Because Precise 3.2.0-79 is missing debug symbols froms ddebs.ubuntu.com I had to compile a 3.2.0-79 kernel in a PPA and expect the symbols to be close to what they used to be in that version.
That led me to a wrong initial analysis that I document here for historical purposes: > 178 2 18 ffff881f716bdc00 RU 0.0 0 0 [khungtaskd] > 3680 2808 38 ffff881f71a5c500 RU 1.6 6629520 4303188 java > 50279 49370 31 ffff883f0c7e8000 RU 0.0 4121160 111120 java > 50757 50322 23 ffff881ef27eae00 RU 0.3 4149720 870892 java crash> bt ffff881ef27eae00 PID: 50757 TASK: ffff881ef27eae00 CPU: 23 COMMAND: "java" #0 [ffff881fbfba6ee0] crash_nmi_callback at ffffffff81031ac9 #1 [ffff881fbfba6ef0] default_do_nmi at ffffffff81666079 #2 [ffff881fbfba6f30] do_nmi at ffffffff816662b0 #3 [ffff881fbfba6f50] nmi at ffffffff81665620 [exception RIP: next_tgid+40] RIP: ffffffff811df248 RSP: ffff881cd4b67da8 RFLAGS: 00000202 RAX: 0000000000000000 RBX: ffffffff81c281a0 RCX: 0000000000000000 RDX: ffff881f72830000 RSI: 0000000000000074 RDI: ffffffff81c281a0 RBP: ffff881cd4b67df8 R8: 000000000000a88d R9: 0000000000000004 R10: ffff883f70922540 R11: 0001f8f579768213 R12: 0000000000000074 R13: ffff881f00000073 R14: ffffffff81c281a0 R15: ffff881f73138000 ORIG_RAX: ffffffffffffffff CS: 0010 SS: 0018 --- <DOUBLEFAULT exception stack> --- #4 [ffff881cd4b67da8] next_tgid at ffffffff811df248 #5 [ffff881cd4b67e00] proc_pid_readdir at ffffffff811e117e #6 [ffff881cd4b67eb0] proc_root_readdir at ffffffff811dbe0a #7 [ffff881cd4b67ee0] vfs_readdir at ffffffff8118e1d0 #8 [ffff881cd4b67f30] sys_getdents at ffffffff8118e4a9 #9 [ffff881cd4b67f80] system_call_fastpath at ffffffff8166d2c2 RIP: 00007f855e94d605 RSP: 00007f853a457bf0 RFLAGS: 00000283 RAX: 000000000000004e RBX: ffffffff8166d2c2 RCX: 0000000000000010 RDX: 0000000000008000 RSI: 00007f8548057980 RDI: 00000000000000fb RBP: 00007f854846b140 R8: 00007f8548057980 R9: 0000000000000008 R10: 0000000000000010 R11: 0000000000000246 R12: 0000000000000016 R13: ffffffffffffffa0 R14: 00007f853a45a4d0 R15: 00007f8548057950 ORIG_RAX: 000000000000004e CS: 0033 SS: 002b 1) khungtaskd is complaining about a hung task 2) task ffff881ef27eae00 is in "next_tgid" from procfs vfs subsystem: We are probably stuck here: rcu_read_lock(); retry: iter.task = NULL; pid = find_ge_pid(iter.tgid, ns); if (pid) { iter.tgid = pid_nr_ns(pid, ns); iter.task = pid_task(pid, PIDTYPE_PID); /* What we to know is if the pid we have find is the * pid of a thread_group_leader. Testing for task * being a thread_group_leader is the obvious thing * todo but there is a window when it fails, due to * the pid transfer logic in de_thread. * * So we perform the straight forward test of seeing * if the pid we have found is the pid of a thread * group leader, and don't worry if the task we have * found doesn't happen to be a thread group leader. * As we don't care in the case of readdir. */ if (!iter.task || !has_group_leader_pid(iter.task)) { iter.tgid += 1; goto retry; } get_task_struct(iter.task); } rcu_read_unlock(); Trying to find a task group leader when reading procfs structure (by the JVM process). I'm still analysing code. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1534413 Title: Precise: lockup during fadvise syscall with POSIX_FADV_DONTNEED To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1534413/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs