From patchwork Mon Jul 3 19:45:03 2017 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Paul B Mahol X-Patchwork-Id: 4207 Delivered-To: ffmpegpatchwork@gmail.com Received: by 10.103.1.76 with SMTP id 73csp168929vsb; Mon, 3 Jul 2017 12:45:21 -0700 (PDT) X-Received: by 10.28.135.79 with SMTP id j76mr23095257wmd.27.1499111121372; Mon, 03 Jul 2017 12:45:21 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1499111121; cv=none; d=google.com; s=arc-20160816; b=cPFKUgkng+56BsyfOQBBiNflBnm8QO0oHMQn8ke/17rZkeVrHjX7qae9IKwbSA9U+I SrUCU7eBYz2pUp+3cDf6XUhYwbSDCtxv34KbJpN5oU+T+MlRLJ5xmYR+du0su5FmY890 /K+M/Bs0i+l+7Udp0vnpR11/m0Bq1YZng9Xs0NiJSD1sD7TH7/EgxtmczheUvSRYQWUU oJHN2vYNKwaIaIcTBGkWBMqWMI33Wz4QfKC6UIAI3f/ip+3YmmbdpGOx4j2KyRqTcH0I DO5Y7s05s885QPepA3B5uL+0flmRWxIrnHZlRMDV4fzKCEZxh1Zi3VBoOTtx/Ey0BIP6 M8hw== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=sender:errors-to:content-transfer-encoding:mime-version:reply-to :list-subscribe:list-help:list-post:list-archive:list-unsubscribe :list-id:precedence:subject:message-id:date:to:from:dkim-signature :delivered-to:arc-authentication-results; bh=+mSPErRLpdUqjZSv0I+ZSZ89RIiYKkOtgqlf7LiOyi0=; b=ruK0BZM2mawsjP2/1IhmlxIk/TswbS9kVr2PwlaJ8dMzXe6uB29LXLN5KH/3YsmbNl 54doX89vyl8Z67ulm6fe1NuI9AJ4wFU53owSPdDhwkiAkKa256keKBdpN7X7Jl/kvsnp Lo7dkJ6gf5Md8qa6KBxuC2pEbi0XD+q6Syz2fiVMZn9KpETGau7nwZaPWmIuNbJQo6y3 TokTDJDSnsLQWaULknfbBY1N6XDcfphWPXjjRMK1qXjLrF6zNYuBGFKE53VgPyzfZSLa yycbvAvL0fRVacSJj3bmYW8gv6UazhVkJnWPrcbQFmMTwSHgzjX1DbMTqbl4IOOFqFma ptfQ== ARC-Authentication-Results: i=1; mx.google.com; dkim=neutral (body hash did not verify) header.i=@gmail.com header.b=Z6Lkf5Ck; 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=gmail.com Return-Path: Received: from ffbox0-bg.mplayerhq.hu (ffbox0-bg.ffmpeg.org. [79.124.17.100]) by mx.google.com with ESMTP id e11si12895141wre.37.2017.07.03.12.45.20; Mon, 03 Jul 2017 12:45:21 -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=@gmail.com header.b=Z6Lkf5Ck; 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=gmail.com Received: from [127.0.1.1] (localhost [127.0.0.1]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTP id CEE4C6882FD; Mon, 3 Jul 2017 22:45:15 +0300 (EEST) X-Original-To: ffmpeg-devel@ffmpeg.org Delivered-To: ffmpeg-devel@ffmpeg.org Received: from mail-wr0-f179.google.com (mail-wr0-f179.google.com [209.85.128.179]) by ffbox0-bg.mplayerhq.hu (Postfix) with ESMTPS id CEE7F680794 for ; Mon, 3 Jul 2017 22:45:13 +0300 (EEST) Received: by mail-wr0-f179.google.com with SMTP id 77so239158175wrb.1 for ; Mon, 03 Jul 2017 12:45:15 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=from:to:subject:date:message-id; bh=U6c65nssoeFtmNezrpuRagTatT0lkenrtyyP2o5+K7o=; b=Z6Lkf5Ckll8WzcJSGUUr4xcXwxIHevhsVD/MRrTImpTfjf0QcGGWprloAjeMK1obCA