解码器主要就是输入NALU,输出YUV数据
AVCodec ff_h264_decoder = { .name = "h264", .long_name = NULL_IF_CONFIG_SMALL("H.264 / AVC / MPEG-4 AVC / MPEG-4 part 10"), .type = AVMEDIA_TYPE_VIDEO, .id = AV_CODEC_ID_H264, .priv_data_size = sizeof(H264Context), .init = h264_decode_init, .close = h264_decode_end, .decode = h264_decode_frame, .capabilities = /*AV_CODEC_CAP_DRAW_HORIZ_BAND |*/ AV_CODEC_CAP_DR1 | AV_CODEC_CAP_DELAY | AV_CODEC_CAP_SLICE_THREADS | AV_CODEC_CAP_FRAME_THREADS, .hw_configs = (const AVCodecHWConfigInternal*[]) { #if CONFIG_H264_DXVA2_HWACCEL HWACCEL_DXVA2(h264), #endif #if CONFIG_H264_D3D11VA_HWACCEL HWACCEL_D3D11VA(h264), #endif #if CONFIG_H264_D3D11VA2_HWACCEL HWACCEL_D3D11VA2(h264), #endif #if CONFIG_H264_NVDEC_HWACCEL HWACCEL_NVDEC(h264), #endif #if CONFIG_H264_VAAPI_HWACCEL HWACCEL_VAAPI(h264), #endif #if CONFIG_H264_VDPAU_HWACCEL HWACCEL_VDPAU(h264), #endif #if CONFIG_H264_VIDEOTOOLBOX_HWACCEL HWACCEL_VIDEOTOOLBOX(h264), #endif NULL }, .caps_internal = FF_CODEC_CAP_INIT_THREADSAFE | FF_CODEC_CAP_EXPORTS_CROPPING, .flush = flush_dpb, .init_thread_copy = ONLY_IF_THREADS_ENABLED(decode_init_thread_copy), .update_thread_context = ONLY_IF_THREADS_ENABLED(ff_h264_update_thread_context), .profiles = NULL_IF_CONFIG_SMALL(ff_h264_profiles), .priv_class = &h264_class, };
h264_decode_init()函数
static av_cold int h264_decode_init(AVCodecContext *avctx) { H264Context *h = avctx->priv_data; int ret; // 基本参数初始化 ret = h264_init_context(avctx, h); if (ret < 0) return ret; // 初始化熵编码器 ret = ff_thread_once(&h264_vlc_init, ff_h264_decode_init_vlc); if (ret != 0) { av_log(avctx, AV_LOG_ERROR, "pthread_once has failed."); return AVERROR_UNKNOWN; } if (avctx->ticks_per_frame == 1) { if(h->avctx->time_base.den < INT_MAX/2) { h->avctx->time_base.den *= 2; } else h->avctx->time_base.num /= 2; } avctx->ticks_per_frame = 2; // 初始化extradata if (avctx->extradata_size > 0 && avctx->extradata) { ret = ff_h264_decode_extradata(avctx->extradata, avctx->extradata_size, &h->ps, &h->is_avc, &h->nal_length_size, avctx->err_recognition, avctx); if (ret < 0) { h264_decode_end(avctx); return ret; } } if (h->ps.sps && h->ps.sps->bitstream_restriction_flag && h->avctx->has_b_frames < h->ps.sps->num_reorder_frames) { h->avctx->has_b_frames = h->ps.sps->num_reorder_frames; } avctx->internal->allocate_progress = 1; ff_h264_flush_change(h); if (h->enable_er < 0 && (avctx->active_thread_type & FF_THREAD_SLICE)) h->enable_er = 0; if (h->enable_er && (avctx->active_thread_type & FF_THREAD_SLICE)) { av_log(avctx, AV_LOG_WARNING, "Error resilience with slice threads is enabled. It is unsafe and unsupported and may crash. " "Use it at your own risk\n"); } return 0; }
h264_decode_init主要是一些基本参数的初始化,还有熵编码器初始化和解析extradata。
h264_decode_frame()函数
