From patchwork Thu Jul 26 11:28:08 2018 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: James Darnley X-Patchwork-Id: 9807 Delivered-To: ffmpegpatchwork@gmail.com Received: by 2002:a02:104:0:0:0:0:0 with SMTP id c4-v6csp336960jad; Thu, 26 Jul 2018 04:34:33 -0700 (PDT) X-Google-Smtp-Source: AAOMgpfDFaV30FUeUh6r/nTVDHFK4XT5Nl8tDiCAHg8O3oJ8JofaPXqJHhA0CuzvyBA2Dl6mIdWW X-Received: by 2002:a1c:2807:: with SMTP id o7-v6mr1235954wmo.11.1532604873648; Thu, 26 Jul 2018 04:34:33 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1532604873; cv=none; d=google.com; s=arc-20160816; b=IfbTqiIb45Qt260doH8dm8vRHONxXu32BO+DlwzzFc9CQNNUxm8z6vqsISvE+843m0 ms9P9b/q26b8bJpDADNT1ypoKEjLMoRzWi1LXTYXfEgr4Sxe+qWriy9TbBMuYnI6Czhi bM1XVcf5O+QT/wvzx7/TApxXY1Ic+ft3tjH/jp3nRaWOfNRcaIY4xx2sUkSYDABD/Exo WY6Rb2NUTlg7NjAzpnrMfVrv1eQ/Y4JMd9cZKBY99Yzj0eYK3L3/wc2PL/F8jqubhbMp 9SrbZM89cNkiTrcYLsKqWhRPbkUW/UpgMDWQ5brhK9nVT65ItUM4A3EdbFdiiXy2/q70 z2rQ== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=sender:errors-to:content-transfer-encoding:mime-version:reply-to :list-subscribe:list-help:list-post:list-archive:list-unsubscribe :list-id:precedence:subject:references:in-reply-to:message-id:date :to:from:dkim-signature:delivered-to:arc-authentication-results; bh=pFwUpLKwmJ2MWLoQx/Q1gi+RrOWA2vgBGYYMswqDn4s=; b=im8aPyRV7x3kFe+ohSnS18XGnq1GsJN8yzvd7b8lJG6z7zMCtvwBDPVd/wX+fUYAmp huG3XXUgJ4RTHOhodTpTG0R5OUntEM0XmDcSSBwZ9aJHkK/ZS7DIwiHS3/wk/T4x0CGP WEHpPkZ2caKHr4ETCjXXt7cfK1o9siAVUDSEt09o1lMDxJi3j5TheonyJw53djZO7tUG hfk/xRb9IgM8/5ZOIpFvznuwU8DMmWR4GUQ2Ns8NkgxHKvxkY/IBQ+56BFybrRQxQFyW 0+elgm0uMikPwHWHSu/69Y0ITRoF5B1lFjsznyEioUx2+40rYOIibluS/VnINeBEI0Lj 3kqg== ARC-Authentication-Results: i=1; mx.google.com; dkim=neutral (body hash did not verify) header.i=@ob-encoder-com.20150623.gappssmtp.com header.s=20150623 header.b=FyloCVYa; 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 o57-v6si941211wrf.258.2018.07.26.04.34.33; Thu, 26 Jul 2018 04:34:33 -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=@ob-encoder-com.20150623.gappssmtp.com header.s=20150623 header.b=FyloCVYa; 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 6F78D68A45B; Thu, 26 Jul 2018 14:34:17 +0300 (EEST) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from mail-lj1-f180.google.com (mail-lj1-f180.google.com [209.85.208.180]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTPS id 96A6268A453 for ; Thu, 26 Jul 2018 14:34:10 +0300 (EEST) Received: by mail-lj1-f180.google.com with SMTP id v9-v6so1201737ljk.4 for ; Thu, 26 Jul 2018 04:34:25 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=ob-encoder-com.20150623.gappssmtp.com; s=20150623; h=sender:from:to:subject:date:message-id:in-reply-to:references; bh=qaO4CX1v13myXvbM4z+nkw3YwXmCMd0OUVNKTpRZxJM=; b=FyloCVYaZj6x4Xzw8Ido87KmZhrratrJWLZoSk5pz1nopqne1hQP80FwDrE0hYF4go