diff options
Diffstat (limited to 'vp8')
41 files changed, 1154 insertions, 1268 deletions
diff --git a/vp8/common/arm/arm_systemdependent.c b/vp8/common/arm/arm_systemdependent.c index 03afd66a8..0147e24a9 100644 --- a/vp8/common/arm/arm_systemdependent.c +++ b/vp8/common/arm/arm_systemdependent.c @@ -11,7 +11,6 @@ #include "vpx_ports/config.h" #include "vpx_ports/arm.h" -#include "vp8/common/g_common.h" #include "vp8/common/pragmas.h" #include "vp8/common/subpixel.h" #include "vp8/common/loopfilter.h" diff --git a/vp8/common/debugmodes.c b/vp8/common/debugmodes.c index ce01ce9af..5cd84c51b 100644 --- a/vp8/common/debugmodes.c +++ b/vp8/common/debugmodes.c @@ -8,13 +8,11 @@ * be found in the AUTHORS file in the root of the source tree. */ - #include <stdio.h> #include "blockd.h" - -void vp9_print_modes_and_motion_vectors(MODE_INFO *mi, int rows, int cols, int frame) { - +void vp9_print_modes_and_motion_vectors(MODE_INFO *mi, int rows, int cols, + int frame) { int mb_row; int mb_col; int mb_index = 0; @@ -120,7 +118,6 @@ void vp9_print_modes_and_motion_vectors(MODE_INFO *mi, int rows, int cols, int f fprintf(mvs, "\n"); - /* print out the block modes */ mb_index = 0; fprintf(mvs, "MVs for Frame %d\n", frame); @@ -145,6 +142,5 @@ void vp9_print_modes_and_motion_vectors(MODE_INFO *mi, int rows, int cols, int f } fprintf(mvs, "\n"); - fclose(mvs); } diff --git a/vp8/common/entropymode.c b/vp8/common/entropymode.c index 67011c1ae..f6921c732 100644 --- a/vp8/common/entropymode.c +++ b/vp8/common/entropymode.c @@ -111,28 +111,23 @@ const vp9_prob vp9_sub_mv_ref_prob2 [SUBMVREF_COUNT][VP9_SUBMVREFS - 1] = { { 208, 1, 1 } }; - - vp9_mbsplit vp9_mbsplits [VP9_NUMMBSPLITS] = { { 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, - }, - { + }, { 0, 0, 1, 1, 0, 0, 1, 1, 0, 0, 1, 1, 0, 0, 1, 1, - }, - { + }, { 0, 0, 1, 1, 0, 0, 1, 1, 2, 2, 3, 3, 2, 2, 3, 3, - }, - { + }, { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, @@ -144,7 +139,6 @@ const int vp9_mbsplit_count [VP9_NUMMBSPLITS] = { 2, 2, 4, 16}; const vp9_prob vp9_mbsplit_probs [VP9_NUMMBSPLITS - 1] = { 110, 111, 150}; - /* Array indices are identical to previously-existing INTRAMODECONTEXTNODES. */ const vp9_tree_index vp9_bmode_tree[VP9_BINTRAMODES * 2 - 2] = /* INTRAMODECONTEXTNODE value */ @@ -241,7 +235,6 @@ const vp9_tree_index vp9_sub_mv_ref_tree[6] = { -ZERO4X4, -NEW4X4 }; - struct vp9_token_struct vp9_bmode_encodings [VP9_BINTRAMODES]; struct vp9_token_struct vp9_ymode_encodings [VP9_YMODES]; #if CONFIG_SUPERBLOCKS @@ -258,48 +251,45 @@ struct vp9_token_struct vp9_sb_mv_ref_encoding_array [VP9_MVREFS]; #endif struct vp9_token_struct vp9_sub_mv_ref_encoding_array [VP9_SUBMVREFS]; - - void vp9_init_mbmode_probs(VP9_COMMON *x) { unsigned int bct [VP9_YMODES] [2]; /* num Ymodes > num UV modes */ vp9_tree_probs_from_distribution(VP9_YMODES, vp9_ymode_encodings, - vp9_ymode_tree, x->fc.ymode_prob, bct, y_mode_cts, 256, 1); + vp9_ymode_tree, x->fc.ymode_prob, + bct, y_mode_cts, 256, 1); { int i; for (i = 0; i < 8; i++) { - vp9_tree_probs_from_distribution( - VP9_YMODES, vp9_kf_ymode_encodings, vp9_kf_ymode_tree, - x->kf_ymode_prob[i], bct, kf_y_mode_cts[i], - 256, 1); + vp9_tree_probs_from_distribution(VP9_YMODES, vp9_kf_ymode_encodings, + vp9_kf_ymode_tree, x->kf_ymode_prob[i], + bct, kf_y_mode_cts[i], 256, 1); #if CONFIG_SUPERBLOCKS - vp9_tree_probs_from_distribution( - VP9_I32X32_MODES, vp9_sb_kf_ymode_encodings, vp9_sb_ymode_tree, - x->sb_kf_ymode_prob[i], bct, kf_y_mode_cts[i], - 256, 1); + vp9_tree_probs_from_distribution(VP9_I32X32_MODES, + vp9_sb_kf_ymode_encodings, + vp9_sb_ymode_tree, + x->sb_kf_ymode_prob[i], bct, + kf_y_mode_cts[i], 256, 1); #endif } } { int i; for (i = 0; i < VP9_YMODES; i++) { - vp9_tree_probs_from_distribution( - VP9_UV_MODES, vp9_uv_mode_encodings, vp9_uv_mode_tree, - x->kf_uv_mode_prob[i], bct, kf_uv_mode_cts[i], - 256, 1); - vp9_tree_probs_from_distribution( - VP9_UV_MODES, vp9_uv_mode_encodings, vp9_uv_mode_tree, - x->fc.uv_mode_prob[i], bct, uv_mode_cts[i], - 256, 1); + vp9_tree_probs_from_distribution(VP9_UV_MODES, vp9_uv_mode_encodings, + vp9_uv_mode_tree, x->kf_uv_mode_prob[i], + bct, kf_uv_mode_cts[i], 256, 1); + vp9_tree_probs_from_distribution(VP9_UV_MODES, vp9_uv_mode_encodings, + vp9_uv_mode_tree, x->fc.uv_mode_prob[i], + bct, uv_mode_cts[i], 256, 1); } } - vp9_tree_probs_from_distribution( - VP9_I8X8_MODES, vp9_i8x8_mode_encodings, vp9_i8x8_mode_tree, - x->fc.i8x8_mode_prob, bct, i8x8_mode_cts, - 256, 1); + vp9_tree_probs_from_distribution(VP9_I8X8_MODES, vp9_i8x8_mode_encodings, + vp9_i8x8_mode_tree, x->fc.i8x8_mode_prob, + bct, i8x8_mode_cts, 256, 1); - vpx_memcpy(x->fc.sub_mv_ref_prob, vp9_sub_mv_ref_prob2, sizeof(vp9_sub_mv_ref_prob2)); + vpx_memcpy(x->fc.sub_mv_ref_prob, vp9_sub_mv_ref_prob2, + sizeof(vp9_sub_mv_ref_prob2)); vpx_memcpy(x->fc.mbsplit_prob, vp9_mbsplit_probs, sizeof(vp9_mbsplit_probs)); vpx_memcpy(x->fc.switchable_interp_prob, vp9_switchable_interp_prob, sizeof(vp9_switchable_interp_prob)); @@ -311,7 +301,8 @@ static void intra_bmode_probs_from_distribution( unsigned int branch_ct [VP9_BINTRAMODES - 1] [2], const unsigned int events [VP9_BINTRAMODES]) { vp9_tree_probs_from_distribution(VP9_BINTRAMODES, vp9_bmode_encodings, - vp9_bmode_tree, p, branch_ct, events, 256, 1); + vp9_bmode_tree, p, branch_ct, + events, 256, 1); } void vp9_default_bmode_probs(vp9_prob p [VP9_BINTRAMODES - 1]) { @@ -319,20 +310,17 @@ void vp9_default_bmode_probs(vp9_prob p [VP9_BINTRAMODES - 1]) { intra_bmode_probs_from_distribution(p, branch_ct, bmode_cts); } -void vp9_kf_default_bmode_probs(vp9_prob p [VP9_BINTRAMODES] [VP9_BINTRAMODES] [VP9_BINTRAMODES - 1]) { - unsigned int branch_ct [VP9_BINTRAMODES - 1] [2]; - - int i = 0; - - do { - int j = 0; +void vp9_kf_default_bmode_probs(vp9_prob p[VP9_BINTRAMODES][VP9_BINTRAMODES] + [VP9_BINTRAMODES - 1]) { + unsigned int branch_ct[VP9_BINTRAMODES - 1][2]; + int i, j; - do { + for (i = 0; i < VP9_BINTRAMODES; i++) { + for (j = 0; j < VP9_BINTRAMODES; j++) { intra_bmode_probs_from_distribution( p[i][j], branch_ct, vp9_kf_default_bmode_counts[i][j]); - - } while (++j < VP9_BINTRAMODES); - } while (++i < VP9_BINTRAMODES); + } + } } #if VP9_SWITCHABLE_FILTERS == 3 @@ -548,10 +536,9 @@ void vp9_adapt_mode_probs(VP9_COMMON *cm) { else cm->fc.ymode_prob[t] = prob; } for (i = 0; i < VP9_YMODES; ++i) { - vp9_tree_probs_from_distribution( - VP9_UV_MODES, vp9_uv_mode_encodings, vp9_uv_mode_tree, - uvmode_probs, branch_ct, cm->fc.uv_mode_counts[i], - 256, 1); + vp9_tree_probs_from_distribution(VP9_UV_MODES, vp9_uv_mode_encodings, + vp9_uv_mode_tree, uvmode_probs, branch_ct, + cm->fc.uv_mode_counts[i], 256, 1); for (t = 0; t < VP9_UV_MODES - 1; ++t) { int prob; count = branch_ct[t][0] + branch_ct[t][1]; @@ -564,10 +551,9 @@ void vp9_adapt_mode_probs(VP9_COMMON *cm) { else cm->fc.uv_mode_prob[i][t] = prob; } } - vp9_tree_probs_from_distribution( - VP9_BINTRAMODES, vp9_bmode_encodings, vp9_bmode_tree, - bmode_probs, branch_ct, cm->fc.bmode_counts, - 256, 1); + vp9_tree_probs_from_distribution(VP9_BINTRAMODES, vp9_bmode_encodings, + vp9_bmode_tree, bmode_probs, branch_ct, + cm->fc.bmode_counts, 256, 1); for (t = 0; t < VP9_BINTRAMODES - 1; ++t) { int prob; count = branch_ct[t][0] + branch_ct[t][1]; @@ -579,10 +565,9 @@ void vp9_adapt_mode_probs(VP9_COMMON *cm) { else if (prob > 255) cm->fc.bmode_prob[t] = 255; else cm->fc.bmode_prob[t] = prob; } - vp9_tree_probs_from_distribution( - VP9_I8X8_MODES, vp9_i8x8_mode_encodings, vp9_i8x8_mode_tree, - i8x8_mode_probs, branch_ct, cm->fc.i8x8_mode_counts, - 256, 1); + vp9_tree_probs_from_distribution(VP9_I8X8_MODES, vp9_i8x8_mode_encodings, + vp9_i8x8_mode_tree, i8x8_mode_probs, + branch_ct, cm->fc.i8x8_mode_counts, 256, 1); for (t = 0; t < VP9_I8X8_MODES - 1; ++t) { int prob; count = branch_ct[t][0] + branch_ct[t][1]; @@ -595,10 +580,11 @@ void vp9_adapt_mode_probs(VP9_COMMON *cm) { else cm->fc.i8x8_mode_prob[t] = prob; } for (i = 0; i < SUBMVREF_COUNT; ++i) { - vp9_tree_probs_from_distribution( - VP9_SUBMVREFS, vp9_sub_mv_ref_encoding_array, vp9_sub_mv_ref_tree, - sub_mv_ref_probs, branch_ct, cm->fc.sub_mv_ref_counts[i], - 256, 1); + vp9_tree_probs_from_distribution(VP9_SUBMVREFS, + vp9_sub_mv_ref_encoding_array, + vp9_sub_mv_ref_tree, sub_mv_ref_probs, + branch_ct, cm->fc.sub_mv_ref_counts[i], + 256, 1); for (t = 0; t < VP9_SUBMVREFS - 1; ++t) { int prob; count = branch_ct[t][0] + branch_ct[t][1]; @@ -611,10 +597,9 @@ void vp9_adapt_mode_probs(VP9_COMMON *cm) { else cm->fc.sub_mv_ref_prob[i][t] = prob; } } - vp9_tree_probs_from_distribution( - VP9_NUMMBSPLITS, vp9_mbsplit_encodings, vp9_mbsplit_tree, - mbsplit_probs, branch_ct, cm->fc.mbsplit_counts, - 256, 1); + vp9_tree_probs_from_distribution(VP9_NUMMBSPLITS, vp9_mbsplit_encodings, + vp9_mbsplit_tree, mbsplit_probs, branch_ct, + cm->fc.mbsplit_counts, 256, 1); for (t = 0; t < VP9_NUMMBSPLITS - 1; ++t) { int prob; count = branch_ct[t][0] + branch_ct[t][1]; diff --git a/vp8/common/entropymode.h b/vp8/common/entropymode.h index 4049eae41..b55e3b0a6 100644 --- a/vp8/common/entropymode.h +++ b/vp8/common/entropymode.h @@ -20,20 +20,21 @@ typedef const int vp9_mbsplit[16]; -extern vp9_mbsplit vp9_mbsplits [VP9_NUMMBSPLITS]; +extern vp9_mbsplit vp9_mbsplits[VP9_NUMMBSPLITS]; -extern const int vp9_mbsplit_count [VP9_NUMMBSPLITS]; /* # of subsets */ +extern const int vp9_mbsplit_count[VP9_NUMMBSPLITS]; /* # of subsets */ -extern const vp9_prob vp9_mbsplit_probs [VP9_NUMMBSPLITS - 1]; +extern const vp9_prob vp9_mbsplit_probs[VP9_NUMMBSPLITS - 1]; extern int vp9_mv_cont(const int_mv *l, const int_mv *a); -extern const vp9_prob vp9_sub_mv_ref_prob [VP9_SUBMVREFS - 1]; -extern const vp9_prob vp9_sub_mv_ref_prob2 [SUBMVREF_COUNT][VP9_SUBMVREFS - 1]; +extern const vp9_prob vp9_sub_mv_ref_prob[VP9_SUBMVREFS - 1]; +extern const vp9_prob vp9_sub_mv_ref_prob2[SUBMVREF_COUNT][VP9_SUBMVREFS - 1]; -extern const unsigned int vp9_kf_default_bmode_counts[VP9_BINTRAMODES][VP9_BINTRAMODES][VP9_BINTRAMODES]; - +extern const unsigned int vp9_kf_default_bmode_counts[VP9_BINTRAMODES] + [VP9_BINTRAMODES] + [VP9_BINTRAMODES]; extern const vp9_tree_index vp9_bmode_tree[]; @@ -47,43 +48,55 @@ extern const vp9_tree_index vp9_mv_ref_tree[]; extern const vp9_tree_index vp9_sb_mv_ref_tree[]; extern const vp9_tree_index vp9_sub_mv_ref_tree[]; -extern struct vp9_token_struct vp9_bmode_encodings [VP9_BINTRAMODES]; -extern struct vp9_token_struct vp9_ymode_encodings [VP9_YMODES]; -extern struct vp9_token_struct vp9_sb_kf_ymode_encodings [VP9_I32X32_MODES]; -extern struct vp9_token_struct vp9_kf_ymode_encodings [VP9_YMODES]; -extern struct vp9_token_struct vp9_i8x8_mode_encodings [VP9_I8X8_MODES]; -extern struct vp9_token_struct vp9_uv_mode_encodings [VP9_UV_MODES]; -extern struct vp9_token_struct vp9_mbsplit_encodings [VP9_NUMMBSPLITS]; +extern struct vp9_token_struct vp9_bmode_encodings[VP9_BINTRAMODES]; +extern struct vp9_token_struct vp9_ymode_encodings[VP9_YMODES]; +extern struct vp9_token_struct vp9_sb_kf_ymode_encodings[VP9_I32X32_MODES]; +extern struct vp9_token_struct vp9_kf_ymode_encodings[VP9_YMODES]; +extern struct vp9_token_struct vp9_i8x8_mode_encodings[VP9_I8X8_MODES]; +extern struct vp9_token_struct vp9_uv_mode_encodings[VP9_UV_MODES]; +extern struct vp9_token_struct vp9_mbsplit_encodings[VP9_NUMMBSPLITS]; /* Inter mode values do not start at zero */ -extern struct vp9_token_struct vp9_mv_ref_encoding_array [VP9_MVREFS]; -extern struct vp9_token_struct vp9_sb_mv_ref_encoding_array [VP9_MVREFS]; -extern struct vp9_token_struct vp9_sub_mv_ref_encoding_array [VP9_SUBMVREFS]; +extern struct vp9_token_struct vp9_mv_ref_encoding_array[VP9_MVREFS]; +extern struct vp9_token_struct vp9_sb_mv_ref_encoding_array[VP9_MVREFS]; +extern struct vp9_token_struct vp9_sub_mv_ref_encoding_array[VP9_SUBMVREFS]; void vp9_entropy_mode_init(void); struct VP9Common; + void vp9_init_mbmode_probs(struct VP9Common *x); + extern void vp9_init_mode_contexts(struct VP9Common *pc); -extern void vp9_update_mode_context(struct VP9Common *pc);; + +extern void vp9_update_mode_context(struct VP9Common *pc); + extern void vp9_accum_mv_refs(struct VP9Common *pc, MB_PREDICTION_MODE m, const int ct[4]); -void vp9_default_bmode_probs(vp9_prob dest [VP9_BINTRAMODES - 1]); -void vp9_kf_default_bmode_probs(vp9_prob dest [VP9_BINTRAMODES] [VP9_BINTRAMODES] [VP9_BINTRAMODES - 1]); +void vp9_default_bmode_probs(vp9_prob dest[VP9_BINTRAMODES - 1]); + +void vp9_kf_default_bmode_probs(vp9_prob dest[VP9_BINTRAMODES][VP9_BINTRAMODES] + [VP9_BINTRAMODES - 1]); void vp9_adapt_mode_probs(struct VP9Common *); #define VP9_SWITCHABLE_FILTERS 2 /* number of switchable filters */ + extern const INTERPOLATIONFILTERTYPE vp9_switchable_interp [VP9_SWITCHABLE_FILTERS]; + extern const int vp9_switchable_interp_map[SWITCHABLE + 1]; + extern const vp9_tree_index vp9_switchable_interp_tree - [2*(VP9_SWITCHABLE_FILTERS - 1)]; + [2 * (VP9_SWITCHABLE_FILTERS - 1)]; + extern struct vp9_token_struct vp9_switchable_interp_encodings [VP9_SWITCHABLE_FILTERS]; -extern const vp9_prob vp9_switchable_interp_prob - [VP9_SWITCHABLE_FILTERS + 1][VP9_SWITCHABLE_FILTERS - 1]; + +extern const vp9_prob vp9_switchable_interp_prob[VP9_SWITCHABLE_FILTERS + 1] + [VP9_SWITCHABLE_FILTERS - 1]; + #endif diff --git a/vp8/common/extend.c b/vp8/common/extend.c index 7632d7148..ad3b89b44 100644 --- a/vp8/common/extend.c +++ b/vp8/common/extend.c @@ -8,24 +8,19 @@ * be found in the AUTHORS file in the root of the source tree. */ - #include "extend.h" #include "vpx_mem/vpx_mem.h" - -static void copy_and_extend_plane -( - unsigned char *s, /* source */ - int sp, /* source pitch */ - unsigned char *d, /* destination */ - int dp, /* destination pitch */ - int h, /* height */ - int w, /* width */ - int et, /* extend top border */ - int el, /* extend left border */ - int eb, /* extend bottom border */ - int er /* extend right border */ -) { +static void copy_and_extend_plane(unsigned char *s, /* source */ + int sp, /* source pitch */ + unsigned char *d, /* destination */ + int dp, /* destination pitch */ + int h, /* height */ + int w, /* width */ + int et, /* extend top border */ + int el, /* extend left border */ + int eb, /* extend bottom border */ + int er) { /* extend right border */ int i; unsigned char *src_ptr1, *src_ptr2; unsigned char *dest_ptr1, *dest_ptr2; @@ -67,7 +62,6 @@ static void copy_and_extend_plane } } - void vp9_copy_and_extend_frame(YV12_BUFFER_CONFIG *src, YV12_BUFFER_CONFIG *dst) { int et = dst->border; @@ -96,7 +90,6 @@ void vp9_copy_and_extend_frame(YV12_BUFFER_CONFIG *src, et, el, eb, er); } - void vp9_copy_and_extend_frame_with_rect(YV12_BUFFER_CONFIG *src, YV12_BUFFER_CONFIG *dst, int srcy, int srcx, @@ -149,9 +142,9 @@ void vp9_copy_and_extend_frame_with_rect(YV12_BUFFER_CONFIG *src, et, el, eb, er); } - /* note the extension is only for the last row, for intra prediction purpose */ -void vp9_extend_mb_row(YV12_BUFFER_CONFIG *ybf, unsigned char *YPtr, unsigned char *UPtr, unsigned char *VPtr) { +void vp9_extend_mb_row(YV12_BUFFER_CONFIG *ybf, unsigned char *YPtr, + unsigned char *UPtr, unsigned char *VPtr) { int i; YPtr += ybf->y_stride * 14; diff --git a/vp8/common/extend.h b/vp8/common/extend.h index 9c30ad10c..c3c590479 100644 --- a/vp8/common/extend.h +++ b/vp8/common/extend.h @@ -8,18 +8,20 @@ * be found in the AUTHORS file in the root of the source tree. */ - #ifndef __INC_EXTEND_H #define __INC_EXTEND_H #include "vpx_scale/yv12config.h" -void vp9_extend_mb_row(YV12_BUFFER_CONFIG *ybf, unsigned char *YPtr, unsigned char *UPtr, unsigned char *VPtr); +void