Bug: 27286867
If the system/app is slow, it might take too long to
stop drawing. Switch the ordering of destroying stuff so
that we switch to the pbuffer surface first, then do
cleanup
Change-Id: If64a3dbb71bb9fd53567231590436a89b2f1a09e
A slow listener could cause a race in the NotifyHandler
where the single reference to the buffer to send would get
updated when it shouldn't have been.
Switch to a queue of available buffers to prevent this race.
Also, stop setting and clearing the observer reference and instead
incStrong/decStrong to mark temporary strong ownership without
colliding with other owners in flight.
Bug: 27097094
Change-Id: Iee647bfae8b80019b6d8290179eed3973230901f
- Rename to FrameMetrics to avoid collision with existing
android.view.FrameStats class
- Make FrameMetricsObserver implementation detail,
exposing FrameMetricsListener interface as public API
and wrapping in FrameStatsObserver to maintain state
- Remove dropped frame count call, in favor of passing as
parameter to callback method.
- Move away from raw timestamp access in favor of Metric IDs
which represent higher-level, more stable stages in a frame
lifecycle and match the categories exposed in the onscreen
bars.
- Support many-to-many Window<->FrameMetricsListener relationship
Change-Id: I00e741d664d4c868b1b6d0131a23f8316bd8c5c2
Not using EGL_SWAP_BEHAVIOR_PRESERVED_BIT as config attribute for
eglChooseConfig doesn't automatically mean the swap behavior is buffer
destroyed. This is driver implementation specific and on some hw this
can still be buffer preserved. Make sure it is buffer destroyed by
explicitly setting it for every new surface when requested.
Change-Id: Ie2c7c89b0d20e35832b488c6263bb4d9dd844a75
Signed-off-by: Christian Poetzsch <christian.potzsch@imgtec.com>
If eglSwapBuffers is called but the under surface was destroyed,
the EGL_BAD_NATIVE_WINDOW error may also be generated according
to the EGL spec 1.4.
This really shouldn't happen from the upper, but add the graceful
handling of this case also.
Change-Id: Ic0a599808b72f401d2a01c3dc40f9e6ea0e0a564
Signed-off-by: Zhang Dongsheng <dongsheng.zhang@intel.com>
Signed-off-by: Zhiquan Liu <zhiquan.liu@intel.com>
Full GLES error checking layer via -include
trickery. Change DEBUG_OPENGL to a level system.
HIGH = every GL call is error checked
MODERATE = checkpointing at interesting spots
LOW = only asserts there are no errors at the end of a frame
or when the FBO changes
NONE = AIN'T GOT NO TIME FOR ERRORS GOTTA GO FAST!
Change-Id: Ibe81aae93d942059c4ddf1cbb11c828b7ce4c10b
Skia's SkCanvas::SaveFlags are being deprecated. This CL introduces
the equivalent android::SaveFlags, converts all internal clients to
the new enum, and switches the saveLayer glue to the
SaveLayerRec-based API.
Change-Id: Icb1785f4e7c0f652b1f04b34a1e3ccb063c408f3
getMaximumBitmapWidth() and getMaximumBitmapHeight() of DisplayListCanvas
need HWUI cache instance. Since the initialization of the cache is
asynchronous it may crash if not yet ready. Add a staticFence() call
to guarantee the cache has been created prior issuing the call.
Change-Id: I5ed9e5cc084444c8d1872a77fef50e294ae14e93
Signed-off-by: Thomas Buhot <thomas.buhot@intel.com>
Signed-off-by: Zhiquan Liu <zhiquan.liu@intel.com>
am: 099bd9ca8f
* commit '099bd9ca8fea02795424d62c05c723290d68ae14':
In CanvasContext::doFrame, make a separate call to computeFrameTimeNanos and save the result. Then pass that value to UiFrameInfoBuilder::setVsync as both arguments.
am: 6222bb0f6a
* commit '6222bb0f6a772c8fa1dc402740399b0ad1017520':
In CanvasContext::doFrame, make a separate call to computeFrameTimeNanos and save the result. Then pass that value to UiFrameInfoBuilder::setVsync as both arguments.
and save the result. Then pass that value to
UiFrameInfoBuilder::setVsync as both arguments.
The order of function argument evaluation is undefined in C++. Because
the value returned from TimeLord::latestVsync may be changed by
the preceding call to TimeLord::computeFrameTimeNanos the values of the
arguments passed to UiFrameInfoBuilder::setVsync is also undefined. This
change removes any ambiguity.
Change-Id: Ie71ee453f9ccc725edfe5f7cc9b277f2a809dfdc
On the critical path of the cold launch of applications
the main thread of the started application tells the RenderThread
to create a surface. This process is synchronous and blocks
the main thread of the application until the creation
of the EGLContext is complete.
As a consequence the launch time of the application is delayed
by time spent allocating the EGL Context in the RenderThread.
With this optimization the launch time of any application
is improved (for example settings by 20 to 40 ms).
Change-Id: Ibf47aaa0abb8dedf7aa00693073db3785d9d6b08
Signed-off-by: Thomas Buhot <thomas.buhot@intel.com>
Signed-off-by: Zhiquan Liu <zhiquan.liu@intel.com>
Bug: 25843358
If the time between last swap & current vsync grew larger
than 2 seconds it would overflow when placed into an int, causing
frames to be dropped as negative numbers are definitely
less than 2_ms.
Change-Id: Icd2136989e5bbf5a0e21611b95a4d515b5ff9b14
