Commit a8b60158 authored by Luca Barbato's avatar Luca Barbato

dsputil: convert remaining functions to use ptrdiff_t strides

Signed-off-by: 's avatarLuca Barbato <lu_zero@gentoo.org>
parent a4472ac0
......@@ -26,73 +26,73 @@
#include "libavutil/arm/cpu.h"
#include "libavcodec/h264qpel.h"
void ff_put_h264_qpel16_mc00_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc10_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc20_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc30_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc01_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc11_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc21_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc31_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc02_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc12_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc22_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc32_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc03_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc13_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc23_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc33_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc00_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel16_mc10_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel16_mc20_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel16_mc30_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel16_mc01_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel16_mc11_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel16_mc21_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel16_mc31_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel16_mc02_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel16_mc12_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel16_mc22_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel16_mc32_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel16_mc03_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel16_mc13_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel16_mc23_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel16_mc33_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel8_mc00_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc10_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc20_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc30_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc01_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc11_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc21_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc31_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc02_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc12_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc22_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc32_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc03_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc13_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc23_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc33_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc00_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel8_mc10_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel8_mc20_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel8_mc30_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel8_mc01_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel8_mc11_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel8_mc21_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel8_mc31_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel8_mc02_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel8_mc12_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel8_mc22_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel8_mc32_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel8_mc03_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel8_mc13_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel8_mc23_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_put_h264_qpel8_mc33_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel16_mc00_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc10_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc20_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc30_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc01_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc11_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc21_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc31_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc02_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc12_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc22_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc32_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc03_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc13_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc23_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc33_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc00_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel16_mc10_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel16_mc20_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel16_mc30_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel16_mc01_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel16_mc11_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel16_mc21_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel16_mc31_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel16_mc02_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel16_mc12_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel16_mc22_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel16_mc32_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel16_mc03_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel16_mc13_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel16_mc23_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel16_mc33_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel8_mc00_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc10_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc20_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc30_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc01_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc11_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc21_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc31_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc02_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc12_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc22_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc32_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc03_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc13_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc23_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc33_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc00_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel8_mc10_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel8_mc20_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel8_mc30_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel8_mc01_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel8_mc11_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel8_mc21_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel8_mc31_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel8_mc02_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel8_mc12_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel8_mc22_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel8_mc32_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel8_mc03_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel8_mc13_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel8_mc23_neon(uint8_t *, uint8_t *, ptrdiff_t);
void ff_avg_h264_qpel8_mc33_neon(uint8_t *, uint8_t *, ptrdiff_t);
av_cold void ff_h264qpel_init_arm(H264QpelContext *c, int bit_depth)
{
......
......@@ -27,7 +27,7 @@
#define DECL_QPEL3(type, w, pos) \
void ff_##type##_rv40_qpel##w##_mc##pos##_neon(uint8_t *dst, uint8_t *src,\
int stride)
ptrdiff_t stride)
#define DECL_QPEL2(w, pos) \
DECL_QPEL3(put, w, pos); \
DECL_QPEL3(avg, w, pos)
......
......@@ -421,63 +421,78 @@ static void OPNAME ## cavs_filt16_hv_ ## NAME(uint8_t *dst, uint8_t *src1, uint8
}\
#define CAVS_MC(OPNAME, SIZE) \
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc10_c(uint8_t *dst, uint8_t *src, int stride){\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc10_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## cavs_filt ## SIZE ## _h_qpel_l(dst, src, stride, stride);\
}\
\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc20_c(uint8_t *dst, uint8_t *src, int stride){\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc20_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## cavs_filt ## SIZE ## _h_hpel(dst, src, stride, stride);\
}\
\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc30_c(uint8_t *dst, uint8_t *src, int stride){\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc30_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## cavs_filt ## SIZE ## _h_qpel_r(dst, src, stride, stride);\
}\
\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc01_c(uint8_t *dst, uint8_t *src, int stride){\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc01_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## cavs_filt ## SIZE ## _v_qpel_l(dst, src, stride, stride);\
}\
\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc02_c(uint8_t *dst, uint8_t *src, int stride){\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc02_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## cavs_filt ## SIZE ## _v_hpel(dst, src, stride, stride);\
}\
\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc03_c(uint8_t *dst, uint8_t *src, int stride){\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc03_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## cavs_filt ## SIZE ## _v_qpel_r(dst, src, stride, stride);\
}\
\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc22_c(uint8_t *dst, uint8_t *src, int stride){\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc22_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## cavs_filt ## SIZE ## _hv_jj(dst, src, NULL, stride, stride); \
}\
\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc11_c(uint8_t *dst, uint8_t *src, int stride){\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc11_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## cavs_filt ## SIZE ## _hv_egpr(dst, src, src, stride, stride); \
}\
\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc13_c(uint8_t *dst, uint8_t *src, int stride){\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc13_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## cavs_filt ## SIZE ## _hv_egpr(dst, src, src+stride, stride, stride); \
}\
\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc31_c(uint8_t *dst, uint8_t *src, int stride){\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc31_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## cavs_filt ## SIZE ## _hv_egpr(dst, src, src+1, stride, stride); \
}\
\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc33_c(uint8_t *dst, uint8_t *src, int stride){\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc33_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## cavs_filt ## SIZE ## _hv_egpr(dst, src, src+stride+1,stride, stride); \
}\
\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc21_c(uint8_t *dst, uint8_t *src, int stride){\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc21_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## cavs_filt ## SIZE ## _hv_ff(dst, src, src+stride+1,stride, stride); \
}\
\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc12_c(uint8_t *dst, uint8_t *src, int stride){\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc12_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## cavs_filt ## SIZE ## _hv_ii(dst, src, src+stride+1,stride, stride); \
}\
\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc32_c(uint8_t *dst, uint8_t *src, int stride){\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc32_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## cavs_filt ## SIZE ## _hv_kk(dst, src, src+stride+1,stride, stride); \
}\
\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc23_c(uint8_t *dst, uint8_t *src, int stride){\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc23_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## cavs_filt ## SIZE ## _hv_qq(dst, src, src+stride+1,stride, stride); \
}\
......
This diff is collapsed.
