On Dec 31, 2005, at 11:52 AM, Jeff Whitaker wrote:
Mike Zulauf wrote:
On Dec 31, 2005, at 6:04 AM, Jeff Whitaker wrote:
Mike: I believe I finally have a workaround for this. I've
updated the package to use '-fno-underscoring' when compiling the
library. At least that way all the symbols now have the same
underscoring convention (identical to xlf's). The fortran tests
now pass.
Thanks for working on this, Jeff. Unfortunately, '-fno-
underscoring' conflicts with how other libraries built against
Fortran typically seem to be built.
For example, when I build a code that's linked with the HDF
libraries, I need to use '-fno-underscoring'. On the other hand,
if I link against pgplot, I can't use that option. I
Unless I update pgplot (and all the other fortran callable C libs,
such as netcdf and ncar graphics) to use -fno-underscoring also.
Well, I assume that would work as well - at least for me. I think
you maintain all of the packages on fink that I use that would be
affected. I've got some packages I've installed myself outside of
fink, but I could easily configure them to do the same.
This problem seems to be a result of NCSA hardcoding xlf name-
mangling conventions for all their libs on OS X (I had to use the
same trick to get hdf5 to work with g95).
That's a bit annoying on their part. I wonder why they did that?
Thanks,
Mike
--
Mike Zulauf (2)
[EMAIL PROTECTED]
-------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems? Stop! Download the new AJAX search engine that makes
searching your log files as easy as surfing the web. DOWNLOAD SPLUNK!
http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click
_______________________________________________
Fink-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/fink-users
