From patchwork Sun May 21 13:03:09 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: Nuo Mi X-Patchwork-Id: 41762 Delivered-To: ffmpegpatchwork2@gmail.com Received: by 2002:a05:6a20:ba91:b0:105:feb:71f2 with SMTP id fb17csp1090667pzb; Sun, 21 May 2023 06:04:32 -0700 (PDT) X-Google-Smtp-Source: ACHHUZ5d9dOkIg1PPsYTom9yL6FGY21EkK5oqWVPdgDoFkEwB6r1iCulA+wzQHnCjrdtsvxCm9B7 X-Received: by 2002:aa7:d8d4:0:b0:50b:bfee:ea26 with SMTP id k20-20020aa7d8d4000000b0050bbfeeea26mr6104492eds.35.1684674271974; Sun, 21 May 2023 06:04:31 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1684674271; cv=none; d=google.com; s=arc-20160816; b=dMDCnjKkqU1uHDCVceZDQoIxUhuYquDTju4fMb5e4Lw3SxEfnScWSkgC6Cp5Qi19KI 5RekHRSWxDIpY9wFS3ZLmKSxU8bNXvLHFcFNWhRgOTLAKbeKuXM+UDpLpRwBHBfTT09e j6k2+IoBlUmK6QhkmYhMmRkBiciD0iZfT6QAdAME+NqD8riVjwnqmkl0p/OOuXQHO+HT fOrsHVRJNp15OkvmfIUhY8J3MMI5PFXw+GDPpKUrudvFzGugGrL+YGc8YqmRIYfHwv6u VEcXnBlWbsxAyjn58Ir8reaurOi4qvQTukR7OswLrXx/kF1BDdmB2FsgPJRwIJUn2qh8 rndA== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=sender:errors-to:content-transfer-encoding:cc:reply-to :list-subscribe:list-help:list-post:list-archive:list-unsubscribe :list-id:precedence:subject:mime-version:references:in-reply-to :message-id:date:to:from:dkim-signature:delivered-to; bh=zgb9qUlLHnsQeG6AKib8agqJ8SqS0GBitKVnD8ygPAg=; b=Fj2zjnVOThO1R1BYXp+qoNtjMV3y1Ezrp9gnt/qBpHhbSBR4Jwn+cUfXY3PR4N2wGt LsITb45nPpC2u0gboffw7ogC0EW0N60pm1Y0G+tNx9neseZ9KNYBnObvt3TL9gp0MAcT IIkuwYiuTAek2O+wQD7ESq4wv0xFkh+NUhBOljHeYB0DzxBMUxX+cK+1Q9nYJBs3QDEB Y3G+ZV1VL31teqmW3erjwgIylNRhdBwsejCbDAX5STicIe7M1PZJJopbyStiBEjs85Xc G3GRgHT/slaISHlXHO+YKgZuvDKL81Q6uepPnrEhfSYGEPtihlWgLOBzF3CDTPLeASCA VxPg== ARC-Authentication-Results: i=1; mx.google.com; dkim=neutral (body hash did not verify) header.i=@gmail.com header.s=20221208 header.b=E7eMq7qX; 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=gmail.com Return-Path: Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org. [79.124.17.100]) by mx.google.com with ESMTP id s2-20020aa7c542000000b0050d8954be10si2912522edr.119.2023.05.21.06.04.31; Sun, 21 May 2023 06:04:31 -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=@gmail.com header.s=20221208 header.b=E7eMq7qX; 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=gmail.com Received: from [127.0.1.1] (localhost [127.0.0.1]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id 0693768C122; Sun, 21 May 2023 16:03:46 +0300 (EEST) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from mail-pf1-f182.google.com (mail-pf1-f182.google.com [209.85.210.182]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTPS id 393FB68C125 for ; Sun, 21 May 2023 16:03:38 +0300 (EEST) Received: by mail-pf1-f182.google.com with SMTP id d2e1a72fcca58-64d30ab1ef2so1751928b3a.2 for ; Sun, 21 May 2023 06:03:38 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20221208; t=1684674216; x=1687266216; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=iA3RV7oVydu4/ZTdJosSHLi2t+/NF2N9OZqI9avpdFA=; b=E7eMq7qXFsCRIaqjDZOL82JYJ7rkRl94lOUSv3/jwKSq3RLhcmwL7M6ODb8uEmQs53 ksXRU9kALCkyOAlBqVyIqwikCiaOj31PVShtIMaVMi4sQu1zCm3pwXNPyDIig3H9oi5f iv+lB2kn9UOP8QWOL4zzSWsCxKOOWftpVS+s+h6EXC+RGlwvMhs5qKoLkv8IfZf/FmTo GWEzA9O7Usyksee0mqwtXy02Eihq6q5xPmC1MCjEBs1zw2p/LCwHtsqt/9MQSv0fpTeQ 2Y0vktrjjXW8IQCtXPER8FvOu34FwGJhRH1qagr6+Mhizoi44px6uaO+IOW5xli9B1yz aJeg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1684674216; x=1687266216; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=iA3RV7oVydu4/ZTdJosSHLi2t+/NF2N9OZqI9avpdFA=; b=Qiv1/piS77cRLWBvosWZ3rw+7GB8XGgzOBh6ND78qKniBSOyqmGrMRMrrcNM9+U2Tx pM0uOMocMdR1W0Yfes5pzk1LPaf4s5Tzef+amUGalUulW2G6JiHJN64pXAUjKl6jdsoO C06yJbP6NPEizzDImzIQWpdUVMfR/naoS2XDrkgGQIH4FxVB9egaN73+KmNuTYTLnsEP 57n7etqzwSJnVErHmJIHyy7aNyZFyXfMX3+hNywIU6C7huR713KqMYhXIryNOuQo+rnX ACthfFUVE3jfx1DXeINPO+Esg+5/9YEM54gWq3HOxXX61+Nt8qoaA48ato6LfMO4TQiS uYPA== X-Gm-Message-State: AC+VfDxurAGRxsXfdsqPmYvZ9+2BjRAVCB9xlGovl3Vyb55O/FeChj4N s/QMNx4GgMbqIJkFM9hCLX2a+G5be36XmQ== X-Received: by 2002:a05:6a21:33aa:b0:10a:b2a2:a301 with SMTP id yy42-20020a056a2133aa00b0010ab2a2a301mr5127153pzb.12.1684674214781; Sun, 21 May 2023 06:03:34 -0700 (PDT) Received: from NuoMi.localdomain ([112.64.8.97]) by smtp.gmail.com with ESMTPSA id t9-20020aa79389000000b0064d45eea573sm1671872pfe.41.2023.05.21.06.03.33 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Sun, 21 May 2023 06:03:34 -0700 (PDT) From: Nuo Mi To: ffmpeg-devel@ffmpeg.org Date: Sun, 21 May 2023 21:03:09 +0800 Message-Id: <20230521130319.13813-5-nuomi2021@gmail.com> X-Mailer: git-send-email 2.25.1 In-Reply-To: <20230521130319.13813-1-nuomi2021@gmail.com> References: <20230521130319.13813-1-nuomi2021@gmail.com> MIME-Version: 1.0 Subject: [FFmpeg-devel] [PATCH 04/14] vvcdec: add cabac decoder 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: Nuo Mi Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" X-TUID: RntFXBDnMTzz add Context-based Adaptive Binary Arithmetic Coding (CABAC) decoder --- libavcodec/vvc/Makefile | 1 + libavcodec/vvc/vvc_cabac.c | 2423 ++++++++++++++++++++++++++++++++++++ libavcodec/vvc/vvc_cabac.h | 126 ++ libavcodec/vvc/vvc_ctu.h | 387 ++++++ libavcodec/vvc/vvcdec.h | 5 + 5 files changed, 2942 insertions(+) create mode 100644 libavcodec/vvc/vvc_cabac.c create mode 100644 libavcodec/vvc/vvc_cabac.h create mode 100644 libavcodec/vvc/vvc_ctu.h diff --git a/libavcodec/vvc/Makefile b/libavcodec/vvc/Makefile index 3dbfb70407..f70cc7a6f2 100644 --- a/libavcodec/vvc/Makefile +++ b/libavcodec/vvc/Makefile @@ -5,3 +5,4 @@ OBJS-$(CONFIG_VVC_DECODER) += vvc/vvcdec.o \ vvc/vvc_executor.o \ vvc/vvc_ps.o \ vvc/vvc_data.o \ + vvc/vvc_cabac.o diff --git a/libavcodec/vvc/vvc_cabac.c b/libavcodec/vvc/vvc_cabac.c new file mode 100644 index 0000000000..73a95c2d6b --- /dev/null +++ b/libavcodec/vvc/vvc_cabac.c @@ -0,0 +1,2423 @@ +/* + * VVC CABAC decoder + * + * Copyright (C) 2021 Nuo Mi + * + * This file is part of FFmpeg. + * + * FFmpeg is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * FFmpeg is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with FFmpeg; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "libavcodec/cabac_functions.h" + +#include "vvc_cabac.h" +#include "vvc_ctu.h" +#include "vvc_data.h" + +#define CABAC_MAX_BIN 31 + +#define CNU 35 + +enum SyntaxElement { + ALF_CTB_FLAG = 0, + ALF_USE_APS_FLAG = ALF_CTB_FLAG + 9, + ALF_CTB_CC_CB_IDC, + ALF_CTB_CC_CR_IDC = ALF_CTB_CC_CB_IDC + 3, + ALF_CTB_FILTER_ALT_IDX = ALF_CTB_CC_CR_IDC + 3, + SAO_MERGE_FLAG = ALF_CTB_FILTER_ALT_IDX + 2, + SAO_TYPE_IDX, + SPLIT_CU_FLAG, + SPLIT_QT_FLAG = SPLIT_CU_FLAG + 9, + MTT_SPLIT_CU_VERTICAL_FLAG = SPLIT_QT_FLAG + 6, + MTT_SPLIT_CU_BINARY_FLAG = MTT_SPLIT_CU_VERTICAL_FLAG + 5, + NON_INTER_FLAG = MTT_SPLIT_CU_BINARY_FLAG + 4, + CU_SKIP_FLAG = NON_INTER_FLAG + 2, + PRED_MODE_IBC_FLAG = CU_SKIP_FLAG + 3, + PRED_MODE_FLAG = PRED_MODE_IBC_FLAG + 3, + PRED_MODE_PLT_FLAG = PRED_MODE_FLAG + 2, + CU_ACT_ENABLED_FLAG, + INTRA_BDPCM_LUMA_FLAG, + INTRA_BDPCM_LUMA_DIR_FLAG, + INTRA_MIP_FLAG, + INTRA_LUMA_REF_IDX = INTRA_MIP_FLAG + 4, + INTRA_SUBPARTITIONS_MODE_FLAG = INTRA_LUMA_REF_IDX + 2, + INTRA_SUBPARTITIONS_SPLIT_FLAG, + INTRA_LUMA_MPM_FLAG, + INTRA_LUMA_NOT_PLANAR_FLAG, + INTRA_BDPCM_CHROMA_FLAG = INTRA_LUMA_NOT_PLANAR_FLAG + 2, + INTRA_BDPCM_CHROMA_DIR_FLAG, + CCLM_MODE_FLAG, + CCLM_MODE_IDX, + INTRA_CHROMA_PRED_MODE, + GENERAL_MERGE_FLAG, + INTER_PRED_IDC, + INTER_AFFINE_FLAG = INTER_PRED_IDC + 6, + CU_AFFINE_TYPE_FLAG = INTER_AFFINE_FLAG + 3, + SYM_MVD_FLAG, + REF_IDX_LX, + MVP_LX_FLAG = REF_IDX_LX + 2, + AMVR_FLAG, + AMVR_PRECISION_IDX = AMVR_FLAG + 2, + BCW_IDX = AMVR_PRECISION_IDX + 3, + CU_CODED_FLAG, + CU_SBT_FLAG, + CU_SBT_QUAD_FLAG = CU_SBT_FLAG + 2, + CU_SBT_HORIZONTAL_FLAG, + CU_SBT_POS_FLAG = CU_SBT_HORIZONTAL_FLAG + 3, + LFNST_IDX, + MTS_IDX = LFNST_IDX + 3, + COPY_ABOVE_PALETTE_INDICES_FLAG = MTS_IDX + 4, + PALETTE_TRANSPOSE_FLAG, + RUN_COPY_FLAG, + REGULAR_MERGE_FLAG = RUN_COPY_FLAG + 8, + MMVD_MERGE_FLAG = REGULAR_MERGE_FLAG + 2, + MMVD_CAND_FLAG, + MMVD_DISTANCE_IDX, + CIIP_FLAG, + MERGE_SUBBLOCK_FLAG, + MERGE_SUBBLOCK_IDX = MERGE_SUBBLOCK_FLAG + 3, + MERGE_IDX, + ABS_MVD_GREATER0_FLAG, + ABS_MVD_GREATER1_FLAG, + TU_Y_CODED_FLAG, + TU_CB_CODED_FLAG = TU_Y_CODED_FLAG + 4, + TU_CR_CODED_FLAG = TU_CB_CODED_FLAG + 2, + CU_QP_DELTA_ABS = TU_CR_CODED_FLAG + 3, + CU_CHROMA_QP_OFFSET_FLAG = CU_QP_DELTA_ABS + 2, + CU_CHROMA_QP_OFFSET_IDX, + TRANSFORM_SKIP_FLAG, + TU_JOINT_CBCR_RESIDUAL_FLAG = TRANSFORM_SKIP_FLAG + 2, + LAST_SIG_COEFF_X_PREFIX = TU_JOINT_CBCR_RESIDUAL_FLAG + 3, + LAST_SIG_COEFF_Y_PREFIX = LAST_SIG_COEFF_X_PREFIX +23, + SB_CODED_FLAG = LAST_SIG_COEFF_Y_PREFIX +23, + SIG_COEFF_FLAG = SB_CODED_FLAG + 7, + PAR_LEVEL_FLAG = SIG_COEFF_FLAG +63, + ABS_LEVEL_GTX_FLAG = PAR_LEVEL_FLAG +33, + COEFF_SIGN_FLAG = ABS_LEVEL_GTX_FLAG +72, + SYNTAX_ELEMENT_LAST = COEFF_SIGN_FLAG + 6, +}; + +static const uint8_t init_values[4][SYNTAX_ELEMENT_LAST] = { + { + //alf_ctb_flag + 62, 39, 39, 54, 39, 39, 31, 39, 39, + //alf_use_aps_flag + 46, + //alf_ctb_cc_cb_idc + 18, 30, 31, + //alf_ctb_cc_cr_idc + 18, 30, 31, + //alf_ctb_filter_alt_idx + 11, 11, + //sao_merge_left_flag and sao_merge_up_flag + 60, + //sao_type_idx_luma and sao_type_idx_chroma + 13, + //split_cu_flag + 19, 28, 38, 27, 29, 38, 20, 30, 31, + //split_qt_flag + 27, 6, 15, 25, 19, 37, + //mtt_split_cu_vertical_flag + 43, 42, 29, 27, 44, + //mtt_split_cu_binary_flag + 36, 45, 36, 45, + //non_inter_flag + CNU, CNU, + //cu_skip_flag + 0, 26, 28, + //pred_mode_ibc_flag + 17, 42, 36, + //pred_mode_flag + CNU, CNU, + //pred_mode_plt_flag + 25, + //cu_act_enabled_flag + 52, + //intra_bdpcm_luma_flag + 19, + //intra_bdpcm_luma_dir_flag + 35, + //intra_mip_flag + 33, 49, 50, 25, + //intra_luma_ref_idx + 25, 60, + //intra_subpartitions_mode_flag + 33, + //intra_subpartitions_split_flag + 43, + //intra_luma_mpm_flag + 45, + //intra_luma_not_planar_flag + 13, 28, + //intra_bdpcm_chroma_flag + 1, + //intra_bdpcm_chroma_dir_flag + 27, + //cclm_mode_flag + 59, + //cclm_mode_idx + 27, + //intra_chroma_pred_mode + 34, + //general_merge_flag + 26, + //inter_pred_idc + CNU, CNU, CNU, CNU, CNU, CNU, + //inter_affine_flag + CNU, CNU, CNU, + //cu_affine_type_flag + CNU, + //sym_mvd_flag + CNU, + //ref_idx_l0 and ref_idx_l1 + CNU, CNU, + //mvp_l0_flag and mvp_l1_flag + 42, + //amvr_flag + CNU, CNU, + //amvr_precision_idx + 35, 34, 35, + //bcw_idx + CNU, + //cu_coded_flag + 6, + //cu_sbt_flag + CNU, CNU, + //cu_sbt_quad_flag + CNU, + //cu_sbt_horizontal_flag + CNU, CNU, CNU, + //cu_sbt_pos_flag + CNU, + //lfnst_idx + 28, 52, 42, + //mts_idx + 29, 0, 28, 0, + //copy_above_palette_indices_flag + 42, + //palette_transpose_flag + 42, + //run_copy_flag + 50, 37, 45, 30, 46, 45, 38, 46, + //regular_merge_flag + CNU, CNU, + //mmvd_merge_flag + CNU, + //mmvd_cand_flag + CNU, + //mmvd_distance_idx + CNU, + //ciip_flag + CNU, + //merge_subblock_flag + CNU, CNU, CNU, + //merge_subblock_idx + CNU, + //merge_idx, merge_gpm_idx0, and merge_gpm_idx1 + 34, + //abs_mvd_greater0_flag + 14, + //abs_mvd_greater1_flag + 45, + //tu_y_coded_flag + 15, 12, 5, 7, + //tu_cb_coded_flag + 12, 21, + //tu_cr_coded_flag + 33, 28, 36, + //cu_qp_delta_abs + CNU, CNU, + //cu_chroma_qp_offset_flag + CNU, + //cu_chroma_qp_offset_idx + CNU, + //transform_skip_flag + 25, 9, + //tu_joint_cbcr_residual_flag + 12, 21, 35, + //last_sig_coeff_x_prefix + 13, 5, 4, 21, 14, 4, 6, 14, 21, 11, 14, 7, 14, 5, 11, 21, + 30, 22, 13, 42, 12, 4, 3, + //last_sig_coeff_y_prefix + 13, 5, 4, 6, 13, 11, 14, 6, 5, 3, 14, 22, 6, 4, 3, 6, + 22, 29, 20, 34, 12, 4, 3, + //sb_coded_flag + 18, 31, 25, 15, 18, 20, 38, + //sig_coeff_flag + 25, 19, 28, 14, 25, 20, 29, 30, 19, 37, 30, 38, 11, 38, 46, 54, + 27, 39, 39, 39, 44, 39, 39, 39, 18, 39, 39, 39, 27, 39, 39, 39, + 0, 39, 39, 39, 25, 27, 28, 37, 34, 53, 53, 46, 19, 46, 38, 39, + 52, 39, 39, 39, 11, 39, 39, 39, 19, 39, 39, 39, 25, 28, 38, + //par_level_flag + 33, 25, 18, 26, 34, 27, 25, 26, 19, 42, 35, 33, 19, 27, 35, 35, + 34, 42, 20, 43, 20, 33, 25, 26, 42, 19, 27, 26, 50, 35, 20, 43, + 11, + //abs_level_gtx_flag + 25, 25, 11, 27, 20, 21, 33, 12, 28, 21, 22, 34, 28, 29, 29, 30, + 36, 29, 45, 30, 23, 40, 33, 27, 28, 21, 37, 36, 37, 45, 38, 46, + 25, 1, 40, 25, 33, 11, 17, 25, 25, 18, 4, 17, 33, 26, 19, 13, + 33, 19, 20, 28, 22, 40, 9, 25, 18, 26, 35, 25, 26, 35, 28, 37, + 11, 5, 5, 14, 10, 3, 3, 3, + //coeff_sign_flag + 12, 17, 46, 28, 25, 46, + }, + { + //alf_ctb_flag + 13, 23, 46, 4, 61, 54, 19, 46, 54, + //alf_use_aps_flag + 46, + //alf_ctb_cc_cb_idc + 18, 21, 38, + //alf_ctb_cc_cr_idc + 18, 21, 38, + //alf_ctb_filter_alt_idx + 20, 12, + //sao_merge_left_flag and sao_merge_up_flag + 60, + //sao_type_idx_luma and sao_type_idx_chroma + 5, + //split_cu_flag + 11, 35, 53, 12, 6, 30, 13, 15, 31, + //split_qt_flag + 20, 14, 23, 18, 19, 6, + //mtt_split_cu_vertical_flag + 43, 35, 37, 34, 52, + //mtt_split_cu_binary_flag + 43, 37, 21, 22, + //non_inter_flag + 25, 12, + //cu_skip_flag + 57, 59, 45, + //pred_mode_ibc_flag + 0, 57, 44, + //pred_mode_flag + 40, 35, + //pred_mode_plt_flag + 0, + //cu_act_enabled_flag + 46, + //intra_bdpcm_luma_flag + 40, + //intra_bdpcm_luma_dir_flag + 36, + //intra_mip_flag + 41, 57, 58, 26, + //intra_luma_ref_idx + 25, 58, + //intra_subpartitions_mode_flag + 33, + //intra_subpartitions_split_flag + 36, + //intra_luma_mpm_flag + 36, + //intra_luma_not_planar_flag + 12, 20, + //intra_bdpcm_chroma_flag + 0, + //intra_bdpcm_chroma_dir_flag + 13, + //cclm_mode_flag + 34, + //cclm_mode_idx + 27, + //intra_chroma_pred_mode + 25, + //general_merge_flag + 21, + //inter_pred_idc + 7, 6, 5, 12, 4, 40, + //inter_affine_flag + 12, 13, 14, + //cu_affine_type_flag + 35, + //sym_mvd_flag + 28, + //ref_idx_l0 and ref_idx_l1 + 20, 35, + //mvp_l0_flag and mvp_l1_flag + 34, + //amvr_flag + 59, 58, + //amvr_precision_idx + 60, 48, 60, + //bcw_idx + 4, + //cu_coded_flag + 5, + //cu_sbt_flag + 56, 57, + //cu_sbt_quad_flag + 42, + //cu_sbt_horizontal_flag + 20, 43, 12, + //cu_sbt_pos_flag + 28, + //lfnst_idx + 37, 45, 27, + //mts_idx + 45, 40, 27, 0, + //copy_above_palette_indices_flag + 59, + //palette_transpose_flag + 42, + //run_copy_flag + 51, 30, 30, 38, 23, 38, 53, 46, + //regular_merge_flag + 38, 7, + //mmvd_merge_flag + 26, + //mmvd_cand_flag + 43, + //mmvd_distance_idx + 60, + //ciip_flag + 57, + //merge_subblock_flag + 48, 57, 44, + //merge_subblock_idx + 5, + //merge_idx, merge_gpm_idx0, and merge_gpm_idx1 + 20, + //abs_mvd_greater0_flag + 44, + //abs_mvd_greater1_flag + 43, + //tu_y_coded_flag + 23, 5, 20, 7, + //tu_cb_coded_flag + 25, 28, + //tu_cr_coded_flag + 25, 29, 45, + //cu_qp_delta_abs + CNU, CNU, + //cu_chroma_qp_offset_flag + CNU, + //cu_chroma_qp_offset_idx + CNU, + //transform_skip_flag + 25, 9, + //tu_joint_cbcr_residual_flag + 27, 36, 45, + //last_sig_coeff_x_prefix + 6, 13, 12, 6, 6, 12, 14, 14, 13, 12, 29, 7, 6, 13, 36, 28, + 14, 13, 5, 26, 12, 4, 18, + //last_sig_coeff_y_prefix + 5, 5, 12, 6, 6, 4, 6, 14, 5, 12, 14, 7, 13, 5, 13, 21, + 14, 20, 12, 34, 11, 4, 18, + //sb_coded_flag + 25, 30, 25, 45, 18, 12, 29, + //sig_coeff_flag + 17, 41, 42, 29, 25, 49, 43, 37, 33, 58, 51, 30, 19, 38, 38, 46, + 34, 54, 54, 39, 6, 39, 39, 39, 19, 39, 54, 39, 19, 39, 39, 39, + 56, 39, 39, 39, 17, 34, 35, 21, 41, 59, 60, 38, 35, 45, 53, 54, + 44, 39, 39, 39, 34, 38, 62, 39, 26, 39, 39, 39, 40, 35, 44, + //par_level_flag + 18, 17, 33, 18, 26, 42, 25, 33, 26, 42, 27, 25, 34, 42, 42, 35, + 26, 27, 42, 20, 20, 25, 25, 26, 11, 19, 27, 33, 42, 35, 35, 43, + 3, + //abs_level_gtx_flag + 0, 17, 26, 19, 35, 21, 25, 34, 20, 28, 29, 33, 27, 28, 29, 22, + 34, 28, 44, 37, 38, 0, 25, 19, 20, 13, 14, 57, 44, 30, 30, 23, + 17, 0, 1, 17, 25, 18, 0, 9, 25, 33, 34, 9, 25, 18, 26, 20, + 25, 18, 19, 27, 29, 17, 9, 25, 10, 18, 4, 17, 33, 19, 20, 29, + 18, 11, 4, 28, 2, 10, 3, 3, + //coeff_sign_flag + 5, 10, 53, 43, 25, 46, + }, + { + //alf_ctb_flag + 33, 52, 46, 25, 61, 54, 25, 61, 54, + //alf_use_aps_flag + 46, + //alf_ctb_cc_cb_idc + 25, 35, 38, + //alf_ctb_cc_cr_idc + 25, 28, 38, + //alf_ctb_filter_alt_idx + 11, 26, + //sao_merge_left_flag and sao_merge_up_flag + 2, + //sao_type_idx_luma and sao_type_idx_chroma + 2, + //split_cu_flag + 18, 27, 15, 18, 28, 45, 26, 7, 23, + //split_qt_flag + 26, 36, 38, 18, 34, 21, + //mtt_split_cu_vertical_flag + 43, 42, 37, 42, 44, + //mtt_split_cu_binary_flag + 28, 29, 28, 29, + //non_inter_flag + 25, 20, + //cu_skip_flag + 57, 60, 46, + //pred_mode_ibc_flag + 0, 43, 45, + //pred_mode_flag + 40, 35, + //pred_mode_plt_flag + 17, + //cu_act_enabled_flag + 46, + //intra_bdpcm_luma_flag + 19, + //intra_bdpcm_luma_dir_flag + 21, + //intra_mip_flag + 56, 57, 50, 26, + //intra_luma_ref_idx + 25, 59, + //intra_subpartitions_mode_flag + 33, + //intra_subpartitions_split_flag + 43, + //intra_luma_mpm_flag + 44, + //intra_luma_not_planar_flag + 13, 6, + //intra_bdpcm_chroma_flag + 0, + //intra_bdpcm_chroma_dir_flag + 28, + //cclm_mode_flag + 26, + //cclm_mode_idx + 27, + //intra_chroma_pred_mode + 25, + //general_merge_flag + 6, + //inter_pred_idc + 14, 13, 5, 4, 3, 40, + //inter_affine_flag + 19, 13, 6, + //cu_affine_type_flag + 35, + //sym_mvd_flag + 28, + //ref_idx_l0 and ref_idx_l1 + 5, 35, + //mvp_l0_flag and mvp_l1_flag + 34, + //amvr_flag + 59, 50, + //amvr_precision_idx + 38, 26, 60, + //bcw_idx + 5, + //cu_coded_flag + 12, + //cu_sbt_flag + 41, 57, + //cu_sbt_quad_flag + 42, + //cu_sbt_horizontal_flag + 35, 51, 27, + //cu_sbt_pos_flag + 28, + //lfnst_idx + 52, 37, 27, + //mts_idx + 45, 25, 27, 0, + //copy_above_palette_indices_flag + 50, + //palette_transpose_flag + 35, + //run_copy_flag + 58, 45, 45, 30, 38, 45, 38, 46, + //regular_merge_flag + 46, 15, + //mmvd_merge_flag + 25, + //mmvd_cand_flag + 43, + //mmvd_distance_idx + 59, + //ciip_flag + 57, + //merge_subblock_flag + 25, 58, 45, + //merge_subblock_idx + 4, + //merge_idx, merge_gpm_idx0, and merge_gpm_idx1 + 18, + //abs_mvd_greater0_flag + 51, + //abs_mvd_greater1_flag + 36, + //tu_y_coded_flag + 15, 6, 5, 14, + //tu_cb_coded_flag + 25, 37, + //tu_cr_coded_flag + 9, 36, 45, + //cu_qp_delta_abs + CNU, CNU, + //cu_chroma_qp_offset_flag + CNU, + //cu_chroma_qp_offset_idx + CNU, + //transform_skip_flag + 25, 17, + //tu_joint_cbcr_residual_flag + 42, 43, 52, + //last_sig_coeff_x_prefix + 6, 6, 12, 14, 6, 4, 14, 7, 6, 4, 29, 7, 6, 6, 12, 28, + 7, 13, 13, 35, 19, 5, 4, + //last_sig_coeff_y_prefix + 5, 5, 20, 13, 13, 19, 21, 6, 12, 12, 14, 14, 5, 4, 12, 13, + 7, 13, 12, 41, 11, 5, 27, + //sb_coded_flag + 25, 45, 25, 14, 18, 35, 45, + //sig_coeff_flag + 17, 41, 49, 36, 1, 49, 50, 37, 48, 51, 58, 45, 26, 45, 53, 46, + 49, 54, 61, 39, 35, 39, 39, 39, 19, 54, 39, 39, 50, 39, 39, 39, + 0, 39, 39, 39, 9, 49, 50, 36, 48, 59, 59, 38, 34, 45, 38, 31, + 58, 39, 39, 39, 34, 38, 54, 39, 41, 39, 39, 39, 25, 50, 37, + //par_level_flag + 33, 40, 25, 41, 26, 42, 25, 33, 26, 34, 27, 25, 41, 42, 42, 35, + 33, 27, 35, 42, 43, 33, 25, 26, 34, 19, 27, 33, 42, 43, 35, 43, + 11, + //abs_level_gtx_flag + 0, 0, 33, 34, 35, 21, 25, 34, 35, 28, 29, 40, 42, 43, 29, 30, + 49, 36, 37, 45, 38, 0, 40, 34, 43, 36, 37, 57, 52, 45, 38, 46, + 25, 0, 0, 17, 25, 26, 0, 9, 25, 33, 19, 0, 25, 33, 26, 20, + 25, 33, 27, 35, 22, 25, 1, 25, 33, 26, 12, 25, 33, 27, 28, 37, + 19, 11, 4, 6, 3, 4, 4, 5, + //coeff_sign_flag + 35, 25, 46, 28, 33, 38, + }, + //shiftIdx + { + //alf_ctb_flag + 0, 0, 0, 4, 0, 0, 1, 0, 0, + //alf_use_aps_flag + 0, + //alf_ctb_cc_cb_idc + 4, 1, 4, + //alf_ctb_cc_cr_idc + 4, 1, 4, + //alf_ctb_filter_alt_idx + 0, 0, + //sao_merge_left_flag and sao_merge_up_flag + 0, + //sao_type_idx_luma and sao_type_idx_chroma + 4, + //split_cu_flag + 12, 13, 8, 8, 13, 12, 5, 9, 9, + //split_qt_flag + 0, 8, 8, 12, 12, 8, + //mtt_split_cu_vertical_flag + 9, 8, 9, 8, 5, + //mtt_split_cu_binary_flag + 12, 13, 12, 13, + //non_inter_flag + 1, 0, + //cu_skip_flag + 5, 4, 8, + //pred_mode_ibc_flag + 1, 5, 8, + //pred_mode_flag + 5, 1, + //pred_mode_plt_flag + 1, + //cu_act_enabled_flag + 1, + //intra_bdpcm_luma_flag + 1, + //intra_bdpcm_luma_dir_flag + 4, + //intra_mip_flag + 9, 10, 9, 6, + //intra_luma_ref_idx + 5, 8, + //intra_subpartitions_mode_flag + 9, + //intra_subpartitions_split_flag + 2, + //intra_luma_mpm_flag + 6, + //intra_luma_not_planar_flag + 1, 5, + //intra_bdpcm_chroma_flag + 1, + //intra_bdpcm_chroma_dir_flag + 0, + //cclm_mode_flag + 4, + //cclm_mode_idx + 9, + //intra_chroma_pred_mode + 5, + //general_merge_flag + 4, + //inter_pred_idc + 0, 0, 1, 4, 4, 0, + //inter_affine_flag + 4, 0, 0, + //cu_affine_type_flag + 4, + //sym_mvd_flag + 5, + //ref_idx_l0 and ref_idx_l1 + 0, 4, + //mvp_l0_flag and mvp_l1_flag + 12, + //amvr_flag + 0, 0, + //amvr_precision_idx + 4, 5, 0, + //bcw_idx + 1, + //cu_coded_flag + 4, + //cu_sbt_flag + 1, 5, + //cu_sbt_quad_flag + 10, + //cu_sbt_horizontal_flag + 8, 4, 1, + //cu_sbt_pos_flag + 13, + //lfnst_idx + 9, 9, 10, + //mts_idx + 8, 0, 9, 0, + //copy_above_palette_indices_flag + 9, + //palette_transpose_flag + 5, + //run_copy_flag + 9, 6, 9, 10, 5, 0, 9, 5, + //regular_merge_flag + 5, 5, + //mmvd_merge_flag + 4, + //mmvd_cand_flag + 10, + //mmvd_distance_idx + 0, + //ciip_flag + 1, + //merge_subblock_flag + 4, 4, 4, + //merge_subblock_idx + 0, + //merge_idx, merge_gpm_idx0, and merge_gpm_idx1 + 4, + //abs_mvd_greater0_flag + 9, + //abs_mvd_greater1_flag + 5, + //tu_y_coded_flag + 5, 1, 8, 9, + //tu_cb_coded_flag + 5, 0, + //tu_cr_coded_flag + 2, 1, 0, + //cu_qp_delta_abs + 8, 8, + //cu_chroma_qp_offset_flag + 8, + //cu_chroma_qp_offset_idx + 8, + //transform_skip_flag + 1, 1, + //tu_joint_cbcr_residual_flag + 1, 1, 0, + //last_sig_coeff_x_prefix + 8, 5, 4, 5, 4, 4, 5, 4, 1, 0, 4, 