lockup on 6.4 while bce in MGETHDR

[John Baldwin]

On Monday 15 June 2009 4:49:06 am pluknet wrote:
> Hi.
> 
> This is on 6.4-S as of April with uptime about 2 months.
> Machine could not be accessed via network, didn't reply on ping.
> Looking at backtrace it's here:
> 
> if_bce.c::bce_get_buf():
>                 /* This is a new mbuf allocation. */
>                 MGETHDR(m_new, M_DONTWAIT, MT_DATA);
> 
> 
> >From brk-seq:
> 
> db> bt
> Tracing pid 31 tid 100034 td 0xc833ad00
> kdb_enter(c097ef95) at kdb_enter+0x2b
> siointr1(c83a3c00) at siointr1+0xce
> siointr(c83a3c00) at siointr+0x5e
> intr_execute_handlers(c80ee4c8,e8963abc,4,e8963b0c,c08cba63,...) at
> intr_execute_handlers+0xe1
> lapic_handle_intr(38) at lapic_handle_intr+0x2e
> Xapic_isr1() at Xapic_isr1+0x33
> --- interrupt, eip = 0xc06a393e, esp = 0xe8963b00, ebp = 0xe8963b0c
> ---
> _mtx_lock_sleep(c0a655c0,c833ad00,0,0,0) at _mtx_lock_sleep+0xb6
> kmem_malloc(c14680c0,1000,101,e8963b8c,c082cadd,...) at
> kmem_malloc+0x328
> page_alloc(c1456000,1000,e8963b7f,101,c8b10016,...) at page_alloc+0x1a
> slab_zalloc(c1456000,101,0,d278ca90,c915aa3c,...) at slab_zalloc+0xdd
> uma_zone_slab(c1456000,1) at uma_zone_slab+0xf0
> uma_zalloc_bucket(c1456000,1) at uma_zalloc_bucket+0x15c
> uma_zalloc_arg(c1456000,d0729e00,1) at uma_zalloc_arg+0x292
> bce_get_buf(c8363000,0,e8963c7c,e8963c7e,e8963c80) at bce_get_buf+0xef
> bce_fill_rx_chain(c8363000,7047,d0729000,cbd86000,59375b37,...) at
> bce_fill_rx_chain+0x48
> bce_rx_intr(c8363000) at bce_rx_intr+0x301
> bce_intr(c8363000) at bce_intr+0xf4
> ithread_execute_handlers(c8337c90,c8230a80) at
> ithread_execute_handlers+0x125
> ithread_loop(c835f860,e8963d38) at ithread_loop+0x55
> fork_exit(c0694a38,c835f860,e8963d38) at fork_exit+0x71
> fork_trampoline() at fork_trampoline+0x8
> --- trap 0x1, eip = 0, esp = 0xe8963d6c, ebp = 0 ---
> 
> db> ps
>   pid  ppid  pgrp   uid   state   wmesg     wchan    cmd
> 17102 17100 15549     0  L      *vm page  0xcce93480 awk
> 17101 17100 15549     0  L      *Giant    0xc8737480 grep
> 17100 17098 15549     0  S       wait     0xc8726a78 sh
> 17098 15570 15549     0  S       wait     0xc98c2a78 sh
> 16935 30771 30771  8382  RL                          httpd
> 16656 16349 16349 10346  S       biord    0xdc389368 php
> 16564 16460 16460 18332  SL      vmpfw    0xc365da98 lynx
> 16563 16351 16351 18332  LL     *vm page  0xcce93480 lynx

I suspect these are your real problems.  Try doing 'show lockchain 16563' 
or 'show lockchain 17101'.

-- 
John Baldwin