virtualenv/FFmpeg
1
0
Fork 0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2024-12-23 12:43:46 +02:00
Code Issues Releases Activity
65212e3ed9
FFmpeg/libavcodec/x86/dsputil_mmx.h
Michael Niedermayer 015903294c Merge remote-tracking branch 'qatar/master' 
* qatar/master: (25 commits)
  rv40dsp x86: MMX/MMX2/3DNow/SSE2/SSSE3 implementations of MC
  ape: Use unsigned integer maths
  arm: dsputil: fix overreads in put/avg_pixels functions
  h264: K&R formatting cosmetics for header files (part II/II)
  h264: K&R formatting cosmetics for header files (part I/II)
  rtmp: Implement check bandwidth notification.
  rtmp: Support 'rtmp_swfurl', an option which specifies the URL of the SWF player.
  rtmp: Support 'rtmp_flashver', an option which overrides the version of the Flash plugin.
  rtmp: Support 'rtmp_tcurl', an option which overrides the URL of the target stream.
  cmdutils: Add fallback case to switch in check_stream_specifier().
  sctp: be consistent with socket option level
  configure: Add _XOPEN_SOURCE=600 to Solaris preprocessor flags.
  vcr1enc: drop pointless empty encode_init() wrapper function
  vcr1: drop pointless write-only AVCodecContext member from VCR1Context
  vcr1: group encoder code together to save #ifdefs
  vcr1: cosmetics: K&R prettyprinting, typos, parentheses, dead code, comments
  mov: make one comment slightly more specific
  lavr: replace the SSE version of ff_conv_fltp_to_flt_6ch() with SSE4 and AVX
  lavfi: move audio-related functions to a separate file.
  lavfi: remove some audio-related function from public API.
  ...

Conflicts:
	cmdutils.c
	libavcodec/h264.h
	libavcodec/h264_mvpred.h
	libavcodec/vcr1.c
	libavfilter/avfilter.c
	libavfilter/avfilter.h
	libavfilter/defaults.c
	libavfilter/internal.h

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2012-05-10 23:30:42 +02:00

224 lines
8.2 KiB
C
Raw Blame History

/*
* MMX optimized DSP utils
* Copyright (c) 2007 Aurelien Jacobs <aurel@gnuage.org>
*
* 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 AVCODEC_X86_DSPUTIL_MMX_H
#define AVCODEC_X86_DSPUTIL_MMX_H
#include <stdint.h>
#include "libavcodec/dsputil.h"
#include "libavutil/x86_cpu.h"
typedef struct { uint64_t a, b; } xmm_reg;
extern const uint64_t ff_bone;
extern const uint64_t ff_wtwo;
extern const uint64_t ff_pdw_80000000[2];
extern const xmm_reg ff_pw_3;
extern const xmm_reg ff_pw_4;
extern const xmm_reg ff_pw_5;
extern const xmm_reg ff_pw_8;
extern const uint64_t ff_pw_15;
extern const xmm_reg ff_pw_16;
extern const xmm_reg ff_pw_18;
extern const uint64_t ff_pw_20;
extern const xmm_reg ff_pw_27;
extern const xmm_reg ff_pw_28;
extern const xmm_reg ff_pw_32;
extern const uint64_t ff_pw_42;
extern const uint64_t ff_pw_53;
extern const xmm_reg ff_pw_63;
extern const xmm_reg ff_pw_64;
