From patchwork Sun Aug 14 16:48:30 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Timo Rothenpieler X-Patchwork-Id: 37272 Delivered-To: ffmpegpatchwork2@gmail.com Received: by 2002:a05:6a20:3d0d:b0:8d:a68e:8a0e with SMTP id y13csp880015pzi; Sun, 14 Aug 2022 09:49:29 -0700 (PDT) X-Google-Smtp-Source: AA6agR5eJaY9DlFfBDoCfAxTTVfiUwe61e31lAbonuTGTjK6dovvJR8+4Erqzvam7tPM97xe5aYh X-Received: by 2002:a05:6402:916:b0:440:9bc5:da7a with SMTP id g22-20020a056402091600b004409bc5da7amr11366022edz.323.1660495769614; Sun, 14 Aug 2022 09:49:29 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1660495769; cv=none; d=google.com; s=arc-20160816; b=QC2QdDyW2VmC0hhNQF9tYcpgbN5uYD6wgfqQsV+5b159BX4+uqIpGWHRYiE0nbqXjZ it6p9wZVBbpcZOe2RNo3cmP7f/JLLeR9/1i7/a0PQ23WdaGwSg0MfuKdux0DtGZ/UCSj HIlVq5ujN6QgCiRCN0sNXqI1hiyxnDlK5j0QJmTt6Mqp+EZzoDe/FwEQSeiV4aK/8Rn5 UoF5+N8iCW3Er1ZasBgodGddbrb2zpd4VqwAPOa5JVMs95b+c/ZH92zJDyk5FaVFzUij byKxguqNmSD58IV7KHcswfmPfp8scco/YRqEJw6FwmugpZe3gk+kwexGD+KoP/x9jiNc aVgQ== 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=Tx2/0KHmR0dgR98IFicQrkWAI6/BXSdlsZ+4ZwlTOQE=; b=t5AmR5yOQsEuFXuvlzVAkG50bd7eZjb9FF9hPuTPit5lyrdLMRGkmMmSlN+i1UnZPk Qcxg3yTNL+Vgcrf6fWNakBaVwU8xbNv4ogiwMdECPfpP2gOB2lB+sZxrlKxUEQuWgZIN dbY+n4cnv1f1j/prkTkfIUiHMmgMPib+9OIGMuOPcLEi6n4ZuwGNrPXwKjiprUI5F1CC pJ3ZxnsWe8f6jNoGCBjwMjnuLU9oq3xnob/taIabOG/lPeNpfEmdj2gNKljTqJkDmvi5 ngb/NsZKKyGU0jdLfU1QkRBUgH918PMjgm58jMVIuvfNxZhM3CHMpp2stRdyyu6kjvwT 9N8g== ARC-Authentication-Results: i=1; mx.google.com; dkim=neutral (body hash did not verify) header.i=@rothenpieler.org header.s=mail header.b=PznUu1g0; 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 du8-20020a17090772c800b00733ca35c0a6si6506875ejc.626.2022.08.14.09.49.29; Sun, 14 Aug 2022 09:49:29 -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=PznUu1g0; 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 5D06D68B94F; Sun, 14 Aug 2022 19:48:57 +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 D42D768B485 for ; Sun, 14 Aug 2022 19:48:46 +0300 (EEST) Received: from [authenticated] by btbn.de (Postfix) with ESMTPSA id 78A062AEB90; 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=9WPbIqU/uYpQobXyQEbdPxocb/q2ix5aNqo9y/XThbM=; h=From:To:Cc:Subject:Date:In-Reply-To:References; b=PznUu1g0UHtHY6WHHcbxN56QTR5E+ZgwGTUYDx/Z/PvzNLsx4sn0FBwKz5GpZL9qB +TiRniqf0WJCFlHoXdXfICaMQavWX8uju6ak6F72crAIQM5sZP+P62Bwjm4ctC9JGN 0G3WJf03Z3cLzyMs7qAHgG/br+AdMUWAC6uicewYpb1WiXB+vzmK7j3HOXHIeXGF9Y lwpPGAVGqgkDRb1QTMqFPJE0W1nTZkJycLfgBdxf8mH0/oDcDIpCKBZqMNIPs89B3+ Aywt9XKFem5DTENdL7fHsEP3qFhpsebYLC8mUA5vv7FsL/nIp68mSpmzFHvjQEVaZq UEB7nx278g+/A== From: Timo Rothenpieler To: ffmpeg-devel@ffmpeg.org Date: Sun, 14 Aug 2022 18:48:30 +0200 Message-Id: <20220814164833.19965-4-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 4/7] avutil/half2float: move non-inline init code out of header 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: IdkmqtzPzGlP --- libavcodec/Makefile | 8 +++--- libavcodec/exr.c | 2 +- libavcodec/exrenc.c | 2 +- libavcodec/float2half.c | 19 +++++++++++++ libavcodec/half2float.c | 19 +++++++++++++ libavcodec/pnmdec.c | 2 +- libavcodec/pnmenc.c | 2 +- libavutil/float2half.c | 53 ++++++++++++++++++++++++++++++++++ libavutil/float2half.h | 36 ++--------------------- libavutil/half2float.c | 63 +++++++++++++++++++++++++++++++++++++++++ libavutil/half2float.h | 46 ++---------------------------- 11 files changed, 166 insertions(+), 86 deletions(-) create mode 100644 libavcodec/float2half.c