Date: Thu, 30 Jan 2003 19:06:31 -0800 From: Peter Wemm <peter@wemm.org> To: John Polstra <jdp@polstra.com> Cc: ia64@freebsd.org Subject: Re: Far enough along for a CVSup port? Message-ID: <20030131030631.0B8A72A89E@canning.wemm.org> In-Reply-To: <200301310244.h0V2i8D9072583@vashon.polstra.com>
next in thread | previous in thread | raw e-mail | index | archive | help
John Polstra wrote:
> In article <20030131024005.78E052A8B4@canning.wemm.org>, Peter Wemm
> <peter@wemm.org> wrote:
> > Oh, one more thing.. possibly important. There is no gdb. If this
> > is a factor, then you need to know about it now rather than later.
>
> Ouch, that does make it harder. I guess I'll still have a go at it.
> Maybe it will just work the first time, ha ha ha. Do we have any
> kind of tool to find out where the program counter was when a
> program died?
dmesg reports all the userland crashes:
fatal user trap (cpu 0):
trap vector = 0x14 (Page Not Present)
cr.iip = 0x60000000402cfa50
cr.ipsr = 0x1013080a6010 (mfl,ic,i,dt,dfh,rt,cpl=3,it,ri=0,bn)
cr.isr = 0x400000000 (code=0,vector=0,r,ei=0)
cr.ifa = 0xdc9a8
cr.iim = 0x3c00e
curthread = 0xe00000003e390a80
pid = 66468, comm = paramtest
Here's how to read it:
iip == program counter which points to the instruction bundle
ipsr == equivalent of eflags. There's some interesting stuff in here, and
some not-so-interesting stuff. Of note, ri=[012] which indicates which
instruction in the current bundle is the one faulting or to be resumed to.
ifa == virtual fault address
While it isn't up to gdb standards, it's better than being completely
in the dark. :-/
Cheers,
-Peter
--
Peter Wemm - peter@wemm.org; peter@FreeBSD.org; peter@yahoo-inc.com
"All of this is for nothing if we don't go to the stars" - JMS/B5
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-ia64" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030131030631.0B8A72A89E>