Hello everybody,
A X86_64 server repeatedly dumps once a while with the signature:
#4 [ffff88003eaf3d30] do_general_protection at ffffffff8163da9e
#5 [ffff88003eaf3d60] general_protection at ffffffff8163d3a8
[exception RIP: __migration_entry_wait+148]
RIP: ffffffff811c5f64 RSP: ffff88003eaf3e18 RFLAGS: 00010203
RAX: 01ffffffffffffff RBX: ffffea0000000030 RCX: 0000000000000000
RDX: 1e001897fe001880 RSI: ffffea0000000030 RDI: f000c4bff000c4be
RBP: ffff88003eaf3e28 R8: 00003ffffffff000 R9: 00000000000000a9
R10: 0000000000000000 R11: f000c4bff000c4be R12: 1e000297fe001880
R13: ffff8820aa4ac870 R14: ffff883be2b06400 R15: 0000000000000080
ORIG_RAX: ffffffffffffffff CS: 0010 SS: 0000
#6 [ffff88003eaf3e10] __migration_entry_wait at ffffffff811c5eea
#7 [ffff88003eaf3e30] migration_entry_wait at ffffffff811c62b3
#8 [ffff88003eaf3e40] handle_mm_fault at ffffffff81197a12
#9 [ffff88003eaf3ed0] __do_page_fault at ffffffff81640e22
#10 [ffff88003eaf3f28] do_page_fault at ffffffff81641113
#11 [ffff88003eaf3f50] page_fault at ffffffff8163d408
RIP: 00000000022d80ba RSP: 00007fbed15b9990 RFLAGS: 00010202
RAX: 0000000000002000 RBX: 00007fbaeaa00e70 RCX: 00007fbaeaa00e70
RDX: 0000000000000000 RSI: 00000000b94bce12 RDI: 0000000000003b82
RBP: 00007fbed15b99a0 R8: 0000000000000000 R9: 000000000e8a3e7e
R10: 0000000000000000 R11: 00007fbaeaa00e70 R12: 00007fbed419d470
R13: 00007fb678172270 R14: 00007fbd0db98620 R15: 00007fb678505da8
ORIG_RAX: ffffffffffffffff CS: 0033 SS: 002b
It seems that this is a bug. I’m not sure if it has been identified and removed. The customer is adviced to disable numa balancing to work around. I’m waiting for the results from them.
Dashi Cao
曹达士
181 0102 1741
?韬{.n?????%??檩??w?{.n???{炳{????塄}?财??j:+v?? ? ???2??璀??摺?囤??z夸z罐?+?????w棹f