extern const uint64_t ff_pw_96;
extern const uint64_t ff_pw_128;
extern const uint64_t ff_pw_255;
extern const xmm_reg ff_pb_1;
extern const xmm_reg ff_pb_3;
extern const uint64_t ff_pb_7;
extern const uint64_t ff_pb_1F;
extern const uint64_t ff_pb_3F;
extern const uint64_t ff_pb_81;
extern const xmm_reg ff_pb_A1;
extern const xmm_reg ff_pb_F8;
extern const uint64_t ff_pb_FC;
extern const xmm_reg ff_pb_FE;
extern const double ff_pd_1[2];
extern const double ff_pd_2[2];
#define LOAD4(stride,in,a,b,c,d)\
"movq 0*"#stride"+"#in", "#a"\n\t"\
"movq 1*"#stride"+"#in", "#b"\n\t"\
"movq 2*"#stride"+"#in", "#c"\n\t"\
"movq 3*"#stride"+"#in", "#d"\n\t"
#define STORE4(stride,out,a,b,c,d)\
"movq "#a", 0*"#stride"+"#out"\n\t"\
"movq "#b", 1*"#stride"+"#out"\n\t"\
"movq "#c", 2*"#stride"+"#out"\n\t"\
"movq "#d", 3*"#stride"+"#out"\n\t"
/* in/out: mma=mma+mmb, mmb=mmb-mma */
#define SUMSUB_BA( a, b ) \
"paddw "#b", "#a" \n\t"\
"paddw "#b", "#b" \n\t"\
"psubw "#a", "#b" \n\t"
#define SBUTTERFLY(a,b,t,n,m)\
"mov" #m " " #a ", " #t " \n\t" /* abcd */\
"punpckl" #n " " #b ", " #a " \n\t" /* aebf */\
"punpckh" #n " " #b ", " #t " \n\t" /* cgdh */\
#define TRANSPOSE4(a,b,c,d,t)\
SBUTTERFLY(a,b,t,wd,q) /* a=aebf t=cgdh */\
SBUTTERFLY(c,d,b,wd,q) /* c=imjn b=kolp */\
SBUTTERFLY(a,c,d,dq,q) /* a=aeim d=bfjn */\
SBUTTERFLY(t,b,c,dq,q) /* t=cgko c=dhlp */
static inline void transpose4x4(uint8_t *dst, uint8_t *src, x86_reg dst_stride, x86_reg src_stride){
__asm__ volatile( //FIXME could save 1 instruction if done as 8x4 ...
"movd (%1), %%mm0 \n\t"
"add %3, %1 \n\t"
"movd (%1), %%mm1 \n\t"
"movd (%1,%3,1), %%mm2 \n\t"
"movd (%1,%3,2), %%mm3 \n\t"
"punpcklbw %%mm1, %%mm0 \n\t"
"punpcklbw %%mm3, %%mm2 \n\t"
"movq %%mm0, %%mm1 \n\t"
"punpcklwd %%mm2, %%mm0 \n\t"
"punpckhwd %%mm2, %%mm1 \n\t"
"movd %%mm0, (%0) \n\t"
"add %2, %0 \n\t"
"punpckhdq %%mm0, %%mm0 \n\t"
"movd %%mm0, (%0) \n\t"
"movd %%mm1, (%0,%2,1) \n\t"
"punpckhdq %%mm1, %%mm1 \n\t"
"movd %%mm1, (%0,%2,2) \n\t"
: "+&r" (dst),
"+&r" (src)
: "r" (dst_stride),
"r" (src_stride)
: "memory"
);
}
// e,f,g,h can be memory
// out: a,d,t,c
#define TRANSPOSE8x4(a,b,c,d,e,f,g,h,t)\
"punpcklbw " #e ", " #a " \n\t" /* a0 e0 a1 e1 a2 e2 a3 e3 */\
"punpcklbw " #f ", " #b " \n\t" /* b0 f0 b1 f1 b2 f2 b3 f3 */\
"punpcklbw " #g ", " #c " \n\t" /* c0 g0 c1 g1 c2 g2 d3 g3 */\
"punpcklbw " #h ", " #d " \n\t" /* d0 h0 d1 h1 d2 h2 d3 h3 */\
SBUTTERFLY(a, b, t, bw, q) /* a= a0 b0 e0 f0 a1 b1 e1 f1 */\
/* t= a2 b2 e2 f2 a3 b3 e3 f3 */\
SBUTTERFLY(c, d, b, bw, q) /* c= c0 d0 g0 h0 c1 d1 g1 h1 */\
/* b= c2 d2 g2 h2 c3 d3 g3 h3 */\
SBUTTERFLY(a, c, d, wd, q) /* a= a0 b0 c0 d0 e0 f0 g0 h0 */\
/* d= a1 b1 c1 d1 e1 f1 g1 h1 */\
SBUTTERFLY(t, b, c, wd, q) /* t= a2 b2 c2 d2 e2 f2 g2 h2 */\
/* c= a3 b3 c3 d3 e3 f3 g3 h3 */
#if ARCH_X86_64
// permutes 01234567 -> 05736421
#define TRANSPOSE8(a,b,c,d,e,f,g,h,t)\
SBUTTERFLY(a,b,%%xmm8,wd,dqa)\
SBUTTERFLY(c,d,b,wd,dqa)\
SBUTTERFLY(e,f,d,wd,dqa)\
SBUTTERFLY(g,h,f,wd,dqa)\
SBUTTERFLY(a,c,h,dq,dqa)\
SBUTTERFLY(%%xmm8,b,c,dq,dqa)\
SBUTTERFLY(e,g,b,dq,dqa)\
SBUTTERFLY(d,f,g,dq,dqa)\
SBUTTERFLY(a,e,f,qdq,dqa)\
SBUTTERFLY(%%xmm8,d,e,qdq,dqa)\
SBUTTERFLY(h,b,d,qdq,dqa)\
SBUTTERFLY(c,g,b,qdq,dqa)\
"movdqa %%xmm8, "#g" \n\t"
#else
#define TRANSPOSE8(a,b,c,d,e,f,g,h,t)\
"movdqa "#h", "#t" \n\t"\
SBUTTERFLY(a,b,h,wd,dqa)\
"movdqa "#h", 16"#t" \n\t"\
"movdqa "#t", "#h" \n\t"\
SBUTTERFLY(c,d,b,wd,dqa)\
SBUTTERFLY(e,f,d,wd,dqa)\
SBUTTERFLY(g,h,f,wd,dqa)\
SBUTTERFLY(a,c,h,dq,dqa)\
"movdqa "#h", "#t" \n\t"\
"movdqa 16"#t", "#h" \n\t"\
SBUTTERFLY(h,b,c,dq,dqa)\
SBUTTERFLY(e,g,b,dq,dqa)\
SBUTTERFLY(d,f,g,dq,dqa)\
SBUTTERFLY(a,e,f,qdq,dqa)\
SBUTTERFLY(h,d,e,qdq,dqa)\
"movdqa "#h", 16"#t" \n\t"\
"movdqa "#t", "#h" \n\t"\
SBUTTERFLY(h,b,d,qdq,dqa)\
SBUTTERFLY(c,g,b,qdq,dqa)\
"movdqa 16"#t", "#g" \n\t"
#endif
#define MOVQ_WONE(regd) \
__asm__ volatile ( \
"pcmpeqd %%" #regd ", %%" #regd " \n\t" \
"psrlw $15, %%" #regd ::)
void ff_dsputilenc_init_mmx(DSPContext* c, AVCodecContext *avctx);
void ff_dsputil_init_pix_mmx(DSPContext* c, AVCodecContext *avctx);
void ff_add_pixels_clamped_mmx(const DCTELEM *block, uint8_t *pixels, int line_size);
void ff_put_pixels_clamped_mmx(const DCTELEM *block, uint8_t *pixels, int line_size);
void ff_put_signed_pixels_clamped_mmx(const DCTELEM *block, uint8_t *pixels, int line_size);
void ff_put_cavs_qpel8_mc00_mmx2(uint8_t *dst, uint8_t *src, int stride);
void ff_avg_cavs_qpel8_mc00_mmx2(uint8_t *dst, uint8_t *src, int stride);
void ff_put_cavs_qpel16_mc00_mmx2(uint8_t *dst, uint8_t *src, int stride);
void ff_avg_cavs_qpel16_mc00_mmx2(uint8_t *dst, uint8_t *src, int stride);
void ff_put_vc1_mspel_mc00_mmx(uint8_t *dst, const uint8_t *src, int stride, int rnd);
void ff_avg_vc1_mspel_mc00_mmx2(uint8_t *dst, const uint8_t *src, int stride, int rnd);
void ff_put_rv40_qpel8_mc33_mmx(uint8_t *block, uint8_t *pixels, int line_size);
void ff_put_rv40_qpel16_mc33_mmx(uint8_t *block, uint8_t *pixels, int line_size);
void ff_avg_rv40_qpel8_mc33_mmx(uint8_t *block, uint8_t *pixels, int line_size);
void ff_avg_rv40_qpel16_mc33_mmx(uint8_t *block, uint8_t *pixels, int line_size);
void ff_mmx_idct(DCTELEM *block);
void ff_mmxext_idct(DCTELEM *block);
void ff_deinterlace_line_mmx(uint8_t *dst,
const uint8_t *lum_m4, const uint8_t *lum_m3,
const uint8_t *lum_m2, const uint8_t *lum_m1,
const uint8_t *lum,
int size);
void ff_deinterlace_line_inplace_mmx(const uint8_t *lum_m4,
const uint8_t *lum_m3,
const uint8_t *lum_m2,
const uint8_t *lum_m1,
const uint8_t *lum, int size);
#endif /* AVCODEC_X86_DSPUTIL_MMX_H */
Reference in New Issue View Git Blame Copy Permalink