Skip to content

GitLab

Commit dba0e5a2 authored by Fiona Glaser's avatar Fiona Glaser
Browse files
Options

Cosmetics and loop unrolling 

GCC is not very good at loop unrolling in cases where it can perform constant propagation, so the unrolling unfortunately has to be done manually.
parent d108f919
Hide whitespace changes
Inline Side-by-side
Showing with 128 additions and 132 deletions
common/cabac.h
View file @ dba0e5a2
......@@ -38,7 +38,7 @@ typedef struct
uint8_t *p;
uint8_t *p_end;
/* aligned for aligned_memcpy starting here */
/* aligned for memcpy_aligned starting here */
DECLARE_ALIGNED_16( int f8_bits_encoded ); // only if using x264_cabac_size_decision()
/* context */
......
common/frame.c
View file @ dba0e5a2
......@@ -594,7 +594,7 @@ void x264_frame_deblock_row( x264_t *h, int mb_y )
const int mb_4x4 = 4 * s4x4 * mb_y + 4 * mb_x;
const int b_8x8_transform = h->mb.mb_transform_size[mb_xy];
const int i_edge_end = (h->mb.type[mb_xy] == P_SKIP) ? 1 : 4;
int i_edge, i_dir;
int i_edge;
int i_pix_y[3] = { 16*mb_y*h->fdec->i_stride[0] + 16*mb_x,
8*mb_y*h->fdec->i_stride[1] + 8*mb_x,
......@@ -610,125 +610,116 @@ void x264_frame_deblock_row( x264_t *h, int mb_y )
/* i_dir == 0 -> vertical edge
* i_dir == 1 -> horizontal edge */
for( i_dir = 0; i_dir < 2; i_dir++ )
{
int i_start = (i_dir ? (mb_y <= b_interlaced) : (mb_x == 0));
int i_qp, i_qpn;
for( i_edge = i_start; i_edge < i_edge_end; i_edge++ )
{
int mbn_xy, mbn_8x8, mbn_4x4;
int bS[4]; /* filtering strength */
if( b_8x8_transform && (i_edge&1) )
continue;
mbn_xy = i_edge > 0 ? mb_xy : ( i_dir == 0 ? mb_xy - 1 : mb_xy - h->mb.i_mb_stride );
mbn_8x8 = i_edge > 0 ? mb_8x8 : ( i_dir == 0 ? mb_8x8 - 2 : mb_8x8 - 2 * s8x8 );
mbn_4x4 = i_edge > 0 ? mb_4x4 : ( i_dir == 0 ? mb_4x4 - 4 : mb_4x4 - 4 * s4x4 );
if( b_interlaced && i_edge == 0 && i_dir == 1 )
{
mbn_xy -= h->mb.i_mb_stride;
mbn_8x8 -= 2 * s8x8;
mbn_4x4 -= 4 * s4x4;
}
/* *** Get bS for each 4px for the current edge *** */
if( IS_INTRA( h->mb.type[mb_xy] ) || IS_INTRA( h->mb.type[mbn_xy] ) )
{
bS[0] = bS[1] = bS[2] = bS[3] = ( i_edge == 0 && !(b_interlaced && i_dir) ? 4 : 3 );
}
else
{
int i;
for( i = 0; i < 4; i++ )
{
int x = i_dir == 0 ? i_edge : i;
int y = i_dir == 0 ? i : i_edge;
int xn = (x - (i_dir == 0 ? 1 : 0 ))&0x03;
int yn = (y - (i_dir == 0 ? 0 : 1 ))&0x03;
if( h->mb.non_zero_count[mb_xy][block_idx_xy[x][y]] != 0 ||
h->mb.non_zero_count[mbn_xy][block_idx_xy[xn][yn]] != 0 )
{
bS[i] = 2;
}
else
{
/* FIXME: A given frame may occupy more than one position in
* the reference list. So we should compare the frame numbers,
* not the indices in the ref list.
