highperfocused/ffmpeg
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge commit '92f8e06ecb431a427ea13d794e5a6bc927a034d2'

Browse Source 
* commit '92f8e06ecb431a427ea13d794e5a6bc927a034d2':
  x86: dsputil hpeldsp: Move shared template functions into separate object

Conflicts:
	libavcodec/x86/Makefile

Merged-by: Michael Niedermayer <michaelni@gmx.at>
This commit is contained in:
Michael Niedermayer 2013-05-08 14:01:06 +02:00
commit 2f9ef60c97
7 changed files with 69 additions and 26 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
libavcodec/x86/Makefile
View File

@ -39,13 +39,15 @@ MMX-OBJS-$(CONFIG_DSPUTIL) += x86/dsputil_mmx.o \
x86/fpel_mmx.o \ x86/fpel_mmx.o \
x86/idct_mmx_xvid.o \ x86/idct_mmx_xvid.o \
x86/idct_sse2_xvid.o \ x86/idct_sse2_xvid.o \
x86/rnd_mmx.o \
x86/simple_idct.o \ x86/simple_idct.o \
MMX-OBJS-$(CONFIG_ENCODERS) += x86/dsputilenc_mmx.o \ MMX-OBJS-$(CONFIG_ENCODERS) += x86/dsputilenc_mmx.o \
x86/motion_est.o x86/motion_est.o
MMX-OBJS-$(CONFIG_DIRAC_DECODER) += x86/dirac_dwt.o MMX-OBJS-$(CONFIG_DIRAC_DECODER) += x86/dirac_dwt.o
MMX-OBJS-$(CONFIG_HPELDSP) += x86/fpel_mmx.o \ MMX-OBJS-$(CONFIG_HPELDSP) += x86/fpel_mmx.o \
x86/hpeldsp_mmx.o x86/hpeldsp_mmx.o \
x86/rnd_mmx.o
MMX-OBJS-$(CONFIG_SNOW_DECODER) += x86/snowdsp.o MMX-OBJS-$(CONFIG_SNOW_DECODER) += x86/snowdsp.o
MMX-OBJS-$(CONFIG_SNOW_ENCODER) += x86/snowdsp.o MMX-OBJS-$(CONFIG_SNOW_ENCODER) += x86/snowdsp.o
MMX-OBJS-$(CONFIG_VC1_DECODER) += x86/vc1dsp_mmx.o MMX-OBJS-$(CONFIG_VC1_DECODER) += x86/vc1dsp_mmx.o

23
libavcodec/x86/dsputil_mmx.c
View File

@ -101,21 +101,6 @@ void ff_put_no_rnd_mpeg4_qpel8_v_lowpass_mmxext(uint8_t *dst, uint8_t *src,
#if HAVE_INLINE_ASM #if HAVE_INLINE_ASM
/***********************************/
/* MMX rounding */
#define DEF(x, y) x ## _ ## y ## _mmx
#define SET_RND MOVQ_WTWO
#define PAVGBP(a, b, c, d, e, f) PAVGBP_MMX(a, b, c, d, e, f)
#define PAVGB(a, b, c, e) PAVGB_MMX(a, b, c, e)
#include "rnd_template.c"
#undef DEF
#undef SET_RND
#undef PAVGBP
#undef PAVGB
/***********************************/ /***********************************/
/* standard MMX */ /* standard MMX */
@ -898,19 +883,19 @@ QPEL_OP(put_no_rnd_, ff_pw_15, _no_rnd_, mmxext)
#if HAVE_INLINE_ASM #if HAVE_INLINE_ASM
void ff_put_rv40_qpel8_mc33_mmx(uint8_t *dst, uint8_t *src, ptrdiff_t stride) void ff_put_rv40_qpel8_mc33_mmx(uint8_t *dst, uint8_t *src, ptrdiff_t stride)
{ {
put_pixels8_xy2_mmx(dst, src, stride, 8); ff_put_pixels8_xy2_mmx(dst, src, stride, 8);
} }
void ff_put_rv40_qpel16_mc33_mmx(uint8_t *dst, uint8_t *src, ptrdiff_t stride) void ff_put_rv40_qpel16_mc33_mmx(uint8_t *dst, uint8_t *src, ptrdiff_t stride)
{ {
put_pixels16_xy2_mmx(dst, src, stride, 16); ff_put_pixels16_xy2_mmx(dst, src, stride, 16);
} }
void ff_avg_rv40_qpel8_mc33_mmx(uint8_t *dst, uint8_t *src, ptrdiff_t stride) void ff_avg_rv40_qpel8_mc33_mmx(uint8_t *dst, uint8_t *src, ptrdiff_t stride)
{ {
avg_pixels8_xy2_mmx(dst, src, stride, 8); ff_avg_pixels8_xy2_mmx(dst, src, stride, 8);
} }
void ff_avg_rv40_qpel16_mc33_mmx(uint8_t *dst, uint8_t *src, ptrdiff_t stride) void ff_avg_rv40_qpel16_mc33_mmx(uint8_t *dst, uint8_t *src, ptrdiff_t stride)
{ {
avg_pixels16_xy2_mmx(dst, src, stride, 16); ff_avg_pixels16_xy2_mmx(dst, src, stride, 16);
} }
typedef void emulated_edge_mc_func(uint8_t *dst, const uint8_t *src, typedef void emulated_edge_mc_func(uint8_t *dst, const uint8_t *src,