bd1AGM0nr5ql42Dnj+dDCgntKwTMsKS8znUjBhoeJl0qU0/1UYMONfBkiInlE1q3MRgf Qjlb9soY/88h6OBVOHo2FwpmqMQ9zEIV/18MTZYnLWMZ7lL4XQAcWQsKjQEpI4K18T6O SKsnWVOC/vIAokq0EgwtiP8zLIs9DRlk03DadYKHvIwvOhjZeelOet8H8gjJck26CmLk 823ksQQLy4SYKCePiP1vX7Izw3GkhQt0+Z4EHJWhbTgfsTre6q5/opHM9VGaDltV+dP/ akeA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:subject:date:message-id; bh=U6c65nssoeFtmNezrpuRagTatT0lkenrtyyP2o5+K7o=; b=qGavkAgVhnk1iAqnd9tojU36xwycs2zRNpW3mrlaUA1mkkmN3YjF53qKvq8ULlTmML K1Ql0gOw4u4g3j82esnORWDgyXNvfyyRkvlBwBliK/Hztpi+4im5cVangJsLCx3WzT+V GKyHKJBxPzb9LTXB7M+EV5UPb4rTtFXULbG/UJar6m7HB4f7KN7kXQxXFzegv1953p5L 8Jea/E59SVS95G9Pi4RhzuEPN6PoQDjao7aKhaLdZT9GymoC+SI6Nri2An82W7tQCn5I VzWd/4g1M8NUzlDc9Gj4kkx2jG/EwOqAIR4KXffz++4liYVmO3W7gRz6FWolaHV35nsi VNpA== X-Gm-Message-State: AKS2vOyh3WE+Xl3FRcYkvIJ0vcIL6tACg9rMa8zeOmHQFkufkbdKVuZJ RazUAa5bqd+HEb1F X-Received: by 10.223.129.74 with SMTP id 68mr22243014wrm.130.1499111114396; Mon, 03 Jul 2017 12:45:14 -0700 (PDT) Received: from localhost.localdomain ([94.250.174.60]) by smtp.gmail.com with ESMTPSA id j125sm10630259wmf.6.2017.07.03.12.45.11 for (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Mon, 03 Jul 2017 12:45:13 -0700 (PDT) From: Paul B Mahol To: ffmpeg-devel@ffmpeg.org Date: Mon, 3 Jul 2017 21:45:03 +0200 Message-Id: <20170703194503.18168-1-onemda@gmail.com> X-Mailer: git-send-email 2.9.3 Subject: [FFmpeg-devel] [PATCH] avfilter: add limiter filter X-BeenThere: ffmpeg-devel@ffmpeg.org X-Mailman-Version: 2.1.20 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 MIME-Version: 1.0 Errors-To: ffmpeg-devel-bounces@ffmpeg.org Sender: "ffmpeg-devel" Signed-off-by: Paul B Mahol --- doc/filters.texi | 17 +++ libavfilter/Makefile | 1 + libavfilter/allfilters.c | 1 + libavfilter/limiter.h | 33 ++++++ libavfilter/vf_limiter.c | 228 ++++++++++++++++++++++++++++++++++++++ libavfilter/x86/Makefile | 2 + libavfilter/x86/vf_limiter.asm | 80 +++++++++++++ libavfilter/x86/vf_limiter_init.c | 41 +++++++ 8 files changed, 403 insertions(+) create mode 100644 libavfilter/limiter.h create mode 100644 libavfilter/vf_limiter.c create mode 100644 libavfilter/x86/vf_limiter.asm create mode 100644 libavfilter/x86/vf_limiter_init.c diff --git a/doc/filters.texi b/doc/filters.texi index 930ca4c..dbf0fc1 100644 --- a/doc/filters.texi +++ b/doc/filters.texi @@ -9639,6 +9639,23 @@ The formula that generates the correction is: where @var{r_0} is halve of the image diagonal and @var{r_src} and @var{r_tgt} are the distances from the focal point in the source and target images, respectively. +@section limiter + +Limits the pixel components values to the specified range [min, max]. + +The filter accepts the following options: + +@table @option +@item min +Lower