Currently, it is possible to notify frame metrics observers that
registered to receive present time about the same frame twice.
This is wasteful, but also requires the metrics listener to guard
against duplicates. See https://b.corp.google.com/issues/169866723#comment23
To avoid this issue, only notify about metrics with present time just
before the frame is advanced to the next position. This ensures that we
are always notifying about the last frame that's about to go away.
Bug: 169866723
Test: added a crash to LatencyTracker for the case where we are notified
about non-existent record. The record disappears once it's complete. So
the first call was completing the record, and the second call was
causing a crash. Now, the crash is gone with this patch.
Change-Id: Ia247b4a0010b8ecb016ac66e44e211ec4f40bf8f
To measure end-to-end touch latency, we need to report the actual
present time to ViewRootImpl. ViewRootImpl, in turn, will report this
information to InputDispatcher. Finally, InputDispatcher will combine
all known information for a specific input event, and will report this
data to westworld.
In another patch, we will add a new call, 'reportLatencyInfo', to
InputPublisher. This call will allow the app to send this latency data
to InputDispatcher.
Bug: 169866723
Test: printed the input event present times inside ViewRootImpl
Change-Id: Ibd3a2cfeb1a340eb15cd2165071df1f8589634af
Present time is available to apps, but is not currently stored inside
FrameInfo. In this CL, we add a location for storing the present time inside
FrameInfo.
Currently, the metrics reporter is triggered after the buffer is sent to
SurfaceFlinger. That means, metrics for anything that happens after gpu
draw completes are currently not available.
In a future CL, we will populate the 'presentTime' field.
In a future CL, we will register a metrics observer inside ViewRootImpl
in order to send this data to InputDispatcher. This will enable
end-to-end touch latency metrics collection.
Bug: 169866723
Test: printed present time inside FrameTimeline.cpp (surfaceflinger
side) and compared to the values printed inside a metrics observer
registered in ViewRootImpl (not done in this CL)
Test: atest ViewFrameInfoTest
Change-Id: I7d0f8c5d7b5a2572abdc4e107123e8938a36f582
- Add SurfaceStatsCallback to TransactionCompletedListener
- Register a callback in RenderProxy to be called when we have
surface stats from SF via the BLAST callback.
- Instead of finishing a frame for frame metrics reporting
immediately, wait until BLAST callback fires, note GPU completion
time and finish frame.
- Expose GPU_COMPLETION in FrameMetrics
- Modify TOTAL_DURATION to also include GPU_COMPLETION
Test: FrameMetricsListenerTest
Fixes: 171046219
Change-Id: I16fa1d80cfc4e7a5527c18fec7e885409f17ee4d
add a method, setSurfaceControl, for java layer to pass surface control to the render thread
Bug: 173671170
Test: call setSurfaceControl method in ViewRootImpl.java
Change-Id: I886a79c377938f19cf38b9058f2bec64e1439000
As HWUI uses vsync id to send buffers to SF, SF applies the right
amount of render ahead by latching the buffers at the configured time.
Test: launch an app and observe systrace
Bug: 178148035
Change-Id: Ifd3e1a2971aad0a085cb35d33e950194046aa634
0 and -1 both meant default when 0 should mean
0 and -1 should mean default
Test: manual
Fixes: 179290765
Change-Id: Ia9aa5e3d83757282bfff776e083d6b3d7d29e9c0
The application receives input events and produces graphic buffers in
response. We are interested in measuring the total time that the input
event takes to process, from the moment it's received on the device, to
the moment that the image is displayed on the screen.
To do this, we need to understand which input event produced a specific
buffer. In this CL, we are sending the input event id from FrameInfo to
SurfaceFlinger. This event id will later be used to identify a specific
frame, and provide the frame timing information to inputflinger.
Inputflinger will be able to use this information to reconstruct the
complete event timeline, and record metrics on the duration of each
input / graphics processing stage. This will allow us to optimize
end-to-end touch latency.
