From patchwork Mon Aug 22 15:26:23 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Hubert Mazur X-Patchwork-Id: 37397 Delivered-To: ffmpegpatchwork2@gmail.com Received: by 2002:a05:6a20:139a:b0:8f:1db5:eae2 with SMTP id w26csp1361361pzh; Mon, 22 Aug 2022 08:26:58 -0700 (PDT) X-Google-Smtp-Source: AA6agR4iw0p486T0/nlYYVnlsRbponLoDfovFG5KvJOE24/gA8wSFI7/wt8eoYREKcxnVesgjAI9 X-Received: by 2002:a17:907:970b:b0:73d:5a29:959 with SMTP id jg11-20020a170907970b00b0073d5a290959mr9193700ejc.183.1661182018484; Mon, 22 Aug 2022 08:26:58 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1661182018; cv=none; d=google.com; s=arc-20160816; b=KvpYjvCYpUdRqwHrLG+GSyaDd9p8mqZ0NaW4gSP6nMKgfIpN/yslO8SwjZIz8FJmcN /k4tHH3q3VY6DsSoqhHDeCD0yphujdnUG9aAFmnKTYq+g2CzgU3T0ZsbNQOZZXbGMWHr QkjmdrZGmJrvyAqs5JtVRMEvVQqJiEZwJUt47rKidOY+poAdvdEm/IFeTr8oYrTglbMf S/Oe47NRvYoLkKeoMSG3XnqmVDhxjGVml0TIGj7jLAWwXnBVXMEprlbOG7hFe+WnB9lr Z7BI0CsGunHUAA0VInavc8nf9FjLLsg3E8h7lOLZ96QulkQt+DK09utnxfT+3k/TyRZR H6+A== 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=9c1i/Y42hKJCCL0HyBnJezNJMaI8sbh1KNEvjSosbX4=; b=AbpXCe5DHX0pkIeUN5hTJZVUezmHq2ZJr/fQbSZePnU7NLgW1D4rXWCuJj3Ny0v+/D Z9f+9WA5MbivsrFb1pghajqjGWUOEXWP1Rr7AiEVrno8UxI0mmiGE/V2wG+nKJgkjc7h 3A3FZqmHIAmGIh5j/SyjvmyCM+xdm1UMnxHUl+Z1/oOcbjW8pLC3FAaeGPAvzGXTeQuZ KXjO74DhLmIeQeJjbmq/saizpEbYjQlXk/QqX8ufIyfTIIX+sqx18enyQG1CBZItW48Z S1eJy3xoL81ut6N2oeLy1617gEsQq88L9VoOOhd1yWtpGFq76uMKxJEXgDiz+3YEePuR 38oQ== ARC-Authentication-Results: i=1; mx.google.com; dkim=neutral (body hash did not verify) header.i=@semihalf.com header.s=google header.b=ICYibZCH; 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=QUARANTINE sp=QUARANTINE dis=NONE) header.from=semihalf.com Return-Path: Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org. [79.124.17.100]) by mx.google.com with ESMTP id cw16-20020a170906479000b0073d8a8fedbfsi1237932ejc.386.2022.08.22.08.26.58; Mon, 22 Aug 2022 08:26:58 -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=@semihalf.com header.s=google header.b=ICYibZCH; 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=QUARANTINE sp=QUARANTINE dis=NONE) header.from=semihalf.com Received: from [127.0.1.1] (localhost [127.0.0.1]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id DF27368BA1E; Mon, 22 Aug 2022 18:26:47 +0300 (EEST) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from mail-lj1-f178.google.com (mail-lj1-f178.google.com [209.85.208.178]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTPS id 6D1C468BA14 for ; Mon, 22 Aug 2022 18:26:41 +0300 (EEST) Received: by mail-lj1-f178.google.com with SMTP id l19so7804288ljg.8 for ; Mon, 22 Aug 2022 08:26:41 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=semihalf.com; s=google; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc; bh=3nwFqcPHpNtl9QykgLjikZYEeqOHyj+nJSfRlC42xTM=; b=ICYibZCHNEu+VpL571B9wb+4hwaVNNnYy3ZJjNIEvqHoYPqo0PC4WCT0KmvbGToagJ 1xyCgPL0ns9syzHoODqw57lRRPvBsIAfJORO0ktPkjyHwaPW3jIcATMhC5dXVkB/Psbr QYjwm3tCbslEuLZIc8zTbHFctw2qqkUU08+cwUYmRT+SP0AFHoR9UewvAMI4LprqhtKf z/cmh+t3vL4NUMpUEiiIybRKQmUpyuZRjXYt+ZWpcloq9RF+SXoRXZkU52l3rc/StV5r j9si0pAXu5fukSOyqasyv3x+IfwCL14dYNUqlpb6kFZhp7m/l9q92hv4ySYiEKLtUMqx F4xQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc; bh=3nwFqcPHpNtl9QykgLjikZYEeqOHyj+nJSfRlC42xTM=; b=iVWbZCMDXdkHL44GkP1HGpLvv0/59gSTW9Ui3j7mZyDr6/eVnl34+D4z3Gmcwh7+J2 hzo+bdCiUNpnvdqnOzoTFhjOFl7nr2u6z25GRBTc73Bf5f7Z5C/heaRi1zgeuLEfunRv QfXbbboG4A77Z44+7YimBjTChoNd9feXD3g8020wQ4Nl6YjDSXGVmtT3x8sLrwxYnJlf IlwMv+R9Y0LxqWkfNP5hXZF8HjYWLTR3AX/JZknvvQQAmZYEw9AXGEj7Ef7SFWUEHL9x rVIM93IrdtB8ZaVI/rSIVFE5DaP/Mq3BmNkA1ac5Y0imqm6eCiPviaWQ1QFcgDYSv43t rt0w== X-Gm-Message-State: ACgBeo1XzkZd8J9PK1OqiRy3u4fB5hy1yKhDH8Ckg1p1DMtQNmLqw7n6 NPF2MR2TGGIHCNRO2azZ/4xjcMYSmJpuQA== X-Received: by 2002:a05:651c:507:b0:25e:73af:1661 with SMTP id o7-20020a05651c050700b0025e73af1661mr6363111ljp.195.1661182000386; Mon, 22 Aug 2022 08:26:40 -0700 (PDT) Received: from hum-HP-ProBook-440-G7.office.semihalf.net ([83.142.187.84]) by smtp.gmail.com with ESMTPSA id z13-20020a056512370d00b00492f21c161esm2921lfr.123.2022.08.22.08.26.39 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 22 Aug 2022 08:26:40 -0700 (PDT) From: Hubert Mazur To: ffmpeg-devel@ffmpeg.org Date: Mon, 22 Aug 2022 17:26:23 +0200 Message-Id: <20220822152627.1992008-2-hum@semihalf.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20220822152627.1992008-1-hum@semihalf.com> References: <20220822152627.1992008-1-hum@semihalf.com> MIME-Version: 1.0 Subject: [FFmpeg-devel] [PATCH 1/5] lavc/aarch64: Add neon implementation for vsad16 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: gjb@semihalf.com, upstream@semihalf.com, jswinney@amazon.com, Hubert Mazur , martin@martin.st, mw@semihalf.com, spop@amazon.com Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" X-TUID: Af0I2hxaWgjB Provide optimized implementation of vsad16 function for arm64. Performance comparison tests are shown below. - vsad_0_c: 285.0 - vsad_0_neon: 42.5 Benchmarks and tests are run with checkasm tool on AWS Graviton 3. Signed-off-by: Hubert Mazur --- libavcodec/aarch64/me_cmp_init_aarch64.c | 5 ++ libavcodec/aarch64/me_cmp_neon.S | 75 ++++++++++++++++++++++++ 2 files changed, 80 insertions(+) diff --git a/libavcodec/aarch64/me_cmp_init_aarch64.c b/libavcodec/aarch64/me_cmp_init_aarch64.c index fb7c3f5059..ddc5d05611 100644 --- a/libavcodec/aarch64/me_cmp_init_aarch64.c +++ b/libavcodec/aarch64/me_cmp_init_aarch64.c @@ -41,6 +41,9 @@ int sse8_neon(MpegEncContext *v, const uint8_t *pix1, const uint8_t *pix2, int sse4_neon(MpegEncContext *v, const uint8_t *pix1, const uint8_t *pix2, ptrdiff_t