[FFmpeg-devel] [PATCH] avcodec/videotoolboxenc: add hevc_videotoolbox encoder
Aman Gupta
ffmpeg at tmm1.net
Sat Nov 11 05:57:33 EET 2017
From: Aman Gupta <aman at tmm1.net>
---
configure | 2 +
libavcodec/allcodecs.c | 1 +
libavcodec/videotoolboxenc.c | 296 ++++++++++++++++++++++++++++++-------------
3 files changed, 214 insertions(+), 85 deletions(-)
diff --git a/configure b/configure
index 2cf18ecc12..39b9d4cb0c 100755
--- a/configure
+++ b/configure
@@ -2928,6 +2928,8 @@ pcm_mulaw_at_encoder_select="audio_frame_queue"
chromaprint_muxer_deps="chromaprint"
h264_videotoolbox_encoder_deps="pthreads"
h264_videotoolbox_encoder_select="videotoolbox_encoder"
+hevc_videotoolbox_encoder_deps="pthreads"
+hevc_videotoolbox_encoder_select="videotoolbox_encoder"
libcelt_decoder_deps="libcelt"
libfdk_aac_decoder_deps="libfdk_aac"
libfdk_aac_encoder_deps="libfdk_aac"
diff --git a/libavcodec/allcodecs.c b/libavcodec/allcodecs.c
index c817003693..d8be53a52a 100644
--- a/libavcodec/allcodecs.c
+++ b/libavcodec/allcodecs.c
@@ -670,6 +670,7 @@ static void register_all(void)
REGISTER_ENCODER(HEVC_QSV, hevc_qsv);
REGISTER_ENCODER(HEVC_V4L2M2M, hevc_v4l2m2m);
REGISTER_ENCODER(HEVC_VAAPI, hevc_vaapi);
+ REGISTER_ENCODER(HEVC_VIDEOTOOLBOX, hevc_videotoolbox);
REGISTER_ENCODER(LIBKVAZAAR, libkvazaar);
REGISTER_DECODER(MJPEG_CUVID, mjpeg_cuvid);
REGISTER_ENCODER(MJPEG_QSV, mjpeg_qsv);
diff --git a/libavcodec/videotoolboxenc.c b/libavcodec/videotoolboxenc.c
index eba6cc672f..9725964e87 100644
--- a/libavcodec/videotoolboxenc.c
+++ b/libavcodec/videotoolboxenc.c
@@ -35,6 +35,17 @@
#include "h264_sei.h"
#include <dlfcn.h>
+#if !HAVE_KCMVIDEOCODECTYPE_HEVC
+enum { kCMVideoCodecType_HEVC = 'hvc1' };
+#endif
+
+typedef OSStatus (*getParameterSetAtIndex)(CMFormatDescriptionRef videoDesc,
+ size_t parameterSetIndex,
+ const uint8_t * _Nullable *parameterSetPointerOut,
+ size_t *parameterSetSizeOut,
+ size_t *parameterSetCountOut,
+ int *NALUnitHeaderLengthOut);
+
//These symbols may not be present
static struct{
CFStringRef kCVImageBufferColorPrimaries_ITU_R_2020;
@@ -65,10 +76,15 @@ static struct{
CFStringRef kVTProfileLevel_H264_High_5_2;
CFStringRef kVTProfileLevel_H264_High_AutoLevel;
+ CFStringRef kVTProfileLevel_HEVC_Main_AutoLevel;
+ CFStringRef kVTProfileLevel_HEVC_Main10_AutoLevel;
+
CFStringRef kVTCompressionPropertyKey_RealTime;
CFStringRef kVTVideoEncoderSpecification_EnableHardwareAcceleratedVideoEncoder;
CFStringRef kVTVideoEncoderSpecification_RequireHardwareAcceleratedVideoEncoder;
+
+ getParameterSetAtIndex CMVideoFormatDescriptionGetHEVCParameterSetAtIndex;
} compat_keys;
#define GET_SYM(symbol, defaultVal) \
@@ -83,6 +99,12 @@ do{ \
static pthread_once_t once_ctrl = PTHREAD_ONCE_INIT;
static void loadVTEncSymbols(){
+ compat_keys.CMVideoFormatDescriptionGetHEVCParameterSetAtIndex =
+ (getParameterSetAtIndex)dlsym(
+ RTLD_DEFAULT,
+ "CMVideoFormatDescriptionGetHEVCParameterSetAtIndex"
+ );
+
GET_SYM(kCVImageBufferColorPrimaries_ITU_R_2020, "ITU_R_2020");
GET_SYM(kCVImageBufferTransferFunction_ITU_R_2020, "ITU_R_2020");
