From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org [79.124.17.100]) by master.gitmailbox.com (Postfix) with ESMTP id 49461444ED for ; Tue, 13 Sep 2022 11:59:31 +0000 (UTC) Received: from [127.0.1.1] (localhost [127.0.0.1]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id 13BF768BB5C; Tue, 13 Sep 2022 14:59:09 +0300 (EEST) 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 D443B68BB45 for ; Tue, 13 Sep 2022 14:59:01 +0300 (EEST) Received: by mail-lj1-f173.google.com with SMTP id y29so14061045ljq.7 for ; Tue, 13 Sep 2022 04:59:01 -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:subject:date; bh=KryZqZEyBO6Td7SuvyCYHfE6SI0SecoQ6GoZw27QFaU=; b=mh/scbH91PF4ZH9CnX567NKj+eY+FrxloNwInmtgF6lnroRHbEk6RYioJ630aFlBUl Dckp4yaRN4f5csXY+u8Jidy+xplYk6lcaYhRj7bWPfHk4HXRyqvOziqz3KBzwAtU7s7a W3w58i3aWn1OzfrGHjz7Fv1sSZHaY7wKZtRM9XmfNpUgKjwnD6a3RTnAKDEUurLi/Nm4 hy+QVF0c5dXeoiLWddj2yojYJwT7hMd9qHmCHegdlJEfzdJxPeqi37P89zc9zOuvNZLk ZxQyLf5epjAtgr6/ObgaLjLak0kLJWUwGIiANzIMbyyG7sIiSpf2vF0EBDZGIKMLSJt9 vH3A== 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 :subject:date; bh=KryZqZEyBO6Td7SuvyCYHfE6SI0SecoQ6GoZw27QFaU=; b=xQktuB4FpsLTDCAAHSwz0+z7msrq1PjaHYcF60bkaJICt7wqlrexZezMFB9jHxhsN0 LiN1XBOuIFmZQkYKQVdXVe0uytC8GR89a/FGx+W90haetodjFiutoEWWZO+3x5k9Rxai LzRAXT5hswpUsEdZFPHxwI0sNxTslVVzQi1C0wdSRA7FV5VGj9PHr0d3vSKmYVEWlwFv mEzmzj408xpu88lurj1p9ekHaLhsgegzy/ly6X6/GwY7+o4NuHb3K+wXTuaSb3zu06mf GYTBPuNOhUJ9OnqbprX6yYnuvoKmFQHtnNnhZMmOqXPgCEc0ZQ/VZPRi4PW+PRQYnbRW IEjg== X-Gm-Message-State: ACgBeo1rKlYPmxothWgh5fvyjGykiQ33bqfkQZYW+soLBI41FY3i3bHa 70o6n6rie6zGcUaCZoCiGGnTE/zNU0ZDww== X-Google-Smtp-Source: AA6agR7fe/mGDNy4RYYUSCnnw25BNcIvU8gVH4Y7aPc2J46/KzRHsTwPQ7GngWzHYRz2e2v8h68JBA== X-Received: by 2002:a05:651c:194d:b0:26c:1109:6cf2 with SMTP id bs13-20020a05651c194d00b0026c11096cf2mr2086556ljb.3.1663070340813; Tue, 13 Sep 2022 04:59:00 -0700 (PDT) Received: from hum-HP-ProBook-440-G7.semihalf.net ([83.142.187.84]) by smtp.gmail.com with ESMTPSA id f20-20020a05651232d400b0048a8586293asm580288lfg.48.2022.09.13.04.58.59 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Tue, 13 Sep 2022 04:59:00 -0700 (PDT) From: Hubert Mazur To: ffmpeg-devel@ffmpeg.org Date: Tue, 13 Sep 2022 13:58:24 +0200 Message-Id: <20220913115824.60792-4-hum@semihalf.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20220913115824.60792-1-hum@semihalf.com> References: <20220913115824.60792-1-hum@semihalf.com> MIME-Version: 1.0 Subject: [FFmpeg-devel] [PATCH 3/3] lavc/aarch64: Add neon implementation for pix_median_abs8 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 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" Archived-At: List-Archive: List-Post: Provide optimized implementation for pix_median_abs16 function. Performance comparison tests are shown below. - median_sad_1_c: 273.7 - median_sad_1_neon: 98.2 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 | 65 ++++++++++++++++++++++++ 2 files changed, 68 insertions(+) diff --git a/libavcodec/aarch64/me_cmp_init_aarch64.c b/libavcodec/aarch64/me_cmp_init_aarch64.c index d3fa047a86..e143f0816e 100644 --- a/libavcodec/aarch64/me_cmp_init_aarch64.c +++ b/libavcodec/aarch64/me_cmp_init_aarch64.c @@ -57,6 +57,8 @@ int nsse16_neon_wrapper(MpegEncContext *c, const uint8_t *s1, const uint8_t *s2, ptrdiff_t stride, int h); int pix_median_abs16_neon(MpegEncContext *v, const uint8_t *pix1, const uint8_t *pix2, ptrdiff_t stride, int h); +int pix_median_abs8_neon(MpegEncContext *v, const uint8_t *pix1, const uint8_t *pix2, + ptrdiff_t stride, int h); av_cold void ff_me_cmp_init_aarch64(MECmpContext *c, AVCodecContext *avctx) { @@ -85,6 +87,7 @@ av_cold void ff_me_cmp_init_aarch64(MECmpContext *c, AVCodecContext *avctx) c->nsse[0] = nsse16_neon_wrapper; c->median_sad[0] = pix_median_abs16_neon; + c->median_sad[1] = pix_median_abs8_neon; } } diff --git a/libavcodec/aarch64/me_cmp_neon.S b/libavcodec/aarch64/me_cmp_neon.S index 73701bd353..2017d0ce63 100644 --- a/libavcodec/aarch64/me_cmp_neon.S +++ b/libavcodec/aarch64/me_cmp_neon.S @@ -1092,3 +1092,68 @@ function vsad_intra8_neon, export=1 ret endfunc + +function pix_median_abs8_neon, export=1 + // x0 unused + // x1 uint8_t *pix1 + // x2 uint8_t *pix2 + // x3 ptrdiff_t stride + // w4 int h + + ld1 {v2.8b}, [x1], x3 + ld1 {v3.8b}, [x2], x3 + movi v31.8h, #0 + ext v0.8b, v2.8b, v2.8b, #1 + ext v1.8b, v3.8b, v3.8b, #1 + usubl v28.8h, v2.8b, v3.8b + usubl v26.8h, v0.8b, v1.8b + sub w4, w4, #1 // we need to make h-1 iterations + saba v31.8h, v26.8h, v28.8h + mov h18, v28.h[0] + cmp w4, #1 + sqabs h18, h18 + + b.lt 2f +1: + ld1 {v6.8b}, [x1], x3 // pix1 vector for V(j-1) + ld1 {v7.8b}, [x2], x3 // pix2 vector for V(j-1) + subs w4, w4, #1 + mov v2.8b, v6.8b + mov v3.8b, v7.8b + ext v4.8b, v6.8b, v6.8b, #1 // pix1 vector for V(j) + ext v5.8b, v7.8b, v7.8b, #1 // pix2 vector for V(j) + + // protected registers: v30, v29, v28, v27, v26, v25, v24, v23 + // scratch registers: v22, v21, v20, v19, v17 + + // To find median of three values, calculate sum of them + // and subtract max and min value from it. + usubl v30.8h, v6.8b, v7.8b // V(j-1) + usubl v24.8h, v4.8b, v5.8b // V(j) + mov v0.16b, v4.16b + mov v1.16b, v5.16b + sabd v20.8h, v30.8h, v28.8h + mov h17, v20.h[0] + add d18, d18, d17 + add v22.8h, v26.8h, v30.8h + smin v20.8h, v26.8h, v30.8h + smax v19.8h, v26.8h, v30.8h + sub v22.8h, v22.8h, v28.8h + smin v17.8h, v19.8h, v22.8h + mov v28.16b, v30.16b + smax v20.8h, v20.8h, v17.8h // median values lower half + smax v19.8h, v25.8h, v29.8h + saba v31.8h, v24.8h, v20.8h + mov v26.16b, v24.16b + smax v17.8h, v22.8h, v19.8h // median values upper half + + b.ne 1b +2: + ins v31.h[7], wzr + uaddlv s17, v31.8h + add d18, d18, d17 + fmov w0, s18 + + ret + +endfunc -- 2.34.1 _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".