From patchwork Mon Oct 17 01:34:40 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Andreas Rheinhardt X-Patchwork-Id: 38757 Delivered-To: ffmpegpatchwork2@gmail.com Received: by 2002:a05:6a20:4a86:b0:9d:28a3:170e with SMTP id fn6csp1375852pzb; Sun, 16 Oct 2022 18:35:20 -0700 (PDT) X-Google-Smtp-Source: AMsMyM5M2QadlURmX4FBqUvIbhNER6lzZQJuctxSFQSyldP6zJJt6l9eaI0jf4mFydUweTJ4B1b2 X-Received: by 2002:a17:907:3ea9:b0:78d:fdf0:88fe with SMTP id hs41-20020a1709073ea900b0078dfdf088femr7051084ejc.667.1665970519796; Sun, 16 Oct 2022 18:35:19 -0700 (PDT) Return-Path: Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org. [79.124.17.100]) by mx.google.com with ESMTP id hs40-20020a1709073ea800b0078df2f22f80si9422925ejc.647.2022.10.16.18.35.19; Sun, 16 Oct 2022 18:35:19 -0700 (PDT) Received-SPF: pass (google.com: domain of ffmpeg-devel-bounces@ffmpeg.org designates 79.124.17.100 as permitted sender) client-ip=79.124.17.100; Authentication-Results: mx.google.com; dkim=neutral (body hash did not verify) header.i=@outlook.com header.s=selector1 header.b=CPskxOkN; arc=fail (body hash mismatch); spf=pass (google.com: domain of ffmpeg-devel-bounces@ffmpeg.org designates 79.124.17.100 as permitted sender) smtp.mailfrom=ffmpeg-devel-bounces@ffmpeg.org; dmarc=fail (p=NONE sp=QUARANTINE dis=NONE) header.from=outlook.com Received: from [127.0.1.1] (localhost [127.0.0.1]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id 4356868BCA6; Mon, 17 Oct 2022 04:35:16 +0300 (EEST) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from EUR04-DB3-obe.outbound.protection.outlook.com (mail-oln040092074017.outbound.protection.outlook.com [40.92.74.17]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTPS id 85F1368BC9B for ; Mon, 17 Oct 2022 04:35:09 +0300 (EEST) ARC-Seal: i=1; a=rsa-sha256; s=arcselector9901; d=microsoft.com; cv=none; b=L2Ih1Z7tPIgyygPWlSNEdIDLugTuZ2UD/HJoBfCUrZB4aAeT+zIhMe4SZRFE+jI3XRLSuBlGSUUoOSQsCsjNfrw5Qu6XLCk7/Ql44L4wPpfL+LuWOkeDlXZIAUcJBvBrdVCvcbrHT91/JdywIr9VBd+lp7mAl18mXqiBCdya9vZJAj9bjmwZVzStvD6qw/idS+IG6N+e673a3UlwS4iQf0EwAliMEhjcaAwDZDeC1Sh2tMH6lAABHhgAQpiBiP16CWYGisuIaUNcnEJbG14heEBgHohRhPhAZLvHrNUHjJMI6sRkr2gYMUROSL5kARscQsldDRs7oEwtW3SGOHvpxw== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=microsoft.com; s=arcselector9901; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-AntiSpam-MessageData-ChunkCount:X-MS-Exchange-AntiSpam-MessageData-0:X-MS-Exchange-AntiSpam-MessageData-1; bh=1oZazUSKme05DuALMHEXr03H65SEnrs+vD2mCV4Nirk=; b=kWQpPxD70oFUtgWVxVNq34u+r1qvkp3PBYsXFOIZ35wPLmZqHnUv4/9fGu/kABVFLlPj/ZpKkuNhaslYb+jMm3kJsaoweGhi3T32D4H4mFAC0V07ll5+bgcQ/IXQjRxWq5D0uPgVgt7HIMIAZ0vbKCgU6Hv0se38+K+svsetejQA6JWONEL3kctbT9AtCFHlfagLpAsdZWs0SeF6cRQlFd4eS5tiOEJeNgFU7zpJx5q2MT4Pgrt/A/W2zJEsVNH1nFzxwAuyI3FI/uRwfcbZEiETTW25hUFgD7dBR1fMxRPMbj0ItEqX+RvAuS0C5ayyefGybmlYL4kvGAXvM82fYw== ARC-Authentication-Results: i=1; mx.microsoft.com 1; spf=none; dmarc=none; dkim=none; arc=none DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=outlook.com; s=selector1; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-SenderADCheck; bh=1oZazUSKme05DuALMHEXr03H65SEnrs+vD2mCV4Nirk=; b=CPskxOkN/pnbZ7yoRD/jtzRs7IifXIiIwpdJ/hHR/192bD076ZoeUYVVXjxKvQ10fmOT/pjNJOVI+Fer+ltNfTU2zSX0/FLYSJ4Qm5lYNQkNc76bzkRkjftIvsRf3LzRCcGdft49ZE70M85dwk2ZXlW4Biuz69rKENtrCST2dfPQuRrcNhSNjslYeJtdsYsdCTKnT/XVl2P9efKxZ3sntlzklYZzF7V13bthzy/iTgrNQPMCU2ysApQSnfkumGQYOeEvGStkh9IyScwdhleYfE6J62u3rV0upnF7N25TmpI2jQwPJCIo71NjT9D8AfUX01QCSgs5Og8ExvBvWL+XHg== Received: from GV1P250MB0737.EURP250.PROD.OUTLOOK.COM (2603:10a6:150:8e::17) by DU0P250MB0916.EURP250.PROD.OUTLOOK.COM (2603:10a6:10:3ec::6) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.5723.26; Mon, 17 Oct 2022 01:35:07 +0000 Received: from GV1P250MB0737.EURP250.PROD.OUTLOOK.COM ([fe80::20c0:9c9e:87f:79f]) by GV1P250MB0737.EURP250.PROD.OUTLOOK.COM ([fe80::20c0:9c9e:87f:79f%2]) with mapi id 15.20.5723.033; Mon, 17 Oct 2022 01:35:07 +0000 From: Andreas Rheinhardt To: ffmpeg-devel@ffmpeg.org Date: Mon, 17 Oct 2022 03:34:40 +0200 Message-ID: X-Mailer: git-send-email 2.34.1 In-Reply-To: References: X-TMN: [DEO2b4aaLu/yg6AjhwtNyRWNjp2+KP/WQ9S9Js+AUT4=] X-ClientProxiedBy: FR3P281CA0056.DEUP281.PROD.OUTLOOK.COM (2603:10a6:d10:4b::6) To GV1P250MB0737.EURP250.PROD.OUTLOOK.COM (2603:10a6:150:8e::17) X-Microsoft-Original-Message-ID: <20221017013442.944352-3-andreas.rheinhardt@outlook.com> MIME-Version: 1.0 X-MS-Exchange-MessageSentRepresentingType: 1 X-MS-PublicTrafficType: Email X-MS-TrafficTypeDiagnostic: GV1P250MB0737:EE_|DU0P250MB0916:EE_ X-MS-Office365-Filtering-Correlation-Id: 493142e8-241b-48de-0814-08daafdfd264 X-MS-Exchange-SLBlob-MailProps: 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 X-Microsoft-Antispam: BCL:0; X-Microsoft-Antispam-Message-Info: 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 X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1 X-MS-Exchange-AntiSpam-MessageData-0: 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 X-OriginatorOrg: outlook.com X-MS-Exchange-CrossTenant-Network-Message-Id: 493142e8-241b-48de-0814-08daafdfd264 X-MS-Exchange-CrossTenant-AuthSource: GV1P250MB0737.EURP250.PROD.OUTLOOK.COM X-MS-Exchange-CrossTenant-AuthAs: Internal X-MS-Exchange-CrossTenant-OriginalArrivalTime: 17 Oct 2022 01:35:07.1329 (UTC) X-MS-Exchange-CrossTenant-FromEntityHeader: Hosted X-MS-Exchange-CrossTenant-Id: 84df9e7f-e9f6-40af-b435-aaaaaaaaaaaa X-MS-Exchange-CrossTenant-RMS-PersistedConsumerOrg: 00000000-0000-0000-0000-000000000000 X-MS-Exchange-Transport-CrossTenantHeadersStamped: DU0P250MB0916 Subject: [FFmpeg-devel] [PATCH 4/6] avcodec/mpegvideo: Split ff_mpv_reconstruct_mb() into de/encoder part X-BeenThere: ffmpeg-devel@ffmpeg.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: FFmpeg development discussions and patches List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Reply-To: FFmpeg development discussions and patches Cc: Andreas Rheinhardt Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" X-TUID: IJxQE62qvcr4 This has the advantage of not having to check for whether a given MpegEncContext is actually a decoder or an encoder context at runtime. To do so, mpv_reconstruct_mb_internal() is moved into a new template file that is included by both mpegvideo_enc.c and mpegvideo_dec.c; the decoder-only code (mainly lowres) are also moved to mpegvideo_dec.c. The is_encoder checks are changed to #if IS_ENCODER in order to avoid having to include headers for decoder-only functions in mpegvideo_enc.c. This approach also has the advantage that it is easy to adapt mpv_reconstruct_mb_internal() to using different structures for decoders and encoders (e.g. the check for whether a macroblock should be processed for the encoder or not uses MpegEncContext elements that make no sense for decoders and should not be part of their context). Signed-off-by: Andreas Rheinhardt --- libavcodec/mpeg_er.c | 1 + libavcodec/mpegvideo.c | 737 ----------------------- libavcodec/mpegvideo.h | 2 - libavcodec/mpegvideo_dec.c | 467 ++++++++++++++ libavcodec/mpegvideo_enc.c | 26 +- libavcodec/mpegvideodec.h | 1 + libavcodec/mpv_reconstruct_mb_template.c | 300 +++++++++ 7 files changed, 792 insertions(+), 742 deletions(-) create mode 100644 libavcodec/mpv_reconstruct_mb_template.c diff --git a/libavcodec/mpeg_er.c b/libavcodec/mpeg_er.c index 02f407d8ea..8034963253 100644 --- a/libavcodec/mpeg_er.c +++ b/libavcodec/mpeg_er.c @@ -18,6 +18,7 @@ #include "error_resilience.h" #include "mpegvideo.h" +#include "mpegvideodec.h" #include "mpeg_er.h" static