create mode 100644 libavcodec/half2float.c create mode 100644 libavutil/float2half.c create mode 100644 libavutil/half2float.c diff --git a/libavcodec/Makefile b/libavcodec/Makefile index 029f1bad3d..cb80f73d99 100644 --- a/libavcodec/Makefile +++ b/libavcodec/Makefile @@ -337,8 +337,8 @@ OBJS-$(CONFIG_EIGHTSVX_FIB_DECODER) += 8svx.o OBJS-$(CONFIG_ESCAPE124_DECODER) += escape124.o OBJS-$(CONFIG_ESCAPE130_DECODER) += escape130.o OBJS-$(CONFIG_EVRC_DECODER) += evrcdec.o acelp_vectors.o lsp.o -OBJS-$(CONFIG_EXR_DECODER) += exr.o exrdsp.o -OBJS-$(CONFIG_EXR_ENCODER) += exrenc.o +OBJS-$(CONFIG_EXR_DECODER) += exr.o exrdsp.o half2float.o +OBJS-$(CONFIG_EXR_ENCODER) += exrenc.o float2half.o OBJS-$(CONFIG_FASTAUDIO_DECODER) += fastaudio.o OBJS-$(CONFIG_FFV1_DECODER) += ffv1dec.o ffv1.o OBJS-$(CONFIG_FFV1_ENCODER) += ffv1enc.o ffv1.o @@ -570,8 +570,8 @@ OBJS-$(CONFIG_PGMYUV_DECODER) += pnmdec.o pnm.o OBJS-$(CONFIG_PGMYUV_ENCODER) += pnmenc.o OBJS-$(CONFIG_PGSSUB_DECODER) += pgssubdec.o OBJS-$(CONFIG_PGX_DECODER) += pgxdec.o -OBJS-$(CONFIG_PHM_DECODER) += pnmdec.o pnm.o -OBJS-$(CONFIG_PHM_ENCODER) += pnmenc.o +OBJS-$(CONFIG_PHM_DECODER) += pnmdec.o pnm.o half2float.o +OBJS-$(CONFIG_PHM_ENCODER) += pnmenc.o float2half.o OBJS-$(CONFIG_PHOTOCD_DECODER) += photocd.o OBJS-$(CONFIG_PICTOR_DECODER) += pictordec.o cga_data.o OBJS-$(CONFIG_PIXLET_DECODER) += pixlet.o diff --git a/libavcodec/exr.c b/libavcodec/exr.c index b0e2e85024..859dd6fedd 100644 --- a/libavcodec/exr.c +++ b/libavcodec/exr.c @@ -2208,7 +2208,7 @@ static av_cold int decode_init(AVCodecContext *avctx) float one_gamma = 1.0f / s->gamma; avpriv_trc_function trc_func = NULL; - init_half2float_tables(&s->h2f_tables); + ff_init_half2float_tables(&s->h2f_tables); s->avctx = avctx; diff --git a/libavcodec/exrenc.c b/libavcodec/exrenc.c index 356bd11543..3dad107d62 100644 --- a/libavcodec/exrenc.c +++ b/libavcodec/exrenc.c @@ -94,7 +94,7 @@ static av_cold int encode_init(AVCodecContext *avctx) { EXRContext *s = avctx->priv_data; - init_float2half_tables(&s->f2h_tables); + ff_init_float2half_tables(&s->f2h_tables); switch (avctx->pix_fmt) { case AV_PIX_FMT_GBRPF32: diff --git a/libavcodec/float2half.c b/libavcodec/float2half.c new file mode 100644 index 0000000000..90a6f63fac --- /dev/null +++ b/libavcodec/float2half.c @@ -0,0 +1,19 @@ +/* + * This file is part of FFmpeg. + * + * FFmpeg is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * FFmpeg is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with FFmpeg; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA + */ + +#include "libavutil/float2half.c" diff --git a/libavcodec/half2float.c b/libavcodec/half2float.c new file mode 100644 index 0000000000..1b023f96a5 --- /dev/null +++ b/libavcodec/half2float.c @@ -0,0 +1,19 @@ +/* + * This file is part of FFmpeg. + * + * FFmpeg is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * FFmpeg is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with FFmpeg; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA + */ + +#include "libavutil/half2float.c" diff --git a/libavcodec/pnmdec.c b/libavcodec/pnmdec.c index 6adc348ec8..fbed282e93 100644 --- a/libavcodec/pnmdec.c +++ b/libavcodec/pnmdec.c @@ -477,7 +477,7 @@ static av_cold int phm_dec_init(AVCodecContext *avctx) { PNMContext *s = avctx->priv_data; - init_half2float_tables(&s->h2f_tables); + ff_init_half2float_tables(&s->h2f_tables); return 0; } diff --git a/libavcodec/pnmenc.c b/libavcodec/pnmenc.c index 