static int h264_decode_frame(AVCodecContext *avctx, void *data, int *got_frame, AVPacket *avpkt) { const uint8_t *buf = avpkt->data; int buf_size = avpkt->size; H264Context *h = avctx->priv_data; AVFrame *pict = data; int buf_index; int ret; h->flags = avctx->flags; h->setup_finished = 0; h->nb_slice_ctx_queued = 0; ff_h264_unref_picture(h, &h->last_pic_for_ec); /* end of stream, output what is still in the buffers */ if (buf_size == 0) return send_next_delayed_frame(h, pict, got_frame, 0); if (h->is_avc && av_packet_get_side_data(avpkt, AV_PKT_DATA_NEW_EXTRADATA, NULL)) { int side_size; uint8_t *side = av_packet_get_side_data(avpkt, AV_PKT_DATA_NEW_EXTRADATA, &side_size); if (is_extra(side, side_size)) ff_h264_decode_extradata(side, side_size, &h->ps, &h->is_avc, &h->nal_length_size, avctx->err_recognition, avctx); } if (h->is_avc && buf_size >= 9 && buf[0]==1 && buf[2]==0 && (buf[4]&0xFC)==0xFC) { if (is_extra(buf, buf_size)) return ff_h264_decode_extradata(buf, buf_size, &h->ps, &h->is_avc, &h->nal_length_size, avctx->err_recognition, avctx); } // 解析nalu buf_index = decode_nal_units(h, buf, buf_size); if (buf_index < 0) return AVERROR_INVALIDDATA; if (!h->cur_pic_ptr && h->nal_unit_type == H264_NAL_END_SEQUENCE) { av_assert0(buf_index <= buf_size); return send_next_delayed_frame(h, pict, got_frame, buf_index); } if (!(avctx->flags2 & AV_CODEC_FLAG2_CHUNKS) && (!h->cur_pic_ptr || !h->has_slice)) { if (avctx->skip_frame >= AVDISCARD_NONREF || buf_size >= 4 && !memcmp("Q264", buf, 4)) return buf_size; av_log(avctx, AV_LOG_ERROR, "no frame!\n"); return AVERROR_INVALIDDATA; } if (!(avctx->flags2 & AV_CODEC_FLAG2_CHUNKS) || (h->mb_y >= h->mb_height && h->mb_height)) { if ((ret = ff_h264_field_end(h, &h->slice_ctx[0], 0)) < 0) return ret; /* Wait for second field. */ // 输出到pict if (h->next_output_pic) { ret = finalize_frame(h, pict, h->next_output_pic, got_frame); if (ret < 0) return ret; } } av_assert0(pict->buf[0] || !*got_frame); ff_h264_unref_picture(h, &h->last_pic_for_ec); return get_consumed_bytes(buf_index, buf_size); }
这里主要就是解析nalu,将结果输出到pict
static int decode_nal_units(H264Context *h, const uint8_t *buf, int buf_size) { AVCodecContext *const avctx = h->avctx; int nals_needed = 0; ///< number of NALs that need decoding before the next frame thread starts int idr_cleared=0; int i, ret = 0; h->has_slice = 0; h->nal_unit_type= 0; if (!(avctx->flags2 & AV_CODEC_FLAG2_CHUNKS)) { h->current_slice = 0; if (!h->first_field) { h->cur_pic_ptr = NULL; ff_h264_sei_uninit(&h->sei); } } if (h->nal_length_size == 4) { if (buf_size > 8 && AV_RB32(buf) == 1 && AV_RB32(buf+5) > (unsigned)buf_size) { h->is_avc = 0; }else if(buf_size > 3 && AV_RB32(buf) > 1 && AV_RB32(buf) <= (unsigned)buf_size) h->is_avc = 1; } ret = ff_h2645_packet_split(&h->pkt, buf, buf_size, avctx, h->is_avc, h->nal_length_size, avctx->codec_id, avctx->flags2 & AV_CODEC_FLAG2_FAST, 0); if (ret < 0) { av_log(avctx, AV_LOG_ERROR, "Error splitting the input into NAL units.