Modified AudioFlinger duplicating output thread so that audio tracks are not mixed until both outputs (A2DP and hardware) have exited standby mode. This avoids to have one output far ahead of the other and audio frames dropped because the compensation mechanism cannot keep up.
Also calculate the maximum wait time in OutputTrack::write() based the on smallest frame count of all output threads instead of the frame count of the thread the OutputTrack is connected to. This avoids starving the thread with the smallest frame count by waiting too long on the other thread.
Since the frame count was reduced on hardware output to reduce latency the difference between A2DP and hardware outputs frame counts had become problematic.
Also increased the number of overflow buffers to cope with bigger timing differences among outputs.
* changes:
docs: add "required" attribute to the <uses-feature> docs and add a note to the CAMERA permission stating that it enables the <uses-feature> for camera. This is an incrimental update for the 2.0.1 docs. More changes to come for the 2.1 docs to include more features
* changes:
docs: update the Bluetooth guide with links to the sample app, scrub the sibling files, and revise the Bluetooth package summary to point to the BT dev guide.
* changes:
docs: fix some typos in the manifest docs samples and remove the "one or more of" phrase from the syntax snippet because it interferes with the syntax coloring.
* changes:
sdk doc change for esr: Add "resources" tab content. Fix links pointing to the old locations. Change Android.mk to output samples files to resources/samples. Misc other fixes.
Merge commit '6f95f461098d5b9e93212cf9cc7c7a6498eecf3a' into eclair
* commit '6f95f461098d5b9e93212cf9cc7c7a6498eecf3a':
Fix the dependencies to make online-sdk-docs.
When a surface is removed from the screen while it holds a "freeze lock", the
release of that lock happens in the destructor as a "safety net". However, it
doesn't trigger an update at that point.
Make sure that "freeze locks" are released from the transaction at the point
a surface is removed from the screen (if it's not on screen, it shouldn't
prevent the screen to redraw, and therefore cannot hold a freeze lock).
The refresh corresponding to that transaction will pick it up as soon as possible.
this was introduced by a recent change. when we try to figure out the size of
the yuv->rgb temporary buffer, the output resolution has not been computed yet
and an invalid buffer size is used. most of the time the allocation fails
and the system reverts to "standard" GL will uses onle the Y plane.
the allocation of the temporary buffer is moved to onDraw(), the first
time it is called, by that time, the window is positioned properly.
* changes:
Partially fix bug 2111240 Detect docking / undocking event by reporting to the AudioPolicyManager a new forced usage AudioSystem::FOR_DOCK which can take the FORCE_NONE, FORCE_BT_DOCK or FORCE_WIRED_ACCESSORY values. This CL is complemented by an update of the APM to take into account the FOR_DOCK usage.
always rescale videos to their target size using copybit during yuv->rgb
conversion. this improves performance of the GPU pass and doesn't require
linear filtering to be enabled. Also always use 16-bits buffers.
the average processing time for 720p dropped from ~50ms to ~30ms
