1
0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2024-11-26 19:01:44 +02:00
FFmpeg/libswscale/sparc/yuv2rgb_vis.c
Michael Niedermayer ac627b3d38 Merge commit '716d413c13981da15323c7a3821860536eefdbbb'
* commit '716d413c13981da15323c7a3821860536eefdbbb':
  Replace PIX_FMT_* -> AV_PIX_FMT_*, PixelFormat -> AVPixelFormat

Conflicts:
	doc/examples/muxing.c
	ffmpeg.h
	ffmpeg_filter.c
	ffmpeg_opt.c
	ffplay.c
	ffprobe.c
	libavcodec/8bps.c
	libavcodec/aasc.c
	libavcodec/aura.c
	libavcodec/avcodec.h
	libavcodec/avs.c
	libavcodec/bfi.c
	libavcodec/bmp.c
	libavcodec/bmpenc.c
	libavcodec/c93.c
	libavcodec/cscd.c
	libavcodec/cyuv.c
	libavcodec/dpx.c
	libavcodec/dpxenc.c
	libavcodec/eatgv.c
	libavcodec/escape124.c
	libavcodec/ffv1.c
	libavcodec/flashsv.c
	libavcodec/fraps.c
	libavcodec/h264.c
	libavcodec/huffyuv.c
	libavcodec/iff.c
	libavcodec/imgconvert.c
	libavcodec/indeo3.c
	libavcodec/kmvc.c
	libavcodec/libopenjpegdec.c
	libavcodec/libopenjpegenc.c
	libavcodec/libx264.c
	libavcodec/ljpegenc.c
	libavcodec/mjpegdec.c
	libavcodec/mjpegenc.c
	libavcodec/motionpixels.c
	libavcodec/mpeg12.c
	libavcodec/mpeg12enc.c
	libavcodec/mpeg4videodec.c
	libavcodec/mpegvideo_enc.c
	libavcodec/pamenc.c
	libavcodec/pcxenc.c
	libavcodec/pgssubdec.c
	libavcodec/pngdec.c
	libavcodec/pngenc.c
	libavcodec/pnm.c
	libavcodec/pnmdec.c
	libavcodec/pnmenc.c
	libavcodec/ptx.c
	libavcodec/qdrw.c
	libavcodec/qpeg.c
	libavcodec/qtrleenc.c
	libavcodec/raw.c
	libavcodec/rawdec.c
	libavcodec/rl2.c
	libavcodec/sgidec.c
	libavcodec/sgienc.c
	libavcodec/snowdec.c
	libavcodec/snowenc.c
	libavcodec/sunrast.c
	libavcodec/targa.c
	libavcodec/targaenc.c
	libavcodec/tiff.c
	libavcodec/tiffenc.c
	libavcodec/tmv.c
	libavcodec/truemotion2.c
	libavcodec/utils.c
	libavcodec/vb.c
	libavcodec/vp3.c
	libavcodec/wnv1.c
	libavcodec/xl.c
	libavcodec/xwddec.c
	libavcodec/xwdenc.c
	libavcodec/yop.c
	libavdevice/v4l2.c
	libavdevice/x11grab.c
	libavfilter/avfilter.c
	libavfilter/avfilter.h
	libavfilter/buffersrc.c
	libavfilter/drawutils.c
	libavfilter/formats.c
	libavfilter/src_movie.c
	libavfilter/vf_ass.c
	libavfilter/vf_drawtext.c
	libavfilter/vf_fade.c
	libavfilter/vf_format.c
	libavfilter/vf_hflip.c
	libavfilter/vf_lut.c
	libavfilter/vf_overlay.c
	libavfilter/vf_pad.c
	libavfilter/vf_scale.c
	libavfilter/vf_transpose.c
	libavfilter/vf_yadif.c
	libavfilter/video.c
