Files
Raboneko 1839d571fb chore(ffmpeg): Properly update spec and patches (#7189) (#7241)
* fix(ffmpeg): Add VERSION_libnpp.txt

* chore: Update properly

* chore: Add new tracking files

(cherry picked from commit cf2a31cb75)

Co-authored-by: Gilver <rockgrub@disroot.org>
2025-11-08 16:44:06 -06:00

3442 lines
139 KiB
Diff

diff -Naur ffmpeg-7.1.2.old/configure ffmpeg-7.1.2/configure
--- ffmpeg-7.1.2.old/configure 2025-10-27 10:07:02.407490033 +0100
+++ ffmpeg-7.1.2/configure 2025-10-27 10:07:02.458434476 +0100
@@ -2466,6 +2466,7 @@
kCMVideoCodecType_HEVC
kCMVideoCodecType_HEVCWithAlpha
kCMVideoCodecType_VP9
+ kCMVideoCodecType_AV1
kCVPixelFormatType_420YpCbCr10BiPlanarVideoRange
kCVPixelFormatType_422YpCbCr8BiPlanarVideoRange
kCVPixelFormatType_422YpCbCr10BiPlanarVideoRange
@@ -3172,6 +3173,8 @@
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"
@@ -3342,6 +3345,7 @@
av1_mediacodec_decoder_deps="mediacodec"
av1_mediacodec_encoder_deps="mediacodec"
av1_mediacodec_encoder_select="extract_extradata_bsf"
+av1_mf_encoder_deps="mediafoundation"
av1_nvenc_encoder_deps="nvenc NV_ENC_PIC_PARAMS_AV1"
av1_nvenc_encoder_select="atsc_a53"
av1_qsv_decoder_select="qsvdec"
@@ -6724,6 +6728,7 @@
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"
diff -Naur ffmpeg-7.1.2.old/libavcodec/ac3dec.c ffmpeg-7.1.2/libavcodec/ac3dec.c
--- ffmpeg-7.1.2.old/libavcodec/ac3dec.c 2025-10-27 10:07:00.254471036 +0100
+++ ffmpeg-7.1.2/libavcodec/ac3dec.c 2025-10-27 10:07:02.463169316 +0100
@@ -253,72 +253,6 @@
}
/**
- * Parse the 'sync info' and 'bit stream info' from the AC-3 bitstream.
- * GetBitContext within AC3DecodeContext must point to
- * the start of the synchronized AC-3 bitstream.
- */
-static int ac3_parse_header(AC3DecodeContext *s)
-{
- GetBitContext *gbc = &s->gbc;
- int i;
-
- /* read the rest of the bsi. read twice for dual mono mode. */
- i = !s->channel_mode;
- do {
- s->dialog_normalization[(!s->channel_mode)-i] = -get_bits(gbc, 5);
- if (s->dialog_normalization[(!s->channel_mode)-i] == 0) {
- s->dialog_normalization[(!s->channel_mode)-i] = -31;
- }
- if (s->target_level != 0) {
- s->level_gain[(!s->channel_mode)-i] = powf(2.0f,
- (float)(s->target_level -
- s->dialog_normalization[(!s->channel_mode)-i])/6.0f);
- }
- if (s->compression_exists[(!s->channel_mode)-i] = get_bits1(gbc)) {
- s->heavy_dynamic_range[(!s->channel_mode)-i] =
- AC3_HEAVY_RANGE(get_bits(gbc, 8));
- }
- if (get_bits1(gbc))
- skip_bits(gbc, 8); //skip language code
- if (get_bits1(gbc))
- skip_bits(gbc, 7); //skip audio production information
- } while (i--);
-
- skip_bits(gbc, 2); //skip copyright bit and original bitstream bit
-
- /* skip the timecodes or parse the Alternate Bit Stream Syntax */
- if (s->bitstream_id != 6) {
- if (get_bits1(gbc))
- skip_bits(gbc, 14); //skip timecode1
- if (get_bits1(gbc))
- skip_bits(gbc, 14); //skip timecode2
- } else {
- if (get_bits1(gbc)) {
- s->preferred_downmix = get_bits(gbc, 2);
- s->center_mix_level_ltrt = get_bits(gbc, 3);
- s->surround_mix_level_ltrt = av_clip(get_bits(gbc, 3), 3, 7);
- s->center_mix_level = get_bits(gbc, 3);
- s->surround_mix_level = av_clip(get_bits(gbc, 3), 3, 7);
- }
- if (get_bits1(gbc)) {
- s->dolby_surround_ex_mode = get_bits(gbc, 2);
- s->dolby_headphone_mode = get_bits(gbc, 2);
- skip_bits(gbc, 10); // skip adconvtyp (1), xbsi2 (8), encinfo (1)
- }
- }
-
- /* skip additional bitstream info */
- if (get_bits1(gbc)) {
- i = get_bits(gbc, 6);
- do {
- skip_bits(gbc, 8);
- } while (i--);
- }
-
- return 0;
-}
-
-/**
* Common function to parse AC-3 or E-AC-3 frame header
*/
static int parse_frame_header(AC3DecodeContext *s)
@@ -375,10 +309,25 @@
s->dba_syntax = 1;
s->skip_syntax = 1;
memset(s->channel_uses_aht, 0, sizeof(s->channel_uses_aht));
- return ac3_parse_header(s);
+ /* volume control params */
+ for (int i = 0; i < (s->channel_mode ? 1 : 2); i++) {
+ s->dialog_normalization[i] = hdr.dialog_normalization[i];
+ if (s->dialog_normalization[i] == 0) {
+ s->dialog_normalization[i] = -31;
+ }
+ if (s->target_level != 0) {
+ s->level_gain[i] = powf(2.0f,
+ (float)(s->target_level - s->dialog_normalization[i])/6.0f);
+ }
+ s->compression_exists[i] = hdr.compression_exists[i];
+ if (s->compression_exists[i]) {
+ s->heavy_dynamic_range[i] = AC3_HEAVY_RANGE(hdr.heavy_dynamic_range[i]);
+ }
+ }
+ return 0;
} else if (CONFIG_EAC3_DECODER) {
s->eac3 = 1;
- return ff_eac3_parse_header(s);
+ return ff_eac3_parse_header(s, &hdr);
} else {
av_log(s->avctx, AV_LOG_ERROR, "E-AC-3 support not compiled in\n");
return AVERROR(ENOSYS);
@@ -1562,6 +1511,9 @@
av_log(avctx, AV_LOG_ERROR, "invalid frame type\n");
}
break;
+ case AC3_PARSE_ERROR_CHANNEL_MAP:
+ av_log(avctx, AV_LOG_ERROR, "invalid channel map\n");
+ return AVERROR_INVALIDDATA;
case AC3_PARSE_ERROR_CRC:
break;
default: // Normal AVERROR do not try to recover.
diff -Naur ffmpeg-7.1.2.old/libavcodec/ac3dec.h ffmpeg-7.1.2/libavcodec/ac3dec.h
--- ffmpeg-7.1.2.old/libavcodec/ac3dec.h 2025-10-27 10:07:00.587473974 +0100
+++ ffmpeg-7.1.2/libavcodec/ac3dec.h 2025-10-27 10:07:02.463287298 +0100
@@ -255,11 +255,12 @@
AVChannelLayout downmix_layout;
} AC3DecodeContext;
+struct AC3HeaderInfo;
/**
* Parse the E-AC-3 frame header.
* This parses both the bit stream info and audio frame header.
*/
-static int ff_eac3_parse_header(AC3DecodeContext *s);
+static int ff_eac3_parse_header(AC3DecodeContext *s, const struct AC3HeaderInfo *hdr);
/**
* Decode mantissas in a single channel for the entire frame.
diff -Naur ffmpeg-7.1.2.old/libavcodec/ac3defs.h ffmpeg-7.1.2/libavcodec/ac3defs.h
--- ffmpeg-7.1.2.old/libavcodec/ac3defs.h 2025-10-27 10:07:00.586473965 +0100
+++ ffmpeg-7.1.2/libavcodec/ac3defs.h 2025-10-27 10:07:02.463340118 +0100
@@ -34,6 +34,8 @@
#define AC3_CRITICAL_BANDS 50
#define AC3_MAX_CPL_BANDS 18
+#define EAC3_SR_CODE_REDUCED 3
+
/* exponent encoding strategy */
#define EXP_REUSE 0
#define EXP_NEW 1
diff -Naur ffmpeg-7.1.2.old/libavcodec/ac3_parser.c ffmpeg-7.1.2/libavcodec/ac3_parser.c
--- ffmpeg-7.1.2.old/libavcodec/ac3_parser.c 2025-10-27 10:07:00.290471354 +0100
+++ ffmpeg-7.1.2/libavcodec/ac3_parser.c 2025-10-27 10:07:02.462962336 +0100
@@ -73,6 +73,217 @@
return i;
}
+/**
+ * Parse the 'sync info' and 'bit stream info' from the AC-3 bitstream.
+ * GetBitContext within AC3DecodeContext must point to
+ * the start of the synchronized AC-3 bitstream.
+ */
+static int ac3_parse_header(GetBitContext *gbc, AC3HeaderInfo *hdr)
+{
+ /* read the rest of the bsi. read twice for dual mono mode. */
+ for (int i = 0; i < (hdr->channel_mode ? 1 : 2); i++) {
+ hdr->dialog_normalization[i] = -get_bits(gbc, 5);
+ hdr->compression_exists[i] = get_bits1(gbc);
+ if (hdr->compression_exists[i])
+ hdr->heavy_dynamic_range[i] = get_bits(gbc, 8);
+ if (get_bits1(gbc))
+ skip_bits(gbc, 8); //skip language code
+ if (get_bits1(gbc))
+ skip_bits(gbc, 7); //skip audio production information
+ }
+
+ skip_bits(gbc, 2); //skip copyright bit and original bitstream bit
+
+ /* skip the timecodes or parse the Alternate Bit Stream Syntax */
+ if (hdr->bitstream_id != 6) {
+ if (get_bits1(gbc))
+ skip_bits(gbc, 14); //skip timecode1
+ if (get_bits1(gbc))
+ skip_bits(gbc, 14); //skip timecode2
+ } else {
+ if (get_bits1(gbc)) {
+ hdr->preferred_downmix = get_bits(gbc, 2);
+ hdr->center_mix_level_ltrt = get_bits(gbc, 3);
+ hdr->surround_mix_level_ltrt = av_clip(get_bits(gbc, 3), 3, 7);
+ hdr->center_mix_level = get_bits(gbc, 3);
+ hdr->surround_mix_level = av_clip(get_bits(gbc, 3), 3, 7);
+ }
+ if (get_bits1(gbc)) {
+ hdr->dolby_surround_ex_mode = get_bits(gbc, 2);
+ hdr->dolby_headphone_mode = get_bits(gbc, 2);
+ skip_bits(gbc, 10); // skip adconvtyp (1), xbsi2 (8), encinfo (1)
+ }
+ }
+
+ /* skip additional bitstream info */
+ if (get_bits1(gbc)) {
+ int i = get_bits(gbc, 6);
+ do {
+ skip_bits(gbc, 8);
+ } while (i--);
+ }
+
+ return 0;
+}
+
+static int eac3_parse_header(GetBitContext *gbc, AC3HeaderInfo *hdr)
+{
+ if (hdr->frame_type == EAC3_FRAME_TYPE_RESERVED)
+ return AC3_PARSE_ERROR_FRAME_TYPE;
+ if (hdr->substreamid)
+ return AC3_PARSE_ERROR_FRAME_TYPE;
+
+ skip_bits(gbc, 5); // skip bitstream id
+
+ /* volume control params */
+ for (int i = 0; i < (hdr->channel_mode ? 1 : 2); i++) {
+ hdr->dialog_normalization[i] = -get_bits(gbc, 5);
+ hdr->compression_exists[i] = get_bits1(gbc);
+ if (hdr->compression_exists[i])
+ hdr->heavy_dynamic_range[i] = get_bits(gbc, 8);
+ }
+
+ /* dependent stream channel map */
+ if (hdr->frame_type == EAC3_FRAME_TYPE_DEPENDENT) {
+ hdr->channel_map_present = get_bits1(gbc);
+ if (hdr->channel_map_present) {
+ int64_t channel_layout = 0;
+ int channel_map = get_bits(gbc, 16);
+
+ for (int i = 0; i < 16; i++)
+ if (channel_map & (1 << (EAC3_MAX_CHANNELS - i - 1)))
+ channel_layout |= ff_eac3_custom_channel_map_locations[i][1];
+
+ if (av_popcount64(channel_layout) > EAC3_MAX_CHANNELS) {
+ return AC3_PARSE_ERROR_CHANNEL_MAP;
+ }
+ hdr->channel_map = channel_map;
+ }
+ }
+
+ /* mixing metadata */
+ if (get_bits1(gbc)) {
+ /* center and surround mix levels */
+ if (hdr->channel_mode > AC3_CHMODE_STEREO) {
+ hdr->preferred_downmix = get_bits(gbc, 2);
+ if (hdr->channel_mode & 1) {
+ /* if three front channels exist */
+ hdr->center_mix_level_ltrt = get_bits(gbc, 3);
+ hdr->center_mix_level = get_bits(gbc, 3);
+ }
+ if (hdr->channel_mode & 4) {
+ /* if a surround channel exists */
+ hdr->surround_mix_level_ltrt = av_clip(get_bits(gbc, 3), 3, 7);
+ hdr->surround_mix_level = av_clip(get_bits(gbc, 3), 3, 7);
+ }
+ }
+
+ /* lfe mix level */
+ if (hdr->lfe_on && (hdr->lfe_mix_level_exists = get_bits1(gbc))) {
+ hdr->lfe_mix_level = get_bits(gbc, 5);
+ }
+
+ /* info for mixing with other streams and substreams */
+ if (hdr->frame_type == EAC3_FRAME_TYPE_INDEPENDENT) {
+ for (int i = 0; i < (hdr->channel_mode ? 1 : 2); i++) {
+ // TODO: apply program scale factor
+ if (get_bits1(gbc)) {
+ skip_bits(gbc, 6); // skip program scale factor
+ }
+ }
+ if (get_bits1(gbc)) {
+ skip_bits(gbc, 6); // skip external program scale factor
+ }
+ /* skip mixing parameter data */
+ switch(get_bits(gbc, 2)) {
+ case 1: skip_bits(gbc, 5); break;
+ case 2: skip_bits(gbc, 12); break;
+ case 3: {
+ int mix_data_size = (get_bits(gbc, 5) + 2) << 3;
+ skip_bits_long(gbc, mix_data_size);
+ break;
+ }
+ }
+ /* skip pan information for mono or dual mono source */
+ if (hdr->channel_mode < AC3_CHMODE_STEREO) {
+ for (int i = 0; i < (hdr->channel_mode ? 1 : 2); i++) {
+ if (get_bits1(gbc)) {
+ /* note: this is not in the ATSC A/52B specification
+ reference: ETSI TS 102 366 V1.1.1
+ section: E.1.3.1.25 */
+ skip_bits(gbc, 8); // skip pan mean direction index
+ skip_bits(gbc, 6); // skip reserved paninfo bits
+ }
+ }
+ }
+ /* skip mixing configuration information */
+ if (get_bits1(gbc)) {
+ for (int i = 0; i < hdr->num_blocks; i++) {
+ if (hdr->num_blocks == 1 || get_bits1(gbc)) {
+ skip_bits(gbc, 5);
+ }
+ }
+ }
+ }
+ }
+
+ /* informational metadata */
+ if (get_bits1(gbc)) {
+ hdr->bitstream_mode = get_bits(gbc, 3);
+ skip_bits(gbc, 2); // skip copyright bit and original bitstream bit
+ if (hdr->channel_mode == AC3_CHMODE_STEREO) {
+ hdr->dolby_surround_mode = get_bits(gbc, 2);
+ hdr->dolby_headphone_mode = get_bits(gbc, 2);
+ }
+ if (hdr->channel_mode >= AC3_CHMODE_2F2R) {
+ hdr->dolby_surround_ex_mode = get_bits(gbc, 2);
+ }
+ for (int i = 0; i < (hdr->channel_mode ? 1 : 2); i++) {
+ if (get_bits1(gbc)) {
+ skip_bits(gbc, 8); // skip mix level, room type, and A/D converter type
+ }
+ }
+ if (hdr->sr_code != EAC3_SR_CODE_REDUCED) {
+ skip_bits1(gbc); // skip source sample rate code
+ }
+ }
+
+ /* converter synchronization flag
+ If frames are less than six blocks, this bit should be turned on
+ once every 6 blocks to indicate the start of a frame set.