bound. Defaults to the lowest allowed value for the input. + +@item max +Upper bound. Defaults to the highest allowed value for the input. + +@item planes +Specify which planes will be processed. +@end table + @section loop Loop video frames. diff --git a/libavfilter/Makefile b/libavfilter/Makefile index f023a0d..4d85f65 100644 --- a/libavfilter/Makefile +++ b/libavfilter/Makefile @@ -216,6 +216,7 @@ OBJS-$(CONFIG_INTERLACE_FILTER) += vf_interlace.o OBJS-$(CONFIG_INTERLEAVE_FILTER) += f_interleave.o OBJS-$(CONFIG_KERNDEINT_FILTER) += vf_kerndeint.o OBJS-$(CONFIG_LENSCORRECTION_FILTER) += vf_lenscorrection.o +OBJS-$(CONFIG_LIMITER_FILTER) += vf_limiter.o OBJS-$(CONFIG_LOOP_FILTER) += f_loop.o OBJS-$(CONFIG_LUMAKEY_FILTER) += vf_lumakey.o OBJS-$(CONFIG_LUT_FILTER) += vf_lut.o diff --git a/libavfilter/allfilters.c b/libavfilter/allfilters.c index c1c5233..0a990ca 100644 --- a/libavfilter/allfilters.c +++ b/libavfilter/allfilters.c @@ -228,6 +228,7 @@ static void register_all(void) REGISTER_FILTER(INTERLEAVE, interleave, vf); REGISTER_FILTER(KERNDEINT, kerndeint, vf); REGISTER_FILTER(LENSCORRECTION, lenscorrection, vf); + REGISTER_FILTER(LIMITER, limiter, vf); REGISTER_FILTER(LOOP, loop, vf); REGISTER_FILTER(LUMAKEY, lumakey, vf); REGISTER_FILTER(LUT, lut, vf); diff --git a/libavfilter/limiter.h b/libavfilter/limiter.h new file mode 100644 index 0000000..54d423d --- /dev/null +++ b/libavfilter/limiter.h @@ -0,0 +1,33 @@ +/* + * 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 + */ + +#ifndef AVFILTER_LIMITER_H +#define AVFILTER_LIMITER_H + +#include +#include + +typedef struct LimiterDSPContext { + void (*limiter)(const uint8_t *src, uint8_t *dst, + ptrdiff_t slinesize, ptrdiff_t dlinesize, + int w, int h, int min, int max); +} LimiterDSPContext; + +void ff_limiter_init_x86(LimiterDSPContext *dsp, int bpp); + +#endif /* AVFILTER_LIMITER_H */ diff --git a/libavfilter/vf_limiter.c b/libavfilter/vf_limiter.c new file mode 100644 index 0000000..d6f5745 --- /dev/null +++ b/libavfilter/vf_limiter.c @@ -0,0 +1,228 @@ +/* + * 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/attributes.h" +#include "libavutil/common.h" +#include "libavutil/eval.h" +#include "libavutil/imgutils.h" +#include "libavutil/opt.h" +#include "libavutil/pixdesc.h" +#include "avfilter.h" +#include "formats.h" +#include "internal.h" +#include "limiter.h" +#include "video.h" + +typedef struct LimiterContext { + const AVClass *class; + int min; + int max; + int planes; + int nb_planes; + int linesize[4]; + int width[4]; + int height[4]; + + LimiterDSPContext dsp; +} LimiterContext; + +#define OFFSET(x) offsetof(LimiterContext, x) +#define FLAGS AV_OPT_FLAG_FILTERING_PARAM|AV_OPT_FLAG_VIDEO_PARAM + +static const AVOption