/*****************************************************************************
* encoder.c: top-level encoder functions
*****************************************************************************
- * Copyright (C) 2003-2010 x264 project
+ * Copyright (C) 2003-2011 x264 project
*
* Authors: Laurent Aimar <fenrir@via.ecp.fr>
* Loren Merritt <lorenm@u.washington.edu>
* For more information, contact us at licensing@x264.com.
*****************************************************************************/
-#include <math.h>
-
#include "common/common.h"
#include "set.h"
if( !h->mb.b_direct_auto_read && sh->i_type == SLICE_TYPE_B )
{
- if( h->fref1[0]->i_poc_l0ref0 == h->fref0[0]->i_poc )
+ if( h->fref[1][0]->i_poc_l0ref0 == h->fref[0][0]->i_poc )
{
if( h->mb.b_direct_auto_write )
sh->b_direct_spatial_mv_pred = ( h->stat.i_direct_score[1] > h->stat.i_direct_score[0] );
sh->i_num_ref_idx_l0_active = 1;
sh->i_num_ref_idx_l1_active = 1;
- sh->b_ref_pic_list_reordering_l0 = h->b_ref_reorder[0];
- sh->b_ref_pic_list_reordering_l1 = h->b_ref_reorder[1];
+ sh->b_ref_pic_list_reordering[0] = h->b_ref_reorder[0];
+ sh->b_ref_pic_list_reordering[1] = h->b_ref_reorder[1];
/* If the ref list isn't in the default order, construct reordering header */
- /* List1 reordering isn't needed yet */
- if( sh->b_ref_pic_list_reordering_l0 )
+ for( int list = 0; list < 2; list++ )
{
- int pred_frame_num = i_frame;
- for( int i = 0; i < h->i_ref0; i++ )
+ if( sh->b_ref_pic_list_reordering[list] )
{
- int diff = h->fref0[i]->i_frame_num - pred_frame_num;
- sh->ref_pic_list_order[0][i].idc = ( diff > 0 );
- sh->ref_pic_list_order[0][i].arg = (abs(diff) - 1) & ((1 << sps->i_log2_max_frame_num) - 1);
- pred_frame_num = h->fref0[i]->i_frame_num;
+ int pred_frame_num = i_frame;
+ for( int i = 0; i < h->i_ref[list]; i++ )
+ {
+ int diff = h->fref[list][i]->i_frame_num - pred_frame_num;
+ sh->ref_pic_list_order[list][i].idc = ( diff > 0 );
+ sh->ref_pic_list_order[list][i].arg = (abs(diff) - 1) & ((1 << sps->i_log2_max_frame_num) - 1);
+ pred_frame_num = h->fref[list][i]->i_frame_num;
+ }
}
}
sh->i_cabac_init_idc = param->i_cabac_init_idc;
- sh->i_qp = i_qp;
- sh->i_qp_delta = i_qp - pps->i_pic_init_qp;
+ sh->i_qp = SPEC_QP(i_qp);
+ sh->i_qp_delta = sh->i_qp - pps->i_pic_init_qp;
sh->b_sp_for_swidth = 0;
sh->i_qs_delta = 0;
/* ref pic list reordering */
if( sh->i_type != SLICE_TYPE_I )
{
- bs_write1( s, sh->b_ref_pic_list_reordering_l0 );
- if( sh->b_ref_pic_list_reordering_l0 )
+ bs_write1( s, sh->b_ref_pic_list_reordering[0] );
+ if( sh->b_ref_pic_list_reordering[0] )
{
for( int i = 0; i < sh->i_num_ref_idx_l0_active; i++ )
{
}
if( sh->i_type == SLICE_TYPE_B )
{
- bs_write1( s, sh->b_ref_pic_list_reordering_l1 );
- if( sh->b_ref_pic_list_reordering_l1 )
+ bs_write1( s, sh->b_ref_pic_list_reordering[1] );
+ if( sh->b_ref_pic_list_reordering[1] )
{
for( int i = 0; i < sh->i_num_ref_idx_l1_active; i++ )
{
return -1;
}
-#if HAVE_PTHREAD
+#if HAVE_THREAD
static void x264_encoder_thread_init( x264_t *h )
{
if( h->param.i_sync_lookahead )
return -1;
}
+ if( (h->param.crop_rect.i_left + h->param.crop_rect.i_right ) >= h->param.i_width ||
+ (h->param.crop_rect.i_top + h->param.crop_rect.i_bottom) >= h->param.i_height )
+ {
+ x264_log( h, X264_LOG_ERROR, "invalid crop-rect %u,%u,%u,%u\n", h->param.crop_rect.i_left,
+ h->param.crop_rect.i_top, h->param.crop_rect.i_right, h->param.crop_rect.i_bottom );
+ return -1;
+ }
+
if( h->param.i_threads == X264_THREADS_AUTO )
h->param.i_threads = x264_cpu_num_processors() * (h->param.b_sliced_threads?2:3)/2;
h->param.i_threads = x264_clip3( h->param.i_threads, 1, X264_THREAD_MAX );
if( h->param.i_threads > 1 )
{
-#if !HAVE_PTHREAD
- x264_log( h, X264_LOG_WARNING, "not compiled with pthread support!\n");
+#if !HAVE_THREAD
+ x264_log( h, X264_LOG_WARNING, "not compiled with thread support!\n");
h->param.i_threads = 1;
#endif
/* Avoid absurdly small thread slices as they can reduce performance
}
}
+ h->param.i_frame_packing = x264_clip3( h->param.i_frame_packing, -1, 5 );