vp9_extend_mb_row(YV12_BUFFER_CONFIG *ybf, unsigned char *YPtr, + unsigned char *UPtr, unsigned char *VPtr); + void vp9_copy_and_extend_frame(YV12_BUFFER_CONFIG *src, YV12_BUFFER_CONFIG *dst); + void vp9_copy_and_extend_frame_with_rect(YV12_BUFFER_CONFIG *src, YV12_BUFFER_CONFIG *dst, int srcy, int srcx, int srch, int srcw); -#endif +#endif // __INC_EXTEND_H diff --git a/vp8/common/filter.c b/vp8/common/filter.c index 83a30a427..4667b461d 100644 --- a/vp8/common/filter.c +++ b/vp8/common/filter.c @@ -140,16 +140,13 @@ DECLARE_ALIGNED(16, const short, vp9_sub_pel_filters_6[SUBPEL_SHIFTS][6]) = { {1, -2, 8, 125, -5, 1} }; -static void filter_block2d_first_pass_6 -( - unsigned char *src_ptr, - int *output_ptr, - unsigned int src_pixels_per_line, - unsigned int pixel_step, - unsigned int output_height, - unsigned int output_width, - const short *vp9_filter -) { +static void filter_block2d_first_pass_6(unsigned char *src_ptr, + int *output_ptr, + unsigned int src_pixels_per_line, + unsigned int pixel_step, + unsigned int output_height, + unsigned int output_width, + const short *vp9_filter) { unsigned int i, j; int Temp; @@ -181,17 +178,14 @@ static void filter_block2d_first_pass_6 } } -static void filter_block2d_second_pass_6 -( - int *src_ptr, - unsigned char *output_ptr, - int output_pitch, - unsigned int src_pixels_per_line, - unsigned int pixel_step, - unsigned int output_height, - unsigned int output_width, - const short *vp9_filter -) { +static void filter_block2d_second_pass_6(int *src_ptr, + unsigned char *output_ptr, + int output_pitch, + unsigned int src_pixels_per_line, + unsigned int pixel_step, + unsigned int output_height, + unsigned int output_width, + const short *vp9_filter) { unsigned int i, j; int Temp; @@ -232,17 +226,14 @@ static void filter_block2d_second_pass_6 * and then averages that with the content already present in the output * ((filter_result + dest + 1) >> 1) and stores that in the output. */ -static void filter_block2d_second_pass_avg_6 -( - int *src_ptr, - unsigned char *output_ptr, - int output_pitch, - unsigned int src_pixels_per_line, - unsigned int pixel_step, - unsigned int output_height, - unsigned int output_width, - const short *vp9_filter -) { +static void filter_block2d_second_pass_avg_6(int *src_ptr, + unsigned char *output_ptr, + int output_pitch, + unsigned int src_pixels_per_line, + unsigned int pixel_step, + unsigned int output_height, + unsigned int output_width, + const short *vp9_filter) { unsigned int i, j; int Temp; @@ -276,15 +267,12 @@ static void filter_block2d_second_pass_avg_6 } #define Interp_Extend 3 -static void filter_block2d_6 -( - unsigned char *src_ptr, - unsigned char *output_ptr, - unsigned int src_pixels_per_line, - int output_pitch, - const short *HFilter, - const short *VFilter -) { +static void filter_block2d_6(unsigned char *src_ptr, + unsigned char *output_ptr, + unsigned int src_pixels_per_line, + int output_pitch, + const short *HFilter, + const short *VFilter) { int FData[(3 + Interp_Extend * 2) * 4]; /* Temp data buffer used in filtering */ /* First filter 1-D horizontally... */ @@ -296,15 +284,12 @@ static void filter_block2d_6 } -void vp9_sixtap_predict_c -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { +void vp9_sixtap_predict_c(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { const short *HFilter; const short *VFilter; @@ -321,15 +306,12 @@ void vp9_sixtap_predict_c * then averages that with the content already present in the output * ((filter_result + dest + 1) >> 1) and stores that in the output. */ -static void filter_block2d_avg_6 -( - unsigned char *src_ptr, - unsigned char *output_ptr, - unsigned int src_pixels_per_line, - int output_pitch, - const short *HFilter, - const short *VFilter -) { +static void filter_block2d_avg_6(unsigned char *src_ptr, + unsigned char *output_ptr, + unsigned int src_pixels_per_line, + int output_pitch, + const short *HFilter, + const short *VFilter) { int FData[(3 + Interp_Extend * 2) * 4]; /* Temp data buffer used in filtering */ /* First filter 1-D horizontally... */ @@ -670,15 +652,12 @@ static void block2d_average_c(unsigned char *src, #define block2d_average block2d_average_c -void vp9_eighttap_predict_c -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { +void vp9_eighttap_predict_c(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { const short *HFilter; const short *VFilter; @@ -690,15 +669,12 @@ void vp9_eighttap_predict_c dst_ptr, dst_pitch); } -void vp9_eighttap_predict_avg4x4_c -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { +void vp9_eighttap_predict_avg4x4_c(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { const short *HFilter = vp9_sub_pel_filters_8[xoffset]; const short *VFilter = vp9_sub_pel_filters_8[yoffset]; unsigned char tmp[4 * 4]; @@ -709,15 +685,12 @@ void vp9_eighttap_predict_avg4x4_c block2d_average(tmp, 4, dst_ptr, dst_pitch, VPX_FILTER_4x4); } -void vp9_eighttap_predict_sharp_c -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { +void vp9_eighttap_predict_sharp_c(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { const short *HFilter; const short *VFilter; @@ -729,15 +702,12 @@ void vp9_eighttap_predict_sharp_c dst_ptr, dst_pitch); } -void vp9_eighttap_predict_avg4x4_sharp_c -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { +void vp9_eighttap_predict_avg4x4_sharp_c(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { const short *HFilter = vp9_sub_pel_filters_8s[xoffset]; const short *VFilter = vp9_sub_pel_filters_8s[yoffset]; unsigned char tmp[4 * 4]; @@ -748,15 +718,12 @@ void vp9_eighttap_predict_avg4x4_sharp_c block2d_average(tmp, 4, dst_ptr, dst_pitch, VPX_FILTER_4x4); } -void vp9_eighttap_predict8x8_c -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { +void vp9_eighttap_predict8x8_c(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { const short *HFilter = vp9_sub_pel_filters_8[xoffset]; const short *VFilter = vp9_sub_pel_filters_8[yoffset]; @@ -765,15 +732,12 @@ void vp9_eighttap_predict8x8_c dst_ptr, dst_pitch); } -void vp9_eighttap_predict8x8_sharp_c -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { +void vp9_eighttap_predict8x8_sharp_c(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { const short *HFilter = vp9_sub_pel_filters_8s[xoffset]; const short *VFilter = vp9_sub_pel_filters_8s[yoffset]; @@ -782,15 +746,12 @@ void vp9_eighttap_predict8x8_sharp_c dst_ptr, dst_pitch); } -void vp9_eighttap_predict_avg8x8_c -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { +void vp9_eighttap_predict_avg8x8_c(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { unsigned char tmp[8 * 8]; const short *HFilter = vp9_sub_pel_filters_8[xoffset]; const short *VFilter = vp9_sub_pel_filters_8[yoffset]; @@ -801,15 +762,12 @@ void vp9_eighttap_predict_avg8x8_c block2d_average(tmp, 8, dst_ptr, dst_pitch, VPX_FILTER_8x8); } -void vp9_eighttap_predict_avg8x8_sharp_c -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { +void vp9_eighttap_predict_avg8x8_sharp_c(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { unsigned char tmp[8 * 8]; const short *HFilter = vp9_sub_pel_filters_8s[xoffset]; const short *VFilter = vp9_sub_pel_filters_8s[yoffset]; @@ -820,16 +778,12 @@ void vp9_eighttap_predict_avg8x8_sharp_c block2d_average(tmp, 8, dst_ptr, dst_pitch, VPX_FILTER_8x8); } -void vp9_eighttap_predict8x4_c -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { - +void vp9_eighttap_predict8x4_c(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { const short *HFilter = vp9_sub_pel_filters_8[xoffset]; const short *VFilter = vp9_sub_pel_filters_8[yoffset]; @@ -838,15 +792,12 @@ void vp9_eighttap_predict8x4_c dst_ptr, dst_pitch); } -void vp9_eighttap_predict8x4_sharp_c -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { +void vp9_eighttap_predict8x4_sharp_c(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { const short *HFilter = vp9_sub_pel_filters_8s[xoffset]; const short *VFilter = vp9_sub_pel_filters_8s[yoffset]; @@ -855,15 +806,12 @@ void vp9_eighttap_predict8x4_sharp_c dst_ptr, dst_pitch); } -void vp9_eighttap_predict16x16_c -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { +void vp9_eighttap_predict16x16_c(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { const short *HFilter = vp9_sub_pel_filters_8[xoffset]; const short *VFilter = vp9_sub_pel_filters_8[yoffset]; @@ -872,15 +820,12 @@ void vp9_eighttap_predict16x16_c dst_ptr, dst_pitch); } -void vp9_eighttap_predict16x16_sharp_c -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { +void vp9_eighttap_predict16x16_sharp_c(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { const short *HFilter = vp9_sub_pel_filters_8s[xoffset]; const short *VFilter = vp9_sub_pel_filters_8s[yoffset]; @@ -889,15 +834,12 @@ void vp9_eighttap_predict16x16_sharp_c dst_ptr, dst_pitch); } -void vp9_eighttap_predict_avg16x16_c -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { +void vp9_eighttap_predict_avg16x16_c(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { DECLARE_ALIGNED_ARRAY(16, unsigned char, tmp, 16 * 16); const short *HFilter = vp9_sub_pel_filters_8[xoffset]; const short *VFilter = vp9_sub_pel_filters_8[yoffset]; @@ -908,15 +850,12 @@ void vp9_eighttap_predict_avg16x16_c block2d_average(tmp, 16, dst_ptr, dst_pitch, VPX_FILTER_16x16); } -void vp9_eighttap_predict_avg16x16_sharp_c -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { +void vp9_eighttap_predict_avg16x16_sharp_c(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { DECLARE_ALIGNED_ARRAY(16, unsigned char, tmp, 16 * 16); const short *HFilter = vp9_sub_pel_filters_8s[xoffset]; const short *VFilter = vp9_sub_pel_filters_8s[yoffset]; @@ -949,15 +888,12 @@ void vp9_eighttap_predict_avg16x16_sharp_c * Two filter taps should sum to VP9_FILTER_WEIGHT. * ****************************************************************************/ -static void filter_block2d_bil_first_pass -( - unsigned char *src_ptr, - unsigned short *dst_ptr, - unsigned int src_stride, - unsigned int height, - unsigned int width, - const short *vp9_filter -) { +static void filter_block2d_bil_first_pass(unsigned char *src_ptr, + unsigned short *dst_ptr, + unsigned int src_stride, + unsigned int height, + unsigned int width, + const short *vp9_filter) { unsigned int i, j; for (i = 0; i < height; i++) { @@ -997,15 +933,12 @@ static void filter_block2d_bil_first_pass * Two filter taps should sum to VP9_FILTER_WEIGHT. * ****************************************************************************/ -static void filter_block2d_bil_second_pass -( - unsigned short *src_ptr, - unsigned char *dst_ptr, - int dst_pitch, - unsigned int height, - unsigned int width, - const short *vp9_filter -) { +static void filter_block2d_bil_second_pass(unsigned short *src_ptr, + unsigned char *dst_ptr, + int dst_pitch, + unsigned int height, + unsigned int width, + const short *vp9_filter) { unsigned int i, j; int Temp; @@ -1033,15 +966,12 @@ static void filter_block2d_bil_second_pass * with the values already present in the output and stores the result of * that back into the output ((filter_result + dest + 1) >> 1). */ -static void filter_block2d_bil_second_pass_avg -( - unsigned short *src_ptr, - unsigned char *dst_ptr, - int dst_pitch, - unsigned int height, - unsigned int width, - const short *vp9_filter -) { +static void filter_block2d_bil_second_pass_avg(unsigned short *src_ptr, + unsigned char *dst_ptr, + int dst_pitch, + unsigned int height, + unsigned int width, + const short *vp9_filter) { unsigned int i, j; int Temp; @@ -1083,17 +1013,14 @@ static void filter_block2d_bil_second_pass_avg * SPECIAL NOTES : The largest block size can be handled here is 16x16 * ****************************************************************************/ -static void filter_block2d_bil -( - unsigned char *src_ptr, - unsigned char *dst_ptr, - unsigned int src_pitch, - unsigned int dst_pitch, - const short *HFilter, - const short *VFilter, - int Width, - int Height -) { +static void filter_block2d_bil(unsigned char *src_ptr, + unsigned char *dst_ptr, + unsigned int src_pitch, + unsigned int dst_pitch, + const short *HFilter, + const short *VFilter, + int Width, + int Height) { unsigned short FData[17 * 16]; /* Temp data buffer used in filtering */ @@ -1104,17 +1031,14 @@ static void filter_block2d_bil filter_block2d_bil_second_pass(FData, dst_ptr, dst_pitch, Height, Width, VFilter); } -static void filter_block2d_bil_avg -( - unsigned char *src_ptr, - unsigned char *dst_ptr, - unsigned int src_pitch, - unsigned int dst_pitch, - const short *HFilter, - const short *VFilter, - int Width, - int Height -) { +static void filter_block2d_bil_avg(unsigned char *src_ptr, + unsigned char *dst_ptr, + unsigned int src_pitch, + unsigned int dst_pitch, + const short *HFilter, + const short *VFilter, + int Width, + int Height) { unsigned short FData[17 * 16]; /* Temp data buffer used in filtering */ /* First filter 1-D horizontally... */ @@ -1124,50 +1048,27 @@ static void filter_block2d_bil_avg filter_block2d_bil_second_pass_avg(FData, dst_ptr, dst_pitch, Height, Width, VFilter); } -void vp9_bilinear_predict4x4_c -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { +void vp9_bilinear_predict4x4_c(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { const short *HFilter; const short *VFilter; HFilter = vp9_bilinear_filters[xoffset]; VFilter = vp9_bilinear_filters[yoffset]; -#if 0 - { - int i; - unsigned char temp1[16]; - unsigned char temp2[16]; - - bilinear_predict4x4_mmx(src_ptr, src_pixels_per_line, xoffset, yoffset, temp1, 4); - filter_block2d_bil(src_ptr, temp2, src_pixels_per_line, 4, HFilter, VFilter, 4, 4); - for (i = 0; i < 16; i++) { - if (temp1[i] != temp2[i]) { - bilinear_predict4x4_mmx(src_ptr, src_pixels_per_line, xoffset, yoffset, temp1, 4); - filter_block2d_bil(src_ptr, temp2, src_pixels_per_line, 4, HFilter, VFilter, 4, 4); - } - } - } -#endif filter_block2d_bil(src_ptr, dst_ptr, src_pixels_per_line, dst_pitch, HFilter, VFilter, 4, 4); - } -void vp9_bilinear_predict_avg4x4_c -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { +void vp9_bilinear_predict_avg4x4_c(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { const short *HFilter; const short *VFilter; @@ -1178,15 +1079,12 @@ void vp9_bilinear_predict_avg4x4_c dst_pitch, HFilter, VFilter, 4, 4); } -void vp9_bilinear_predict8x8_c -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { +void vp9_bilinear_predict8x8_c(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { const short *HFilter; const short *VFilter; @@ -1197,15 +1095,12 @@ void vp9_bilinear_predict8x8_c } -void vp9_bilinear_predict_avg8x8_c -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { +void vp9_bilinear_predict_avg8x8_c(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { const short *HFilter; const short *VFilter; @@ -1216,15 +1111,12 @@ void vp9_bilinear_predict_avg8x8_c dst_pitch, HFilter, VFilter, 8, 8); } -void vp9_bilinear_predict8x4_c -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { +void vp9_bilinear_predict8x4_c(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { const short *HFilter; const short *VFilter; @@ -1235,15 +1127,12 @@ void vp9_bilinear_predict8x4_c } -void vp9_bilinear_predict16x16_c -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { +void vp9_bilinear_predict16x16_c(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { const short *HFilter; const short *VFilter; @@ -1253,15 +1142,12 @@ void vp9_bilinear_predict16x16_c filter_block2d_bil(src_ptr, dst_ptr, src_pixels_per_line, dst_pitch, HFilter, VFilter, 16, 16); } -void vp9_bilinear_predict_avg16x16_c -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { +void vp9_bilinear_predict_avg16x16_c(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { const short *HFilter; const short *VFilter; diff --git a/vp8/common/findnearmv.h b/vp8/common/findnearmv.h index e927c22af..3cba90d14 100644 --- a/vp8/common/findnearmv.h +++ b/vp8/common/findnearmv.h @@ -81,24 +81,20 @@ static unsigned int check_mv_bounds(int_mv *mv, (mv->as_mv.row > mb_to_bottom_edge); } -void vp9_find_near_mvs -( - MACROBLOCKD *xd, - const MODE_INFO *here, - const MODE_INFO *lfhere, - int_mv *nearest, int_mv *nearby, int_mv *best, - int near_mv_ref_cts[4], - int refframe, - int *ref_frame_sign_bias -); +void vp9_find_near_mvs(MACROBLOCKD *xd, + const MODE_INFO *here, + const MODE_INFO *lfhere, + int_mv *nearest, int_mv *nearby, int_mv *best, + int near_mv_ref_cts[4], + int refframe, + int *ref_frame_sign_bias); vp9_prob *vp9_mv_ref_probs(VP9_COMMON *pc, - vp9_prob p[VP9_MVREFS - 1], const int near_mv_ref_ct[4] - ); + vp9_prob p[VP9_MVREFS - 1], + const int near_mv_ref_ct[4]); extern const unsigned char vp9_mbsplit_offset[4][16]; - static int left_block_mv(const MODE_INFO *cur_mb, int b) { if (!