mirror of https://git.ffmpeg.org/ffmpeg.git
avcodec/videotoolbox: add AV1 hardware acceleration
Use AV1DecContext's current_obu to access the original OBUs, and feed them to videotoolbox, rather than the bare slice data passed via decode_slice. This requires a small addition to AV1DecContext, for keeping track of the current range of OBUs that belong to the current frame. Co-authored-by: Ruslan Chernenko <ractyfree@gmail.com> Co-authored-by: Martin Storsjö <martin@martin.st> Signed-off-by: Martin Storsjö <martin@martin.st>
This commit is contained in:
parent
7151081e33
commit
f9c5c5358c
|
@ -2467,6 +2467,7 @@ TYPES_LIST="
|
|||
kCMVideoCodecType_HEVC
|
||||
kCMVideoCodecType_HEVCWithAlpha
|
||||
kCMVideoCodecType_VP9
|
||||
kCMVideoCodecType_AV1
|
||||
kCVPixelFormatType_420YpCbCr10BiPlanarVideoRange
|
||||
kCVPixelFormatType_422YpCbCr8BiPlanarVideoRange
|
||||
kCVPixelFormatType_422YpCbCr10BiPlanarVideoRange
|
||||
|
@ -3174,6 +3175,8 @@ av1_vaapi_hwaccel_deps="vaapi VADecPictureParameterBufferAV1_bit_depth_idx"
|
|||
av1_vaapi_hwaccel_select="av1_decoder"
|
||||
av1_vdpau_hwaccel_deps="vdpau VdpPictureInfoAV1"
|
||||
av1_vdpau_hwaccel_select="av1_decoder"
|
||||
av1_videotoolbox_hwaccel_deps="videotoolbox"
|
||||
av1_videotoolbox_hwaccel_select="av1_decoder"
|
||||
av1_vulkan_hwaccel_deps="vulkan"
|
||||
av1_vulkan_hwaccel_select="av1_decoder"
|
||||
h263_vaapi_hwaccel_deps="vaapi"
|
||||
|
@ -6708,6 +6711,7 @@ enabled videotoolbox && {
|
|||
check_func_headers CoreMedia/CMFormatDescription.h kCMVideoCodecType_HEVC "-framework CoreMedia"
|
||||
check_func_headers CoreMedia/CMFormatDescription.h kCMVideoCodecType_HEVCWithAlpha "-framework CoreMedia"
|
||||
check_func_headers CoreMedia/CMFormatDescription.h kCMVideoCodecType_VP9 "-framework CoreMedia"
|
||||
check_func_headers CoreMedia/CMFormatDescription.h kCMVideoCodecType_AV1 "-framework CoreMedia"
|
||||
check_func_headers CoreVideo/CVPixelBuffer.h kCVPixelFormatType_420YpCbCr10BiPlanarVideoRange "-framework CoreVideo"
|
||||
check_func_headers CoreVideo/CVPixelBuffer.h kCVPixelFormatType_422YpCbCr8BiPlanarVideoRange "-framework CoreVideo"
|
||||
check_func_headers CoreVideo/CVPixelBuffer.h kCVPixelFormatType_422YpCbCr10BiPlanarVideoRange "-framework CoreVideo"
|
||||
|
|
|
@ -1008,6 +1008,7 @@ OBJS-$(CONFIG_AV1_D3D12VA_HWACCEL) += dxva2_av1.o d3d12va_av1.o