......@@ -50,10 +50,10 @@ extern uint32_t ff_squareTbl[512];
extern uint8_t ff_cropTbl[256 + 2 * MAX_NEG_CROP];
#define PUTAVG_PIXELS(depth)\
void ff_put_pixels8x8_ ## depth ## _c(uint8_t *dst, uint8_t *src, int stride);\
void ff_avg_pixels8x8_ ## depth ## _c(uint8_t *dst, uint8_t *src, int stride);\
void ff_put_pixels16x16_ ## depth ## _c(uint8_t *dst, uint8_t *src, int stride);\
void ff_avg_pixels16x16_ ## depth ## _c(uint8_t *dst, uint8_t *src, int stride);
void ff_put_pixels8x8_ ## depth ## _c(uint8_t *dst, uint8_t *src, ptrdiff_t stride);\
void ff_avg_pixels8x8_ ## depth ## _c(uint8_t *dst, uint8_t *src, ptrdiff_t stride);\
void ff_put_pixels16x16_ ## depth ## _c(uint8_t *dst, uint8_t *src, ptrdiff_t stride);\
void ff_avg_pixels16x16_ ## depth ## _c(uint8_t *dst, uint8_t *src, ptrdiff_t stride);
PUTAVG_PIXELS( 8)
PUTAVG_PIXELS( 9)
......@@ -65,10 +65,10 @@ PUTAVG_PIXELS(10)
#define ff_avg_pixels16x16_c ff_avg_pixels16x16_8_c
/* RV40 functions */
void ff_put_rv40_qpel16_mc33_c(uint8_t *dst, uint8_t *src, int stride);
void ff_avg_rv40_qpel16_mc33_c(uint8_t *dst, uint8_t *src, int stride);
void ff_put_rv40_qpel8_mc33_c(uint8_t *dst, uint8_t *src, int stride);
void ff_avg_rv40_qpel8_mc33_c(uint8_t *dst, uint8_t *src, int stride);
void ff_put_rv40_qpel16_mc33_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride);
void ff_avg_rv40_qpel16_mc33_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride);
void ff_put_rv40_qpel8_mc33_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride);
void ff_avg_rv40_qpel8_mc33_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride);
void ff_gmc_c(uint8_t *dst, uint8_t *src, int stride, int h, int ox, int oy,
int dxx, int dxy, int dyx, int dyy, int shift, int r, int width, int height);
......@@ -91,14 +91,14 @@ could be reached easily ...
//h for op_pixels_func is limited to {width/2, width} but never larger than 16 and never smaller than 4
typedef void (*op_pixels_func)(uint8_t *block/*align width (8 or 16)*/, const uint8_t *pixels/*align 1*/, ptrdiff_t line_size, int h);
typedef void (*tpel_mc_func)(uint8_t *block/*align width (8 or 16)*/, const uint8_t *pixels/*align 1*/, int line_size, int w, int h);
typedef void (*qpel_mc_func)(uint8_t *dst/*align width (8 or 16)*/, uint8_t *src/*align 1*/, int stride);
typedef void (*qpel_mc_func)(uint8_t *dst/*align width (8 or 16)*/, uint8_t *src/*align 1*/, ptrdiff_t stride);
typedef void (*op_fill_func)(uint8_t *block/*align width (8 or 16)*/, uint8_t value, int line_size, int h);
#define DEF_OLD_QPEL(name)\
void ff_put_ ## name (uint8_t *dst/*align width (8 or 16)*/, uint8_t *src/*align 1*/, int stride);\
void ff_put_no_rnd_ ## name (uint8_t *dst/*align width (8 or 16)*/, uint8_t *src/*align 1*/, int stride);\
void ff_avg_ ## name (uint8_t *dst/*align width (8 or 16)*/, uint8_t *src/*align 1*/, int stride);
void ff_put_ ## name (uint8_t *dst/*align width (8 or 16)*/, uint8_t *src/*align 1*/, ptrdiff_t stride);\
void ff_put_no_rnd_ ## name (uint8_t *dst/*align width (8 or 16)*/, uint8_t *src/*align 1*/, ptrdiff_t stride);\
void ff_avg_ ## name (uint8_t *dst/*align width (8 or 16)*/, uint8_t *src/*align 1*/, ptrdiff_t stride);
DEF_OLD_QPEL(qpel16_mc11_old_c)
DEF_OLD_QPEL(qpel16_mc31_old_c)
......
......@@ -421,15 +421,19 @@ PIXOP2(put, op_put)
#undef op_avg
#undef op_put
void FUNCC(ff_put_pixels8x8)(uint8_t *dst, uint8_t *src, int stride) {
void FUNCC(ff_put_pixels8x8)(uint8_t *dst, uint8_t *src, ptrdiff_t stride)
{
FUNCC(put_pixels8)(dst, src, stride, 8);
}
void FUNCC(ff_avg_pixels8x8)(uint8_t *dst, uint8_t *src, int stride) {
void FUNCC(ff_avg_pixels8x8)(uint8_t *dst, uint8_t *src, ptrdiff_t stride)
{
FUNCC(avg_pixels8)(dst, src, stride, 8);
}
void FUNCC(ff_put_pixels16x16)(uint8_t *dst, uint8_t *src, int stride) {
void FUNCC(ff_put_pixels16x16)(uint8_t *dst, uint8_t *src, ptrdiff_t stride)
{
FUNCC(put_pixels16)(dst, src, stride, 16);
}
void FUNCC(ff_avg_pixels16x16)(uint8_t *dst, uint8_t *src, int stride) {
void FUNCC(ff_avg_pixels16x16)(uint8_t *dst, uint8_t *src, ptrdiff_t stride)
{
FUNCC(avg_pixels16)(dst, src, stride, 16);
}
......@@ -375,27 +375,32 @@ static void FUNC(OPNAME ## h264_qpel16_hv_lowpass)(uint8_t *dst, int16_t *tmp, u
}\
#define H264_MC(OPNAME, SIZE) \
static av_unused void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc00)(uint8_t *dst, uint8_t *src, int stride){\
static av_unused void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc00)(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
FUNCC(OPNAME ## pixels ## SIZE)(dst, src, stride, SIZE);\
}\
\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc10)(uint8_t *dst, uint8_t *src, int stride){\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc10)(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
uint8_t half[SIZE*SIZE*sizeof(pixel)];\
FUNC(put_h264_qpel ## SIZE ## _h_lowpass)(half, src, SIZE*sizeof(pixel), stride);\
FUNC(OPNAME ## pixels ## SIZE ## _l2)(dst, src, half, stride, stride, SIZE*sizeof(pixel), SIZE);\
}\
\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc20)(uint8_t *dst, uint8_t *src, int stride){\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc20)(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
FUNC(OPNAME ## h264_qpel ## SIZE ## _h_lowpass)(dst, src, stride, stride);\
}\
\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc30)(uint8_t *dst, uint8_t *src, int stride){\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc30)(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
uint8_t half[SIZE*SIZE*sizeof(pixel)];\
FUNC(put_h264_qpel ## SIZE ## _h_lowpass)(half, src, SIZE*sizeof(pixel), stride);\
FUNC(OPNAME ## pixels ## SIZE ## _l2)(dst, src+sizeof(pixel), half, stride, stride, SIZE*sizeof(pixel), SIZE);\
}\
\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc01)(uint8_t *dst, uint8_t *src, int stride){\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc01)(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
uint8_t full[SIZE*(SIZE+5)*sizeof(pixel)];\
uint8_t * const full_mid= full + SIZE*2*sizeof(pixel);\
uint8_t half[SIZE*SIZE*sizeof(pixel)];\
