I've seen a variety of glibc test failures, most of which I don't
have a clue about at the moment. For x86, I think the only failure
now is posix/annexc.out which doesn't really count. I haven't built
ppc for a while (my G5 is only building ppc64 now, my G4 ibook
hasn't found a usable kernel/pbbuttonsd combination since the kernel
ARCH changed to powerpc :-( I haven't attempted x86_64 multilib
recently, but I've still managed to collect the following errors:
1. ppc64 32-bit multilib
1.0 nptl/tst-cancel24.out error while loading shared libraries:
libstdc++.so.6: cannot open shared object file: No such file or
directory
1.1 elf/check-textrel.out [ no information about the error ]
2. ppc64 64-bit multilib
2.0 iconvdata/tst-tables.out
Testing ASCIItst-tables.sh: line 251: bash-3.1: command not found
I guess this is from line 251 of iconvdata/tst-tables.sh :
if ${SHELL} tst-table.sh ${common_objpfx} ${objpfx} ${charset}
${charmap}; then
echo 1>&2
else
echo "failed: ./tst-table.sh ${common_objpfx} ${objpfx}
${charset} ${charmap}"
echo " *** FAILED ***" 1>&2
but I've no idea why it is apparently evaluating ${SHELL} as
bash-3.1 instead of /bin/bash.
2.1 misc/tst-pselect.out
Timed out: killed the child process
Testing without the iconv_fix patch (thought it might be implicated
in 2.0, but no) caused this test to pass.
2.2 nptl/tst-cancel24.out - as 1.0
2.3 rt/tst-timer.out Error 139
Segmentation fault
This looks similar to the x86_64 multilib report
http://sourceware.org/bugzilla/show_bug.cgi?id=2529
which was marked as a duplicate of their bug 333 "inappropriate user
report".
3. x86_64-64 'pure64'
3.0 iconvdata/test-tables.out - as 2.0
3.1 tst-pselect.out
Timed out: killed the child process
I suppose that the iconvdata/test-tables thing might be common to
other 64-bit builds. Hopefully, the other failures might be
arch-specific.
Any other failures ? Any suggestions, apart from "don't run the
testsuites" ?
Ken
--
das eine Mal als Tragödie, das andere Mal als Farce
_______________________________________________
Clfs-dev mailing list
[email protected]
http://lists.cross-lfs.org/cgi-bin/mailman/listinfo/clfs-dev