	libavfilter/vsrc_testsrc.c
	libavformat/movenc.c
	libavformat/mxf.h
	libavformat/utils.c
	libavformat/yuv4mpeg.c
	libavutil/imgutils.c
	libavutil/pixdesc.c
	libswscale/input.c
	libswscale/output.c
	libswscale/swscale_internal.h
	libswscale/swscale_unscaled.c
	libswscale/utils.c
	libswscale/x86/swscale_template.c
	libswscale/x86/yuv2rgb.c
	libswscale/x86/yuv2rgb_template.c
	libswscale/yuv2rgb.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2012-10-08 21:06:57 +02:00

213 lines
9.5 KiB
C

/*
* VIS optimized software YUV to RGB converter
* Copyright (c) 2007 Denes Balatoni <dbalatoni@programozo.hu>
*
* 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 <inttypes.h>
#include <stdlib.h>
#include "libavutil/attributes.h"
#include "libswscale/swscale.h"
#include "libswscale/swscale_internal.h"
#define YUV2RGB_INIT \
"wr %%g0, 0x10, %%gsr \n\t" \
"ldd [%5], %%f32 \n\t" \
"ldd [%5 + 8], %%f34 \n\t" \
"ldd [%5 + 16], %%f36 \n\t" \
"ldd [%5 + 24], %%f38 \n\t" \
"ldd [%5 + 32], %%f40 \n\t" \
"ldd [%5 + 40], %%f42 \n\t" \
"ldd [%5 + 48], %%f44 \n\t" \
"ldd [%5 + 56], %%f46 \n\t" \
"ldd [%5 + 64], %%f48 \n\t" \
"ldd [%5 + 72], %%f50 \n\t"
#define YUV2RGB_KERNEL \
/* ^^^^ f0=Y f3=u f5=v */ \
"fmul8x16 %%f3, %%f48, %%f6 \n\t" \
"fmul8x16 %%f19, %%f48, %%f22 \n\t" \
"fmul8x16 %%f5, %%f44, %%f8 \n\t" \
"fmul8x16 %%f21, %%f44, %%f24 \n\t" \
"fmul8x16 %%f0, %%f42, %%f0 \n\t" \
"fmul8x16 %%f16, %%f42, %%f16 \n\t" \
"fmul8x16 %%f3, %%f50, %%f2 \n\t" \
"fmul8x16 %%f19, %%f50, %%f18 \n\t" \
"fmul8x16 %%f5, %%f46, %%f4 \n\t" \
"fmul8x16 %%f21, %%f46, %%f20 \n\t" \
\
"fpsub16 %%f6, %%f34, %%f6 \n\t" /* 1 */ \
"fpsub16 %%f22, %%f34, %%f22 \n\t" /* 1 */ \
"fpsub16 %%f8, %%f38, %%f8 \n\t" /* 3 */ \
"fpsub16 %%f24, %%f38, %%f24 \n\t" /* 3 */ \
"fpsub16 %%f0, %%f32, %%f0 \n\t" /* 0 */ \
"fpsub16 %%f16, %%f32, %%f16 \n\t" /* 0 */ \
"fpsub16 %%f2, %%f36, %%f2 \n\t" /* 2 */ \
"fpsub16 %%f18, %%f36, %%f18 \n\t" /* 2 */ \
"fpsub16 %%f4, %%f40, %%f4 \n\t" /* 4 */ \
"fpsub16 %%f20, %%f40, %%f20 \n\t" /* 4 */ \
\
"fpadd16 %%f0, %%f8, %%f8 \n\t" /* Gt */ \
"fpadd16 %%f16, %%f24, %%f24 \n\t" /* Gt */ \
"fpadd16 %%f0, %%f4, %%f4 \n\t" /* R */ \
"fpadd16 %%f16, %%f20, %%f20 \n\t" /* R */ \
"fpadd16 %%f0, %%f6, %%f6 \n\t" /* B */ \
"fpadd16 %%f16, %%f22, %%f22 \n\t" /* B */ \
"fpadd16 %%f8, %%f2, %%f2 \n\t" /* G */ \
"fpadd16 %%f24, %%f18, %%f18 \n\t" /* G */ \
