This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug fortran/23925] HDF5 check fails--type conversions
- From: "swhite at aei dot mpg dot de" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: 16 Sep 2005 23:12:07 -0000
- Subject: [Bug fortran/23925] HDF5 check fails--type conversions
- References: <20050916225809.23925.swhite@aei.mpg.de>
- Reply-to: gcc-bugzilla at gcc dot gnu dot org
------- Additional Comments From swhite at aei dot mpg dot de 2005-09-16 23:12 -------
Andrew,
The "make check" command runs through a battery of tests. With gfortran, it
reports 72 failures, while with the Intel compiler, it passes everything.
I'm not sure what you're referring to regarding a simple test case. It's a test
suite. Maybe you could pull out a single item from the suite and call that a case.
The failures seem to be of a certain type. For example, here's an excerpt:
Testing random hw short -> unsigned short conversions PASSED
Testing random hw short -> int conversions *FAILED*
test 1 elmt 0
src = bf 92 -16494
dst = 00 00 00 00 0
I encourage you to download and make this package. It's a very important one
for the working scientists who are your end users.
(And thanks for fixing the internal error bug that stopped it from compiling at
all!)
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23925