+ reference: RFC 4598, Section 2.1.3 Frame Sets */
+ if (hdr->frame_type == EAC3_FRAME_TYPE_INDEPENDENT && hdr->num_blocks != 6) {
+ skip_bits1(gbc); // skip converter synchronization flag
+ }
+
+ /* original frame size code if this stream was converted from AC-3 */
+ if (hdr->frame_type == EAC3_FRAME_TYPE_AC3_CONVERT &&
+ (hdr->num_blocks == 6 || get_bits1(gbc))) {
+ skip_bits(gbc, 6); // skip frame size code
+ }
+
+ /* additional bitstream info */
+ if (get_bits1(gbc)) {
+ int addbsil = get_bits(gbc, 6);
+ for (int i = 0; i < addbsil + 1; i++) {
+ if (i == 0) {
+ /* In this 8 bit chunk, the LSB is equal to flag_ec3_extension_type_a
+ which can be used to detect Atmos presence */
+ skip_bits(gbc, 7);
+ hdr->eac3_extension_type_a = get_bits1(gbc);
+ if (hdr->eac3_extension_type_a) {
+ hdr->complexity_index_type_a = get_bits(gbc, 8);
+ i++;
+ }
+ } else {
+ skip_bits(gbc, 8); // skip additional bit stream info
+ }
+ }
+ }
+
+ return 0;
+}
+
int ff_ac3_parse_header(GetBitContext *gbc, AC3HeaderInfo *hdr)
{
int frame_size_code;
@@ -133,6 +344,10 @@
hdr->frame_size = ff_ac3_frame_size_tab[frame_size_code][hdr->sr_code] * 2;
hdr->frame_type = EAC3_FRAME_TYPE_AC3_CONVERT; //EAC3_FRAME_TYPE_INDEPENDENT;
hdr->substreamid = 0;
+
+ int ret = ac3_parse_header(gbc, hdr);
+ if (ret < 0)
+ return ret;
} else {
/* Enhanced AC-3 */
hdr->crc1 = 0;
@@ -165,6 +380,10 @@
hdr->bit_rate = 8LL * hdr->frame_size * hdr->sample_rate /
(hdr->num_blocks * 256);
hdr->channels = ff_ac3_channels_tab[hdr->channel_mode] + hdr->lfe_on;
+
+ int ret = eac3_parse_header(gbc, hdr);
+ if (ret < 0)
+ return ret;
}
hdr->channel_layout = ff_ac3_channel_layout_tab[hdr->channel_mode];
if (hdr->lfe_on)
@@ -202,9 +421,13 @@
{
GetBitContext gb;
AC3HeaderInfo hdr;
+ uint8_t tmp[32 + AV_INPUT_BUFFER_PADDING_SIZE];
int err;
- err = init_get_bits8(&gb, buf, size);
+ size = FFMIN(32, size);
+ memcpy(tmp, buf, size);
+ memset(tmp + size, 0, AV_INPUT_BUFFER_PADDING_SIZE);
+ err = init_get_bits8(&gb, tmp, size);
if (err < 0)
return AVERROR_INVALIDDATA;
err = ff_ac3_parse_header(&gb, &hdr);
diff -Naur ffmpeg-7.1.2.old/libavcodec/ac3_parser_internal.h ffmpeg-7.1.2/libavcodec/ac3_parser_internal.h
--- ffmpeg-7.1.2.old/libavcodec/ac3_parser_internal.h 2025-10-27 10:07:00.519473374 +0100
+++ ffmpeg-7.1.2/libavcodec/ac3_parser_internal.h 2025-10-27 10:07:02.463059088 +0100
@@ -46,6 +46,7 @@
int substreamid; ///< substream identification
int center_mix_level; ///< Center mix level index
int surround_mix_level; ///< Surround mix level index
+ uint8_t channel_map_present;
uint16_t channel_map;
int num_blocks; ///< number of audio blocks
int dolby_surround_mode;
@@ -62,6 +63,23 @@
uint64_t channel_layout;
int8_t ac3_bit_rate_code;
/** @} */
+
+ /** @name enhanced eac3 extension coded elements
+ * @{
+ */
+ int8_t dialog_normalization[2];
+ uint8_t compression_exists[2];
+ uint8_t heavy_dynamic_range[2];
+ uint8_t center_mix_level_ltrt; ///< Center mix level index
+ uint8_t surround_mix_level_ltrt; ///< Surround mix level index
+ uint8_t dolby_headphone_mode;
+ uint8_t dolby_surround_ex_mode;
+ uint8_t lfe_mix_level_exists;
+ uint8_t lfe_mix_level;
+ uint8_t preferred_downmix;
+ uint8_t eac3_extension_type_a;
+ uint8_t complexity_index_type_a;
+ /** @} */
} AC3HeaderInfo;
typedef enum {
@@ -71,6 +89,7 @@
AC3_PARSE_ERROR_FRAME_SIZE = -0x4030c0a,
AC3_PARSE_ERROR_FRAME_TYPE = -0x5030c0a,
AC3_PARSE_ERROR_CRC = -0x6030c0a,
+ AC3_PARSE_ERROR_CHANNEL_MAP = -0x7030c0a,
} AC3ParseError;
/**
diff -Naur ffmpeg-7.1.2.old/libavcodec/allcodecs.c ffmpeg-7.1.2/libavcodec/allcodecs.c
--- ffmpeg-7.1.2.old/libavcodec/allcodecs.c 2025-10-27 10:07:01.864539025 +0100
+++ ffmpeg-7.1.2/libavcodec/allcodecs.c 2025-10-27 10:07:02.458796899 +0100
@@ -839,6 +839,7 @@
extern const FFCodec ff_av1_qsv_decoder;
extern const FFCodec ff_av1_qsv_encoder;
extern const FFCodec ff_av1_amf_encoder;
+extern const FFCodec ff_av1_mf_encoder;
extern const FFCodec ff_av1_vaapi_encoder;
extern const FFCodec ff_libopenh264_encoder;
extern const FFCodec ff_libopenh264_decoder;
diff -Naur ffmpeg-7.1.2.old/libavcodec/amfenc_av1.c ffmpeg-7.1.2/libavcodec/amfenc_av1.c
--- ffmpeg-7.1.2.old/libavcodec/amfenc_av1.c 2025-10-27 10:07:00.536473524 +0100
+++ ffmpeg-7.1.2/libavcodec/amfenc_av1.c 2025-10-27 10:07:02.440481173 +0100
@@ -116,6 +116,7 @@
{ "none", "no adaptive quantization", 0, AV_OPT_TYPE_CONST, {.i64 = AMF_VIDEO_ENCODER_AV1_AQ_MODE_NONE }, 0, 0, VE, .unit = "adaptive_quantisation_mode" },
{ "caq", "context adaptive quantization", 0, AV_OPT_TYPE_CONST, {.i64 = AMF_VIDEO_ENCODER_AV1_AQ_MODE_CAQ }, 0, 0, VE, .unit = "adaptive_quantisation_mode" },
+ { "forced_idr", "Force I frames to be IDR frames", OFFSET(forced_idr), AV_OPT_TYPE_BOOL, { .i64 = 0 }, 0, 1, VE },
{ "align", "alignment mode", OFFSET(align), AV_OPT_TYPE_INT, {.i64 = AMF_VIDEO_ENCODER_AV1_ALIGNMENT_MODE_NO_RESTRICTIONS }, AMF_VIDEO_ENCODER_AV1_ALIGNMENT_MODE_64X16_ONLY, AMF_VIDEO_ENCODER_AV1_ALIGNMENT_MODE_NO_RESTRICTIONS, VE, .unit = "align" },
{ "64x16", "", 0, AV_OPT_TYPE_CONST, {.i64 = AMF_VIDEO_ENCODER_AV1_ALIGNMENT_MODE_64X16_ONLY }, 0, 0, VE, .unit = "align" },
@@ -186,6 +187,8 @@
AMFRate framerate;
AMFSize framesize = AMFConstructSize(avctx->width, avctx->height);
amf_int64 color_depth;
+ amf_int64 color_primaries;
+ amf_int64 transfer_characteristic;
amf_int64 color_profile;
enum AVPixelFormat pix_fmt;
@@ -238,7 +241,11 @@
}
/// Color profile
+ color_primaries = ff_amf_get_color_primaries(avctx);
+ transfer_characteristic = ff_amf_get_transfer_characteristic(avctx);
color_profile = ff_amf_get_color_profile(avctx);
+ AMF_ASSIGN_PROPERTY_INT64(res, ctx->encoder, AMF_VIDEO_ENCODER_AV1_OUTPUT_COLOR_PRIMARIES, color_primaries);
+ AMF_ASSIGN_PROPERTY_INT64(res, ctx->encoder, AMF_VIDEO_ENCODER_AV1_OUTPUT_TRANSFER_CHARACTERISTIC, transfer_characteristic);
AMF_ASSIGN_PROPERTY_INT64(res, ctx->encoder, AMF_VIDEO_ENCODER_AV1_OUTPUT_COLOR_PROFILE, color_profile);
/// Color Depth
@@ -250,16 +257,6 @@
}
AMF_ASSIGN_PROPERTY_INT64(res, ctx->encoder, AMF_VIDEO_ENCODER_AV1_COLOR_BIT_DEPTH, color_depth);
- AMF_ASSIGN_PROPERTY_INT64(res, ctx->encoder, AMF_VIDEO_ENCODER_AV1_OUTPUT_COLOR_PROFILE, color_profile);
- if (color_depth == AMF_COLOR_BIT_DEPTH_8) {
- /// Color Transfer Characteristics (AMF matches ISO/IEC)
- AMF_ASSIGN_PROPERTY_INT64(res, ctx->encoder, AMF_VIDEO_ENCODER_AV1_OUTPUT_TRANSFER_CHARACTERISTIC, AMF_COLOR_TRANSFER_CHARACTERISTIC_BT709);
- /// Color Primaries (AMF matches ISO/IEC)
- AMF_ASSIGN_PROPERTY_INT64(res, ctx->encoder, AMF_VIDEO_ENCODER_AV1_OUTPUT_COLOR_PRIMARIES, AMF_COLOR_PRIMARIES_BT709);
- } else {
- AMF_ASSIGN_PROPERTY_INT64(res, ctx->encoder, AMF_VIDEO_ENCODER_AV1_OUTPUT_TRANSFER_CHARACTERISTIC, AMF_COLOR_TRANSFER_CHARACTERISTIC_SMPTE2084);
- AMF_ASSIGN_PROPERTY_INT64(res, ctx->encoder, AMF_VIDEO_ENCODER_AV1_OUTPUT_COLOR_PRIMARIES, AMF_COLOR_PRIMARIES_BT2020);
- }
profile_level = avctx->level;
if (profile_level == AV_LEVEL_UNKNOWN) {
diff -Naur ffmpeg-7.1.2.old/libavcodec/amfenc.c ffmpeg-7.1.2/libavcodec/amfenc.c
--- ffmpeg-7.1.2.old/libavcodec/amfenc.c 2025-10-27 10:07:00.387472209 +0100
+++ ffmpeg-7.1.2/libavcodec/amfenc.c 2025-10-27 10:07:02.445244447 +0100
@@ -415,10 +415,6 @@
else
pix_fmt = avctx->pix_fmt;
- if (pix_fmt == AV_PIX_FMT_P010) {
- AMF_RETURN_IF_FALSE(ctx, ctx->version >= AMF_MAKE_FULL_VERSION(1, 4, 32, 0), AVERROR_UNKNOWN, "10-bit encoder is not supported by AMD GPU drivers versions lower than 23.30.\n");
- }
-
ctx->format = amf_av_to_amf_format(pix_fmt);
AMF_RETURN_IF_FALSE(ctx, ctx->format != AMF_SURFACE_UNKNOWN, AVERROR(EINVAL),
"Format %s is not supported\n", av_get_pix_fmt_name(pix_fmt));
@@ -766,11 +762,50 @@
switch (avctx->codec->id) {
case AV_CODEC_ID_H264:
AMF_ASSIGN_PROPERTY_INT64(res, surface, AMF_VIDEO_ENCODER_INSERT_AUD, !!ctx->aud);
+ switch (frame->pict_type) {
+ case AV_PICTURE_TYPE_I:
+ if (ctx->forced_idr) {
+ AMF_ASSIGN_PROPERTY_INT64(res, surface, AMF_VIDEO_ENCODER_INSERT_SPS, 1);
+ AMF_ASSIGN_PROPERTY_INT64(res, surface, AMF_VIDEO_ENCODER_INSERT_PPS, 1);
+ AMF_ASSIGN_PROPERTY_INT64(res, surface, AMF_VIDEO_ENCODER_FORCE_PICTURE_TYPE, AMF_VIDEO_ENCODER_PICTURE_TYPE_IDR);
+ } else {
+ AMF_ASSIGN_PROPERTY_INT64(res, surface, AMF_VIDEO_ENCODER_FORCE_PICTURE_TYPE, AMF_VIDEO_ENCODER_PICTURE_TYPE_I);
+ }
+ break;
+ case AV_PICTURE_TYPE_P:
+ AMF_ASSIGN_PROPERTY_INT64(res, surface, AMF_VIDEO_ENCODER_FORCE_PICTURE_TYPE, AMF_VIDEO_ENCODER_PICTURE_TYPE_P);
+ break;
+ case AV_PICTURE_TYPE_B:
+ AMF_ASSIGN_PROPERTY_INT64(res, surface, AMF_VIDEO_ENCODER_FORCE_PICTURE_TYPE, AMF_VIDEO_ENCODER_PICTURE_TYPE_B);
+ break;
+ }
break;
case AV_CODEC_ID_HEVC:
AMF_ASSIGN_PROPERTY_INT64(res, surface, AMF_VIDEO_ENCODER_HEVC_INSERT_AUD, !!ctx->aud);
+ switch (frame->pict_type) {
+ case AV_PICTURE_TYPE_I:
+ if (ctx->forced_idr) {
+ AMF_ASSIGN_PROPERTY_INT64(res, surface, AMF_VIDEO_ENCODER_HEVC_INSERT_HEADER, 1);
+ AMF_ASSIGN_PROPERTY_INT64(res, surface, AMF_VIDEO_ENCODER_HEVC_FORCE_PICTURE_TYPE, AMF_VIDEO_ENCODER_HEVC_PICTURE_TYPE_IDR);
+ } else {
+ AMF_ASSIGN_PROPERTY_INT64(res, surface, AMF_VIDEO_ENCODER_HEVC_FORCE_PICTURE_TYPE, AMF_VIDEO_ENCODER_HEVC_PICTURE_TYPE_I);
+ }
+ break;
+ case AV_PICTURE_TYPE_P:
+ AMF_ASSIGN_PROPERTY_INT64(res, surface, AMF_VIDEO_ENCODER_HEVC_FORCE_PICTURE_TYPE, AMF_VIDEO_ENCODER_HEVC_PICTURE_TYPE_P);
+ break;
+ }
+ break;
+ case AV_CODEC_ID_AV1:
+ if (frame->pict_type == AV_PICTURE_TYPE_I) {
+ if (ctx->forced_idr) {
+ AMF_ASSIGN_PROPERTY_INT64(res, surface, AMF_VIDEO_ENCODER_AV1_FORCE_INSERT_SEQUENCE_HEADER, 1);
+ AMF_ASSIGN_PROPERTY_INT64(res, surface, AMF_VIDEO_ENCODER_AV1_FORCE_FRAME_TYPE, AMF_VIDEO_ENCODER_AV1_FORCE_FRAME_TYPE_KEY);
+ } else {
+ AMF_ASSIGN_PROPERTY_INT64(res, surface, AMF_VIDEO_ENCODER_AV1_FORCE_FRAME_TYPE, AMF_VIDEO_ENCODER_AV1_FORCE_FRAME_TYPE_INTRA_ONLY);
+ }
+ }
break;
- //case AV_CODEC_ID_AV1 not supported
default:
break;
}
@@ -878,6 +913,115 @@
return ret;
}
+int ff_amf_get_color_primaries(AVCodecContext *avctx)
+{
+ amf_int64 color_primaries = AMF_COLOR_PRIMARIES_UNDEFINED;
+ switch (avctx->color_primaries) {
+ case AVCOL_PRI_BT709:
+ color_primaries = AMF_COLOR_PRIMARIES_BT709;
+ break;
+ case AVCOL_PRI_UNSPECIFIED:
+ color_primaries = AMF_COLOR_PRIMARIES_UNSPECIFIED;
+ break;
+ case AVCOL_PRI_RESERVED:
+ color_primaries = AMF_COLOR_PRIMARIES_RESERVED;
+ break;
+ case AVCOL_PRI_BT470M:
+ color_primaries = AMF_COLOR_PRIMARIES_BT470M;
+ break;
+ case AVCOL_PRI_BT470BG:
+ color_primaries = AMF_COLOR_PRIMARIES_BT470BG;
+ break;
+ case AVCOL_PRI_SMPTE170M:
+ color_primaries = AMF_COLOR_PRIMARIES_SMPTE170M;
+ break;
+ case AVCOL_PRI_SMPTE240M:
+ color_primaries = AMF_COLOR_PRIMARIES_SMPTE240M;
+ break;
+ case AVCOL_PRI_FILM:
+ color_primaries = AMF_COLOR_PRIMARIES_FILM;
+ break;
+ case AVCOL_PRI_BT2020:
+ color_primaries = AMF_COLOR_PRIMARIES_BT2020;
+ break;
+ case AVCOL_PRI_SMPTE428:
+ color_primaries = AMF_COLOR_PRIMARIES_SMPTE428;
+ break;
+ case AVCOL_PRI_SMPTE431:
+ color_primaries = AMF_COLOR_PRIMARIES_SMPTE431;
+ break;
+ case AVCOL_PRI_SMPTE432:
+ color_primaries = AMF_COLOR_PRIMARIES_SMPTE432;
+ break;
+ case AVCOL_PRI_EBU3213:
+ color_primaries = AMF_COLOR_PRIMARIES_JEDEC_P22;
+ break;
+ }
+ return color_primaries;
+}
+
+int ff_amf_get_transfer_characteristic(AVCodecContext *avctx)
+{
+ amf_int64 transfer_characteristic = AMF_COLOR_TRANSFER_CHARACTERISTIC_UNDEFINED;
+ switch (avctx->color_trc) {
+ case AVCOL_TRC_BT709:
+ transfer_characteristic = AMF_COLOR_TRANSFER_CHARACTERISTIC_BT709;
+ break;
+ case AVCOL_TRC_UNSPECIFIED:
+ transfer_characteristic = AMF_COLOR_TRANSFER_CHARACTERISTIC_UNSPECIFIED;
+ break;
+ case AVCOL_TRC_RESERVED:
+ transfer_characteristic = AMF_COLOR_TRANSFER_CHARACTERISTIC_RESERVED;
+ break;
+ case AVCOL_TRC_GAMMA22:
+ transfer_characteristic = AMF_COLOR_TRANSFER_CHARACTERISTIC_GAMMA22;
+ break;
+ case AVCOL_TRC_GAMMA28:
+ transfer_characteristic = AMF_COLOR_TRANSFER_CHARACTERISTIC_GAMMA28;
+ break;
+ case AVCOL_TRC_SMPTE170M:
+ transfer_characteristic = AMF_COLOR_TRANSFER_CHARACTERISTIC_SMPTE170M;
+ break;
+ case AVCOL_TRC_SMPTE240M:
+ transfer_characteristic = AMF_COLOR_TRANSFER_CHARACTERISTIC_SMPTE240M;
+ break;
+ case AVCOL_TRC_LINEAR:
+ transfer_characteristic = AMF_COLOR_TRANSFER_CHARACTERISTIC_LINEAR;
+ break;
+ case AVCOL_TRC_LOG:
+ transfer_characteristic = AMF_COLOR_TRANSFER_CHARACTERISTIC_LOG;
+ break;
+ case AVCOL_TRC_LOG_SQRT:
+ transfer_characteristic = AMF_COLOR_TRANSFER_CHARACTERISTIC_LOG_SQRT;
+ break;
+ case AVCOL_TRC_IEC61966_2_4:
+ transfer_characteristic = AMF_COLOR_TRANSFER_CHARACTERISTIC_IEC61966_2_4;
+ break;
+ case AVCOL_TRC_BT1361_ECG:
+ transfer_characteristic = AMF_COLOR_TRANSFER_CHARACTERISTIC_BT1361_ECG;
+ break;
+ case AVCOL_TRC_IEC61966_2_1:
+ transfer_characteristic = AMF_COLOR_TRANSFER_CHARACTERISTIC_IEC61966_2_1;
+ break;
+ case AVCOL_TRC_BT2020_10:
+ transfer_characteristic = AMF_COLOR_TRANSFER_CHARACTERISTIC_BT2020_10;
+ break;
+ case AVCOL_TRC_BT2020_12:
+ transfer_characteristic = AMF_COLOR_TRANSFER_CHARACTERISTIC_BT2020_12;
+ break;
+ case AVCOL_TRC_SMPTE2084:
+ transfer_characteristic = AMF_COLOR_TRANSFER_CHARACTERISTIC_SMPTE2084;
+ break;
+ case AVCOL_TRC_SMPTE428:
+ transfer_characteristic = AMF_COLOR_TRANSFER_CHARACTERISTIC_SMPTE428;
+ break;
+ case AVCOL_TRC_ARIB_STD_B67:
+ transfer_characteristic = AMF_COLOR_TRANSFER_CHARACTERISTIC_ARIB_STD_B67;
+ break;
+ }
+ return transfer_characteristic;
+}
+
int ff_amf_get_color_profile(AVCodecContext *avctx)
{
amf_int64 color_profile = AMF_VIDEO_CONVERTER_COLOR_PROFILE_UNKNOWN;
diff -Naur ffmpeg-7.1.2.old/libavcodec/amfenc.h ffmpeg-7.1.2/libavcodec/amfenc.h
--- ffmpeg-7.1.2.old/libavcodec/amfenc.h 2025-10-27 10:07:00.322471636 +0100
+++ ffmpeg-7.1.2/libavcodec/amfenc.h 2025-10-27 10:07:02.440421671 +0100
@@ -114,6 +114,7 @@
int max_b_frames;
