Index: source/config/mac/ia32/vp9_rtcd.h |
=================================================================== |
--- source/config/mac/ia32/vp9_rtcd.h (revision 264025) |
+++ source/config/mac/ia32/vp9_rtcd.h (working copy) |
@@ -44,7 +44,6 @@ |
void vp9_convolve8_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h); |
void vp9_convolve8_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h); |
void vp9_convolve8_ssse3(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h); |
-void vp9_convolve8_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h); |
RTCD_EXTERN void (*vp9_convolve8)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h); |
void vp9_convolve8_avg_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h); |
@@ -65,13 +64,11 @@ |
void vp9_convolve8_horiz_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h); |
void vp9_convolve8_horiz_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h); |
void vp9_convolve8_horiz_ssse3(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h); |
-void vp9_convolve8_horiz_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h); |
RTCD_EXTERN void (*vp9_convolve8_horiz)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h); |
void vp9_convolve8_vert_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h); |
void vp9_convolve8_vert_sse2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h); |
void vp9_convolve8_vert_ssse3(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h); |
-void vp9_convolve8_vert_avx2(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h); |
RTCD_EXTERN void (*vp9_convolve8_vert)(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h); |
void vp9_convolve_avg_c(const uint8_t *src, ptrdiff_t src_stride, uint8_t *dst, ptrdiff_t dst_stride, const int16_t *filter_x, int x_step_q4, const int16_t *filter_y, int y_step_q4, int w, int h); |
@@ -206,42 +203,34 @@ |
void vp9_fdct16x16_c(const int16_t *input, int16_t *output, int stride); |
void vp9_fdct16x16_sse2(const int16_t *input, int16_t *output, int stride); |
-void vp9_fdct16x16_avx2(const int16_t *input, int16_t *output, int stride); |
RTCD_EXTERN void (*vp9_fdct16x16)(const int16_t *input, int16_t *output, int stride); |
void vp9_fdct32x32_c(const int16_t *input, int16_t *output, int stride); |
void vp9_fdct32x32_sse2(const int16_t *input, int16_t *output, int stride); |
-void vp9_fdct32x32_avx2(const int16_t *input, int16_t *output, int stride); |
RTCD_EXTERN void (*vp9_fdct32x32)(const int16_t *input, int16_t *output, int stride); |
void vp9_fdct32x32_rd_c(const int16_t *input, int16_t *output, int stride); |
void vp9_fdct32x32_rd_sse2(const int16_t *input, int16_t *output, int stride); |
-void vp9_fdct32x32_rd_avx2(const int16_t *input, int16_t *output, int stride); |
RTCD_EXTERN void (*vp9_fdct32x32_rd)(const int16_t *input, int16_t *output, int stride); |
void vp9_fdct4x4_c(const int16_t *input, int16_t *output, int stride); |
void vp9_fdct4x4_sse2(const int16_t *input, int16_t *output, int stride); |
-void vp9_fdct4x4_avx2(const int16_t *input, int16_t *output, int stride); |