(b & 3)) { /* On L edge, get from MB to left of us */ diff --git a/vp8/common/g_common.h b/vp8/common/g_common.h deleted file mode 100644 index 7dc942056..000000000 --- a/vp8/common/g_common.h +++ /dev/null @@ -1,21 +0,0 @@ -/* - * Copyright (c) 2010 The WebM project authors. All Rights Reserved. - * - * Use of this source code is governed by a BSD-style license - * that can be found in the LICENSE file in the root of the source - * tree. An additional intellectual property rights grant can be found - * in the file PATENTS. All contributing project authors may - * be found in the AUTHORS file in the root of the source tree. - */ - - -extern void (*vp9_clear_system_state)(void); -extern void (*vp9_plane_add_noise)(unsigned char *Start, unsigned int Width, unsigned int Height, int Pitch, int DPitch, int q); -extern void (*de_interlace) -( - unsigned char *src_ptr, - unsigned char *dst_ptr, - int Width, - int Height, - int Stride -); diff --git a/vp8/common/generic/systemdependent.c b/vp8/common/generic/systemdependent.c index 55b66461a..8924f62aa 100644 --- a/vp8/common/generic/systemdependent.c +++ b/vp8/common/generic/systemdependent.c @@ -11,7 +11,6 @@ #include "vpx_ports/config.h" #include "vpx_rtcd.h" -#include "vp8/common/g_common.h" #include "vp8/common/subpixel.h" #include "vp8/common/loopfilter.h" #include "vp8/common/idct.h" @@ -41,13 +40,13 @@ void vp9_machine_specific_config(VP9_COMMON *ctx) { rtcd->subpix.eighttap_avg4x4 = vp9_eighttap_predict_avg4x4_c; rtcd->subpix.eighttap8x4 = vp9_eighttap_predict8x4_c; rtcd->subpix.eighttap4x4 = vp9_eighttap_predict_c; - rtcd->subpix.eighttap16x16_sharp = vp9_eighttap_predict16x16_sharp_c; - rtcd->subpix.eighttap8x8_sharp = vp9_eighttap_predict8x8_sharp_c; - rtcd->subpix.eighttap_avg16x16_sharp = vp9_eighttap_predict_avg16x16_sharp_c; - rtcd->subpix.eighttap_avg8x8_sharp = vp9_eighttap_predict_avg8x8_sharp_c; - rtcd->subpix.eighttap_avg4x4_sharp = vp9_eighttap_predict_avg4x4_sharp_c; - rtcd->subpix.eighttap8x4_sharp = vp9_eighttap_predict8x4_sharp_c; - rtcd->subpix.eighttap4x4_sharp = vp9_eighttap_predict_sharp_c; + rtcd->subpix.eighttap16x16_sharp = vp9_eighttap_predict16x16_sharp_c; + rtcd->subpix.eighttap8x8_sharp = vp9_eighttap_predict8x8_sharp_c; + rtcd->subpix.eighttap_avg16x16_sharp = vp9_eighttap_predict_avg16x16_sharp_c; + rtcd->subpix.eighttap_avg8x8_sharp = vp9_eighttap_predict_avg8x8_sharp_c; + rtcd->subpix.eighttap_avg4x4_sharp = vp9_eighttap_predict_avg4x4_sharp_c; + rtcd->subpix.eighttap8x4_sharp = vp9_eighttap_predict8x4_sharp_c; + rtcd->subpix.eighttap4x4_sharp = vp9_eighttap_predict_sharp_c; rtcd->subpix.sixtap16x16 = vp9_sixtap_predict16x16_c; rtcd->subpix.sixtap8x8 = vp9_sixtap_predict8x8_c; @@ -80,7 +79,6 @@ void vp9_machine_specific_config(VP9_COMMON *ctx) { vp9_arch_x86_common_init(ctx); #endif - #if ARCH_ARM vp9_arch_arm_common_init(ctx); #endif diff --git a/vp8/common/idctllm.c b/vp8/common/idctllm.c index efbb6432c..9958f9508 100644 --- a/vp8/common/idctllm.c +++ b/vp8/common/idctllm.c @@ -413,7 +413,7 @@ void vp9_ihtllm_float_c(const int16_t *input, int16_t *output, int pitch, #define HORIZONTAL_SHIFT 17 // 15 #define HORIZONTAL_ROUNDING ((1 << (HORIZONTAL_SHIFT - 1)) - 1) void vp9_ihtllm_int_c(const int16_t *input, int16_t *output, int pitch, - TX_TYPE tx_type, int tx_dim) { + TX_TYPE tx_type, int tx_dim) { int i, j, k; int16_t imbuf[256]; @@ -564,7 +564,8 @@ void vp9_short_idct4x4llm_1_c(short *input, short *output, int pitch) { } } -void vp9_dc_only_idct_add_c(short input_dc, unsigned char *pred_ptr, unsigned char *dst_ptr, int pitch, int stride) { +void vp9_dc_only_idct_add_c(short input_dc, unsigned char *pred_ptr, + unsigned char *dst_ptr, int pitch, int stride) { int a1 = ((input_dc + 16) >> 5); int r, c; @@ -584,7 +585,6 @@ void vp9_dc_only_idct_add_c(short input_dc, unsigned char *pred_ptr, unsigned ch dst_ptr += stride; pred_ptr += pitch; } - } void vp9_short_inv_walsh4x4_c(short *input, short *output) { @@ -765,7 +765,9 @@ void vp9_short_inv_walsh4x4_1_x8_c(short *in, short *out, int pitch) { } } -void vp9_dc_only_inv_walsh_add_c(short input_dc, unsigned char *pred_ptr, unsigned char *dst_ptr, int pitch, int stride) { +void vp9_dc_only_inv_walsh_add_c(short input_dc, unsigned char *pred_ptr, + unsigned char *dst_ptr, + int pitch, int stride) { int r, c; short tmp[16]; vp9_short_inv_walsh4x4_1_x8_c(&input_dc, tmp, 4 << 1); diff --git a/vp8/common/invtrans.h b/vp8/common/invtrans.h index 00dc815cd..370964f1c 100644 --- a/vp8/common/invtrans.h +++ b/vp8/common/invtrans.h @@ -8,7 +8,6 @@ * be found in the AUTHORS file in the root of the source tree. */ - #ifndef __INC_INVTRANS_H #define __INC_INVTRANS_H @@ -16,19 +15,39 @@ #include "idct.h" #include "blockd.h" -extern void vp9_inverse_transform_b_4x4(const vp9_idct_rtcd_vtable_t *rtcd, BLOCKD *b, int pitch); -extern void vp9_inverse_transform_mb_4x4(const vp9_idct_rtcd_vtable_t *rtcd, MACROBLOCKD *xd); -extern void vp9_inverse_transform_mby_4x4(const vp9_idct_rtcd_vtable_t *rtcd, MACROBLOCKD *xd); -extern void vp9_inverse_transform_mbuv_4x4(const vp9_idct_rtcd_vtable_t *rtcd, MACROBLOCKD *xd); +extern void vp9_inverse_transform_b_4x4(const vp9_idct_rtcd_vtable_t *rtcd, + BLOCKD *b, int pitch); + +extern void vp9_inverse_transform_mb_4x4(const vp9_idct_rtcd_vtable_t *rtcd, + MACROBLOCKD *xd); + +extern void vp9_inverse_transform_mby_4x4(const vp9_idct_rtcd_vtable_t *rtcd, + MACROBLOCKD *xd); + +extern void vp9_inverse_transform_mbuv_4x4(const vp9_idct_rtcd_vtable_t *rtcd, + MACROBLOCKD *xd); + +extern void vp9_inverse_transform_b_8x8(const vp9_idct_rtcd_vtable_t *rtcd, + short *input_dqcoeff, + short *output_coeff, int pitch); -extern void vp9_inverse_transform_b_8x8(const vp9_idct_rtcd_vtable_t *rtcd, short *input_dqcoeff, short *output_coeff, int pitch); -extern void vp9_inverse_transform_mb_8x8(const vp9_idct_rtcd_vtable_t *rtcd, MACROBLOCKD *xd); -extern void vp9_inverse_transform_mby_8x8(const vp9_idct_rtcd_vtable_t *rtcd, MACROBLOCKD *xd); -extern void vp9_inverse_transform_mbuv_8x8(const vp9_idct_rtcd_vtable_t *rtcd, MACROBLOCKD *xd); +extern void vp9_inverse_transform_mb_8x8(const vp9_idct_rtcd_vtable_t *rtcd, + MACROBLOCKD *xd); + +extern void vp9_inverse_transform_mby_8x8(const vp9_idct_rtcd_vtable_t *rtcd, + MACROBLOCKD *xd); + +extern void vp9_inverse_transform_mbuv_8x8(const vp9_idct_rtcd_vtable_t *rtcd, + MACROBLOCKD *xd); extern void vp9_inverse_transform_b_16x16(const vp9_idct_rtcd_vtable_t *rtcd, - short *input_dqcoeff, short *output_coeff, - int pitch); -extern void vp9_inverse_transform_mb_16x16(const vp9_idct_rtcd_vtable_t *rtcd, MACROBLOCKD *xd); -extern void vp9_inverse_transform_mby_16x16(const vp9_idct_rtcd_vtable_t *rtcd, MACROBLOCKD *xd); -#endif + short *input_dqcoeff, + short *output_coeff, int pitch); + +extern void vp9_inverse_transform_mb_16x16(const vp9_idct_rtcd_vtable_t *rtcd, + MACROBLOCKD *xd); + +extern void vp9_inverse_transform_mby_16x16(const vp9_idct_rtcd_vtable_t *rtcd, + MACROBLOCKD *xd); + +#endif // __INC_INVTRANS_H diff --git a/vp8/common/loopfilter.c b/vp8/common/loopfilter.c index 25ab30850..a2bf14d82 100644 --- a/vp8/common/loopfilter.c +++ b/vp8/common/loopfilter.c @@ -8,7 +8,6 @@ * be found in the AUTHORS file in the root of the source tree. */ - #include "vpx_config.h" #include "loopfilter.h" #include "onyxc_int.h" @@ -52,7 +51,6 @@ static void lf_init_lut(loop_filter_info_n *lfi) { lfi->mode_lf_lut[NEARMV] = 2; lfi->mode_lf_lut[NEWMV] = 2; lfi->mode_lf_lut[SPLITMV] = 3; - } void vp9_loop_filter_update_sharpness(loop_filter_info_n *lfi, @@ -178,11 +176,7 @@ void vp9_loop_filter_frame_init(VP9_COMMON *cm, } } -void vp9_loop_filter_frame -( - VP9_COMMON *cm, - MACROBLOCKD *xd -) { +void vp9_loop_filter_frame(VP9_COMMON *cm, MACROBLOCKD *xd) { YV12_BUFFER_CONFIG *post = cm->frame_to_show; loop_filter_info_n *lfi_n = &cm->lf_info; struct loop_filter_info lfi; @@ -316,12 +310,8 @@ void vp9_loop_filter_frame } } -void vp9_loop_filter_frame_yonly -( - VP9_COMMON *cm, - MACROBLOCKD *xd, - int default_filt_lvl -) { +void vp9_loop_filter_frame_yonly(VP9_COMMON *cm, MACROBLOCKD *xd, + int default_filt_lvl) { YV12_BUFFER_CONFIG *post = cm->frame_to_show; unsigned char *y_ptr; @@ -413,21 +403,15 @@ void vp9_loop_filter_frame_yonly y_ptr += 16; mode_info_context++; /* step to next MB */ - } y_ptr += post->y_stride * 16 - post->y_width; mode_info_context++; /* Skip border mb */ } - } -void vp9_loop_filter_partial_frame -( - VP9_COMMON *cm, - MACROBLOCKD *xd, - int default_filt_lvl -) { +void vp9_loop_filter_partial_frame(VP9_COMMON *cm, MACROBLOCKD *xd, + int default_filt_lvl) { YV12_BUFFER_CONFIG *post = cm->frame_to_show; unsigned char *y_ptr; diff --git a/vp8/common/loopfilter.h b/vp8/common/loopfilter.h index 99c135742..0b7de682c 100644 --- a/vp8/common/loopfilter.h +++ b/vp8/common/loopfilter.h @@ -8,7 +8,6 @@ * be found in the AUTHORS file in the root of the source tree. */ - #ifndef loopfilter_h #define loopfilter_h @@ -33,10 +32,14 @@ typedef enum { * passed it can be loaded into vector registers. */ typedef struct { - DECLARE_ALIGNED(SIMD_WIDTH, unsigned char, mblim[MAX_LOOP_FILTER + 1][SIMD_WIDTH]); - DECLARE_ALIGNED(SIMD_WIDTH, unsigned char, blim[MAX_LOOP_FILTER + 1][SIMD_WIDTH]); - DECLARE_ALIGNED(SIMD_WIDTH, unsigned char, lim[MAX_LOOP_FILTER + 1][SIMD_WIDTH]); - DECLARE_ALIGNED(SIMD_WIDTH, unsigned char, hev_thr[4][SIMD_WIDTH]); + DECLARE_ALIGNED(SIMD_WIDTH, unsigned char, + mblim[MAX_LOOP_FILTER + 1][SIMD_WIDTH]); + DECLARE_ALIGNED(SIMD_WIDTH, unsigned char, + blim[MAX_LOOP_FILTER + 1][SIMD_WIDTH]); + DECLARE_ALIGNED(SIMD_WIDTH, unsigned char, + lim[MAX_LOOP_FILTER + 1][SIMD_WIDTH]); + DECLARE_ALIGNED(SIMD_WIDTH, unsigned char, + hev_thr[4][SIMD_WIDTH]); unsigned char lvl[4][4][4]; unsigned char hev_thr_lut[2][MAX_LOOP_FILTER + 1]; unsigned char mode_lf_lut[MB_MODE_COUNT]; @@ -49,7 +52,6 @@ struct loop_filter_info { const unsigned char *hev_thr; }; - #define prototype_loopfilter(sym) \ void sym(unsigned char *src, int pitch, const unsigned char *blimit,\ const unsigned char *limit, const unsigned char *thresh, int count) @@ -69,15 +71,12 @@ struct loop_filter_info { #include "arm/loopfilter_arm.h" #endif -typedef void loop_filter_uvfunction -( - unsigned char *u, /* source pointer */ - int p, /* pitch */ - const unsigned char *blimit, - const unsigned char *limit, - const unsigned char *thresh, - unsigned char *v -); +typedef void loop_filter_uvfunction(unsigned char *u, /* source pointer */ + int p, /* pitch */ + const unsigned char *blimit, + const unsigned char *limit, + const unsigned char *thresh, + unsigned char *v); /* assorted loopfilter functions which get used elsewhere */ struct VP9Common; @@ -102,4 +101,4 @@ void vp9_loop_filter_frame_yonly(struct VP9Common *cm, void vp9_loop_filter_update_sharpness(loop_filter_info_n *lfi, int sharpness_lvl); -#endif +#endif // loopfilter_h diff --git a/vp8/common/loopfilter_filters.c b/vp8/common/loopfilter_filters.c index 9808287b3..c719b44c3 100644 --- a/vp8/common/loopfilter_filters.c +++ b/vp8/common/loopfilter_filters.c @@ -7,6 +7,7 @@ * in the file PATENTS. All contributing project authors may * be found in the AUTHORS file in the root of the source tree. */ + #include <stdlib.h> #include "vpx_config.h" #include "loopfilter.h" @@ -122,15 +123,12 @@ void vp9_loop_filter_horizontal_edge_c } while (++i < count * 8); } -void vp9_loop_filter_vertical_edge_c -( - unsigned char *s, - int p, - const unsigned char *blimit, - const unsigned char *limit, - const unsigned char *thresh, - int count -) { +void vp9_loop_filter_vertical_edge_c(unsigned char *s, + int p, + const unsigned char *blimit, + const unsigned char *limit, + const unsigned char *thresh, + int count) { int hev = 0; /* high edge variance */ signed char mask = 0; int i = 0; @@ -382,56 +380,72 @@ void vp9_loop_filter_simple_vertical_edge_c void vp9_loop_filter_mbv_c(unsigned char *y_ptr, unsigned char *u_ptr, unsigned char *v_ptr, int y_stride, int uv_stride, struct loop_filter_info *lfi) { - vp9_mbloop_filter_vertical_edge_c(y_ptr, y_stride, lfi->mblim, lfi->lim, lfi->hev_thr, 2); + vp9_mbloop_filter_vertical_edge_c(y_ptr, y_stride, + lfi->mblim, lfi->lim, lfi->hev_thr, 2); if (u_ptr) - vp9_mbloop_filter_vertical_edge_c(u_ptr, uv_stride, lfi->mblim, lfi->lim, lfi->hev_thr, 1); + vp9_mbloop_filter_vertical_edge_c(u_ptr, uv_stride, + lfi->mblim, lfi->lim, lfi->hev_thr, 1); if (v_ptr) - vp9_mbloop_filter_vertical_edge_c(v_ptr, uv_stride, lfi->mblim, lfi->lim, lfi->hev_thr, 1); + vp9_mbloop_filter_vertical_edge_c(v_ptr, uv_stride, + lfi->mblim, lfi->lim, lfi->hev_thr, 1); } /* Vertical B Filtering */ void vp9_loop_filter_bv_c(unsigned char *y_ptr, unsigned char *u_ptr, unsigned char *v_ptr, int y_stride, int uv_stride, struct loop_filter_info *lfi) { - vp9_loop_filter_vertical_edge_c(y_ptr + 4, y_stride, lfi->blim, lfi->lim, lfi->hev_thr, 2); - vp9_loop_filter_vertical_edge_c(y_ptr + 8, y_stride, lfi->blim, lfi->lim, lfi->hev_thr, 2); - vp9_loop_filter_vertical_edge_c(y_ptr + 12, y_stride, lfi->blim, lfi->lim, lfi->hev_thr, 2); + vp9_loop_filter_vertical_edge_c(y_ptr + 4, y_stride, + lfi->blim, lfi->lim, lfi->hev_thr, 2); + vp9_loop_filter_vertical_edge_c(y_ptr + 8, y_stride, + lfi->blim, lfi->lim, lfi->hev_thr, 2); + vp9_loop_filter_vertical_edge_c(y_ptr + 12, y_stride, + lfi->blim, lfi->lim, lfi->hev_thr, 2); if (u_ptr) - vp9_loop_filter_vertical_edge_c(u_ptr + 4, uv_stride, lfi->blim, lfi->lim, lfi->hev_thr, 1); + vp9_loop_filter_vertical_edge_c(u_ptr + 4, uv_stride, + lfi->blim, lfi->lim, lfi->hev_thr, 1); if (v_ptr) - vp9_loop_filter_vertical_edge_c(v_ptr + 4, uv_stride, lfi->blim, lfi->lim, lfi->hev_thr, 1); + vp9_loop_filter_vertical_edge_c(v_ptr + 4, uv_stride, + lfi->blim, lfi->lim, lfi->hev_thr, 1); } /* Horizontal MB filtering */ void vp9_loop_filter_mbh_c(unsigned char *y_ptr, unsigned char *u_ptr, unsigned char *v_ptr, int y_stride, int uv_stride, struct loop_filter_info *lfi) { - vp9_mbloop_filter_horizontal_edge_c(y_ptr, y_stride, lfi->mblim, lfi->lim, lfi->hev_thr, 2); + vp9_mbloop_filter_horizontal_edge_c(y_ptr, y_stride, + lfi->mblim, lfi->lim, lfi->hev_thr, 2); if (u_ptr) - vp9_mbloop_filter_horizontal_edge_c(u_ptr, uv_stride, lfi->mblim, lfi->lim, lfi->hev_thr, 1); + vp9_mbloop_filter_horizontal_edge_c(u_ptr, uv_stride, + lfi->mblim, lfi->lim, lfi->hev_thr, 1); if (v_ptr) - vp9_mbloop_filter_horizontal_edge_c(v_ptr, uv_stride, lfi->mblim, lfi->lim, lfi->hev_thr, 