This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug testsuite/38946] [trunk regression] gcc trunk 143562 - Testsuite - gfortran failing tests that worked previously
- From: "rob1weld at aol dot com" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: 27 Jan 2009 23:26:43 -0000
- Subject: [Bug testsuite/38946] [trunk regression] gcc trunk 143562 - Testsuite - gfortran failing tests that worked previously
- References: <bug-38946-13830@http.gcc.gnu.org/bugzilla/>
- Reply-to: gcc-bugzilla at gcc dot gnu dot org
------- Comment #12 from rob1weld at aol dot com 2009-01-27 23:26 -------
(In reply to comment #11)
> (In reply to comment #9)
> Two alternatives are:
I guess there is three.
3. The gcc Testsuite is testing 'outside of gcc' and exercising the
libraries of the Operating System (maybe GNU libc / libmath, maybe not).
Features of the Operating System are checked for in ./configure and
compensated for with: 1. alternate source, 2. "#ifdef"s, 3. Libiberty.
Using the Testsuite to check outside the scope of the Testsuite is not
valid and should be removed _OR_ at the very least we should split this
into a different test and have it excluded on all platforms that do not
support this ONE Operating System specific feature.
That course of action means that other system libraries would also
need to be excluded. You would not want to test Sun's libm for errors,
compare that with any other library, and on that basis determine that
the gcc Testsuite passed or failed.
Rob
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38946