From patchwork Fri Jul 15 08:02:26 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Hubert Mazur X-Patchwork-Id: 36799 Delivered-To: ffmpegpatchwork2@gmail.com Received: by 2002:a05:6a20:43a1:b0:8b:53c6:bec9 with SMTP id i33csp125377pzl; Fri, 15 Jul 2022 01:03:43 -0700 (PDT) X-Google-Smtp-Source: AGRyM1tAazegxettjcgzjRW0VXnXs6zyF2vu23rRX3GPJAWR3SC9GpvHfXx3uE23ZX3PV6ItBHyV X-Received: by 2002:a17:906:cc0e:b0:72b:311b:8ed9 with SMTP id ml14-20020a170906cc0e00b0072b311b8ed9mr12523374ejb.186.1657872223637; Fri, 15 Jul 2022 01:03:43 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1657872223; cv=none; d=google.com; s=arc-20160816; b=TVn5XBRan/LG4aH1ofm60pwGZk5pEHFlCXZ3XRljgn40Z8zpQhcMtoH/M+wHVNT5Qh VfghUCLnJWuHLlpWJiq1g0usYTLsFn8YWN1IHslpxf06hNw4hLqR/UdgUMhe+UjF4P0C LRXon/LIST95ZdUmwedDRj8WY31Dzj/K0G5aNbD/TPJtEVgY9OyhpB2EtoFNSfUa/JzY Ms2+7IZ8dBilwVJDz1hHGBReJ96l4+kRi14sg7trbiegzFvjzPcYGA30dutYIVkRf+Vb iUmrHOMm7lISnhVRBEjLULozR+N1DboQ2gorWvjgvcUost8FEFZHumw7AKmgZ7HIU0Em oKUw== 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=hjEcwsqAaCfYYF4R4kboc/goTgYZQhCO5zW1xWOPhD0=; b=H9iugUy87KbBFwRq6AHLzbpU1CCNPnhOIowRCuxA3moaa7puyMV9DV801/hxXeD9zu K1w+uWj9LcxrfLp7VCBJq67pVO23W1ZJdP3m7ZIpHF7xox0jqPT71ktytQd+/qlGlCVE Iqf2yQBGvBu37oxJSzzNuqVZxtLi5uCG+yFXCPmFg0s4LRSoVPWatJxvwkVMaFUZ1oS2 PSSkGp7ttmzdw8iAdsSDt2SOTMkEO7NBHIvZeT1o2u3XB6R8or7fPDhgr4vQ2/vyadO/ N5EhhIl9XW9vGLBsbWU88nAp4/DXfPPWj+1rtA7djgb2Zsqbcju1HL5peU1rDa0MgSMj Bl+A== ARC-Authentication-Results: i=1; mx.google.com; dkim=neutral (body hash did not verify) header.i=@semihalf.com header.s=google header.b="d1u+g/du"; 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 i9-20020a0564020f0900b00435a9115e94si4805855eda.38.2022.07.15.01.03.42; Fri, 15 Jul 2022 01:03:43 -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="d1u+g/du"; 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 E0E1A68B98D; Fri, 15 Jul 2022 11:03:10 +0300 (EEST) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from mail-lj1-f169.google.com (mail-lj1-f169.google.com [209.85.208.169]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTPS id 8E83E68B9BF for ; Fri, 15 Jul 2022 11:03:04 +0300 (EEST) Received: by mail-lj1-f169.google.com with SMTP id p6so4863890ljc.8 for ; Fri, 15 Jul 2022 01:03:04 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=semihalf.com; s=google; h=from:to:cc:subject:date:message-id:in-reply-to:references :mime-version:content-transfer-encoding; bh=J6EuxDY1SXY5APyH+DpJ8Q0rPOWudzi3L2/8LZuA2Ag=; b=d1u+g/duGYewl2+ahEE/3klL7nttYyGhhShBFS6q6G28Dz2cYrE2f/Zlb4XWsHmnbK 753+SQs6CIC6ggDzMLuSSdJrWGjKYGGfQoTyJn8fZiUUiKzH4z2n9HNXO6GJedXX9EwT Z+8fH6W1Gsw25qFKOXu0ekVcXfPMdvTp7V16tQ9gwOS1OdX6OIvpwqIle9XB/dWci6Dm 9brKh0q3Yok6K5btVQ6pDq3hb40oliUC+4K1SJWNaIELbX58qZ64UtU5Htb5Kz/5THjg andznbB3r8a6QRXnQDndzUojl1/WFb0MZTfeR+cojCq7wRc++AnUKByx4xeGs7+dcN5C 5p4A== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references:mime-version:content-transfer-encoding; bh=J6EuxDY1SXY5APyH+DpJ8Q0rPOWudzi3L2/8LZuA2Ag=; b=3FqAC8wPI5VI/EbsCK+3ZoEVALmqje7sdcdxjUbv3DweytKgwgaGRC3N3sq//48ECo pXULtHEMlZd3JWldASySO1u2RPwC4xIEPM2V3TV4ubyjhvcHlydJd+VfqQVPRxYffAPC XulJn/eg7TIW9kF3ckTKqiETwbSvZOTn0X++Qrig9KS5vei9kYRUJwC4pP584iw+Le8M HaRUsg6YztQI3C0945vxsLap7xI7YJut5BDv8OoQX29fRJ35rLwq0gyx1WRW3S4GaB10 EiRxoDY6sUqRYdnGyT0tfVZBfdJCzUTsUyULl9rhqXXpEMOwDWxgMAMTT8MNF7/WruUU YhHg== X-Gm-Message-State: AJIora9WznqCoz+PT1T1BAc4E2JtAdQ8fvQWe7bqRZO+4Uu29IkidKOQ Hd2ynDanhRd/XKkgn9g/VoPoB15ws/v1qu8W X-Received: by 2002:a2e:a9a6:0:b0:25d:601a:d3d8 with SMTP id x38-20020a2ea9a6000000b0025d601ad3d8mr6265059ljq.141.1657872183534; Fri, 15 Jul 2022 01:03:03 -0700 (PDT) Received: from hum-HP-ProBook-440-G7.semihalf.net (adrf243.neoplus.adsl.tpnet.pl. [79.185.165.243]) by smtp.gmail.com with ESMTPSA id k1-20020a2eb741000000b0025d542731ffsm645548ljo.5.2022.07.15.01.03.02 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Fri, 15 Jul 2022 01:03:03 -0700 (PDT) From: Hubert Mazur To: ffmpeg-devel@ffmpeg.org Date: Fri, 15 Jul 2022 10:02:26 +0200 Message-Id: <20220715080228.686736-4-hum@semihalf.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20220715080228.686736-1-hum@semihalf.com> References: <20220715080228.686736-1-hum@semihalf.com> MIME-Version: 1.0 Subject: [FFmpeg-devel] [PATCH 3/5] lavc/aarch64: Add neon implementation for pix_abs16_y2 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: 3wiWQYEIJxN/ Provide optimized implementation of pix_abs16_y2 function for arm64. Performance comparison tests are shown below. pix_abs_0_2_c: 308.5 pix_abs_0_2_neon: 39.2 Benchmarks and tests 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 | 74 ++++++++++++++++++++++++ 2 files changed, 77 insertions(+) diff --git a/libavcodec/aarch64/me_cmp_init_aarch64.c b/libavcodec/aarch64/me_cmp_init_aarch64.c index 72a2062e7e..07d62cc1e5 100644 --- a/libavcodec/aarch64/me_cmp_init_aarch64.c +++ b/libavcodec/aarch64/me_cmp_init_aarch64.c @@ -29,6 +29,8 @@ int