/*****************************************************************************
* 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++ )
{
*
****************************************************************************/
-static int x264_validate_parameters( x264_t *h )
+static int x264_validate_parameters( x264_t *h, int b_open )
{
#if HAVE_MMX
#ifdef __SSE__
- if( !(x264_cpu_detect() & X264_CPU_SSE) )
+ if( b_open && !(x264_cpu_detect() & X264_CPU_SSE) )
{
x264_log( h, X264_LOG_ERROR, "your cpu does not support SSE1, but x264 was compiled with asm support\n");
#else
- if( !(x264_cpu_detect() & X264_CPU_MMXEXT) )
+ if( b_open && !(x264_cpu_detect() & X264_CPU_MMXEXT) )
{
x264_log( h, X264_LOG_ERROR, "your cpu does not support MMXEXT, but x264 was compiled with asm support\n");
#endif
h->param.analyse.i_weighted_pred = 0;
}
- if( h->param.b_interlaced )
- {
- if( h->param.analyse.i_me_method >= X264_ME_ESA )
- {
- x264_log( h, X264_LOG_WARNING, "interlace + me=esa is not implemented\n" );
- h->param.analyse.i_me_method = X264_ME_UMH;
- }
- if( h->param.analyse.i_weighted_pred > 0 )
- {
- x264_log( h, X264_LOG_WARNING, "interlace + weightp is not implemented\n" );
- h->param.analyse.i_weighted_pred = X264_WEIGHTP_NONE;
- }
- }
-
h->param.i_frame_packing = x264_clip3( h->param.i_frame_packing, -1, 5 );
/* Detect default ffmpeg settings and terminate with an error. */
+ if( b_open )
{
int score = 0;
score += h->param.analyse.i_me_range == 0;
return -1;
}
h->param.rc.f_rf_constant = x264_clip3f( h->param.rc.f_rf_constant, -QP_BD_OFFSET, 51 );
+ h->param.rc.f_rf_constant_max = x264_clip3f( h->param.rc.f_rf_constant_max, -QP_BD_OFFSET, 51 );
h->param.rc.i_qp_constant = x264_clip3( h->param.rc.i_qp_constant, 0, QP_MAX );
+ h->param.analyse.i_subpel_refine = x264_clip3( h->param.analyse.i_subpel_refine, 0, 10 );
+ h->param.rc.f_ip_factor = X264_MAX( h->param.rc.f_ip_factor, 0.01f );
+ h->param.rc.f_pb_factor = X264_MAX( h->param.rc.f_pb_factor, 0.01f );
if( h->param.rc.i_rc_method == X264_RC_CRF )
{
h->param.rc.i_qp_constant = h->param.rc.f_rf_constant + QP_BD_OFFSET;
h->param.rc.i_qp_max = x264_clip3( (int)(X264_MAX3( qp_p, qp_i, qp_b ) + .999), 0, QP_MAX );
h->param.rc.i_aq_mode = 0;
h->param.rc.b_mb_tree = 0;
+ h->param.rc.i_bitrate = 0;
}
h->param.rc.i_qp_max = x264_clip3( h->param.rc.i_qp_max, 0, QP_MAX );
h->param.rc.i_qp_min = x264_clip3( h->param.rc.i_qp_min, 0, h->param.rc.i_qp_max );
+ h->param.rc.i_qp_step = x264_clip3( h->param.rc.i_qp_step, 0, QP_MAX );
+ h->param.rc.i_bitrate = x264_clip3( h->param.rc.i_bitrate, 0, 2000000 );
+ h->param.rc.i_vbv_buffer_size = x264_clip3( h->param.rc.i_vbv_buffer_size, 0, 2000000 );
+ h->param.rc.i_vbv_max_bitrate = x264_clip3( h->param.rc.i_vbv_max_bitrate, 0, 2000000 );
+ h->param.rc.f_vbv_buffer_init = x264_clip3f( h->param.rc.f_vbv_buffer_init, 0, 2000000 );
if( h->param.rc.i_vbv_buffer_size )
{
if( h->param.rc.i_rc_method == X264_RC_CQP )