int qvbr_quality_level;
int hw_high_motion_quality_boost;
+ int forced_idr;
// HEVC - specific options
@@ -173,6 +174,8 @@
*/
extern const enum AVPixelFormat ff_amf_pix_fmts[];
+int ff_amf_get_color_primaries(AVCodecContext *avctx);
+int ff_amf_get_transfer_characteristic(AVCodecContext *avctx);
int ff_amf_get_color_profile(AVCodecContext *avctx);
/**
diff -Naur ffmpeg-7.1.2.old/libavcodec/amfenc_h264.c ffmpeg-7.1.2/libavcodec/amfenc_h264.c
--- ffmpeg-7.1.2.old/libavcodec/amfenc_h264.c 2025-10-27 10:07:00.604474124 +0100
+++ ffmpeg-7.1.2/libavcodec/amfenc_h264.c 2025-10-27 10:07:02.440571423 +0100
@@ -133,6 +133,7 @@
{ "me_half_pel", "Enable ME Half Pixel", OFFSET(me_half_pel), AV_OPT_TYPE_BOOL, { .i64 = -1 }, -1, 1, VE },
{ "me_quarter_pel", "Enable ME Quarter Pixel", OFFSET(me_quarter_pel),AV_OPT_TYPE_BOOL, { .i64 = -1 }, -1, 1, VE },
+ { "forced_idr", "Force I frames to be IDR frames", OFFSET(forced_idr) , AV_OPT_TYPE_BOOL, { .i64 = 0 }, 0, 1, VE },
{ "aud", "Inserts AU Delimiter NAL unit", OFFSET(aud) , AV_OPT_TYPE_BOOL, { .i64 = -1 }, -1, 1, VE },
@@ -201,6 +202,8 @@
AMFRate framerate;
AMFSize framesize = AMFConstructSize(avctx->width, avctx->height);
int deblocking_filter = (avctx->flags & AV_CODEC_FLAG_LOOP_FILTER) ? 1 : 0;
+ amf_int64 color_primaries;
+ amf_int64 transfer_characteristic;
amf_int64 color_profile;
enum AVPixelFormat pix_fmt;
@@ -273,7 +276,11 @@
AMF_ASSIGN_PROPERTY_RATIO(res, ctx->encoder, AMF_VIDEO_ENCODER_ASPECT_RATIO, ratio);
}
+ color_primaries = ff_amf_get_color_primaries(avctx);
+ transfer_characteristic = ff_amf_get_transfer_characteristic(avctx);
color_profile = ff_amf_get_color_profile(avctx);
+ AMF_ASSIGN_PROPERTY_INT64(res, ctx->encoder, AMF_VIDEO_ENCODER_OUTPUT_COLOR_PRIMARIES, color_primaries);
+ AMF_ASSIGN_PROPERTY_INT64(res, ctx->encoder, AMF_VIDEO_ENCODER_OUTPUT_TRANSFER_CHARACTERISTIC, transfer_characteristic);
AMF_ASSIGN_PROPERTY_INT64(res, ctx->encoder, AMF_VIDEO_ENCODER_OUTPUT_COLOR_PROFILE, color_profile);
/// Color Range (Support for older Drivers)
@@ -287,10 +294,6 @@
AMF_RETURN_IF_FALSE(ctx, pix_fmt != AV_PIX_FMT_P010, AVERROR_INVALIDDATA, "10-bit input video is not supported by AMF H264 encoder\n");
AMF_ASSIGN_PROPERTY_INT64(res, ctx->encoder, AMF_VIDEO_ENCODER_COLOR_BIT_DEPTH, AMF_COLOR_BIT_DEPTH_8);
- /// Color Transfer Characteristics (AMF matches ISO/IEC)
- AMF_ASSIGN_PROPERTY_INT64(res, ctx->encoder, AMF_VIDEO_ENCODER_OUTPUT_TRANSFER_CHARACTERISTIC, (amf_int64)avctx->color_trc);
- /// Color Primaries (AMF matches ISO/IEC)
- AMF_ASSIGN_PROPERTY_INT64(res, ctx->encoder, AMF_VIDEO_ENCODER_OUTPUT_COLOR_PRIMARIES, (amf_int64)avctx->color_primaries);
// autodetect rate control method
if (ctx->rate_control_mode == AMF_VIDEO_ENCODER_RATE_CONTROL_METHOD_UNKNOWN) {
diff -Naur ffmpeg-7.1.2.old/libavcodec/amfenc_hevc.c ffmpeg-7.1.2/libavcodec/amfenc_hevc.c
--- ffmpeg-7.1.2.old/libavcodec/amfenc_hevc.c 2025-10-27 10:07:00.324471654 +0100
+++ ffmpeg-7.1.2/libavcodec/amfenc_hevc.c 2025-10-27 10:07:02.440657756 +0100
@@ -100,6 +100,7 @@
{ "me_half_pel", "Enable ME Half Pixel", OFFSET(me_half_pel), AV_OPT_TYPE_BOOL,{ .i64 = -1 }, -1, 1, VE },
{ "me_quarter_pel", "Enable ME Quarter Pixel ", OFFSET(me_quarter_pel),AV_OPT_TYPE_BOOL,{ .i64 = -1 }, -1, 1, VE },
+ { "forced_idr", "Force I frames to be IDR frames", OFFSET(forced_idr) ,AV_OPT_TYPE_BOOL,{ .i64 = 0 }, 0, 1, VE },
{ "aud", "Inserts AU Delimiter NAL unit", OFFSET(aud) ,AV_OPT_TYPE_BOOL,{ .i64 = -1 }, -1, 1, VE },
@@ -167,6 +168,8 @@
AMFSize framesize = AMFConstructSize(avctx->width, avctx->height);
int deblocking_filter = (avctx->flags & AV_CODEC_FLAG_LOOP_FILTER) ? 1 : 0;
amf_int64 color_depth;
+ amf_int64 color_primaries;
+ amf_int64 transfer_characteristic;
amf_int64 color_profile;
enum AVPixelFormat pix_fmt;
@@ -241,7 +244,11 @@
AMF_ASSIGN_PROPERTY_RATIO(res, ctx->encoder, AMF_VIDEO_ENCODER_HEVC_ASPECT_RATIO, ratio);
}
+ color_primaries = ff_amf_get_color_primaries(avctx);
+ transfer_characteristic = ff_amf_get_transfer_characteristic(avctx);
color_profile = ff_amf_get_color_profile(avctx);
+ AMF_ASSIGN_PROPERTY_INT64(res, ctx->encoder, AMF_VIDEO_ENCODER_HEVC_OUTPUT_COLOR_PRIMARIES, color_primaries);
+ AMF_ASSIGN_PROPERTY_INT64(res, ctx->encoder, AMF_VIDEO_ENCODER_HEVC_OUTPUT_TRANSFER_CHARACTERISTIC, transfer_characteristic);
AMF_ASSIGN_PROPERTY_INT64(res, ctx->encoder, AMF_VIDEO_ENCODER_HEVC_OUTPUT_COLOR_PROFILE, color_profile);
/// Color Range (Support for older Drivers)
AMF_ASSIGN_PROPERTY_BOOL(res, ctx->encoder, AMF_VIDEO_ENCODER_HEVC_NOMINAL_RANGE, !!(avctx->color_range == AVCOL_RANGE_JPEG));
@@ -253,15 +260,6 @@
color_depth = AMF_COLOR_BIT_DEPTH_10;
}
AMF_ASSIGN_PROPERTY_INT64(res, ctx->encoder, AMF_VIDEO_ENCODER_HEVC_COLOR_BIT_DEPTH, color_depth);
- if (color_depth == AMF_COLOR_BIT_DEPTH_8) {
- /// Color Transfer Characteristics (AMF matches ISO/IEC)
- AMF_ASSIGN_PROPERTY_INT64(res, ctx->encoder, AMF_VIDEO_ENCODER_HEVC_OUTPUT_TRANSFER_CHARACTERISTIC, AMF_COLOR_TRANSFER_CHARACTERISTIC_BT709);
- /// Color Primaries (AMF matches ISO/IEC)
- AMF_ASSIGN_PROPERTY_INT64(res, ctx->encoder, AMF_VIDEO_ENCODER_HEVC_OUTPUT_COLOR_PRIMARIES, AMF_COLOR_PRIMARIES_BT709);
- } else {
- AMF_ASSIGN_PROPERTY_INT64(res, ctx->encoder, AMF_VIDEO_ENCODER_HEVC_OUTPUT_TRANSFER_CHARACTERISTIC, AMF_COLOR_TRANSFER_CHARACTERISTIC_SMPTE2084);
- AMF_ASSIGN_PROPERTY_INT64(res, ctx->encoder, AMF_VIDEO_ENCODER_HEVC_OUTPUT_COLOR_PRIMARIES, AMF_COLOR_PRIMARIES_BT2020);
- }
// Picture control properties
AMF_ASSIGN_PROPERTY_INT64(res, ctx->encoder, AMF_VIDEO_ENCODER_HEVC_NUM_GOPS_PER_IDR, ctx->gops_per_idr);
diff -Naur ffmpeg-7.1.2.old/libavcodec/av1dec.c ffmpeg-7.1.2/libavcodec/av1dec.c
--- ffmpeg-7.1.2.old/libavcodec/av1dec.c 2025-10-27 10:07:00.478473013 +0100
+++ ffmpeg-7.1.2/libavcodec/av1dec.c 2025-10-27 10:07:02.454638605 +0100
@@ -541,6 +541,7 @@
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 @@
#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 @@
#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
@@ -1002,6 +1009,8 @@
break;
}
case ITU_T_T35_PROVIDER_CODE_DOLBY: {
+ AVBufferRef *rpu_buf;
+ AVFrameSideData *rpu;
int provider_oriented_code = bytestream2_get_be32(&gb);
if (itut_t35->itu_t_t35_country_code != ITU_T_T35_COUNTRY_CODE_US ||
provider_oriented_code != 0x800)
@@ -1014,6 +1023,18 @@
break; // ignore
}
+ rpu_buf = av_buffer_alloc(itut_t35->payload_size);
+ if (rpu_buf) {
+ memcpy(rpu_buf->data, itut_t35->payload, itut_t35->payload_size);
+ rpu = av_frame_new_side_data_from_buf(frame, AV_FRAME_DATA_DOVI_RPU_BUFFER_T35, rpu_buf);
+ if (!rpu) {
+ av_buffer_unref(&rpu_buf);
+ return AVERROR(ENOMEM);
+ }
+ } else {
+ return AVERROR(ENOMEM);
+ }
+
ret = ff_dovi_attach_side_data(&s->dovi, frame);
if (ret < 0)
return ret;
@@ -1439,6 +1460,10 @@
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 +1474,8 @@
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 +1505,7 @@
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 +1532,7 @@
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 +1553,7 @@
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 +1621,9 @@
#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
diff -Naur ffmpeg-7.1.2.old/libavcodec/av1dec.h ffmpeg-7.1.2/libavcodec/av1dec.h
--- ffmpeg-7.1.2.old/libavcodec/av1dec.h 2025-10-27 10:07:00.402472342 +0100
+++ ffmpeg-7.1.2/libavcodec/av1dec.h 2025-10-27 10:07:02.454762738 +0100
@@ -114,7 +114,8 @@
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;
diff -Naur ffmpeg-7.1.2.old/libavcodec/bsf/hevc_mp4toannexb.c ffmpeg-7.1.2/libavcodec/bsf/hevc_mp4toannexb.c
--- ffmpeg-7.1.2.old/libavcodec/bsf/hevc_mp4toannexb.c 2025-10-27 10:07:00.449472757 +0100
+++ ffmpeg-7.1.2/libavcodec/bsf/hevc_mp4toannexb.c 2025-10-27 10:07:02.435022530 +0100
@@ -125,6 +125,7 @@
AVPacket *in;
GetByteContext gb;
+ int has_sps = 0, has_pps = 0;
int got_irap = 0;
int i, ret = 0;
@@ -158,11 +159,14 @@
}
nalu_type = (bytestream2_peek_byte(&gb) >> 1) & 0x3f;
+ has_sps = (has_sps || nalu_type == HEVC_NAL_SPS);
+ has_pps = (has_pps || nalu_type == HEVC_NAL_PPS);
/* prepend extradata to IRAP frames */
is_irap = nalu_type >= HEVC_NAL_BLA_W_LP &&
nalu_type <= HEVC_NAL_RSV_IRAP_VCL23;
- add_extradata = is_irap && !got_irap;
+ /* ignore the extradata if IRAP frame has sps and pps */
+ add_extradata = is_irap && !got_irap && !(has_sps && has_pps);
extra_size = add_extradata * ctx->par_out->extradata_size;
got_irap |= is_irap;
diff -Naur ffmpeg-7.1.2.old/libavcodec/ccaption_dec.c ffmpeg-7.1.2/libavcodec/ccaption_dec.c
--- ffmpeg-7.1.2.old/libavcodec/ccaption_dec.c 2025-10-27 10:07:00.593474027 +0100
+++ ffmpeg-7.1.2/libavcodec/ccaption_dec.c 2025-10-27 10:07:02.429490227 +0100
@@ -889,12 +889,13 @@
if (ctx->buffer[bidx].str[0] || ctx->real_time) {
ff_dlog(ctx, "cdp writing data (%s)\n", ctx->buffer[bidx].str);
- start_time = ctx->buffer_time[0];
- sub->pts = start_time;
- end_time = ctx->buffer_time[1];
- if (!ctx->real_time)
+ if (!ctx->real_time) {
+ start_time = ctx->buffer_time[0];
+ sub->pts = start_time;
+ end_time = ctx->buffer_time[1];
sub->end_display_time = av_rescale_q(end_time - start_time,
AV_TIME_BASE_Q, ms_tb);
+ }
else
sub->end_display_time = -1;
ret = ff_ass_add_rect2(sub, ctx->buffer[bidx].str, ctx->readorder++, 0, NULL, NULL, &nb_rect_allocated);
diff -Naur ffmpeg-7.1.2.old/libavcodec/dvdsubdec.c ffmpeg-7.1.2/libavcodec/dvdsubdec.c
--- ffmpeg-7.1.2.old/libavcodec/dvdsubdec.c 2025-10-27 10:07:00.303471468 +0100
+++ ffmpeg-7.1.2/libavcodec/dvdsubdec.c 2025-10-27 10:07:02.428490218 +0100
@@ -45,6 +45,8 @@
int buf_size;
int forced_subs_only;
uint8_t used_color[256];
+ int64_t pts;
+ int output_empty_rects;
} DVDSubContext;
static void yuv_a_to_rgba(const uint8_t *ycbcr, const uint8_t *alpha, uint32_t *rgba, int num_values)
@@ -230,7 +232,10 @@
uint32_t size;
int64_t offset1, offset2;
- if (buf_size < 10)
+ if (buf_size < 2)
+ return AVERROR(EAGAIN);
+
+ if (buf_size == 2 && AV_RB16(buf) == 0)
return -1;
if (AV_RB16(buf) == 0) { /* HD subpicture with 4-byte offsets */
@@ -243,15 +248,22 @@
cmd_pos = 2;
}
+ if (big_offsets && buf_size < 6)
+ return AVERROR(EAGAIN);
+
size = READ_OFFSET(buf + (big_offsets ? 2 : 0));
- cmd_pos = READ_OFFSET(buf + cmd_pos);
- if (cmd_pos < 0 || cmd_pos > buf_size - 2 - offset_size) {
- if (cmd_pos > size) {
- av_log(ctx, AV_LOG_ERROR, "Discarding invalid packet\n");
- return 0;
- }
+ if (size == 0)
+ return -1;
+
+ if (buf_size < size)
return AVERROR(EAGAIN);
+
+ cmd_pos = READ_OFFSET(buf + cmd_pos);
+
+ if (cmd_pos < 0 || cmd_pos > size) {
+ av_log(ctx, AV_LOG_ERROR, "Discarding invalid packet\n");
+ return AVERROR_INVALIDDATA;
}
while (cmd_pos > 0 && cmd_pos < buf_size - 2 - offset_size) {
@@ -524,10 +536,13 @@
int appended = 0;
int is_menu;
+ if (ctx->pts == AV_NOPTS_VALUE && sub->pts != AV_NOPTS_VALUE)
+ ctx->pts = sub->pts;
if (ctx->buf_size) {
int ret = append_to_cached_buf(avctx, buf, buf_size);
if (ret < 0) {
*data_size = 0;
+ ctx->pts = AV_NOPTS_VALUE;
return ret;
}
buf = ctx->buf;
@@ -538,7 +553,12 @@
is_menu = decode_dvd_subtitles(ctx, sub, buf, buf_size);
if (is_menu == AVERROR(EAGAIN)) {
*data_size = 0;
- return appended ? 0 : append_to_cached_buf(avctx, buf, buf_size);
+ int ret = appended ? 0 : append_to_cached_buf(avctx, buf, buf_size);
+ if (ret < 0) {
+ ctx->pts = AV_NOPTS_VALUE;
+ return ret;
+ }
+ return buf_size;
}
if (is_menu < 0) {
@@ -547,9 +567,10 @@
reset_rects(sub);
*data_size = 0;
+ ctx->pts = AV_NOPTS_VALUE;
return buf_size;
}
- if (!is_menu && find_smallest_bounding_rectangle(ctx, sub) == 0)
+ if (!is_menu && !ctx->output_empty_rects && find_smallest_bounding_rectangle(ctx, sub) == 0)
goto no_subtitle;
if (ctx->forced_subs_only && !(sub->rects[0]->flags & AV_SUBTITLE_FLAG_FORCED))
@@ -557,6 +578,8 @@
ctx->buf_size = 0;
*data_size = 1;
+ sub->pts = ctx->pts;
+ ctx->pts = AV_NOPTS_VALUE;
return buf_size;
}
@@ -682,6 +705,7 @@
av_log(avctx, AV_LOG_DEBUG, " 0x%06"PRIx32, ctx->palette[i]);
av_log(avctx, AV_LOG_DEBUG, "\n");
}
+ ctx->pts = AV_NOPTS_VALUE;
return 1;
}
@@ -698,6 +722,7 @@
{ "palette", "set the global palette", OFFSET(palette_str), AV_OPT_TYPE_STRING, { .str = NULL }, 0, 0, SD },
{ "ifo_palette", "obtain the global palette from .IFO file", OFFSET(ifo_str), AV_OPT_TYPE_STRING, { .str = NULL }, 0, 0, SD },
{ "forced_subs_only", "Only show forced subtitles", OFFSET(forced_subs_only), AV_OPT_TYPE_BOOL, {.i64 = 0}, 0, 1, SD},
+ { "output_empty_rects", "Output subtitles with empty or fully transparent rects", OFFSET(output_empty_rects), AV_OPT_TYPE_BOOL, {.i64 = 0}, 0, 1, SD},
{ NULL }
};
static const AVClass dvdsub_class = {
diff -Naur ffmpeg-7.1.2.old/libavcodec/eac3dec.c ffmpeg-7.1.2/libavcodec/eac3dec.c
--- ffmpeg-7.1.2.old/libavcodec/eac3dec.c 2025-10-27 10:07:00.490473118 +0100
+++ ffmpeg-7.1.2/libavcodec/eac3dec.c 2025-10-27 10:07:02.463392296 +0100
@@ -53,8 +53,6 @@
EAC3_GAQ_124
} EAC3GaqMode;
-#define EAC3_SR_CODE_REDUCED 3
-
static void ff_eac3_apply_spectral_extension(AC3DecodeContext *s)
{
int bin, bnd, ch, i;
@@ -287,7 +285,7 @@
}
}
-static int ff_eac3_parse_header(AC3DecodeContext *s)
+static int ff_eac3_parse_header(AC3DecodeContext *s, const AC3HeaderInfo *hdr)
{
int i, blk, ch;
int ac3_exponent_strategy, parse_aht_info, parse_spx_atten_data;
@@ -323,11 +321,10 @@
avpriv_request_sample(s->avctx, "Reduced sampling rate");
return AVERROR_PATCHWELCOME;
}
- skip_bits(gbc, 5); // skip bitstream id
/* volume control params */
for (i = 0; i < (s->channel_mode ? 1 : 2); i++) {
- s->dialog_normalization[i] = -get_bits(gbc, 5);
+ s->dialog_normalization[i] = hdr->dialog_normalization[i];
if (s->dialog_normalization[i] == 0) {
s->dialog_normalization[i] = -31;
}
@@ -335,147 +332,30 @@
s->level_gain[i] = powf(2.0f,
(float)(s->target_level - s->dialog_normalization[i])/6.0f);
}
- s->compression_exists[i] = get_bits1(gbc);