void set_erpic(ERPicture *dst, Picture *src) diff --git a/libavcodec/mpegvideo.c b/libavcodec/mpegvideo.c index 43f3ec5a47..850d8f2e94 100644 --- a/libavcodec/mpegvideo.c +++ b/libavcodec/mpegvideo.c @@ -27,8 +27,6 @@ * The simplest mpeg encoder (well, it was the simplest!). */ -#include "config_components.h" - #include "libavutil/attributes.h" #include "libavutil/avassert.h" #include "libavutil/imgutils.h" @@ -42,12 +40,7 @@ #include "mpeg_er.h" #include "mpegutils.h" #include "mpegvideo.h" -#include "mpeg4videodec.h" #include "mpegvideodata.h" -#include "qpeldsp.h" -#include "threadframe.h" -#include "wmv2dec.h" -#include static void dct_unquantize_mpeg1_intra_c(MpegEncContext *s, int16_t *block, int n, int qscale) @@ -811,456 +804,6 @@ void ff_mpv_common_end(MpegEncContext *s) } -static inline int hpel_motion_lowres(MpegEncContext *s, - uint8_t *dest, const uint8_t *src, - int field_based, int field_select, - int src_x, int src_y, - int width, int height, ptrdiff_t stride, - int h_edge_pos, int v_edge_pos, - int w, int h, const h264_chroma_mc_func *pix_op, - int motion_x, int motion_y) -{ - const int lowres = s->avctx->lowres; - const int op_index = FFMIN(lowres, 3); - const int s_mask = (2 << lowres) - 1; - int emu = 0; - int sx, sy; - - if (s->quarter_sample) { - motion_x /= 2; - motion_y /= 2; - } - - sx = motion_x & s_mask; - sy = motion_y & s_mask; - src_x += motion_x >> lowres + 1; - src_y += motion_y >> lowres + 1; - - src += src_y * stride + src_x; - - if ((unsigned)src_x > FFMAX( h_edge_pos - (!!sx) - w, 0) || - (unsigned)src_y > FFMAX((v_edge_pos >> field_based) - (!!sy) - h, 0)) { - s->vdsp.emulated_edge_mc(s->sc.edge_emu_buffer, src, - s->linesize, s->linesize, - w + 1, (h + 1) << field_based, - src_x, src_y * (1 << field_based), - h_edge_pos, v_edge_pos); - src = s->sc.edge_emu_buffer; - emu = 1; - } - - sx = (sx << 2) >> lowres; - sy = (sy << 2) >> lowres; - if (field_select) - src += s->linesize; - pix_op[op_index](dest, src, stride, h, sx, sy); - return emu; -} - -/* apply one mpeg motion vector to the three components */ -static av_always_inline void mpeg_motion_lowres(MpegEncContext *s, - uint8_t *dest_y, - uint8_t *dest_cb, - uint8_t *dest_cr, - int field_based, - int bottom_field, - int field_select, - uint8_t *const *ref_picture, - const h264_chroma_mc_func *pix_op, - int motion_x, int motion_y, - int h, int mb_y) -{ - const uint8_t *ptr_y, *ptr_cb, *ptr_cr; - int mx, my, src_x, src_y, uvsrc_x, uvsrc_y, sx, sy, uvsx, uvsy; - ptrdiff_t uvlinesize, linesize; - const int lowres = s->avctx->lowres; - const int op_index = FFMIN(lowres-1+s->chroma_x_shift, 3); - const int block_s = 8>>lowres; - const int s_mask = (2 << lowres) - 1; - const int h_edge_pos = s->h_edge_pos >> lowres; - const int v_edge_pos = s->v_edge_pos >> lowres; - linesize = s->current_picture.f->linesize[0] << field_based; - uvlinesize = s->current_picture.f->linesize[1] << field_based; - - // FIXME obviously not perfect but qpel will not work in lowres anyway - if (s->quarter_sample) { - motion_x /= 2; - motion_y /= 2; - } - - if(field_based){ - motion_y += (bottom_field - field_select)*((1 << lowres)-1); - } - - sx = motion_x & s_mask; - sy = motion_y & s_mask; - src_x = s->mb_x * 2 * block_s + (motion_x >> lowres + 1); - src_y = (mb_y * 2 * block_s >> field_based) + (motion_y >> lowres + 1); - - if (s->out_format == FMT_H263) { - uvsx = ((motion_x >> 1) & s_mask) | (sx & 1); - uvsy = ((motion_y >> 1) & s_mask) | (sy & 1); - uvsrc_x = src_x >> 1; - uvsrc_y = src_y >> 1; - } else if (s->out_format == FMT_H261) { - // even chroma mv's are full pel in H261 - mx = motion_x / 4; - my = motion_y / 4; - uvsx = (2 * mx) & s_mask; - uvsy = (2 * my) & s_mask; - uvsrc_x = s->mb_x * block_s + (mx >> lowres); - uvsrc_y = mb_y * block_s + (my >> lowres); - } else { - if(s->chroma_y_shift){ - mx = motion_x / 2; - my = motion_y / 2; - uvsx = mx & s_mask; - uvsy = my & s_mask; - uvsrc_x = s->mb_x * block_s + (mx >> lowres + 1); - uvsrc_y = (mb_y * block_s >> field_based) + (my >> lowres + 1); - } else { - if(s->chroma_x_shift){ - //Chroma422 - mx = motion_x / 2; - uvsx = mx & s_mask; - uvsy = motion_y & s_mask; - uvsrc_y = src_y; - uvsrc_x = s->mb_x*block_s + (mx >> (lowres+1)); - } else { - //Chroma444 - uvsx = motion_x & s_mask; - uvsy = motion_y & s_mask; - uvsrc_x = src_x; - uvsrc_y = src_y; - } - } - } - - ptr_y = ref_picture[0] + src_y * linesize + src_x; - ptr_cb = ref_picture[1] + uvsrc_y * uvlinesize + uvsrc_x; - ptr_cr = ref_picture[2] + uvsrc_y * uvlinesize + uvsrc_x; - - if ((unsigned) src_x > FFMAX( h_edge_pos - (!!sx) - 2 * block_s, 0) || uvsrc_y<0 || - (unsigned) src_y > FFMAX((v_edge_pos >> field_based) - (!!sy) - h, 0)) { - s->vdsp.emulated_edge_mc(s->sc.edge_emu_buffer, ptr_y, - linesize >> field_based, linesize >> field_based, - 17, 17 + field_based, - src_x, src_y * (1 << field_based), h_edge_pos, - v_edge_pos); - ptr_y = s->sc.edge_emu_buffer; - if (!CONFIG_GRAY || !(s->avctx->flags & AV_CODEC_FLAG_GRAY)) { - uint8_t *ubuf = s->sc.edge_emu_buffer + 18 * s->linesize; - uint8_t *vbuf =ubuf + 10 * s->uvlinesize; - if (s->workaround_bugs & FF_BUG_IEDGE) - vbuf -= s->uvlinesize; - s->vdsp.emulated_edge_mc(ubuf, ptr_cb, - uvlinesize >> field_based, uvlinesize >> field_based, - 9, 9 + field_based, - uvsrc_x, uvsrc_y * (1 << field_based), - h_edge_pos >> 1, v_edge_pos >> 1); - s->vdsp.emulated_edge_mc(vbuf, ptr_cr, - uvlinesize >> field_based,uvlinesize >> field_based, - 9, 9 + field_based, - uvsrc_x, uvsrc_y * (1 << field_based), - h_edge_pos >> 1, v_edge_pos >> 1); - ptr_cb = ubuf; - ptr_cr = vbuf; - } - } - - // FIXME use this for field pix too instead of the obnoxious hack which changes picture.f->data - if (bottom_field) { - dest_y += s->linesize; - dest_cb += s->uvlinesize; - dest_cr += s->uvlinesize; - } - - if (field_select) { - ptr_y += s->linesize; - ptr_cb += s->uvlinesize; - ptr_cr += s->uvlinesize; - } - - sx = (sx << 2) >> lowres; - sy = (sy << 2) >> lowres; - pix_op[lowres - 1](dest_y, ptr_y, linesize, h, sx, sy); - - if (!CONFIG_GRAY || !(s->avctx->flags & AV_CODEC_FLAG_GRAY)) { - int hc = s->chroma_y_shift ? (h+1-bottom_field)>>1 : h; - uvsx = (uvsx << 2) >> lowres; - uvsy = (uvsy << 2) >> lowres; - if (hc) { - pix_op[op_index](dest_cb, ptr_cb, uvlinesize, hc, uvsx, uvsy); - pix_op[op_index](dest_cr, ptr_cr, uvlinesize, hc, uvsx, uvsy); - } - } - // FIXME h261 lowres loop filter -} - -static inline void chroma_4mv_motion_lowres(MpegEncContext *s, - uint8_t *dest_cb, uint8_t *dest_cr, - uint8_t *const *ref_picture, - const h264_chroma_mc_func * pix_op, - int mx, int my) -{ - const int lowres = s->avctx->lowres; - const int op_index = FFMIN(lowres, 3); - const int block_s = 8 >> lowres; - const int s_mask = (2 << lowres) - 1; - const int h_edge_pos = s->h_edge_pos >> lowres + 1; - const int v_edge_pos = s->v_edge_pos >> lowres + 1; - int emu = 0, src_x, src_y, sx, sy; - ptrdiff_t offset; - const uint8_t *ptr; - - if (s->quarter_sample) { - mx /= 2; - my /= 2; - } - - /* In case of 8X8, we construct a single chroma motion vector - with a special rounding */ - mx = ff_h263_round_chroma(mx); - my = ff_h263_round_chroma(my); - - sx = mx & s_mask; - sy = my & s_mask; - src_x = s->mb_x * block_s + (mx >> lowres + 1); - src_y = s->mb_y * block_s + (my >> lowres + 1); - - offset = src_y * s->uvlinesize + src_x; - ptr = ref_picture[1] + offset; - if ((unsigned) src_x > FFMAX(h_edge_pos - (!!sx) - block_s, 0) || - (unsigned) src_y > FFMAX(v_edge_pos - (!!sy) - block_s, 