bug#50902: 28.0.50; emacs-module-tests time out

[Philipp Stephani]

Am Fr., 8. Okt. 2021 um 13:45 Uhr schrieb Michael Albinus
<michael.albinus@gmx.de>:
>
> Michael Albinus <michael.albinus@gmx.de> writes:
>
> Hi,
>
> > on emba.gnu.org, emacs-module-tests time out.
>
> The log file contains
>
> --8<---------------cut here---------------start------------->8---
> Running 38 tests (2021-10-08 04:35:34+0000, selector `(not (or (tag 
> :unstable) (tag :nativecomp)))')
>    passed   1/38  emacs-module-tests/interleaved-threads (0.000996 sec)
>    passed   2/38  mod-test-add-nanosecond/invalid (0.000242 sec)
>    passed   3/38  mod-test-add-nanosecond/nil (0.000117 sec)
>    passed   4/38  mod-test-add-nanosecond/valid (0.000522 sec)
>    passed   5/38  mod-test-double (0.000357 sec)
>    passed   6/38  mod-test-globref-free-test (0.000120 sec)
>    passed   7/38  mod-test-globref-make-test (0.023353 sec)
>    passed   8/38  mod-test-globref-reordered (0.000132 sec)
>    passed   9/38  mod-test-make-string/empty (0.000160 sec)
>    passed  10/38  mod-test-make-string/nonempty (0.000348 sec)
>    passed  11/38  mod-test-nanoseconds (0.000654 sec)
>    passed  12/38  mod-test-non-local-exit-funcall-normal (0.000135 sec)
>    passed  13/38  mod-test-non-local-exit-funcall-signal (0.000165 sec)
>    passed  14/38  mod-test-non-local-exit-funcall-throw (0.000132 sec)
>    passed  15/38  mod-test-non-local-exit-signal-test (0.230399 sec)
>    passed  16/38  mod-test-non-local-exit-throw-test (0.000115 sec)
>    passed  17/38  mod-test-sleep-until (0.218033 sec)
>    passed  18/38  mod-test-string-a-to-b-test (0.000160 sec)
>    passed  19/38  mod-test-sum-docstring (0.000142 sec)
>    passed  20/38  mod-test-sum-test (0.000398 sec)
>    passed  21/38  mod-test-userptr-fun-test (0.000219 sec)
>    passed  22/38  mod-test-vector-test (0.021364 sec)
>    passed  23/38  module--func-arity (0.000202 sec)
>    passed  24/38  module--help-function-arglist (0.000273 sec)
> --8<---------------cut here---------------end--------------->8---
>
> So I suspect the timeout happens in 
> module--test-assertions--call-emacs-from-gc.
> Philipp, does this tell you anything?
>


Not really. That test uses the module assertion functionality to abort
Emacs on some invalid usage patterns. It starts a subprocess which
should quickly call abort().
Would it be possible to capture a core dump or similar while the test
is hanging and analyze it with a debugger?