Lx09mt6dwkLxrRZWQ/q3DODVnq9cGWxaMypylshkNDeuLrxBooODCCP++zQM48dIGzlx /RYr93+Qgdvra2jRAgYvVuI7DWkH0o/DjRG8oApbmtQd3ZDW8WYShHewDNKNXZkxvbgy bz9CbDv1+N8Y9gGln+t1qWhPOgowo8KdSV6RGLW5aMy6LHS34mcXjsu01bJT50WGAvNm ce8gD5oBRZFA2uAreFsOPp6KFEt1rPW/UBeIL6jZ3A04aYUHUXZdcWvGqhfNOpGncsLs +Y3A== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:sender:from:to:subject:date:message-id :in-reply-to:references; bh=qaO4CX1v13myXvbM4z+nkw3YwXmCMd0OUVNKTpRZxJM=; b=XHW0SZvkTlKLi2OaIBMSfHf5IHLg8a1shXyKtrTyhTQFh9z0YrD1W7mrdEKrNULkJY 2jBQgxCwfxSBck/68os4Vi9KKKqfBcykPJuLgziZAavKcYaMwCcFbpRPCzReT/rgDPBI WvvO5saJvkvifkpB0u+aTdvHo/anFcesv+FkH/LYtcrrx62XmMWXuYwdF2oCcb97fc0r UmZHt6F6wk3JUPXifvOToZqtF3ZwL3qUOOZuli5iuUFMft3D94TE4IGkvE+JPYsM2W2y /BG5WjanXlWCIEMcsyuk1E5Z1sADb0fQO+B25dEEjOQ1jacT04KtWPx+h0tNBex1deuH Gr7g== X-Gm-Message-State: AOUpUlGx0bdwFhvQEGzyGauu42VI4xvrtqCg1F1JLoyRLGFvzT3NHFAp 4u4lfX1pVkAPlA/+LUZhYztEnai6qUo= X-Received: by 2002:a2e:40c6:: with SMTP id r67-v6mr1359846lje.41.1532604536362; Thu, 26 Jul 2018 04:28:56 -0700 (PDT) Received: from Ifrit.systemlords.lan (d51a44418.access.telenet.be. [81.164.68.24]) by smtp.gmail.com with ESMTPSA id r73-v6sm182286ljb.16.2018.07.26.04.28.55 for (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Thu, 26 Jul 2018 04:28:55 -0700 (PDT) From: James Darnley To: ffmpeg-devel@ffmpeg.org Date: Thu, 26 Jul 2018 13:28:08 +0200 Message-Id: <20180726112808.11792-4-jdarnley@obe.tv> X-Mailer: git-send-email 2.18.0 In-Reply-To: <20180726112808.11792-1-jdarnley@obe.tv> References: <20180726112808.11792-1-jdarnley@obe.tv> Subject: [FFmpeg-devel] [PATCH 3/3] diracdec: add 10-bit Deslauriers-Dubuc 9, 7 (9_7) vertical high-pass function 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 MIME-Version: 1.0 Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" Speed of ffmpeg when decoding a 720p yuv422p10 file encoded with the relevant transform. C: 84fps SSE2: 111fps AVX2: 115fps dd97 vertical hi sse2: 2.77x faster (31773 vs. 11457 decicycles) compared with C avx2: 3.83x faster (31773 vs. 8297 decicycles) compared with C --- libavcodec/x86/dirac_dwt_10bit.asm | 39 +++++++++++++++++++++++++++ libavcodec/x86/dirac_dwt_init_10bit.c | 29 ++++++++++++++++++++ 2 files changed, 68 insertions(+) diff --git a/libavcodec/x86/dirac_dwt_10bit.asm b/libavcodec/x86/dirac_dwt_10bit.asm index 0295e6f554..2ed77fe3b0 100644 --- a/libavcodec/x86/dirac_dwt_10bit.asm +++ b/libavcodec/x86/dirac_dwt_10bit.asm @@ -25,6 +25,7 @@ SECTION_RODATA 32 cextern pd_1 pd_2: times 8 dd 2 +pd_8: times 8 dd 8 SECTION .text @@ -246,7 +247,44 @@ RET %endmacro +%macro DD97_VERTICAL_HI 