On Mon, 22 Jan 2018, Jeff Law wrote:
> On 01/22/2018 01:57 AM, Jakub Jelinek wrote:
> > On Mon, Jan 22, 2018 at 09:37:05AM +0100, Richard Biener wrote:
> >>
> >> Currently gcc_release builds GCC (for generating in-tree generated
> >> files) serially - that's prohibitly expensive and takges ages (>4h for
> >> me). I'm using (when I remember to edit gcc_release ...) -j6 without
> >> a problem for some years, thus the following proposal.
> >>
> >> Any recommendation for the default N? 4 might work just fine as well
> >> and no release manager should do without at least 4 cores...
> >
> > Perhaps
> > num_cpus=1
> > if type -p getconf 2>/dev/null; then
> > num_cpus=`getconf _NPROCESSORS_ONLN 2>/dev/null`
> > case "$num_cpus" in
> > '' | 0* | *[!0-9]*) num_cpus=1;;
> > esac
> > fi
> >
> > and "-j$num_cpus" ?
> Seems reasonable to me -- I believe we do something similar on our
> various internal scripts ({redhat,fedora}-rpm-config).
>
> jeff
Ok, I'll commit the following then if it all goes well when doing
the final GCC 7.3 release.
Richard.
Index: maintainer-scripts/gcc_release
===================================================================
--- maintainer-scripts/gcc_release (revision 256973)
+++ maintainer-scripts/gcc_release (working copy)
@@ -209,8 +209,16 @@ EOF
# on at least one platform.
inform "Building compiler"
OBJECT_DIRECTORY=../objdir
+ num_cpus=1
+ if type -p getconf 2>/dev/null; then
+ num_cpus=`getconf _NPROCESSORS_ONLN 2>/dev/null`
+ case "$num_cpus" in
+ '' | 0* | *[!0-9]*) num_cpus=1;;
+ esac
+ fi
contrib/gcc_build -d ${SOURCE_DIRECTORY} -o ${OBJECT_DIRECTORY} \
- -c "--enable-generated-files-in-srcdir --disable-multilib" build || \
+ -c "--enable-generated-files-in-srcdir --disable-multilib" \
+ -m "-j$num_cpus" build || \
error "Could not rebuild GCC"
fi