In the current CL, we are using 'NewestInputEvent' as the inputEventId.
That's not quite correct. Today, this field contains the timestamp of
the input event. Therefore, we will simply pass the truncated
timestamp instead of proper input event id. We will fix this in a
separate CL by providing input event id to FrameInfo.
Design doc: https://docs.google.com/document/d/1G3bLaZYSmbe6AKcL-6ZChvrw_B_LXEz29Z6Ed9QoYXY/edit#
Bug: 169866723
Test: printed input event id at the site of SurfaceFrame creation in
FrameTimeline.cpp
Change-Id: Ia90337bb2f000e9c93a4db04d9dd6ea9ea153520
If returning early from a draw, there is a chance that the frame
complete callbacks can execute before the frame callbacks. To fix
this, explicitly wait on frame callbacks to complete.
Test: Enable blast and test split screen resize
Change-Id: Id0eabd4dacdccbf1ce774efaad51abae7c57d5ea
Pass the frame deadline calculated by SF to Choreographer so
hwui would be able to improve its stats by knowing if a frame is
likely to be late.
Bug: 169858174
Test: manual
Change-Id: Ib9fd93638b54f08d8dc72fa6b023e2dd7c276dc7
The logic in setBufferCount was still assuming it needed
to inject the extra buffers, however that injection was
already happening in the min_undequeued query. So it was
increasing buffercount by 2x the extraCount, or concretely
5 buffers when it should have been 4 on >75hz devices
Bug: 168928692
Test: Boot coral, verify 4 buffers instead of 5
Change-Id: I0908228a6791d90544fbf4cb21170931bd31a9db
Add plumbing to get shared timeline data from Surface Flinger to HWUI
and back.
Bug: 162890382
Bug: 162888881
Bug: 162890407
Test: manual using ALOG messages in SF
Change-Id: I85c5eb56d2fa933889518854e7b79b1e65e29af3
Hook MIN_UNDEQUEUED_BUFFERS if possible to avoid thrashing kgsl
maps when render_ahead is being used
Bug: 143555869
Test: verified kgsl maps only happened once per buffer
Change-Id: I985fae0a9a7635be3a1cf6177186e5541a1169df
There are cases where ViewRootImpl requests to draw, but there was
nothing new to draw. In that case, the callback will never be invoked
and ViewRootImpl will wait forever.
This change will invoke the callback even if there is nothing to draw.
It will use the last frameNumber since nothing new has drawn
Test: Request draw with nothing new. Callback is invoked
Fixes: 155429223
Change-Id: I7c9ed7fd63a451b17133a11ffbcf8fb64be558e5
If ReliableSurface is forced to acquire a fallback buffer then the
Surface may be abandoned. When getting frame timestamps we need to check
that the Surface still exists.
Bug: 152262035
Test: builds, boots
Test: dumpsys gfxinfo
Change-Id: Ifdb198ebf74cc9dc681c4ab51c4901176a7f5fc9
ANativeWindow usage now has enough feature parity so that we can use
that instead.
Bug: 137012798
Test: builds
Test: Scroll through settings
Change-Id: I0054315058b28bcb5e779a6f71a3cfb164625a5f
These aren't finalized as stable yet until we sort out some BufferQueue
challenges.
Bug: 137012798
Test: builds
Change-Id: Icd5eb02afc4cf16a2be485e6710f71818b300b9d
Previously we didn't assign a colorspace to the pipeline until it
was provided a surface to render into. This resulted in undefined
behavior if the application attempted to render an offscreen layer
before the OS provided the main window with its surface. Now instead
of deferring setting whether or not the application is wide gamut we
do initialize it to a default setting when the pipeline is created.
Bug: 148042673
Test: apct/device_boot_health_check_extra_postsubmit
Change-Id: I84d743511e949ac977486470bb14eec936de7f88
Adjust the target present time when using
render_ahead to better fit in the gaps between
vsync pulses. This prevents small clock
drifts from causing jank.
