With Gradle integration centralized in gradle.configure, changing
these integration points will need to trigger the android-* tasks.
MozReview-Commit-ID: DuOuW1RIgCh
--HG--
extra : rebase_source : a3f0a76b1ae6b3cd952271782f9d0c2463b704e0
This pushes all of the magic Gradle targets into gradle.configure,
which is the most accessible central place for them. Some impact the
build, so they almost certainly need to be in moz.configure; and its
better to have one central place to update than many places.
Gradle has a notion of configurations. Fennec determines many
configurations, only one of which is used in automation; and right
now, that one is "officialPhotonDebug". Evolving the "one true"
configuration is frustrating, and this helps with that. Post
Android-Gradle plugin 3.0+, we can do better and also extract the
configuration details for the :geckoview and :geckoview_example Gradle
projects, but right now they are ad-hoc and can't really be improved.
MozReview-Commit-ID: LdGE05zn2H1
--HG--
extra : rebase_source : db7e9fe90f2bd1ae5058af046a66edd0cd549141
These magic locations evolve over time. Baking them into
moz.configure is the easiest way to share them across the build
system, and pushing them into a new |mach android *| command continues
a pattern that has been very successful.
MozReview-Commit-ID: CyxVQ0LHHgl
--HG--
extra : rebase_source : 8350d71665f0126aa4ee2c8fec32c4b8e34dc772
This was in Gradle due to history. When this first landed, we invoked
Gradle directly from the mozharness, and the best way to print the
report URLs was from Gradle itself. When the Android Gradle suites
were made tier 1, little harnesses (|mach android
{checkstyle,findbugs,lint,test}|) were written and invoked locally and
in automation. This functionality should have migrated with them.
This removes the special Gradle target names from the Gradle
configuration, making it easier to change them in the future.
MozReview-Commit-ID: 1KPd3J5t82Q
--HG--
extra : rebase_source : 1da85e31c113bc9da138817bebf981af8b9b66dd
We've had good success with |mach android *| for invoking test suites
and other Gradle-related features; this was just an oversight. The
real advantage is that almost all of the magic Gradle targets are now
in mobile/android/mach_commands.py and not scattered throughout the
tree. That'll make it easier to move the actual Gradle configuration
choices forward in the future.
MozReview-Commit-ID: GI1pUHHfXeO
--HG--
extra : rebase_source : 9d7b98e968a2537f8dd1ffd0646c51ffe429bd57
This patch is effectively just resurrecting a function that was previously removed in https://hg.mozilla.org/mozilla-central/rev/173a4f49dfe3#l1.96 .
MozReview-Commit-ID: 4PXSOu4tzzU
--HG--
extra : rebase_source : a8ada7319ea86ccb6178870510fdbb1dac6bd027
This code used to insert duplicate entries to avoid expensive shuffling of the LRU cache. With uluru this is no longer necessary, because reordering the cache is cheap.
Now it uses the `LRUCache::find` method from uluru 0.2 to update entries in-place. This should increase cache hits by eliminating unnecessary evictions.
This PR also updates `arrayvec` because `uluru` depends on it, and `immeta` because it depends on `arrayvec`.
---
- [x] `./mach build -d` does not report any errors
- [x] `./mach test-tidy` does not report any errors
- [x] These changes do not require tests because they don't change behavior
Source-Repo: https://github.com/servo/servo
Source-Revision: 0c13f14aaa151764e1b3dd2d4503cb6fc3c9595b
--HG--
extra : subtree_source : https%3A//hg.mozilla.org/projects/converted-servo-linear
extra : subtree_revision : b1551f8e564b25ef5a332d0492584e8e179e9325
It looks like the main cause of intermittent failures in getDirectory is
that the adb pull command fails because the emulator has hung. For other
commands, we usually handle this by checking the return code and raising
DMError if anything fails. There is mozharness/taskcluster code in
place to automatically retry tasks that throw DMError.
In the first frame after the initial paint, we skip restyling if the initial
paint took over vsync refresh rate as an optimization and to avoid jumpy
animations. To avoid checking this skipped restyling that we'd actually expect
a restyle maker there, firstly we check there is no restyle marker after the
initial paint for scrolled out animation for five frames, and we then make the
element visible and check a restyle marker there.
MozReview-Commit-ID: 5XkJhdtUly5
--HG--
extra : rebase_source : 1a441296ad6f6cc42b50d300ebacd66e5dee77a1