......@@ -404,14 +409,16 @@ static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc01)(uint8_t *dst, uint8_t *s
FUNC(OPNAME ## pixels ## SIZE ## _l2)(dst, full_mid, half, stride, SIZE*sizeof(pixel), SIZE*sizeof(pixel), SIZE);\
}\
\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc02)(uint8_t *dst, uint8_t *src, int stride){\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc02)(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
uint8_t full[SIZE*(SIZE+5)*sizeof(pixel)];\
uint8_t * const full_mid= full + SIZE*2*sizeof(pixel);\
FUNC(copy_block ## SIZE )(full, src - stride*2, SIZE*sizeof(pixel), stride, SIZE + 5);\
FUNC(OPNAME ## h264_qpel ## SIZE ## _v_lowpass)(dst, full_mid, stride, SIZE*sizeof(pixel));\
}\
\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc03)(uint8_t *dst, uint8_t *src, int stride){\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc03)(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
uint8_t full[SIZE*(SIZE+5)*sizeof(pixel)];\
uint8_t * const full_mid= full + SIZE*2*sizeof(pixel);\
uint8_t half[SIZE*SIZE*sizeof(pixel)];\
......@@ -420,7 +427,8 @@ static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc03)(uint8_t *dst, uint8_t *s
FUNC(OPNAME ## pixels ## SIZE ## _l2)(dst, full_mid+SIZE*sizeof(pixel), half, stride, SIZE*sizeof(pixel), SIZE*sizeof(pixel), SIZE);\
}\
\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc11)(uint8_t *dst, uint8_t *src, int stride){\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc11)(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
uint8_t full[SIZE*(SIZE+5)*sizeof(pixel)];\
uint8_t * const full_mid= full + SIZE*2*sizeof(pixel);\
uint8_t halfH[SIZE*SIZE*sizeof(pixel)];\
......@@ -431,7 +439,8 @@ static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc11)(uint8_t *dst, uint8_t *s
FUNC(OPNAME ## pixels ## SIZE ## _l2)(dst, halfH, halfV, stride, SIZE*sizeof(pixel), SIZE*sizeof(pixel), SIZE);\
}\
\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc31)(uint8_t *dst, uint8_t *src, int stride){\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc31)(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
uint8_t full[SIZE*(SIZE+5)*sizeof(pixel)];\
uint8_t * const full_mid= full + SIZE*2*sizeof(pixel);\
uint8_t halfH[SIZE*SIZE*sizeof(pixel)];\
......@@ -442,7 +451,8 @@ static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc31)(uint8_t *dst, uint8_t *s
FUNC(OPNAME ## pixels ## SIZE ## _l2)(dst, halfH, halfV, stride, SIZE*sizeof(pixel), SIZE*sizeof(pixel), SIZE);\
}\
\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc13)(uint8_t *dst, uint8_t *src, int stride){\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc13)(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
uint8_t full[SIZE*(SIZE+5)*sizeof(pixel)];\
uint8_t * const full_mid= full + SIZE*2*sizeof(pixel);\
uint8_t halfH[SIZE*SIZE*sizeof(pixel)];\
......@@ -453,7 +463,8 @@ static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc13)(uint8_t *dst, uint8_t *s
FUNC(OPNAME ## pixels ## SIZE ## _l2)(dst, halfH, halfV, stride, SIZE*sizeof(pixel), SIZE*sizeof(pixel), SIZE);\
}\
\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc33)(uint8_t *dst, uint8_t *src, int stride){\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc33)(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
uint8_t full[SIZE*(SIZE+5)*sizeof(pixel)];\
uint8_t * const full_mid= full + SIZE*2*sizeof(pixel);\
uint8_t halfH[SIZE*SIZE*sizeof(pixel)];\
......@@ -464,12 +475,14 @@ static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc33)(uint8_t *dst, uint8_t *s
FUNC(OPNAME ## pixels ## SIZE ## _l2)(dst, halfH, halfV, stride, SIZE*sizeof(pixel), SIZE*sizeof(pixel), SIZE);\
}\
\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc22)(uint8_t *dst, uint8_t *src, int stride){\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc22)(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
int16_t tmp[SIZE*(SIZE+5)*sizeof(pixel)];\
FUNC(OPNAME ## h264_qpel ## SIZE ## _hv_lowpass)(dst, tmp, src, stride, SIZE*sizeof(pixel), stride);\
}\
\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc21)(uint8_t *dst, uint8_t *src, int stride){\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc21)(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
int16_t tmp[SIZE*(SIZE+5)*sizeof(pixel)];\
uint8_t halfH[SIZE*SIZE*sizeof(pixel)];\
uint8_t halfHV[SIZE*SIZE*sizeof(pixel)];\
......@@ -478,7 +491,8 @@ static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc21)(uint8_t *dst, uint8_t *s
FUNC(OPNAME ## pixels ## SIZE ## _l2)(dst, halfH, halfHV, stride, SIZE*sizeof(pixel), SIZE*sizeof(pixel), SIZE);\
}\
\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc23)(uint8_t *dst, uint8_t *src, int stride){\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc23)(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
int16_t tmp[SIZE*(SIZE+5)*sizeof(pixel)];\
uint8_t halfH[SIZE*SIZE*sizeof(pixel)];\
uint8_t halfHV[SIZE*SIZE*sizeof(pixel)];\
......@@ -487,7 +501,8 @@ static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc23)(uint8_t *dst, uint8_t *s
FUNC(OPNAME ## pixels ## SIZE ## _l2)(dst, halfH, halfHV, stride, SIZE*sizeof(pixel), SIZE*sizeof(pixel), SIZE);\
}\
\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc12)(uint8_t *dst, uint8_t *src, int stride){\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc12)(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
uint8_t full[SIZE*(SIZE+5)*sizeof(pixel)];\
uint8_t * const full_mid= full + SIZE*2*sizeof(pixel);\
int16_t tmp[SIZE*(SIZE+5)*sizeof(pixel)];\
......@@ -499,7 +514,8 @@ static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc12)(uint8_t *dst, uint8_t *s
FUNC(OPNAME ## pixels ## SIZE ## _l2)(dst, halfV, halfHV, stride, SIZE*sizeof(pixel), SIZE*sizeof(pixel), SIZE);\
}\
\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc32)(uint8_t *dst, uint8_t *src, int stride){\
static void FUNCC(OPNAME ## h264_qpel ## SIZE ## _mc32)(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
uint8_t full[SIZE*(SIZE+5)*sizeof(pixel)];\
uint8_t * const full_mid= full + SIZE*2*sizeof(pixel);\
int16_t tmp[SIZE*(SIZE+5)*sizeof(pixel)];\
......