- if (s->compression_exists[i]) {
- s->heavy_dynamic_range[i] = AC3_HEAVY_RANGE(get_bits(gbc, 8));
+ if (hdr->compression_exists[i]) {
+ s->heavy_dynamic_range[i] = AC3_HEAVY_RANGE(hdr->heavy_dynamic_range[i]);
}
}
- /* dependent stream channel map */
- if (s->frame_type == EAC3_FRAME_TYPE_DEPENDENT) {
- if (get_bits1(gbc)) {
- int64_t channel_layout = 0;
- int channel_map = get_bits(gbc, 16);
- av_log(s->avctx, AV_LOG_DEBUG, "channel_map: %0X\n", channel_map);
-
- for (i = 0; i < 16; i++)
- if (channel_map & (1 << (EAC3_MAX_CHANNELS - i - 1)))
- channel_layout |= ff_eac3_custom_channel_map_locations[i][1];
-
- if (av_popcount64(channel_layout) > EAC3_MAX_CHANNELS) {
- return AVERROR_INVALIDDATA;
- }
- s->channel_map = channel_map;
- }
- }
+ s->channel_map = hdr->channel_map;
/* mixing metadata */
- if (get_bits1(gbc)) {
- /* center and surround mix levels */
- if (s->channel_mode > AC3_CHMODE_STEREO) {
- s->preferred_downmix = get_bits(gbc, 2);
- if (s->channel_mode & 1) {
- /* if three front channels exist */
- s->center_mix_level_ltrt = get_bits(gbc, 3);
- s->center_mix_level = get_bits(gbc, 3);
- }
- if (s->channel_mode & 4) {
- /* if a surround channel exists */
- s->surround_mix_level_ltrt = av_clip(get_bits(gbc, 3), 3, 7);
- s->surround_mix_level = av_clip(get_bits(gbc, 3), 3, 7);
- }
- }
-
- /* lfe mix level */
- if (s->lfe_on && (s->lfe_mix_level_exists = get_bits1(gbc))) {
- s->lfe_mix_level = get_bits(gbc, 5);
- }
-
- /* info for mixing with other streams and substreams */
- if (s->frame_type == EAC3_FRAME_TYPE_INDEPENDENT) {
- for (i = 0; i < (s->channel_mode ? 1 : 2); i++) {
- // TODO: apply program scale factor
- if (get_bits1(gbc)) {
- skip_bits(gbc, 6); // skip program scale factor
- }
- }
- if (get_bits1(gbc)) {
- skip_bits(gbc, 6); // skip external program scale factor
- }
- /* skip mixing parameter data */
- switch(get_bits(gbc, 2)) {
- case 1: skip_bits(gbc, 5); break;
- case 2: skip_bits(gbc, 12); break;
- case 3: {
- int mix_data_size = (get_bits(gbc, 5) + 2) << 3;
- skip_bits_long(gbc, mix_data_size);
- break;
- }
- }
- /* skip pan information for mono or dual mono source */
- if (s->channel_mode < AC3_CHMODE_STEREO) {
- for (i = 0; i < (s->channel_mode ? 1 : 2); i++) {
- if (get_bits1(gbc)) {
- /* note: this is not in the ATSC A/52B specification
- reference: ETSI TS 102 366 V1.1.1
- section: E.1.3.1.25 */
- skip_bits(gbc, 8); // skip pan mean direction index
- skip_bits(gbc, 6); // skip reserved paninfo bits
- }
- }
- }
- /* skip mixing configuration information */
- if (get_bits1(gbc)) {
- for (blk = 0; blk < s->num_blocks; blk++) {
- if (s->num_blocks == 1 || get_bits1(gbc)) {
- skip_bits(gbc, 5);
- }
- }
- }
- }
- }
+ s->preferred_downmix = hdr->preferred_downmix;
+ s->center_mix_level_ltrt = hdr->center_mix_level_ltrt;
+ s->center_mix_level = hdr->center_mix_level;
+ s->surround_mix_level_ltrt = hdr->surround_mix_level_ltrt;
+ s->surround_mix_level = hdr->surround_mix_level;
+ s->lfe_mix_level_exists = hdr->lfe_mix_level_exists;
+ s->lfe_mix_level = hdr->lfe_mix_level;
+ s->dolby_surround_mode = hdr->dolby_surround_mode;
+ s->dolby_headphone_mode = hdr->dolby_headphone_mode;
+ s->dolby_surround_ex_mode = hdr->dolby_surround_ex_mode;
/* informational metadata */
- if (get_bits1(gbc)) {
- s->bitstream_mode = get_bits(gbc, 3);
- skip_bits(gbc, 2); // skip copyright bit and original bitstream bit
- if (s->channel_mode == AC3_CHMODE_STEREO) {
- s->dolby_surround_mode = get_bits(gbc, 2);
- s->dolby_headphone_mode = get_bits(gbc, 2);
- }
- if (s->channel_mode >= AC3_CHMODE_2F2R) {
- s->dolby_surround_ex_mode = get_bits(gbc, 2);
- }
- for (i = 0; i < (s->channel_mode ? 1 : 2); i++) {
- if (get_bits1(gbc)) {
- skip_bits(gbc, 8); // skip mix level, room type, and A/D converter type
- }
- }
- if (s->bit_alloc_params.sr_code != EAC3_SR_CODE_REDUCED) {
- skip_bits1(gbc); // skip source sample rate code
- }
- }
-
- /* converter synchronization flag
- If frames are less than six blocks, this bit should be turned on
- once every 6 blocks to indicate the start of a frame set.
- reference: RFC 4598, Section 2.1.3 Frame Sets */
- if (s->frame_type == EAC3_FRAME_TYPE_INDEPENDENT && s->num_blocks != 6) {
- skip_bits1(gbc); // skip converter synchronization flag
- }
-
- /* original frame size code if this stream was converted from AC-3 */
- if (s->frame_type == EAC3_FRAME_TYPE_AC3_CONVERT &&
- (s->num_blocks == 6 || get_bits1(gbc))) {
- skip_bits(gbc, 6); // skip frame size code
- }
+ s->bitstream_mode = hdr->bitstream_mode;
/* additional bitstream info */
- if (get_bits1(gbc)) {
- int addbsil = get_bits(gbc, 6);
- for (i = 0; i < addbsil + 1; i++) {
- if (i == 0) {
- /* In this 8 bit chunk, the LSB is equal to flag_ec3_extension_type_a
- which can be used to detect Atmos presence */
- skip_bits(gbc, 7);
- if (get_bits1(gbc)) {
- s->eac3_extension_type_a = 1;
- }
- } else {
- skip_bits(gbc, 8); // skip additional bit stream info
- }
- }
- }
+ s->eac3_extension_type_a = hdr->eac3_extension_type_a;
/* audio frame syntax flags, strategy data, and per-frame data */
diff -Naur ffmpeg-7.1.2.old/libavcodec/hwaccels.h ffmpeg-7.1.2/libavcodec/hwaccels.h
--- ffmpeg-7.1.2.old/libavcodec/hwaccels.h 2025-10-27 10:07:00.251471009 +0100
+++ ffmpeg-7.1.2/libavcodec/hwaccels.h 2025-10-27 10:07:02.454807062 +0100
@@ -26,6 +26,7 @@
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;
diff -Naur ffmpeg-7.1.2.old/libavcodec/libdav1d.c ffmpeg-7.1.2/libavcodec/libdav1d.c
--- ffmpeg-7.1.2.old/libavcodec/libdav1d.c 2025-10-27 10:07:00.474472977 +0100
+++ ffmpeg-7.1.2/libavcodec/libdav1d.c 2025-10-27 10:07:02.436286040 +0100
@@ -563,6 +563,8 @@
break;
}
case ITU_T_T35_PROVIDER_CODE_DOLBY: {
+ AVBufferRef *rpu_buf;
+ AVFrameSideData *rpu;
int provider_oriented_code = bytestream2_get_be32(&gb);
if (itut_t35->country_code != ITU_T_T35_COUNTRY_CODE_US ||
provider_oriented_code != 0x800)
@@ -575,6 +577,18 @@
break; // ignore
}
+ rpu_buf = av_buffer_alloc(itut_t35->payload_size);
+ if (rpu_buf) {
+ memcpy(rpu_buf->data, itut_t35->payload, itut_t35->payload_size);
+ rpu = av_frame_new_side_data_from_buf(frame, AV_FRAME_DATA_DOVI_RPU_BUFFER_T35, rpu_buf);
+ if (!rpu) {
+ av_buffer_unref(&rpu_buf);
+ goto fail;
+ }
+ } else {
+ goto fail;
+ }
+
res = ff_dovi_attach_side_data(&dav1d->dovi, frame);
if (res < 0)
goto fail;
diff -Naur ffmpeg-7.1.2.old/libavcodec/Makefile ffmpeg-7.1.2/libavcodec/Makefile
--- ffmpeg-7.1.2.old/libavcodec/Makefile 2025-10-27 10:07:01.865180885 +0100
+++ ffmpeg-7.1.2/libavcodec/Makefile 2025-10-27 10:07:02.454516424 +0100
@@ -1008,6 +1008,7 @@
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
diff -Naur ffmpeg-7.1.2.old/libavcodec/mfenc.c ffmpeg-7.1.2/libavcodec/mfenc.c
--- ffmpeg-7.1.2.old/libavcodec/mfenc.c 2025-10-27 10:07:00.524473418 +0100
+++ ffmpeg-7.1.2/libavcodec/mfenc.c 2025-10-27 10:07:02.459027162 +0100
@@ -1315,3 +1315,4 @@
MF_ENCODER(VIDEO, h264, H264, venc_opts, VFMTS, VCAPS, defaults);
MF_ENCODER(VIDEO, hevc, HEVC, venc_opts, VFMTS, VCAPS, defaults);
+MF_ENCODER(VIDEO, av1, AV1, venc_opts, VFMTS, VCAPS, defaults);
diff -Naur ffmpeg-7.1.2.old/libavcodec/mf_utils.c ffmpeg-7.1.2/libavcodec/mf_utils.c
--- ffmpeg-7.1.2.old/libavcodec/mf_utils.c 2025-10-27 10:07:00.467472915 +0100
+++ ffmpeg-7.1.2/libavcodec/mf_utils.c 2025-10-27 10:07:02.458887429 +0100
@@ -240,6 +240,7 @@
GUID_ENTRY(MFMediaType_Video),
GUID_ENTRY(MFAudioFormat_PCM),
GUID_ENTRY(MFAudioFormat_Float),
+ GUID_ENTRY(ff_MFVideoFormat_AV1),
GUID_ENTRY(MFVideoFormat_H264),
GUID_ENTRY(MFVideoFormat_H264_ES),
GUID_ENTRY(ff_MFVideoFormat_HEVC),
@@ -507,6 +508,7 @@
const CLSID *ff_codec_to_mf_subtype(enum AVCodecID codec)
{
switch (codec) {
+ case AV_CODEC_ID_AV1: return &ff_MFVideoFormat_AV1;
case AV_CODEC_ID_H264: return &MFVideoFormat_H264;
case AV_CODEC_ID_HEVC: return &ff_MFVideoFormat_HEVC;
case AV_CODEC_ID_AC3: return &MFAudioFormat_Dolby_AC3;
diff -Naur ffmpeg-7.1.2.old/libavcodec/mf_utils.h ffmpeg-7.1.2/libavcodec/mf_utils.h
--- ffmpeg-7.1.2.old/libavcodec/mf_utils.h 2025-10-27 10:07:00.429472580 +0100
+++ ffmpeg-7.1.2/libavcodec/mf_utils.h 2025-10-27 10:07:02.458955668 +0100
@@ -113,6 +113,7 @@
DEFINE_MEDIATYPE_GUID(ff_MFVideoFormat_HEVC, 0x43564548); // FCC('HEVC')
DEFINE_MEDIATYPE_GUID(ff_MFVideoFormat_HEVC_ES, 0x53564548); // FCC('HEVS')
+DEFINE_MEDIATYPE_GUID(ff_MFVideoFormat_AV1, 0x31305641); // FCC('AV01')
// This enum is missing from mingw-w64's codecapi.h by v7.0.0.
diff -Naur ffmpeg-7.1.2.old/libavcodec/pgssubdec.c ffmpeg-7.1.2/libavcodec/pgssubdec.c
--- ffmpeg-7.1.2.old/libavcodec/pgssubdec.c 2025-10-27 10:07:00.318471601 +0100
+++ ffmpeg-7.1.2/libavcodec/pgssubdec.c 2025-10-27 10:07:02.446191090 +0100
@@ -35,9 +35,11 @@
#include "libavutil/opt.h"
#define RGBA(r,g,b,a) (((unsigned)(a) << 24) | ((r) << 16) | ((g) << 8) | (b))
-#define MAX_EPOCH_PALETTES 8 // Max 8 allowed per PGS epoch
-#define MAX_EPOCH_OBJECTS 64 // Max 64 allowed per PGS epoch
-#define MAX_OBJECT_REFS 2 // Max objects per display set
+#define MAX_EPOCH_PALETTES 8 // Max 8 allowed per PGS epoch
+#define MAX_EPOCH_OBJECTS 64 // Max 64 allowed per PGS epoch
+#define MAX_OBJECT_REFS 2 // Max objects per display set
+#define MAX_OBJECT_WH 4096 // Max object width/height
+
enum SegmentType {
PALETTE_SEGMENT = 0x14,
@@ -48,57 +50,80 @@
};
typedef struct PGSSubObjectRef {
- int id;
- int window_id;
- uint8_t composition_flag;
- int x;
- int y;
- int crop_x;
- int crop_y;
- int crop_w;
- int crop_h;
+ uint16_t id;
+ uint8_t window_id;
+ uint8_t composition_flag;
+ uint16_t x;
+ uint16_t y;
+ uint16_t crop_x;
+ uint16_t crop_y;
+ uint16_t crop_w;
+ uint16_t crop_h;
} PGSSubObjectRef;
typedef struct PGSSubPresentation {
- int id_number;
- int palette_id;
- int object_count;
+ uint8_t palette_flag;
+ uint8_t palette_id;
+ uint8_t object_count;
PGSSubObjectRef objects[MAX_OBJECT_REFS];
- int64_t pts;
+ int64_t pts;
} PGSSubPresentation;
typedef struct PGSSubObject {
- int id;
- int w;
- int h;
- uint8_t *rle;
- unsigned int rle_buffer_size, rle_data_len;
- unsigned int rle_remaining_len;
+ uint16_t id;
+ uint16_t w;
+ uint16_t h;
+ uint8_t *rle;
+ uint8_t *bitmap;
+ uint32_t rle_buffer_size;
+ uint32_t rle_data_len;
+ uint32_t rle_remaining_len;
+ uint32_t bitmap_buffer_size;
+ uint32_t bitmap_size;
} PGSSubObject;
typedef struct PGSSubObjects {
- int count;
+ uint8_t count;
PGSSubObject object[MAX_EPOCH_OBJECTS];
} PGSSubObjects;
typedef struct PGSSubPalette {
- int id;
- uint32_t clut[256];
+ uint8_t id;
+ uint32_t clut[AVPALETTE_COUNT];
} PGSSubPalette;
typedef struct PGSSubPalettes {
- int count;
+ uint8_t count;
PGSSubPalette palette[MAX_EPOCH_PALETTES];
} PGSSubPalettes;
+typedef struct PGSGraphicPlane {
+ uint8_t count;
+ uint8_t writable;
+ AVSubtitleRect visible_rect[MAX_OBJECT_REFS];
+} PGSGraphicPlane;
+
typedef struct PGSSubContext {
AVClass *class;
PGSSubPresentation presentation;
PGSSubPalettes palettes;
PGSSubObjects objects;
+ PGSGraphicPlane plane;
int forced_subs_only;
} PGSSubContext;
+static void clear_graphic_plane(PGSSubContext *ctx)
+{
+ int i;
+
+ for (i = 0; i < ctx->plane.count; i++) {
+ av_freep(&ctx->plane.visible_rect[i].data[0]);
+ memset(&ctx->plane.visible_rect[i], 0, sizeof(ctx->plane.visible_rect[i]));
+ }
+ ctx->plane.writable = 0;
+ ctx->plane.count = 0;
+}
+
static void flush_cache(AVCodecContext *avctx)
{
PGSSubContext *ctx = avctx->priv_data;
@@ -106,8 +131,11 @@
for (i = 0; i < ctx->objects.count; i++) {
av_freep(&ctx->objects.object[i].rle);
- ctx->objects.object[i].rle_buffer_size = 0;
+ ctx->objects.object[i].rle_buffer_size = 0;
ctx->objects.object[i].rle_remaining_len = 0;
+ av_freep(&ctx->objects.object[i].bitmap);
+ ctx->objects.object[i].bitmap_buffer_size = 0;
+ ctx->objects.object[i].bitmap_size = 0;
}
ctx->objects.count = 0;
ctx->palettes.count = 0;
@@ -144,6 +172,7 @@
static av_cold int close_decoder(AVCodecContext *avctx)
{
+ clear_graphic_plane((PGSSubContext *)avctx->priv_data);
flush_cache(avctx);
return 0;
@@ -159,48 +188,51 @@
* @param buf pointer to the RLE data to process
* @param buf_size size of the RLE data to process
*/
-static int decode_rle(AVCodecContext *avctx, AVSubtitleRect *rect,
- const uint8_t *buf, unsigned int buf_size)
+static int decode_object_rle(AVCodecContext *avctx, PGSSubObject *object)
{
- const uint8_t *rle_bitmap_end;
+ const uint8_t *rle_buf;
+ const uint8_t *rle_end;
int pixel_count, line_count;
+ rle_buf = object->rle;
+ rle_end = object->rle + object->rle_data_len;
- rle_bitmap_end = buf + buf_size;
+ object->bitmap_size = object->w * object->h;
+ av_fast_padded_malloc(&object->bitmap, &object->bitmap_buffer_size,
+ object->bitmap_size);
- rect->data[0] = av_malloc_array(rect->w, rect->h);
-
- if (!rect->data[0])
+ if (!object->bitmap)
return AVERROR(ENOMEM);
pixel_count = 0;
line_count = 0;
- while (buf < rle_bitmap_end && line_count < rect->h) {
+ while (rle_buf < rle_end && line_count < object->h) {
uint8_t flags, color;
int run;
- color = bytestream_get_byte(&buf);
+ color = bytestream_get_byte(&rle_buf);
run = 1;
if (color == 0x00) {
- flags = bytestream_get_byte(&buf);
+ flags = bytestream_get_byte(&rle_buf);
run = flags & 0x3f;
if (flags & 0x40)
- run = (run << 8) + bytestream_get_byte(&buf);
- color = flags & 0x80 ? bytestream_get_byte(&buf) : 0;
+ run = (run << 8) + bytestream_get_byte(&rle_buf);
+ color = flags & 0x80 ? bytestream_get_byte(&rle_buf) : 0;
}
- if (run > 0 && pixel_count + run <= rect->w * rect->h) {
- memset(rect->data[0] + pixel_count, color, run);
+ if (run > 0 && pixel_count + run <= object->w * object->h) {
+ memset(object->bitmap + pixel_count, color, run);
pixel_count += run;
} else if (!run) {
/*
* New Line. Check if correct pixels decoded, if not display warning
* and adjust bitmap pointer to correct new line position.
