Brian Inglis via Cygwin-apps wrote:
On 2025-04-07 09:34, Christian Franke via Cygwin-apps wrote:
Jon Turney via Cygwin-apps wrote:
On 15/09/2024 16:11, Jon Turney via Cygwin-apps wrote:
On 13/09/2024 18:33, Christian Franke via Cygwin-apps wrote:
I would like to contribute stress-ng. Also present in Debian,
Fedora, FreeBSD, Ubuntu, ...
I was thinking about adding a step to our CI to run stress-ng, as it
seems quite good at finding some kinds of regressions.
I wonder if you could suggest a suitable set of options to invoke it
with?
Attached is an enhanced version of my local script used to test
stress-ng before packaging. I agree that it should provide some more
comments :-)
Try
./cygstress -n CI
to see a selection of tests for CI. Remove -n to run these (~ 5min).
It would also be possible to run the tests with a single stress-ng
command.
The tests which detected Cygwin bugs that are already fixed are
included, see comments "fixed in Cygwin ...".
Use WORK (or FAIL) instead of CI to run all working (or failing)
tests. See comments "TODO Cygwin: ..." for brief info about the
failures. See three links for testcases posted to ML.
Tests are intentionally run separately because failing tests
sometimes hang and may even ignore SIGKILL. Therefore an external
tool is used, pskill from Sysinternals Suite is the default.
Have you tried the Cygwin binary kill /usr/bin/kill -f PID..., which
goes via Win32, or even with -W WINPID, which I have found adequate
for all purposes so far, rather than the normal default shell kill
commands, which are lacking.
Yes, but it cannot kill processes by name like 'killall'. This is
required because a hanging stress-ng may leave many child processes
behind. 'killall' from psmisc lacks a -f.
An alternative would be to obtain the list of processes via ps or procps
and then run kill -f. Unfortunately ps may also hang when stress-ng
hangs in some rare cases.
--
Regards,
Christian