......@@ -65,43 +65,51 @@
#undef PREFIX_h264_qpel16_hv_lowpass_num
#define H264_MC(OPNAME, SIZE, CODETYPE) \
static void OPNAME ## h264_qpel ## SIZE ## _mc00_ ## CODETYPE (uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc00_ ## CODETYPE (uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
ff_ ## OPNAME ## pixels ## SIZE ## _ ## CODETYPE(dst, src, stride, SIZE);\
}\
\
static void OPNAME ## h264_qpel ## SIZE ## _mc10_ ## CODETYPE(uint8_t *dst, uint8_t *src, int stride){ \
static void OPNAME ## h264_qpel ## SIZE ## _mc10_ ## CODETYPE(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{ \
DECLARE_ALIGNED(16, uint8_t, half)[SIZE*SIZE];\
put_h264_qpel ## SIZE ## _h_lowpass_ ## CODETYPE(half, src, SIZE, stride);\
OPNAME ## pixels ## SIZE ## _l2_ ## CODETYPE(dst, src, half, stride, stride, SIZE);\
}\
\
static void OPNAME ## h264_qpel ## SIZE ## _mc20_ ## CODETYPE(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc20_ ## CODETYPE(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## h264_qpel ## SIZE ## _h_lowpass_ ## CODETYPE(dst, src, stride, stride);\
}\
\
static void OPNAME ## h264_qpel ## SIZE ## _mc30_ ## CODETYPE(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc30_ ## CODETYPE(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
DECLARE_ALIGNED(16, uint8_t, half)[SIZE*SIZE];\
put_h264_qpel ## SIZE ## _h_lowpass_ ## CODETYPE(half, src, SIZE, stride);\
OPNAME ## pixels ## SIZE ## _l2_ ## CODETYPE(dst, src+1, half, stride, stride, SIZE);\
}\
\
static void OPNAME ## h264_qpel ## SIZE ## _mc01_ ## CODETYPE(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc01_ ## CODETYPE(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
DECLARE_ALIGNED(16, uint8_t, half)[SIZE*SIZE];\
put_h264_qpel ## SIZE ## _v_lowpass_ ## CODETYPE(half, src, SIZE, stride);\
OPNAME ## pixels ## SIZE ## _l2_ ## CODETYPE(dst, src, half, stride, stride, SIZE);\
}\
\
static void OPNAME ## h264_qpel ## SIZE ## _mc02_ ## CODETYPE(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc02_ ## CODETYPE(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## h264_qpel ## SIZE ## _v_lowpass_ ## CODETYPE(dst, src, stride, stride);\
}\
\
static void OPNAME ## h264_qpel ## SIZE ## _mc03_ ## CODETYPE(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc03_ ## CODETYPE(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
DECLARE_ALIGNED(16, uint8_t, half)[SIZE*SIZE];\
put_h264_qpel ## SIZE ## _v_lowpass_ ## CODETYPE(half, src, SIZE, stride);\
OPNAME ## pixels ## SIZE ## _l2_ ## CODETYPE(dst, src+stride, half, stride, stride, SIZE);\
}\
\
static void OPNAME ## h264_qpel ## SIZE ## _mc11_ ## CODETYPE(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc11_ ## CODETYPE(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
DECLARE_ALIGNED(16, uint8_t, halfH)[SIZE*SIZE];\
DECLARE_ALIGNED(16, uint8_t, halfV)[SIZE*SIZE];\
put_h264_qpel ## SIZE ## _h_lowpass_ ## CODETYPE(halfH, src, SIZE, stride);\
......@@ -109,7 +117,8 @@ static void OPNAME ## h264_qpel ## SIZE ## _mc11_ ## CODETYPE(uint8_t *dst, uint
OPNAME ## pixels ## SIZE ## _l2_ ## CODETYPE(dst, halfH, halfV, stride, SIZE, SIZE);\
}\
\
static void OPNAME ## h264_qpel ## SIZE ## _mc31_ ## CODETYPE(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc31_ ## CODETYPE(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
DECLARE_ALIGNED(16, uint8_t, halfH)[SIZE*SIZE];\
DECLARE_ALIGNED(16, uint8_t, halfV)[SIZE*SIZE];\
put_h264_qpel ## SIZE ## _h_lowpass_ ## CODETYPE(halfH, src, SIZE, stride);\
......@@ -117,7 +126,8 @@ static void OPNAME ## h264_qpel ## SIZE ## _mc31_ ## CODETYPE(uint8_t *dst, uint
OPNAME ## pixels ## SIZE ## _l2_ ## CODETYPE(dst, halfH, halfV, stride, SIZE, SIZE);\
}\
\
static void OPNAME ## h264_qpel ## SIZE ## _mc13_ ## CODETYPE(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc13_ ## CODETYPE(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
DECLARE_ALIGNED(16, uint8_t, halfH)[SIZE*SIZE];\
DECLARE_ALIGNED(16, uint8_t, halfV)[SIZE*SIZE];\
put_h264_qpel ## SIZE ## _h_lowpass_ ## CODETYPE(halfH, src + stride, SIZE, stride);\
......@@ -125,7 +135,8 @@ static void OPNAME ## h264_qpel ## SIZE ## _mc13_ ## CODETYPE(uint8_t *dst, uint
OPNAME ## pixels ## SIZE ## _l2_ ## CODETYPE(dst, halfH, halfV, stride, SIZE, SIZE);\
}\
\
static void OPNAME ## h264_qpel ## SIZE ## _mc33_ ## CODETYPE(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc33_ ## CODETYPE(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
DECLARE_ALIGNED(16, uint8_t, halfH)[SIZE*SIZE];\
DECLARE_ALIGNED(16, uint8_t, halfV)[SIZE*SIZE];\
put_h264_qpel ## SIZE ## _h_lowpass_ ## CODETYPE(halfH, src + stride, SIZE, stride);\
