From patchwork Sun Aug 14 16:48:31 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Timo Rothenpieler X-Patchwork-Id: 37273 Delivered-To: ffmpegpatchwork2@gmail.com Received: by 2002:a05:6a20:3d0d:b0:8d:a68e:8a0e with SMTP id y13csp880071pzi; Sun, 14 Aug 2022 09:49:39 -0700 (PDT) X-Google-Smtp-Source: AA6agR5CKFkXcoZd1ul3mCdfXQed8EAiwBoA9JrutKHUPNq4h/mljM5uOML/4JMlYseVhrOOW33k X-Received: by 2002:a05:6402:51cb:b0:43e:8f40:858d with SMTP id r11-20020a05640251cb00b0043e8f40858dmr11100091edd.391.1660495779296; Sun, 14 Aug 2022 09:49:39 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1660495779; cv=none; d=google.com; s=arc-20160816; b=Wj88vVwkWA1tN0QsSx/hUdtu/iOB0EDWIX0KYuiL1mlTRdcSZzR/ur5ohT+O0Ao0hv ea4MOtrmqCARQtLQ+RP8S8Sj4gwqlStxuLUegGr8Pko9f4rsgDeTjMG3xzuzfjfVx2tb 5Te2m+NFg6/EHgzPvBWN0bF3c5+oFew+Q7cdSn4l3Sn/QsjAXdhB1860MjzD9P9ml+xY /Mjx7FSDOnWQlWc0QzLPtaYTc7oGU8JipxESKMioZB2hz66sghXa4iaDVFWLoKMNg4fi +J+VrkrWoIVGldm3yTrch3CO8YqEhxkbjxrxWNIV5b/+CHzwDCwTdMPmoO77qnmoBOGd 9Kzg== 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=1AbL9FFH41K7EHK3OZc/ySirFpeZ2gPNS/Ksn6juQkA=; b=ZnpY+IMHvY+8YFpMiOf3K/lBugZoBYzsIVUiXx3Cy1uZFtTjlWCPKSlsC7ZcKs46XR gy5MhbdNMNKVXST4YsCNl03FUo0BjeFH6CB6EOlM4oBHcW8VeSYXXxrdrNZ+dP2fnI1o ijydqB3//OF+mH6LSJKlqUgpv//x8KduDzZPEoL51wjuLlceHSywknmvXpI2q9820+eI xnRRRp9VItZnDoRhxOYsj6AYYSLB6US1ZbMuHWY+ZVPJy3edkuq1YORx+sadcs0g/9Ej KXFcsQ21uYh/RIrbRxwFFOZVlBXp88rmmGwuVZL7/QRgUmTlUCF5am6MAS+iXcqP8Gds ceQg== ARC-Authentication-Results: i=1; mx.google.com; dkim=neutral (body hash did not verify) header.i=@rothenpieler.org header.s=mail header.b=kuwBVNKi; 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=rothenpieler.org Return-Path: Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org. [79.124.17.100]) by mx.google.com with ESMTP id hc34-20020a17090716a200b0073832e13344si3591541ejc.86.2022.08.14.09.49.38; Sun, 14 Aug 2022 09:49:39 -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=@rothenpieler.org header.s=mail header.b=kuwBVNKi; 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=rothenpieler.org Received: from [127.0.1.1] (localhost [127.0.0.1]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id 6978A68B959; Sun, 14 Aug 2022 19:48:58 +0300 (EEST) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from btbn.de (btbn.de [136.243.74.85]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTPS id F092568B8C1 for ; Sun, 14 Aug 2022 19:48:46 +0300 (EEST) Received: from [authenticated] by btbn.de (Postfix) with ESMTPSA id 95A002AEB91; Sun, 14 Aug 2022 18:48:46 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=rothenpieler.org; s=mail; t=1660495726; bh=Df/+cqjfn0qlYxkuWAJIdtwvGQfuxzhyrsuK/yhR48g=; h=From:To:Cc:Subject:Date:In-Reply-To:References; b=kuwBVNKil6sCmvBj+zWmxMUmLORCctdy5PcvJF0c0xaUEHoMwyTlTtq0KnuUCsYTo CFkMZlQ/kFHn9Qmi29VHoUSAlFvolFSmeTz6hynJB7G2nV7bLGiKeuk5NdcgIwhbb0 5zWAOpV2QzFhNPf1jYo8oapws2nUXHMR4A9zSg70N38W10cg6ibuEZeWY/h+Qj3uXQ 61FdrnlQeduXQDScbAPwRz61nvV92t2roKYX/pWKDGyiFlrycOgSX2RYwGn1cG4Dnq B6Sv0pLX5GxMnBCewWqHMPWiXhYF1k/rEAnJBn22VLSHUgwUyeo1ZQkRC7D3DpcrUb