Mail Archives: pgcc/1998/03/25/03:10:34
Hello everybody,
after trying pgcc on one quantum chemistry code and getting speed
improvements of about 40% (almost all it does is floating point
arithmetic), I also wanted to recompile another one. During the course
of this, I discovered the following problem in pg77-1.0.1 (offical
binaries for Linux) with this Fortran program:
program test
implicit none
integer a(-1:5)
call sub(a)
end
subroutine sub(b)
integer b(-1:*)
end
This is valid Fortran77 (checked in the standard) and compiles without
error using g77/gcc 2.7.2.1, but produces the following error with pg77:
fortran/test> g77 -v test.f
Reading specs from
/usr/lib/gcc-lib/i586-pc-linux-gnulibc1/pgcc-2.90.23/specs
gcc version pgcc-2.90.23 980102 (egcs-1.0.1 release)
/usr/lib/gcc-lib/i586-pc-linux-gnulibc1/pgcc-2.90.23/f771 test.f
-fset-g77-defaults -quiet -dumpbase test.f -version -fversion -o
/tmp/cca01038.s
GNU F77 version pgcc-2.90.23 980102 (egcs-1.0.1 release)
(i586-pc-linux-gnulibc1) compiled by GNU C version pgcc-2.90.23 980102
(egcs-1.0.1 release).
GNU Fortran Front End version 0.5.22-19970929
test.f: In subroutine `sub':
test.f:12:
subroutine sub(b)
^
Array `b' at (^) is too large to handle
Any ideas? Does it use an unsigned int?
Any help would be appreciated, it would be possible to rewrite a couple
of subroutines, but this construct is used too often for that.
I just joined the mailing list and I'm not sure the majodomo got me, so
please also reply directly.
Thanks a lot in advance
Jochen Heyd
--
____________________________________________________________________
_/_/_/ _/_/_/ _/ _/ Jochen Heyd
_/ _/ _/ _/ Manager of Information Systems
_/_/ _/ _/_/_/ Safety & Environmental Technology Group
_/ _/ _/ _/ ETH-Zentrum, CAB D46
_/_/_/ _/ _/ _/ CH-8092 Zurich
Switzerland
Swiss Federal Institute tel: ++41-1-632 71 42
of Technology, Zurich cell: ++41-79-447 64 27
FAX: ++41-1-632 11 89
e-mail:jheyd AT tech DOT chem DOT ethz DOT ch
____________________________________________________________________
- Raw text -