* No harm yet, as we don't generate that case.*/
int i8p= mb_8x8+(x/2)+(y/2)*s8x8;
int i8q= mbn_8x8+(xn/2)+(yn/2)*s8x8;
int i4p= mb_4x4+x+y*s4x4;
int i4q= mbn_4x4+xn+yn*s4x4;
int l;
bS[i] = 0;
for( l = 0; l < 1 + (h->sh.i_type == SLICE_TYPE_B); l++ )
{
if( h->mb.ref[l][i8p] != h->mb.ref[l][i8q] ||
abs( h->mb.mv[l][i4p][0] - h->mb.mv[l][i4q][0] ) >= 4 ||
abs( h->mb.mv[l][i4p][1] - h->mb.mv[l][i4q][1] ) >= mvy_limit )
{
bS[i] = 1;
break;
}
}
}
}
}
/* *** filter *** */
/* Y plane */
i_qp = h->mb.qp[mb_xy];
i_qpn= h->mb.qp[mbn_xy];
if( i_dir == 0 )
{
/* vertical edge */
deblock_edge( h, &h->fdec->plane[0][i_pix_y[0] + 4*i_edge],
i_stride2[0], bS, (i_qp+i_qpn+1) >> 1, 0,
h->loopf.deblock_h_luma, h->loopf.deblock_h_luma_intra );
if( !(i_edge & 1) )
{
/* U/V planes */
int i_qpc = ( i_chroma_qp_table[x264_clip3( i_qp + h->pps->i_chroma_qp_index_offset, 0, 51 )] +
i_chroma_qp_table[x264_clip3( i_qpn + h->pps->i_chroma_qp_index_offset, 0, 51 )] + 1 ) >> 1;
deblock_edge( h, &h->fdec->plane[1][i_pix_y[1] + 2*i_edge],
i_stride2[1], bS, i_qpc, 1,
h->loopf.deblock_h_chroma, h->loopf.deblock_h_chroma_intra );
deblock_edge( h, &h->fdec->plane[2][i_pix_y[2] + 2*i_edge],
i_stride2[2], bS, i_qpc, 1,
h->loopf.deblock_h_chroma, h->loopf.deblock_h_chroma_intra );
}
}
else
{
/* horizontal edge */
deblock_edge( h, &h->fdec->plane[0][i_pix_y[0] + 4*i_edge*i_stride2[0]],
i_stride2[0], bS, (i_qp+i_qpn+1) >> 1, 0,
h->loopf.deblock_v_luma, h->loopf.deblock_v_luma_intra );
/* U/V planes */
if( !(i_edge & 1) )
{
int i_qpc = ( i_chroma_qp_table[x264_clip3( i_qp + h->pps->i_chroma_qp_index_offset, 0, 51 )] +
i_chroma_qp_table[x264_clip3( i_qpn + h->pps->i_chroma_qp_index_offset, 0, 51 )] + 1 ) >> 1;
deblock_edge( h, &h->fdec->plane[1][i_pix_y[1] + 2*i_edge*i_stride2[1]],
i_stride2[1], bS, i_qpc, 1,
h->loopf.deblock_v_chroma, h->loopf.deblock_v_chroma_intra );
deblock_edge( h, &h->fdec->plane[2][i_pix_y[2] + 2*i_edge*i_stride2[2]],
i_stride2[2], bS, i_qpc, 1,
h->loopf.deblock_v_chroma, h->loopf.deblock_v_chroma_intra );
}
}
}
#define deblock_dir(i_dir)\
{\
int i_start = (i_dir ? (mb_y <= b_interlaced) : (mb_x == 0));\
int i_qp, i_qpn;\
for( i_edge = i_start; i_edge < i_edge_end; i_edge++ )\
{\
int mbn_xy, mbn_8x8, mbn_4x4;\
int bS[4]; /* filtering strength */\
if( b_8x8_transform && (i_edge&1) )\
continue;\
mbn_xy = i_edge > 0 ? mb_xy : ( i_dir == 0 ? mb_xy - 1 : mb_xy - h->mb.i_mb_stride );\
mbn_8x8 = i_edge > 0 ? mb_8x8 : ( i_dir == 0 ? mb_8x8 - 2 : mb_8x8 - 2 * s8x8 );\
mbn_4x4 = i_edge > 0 ? mb_4x4 : ( i_dir == 0 ? mb_4x4 - 4 : mb_4x4 - 4 * s4x4 );\
if( b_interlaced && i_edge == 0 && i_dir == 1 )\
{\
mbn_xy -= h->mb.i_mb_stride;\
mbn_8x8 -= 2 * s8x8;\
mbn_4x4 -= 4 * s4x4;\
}\
/* *** Get bS for each 4px for the current edge *** */\
if( IS_INTRA( h->mb.type[mb_xy] ) || IS_INTRA( h->mb.type[mbn_xy] ) )\
