From patchwork Sun Aug 18 08:29:02 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Michael Niedermayer X-Patchwork-Id: 14572 Return-Path: X-Original-To: patchwork@ffaux-bg.ffmpeg.org Delivered-To: patchwork@ffaux-bg.ffmpeg.org Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org [79.124.17.100]) by ffaux.localdomain (Postfix) with ESMTP id AB9B7443BAC for ; Sun, 18 Aug 2019 11:31:02 +0300 (EEST) Received: from [127.0.1.1] (localhost [127.0.0.1]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id 7F211689AB4; Sun, 18 Aug 2019 11:31:02 +0300 (EEST) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from vie01a-dmta-pe08-2.mx.upcmail.net (vie01a-dmta-pe08-2.mx.upcmail.net [84.116.36.21]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTPS id B5833680109 for ; Sun, 18 Aug 2019 11:30:55 +0300 (EEST) Received: from [172.31.216.235] (helo=vie01a-pemc-psmtp-pe12.mail.upcmail.net) by vie01a-dmta-pe08.mx.upcmail.net with esmtp (Exim 4.92) (envelope-from ) id 1hzGaV-0002xZ-4Q for ffmpeg-devel@ffmpeg.org; Sun, 18 Aug 2019 10:30:55 +0200 Received: from localhost ([213.47.41.20]) by vie01a-pemc-psmtp-pe12.mail.upcmail.net with ESMTP id zGZWhuUZhwlyszGZWh3G2Y; Sun, 18 Aug 2019 10:29:55 +0200 X-Env-Mailfrom: michael@niedermayer.cc X-Env-Rcptto: ffmpeg-devel@ffmpeg.org X-SourceIP: 213.47.41.20 X-CNFS-Analysis: v=2.3 cv=E5OzWpVl c=1 sm=1 tr=0 a=I1eytVlZLDX1BM2VTtTtSw==:117 a=I1eytVlZLDX1BM2VTtTtSw==:17 a=MKtGQD3n3ToA:10 a=1oJP67jkp3AA:10 a=GEAsPZ9sns4A:10 a=ZZnuYtJkoWoA:10 a=nZOtpAppAAAA:20 a=ibYtI8ihYU65C1DyFysA:9 a=1fhp2MxaeJtTNGEnv6mo:22 a=Z5ABNNGmrOfJ6cZ5bIyy:22 a=jd6J4Gguk5HxikPWLKER:22 From: Michael Niedermayer To: FFmpeg development discussions and patches Date: Sun, 18 Aug 2019 10:29:02 +0200 Message-Id: <20190818082902.2425-1-michael@niedermayer.cc> X-Mailer: git-send-email 2.22.1 MIME-Version: 1.0 X-CMAE-Envelope: MS4wfAxbMLTcnu7imR/eijpqLKs6kcA0LtGCx5H0H8D80A/6L2IBT0guERtTcw0flwu3HtzGLSWvagWO8OFDILGZUI+SIUYJs8OUF6YpvO2au1ftPMlg+2X/ XM/9IvYx6xCCJ+SFZR3eJe674gvLs4ZR9NKiX1BIhoexc9qH/VPAglJY Subject: [FFmpeg-devel] [PATCH] avcodec/pngdec: Optimize has_trns code X-BeenThere: ffmpeg-devel@ffmpeg.org X-Mailman-Version: 2.1.20 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 Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" add inner loop specialisations for 2 bpp and 4 bpp These are all cases for which i found testsamples. 30M cycles -> 5M cycles Testcase: fate-rgbapng-4816 Testcase: 16097/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_APNG_fuzzer-5664690889293824 Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg Signed-off-by: Michael Niedermayer --- libavcodec/pngdec.c | 37 +++++++++++++++++++++++++++++-------- 1 file changed, 29 insertions(+), 8 deletions(-) diff --git a/libavcodec/pngdec.c b/libavcodec/pngdec.c index cad5796545..2d6c1b218e 100644 --- a/libavcodec/pngdec.c +++ b/libavcodec/pngdec.c @@ -24,6 +24,7 @@ #include "libavutil/avassert.h" #include "libavutil/bprint.h" #include "libavutil/imgutils.h" +#include "libavutil/intreadwrite.h" #include "libavutil/stereo3d.h" #include "libavutil/mastering_display_metadata.h" @@ -1367,15 +1368,35 @@ exit_loop: for (y = 0; y < s->height; ++y) { uint8_t *row = &s->image_buf[s->image_linesize * y]; - /* since we're updating in-place, we have to go from right to left */ - for (x = s->width; x > 0; --x) { - uint8_t *pixel = &row[s->bpp * (x - 1)]; - memmove(pixel, &row[raw_bpp * (x - 1)], raw_bpp); + if (s->bpp == 2 && byte_depth == 1) { + uint8_t *pixel = &row[2 * s->width - 1]; + uint8_t *rowp = &row[1 * s->width - 1]; + int tcolor = s->transparent_color_be[0]; + for (x = s->width; x > 0; --x) { + *pixel-- = *rowp == tcolor ? 0 : 0xff; + *pixel-- = *rowp--; + } + } else if (s->bpp == 4 && byte_depth == 1) { + uint8_t *pixel = &row[4 * s->width - 1]; + uint8_t *rowp = &row[3 * s->width - 1]; + int tcolor = AV_RL24(s->transparent_color_be); + for (x = s->width; x > 0; --x) { + *pixel-- = AV_RL24(rowp-2) == tcolor ? 0 : 0xff; + *pixel-- = *rowp--; + *pixel-- = *rowp--; + *pixel-- = *rowp--; + } + } else { + /* since we're updating in-place, we have to go from right to left */ + for (x = s->width; x > 0; --x) { + uint8_t *pixel = &row[s->bpp * (x - 1)]; + memmove(pixel, &row[raw_bpp * (x - 1)], raw_bpp); - if (!memcmp(pixel, s->transparent_color_be, raw_bpp)) { - memset(&pixel[raw_bpp], 0, byte_depth); - } else { - memset(&pixel[raw_bpp], 0xff, byte_depth); + if (!memcmp(pixel, s->transparent_color_be, raw_bpp)) { + memset(&pixel[raw_bpp], 0, byte_depth); + } else { + memset(&pixel[raw_bpp], 0xff, byte_depth); + } } } }