From patchwork Fri Mar 25 18:52:51 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Ben Avison X-Patchwork-Id: 34987 Delivered-To: ffmpegpatchwork2@gmail.com Received: by 2002:ab0:5fda:0:0:0:0:0 with SMTP id g26csp1432467uaj; Fri, 25 Mar 2022 11:54:27 -0700 (PDT) X-Google-Smtp-Source: ABdhPJz1lWpXn77kLG6XZRT7n2RCmWBE18OCmObnD4Z4Ty5GWp3cB1ERiufP0oTvpARd122yWci2 X-Received: by 2002:a17:906:99c3:b0:6df:a289:ed3a with SMTP id s3-20020a17090699c300b006dfa289ed3amr13212980ejn.660.1648234467324; Fri, 25 Mar 2022 11:54:27 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1648234467; cv=none; d=google.com; s=arc-20160816; b=EAhIAKs8WRO5GqqlgpUCI3XF107SAImFlvaqNAZgrj/D7QyplSVzmDCPGEw++ObmpQ snc/VQU3VIECPADA0lJamjcH8blekaQYROm+DjOBBgryWAHmryz5UEIJPn6ZyppmEE+h lOg7273bup04aiO3mB/KlvvJgrXj7NNKvyT14wPGue5y9w+l1AsffspXgOl9UUSc9BG3 vG7l/A3kehGtam2CRCBLzOEv0zVj/X06kvJa5L3+edwJqSIgTRxAoae9QRmiwJWv7nW3 vun5FxHexK92FlAx5W8nbdtwf1hO1xUE9fu9/ykTiqu2k/H+bdS9At+olnUKx5IHSAZV EzvQ== 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:delivered-to; bh=fbSIXWtYjRFa/1nG6mg1xxDPymF6c2TgZSuUHh2qelg=; b=0P4c0yjThm5A0PSm9gb7vKHnfPz4dsg1LfjibxAtFiiG2vnWrKu1DnpuJKvhHZBkHC Psv8ryROHzBCBs9/cc5mw/qyOO2LTdrix14eI16ZukF68vk3QDfNtSyUcDo30Ze+fMWR 7kU/ZEwzgCusHc1G9Qz9cNQu9OK1L986k2EXQAS6bL2cwpu62UarAcbCpZJzTLsUc5dy QplHJuFx8Vhdk+cW9rodHCv8tkELazwuouvJCWTs5Kqg6+Y03gw5w/Y6jaVmY92Y0TCq gC5Zrh21GADdtTKUbidLDgVcxvdNi0krBHFaFc8bizcBa6ydfK67lZok0GFcMSvrGEQk qrcg== ARC-Authentication-Results: i=1; mx.google.com; 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 Return-Path: Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org. [79.124.17.100]) by mx.google.com with ESMTP id hb17-20020a170906b89100b006df84b138c8si3484968ejb.363.2022.03.25.11.54.27; Fri, 25 Mar 2022 11:54:27 -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; 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 Received: from [127.0.1.1] (localhost [127.0.0.1]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id C0B0E68B246; Fri, 25 Mar 2022 20:53:51 +0200 (EET) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from outmail148096.authsmtp.net (outmail148096.authsmtp.net [62.13.148.96]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTPS id B03BA68B23A for ; Fri, 25 Mar 2022 20:53:44 +0200 (EET) Received: from mail-c233.authsmtp.com (mail-c233.authsmtp.com [62.13.128.233]) by punt17.authsmtp.com. (8.15.2/8.15.2) with ESMTP id 22PIrieB052390; Fri, 25 Mar 2022 18:53:44 GMT (envelope-from bavison@riscosopen.org) Received: from rpi2021 (237.63.9.51.dyn.plus.net [51.9.63.237]) (authenticated bits=0) by mail.authsmtp.com (8.15.2/8.15.2) with ESMTPSA id 22PIrf2a046518 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NO); Fri, 25 Mar 2022 18:53:42 GMT (envelope-from bavison@riscosopen.org) Received: by rpi2021 (sSMTP sendmail emulation); Fri, 25 Mar 2022 18:53:41 +0000 From: Ben Avison To: ffmpeg-devel@ffmpeg.org Date: Fri, 25 Mar 2022 18:52:51 +0000 Message-Id: <20220325185257.513933-5-bavison@riscosopen.org> X-Mailer: git-send-email 2.25.1 In-Reply-To: <20220325185257.513933-1-bavison@riscosopen.org> References: <20220317185819.466470-1-bavison@riscosopen.org> <20220325185257.513933-1-bavison@riscosopen.org> MIME-Version: 1.0 