x264_log( h, X264_LOG_WARNING, "interlaced + slice-max-mbs is not implemented\n" );
h->param.i_slice_max_mbs = 0;
}
+ h->param.i_slice_max_size = X264_MAX( h->param.i_slice_max_size, 0 );
+ h->param.i_slice_max_mbs = X264_MAX( h->param.i_slice_max_mbs, 0 );
+
+ h->param.b_interlaced = !!h->param.b_interlaced;
int max_slices = (h->param.i_height+((16<<h->param.b_interlaced)-1))/(16<<h->param.b_interlaced);
if( h->param.b_sliced_threads )
h->param.i_slice_count = x264_clip3( h->param.i_threads, 0, max_slices );
else
{
h->param.i_slice_count = x264_clip3( h->param.i_slice_count, 0, max_slices );
- h->param.i_slice_max_size = X264_MAX( h->param.i_slice_max_size, 0 );
- h->param.i_slice_max_mbs = X264_MAX( h->param.i_slice_max_mbs, 0 );
if( h->param.i_slice_max_mbs || h->param.i_slice_max_size )
h->param.i_slice_count = 0;
}
+ if( h->param.b_bluray_compat )
+ {
+ h->param.i_bframe_pyramid = X264_MIN( X264_B_PYRAMID_STRICT, h->param.i_bframe_pyramid );
+ h->param.i_bframe = X264_MIN( h->param.i_bframe, 3 );
+ h->param.b_aud = 1;
+ h->param.i_nal_hrd = X264_MAX( h->param.i_nal_hrd, X264_NAL_HRD_VBR );
+ h->param.i_slice_max_size = 0;
+ h->param.i_slice_max_mbs = 0;
+ h->param.b_intra_refresh = 0;
+ h->param.i_frame_reference = X264_MIN( h->param.i_frame_reference, 6 );
+ h->param.i_dpb_size = X264_MIN( h->param.i_dpb_size, 6 );
+ /* Due to the proliferation of broken players that don't handle dupes properly. */
+ h->param.analyse.i_weighted_pred = X264_MIN( h->param.analyse.i_weighted_pred, X264_WEIGHTP_SIMPLE );
+ if( h->param.b_fake_interlaced )
+ h->param.b_pic_struct = 1;
+ }
+
h->param.i_frame_reference = x264_clip3( h->param.i_frame_reference, 1, X264_REF_MAX );
h->param.i_dpb_size = x264_clip3( h->param.i_dpb_size, 1, X264_REF_MAX );
if( h->param.i_scenecut_threshold < 0 )
h->param.i_scenecut_threshold = 0;
+ h->param.analyse.i_direct_mv_pred = x264_clip3( h->param.analyse.i_direct_mv_pred, X264_DIRECT_PRED_NONE, X264_DIRECT_PRED_AUTO );
if( !h->param.analyse.i_subpel_refine && h->param.analyse.i_direct_mv_pred > X264_DIRECT_PRED_SPATIAL )
{
x264_log( h, X264_LOG_WARNING, "subme=0 + direct=temporal is not supported\n" );
h->param.analyse.i_direct_mv_pred = X264_DIRECT_PRED_SPATIAL;
}
h->param.i_bframe = x264_clip3( h->param.i_bframe, 0, X264_MIN( X264_BFRAME_MAX, h->param.i_keyint_max-1 ) );
- h->param.i_open_gop = x264_clip3( h->param.i_open_gop, X264_OPEN_GOP_NONE, X264_OPEN_GOP_BLURAY );
h->param.i_bframe_bias = x264_clip3( h->param.i_bframe_bias, -90, 100 );
if( h->param.i_bframe <= 1 )
h->param.i_bframe_pyramid = X264_B_PYRAMID_NONE;
h->param.i_bframe_pyramid = x264_clip3( h->param.i_bframe_pyramid, X264_B_PYRAMID_NONE, X264_B_PYRAMID_NORMAL );
+ h->param.i_bframe_adaptive = x264_clip3( h->param.i_bframe_adaptive, X264_B_ADAPT_NONE, X264_B_ADAPT_TRELLIS );
if( !h->param.i_bframe )
{
h->param.i_bframe_adaptive = X264_B_ADAPT_NONE;
h->param.analyse.i_direct_mv_pred = 0;
h->param.analyse.b_weighted_bipred = 0;
