AnsweredAssumed Answered

t2080 machine check investigation

Question asked by Matthew Dahl on Feb 14, 2020
Latest reply on Feb 17, 2020 by Matthew Dahl
Hello all,
My company is facing a problem with executing our highly accelerated testing of a handful of t2080 based systems.
Background, we have delivered successfully a dozen or so of these systems that have all cycled through this highly accelerated test set with out issue.
Issue that we are seeing is a machine check after power cycle of the unit. This issue doesn't happen on every power cycle, during the testing the ambient temperature is either ramped from high temp to low temp or vice versa (believe 15 - 30 degree per minute) . When the temp has settled at high or low the test waits a handful of seconds and then power cycles the unit. Our t2080 is running in a hypervisor environment (topaz based), with a Linux partition (cores 0-5), and a RTEMS partition (cores 6-7). The Linux partition is the manager of the RTEMS partition and starts it during Linux boot.
The Linux partition has completed system init and has begun launching our custom applications when we would receive the machine check (output below).
I have inspected the output and do not see anything SW related that would have caused the machine check. The documentation leads me to believe this is something internal to the SoC and have no ability to correct nor determine the root cause.
Is there better way to determine the cause (some additional registers to inspect, some hardware to be instrumented, etc...)
One last note, once this machine check has occurred, the system will startup correctly on power cycle without changing the physical environment.
At this point when the machine check happens, the Linux partition is running and the RTEMs side was previous launched  and is starting up.

[0] Recursive machine check
[0] powerpc_mchk_interrupt: machine check interrupt!
[0] mcheck registers: mcsr = 80000000, mcssr0 = 4009c0, mcsrr1 = 80001000
[0] machine check
[0] NIP 0x004009c0 MSR 0x80001000 LR 0xc00000000001aaa0 ESR 0x00000080 EXC 1
CTR 0xc00000000001aa7c CR 0x22000088 XER 0x00000000 DEAR 0x8000080080901120 PIR 0
Prev trap level 0
r00 0x00000001  r01 0x004816e0  r02 0x0048d880  r03 0xc000000000bdfd00 
r04 0xc00000003fff5000  r05 0xc00000000001aaa0  r06 0xc00000000001aa7c  r07 0x100000000000000 
r08 0x00000000  r09 0xc000000000b43518  r10 0x00000001  r11 0xc000000000bdc000 
r12 0xc000000000bdfbc8  r13 0x00478b40  r14 0xc000000000bdc000  r15 0xc000000000be91c8 
r16 0xc000000000b1691f  r17 0xc0000000007addc8  r18 0xc000000000bdc040  r19 0xc000000000bdc040 
r20 0x00000001  r21 0xc000000000bdc040  r22 0xc000000000acf4c8  r23 0xc000000000bdc000 
r24 0xc000000000bdc040  r25 0xc000000000bdc000  r26 0xc000000000bdc000  r27 0xc000000000bdc000 
r28 0x00080000  r29 0xc000000000bdc000  r30 0xc000000000bdc000  r31 0xc000000000be9138 
[0] sp 4816e0 0 0 400a84 0
[0] Traceback:
[1] doorbell critical