RTCD_EXTERN void (*vp9_fdct4x4)(const int16_t *input, int16_t *output, int stride); |
void vp9_fdct8x8_c(const int16_t *input, int16_t *output, int stride); |
void vp9_fdct8x8_sse2(const int16_t *input, int16_t *output, int stride); |
-void vp9_fdct8x8_avx2(const int16_t *input, int16_t *output, int stride); |
RTCD_EXTERN void (*vp9_fdct8x8)(const int16_t *input, int16_t *output, int stride); |
void vp9_fht16x16_c(const int16_t *input, int16_t *output, int stride, int tx_type); |
void vp9_fht16x16_sse2(const int16_t *input, int16_t *output, int stride, int tx_type); |
-void vp9_fht16x16_avx2(const int16_t *input, int16_t *output, int stride, int tx_type); |
RTCD_EXTERN void (*vp9_fht16x16)(const int16_t *input, int16_t *output, int stride, int tx_type); |
void vp9_fht4x4_c(const int16_t *input, int16_t *output, int stride, int tx_type); |
void vp9_fht4x4_sse2(const int16_t *input, int16_t *output, int stride, int tx_type); |
-void vp9_fht4x4_avx2(const int16_t *input, int16_t *output, int stride, int tx_type); |
RTCD_EXTERN void (*vp9_fht4x4)(const int16_t *input, int16_t *output, int stride, int tx_type); |
void vp9_fht8x8_c(const int16_t *input, int16_t *output, int stride, int tx_type); |
void vp9_fht8x8_sse2(const int16_t *input, int16_t *output, int stride, int tx_type); |
-void vp9_fht8x8_avx2(const int16_t *input, int16_t *output, int stride, int tx_type); |
RTCD_EXTERN void (*vp9_fht8x8)(const int16_t *input, int16_t *output, int stride, int tx_type); |
int vp9_full_range_search_c(const struct macroblock *x, struct mv *ref_mv, struct mv *best_mv, int search_param, int sad_per_bit, int *num00, const struct vp9_variance_vtable *fn_ptr, DEC_MVCOSTS, const struct mv *center_mv); |
@@ -340,7 +329,6 @@ |
void vp9_lpf_horizontal_16_c(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int count); |
void vp9_lpf_horizontal_16_sse2(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int count); |
-void vp9_lpf_horizontal_16_avx2(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int count); |
RTCD_EXTERN void (*vp9_lpf_horizontal_16)(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int count); |
void vp9_lpf_horizontal_4_c(uint8_t *s, int pitch, const uint8_t *blimit, const uint8_t *limit, const uint8_t *thresh, int count); |
@@ -475,7 +463,6 @@ |
void vp9_sad32x32x4d_c(const uint8_t *src_ptr, int src_stride, const uint8_t* const ref_ptr[], int ref_stride, unsigned int *sad_array); |
void vp9_sad32x32x4d_sse2(const uint8_t *src_ptr, int src_stride, const uint8_t* const ref_ptr[], int ref_stride, unsigned int *sad_array); |
-void vp9_sad32x32x4d_avx2(const uint8_t *src_ptr, int src_stride, const uint8_t* const ref_ptr[], int ref_stride, unsigned int *sad_array); |
RTCD_EXTERN void (*vp9_sad32x32x4d)(const uint8_t *src_ptr, int src_stride, const uint8_t* const ref_ptr[], int ref_stride, unsigned int *sad_array); |
void vp9_sad32x32x8_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, uint32_t *sad_array); |
@@ -543,7 +530,6 @@ |
void vp9_sad64x64x4d_c(const uint8_t *src_ptr, int src_stride, const uint8_t* const ref_ptr[], int ref_stride, unsigned int *sad_array); |