\n"); return ret; } if (avctx->active_thread_type & FF_THREAD_FRAME) nals_needed = get_last_needed_nal(h); if (nals_needed < 0) return nals_needed; for (i = 0; i < h->pkt.nb_nals; i++) { H2645NAL *nal = &h->pkt.nals[i]; int max_slice_ctx, err; if (avctx->skip_frame >= AVDISCARD_NONREF && nal->ref_idc == 0 && nal->type != H264_NAL_SEI) continue; // FIXME these should stop being context-global variables h->nal_ref_idc = nal->ref_idc; h->nal_unit_type = nal->type; err = 0; switch (nal->type) { case H264_NAL_IDR_SLICE: if ((nal->data[1] & 0xFC) == 0x98) { av_log(h->avctx, AV_LOG_ERROR, "Invalid inter IDR frame\n"); h->next_outputed_poc = INT_MIN; ret = -1; goto end; } if(!idr_cleared) { idr(h); // FIXME ensure we don‘t lose some frames if there is reordering } idr_cleared = 1; h->has_recovery_point = 1; case H264_NAL_SLICE: h->has_slice = 1; if ((err = ff_h264_queue_decode_slice(h, nal))) { H264SliceContext *sl = h->slice_ctx + h->nb_slice_ctx_queued; sl->ref_count[0] = sl->ref_count[1] = 0; break; } if (h->current_slice == 1) { if (avctx->active_thread_type & FF_THREAD_FRAME && i >= nals_needed && !h->setup_finished && h->cur_pic_ptr) { ff_thread_finish_setup(avctx); h->setup_finished = 1; } if (h->avctx->hwaccel && (ret = h->avctx->hwaccel->start_frame(h->avctx, buf, buf_size)) < 0) goto end; } max_slice_ctx = avctx->hwaccel ? 1 : h->nb_slice_ctx; if (h->nb_slice_ctx_queued == max_slice_ctx) { if (h->avctx->hwaccel) { ret = avctx->hwaccel->decode_slice(avctx, nal->raw_data, nal->raw_size); h->nb_slice_ctx_queued = 0; } else ret = ff_h264_execute_decode_slices(h); if (ret < 0 && (h->avctx->err_recognition & AV_EF_EXPLODE)) goto end; } break; case H264_NAL_DPA: case H264_NAL_DPB: case H264_NAL_DPC: avpriv_request_sample(avctx, "data partitioning"); break; case H264_NAL_SEI: ret = ff_h264_sei_decode(&h->sei, &nal->gb, &h->ps, avctx); h->has_recovery_point = h->has_recovery_point || h->sei.recovery_point.recovery_frame_cnt != -1; if (avctx->debug & FF_DEBUG_GREEN_MD) debug_green_metadata(&h->sei.green_metadata, h->avctx); if (ret < 0 && (h->avctx->err_recognition & AV_EF_EXPLODE)) goto end; break; case H264_NAL_SPS: { GetBitContext tmp_gb = nal->gb; if (avctx->hwaccel && avctx->hwaccel->decode_params) { ret = avctx->hwaccel->decode_params(avctx, nal->type, nal->raw_data, nal->raw_size); if (ret < 0) goto end; } if (ff_h264_decode_seq_parameter_set(&tmp_gb, avctx, &h->ps, 0) >= 0) break; av_log(h->avctx, AV_LOG_DEBUG, "SPS decoding failure, trying again with the complete NAL\n"); init_get_bits8(&tmp_gb, nal->raw_data + 1, nal->raw_size - 1); if (ff_h264_decode_seq_parameter_set(&tmp_gb, avctx, &h->ps, 0) >= 0) break; ff_h264_decode_seq_parameter_set(&nal->gb, avctx, &h->ps, 1); break; } case H264_NAL_PPS: if (avctx->hwaccel && avctx->hwaccel->decode_params) { ret = avctx->hwaccel->decode_params(avctx, nal->type, nal->raw_data, nal->raw_size); if (ret < 0) goto end; } ret = ff_h264_decode_picture_parameter_set(&nal->gb, avctx, &h->ps, nal->size_bits); if (ret < 0 && (h->avctx->err_recognition & AV_EF_EXPLODE)) goto end; break; case H264_NAL_AUD: case H264_NAL_END_SEQUENCE: case H264_NAL_END_STREAM: case H264_NAL_FILLER_DATA: case H264_NAL_SPS_EXT: case H264_NAL_AUXILIARY_SLICE: break; default: av_log(avctx, AV_LOG_DEBUG, "Unknown NAL code: %d (%d