rtems-tester lock failure

[Chris Johns]

On 30/1/20 4:14 am, Joel Sherrill wrote:
> I have seen similar failures before but Chris and I discussed it was likely 
> related to Python 2. Using Python 3, I am getting the following error:
> 
> + /home/joel/rtems-work/rtems-tools//tester/rtems-test
> --rtems-tools=/home/joel/rtems-work/tools/5 --rtems-bsp=leon3-sis --log=run.log .
> RTEMS Testing - Tester, 5 (4cb2ddc568b1)
> [  2/594] p:0   f:0   u:0   e:0   I:0   B:0   t:0   i:0   W:0   | sparc/leon3:
> linpack.exe
> [  1/594] p:0   f:0   u:0   e:0   I:0   B:0   t:0   i:0   W:0   | sparc/leon3:
> dhrystone.exe
> [  6/594] p:0   f:0   u:0   e:0   I:0   B:0   t:0   i:0   W:0   | sparc/leon3:
> fsdosfsformat01.exe
> [  7/594] p:0   f:0   u:0   e:0   I:0   B:0   t:0   i:0   W:0   | sparc/leon3:
> fsdosfsname01.exe
> [  3/594] p:0   f:0   u:0   e:0   I:0   B:0   t:0   i:0   W:0   | sparc/leon3:
> whetstone.exe
> [  5/594] p:0   f:0   u:0   e:0   I:0   B:0   t:0   i:0   W:0   | sparc/leon3:
> fsclose01.exe
> [  8/594] p:0   f:0   u:0   e:0   I:0   B:0   t:0   i:0   W:0   | sparc/leon3:
> fsdosfsname02.exe
> [  4/594] p:0   f:0   u:0   e:0   I:0   B:0   t:0   i:0   W:0   | sparc/leon3:
> fsbdpart01.exe
> error: invalid test state:
> ./sparc-rtems5/c/leon3/testsuites/fstests/fsbdpart01.exe: EXPECTED_PASS

This is the error that is causing the problem. I suspect it is related to the
patches that went in recently to change `-` to `_` in all the labels which
caused a parsing or something like that issue. I suspect we would see these
issues pop up and we are.

> Exception ignored in: <module 'threading' from
> '/opt/rh/rh-python36/root/usr/lib64/python3.6/threading.py'>
> Traceback (most recent call last):
>   File "/opt/rh/rh-python36/root/usr/lib64/python3.6/threading.py", line 1294,
> in _shutdown
>     t.join()
>   File "/opt/rh/rh-python36/root/usr/lib64/python3.6/threading.py", line 1056,
> in join
>     self._wait_for_tstate_lock()
>   File "/opt/rh/rh-python36/root/usr/lib64/python3.6/threading.py", line 1072,
> in _wait_for_tstate_lock
>     elif lock.acquire(block, timeout):

This is in the thread shutdown. I am not sure what is wrong here.

> Hopefully someone can reproduce this.

It should be with a hack. Start a bunch of simulation jobs then raise a general
error exception.

Chris