void vp9_sad64x64x4d_sse2(const uint8_t *src_ptr, int src_stride, const uint8_t* const ref_ptr[], int ref_stride, unsigned int *sad_array); |
-void vp9_sad64x64x4d_avx2(const uint8_t *src_ptr, int src_stride, const uint8_t* const ref_ptr[], int ref_stride, unsigned int *sad_array); |
RTCD_EXTERN void (*vp9_sad64x64x4d)(const uint8_t *src_ptr, int src_stride, const uint8_t* const ref_ptr[], int ref_stride, unsigned int *sad_array); |
void vp9_sad64x64x8_c(const uint8_t *src_ptr, int src_stride, const uint8_t *ref_ptr, int ref_stride, uint32_t *sad_array); |
@@ -611,8 +597,7 @@ |
#define vp9_sub_pixel_avg_variance32x16 vp9_sub_pixel_avg_variance32x16_c |
unsigned int vp9_sub_pixel_avg_variance32x32_c(const uint8_t *src_ptr, int source_stride, int xoffset, int yoffset, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse, const uint8_t *second_pred); |
-unsigned int vp9_sub_pixel_avg_variance32x32_avx2(const uint8_t *src_ptr, int source_stride, int xoffset, int yoffset, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse, const uint8_t *second_pred); |
-RTCD_EXTERN unsigned int (*vp9_sub_pixel_avg_variance32x32)(const uint8_t *src_ptr, int source_stride, int xoffset, int yoffset, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse, const uint8_t *second_pred); |
+#define vp9_sub_pixel_avg_variance32x32 vp9_sub_pixel_avg_variance32x32_c |
unsigned int vp9_sub_pixel_avg_variance32x64_c(const uint8_t *src_ptr, int source_stride, int xoffset, int yoffset, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse, const uint8_t *second_pred); |
#define vp9_sub_pixel_avg_variance32x64 vp9_sub_pixel_avg_variance32x64_c |
@@ -627,8 +612,7 @@ |
#define vp9_sub_pixel_avg_variance64x32 vp9_sub_pixel_avg_variance64x32_c |
unsigned int vp9_sub_pixel_avg_variance64x64_c(const uint8_t *src_ptr, int source_stride, int xoffset, int yoffset, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse, const uint8_t *second_pred); |
-unsigned int vp9_sub_pixel_avg_variance64x64_avx2(const uint8_t *src_ptr, int source_stride, int xoffset, int yoffset, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse, const uint8_t *second_pred); |
-RTCD_EXTERN unsigned int (*vp9_sub_pixel_avg_variance64x64)(const uint8_t *src_ptr, int source_stride, int xoffset, int yoffset, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse, const uint8_t *second_pred); |
+#define vp9_sub_pixel_avg_variance64x64 vp9_sub_pixel_avg_variance64x64_c |
unsigned int vp9_sub_pixel_avg_variance8x16_c(const uint8_t *src_ptr, int source_stride, int xoffset, int yoffset, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse, const uint8_t *second_pred); |
#define vp9_sub_pixel_avg_variance8x16 vp9_sub_pixel_avg_variance8x16_c |
@@ -658,8 +642,7 @@ |
#define vp9_sub_pixel_variance32x16 vp9_sub_pixel_variance32x16_c |
unsigned int vp9_sub_pixel_variance32x32_c(const uint8_t *src_ptr, int source_stride, int xoffset, int yoffset, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); |
-unsigned int vp9_sub_pixel_variance32x32_avx2(const uint8_t *src_ptr, int source_stride, int xoffset, int yoffset, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); |
