On Thu, Jul 12, 2001 at 08:06:09PM +0200, Artur Bergman wrote:
> 01-07-12 19.54, skrev Doug MacEachern p� [EMAIL PROTECTED] f�ljande:
>
> > On Thu, 12 Jul 2001, Doug MacEachern wrote:
> >
> >> i don't see any regcomp.h change from you, you mean your regexec.c change?
> >> anyhow, that change did not fix the problem for me.
> >> i just backed out your change (11299), with mine still in there and all
> >> tests pass.
> >
> > actually, 2 tests fail after backing out 11299:
> > lib/extutils.......................make[3]: *** [ExtTest.c] Error 139
> > FAILED at test 4
> > lib/ExtUtils.......................make[3]: *** [ExtTest.c] Error 139
> > FAILED at test 4
> >
> > looking into it.
> >
> >
>
> Ok, this is wierd. I only got failures in pat regexp and regexp-something
> with the PMOP fix patch. I didn't get any segfaults in perl_destruct except
> in those. Backing out 11299 should fail pat.t for perl_destruct.
>
> I can't understand why those SvIVs ened a higher refcount.
I guess it's time for a snapshot so that we have a common ground
to talk about.
--
$jhi++; # http://www.iki.fi/jhi/
# There is this special biologist word we use for 'stable'.
# It is 'dead'. -- Jack Cohen