FE3/HRXCfegiA== From: Timo Rothenpieler To: ffmpeg-devel@ffmpeg.org Date: Sun, 14 Aug 2022 18:48:31 +0200 Message-Id: <20220814164833.19965-5-timo@rothenpieler.org> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20220814164833.19965-1-timo@rothenpieler.org> References: <20220814164833.19965-1-timo@rothenpieler.org> MIME-Version: 1.0 Subject: [FFmpeg-devel] [PATCH v2 5/7] avutil/half2float: use native _Float16 if available 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: Timo Rothenpieler Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" X-TUID: WMykrLZh111e _Float16 support was available on arm/aarch64 for a while, and with gcc 12 was enabled on x86 as long as SSE2 is supported. If the target arch supports f16c, gcc emits fairly efficient assembly, taking advantage of it. This is the case on x86-64-v3 or higher. Same goes on arm, which has native float16 support. On x86, without f16c, it emulates it in software using sse2 instructions. This has shown to perform rather poorly: _Float16 full SSE2 emulation: frame=50074 fps=848 q=-0.0 size=N/A time=00:33:22.96 bitrate=N/A speed=33.9x _Float16 f16c accelerated (Zen2, --cpu=znver2): frame=50636 fps=1965 q=-0.0 Lsize=N/A time=00:33:45.40 bitrate=N/A speed=78.6x classic half2float full software implementation: frame=49926 fps=1605 q=-0.0 Lsize=N/A time=00:33:17.00 bitrate=N/A speed=64.2x Hence an additional check was introduced, that only enables use of _Float16 on x86 if f16c is being utilized. On aarch64, a similar uplift in performance is seen: RPi4 half2float full software implementation: frame= 6088 fps=126 q=-0.0 Lsize=N/A time=00:04:03.48 bitrate=N/A speed=5.06x RPi4 _Float16: frame= 6103 fps=158 q=-0.0 Lsize=N/A time=00:04:04.08 bitrate=N/A speed=6.32x Since arm/aarch64 always natively support 16 bit floats, it can always be considered fast there. I'm not aware of any additional platforms that currently support _Float16. And if there are, they should be considered non-fast until proven fast. --- configure | 12 ++++++++++++ libavutil/float2half.c | 2 ++ libavutil/float2half.h | 16 ++++++++++++++++ libavutil/half2float.c | 4 ++++ libavutil/half2float.h | 16 ++++++++++++++++ 5 files changed, 50 insertions(+) diff --git a/configure b/configure index fe94941a03..ea50c94002 100755 --- a/configure +++ b/configure @@ -2145,6 +2145,7 @@ ARCH_FEATURES=" fast_64bit fast_clz fast_cmov + fast_float16 local_aligned simd_align_16 simd_align_32 @@ -5127,6 +5128,8 @@ elif enabled arm; then ;; esac + test_cflags -mfp16-format=ieee && add_cflags -mfp16-format=ieee + elif enabled avr32; then case $cpu in @@ -6231,6 +6234,15 @@ check_builtin sync_val_compare_and_swap "" "int *ptr; int oldval, newval; __sync check_builtin gmtime_r time.h "time_t *time; struct tm *tm; gmtime_r(time, tm)" check_builtin localtime_r time.h "time_t *time; struct tm *tm; localtime_r(time, tm)" +check_builtin float16 "" "_Float16 f16var" +if enabled float16; then + if enabled x86; then + test_cpp_condition stddef.h "defined(__F16C__)" && enable