- h->param.i_open_gop = X264_OPEN_GOP_NONE;
+ h->param.b_open_gop = 0;
}
if( h->param.b_intra_refresh && h->param.i_bframe_pyramid == X264_B_PYRAMID_NORMAL )
{
h->param.i_frame_reference = 1;
h->param.i_dpb_size = 1;
}
- if( h->param.b_intra_refresh && h->param.i_open_gop )
+ if( h->param.b_intra_refresh && h->param.b_open_gop )
{
x264_log( h, X264_LOG_WARNING, "intra-refresh is not compatible with open-gop\n" );
- h->param.i_open_gop = X264_OPEN_GOP_NONE;
+ h->param.b_open_gop = 0;
}
float fps = h->param.i_fps_num > 0 && h->param.i_fps_den > 0 ? (float) h->param.i_fps_num / h->param.i_fps_den : 25.0;
if( h->param.i_keyint_min == X264_KEYINT_MIN_AUTO )
if( h->param.analyse.i_me_method < X264_ME_DIA ||
h->param.analyse.i_me_method > X264_ME_TESA )
h->param.analyse.i_me_method = X264_ME_HEX;
- if( h->param.analyse.i_me_range < 4 )
- h->param.analyse.i_me_range = 4;
+ h->param.analyse.i_me_range = x264_clip3( h->param.analyse.i_me_range, 4, 1024 );
if( h->param.analyse.i_me_range > 16 && h->param.analyse.i_me_method <= X264_ME_HEX )
h->param.analyse.i_me_range = 16;
if( h->param.analyse.i_me_method == X264_ME_TESA &&
(h->mb.b_lossless || h->param.analyse.i_subpel_refine <= 1) )
h->param.analyse.i_me_method = X264_ME_ESA;
- h->param.analyse.i_subpel_refine = x264_clip3( h->param.analyse.i_subpel_refine, 0, 10 );
h->param.analyse.b_mixed_references = h->param.analyse.b_mixed_references && h->param.i_frame_reference > 1;
h->param.analyse.inter &= X264_ANALYSE_PSUB16x16|X264_ANALYSE_PSUB8x8|X264_ANALYSE_BSUB16x16|
X264_ANALYSE_I4x4|X264_ANALYSE_I8x8;
}
h->param.analyse.i_chroma_qp_offset = x264_clip3(h->param.analyse.i_chroma_qp_offset, -12, 12);
h->param.analyse.i_trellis = x264_clip3( h->param.analyse.i_trellis, 0, 2 );
+ h->param.rc.i_aq_mode = x264_clip3( h->param.rc.i_aq_mode, 0, 2 );
+ h->param.rc.f_aq_strength = x264_clip3f( h->param.rc.f_aq_strength, 0, 3 );
+ if( h->param.rc.f_aq_strength == 0 )
+ h->param.rc.i_aq_mode = 0;
+
+ if( h->param.i_log_level < X264_LOG_INFO )
+ {
+ h->param.analyse.b_psnr = 0;
+ h->param.analyse.b_ssim = 0;
+ }
+ /* Warn users trying to measure PSNR/SSIM with psy opts on. */
+ if( b_open && (h->param.analyse.b_psnr || h->param.analyse.b_ssim) )
+ {
+ char *s = NULL;
+
+ if( h->param.analyse.b_psy )
+ {
+ s = h->param.analyse.b_psnr ? "psnr" : "ssim";
+ x264_log( h, X264_LOG_WARNING, "--%s used with psy on: results will be invalid!\n", s );
+ }
+ else if( !h->param.rc.i_aq_mode && h->param.analyse.b_ssim )
+ {
+ x264_log( h, X264_LOG_WARNING, "--ssim used with AQ off: results will be invalid!\n" );
+ s = "ssim";
+ }
+ else if( h->param.rc.i_aq_mode && h->param.analyse.b_psnr )
+ {
+ x264_log( h, X264_LOG_WARNING, "--psnr used with AQ on: results will be invalid!\n" );
+ s = "psnr";
+ }
+ if( s )
+ x264_log( h, X264_LOG_WARNING, "--tune %s should be used if attempting to benchmark %s!\n", s, s );
+ }
+
if( !h->param.analyse.b_psy )
{
h->param.analyse.f_psy_rd = 0;
h->param.analyse.f_psy_trellis = 0;
