vpx/vp9/decoder/vp9_decoder.h

149 lines
4.6 KiB
C
Raw Normal View History

2010-05-18 17:58:33 +02:00
/*
* Copyright (c) 2010 The WebM project authors. All Rights Reserved.
2010-05-18 17:58:33 +02:00
*
* Use of this source code is governed by a BSD-style license
* that can be found in the LICENSE file in the root of the source
* tree. An additional intellectual property rights grant can be found
* in the file PATENTS. All contributing project authors may
* be found in the AUTHORS file in the root of the source tree.
2010-05-18 17:58:33 +02:00
*/
#ifndef VP9_DECODER_VP9_DECODER_H_
#define VP9_DECODER_VP9_DECODER_H_
2010-05-18 17:58:33 +02:00
#include "./vpx_config.h"
#include "vpx/vpx_codec.h"
#include "vpx_dsp/bitreader.h"
2010-05-18 17:58:33 +02:00
#include "vpx_scale/yv12config.h"
#include "vpx_util/vpx_thread.h"
#include "vp9/common/vp9_thread_common.h"
#include "vp9/common/vp9_onyxc_int.h"
#include "vp9/common/vp9_ppflags.h"
Try again to merge branch 'frame-parallel' into master branch. In frame parallel decode, libvpx decoder decodes several frames on all cpus in parallel fashion. If not being flushed, it will only return frame when all the cpus are busy. If getting flushed, it will return all the frames in the decoder. Compare with current serial decode mode in which libvpx decoder is idle between decode calls, libvpx decoder is busy between decode calls. Current frame parallel decode will only speed up the decoding for frame parallel encoded videos. For non frame parallel encoded videos, frame parallel decode is slower than serial decode due to lack of loopfilter worker thread. There are still some known issues that need to be addressed. For example: decode frame parallel videos with segmentation enabled is not right sometimes. * frame-parallel: Add error handling for frame parallel decode and unit test for that. Fix a bug in frame parallel decode and add a unit test for that. Add two test vectors to test frame parallel decode. Add key frame seeking to webmdec and webm_video_source. Implement frame parallel decode for VP9. Increase the thread test range to cover 5, 6, 7, 8 threads. Fix a bug in adding frame parallel unit test. Add VP9 frame-parallel unit test. Manually pick "Make the api behavior conform to api spec." from master branch. Move vp9_dec_build_inter_predictors_* to decoder folder. Add segmentation map array for current and last frame segmentation. Include the right header for VP9 worker thread. Move vp9_thread.* to common. ctrl_get_reference does not need user_priv. Seperate the frame buffers from VP9 encoder/decoder structure. Revert "Revert "Revert "Revert 3 patches from Hangyu to get Chrome to build:""" Conflicts: test/codec_factory.h test/decode_test_driver.cc test/decode_test_driver.h test/invalid_file_test.cc test/test-data.sha1 test/test.mk test/test_vectors.cc vp8/vp8_dx_iface.c vp9/common/vp9_alloccommon.c vp9/common/vp9_entropymode.c vp9/common/vp9_loopfilter_thread.c vp9/common/vp9_loopfilter_thread.h vp9/common/vp9_mvref_common.c vp9/common/vp9_onyxc_int.h vp9/common/vp9_reconinter.c vp9/decoder/vp9_decodeframe.c vp9/decoder/vp9_decodeframe.h vp9/decoder/vp9_decodemv.c vp9/decoder/vp9_decoder.c vp9/decoder/vp9_decoder.h vp9/encoder/vp9_encoder.c vp9/encoder/vp9_pickmode.c vp9/encoder/vp9_rdopt.c vp9/vp9_cx_iface.c vp9/vp9_dx_iface.c This reverts commit a18da9760a74d9ce6fb9f875706dc639c95402f5. Change-Id: I361442ffec1586d036ea2e0ee97ce4f077585f02
2015-01-27 21:26:28 +01:00
#include "vp9/decoder/vp9_dthread.h"
2010-05-18 17:58:33 +02:00
#ifdef __cplusplus
extern "C" {
#endif
// TODO(hkuang): combine this with TileWorkerData.