0)) { - s->vdsp.emulated_edge_mc(s->sc.edge_emu_buffer, ptr, - s->uvlinesize, s->uvlinesize, - 9, 9, - src_x, src_y, h_edge_pos, v_edge_pos); - ptr = s->sc.edge_emu_buffer; - emu = 1; - } - sx = (sx << 2) >> lowres; - sy = (sy << 2) >> lowres; - pix_op[op_index](dest_cb, ptr, s->uvlinesize, block_s, sx, sy); - - ptr = ref_picture[2] + offset; - if (emu) { - s->vdsp.emulated_edge_mc(s->sc.edge_emu_buffer, ptr, - s->uvlinesize, s->uvlinesize, - 9, 9, - src_x, src_y, h_edge_pos, v_edge_pos); - ptr = s->sc.edge_emu_buffer; - } - pix_op[op_index](dest_cr, ptr, s->uvlinesize, block_s, sx, sy); -} - -/** - * motion compensation of a single macroblock - * @param s context - * @param dest_y luma destination pointer - * @param dest_cb chroma cb/u destination pointer - * @param dest_cr chroma cr/v destination pointer - * @param dir direction (0->forward, 1->backward) - * @param ref_picture array[3] of pointers to the 3 planes of the reference picture - * @param pix_op halfpel motion compensation function (average or put normally) - * the motion vectors are taken from s->mv and the MV type from s->mv_type - */ -static inline void MPV_motion_lowres(MpegEncContext *s, - uint8_t *dest_y, uint8_t *dest_cb, - uint8_t *dest_cr, - int dir, uint8_t *const *ref_picture, - const h264_chroma_mc_func *pix_op) -{ - int mx, my; - int mb_x, mb_y, i; - const int lowres = s->avctx->lowres; - const int block_s = 8 >>lowres; - - mb_x = s->mb_x; - mb_y = s->mb_y; - - switch (s->mv_type) { - case MV_TYPE_16X16: - mpeg_motion_lowres(s, dest_y, dest_cb, dest_cr, - 0, 0, 0, - ref_picture, pix_op, - s->mv[dir][0][0], s->mv[dir][0][1], - 2 * block_s, mb_y); - break; - case MV_TYPE_8X8: - mx = 0; - my = 0; - for (i = 0; i < 4; i++) { - hpel_motion_lowres(s, dest_y + ((i & 1) + (i >> 1) * - s->linesize) * block_s, - ref_picture[0], 0, 0, - (2 * mb_x + (i & 1)) * block_s, - (2 * mb_y + (i >> 1)) * block_s, - s->width, s->height, s->linesize, - s->h_edge_pos >> lowres, s->v_edge_pos >> lowres, - block_s, block_s, pix_op, - s->mv[dir][i][0], s->mv[dir][i][1]); - - mx += s->mv[dir][i][0]; - my += s->mv[dir][i][1]; - } - - if (!CONFIG_GRAY || !(s->avctx->flags & AV_CODEC_FLAG_GRAY)) - chroma_4mv_motion_lowres(s, dest_cb, dest_cr, ref_picture, - pix_op, mx, my); - break; - case MV_TYPE_FIELD: - if (s->picture_structure == PICT_FRAME) { - /* top field */ - mpeg_motion_lowres(s, dest_y, dest_cb, dest_cr, - 1, 0, s->field_select[dir][0], - ref_picture, pix_op, - s->mv[dir][0][0], s->mv[dir][0][1], - block_s, mb_y); - /* bottom field */ - mpeg_motion_lowres(s, dest_y, dest_cb, dest_cr, - 1, 1, s->field_select[dir][1], - ref_picture, pix_op, - s->mv[dir][1][0], s->mv[dir][1][1], - block_s, mb_y); - } else { - if (s->picture_structure != s->field_select[dir][0] + 1 && - s->pict_type != AV_PICTURE_TYPE_B && !s->first_field) { - ref_picture = s->current_picture_ptr->f->data; - - } - mpeg_motion_lowres(s, dest_y, dest_cb, dest_cr, - 0, 0, s->field_select[dir][0], - ref_picture, pix_op, - s->mv[dir][0][0], - s->mv[dir][0][1], 2 * block_s, mb_y >> 1); - } - break; - case MV_TYPE_16X8: - for (i = 0; i < 2; i++) { - uint8_t *const *ref2picture; - - if (s->picture_structure == s->field_select[dir][i] + 1 || - s->pict_type == AV_PICTURE_TYPE_B || s->first_field) { - ref2picture = ref_picture; - } else { - ref2picture = s->current_picture_ptr->f->data; - } - - mpeg_motion_lowres(s, dest_y, dest_cb, dest_cr, - 0, 0, s->field_select[dir][i], - ref2picture, pix_op, - s->mv[dir][i][0], s->mv[dir][i][1] + - 2 * block_s * i, block_s, mb_y >> 1); - - dest_y += 2 * block_s * s->linesize; - dest_cb += (2 * block_s >> s->chroma_y_shift) * s->uvlinesize; - dest_cr += (2 * block_s >> s->chroma_y_shift) * s->uvlinesize; - } - break; - case MV_TYPE_DMV: - if (s->picture_structure == PICT_FRAME) { - for (i = 0; i < 2; i++) { - int j; - for (j = 0; j < 2; j++) { - mpeg_motion_lowres(s, dest_y, dest_cb, dest_cr, - 1, j, j ^ i, - ref_picture, pix_op, - s->mv[dir][2 * i + j][0], - s->mv[dir][2 * i + j][1], - block_s, mb_y); - } - pix_op = s->h264chroma.avg_h264_chroma_pixels_tab; - } - } else { - for (i = 0; i < 2; i++) { - mpeg_motion_lowres(s, dest_y, dest_cb, dest_cr, - 0, 0, s->picture_structure != i + 1, - ref_picture, pix_op, - s->mv[dir][2 * i][0],s->mv[dir][2 * i][1], - 2 * block_s, mb_y >> 1); - - // after put we make avg of the same block - pix_op = s->h264chroma.avg_h264_chroma_pixels_tab; - - // opposite parity is always in the same - // frame if this is second field - if (!s->first_field) { - ref_picture = s->current_picture_ptr->f->data; - } - } - } - break; - default: - av_assert2(0); - } -} - -/** - * find the lowest MB row referenced in the MVs - */ -static int lowest_referenced_row(MpegEncContext *s, int dir) -{ - int my_max = INT_MIN, my_min = INT_MAX, qpel_shift = !s->quarter_sample; - int my, off, i, mvs; - - if (s->picture_structure != PICT_FRAME || s->mcsel) - goto unhandled; - - switch (s->mv_type) { - case MV_TYPE_16X16: - mvs = 1; - break; - case MV_TYPE_16X8: - mvs = 2; - break; - case MV_TYPE_8X8: - mvs = 4; - break; - default: - goto unhandled; - } - - for (i = 0; i < mvs; i++) { - my = s->mv[dir][i][1]; - my_max = FFMAX(my_max, my); - my_min = FFMIN(my_min, my); - } - - off = ((FFMAX(-my_min, my_max)<> 6; - - return av_clip(s->mb_y + off, 0, s->mb_height - 1); -unhandled: - return s->mb_height-1; -} - -/* put block[] to dest[] */ -static inline void put_dct(MpegEncContext *s, - int16_t *block, int i, uint8_t *dest, int line_size, int qscale) -{ - s->dct_unquantize_intra(s, block, i, qscale); - s->idsp.idct_put(dest, line_size, block); -} - -/* add block[] to dest[] */ -static inline void add_dct(MpegEncContext *s, - int16_t *block, int i, uint8_t *dest, int line_size) -{ - if (s->block_last_index[i] >= 0) { - s->idsp.idct_add(dest, line_size, block); - } -} - -static inline void add_dequant_dct(MpegEncContext *s, - int16_t *block, int i, uint8_t *dest, int line_size, int qscale) -{ - if (s->block_last_index[i] >= 0) { - s->dct_unquantize_inter(s, block, i, qscale); - - s->idsp.idct_add(dest, line_size, block); - } -} - /** * Clean dc, ac, coded_block for the current non-intra MB. */ @@ -1294,286 +837,6 @@ void ff_clean_intra_table_entries(MpegEncContext *s) s->mbintra_table[xy]= 0; } -#define NOT_MPEG12 0 -#define MAY_BE_MPEG12 1 -#define DEFINITELY_MPEG12 2 - -/* generic function called after a macroblock has been parsed by the - decoder or after it has been encoded by the encoder. - - Important variables used: - s->mb_intra : true if intra macroblock - s->mv_dir : motion vector direction - s->mv_type : motion vector type - s->mv : motion vector - s->interlaced_dct : true if interlaced dct used (mpeg2) - */ -static av_always_inline -void mpv_reconstruct_mb_internal(MpegEncContext *s, int16_t block[12][64], - int lowres_flag, int is_mpeg12, int is_encoder) -{ -#define IS_MPEG12(s) (is_mpeg12 == MAY_BE_MPEG12 ? ((s)->out_format == FMT_MPEG1) : is_mpeg12) - const int mb_xy = s->mb_y * s->mb_stride + s->mb_x; - - s->current_picture.qscale_table[mb_xy] = s->qscale; - - /* update DC predictors for P macroblocks */ - if (!s->mb_intra) { - if (is_mpeg12 != DEFINITELY_MPEG12 && (s->h263_pred || s->h263_aic)) { - if(s->mbintra_table[mb_xy]) - ff_clean_intra_table_entries(s); - } else { - s->last_dc[0] = - s->last_dc[1] = - s->last_dc[2] = 128 << s->intra_dc_precision; - } - } else if (is_mpeg12 != DEFINITELY_MPEG12 && (s->h263_pred || s->h263_aic)) - s->mbintra_table[mb_xy]=1; - - if (!is_encoder || (s->avctx->flags & AV_CODEC_FLAG_PSNR) || s->frame_skip_threshold || s->frame_skip_factor || - !