This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] Re: 3 new fails with builtin patch
- From: Paolo Carlini <pcarlini at unitus dot it>
- To: Benjamin Kosnik <bkoz at redhat dot com>
- Cc: Roger Sayle <roger at eyesopen dot com>, Mark Mitchell<mark at codesourcery dot com>, gcc-patches at gcc dot gnu dot org, libstdc++ at gcc dot gnu dot org
- Date: Sat, 30 Mar 2002 17:25:05 +0100
- Subject: Re: [PATCH] Re: 3 new fails with builtin patch
- References: <Pine.SOL.3.91.1020330081515.21856B-100000@taarna.cygnus.com>
Benjamin Kosnik wrote:
>>If I try to build it "by hand" -O0, the link fails. Please try yourself.
>>
>
>Right. The failures are in the libstdc++ testsuite. Seems like every platform
>that used to have no failures now has 3 or 4...
>
Hi. However, why g++.old-deja/g++.other/builtins9.C does not fail during
g++ regression testing? This I cannot understand because, to be honest,
I don't know in detail how those tests are carried out. Perhaps are
build not link?!? Roger (and Mark) of course meant to link too, but...
>...the libstdc++ testsuite is run at -00, not -O2. Perhaps this should
>change.
>
Notice, however, that running it -O0 helped us catching this problem ;-)
But I agree that -O2 is much more realistic. Ideally, we should iterate
over various optmization levels, perhaps...
Ciao, P.