|
|||
OBJS-$(CONFIG_AV1_NVDEC_HWACCEL) += nvdec_av1.o
|
||||
OBJS-$(CONFIG_AV1_VAAPI_HWACCEL) += vaapi_av1.o
|
||||
OBJS-$(CONFIG_AV1_VDPAU_HWACCEL) += vdpau_av1.o
|
||||
OBJS-$(CONFIG_AV1_VIDEOTOOLBOX_HWACCEL) += videotoolbox_av1.o
|
||||
OBJS-$(CONFIG_AV1_VULKAN_HWACCEL) += vulkan_decode.o vulkan_av1.o
|
||||
OBJS-$(CONFIG_H263_VAAPI_HWACCEL) += vaapi_mpeg4.o
|
||||
OBJS-$(CONFIG_H263_VIDEOTOOLBOX_HWACCEL) += videotoolbox.o
|
||||
|
|
|
@ -541,6 +541,7 @@ static int get_pixel_format(AVCodecContext *avctx)
|
|||
CONFIG_AV1_NVDEC_HWACCEL + \
|
||||
CONFIG_AV1_VAAPI_HWACCEL + \
|
||||
CONFIG_AV1_VDPAU_HWACCEL + \
|
||||
CONFIG_AV1_VIDEOTOOLBOX_HWACCEL + \
|
||||
CONFIG_AV1_VULKAN_HWACCEL)
|
||||
enum AVPixelFormat pix_fmts[HWACCEL_MAX + 2], *fmtp = pix_fmts;
|
||||
|
||||
|
@ -568,6 +569,9 @@ static int get_pixel_format(AVCodecContext *avctx)
|
|||
#if CONFIG_AV1_VDPAU_HWACCEL
|
||||
*fmtp++ = AV_PIX_FMT_VDPAU;
|
||||
#endif
|
||||
#if CONFIG_AV1_VIDEOTOOLBOX_HWACCEL
|
||||
*fmtp++ = AV_PIX_FMT_VIDEOTOOLBOX;
|
||||
#endif
|
||||
#if CONFIG_AV1_VULKAN_HWACCEL
|
||||
*fmtp++ = AV_PIX_FMT_VULKAN;
|
||||
#endif
|
||||
|
@ -592,6 +596,9 @@ static int get_pixel_format(AVCodecContext *avctx)
|
|||
#if CONFIG_AV1_VDPAU_HWACCEL
|
||||
*fmtp++ = AV_PIX_FMT_VDPAU;
|
||||
#endif
|
||||
#if CONFIG_AV1_VIDEOTOOLBOX_HWACCEL
|
||||
*fmtp++ = AV_PIX_FMT_VIDEOTOOLBOX;
|
||||
#endif
|
||||
#if CONFIG_AV1_VULKAN_HWACCEL
|
||||
*fmtp++ = AV_PIX_FMT_VULKAN;
|
||||
#endif
|
||||
|
@ -1439,6 +1446,10 @@ static int av1_receive_frame_internal(AVCodecContext *avctx, AVFrame *frame)
|
|||
|
||||
if (raw_tile_group && (s->tile_num == raw_tile_group->tg_end + 1)) {
|
||||
int show_frame = s->raw_frame_header->show_frame;
|
||||
// Set nb_unit to point at the next OBU, to indicate which
|
||||
// OBUs have been processed for this current frame. (If this
|
||||
// frame gets output, we set nb_unit to this value later too.)
|
||||
s->nb_unit = i + 1;
|
||||
if (avctx->hwaccel && s->cur_frame.f) {
|
||||
ret = FF_HW_SIMPLE_CALL(avctx, end_frame);
|
||||
if (ret < 0) {
|
||||
|
@ -1449,6 +1460,8 @@ static int av1_receive_frame_internal(AVCodecContext *avctx, AVFrame *frame)
|
|||
|
||||
update_reference_list(avctx);
|
||||
|
||||
// Set start_unit to indicate the first OBU of the next frame.
