| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
| |
Apparently there are a number of people and/or tools that name their
opus coded audio in ogg containers *.opus. Reflect that fact.
Change-Id: I0617a6d93fe793cdef06ffd4e334001e5f9b054e
|
|
|
|
|
|
|
|
| |
Usage of boot-time as the timestamp reference isn't necessarily limited to
HAL3. A single property check for media.camera.ts.monotonic is enough to
cover all cases.
Change-Id: Ic56dbcc6ba32c25bccde6dd0bbf07d4c918af43e
|
|\
| |
| |
| |
| |
| | |
Android 6.0.1 Release 61 (MOB30Z)
Change-Id: Ic86e42d5fb8f81fe5c4f58d2e76852829c609760
|
| |
| |
| |
| |
| |
| |
| |
| | |
Subtract address of a random static object from pointers being routed
through app process.
Bug: 28466701
Change-Id: Idcbfe81e9507433769672f3dc6d67db5eeed4e04
|
| |
| |
| |
| |
| | |
Bug: 28816827
Change-Id: I3d5bad4a1ef96dec544b05bb31cc6f7109aae0a5
|
| |
| |
| |
| |
| | |
Bug: 29251553
Change-Id: I1bcc1281f1f0542bb645f6358ce31631f2a8ffbf
|
| |
| |
| |
| |
| |
| |
| |
| | |
Exit for error in allocation and unsupported resolutions
Bug: 28816956
Change-Id: Ieb830bedeb3a7431d1d21a024927df630f7eda1e
|
| |
| |
| |
| |
| |
| |
| | |
because they may end up pointing to NULL after a NULL check was performed.
Bug: 28166152
Change-Id: Iab2ea30395b620628cc6f3d067dd4f6fcda824fe
|
| |
| |
| |
| |
| | |
Bug: 29153599
Change-Id: I1cb81c054098b86cf24f024f8479909ca7bc85a6
|
| |
| |
| |
| |
| | |
Bug: 28533562
Change-Id: I798ab24caa4c81f3ba564cad7c9ee019284fb702
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
* We're really only interested in music streams right now, but
events are being generated for all streams (system sounds, etc).
* Constrain for now, in the future we will filter based on client
registrations.
Change-Id: Ic445052028c454eed146addebcdb28c4b26c4f20
|
| |
| |
| |
| | |
Change-Id: I13f3ba8fc9caaeee21597a2405a8c3629e985502
|
|\ \
| |/
| |
| |
| |
| | |
Ticket: CYNGNOS-3020
Change-Id: I7e8d69c5f7041b66893ea643c4bc19c3b7bcdda5
|
| |
| |
| |
| |
| |
| | |
Change-Id: I336cb003fb7f50fd7d95c30ca47e45530a7ad503
(cherry picked from commit 33f6da1092834f1e4be199cfa3b6310d66b521c0)
(cherry picked from commit bb3a0338b58fafb01ac5b34efc450b80747e71e4)
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
track.
GenericSource: return error when no track exists.
SampleIterator: make sure mSamplesPerChunk is not zero before using it as divisor.
Bug: 21657957
Bug: 23705695
Bug: 22802344
Bug: 28799341
Change-Id: I7664992ade90b935d3f255dcd43ecc2898f30b04
(cherry picked from commit 0386c91b8a910a134e5898ffa924c1b6c7560b13)
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Limit mediaserver using rlimit, to prevent it from bringing down the system
via the low memory killer.
Default max is 65% of total RAM, but can be customized via system property.
Bug: 28471206
Bug: 28615448
Change-Id: Ic84137435d1ef0a6883e9789a4b4f399e4283f05
|
| |
| |
| |
| |
| | |
Bug: 28471206
Change-Id: Id5d055d76893d6f53a2e524ff5f282d1ddca3345
|
| |
| |
| |
| |
| | |
bug: 28175045
Change-Id: Icee6c7eb5b761da4aa3e412fb71825508d74d38f
|
| |
| |
| |
| |
| |
| |
| |
| | |
Report MEMORY_ALLOCATION_ERROR if pStorage->picSizeInMbs would
exceed UINT32_MAX bytes.
Bug: 28532266
Change-Id: Ia6f11efb18818afcdb5fa2a38a14f2a2d8c8447a
|
| |
| |
| |
| |
| | |
Bug: 28377502
Change-Id: Iad5ebfb0a9ef89f09755bb332579dbd3534f9c98
|
| |
| |
| |
| |
| | |
Bug: 28333006
Change-Id: Ief7a2da848face78f0edde21e2f2009316076679
|
| |
| |
| |
| |
| |
| |
| |
| | |
Upstream commit 0bb5ced removed support for this flag.
Reference: https://github.com/CyanogenMod/android_frameworks_av/commit/0bb5ced60304da7f61478ffd359e7ba65d72f181
Change-Id: I56c8c0957b556f70b1a52cd6dd8a2bdd8f958381
|
| |
| |
| |
| | |
Change-Id: Icdfe4662ff6b53cff53142706bbd5ade1636fb27
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
* getStorage is called both internally with mMutex held and externally
without mMutex held. Create getStorageLocked for internal use and
make getStorage a wrapper.
* hasStorage is only called internally with mMutex held. Make it a
private method.
Change-Id: I8f73310ad6cca14cd88b8e29f20cc181b3a4fac3
|
| |
| |
| |
| | |
Change-Id: I6d7202ade46a5d781a3db5a1a3bdde17c8e70a60
|
|\ \
| |/
| |
| |
| |
| | |
Android 6.0.1 release 46
Change-Id: I4fcabf5775aebd2cb8dc0e352d1b1460d3214573
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
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: 27855419
Change-Id: Idabedca52913ec31ea5cb6a6109ab94e3fb2badd
|
| |
| |
| |
| |
| |
| | |
Bug: 27833616
Change-Id: Ic4045a3f56f53b08d0b1264b2a91b8f43e91b738
(cherry picked from commit 87fdee0bc9e3ac4d2a88ef0a8e150cfdf08c161d)
|
| |
| |
| |
| |
| |
| | |
Bug: 27793163
Bug: 27793367
Change-Id: Iec3de8a237ee2379d87a8371c13e543878c6652c
|
| |
| |
| |
| |
| |
| |
| | |
to prevent info leak
Bug: 27855172
Change-Id: I3d33e0a9cc5cf8a758d7b0794590b09c43a24561
|
| |
| |
| |
| |
| | |
Bug: 27793371
Change-Id: I0fe40a4cfd0a5b488f93d3f3ba6f9495235926ac
|
| |
| |
| |
| |
| | |
Bug: 27833616
Change-Id: I0fd599b3da431425d89236ffdd9df423c11947c0
|
| |
| |
| |
| |
| | |
Bug: 27833616
Change-Id: I1ccdd16a00741da072527a6d13e87fd7c7fe8c54
|
| |
| |
| |
| |
| | |
Bug: 27207275
Change-Id: I9a7c9fb22a0e84a490ff09c151bd2f88141fdbc0
|
| |
| |
| |
| |
| |
| |
| |
| | |
since it doesn't follow the OMX convention. And remove support
for the kClientNeedsFrameBuffer flag.
Bug: 27207275
Change-Id: Ia2c119e2456ebf9e2f4e1de5104ef9032a212255
|
| |
| |
| |
| |
| |
| |
| | |
Otherwise the new size checks trip on this.
Bug: 27207275
Change-Id: I1f8f01097e3a88ff041b69279a6121be842f1766
|
| |
| |
| |
| |
| | |
Bug: 27207275
Change-Id: I4412825d1ee233d993af0a67708bea54304ff62d
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
When using the configurable policy engine, the audiopolicy makefile
implicitly declares dependencies on the configuration files. However,
the example folder already contains the example files which we wan't to
override.
This patch removes the example from the makefile inclusion path so that
we can declare it on a per-device basis without getting errored out by
duplicate module names.
Change-Id: Ie47ebff6b13606c93630c017252ebaa204203a76
|
| |
| |
| |
| |
| |
| |
| |
| | |
For the OMX_COLOR_FormatYUV420Planar color format case, Google's VP9
codec is used, in that case the colors were shifting 16 bytes per
horizontal line, causing green lines to appear on video playback.
Change-Id: I4cf0fd40b79e53882d99f1c3f263c8a98fff1f30
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
8909 is using soft-ip divx decoder using component
name starts with "OMX.ittiam." instead of "OMX.qcom."
leading to FFMPEGSoftCodec override with wrong compression
format using OMX_VIDEO_CodingDIVX (0x7F000003) defined in
OMX_FFMPEG_Extn.h. But qcom omx component actually use
QOMX_VIDEO_CodingDivx (0x7FA30C02) which defined in OMX_QCOMExtns.h.
Add checking for OMX.ittiam in additional to OMX.qcom solved the
problem.
FEIJ-700
Change-Id: I4071eb3978974d1072373f0c88ea83653c0c53a9
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This patch is buggy, and causing notifyConsume not called
when the buffer is erased, causing audio discontinous cannot
resume properly. This patch is not in 8994 CAF also.
FEIJ-276
This reverts commit dc5f0eb0bb08885f63a97a111b02e234efc5c413.
Change-Id: Idcb5ec9a419a979afe170d3a518ee7696a6fac6c
|
| |
| |
| |
| |
| |
| |
| |
| | |
Make sure all the items in the port param config are filled out, not
just those we expect to use.
Ref: CYNGNOS-2797
Change-Id: I9271a83fc6496cfe5ad14e45841cd8a05ffece30
|
| |
| |
| |
| |
| |
| |
| | |
Reverting the Temp change.
This reverts commit f300e17bf3451bf821ccc48c45b243c84ebaff7f.
Change-Id: Ib11e38eb8edd7523cb6cdb21caa511152ee62e45
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
We seem to get an improper offset when getting the value from the DSP
(offset 100, data size 1), when this happens, return a new phrase event
with valid extras which are initialized to some default values.
Change-Id: Ie041f78fec6bdbcd82286b54a7b7fb0c98c740f5
Signed-off-by: Roman Birg <roman@cyngn.com>
|
|\ \
| |/
| |
| |
| |
| |
| | |
Ticket: CYNGNOS-2373
Android 6.0.1 release 43 (MOB30J)
Change-Id: Ie0ab546349bd8a7eed51c839fba28b809e433fd4
|
| |
| |
| |
| |
| |
| |
| |
| | |
Previous change caused EOS to be ignored.
Bug: 27843673
Related-to-bug: 27662364
Change-Id: Ia148a88abc861a9b393f42bc7cd63d8d3ae349bc
|
| |
| |
| |
| |
| | |
Bug: 27662364
Change-Id: I47380545ea7d85845e141e722b0d84f498d27145
|
| |
| |
| |
| |
| | |
Bug: 27662364
Change-Id: I7b26892c41d6f2e690e77478ab855c2fed1ff6b0
|