This is the mail archive of the
mailing list for the GCC project.
Re: PATCH: fix breakage from "[PATCH] Fix genmatch linking"
- From: Richard Biener <rguenther at suse dot de>
- To: Hans-Peter Nilsson <hans-peter dot nilsson at axis dot com>
- Cc: gcc-patches at gcc dot gnu dot org,jakub at redhat dot com
- Date: Sat, 25 Oct 2014 10:08:24 +0200
- Subject: Re: PATCH: fix breakage from "[PATCH] Fix genmatch linking"
- Authentication-results: sourceware.org; auth=none
- References: <201410241647 dot s9OGlAUU020641 at ignucius dot se dot axis dot com>
On October 24, 2014 6:47:10 PM CEST, Hans-Peter Nilsson <firstname.lastname@example.org> wrote:
>> From: Richard Biener <email@example.com>
>> Date: Fri, 24 Oct 2014 09:56:51 +0200
>> On Fri, 24 Oct 2014, Hans-Peter Nilsson wrote:
>> > Still, I don't understand exactly how your patch
>> > introduces build-subdirectories where there were none before.
>> > Maybe that "+all-gcc: maybe-all-build-libcpp" was wrong and
>> > should be different?
>> No, we do need a build-libcpp to build gcc/build/genmatch.
>> Not sure how you got around without a build-libiberty as other
>> gen* programs surely require that.
>Regular cross-configurations got around fine as they used the
>"host"-build libiberty, which for crosses seemed to differ from
>"build"-builds(!) only in that they're built at the objdir top
>instead of objdir/build-<hosttuple>. Crosses *could* still use
>the host libraries, but whatever; we're avoiding a
>cross-or-native-conditional now. I haven't given
>canadian-crosses any thought, maybe they were broken before.
Most definitely. I wonder why we don't use the same trick with regular bootstrap. There I see build- variants of libiberty used for stage1.