fast_float16 + elif enabled arm || enabled aarch64; then + enable fast_float16 + fi +fi + case "$custom_allocator" in jemalloc) # jemalloc by default does not use a prefix diff --git a/libavutil/float2half.c b/libavutil/float2half.c index c79a3abfa1..1a283956e7 100644 --- a/libavutil/float2half.c +++ b/libavutil/float2half.c @@ -20,6 +20,7 @@ void ff_init_float2half_tables(Float2HalfTables *t) { +#if !HAVE_FAST_FLOAT16 for (int i = 0; i < 256; i++) { int e = i - 127; @@ -50,4 +51,5 @@ void ff_init_float2half_tables(Float2HalfTables *t) t->shifttable[i|0x100] = 13; } } +#endif } diff --git a/libavutil/float2half.h b/libavutil/float2half.h index 20fdc2a36b..e619046911 100644 --- a/libavutil/float2half.h +++ b/libavutil/float2half.h @@ -20,21 +20,37 @@ #define AVUTIL_FLOAT2HALF_H #include +#include "intfloat.h" + +#include "config.h" typedef struct Float2HalfTables { +#if HAVE_FAST_FLOAT16 + uint8_t dummy; +#else uint16_t basetable[512]; uint8_t shifttable[512]; +#endif } Float2HalfTables; void ff_init_float2half_tables(Float2HalfTables *t); static inline uint16_t float2half(uint32_t f, const Float2HalfTables *t) { +#if HAVE_FAST_FLOAT16 + union { + _Float16 f; + uint16_t i; + } u; + u.f = av_int2float(f); + return u.i; +#else uint16_t h; h = t->basetable[(f >> 23) & 0x1ff] + ((f & 0x007fffff) >> t->shifttable[(f >> 23) & 0x1ff]); return h; +#endif } #endif /* AVUTIL_FLOAT2HALF_H */ diff --git a/libavutil/half2float.c b/libavutil/half2float.c index 1967126f76..4de2180a19 100644 --- a/libavutil/half2float.c +++ b/libavutil/half2float.c @@ -18,6 +18,7 @@ #include "libavutil/half2float.h" +#if !HAVE_FAST_FLOAT16 static uint32_t convertmantissa(uint32_t i) { int32_t m = i << 13; // Zero pad mantissa bits @@ -33,9 +34,11 @@ static uint32_t convertmantissa(uint32_t i) return m | e; // Return combined number } +#endif void ff_init_half2float_tables(Half2FloatTables *t) { +#if !HAVE_FAST_FLOAT16 t->mantissatable[0] = 0; for (int i = 1; i < 1024; i++) t->mantissatable[i] = convertmantissa(i); @@ -60,4 +63,5 @@ void ff_init_half2float_tables(Half2FloatTables *t) t->offsettable[31] = 2048; t->offsettable[32] = 0; t->offsettable[63] = 2048; +#endif } diff --git a/libavutil/half2float.h b/libavutil/half2float.h index 428a27a19f..dbd5e7150f 100644 --- a/libavutil/half2float.h +++ b/libavutil/half2float.h @@ -20,22 +20,38 @@ #define AVUTIL_HALF2FLOAT_H #include +#include "intfloat.h" + +#include "config.h" typedef struct Half2FloatTables { +#if HAVE_FAST_FLOAT16 + uint8_t dummy; +#else uint32_t mantissatable[3072]; uint32_t exponenttable[64]; uint16_t offsettable[64]; +#endif } Half2FloatTables; void ff_init_half2float_tables(Half2FloatTables *t); static inline uint32_t half2float(uint16_t h, const Half2FloatTables *t) { +#if HAVE_FAST_FLOAT16 + union { + _Float16 f; + uint16_t i; + } u; + u.i = h; + return av_float2int(u.f); +#else uint32_t f; f = t->mantissatable[t->offsettable[h >> 10] + (h & 0x3ff)] + t->exponenttable[h >> 10]; return f; +#endif } #endif /* AVUTIL_HALF2FLOAT_H */