......@@ -133,12 +144,14 @@ static void OPNAME ## h264_qpel ## SIZE ## _mc33_ ## CODETYPE(uint8_t *dst, uint
OPNAME ## pixels ## SIZE ## _l2_ ## CODETYPE(dst, halfH, halfV, stride, SIZE, SIZE);\
}\
\
static void OPNAME ## h264_qpel ## SIZE ## _mc22_ ## CODETYPE(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc22_ ## CODETYPE(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
DECLARE_ALIGNED(16, int16_t, tmp)[SIZE*(SIZE+8)];\
OPNAME ## h264_qpel ## SIZE ## _hv_lowpass_ ## CODETYPE(dst, tmp, src, stride, SIZE, stride);\
}\
\
static void OPNAME ## h264_qpel ## SIZE ## _mc21_ ## CODETYPE(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc21_ ## CODETYPE(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
DECLARE_ALIGNED(16, uint8_t, halfH)[SIZE*SIZE];\
DECLARE_ALIGNED(16, uint8_t, halfHV)[SIZE*SIZE];\
DECLARE_ALIGNED(16, int16_t, tmp)[SIZE*(SIZE+8)];\
......@@ -147,7 +160,8 @@ static void OPNAME ## h264_qpel ## SIZE ## _mc21_ ## CODETYPE(uint8_t *dst, uint
OPNAME ## pixels ## SIZE ## _l2_ ## CODETYPE(dst, halfH, halfHV, stride, SIZE, SIZE);\
}\
\
static void OPNAME ## h264_qpel ## SIZE ## _mc23_ ## CODETYPE(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc23_ ## CODETYPE(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
DECLARE_ALIGNED(16, uint8_t, halfH)[SIZE*SIZE];\
DECLARE_ALIGNED(16, uint8_t, halfHV)[SIZE*SIZE];\
DECLARE_ALIGNED(16, int16_t, tmp)[SIZE*(SIZE+8)];\
......@@ -156,7 +170,8 @@ static void OPNAME ## h264_qpel ## SIZE ## _mc23_ ## CODETYPE(uint8_t *dst, uint
OPNAME ## pixels ## SIZE ## _l2_ ## CODETYPE(dst, halfH, halfHV, stride, SIZE, SIZE);\
}\
\
static void OPNAME ## h264_qpel ## SIZE ## _mc12_ ## CODETYPE(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc12_ ## CODETYPE(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
DECLARE_ALIGNED(16, uint8_t, halfV)[SIZE*SIZE];\
DECLARE_ALIGNED(16, uint8_t, halfHV)[SIZE*SIZE];\
DECLARE_ALIGNED(16, int16_t, tmp)[SIZE*(SIZE+8)];\
......@@ -165,7 +180,8 @@ static void OPNAME ## h264_qpel ## SIZE ## _mc12_ ## CODETYPE(uint8_t *dst, uint
OPNAME ## pixels ## SIZE ## _l2_ ## CODETYPE(dst, halfV, halfHV, stride, SIZE, SIZE);\
}\
\
static void OPNAME ## h264_qpel ## SIZE ## _mc32_ ## CODETYPE(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc32_ ## CODETYPE(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
DECLARE_ALIGNED(16, uint8_t, halfV)[SIZE*SIZE];\
DECLARE_ALIGNED(16, uint8_t, halfHV)[SIZE*SIZE];\
DECLARE_ALIGNED(16, int16_t, tmp)[SIZE*(SIZE+8)];\
......
......@@ -210,35 +210,43 @@ static void OPNAME ## rv30_tpel16_hhvv_lowpass(uint8_t *dst, uint8_t *src, int d
\
#define RV30_MC(OPNAME, SIZE) \
static void OPNAME ## rv30_tpel ## SIZE ## _mc10_c(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## rv30_tpel ## SIZE ## _mc10_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## rv30_tpel ## SIZE ## _h_lowpass(dst, src, stride, stride, 12, 6);\
}\
\
static void OPNAME ## rv30_tpel ## SIZE ## _mc20_c(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## rv30_tpel ## SIZE ## _mc20_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## rv30_tpel ## SIZE ## _h_lowpass(dst, src, stride, stride, 6, 12);\
}\
\
static void OPNAME ## rv30_tpel ## SIZE ## _mc01_c(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## rv30_tpel ## SIZE ## _mc01_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## rv30_tpel ## SIZE ## _v_lowpass(dst, src, stride, stride, 12, 6);\
}\
\
static void OPNAME ## rv30_tpel ## SIZE ## _mc02_c(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## rv30_tpel ## SIZE ## _mc02_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## rv30_tpel ## SIZE ## _v_lowpass(dst, src, stride, stride, 6, 12);\
}\
\
static void OPNAME ## rv30_tpel ## SIZE ## _mc11_c(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## rv30_tpel ## SIZE ## _mc11_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## rv30_tpel ## SIZE ## _hv_lowpass(dst, src, stride, stride);\
}\
\
static void OPNAME ## rv30_tpel ## SIZE ## _mc12_c(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## rv30_tpel ## SIZE ## _mc12_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## rv30_tpel ## SIZE ## _hvv_lowpass(dst, src, stride, stride);\
}\
\
static void OPNAME ## rv30_tpel ## SIZE ## _mc21_c(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## rv30_tpel ## SIZE ## _mc21_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## rv30_tpel ## SIZE ## _hhv_lowpass(dst, src, stride, stride);\
}\
\
static void OPNAME ## rv30_tpel ## SIZE ## _mc22_c(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## rv30_tpel ## SIZE ## _mc22_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## rv30_tpel ## SIZE ## _hhvv_lowpass(dst, src, stride, stride);\
}\
\
......