ff_pix_abs16_xy2_neon(MpegEncContext *s, uint8_t *blk1, uint8_t *blk2, ptrdiff_t stride, int h); int ff_pix_abs16_x2_neon(MpegEncContext *v, uint8_t *pix1, uint8_t *pix2, ptrdiff_t stride, int h); +int ff_pix_abs16_y2_neon(MpegEncContext *v, uint8_t *pix1, uint8_t *pix2, + ptrdiff_t stride, int h); int sse16_neon(MpegEncContext *v, uint8_t *pix1, uint8_t *pix2, ptrdiff_t stride, int h); @@ -42,6 +44,7 @@ av_cold void ff_me_cmp_init_aarch64(MECmpContext *c, AVCodecContext *avctx) if (have_neon(cpu_flags)) { c->pix_abs[0][0] = ff_pix_abs16_neon; c->pix_abs[0][1] = ff_pix_abs16_x2_neon; + c->pix_abs[0][2] = ff_pix_abs16_y2_neon; c->pix_abs[0][3] = ff_pix_abs16_xy2_neon; c->sad[0] = ff_pix_abs16_neon; diff --git a/libavcodec/aarch64/me_cmp_neon.S b/libavcodec/aarch64/me_cmp_neon.S index bacf151314..858833b0ae 100644 --- a/libavcodec/aarch64/me_cmp_neon.S +++ b/libavcodec/aarch64/me_cmp_neon.S @@ -425,3 +425,77 @@ function sse4_neon, export=1 ret endfunc + +function ff_pix_abs16_y2_neon, export=1 + // x0 unused + // x1 uint8_t *pix1 + // x2 uint8_t *pix2 + // x3 ptrdiff_t stride + // x4 int h + // x5 uint8_t *pix2 + stride + + // initialize buffers + movi d18, #0 + add x5, x2, x3 + cmp w4, #4 + b.lt 2f + +// make 4 iterations at once +1: + + // abs(pix1[0], avg2(pix2[0], pix2[0 + stride])) + // avg2(a, b) = (((a) + (b) + 1) >> 1) + // abs(x) = (x < 0 ? (-x) : (x)) + + ld1 {v1.16b}, [x2], x3 + ld1 {v2.16b}, [x5], x3 + urhadd v30.16b, v1.16b, v2.16b + ld1 {v0.16b}, [x1], x3 + uabdl v29.8h, v0.8b, v30.8b + ld1 {v4.16b}, [x2], x3 + uabdl2 v28.8h, v0.16b, v30.16b + ld1 {v5.16b}, [x5], x3 + urhadd v27.16b, v4.16b, v5.16b + ld1 {v3.16b}, [x1], x3 + uabal v29.8h, v3.8b, v27.8b + ld1 {v7.16b}, [x2], x3 + uabal2 v28.8h, v3.16b, v27.16b + ld1 {v20.16b}, [x5], x3 + urhadd v26.16b, v7.16b, v20.16b + ld1 {v6.16b}, [x1], x3 + uabal v29.8h, v6.8b, v26.8b + ld1 {v22.16b}, [x2], x3 + uabal2 v28.8h, v6.16b, v26.16b + ld1 {v23.16b}, [x5], x3 + urhadd v25.16b, v22.16b, v23.16b + ld1 {v21.16b}, [x1], x3 + uabal v29.8h, v21.8b, v25.8b + uabal2 v28.8h, v21.16b, v25.16b + + add v29.8h, v29.8h, v28.8h + sub w4, w4, #4 + uaddlv s16, v29.8h + cmp w4, #4 + add d18, d18, d16 + b.ge 1b + cbz w4, 3f + +// iterate by one +2: + + ld1 {v0.16b}, [x1], x3 + ld1 {v1.16b}, [x2], x3 + urhadd v30.16b, v1.16b, v2.16b + ld1 {v2.16b}, [x5], x3 + uabd v30.16b, v30.16b, v30.16b + + uaddlv h17, v30.16b + subs w4, w4, #1 + add d18, d18, d17 + b.ne 2b + +3: + fmov w0, s18 + + ret +endfunc