFO 2020-09-15 15:00:49.052 [System] PVR_K: (HYD) EUR_CR_MASTER_VDM_CONTEXT_STORE_STATUS: 00000000

INFO 2020-09-15 15:00:49.052 [System] PVR_K: (HYD) EUR_CR_MASTER_VDM_CONTEXT_STORE_STATE0: 00000000

INFO 2020-09-15 15:00:49.052 [System] PVR_K: (HYD) EUR_CR_MASTER_VDM_CONTEXT_STORE_STATE1: 00000000

INFO 2020-09-15 15:00:49.052 [System] PVR_K: (HYD) EUR_CR_MASTER_VDM_WAIT_FOR_KICK: 00000001

INFO 2020-09-15 15:00:49.053 [System] PVR_K: (P0) EUR_CR_EVENT_STATUS: 24242500

INFO 2020-09-15 15:00:49.053 [System] PVR_K: (P0) EUR_CR_EVENT_STATUS2: 000000B8

INFO 2020-09-15 15:00:49.053 [System] PVR_K: (P0) EUR_CR_BIF_CTRL: 00000000

INFO 2020-09-15 15:00:49.053 [System] PVR_K: (P0) EUR_CR_BIF_BANK0: 00001007

INFO 2020-09-15 15:00:49.053 [System] PVR_K: (P0) EUR_CR_BIF_INT_STAT: 00092000

INFO 2020-09-15 15:00:49.053 [System] PVR_K: (P0) EUR_CR_BIF_FAULT: F0002011

INFO 2020-09-15 15:00:49.053 [System] PVR_K: (P0) EUR_CR_BIF_MEM_REQ_STAT: 00000000

INFO 2020-09-15 15:00:49.053 [System] PVR_K: (P0) EUR_CR_CLKGATECTL: 002AAAAA

INFO 2020-09-15 15:00:49.053 [System] PVR_K: (P0) EUR_CR_BIF_BANK_SET: 00000000

INFO 2020-09-15 15:00:49.054 [System] PVR_K: (P0) EUR_CR_CLKGATECTL: 002AAAAA

INFO 2020-09-15 15:00:49.054 [System] PVR_K: (P0) EUR_CR_CLKGATESTATUS: 01600887

INFO 2020-09-15 15:00:49.054 [System] PVR_K: (P0) EUR_CR_MTE_CTRL: 000000A2

INFO 2020-09-15 15:00:49.054 [System] PVR_K: (P0) EUR_CR_EVENT_OTHER_PDS_EXEC: F2011E30

INFO 2020-09-15 15:00:49.054 [System] PVR_K: (P0) EUR_CR_EVENT_OTHER_PDS_DATA: 00000005

INFO 2020-09-15 15:00:49.054 [System] PVR_K: (P0) EUR_CR_EVENT_OTHER_PDS_INFO: 00012001

INFO 2020-09-15 15:00:49.184 [System] PVR_K: (P0) EUR_CR_DPM_ZLS_PAGE_THRESHOLD: 00000000

INFO 2020-09-15 15:00:49.184 [System] PVR_K: (P0) EUR_CR_DPM_TA_GLOBAL_LIST: 00000000

INFO 2020-09-15 15:00:49.184 [System] PVR_K: (P0) EUR_CR_DPM_STATE_CONTEXT_ID: 00000000

INFO 2020-09-15 15:00:49.184 [System] PVR_K: (P0) EUR_CR_DPM_CONTEXT_PB_BASE: 00000000

INFO 2020-09-15 15:00:49.184 [System] PVR_K: (P0) EUR_CR_DPM_TA_ALLOC_FREE_LIST_STATUS1: 00000000

INFO 2020-09-15 15:00:49.184 [System] PVR_K: (P0) EUR_CR_DPM_3D_FREE_LIST_STATUS1: 00000000

INFO 2020-09-15 15:00:49.185 [System] PVR_K: (P0) EUR_CR_DPM_TA_ALLOC_FREE_LIST_STATUS2: 00000000

INFO 2020-09-15 15:00:49.185 [System] PVR_K: (P0) EUR_CR_DPM_3D_FREE_LIST_STATUS2: 00000000

