This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH, RFC] Enable IBM long double for PPC32 Linux
- From: Jakub Jelinek <jakub at redhat dot com>
- To: David Edelsohn <dje at watson dot ibm dot com>
- Cc: gcc-patches at gcc dot gnu dot org
- Date: Tue, 24 Jan 2006 10:36:30 -0500
- Subject: Re: [PATCH, RFC] Enable IBM long double for PPC32 Linux
- References: <200601231656.k0NGuED31052@makai.watson.ibm.com>
- Reply-to: Jakub Jelinek <jakub at redhat dot com>
On Mon, Jan 23, 2006 at 11:56:14AM -0500, David Edelsohn wrote:
> The appended patch implements this proposal and we would like to
> include it in mainline and GCC 4.1.
I believe in addition to this patch we need to make sure ppc-linux
(as well as ppc64-linux -m32) libgcc is compiled with -mlong-double-128,
to export all the needed TFmode exports. Haven't tried your patch yet,
but saw similar problem on sparc32 a few days ago (though there it will
be harder to solve, as we have an chicken-and-egg issue).
rs6000/t-linux64 does that for ppc64-linux -m64 already:
TARGET_LIBGCC2_CFLAGS = -mno-minimal-toc -fPIC -specs=bispecs
...
# Hack to use -mlong-double-128 just for compiling 64 bit libgcc
mklibgcc: bispecs
bispecs: specs
if [ x$(GCC_FOR_TARGET) -print-multi-os-directory = x../lib ]; then \
sed -e '/cc1_options/{ n; s/$$/ %{m64:-mlong-double-128}/; }' < specs > $@; \
else \
sed -e '/cc1_options/{ n; s/$$/ %{!m32:-mlong-double-128}/; }' < specs > $@; \
fi
I'd say we should just change that to:
TARGET_LIBGCC2_CFLAGS = -mno-minimal-toc -fPIC -mlong-double-128
and
get rid of the rest (and need also
TARGET_LIBGCC2_CFLAGS = -fPIC -mlong-double-128
in newly created rs6000/t-linux).
Jakub