1); + vp9_mbloop_filter_horizontal_edge_c(v_ptr, uv_stride, + lfi->mblim, lfi->lim, lfi->hev_thr, 1); } /* Horizontal B Filtering */ void vp9_loop_filter_bh_c(unsigned char *y_ptr, unsigned char *u_ptr, unsigned char *v_ptr, int y_stride, int uv_stride, struct loop_filter_info *lfi) { - vp9_loop_filter_horizontal_edge_c(y_ptr + 4 * y_stride, y_stride, lfi->blim, lfi->lim, lfi->hev_thr, 2); - vp9_loop_filter_horizontal_edge_c(y_ptr + 8 * y_stride, y_stride, lfi->blim, lfi->lim, lfi->hev_thr, 2); - vp9_loop_filter_horizontal_edge_c(y_ptr + 12 * y_stride, y_stride, lfi->blim, lfi->lim, lfi->hev_thr, 2); + vp9_loop_filter_horizontal_edge_c(y_ptr + 4 * y_stride, y_stride, + lfi->blim, lfi->lim, lfi->hev_thr, 2); + vp9_loop_filter_horizontal_edge_c(y_ptr + 8 * y_stride, y_stride, + lfi->blim, lfi->lim, lfi->hev_thr, 2); + vp9_loop_filter_horizontal_edge_c(y_ptr + 12 * y_stride, y_stride, + lfi->blim, lfi->lim, lfi->hev_thr, 2); if (u_ptr) - vp9_loop_filter_horizontal_edge_c(u_ptr + 4 * uv_stride, uv_stride, lfi->blim, lfi->lim, lfi->hev_thr, 1); + vp9_loop_filter_horizontal_edge_c(u_ptr + 4 * uv_stride, uv_stride, + lfi->blim, lfi->lim, lfi->hev_thr, 1); if (v_ptr) - vp9_loop_filter_horizontal_edge_c(v_ptr + 4 * uv_stride, uv_stride, lfi->blim, lfi->lim, lfi->hev_thr, 1); + vp9_loop_filter_horizontal_edge_c(v_ptr + 4 * uv_stride, uv_stride, + lfi->blim, lfi->lim, lfi->hev_thr, 1); } void vp9_loop_filter_bh8x8_c(unsigned char *y_ptr, unsigned char *u_ptr, @@ -443,9 +457,12 @@ void vp9_loop_filter_bh8x8_c(unsigned char *y_ptr, unsigned char *u_ptr, void vp9_loop_filter_bhs_c(unsigned char *y_ptr, int y_stride, const unsigned char *blimit) { - vp9_loop_filter_simple_horizontal_edge_c(y_ptr + 4 * y_stride, y_stride, blimit); - vp9_loop_filter_simple_horizontal_edge_c(y_ptr + 8 * y_stride, y_stride, blimit); - vp9_loop_filter_simple_horizontal_edge_c(y_ptr + 12 * y_stride, y_stride, blimit); + vp9_loop_filter_simple_horizontal_edge_c(y_ptr + 4 * y_stride, + y_stride, blimit); + vp9_loop_filter_simple_horizontal_edge_c(y_ptr + 8 * y_stride, + y_stride, blimit); + vp9_loop_filter_simple_horizontal_edge_c(y_ptr + 12 * y_stride, + y_stride, blimit); } void vp9_loop_filter_bv8x8_c(unsigned char *y_ptr, unsigned char *u_ptr, diff --git a/vp8/common/postproc.c b/vp8/common/postproc.c index 4dea88b91..a05ed1236 100644 --- a/vp8/common/postproc.c +++ b/vp8/common/postproc.c @@ -20,10 +20,13 @@ #include <stdlib.h> #include <stdio.h> -#define RGB_TO_YUV(t) \ - ( (0.257*(float)(t>>16)) + (0.504*(float)(t>>8&0xff)) + (0.098*(float)(t&0xff)) + 16), \ - (-(0.148*(float)(t>>16)) - (0.291*(float)(t>>8&0xff)) + (0.439*(float)(t&0xff)) + 128), \ - ( (0.439*(float)(t>>16)) - (0.368*(float)(t>>8&0xff)) - (0.071*(float)(t&0xff)) + 128) +#define RGB_TO_YUV(t) \ + ( (0.257*(float)(t >> 16)) + (0.504*(float)(t >> 8 & 0xff)) + \ + (0.098*(float)(t & 0xff)) + 16), \ + (-(0.148*(float)(t >> 16)) - (0.291*(float)(t >> 8 & 0xff)) + \ + (0.439*(float)(t & 0xff)) + 128), \ + ( (0.439*(float)(t >> 16)) - (0.368*(float)(t >> 8 & 0xff)) - \ + (0.071*(float)(t & 0xff)) + 128) /* global constants */ #if CONFIG_POSTPROC_VISUALIZER @@ -124,20 +127,19 @@ const short vp9_rv[] = { }; -extern void vp9_blit_text(const char *msg, unsigned char *address, const int pitch); -extern void vp9_blit_line(int x0, int x1, int y0, int y1, unsigned char *image, const int pitch); -/*********************************************************************************************************** +extern void vp9_blit_text(const char *msg, unsigned char *address, + const int pitch); +extern void vp9_blit_line(int x0, int x1, int y0, int y1, + unsigned char *image, const int pitch); +/**************************************************************************** */ -void vp9_post_proc_down_and_across_c -( - unsigned char *src_ptr, - unsigned char *dst_ptr, - int src_pixels_per_line, - int dst_pixels_per_line, - int rows, - int cols, - int flimit -) { +void vp9_post_proc_down_and_across_c(unsigned char *src_ptr, + unsigned char *dst_ptr, + int src_pixels_per_line, + int dst_pixels_per_line, + int rows, + int cols, + int flimit) { unsigned char *p_src, *p_dst; int row; int col; @@ -213,7 +215,9 @@ static int q2mbl(int x) { x = 50 + (x - 50) * 10 / 8; return x * x / 3; } -void vp9_mbpost_proc_across_ip_c(unsigned char *src, int pitch, int rows, int cols, int flimit) { + +void vp9_mbpost_proc_across_ip_c(unsigned char *src, int pitch, + int rows, int cols, int flimit) { int r, c, i; unsigned char *s = src; @@ -250,11 +254,8 @@ void vp9_mbpost_proc_across_ip_c(unsigned char *src, int pitch, int rows, int co } } - - - - -void vp9_mbpost_proc_down_c(unsigned char *dst, int pitch, int rows, int cols, int flimit) { +void vp9_mbpost_proc_down_c(unsigned char *dst, int pitch, + int rows, int cols, int flimit) { int r, c, i; const short *rv3 = &vp9_rv[63 & rand()]; @@ -296,13 +297,20 @@ static void deblock_and_de_macro_block(YV12_BUFFER_CONFIG *source, (void) low_var_thresh; (void) flag; - POSTPROC_INVOKE(rtcd, downacross)(source->y_buffer, post->y_buffer, source->y_stride, post->y_stride, source->y_height, source->y_width, ppl); - POSTPROC_INVOKE(rtcd, across)(post->y_buffer, post->y_stride, post->y_height, post->y_width, q2mbl(q)); - POSTPROC_INVOKE(rtcd, down)(post->y_buffer, post->y_stride, post->y_height, post->y_width, q2mbl(q)); - - POSTPROC_INVOKE(rtcd, downacross)(source->u_buffer, post->u_buffer, source->uv_stride, post->uv_stride, source->uv_height, source->uv_width, ppl); - POSTPROC_INVOKE(rtcd, downacross)(source->v_buffer, post->v_buffer, source->uv_stride, post->uv_stride, source->uv_height, source->uv_width, ppl); - + POSTPROC_INVOKE(rtcd, downacross)(source->y_buffer, post->y_buffer, + source->y_stride, post->y_stride, + source->y_height, source->y_width, ppl); + POSTPROC_INVOKE(rtcd, across)(post->y_buffer, post->y_stride, + post->y_height, post->y_width, q2mbl(q)); + POSTPROC_INVOKE(rtcd, down)(post->y_buffer, post->y_stride, + post->y_height, post->y_width, q2mbl(q)); + + POSTPROC_INVOKE(rtcd, downacross)(source->u_buffer, post->u_buffer, + source->uv_stride, post->uv_stride, + source->uv_height, source->uv_width, ppl); + POSTPROC_INVOKE(rtcd, downacross)(source->v_buffer, post->v_buffer, + source->uv_stride, post->uv_stride, + source->uv_height, source->uv_width, ppl); } void vp9_deblock(YV12_BUFFER_CONFIG *source, @@ -316,12 +324,18 @@ void vp9_deblock(YV12_BUFFER_CONFIG *source, (void) low_var_thresh; (void) flag; - POSTPROC_INVOKE(rtcd, downacross)(source->y_buffer, post->y_buffer, source->y_stride, post->y_stride, source->y_height, source->y_width, ppl); - POSTPROC_INVOKE(rtcd, downacross)(source->u_buffer, post->u_buffer, source->uv_stride, post->uv_stride, source->uv_height, source->uv_width, ppl); - POSTPROC_INVOKE(rtcd, downacross)(source->v_buffer, post->v_buffer, source->uv_stride, post->uv_stride, source->uv_height, source->uv_width, ppl); + POSTPROC_INVOKE(rtcd, downacross)(source->y_buffer, post->y_buffer, + source->y_stride, post->y_stride, + source->y_height, source->y_width, ppl); + POSTPROC_INVOKE(rtcd, downacross)(source->u_buffer, post->u_buffer, + source->uv_stride, post->uv_stride, + source->uv_height, source->uv_width, ppl); + POSTPROC_INVOKE(rtcd, downacross)(source->v_buffer, post->v_buffer, + source->uv_stride, post->uv_stride, + source->uv_height, source->uv_width, ppl); } -void vp9_de_noise(YV12_BUFFER_CONFIG *source, +void vp9_de_noise(YV12_BUFFER_CONFIG *src, YV12_BUFFER_CONFIG *post, int q, int low_var_thresh, @@ -333,29 +347,25 @@ void vp9_de_noise(YV12_BUFFER_CONFIG *source, (void) low_var_thresh; (void) flag; - POSTPROC_INVOKE(rtcd, downacross)( - source->y_buffer + 2 * source->y_stride + 2, - source->y_buffer + 2 * source->y_stride + 2, - source->y_stride, - source->y_stride, - source->y_height - 4, - source->y_width - 4, - ppl); - POSTPROC_INVOKE(rtcd, downacross)( - source->u_buffer + 2 * source->uv_stride + 2, - source->u_buffer + 2 * source->uv_stride + 2, - source->uv_stride, - source->uv_stride, - source->uv_height - 4, - source->uv_width - 4, ppl); - POSTPROC_INVOKE(rtcd, downacross)( - source->v_buffer + 2 * source->uv_stride + 2, - source->v_buffer + 2 * source->uv_stride + 2, - source->uv_stride, - source->uv_stride, - source->uv_height - 4, - source->uv_width - 4, ppl); - + POSTPROC_INVOKE(rtcd, downacross)(src->y_buffer + 2 * src->y_stride + 2, + src->y_buffer + 2 * src->y_stride + 2, + src->y_stride, + src->y_stride, + src->y_height - 4, + src->y_width - 4, + ppl); + POSTPROC_INVOKE(rtcd, downacross)(src->u_buffer + 2 * src->uv_stride + 2, + src->u_buffer + 2 * src->uv_stride + 2, + src->uv_stride, + src->uv_stride, + src->uv_height - 4, + src->uv_width - 4, ppl); + POSTPROC_INVOKE(rtcd, downacross)(src->v_buffer + 2 * src->uv_stride + 2, + src->v_buffer + 2 * src->uv_stride + 2, + src->uv_stride, + src->uv_stride, + src->uv_height - 4, + src->uv_width - 4, ppl); } double vp9_gaussian(double sigma, double mu, double x) { @@ -363,9 +373,6 @@ double vp9_gaussian(double sigma, double mu, double x) { (exp(-(x - mu) * (x - mu) / (2 * sigma * sigma))); } -extern void (*vp9_clear_system_state)(void); - - static void fillrd(struct postproc_state *state, int q, int a) { char char_dist[300]; @@ -374,7 +381,6 @@ static void fillrd(struct postproc_state *state, int q, int a) { vp9_clear_system_state(); - sigma = ai + .5 + .6 * (63 - qi) / 63.0; /* set up a lookup table of 256 entries that matches @@ -401,7 +407,6 @@ static void fillrd(struct postproc_state *state, int q, int a) { for (next = next; next < 256; next++) char_dist[next] = 0; - } for (i = 0; i < 3072; i++) { @@ -422,8 +427,8 @@ static void fillrd(struct postproc_state *state, int q, int a) { * * ROUTINE : plane_add_noise_c * - * INPUTS : unsigned char *Start starting address of buffer to add gaussian - * noise to + * INPUTS : unsigned char *Start starting address of buffer to + * add gaussian noise to * unsigned int Width width of plane * unsigned int Height height of plane * int Pitch distance between subsequent lines of frame @@ -580,7 +585,8 @@ void vp9_blend_b_c(unsigned char *y, unsigned char *u, unsigned char *v, } } -static void constrain_line(int x0, int *x1, int y0, int *y1, int width, int height) { +static void constrain_line(int x0, int *x1, int y0, int *y1, + int width, int height) { int dx; int dy; @@ -625,7 +631,8 @@ static void constrain_line(int x0, int *x1, int y0, int *y1, int width, int heig #define RTCD_VTABLE(oci) NULL #endif -int vp9_post_proc_frame(VP9_COMMON *oci, YV12_BUFFER_CONFIG *dest, vp9_ppflags_t *ppflags) { +int vp9_post_proc_frame(VP9_COMMON *oci, YV12_BUFFER_CONFIG *dest, + vp9_ppflags_t *ppflags) { int q = oci->filter_level * 10 / 6; int flags = ppflags->post_proc_flag; int deblock_level = ppflags->deblocking_level; @@ -669,14 +676,14 @@ int vp9_post_proc_frame(VP9_COMMON *oci, YV12_BUFFER_CONFIG *dest, vp9_ppflags_t fillrd(&oci->postproc_state, 63 - q, noise_level); } - POSTPROC_INVOKE(RTCD_VTABLE(oci), addnoise) - (oci->post_proc_buffer.y_buffer, - oci->postproc_state.noise, - oci->postproc_state.blackclamp, - oci->postproc_state.whiteclamp, - oci->postproc_state.bothclamp, - oci->post_proc_buffer.y_width, oci->post_proc_buffer.y_height, - oci->post_proc_buffer.y_stride); + POSTPROC_INVOKE(RTCD_VTABLE(oci), addnoise)(oci->post_proc_buffer.y_buffer, + oci->postproc_state.noise, + oci->postproc_state.blackclamp, + oci->postproc_state.whiteclamp, + oci->postproc_state.bothclamp, + oci->post_proc_buffer.y_width, + oci->post_proc_buffer.y_height, + oci->post_proc_buffer.y_stride); } #if CONFIG_POSTPROC_VISUALIZER @@ -689,7 +696,8 @@ int vp9_post_proc_frame(VP9_COMMON *oci, YV12_BUFFER_CONFIG *dest, vp9_ppflags_t oci->filter_level, flags, oci->mb_cols, oci->mb_rows); - vp9_blit_text(message, oci->post_proc_buffer.y_buffer, oci->post_proc_buffer.y_stride); + vp9_blit_text(message, oci->post_proc_buffer.y_buffer, + oci->post_proc_buffer.y_stride); } if (flags & VP9D_DEBUG_TXT_MBLK_MODES) { @@ -758,8 +766,11 @@ int vp9_post_proc_frame(VP9_COMMON *oci, YV12_BUFFER_CONFIG *dest, vp9_ppflags_t if (flags & VP9D_DEBUG_TXT_RATE_INFO) { char message[512]; - sprintf(message, "Bitrate: %10.2f frame_rate: %10.2f ", oci->bitrate, oci->framerate); - vp9_blit_text(message, oci->post_proc_buffer.y_buffer, oci->post_proc_buffer.y_stride); + snprintf(message, sizeof(message), + "Bitrate: %10.2f frame_rate: %10.2f ", + oci->bitrate, oci->framerate); + vp9_blit_text(message, oci->post_proc_buffer.y_buffer, + oci->post_proc_buffer.y_stride); } /* Draw motion vectors */ @@ -920,7 +931,8 @@ int vp9_post_proc_frame(VP9_COMMON *oci, YV12_BUFFER_CONFIG *dest, vp9_ppflags_t int Y = 0, U = 0, V = 0; if (mi->mbmi.mode == B_PRED && - ((ppflags->display_mb_modes_flag & B_PRED) || ppflags->display_b_modes_flag)) { + ((ppflags->display_mb_modes_flag & B_PRED) || + ppflags->display_b_modes_flag)) { int by, bx; unsigned char *yl, *ul, *vl; union b_mode_info *bmi = mi->bmi; @@ -937,8 +949,11 @@ int vp9_post_proc_frame(VP9_COMMON *oci, YV12_BUFFER_CONFIG *dest, vp9_ppflags_t U = B_PREDICTION_MODE_colors[bmi->as_mode.first][1]; V = B_PREDICTION_MODE_colors[bmi->as_mode.first][2]; - POSTPROC_INVOKE(RTCD_VTABLE(oci), blend_b) - (yl + bx, ul + (bx >> 1), vl + (bx >> 1), Y, U, V, 0xc000, y_stride); + POSTPROC_INVOKE(RTCD_VTABLE(oci), blend_b)(yl + bx, + ul + (bx >> 1), + vl + (bx >> 1), + Y, U, V, + 0xc000, y_stride); } bmi++; } @@ -952,8 +967,11 @@ int vp9_post_proc_frame(VP9_COMMON *oci, YV12_BUFFER_CONFIG *dest, vp9_ppflags_t U = MB_PREDICTION_MODE_colors[mi->mbmi.mode][1]; V = MB_PREDICTION_MODE_colors[mi->mbmi.mode][2]; - POSTPROC_INVOKE(RTCD_VTABLE(oci), blend_mb_inner) - (y_ptr + x, u_ptr + (x >> 1), v_ptr + (x >> 1), Y, U, V, 0xc000, y_stride); + POSTPROC_INVOKE(RTCD_VTABLE(oci), blend_mb_inner)(y_ptr + x, + u_ptr + (x >> 1), + v_ptr + (x >> 1), + Y, U, V, + 0xc000, y_stride); } mi++; @@ -967,7 +985,8 @@ int vp9_post_proc_frame(VP9_COMMON *oci, YV12_BUFFER_CONFIG *dest, vp9_ppflags_t } /* Color in frame reference blocks */ - if ((flags & VP9D_DEBUG_CLR_FRM_REF_BLKS) && ppflags->display_ref_frame_flag) { + if ((flags & VP9D_DEBUG_CLR_FRM_REF_BLKS) && + ppflags->display_ref_frame_flag) { int y, x; YV12_BUFFER_CONFIG *post = &oci->post_proc_buffer; int width = post->y_width; @@ -987,8 +1006,11 @@ int vp9_post_proc_frame(VP9_COMMON *oci, YV12_BUFFER_CONFIG *dest, vp9_ppflags_t U = MV_REFERENCE_FRAME_colors[mi->mbmi.ref_frame][1]; V = MV_REFERENCE_FRAME_colors[mi->mbmi.ref_frame][2]; - POSTPROC_INVOKE(RTCD_VTABLE(oci), blend_mb_outer) - (y_ptr + x, u_ptr + (x >> 1), v_ptr + (x >> 1), Y, U, V, 0xc000, y_stride); + POSTPROC_INVOKE(RTCD_VTABLE(oci), blend_mb_outer)(y_ptr + x, + u_ptr + (x >> 1), + v_ptr + (x >> 1), + Y, U, V, + 0xc000, y_stride); } mi++; @@ -1008,5 +1030,6 @@ int vp9_post_proc_frame(VP9_COMMON *oci, YV12_BUFFER_CONFIG *dest, vp9_ppflags_t dest->y_width = oci->Width; dest->y_height = oci->Height; dest->uv_height = dest->y_height / 2; + return 0; } diff --git a/vp8/common/postproc.h b/vp8/common/postproc.h index 226f89dc2..f43a8f08a 100644 --- a/vp8/common/postproc.h +++ b/vp8/common/postproc.h @@ -13,28 +13,28 @@ #define POSTPROC_H #define prototype_postproc_inplace(sym)\ - void sym (unsigned char *dst, int pitch, int rows, int cols,int flimit) + void sym(unsigned char *dst, int pitch, int rows, int cols, int flimit) #define prototype_postproc(sym)\ - void sym (unsigned char *src, unsigned char *dst, int src_pitch,\ - int dst_pitch, int rows, int cols, int flimit) + void sym(unsigned char *src, unsigned char *dst, int src_pitch, \ + int