+
/* Detect default ffmpeg settings and terminate with an error. */
{
int score = 0;
h->param.analyse.i_noise_reduction = 0;
h->param.analyse.b_psy = 0;
h->param.i_bframe = 0;
- /* 8x8dct is not useful at all in CAVLC lossless */
- if( !h->param.b_cabac )
+ /* 8x8dct is not useful without RD in CAVLC lossless */
+ if( !h->param.b_cabac && h->param.analyse.i_subpel_refine < 6 )
h->param.analyse.b_transform_8x8 = 0;
}
if( h->param.rc.i_rc_method == X264_RC_CQP )
}
if( h->param.rc.b_stat_read )
h->param.rc.i_lookahead = 0;
-#if HAVE_PTHREAD
+#if HAVE_THREAD
if( h->param.i_sync_lookahead < 0 )
h->param.i_sync_lookahead = h->param.i_bframe + 1;
h->param.i_sync_lookahead = X264_MIN( h->param.i_sync_lookahead, X264_LOOKAHEAD_MAX );
}
h->param.analyse.i_weighted_pred = x264_clip3( h->param.analyse.i_weighted_pred, X264_WEIGHTP_NONE, X264_WEIGHTP_SMART );
- if( !h->param.analyse.i_weighted_pred && h->param.rc.b_mb_tree && h->param.analyse.b_psy && !h->param.b_interlaced )
+ if( !h->param.analyse.i_weighted_pred && h->param.rc.b_mb_tree && h->param.analyse.b_psy )
h->param.analyse.i_weighted_pred = X264_WEIGHTP_FAKE;
- if( h->param.analyse.i_weighted_pred == X264_WEIGHTP_BLIND && BIT_DEPTH > 8 )
- h->param.analyse.i_weighted_pred = X264_WEIGHTP_NONE;
if( h->i_thread_frames > 1 )
{
h->param.vui.i_sar_width = i_w;
h->param.vui.i_sar_height = i_h;
}
+ x264_sps_init( h->sps, h->param.i_sps_id, &h->param );
}
}
}
if( param->param_free )
param->param_free( param );
+ if( x264_threading_init() )
+ {
+ x264_log( h, X264_LOG_ERROR, "unable to initialize threading\n" );
+ goto fail;
+ }
+
if( x264_validate_parameters( h ) < 0 )
goto fail;
if( h->param.rc.psz_stat_in )
h->param.rc.psz_stat_in = strdup( h->param.rc.psz_stat_in );
- x264_set_aspect_ratio( h, &h->param, 1 );
-
x264_reduce_fraction( &h->param.i_fps_num, &h->param.i_fps_den );
x264_reduce_fraction( &h->param.i_timebase_num, &h->param.i_timebase_den );
h->pps = &h->pps_array[0];
x264_pps_init( h->pps, h->param.i_sps_id, &h->param, h->sps );
+ x264_set_aspect_ratio( h, &h->param, 1 );
+
x264_validate_levels( h, 1 );
h->chroma_qp_table = i_chroma_qp_table + 12 + h->pps->i_chroma_qp_index_offset;
i_slicetype_length = h->frames.i_delay;
h->frames.i_delay += h->i_thread_frames - 1;
h->frames.i_delay += h->param.i_sync_lookahead;
- h->frames.i_delay += h->param.b_vfr_input && (h->param.rc.i_rc_method == X264_RC_ABR || h->param.rc.b_stat_write
- || h->param.rc.i_vbv_buffer_size);
+ h->frames.i_delay += h->param.b_vfr_input;
h->frames.i_bframe_delay = h->param.i_bframe ? (h->param.i_bframe_pyramid ? 2 : 1) : 0;
h->frames.i_max_ref0 = h->param.i_frame_reference;
|| h->param.i_bframe_adaptive
|| h->param.i_scenecut_threshold
|| h->param.rc.b_mb_tree
- || h->param.analyse.i_weighted_pred == X264_WEIGHTP_SMART );
+ || h->param.analyse.i_weighted_pred );
h->frames.b_have_lowres |= h->param.rc.b_stat_read && h->param.rc.i_vbv_buffer_size > 0;
h->frames.b_have_sub8x8_esa = !!(h->param.analyse.inter & X264_ANALYSE_PSUB8x8);
+ h->i_thread_frames + 3) * sizeof(x264_frame_t *) );
if( h->param.analyse.i_weighted_pred > 0 )
CHECKED_MALLOCZERO( h->frames.blank_unused, h->i_thread_frames * 4 * sizeof(x264_frame_t *) );
- h->i_ref0 = 0;
- h->i_ref1 = 0;
+ h->i_ref[0] = h->i_ref[1] = 0;
h->i_cpb_delay = h->i_coded_fields = h->i_disp_fields = 0;
h->i_prev_duration = ((uint64_t)h->param.i_fps_den * h->sps->vui.i_time_scale) / ((uint64_t)h->param.i_fps_num * h->sps->vui.i_num_units_in_tick);
h->i_disp_fields_last_frame = -1;
p += sprintf( p, " none!" );
x264_log( h, X264_LOG_INFO, "%s\n", buf );
- for( qp = h->param.rc.i_qp_min; qp <= h->param.rc.i_qp_max; qp++ )
+ for( qp = X264_MIN( h->param.rc.i_qp_min, QP_MAX_SPEC ); qp <= h->param.rc.i_qp_max; qp++ )
if( x264_analyse_init_costs( h, qp ) )
goto fail;
if( x264_analyse_init_costs( h, X264_LOOKAHEAD_QP ) )
static const uint16_t cost_mv_correct[7] = { 24, 47, 95, 189, 379, 757, 1515 };
/* Checks for known miscompilation issues. */