((s->intra_only || s->pict_type == AV_PICTURE_TYPE_B) && - s->avctx->mb_decision != FF_MB_DECISION_RD)) { // FIXME precalc - uint8_t *dest_y, *dest_cb, *dest_cr; - int dct_linesize, dct_offset; - op_pixels_func (*op_pix)[4]; - qpel_mc_func (*op_qpix)[16]; - const int linesize = s->current_picture.f->linesize[0]; //not s->linesize as this would be wrong for field pics - const int uvlinesize = s->current_picture.f->linesize[1]; - const int readable = s->pict_type != AV_PICTURE_TYPE_B || is_encoder || s->avctx->draw_horiz_band || lowres_flag; - const int block_size= lowres_flag ? 8>>s->avctx->lowres : 8; - - /* avoid copy if macroblock skipped in last frame too */ - /* skip only during decoding as we might trash the buffers during encoding a bit */ - if (!is_encoder) { - uint8_t *mbskip_ptr = &s->mbskip_table[mb_xy]; - - if (s->mb_skipped) { - s->mb_skipped= 0; - av_assert2(s->pict_type!=AV_PICTURE_TYPE_I); - *mbskip_ptr = 1; - } else if(!s->current_picture.reference) { - *mbskip_ptr = 1; - } else{ - *mbskip_ptr = 0; /* not skipped */ - } - } - - dct_linesize = linesize << s->interlaced_dct; - dct_offset = s->interlaced_dct ? linesize : linesize * block_size; - - if(readable){ - dest_y= s->dest[0]; - dest_cb= s->dest[1]; - dest_cr= s->dest[2]; - }else{ - dest_y = s->sc.b_scratchpad; - dest_cb= s->sc.b_scratchpad+16*linesize; - dest_cr= s->sc.b_scratchpad+32*linesize; - } - - if (!s->mb_intra) { - /* motion handling */ - /* decoding or more than one mb_type (MC was already done otherwise) */ - if (!is_encoder) { - - if(HAVE_THREADS && s->avctx->active_thread_type&FF_THREAD_FRAME) { - if (s->mv_dir & MV_DIR_FORWARD) { - ff_thread_await_progress(&s->last_picture_ptr->tf, - lowest_referenced_row(s, 0), - 0); - } - if (s->mv_dir & MV_DIR_BACKWARD) { - ff_thread_await_progress(&s->next_picture_ptr->tf, - lowest_referenced_row(s, 1), - 0); - } - } - - if(lowres_flag){ - const h264_chroma_mc_func *op_pix = s->h264chroma.put_h264_chroma_pixels_tab; - - if (s->mv_dir & MV_DIR_FORWARD) { - MPV_motion_lowres(s, dest_y, dest_cb, dest_cr, 0, s->last_picture.f->data, op_pix); - op_pix = s->h264chroma.avg_h264_chroma_pixels_tab; - } - if (s->mv_dir & MV_DIR_BACKWARD) { - MPV_motion_lowres(s, dest_y, dest_cb, dest_cr, 1, s->next_picture.f->data, op_pix); - } - }else{ - op_qpix = s->me.qpel_put; - if ((is_mpeg12 == DEFINITELY_MPEG12 || !s->no_rounding) || s->pict_type == AV_PICTURE_TYPE_B) { - op_pix = s->hdsp.put_pixels_tab; - }else{ - op_pix = s->hdsp.put_no_rnd_pixels_tab; - } - if (s->mv_dir & MV_DIR_FORWARD) { - ff_mpv_motion(s, dest_y, dest_cb, dest_cr, 0, s->last_picture.f->data, op_pix, op_qpix); - op_pix = s->hdsp.avg_pixels_tab; - op_qpix= s->me.qpel_avg; - } - if (s->mv_dir & MV_DIR_BACKWARD) { - ff_mpv_motion(s, dest_y, dest_cb, dest_cr, 1, s->next_picture.f->data, op_pix, op_qpix); - } - } - - /* skip dequant / idct if we are really late ;) */ - if(s->avctx->skip_idct){ - if( (s->avctx->skip_idct >= AVDISCARD_NONREF && s->pict_type == AV_PICTURE_TYPE_B) - ||(s->avctx->skip_idct >= AVDISCARD_NONKEY && s->pict_type != AV_PICTURE_TYPE_I) - || s->avctx->skip_idct >= AVDISCARD_ALL) - goto skip_idct; - } - } - - /* add dct residue */ - if (is_encoder || !(IS_MPEG12(s) || s->msmpeg4_version - || (s->codec_id==AV_CODEC_ID_MPEG4 && !s->mpeg_quant))){ - add_dequant_dct(s, block[0], 0, dest_y , dct_linesize, s->qscale); - add_dequant_dct(s, block[1], 1, dest_y + block_size, dct_linesize, s->qscale); - add_dequant_dct(s, block[2], 2, dest_y + dct_offset , dct_linesize, s->qscale); - add_dequant_dct(s, block[3], 3, dest_y + dct_offset + block_size, dct_linesize, s->qscale); - - if (!CONFIG_GRAY || !(s->avctx->flags & AV_CODEC_FLAG_GRAY)) { - if (s->chroma_y_shift){ - add_dequant_dct(s, block[4], 4, dest_cb, uvlinesize, s->chroma_qscale); - add_dequant_dct(s, block[5], 5, dest_cr, uvlinesize, s->chroma_qscale); - }else{ - dct_linesize >>= 1; - dct_offset >>=1; - add_dequant_dct(s, block[4], 4, dest_cb, dct_linesize, s->chroma_qscale); - add_dequant_dct(s, block[5], 5, dest_cr, dct_linesize, s->chroma_qscale); - add_dequant_dct(s, block[6], 6, dest_cb + dct_offset, dct_linesize, s->chroma_qscale); - add_dequant_dct(s, block[7], 7, dest_cr + dct_offset, dct_linesize, s->chroma_qscale); - } - } - } else if (is_mpeg12 == DEFINITELY_MPEG12 || (s->codec_id != AV_CODEC_ID_WMV2)){ - add_dct(s, block[0], 0, dest_y , dct_linesize); - add_dct(s, block[1], 1, dest_y + block_size, dct_linesize); - add_dct(s, block[2], 2, dest_y + dct_offset , dct_linesize); - add_dct(s, block[3], 3, dest_y + dct_offset + block_size, dct_linesize); - - if (!CONFIG_GRAY || !(s->avctx->flags & AV_CODEC_FLAG_GRAY)) { - if(s->chroma_y_shift){//Chroma420 - add_dct(s, block[4], 4, dest_cb, uvlinesize); - add_dct(s, block[5], 5, dest_cr, uvlinesize); - }else{ - //chroma422 - dct_linesize = uvlinesize << s->interlaced_dct; - dct_offset = s->interlaced_dct ? uvlinesize : uvlinesize*block_size; - - add_dct(s, block[4], 4, dest_cb, dct_linesize); - add_dct(s, block[5], 5, dest_cr, dct_linesize); - add_dct(s, block[6], 6, dest_cb+dct_offset, dct_linesize); - add_dct(s, block[7], 7, dest_cr+dct_offset, dct_linesize); - if(!s->chroma_x_shift){//Chroma444 - add_dct(s, block[8], 8, dest_cb+block_size, dct_linesize); - add_dct(s, block[9], 9, dest_cr+block_size, dct_linesize); - add_dct(s, block[10], 10, dest_cb+block_size+dct_offset, dct_linesize); - add_dct(s, block[11], 11, dest_cr+block_size+dct_offset, dct_linesize); - } - } - }//fi gray - } else if (CONFIG_WMV2_DECODER) { - ff_wmv2_add_mb(s, block, dest_y, dest_cb, dest_cr); - } - } else { - /* Only MPEG-4 Simple Studio Profile is supported in > 8-bit mode. - TODO: Integrate 10-bit properly into mpegvideo.c so that ER works properly */ - if (!is_encoder && is_mpeg12 != DEFINITELY_MPEG12 && CONFIG_MPEG4_DECODER && - /* s->codec_id == AV_CODEC_ID_MPEG4 && */ - s->avctx->bits_per_raw_sample > 8) { - ff_mpeg4_decode_studio(s, dest_y, dest_cb, dest_cr, block_size, - uvlinesize, dct_linesize, dct_offset); - } - /* dct only in intra block */ - else if (is_encoder || !IS_MPEG12(s)) { - put_dct(s, block[0], 0, dest_y , dct_linesize, s->qscale); - put_dct(s, block[1], 1, dest_y + block_size, dct_linesize, s->qscale); - put_dct(s, block[2], 2, dest_y + dct_offset , dct_linesize, s->qscale); - put_dct(s, block[3], 3, dest_y + dct_offset + block_size, dct_linesize, s->qscale); - - if (!CONFIG_GRAY || !(s->avctx->flags & AV_CODEC_FLAG_GRAY)) { - if(s->chroma_y_shift){ - put_dct(s, block[4], 4, dest_cb, uvlinesize, s->chroma_qscale); - put_dct(s, block[5], 5, dest_cr, uvlinesize, s->chroma_qscale); - }else{ - dct_offset >>=1; - dct_linesize >>=1; - put_dct(s, block[4], 4, dest_cb, dct_linesize, s->chroma_qscale); - put_dct(s, block[5], 5, dest_cr, dct_linesize, s->chroma_qscale); - put_dct(s, block[6], 6, dest_cb + dct_offset, dct_linesize, s->chroma_qscale); - put_dct(s, block[7], 7, dest_cr + dct_offset, dct_linesize, s->chroma_qscale); - } - } - }else{ - s->idsp.idct_put(dest_y, dct_linesize, block[0]); - s->idsp.idct_put(dest_y + block_size, dct_linesize, block[1]); - s->idsp.idct_put(dest_y + dct_offset, dct_linesize, block[2]); - s->idsp.idct_put(dest_y + dct_offset + block_size, dct_linesize, block[3]); - - if (!CONFIG_GRAY || !