Mail Archives: cygwin/2009/03/27/08:28:38
On Mar 27 14:21, Corinna Vinschen wrote:
> On Mar 27 07:14, Eric Blake wrote:
> > >> The relevant portion of the strace output is:
> > >> --------------------------------------------------------------------------------
> > >> 847 506042200 [main] git-index-pack 2084 readv: readv (0, 0x22C844, 1) blocking, sigcatchers 1
> > >> 29196 506071396 [main] git-index-pack 2084 seterrno_from_win_error: /ext/build/netrel/src/cygwin-1.7.0-44/winsup/cygwin/fhandler.cc:1747 windows error 998
> > >> 77 506071473 [main] git-index-pack 2084 geterrno_from_win_error: windows error 998 == errno 14
> > >> 134 506071607 [main] git-index-pack 2084 readv: -1 = readv (0, 0x22C844, 1), errno 14
> > >> --------------------------------------------------------------------------------
> > >
> > > This needs help by the git maintainer. Eric?
> >
> > I've been meaning to get enough free time to look into this. It appears
> > to be a heisenbug, where simply stracing things changes the behavior:
> > http://cygwin.com/ml/cygwin/2009-02/msg00567.html
> > I'm also wondering if today's fix to pathconf may have any impact.
>
> No, it hasn't. I tested this with the latest from CVS.
>
> If you have a git version built with CFLAGS=-g, then the git command
> takes long enough to attach to it with GDB from another window before
> the problem actually occurs.
...which should take the heisenbug component out of the picture. I could
easily reproduce the problem this way but I have no explanation for the
weird buffer address.
Corinna
--
Corinna Vinschen Please, send mails regarding Cygwin to
Cygwin Project Co-Leader cygwin AT cygwin DOT com
Red Hat
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
- Raw text -