This is the mail archive of the
gcc-testresults@gcc.gnu.org
mailing list for the GCC project.
Re: Results for haifa-enabled 2.96 19991201 (experimental) testsuite on i686-redhat-linux-gnu
- To: vonbrand at sleipnir dot valparaiso dot cl
- Subject: Re: Results for haifa-enabled 2.96 19991201 (experimental) testsuite on i686-redhat-linux-gnu
- From: Manfred Hollstein <mhollstein at cygnus dot com>
- Date: Wed, 8 Dec 1999 10:46:45 +0100 (MET)
- Cc: gcc-testresults at gcc dot gnu dot org
- References: <mhollstein@cygnus.com><14409.34708.736787.287738@saturn.hollstein.net><199912051619.NAA14009@sleipnir.valparaiso.cl>
- Reply-To: Manfred Hollstein <manfred dot h at gmx dot net>
On Sun, 5 December 1999, 13:19:02, vonbrand@sleipnir.valparaiso.cl wrote:
> Manfred Hollstein <mhollstein@cygnus.com> said:
> > What were your BOOT_CFLAGS? I've seen '-fomit-frame-pointer' resulting in
> > similar failures like your's; passing '-fno-omit-frame-pointer' cured it.
>
> BOOT_CFLAGS='-O2 -fomit-frame-pointer -march=i686'
>
> > Haven't sent a bug report so far, though.
>
> Those are standard flags for compiling much stuff on Linux (particularly
> the kernel), so any problem there is _serious_ IMVHO.
Yup, absolutely! But, as only the C++ tests are failing, I guess some
C++ module in libgcc2 gets miscompiled. I'll look into this.
>
> BTW, the compiler build decided that 104 of my header files were broken and
> fixed them (This is Red Hat 6.1, except for binutils-2.9.5.0.22), usually
> everything was deemed fine.
I also do not like running fixincludes on Linux! My usual attempt on this
is this
$ rm -rf gcc/include/{asm,linux,root/usr/src/linux}
> --
> Horst von Brand vonbrand@sleipnir.valparaiso.cl
> Casilla 9G, Viņa del Mar, Chile +56 32 672616
manfred