* Revert "Fix eglPresentationTimeANDROID is no effective (#30182)"
This reverts commit 0d7ba05d3456807e7e24353fe911738952a02888.
* Revert "Use eglPresentationTimeANDROID to avoid bogging down the GPU (#29727)"
This reverts commit edb87942de0404a2802351c050a4f1b6de239bd7.
Fixesflutter/flutter#93352
Improves Android benchmarks on both Pixel 4 and a lower end Android Go device for 99th percentile and average raster times.
This works by telling the system compositor what timestamp we intended to show this frame for. This way, if we end up with a frame that gets submitted right at the beginning of a vsync and then a second frame submitted on the same vsync, the compositor will only try to show the second frame on the screen and save the GPU some work.
Without this, a situation like that results in an "avalanche" of calls where the GPU is behind the CPU and keeps delaying CPU work until we finally stop submitting frames. This can be observed as a lengthy dequeuBuffer in a systrace enabled trace, as shown in the linked issue. This avalanche is often triggered by a frame that does a shader compile through a couple vsyncs and then is followed by a bunch of very fast frames that take less than a vsync to render - the first of those fast frames gets delivered before the end of the vsync that the slow frame ended in.
We cannot implement this ourselves because we don't know how long the swap buffers call will take on the system side, and if we try to guess we can very well get it wrong.
I've filed issues to look into adding this for Vulkan and Metal, although we should also first take traces there to make sure it's warranted.
See also: https://android-developers.googleblog.com/2020/04/high-refresh-rate-rendering-on-android.html
This PR changes how embedder API's SendKeyData sends ui.KeyData to the framework. The packets are now sent over the existing platform messenger, reusing the entirety of its code path and functionalities while keeping the embedder API unchanged
This makes sure the frame timings recorder and vsync waiter implementations get
the correct refresh rate if or when it adjusts at runtime. This should be OK
because we'll only query the refresh rate when the display metrics actually
change, which is much less frequent than every frame. I experimented with an NDK
implementation in the previous patch, but that vastly restricts the API levels
we can support, and currently on API 30 and 31 it just calls Java methods
through JNI anyway.
I've refactored the way Display updates are reported so that AndroidDisplay can
just dynamically get the refresh rate correctly. These values are used by a
service protocol extension and by some Stopwatches on the CompositorContext.
See also #29765Fixesflutter/flutter#93688Fixesflutter/flutter#93698
Several arrays pointers on semantics update objects were being passed as
the address of the 0th element; however, this is UB in the case where
the array length is 0.
We already test for this in embedder_a11y_unittests.cc and the
associated fixture:
78a2941957/shell/platform/embedder/fixtures/main.dart (L182-L184)
We didn't catch this since we don't currently run the embedder tests on
Windows.
Issue: flutter/flutter#93338
Context creation options for each backend were spread across multiple
translation units. This makes setting options common across all backends hard to
configure. I have moved the creation of such common options into a separate
translation unit.
Fixes https://github.com/flutter/flutter/issues/84213
