This is the mail archive of the
egcs-testresults@egcs.cygnus.com
mailing list for the EGCS project.
Re: Results for gcc-2.95 19990618 (prerelease) testsuite on i686-pc-sco3.2v5.0.5
- To: Robert Lipe <robertlipe@usa.net>
- Subject: Re: Results for gcc-2.95 19990618 (prerelease) testsuite on i686-pc-sco3.2v5.0.5
- From: Jeffrey A Law <law@cygnus.com>
- Date: Fri, 02 Jul 1999 03:24:36 -0600
- cc: egcs-testresults@egcs.cygnus.com
- Reply-To: law@cygnus.com
In message <19990701102431.N17422@rjlhome.sco.com>you write:
> > 990127-2.c execution failure -O3 (-fPIC)
> > conversion.c execution failure -O3 (-fPIC)
>
> I don't know what these two are about. Only fails for PIC and only at -O3.
Nevermind the analysis. I tracked these down.
-O3 -fPIC seems to be tweaking the generated code enough to trip the infamous
extended precision vs double precision issues on the x86. Joys. I'll
remove them from the regression list.
jeff