|
||||
s->start_unit = s->nb_unit;
|
||||
raw_tile_group = NULL;
|
||||
s->raw_frame_header = NULL;
|
||||
|
||||
|
@ -1478,7 +1491,7 @@ end:
|
|||
s->raw_frame_header = NULL;
|
||||
av_packet_unref(s->pkt);
|
||||
ff_cbs_fragment_reset(&s->current_obu);
|
||||
s->nb_unit = 0;
|
||||
s->nb_unit = s->start_unit = 0;
|
||||
}
|
||||
if (!ret && !frame->buf[0])
|
||||
ret = AVERROR(EAGAIN);
|
||||
|
@ -1505,7 +1518,7 @@ static int av1_receive_frame(AVCodecContext *avctx, AVFrame *frame)
|
|||
return ret;
|
||||
}
|
||||
|
||||
s->nb_unit = 0;
|
||||
s->nb_unit = s->start_unit = 0;
|
||||
av_log(avctx, AV_LOG_DEBUG, "Total OBUs on this packet: %d.\n",
|
||||
s->current_obu.nb_units);
|
||||
}
|
||||
|
@ -1526,7 +1539,7 @@ static void av1_decode_flush(AVCodecContext *avctx)
|
|||
|
||||
av1_frame_unref(&s->cur_frame);
|
||||
s->operating_point_idc = 0;
|
||||
s->nb_unit = 0;
|
||||
s->nb_unit = s->start_unit = 0;
|
||||
s->raw_frame_header = NULL;
|
||||
s->raw_seq = NULL;
|
||||
s->cll = NULL;
|
||||
|
@ -1594,6 +1607,9 @@ const FFCodec ff_av1_decoder = {
|
|||
#if CONFIG_AV1_VDPAU_HWACCEL
|
||||
HWACCEL_VDPAU(av1),
|
||||
#endif
|
||||
#if CONFIG_AV1_VIDEOTOOLBOX_HWACCEL
|
||||
HWACCEL_VIDEOTOOLBOX(av1),
|
||||
#endif
|
||||
#if CONFIG_AV1_VULKAN_HWACCEL
|
||||
HWACCEL_VULKAN(av1),
|
||||
#endif
|
||||
|
|
|
@ -114,7 +114,8 @@ typedef struct AV1DecContext {
|
|||
AV1Frame ref[AV1_NUM_REF_FRAMES];
|
||||
AV1Frame cur_frame;
|
||||
|
||||
int nb_unit;
|
||||
int nb_unit; ///< The index of the next OBU to be processed.
|
||||
int start_unit; ///< The index of the first OBU of the current frame.
|
||||
|
||||
// AVOptions
|
||||
int operating_point;
|
||||
|
|
|
@ -26,6 +26,7 @@ extern const struct FFHWAccel ff_av1_dxva2_hwaccel;
|
|||
extern const struct FFHWAccel ff_av1_nvdec_hwaccel;
|
||||
extern const struct FFHWAccel ff_av1_vaapi_hwaccel;
|
||||
extern const struct FFHWAccel ff_av1_vdpau_hwaccel;
|
||||
extern const struct FFHWAccel ff_av1_videotoolbox_hwaccel;
|
||||
extern const struct FFHWAccel ff_av1_vulkan_hwaccel;
|
||||
extern const struct FFHWAccel ff_h263_vaapi_hwaccel;
|
||||
extern const struct FFHWAccel ff_h263_videotoolbox_hwaccel;
|
||||
|
|
|
@ -29,7 +29,7 @@
|
|||
|
||||
#include "version_major.h"
|
||||
|
||||
#define LIBAVCODEC_VERSION_MINOR 20
|
||||
#define LIBAVCODEC_VERSION_MINOR 21
|
||||
#define LIBAVCODEC_VERSION_MICRO 100
|
||||
|
||||
#define LIBAVCODEC_VERSION_INT AV_VERSION_INT(LIBAVCODEC_VERSION_MAJOR, \
|
||||
|
|
|
@ -56,6 +56,10 @@ enum { kCMVideoCodecType_HEVC = 'hvc1' };
|
|||
enum { kCMVideoCodecType_VP9 = 'vp09' };
|
||||
#endif
|
||||
|
||||
#if !HAVE_KCMVIDEOCODECTYPE_AV1
|
||||
enum { kCMVideoCodecType_AV1 = 'av01' };
|
||||
#endif
|
||||
|
||||
#define VIDEOTOOLBOX_ESDS_EXTRADATA_PADDING 12
|
||||
|
||||
typedef struct VTHWFrame {
|
||||
|
@ -92,6 +96,26 @@ int ff_videotoolbox_buffer_copy(VTContext *vtctx,
|
|||
return 0;
|
||||
}
|
||||
|
||||
int ff_videotoolbox_buffer_append(VTContext *vtctx,
|
||||
const uint8_t *buffer,
|
||||
uint32_t size)
|
||||
{
|
||||
void *tmp;
|
||||
|
||||
tmp = av_fast_realloc(vtctx->bitstream,
