| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
| |
Impose a restriction of 200MiB for tables in stsc, stts, ctts and stss
boxes. Also change mTimeToSample from Vector to array.
Bug: 29367429
Change-Id: I953bea9fe0590268cf27376740f582dc88563d42
Merge conflict resolution of ag/1170200 to mnc-mr2-release
|
|
|
|
|
| |
Bug: 28816827
Change-Id: I3d5bad4a1ef96dec544b05bb31cc6f7109aae0a5
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
28076789.
Detail: Before the original fix
(Id207f369ab7b27787d83f5d8fc48dc53ed9fcdc9) for 28076789, the
code allowed a time-to-sample table size to be 0. The change
made in that fix disallowed such situation, which in fact should
be allowed. This current patch allows it again while maintaining
the security of the previous fix.
Bug: 28288202
Bug: 28076789
Change-Id: I1c9a60c7f0cfcbd3d908f24998dde15d5136a295
|
|
|
|
| |
Change-Id: I079d1db2d30d126f8aed348bd62451acf741037d
|
|
|
|
|
| |
Bug: 27207275
Change-Id: I4412825d1ee233d993af0a67708bea54304ff62d
|
|
|
|
|
| |
Bug: 24310423
Change-Id: Iebcfc58b447f925ec2134898060af2ef227266a3
|
|
|
|
|
|
|
|
|
|
|
|
| |
Because the constructor of NuCachedSource2 sent a message to
AHandlerReflector object, AHandlerReflector::onMessageReceived could
have executed just before the object gets wrapped in a strong
pointer, resulting in erroneous early free. Fix the issue by using
static Create function to ensure the message is sent after the
object is wrapped in a sp.
Bug: 23882800
Change-Id: I38a9d7a3083f184b4c81d0b00ba1661721278855
|
|
|
|
|
|
|
| |
SampleTable.
Bug: 23247055
Change-Id: I29ef59c7ff09248063714e5013f7c33f66c5eebd
|
|
|
|
|
| |
bug: 13222807
Change-Id: I6f6f417422d3a18117b594670bb23e3019d449bb
|
|
|
|
|
|
|
|
|
|
|
| |
- Added FRAME_RENDERED event in OMX, used by tunneled video decoders
to signal rendered event timing
- Track buffers sent for rendering in ACodec and in SoftwareRenderer, and
determine when they have rendered
- Propagate render times to MediaCodec
Bug: 20503131
Change-Id: Idf0a8714d5368b237c2285dd39fa82db847c232f
|
|
|
|
|
| |
Bug: 20503131
Change-Id: I762c419ed1245f8b83fb1f6bf61e5557213ca07b
|
|\ |
|
| |
| |
| |
| |
| | |
Bug: 19286916
Change-Id: I176d803aa5a16038d074aa06b41fc76da0be8952
|
| |
| |
| |
| |
| |
| |
| | |
Pass Fence between Surface and ACodec, and between ACodec and IOMX.
Bug: 12386081
Change-Id: Ifdc566979dec0d91ed8b07c3b69d2cf092accc73
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
- Add fence to new metadata buffer mode structure, but don't use it
yet, so that we don't have to add another mode soon.
- Change GraphicBuffers to ANativeWindowBuffer in new metadata mode.
This removes internal depencency from vendor codecs.
- Extend new metadata mode from SW encoders to all codecs.
- Fallback from new metadata mode to old mode (so Camera can always
use the new mode even with legacy codecs)
Related-bug: 19614055
Bug: 13222807
Change-Id: I405a21dd6ce8e99808f633bfeab8f14278eb7079
|
|/
|
|
|
|
|
|
|
| |
Specify allottedSize for useBuffer and allocateBufferWithBackup,
where OMX will see only the allottedSize, even though framework may
allocate and use a bigger buffer.
Bug: 13222807
Change-Id: Ibfa93f6c7b7c8ca0853ef17d4a0758e496eaf921
|
|
|
|
|
| |
Bug: 20634516
Change-Id: Iac9eac3cb251eccd9bbad5df7421a07edc21da0c
|
|
|
|
|
|
|
| |
TODO: Use bandwidth-based estimation to seek forward long period.
