This is the mail archive of the
mailing list for the GCC project.
Re: [PATCH] PR fortran/8485: Misc INTEGER*8 fixes
- From: Toon Moene <toon at moene dot indiv dot nluug dot nl>
- To: Roger Sayle <roger at www dot eyesopen dot com>
- Cc: gcc-patches at gcc dot gnu dot org
- Date: Mon, 05 May 2003 13:42:13 +0200
- Subject: Re: [PATCH] PR fortran/8485: Misc INTEGER*8 fixes
- Organization: Moene Computational Physics, Maartensdijk, The Netherlands
- References: <Pine.LNX.email@example.com>
Roger Sayle wrote:
This patch fixes GNATS PR fortran/8485.
These original changes bootstrapped fine, fixing the PR, but alas
resulted in 25 testsuite regressions. It turns out that there are
other bugs in g77, when passing "integer4" constants to the middle-end.
Yeah, the support for all this *4, *8 stuff is really "incomplete" as
the manual states. Thanks for finding and fixing some of the holes !
The following patch has been tested on i686-pc-linux-gnu with a
bootstrap that included the g77 front-end, and regression tested
by running the g77 testsuite. Please don't ask for an exhaustive
set of test cases for INTEGER*8 functionality, adding the PRINT
statement to the testcase below (to confirm that we generate the
correct value) pushed my knowledge of FORTRAN to its limits.
I'll also test it on alphaev67-dec-osf5.1 tomorrow to test the
"trivial" 64-bit code paths. Unfortunately, I need to install
the Tru64 license PAKs to reenable user logins after 1 May :}
Ok for mainline?
It's OK with me for mainline (testcase also). I tried to test it on
alphaev6-unknown-linux-gnu, but mainline doesn't build on that platform
In file included from ../../gcc/gcc/gengenrtl.c:31:
../../gcc/gcc/real.h:96:9: #error "REAL_WIDTH > 6 not supported"
Toon Moene - mailto:firstname.lastname@example.org - phoneto: +31 346 214290
Saturnushof 14, 3738 XG Maartensdijk, The Netherlands
Maintainer, GNU Fortran 77: http://gcc.gnu.org/onlinedocs/g77_news.html
GNU Fortran 95: http://gcc-g95.sourceforge.net/ (under construction)