From patchwork Wed Apr 3 06:43:26 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Geoff Hill X-Patchwork-Id: 47744 Delivered-To: ffmpegpatchwork2@gmail.com Received: by 2002:a05:6a20:9f96:b0:1a3:b6bb:3029 with SMTP id mm22csp213738pzb; Tue, 2 Apr 2024 23:43:52 -0700 (PDT) X-Forwarded-Encrypted: i=2; AJvYcCV+jrvg4N3uq6GwIFIVNQg0cuzk4FNRQLXycH4d6Yy12s2fasj787lbBcQz5oIzTq6k1CA4MeyfuplswqpzZyp0zacMaCwYndKvGg== X-Google-Smtp-Source: AGHT+IHy7ZXcTUx+fo0zB07hDU1uMb/EqfS0Tbt3PiM2q0DwnKS8uOVNuv4v7yG0wvug/q6mNKhy X-Received: by 2002:a50:c34c:0:b0:568:b0f4:fe69 with SMTP id q12-20020a50c34c000000b00568b0f4fe69mr1447169edb.12.1712126632091; Tue, 02 Apr 2024 23:43:52 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1712126632; cv=none; d=google.com; s=arc-20160816; b=OTKQhSYJ1PpVpF9tIGumViP4b/UEaDplyK/434ZHE7vhvneYupM/iWk2X0dCMS17vB /uzgsM4ypzCrnUEQDH3ZF7xOpJn8EU20AUzAVTwDw/5ZF8OpYlYQZa+vBUyNz/RhX9fW tkR2cU7UwiwuTfpbJXQbnQxeZZnZwKLRin7TRSlgyzifjffZCS95rsxLY8DEXHHZ/pHr tbKUR9baF24r4VCz9oCTifhEpE+k0ojfew83wX5EYCajxJ4tpXaLFze9vwDUfABPxuK0 EAkEkQ5Dr98fqmpLT5SA6p5/KV6xlAt/M0t+rEdWpMAXj0iDbcZnw5DQzDCoeMqQMlis yH1g== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=sender:errors-to:content-transfer-encoding:reply-to:list-subscribe :list-help:list-post:list-archive:list-unsubscribe:list-id :precedence:subject:content-language:to:from:user-agent:mime-version :date:message-id:dkim-signature:delivered-to; bh=4HUopN05UwVSIZKBVJAXjxOLTQ8Fxa0AQnoy4v77MPc=; fh=YOA8vD9MJZuwZ71F/05pj6KdCjf6jQRmzLS+CATXUQk=; b=ZiI+kDGlY/EJ/ZK0XKK1zuB8qzvZzyIRPjYoJdgIBqy5JUTIwpKTgjBwA0ZO6013W1 vhfPgHONviuOASSvVHZmb9xfdWz89/COQPK4+AChx+KIeTBMhN63clasTKOVGXAIq9sS E5WvIpUpa7FQy8NaKB9sFHKMRfz9ynmIX4P/h8ntMa++113LCDPkLu8VW77WiGKiTz18 /gjHNRY8tt4HqfrHkEW8k6FQv80DxN4qa/f+sBxKAqrGoteuU0KPndXHVDwM+WdWn3X2 l8Dh9JANAEjD65UI3CNHUf8pThaKl2OVcIwgKSj5jilYa2ktNfj8jkVgvlHV7wMzYOTI YQhw==; dara=google.com ARC-Authentication-Results: i=1; mx.google.com; dkim=neutral (body hash did not verify) header.i=@geoffhill.org header.s=20220510 header.b=iGOfKlLH; 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=NONE sp=NONE dis=NONE) header.from=geoffhill.org Return-Path: Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org. [79.124.17.100]) by mx.google.com with ESMTP id k7-20020a508ac7000000b0056bf30ff6b4si6345728edk.122.2024.04.02.23.43.51; Tue, 02 Apr 2024 23:43:52 -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=@geoffhill.org header.s=20220510 header.b=iGOfKlLH; 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=NONE sp=NONE dis=NONE) header.from=geoffhill.org Received: from [127.0.1.1] (localhost [127.0.0.1]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id A534968D0B7; Wed, 3 Apr 2024 09:43:39 +0300 (EEST) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from riki.gfrh.net (riki.gfrh.net [52.40.11.205]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTPS id 0673068D095 for ; Wed, 3 Apr 2024 09:43:32 +0300 (EEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=geoffhill.org; s=20220510; t=1712126606; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:mime-version:mime-version:content-type:content-type: content-transfer-encoding:content-transfer-encoding; bh=y6CA4nDbpmw8mTCNkHyvCIleOCyrruqldrbp9kHOWMk=; b=iGOfKlLHl5/xYXXno9ERZwbYGjaIxbqe+k0XTWwSzC6RC4XiEO9KERxfB3zJTwg9N9LfVT kMKeqgbfdkZhmMN4pMJMluRXZvbQox4oYZKNqBig2b7WkNE+QIVfVogmeDYE0Jt8LoSqlv RuJubgE5pkhxM77RUrfEt2F/GgAvD14= Received: by riki.gfrh.net (OpenSMTPD) with ESMTPSA id 08392bfe (TLSv1.3:TLS_AES_256_GCM_SHA384:256:NO) for ; Wed, 3 Apr 2024 06:43:26 +0000 (UTC) Message-ID: <35877d65-6a5c-4897-bf83-d6216b171aaf@geoffhill.org> Date: Tue, 2 Apr 2024 23:43:26 -0700 MIME-Version: 1.0 User-Agent: Mozilla Thunderbird From: Geoff Hill To: ffmpeg-devel@ffmpeg.org Content-Language: en-US Subject: [FFmpeg-devel] [PATCH v3 2/5] avcodec/ac3: Implement ac3_exponent_min for aarch64 NEON 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 Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" X-TUID: UHsEim38jQFc Signed-off-by: Geoff Hill --- libavcodec/aarch64/ac3dsp_init_aarch64.c | 2 ++ libavcodec/aarch64/ac3dsp_neon.S | 16 +++++++++ tests/checkasm/ac3dsp.c | 41 ++++++++++++++++++++++++ 3 files changed, 59 insertions(+) diff --git a/libavcodec/aarch64/ac3dsp_init_aarch64.c b/libavcodec/aarch64/ac3dsp_init_aarch64.c index e3320de0f5..8874b41393 100644 --- a/libavcodec/aarch64/ac3dsp_init_aarch64.c +++ b/libavcodec/aarch64/ac3dsp_init_aarch64.c @@ -25,6 +25,7 @@ #include "libavcodec/ac3dsp.h" #include "config.h" +void ff_ac3_exponent_min_neon(uint8_t *exp, int num_reuse_blocks, int nb_coefs); void ff_float_to_fixed24_neon(int32_t *dst, const float *src, size_t len); av_cold void ff_ac3dsp_init_aarch64(AC3DSPContext *c) @@ -32,5 +33,6 @@ av_cold void ff_ac3dsp_init_aarch64(AC3DSPContext *c) int cpu_flags = av_get_cpu_flags(); if (!have_neon(cpu_flags)) return; + c->ac3_exponent_min = ff_ac3_exponent_min_neon; c->float_to_fixed24 = ff_float_to_fixed24_neon; } diff --git a/libavcodec/aarch64/ac3dsp_neon.S b/libavcodec/aarch64/ac3dsp_neon.S index 6924645b7e..c8bdbb1bd3 100644 --- a/libavcodec/aarch64/ac3dsp_neon.S +++ b/libavcodec/aarch64/ac3dsp_neon.S @@ -20,6 +20,22 @@ #include "libavutil/aarch64/asm.S" +function ff_ac3_exponent_min_neon, export=1 + cbz w1, 2f +0: ld1 {v0.16b}, [x0] + mov w3, w1 + add x4, x0, #256 +1: ld1 {v1.16b}, [x4] + umin v0.16b, v0.16b, v1.16b + add x4, x4, #256 + subs w3, w3, #1 + b.gt 1b + st1 {v0.16b}, [x0], #16 + subs w2, w2, #16 + b.gt 0b +2: ret +endfunc + function ff_float_to_fixed24_neon, export=1 0: ld1 {v0.4s, v1.4s}, [x1], #32 fcvtzs v0.4s, v0.4s, #24 diff --git a/tests/checkasm/ac3dsp.c b/tests/checkasm/ac3dsp.c index 344e1fe5c2..acb00b6fe1 100644 --- a/tests/checkasm/ac3dsp.c +++ b/tests/checkasm/ac3dsp.c @@ -27,6 +27,14 @@ #include "checkasm.h" +#define randomize_exp(buf, len) \ + do { \ + int i; \ + for (i = 0; i < len; i++) { \ + buf[i] = (uint8_t)rnd(); \ + } \ + } while (0) \ + #define randomize_float(buf, len) \ do { \ int i; \ @@ -36,6 +44,38 @@ } \ } while (0) +static void check_ac3_exponent_min(AC3DSPContext *c) { +#define MAX_COEFS 256 +#define MAX_CTXT 6 +#define EXP_SIZE (MAX_CTXT * MAX_COEFS) + + LOCAL_ALIGNED_16(uint8_t, src, [EXP_SIZE]); + LOCAL_ALIGNED_16(uint8_t, v1, [EXP_SIZE]); + LOCAL_ALIGNED_16(uint8_t, v2, [EXP_SIZE]); + int n; + + declare_func(void, uint8_t *, int, int); + + for (n = 0; n < MAX_CTXT; ++n) { + if (check_func(c->ac3_exponent_min, "ac3_exponent_min_reuse%d", n)) { + randomize_exp(src, EXP_SIZE); + + memcpy(v1, src, EXP_SIZE); + memcpy(v2, src, EXP_SIZE); + + call_ref(v1, n, MAX_COEFS); + call_new(v2, n, MAX_COEFS); + + if (memcmp(v1, v2, EXP_SIZE) != 0) + fail(); + + bench_new(v2, n, MAX_COEFS); + } + } + + report("ac3_exponent_min"); +} + static void check_float_to_fixed24(AC3DSPContext *c) { #define BUF_SIZE 1024 LOCAL_ALIGNED_32(float, src, [BUF_SIZE]); @@ -66,5 +106,6 @@ void checkasm_check_ac3dsp(void) AC3DSPContext c; ff_ac3dsp_init(&c); + check_ac3_exponent_min(&c); check_float_to_fixed24(&c); }