Traneptora/FFmpeg
1
Fork 0
forked from FFmpeg/FFmpeg
Code Pull requests Activity

avcodec/h264chroma: Constify src in h264_chroma_mc_func

Browse source 
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
This commit is contained in:
Andreas Rheinhardt 2022-07-26 07:46:45 +02:00
parent 2c8dc7e953
commit 333b32af8e
27 changed files with 211 additions and 211 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

12
libavcodec/aarch64/h264chroma_init_aarch64.c
View file

@ -28,18 +28,18 @@
#include "config.h" #include "config.h"
void ff_put_h264_chroma_mc8_neon(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_put_h264_chroma_mc8_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
void ff_put_h264_chroma_mc4_neon(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_put_h264_chroma_mc4_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
void ff_put_h264_chroma_mc2_neon(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_put_h264_chroma_mc2_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
void ff_avg_h264_chroma_mc8_neon(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_avg_h264_chroma_mc8_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
void ff_avg_h264_chroma_mc4_neon(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_avg_h264_chroma_mc4_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
void ff_avg_h264_chroma_mc2_neon(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_avg_h264_chroma_mc2_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
av_cold void ff_h264chroma_init_aarch64(H264ChromaContext *c, int bit_depth) av_cold void ff_h264chroma_init_aarch64(H264ChromaContext *c, int bit_depth)

4
libavcodec/aarch64/h264cmc_neon.S
View file

@ -23,7 +23,7 @@
#include "libavutil/aarch64/asm.S" #include "libavutil/aarch64/asm.S"
/* chroma_mc8(uint8_t *dst, uint8_t *src, ptrdiff_t stride, int h, int x, int y) */ /* chroma_mc8(uint8_t *dst, const uint8_t *src, ptrdiff_t stride, int h, int x, int y) */
.macro h264_chroma_mc8 type, codec=h264 .macro h264_chroma_mc8 type, codec=h264
function ff_\type\()_\codec\()_chroma_mc8_neon, export=1 function ff_\type\()_\codec\()_chroma_mc8_neon, export=1
.ifc \type,avg .ifc \type,avg
@ -193,7 +193,7 @@ function ff_\type\()_\codec\()_chroma_mc8_neon, export=1
endfunc endfunc
.endm .endm
/* chroma_mc4(uint8_t *dst, uint8_t *src, ptrdiff_t stride, int h, int x, int y) */ /* chroma_mc4(uint8_t *dst, const uint8_t *src, ptrdiff_t stride, int h, int x, int y) */
.macro h264_chroma_mc4 type, codec=h264 .macro h264_chroma_mc4 type, codec=h264
function ff_\type\()_\codec\()_chroma_mc4_neon, export=1 function ff_\type\()_\codec\()_chroma_mc4_neon, export=1
.ifc \type,avg .ifc \type,avg

8
libavcodec/aarch64/rv40dsp_init_aarch64.c
View file

@ -25,14 +25,14 @@
#include "config.h" #include "config.h"
void ff_put_rv40_chroma_mc8_neon(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_put_rv40_chroma_mc8_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
void ff_put_rv40_chroma_mc4_neon(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_put_rv40_chroma_mc4_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
void ff_avg_rv40_chroma_mc8_neon(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_avg_rv40_chroma_mc8_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
void ff_avg_rv40_chroma_mc4_neon(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_avg_rv40_chroma_mc4_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
av_cold void ff_rv40dsp_init_aarch64(RV34DSPContext *c) av_cold void ff_rv40dsp_init_aarch64(RV34DSPContext *c)

8
libavcodec/aarch64/vc1dsp_init_aarch64.c
View file

@ -43,13 +43,13 @@ void ff_vc1_h_loop_filter8_neon(uint8_t *src, ptrdiff_t stride, int pq);
void ff_vc1_v_loop_filter16_neon(uint8_t *src, ptrdiff_t stride, int pq); void ff_vc1_v_loop_filter16_neon(uint8_t *src, ptrdiff_t stride, int pq);
void ff_vc1_h_loop_filter16_neon(uint8_t *src, ptrdiff_t stride, int pq); void ff_vc1_h_loop_filter16_neon(uint8_t *src, ptrdiff_t stride, int pq);
void ff_put_vc1_chroma_mc8_neon(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_put_vc1_chroma_mc8_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
void ff_avg_vc1_chroma_mc8_neon(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_avg_vc1_chroma_mc8_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
void ff_put_vc1_chroma_mc4_neon(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_put_vc1_chroma_mc4_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
void ff_avg_vc1_chroma_mc4_neon(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_avg_vc1_chroma_mc4_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
int ff_vc1_unescape_buffer_helper_neon(const uint8_t *src, int size, uint8_t *dst); int ff_vc1_unescape_buffer_helper_neon(const uint8_t *src, int size, uint8_t *dst);

12
libavcodec/arm/h264chroma_init_arm.c
View file

@ -26,18 +26,18 @@
#include "libavutil/arm/cpu.h" #include "libavutil/arm/cpu.h"
#include "libavcodec/h264chroma.h" #include "libavcodec/h264chroma.h"
void ff_put_h264_chroma_mc8_neon(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_put_h264_chroma_mc8_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
void ff_put_h264_chroma_mc4_neon(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_put_h264_chroma_mc4_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
void ff_put_h264_chroma_mc2_neon(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_put_h264_chroma_mc2_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
void ff_avg_h264_chroma_mc8_neon(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_avg_h264_chroma_mc8_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
void ff_avg_h264_chroma_mc4_neon(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_avg_h264_chroma_mc4_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
void ff_avg_h264_chroma_mc2_neon(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_avg_h264_chroma_mc2_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
av_cold void ff_h264chroma_init_arm(H264ChromaContext *c, int bit_depth) av_cold void ff_h264chroma_init_arm(H264ChromaContext *c, int bit_depth)

4
libavcodec/arm/h264cmc_neon.S
View file

@ -22,7 +22,7 @@
#include "libavutil/arm/asm.S" #include "libavutil/arm/asm.S"
/* chroma_mc8(uint8_t *dst, uint8_t *src, ptrdiff_t stride, int h, int x, int y) */ /* chroma_mc8(uint8_t *dst, const uint8_t *src, ptrdiff_t stride, int h, int x, int y) */
.macro h264_chroma_mc8 type, codec=h264 .macro h264_chroma_mc8 type, codec=h264
function ff_\type\()_\codec\()_chroma_mc8_neon, export=1 function ff_\type\()_\codec\()_chroma_mc8_neon, export=1
push {r4-r7, lr} push {r4-r7, lr}
@ -197,7 +197,7 @@ T cmp r7, #0
endfunc endfunc
.endm .endm
/* chroma_mc4(uint8_t *dst, uint8_t *src, ptrdiff_t stride, int h, int x, int y) */ /* chroma_mc4(uint8_t *dst, const uint8_t *src, ptrdiff_t stride, int h, int x, int y) */
.macro h264_chroma_mc4 type, codec=h264 .macro h264_chroma_mc4 type, codec=h264
function ff_\type\()_\codec\()_chroma_mc4_neon, export=1 function ff_\type\()_\codec\()_chroma_mc4_neon, export=1
push {r4-r7, lr} push {r4-r7, lr}

8
libavcodec/arm/rv40dsp_init_arm.c
View file

@ -49,11 +49,11 @@ DECL_QPEL_Y(1);
DECL_QPEL_Y(2); DECL_QPEL_Y(2);
DECL_QPEL_Y(3); DECL_QPEL_Y(3);
void ff_put_rv40_chroma_mc8_neon(uint8_t *, uint8_t *, int, int, int, int); void ff_put_rv40_chroma_mc8_neon(uint8_t *, const uint8_t *, int, int, int, int);
void ff_put_rv40_chroma_mc4_neon(uint8_t *, uint8_t *, int, int, int, int); void ff_put_rv40_chroma_mc4_neon(uint8_t *, const uint8_t *, int, int, int, int);
void ff_avg_rv40_chroma_mc8_neon(uint8_t *, uint8_t *, int, int, int, int); void ff_avg_rv40_chroma_mc8_neon(uint8_t *, const uint8_t *, int, int, int, int);
void ff_avg_rv40_chroma_mc4_neon(uint8_t *, uint8_t *, int, int, int, int); void ff_avg_rv40_chroma_mc4_neon(uint8_t *, const uint8_t *, int, int, int, int);
void ff_rv40_weight_func_16_neon(uint8_t *, uint8_t *, uint8_t *, int, int, ptrdiff_t); void ff_rv40_weight_func_16_neon(uint8_t *, uint8_t *, uint8_t *, int, int, ptrdiff_t);
void ff_rv40_weight_func_8_neon(uint8_t *, uint8_t *, uint8_t *, int, int, ptrdiff_t); void ff_rv40_weight_func_8_neon(uint8_t *, uint8_t *, uint8_t *, int, int, ptrdiff_t);

8
libavcodec/arm/vc1dsp_init_neon.c
View file

@ -76,13 +76,13 @@ DECL_PUT(3, 1)
DECL_PUT(3, 2) DECL_PUT(3, 2)
DECL_PUT(3, 3) DECL_PUT(3, 3)
void ff_put_vc1_chroma_mc8_neon(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_put_vc1_chroma_mc8_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
void ff_avg_vc1_chroma_mc8_neon(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_avg_vc1_chroma_mc8_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
void ff_put_vc1_chroma_mc4_neon(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_put_vc1_chroma_mc4_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
void ff_avg_vc1_chroma_mc4_neon(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_avg_vc1_chroma_mc4_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
int ff_vc1_unescape_buffer_helper_neon(const uint8_t *src, int size, uint8_t *dst); int ff_vc1_unescape_buffer_helper_neon(const uint8_t *src, int size, uint8_t *dst);

2
libavcodec/h264chroma.h
View file

@ -22,7 +22,7 @@
#include <stddef.h> #include <stddef.h>
#include <stdint.h> #include <stdint.h>
typedef void (*h264_chroma_mc_func)(uint8_t *dst /*align 8*/, uint8_t *src /*align 1*/, ptrdiff_t srcStride, int h, int x, int y); typedef void (*h264_chroma_mc_func)(uint8_t *dst /*align 8*/, const uint8_t *src /*align 1*/, ptrdiff_t srcStride, int h, int x, int y);
typedef struct H264ChromaContext { typedef struct H264ChromaContext {
h264_chroma_mc_func put_h264_chroma_pixels_tab[4]; h264_chroma_mc_func put_h264_chroma_pixels_tab[4];

16
libavcodec/h264chroma_template.c
View file

@ -26,9 +26,9 @@
#include "bit_depth_template.c" #include "bit_depth_template.c"
#define H264_CHROMA_MC(OPNAME, OP)\ #define H264_CHROMA_MC(OPNAME, OP)\
static void FUNCC(OPNAME ## h264_chroma_mc1)(uint8_t *_dst /*align 8*/, uint8_t *_src /*align 1*/, ptrdiff_t stride, int h, int x, int y){\ static void FUNCC(OPNAME ## h264_chroma_mc1)(uint8_t *_dst /*align 8*/, const uint8_t *_src /*align 1*/, ptrdiff_t stride, int h, int x, int y){\
pixel *dst = (pixel*)_dst;\ pixel *dst = (pixel*)_dst;\
pixel *src = (pixel*)_src;\ const pixel *src = (const pixel*)_src;\
const int A=(8-x)*(8-y);\ const int A=(8-x)*(8-y);\
const int B=( x)*(8-y);\ const int B=( x)*(8-y);\
const int C=(8-x)*( y);\ const int C=(8-x)*( y);\
@ -60,10 +60,10 @@ static void FUNCC(OPNAME ## h264_chroma_mc1)(uint8_t *_dst /*align 8*/, uint8_t
}\ }\
}\ }\
}\ }\
static void FUNCC(OPNAME ## h264_chroma_mc2)(uint8_t *_dst /*align 8*/, uint8_t *_src /*align 1*/, ptrdiff_t stride, int h, int x, int y)\ static void FUNCC(OPNAME ## h264_chroma_mc2)(uint8_t *_dst /*align 8*/, const uint8_t *_src /*align 1*/, ptrdiff_t stride, int h, int x, int y)\
{\ {\
pixel *dst = (pixel*)_dst;\ pixel *dst = (pixel*)_dst;\
pixel *src = (pixel*)_src;\ const pixel *src = (const pixel*)_src;\
const int A=(8-x)*(8-y);\ const int A=(8-x)*(8-y);\
const int B=( x)*(8-y);\ const int B=( x)*(8-y);\
const int C=(8-x)*( y);\ const int C=(8-x)*( y);\
@ -99,10 +99,10 @@ static void FUNCC(OPNAME ## h264_chroma_mc2)(uint8_t *_dst /*align 8*/, uint8_t
}\ }\
}\ }\
\ \
static void FUNCC(OPNAME ## h264_chroma_mc4)(uint8_t *_dst /*align 8*/, uint8_t *_src /*align 1*/, ptrdiff_t stride, int h, int x, int y)\ static void FUNCC(OPNAME ## h264_chroma_mc4)(uint8_t *_dst /*align 8*/, const uint8_t *_src /*align 1*/, ptrdiff_t stride, int h, int x, int y)\
{\ {\
pixel *dst = (pixel*)_dst;\ pixel *dst = (pixel*)_dst;\
pixel *src = (pixel*)_src;\ const pixel *src = (const pixel*)_src;\
const int A=(8-x)*(8-y);\ const int A=(8-x)*(8-y);\
const int B=( x)*(8-y);\ const int B=( x)*(8-y);\
const int C=(8-x)*( y);\ const int C=(8-x)*( y);\
@ -144,10 +144,10 @@ static void FUNCC(OPNAME ## h264_chroma_mc4)(uint8_t *_dst /*align 8*/, uint8_t
}\ }\
}\ }\
\ \
static void FUNCC(OPNAME ## h264_chroma_mc8)(uint8_t *_dst /*align 8*/, uint8_t *_src /*align 1*/, ptrdiff_t stride, int h, int x, int y)\ static void FUNCC(OPNAME ## h264_chroma_mc8)(uint8_t *_dst /*align 8*/, const uint8_t *_src /*align 1*/, ptrdiff_t stride, int h, int x, int y)\
{\ {\
pixel *dst = (pixel*)_dst;\ pixel *dst = (pixel*)_dst;\
pixel *src = (pixel*)_src;\ const pixel *src = (const pixel*)_src;\
const int A=(8-x)*(8-y);\ const int A=(8-x)*(8-y);\
const int B=( x)*(8-y);\ const int B=( x)*(8-y);\
const int C=(8-x)*( y);\ const int C=(8-x)*( y);\

