| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
| |
Change replyID-s from uint32_t to an object
Move reply handling into the loopers (to reuse a common mutex)
Bug: 19607784
Change-Id: Iaa035b846c424c5687ed17ce1079b325e86c54be
|
|
|
|
|
| |
Bug: 19607784
Change-Id: I94cddcb81f671422ad4982a23dc4acfe57a9f1aa
|
|
|
|
|
|
|
|
|
| |
Async mode reduces the number of messages posted between
MediaCodec and MediaCodecSource. This reduces thread
wakeups and helps reduce CPU utilization.
Bug: 18246026
Change-Id: I4b0837f309fdd12e323c1dfa72525f5a31971a03
|
|
|
|
|
| |
Bug: 17514968
Change-Id: If232f92d163deb2440b927315c69e0c4d51ca290
|
|
|
|
|
| |
Bug: 17454455
Change-Id: Ia423bcc2e1fa39137f114eac44912ed15357bb99
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Currently CameraSource/AudioSource's stop() and read() are both called
from the puller's looper. This works if source operates normally (i.e.
read() returns regularly before source is stopped), as the stop() will
eventually be handled by the looper. However, if for some reason the
source hang, it will get stuck in read(), and the stop() will never
be processed, which could lead to ANR (in addition to the source hang).
We need to move the source's stop out of the puller's looper. It also
can't be on MediaCodecSource's looper, because the source's stop
synchrounously waits for all outstanding buffers to return, these
are only returned when MediaCodecSource's looper processes the buffer.
This change moves the stop to MediaCodecSource::stop, after encoder
is shutdown.
Bug: 16522726
Change-Id: Ie91f563c5d8a98ab091bf1945af4e51f662b9403
|
|
|
|
| |
Change-Id: I3d1146714fa23be3d4e696599b6f70cac1f9d28b
|
|
|
|
| |
Change-Id: Ie3bae3f037730e316d7fca12e7a3527973f752ef
|
|
Bug: 12305192
Change-Id: I72d7cb571be5bd348b58ad650f3269d24c15d350
|