As a side note to this thread, I cannot get the current gcc4 package
in 10.4/unstable (which has gfortran as a SplitOff) to compile on
intel. The build log ends like this:
/bin/sh ./libtool --mode=compile /Volumes/Fink/fink.build/
gcc4-4.1.0-1/gcc-4.1.0/darwin/./gcc/xgcc -B/Volumes/Fink/fink.build/
gcc4-4.1.0-1/gcc-4.1.0/darwin/./gcc/ -B/fink/lib/gcc4/i386-apple-
darwin8/bin/ -B/fink/lib/gcc4/i386-apple-darwin8/lib/ -isystem /fink/
lib/gcc4/i386-apple-darwin8/include -isystem /fink/lib/gcc4/i386-
apple-darwin8/sys-include -DHAVE_CONFIG_H -I. -I../../../libgfortran -
I. -iquote../../../libgfortran/io -I../../../libgfortran/../gcc -
I../../../libgfortran/../gcc/config -I../.././gcc -D_GNU_SOURCE -
std=gnu99 -Wall -Wstrict-prototypes -Wmissing-prototypes -Wold-style-
definition -Wextra -Wwrite-strings -O2 -g -O2 -c -o
compile_options.lo `test -f 'runtime/compile_options.c' || echo
'../../../libgfortran/'`runtime/compile_options.c
mkdir .libs
/Volumes/Fink/fink.build/gcc4-4.1.0-1/gcc-4.1.0/darwin/./gcc/xgcc -B/
Volumes/Fink/fink.build/gcc4-4.1.0-1/gcc-4.1.0/darwin/./gcc/ -B/fink/
lib/gcc4/i386-apple-darwin8/bin/ -B/fink/lib/gcc4/i386-apple-darwin8/
lib/ -isystem /fink/lib/gcc4/i386-apple-darwin8/include -isystem /
fink/lib/gcc4/i386-apple-darwin8/sys-include -DHAVE_CONFIG_H -I. -
I../../../libgfortran -I. -iquote../../../libgfortran/io -I../../../
libgfortran/../gcc -I../../../libgfortran/../gcc/config -I../.././gcc
-D_GNU_SOURCE -std=gnu99 -Wall -Wstrict-prototypes -Wmissing-
prototypes -Wold-style-definition -Wextra -Wwrite-strings -O2 -g -O2 -
c ../../../libgfortran/runtime/compile_options.c -fno-common -DPIC -
o .libs/compile_options.o
/Volumes/Fink/fink.build/gcc4-4.1.0-1/gcc-4.1.0/darwin/./gcc/xgcc -B/
Volumes/Fink/fink.build/gcc4-4.1.0-1/gcc-4.1.0/darwin/./gcc/ -B/fink/
lib/gcc4/i386-apple-darwin8/bin/ -B/fink/lib/gcc4/i386-apple-darwin8/
lib/ -isystem /fink/lib/gcc4/i386-apple-darwin8/include -isystem /
fink/lib/gcc4/i386-apple-darwin8/sys-include -DHAVE_CONFIG_H -I. -
I../../../libgfortran -I. -iquote../../../libgfortran/io -I../../../
libgfortran/../gcc -I../../../libgfortran/../gcc/config -I../.././gcc
-D_GNU_SOURCE -std=gnu99 -Wall -Wstrict-prototypes -Wmissing-
prototypes -Wold-style-definition -Wextra -Wwrite-strings -O2 -g -O2 -
c ../../../libgfortran/runtime/compile_options.c -o compile_options.o
>/dev/null 2>&1
make[3]: *** [compile_options.lo] Error 1
make[2]: *** [all] Error 2
make[1]: *** [all-target-libgfortran] Error 2
make: *** [all] Error 2
### execution of /var/tmp/tmp.2.iunI1z failed, exit code 2
-- Dave
On Mar 6, 2006, at 11:56 PM, William Scott wrote:
I've been contacting the maintainers of some software (ccp4) that I
have packaged for fink, and I got this rather disturbing reply this
morning. It sounds like gfortran on intel is problematic. Has
anyone else had experiences, good or bad, with gfortran on intel? I
was actually about to buy one but this put me off.
Hi Bill
I have not been down the fort77 line yet. The results from gcc
4.1.0 on a linux box are not encouraging; we get the os x failures
plus a few more.
Still, this is better than gfortran on an intel Mac. Every code
that uses a maths function fails at runtime with an
EXC_ARITHMETIC, including the gcc testsuite math.f90 . gcc/g++
seem to work nicely though.
Charles
On Monday, March 6, 2006, at 03:00 PM, William Scott wrote:
Hi Charles:
Thanks for confirming. I did get sfcheck to pass the tests when
I recompiled using gfortran -0O. No such luck with the other
two. If the bulk problem is minor, then there is only one
problem: oasis.
As a temporary workaround, would it be possible to use fort77 (or
f2c) to compile it, or would that require recompiling all the
libraries to which it links?
g77-3.4.3 works great, but apparently won't be available on intel
macs.
All the best,
Bill
On Mon, 6 Mar 2006, Charles Ballard wrote:
Dear All,
I can confirm Bill's observations. gfortran 4.1 seems to be a
step back from the pre-release, at least on os x.
bulk.exam : the example fails, but the programs are running
correctly. The tee does not seem to be picking up the fortran
>> output.
oasis_sir/oas.exam : end of file. Looks like a failure to
rewind a file.
sfcheck.exam : unknown reason.
Francois is checking how 4.1 performs under linux.
Given this, if you have the option of using g77, this would be
the way to go.
Charles
CCP4
-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting
language
that extends applications into web and mobile media. Attend the
live webcast
and join the prime developer group breaking into this new coding
territory!
http://sel.as-us.falkag.net/sel?
cmd=lnk&kid=110944&bid=241720&dat=121642
_______________________________________________
Fink-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/fink-devel
-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
_______________________________________________
Fink-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/fink-devel