82
libavcodec/loongarch/h264chroma_lasx.c
View file

@ -33,7 +33,7 @@ static const uint8_t chroma_mask_arr[64] = {
0, 1, 1, 2, 2, 3, 3, 4, 16, 17, 17, 18, 18, 19, 19, 20 0, 1, 1, 2, 2, 3, 3, 4, 16, 17, 17, 18, 18, 19, 19, 20
}; };
static av_always_inline void avc_chroma_hv_8x4_lasx(uint8_t *src, uint8_t *dst, static av_always_inline void avc_chroma_hv_8x4_lasx(const uint8_t *src, uint8_t *dst,
ptrdiff_t stride, uint32_t coef_hor0, ptrdiff_t stride, uint32_t coef_hor0,
uint32_t coef_hor1, uint32_t coef_ver0, uint32_t coef_hor1, uint32_t coef_ver0,
uint32_t coef_ver1) uint32_t coef_ver1)
@ -71,7 +71,7 @@ static av_always_inline void avc_chroma_hv_8x4_lasx(uint8_t *src, uint8_t *dst,
__lasx_xvstelm_d(out, dst + stride_3x, 0, 3); __lasx_xvstelm_d(out, dst + stride_3x, 0, 3);
} }
static av_always_inline void avc_chroma_hv_8x8_lasx(uint8_t *src, uint8_t *dst, static av_always_inline void avc_chroma_hv_8x8_lasx(const uint8_t *src, uint8_t *dst,
ptrdiff_t stride, uint32_t coef_hor0, ptrdiff_t stride, uint32_t coef_hor0,
uint32_t coef_hor1, uint32_t coef_ver0, uint32_t coef_hor1, uint32_t coef_ver0,
uint32_t coef_ver1) uint32_t coef_ver1)
@ -127,7 +127,7 @@ static av_always_inline void avc_chroma_hv_8x8_lasx(uint8_t *src, uint8_t *dst,
__lasx_xvstelm_d(out1, dst + stride_3x, 0, 3); __lasx_xvstelm_d(out1, dst + stride_3x, 0, 3);
} }
static av_always_inline void avc_chroma_hz_8x4_lasx(uint8_t *src, uint8_t *dst, static av_always_inline void avc_chroma_hz_8x4_lasx(const uint8_t *src, uint8_t *dst,
ptrdiff_t stride, uint32_t coeff0, uint32_t coeff1) ptrdiff_t stride, uint32_t coeff0, uint32_t coeff1)
{ {
ptrdiff_t stride_2x = stride << 1; ptrdiff_t stride_2x = stride << 1;
@ -154,7 +154,7 @@ static av_always_inline void avc_chroma_hz_8x4_lasx(uint8_t *src, uint8_t *dst,
} }
static av_always_inline void avc_chroma_hz_8x8_lasx(uint8_t *src, uint8_t *dst, static av_always_inline void avc_chroma_hz_8x8_lasx(const uint8_t *src, uint8_t *dst,
ptrdiff_t stride, uint32_t coeff0, uint32_t coeff1) ptrdiff_t stride, uint32_t coeff0, uint32_t coeff1)
{ {
ptrdiff_t stride_2x = stride << 1; ptrdiff_t stride_2x = stride << 1;
@ -193,7 +193,7 @@ static av_always_inline void avc_chroma_hz_8x8_lasx(uint8_t *src, uint8_t *dst,
__lasx_xvstelm_d(out1, dst + stride_3x, 0, 3); __lasx_xvstelm_d(out1, dst + stride_3x, 0, 3);
} }
static av_always_inline void avc_chroma_hz_nonmult_lasx(uint8_t *src, static av_always_inline void avc_chroma_hz_nonmult_lasx(const uint8_t *src,
uint8_t *dst, ptrdiff_t stride, uint32_t coeff0, uint8_t *dst, ptrdiff_t stride, uint32_t coeff0,
uint32_t coeff1, int32_t height) uint32_t coeff1, int32_t height)
{ {
@ -239,7 +239,7 @@ static av_always_inline void avc_chroma_hz_nonmult_lasx(uint8_t *src,
} }
} }
static av_always_inline void avc_chroma_vt_8x4_lasx(uint8_t *src, uint8_t *dst, static av_always_inline void avc_chroma_vt_8x4_lasx(const uint8_t *src, uint8_t *dst,
ptrdiff_t stride, uint32_t coeff0, uint32_t coeff1) ptrdiff_t stride, uint32_t coeff0, uint32_t coeff1)
{ {
ptrdiff_t stride_2x = stride << 1; ptrdiff_t stride_2x = stride << 1;
@ -266,7 +266,7 @@ static av_always_inline void avc_chroma_vt_8x4_lasx(uint8_t *src, uint8_t *dst,
__lasx_xvstelm_d(out, dst + stride_3x, 0, 3); __lasx_xvstelm_d(out, dst + stride_3x, 0, 3);
} }
static av_always_inline void avc_chroma_vt_8x8_lasx(uint8_t *src, uint8_t *dst, static av_always_inline void avc_chroma_vt_8x8_lasx(const uint8_t *src, uint8_t *dst,
ptrdiff_t stride, uint32_t coeff0, uint32_t coeff1) ptrdiff_t stride, uint32_t coeff0, uint32_t coeff1)
{ {
ptrdiff_t stride_2x = stride << 1; ptrdiff_t stride_2x = stride << 1;
@ -307,7 +307,7 @@ static av_always_inline void avc_chroma_vt_8x8_lasx(uint8_t *src, uint8_t *dst,
__lasx_xvstelm_d(out1, dst + stride_3x, 0, 3); __lasx_xvstelm_d(out1, dst + stride_3x, 0, 3);
} }
static av_always_inline void copy_width8x8_lasx(uint8_t *src, uint8_t *dst, static av_always_inline void copy_width8x8_lasx(const uint8_t *src, uint8_t *dst,
ptrdiff_t stride) ptrdiff_t stride)
{ {
uint64_t tmp[8]; uint64_t tmp[8];
@ -347,7 +347,7 @@ static av_always_inline void copy_width8x8_lasx(uint8_t *src, uint8_t *dst,
); );
} }
static av_always_inline void copy_width8x4_lasx(uint8_t *src, uint8_t *dst, static av_always_inline void copy_width8x4_lasx(const uint8_t *src, uint8_t *dst,
ptrdiff_t stride) ptrdiff_t stride)
{ {
uint64_t tmp[4]; uint64_t tmp[4];
@ -372,7 +372,7 @@ static av_always_inline void copy_width8x4_lasx(uint8_t *src, uint8_t *dst,
); );
} }
static void avc_chroma_hv_8w_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride, static void avc_chroma_hv_8w_lasx(const uint8_t *src, uint8_t *dst, ptrdiff_t stride,
uint32_t coef_hor0, uint32_t coef_hor1, uint32_t coef_hor0, uint32_t coef_hor1,
uint32_t coef_ver0, uint32_t coef_ver1, uint32_t coef_ver0, uint32_t coef_ver1,
int32_t height) int32_t height)
@ -386,7 +386,7 @@ static void avc_chroma_hv_8w_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride,
} }
} }
static void avc_chroma_hv_4x2_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride, static void avc_chroma_hv_4x2_lasx(const uint8_t *src, uint8_t *dst, ptrdiff_t stride,
uint32_t coef_hor0, uint32_t coef_hor1, uint32_t coef_hor0, uint32_t coef_hor1,
uint32_t coef_ver0, uint32_t coef_ver1) uint32_t coef_ver0, uint32_t coef_ver1)
{ {
@ -414,7 +414,7 @@ static void avc_chroma_hv_4x2_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride,
__lasx_xvstelm_w(res_vt, dst + stride, 0, 1); __lasx_xvstelm_w(res_vt, dst + stride, 0, 1);
} }
static void avc_chroma_hv_4x4_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride, static void avc_chroma_hv_4x4_lasx(const uint8_t *src, uint8_t *dst, ptrdiff_t stride,
uint32_t coef_hor0, uint32_t coef_hor1, uint32_t coef_hor0, uint32_t coef_hor1,
uint32_t coef_ver0, uint32_t coef_ver1) uint32_t coef_ver0, uint32_t coef_ver1)
{ {
@ -446,7 +446,7 @@ static void avc_chroma_hv_4x4_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride,
__lasx_xvstelm_w(res_hz0, dst + stride_3, 0, 5); __lasx_xvstelm_w(res_hz0, dst + stride_3, 0, 5);
} }
static void avc_chroma_hv_4x8_lasx(uint8_t *src, uint8_t * dst, ptrdiff_t stride, static void avc_chroma_hv_4x8_lasx(const uint8_t *src, uint8_t * dst, ptrdiff_t stride,
uint32_t coef_hor0, uint32_t coef_hor1, uint32_t coef_hor0, uint32_t coef_hor1,
uint32_t coef_ver0, uint32_t coef_ver1) uint32_t coef_ver0, uint32_t coef_ver1)
{ {
@ -492,7 +492,7 @@ static void avc_chroma_hv_4x8_lasx(uint8_t *src, uint8_t * dst, ptrdiff_t stride
__lasx_xvstelm_w(res_hz0, dst + stride_3, 0, 7); __lasx_xvstelm_w(res_hz0, dst + stride_3, 0, 7);
} }
static void avc_chroma_hv_4w_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride, static void avc_chroma_hv_4w_lasx(const uint8_t *src, uint8_t *dst, ptrdiff_t stride,
uint32_t coef_hor0, uint32_t coef_hor1, uint32_t coef_hor0, uint32_t coef_hor1,
uint32_t coef_ver0, uint32_t coef_ver1, uint32_t coef_ver0, uint32_t coef_ver1,
int32_t height) int32_t height)
@ -509,7 +509,7 @@ static void avc_chroma_hv_4w_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride,
} }
} }
static void avc_chroma_hz_4x2_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride, static void avc_chroma_hz_4x2_lasx(const uint8_t *src, uint8_t *dst, ptrdiff_t stride,
uint32_t coeff0, uint32_t coeff1) uint32_t coeff0, uint32_t coeff1)
{ {
__m256i src0, src1; __m256i src0, src1;
@ -528,7 +528,7 @@ static void avc_chroma_hz_4x2_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride,
__lasx_xvstelm_w(res, dst + stride, 0, 1); __lasx_xvstelm_w(res, dst + stride, 0, 1);
} }
static void avc_chroma_hz_4x4_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride, static void avc_chroma_hz_4x4_lasx(const uint8_t *src, uint8_t *dst, ptrdiff_t stride,
uint32_t coeff0, uint32_t coeff1) uint32_t coeff0, uint32_t coeff1)
{ {
ptrdiff_t stride_2 = stride << 1; ptrdiff_t stride_2 = stride << 1;
@ -553,7 +553,7 @@ static void avc_chroma_hz_4x4_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride,
__lasx_xvstelm_w(res, dst + stride_3, 0, 5); __lasx_xvstelm_w(res, dst + stride_3, 0, 5);
} }
static void avc_chroma_hz_4x8_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride, static void avc_chroma_hz_4x8_lasx(const uint8_t *src, uint8_t *dst, ptrdiff_t stride,
uint32_t coeff0, uint32_t coeff1) uint32_t coeff0, uint32_t coeff1)
{ {
ptrdiff_t stride_2 = stride << 1; ptrdiff_t stride_2 = stride << 1;
@ -588,7 +588,7 @@ static void avc_chroma_hz_4x8_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride,
__lasx_xvstelm_w(res0, dst + stride_3, 0, 7); __lasx_xvstelm_w(res0, dst + stride_3, 0, 7);
} }
static void avc_chroma_hz_4w_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride, static void avc_chroma_hz_4w_lasx(const uint8_t *src, uint8_t *dst, ptrdiff_t stride,
uint32_t coeff0, uint32_t coeff1, uint32_t coeff0, uint32_t coeff1,
int32_t height) int32_t height)
{ {
@ -601,7 +601,7 @@ static void avc_chroma_hz_4w_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride,
} }
} }
static void avc_chroma_hz_8w_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride, static void avc_chroma_hz_8w_lasx(const uint8_t *src, uint8_t *dst, ptrdiff_t stride,
uint32_t coeff0, uint32_t coeff1, uint32_t coeff0, uint32_t coeff1,
int32_t height) int32_t height)
{ {
@ -614,7 +614,7 @@ static void avc_chroma_hz_8w_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride,
} }
} }
static void avc_chroma_vt_4x2_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride, static void avc_chroma_vt_4x2_lasx(const uint8_t *src, uint8_t *dst, ptrdiff_t stride,
