uma_ref_cnt: vm_fault: fault on nofault entry

[pluknet]

On 3 August 2010 15:29, pluknet <pluknet at gmail.com> wrote:
>
> The second panic seen today on the same box at uma_find_refcnt().
> I'm unsure if it might be caused by Xen hvm setup.

Okey, updating the system to 8.1-R seems to fix this.
13 days of uptime since then, while with 8.0 it panicked every 3-4 days.

>
> db> bt
> Tracing pid 12 tid 100035 td 0xc776f480
> kdb_enter(c0c8dce2,c0c8dce2,c0cab939,c73678b8,0,...) at kdb_enter+0x3a
> panic(c0cab939,c14b3000,1,c73679e8,c73679d8,...) at panic+0x136
> vm_fault(c1490000,c14b3000,1,0,c7a81760,...) at vm_fault+0x197
> trap_pfault(c7bb1748,c7367aa0,c7f7e7bf,c7bb1700,c755c7f8,...) at
> trap_pfault+0x20e
> trap(c7367b04) at trap+0x455
> calltrap() at calltrap+0x6
> --- trap 0xc, eip = 0xc0af336b, esp = 0xc7367b44, ebp = 0xc7367b4c ---
> uma_find_refcnt(c147f700,cbe96800,1,1,cbe96800,...) at uma_find_refcnt+0x5b
> mb_ctor_clust(cbe96800,800,c7f2a700,1,c0dd7b40,...) at mb_ctor_clust+0x9c
> uma_zalloc_arg(c147f700,c7f2a700,1,c776f000,0,...) at uma_zalloc_arg+0x8a
> igb_get_buf(c0dd6e40,c0dd6e40,c0dd6e40,c7367c38,c08a9d00,...) at
> igb_get_buf+0x146
> igb_rxeof(c775f540,0,0,c775f5c0,c7762a00,...) at igb_rxeof+0x2b0
> igb_msix_rx(c7760a00,0,109,e3c02e88,2a1bf,...) at igb_msix_rx+0x29
> intr_event_execute_handlers(c755c7f8,c7762a00,c0c8ab0b,4f6,c7762a70,...)
> at intr_event_execute_handlers+0x14b
> ithread_loop(c7772930,c7367d38,0,0,0,...) at ithread_loop+0x6b
> fork_exit(c0861b20,c7772930,c7367d38) at fork_exit+0x91
> fork_trampoline() at fork_trampoline+0x8
> --- trap 0, eip = 0, esp = 0xc7367d70, ebp = 0 ---
>


-- 
wbr,
pluknet