(s->avctx->flags & AV_CODEC_FLAG_GRAY)) { - if(s->chroma_y_shift){ - s->idsp.idct_put(dest_cb, uvlinesize, block[4]); - s->idsp.idct_put(dest_cr, uvlinesize, block[5]); - }else{ - - dct_linesize = uvlinesize << s->interlaced_dct; - dct_offset = s->interlaced_dct ? uvlinesize : uvlinesize*block_size; - - s->idsp.idct_put(dest_cb, dct_linesize, block[4]); - s->idsp.idct_put(dest_cr, dct_linesize, block[5]); - s->idsp.idct_put(dest_cb + dct_offset, dct_linesize, block[6]); - s->idsp.idct_put(dest_cr + dct_offset, dct_linesize, block[7]); - if(!s->chroma_x_shift){//Chroma444 - s->idsp.idct_put(dest_cb + block_size, dct_linesize, block[8]); - s->idsp.idct_put(dest_cr + block_size, dct_linesize, block[9]); - s->idsp.idct_put(dest_cb + block_size + dct_offset, dct_linesize, block[10]); - s->idsp.idct_put(dest_cr + block_size + dct_offset, dct_linesize, block[11]); - } - } - }//gray - } - } -skip_idct: - if(!readable){ - s->hdsp.put_pixels_tab[0][0](s->dest[0], dest_y , linesize,16); - if (!CONFIG_GRAY || !(s->avctx->flags & AV_CODEC_FLAG_GRAY)) { - s->hdsp.put_pixels_tab[s->chroma_x_shift][0](s->dest[1], dest_cb, uvlinesize,16 >> s->chroma_y_shift); - s->hdsp.put_pixels_tab[s->chroma_x_shift][0](s->dest[2], dest_cr, uvlinesize,16 >> s->chroma_y_shift); - } - } - } -} - -void ff_mpv_reconstruct_mb(MpegEncContext *s, int16_t block[12][64]) -{ - if (s->avctx->debug & FF_DEBUG_DCT_COEFF) { - /* print DCT coefficients */ - av_log(s->avctx, AV_LOG_DEBUG, "DCT coeffs of MB at %dx%d:\n", s->mb_x, s->mb_y); - for (int i = 0; i < 6; i++) { - for (int j = 0; j < 64; j++) { - av_log(s->avctx, AV_LOG_DEBUG, "%5d", - block[i][s->idsp.idct_permutation[j]]); - } - av_log(s->avctx, AV_LOG_DEBUG, "\n"); - } - } - - if (s->encoding) { - mpv_reconstruct_mb_internal(s, block, 0, MAY_BE_MPEG12, 1); - } else if (!s->avctx->lowres) { -#if !CONFIG_SMALL - if (s->out_format == FMT_MPEG1) - mpv_reconstruct_mb_internal(s, block, 0, DEFINITELY_MPEG12, 0); - else - mpv_reconstruct_mb_internal(s, block, 0, NOT_MPEG12, 0); -#else - mpv_reconstruct_mb_internal(s, block, 0, MAY_BE_MPEG12, 0); -#endif - } else - mpv_reconstruct_mb_internal(s, block, 1, MAY_BE_MPEG12, 0); -} - void ff_init_block_index(MpegEncContext *s){ //FIXME maybe rename const int linesize = s->current_picture.f->linesize[0]; //not s->linesize as this would be wrong for field pics const int uvlinesize = s->current_picture.f->linesize[1]; diff --git a/libavcodec/mpegvideo.h b/libavcodec/mpegvideo.h index 1ddf8034aa..6e458b0379 100644 --- a/libavcodec/mpegvideo.h +++ b/libavcodec/mpegvideo.h @@ -582,8 +582,6 @@ void ff_mpv_free_context_frame(MpegEncContext *s); void ff_mpv_common_end(MpegEncContext *s); -void ff_mpv_reconstruct_mb(MpegEncContext *s, int16_t block[12][64]); - void ff_clean_intra_table_entries(MpegEncContext *s); int ff_update_duplicate_context(MpegEncContext *dst, const MpegEncContext *src); diff --git a/libavcodec/mpegvideo_dec.c b/libavcodec/mpegvideo_dec.c index 6d1edc027a..c2d6d8bdd7 100644 --- a/libavcodec/mpegvideo_dec.c +++ b/libavcodec/mpegvideo_dec.c @@ -22,17 +22,22 @@ #include +#include "config_components.h" + #include "libavutil/avassert.h" #include "libavutil/imgutils.h" #include "libavutil/internal.h" #include "libavutil/video_enc_params.h" #include "avcodec.h" +#include "h264chroma.h" #include "internal.h" #include "mpegutils.h" #include "mpegvideo.h" #include "mpegvideodec.h" +#include "mpeg4videodec.h" #include "threadframe.h" +#include "wmv2dec.h" void ff_mpv_decode_init(MpegEncContext *s, AVCodecContext *avctx) { @@ -563,3 +568,465 @@ void ff_mpv_report_decode_progress(MpegEncContext *s) if (s->pict_type != AV_PICTURE_TYPE_B && !s->partitioned_frame && !s->er.error_occurred) ff_thread_report_progress(&s->current_picture_ptr->tf, s->mb_y, 0); } + + +static inline int hpel_motion_lowres(MpegEncContext *s, + uint8_t *dest, const uint8_t *src, + int field_based, int field_select, + int src_x, int src_y, + int width, int height, ptrdiff_t stride, + int h_edge_pos, int v_edge_pos, + int w, int h, const h264_chroma_mc_func *pix_op, + int motion_x, int motion_y) +{ + const int lowres = s->avctx->lowres; + const int op_index = FFMIN(lowres, 3); + const int s_mask = (2 << lowres) - 1; + int emu = 0; + int sx, sy; + + if (s->quarter_sample) { + motion_x /= 2; + motion_y /= 2; + } + + sx = motion_x & s_mask; + sy = motion_y & s_mask; + src_x += motion_x >> lowres + 1; + src_y += motion_y >> lowres + 1; + + src += src_y * stride + src_x; + + if ((unsigned)src_x > FFMAX( h_edge_pos - (!!sx) - w, 0) || + (unsigned)src_y > FFMAX((v_edge_pos >> field_based) - (!!sy) - h, 0)) { + s->vdsp.emulated_edge_mc(s->sc.edge_emu_buffer, src, + s->linesize, s->linesize, + w + 1, (h + 1) << field_based, + src_x, src_y * (1 << field_based), + h_edge_pos, v_edge_pos); + src = s->sc.edge_emu_buffer; + emu = 1; + } + + sx = (sx << 2) >> lowres; + sy = (sy << 2) >> lowres; + if (field_select) + src += s->linesize; + pix_op[op_index](dest, src, stride, h, sx, sy); + return emu; +} + +/* apply one mpeg motion vector to the three components */ +static av_always_inline void mpeg_motion_lowres(MpegEncContext *s, + uint8_t *dest_y, + uint8_t *dest_cb, + uint8_t *dest_cr, + int field_based, + int bottom_field, + int field_select, + uint8_t *const *ref_picture, + const h264_chroma_mc_func *pix_op, + int motion_x, int motion_y, + int h, int mb_y) +{ + const uint8_t *ptr_y, *ptr_cb, *ptr_cr; + int mx, my, src_x, src_y, uvsrc_x, uvsrc_y, sx, sy, uvsx, uvsy; + ptrdiff_t uvlinesize, linesize; + const int lowres = s->avctx->lowres; + const int op_index = FFMIN(lowres - 1 + s->chroma_x_shift, 3); + const int block_s = 8 >> lowres; + const int s_mask = (2 << lowres) - 1; + const int h_edge_pos = s->h_edge_pos >> lowres; + const int v_edge_pos = s->v_edge_pos >> lowres; + linesize = s->current_picture.f->linesize[0] << field_based; + uvlinesize = s->current_picture.f->linesize[1] << field_based; + + // FIXME obviously not perfect but qpel will not work in lowres anyway + if (s->quarter_sample) { + motion_x /= 2; + motion_y /= 2; + } + + if (field_based) { + motion_y += (bottom_field - field_select)*((1 << lowres)-1); + } + + sx = motion_x & s_mask; + sy = motion_y & s_mask; + src_x = s->mb_x * 2 * block_s + (motion_x >> lowres + 1); + src_y = (mb_y * 2 * block_s >> field_based) + (motion_y >> lowres + 1); + + if (s->out_format == FMT_H263) { + uvsx = ((motion_x >> 1) & s_mask) | (sx & 1); + uvsy = ((motion_y >> 1) & s_mask) | (sy & 1); + uvsrc_x = src_x >> 1; + uvsrc_y = src_y >> 1; + } else if (s->out_format == FMT_H261) { + // even chroma mv's are full pel in H261 + mx = motion_x / 4; + my = motion_y / 4; + uvsx = (2 * mx) & s_mask; + uvsy = (2 * my) & s_mask; + uvsrc_x = s->mb_x * block_s + (mx >> lowres); + uvsrc_y = mb_y * block_s + (my >> lowres); + } else { + if (s->chroma_y_shift) { + mx = motion_x / 2; + my = motion_y / 2; + uvsx = mx & s_mask; + uvsy = my & s_mask; + uvsrc_x = s->mb_x * block_s + (mx >> lowres + 1); + uvsrc_y = (mb_y * block_s >> field_based) + (my >> lowres + 1); + } else { + if (s->chroma_x_shift) { + //Chroma422 + mx = motion_x / 2; + uvsx = mx & s_mask; + uvsy = motion_y & s_mask; + uvsrc_y = src_y; + uvsrc_x = s->mb_x*block_s + (mx >> (lowres+1)); + } else { + //Chroma444 + uvsx = motion_x & s_mask; + uvsy = motion_y & s_mask; + uvsrc_x = src_x; + uvsrc_y = src_y; + } + } + } + + ptr_y = ref_picture[0] + src_y * linesize + src_x; + ptr_cb = ref_picture[1] + uvsrc_y * uvlinesize + uvsrc_x; + ptr_cr = ref_picture[2] + uvsrc_y * uvlinesize + uvsrc_x; + + if ((unsigned) src_x > FFMAX( h_edge_pos - (!!sx) - 2 * block_s, 0) || uvsrc_y<0 || + (unsigned) src_y > FFMAX((v_edge_pos >> field_based) - (!!sy) - h, 0)) { + s->vdsp.emulated_edge_mc(s->sc.edge_emu_buffer, ptr_y, + linesize >> field_based, linesize >> field_based, + 17, 17 + field_based, + src_x, src_y * (1 << field_based), h_edge_pos, + v_edge_pos); + ptr_y = s->sc.edge_emu_buffer; + if (!CONFIG_GRAY || !