INFO 2020-09-15 15:00:49.185 [System] PVR_K: (P0) EUR_CR_DPM_ABORT_STATUS_MTILE: 00000000

INFO 2020-09-15 15:00:49.185 [System] PVR_K: (P0) EUR_CR_DPM_PAGE_STATUS: 00000000

INFO 2020-09-15 15:00:49.185 [System] PVR_K: (P0) EUR_CR_DPM_PAGE: 00000000

INFO 2020-09-15 15:00:49.185 [System] PVR_K: (P0) EUR_CR_DPM_GLOBAL_PAGE_STATUS: 00000000

INFO 2020-09-15 15:00:49.185 [System] PVR_K: (P0) EUR_CR_VDM_CONTEXT_LOAD_STATUS: 00000000

INFO 2020-09-15 15:00:49.185 [System] PVR_K: (P0) EUR_CR_VDM_CONTEXT_STORE_STATUS: 00000000

INFO 2020-09-15 15:00:49.185 [System] PVR_K: (P0) EUR_CR_VDM_TASK_KICK_STATUS: 00000000

INFO 2020-09-15 15:00:49.185 [System] PVR_K: (P0) EUR_CR_VDM_CONTEXT_STORE_STATE0: 00000000

INFO 2020-09-15 15:00:49.185 [System] PVR_K: (P0) EUR_CR_VDM_CONTEXT_STORE_STATE1: 00000000

INFO 2020-09-15 15:00:49.185 [System] PVR_K: (P0) EUR_CR_DPM_REQUESTING: 00000000

INFO 2020-09-15 15:00:49.186 [System] PVR_K: (P0) EUR_CR_DPM_REQUESTING: 00000000

INFO 2020-09-15 15:00:49.186 [System] PVR_K: (P1) EUR_CR_EVENT_STATUS: 04242500

INFO 2020-09-15 15:00:49.186 [System] PVR_K: (P1) EUR_CR_EVENT_STATUS2: 000000B8

INFO 2020-09-15 15:00:49.186 [System] PVR_K: (P1) EUR_CR_BIF_CTRL: 00000000

INFO 2020-09-15 15:00:49.186 [System] PVR_K: (P1) EUR_CR_BIF_BANK0: 00001007

INFO 2020-09-15 15:00:49.186 [System] PVR_K: (P1) EUR_CR_BIF_INT_STAT: 00092000

INFO 2020-09-15 15:00:49.186 [System] PVR_K: (P1) EUR_CR_BIF_FAULT: F0002011

INFO 2020-09-15 15:00:49.186 [System] PVR_K: (P1) EUR_CR_BIF_MEM_REQ_STAT: 00000000

INFO 2020-09-15 15:00:49.187 [System] PVR_K: (P1) EUR_CR_CLKGATECTL: 002AAAAA

INFO 2020-09-15 15:00:49.187 [System] PVR_K: (P1) EUR_CR_BIF_BANK_SET: 00000000

INFO 2020-09-15 15:00:49.187 [System] PVR_K: (P1) EUR_CR_CLKGATECTL: 002AAAAA

INFO 2020-09-15 15:00:49.187 [System] PVR_K: (P1) EUR_CR_CLKGATESTATUS: 01000087

INFO 2020-09-15 15:00:49.187 [System] PVR_K: (P1) EUR_CR_MTE_CTRL: 000000A2

INFO 2020-09-15 15:00:49.187 [System] PVR_K: (P1) EUR_CR_EVENT_OTHER_PDS_EXEC: F2011D60

INFO 2020-09-15 15:00:49.325 [System] PVR_K: (P1) EUR_CR_EVENT_OTHER_PDS_DATA: 00000003

INFO 2020-09-15 15:00:49.325 [System] PVR_K: (P1) EUR_CR_EVENT_OTHER_PDS_INFO: 00012001

INFO 2020-09-15 15:00:49.325 [System] PVR_K: (P1) EUR_CR_DPM_ZLS_PAGE_THRESHOLD: 00000000