stride, int h); +int vsad16_neon(MpegEncContext *c, const uint8_t *s1, const uint8_t *s2, + ptrdiff_t stride, int h); + av_cold void ff_me_cmp_init_aarch64(MECmpContext *c, AVCodecContext *avctx) { int cpu_flags = av_get_cpu_flags(); @@ -57,5 +60,7 @@ av_cold void ff_me_cmp_init_aarch64(MECmpContext *c, AVCodecContext *avctx) c->sse[0] = sse16_neon; c->sse[1] = sse8_neon; c->sse[2] = sse4_neon; + + c->vsad[0] = vsad16_neon; } } diff --git a/libavcodec/aarch64/me_cmp_neon.S b/libavcodec/aarch64/me_cmp_neon.S index 4198985c6c..d4c0099854 100644 --- a/libavcodec/aarch64/me_cmp_neon.S +++ b/libavcodec/aarch64/me_cmp_neon.S @@ -584,3 +584,78 @@ function sse4_neon, export=1 ret endfunc + +function vsad16_neon, export=1 + // x0 unused + // x1 uint8_t *pix1 + // x2 uint8_t *pix2 + // x3 ptrdiff_t stride + // w4 int h + + sub w4, w4, #1 // we need to make h-1 iterations + movi v16.8h, #0 + + cmp w4, #3 // check if we can make 3 iterations at once + add x5, x1, x3 // pix1 + stride + add x6, x2, x3 // pix2 + stride + b.le 2f + +1: + // abs(pix1[0] - pix2[0] - pix1[0 + stride] + pix2[0 + stride]) + // abs(x) = (x < 0 ? (-x) : (x)) + ld1 {v0.16b}, [x1], x3 // Load pix1[0], first iteration + ld1 {v1.16b}, [x2], x3 // Load pix2[0], first iteration + ld1 {v2.16b}, [x5], x3 // Load pix1[0 + stride], first iteration + usubl v31.8h, v0.8b, v1.8b // Signed difference pix1[0] - pix2[0], first iteration + ld1 {v3.16b}, [x6], x3 // Load pix2[0 + stride], first iteration + usubl2 v30.8h, v0.16b, v1.16b // Signed difference pix1[0] - pix2[0], first iteration + usubl v29.8h, v2.8b, v3.8b // Signed difference pix1[0 + stride] - pix2[0 + stride], first iteration + ld1 {v4.16b}, [x1], x3 // Load pix1[0], second iteration + usubl2 v28.8h, v2.16b, v3.16b // Signed difference pix1[0 + stride] - pix2[0 + stride], first iteration + ld1 {v5.16b}, [x2], x3 // Load pix2[0], second iteration + saba v16.8h, v31.8h, v29.8h // Signed absolute difference and accumulate the result. first iteration + ld1 {v6.16b}, [x5], x3 // Load pix1[0 + stride], second iteration + saba v16.8h, v30.8h, v28.8h // Signed absolute difference and accumulate the result. first iteration + usubl v27.8h, v4.8b, v5.8b // Signed difference pix1[0] - pix2[0], second iteration + ld1 {v7.16b}, [x6], x3 // Load pix2[0 + stride], second iteration + usubl2 v26.8h, v4.16b, v5.16b // Signed difference pix1[0] - pix2[0], second iteration + usubl v25.8h, v6.8b, v7.8b // Signed difference pix1[0 + stride] - pix2[0 + stride], second iteration + ld1 {v17.16b}, [x1], x3 // Load pix1[0], third iteration + usubl2 v24.8h, v6.16b, v7.16b // Signed difference pix1[0 + stride] - pix2[0 + stride], second iteration + ld1 {v18.16b}, [x2], x3 // Load pix2[0], second iteration + saba v16.8h, v27.8h, v25.8h // Signed absolute difference and accumulate the result. second iteration + ld1 {v19.16b}, [x5], x3 // Load pix1[0 + stride], third iteration + saba v16.8h, v26.8h, v24.8h // Signed absolute difference and accumulate the result. second iteration + usubl v23.8h, v17.8b, v18.8b // Signed difference pix1[0] - pix2[0], third iteration + ld1 {v20.16b}, [x6], x3 // Load pix2[0 + stride], third iteration + usubl2 v22.8h, v17.16b, v18.16b // Signed difference pix1[0] - pix2[0], third iteration + usubl v21.8h, v19.8b, v20.8b // Signed difference pix1[0 + stride] - pix2[0 + stride], third iteration + sub w4, w4, #3 // h -= 3 + saba v16.8h, v23.8h, v21.8h // Signed absolute difference and accumulate the result. third iteration + usubl2 v31.8h, v19.16b, v20.16b // Signed difference pix1[0 + stride] - pix2[0 + stride], third iteration + cmp w4, #3 + saba v16.8h, v22.8h, v31.8h // Signed absolute difference and accumulate the result. third iteration + + b.ge 1b + cbz w4, 3f +2: + + ld1 {v0.16b}, [x1], x3 + ld1 {v1.16b}, [x2], x3 + ld1 {v2.16b}, [x5], x3 + usubl v30.8h, v0.8b, v1.8b + ld1 {v3.16b}, [x6], x3 + usubl2 v29.8h, v0.16b, v1.16b + usubl v28.8h, v2.8b, v3.8b + usubl2 v27.8h, v2.16b, v3.16b + saba v16.8h, v30.8h, v28.8h + subs w4, w4, #1 + saba v16.8h, v29.8h, v27.8h + + b.ne 2b +3: + uaddlv s17, v16.8h + fmov w0, s17 + + ret +endfunc From patchwork Mon Aug 22 15:26:24 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Hubert Mazur X-Patchwork-Id: 37398 Delivered-To: ffmpegpatchwork2@gmail.com Received: by 2002:a05:6a20:139a:b0:8f:1db5:eae2 with SMTP id w26csp1361440pzh; Mon, 22 Aug 2022 08:27:07 -0700 (PDT) X-Google-Smtp-Source: AA6agR6Y+uq61hHclPsQsVU2sYJ8wIEcpK70Eqay/ObDlagOUVTINfsK/44H/8dVfhVTBrJd0ILc X-Received: by 2002:a05:6402:14cd:b0:446:52ed:f131 with SMTP id f13-20020a05640214cd00b0044652edf131mr11871795edx.341.1661182026845; Mon, 22 Aug 2022 08:27:06 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1661182026; cv=none; d=google.com; s=arc-20160816; b=auDg/Jdx4z0VTjnLQFF3odeOVTNM4GbOcYfnjOIUBrIrOltiFp3ZDvM+VTqCjDCBhE W07z79GyD/0MDXYXL4zZudLTIpIL3F8gN6Lm9VKbhP1APgpuj8BCfRy/lLCoQ63reVLV VWOso1r2+FXnjU1W7P9lgdVqH3EK9tCbbs2bhRRBUOyLybIwyH+K0NsHwgF1hQknmxl7 R7s3RBgEILXYIlE0xQ+xaZWaRXb7vOBpv0gqtK2Z8fAhxzYO2i2xrDG2OY3Pb08F6vV/ L4IYP4LfC2OqUDLtF6rHweHBVaXmzB+hWQel9rE7ppYz2aMZR2YV3/8E18lUIA6lhCv6 fWhA== 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=M/+ExitrdNGyeY00ipUOV/Hux4cVTCUlWowFnvmvwcU=; b=Vs8/vaFCvEwaZ2kzHkTJ3lbtv3xLmQdCyIUM/urPPEDwFRroEpcbmAyD7dojuo9LJp h9wde9+O7vXW4nVt/WUv9FJs5RbR0dq39W3DuIQp5iUeF9SQDLl3T7od3nPPybmWa/XL vAnR46H0eAqoEuyVMa2XYv00VbE77QpMC6XB8STA+d5F81DNlziM/rmUxSx9234q3NuC kbBIE4aiaUJdf/Z08aYjZICvt01W7L2uKmayWRmSUh2myziv3r8UNi1796+MwL+nGms9 yS8657Rj5NDLQ3YwC8VD6QkcQTE/YmsxFSM/fBY98UwINg0yDHaeNiPyfwF7tehlvYHH wszw== ARC-Authentication-Results: i=1; mx.google.com; dkim=neutral (body hash