Replace the existing regex-based hack for recognizing reftest reference files
with a reliable method based on the reftest manifest.
Differential Revision: https://phabricator.services.mozilla.com/D51172
--HG--
extra : moz-landing-system : lando
nsXULWindow is no longer XUL specific and is somewhat confusing name.
Differential Revision: https://phabricator.services.mozilla.com/D51486
--HG--
rename : xpfe/appshell/nsXULWindow.cpp => xpfe/appshell/AppWindow.cpp
rename : xpfe/appshell/nsXULWindow.h => xpfe/appshell/AppWindow.h
rename : xpfe/appshell/nsIXULWindow.idl => xpfe/appshell/nsIAppWindow.idl
extra : moz-landing-system : lando
Standardize the jsapi-tests logic to match the other spidermonkey check
commands in testing/mach_commands.py.
Differential Revision: https://phabricator.services.mozilla.com/D51530
--HG--
extra : moz-landing-system : lando
* Adds a new moz-http2-child.js file which gets spawned into a new process. When calling NodeServer.execute, the code gets passed to the existing moz-http2.js process which then sends it to be evaluated in the child process. Any crash in the child should not be able to kill the main node process.
* Moves the proxy creation code into test_http2-proxy.js
* Adds the new NodeServer.fork() and NodeServer.kill() static methods to spawn a new server
* Makes it easier to isolate a test's behaviour from another's. It also opens the way to moving some of the logic to individual unit tests, like we do for the proxy creation code, rather than keeping it all in moz-http2.js
Differential Revision: https://phabricator.services.mozilla.com/D49961
--HG--
extra : moz-landing-system : lando
Running this test leads to intermittent "unable to find test window"
failures on 32 bit windows. If the timeout to get_test_window() in
executormarionette.py is increased, the failure goes away, but there
is no way in the framework to increase this timeout for a single
test.
Differential Revision: https://phabricator.services.mozilla.com/D51618
--HG--
extra : moz-landing-system : lando
This is the third, and final, part of a patch that adds chrome support to raptor-browsertime.
In this part, all the changes from the previous 2 parts are integrated in raptor and browsertime. The main change is that the browsertime-chromedriver paths created in taskcluster's `tests.py` are changed to include a `{}` for inserting the chrome version that is used in production (formatting is done within raptor). The Perftest class structure is slightly modified as well here to accommodate these changes.
There are also some changes to how the taskcluster definitions are made to make it easier to specify how chrome must be handled. The 'limit-platforms' and 'run-visual-metrics' fields are added for this purpose to limit chrome's platforms and disable visual-metric tasks.
Finally, chrome is missing some results in browsertime (relative to firefox) and at least one of the other results is scattered into a different location. The results.py and output.py changes handle this issue.
Differential Revision: https://phabricator.services.mozilla.com/D48898
--HG--
extra : moz-landing-system : lando
This is the first part of a patch that adds chrome support to raptor-browsertime.
In this patch, a get_browser_meta function is added which returns the name and version of the browser being tested. This version will then be used (in part 3) in the variable passed in through --browsertime-chromedriver (only when running not locally, or if we find {} in the string).
Differential Revision: https://phabricator.services.mozilla.com/D48895
--HG--
extra : moz-landing-system : lando
The glyph pixel space in which we rasterized glyphs differed from how we
rendered the rasterized glyphs in the shader. They need to be in
agreement because the glyph subpixel offset selected during
rasterization depends on it. This patch should make the paths consistent
with each other.
Additionally, during animations, we now snap the reference frame
relative offset ignoring the impact of any animated transforms. This
helps with minimizing glyph wiggling during the transition.
Differential Revision: https://phabricator.services.mozilla.com/D51305
--HG--
extra : moz-landing-system : lando
promise rejection event was enabled by default on 69 (bug 1525554).
We could get rid of this preference.
Differential Revision: https://phabricator.services.mozilla.com/D51491
--HG--
extra : moz-landing-system : lando
For some reasons, the remote type doesn't support on Beta and Release Channels.
Disable the wpt tests for now, and will reenable them after suspicious bug is
fixed.
Differential Revision: https://phabricator.services.mozilla.com/D51633
--HG--
extra : moz-landing-system : lando
nsXULWindow is no longer XUL specific and is somewhat confusing name.
