| Age | Commit message (Collapse) | Author |
|---|
|
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/15463251
Change-Id: I1e0ba658e2230d666dd8eaa7497a4986fece264b
|
|
With the current implementation we might mark layers as NoVote
incorrectly when a sibling has a frame rate.
Test: adb shell /data/nativetest64/libsurfaceflinger_unittest/libsurfaceflinger_unittest
Bug: 195205467
Change-Id: I62641f855a027e1192f9a85a4bc50692b1744764
|
|
Move RefreshRateConfigs to DisplayDevice to be able to maintain
a per-display refresh rate switching policy.
Test: SF unit tests
Test: refresh rate switching is working on device with
more than one display
Bug: 187539899
Change-Id: Ica6a955e8ad0e563a0740f6579b61fc592eb982c
|
|
Now that we only write to current state from the main thread
we can actually just write directly in to drawing state and
avoid copying current to drawing state on each frame. simpleperf
on bouncing ball indicates this was saving around 8% CPU or so
per frame. As a next target we will eliminate the remaining work
in doTransaction and eliminate the doTransaction pass.
Test: Existing tests pass. simpleperf
Bug: 186200583
Change-Id: I59b75747d7371b3ffe870129c8be100c7daa6c4e
|
|
Allocate HAL virtual display and generate HAL/GPU virtual display ID in
SF rather than CE. This centralizes the HAL vs. GPU decision as a first
step in isolating display configuration (e.g. hotplug, modeset) to be a
distinct stage from invalidate/refresh.
Rework SF backend hooks for screen capture. Plumb the PhysicalDisplayId
to be mirrored by the virtual display to Composer::createVirtualDisplay.
This enables the ARC backend to know which display to mirror (instead of
making assumptions about the layer stack) or error out if not mirroring
(previously done through maybeAllocateDisplayIdForVirtualDisplay), such
that SF falls back to creating a GPU virtual display.
Bug: 182939859
Bug: 129481165
Test: Enable overlay display and toggle HAL/GPU
Test: libsurfaceflinger_unittest
Test: libcompositionengine_test
Change-Id: I209b245966e544d5ff55d5d118140cfcfa85db15
Merged-In: I209b245966e544d5ff55d5d118140cfcfa85db15
|
|
There are no users left
Bug: 168505645
Change-Id: I81725bf3c0ef4704e9da25da9a75854b4f172885
|
|
No callers left outside of tests.
Bug: 161937501
Test: Existing tests pass
Change-Id: Id5c2b68d1be05fce9f698813d83044b02bb2764d
|
|
wp::promote has a significant performace impact. Instead we maintain
a copy of the relevant layer information in LayerInfo.
This shows reduction in simpleperf for the main thread
3.07% -> 0.81%
Test: simpleperf for PIP + Notification shade expansion
Test: atest SetFrameRateTest
Bug: 169873384
Change-Id: Ib1414a2db73a4c30a26379a0ba8ba9e639f920ce
|
|
Change the shouldBeSeamless parameter to an enum
in order to make the API easier to understand.
This changes
- SurfaceControl.setFrameRate
- Surface.setFrameRate
- ANativeWindow_setFrameRateWithChangeStrategy
- ASurfaceTransaction_setFrameRateWithChangeStrategy
Bug: 179116474
Test: atest SetFrameRateTest
Change-Id: I28a8863ea77101f90b878fbda5f00d98e075b7cc
|
|
When a layer frame rate vote changes (via setFrameRate call),
the layers needs to be activated in scheduler. This change fixes a
bug that was activating all layers instead of just the layers that
has a vote change
Bug: 163079696
Bug: 180014293
Test: run a test app that calls setFrameRate
Change-Id: Iaa538d04c535b185161a24a4bbadeff2ac99bcd8
|
|
The display modes should be stored only in DisplayDevice.
Having the state also in HWComposer is unnesesary and hard to
keep in sync with SF, e.g. during hotplug processing of
displays which can change their supported modes.
Any HWC calls which need to validate their parameters need
to go through display device. This additinally makes the
code more undestandable.
Bug: 159590486
Test: presubmit
Change-Id: I40b03c09a5fd6092fca0682d602deb70db022fa5
|
|
Add visibility to SurfaceFlinger into the high refresh rate deny list
and let SurfaceFlinger handle it. Previously WM was setting the
preferredDisplayModeId on the denied app's window. The old way prevented
SurfaceFlinger to use the frame rate override feature as it didn't
know that a specific app is causing the refresh rate spec to be limited.