1, 0, 0, 0, 0, + 1, 0, 0, 0, 5, 4, 4, + //last_sig_coeff_y_prefix + 8, 5, 8, 5, 5, 4, 5, 5, 4, 0, 5, 4, 1, 0, 0, 1, + 4, 0, 0, 0, 6, 5, 5, + //sb_coded_flag + 8, 5, 5, 8, 5, 8, 8, + //sig_coeff_flag + 12, 9, 9, 10, 9, 9, 9, 10, 8, 8, 8, 10, 9, 13, 8, 8, + 8, 8, 8, 5, 8, 0, 0, 0, 8, 8, 8, 8, 8, 0, 4, 4, + 0, 0, 0, 0, 12, 12, 9, 13, 4, 5, 8, 9, 8, 12, 12, 8, + 4, 0, 0, 0, 8, 8, 8, 8, 4, 0, 0, 0, 13, 13, 8, + //par_level_flag + 8, 9, 12, 13, 13, 13, 10, 13, 13, 13, 13, 13, 13, 13, 13, 13, + 10, 13, 13, 13, 13, 8, 12, 12, 12, 13, 13, 13, 13, 13, 13, 13, + 6, + //abs_level_gtx_flag + 9, 5, 10, 13, 13, 10, 9, 10, 13, 13, 13, 9, 10, 10, 10, 13, + 8, 9, 10, 10, 13, 8, 8, 9, 12, 12, 10, 5, 9, 9, 9, 13, + 1, 5, 9, 9, 9, 6, 5, 9, 10, 10, 9, 9, 9, 9, 9, 9, + 6, 8, 9, 9, 10, 1, 5, 8, 8, 9, 6, 6, 9, 8, 8, 9, + 4, 2, 1, 6, 1, 1, 1, 1, + //coeff_sign_flag + 1, 4, 4, 5, 8, 8, + } +}; + +#define MAX_SUB_BLOCKS 16 +#define MAX_SUB_BLOCK_SIZE 4 +#define MAX_TB_SIZE 64 + +typedef struct ResidualCoding { + //common for ts and non ts + TransformBlock *tb; + + int log2_sb_w; + int log2_sb_h; + int last_sub_block; + int num_sb_coeff; + int rem_bins_pass1; + + int width_in_sbs; + int height_in_sbs; + int nb_sbs; + + const uint8_t *sb_scan_x_off; + const uint8_t *sb_scan_y_off; + const uint8_t *scan_x_off; + const uint8_t *scan_y_off; + + uint8_t sb_coded_flag[MAX_SUB_BLOCKS * MAX_SUB_BLOCKS]; + int sig_coeff_flag[MAX_TB_SIZE * MAX_TB_SIZE]; + int abs_level_pass1[MAX_TB_SIZE * MAX_TB_SIZE]; ///< AbsLevelPass1[][] + int abs_level[MAX_TB_SIZE * MAX_TB_SIZE]; + + //for ts only + uint8_t infer_sb_cbf; + int coeff_sign_level[MAX_TB_SIZE * MAX_TB_SIZE]; ///< CoeffSignLevel[][] + + //for non ts only + int qstate; + int last_scan_pos; + int last_significant_coeff_x; + int last_significant_coeff_y; +} ResidualCoding; + +static int cabac_reinit(VVCLocalContext *lc) +{ + return skip_bytes(&lc->ep->cc, 0) == NULL ? AVERROR_INVALIDDATA : 0; +} + +static void cabac_init_state(VVCLocalContext *lc) +{ + const VVCSH *sh = &lc->sc->sh; + const int qp = av_clip_uintp2(sh->slice_qp_y, 6); + int init_type = 2 - sh->slice_type; + + av_assert0(VVC_CONTEXTS == SYNTAX_ELEMENT_LAST); + + if (sh->cabac_init_flag && sh->slice_type != VVC_SLICE_TYPE_I) + init_type ^= 3; + + for (int i = 0; i < VVC_CONTEXTS; i++) { + VVCCabacState *state = &lc->ep->cabac_state[i]; + const int init_value = init_values[init_type][i]; + const int shift_idx = init_values[3][i]; + const int m = (init_value >> 3) - 4; + const int n = ((init_value & 7) * 18) + 1; + const int pre = av_clip(((m * (qp - 16)) >> 1) + n, 1, 127); + + state->state[0] = pre << 3; + state->state[1] = pre << 7; + state->shift[0] = (shift_idx >> 2 ) + 2; + state->shift[1] = (shift_idx & 3 ) + 3 + state->shift[0]; + } +} + +int ff_vvc_cabac_init(VVCLocalContext *lc, const int ctb_addr, + const int ctb_addr_x, const int ctb_addr_y) +{ + int ret = 0; + const VVCPPS *pps = lc->fc->ps.pps; + const int first_ctb_in_slice = !ctb_addr; + const int first_ctb_in_tile = ctb_addr_x == pps->ctb_to_col_bd[ctb_addr_x] && ctb_addr_y == pps->ctb_to_row_bd[ctb_addr_y]; + + if (first_ctb_in_slice|| first_ctb_in_tile) { + if (lc->sc->nb_eps == 1 && !first_ctb_in_slice) + ret = cabac_reinit(lc); + if (ret == 0) + cabac_init_state(lc); + } + return ret; +} + +//fixme +static void vvc_refill2(CABACContext* c) { + int i; + unsigned x; +#if !HAVE_FAST_CLZ + x = c->low ^ (c->low - 1); + i = 7 - ff_h264_norm_shift[x >> (CABAC_BITS - 1)]; +#else + i = ff_ctz(c->low) - CABAC_BITS; +#endif + + x = -CABAC_MASK; + +#if CABAC_BITS == 16 + x += (c->bytestream[0] << 9) + (c->bytestream[1] << 1); +#else + x += c->bytestream[0] << 1; +#endif + + c->low += x << i; +#if !UNCHECKED_BITSTREAM_READER + if (c->bytestream < c->bytestream_end) +#endif + c->bytestream += CABAC_BITS / 8; +} + +static int inline vvc_get_cabac(CABACContext *c, VVCCabacState* base, const int ctx) +{ + VVCCabacState *s = base + ctx; + const int qRangeIdx = c->range >> 5; + const int pState = s->state[1] + (s->state[0] << 4); + const int valMps = pState >> 14; + const int RangeLPS = (qRangeIdx * ((valMps ? 32767 - pState : pState) >> 9 ) >> 1) + 4; + int bit, lps_mask; + + c->range -= RangeLPS; + lps_mask = ((c->range<<(CABAC_BITS+1)) - c->low)>>31; + + c->low -= (c->range<<(CABAC_BITS+1)) & lps_mask; + c->range += (RangeLPS - c->range) & lps_mask; + + bit = valMps ^ (lps_mask & 1); + + lps_mask = ff_h264_norm_shift[c->range]; + c->range <<= lps_mask; + c->low <<= lps_mask; + + if (!(c->low & CABAC_MASK)) + vvc_refill2(c); + s->state[0] = s->state[0] - (s->state[0] >> s->shift[0]) + (1023 * bit >> s->shift[0]); + s->state[1] = s->state[1] - (s->state[1] >> s->shift[1]) + (16383 * bit >> s->shift[1]); + return bit; +} + +#define GET_CABAC(ctx) vvc_get_cabac(&lc->ep->cc, lc->ep->cabac_state, ctx) + +static av_always_inline int vvc_get_cabac_bypass(CABACContext *c) +{ + return get_cabac_bypass(c); +} + +//9.3.3.4 Truncated binary (TB) binarization process +static int truncated_binary_decode(VVCLocalContext *lc, const int c_max) +{ + const int n = c_max + 1; + const int k = av_log2(n); + const int u = (1 << (k+1)) - n; + int v = 0; + for (int i = 0; i < k; i++) + v = (v << 1) | vvc_get_cabac_bypass(&lc->ep->cc); + if (v >= u) { + v = (v << 1) | vvc_get_cabac_bypass(&lc->ep->cc); + v -= u; + } + return v; +} + +// 9.3.3.6 Limited k-th order Exp-Golomb binarization process +static int limited_kth_order_egk_decode(CABACContext *c, const int k, const int max_pre_ext_len, const int trunc_suffix_len) +{ + int pre_ext_len = 0; + int escape_length; + int val = 0; + while ((pre_ext_len < max_pre_ext_len) && vvc_get_cabac_bypass(c)) + pre_ext_len++; + if (pre_ext_len == max_pre_ext_len) + escape_length = trunc_suffix_len; + else + escape_length = pre_ext_len + k; + while (escape_length-- > 0) { + val = (val << 1) + vvc_get_cabac_bypass(c); + } + val += ((1 << pre_ext_len) - 1) << k; + return val; +} + +static av_always_inline +void get_left_top(const VVCLocalContext *lc, uint8_t *left, uint8_t *top, + const int x0, const int y0, const uint8_t *left_ctx, const uint8_t *top_ctx) +{ + const VVCFrameContext *fc = lc->fc; + const VVCSPS *sps = fc->ps.sps; + const int min_cb_width = fc->ps.pps->min_cb_width; + const int x0b = av_mod_uintp2(x0, sps->ctb_log2_size_y); + const int y0b = av_mod_uintp2(y0, sps->ctb_log2_size_y); + const int x_cb = x0 >> sps->min_cb_log2_size_y; + const int y_cb = y0 >> sps->min_cb_log2_size_y; + + if (lc->ctb_left_flag || x0b) + *left = SAMPLE_CTB(left_ctx, x_cb - 1, y_cb); + if (lc->ctb_up_flag || y0b) + *top = SAMPLE_CTB(top_ctx, x_cb, y_cb - 1); +} + +static av_always_inline +uint8_t get_inc(VVCLocalContext *lc, const uint8_t *ctx) +{ + uint8_t left = 0, top = 0; + get_left_top(lc, &left, &top, lc->cu->x0, lc->cu->y0, ctx, ctx); + return left + top; +} + +int ff_vvc_sao_merge_flag_decode(VVCLocalContext *lc) +{ + return GET_CABAC(SAO_MERGE_FLAG); +} + +int ff_vvc_sao_type_idx_decode(VVCLocalContext *lc) +{ + if (!GET_CABAC(SAO_TYPE_IDX)) + return SAO_NOT_APPLIED; + + if (!vvc_get_cabac_bypass(&lc->ep->cc)) + return SAO_BAND; + return SAO_EDGE; +} + +int ff_vvc_sao_band_position_decode(VVCLocalContext *lc) +{ + int i; + int value = vvc_get_cabac_bypass(&lc->ep->cc); + + for (i = 0; i < 4; i++) + value = (value << 1) | vvc_get_cabac_bypass(&lc->ep->cc); + return value; +} + +int ff_vvc_sao_offset_abs_decode(VVCLocalContext *lc) +{ + int i = 0; + int length = (1 << (FFMIN(lc->fc->ps.sps->bit_depth, 10) - 5)) - 1; + + while (i < length && vvc_get_cabac_bypass(&lc->ep->cc)) + i++; + return i; +} + +int ff_vvc_sao_offset_sign_decode(VVCLocalContext *lc) +{ + return vvc_get_cabac_bypass(&lc->ep->cc); +} + +int ff_vvc_sao_eo_class_decode(VVCLocalContext *lc) +{ + int ret = vvc_get_cabac_bypass(&lc->ep->cc) << 1; + ret |= vvc_get_cabac_bypass(&lc->ep->cc); + return ret; +} + +int ff_vvc_alf_ctb_flag(VVCLocalContext *lc, const int rx, const int ry, const int c_idx) +{ + int inc = c_idx * 3; + const VVCFrameContext *fc = lc->fc; + if (lc->ctb_left_flag) { + const ALFParams *left = &CTB(fc->tab.alf, rx - 1, ry); + inc += left->ctb_flag[c_idx]; + } + if (lc->ctb_up_flag) { + const ALFParams *above = &CTB(fc->tab.alf, rx, ry - 1); + inc += above->ctb_flag[c_idx]; + } + return GET_CABAC(ALF_CTB_FLAG + inc); +} + +int ff_vvc_alf_use_aps_flag(VVCLocalContext *lc) +{ + return GET_CABAC(ALF_USE_APS_FLAG); +} + +int ff_vvc_alf_luma_prev_filter_idx(VVCLocalContext *lc) +{ + return truncated_binary_decode(lc, lc->sc->sh.alf.num_aps_ids_luma - 1); +} + +int ff_vvc_alf_luma_fixed_filter_idx(VVCLocalContext *lc) +{ + return truncated_binary_decode(lc, 15); +} + +int ff_vvc_alf_ctb_filter_alt_idx(VVCLocalContext *lc, const int c_idx, const int num_chroma_filters) +{ + int i = 0; + const int length = num_chroma_filters - 1; + + while (i < length && GET_CABAC(ALF_CTB_FILTER_ALT_IDX + c_idx - 1)) + i++; + return i; +} + +int ff_vvc_alf_ctb_cc_idc(VVCLocalContext *lc, const int rx, const int ry, const int idx, const int cc_filters_signalled) +{ + int inc = !idx ? ALF_CTB_CC_CB_IDC : ALF_CTB_CC_CR_IDC; + int i = 0; + const VVCFrameContext *fc = lc->fc; + if (lc->ctb_left_flag) { + const ALFParams *left = &CTB(fc->tab.alf, rx - 1, ry); + inc += left->ctb_cc_idc[idx] != 0; + } + if (lc->ctb_up_flag) { + const ALFParams *above = &CTB(fc->tab.alf, rx, ry - 1); + inc += above->ctb_cc_idc[idx] != 0; + } + + if (!GET_CABAC(inc)) + return 0; + i++; + while (i < cc_filters_signalled && vvc_get_cabac_bypass(&lc->ep->cc)) + i++; + return i; +} + +int ff_vvc_split_cu_flag(VVCLocalContext *lc, const int x0, const int y0, + const int cb_width, const int cb_height, const int is_chroma, const VVCAllowedSplit *a) +{ + const VVCFrameContext *fc = lc->fc; + const VVCPPS *pps = fc->ps.pps; + const int is_inside = (x0 + cb_width <= pps->width) && (y0 + cb_height <= pps->height); + + if ((a->btv || a->bth || a->ttv || a->tth || a->qt) && is_inside) + { + uint8_t inc = 0, left_height = cb_height, top_width = cb_width; + + get_left_top(lc, &left_height, &top_width, x0, y0, fc->tab.cb_height[is_chroma], fc->tab.cb_width[is_chroma]); + inc += left_height < cb_height; + inc += top_width < cb_width; + inc += (a->btv + a->bth + a->ttv + a->tth + 2 * a->qt - 1) / 2 * 3; + + return GET_CABAC(SPLIT_CU_FLAG + inc); + + } + return !is_inside; +} + +static int split_qt_flag_decode(VVCLocalContext *lc, const int x0, const int y0, const int ch_type, const int cqt_depth) +{ + const VVCFrameContext *fc = lc->fc; + int inc = 0; + uint8_t depth_left = 0, depth_top = 0; + + get_left_top(lc, &depth_left, &depth_top, x0, y0, fc->tab.cqt_depth[ch_type], fc->tab.cqt_depth[ch_type]); + inc += depth_left > cqt_depth; + inc += depth_top > cqt_depth; + inc += (cqt_depth >= 2) * 3; + + return GET_CABAC(SPLIT_QT_FLAG + inc); +} + +static int mtt_split_cu_vertical_flag_decode(VVCLocalContext *lc, const int x0, const int y0, + const int cb_width, const int cb_height, const int ch_type, const VVCAllowedSplit* a) +{ + if ((a->bth || a->tth) && (a->btv || a->ttv)) { + int inc; + const int v = a->btv + a->ttv; + const int h = a->bth + a->tth; + if (v > h) + inc = 4; + else if (v < h) + inc = 3; + else { + const VVCFrameContext *fc = lc->fc; + const VVCSPS *sps = fc->ps.sps; + const int min_cb_width = fc->ps.pps->min_cb_width; + const int x0b = av_mod_uintp2(x0, sps->ctb_log2_size_y); + const int y0b = av_mod_uintp2(y0, sps->ctb_log2_size_y); + const int x_cb = x0 >> sps->min_cb_log2_size_y; + const int y_cb = y0 >> sps->min_cb_log2_size_y; + const int available_a = lc->ctb_up_flag || y0b; + const int available_l = lc->ctb_left_flag || x0b; + const int da = cb_width / (available_a ? SAMPLE_CTB(fc->tab.cb_width[ch_type], x_cb, y_cb - 1) : 1); + const int dl = cb_height / (available_l ? SAMPLE_CTB(fc->tab.cb_height[ch_type], x_cb - 1, y_cb) : 1); + + if (da == dl || !available_a || !available_l) + inc = 0; + else if (da < dl) + inc = 1; + else + inc = 2; + } + return GET_CABAC(MTT_SPLIT_CU_VERTICAL_FLAG + inc); + } + return !