}
- if( !h->param.analyse.i_trellis )
- h->param.analyse.f_psy_trellis = 0;
h->param.analyse.f_psy_rd = x264_clip3f( h->param.analyse.f_psy_rd, 0, 10 );
h->param.analyse.f_psy_trellis = x264_clip3f( h->param.analyse.f_psy_trellis, 0, 10 );
- if( h->param.analyse.i_subpel_refine < 6 )
- h->param.analyse.f_psy_rd = 0;
- h->mb.i_psy_rd = FIX8( h->param.analyse.f_psy_rd );
+ h->mb.i_psy_rd = h->param.analyse.i_subpel_refine >= 6 ? FIX8( h->param.analyse.f_psy_rd ) : 0;
+ h->mb.i_psy_trellis = h->param.analyse.i_trellis ? FIX8( h->param.analyse.f_psy_trellis / 4 ) : 0;
/* Psy RDO increases overall quantizers to improve the quality of luma--this indirectly hurts chroma quality */
/* so we lower the chroma QP offset to compensate */
- /* This can be triggered repeatedly on multiple calls to parameter_validate, but since encoding
- * uses the pps chroma qp offset not the param chroma qp offset, this is not a problem. */
- if( h->mb.i_psy_rd )
+ if( b_open && h->mb.i_psy_rd )
h->param.analyse.i_chroma_qp_offset -= h->param.analyse.f_psy_rd < 0.25 ? 1 : 2;
- h->mb.i_psy_trellis = FIX8( h->param.analyse.f_psy_trellis / 4 );
/* Psy trellis has a similar effect. */
- if( h->mb.i_psy_trellis )
+ if( b_open && h->mb.i_psy_trellis )
h->param.analyse.i_chroma_qp_offset -= h->param.analyse.f_psy_trellis < 0.25 ? 1 : 2;
h->param.analyse.i_chroma_qp_offset = x264_clip3(h->param.analyse.i_chroma_qp_offset, -12, 12);
- h->param.rc.i_aq_mode = x264_clip3( h->param.rc.i_aq_mode, 0, 2 );
- h->param.rc.f_aq_strength = x264_clip3f( h->param.rc.f_aq_strength, 0, 3 );
- if( h->param.rc.f_aq_strength == 0 )
- h->param.rc.i_aq_mode = 0;
/* MB-tree requires AQ to be on, even if the strength is zero. */
if( !h->param.rc.i_aq_mode && h->param.rc.b_mb_tree )
{
}
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.b_interlaced )
+ {
+ if( h->param.analyse.i_me_method >= X264_ME_ESA )
+ {
+ x264_log( h, X264_LOG_WARNING, "interlace + me=esa is not implemented\n" );
+ h->param.analyse.i_me_method = X264_ME_UMH;
+ }
+ if( h->param.analyse.i_weighted_pred > 0 )
+ {
+ x264_log( h, X264_LOG_WARNING, "interlace + weightp is not implemented\n" );
+ h->param.analyse.i_weighted_pred = X264_WEIGHTP_NONE;
+ }
+ }
+
+ 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->i_thread_frames > 1 )
h->param.analyse.i_mv_range_thread = r2;
}
+ if( h->param.rc.f_rate_tolerance < 0 )
+ h->param.rc.f_rate_tolerance = 0;
if( h->param.rc.f_qblur < 0 )
h->param.rc.f_qblur = 0;
if( h->param.rc.f_complexity_blur < 0 )
h->param.i_sps_id &= 31;
- if( h->param.i_log_level < X264_LOG_INFO )
- {
- h->param.analyse.b_psnr = 0;
- h->param.analyse.b_ssim = 0;
- }
-
if( h->param.b_interlaced )
h->param.b_pic_struct = 1;
+ h->param.i_nal_hrd = x264_clip3( h->param.i_nal_hrd, X264_NAL_HRD_NONE, X264_NAL_HRD_CBR );
+
if( h->param.i_nal_hrd && !h->param.rc.i_vbv_buffer_size )
{
x264_log( h, X264_LOG_WARNING, "NAL HRD parameters require VBV parameters\n" );
