/*****************************************************************************
- * x264: h264 encoder
+ * encoder.c: top-level encoder functions
*****************************************************************************
- * Copyright (C) 2003-2008 x264 project
+ * Copyright (C) 2003-2011 x264 project
*
* Authors: Laurent Aimar <fenrir@via.ecp.fr>
* Loren Merritt <lorenm@u.washington.edu>
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02111, USA.
+ *
+ * This program is also available under a commercial proprietary license.
+ * For more information, contact us at licensing@x264.com.
*****************************************************************************/
-#include <math.h>
-
#include "common/common.h"
#include "set.h"
******************************* x264 libs **********************************
*
****************************************************************************/
-static float x264_psnr( int64_t i_sqe, int64_t i_size )
+static double x264_psnr( double sqe, double size )
{
- double f_mse = (double)i_sqe / (PIXEL_MAX*PIXEL_MAX * (double)i_size);
- if( f_mse <= 0.0000000001 ) /* Max 100dB */
+ double mse = sqe / (PIXEL_MAX*PIXEL_MAX * size);
+ if( mse <= 0.0000000001 ) /* Max 100dB */
return 100;
- return -10.0 * log10( f_mse );
+ return -10.0 * log10( mse );
}
-static float x264_ssim( float ssim )
+static double x264_ssim( double ssim )
{
- return -10.0 * log10( 1 - ssim );
+ double inv_ssim = 1 - ssim;
+ if( inv_ssim <= 0.0000000001 ) /* Max 100dB */
+ return 100;
+
+ return -10.0 * log10( inv_ssim );
}
static void x264_frame_dump( x264_t *h )
FILE *f = fopen( h->param.psz_dump_yuv, "r+b" );
if( !f )
return;
+
/* Write the frame in display order */
- fseek( f, (uint64_t)h->fdec->i_frame * h->param.i_height * h->param.i_width * 3/2 * sizeof(pixel), SEEK_SET );
- for( int y = 0; y < h->param.i_height; y++ )
- fwrite( &h->fdec->plane[0][y*h->fdec->i_stride[0]], sizeof(pixel), h->param.i_width, f );
- int cw = h->param.i_width>>1;
- int ch = h->param.i_height>>1;
- pixel *planeu = x264_malloc( cw*ch*2*sizeof(pixel) );
- pixel *planev = planeu + cw*ch;
- h->mc.plane_copy_deinterleave( planeu, cw, planev, cw, h->fdec->plane[1], h->fdec->i_stride[1], cw, ch );
- fwrite( planeu, 1, cw*ch*2*sizeof(pixel), f );
- x264_free( planeu );
+ int frame_size = FRAME_SIZE( h->param.i_height * h->param.i_width * sizeof(pixel) );
+ fseek( f, (uint64_t)h->fdec->i_frame * frame_size, SEEK_SET );
+ for( int p = 0; p < (CHROMA444 ? 3 : 1); p++ )
+ for( int y = 0; y < h->param.i_height; y++ )
+ fwrite( &h->fdec->plane[p][y*h->fdec->i_stride[p]], sizeof(pixel), h->param.i_width, f );
+ if( !CHROMA444 )
+ {
+ int cw = h->param.i_width>>1;
+ int ch = h->param.i_height>>CHROMA_V_SHIFT;
+ pixel *planeu = x264_malloc( (cw*ch*2+32)*sizeof(pixel) );
+ pixel *planev = planeu + cw*ch + 16;
+ h->mc.plane_copy_deinterleave( planeu, cw, planev, cw, h->fdec->plane[1], h->fdec->i_stride[1], cw, ch );
+ fwrite( planeu, 1, cw*ch*sizeof(pixel), f );
+ fwrite( planev, 1, cw*ch*sizeof(pixel), f );
+ x264_free( planeu );
+ }
fclose( f );
}
-
/* Fill "default" values */
static void x264_slice_header_init( x264_t *h, x264_slice_header_t *sh,
x264_sps_t *sps, x264_pps_t *pps,
sh->i_frame_num = i_frame;
- sh->b_mbaff = h->param.b_interlaced;
+ sh->b_mbaff = PARAM_INTERLACED;
sh->b_field_pic = 0; /* no field support for now */
sh->b_bottom_field = 0; /* not yet used */
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;
{
if( sh->b_mbaff )
{
- assert( sh->i_first_mb % (2*sh->sps->i_mb_width) == 0 );
- bs_write_ue( s, sh->i_first_mb >> 1 );
+ int first_x = sh->i_first_mb % sh->sps->i_mb_width;
+ int first_y = sh->i_first_mb / sh->sps->i_mb_width;
+ assert( (first_y&1) == 0 );
+ bs_write_ue( s, (2*first_x + sh->sps->i_mb_width*(first_y&~1) + (first_y&1)) >> 1 );
}
else
bs_write_ue( s, sh->i_first_mb );
if( sh->pps->b_pic_order && !sh->b_field_pic )
bs_write_se( s, sh->i_delta_poc_bottom );
}
- else if( sh->sps->i_poc_type == 1 && !sh->sps->b_delta_pic_order_always_zero )
- {
- bs_write_se( s, sh->i_delta_poc[0] );
- if( sh->pps->b_pic_order && !sh->b_field_pic )
- bs_write_se( s, sh->i_delta_poc[1] );
- }
if( sh->pps->b_redundant_pic_cnt )
bs_write_ue( s, sh->i_redundant_pic_cnt );
/* 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_bitstream_check_buffer( x264_t *h )
{
uint8_t *bs_bak = h->out.p_bitstream;
- if( (h->param.b_cabac && (h->cabac.p_end - h->cabac.p < 2500)) ||
- (h->out.bs.p_end - h->out.bs.p < 2500) )
+ int max_mb_size = 2500 << SLICE_MBAFF;
+ if( (h->param.b_cabac && (h->cabac.p_end - h->cabac.p < max_mb_size)) ||
+ (h->out.bs.p_end - h->out.bs.p < max_mb_size) )
{
h->out.i_bitstream += 100000;
CHECKED_MALLOC( h->out.p_bitstream, h->out.i_bitstream );
return -1;
}
-#if HAVE_PTHREAD
+#if HAVE_THREAD
static void x264_encoder_thread_init( x264_t *h )
{
if( h->param.i_sync_lookahead )
*
****************************************************************************/
-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_MMX2) )
{
x264_log( h, X264_LOG_ERROR, "your cpu does not support MMXEXT, but x264 was compiled with asm support\n");
#endif
return -1;
}
#endif
+
+#if HAVE_INTERLACED
+ h->param.b_interlaced = !!PARAM_INTERLACED;
+#else
+ if( h->param.b_interlaced )
+ {
+ x264_log( h, X264_LOG_ERROR, "not compiled with interlaced support\n" );
+ return -1;
+ }
+#endif
+
if( h->param.i_width <= 0 || h->param.i_height <= 0 )
{
x264_log( h, X264_LOG_ERROR, "invalid width x height (%dx%d)\n",
return -1;
}
- if( h->param.i_width % 2 || h->param.i_height % 2 )
+ int i_csp = h->param.i_csp & X264_CSP_MASK;
+#if X264_CHROMA_FORMAT
+ if( CHROMA_FORMAT != CHROMA_420 && i_csp >= X264_CSP_I420 && i_csp <= X264_CSP_NV12 )
{
- x264_log( h, X264_LOG_ERROR, "width or height not divisible by 2 (%dx%d)\n",
- h->param.i_width, h->param.i_height );
+ x264_log( h, X264_LOG_ERROR, "not compiled with 4:2:0 support\n" );
return -1;
}
- int i_csp = h->param.i_csp & X264_CSP_MASK;
+ else if( CHROMA_FORMAT != CHROMA_422 && i_csp >= X264_CSP_I422 && i_csp <= X264_CSP_NV16 )
+ {
+ x264_log( h, X264_LOG_ERROR, "not compiled with 4:2:2 support\n" );
+ return -1;
+ }
+ else if( CHROMA_FORMAT != CHROMA_444 && i_csp >= X264_CSP_I444 && i_csp <= X264_CSP_RGB )
+ {
+ x264_log( h, X264_LOG_ERROR, "not compiled with 4:4:4 support\n" );
+ return -1;
+ }
+#endif
if( i_csp <= X264_CSP_NONE || i_csp >= X264_CSP_MAX )
{
- x264_log( h, X264_LOG_ERROR, "invalid CSP (only I420/YV12/NV12 supported)\n" );
+ x264_log( h, X264_LOG_ERROR, "invalid CSP (only I420/YV12/NV12/I422/YV16/NV16/I444/YV24/BGR/BGRA/RGB supported)\n" );
+ return -1;
+ }
+
+ if( i_csp < X264_CSP_I444 && h->param.i_width % 2 )
+ {
+ x264_log( h, X264_LOG_ERROR, "width not divisible by 2 (%dx%d)\n",
+ h->param.i_width, h->param.i_height );
+ return -1;
+ }
+
+ if( i_csp < X264_CSP_I422 && PARAM_INTERLACED && h->param.i_height % 4 )
+ {
+ x264_log( h, X264_LOG_ERROR, "height not divisible by 4 (%dx%d)\n",