With this change, SurfaceFlinger will limit the display refresh rate based
on the high refresh rate deny list, and if possible, will use the frame
rate override feature to change the display rate to a multiple, allowing
other animations to be smooth while the denied app remains in the low
refresh rate.
Bug: 170502573
Test: SF unit tests
Change-Id: Idc8a5fe6bc12dbd949ad5e09ff50e339ffaeac36
Merged-In: Idc8a5fe6bc12dbd949ad5e09ff50e339ffaeac36
|
|
If an ancestor of a layer set a frame rate, and that layer doesn't
have a frame rate configured, use the one of the ancestor. The reason
for this is to support the high refresh rate deny list where WM will
set the frame rate on a container layer which is not visible. With this
change we will treat all child layers of that container as if they voted
themselves.
Test: test app that sets preferredDisplayModeId
Bug: 163079696
Change-Id: I3c55d393af72e19cd7b4f107d8cc0b2e85289d96
Merged-In: I3c55d393af72e19cd7b4f107d8cc0b2e85289d96
(cherry picked from commit 20dcde8e35b003a82e4c14d39c76f93a16ec6e9d)
|
|
Test: presubmit
Change-Id: I7e7ec1eab615b183a0d2074d96c0a444e5b97e79
|
|
Add a class to wrap a fps value. This is useful because across
the code we
- convert between vsyncPeriod and fps
- compare with tolerance (this ensures we use consistent tolerance)
- consistent toString method
Bug: 159590486
Test: presubmit
Change-Id: Iebb77a33a2f822056642aa61bd6fac6514aa656d
|
|
Add a new event to DisplayEventReceiver that conveys a list
of apps whose frame rates have been overriden.
This event is processed by the DisplayManager to advertise the
new frame rate to the overridden app.
Change-Id: I89fce8b5b4d9db65ec5db7dd2393c384c0fcfd82
Bug: 169271059
Bug: 169271062
Bug: 170503758
Test: manual test using SF backdoor
|
|
Add the ability for SurfaceFlinger to be able to throttle down to
a divider of the refresh rate (i.e. for 30/45 for 90Hz)
Change-Id: I6bfd6f43ee1f30e771a136c558d8ae9a6d7fbe0f
Test: Manually via 1039 SF backdoor
Bug: 170502573
Bug: 169270763
Bug: 169271059
|
|
Allow transaction clients to specify the vsyncId that started the
transaction and plumb this data to FrameTimeline to be able
to track jank associated with transactions
Bug: 166302754
Test: manually see transactions in frame timeline
Change-Id: Id05e0d0a73039204943d93b666cb67e3e7515a69
|
|
Refactor the old DispSync to separate between Vsync math (VsyncTracker)
and fence / vsync callback management (VsyncController)
Bug: 162888874
Test: examine systraces
Test: SF unit tests
Change-Id: Id275620380a21aeb0017e966910cbf24860cecef
|
|
EventControlThread was a HWC workaround dating back to K for toggling
VSYNC off the main thread, but as of R it defers back to main only to
incur context switches.
Clean up TestableScheduler construction to skip creating DispSync and
timer threads, which fixes several gMock warnings of unexpected calls.
Remove virtual destructors for non-polymorphic interfaces.
Bug: 160012986
Test: systrace
Test: libsurfaceflinger_unittest
Change-Id: I01360016a7dba79dc905f250753e6fce34af0a90
|
|
If SurfaceFlinger wakes up too late (due to the main thread being busy)
it may calculate a wrong presentation time as it is based on the wake up time.
Instead, cache the intended wake up time and use it as the base time.
Change-Id: I84c37bd5f164fbdef3151c0898d3cea624cea515
Bug: 154199567
Test: collect systrace for launcher scrolling
|
|
If an app set a framerate on a layer, then getFrameRate() on
ancestors/successors of this layer should return NoVote to allow the
refresh rate heuristic to be based on the the actual layers that voted.
Bug: 151274728
Test: Swappy with ANativeWindow_setFrateRate
Change-Id: Icfdf8e8fd4d92ba520bbc894bb9971b980691518
|
