diff -Naur ffmpeg-7.1.1.old/configure ffmpeg-7.1.1/configure --- ffmpeg-7.1.1.old/configure 2025-03-24 10:54:39.093543820 +0100 +++ ffmpeg-7.1.1/configure 2025-03-24 10:54:39.281929795 +0100 @@ -3338,6 +3338,7 @@ 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" diff -Naur ffmpeg-7.1.1.old/libavcodec/allcodecs.c ffmpeg-7.1.1/libavcodec/allcodecs.c --- ffmpeg-7.1.1.old/libavcodec/allcodecs.c 2025-03-24 10:54:39.094862193 +0100 +++ ffmpeg-7.1.1/libavcodec/allcodecs.c 2025-03-24 10:54:39.282298069 +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-7.1.1.old/libavcodec/amfenc_av1.c ffmpeg-7.1.1/libavcodec/amfenc_av1.c --- ffmpeg-7.1.1.old/libavcodec/amfenc_av1.c 2025-03-24 10:54:38.133756381 +0100 +++ ffmpeg-7.1.1/libavcodec/amfenc_av1.c 2025-03-24 10:54:39.277830845 +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-7.1.1.old/libavcodec/amfenc.c ffmpeg-7.1.1/libavcodec/amfenc.c --- ffmpeg-7.1.1.old/libavcodec/amfenc.c 2025-03-24 10:54:38.032755166 +0100 +++ ffmpeg-7.1.1/libavcodec/amfenc.c 2025-03-24 10:54:39.279550845 +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-7.1.1.old/libavcodec/amfenc.h ffmpeg-7.1.1/libavcodec/amfenc.h --- ffmpeg-7.1.1.old/libavcodec/amfenc.h 2025-03-24 10:54:37.990754660 +0100 +++ ffmpeg-7.1.1/libavcodec/amfenc.h 2025-03-24 10:54:39.277763348 +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-7.1.1.old/libavcodec/amfenc_h264.c ffmpeg-7.1.1/libavcodec/amfenc_h264.c --- ffmpeg-7.1.1.old/libavcodec/amfenc_h264.c 2025-03-24 10:54:38.175756886 +0100 +++ ffmpeg-7.1.1/libavcodec/amfenc_h264.c 2025-03-24 10:54:39.277925894 +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-7.1.1.old/libavcodec/amfenc_hevc.c ffmpeg-7.1.1/libavcodec/amfenc_hevc.c --- ffmpeg-7.1.1.old/libavcodec/amfenc_hevc.c 2025-03-24 10:54:37.991754672 +0100 +++ ffmpeg-7.1.1/libavcodec/amfenc_hevc.c 2025-03-24 10:54:39.278019962 +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-7.1.1.old/libavcodec/av1dec.c ffmpeg-7.1.1/libavcodec/av1dec.c --- ffmpeg-7.1.1.old/libavcodec/av1dec.c 2025-03-24 10:54:38.093755899 +0100 +++ ffmpeg-7.1.1/libavcodec/av1dec.c 2025-03-24 10:54:39.273932914 +0100 @@ -1002,6 +1002,8 @@ 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 +1016,18 @@ 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; diff -Naur ffmpeg-7.1.1.old/libavcodec/bsf/hevc_mp4toannexb.c ffmpeg-7.1.1/libavcodec/bsf/hevc_mp4toannexb.c --- ffmpeg-7.1.1.old/libavcodec/bsf/hevc_mp4toannexb.c 2025-03-24 10:54:38.075755683 +0100 +++ ffmpeg-7.1.1/libavcodec/bsf/hevc_mp4toannexb.c 2025-03-24 10:54:39.273106358 +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-7.1.1.old/libavcodec/ccaption_dec.c ffmpeg-7.1.1/libavcodec/ccaption_dec.c --- ffmpeg-7.1.1.old/libavcodec/ccaption_dec.c 2025-03-24 10:54:38.169756814 +0100 +++ ffmpeg-7.1.1/libavcodec/ccaption_dec.c 2025-03-24 10:54:39.268544045 +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-7.1.1.old/libavcodec/dvdsubdec.c ffmpeg-7.1.1/libavcodec/dvdsubdec.c --- ffmpeg-7.1.1.old/libavcodec/dvdsubdec.c 2025-03-24 10:54:37.975754480 +0100 +++ ffmpeg-7.1.1/libavcodec/dvdsubdec.c 2025-03-24 10:54:39.267391144 +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-7.1.1.old/libavcodec/libdav1d.c ffmpeg-7.1.1/libavcodec/libdav1d.c --- ffmpeg-7.1.1.old/libavcodec/libdav1d.c 2025-03-24 10:54:38.090755863 +0100 +++ ffmpeg-7.1.1/libavcodec/libdav1d.c 2025-03-24 10:54:39.274085311 +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-7.1.1.old/libavcodec/mfenc.c ffmpeg-7.1.1/libavcodec/mfenc.c --- ffmpeg-7.1.1.old/libavcodec/mfenc.c 2025-03-24 10:54:38.125756284 +0100 +++ ffmpeg-7.1.1/libavcodec/mfenc.c 2025-03-24 10:54:39.282547759 +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-7.1.1.old/libavcodec/mf_utils.c ffmpeg-7.1.1/libavcodec/mf_utils.c --- ffmpeg-7.1.1.old/libavcodec/mf_utils.c 2025-03-24 10:54:38.086755815 +0100 +++ ffmpeg-7.1.1/libavcodec/mf_utils.c 2025-03-24 10:54:39.282392106 +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-7.1.1.old/libavcodec/mf_utils.h ffmpeg-7.1.1/libavcodec/mf_utils.h --- ffmpeg-7.1.1.old/libavcodec/mf_utils.h 2025-03-24 