38a5d8172d..b052c03b21 100644 --- a/libavcodec/pnmenc.c +++ b/libavcodec/pnmenc.c @@ -294,7 +294,7 @@ static av_cold int phm_enc_init(AVCodecContext *avctx) { PHMEncContext *s = avctx->priv_data; - init_float2half_tables(&s->f2h_tables); + ff_init_float2half_tables(&s->f2h_tables); return 0; } diff --git a/libavutil/float2half.c b/libavutil/float2half.c new file mode 100644 index 0000000000..c79a3abfa1 --- /dev/null +++ b/libavutil/float2half.c @@ -0,0 +1,53 @@ +/* + * This file is part of FFmpeg. + * + * FFmpeg is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * FFmpeg is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with FFmpeg; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA + */ + +#include "libavutil/float2half.h" + +void ff_init_float2half_tables(Float2HalfTables *t) +{ + for (int i = 0; i < 256; i++) { + int e = i - 127; + + if (e < -24) { // Very small numbers map to zero + t->basetable[i|0x000] = 0x0000; + t->basetable[i|0x100] = 0x8000; + t->shifttable[i|0x000] = 24; + t->shifttable[i|0x100] = 24; + } else if (e < -14) { // Small numbers map to denorms + t->basetable[i|0x000] = (0x0400>>(-e-14)); + t->basetable[i|0x100] = (0x0400>>(-e-14)) | 0x8000; + t->shifttable[i|0x000] = -e-1; + t->shifttable[i|0x100] = -e-1; + } else if (e <= 15) { // Normal numbers just lose precision + t->basetable[i|0x000] = ((e + 15) << 10); + t->basetable[i|0x100] = ((e + 15) << 10) | 0x8000; + t->shifttable[i|0x000] = 13; + t->shifttable[i|0x100] = 13; + } else if (e < 128) { // Large numbers map to Infinity + t->basetable[i|0x000] = 0x7C00; + t->basetable[i|0x100] = 0xFC00; + t->shifttable[i|0x000] = 24; + t->shifttable[i|0x100] = 24; + } else { // Infinity and NaN's stay Infinity and NaN's + t->basetable[i|0x000] = 0x7C00; + t->basetable[i|0x100] = 0xFC00; + t->shifttable[i|0x000] = 13; + t->shifttable[i|0x100] = 13; + } + } +} diff --git a/libavutil/float2half.h b/libavutil/float2half.h index 3548aa1d45..20fdc2a36b 100644 --- a/libavutil/float2half.h +++ b/libavutil/float2half.h @@ -26,41 +26,9 @@ typedef struct Float2HalfTables { uint8_t shifttable[512]; } Float2HalfTables; -static void init_float2half_tables(Float2HalfTables *t) -{ - for (int i = 0; i < 256; i++) { - int e = i - 127; - - if (e < -24) { // Very small numbers map to zero - t->basetable[i|0x000] = 0x0000; - t->basetable[i|0x100] = 0x8000; - t->shifttable[i|0x000] = 24; - t->shifttable[i|0x100] = 24; - } else if (e < -14) { // Small numbers map to denorms - t->basetable[i|0x000] = (0x0400>>(-e-14)); - t->basetable[i|0x100] = (0x0400>>(-e-14)) | 0x8000; - t->shifttable[i|0x000] = -e-1; - t->shifttable[i|0x100] = -e-1; - } else if (e <= 15) { // Normal numbers just lose precision - t->basetable[i|0x000] = ((e + 15) << 10); - t->basetable[i|0x100] = ((e + 15) << 10) | 0x8000; - t->shifttable[i|0x000] = 13; - t->shifttable[i|0x100] = 13; - } else if (e < 128) { // Large numbers map to Infinity - t->basetable[i|0x000] = 0x7C00; - t->basetable[i|0x100] = 0xFC00; - t->shifttable[i|0x000] = 24; - t->shifttable[i|0x100] = 24; - } else { // Infinity and NaN's stay Infinity and NaN's - t->basetable[i|0x000] = 0x7C00; - t->basetable[i|0x100] = 0xFC00; - t->shifttable[i|0x000] = 13; - t->shifttable[i|0x100] = 13; - } - } -} +void ff_init_float2half_tables(Float2HalfTables *t); -static uint16_t float2half(uint32_t f, const Float2HalfTables *t) +static inline uint16_t float2half(uint32_t f, const Float2HalfTables *t) { uint16_t h; diff --git a/libavutil/half2float.c