GET_SYM(kCVImageBufferYCbCrMatrix_ITU_R_2020, "ITU_R_2020");
@@ -111,6 +133,9 @@ static void loadVTEncSymbols(){
GET_SYM(kVTProfileLevel_H264_High_5_2, "H264_High_5_2");
GET_SYM(kVTProfileLevel_H264_High_AutoLevel, "H264_High_AutoLevel");
+ GET_SYM(kVTProfileLevel_HEVC_Main_AutoLevel, "HEVC_Main_AutoLevel");
+ GET_SYM(kVTProfileLevel_HEVC_Main10_AutoLevel, "HEVC_Main10_AutoLevel");
+
GET_SYM(kVTCompressionPropertyKey_RealTime, "RealTime");
GET_SYM(kVTVideoEncoderSpecification_EnableHardwareAcceleratedVideoEncoder,
@@ -133,6 +158,13 @@ typedef enum VTH264Entropy{
VT_CABAC
} VTH264Entropy;
+typedef enum VT_HEVCProfile {
+ HEVC_PROF_AUTO,
+ HEVC_PROF_MAIN,
+ HEVC_PROF_MAIN10,
+ HEVC_PROF_COUNT
+} VT_HEVCProfile;
+
static const uint8_t start_code[] = { 0, 0, 0, 1 };
typedef struct ExtraSEI {
@@ -149,10 +181,12 @@ typedef struct BufNode {
typedef struct VTEncContext {
AVClass *class;
+ enum AVCodecID codec_id;
VTCompressionSessionRef session;
CFStringRef ycbcr_matrix;
CFStringRef color_primaries;
CFStringRef transfer_function;
+ getParameterSetAtIndex get_param_set_func;
pthread_mutex_t lock;
pthread_cond_t cv_sample_sent;
@@ -348,6 +382,7 @@ static CMVideoCodecType get_cm_codec_type(enum AVCodecID id)
{
switch (id) {
case AV_CODEC_ID_H264: return kCMVideoCodecType_H264;
+ case AV_CODEC_ID_HEVC: return kCMVideoCodecType_HEVC;
default: return 0;
}
}
@@ -365,17 +400,18 @@ static int get_params_size(
CMVideoFormatDescriptionRef vid_fmt,
size_t *size)
{
+ VTEncContext *vtctx = avctx->priv_data;
size_t total_size = 0;
size_t ps_count;
int is_count_bad = 0;
size_t i;
int status;
- status = CMVideoFormatDescriptionGetH264ParameterSetAtIndex(vid_fmt,
- 0,
- NULL,
- NULL,
- &ps_count,
- NULL);
+ status = vtctx->get_param_set_func(vid_fmt,
+ 0,
+ NULL,
+ NULL,
+ &ps_count,
+ NULL);
if (status) {
is_count_bad = 1;
ps_count = 0;
@@ -385,12 +421,12 @@ static int get_params_size(
for (i = 0; i < ps_count || is_count_bad; i++) {
const uint8_t *ps;
size_t ps_size;
- status = CMVideoFormatDescriptionGetH264ParameterSetAtIndex(vid_fmt,
- i,
- &ps,
- &ps_size,
- NULL,
- NULL);
+ status = vtctx->get_param_set_func(vid_fmt,
+ i,
+ &ps,
+ &ps_size,
+ NULL,
+ NULL);
if (status) {
/*
* When ps_count is invalid, status != 0 ends the loop normally
@@ -419,18 +455,19 @@ static int copy_param_sets(
uint8_t *dst,
size_t dst_size)
{
+ VTEncContext *vtctx = avctx->priv_data;
size_t ps_count;
int is_count_bad = 0;
int status;
size_t offset = 0;
size_t i;
- status = CMVideoFormatDescriptionGetH264ParameterSetAtIndex(vid_fmt,
- 0,
- NULL,
- NULL,
- &ps_count,
- NULL);
+ status = vtctx->get_param_set_func(vid_fmt,
+ 0,
+ NULL,
+ NULL,
+ &ps_count,
+ NULL);
if (status) {
is_count_bad = 1;
ps_count = 0;
@@ -443,12 +480,12 @@ static int copy_param_sets(
size_t ps_size;
size_t next_offset;
- status = CMVideoFormatDescriptionGetH264ParameterSetAtIndex(vid_fmt,
- i,
- &ps,
- &ps_size,
- NULL,
- NULL);
+ status = vtctx->get_param_set_func(vid_fmt,
+ i,
+ &ps,
+ &ps_size,
+ NULL,
+ NULL);
if (status) {
if (i > 0 && is_count_bad) status = 0;
