generic-library/vpx
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Refactor fast loop filter code to handle 444.

Browse Source 
Change-Id: I921b1ebabdf617049f8fa26fbe462c3ff115c1ce
This commit is contained in:
Alex Converse 2015-03-20 14:51:59 -07:00
parent c77d4dcb35
commit a1e20ec58f
3 changed files with 245 additions and 249 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

438
vp9/common/vp9_loopfilter.c
View File

@ -293,7 +293,7 @@ void vp9_loop_filter_frame_init(VP9_COMMON *cm, int default_filt_lvl) {
} }
} }
static void filter_selectively_vert_row2(PLANE_TYPE plane_type, static void filter_selectively_vert_row2(int subsampling_factor,
uint8_t *s, int pitch, uint8_t *s, int pitch,
unsigned int mask_16x16_l, unsigned int mask_16x16_l,
unsigned int mask_8x8_l, unsigned int mask_8x8_l,
@ -301,9 +301,9 @@ static void filter_selectively_vert_row2(PLANE_TYPE plane_type,
unsigned int mask_4x4_int_l, unsigned int mask_4x4_int_l,
const loop_filter_info_n *lfi_n, const loop_filter_info_n *lfi_n,
const uint8_t *lfl) { const uint8_t *lfl) {
const int mask_shift = plane_type ? 4 : 8; const int mask_shift = subsampling_factor ? 4 : 8;
const int mask_cutoff = plane_type ? 0xf : 0xff; const int mask_cutoff = subsampling_factor ? 0xf : 0xff;
const int lfl_forward = plane_type ? 4 : 8; const int lfl_forward = subsampling_factor ? 4 : 8;
unsigned int mask_16x16_0 = mask_16x16_l & mask_cutoff; unsigned int mask_16x16_0 = mask_16x16_l & mask_cutoff;
unsigned int mask_8x8_0 = mask_8x8_l & mask_cutoff; unsigned int mask_8x8_0 = mask_8x8_l & mask_cutoff;
@ -393,7 +393,7 @@ static void filter_selectively_vert_row2(PLANE_TYPE plane_type,
} }
#if CONFIG_VP9_HIGHBITDEPTH #if CONFIG_VP9_HIGHBITDEPTH
static void highbd_filter_selectively_vert_row2(PLANE_TYPE plane_type, static void highbd_filter_selectively_vert_row2(int subsampling_factor,
uint16_t *s, int pitch, uint16_t *s, int pitch,
unsigned int mask_16x16_l, unsigned int mask_16x16_l,
unsigned int mask_8x8_l, unsigned int mask_8x8_l,
@ -401,9 +401,9 @@ static void highbd_filter_selectively_vert_row2(PLANE_TYPE plane_type,
unsigned int mask_4x4_int_l, unsigned int mask_4x4_int_l,
const loop_filter_info_n *lfi_n, const loop_filter_info_n *lfi_n,
const uint8_t *lfl, int bd) { const uint8_t *lfl, int bd) {
const int mask_shift = plane_type ? 4 : 8; const int mask_shift = subsampling_factor ? 4 : 8;
const int mask_cutoff = plane_type ? 0xf : 0xff; const int mask_cutoff = subsampling_factor ? 0xf : 0xff;
const int lfl_forward = plane_type ? 4 : 8; const int lfl_forward = subsampling_factor ? 4 : 8;
unsigned int mask_16x16_0 = mask_16x16_l & mask_cutoff; unsigned int mask_16x16_0 = mask_16x16_l & mask_cutoff;
unsigned int mask_8x8_0 = mask_8x8_l & mask_cutoff; unsigned int mask_8x8_0 = mask_8x8_l & mask_cutoff;
@ -1326,248 +1326,203 @@ void vp9_filter_block_plane_non420(VP9_COMMON *cm,
} }
} }
void vp9_filter_block_plane(VP9_COMMON *const cm, void vp9_filter_block_plane_ss00(VP9_COMMON *const cm,
struct macroblockd_plane *const plane, struct macroblockd_plane *const plane,
