After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 761472 - common: tests timeout is different on top-level make check
common: tests timeout is different on top-level make check
Status: RESOLVED FIXED
Product: GStreamer
Classification: Platform
Component: common
git master
Other All
: Normal enhancement
: 1.7.2
Assigned To: GStreamer Maintainers
GStreamer Maintainers
Depends on:
Blocks:
 
 
Reported: 2016-02-02 20:20 UTC by Thiago Sousa Santos
Modified: 2016-02-05 23:53 UTC
See Also:
GNOME target: ---
GNOME version: ---

Attachments
check.mak: set CK_DEFAULT_TIMEOUT for all tests (1.54 KB, patch)
2016-02-02 20:21 UTC, Thiago Sousa Santos 		committed Details | Review
tests: avoid overwriting AM_TESTS_ENVIRONMENT from common/check.mak (1.06 KB, patch)
2016-02-02 20:21 UTC, Thiago Sousa Santos 		committed Details | Review

Description Thiago Sousa Santos 2016-02-02 20:20:39 UTC 
The timeout on the top-level make check run is different than when you run element.check.

This causes some tests to fail on the top-level while they work when run separately. CK_DEFAULT_TIMEOUT is only being set for the tests/check subdir makefile
Comment 1 Thiago Sousa Santos 2016-02-02 20:21:42 UTC 
Created attachment 320291 [details] [review]
check.mak: set CK_DEFAULT_TIMEOUT for all tests

Set it as part of the AM_TESTS_ENVIRONMENT variable so that it
is the same when running make-check from the top-level dir or
when running the element.check variations
Comment 2 Thiago Sousa Santos 2016-02-02 20:21:56 UTC 
Created attachment 320292 [details] [review]
tests: avoid overwriting AM_TESTS_ENVIRONMENT from common/check.mak

So that CK_DEFAULT_TIMEOUT is properly set for all tests
Comment 3 Thiago Sousa Santos 2016-02-02 20:22:45 UTC 
I'm not sure if there a saner way to have a default, the same patch might have to be replicated for the other modules.
Comment 4 Tim-Philipp Müller 2016-02-02 20:40:50 UTC 
Is 'check-valgrind' also affected then, which has a longer timeout?
Comment 5 Thiago Sousa Santos 2016-02-03 12:34:20 UTC 
No. For some reason it already uses the correct 360 timeout for valgrind tests.

It is defined as:
check-valgrind:
        $(MAKE) -C tests/check check-valgrind
Comment 6 Tim-Philipp Müller 2016-02-05 19:42:29 UTC 
I'd say let's get this in then.

We could just use the default for CK_TIMEOUT which is 30 seconds IIRC. Which seems just as reasonable as 20 really.
Comment 7 Thiago Sousa Santos 2016-02-05 23:50:38 UTC 
commit 33f8051a3a5fee6ea5e63f22be3034a81e295038
Author: Thiago Santos <thiagoss@osg.samsung.com>
Date:   Tue Feb 2 17:12:18 2016 -0300

    check.mak: set CK_DEFAULT_TIMEOUT for all tests
    
    Set it as part of the AM_TESTS_ENVIRONMENT variable so that it
    is the same when running make-check from the top-level dir or
    when running the element.check variations
    
    https://bugzilla.gnome.org/show_bug.cgi?id=761472


Also updated all modules to extend AM_TESTS_ENVIRONMENT in their makefiles
Comment 8 Thiago Sousa Santos 2016-02-05 23:53:25 UTC 
core: 682516402d2e2179238ee8e388ac7a304286c1c8
base: 681b2b8d81d9584ff8f45f48916a8601afd46580
good: 522de42381cd6c68417b41e41610c78e4f71e1e6
bad: 669b360ff7ab03963105564e8b011b322b2ac175
ugly: 7e2f2f9e5b0b0da16fad50f23d184a273d025386
ges: c0dbfeb433a1f328885ebeff9a84d79399a9067e
rtsp-server: 332846173848205a2c0ea72e40f060477dade8d2