typedef struct TileData {
VP9_COMMON *cm;
vpx_reader bit_reader;
DECLARE_ALIGNED(16, MACROBLOCKD, xd);
/* dqcoeff are shared by all the planes. So planes must be decoded serially */
DECLARE_ALIGNED(16, tran_low_t, dqcoeff[32 * 32]);
} TileData;
typedef struct TileBuffer {
const uint8_t *data;
size_t size;
int col; // only used with multi-threaded decoding
} TileBuffer;
typedef struct TileWorkerData {
const uint8_t *data_end;
int buf_start, buf_end; // pbi->tile_buffers to decode, inclusive
vpx_reader bit_reader;
FRAME_COUNTS counts;
DECLARE_ALIGNED(16, MACROBLOCKD, xd);
/* dqcoeff are shared by all the planes. So planes must be decoded serially */
DECLARE_ALIGNED(16, tran_low_t, dqcoeff[32 * 32]);
struct vpx_internal_error_info error_info;
} TileWorkerData;
typedef struct VP9Decoder {
DECLARE_ALIGNED(16, MACROBLOCKD, mb);
DECLARE_ALIGNED(16, VP9_COMMON, common);
int ready_for_new_data;
int refresh_frame_flags;
int frame_parallel_decode; // frame-based threading.
Try again to merge branch 'frame-parallel' into master branch. In frame parallel decode, libvpx decoder decodes several frames on all cpus in parallel fashion. If not being flushed, it will only return frame when all the cpus are busy. If getting flushed, it will return all the frames in the decoder. Compare with current serial decode mode in which libvpx decoder is idle between decode calls, libvpx decoder is busy between decode calls. Current frame parallel decode will only speed up the decoding for frame parallel encoded videos. For non frame parallel encoded videos, frame parallel decode is slower than serial decode due to lack of loopfilter worker thread. There are still some known issues that need to be addressed. For example: decode frame parallel videos with segmentation enabled is not right sometimes. * frame-parallel: Add error handling for frame parallel decode and unit test for that. Fix a bug in frame parallel decode and add a unit test for that. Add two test vectors to test frame parallel decode. Add key frame seeking to webmdec and webm_video_source. Implement frame parallel decode for VP9. Increase the thread test range to cover 5, 6, 7, 8 threads. Fix a bug in adding frame parallel unit test. Add VP9 frame-parallel unit test. Manually pick "Make the api behavior conform to api spec." from master branch. Move vp9_dec_build_inter_predictors_* to decoder folder. Add segmentation map array for current and last frame segmentation. Include the right header for VP9 worker thread. Move vp9_thread.* to common. ctrl_get_reference does not need user_priv. Seperate the frame buffers from VP9 encoder/decoder structure. Revert "Revert "Revert "Revert 3 patches from Hangyu to get Chrome to build:""" Conflicts: test/codec_factory.h test/decode_test_driver.cc test/decode_test_driver.h test/invalid_file_test.cc test/test-data.sha1 test/test.mk test/test_vectors.cc vp8/vp8_dx_iface.c vp9/common/vp9_alloccommon.c vp9/common/vp9_entropymode.c vp9/common/vp9_loopfilter_thread.c vp9/common/vp9_loopfilter_thread.h vp9/common/vp9_mvref_common.c vp9/common/vp9_onyxc_int.h vp9/common/vp9_reconinter.c vp9/decoder/vp9_decodeframe.c vp9/decoder/vp9_decodeframe.h vp9/decoder/vp9_decodemv.c vp9/decoder/vp9_decoder.c vp9/decoder/vp9_decoder.h vp9/encoder/vp9_encoder.c vp9/encoder/vp9_pickmode.c vp9/encoder/vp9_rdopt.c vp9/vp9_cx_iface.c vp9/vp9_dx_iface.c This reverts commit a18da9760a74d9ce6fb9f875706dc639c95402f5. Change-Id: I361442ffec1586d036ea2e0ee97ce4f077585f02
2015-01-27 21:26:28 +01:00
// TODO(hkuang): Combine this with cur_buf in macroblockd as they are
// the same.
RefCntBuffer *cur_buf; // Current decoding frame buffer.