......@@ -103,72 +103,84 @@ static void OPNAME ## rv40_qpel16_h_lowpass(uint8_t *dst, uint8_t *src, int dstS
\
#define RV40_MC(OPNAME, SIZE) \
static void OPNAME ## rv40_qpel ## SIZE ## _mc10_c(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## rv40_qpel ## SIZE ## _mc10_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## rv40_qpel ## SIZE ## _h_lowpass(dst, src, stride, stride, SIZE, 52, 20, 6);\
}\
\
static void OPNAME ## rv40_qpel ## SIZE ## _mc30_c(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## rv40_qpel ## SIZE ## _mc30_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## rv40_qpel ## SIZE ## _h_lowpass(dst, src, stride, stride, SIZE, 20, 52, 6);\
}\
\
static void OPNAME ## rv40_qpel ## SIZE ## _mc01_c(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## rv40_qpel ## SIZE ## _mc01_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## rv40_qpel ## SIZE ## _v_lowpass(dst, src, stride, stride, SIZE, 52, 20, 6);\
}\
\
static void OPNAME ## rv40_qpel ## SIZE ## _mc11_c(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## rv40_qpel ## SIZE ## _mc11_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
uint8_t full[SIZE*(SIZE+5)];\
uint8_t * const full_mid = full + SIZE*2;\
put_rv40_qpel ## SIZE ## _h_lowpass(full, src - 2*stride, SIZE, stride, SIZE+5, 52, 20, 6);\
OPNAME ## rv40_qpel ## SIZE ## _v_lowpass(dst, full_mid, stride, SIZE, SIZE, 52, 20, 6);\
}\
\
static void OPNAME ## rv40_qpel ## SIZE ## _mc21_c(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## rv40_qpel ## SIZE ## _mc21_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
uint8_t full[SIZE*(SIZE+5)];\
uint8_t * const full_mid = full + SIZE*2;\
put_rv40_qpel ## SIZE ## _h_lowpass(full, src - 2*stride, SIZE, stride, SIZE+5, 20, 20, 5);\
OPNAME ## rv40_qpel ## SIZE ## _v_lowpass(dst, full_mid, stride, SIZE, SIZE, 52, 20, 6);\
}\
\
static void OPNAME ## rv40_qpel ## SIZE ## _mc31_c(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## rv40_qpel ## SIZE ## _mc31_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
uint8_t full[SIZE*(SIZE+5)];\
uint8_t * const full_mid = full + SIZE*2;\
put_rv40_qpel ## SIZE ## _h_lowpass(full, src - 2*stride, SIZE, stride, SIZE+5, 20, 52, 6);\
OPNAME ## rv40_qpel ## SIZE ## _v_lowpass(dst, full_mid, stride, SIZE, SIZE, 52, 20, 6);\
}\
\
static void OPNAME ## rv40_qpel ## SIZE ## _mc12_c(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## rv40_qpel ## SIZE ## _mc12_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
uint8_t full[SIZE*(SIZE+5)];\
uint8_t * const full_mid = full + SIZE*2;\
put_rv40_qpel ## SIZE ## _h_lowpass(full, src - 2*stride, SIZE, stride, SIZE+5, 52, 20, 6);\
OPNAME ## rv40_qpel ## SIZE ## _v_lowpass(dst, full_mid, stride, SIZE, SIZE, 20, 20, 5);\
}\
\
static void OPNAME ## rv40_qpel ## SIZE ## _mc22_c(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## rv40_qpel ## SIZE ## _mc22_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
uint8_t full[SIZE*(SIZE+5)];\
uint8_t * const full_mid = full + SIZE*2;\
put_rv40_qpel ## SIZE ## _h_lowpass(full, src - 2*stride, SIZE, stride, SIZE+5, 20, 20, 5);\
OPNAME ## rv40_qpel ## SIZE ## _v_lowpass(dst, full_mid, stride, SIZE, SIZE, 20, 20, 5);\
}\
\
static void OPNAME ## rv40_qpel ## SIZE ## _mc32_c(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## rv40_qpel ## SIZE ## _mc32_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
uint8_t full[SIZE*(SIZE+5)];\
uint8_t * const full_mid = full + SIZE*2;\
put_rv40_qpel ## SIZE ## _h_lowpass(full, src - 2*stride, SIZE, stride, SIZE+5, 20, 52, 6);\
OPNAME ## rv40_qpel ## SIZE ## _v_lowpass(dst, full_mid, stride, SIZE, SIZE, 20, 20, 5);\
}\
\
static void OPNAME ## rv40_qpel ## SIZE ## _mc03_c(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## rv40_qpel ## SIZE ## _mc03_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## rv40_qpel ## SIZE ## _v_lowpass(dst, src, stride, stride, SIZE, 20, 52, 6);\
}\
\
static void OPNAME ## rv40_qpel ## SIZE ## _mc13_c(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## rv40_qpel ## SIZE ## _mc13_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
uint8_t full[SIZE*(SIZE+5)];\
uint8_t * const full_mid = full + SIZE*2;\
put_rv40_qpel ## SIZE ## _h_lowpass(full, src - 2*stride, SIZE, stride, SIZE+5, 52, 20, 6);\
OPNAME ## rv40_qpel ## SIZE ## _v_lowpass(dst, full_mid, stride, SIZE, SIZE, 20, 52, 6);\
}\
\
static void OPNAME ## rv40_qpel ## SIZE ## _mc23_c(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## rv40_qpel ## SIZE ## _mc23_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
uint8_t full[SIZE*(SIZE+5)];\
uint8_t * const full_mid = full + SIZE*2;\
put_rv40_qpel ## SIZE ## _h_lowpass(full, src - 2*stride, SIZE, stride, SIZE+5, 20, 20, 5);\
......
......@@ -627,11 +627,17 @@ VC1_MSPEL_MC(op_avg, avg_)
/* pixel functions - really are entry points to vc1_mspel_mc */
#define PUT_VC1_MSPEL(a, b)\
static void put_vc1_mspel_mc ## a ## b ##_c(uint8_t *dst, const uint8_t *src, int stride, int rnd) { \
put_vc1_mspel_mc(dst, src, stride, a, b, rnd); \
}\
static void avg_vc1_mspel_mc ## a ## b ##_c(uint8_t *dst, const uint8_t *src, int stride, int rnd) { \
avg_vc1_mspel_mc(dst, src, stride, a, b, rnd); \
static void put_vc1_mspel_mc ## a ## b ##_c(uint8_t *dst, \
const uint8_t *src, \
ptrdiff_t stride, int rnd) \
{ \
put_vc1_mspel_mc(dst, src, stride, a, b, rnd); \
} \
static void avg_vc1_mspel_mc ## a ## b ##_c(uint8_t *dst, \
const uint8_t *src, \
ptrdiff_t stride, int rnd) \
{ \
avg_vc1_mspel_mc(dst, src, stride, a, b, rnd); \
}
PUT_VC1_MSPEL(1, 0)
......
......@@ -409,19 +409,23 @@ static void OPNAME ## cavs_qpel16_h_ ## MMX(uint8_t *dst, uint8_t *src, int dstS
}\
#define CAVS_MC(OPNAME, SIZE, MMX) \
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc20_ ## MMX(uint8_t *dst, uint8_t *src, int stride){\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc20_ ## MMX(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## cavs_qpel ## SIZE ## _h_ ## MMX(dst, src, stride, stride);\
}\
\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc01_ ## MMX(uint8_t *dst, uint8_t *src, int stride){\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc01_ ## MMX(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## cavs_qpel ## SIZE ## _v1_ ## MMX(dst, src, stride, stride);\
}\
\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc02_ ## MMX(uint8_t *dst, uint8_t *src, int stride){\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc02_ ## MMX(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## cavs_qpel ## SIZE ## _v2_ ## MMX(dst, src, stride, stride);\
}\
\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc03_ ## MMX(uint8_t *dst, uint8_t *src, int stride){\
static void ff_ ## OPNAME ## cavs_qpel ## SIZE ## _mc03_ ## MMX(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
OPNAME ## cavs_qpel ## SIZE ## _v3_ ## MMX(dst, src, stride, stride);\
}\
......
This diff is collapsed.
