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 8FCB74A77D for ; Sat, 6 Apr 2024 14:23:52 +0000 (UTC) Received: from [127.0.1.1] (localhost [127.0.0.1]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id 9536468D00E; Sat, 6 Apr 2024 17:23:48 +0300 (EEST) Received: from riki.gfrh.net (riki.gfrh.net [52.40.11.205]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTPS id C02AE68CB57 for ; Sat, 6 Apr 2024 17:23:41 +0300 (EEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=geoffhill.org; s=20220510; t=1712413418; 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=Wv1ARWCFFEyUocr3Fkr3Je8fmkViIccG+CWk8xhkIt4=; b=n8nfoujzwHjNl9SIXwQwPTwLbQglkhNFz64wJcYbMPsoV2LW7f+6vII5VlFUU1DT5IQD2D B8P7Jq7UFlhpxnw4cAENv+7ogZVoDvGghf4zn5uE3Th05UHB94sStcaiDxU+tFftB54QCR MZ/0d8m8QXBCsfskOFXqSVCcNdLrrJY= Received: by riki.gfrh.net (OpenSMTPD) with ESMTPSA id d063ae45 (TLSv1.3:TLS_AES_256_GCM_SHA384:256:NO); Sat, 6 Apr 2024 14:23:38 +0000 (UTC) Message-ID: <51f7be0a-4267-47bf-ab0b-bd6585806da7@geoffhill.org> Date: Sat, 6 Apr 2024 07:23:37 -0700 MIME-Version: 1.0 User-Agent: Mozilla Thunderbird From: Geoff Hill To: ffmpeg-devel@ffmpeg.org, =?UTF-8?Q?Martin_Storsj=C3=B6?= Content-Language: en-US Subject: [FFmpeg-devel] [PATCH v4 0/5] avcodec/ac3: Add aarch64 NEON DSP 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 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: Thanks Martin for your review and testing. Here's v4 with the following changes: * Use fmal in sum_square_butterfly_float loop. Faster. * Removed redundant loop bound zero checks in extract_exponents, sum_square_bufferfly_int32 and sum_square_bufferfly_float. * Fixed randomize_int24() to also use negative values. * Carry copyright from arm implementation over to aarch64. I did use this version as reference. * Fix indentation to match existing aarch64 assembly style. Tested once again on aarch64 and x86. On AWS Graviton2 (t4g.medium), GCC 12.3: $ tests/checkasm/checkasm --bench --test=ac3dsp ... NEON: - ac3dsp.ac3_exponent_min [OK] - ac3dsp.ac3_extract_exponents [OK] - ac3dsp.float_to_fixed24 [OK] - ac3dsp.ac3_sum_square_butterfly_int32 [OK] - ac3dsp.ac3_sum_square_butterfly_float [OK] checkasm: all 20 tests passed ac3_exponent_min_reuse0_c: 7.5 ac3_exponent_min_reuse0_neon: 7.5 ac3_exponent_min_reuse1_c: 1044.0 ac3_exponent_min_reuse1_neon: 57.0 ac3_exponent_min_reuse2_c: 2073.0 ac3_exponent_min_reuse2_neon: 73.7 ac3_exponent_min_reuse3_c: 2596.2 ac3_exponent_min_reuse3_neon: 154.0 ac3_exponent_min_reuse4_c: 3107.2 ac3_exponent_min_reuse4_neon: 169.2 ac3_exponent_min_reuse5_c: 3615.2 ac3_exponent_min_reuse5_neon: 185.2 ac3_extract_exponents_n512_c: 1672.0 ac3_extract_exponents_n512_neon: 517.5 ac3_extract_exponents_n768_c: 2505.0 ac3_extract_exponents_n768_neon: 770.5 ac3_extract_exponents_n1024_c: 3304.0 ac3_extract_exponents_n1024_neon: 1022.0 ac3_extract_exponents_n1280_c: 4163.5 ac3_extract_exponents_n1280_neon: 1279.5 ac3_extract_exponents_n1536_c: 5001.2 ac3_extract_exponents_n1536_neon: 1553.2 ac3_extract_exponents_n1792_c: 5823.5 ac3_extract_exponents_n1792_neon: 1851.7 ac3_extract_exponents_n2048_c: 6601.5 ac3_extract_exponents_n2048_neon: 2116.2 ac3_extract_exponents_n2304_c: 7425.2 ac3_extract_exponents_n2304_neon: 2382.7 ac3_extract_exponents_n2560_c: 8278.5 ac3_extract_exponents_n2560_neon: 2620.5 ac3_extract_exponents_n2816_c: 9079.5 ac3_extract_exponents_n2816_neon: 2893.2 ac3_extract_exponents_n3072_c: 10026.5 ac3_extract_exponents_n3072_neon: 3127.0 ac3_sum_square_bufferfly_float_c: 1647.5 ac3_sum_square_bufferfly_float_neon: 229.5 ac3_sum_square_bufferfly_int32_c: 963.5 ac3_sum_square_bufferfly_int32_neon: 546.5 float_to_fixed24_c: 2460.5 float_to_fixed24_neon: 561.5 Geoff Hill (5): avcodec/ac3: Implement float_to_fixed24 for aarch64 NEON avcodec/ac3: Implement ac3_exponent_min for aarch64 NEON avcodec/ac3: Implement ac3_extract_exponents for aarch64 NEON avcodec/ac3: Implement sum_square_butterfly_int32 for aarch64 NEON avcodec/ac3: Implement sum_square_butterfly_float for aarch64 NEON libavcodec/aarch64/Makefile | 2 + libavcodec/aarch64/ac3dsp_init_aarch64.c | 50 +++++++++ libavcodec/aarch64/ac3dsp_neon.S | 119 ++++++++++++++++++++ libavcodec/ac3dsp.c | 4 +- libavcodec/ac3dsp.h | 3 +- tests/checkasm/ac3dsp.c | 133 +++++++++++++++++++++++ 6 files changed, 309 insertions(+), 2 deletions(-) create mode 100644 libavcodec/aarch64/ac3dsp_init_aarch64.c create mode 100644 libavcodec/aarch64/ac3dsp_neon.S -- 2.42.0 _______________________________________________ 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".