Bug: 20126845
Change-Id: I5e2f90784a9ce0dce348715dfcfc4f83ee196170
|
|
|
|
|
|
|
|
| |
and clear persistent surface pointer when MediaRecorder is reset
bug: 21045118
bug: 21045402
Change-Id: Ifd0d6deeb969f2252123929541b30b518cecbf9a
|
|\ |
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Propagate OMX_ErrorInsufficientResources when handling
node-allocation failure any other error from the component.
bug: 20132289
Change-Id: I1ed9626dab947227e3d5e983bbbeee89ad7b2d2a
|
| |
| |
| |
| |
| |
| |
| | |
Bug: 19127604
Bug: 19489395
Change-Id: Idaf1cc9008016f66903e93907a676f54e342e1a3
|
|/
|
|
|
|
|
| |
Bug: 18249558
Bug: 19666434
Bug: 20057497
Change-Id: I5868b17423d7c20cfaf4a399f3eb67bfba440605
|
|
|
|
|
|
|
|
|
|
|
|
| |
Without a cache the mediaserver does a lot of small reads which result
in round trips through binder and jni to the app MediaDataSource.
On a Nexus 5 I measured time to first frame from MediaPlayer for
1) 1350kbps h264, and 2) 20480kbps vp8. Without a cache, MediaDataSource
was ~250ms slower than an fd. With a 2kb cache it's 30ms slower for (1)
and 70ms slower for (2).
Change-Id: If1e811db7b853c4f79430603318d4744ac30acb9
|
|
|
|
| |
Change-Id: I81f438ae444f04c12ae27ae4ef6d073033de172c
|
|\ |
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This allows apps to implement MediaDataSource, which is modeled on
stagefright's DataSource, to supply media data to the framework. This
was already implemented for MediaExtractor, but it was renamed from
DataSource.
MediaExtractor, MediaPlayer and MediaMetadataRetriever each have a new
overload: #setDataSource(android.media.MediaDataSource)
Only NuPlayer supports this new data source.
The change introduces:
* IDataSource: The binder interface for DataSource.
* JMediaDataSource: The native counterpart to the java interface. It
implements IDataSource.
* CallbackDataSource: A stagefright DataSource that wraps an
IDataSource.
Change-Id: Ib3c944b49cc8a792c8eb9c85e5015c07f298ebc1
|
|/
|
|
|
| |
bug: 19460202
Change-Id: I20afded0b7eec45b2f25b51cf59d9f48eada74d8
|
|\ |
|
| |
| |
| |
| |
| |
| | |
bug: 18520722
Change-Id: I213e4bde92a7830543c1815b324164b27d853049
|
|/
|
|
|
| |
bug: 19297083
Change-Id: I5b89e35cc2bd5fb2f332872941e7978f15a6f403
|
|\
| |
| |
| |
| |
| |
| | |
bandwidth estimate
* commit 'ce25d85ad22e6df4b861d17e9e67cb6d0e62c363':
stagefright: httplive: Decouple block size from bandwidth estimate
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
A very small block size in PlaylistFetcher can lead to framework
overhead and difficulty streaming high bitrate content, but since
HTTPBase keeps a constant history of the past 100 HTTP reads, the
block size directly affects bandwidth estimation and in turn,
switching latency.
Add setBandwidthHistorySize() to HTTPBase to allow setting the
history size for bandwidth estimation. Call this within LiveSession
based on the current block size to ensure that the number of bytes
used for estimating bandwidth does not change if the block size is
changed in PlaylistFetcher.
Since a single TCP/IP packet can contain up to 64k of data, increase
the block size in PlaylistFetcher from 2k to lcm(188, 1024) or 47k to
avoid inaccuracies in read timings due to up to a comparable 47 reads
from the same locally-cached packet instead of from the network.
Also make HTTPBase::addBandwidthMeasurement() virtual to allow
bandwidth estimation extensions that do not rely on a history list.