dst_pitch, int rows, int cols, int flimit) #define prototype_postproc_addnoise(sym) \ - void sym (unsigned char *s, char *noise, char blackclamp[16],\ - char whiteclamp[16], char bothclamp[16],\ - unsigned int w, unsigned int h, int pitch) + void sym(unsigned char *s, char *noise, char blackclamp[16], \ + char whiteclamp[16], char bothclamp[16], \ + unsigned int w, unsigned int h, int pitch) #define prototype_postproc_blend_mb_inner(sym)\ - void sym (unsigned char *y, unsigned char *u, unsigned char *v,\ - int y1, int u1, int v1, int alpha, int stride) + void sym(unsigned char *y, unsigned char *u, unsigned char *v, \ + int y1, int u1, int v1, int alpha, int stride) #define prototype_postproc_blend_mb_outer(sym)\ - void sym (unsigned char *y, unsigned char *u, unsigned char *v,\ - int y1, int u1, int v1, int alpha, int stride) + void sym(unsigned char *y, unsigned char *u, unsigned char *v, \ + int y1, int u1, int v1, int alpha, int stride) #define prototype_postproc_blend_b(sym)\ - void sym (unsigned char *y, unsigned char *u, unsigned char *v,\ - int y1, int u1, int v1, int alpha, int stride) + void sym(unsigned char *y, unsigned char *u, unsigned char *v, \ + int y1, int u1, int v1, int alpha, int stride) #if ARCH_X86 || ARCH_X86_64 #include "x86/postproc_x86.h" diff --git a/vp8/common/ppc/systemdependent.c b/vp8/common/ppc/systemdependent.c index 9ed17dbcb..ba4f236a8 100644 --- a/vp8/common/ppc/systemdependent.c +++ b/vp8/common/ppc/systemdependent.c @@ -8,8 +8,6 @@ * be found in the AUTHORS file in the root of the source tree. */ - -#include "g_common.h" #include "subpixel.h" #include "loopfilter.h" #include "recon.h" diff --git a/vp8/common/reconinter.c b/vp8/common/reconinter.c index e404c634b..252485fe0 100644 --- a/vp8/common/reconinter.c +++ b/vp8/common/reconinter.c @@ -85,12 +85,10 @@ void vp9_setup_interp_filters(MACROBLOCKD *xd, } } -void vp9_copy_mem16x16_c( - unsigned char *src, - int src_stride, - unsigned char *dst, - int dst_stride) { - +void vp9_copy_mem16x16_c(unsigned char *src, + int src_stride, + unsigned char *dst, + int dst_stride) { int r; for (r = 0; r < 16; r++) { @@ -121,16 +119,13 @@ void vp9_copy_mem16x16_c( #endif src += src_stride; dst += dst_stride; - } - } -void vp9_avg_mem16x16_c( - unsigned char *src, - int src_stride, - unsigned char *dst, - int dst_stride) { +void vp9_avg_mem16x16_c(unsigned char *src, + int src_stride, + unsigned char *dst, + int dst_stride) { int r; for (r = 0; r < 16; r++) { @@ -145,11 +140,10 @@ void vp9_avg_mem16x16_c( } } -void vp9_copy_mem8x8_c( - unsigned char *src, - int src_stride, - unsigned char *dst, - int dst_stride) { +void vp9_copy_mem8x8_c(unsigned char *src, + int src_stride, + unsigned char *dst, + int dst_stride) { int r; for (r = 0; r < 8; r++) { @@ -168,16 +162,13 @@ void vp9_copy_mem8x8_c( #endif src += src_stride; dst += dst_stride; - } - } -void vp9_avg_mem8x8_c( - unsigned char *src, - int src_stride, - unsigned char *dst, - int dst_stride) { +void vp9_avg_mem8x8_c(unsigned char *src, + int src_stride, + unsigned char *dst, + int dst_stride) { int r; for (r = 0; r < 8; r++) { @@ -192,11 +183,10 @@ void vp9_avg_mem8x8_c( } } -void vp9_copy_mem8x4_c( - unsigned char *src, - int src_stride, - unsigned char *dst, - int dst_stride) { +void vp9_copy_mem8x4_c(unsigned char *src, + int src_stride, + unsigned char *dst, + int dst_stride) { int r; for (r = 0; r < 4; r++) { @@ -215,13 +205,9 @@ void vp9_copy_mem8x4_c( #endif src += src_stride; dst += dst_stride; - } - } - - void vp9_build_inter_predictors_b(BLOCKD *d, int pitch, vp9_subpix_fn_t sppf) { int r; unsigned char *ptr_base; @@ -233,10 +219,13 @@ void vp9_build_inter_predictors_b(BLOCKD *d, int pitch, vp9_subpix_fn_t sppf) { mv.as_int = d->bmi.as_mv.first.as_int; if (mv.as_mv.row & 7 || mv.as_mv.col & 7) { - ptr = ptr_base + d->pre + (mv.as_mv.row >> 3) * d->pre_stride + (mv.as_mv.col >> 3); - sppf(ptr, d->pre_stride, (mv.as_mv.col & 7) << 1, (mv.as_mv.row & 7) << 1, pred_ptr, pitch); + ptr = ptr_base + d->pre + (mv.as_mv.row >> 3) * d->pre_stride + + (mv.as_mv.col >> 3); + sppf(ptr, d->pre_stride, (mv.as_mv.col & 7) << 1, (mv.as_mv.row & 7) << 1, + pred_ptr, pitch); } else { - ptr_base += d->pre + (mv.as_mv.row >> 3) * d->pre_stride + (mv.as_mv.col >> 3); + ptr_base += d->pre + (mv.as_mv.row >> 3) * d->pre_stride + + (mv.as_mv.col >> 3); ptr = ptr_base; for (r = 0; r < 4; r++) { @@ -260,7 +249,8 @@ void vp9_build_inter_predictors_b(BLOCKD *d, int pitch, vp9_subpix_fn_t sppf) { * come from an earlier call to vp9_build_inter_predictors_b()) with the * predictor of the second reference frame / motion vector. */ -void vp9_build_2nd_inter_predictors_b(BLOCKD *d, int pitch, vp9_subpix_fn_t sppf) { +void vp9_build_2nd_inter_predictors_b(BLOCKD *d, int pitch, + vp9_subpix_fn_t sppf) { int r; unsigned char *ptr_base; unsigned char *ptr; @@ -271,10 +261,13 @@ void vp9_build_2nd_inter_predictors_b(BLOCKD *d, int pitch, vp9_subpix_fn_t sppf mv.as_int = d->bmi.as_mv.second.as_int; if (mv.as_mv.row & 7 || mv.as_mv.col & 7) { - ptr = ptr_base + d->pre + (mv.as_mv.row >> 3) * d->pre_stride + (mv.as_mv.col >> 3); - sppf(ptr, d->pre_stride, (mv.as_mv.col & 7) << 1, (mv.as_mv.row & 7) << 1, pred_ptr, pitch); + ptr = ptr_base + d->pre + (mv.as_mv.row >> 3) * d->pre_stride + + (mv.as_mv.col >> 3); + sppf(ptr, d->pre_stride, (mv.as_mv.col & 7) << 1, (mv.as_mv.row & 7) << 1, + pred_ptr, pitch); } else { - ptr_base += d->pre + (mv.as_mv.row >> 3) * d->pre_stride + (mv.as_mv.col >> 3); + ptr_base += d->pre + (mv.as_mv.row >> 3) * d->pre_stride + + (mv.as_mv.col >> 3); ptr = ptr_base; for (r = 0; r < 4; r++) { diff --git a/vp8/common/reconinter.h b/vp8/common/reconinter.h index 3b72b0a7e..ad1cf9197 100644 --- a/vp8/common/reconinter.h +++ b/vp8/common/reconinter.h @@ -8,7 +8,6 @@ * be found in the AUTHORS file in the root of the source tree. */ - #ifndef __INC_RECONINTER_H #define __INC_RECONINTER_H @@ -18,10 +17,12 @@ extern void vp9_build_1st_inter16x16_predictors_mby(MACROBLOCKD *xd, unsigned char *dst_y, int dst_ystride, int clamp_mvs); + extern void vp9_build_1st_inter16x16_predictors_mbuv(MACROBLOCKD *xd, unsigned char *dst_u, unsigned char *dst_v, int dst_uvstride); + extern void vp9_build_1st_inter16x16_predictors_mb(MACROBLOCKD *xd, unsigned char *dst_y, unsigned char *dst_u, @@ -32,10 +33,12 @@ extern void vp9_build_1st_inter16x16_predictors_mb(MACROBLOCKD *xd, extern void vp9_build_2nd_inter16x16_predictors_mby(MACROBLOCKD *xd, unsigned char *dst_y, int dst_ystride); + extern void vp9_build_2nd_inter16x16_predictors_mbuv(MACROBLOCKD *xd, unsigned char *dst_u, unsigned char *dst_v, int dst_uvstride); + extern void vp9_build_2nd_inter16x16_predictors_mb(MACROBLOCKD *xd, unsigned char *dst_y, unsigned char *dst_u, @@ -56,17 +59,20 @@ extern void vp9_build_inter_predictors_mb(MACROBLOCKD *xd); extern void vp9_build_inter_predictors_b(BLOCKD *d, int pitch, vp9_subpix_fn_t sppf); + extern void vp9_build_2nd_inter_predictors_b(BLOCKD *d, int pitch, vp9_subpix_fn_t sppf); extern void vp9_build_inter_predictors4b(MACROBLOCKD *xd, BLOCKD *d, int pitch); + extern void vp9_build_2nd_inter_predictors4b(MACROBLOCKD *xd, BLOCKD *d, int pitch); extern void vp9_build_inter4x4_predictors_mbuv(MACROBLOCKD *xd); + extern void vp9_setup_interp_filters(MACROBLOCKD *xd, INTERPOLATIONFILTERTYPE filter, VP9_COMMON *cm); -#endif +#endif // __INC_RECONINTER_H diff --git a/vp8/common/reconintra.c b/vp8/common/reconintra.c index 16e6edbc9..97f782c52 100644 --- a/vp8/common/reconintra.c +++ b/vp8/common/reconintra.c @@ -201,7 +201,7 @@ void vp9_recon_intra_mbuv(MACROBLOCKD *xd) { for (i = 16; i < 24; i += 2) { BLOCKD *b = &xd->block[i]; - vp9_recon2b(b->predictor, b->diff,*(b->base_dst) + b->dst, b->dst_stride); + vp9_recon2b(b->predictor, b->diff, *(b->base_dst) + b->dst, b->dst_stride); } } diff --git a/vp8/common/reconintra.h b/vp8/common/reconintra.h index 72de5d99f..bf990b4c6 100644 --- a/vp8/common/reconintra.h +++ b/vp8/common/reconintra.h @@ -8,7 +8,6 @@ * be found in the AUTHORS file in the root of the source tree. */ - #ifndef __INC_RECONINTRA_H #define __INC_RECONINTRA_H @@ -16,4 +15,4 @@ extern void init_intra_left_above_pixels(MACROBLOCKD *xd); -#endif +#endif // __INC_RECONINTRA_H diff --git a/vp8/common/swapyv12buffer.c b/vp8/common/swapyv12buffer.c index dd5246a66..13fc8d4bc 100644 --- a/vp8/common/swapyv12buffer.c +++ b/vp8/common/swapyv12buffer.c @@ -8,10 +8,10 @@ * be found in the AUTHORS file in the root of the source tree. */ - #include "swapyv12buffer.h" -void vp9_swap_yv12_buffer(YV12_BUFFER_CONFIG *new_frame, YV12_BUFFER_CONFIG *last_frame) { +void vp9_swap_yv12_buffer(YV12_BUFFER_CONFIG *new_frame, + YV12_BUFFER_CONFIG *last_frame) { unsigned char *temp; temp = last_frame->buffer_alloc; @@ -29,5 +29,4 @@ void vp9_swap_yv12_buffer(YV12_BUFFER_CONFIG *new_frame, YV12_BUFFER_CONFIG *las temp = last_frame->v_buffer; last_frame->v_buffer = new_frame->v_buffer; new_frame->v_buffer = temp; - } diff --git a/vp8/common/swapyv12buffer.h b/vp8/common/swapyv12buffer.h index d1c7d18c9..44ed5e84d 100644 --- a/vp8/common/swapyv12buffer.h +++ b/vp8/common/swapyv12buffer.h @@ -8,12 +8,12 @@ * be found in the AUTHORS file in the root of the source tree. */ - -#ifndef SWAPYV12_BUFFER_H -#define SWAPYV12_BUFFER_H +#ifndef __SWAPYV12_BUFFER_H +#define __SWAPYV12_BUFFER_H #include "vpx_scale/yv12config.h" -void vp9_swap_yv12_buffer(YV12_BUFFER_CONFIG *new_frame, YV12_BUFFER_CONFIG *last_frame); +void vp9_swap_yv12_buffer(YV12_BUFFER_CONFIG *new_frame, + YV12_BUFFER_CONFIG *last_frame); -#endif +#endif // __SWAPYV12_BUFFER_H diff --git a/vp8/common/x86/loopfilter_x86.c b/vp8/common/x86/loopfilter_x86.c index 30933d2d9..faf857740 100644 --- a/vp8/common/x86/loopfilter_x86.c +++ b/vp8/common/x86/loopfilter_x86.c @@ -8,7 +8,6 @@ * be found in the AUTHORS file in the root of the source tree. */ - #include <emmintrin.h> // SSE2 #include "vpx_config.h" #include "vp8/common/loopfilter.h" @@ -24,64 +23,73 @@ extern loop_filter_uvfunction vp9_loop_filter_vertical_edge_uv_sse2; #if HAVE_MMX /* Horizontal MB filtering */ -void vp9_loop_filter_mbh_mmx(unsigned char *y_ptr, unsigned char *u_ptr, unsigned char *v_ptr, - int y_stride, int uv_stride, struct loop_filter_info *lfi) { +void vp9_loop_filter_mbh_mmx(unsigned char *y_ptr, + unsigned char *u_ptr, unsigned char *v_ptr, + int y_stride, int uv_stride, + struct loop_filter_info *lfi) { } - /* Vertical MB Filtering */ -void vp9_loop_filter_mbv_mmx(unsigned char *y_ptr, unsigned char *u_ptr, unsigned char *v_ptr, - int y_stride, int uv_stride, struct loop_filter_info *lfi) { +void vp9_loop_filter_mbv_mmx(unsigned char *y_ptr, + unsigned char *u_ptr, unsigned char *v_ptr, + int y_stride, int uv_stride, + struct loop_filter_info *lfi) { } - /* Horizontal B Filtering */ -void vp9_loop_filter_bh_mmx(unsigned char *y_ptr, unsigned char *u_ptr, unsigned char *v_ptr, - int y_stride, int uv_stride, struct loop_filter_info *lfi) { +void vp9_loop_filter_bh_mmx(unsigned char *y_ptr, + unsigned char *u_ptr, unsigned char *v_ptr, + int y_stride, int uv_stride, + struct loop_filter_info *lfi) { } - -void vp9_loop_filter_bhs_mmx(unsigned char *y_ptr, int y_stride, const unsigned char *blimit) { - vp9_loop_filter_simple_horizontal_edge_mmx(y_ptr + 4 * y_stride, y_stride, blimit); - vp9_loop_filter_simple_horizontal_edge_mmx(y_ptr + 8 * y_stride, y_stride, blimit); - vp9_loop_filter_simple_horizontal_edge_mmx(y_ptr + 12 * y_stride, y_stride, blimit); +void vp9_loop_filter_bhs_mmx(unsigned char *y_ptr, int y_stride, + const unsigned char *blimit) { + vp9_loop_filter_simple_horizontal_edge_mmx(y_ptr + 4 * y_stride, + y_stride, blimit); + vp9_loop_filter_simple_horizontal_edge_mmx(y_ptr + 8 * y_stride, + y_stride, blimit); + vp9_loop_filter_simple_horizontal_edge_mmx(y_ptr + 12 * y_stride, + y_stride, blimit); } - /* Vertical B Filtering */ -void vp9_loop_filter_bv_mmx(unsigned char *y_ptr, unsigned char *u_ptr, unsigned char *v_ptr, - int y_stride, int uv_stride, struct loop_filter_info *lfi) { - vp9_loop_filter_vertical_edge_mmx(y_ptr + 4, y_stride, lfi->blim, lfi->lim, lfi->hev_thr, 2); - vp9_loop_filter_vertical_edge_mmx(y_ptr + 8, y_stride, lfi->blim, lfi->lim, lfi->hev_thr, 2); - vp9_loop_filter_vertical_edge_mmx(y_ptr + 12, y_stride, lfi->blim, lfi->lim, lfi->hev_thr, 2); +void vp9_loop_filter_bv_mmx(unsigned char *y_ptr, + unsigned char *u_ptr, unsigned char *v_ptr, + int y_stride, int uv_stride, + struct loop_filter_info *lfi) { + vp9_loop_filter_vertical_edge_mmx(y_ptr + 4, y_stride, + lfi->blim, lfi->lim, lfi->hev_thr, 2); + vp9_loop_filter_vertical_edge_mmx(y_ptr + 8, y_stride, + lfi->blim, lfi->lim, lfi->hev_thr, 2); + vp9_loop_filter_vertical_edge_mmx(y_ptr + 12, y_stride, + lfi->blim, lfi->lim, lfi->hev_thr, 2); if (u_ptr) - vp9_loop_filter_vertical_edge_mmx(u_ptr + 4, uv_stride, lfi->blim, lfi->lim, lfi->hev_thr, 1); + vp9_loop_filter_vertical_edge_mmx(u_ptr + 4, uv_stride, + lfi->blim, lfi->lim, lfi->hev_thr, 1); if (v_ptr) - vp9_loop_filter_vertical_edge_mmx(v_ptr + 4, uv_stride, lfi->blim, lfi->lim, lfi->hev_thr, 1); + vp9_loop_filter_vertical_edge_mmx(v_ptr + 4, uv_stride, + lfi->blim, lfi->lim, lfi->hev_thr, 1); } - -void vp9_loop_filter_bvs_mmx(unsigned char *y_ptr, int y_stride, const unsigned char *blimit) { +void vp9_loop_filter_bvs_mmx(unsigned char *y_ptr, int y_stride, + const unsigned char *blimit) { vp9_loop_filter_simple_vertical_edge_mmx(y_ptr + 4, y_stride, blimit); vp9_loop_filter_simple_vertical_edge_mmx(y_ptr + 8, y_stride, blimit); vp9_loop_filter_simple_vertical_edge_mmx(y_ptr + 12, y_stride, blimit); } #endif - #if HAVE_SSE2 -void vp9_mbloop_filter_horizontal_edge_c_sse2 -( - unsigned char *s, - int p, - const unsigned char *_blimit, - const unsigned char *_limit, - const unsigned char *_thresh, - int count -) { +void vp9_mbloop_filter_horizontal_edge_c_sse2(unsigned char *s, + int p, + const unsigned char *_blimit, + const unsigned char *_limit, + const unsigned char *_thresh, + int count) { DECLARE_ALIGNED(16, unsigned char, flat_op2[16]); DECLARE_ALIGNED(16, unsigned char, flat_op1[16]); DECLARE_ALIGNED(16, unsigned char, flat_op0[16]); @@ -330,6 +338,7 @@ void vp9_mbloop_filter_horizontal_edge_c_sse2 } } } + static __inline void transpose(unsigned char *src[], int in_p, unsigned char *dst[], int out_p, int num_8x8_to_transpose) { @@ -392,15 +401,13 @@ static __inline void transpose(unsigned char *src[], int in_p, _mm_castsi128_pd(x7)); // 07 17 27 37 47 57 67 77 } while (++idx8x8 < num_8x8_to_transpose); } -void vp9_mbloop_filter_vertical_edge_c_sse2 -( - unsigned char *s, - int p, - const unsigned char *blimit, - const unsigned char *limit, - const unsigned char *thresh, - int count -) { + +void vp9_mbloop_filter_vertical_edge_c_sse2(unsigned char *s, + int p, + const unsigned char *blimit, + const unsigned char *limit, + const unsigned char *thresh, + int count) { DECLARE_ALIGNED(16, unsigned char, t_dst[16 * 16]); unsigned char *src[4]; unsigned char *dst[4]; @@ -432,9 +439,10 @@ void vp9_mbloop_filter_vertical_edge_c_sse2 } /* Horizontal MB filtering */ -void vp9_loop_filter_mbh_sse2(unsigned char *y_ptr, unsigned char *u_ptr, unsigned char *v_ptr, - int y_stride, int uv_stride, struct loop_filter_info *lfi) { - +void vp9_loop_filter_mbh_sse2(unsigned char *y_ptr, + unsigned char *u_ptr, unsigned char *v_ptr, + int y_stride, int uv_stride, + struct loop_filter_info *lfi) { vp9_mbloop_filter_horizontal_edge_c_sse2(y_ptr, y_stride, lfi->mblim, lfi->lim, lfi->hev_thr, 2); @@ -480,37 +488,53 @@ void vp9_loop_filter_bv8x8_sse2(unsigned char *y_ptr, unsigned