uint32_t coeff0, uint32_t coeff1) uint32_t coeff0, uint32_t coeff1)
{ {
__m256i src0, src1, src2; __m256i src0, src1, src2;
@ -635,7 +635,7 @@ static void avc_chroma_vt_4x2_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride,
__lasx_xvstelm_w(res, dst + stride, 0, 1); __lasx_xvstelm_w(res, dst + stride, 0, 1);
} }
static void avc_chroma_vt_4x4_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride, static void avc_chroma_vt_4x4_lasx(const uint8_t *src, uint8_t *dst, ptrdiff_t stride,
uint32_t coeff0, uint32_t coeff1) uint32_t coeff0, uint32_t coeff1)
{ {
ptrdiff_t stride_2 = stride << 1; ptrdiff_t stride_2 = stride << 1;
@ -664,7 +664,7 @@ static void avc_chroma_vt_4x4_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride,
__lasx_xvstelm_w(res, dst + stride_3, 0, 5); __lasx_xvstelm_w(res, dst + stride_3, 0, 5);
} }
static void avc_chroma_vt_4x8_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride, static void avc_chroma_vt_4x8_lasx(const uint8_t *src, uint8_t *dst, ptrdiff_t stride,
uint32_t coeff0, uint32_t coeff1) uint32_t coeff0, uint32_t coeff1)
{ {
ptrdiff_t stride_2 = stride << 1; ptrdiff_t stride_2 = stride << 1;
@ -705,7 +705,7 @@ static void avc_chroma_vt_4x8_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride,
__lasx_xvstelm_w(res0, dst + stride_3, 0, 7); __lasx_xvstelm_w(res0, dst + stride_3, 0, 7);
} }
static void avc_chroma_vt_4w_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride, static void avc_chroma_vt_4w_lasx(const uint8_t *src, uint8_t *dst, ptrdiff_t stride,
uint32_t coeff0, uint32_t coeff1, uint32_t coeff0, uint32_t coeff1,
int32_t height) int32_t height)
{ {
@ -718,7 +718,7 @@ static void avc_chroma_vt_4w_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride,
} }
} }
static void avc_chroma_vt_8w_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride, static void avc_chroma_vt_8w_lasx(const uint8_t *src, uint8_t *dst, ptrdiff_t stride,
uint32_t coeff0, uint32_t coeff1, uint32_t coeff0, uint32_t coeff1,
int32_t height) int32_t height)
{ {
@ -729,7 +729,7 @@ static void avc_chroma_vt_8w_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride,
} }
} }
static void copy_width4_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride, static void copy_width4_lasx(const uint8_t *src, uint8_t *dst, ptrdiff_t stride,
int32_t height) int32_t height)
{ {
uint32_t tp0, tp1, tp2, tp3, tp4, tp5, tp6, tp7; uint32_t tp0, tp1, tp2, tp3, tp4, tp5, tp6, tp7;
@ -799,7 +799,7 @@ static void copy_width4_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride,
} }
} }
static void copy_width8_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride, static void copy_width8_lasx(const uint8_t *src, uint8_t *dst, ptrdiff_t stride,
int32_t height) int32_t height)
{ {
if (8 == height) { if (8 == height) {
@ -809,7 +809,7 @@ static void copy_width8_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride,
} }
} }
void ff_put_h264_chroma_mc4_lasx(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_put_h264_chroma_mc4_lasx(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int height, int x, int y) int height, int x, int y)
{ {
av_assert2(x < 8 && y < 8 && x >= 0 && y >= 0); av_assert2(x < 8 && y < 8 && x >= 0 && y >= 0);
@ -825,7 +825,7 @@ void ff_put_h264_chroma_mc4_lasx(uint8_t *dst, uint8_t *src, ptrdiff_t stride,
} }
} }
void ff_put_h264_chroma_mc8_lasx(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_put_h264_chroma_mc8_lasx(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int height, int x, int y) int height, int x, int y)
{ {
av_assert2(x < 8 && y < 8 && x >= 0 && y >= 0); av_assert2(x < 8 && y < 8 && x >= 0 && y >= 0);
@ -841,7 +841,7 @@ void ff_put_h264_chroma_mc8_lasx(uint8_t *dst, uint8_t *src, ptrdiff_t stride,
} }
} }
static av_always_inline void avc_chroma_hv_and_aver_dst_8x4_lasx(uint8_t *src, static av_always_inline void avc_chroma_hv_and_aver_dst_8x4_lasx(const uint8_t *src,
uint8_t *dst, ptrdiff_t stride, uint32_t coef_hor0, uint8_t *dst, ptrdiff_t stride, uint32_t coef_hor0,
uint32_t coef_hor1, uint32_t coef_ver0, uint32_t coef_hor1, uint32_t coef_ver0,
uint32_t coef_ver1) uint32_t coef_ver1)
@ -885,7 +885,7 @@ static av_always_inline void avc_chroma_hv_and_aver_dst_8x4_lasx(uint8_t *src,
__lasx_xvstelm_d(out, dst + stride_3x, 0, 3); __lasx_xvstelm_d(out, dst + stride_3x, 0, 3);
} }
static av_always_inline void avc_chroma_hv_and_aver_dst_8x8_lasx(uint8_t *src, static av_always_inline void avc_chroma_hv_and_aver_dst_8x8_lasx(const uint8_t *src,
uint8_t *dst, ptrdiff_t stride, uint32_t coef_hor0, uint8_t *dst, ptrdiff_t stride, uint32_t coef_hor0,
uint32_t coef_hor1, uint32_t coef_ver0, uint32_t coef_hor1, uint32_t coef_ver0,
uint32_t coef_ver1) uint32_t coef_ver1)
@ -957,7 +957,7 @@ static av_always_inline void avc_chroma_hv_and_aver_dst_8x8_lasx(uint8_t *src,
__lasx_xvstelm_d(out1, dst + stride_3x, 0, 3); __lasx_xvstelm_d(out1, dst + stride_3x, 0, 3);
} }
static av_always_inline void avc_chroma_hz_and_aver_dst_8x4_lasx(uint8_t *src, static av_always_inline void avc_chroma_hz_and_aver_dst_8x4_lasx(const uint8_t *src,
uint8_t *dst, ptrdiff_t stride, uint32_t coeff0, uint8_t *dst, ptrdiff_t stride, uint32_t coeff0,
uint32_t coeff1) uint32_t coeff1)
{ {
@ -990,7 +990,7 @@ static av_always_inline void avc_chroma_hz_and_aver_dst_8x4_lasx(uint8_t *src,
__lasx_xvstelm_d(out, dst + stride_3x, 0, 3); __lasx_xvstelm_d(out, dst + stride_3x, 0, 3);
} }
static av_always_inline void avc_chroma_hz_and_aver_dst_8x8_lasx(uint8_t *src, static av_always_inline void avc_chroma_hz_and_aver_dst_8x8_lasx(const uint8_t *src,
uint8_t *dst, ptrdiff_t stride, uint32_t coeff0, uint8_t *dst, ptrdiff_t stride, uint32_t coeff0,
uint32_t coeff1) uint32_t coeff1)
{ {
@ -1043,7 +1043,7 @@ static av_always_inline void avc_chroma_hz_and_aver_dst_8x8_lasx(uint8_t *src,
__lasx_xvstelm_d(out1, dst + stride_3x, 0, 3); __lasx_xvstelm_d(out1, dst + stride_3x, 0, 3);
} }
static av_always_inline void avc_chroma_vt_and_aver_dst_8x4_lasx(uint8_t *src, static av_always_inline void avc_chroma_vt_and_aver_dst_8x4_lasx(const uint8_t *src,
uint8_t *dst, ptrdiff_t stride, uint32_t coeff0, uint8_t *dst, ptrdiff_t stride, uint32_t coeff0,
uint32_t coeff1) uint32_t coeff1)
{ {
@ -1077,7 +1077,7 @@ static av_always_inline void avc_chroma_vt_and_aver_dst_8x4_lasx(uint8_t *src,
__lasx_xvstelm_d(out, dst + stride_3x, 0, 3); __lasx_xvstelm_d(out, dst + stride_3x, 0, 3);
} }
static av_always_inline void avc_chroma_vt_and_aver_dst_8x8_lasx(uint8_t *src, static av_always_inline void avc_chroma_vt_and_aver_dst_8x8_lasx(const uint8_t *src,
uint8_t *dst, ptrdiff_t stride, uint32_t coeff0, uint8_t *dst, ptrdiff_t stride, uint32_t coeff0,
uint32_t coeff1) uint32_t coeff1)
{ {
@ -1132,7 +1132,7 @@ static av_always_inline void avc_chroma_vt_and_aver_dst_8x8_lasx(uint8_t *src,
__lasx_xvstelm_d(out1, dst + stride_3x, 0, 3); __lasx_xvstelm_d(out1, dst + stride_3x, 0, 3);
} }
static av_always_inline void avg_width8x8_lasx(uint8_t *src, uint8_t *dst, static av_always_inline void avg_width8x8_lasx(const uint8_t *src, uint8_t *dst,
ptrdiff_t stride) ptrdiff_t stride)
{ {
__m256i src0, src1, src2, src3; __m256i src0, src1, src2, src3;
@ -1184,7 +1184,7 @@ static av_always_inline void avg_width8x8_lasx(uint8_t *src, uint8_t *dst,
__lasx_xvstelm_d(dst0, dst + stride_3x, 0, 3); __lasx_xvstelm_d(dst0, dst + stride_3x, 0, 3);
} }
static av_always_inline void avg_width8x4_lasx(uint8_t *src, uint8_t *dst, static av_always_inline void avg_width8x4_lasx(const uint8_t *src, uint8_t *dst,
ptrdiff_t stride) ptrdiff_t stride)
{ {
__m256i src0, src1, src2, src3; __m256i src0, src1, src2, src3;
@ -1213,7 +1213,7 @@ static av_always_inline void avg_width8x4_lasx(uint8_t *src, uint8_t *dst,
__lasx_xvstelm_d(dst0, dst + stride_3x, 0, 3); __lasx_xvstelm_d(dst0, dst + stride_3x, 0, 3);
} }
static void avc_chroma_hv_and_aver_dst_8w_lasx(uint8_t *src, uint8_t *dst, static void avc_chroma_hv_and_aver_dst_8w_lasx(const uint8_t *src, uint8_t *dst,
ptrdiff_t stride, ptrdiff_t stride,
uint32_t coef_hor0, uint32_t coef_hor0,
uint32_t coef_hor1, uint32_t coef_hor1,
@ -1230,7 +1230,7 @@ static void avc_chroma_hv_and_aver_dst_8w_lasx(uint8_t *src, uint8_t *dst,
} }
} }
static void avc_chroma_hz_and_aver_dst_8w_lasx(uint8_t *src, uint8_t *dst, static void avc_chroma_hz_and_aver_dst_8w_lasx(const uint8_t *src, uint8_t *dst,
ptrdiff_t stride, uint32_t coeff0, ptrdiff_t stride, uint32_t coeff0,
uint32_t coeff1, int32_t height) uint32_t coeff1, int32_t height)
{ {
@ -1241,7 +1241,7 @@ static void avc_chroma_hz_and_aver_dst_8w_lasx(uint8_t *src, uint8_t *dst,
} }
} }
static void avc_chroma_vt_and_aver_dst_8w_lasx(uint8_t *src, uint8_t *dst, static void avc_chroma_vt_and_aver_dst_8w_lasx(const uint8_t *src, uint8_t *dst,
ptrdiff_t stride, uint32_t coeff0, ptrdiff_t stride, uint32_t coeff0,
uint32_t coeff1, int32_t height) uint32_t coeff1, int32_t height)
{ {
@ -1252,7 +1252,7 @@ static void avc_chroma_vt_and_aver_dst_8w_lasx(uint8_t *src, uint8_t *dst,
} }
} }
static void avg_width8_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride, static void avg_width8_lasx(const uint8_t *src, uint8_t *dst, ptrdiff_t stride,
int32_t height) int32_t height)
{ {
if (8 == height) { if (8 == height) {
@ -1262,7 +1262,7 @@ static void avg_width8_lasx(uint8_t *src, uint8_t *dst, ptrdiff_t stride,
} }
} }
void ff_avg_h264_chroma_mc8_lasx(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_avg_h264_chroma_mc8_lasx(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int height, int x, int y) int height, int x, int y)
{ {
av_assert2(x < 8 && y < 8 && x >= 0 && y >= 0); av_assert2(x < 8 && y < 8 && x >= 0 && y >= 0);