limiter_options[] = { + { "min", "set min value", OFFSET(min), AV_OPT_TYPE_INT, {.i64=0}, 0, 65535, .flags = FLAGS }, + { "max", "set max value", OFFSET(max), AV_OPT_TYPE_INT, {.i64=65535}, 0, 65535, .flags = FLAGS }, + { "planes", "set planes", OFFSET(planes), AV_OPT_TYPE_INT, {.i64=15}, 0, 65535, .flags = FLAGS }, + { NULL } +}; + +AVFILTER_DEFINE_CLASS(limiter); + +static av_cold int init(AVFilterContext *ctx) +{ + LimiterContext *s = ctx->priv; + + if (s->min > s->max) + return AVERROR(EINVAL); + return 0; +} + +static int query_formats(AVFilterContext *ctx) +{ + static const enum AVPixelFormat pix_fmts[] = { + AV_PIX_FMT_YUVA444P, AV_PIX_FMT_YUV444P, AV_PIX_FMT_YUV440P, + AV_PIX_FMT_YUVJ444P, AV_PIX_FMT_YUVJ440P, + AV_PIX_FMT_YUVA422P, AV_PIX_FMT_YUV422P, AV_PIX_FMT_YUVA420P, AV_PIX_FMT_YUV420P, + AV_PIX_FMT_YUVJ422P, AV_PIX_FMT_YUVJ420P, + AV_PIX_FMT_YUVJ411P, AV_PIX_FMT_YUV411P, AV_PIX_FMT_YUV410P, + AV_PIX_FMT_YUV420P9, AV_PIX_FMT_YUV422P9, AV_PIX_FMT_YUV444P9, + AV_PIX_FMT_YUV420P10, AV_PIX_FMT_YUV422P10, AV_PIX_FMT_YUV444P10, + AV_PIX_FMT_YUV420P12, AV_PIX_FMT_YUV422P12, AV_PIX_FMT_YUV444P12, AV_PIX_FMT_YUV440P12, + AV_PIX_FMT_YUV420P14, AV_PIX_FMT_YUV422P14, AV_PIX_FMT_YUV444P14, + AV_PIX_FMT_YUV420P16, AV_PIX_FMT_YUV422P16, AV_PIX_FMT_YUV444P16, + AV_PIX_FMT_YUVA420P9, AV_PIX_FMT_YUVA422P9, AV_PIX_FMT_YUVA444P9, + AV_PIX_FMT_YUVA420P10, AV_PIX_FMT_YUVA422P10, AV_PIX_FMT_YUVA444P10, + AV_PIX_FMT_YUVA420P16, AV_PIX_FMT_YUVA422P16, AV_PIX_FMT_YUVA444P16, + AV_PIX_FMT_GBRP, AV_PIX_FMT_GBRP9, AV_PIX_FMT_GBRP10, + AV_PIX_FMT_GBRP12, AV_PIX_FMT_GBRP14, AV_PIX_FMT_GBRP16, + AV_PIX_FMT_GBRAP, AV_PIX_FMT_GBRAP12, AV_PIX_FMT_GBRAP16, + AV_PIX_FMT_GRAY8, AV_PIX_FMT_GRAY10, AV_PIX_FMT_GRAY12, AV_PIX_FMT_GRAY16, + AV_PIX_FMT_NONE + }; + + AVFilterFormats *fmts_list = ff_make_format_list(pix_fmts); + if (!fmts_list) + return AVERROR(ENOMEM); + return ff_set_common_formats(ctx, fmts_list); +} + +static void limiter8(const uint8_t *src, uint8_t *dst, + ptrdiff_t slinesize, ptrdiff_t dlinesize, + int w, int h, int min, int max) +{ + int x, y; + + for (y = 0; y < h; y++) { + for (x = 0; x < w; x++) { + dst[x] = av_clip(src[x], min, max); + } + + dst += dlinesize; + src += slinesize; + } +} + +static void limiter16(const uint8_t *ssrc, uint8_t *ddst, + ptrdiff_t slinesize, ptrdiff_t dlinesize, + int w, int h, int min, int max) +{ + const uint16_t *src = (const uint16_t *)ssrc; + uint16_t *dst = (uint16_t *)ddst; + int x, y; + + dlinesize /= 2; + slinesize /= 2; + + for (y = 0; y < h; y++) { + for (x = 0; x < w; x++) { + dst[x] = av_clip(src[x], min, max); + } + + dst += dlinesize; + src += slinesize; + } +} + +static int config_props(AVFilterLink *inlink) +{ + AVFilterContext *ctx = inlink->dst; + LimiterContext *s = ctx->priv; + const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(inlink->format); + int vsub, hsub, ret; + + s->nb_planes = av_pix_fmt_count_planes(inlink->format); + + if ((ret = av_image_fill_linesizes(s->linesize, inlink->format, inlink->w)) < 0) + return ret; + + hsub = desc->log2_chroma_w; + vsub = desc->log2_chroma_h; + s->height[1] = s->height[2] = AV_CEIL_RSHIFT(inlink->h, vsub); + s->height[0] = s->height[3] = inlink->h; + s->width[1] = s->width[2] = AV_CEIL_RSHIFT(inlink->w, hsub); + s->width[0] = s->width[3] = inlink->w; + + if (desc->comp[0].depth == 8) + s->dsp.limiter = limiter8; + else + s->dsp.limiter = limiter16; + + if (ARCH_X86) + ff_limiter_init_x86(&s->dsp, desc->comp[0].depth); + + return 0; +} + +static int filter_frame(AVFilterLink *inlink, AVFrame *in) +{ + AVFilterContext *ctx = inlink->dst; + LimiterContext *s = ctx->priv; + AVFilterLink *outlink = ctx->outputs[0]; + AVFrame *out; + int p; + + if (av_frame_is_writable(in)) { + out = in; + } else { + out = ff_get_video_buffer(outlink, outlink->w, outlink->h); + if (!out) { + av_frame_free(&in); + return AVERROR(ENOMEM); + } + av_frame_copy_props(out, in); + } + + for (p = 0; p < s->nb_planes; p++) { + if (!((1 << p) & s->planes)) { + if (out != in) + av_image_copy_plane(out->data[p], out->linesize[p], in->data[p], in->linesize[p], + s->linesize[p], s->height[p]); + continue; + } + + s->dsp.limiter(in->data[p], out->data[p], + in->linesize[p], out->linesize[p], + s->width[p], s->height[p], + s->min, s->max); + } + + if (out != in) + av_frame_free(&in); + + return ff_filter_frame(outlink, out); +} + +static const AVFilterPad inputs[] = { + { + .name = "default", + .type = AVMEDIA_TYPE_VIDEO, + .filter_frame = filter_frame, + .config_props = config_props, + }, + { NULL } +}; + +static const AVFilterPad outputs[] = { + { + .name = "default", + .type = AVMEDIA_TYPE_VIDEO, + }, + { NULL } +}; + +AVFilter ff_vf_limiter = { + .name = "limiter", + .description = NULL_IF_CONFIG_SMALL("Limit pixels components to the specified range."), + .priv_size = sizeof(LimiterContext), + .priv_class = &limiter_class, + .init = init, + .query_formats = query_formats, + .inputs = inputs, + .outputs = outputs, + .flags = AVFILTER_FLAG_SUPPORT_TIMELINE_GENERIC, +}; diff --git a/libavfilter/x86/Makefile b/libavfilter/x86/Makefile index f598ceb..3431625 100644 --- a/libavfilter/x86/Makefile +++ b/libavfilter/x86/Makefile @@ -8,6 +8,7 @@ OBJS-$(CONFIG_GRADFUN_FILTER) += x86/vf_gradfun_init.o OBJS-$(CONFIG_HQDN3D_FILTER) += x86/vf_hqdn3d_init.o OBJS-$(CONFIG_IDET_FILTER) += x86/vf_idet_init.o OBJS-$(CONFIG_INTERLACE_FILTER) += x86/vf_interlace_init.o +OBJS-$(CONFIG_LIMITER_FILTER) += x86/vf_limiter_init.o OBJS-$(CONFIG_MASKEDMERGE_FILTER) += x86/vf_maskedmerge_init.o