did not verify) header.i=@semihalf.com header.s=google header.b="DK+0Y/gZ"; 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=QUARANTINE sp=QUARANTINE dis=NONE) header.from=semihalf.com Return-Path: Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org. [79.124.17.100]) by mx.google.com with ESMTP id cx1-20020a05640222a100b00445e63a702fsi8607074edb.215.2022.08.22.08.27.06; Mon, 22 Aug 2022 08:27:06 -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=@semihalf.com header.s=google header.b="DK+0Y/gZ"; 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=QUARANTINE sp=QUARANTINE dis=NONE) header.from=semihalf.com Received: from [127.0.1.1] (localhost [127.0.0.1]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id E5D1668BA2D; Mon, 22 Aug 2022 18:26:48 +0300 (EEST) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from mail-lf1-f47.google.com (mail-lf1-f47.google.com [209.85.167.47]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTPS id CC46568BA1C for ; Mon, 22 Aug 2022 18:26:42 +0300 (EEST) Received: by mail-lf1-f47.google.com with SMTP id be9so7832329lfb.12 for ; Mon, 22 Aug 2022 08:26:42 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=semihalf.com; s=google; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc; bh=JOowWCXKUNlSQSlUTEaEnzMt07wxf7+1QgxYoBrQbjk=; b=DK+0Y/gZ4jBg2t/1j4Qttv3WJzLVTvfxnGCpKw9G4mEeHzkn/A3TmfaLl7JmhND8X0 wUcKEXOfK/2gtOwwF5bhn/Rp24W7WCDPdNNzde+AyTH/0R4GvnN6ifmYwicxvJZBamCH dDW7PHAOJ8dwLuOc+Dbpn3nML5HPzzi3RdHglTBxQLOgHNWCGH59dt54P4qgDsZ0HTNs QH8INjUNM1qGk2TpjRMQ7mYfXyC+ZA+cIFvHrnZOL9c4n+udmxnWEfd/dzmripBca1ay mhaflXltDB8ro+ALrbQ3fbFGpa3iEOQAzHQVCM2PctzvXeleClx/rI3wwYGlLElTyQRM tOLw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc; bh=JOowWCXKUNlSQSlUTEaEnzMt07wxf7+1QgxYoBrQbjk=; b=MUvoJyg+3p2386Yq4I9fpKMJk8utSUMhUo0eqdQ0Zr8TE4MQUrKYigYQPvmejo7Jnw y36eGwsbaLNsklVyM9mEkjNBhb+p45OxqoHbipYqdrYT/4VcNgnygzn0Kps697q6oB5e 3MeahnYCMkU/1/+4fIVpUuGrpi360hc+H8Y18emlAZ0h55wEJBJt7JoTt+YSqNzhgQp9 7d3RBxbWKMRagkwE6xU/G73N1KpvhHXnOl4pg7f8HDHm+y+vKeMKyL8JdDbzn3DsIeaZ msa9T8VuGLBGfGEa/3+zxUgJBLKnwywhWPRDTDFTqWIK9J7QMACb/qpjF2RIwr6gZy3G 3WXA== X-Gm-Message-State: ACgBeo0Cc5ptRBHVsC8hCCgj+nXNIfCR5r3Bki5ekiGLLEEj7XGXf78c kNo2/bLmmz2sj9ZkGRE7+s4iJ0TsHmNBAA== X-Received: by 2002:a05:6512:33d6:b0:492:c423:1672 with SMTP id d22-20020a05651233d600b00492c4231672mr6840297lfg.432.1661182001855; Mon, 22 Aug 2022 08:26:41 -0700 (PDT) Received: from hum-HP-ProBook-440-G7.office.semihalf.net ([83.142.187.84]) by smtp.gmail.com with ESMTPSA id z13-20020a056512370d00b00492f21c161esm2921lfr.123.2022.08.22.08.26.41 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 22 Aug 2022 08:26:41 -0700 (PDT) From: Hubert Mazur To: ffmpeg-devel@ffmpeg.org Date: Mon, 22 Aug 2022 17:26:24 +0200 Message-Id: <20220822152627.1992008-3-hum@semihalf.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20220822152627.1992008-1-hum@semihalf.com> References: <20220822152627.1992008-1-hum@semihalf.com> MIME-Version: 1.0 Subject: [FFmpeg-devel] [PATCH 2/5] lavc/aarch64: Add neon implementation of vsse16 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: gjb@semihalf.com, upstream@semihalf.com, jswinney@amazon.com, Hubert Mazur , martin@martin.st, mw@semihalf.com, spop@amazon.com Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" X-TUID: RLrAzIeb/+WC Provide optimized implementation of vsse16 for arm64. Performance comparison tests are shown below. - vsse_0_c: 254.4 - vsse_0_neon: 64.7 Benchmarks and tests are run with checkasm tool on AWS Graviton 3. Signed-off-by: Hubert Mazur --- libavcodec/aarch64/me_cmp_init_aarch64.c | 4 + libavcodec/aarch64/me_cmp_neon.S | 97 ++++++++++++++++++++++++ 2 files changed, 101 insertions(+) diff --git a/libavcodec/aarch64/me_cmp_init_aarch64.c b/libavcodec/aarch64/me_cmp_init_aarch64.c index ddc5d05611..7b81e48d16 100644 --- a/libavcodec/aarch64/me_cmp_init_aarch64.c +++ b/libavcodec/aarch64/me_cmp_init_aarch64.c @@ -43,6 +43,8 @@ int sse4_neon(MpegEncContext *v, const uint8_t *pix1, const uint8_t *pix2, int vsad16_neon(MpegEncContext *c, const uint8_t *s1, const uint8_t *s2, ptrdiff_t stride, int h); +int vsse16_neon(MpegEncContext *c, const uint8_t *s1, const uint8_t *s2, + ptrdiff_t stride, int h); av_cold void ff_me_cmp_init_aarch64(MECmpContext *c, AVCodecContext *avctx) { @@ -62,5 +64,7 @@ av_cold void ff_me_cmp_init_aarch64(MECmpContext *c, AVCodecContext *avctx) c->sse[2] = sse4_neon; c->vsad[0] = vsad16_neon; + + c->vsse[0] = vsse16_neon; } } diff --git a/libavcodec/aarch64/me_cmp_neon.S b/libavcodec/aarch64/me_cmp_neon.S index d4c0099854..279bae7cb5 100644 --- a/libavcodec/aarch64/me_cmp_neon.S +++ b/libavcodec/aarch64/me_cmp_neon.S @@ -659,3 +659,100 @@ function vsad16_neon, export=1 ret endfunc + +function vsse16_neon, export=1 + // x0 unused + // x1 uint8_t *pix1 + // x2 uint8_t *pix2 + // x3 ptrdiff_t stride + // w4 int h + + movi v30.4s, #0 + movi v29.4s, #0 + + add x5, x1, x3 // pix1 + stride + add x6, x2, x3 // pix2 + stride + sub w4, w4, #1 // we need to make h-1 iterations + cmp w4, #3 // check if we can make 4 iterations at once + b.le 2f + +// make 4 iterations at once +1: + // x = abs(pix1[0] - pix2[0] - pix1[0 + stride] + pix2[0 + stride]) = + // res = (x) * (x) + ld1 {v0.16b}, [x1], x3 // Load pix1[0], first iteration + ld1 {v1.16b}, [x2], x3 // Load pix2[0], first iteration + ld1 {v2.16b}, [x5], x3 // Load pix1[0 + stride], first iteration + usubl v28.8h, v0.8b, v1.8b // Signed difference of pix1[0] - pix2[0], first iteration + ld1 {v3.16b}, [x6], x3 // Load pix2[0 + stride], first iteration + usubl2 v27.8h, v0.16b, v1.16b // Signed difference of pix1[0] - pix2[0], first