-RTCD_EXTERN unsigned int (*vp9_sub_pixel_variance32x32)(const uint8_t *src_ptr, int source_stride, int xoffset, int yoffset, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); |
+#define vp9_sub_pixel_variance32x32 vp9_sub_pixel_variance32x32_c |
unsigned int vp9_sub_pixel_variance32x64_c(const uint8_t *src_ptr, int source_stride, int xoffset, int yoffset, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); |
#define vp9_sub_pixel_variance32x64 vp9_sub_pixel_variance32x64_c |
@@ -674,8 +657,7 @@ |
#define vp9_sub_pixel_variance64x32 vp9_sub_pixel_variance64x32_c |
unsigned int vp9_sub_pixel_variance64x64_c(const uint8_t *src_ptr, int source_stride, int xoffset, int yoffset, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); |
-unsigned int vp9_sub_pixel_variance64x64_avx2(const uint8_t *src_ptr, int source_stride, int xoffset, int yoffset, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); |
-RTCD_EXTERN unsigned int (*vp9_sub_pixel_variance64x64)(const uint8_t *src_ptr, int source_stride, int xoffset, int yoffset, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); |
+#define vp9_sub_pixel_variance64x64 vp9_sub_pixel_variance64x64_c |
unsigned int vp9_sub_pixel_variance8x16_c(const uint8_t *src_ptr, int source_stride, int xoffset, int yoffset, const uint8_t *ref_ptr, int ref_stride, unsigned int *sse); |
#define vp9_sub_pixel_variance8x16 vp9_sub_pixel_variance8x16_c |
@@ -801,7 +783,6 @@ |
vp9_convolve8 = vp9_convolve8_c; |
if (flags & HAS_SSE2) vp9_convolve8 = vp9_convolve8_sse2; |
if (flags & HAS_SSSE3) vp9_convolve8 = vp9_convolve8_ssse3; |
- if (flags & HAS_AVX2) vp9_convolve8 = vp9_convolve8_avx2; |
vp9_convolve8_avg = vp9_convolve8_avg_c; |
if (flags & HAS_SSE2) vp9_convolve8_avg = vp9_convolve8_avg_sse2; |
if (flags & HAS_SSSE3) vp9_convolve8_avg = vp9_convolve8_avg_ssse3; |
@@ -814,37 +795,27 @@ |
vp9_convolve8_horiz = vp9_convolve8_horiz_c; |
if (flags & HAS_SSE2) vp9_convolve8_horiz = vp9_convolve8_horiz_sse2; |
if (flags & HAS_SSSE3) vp9_convolve8_horiz = vp9_convolve8_horiz_ssse3; |
- if (flags & HAS_AVX2) vp9_convolve8_horiz = vp9_convolve8_horiz_avx2; |
vp9_convolve8_vert = vp9_convolve8_vert_c; |
if (flags & HAS_SSE2) vp9_convolve8_vert = vp9_convolve8_vert_sse2; |
if (flags & HAS_SSSE3) vp9_convolve8_vert = vp9_convolve8_vert_ssse3; |
- if (flags & HAS_AVX2) vp9_convolve8_vert = vp9_convolve8_vert_avx2; |
vp9_diamond_search_sad = vp9_diamond_search_sad_c; |
if (flags & HAS_SSE3) vp9_diamond_search_sad = vp9_diamond_search_sadx4; |
vp9_fdct16x16 = vp9_fdct16x16_c; |
if (flags & HAS_SSE2) vp9_fdct16x16 = vp9_fdct16x16_sse2; |
- if (flags & HAS_AVX2) vp9_fdct16x16 = vp9_fdct16x16_avx2; |
vp9_fdct32x32 = vp9_fdct32x32_c; |
if (flags & HAS_SSE2) vp9_fdct32x32 = vp9_fdct32x32_sse2; |
- if (flags & HAS_AVX2) vp9_fdct32x32 = vp9_fdct32x32_avx2; |
vp9_fdct32x32_rd = vp9_fdct32x32_rd_c; |
if (flags & HAS_SSE2) vp9_fdct32x32_rd = vp9_fdct32x32_rd_sse2; |
- if (flags & HAS_AVX2) vp9_fdct32x32_rd = vp9_fdct32x32_rd_avx2; |
vp9_fdct4x4 = vp9_fdct4x4_c; |
