[x264-devel] x264_encoder_encode too slow
mm1201 at mail.ustc.edu.cn
mm1201 at mail.ustc.edu.cn
Thu Nov 13 15:42:22 CET 2014
It works,thank you a lot,interesting!
> -----原始邮件-----
> 发件人: x264-devel-request at videolan.org
> 发送时间: 2014-11-13 20:52:05 (星期四)
> 收件人: x264-devel at videolan.org
> 抄送:
> 主题: x264-devel Digest, Vol 90, Issue 8
>
> Send x264-devel mailing list submissions to
> x264-devel at videolan.org
>
> To subscribe or unsubscribe via the World Wide Web, visit
> https://mailman.videolan.org/listinfo/x264-devel
> or, via email, send a message with subject or body 'help' to
> x264-devel-request at videolan.org
>
> You can reach the person managing the list at
> x264-devel-owner at videolan.org
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of x264-devel digest..."
>
>
> Today's Topics:
>
> 1. Re: x264_encoder_encode too slow (Daniel Tisch)
> 2. Fix VBV (Anton Mitrofanov)
> 3. Fix VBV with true VFR streams (Anton Mitrofanov)
> 4. Fix inappropriate instruction use (Anton Mitrofanov)
> 5. x264asm: warn when inappropriate instruction used in function
> with specified cpuflags (Anton Mitrofanov)
> 6. Fix few clang -Wunused-* warnings (Anton Mitrofanov)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Thu, 13 Nov 2014 12:45:31 +0100
> From: Daniel Tisch <tisch.daniel at gmail.com>
> To: Mailing list for x264 developers <x264-devel at videolan.org>
> Subject: Re: [x264-devel] x264_encoder_encode too slow
> Message-ID:
> <CA+mv---31KCoT6nh3y6C0f2uftQuK6qjkrEuifkMsCZmbUct0w at mail.gmail.com>
> Content-Type: text/plain; charset="utf-8"
>
> H.264 encoding can be *VERY* slow depending on your settings, source
> resolution and CPU. Try using presets "ultrafast", "superfast", "veryfast",
> "faster" or "fast" (in decreasing encoding speed) with
> x264_param_default_preset (instead of x264_param_default)!
>
>
> Good luck! ;)
>
> Cheers,
>
> Daniel
>
>
> 2014.11.13. 9:27 ezt ?rta ( <mm1201 at mail.ustc.edu.cn>):
>
> > i call x264_encoder_encoed() in my code,and it works,but the problem is
> > that it runs slow,costs about 2 seconds,i don't know why.Is there anynoe
> > who ever ran into this,or knows what could cause this?
> > _______________________________________________
> > x264-devel mailing list
> > x264-devel at videolan.org
> > https://mailman.videolan.org/listinfo/x264-devel
> >
> >
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <http://mailman.videolan.org/pipermail/x264-devel/attachments/20141113/8577e375/attachment-0001.html>
>
> ------------------------------
>
> Message: 2
> Date: Thu, 13 Nov 2014 13:52:02 +0100 (CET)
> From: git at videolan.org (Anton Mitrofanov)
> To: x264-devel at videolan.org
> Subject: [x264-devel] Fix VBV
> Message-ID: <20141113125202.C8FC9151F6F at albiero.videolan.org>
> Content-Type: text/plain; charset=UTF-8
>
> x264 | branch: master | Anton Mitrofanov <BugMaster at narod.ru> | Mon Sep 1 22:45:00 2014 +0400| [b36d44c68cddff00c5b6de1e6cb6a86c1af2cbfc] | committer: Fiona Glaser
>
> Fix VBV
>
> > http://git.videolan.org/gitweb.cgi/x264.git/?a=commit;h=b36d44c68cddff00c5b6de1e6cb6a86c1af2cbfc
> ---
>
> encoder/ratecontrol.c | 27 ++++++++++++++++-----------
> 1 file changed, 16 insertions(+), 11 deletions(-)
>
> diff --git a/encoder/ratecontrol.c b/encoder/ratecontrol.c
> index f71eb5f..8893c1f 100644
> --- a/encoder/ratecontrol.c
> +++ b/encoder/ratecontrol.c
> @@ -2255,32 +2255,31 @@ static double clip_qscale( x264_t *h, int pict_type, double q )
> /* Now a hard threshold to make sure the frame fits in VBV.