OBJS-$(CONFIG_NOISE_FILTER) += x86/vf_noise.o OBJS-$(CONFIG_PP7_FILTER) += x86/vf_pp7_init.o @@ -33,6 +34,7 @@ X86ASM-OBJS-$(CONFIG_GRADFUN_FILTER) += x86/vf_gradfun.o X86ASM-OBJS-$(CONFIG_HQDN3D_FILTER) += x86/vf_hqdn3d.o X86ASM-OBJS-$(CONFIG_IDET_FILTER) += x86/vf_idet.o X86ASM-OBJS-$(CONFIG_INTERLACE_FILTER) += x86/vf_interlace.o +X86ASM-OBJS-$(CONFIG_LIMITER_FILTER) += x86/vf_limiter.o X86ASM-OBJS-$(CONFIG_MASKEDMERGE_FILTER) += x86/vf_maskedmerge.o X86ASM-OBJS-$(CONFIG_PP7_FILTER) += x86/vf_pp7.o X86ASM-OBJS-$(CONFIG_PSNR_FILTER) += x86/vf_psnr.o diff --git a/libavfilter/x86/vf_limiter.asm b/libavfilter/x86/vf_limiter.asm new file mode 100644 index 0000000..7d61c26 --- /dev/null +++ b/libavfilter/x86/vf_limiter.asm @@ -0,0 +1,80 @@ +;***************************************************************************** +;* x86-optimized functions for limiter filter +;* +;* 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/x86/x86util.asm" + +SECTION_RODATA + +pb_0: times 16 db 0 + +SECTION .text + +INIT_XMM sse2 + +cglobal limiter_8bit, 8, 9, 3, src, dst, slinesize, dlinesize, w, h, min, max, x + movsxdifnidn wq, wd + add srcq, wq + add dstq, wq + neg wq + SPLATB_REG m1, min, [pb_0] + SPLATB_REG m2, max, [pb_0] +.nextrow: + mov xq, wq + + .loop: + movu m0, [srcq + xq] + CLIPUB m0, m1, m2 + mova [dstq+xq], m0 + add xq, mmsize + jl .loop + + add srcq, slinesizeq + add dstq, dlinesizeq + sub hd, 1 + jg .nextrow + ret + +cglobal limiter_16bit, 8, 9, 3, src, dst, slinesize, dlinesize, w, h, min, max, x + shl wd, 1 + add srcq, wq + add dstq, wq + neg wq + pinsrw m1, minw, 0 + SPLATW m1, m1 + pinsrw m2, maxw, 0 + SPLATW m2, m2 +.nextrow: + mov xq, wq + + .loop: + movu m0, [srcq + xq] + pmaxuw m0, m1 + pminuw m0, m2 + mova [dstq+xq], m0 + add xq, mmsize + jl .loop + + add srcq, slinesizeq + add dstq, dlinesizeq + sub hd, 1 + jg .nextrow + ret + + ret diff --git a/libavfilter/x86/vf_limiter_init.c b/libavfilter/x86/vf_limiter_init.c new file mode 100644 index 0000000..348bb23 --- /dev/null +++ b/libavfilter/x86/vf_limiter_init.c @@ -0,0 +1,41 @@ +/* + * 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/x86/cpu.h" + +#include "libavfilter/limiter.h" + +void ff_limiter_8bit_sse2(const uint8_t *src, uint8_t *dst, + ptrdiff_t slinesize, ptrdiff_t dlinesize, + int w, int h, int min, int max); +void ff_limiter_16bit_sse2(const uint8_t *src, uint8_t *dst, + ptrdiff_t slinesize, ptrdiff_t dlinesize, + int w, int h, int min, int max); + +void ff_limiter_init_x86(LimiterDSPContext *dsp, int bpp) +{ + int cpu_flags = av_get_cpu_flags(); + + if (EXTERNAL_SSE2(cpu_flags)) { + if (bpp <= 8) { + dsp->limiter = ff_limiter_8bit_sse2; + } else if (bpp <= 16) { + dsp->limiter = ff_limiter_16bit_sse2; + } + } +}