Got another dump, with a bunch of

uvm_fault(0x30cec900, 0xf48ad00, 0, 1) -> 1;

followd by
kernel page fault trap 0

show reg:
ds	0x10
es	0x10
fs	0x10
gs	0x10
edi	0x6
esi	0x54
ebp	0xcfa3b1a8 end+0xf61dc5c
ebx	0xf48ad4f6
edx	0xc07c2400 end+0x3a4eb4
ecx	0x1680
eax	0x54
eip	0xc017cf50 pfind+0x28
cs	0x8
eflags	0x10286
esp	0xcfa3b190 end+0xf61dc44
ss	0x10
pfind+0x28: cmpl %esi, 0x3c(%ecx)


trace loops with

pfind(
selwakeup(
logwakeup(
printf(
trap() at trap+539
trap (number 6)

On 19 Feb 2002, enami tsugutomo wrote:

> Hisashi T Fujinaka <htodd@twofifty.com> writes:
>
> > Whoops. I'm putting in diagnostic and debug.
>
> I asked just to make sure the following fault place is same place
> where i imagine.
>
> 	genfs_putpages+268: movl 0x24(%esdi), %eax
>
> If I removed DIAGNOSTIC and UVM_PAGE_TRKOWN from my config, there was
> such instrunctions at line 268.
>
> > Is there a "howto" of using a serial console so I don't have to retype
> > the error messages? :)
>
> Typing `consdev com0' is easy way for few times try (see boot(9)).  Or
> install biosboot_com0.sym for permanent change.
>
> > > > uvmfault(0xc03f5620, 0xc0714000, 0, 1) -> e
> > > > kernel, page fault trap, code 0
> > > > Stopped in pid 9 (ioflush) at genfs_putpages+268: movl 0x24(%esdi), %eax
> > > > db> trace
> > > > genfs_putpages(cf581d40,0,cf581d58,c492ad58,1) at genfs_putpages+0x268
> > > > VOP_PUTPAGES(cfb39d20,0,0,0,0,11,c08a6800,0) at VOP_PUTPAGES+0x49
>
> show reg
> show vnode/f cfb39d20
>
> will give more info.
>
> enami.
>

-- 
Hisashi T Fujinaka - htodd@twofifty.com
BSEE (6/86) + BSChem (3/95) + BAEnglish (8/95) + $2.50 = mocha latte