This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Testsuites
- To: Mike Stump <mrs at windriver dot com>
- Subject: Re: Testsuites
- From: Toon Moene <toon at moene dot indiv dot nluug dot nl>
- Date: Tue, 06 Jun 2000 20:52:04 +0200
- CC: David dot Billinghurst at riotinto dot com dot au, mark at codesourcery dot com,gcc at gcc dot gnu dot org
- Organization: Moene Computational Physics, Maartensdijk, The Netherlands
- References: <200006052339.QAA24863@kankakee.wrs.com>
Mike Stump wrote:
> >>>>> "Billinghurst," == Billinghurst, David (CRTS) <David.Billinghurst@riotinto.com.au> writes:
> Billinghurst,> A couple of questions: * How do you mark a g77 test
> Billinghurst,> as XFAIL * How do you do it for a particular target
> Actually, Fortran was one of the few that had 0 unexpected failures.
> :-( Then someone broke it, and I've never seen it recover.
It's not as bad as you think. The bitsize/bytesize changes broke some
tests - fixed by Zack Weinberg. Then there was the mismatch between the
definition of int and long int in the frontend as compared with libf2c,
which was fixed by Chris Demetriou. Then there are two new testcases by
David Billinghurst, derived from testing with LAPACK 3.0: 20000503-1.f
(which might be a induction variable problem); another one was submitted
by him on the 1st of June, a compile error:
+* ICE in `change_address', at emit-rtl.c:1589 with -O1 and above
+* g77 version 2.96 20000530 (experimental) on
mips-sgi-irix6.5/-mabi=64
+*
--
Toon Moene - mailto:toon@moene.indiv.nluug.nl - phoneto: +31 346 214290
Saturnushof 14, 3738 XG Maartensdijk, The Netherlands
GNU Fortran 77: http://gcc.gnu.org/onlinedocs/g77_news.html
GNU Fortran 95: http://g95.sourceforge.net/ (under construction)