This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
Re: Fix testing libstdc++-v3 with multilibs
- From: Nick Clifton <nickc at cambridge dot redhat dot com>
- To: Phil Edwards <phil at jaj dot com>
- Cc: dosreis at cmla dot ens-cachan dot fr, drepper at redhat dot com, pme at gcc dot gnu dot org,bkoz at redhat dot com, gcc-patches at gcc dot gnu dot org, libstdc++ at gcc dot gnu dot org
- Date: 29 May 2002 09:57:51 +0100
- Subject: Re: Fix testing libstdc++-v3 with multilibs
- References: <m37klo9qsm.fsf@north-pole.nickc.cambridge.redhat.com><20020528142714.A10652@disaster.basement.lan>
Hi Phil,
> Looks safe to me. (BTW, it's usually easier and better to just mail
> libstdc++ rather than each of the maintainers individually.)
Ah - I looked in the top level MAINTAINERS file to find out who looked
after this kind of thing and it only had your individual names and not
the name of the list. Sorry :-(
> Will "file delete" be okay the first time through, when the gluefile
> doesn't exist?
Yup - it is like doing "rm -f" except that it will work over a remote
link.
> A question: I plan to check in the testsuite-library patch described in
>
> http://gcc.gnu.org/ml/libstdc++/2002-05/msg00295.html
>
> I'd forgotten about gluefile.o, but my subconscious made me write that
> second paragraph. :-) Do you think this library could make gluefile
> requirements easier to maintain?
I doubt it - but then it should not make them harder either. The
gluefile is (or should be) just a simple wrapper to catch abort, exit
and main. It could be archived in your libv3test.a, which might
simplify the linker's command line but other than that I doubt if it
will have any impact on your change.
> Either way, would you be willing to test that patch on your
> multilibbed target?
Certainly. Just ping me when you want some tests run :-)
Cheers
Nick