GPUOpen-LibrariesAndSDKs / MxGPU-Virtualization

MIT License
182 stars 83 forks source link

switch vf failed #1

Open zlxing opened 6 years ago

zlxing commented 6 years ago

log in syslog

Dec 22 01:26:33 ubuntu-sriov kernel: [ 1669.414771] perf interrupt took too long (2580 > 2500), lowering kernel.perf_event_max_sample_rate to 50000 Dec 22 01:31:59 ubuntu-sriov kernel: [ 1994.969792] gim error:(wait_cmd_complete:1643) wait_cmd_complete -- time out after 0.100007496 sec Dec 22 01:31:59 ubuntu-sriov kernel: [ 1994.972044] gim error:(wait_cmd_complete:1650) Cmd = 0x11, Status = 0x11 Dec 22 01:31:59 ubuntu-sriov kernel: [ 1994.973946] gim error:(dump_gpu_status:1271) dump gpu status begin for struct adapter 129:00.00 Dec 22 01:31:59 ubuntu-sriov kernel: [ 1994.975800] gim info:(check_base_addrs:1259) CP_MQD_BASE_ADDR = 0xf4:0f9ff000 Dec 22 01:31:59 ubuntu-sriov kernel: [ 1994.975804] gim error:(dump_gpu_status:1278) CP Ring buffer is not empty, Dec 22 01:31:59 ubuntu-sriov kernel: [ 1994.975847] gim error:(dump_gpu_status:1279) RPTR = 0x00003188, WPTR = 0x00000000 Dec 22 01:31:59 ubuntu-sriov kernel: [ 1994.979473] gim error:(dump_gpu_status:1281) When IDLE_GPU was sent RPTR = 0x00003188, WPTR = 0x00000000 Dec 22 01:31:59 ubuntu-sriov kernel: [ 1994.981322] gim warning:(ring_is_empty:1119) CP_RB_WPTR (0x00000000) != CP_RB_RPTR (0x00003188) Dec 22 01:31:59 ubuntu-sriov kernel: [ 1994.981324] gim error:(dump_gpu_status:1285) At least one ring is active Dec 22 01:31:59 ubuntu-sriov kernel: [ 1994.983165] gim error:(dump_gpu_status:1308) mmGRBM_STATUS = 0xa0003028 Dec 22 01:31:59 ubuntu-sriov kernel: [ 1994.984893] gim error:(dump_gpu_status:1311) mmGRBM_STATUS2 = 0x71000808 Dec 22 01:31:59 ubuntu-sriov kernel: [ 1994.986578] gim error:(dump_gpu_status:1314) mmSRBM_STATUS = 0x20020040 Dec 22 01:31:59 ubuntu-sriov kernel: [ 1994.988238] gim error:(dump_gpu_status:1317) mmSRBM_STATUS2 = 0x0 Dec 22 01:31:59 ubuntu-sriov kernel: [ 1994.989890] gim error:(dump_gpu_status:1320) mmSDMA0_STATUS_REG = 0x46deed57 Dec 22 01:31:59 ubuntu-sriov kernel: [ 1994.991533] gim error:(dump_gpu_status:1323) mmSDMA1_STATUS_REG = 0x46deed57 Dec 22 01:31:59 ubuntu-sriov kernel: [ 1994.993159] gim error:(dump_gpu_status:1337) CP busy Dec 22 01:31:59 ubuntu-sriov kernel: [ 1994.994808] gim error:(dump_gpu_status:1342) RLC busy Dec 22 01:31:59 ubuntu-sriov kernel: [ 1994.996335] gim error:(dump_gpu_status:1345) RLC_STAT = 0x00000003 Dec 22 01:31:59 ubuntu-sriov kernel: [ 1994.997824] gim error:(dump_gpu_status:1347) RLC busy processing a context Dec 22 01:31:59 ubuntu-sriov kernel: [ 1994.997863] gim error:(dump_gpu_status:1348) switch Dec 22 01:31:59 ubuntu-sriov kernel: [ 1995.000766] gim error:(dump_gpu_status:1352) RLC Graphics Power Management Dec 22 01:31:59 ubuntu-sriov kernel: [ 1995.000804] gim error:(dump_gpu_status:1353) unit is busy Dec 22 01:31:59 ubuntu-sriov kernel: [ 1995.003717] gim error:(dump_gpu_status:1364) RLC_GPM_STAT = 0x00000017 Dec 22 01:31:59 ubuntu-sriov kernel: [ 1995.003751] gim error:(dump_gpu_status:1365) - RLC GPM module is busy Dec 22 01:31:59 ubuntu-sriov kernel: [ 1995.006642] gim error:(dump_gpu_status:1372) CP busy Dec 22 01:31:59 ubuntu-sriov kernel: [ 1995.008030] gim error:(dump_gpu_status:1418) CP_CPF_STATUS = 0xb4000223 Dec 22 01:31:59 ubuntu-sriov kernel: [ 1995.009370] gim error:(dump_gpu_status:1420) The write pointer has been updated and Dec 22 01:31:59 ubuntu-sriov kernel: [ 1995.009408] gim error:(dump_gpu_status:1421) the initiated work is still being processed<3>[ 1995.010767] gim error:(dump_gpu_status:1422) by the GFX pipe Dec 22 01:31:59 ubuntu-sriov kernel: [ 1995.012140] gim info:(check_me_cntl:1247) ME/PFP/CE running GPU dump Dec 22 01:31:59 ubuntu-sriov kernel: [ 1995.012142] gim error:(dump_gpu_status:1438) CP_CPF_BUSY_STAT = 0x00000002 Dec 22 01:31:59 ubuntu-sriov kernel: [ 1995.013537] gim error:(dump_gpu_status:1443) dump gpu status end Dec 22 01:31:59 ubuntu-sriov kernel: [ 1995.014946] gim error:(world_switch:2531) Schedule VF0 to VF1 failed;Failure reason is 3, try to reset Dec 22 01:31:59 ubuntu-sriov kernel: [ 1995.016376] gim info:(gim_notify_reset_per_vf:3661) Notify reset to VF0 Dec 22 01:31:59 ubuntu-sriov kernel: [ 1995.016380] gim info:(mailbox_update_index:833) write mmMAILBOX_INDEX: 0x0