BOOLIFY( b_repeat_headers );
BOOLIFY( b_annexb );
BOOLIFY( b_vfr_input );
+ BOOLIFY( b_pulldown );
+ BOOLIFY( b_tff );
BOOLIFY( b_pic_struct );
BOOLIFY( b_fake_interlaced );
+ BOOLIFY( b_open_gop );
+ BOOLIFY( b_bluray_compat );
BOOLIFY( analyse.b_transform_8x8 );
BOOLIFY( analyse.b_weighted_bipred );
BOOLIFY( analyse.b_chroma_me );
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 );
}
}
}
goto fail;
}
- if( x264_validate_parameters( h ) < 0 )
+ if( x264_validate_parameters( h, 1 ) < 0 )
goto fail;
if( h->param.psz_cqm_file )
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;
h->mb.i_mb_width = h->sps->i_mb_width;
h->mb.i_mb_height = h->sps->i_mb_height;
h->mb.i_mb_count = h->mb.i_mb_width * h->mb.i_mb_height;
+ /* Adaptive MBAFF and subme 0 are not supported as we require halving motion
+ * vectors during prediction, resulting in hpel mvs.
+ * The chosen solution is to make MBAFF non-adaptive in this case. */
+ h->mb.b_adaptive_mbaff = h->param.b_interlaced && h->param.analyse.i_subpel_refine;
/* Init frames. */
if( h->param.i_bframe_adaptive == X264_B_ADAPT_TRELLIS && !h->param.rc.b_stat_read )
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->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;
x264_predict_8x8c_init( h->param.cpu, h->predict_8x8c );
x264_predict_8x8_init( h->param.cpu, h->predict_8x8, &h->predict_8x8_filter );
x264_predict_4x4_init( h->param.cpu, h->predict_4x4 );
- if( !h->param.b_cabac )
- x264_init_vlc_tables();
+ if( h->param.b_cabac )
+ x264_cabac_init();
+ else
+ x264_cavlc_init();
x264_pixel_init( h->param.cpu, &h->pixf );
x264_dct_init( h->param.cpu, &h->dctf );
x264_zigzag_init( h->param.cpu, &h->zigzagf, h->param.b_interlaced );
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++ )
- if( x264_analyse_init_costs( h, qp ) )
+ float *logs = x264_analyse_prepare_costs( h );
+ if( !logs )
+ goto fail;
+ 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, logs, qp ) )
goto fail;
- if( x264_analyse_init_costs( h, X264_LOOKAHEAD_QP ) )
+ if( x264_analyse_init_costs( h, logs, X264_LOOKAHEAD_QP ) )
goto fail;
+ x264_free( logs );
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;
* ( h->param.rc.i_rc_method == X264_RC_ABR ? pow( 0.95, h->param.rc.i_qp_min )
: pow( 0.95, h->param.rc.i_qp_constant ) * X264_MAX( 1, h->param.rc.f_ip_factor )));
- CHECKED_MALLOC( h->nal_buffer, h->out.i_bitstream * 3/2 + 4 );
h->nal_buffer_size = h->out.i_bitstream * 3/2 + 4;
+ CHECKED_MALLOC( h->nal_buffer, h->nal_buffer_size );
if( h->param.i_threads > 1 &&
x264_threadpool_init( &h->threadpool, h->param.i_threads, (void*)x264_encoder_thread_init, h ) )
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) );
if( h->param.rc.i_vbv_max_bitrate > 0 && h->param.rc.i_vbv_buffer_size > 0 &&
param->rc.i_vbv_max_bitrate > 0 && param->rc.i_vbv_buffer_size > 0 )
{
+ rc_reconfig |= h->param.rc.i_vbv_max_bitrate != param->rc.i_vbv_max_bitrate;
+ rc_reconfig |= h->param.rc.i_vbv_buffer_size != param->rc.i_vbv_buffer_size;
