The release_input_buffer callback will be called when the library
is done consuming an "input buffer". The buffer passed into
EnqueueFrame must be kept valid until this callback is called. If
frame parallel is false, then this callback can be nullptr (in
this case the buffer has to be kept valid until the next call to
DequeueFrame). If frame parallel is true, this callback cannot be
nullptr.
PiperOrigin-RevId: 296276083
1. Have frame buffer callbacks return Libgav1StatusCode instead of int.
The 0 (success), -1 (failure) return value convention is less obvious.
Note: The callers of frame buffer callbacks,
BufferPool::OnFrameBufferSizeChanged() and YuvBuffer::Realloc(),
currently return bool, so more work is needed to propagate the frame
buffer callbacks' Libgav1StatusCode return value to the Decoder API.
2. Allow the FrameBufferSizeChangedCallback to be omitted if the frame
buffer size information is not useful to the application.
3. Remove the old (version 1) frame buffer callback API. Remove the
frame buffer callback adaptor.
frame_buffer2.h is renamed frame_buffer.h.
Libgav1FrameBuffer2 is renamed Libgav1FrameBuffer.
GetFrameBufferCallback2 and ReleaseFrameBufferCallback2 are renamed
GetFrameBufferCallback and ReleaseFrameBufferCallback.
PiperOrigin-RevId: 295971183
This avoids the issue of whether it is defined behaviour to cast an
arbitrary int (or even intptr_t) value to a void* pointer. This is the
original approach used before commit 0915998add5918214fa0282a69b50a159168a6d5.
PiperOrigin-RevId: 295552115
Libgav1 recently added the ComputeFrameBufferInfo() and SetFrameBuffer()
helper functions for writing frame buffer callbacks. Using them
simplifies the Libgav1GetFrameBuffer() function.
Also resurrect the AlignTo16() function.
PiperOrigin-RevId: 295548330
fixes:
gav1_jni.cc:446:25: error: cast from pointer to smaller type 'int' loses information
const int buffer_id = reinterpret_cast<int>(buffer_private_data);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
gav1_jni.cc:730:9: error: cast from pointer to smaller type 'int' loses information
reinterpret_cast<int>(decoder_buffer->buffer_private_data);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
after 0915998add
PiperOrigin-RevId: 295211245
The new code in Libgav1GetFrameBuffer is copied from
libgav1/src/frame_buffer_callback_adaptor.cc. It may become libgav1
utility functions available to libgav1 clients in the future.
The Libgav1FrameBuffer struct in the old frame buffer callback API is
defined as follows:
typedef struct Libgav1FrameBuffer {
uint8_t* data[3];
size_t size[3];
void* private_data;
} Libgav1FrameBuffer;
Copy these three fields to the JniFrameBuffer class as private data
members and add the RawBuffer() and Id() getter methods.
The existing AlignTo16 function is replaced by the copied Align template
function.
PiperOrigin-RevId: 293709205
This check is not needed because the FLAC specification does not
restrict the sample rate value and because the extension reads files
with other sample rates properly.
PiperOrigin-RevId: 292909391
+ force arm (over thumb) mode for 32-bit builds
-O2 improves performance ~30-40% over the default -Oz depending on the
resolution; this is similar to what is done for vp9 which uses -O3.
PiperOrigin-RevId: 290318121
- DownloadManagerHelper now passes all downloads to the
DownloadService when the service is attached (and once
the downloads are known). The service then starts the
foreground notification updater if necessary. This fixes
the ref'd issue.
- Don't call getScheduler() if the service is background
only. This was already documented to be the case on the
DownloadService constructor.
- If the service is started in the foreground on SDK level
26 and higher, satisfy the condition to move the service
to the foreground in onStartCommand rather than in stop().
It's much more obviously correct, and should produce the
same end result.
Issue: #6798
PiperOrigin-RevId: 290050024
This makes the format metadata null (instead of an empty Metadata
object) when it is not provided, and is therefore consistent with the
other extractors behavior.
PiperOrigin-RevId: 284171148
When transitioning to a new stream in a different format, the audio
processors are reconfigured. After this, they are drained and then
flushed so that they are ready to handle data in updated formats for the
new stream.
Before this change, some audio processors made the assumption that after
reconfiguration no more input would be queued in their old input format,
but this assumption is not correct: during draining more input may be
queued. Fix this behavior so that the new configuration is not referred
to while draining and only becomes active once flushed.
Issue: #6601
PiperOrigin-RevId: 282515359
This simplifies the contract of configure and is in preparation for
fixing a bug where more input can't be queued when draining audio
processors for a configuration change.
Issue: #6601
PiperOrigin-RevId: 282514367
Assuming that a flac stream starts with bytes
['f', 'L', 'a', 'C', 0, 0, 0, 0x22] is not always correct as it could
also start with ['f', 'L', 'a', 'C', 0x80, 0, 0, 0x22]
(see https://xiph.org/flac/format.html#metadata_block_streaminfo).
PiperOrigin-RevId: 279080562