bS[0] = bS[1] = bS[2] = bS[3] = ( i_edge == 0 && !(b_interlaced && i_dir) ? 4 : 3 );\
else\
{\
int i;\
for( i = 0; i < 4; i++ )\
{\
int x = i_dir == 0 ? i_edge : i;\
int y = i_dir == 0 ? i : i_edge;\
int xn = (x - (i_dir == 0 ? 1 : 0 ))&0x03;\
int yn = (y - (i_dir == 0 ? 0 : 1 ))&0x03;\
if( h->mb.non_zero_count[mb_xy][block_idx_xy[x][y]] != 0 ||\
h->mb.non_zero_count[mbn_xy][block_idx_xy[xn][yn]] != 0 )\
{\
bS[i] = 2;\
}\
else\
{\
/* FIXME: A given frame may occupy more than one position in\
* the reference list. So we should compare the frame numbers,\
* not the indices in the ref list.\
* No harm yet, as we don't generate that case.*/\
int i8p= mb_8x8+(x/2)+(y/2)*s8x8;\
int i8q= mbn_8x8+(xn/2)+(yn/2)*s8x8;\
int i4p= mb_4x4+x+y*s4x4;\
int i4q= mbn_4x4+xn+yn*s4x4;\
int l;\
bS[i] = 0;\
for( l = 0; l < 1 + (h->sh.i_type == SLICE_TYPE_B); l++ )\
{\
if( h->mb.ref[l][i8p] != h->mb.ref[l][i8q] ||\
abs( h->mb.mv[l][i4p][0] - h->mb.mv[l][i4q][0] ) >= 4 ||\
abs( h->mb.mv[l][i4p][1] - h->mb.mv[l][i4q][1] ) >= mvy_limit )\
{\
bS[i] = 1;\
break;\
}\
}\
}\
}\
}\
/* *** filter *** */\
/* Y plane */\
i_qp = h->mb.qp[mb_xy];\
i_qpn= h->mb.qp[mbn_xy];\
if( i_dir == 0 )\
{\
/* vertical edge */\
deblock_edge( h, &h->fdec->plane[0][i_pix_y[0] + 4*i_edge],\
i_stride2[0], bS, (i_qp+i_qpn+1) >> 1, 0,\
h->loopf.deblock_h_luma, h->loopf.deblock_h_luma_intra );\
if( !(i_edge & 1) )\
{\
/* U/V planes */\
int i_qpc = ( i_chroma_qp_table[x264_clip3( i_qp + h->pps->i_chroma_qp_index_offset, 0, 51 )] +\
i_chroma_qp_table[x264_clip3( i_qpn + h->pps->i_chroma_qp_index_offset, 0, 51 )] + 1 ) >> 1;\
deblock_edge( h, &h->fdec->plane[1][i_pix_y[1] + 2*i_edge],\
i_stride2[1], bS, i_qpc, 1,\
h->loopf.deblock_h_chroma, h->loopf.deblock_h_chroma_intra );\
deblock_edge( h, &h->fdec->plane[2][i_pix_y[2] + 2*i_edge],\
i_stride2[2], bS, i_qpc, 1,\
h->loopf.deblock_h_chroma, h->loopf.deblock_h_chroma_intra );\
}\
}\
else\
{\
/* horizontal edge */\
deblock_edge( h, &h->fdec->plane[0][i_pix_y[0] + 4*i_edge*i_stride2[0]],\
i_stride2[0], bS, (i_qp+i_qpn+1) >> 1, 0,\
h->loopf.deblock_v_luma, h->loopf.deblock_v_luma_intra );\
/* U/V planes */\
if( !(i_edge & 1) )\
{\
int i_qpc = ( i_chroma_qp_table[x264_clip3( i_qp + h->pps->i_chroma_qp_index_offset, 0, 51 )] +\
i_chroma_qp_table[x264_clip3( i_qpn + h->pps->i_chroma_qp_index_offset, 0, 51 )] + 1 ) >> 1;\
deblock_edge( h, &h->fdec->plane[1][i_pix_y[1] + 2*i_edge*i_stride2[1]],\
i_stride2[1], bS, i_qpc, 1,\
h->loopf.deblock_v_chroma, h->loopf.deblock_v_chroma_intra );\
deblock_edge( h, &h->fdec->plane[2][i_pix_y[2] + 2*i_edge*i_stride2[2]],\
i_stride2[2], bS, i_qpc, 1,\
h->loopf.deblock_v_chroma, h->loopf.deblock_v_chroma_intra );\
}\
}\
}\
}
deblock_dir(0);
deblock_dir(1);
/* next mb */
if( !b_interlaced || (mb_y&1) )
mb_x++;
......