6
libavcodec/loongarch/h264chroma_lasx.h
View file

@ -26,11 +26,11 @@
#include <stddef.h> #include <stddef.h>
#include "libavcodec/h264.h" #include "libavcodec/h264.h"
void ff_put_h264_chroma_mc4_lasx(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_put_h264_chroma_mc4_lasx(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
void ff_put_h264_chroma_mc8_lasx(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_put_h264_chroma_mc8_lasx(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
void ff_avg_h264_chroma_mc8_lasx(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_avg_h264_chroma_mc8_lasx(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
#endif /* AVCODEC_LOONGARCH_H264CHROMA_LASX_H */ #endif /* AVCODEC_LOONGARCH_H264CHROMA_LASX_H */

2
libavcodec/loongarch/vc1dsp_lasx.c
View file

@ -698,7 +698,7 @@ PUT_VC1_MSPEL_MC_LASX(3, 2);
PUT_VC1_MSPEL_MC_LASX(3, 3); PUT_VC1_MSPEL_MC_LASX(3, 3);
void ff_put_no_rnd_vc1_chroma_mc8_lasx(uint8_t *dst /* align 8 */, void ff_put_no_rnd_vc1_chroma_mc8_lasx(uint8_t *dst /* align 8 */,
uint8_t *src /* align 1 */, const uint8_t *src /* align 1 */,
ptrdiff_t stride, int h, int x, int y) ptrdiff_t stride, int h, int x, int y)
{ {
const int intA = (8 - x) * (8 - y); const int intA = (8 - x) * (8 - y);

2
libavcodec/loongarch/vc1dsp_loongarch.h
View file

@ -73,7 +73,7 @@ FF_PUT_VC1_MSPEL_MC_H_LASX(2);
FF_PUT_VC1_MSPEL_MC_H_LASX(3); FF_PUT_VC1_MSPEL_MC_H_LASX(3);
void ff_put_no_rnd_vc1_chroma_mc8_lasx(uint8_t *dst /* align 8 */, void ff_put_no_rnd_vc1_chroma_mc8_lasx(uint8_t *dst /* align 8 */,
uint8_t *src /* align 1 */, const uint8_t *src /* align 1 */,
ptrdiff_t stride, int h, int x, int y); ptrdiff_t stride, int h, int x, int y);
#endif /* AVCODEC_LOONGARCH_VC1DSP_LOONGARCH_H */ #endif /* AVCODEC_LOONGARCH_VC1DSP_LOONGARCH_H */

20
libavcodec/mips/h264chroma_mips.h
View file

@ -22,26 +22,26 @@
#define AVCODEC_MIPS_H264CHROMA_MIPS_H #define AVCODEC_MIPS_H264CHROMA_MIPS_H
#include "libavcodec/h264dec.h" #include "libavcodec/h264dec.h"
void ff_put_h264_chroma_mc8_msa(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_put_h264_chroma_mc8_msa(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int height, int x, int y); int height, int x, int y);
void ff_put_h264_chroma_mc4_msa(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_put_h264_chroma_mc4_msa(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int height, int x, int y); int height, int x, int y);
void ff_put_h264_chroma_mc2_msa(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_put_h264_chroma_mc2_msa(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int height, int x, int y); int height, int x, int y);
void ff_avg_h264_chroma_mc8_msa(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_avg_h264_chroma_mc8_msa(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int height, int x, int y); int height, int x, int y);
void ff_avg_h264_chroma_mc4_msa(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_avg_h264_chroma_mc4_msa(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int height, int x, int y); int height, int x, int y);
void ff_avg_h264_chroma_mc2_msa(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_avg_h264_chroma_mc2_msa(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int height, int x, int y); int height, int x, int y);
void ff_put_h264_chroma_mc8_mmi(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_put_h264_chroma_mc8_mmi(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
void ff_avg_h264_chroma_mc8_mmi(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_avg_h264_chroma_mc8_mmi(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
void ff_put_h264_chroma_mc4_mmi(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_put_h264_chroma_mc4_mmi(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
void ff_avg_h264_chroma_mc4_mmi(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_avg_h264_chroma_mc4_mmi(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y); int h, int x, int y);
#endif /* AVCODEC_MIPS_H264CHROMA_MIPS_H */ #endif /* AVCODEC_MIPS_H264CHROMA_MIPS_H */

8
libavcodec/mips/h264chroma_mmi.c
View file

@ -26,7 +26,7 @@
#include "constants.h" #include "constants.h"
#include "libavutil/mips/mmiutils.h" #include "libavutil/mips/mmiutils.h"
void ff_put_h264_chroma_mc8_mmi(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_put_h264_chroma_mc8_mmi(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y) int h, int x, int y)
{ {
double ftmp[12]; double ftmp[12];
@ -289,7 +289,7 @@ void ff_put_h264_chroma_mc8_mmi(uint8_t *dst, uint8_t *src, ptrdiff_t stride,
} }
} }
void ff_avg_h264_chroma_mc8_mmi(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_avg_h264_chroma_mc8_mmi(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y) int h, int x, int y)
{ {
double ftmp[10]; double ftmp[10];
@ -497,7 +497,7 @@ void ff_avg_h264_chroma_mc8_mmi(uint8_t *dst, uint8_t *src, ptrdiff_t stride,
} }
} }
void ff_put_h264_chroma_mc4_mmi(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_put_h264_chroma_mc4_mmi(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y) int h, int x, int y)
{ {
double ftmp[8]; double ftmp[8];
@ -618,7 +618,7 @@ void ff_put_h264_chroma_mc4_mmi(uint8_t *dst, uint8_t *src, ptrdiff_t stride,
} }
} }
void ff_avg_h264_chroma_mc4_mmi(uint8_t *dst, uint8_t *src, ptrdiff_t stride, void ff_avg_h264_chroma_mc4_mmi(uint8_t *dst, const uint8_t *src, ptrdiff_t stride,
int h, int x, int y) int h, int x, int y)
{ {
double ftmp[8]; double ftmp[8];