VPxWorker *frame_worker_owner; // frame_worker that owns this pbi.
VPxWorker lf_worker;
VPxWorker *tile_workers;
TileWorkerData *tile_worker_data;
TileBuffer tile_buffers[64];
int num_tile_workers;
TileData *tile_data;
int total_tiles;
VP9LfSync lf_row_sync;
vpx_decrypt_cb decrypt_cb;
void *decrypt_state;
int max_threads;
int inv_tile_order;
Try again to merge branch 'frame-parallel' into master branch. In frame parallel decode, libvpx decoder decodes several frames on all cpus in parallel fashion. If not being flushed, it will only return frame when all the cpus are busy. If getting flushed, it will return all the frames in the decoder. Compare with current serial decode mode in which libvpx decoder is idle between decode calls, libvpx decoder is busy between decode calls. Current frame parallel decode will only speed up the decoding for frame parallel encoded videos. For non frame parallel encoded videos, frame parallel decode is slower than serial decode due to lack of loopfilter worker thread. There are still some known issues that need to be addressed. For example: decode frame parallel videos with segmentation enabled is not right sometimes. * frame-parallel: Add error handling for frame parallel decode and unit test for that. Fix a bug in frame parallel decode and add a unit test for that. Add two test vectors to test frame parallel decode. Add key frame seeking to webmdec and webm_video_source. Implement frame parallel decode for VP9. Increase the thread test range to cover 5, 6, 7, 8 threads. Fix a bug in adding frame parallel unit test. Add VP9 frame-parallel unit test. Manually pick "Make the api behavior conform to api spec." from master branch. Move vp9_dec_build_inter_predictors_* to decoder folder. Add segmentation map array for current and last frame segmentation. Include the right header for VP9 worker thread. Move vp9_thread.* to common. ctrl_get_reference does not need user_priv. Seperate the frame buffers from VP9 encoder/decoder structure. Revert "Revert "Revert "Revert 3 patches from Hangyu to get Chrome to build:""" Conflicts: test/codec_factory.h test/decode_test_driver.cc test/decode_test_driver.h test/invalid_file_test.cc test/test-data.sha1 test/test.mk test/test_vectors.cc vp8/vp8_dx_iface.c vp9/common/vp9_alloccommon.c vp9/common/vp9_entropymode.c vp9/common/vp9_loopfilter_thread.c vp9/common/vp9_loopfilter_thread.h vp9/common/vp9_mvref_common.c vp9/common/vp9_onyxc_int.h vp9/common/vp9_reconinter.c vp9/decoder/vp9_decodeframe.c vp9/decoder/vp9_decodeframe.h vp9/decoder/vp9_decodemv.c vp9/decoder/vp9_decoder.c vp9/decoder/vp9_decoder.h vp9/encoder/vp9_encoder.c vp9/encoder/vp9_pickmode.c vp9/encoder/vp9_rdopt.c vp9/vp9_cx_iface.c vp9/vp9_dx_iface.c This reverts commit a18da9760a74d9ce6fb9f875706dc639c95402f5. Change-Id: I361442ffec1586d036ea2e0ee97ce4f077585f02
2015-01-27 21:26:28 +01:00
int need_resync; // wait for key/intra-only frame.
int hold_ref_buf; // hold the reference buffer.
} VP9Decoder;
int vp9_receive_compressed_data(struct VP9Decoder *pbi,
size_t size, const uint8_t **dest);
int vp9_get_raw_frame(struct VP9Decoder *pbi, YV12_BUFFER_CONFIG *sd,
vp9_ppflags_t *flags);
vpx_codec_err_t vp9_copy_reference_dec(struct VP9Decoder *pbi,
VP9_REFFRAME ref_frame_flag,
YV12_BUFFER_CONFIG *sd);
vpx_codec_err_t vp9_set_reference_dec(VP9_COMMON *cm,
VP9_REFFRAME ref_frame_flag,
YV12_BUFFER_CONFIG *sd);
static INLINE uint8_t read_marker(vpx_decrypt_cb decrypt_cb,
void *decrypt_state,
const uint8_t *data) {
if (decrypt_cb) {
uint8_t marker;
decrypt_cb(decrypt_state, data, &marker, 1);
return marker;
}
return *data;
}
// This function is exposed for use in tests, as well as the inlined function
// "read_marker".
