GNU bug report logs -
#30807
Tramp tests take waaayyy too long and refuse to exit
Previous Next
Reported by: Paul Eggert <eggert <at> cs.ucla.edu>
Date: Tue, 13 Mar 2018 20:50:02 UTC
Severity: normal
Done: Paul Eggert <eggert <at> cs.ucla.edu>
Bug is archived. No further changes may be made.
Full log
View this message in rfc822 format
Paul Eggert <eggert <at> cs.ucla.edu> writes:
Hi Paul,
> Thanks, I did a 'git pull' and a 'make bootstrap' followed by a 'make
> check' and got better numbers. Here are the top four now:
>
> tramp-test21-file-links (9.909445 sec)
> cl-seq-test-bug24264 (12.105374 sec)
> tramp-test38-special-characters (13.106871 sec)
> auto-revert-test00-auto-revert-mode (13.113220 sec)
I've profiled tramp-tests.el, and could optimize some of the
tests. Surprisingly, should-error clauses take much more time than other
forms.
Furthermore, there was an error in how tramp-tests determine whether
expensive tests run. This I've fixed as well.
As a result, I have now
passed 13/42 tramp-test11-copy-file (1.050841 sec)
passed 14/42 tramp-test12-rename-file (1.175471 sec)
passed 25/42 tramp-test21-file-links (0.606194 sec)
passed 36/42 tramp-test38-special-characters (0.535814 sec)
Ran 42 tests, 41 results as expected, 1 skipped (2018-03-17 15:49:10+0100, 13.198513 sec)
The high runners take less time, and the whole duration for tramp-tests
is 13 sec. This is acceptable, isn't it?
I will continue to analyze other long-running tests I feel responsible
(autorevert-tests, filenotify-tests), but for this bug report I believe
we could close it.
Best regards, Michael.
This bug report was last modified 6 years and 280 days ago.
Previous Next
GNU bug tracking system
Copyright (C) 1999 Darren O. Benham,
1997,2003 nCipher Corporation Ltd,
1994-97 Ian Jackson.