int mi_row, int mi_row,
LOOP_FILTER_MASK *lfm) { LOOP_FILTER_MASK *lfm) {
struct buf_2d *const dst = &plane->dst; struct buf_2d *const dst = &plane->dst;
uint8_t* const dst0 = dst->buf; uint8_t *const dst0 = dst->buf;
int r, c; int r;
uint64_t mask_16x16 = lfm->left_y[TX_16X16];
uint64_t mask_8x8 = lfm->left_y[TX_8X8];
uint64_t mask_4x4 = lfm->left_y[TX_4X4];
uint64_t mask_4x4_int = lfm->int_4x4_y;
if (!plane->plane_type) { assert(plane->subsampling_x == 0 && plane->subsampling_y == 0);
uint64_t mask_16x16 = lfm->left_y[TX_16X16];
uint64_t mask_8x8 = lfm->left_y[TX_8X8];
uint64_t mask_4x4 = lfm->left_y[TX_4X4];
uint64_t mask_4x4_int = lfm->int_4x4_y;
// Vertical pass: do 2 rows at one time // Vertical pass: do 2 rows at one time
for (r = 0; r < MI_BLOCK_SIZE && mi_row + r < cm->mi_rows; r += 2) { for (r = 0; r < MI_BLOCK_SIZE && mi_row + r < cm->mi_rows; r += 2) {
unsigned int mask_16x16_l = mask_16x16 & 0xffff; unsigned int mask_16x16_l = mask_16x16 & 0xffff;
unsigned int mask_8x8_l = mask_8x8 & 0xffff; unsigned int mask_8x8_l = mask_8x8 & 0xffff;
unsigned int mask_4x4_l = mask_4x4 & 0xffff; unsigned int mask_4x4_l = mask_4x4 & 0xffff;
unsigned int mask_4x4_int_l = mask_4x4_int & 0xffff; unsigned int mask_4x4_int_l = mask_4x4_int & 0xffff;
// Disable filtering on the leftmost column. // Disable filtering on the leftmost column.
#if CONFIG_VP9_HIGHBITDEPTH #if CONFIG_VP9_HIGHBITDEPTH
if (cm->use_highbitdepth) { if (cm->use_highbitdepth) {
highbd_filter_selectively_vert_row2(plane->plane_type, highbd_filter_selectively_vert_row2(
CONVERT_TO_SHORTPTR(dst->buf), plane->subsampling_x, CONVERT_TO_SHORTPTR(dst->buf), dst->stride,
dst->stride, mask_16x16_l, mask_8x8_l, mask_4x4_l, mask_4x4_int_l, &cm->lf_info,
mask_16x16_l, &lfm->lfl_y[r << 3], (int)cm->bit_depth);
mask_8x8_l, } else {
mask_4x4_l, filter_selectively_vert_row2(
mask_4x4_int_l, plane->subsampling_x, dst->buf, dst->stride, mask_16x16_l, mask_8x8_l,
&cm->lf_info, &lfm->lfl_y[r << 3], mask_4x4_l, mask_4x4_int_l, &cm->lf_info, &lfm->lfl_y[r << 3]);
(int)cm->bit_depth); }
} else {
filter_selectively_vert_row2(plane->plane_type,
dst->buf, dst->stride,
mask_16x16_l,
mask_8x8_l,
mask_4x4_l,
mask_4x4_int_l,
&cm->lf_info,
&lfm->lfl_y[r << 3]);
}
#else #else
filter_selectively_vert_row2(plane->plane_type, filter_selectively_vert_row2(
dst->buf, dst->stride, plane->subsampling_x, dst->buf, dst->stride, mask_16x16_l, mask_8x8_l,
mask_16x16_l, mask_4x4_l, mask_4x4_int_l, &cm->lf_info, &lfm->lfl_y[r << 3]);
mask_8x8_l,
mask_4x4_l,
mask_4x4_int_l,
&cm->lf_info, &lfm->lfl_y[r << 3]);
#endif // CONFIG_VP9_HIGHBITDEPTH #endif // CONFIG_VP9_HIGHBITDEPTH
dst->buf += 16 * dst->stride; dst->buf += 16 * dst->stride;
mask_16x16 >>= 16; mask_16x16 >>= 16;
mask_8x8 >>= 16; mask_8x8 >>= 16;
mask_4x4 >>= 16; mask_4x4 >>= 16;
mask_4x4_int >>= 16; mask_4x4_int >>= 16;
}
// Horizontal pass
dst->buf = dst0;
mask_16x16 = lfm->above_y[TX_16X16];
mask_8x8 = lfm->above_y[TX_8X8];
mask_4x4 = lfm->above_y[TX_4X4];
mask_4x4_int = lfm->int_4x4_y;
for (r = 0; r < MI_BLOCK_SIZE && mi_row + r < cm->mi_rows; r++) {
