mutter/tests/README
Øyvind Kolås aa05b66a01 tests: Add performance tracking framework
This adds a performance tracking framework that can run a set of tests over
specified git revisions. The ruby script for generating the reports comes from
similar performance tracking in GEGL. The framework permits evaluating new
tests against older version of clutter.

The tests themselves go through a few hoops for disabling framerate limiting in
both mesa and clutter.

When running make check the tests will be run and lines of the form:

@ test-state: 40.51 fps

will be left in the output, a script can scrape these lines out of a build log
on a buildbot to in other ways track performance.
2011-07-04 17:27:48 +01:00

41 lines
1.8 KiB
Plaintext

Outline of test categories:
The conform/ tests should be non-interactive unit-tests that verify a single
feature is behaving as documented. See conform/ADDING_NEW_TESTS for more
details.
The performance/ tests are performance tests, both focused tests testing single
metrics and larger tests. These tests are used to report one or more
performance markers for the build of Clutter. Each performance marker is picked
up from the standard output of running the tests from strings having the form
"\n@ marker-name: 42.23" where 'marker-name' and '42.23' are the key/value pairs
of a single metric. Each test can provide multiple key/value pairs. Note that
if framerate is the feedback metric the test should forcibly enable FPS
debugging itself. The file test-common.h contains utility function helping to
do fps reporting.
The interactive/ tests are any tests whose status can not be determined without
a user looking at some visual output, or providing some manual input etc. This
covers most of the original Clutter tests. Ideally some of these tests will be
migrated into the conformance/ directory so they can be used in automated
nightly tests.
The accessibility/ tests are tests created to test the accessibility support of
clutter, testing some of the atk interfaces.
The data/ directory contains optional data (like images and ClutterScript
definitions) that can be referenced by a test.
Other notes:
• All tests should ideally include a detailed description in the source
explaining exactly what the test is for, how the test was designed to work,
and possibly a rationale for the approach taken for testing.
• When running tests under Valgrind, you should follow the instructions
available here:
http://live.gnome.org/Valgrind
and also use the suppression file available inside the data/ directory.