RE: 2.49e problems with gcc

[Tim Van Holder]

> > Tim> And as for the autoheader problem: is it supposed to fail
> > Tim> silently (as it does for me),
> >
> > No, it should not.
> OK - I'll look into possible problems (I expect there'll be some issue
> with FDs/redirection again).  For further evidence, is autoheader
> supposed to clean up after itself if traces.sh fails (for me, traces.sh
> gets left behind).
It seems that the DJGPP port of bash hash a bug in its 'set -e';
autoheader uses

if (set -e && . traces.sh) >/dev/null 2>&1; then
  . traces.sh
else
  echo failed
fi

But on the DJGPP port, bash exits completely due to the set -e (instead
of just the subshell exiting).
I've reported this to the DJGPP maintainer of bash.

Note that I think you shouldn't redirect the first run; this masks any
useful error messages sourcing traces.sh might produce.