unsigned int mask_16x16_r;
unsigned int mask_8x8_r;
unsigned int mask_4x4_r;
if (mi_row + r == 0) {
mask_16x16_r = 0;
mask_8x8_r = 0;
mask_4x4_r = 0;
} else {
mask_16x16_r = mask_16x16 & 0xff;
mask_8x8_r = mask_8x8 & 0xff;
mask_4x4_r = mask_4x4 & 0xff;
} }
// Horizontal pass
dst->buf = dst0;
mask_16x16 = lfm->above_y[TX_16X16];
mask_8x8 = lfm->above_y[TX_8X8];
mask_4x4 = lfm->above_y[TX_4X4];
mask_4x4_int = lfm->int_4x4_y;
for (r = 0; r < MI_BLOCK_SIZE && mi_row + r < cm->mi_rows; r++) {
unsigned int mask_16x16_r;
unsigned int mask_8x8_r;
unsigned int mask_4x4_r;
if (mi_row + r == 0) {
mask_16x16_r = 0;
mask_8x8_r = 0;
mask_4x4_r = 0;
} else {
mask_16x16_r = mask_16x16 & 0xff;
mask_8x8_r = mask_8x8 & 0xff;
mask_4x4_r = mask_4x4 & 0xff;
}
#if CONFIG_VP9_HIGHBITDEPTH #if CONFIG_VP9_HIGHBITDEPTH
if (cm->use_highbitdepth) { if (cm->use_highbitdepth) {
highbd_filter_selectively_horiz(CONVERT_TO_SHORTPTR(dst->buf), highbd_filter_selectively_horiz(
dst->stride, CONVERT_TO_SHORTPTR(dst->buf), dst->stride, mask_16x16_r, mask_8x8_r,
mask_16x16_r, mask_4x4_r, mask_4x4_int & 0xff, &cm->lf_info, &lfm->lfl_y[r << 3],
mask_8x8_r, (int)cm->bit_depth);
mask_4x4_r, } else {
mask_4x4_int & 0xff, filter_selectively_horiz(dst->buf, dst->stride, mask_16x16_r, mask_8x8_r,
&cm->lf_info, mask_4x4_r, mask_4x4_int & 0xff, &cm->lf_info,
&lfm->lfl_y[r << 3],
(int)cm->bit_depth);
} else {
filter_selectively_horiz(dst->buf, dst->stride,
mask_16x16_r,
mask_8x8_r,
mask_4x4_r,
mask_4x4_int & 0xff,
&cm->lf_info,
&lfm->lfl_y[r << 3]);
}
#else
filter_selectively_horiz(dst->buf, dst->stride,
mask_16x16_r,
mask_8x8_r,
mask_4x4_r,
mask_4x4_int & 0xff,
&cm->lf_info,
&lfm->lfl_y[r << 3]); &lfm->lfl_y[r << 3]);
}
#else
filter_selectively_horiz(dst->buf, dst->stride, mask_16x16_r, mask_8x8_r,
mask_4x4_r, mask_4x4_int & 0xff, &cm->lf_info,
&lfm->lfl_y[r << 3]);
#endif // CONFIG_VP9_HIGHBITDEPTH #endif // CONFIG_VP9_HIGHBITDEPTH
dst->buf += 8 * dst->stride; dst->buf += 8 * dst->stride;
mask_16x16 >>= 8;
mask_8x8 >>= 8;
mask_4x4 >>= 8;
mask_4x4_int >>= 8;
}
}
void vp9_filter_block_plane_ss11(VP9_COMMON *const cm,
struct macroblockd_plane *const plane,
int mi_row,
LOOP_FILTER_MASK *lfm) {
struct buf_2d *const dst = &plane->dst;
uint8_t *const dst0 = dst->buf;
int r, c;
uint16_t mask_16x16 = lfm->left_uv[TX_16X16];
uint16_t mask_8x8 = lfm->left_uv[TX_8X8];
uint16_t mask_4x4 = lfm->left_uv[TX_4X4];
uint16_t mask_4x4_int = lfm->int_4x4_uv;
assert(plane->subsampling_x == 1 && plane->subsampling_y == 1);
// Vertical pass: do 2 rows at one time
for (r = 0; r < MI_BLOCK_SIZE && mi_row + r < cm->mi_rows; r += 4) {
if (plane->plane_type == 1) {
for (c = 0; c < (MI_BLOCK_SIZE >> 1); c++) {
lfm->lfl_uv[(r << 1) + c] = lfm->lfl_y[(r << 3) + (c << 1)];
lfm->lfl_uv[((r + 2) << 1) + c] = lfm->lfl_y[((r + 2) << 3) + (c << 1)];
}
}
{
unsigned int mask_16x16_l = mask_16x16 & 0xff;
unsigned int mask_8x8_l = mask_8x8 & 0xff;
unsigned int mask_4x4_l = mask_4x4 & 0xff;
unsigned int mask_4x4_int_l = mask_4x4_int & 0xff;
// Disable filtering on the leftmost column.