- if( h->cost_mv[x264_lambda_tab[X264_LOOKAHEAD_QP]][2013] != cost_mv_correct[BIT_DEPTH-8] )
+ if( h->cost_mv[X264_LOOKAHEAD_QP][2013] != cost_mv_correct[BIT_DEPTH-8] )
{
x264_log( h, X264_LOG_ERROR, "MV cost test failed: x264 has been miscompiled!\n" );
goto fail;
else
h->thread[i]->fdec = h->thread[0]->fdec;
+ h->thread[i]->sps = &h->thread[i]->sps_array[0];
+ h->thread[i]->pps = &h->thread[i]->pps_array[0];
+
CHECKED_MALLOC( h->thread[i]->out.p_bitstream, h->out.i_bitstream );
/* Start each thread with room for init_nal_count NAL units; it'll realloc later if needed. */
CHECKED_MALLOC( h->thread[i]->out.nal, init_nal_count*sizeof(x264_nal_t) );
COPY( b_deblocking_filter );
COPY( i_deblocking_filter_alphac0 );
COPY( i_deblocking_filter_beta );
+ COPY( i_frame_packing );
COPY( analyse.inter );
COPY( analyse.intra );
COPY( analyse.i_direct_mv_pred );
COPY( analyse.b_mixed_references );
COPY( analyse.f_psy_rd );
COPY( analyse.f_psy_trellis );
+ COPY( crop_rect );
// can only twiddle these if they were enabled to begin with:
if( h->param.analyse.i_me_method >= X264_ME_ESA || param->analyse.i_me_method < X264_ME_ESA )
COPY( analyse.i_me_method );
COPY( i_slice_max_size );
COPY( i_slice_max_mbs );
COPY( i_slice_count );
+ COPY( b_tff );
/* VBV can't be turned on if it wasn't on to begin with */
if( h->param.rc.i_vbv_max_bitrate > 0 && h->param.rc.i_vbv_buffer_size > 0 &&
h->b_ref_reorder[0] = 1;
return;
}
- for( int i = 0; i < h->i_ref0 - 1; i++ )
- /* P and B-frames use different default orders. */
- if( h->sh.i_type == SLICE_TYPE_P ? h->fref0[i]->i_frame_num < h->fref0[i+1]->i_frame_num
- : h->fref0[i]->i_poc < h->fref0[i+1]->i_poc )
+ for( int list = 0; list <= (h->sh.i_type == SLICE_TYPE_B); list++ )
+ for( int i = 0; i < h->i_ref[list] - 1; i++ )
{
- h->b_ref_reorder[0] = 1;
- return;
+ int framenum_diff = h->fref[list][i+1]->i_frame_num - h->fref[list][i]->i_frame_num;
+ int poc_diff = h->fref[list][i+1]->i_poc - h->fref[list][i]->i_poc;
+ /* P and B-frames use different default orders. */
+ if( h->sh.i_type == SLICE_TYPE_P ? framenum_diff > 0 : list == 1 ? poc_diff < 0 : poc_diff > 0 )
+ {
+ h->b_ref_reorder[list] = 1;
+ return;
+ }
}
}
/* return -1 on failure, else return the index of the new reference frame */
int x264_weighted_reference_duplicate( x264_t *h, int i_ref, const x264_weight_t *w )
{
- int i = h->i_ref0;
+ int i = h->i_ref[0];
int j = 1;
x264_frame_t *newframe;
if( i <= 1 ) /* empty list, definitely can't duplicate frame */
return -1;
+ //Duplication is only used in X264_WEIGHTP_SMART
+ if( h->param.analyse.i_weighted_pred != X264_WEIGHTP_SMART )
+ return -1;
+
/* Duplication is a hack to compensate for crappy rounding in motion compensation.
* With high bit depth, it's not worth doing, so turn it off except in the case of
* unweighted dupes. */
return -1;
newframe = x264_frame_pop_blank_unused( h );
+ if( !newframe )
+ return -1;
//FIXME: probably don't need to copy everything
- *newframe = *h->fref0[i_ref];
+ *newframe = *h->fref[0][i_ref];
newframe->i_reference_count = 1;
- newframe->orig = h->fref0[i_ref];
+ newframe->orig = h->fref[0][i_ref];
newframe->b_duplicate = 1;
memcpy( h->fenc->weight[j], w, sizeof(h->fenc->weight[i]) );
/* shift the frames to make space for the dupe. */
h->b_ref_reorder[0] = 1;
- if( h->i_ref0 < X264_REF_MAX )
- ++h->i_ref0;
- h->fref0[X264_REF_MAX-1] = NULL;
- x264_frame_unshift( &h->fref0[j], newframe );
+ if( h->i_ref[0] < X264_REF_MAX )
+ ++h->i_ref[0];
+ h->fref[0][X264_REF_MAX-1] = NULL;
+ x264_frame_unshift( &h->fref[0][j], newframe );
return j;
}
static void x264_weighted_pred_init( x264_t *h )
{
/* for now no analysis and set all weights to nothing */
- for( int i_ref = 0; i_ref < h->i_ref0; i_ref++ )
- h->fenc->weighted[i_ref] = h->fref0[i_ref]->filtered[0];
+ for( int i_ref = 0; i_ref < h->i_ref[0]; i_ref++ )
+ h->fenc->weighted[i_ref] = h->fref[0][i_ref]->filtered[0];
// FIXME: This only supports weighting of one reference frame
// and duplicates of that frame.
h->fenc->i_lines_weighted = 0;
- for( int i_ref = 0; i_ref < (h->i_ref0 << h->sh.b_mbaff); i_ref++ )
+ for( int i_ref = 0; i_ref < (h->i_ref[0] << h->sh.b_mbaff); i_ref++ )
for( int i = 0; i < 3; i++ )
h->sh.weight[i_ref][i].weightfn = NULL;
int i_padv = PADV << h->param.b_interlaced;
int denom = -1;
- int weightluma = 0;
+ int weightplane[2] = { 0, 0 };
int buffer_next = 0;
- //FIXME: when chroma support is added, move this into loop
- h->sh.weight[0][1].weightfn = h->sh.weight[0][2].weightfn = NULL;
- h->sh.weight[0][1].i_denom = h->sh.weight[0][2].i_denom = 0;
- for( int j = 0; j < h->i_ref0; j++ )
+ for( int i = 0; i < 3; i++ )
{
- if( h->fenc->weight[j][0].weightfn )
+ for( int j = 0; j < h->i_ref[0]; j++ )
{
- h->sh.weight[j][0] = h->fenc->weight[j][0];
- // if weight is useless, don't write it to stream
- if( h->sh.weight[j][0].i_scale == 1<<h->sh.weight[j][0].i_denom && h->sh.weight[j][0].i_offset == 0 )
- h->sh.weight[j][0].weightfn = NULL;
- else
+ if( h->fenc->weight[j][i].weightfn )
{
- if( !weightluma )
+ h->sh.weight[j][i] = h->fenc->weight[j][i];
+ // if weight is useless, don't write it to stream
+ if( h->sh.weight[j][i].i_scale == 1<<h->sh.weight[j][i].i_denom && h->sh.weight[j][i].i_offset == 0 )
+ h->sh.weight[j][i].weightfn = NULL;
+ else
{
- weightluma = 1;
- h->sh.weight[0][0].i_denom = denom = h->sh.weight[j][0].i_denom;
- assert( x264_clip3( denom, 0, 7 ) == denom );
+ if( !weightplane[!!i] )
+ {
+ weightplane[!!i] = 1;
+ h->sh.weight[0][!!i].i_denom = denom = h->sh.weight[j][i].i_denom;
+ assert( x264_clip3( denom, 0, 7 ) == denom );
+ }
+
+ assert( h->sh.weight[j][i].i_denom == denom );
+ if( !i )
+ {
+ h->fenc->weighted[j] = h->mb.p_weight_buf[buffer_next++] + h->fenc->i_stride[0] * i_padv + PADH;
+ //scale full resolution frame
+ if( h->param.i_threads == 1 )
+ {
+ pixel *src = h->fref[0][j]->filtered[0] - h->fref[0][j]->i_stride[0]*i_padv - PADH;
+ pixel *dst = h->fenc->weighted[j] - h->fenc->i_stride[0]*i_padv - PADH;
+ int stride = h->fenc->i_stride[0];
+ int width = h->fenc->i_width[0] + PADH*2;
+ int height = h->fenc->i_lines[0] + i_padv*2;
+ x264_weight_scale_plane( h, dst, stride, src, stride, width, height, &h->sh.weight[j][0] );
+ h->fenc->i_lines_weighted = height;
+ }
+ }
}
- assert( h->sh.weight[j][0].i_denom == denom );
- assert( x264_clip3( h->sh.weight[j][0].i_scale, 0, 127 ) == h->sh.weight[j][0].i_scale );
- assert( x264_clip3( h->sh.weight[j][0].i_offset, -128, 127 ) == h->sh.weight[j][0].i_offset );
- h->fenc->weighted[j] = h->mb.p_weight_buf[buffer_next++] +
- h->fenc->i_stride[0] * i_padv + PADH;
}
}
+ }
- //scale full resolution frame
- if( h->sh.weight[j][0].weightfn && h->param.i_threads == 1 )
+ if( weightplane[1] )
+ for( int i = 0; i < h->i_ref[0]; i++ )
{
- pixel *src = h->fref0[j]->filtered[0] - h->fref0[j]->i_stride[0]*i_padv - PADH;
- pixel *dst = h->fenc->weighted[j] - h->fenc->i_stride[0]*i_padv - PADH;
- int stride = h->fenc->i_stride[0];
- int width = h->fenc->i_width[0] + PADH*2;
- int height = h->fenc->i_lines[0] + i_padv*2;
- x264_weight_scale_plane( h, dst, stride, src, stride, width, height, &h->sh.weight[j][0] );
- h->fenc->i_lines_weighted = height;
+ if( h->sh.weight[i][1].weightfn && !h->sh.weight[i][2].weightfn )
+ {
+ h->sh.weight[i][2].i_scale = 1 << h->sh.weight[0][1].i_denom;
+ h->sh.weight[i][2].i_offset = 0;
+ }
+ else if( h->sh.weight[i][2].weightfn && !h->sh.weight[i][1].weightfn )
+ {
+ h->sh.weight[i][1].i_scale = 1 << h->sh.weight[0][1].i_denom;
+ h->sh.weight[i][1].i_offset = 0;
+ }
}
- }
- if( !weightluma )
+
+ if( !weightplane[0] )
h->sh.weight[0][0].i_denom = 0;
+ if( !weightplane[1] )