[1] NIP 0xc00000000001aac4 MSR 0x90029000 LR 0xc00000000001aaa0 ESR 0x00000080 EXC 37
CTR 0xc00000000001aa7c CR 0x22000088 XER 0x00000000 DEAR 0x800003fffff2d058 PIR 1
Prev trap level 0
r00 0x00000001  r01 0xc0000001730fbd40  r02 0xc000000000be1a00  r03 0x00000000 
r04 0x00000000  r05 0x00000001  r06 0x17db51000  r07 0x100000000000000 
r08 0x140000000000000  r09 0xc000000000b43518  r10 0x00000001  r11 0xc0000001730f8000 
r12 0xc0000001730fbc08  r13 0xc00000003fff56c0  r14 0xc0000001730f8000  r15 0xc000000000be91c8 
r16 0xc000000000b1691f  r17 0xc0000000007addc8  r18 0xc0000001730f8040  r19 0xc0000001730f8040 
r20 0x00000001  r21 0xc0000001730f8040  r22 0xc000000000acf4c8  r23 0xc0000001730f8000 
r24 0xc0000001730f8040  r25 0xc0000001730f8000  r26 0xc0000001730f8000  r27 0xc0000001730f8000 
r28 0x00080000  r29 0xc0000001730f8000  r30 0xc0000001730f8000  r31 0xc000000000be9138 
[2] doorbell critical
[2] NIP 0xc00000000001aac4 MSR 0x90029000 LR 0xc00000000001aaa0 ESR 0x00000080 EXC 37
CTR 0xc00000000001aa7c CR 0x22000028 XER 0x00000000 DEAR 0x3fff86af7ae8 PIR 2
Prev trap level 0
r00 0x00000001  r01 0xc0000001730ffd40  r02 0xc000000000be1a00  r03 0x00000000 
r04 0xc0000001730f6150  r05 0x00000002  r06 0x17db71000  r07 0x100000000000000 
r08 0x140000000000000  r09 0xc000000000b43518  r10 0x00000001  r11 0xc0000001730fc000 
r12 0x84a28e22  r13 0xc00000003fff5d80  r14 0xc0000001730fc000  r15 0xc000000000be91c8 
r16 0xc000000000b1691f  r17 0xc0000000007addc8  r18 0xc0000001730fc040  r19 0xc0000001730fc040 
r20 0x00000001  r21 0xc0000001730fc040  r22 0xc000000000acf4c8  r23 0xc0000001730fc000 
r24 0xc0000001730fc040  r25 0xc0000001730fc000  r26 0xc0000001730fc000  r27 0xc0000001730fc000 
r28 0x00080000  r29 0xc0000001730fc000  r30 0xc0000001730fc000  r31 0xc000000000be9138 
[3] doorbell critical
[3] NIP 0xc000000000173b00 MSR 0x90029000 LR 0xc000000000173be4 ESR 0x00000080 EXC 37
CTR 0xc0000000002a8f70 CR 0x24000222 XER 0x20000000 DEAR 0x80000800809e8008 PIR 3
Prev trap level 0
r00 0xc000000000173be4  r01 0xc00000016e98fb50  r02 0xc000000000be1a00  r03 0x00000001 
r04 0x00003e40  r05 0xc000000000b1c340  r06 0x00000000  r07 0xe000000000000000 
r08 0x00000000  r09 0x00000000  r10 0x00000000  r11 0x00000c00 
r12 0x24000224  r13 0xc00000003fff6440  r14 0x1000d3e8  r15 0x00000000 
r16 0x00000000  r17 0x1000d380  r18 0x1000d438  r19 0x00000001 
r20 0x10021420  r21 0x00000001  r22 0x1000cf18  r23 0x100225c8 
r24 0x100e5c48  r25 0x101025c8  r26 0xc0000001732da410  r27 0xc000000172358798 
r28 0xc000000000b3b160  r29 0xc0000001723587f0  r30 0xc00000007a39def0  r31 0xc00000007a39ddf0 
[4] doorbell critical
[4] NIP 0xc00000000001aac4 MSR 0x90029000 LR 0xc00000000001aaa0 ESR 0x00000080 EXC 37
CTR 0xc00000000001aa7c CR 0x22000028 XER 0x00000000 DEAR 0x80000800809e0008 PIR 4
Prev trap level 0
r00 0x00000001  r01 0xc00000017310bd40  r02 0xc000000000be1a00  r03 0x00000000 