char *u_ptr, } /* Horizontal B Filtering */ -void vp9_loop_filter_bh_sse2(unsigned char *y_ptr, unsigned char *u_ptr, unsigned char *v_ptr, - int y_stride, int uv_stride, struct loop_filter_info *lfi) { - vp9_loop_filter_horizontal_edge_sse2(y_ptr + 4 * y_stride, y_stride, lfi->blim, lfi->lim, lfi->hev_thr, 2); - vp9_loop_filter_horizontal_edge_sse2(y_ptr + 8 * y_stride, y_stride, lfi->blim, lfi->lim, lfi->hev_thr, 2); - vp9_loop_filter_horizontal_edge_sse2(y_ptr + 12 * y_stride, y_stride, lfi->blim, lfi->lim, lfi->hev_thr, 2); +void vp9_loop_filter_bh_sse2(unsigned char *y_ptr, + unsigned char *u_ptr, unsigned char *v_ptr, + int y_stride, int uv_stride, + struct loop_filter_info *lfi) { + vp9_loop_filter_horizontal_edge_sse2(y_ptr + 4 * y_stride, y_stride, + lfi->blim, lfi->lim, lfi->hev_thr, 2); + vp9_loop_filter_horizontal_edge_sse2(y_ptr + 8 * y_stride, y_stride, + lfi->blim, lfi->lim, lfi->hev_thr, 2); + vp9_loop_filter_horizontal_edge_sse2(y_ptr + 12 * y_stride, y_stride, + lfi->blim, lfi->lim, lfi->hev_thr, 2); if (u_ptr) - vp9_loop_filter_horizontal_edge_uv_sse2(u_ptr + 4 * uv_stride, uv_stride, lfi->blim, lfi->lim, lfi->hev_thr, v_ptr + 4 * uv_stride); + vp9_loop_filter_horizontal_edge_uv_sse2(u_ptr + 4 * uv_stride, uv_stride, + lfi->blim, lfi->lim, lfi->hev_thr, + v_ptr + 4 * uv_stride); } - -void vp9_loop_filter_bhs_sse2(unsigned char *y_ptr, int y_stride, const unsigned char *blimit) { - vp9_loop_filter_simple_horizontal_edge_sse2(y_ptr + 4 * y_stride, y_stride, blimit); - vp9_loop_filter_simple_horizontal_edge_sse2(y_ptr + 8 * y_stride, y_stride, blimit); - vp9_loop_filter_simple_horizontal_edge_sse2(y_ptr + 12 * y_stride, y_stride, blimit); +void vp9_loop_filter_bhs_sse2(unsigned char *y_ptr, int y_stride, + const unsigned char *blimit) { + vp9_loop_filter_simple_horizontal_edge_sse2(y_ptr + 4 * y_stride, + y_stride, blimit); + vp9_loop_filter_simple_horizontal_edge_sse2(y_ptr + 8 * y_stride, + y_stride, blimit); + vp9_loop_filter_simple_horizontal_edge_sse2(y_ptr + 12 * y_stride, + y_stride, blimit); } - /* Vertical B Filtering */ -void vp9_loop_filter_bv_sse2(unsigned char *y_ptr, unsigned char *u_ptr, unsigned char *v_ptr, - int y_stride, int uv_stride, struct loop_filter_info *lfi) { - vp9_loop_filter_vertical_edge_sse2(y_ptr + 4, y_stride, lfi->blim, lfi->lim, lfi->hev_thr, 2); - vp9_loop_filter_vertical_edge_sse2(y_ptr + 8, y_stride, lfi->blim, lfi->lim, lfi->hev_thr, 2); - vp9_loop_filter_vertical_edge_sse2(y_ptr + 12, y_stride, lfi->blim, lfi->lim, lfi->hev_thr, 2); +void vp9_loop_filter_bv_sse2(unsigned char *y_ptr, + unsigned char *u_ptr, unsigned char *v_ptr, + int y_stride, int uv_stride, + struct loop_filter_info *lfi) { + vp9_loop_filter_vertical_edge_sse2(y_ptr + 4, y_stride, + lfi->blim, lfi->lim, lfi->hev_thr, 2); + vp9_loop_filter_vertical_edge_sse2(y_ptr + 8, y_stride, + lfi->blim, lfi->lim, lfi->hev_thr, 2); + vp9_loop_filter_vertical_edge_sse2(y_ptr + 12, y_stride, + lfi->blim, lfi->lim, lfi->hev_thr, 2); if (u_ptr) - vp9_loop_filter_vertical_edge_uv_sse2(u_ptr + 4, uv_stride, lfi->blim, lfi->lim, lfi->hev_thr, v_ptr + 4); + vp9_loop_filter_vertical_edge_uv_sse2(u_ptr + 4, uv_stride, + lfi->blim, lfi->lim, lfi->hev_thr, + v_ptr + 4); } - -void vp9_loop_filter_bvs_sse2(unsigned char *y_ptr, int y_stride, const unsigned char *blimit) { +void vp9_loop_filter_bvs_sse2(unsigned char *y_ptr, int y_stride, + const unsigned char *blimit) { vp9_loop_filter_simple_vertical_edge_sse2(y_ptr + 4, y_stride, blimit); vp9_loop_filter_simple_vertical_edge_sse2(y_ptr + 8, y_stride, blimit); vp9_loop_filter_simple_vertical_edge_sse2(y_ptr + 12, y_stride, blimit); diff --git a/vp8/common/x86/loopfilter_x86.h b/vp8/common/x86/loopfilter_x86.h index da5ed315e..25cf383c9 100644 --- a/vp8/common/x86/loopfilter_x86.h +++ b/vp8/common/x86/loopfilter_x86.h @@ -8,7 +8,6 @@ * be found in the AUTHORS file in the root of the source tree. */ - #ifndef LOOPFILTER_X86_H #define LOOPFILTER_X86_H @@ -30,7 +29,6 @@ extern prototype_simple_loopfilter(vp9_loop_filter_simple_horizontal_edge_mmx); extern prototype_simple_loopfilter(vp9_loop_filter_bhs_mmx); #endif - #if HAVE_SSE2 extern prototype_loopfilter_block(vp9_loop_filter_mbv_sse2); extern prototype_loopfilter_block(vp9_loop_filter_bv_sse2); @@ -42,5 +40,4 @@ extern prototype_simple_loopfilter(vp9_loop_filter_simple_horizontal_edge_sse2); extern prototype_simple_loopfilter(vp9_loop_filter_bhs_sse2); #endif - -#endif +#endif // LOOPFILTER_X86_H diff --git a/vp8/common/x86/vp8_asm_stubs.c b/vp8/common/x86/vp8_asm_stubs.c index 86ae5a674..671e05b18 100644 --- a/vp8/common/x86/vp8_asm_stubs.c +++ b/vp8/common/x86/vp8_asm_stubs.c @@ -14,516 +14,489 @@ #include "vp8/common/subpixel.h" extern const short vp9_six_tap_mmx[16][6 * 8]; + extern const short vp9_bilinear_filters_8x_mmx[16][2 * 8]; -extern void vp9_filter_block1d_h6_mmx -( - unsigned char *src_ptr, - unsigned short *output_ptr, - unsigned int src_pixels_per_line, - unsigned int pixel_step, - unsigned int output_height, - unsigned int output_width, - const short *vp9_filter -); -extern void vp9_filter_block1dc_v6_mmx -( - unsigned short *src_ptr, - unsigned char *output_ptr, - int output_pitch, - unsigned int pixels_per_line, - unsigned int pixel_step, - unsigned int output_height, - unsigned int output_width, - const short *vp9_filter -); -extern void vp9_filter_block1d8_h6_sse2 -( - unsigned char *src_ptr, - unsigned short *output_ptr, - unsigned int src_pixels_per_line, - unsigned int pixel_step, - unsigned int output_height, - unsigned int output_width, - const short *vp9_filter -); -extern void vp9_filter_block1d16_h6_sse2 -( - unsigned char *src_ptr, - unsigned short *output_ptr, - unsigned int src_pixels_per_line, - unsigned int pixel_step, - unsigned int output_height, - unsigned int output_width, - const short *vp9_filter -); -extern void vp9_filter_block1d8_v6_sse2 -( - unsigned short *src_ptr, - unsigned char *output_ptr, - int dst_ptich, - unsigned int pixels_per_line, - unsigned int pixel_step, - unsigned int output_height, - unsigned int output_width, - const short *vp9_filter -); -extern void vp9_filter_block1d16_v6_sse2 -( - unsigned short *src_ptr, - unsigned char *output_ptr, - int dst_ptich, - unsigned int pixels_per_line, - unsigned int pixel_step, - unsigned int output_height, - unsigned int output_width, - const short *vp9_filter -); -extern void vp9_unpack_block1d16_h6_sse2 -( - unsigned char *src_ptr, - unsigned short *output_ptr, - unsigned int src_pixels_per_line, - unsigned int output_height, - unsigned int output_width -); -extern void vp9_filter_block1d8_h6_only_sse2 -( - unsigned char *src_ptr, - unsigned int src_pixels_per_line, - unsigned char *output_ptr, - int dst_ptich, - unsigned int output_height, - const short *vp9_filter -); -extern void vp9_filter_block1d16_h6_only_sse2 -( - unsigned char *src_ptr, - unsigned int src_pixels_per_line, - unsigned char *output_ptr, - int dst_ptich, - unsigned int output_height, - const short *vp9_filter -); -extern void vp9_filter_block1d8_v6_only_sse2 -( - unsigned char *src_ptr, - unsigned int src_pixels_per_line, - unsigned char *output_ptr, - int dst_ptich, - unsigned int output_height, - const short *vp9_filter -); -extern prototype_subpixel_predict(vp9_bilinear_predict8x8_mmx); +extern void vp9_filter_block1d_h6_mmx(unsigned char *src_ptr, + unsigned short *output_ptr, + unsigned int src_pixels_per_line, + unsigned int pixel_step, + unsigned int output_height, + unsigned int output_width, + const short *vp9_filter); + +extern void vp9_filter_block1dc_v6_mmx(unsigned short *src_ptr, + unsigned char *output_ptr, + int output_pitch, + unsigned int pixels_per_line, + unsigned int pixel_step, + unsigned int output_height, + unsigned int output_width, + const short *vp9_filter); + +extern void vp9_filter_block1d8_h6_sse2(unsigned char *src_ptr, + unsigned short *output_ptr, + unsigned int src_pixels_per_line, + unsigned int pixel_step, + unsigned int output_height, + unsigned int output_width, + const short *vp9_filter); + +extern void vp9_filter_block1d16_h6_sse2(unsigned char *src_ptr, + unsigned short *output_ptr, + unsigned int src_pixels_per_line, + unsigned int pixel_step, + unsigned int output_height, + unsigned int output_width, + const short *vp9_filter); + +extern void vp9_filter_block1d8_v6_sse2(unsigned short *src_ptr, + unsigned char *output_ptr, + int dst_ptich, + unsigned int pixels_per_line, + unsigned int pixel_step, + unsigned int output_height, + unsigned int output_width, + const short *vp9_filter); + +extern void vp9_filter_block1d16_v6_sse2(unsigned short *src_ptr, + unsigned char *output_ptr, + int dst_ptich, + unsigned int pixels_per_line, + unsigned int pixel_step, + unsigned int output_height, + unsigned int output_width, + const short *vp9_filter); + +extern void vp9_unpack_block1d16_h6_sse2(unsigned char *src_ptr, + unsigned short *output_ptr, + unsigned int src_pixels_per_line, + unsigned int output_height, + unsigned int output_width); + +extern void vp9_filter_block1d8_h6_only_sse2(unsigned char *src_ptr, + unsigned int src_pixels_per_line, + unsigned char *output_ptr, + int dst_pitch, + unsigned int output_height, + const short *vp9_filter); + +extern void vp9_filter_block1d16_h6_only_sse2(unsigned char *src_ptr, + unsigned int src_pixels_per_lin, + unsigned char *output_ptr, + int dst_pitch, + unsigned int output_height, + const short *vp9_filter); + +extern void vp9_filter_block1d8_v6_only_sse2(unsigned char *src_ptr, + unsigned int src_pixels_per_line, + unsigned char *output_ptr, + int dst_pitch, + unsigned int output_height, + const short *vp9_filter); +extern prototype_subpixel_predict(vp9_bilinear_predict8x8_mmx); #if HAVE_MMX -void vp9_sixtap_predict4x4_mmx -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { +void vp9_sixtap_predict4x4_mmx(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { #ifdef ANNOUNCE_FUNCTION printf("vp9_sixtap_predict4x4_mmx\n"); #endif - DECLARE_ALIGNED_ARRAY(16, unsigned short, FData2, 16 * 16); /* Temp data bufffer used in filtering */ - const short *HFilter, *VFilter; - HFilter = vp9_six_tap_mmx[xoffset]; - vp9_filter_block1d_h6_mmx(src_ptr - (2 * src_pixels_per_line), FData2, src_pixels_per_line, 1, 9, 8, HFilter); - VFilter = vp9_six_tap_mmx[yoffset]; - vp9_filter_block1dc_v6_mmx(FData2 + 8, dst_ptr, dst_pitch, 8, 4, 4, 4, VFilter); - + /* Temp data bufffer used in filtering */ + DECLARE_ALIGNED_ARRAY(16, unsigned short, fdata2, 16 * 16); + const short *hfilter, *vfilter; + hfilter = vp9_six_tap_mmx[xoffset]; + vp9_filter_block1d_h6_mmx(src_ptr - (2 * src_pixels_per_line), fdata2, + src_pixels_per_line, 1, 9, 8, hfilter); + vfilter = vp9_six_tap_mmx[yoffset]; + vp9_filter_block1dc_v6_mmx(fdata2 + 8, dst_ptr, dst_pitch, + 8, 4, 4, 4, vfilter); } - -void vp9_sixtap_predict16x16_mmx -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { - +void vp9_sixtap_predict16x16_mmx(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { #ifdef ANNOUNCE_FUNCTION printf("vp9_sixtap_predict16x16_mmx\n"); #endif - DECLARE_ALIGNED_ARRAY(16, unsigned short, FData2, 24 * 24); /* Temp data bufffer used in filtering */ - - const short *HFilter, *VFilter; - - - HFilter = vp9_six_tap_mmx[xoffset]; - - vp9_filter_block1d_h6_mmx(src_ptr - (2 * src_pixels_per_line), FData2, src_pixels_per_line, 1, 21, 32, HFilter); - vp9_filter_block1d_h6_mmx(src_ptr - (2 * src_pixels_per_line) + 4, FData2 + 4, src_pixels_per_line, 1, 21, 32, HFilter); - vp9_filter_block1d_h6_mmx(src_ptr - (2 * src_pixels_per_line) + 8, FData2 + 8, src_pixels_per_line, 1, 21, 32, HFilter); - vp9_filter_block1d_h6_mmx(src_ptr - (2 * src_pixels_per_line) + 12, FData2 + 12, src_pixels_per_line, 1, 21, 32, HFilter); - - VFilter = vp9_six_tap_mmx[yoffset]; - vp9_filter_block1dc_v6_mmx(FData2 + 32, dst_ptr, dst_pitch, 32, 16, 16, 16, VFilter); - vp9_filter_block1dc_v6_mmx(FData2 + 36, dst_ptr + 4, dst_pitch, 32, 16, 16, 16, VFilter); - vp9_filter_block1dc_v6_mmx(FData2 + 40, dst_ptr + 8, dst_pitch, 32, 16, 16, 16, VFilter); - vp9_filter_block1dc_v6_mmx(FData2 + 44, dst_ptr + 12, dst_pitch, 32, 16, 16, 16, VFilter); - + /* Temp data bufffer used in filtering */ + DECLARE_ALIGNED_ARRAY(16, unsigned short, fdata2, 24 * 24); + const short *hfilter, *vfilter; + + hfilter = vp9_six_tap_mmx[xoffset]; + vp9_filter_block1d_h6_mmx(src_ptr - (2 * src_pixels_per_line), + fdata2, src_pixels_per_line, 1, 21, 32, + hfilter); + vp9_filter_block1d_h6_mmx(src_ptr - (2 * src_pixels_per_line) + 4, + fdata2 + 4, src_pixels_per_line, 1, 21, 32, + hfilter); + vp9_filter_block1d_h6_mmx(src_ptr - (2 * src_pixels_per_line) + 8, + fdata2 + 8, src_pixels_per_line, 1, 21, 32, + hfilter); + vp9_filter_block1d_h6_mmx(src_ptr - (2 * src_pixels_per_line) + 12, + fdata2 + 12, src_pixels_per_line, 1, 21, 32, + hfilter); + + vfilter = vp9_six_tap_mmx[yoffset]; + vp9_filter_block1dc_v6_mmx(fdata2 + 32, dst_ptr, dst_pitch, + 32, 16, 16, 16, vfilter); + vp9_filter_block1dc_v6_mmx(fdata2 + 36, dst_ptr + 4, dst_pitch, + 32, 16, 16, 16, vfilter); + vp9_filter_block1dc_v6_mmx(fdata2 + 40, dst_ptr + 8, dst_pitch, + 32, 16, 16, 16, vfilter); + vp9_filter_block1dc_v6_mmx(fdata2 + 44, dst_ptr + 12, dst_pitch, + 32, 16, 16, 16, vfilter); } - -void vp9_sixtap_predict8x8_mmx -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { - +void vp9_sixtap_predict8x8_mmx(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { #ifdef ANNOUNCE_FUNCTION printf("vp9_sixtap_predict8x8_mmx\n"); #endif - DECLARE_ALIGNED_ARRAY(16, unsigned short, FData2, 256); /* Temp data bufffer used in filtering */ - - const short *HFilter, *VFilter; - - HFilter = vp9_six_tap_mmx[xoffset]; - vp9_filter_block1d_h6_mmx(src_ptr - (2 * src_pixels_per_line), FData2, src_pixels_per_line, 1, 13, 16, HFilter); - vp9_filter_block1d_h6_mmx(src_ptr - (2 * src_pixels_per_line) + 4, FData2 + 4, src_pixels_per_line, 1, 13, 16, HFilter); - - VFilter = vp9_six_tap_mmx[yoffset]; - vp9_filter_block1dc_v6_mmx(FData2 + 16, dst_ptr, dst_pitch, 16, 8, 8, 8, VFilter); - vp9_filter_block1dc_v6_mmx(FData2 + 20, dst_ptr + 4, dst_pitch, 16, 8, 8, 8, VFilter); - + /* Temp data bufffer used in filtering */ + DECLARE_ALIGNED_ARRAY(16, unsigned short, fdata2, 256); + const short *hfilter, *vfilter; + + hfilter = vp9_six_tap_mmx[xoffset]; + vp9_filter_block1d_h6_mmx(src_ptr - (2 * src_pixels_per_line), + fdata2, src_pixels_per_line, 1, 13, 16, + hfilter); + vp9_filter_block1d_h6_mmx(src_ptr - (2 * src_pixels_per_line) + 4, + fdata2 + 4, src_pixels_per_line, 1, 13, 16, + hfilter); + + vfilter = vp9_six_tap_mmx[yoffset]; + vp9_filter_block1dc_v6_mmx(fdata2 + 16, dst_ptr, dst_pitch, + 16, 8, 8, 8, vfilter); + vp9_filter_block1dc_v6_mmx(fdata2 + 20, dst_ptr + 4, dst_pitch, + 16, 8, 8, 8, vfilter); } - -void vp9_sixtap_predict8x4_mmx -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { +void vp9_sixtap_predict8x4_mmx(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { #ifdef ANNOUNCE_FUNCTION printf("vp9_sixtap_predict8x4_mmx\n"); #endif - DECLARE_ALIGNED_ARRAY(16, unsigned short, FData2, 256); /* Temp data bufffer used in filtering */ - - const short *HFilter, *VFilter; - - HFilter = vp9_six_tap_mmx[xoffset]; - vp9_filter_block1d_h6_mmx(src_ptr - (2 * src_pixels_per_line), FData2, src_pixels_per_line, 1, 9, 16, HFilter); - vp9_filter_block1d_h6_mmx(src_ptr - (2 * src_pixels_per_line) + 4, FData2 + 4, src_pixels_per_line, 1, 9, 16, HFilter); - - VFilter = vp9_six_tap_mmx[yoffset]; - vp9_filter_block1dc_v6_mmx(FData2 + 16, dst_ptr, dst_pitch, 16, 8, 4, 8, VFilter); - vp9_filter_block1dc_v6_mmx(FData2 + 20, dst_ptr + 4, dst_pitch, 16, 8, 4, 8, VFilter); - + /* Temp data bufffer used in filtering */ + DECLARE_ALIGNED_ARRAY(16, unsigned short, fdata2, 256); + const short *hfilter, *vfilter; + + hfilter = vp9_six_tap_mmx[xoffset]; + vp9_filter_block1d_h6_mmx(src_ptr - (2 * src_pixels_per_line), + fdata2, src_pixels_per_line, 1, 9, 16, hfilter); + vp9_filter_block1d_h6_mmx(src_ptr - (2 * src_pixels_per_line) + 4, + fdata2 + 4, src_pixels_per_line, 1, 9, 16, hfilter); + + vfilter = vp9_six_tap_mmx[yoffset]; + vp9_filter_block1dc_v6_mmx(fdata2 + 16, dst_ptr, dst_pitch, + 16, 8, 4, 8, vfilter); + vp9_filter_block1dc_v6_mmx(fdata2 + 20, dst_ptr + 4, dst_pitch, + 16, 8, 4, 8, vfilter); } - - -void vp9_bilinear_predict16x16_mmx -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { - vp9_bilinear_predict8x8_mmx(src_ptr, src_pixels_per_line, xoffset, yoffset, dst_ptr, dst_pitch); - vp9_bilinear_predict8x8_mmx(src_ptr + 8, src_pixels_per_line, xoffset, yoffset, dst_ptr + 8, dst_pitch); - vp9_bilinear_predict8x8_mmx(src_ptr + 8 * src_pixels_per_line, src_pixels_per_line, xoffset, yoffset, dst_ptr + dst_pitch * 8, dst_pitch); - vp9_bilinear_predict8x8_mmx(src_ptr + 8 * src_pixels_per_line + 8, src_pixels_per_line, xoffset, yoffset, dst_ptr + dst_pitch * 8 + 8, dst_pitch); +void vp9_bilinear_predict16x16_mmx(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { + vp9_bilinear_predict8x8_mmx(src_ptr, + src_pixels_per_line, xoffset, yoffset, + dst_ptr, dst_pitch); + vp9_bilinear_predict8x8_mmx(src_ptr + 8, + src_pixels_per_line, xoffset, yoffset, + dst_ptr + 8, dst_pitch); + vp9_bilinear_predict8x8_mmx(src_ptr + 8 * src_pixels_per_line, + src_pixels_per_line, xoffset, yoffset, + dst_ptr + dst_pitch * 8, dst_pitch); + vp9_bilinear_predict8x8_mmx(src_ptr + 8 * src_pixels_per_line + 8, + src_pixels_per_line, xoffset, yoffset, + dst_ptr + dst_pitch * 8 + 8, dst_pitch); } #endif - #if HAVE_SSE2 -void vp9_sixtap_predict16x16_sse2 -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch - -) { - DECLARE_ALIGNED_ARRAY(16, unsigned short, FData2, 24 * 24); /* Temp data bufffer used in filtering */ - - const short *HFilter, *VFilter; +void vp9_sixtap_predict16x16_sse2(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { + /* Temp data bufffer used in filtering */ + DECLARE_ALIGNED_ARRAY(16, unsigned short, fdata2, 24 * 24); + const short *hfilter, *vfilter; #ifdef ANNOUNCE_FUNCTION printf("vp9_sixtap_predict16x16_sse2\n"); #endif if (xoffset) { if (yoffset) { - HFilter = vp9_six_tap_mmx[xoffset]; - vp9_filter_block1d16_h6_sse2(src_ptr - (2 * src_pixels_per_line), FData2, src_pixels_per_line, 1, 21, 32, HFilter); - VFilter = vp9_six_tap_mmx[yoffset]; - vp9_filter_block1d16_v6_sse2(FData2 + 32, dst_ptr, dst_pitch, 32, 16, 16, dst_pitch, VFilter); + hfilter = vp9_six_tap_mmx[xoffset]; + vp9_filter_block1d16_h6_sse2(src_ptr - (2 * src_pixels_per_line), fdata2, + src_pixels_per_line, 1, 21, 32, hfilter); + vfilter = vp9_six_tap_mmx[yoffset]; + vp9_filter_block1d16_v6_sse2(fdata2 + 32, dst_ptr, dst_pitch, + 32, 16, 16, dst_pitch, vfilter); } else { /* First-pass only */ - HFilter = vp9_six_tap_mmx[xoffset]; - vp9_filter_block1d16_h6_only_sse2(src_ptr, src_pixels_per_line, dst_ptr, dst_pitch, 16, HFilter); + hfilter = vp9_six_tap_mmx[xoffset]; + vp9_filter_block1d16_h6_only_sse2(src_ptr, src_pixels_per_line, + dst_ptr, dst_pitch, 16, hfilter); } } else { /* Second-pass only */ - VFilter = vp9_six_tap_mmx[yoffset]; - vp9_unpack_block1d16_h6_sse2(src_ptr - (2 * src_pixels_per_line), FData2, src_pixels_per_line, 21, 32); - vp9_filter_block1d16_v6_sse2(FData2 + 32, dst_ptr, dst_pitch, 32, 16, 16, dst_pitch, VFilter); + vfilter = vp9_six_tap_mmx[yoffset]; + vp9_unpack_block1d16_h6_sse2(src_ptr - (2 * src_pixels_per_line), fdata2, + src_pixels_per_line, 21, 32); + vp9_filter_block1d16_v6_sse2(fdata2 + 32, dst_ptr, dst_pitch, + 32, 16, 16, dst_pitch, vfilter); } } - -void vp9_sixtap_predict8x8_sse2 -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { - DECLARE_ALIGNED_ARRAY(16, unsigned short, FData2, 256); /* Temp data bufffer used in filtering */ - const short *HFilter, *VFilter; +void vp9_sixtap_predict8x8_sse2(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { + /* Temp data bufffer used in filtering */ + DECLARE_ALIGNED_ARRAY(16, unsigned short, fdata2, 256); + const short *hfilter, *vfilter; #ifdef ANNOUNCE_FUNCTION printf("vp9_sixtap_predict8x8_sse2\n"); #endif if (xoffset) { if (yoffset) { - HFilter = vp9_six_tap_mmx[xoffset]; - vp9_filter_block1d8_h6_sse2(src_ptr - (2 * src_pixels_per_line), FData2, src_pixels_per_line, 1, 13, 16, HFilter); - VFilter = vp9_six_tap_mmx[yoffset]; - vp9_filter_block1d8_v6_sse2(FData2 + 16, dst_ptr, dst_pitch, 16, 8, 8, dst_pitch, VFilter); + hfilter = vp9_six_tap_mmx[xoffset]; + vp9_filter_block1d8_h6_sse2(src_ptr - (2 * src_pixels_per_line), fdata2, + src_pixels_per_line, 1, 13, 16, hfilter); + vfilter = vp9_six_tap_mmx[yoffset]; + vp9_filter_block1d8_v6_sse2(fdata2 + 16, dst_ptr, dst_pitch, + 16, 8, 8, dst_pitch, vfilter); } else { /* First-pass only */ - HFilter = vp9_six_tap_mmx[xoffset]; - vp9_filter_block1d8_h6_only_sse2(src_ptr, src_pixels_per_line, dst_ptr, dst_pitch, 8, HFilter); + hfilter = vp9_six_tap_mmx[xoffset]; + vp9_filter_block1d8_h6_only_sse2(src_ptr, src_pixels_per_line, + dst_ptr, dst_pitch, 8, hfilter); } } else { /* Second-pass only */ - VFilter = vp9_six_tap_mmx[yoffset]; - vp9_filter_block1d8_v6_only_sse2(src_ptr - (2 * src_pixels_per_line), src_pixels_per_line, dst_ptr, dst_pitch, 8, VFilter); + vfilter = vp9_six_tap_mmx[yoffset]; + vp9_filter_block1d8_v6_only_sse2(src_ptr - (2 * src_pixels_per_line), + src_pixels_per_line, + dst_ptr, dst_pitch, 8, vfilter); } } - -void vp9_sixtap_predict8x4_sse2 -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { - DECLARE_ALIGNED_ARRAY(16, unsigned short, FData2, 256); /* Temp data bufffer used in filtering */ - const short *HFilter, *VFilter; +void vp9_sixtap_predict8x4_sse2(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { + /* Temp data bufffer used in filtering */ + DECLARE_ALIGNED_ARRAY(16, unsigned short, fdata2, 256); + const short *hfilter, *vfilter; #ifdef ANNOUNCE_FUNCTION printf("vp9_sixtap_predict8x4_sse2\n"); #endif if (xoffset) { if (yoffset) { - HFilter = vp9_six_tap_mmx[xoffset]; - vp9_filter_block1d8_h6_sse2(src_ptr - (2 * src_pixels_per_line), FData2, src_pixels_per_line, 1, 9, 16, HFilter); - VFilter = vp9_six_tap_mmx[yoffset]; - vp9_filter_block1d8_v6_sse2(FData2 + 16, dst_ptr, dst_pitch, 16, 8, 4, dst_pitch, VFilter); + hfilter = vp9_six_tap_mmx[xoffset]; + vp9_filter_block1d8_h6_sse2(src_ptr - (2 * src_pixels_per_line), fdata2, + src_pixels_per_line, 1, 9, 16, hfilter); + vfilter = vp9_six_tap_mmx[yoffset]; + vp9_filter_block1d8_v6_sse2(fdata2 + 16, dst_ptr, dst_pitch, + 16, 8, 4, dst_pitch, vfilter); } else { /* First-pass only */ - HFilter = vp9_six_tap_mmx[xoffset]; - vp9_filter_block1d8_h6_only_sse2(src_ptr, src_pixels_per_line, dst_ptr, dst_pitch, 4, HFilter); + hfilter = vp9_six_tap_mmx[xoffset]; + vp9_filter_block1d8_h6_only_sse2(src_ptr, src_pixels_per_line, + dst_ptr, dst_pitch, 4, hfilter); } } else { /* Second-pass only */ - VFilter = vp9_six_tap_mmx[yoffset]; - vp9_filter_block1d8_v6_only_sse2(src_ptr - (2 * src_pixels_per_line), src_pixels_per_line, dst_ptr, dst_pitch, 4, VFilter); + vfilter = vp9_six_tap_mmx[yoffset]; + vp9_filter_block1d8_v6_only_sse2(src_ptr - (2 * src_pixels_per_line), + src_pixels_per_line, + dst_ptr, dst_pitch, 4, vfilter); } } - #endif #if HAVE_SSSE3 - -extern void vp9_filter_block1d8_h6_ssse3 -( - unsigned char *src_ptr, - unsigned int src_pixels_per_line, - unsigned char *output_ptr, - unsigned int output_pitch, - unsigned int output_height, - unsigned int vp9_filter_index -); - -extern void vp9_filter_block1d16_h6_ssse3 -( - unsigned char *src_ptr, - unsigned int src_pixels_per_line, - unsigned char *output_ptr, - unsigned int output_pitch, - unsigned int output_height, - unsigned int vp9_filter_index -); - -extern void vp9_filter_block1d16_v6_ssse3 -( - unsigned char *src_ptr, - unsigned int src_pitch, - unsigned char *output_ptr, - unsigned int out_pitch, - unsigned int output_height, - unsigned int vp9_filter_index -); - -extern void vp9_filter_block1d8_v6_ssse3 -( - unsigned char *src_ptr, - unsigned int src_pitch, - unsigned char *output_ptr, - unsigned int out_pitch, - unsigned int output_height, - unsigned int vp9_filter_index -); - -extern void vp9_filter_block1d4_h6_ssse3 -( - unsigned char *src_ptr, - unsigned int src_pixels_per_line, - unsigned char *output_ptr, - unsigned int output_pitch, - unsigned int output_height, - unsigned int vp9_filter_index -); - -extern void vp9_filter_block1d4_v6_ssse3 -( - unsigned char *src_ptr, - unsigned int src_pitch, - unsigned char *output_ptr, - unsigned int out_pitch, - unsigned int output_height, - unsigned int vp9_filter_index -); - -void vp9_sixtap_predict16x16_ssse3 -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch - -) { - DECLARE_ALIGNED_ARRAY(16, unsigned char, FData2, 24 * 24); +extern void vp9_filter_block1d8_h6_ssse3(unsigned char *src_ptr, + unsigned int src_pixels_per_line, + unsigned char *output_ptr, + unsigned int output_pitch, + unsigned int output_height, + unsigned int vp9_filter_index); + +extern void vp9_filter_block1d16_h6_ssse3(unsigned char *src_ptr, + unsigned int src_pixels_per_line, + unsigned char *output_ptr, + unsigned int output_pitch, + unsigned int output_height, + unsigned int vp9_filter_index); + +extern void vp9_filter_block1d16_v6_ssse3(unsigned char *src_ptr, + unsigned int src_pitch, + unsigned char *output_ptr, + unsigned int out_pitch, + unsigned int output_height, + unsigned int vp9_filter_index); + +extern void vp9_filter_block1d8_v6_ssse3(unsigned char *src_ptr, + unsigned int src_pitch, + unsigned char *output_ptr, + unsigned int out_pitch, + unsigned int output_height, + unsigned int vp9_filter_index); + +extern void vp9_filter_block1d4_h6_ssse3(unsigned char *src_ptr, + unsigned int src_pixels_per_line, + unsigned char *output_ptr, + unsigned int output_pitch, + unsigned int output_height, + unsigned int vp9_filter_index); + +extern void vp9_filter_block1d4_v6_ssse3(unsigned char *src_ptr, + unsigned int src_pitch, + unsigned char *output_ptr, + unsigned int out_pitch, + unsigned int output_height, + unsigned int vp9_filter_index); + +void vp9_sixtap_predict16x16_ssse3(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { + DECLARE_ALIGNED_ARRAY(16, unsigned char, fdata2, 24 * 24); #ifdef ANNOUNCE_FUNCTION printf("vp9_sixtap_predict16x16_ssse3\n"); #endif if (xoffset) { if (yoffset) { - vp9_filter_block1d16_h6_ssse3(src_ptr - (2 * src_pixels_per_line), src_pixels_per_line, FData2, 16, 21, xoffset); - vp9_filter_block1d16_v6_ssse3(FData2, 16, dst_ptr, dst_pitch, 16, yoffset); + vp9_filter_block1d16_h6_ssse3(src_ptr - (2 * src_pixels_per_line), + src_pixels_per_line, + fdata2, 16, 21, xoffset); + vp9_filter_block1d16_v6_ssse3(fdata2, 16, dst_ptr, dst_pitch, + 16, yoffset); } else { /* First-pass only */ - vp9_filter_block1d16_h6_ssse3(src_ptr, src_pixels_per_line, dst_ptr, dst_pitch, 16, xoffset); + vp9_filter_block1d16_h6_ssse3(src_ptr, src_pixels_per_line, + dst_ptr, dst_pitch, 16, xoffset); } } else { /* Second-pass only */ - vp9_filter_block1d16_v6_ssse3(src_ptr - (2 * src_pixels_per_line), src_pixels_per_line, dst_ptr, dst_pitch, 16, yoffset); + vp9_filter_block1d16_v6_ssse3(src_ptr - (2 * src_pixels_per_line), + src_pixels_per_line, + dst_ptr, dst_pitch, 16, yoffset); } } -void vp9_sixtap_predict8x8_ssse3 -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { - DECLARE_ALIGNED_ARRAY(16, unsigned char, FData2, 256); +void vp9_sixtap_predict8x8_ssse3(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { + DECLARE_ALIGNED_ARRAY(16, unsigned char, fdata2, 256); #ifdef ANNOUNCE_FUNCTION printf("vp9_sixtap_predict8x8_ssse3\n"); #endif if (xoffset) { if (yoffset) { - vp9_filter_block1d8_h6_ssse3(src_ptr - (2 * src_pixels_per_line), src_pixels_per_line, FData2, 8, 13, xoffset); - vp9_filter_block1d8_v6_ssse3(FData2, 8, dst_ptr, dst_pitch, 8, yoffset); + vp9_filter_block1d8_h6_ssse3(src_ptr - (2 * src_pixels_per_line), + src_pixels_per_line, fdata2, 8, 13, xoffset); + vp9_filter_block1d8_v6_ssse3(fdata2, 8, dst_ptr, dst_pitch, 8, yoffset); } else { - vp9_filter_block1d8_h6_ssse3(src_ptr, src_pixels_per_line, dst_ptr, dst_pitch, 8, xoffset); + vp9_filter_block1d8_h6_ssse3(src_ptr, src_pixels_per_line, + dst_ptr, dst_pitch, 8, xoffset); } } else { /* Second-pass only */ - vp9_filter_block1d8_v6_ssse3(src_ptr - (2 * src_pixels_per_line), src_pixels_per_line, dst_ptr, dst_pitch, 8, yoffset); + vp9_filter_block1d8_v6_ssse3(src_ptr - (2 * src_pixels_per_line), + src_pixels_per_line, + dst_ptr, dst_pitch, 8, yoffset); } } - -void vp9_sixtap_predict8x4_ssse3 -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { - DECLARE_ALIGNED_ARRAY(16, unsigned char, FData2, 256); +void vp9_sixtap_predict8x4_ssse3(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { + DECLARE_ALIGNED_ARRAY(16, unsigned char, fdata2, 256); #ifdef ANNOUNCE_FUNCTION printf("vp9_sixtap_predict8x4_ssse3\n"); #endif if (xoffset) { if (yoffset) { - vp9_filter_block1d8_h6_ssse3(src_ptr - (2 * src_pixels_per_line), src_pixels_per_line, FData2, 8, 9, xoffset); - vp9_filter_block1d8_v6_ssse3(FData2, 8, dst_ptr, dst_pitch, 4, yoffset); + vp9_filter_block1d8_h6_ssse3(src_ptr - (2 * src_pixels_per_line), + src_pixels_per_line, fdata2, 8, 9, xoffset); + vp9_filter_block1d8_v6_ssse3(fdata2, 8, dst_ptr, dst_pitch, 4, yoffset); } else { /* First-pass only */ - vp9_filter_block1d8_h6_ssse3(src_ptr, src_pixels_per_line, dst_ptr, dst_pitch, 4, xoffset); + vp9_filter_block1d8_h6_ssse3(src_ptr, src_pixels_per_line, + dst_ptr, dst_pitch, 4, xoffset); } } else { /* Second-pass only */ - vp9_filter_block1d8_v6_ssse3(src_ptr - (2 * src_pixels_per_line), src_pixels_per_line, dst_ptr, dst_pitch, 4, yoffset); + vp9_filter_block1d8_v6_ssse3(src_ptr - (2 * src_pixels_per_line), + src_pixels_per_line, + dst_ptr, dst_pitch, 4, yoffset); } } -void vp9_sixtap_predict4x4_ssse3 -( - unsigned char *src_ptr, - int src_pixels_per_line, - int xoffset, - int yoffset, - unsigned char *dst_ptr, - int dst_pitch -) { - DECLARE_ALIGNED_ARRAY(16, unsigned char, FData2, 4 * 9); +void vp9_sixtap_predict4x4_ssse3(unsigned char *src_ptr, + int src_pixels_per_line, + int xoffset, + int yoffset, + unsigned char *dst_ptr, + int dst_pitch) { + DECLARE_ALIGNED_ARRAY(16, unsigned char, fdata2, 4 * 9); #ifdef ANNOUNCE_FUNCTION printf("vp9_sixtap_predict4x4_ssse3\n"); #endif if (xoffset) { if (yoffset) { - vp9_filter_block1d4_h6_ssse3(src_ptr - (2 * src_pixels_per_line), src_pixels_per_line, FData2, 4, 9, xoffset); - vp9_filter_block1d4_v6_ssse3(FData2, 4, dst_ptr, dst_pitch, 4, yoffset); + vp9_filter_block1d4_h6_ssse3(src_ptr - (2 * src_pixels_per_line), + src_pixels_per_line, fdata2, 4, 9, xoffset); + vp9_filter_block1d4_v6_ssse3(fdata2, 4, dst_ptr, dst_pitch, 4, yoffset); } else { - vp9_filter_block1d4_h6_ssse3(src_ptr, src_pixels_per_line, dst_ptr, dst_pitch, 4, xoffset); + vp9_filter_block1d4_h6_ssse3(src_ptr, src_pixels_per_line, + dst_ptr, dst_pitch, 4, xoffset); } } else { - vp9_filter_block1d4_v6_ssse3(src_ptr - (2 * src_pixels_per_line), src_pixels_per_line, dst_ptr, dst_pitch, 4, yoffset); + vp9_filter_block1d4_v6_ssse3(src_ptr - (2 * src_pixels_per_line), + src_pixels_per_line, + dst_ptr, dst_pitch, 4, yoffset); } } @@ -533,6 +506,7 @@ void vp9_filter_block1d16_v8_ssse3(const unsigned char *src_ptr, unsigned int out_pitch, unsigned int output_height, const short *filter); + void vp9_filter_block1d16_h8_ssse3(const unsigned char *src_ptr, const unsigned int src_pitch, unsigned char *output_ptr, @@ -540,25 +514,26 @@ void vp9_filter_block1d16_h8_ssse3(const unsigned char *src_ptr, unsigned int output_height, const short *filter); -void vp9_filter_block2d_16x16_8_ssse3 -( - const unsigned char *src_ptr, const unsigned int src_stride, - const short *HFilter_aligned16, const short *VFilter_aligned16, - unsigned char *dst_ptr, unsigned int dst_stride -) { - if (HFilter_aligned16[3] !