X-Server-Quench: e4786446-ac6c-11ec-a0f2-84349711df28 X-AuthReport-Spam: If SPAM / abuse - report it at: http://www.authsmtp.com/abuse X-AuthRoute: OCd1YggXA1ZfRRob ESQCJDVBUg4iPRpU DBlFKhFVNl8UURhQ KkJXbgASJgdBAnRQ QXkJW1ZWQFx5U2Z2 YQ5SIwBcfENQWQZ0 UktOXVBXFgB3AFID BH5nEE9wJgVCfntx bQhgXnZYXgovcUF1 REpQE3BXNmEzdWEe BRNFJgMCch5CehxB Y1d+VSdbY21JDRoR IyQTd2htemwHHWxc XAoKIV8ZBlgAR2x0 ZhYeBzIjAEYIDys6 NABuMlcAHA4fNV96 LF0sQlkRKx4UBUVU Hl1WHDVULlxJTTcl Aw5ANQAA X-Authentic-SMTP: 61633632303230.1021:7600 X-AuthFastPath: 0 (Was 255) X-AuthSMTP-Origin: 51.9.63.237/2525 X-AuthVirus-Status: No virus detected - but ensure you scan with your own anti-virus system. Subject: [FFmpeg-devel] [PATCH 04/10] avcodec/vc1: Introduce fast path for unescaping bitstream buffer 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: Ben Avison Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" X-TUID: QF879Ft12wdj Includes a checkasm test. Signed-off-by: Ben Avison --- libavcodec/vc1dec.c | 20 +++++++------- libavcodec/vc1dsp.c | 2 ++ libavcodec/vc1dsp.h | 3 +++ tests/checkasm/vc1dsp.c | 59 +++++++++++++++++++++++++++++++++++++++++ 4 files changed, 74 insertions(+), 10 deletions(-) diff --git a/libavcodec/vc1dec.c b/libavcodec/vc1dec.c index 1c92b9d401..6a30b5b664 100644 --- a/libavcodec/vc1dec.c +++ b/libavcodec/vc1dec.c @@ -490,7 +490,7 @@ static av_cold int vc1_decode_init(AVCodecContext *avctx) size = next - start - 4; if (size <= 0) continue; - buf2_size = vc1_unescape_buffer(start + 4, size, buf2); + buf2_size = v->vc1dsp.vc1_unescape_buffer(start + 4, size, buf2); init_get_bits(&gb, buf2, buf2_size * 8); switch (AV_RB32(start)) { case VC1_CODE_SEQHDR: @@ -680,7 +680,7 @@ static int vc1_decode_frame(AVCodecContext *avctx, void *data, case VC1_CODE_FRAME: if (avctx->hwaccel) buf_start = start; - buf_size2 = vc1_unescape_buffer(start + 4, size, buf2); + buf_size2 = v->vc1dsp.vc1_unescape_buffer(start + 4, size, buf2); break; case VC1_CODE_FIELD: { int buf_size3; @@ -697,8 +697,8 @@ static int vc1_decode_frame(AVCodecContext *avctx, void *data, ret = AVERROR(ENOMEM); goto err; } - buf_size3 = vc1_unescape_buffer(start + 4, size, - slices[n_slices].buf); + buf_size3 = v->vc1dsp.vc1_unescape_buffer(start + 4, size, + slices[n_slices].buf); init_get_bits(&slices[n_slices].gb, slices[n_slices].buf, buf_size3 << 3); slices[n_slices].mby_start = avctx->coded_height + 31 >> 5; @@ -709,7 +709,7 @@ static int vc1_decode_frame(AVCodecContext *avctx, void *data, break; } case VC1_CODE_ENTRYPOINT: /* it should be before frame data */ - buf_size2 = vc1_unescape_buffer(start + 4, size, buf2); + buf_size2 = v->vc1dsp.vc1_unescape_buffer(start + 4, size, buf2); init_get_bits(&s->gb, buf2, buf_size2 * 8); ff_vc1_decode_entry_point(avctx, v, &s->gb); break; @@ -726,8 +726,8 @@ static int vc1_decode_frame(AVCodecContext *avctx, void *data, ret = AVERROR(ENOMEM); goto err; } - buf_size3 = vc1_unescape_buffer(start + 4, size, - slices[n_slices].buf); + buf_size3 = v->vc1dsp.vc1_unescape_buffer(start + 4, size, + slices[n_slices].buf); init_get_bits(&slices[n_slices].gb, slices[n_slices].buf, buf_size3 << 3); slices[n_slices].mby_start = get_bits(&slices[n_slices].gb, 9); @@ -761,7 +761,7 @@ static int