|
||||
&vtctx->allocated_size,
|
||||
vtctx->bitstream_size + size);
|
||||
|
||||
if (!tmp)
|
||||
return AVERROR(ENOMEM);
|
||||
|
||||
vtctx->bitstream = tmp;
|
||||
memcpy(vtctx->bitstream + vtctx->bitstream_size, buffer, size);
|
||||
vtctx->bitstream_size += size;
|
||||
|
||||
return 0;
|
||||
}
|
||||
|
||||
static int videotoolbox_postproc_frame(void *avctx, AVFrame *frame)
|
||||
{
|
||||
int ret;
|
||||
|
@ -846,6 +870,13 @@ static CFDictionaryRef videotoolbox_decoder_config_create(CMVideoCodecType codec
|
|||
if (data)
|
||||
CFDictionarySetValue(avc_info, CFSTR("vpcC"), data);
|
||||
break;
|
||||
#endif
|
||||
#if CONFIG_AV1_VIDEOTOOLBOX_HWACCEL
|
||||
case kCMVideoCodecType_AV1 :
|
||||
data = ff_videotoolbox_av1c_extradata_create(avctx);
|
||||
if (data)
|
||||
CFDictionarySetValue(avc_info, CFSTR("av1C"), data);
|
||||
break;
|
||||
#endif
|
||||
default:
|
||||
break;
|
||||
|
@ -912,6 +943,9 @@ static int videotoolbox_start(AVCodecContext *avctx)
|
|||
case AV_CODEC_ID_VP9 :
|
||||
videotoolbox->cm_codec_type = kCMVideoCodecType_VP9;
|
||||
break;
|
||||
case AV_CODEC_ID_AV1 :
|
||||
videotoolbox->cm_codec_type = kCMVideoCodecType_AV1;
|
||||
break;
|
||||
default :
|
||||
break;
|
||||
}
|
||||
|
|
|
@ -0,0 +1,105 @@
|
|||
/*
|
||||
* Videotoolbox hardware acceleration for AV1
|
||||
* Copyright (c) 2023 Jan Ekström
|
||||
* Copyright (c) 2024 Ruslan Chernenko
|
||||
* Copyright (c) 2024 Martin Storsjö
|
||||
*
|
||||
* This file is part of FFmpeg.
|
||||
*
|
||||
* FFmpeg is free software; you can redistribute it and/or
|
||||
* modify it under the terms of the GNU Lesser General Public
|
||||
* License as published by the Free Software Foundation; either
|
||||
* version 2.1 of the License, or (at your option) any later version.
|
||||
*
|
||||
* FFmpeg is distributed in the hope that it will be useful,
|
||||
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
||||
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
||||
* Lesser General Public License for more details.
|
||||
*
|
||||
* You should have received a copy of the GNU Lesser General Public
|
||||
* License along with FFmpeg; if not, write to the Free Software
|
||||
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
|
||||
*/
|
||||
|
||||
#include "libavutil/mem.h"
|
||||
|
||||
#include "av1dec.h"
|
||||
#include "hwaccel_internal.h"
|
||||
#include "internal.h"
|
||||
#include "vt_internal.h"
|
||||
|
||||
CFDataRef ff_videotoolbox_av1c_extradata_create(AVCodecContext *avctx)
|
||||
{
|
||||
AV1DecContext *s = avctx->priv_data;
|
||||
uint8_t *buf;
|
||||
CFDataRef data;
|
||||
if (!s->raw_seq)
|
||||
return NULL;
|
||||
|
||||
buf = av_malloc(s->seq_data_ref->size + 4);
|
||||
if (!buf)
|
||||
return NULL;
|
||||
buf[0] = 0x81; // version and marker (constant)
|
||||
buf[1] = s->raw_seq->seq_profile << 5 | s->raw_seq->seq_level_idx[0];
|
||||
buf[2] = s->raw_seq->seq_tier[0] << 7 |
|
||||
s->raw_seq->color_config.high_bitdepth << 6 |
|
||||
s->raw_seq->color_config.twelve_bit << 5 |
|
||||
s->raw_seq->color_config.mono_chrome << 4 |