=128 && VFilter_aligned16[3] != 128) { - DECLARE_ALIGNED_ARRAY(16, unsigned char, FData2, 23 * 16); +void vp9_filter_block2d_16x16_8_ssse3(const unsigned char *src_ptr, + const unsigned int src_stride, + const short *hfilter_aligned16, + const short *vfilter_aligned16, + unsigned char *dst_ptr, + unsigned int dst_stride) { + if (hfilter_aligned16[3] != 128 && vfilter_aligned16[3] != 128) { + DECLARE_ALIGNED_ARRAY(16, unsigned char, fdata2, 23 * 16); + vp9_filter_block1d16_h8_ssse3(src_ptr - (3 * src_stride), src_stride, - FData2, 16, 23, HFilter_aligned16); - vp9_filter_block1d16_v8_ssse3(FData2, 16, dst_ptr, dst_stride, 16, - VFilter_aligned16); + fdata2, 16, 23, hfilter_aligned16); + vp9_filter_block1d16_v8_ssse3(fdata2, 16, dst_ptr, dst_stride, 16, + vfilter_aligned16); } else { - if (HFilter_aligned16[3] !=128) { + if (hfilter_aligned16[3] != 128) { vp9_filter_block1d16_h8_ssse3(src_ptr, src_stride, dst_ptr, dst_stride, - 16, HFilter_aligned16); + 16, hfilter_aligned16); } else { vp9_filter_block1d16_v8_ssse3(src_ptr - (3 * src_stride), src_stride, - dst_ptr, dst_stride, 16, VFilter_aligned16); + dst_ptr, dst_stride, 16, vfilter_aligned16); } } } @@ -569,54 +544,58 @@ void vp9_filter_block1d8_v8_ssse3(const unsigned char *src_ptr, unsigned int out_pitch, unsigned int output_height, const short *filter); + void vp9_filter_block1d8_h8_ssse3(const unsigned char *src_ptr, const unsigned int src_pitch, unsigned char *output_ptr, unsigned int out_pitch, unsigned int output_height, const short *filter); -void vp9_filter_block2d_8x8_8_ssse3 -( - const unsigned char *src_ptr, const unsigned int src_stride, - const short *HFilter_aligned16, const short *VFilter_aligned16, - unsigned char *dst_ptr, unsigned int dst_stride -) { - if (HFilter_aligned16[3] !=128 && VFilter_aligned16[3] != 128) { - DECLARE_ALIGNED_ARRAY(16, unsigned char, FData2, 23 * 16); + +void vp9_filter_block2d_8x8_8_ssse3(const unsigned char *src_ptr, + const unsigned int src_stride, + const short *hfilter_aligned16, + const short *vfilter_aligned16, + unsigned char *dst_ptr, + unsigned int dst_stride) { + if (hfilter_aligned16[3] != 128 && vfilter_aligned16[3] != 128) { + DECLARE_ALIGNED_ARRAY(16, unsigned char, fdata2, 23 * 16); + vp9_filter_block1d8_h8_ssse3(src_ptr - (3 * src_stride), src_stride, - FData2, 16, 15, HFilter_aligned16); - vp9_filter_block1d8_v8_ssse3(FData2, 16, dst_ptr, dst_stride, 8, - VFilter_aligned16); + fdata2, 16, 15, hfilter_aligned16); + vp9_filter_block1d8_v8_ssse3(fdata2, 16, dst_ptr, dst_stride, 8, + vfilter_aligned16); } else { - if (HFilter_aligned16[3] !=128) { + if (hfilter_aligned16[3] != 128) { vp9_filter_block1d8_h8_ssse3(src_ptr, src_stride, dst_ptr, dst_stride, 8, - HFilter_aligned16); + hfilter_aligned16); } else { vp9_filter_block1d8_v8_ssse3(src_ptr - (3 * src_stride), src_stride, - dst_ptr, dst_stride, 8, VFilter_aligned16); + dst_ptr, dst_stride, 8, vfilter_aligned16); } } } -void vp9_filter_block2d_8x4_8_ssse3 -( - const unsigned char *src_ptr, const unsigned int src_stride, - const short *HFilter_aligned16, const short *VFilter_aligned16, - unsigned char *dst_ptr, unsigned int dst_stride -) { - if (HFilter_aligned16[3] !=128 && VFilter_aligned16[3] != 128) { - DECLARE_ALIGNED_ARRAY(16, unsigned char, FData2, 23 * 16); +void vp9_filter_block2d_8x4_8_ssse3(const unsigned char *src_ptr, + const unsigned int src_stride, + const short *hfilter_aligned16, + const short *vfilter_aligned16, + unsigned char *dst_ptr, + unsigned int dst_stride) { + if (hfilter_aligned16[3] !=128 && vfilter_aligned16[3] != 128) { + DECLARE_ALIGNED_ARRAY(16, unsigned char, fdata2, 23 * 16); + vp9_filter_block1d8_h8_ssse3(src_ptr - (3 * src_stride), src_stride, - FData2, 16, 11, HFilter_aligned16); - vp9_filter_block1d8_v8_ssse3(FData2, 16, dst_ptr, dst_stride, 4, - VFilter_aligned16); + fdata2, 16, 11, hfilter_aligned16); + vp9_filter_block1d8_v8_ssse3(fdata2, 16, dst_ptr, dst_stride, 4, + vfilter_aligned16); } else { - if (HFilter_aligned16[3] !=128) { + if (hfilter_aligned16[3] != 128) { vp9_filter_block1d8_h8_ssse3(src_ptr, src_stride, dst_ptr, dst_stride, 4, - HFilter_aligned16); + hfilter_aligned16); } else { vp9_filter_block1d8_v8_ssse3(src_ptr - (3 * src_stride), src_stride, - dst_ptr, dst_stride, 4, VFilter_aligned16); + dst_ptr, dst_stride, 4, vfilter_aligned16); } } } diff --git a/vp8/common/x86/x86_systemdependent.c b/vp8/common/x86/x86_systemdependent.c index bf6456323..e814755b6 100644 --- a/vp8/common/x86/x86_systemdependent.c +++ b/vp8/common/x86/x86_systemdependent.c @@ -8,10 +8,8 @@ * be found in the AUTHORS file in the root of the source tree. */ - #include "vpx_config.h" #include "vpx_ports/x86.h" -#include "vp8/common/g_common.h" #include "vp8/common/subpixel.h" #include "vp8/common/loopfilter.h" #include "vp8/common/idct.h" diff --git a/vp8/decoder/onyxd_if.c b/vp8/decoder/onyxd_if.c index bce5075b6..c1630b008 100644 --- a/vp8/decoder/onyxd_if.c +++ b/vp8/decoder/onyxd_if.c @@ -20,7 +20,6 @@ #include "vpx_scale/yv12extend.h" #include "vp8/common/loopfilter.h" #include "vp8/common/swapyv12buffer.h" -#include "vp8/common/g_common.h" #include <stdio.h> #include <assert.h> diff --git a/vp8/decoder/x86/idct_blk_mmx.c b/vp8/decoder/x86/idct_blk_mmx.c index 6e58ab8ac..971b80cc6 100644 --- a/vp8/decoder/x86/idct_blk_mmx.c +++ b/vp8/decoder/x86/idct_blk_mmx.c @@ -12,9 +12,10 @@ #include "vp8/common/idct.h" #include "vp8/decoder/dequantize.h" -void vp9_dequant_dc_idct_add_y_block_mmx -(short *q, short *dq, unsigned char *pre, - unsigned char *dst, int stride, char *eobs, short *dc) { +void vp9_dequant_dc_idct_add_y_block_mmx(short *q, short *dq, + unsigned char *pre, + unsigned char *dst, + int stride, char *eobs, short *dc) { int i; for (i = 0; i < 4; i++) { @@ -49,9 +50,10 @@ void vp9_dequant_dc_idct_add_y_block_mmx } } -void vp9_dequant_idct_add_y_block_mmx -(short *q, short *dq, unsigned char *pre, - unsigned char *dst, int stride, char *eobs) { +void vp9_dequant_idct_add_y_block_mmx(short *q, short *dq, + unsigned char *pre, + unsigned char *dst, + int stride, char *eobs) { int i; for (i = 0; i < 4; i++) { @@ -90,9 +92,11 @@ void vp9_dequant_idct_add_y_block_mmx } } -void vp9_dequant_idct_add_uv_block_mmx -(short *q, short *dq, unsigned char *pre, - unsigned char *dstu, unsigned char *dstv, int stride, char *eobs) { +void vp9_dequant_idct_add_uv_block_mmx(short *q, short *dq, + unsigned char *pre, + unsigned char *dstu, + unsigned char *dstv, + int stride, char *eobs) { int i; for (i = 0; i < 2; i++) { diff --git a/vp8/decoder/x86/idct_blk_sse2.c b/vp8/decoder/x86/idct_blk_sse2.c index c70739961..dbda3ee8b 100644 --- a/vp8/decoder/x86/idct_blk_sse2.c +++ b/vp8/decoder/x86/idct_blk_sse2.c @@ -12,23 +12,26 @@ #include "vp8/common/idct.h" #include "vp8/decoder/dequantize.h" -void vp9_idct_dequant_dc_0_2x_sse2 -(short *q, short *dq, unsigned char *pre, - unsigned char *dst, int dst_stride, short *dc); -void vp9_idct_dequant_dc_full_2x_sse2 -(short *q, short *dq, unsigned char *pre, - unsigned char *dst, int dst_stride, short *dc); - -void vp9_idct_dequant_0_2x_sse2 -(short *q, short *dq, unsigned char *pre, - unsigned char *dst, int dst_stride, int blk_stride); -void vp9_idct_dequant_full_2x_sse2 -(short *q, short *dq, unsigned char *pre, - unsigned char *dst, int dst_stride, int blk_stride); - -void vp9_dequant_dc_idct_add_y_block_sse2 -(short *q, short *dq, unsigned char *pre, - unsigned char *dst, int stride, char *eobs, short *dc) { +void vp9_idct_dequant_dc_0_2x_sse2(short *q, short *dq, + unsigned char *pre, unsigned char *dst, + int dst_stride, short *dc); + +void vp9_idct_dequant_dc_full_2x_sse2(short *q, short *dq, + unsigned char *pre, unsigned char *dst, + int dst_stride, short *dc); + +void vp9_idct_dequant_0_2x_sse2(short *q, short *dq, + unsigned char *pre, unsigned char *dst, + int dst_stride, int blk_stride); + +void vp9_idct_dequant_full_2x_sse2(short *q, short *dq, + unsigned char *pre, unsigned char *dst, + int dst_stride, int blk_stride); + +void vp9_dequant_dc_idct_add_y_block_sse2(short *q, short *dq, + unsigned char *pre, + unsigned char *dst, + int stride, char *eobs, short *dc) { int i; for (i = 0; i < 4; i++) { @@ -38,9 +41,11 @@ void vp9_dequant_dc_idct_add_y_block_sse2 vp9_idct_dequant_dc_0_2x_sse2(q, dq, pre, dst, stride, dc); if (((short *)(eobs))[1] & 0xfefe) - vp9_idct_dequant_dc_full_2x_sse2(q + 32, dq, pre + 8, dst + 8, stride, dc + 2); + vp9_idct_dequant_dc_full_2x_sse2(q + 32, dq, pre + 8, dst + 8, + stride, dc + 2); else - vp9_idct_dequant_dc_0_2x_sse2(q + 32, dq, pre + 8, dst + 8, stride, dc + 2); + vp9_idct_dequant_dc_0_2x_sse2(q + 32, dq, pre + 8, dst + 8, + stride, dc + 2); q += 64; dc += 4; @@ -50,9 +55,9 @@ void vp9_dequant_dc_idct_add_y_block_sse2 } } -void vp9_dequant_idct_add_y_block_sse2 -(short *q, short *dq, unsigned char *pre, - unsigned char *dst, int stride, char *eobs) { +void vp9_dequant_idct_add_y_block_sse2(short *q, short *dq, + unsigned char *pre, unsigned char *dst, + int stride, char *eobs) { int i; for (i = 0; i < 4; i++) { @@ -73,9 +78,11 @@ void vp9_dequant_idct_add_y_block_sse2 } } -void vp9_dequant_idct_add_uv_block_sse2 -(short *q, short *dq, unsigned char *pre, - unsigned char *dstu, unsigned char *dstv, int stride, char *eobs) { +void vp9_dequant_idct_add_uv_block_sse2(short *q, short *dq, + unsigned char *pre, + unsigned char *dstu, + unsigned char *dstv, + int stride, char *eobs) { if (((short *)(eobs))[0] & 0xfefe) vp9_idct_dequant_full_2x_sse2(q, dq, pre, dstu, stride, 8); else diff --git a/vp8/decoder/x86/x86_dsystemdependent.c b/vp8/decoder/x86/x86_dsystemdependent.c index b1cd49440..19c03659a 100644 --- a/vp8/decoder/x86/x86_dsystemdependent.c +++ b/vp8/decoder/x86/x86_dsystemdependent.c @@ -8,12 +8,10 @@ * be found in the AUTHORS file in the root of the source tree. */ - #include "vpx_ports/config.h" #include "vpx_ports/x86.h" #include "vp8/decoder/onyxd_int.h" - #if HAVE_MMX void vp9_dequantize_b_impl_mmx(short *sq, short *dq, short *q); diff --git a/vp8/encoder/encodeframe.c b/vp8/encoder/encodeframe.c index 8695cb0f4..32777876f 100644 --- a/vp8/encoder/encodeframe.c +++ b/vp8/encoder/encodeframe.c @@ -2296,7 +2296,7 @@ void vp9_encode_inter_superblock(VP9_COMP *cpi, MACROBLOCK *x, TOKENEXTRA **t, vp9_optimize_mbuv_8x8(x, rtcd); } vp9_inverse_transform_mb_8x8(IF_RTCD(&rtcd->common->idct), &x->e_mbd); - vp9_recon_mby_s_c( &x->e_mbd, + vp9_recon_mby_s_c(&x->e_mbd, dst + x_idx * 16 + y_idx * 16 * dst_y_stride); vp9_recon_mbuv_s_c(&x->e_mbd, udst + x_idx * 8 + y_idx * 8 * dst_uv_stride, diff --git a/vp8/encoder/encodeintra.c b/vp8/encoder/encodeintra.c index b9354029d..e89e96c44 100644 --- a/vp8/encoder/encodeintra.c +++ b/vp8/encoder/encodeintra.c @@ -16,7 +16,6 @@ #include "vp8/common/reconintra4x4.h" #include "encodemb.h" #include "vp8/common/invtrans.h" -#include "vp8/common/g_common.h" #include "encodeintra.h" #if CONFIG_RUNTIME_CPU_DETECT @@ -79,7 +78,7 @@ void vp9_encode_intra4x4block(const VP9_ENCODER_RTCD *rtcd, } else { x->vp9_short_fdct4x4(be->src_diff, be->coeff, 32); x->quantize_b_4x4(be, b) ; - vp9_inverse_transform_b_4x4(IF_RTCD(&rtcd->common->idct), b, 32) ; + vp9_inverse_transform_b_4x4(IF_RTCD(&rtcd->common->idct), b, 32); } vp9_recon_b(b->predictor, b->diff, *(b->base_dst) + b->dst, b->dst_stride); @@ -265,7 +264,8 @@ void vp9_encode_intra_uv4x4(const VP9_ENCODER_RTCD *rtcd, x->quantize_b_4x4(be, b); vp9_inverse_transform_b_4x4(IF_RTCD(&rtcd->common->idct), b, 16); - vp9_recon_uv_b_c(b->predictor,b->diff, *(b->base_dst) + b->dst, b->dst_stride); + vp9_recon_uv_b_c(b->predictor, b->diff, *(b->base_dst) + b->dst, + b->dst_stride); } void vp9_encode_intra8x8mbuv(const VP9_ENCODER_RTCD *rtcd, MACROBLOCK *x) { diff --git a/vp8/encoder/encodeintra.h b/vp8/encoder/encodeintra.h index 905ccc183..38b42b71e 100644 --- a/vp8/encoder/encodeintra.h +++ b/vp8/encoder/encodeintra.h @@ -8,9 +8,9 @@ * be found in the AUTHORS file in the root of the source tree. */ +#ifndef __ENCODEINTRA_H_ +#define __ENCODEINTRA_H_ -#ifndef _ENCODEINTRA_H_ -#define _ENCODEINTRA_H_ #include "onyx_int.h" int vp9_encode_intra(VP9_COMP *cpi, MACROBLOCK *x, int use_16x16_pred); @@ -24,4 +24,4 @@ void vp9_encode_intra8x8mbuv(const VP9_ENCODER_RTCD *rtcd, MACROBLOCK *x); void vp9_encode_intra8x8(const VP9_ENCODER_RTCD *rtcd, MACROBLOCK *x, int ib); -#endif +#endif // __ENCODEINTRA_H_ diff --git a/vp8/encoder/firstpass.c b/vp8/encoder/firstpass.c index 8d015c49e..a7ae9189a 100644 --- a/vp8/encoder/firstpass.c +++ b/vp8/encoder/firstpass.c @@ -692,7 +692,8 @@ void vp9_first_pass(VP9_COMP *cpi) { x->src.v_buffer += 8 * x->src.uv_stride - 8 * cm->mb_cols; // extend the recon for intra prediction - vp9_extend_mb_row(new_yv12, xd->dst.y_buffer + 16, xd->dst.u_buffer + 8, xd->dst.v_buffer + 8); + vp9_extend_mb_row(new_yv12, xd->dst.y_buffer + 16, + xd->dst.u_buffer + 8, xd->dst.v_buffer + 8); vp9_clear_system_state(); // __asm emms; } diff --git a/vp8/encoder/mbgraph.c b/vp8/encoder/mbgraph.c index df852c715..d2e609876 100644 --- a/vp8/encoder/mbgraph.c +++ b/vp8/encoder/mbgraph.c @@ -18,12 +18,9 @@ #include <vpx_mem/vpx_mem.h> #include <vp8/encoder/segmentation.h> -static unsigned int do_16x16_motion_iteration -( - VP9_COMP *cpi, - int_mv *ref_mv, - int_mv *dst_mv -) { +static unsigned int do_16x16_motion_iteration(VP9_COMP *cpi, + int_mv *ref_mv, + int_mv *dst_mv) { MACROBLOCK *const x = &cpi->mb; MACROBLOCKD *const xd = &x->e_mbd; BLOCK *b = &x->block[0]; diff --git a/vp8/encoder/onyx_if.c b/vp8/encoder/onyx_if.c index fbce2020e..4fd6f84d4 100644 --- a/vp8/encoder/onyx_if.c +++ b/vp8/encoder/onyx_if.c @@ -23,7 +23,6 @@ #include "ratectrl.h" #include "vp8/common/quant_common.h" #include "segmentation.h" -#include "vp8/common/g_common.h" #include "vpx_scale/yv12extend.h" #if CONFIG_POSTPROC #include "vp8/common/postproc.h" diff --git a/vp8/encoder/rdopt.c b/vp8/encoder/rdopt.c index 78302d1f8..a30f9ce2a 100644 --- a/vp8/encoder/rdopt.c +++ b/vp8/encoder/rdopt.c @@ -29,7 +29,6 @@ #include "encodemb.h" #include "quantize.h" #include "vp8/common/idct.h" -#include "vp8/common/g_common.h" #include "variance.h" #include "mcomp.h" #include "rdopt.h" diff --git a/vp8/encoder/temporal_filter.c b/vp8/encoder/temporal_filter.c index 276248c70..7e532ea6c 100644 --- a/vp8/encoder/temporal_filter.c +++ b/vp8/encoder/temporal_filter.c @@ -22,7 +22,6 @@ #include "ratectrl.h" #include "vp8/common/quant_common.h" #include "segmentation.h" -#include "vp8/common/g_common.h" #include "vpx_scale/yv12extend.h" #include "vpx_mem/vpx_mem.h" #include "vp8/common/swapyv12buffer.h" diff --git a/vp8/vp8_common.mk b/vp8/vp8_common.mk index ed1923aa5..87c1e6c6d 100644 --- a/vp8/vp8_common.mk +++ b/vp8/vp8_common.mk @@ -37,7 +37,6 @@ VP8_COMMON_SRCS-yes += common/entropymode.h VP8_COMMON_SRCS-yes += common/entropymv.h VP8_COMMON_SRCS-yes += common/extend.h VP8_COMMON_SRCS-yes += common/findnearmv.h -VP8_COMMON_SRCS-yes += common/g_common.h VP8_COMMON_SRCS-yes += common/header.h VP8_COMMON_SRCS-yes += common/idct.h VP8_COMMON_SRCS-yes += common/invtrans.h |