142
libavcodec/mips/h264chroma_msa.c
View file

@ -29,7 +29,7 @@ static const uint8_t chroma_mask_arr[16 * 5] = {
0, 1, 1, 2, 16, 17, 17, 18, 16, 17, 17, 18, 18, 19, 19, 20 0, 1, 1, 2, 16, 17, 17, 18, 16, 17, 17, 18, 18, 19, 19, 20
}; };
static void avc_chroma_hz_2x2_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_hz_2x2_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coeff0, uint32_t coeff1) uint32_t coeff0, uint32_t coeff1)
{ {
uint16_t out0, out1; uint16_t out0, out1;
@ -60,7 +60,7 @@ static void avc_chroma_hz_2x2_msa(uint8_t *src, uint8_t *dst, int32_t stride,
SH(out1, dst); SH(out1, dst);
} }
static void avc_chroma_hz_2x4_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_hz_2x4_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coeff0, uint32_t coeff1) uint32_t coeff0, uint32_t coeff1)
{ {
v16u8 src0, src1, src2, src3; v16u8 src0, src1, src2, src3;
@ -88,7 +88,7 @@ static void avc_chroma_hz_2x4_msa(uint8_t *src, uint8_t *dst, int32_t stride,
ST_H4(res, 0, 1, 2, 3, dst, stride); ST_H4(res, 0, 1, 2, 3, dst, stride);
} }
static void avc_chroma_hz_2w_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_hz_2w_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coeff0, uint32_t coeff1, uint32_t coeff0, uint32_t coeff1,
int32_t height) int32_t height)
{ {
@ -99,7 +99,7 @@ static void avc_chroma_hz_2w_msa(uint8_t *src, uint8_t *dst, int32_t stride,
} }
} }
static void avc_chroma_hz_4x2_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_hz_4x2_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coeff0, uint32_t coeff1) uint32_t coeff0, uint32_t coeff1)
{ {
v16i8 src0, src1; v16i8 src0, src1;
@ -124,7 +124,7 @@ static void avc_chroma_hz_4x2_msa(uint8_t *src, uint8_t *dst, int32_t stride,
ST_W2(res, 0, 1, dst, stride); ST_W2(res, 0, 1, dst, stride);
} }
static void avc_chroma_hz_4x4_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_hz_4x4_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coeff0, uint32_t coeff1) uint32_t coeff0, uint32_t coeff1)
{ {
v16u8 src0, src1, src2, src3, out; v16u8 src0, src1, src2, src3, out;
@ -147,7 +147,7 @@ static void avc_chroma_hz_4x4_msa(uint8_t *src, uint8_t *dst, int32_t stride,
ST_W4(out, 0, 1, 2, 3, dst, stride); ST_W4(out, 0, 1, 2, 3, dst, stride);
} }
static void avc_chroma_hz_4x8_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_hz_4x8_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coeff0, uint32_t coeff1) uint32_t coeff0, uint32_t coeff1)
{ {
v16u8 src0, src1, src2, src3, src4, src5, src6, src7, out0, out1; v16u8 src0, src1, src2, src3, src4, src5, src6, src7, out0, out1;
@ -171,7 +171,7 @@ static void avc_chroma_hz_4x8_msa(uint8_t *src, uint8_t *dst, int32_t stride,
ST_W8(out0, out1, 0, 1, 2, 3, 0, 1, 2, 3, dst, stride); ST_W8(out0, out1, 0, 1, 2, 3, 0, 1, 2, 3, dst, stride);
} }
static void avc_chroma_hz_4w_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_hz_4w_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coeff0, uint32_t coeff1, uint32_t coeff0, uint32_t coeff1,
int32_t height) int32_t height)
{ {
@ -184,7 +184,7 @@ static void avc_chroma_hz_4w_msa(uint8_t *src, uint8_t *dst, int32_t stride,
} }
} }
static void avc_chroma_hz_8x4_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_hz_8x4_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coeff0, uint32_t coeff1) uint32_t coeff0, uint32_t coeff1)
{ {
v16u8 src0, src1, src2, src3, out0, out1; v16u8 src0, src1, src2, src3, out0, out1;
@ -207,7 +207,7 @@ static void avc_chroma_hz_8x4_msa(uint8_t *src, uint8_t *dst, int32_t stride,
ST_D4(out0, out1, 0, 1, 0, 1, dst, stride); ST_D4(out0, out1, 0, 1, 0, 1, dst, stride);
} }
static void avc_chroma_hz_8x8_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_hz_8x8_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coeff0, uint32_t coeff1) uint32_t coeff0, uint32_t coeff1)
{ {
v16u8 src0, src1, src2, src3, src4, src5, src6, src7; v16u8 src0, src1, src2, src3, src4, src5, src6, src7;
@ -240,7 +240,7 @@ static void avc_chroma_hz_8x8_msa(uint8_t *src, uint8_t *dst, int32_t stride,
ST_D8(out0, out1, out2, out3, 0, 1, 0, 1, 0, 1, 0, 1, dst, stride); ST_D8(out0, out1, out2, out3, 0, 1, 0, 1, 0, 1, 0, 1, dst, stride);
} }
static void avc_chroma_hz_nonmult_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_hz_nonmult_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, uint32_t coeff0, int32_t stride, uint32_t coeff0,
uint32_t coeff1, int32_t height) uint32_t coeff1, int32_t height)
{ {
@ -289,7 +289,7 @@ static void avc_chroma_hz_nonmult_msa(uint8_t *src, uint8_t *dst,
} }
} }
static void avc_chroma_hz_8w_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_hz_8w_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coeff0, uint32_t coeff1, uint32_t coeff0, uint32_t coeff1,
int32_t height) int32_t height)
{ {
@ -302,7 +302,7 @@ static void avc_chroma_hz_8w_msa(uint8_t *src, uint8_t *dst, int32_t stride,
} }
} }
static void avc_chroma_vt_2x2_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_vt_2x2_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coeff0, uint32_t coeff1) uint32_t coeff0, uint32_t coeff1)
{ {
uint16_t out0, out1; uint16_t out0, out1;
@ -334,7 +334,7 @@ static void avc_chroma_vt_2x2_msa(uint8_t *src, uint8_t *dst, int32_t stride,
SH(out1, dst); SH(out1, dst);
} }
static void avc_chroma_vt_2x4_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_vt_2x4_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coeff0, uint32_t coeff1) uint32_t coeff0, uint32_t coeff1)
{ {
v16u8 src0, src1, src2, src3, src4; v16u8 src0, src1, src2, src3, src4;
@ -362,7 +362,7 @@ static void avc_chroma_vt_2x4_msa(uint8_t *src, uint8_t *dst, int32_t stride,
ST_H4(res, 0, 1, 2, 3, dst, stride); ST_H4(res, 0, 1, 2, 3, dst, stride);
} }
static void avc_chroma_vt_2w_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_vt_2w_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coeff0, uint32_t coeff1, uint32_t coeff0, uint32_t coeff1,
int32_t height) int32_t height)
{ {
@ -373,7 +373,7 @@ static void avc_chroma_vt_2w_msa(uint8_t *src, uint8_t *dst, int32_t stride,
} }
} }
static void avc_chroma_vt_4x2_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_vt_4x2_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coeff0, uint32_t coeff1) uint32_t coeff0, uint32_t coeff1)
{ {
v16u8 src0, src1, src2; v16u8 src0, src1, src2;
@ -397,7 +397,7 @@ static void avc_chroma_vt_4x2_msa(uint8_t *src, uint8_t *dst, int32_t stride,
ST_W2(res, 0, 1, dst, stride); ST_W2(res, 0, 1, dst, stride);
} }
static void avc_chroma_vt_4x4_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_vt_4x4_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coeff0, uint32_t coeff1) uint32_t coeff0, uint32_t coeff1)
{ {
v16u8 src0, src1, src2, src3, src4; v16u8 src0, src1, src2, src3, src4;
@ -421,7 +421,7 @@ static void avc_chroma_vt_4x4_msa(uint8_t *src, uint8_t *dst, int32_t stride,
ST_W4(out, 0, 1, 2, 3, dst, stride); ST_W4(out, 0, 1, 2, 3, dst, stride);
} }
static void avc_chroma_vt_4x8_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_vt_4x8_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coeff0, uint32_t coeff1) uint32_t coeff0, uint32_t coeff1)
{ {
v16u8 src0, src1, src2, src3, src4, src5, src6, src7, src8; v16u8 src0, src1, src2, src3, src4, src5, src6, src7, src8;
@ -449,7 +449,7 @@ static void avc_chroma_vt_4x8_msa(uint8_t *src, uint8_t *dst, int32_t stride,
ST_W8(out0, out1, 0, 1, 2, 3, 0, 1, 2, 3, dst, stride); ST_W8(out0, out1, 0, 1, 2, 3, 0, 1, 2, 3, dst, stride);
} }
static void avc_chroma_vt_4w_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_vt_4w_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coeff0, uint32_t coeff1, uint32_t coeff0, uint32_t coeff1,
int32_t height) int32_t height)
{ {
@ -462,7 +462,7 @@ static void avc_chroma_vt_4w_msa(uint8_t *src, uint8_t *dst, int32_t stride,
} }
} }
static void avc_chroma_vt_8x4_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_vt_8x4_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coeff0, uint32_t coeff1) uint32_t coeff0, uint32_t coeff1)
{ {
v16u8 src0, src1, src2, src3, src4, out0, out1; v16u8 src0, src1, src2, src3, src4, out0, out1;
@ -483,7 +483,7 @@ static void avc_chroma_vt_8x4_msa(uint8_t *src, uint8_t *dst, int32_t stride,
ST_D4(out0, out1, 0, 1, 0, 1, dst, stride); ST_D4(out0, out1, 0, 1, 0, 1, dst, stride);
} }
static void avc_chroma_vt_8x8_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_vt_8x8_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coeff0, uint32_t coeff1) uint32_t coeff0, uint32_t coeff1)
{ {
v16u8 src0, src1, src2, src3, src4, src5, src6, src7, src8; v16u8 src0, src1, src2, src3, src4, src5, src6, src7, src8;
@ -515,7 +515,7 @@ static void avc_chroma_vt_8x8_msa(uint8_t *src, uint8_t *dst, int32_t stride,
ST_D8(out0, out1, out2, out3, 0, 1, 0, 1, 0, 1, 0, 1, dst, stride); ST_D8(out0, out1, out2, out3, 0, 1, 0, 1, 0, 1, 0, 1, dst, stride);
} }
static void avc_chroma_vt_8w_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_vt_8w_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coeff0, uint32_t coeff1, uint32_t coeff0, uint32_t coeff1,
int32_t height) int32_t height)
{ {
@ -526,7 +526,7 @@ static void avc_chroma_vt_8w_msa(uint8_t *src, uint8_t *dst, int32_t stride,
} }
} }
static void avc_chroma_hv_2x2_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_hv_2x2_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coef_hor0, uint32_t coef_hor1, uint32_t coef_hor0, uint32_t coef_hor1,
uint32_t coef_ver0, uint32_t coef_ver1) uint32_t coef_ver0, uint32_t coef_ver1)
{ {
@ -561,7 +561,7 @@ static void avc_chroma_hv_2x2_msa(uint8_t *src, uint8_t *dst, int32_t stride,
SH(out1, dst); SH(out1, dst);
} }
static void avc_chroma_hv_2x4_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_hv_2x4_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coef_hor0, uint32_t coef_hor1, uint32_t coef_hor0, uint32_t coef_hor1,
uint32_t coef_ver0, uint32_t coef_ver1) uint32_t coef_ver0, uint32_t coef_ver1)
{ {
@ -595,7 +595,7 @@ static void avc_chroma_hv_2x4_msa(uint8_t *src, uint8_t *dst, int32_t stride,
ST_H4(res, 0, 1, 2, 3, dst, stride); ST_H4(res, 0, 1, 2, 3, dst, stride);
} }
static void avc_chroma_hv_2w_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_hv_2w_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coef_hor0, uint32_t coef_hor1, uint32_t coef_hor0, uint32_t coef_hor1,
uint32_t coef_ver0, uint32_t coef_ver1, uint32_t coef_ver0, uint32_t coef_ver1,
int32_t height) int32_t height)
@ -609,7 +609,7 @@ static void avc_chroma_hv_2w_msa(uint8_t *src, uint8_t *dst, int32_t stride,
} }
} }
static void avc_chroma_hv_4x2_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_hv_4x2_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coef_hor0, uint32_t coef_hor1, uint32_t coef_hor0, uint32_t coef_hor1,
uint32_t coef_ver0, uint32_t coef_ver1) uint32_t coef_ver0, uint32_t coef_ver1)
{ {
@ -637,7 +637,7 @@ static void avc_chroma_hv_4x2_msa(uint8_t *src, uint8_t *dst, int32_t stride,
ST_W2(res, 0, 1, dst, stride); ST_W2(res, 0, 1, dst, stride);
} }
static void avc_chroma_hv_4x4_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_hv_4x4_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coef_hor0, uint32_t coef_hor1, uint32_t coef_hor0, uint32_t coef_hor1,
uint32_t coef_ver0, uint32_t coef_ver1) uint32_t coef_ver0, uint32_t coef_ver1)
{ {
@ -670,7 +670,7 @@ static void avc_chroma_hv_4x4_msa(uint8_t *src, uint8_t *dst, int32_t stride,
ST_W2(res1, 0, 1, dst + 2 * stride, stride); ST_W2(res1, 0, 1, dst + 2 * stride, stride);
} }
static void avc_chroma_hv_4x8_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_hv_4x8_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coef_hor0, uint32_t coef_hor1, uint32_t coef_hor0, uint32_t coef_hor1,
uint32_t coef_ver0, uint32_t coef_ver1) uint32_t coef_ver0, uint32_t coef_ver1)
{ {
@ -710,7 +710,7 @@ static void avc_chroma_hv_4x8_msa(uint8_t *src, uint8_t *dst, int32_t stride,
ST_W8(res0, res1, 0, 1, 2, 3, 0, 1, 2, 3, dst, stride); ST_W8(res0, res1, 0, 1, 2, 3, 0, 1, 2, 3, dst, stride);
} }
static void avc_chroma_hv_4w_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_hv_4w_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coef_hor0, uint32_t coef_hor1, uint32_t coef_hor0, uint32_t coef_hor1,
uint32_t coef_ver0, uint32_t coef_ver1, uint32_t coef_ver0, uint32_t coef_ver1,
int32_t height) int32_t height)
@ -727,7 +727,7 @@ static void avc_chroma_hv_4w_msa(uint8_t *src, uint8_t *dst, int32_t stride,
} }
} }
static void avc_chroma_hv_8x4_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_hv_8x4_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coef_hor0, uint32_t coef_hor1, uint32_t coef_hor0, uint32_t coef_hor1,
uint32_t coef_ver0, uint32_t coef_ver1) uint32_t coef_ver0, uint32_t coef_ver1)
{ {
@ -770,7 +770,7 @@ static void avc_chroma_hv_8x4_msa(uint8_t *src, uint8_t *dst, int32_t stride,
ST_D4(out0, out1, 0, 1, 0, 1, dst, stride); ST_D4(out0, out1, 0, 1, 0, 1, dst, stride);
} }
static void avc_chroma_hv_8x8_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_hv_8x8_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coef_hor0, uint32_t coef_hor1, uint32_t coef_hor0, uint32_t coef_hor1,
uint32_t coef_ver0, uint32_t coef_ver1) uint32_t coef_ver0, uint32_t coef_ver1)
{ {
@ -826,7 +826,7 @@ static void avc_chroma_hv_8x8_msa(uint8_t *src, uint8_t *dst, int32_t stride,
ST_D8(out0, out1, out2, out3, 0, 1, 0, 1, 0, 1, 0, 1, dst, stride); ST_D8(out0, out1, out2, out3, 0, 1, 0, 1, 0, 1, 0, 1, dst, stride);
} }
static void avc_chroma_hv_8w_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avc_chroma_hv_8w_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
uint32_t coef_hor0, uint32_t coef_hor1, uint32_t coef_hor0, uint32_t coef_hor1,
uint32_t coef_ver0, uint32_t coef_ver1, uint32_t coef_ver0, uint32_t coef_ver1,
int32_t height) int32_t height)
@ -840,7 +840,7 @@ static void avc_chroma_hv_8w_msa(uint8_t *src, uint8_t *dst, int32_t stride,
} }
} }
static void avc_chroma_hz_and_aver_dst_2x2_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_hz_and_aver_dst_2x2_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, uint32_t coeff0, int32_t stride, uint32_t coeff0,
uint32_t coeff1) uint32_t coeff1)
{ {
@ -882,7 +882,7 @@ static void avc_chroma_hz_and_aver_dst_2x2_msa(uint8_t *src, uint8_t *dst,
SH(out1, dst); SH(out1, dst);
} }
static void avc_chroma_hz_and_aver_dst_2x4_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_hz_and_aver_dst_2x4_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, uint32_t coeff0, int32_t stride, uint32_t coeff0,
uint32_t coeff1) uint32_t coeff1)
{ {
@ -922,7 +922,7 @@ static void avc_chroma_hz_and_aver_dst_2x4_msa(uint8_t *src, uint8_t *dst,
ST_H4(dst0, 0, 1, 2, 3, dst, stride); ST_H4(dst0, 0, 1, 2, 3, dst, stride);
} }
static void avc_chroma_hz_and_aver_dst_2w_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_hz_and_aver_dst_2w_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, uint32_t coeff0, int32_t stride, uint32_t coeff0,
uint32_t coeff1, int32_t height) uint32_t coeff1, int32_t height)
{ {
@ -933,7 +933,7 @@ static void avc_chroma_hz_and_aver_dst_2w_msa(uint8_t *src, uint8_t *dst,
} }
} }