+ h->param.i_width, h->param.i_height );
+ return -1;
+ }
+
+ if( (i_csp < X264_CSP_I422 || PARAM_INTERLACED) && h->param.i_height % 2 )
+ {
+ x264_log( h, X264_LOG_ERROR, "height not divisible by 2 (%dx%d)\n",
+ h->param.i_width, h->param.i_height );
+ 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_threads = X264_MIN( h->param.i_threads, max_threads );
}
}
- else
+ h->param.i_threads = x264_clip3( h->param.i_threads, 1, X264_THREAD_MAX );
+ if( h->param.i_threads == 1 )
h->param.b_sliced_threads = 0;
h->i_thread_frames = h->param.b_sliced_threads ? 1 : h->param.i_threads;
if( h->i_thread_frames > 1 )
h->param.nalu_process = NULL;
- if( h->param.b_interlaced )
+ h->param.i_keyint_max = x264_clip3( h->param.i_keyint_max, 1, X264_KEYINT_MAX_INFINITE );
+ if( h->param.i_keyint_max == 1 )
{
- 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.b_intra_refresh = 0;
+ h->param.analyse.i_weighted_pred = 0;
}
+ 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;
if( score >= 5 )
{
x264_log( h, X264_LOG_ERROR, "broken ffmpeg default settings detected\n" );
- x264_log( h, X264_LOG_ERROR, "use an encoding preset (vpre)\n" );
+ x264_log( h, X264_LOG_ERROR, "use an encoding preset (e.g. -vpre medium)\n" );
+ x264_log( h, X264_LOG_ERROR, "preset usage: -vpre <speed> -vpre <profile>\n" );
+ x264_log( h, X264_LOG_ERROR, "speed presets are listed in x264 --help\n" );
+ x264_log( h, X264_LOG_ERROR, "profile is optional; x264 defaults to high\n" );
return -1;
}
}
x264_log( h, X264_LOG_ERROR, "no ratecontrol method specified\n" );
return -1;
}
- h->param.rc.f_rf_constant = x264_clip3f( h->param.rc.f_rf_constant, 0, QP_MAX );
+ 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, 11 );
+ 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;
+ h->param.rc.i_qp_constant = h->param.rc.f_rf_constant + QP_BD_OFFSET;
h->param.rc.i_bitrate = 0;
}
if( (h->param.rc.i_rc_method == X264_RC_CQP || h->param.rc.i_rc_method == X264_RC_CRF)
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 )
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 )
h->param.rc.i_vbv_max_bitrate = 0;
}
- if( h->param.b_interlaced && h->param.i_slice_max_size )
- {
- x264_log( h, X264_LOG_WARNING, "interlaced + slice-max-size is not implemented\n" );
- h->param.i_slice_max_size = 0;
- }
- if( h->param.b_interlaced && h->param.i_slice_max_mbs )
- {
- x264_log( h, X264_LOG_WARNING, "interlaced + slice-max-mbs is not implemented\n" );
- h->param.i_slice_max_mbs = 0;
- }
- int max_slices = (h->param.i_height+((16<<h->param.b_interlaced)-1))/(16<<h->param.b_interlaced);
+ 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 );
+
+ int max_slices = (h->param.i_height+((16<<PARAM_INTERLACED)-1))/(16<<PARAM_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 );
- h->param.i_keyint_max = x264_clip3( h->param.i_keyint_max, 1, X264_KEYINT_MAX_INFINITE );
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 );
- if( h->param.i_keyint_max == 1 )
- h->param.b_intra_refresh = 0;
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 )
h->param.rc.i_lookahead = X264_MIN( h->param.rc.i_lookahead, X264_MAX( h->param.i_keyint_max, bufsize*fps ) );
}
- if( !h->param.i_timebase_num || !h->param.i_timebase_den )
+ if( !h->param.i_timebase_num || !h->param.i_timebase_den || !(h->param.b_vfr_input || h->param.b_pulldown) )
{
h->param.i_timebase_num = h->param.i_fps_den;
h->param.i_timebase_den = h->param.i_fps_num;
}
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 );
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.inter &= ~X264_ANALYSE_I8x8;
h->param.analyse.intra &= ~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;
+ h->param.analyse.i_chroma_qp_offset = x264_clip3(h->param.analyse.i_chroma_qp_offset, -32, 32);
+ /* In 4:4:4 mode, chroma gets twice as much resolution, so we can halve its quality. */
+ if( b_open && i_csp >= X264_CSP_I444 && i_csp < X264_CSP_BGR && h->param.analyse.b_psy )
+ h->param.analyse.i_chroma_qp_offset += 6;
/* 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.rc.f_aq_strength = 0;
}
h->param.analyse.i_noise_reduction = x264_clip3( h->param.analyse.i_noise_reduction, 0, 1<<16 );
- if( h->param.analyse.i_subpel_refine == 10 && (h->param.analyse.i_trellis != 2 || !h->param.rc.i_aq_mode) )
+ if( h->param.analyse.i_subpel_refine >= 10 && (h->param.analyse.i_trellis != 2 || !h->param.rc.i_aq_mode) )
h->param.analyse.i_subpel_refine = 9;
{
int maxrate_bak = h->param.rc.i_vbv_max_bitrate;
if( h->param.rc.i_rc_method == X264_RC_ABR && h->param.rc.i_vbv_buffer_size <= 0 )
h->param.rc.i_vbv_max_bitrate = h->param.rc.i_bitrate * 2;
- h->sps = h->sps_array;
x264_sps_init( h->sps, h->param.i_sps_id, &h->param );
do h->param.i_level_idc = l->level_idc;
while( l[1].level_idc && x264_validate_levels( h, 0 ) && l++ );
}
}
if( h->param.analyse.i_mv_range <= 0 )
- h->param.analyse.i_mv_range = l->mv_range >> h->param.b_interlaced;
+ h->param.analyse.i_mv_range = l->mv_range >> PARAM_INTERLACED;
else
- h->param.analyse.i_mv_range = x264_clip3(h->param.analyse.i_mv_range, 32, 512 >> h->param.b_interlaced);
+ h->param.analyse.i_mv_range = x264_clip3(h->param.analyse.i_mv_range, 32, 512 >> PARAM_INTERLACED);
}
- h->param.analyse.i_weighted_pred = x264_clip3( h->param.analyse.i_weighted_pred, 0, 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 )
+ h->param.analyse.i_weighted_pred = x264_clip3( h->param.analyse.i_weighted_pred, X264_WEIGHTP_NONE, X264_WEIGHTP_SMART );
+
+ if( PARAM_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 )
+ if( PARAM_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 );
memcpy( h->pixf.mbcmp, satd ? h->pixf.satd : h->pixf.sad_aligned, sizeof(h->pixf.mbcmp) );
memcpy( h->pixf.mbcmp_unaligned, satd ? h->pixf.satd : h->pixf.sad, sizeof(h->pixf.mbcmp_unaligned) );
h->pixf.intra_mbcmp_x3_16x16 = satd ? h->pixf.intra_satd_x3_16x16 : h->pixf.intra_sad_x3_16x16;
- h->pixf.intra_mbcmp_x3_8x8c = satd ? h->pixf.intra_satd_x3_8x8c : h->pixf.intra_sad_x3_8x8c;
+ h->pixf.intra_mbcmp_x3_8x16c = satd ? h->pixf.intra_satd_x3_8x16c : h->pixf.intra_sad_x3_8x16c;
+ h->pixf.intra_mbcmp_x3_8x8c = satd ? h->pixf.intra_satd_x3_8x8c : h->pixf.intra_sad_x3_8x8c;
h->pixf.intra_mbcmp_x3_8x8 = satd ? h->pixf.intra_sa8d_x3_8x8 : h->pixf.intra_sad_x3_8x8;
h->pixf.intra_mbcmp_x3_4x4 = satd ? h->pixf.intra_satd_x3_4x4 : h->pixf.intra_sad_x3_4x4;
+ h->pixf.intra_mbcmp_x9_4x4 = h->param.b_cpu_independent || h->mb.b_lossless ? NULL
+ : satd ? h->pixf.intra_satd_x9_4x4 : h->pixf.intra_sad_x9_4x4;
+ h->pixf.intra_mbcmp_x9_8x8 = h->param.b_cpu_independent || h->mb.b_lossless ? NULL
+ : satd ? h->pixf.intra_sa8d_x9_8x8 : h->pixf.intra_sad_x9_8x8;
satd &= h->param.analyse.i_me_method == X264_ME_TESA;
memcpy( h->pixf.fpelcmp, satd ? h->pixf.satd : h->pixf.sad, sizeof(h->pixf.fpelcmp) );
memcpy( h->pixf.fpelcmp_x3, satd ? h->pixf.satd_x3 : h->pixf.sad_x3, sizeof(h->pixf.fpelcmp_x3) );