vc1_decode_frame(AVCodecContext *avctx, void *data, ret = AVERROR(ENOMEM); goto err; } - buf_size3 = vc1_unescape_buffer(divider + 4, buf + buf_size - divider - 4, slices[n_slices].buf); + buf_size3 = v->vc1dsp.vc1_unescape_buffer(divider + 4, buf + buf_size - divider - 4, slices[n_slices].buf); init_get_bits(&slices[n_slices].gb, slices[n_slices].buf, buf_size3 << 3); slices[n_slices].mby_start = s->mb_height + 1 >> 1; @@ -770,9 +770,9 @@ static int vc1_decode_frame(AVCodecContext *avctx, void *data, n_slices1 = n_slices - 1; n_slices++; } - buf_size2 = vc1_unescape_buffer(buf, divider - buf, buf2); + buf_size2 = v->vc1dsp.vc1_unescape_buffer(buf, divider - buf, buf2); } else { - buf_size2 = vc1_unescape_buffer(buf, buf_size, buf2); + buf_size2 = v->vc1dsp.vc1_unescape_buffer(buf, buf_size, buf2); } init_get_bits(&s->gb, buf2, buf_size2*8); } else{ diff --git a/libavcodec/vc1dsp.c b/libavcodec/vc1dsp.c index a29b91bf3d..11d493f002 100644 --- a/libavcodec/vc1dsp.c +++ b/libavcodec/vc1dsp.c @@ -34,6 +34,7 @@ #include "rnd_avg.h" #include "vc1dsp.h" #include "startcode.h" +#include "vc1_common.h" /* Apply overlap transform to horizontal edge */ static void vc1_v_overlap_c(uint8_t *src, int stride) @@ -1030,6 +1031,7 @@ av_cold void ff_vc1dsp_init(VC1DSPContext *dsp) #endif /* CONFIG_WMV3IMAGE_DECODER || CONFIG_VC1IMAGE_DECODER */ dsp->startcode_find_candidate = ff_startcode_find_candidate_c; + dsp->vc1_unescape_buffer = vc1_unescape_buffer; if (ARCH_AARCH64) ff_vc1dsp_init_aarch64(dsp); diff --git a/libavcodec/vc1dsp.h b/libavcodec/vc1dsp.h index c6443acb20..8be1198071 100644 --- a/libavcodec/vc1dsp.h +++ b/libavcodec/vc1dsp.h @@ -80,6 +80,9 @@ typedef struct VC1DSPContext { * one or more further zero bytes and a one byte. */ int (*startcode_find_candidate)(const uint8_t *buf, int size); + + /* Copy a buffer, removing startcode emulation escape bytes as we go */ + int (*vc1_unescape_buffer)(const uint8_t *src, int size, uint8_t *dst); } VC1DSPContext; void ff_vc1dsp_init(VC1DSPContext* c); diff --git a/tests/checkasm/vc1dsp.c b/tests/checkasm/vc1dsp.c index 0823ccad31..0ab5892403 100644 --- a/tests/checkasm/vc1dsp.c +++ b/tests/checkasm/vc1dsp.c @@ -286,6 +286,20 @@ static matrix *generate_inverse_quantized_transform_coefficients(size_t width, s } \ } while (0) +#define TEST_UNESCAPE \ + do { \ + for (int count = 100; count > 0; --count) { \ + escaped_offset = rnd() & 7; \ + unescaped_offset = rnd() & 7; \ + escaped_len = (1u << (rnd() % 8) + 3) - (rnd() & 7); \ + RANDOMIZE_BUFFER8(unescaped, UNESCAPE_BUF_SIZE); \ + len0 = call_ref(escaped0 + escaped_offset, escaped_len, unescaped0 + unescaped_offset); \ + len1 = call_new(escaped1 + escaped_offset, escaped_len, unescaped1 + unescaped_offset); \ + if (len0 != len1 || memcmp(unescaped0, unescaped1, len0)) \ + fail(); \ + } \ + } while (0) + void checkasm_check_vc1dsp(void) { /* Inverse transform input coefficients are stored in a 16-bit buffer @@ -309,6 +323,14 @@ void checkasm_check_vc1dsp(void) LOCAL_ALIGNED_4(uint8_t, filter_buf0, [24 * 24]); LOCAL_ALIGNED_4(uint8_t, filter_buf1, [24 * 24]); + /* This appears to be a typical length of buffer in use */ +#define LOG2_UNESCAPE_BUF_SIZE 17 +#define UNESCAPE_BUF_SIZE (1u<