Bug: 143540628
Test: manual, looked at systrace
Change-Id: I26ae6ebf16a3dea48dfd8e1c420d9e5e43e42bcb
This is part of the work to move away from gui/Surface and onto
ANativeWindow in HWUI.
Bug: 137012798
Test: builds
Change-Id: I5862d7833b583b8295d1677e725ed58d97808b27
Add ATRACE memory counters to track memory usage at the end
of each frame in HWUI.
There are 3 catagories: CPU, GPU and Texture memory.
There are 3 more counters for memory that can be purged.
This CL deletes GpuMemoryTracker class, which implemented
similar function for Android O HWUI renderer.
Test: Collected systrace with gmail and setting.
Test: memory tracing adds ~0.1ms per frame when ATRACE is enabled
Bug: 146580770
Change-Id: Icbcc0478bc426dff578e83726fe7c95df171ed93
By avoiding setting a dequeue buffer timeout we avoid
hitting a different path in BufferQueue that prevents
async behavior from happening. This restores P's
behavior in this path.
Bug: 143860379
Test: repro app in bug
Change-Id: Iffbd9f9e6689a40876ff3aa74c10020e3f09fc6a
Invoke Surface::enableFrameTimestamps after eglDestroySurface.
eglDestroySurface internally disables time stats. Order is
important, when CanvasContext::setSurface is invoked with
a surface, that is already current (which happens all the time).
Test: ran UiBench microbenchmark tests
Change-Id: I3d023c3a87da6329c556426d553c744e541b9dff
A new instance snuck in since I haven't completed the platform toolchain
upgrade yet. There was imprecision in this call; accept it by making the
implicit cast explicit.
Bug: 139945549
Test: mm
Change-Id: Id704d9741b480eba4cf3955e70119d935776faf6
Signed-off-by: Nick Desaulniers <ndesaulniers@google.com>
We only need to redraw if the size /changed/ not
if it was the same. Also fix damageId to not use
frameNumber as repeated redraws of the same frame
would toggle.
Bug: 143711430
Test: systrace
Change-Id: I8ac4629c9ff4fd51de33d1be7aa46ccc995ba342
gui/Surface will eventually be removed since HWUI needs to depend on a
stable ABI, so use the NDK methods for querying width and height that
operates on ANativeWindow instead.
Bug: 137012798
Test: builds
Change-Id: I1309e31992190e7b44c6ea83f962f372b6b0afcf
This also removes the dependency on ui/DeviceInfo other than in test
code.
Bug: 136263392
Bug: 136263238
Test: builds, boots
Change-Id: I6a4687e981359f0e6beb83be8a5501ed7fd16f15
Poking around in a few apps it doesn't appear that
the VectorDrawableAtlas is achieving sufficient
utilization to justify its existence. The potential for
draw call merging doesn't seem warranted for the
RAM cost of the atlas.
Bug: 137853925
Test: builds
Change-Id: Id2419bc6dccb6316636d50c568f8fac75a2d563f
Now that they return nanoseconds, there's no reason to lose precision.
Bug: 137012798
Test: builds
Change-Id: I52ca20cb010b4f1829596e0a765b7013ae3665ca
Replaces the query() call for getting the duration of the internal
dequeue() call a call that will be abi stable.
Bug: 137012798
Test: builds
Change-Id: I6d2118a7c093dd2ecc4ebfb4b4ee8556f7e709a7
Handle TIMED_OUT better by rescheduling (TODO: give up after N
attempts?)
Fix SYNC_SURFACE_LOST_REWARD_IF_FOUND path to actually go fetch
a new surface.
Bug: 137509524
Test: Injected errors randomly, verified nothing got permanently dead.
Change-Id: Id30f8ad1dd7196041ee84c16c8cf5c814002a6ce
Timeout recovery isn't working properly, so increase it
to avoid any false-positives from something like a
slow screen-on
Bug: 137509524
Test: none
Change-Id: Ic4fae04b41f71600479f9557e334f67c27851db4