iteration + usubl v26.8h, v3.8b, v2.8b // Signed difference of pix1[0 + stride] - pix2[0 + stride], first iteration + usubl2 v25.8h, v3.16b, v2.16b // Signed difference of pix1[0 + stride] - pix2[0 + stride], first iteration + ld1 {v4.16b}, [x1], x3 // Load pix1[0], second iteration + sqadd v28.8h, v28.8h, v26.8h // Add first iteration + ld1 {v6.16b}, [x5], x3 // Load pix1[0 + stride], second iteration + sqadd v27.8h, v27.8h, v25.8h // Add first iteration + ld1 {v5.16b}, [x2], x3 // Load pix2[0], second iteration + smlal v30.4s, v28.4h, v28.4h // Multiply-accumulate first iteration + ld1 {v7.16b}, [x6], x3 // Load pix2[0 + stride], second iteration + usubl v26.8h, v4.8b, v5.8b // Signed difference of pix1[0] - pix2[0], second iteration + smlal2 v29.4s, v28.8h, v28.8h // Multiply-accumulate first iteration + usubl2 v25.8h, v4.16b, v5.16b // Signed difference of pix1[0] - pix2[0], second iteration + usubl v24.8h, v7.8b, v6.8b // Signed difference of pix1[0 + stride] - pix2[0 + stride], first iteration + smlal v30.4s, v27.4h, v27.4h // Multiply-accumulate first iteration + usubl2 v23.8h, v7.16b, v6.16b // Signed difference of pix1[0 + stride] - pix2[0 + stride], first iteration + sqadd v24.8h, v26.8h, v24.8h // Add second iteration + smlal2 v29.4s, v27.8h, v27.8h // Multiply-accumulate first iteration + sqadd v23.8h, v25.8h, v23.8h // Add second iteration + ld1 {v18.16b}, [x1], x3 // Load pix1[0], third iteration + smlal v30.4s, v24.4h, v24.4h // Multiply-accumulate second iteration + ld1 {v31.16b}, [x2], x3 // Load pix2[0], third iteration + ld1 {v17.16b}, [x5], x3 // Load pix1[0 + stride], third iteration + smlal2 v29.4s, v24.8h, v24.8h // Multiply-accumulate second iteration + ld1 {v16.16b}, [x6], x3 // Load pix2[0 + stride], third iteration + usubl v22.8h, v18.8b, v31.8b // Signed difference of pix1[0] - pix2[0], third iteration + smlal v30.4s, v23.4h, v23.4h // Multiply-accumulate second iteration + usubl2 v21.8h, v18.16b, v31.16b // Signed difference of pix1[0] - pix2[0], third iteration + usubl v20.8h, v16.8b, v17.8b // Signed difference of pix1[0 + stride] - pix2[0 + stride], first iteration + smlal2 v29.4s, v23.8h, v23.8h // Multiply-accumulate second iteration + sqadd v20.8h, v22.8h, v20.8h // Add third iteration + usubl2 v19.8h, v16.16b, v17.16b // Signed difference of pix1[0 + stride] - pix2[0 + stride], first iteration + smlal v30.4s, v20.4h, v20.4h // Multiply-accumulate third iteration + sqadd v19.8h, v21.8h, v19.8h // Add third iteration + smlal2 v29.4s, v20.8h, v20.8h // Multiply-accumulate third iteration + sub w4, w4, #3 + smlal v30.4s, v19.4h, v19.4h // Multiply-accumulate third iteration + cmp w4, #3 + smlal2 v29.4s, v19.8h, v19.8h // Multiply-accumulate third iteration + + b.ge 1b + + cbz w4, 3f + +// iterate by once +2: + ld1 {v0.16b}, [x1], x3 + ld1 {v1.16b}, [x2], x3 + ld1 {v2.16b}, [x5], x3 + usubl v28.8h, v0.8b, v1.8b + ld1 {v3.16b}, [x6], x3 + usubl2 v27.8h, v0.16b, v1.16b + usubl v26.8h, v3.8b, v2.8b + usubl2 v25.8h, v3.16b, v2.16b + sqadd v28.8h, v28.8h, v26.8h + sqadd v27.8h, v27.8h, v25.8h + smlal v30.4s, v28.4h, v28.4h + smlal2 v29.4s, v28.8h, v28.8h + subs w4, w4, #1 + smlal v30.4s, v27.4h, v27.4h + smlal2 v29.4s, v27.8h, v27.8h + + b.ne 2b + +3: + add v30.4s, v30.4s, v29.4s + saddlv d17, v30.4s + fmov w0, s17 + + ret +endfunc From patchwork Mon Aug 22 15:26:25 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Hubert Mazur X-Patchwork-Id: 37399 Delivered-To: ffmpegpatchwork2@gmail.com Received: by 2002:a05:6a20:139a:b0:8f:1db5:eae2 with SMTP id w26csp1361527pzh; Mon, 22 Aug 2022 08:27:16 -0700 (PDT) X-Google-Smtp-Source: AA6agR7dG6vtlX/T1aUKUsD1IPgeVl/Tuql+jM5KtXydX4KfXysRo5R4wucMtns8gl/hhFdaI7yT X-Received: by 2002:a05:6402:4311:b0:447:9fb:b710 with SMTP id m17-20020a056402431100b0044709fbb710mr15037edc.272.1661182035979; Mon, 22 Aug 2022 08:27:15 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1661182035; cv=none; d=google.com; s=arc-20160816; b=0i27imgud4wI1KGMdIPSIuxT9dQkltAbbQGaD2dfibD+emUsmqx6RYspxtahKLYran bZnUBgca/gCW72xuQa+KQ5g79BePg5fZJbCrarJWSi/Y5I+nuh/xvz0p+LKc+KqRK+R9 ky6A3L+21h5D4/j9ml0DX/bCHSTpWrhvZwoRfpCcqHnZNA1PW19agZH06QxijGZ25GoF 6qdlkkIjNoR5euby8XqtO08XrnLoxlYaKtkXBn7pVbyMBn1ReWIr6oeUB87NS/KbKWgV FrGWQ6nNYGxJsDB5ydX0zutirypFoG4jW58oXFXJ37tFnlPwgbohNf6qRp7EBCK18W/E 9pHQ== 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=tE1czw2bfjfHsq/FD+f1DwBQGhLHanwe9COMO3nLDek=; b=URKttyZfKv1Af8n9qssylkuGj6qpFP7njWYjoapNRbRpxnZIAcExknE695djBpBmZv aoHW8bpnkALs0+l4Ph2O127MLh+GbbAOQHuwA4aD7jTNVNii9oFbC7S+tcuLthPufHBP 9Lh9/0O67/vWrJzIttNwC5X+XopIfN0mc/L5pb1DGq10b/1yjWyzgQYTb5n6RtSmR+sq DcOOLvyoPTSq9O7NyphtO0OqbE8zSY3jjINiY+AR/P7rSHmacI71gOQauH4p+5r4OKJD dy9isyBg5RxftNVg23v2p6nr6rGcGw1N7Aafg9iUznISaxnospgD346/EoT3Y3uF09ot YT0Q== ARC-Authentication-Results: i=1; mx.google.com; dkim=neutral (body hash did not verify) header.i=@semihalf.com header.s=google header.b=CSi1nUtN; 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=QUARANTINE sp=QUARANTINE dis=NONE) header.from=semihalf.com Return-Path: Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org. [79.124.17.100]) by mx.google.com with ESMTP id ec40-20020a0564020d6800b0043d0c49375bsi8596589edb.389.2022.08.22.08.27.15; Mon, 22 Aug 2022 08:27:15 -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=@semihalf.com header.s=google header.b=CSi1nUtN; 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=QUARANTINE sp=QUARANTINE dis=NONE) header.from=semihalf.com Received: from [127.0.1.1] (localhost [127.0.0.1]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id 0FDA068BA0C; Mon, 22 Aug 2022 18:26:51 +0300 (EEST) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from mail-lf1-f49.google.com (mail-lf1-f49.google.com [209.85.167.49]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTPS id 5F95A68BA2F for ; Mon, 22 Aug 2022 18:26:44 +0300 (EEST) Received: by mail-lf1-f49.google.com with SMTP id be9so7832405lfb.12 for ; Mon, 22 Aug 2022 08:26:44 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=semihalf.com; s=google; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc; bh=Ab2fXCqaExS7bHmukV+siuKUQ1DWa+I6kb+MaKV06bc=; b=CSi1nUtNDgA7RWmOEUC/7C9yiKaFaxy2QPwGpFHOzj4is1/mX/mni5BO/FefcryXr8 sqlfqRY9t6QEQcMjp1uORokkL+KeSsRqIIzc3gwn1upXoGw685gQ7DU8a+khg8HNDHT1 I0QcJjtTX1xaM5XnOct1U9BkWM/22z6jekzaUZbLodE8Whfukq6nx2BuqGe2/EOLG7/a MaWx8KUcAEWfpLzdZRcG7MJPdX9BCZHLmLhotl4n5IaEu3KUlN3xvi7cQl7Y4FCPF/Yi p7huRJOmWIUqxcDlk108DmBM5Rx/bZ5IrquQKv7302mqAZQTRDnwPXbQLAZ7eQWR8q5M CjOw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc; bh=Ab2fXCqaExS7bHmukV+siuKUQ1DWa+I6kb+MaKV06bc=; b=Wkc3FyEE1wNOsIHDXshRwNMKuOI9sGdYvp7LIk22l2K2ShAVyxHflfrcYsd6TP8lQn qoFvverfG+IAAhOUeIVZ4mf0ja2TgILbPfr8ym0fNcakNOrO7s57Q3QO3x2SpGprOw1d rP2EnNA67kgIS57GUBKoDZAE5hO5nisFAtbm+ZheGICDqmgyL3RaCz6OusowLSN4JAvp ERj6V9vWAEp1QWBcnkK99x+/xlkQcAD+ELONd4CSI4cRo24q3iD/GdZvpRRcOWI6yQXn L2/cH7f7EmYC8Hl0Kk961ynmEWX8hjADF2RMMyJNpqOBRHy/FHWN1mpJvAR/tf6IS+3V aASQ== X-Gm-Message-State: ACgBeo15tzt7eJkURDjdgoJ2+CdGuA3FXlbaL1VhK8PgDB0a6ycU4hqz cYjh26w8ke9qm7CaYUliKTzVW0IgBUkAtw== X-Received: by 2002:a05:6512:2294:b0:492:e136:5211 with SMTP id f20-20020a056512229400b00492e1365211mr2601676lfu.273.1661182003381; Mon, 22 Aug 2022 08:26:43 -0700 (PDT) Received: from hum-HP-ProBook-440-G7.office.semihalf.net ([83.142.187.84]) by smtp.gmail.com with ESMTPSA id z13-20020a056512370d00b00492f21c161esm2921lfr.123.2022.08.22.08.26.42 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 22 Aug 2022 08:26:43 -0700 (PDT) From: Hubert Mazur To: ffmpeg-devel@ffmpeg.org Date: Mon, 22 Aug 2022 17:26:25 +0200 Message-Id: <20220822152627.1992008-4-hum@semihalf.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20220822152627.1992008-1-hum@semihalf.com> References: <20220822152627.1992008-1-hum@semihalf.com> MIME-Version: 1.0 Subject: [FFmpeg-devel] [PATCH 3/5] lavc/aarch64: Add neon implementation for vsad_intra16 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: gjb@semihalf.com, upstream@semihalf.com, jswinney@amazon.com, Hubert Mazur , martin@martin.st, mw@semihalf.com, spop@amazon.com Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" X-TUID: udgP2wqu+Qe+ Provide optimized implementation for vsad_intra16 function for arm64. Performance comparison tests are shown below. - vsad_4_c: 177.2 - vsad_4_neon: 24.5 Benchmarks and tests are run with checkasm tool on AWS Gravtion 3. Signed-off-by: Hubert Mazur --- libavcodec/aarch64/me_cmp_init_aarch64.c | 3 ++ libavcodec/aarch64/me_cmp_neon.S | 58 ++++++++++++++++++++++++ 2 files changed, 61 insertions(+) diff --git a/libavcodec/aarch64/me_cmp_init_aarch64.c b/libavcodec/aarch64/me_cmp_init_aarch64.c index 7b81e48d16..af83f7ed1e 100644 --- a/libavcodec/aarch64/me_cmp_init_aarch64.c +++ b/libavcodec/aarch64/me_cmp_init_aarch64.c @@ -43,6 +43,8 @@ int sse4_neon(MpegEncContext *v, const uint8_t *pix1, const uint8_t *pix2, int vsad16_neon(MpegEncContext *c, const uint8_t *s1, const uint8_t *s2, ptrdiff_t stride, int h); +int vsad_intra16_neon(MpegEncContext *c, const uint8_t *s, const uint8_t *dummy, + ptrdiff_t stride, int h) ; int vsse16_neon(MpegEncContext *c, const uint8_t *s1, const uint8_t *s2, ptrdiff_t stride, int h); @@ -64,6 +66,7 @@ av_cold void ff_me_cmp_init_aarch64(MECmpContext *c, AVCodecContext *avctx) c->sse[2] = sse4_neon; c->vsad[0] = vsad16_neon; + c->vsad[4] = vsad_intra16_neon; c->vsse[0] = vsse16_neon; } diff --git a/libavcodec/aarch64/me_cmp_neon.S b/libavcodec/aarch64/me_cmp_neon.S index 279bae7cb5..126e267fdc 100644 --- a/libavcodec/aarch64/me_cmp_neon.S +++ b/libavcodec/aarch64/me_cmp_neon.S @@ -756,3 +756,61 @@ function vsse16_neon, export=1 ret endfunc + +function vsad_intra16_neon, export=1 + // x0 unused + // x1 uint8_t *pix1 + // x2 uint8_t *dummy + // x3 ptrdiff_t stride + // w4 int h + + sub w4, w4, #1 // we need to make h-1 iterations + add x5, x1, x3 // pix1 + stride + cmp w4, #4 + movi v16.8h, #0 + + b.lt 2f + +// make 4 iterations at once +1: + // v = abs( pix1[0] - pix1[0 + stride] ) + // score = sum(v) + // abs(x) = ( (x > 0) ? (x) : (-x) ) + + ld1 {v0.16b}, [x1], x3 + ld1 {v1.16b}, [x5], x3 + ld1 {v2.16b}, [x1], x3 + uabal v16.8h, v0.8b, v1.8b + ld1 {v3.16b}, [x5], x3 + uabal2 v16.8h, v0.16b, v1.16b + ld1 {v4.16b}, [x1], x3 + uabal v16.8h, v2.8b, v3.8b + ld1 {v5.16b}, [x5], x3 + uabal2 v16.8h, v2.16b, v3.16b + ld1 {v6.16b}, [x1], x3 + uabal v16.8h, v4.8b, v5.8b + ld1 {v7.16b}, [x5], x3 + uabal2 v16.8h, v4.16b, v5.16b + sub w4, w4, #4 + uabal v16.8h, v6.8b, v7.8b + cmp w4, #4 + uabal2 v16.8h, v6.16b, v7.16b + + b.ge 1b + cbz w4, 3f + +// iterate by one +2: + ld1 {v0.16b}, [x1], x3 + ld1 {v1.16b}, [x5], x3 + subs w4, w4, #1 + uabal v16.8h, v0.8b, v1.8b + uabal2 v16.8h, v0.16b, v1.16b + cbnz w4, 2b + +3: + uaddlv s17, v16.8h + fmov w0, s17 + + ret +endfunc From patchwork Mon Aug 22 15:26:26 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Hubert Mazur X-Patchwork-Id: 37400 Delivered-To: ffmpegpatchwork2@gmail.com Received: by 2002:a05:6a20:139a:b0:8f:1db5:eae2 with SMTP id w26csp1361672pzh; Mon, 22 Aug 2022 08:27:25 -0700 (PDT) X-Google-Smtp-Source: AA6agR6x0KngcacQBn++HUZPdlYkX/dn66OtY1Pdpxn+GhCTq3LDF1eWjNpsg+6ug9SIGq8fcLLq X-Received: by 2002:a17:907:69b0:b0:73d:6cda:93ac with SMTP id