\
"fpack16 %%f4, %%f4 \n\t" \
"fpack16 %%f20, %%f20 \n\t" \
"fpack16 %%f6, %%f6 \n\t" \
"fpack16 %%f22, %%f22 \n\t" \
"fpack16 %%f2, %%f2 \n\t" \
"fpack16 %%f18, %%f18 \n\t"
// FIXME: must be changed to set alpha to 255 instead of 0
static int vis_420P_ARGB32(SwsContext *c, uint8_t *src[], int srcStride[],
int srcSliceY, int srcSliceH,
uint8_t *dst[], int dstStride[])
{
int y, out1, out2, out3, out4, out5, out6;
for (y = 0; y < srcSliceH; ++y)
__asm__ volatile (
YUV2RGB_INIT
"wr %%g0, 0xd2, %%asi \n\t" /* ASI_FL16_P */
"1: \n\t"
"ldda [%1] %%asi, %%f2 \n\t"
"ldda [%1 + 2] %%asi, %%f18 \n\t"
"ldda [%2] %%asi, %%f4 \n\t"
"ldda [%2 + 2] %%asi, %%f20 \n\t"
"ld [%0], %%f0 \n\t"
"ld [%0+4], %%f16 \n\t"
"fpmerge %%f3, %%f3, %%f2 \n\t"
"fpmerge %%f19, %%f19, %%f18 \n\t"
"fpmerge %%f5, %%f5, %%f4 \n\t"
"fpmerge %%f21, %%f21, %%f20 \n\t"
YUV2RGB_KERNEL
"fzero %%f0 \n\t"
"fpmerge %%f4, %%f6, %%f8 \n\t" // r, b, t1
"fpmerge %%f20, %%f22, %%f24 \n\t" // r, b, t1
"fpmerge %%f0, %%f2, %%f10 \n\t" // 0, g, t2
"fpmerge %%f0, %%f18, %%f26 \n\t" // 0, g, t2
"fpmerge %%f10, %%f8, %%f4 \n\t" // t2, t1, msb
"fpmerge %%f26, %%f24, %%f20 \n\t" // t2, t1, msb
"fpmerge %%f11, %%f9, %%f6 \n\t" // t2, t1, lsb
"fpmerge %%f27, %%f25, %%f22 \n\t" // t2, t1, lsb
"std %%f4, [%3] \n\t"
"std %%f20, [%3 + 16] \n\t"
"std %%f6, [%3 + 8] \n\t"
"std %%f22, [%3 + 24] \n\t"
"add %0, 8, %0 \n\t"
"add %1, 4, %1 \n\t"
"add %2, 4, %2 \n\t"
"subcc %4, 8, %4 \n\t"
"bne 1b \n\t"
"add %3, 32, %3 \n\t" // delay slot
: "=r" (out1), "=r" (out2), "=r" (out3), "=r" (out4), "=r" (out5), "=r" (out6)
: "0" (src[0] + (y + srcSliceY) * srcStride[0]), "1" (src[1] + ((y + srcSliceY) >> 1) * srcStride[1]),
"2" (src[2] + ((y + srcSliceY) >> 1) * srcStride[2]), "3" (dst[0] + (y + srcSliceY) * dstStride[0]),
"4" (c->dstW),
"5" (c->sparc_coeffs)
);
return srcSliceH;
}
// FIXME: must be changed to set alpha to 255 instead of 0
static int vis_422P_ARGB32(SwsContext *c, uint8_t *src[], int srcStride[],
int srcSliceY, int srcSliceH,
uint8_t *dst[], int dstStride[])
{
int y, out1, out2, out3, out4, out5, out6;
for (y = 0; y < srcSliceH; ++y)
__asm__ volatile (
YUV2RGB_INIT
"wr %%g0, 0xd2, %%asi \n\t" /* ASI_FL16_P */
"1: \n\t"
"ldda [%1] %%asi, %%f2 \n\t"
"ldda [%1 + 2] %%asi, %%f18 \n\t"
"ldda [%2] %%asi, %%f4 \n\t"
"ldda [%2 + 2] %%asi, %%f20 \n\t"
"ld [%0], %%f0 \n\t"
"ld [%0 + 4], %%f16 \n\t"
"fpmerge %%f3, %%f3, %%f2 \n\t"
"fpmerge %%f19, %%f19, %%f18 \n\t"
"fpmerge %%f5, %%f5, %%f4 \n\t"
"fpmerge %%f21, %%f21, %%f20 \n\t"
YUV2RGB_KERNEL
"fzero %%f0 \n\t"
"fpmerge %%f4, %%f6, %%f8 \n\t" // r,b,t1
"fpmerge %%f20, %%f22, %%f24 \n\t" // r,b,t1
"fpmerge %%f0, %%f2, %%f10 \n\t" // 0,g,t2
"fpmerge %%f0, %%f18, %%f26 \n\t" // 0,g,t2
"fpmerge %%f10, %%f8, %%f4 \n\t" // t2,t1,msb
"fpmerge %%f26, %%f24, %%f20 \n\t" // t2,t1,msb
"fpmerge %%f11, %%f9, %%f6 \n\t" // t2,t1,lsb
"fpmerge %%f27, %%f25, %%f22 \n\t" // t2,t1,lsb
"std %%f4, [%3] \n\t"
"std %%f20, [%3 + 16] \n\t"
"std %%f6, [%3 + 8] \n\t"
"std %%f22, [%3 + 24] \n\t"
"add %0, 8, %0 \n\t"
"add %1, 4, %1 \n\t"
"add %2, 4, %2 \n\t"
"subcc %4, 8, %4 \n\t"
"bne 1b \n\t"
"add %3, 32, %3 \n\t" //delay slot
: "=r" (out1), "=r" (out2), "=r" (out3), "=r" (out4), "=r" (out5), "=r" (out6)
: "0" (src[0] + (y + srcSliceY) * srcStride[0]), "1" (src[1] + (y + srcSliceY) * srcStride[1]),
"2" (src[2] + (y + srcSliceY) * srcStride[2]), "3" (dst[0] + (y + srcSliceY) * dstStride[0]),
"4" (c->dstW),
"5" (c->sparc_coeffs)
);
return srcSliceH;
}
av_cold SwsFunc ff_yuv2rgb_init_vis(SwsContext *c)
{
c->sparc_coeffs[5] = c->yCoeff;
c->sparc_coeffs[6] = c->vgCoeff;
c->sparc_coeffs[7] = c->vrCoeff;
c->sparc_coeffs[8] = c->ubCoeff;
c->sparc_coeffs[9] = c->ugCoeff;
c->sparc_coeffs[0] = (((int16_t)c->yOffset * (int16_t)c->yCoeff >> 11) & 0xffff) * 0x0001000100010001ULL;
c->sparc_coeffs[1] = (((int16_t)c->uOffset * (int16_t)c->ubCoeff >> 11) & 0xffff) * 0x0001000100010001ULL;
c->sparc_coeffs[2] = (((int16_t)c->uOffset * (int16_t)c->ugCoeff >> 11) & 0xffff) * 0x0001000100010001ULL;
c->sparc_coeffs[3] = (((int16_t)c->vOffset * (int16_t)c->vgCoeff >> 11) & 0xffff) * 0x0001000100010001ULL;
c->sparc_coeffs[4] = (((int16_t)c->vOffset * (int16_t)c->vrCoeff >> 11) & 0xffff) * 0x0001000100010001ULL;
if (c->dstFormat == AV_PIX_FMT_RGB32 && c->srcFormat == AV_PIX_FMT_YUV422P && (c->dstW & 7) == 0) {
av_log(c, AV_LOG_INFO,
"SPARC VIS accelerated YUV422P -> RGB32 (WARNING: alpha value is wrong)\n");
return vis_422P_ARGB32;
} else if (c->dstFormat == AV_PIX_FMT_RGB32 && c->srcFormat == AV_PIX_FMT_YUV420P && (c->dstW & 7) == 0) {
av_log(c, AV_LOG_INFO,
"SPARC VIS accelerated YUV420P -> RGB32 (WARNING: alpha value is wrong)\n");
return vis_420P_ARGB32;
}
return NULL;
}