INFO 2020-09-15 15:00:49.325 [System] PVR_K: (P1) EUR_CR_DPM_TA_GLOBAL_LIST: 00000000

INFO 2020-09-15 15:00:49.325 [System] PVR_K: (P1) EUR_CR_DPM_STATE_CONTEXT_ID: 00000000

INFO 2020-09-15 15:00:49.325 [System] PVR_K: (P1) EUR_CR_DPM_CONTEXT_PB_BASE: 00000000

INFO 2020-09-15 15:00:49.325 [System] PVR_K: (P1) EUR_CR_DPM_TA_ALLOC_FREE_LIST_STATUS1: 00000000

INFO 2020-09-15 15:00:49.325 [System] PVR_K: (P1) EUR_CR_DPM_3D_FREE_LIST_STATUS1: 00000000

INFO 2020-09-15 15:00:49.325 [System] PVR_K: (P1) EUR_CR_DPM_TA_ALLOC_FREE_LIST_STATUS2: 00000000

INFO 2020-09-15 15:00:49.325 [System] PVR_K: (P1) EUR_CR_DPM_3D_FREE_LIST_STATUS2: 00000000

INFO 2020-09-15 15:00:49.326 [System] PVR_K: (P1) EUR_CR_DPM_ABORT_STATUS_MTILE: 00000000

INFO 2020-09-15 15:00:49.326 [System] PVR_K: (P1) EUR_CR_DPM_PAGE_STATUS: 00000000

INFO 2020-09-15 15:00:49.326 [System] PVR_K: (P1) EUR_CR_DPM_PAGE: 00000000

INFO 2020-09-15 15:00:49.326 [System] PVR_K: (P1) EUR_CR_DPM_GLOBAL_PAGE_STATUS: 00000000

INFO 2020-09-15 15:00:49.326 [System] PVR_K: (P1) EUR_CR_VDM_CONTEXT_LOAD_STATUS: 00000000

INFO 2020-09-15 15:00:49.326 [System] PVR_K: (P1) EUR_CR_VDM_CONTEXT_STORE_STATUS: 00000000

INFO 2020-09-15 15:00:49.326 [System] PVR_K: (P1) EUR_CR_VDM_TASK_KICK_STATUS: 00000000

INFO 2020-09-15 15:00:49.326 [System] PVR_K: (P1) EUR_CR_VDM_CONTEXT_STORE_STATE0: 00000000

INFO 2020-09-15 15:00:49.326 [System] PVR_K: (P1) EUR_CR_VDM_CONTEXT_STORE_STATE1: 00000000

INFO 2020-09-15 15:00:49.326 [System] PVR_K: (P1) EUR_CR_DPM_REQUESTING: 00000000

INFO 2020-09-15 15:00:49.327 [System] PVR_K: Checking EDM memory context (index = 7, PD = 0xb16fd000)

INFO 2020-09-15 15:00:49.327 [System] PVR_K: Found MMU context for page fault 0xf0002000

INFO 2020-09-15 15:00:49.327 [System] PVR_K: GPU memory context is for PID=-1 ()

INFO 2020-09-15 15:00:49.327 [System] PVR_K: No PDE found

INFO 2020-09-15 15:00:49.327 [System] PVR_K: Checking TA memory context (index = 0, PD = 0xb016c000)

INFO 2020-09-15 15:00:49.327 [System] PVR_K: Found MMU context for page fault 0xf0002000

INFO 2020-09-15 15:00:49.327 [System] PVR_K: GPU memory context is for PID=-1 ()

INFO 2020-09-15 15:00:49.327 [System] PVR_K: PDE valid: PTE = 0xaf06100d (PhysAddr = 0xaf061000, valid)

INFO 2020-09-15 15:00:49.463 [System] PVR_K: Checking 3D memory context (index = 1, PD = 0xb1245000)