0 + +cglobal dd97_vertical_hi, 6, 6, 8, b0, b1, b2, b3, b4, w + mova m7, [pd_8] + shl wd, 2 + add b0q, wq + add b1q, wq + add b2q, wq + add b3q, wq + add b4q, wq + neg wq + + ALIGN 16 + .loop: + mova m0, [b0q + wq] + mova m1, [b1q + wq] + mova m2, [b2q + wq] + mova m3, [b3q + wq] + mova m4, [b4q + wq] + pslld m5, m1, 3 + pslld m6, m3, 3 + paddd m5, m1 + paddd m6, m3 + psubd m5, m0 + psubd m6, m4 + paddd m5, m7 + paddd m5, m6 + psrad m5, 4 + paddd m2, m5 + mova [b2q + wq], m2 + add wq, mmsize + jl .loop +RET + +%endmacro + INIT_XMM sse2 +DD97_VERTICAL_HI HAAR_HORIZONTAL HAAR_VERTICAL LEGALL53_VERTICAL_HI @@ -257,6 +295,7 @@ HAAR_HORIZONTAL HAAR_VERTICAL INIT_YMM avx2 +DD97_VERTICAL_HI HAAR_HORIZONTAL HAAR_VERTICAL LEGALL53_VERTICAL_HI diff --git a/libavcodec/x86/dirac_dwt_init_10bit.c b/libavcodec/x86/dirac_dwt_init_10bit.c index d1234efac5..a9ac603bc5 100644 --- a/libavcodec/x86/dirac_dwt_init_10bit.c +++ b/libavcodec/x86/dirac_dwt_init_10bit.c @@ -23,6 +23,9 @@ #include "libavutil/x86/cpu.h" #include "libavcodec/dirac_dwt.h" +void ff_dd97_vertical_hi_sse2(int32_t *b0, int32_t *b1, int32_t *b2, int32_t *b3, int32_t *b4, int width); +void ff_dd97_vertical_hi_avx2(int32_t *b0, int32_t *b1, int32_t *b2, int32_t *b3, int32_t *b4, int width); + void ff_legall53_vertical_hi_sse2(int32_t *b0, int32_t *b1, int32_t *b2, int width); void ff_legall53_vertical_lo_sse2(int32_t *b0, int32_t *b1, int32_t *b2, int width); void ff_legall53_vertical_hi_avx2(int32_t *b0, int32_t *b1, int32_t *b2, int width); @@ -36,6 +39,24 @@ void ff_vertical_compose_haar_10bit_sse2(int32_t *b0, int32_t *b1, int width_ali void ff_vertical_compose_haar_10bit_avx(int32_t *b0, int32_t *b1, int width_align); void ff_vertical_compose_haar_10bit_avx2(int32_t *b0, int32_t *b1, int width_align); +static void dd97_vertical_hi_sse2(int32_t *b0, int32_t *b1, int32_t *b2, + int32_t *b3, int32_t *b4, int width) +{ + int i = width & ~3; + ff_dd97_vertical_hi_sse2(b0, b1, b2, b3, b4, i); + for(; ivertical_compose_h0 = (void*)dd97_vertical_hi_sse2; + d->vertical_compose_l0 = (void*)ff_legall53_vertical_lo_sse2; + break; case DWT_DIRAC_LEGALL5_3: d->vertical_compose_h0 = (void*)ff_legall53_vertical_hi_sse2; d->vertical_compose_l0 = (void*)ff_legall53_vertical_lo_sse2; @@ -71,6 +96,10 @@ av_cold void ff_spatial_idwt_init_10bit_x86(DWTContext *d, enum dwt_type type) if (EXTERNAL_AVX2(cpu_flags)) { switch (type) { + case DWT_DIRAC_DD9_7: + d->vertical_compose_h0 = (void*)dd97_vertical_hi_avx2; + d->vertical_compose_l0 = (void*)ff_legall53_vertical_lo_avx2; + break; case DWT_DIRAC_LEGALL5_3: d->vertical_compose_h0 = (void*)ff_legall53_vertical_hi_avx2; d->vertical_compose_l0 = (void*)ff_legall53_vertical_lo_avx2;