Thanks to the new faulthandler module (#11393) and regrtest timeout (#11727, timeout of 15 minutes), I finally found why test_threadsignals hangs on PPC Tiger 3.x:
-------------------------------------
...
[ 25/354] test_threadsignals
Thread 0xa000ed88:
  File "/Users/buildbot/buildarea/3.x.parc-tiger-1/build/Lib/test/test_threadsignals.py", line 46 in test_signals
  File "/Users/buildbot/buildarea/3.x.parc-tiger-1/build/Lib/unittest/case.py", line 387 in _executeTestPart
  File "/Users/buildbot/buildarea/3.x.parc-tiger-1/build/Lib/unittest/case.py", line 442 in run
  File "/Users/buildbot/buildarea/3.x.parc-tiger-1/build/Lib/unittest/case.py", line 494 in __call__
  File "/Users/buildbot/buildarea/3.x.parc-tiger-1/build/Lib/unittest/suite.py", line 105 in run
  File "/Users/buildbot/buildarea/3.x.parc-tiger-1/build/Lib/unittest/suite.py", line 67 in __call__
  File "/Users/buildbot/buildarea/3.x.parc-tiger-1/build/Lib/unittest/suite.py", line 105 in run
  File "/Users/buildbot/buildarea/3.x.parc-tiger-1/build/Lib/unittest/suite.py", line 67 in __call__
  File "/Users/buildbot/buildarea/3.x.parc-tiger-1/build/Lib/test/support.py", line 1078 in run
  File "/Users/buildbot/buildarea/3.x.parc-tiger-1/build/Lib/test/support.py", line 1166 in _run_suite
  File "/Users/buildbot/buildarea/3.x.parc-tiger-1/build/Lib/test/support.py", line 1192 in run_unittest
  File "/Users/buildbot/buildarea/3.x.parc-tiger-1/build/Lib/test/test_threadsignals.py", line 210 in test_main
  File "./Lib/test/regrtest.py", line 1032 in runtest_inner
  File "./Lib/test/regrtest.py", line 826 in runtest
  File "./Lib/test/regrtest.py", line 650 in main
  File "./Lib/test/regrtest.py", line 1607 in <module>
make: *** [buildbottest] Error 1
program finished with exit code 2
elapsedTime=1103.489396
-------------------------------------
http://www.python.org/dev/buildbot/all/builders/PPC%20Tiger%203.x/builds/1685/steps/test/logs/stdio

test_threadsignals hangs at:
----------------
class ThreadSignals(unittest.TestCase):
    def test_signals(self):
        signalled_all.acquire()
        self.spawnSignallingThread()
        signalled_all.acquire() <~~~~ here
        ...
----------------

self.spawnSignallingThread() calls:
----------------
def send_signals():
    os.kill(process_pid, signal.SIGUSR1)
    os.kill(process_pid, signal.SIGUSR2)
    signalled_all.release()
----------------

Before fixing the bug, we can workaroung the hang in the test using a timeout on the second acquire.

This issue may be related to #11223.

The trace is supposed to contain the traceback of all threads, and I see only one thread. So I suppose that send_signals() thread has exited. I don't know if it raises an exception or was interrupted before calling signalled_all.release().

May be related to #11753.

New changeset 9d59ae98013c by Victor Stinner in branch 'default':
Reenable regrtest.py timeout (30 min): #11738 and #11753 looks to be fixed
http://hg.python.org/cpython/rev/9d59ae98013c

I think that the last test_threadsignals failures on PPC Tiger were related to the new regrtest timeout, and it is now fixed.

> I think that the last test_threadsignals failures on PPC Tiger
> were related to the new regrtest timeout, and it is now fixed.

Wrong, it was a deadlock. It is a duplicate of #11768 and it is now fixed.