Steve Hay wrote:
>  Automated smoke report for 5.11.0 patch 34150
>  maldoror.bath.planit.group: Intel(R) Core(TM)2 CPU 6700 @
>      2.66GHz(~2660 MHz) (x86/2 cpu) on        MSWin32 - WinXP/.Net SP3
>      using     cl version 13.10.3077
>      smoketime 5 hours 18 minutes (average 15 minutes 54 seconds)
>  
>  Summary: FAIL(F)
>  
[...]
>  ../ext/threads/t/free2.t....................................FAILED
>      Bad plan.  You planned 78 tests but ran 67.

This failure arises from #34150, which added a watchdog() to the
threads/t/free*.t tests. Apparently the test hung and the watchdog() was
called into play. It did it's job nicely, killing the hung test and
allowing the remainder of the smoke to continue. (Nice work, Jerry.)
Here's the output:

[...]
[03:54:26] ../ext/threads/t/exit...........ok      267 ms
[03:54:27] ../ext/threads/t/free...........ok    12188 ms
# Test process timed out - terminating
[03:54:39] ../ext/threads/t/free2......... Dubious, test returned 1
(wstat 256, 0x100)
 Failed 11/78 subtests 
[03:55:39] ../ext/threads/t/join...........ok     2296 ms
[...]

It would be even better if the parsers of that output were able to
recognize the "# Test process timed out - terminating" line and report
it in the final summary shown above, rather than "Bad plan...". It
wasn't a bad plan: It was a good plan, it just didn't work out.

What needs changing for that to happen? Test-Harness or Test-Smoke or
both?