Bug: 18821145
Change-Id: I5f957be01f5346e74cfb7eeb150ca4b397ad5798
|
|\ \
| |/
| |
| | |
Change-Id: I39fdc2e8895e1e943749b9a2628656a8fa5bb72b
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This is to restore patch attributions
This reverts commit f580806d893c4631f5324ff0af5c2db68a40ef42.
Bug: 18821145
Change-Id: Idc49385fffccfde2a3915388fe3fe4e2b740d787
|
|\ \
| | |
| | |
| | |
| | | |
* commit 'cec272dfe2cf6bf6cdb8a4afa5afdd0e910c915f':
Replace MidiFile player with a Midi extractor
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
This gets rids of a bunch of special midi handling and replaces it
with an extractor that works with NuPlayer and MediaMetadataRetriever.
Change-Id: I8d0f5bbdde2ca24267cf4d62ab26afe9630e0217
|
|\ \ \
| | |/
| |/|
| | | |
Change-Id: I4313941f3561176ce9f6ab055678fb626e570107
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
This commit consists of:
http://go/pag/c/188753 Add NULL check for empty playlist
http://go/pag/c/188754 Fix deadlock for low duration clips
http://go/pag/c/188757 Create a copy of last enqueued metadata
http://go/pag/c/188755 Propagate target duration to LiveSession
http://go/pag/c/188762 Decouple block size from bandwidth estimate
http://go/pag/c/188756 Reduce memcpy calls for chunked content
http://go/pag/c/188758 Dont resume if we have almost fetched till stop time
Bug: 18821145
Change-Id: I7fd650999c6c50bbadffd65adee9020e669dfe62
|
|\ \ \
| |/ /
| | |
| | | |
Change-Id: Ibaea3a0766f75a61964615482a9ef28651a42b37
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
- move logic to set encoder parameters to common encoder class
(similarly to what was done for decoders)
- set compressed buffer size based on frame size and compression ratio,
and codec-specific minimum size
- set raw buffer size based on frame size and metadata mode
- do not set stride and slice height on compressed ports
- advertise only QCIF support for google H263 encoder
- set large-enough input size for video decoders to support adaptive
playback
- do not change input buffer size on output port reconfiguration, as
no input buffer reallocation takes place
- do not return empty buffers with EOS after EOS has been signaled
Bug: 18513091
Bug: 18639027
Change-Id: Ib13492ef66adf331aa4572c67d2b283ea020cb41
|
|\ \ \
| |/ /
|/| /
| |/
| | |
* commit 'd86450b93ab3960ccd68aed14563aaa76b43ce9b':
Fix struct vs. class mismatches
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Fix something being declared a struct and forward-declared a class and
vice versa.
gcc doesn't care, clang complains about the mismatch.
Change-Id: I7a7a8e9040e3c86fa1d73eb112631ea140861657
Signed-off-by: Bernhard Rosenkränzer <Bernhard.Rosenkranzer@linaro.org>
|
|/
|
|
|
| |
Bug: 18285408
Change-Id: I56346f4652b2c5f7ef5fa3003fd8676051339384
|
|
|
|
|
|
|
|
|
| |
- added SoftVideoEncoder for common color conversion and
extension handling logic
- fix YUV420 SemiPlanar handling that should be NV12 not NV21
Bug: 17935149
Change-Id: I9b8d05678b1862dd37bf349ea83d67bdf1bb5560
|
|
|
|
|
| |
Bug: 17935149
Change-Id: I6bb5dd654e498a7153410afc052c2c8f7f35e44d
|
|
|
|
|
| |
Bug: 17326758
Change-Id: I9b0c281d92cf1803e275b1768e9edab9404ea577
|
|
|
|
|
| |
Bug: 17326758
Change-Id: I245e2bd1490e810094ca550fdddfa87075bc2056
|
|
|
|
|
|
| |
Bug: 17425250
Change-Id: Ieb4c93fd9848489f9dbf35a36474376924fd8eb9
|
|
|
|
|
|
|
| |
This covers both MPEG4 and H263 adaptive playback.
Bug: 17326758
Change-Id: I80a67b7f3ceab05e792f0a459439a8274bd78e20
|