@@ -548,6 +585,7 @@ static int get_length_code_size(
CMSampleBufferRef sample_buffer,
size_t *size)
{
+ VTEncContext *vtctx = avctx->priv_data;
CMVideoFormatDescriptionRef vid_fmt;
int isize;
int status;
@@ -558,12 +596,12 @@ static int get_length_code_size(
return AVERROR_EXTERNAL;
}
- status = CMVideoFormatDescriptionGetH264ParameterSetAtIndex(vid_fmt,
- 0,
- NULL,
- NULL,
- NULL,
- &isize);
+ status = vtctx->get_param_set_func(vid_fmt,
+ 0,
+ NULL,
+ NULL,
+ NULL,
+ &isize);
if (status) {
av_log(avctx, AV_LOG_ERROR, "Error getting length code size: %d\n", status);
return AVERROR_EXTERNAL;
@@ -579,8 +617,8 @@ static int get_length_code_size(
* If profile_level_val is NULL and this method returns true, don't specify the
* profile/level to the encoder.
*/
-static bool get_vt_profile_level(AVCodecContext *avctx,
- CFStringRef *profile_level_val)
+static bool get_vt_h264_profile_level(AVCodecContext *avctx,
+ CFStringRef *profile_level_val)
{
VTEncContext *vtctx = avctx->priv_data;
int64_t profile = vtctx->profile;
@@ -670,6 +708,41 @@ static bool get_vt_profile_level(AVCodecContext *avctx,
return true;
}
+/*
+ * Returns true on success.
+ *
+ * If profile_level_val is NULL and this method returns true, don't specify the
+ * profile/level to the encoder.
+ */
+static bool get_vt_hevc_profile_level(AVCodecContext *avctx,
+ CFStringRef *profile_level_val)
+{
+ VTEncContext *vtctx = avctx->priv_data;
+ int64_t profile = vtctx->profile;
+
+ *profile_level_val = NULL;
+
+ switch (profile) {
+ case HEVC_PROF_AUTO:
+ return true;
+ case HEVC_PROF_MAIN:
+ *profile_level_val =
+ compat_keys.kVTProfileLevel_HEVC_Main_AutoLevel;
+ break;
+ case HEVC_PROF_MAIN10:
+ *profile_level_val =
+ compat_keys.kVTProfileLevel_HEVC_Main10_AutoLevel;
+ break;
+ }
+
+ if (!*profile_level_val) {
+ av_log(avctx, AV_LOG_ERROR, "Invalid Profile/Level.\n");
+ return false;
+ }
+
+ return true;
+}
+
static int get_cv_pixel_format(AVCodecContext* avctx,
enum AVPixelFormat fmt,
enum AVColorRange range,
@@ -944,52 +1017,54 @@ static int vtenc_create_encoder(AVCodecContext *avctx,
return AVERROR_EXTERNAL;
}
- bytes_per_second_value = max_rate >> 3;
- bytes_per_second = CFNumberCreate(kCFAllocatorDefault,
- kCFNumberSInt64Type,
- &bytes_per_second_value);
- if (!bytes_per_second) {
- return AVERROR(ENOMEM);
- }
- one_second_value = 1;
- one_second = CFNumberCreate(kCFAllocatorDefault,
- kCFNumberSInt64Type,
- &one_second_value);
- if (!one_second) {
- CFRelease(bytes_per_second);
- return AVERROR(ENOMEM);
- }
- nums[0] = bytes_per_second;
- nums[1] = one_second;
- data_rate_limits = CFArrayCreate(kCFAllocatorDefault,
- nums,
- 2,
- &kCFTypeArrayCallBacks);
+ if (vtctx->codec_id == AV_CODEC_ID_H264) {
+ bytes_per_second_value = max_rate >> 3;
+ bytes_per_second = CFNumberCreate(kCFAllocatorDefault,
+ kCFNumberSInt64Type,
+ &bytes_per_second_value);
+ if (!bytes_per_second) {
+ return AVERROR(ENOMEM);
+ }
+ one_second_value = 1;
+ one_second = CFNumberCreate(kCFAllocatorDefault,
+ kCFNumberSInt64Type,
+ &one_second_value);
+ if (!one_second) {
