- Sep 13, 2019
-
-
Pierre-Yves David authored
this will result in extra processing and output.
-
Raphaël Gomès authored
-
Pierre-Yves David authored
1) this is what it is meant for so let us use it. 2) if we don't use it, the user's `local.repos` will be used instead of the `default.repos` we carefully built for the test.
-
- Aug 15, 2019
-
-
Pierre-Yves David authored
We want to be able to use a virtual-env here.
-
- Aug 09, 2019
-
-
Pierre-Yves David authored
We are warming more of the branchcache. So let's use them
-
- Aug 06, 2019
-
-
Raphaël Gomès authored
-
- Aug 01, 2019
-
-
Raphaël Gomès authored
We can't make sure that no Rust code will be run before the rust module policy was introduced in 94167e701e125dce1788e19b1e1489958235e40c, in fact, we can be pretty sure of it.
-
- Jul 30, 2019
-
-
Raphaël Gomès authored
-
- Aug 06, 2019
-
-
Raphaël Gomès authored
Forgetting to use $PYTHON and just calling the script as an executable will make it use the system version of python instead of the one running the tests, which has very little chance of a) being the right version and b) having the right packages installed.
-
- Aug 01, 2019
-
-
Raphaël Gomès authored
-
- Jul 30, 2019
-
-
Raphaël Gomès authored
-
Raphaël Gomès authored
-
- Jul 22, 2019
-
-
Raphaël Gomès authored
If the `setup` method fails, the tests session waits for timeout (1800s per benchmark!), because BaseExchangeMixin's setup launches `hg serve`. The ASV test runner does not understand that setup failed and it should move on to the next benchmark. This shuts down the `hg serve` in case the `setup` breaks.
-
- Jul 18, 2019
-
-
Pierre-Yves David authored
`self.check_output` call signature is sightly different from `subprocess.check_output`, so the call was wrong.
-
Pierre-Yves David authored
taskset is kind of machine specific and also linux specific. Moving it to the "main" script make it simpler to use launch.py in many context. Ideally, ASV would deal with it itself when spawning process
-
Pierre-Yves David authored
-
Pierre-Yves David authored
-
- May 24, 2019
-
-
Pierre-Yves David authored
-
- May 21, 2019
-
-
Raphaël Gomès authored
-
- May 18, 2019
-
-
Pierre-Yves David authored
If the `local.repos` file is present, it will be read instead of the `default.repos` file. The file itself is not tracked by version control, it is... local
-
- May 17, 2019
-
-
Pierre-Yves David authored
Exchange test takes a lot of time, rsync is one of the factor. Factoring the code will make it easier to improve and measure.
-
Pierre-Yves David authored
-
Pierre-Yves David authored
I started seeing benchmark failing because `hg debugupdatecache` covers more data in some core changesets (so taking extra time to warm some cold cache). Instead we perform an extensive cache warming in the preparation steps. Once the cache are warm the `hg debugupdatecache` run are "fast". See inline documentation for details. We also introduce a way to skip this step as it might be slow.
-
- May 19, 2019
-
-
Pierre-Yves David authored
Since 481259af4bdf the `debugupdatecache` command also warming the `hgtagsfnodes` cache too. We update the test to expect this file.
-
Pierre-Yves David authored
This new experimental mode need to be tracked. The initial implementation is quite slow but things could get better over time.
-
Pierre-Yves David authored
Before, not passing a filter_fn means the function will always return False
-
Pierre-Yves David authored
There is some delay with the mozilla-try one. We push the other one in the mean-time.
-
Pierre-Yves David authored
strictly speaking, I could have update the partials variants keeping the same identifier. However I got lazy and the whole things is long enough to generate.
-
- May 18, 2019
-
-
Pierre-Yves David authored
This benchmark measure cases where both side has missing changed. pypy and mozilla-try has partial variant for it.
-
- May 19, 2019
-
-
Pierre-Yves David authored
Otherwise, we can modify an inherited list, affecting all other benchmark.
-
- May 16, 2019
-
-
Pierre-Yves David authored
-
- May 15, 2019
-
-
Pierre-Yves David authored
-
Pierre-Yves David authored
One more step toward replacing the makefile.
-
Pierre-Yves David authored
This will eventually replace the makefile to setup repository
-
- May 07, 2019
-
-
Pierre-Yves David authored
There have been various behavior and format change after a5e03453eae8. Tagging it will help us to move back to older version if necessary.
-
- Apr 28, 2019
-
-
Pierre-Yves David authored
I fear that the lack of the "same" params value invalidate previous result file that has this value. So I am adding it again. However I am not sure I got that right, that this is in an independent changeset easy to be dropped.
-
Pierre-Yves David authored
This one is dedicated to tracking the time taken for two identical repository to realise they are identical. We take the addition of a new test as an opportunity to give it a proper name to this test.
-
Pierre-Yves David authored
The existing discovery test are now using the new roles information. To keep the compatibility witht he older class structure. We are using the `benchmark_name` feature to keep the same name. We'll probably want to use the very same feature to rename all test later into something more sensible. However, as a side effect, this means we are dropping the "same" variant of the subset/super set test. So ASV will be confused by the params value change... <sigh>. Important note: for this discovery to work, benchmark now needs to be have an upgraded reference.
-
- Apr 27, 2019
-
-
Pierre-Yves David authored
benchmark test can now "easily" use the new role information. A `BaseExchangeTimeSuite` subclass can use the `role_action` and `role_subtype` to declare the role the class is using. Then decorate the class with `@setup_role` to make sure the parameters variants are properly set. Finally the appropriate data can we accessed through the `role_data` property.
-
Pierre-Yves David authored
We need this information to be available to the benchrepo code before we can use it.
-