ra48-20020a17090769b000b0073d6cda93acmr6088297ejc.226.1661182045224; Mon, 22 Aug 2022 08:27:25 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1661182045; cv=none; d=google.com; s=arc-20160816; b=fw0EM6cZWKz40Vc4hqnNGcQivnYkcQo8J8Z/1Mgx+i4TjkgOuYrAuHKA1/Z6STHs+G DqHLIoqiyTAyKyPW8LHaZk3nUi78DLpfsEkM6oYIeDuPMilfLaayrZG5MQX1Disa2fj6 NnyZ6j+bmqN8AgnP5i9jwzwIJryBOI57XvHx1MiCZnoMiWsPNIUjFXgEXBxhrRUnGDFq 9le9CJAaZlmPYNRe37d96HBRQliGNrkBYIejVnw98PQsW9UyH/lk7yXgmbvKl1eRG+uX hiuZAexJilDWYooWCuW3cG1Q03vXD+LViQR4sh6/2ZQ2/5VhXJ97Us/ljcnDSRHcQ763 QdhA== 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=wviSH1A8befJlr/ZpGPHPxRL5lM4EeGntI0IDpzr8dI=; b=Rlhsj+KQaM0ZPmCA6FbLGU9q0kIHVDG/6/6a4EpfMhbBxAKuVE84FCtOH28oSlL3zX /X9tJu3EQRhCKY4JZgKdfG+g9YBuc+YE7HqIT9AiORNn647zPrcElKyjY7QSIagZ/dTn 7JqRxXmko3KqYIfHvhGwpS1U7R1Mr983DWmplwPf7kIbseENMwN7L36x5XrFXr5wA5Yc naDoxuM6cclCxtI+M50Dkl+3eFCRnz7J9L0IYF+XgXddUbTe9OczGPjuzJoMXwSlrkf3 1dPd7FgDmLXJryMTGNWupJJqRlP3Z5t1GUDLTi6giu54hKYFpII4j1raUFdd2sahuuk4 o3aQ== ARC-Authentication-Results: i=1; mx.google.com; dkim=neutral (body hash did not verify) header.i=@semihalf.com header.s=google header.b=infi28gR; 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=QUARANTINE sp=QUARANTINE dis=NONE) header.from=semihalf.com Return-Path: Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org. [79.124.17.100]) by mx.google.com with ESMTP id gb6-20020a170907960600b00734b96e0cbdsi1678949ejc.167.2022.08.22.08.27.24; Mon, 22 Aug 2022 08:27:25 -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=@semihalf.com header.s=google header.b=infi28gR; 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=QUARANTINE sp=QUARANTINE dis=NONE) header.from=semihalf.com Received: from [127.0.1.1] (localhost [127.0.0.1]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id 0B11E68BA4D; Mon, 22 Aug 2022 18:26:53 +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 A993168B75A for ; Mon, 22 Aug 2022 18:26:45 +0300 (EEST) Received: by mail-lj1-f180.google.com with SMTP id x25so10935586ljm.5 for ; Mon, 22 Aug 2022 08:26:45 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=semihalf.com; s=google; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc; bh=giiiwYOsrCuGx3N0Xls5tGAuY1W1s0VS+2iObUdWG7A=; b=infi28gRJqfOTDxJXnlg+3URX7B4LKM14RAnRFN2rcBLtyGphECXLzkIw6XJKcueBq CAy9/1s8w60E+PpTwvBpQcLKQ1a5Y+OoARgCqyOvK3pqQEXP7YXY9ryCMZaevMB7FXma KezCeK12l1LAqllaOLn6aaVYULA8cGYj/YUAfbGWShIvbRMdVFKVKj22G/x3kkS3BzqU EkXuyodyr3gQBJo0mQhEaudabIszsYW1I0J9XvE3A3sLn3ZqISEuj0dEpG6XA6QXcGtI RklXrx4ONaEXgSJVs7PwYb0GucrW5c0ERJaKqtaaEC5UJbWknT+ZX/B4HDObmNWuHoTW aAIQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc; bh=giiiwYOsrCuGx3N0Xls5tGAuY1W1s0VS+2iObUdWG7A=; b=IeW9iBcBvXWNlShrXY0LjLSvsSpF1tZ/aNrYVWENzFYqunq8EyhFLCq2FlBg2qDGJC 9QnHbJfOul0zmKLx0cqb4GFGbOFgF8JbsPFqPPUpjdlY95ToczIheQ1QObElG0Gip1BL d1PuGtLeyYL/oPajsyY8+xuiEBeLudOlDg+YlW8yZng0clfr1nJfBMj9u7QdZKrJJRnn uoeX0jSH3KWgBbu0G4DJCJbW22t9K+LrIIDGml0uEIYvqkgnoKiHz/NXpebdyCcwbFqN RfjaG4zRWvLFfBehfKtztofXvuF2eV/039gSPYdvrqrf90S6FGryBPcKCTHqn37TrQYd KwVQ== X-Gm-Message-State: ACgBeo1kprY6FdZYgBFwasutkJ75cuxoVcbA+6iHDkd2a4Ow/L9CW56H gP7+vw5Ab5747qxAJ7wzAg7bWPH8ozsfaw== X-Received: by 2002:a2e:9dc5:0:b0:25e:6fa0:1243 with SMTP id x5-20020a2e9dc5000000b0025e6fa01243mr6302679ljj.513.1661182004766; Mon, 22 Aug 2022 08:26:44 -0700 (PDT) Received: from hum-HP-ProBook-440-G7.office.semihalf.net ([83.142.187.84]) by smtp.gmail.com with ESMTPSA id z13-20020a056512370d00b00492f21c161esm2921lfr.123.2022.08.22.08.26.43 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 22 Aug 2022 08:26:44 -0700 (PDT) From: Hubert Mazur To: ffmpeg-devel@ffmpeg.org Date: Mon, 22 Aug 2022 17:26:26 +0200 Message-Id: <20220822152627.1992008-5-hum@semihalf.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20220822152627.1992008-1-hum@semihalf.com> References: <20220822152627.1992008-1-hum@semihalf.com> MIME-Version: 1.0 Subject: [FFmpeg-devel] [PATCH 4/5] lavc/aarch64: Add neon implementation for vsse_intra16 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: gjb@semihalf.com, upstream@semihalf.com, jswinney@amazon.com, Hubert Mazur , martin@martin.st, mw@semihalf.com, spop@amazon.com Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" X-TUID: y3emJsF6WFk4 Provide optimized implementation for vsse_intra16 for arm64. Performance tests are shown below. - vsse_4_c: 153.7 - vsse_4_neon: 34.2 Benchmarks and tests are run with checkasm tool on AWS Graviton 3. Signed-off-by: Hubert Mazur --- libavcodec/aarch64/me_cmp_init_aarch64.c | 3 + libavcodec/aarch64/me_cmp_neon.S | 75 ++++++++++++++++++++++++ 2 files changed, 78 insertions(+) diff --git a/libavcodec/aarch64/me_cmp_init_aarch64.c b/libavcodec/aarch64/me_cmp_init_aarch64.c index af83f7ed1e..8c295d5457 100644 --- a/libavcodec/aarch64/me_cmp_init_aarch64.c +++ b/libavcodec/aarch64/me_cmp_init_aarch64.c @@ -47,6 +47,8 @@ int vsad_intra16_neon(MpegEncContext *c, const uint8_t *s, const uint8_t *dummy, ptrdiff_t stride, int h) ; int vsse16_neon(MpegEncContext *c, const uint8_t *s1, const uint8_t *s2, ptrdiff_t stride, int h); +int vsse_intra16_neon(MpegEncContext *c, const uint8_t *s, const uint8_t *dummy, + ptrdiff_t stride, int h); av_cold void ff_me_cmp_init_aarch64(MECmpContext *c, AVCodecContext *avctx) { @@ -69,5 +71,6 @@ av_cold void ff_me_cmp_init_aarch64(MECmpContext *c, AVCodecContext *avctx) c->vsad[4] = vsad_intra16_neon; c->vsse[0] = vsse16_neon; + c->vsse[4] = vsse_intra16_neon; } } diff --git a/libavcodec/aarch64/me_cmp_neon.S b/libavcodec/aarch64/me_cmp_neon.S index 126e267fdc..46d4dade5d 100644 --- a/libavcodec/aarch64/me_cmp_neon.S +++ b/libavcodec/aarch64/me_cmp_neon.S @@ -814,3 +814,78 @@ function vsad_intra16_neon, export=1 ret endfunc + +function vsse_intra16_neon, export=1 + // x0 unused + // x1 uint8_t *pix1 + // x2 uint8_t *dummy + // x3 ptrdiff_t stride + // w4 int h + + add x5, x1, x3 // pix1 + stride + movi v16.4s, #0 + movi v17.4s, #0 + + sub w4, w4, #1 // we need to make h-1 iterations + cmp w4, #4 + b.lt 2f + +// make 4 iterations at once +1: + // v = abs( pix1[0] - pix1[0 + stride] ) + // score = sum( v * v ) + // abs(x) = ( (x > 0) ? (x) : (-x) ) + + ld1 {v0.16b}, [x1], x3 + ld1 {v1.16b}, [x5], x3 + ld1 {v2.16b}, [x1], x3 + uabd v30.16b, v0.16b, v1.16b + ld1 {v3.16b}, [x5], x3 + ld1 {v4.16b}, [x1], x3 + umull v29.8h, v30.8b, v30.8b + umull2 v28.8h, v30.16b, v30.16b + uabd v27.16b, v2.16b, v3.16b + uadalp v16.4s, v29.8h + ld1 {v5.16b}, [x5], x3 + umull v26.8h, v27.8b, v27.8b + uadalp v17.4s, v28.8h + ld1 {v6.16b}, [x1], x3 + umull2 v27.8h, v27.16b, v27.16b + uabd v25.16b, v4.16b, v5.16b + uadalp v16.4s, v26.8h + umull v24.8h, v25.8b, v25.8b + ld1 {v7.16b}, [x5], x3 + uadalp v17.4s, v27.8h + umull2 v25.8h, v25.16b, v25.16b + uabd v23.16b, v6.16b, v7.16b + uadalp v16.4s, v24.8h + umull v22.8h, v23.8b, v23.8b + uadalp v17.4s, v25.8h + umull2 v23.8h, v23.16b, v23.16b + sub w4, w4, #4 + uadalp v16.4s, v22.8h + cmp w4, #4 + uadalp v17.4s, v23.8h + + b.ge 1b + cbz w4, 3f + +// iterate by one +2: + ld1 {v0.16b}, [x1], x3 + ld1 {v1.16b}, [x5], x3 + subs w4, w4, #1 + uabd v30.16b, v0.16b, v1.16b + umull v29.8h, v30.8b, v30.8b + umull2 v30.8h, v30.16b, v30.16b + uadalp v16.4s, v29.8h + uadalp v17.4s, v30.8h + cbnz w4, 2b + +3: + add v16.4s, v16.4s, v17.4S + uaddlv d17, v16.4s + fmov w0, s17 + + ret +endfunc From patchwork Mon Aug 22 15:26:27 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Hubert Mazur X-Patchwork-Id: 37401 Delivered-To: ffmpegpatchwork2@gmail.com Received: by 2002:a05:6a20:139a:b0:8f:1db5:eae2 with SMTP id w26csp1361780pzh; Mon, 22 Aug 2022 08:27:33 -0700 (PDT) X-Google-Smtp-Source: AA6agR7gAB7MRxVSJmLq48pi8j44vCNr7BV5Z2/bZA/MpcmhXCBoT2dbLazHge7MakNwADEYM9ii X-Received: by 2002:a05:6402:4282:b0:43e:612c:fcf7 with SMTP id g2-20020a056402428200b0043e612cfcf7mr16440419edc.242.1661182053792; Mon, 22 Aug 2022 08:27:33 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1661182053; cv=none; d=google.com; s=arc-20160816; b=m0AUs02YaXX5+yg/RFSChhbHUoFwfsd2ERsdfBBpV5GJnQo0HbwGMzV5TLWxhdQ1Kv LfV+nC1eHsrU3lvkas7YLnV7VrMEnWs+Dl4YEIYyFnQhOC94xp1ZDMKxTxir1LPWCnU/ dlVNd6LYxQf2DFvlIaWR1+3IqoZVJqT4Za3o0fUGtjuqqGGcBt0MjUu4+o0XM9chAxrl LTd+LkELzWr8cvqUFTu8MuYSQnTVurugnchzhyGjE8qQL0JpmNuGitJPjCGMFr8xRsEv ueBfWn/+bi1TBdV79uQgLvk3/CaZdKl7BYooANpPLpeCP4DnCiBjg+U18PmSwk1PR91t 0zGw== 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=CBhxVCd2y9L7APUtzQunzFIGE6NleZXhjjp/lqiDiGM=; b=dTPSiK+ASTGyBXAyEgeZkMHcCXwxWg0VTdYgBx1D8U+TyYywMTCi0DCoS+TuPnS6cY YmdCJnZR06C6L4kzo+FBgeUZAfAWpkmu6QsOqPhB5CD7h6ozk/aC1a+rQi5P6TeSCq8J G2xiSob2cnD2y3T5nEKLNLVPN6DeAquwYT8r/GNxKBGbnDqwrrSFlKZ9nbRCKOWdMg8q vPTHot966rLetKeawo8P1ipjYJSMfZ14r7soPtLU+3JlRoZhdG9GT07vFMFrop+jU+yR 8N5HFgKAoQgdkDIhECkHENioz8qmXCrnrBjbfAyWjdzwDJXNiGZyH8YXcoAvbxOKWUrh DwgQ== ARC-Authentication-Results: i=1; mx.google.com; dkim=neutral (body hash did not verify) header.i=@semihalf.com header.s=google header.b=QyJbjsix; 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=QUARANTINE sp=QUARANTINE dis=NONE) header.from=semihalf.com Return-Path: Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org. [79.124.17.100]) by mx.google.com with ESMTP id js18-20020a17090797d200b007308396d8eesi11029214ejc.780.2022.08.22.08.27.33; Mon, 22 Aug 2022 08:27:33 -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=@semihalf.com header.s=google header.b=QyJbjsix; 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=QUARANTINE sp=QUARANTINE dis=NONE) header.from=semihalf.com Received: from [127.0.1.1] (localhost [127.0.0.1]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id 2890268BA52; Mon, 22 Aug 2022 18:26:54 +0300 (EEST) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from mail-lj1-f173.google.com (mail-lj1-f173.google.com [209.85.208.173]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTPS id 17D7468BA23 for ; Mon, 22 Aug 2022 18:26:47 +0300 (EEST) Received: by mail-lj1-f173.google.com with SMTP id z20so10963788ljq.3 for ; Mon, 22 Aug 2022 08:26:47 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=semihalf.com; s=google; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc; bh=IvYM9kGJkMv+rHITVPbKft9RJcuV/aOsCiCePaQ0Utg=; b=QyJbjsixshHDDk5KhbqqOLitIQA5W5Fle0iwX1405NZHmTbZBPAJBotWPiDZCNAbu4 bunhzOUK6bfyzARySZOKVKLz0W+waSgbsv7r/mz+4ZGAm2LjpZnoY+JQV3N//HSZAbgj LnwCp8/Mmb9jRLsGexFM52ekIUfgUtNuhffhixz9I5uZVIqxzBdabN4R7Kz05B0E8IWF Uv33qC0WfXXLSDd6Xg8oW2ngnALY739GpUIfZ4a15U+KrKTSOnkk7GwLB5xOxlAKkJH6 b1MyeXvBWScFi/uCFcjYgr5IPIX/U7a7IGpoSfRTeiVLErT8OG9TDQY0HL27oR6GcGrV a71g== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc; bh=IvYM9kGJkMv+rHITVPbKft9RJcuV/aOsCiCePaQ0Utg=; b=EdaIinAu1gUL/8k/jaBuDBi6l6wQkMB7qDzurer8G01JsaXgaKM6xONbcSoblzR6eP 4PA2+CAK2mtcSqxnmPHYjf/p/9MUdaifqLT7Alxit2oTGJE3j2bgNlETHNk12kHLeut2 isijTXvsxzqI+7cNN2zB/K5zdI6GyknSiRE+xh4dwze8S9tWOCtcSxuxKg2fkverf1hM rorRdkpHM/BdeIl+/PkM7Jvb+qMeYac74KWObCU5wdOv3VEkZDjovHWGg0H8MuyJSYex XYadeL1mQCqw/ROYNR8xm8UfDQyZ+OMgbRnVvFxIUxCFCABclqn8M/x87K2I/V3fw09a P8vw== X-Gm-Message-State: ACgBeo0DPrFve73rAvliXERH15tlqoO6/c0NQGh0Cfg6EjwplhfMqviT uKfGYFiaIjBkl1ifeY8Hk0z3L/lO2KWjOA== X-Received: by 2002:a05:651c:a0c:b0:25e:6d4e:1bdb with SMTP id k12-20020a05651c0a0c00b0025e6d4e1bdbmr6372771ljq.111.1661182006138; Mon, 22 Aug 2022 08:26:46 -0700 (PDT) Received: from hum-HP-ProBook-440-G7.office.semihalf.net ([83.142.187.84]) by smtp.gmail.com with ESMTPSA id z13-20020a056512370d00b00492f21c161esm2921lfr.123.2022.08.22.08.26.45 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 22 Aug 2022 08:26:45 -0700 (PDT) From: Hubert Mazur To: ffmpeg-devel@ffmpeg.org Date: Mon, 22 Aug 2022 17:26:27 +0200 Message-Id: <20220822152627.1992008-6-hum@semihalf.