10
libavcodec/x86/dsputil_mmx.h
View File

@ -176,6 +176,16 @@ void ff_put_pixels16_sse2(uint8_t *block, const uint8_t *pixels,
void ff_avg_pixels8_x2_mmx(uint8_t *block, const uint8_t *pixels, void ff_avg_pixels8_x2_mmx(uint8_t *block, const uint8_t *pixels,
ptrdiff_t line_size, int h); ptrdiff_t line_size, int h);
void ff_avg_pixels8_xy2_mmx(uint8_t *block, const uint8_t *pixels,
ptrdiff_t line_size, int h);
void ff_avg_pixels16_xy2_mmx(uint8_t *block, const uint8_t *pixels,
ptrdiff_t line_size, int h);
void ff_put_pixels8_xy2_mmx(uint8_t *block, const uint8_t *pixels,
ptrdiff_t line_size, int h);
void ff_put_pixels16_xy2_mmx(uint8_t *block, const uint8_t *pixels,
ptrdiff_t line_size, int h);
void ff_put_rv40_qpel8_mc33_mmx(uint8_t *block, uint8_t *pixels, ptrdiff_t stride); void ff_put_rv40_qpel8_mc33_mmx(uint8_t *block, uint8_t *pixels, ptrdiff_t stride);
void ff_put_rv40_qpel16_mc33_mmx(uint8_t *block, uint8_t *pixels, ptrdiff_t stride); void ff_put_rv40_qpel16_mc33_mmx(uint8_t *block, uint8_t *pixels, ptrdiff_t stride);
void ff_avg_rv40_qpel8_mc33_mmx(uint8_t *block, uint8_t *pixels, ptrdiff_t stride); void ff_avg_rv40_qpel8_mc33_mmx(uint8_t *block, uint8_t *pixels, ptrdiff_t stride);

7
libavcodec/x86/hpeldsp_init.c
View File

@ -77,8 +77,12 @@ void ff_avg_pixels8_xy2_3dnow(uint8_t *block, const uint8_t *pixels,
#define avg_pixels8_mmx ff_avg_pixels8_mmx #define avg_pixels8_mmx ff_avg_pixels8_mmx
#define avg_pixels8_x2_mmx ff_avg_pixels8_x2_mmx #define avg_pixels8_x2_mmx ff_avg_pixels8_x2_mmx
#define avg_pixels16_mmx ff_avg_pixels16_mmx #define avg_pixels16_mmx ff_avg_pixels16_mmx
#define avg_pixels8_xy2_mmx ff_avg_pixels8_xy2_mmx
#define avg_pixels16_xy2_mmx ff_avg_pixels16_xy2_mmx
#define put_pixels8_mmx ff_put_pixels8_mmx #define put_pixels8_mmx ff_put_pixels8_mmx
#define put_pixels16_mmx ff_put_pixels16_mmx #define put_pixels16_mmx ff_put_pixels16_mmx
#define put_pixels8_xy2_mmx ff_put_pixels8_xy2_mmx
#define put_pixels16_xy2_mmx ff_put_pixels16_xy2_mmx
#define avg_no_rnd_pixels16_mmx ff_avg_pixels16_mmx #define avg_no_rnd_pixels16_mmx ff_avg_pixels16_mmx
#define put_no_rnd_pixels8_mmx ff_put_pixels8_mmx #define put_no_rnd_pixels8_mmx ff_put_pixels8_mmx
#define put_no_rnd_pixels16_mmx ff_put_pixels16_mmx #define put_no_rnd_pixels16_mmx ff_put_pixels16_mmx
@ -91,13 +95,16 @@ void ff_avg_pixels8_xy2_3dnow(uint8_t *block, const uint8_t *pixels,
#define SET_RND MOVQ_WONE #define SET_RND MOVQ_WONE
#define PAVGBP(a, b, c, d, e, f) PAVGBP_MMX_NO_RND(a, b, c, d, e, f) #define PAVGBP(a, b, c, d, e, f) PAVGBP_MMX_NO_RND(a, b, c, d, e, f)
#define PAVGB(a, b, c, e) PAVGB_MMX_NO_RND(a, b, c, e) #define PAVGB(a, b, c, e) PAVGB_MMX_NO_RND(a, b, c, e)
#define STATIC static
#include "rnd_template.c"
#include "hpeldsp_rnd_template.c" #include "hpeldsp_rnd_template.c"
#undef DEF #undef DEF
#undef SET_RND #undef SET_RND
#undef PAVGBP #undef PAVGBP
#undef PAVGB #undef PAVGB
#undef STATIC
/***********************************/ /***********************************/
/* MMX rounding */ /* MMX rounding */