+ rc_reconfig |= h->param.rc.i_bitrate != param->rc.i_bitrate;
COPY( rc.i_vbv_max_bitrate );
COPY( rc.i_vbv_buffer_size );
COPY( rc.i_bitrate );
- rc_reconfig = 1;
- }
- if( h->param.rc.f_rf_constant != param->rc.f_rf_constant )
- {
- COPY( rc.f_rf_constant );
- rc_reconfig = 1;
- }
- if( h->param.rc.f_rf_constant_max != param->rc.f_rf_constant_max )
- {
- COPY( rc.f_rf_constant_max );
- rc_reconfig = 1;
}
-
+ rc_reconfig |= h->param.rc.f_rf_constant != param->rc.f_rf_constant;
+ rc_reconfig |= h->param.rc.f_rf_constant_max != param->rc.f_rf_constant_max;
+ COPY( rc.f_rf_constant );
+ COPY( rc.f_rf_constant_max );
#undef COPY
mbcmp_init( h );
- int ret = x264_validate_parameters( h );
+ int ret = x264_validate_parameters( h, 0 );
/* Supported reconfiguration options (1-pass only):
* vbv-maxrate
nal_size += h->out.nal[i].i_payload;
/* Worst-case NAL unit escaping: reallocate the buffer if it's too small. */
- if( h->nal_buffer_size < nal_size * 3/2 + h->out.i_nal * 4 )
+ int necessary_size = nal_size * 3/2 + h->out.i_nal * 4;
+ if( h->nal_buffer_size < necessary_size )
{
- uint8_t *buf = x264_malloc( nal_size * 2 + h->out.i_nal * 4 );
+ h->nal_buffer_size = necessary_size * 2;
+ uint8_t *buf = x264_malloc( h->nal_buffer_size );
if( !buf )
return -1;
if( previous_nal_size )
return -1;
frame_size = x264_encoder_encapsulate_nals( h, 0 );
+ if( frame_size < 0 )
+ return -1;
/* now set output*/
*pi_nal = h->out.i_nal;
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;
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 buffer_next = 0;
for( int i = 0; i < 3; i++ )
{
- for( int j = 0; j < h->i_ref0; j++ )
+ for( int j = 0; j < h->i_ref[0]; j++ )
{
if( h->fenc->weight[j][i].weightfn )
{
//scale full resolution frame
if( h->param.i_threads == 1 )
{
- pixel *src = h->fref0[j]->filtered[0] - h->fref0[j]->i_stride[0]*i_padv - PADH;
+ 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;
}
if( weightplane[1] )
- for( int i = 0; i < h->i_ref0; i++ )
+ for( int i = 0; i < h->i_ref[0]; i++ )
{
if( h->sh.weight[i][1].weightfn && !h->sh.weight[i][2].weightfn )
{
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
+
+ /* For Blu-ray compliance, don't reference frames outside of the minigop. */
+ if( IS_X264_TYPE_B( h->fenc->i_type ) && h->param.b_bluray_compat )
+ h->i_ref[0] = X264_MIN( h->i_ref[0], IS_X264_TYPE_B( h->fref[0][0]->i_type ) + 1 );
/* add duplicates */
if( h->fenc->i_type == X264_TYPE_P )
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 )
int b_measure_quality = 1;
int min_y = mb_y - (1 << h->sh.b_mbaff);
int b_start = min_y == h->i_threadslice_start;
- int max_y = b_end ? h->i_threadslice_end : mb_y;
+ /* Even in interlaced mode, deblocking never modifies more than 4 pixels
+ * above each MB, as bS=4 doesn't happen for the top of interlaced mbpairs. */
+ int minpix_y = min_y*16 - 4 * !b_start;
+ int maxpix_y = mb_y*16 - 4 * !b_end;
b_deblock &= b_hpel || h->param.psz_dump_yuv;
if( h->param.b_sliced_threads && b_start && min_y && !b_inloop )
{
return;
if( b_deblock )
- for( int y = min_y; y < max_y; y += (1 << h->sh.b_mbaff) )
+ for( int y = min_y; y < mb_y; y += (1 << h->sh.b_mbaff) )
x264_frame_deblock_row( h, y );
+ /* FIXME: Prediction requires different borders for interlaced/progressive mc,
+ * but the actual image data is equivalent. For now, maintain this
+ * consistency by copying deblocked pixels between planes. */
+ if( h->param.b_interlaced )
+ for( int p = 0; p < 2; p++ )
+ for( int i = minpix_y>>p; i < maxpix_y>>p; i++ )
+ memcpy( h->fdec->plane_fld[p] + i*h->fdec->i_stride[p],
+ h->fdec->plane[p] + i*h->fdec->i_stride[p],
+ h->mb.i_mb_width*16*sizeof(pixel) );
+
if( b_hpel )
{
int end = mb_y == h->mb.i_mb_height;
}
}
+ if( h->sh.b_mbaff )
+ for( int i = 0; i < 2; i++ )
+ {
+ XCHG( pixel *, h->intra_border_backup[0][i], h->intra_border_backup[3][i] );
+ XCHG( pixel *, h->intra_border_backup[1][i], h->intra_border_backup[4][i] );
+ }
+
if( h->i_thread_frames > 1 && h->fdec->b_kept_as_ref )
x264_frame_cond_broadcast( h->fdec, mb_y*16 + (b_end ? 10000 : -(X264_THREAD_HEIGHT << h->sh.b_mbaff)) );
- min_y = min_y*16 - 8 * !b_start;
- max_y = b_end ? X264_MIN( h->i_threadslice_end*16 , h->param.i_height ) : mb_y*16 - 8;
-
if( b_measure_quality )
{
+ maxpix_y = X264_MIN( maxpix_y, h->param.i_height );
if( h->param.analyse.b_psnr )
{
uint64_t ssd_y = x264_pixel_ssd_wxh( &h->pixf,
- h->fdec->plane[0] + min_y * h->fdec->i_stride[0], h->fdec->i_stride[0],
- h->fenc->plane[0] + min_y * h->fenc->i_stride[0], h->fenc->i_stride[0],
- h->param.i_width, max_y-min_y );
+ h->fdec->plane[0] + minpix_y * h->fdec->i_stride[0], h->fdec->i_stride[0],
+ h->fenc->plane[0] + minpix_y * h->fenc->i_stride[0], h->fenc->i_stride[0],
+ h->param.i_width, maxpix_y-minpix_y );
uint64_t ssd_u, ssd_v;
x264_pixel_ssd_nv12( &h->pixf,
- h->fdec->plane[1] + (min_y>>1) * h->fdec->i_stride[1], h->fdec->i_stride[1],
- h->fenc->plane[1] + (min_y>>1) * h->fenc->i_stride[1], h->fenc->i_stride[1],
- h->param.i_width>>1, (max_y-min_y)>>1, &ssd_u, &ssd_v );
+ h->fdec->plane[1] + (minpix_y>>1) * h->fdec->i_stride[1], h->fdec->i_stride[1],
+ h->fenc->plane[1] + (minpix_y>>1) * h->fenc->i_stride[1], h->fenc->i_stride[1],
+ h->param.i_width>>1, (maxpix_y-minpix_y)>>1, &ssd_u, &ssd_v );
h->stat.frame.i_ssd[0] += ssd_y;
h->stat.frame.i_ssd[1] += ssd_u;
h->stat.frame.i_ssd[2] += ssd_v;
x264_emms();
/* offset by 2 pixels to avoid alignment of ssim blocks with dct blocks,
* and overlap by 4 */
- min_y += b_start ? 2 : -6;
+ minpix_y += b_start ? 2 : -6;
h->stat.frame.f_ssim +=
x264_pixel_ssim_wxh( &h->pixf,
- h->fdec->plane[0] + 2+min_y*h->fdec->i_stride[0], h->fdec->i_stride[0],
- h->fenc->plane[0] + 2+min_y*h->fenc->i_stride[0], h->fenc->i_stride[0],
- h->param.i_width-2, max_y-min_y, h->scratch_buffer );
+ h->fdec->plane[0] + 2+minpix_y*h->fdec->i_stride[0], h->fdec->i_stride[0],
