2018-05-16 15:56 GMT+02:00 Michael Biebl <[email protected]>: > 2018-05-16 14:38 GMT+02:00 Rainer Gerhards <[email protected]>: >> As a side-note, I would like to change that behaviour for rsyslog at >> least, trying to select ports dynamically. That would potentially >> introduce a noticable speedup. But unfortunately there is so much too >> do... >> >> I think I remember there was a way to tell autoconf to run the tests >> in a given series, but I could not find the info when I just searched. >> Do you happen to know (the link)? > > Not off-hand.
Thx. Just FYI: I did a more elaborate search. It's described here: https://www.gnu.org/software/automake/manual/html_node/Parallel-Test-Harness.html Starts at bottom of page with sentence "In order to guarantee an ordering ". But I tried it and it does not work. Just if you are curios, this is the patch: https://github.com/rgerhards/librelp/commit/e45a8656e59cb708c3b29e9ef0a3ac47ac6aab55 Travis result: https://www.travis-ci.org/rgerhards/librelp Even if it would work, it would be way to clumpsy for a beast as complex as the rsyslog testbench, where we enable and disable various tests depending on configure options selected. I guess it would also not work when TESTS="xxx" is used as the tests on which xxx then depends would not be executed. > > What I did for the time being is to tell debhelper to run the checks > without parallelism: > > https://salsa.debian.org/debian/librelp/commit/ede9a6837cefd195315bb0044f49cb0c3cbce9eb > in conclusion, this looks like the only decent way to handle it for such cases. Rainer _______________________________________________ rsyslog mailing list http://lists.adiscon.net/mailman/listinfo/rsyslog http://www.rsyslog.com/professional-services/ What's up with rsyslog? Follow https://twitter.com/rgerhards NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a myriad of sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if you DON'T LIKE THAT.