bits)\n", nal->type, nal->size_bits); } if (err < 0) { av_log(h->avctx, AV_LOG_ERROR, "decode_slice_header error\n"); } } ret = ff_h264_execute_decode_slices(h); if (ret < 0 && (h->avctx->err_recognition & AV_EF_EXPLODE)) goto end; // set decode_error_flags to allow users to detect concealed decoding errors if ((ret < 0 || h->slice_ctx->er.error_occurred) && h->cur_pic_ptr) { h->cur_pic_ptr->f->decode_error_flags |= FF_DECODE_ERROR_DECODE_SLICES; } ret = 0; end: #if CONFIG_ERROR_RESILIENCE /* * FIXME: Error handling code does not seem to support interlaced * when slices span multiple rows * The ff_er_add_slice calls don‘t work right for bottom * fields; they cause massive erroneous error concealing * Error marking covers both fields (top and bottom). * This causes a mismatched s->error_count * and a bad error table. Further, the error count goes to * INT_MAX when called for bottom field, because mb_y is * past end by one (callers fault) and resync_mb_y != 0 * causes problems for the first MB line, too. */ if (!FIELD_PICTURE(h) && h->current_slice && h->ps.sps == (const SPS*)h->ps.sps_list[h->ps.pps->sps_id]->data && h->enable_er) { H264SliceContext *sl = h->slice_ctx; int use_last_pic = h->last_pic_for_ec.f->buf[0] && !sl->ref_count[0]; ff_h264_set_erpic(&sl->er.cur_pic, h->cur_pic_ptr); if (use_last_pic) { ff_h264_set_erpic(&sl->er.last_pic, &h->last_pic_for_ec); sl->ref_list[0][0].parent = &h->last_pic_for_ec; memcpy(sl->ref_list[0][0].data, h->last_pic_for_ec.f->data, sizeof(sl->ref_list[0][0].data)); memcpy(sl->ref_list[0][0].linesize, h->last_pic_for_ec.f->linesize, sizeof(sl->ref_list[0][0].linesize)); sl->ref_list[0][0].reference = h->last_pic_for_ec.reference; } else if (sl->ref_count[0]) { ff_h264_set_erpic(&sl->er.last_pic, sl->ref_list[0][0].parent); } else ff_h264_set_erpic(&sl->er.last_pic, NULL); if (sl->ref_count[1]) ff_h264_set_erpic(&sl->er.next_pic, sl->ref_list[1][0].parent); sl->er.ref_count = sl->ref_count[0]; ff_er_frame_end(&sl->er); if (use_last_pic) memset(&sl->ref_list[0][0], 0, sizeof(sl->ref_list[0][0])); } #endif /* CONFIG_ERROR_RESILIENCE */ /* clean up */ if (h->cur_pic_ptr && !h->droppable && h->has_slice) { ff_thread_report_progress(&h->cur_pic_ptr->tf, INT_MAX, h->picture_structure == PICT_BOTTOM_FIELD); } return (ret < 0) ? ret : buf_size; }
基于不同的类型,分别解析。
h264_decode_end()函数
static av_cold int h264_decode_end(AVCodecContext *avctx) { H264Context *h = avctx->priv_data; int i; ff_h264_remove_all_refs(h); ff_h264_free_tables(h); for (i = 0; i < H264_MAX_PICTURE_COUNT; i++) { ff_h264_unref_picture(h, &h->DPB[i]); av_frame_free(&h->DPB[i].f); } memset(h->delayed_pic, 0, sizeof(h->delayed_pic)); h->cur_pic_ptr = NULL; av_freep(&h->slice_ctx); h->nb_slice_ctx = 0; ff_h264_sei_uninit(&h->sei); ff_h264_ps_uninit(&h->ps); ff_h2645_packet_uninit(&h->pkt); ff_h264_unref_picture(h, &h->cur_pic); av_frame_free(&h->cur_pic.f); ff_h264_unref_picture(h, &h->last_pic_for_ec); av_frame_free(&h->last_pic_for_ec.f); return 0; }
这里主要就是释放申请的一些内存