(a->bth || a->tth); +} + +static int mtt_split_cu_binary_flag_decode(VVCLocalContext *lc, const int mtt_split_cu_vertical_flag, const int mtt_depth) +{ + int inc = (2 * mtt_split_cu_vertical_flag) + ((mtt_depth <= 1) ? 1 : 0); + return GET_CABAC(MTT_SPLIT_CU_BINARY_FLAG + inc); +} + +VVCSplitMode ff_vvc_split_mode(VVCLocalContext *lc, const int x0, const int y0, const int cb_width, const int cb_height, + const int cqt_depth, const int mtt_depth, const int ch_type, const VVCAllowedSplit *a) +{ + int allow_no_qt = a->btv || a->bth || a->ttv || a->tth; + int split_qt_flag; + int mtt_split_cu_vertical_flag; + int mtt_split_cu_binary_flag; + const VVCSplitMode mtt_split_modes[] = { + SPLIT_TT_HOR, SPLIT_BT_HOR, SPLIT_TT_VER, SPLIT_BT_VER, + }; + if (allow_no_qt && a->qt) { + split_qt_flag = split_qt_flag_decode(lc, x0, y0, ch_type, cqt_depth); + } else { + split_qt_flag = !allow_no_qt || a->qt; + } + if (split_qt_flag) + return SPLIT_QT; + mtt_split_cu_vertical_flag = mtt_split_cu_vertical_flag_decode(lc, x0, y0, cb_width, cb_height, ch_type, a); + if ((a->btv && a->ttv && mtt_split_cu_vertical_flag) || + (a->bth && a->tth && !mtt_split_cu_vertical_flag)) { + mtt_split_cu_binary_flag = mtt_split_cu_binary_flag_decode(lc, mtt_split_cu_vertical_flag, mtt_depth); + } else { + if (!a->btv && !a->bth) + mtt_split_cu_binary_flag = 0; + else if (!a->ttv && !a->tth) + mtt_split_cu_binary_flag = 1; + else if (a->bth && a->ttv) + mtt_split_cu_binary_flag = 1 - mtt_split_cu_vertical_flag; + else + mtt_split_cu_binary_flag = mtt_split_cu_vertical_flag; + } + return mtt_split_modes[(mtt_split_cu_vertical_flag << 1) + mtt_split_cu_binary_flag]; +} + +int ff_vvc_non_inter_flag(VVCLocalContext *lc, const int x0, const int y0, const int ch_type) +{ + const VVCFrameContext *fc = lc->fc; + uint8_t inc, left = 0, top = 0; + + get_left_top(lc, &left, &top, x0, y0, fc->tab.cpm[ch_type], fc->tab.cpm[ch_type]); + inc = left || top; + return GET_CABAC(NON_INTER_FLAG + inc); +} + +int ff_vvc_pred_mode_flag(VVCLocalContext *lc, const int is_chroma) +{ + const VVCFrameContext *fc = lc->fc; + const CodingUnit *cu = lc->cu; + uint8_t inc, left = 0, top = 0; + + get_left_top(lc, &left, &top, cu->x0, cu->y0, fc->tab.cpm[is_chroma], fc->tab.cpm[is_chroma]); + inc = left || top; + return GET_CABAC(PRED_MODE_FLAG + inc); +} + +int ff_vvc_pred_mode_plt_flag(VVCLocalContext *lc) +{ + return GET_CABAC(PRED_MODE_PLT_FLAG); +} + +int ff_vvc_intra_bdpcm_luma_flag(VVCLocalContext *lc) +{ + return GET_CABAC(INTRA_BDPCM_LUMA_FLAG); +} + +int ff_vvc_intra_bdpcm_luma_dir_flag(VVCLocalContext *lc) +{ + return GET_CABAC(INTRA_BDPCM_LUMA_DIR_FLAG); +} + +int ff_vvc_intra_bdpcm_chroma_flag(VVCLocalContext *lc) +{ + return GET_CABAC(INTRA_BDPCM_CHROMA_FLAG); +} + +int ff_vvc_intra_bdpcm_chroma_dir_flag(VVCLocalContext *lc) +{ + return GET_CABAC(INTRA_BDPCM_CHROMA_DIR_FLAG); +} + +int ff_vvc_cu_skip_flag(VVCLocalContext *lc, const uint8_t *cu_skip_flag) +{ + const int inc = get_inc(lc, cu_skip_flag); + return GET_CABAC(CU_SKIP_FLAG + inc); +} + +int ff_vvc_pred_mode_ibc_flag(VVCLocalContext *lc, const int is_chroma) +{ + const VVCFrameContext *fc = lc->fc; + const CodingUnit *cu = lc->cu; + uint8_t left_mode = MODE_INTER, top_mode = MODE_INTER; + int inc; + + get_left_top(lc, &left_mode, &top_mode, cu->x0, cu->y0, fc->tab.cpm[is_chroma], fc->tab.cpm[is_chroma]); + inc = (left_mode == MODE_IBC) + (top_mode == MODE_IBC); + return GET_CABAC(PRED_MODE_IBC_FLAG + inc); +} + +int ff_vvc_intra_mip_flag(VVCLocalContext *lc, const uint8_t *intra_mip_flag) +{ + const int w = lc->cu->cb_width; + const int h = lc->cu->cb_height; + const int inc = (w > h * 2 || h > w * 2) ? 3 : get_inc(lc, intra_mip_flag); + return GET_CABAC(INTRA_MIP_FLAG + inc); +} + +int ff_vvc_intra_mip_transposed_flag(VVCLocalContext *lc) +{ + return vvc_get_cabac_bypass(&lc->ep->cc); +} + +int ff_vvc_intra_mip_mode(VVCLocalContext *lc) +{ + const int w = lc->cu->cb_width; + const int h = lc->cu->cb_height; + const int c_max = (w == 4 && h == 4) ? 15 : + ((w == 4 || h == 4) || (w == 8 && h == 8)) ? 7: 5; + return truncated_binary_decode(lc, c_max); +} + +int ff_vvc_intra_luma_ref_idx(VVCLocalContext *lc) +{ + int i; + for (i = 0; i < 2; i++) { + if (!GET_CABAC(INTRA_LUMA_REF_IDX + i)) + return i; + } + return i; +} + +int ff_vvc_intra_subpartitions_mode_flag(VVCLocalContext *lc) +{ + return GET_CABAC(INTRA_SUBPARTITIONS_MODE_FLAG); +} + +enum IspType ff_vvc_isp_split_type(VVCLocalContext *lc, const int intra_subpartitions_mode_flag) +{ + if (!intra_subpartitions_mode_flag) + return ISP_NO_SPLIT; + return 1 + GET_CABAC(INTRA_SUBPARTITIONS_SPLIT_FLAG); +} + +int ff_vvc_intra_luma_mpm_flag(VVCLocalContext *lc) +{ + return GET_CABAC(INTRA_LUMA_MPM_FLAG); +} + +int ff_vvc_intra_luma_not_planar_flag(VVCLocalContext *lc, const int intra_subpartitions_mode_flag) +{ + return GET_CABAC(INTRA_LUMA_NOT_PLANAR_FLAG + !intra_subpartitions_mode_flag); +} + +int ff_vvc_intra_luma_mpm_idx(VVCLocalContext *lc) +{ + int i; + for (i = 0; i < 4 && vvc_get_cabac_bypass(&lc->ep->cc); i++) + /* nothing */; + return i; +} + +int ff_vvc_intra_luma_mpm_remainder(VVCLocalContext *lc) +{ + return truncated_binary_decode(lc, 60); +} + +int ff_vvc_cclm_mode_flag(VVCLocalContext *lc) +{ + return GET_CABAC(CCLM_MODE_FLAG); +} + +int ff_vvc_cclm_mode_idx(VVCLocalContext *lc) +{ + if (!GET_CABAC(CCLM_MODE_IDX)) + return 0; + return vvc_get_cabac_bypass(&lc->ep->cc) + 1; +} + +int ff_vvc_intra_chroma_pred_mode(VVCLocalContext *lc) +{ + if (!GET_CABAC(INTRA_CHROMA_PRED_MODE)) + return 4; + return (vvc_get_cabac_bypass(&lc->ep->cc) << 1) | vvc_get_cabac_bypass(&lc->ep->cc); +} + +int ff_vvc_general_merge_flag(VVCLocalContext *lc) +{ + return GET_CABAC(GENERAL_MERGE_FLAG); +} + +static int get_inter_flag_inc(VVCLocalContext *lc, const int x0, const int y0) +{ + uint8_t left_merge = 0, top_merge = 0; + uint8_t left_affine = 0, top_affine = 0; + const VVCFrameContext *fc = lc->fc; + + get_left_top(lc, &left_merge, &top_merge, x0, y0, fc->tab.msf, fc->tab.msf); + get_left_top(lc, &left_affine, &top_affine, x0, y0, fc->tab.iaf, fc->tab.iaf); + return (left_merge || left_affine) + (top_merge + top_affine); +} + +int ff_vvc_merge_subblock_flag(VVCLocalContext *lc) +{ + const int inc = get_inter_flag_inc(lc, lc->cu->x0, lc->cu->y0); + return GET_CABAC(MERGE_SUBBLOCK_FLAG + inc); +} + +int ff_vvc_merge_subblock_idx(VVCLocalContext *lc, const int max_num_subblock_merge_cand) +{ + int i; + if (!GET_CABAC(MERGE_SUBBLOCK_IDX)) + return 0; + for (i = 1; i < max_num_subblock_merge_cand - 1 && vvc_get_cabac_bypass(&lc->ep->cc); i++) + /* nothing */; + return i; +} + +int ff_vvc_regular_merge_flag(VVCLocalContext *lc, const int cu_skip_flag) +{ + int inc = !cu_skip_flag; + return GET_CABAC(REGULAR_MERGE_FLAG + inc); +} + +int ff_vvc_mmvd_merge_flag(VVCLocalContext *lc) +{ + return GET_CABAC(MMVD_MERGE_FLAG); +} + +int ff_vvc_mmvd_cand_flag(VVCLocalContext *lc) +{ + return GET_CABAC(MMVD_CAND_FLAG); +} + +static int mmvd_distance_idx_decode(VVCLocalContext *lc) +{ + int i; + if (!GET_CABAC(MMVD_DISTANCE_IDX)) + return 0; + for (i = 1; i < 7 && vvc_get_cabac_bypass(&lc->ep->cc); i++) + /* nothing */; + return i; +} + +static int mmvd_direction_idx_decode(VVCLocalContext *lc) +{ + return (vvc_get_cabac_bypass(&lc->ep->cc) << 1) | vvc_get_cabac_bypass(&lc->ep->cc); +} + +void ff_vvc_mmvd_offset_coding(VVCLocalContext *lc, Mv *mmvd_offset, const int ph_mmvd_fullpel_only_flag) +{ + const int shift = ph_mmvd_fullpel_only_flag ? 4 : 2; + const int mmvd_distance = 1 << (mmvd_distance_idx_decode(lc) + shift); + const int mmvd_direction_idx = mmvd_direction_idx_decode(lc); + const int mmvd_signs[][2] = { {1, 0}, {-1, 0}, {0, 1}, {0, -1} }; + mmvd_offset->x = mmvd_distance * mmvd_signs[mmvd_direction_idx][0]; + mmvd_offset->y = mmvd_distance * mmvd_signs[mmvd_direction_idx][1]; +} + +static PredMode get_luma_pred_mode(VVCLocalContext *lc) +{ + const VVCFrameContext *fc = lc->fc; + const CodingUnit *cu = lc->cu; + PredMode pred_mode; + if (cu->tree_type != DUAL_TREE_CHROMA) { + pred_mode = cu->pred_mode; + } else { + const int x_cb = cu->x0 >> fc->ps.sps->min_cb_log2_size_y; + const int y_cb = cu->y0 >> fc->ps.sps->min_cb_log2_size_y; + const int min_cb_width = fc->ps.pps->min_cb_width; + pred_mode = SAMPLE_CTB(fc->tab.cpm[0], x_cb, y_cb); + } + return pred_mode; +} + +int ff_vvc_merge_idx(VVCLocalContext *lc) +{ + const VVCSPS *sps = lc->fc->ps.sps; + const int is_ibc = get_luma_pred_mode(lc) == MODE_IBC; + const int c_max = (is_ibc ? sps->max_num_ibc_merge_cand : sps->max_num_merge_cand) - 1; + int i; + + if (!GET_CABAC(MERGE_IDX)) + return 0; + + for (i = 1; i < c_max && vvc_get_cabac_bypass(&lc->ep->cc); i++) + /* nothing */; + return i; +} + +int ff_vvc_merge_gpm_partition_idx(VVCLocalContext *lc) +{ + int i = 0; + + for (int j = 0; j < 6; j++) + i = (i << 1) | vvc_get_cabac_bypass(&lc->ep->cc); + + return i; +} + +int ff_vvc_merge_gpm_idx(VVCLocalContext *lc, const int idx) +{ + const int c_max = lc->fc->ps.sps->max_num_gpm_merge_cand - idx - 1; + int i; + + if (!GET_CABAC(MERGE_IDX)) + return 0; + + for (i = 1; i < c_max && vvc_get_cabac_bypass(&lc->ep->cc); i++) + /* nothing */; + + return i; +} + +int ff_vvc_ciip_flag(VVCLocalContext *lc) +{ + return GET_CABAC(CIIP_FLAG); +} + +PredFlag ff_vvc_pred_flag(VVCLocalContext *lc, const int is_b) +{ + const int w = lc->cu->cb_width; + const int h = lc->cu->cb_height; + if (!is_b) + return PF_L0; + if (w + h > 12) { + const int log2 = av_log2(w) + av_log2(h); + const int inc = 7 - ((1 + log2)>>1); + if (GET_CABAC(INTER_PRED_IDC + inc)) + return PF_BI; + } + return PF_L0 + GET_CABAC(INTER_PRED_IDC + 5); +} + +int ff_vvc_inter_affine_flag(VVCLocalContext *lc) +{ + const int inc = get_inter_flag_inc(lc, lc->cu->x0, lc->cu->y0); + return GET_CABAC(INTER_AFFINE_FLAG + inc); +} + +int ff_vvc_cu_affine_type_flag(VVCLocalContext *lc) +{ + return GET_CABAC(CU_AFFINE_TYPE_FLAG); +} + +int ff_vvc_sym_mvd_flag(VVCLocalContext *lc) +{ + return GET_CABAC(SYM_MVD_FLAG); +} + +int ff_vvc_ref_idx_lx(VVCLocalContext *lc, const uint8_t nb_refs) +{ + const int c_max = nb_refs - 1; + const int max_ctx = FFMIN(c_max, 2); + int i = 0; + + while (i < max_ctx && GET_CABAC(REF_IDX_LX + i)) + i++; + if (i == 2) { + while (i < c_max && vvc_get_cabac_bypass(&lc->ep->cc)) + i++; + } + return i; +} + +int ff_vvc_abs_mvd_greater0_flag(VVCLocalContext *lc) +{ + return GET_CABAC(ABS_MVD_GREATER0_FLAG); +} + +int ff_vvc_abs_mvd_greater1_flag(VVCLocalContext *lc) +{ + return GET_CABAC(ABS_MVD_GREATER1_FLAG); +} + +int ff_vvc_abs_mvd_minus2(VVCLocalContext *lc) +{ + return limited_kth_order_egk_decode(&lc->ep->cc, 1, 15, 17); +} + +int ff_vvc_mvd_sign_flag(VVCLocalContext *lc) +{ + return vvc_get_cabac_bypass(&lc->ep->cc); +} + +int ff_vvc_mvp_lx_flag(VVCLocalContext *lc) +{ + return GET_CABAC(MVP_LX_FLAG); +} + +static int amvr_flag(VVCLocalContext *lc, const int inter_affine_flag) +{ + return GET_CABAC(AMVR_FLAG + inter_affine_flag); +} + +static int amvr_precision_idx(VVCLocalContext *lc, const int inc, const int c_max) +{ + int i = 0; + if (!GET_CABAC(AMVR_PRECISION_IDX + inc)) + return 0; + i++; + if (i < c_max && GET_CABAC(AMVR_PRECISION_IDX + 1)) + i++; + return i; +} + +int ff_vvc_amvr_shift(VVCLocalContext *lc, const int inter_affine_flag, + const PredMode pred_mode, const int has_amvr_flag) +{ + int amvr_shift = 2; + if (has_amvr_flag) { + if (amvr_flag(lc, inter_affine_flag)) { + int idx; + if (inter_affine_flag) { + idx = amvr_precision_idx(lc, 2, 1); + amvr_shift = idx * 4; + } else if (pred_mode == MODE_IBC) { + idx = amvr_precision_idx(lc, 1, 1); + amvr_shift = 4 + idx * 2; + } else { + static const int shifts[] = {3, 4, 6}; + idx = amvr_precision_idx(lc, 0, 2); + amvr_shift = shifts[idx]; + } + } + } + return amvr_shift; +} + +int ff_vvc_bcw_idx(VVCLocalContext *lc, const int no_backward_pred_flag) +{ + const int c_max = no_backward_pred_flag ? 4 : 2; + int i = 1; + if (!GET_CABAC(BCW_IDX)) + return 0; + while (i < c_max && vvc_get_cabac_bypass(&lc->ep->cc)) + i++; + return i; +} + +int ff_vvc_tu_cb_coded_flag(VVCLocalContext *lc) +{ + return GET_CABAC(TU_CB_CODED_FLAG + lc->cu->bdpcm_flag[1]); +} + +int ff_vvc_tu_cr_coded_flag(VVCLocalContext *lc, int tu_cb_coded_flag) +{ + return GET_CABAC(TU_CR_CODED_FLAG + (lc->cu->bdpcm_flag[1] ? 2 : tu_cb_coded_flag)); +} + +int ff_vvc_tu_y_coded_flag(VVCLocalContext *lc) +{ + const CodingUnit *cu = lc->cu; + int inc; + if (cu->bdpcm_flag[0]) + inc = 1; + else if (cu->isp_split_type == ISP_NO_SPLIT) + inc = 0; + else + inc = 2 + lc->parse.prev_tu_cbf_y; + lc->parse.prev_tu_cbf_y = GET_CABAC(TU_Y_CODED_FLAG + inc); + return lc->parse.prev_tu_cbf_y; +} + +int ff_vvc_cu_qp_delta_abs(VVCLocalContext *lc) +{ + int v, i, k; + if (!GET_CABAC(CU_QP_DELTA_ABS)) + return 0; + + // prefixVal + for (v = 1; v < 5 && GET_CABAC(CU_QP_DELTA_ABS + 1); v++) + /* nothing */; + if (v < 5) + return v; + + // 9.3.3.5 k-th order Exp-Golomb binarization process + // suffixVal + + // CuQpDeltaVal shall in the range of −( 32 + QpBdOffset / 2 ) to +( 31 + QpBdOffset / 2 ) + // so k = 6 should enough + for (k = 0; k < 6 && vvc_get_cabac_bypass(&lc->ep->cc); k++) + /* nothing */; + i = (1 << k) - 1; + v = 0; + while (k--) + v = (v << 1) + vvc_get_cabac_bypass(&lc->ep->cc); + v += i; + + return v + 5; +} + +int ff_vvc_cu_qp_delta_sign_flag(VVCLocalContext *lc) +{ + return vvc_get_cabac_bypass(&lc->ep->cc); +} + +int ff_vvc_cu_chroma_qp_offset_flag(VVCLocalContext *lc) +{ + return GET_CABAC(CU_CHROMA_QP_OFFSET_FLAG); +} +int ff_vvc_cu_chroma_qp_offset_idx(VVCLocalContext *lc) +{ + const int c_max = lc->fc->ps.pps->chroma_qp_offset_list_len_minus1; + int i; + for (i = 0; i < c_max && GET_CABAC(CU_CHROMA_QP_OFFSET_IDX); i++) + /* nothing */; + return i; +} + +static av_always_inline int last_significant_coeff_xy_prefix(VVCLocalContext *lc, + const int log2_tb_size, const int log2_zo_tb_size, const int c_idx, const int ctx) +{ + int i = 0; + int max = (log2_zo_tb_size << 1) - 1; + int ctx_offset, ctx_shift; + if (!log2_tb_size) + return 0; + if (!c_idx) { + const int offset_y[] = {0, 0, 3, 6, 10, 15}; + ctx_offset = offset_y[log2_tb_size - 1]; + ctx_shift = (log2_tb_size + 1) >> 2; + } else { + const int shifts[] = {0, 0, 0, 1, 2, 2, 2}; + ctx_offset = 20; + ctx_shift = shifts[log2_tb_size]; + } + while (i < max && GET_CABAC(ctx + (i >> ctx_shift) + ctx_offset)) + i++; + return i; +} + +static av_always_inline int last_significant_coeff_x_prefix_decode(VVCLocalContext *lc, + const int log2_tb_width, const int log2_zo_tb_width, const int c_idx) +{ + return last_significant_coeff_xy_prefix(lc, log2_tb_width, log2_zo_tb_width, c_idx, LAST_SIG_COEFF_X_PREFIX); +} + +static av_always_inline int last_significant_coeff_y_prefix_decode(VVCLocalContext *lc, + const int log2_tb_height, const int log2_zo_tb_height, const int c_idx) +{ + return last_significant_coeff_xy_prefix(lc, log2_tb_height, log2_zo_tb_height, c_idx, LAST_SIG_COEFF_Y_PREFIX); +} + +static av_always_inline int last_sig_coeff_suffix_decode(VVCLocalContext *lc, + const int last_significant_coeff_y_prefix) +{ + int i; + const int length = (last_significant_coeff_y_prefix >> 1) - 1; + int value = vvc_get_cabac_bypass(&lc->ep->cc); + + for (i = 1; i < length; i++) + value = (value << 1) | vvc_get_cabac_bypass(&lc->ep->cc); + return value; +} + +int ff_vvc_tu_joint_cbcr_residual_flag(VVCLocalContext *lc, const int tu_cb_coded_flag, const int tu_cr_coded_flag) +{ + return GET_CABAC(TU_JOINT_CBCR_RESIDUAL_FLAG + 2 * tu_cb_coded_flag + tu_cr_coded_flag - 1); +} + +int ff_vvc_transform_skip_flag(VVCLocalContext *lc, const int inc) +{ + return GET_CABAC(TRANSFORM_SKIP_FLAG + inc); +} + +//9.3.4.2.7 Derivation process for the variables locNumSig, locSumAbsPass1 +static int get_local_sum(const int *level, const int w, const int h, + const int xc, const int yc) +{ + int loc_sum = 0; + level += w * yc + xc; + if (xc < w - 1) { + loc_sum += level[1]; + if (xc < w - 2) + loc_sum += level[2]; + if (yc < h - 1) + loc_sum += level[w + 1]; + } + if (yc < h - 1) { + loc_sum += level[w]; + if (yc < h - 2) + loc_sum += level[w << 1]; + } + return loc_sum; +} + +//9.3.4.2.7 Derivation process for the variables locNumSig, locSumAbsPass1 +static int get_local_sum_ts(const int *level, const int w, const int h, const int xc, const int yc) +{ + int loc_sum = 0; + level += w * yc + xc; + if (xc > 0) + loc_sum += level[-1]; + if (yc > 0) + loc_sum += level[-w]; + return loc_sum; +} + +static int get_gtx_flag_inc(const ResidualCoding* rc, const int xc, const int yc, const int last) +{ + const TransformBlock *tb = rc->tb; + int inc; + if (last) { + const int incs[] = {0, 21, 21}; + inc = incs[tb->c_idx]; + } else { + const int d = xc + yc; + const int local_sum_sig = get_local_sum(rc->sig_coeff_flag, tb->tb_width, tb->tb_height, xc, yc); + const int loc_sum_abs_pass1 = get_local_sum(rc->abs_level_pass1, tb->tb_width, tb->tb_height, xc, yc); + const int offset = FFMIN(loc_sum_abs_pass1 - local_sum_sig, 4); + + if (!tb->c_idx) + inc = 1 + offset + (!d ? 15 : (d < 3 ? 10 : (d < 10 ? 5 : 0))); + else + inc = 22 + offset + (!d ? 5 : 0); + } + return inc; +} + +static int abs_level_gtx_flag_decode(VVCLocalContext *lc, const int inc) +{ + return GET_CABAC(ABS_LEVEL_GTX_FLAG + inc); +} + +static int par_level_flag_decode(VVCLocalContext *lc, const int inc) +{ + return GET_CABAC(PAR_LEVEL_FLAG + inc); +} + +static int par_level_flag_ts_decode(VVCLocalContext *lc) +{ + const int inc = 32; + return GET_CABAC(PAR_LEVEL_FLAG + inc); +} + +static int sb_coded_flag_decode(VVCLocalContext *lc, const uint8_t *sb_coded_flag, + const ResidualCoding *rc, const int xs, const int ys) +{ + const VVCSH *sh = &lc->sc->sh; + const TransformBlock *tb = rc->tb; + const int w = rc->width_in_sbs; + const int h = rc->height_in_sbs; + int inc; + + if (tb->ts && !sh->ts_residual_coding_disabled_flag) { + const int left = xs > 0 ? sb_coded_flag[-1] : 0; + const int above = ys > 0 ? sb_coded_flag[-w] : 0; + inc = left + above + 4; + } else { + const int right = (xs < w - 1) ? sb_coded_flag[1] : 0; + const int bottom = (ys < h - 1) ? sb_coded_flag[w] : 0; + inc = (right | bottom) + (tb->c_idx ? 2 : 0); + } + return GET_CABAC(SB_CODED_FLAG + inc); +} + +static int sig_coeff_flag_decode(VVCLocalContext *lc, const ResidualCoding* rc, const int xc, const int yc) +{ + const VVCSH *sh = &lc->sc->sh; + const TransformBlock *tb = rc->tb; + int inc; + + if (tb->ts && !sh->ts_residual_coding_disabled_flag) { + const int local_num_sig = get_local_sum_ts(rc->sig_coeff_flag, tb->tb_width, tb->tb_height, xc, yc); + inc = 60 + local_num_sig; + } else { + const int d = xc + yc; + const int loc_sum_abs_pass1 = get_local_sum(rc->abs_level_pass1, tb->tb_width, tb->tb_height, xc, yc); + + if (!tb->c_idx) { + inc = 12 * FFMAX(0, rc->qstate - 1) + FFMIN((loc_sum_abs_pass1 + 1) >> 1, 3) + ((d < 2) ? 8 : (d < 5 ? 4 : 0)); + } else { + inc = 36 + 8 * FFMAX(0, rc->qstate - 1) + FFMIN((loc_sum_abs_pass1 + 1) >> 1, 3) + (d < 2 ? 4 : 0); + } + } + return GET_CABAC(SIG_COEFF_FLAG + inc); +} + +static int abs_get_rice_param(const ResidualCoding* rc, const int xc, const int yc, const int base_level) +{ + const TransformBlock* tb = rc->tb; + const int rice_params[] = { + 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 3, 3, 3, 3, + }; + int loc_sum_abs; + + loc_sum_abs = get_local_sum(rc->abs_level, tb->tb_width, tb->tb_height, xc, yc); + loc_sum_abs = av_clip_uintp2(loc_sum_abs - base_level * 5, 5); + return rice_params[loc_sum_abs]; +} + +static int abs_decode(VVCLocalContext *lc, const int c_rice_param) +{ + const int MAX_BIN = 6; + int prefix = 0; + int suffix = 0; + int i; + + while (prefix < MAX_BIN && vvc_get_cabac_bypass(&lc->ep->cc)) + prefix++; + if (prefix < MAX_BIN) { + for (i = 0; i < c_rice_param; i++) { + suffix = (suffix << 1) | vvc_get_cabac_bypass(&lc->ep->cc); + } + } else { + suffix = limited_kth_order_egk_decode(&lc->ep->cc, c_rice_param + 1, 11, 15); + } + return suffix + (prefix << c_rice_param); +} + +static int abs_remainder_decode(VVCLocalContext *lc, const ResidualCoding* rc, const int xc, const int yc) +{ + const int c_rice_param = abs_get_rice_param(rc, xc, yc, 4); + const int rem = abs_decode(lc, c_rice_param); + return rem; +} + +static int abs_remainder_ts_decode(VVCLocalContext *lc, const ResidualCoding* rc, const int xc, const int yc) +{ + const int c_rice_param = 1; + const int rem = abs_decode(lc, c_rice_param); + return rem; +} + +static int coeff_sign_flag_decode(VVCLocalContext *lc) +{ + return vvc_get_cabac_bypass(&lc->ep->cc); +} + +//9.3.4.2.10 Derivation process of ctxInc for the syntax element coeff_sign_flag for transform skip mode +static int coeff_sign_flag_ts_decode(VVCLocalContext *lc, const CodingUnit *cu, const ResidualCoding *rc, const int xc, const int yc) +{ + const TransformBlock *tb = rc->tb; + const int w = tb->tb_width; + const int *level = rc->coeff_sign_level + yc * w + xc; + const int left_sign = (xc == 0) ? 