Fedora-x86_64-native-gdbserver-m64 build is testing the wrong target (Re: Failures on Fedora-x86_64-native-gdbserver-m64, branch master)

Pedro Alves palves@redhat.com
Fri Feb 6 15:28:00 GMT 2015


On 02/06/2015 02:44 PM, sergiodj@redhat.com wrote:
> Buildslave:
> 	fedora-x86-64-2
> 
> Full Build URL:
> 	<http://gdb-build.sergiodj.net/builders/Fedora-x86_64-native-gdbserver-m64/builds/148>
> 
> Commit(s) tested:
> 	e584fdbc6a9b0da786c5a55855f9fd9f7e676198
> 
> Author(s) (in the same order as the commits):
> 	Pedro Alves <palves@redhat.com>
> 
> Testsuite log (gdb.sum and gdb.log) URL(s):
> 	<http://gdb-build.sergiodj.net/cgit/Fedora-x86_64-native-gdbserver-m64/.git/tree/?h=master&id=d7a565fb9782a8824f5248fabca0b9e46a840215>
> 
> *** Regressions found ***
> ============================
> new FAIL: gdb.threads/attach-many-short-lived-threads.exp: iter 10: attach (got interactive prompt)

It looks like something is broken in this builder's setup.  This
test shouldn't be running at all with the --target_board=native-gdbserver
board.

Looking at the test logs, we see:

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Test Run By gdb-buildbot on Fri Feb  6 08:46:03 2015
Native configuration is x86_64-unknown-linux-gnu

                === gdb tests ===

Schedule of variations:
    unix

Running target unix
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

"unix" means that the tests are actually running with the
default native target.




It just read native-gdbserver instead, like:

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Test Run By gdbslave on Wed Feb  4 09:48:13 2015
Native configuration is x86_64-unknown-linux-gnu

                === gdb tests ===

Schedule of variations:
    native-gdbserver
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~


Sounds like --target_board= is getting lost somewhere.  I'd
suspect the switch-to-parallel-mode changes.

Thanks,
Pedro Alves



More information about the Gdb-testers mailing list