memcpy( h->pixf.fpelcmp_x4, satd ? h->pixf.satd_x4 : h->pixf.sad_x4, sizeof(h->pixf.fpelcmp_x4) );
}
+static void chroma_dsp_init( x264_t *h )
+{
+ memcpy( h->luma2chroma_pixel, x264_luma2chroma_pixel[CHROMA_FORMAT], sizeof(h->luma2chroma_pixel) );
+
+ switch( CHROMA_FORMAT )
+ {
+ case CHROMA_420:
+ memcpy( h->predict_chroma, h->predict_8x8c, sizeof(h->predict_chroma) );
+ h->mc.prefetch_fenc = h->mc.prefetch_fenc_420;
+ h->loopf.deblock_chroma[0] = h->loopf.deblock_h_chroma_420;
+ h->loopf.deblock_chroma_intra[0] = h->loopf.deblock_h_chroma_420_intra;
+ h->loopf.deblock_chroma_mbaff = h->loopf.deblock_chroma_420_mbaff;
+ h->loopf.deblock_chroma_intra_mbaff = h->loopf.deblock_chroma_420_intra_mbaff;
+ h->pixf.intra_mbcmp_x3_chroma = h->pixf.intra_mbcmp_x3_8x8c;
+ h->quantf.coeff_last[DCT_CHROMA_DC] = h->quantf.coeff_last4;
+ h->quantf.coeff_level_run[DCT_CHROMA_DC] = h->quantf.coeff_level_run4;
+ break;
+ case CHROMA_422:
+ memcpy( h->predict_chroma, h->predict_8x16c, sizeof(h->predict_chroma) );
+ h->mc.prefetch_fenc = h->mc.prefetch_fenc_422;
+ h->loopf.deblock_chroma[0] = h->loopf.deblock_h_chroma_422;
+ h->loopf.deblock_chroma_intra[0] = h->loopf.deblock_h_chroma_422_intra;
+ h->loopf.deblock_chroma_mbaff = h->loopf.deblock_chroma_422_mbaff;
+ h->loopf.deblock_chroma_intra_mbaff = h->loopf.deblock_chroma_422_intra_mbaff;
+ h->pixf.intra_mbcmp_x3_chroma = h->pixf.intra_mbcmp_x3_8x16c;
+ h->quantf.coeff_last[DCT_CHROMA_DC] = h->quantf.coeff_last8;
+ h->quantf.coeff_level_run[DCT_CHROMA_DC] = h->quantf.coeff_level_run8;
+ break;
+ case CHROMA_444:
+ h->mc.prefetch_fenc = h->mc.prefetch_fenc_422; /* FIXME: doesn't cover V plane */
+ h->loopf.deblock_chroma_mbaff = h->loopf.deblock_luma_mbaff;
+ h->loopf.deblock_chroma_intra_mbaff = h->loopf.deblock_luma_intra_mbaff;
+ break;
+ }
+}
+
static void x264_set_aspect_ratio( x264_t *h, x264_param_t *param, int initial )
{
/* VUI */
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_validate_parameters( h ) < 0 )
+ if( x264_threading_init() )
+ {
+ x264_log( h, X264_LOG_ERROR, "unable to initialize threading\n" );
+ goto fail;
+ }
+
+ 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->i_frame = -1;
h->i_frame_num = 0;
h->i_idr_pic_id = 0;
- uint64_t new_timebase_den = h->param.i_timebase_den;
- if( h->param.b_dts_compress )
- {
- /* h->i_dts_compress_multiplier == h->frames.i_bframe_delay + 1 */
- h->i_dts_compress_multiplier = h->param.i_bframe ? (h->param.i_bframe_pyramid ? 3 : 2) : 1;
- if( h->i_dts_compress_multiplier != 1 )
- {
- new_timebase_den = h->param.i_timebase_den * h->i_dts_compress_multiplier;
- x264_log( h, X264_LOG_DEBUG, "DTS compression changed timebase: %u/%u -> %u/%"PRIu64"\n",
- h->param.i_timebase_num, h->param.i_timebase_den,
- h->param.i_timebase_num, new_timebase_den );
- }
- }
- else
- h->i_dts_compress_multiplier = 1;
- if( new_timebase_den * 2 > UINT32_MAX )
+ if( (uint64_t)h->param.i_timebase_den * 2 > UINT32_MAX )
{
- x264_log( h, X264_LOG_ERROR, "Effective timebase denominator %"PRIu64" exceeds H.264 maximum\n", new_timebase_den );
+ x264_log( h, X264_LOG_ERROR, "Effective timebase denominator %u exceeds H.264 maximum\n", h->param.i_timebase_den );
goto fail;
}
- h->param.i_timebase_den = new_timebase_den;
- h->sps = &h->sps_array[0];
x264_sps_init( h->sps, h->param.i_sps_id, &h->param );
-
- 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_height = h->sps->i_mb_height;
h->mb.i_mb_count = h->mb.i_mb_width * h->mb.i_mb_height;
+ h->mb.chroma_h_shift = CHROMA_FORMAT == CHROMA_420 || CHROMA_FORMAT == CHROMA_422;
+ h->mb.chroma_v_shift = CHROMA_FORMAT == CHROMA_420;
+
+ /* 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 = PARAM_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 )
h->frames.i_delay = X264_MAX(h->param.i_bframe,3)*4;
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_cpb_delay = h->i_coded_fields = h->i_disp_fields = h->i_prev_duration = 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_rdo_init();
/* init CPU functions */
x264_predict_16x16_init( h->param.cpu, h->predict_16x16 );
x264_predict_8x8c_init( h->param.cpu, h->predict_8x8c );
+ x264_predict_8x16c_init( h->param.cpu, h->predict_8x16c );
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();
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 );
+ x264_zigzag_init( h->param.cpu, &h->zigzagf_progressive, &h->zigzagf_interlaced );
+ memcpy( &h->zigzagf, PARAM_INTERLACED ? &h->zigzagf_interlaced : &h->zigzagf_progressive, sizeof(h->zigzagf) );
x264_mc_init( h->param.cpu, &h->mc );
x264_quant_init( h, h->param.cpu, &h->quantf );
- x264_deblock_init( h->param.cpu, &h->loopf );
+ x264_deblock_init( h->param.cpu, &h->loopf, PARAM_INTERLACED );
x264_bitstream_init( h->param.cpu, &h->bsf );
- x264_dct_init_weights();
+ if( h->param.b_cabac )
+ x264_cabac_init( h );
+ else
+ x264_cavlc_init( h );
mbcmp_init( h );
+ chroma_dsp_init( h );
p = buf + sprintf( buf, "using cpu capabilities:" );
for( int i = 0; x264_cpu_names[i].flags; i++ )
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 ) )
fclose( f );
}
- const char *profile = h->sps->i_profile_idc == PROFILE_BASELINE ? "Baseline" :
+ const char *profile = h->sps->i_profile_idc == PROFILE_BASELINE ? "Constrained Baseline" :
h->sps->i_profile_idc == PROFILE_MAIN ? "Main" :
h->sps->i_profile_idc == PROFILE_HIGH ? "High" :
- h->sps->i_profile_idc == PROFILE_HIGH10 ? "High 10" :
- "High 4:4:4 Predictive";
+ h->sps->i_profile_idc == PROFILE_HIGH10 ? (h->sps->b_constraint_set3 == 1 ? "High 10 Intra" : "High 10") :
+ h->sps->i_profile_idc == PROFILE_HIGH422 ? (h->sps->b_constraint_set3 == 1 ? "High 4:2:2 Intra" : "High 4:2:2") :
+ h->sps->b_constraint_set3 == 1 ? "High 4:4:4 Intra" : "High 4:4:4 Predictive";
char level[4];
snprintf( level, sizeof(level), "%d.%d", h->sps->i_level_idc/10, h->sps->i_level_idc%10 );
- if( h->sps->i_level_idc == 9 || ( h->sps->i_level_idc == 11 && h->sps->b_constraint_set3 ) )
+ if( h->sps->i_level_idc == 9 || ( h->sps->i_level_idc == 11 && h->sps->b_constraint_set3 &&
+ (h->sps->i_profile_idc >= PROFILE_BASELINE && h->sps->i_profile_idc <= PROFILE_EXTENDED) ) )
strcpy( level, "1b" );
if( h->sps->i_profile_idc < PROFILE_HIGH10 )
}
else
{
- x264_log( h, X264_LOG_INFO, "profile %s, level %s, bit depth %d\n",
- profile, level, BIT_DEPTH );
+ static const char * const subsampling[4] = { "4:0:0", "4:2:0", "4:2:2", "4:4:4" };
+ x264_log( h, X264_LOG_INFO, "profile %s, level %s, %s %d-bit\n",
+ profile, level, subsampling[CHROMA_FORMAT], BIT_DEPTH );
}
return h;
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 &&
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
static int x264_nal_end( x264_t *h )
{
x264_nal_t *nal = &h->out.nal[h->out.i_nal];
- nal->i_payload = &h->out.p_bitstream[bs_pos( &h->out.bs ) / 8] - nal->p_payload;
+ uint8_t *end = &h->out.p_bitstream[bs_pos( &h->out.bs ) / 8];
+ nal->i_payload = end - nal->p_payload;
+ /* nal_escape_mmx reads past the end of the input.