+ CFRelease(bytes_per_second);
+ return AVERROR(ENOMEM);
+ }
+ nums[0] = (void *)bytes_per_second;
+ nums[1] = (void *)one_second;
+ data_rate_limits = CFArrayCreate(kCFAllocatorDefault,
+ (const void **)nums,
+ 2,
+ &kCFTypeArrayCallBacks);
+
+ if (!data_rate_limits) {
+ CFRelease(bytes_per_second);
+ CFRelease(one_second);
+ return AVERROR(ENOMEM);
+ }
+ status = VTSessionSetProperty(vtctx->session,
+ kVTCompressionPropertyKey_DataRateLimits,
+ data_rate_limits);
- if (!data_rate_limits) {
CFRelease(bytes_per_second);
CFRelease(one_second);
- return AVERROR(ENOMEM);
- }
- status = VTSessionSetProperty(vtctx->session,
- kVTCompressionPropertyKey_DataRateLimits,
- data_rate_limits);
-
- CFRelease(bytes_per_second);
- CFRelease(one_second);
- CFRelease(data_rate_limits);
+ CFRelease(data_rate_limits);
- if (status) {
- av_log(avctx, AV_LOG_ERROR, "Error setting max bitrate property: %d\n", status);
- return AVERROR_EXTERNAL;
- }
-
- if (profile_level) {
- status = VTSessionSetProperty(vtctx->session,
- kVTCompressionPropertyKey_ProfileLevel,
- profile_level);
if (status) {
- av_log(avctx, AV_LOG_ERROR, "Error setting profile/level property: %d\n", status);
+ av_log(avctx, AV_LOG_ERROR, "Error setting max bitrate property: %d\n", status);
+ return AVERROR_EXTERNAL;
+ }
+
+ if (profile_level) {
+ status = VTSessionSetProperty(vtctx->session,
+ kVTCompressionPropertyKey_ProfileLevel,
+ profile_level);
+ if (status) {
+ av_log(avctx, AV_LOG_ERROR, "Error setting profile/level property: %d\n", status);
+ }
}
}
@@ -1205,18 +1280,27 @@ static av_cold int vtenc_init(AVCodecContext *avctx)
return AVERROR(EINVAL);
}
- vtctx->has_b_frames = avctx->max_b_frames > 0;
- if(vtctx->has_b_frames && vtctx->profile == H264_PROF_BASELINE){
- av_log(avctx, AV_LOG_WARNING, "Cannot use B-frames with baseline profile. Output will not contain B-frames.\n");
- vtctx->has_b_frames = false;
- }
+ vtctx->codec_id = avctx->codec_id;
- if (vtctx->entropy == VT_CABAC && vtctx->profile == H264_PROF_BASELINE) {
- av_log(avctx, AV_LOG_WARNING, "CABAC entropy requires 'main' or 'high' profile, but baseline was requested. Encode will not use CABAC entropy.\n");
- vtctx->entropy = VT_ENTROPY_NOT_SET;
- }
+ if (vtctx->codec_id == AV_CODEC_ID_H264) {
+ vtctx->get_param_set_func = CMVideoFormatDescriptionGetH264ParameterSetAtIndex;
+
+ vtctx->has_b_frames = avctx->max_b_frames > 0;
+ if(vtctx->has_b_frames && vtctx->profile == H264_PROF_BASELINE){
+ av_log(avctx, AV_LOG_WARNING, "Cannot use B-frames with baseline profile. Output will not contain B-frames.\n");
+ vtctx->has_b_frames = false;
+ }
- if (!get_vt_profile_level(avctx, &profile_level)) return AVERROR(EINVAL);
+ if (vtctx->entropy == VT_CABAC && vtctx->profile == H264_PROF_BASELINE) {
+ av_log(avctx, AV_LOG_WARNING, "CABAC entropy requires 'main' or 'high' profile, but baseline was requested. Encode will not use CABAC entropy.\n");
+ vtctx->entropy = VT_ENTROPY_NOT_SET;
+ }
+
+ if (!get_vt_h264_profile_level(avctx, &profile_level)) return AVERROR(EINVAL);
+ } else {