b/libavutil/half2float.c new file mode 100644 index 0000000000..1967126f76 --- /dev/null +++ b/libavutil/half2float.c @@ -0,0 +1,63 @@ +/* + * This file is part of FFmpeg. + * + * FFmpeg is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * FFmpeg is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with FFmpeg; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA + */ + +#include "libavutil/half2float.h" + +static uint32_t convertmantissa(uint32_t i) +{ + int32_t m = i << 13; // Zero pad mantissa bits + int32_t e = 0; // Zero exponent + + while (!(m & 0x00800000)) { // While not normalized + e -= 0x00800000; // Decrement exponent (1<<23) + m <<= 1; // Shift mantissa + } + + m &= ~0x00800000; // Clear leading 1 bit + e += 0x38800000; // Adjust bias ((127-14)<<23) + + return m | e; // Return combined number +} + +void ff_init_half2float_tables(Half2FloatTables *t) +{ + t->mantissatable[0] = 0; + for (int i = 1; i < 1024; i++) + t->mantissatable[i] = convertmantissa(i); + for (int i = 1024; i < 2048; i++) + t->mantissatable[i] = 0x38000000UL + ((i - 1024) << 13UL); + for (int i = 2048; i < 3072; i++) + t->mantissatable[i] = t->mantissatable[i - 1024] | 0x400000UL; + t->mantissatable[2048] = t->mantissatable[1024]; + + t->exponenttable[0] = 0; + for (int i = 1; i < 31; i++) + t->exponenttable[i] = i << 23; + for (int i = 33; i < 63; i++) + t->exponenttable[i] = 0x80000000UL + ((i - 32) << 23UL); + t->exponenttable[31]= 0x47800000UL; + t->exponenttable[32]= 0x80000000UL; + t->exponenttable[63]= 0xC7800000UL; + + t->offsettable[0] = 0; + for (int i = 1; i < 64; i++) + t->offsettable[i] = 1024; + t->offsettable[31] = 2048; + t->offsettable[32] = 0; + t->offsettable[63] = 2048; +} diff --git a/libavutil/half2float.h b/libavutil/half2float.h index 5696567a8c..428a27a19f 100644 --- a/libavutil/half2float.h +++ b/libavutil/half2float.h @@ -27,51 +27,9 @@ typedef struct Half2FloatTables { uint16_t offsettable[64]; } Half2FloatTables; -static uint32_t convertmantissa(uint32_t i) -{ - int32_t m = i << 13; // Zero pad mantissa bits - int32_t e = 0; // Zero exponent - - while (!(m & 0x00800000)) { // While not normalized - e -= 0x00800000; // Decrement exponent (1<<23) - m <<= 1; // Shift mantissa - } - - m &= ~0x00800000; // Clear leading 1 bit - e += 0x38800000; // Adjust bias ((127-14)<<23) - - return m | e; // Return combined number -} - -static void init_half2float_tables(Half2FloatTables *t) -{ - t->mantissatable[0] = 0; - for (int i = 1; i < 1024; i++) - t->mantissatable[i] = convertmantissa(i); - for (int i = 1024; i < 2048; i++) - t->mantissatable[i] = 0x38000000UL + ((i - 1024) << 13UL); - for (int i = 2048; i < 3072; i++) - t->mantissatable[i] = t->mantissatable[i - 1024] | 0x400000UL; - t->mantissatable[2048] = t->mantissatable[1024]; - - t->exponenttable[0] = 0; - for (int i = 1; i < 31; i++) - t->exponenttable[i] = i << 23; - for (int i = 33; i < 63; i++) - t->exponenttable[i] = 0x80000000UL + ((i - 32) << 23UL); - t->exponenttable[31]= 0x47800000UL; - t->exponenttable[32]= 0x80000000UL; - t->exponenttable[63]= 0xC7800000UL; - - t->offsettable[0] = 0; - for (int i = 1; i < 64; i++) - t->offsettable[i] = 1024; - t->offsettable[31] = 2048; - t->offsettable[32] = 0; - t->offsettable[63] = 2048; -} +void ff_init_half2float_tables(Half2FloatTables *t); -static uint32_t half2float(uint16_t h, const Half2FloatTables *t) +static inline uint32_t half2float(uint16_t h, const Half2FloatTables *t) { uint32_t f;