0 : level[-1]; + const int above_sign = (yc == 0) ? 0 : level[-w]; + const int bdpcm_flag = cu->bdpcm_flag[tb->c_idx]; + int inc; + + if (left_sign == -above_sign) + inc = bdpcm_flag == 0 ? 0 : 3; + else if (left_sign >= 0 && above_sign >= 0) + inc = bdpcm_flag == 0 ? 1 : 4; + else + inc = bdpcm_flag == 0 ? 2 : 5; + return GET_CABAC(COEFF_SIGN_FLAG + inc); +} + +static int abs_level_gt1_flag_ts_decode(VVCLocalContext *lc, const CodingUnit *cu, const ResidualCoding *rc, const int xc, const int yc) +{ + const TransformBlock *tb = rc->tb; + const int *sig_coeff_flag = rc->sig_coeff_flag + yc * tb->tb_width + xc; + int inc; + + if (cu->bdpcm_flag[tb->c_idx]) { + inc = 67; + } else { + const int l = xc > 0 ? sig_coeff_flag[-1] : 0; + const int a = yc > 0 ? sig_coeff_flag[-tb->tb_width] : 0; + inc = 64 + a + l; + } + return GET_CABAC(ABS_LEVEL_GTX_FLAG + inc); +} + +static int abs_level_gtx_flag_ts_decode(VVCLocalContext *lc, const int j) +{ + const int inc = 67 + j; + return GET_CABAC(ABS_LEVEL_GTX_FLAG + inc); +} + +static const uint8_t qstate_translate_table[][2] = { + { 0, 2 }, { 2, 0 }, { 1, 3 }, { 3, 1 } +}; + +static int abs_level_decode(VVCLocalContext *lc, const ResidualCoding *rc, const int xc, const int yc) +{ + const int c_rice_param = abs_get_rice_param(rc, xc, yc, 0); + int abs_dec_level = abs_decode(lc, c_rice_param); + const int zero_pos = (rc->qstate < 2 ? 1 : 2) << c_rice_param; + if (abs_dec_level != zero_pos) { + if (abs_dec_level < zero_pos) + abs_dec_level++; + } else { + abs_dec_level = 0; + } + return abs_dec_level; +} + +static void init_residual_coding(ResidualCoding *rc, const int log2_zo_tb_width, const int log2_zo_tb_height, TransformBlock *tb) +{ + int log2_sb_w = (FFMIN(log2_zo_tb_width, log2_zo_tb_height ) < 2 ? 1 : 2 ); + int log2_sb_h = log2_sb_w; + + if ( log2_zo_tb_width + log2_zo_tb_height > 3 ) { + if ( log2_zo_tb_width < 2 ) { + log2_sb_w = log2_zo_tb_width; + log2_sb_h = 4 - log2_sb_w; + } else if ( log2_zo_tb_height < 2 ) { + log2_sb_h = log2_zo_tb_height; + log2_sb_w = 4 - log2_sb_h; + } + } + rc->log2_sb_w = log2_sb_w; + rc->log2_sb_h = log2_sb_h; + rc->num_sb_coeff = 1 << (log2_sb_w + log2_sb_h); + rc->last_sub_block = ( 1 << ( log2_zo_tb_width + log2_zo_tb_height - (log2_sb_w + log2_sb_h))) - 1; + rc->rem_bins_pass1 = (( 1 << ( log2_zo_tb_width + log2_zo_tb_height)) * 7 ) >> 2; + + + rc->sb_scan_x_off = ff_vvc_diag_scan_x[log2_zo_tb_width - log2_sb_w][log2_zo_tb_height - log2_sb_h]; + rc->sb_scan_y_off = ff_vvc_diag_scan_y[log2_zo_tb_width - log2_sb_w][log2_zo_tb_height - log2_sb_h]; + + rc->scan_x_off = ff_vvc_diag_scan_x[log2_sb_w][log2_sb_h]; + rc->scan_y_off = ff_vvc_diag_scan_y[log2_sb_w][log2_sb_h]; + + rc->infer_sb_cbf = 1; + + rc->width_in_sbs = (1 << (log2_zo_tb_width - log2_sb_w)); + rc->height_in_sbs = (1 << (log2_zo_tb_height - log2_sb_h)); + rc->nb_sbs = rc->width_in_sbs * rc->height_in_sbs; + + rc->last_scan_pos = rc->num_sb_coeff; + rc->qstate = 0; + + rc->tb = tb; +} + +static int residual_ts_coding_subblock(VVCLocalContext *lc, ResidualCoding* rc, const int i) +{ + const CodingUnit *cu = lc->cu; + TransformBlock *tb = rc->tb; + const int bdpcm_flag = cu->bdpcm_flag[tb->c_idx]; + const int xs = rc->sb_scan_x_off[i]; + const int ys = rc->sb_scan_y_off[i]; + uint8_t *sb_coded_flag = rc->sb_coded_flag + ys * rc->width_in_sbs + xs; + int infer_sb_sig_coeff_flag = 1; + int last_scan_pos_pass1 = -1, last_scan_pos_pass2 = -1, n; + int abs_level_gtx_flag[MAX_SUB_BLOCK_SIZE * MAX_SUB_BLOCK_SIZE]; + int abs_level_pass2[MAX_SUB_BLOCK_SIZE * MAX_SUB_BLOCK_SIZE]; ///< AbsLevelPass2 + + if (i != rc->last_sub_block || !rc->infer_sb_cbf) + *sb_coded_flag = sb_coded_flag_decode(lc, sb_coded_flag, rc, xs, ys); + else + *sb_coded_flag = 1; + if (*sb_coded_flag && i < rc->last_sub_block) + rc->infer_sb_cbf = 0; + + //first scan pass + for (n = 0; n < rc->num_sb_coeff && rc->rem_bins_pass1 >= 4; n++) { + const int xc = (xs << rc->log2_sb_w) + rc->scan_x_off[n]; + const int yc = (ys << rc->log2_sb_h) + rc->scan_y_off[n]; + const int off = yc * tb->tb_width + xc; + int *sig_coeff_flag = rc->sig_coeff_flag + off; + int *abs_level_pass1 = rc->abs_level_pass1 + off; + int *coeff_sign_level = rc->coeff_sign_level + off; + int par_level_flag = 0; + + abs_level_gtx_flag[n] = 0; + last_scan_pos_pass1 = n; + if (*sb_coded_flag && (n != rc->num_sb_coeff - 1 || !infer_sb_sig_coeff_flag)) { + *sig_coeff_flag = sig_coeff_flag_decode(lc, rc, xc, yc); + rc->rem_bins_pass1--; + if (*sig_coeff_flag) + infer_sb_sig_coeff_flag = 0; + } else { + *sig_coeff_flag = (n == rc->num_sb_coeff - 1) && infer_sb_sig_coeff_flag && *sb_coded_flag; + } + *coeff_sign_level = 0; + if (*sig_coeff_flag) { + *coeff_sign_level = 1 - 2 * coeff_sign_flag_ts_decode(lc, cu, rc, xc, yc); + abs_level_gtx_flag[n] = abs_level_gt1_flag_ts_decode(lc, cu, rc, xc, yc); + rc->rem_bins_pass1 -= 2; + if (abs_level_gtx_flag[n]) { + par_level_flag = par_level_flag_ts_decode(lc); + rc->rem_bins_pass1--; + } + } + *abs_level_pass1 = *sig_coeff_flag + par_level_flag + abs_level_gtx_flag[n]; + } + + //greater than x scan pass + for (n = 0; n < rc->num_sb_coeff && rc->rem_bins_pass1 >= 4; n++) { + const int xc = (xs << rc->log2_sb_w) + rc->scan_x_off[n]; + const int yc = (ys << rc->log2_sb_h) + rc->scan_y_off[n]; + const int off = yc * tb->tb_width + xc; + + abs_level_pass2[n] = rc->abs_level_pass1[off]; + for (int j = 1; j < 5 && abs_level_gtx_flag[n]; j++) { + abs_level_gtx_flag[n] = abs_level_gtx_flag_ts_decode(lc, j); + abs_level_pass2[n] += abs_level_gtx_flag[n] << 1; + rc->rem_bins_pass1--; + } + last_scan_pos_pass2 = n; + } + + /* remainder scan pass */ + for (n = 0; n < rc->num_sb_coeff; n++) { + const int xc = (xs << rc->log2_sb_w) + rc->scan_x_off[n]; + const int yc = (ys << rc->log2_sb_h) + rc->scan_y_off[n]; + const int off = yc * tb->tb_width + xc; + const int *abs_level_pass1 = rc->abs_level_pass1 + off; + int *abs_level = rc->abs_level + off; + int *coeff_sign_level = rc->coeff_sign_level + off; + int abs_remainder = 0; + + if ((n <= last_scan_pos_pass2 && abs_level_pass2[n] >= 10) || + (n > last_scan_pos_pass2 && n <= last_scan_pos_pass1 && + *abs_level_pass1 >= 2) || + (n > last_scan_pos_pass1 && *sb_coded_flag)) + abs_remainder = abs_remainder_ts_decode(lc, rc, xc, yc); + if (n <= last_scan_pos_pass2) { + *abs_level = abs_level_pass2[n] + 2 * abs_remainder; + } else if (n <= last_scan_pos_pass1) { + *abs_level = *abs_level_pass1 + 2 * abs_remainder; + } else { + *abs_level = abs_remainder; + if (abs_remainder) { + //n > lastScanPosPass1 + *coeff_sign_level = 1 - 2 * coeff_sign_flag_decode(lc); + } + } + if (!bdpcm_flag && n <= last_scan_pos_pass1) { + const int left = xc > 0 ? abs_level[-1] : 0; + const int above = yc > 0 ? abs_level[-tb->tb_width] : 0; + const int pred = FFMAX(left, above); + + if (*abs_level == 1 && pred > 0) + *abs_level = pred; + else if (*abs_level > 0 && *abs_level <= pred) + (*abs_level)--; + } + if (*abs_level) { + tb->coeffs[off] = *coeff_sign_level * *abs_level; + tb->max_scan_x = FFMAX(xc, tb->max_scan_x); + tb->max_scan_y = FFMAX(yc, tb->max_scan_y); + tb->min_scan_x = FFMIN(xc, tb->min_scan_x); + tb->min_scan_y = FFMIN(yc, tb->min_scan_y); + } else { + tb->coeffs[off] = 0; + } + } + + return 0; +} + +static int hls_residual_ts_coding(VVCLocalContext *lc, TransformBlock *tb) +{ + ResidualCoding rc; + tb->min_scan_x = tb->min_scan_y = INT_MAX; + init_residual_coding(&rc, tb->log2_tb_width, tb->log2_tb_height, tb); + for (int i = 0; i <= rc.last_sub_block; i++) { + int ret = residual_ts_coding_subblock(lc, &rc, i); + if (ret < 0) + return ret; + } + + return 0; +} + +static inline int residual_coding_subblock(VVCLocalContext *lc, ResidualCoding *rc, const int i) +{ + const VVCSH *sh = &lc->sc->sh; + TransformBlock *tb = rc->tb; + int first_sig_scan_pos_sb, last_sig_scan_pos_sb; + int first_pos_mode0, first_pos_mode1; + int infer_sb_dc_sig_coeff_flag = 0; + int n, sig_hidden_flag, sum = 0; + int abs_level_gt2_flag[MAX_SUB_BLOCK_SIZE * MAX_SUB_BLOCK_SIZE]; + const int start_qstate_sb = rc->qstate; + const int xs = rc->sb_scan_x_off[i]; + const int ys = rc->sb_scan_y_off[i]; + uint8_t *sb_coded_flag = rc->sb_coded_flag + ys * rc->width_in_sbs + xs; + + + av_assert0(rc->num_sb_coeff <= MAX_SUB_BLOCK_SIZE * MAX_SUB_BLOCK_SIZE); + if (i < rc->last_sub_block && i > 0) { + *sb_coded_flag = sb_coded_flag_decode(lc, sb_coded_flag, rc, xs, ys); + infer_sb_dc_sig_coeff_flag = 1; + } else { + *sb_coded_flag = 1; + } + if (*sb_coded_flag && (xs > 3 || ys > 3) && tb->c_idx == 0) + lc->parse.mts_zero_out_sig_coeff_flag = 0; + + if (!*sb_coded_flag) + return 0; + + first_sig_scan_pos_sb = rc->num_sb_coeff; + last_sig_scan_pos_sb = -1; + first_pos_mode0 = (i == rc->last_sub_block ? rc->last_scan_pos : rc->num_sb_coeff -1); + first_pos_mode1 = first_pos_mode0; + for (n = first_pos_mode0; n >= 0 && rc->rem_bins_pass1 >= 4; n--) { + const int xc = (xs << rc->log2_sb_w) + rc->scan_x_off[n]; + const int yc = (ys << rc->log2_sb_h) + rc->scan_y_off[n]; + const int last = (xc == rc->last_significant_coeff_x && yc == rc->last_significant_coeff_y); + int *abs_level_pass1 = rc->abs_level_pass1 + yc * tb->tb_width + xc; + int *sig_coeff_flag = rc->sig_coeff_flag + yc * tb->tb_width + xc; + + if ((n > 0 || !infer_sb_dc_sig_coeff_flag ) && !last) { + *sig_coeff_flag = sig_coeff_flag_decode(lc, rc, xc, yc); + rc->rem_bins_pass1--; + if (*sig_coeff_flag) + infer_sb_dc_sig_coeff_flag = 0; + } else { + *sig_coeff_flag = last || (!rc->scan_x_off[n] && !rc ->scan_y_off[n] && + infer_sb_dc_sig_coeff_flag); + } + *abs_level_pass1 = 0; + if (*sig_coeff_flag) { + int abs_level_gt1_flag, par_level_flag = 0; + const int inc = get_gtx_flag_inc(rc, xc, yc, last); + abs_level_gt1_flag = abs_level_gtx_flag_decode(lc, inc); + rc->rem_bins_pass1--; + if (abs_level_gt1_flag) { + par_level_flag = par_level_flag_decode(lc, inc); + abs_level_gt2_flag[n] = abs_level_gtx_flag_decode(lc, inc + 32); + rc->rem_bins_pass1 -= 2; + } else { + abs_level_gt2_flag[n] = 0; + } + if (last_sig_scan_pos_sb == -1) + last_sig_scan_pos_sb = n; + first_sig_scan_pos_sb = n; + + *abs_level_pass1 = + 1 + par_level_flag + abs_level_gt1_flag + (abs_level_gt2_flag[n] << 1); + } else { + abs_level_gt2_flag[n] = 0; + } + + if (sh->dep_quant_used_flag) + rc->qstate = qstate_translate_table[rc->qstate][*abs_level_pass1 & 1]; + + first_pos_mode1 = n - 1; + } + for (n = first_pos_mode0; n > first_pos_mode1; n--) { + const int xc = (xs << rc->log2_sb_w) + rc->scan_x_off[n]; + const int yc = (ys << rc->log2_sb_h) + rc->scan_y_off[n]; + const int *abs_level_pass1 = rc->abs_level_pass1 + yc * tb->tb_width + xc; + int *abs_level = rc->abs_level + yc * tb->tb_width + xc; + + *abs_level = *abs_level_pass1; + if (abs_level_gt2_flag[n]) + *abs_level += abs_remainder_decode(lc, rc, xc, yc) << 1; + } + for (n = first_pos_mode1; n >= 0; n--) { + const int xc = (xs << rc->log2_sb_w) + rc->scan_x_off[n]; + const int yc = (ys << rc->log2_sb_h) + rc->scan_y_off[n]; + int *abs_level = rc->abs_level + yc * tb->tb_width + xc; + + *abs_level = abs_level_decode(lc, rc, xc, yc); + if (*abs_level > 0) { + if (last_sig_scan_pos_sb == -1) + last_sig_scan_pos_sb = n; + first_sig_scan_pos_sb = n; + } + if (sh->dep_quant_used_flag) + rc->qstate = qstate_translate_table[rc->qstate][*abs_level & 1]; + } + sig_hidden_flag = sh->sign_data_hiding_used_flag && + (last_sig_scan_pos_sb - first_sig_scan_pos_sb > 3 ? 