INFO 2020-09-15 15:00:49.464 [System] PVR_K: Found MMU context for page fault 0xf0002000

INFO 2020-09-15 15:00:49.464 [System] PVR_K: GPU memory context is for PID=-1 ()

INFO 2020-09-15 15:00:49.464 [System] PVR_K: PDE valid: PTE = 0x00000000 (PhysAddr = 0x00000000, Invalid)

INFO 2020-09-15 15:00:49.464 [System] PVR_K: Checking PTLA memory context (index = 0, PD = 0xb016c000)

INFO 2020-09-15 15:00:49.464 [System] PVR_K: Found MMU context for page fault 0xf0002000

INFO 2020-09-15 15:00:49.464 [System] PVR_K: GPU memory context is for PID=-1 ()

INFO 2020-09-15 15:00:49.464 [System] PVR_K: PDE valid: PTE = 0xaf06100d (PhysAddr = 0xaf061000, valid)

INFO 2020-09-15 15:00:49.464 [System] PVR_K: Host Ctl flags= 0000000c

INFO 2020-09-15 15:00:49.464 [System] PVR_K: SGX Host control:

INFO 2020-09-15 15:00:49.465 [System] PVR_K: (HC-0) 0x00000001 0x00000000 0x00000000 0x00000000

INFO 2020-09-15 15:00:49.465 [System] PVR_K: (HC-10) 0x00000001 0x0000000A 0x0002B4B5 0x00000007

INFO 2020-09-15 15:00:49.465 [System] PVR_K: (HC-20) 0x00000000 0x00000003 0x00000000 0x00000363

INFO 2020-09-15 15:00:49.465 [System] PVR_K: (HC-30) 0x000062D3 0x01F5A7BD 0x00000000 0x00000000

INFO 2020-09-15 15:00:49.465 [System] PVR_K: (HC-40) 0x00000000 0x00000000 0x00000000 0x00000000

INFO 2020-09-15 15:00:49.465 [System] PVR_K: (HC-50) 0x00000000 0x00000000 0x00000000 0x00000000

INFO 2020-09-15 15:00:49.465 [System] PVR_K: (HC-60) 0x00000000 0x00000000 0x00000000 0x00000000

INFO 2020-09-15 15:00:49.466 [System] PVR_K: (HC-70) 0x00000000 0x00000000 0x00000000 0x00000000

INFO 2020-09-15 15:00:49.466 [System] PVR_K: (HC-80) 0x00000000 0x00000000 0x00000000 0x37073438

INFO 2020-09-15 15:00:49.466 [System] PVR_K: SGX TA/3D control:

INFO 2020-09-15 15:00:49.466 [System] PVR_K: (T3C-0) 0xF4003000 0xF4003120 0xF4002000 0xF414D200

INFO 2020-09-15 15:00:49.466 [System] PVR_K: (T3C-10) 0x00000000 0x00000000 0x00000000 0xF4002C80

INFO 2020-09-15 15:00:49.466 [System] PVR_K: (T3C-20) 0x00000000 0x00000000 0x00000000 0x00000000

INFO 2020-09-15 15:00:49.466 [System] PVR_K: (T3C-30) 0x00000000 0x00000000 0x00000000 0xF4002C80

INFO 2020-09-15 15:00:49.467 [System] PVR_K: (T3C-40) 0xF4002C80 0xF400FC40 0xF400FC40 0x00000002

INFO 2020-09-15 15:00:49.467 [System] PVR_K: (T3C-50) 0x00000000 0x00000001 0x0000031E 0x00000007

INFO 2020-09-15 15:00:49.501 [System] screen-pvr2d: pvr2d_finish: PVR2DQueryBlitsComplete failed [status:-8]
INFO 2020-09-15 15:00:49.603 [System] PVR_K: (T3C-60) 0x00000000 0x00000000 0xF4000000 0xB16FD000

INFO 2020-09-15 15:00:49.603 [System] PVR_K: (T3C-70) 0xF4161940 0x00000000 0xF411F7A0 0xF4002C80