vpx_codec_err_t vp9_parse_superframe_index(const uint8_t *data,
size_t data_sz,
uint32_t sizes[8], int *count,
vpx_decrypt_cb decrypt_cb,
void *decrypt_state);
Try again to merge branch 'frame-parallel' into master branch. In frame parallel decode, libvpx decoder decodes several frames on all cpus in parallel fashion. If not being flushed, it will only return frame when all the cpus are busy. If getting flushed, it will return all the frames in the decoder. Compare with current serial decode mode in which libvpx decoder is idle between decode calls, libvpx decoder is busy between decode calls. Current frame parallel decode will only speed up the decoding for frame parallel encoded videos. For non frame parallel encoded videos, frame parallel decode is slower than serial decode due to lack of loopfilter worker thread. There are still some known issues that need to be addressed. For example: decode frame parallel videos with segmentation enabled is not right sometimes. * frame-parallel: Add error handling for frame parallel decode and unit test for that. Fix a bug in frame parallel decode and add a unit test for that. Add two test vectors to test frame parallel decode. Add key frame seeking to webmdec and webm_video_source. Implement frame parallel decode for VP9. Increase the thread test range to cover 5, 6, 7, 8 threads. Fix a bug in adding frame parallel unit test. Add VP9 frame-parallel unit test. Manually pick "Make the api behavior conform to api spec." from master branch. Move vp9_dec_build_inter_predictors_* to decoder folder. Add segmentation map array for current and last frame segmentation. Include the right header for VP9 worker thread. Move vp9_thread.* to common. ctrl_get_reference does not need user_priv. Seperate the frame buffers from VP9 encoder/decoder structure. Revert "Revert "Revert "Revert 3 patches from Hangyu to get Chrome to build:""" Conflicts: test/codec_factory.h test/decode_test_driver.cc test/decode_test_driver.h test/invalid_file_test.cc test/test-data.sha1 test/test.mk test/test_vectors.cc vp8/vp8_dx_iface.c vp9/common/vp9_alloccommon.c vp9/common/vp9_entropymode.c vp9/common/vp9_loopfilter_thread.c vp9/common/vp9_loopfilter_thread.h vp9/common/vp9_mvref_common.c vp9/common/vp9_onyxc_int.h vp9/common/vp9_reconinter.c vp9/decoder/vp9_decodeframe.c vp9/decoder/vp9_decodeframe.h vp9/decoder/vp9_decodemv.c vp9/decoder/vp9_decoder.c vp9/decoder/vp9_decoder.h vp9/encoder/vp9_encoder.c vp9/encoder/vp9_pickmode.c vp9/encoder/vp9_rdopt.c vp9/vp9_cx_iface.c vp9/vp9_dx_iface.c This reverts commit a18da9760a74d9ce6fb9f875706dc639c95402f5. Change-Id: I361442ffec1586d036ea2e0ee97ce4f077585f02
2015-01-27 21:26:28 +01:00
struct VP9Decoder *vp9_decoder_create(BufferPool *const pool);
void vp9_decoder_remove(struct VP9Decoder *pbi);
static INLINE void decrease_ref_count(int idx, RefCntBuffer *const frame_bufs,
BufferPool *const pool) {
if (idx >= 0) {
--frame_bufs[idx].ref_count;
// A worker may only get a free framebuffer index when calling get_free_fb.
// But the private buffer is not set up until finish decoding header.
// So any error happens during decoding header, the frame_bufs will not
// have valid priv buffer.
if (frame_bufs[idx].ref_count == 0 &&
frame_bufs[idx].raw_frame_buffer.priv) {
pool->release_fb_cb(pool->cb_priv, &frame_bufs[idx].raw_frame_buffer);
}
}
}
2010-05-18 17:58:33 +02:00
#ifdef __cplusplus
} // extern "C"
2010-05-18 17:58:33 +02:00
#endif
#endif // VP9_DECODER_VP9_DECODER_H_