+ vtctx->get_param_set_func = compat_keys.CMVideoFormatDescriptionGetHEVCParameterSetAtIndex;
+ if (!get_vt_hevc_profile_level(avctx, &profile_level)) return AVERROR(EINVAL);
+ }
vtctx->session = NULL;
@@ -2426,7 +2510,7 @@ static const enum AVPixelFormat pix_fmts[] = {
#define OFFSET(x) offsetof(VTEncContext, x)
#define VE AV_OPT_FLAG_VIDEO_PARAM | AV_OPT_FLAG_ENCODING_PARAM
-static const AVOption options[] = {
+static const AVOption h264_options[] = {
{ "profile", "Profile", OFFSET(profile), AV_OPT_TYPE_INT, { .i64 = H264_PROF_AUTO }, H264_PROF_AUTO, H264_PROF_COUNT, VE, "profile" },
{ "baseline", "Baseline Profile", 0, AV_OPT_TYPE_CONST, { .i64 = H264_PROF_BASELINE }, INT_MIN, INT_MAX, VE, "profile" },
{ "main", "Main Profile", 0, AV_OPT_TYPE_CONST, { .i64 = H264_PROF_MAIN }, INT_MIN, INT_MAX, VE, "profile" },
@@ -2469,7 +2553,7 @@ static const AVOption options[] = {
static const AVClass h264_videotoolbox_class = {
.class_name = "h264_videotoolbox",
.item_name = av_default_item_name,
- .option = options,
+ .option = h264_options,
.version = LIBAVUTIL_VERSION_INT,
};
@@ -2488,3 +2572,45 @@ AVCodec ff_h264_videotoolbox_encoder = {
.caps_internal = FF_CODEC_CAP_INIT_THREADSAFE |
FF_CODEC_CAP_INIT_CLEANUP,
};
+
+static const AVOption hevc_options[] = {
+ { "profile", "Profile", OFFSET(profile), AV_OPT_TYPE_INT, { .i64 = HEVC_PROF_AUTO }, HEVC_PROF_AUTO, HEVC_PROF_COUNT, VE, "profile" },
+ { "main", "Main Profile", 0, AV_OPT_TYPE_CONST, { .i64 = HEVC_PROF_MAIN }, INT_MIN, INT_MAX, VE, "profile" },
+ { "main10", "Main10 Profile", 0, AV_OPT_TYPE_CONST, { .i64 = HEVC_PROF_MAIN10 }, INT_MIN, INT_MAX, VE, "profile" },
+
+ { "allow_sw", "Allow software encoding", OFFSET(allow_sw), AV_OPT_TYPE_BOOL,
+ { .i64 = 0 }, 0, 1, VE },
+
+ { "realtime", "Hint that encoding should happen in real-time if not faster (e.g. capturing from camera).",
+ OFFSET(realtime), AV_OPT_TYPE_BOOL, { .i64 = 0 }, 0, 1, VE },
+
+ { "frames_before", "Other frames will come before the frames in this session. This helps smooth concatenation issues.",
+ OFFSET(frames_before), AV_OPT_TYPE_BOOL, { .i64 = 0 }, 0, 1, VE },
+ { "frames_after", "Other frames will come after the frames in this session. This helps smooth concatenation issues.",
+ OFFSET(frames_after), AV_OPT_TYPE_BOOL, { .i64 = 0 }, 0, 1, VE },
+
+ { NULL },
+};
+
+static const AVClass hevc_videotoolbox_class = {
+ .class_name = "hevc_videotoolbox",
+ .item_name = av_default_item_name,
+ .option = hevc_options,
+ .version = LIBAVUTIL_VERSION_INT,
+};
+
+AVCodec ff_hevc_videotoolbox_encoder = {
+ .name = "hevc_videotoolbox",
+ .long_name = NULL_IF_CONFIG_SMALL("VideoToolbox H.265 Encoder"),
+ .type = AVMEDIA_TYPE_VIDEO,
+ .id = AV_CODEC_ID_HEVC,
+ .priv_data_size = sizeof(VTEncContext),
+ .pix_fmts = pix_fmts,
+ .init = vtenc_init,
+ .encode2 = vtenc_frame,
+ .close = vtenc_close,
+ .capabilities = AV_CODEC_CAP_DELAY,
+ .priv_class = &hevc_videotoolbox_class,
+ .caps_internal = FF_CODEC_CAP_INIT_THREADSAFE |
+ FF_CODEC_CAP_INIT_CLEANUP,
+};
--
2.14.2
More information about the ffmpeg-devel
mailing list