INFO 2020-09-15 15:00:49.603 [System] PVR_K: (T3C-80) 0xF4178538 0xF411F7A0 0x004C322E 0x004C322E

INFO 2020-09-15 15:00:49.603 [System] PVR_K: (T3C-90) 0x00000000 0x00000000 0x00000000 0x00000000

INFO 2020-09-15 15:00:49.603 [System] PVR_K: (T3C-A0) 0x00000000 0x00000000 0x00000000 0x00000000

INFO 2020-09-15 15:00:49.603 [System] PVR_K: (T3C-B0) 0x00000000 0x00000000 0x00000000 0x00000000

INFO 2020-09-15 15:00:49.603 [System] PVR_K: (T3C-C0) 0x00000000 0x00000000 0x00000185 0x00000185

INFO 2020-09-15 15:00:49.603 [System] PVR_K: (T3C-D0) 0xF4004000 0xF400F820 0xF4002020 0xF411DF00

INFO 2020-09-15 15:00:49.603 [System] PVR_K: (T3C-E0) 0xF411DF00 0x00000000 0x00000000 0x00049017

INFO 2020-09-15 15:00:49.603 [System] PVR_K: (T3C-F0) 0x0003600D 0x00000000 0x00000003 0x00000006

INFO 2020-09-15 15:00:49.604 [System] PVR_K: (T3C-100) 0x60D69780 0x00000000 0xA0F8442E 0x6A1F4A86

INFO 2020-09-15 15:00:49.604 [System] PVR_K: SGX Kernel CCB WO:0x8C RO:0x88

INFO 2020-09-15 15:00:49.604 [System] PVR_K: HWRecoveryResetSGX: SGX Hardware Recovery done

over 4 years ago

0 Hemant Hariyani over 4 years ago

TI__Expert 8385 points

Hello,

This is GPU HW recovery. The GPU locked up and a HW recovery was triggered. There could be several reasons behind this. First thing we need to check is to see if a HW recovery was successful at all. If you had serial terminal connected, can you disconnect it and see if you have a visual impact? Logging HW recovery over serial terminal takes time and cause stutter.

In the meanwhile, I will send these logs to Imagination for review.

Regards

Hemant

0 SangGyu Kim over 4 years ago in reply to Hemant Hariyani

Prodigy 60 points

Dear Hemant,

This issue was occurred in fields after the mass production. So, I can't connect serial port to HW board. (This log was from the saved file)

And, this is one time issue, so it was recovered after power on reset.

I am not sure HW recovery was successful because the GPU HW recovery was triggered infinitely before the power on reset.

I need some possible reasons for this issue, ASAP.

(OEM have asked the root cause everyday...)

Thanks,

Sanggyu.

0 SangGyu Kim over 3 years ago in reply to SangGyu Kim

Prodigy 60 points

Dear Hemant,

Could you reply for this issue, ASAP..?

Thanks,

0 James Lee over 3 years ago in reply to SangGyu Kim

TI__Expert 3145 points

Hello SangGyu,

Here is the initial response from IMG:

——-

Logs show that it is a PDS CSC fault in PDSPIXEL_CODEDATA heap.

(EUR_CR_EVENT_STATUS2: 0x00000018) - ta_lockup[locked|0x1], 3d_lockup[locked|0x1],

(EUR_CR_BIF_INT_STAT: 0x00092000) - fault_type[pg_fault|0x1], fault_req[pds|0x2000],

(EUR_CR_BIF_BANK0: 0x00001007) - index_3d[raw|0x1], index_ta[raw|0x0], index_edm[raw|0x7],

(EUR_CR_BIF_FAULT: 0xF0002011) - fault_Addr[raw|0xF0002000], sideband[raw|0x1] csc read, cid[raw|0x1]

——-

Beyond this, they would need more debug that would involve being able to reproduce the issue.

Regards,

James

Processors

Processors forum

DRA746: [j6-dra74x] GPU Error log