1 : 0); + + if (sh->dep_quant_used_flag) + rc->qstate = start_qstate_sb; + n = (i == rc->last_sub_block ? rc->last_scan_pos : rc->num_sb_coeff -1); + for (/* nothing */; n >= 0; n--) { + int trans_coeff_level; + const int xc = (xs << rc->log2_sb_w) + rc->scan_x_off[n]; + const int yc = (ys << rc->log2_sb_h) + rc->scan_y_off[n]; + const int off = yc * tb->tb_width + xc; + const int *abs_level = rc->abs_level + off; + + if (*abs_level > 0) { + int sign = 1; + if (!sig_hidden_flag || (n != first_sig_scan_pos_sb)) + sign = 1 - 2 * coeff_sign_flag_decode(lc); + if (sh->dep_quant_used_flag) { + trans_coeff_level = (2 * *abs_level - (rc->qstate > 1)) * sign; + } else { + trans_coeff_level = *abs_level * sign; + if (sig_hidden_flag) { + sum += *abs_level; + if (n == first_sig_scan_pos_sb && (sum % 2)) + trans_coeff_level = -trans_coeff_level; + } + } + tb->coeffs[off] = trans_coeff_level; + tb->max_scan_x = FFMAX(xc, tb->max_scan_x); + tb->max_scan_y = FFMAX(yc, tb->max_scan_y); + } + if (sh->dep_quant_used_flag) + rc->qstate = qstate_translate_table[rc->qstate][*abs_level & 1]; + } + + return 0; +} + +static void derive_last_scan_pos(ResidualCoding *rc, + const int log2_zo_tb_width, int log2_zo_tb_height) +{ + int xc, yc, xs, ys; + do { + if (rc->last_scan_pos == 0) { + rc->last_scan_pos = rc->num_sb_coeff; + rc->last_sub_block--; + } + rc->last_scan_pos--; + xs = rc->sb_scan_x_off[rc->last_sub_block]; + ys = rc->sb_scan_y_off[rc->last_sub_block]; + xc = (xs << rc->log2_sb_w) + rc->scan_x_off[rc->last_scan_pos]; + yc = (ys << rc->log2_sb_h) + rc->scan_y_off[rc->last_scan_pos]; + } while ((xc != rc->last_significant_coeff_x) || (yc != rc->last_significant_coeff_y)); +} + +static void last_significant_coeff_x_y_decode(ResidualCoding *rc, VVCLocalContext *lc, + const int log2_zo_tb_width, const int log2_zo_tb_height) +{ + const TransformBlock *tb = rc->tb; + int last_significant_coeff_x, last_significant_coeff_y; + + last_significant_coeff_x = last_significant_coeff_x_prefix_decode(lc, + tb->log2_tb_width, log2_zo_tb_width, tb->c_idx); + + last_significant_coeff_y = last_significant_coeff_y_prefix_decode(lc, + tb->log2_tb_height, log2_zo_tb_height, tb->c_idx); + + if (last_significant_coeff_x > 3) { + int suffix = last_sig_coeff_suffix_decode(lc, last_significant_coeff_x); + last_significant_coeff_x = (1 << ((last_significant_coeff_x >> 1) - 1)) * + (2 + (last_significant_coeff_x & 1)) + suffix; + } + if (last_significant_coeff_y > 3) { + int suffix = last_sig_coeff_suffix_decode(lc, last_significant_coeff_y); + last_significant_coeff_y = (1 << ((last_significant_coeff_y >> 1) - 1)) * + (2 + (last_significant_coeff_y & 1)) + suffix; + } + rc->last_significant_coeff_x = last_significant_coeff_x; + rc->last_significant_coeff_y = last_significant_coeff_y; +} + +static int hls_residual_coding(VVCLocalContext *lc, TransformBlock *tb) +{ + const VVCSPS *sps = lc->fc->ps.sps; + const CodingUnit *cu = lc->cu; + const int log2_tb_width = tb->log2_tb_width; + const int log2_tb_height = tb->log2_tb_height; + const int c_idx = tb->c_idx; + int log2_zo_tb_width, log2_zo_tb_height; + ResidualCoding rc; + + if (sps->mts_enabled_flag && cu->sbt_flag && c_idx == 0 && log2_tb_width == 5 && log2_tb_height < 6) + log2_zo_tb_width = 4; + else + log2_zo_tb_width = FFMIN(log2_tb_width, 5 ); + + if (sps->mts_enabled_flag && cu->sbt_flag && c_idx == 0 && log2_tb_width < 6 && log2_tb_height == 5 ) + log2_zo_tb_height = 4; + else + log2_zo_tb_height = FFMIN(log2_tb_height, 5); + + init_residual_coding(&rc, log2_zo_tb_width, log2_zo_tb_height, tb); + last_significant_coeff_x_y_decode(&rc, lc, log2_zo_tb_width, log2_zo_tb_height); + derive_last_scan_pos(&rc, log2_zo_tb_width, log2_zo_tb_height); + + if (rc.last_sub_block == 0 && log2_tb_width >= 2 && log2_tb_height >= 2 && !tb->ts && rc.last_scan_pos > 0) + lc->parse.lfnst_dc_only = 0; + if ((rc.last_sub_block > 0 && log2_tb_width >= 2 && log2_tb_height >= 2 ) || + (rc.last_scan_pos > 7 && (log2_tb_width == 2 || log2_tb_width == 3 ) && + log2_tb_width == log2_tb_height)) + lc->parse.lfnst_zero_out_sig_coeff_flag = 0; + if ((rc.last_sub_block > 0 || rc.last_scan_pos > 0 ) && c_idx == 0) + lc->parse.mts_dc_only = 0; + + memset(tb->coeffs, 0, tb->tb_width * tb->tb_height * sizeof(*tb->coeffs)); + memset(rc.abs_level, 0, tb->tb_width * tb->tb_height * sizeof(rc.abs_level[0])); + memset(rc.sb_coded_flag, 0, rc.nb_sbs); + memset(rc.abs_level_pass1, 0, tb->tb_width * tb->tb_height * sizeof(rc.abs_level_pass1[0])); + memset(rc.sig_coeff_flag, 0, tb->tb_width * tb->tb_height * sizeof(rc.sig_coeff_flag[0])); + + for (int i = rc.last_sub_block; i >= 0; i--) { + int ret = residual_coding_subblock(lc, &rc, i); + if (ret < 0) + return ret; + } + + return 0; +} + +int ff_vvc_residual_coding(VVCLocalContext *lc, TransformBlock *tb) +{ + const VVCSH *sh = &lc->sc->sh; + const int ts = !sh->ts_residual_coding_disabled_flag && tb->ts; + + return ts ? hls_residual_ts_coding(lc, tb) : hls_residual_coding(lc, tb); +} + +int ff_vvc_cu_coded_flag(VVCLocalContext *lc) +{ + return GET_CABAC(CU_CODED_FLAG); +} + +int ff_vvc_sbt_flag(VVCLocalContext *lc) +{ + const int w = lc->cu->cb_width; + const int h = lc->cu->cb_height; + const int inc = w * h <= 256; + return GET_CABAC(CU_SBT_FLAG + inc); +} + +int ff_vvc_sbt_quad_flag(VVCLocalContext *lc) +{ + return GET_CABAC(CU_SBT_QUAD_FLAG); +} + +int ff_vvc_sbt_horizontal_flag(VVCLocalContext *lc) +{ + const int w = lc->cu->cb_width; + const int h = lc->cu->cb_height; + const int inc = (w == h) ? 0 : ((w < h) ? 1 : 2); + return GET_CABAC(CU_SBT_HORIZONTAL_FLAG + inc); +} + +int ff_vvc_sbt_pos_flag(VVCLocalContext *lc) +{ + return GET_CABAC(CU_SBT_POS_FLAG); +} + +int ff_vvc_lfnst_idx(VVCLocalContext *lc, const int inc) +{ + if (!GET_CABAC(LFNST_IDX + inc)) + return 0; + if (!GET_CABAC(LFNST_IDX + 2)) + return 1; + return 2; +} + +int ff_vvc_mts_idx(VVCLocalContext *lc) +{ + int i; + for (i = 0; i < 4; i++) { + if (!GET_CABAC(MTS_IDX + i)) + return i; + } + return i; +} + +int ff_vvc_end_of_slice_flag_decode(VVCLocalContext *lc) +{ + return get_cabac_terminate(&lc->ep->cc); +} + +int ff_vvc_end_of_tile_one_bit(VVCLocalContext *lc) +{ + return get_cabac_terminate(&lc->ep->cc); +} + +int ff_vvc_end_of_subset_one_bit(VVCLocalContext *lc) +{ + return get_cabac_terminate(&lc->ep->cc); +} diff --git a/libavcodec/vvc/vvc_cabac.h b/libavcodec/vvc/vvc_cabac.h new file mode 100644 index 0000000000..6f56f63852 --- /dev/null +++ b/libavcodec/vvc/vvc_cabac.h @@ -0,0 +1,126 @@ +/* + * VVC CABAC decoder + * + * Copyright (C) 2022 Nuo Mi + * + * 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 + */ + +#ifndef AVCODEC_VVC_CABAC_H +#define AVCODEC_VVC_CABAC_H + +#include "vvc_ctu.h" + +int ff_vvc_cabac_init(VVCLocalContext *lc, int ctb_addr, int ctb_addr_x, int ctb_addr_y); + +//sao +int ff_vvc_sao_merge_flag_decode(VVCLocalContext *lc); +int ff_vvc_sao_type_idx_decode(VVCLocalContext *lc); +int ff_vvc_sao_band_position_decode(VVCLocalContext *lc); +int ff_vvc_sao_offset_abs_decode(VVCLocalContext *lc); +int ff_vvc_sao_offset_sign_decode(VVCLocalContext *lc); +int ff_vvc_sao_eo_class_decode(VVCLocalContext *lc); + +//alf +int ff_vvc_alf_ctb_flag(VVCLocalContext *lc, int rx, int ry, int c_idx); +int ff_vvc_alf_use_aps_flag(VVCLocalContext *lc); +int ff_vvc_alf_luma_prev_filter_idx(VVCLocalContext *lc); +int ff_vvc_alf_luma_fixed_filter_idx(VVCLocalContext *lc); +int ff_vvc_alf_ctb_filter_alt_idx(VVCLocalContext *lc, int c_idx, int num_chroma_filters); +int ff_vvc_alf_ctb_cc_idc(VVCLocalContext *lc, int rx, int ry, int idx, int cc_filters_signalled); + +//coding_tree +int ff_vvc_split_cu_flag(VVCLocalContext* lc, int x0, int y0, int cb_width, int cb_height, + int ch_type, const VVCAllowedSplit *a); +VVCSplitMode ff_vvc_split_mode(VVCLocalContext *lc, int x0, int y0, int cb_width, int cb_height, + int cqt_depth, int mtt_depth, int ch_type, const VVCAllowedSplit *a); +int ff_vvc_non_inter_flag(VVCLocalContext *lc, int x0, int y0, int ch_type); + +//coding unit +int ff_vvc_pred_mode_flag(VVCLocalContext *lc, int is_chroma); +int ff_vvc_pred_mode_plt_flag(VVCLocalContext *lc); +int ff_vvc_intra_bdpcm_luma_flag(VVCLocalContext *lc); +int ff_vvc_intra_bdpcm_luma_dir_flag(VVCLocalContext *lc); +int ff_vvc_intra_bdpcm_chroma_flag(VVCLocalContext *lc); +int ff_vvc_intra_bdpcm_chroma_dir_flag(VVCLocalContext *lc); +int ff_vvc_cu_skip_flag(VVCLocalContext *lc, const uint8_t *cu_skip_flag); +int ff_vvc_pred_mode_ibc_flag(VVCLocalContext *lc, int ch_type); +int ff_vvc_cu_coded_flag(VVCLocalContext *lc); +int ff_vvc_cu_qp_delta_abs(VVCLocalContext *lc); +int ff_vvc_cu_qp_delta_sign_flag(VVCLocalContext *lc); +int ff_vvc_sbt_flag(VVCLocalContext *lc); +int ff_vvc_sbt_quad_flag(VVCLocalContext *lc); +int ff_vvc_sbt_horizontal_flag(VVCLocalContext *lc); +int ff_vvc_sbt_pos_flag(VVCLocalContext *lc); + +//intra +int ff_vvc_intra_mip_flag(VVCLocalContext *lc, const uint8_t *intra_mip_flag); +int ff_vvc_intra_mip_transposed_flag(VVCLocalContext *lc); +int ff_vvc_intra_mip_mode(VVCLocalContext *lc); +int ff_vvc_intra_luma_ref_idx(VVCLocalContext *lc); +int ff_vvc_intra_subpartitions_mode_flag(VVCLocalContext *lc); +enum IspType ff_vvc_isp_split_type(VVCLocalContext *lc, int intra_subpartitions_mode_flag); +int ff_vvc_intra_luma_mpm_flag(VVCLocalContext *lc); +int ff_vvc_intra_luma_not_planar_flag(VVCLocalContext *lc, int intra_subpartitions_mode_flag); +int ff_vvc_intra_luma_mpm_idx(VVCLocalContext *lc); +int