panic: vn_finished_write: neg cnt

From: Kris Kennaway (kris_at_obsecurity.org)
Date: 11/29/05

  • Next message: Mark Atkinson: "Re: problems with -current" 
    Date: Tue, 29 Nov 2005 16:45:11 -0500
To: current@FreeBSD.org

    This was on a quad amd64 machine running 6.0:

    panic() at panic+0x1e6
    vn_finished_write() at vn_finished_write+0x70
    flushbufqueues() at flushbufqueues+0x541
    buf_daemon() at buf_daemon+0x12a
    fork_exit() at fork_exit+0xaa
    fork_trampoline() at fork_trampoline+0xe

    db> show allpcpu
    cpuid = 0
    curthread = 0xffffff032594d980: pid 96815 "gtar"
    curpcb = 0xffffffffbfc09d10
    fpcurthread = none
    idlethread = 0xffffff03e1626be0: pid 14 "idle: cpu0"

    cpuid = 1
    curthread = 0xffffff037e55d4c0: pid 96821 "gtar"
    curpcb = 0xffffffffbfe57d10
    fpcurthread = none
    idlethread = 0xffffff03e1626000: pid 13 "idle: cpu1"

    cpuid = 2
    curthread = 0xffffff02fde554c0: pid 96908 "gtar"
    curpcb = 0xffffffffbff0bd10
    fpcurthread = none
    idlethread = 0xffffff03e1625be0: pid 12 "idle: cpu2"

    cpuid = 3
    curthread = 0xffffff03e152b260: pid 69 "bufdaemon"
    curpcb = 0xffffffffbd78fd10
    fpcurthread = none
    idlethread = 0xffffff03e1625980: pid 11 "idle: cpu3"

    db> wh 96815
    Tracing pid 96815 tid 100159 td 0xffffff032594d980
    ipi_nmi_handler() at ipi_nmi_handler+0x5e
    trap() at trap+0x5a
    nmi_calltrap() at nmi_calltrap+0x5
    --- trap 0x13, rip = 0xffffffff8024eaec, rsp = 0xffffffffbfc09930, rbp = 0xffffffffbfc09960 ---
    _mtx_lock_sleep() at _mtx_lock_sleep+0xdc
    vn_start_write() at vn_start_write+0xc0
    vn_write() at vn_write+0x128
    dofilewrite() at dofilewrite+0x90
    kern_writev() at kern_writev+0x54
    write() at write+0x4b
    syscall() at syscall+0x404
    Xfast_syscall() at Xfast_syscall+0xa8
    --- syscall (4, FreeBSD ELF64, write), rip = 0x8007160f8, rsp = 0x7fffffffe838, rbp = 0x52ca00 ---
    db> wh 96821
    Tracing pid 96821 tid 100243 td 0xffffff037e55d4c0
    ipi_nmi_handler() at ipi_nmi_handler+0x5e
    trap() at trap+0x5a
    nmi_calltrap() at nmi_calltrap+0x5
    --- trap 0x13, rip = 0xffffffff8024eb0e, rsp = 0xffffffffbfe57930, rbp = 0xffffffffbfe57960 ---
    _mtx_lock_sleep() at _mtx_lock_sleep+0xfe
    vn_start_write() at vn_start_write+0xc0
    vn_write() at vn_write+0x128
    dofilewrite() at dofilewrite+0x90
    kern_writev() at kern_writev+0x54
    write() at write+0x4b
    syscall() at syscall+0x404
    Xfast_syscall() at Xfast_syscall+0xa8
    --- syscall (4, FreeBSD ELF64, write), rip = 0x8007160f8, rsp = 0x7fffffffe838, rbp = 0x52b000 ---
    db> wh 96908
    Tracing pid 96908 tid 100310 td 0xffffff02fde554c0
    ipi_nmi_handler() at ipi_nmi_handler+0x5e
    trap() at trap+0x5a
    nmi_calltrap() at nmi_calltrap+0x5
    --- trap 0x13, rip = 0xffffffff803f2a37, rsp = 0xffffffffbff0b710, rbp = 0xffffffffbff0b750 ---
    siointr1() at siointr1+0x97
    siointr() at siointr+0x78
    intr_execute_handlers() at intr_execute_handlers+0xaa
    lapic_handle_intr() at lapic_handle_intr+0x38
    Xapic_isr1() at Xapic_isr1+0x7d
    --- interrupt, rip = 0xffffffff8024eaf7, rsp = 0xffffffffbff0b880, rbp = 0xffffffffbff0b8b0 ---
    _mtx_lock_sleep() at _mtx_lock_sleep+0xe7
    vn_start_write() at vn_start_write+0xc0
    setfown() at setfown+0x3f
    kern_chown() at kern_chown+0x92
    chown() at chown+0x18
    syscall() at syscall+0x404
    Xfast_syscall() at Xfast_syscall+0xa8
    --- syscall (16, FreeBSD ELF64, chown), rip = 0x8006a4b18, rsp = 0x7fffffffe818, rbp = 0x5261c0 ---
    db>

    It was doing a loop of 22 simultaneous tarball extractions into an
    async swap-backed md. I can't dump it, but can leave it in ddb for a
    while in case there's further information I can provide that can be of
    use.

    Kris

  • Next message: Mark Atkinson: "Re: problems with -current"

    Relevant Pages