......@@ -92,17 +92,17 @@ void ff_put_signed_pixels_clamped_mmx(const int16_t *block, uint8_t *pixels, int
void ff_avg_pixels8_mmxext(uint8_t *block, const uint8_t *pixels,
ptrdiff_t line_size, int h);
void ff_put_cavs_qpel8_mc00_mmxext(uint8_t *dst, uint8_t *src, int stride);
void ff_avg_cavs_qpel8_mc00_mmxext(uint8_t *dst, uint8_t *src, int stride);
void ff_put_cavs_qpel16_mc00_mmxext(uint8_t *dst, uint8_t *src, int stride);
void ff_avg_cavs_qpel16_mc00_mmxext(uint8_t *dst, uint8_t *src, int stride);
void ff_put_cavs_qpel8_mc00_mmxext(uint8_t *dst, uint8_t *src, ptrdiff_t stride);
void ff_avg_cavs_qpel8_mc00_mmxext(uint8_t *dst, uint8_t *src, ptrdiff_t stride);
void ff_put_cavs_qpel16_mc00_mmxext(uint8_t *dst, uint8_t *src, ptrdiff_t stride);
void ff_avg_cavs_qpel16_mc00_mmxext(uint8_t *dst, uint8_t *src, ptrdiff_t stride);
void ff_put_vc1_mspel_mc00_mmx(uint8_t *dst, const uint8_t *src, int stride, int rnd);
void ff_put_vc1_mspel_mc00_mmx(uint8_t *dst, const uint8_t *src, ptrdiff_t 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_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_avg_rv40_qpel8_mc33_mmx(uint8_t *block, uint8_t *pixels, ptrdiff_t stride);
void ff_avg_rv40_qpel16_mc33_mmx(uint8_t *block, uint8_t *pixels, ptrdiff_t stride);
void ff_deinterlace_line_mmx(uint8_t *dst,
const uint8_t *lum_m4, const uint8_t *lum_m3,
......
......@@ -250,81 +250,98 @@ H264_MC_V(OPNAME, SIZE, MMX, ALIGN)\
H264_MC_H(OPNAME, SIZE, MMX, ALIGN)\
H264_MC_HV(OPNAME, SIZE, MMX, ALIGN)\
static void put_h264_qpel16_mc00_sse2 (uint8_t *dst, uint8_t *src, int stride){
static void put_h264_qpel16_mc00_sse2 (uint8_t *dst, uint8_t *src,
ptrdiff_t stride)
{
ff_put_pixels16_sse2(dst, src, stride, 16);
}
static void avg_h264_qpel16_mc00_sse2 (uint8_t *dst, uint8_t *src, int stride){
static void avg_h264_qpel16_mc00_sse2 (uint8_t *dst, uint8_t *src,
ptrdiff_t stride)
{
ff_avg_pixels16_sse2(dst, src, stride, 16);
}
#define put_h264_qpel8_mc00_sse2 put_h264_qpel8_mc00_mmxext
#define avg_h264_qpel8_mc00_sse2 avg_h264_qpel8_mc00_mmxext
#define H264_MC_C(OPNAME, SIZE, MMX, ALIGN) \
static void OPNAME ## h264_qpel ## SIZE ## _mc00_ ## MMX (uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc00_ ## MMX (uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
ff_ ## OPNAME ## pixels ## SIZE ## _ ## MMX(dst, src, stride, SIZE);\
}\
#define H264_MC_H(OPNAME, SIZE, MMX, ALIGN) \
static void OPNAME ## h264_qpel ## SIZE ## _mc10_ ## MMX(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc10_ ## MMX(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
ff_ ## OPNAME ## h264_qpel ## SIZE ## _h_lowpass_l2_ ## MMX(dst, src, src, stride, stride);\
}\
\
static void OPNAME ## h264_qpel ## SIZE ## _mc20_ ## MMX(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc20_ ## MMX(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
ff_ ## OPNAME ## h264_qpel ## SIZE ## _h_lowpass_ ## MMX(dst, src, stride, stride);\
}\
\
static void OPNAME ## h264_qpel ## SIZE ## _mc30_ ## MMX(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc30_ ## MMX(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
ff_ ## OPNAME ## h264_qpel ## SIZE ## _h_lowpass_l2_ ## MMX(dst, src, src+1, stride, stride);\
}\
#define H264_MC_V(OPNAME, SIZE, MMX, ALIGN) \
static void OPNAME ## h264_qpel ## SIZE ## _mc01_ ## MMX(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc01_ ## MMX(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
DECLARE_ALIGNED(ALIGN, uint8_t, temp)[SIZE*SIZE];\
ff_put_h264_qpel ## SIZE ## _v_lowpass_ ## MMX(temp, src, SIZE, stride);\
ff_ ## OPNAME ## pixels ## SIZE ## _l2_ ## MMX(dst, src, temp, stride, stride, SIZE);\
}\
\
static void OPNAME ## h264_qpel ## SIZE ## _mc02_ ## MMX(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc02_ ## MMX(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
ff_ ## OPNAME ## h264_qpel ## SIZE ## _v_lowpass_ ## MMX(dst, src, stride, stride);\
}\
\
static void OPNAME ## h264_qpel ## SIZE ## _mc03_ ## MMX(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc03_ ## MMX(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
DECLARE_ALIGNED(ALIGN, uint8_t, temp)[SIZE*SIZE];\
ff_put_h264_qpel ## SIZE ## _v_lowpass_ ## MMX(temp, src, SIZE, stride);\
ff_ ## OPNAME ## pixels ## SIZE ## _l2_ ## MMX(dst, src+stride, temp, stride, stride, SIZE);\
}\
#define H264_MC_HV(OPNAME, SIZE, MMX, ALIGN) \
static void OPNAME ## h264_qpel ## SIZE ## _mc11_ ## MMX(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc11_ ## MMX(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
DECLARE_ALIGNED(ALIGN, uint8_t, temp)[SIZE*SIZE];\
ff_put_h264_qpel ## SIZE ## _v_lowpass_ ## MMX(temp, src, SIZE, stride);\
ff_ ## OPNAME ## h264_qpel ## SIZE ## _h_lowpass_l2_ ## MMX(dst, src, temp, stride, SIZE);\
}\
\
static void OPNAME ## h264_qpel ## SIZE ## _mc31_ ## MMX(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc31_ ## MMX(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