common/macroblock.c
View file @ dba0e5a2
......@@ -467,16 +467,16 @@ void x264_mb_load_mv_direct8x8( x264_t *h, int idx )
{
const int x = 2*(idx%2);
const int y = 2*(idx/2);
int l;
x264_macroblock_cache_ref( h, x, y, 2, 2, 0, h->mb.cache.direct_ref[0][idx] );
x264_macroblock_cache_ref( h, x, y, 2, 2, 1, h->mb.cache.direct_ref[1][idx] );
for( l = 0; l < 2; l++ )
{
*(uint64_t*)h->mb.cache.mv[l][x264_scan8[idx*4]] =
*(uint64_t*)h->mb.cache.direct_mv[l][x264_scan8[idx*4]];
*(uint64_t*)h->mb.cache.mv[l][x264_scan8[idx*4]+8] =
*(uint64_t*)h->mb.cache.direct_mv[l][x264_scan8[idx*4]+8];
}
*(uint64_t*)h->mb.cache.mv[0][x264_scan8[idx*4]] =
*(uint64_t*)h->mb.cache.direct_mv[0][x264_scan8[idx*4]];
*(uint64_t*)h->mb.cache.mv[0][x264_scan8[idx*4]+8] =
*(uint64_t*)h->mb.cache.direct_mv[0][x264_scan8[idx*4]+8];
*(uint64_t*)h->mb.cache.mv[1][x264_scan8[idx*4]] =
*(uint64_t*)h->mb.cache.direct_mv[1][x264_scan8[idx*4]];
*(uint64_t*)h->mb.cache.mv[1][x264_scan8[idx*4]+8] =
*(uint64_t*)h->mb.cache.direct_mv[1][x264_scan8[idx*4]+8];
}
#define FIXED_SCALE 256
......@@ -979,7 +979,7 @@ static void ALWAYS_INLINE x264_macroblock_load_pic_pointers( x264_t *h, int i_mb
int ref_pix_offset[2] = { i_pix_offset, i_pix_offset };
const uint8_t *intra_fdec = &h->mb.intra_border_backup[i_mb_y & h->sh.b_mbaff][i][i_mb_x*16>>!!i];
x264_frame_t **fref[2] = { h->fref0, h->fref1 };
int j, k, l;
int j, k;
if( h->mb.b_interlaced )
ref_pix_offset[1] += (1-2*(i_mb_y&1)) * i_stride;
h->mb.pic.i_stride[i] = i_stride2;
......@@ -992,16 +992,21 @@ static void ALWAYS_INLINE x264_macroblock_load_pic_pointers( x264_t *h, int i_mb
for( j = 0; j < w; j++ )
h->mb.pic.p_fdec[i][-1+j*FDEC_STRIDE] = plane_fdec[-1+j*i_stride2];
}
for( l=0; l<2; l++ )
for( j = 0; j < h->mb.pic.i_fref[0]; j++ )
{
for( j=0; j<h->mb.pic.i_fref[l]; j++ )
h->mb.pic.p_fref[0][j][i==0 ? 0:i+3] = &fref[0][j >> h->mb.b_interlaced]->plane[i][ref_pix_offset[j&1]];
if( i == 0 )
for( k = 1; k < 4; k++ )
h->mb.pic.p_fref[0][j][k] = &fref[0][j >> h->mb.b_interlaced]->filtered[k][ref_pix_offset[j&1]];
}
if( h->sh.i_type == SLICE_TYPE_B )
for( j = 0; j < h->mb.pic.i_fref[1]; j++ )
{
h->mb.pic.p_fref[l][j][i==0 ? 0:i+3] = &fref[l][j >> h->mb.b_interlaced]->plane[i][ref_pix_offset[j&1]];
h->mb.pic.p_fref[1][j][i==0 ? 0:i+3] = &fref[1][j >> h->mb.b_interlaced]->plane[i][ref_pix_offset[j&1]];
if( i == 0 )
for( k = 1; k < 4; k++ )
h->mb.pic.p_fref[l][j][k] = &fref[l][j >> h->mb.b_interlaced]->filtered[k][ref_pix_offset[j&1]];
h->mb.pic.p_fref[1][j][k] = &fref[1][j >> h->mb.b_interlaced]->filtered[k][ref_pix_offset[j&1]];
}
}
}
void x264_macroblock_cache_load( x264_t *h, int i_mb_x, int i_mb_y )
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment