From patchwork Wed Aug 10 20:47:09 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Timo Rothenpieler X-Patchwork-Id: 37227 Delivered-To: ffmpegpatchwork2@gmail.com Received: by 2002:a05:6a20:3d0d:b0:8d:a68e:8a0e with SMTP id y13csp1142710pzi; Wed, 10 Aug 2022 13:48:47 -0700 (PDT) X-Google-Smtp-Source: AA6agR4NSLG8xklyYBEY9gN/yI+MUrFXUUixkw4jalxYOr+w0ncjOMNMHpRbh4LCa9cLz1m0bJXL X-Received: by 2002:a05:6402:3220:b0:43d:ca4f:d2b9 with SMTP id g32-20020a056402322000b0043dca4fd2b9mr27936753eda.177.1660164526998; Wed, 10 Aug 2022 13:48:46 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1660164526; cv=none; d=google.com; s=arc-20160816; b=T+6G8vxpf9ZLEl3WdCJ56v3LEfCiH1t42VsTphccSAOy78FIAQWrVrZHiNjWGFHKtx L76+CITfcaf9soG0GnCeCYVgZVE2sgqf437WvpRrmHuQ6P5Z8Yz3crNtK3gIqFqXdozV RFPkxyf1IB7H7G3Du9w/MqXU8MFCaY75OKvvkYcHIkodIQRGeu8S/zvSocaklLliqF/V KDAJaK8+tN1Ctbx+veKZJBpHN1g0lT8/N5A2MWfhfe4zKwsPO/tqiRLveoCnKGF5WHPC YCcRJ1UXGHHY1ZYF8AGgp6M7gPUBJ4lu5ylBT+3L0+B7XHgjodBwW/n6qzCrDIv9+SBd wHyg== 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=VzCY+C4+LMrFrtHF21V6NKxiRcHRk3TGsnlQrWw1o4w=; b=jC/8jeFIh4msD0BEt5xBC4zE/ccqpqVLtV12mSZAWxMemEx3FlIGbM4WIKMHweuyGg NsFbuzM2ezUO3M0hQgN2z8lqjSqv2bY2DnO5c6vArqSHx3qoASviaFVvevCprIp6ZpYh rO7N5RkFSJI4oayuhvCDPoxTGl+31QMakA3jIThDDWAaDxIGmjMhZWCO2MrFJiaROr+C n2KpEYIRu92SSxkNlNohrb0fQY7BHraDdHu22ruStzdf9C5u9BzOHI/jVXKkM4QaoFzU yFvQqRFKrYm0BX41Q3lAsfvbcnYuqrIHiJ2PqVJKpChH3CIpipD8txX9VJKNuCIwQw4b Ipnw== ARC-Authentication-Results: i=1; mx.google.com; dkim=neutral (body hash did not verify) header.i=@rothenpieler.org header.s=mail header.b=UCBUG2PJ; 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 z11-20020a05640235cb00b004377e09fb70si13252708edc.551.2022.08.10.13.48.46; Wed, 10 Aug 2022 13:48:46 -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=UCBUG2PJ; 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 2FF0E68B871; Wed, 10 Aug 2022 23:47:39 +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 1B14568B810 for ; Wed, 10 Aug 2022 23:47:31 +0300 (EEST) Received: from [authenticated] by btbn.de (Postfix) with ESMTPSA id 45AC92F260E; Wed, 10 Aug 2022 22:47:25 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=rothenpieler.org; s=mail; t=1660164445; bh=XwRiEcr4iHYNfeISTznrmlj6RGS4De+KGp7K6Gmz+h0=; h=From:To:Cc:Subject:Date:In-Reply-To:References; b=UCBUG2PJ/yUQIklsfkbQ0M4fcy8enHrODGtd8HPG9NJAHeGUGlgHD8nRS5mAJlkgS ZZtwn7s6YLWuuwI2Sgj5RD+qCf0WeHxRm8EaCav0hg8ndcZWnedUAVwOolawHAj20K QSHyCXnaP4SFo0YXvYcSvy93dCt8wdcaITEiucx6Jc7x5TzCyCRN0XtO+AQFmLmWqi sy9XxGugxlMEoBehOuJuJGil6dZXAcUSJ4kbl19C7Xdq/KDkVOlzbHnHrwsNAYC0HZ an4sGsnzoiuzXhdj5haF/fZAMNXgzzGEfRHhDUFtfHZk5RSM4BXEwir3qtXmRjwDIv EgNA/uVRRB0vQ== From: Timo Rothenpieler To: ffmpeg-devel@ffmpeg.org Date: Wed, 10 Aug 2022 22:47:09 +0200 Message-Id: <20220810204712.3123-8-timo@rothenpieler.org> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20220810204712.3123-1-timo@rothenpieler.org> References: <20220810204712.3123-1-timo@rothenpieler.org> MIME-Version: 1.0 Subject: [FFmpeg-devel] [PATCH 08/11] 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: +rWCTh6TxK17 --- 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 825354873d..a3582bfdd6 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 6ab9400b7c..77b1ce052b 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 70992531bf..50f55bb1b9 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..dba14cef5d --- /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(float2half_tables *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 9252560649..b8c9cdfc4f 100644 --- a/libavutil/float2half.h +++ b/libavutil/float2half.h @@ -26,41 +26,9 @@ typedef struct float2half_tables { uint8_t shifttable[512]; } float2half_tables; -static void init_float2half_tables(float2half_tables *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(float2half_tables *t); -static uint16_t float2half(uint32_t f, const float2half_tables *t) +static inline uint16_t float2half(uint32_t f, const float2half_tables *t) { uint16_t h; diff --git a/libavutil/half2float.c b/libavutil/half2float.c new file mode 100644 index 0000000000..baac8e4093 --- /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(half2float_tables *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 10b6fef4e6..cb58e44a1c 100644 --- a/libavutil/half2float.h +++ b/libavutil/half2float.h @@ -27,51 +27,9 @@ typedef struct half2float_tables { uint16_t offsettable[64]; } half2float_tables; -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(half2float_tables *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(half2float_tables *t); -static uint32_t half2float(uint16_t h, const half2float_tables *t) +static inline uint32_t half2float(uint16_t h, const half2float_tables *t) { uint32_t f;