> * This one is mostly for I-frames. */
> double bits = predict_size( &rcc->pred[h->sh.i_type], q, rcc->last_satd );
> - double qf = 1.0;
> /* For small VBVs, allow the frame to use up the entire VBV. */
> double max_fill_factor = h->param.rc.i_vbv_buffer_size >= 5*h->param.rc.i_vbv_max_bitrate / rcc->fps ? 2 : 1;
> /* For single-frame VBVs, request that the frame use up the entire VBV. */
> double min_fill_factor = rcc->single_frame_vbv ? 1 : 2;
>
> if( bits > rcc->buffer_fill/max_fill_factor )
> - qf = x264_clip3f( rcc->buffer_fill/(max_fill_factor*bits), 0.2, 1.0 );
> - q /= qf;
> - bits *= qf;
> + {
> + double qf = x264_clip3f( rcc->buffer_fill/(max_fill_factor*bits), 0.2, 1.0 );
> + q /= qf;
> + bits *= qf;
> + }
> if( bits < rcc->buffer_rate/min_fill_factor )
> - q *= bits*min_fill_factor/rcc->buffer_rate;
> + {
> + double qf = x264_clip3f( bits*min_fill_factor/rcc->buffer_rate, 0.001, 1.0 );
> + q *= qf;
> + }
> q = X264_MAX( q0, q );
> }
>
> - /* Apply MinCR restrictions */
> - double bits = predict_size( &rcc->pred[h->sh.i_type], q, rcc->last_satd );
> - if( bits > rcc->frame_size_maximum )
> - q *= bits / rcc->frame_size_maximum;
> - bits = predict_size( &rcc->pred[h->sh.i_type], q, rcc->last_satd );
> -
> /* Check B-frame complexity, and use up any bits that would
> * overflow before the next P-frame. */
> if( h->sh.i_type == SLICE_TYPE_P && !rcc->single_frame_vbv )
> {
> int nb = rcc->bframes;
> + double bits = predict_size( &rcc->pred[h->sh.i_type], q, rcc->last_satd );
> double pbbits = bits;
> double bbits = predict_size( rcc->pred_b_from_p, q * h->param.rc.f_pb_factor, rcc->last_satd );
> double space;
> @@ -2302,6 +2301,12 @@ static double clip_qscale( x264_t *h, int pict_type, double q )
> q = X264_MAX( q0/2, q );
> }
>
> + /* Apply MinCR and buffer fill restrictions */
> + double bits = predict_size( &rcc->pred[h->sh.i_type], q, rcc->last_satd );
> + double frame_size_maximum = X264_MIN( rcc->frame_size_maximum, X264_MAX( rcc->buffer_fill, 0.001 ) );
> + if( bits > frame_size_maximum )
> + q *= bits / frame_size_maximum;
> +
> if( !rcc->b_vbv_min_rate )
> q = X264_MAX( q0, q );
> }
>
>
>
> ------------------------------
>
> Message: 3
> Date: Thu, 13 Nov 2014 13:52:02 +0100 (CET)
> From: git at videolan.org (Anton Mitrofanov)
> To: x264-devel at videolan.org
> Subject: [x264-devel] Fix VBV with true VFR streams
> Message-ID: <20141113125202.DF804151DC1 at albiero.videolan.org>
> Content-Type: text/plain; charset=UTF-8
>
> x264 | branch: master | Anton Mitrofanov <BugMaster at narod.ru> | Tue Sep 2 01:48:00 2014 +0400| [204a9bd0a1bc507cbd69a77f3318afcb56ede65d] | committer: Fiona Glaser
>
> Fix VBV with true VFR streams
>
> > http://git.videolan.org/gitweb.cgi/x264.git/?a=commit;h=204a9bd0a1bc507cbd69a77f3318afcb56ede65d
> ---
>
> encoder/ratecontrol.c | 12 ++++++++----
> encoder/slicetype.c | 5 +----
> 2 files changed, 9 insertions(+), 8 deletions(-)