*/
- if (pixel_count % rect->w > 0) {
- av_log(avctx, AV_LOG_ERROR, "Decoded %d pixels, when line should be %d pixels\n",
- pixel_count % rect->w, rect->w);
+ if (pixel_count % object->w > 0) {
+ av_log(avctx, AV_LOG_ERROR,
+ "Decoded %d pixels, when object line should be %d pixels\n",
+ pixel_count % object->w, object->w);
if (avctx->err_recognition & AV_EF_EXPLODE) {
return AVERROR_INVALIDDATA;
}
@@ -209,13 +241,11 @@
}
}
- if (pixel_count < rect->w * rect->h) {
- av_log(avctx, AV_LOG_ERROR, "Insufficient RLE data for subtitle\n");
+ if (pixel_count < object->w * object->h) {
+ av_log(avctx, AV_LOG_ERROR, "Insufficient RLE data for object\n");
return AVERROR_INVALIDDATA;
}
-
- ff_dlog(avctx, "Pixel Count = %d, Area = %d\n", pixel_count, rect->w * rect->h);
-
+ ff_dlog(avctx, "Pixel Count = %d, Area = %d\n", pixel_count, object->w * object->h);
return 0;
}
@@ -237,7 +267,7 @@
uint8_t sequence_desc;
unsigned int rle_bitmap_len, width, height;
- int id;
+ int id, ret;
if (buf_size <= 4)
return AVERROR_INVALIDDATA;
@@ -260,57 +290,71 @@
/* Read the Sequence Description to determine if start of RLE data or appended to previous RLE */
sequence_desc = bytestream_get_byte(&buf);
- if (!(sequence_desc & 0x80)) {
- /* Additional RLE data */
- if (buf_size > object->rle_remaining_len)
+ /* First in sequence object definition segment */
+ if (sequence_desc & 0x80) {
+ if (buf_size <= 7)
return AVERROR_INVALIDDATA;
+ buf_size -= 7;
- memcpy(object->rle + object->rle_data_len, buf, buf_size);
- object->rle_data_len += buf_size;
- object->rle_remaining_len -= buf_size;
-
- return 0;
- }
-
- if (buf_size <= 7)
- return AVERROR_INVALIDDATA;
- buf_size -= 7;
+ /* Decode rle bitmap length, stored size includes width/height data */
+ rle_bitmap_len = bytestream_get_be24(&buf) - 2*2;
- /* Decode rle bitmap length, stored size includes width/height data */
- rle_bitmap_len = bytestream_get_be24(&buf) - 2*2;
+ if (buf_size > rle_bitmap_len) {
+ av_log(avctx, AV_LOG_ERROR,
+ "Buffer dimension %d larger than the expected RLE data %d\n",
+ buf_size, rle_bitmap_len);
+ return AVERROR_INVALIDDATA;
+ }
- if (buf_size > rle_bitmap_len) {
- av_log(avctx, AV_LOG_ERROR,
- "Buffer dimension %d larger than the expected RLE data %d\n",
- buf_size, rle_bitmap_len);
- return AVERROR_INVALIDDATA;
- }
+ /* Get bitmap dimensions from data */
+ width = bytestream_get_be16(&buf);
+ height = bytestream_get_be16(&buf);
+
+ /* Make sure the bitmap is not too large */
+ if (MAX_OBJECT_WH < width || MAX_OBJECT_WH < height || !width || !height) {
+ av_log(avctx, AV_LOG_ERROR, "Bitmap dimensions (%dx%d) invalid.\n", width, height);
+ return AVERROR_INVALIDDATA;
+ }
- /* Get bitmap dimensions from data */
- width = bytestream_get_be16(&buf);
- height = bytestream_get_be16(&buf);
-
- /* Make sure the bitmap is not too large */
- if (avctx->width < width || avctx->height < height || !width || !height) {
- av_log(avctx, AV_LOG_ERROR, "Bitmap dimensions (%dx%d) invalid.\n", width, height);
- return AVERROR_INVALIDDATA;
- }
+ object->rle_data_len = 0;
+ object->w = width;
+ object->h = height;
+ /* Dimensions against video are checked at decode after cropping. */
+ av_fast_padded_malloc(&object->rle, &object->rle_buffer_size, rle_bitmap_len);
- object->w = width;
- object->h = height;
+ if (!object->rle) {
+ object->rle_remaining_len = 0;
+ return AVERROR(ENOMEM);
+ }
- av_fast_padded_malloc(&object->rle, &object->rle_buffer_size, rle_bitmap_len);
+ memcpy(object->rle, buf, buf_size);
+ object->rle_remaining_len = rle_bitmap_len;
+ } else {
+ /* Additional RLE data */
+ if (buf_size > object->rle_remaining_len)
+ return AVERROR_INVALIDDATA;
- if (!object->rle) {
- object->rle_data_len = 0;
- object->rle_remaining_len = 0;
- return AVERROR(ENOMEM);
+ memcpy(object->rle + object->rle_data_len, buf, buf_size);
}
+ object->rle_data_len += buf_size;
+ object->rle_remaining_len -= buf_size;
- memcpy(object->rle, buf, buf_size);
- object->rle_data_len = buf_size;
- object->rle_remaining_len = rle_bitmap_len - buf_size;
-
+ /* Last in sequence object definition (can be both first and last) */
+ if (sequence_desc & 0x40) {
+ /* Attempt decoding if data is valid */
+ if (0 == object->rle_remaining_len) {
+ ret = decode_object_rle(avctx, object);
+ if (ret < 0 && (avctx->err_recognition & AV_EF_EXPLODE || ret == AVERROR(ENOMEM))) {
+ return ret;
+ }
+ } else {
+ av_log(avctx, AV_LOG_ERROR,
+ "RLE data length %u is %u bytes shorter than expected\n",
+ object->rle_data_len, object->rle_remaining_len);
+ if (avctx->err_recognition & AV_EF_EXPLODE)
+ return AVERROR_INVALIDDATA;
+ }
+ }
return 0;
}
@@ -318,7 +362,7 @@
* Parse the palette segment packet.
*
* The palette segment contains details of the palette,
- * a maximum of 256 colors can be defined.
+ * a maximum of 256 colors (AVPALETTE_COUNT) can be defined.
*
* @param avctx contains the current codec context
* @param buf pointer to the packet to process
@@ -391,13 +435,17 @@
int64_t pts)
{
PGSSubContext *ctx = avctx->priv_data;
- int i, state, ret;
+ int ret;
+ uint8_t i, state;
const uint8_t *buf_end = buf + buf_size;
// Video descriptor
int w = bytestream_get_be16(&buf);
int h = bytestream_get_be16(&buf);
+ // On a new display set, reset writability of the graphic plane
+ ctx->plane.writable = 0;
+
ctx->presentation.pts = pts;
ff_dlog(avctx, "Video Dimensions %dx%d\n",
@@ -406,88 +454,121 @@
if (ret < 0)
return ret;
- /* Skip 1 bytes of unknown, frame rate */
- buf++;
+ /* Skip 3 bytes: framerate (1), presentation id number (2) */
+ buf+=3;
- // Composition descriptor
- ctx->presentation.id_number = bytestream_get_be16(&buf);
/*
- * state is a 2 bit field that defines pgs epoch boundaries
+ * State is a 2 bit field that defines pgs epoch boundaries
* 00 - Normal, previously defined objects and palettes are still valid
* 01 - Acquisition point, previous objects and palettes can be released
* 10 - Epoch start, previous objects and palettes can be released
* 11 - Epoch continue, previous objects and palettes can be released
*
- * reserved 6 bits discarded
+ * Reserved 6 bits discarded
*/
state = bytestream_get_byte(&buf) >> 6;
if (state != 0) {
+ /* Epoch start always wipes the graphic plane. Epoch continue does only if
+ * playback is not seamless, which should not happen with a proper stream.
+ */
+ if (0b10 == state)
+ clear_graphic_plane((PGSSubContext *)avctx->priv_data);
flush_cache(avctx);
}
+ /* Reserved 7 bits discarded. */
+ ctx->presentation.palette_flag = bytestream_get_byte(&buf) & 0x80;
+ ctx->presentation.palette_id = bytestream_get_byte(&buf);
+
/*
- * skip palette_update_flag (0x80),
+ * On palette update, don't parse the compositions references,
+ * just evaluate the existing graphic plane with the new palette.
*/
- buf += 1;
- ctx->presentation.palette_id = bytestream_get_byte(&buf);
- ctx->presentation.object_count = bytestream_get_byte(&buf);
- if (ctx->presentation.object_count > MAX_OBJECT_REFS) {
- av_log(avctx, AV_LOG_ERROR,
- "Invalid number of presentation objects %d\n",
- ctx->presentation.object_count);
- ctx->presentation.object_count = 2;
- if (avctx->err_recognition & AV_EF_EXPLODE) {
- return AVERROR_INVALIDDATA;
+ if (!ctx->presentation.palette_flag) {
+ ctx->presentation.object_count = bytestream_get_byte(&buf);
+ if (ctx->presentation.object_count > MAX_OBJECT_REFS) {
+ av_log(avctx, AV_LOG_ERROR,
+ "Invalid number of presentation objects %d\n",
+ ctx->presentation.object_count);
+ ctx->presentation.object_count = 2;
+ if (avctx->err_recognition & AV_EF_EXPLODE) {
+ return AVERROR_INVALIDDATA;
+ }
}
- }
+ for (i = 0; i < ctx->presentation.object_count; i++) {
+ PGSSubObjectRef *const object = &ctx->presentation.objects[i];
- for (i = 0; i < ctx->presentation.object_count; i++)
- {
- PGSSubObjectRef *const object = &ctx->presentation.objects[i];
-
- if (buf_end - buf < 8) {
- av_log(avctx, AV_LOG_ERROR, "Insufficent space for object\n");
- ctx->presentation.object_count = i;
- return AVERROR_INVALIDDATA;
- }
-
- object->id = bytestream_get_be16(&buf);
- object->window_id = bytestream_get_byte(&buf);
- object->composition_flag = bytestream_get_byte(&buf);
-
- object->x = bytestream_get_be16(&buf);
- object->y = bytestream_get_be16(&buf);
-
- // If cropping
- if (object->composition_flag & 0x80) {
- object->crop_x = bytestream_get_be16(&buf);
- object->crop_y = bytestream_get_be16(&buf);
- object->crop_w = bytestream_get_be16(&buf);
- object->crop_h = bytestream_get_be16(&buf);
- }
-
- ff_dlog(avctx, "Subtitle Placement x=%d, y=%d\n",
- object->x, object->y);
-
- if (object->x > avctx->width || object->y > avctx->height) {
- av_log(avctx, AV_LOG_ERROR, "Subtitle out of video bounds. x = %d, y = %d, video width = %d, video height = %d.\n",
- object->x, object->y,
- avctx->width, avctx->height);
- object->y = object->x = 0;
- if (avctx->err_recognition & AV_EF_EXPLODE) {
+ if (buf_end - buf < 8) {
+ av_log(avctx, AV_LOG_ERROR, "Insufficent space for object\n");
+ ctx->presentation.object_count = i;
return AVERROR_INVALIDDATA;
}
+
+ object->id = bytestream_get_be16(&buf);
+ object->window_id = bytestream_get_byte(&buf);
+ object->composition_flag = bytestream_get_byte(&buf);
+
+ object->x = bytestream_get_be16(&buf);
+ object->y = bytestream_get_be16(&buf);
+
+ // If cropping
+ if (object->composition_flag & 0x80) {
+ object->crop_x = bytestream_get_be16(&buf);
+ object->crop_y = bytestream_get_be16(&buf);
+ object->crop_w = bytestream_get_be16(&buf);
+ object->crop_h = bytestream_get_be16(&buf);
+ }
+
+ /* Placement is checked at decode after cropping. */
+ ff_dlog(avctx, "Subtitle Placement x=%d, y=%d\n",
+ object->x, object->y);
}
}
+ return 0;
+}
+
+/**
+ * Parse the window segment packet.
+ *
+ * The window segment instructs the decoder to redraw the graphic plane
+ * with the composition references provided in the presentation segment
+ *
+ * @param avctx contains the current codec context
+ */
+static int parse_window_segment(AVCodecContext *avctx, const uint8_t *buf,
+ int buf_size)
+{
+ PGSSubContext *ctx = (PGSSubContext *)avctx->priv_data;
+
+ // 1 byte: number of windows defined
+ if (bytestream_get_byte(&buf) > MAX_OBJECT_REFS) {
+ av_log(avctx, AV_LOG_ERROR, "Too many windows defined.\n");
+ return AVERROR_INVALIDDATA;
+ }
+ /* TODO: mask objects with windows when transfering to the graphic plane
+ * Window Segment Structure
+ * {
+ * 1 byte : window id,
+ * 2 bytes: X position of window,
+ * 2 bytes: Y position of window,
+ * 2 bytes: Width of window,
+ * 2 bytes: Height of window.
+ * }
+ */
+ // Flush the graphic plane, it will be redrawn.
+ clear_graphic_plane(ctx);
+ ctx->plane.writable = 1;
+ ctx->plane.count = ctx->presentation.object_count;
return 0;
}
/**
* Parse the display segment packet.
*
- * The display segment controls the updating of the display.
+ * The display segment closes the display set. The inferred data is used
+ * to decide if the display should be updated.
*
* @param avctx contains the current codec context
* @param data pointer to the data pertaining the subtitle to display
@@ -500,26 +581,33 @@
PGSSubContext *ctx = avctx->priv_data;
int64_t pts;
PGSSubPalette *palette;
- int i, ret;
+ int i;
pts = ctx->presentation.pts != AV_NOPTS_VALUE ? ctx->presentation.pts : sub->pts;
memset(sub, 0, sizeof(*sub));
sub->pts = pts;
ctx->presentation.pts = AV_NOPTS_VALUE;
- sub->start_display_time = 0;
// There is no explicit end time for PGS subtitles. The end time
// is defined by the start of the next sub which may contain no
// objects (i.e. clears the previous sub)
sub->end_display_time = UINT32_MAX;
- sub->format = 0;
- // Blank if last object_count was 0.
- if (!ctx->presentation.object_count)
+ // Object count is zero only on an epoch start with no WDS
+ // or the last DS with a WDS had no presentation object.
+ if (!ctx->plane.count) {
return 1;
- sub->rects = av_calloc(ctx->presentation.object_count, sizeof(*sub->rects));
- if (!sub->rects) {
- return AVERROR(ENOMEM);
}
+
+ if (!ctx->presentation.palette_flag && !ctx->plane.writable) {
+ // This display set does not perform a display update
+ // E.g. it only defines new objects or palettes for future usage.
+ return 0;
+ }
+
+ sub->rects = av_calloc(ctx->plane.count, sizeof(*sub->rects));
+ if (!sub->rects)
+ return AVERROR(ENOMEM);
+
palette = find_palette(ctx->presentation.palette_id, &ctx->palettes);
if (!palette) {
// Missing palette. Should only happen with damaged streams.
@@ -528,57 +616,128 @@
avsubtitle_free(sub);
return AVERROR_INVALIDDATA;
}
- for (i = 0; i < ctx->presentation.object_count; i++) {
- AVSubtitleRect *const rect = av_mallocz(sizeof(*rect));
- PGSSubObject *object;
- if (!rect)
- return AVERROR(ENOMEM);
- sub->rects[sub->num_rects++] = rect;
- rect->type = SUBTITLE_BITMAP;
+ for (i = 0; i < ctx->plane.count; i++) {
+ const PGSSubObjectRef *sub_object = &ctx->presentation.objects[i];
+ AVSubtitleRect *const gp_rect = &ctx->plane.visible_rect[i];
+ AVSubtitleRect *rect;
+ gp_rect->type = SUBTITLE_BITMAP;
+
+ // Compose the graphic plane if a window segment has been provided
+ if (ctx->plane.writable) {
+ PGSSubObject *object;
+
+ // Process bitmap
+ object = find_object(sub_object->id, &ctx->objects);
+ if (!object) {
+ // Missing object. Should only happen with damaged streams.
+ av_log(avctx, AV_LOG_ERROR, "Invalid object id %d\n", sub_object->id);
+ if (avctx->err_recognition & AV_EF_EXPLODE)
+ return AVERROR_INVALIDDATA;
+ // Leaves rect empty with 0 width and height.
+ continue;
+ }
+ if (sub_object->composition_flag & 0x40)
+ gp_rect->flags |= AV_SUBTITLE_FLAG_FORCED;
- /* Process bitmap */
- object = find_object(ctx->presentation.objects[i].id, &ctx->objects);
- if (!object) {
- // Missing object. Should only happen with damaged streams.
- av_log(avctx, AV_LOG_ERROR, "Invalid object id %d\n",
- ctx->presentation.objects[i].id);
- if (avctx->err_recognition & AV_EF_EXPLODE)
- return AVERROR_INVALIDDATA;
- // Leaves rect empty with 0 width and height.
- continue;
- }
- if (ctx->presentation.objects[i].composition_flag & 0x40)
- rect->flags |= AV_SUBTITLE_FLAG_FORCED;
+ gp_rect->x = sub_object->x;
+ gp_rect->y = sub_object->y;
- rect->x = ctx->presentation.objects[i].x;
- rect->y = ctx->presentation.objects[i].y;
+ if (object->rle) {
+ int out_of_picture = 0;
+ gp_rect->w = object->w;
+ gp_rect->h = object->h;
+
+ gp_rect->linesize[0] = object->w;
+
+ // Check for cropping.
+ if (sub_object->composition_flag & 0x80) {
+ int out_of_object = 0;
+
+ if (object->w < sub_object->crop_x + sub_object->crop_w)
+ out_of_object = 1;
+ if (object->h < sub_object->crop_y + sub_object->crop_h)
+ out_of_object = 1;
+
+ if (out_of_object) {
+ av_log(avctx, AV_LOG_ERROR,
+ "Subtitle cropping values are out of object. "
+ "obj_w = %d, obj_h = %d, crop_x = %d, crop_y = %d, "
+ "crop_w = %d, crop_h = %d.\n",
+ object->w,
+ object->h,
+ sub_object->crop_x,
+ sub_object->crop_y,
+ sub_object->crop_w,
+ sub_object->crop_h);
+ if (avctx->err_recognition & AV_EF_EXPLODE)
+ return AVERROR_INVALIDDATA;
+ } else {
+ // Replace subtitle dimensions with cropping ones.
