[Soc-coordination] GCC internal compiler error (was: Weekly report (8th week) - Debian GNU/Hurd Debianish initialization)

Thomas Schwinge thomas at codesourcery.com
Tue Aug 13 11:13:43 UTC 2013


Hi!

On Tue, 13 Aug 2013 12:36:43 +0200, Justus Winter <4winter at informatik.uni-hamburg.de> wrote:
> Quoting Thomas Schwinge (2013-08-13 11:52:25)
> > On Fri, 09 Aug 2013 21:52:16 +0200, 4winter at informatik.uni-hamburg.de wrote:
> > >   In file included from printf-parsewc.c:2:0:
> > >   printf-parsemb.c: In function ‘__parse_one_specwc’:
> > >   printf-parsemb.c:407:1: internal compiler error: Segmentation fault
> > >   Please submit a full bug report,
> > >   with preprocessed source if appropriate.
> > >   See <file:///usr/share/doc/gcc-4.7/README.Bugs> for instructions.
> > >   The bug is not reproducible, so it is likely a hardware or OS problem.
> > > 
> > > A similar error came up like five hours into my libc rebuild.
> > 
> > Should you ever find a way to reproduce this, please send me the
> > preprocessed source file (»-save-temps«), and any relevant GCC details
> > (as written in the README.Bugs file).
> > 
> > Is the input file a particularly big/complex one (also taking into
> > consideration any #include directives)?
> 
> No, it's probably not gccs fault, it might be a problem with my
> virtualization environment or the host system.

My thoughts were that maybe GCC is hitting memory pressure, and an
allocation returns NULL, which it fails to catch.  This might explain the
"arbitrariness".

> I've seen seemingly
> random failures, like dpkg-query dying with SIGILL.

Oh, hmm...

> > > How nice
> > > of gcc to retry the build to figure out whether it is to blame or the
> > > environment
> > 
> > Huh, I didn't know it was doing that.  (Aha, a 2004 patch by Jakub
> > Jelinek that is applied to Debian GCC.)
> 
> I hope the sarcasm came across here. While I see the benefit of being
> notified that something fishy is going on, I'd have been more happy
> with a successful build.

Yeah, I understand.  The reasoning is that GCC detected that something's
fishy, and decided in that case it's better to complain loudly instead of
silently continuing (which you'd then do manually, after taking notice of
the error).


Grüße,
 Thomas
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 489 bytes
Desc: not available
URL: <http://lists.alioth.debian.org/pipermail/soc-coordination/attachments/20130813/ac68d0a0/attachment.sig>


More information about the Soc-coordination mailing list