>
> diff --git a/encoder/ratecontrol.c b/encoder/ratecontrol.c
> index 8893c1f..86aad96 100644
> --- a/encoder/ratecontrol.c
> +++ b/encoder/ratecontrol.c
> @@ -2191,6 +2191,8 @@ static double clip_qscale( x264_t *h, int pict_type, double q )
>
> if( rcc->b_vbv && rcc->last_satd > 0 )
> {
> + double fenc_cpb_duration = (double)h->fenc->i_cpb_duration *
> + h->sps->vui.i_num_units_in_tick / h->sps->vui.i_time_scale;
> /* Lookahead VBV: raise the quantizer as necessary such that no frames in
> * the lookahead overflow and such that the buffer is in a reasonable state
> * by the end of the lookahead. */
> @@ -2206,6 +2208,7 @@ static double clip_qscale( x264_t *h, int pict_type, double q )
> double buffer_fill_cur = rcc->buffer_fill - cur_bits;
> double target_fill;
> double total_duration = 0;
> + double last_duration = fenc_cpb_duration;
> frame_q[0] = h->sh.i_type == SLICE_TYPE_I ? q * h->param.rc.f_ip_factor : q;
> frame_q[1] = frame_q[0] * h->param.rc.f_pb_factor;
> frame_q[2] = frame_q[0] / h->param.rc.f_ip_factor;
> @@ -2213,8 +2216,8 @@ static double clip_qscale( x264_t *h, int pict_type, double q )
> /* Loop over the planned future frames. */
> for( int j = 0; buffer_fill_cur >= 0 && buffer_fill_cur <= rcc->buffer_size; j++ )
> {
> - total_duration += h->fenc->f_planned_cpb_duration[j];
> - buffer_fill_cur += rcc->vbv_max_rate * h->fenc->f_planned_cpb_duration[j];
> + total_duration += last_duration;
> + buffer_fill_cur += rcc->vbv_max_rate * last_duration;
> int i_type = h->fenc->i_planned_type[j];
> int i_satd = h->fenc->i_planned_satd[j];
> if( i_type == X264_TYPE_AUTO )
> @@ -2222,6 +2225,7 @@ static double clip_qscale( x264_t *h, int pict_type, double q )
> i_type = IS_X264_TYPE_I( i_type ) ? SLICE_TYPE_I : IS_X264_TYPE_B( i_type ) ? SLICE_TYPE_B : SLICE_TYPE_P;
> cur_bits = predict_size( &rcc->pred[i_type], frame_q[i_type], i_satd );
> buffer_fill_cur -= cur_bits;
> + last_duration = h->fenc->f_planned_cpb_duration[j];
> }
> /* Try to get to get the buffer at least 50% filled, but don't set an impossible goal. */
> target_fill = X264_MIN( rcc->buffer_fill + total_duration * rcc->vbv_max_rate * 0.5, rcc->buffer_size * 0.5 );
> @@ -2286,13 +2290,13 @@ static double clip_qscale( x264_t *h, int pict_type, double q )
> double bframe_cpb_duration = 0;
> double minigop_cpb_duration;
> for( int i = 0; i < nb; i++ )
> - bframe_cpb_duration += h->fenc->f_planned_cpb_duration[1+i];
> + bframe_cpb_duration += h->fenc->f_planned_cpb_duration[i];
>
> if( bbits * nb > bframe_cpb_duration * rcc->vbv_max_rate )
> nb = 0;
> pbbits += nb * bbits;
>
> - minigop_cpb_duration = bframe_cpb_duration + h->fenc->f_planned_cpb_duration[0];
> + minigop_cpb_duration = bframe_cpb_duration + fenc_cpb_duration;
> space = rcc->buffer_fill + minigop_cpb_duration*rcc->vbv_max_rate - rcc->buffer_size;
> if( pbbits < space )
> {