+ h->fenc->plane[0] + 2+minpix_y*h->fenc->i_stride[0], h->fenc->i_stride[0],
+ h->param.i_width-2, maxpix_y-minpix_y, h->scratch_buffer );
}
}
}
{
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->fenc->i_type == X264_TYPE_BREF && h->param.b_bluray_compat && h->sh.i_mmco_command_count )
+ {
+ h->b_sh_backup = 1;
+ h->sh_backup = h->sh;
+ }
+
h->fdec->i_frame_num = h->sh.i_frame_num;
if( h->sps->i_poc_type == 0 )
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;
if( i_mb_x == 0 && !h->mb.b_reencode_mb )
x264_fdec_filter_row( h, i_mb_y, 1 );
+ if( h->param.b_interlaced )
+ {
+ if( h->mb.b_adaptive_mbaff )
+ {
+ if( !(i_mb_y&1) )
+ h->mb.b_interlaced = 1;
+ x264_zigzag_init( h->param.cpu, &h->zigzagf, h->mb.b_interlaced );
+ }
+ h->mb.field[mb_xy] = h->mb.b_interlaced;
+ }
+
/* load cache */
x264_macroblock_cache_load( h, i_mb_x, i_mb_y );
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 )
{
memcpy( &dst->i_frame, &src->i_frame, offsetof(x264_t, mb.type) - offsetof(x264_t, i_frame) );
dst->param = src->param;
dst->stat = src->stat;
+ dst->pixf = src->pixf;
}
static void x264_thread_sync_stat( x264_t *dst, x264_t *src )
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
i_nal_ref_idc = NAL_PRIORITY_HIGH; /* Not completely true but for now it is (as all I/P are kept as ref)*/
h->sh.i_type = SLICE_TYPE_I;
x264_reference_hierarchy_reset( h );
- if( h->param.i_open_gop )
+ if( h->param.b_open_gop )
h->frames.i_poc_last_open_gop = h->fenc->b_keyframe ? h->fenc->i_poc : -1;
}
else if( h->fenc->i_type == X264_TYPE_P )
}
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) )
{
if( h->fenc->i_type != X264_TYPE_IDR )
{
- 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;
+ int time_to_recovery = h->param.b_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 );
if( x264_nal_end( h ) )
overhead += h->out.nal[h->out.i_nal-1].i_payload + NALU_OVERHEAD - (h->param.b_annexb && h->out.i_nal-1);
}
+ /* As required by Blu-ray. */
+ if( !IS_X264_TYPE_B( h->fenc->i_type ) && h->b_sh_backup )
+ {
+ h->b_sh_backup = 0;
+ x264_nal_start( h, NAL_SEI, NAL_PRIORITY_DISPOSABLE );
+ x264_sei_dec_ref_pic_marking_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);
+ }
+
if( h->fenc->b_keyframe && h->param.b_intra_refresh )
h->i_cpb_delay_pir_offset = h->fenc->i_cpb_delay;
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++;
}
int frame_size = x264_encoder_encapsulate_nals( h, 0 );
+ if( frame_size < 0 )
+ return -1;
/* Set output picture properties */
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;
if( x264_nal_end( h ) )
return -1;
int total_size = x264_encoder_encapsulate_nals( h, h->out.i_nal-1 );
+ if( total_size < 0 )
+ return -1;
frame_size += total_size;
filler -= total_size;
}
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_SIMPLE )
- {
- 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;
- }
+ 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++ )
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 );