#if CONFIG_VP9_HIGHBITDEPTH
if (cm->use_highbitdepth) {
highbd_filter_selectively_vert_row2(
plane->subsampling_x, CONVERT_TO_SHORTPTR(dst->buf), dst->stride,
mask_16x16_l, mask_8x8_l, mask_4x4_l, mask_4x4_int_l, &cm->lf_info,
&lfm->lfl_uv[r << 1], (int)cm->bit_depth);
} else {
filter_selectively_vert_row2(
plane->subsampling_x, dst->buf, dst->stride,
mask_16x16_l, mask_8x8_l, mask_4x4_l, mask_4x4_int_l, &cm->lf_info,
&lfm->lfl_uv[r << 1]);
}
#else
filter_selectively_vert_row2(
plane->subsampling_x, dst->buf, dst->stride,
mask_16x16_l, mask_8x8_l, mask_4x4_l, mask_4x4_int_l, &cm->lf_info,
&lfm->lfl_uv[r << 1]);
#endif // CONFIG_VP9_HIGHBITDEPTH
dst->buf += 16 * dst->stride;
mask_16x16 >>= 8; mask_16x16 >>= 8;
mask_8x8 >>= 8; mask_8x8 >>= 8;
mask_4x4 >>= 8; mask_4x4 >>= 8;
mask_4x4_int >>= 8; mask_4x4_int >>= 8;
} }
} else { }
uint16_t mask_16x16 = lfm->left_uv[TX_16X16];
uint16_t mask_8x8 = lfm->left_uv[TX_8X8];
uint16_t mask_4x4 = lfm->left_uv[TX_4X4];
uint16_t mask_4x4_int = lfm->int_4x4_uv;
// Vertical pass: do 2 rows at one time // Horizontal pass
for (r = 0; r < MI_BLOCK_SIZE && mi_row + r < cm->mi_rows; r += 4) { dst->buf = dst0;
if (plane->plane_type == 1) { mask_16x16 = lfm->above_uv[TX_16X16];
for (c = 0; c < (MI_BLOCK_SIZE >> 1); c++) { mask_8x8 = lfm->above_uv[TX_8X8];
lfm->lfl_uv[(r << 1) + c] = lfm->lfl_y[(r << 3) + (c << 1)]; mask_4x4 = lfm->above_uv[TX_4X4];
lfm->lfl_uv[((r + 2) << 1) + c] = lfm->lfl_y[((r + 2) << 3) + mask_4x4_int = lfm->int_4x4_uv;
(c << 1)];
}
}
{ for (r = 0; r < MI_BLOCK_SIZE && mi_row + r < cm->mi_rows; r += 2) {
unsigned int mask_16x16_l = mask_16x16 & 0xff; const int skip_border_4x4_r = mi_row + r == cm->mi_rows - 1;
unsigned int mask_8x8_l = mask_8x8 & 0xff; const unsigned int mask_4x4_int_r =
unsigned int mask_4x4_l = mask_4x4 & 0xff; skip_border_4x4_r ? 0 : (mask_4x4_int & 0xf);
unsigned int mask_4x4_int_l = mask_4x4_int & 0xff; unsigned int mask_16x16_r;
unsigned int mask_8x8_r;
unsigned int mask_4x4_r;
// Disable filtering on the leftmost column. if (mi_row + r == 0) {
#if CONFIG_VP9_HIGHBITDEPTH mask_16x16_r = 0;
if (cm->use_highbitdepth) { mask_8x8_r = 0;
highbd_filter_selectively_vert_row2(plane->plane_type, mask_4x4_r = 0;
CONVERT_TO_SHORTPTR(dst->buf), } else {
dst->stride, mask_16x16_r = mask_16x16 & 0xf;
mask_16x16_l, mask_8x8_r = mask_8x8 & 0xf;
mask_8x8_l, mask_4x4_r = mask_4x4 & 0xf;
mask_4x4_l,
mask_4x4_int_l,
&cm->lf_info,
&lfm->lfl_uv[r << 1],
(int)cm->bit_depth);
} else {
filter_selectively_vert_row2(plane->plane_type,
dst->buf, dst->stride,
mask_16x16_l,
mask_8x8_l,
mask_4x4_l,
mask_4x4_int_l,
&cm->lf_info,
&lfm->lfl_uv[r << 1]);
}
#else
filter_selectively_vert_row2(plane->plane_type,
dst->buf, dst->stride,