> diff --git a/encoder/slicetype.c b/encoder/slicetype.c
> index 0f4d831..dd5e318 100644
> --- a/encoder/slicetype.c
> +++ b/encoder/slicetype.c
> @@ -1853,14 +1853,11 @@ void x264_slicetype_decide( x264_t *h )
> if( i )
> {
> x264_calculate_durations( h, h->lookahead->next.list[i], h->lookahead->next.list[i-1], &h->i_cpb_delay, &h->i_coded_fields );
> - h->lookahead->next.list[0]->f_planned_cpb_duration[i-1] = (double)h->lookahead->next.list[i-1]->i_cpb_duration *
> + h->lookahead->next.list[0]->f_planned_cpb_duration[i-1] = (double)h->lookahead->next.list[i]->i_cpb_duration *
> h->sps->vui.i_num_units_in_tick / h->sps->vui.i_time_scale;
> }
> else
> x264_calculate_durations( h, h->lookahead->next.list[i], NULL, &h->i_cpb_delay, &h->i_coded_fields );
> -
> - h->lookahead->next.list[0]->f_planned_cpb_duration[i] = (double)h->lookahead->next.list[i]->i_cpb_duration *
> - h->sps->vui.i_num_units_in_tick / h->sps->vui.i_time_scale;
> }
> }
>
>
>
>
> ------------------------------
>
> Message: 4
> Date: Thu, 13 Nov 2014 13:52:03 +0100 (CET)
> From: git at videolan.org (Anton Mitrofanov)
> To: x264-devel at videolan.org
> Subject: [x264-devel] Fix inappropriate instruction use
> Message-ID: <20141113125203.366EF1527EE at albiero.videolan.org>
> Content-Type: text/plain; charset=UTF-8
>
> x264 | branch: master | Anton Mitrofanov <BugMaster at narod.ru> | Thu Aug 28 20:13:13 2014 +0400| [9df377f87702c82a2202d34919c07e32c60b40ae] | committer: Fiona Glaser
>
> Fix inappropriate instruction use
>
> > http://git.videolan.org/gitweb.cgi/x264.git/?a=commit;h=9df377f87702c82a2202d34919c07e32c60b40ae
> ---
>
> common/dct.c | 2 +-
> common/quant.c | 4 ++--
> common/x86/dct-a.asm | 2 +-
> common/x86/dct.h | 2 +-
> common/x86/pixel-a.asm | 2 +-
> common/x86/quant-a.asm | 2 +-
> common/x86/quant.h | 4 ++--
> 7 files changed, 9 insertions(+), 9 deletions(-)
>
> diff --git a/common/dct.c b/common/dct.c
> index f5900ef..08f4e89 100644
> --- a/common/dct.c
> +++ b/common/dct.c
> @@ -611,7 +611,6 @@ void x264_dct_init( int cpu, x264_dct_function_t *dctf )
> {
> dctf->sub4x4_dct = x264_sub4x4_dct_mmx;
> dctf->add4x4_idct = x264_add4x4_idct_mmx;
> - dctf->dct4x4dc = x264_dct4x4dc_mmx;
> dctf->idct4x4dc = x264_idct4x4dc_mmx;
> dctf->sub8x8_dct_dc = x264_sub8x8_dct_dc_mmx2;
>
> @@ -630,6 +629,7 @@ void x264_dct_init( int cpu, x264_dct_function_t *dctf )
>
> if( cpu&X264_CPU_MMX2 )
> {
> + dctf->dct4x4dc = x264_dct4x4dc_mmx2;
> dctf->add8x8_idct_dc = x264_add8x8_idct_dc_mmx2;
> dctf->add16x16_idct_dc = x264_add16x16_idct_dc_mmx2;
> }
> diff --git a/common/quant.c b/common/quant.c
> index d7b6911..31d8901 100644
> --- a/common/quant.c
> +++ b/common/quant.c
> @@ -558,8 +558,6 @@ void x264_quant_init( x264_t *h, int cpu, x264_quant_function_t *pf )
> if( cpu&X264_CPU_MMX )
> {
> #if ARCH_X86
> - pf->quant_4x4 = x264_quant_4x4_mmx;
> - pf->quant_8x8 = x264_quant_8x8_mmx;
> pf->dequant_4x4 = x264_dequant_4x4_mmx;
> pf->dequant_4x4_dc = x264_dequant_4x4dc_mmx2;
> pf->dequant_8x8 = x264_dequant_8x8_mmx;
> @@ -576,6 +574,8 @@ void x264_quant_init( x264_t *h, int cpu, x264_quant_function_t *pf )
> {
> pf->quant_2x2_dc = x264_quant_2x2_dc_mmx2;
> #if ARCH_X86
> + pf->quant_4x4 = x264_quant_4x4_mmx2;
> + pf->quant_8x8 = x264_quant_8x8_mmx2;
> pf->quant_4x4_dc = x264_quant_4x4_dc_mmx2;
> pf->decimate_score15 = x264_decimate_score15_mmx2;
> pf->decimate_score16 = x264_decimate_score16_mmx2;
> diff --git a/common/x86/dct-a.asm b/common/x86/dct-a.asm
> index 4376e36..bc82ff6 100644
> --- a/common/x86/dct-a.asm
> +++ b/common/x86/dct-a.asm
> @@ -143,7 +143,7 @@ INIT_XMM avx
> DCT4x4_DC
> %else
>
> -INIT_MMX mmx
> +INIT_MMX mmx2
> cglobal dct4x4dc, 1,1
> movq m3, [r0+24]
> movq m2, [r0+16]
> diff --git a/common/x86/dct.h b/common/x86/dct.h
> index 337a632..f22a979 100644
> --- a/common/x86/dct.h
> +++ b/common/x86/dct.h
> @@ -70,7 +70,7 @@ void x264_add8x8_idct_dc_avx ( pixel *p_dst, dctcoef dct [ 4] );
> void x264_add16x16_idct_dc_avx ( pixel *p_dst, dctcoef dct [16] );
> void x264_add16x16_idct_dc_avx2 ( uint8_t *p_dst, int16_t dct [16] );
>
> -void x264_dct4x4dc_mmx ( int16_t d[16] );
> +void x264_dct4x4dc_mmx2 ( int16_t d[16] );
> void x264_dct4x4dc_sse2 ( int32_t d[16] );
> void x264_dct4x4dc_avx ( int32_t d[16] );
> void x264_idct4x4dc_mmx ( int16_t d[16] );
> diff --git a/common/x86/pixel-a.asm b/common/x86/pixel-a.asm
> index 262c537..f5f6a82 100644
> --- a/common/x86/pixel-a.asm
> +++ b/common/x86/pixel-a.asm
> @@ -1600,7 +1600,7 @@ cglobal pixel_satd_4x4, 4,6
> %macro SATDS_SSE2 0
> %define vertical ((notcpuflag(ssse3) || cpuflag(atom)) || HIGH_BIT_DEPTH)
>
> -%if vertical==0 || HIGH_BIT_DEPTH
> +%if cpuflag(ssse3) && (vertical==0 || HIGH_BIT_DEPTH)
> cglobal pixel_satd_4x4, 4, 6, 6
> SATD_START_MMX
> mova m4, [hmul_4p]
> diff --git a/common/x86/quant-a.asm b/common/x86/quant-a.asm
> index fb588d3..731f7d1 100644
> --- a/common/x86/quant-a.asm
> +++ b/common/x86/quant-a.asm
> @@ -453,7 +453,7 @@ INIT_MMX mmx2
> QUANT_DC quant_2x2_dc, 1
> %if ARCH_X86_64 == 0 ; not needed because sse2 is faster
> QUANT_DC quant_4x4_dc, 4
> -INIT_MMX mmx
> +INIT_MMX mmx2
> QUANT_AC quant_4x4, 4
> QUANT_AC quant_8x8, 16
> %endif
> diff --git a/common/x86/quant.h b/common/x86/quant.h
> index 1fcb800..c6a8a9b 100644
> --- a/common/x86/quant.h
> +++ b/common/x86/quant.h
> @@ -30,8 +30,8 @@
>
> int x264_quant_2x2_dc_mmx2( dctcoef dct[4], int mf, int bias );
> int x264_quant_4x4_dc_mmx2( dctcoef dct[16], int mf, int bias );
> -int x264_quant_4x4_mmx( dctcoef dct[16], udctcoef mf[16], udctcoef bias[16] );
> -int x264_quant_8x8_mmx( dctcoef dct[64], udctcoef mf[64], udctcoef bias[64] );
> +int x264_quant_4x4_mmx2( dctcoef dct[16], udctcoef mf[16], udctcoef bias[16] );
> +int x264_quant_8x8_mmx2( dctcoef dct[64], udctcoef mf[64], udctcoef bias[64] );
