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 249A9403D5 for ; Mon, 22 Aug 2022 15:27:31 +0000 (UTC) Received: from [127.0.1.1] (localhost [127.0.0.1]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id DB92A68BA50; Mon, 22 Aug 2022 18:26:53 +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 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-Google-Smtp-Source: AA6agR6BSfr5PPYjEbLFqHzIRqEvH+vDW00X01edhtknBHSSk4x66SiIdyohLlJ5+vw/yw51MH0P+g== 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 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: 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 -- 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".