if (flags & HAS_SSE2) vp9_fdct4x4 = vp9_fdct4x4_sse2; |
- if (flags & HAS_AVX2) vp9_fdct4x4 = vp9_fdct4x4_avx2; |
vp9_fdct8x8 = vp9_fdct8x8_c; |
if (flags & HAS_SSE2) vp9_fdct8x8 = vp9_fdct8x8_sse2; |
- if (flags & HAS_AVX2) vp9_fdct8x8 = vp9_fdct8x8_avx2; |
vp9_fht16x16 = vp9_fht16x16_c; |
if (flags & HAS_SSE2) vp9_fht16x16 = vp9_fht16x16_sse2; |
- if (flags & HAS_AVX2) vp9_fht16x16 = vp9_fht16x16_avx2; |
vp9_fht4x4 = vp9_fht4x4_c; |
if (flags & HAS_SSE2) vp9_fht4x4 = vp9_fht4x4_sse2; |
- if (flags & HAS_AVX2) vp9_fht4x4 = vp9_fht4x4_avx2; |
vp9_fht8x8 = vp9_fht8x8_c; |
if (flags & HAS_SSE2) vp9_fht8x8 = vp9_fht8x8_sse2; |
- if (flags & HAS_AVX2) vp9_fht8x8 = vp9_fht8x8_avx2; |
vp9_full_search_sad = vp9_full_search_sad_c; |
if (flags & HAS_SSE3) vp9_full_search_sad = vp9_full_search_sadx3; |
if (flags & HAS_SSE4_1) vp9_full_search_sad = vp9_full_search_sadx8; |
@@ -883,7 +854,6 @@ |
if (flags & HAS_SSE2) vp9_iht8x8_64_add = vp9_iht8x8_64_add_sse2; |
vp9_lpf_horizontal_16 = vp9_lpf_horizontal_16_c; |
if (flags & HAS_SSE2) vp9_lpf_horizontal_16 = vp9_lpf_horizontal_16_sse2; |
- if (flags & HAS_AVX2) vp9_lpf_horizontal_16 = vp9_lpf_horizontal_16_avx2; |
vp9_lpf_horizontal_4 = vp9_lpf_horizontal_4_c; |
if (flags & HAS_MMX) vp9_lpf_horizontal_4 = vp9_lpf_horizontal_4_mmx; |
vp9_lpf_horizontal_4_dual = vp9_lpf_horizontal_4_dual_c; |
@@ -928,7 +898,6 @@ |
if (flags & HAS_SSE2) vp9_sad32x16x4d = vp9_sad32x16x4d_sse2; |
vp9_sad32x32x4d = vp9_sad32x32x4d_c; |
if (flags & HAS_SSE2) vp9_sad32x32x4d = vp9_sad32x32x4d_sse2; |
- if (flags & HAS_AVX2) vp9_sad32x32x4d = vp9_sad32x32x4d_avx2; |
vp9_sad32x64x4d = vp9_sad32x64x4d_c; |
if (flags & HAS_SSE2) vp9_sad32x64x4d = vp9_sad32x64x4d_sse2; |
vp9_sad4x4 = vp9_sad4x4_c; |
@@ -943,7 +912,6 @@ |
if (flags & HAS_SSE2) vp9_sad64x32x4d = vp9_sad64x32x4d_sse2; |
vp9_sad64x64x4d = vp9_sad64x64x4d_c; |
if (flags & HAS_SSE2) vp9_sad64x64x4d = vp9_sad64x64x4d_sse2; |
- if (flags & HAS_AVX2) vp9_sad64x64x4d = vp9_sad64x64x4d_avx2; |
vp9_sad8x16 = vp9_sad8x16_c; |
if (flags & HAS_MMX) vp9_sad8x16 = vp9_sad8x16_mmx; |
vp9_sad8x16x3 = vp9_sad8x16x3_c; |
@@ -958,14 +926,6 @@ |
if (flags & HAS_SSE3) vp9_sad8x8x3 = vp9_sad8x8x3_sse3; |
vp9_sad8x8x4d = vp9_sad8x8x4d_c; |
if (flags & HAS_SSE2) vp9_sad8x8x4d = vp9_sad8x8x4d_sse2; |
- vp9_sub_pixel_avg_variance32x32 = vp9_sub_pixel_avg_variance32x32_c; |
- if (flags & HAS_AVX2) vp9_sub_pixel_avg_variance32x32 = vp9_sub_pixel_avg_variance32x32_avx2; |
- vp9_sub_pixel_avg_variance64x64 = vp9_sub_pixel_avg_variance64x64_c; |
- if (flags & HAS_AVX2) vp9_sub_pixel_avg_variance64x64 = vp9_sub_pixel_avg_variance64x64_avx2; |
- vp9_sub_pixel_variance32x32 = vp9_sub_pixel_variance32x32_c; |
- if (flags & HAS_AVX2) vp9_sub_pixel_variance32x32 = vp9_sub_pixel_variance32x32_avx2; |
- vp9_sub_pixel_variance64x64 = vp9_sub_pixel_variance64x64_c; |
- if (flags & HAS_AVX2) vp9_sub_pixel_variance64x64 = vp9_sub_pixel_variance64x64_avx2; |
vp9_temporal_filter_apply = vp9_temporal_filter_apply_c; |
if (flags & HAS_SSE2) vp9_temporal_filter_apply = vp9_temporal_filter_apply_sse2; |
vp9_variance16x16 = vp9_variance16x16_c; |