mask_16x16_l,
mask_8x8_l,
mask_4x4_l,
mask_4x4_int_l,
&cm->lf_info,
&lfm->lfl_uv[r << 1]);
#endif // CONFIG_VP9_HIGHBITDEPTH
dst->buf += 16 * dst->stride;
mask_16x16 >>= 8;
mask_8x8 >>= 8;
mask_4x4 >>= 8;
mask_4x4_int >>= 8;
}
} }
// Horizontal pass
dst->buf = dst0;
mask_16x16 = lfm->above_uv[TX_16X16];
mask_8x8 = lfm->above_uv[TX_8X8];
mask_4x4 = lfm->above_uv[TX_4X4];
mask_4x4_int = lfm->int_4x4_uv;
for (r = 0; r < MI_BLOCK_SIZE && mi_row + r < cm->mi_rows; r += 2) {
const int skip_border_4x4_r = mi_row + r == cm->mi_rows - 1;
const unsigned int mask_4x4_int_r = skip_border_4x4_r ?
0 : (mask_4x4_int & 0xf);
unsigned int mask_16x16_r;
unsigned int mask_8x8_r;
unsigned int mask_4x4_r;
if (mi_row + r == 0) {
mask_16x16_r = 0;
mask_8x8_r = 0;
mask_4x4_r = 0;
} else {
mask_16x16_r = mask_16x16 & 0xf;
mask_8x8_r = mask_8x8 & 0xf;
mask_4x4_r = mask_4x4 & 0xf;
}
#if CONFIG_VP9_HIGHBITDEPTH #if CONFIG_VP9_HIGHBITDEPTH
if (cm->use_highbitdepth) { if (cm->use_highbitdepth) {
highbd_filter_selectively_horiz(CONVERT_TO_SHORTPTR(dst->buf), highbd_filter_selectively_horiz(CONVERT_TO_SHORTPTR(dst->buf),
dst->stride, dst->stride, mask_16x16_r, mask_8x8_r,
mask_16x16_r, mask_4x4_r, mask_4x4_int_r, &cm->lf_info,
mask_8x8_r, &lfm->lfl_uv[r << 1], (int)cm->bit_depth);
mask_4x4_r, } else {
mask_4x4_int_r, filter_selectively_horiz(dst->buf, dst->stride, mask_16x16_r, mask_8x8_r,
&cm->lf_info, mask_4x4_r, mask_4x4_int_r, &cm->lf_info,
&lfm->lfl_uv[r << 1],
(int)cm->bit_depth);
} else {
filter_selectively_horiz(dst->buf, dst->stride,
mask_16x16_r,
mask_8x8_r,
mask_4x4_r,
mask_4x4_int_r,
&cm->lf_info,
&lfm->lfl_uv[r << 1]);
}
#else
filter_selectively_horiz(dst->buf, dst->stride,
mask_16x16_r,
mask_8x8_r,
mask_4x4_r,
mask_4x4_int_r,
&cm->lf_info,
&lfm->lfl_uv[r << 1]); &lfm->lfl_uv[r << 1]);
}
#else
filter_selectively_horiz(dst->buf, dst->stride, mask_16x16_r, mask_8x8_r,
mask_4x4_r, mask_4x4_int_r, &cm->lf_info,
&lfm->lfl_uv[r << 1]);
#endif // CONFIG_VP9_HIGHBITDEPTH #endif // CONFIG_VP9_HIGHBITDEPTH
dst->buf += 8 * dst->stride; dst->buf += 8 * dst->stride;
mask_16x16 >>= 4; mask_16x16 >>= 4;
mask_8x8 >>= 4; mask_8x8 >>= 4;
mask_4x4 >>= 4; mask_4x4 >>= 4;
mask_4x4_int >>= 4; mask_4x4_int >>= 4;
}
} }
} }
@ -1576,11 +1531,19 @@ void vp9_loop_filter_rows(YV12_BUFFER_CONFIG *frame_buffer,
struct macroblockd_plane planes[MAX_MB_PLANE], struct macroblockd_plane planes[MAX_MB_PLANE],
int start, int stop, int y_only) { int start, int stop, int y_only) {
const int num_planes = y_only ? 1 : MAX_MB_PLANE; const int num_planes = y_only ? 1 : MAX_MB_PLANE;
const int use_420 = y_only || (planes[1].subsampling_y == 1 && enum lf_path path;
planes[1].subsampling_x == 1);
