This is the mail archive of the fortran@gcc.gnu.org mailing list for the GNU Fortran project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: more new test failures on Darwin


I posted what the entries were for stat_1 in the previous message.

Executing on host: /sw/src/fink.build/gcc4-4.1.999-20060726/darwin_objdir/gcc/testsuite/gfortran/../../gfortran -B/sw/src/fink.build/gcc4-4
.1.999-20060726/darwin_objdir/gcc/testsuite/gfortran/../../ /sw/src/fink.build/gcc4-4.1.999-20060726/gcc-4.2-20060726/gcc/testsuite/gfortra
n.dg/stat_1.f90   -O0  -std=gnu  -L/sw/src/fink.build/gcc4-4.1.999-20060726/darwin_objdir/powerpc-apple-darwin8/./libgfortran/.libs -L/sw/s
rc/fink.build/gcc4-4.1.999-20060726/darwin_objdir/powerpc-apple-darwin8/./libgfortran/.libs -L/sw/src/fink.build/gcc4-4.1.999-20060726/darw
in_objdir/powerpc-apple-darwin8/./libiberty  -lm   -o ./stat_1.exe    (timeout = 300)
PASS: gfortran.dg/stat_1.f90  -O0  (test for excess errors)
Setting LD_LIBRARY_PATH to .:/sw/src/fink.build/gcc4-4.1.999-20060726/darwin_objdir/powerpc-apple-darwin8/./libgfortran/.libs:/sw/src/fink.
build/gcc4-4.1.999-20060726/darwin_objdir/powerpc-apple-darwin8/./libgfortran/.libs:/sw/src/fink.build/gcc4-4.1.999-20060726/darwin_objdir/
gcc:.:/sw/src/fink.build/gcc4-4.1.999-20060726/darwin_objdir/powerpc-apple-darwin8/./libgfortran/.libs:/sw/src/fink.build/gcc4-4.1.999-2006
0726/darwin_objdir/powerpc-apple-darwin8/./libgfortran/.libs:/sw/src/fink.build/gcc4-4.1.999-20060726/darwin_objdir/gcc
FAIL: gfortran.dg/stat_1.f90  -O0  execution test

The test for excess errors passed but the execution test failed. Frankly I find the output confusing. It looks like exactly the same test except the
second one has LD_LIBRARY_PATH set.

The test for excess errors is a compilation test (it compiles stat_1.f90 at -O0 and tests that no error is issued during compilation other than those supposed to happen, if errors are supposed to happen (they are marked by a dg-error directive in the testcase)).


Then execution test is simply "run the program and see if it fails". So, could you compile the source stat_1.f90 with the same options, and see why its execution failed? The exact command-line used by the testsuite framework is given in what you quoted in your mail:

Executing on host: /sw/src/fink.build/gcc4-4.1.999-20060726/darwin_objdir/gcc/testsuite/gfortran/../../gfortran -B/sw/src/fink.build/gcc4-4.1.999-20060726/darwin_objdir/gcc/testsuite/gfortran/../../ /sw/src/fink.build/gcc4-4.1.999-20060726/gcc-4.2-20060726/gcc/testsuite/gfortran.dg/stat_1.f90 -O0 -std=gnu -L/sw/src/fink.build/gcc4-4.1.999-20060726/darwin_objdir/powerpc-apple-darwin8/./libgfortran/.libs -L/sw/src/fink.build/gcc4-4.1.999-20060726/darwin_objdir/powerpc-apple-darwin8/./libgfortran/.libs -L/sw/src/fink.build/gcc4-4.1.999-20060726/darwin_objdir/powerpc-apple-darwin8/./libiberty -lm -o ./stat_1.exe (timeout = 300)

Once we identify the reason of the failure, I can work something out.


Thanks,
FX


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]