r04 0xc000000173100190  r05 0x00000004  r06 0x17dbb1000  r07 0x100000000000000 
r08 0x140000000000000  r09 0xc000000000b43518  r10 0x00000001  r11 0xc000000173108000 
r12 0x84adbe82  r13 0xc00000003fff6b00  r14 0xc000000173108000  r15 0xc000000000be91c8 
r16 0xc000000000b1691f  r17 0xc0000000007addc8  r18 0xc000000173108040  r19 0xc000000173108040 
r20 0x00000001  r21 0xc000000173108040  r22 0xc000000000acf4c8  r23 0xc000000173108000 
r24 0xc000000173108040  r25 0xc000000173108000  r26 0xc000000173108000  r27 0xc000000173108000 
r28 0x00080000  r29 0xc000000173108000  r30 0xc000000173108000  r31 0xc000000000be9138 
[5] doorbell critical
[5] NIP 0xc00000000001aac4 MSR 0x90029000 LR 0xc00000000001aaa0 ESR 0x00000080 EXC 37
CTR 0xc00000000001aa7c CR 0x22000028 XER 0x00000000 DEAR 0x8000080080010e04 PIR 5
Prev trap level 0
r00 0x00000001  r01 0xc000000173113d40  r02 0xc000000000be1a00  r03 0x00000000 
r04 0xc00000017310cd50  r05 0x00000005  r06 0x17dbd1000  r07 0x100000000000000 
r08 0x140000000000000  r09 0xc000000000b43518  r10 0x00000001  r11 0xc000000173110000 
r12 0x84a28e82  r13 0xc00000003fff71c0  r14 0xc000000173110000  r15 0xc000000000be91c8 
r16 0xc000000000b1691f  r17 0xc0000000007addc8  r18 0xc000000173110040  r19 0xc000000173110040 
r20 0x00000001  r21 0xc000000173110040  r22 0xc000000000acf4c8  r23 0xc000000173110000 
r24 0xc000000173110040  r25 0xc000000173110000  r26 0xc000000173110000  r27 0xc000000173110000 
r28 0x00080000  r29 0xc000000173110000  r30 0xc000000173110000  r31 0xc000000000be9138 
[7] doorbell critical
[7] NIP 0x00421090 MSR 0x80029002 LR 0x004210a4 ESR 0x00000000 EXC 37
CTR 0x00000000 CR 0x20000088 XER 0x00000000 DEAR 0x00000000 PIR 7
Prev trap level 0
r00 0x00400500  r01 0x006411a0  r02 0x0048d880  r03 0x7e3bd040 
r04 0x7e3bd040  r05 0x00000000  r06 0x00000000  r07 0x0049d880 
r08 0x00000000  r09 0x00000000  r10 0x00000020  r11 0x00000000 
r12 0x00000000  r13 0x004e8280  r14 0x00000000  r15 0x00000000 
r16 0x00000000  r17 0x00000000  r18 0x00000000  r19 0x00000000 
r20 0x00000000  r21 0x00000000  r22 0x00000000  r23 0x00000000 
r24 0x00000007  r25 0x00472260  r26 0x00000001  r27 0x00641250 
r28 0x00000040  r29 0x0068d278  r30 0x00000001  r31 0x0063d3e0 
[7] sp 6411a0 641250 0 0 0
[7] Traceback: 0xfffffffffffffffc
[6] doorbell critical
[6] NIP 0x0011f3ac MSR 0x1200a200 LR 0x00137394 ESR 0x00000000 EXC 37
CTR 0x0011f3a0 CR 0x20000008 XER 0x00000000 DEAR 0x14041000 PIR 6
Prev trap level 0
r00 0x00137374  r01 0x05089bf0  r02 0x00000000  r03 0x1204a200 
r04 0x050694c0  r05 0x0500a198  r06 0x1200a200  r07 0x0500a1a0 
r08 0x0500a1a4  r09 0x0011f3a0  r10 0x09010001  r11 0x00000000 
r12 0x20000002  r13 0x050375d8  r14 0x00000000  r15 0x00000000 
r16 0x00000000  r17 0x00000000  r18 0x00000000  r19 0x00000000 
r20 0x00000000  r21 0x00000000  r22 0x00000000  r23 0x00000000 
r24 0x00000000  r25 0x00000000  r26 0x00000000  r27 0x00000000 
r28 0x00000000  r29 0x00000000  r30 0x00000000  r31 0x050694c0 

Outcomes