LOOP_FILTER_MASK lfm; LOOP_FILTER_MASK lfm;
int mi_row, mi_col; int mi_row, mi_col;
if (y_only)
path = LF_PATH_444;
else if (planes[1].subsampling_y == 1 && planes[1].subsampling_x == 1)
path = LF_PATH_420;
else if (planes[1].subsampling_y == 0 && planes[1].subsampling_x == 0)
path = LF_PATH_444;
else
path = LF_PATH_SLOW;
for (mi_row = start; mi_row < stop; mi_row += MI_BLOCK_SIZE) { for (mi_row = start; mi_row < stop; mi_row += MI_BLOCK_SIZE) {
MODE_INFO *mi = cm->mi + mi_row * cm->mi_stride; MODE_INFO *mi = cm->mi + mi_row * cm->mi_stride;
@ -1590,16 +1553,23 @@ void vp9_loop_filter_rows(YV12_BUFFER_CONFIG *frame_buffer,
vp9_setup_dst_planes(planes, frame_buffer, mi_row, mi_col); vp9_setup_dst_planes(planes, frame_buffer, mi_row, mi_col);
// TODO(JBB): Make setup_mask work for non 420. // TODO(JBB): Make setup_mask work for non 420.
if (use_420) vp9_setup_mask(cm, mi_row, mi_col, mi + mi_col, cm->mi_stride,
vp9_setup_mask(cm, mi_row, mi_col, mi + mi_col, cm->mi_stride, &lfm);
&lfm);
for (plane = 0; plane < num_planes; ++plane) { vp9_filter_block_plane_ss00(cm, &planes[0], mi_row, &lfm);
if (use_420) for (plane = 1; plane < num_planes; ++plane) {
vp9_filter_block_plane(cm, &planes[plane], mi_row, &lfm); switch (path) {
else case LF_PATH_420:
vp9_filter_block_plane_non420(cm, &planes[plane], mi + mi_col, vp9_filter_block_plane_ss11(cm, &planes[plane], mi_row, &lfm);
mi_row, mi_col); break;
case LF_PATH_444:
vp9_filter_block_plane_ss00(cm, &planes[plane], mi_row, &lfm);
break;
case LF_PATH_SLOW:
vp9_filter_block_plane_non420(cm, &planes[plane], mi + mi_col,
mi_row, mi_col);
break;
}
} }
} }
} }

19
vp9/common/vp9_loopfilter.h
View File

@ -29,6 +29,12 @@ extern "C" {
#define MAX_REF_LF_DELTAS 4 #define MAX_REF_LF_DELTAS 4
#define MAX_MODE_LF_DELTAS 2 #define MAX_MODE_LF_DELTAS 2
enum lf_path {
LF_PATH_420,
LF_PATH_444,
LF_PATH_SLOW,
};
struct loopfilter { struct loopfilter {
int filter_level; int filter_level;
@ -92,10 +98,15 @@ void vp9_setup_mask(struct VP9Common *const cm,
MODE_INFO *mi_8x8, const int mode_info_stride, MODE_INFO *mi_8x8, const int mode_info_stride,
LOOP_FILTER_MASK *lfm); LOOP_FILTER_MASK *lfm);
void vp9_filter_block_plane(struct VP9Common *const cm, void vp9_filter_block_plane_ss00(struct VP9Common *const cm,
struct macroblockd_plane *const plane, struct macroblockd_plane *const plane,
int mi_row, int mi_row,
LOOP_FILTER_MASK *lfm); LOOP_FILTER_MASK *lfm);
void vp9_filter_block_plane_ss11(struct VP9Common *const cm,
struct macroblockd_plane *const plane,
int mi_row,
LOOP_FILTER_MASK *lfm);
void vp9_filter_block_plane_non420(struct VP9Common *cm, void vp9_filter_block_plane_non420(struct VP9Common *cm,
struct macroblockd_plane *plane, struct macroblockd_plane *plane,