Differential Revision: https://phabricator.services.mozilla.com/D51486
--HG--
rename : xpfe/appshell/nsXULWindow.cpp => xpfe/appshell/AppWindow.cpp
rename : xpfe/appshell/nsXULWindow.h => xpfe/appshell/AppWindow.h
rename : xpfe/appshell/nsIXULWindow.idl => xpfe/appshell/nsIAppWindow.idl
extra : moz-landing-system : lando
This minor optimization is easily implemented: If there are no tests to verify, call fatal(0).
However, I encountered a minor obstacle: If the task exits before creating the upload directory,
the task fails, regardless of exit value; the remaining changes overcome this by creating the
upload directory earlier.
Differential Revision: https://phabricator.services.mozilla.com/D51684
--HG--
extra : moz-landing-system : lando
This allows running the jit_test.py harness using mach. It works very
similarly to the existing 'mach jstests' command.
Differential Revision: https://phabricator.services.mozilla.com/D51518
--HG--
extra : moz-landing-system : lando
* Adds a new moz-http2-child.js file which gets spawned into a new process. When calling NodeServer.execute, the code gets passed to the existing moz-http2.js process which then sends it to be evaluated in the child process. Any crash in the child should not be able to kill the main node process.
* Moves the proxy creation code into test_http2-proxy.js
* Adds the new NodeServer.fork() and NodeServer.kill() static methods to spawn a new server
* Makes it easier to isolate a test's behaviour from another's. It also opens the way to moving some of the logic to individual unit tests, like we do for the proxy creation code, rather than keeping it all in moz-http2.js
Differential Revision: https://phabricator.services.mozilla.com/D49961
--HG--
extra : moz-landing-system : lando
The previous code tried to do it, but it did it wrongly, as the overflow clip
comes from the parent, not the child.
Thus when we change a style that influences it, we weren't invalidating the
SIMPLE_DISPLAY_LIST bit, and such.
Make the reftest that caught this fail more reliable.
Differential Revision: https://phabricator.services.mozilla.com/D51683
--HG--
extra : moz-landing-system : lando
nsXULWindow is no longer XUL specific and is somewhat confusing name.
Differential Revision: https://phabricator.services.mozilla.com/D51486
--HG--
rename : xpfe/appshell/nsXULWindow.cpp => xpfe/appshell/AppWindow.cpp
rename : xpfe/appshell/nsXULWindow.h => xpfe/appshell/AppWindow.h
rename : xpfe/appshell/nsIXULWindow.idl => xpfe/appshell/nsIAppWindow.idl
extra : moz-landing-system : lando
Split off of Bug 1590894
Rename these to support unprefixed version
Also add alias to keep compatibility
Differential Revision: https://phabricator.services.mozilla.com/D50989
--HG--
extra : moz-landing-system : lando
I'm making this refactor now because a future commit is going to completely re-write the
test data (so that it matches the data that a future TestManifestLoader is going to use).
With this method, updating the data will be a lot less tedious.
Differential Revision: https://phabricator.services.mozilla.com/D51278
--HG--
extra : moz-landing-system : lando
Having a distinction between -chunked and not adds unnecessary complexity. It's
possible to simply remove them because:
1. The mozharness definitions for 'jittest' and 'jittest-chunked' are
identical, so it is literally not serving any purpose.
2. The definitions for 'mochitest' only add either '--chunk-by-dir' or
'--chunk-by-runtime'. Both of these are no-ops in the mochitest harness
unless '--total-chunks' is also supplied. Therefore, if we ever use these
suites without chunking (which I don't think we do anyway), then it'll
still work fine as those options won't have any affect.
Differential Revision: https://phabricator.services.mozilla.com/D51173
--HG--
extra : moz-landing-system : lando
Geoff Brown
Brendan Dahl
Cosmin Sabou
Ryan Hunt
Ted Campbell
Valentin Gosu
Dan Minor
Gregory Mierzwinski
Andrew Halberstadt
Andrew Osmond
Ionut Goldan
Edgar Chen
Tom Tung
Mihai Alexandru Michis
Tarek Ziadé
Ciure Andrei
Karl Tomlinson
Csoregi Natalia
Emilio Cobos Álvarez
Dorel Luca
Andreas Pehrson
Sam Mauldin
Andreea Pavel
Florin Strugariu
Rishi Gupta
Gurzau Raul
Sebastian Hengst
Oriol Brufau