+ h->sh.weight[0][1].i_denom = 0;
+ h->sh.weight[0][2].i_denom = h->sh.weight[0][1].i_denom;
+}
+
+static inline int x264_reference_distance( x264_t *h, x264_frame_t *frame )
+{
+ if( h->param.i_frame_packing == 5 )
+ return abs((h->fenc->i_frame&~1) - (frame->i_frame&~1)) +
+ ((h->fenc->i_frame&1) != (frame->i_frame&1));
+ else
+ return abs(h->fenc->i_frame - frame->i_frame);
}
static inline void x264_reference_build_list( x264_t *h, int i_poc )
int b_ok;
/* build ref list 0/1 */
- h->mb.pic.i_fref[0] = h->i_ref0 = 0;
- h->mb.pic.i_fref[1] = h->i_ref1 = 0;
+ h->mb.pic.i_fref[0] = h->i_ref[0] = 0;
+ h->mb.pic.i_fref[1] = h->i_ref[1] = 0;
if( h->sh.i_type == SLICE_TYPE_I )
return;
if( h->frames.reference[i]->b_corrupt )
continue;
if( h->frames.reference[i]->i_poc < i_poc )
- h->fref0[h->i_ref0++] = h->frames.reference[i];
+ h->fref[0][h->i_ref[0]++] = h->frames.reference[i];
else if( h->frames.reference[i]->i_poc > i_poc )
- h->fref1[h->i_ref1++] = h->frames.reference[i];
+ h->fref[1][h->i_ref[1]++] = h->frames.reference[i];
}
- /* Order ref0 from higher to lower poc */
- do
+ /* Order reference lists by distance from the current frame. */
+ for( int list = 0; list < 2; list++ )
{
- b_ok = 1;
- for( int i = 0; i < h->i_ref0 - 1; i++ )
+ h->fref_nearest[list] = h->fref[list][0];
+ do
{
- if( h->fref0[i]->i_poc < h->fref0[i+1]->i_poc )
+ b_ok = 1;
+ for( int i = 0; i < h->i_ref[list] - 1; i++ )
{
- XCHG( x264_frame_t*, h->fref0[i], h->fref0[i+1] );
- b_ok = 0;
- break;
+ if( list ? h->fref[list][i+1]->i_poc < h->fref_nearest[list]->i_poc
+ : h->fref[list][i+1]->i_poc > h->fref_nearest[list]->i_poc )
+ h->fref_nearest[list] = h->fref[list][i+1];
+ if( x264_reference_distance( h, h->fref[list][i] ) > x264_reference_distance( h, h->fref[list][i+1] ) )
+ {
+ XCHG( x264_frame_t*, h->fref[list][i], h->fref[list][i+1] );
+ b_ok = 0;
+ break;
+ }
}
- }
- } while( !b_ok );
+ } while( !b_ok );
+ }
if( h->sh.i_mmco_remove_from_end )
- for( int i = h->i_ref0-1; i >= h->i_ref0 - h->sh.i_mmco_remove_from_end; i-- )
+ for( int i = h->i_ref[0]-1; i >= h->i_ref[0] - h->sh.i_mmco_remove_from_end; i-- )
{
- int diff = h->i_frame_num - h->fref0[i]->i_frame_num;
- h->sh.mmco[h->sh.i_mmco_command_count].i_poc = h->fref0[i]->i_poc;
+ int diff = h->i_frame_num - h->fref[0][i]->i_frame_num;
+ h->sh.mmco[h->sh.i_mmco_command_count].i_poc = h->fref[0][i]->i_poc;
h->sh.mmco[h->sh.i_mmco_command_count++].i_difference_of_pic_nums = diff;
}
- /* Order ref1 from lower to higher poc (bubble sort) for B-frame */
- do
- {
- b_ok = 1;
- for( int i = 0; i < h->i_ref1 - 1; i++ )
- {
- if( h->fref1[i]->i_poc > h->fref1[i+1]->i_poc )
- {
- XCHG( x264_frame_t*, h->fref1[i], h->fref1[i+1] );
- b_ok = 0;
- break;
- }
- }
- } while( !b_ok );
-
x264_reference_check_reorder( h );
- h->i_ref1 = X264_MIN( h->i_ref1, h->frames.i_max_ref1 );
- h->i_ref0 = X264_MIN( h->i_ref0, h->frames.i_max_ref0 );
- h->i_ref0 = X264_MIN( h->i_ref0, h->param.i_frame_reference ); // if reconfig() has lowered the limit
+ h->i_ref[1] = X264_MIN( h->i_ref[1], h->frames.i_max_ref1 );
+ h->i_ref[0] = X264_MIN( h->i_ref[0], h->frames.i_max_ref0 );
+ h->i_ref[0] = X264_MIN( h->i_ref[0], h->param.i_frame_reference ); // if reconfig() has lowered the limit
/* add duplicates */
if( h->fenc->i_type == X264_TYPE_P )
{
int idx = -1;
- if( h->param.analyse.i_weighted_pred == X264_WEIGHTP_SMART )
+ if( h->param.analyse.i_weighted_pred >= X264_WEIGHTP_SIMPLE )
{
x264_weight_t w[3];
w[1].weightfn = w[2].weightfn = NULL;
}
}
}
- else if( h->param.analyse.i_weighted_pred == X264_WEIGHTP_BLIND )
- {
- //weighted offset=-1
- x264_weight_t w[3];
- SET_WEIGHT( w[0], 1, 1, 0, -1 );
- h->fenc->weight[0][0].i_denom = 0;
- w[1].weightfn = w[2].weightfn = NULL;
- idx = x264_weighted_reference_duplicate( h, 0, w );
- }
h->mb.ref_blind_dupe = idx;
}
- assert( h->i_ref0 + h->i_ref1 <= X264_REF_MAX );
- h->mb.pic.i_fref[0] = h->i_ref0;
- h->mb.pic.i_fref[1] = h->i_ref1;
+ assert( h->i_ref[0] + h->i_ref[1] <= X264_REF_MAX );
+ h->mb.pic.i_fref[0] = h->i_ref[0];
+ h->mb.pic.i_fref[1] = h->i_ref[1];
}
static void x264_fdec_filter_row( x264_t *h, int mb_y, int b_inloop )
{
x264_slice_header_init( h, &h->sh, h->sps, h->pps, -1, h->i_frame_num, i_global_qp );
- h->sh.i_num_ref_idx_l0_active = h->i_ref0 <= 0 ? 1 : h->i_ref0;
- h->sh.i_num_ref_idx_l1_active = h->i_ref1 <= 0 ? 1 : h->i_ref1;
+ h->sh.i_num_ref_idx_l0_active = h->i_ref[0] <= 0 ? 1 : h->i_ref[0];
+ h->sh.i_num_ref_idx_l1_active = h->i_ref[1] <= 0 ? 1 : h->i_ref[1];
if( h->sh.i_num_ref_idx_l0_active != h->pps->i_num_ref_idx_l0_default_active ||
(h->sh.i_type == SLICE_TYPE_B && h->sh.i_num_ref_idx_l1_active != h->pps->i_num_ref_idx_l1_default_active) )
{
if( h->param.b_interlaced )
{
h->sh.i_delta_poc_bottom = h->param.b_tff ? 1 : -1;
- if( h->sh.i_delta_poc_bottom == -1 )
- h->sh.i_poc = h->fdec->i_poc + 1;
+ h->sh.i_poc += h->sh.i_delta_poc_bottom == -1;
}
else
h->sh.i_delta_poc_bottom = 0;
+ h->fdec->i_delta_poc[0] = h->sh.i_delta_poc_bottom == -1;
+ h->fdec->i_delta_poc[1] = h->sh.i_delta_poc_bottom == 1;
}
else if( h->sps->i_poc_type == 1 )
{
int i_skip;
int mb_xy, i_mb_x, i_mb_y;
int i_skip_bak = 0; /* Shut up GCC. */
- bs_t bs_bak;
+ bs_t UNINIT(bs_bak);
x264_cabac_t cabac_bak;
uint8_t cabac_prevbyte_bak = 0; /* Shut up GCC. */
int mv_bits_bak = 0;
* other inaccuracies. */
int overhead_guess = (NALU_OVERHEAD - (h->param.b_annexb && h->out.i_nal)) + 1 + h->param.b_cabac + 5;
int slice_max_size = h->param.i_slice_max_size > 0 ? (h->param.i_slice_max_size-overhead_guess)*8 : 0;
+ int back_up_bitstream = slice_max_size || (!h->param.b_cabac && h->sps->i_profile_idc < PROFILE_HIGH);
int starting_bits = bs_pos(&h->out.bs);
int b_deblock = h->sh.i_disable_deblocking_filter_idc != 1;
int b_hpel = h->fdec->b_kept_as_ref;
+ uint8_t *last_emu_check;
b_deblock &= b_hpel || h->param.psz_dump_yuv;
bs_realign( &h->out.bs );
/* init cabac */
x264_cabac_context_init( &h->cabac, h->sh.i_type, x264_clip3( h->sh.i_qp-QP_BD_OFFSET, 0, 51 ), h->sh.i_cabac_init_idc );
x264_cabac_encode_init ( &h->cabac, h->out.bs.p, h->out.bs.p_end );
+ last_emu_check = h->cabac.p;
}
+ else
+ last_emu_check = h->out.bs.p;
h->mb.i_last_qp = h->sh.i_qp;
h->mb.i_last_dqp = 0;
if( x264_bitstream_check_buffer( h ) )
return -1;
- if( slice_max_size )
+ if( back_up_bitstream )
{
mv_bits_bak = h->stat.frame.i_mv_bits;
tex_bits_bak = h->stat.frame.i_tex_bits;
x264_macroblock_analyse( h );
/* encode this macroblock -> be careful it can change the mb type to P_SKIP if needed */
+reencode:
x264_macroblock_encode( h );
if( h->param.b_cabac )
i_skip = 0;
}
x264_macroblock_write_cavlc( h );
+ /* If there was a CAVLC level code overflow, try again at a higher QP. */
+ if( h->mb.b_overflow )
+ {
+ h->mb.i_chroma_qp = h->chroma_qp_table[++h->mb.i_qp];
+ h->mb.i_skip_intra = 0;
+ h->mb.b_skip_mc = 0;
+ h->mb.b_overflow = 0;
+ h->out.bs = bs_bak;
+ i_skip = i_skip_bak;
+ h->stat.frame.i_mv_bits = mv_bits_bak;
+ h->stat.frame.i_tex_bits = tex_bits_bak;
+ goto reencode;
+ }
}
}
/* Count the skip run, just in case. */
if( !h->param.b_cabac )
total_bits += bs_size_ue_big( i_skip );
- /* HACK: we assume no more than 3 bytes of NALU escaping, but
- * this can fail in CABAC streams with an extremely large number of identical
- * blocks in sequence (e.g. all-black intra blocks).