37
vp9/common/vp9_thread_common.c
View File

@ -13,6 +13,7 @@
#include "vp9/common/vp9_entropymode.h" #include "vp9/common/vp9_entropymode.h"
#include "vp9/common/vp9_thread_common.h" #include "vp9/common/vp9_thread_common.h"
#include "vp9/common/vp9_reconinter.h" #include "vp9/common/vp9_reconinter.h"
#include "vp9/common/vp9_loopfilter.h"
#if CONFIG_MULTITHREAD #if CONFIG_MULTITHREAD
static INLINE void mutex_lock(pthread_mutex_t *const mutex) { static INLINE void mutex_lock(pthread_mutex_t *const mutex) {
@ -92,10 +93,17 @@ void thread_loop_filter_rows(const YV12_BUFFER_CONFIG *const frame_buffer,
int start, int stop, int y_only, int start, int stop, int y_only,
VP9LfSync *const lf_sync) { VP9LfSync *const lf_sync) {
const int num_planes = y_only ? 1 : MAX_MB_PLANE; const int num_planes = y_only ? 1 : MAX_MB_PLANE;
const int use_420 = y_only || (planes[1].subsampling_y == 1 &&
planes[1].subsampling_x == 1);
const int sb_cols = mi_cols_aligned_to_sb(cm->mi_cols) >> MI_BLOCK_SIZE_LOG2; const int sb_cols = mi_cols_aligned_to_sb(cm->mi_cols) >> MI_BLOCK_SIZE_LOG2;
int mi_row, mi_col; int mi_row, mi_col;
enum lf_path path;
if (y_only)
path = LF_PATH_444;
else if (planes[1].subsampling_y == 1 && planes[1].subsampling_x == 1)
path = LF_PATH_420;
else if (planes[1].subsampling_y == 0 && planes[1].subsampling_x == 0)
path = LF_PATH_444;
else
path = LF_PATH_SLOW;
for (mi_row = start; mi_row < stop; for (mi_row = start; mi_row < stop;
mi_row += lf_sync->num_workers * MI_BLOCK_SIZE) { mi_row += lf_sync->num_workers * MI_BLOCK_SIZE) {
@ -112,16 +120,23 @@ void thread_loop_filter_rows(const YV12_BUFFER_CONFIG *const frame_buffer,
vp9_setup_dst_planes(planes, frame_buffer, mi_row, mi_col); vp9_setup_dst_planes(planes, frame_buffer, mi_row, mi_col);
// TODO(JBB): Make setup_mask work for non 420. // TODO(JBB): Make setup_mask work for non 420.
if (use_420) vp9_setup_mask(cm, mi_row, mi_col, mi + mi_col, cm->mi_stride,
vp9_setup_mask(cm, mi_row, mi_col, mi + mi_col, cm->mi_stride, &lfm);
&lfm);
for (plane = 0; plane < num_planes; ++plane) { vp9_filter_block_plane_ss00(cm, &planes[0], mi_row, &lfm);
if (use_420) for (plane = 1; plane < num_planes; ++plane) {
vp9_filter_block_plane(cm, &planes[plane], mi_row, &lfm); switch (path) {
else case LF_PATH_420:
vp9_filter_block_plane_non420(cm, &planes[plane], mi + mi_col, vp9_filter_block_plane_ss11(cm, &planes[plane], mi_row, &lfm);
mi_row, mi_col); break;
case LF_PATH_444:
vp9_filter_block_plane_ss00(cm, &planes[plane], mi_row, &lfm);
break;
case LF_PATH_SLOW:
vp9_filter_block_plane_non420(cm, &planes[plane], mi + mi_col,
mi_row, mi_col);
break;
}
} }
sync_write(lf_sync, r, c, sb_cols); sync_write(lf_sync, r, c, sb_cols);