This is the mail archive of the
cygwin
mailing list for the Cygwin project.
Re: crash on latest cygwin snapshot
On Jun 25 09:34, marco atzeri wrote:
> On 6/25/2012 6:31 AM, Christopher Faylor wrote:
> >On Sun, Jun 24, 2012 at 11:30:02PM +0200, marco atzeri wrote:
> >>Cgf, Corinna,
> >>
> >>while building latest atlas I hit a bug that is crashing all the
> >>running cygwin process.
> >>It crashes on 20120619 snapshot and also latest cvs source.
> >>No issue on 20120611 snapshot.
> >>
> >>The binary to replicate and a extract of the strace's is at
> >>http://matzeri.altervista.org/works/atlas/
> >>
> >>on W7/64 just run this command
> >>---------------------------------------------------------------------
> >> ./xprobe_comp -v 2 -o atlconf.txt -O 9 -A 25 -Si nof77 0 -V 448 -Fa
> >>ic '-fno-common' -Fa sm '-fno-common' -Fa dm '-fno-common' -Fa sk
> >>'-fno-common' -Fa dk '-fno-common' -Fa xc '-fno-common' -Fa gc
> >>'-fno-common' -Fa if '-fno-common' -b 32 -d b .
> >>----------------------------------------------------------------------
> >>
> >>no stackdump is generated.
> >
> >I think I need a better test case.
>
> I was afraid of that.
> xprobe_comp is a sytem/processor probe and it is really
> resource consuming during its normal run.
>
> I tried to strace but it produces an enormous amount of data
>
> $ ls -sh1 xprobe_comp.strace*
> 330M xprobe_comp.strace_fail
> 761M xprobe_comp.strace_good
>
> and there is no guarantee that the root cause is traced.
Does the failing strace contain a line like this?
--- Process 3708, exception C0000094 at 00401198
If so, it might give a clue. There's a chance that an exception
generates another exception, so you might have more than one such
line. The first one rules. Did you try from CVS again with cgf's
latest patch?
Corinna
--
Corinna Vinschen Please, send mails regarding Cygwin to
Cygwin Project Co-Leader cygwin AT cygwin DOT com
Red Hat
--
Problem reports: http://cygwin.com/problems.html
FAQ: http://cygwin.com/faq/
Documentation: http://cygwin.com/docs.html
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple