| Commit message (Collapse) | Author | Age | Files | Lines |
|\
| |
| |
| |
| |
| | |
Android 6.0.1 Release 61 (MOB30Z)
Change-Id: Ic86e42d5fb8f81fe5c4f58d2e76852829c609760
|
| |
| |
| |
| |
| | |
Bug: 29153599
Change-Id: I1cb81c054098b86cf24f024f8479909ca7bc85a6
|
|\ \
| |/
| |
| |
| |
| | |
Android 6.0.1 release 46
Change-Id: I4fcabf5775aebd2cb8dc0e352d1b1460d3214573
|
| |
| |
| |
| |
| | |
Bug: 27207275
Change-Id: I4412825d1ee233d993af0a67708bea54304ff62d
|
|/
|
|
|
|
|
| |
- handle non default input buffer size
- fix issues associated with non-default buffer size
Change-Id: Iad31f5c85601983b01e18e130b1c3eecb3988e30
|
|
|
|
|
|
|
|
| |
Detected with AddressSanitizer.
Bug: 18202869
Change-Id: I7fa9ee722766968e33faa4d9312385ca303ca274
(cherry picked from commit a37f84ecce59a0d11a687b55527ccb18d4db7f52)
|
|
|
|
|
| |
bug:19656539
Change-Id: I27f41d885f4fd5bb5a0e29510f808fb84a0f75f3
|
|
|
|
|
| |
bug: 19602975
Change-Id: Ie2e5a61d627980762080c4a8701b578143d123fa
|
|\
| |
| |
| |
| | |
* commit 'f3c253464e0047a28960b27bbee7366d5bd63327':
Cast int64_t values to long long before print.
|
| |
| |
| |
| |
| |
| |
| |
| | |
So the "%lld" format will not have Clang warnings in 32-bit and
64-bit builds. Alternative is to use PRId64.
Change-Id: I5fafc506fae012748a04ec261af62d9d13475cde
Signed-off-by: Bernhard Rosenkränzer <Bernhard.Rosenkranzer@linaro.org>
|
|\ \
| |/
| |
| | |
Change-Id: Iea12c8a6cabf84584e4a89ad80e298c1f4ea3dd7
|
| |
| |
| |
| |
| |
| |
| | |
For build-system CFLAGS clean-up, remove unused functions and
variables.
Change-Id: Ic3dee56b589ea9a693efa1d72ba394036efff168
|
|/
|
|
|
|
|
| |
Just signaling EOS is sufficient
Bug: 17933838
Change-Id: I04a1af57378115731febe7cacb35af5e55d5db83
|
|
|
|
|
|
|
|
|
| |
When the AAC decoder encounters an error, the lists of submitted
and consumed sizes need to be resynchronized, to prevent timestamp
drift.
Bug: 17638652
Change-Id: Ia6d5f48dc36be355803589e4a5796b01559942b8
|
|\ |
|
| |
| |
| |
| |
| | |
Bug: 17470065
Change-Id: Ib67aa1c50e3b6c24c4b12b0f31a996cc6874abd1
|
|/
|
|
|
|
|
|
|
| |
If there were less than a full frame worth of samples in the ring
buffer, then flush would loop forever trying to empty the ring
buffer.
Bug: 17646525
Change-Id: I68ec87352a91ce3a96d05e9b3f60a6e7975f9156
|
|
|
|
|
|
|
|
|
|
|
| |
Explicitly keep track of the number of samples in the ring buffer,
rather than inferring it from the difference between the read and
write pointer, since the latter cannot distinguish between a
completely full and a completely empty buffer.
Bug: 17582331
Change-Id: I24d16ce96710209b7457ffad7c4c60201451980f
|
|
|
|
|
|
|
|
| |
for backwards compatibility. Also fix a problem where writing 0
bytes to an empty ringbuffer would treat it as overflowed.
Bug: 17567501
Change-Id: I8019da76466576e1b2984e9eccf09bdec593038a
|
|
|
|
|
| |
Bug: 17544813
Change-Id: I6c94cd3056a99bb42fc71ee583938a2e3806fdc7
|
|
|
|
|
|
|
| |
Support multiple aac frames per input buffer also for non-ADTS streams,
now also works with 5.1 audio.
Bug: 16715379
|
|
|
|
|
|
|
|
| |
This reverts commit e086387c805311ac87904c3c4d6d4eb08d4b4ee2.
(broke multichannel movie playback)
Bug: 17454025
Change-Id: I95b82359f87f8beca66f35c34b7e125850a44747
|
|
|
|
|
|
|
| |
Support multiple aac frames per input buffer also for non-ADTS streams.
Bug: 16715379
Change-Id: I84f33d9bb65b6821b2a697f6750356cea98777c4
|
|
|
|
|
|
|
|
| |
CSD is always on the input port, so it does not matter which port
a buffer was last queued to.
Bug: 17216852
Change-Id: I6d00eb4bf3c2e9b9cc3c9918229de958bc465693
|
|
|
|
|
| |
Bug: 17118001
Change-Id: I45469b9aa6146edf3265b0f05f5ce3592b5daca6
|
|
|
|
|
| |
Bug: 17134697
Change-Id: Iaa8bdccbf1a8ccd3f898e48cfd22a0a443710fdd
|
|
|
|
|
|
|
|
|
| |
Control the dowmixing of the AAC decoder
Control the DRC processing of the AAC decoder
Bug 16740915
Change-Id: I9b06ea8785f56213ea120ed85eeb360c88223297
|
|
|
|
|
|
|
|
|
|
|
| |
When playing an mpeg-2 transport stream with AAC audio, there will be
more than one packet per input buffer, resulting in multiple output
buffers for each input buffers. Additional timestamps needs to be
generated in that case.
b/15755476
Change-Id: I2df50b0cb3690eb36d56871daa1263de0028db06
|
|
|
|
| |
Change-Id: I33c7424f1858e9f3f7f7ab5fb46db09c2714e02e
|
|
|
|
|
|
| |
Timestamps don't start at zero, but at least they're consistent.
Change-Id: I2ce9cd60655a4ece787802902b679fda01cb6182
|
|
|
|
|
|
|
|
|
|
|
| |
Add decoder limiter delay compensation to decoder wrapper
Includes a wrapper function for SoftAAC2.cpp which selects
DRC-related decoder parameters according to information in the bitstream
and desired DRC characteristics for different playback modes.
Bug 9428126
Change-Id: I5041b68760e95cf54073c3addf2b6026b9cfe8c5
|
|\
| |
| |
| | |
Change-Id: I7a6105f467a2fa3631c62be3d306845e9cf9b365
|
| |
| |
| |
| | |
Change-Id: I9c41bb34c35595ac19a7ab8faaacde8cb4fa9a15
|
| |
| |
| |
| |
| |
| | |
b/8543366
Change-Id: I746ffed6289486b802a2292bfc492ea6c780ed0b
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Conceptually it should be the same whether EOS is signalled on the last
buffer holding data, or an empty buffer that follows. Make it so that
this actually behaves the same for mp3, AAC and Vorbis.
b/8747869
Change-Id: Idece8ef45689a3ffaf70fb45d19862d7b93b2f92
|
| |
| |
| |
| |
| |
| | |
CTS DecoderTest.testBug11696552 started failing again, and this fixes that
Change-Id: I422c3b249a1d5eb25422573d2106d1f142ccfab6
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Rename decoder parameter for the maximum number of PCM output channels,
according to FDK AAC decoder interface, as defined in aacdecoder_lib.h
Bug 9428126
Change-Id: I2f0f6ca848bdbc8657d8dea589b03238245c0eaf
|
| |
| |
| |
| |
| |
| | |
b/8543366
Change-Id: I746ffed6289486b802a2292bfc492ea6c780ed0b
|
|/
|
|
|
|
|
|
|
|
| |
Conceptually it should be the same whether EOS is signalled on the last
buffer holding data, or an empty buffer that follows. Make it so that
this actually behaves the same for mp3, AAC and Vorbis.
b/8747869
Change-Id: Idece8ef45689a3ffaf70fb45d19862d7b93b2f92
|
|
|
|
|
|
|
| |
This reverts commit 9da36a6c8df70a5c8179ac78fab33cfbb5078cb2.
Bug: 10571297.
Change-Id: I76f20fe34872ea54fce626077462fb86c8c3f02e
|
|
|
|
|
|
|
|
|
|
| |
Conceptually it should be the same whether EOS is signalled on the last
buffer holding data, or an empty buffer that follows. Make it so that
this actually behaves the same for mp3, AAC and Vorbis.
b/8747869
Change-Id: Idece8ef45689a3ffaf70fb45d19862d7b93b2f92
|
|
|
|
|
|
|
| |
properly on a transition from idle->loaded.
Change-Id: I56ccfeef24c391e50e42b522194206e35c7ab700
related-to-bug: 9105408
|
|
|
|
|
|
| |
b/8614909
Change-Id: I9d973dcd74100b793791359c262b821207ff9ddd
|
|\ |
|
| |
| |
| |
| |
| |
| |
| |
| | |
By default, use boost and attenuation together.
bug 7140036
Change-Id: Ie0565f9e5cd24f7a59b5a218b4e6186661832601
|
|/
|
|
|
|
|
|
| |
The recent flush() changed made the codec behave differently after
a reconfigure. Now we reset its state properly again.
b/8543366
Change-Id: I8807b5ab02249b43fc1cf315d4e8d4ceb3f9b298
|
|
|
|
|
|
|
|
|
|
| |
There were two problems here. One was that the skip/cut buffer
wasn't cleared when it should be, and the second was that we
were always sending the first buffer of encoded data to the AAC
decoder twice.
b/8543366
Change-Id: Ic040edabf16cccd1f6ef8c9e5c9cfbacbdd8a089
|
|
|
|
| |
Change-Id: I3dd46cb4401493becbf6152f4dcd5a8f1e9a0b44
|
|
|
|
|
|
|
|
|
| |
Use -16dB as the default DRC reference level when decoding streams
with DRC metadata.
Bug 7370764
Change-Id: I900cee22f32384a5657fb041b69d42657bcddf09
|
|
|
|
|
|
|
|
|
|
|
|
| |
Configure the AAC decoder to use the modile DRC settings when
decoding streams with DRC metadata.
Settings are:
- target reference level of -12dB
- DRC compression factor (attenuation) at max
These settings can be overridden by platform properties if the
default values are to be modified.
Change-Id: If329fe566497685b735abe4e4b9a5e01697c063b
|