ff_vvc_intra_luma_mpm_remainder(VVCLocalContext *lc); +int ff_vvc_cclm_mode_flag(VVCLocalContext *lc); +int ff_vvc_cclm_mode_idx(VVCLocalContext *lc); +int ff_vvc_intra_chroma_pred_mode(VVCLocalContext *lc); + +//inter +int ff_vvc_general_merge_flag(VVCLocalContext *lc); +int ff_vvc_merge_subblock_flag(VVCLocalContext *lc); +int ff_vvc_merge_subblock_idx(VVCLocalContext *lc, int max_num_subblock_merge_cand); +int ff_vvc_regular_merge_flag(VVCLocalContext *lc, int cu_skip_flag); +int ff_vvc_merge_idx(VVCLocalContext *lc); +int ff_vvc_mmvd_merge_flag(VVCLocalContext *lc); +int ff_vvc_mmvd_cand_flag(VVCLocalContext *lc); +void ff_vvc_mmvd_offset_coding(VVCLocalContext *lc, Mv *mvd_offset, int ph_mmvd_fullpel_only_flag); +int ff_vvc_ciip_flag(VVCLocalContext *lc); +int ff_vvc_merge_gpm_partition_idx(VVCLocalContext *lc); +int ff_vvc_merge_gpm_idx(VVCLocalContext *lc, int idx); +PredFlag ff_vvc_pred_flag(VVCLocalContext *lc, int is_b); +int ff_vvc_inter_affine_flag(VVCLocalContext *lc); +int ff_vvc_cu_affine_type_flag(VVCLocalContext *lc); +int ff_vvc_sym_mvd_flag(VVCLocalContext *lc); +int ff_vvc_ref_idx_lx(VVCLocalContext *lc, uint8_t nb_refs); +int ff_vvc_abs_mvd_greater0_flag(VVCLocalContext *lc); +int ff_vvc_abs_mvd_greater1_flag(VVCLocalContext *lc); +int ff_vvc_abs_mvd_minus2(VVCLocalContext *lc); +int ff_vvc_mvd_sign_flag(VVCLocalContext *lc); +int ff_vvc_mvp_lx_flag(VVCLocalContext *lc); +int ff_vvc_amvr_shift(VVCLocalContext *lc, int inter_affine_flag, PredMode pred_mode, int has_amvr_flag); +int ff_vvc_bcw_idx(VVCLocalContext *lc, int no_backward_pred_flag); + +//transform +int ff_vvc_tu_cb_coded_flag(VVCLocalContext *lc); +int ff_vvc_tu_cr_coded_flag(VVCLocalContext *lc, int tu_cb_coded_flag); +int ff_vvc_tu_y_coded_flag(VVCLocalContext *lc); +int ff_vvc_cu_chroma_qp_offset_flag(VVCLocalContext *lc); +int ff_vvc_cu_chroma_qp_offset_idx(VVCLocalContext *lc); +int ff_vvc_tu_joint_cbcr_residual_flag(VVCLocalContext *lc, int tu_cb_coded_flag, int tu_cr_coded_flag); +int ff_vvc_transform_skip_flag(VVCLocalContext *lc, int ctx); +int ff_vvc_residual_coding(VVCLocalContext *lc, TransformBlock *tb); +int ff_vvc_lfnst_idx(VVCLocalContext *lc, int inc); +int ff_vvc_mts_idx(VVCLocalContext *lc); + +int ff_vvc_end_of_slice_flag_decode(VVCLocalContext *lc); +int ff_vvc_end_of_tile_one_bit(VVCLocalContext *lc); +int ff_vvc_end_of_subset_one_bit(VVCLocalContext *lc); + +#endif //AVCODEC_VVC_CABAC_H diff --git a/libavcodec/vvc/vvc_ctu.h b/libavcodec/vvc/vvc_ctu.h new file mode 100644 index 0000000000..7da9e15b45 --- /dev/null +++ b/libavcodec/vvc/vvc_ctu.h @@ -0,0 +1,387 @@ +/* + * VVC CTU(Coding Tree Unit) parser + * + * Copyright (C) 2022 Nuo Mi + * + * 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 + */ + +#ifndef AVCODEC_VVC_CTU_H +#define AVCODEC_VVC_CTU_H + +#include "vvcdec.h" + +enum SAOType { + SAO_NOT_APPLIED = 0, + SAO_BAND, + SAO_EDGE, + SAO_APPLIED +}; + +enum SAOEOClass { + SAO_EO_HORIZ = 0, + SAO_EO_VERT, + SAO_EO_135D, + SAO_EO_45D, +}; + +typedef struct NeighbourAvailable { + int cand_left; + int cand_up; + int cand_up_left; + int cand_up_right; + int cand_up_right_sap; +} NeighbourAvailable; + +enum IspType{ + ISP_NO_SPLIT, + ISP_HOR_SPLIT, + ISP_VER_SPLIT, +}; + +typedef enum VVCSplitMode { + SPLIT_NONE, + SPLIT_TT_HOR, + SPLIT_BT_HOR, + SPLIT_TT_VER, + SPLIT_BT_VER, + SPLIT_QT, +} VVCSplitMode; + +typedef enum MtsIdx { + MTS_DCT2_DCT2, + MTS_DST7_DST7, + MTS_DST7_DCT8, + MTS_DCT8_DST7, + MTS_DCT8_DCT8, +} MtsIdx; + +typedef struct TransformBlock { + uint8_t has_coeffs; + uint8_t c_idx; + uint8_t ts; ///< transform_skip_flag + int x0; + int y0; + + int tb_width; + int tb_height; + int log2_tb_width; + int log2_tb_height; + + int max_scan_x; + int max_scan_y; + int min_scan_x; + int min_scan_y; + + int qp; + int rect_non_ts_flag; + int bd_shift; + int bd_offset; + + int *coeffs; +} TransformBlock; + +typedef enum VVCTreeType { + SINGLE_TREE, + DUAL_TREE_LUMA, + DUAL_TREE_CHROMA, +} VVCTreeType; + +typedef struct TransformUnit { + int x0; + int y0; + int width; + int height; + + uint8_t joint_cbcr_residual_flag; ///< tu_joint_cbcr_residual_flag + + uint8_t coded_flag[VVC_MAX_SAMPLE_ARRAYS]; ///< tu_y_coded_flag, tu_cb_coded_flag, tu_cr_coded_flag + uint8_t nb_tbs; + TransformBlock tbs[VVC_MAX_SAMPLE_ARRAYS]; +} TransformUnit; + +typedef enum PredMode { + MODE_INTER, + MODE_INTRA, + MODE_SKIP, + MODE_PLT, + MODE_IBC, +} PredMode; + +typedef struct Mv { + int x; ///< horizontal component of motion vector + int y; ///< vertical component of motion vector +} Mv; + +typedef struct MvField { + DECLARE_ALIGNED(4, Mv, mv)[2]; ///< mvL0, vvL1 + int8_t ref_idx[2]; ///< refIdxL0, refIdxL1 + uint8_t hpel_if_idx; ///< hpelIfIdx + uint8_t bcw_idx; ///< bcwIdx + uint8_t pred_flag; + uint8_t ciip_flag; ///< ciip_flag +} MvField; + +typedef struct DMVRInfo { + DECLARE_ALIGNED(4, Mv, mv)[2]; ///< mvL0, vvL1 + uint8_t dmvr_enabled; +} DMVRInfo; + +typedef enum MotionModelIdc { + MOTION_TRANSLATION, + MOTION_4_PARAMS_AFFINE, + MOTION_6_PARAMS_AFFINE, +} MotionModelIdc; + +typedef enum PredFlag { + PF_INTRA = 0x0, + PF_L0 = 0x1, + PF_L1 = 0x2, + PF_BI = 0x3, +} PredFlag; + +typedef enum IntraPredMode { + INTRA_PLANAR = 0, + INTRA_DC, + INTRA_HORZ = 18, + INTRA_DIAG = 34, + INTRA_VERT = 50, + INTRA_VDIAG = 66, + INTRA_LT_CCLM = 81, + INTRA_L_CCLM, + INTRA_T_CCLM +} IntraPredMode; + +typedef struct MotionInfo { + MotionModelIdc motion_model_idc; ///< MotionModelIdc + int8_t ref_idx[2]; ///< refIdxL0, refIdxL1 + uint8_t hpel_if_idx; ///< hpelIfIdx + uint8_t bcw_idx; ///< bcwIdx + PredFlag pred_flag; + + Mv mv[2][MAX_CONTROL_POINTS]; + + int num_sb_x, num_sb_y; +} MotionInfo; + +typedef struct PredictionUnit { + uint8_t general_merge_flag; + uint8_t mmvd_merge_flag; + //InterPredIdc inter_pred_idc; + uint8_t inter_affine_flag; + + //subblock predict + uint8_t merge_subblock_flag; + + uint8_t merge_gpm_flag; + uint8_t gpm_partition_idx; + MvField gpm_mv[2]; + + int sym_mvd_flag; + + MotionInfo mi; + + int16_t diff_mv_x[2][AFFINE_MIN_BLOCK_SIZE * AFFINE_MIN_BLOCK_SIZE]; ///< diffMvLX + int16_t diff_mv_y[2][AFFINE_MIN_BLOCK_SIZE * AFFINE_MIN_BLOCK_SIZE]; ///< diffMvLX + int cb_prof_flag[2]; +} PredictionUnit; + +struct CodingUnit; +typedef struct CodingUnit { + AVBufferRef *buf; + + VVCTreeType tree_type; + int x0; + int y0; + int cb_width; + int cb_height; + int ch_type; + int cqt_depth; + + uint8_t coded_flag; + + uint8_t sbt_flag; + uint8_t sbt_horizontal_flag; + uint8_t sbt_pos_flag; + + int lfnst_idx; + MtsIdx mts_idx; + + uint8_t act_enabled_flag; + + uint8_t intra_luma_ref_idx; ///< IntraLumaRefLineIdx[][] + uint8_t intra_mip_flag; ///< intra_mip_flag + uint8_t skip_flag; ///< cu_skip_flag; + + //inter + uint8_t ciip_flag; + + // Inferred parameters + enum IspType isp_split_type; ///< IntraSubPartitionsSplitType + + enum PredMode pred_mode; ///< PredMode + + int num_intra_subpartitions; + + IntraPredMode intra_pred_mode_y; ///< IntraPredModeY + IntraPredMode intra_pred_mode_c; ///< IntraPredModeC + int mip_chroma_direct_flag; ///< MipChromaDirectFlag + + int bdpcm_flag[VVC_MAX_SAMPLE_ARRAYS]; ///< BdpcmFlag + + int apply_lfnst_flag[VVC_MAX_SAMPLE_ARRAYS]; ///< ApplyLfnstFlag[] + + TransformUnit tus[MAX_TUS_IN_CU]; + int num_tus; + + int8_t qp[4]; ///< QpY, Qp′Cb, Qp′Cr, Qp′CbCr + + PredictionUnit pu; + + struct CodingUnit *next; +} CodingUnit; + +struct CTU { + CodingUnit *cus; +}; + +typedef struct ReconstructedArea { + int x; + int y; + int w; + int h; +} ReconstructedArea; + +typedef struct VVCCabacState { + uint16_t state[2]; + uint8_t shift[2]; +} VVCCabacState; + +// VVC_CONTEXTS matched with SYNTAX_ELEMENT_LAST, it's checked by cabac_init_state. +#define VVC_CONTEXTS 378 +typedef struct EntryPoint { + int8_t qp_y; //< QpY + + VVCCabacState cabac_state[VVC_CONTEXTS]; + CABACContext cc; + + VVCTask *parse_task; + int ctu_addr_last; + + uint8_t is_first_qg; + MvField hmvp[MAX_NUM_HMVP_CANDS]; ///< HmvpCandList + int num_hmvp; ///< NumHmvpCand +} EntryPoint; + +struct VVCLocalContext { + uint8_t ctb_left_flag; + uint8_t ctb_up_flag; + uint8_t ctb_up_right_flag; + uint8_t ctb_up_left_flag; + int end_of_tiles_x; + int end_of_tiles_y; + + /* +7 is for subpixel interpolation, *2 for high bit depths */ + DECLARE_ALIGNED(32, uint8_t, edge_emu_buffer)[(MAX_PB_SIZE + 7) * EDGE_EMU_BUFFER_STRIDE * 2]; + /* The extended size between the new edge emu buffer is abused by SAO */ + DECLARE_ALIGNED(32, uint8_t, edge_emu_buffer2)[(MAX_PB_SIZE + 7) * EDGE_EMU_BUFFER_STRIDE * 2]; + DECLARE_ALIGNED(32, int16_t, tmp)[MAX_PB_SIZE * MAX_PB_SIZE]; + DECLARE_ALIGNED(32, int16_t, tmp1)[MAX_PB_SIZE * MAX_PB_SIZE]; + DECLARE_ALIGNED(32, uint8_t, ciip_tmp1)[MAX_PB_SIZE * MAX_PB_SIZE * 2]; + DECLARE_ALIGNED(32, uint8_t, ciip_tmp2)[MAX_PB_SIZE * MAX_PB_SIZE * 2]; + DECLARE_ALIGNED(32, uint8_t, sao_buffer)[(MAX_CTU_SIZE + 2 * SAO_PADDING_SIZE) * EDGE_EMU_BUFFER_STRIDE * 2]; + DECLARE_ALIGNED(32, uint8_t, alf_buffer_luma)[(MAX_CTU_SIZE + 2 * ALF_PADDING_SIZE) * EDGE_EMU_BUFFER_STRIDE * 2]; + DECLARE_ALIGNED(32, uint8_t, alf_buffer_chroma)[(MAX_CTU_SIZE + 2 * ALF_PADDING_SIZE) * EDGE_EMU_BUFFER_STRIDE * 2]; + DECLARE_ALIGNED(32, int32_t, alf_gradient_tmp)[ALF_GRADIENT_SIZE * ALF_GRADIENT_SIZE * ALF_NUM_DIR]; + + struct { + int sbt_num_fourths_tb0; ///< SbtNumFourthsTb0 + + uint8_t is_cu_qp_delta_coded; ///< IsCuQpDeltaCoded + int cu_qg_top_left_x; ///< CuQgTopLeftX + int cu_qg_top_left_y; ///< CuQgTopLeftY + int is_cu_chroma_qp_offset_coded; ///< IsCuChromaQpOffsetCoded + int chroma_qp_offset[3]; ///< CuQpOffsetCb, CuQpOffsetCr, CuQpOffsetCbCr + + int infer_tu_cbf_luma; ///< InferTuCbfLuma + int prev_tu_cbf_y; ///< prevTuCbfY; + + int lfnst_dc_only; ///< LfnstDcOnly + int lfnst_zero_out_sig_coeff_flag; ///< LfnstZeroOutSigCoeffFlag + + int mts_dc_only; ///< MtsDcOnly + int mts_zero_out_sig_coeff_flag; ///< MtsZeroOutSigCoeffFlag; + } parse; + + struct { + // lmcs cache, for recon only + int chroma_scale; + int x_vpdu; + int y_vpdu; + } lmcs; + + CodingUnit *cu; + ReconstructedArea ras[2][MAX_PARTS_IN_CTU]; + int num_ras[2]; + + NeighbourAvailable na; + +#define BOUNDARY_LEFT_SLICE (1 << 0) +#define BOUNDARY_LEFT_TILE (1 << 1) +#define BOUNDARY_UPPER_SLICE (1 << 2) +#define BOUNDARY_UPPER_TILE (1 << 3) + /* properties of the boundary of the current CTB for the purposes + * of the deblocking filter */ + int boundary_flags; + + SliceContext *sc; + VVCFrameContext *fc; + EntryPoint *ep; + int *coeffs; +} ; + +typedef struct VVCAllowedSplit { + int qt; + int btv; + int bth; + int ttv; + int tth; +} VVCAllowedSplit; + +struct SAOParams { + int offset_abs[3][4]; ///< sao_offset_abs + int offset_sign[3][4]; ///< sao_offset_sign + + uint8_t band_position[3]; ///< sao_band_position + + int eo_class[3]; ///< sao_eo_class + + int16_t offset_val[3][5]; ///