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20220822152627.1992008-1-hum@semihalf.com> References: <20220822152627.1992008-1-hum@semihalf.com> MIME-Version: 1.0 Subject: [FFmpeg-devel] [PATCH 5/5] lavc/aarch64: Provide neon implementation of nsse16 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: gjb@semihalf.com, upstream@semihalf.com, jswinney@amazon.com, Hubert Mazur , martin@martin.st, mw@semihalf.com, spop@amazon.com Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" X-TUID: GfW3uqxddwFE Add vectorized implementation of nsse16 function. Performance comparison tests are shown below. - nsse_0_c: 707.0 - nsse_0_neon: 120.0 Benchmarks and tests run with checkasm tool on AWS Graviton 3. Signed-off-by: Hubert Mazur --- libavcodec/aarch64/me_cmp_init_aarch64.c | 15 +++ libavcodec/aarch64/me_cmp_neon.S | 126 +++++++++++++++++++++++ 2 files changed, 141 insertions(+) diff --git a/libavcodec/aarch64/me_cmp_init_aarch64.c b/libavcodec/aarch64/me_cmp_init_aarch64.c index 8c295d5457..146ef04345 100644 --- a/libavcodec/aarch64/me_cmp_init_aarch64.c +++ b/libavcodec/aarch64/me_cmp_init_aarch64.c @@ -49,6 +49,10 @@ int vsse16_neon(MpegEncContext *c, const uint8_t *s1, const uint8_t *s2, ptrdiff_t stride, int h); int vsse_intra16_neon(MpegEncContext *c, const uint8_t *s, const uint8_t *dummy, ptrdiff_t stride, int h); +int nsse16_neon(int multiplier, const uint8_t *s, const uint8_t *s2, + ptrdiff_t stride, int h); +int nsse16_neon_wrapper(MpegEncContext *c, const uint8_t *s1, const uint8_t *s2, + ptrdiff_t stride, int h); av_cold void ff_me_cmp_init_aarch64(MECmpContext *c, AVCodecContext *avctx) { @@ -72,5 +76,16 @@ av_cold void ff_me_cmp_init_aarch64(MECmpContext *c, AVCodecContext *avctx) c->vsse[0] = vsse16_neon; c->vsse[4] = vsse_intra16_neon; + + c->nsse[0] = nsse16_neon_wrapper; } } + +int nsse16_neon_wrapper(MpegEncContext *c, const uint8_t *s1, const uint8_t *s2, + ptrdiff_t stride, int h) +{ + if (c) + return nsse16_neon(c->avctx->nsse_weight, s1, s2, stride, h); + else + return nsse16_neon(8, s1, s2, stride, h); +} \ No newline at end of file diff --git a/libavcodec/aarch64/me_cmp_neon.S b/libavcodec/aarch64/me_cmp_neon.S index 46d4dade5d..9fe96e111c 100644 --- a/libavcodec/aarch64/me_cmp_neon.S +++ b/libavcodec/aarch64/me_cmp_neon.S @@ -889,3 +889,129 @@ function vsse_intra16_neon, export=1 ret endfunc + +function nsse16_neon, export=1 + // x0 multiplier + // x1 uint8_t *pix1 + // x2 uint8_t *pix2 + // x3 ptrdiff_t stride + // w4 int h + + str x0, [sp, #-0x40]! + stp x1, x2, [sp, #0x10] + stp x3, x4, [sp, #0x20] + str lr, [sp, #0x30] + bl sse16_neon + ldr lr, [sp, #0x30] + mov w9, w0 // here we store score1 + ldr x5, [sp] + ldp x1, x2, [sp, #0x10] + ldp x3, x4, [sp, #0x20] + add sp, sp, #0x40 + + movi v16.8h, #0 + movi v17.8h, #0 + movi v18.8h, #0 + movi v19.8h, #0 + + mov x10, x1 // x1 + mov x14, x2 // x2 + add x11, x1, x3 // x1 + stride + add x15, x2, x3 // x2 + stride + add x12, x1, #1 // x1 + 1 + add x16, x2, #1 // x2 + 1 + add x13, x11, #1 // x1 + stride + 1 + add x17, x15, #1 // x2 + stride + 1 + + subs w4, w4, #1 // we need to make h-1 iterations + cmp w4, #2 + b.lt 2f + +// make 2 iterations at once +1: + ld1 {v0.16b}, [x10], x3 + ld1 {v1.16b}, [x11], x3 + ld1 {v2.16b}, [x12], x3 + usubl v31.8h, v0.8b, v1.8b + ld1 {v3.16b}, [x13], x3 + usubl2 v30.8h, v0.16b, v1.16b + usubl v29.8h, v2.8b, v3.8b + usubl2 v28.8h, v2.16b, v3.16b + ld1 {v4.16b}, [x14], x3 + saba v16.8h, v31.8h, v29.8h + ld1 {v5.16b}, [x15], x3 + ld1 {v6.16b}, [x16], x3 + saba v17.8h, v30.8h, v28.8h + ld1 {v7.16b}, [x17], x3 + usubl v27.8h, v4.8b, v5.8b + usubl2 v26.8h, v4.16b, v5.16b + usubl v25.8h, v6.8b, v7.8b + ld1 {v31.16b}, [x10], x3 + saba v18.8h, v27.8h, v25.8h + usubl2 v24.8h, v6.16b, v7.16b + ld1 {v30.16b}, [x11], x3 + ld1 {v29.16b}, [x12], x3 + saba v19.8h, v26.8h, v24.8h + usubl v23.8h, v31.8b, v30.8b + ld1 {v28.16b}, [x13], x3 + usubl2 v22.8h, v31.16b, v30.16b + usubl v21.8h, v29.8b, v28.8b + ld1 {v27.16b}, [x14], x3 + usubl2 v20.8h, v29.16b, v28.16b + saba v16.8h, v23.8h, v21.8h + ld1 {v26.16b}, [x15], x3 + ld1 {v25.16b}, [x16], x3 + saba v17.8h, v22.8h, v20.8h + ld1 {v24.16b}, [x17], x3 + usubl v31.8h, v27.8b, v26.8b + usubl v29.8h, v25.8b, v24.8b + usubl2 v30.8h, v27.16b, v26.16b + saba v18.8h, v31.8h, v29.8h + usubl2 v28.8h, v25.16b, v24.16b + sub w4, w4, #2 + cmp w4, #2 + saba v19.8h, v30.8h, v28.8h + + b.ge 1b + cbz w4, 3f + +// iterate by one +2: + ld1 {v0.16b}, [x10], x3 + ld1 {v1.16b}, [x11], x3 + ld1 {v2.16b}, [x12], x3 + usubl v31.8h, v0.8b, v1.8b + ld1 {v3.16b}, [x13], x3 + usubl2 v30.8h, v0.16b, v1.16b + usubl v29.8h, v2.8b, v3.8b + usubl2 v28.8h, v2.16b, v3.16b + saba v16.8h, v31.8h, v29.8h + ld1 {v4.16b}, [x14], x3 + ld1 {v5.16b}, [x15], x3 + saba v17.8h, v30.8h, v28.8h + ld1 {v6.16b}, [x16], x3 + usubl v27.8h, v4.8b, v5.8b + ld1 {v7.16b}, [x17], x3 + usubl2 v26.8h, v4.16b, v5.16b + usubl v25.8h, v6.8b, v7.8b + usubl2 v24.8h, v6.16b, v7.16b + saba v18.8h, v27.8h, v25.8h + subs w4, w4, #1 + saba v19.8h, v26.8h, v24.8h + + cbnz w4, 2b + +3: + sqsub v16.8h, v16.8h, v18.8h + sqsub v17.8h, v17.8h, v19.8h + ins v17.h[7], wzr + sqadd v16.8h, v16.8h, v17.8h + saddlv s16, v16.8h + sqabs s16, s16 + fmov w0, s16 + + mul w0, w0, w5 + add w0, w0, w9 + + ret +endfunc