- * Thus, every 64 blocks, pretend we've used a byte.
- * For reference, a seqeuence of identical empty-CBP i16x16 blocks will use
- * one byte after 26 macroblocks, assuming a perfectly adapted CABAC.
- * That's 78 macroblocks to generate the 3-byte sequence to trigger an escape. */
- else if( ((mb_xy - h->sh.i_first_mb) & 63) == 63 )
- slice_max_size -= 8;
+ /* Check for escape bytes. */
+ uint8_t *end = h->param.b_cabac ? h->cabac.p : h->out.bs.p;
+ for( ; last_emu_check < end - 2; last_emu_check++ )
+ if( last_emu_check[0] == 0 && last_emu_check[1] == 0 && last_emu_check[2] <= 3 )
+ {
+ slice_max_size -= 8;
+ last_emu_check++;
+ }
/* We'll just re-encode this last macroblock if we go over the max slice size. */
if( total_bits - starting_bits > slice_max_size && !h->mb.b_reencode_mb )
{
thread_oldest = h;
}
#if HAVE_MMX
- if( h->i_thread_frames == 1 && h->param.cpu&X264_CPU_SSE_MISALIGN )
+ if( h->param.cpu&X264_CPU_SSE_MISALIGN )
x264_cpu_mask_misalign_sse();
#endif
}
else if( h->fenc->i_type == X264_TYPE_P )
{
- int pocdiff = (h->fdec->i_poc - h->fref0[0]->i_poc)/2;
+ int pocdiff = (h->fdec->i_poc - h->fref[0][0]->i_poc)/2;
float increment = X264_MAX( ((float)h->mb.i_mb_width-1) / h->param.i_keyint_max, 1 );
- h->fdec->f_pir_position = h->fref0[0]->f_pir_position;
- h->fdec->i_frames_since_pir = h->fref0[0]->i_frames_since_pir + pocdiff;
+ h->fdec->f_pir_position = h->fref[0][0]->f_pir_position;
+ h->fdec->i_frames_since_pir = h->fref[0][0]->i_frames_since_pir + pocdiff;
if( h->fdec->i_frames_since_pir >= h->param.i_keyint_max ||
(h->b_queued_intra_refresh && h->fdec->f_pir_position + 0.5 >= h->mb.i_mb_width) )
{
int time_to_recovery = h->param.i_open_gop ? 0 : X264_MIN( h->mb.i_mb_width - 1, h->param.i_keyint_max ) + h->param.i_bframe - 1;
x264_nal_start( h, NAL_SEI, NAL_PRIORITY_DISPOSABLE );
x264_sei_recovery_point_write( h, &h->out.bs, time_to_recovery );
- x264_nal_end( h );
+ if( x264_nal_end( h ) )
+ return -1;
+ overhead += h->out.nal[h->out.i_nal-1].i_payload + NALU_OVERHEAD - (h->param.b_annexb && h->out.i_nal-1);
+ }
+
+ if ( h->param.i_frame_packing >= 0 )
+ {
+ x264_nal_start( h, NAL_SEI, NAL_PRIORITY_DISPOSABLE );
+ x264_sei_frame_packing_write( h, &h->out.bs );
+ if( x264_nal_end( h ) )
+ return -1;
overhead += h->out.nal[h->out.i_nal-1].i_payload + NALU_OVERHEAD - (h->param.b_annexb && h->out.i_nal-1);
}
}
x264_reference_check_reorder( h );
}
- if( h->i_ref0 )
- h->fdec->i_poc_l0ref0 = h->fref0[0]->i_poc;
+ if( h->i_ref[0] )
+ h->fdec->i_poc_l0ref0 = h->fref[0][0]->i_poc;
+ /* ------------------------ Create slice header ----------------------- */
+ x264_slice_init( h, i_nal_type, i_global_qp );
+
+ /*------------------------- Weights -------------------------------------*/
if( h->sh.i_type == SLICE_TYPE_B )
x264_macroblock_bipred_init( h );
- /*------------------------- Weights -------------------------------------*/
x264_weighted_pred_init( h );
- /* ------------------------ Create slice header ----------------------- */
- x264_slice_init( h, i_nal_type, i_global_qp );
-
if( i_nal_ref_idc != NAL_PRIORITY_DISPOSABLE )
h->i_frame_num++;
pic_out->i_type = h->fenc->i_type;
pic_out->b_keyframe = h->fenc->b_keyframe;
+ pic_out->i_pic_struct = h->fenc->i_pic_struct;
pic_out->i_pts = h->fdec->i_pts;
pic_out->i_dts = h->fdec->i_dts;
x264_log( h, X264_LOG_WARNING, "invalid DTS: PTS is less than DTS\n" );
pic_out->img.i_csp = X264_CSP_NV12;
-#if X264_HIGH_BIT_DEPTH
+#if HIGH_BIT_DEPTH
pic_out->img.i_csp |= X264_CSP_HIGH_DEPTH;
#endif
pic_out->img.i_plane = h->fdec->i_plane;
for( int i_list = 0; i_list < 2; i_list++ )
for( int i = 0; i < X264_REF_MAX*2; i++ )
h->stat.i_mb_count_ref[h->sh.i_type][i_list][i] += h->stat.frame.i_mb_count_ref[i_list][i];
- if( h->sh.i_type == SLICE_TYPE_P )
+ if( h->sh.i_type == SLICE_TYPE_P && h->param.analyse.i_weighted_pred >= X264_WEIGHTP_SIMPLE )
{
- h->stat.i_consecutive_bframes[h->fdec->i_frame - h->fref0[0]->i_frame - 1]++;
- if( h->param.analyse.i_weighted_pred == X264_WEIGHTP_SMART )
- for( int i = 0; i < 3; i++ )
- for( int j = 0; j < h->i_ref0; j++ )
- if( h->sh.weight[0][i].i_denom != 0 )
- {
- h->stat.i_wpred[i]++;
- break;
- }
+ h->stat.i_wpred[0] += !!h->sh.weight[0][0].weightfn;
+ h->stat.i_wpred[1] += !!h->sh.weight[0][1].weightfn || !!h->sh.weight[0][2].weightfn;
}
if( h->sh.i_type == SLICE_TYPE_B )
{
h->stat.i_direct_score[i] += h->stat.frame.i_direct_score[i];
}
}
+ else
+ h->stat.i_consecutive_bframes[h->fenc->i_bframes]++;
psz_message[0] = '\0';
double dur = h->fenc->f_duration;
/* Remove duplicates, must be done near the end as breaks h->fref0 array
* by freeing some of its pointers. */
- for( int i = 0; i < h->i_ref0; i++ )
- if( h->fref0[i] && h->fref0[i]->b_duplicate )
+ for( int i = 0; i < h->i_ref[0]; i++ )
+ if( h->fref[0][i] && h->fref[0][i]->b_duplicate )
{
- x264_frame_push_blank_unused( h, h->fref0[i] );
- h->fref0[i] = 0;
+ x264_frame_push_blank_unused( h, h->fref[0][i] );
+ h->fref[0][i] = 0;
}
if( h->param.psz_dump_yuv )
}
}
}
- if( h->param.i_bframe && h->stat.i_frame_count[SLICE_TYPE_P] )
+ if( h->param.i_bframe && h->stat.i_frame_count[SLICE_TYPE_B] )
{
char *p = buf;
int den = 0;
- // weight by number of frames (including the P-frame) that are in a sequence of N B-frames
+ // weight by number of frames (including the I/P-frames) that are in a sequence of N B-frames
for( int i = 0; i <= h->param.i_bframe; i++ )
den += (i+1) * h->stat.i_consecutive_bframes[i];
for( int i = 0; i <= h->param.i_bframe; i++ )
fixed_pred_modes[3][2] * 100.0 / sum_pred_modes[3],
fixed_pred_modes[3][3] * 100.0 / sum_pred_modes[3] );
- if( h->param.analyse.i_weighted_pred == X264_WEIGHTP_SMART && h->stat.i_frame_count[SLICE_TYPE_P] > 0 )
- x264_log( h, X264_LOG_INFO, "Weighted P-Frames: Y:%.1f%%\n",
- h->stat.i_wpred[0] * 100.0 / h->stat.i_frame_count[SLICE_TYPE_P] );
+ if( h->param.analyse.i_weighted_pred >= X264_WEIGHTP_SIMPLE && h->stat.i_frame_count[SLICE_TYPE_P] > 0 )
+ x264_log( h, X264_LOG_INFO, "Weighted P-Frames: Y:%.1f%% UV:%.1f%%\n",
+ h->stat.i_wpred[0] * 100.0 / h->stat.i_frame_count[SLICE_TYPE_P],
+ h->stat.i_wpred[1] * 100.0 / h->stat.i_frame_count[SLICE_TYPE_P] );
for( int i_list = 0; i_list < 2; i_list++ )
for( int i_slice = 0; i_slice < 2; i_slice++ )
h = h->thread[0];
+ for( int i = 0; i < h->i_thread_frames; i++ )
+ if( h->thread[i]->b_thread_active )
+ for( int j = 0; j < h->thread[i]->i_ref[0]; j++ )
+ if( h->thread[i]->fref[0][j] && h->thread[i]->fref[0][j]->b_duplicate )
+ x264_frame_delete( h->thread[i]->fref[0][j] );
+
for( int i = h->param.i_threads - 1; i >= 0; i-- )
{
x264_frame_t **frame;
x264_frame_delete( *frame );
}
frame = &h->thread[i]->fdec;
- assert( (*frame)->i_reference_count > 0 );
- (*frame)->i_reference_count--;
- if( (*frame)->i_reference_count == 0 )
- x264_frame_delete( *frame );
+ if( *frame )
+ {
+ assert( (*frame)->i_reference_count > 0 );
+ (*frame)->i_reference_count--;
+ if( (*frame)->i_reference_count == 0 )
+ x264_frame_delete( *frame );
+ }
x264_macroblock_cache_free( h->thread[i] );
}
x264_macroblock_thread_free( h->thread[i], 0 );
}
}
-/****************************************************************************
- * x264_encoder_delayed_frames:
- ****************************************************************************/
int x264_encoder_delayed_frames( x264_t *h )
{
int delayed_frames = 0;
x264_pthread_mutex_unlock( &h->lookahead->ofbuf.mutex );
return delayed_frames;
}
+
+int x264_encoder_maximum_delayed_frames( x264_t *h )
+{
+ return h->frames.i_delay;
+}
projects / x264 / blobdiff