2
libavcodec/x86/hpeldsp_rnd_template.c
View File

@ -24,8 +24,6 @@
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/ */
#include "rnd_template.c"
// put_pixels // put_pixels
static void DEF(put, pixels8_x2)(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h) static void DEF(put, pixels8_x2)(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h)
{ {

32
libavcodec/x86/rnd_mmx.c Normal file
View File

@ -0,0 +1,32 @@
/*
* 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 "config.h"
#include "dsputil_mmx.h"
#if HAVE_INLINE_ASM
#define DEF(x, y) ff_ ## x ## _ ## y ## _mmx
#define SET_RND MOVQ_WTWO
#define PAVGBP(a, b, c, d, e, f) PAVGBP_MMX(a, b, c, d, e, f)
#define PAVGB(a, b, c, e) PAVGB_MMX(a, b, c, e)
#define STATIC
#include "rnd_template.c"
#endif /* HAVE_INLINE_ASM */

17
libavcodec/x86/rnd_template.c
View File

@ -24,8 +24,12 @@
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/ */
#include <stddef.h>
#include <stdint.h>
// put_pixels // put_pixels
static void DEF(put, pixels8_xy2)(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h) STATIC void DEF(put, pixels8_xy2)(uint8_t *block, const uint8_t *pixels,
ptrdiff_t line_size, int h)
{ {
MOVQ_ZERO(mm7); MOVQ_ZERO(mm7);
SET_RND(mm6); // =2 for rnd and =1 for no_rnd version SET_RND(mm6); // =2 for rnd and =1 for no_rnd version
@ -93,7 +97,8 @@ static void DEF(put, pixels8_xy2)(uint8_t *block, const uint8_t *pixels, ptrdiff
// avg_pixels // avg_pixels
// this routine is 'slightly' suboptimal but mostly unused // this routine is 'slightly' suboptimal but mostly unused
static void DEF(avg, pixels8_xy2)(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h) STATIC void DEF(avg, pixels8_xy2)(uint8_t *block, const uint8_t *pixels,
ptrdiff_t line_size, int h)
{ {
MOVQ_ZERO(mm7); MOVQ_ZERO(mm7);
SET_RND(mm6); // =2 for rnd and =1 for no_rnd version SET_RND(mm6); // =2 for rnd and =1 for no_rnd version
@ -168,12 +173,16 @@ static void DEF(avg, pixels8_xy2)(uint8_t *block, const uint8_t *pixels, ptrdiff
} }
//FIXME optimize //FIXME optimize
static void DEF(put, pixels16_xy2)(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h){ STATIC void DEF(put, pixels16_xy2)(uint8_t *block, const uint8_t *pixels,
ptrdiff_t line_size, int h)
{
DEF(put, pixels8_xy2)(block , pixels , line_size, h); DEF(put, pixels8_xy2)(block , pixels , line_size, h);
DEF(put, pixels8_xy2)(block+8, pixels+8, line_size, h); DEF(put, pixels8_xy2)(block+8, pixels+8, line_size, h);
} }
static void DEF(avg, pixels16_xy2)(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h){ STATIC void DEF(avg, pixels16_xy2)(uint8_t *block, const uint8_t *pixels,
ptrdiff_t line_size, int h)
{
DEF(avg, pixels8_xy2)(block , pixels , line_size, h); DEF(avg, pixels8_xy2)(block , pixels , line_size, h);
DEF(avg, pixels8_xy2)(block+8, pixels+8, line_size, h); DEF(avg, pixels8_xy2)(block+8, pixels+8, line_size, h);
} }