10:54:38.061755515 +0100 +++ ffmpeg-7.1.1/libavcodec/mf_utils.h 2025-03-24 10:54:39.282464482 +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-7.1.1.old/libavcodec/pgssubdec.c ffmpeg-7.1.1/libavcodec/pgssubdec.c --- ffmpeg-7.1.1.old/libavcodec/pgssubdec.c 2025-03-24 10:54:37.986754612 +0100 +++ ffmpeg-7.1.1/libavcodec/pgssubdec.c 2025-03-24 10:54:39.280528236 +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-7.1.1.old/libavcodec/videotoolbox.c ffmpeg-7.1.1/libavcodec/videotoolbox.c --- ffmpeg-7.1.1.old/libavcodec/videotoolbox.c 2025-03-24 10:54:38.027755105 +0100 +++ ffmpeg-7.1.1/libavcodec/videotoolbox.c 2025-03-24 10:54:39.271196851 +0100 @@ -916,6 +916,23 @@ 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-7.1.1.old/libavformat/isom.h ffmpeg-7.1.1/libavformat/isom.h --- ffmpeg-7.1.1.old/libavformat/isom.h 2025-03-24 10:54:38.366759184 +0100 +++ ffmpeg-7.1.1/libavformat/isom.h 2025-03-24 10:54:39.261252302 +0100 @@ -272,6 +272,9 @@ MOVEncryptionIndex *encryption_index; } cenc; + int has_fallback; // Audio fallback track + int fallback; + struct IAMFDemuxContext *iamf; } MOVStreamContext; diff -Naur ffmpeg-7.1.1.old/libavformat/matroskaenc.c ffmpeg-7.1.1/libavformat/matroskaenc.c --- ffmpeg-7.1.1.old/libavformat/matroskaenc.c 2025-03-24 10:54:38.396759545 +0100 +++ ffmpeg-7.1.1/libavformat/matroskaenc.c 2025-03-24 10:54:39.269651231 +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-7.1.1.old/libavformat/mov.c ffmpeg-7.1.1/libavformat/mov.c --- ffmpeg-7.1.1.old/libavformat/mov.c 2025-03-24 10:54:38.353759028 +0100 +++ ffmpeg-7.1.1/libavformat/mov.c 2025-03-24 10:54:39.261629363 +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" @@ -342,6 +343,73 @@ 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 +422,7 @@ 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; @@ -367,15 +436,33 @@ case MKTAG( 'c','a','t','g'): key = "category"; 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','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'): @@ -386,6 +473,7 @@ 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"; @@ -445,6 +533,15 @@ case MKTAG(0xa9,'w','r','n'): key = "warning"; break; case MKTAG(0xa9,'w','r','t'): key = "composer"; 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 +668,23 @@ } 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, µ) == 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, µ) == 3) { - c->handbrake_version = 1000000*major + 1000*minor + micro; - } + av_dict_set(&metadata, key2, str, 0); } } @@ -9081,6 +9184,23 @@ 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 }, @@ -9183,6 +9303,7 @@ { 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 }, @@ -10367,6 +10488,23 @@ 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-7.1.1.old/libavformat/movenc.c ffmpeg-7.1.1/libavformat/movenc.c --- ffmpeg-7.1.1.old/libavformat/movenc.c 2025-03-24 10:54:38.381759365 +0100 +++ ffmpeg-7.1.1/libavformat/movenc.c 2025-03-24 10:54:39.258769917 +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" @@ -4055,6 +4056,41 @@ 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 +4165,23 @@ 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) @@ -4727,35 +4778,6 @@ 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 +4816,14 @@ 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-7.1.1.old/libavutil/frame.c ffmpeg-7.1.1/libavutil/frame.c --- ffmpeg-7.1.1.old/libavutil/frame.c 2025-03-24 10:54:38.298758366 +0100 +++ ffmpeg-7.1.1/libavutil/frame.c 2025-03-24 10:54:39.274197994 +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-7.1.1.old/libavutil/frame.h ffmpeg-7.1.1/libavutil/frame.h --- ffmpeg-7.1.1.old/libavutil/frame.h 2025-03-24 10:54:38.287758234 +0100 +++ ffmpeg-7.1.1/libavutil/frame.h 2025-03-24 10:54:39.274296419 +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-7.1.1.old/libswscale/swscale_unscaled.c ffmpeg-7.1.1/libswscale/swscale_unscaled.c --- ffmpeg-7.1.1.old/libswscale/swscale_unscaled.c 2025-03-24 10:54:38.201757199 +0100 +++ ffmpeg-7.1.1/libswscale/swscale_unscaled.c 2025-03-24 10:54:39.272233834 +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];