This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Arraysizes in g77
- From: Toon Moene <toon at moene dot indiv dot nluug dot nl>
- To: Alex Yu <yua at yudesigns dot com>, gcc at gcc dot gnu dot org
- Date: Sat, 05 Jan 2002 12:29:54 +0100
- Subject: Re: Arraysizes in g77
- Organization: Moene Computational Physics, Maartensdijk, The Netherlands
- References: <000901c1955a$11048a30$6401000a@monkey> <3C36E20E.7DA71327@moene.indiv.nluug.nl>
I wrote:
> In fact, it's probably an error that (currently) g77-3.1 compiles the
> following:
>
> DIMENSION A(520 000 000)
> END
>
> because this code has the same problem.
Well, in fact it does error when the size is wrong (larger than 2**29-1
reals):
DIMENSION A(540 000 000)
END
but obviously, there something wrong there too:
toon@laptop:~/g77-bugs$ /usr/snp/bin/g77 yu2.f
yu2.f: In program `MAIN__':
yu2.f:1: size of variable `a' is too large
f771: ../../combined/gcc/f/com.c:7550: ffecom_sym_transform_: Assertion
`0 == compare_tree_int ((__extension__ ({ const tree __t = (t); if
(tree_code_type[(int) (((enum tree_code) (__t)->common.code))] != ('d'))
tree_class_check_failed (__t, 'd', "../../combined/gcc/f/com.c", 7550,
__FUNCTION__); __t; })->decl.size_unit), ((st)->size))' failed.
yu2.f:1: confused by earlier errors, bailing out
Somehow it doesn't stop processing right after the fatal error ...
Goes on the bug list ;-)
--
Toon Moene - mailto:toon@moene.indiv.nluug.nl - phoneto: +31 346 214290
Saturnushof 14, 3738 XG Maartensdijk, The Netherlands
Maintainer, GNU Fortran 77: http://gcc.gnu.org/onlinedocs/g77_news.html
Join GNU Fortran 95: http://g95.sourceforge.net/ (under construction)