+ gp_rect->w = sub_object->crop_w;
+ gp_rect->h = sub_object->crop_h;
+ gp_rect->linesize[0] = sub_object->crop_w;
+ }
+ }
- if (object->rle) {
- rect->w = object->w;
- rect->h = object->h;
+ /* Make sure the subtitle is not out of picture. */
+ if (avctx->width < gp_rect->x + gp_rect->w || !gp_rect->w)
+ out_of_picture = 1;
+ if (avctx->height < gp_rect->y + gp_rect->h || !gp_rect->h)
+ out_of_picture = 1;
+ if (out_of_picture) {
+ av_log(avctx, AV_LOG_ERROR,
+ "Subtitle out of video bounds. "
+ "x = %d, y = %d, width = %d, height = %d.\n",
+ gp_rect->x, gp_rect->y, gp_rect->w, gp_rect->h);
+ if (avctx->err_recognition & AV_EF_EXPLODE)
+ return AVERROR_INVALIDDATA;
+ gp_rect->w = 0;
+ gp_rect->h = 0;
+ continue;
+ }
- rect->linesize[0] = object->w;
+ if (!object->bitmap_size || object->rle_remaining_len) {
+ gp_rect->w = 0;
+ gp_rect->h = 0;
+ continue;
+ }
- if (object->rle_remaining_len) {
- av_log(avctx, AV_LOG_ERROR, "RLE data length %u is %u bytes shorter than expected\n",
- object->rle_data_len, object->rle_remaining_len);
- if (avctx->err_recognition & AV_EF_EXPLODE)
- return AVERROR_INVALIDDATA;
- }
- ret = decode_rle(avctx, rect, object->rle, object->rle_data_len);
- if (ret < 0) {
- if ((avctx->err_recognition & AV_EF_EXPLODE) ||
- ret == AVERROR(ENOMEM)) {
- return ret;
+ gp_rect->data[0] = av_malloc_array(gp_rect->w, gp_rect->h);
+ if (!gp_rect->data[0])
+ return AVERROR(ENOMEM);
+
+ if (sub_object->composition_flag & 0x80) {
+ /* Copy cropped bitmap. */
+ int y;
+
+ for (y = 0; y < sub_object->crop_h; y++) {
+ memcpy(&gp_rect->data[0][y * sub_object->crop_w],
+ &object->bitmap[(sub_object->crop_y + y) *
+ object->w + sub_object->crop_x],
+ sub_object->crop_w);
+ }
+ }
+ else {
+ /* copy full object */
+ memcpy(gp_rect->data[0], object->bitmap, object->bitmap_size);
}
- rect->w = 0;
- rect->h = 0;
- continue;
}
}
- /* Allocate memory for colors */
- rect->nb_colors = 256;
+ // Export graphic plane content with latest palette
+ rect = av_memdup(gp_rect, sizeof(*gp_rect));
+ if (!rect)
+ return AVERROR(ENOMEM);
+
+ sub->rects[sub->num_rects++] = rect;
+ if (gp_rect->data[0]) {
+ rect->data[0] = av_memdup(gp_rect->data[0], rect->w*rect->h);
+ if (!rect->data[0])
+ return AVERROR(ENOMEM);
+ }
+
+ // Allocate memory for colors
+ rect->nb_colors = AVPALETTE_COUNT;
rect->data[1] = av_mallocz(AVPALETTE_SIZE);
if (!rect->data[1])
return AVERROR(ENOMEM);
@@ -641,14 +800,7 @@
ret = parse_presentation_segment(avctx, buf, segment_length, sub->pts);
break;
case WINDOW_SEGMENT:
- /*
- * Window Segment Structure (No new information provided):
- * 2 bytes: Unknown,
- * 2 bytes: X position of subtitle,
- * 2 bytes: Y position of subtitle,
- * 2 bytes: Width of subtitle,
- * 2 bytes: Height of subtitle.
- */
+ ret = parse_window_segment(avctx, buf, segment_length);
break;
case DISPLAY_SEGMENT:
if (*got_sub_ptr) {
diff -Naur ffmpeg-7.1.2.old/libavcodec/qsvenc_av1.c ffmpeg-7.1.2/libavcodec/qsvenc_av1.c
--- ffmpeg-7.1.2.old/libavcodec/qsvenc_av1.c 2025-10-27 10:07:00.531473480 +0100
+++ ffmpeg-7.1.2/libavcodec/qsvenc_av1.c 2025-10-27 10:07:02.443863055 +0100
@@ -189,6 +189,10 @@
{ "tile_cols", "Number of columns for tiled encoding", OFFSET(qsv.tile_cols), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, UINT16_MAX, VE },
{ "tile_rows", "Number of rows for tiled encoding", OFFSET(qsv.tile_rows), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, UINT16_MAX, VE },
{ "look_ahead_depth", "Depth of look ahead in number frames, available when extbrc option is enabled", OFFSET(qsv.look_ahead_depth), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 100, VE },
+#if QSV_HAVE_EXT_AV1_SCC
+ { "palette_mode", "Enable palette mode of Screen Content Tool for encoding", OFFSET(qsv.palette_mode), AV_OPT_TYPE_BOOL, { .i64 = 0}, 0, 1, VE},
+ { "intrabc", "Enable intra block copy of Screen Content Tool for encoding", OFFSET(qsv.intrabc), AV_OPT_TYPE_BOOL, { .i64 = 0}, 0, 1, VE},
+#endif
{ NULL },
};
diff -Naur ffmpeg-7.1.2.old/libavcodec/qsvenc.c ffmpeg-7.1.2/libavcodec/qsvenc.c
--- ffmpeg-7.1.2.old/libavcodec/qsvenc.c 2025-10-27 10:07:00.592474018 +0100
+++ ffmpeg-7.1.2/libavcodec/qsvenc.c 2025-10-27 10:07:02.443621469 +0100
@@ -494,6 +494,9 @@
mfxExtAV1BitstreamParam *av1_bs_param = (mfxExtAV1BitstreamParam *)coding_opts[1];
mfxExtCodingOption2 *co2 = (mfxExtCodingOption2*)coding_opts[2];
mfxExtCodingOption3 *co3 = (mfxExtCodingOption3*)coding_opts[3];
+#if QSV_HAVE_EXT_AV1_SCC
+ mfxExtAV1ScreenContentTools *scc = (mfxExtAV1ScreenContentTools*)coding_opts[4];
+#endif
av_log(avctx, AV_LOG_VERBOSE, "profile: %s; level: %"PRIu16"\n",
print_profile(avctx->codec_id, info->CodecProfile), info->CodecLevel);
@@ -566,6 +569,13 @@
print_threestate(av1_bs_param->WriteIVFHeaders));
av_log(avctx, AV_LOG_VERBOSE, "LowDelayBRC: %s\n", print_threestate(co3->LowDelayBRC));
av_log(avctx, AV_LOG_VERBOSE, "MaxFrameSize: %d;\n", co2->MaxFrameSize);
+#if QSV_HAVE_EXT_AV1_SCC
+ if (scc) {
+ av_log(avctx, AV_LOG_VERBOSE,
+ "Palette: %s; IntraBlockCopy: %s\n",
+ print_threestate(scc->Palette), print_threestate(scc->IntraBlockCopy));
+ }
+#endif
}
#endif
@@ -1282,6 +1292,28 @@
}
#endif
+#if QSV_HAVE_EXT_AV1_SCC
+ if (q->palette_mode || q->intrabc) {
+ if (QSV_RUNTIME_VERSION_ATLEAST(q->ver, 2, 13)) {
+ if (q->param.mfx.CodecId != MFX_CODEC_AV1) {
+ av_log(avctx, AV_LOG_ERROR, "Not supported encoder for Screen Content Tool Encode. "
+ "Supported: av1_qsv \n");
+ return AVERROR_UNKNOWN;
+ }
+
+ q->extsccparam.Header.BufferId = MFX_EXTBUFF_AV1_SCREEN_CONTENT_TOOLS;
+ q->extsccparam.Header.BufferSz = sizeof(q->extsccparam);
+ q->extsccparam.Palette = q->palette_mode ? MFX_CODINGOPTION_ON : MFX_CODINGOPTION_OFF;
+ q->extsccparam.IntraBlockCopy = q->intrabc ? MFX_CODINGOPTION_ON : MFX_CODINGOPTION_OFF;
+ q->extparam_internal[q->nb_extparam_internal++] = (mfxExtBuffer *)&q->extsccparam;
+ } else {
+ av_log(avctx, AV_LOG_ERROR,
+ "This version of runtime doesn't support Screen Content Tool Encode\n");
+ return AVERROR_UNKNOWN;
+ }
+ }
+#endif
+
if (!check_enc_param(avctx,q)) {
av_log(avctx, AV_LOG_ERROR,
"some encoding parameters are not supported by the QSV "
@@ -1389,11 +1421,21 @@
.Header.BufferSz = sizeof(co3),
};
+#if QSV_HAVE_EXT_AV1_SCC
+ mfxExtAV1ScreenContentTools scc_buf = {
+ .Header.BufferId = MFX_EXTBUFF_AV1_SCREEN_CONTENT_TOOLS,
+ .Header.BufferSz = sizeof(scc_buf),
+ };
+#endif
+
mfxExtBuffer *ext_buffers[] = {
(mfxExtBuffer*)&av1_extend_tile_buf,
(mfxExtBuffer*)&av1_bs_param,
(mfxExtBuffer*)&co2,
(mfxExtBuffer*)&co3,
+#if QSV_HAVE_EXT_AV1_SCC
+ (mfxExtBuffer*)&scc_buf,
+#endif
};
if (!QSV_RUNTIME_VERSION_ATLEAST(q->ver, 2, 5)) {
@@ -1842,6 +1884,13 @@
return ret;
}
+ // Update AVCodecContext with actual encoding parameters
+ mfxInfoMFX *info = &q->param.mfx;
+ avctx->has_b_frames = 0;
+ if (info->GopRefDist > 1) {
+ avctx->has_b_frames = info->GopRefDist - 1;
+ }
+
q->avctx = avctx;
return 0;
diff -Naur ffmpeg-7.1.2.old/libavcodec/qsvenc.h ffmpeg-7.1.2/libavcodec/qsvenc.h
--- ffmpeg-7.1.2.old/libavcodec/qsvenc.h 2025-10-27 10:07:00.282471283 +0100
+++ ffmpeg-7.1.2/libavcodec/qsvenc.h 2025-10-27 10:07:02.443799916 +0100
@@ -38,6 +38,7 @@
#define QSV_HAVE_EXT_VP9_TILES QSV_VERSION_ATLEAST(1, 29)
#define QSV_HAVE_EXT_AV1_PARAM QSV_VERSION_ATLEAST(2, 5)
+#define QSV_HAVE_EXT_AV1_SCC QSV_VERSION_ATLEAST(2, 13)
#if defined(_WIN32) || defined(__CYGWIN__)
#define QSV_HAVE_AVBR 1
@@ -188,10 +189,13 @@
mfxFrameSurface1 **opaque_surfaces;
AVBufferRef *opaque_alloc_buf;
#endif
+#if QSV_HAVE_EXT_AV1_SCC
+ mfxExtAV1ScreenContentTools extsccparam;
+#endif
mfxExtVideoSignalInfo extvsi;
- mfxExtBuffer *extparam_internal[5 + (QSV_HAVE_MF * 2) + (QSV_HAVE_EXT_AV1_PARAM * 2) + QSV_HAVE_HE];
+ mfxExtBuffer *extparam_internal[5 + (QSV_HAVE_MF * 2) + (QSV_HAVE_EXT_AV1_PARAM * 2) + QSV_HAVE_HE + QSV_HAVE_EXT_AV1_SCC];
int nb_extparam_internal;
mfxExtBuffer **extparam_str;
@@ -319,6 +323,8 @@
int dual_gfx;
AVDictionary *qsv_params;
+ int palette_mode;
+ int intrabc;
} QSVEncContext;
int ff_qsv_enc_init(AVCodecContext *avctx, QSVEncContext *q);
diff -Naur ffmpeg-7.1.2.old/libavcodec/videotoolbox_av1.c ffmpeg-7.1.2/libavcodec/videotoolbox_av1.c
--- ffmpeg-7.1.2.old/libavcodec/videotoolbox_av1.c 1970-01-01 01:00:00.000000000 +0100
+++ ffmpeg-7.1.2/libavcodec/videotoolbox_av1.c 2025-10-27 10:07:02.455024552 +0100
@@ -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),
+};
diff -Naur ffmpeg-7.1.2.old/libavcodec/videotoolbox.c ffmpeg-7.1.2/libavcodec/videotoolbox.c
--- ffmpeg-7.1.2.old/libavcodec/videotoolbox.c 2025-10-27 10:07:00.379472139 +0100
+++ ffmpeg-7.1.2/libavcodec/videotoolbox.c 2025-10-27 10:07:02.456490466 +0100
@@ -56,6 +56,10 @@
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 @@
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;
@@ -108,9 +132,6 @@
frame->crop_top = 0;
frame->crop_bottom = 0;
- if ((ret = av_vt_pixbuf_set_attachments(avctx, ref->pixbuf, frame)) < 0)
- return ret;
-
frame->data[3] = (uint8_t*)ref->pixbuf;
if (ref->hw_frames_ctx) {
@@ -790,7 +811,7 @@
#if TARGET_OS_IPHONE
CFDictionarySetValue(buffer_attributes, kCVPixelBufferOpenGLESCompatibilityKey, kCFBooleanTrue);
#else
- CFDictionarySetValue(buffer_attributes, kCVPixelBufferIOSurfaceOpenGLTextureCompatibilityKey, kCFBooleanTrue);
+ CFDictionarySetValue(buffer_attributes, kCVPixelBufferMetalCompatibilityKey, kCFBooleanTrue);
#endif
CFRelease(io_surface_properties);
@@ -847,6 +868,13 @@
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;
}
@@ -912,10 +940,30 @@
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;
}
+#if ARCH_X86_64
+ if (avctx->codec_id == AV_CODEC_ID_H264 &&
+ avctx->sw_pix_fmt == AV_PIX_FMT_YUV420P10)
+ {
+ // 10-bit H.264 is not supported on x86_64
+ return AVERROR(ENOSYS);
+ }
+#endif
+
+ if (avctx->codec_id == AV_CODEC_ID_H264 &&
+ (avctx->level == 61 || avctx->level == 62))
+ {
+ // H.264 Level 6.1 and 6.2 can't be
+ // decoded properly
+ return AVERROR(ENOSYS);
+ }
+
#if defined(MAC_OS_X_VERSION_10_9) && !TARGET_OS_IPHONE && (MAC_OS_X_VERSION_MAX_ALLOWED >= MAC_OS_X_VERSION_10_9) && AV_HAS_BUILTIN(__builtin_available)
if (avctx->codec_id == AV_CODEC_ID_PRORES) {
if (__builtin_available(macOS 10.9, *)) {
diff -Naur ffmpeg-7.1.2.old/libavcodec/vt_internal.h ffmpeg-7.1.2/libavcodec/vt_internal.h
--- ffmpeg-7.1.2.old/libavcodec/vt_internal.h 2025-10-27 10:07:00.531473480 +0100
+++ ffmpeg-7.1.2/libavcodec/vt_internal.h 2025-10-27 10:07:02.455086969 +0100
@@ -56,6 +56,9 @@
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 @@
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);
diff -Naur ffmpeg-7.1.2.old/libavfilter/vf_vpp_qsv.c ffmpeg-7.1.2/libavfilter/vf_vpp_qsv.c
--- ffmpeg-7.1.2.old/libavfilter/vf_vpp_qsv.c 2025-10-27 10:07:00.764475536 +0100
+++ ffmpeg-7.1.2/libavfilter/vf_vpp_qsv.c 2025-10-27 10:07:02.461658339 +0100
@@ -494,9 +494,9 @@
outvsi_conf.Header.BufferId = MFX_EXTBUFF_VIDEO_SIGNAL_INFO_OUT;
outvsi_conf.Header.BufferSz = sizeof(mfxExtVideoSignalInfo);
outvsi_conf.VideoFullRange = (out->color_range == AVCOL_RANGE_JPEG);
- outvsi_conf.ColourPrimaries = (out->color_primaries == AVCOL_PRI_UNSPECIFIED) ? AVCOL_PRI_BT709 : out->color_primaries;
- outvsi_conf.TransferCharacteristics = (out->color_trc == AVCOL_TRC_UNSPECIFIED) ? AVCOL_TRC_BT709 : out->color_trc;
- outvsi_conf.MatrixCoefficients = (out->colorspace == AVCOL_SPC_UNSPECIFIED) ? AVCOL_SPC_BT709 : out->colorspace;
+ outvsi_conf.ColourPrimaries = (out->color_primaries == AVCOL_PRI_UNSPECIFIED) ? invsi_conf.ColourPrimaries : out->color_primaries;
+ outvsi_conf.TransferCharacteristics = (out->color_trc == AVCOL_TRC_UNSPECIFIED) ? invsi_conf.TransferCharacteristics : out->color_trc;
+ outvsi_conf.MatrixCoefficients = (out->colorspace == AVCOL_SPC_UNSPECIFIED) ? invsi_conf.MatrixCoefficients : out->colorspace;
outvsi_conf.ColourDescriptionPresent = 1;
if (memcmp(&vpp->invsi_conf, &invsi_conf, sizeof(mfxExtVideoSignalInfo)) ||
diff -Naur ffmpeg-7.1.2.old/libavformat/isom.h ffmpeg-7.1.2/libavformat/isom.h
--- ffmpeg-7.1.2.old/libavformat/isom.h 2025-10-27 10:07:00.909476815 +0100
+++ ffmpeg-7.1.2/libavformat/isom.h 2025-10-27 10:07:02.422783282 +0100
@@ -272,6 +272,9 @@
MOVEncryptionIndex *encryption_index;
} cenc;
+ int has_fallback; // Audio fallback track
+ int fallback;
+
struct IAMFDemuxContext *iamf;
} MOVStreamContext;
diff -Naur ffmpeg-7.1.2.old/libavformat/matroskaenc.c ffmpeg-7.1.2/libavformat/matroskaenc.c
--- ffmpeg-7.1.2.old/libavformat/matroskaenc.c 2025-10-27 10:07:00.955477221 +0100
+++ ffmpeg-7.1.2/libavformat/matroskaenc.c 2025-10-27 10:07:02.430490236 +0100
@@ -2933,6 +2933,16 @@
case AV_CODEC_ID_AAC:
if (side_data_size && mkv->track.bc) {
int output_sample_rate = 0;
+ if (par->extradata && par->extradata_size) {
+ if (par->extradata_size != side_data_size ||
+ memcmp(par->extradata, side_data, side_data_size)) {
+ av_log(s, AV_LOG_ERROR, "Error, AAC extradata changed mid-stream.\n");
+ return AVERROR(EINVAL);
+ } else {
+ // Already written
+ break;
+ }
+ }
ret = get_aac_sample_rates(s, mkv, side_data, side_data_size,
&track->sample_rate, &output_sample_rate);
if (ret < 0)
diff -Naur ffmpeg-7.1.2.old/libavformat/mov.c ffmpeg-7.1.2/libavformat/mov.c
--- ffmpeg-7.1.2.old/libavformat/mov.c 2025-10-27 10:07:00.889476639 +0100
+++ ffmpeg-7.1.2/libavformat/mov.c 2025-10-27 10:07:02.447584145 +0100
@@ -56,6 +56,7 @@
#include "libavcodec/mpegaudiodecheader.h"
#include "libavcodec/mlp_parse.h"
#include "avformat.h"
+#include "avlanguage.h"
#include "internal.h"
#include "avio_internal.h"
#include "demux.h"
@@ -131,6 +132,33 @@
return 0;
}
+static int mov_metadata_int16_no_padding(MOVContext *c, AVIOContext *pb,
+ unsigned len, const char *key)
+{
+ c->fc->event_flags |= AVFMT_EVENT_FLAG_METADATA_UPDATED;
+ av_dict_set_int(&c->fc->metadata, key, avio_rb16(pb), 0);
+
+ return 0;
+}
+
+static int mov_metadata_int32_no_padding(MOVContext *c, AVIOContext *pb,
+ unsigned len, const char *key)
+{
+ c->fc->event_flags |= AVFMT_EVENT_FLAG_METADATA_UPDATED;
+ av_dict_set_int(&c->fc->metadata, key, avio_rb32(pb), 0);
+
+ return 0;
+}
+
+static int mov_metadata_int64_no_padding(MOVContext *c, AVIOContext *pb,
+ unsigned len, const char *key)
+{