+ * While undefined padding wouldn't actually affect the output, it makes valgrind unhappy. */
+ memset( end, 0xff, 32 );
if( h->param.nalu_process )
h->param.nalu_process( h, nal );
h->out.i_nal++;
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 )
/* generate picture parameters */
x264_nal_start( h, NAL_PPS, NAL_PRIORITY_HIGHEST );
- x264_pps_write( &h->out.bs, h->pps );
+ x264_pps_write( &h->out.bs, h->sps, h->pps );
if( x264_nal_end( h ) )
return -1;
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;
+ //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. */
+ if( BIT_DEPTH > 8 && w != x264_weight_none )
+ 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][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] << SLICE_MBAFF); i_ref++ )
for( int i = 0; i < 3; i++ )
h->sh.weight[i_ref][i].weightfn = NULL;
if( h->sh.i_type != SLICE_TYPE_P || h->param.analyse.i_weighted_pred <= 0 )
return;
- int i_padv = PADV << h->param.b_interlaced;
+ int i_padv = PADV << PARAM_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][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
+
+ /* 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 )
{
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;
{
SET_WEIGHT( h->fenc->weight[0][0], 1, 1, 0, h->fenc->weight[0][0].i_offset );
}
- x264_weighted_reference_duplicate( h, 0, weight_none );
+ x264_weighted_reference_duplicate( h, 0, x264_weight_none );
if( h->fenc->weight[0][0].i_offset > -128 )
{
w[0] = h->fenc->weight[0][0];
}
}
}
- 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 )
int b_deblock = h->sh.i_disable_deblocking_filter_idc != 1;
int b_end = mb_y == h->i_threadslice_end;
int b_measure_quality = 1;
- int min_y = mb_y - (1 << h->sh.b_mbaff);
+ int min_y = mb_y - (1 << SLICE_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 )
{
b_deblock = 0; /* We already deblocked on the inloop pass. */
b_measure_quality = 0; /* We already measured quality on the inloop pass. */
}
- if( mb_y & h->sh.b_mbaff )
+ if( mb_y & SLICE_MBAFF )
return;
if( min_y < h->i_threadslice_start )
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 << SLICE_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( PARAM_INTERLACED )
+ for( int p = 0; p < h->fdec->i_plane; p++ )
+ for( int i = minpix_y>>(CHROMA_V_SHIFT && p); i < maxpix_y>>(CHROMA_V_SHIFT && 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->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)) );
+ if( SLICE_MBAFF )
+ for( int i = 0; i < 3; 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] );
+ }
- 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( 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 << SLICE_MBAFF)) );
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 );
- uint64_t ssd_uv = 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 );
- h->stat.frame.i_ssd[0] += ssd_y;
- h->stat.frame.i_ssd[1] += (uint32_t)ssd_uv;
- h->stat.frame.i_ssd[2] += ssd_uv>>32;
+ for( int p = 0; p < (CHROMA444 ? 3 : 1); p++ )
+ h->stat.frame.i_ssd[p] += x264_pixel_ssd_wxh( &h->pixf,
+ h->fdec->plane[p] + minpix_y * h->fdec->i_stride[p], h->fdec->i_stride[p],
+ h->fenc->plane[p] + minpix_y * h->fenc->i_stride[p], h->fenc->i_stride[p],
+ h->param.i_width, maxpix_y-minpix_y );
+ if( !CHROMA444 )
+ {
+ uint64_t ssd_u, ssd_v;
+ int v_shift = CHROMA_V_SHIFT;
+ x264_pixel_ssd_nv12( &h->pixf,
+ h->fdec->plane[1] + (minpix_y>>v_shift) * h->fdec->i_stride[1], h->fdec->i_stride[1],
+ h->fenc->plane[1] + (minpix_y>>v_shift) * h->fenc->i_stride[1], h->fenc->i_stride[1],
+ h->param.i_width>>1, (maxpix_y-minpix_y)>>v_shift, &ssd_u, &ssd_v );
+ h->stat.frame.i_ssd[1] += ssd_u;
+ h->stat.frame.i_ssd[2] += ssd_v;
+ }
}
if( h->param.analyse.b_ssim )
{
+ int ssim_cnt;
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, &ssim_cnt );
+ h->stat.frame.i_ssim_cnt += ssim_cnt;
}
}
}
{
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 )
{
h->sh.i_poc = h->fdec->i_poc;
- if( h->param.b_interlaced )
+ if( PARAM_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;
- }
- else if( h->sps->i_poc_type == 1 )
- {
- /* FIXME TODO FIXME */
+ 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
{
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;
int tex_bits_bak = 0;
- /* Assume no more than 3 bytes of NALU escaping.
- * NALUs other than the first use a 3-byte startcode. */
- int overhead_guess = (NALU_OVERHEAD - (h->param.b_annexb && h->out.i_nal)) + 3;
- int slice_max_size = h->param.i_slice_max_size > 0 ? (h->param.i_slice_max_size-overhead_guess)*8 : INT_MAX;
+ /* NALUs other than the first use a 3-byte startcode.
+ * Add one extra byte for the rbsp, and one more for the final CABAC putbyte.