static void avc_chroma_hz_and_aver_dst_4x2_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_hz_and_aver_dst_4x2_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, uint32_t coeff0, int32_t stride, uint32_t coeff0,
uint32_t coeff1) uint32_t coeff1)
{ {
@ -966,7 +966,7 @@ static void avc_chroma_hz_and_aver_dst_4x2_msa(uint8_t *src, uint8_t *dst,
ST_W2(dst_data, 0, 1, dst, stride); ST_W2(dst_data, 0, 1, dst, stride);
} }
static void avc_chroma_hz_and_aver_dst_4x4_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_hz_and_aver_dst_4x4_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, uint32_t coeff0, int32_t stride, uint32_t coeff0,
uint32_t coeff1) uint32_t coeff1)
{ {
@ -995,7 +995,7 @@ static void avc_chroma_hz_and_aver_dst_4x4_msa(uint8_t *src, uint8_t *dst,
ST_W4(out, 0, 1, 2, 3, dst, stride); ST_W4(out, 0, 1, 2, 3, dst, stride);
} }
static void avc_chroma_hz_and_aver_dst_4x8_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_hz_and_aver_dst_4x8_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, uint32_t coeff0, int32_t stride, uint32_t coeff0,
uint32_t coeff1) uint32_t coeff1)
{ {
@ -1027,7 +1027,7 @@ static void avc_chroma_hz_and_aver_dst_4x8_msa(uint8_t *src, uint8_t *dst,
ST_W8(out0, out1, 0, 1, 2, 3, 0, 1, 2, 3, dst, stride); ST_W8(out0, out1, 0, 1, 2, 3, 0, 1, 2, 3, dst, stride);
} }
static void avc_chroma_hz_and_aver_dst_4w_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_hz_and_aver_dst_4w_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, uint32_t coeff0, int32_t stride, uint32_t coeff0,
uint32_t coeff1, int32_t height) uint32_t coeff1, int32_t height)
{ {
@ -1040,7 +1040,7 @@ static void avc_chroma_hz_and_aver_dst_4w_msa(uint8_t *src, uint8_t *dst,
} }
} }
static void avc_chroma_hz_and_aver_dst_8x4_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_hz_and_aver_dst_8x4_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, uint32_t coeff0, int32_t stride, uint32_t coeff0,
uint32_t coeff1) uint32_t coeff1)
{ {
@ -1070,7 +1070,7 @@ static void avc_chroma_hz_and_aver_dst_8x4_msa(uint8_t *src, uint8_t *dst,
ST_D4(dst0, dst1, 0, 1, 0, 1, dst, stride); ST_D4(dst0, dst1, 0, 1, 0, 1, dst, stride);
} }
static void avc_chroma_hz_and_aver_dst_8x8_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_hz_and_aver_dst_8x8_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, uint32_t coeff0, int32_t stride, uint32_t coeff0,
uint32_t coeff1) uint32_t coeff1)
{ {
@ -1114,7 +1114,7 @@ static void avc_chroma_hz_and_aver_dst_8x8_msa(uint8_t *src, uint8_t *dst,
ST_D8(out0, out1, out2, out3, 0, 1, 0, 1, 0, 1, 0, 1, dst, stride); ST_D8(out0, out1, out2, out3, 0, 1, 0, 1, 0, 1, 0, 1, dst, stride);
} }
static void avc_chroma_hz_and_aver_dst_8w_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_hz_and_aver_dst_8w_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, uint32_t coeff0, int32_t stride, uint32_t coeff0,
uint32_t coeff1, int32_t height) uint32_t coeff1, int32_t height)
{ {
@ -1125,7 +1125,7 @@ static void avc_chroma_hz_and_aver_dst_8w_msa(uint8_t *src, uint8_t *dst,
} }
} }
static void avc_chroma_vt_and_aver_dst_2x2_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_vt_and_aver_dst_2x2_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, uint32_t coeff0, int32_t stride, uint32_t coeff0,
uint32_t coeff1) uint32_t coeff1)
{ {
@ -1162,7 +1162,7 @@ static void avc_chroma_vt_and_aver_dst_2x2_msa(uint8_t *src, uint8_t *dst,
SH(out1, dst); SH(out1, dst);
} }
static void avc_chroma_vt_and_aver_dst_2x4_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_vt_and_aver_dst_2x4_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, uint32_t coeff0, int32_t stride, uint32_t coeff0,
uint32_t coeff1) uint32_t coeff1)
{ {
@ -1204,7 +1204,7 @@ static void avc_chroma_vt_and_aver_dst_2x4_msa(uint8_t *src, uint8_t *dst,
ST_H4(res, 0, 1, 2, 3, dst, stride); ST_H4(res, 0, 1, 2, 3, dst, stride);
} }
static void avc_chroma_vt_and_aver_dst_2w_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_vt_and_aver_dst_2w_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, uint32_t coeff0, int32_t stride, uint32_t coeff0,
uint32_t coeff1, int32_t height) uint32_t coeff1, int32_t height)
{ {
@ -1215,7 +1215,7 @@ static void avc_chroma_vt_and_aver_dst_2w_msa(uint8_t *src, uint8_t *dst,
} }
} }
static void avc_chroma_vt_and_aver_dst_4x2_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_vt_and_aver_dst_4x2_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, uint32_t coeff0, int32_t stride, uint32_t coeff0,
uint32_t coeff1) uint32_t coeff1)
{ {
@ -1247,7 +1247,7 @@ static void avc_chroma_vt_and_aver_dst_4x2_msa(uint8_t *src, uint8_t *dst,
ST_W2(res, 0, 1, dst, stride); ST_W2(res, 0, 1, dst, stride);
} }
static void avc_chroma_vt_and_aver_dst_4x4_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_vt_and_aver_dst_4x4_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, uint32_t coeff0, int32_t stride, uint32_t coeff0,
uint32_t coeff1) uint32_t coeff1)
{ {
@ -1277,7 +1277,7 @@ static void avc_chroma_vt_and_aver_dst_4x4_msa(uint8_t *src, uint8_t *dst,
ST_W4(out, 0, 1, 2, 3, dst, stride); ST_W4(out, 0, 1, 2, 3, dst, stride);
} }
static void avc_chroma_vt_and_aver_dst_4x8_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_vt_and_aver_dst_4x8_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, uint32_t coeff0, int32_t stride, uint32_t coeff0,
uint32_t coeff1) uint32_t coeff1)
{ {
@ -1313,7 +1313,7 @@ static void avc_chroma_vt_and_aver_dst_4x8_msa(uint8_t *src, uint8_t *dst,
ST_W8(out0, out1, 0, 1, 2, 3, 0, 1, 2, 3, dst, stride); ST_W8(out0, out1, 0, 1, 2, 3, 0, 1, 2, 3, dst, stride);
} }
static void avc_chroma_vt_and_aver_dst_4w_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_vt_and_aver_dst_4w_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, uint32_t coeff0, int32_t stride, uint32_t coeff0,
uint32_t coeff1, int32_t height) uint32_t coeff1, int32_t height)
{ {
@ -1326,7 +1326,7 @@ static void avc_chroma_vt_and_aver_dst_4w_msa(uint8_t *src, uint8_t *dst,
} }
} }
static void avc_chroma_vt_and_aver_dst_8x4_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_vt_and_aver_dst_8x4_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, uint32_t coeff0, int32_t stride, uint32_t coeff0,
uint32_t coeff1) uint32_t coeff1)
{ {
@ -1355,7 +1355,7 @@ static void avc_chroma_vt_and_aver_dst_8x4_msa(uint8_t *src, uint8_t *dst,
ST_D4(out0, out1, 0, 1, 0, 1, dst, stride); ST_D4(out0, out1, 0, 1, 0, 1, dst, stride);
} }
static void avc_chroma_vt_and_aver_dst_8x8_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_vt_and_aver_dst_8x8_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, uint32_t coeff0, int32_t stride, uint32_t coeff0,
uint32_t coeff1) uint32_t coeff1)
{ {
@ -1398,7 +1398,7 @@ static void avc_chroma_vt_and_aver_dst_8x8_msa(uint8_t *src, uint8_t *dst,
ST_D8(out0, out1, out2, out3, 0, 1, 0, 1, 0, 1, 0, 1, dst, stride); ST_D8(out0, out1, out2, out3, 0, 1, 0, 1, 0, 1, 0, 1, dst, stride);
} }
static void avc_chroma_vt_and_aver_dst_8w_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_vt_and_aver_dst_8w_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, uint32_t coeff0, int32_t stride, uint32_t coeff0,
uint32_t coeff1, int32_t height) uint32_t coeff1, int32_t height)
{ {
@ -1409,7 +1409,7 @@ static void avc_chroma_vt_and_aver_dst_8w_msa(uint8_t *src, uint8_t *dst,
} }
} }
static void avc_chroma_hv_and_aver_dst_2x2_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_hv_and_aver_dst_2x2_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, int32_t stride,
uint32_t coef_hor0, uint32_t coef_hor0,
uint32_t coef_hor1, uint32_t coef_hor1,
@ -1451,7 +1451,7 @@ static void avc_chroma_hv_and_aver_dst_2x2_msa(uint8_t *src, uint8_t *dst,
SH(out1, dst); SH(out1, dst);
} }
static void avc_chroma_hv_and_aver_dst_2x4_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_hv_and_aver_dst_2x4_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, int32_t stride,
uint32_t coef_hor0, uint32_t coef_hor0,
uint32_t coef_hor1, uint32_t coef_hor1,
@ -1496,7 +1496,7 @@ static void avc_chroma_hv_and_aver_dst_2x4_msa(uint8_t *src, uint8_t *dst,
ST_H4(dst0, 0, 1, 2, 3, dst, stride); ST_H4(dst0, 0, 1, 2, 3, dst, stride);
} }
static void avc_chroma_hv_and_aver_dst_2w_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_hv_and_aver_dst_2w_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, int32_t stride,
uint32_t coef_hor0, uint32_t coef_hor0,
uint32_t coef_hor1, uint32_t coef_hor1,
@ -1513,7 +1513,7 @@ static void avc_chroma_hv_and_aver_dst_2w_msa(uint8_t *src, uint8_t *dst,
} }
} }
static void avc_chroma_hv_and_aver_dst_4x2_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_hv_and_aver_dst_4x2_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, int32_t stride,
uint32_t coef_hor0, uint32_t coef_hor0,
uint32_t coef_hor1, uint32_t coef_hor1,
@ -1549,7 +1549,7 @@ static void avc_chroma_hv_and_aver_dst_4x2_msa(uint8_t *src, uint8_t *dst,
ST_W2(dst0, 0, 1, dst, stride); ST_W2(dst0, 0, 1, dst, stride);
} }
static void avc_chroma_hv_and_aver_dst_4x4_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_hv_and_aver_dst_4x4_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, int32_t stride,
uint32_t coef_hor0, uint32_t coef_hor0,
uint32_t coef_hor1, uint32_t coef_hor1,
@ -1588,7 +1588,7 @@ static void avc_chroma_hv_and_aver_dst_4x4_msa(uint8_t *src, uint8_t *dst,
ST_W4(out, 0, 1, 2, 3, dst, stride); ST_W4(out, 0, 1, 2, 3, dst, stride);
} }
static void avc_chroma_hv_and_aver_dst_4x8_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_hv_and_aver_dst_4x8_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, int32_t stride,
uint32_t coef_hor0, uint32_t coef_hor0,
uint32_t coef_hor1, uint32_t coef_hor1,
@ -1637,7 +1637,7 @@ static void avc_chroma_hv_and_aver_dst_4x8_msa(uint8_t *src, uint8_t *dst,
ST_W8(res0, res1, 0, 1, 2, 3, 0, 1, 2, 3, dst, stride); ST_W8(res0, res1, 0, 1, 2, 3, 0, 1, 2, 3, dst, stride);
} }
static void avc_chroma_hv_and_aver_dst_4w_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_hv_and_aver_dst_4w_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, int32_t stride,
uint32_t coef_hor0, uint32_t coef_hor0,
uint32_t coef_hor1, uint32_t coef_hor1,
@ -1657,7 +1657,7 @@ static void avc_chroma_hv_and_aver_dst_4w_msa(uint8_t *src, uint8_t *dst,
} }
} }
static void avc_chroma_hv_and_aver_dst_8x4_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_hv_and_aver_dst_8x4_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, int32_t stride,
uint32_t coef_hor0, uint32_t coef_hor0,
uint32_t coef_hor1, uint32_t coef_hor1,
@ -1705,7 +1705,7 @@ static void avc_chroma_hv_and_aver_dst_8x4_msa(uint8_t *src, uint8_t *dst,
ST_D4(out0, out1, 0, 1, 0, 1, dst, stride); ST_D4(out0, out1, 0, 1, 0, 1, dst, stride);
} }
static void avc_chroma_hv_and_aver_dst_8x8_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_hv_and_aver_dst_8x8_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, int32_t stride,
uint32_t coef_hor0, uint32_t coef_hor0,
uint32_t coef_hor1, uint32_t coef_hor1,
@ -1774,7 +1774,7 @@ static void avc_chroma_hv_and_aver_dst_8x8_msa(uint8_t *src, uint8_t *dst,
ST_D8(out0, out1, out2, out3, 0, 1, 0, 1, 0, 1, 0, 1, dst, stride); ST_D8(out0, out1, out2, out3, 0, 1, 0, 1, 0, 1, 0, 1, dst, stride);
} }
static void avc_chroma_hv_and_aver_dst_8w_msa(uint8_t *src, uint8_t *dst, static void avc_chroma_hv_and_aver_dst_8w_msa(const uint8_t *src, uint8_t *dst,
int32_t stride, int32_t stride,
uint32_t coef_hor0, uint32_t coef_hor0,
uint32_t coef_hor1, uint32_t coef_hor1,
@ -1791,7 +1791,7 @@ static void avc_chroma_hv_and_aver_dst_8w_msa(uint8_t *src, uint8_t *dst,
} }
} }
static void copy_width4_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void copy_width4_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
int32_t height) int32_t height)
{ {
uint32_t tp0, tp1, tp2, tp3, tp4, tp5, tp6, tp7; uint32_t tp0, tp1, tp2, tp3, tp4, tp5, tp6, tp7;
@ -1814,7 +1814,7 @@ static void copy_width4_msa(uint8_t *src, uint8_t *dst, int32_t stride,
} }
} }
static void copy_width8_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void copy_width8_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
int32_t height) int32_t height)
{ {
uint64_t src0, src1, src2, src3, src4, src5, src6, src7; uint64_t src0, src1, src2, src3, src4, src5, src6, src7;
@ -1832,7 +1832,7 @@ static void copy_width8_msa(uint8_t *src, uint8_t *dst, int32_t stride,
} }
} }
static void avg_width4_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avg_width4_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
int32_t height) int32_t height)
{ {
uint32_t tp0, tp1, tp2, tp3; uint32_t tp0, tp1, tp2, tp3;
@ -1867,7 +1867,7 @@ static void avg_width4_msa(uint8_t *src, uint8_t *dst, int32_t stride,
} }
} }
static void avg_width8_msa(uint8_t *src, uint8_t *dst, int32_t stride, static void avg_width8_msa(const uint8_t *src, uint8_t *dst, int32_t stride,
int32_t height) int32_t height)
{ {
uint64_t tp0, tp1, tp2, tp3, tp4, tp5, tp6, tp7; uint64_t tp0, tp1, tp2, tp3, tp4, tp5, tp6, tp7;
@ -1903,7 +1903,7 @@ static void avg_width8_msa(uint8_t *src, uint8_t *dst, int32_t stride,
} }
} }
void ff_put_h264_chroma_mc8_msa(uint8_t *dst, uint8_t *src, void ff_put_h264_chroma_mc8_msa(uint8_t *dst, const uint8_t *src,
ptrdiff_t stride, int height, int x, int y) ptrdiff_t stride, int height, int x, int y)
{ {
av_assert2(x < 8 && y < 8 && x >= 0 && y >= 0); av_assert2(x < 8 && y < 8 && x >= 0 && y >= 0);
@ -1919,7 +1919,7 @@ void ff_put_h264_chroma_mc8_msa(uint8_t *dst, uint8_t *src,
} }
} }
void ff_put_h264_chroma_mc4_msa(uint8_t *dst, uint8_t *src, void ff_put_h264_chroma_mc4_msa(uint8_t *dst, const uint8_t *src,
ptrdiff_t stride, int height, int x, int y) ptrdiff_t stride, int height, int x, int y)
{ {
av_assert2(x < 8 && y < 8 && x >= 0 && y >= 0); av_assert2(x < 8 && y < 8 && x >= 0 && y >= 0);
@ -1935,7 +1935,7 @@ void ff_put_h264_chroma_mc4_msa(uint8_t *dst, uint8_t *src,
} }
} }
void ff_put_h264_chroma_mc2_msa(uint8_t *dst, uint8_t *src, void ff_put_h264_chroma_mc2_msa(uint8_t *dst, const uint8_t *src,
ptrdiff_t stride, int height, int x, int y) ptrdiff_t stride, int height, int x, int y)
{ {
int32_t cnt; int32_t cnt;
@ -1958,7 +1958,7 @@ void ff_put_h264_chroma_mc2_msa(uint8_t *dst, uint8_t *src,
} }
} }
void ff_avg_h264_chroma_mc8_msa(uint8_t *dst, uint8_t *src, void ff_avg_h264_chroma_mc8_msa(uint8_t *dst, const uint8_t *src,
ptrdiff_t stride, int height, int x, int y) ptrdiff_t stride, int height, int x, int y)
{ {
av_assert2(x < 8 && y < 8 && x >= 0 && y >= 0); av_assert2(x < 8 && y < 8 && x >= 0 && y >= 0);
@ -1976,7 +1976,7 @@ void ff_avg_h264_chroma_mc8_msa(uint8_t *dst, uint8_t *src,
} }
} }
void ff_avg_h264_chroma_mc4_msa(uint8_t *dst, uint8_t *src, void ff_avg_h264_chroma_mc4_msa(uint8_t *dst, const uint8_t *src,
ptrdiff_t stride, int height, int x, int y) ptrdiff_t stride, int height, int x, int y)
{ {
av_assert2(x < 8 && y < 8 && x >= 0 && y >= 0); av_assert2(x < 8 && y < 8 && x >= 0 && y >= 0);
@ -1993,7 +1993,7 @@ void ff_avg_h264_chroma_mc4_msa(uint8_t *dst, uint8_t *src,
} }
} }
void ff_avg_h264_chroma_mc2_msa(uint8_t *dst, uint8_t *src, void ff_avg_h264_chroma_mc2_msa(uint8_t *dst, const uint8_t *src,
ptrdiff_t stride, int height, int x, int y) ptrdiff_t stride, int height, int x, int y)
{ {
int32_t cnt; int32_t cnt;

8
libavcodec/mips/vc1dsp_mips.h
View file

@ -179,16 +179,16 @@ void ff_vc1_v_loop_filter16_mmi(uint8_t *src, ptrdiff_t stride, int pq);
void ff_vc1_h_loop_filter16_mmi(uint8_t *src, ptrdiff_t stride, int pq); void ff_vc1_h_loop_filter16_mmi(uint8_t *src, ptrdiff_t stride, int pq);
void ff_put_no_rnd_vc1_chroma_mc8_mmi(uint8_t *dst /* align 8 */, void ff_put_no_rnd_vc1_chroma_mc8_mmi(uint8_t *dst /* align 8 */,
uint8_t *src /* align 1 */, const uint8_t *src /* align 1 */,
ptrdiff_t stride, int h, int x, int y); ptrdiff_t stride, int h, int x, int y);
void ff_put_no_rnd_vc1_chroma_mc4_mmi(uint8_t *dst /* align 8 */, void ff_put_no_rnd_vc1_chroma_mc4_mmi(uint8_t *dst /* align 8 */,
uint8_t *src /* align 1 */, const uint8_t *src /* align 1 */,
ptrdiff_t stride, int h, int x, int y); ptrdiff_t stride, int h, int x, int y);
void ff_avg_no_rnd_vc1_chroma_mc8_mmi(uint8_t *dst /* align 8 */, void ff_avg_no_rnd_vc1_chroma_mc8_mmi(uint8_t *dst /* align 8 */,
uint8_t *src /* align 1 */, const uint8_t *src /* align 1 */,
ptrdiff_t stride, int h, int x, int y); ptrdiff_t stride, int h, int x, int y);
void ff_avg_no_rnd_vc1_chroma_mc4_mmi(uint8_t *dst /* align 8 */, void ff_avg_no_rnd_vc1_chroma_mc4_mmi(uint8_t *dst /* align 8 */,
uint8_t *src /* align 1 */, const uint8_t *src /* align 1 */,
ptrdiff_t stride, int h, int x, int y); ptrdiff_t stride, int h, int x, int y);
void ff_vc1_inv_trans_8x8_msa(int16_t block[64]); void ff_vc1_inv_trans_8x8_msa(int16_t block[64]);

8
libavcodec/mips/vc1dsp_mmi.c
View file

@ -2235,7 +2235,7 @@ DECLARE_FUNCTION(3, 3)
void ff_put_no_rnd_vc1_chroma_mc8_mmi(uint8_t *dst /* align 8 */, void ff_put_no_rnd_vc1_chroma_mc8_mmi(uint8_t *dst /* align 8 */,
uint8_t *src /* align 1 */, const uint8_t *src /* align 1 */,
ptrdiff_t stride, int h, int x, int y) ptrdiff_t stride, int h, int x, int y)
{ {
union mmi_intfloat64 A, B, C, D; union mmi_intfloat64 A, B, C, D;
@ -2291,7 +2291,7 @@ void ff_put_no_rnd_vc1_chroma_mc8_mmi(uint8_t *dst /* align 8 */,
} }
void ff_put_no_rnd_vc1_chroma_mc4_mmi(uint8_t *dst /* align 8 */, void ff_put_no_rnd_vc1_chroma_mc4_mmi(uint8_t *dst /* align 8 */,
uint8_t *src /* align 1 */, const uint8_t *src /* align 1 */,
ptrdiff_t stride, int h, int x, int y) ptrdiff_t stride, int h, int x, int y)
{ {
union mmi_intfloat64 A, B, C, D; union mmi_intfloat64 A, B, C, D;
@ -2345,7 +2345,7 @@ void ff_put_no_rnd_vc1_chroma_mc4_mmi(uint8_t *dst /* align 8 */,
} }
void ff_avg_no_rnd_vc1_chroma_mc8_mmi(uint8_t *dst /* align 8 */, void ff_avg_no_rnd_vc1_chroma_mc8_mmi(uint8_t *dst /* align 8 */,
uint8_t *src /* align 1 */, const uint8_t *src /* align 1 */,
ptrdiff_t stride, int h, int x, int y) ptrdiff_t stride, int h, int x, int y)
{ {
union mmi_intfloat64 A, B, C, D; union mmi_intfloat64 A, B, C, D;
@ -2404,7 +2404,7 @@ void ff_avg_no_rnd_vc1_chroma_mc8_mmi(uint8_t *dst /* align 8 */,
} }
void ff_avg_no_rnd_vc1_chroma_mc4_mmi(uint8_t *dst /* align 8 */, void ff_avg_no_rnd_vc1_chroma_mc4_mmi(uint8_t *dst /* align 8 */,
uint8_t *src /* align 1 */, const uint8_t *src /* align 1 */,
ptrdiff_t stride, int h, int x, int y) ptrdiff_t stride, int h, int x, int y)
{ {
union mmi_intfloat64 A, B, C, D; union mmi_intfloat64 A, B, C, D;

4
libavcodec/ppc/h264chroma_template.c
View file

@ -109,7 +109,7 @@
#endif /* HAVE_BIGENDIAN */ #endif /* HAVE_BIGENDIAN */
#ifdef PREFIX_h264_chroma_mc8_altivec #ifdef PREFIX_h264_chroma_mc8_altivec
static void PREFIX_h264_chroma_mc8_altivec(uint8_t * dst, uint8_t * src, static void PREFIX_h264_chroma_mc8_altivec(uint8_t * dst, const uint8_t * src,
ptrdiff_t stride, int h, ptrdiff_t stride, int h,
int x, int y) int x, int y)
{ {
@ -184,7 +184,7 @@ static void PREFIX_h264_chroma_mc8_altivec(uint8_t * dst, uint8_t * src,
/* this code assume that stride % 16 == 0 */ /* this code assume that stride % 16 == 0 */
#ifdef PREFIX_no_rnd_vc1_chroma_mc8_altivec #ifdef PREFIX_no_rnd_vc1_chroma_mc8_altivec
static void PREFIX_no_rnd_vc1_chroma_mc8_altivec(uint8_t *dst, uint8_t *src, static void PREFIX_no_rnd_vc1_chroma_mc8_altivec(uint8_t *dst, const uint8_t *src,
ptrdiff_t stride, int h, ptrdiff_t stride, int h,
int x, int y) int x, int y)
{ {

4
libavcodec/rv40dsp.c
View file

@ -292,7 +292,7 @@ static const int rv40_bias[4][4] = {
#define RV40_CHROMA_MC(OPNAME, OP)\ #define RV40_CHROMA_MC(OPNAME, OP)\
static void OPNAME ## rv40_chroma_mc4_c(uint8_t *dst /*align 8*/,\ static void OPNAME ## rv40_chroma_mc4_c(uint8_t *dst /*align 8*/,\
uint8_t *src /*align 1*/,\ const uint8_t *src /*align 1*/,\
ptrdiff_t stride, int h, int x, int y)\ ptrdiff_t stride, int h, int x, int y)\
{\ {\
const int A = (8-x) * (8-y);\ const int A = (8-x) * (8-y);\
@ -328,7 +328,7 @@ static void OPNAME ## rv40_chroma_mc4_c(uint8_t *dst /*align 8*/,\
}\ }\
\ \
static void OPNAME ## rv40_chroma_mc8_c(uint8_t *dst/*align 8*/,\ static void OPNAME ## rv40_chroma_mc8_c(uint8_t *dst/*align 8*/,\
uint8_t *src/*align 1*/,\ const uint8_t *src/*align 1*/,\
ptrdiff_t stride, int h, int x, int y)\ ptrdiff_t stride, int h, int x, int y)\
{\ {\
const int A = (8-x) * (8-y);\ const int A = (8-x) * (8-y);\