+ c->fc->event_flags |= AVFMT_EVENT_FLAG_METADATA_UPDATED;
+ av_dict_set_int(&c->fc->metadata, key, avio_rb64(pb), 0);
+
+ return 0;
+}
+
static int mov_metadata_gnre(MOVContext *c, AVIOContext *pb,
unsigned len, const char *key)
{
@@ -342,6 +370,73 @@
return 0;
}
+static int mov_read_3gp_udta_tag(MOVContext *c, AVIOContext *pb, MOVAtom atom)
+{
+ const char *key;
+ AVDictionary *metadata;
+ uint16_t langcode = 0;
+ char key2[32], language[4] = {0};
+ uint32_t str_size, version;
+ char *str;
+ if (atom.size < 6)
+ return AVERROR_INVALIDDATA;
+ switch (atom.type) {
+ case MKTAG( 'a','l','b','m'): key = "album"; break;
+ case MKTAG( 'a','u','t','h'): key = "author"; break;
+ case MKTAG( 'c','p','r','t'): key = "copyright"; break;
+ case MKTAG( 'd','s','c','p'): key = "comment"; break;
+ case MKTAG( 'g','n','r','e'): key = "genre"; break;
+ case MKTAG( 'p','e','r','f'): key = "artist"; break;
+ case MKTAG( 't','i','t','l'): key = "title"; break;
+ case MKTAG( 'y','r','r','c'): key = "date"; break;
+ default: return 0;
+ }
+ version = avio_rb32(pb); // version + flags
+ if (version != 0)
+ av_log(c->fc, AV_LOG_WARNING, "udta %s unknown version number: %u\n", str, version);
+ if (MKTAG( 'y','r','r','c') == atom.type) {
+ int year;
+ year = avio_rb16(pb);
+ str = av_asprintf("%d", year);
+ if (!str)
+ return AVERROR(ENOMEM);
+ } else {
+ int ret;
+ const char *tmp;
+ langcode = avio_rb16(pb);
+ ff_mov_lang_to_iso639(langcode, language);
+ tmp = ff_convert_lang_to(language, AV_LANG_ISO639_2_BIBL);
+ if (!tmp)
+ av_log(c->fc, AV_LOG_WARNING, "udta %s unknown language code: %u\n", str, langcode);
+ str_size = atom.size - 6;
+ if (str_size <= 0 || str_size >= INT_MAX/2)
+ return AVERROR_INVALIDDATA;
+ str = av_mallocz(str_size + 1);
+ if (!str)
+ return AVERROR(ENOMEM);
+ ret = ffio_read_size(pb, str, str_size);
+ if (ret < 0) {
+ av_free(str);
+ return ret;
+ }
+ str[str_size] = 0;
+ }
+ if (c->trak_index < 0) {
+ metadata = c->fc->metadata;
+ c->fc->event_flags |= AVFMT_EVENT_FLAG_METADATA_UPDATED;
+ }
+ else {
+ metadata = c->fc->streams[c->trak_index]->metadata;
+ }
+ av_dict_set(&metadata, key, str, 0);
+ if (*language && strcmp(language, "und")) {
+ snprintf(key2, sizeof(key2), "%s-%s", key, language);
+ av_dict_set(&metadata, key2, str, 0);
+ }
+ av_freep(&str);
+ return 0;
+}
+
static int mov_read_udta_string(MOVContext *c, AVIOContext *pb, MOVAtom atom)
{
char tmp_key[AV_FOURCC_MAX_STRING_SIZE] = {0};
@@ -354,6 +449,7 @@
int (*parse)(MOVContext*, AVIOContext*, unsigned, const char*) = NULL;
int raw = 0;
int num = 0;
+ AVDictionary *metadata;
switch (atom.type) {
case MKTAG( '@','P','R','M'): key = "premiere_version"; raw = 1; break;
@@ -364,35 +460,71 @@
case MKTAG( 'a','k','I','D'): key = "account_type";
parse = mov_metadata_int8_no_padding; break;
case MKTAG( 'a','p','I','D'): key = "account_id"; break;
+ case MKTAG( 'a','t','I','D'): key = "artist_id";
+ parse = mov_metadata_int32_no_padding; break;
case MKTAG( 'c','a','t','g'): key = "category"; break;
+ case MKTAG( 'c','m','I','D'): key = "composer_id";
+ parse = mov_metadata_int32_no_padding; break;
+ case MKTAG( 'c','n','I','D'): key = "content_id";
+ parse = mov_metadata_int32_no_padding; break;
case MKTAG( 'c','p','i','l'): key = "compilation";
parse = mov_metadata_int8_no_padding; break;
- case MKTAG( 'c','p','r','t'): key = "copyright"; break;
+ case MKTAG( 'c','p','r','t'):
+ key = "copyright";
+ if (!c->itunes_metadata) {
+ int64_t pos = avio_tell(pb);
+ int ret = mov_read_3gp_udta_tag(c, pb, atom);
+ if (ret != AVERROR_INVALIDDATA)
+ return ret;
+ avio_seek(pb, pos, SEEK_SET);
+ }
+ break;
case MKTAG( 'd','e','s','c'): key = "description"; break;
case MKTAG( 'd','i','s','k'): key = "disc";
parse = mov_metadata_track_or_disc_number; break;
case MKTAG( 'e','g','i','d'): key = "episode_uid";
parse = mov_metadata_int8_no_padding; break;
case MKTAG( 'F','I','R','M'): key = "firmware"; raw = 1; break;
- case MKTAG( 'g','n','r','e'): key = "genre";
- parse = mov_metadata_gnre; break;
+ case MKTAG( 'g','e','I','D'): key = "genre_id";
+ parse = mov_metadata_int32_no_padding; break;
+ case MKTAG( 'g','n','r','e'):
+ key = "genre";
+ parse = mov_metadata_gnre;
+ if (!c->itunes_metadata) {
+ int64_t pos = avio_tell(pb);
+ int ret = mov_read_3gp_udta_tag(c, pb, atom);
+ if (ret != AVERROR_INVALIDDATA)
+ return ret;
+ avio_seek(pb, pos, SEEK_SET);
+ }
+ break;
case MKTAG( 'h','d','v','d'): key = "hd_video";
parse = mov_metadata_int8_no_padding; break;
case MKTAG( 'H','M','M','T'):
return mov_metadata_hmmt(c, pb, atom.size);
+ case MKTAG( 'i','t','n','u'): key = "itunes_u";
+ parse = mov_metadata_int8_no_padding; break;
case MKTAG( 'k','e','y','w'): key = "keywords"; break;
case MKTAG( 'l','d','e','s'): key = "synopsis"; break;
case MKTAG( 'l','o','c','i'):
return mov_metadata_loci(c, pb, atom.size);
case MKTAG( 'm','a','n','u'): key = "make"; break;
case MKTAG( 'm','o','d','l'): key = "model"; break;
+ case MKTAG( 'n','a','m','e'): key = "title"; raw = 1; break;
case MKTAG( 'p','c','s','t'): key = "podcast";
parse = mov_metadata_int8_no_padding; break;
case MKTAG( 'p','g','a','p'): key = "gapless_playback";
parse = mov_metadata_int8_no_padding; break;
+ case MKTAG( 'p','l','I','D'): key = "playlist_id";
+ parse = mov_metadata_int64_no_padding; break;
case MKTAG( 'p','u','r','d'): key = "purchase_date"; break;
case MKTAG( 'r','t','n','g'): key = "rating";
parse = mov_metadata_int8_no_padding; break;
+ case MKTAG( 's','f','I','D'): key = "itunes_country";
+ parse = mov_metadata_int32_no_padding; break;
+ case MKTAG( 's','d','e','s'): key = "series_description"; break;
+ case MKTAG( 's','h','w','m'): key = "show_work_and_movement";
+ parse = mov_metadata_int8_no_padding; break;
case MKTAG( 's','o','a','a'): key = "sort_album_artist"; break;
case MKTAG( 's','o','a','l'): key = "sort_album"; break;
case MKTAG( 's','o','a','r'): key = "sort_artist"; break;
@@ -401,6 +533,8 @@
case MKTAG( 's','o','s','n'): key = "sort_show"; break;
case MKTAG( 's','t','i','k'): key = "media_type";
parse = mov_metadata_int8_no_padding; break;
+ case MKTAG( 't','m','p','o'): key = "tmpo";
+ parse = mov_metadata_int16_no_padding; break;
case MKTAG( 't','r','k','n'): key = "track";
parse = mov_metadata_track_or_disc_number; break;
case MKTAG( 't','v','e','n'): key = "episode_id"; break;
@@ -410,17 +544,23 @@
case MKTAG( 't','v','s','h'): key = "show"; break;
case MKTAG( 't','v','s','n'): key = "season_number";
parse = mov_metadata_int8_bypass_padding; break;
+ case MKTAG( 'x','i','d',' '): key = "xid"; break;
case MKTAG(0xa9,'A','R','T'): key = "artist"; break;
case MKTAG(0xa9,'P','R','D'): key = "producer"; break;
case MKTAG(0xa9,'a','l','b'): key = "album"; break;
- case MKTAG(0xa9,'a','u','t'): key = "artist"; break;
+ case MKTAG(0xa9,'a','r','d'): key = "art_director"; break;
+ case MKTAG(0xa9,'a','r','g'): key = "arranger"; break;
+ case MKTAG(0xa9,'a','u','t'): key = "author"; break;
+ case MKTAG(0xa9,'c','a','k'): key = "acknowledgement"; break;
case MKTAG(0xa9,'c','h','p'): key = "chapter"; break;
case MKTAG(0xa9,'c','m','t'): key = "comment"; break;
case MKTAG(0xa9,'c','o','m'): key = "composer"; break;
+ case MKTAG(0xa9,'c','o','n'): key = "conductor"; break;
case MKTAG(0xa9,'c','p','y'): key = "copyright"; break;
case MKTAG(0xa9,'d','a','y'): key = "date"; break;
case MKTAG(0xa9,'d','i','r'): key = "director"; break;
case MKTAG(0xa9,'d','i','s'): key = "disclaimer"; break;
+ case MKTAG(0xa9,'d','e','s'): key = "song_description"; break;
case MKTAG(0xa9,'e','d','1'): key = "edit_date"; break;
case MKTAG(0xa9,'e','n','c'): key = "encoder"; break;
case MKTAG(0xa9,'f','m','t'): key = "original_format"; break;
@@ -428,23 +568,45 @@
case MKTAG(0xa9,'g','r','p'): key = "grouping"; break;
case MKTAG(0xa9,'h','s','t'): key = "host_computer"; break;
case MKTAG(0xa9,'i','n','f'): key = "comment"; break;
+ case MKTAG(0xa9,'l','n','t'): key = "linear_notes"; break;
case MKTAG(0xa9,'l','y','r'): key = "lyrics"; break;
case MKTAG(0xa9,'m','a','k'): key = "make"; break;
case MKTAG(0xa9,'m','o','d'): key = "model"; break;
+ case MKTAG(0xa9,'m','v','n'): key = "movement_name"; break;
+ case MKTAG(0xa9,'m','v','i'): key = "movement_number";
+ parse = mov_metadata_int16_no_padding; break;
+ case MKTAG(0xa9,'m','v','c'): key = "movement_count";
+ parse = mov_metadata_int16_no_padding; break;
case MKTAG(0xa9,'n','a','m'): key = "title"; break;
case MKTAG(0xa9,'o','p','e'): key = "original_artist"; break;
+ case MKTAG(0xa9,'p','h','g'): key = "phonogram_rights"; break;
case MKTAG(0xa9,'p','r','d'): key = "producer"; break;
case MKTAG(0xa9,'p','r','f'): key = "performers"; break;
+ case MKTAG(0xa9,'p','u','b'): key = "publisher"; break;
case MKTAG(0xa9,'r','e','q'): key = "playback_requirements"; break;
+ case MKTAG(0xa9,'s','n','e'): key = "sound_engineer"; break;
+ case MKTAG(0xa9,'s','o','l'): key = "soloist"; break;
case MKTAG(0xa9,'s','r','c'): key = "original_source"; break;
case MKTAG(0xa9,'s','t','3'): key = "subtitle"; break;
case MKTAG(0xa9,'s','w','r'): key = "encoder"; break;
+ case MKTAG(0xa9,'t','h','x'): key = "thanks"; break;
case MKTAG(0xa9,'t','o','o'): key = "encoder"; break;
case MKTAG(0xa9,'t','r','k'): key = "track"; break;
case MKTAG(0xa9,'u','r','l'): key = "URL"; break;
+ case MKTAG(0xa9,'w','r','k'): key = "work_name"; break;
case MKTAG(0xa9,'w','r','n'): key = "warning"; break;
case MKTAG(0xa9,'w','r','t'): key = "composer"; break;
+ case MKTAG(0xa9,'x','p','d'): key = "executive_producer"; break;
case MKTAG(0xa9,'x','y','z'): key = "location"; break;
+ case MKTAG( 'a','l','b','m'):
+ case MKTAG( 'a','u','t','h'):
+ case MKTAG( 'd','s','c','p'):
+ case MKTAG( 'p','e','r','f'):
+ case MKTAG( 't','i','t','l'):
+ case MKTAG( 'y','r','r','c'):
+ if (!c->itunes_metadata) {
+ return mov_read_3gp_udta_tag(c, pb, atom);
+ }
}
retry:
if (c->itunes_metadata && atom.size > 8) {
@@ -571,17 +733,23 @@
}
str[str_size] = 0;
}
- c->fc->event_flags |= AVFMT_EVENT_FLAG_METADATA_UPDATED;
- av_dict_set(&c->fc->metadata, key, str, 0);
+ if (c->trak_index < 0) {
+ metadata = c->fc->metadata;
+ c->fc->event_flags |= AVFMT_EVENT_FLAG_METADATA_UPDATED;
+ if (!strcmp(key, "encoder")) {
+ int major, minor, micro;
+ if (sscanf(str, "HandBrake %d.%d.%d", &major, &minor, &micro) == 3) {
+ c->handbrake_version = 1000000*major + 1000*minor + micro;
+ }
+ }
+ }
+ else {
+ metadata = c->fc->streams[c->trak_index]->metadata;
+ }
+ av_dict_set(&metadata, key, str, 0);
if (*language && strcmp(language, "und")) {
snprintf(key2, sizeof(key2), "%s-%s", key, language);
- av_dict_set(&c->fc->metadata, key2, str, 0);
- }
- if (!strcmp(key, "encoder")) {
- int major, minor, micro;
- if (sscanf(str, "HandBrake %d.%d.%d", &major, &minor, &micro) == 3) {
- c->handbrake_version = 1000000*major + 1000*minor + micro;
- }
+ av_dict_set(&metadata, key2, str, 0);
}
}
@@ -9088,6 +9256,23 @@
return ret;
}
+static int mov_read_fall(MOVContext *c, AVIOContext *pb, MOVAtom atom)
+{
+ AVStream *st;
+ MOVStreamContext *sc;
+
+ if (c->fc->nb_streams < 1)
+ return 0;
+ st = c->fc->streams[c->fc->nb_streams-1];
+ sc = st->priv_data;
+
+ sc->fallback = avio_rb32(pb);
+ sc->has_fallback = 1;
+
+ return 0;
+}
+
+
static const MOVParseTableEntry mov_default_parse_table[] = {
{ MKTAG('A','C','L','R'), mov_read_aclr },
{ MKTAG('A','P','R','G'), mov_read_avid },
@@ -9190,6 +9375,7 @@
{ MKTAG('v','p','c','C'), mov_read_vpcc },
{ MKTAG('m','d','c','v'), mov_read_mdcv },
{ MKTAG('c','l','l','i'), mov_read_clli },
+{ MKTAG('f','a','l','l'), mov_read_fall },
{ MKTAG('d','v','c','C'), mov_read_dvcc_dvvc },
{ MKTAG('d','v','v','C'), mov_read_dvcc_dvvc },
{ MKTAG('d','v','w','C'), mov_read_dvcc_dvvc },
@@ -10374,6 +10560,23 @@
err = ff_replaygain_export(st, s->metadata);
if (err < 0)
return err;
+ if (sc->has_fallback) {
+ for (j = 0; j < s->nb_streams; j++) {
+ if (s->streams[j]->id == sc->fallback) {
+ AVPacketSideData *sd;
+ int *fallback;
+ sd = av_packet_side_data_new(&st->codecpar->coded_side_data,
+ &st->codecpar->nb_coded_side_data,
+ AV_PKT_DATA_FALLBACK_TRACK,
+ sizeof(int), 0);
+ if (!sd)
+ return AVERROR(ENOMEM);
+ fallback = (int*)sd->data;
+ *fallback = j;
+ break;
+ }
+ }
+ }
break;
case AVMEDIA_TYPE_VIDEO:
if (sc->display_matrix) {
diff -Naur ffmpeg-7.1.2.old/libavformat/movenc.c ffmpeg-7.1.2/libavformat/movenc.c
--- ffmpeg-7.1.2.old/libavformat/movenc.c 2025-10-27 10:07:00.933477027 +0100
+++ ffmpeg-7.1.2/libavformat/movenc.c 2025-10-27 10:07:02.463726465 +0100
@@ -28,6 +28,7 @@
#include "movenc.h"
#include "avformat.h"
+#include "avlanguage.h"
#include "avio_internal.h"
#include "dovi_isom.h"
#include "riff.h"
@@ -393,6 +394,8 @@
uint16_t chan_loc;
/* if there is no dependent substream, then one bit reserved instead */
} substream[1]; /* TODO: support 8 independent substreams */
+ /* indicates the decoding complexity, 8 bits */
+ uint8_t complexity_index_type_a;
};
static int mov_write_ac3_tag(AVFormatContext *s, AVIOContext *pb, MOVTrack *track)
@@ -474,6 +477,8 @@
info->data_rate = FFMAX(info->data_rate, hdr->bit_rate / 1000);
info->ac3_bit_rate_code = FFMAX(info->ac3_bit_rate_code,
hdr->ac3_bit_rate_code);
+ info->complexity_index_type_a = hdr->complexity_index_type_a;
+
num_blocks = hdr->num_blocks;
if (!info->ec3_done) {
@@ -532,8 +537,6 @@
int parent = hdr->substreamid;
while (cumul_size != pkt->size) {
- GetBitContext gbc;
- int i;
ret = avpriv_ac3_parse_header(&hdr, pkt->data + cumul_size, pkt->size - cumul_size);
if (ret < 0)
goto end;
@@ -544,20 +547,9 @@
info->substream[parent].num_dep_sub++;
ret /= 8;
- /* header is parsed up to lfeon, but custom channel map may be needed */
- init_get_bits8(&gbc, pkt->data + cumul_size + ret, pkt->size - cumul_size - ret);
- /* skip bsid */
- skip_bits(&gbc, 5);
- /* skip volume control params */
- for (i = 0; i < (hdr->channel_mode ? 1 : 2); i++) {
- skip_bits(&gbc, 5); // skip dialog normalization
- if (get_bits1(&gbc)) {
- skip_bits(&gbc, 8); // skip compression gain word
- }
- }
/* get the dependent stream channel map, if exists */
- if (get_bits1(&gbc))
- info->substream[parent].chan_loc |= (get_bits(&gbc, 16) >> 5) & 0x1f;
+ if (hdr->channel_map_present)