+ * Then add an extra 5 bytes just in case, to account for random NAL escapes and
+ * 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 );
bs_align_1( &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_context_init( h, &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;
+ h->mb.field_decoding_flag = 0;
i_mb_y = h->sh.i_first_mb / h->mb.i_mb_width;
i_mb_x = h->sh.i_first_mb % h->mb.i_mb_width;
i_skip = 0;
- while( (mb_xy = i_mb_x + i_mb_y * h->mb.i_mb_width) <= h->sh.i_last_mb )
+ while( 1 )
{
+ mb_xy = i_mb_x + i_mb_y * h->mb.i_mb_width;
int mb_spos = bs_pos(&h->out.bs) + x264_cabac_pos(&h->cabac);
- if( x264_bitstream_check_buffer( h ) )
- return -1;
-
- if( h->param.i_slice_max_size > 0 )
+ if( !(i_mb_y & SLICE_MBAFF) )
{
- mv_bits_bak = h->stat.frame.i_mv_bits;
- tex_bits_bak = h->stat.frame.i_tex_bits;
- /* We don't need the contexts because flushing the CABAC encoder has no context
- * dependency and macroblocks are only re-encoded in the case where a slice is
- * ended (and thus the content of all contexts are thrown away). */
- if( h->param.b_cabac )
- {
- memcpy( &cabac_bak, &h->cabac, offsetof(x264_cabac_t, f8_bits_encoded) );
- /* x264's CABAC writer modifies the previous byte during carry, so it has to be
- * backed up. */
- cabac_prevbyte_bak = h->cabac.p[-1];
- }
- else
+ if( x264_bitstream_check_buffer( h ) )
+ return -1;
+
+ if( back_up_bitstream )
{
- bs_bak = h->out.bs;
- i_skip_bak = i_skip;
+ mv_bits_bak = h->stat.frame.i_mv_bits;
+ tex_bits_bak = h->stat.frame.i_tex_bits;
+ /* We don't need the contexts because flushing the CABAC encoder has no context
+ * dependency and macroblocks are only re-encoded in the case where a slice is
+ * ended (and thus the content of all contexts are thrown away). */
+ if( h->param.b_cabac )
+ {
+ memcpy( &cabac_bak, &h->cabac, offsetof(x264_cabac_t, f8_bits_encoded) );
+ /* x264's CABAC writer modifies the previous byte during carry, so it has to be
+ * backed up. */
+ cabac_prevbyte_bak = h->cabac.p[-1];
+ }
+ else
+ {
+ bs_bak = h->out.bs;
+ i_skip_bak = i_skip;
+ }
}
}
if( i_mb_x == 0 && !h->mb.b_reencode_mb )
x264_fdec_filter_row( h, i_mb_y, 1 );
+ if( PARAM_INTERLACED )
+ {
+ if( h->mb.b_adaptive_mbaff )
+ {
+ if( !(i_mb_y&1) )
+ {
+ /* FIXME: VSAD is fast but fairly poor at choosing the best interlace type. */
+ h->mb.b_interlaced = x264_field_vsad( h, i_mb_x, i_mb_y );
+ memcpy( &h->zigzagf, MB_INTERLACED ? &h->zigzagf_interlaced : &h->zigzagf_progressive, sizeof(h->zigzagf) );
+ if( !MB_INTERLACED && (i_mb_y+2) == h->mb.i_mb_height )
+ x264_expand_border_mbpair( h, i_mb_x, i_mb_y );
+ }
+ }
+ h->mb.field[mb_xy] = MB_INTERLACED;
+ }
+
/* load cache */
- x264_macroblock_cache_load( h, i_mb_x, i_mb_y );
+ if( SLICE_MBAFF )
+ x264_macroblock_cache_load_interlaced( h, i_mb_x, i_mb_y );
+ else
+ x264_macroblock_cache_load_progressive( 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 )
{
- if( mb_xy > h->sh.i_first_mb && !(h->sh.b_mbaff && (i_mb_y&1)) )
+ if( mb_xy > h->sh.i_first_mb && !(SLICE_MBAFF && (i_mb_y&1)) )
x264_cabac_encode_terminal( &h->cabac );
if( IS_SKIP( h->mb.i_type ) )
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;
+ }
}
}
int total_bits = bs_pos(&h->out.bs) + x264_cabac_pos(&h->cabac);
int mb_size = total_bits - mb_spos;
- /* 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 )
+ if( slice_max_size )
{
- if( mb_xy != h->sh.i_first_mb )
+ /* Count the skip run, just in case. */
+ if( !h->param.b_cabac )
+ total_bits += bs_size_ue_big( i_skip );
+ /* 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 )
{
- h->stat.frame.i_mv_bits = mv_bits_bak;
- h->stat.frame.i_tex_bits = tex_bits_bak;
- if( h->param.b_cabac )
+ if( mb_xy != h->sh.i_first_mb )
{
- memcpy( &h->cabac, &cabac_bak, offsetof(x264_cabac_t, f8_bits_encoded) );
- h->cabac.p[-1] = cabac_prevbyte_bak;
+ h->stat.frame.i_mv_bits = mv_bits_bak;
+ h->stat.frame.i_tex_bits = tex_bits_bak;
+ if( h->param.b_cabac )
+ {
+ memcpy( &h->cabac, &cabac_bak, offsetof(x264_cabac_t, f8_bits_encoded) );
+ h->cabac.p[-1] = cabac_prevbyte_bak;
+ }
+ else
+ {
+ h->out.bs = bs_bak;
+ i_skip = i_skip_bak;
+ }
+ h->mb.b_reencode_mb = 1;
+ if( SLICE_MBAFF )
+ {
+ // set to bottom of previous mbpair
+ if( i_mb_x )
+ h->sh.i_last_mb = mb_xy-1+h->mb.i_mb_stride*(!(i_mb_y&1));
+ else
+ h->sh.i_last_mb = (i_mb_y-2+!(i_mb_y&1))*h->mb.i_mb_stride + h->mb.i_mb_width - 1;
+ }
+ else
+ h->sh.i_last_mb = mb_xy-1;
+ break;
}
else
{
- h->out.bs = bs_bak;
- i_skip = i_skip_bak;
+ h->sh.i_last_mb = mb_xy;
+ h->mb.b_reencode_mb = 0;
}
- h->mb.b_reencode_mb = 1;
- h->sh.i_last_mb = mb_xy-1;
- break;
}
else
- {
- h->sh.i_last_mb = mb_xy;
h->mb.b_reencode_mb = 0;
- }
}
- else
- h->mb.b_reencode_mb = 0;
#if HAVE_VISUALIZE
if( h->param.b_visualize )
h->stat.frame.i_mb_count[h->mb.i_type]++;
int b_intra = IS_INTRA( h->mb.i_type );
+ int b_skip = IS_SKIP( h->mb.i_type );
if( h->param.i_log_level >= X264_LOG_INFO || h->param.rc.b_stat_write )
{
- if( !b_intra && !IS_SKIP( h->mb.i_type ) && !IS_DIRECT( h->mb.i_type ) )
+ if( !b_intra && !b_skip && !IS_DIRECT( h->mb.i_type ) )
{
if( h->mb.i_partition != D_8x8 )
h->stat.frame.i_mb_partition[h->mb.i_partition] += 4;
{
if( h->mb.i_cbp_luma | h->mb.i_cbp_chroma )
{
- int cbpsum = (h->mb.i_cbp_luma&1) + ((h->mb.i_cbp_luma>>1)&1)
- + ((h->mb.i_cbp_luma>>2)&1) + (h->mb.i_cbp_luma>>3);
- h->stat.frame.i_mb_cbp[!b_intra + 0] += cbpsum;
- h->stat.frame.i_mb_cbp[!b_intra + 2] += !!h->mb.i_cbp_chroma;
- h->stat.frame.i_mb_cbp[!b_intra + 4] += h->mb.i_cbp_chroma >> 1;
+ if( CHROMA444 )
+ {
+ for( int i = 0; i < 4; i++ )
+ if( h->mb.i_cbp_luma & (1 << i) )
+ for( int p = 0; p < 3; p++ )
+ {
+ int s8 = i*4+p*16;
+ int nnz8x8 = M16( &h->mb.cache.non_zero_count[x264_scan8[s8]+0] )
+ | M16( &h->mb.cache.non_zero_count[x264_scan8[s8]+8] );
+ h->stat.frame.i_mb_cbp[!b_intra + p*2] += !!nnz8x8;
+ }
+ }
+ else
+ {
+ int cbpsum = (h->mb.i_cbp_luma&1) + ((h->mb.i_cbp_luma>>1)&1)
+ + ((h->mb.i_cbp_luma>>2)&1) + (h->mb.i_cbp_luma>>3);
+ h->stat.frame.i_mb_cbp[!b_intra + 0] += cbpsum;
+ h->stat.frame.i_mb_cbp[!b_intra + 2] += !!h->mb.i_cbp_chroma;
+ h->stat.frame.i_mb_cbp[!b_intra + 4] += h->mb.i_cbp_chroma >> 1;
+ }