|
||||
s->raw_seq->color_config.subsampling_x << 3 |
|
||||
s->raw_seq->color_config.subsampling_y << 2 |
|
||||
s->raw_seq->color_config.chroma_sample_position;
|
||||
|
||||
if (s->raw_seq->initial_display_delay_present_flag)
|
||||
buf[3] = 0 << 5 |
|
||||
s->raw_seq->initial_display_delay_present_flag << 4 |
|
||||
s->raw_seq->initial_display_delay_minus_1[0];
|
||||
else
|
||||
buf[3] = 0x00;
|
||||
memcpy(buf + 4, s->seq_data_ref->data, s->seq_data_ref->size);
|
||||
data = CFDataCreate(kCFAllocatorDefault, buf, s->seq_data_ref->size + 4);
|
||||
av_free(buf);
|
||||
return data;
|
||||
};
|
||||
|
||||
|
||||
static int videotoolbox_av1_start_frame(AVCodecContext *avctx,
|
||||
const uint8_t *buffer,
|
||||
uint32_t size)
|
||||
{
|
||||
return 0;
|
||||
}
|
||||
|
||||
static int videotoolbox_av1_decode_slice(AVCodecContext *avctx,
|
||||
const uint8_t *buffer,
|
||||
uint32_t size)
|
||||
{
|
||||
return 0;
|
||||
}
|
||||
|
||||
static int videotoolbox_av1_end_frame(AVCodecContext *avctx)
|
||||
{
|
||||
const AV1DecContext *s = avctx->priv_data;
|
||||
VTContext *vtctx = avctx->internal->hwaccel_priv_data;
|
||||
AVFrame *frame = s->cur_frame.f;
|
||||
|
||||
vtctx->bitstream_size = 0;
|
||||
for (int i = s->start_unit; i < s->nb_unit; i++)
|
||||
ff_videotoolbox_buffer_append(vtctx, s->current_obu.units[i].data,
|
||||
s->current_obu.units[i].data_size);
|
||||
return ff_videotoolbox_common_end_frame(avctx, frame);
|
||||
}
|
||||
|
||||
const FFHWAccel ff_av1_videotoolbox_hwaccel = {
|
||||
.p.name = "av1_videotoolbox",
|
||||
.p.type = AVMEDIA_TYPE_VIDEO,
|
||||
.p.id = AV_CODEC_ID_AV1,
|
||||
.p.pix_fmt = AV_PIX_FMT_VIDEOTOOLBOX,
|
||||
.alloc_frame = ff_videotoolbox_alloc_frame,
|
||||
.start_frame = videotoolbox_av1_start_frame,
|
||||
.decode_slice = videotoolbox_av1_decode_slice,
|
||||
.end_frame = videotoolbox_av1_end_frame,
|
||||
.frame_params = ff_videotoolbox_frame_params,
|
||||
.init = ff_videotoolbox_common_init,
|
||||
.uninit = ff_videotoolbox_uninit,
|
||||
.priv_data_size = sizeof(VTContext),
|
||||
};
|
|
@ -56,6 +56,9 @@ int ff_videotoolbox_frame_params(AVCodecContext *avctx,
|
|||
int ff_videotoolbox_buffer_copy(VTContext *vtctx,
|
||||
const uint8_t *buffer,
|
||||
uint32_t size);
|
||||
int ff_videotoolbox_buffer_append(VTContext *vtctx,
|
||||
const uint8_t *buffer,
|
||||
uint32_t size);
|
||||
int ff_videotoolbox_uninit(AVCodecContext *avctx);
|
||||
int ff_videotoolbox_h264_start_frame(AVCodecContext *avctx,
|
||||
const uint8_t *buffer,
|
||||
|
@ -64,6 +67,7 @@ int ff_videotoolbox_h264_decode_slice(AVCodecContext *avctx,
|
|||
const uint8_t *buffer,
|
||||
uint32_t size);
|
||||
int ff_videotoolbox_common_end_frame(AVCodecContext *avctx, AVFrame *frame);
|
||||
CFDataRef ff_videotoolbox_av1c_extradata_create(AVCodecContext *avctx);
|
||||
CFDataRef ff_videotoolbox_avcc_extradata_create(AVCodecContext *avctx);
|
||||
CFDataRef ff_videotoolbox_hvcc_extradata_create(AVCodecContext *avctx);
|
||||
CFDataRef ff_videotoolbox_vpcc_extradata_create(AVCodecContext *avctx);
|
||||
|
|
Loading…
Reference in New Issue