+ info->substream[parent].chan_loc |= (hdr->channel_map >> 5) & 0x1f;
else
info->substream[parent].chan_loc |= hdr->channel_mode;
cumul_size += hdr->frame_size;
@@ -614,7 +606,7 @@
}
info = track->eac3_priv;
- size = 2 + ((32 * (info->num_ind_sub + 1) + 7) >> 3);
+ size = 2 + (4 * (info->num_ind_sub + 1)) + (2 * !!info->complexity_index_type_a);
buf = av_malloc(size);
if (!buf) {
return AVERROR(ENOMEM);
@@ -639,6 +631,11 @@
put_bits(&pbc, 9, info->substream[i].chan_loc);
}
}
+ if (info->complexity_index_type_a) {
+ put_bits(&pbc, 7, 0); /* reserved */
+ put_bits(&pbc, 1, 1); // flag_eac3_extension_type_a
+ put_bits(&pbc, 8, info->complexity_index_type_a);
+ }
flush_put_bits(&pbc);
size = put_bytes_output(&pbc);
@@ -4055,6 +4052,41 @@
return len + 24;
}
+static uint16_t language_code(const char *str)
+{
+ return (((str[0] - 0x60) & 0x1F) << 10) +
+ (((str[1] - 0x60) & 0x1F) << 5) +
+ (( str[2] - 0x60) & 0x1F);
+}
+static int mov_write_3gp_udta_tag(AVIOContext *pb, AVDictionary *metadata,
+ const char *tag, const char *str)
+{
+ int64_t pos = avio_tell(pb);
+ AVDictionaryEntry *t = av_dict_get(metadata, str, NULL, 0);
+ if (!t || !utf8len(t->value))
+ return 0;
+ avio_wb32(pb, 0); /* size */
+ ffio_wfourcc(pb, tag); /* type */
+ avio_wb32(pb, 0); /* version + flags */
+ if (!strcmp(tag, "yrrc"))
+ avio_wb16(pb, atoi(t->value));
+ else {
+ int lang = 0, len;
+ len = strlen(t->key);
+ if (t->key[len - 4] == '-') {
+ lang = ff_mov_iso639_to_lang(&t->key[len - 3], 1);
+ }
+ if (!lang)
+ lang = ff_mov_iso639_to_lang("und", 1);
+ avio_wb16(pb, lang); /* language */
+ avio_write(pb, t->value, strlen(t->value) + 1); /* UTF8 string value */
+ if (!strcmp(tag, "albm") &&
+ (t = av_dict_get(metadata, "track", NULL, 0)))
+ avio_w8(pb, atoi(t->value));
+ }
+ return update_size(pb, pos);
+}
+
static int mov_write_track_metadata(AVIOContext *pb, AVStream *st,
const char *tag, const char *str)
{
@@ -4129,8 +4161,23 @@
if (ret < 0)
return ret;
- if (mov->mode & (MODE_MP4|MODE_MOV))
+ if (mov->mode & (MODE_MP4|MODE_MOV)) {
+ AVDictionaryEntry *t = NULL;
+ int und = 0;
+
mov_write_track_metadata(pb_buf, st, "name", "title");
+ while ((t = av_dict_get(st->metadata, "title-", t, AV_DICT_IGNORE_SUFFIX))) {
+ int len = strlen(t->key);
+ if (len == 10 &&
+ ff_convert_lang_to(&t->key[len - 3], AV_LANG_ISO639_2_BIBL)) {
+ mov_write_3gp_udta_tag(pb_buf, st->metadata, "titl", t->key);
+ if (!strcmp("und", &t->key[len - 3]))
+ und = 1;
+ }
+ }
+ if (!und)
+ mov_write_3gp_udta_tag(pb_buf, st->metadata, "titl", "title");
+ }
if (mov->mode & MODE_MP4) {
if ((ret = mov_write_track_kinds(pb_buf, st)) < 0)
@@ -4505,20 +4552,66 @@
return size;
}
+/* iTunes iTMF metadata */
+static int mov_write_itmf_tag(AVIOContext *pb, AVFormatContext *s,
+ const char *name, const char *tag)
+{
+ AVDictionaryEntry *t = av_dict_get(s->metadata, tag, NULL, 0);
+ uint64_t size = 0;
+ if (t) {
+ size_t name_len = strlen(name);
+ size_t value_len = strlen(t->value);
+ size = 8 + 28 + 12 + name_len + 16 + value_len;
+
+ if (size > INT_MAX) {
+ av_log(s, AV_LOG_WARNING,
+ "Unable to write %s, invalid value len %ld\n",
+ name, value_len);
+ return 0;
+ }
+
+ avio_wb32(pb, size); // size
+ ffio_wfourcc(pb, "----"); // type
+
+ avio_wb32(pb, 28); // size
+ ffio_wfourcc(pb, "mean"); // type
+ avio_wb32(pb, 0); // version + flags
+ avio_write(pb, "com.apple.iTunes", 16);
+
+ avio_wb32(pb, 12 + name_len); // size
+ ffio_wfourcc(pb, "name"); // type
+ avio_wb32(pb, 0); // version + flags
+ avio_write(pb, name, name_len);
+
+ avio_wb32(pb, 16 + value_len); // size
+ ffio_wfourcc(pb, "data"); // type
+ avio_wb16(pb, 0); // typeReserved
+ avio_w8(pb, 0); // typeSetIdentifier
+ avio_w8(pb, 1); // typeCode = UTF-8
+ avio_wb32(pb, 0); // locale
+ avio_write(pb, t->value, value_len);
+ }
+ return size;
+}
+
static int mov_write_int8_metadata(AVFormatContext *s, AVIOContext *pb,
const char *name, const char *tag,
int len)
{
AVDictionaryEntry *t = NULL;
- uint8_t num;
+ uint64_t num;
int size = 24 + len;
- if (len != 1 && len != 4)
+ if (len != 1 && len != 4 &&
+ len != 2 && len != 8)
return -1;
if (!(t = av_dict_get(s->metadata, tag, NULL, 0)))
return 0;
- num = atoi(t->value);
+ if (len <= 4)
+ num = atoi(t->value);
+ else
+ num = atol(t->value);
avio_wb32(pb, size);
ffio_wfourcc(pb, name);
@@ -4526,7 +4619,9 @@
ffio_wfourcc(pb, "data");
avio_wb32(pb, 0x15);
avio_wb32(pb, 0);
- if (len==4) avio_wb32(pb, num);
+ if (len==8) avio_wb64(pb, num);
+ else if (len==4) avio_wb32(pb, num);
+ else if (len==2) avio_wb16(pb, num);
else avio_w8 (pb, num);
return size;
@@ -4582,6 +4677,8 @@
mov_write_string_metadata(s, pb, "\251lyr", "lyrics" , 1);
mov_write_string_metadata(s, pb, "desc", "description",1);
mov_write_string_metadata(s, pb, "ldes", "synopsis" , 1);
+ mov_write_string_metadata(s, pb, "sdes", "series_description", 1);
+ mov_write_string_metadata(s, pb, "rtng", "rating", 1);
mov_write_string_metadata(s, pb, "tvsh", "show" , 1);
mov_write_string_metadata(s, pb, "tven", "episode_id",1);
mov_write_string_metadata(s, pb, "tvnn", "network" , 1);
@@ -4592,10 +4689,67 @@
mov_write_int8_metadata (s, pb, "hdvd", "hd_video", 1);
mov_write_int8_metadata (s, pb, "pgap", "gapless_playback",1);
mov_write_int8_metadata (s, pb, "cpil", "compilation", 1);
+
+ mov_write_string_metadata(s, pb, "\251st3", "subtitle" , 1);
+ mov_write_string_metadata(s, pb, "\251des", "song_description", 1);
+ mov_write_string_metadata(s, pb, "\251dir", "director" , 1);
+ mov_write_string_metadata(s, pb, "\251ard", "art_director" , 1);
+ mov_write_string_metadata(s, pb, "\251arg", "arranger" , 1);
+ mov_write_string_metadata(s, pb, "\251aut", "author" , 1);
+ mov_write_string_metadata(s, pb, "\251cak", "acknowledgement" , 1);
+ mov_write_string_metadata(s, pb, "\251con", "conductor" , 1);
+
+ mov_write_string_metadata(s, pb, "\251wrk", "work_name" , 1);
+ mov_write_string_metadata(s, pb, "\251mvn", "movement_name" , 1);
+ mov_write_int8_metadata (s, pb, "\251mvi", "movement_number", 2);
+ mov_write_int8_metadata (s, pb, "\251mvc", "movement_count" , 2);
+ mov_write_int8_metadata (s, pb, "shwm", "show_work_and_movement", 1);
+
+ mov_write_string_metadata(s, pb, "\251lnt", "linear_notes" , 1);
+ mov_write_string_metadata(s, pb, "\251mak", "make" , 1); // Record company
+ mov_write_string_metadata(s, pb, "\251ope", "original_artist" , 1);
+ mov_write_string_metadata(s, pb, "\251phg", "phonogram_rights" , 1);
+ mov_write_string_metadata(s, pb, "\251prd", "producer" , 1);
+ mov_write_string_metadata(s, pb, "\251prf", "performers" , 1);
+ mov_write_string_metadata(s, pb, "\251pub", "publisher" , 1);
+ mov_write_string_metadata(s, pb, "\251sne", "sound_engineer" , 1);
+ mov_write_string_metadata(s, pb, "\251sol", "soloist" , 1);
+ mov_write_string_metadata(s, pb, "\251src", "original_source" , 1); // Credits
+ mov_write_string_metadata(s, pb, "\251thx", "thanks" , 1);
+ mov_write_string_metadata(s, pb, "\251url", "URL" , 1); // Online extras
+ mov_write_string_metadata(s, pb, "\251xpd", "executive_producer", 1);
+
+ mov_write_string_metadata(s, pb, "sonm", "sort_name" , 1);
+ mov_write_string_metadata(s, pb, "soar", "sort_artist" , 1);
+ mov_write_string_metadata(s, pb, "soaa", "sort_album_artist", 1);
+ mov_write_string_metadata(s, pb, "soal", "sort_album" , 1);
+ mov_write_string_metadata(s, pb, "soco", "sort_composer" , 1);
+ mov_write_string_metadata(s, pb, "sosn", "sort_show" , 1);
+
+ mov_write_string_metadata(s, pb, "\251enc", "encoder" , 1); // Encoded by
+ mov_write_string_metadata(s, pb, "purd", "purchase_date" , 1);
+
+ mov_write_int8_metadata (s, pb, "itnu", "itunes_u" , 1);
+ mov_write_int8_metadata (s, pb, "pcst", "podcast" , 1);
+ mov_write_string_metadata(s, pb, "catg", "category" , 1);
+
+ mov_write_string_metadata(s, pb, "apID", "account_id" , 1);
+ mov_write_int8_metadata (s, pb, "akID", "account_type" , 1);
+ mov_write_int8_metadata (s, pb, "sfID", "itunes_country", 4);
+ mov_write_int8_metadata (s, pb, "cnID", "content_id" , 4);
+ mov_write_int8_metadata (s, pb, "atID", "artist_id" , 4);
+ mov_write_int8_metadata (s, pb, "plID", "playlist_id" , 8);
+ mov_write_int8_metadata (s, pb, "geID", "genre_id" , 4);
+ mov_write_int8_metadata (s, pb, "cmID", "composer_id" , 4);
+ mov_write_string_metadata(s, pb, "xid ", "xid" , 1);
+
mov_write_covr(pb, s);
mov_write_trkn_tag(pb, mov, s, 0); // track number
mov_write_trkn_tag(pb, mov, s, 1); // disc number
mov_write_tmpo_tag(pb, s);
+ mov_write_itmf_tag(pb, s, "iTunEXTC", "iTunEXTC");
+ mov_write_itmf_tag(pb, s, "iTunMOVI", "iTunMOVI");
+
return update_size(pb, pos);
}
@@ -4727,35 +4881,6 @@
return 0;
}
-static uint16_t language_code(const char *str)
-{
- return (((str[0] - 0x60) & 0x1F) << 10) +
- (((str[1] - 0x60) & 0x1F) << 5) +
- (( str[2] - 0x60) & 0x1F);
-}
-
-static int mov_write_3gp_udta_tag(AVIOContext *pb, AVFormatContext *s,
- const char *tag, const char *str)
-{
- int64_t pos = avio_tell(pb);
- AVDictionaryEntry *t = av_dict_get(s->metadata, str, NULL, 0);
- if (!t || !utf8len(t->value))
- return 0;
- avio_wb32(pb, 0); /* size */
- ffio_wfourcc(pb, tag); /* type */
- avio_wb32(pb, 0); /* version + flags */
- if (!strcmp(tag, "yrrc"))
- avio_wb16(pb, atoi(t->value));
- else {
- avio_wb16(pb, language_code("eng")); /* language */
- avio_write(pb, t->value, strlen(t->value) + 1); /* UTF8 string value */
- if (!strcmp(tag, "albm") &&
- (t = av_dict_get(s->metadata, "track", NULL, 0)))
- avio_w8(pb, atoi(t->value));
- }
- return update_size(pb, pos);
-}
-
static int mov_write_chpl_tag(AVIOContext *pb, AVFormatContext *s)
{
int64_t pos = avio_tell(pb);
@@ -4794,14 +4919,14 @@
return ret;
if (mov->mode & MODE_3GP) {
- mov_write_3gp_udta_tag(pb_buf, s, "perf", "artist");
- mov_write_3gp_udta_tag(pb_buf, s, "titl", "title");
- mov_write_3gp_udta_tag(pb_buf, s, "auth", "author");
- mov_write_3gp_udta_tag(pb_buf, s, "gnre", "genre");
- mov_write_3gp_udta_tag(pb_buf, s, "dscp", "comment");
- mov_write_3gp_udta_tag(pb_buf, s, "albm", "album");
- mov_write_3gp_udta_tag(pb_buf, s, "cprt", "copyright");
- mov_write_3gp_udta_tag(pb_buf, s, "yrrc", "date");
+ mov_write_3gp_udta_tag(pb_buf, s->metadata, "perf", "artist");
+ mov_write_3gp_udta_tag(pb_buf, s->metadata, "titl", "title");
+ mov_write_3gp_udta_tag(pb_buf, s->metadata, "auth", "author");
+ mov_write_3gp_udta_tag(pb_buf, s->metadata, "gnre", "genre");
+ mov_write_3gp_udta_tag(pb_buf, s->metadata, "dscp", "comment");
+ mov_write_3gp_udta_tag(pb_buf, s->metadata, "albm", "album");
+ mov_write_3gp_udta_tag(pb_buf, s->metadata, "cprt", "copyright");
+ mov_write_3gp_udta_tag(pb_buf, s->metadata, "yrrc", "date");
mov_write_loci_tag(s, pb_buf);
} else if (mov->mode == MODE_MOV && !(mov->flags & FF_MOV_FLAG_USE_MDTA)) { // the title field breaks gtkpod with mp4 and my suspicion is that stuff is not valid in mp4
mov_write_string_metadata(s, pb_buf, "\251ART", "artist", 0);
diff -Naur ffmpeg-7.1.2.old/libavutil/frame.c ffmpeg-7.1.2/libavutil/frame.c
--- ffmpeg-7.1.2.old/libavutil/frame.c 2025-10-27 10:07:00.804475889 +0100
+++ ffmpeg-7.1.2/libavutil/frame.c 2025-10-27 10:07:02.436386459 +0100
@@ -45,6 +45,7 @@
[AV_FRAME_DATA_FILM_GRAIN_PARAMS] = { "Film grain parameters" },
[AV_FRAME_DATA_DETECTION_BBOXES] = { "Bounding boxes for object detection and classification" },
[AV_FRAME_DATA_DOVI_RPU_BUFFER] = { "Dolby Vision RPU Data" },
+ [AV_FRAME_DATA_DOVI_RPU_BUFFER_T35] = { "Dolby Vision RPU ITU T35 Data" },
[AV_FRAME_DATA_DOVI_METADATA] = { "Dolby Vision Metadata" },
[AV_FRAME_DATA_LCEVC] = { "LCEVC NAL data" },
[AV_FRAME_DATA_VIEW_ID] = { "View ID" },
diff -Naur ffmpeg-7.1.2.old/libavutil/frame.h ffmpeg-7.1.2/libavutil/frame.h
--- ffmpeg-7.1.2.old/libavutil/frame.h 2025-10-27 10:07:00.786475730 +0100
+++ ffmpeg-7.1.2/libavutil/frame.h 2025-10-27 10:07:02.436480126 +0100
@@ -201,6 +201,12 @@
AV_FRAME_DATA_DOVI_RPU_BUFFER,
/**
+ * Dolby Vision RPU ITU T35 raw data, suitable for passing to SVT-AV1
+ * or other libraries. Array of uint8_t.
+ */
+ AV_FRAME_DATA_DOVI_RPU_BUFFER_T35,
+
+ /**
* Parsed Dolby Vision metadata, suitable for passing to a software
* implementation. The payload is the AVDOVIMetadata struct defined in
* libavutil/dovi_meta.h.
diff -Naur ffmpeg-7.1.2.old/libavutil/hwcontext_videotoolbox.c ffmpeg-7.1.2/libavutil/hwcontext_videotoolbox.c
--- ffmpeg-7.1.2.old/libavutil/hwcontext_videotoolbox.c 2025-10-27 10:07:00.813475968 +0100
+++ ffmpeg-7.1.2/libavutil/hwcontext_videotoolbox.c 2025-10-27 10:07:02.457067390 +0100
@@ -183,7 +183,8 @@
VTFramesContext *fctx = ctx->hwctx;
AVVTFramesContext *hw_ctx = &fctx->p;
CVReturn err;
- CFNumberRef w, h, pixfmt;
+ CFNumberRef w, h, extend_w_num, extend_h_num, pixfmt;
+ int extend_w, extend_h;
uint32_t cv_pixfmt;
CFMutableDictionaryRef attributes, iosurface_properties;
@@ -216,6 +217,15 @@
CFRelease(w);
CFRelease(h);
+ extend_w = FFALIGN(ctx->width, 16) - ctx->width;
+ extend_h = FFALIGN(ctx->height, 16) - ctx->height;
+ extend_w_num = CFNumberCreate(NULL, kCFNumberSInt32Type, &extend_w);
+ extend_h_num = CFNumberCreate(NULL, kCFNumberSInt32Type, &extend_h);
+ CFDictionarySetValue(attributes, kCVPixelBufferExtendedPixelsRightKey, extend_w_num);
+ CFDictionarySetValue(attributes, kCVPixelBufferExtendedPixelsBottomKey, extend_h_num);
+ CFRelease(extend_w_num);
+ CFRelease(extend_h_num);
+
err = CVPixelBufferPoolCreate(
NULL,
NULL,
diff -Naur ffmpeg-7.1.2.old/libswscale/swscale_unscaled.c ffmpeg-7.1.2/libswscale/swscale_unscaled.c
--- ffmpeg-7.1.2.old/libswscale/swscale_unscaled.c 2025-10-27 10:07:00.644474477 +0100
+++ ffmpeg-7.1.2/libswscale/swscale_unscaled.c 2025-10-27 10:07:02.433490262 +0100
@@ -341,7 +341,7 @@
const uint8_t *tsrc0 = src[0];
for (x = c->srcW; x > 0; x--) {
t = *tsrc0++;
- output_pixel(tdstY++, t | (t << 8));
+ output_pixel(tdstY++, (t << 8));
}
src[0] += srcStride[0];
dstY += dstStride[0] / 2;
@@ -352,9 +352,9 @@
const uint8_t *tsrc2 = src[2];
for (x = c->srcW / 2; x > 0; x--) {
t = *tsrc1++;
- output_pixel(tdstUV++, t | (t << 8));
+ output_pixel(tdstUV++, (t << 8));
t = *tsrc2++;
- output_pixel(tdstUV++, t | (t << 8));
+ output_pixel(tdstUV++, (t << 8));
}
src[1] += srcStride[1];
src[2] += srcStride[2];