8
libavcodec/vc1dsp.c
View file

@ -787,7 +787,7 @@ PUT_VC1_MSPEL(3, 3)
((A * src[a] + B * src[a + 1] + \ ((A * src[a] + B * src[a + 1] + \
C * src[stride + a] + D * src[stride + a + 1] + 32 - 4) >> 6) C * src[stride + a] + D * src[stride + a + 1] + 32 - 4) >> 6)
static void put_no_rnd_vc1_chroma_mc8_c(uint8_t *dst /* align 8 */, static void put_no_rnd_vc1_chroma_mc8_c(uint8_t *dst /* align 8 */,
uint8_t *src /* align 1 */, const uint8_t *src /* align 1 */,
ptrdiff_t stride, int h, int x, int y) ptrdiff_t stride, int h, int x, int y)
{ {
const int A = (8 - x) * (8 - y); const int A = (8 - x) * (8 - y);
@ -812,7 +812,7 @@ static void put_no_rnd_vc1_chroma_mc8_c(uint8_t *dst /* align 8 */,
} }
} }
static void put_no_rnd_vc1_chroma_mc4_c(uint8_t *dst, uint8_t *src, static void put_no_rnd_vc1_chroma_mc4_c(uint8_t *dst, const uint8_t *src,
ptrdiff_t stride, int h, int x, int y) ptrdiff_t stride, int h, int x, int y)
{ {
const int A = (8 - x) * (8 - y); const int A = (8 - x) * (8 - y);
@ -835,7 +835,7 @@ static void put_no_rnd_vc1_chroma_mc4_c(uint8_t *dst, uint8_t *src,
#define avg2(a, b) (((a) + (b) + 1) >> 1) #define avg2(a, b) (((a) + (b) + 1) >> 1)
static void avg_no_rnd_vc1_chroma_mc8_c(uint8_t *dst /* align 8 */, static void avg_no_rnd_vc1_chroma_mc8_c(uint8_t *dst /* align 8 */,
uint8_t *src /* align 1 */, const uint8_t *src /* align 1 */,
ptrdiff_t stride, int h, int x, int y) ptrdiff_t stride, int h, int x, int y)
{ {
const int A = (8 - x) * (8 - y); const int A = (8 - x) * (8 - y);
@ -861,7 +861,7 @@ static void avg_no_rnd_vc1_chroma_mc8_c(uint8_t *dst /* align 8 */,
} }
static void avg_no_rnd_vc1_chroma_mc4_c(uint8_t *dst /* align 8 */, static void avg_no_rnd_vc1_chroma_mc4_c(uint8_t *dst /* align 8 */,
uint8_t *src /* align 1 */, const uint8_t *src /* align 1 */,
ptrdiff_t stride, int h, int x, int y) ptrdiff_t stride, int h, int x, int y)
{ {
const int A = (8 - x) * (8 - y); const int A = (8 - x) * (8 - y);

2
libavcodec/x86/h264_chromamc.asm
View file

@ -104,7 +104,7 @@ SECTION .text
%define extra_regs 0 %define extra_regs 0
%endif ; rv40 %endif ; rv40
; void ff_put/avg_h264_chroma_mc8_*(uint8_t *dst /* align 8 */, ; void ff_put/avg_h264_chroma_mc8_*(uint8_t *dst /* align 8 */,
; uint8_t *src /* align 1 */, ; const uint8_t *src /* align 1 */,
; ptrdiff_t stride, int h, int mx, int my) ; ptrdiff_t stride, int h, int mx, int my)
cglobal %1_%2_chroma_mc8%3, 6, 7 + extra_regs, 0 cglobal %1_%2_chroma_mc8%3, 6, 7 + extra_regs, 0
mov r6d, r5d mov r6d, r5d

4
libavcodec/x86/h264_chromamc_10bit.asm
View file

@ -57,7 +57,7 @@ SECTION .text
%endmacro %endmacro
;----------------------------------------------------------------------------- ;-----------------------------------------------------------------------------
; void ff_put/avg_h264_chroma_mc8(pixel *dst, pixel *src, ptrdiff_t stride, ; void ff_put/avg_h264_chroma_mc8(pixel *dst, const pixel *src, ptrdiff_t stride,
; int h, int mx, int my) ; int h, int mx, int my)
;----------------------------------------------------------------------------- ;-----------------------------------------------------------------------------
%macro CHROMA_MC8 1 %macro CHROMA_MC8 1
@ -198,7 +198,7 @@ cglobal %1_h264_chroma_mc4_10, 6,6,7
%endmacro %endmacro
;----------------------------------------------------------------------------- ;-----------------------------------------------------------------------------
; void ff_put/avg_h264_chroma_mc2(pixel *dst, pixel *src, ptrdiff_t stride, ; void ff_put/avg_h264_chroma_mc2(pixel *dst, const pixel *src, ptrdiff_t stride,
; int h, int mx, int my) ; int h, int mx, int my)
;----------------------------------------------------------------------------- ;-----------------------------------------------------------------------------
%macro CHROMA_MC2 1 %macro CHROMA_MC2 1

24
libavcodec/x86/h264chroma_init.c
View file

@ -24,35 +24,35 @@
#include "libavutil/x86/cpu.h" #include "libavutil/x86/cpu.h"
#include "libavcodec/h264chroma.h" #include "libavcodec/h264chroma.h"
void ff_put_h264_chroma_mc8_rnd_mmx (uint8_t *dst, uint8_t *src, void ff_put_h264_chroma_mc8_rnd_mmx (uint8_t *dst, const uint8_t *src,
ptrdiff_t stride, int h, int x, int y); ptrdiff_t stride, int h, int x, int y);
void ff_avg_h264_chroma_mc8_rnd_mmxext(uint8_t *dst, uint8_t *src, void ff_avg_h264_chroma_mc8_rnd_mmxext(uint8_t *dst, const uint8_t *src,
ptrdiff_t stride, int h, int x, int y); ptrdiff_t stride, int h, int x, int y);
void ff_put_h264_chroma_mc4_mmx (uint8_t *dst, uint8_t *src, void ff_put_h264_chroma_mc4_mmx (uint8_t *dst, const uint8_t *src,
ptrdiff_t stride, int h, int x, int y); ptrdiff_t stride, int h, int x, int y);
void ff_avg_h264_chroma_mc4_mmxext (uint8_t *dst, uint8_t *src, void ff_avg_h264_chroma_mc4_mmxext (uint8_t *dst, const uint8_t *src,
ptrdiff_t stride, int h, int x, int y); ptrdiff_t stride, int h, int x, int y);
void ff_put_h264_chroma_mc2_mmxext (uint8_t *dst, uint8_t *src, void ff_put_h264_chroma_mc2_mmxext (uint8_t *dst, const uint8_t *src,
ptrdiff_t stride, int h, int x, int y); ptrdiff_t stride, int h, int x, int y);
void ff_avg_h264_chroma_mc2_mmxext (uint8_t *dst, uint8_t *src, void ff_avg_h264_chroma_mc2_mmxext (uint8_t *dst, const uint8_t *src,
ptrdiff_t stride, int h, int x, int y); ptrdiff_t stride, int h, int x, int y);
void ff_put_h264_chroma_mc8_rnd_ssse3(uint8_t *dst, uint8_t *src, void ff_put_h264_chroma_mc8_rnd_ssse3(uint8_t *dst, const uint8_t *src,
ptrdiff_t stride, int h, int x, int y); ptrdiff_t stride, int h, int x, int y);
void ff_put_h264_chroma_mc4_ssse3 (uint8_t *dst, uint8_t *src, void ff_put_h264_chroma_mc4_ssse3 (uint8_t *dst, const uint8_t *src,
ptrdiff_t stride, int h, int x, int y); ptrdiff_t stride, int h, int x, int y);
void ff_avg_h264_chroma_mc8_rnd_ssse3(uint8_t *dst, uint8_t *src, void ff_avg_h264_chroma_mc8_rnd_ssse3(uint8_t *dst, const uint8_t *src,
ptrdiff_t stride, int h, int x, int y); ptrdiff_t stride, int h, int x, int y);
void ff_avg_h264_chroma_mc4_ssse3 (uint8_t *dst, uint8_t *src, void ff_avg_h264_chroma_mc4_ssse3 (uint8_t *dst, const uint8_t *src,
ptrdiff_t stride, int h, int x, int y); ptrdiff_t stride, int h, int x, int y);
#define CHROMA_MC(OP, NUM, DEPTH, OPT) \ #define CHROMA_MC(OP, NUM, DEPTH, OPT) \
void ff_ ## OP ## _h264_chroma_mc ## NUM ## _ ## DEPTH ## _ ## OPT \ void ff_ ## OP ## _h264_chroma_mc ## NUM ## _ ## DEPTH ## _ ## OPT \
(uint8_t *dst, uint8_t *src, \ (uint8_t *dst, const uint8_t *src, \
ptrdiff_t stride, int h, int x, int y); ptrdiff_t stride, int h, int x, int y);
CHROMA_MC(put, 2, 10, mmxext) CHROMA_MC(put, 2, 10, mmxext)
CHROMA_MC(avg, 2, 10, mmxext) CHROMA_MC(avg, 2, 10, mmxext)

8
libavcodec/x86/rv40dsp_init.c
View file

@ -40,14 +40,14 @@ static void op##_rv40_qpel##size##_mc33_##insn(uint8_t *dst, const uint8_t *src,
} }
#if HAVE_X86ASM #if HAVE_X86ASM
void ff_put_rv40_chroma_mc8_mmx (uint8_t *dst, uint8_t *src, void ff_put_rv40_chroma_mc8_mmx (uint8_t *dst, const uint8_t *src,
ptrdiff_t stride, int h, int x, int y); ptrdiff_t stride, int h, int x, int y);
void ff_avg_rv40_chroma_mc8_mmxext(uint8_t *dst, uint8_t *src, void ff_avg_rv40_chroma_mc8_mmxext(uint8_t *dst, const uint8_t *src,
ptrdiff_t stride, int h, int x, int y); ptrdiff_t stride, int h, int x, int y);
void ff_put_rv40_chroma_mc4_mmx (uint8_t *dst, uint8_t *src, void ff_put_rv40_chroma_mc4_mmx (uint8_t *dst, const uint8_t *src,
ptrdiff_t stride, int h, int x, int y); ptrdiff_t stride, int h, int x, int y);
void ff_avg_rv40_chroma_mc4_mmxext(uint8_t *dst, uint8_t *src, void ff_avg_rv40_chroma_mc4_mmxext(uint8_t *dst, const uint8_t *src,
ptrdiff_t stride, int h, int x, int y); ptrdiff_t stride, int h, int x, int y);
#define DECLARE_WEIGHT(opt) \ #define DECLARE_WEIGHT(opt) \

8
libavcodec/x86/vc1dsp_init.c
View file

@ -80,13 +80,13 @@ DECLARE_FUNCTION(avg_, 16, _sse2)
#endif /* HAVE_X86ASM */ #endif /* HAVE_X86ASM */
void ff_put_vc1_chroma_mc8_nornd_mmx (uint8_t *dst, uint8_t *src, void ff_put_vc1_chroma_mc8_nornd_mmx (uint8_t *dst, const uint8_t *src,
ptrdiff_t stride, int h, int x, int y); ptrdiff_t stride, int h, int x, int y);
void ff_avg_vc1_chroma_mc8_nornd_mmxext(uint8_t *dst, uint8_t *src, void ff_avg_vc1_chroma_mc8_nornd_mmxext(uint8_t *dst, const uint8_t *src,
ptrdiff_t stride, int h, int x, int y); ptrdiff_t stride, int h, int x, int y);
void ff_put_vc1_chroma_mc8_nornd_ssse3(uint8_t *dst, uint8_t *src, void ff_put_vc1_chroma_mc8_nornd_ssse3(uint8_t *dst, const uint8_t *src,
ptrdiff_t stride, int h, int x, int y); ptrdiff_t stride, int h, int x, int y);
void ff_avg_vc1_chroma_mc8_nornd_ssse3(uint8_t *dst, uint8_t *src, void ff_avg_vc1_chroma_mc8_nornd_ssse3(uint8_t *dst, const uint8_t *src,
ptrdiff_t stride, int h, int x, int y); ptrdiff_t stride, int h, int x, int y);
void ff_vc1_inv_trans_4x4_dc_mmxext(uint8_t *dest, ptrdiff_t linesize, void ff_vc1_inv_trans_4x4_dc_mmxext(uint8_t *dest, ptrdiff_t linesize,
int16_t *block); int16_t *block);