DECLARE_ALIGNED(ALIGN, uint8_t, temp)[SIZE*SIZE];\
ff_put_h264_qpel ## SIZE ## _v_lowpass_ ## MMX(temp, src+1, SIZE, stride);\
ff_ ## OPNAME ## h264_qpel ## SIZE ## _h_lowpass_l2_ ## MMX(dst, src, temp, stride, SIZE);\
}\
\
static void OPNAME ## h264_qpel ## SIZE ## _mc13_ ## MMX(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc13_ ## MMX(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
DECLARE_ALIGNED(ALIGN, uint8_t, temp)[SIZE*SIZE];\
ff_put_h264_qpel ## SIZE ## _v_lowpass_ ## MMX(temp, src, SIZE, stride);\
ff_ ## OPNAME ## h264_qpel ## SIZE ## _h_lowpass_l2_ ## MMX(dst, src+stride, temp, stride, SIZE);\
}\
\
static void OPNAME ## h264_qpel ## SIZE ## _mc33_ ## MMX(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc33_ ## MMX(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
DECLARE_ALIGNED(ALIGN, uint8_t, temp)[SIZE*SIZE];\
ff_put_h264_qpel ## SIZE ## _v_lowpass_ ## MMX(temp, src+1, SIZE, stride);\
ff_ ## OPNAME ## h264_qpel ## SIZE ## _h_lowpass_l2_ ## MMX(dst, src+stride, temp, stride, SIZE);\
}\
\
static void OPNAME ## h264_qpel ## SIZE ## _mc22_ ## MMX(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc22_ ## MMX(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
DECLARE_ALIGNED(ALIGN, uint16_t, temp)[SIZE*(SIZE<8?12:24)];\
ff_ ## OPNAME ## h264_qpel ## SIZE ## _hv_lowpass_ ## MMX(dst, temp, src, stride, SIZE, stride);\
}\
\
static void OPNAME ## h264_qpel ## SIZE ## _mc21_ ## MMX(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc21_ ## MMX(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
DECLARE_ALIGNED(ALIGN, uint8_t, temp)[SIZE*(SIZE<8?12:24)*2 + SIZE*SIZE];\
uint8_t * const halfHV= temp;\
int16_t * const halfV= (int16_t*)(temp + SIZE*SIZE);\
......@@ -333,7 +350,8 @@ static void OPNAME ## h264_qpel ## SIZE ## _mc21_ ## MMX(uint8_t *dst, uint8_t *
ff_ ## OPNAME ## h264_qpel ## SIZE ## _h_lowpass_l2_ ## MMX(dst, src, halfHV, stride, SIZE);\
}\
\
static void OPNAME ## h264_qpel ## SIZE ## _mc23_ ## MMX(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc23_ ## MMX(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
DECLARE_ALIGNED(ALIGN, uint8_t, temp)[SIZE*(SIZE<8?12:24)*2 + SIZE*SIZE];\
uint8_t * const halfHV= temp;\
int16_t * const halfV= (int16_t*)(temp + SIZE*SIZE);\
......@@ -342,7 +360,8 @@ static void OPNAME ## h264_qpel ## SIZE ## _mc23_ ## MMX(uint8_t *dst, uint8_t *
ff_ ## OPNAME ## h264_qpel ## SIZE ## _h_lowpass_l2_ ## MMX(dst, src+stride, halfHV, stride, SIZE);\
}\
\
static void OPNAME ## h264_qpel ## SIZE ## _mc12_ ## MMX(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc12_ ## MMX(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
DECLARE_ALIGNED(ALIGN, uint8_t, temp)[SIZE*(SIZE<8?12:24)*2 + SIZE*SIZE];\
uint8_t * const halfHV= temp;\
int16_t * const halfV= (int16_t*)(temp + SIZE*SIZE);\
......@@ -351,7 +370,8 @@ static void OPNAME ## h264_qpel ## SIZE ## _mc12_ ## MMX(uint8_t *dst, uint8_t *
ff_ ## OPNAME ## pixels ## SIZE ## _l2_shift5_mmxext(dst, halfV+2, halfHV, stride, SIZE, SIZE);\
}\
\
static void OPNAME ## h264_qpel ## SIZE ## _mc32_ ## MMX(uint8_t *dst, uint8_t *src, int stride){\
static void OPNAME ## h264_qpel ## SIZE ## _mc32_ ## MMX(uint8_t *dst, uint8_t *src, ptrdiff_t stride)\
{\
DECLARE_ALIGNED(ALIGN, uint8_t, temp)[SIZE*(SIZE<8?12:24)*2 + SIZE*SIZE];\
uint8_t * const halfHV= temp;\
int16_t * const halfV= (int16_t*)(temp + SIZE*SIZE);\
......@@ -398,7 +418,7 @@ H264_MC_816(H264_MC_HV, ssse3)
//10bit
#define LUMA_MC_OP(OP, NUM, DEPTH, TYPE, OPT) \
void ff_ ## OP ## _h264_qpel ## NUM ## _ ## TYPE ## _ ## DEPTH ## _ ## OPT \
(uint8_t *dst, uint8_t *src, int stride);
(uint8_t *dst, uint8_t *src, ptrdiff_t stride);
#define LUMA_MC_ALL(DEPTH, TYPE, OPT) \
LUMA_MC_OP(put, 4, DEPTH, TYPE, OPT) \
......
......@@ -71,7 +71,7 @@ DECLARE_WEIGHT(ssse3)
#define QPEL_FUNC_DECL(OP, SIZE, PH, PV, OPT) \
static void OP ## rv40_qpel ##SIZE ##_mc ##PH ##PV ##OPT(uint8_t *dst, \
uint8_t *src, \
int stride) \
ptrdiff_t stride) \
{ \
int i; \
if (PH && PV) { \
......
......@@ -63,7 +63,7 @@ static void vc1_h_loop_filter16_sse4(uint8_t *src, int stride, int pq)
}
static void avg_vc1_mspel_mc00_mmxext(uint8_t *dst, const uint8_t *src,
int stride, int rnd)
ptrdiff_t stride, int rnd)
{
ff_avg_pixels8_mmxext(dst, src, stride, 8);
}
......
......@@ -463,12 +463,17 @@ VC1_MSPEL_MC(avg_)
/** Macro to ease bicubic filter interpolation functions declarations */
#define DECLARE_FUNCTION(a, b) \
static void put_vc1_mspel_mc ## a ## b ## _mmx(uint8_t *dst, const uint8_t *src, int stride, int rnd) { \
static void put_vc1_mspel_mc ## a ## b ## _mmx(uint8_t *dst, \
const uint8_t *src, \
ptrdiff_t stride, \
int rnd) \
{ \
put_vc1_mspel_mc(dst, src, stride, a, b, rnd); \
}\
static void avg_vc1_mspel_mc ## a ## b ## _mmxext(uint8_t *dst, \
const uint8_t *src, \
int stride, int rnd) \
ptrdiff_t stride, \
int rnd) \
{ \
avg_vc1_mspel_mc(dst, src, stride, a, b, rnd); \
}
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment