Re: Bug#428582: xulrunner_1.8.1.4-2 (mips/unstable): FTBFS: relocation overflow
Aurelien Jarno writes:
> Mike Hommey a écrit :
> > On Tue, Jun 12, 2007 at 08:58:33PM +0200, Mike Hommey <mh@glandium.org> wrote:
> >> Hi,
> >>
> >> I'm following up here a bug report against xulrunner: It failed to
> >> build on mips and mipsel. Here is what Steve Langasek has to say about
> >> the issue:
> >>
> >>> Since 1.8.1.4-1 built successfully on both archs, and there are no changes
> >>> in -2 that seem to explain the build failure, it's possible that this is the
> >>> result of a toolchain regression; gcc-4.1 4.1.2-11 was uploaded Jun 1, with
> >>> a changelog entry of "Update to SVN 20070601", and this was the version of
> >>> g++-4.1 used in at least the mips build. You might want to talk to
> >>> debian-mips and/or debian-gcc about this.
> >
> > FWIW, the only mips related change i can see in gcc-4.1 between -8 and -11 is
> > debian/patches/pr20218-mips.dpatch.
> >
>
> It may not be a mips specific problem. The glibc fails to build on hppa
> when switching from gcc-4.1 -8 to -11 (see #428509). The changes in gcc
> that trigger the failure (currently reduced to about 300 lines) are not
> hppa specific. Maybe the two problems are related.
In http://lists.parisc-linux.org/pipermail/parisc-linux/2007-June/031712.html
John David Anglin identified the backport for 20218 as cause for the
failure for #428509.
#428582 isn't seen anymore on FC's 4.1 branch, but with a different
configuration for xulrunner.
see https://bugzilla.mozilla.org/show_bug.cgi?id=367208
Reply to: