h->mb.i_mb_width = h->sps->i_mb_width;
h->mb.i_mb_height = h->sps->i_mb_height;
h->mb.i_mb_count = h->mb.i_mb_width * h->mb.i_mb_height;
+ /* Adaptive MBAFF and subme 0 are not supported as we require halving motion
+ * vectors during prediction, resulting in hpel mvs.
+ * The chosen solution is to make MBAFF non-adaptive in this case. */
+ h->mb.b_adaptive_mbaff = h->param.b_interlaced && h->param.analyse.i_subpel_refine;
/* Init frames. */
if( h->param.i_bframe_adaptive == X264_B_ADAPT_TRELLIS && !h->param.rc.b_stat_read )
* ( 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 ) )
nal_size += h->out.nal[i].i_payload;
/* Worst-case NAL unit escaping: reallocate the buffer if it's too small. */
- if( h->nal_buffer_size < nal_size * 3/2 + h->out.i_nal * 4 )
+ int necessary_size = nal_size * 3/2 + h->out.i_nal * 4;
+ if( h->nal_buffer_size < necessary_size )
{
- uint8_t *buf = x264_malloc( nal_size * 2 + h->out.i_nal * 4 );
+ h->nal_buffer_size = necessary_size * 2;
+ uint8_t *buf = x264_malloc( h->nal_buffer_size );
if( !buf )
return -1;
if( previous_nal_size )
return -1;
frame_size = x264_encoder_encapsulate_nals( h, 0 );
+ if( frame_size < 0 )
+ return -1;
/* now set output*/
*pi_nal = h->out.i_nal;
int b_measure_quality = 1;
int min_y = mb_y - (1 << h->sh.b_mbaff);
int b_start = min_y == h->i_threadslice_start;
- int max_y = b_end ? h->i_threadslice_end : mb_y;
+ /* Even in interlaced mode, deblocking never modifies more than 4 pixels
+ * above each MB, as bS=4 doesn't happen for the top of interlaced mbpairs. */
+ int minpix_y = min_y*16 - 4 * !b_start;
+ int maxpix_y = mb_y*16 - 4 * !b_end;
b_deblock &= b_hpel || h->param.psz_dump_yuv;
if( h->param.b_sliced_threads && b_start && min_y && !b_inloop )
{
return;
if( b_deblock )
- for( int y = min_y; y < max_y; y += (1 << h->sh.b_mbaff) )
+ for( int y = min_y; y < mb_y; y += (1 << h->sh.b_mbaff) )
x264_frame_deblock_row( h, y );
+ /* FIXME: Prediction requires different borders for interlaced/progressive mc,
+ * but the actual image data is equivalent. For now, maintain this
+ * consistency by copying deblocked pixels between planes. */
+ if( h->param.b_interlaced )
+ for( int p = 0; p < 2; p++ )
+ for( int i = minpix_y>>p; i < maxpix_y>>p; i++ )
+ memcpy( h->fdec->plane_fld[p] + i*h->fdec->i_stride[p],
+ h->fdec->plane[p] + i*h->fdec->i_stride[p],
+ h->mb.i_mb_width*16*sizeof(pixel) );
+
if( b_hpel )
{
int end = mb_y == h->mb.i_mb_height;
}
}
+ if( h->sh.b_mbaff )
+ for( int i = 0; i < 2; i++ )
+ {
+ XCHG( pixel *, h->intra_border_backup[0][i], h->intra_border_backup[3][i] );
+ XCHG( pixel *, h->intra_border_backup[1][i], h->intra_border_backup[4][i] );
+ }
+
if( h->i_thread_frames > 1 && h->fdec->b_kept_as_ref )
x264_frame_cond_broadcast( h->fdec, mb_y*16 + (b_end ? 10000 : -(X264_THREAD_HEIGHT << h->sh.b_mbaff)) );
- min_y = min_y*16 - 8 * !b_start;
- max_y = b_end ? X264_MIN( h->i_threadslice_end*16 , h->param.i_height ) : mb_y*16 - 8;
-
if( b_measure_quality )
{
+ maxpix_y = X264_MIN( maxpix_y, h->param.i_height );
if( h->param.analyse.b_psnr )
{
uint64_t ssd_y = x264_pixel_ssd_wxh( &h->pixf,
- h->fdec->plane[0] + min_y * h->fdec->i_stride[0], h->fdec->i_stride[0],
- h->fenc->plane[0] + min_y * h->fenc->i_stride[0], h->fenc->i_stride[0],
- h->param.i_width, max_y-min_y );
+ h->fdec->plane[0] + minpix_y * h->fdec->i_stride[0], h->fdec->i_stride[0],
+ h->fenc->plane[0] + minpix_y * h->fenc->i_stride[0], h->fenc->i_stride[0],
+ h->param.i_width, maxpix_y-minpix_y );
uint64_t ssd_u, ssd_v;
x264_pixel_ssd_nv12( &h->pixf,
- h->fdec->plane[1] + (min_y>>1) * h->fdec->i_stride[1], h->fdec->i_stride[1],
- h->fenc->plane[1] + (min_y>>1) * h->fenc->i_stride[1], h->fenc->i_stride[1],
- h->param.i_width>>1, (max_y-min_y)>>1, &ssd_u, &ssd_v );
+ h->fdec->plane[1] + (minpix_y>>1) * h->fdec->i_stride[1], h->fdec->i_stride[1],
+ h->fenc->plane[1] + (minpix_y>>1) * h->fenc->i_stride[1], h->fenc->i_stride[1],
+ h->param.i_width>>1, (maxpix_y-minpix_y)>>1, &ssd_u, &ssd_v );
h->stat.frame.i_ssd[0] += ssd_y;
h->stat.frame.i_ssd[1] += ssd_u;
h->stat.frame.i_ssd[2] += ssd_v;
x264_emms();
/* offset by 2 pixels to avoid alignment of ssim blocks with dct blocks,
* and overlap by 4 */
- min_y += b_start ? 2 : -6;
+ minpix_y += b_start ? 2 : -6;
h->stat.frame.f_ssim +=
x264_pixel_ssim_wxh( &h->pixf,
- h->fdec->plane[0] + 2+min_y*h->fdec->i_stride[0], h->fdec->i_stride[0],
- h->fenc->plane[0] + 2+min_y*h->fenc->i_stride[0], h->fenc->i_stride[0],
- h->param.i_width-2, max_y-min_y, h->scratch_buffer );
+ h->fdec->plane[0] + 2+minpix_y*h->fdec->i_stride[0], h->fdec->i_stride[0],
+ h->fenc->plane[0] + 2+minpix_y*h->fenc->i_stride[0], h->fenc->i_stride[0],
+ h->param.i_width-2, maxpix_y-minpix_y, h->scratch_buffer );
}
}
}
if( i_mb_x == 0 && !h->mb.b_reencode_mb )
x264_fdec_filter_row( h, i_mb_y, 1 );
+ if( h->param.b_interlaced )
+ {
+ if( h->mb.b_adaptive_mbaff )
+ {
+ if( !(i_mb_y&1) )
+ h->mb.b_interlaced = 1;
+ x264_zigzag_init( h->param.cpu, &h->zigzagf, h->mb.b_interlaced );
+ }
+ h->mb.field[mb_xy] = h->mb.b_interlaced;
+ }
+
/* load cache */
x264_macroblock_cache_load( h, i_mb_x, i_mb_y );
}
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;
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;
}
projects / x264 / blobdiff