}
if( h->mb.i_cbp_luma && !b_intra )
{
else //if( h->mb.i_type == I_4x4 )
for( int i = 0; i < 16; i++ )
h->stat.frame.i_mb_pred_mode[2][h->mb.cache.intra4x4_pred_mode[x264_scan8[i]]]++;
- h->stat.frame.i_mb_pred_mode[3][x264_mb_pred_mode8x8c_fix[h->mb.i_chroma_pred_mode]]++;
+ h->stat.frame.i_mb_pred_mode[3][x264_mb_chroma_pred_mode_fix[h->mb.i_chroma_pred_mode]]++;
}
+ h->stat.frame.i_mb_field[b_intra?0:b_skip?2:1] += MB_INTERLACED;
}
/* calculate deblock strength values (actual deblocking is done per-row along with hpel) */
if( b_deblock )
- {
- int mvy_limit = 4 >> h->sh.b_mbaff;
- uint8_t (*bs)[4][4] = h->deblock_strength[h->mb.i_mb_y&1][h->mb.i_mb_x];
- x264_macroblock_cache_load_deblock( h );
- if( IS_INTRA( h->mb.type[h->mb.i_mb_xy] ) )
- memset( bs, 3, 2*4*4*sizeof(uint8_t) );
- else
- h->loopf.deblock_strength( h->mb.cache.non_zero_count, h->mb.cache.ref, h->mb.cache.mv,
- bs, mvy_limit, h->sh.i_type == SLICE_TYPE_B );
- }
+ x264_macroblock_deblock_strength( h );
x264_ratecontrol_mb( h, mb_size );
- if( h->sh.b_mbaff )
+ if( mb_xy == h->sh.i_last_mb )
+ break;
+
+ if( SLICE_MBAFF )
{
i_mb_x += i_mb_y & 1;
i_mb_y ^= i_mb_x < h->mb.i_mb_width;
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 )
/* init stats */
memset( &h->stat.frame, 0, sizeof(h->stat.frame) );
h->mb.b_reencode_mb = 0;
- while( h->sh.i_first_mb <= last_thread_mb )
+ while( h->sh.i_first_mb + SLICE_MBAFF*h->mb.i_mb_stride <= last_thread_mb )
{
h->sh.i_last_mb = last_thread_mb;
if( h->param.i_slice_max_mbs )
- h->sh.i_last_mb = h->sh.i_first_mb + h->param.i_slice_max_mbs - 1;
+ {
+ if( SLICE_MBAFF )
+ {
+ // convert first to mbaff form, add slice-max-mbs, then convert back to normal form
+ int last_mbaff = 2*(h->sh.i_first_mb % h->mb.i_mb_width)
+ + h->mb.i_mb_width*(h->sh.i_first_mb / h->mb.i_mb_width)
+ + h->param.i_slice_max_mbs - 1;
+ int last_x = (last_mbaff % (2*h->mb.i_mb_width))/2;
+ int last_y = (last_mbaff / (2*h->mb.i_mb_width))*2 + 1;
+ h->sh.i_last_mb = last_x + h->mb.i_mb_stride*last_y;
+ }
+ else
+ h->sh.i_last_mb = h->sh.i_first_mb + h->param.i_slice_max_mbs - 1;
+ }
else if( h->param.i_slice_count && !h->param.b_sliced_threads )
{
- int height = h->mb.i_mb_height >> h->param.b_interlaced;
- int width = h->mb.i_mb_width << h->param.b_interlaced;
+ int height = h->mb.i_mb_height >> PARAM_INTERLACED;
+ int width = h->mb.i_mb_width << PARAM_INTERLACED;
i_slice_num++;
h->sh.i_last_mb = (height * i_slice_num + h->param.i_slice_count/2) / h->param.i_slice_count * width - 1;
}
if( x264_stack_align( x264_slice_write, h ) )
return (void *)-1;
h->sh.i_first_mb = h->sh.i_last_mb + 1;
+ // if i_first_mb is not the last mb in a row then go to the next mb in MBAFF order
+ if( SLICE_MBAFF && h->sh.i_first_mb % h->mb.i_mb_width )
+ h->sh.i_first_mb -= h->mb.i_mb_stride;
}
#if HAVE_VISUALIZE
t->param = h->param;
memcpy( &t->i_frame, &h->i_frame, offsetof(x264_t, rc) - offsetof(x264_t, i_frame) );
}
- int height = h->mb.i_mb_height >> h->param.b_interlaced;
- t->i_threadslice_start = ((height * i + h->param.i_slice_count/2) / h->param.i_threads) << h->param.b_interlaced;
- t->i_threadslice_end = ((height * (i+1) + h->param.i_slice_count/2) / h->param.i_threads) << h->param.b_interlaced;
+ int height = h->mb.i_mb_height >> PARAM_INTERLACED;
+ t->i_threadslice_start = ((height * i + h->param.i_slice_count/2) / h->param.i_threads) << PARAM_INTERLACED;
+ t->i_threadslice_end = ((height * (i+1) + h->param.i_slice_count/2) / h->param.i_threads) << PARAM_INTERLACED;
t->sh.i_first_mb = t->i_threadslice_start * h->mb.i_mb_width;
t->sh.i_last_mb = t->i_threadslice_end * h->mb.i_mb_width - 1;
}
for( int i = 1; i < h->param.i_threads; i++ )
{
x264_fdec_filter_row( h->thread[i], h->thread[i]->i_threadslice_start + 1, 0 );
- if( h->sh.b_mbaff )
+ if( SLICE_MBAFF )
x264_fdec_filter_row( h->thread[i], h->thread[i]->i_threadslice_start + 2, 0 );
}
for( int j = 0; j < 3; j++ )
h->stat.frame.i_ssd[j] += t->stat.frame.i_ssd[j];
h->stat.frame.f_ssim += t->stat.frame.f_ssim;
+ h->stat.frame.i_ssim_cnt += t->stat.frame.i_ssim_cnt;
}
return 0;
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
fenc->i_frame = h->frames.i_input++;
+ if( fenc->i_frame == 0 )
+ h->frames.i_first_pts = fenc->i_pts;
if( h->frames.i_bframe_delay && fenc->i_frame == h->frames.i_bframe_delay )
- h->frames.i_bframe_delay_time = fenc->i_pts;
+ h->frames.i_bframe_delay_time = fenc->i_pts - h->frames.i_first_pts;
if( h->param.b_vfr_input && fenc->i_pts <= h->frames.i_largest_pts )
x264_log( h, X264_LOG_WARNING, "non-strictly-monotonic PTS\n" );
if( fenc->i_pic_struct == PIC_STRUCT_AUTO )
{
+#if HAVE_INTERLACED
int b_interlaced = fenc->param ? fenc->param->b_interlaced : h->param.b_interlaced;
+#else
+ int b_interlaced = 0;
+#endif
if( b_interlaced )
{
int b_tff = fenc->param ? fenc->param->b_tff : h->param.b_tff;
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 )
h->fenc->b_kept_as_ref =
h->fdec->b_kept_as_ref = i_nal_ref_idc != NAL_PRIORITY_DISPOSABLE && h->param.i_keyint_max > 1;
- h->fdec->i_pts = h->fenc->i_pts *= h->i_dts_compress_multiplier;
+ h->fdec->i_pts = h->fenc->i_pts;
if( h->frames.i_bframe_delay )
{
int64_t *prev_reordered_pts = thread_current->frames.i_prev_reordered_pts;
- if( h->i_frame <= h->frames.i_bframe_delay )
- {
- if( h->i_dts_compress_multiplier == 1 )
- h->fdec->i_dts = h->fenc->i_reordered_pts - h->frames.i_bframe_delay_time;
- else
- {
- /* DTS compression */
- if( h->i_frame == 1 )
- thread_current->frames.i_init_delta = h->fenc->i_reordered_pts * h->i_dts_compress_multiplier;
- h->fdec->i_dts = h->i_frame * thread_current->frames.i_init_delta / h->i_dts_compress_multiplier;
- }
- }
- else
- h->fdec->i_dts = prev_reordered_pts[ (h->i_frame - h->frames.i_bframe_delay) % h->frames.i_bframe_delay ];
- prev_reordered_pts[ h->i_frame % h->frames.i_bframe_delay ] = h->fenc->i_reordered_pts * h->i_dts_compress_multiplier;
+ h->fdec->i_dts = h->i_frame > h->frames.i_bframe_delay
+ ? prev_reordered_pts[ (h->i_frame - h->frames.i_bframe_delay) % h->frames.i_bframe_delay ]
+ : h->fenc->i_reordered_pts - h->frames.i_bframe_delay_time;
+ prev_reordered_pts[ h->i_frame % h->frames.i_bframe_delay ] = h->fenc->i_reordered_pts;
}
else
h->fdec->i_dts = h->fenc->i_reordered_pts;
}
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) )
{
/* generate picture parameters */
x264_nal_start( h, NAL_PPS, NAL_PRIORITY_HIGHEST );
- x264_pps_write( &h->out.bs, h->pps );
+ x264_pps_write( &h->out.bs, h->sps, h->pps );
if( x264_nal_end( h ) )
return -1;
overhead += h->out.nal[h->out.i_nal-1].i_payload + NALU_OVERHEAD;
}
- /* buffering period sei is written in x264_encoder_frame_end */
+ /* when frame threading is used, buffering period sei is written in x264_encoder_frame_end */
+ if( h->i_thread_frames == 1 && h->sps->vui.b_nal_hrd_parameters_present )
+ {
+ x264_hrd_fullness( h );
+ x264_nal_start( h, NAL_SEI, NAL_PRIORITY_DISPOSABLE );
+ x264_sei_buffering_period_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;
+ }
+ }
+ /* write extra sei */
+ for( int i = 0; i < h->fenc->extra_sei.num_payloads; i++ )
+ {
+ x264_nal_start( h, NAL_SEI, NAL_PRIORITY_DISPOSABLE );
+ x264_sei_write( &h->out.bs, h->fenc->extra_sei.payloads[i].payload, h->fenc->extra_sei.payloads[i].payload_size,
+ h->fenc->extra_sei.payloads[i].payload_type );
+ 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->extra_sei.sei_free && h->fenc->extra_sei.payloads[i].payload )
+ h->fenc->extra_sei.sei_free( h->fenc->extra_sei.payloads[i].payload );
+ }
+
+ if( h->fenc->extra_sei.sei_free && h->fenc->extra_sei.payloads )
+ h->fenc->extra_sei.sei_free( h->fenc->extra_sei.payloads );
+
+ if( h->fenc->b_keyframe )
+ {
if( h->param.b_repeat_headers && h->fenc->i_frame == 0 )
{
/* identify ourself */
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;
+ 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 );
- 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);
}
}
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;
+
/* Init the rate control */
/* FIXME: Include slice header bit cost. */
x264_ratecontrol_start( h, h->fenc->i_qpplus1, overhead*8 );
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++;
}
x264_emms();
- /* generate sei buffering period and insert it into place */
- if( h->fenc->b_keyframe && h->sps->vui.b_nal_hrd_parameters_present )
+ /* generate buffering period sei and insert it into place */
+ if( h->i_thread_frames > 1 && h->fenc->b_keyframe && h->sps->vui.b_nal_hrd_parameters_present )
{
x264_hrd_fullness( h );
x264_nal_start( h, NAL_SEI, NAL_PRIORITY_DISPOSABLE );
}
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( pic_out->i_pts < pic_out->i_dts )
x264_log( h, X264_LOG_WARNING, "invalid DTS: PTS is less than DTS\n" );
- pic_out->img.i_csp = X264_CSP_NV12;
+ pic_out->opaque = h->fenc->opaque;
+
+ pic_out->img.i_csp = h->fdec->i_csp;
+#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 = 0; i < 2; i++ )
+ for( int i = 0; i < pic_out->img.i_plane; i++ )
{
- pic_out->img.i_stride[i] = h->fdec->i_stride[i];
- // FIXME This breaks the API when pixel != uint8_t.
- pic_out->img.plane[i] = h->fdec->plane[i];
+ pic_out->img.i_stride[i] = h->fdec->i_stride[i] * sizeof(pixel);
+ pic_out->img.plane[i] = (uint8_t*)h->fdec->plane[i];
}
x264_frame_push_unused( thread_current, h->fenc );
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 )
+ for( int i = 0; i < 3; i++ )
+ h->stat.i_mb_field[i] += h->stat.frame.i_mb_field[i];
+ 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;
+ h->stat.f_frame_duration[h->sh.i_type] += dur;
if( h->param.analyse.b_psnr )
{
- int64_t ssd[3] = {
+ int64_t ssd[3] =
+ {
h->stat.frame.i_ssd[0],
h->stat.frame.i_ssd[1],
h->stat.frame.i_ssd[2],
};
+ int luma_size = h->param.i_width * h->param.i_height;
+ int chroma_size = CHROMA_SIZE( luma_size );
+ double psnr_y = x264_psnr( ssd[0], luma_size );
+ double psnr_u = x264_psnr( ssd[1], chroma_size );
+ double psnr_v = x264_psnr( ssd[2], chroma_size );
- h->stat.i_ssd_global[h->sh.i_type] += ssd[0] + ssd[1] + ssd[2];
- h->stat.f_psnr_average[h->sh.i_type] += x264_psnr( ssd[0] + ssd[1] + ssd[2], 3 * h->param.i_width * h->param.i_height / 2 );
- h->stat.f_psnr_mean_y[h->sh.i_type] += x264_psnr( ssd[0], h->param.i_width * h->param.i_height );
- h->stat.f_psnr_mean_u[h->sh.i_type] += x264_psnr( ssd[1], h->param.i_width * h->param.i_height / 4 );
- h->stat.f_psnr_mean_v[h->sh.i_type] += x264_psnr( ssd[2], h->param.i_width * h->param.i_height / 4 );
+ h->stat.f_ssd_global[h->sh.i_type] += dur * (ssd[0] + ssd[1] + ssd[2]);
+ h->stat.f_psnr_average[h->sh.i_type] += dur * x264_psnr( ssd[0] + ssd[1] + ssd[2], luma_size + chroma_size*2 );
+ h->stat.f_psnr_mean_y[h->sh.i_type] += dur * psnr_y;
+ h->stat.f_psnr_mean_u[h->sh.i_type] += dur * psnr_u;
+ h->stat.f_psnr_mean_v[h->sh.i_type] += dur * psnr_v;
- snprintf( psz_message, 80, " PSNR Y:%5.2f U:%5.2f V:%5.2f",
- x264_psnr( ssd[0], h->param.i_width * h->param.i_height ),
- x264_psnr( ssd[1], h->param.i_width * h->param.i_height / 4),
- x264_psnr( ssd[2], h->param.i_width * h->param.i_height / 4) );
+ snprintf( psz_message, 80, " PSNR Y:%5.2f U:%5.2f V:%5.2f", psnr_y, psnr_u, psnr_v );
}
if( h->param.analyse.b_ssim )
{
double ssim_y = h->stat.frame.f_ssim
- / (((h->param.i_width-6)>>2) * ((h->param.i_height-6)>>2));
- h->stat.f_ssim_mean_y[h->sh.i_type] += ssim_y;
+ / h->stat.frame.i_ssim_cnt;
+ h->stat.f_ssim_mean_y[h->sh.i_type] += ssim_y * dur;
snprintf( psz_message + strlen(psz_message), 80 - strlen(psz_message),
" SSIM Y:%.5f", ssim_y );
}
/* 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 )
- {
- x264_frame_push_blank_unused( h, h->fref0[i] );
- h->fref0[i] = 0;
- }
+ 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->fref[0][i] );
+ h->fref[0][i] = 0;
+ }
if( h->param.psz_dump_yuv )
x264_frame_dump( h );
+ x264_emms();
return frame_size;
}
****************************************************************************/
void x264_encoder_close ( x264_t *h )
{
- int64_t i_yuv_size = 3 * h->param.i_width * h->param.i_height / 2;
+ int64_t i_yuv_size = FRAME_SIZE( h->param.i_width * h->param.i_height );
int64_t i_mb_count_size[2][7] = {{0}};
char buf[200];
int b_print_pcm = h->stat.i_mb_count[SLICE_TYPE_I][I_PCM]
if( h->stat.i_frame_count[i_slice] > 0 )
{
- const int i_count = h->stat.i_frame_count[i_slice];
+ int i_count = h->stat.i_frame_count[i_slice];
+ double dur = h->stat.f_frame_duration[i_slice];
if( h->param.analyse.b_psnr )
{
x264_log( h, X264_LOG_INFO,
i_count,
h->stat.f_frame_qp[i_slice] / i_count,
(double)h->stat.i_frame_size[i_slice] / i_count,
- h->stat.f_psnr_mean_y[i_slice] / i_count, h->stat.f_psnr_mean_u[i_slice] / i_count, h->stat.f_psnr_mean_v[i_slice] / i_count,
- h->stat.f_psnr_average[i_slice] / i_count,
- x264_psnr( h->stat.i_ssd_global[i_slice], i_count * i_yuv_size ) );
+ h->stat.f_psnr_mean_y[i_slice] / dur, h->stat.f_psnr_mean_u[i_slice] / dur, h->stat.f_psnr_mean_v[i_slice] / dur,
+ h->stat.f_psnr_average[i_slice] / dur,
+ x264_psnr( h->stat.f_ssd_global[i_slice], dur * i_yuv_size ) );
}
else
{
}
}
}
- 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++ )
list_count[2] += h->stat.i_mb_partition[SLICE_TYPE_B][D_BI_8x8];
i_mb_count[B_DIRECT] += (h->stat.i_mb_partition[SLICE_TYPE_B][D_DIRECT_8x8]+2)/4;
i_mb_list_count = (list_count[0] + list_count[1] + list_count[2]) / 100.0;
- x264_log( h, X264_LOG_INFO,
- "mb B %s B16..8: %4.1f%% %4.1f%% %4.1f%% direct:%4.1f%% skip:%4.1f%% L0:%4.1f%% L1:%4.1f%% BI:%4.1f%%\n",
- buf,
- i_mb_size[PIXEL_16x16] / (i_count*4),
- (i_mb_size[PIXEL_16x8] + i_mb_size[PIXEL_8x16]) / (i_count*4),
- i_mb_size[PIXEL_8x8] / (i_count*4),
- i_mb_count[B_DIRECT] / i_count,
- i_mb_count[B_SKIP] / i_count,
- list_count[0] / i_mb_list_count,
- list_count[1] / i_mb_list_count,
- list_count[2] / i_mb_list_count );
+ sprintf( buf + strlen(buf), " B16..8: %4.1f%% %4.1f%% %4.1f%% direct:%4.1f%% skip:%4.1f%%",
+ i_mb_size[PIXEL_16x16] / (i_count*4),
+ (i_mb_size[PIXEL_16x8] + i_mb_size[PIXEL_8x16]) / (i_count*4),
+ i_mb_size[PIXEL_8x8] / (i_count*4),
+ i_mb_count[B_DIRECT] / i_count,
+ i_mb_count[B_SKIP] / i_count );
+ if( i_mb_list_count != 0 )
+ sprintf( buf + strlen(buf), " L0:%4.1f%% L1:%4.1f%% BI:%4.1f%%",
+ list_count[0] / i_mb_list_count,
+ list_count[1] / i_mb_list_count,
+ list_count[2] / i_mb_list_count );
+ x264_log( h, X264_LOG_INFO, "mb B %s\n", buf );
}
x264_ratecontrol_summary( h );
int64_t i_intra = i_i8x8 + SUM3b( h->stat.i_mb_count, I_4x4 )
+ SUM3b( h->stat.i_mb_count, I_16x16 );
int64_t i_all_intra = i_intra + SUM3b( h->stat.i_mb_count, I_PCM);
+ int64_t i_skip = SUM3b( h->stat.i_mb_count, P_SKIP )
+ + SUM3b( h->stat.i_mb_count, B_SKIP );
const int i_count = h->stat.i_frame_count[SLICE_TYPE_I] +
h->stat.i_frame_count[SLICE_TYPE_P] +
h->stat.i_frame_count[SLICE_TYPE_B];
int64_t i_mb_count = (int64_t)i_count * h->mb.i_mb_count;
- float fps = (float) h->param.i_fps_num / h->param.i_fps_den;
- float f_bitrate;
- /* duration algorithm fails with one frame */
- if( !h->param.b_vfr_input || i_count == 1 )
- f_bitrate = fps * SUM3(h->stat.i_frame_size) / i_count / 125;
- else
+ int64_t i_inter = i_mb_count - i_skip - i_intra;
+ const double duration = h->stat.f_frame_duration[SLICE_TYPE_I] +
+ h->stat.f_frame_duration[SLICE_TYPE_P] +
+ h->stat.f_frame_duration[SLICE_TYPE_B];
+ float f_bitrate = SUM3(h->stat.i_frame_size) / duration / 125;
+
+ if( PARAM_INTERLACED )
{
- float duration = (float)(2 * h->frames.i_largest_pts - h->frames.i_second_largest_pts) * h->param.i_timebase_num / h->param.i_timebase_den;
- f_bitrate = SUM3(h->stat.i_frame_size) / duration / 125;
+ char *fieldstats = buf;
+ fieldstats[0] = 0;
+ if( i_inter )
+ fieldstats += sprintf( fieldstats, " inter:%.1f%%", h->stat.i_mb_field[1] * 100.0 / i_inter );
+ if( i_skip )
+ fieldstats += sprintf( fieldstats, " skip:%.1f%%", h->stat.i_mb_field[2] * 100.0 / i_skip );
+ x264_log( h, X264_LOG_INFO, "field mbs: intra: %.1f%%%s\n",
+ h->stat.i_mb_field[0] * 100.0 / i_intra, buf );
}
if( h->pps->b_transform_8x8_mode )
}
buf[0] = 0;
+ int csize = CHROMA444 ? 4 : 1;
if( i_mb_count != i_all_intra )
sprintf( buf, " inter: %.1f%% %.1f%% %.1f%%",
h->stat.i_mb_cbp[1] * 100.0 / ((i_mb_count - i_all_intra)*4),
- h->stat.i_mb_cbp[3] * 100.0 / ((i_mb_count - i_all_intra) ),
- h->stat.i_mb_cbp[5] * 100.0 / ((i_mb_count - i_all_intra)) );
- x264_log( h, X264_LOG_INFO, "coded y,uvDC,uvAC intra: %.1f%% %.1f%% %.1f%%%s\n",
+ h->stat.i_mb_cbp[3] * 100.0 / ((i_mb_count - i_all_intra)*csize),
+ h->stat.i_mb_cbp[5] * 100.0 / ((i_mb_count - i_all_intra)*csize) );
+ x264_log( h, X264_LOG_INFO, "coded y,%s,%s intra: %.1f%% %.1f%% %.1f%%%s\n",
+ CHROMA444?"u":"uvDC", CHROMA444?"v":"uvAC",
h->stat.i_mb_cbp[0] * 100.0 / (i_all_intra*4),
- h->stat.i_mb_cbp[2] * 100.0 / (i_all_intra ),
- h->stat.i_mb_cbp[4] * 100.0 / (i_all_intra ), buf );
+ h->stat.i_mb_cbp[2] * 100.0 / (i_all_intra*csize),
+ h->stat.i_mb_cbp[4] * 100.0 / (i_all_intra*csize), buf );
int64_t fixed_pred_modes[4][9] = {{0}};
int64_t sum_pred_modes[4] = {0};
}
for( int i = 0; i <= I_PRED_CHROMA_DC_128; i++ )
{
- fixed_pred_modes[3][x264_mb_pred_mode8x8c_fix[i]] += h->stat.i_mb_pred_mode[3][i];
+ fixed_pred_modes[3][x264_mb_chroma_pred_mode_fix[i]] += h->stat.i_mb_pred_mode[3][i];
sum_pred_modes[3] += h->stat.i_mb_pred_mode[3][i];
}
- if( sum_pred_modes[3] )
+ if( sum_pred_modes[3] && !CHROMA444 )
x264_log( h, X264_LOG_INFO, "i8c dc,h,v,p: %2.0f%% %2.0f%% %2.0f%% %2.0f%%\n",
fixed_pred_modes[3][0] * 100.0 / sum_pred_modes[3],
fixed_pred_modes[3][1] * 100.0 / sum_pred_modes[3],
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++ )
if( h->param.analyse.b_ssim )
{
- float ssim = SUM3( h->stat.f_ssim_mean_y ) / i_count;
+ float ssim = SUM3( h->stat.f_ssim_mean_y ) / duration;
x264_log( h, X264_LOG_INFO, "SSIM Mean Y:%.7f (%6.3fdb)\n", ssim, x264_ssim( ssim ) );
}
if( h->param.analyse.b_psnr )
{
x264_log( h, X264_LOG_INFO,
"PSNR Mean Y:%6.3f U:%6.3f V:%6.3f Avg:%6.3f Global:%6.3f kb/s:%.2f\n",
- SUM3( h->stat.f_psnr_mean_y ) / i_count,
- SUM3( h->stat.f_psnr_mean_u ) / i_count,
- SUM3( h->stat.f_psnr_mean_v ) / i_count,
- SUM3( h->stat.f_psnr_average ) / i_count,
- x264_psnr( SUM3( h->stat.i_ssd_global ), i_count * i_yuv_size ),
+ SUM3( h->stat.f_psnr_mean_y ) / duration,
+ SUM3( h->stat.f_psnr_mean_u ) / duration,
+ SUM3( h->stat.f_psnr_mean_v ) / duration,
+ SUM3( h->stat.f_psnr_average ) / duration,
+ x264_psnr( SUM3( h->stat.f_ssd_global ), duration * i_yuv_size ),
f_bitrate );
}
else
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;
+}