(s->avctx->flags & AV_CODEC_FLAG_GRAY)) { + uint8_t *ubuf = s->sc.edge_emu_buffer + 18 * s->linesize; + uint8_t *vbuf =ubuf + 10 * s->uvlinesize; + if (s->workaround_bugs & FF_BUG_IEDGE) + vbuf -= s->uvlinesize; + s->vdsp.emulated_edge_mc(ubuf, ptr_cb, + uvlinesize >> field_based, uvlinesize >> field_based, + 9, 9 + field_based, + uvsrc_x, uvsrc_y * (1 << field_based), + h_edge_pos >> 1, v_edge_pos >> 1); + s->vdsp.emulated_edge_mc(vbuf, ptr_cr, + uvlinesize >> field_based,uvlinesize >> field_based, + 9, 9 + field_based, + uvsrc_x, uvsrc_y * (1 << field_based), + h_edge_pos >> 1, v_edge_pos >> 1); + ptr_cb = ubuf; + ptr_cr = vbuf; + } + } + + // FIXME use this for field pix too instead of the obnoxious hack which changes picture.f->data + if (bottom_field) { + dest_y += s->linesize; + dest_cb += s->uvlinesize; + dest_cr += s->uvlinesize; + } + + if (field_select) { + ptr_y += s->linesize; + ptr_cb += s->uvlinesize; + ptr_cr += s->uvlinesize; + } + + sx = (sx << 2) >> lowres; + sy = (sy << 2) >> lowres; + pix_op[lowres - 1](dest_y, ptr_y, linesize, h, sx, sy); + + if (!CONFIG_GRAY || !(s->avctx->flags & AV_CODEC_FLAG_GRAY)) { + int hc = s->chroma_y_shift ? (h+1-bottom_field)>>1 : h; + uvsx = (uvsx << 2) >> lowres; + uvsy = (uvsy << 2) >> lowres; + if (hc) { + pix_op[op_index](dest_cb, ptr_cb, uvlinesize, hc, uvsx, uvsy); + pix_op[op_index](dest_cr, ptr_cr, uvlinesize, hc, uvsx, uvsy); + } + } + // FIXME h261 lowres loop filter +} + +static inline void chroma_4mv_motion_lowres(MpegEncContext *s, + uint8_t *dest_cb, uint8_t *dest_cr, + uint8_t *const *ref_picture, + const h264_chroma_mc_func * pix_op, + int mx, int my) +{ + const int lowres = s->avctx->lowres; + const int op_index = FFMIN(lowres, 3); + const int block_s = 8 >> lowres; + const int s_mask = (2 << lowres) - 1; + const int h_edge_pos = s->h_edge_pos >> lowres + 1; + const int v_edge_pos = s->v_edge_pos >> lowres + 1; + int emu = 0, src_x, src_y, sx, sy; + ptrdiff_t offset; + const uint8_t *ptr; + + if (s->quarter_sample) { + mx /= 2; + my /= 2; + } + + /* In case of 8X8, we construct a single chroma motion vector + with a special rounding */ + mx = ff_h263_round_chroma(mx); + my = ff_h263_round_chroma(my); + + sx = mx & s_mask; + sy = my & s_mask; + src_x = s->mb_x * block_s + (mx >> lowres + 1); + src_y = s->mb_y * block_s + (my >> lowres + 1); + + offset = src_y * s->uvlinesize + src_x; + ptr = ref_picture[1] + offset; + if ((unsigned) src_x > FFMAX(h_edge_pos - (!!sx) - block_s, 0) || + (unsigned) src_y > FFMAX(v_edge_pos - (!!sy) - block_s, 0)) { + s->vdsp.emulated_edge_mc(s->sc.edge_emu_buffer, ptr, + s->uvlinesize, s->uvlinesize, + 9, 9, + src_x, src_y, h_edge_pos, v_edge_pos); + ptr = s->sc.edge_emu_buffer; + emu = 1; + } + sx = (sx << 2) >> lowres; + sy = (sy << 2) >> lowres; + pix_op[op_index](dest_cb, ptr, s->uvlinesize, block_s, sx, sy); + + ptr = ref_picture[2] + offset; + if (emu) { + s->vdsp.emulated_edge_mc(s->sc.edge_emu_buffer, ptr, + s->uvlinesize, s->uvlinesize, + 9, 9, + src_x, src_y, h_edge_pos, v_edge_pos); + ptr = s->sc.edge_emu_buffer; + } + pix_op[op_index](dest_cr, ptr, s->uvlinesize, block_s, sx, sy); +} + +/** + * motion compensation of a single macroblock + * @param s context + * @param dest_y luma destination pointer + * @param dest_cb chroma cb/u destination pointer + * @param dest_cr chroma cr/v destination pointer + * @param dir direction (0->forward, 1->backward) + * @param ref_picture array[3] of pointers to the 3 planes of the reference picture + * @param pix_op halfpel motion compensation function (average or put normally) + * the motion vectors are taken from s->mv and the MV type from s->mv_type + */ +static inline void MPV_motion_lowres(MpegEncContext *s, + uint8_t *dest_y, uint8_t *dest_cb, + uint8_t *dest_cr, + int dir, uint8_t *const *ref_picture, + const h264_chroma_mc_func *pix_op) +{ + int mx, my; + int mb_x, mb_y; + const int lowres = s->avctx->lowres; + const int block_s = 8 >>lowres; + + mb_x = s->mb_x; + mb_y = s->mb_y; + + switch (s->mv_type) { + case MV_TYPE_16X16: + mpeg_motion_lowres(s, dest_y, dest_cb, dest_cr, + 0, 0, 0, + ref_picture, pix_op, + s->mv[dir][0][0], s->mv[dir][0][1], + 2 * block_s, mb_y); + break; + case MV_TYPE_8X8: + mx = 0; + my = 0; + for (int i = 0; i < 4; i++) { + hpel_motion_lowres(s, dest_y + ((i & 1) + (i >> 1) * + s->linesize) * block_s, + ref_picture[0], 0, 0, + (2 * mb_x + (i & 1)) * block_s, + (2 * mb_y + (i >> 1)) * block_s, + s->width, s->height, s->linesize, + s->h_edge_pos >> lowres, s->v_edge_pos >> lowres, + block_s, block_s, pix_op, + s->mv[dir][i][0], s->mv[dir][i][1]); + + mx += s->mv[dir][i][0]; + my += s->mv[dir][i][1]; + } + + if (!CONFIG_GRAY || !(s->avctx->flags & AV_CODEC_FLAG_GRAY)) + chroma_4mv_motion_lowres(s, dest_cb, dest_cr, ref_picture, + pix_op, mx, my); + break; + case MV_TYPE_FIELD: + if (s->picture_structure == PICT_FRAME) { + /* top field */ + mpeg_motion_lowres(s, dest_y, dest_cb, dest_cr, + 1, 0, s->field_select[dir][0], + ref_picture, pix_op, + s->mv[dir][0][0], s->mv[dir][0][1], + block_s, mb_y); + /* bottom field */ + mpeg_motion_lowres(s, dest_y, dest_cb, dest_cr, + 1, 1, s->field_select[dir][1], + ref_picture, pix_op, + s->mv[dir][1][0], s->mv[dir][1][1], + block_s, mb_y); + } else { + if (s->picture_structure != s->field_select[dir][0] + 1 && + s->pict_type != AV_PICTURE_TYPE_B && !s->first_field) { + ref_picture = s->current_picture_ptr->f->data; + + } + mpeg_motion_lowres(s, dest_y, dest_cb, dest_cr, + 0, 0, s->field_select[dir][0], + ref_picture, pix_op, + s->mv[dir][0][0], + s->mv[dir][0][1], 2 * block_s, mb_y >> 1); + } + break; + case MV_TYPE_16X8: + for (int i = 0; i < 2; i++) { + uint8_t *const *ref2picture; + + if (s->picture_structure == s->field_select[dir][i] + 1 || + s->pict_type == AV_PICTURE_TYPE_B || s->first_field) { + ref2picture = ref_picture; + } else { + ref2picture = s->current_picture_ptr->f->data; + } + + mpeg_motion_lowres(s, dest_y, dest_cb, dest_cr, + 0, 0, s->field_select[dir][i], + ref2picture, pix_op, + s->mv[dir][i][0], s->mv[dir][i][1] + + 2 * block_s * i, block_s, mb_y >> 1); + + dest_y += 2 * block_s * s->linesize; + dest_cb += (2 * block_s >> s->chroma_y_shift) * s->uvlinesize; + dest_cr += (2 * block_s >> s->chroma_y_shift) * s->uvlinesize; + } + break; + case MV_TYPE_DMV: + if (s->picture_structure == PICT_FRAME) { + for (int i = 0; i < 2; i++) { + for (int j = 0; j < 2; j++) { + mpeg_motion_lowres(s, dest_y, dest_cb, dest_cr, + 1, j, j ^ i, + ref_picture, pix_op, + s->mv[dir][2 * i + j][0], + s->mv[dir][2 * i + j][1], + block_s, mb_y); + } + pix_op = s->h264chroma.avg_h264_chroma_pixels_tab; + } + } else { + for (int i = 0; i < 2; i++) { + mpeg_motion_lowres(s, dest_y, dest_cb, dest_cr, + 0, 0, s->picture_structure != i + 1, + ref_picture, pix_op, + s->mv[dir][2 * i][0],s->mv[dir][2 * i][1], + 2 * block_s, mb_y >> 1); + + // after put we make avg of the same block + pix_op = s->h264chroma.avg_h264_chroma_pixels_tab; + + // opposite parity is always in the same + // frame if this is second field + if (!s->first_field) { + ref_picture = s->current_picture_ptr->f->data; + } + } + } + break; + default: + av_assert2(0); + } +} + +/** + * find the lowest MB row referenced in the MVs + */ +static int lowest_referenced_row(MpegEncContext *s, int dir) +{ + int my_max = INT_MIN, my_min = INT_MAX, qpel_shift = !s->quarter_sample; + int off, mvs; + + if (s->picture_structure != PICT_FRAME || s->mcsel) + goto unhandled; + + switch (s->mv_type) { + case MV_TYPE_16X16: + mvs = 1; + break; + case MV_TYPE_16X8: + mvs = 2; + break; + case MV_TYPE_8X8: + mvs = 4; + break; + default: + goto unhandled; + } + + for (int i = 0; i < mvs; i++) { + int my = s->mv[dir][i][1]; + my_max = FFMAX(my_max, my); + my_min = FFMIN(my_min, my); + } + + off = ((FFMAX(-my_min, my_max) << qpel_shift) + 63) >> 6; + + return av_clip(s->mb_y + off, 0, s->mb_height - 1); +unhandled: + return s->mb_height - 1; +} + +/* add block[] to dest[] */ +static inline void add_dct(MpegEncContext *s, + int16_t *block, int i, uint8_t *dest, int line_size) +{ + if (s->block_last_index[i] >= 0) { + s->idsp.idct_add(dest, line_size, block); + } +} + +#define IS_ENCODER 0 +#include "mpv_reconstruct_mb_template.c" + +void ff_mpv_reconstruct_mb(MpegEncContext *s, int16_t block[12][64]) +{ + if (s->avctx->debug & FF_DEBUG_DCT_COEFF) { + /* print DCT coefficients */ + av_log(s->avctx, AV_LOG_DEBUG, "DCT coeffs of MB at %dx%d:\n", s->mb_x, s->mb_y); + for (int i = 0; i < 6; i++) { + for (int j = 0; j < 64; j++) { + av_log(s->avctx, AV_LOG_DEBUG, "%5d", + block[i][s->idsp.idct_permutation[j]]); + } + av_log(s->avctx, AV_LOG_DEBUG, "\n"); + } + } + + if (!s->avctx->lowres) { +#if !CONFIG_SMALL + if (s->out_format == FMT_MPEG1) + mpv_reconstruct_mb_internal(s, block, 0, DEFINITELY_MPEG12); + else + mpv_reconstruct_mb_internal(s, block, 0, NOT_MPEG12); +#else + mpv_reconstruct_mb_internal(s, block, 0, MAY_BE_MPEG12); +#endif + } else + mpv_reconstruct_mb_internal(s, block, 1, MAY_BE_MPEG12); +} diff --git a/libavcodec/mpegvideo_enc.c b/libavcodec/mpegvideo_enc.c index 2cbb856866..ce363a585d 100644 --- a/libavcodec/mpegvideo_enc.c +++ b/libavcodec/mpegvideo_enc.c @@ -1017,6 +1017,26 @@ av_cold int ff_mpv_encode_end(AVCodecContext *avctx) return 0; } +#define IS_ENCODER 1 +#include "mpv_reconstruct_mb_template.c" + +static void mpv_reconstruct_mb(MpegEncContext *s, int16_t block[12][64]) +{ + if (s->avctx->debug & FF_DEBUG_DCT_COEFF) { + /* print DCT coefficients */ + av_log(s->avctx, AV_LOG_DEBUG, "DCT coeffs of MB at %dx%d:\n", s->mb_x, s->mb_y); + for (int i = 0; i < 6; i++) { + for (int j = 0; j < 64; j++) { + av_log(s->avctx, AV_LOG_DEBUG, "%5d", + block[i][s->idsp.idct_permutation[j]]); + } + av_log(s->avctx, AV_LOG_DEBUG, "\n"); + } + } + + mpv_reconstruct_mb_internal(s, block, 0, MAY_BE_MPEG12); +} + static int get_sae(const uint8_t *src, int ref, int stride) { int x,y; @@ -2577,7 +2597,7 @@ static inline void encode_mb_hq(MpegEncContext *s, MpegEncContext *backup, MpegE } if(s->avctx->mb_decision == FF_MB_DECISION_RD){ - ff_mpv_reconstruct_mb(s, s->block); + mpv_reconstruct_mb(s, s->block); score *= s->lambda2; score += sse_mb(s) << FF_LAMBDA_SHIFT; @@ -3287,7 +3307,7 @@ static int encode_thread(AVCodecContext *c, void *arg){ } if(s->avctx->mb_decision == FF_MB_DECISION_BITS) - ff_mpv_reconstruct_mb(s, s->block); + mpv_reconstruct_mb(s, s->block); } else { int motion_x = 0, motion_y = 0; s->mv_type=MV_TYPE_16X16; @@ -3406,7 +3426,7 @@ static int encode_thread(AVCodecContext *c, void *arg){ s->out_format == FMT_H263 && s->pict_type!=AV_PICTURE_TYPE_B) ff_h263_update_motion_val(s); - ff_mpv_reconstruct_mb(s, s->block); + mpv_reconstruct_mb(s, s->block); } /* clean the MV table in IPS frames for direct mode in B-frames */ diff --git a/libavcodec/mpegvideodec.h b/libavcodec/mpegvideodec.h index 250034b486..0b841bc1a1 100644 --- a/libavcodec/mpegvideodec.h +++ b/libavcodec/mpegvideodec.h @@ -50,6 +50,7 @@ void ff_mpv_decode_init(MpegEncContext *s, AVCodecContext *avctx); int ff_mpv_common_frame_size_change(MpegEncContext *s); int ff_mpv_frame_start(MpegEncContext *s, AVCodecContext *avctx); +void ff_mpv_reconstruct_mb(MpegEncContext *s, int16_t block[12][64]); void ff_mpv_report_decode_progress(MpegEncContext *s); void ff_mpv_frame_end(MpegEncContext *s); diff --git a/libavcodec/mpv_reconstruct_mb_template.c b/libavcodec/mpv_reconstruct_mb_template.c new file mode 100644 index 0000000000..7d74ec1f5c --- /dev/null +++ b/libavcodec/mpv_reconstruct_mb_template.c @@ -0,0 +1,300 @@ +/* + * MPEG macroblock reconstruction + * Copyright (c) 2000,2001 Fabrice Bellard + * Copyright (c) 2002-2004 Michael Niedermayer + * + * 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 + */ + +#define NOT_MPEG12 0 +#define MAY_BE_MPEG12 1 +#define DEFINITELY_MPEG12 2 + +/* put block[] to dest[] */ +static inline void put_dct(MpegEncContext *s, + int16_t *block, int i, uint8_t *dest, int line_size, int qscale) +{ + s->dct_unquantize_intra(s, block, i, qscale); + s->idsp.idct_put(dest, line_size, block); +} + +static inline void add_dequant_dct(MpegEncContext *s, + int16_t *block, int i, uint8_t *dest, int line_size, int qscale) +{ + if (s->block_last_index[i] >= 0) { + s->dct_unquantize_inter(s, block, i, qscale); + + s->idsp.idct_add(dest, line_size, block); + } +} + +/* generic function called after a macroblock has been parsed by the + decoder or after it has been encoded by the encoder. + + Important variables used: + s->mb_intra : true if intra macroblock + s->mv_dir : motion vector direction + s->mv_type : motion vector type + s->mv : motion vector + s->interlaced_dct : true if interlaced dct used (mpeg2) + */ +static av_always_inline +void mpv_reconstruct_mb_internal(MpegEncContext *s, int16_t block[12][64], + int lowres_flag, int is_mpeg12) +{ +#define IS_MPEG12(s) (is_mpeg12 == MAY_BE_MPEG12 ? ((s)->out_format == FMT_MPEG1) : is_mpeg12) + const int mb_xy = s->mb_y * s->mb_stride + s->mb_x; + + s->current_picture.qscale_table[mb_xy] = s->qscale; + + /* update DC predictors for P macroblocks */ + if (!s->mb_intra) { + if (is_mpeg12 != DEFINITELY_MPEG12 && (s->h263_pred || s->h263_aic)) { + if (s->mbintra_table[mb_xy]) + ff_clean_intra_table_entries(s); + } else { + s->last_dc[0] = + s->last_dc[1] = + s->last_dc[2] = 128 << s->intra_dc_precision; + } + } else if (is_mpeg12 != DEFINITELY_MPEG12 && (s->h263_pred || s->h263_aic)) + s->mbintra_table[mb_xy] = 1; + +#if IS_ENCODER + if ((s->avctx->flags & AV_CODEC_FLAG_PSNR) || s->frame_skip_threshold || s->frame_skip_factor || + !((s->intra_only || s->pict_type == AV_PICTURE_TYPE_B) && + s->avctx->mb_decision != FF_MB_DECISION_RD)) // FIXME precalc +#endif /* IS_ENCODER */ + { + uint8_t *dest_y, *dest_cb, *dest_cr; + int dct_linesize, dct_offset; + const int linesize = s->current_picture.f->linesize[0]; //not s->linesize as this would be wrong for field pics + const int uvlinesize = s->current_picture.f->linesize[1]; + const int readable = IS_ENCODER || lowres_flag || s->pict_type != AV_PICTURE_TYPE_B || s->avctx->draw_horiz_band; + const int block_size = lowres_flag ? 8 >> s->avctx->lowres : 8; + + /* avoid copy if macroblock skipped in last frame too */ + /* skip only during decoding as we might trash the buffers during encoding a bit */ + if (!IS_ENCODER) { + uint8_t *mbskip_ptr = &s->mbskip_table[mb_xy]; + + if (s->mb_skipped) { + s->mb_skipped = 0; + av_assert2(s->pict_type!=AV_PICTURE_TYPE_I); + *mbskip_ptr = 1; + } else if(!s->current_picture.reference) { + *mbskip_ptr = 1; + } else{ + *mbskip_ptr = 0; /* not skipped */ + } + } + + dct_linesize = linesize << s->interlaced_dct; + dct_offset = s->interlaced_dct ? linesize : linesize * block_size; + + if (readable) { + dest_y = s->dest[0]; + dest_cb = s->dest[1]; + dest_cr = s->dest[2]; + } else { + dest_y = s->sc.b_scratchpad; + dest_cb = s->sc.b_scratchpad + 16 * linesize; + dest_cr = s->sc.b_scratchpad + 32 * linesize; + } + + if (!s->mb_intra) { + /* motion handling */ + /* decoding or more than one mb_type (MC was already done otherwise) */ + +#if !IS_ENCODER + if (HAVE_THREADS && s->avctx->active_thread_type & FF_THREAD_FRAME) { + if (s->mv_dir & MV_DIR_FORWARD) { + ff_thread_await_progress(&s->last_picture_ptr->tf, + lowest_referenced_row(s, 0), 0); + } + if (s->mv_dir & MV_DIR_BACKWARD) { + ff_thread_await_progress(&s->next_picture_ptr->tf, + lowest_referenced_row(s, 1), 0); + } + } + + if (lowres_flag) { + const h264_chroma_mc_func *op_pix = s->h264chroma.put_h264_chroma_pixels_tab; + + if (s->mv_dir & MV_DIR_FORWARD) { + MPV_motion_lowres(s, dest_y, dest_cb, dest_cr, 0, s->last_picture.f->data, op_pix); + op_pix = s->h264chroma.avg_h264_chroma_pixels_tab; + } + if (s->mv_dir & MV_DIR_BACKWARD) { + MPV_motion_lowres(s, dest_y, dest_cb, dest_cr, 1, s->next_picture.f->data, op_pix); + } + } else { + op_pixels_func (*op_pix)[4]; + qpel_mc_func (*op_qpix)[16] = s->me.qpel_put; + + if ((is_mpeg12 == DEFINITELY_MPEG12 || !s->no_rounding) || s->pict_type == AV_PICTURE_TYPE_B) { + op_pix = s->hdsp.put_pixels_tab; + } else { + op_pix = s->hdsp.put_no_rnd_pixels_tab; + } + if (s->mv_dir & MV_DIR_FORWARD) { + ff_mpv_motion(s, dest_y, dest_cb, dest_cr, 0, s->last_picture.f->data, op_pix, op_qpix); + op_pix = s->hdsp.avg_pixels_tab; + op_qpix = s->me.qpel_avg; + } + if (s->mv_dir & MV_DIR_BACKWARD) { + ff_mpv_motion(s, dest_y, dest_cb, dest_cr, 1, s->next_picture.f->data, op_pix, op_qpix); + } + } + + /* skip dequant / idct if we are really late ;) */ + if (s->avctx->skip_idct) { + if( (s->avctx->skip_idct >= AVDISCARD_NONREF && s->pict_type == AV_PICTURE_TYPE_B) + ||(s->avctx->skip_idct >= AVDISCARD_NONKEY && s->pict_type != AV_PICTURE_TYPE_I) + || s->avctx->skip_idct >= AVDISCARD_ALL) + goto skip_idct; + } + + /* add dct residue */ + if (!(IS_MPEG12(s) || s->msmpeg4_version || + (s->codec_id == AV_CODEC_ID_MPEG4 && !s->mpeg_quant))) +#endif /* !IS_ENCODER */ + { + add_dequant_dct(s, block[0], 0, dest_y , dct_linesize, s->qscale); + add_dequant_dct(s, block[1], 1, dest_y + block_size, dct_linesize, s->qscale); + add_dequant_dct(s, block[2], 2, dest_y + dct_offset , dct_linesize, s->qscale); + add_dequant_dct(s, block[3], 3, dest_y + dct_offset + block_size, dct_linesize, s->qscale); + + if (!CONFIG_GRAY || !(s->avctx->flags & AV_CODEC_FLAG_GRAY)) { + if (s->chroma_y_shift) { + add_dequant_dct(s, block[4], 4, dest_cb, uvlinesize, s->chroma_qscale); + add_dequant_dct(s, block[5], 5, dest_cr, uvlinesize, s->chroma_qscale); + } else { + dct_linesize >>= 1; + dct_offset >>= 1; + add_dequant_dct(s, block[4], 4, dest_cb, dct_linesize, s->chroma_qscale); + add_dequant_dct(s, block[5], 5, dest_cr, dct_linesize, s->chroma_qscale); + add_dequant_dct(s, block[6], 6, dest_cb + dct_offset, dct_linesize, s->chroma_qscale); + add_dequant_dct(s, block[7], 7, dest_cr + dct_offset, dct_linesize, s->chroma_qscale); + } + } + } +#if !IS_ENCODER + else if (is_mpeg12 == DEFINITELY_MPEG12 || (s->codec_id != AV_CODEC_ID_WMV2)) { + add_dct(s, block[0], 0, dest_y , dct_linesize); + add_dct(s, block[1], 1, dest_y + block_size, dct_linesize); + add_dct(s, block[2], 2, dest_y + dct_offset , dct_linesize); + add_dct(s, block[3], 3, dest_y + dct_offset + block_size, dct_linesize); + + if (!CONFIG_GRAY || !(s->avctx->flags & AV_CODEC_FLAG_GRAY)) { + if (s->chroma_y_shift) {//Chroma420 + add_dct(s, block[4], 4, dest_cb, uvlinesize); + add_dct(s, block[5], 5, dest_cr, uvlinesize); + } else { + //chroma422 + dct_linesize = uvlinesize << s->interlaced_dct; + dct_offset = s->interlaced_dct ? uvlinesize : uvlinesize*block_size; + + add_dct(s, block[4], 4, dest_cb, dct_linesize); + add_dct(s, block[5], 5, dest_cr, dct_linesize); + add_dct(s, block[6], 6, dest_cb+dct_offset, dct_linesize); + add_dct(s, block[7], 7, dest_cr+dct_offset, dct_linesize); + if (!s->chroma_x_shift) {//Chroma444 + add_dct(s, block[8], 8, dest_cb+block_size, dct_linesize); + add_dct(s, block[9], 9, dest_cr+block_size, dct_linesize); + add_dct(s, block[10], 10, dest_cb+block_size+dct_offset, dct_linesize); + add_dct(s, block[11], 11, dest_cr+block_size+dct_offset, dct_linesize); + } + } + } //fi gray + } else if (CONFIG_WMV2_DECODER) { + ff_wmv2_add_mb(s, block, dest_y, dest_cb, dest_cr); + } +#endif /* !IS_ENCODER */ + } else { +#if !IS_ENCODER + /* Only MPEG-4 Simple Studio Profile is supported in > 8-bit mode. + TODO: Integrate 10-bit properly into mpegvideo.c so that ER works properly */ + if (is_mpeg12 != DEFINITELY_MPEG12 && CONFIG_MPEG4_DECODER && + /* s->codec_id == AV_CODEC_ID_MPEG4 && */ + s->avctx->bits_per_raw_sample > 8) { + ff_mpeg4_decode_studio(s, dest_y, dest_cb, dest_cr, block_size, + uvlinesize, dct_linesize, dct_offset); + } else if (!IS_MPEG12(s)) +#endif /* !IS_ENCODER */ + { + /* dct only in intra block */ + put_dct(s, block[0], 0, dest_y , dct_linesize, s->qscale); + put_dct(s, block[1], 1, dest_y + block_size, dct_linesize, s->qscale); + put_dct(s, block[2], 2, dest_y + dct_offset , dct_linesize, s->qscale); + put_dct(s, block[3], 3, dest_y + dct_offset + block_size, dct_linesize, s->qscale); + + if (!CONFIG_GRAY || !(s->avctx->flags & AV_CODEC_FLAG_GRAY)) { + if (s->chroma_y_shift) { + put_dct(s, block[4], 4, dest_cb, uvlinesize, s->chroma_qscale); + put_dct(s, block[5], 5, dest_cr, uvlinesize, s->chroma_qscale); + } else { + dct_offset >>=1; + dct_linesize >>=1; + put_dct(s, block[4], 4, dest_cb, dct_linesize, s->chroma_qscale); + put_dct(s, block[5], 5, dest_cr, dct_linesize, s->chroma_qscale); + put_dct(s, block[6], 6, dest_cb + dct_offset, dct_linesize, s->chroma_qscale); + put_dct(s, block[7], 7, dest_cr + dct_offset, dct_linesize, s->chroma_qscale); + } + } + } +#if !IS_ENCODER + else { + s->idsp.idct_put(dest_y, dct_linesize, block[0]); + s->idsp.idct_put(dest_y + block_size, dct_linesize, block[1]); + s->idsp.idct_put(dest_y + dct_offset, dct_linesize, block[2]); + s->idsp.idct_put(dest_y + dct_offset + block_size, dct_linesize, block[3]); + + if (!CONFIG_GRAY || !(s->avctx->flags & AV_CODEC_FLAG_GRAY)) { + if (s->chroma_y_shift) { + s->idsp.idct_put(dest_cb, uvlinesize, block[4]); + s->idsp.idct_put(dest_cr, uvlinesize, block[5]); + } else { + dct_linesize = uvlinesize << s->interlaced_dct; + dct_offset = s->interlaced_dct ? uvlinesize : uvlinesize*block_size; + + s->idsp.idct_put(dest_cb, dct_linesize, block[4]); + s->idsp.idct_put(dest_cr, dct_linesize, block[5]); + s->idsp.idct_put(dest_cb + dct_offset, dct_linesize, block[6]); + s->idsp.idct_put(dest_cr + dct_offset, dct_linesize, block[7]); + if (!s->chroma_x_shift) { //Chroma444 + s->idsp.idct_put(dest_cb + block_size, dct_linesize, block[8]); + s->idsp.idct_put(dest_cr + block_size, dct_linesize, block[9]); + s->idsp.idct_put(dest_cb + block_size + dct_offset, dct_linesize, block[10]); + s->idsp.idct_put(dest_cr + block_size + dct_offset, dct_linesize, block[11]); + } + } + } //gray + } + } +skip_idct: + if (!readable) { + s->hdsp.put_pixels_tab[0][0](s->dest[0], dest_y, linesize, 16); + if (!CONFIG_GRAY || !(s->avctx->flags & AV_CODEC_FLAG_GRAY)) { + s->hdsp.put_pixels_tab[s->chroma_x_shift][0](s->dest[1], dest_cb, uvlinesize, 16 >> s->chroma_y_shift); + s->hdsp.put_pixels_tab[s->chroma_x_shift][0](s->dest[2], dest_cr, uvlinesize, 16 >> s->chroma_y_shift); + } +#endif /* !IS_ENCODER */ + } + } +} +