Subject: Re: i386-current page fault trap in profiled code
To: Frank van der Linden <fvdl@wasabisystems.com>
From: john heasley <heas@shrubbery.net>
List: port-i386
Date: 11/12/2002 10:18:23
Tue, Nov 12, 2002 at 12:47:51PM +0100, Frank van der Linden:
> On Mon, Nov 11, 2002 at 11:48:41PM -0800, john heasley wrote:
> > kernel: page fault trap, code=0
> > Stopped at      0xc0131425:     movl    %eax,         0(%edx)
> > db{1}> trace
> > gdt_desc(c1626000) at 0xc0131425
> > gdt_desc(10,30,10,10,e3420000) at 0xc01024d8
> > Bad frame pointer: 0xc1626000
> 
> Without any symbol information thise trace is useless, unfortunately.
> Do you have a nm -n output for that kernel binary?

i do.  ftp://ftp.shrubbery.net/pub/kernel.sym

btw, thanks!  i've been looking for that glue.  if i've got it, its

c0131264 T ex_intr
c010246f t Xintrioapic4

i've had several page faults in ex_intr (with and w/o offloaded chksums).
i installed a spare fxp in hopes of avoiding it while at ietf.