> int x264_quant_2x2_dc_sse2( dctcoef dct[16], int mf, int bias );
> int x264_quant_4x4_dc_sse2( dctcoef dct[16], int mf, int bias );
> int x264_quant_4x4_sse2( dctcoef dct[16], udctcoef mf[16], udctcoef bias[16] );
>
>
>
> ------------------------------
>
> Message: 5
> Date: Thu, 13 Nov 2014 13:52:02 +0100 (CET)
> From: git at videolan.org (Anton Mitrofanov)
> To: x264-devel at videolan.org
> Subject: [x264-devel] x264asm: warn when inappropriate instruction
> used in function with specified cpuflags
> Message-ID: <20141113125203.02ECB152262 at albiero.videolan.org>
> Content-Type: text/plain; charset=UTF-8
>
> x264 | branch: master | Anton Mitrofanov <BugMaster at narod.ru> | Thu Aug 28 18:38:53 2014 +0400| [73b8686fc22c9247d90963983d406cd7b9131068] | committer: Fiona Glaser
>
> x264asm: warn when inappropriate instruction used in function with specified cpuflags
>
> > http://git.videolan.org/gitweb.cgi/x264.git/?a=commit;h=73b8686fc22c9247d90963983d406cd7b9131068
> ---
>
> common/x86/x86inc.asm | 581 +++++++++++++++++++++++++------------------------
> 1 file changed, 295 insertions(+), 286 deletions(-)
>
> Diff: http://git.videolan.org/gitweb.cgi/x264.git/?a=commitdiff;h=73b8686fc22c9247d90963983d406cd7b9131068
>
>
> ------------------------------
>
> Message: 6
> Date: Thu, 13 Nov 2014 13:52:03 +0100 (CET)
> From: git at videolan.org (Anton Mitrofanov)
> To: x264-devel at videolan.org
> Subject: [x264-devel] Fix few clang -Wunused-* warnings
> Message-ID: <20141113125203.5959D151DC1 at albiero.videolan.org>
> Content-Type: text/plain; charset=UTF-8
>
> x264 | branch: master | Anton Mitrofanov <BugMaster at narod.ru> | Fri Sep 5 19:30:47 2014 +0400| [01204b60367f4959e8393652dd30f0cfba2d2c80] | committer: Fiona Glaser
>
> Fix few clang -Wunused-* warnings
>
> > http://git.videolan.org/gitweb.cgi/x264.git/?a=commit;h=01204b60367f4959e8393652dd30f0cfba2d2c80
> ---
>
> common/x86/predict-c.c | 5 +++++
> encoder/cavlc.c | 2 ++
> 2 files changed, 7 insertions(+)
>
> diff --git a/common/x86/predict-c.c b/common/x86/predict-c.c
> index 086c45d..0c668fa 100644
> --- a/common/x86/predict-c.c
> +++ b/common/x86/predict-c.c
> @@ -65,12 +65,17 @@ PREDICT_16x16_DC_LEFT( avx2 )
> H += i * ( src[j+i - FDEC_STRIDE ] - src[j-i - FDEC_STRIDE ] );\
> V += i * ( src[(j+i)*FDEC_STRIDE -1] - src[(j-i)*FDEC_STRIDE -1] );
>
> +#if HAVE_X86_INLINE_ASM
> +#if HIGH_BIT_DEPTH
> ALIGNED_16( static const int16_t pw_12345678[8] ) = {1,2,3,4,5,6,7,8};
> ALIGNED_16( static const int16_t pw_m87654321[8] ) = {-8,-7,-6,-5,-4,-3,-2,-1};
> ALIGNED_16( static const int16_t pw_m32101234[8] ) = {-3,-2,-1,0,1,2,3,4};
> +#else // !HIGH_BIT_DEPTH
> ALIGNED_8( static const int8_t pb_12345678[8] ) = {1,2,3,4,5,6,7,8};
> ALIGNED_8( static const int8_t pb_m87654321[8] ) = {-8,-7,-6,-5,-4,-3,-2,-1};
> ALIGNED_8( static const int8_t pb_m32101234[8] ) = {-3,-2,-1,0,1,2,3,4};
> +#endif // HIGH_BIT_DEPTH
> +#endif // HAVE_X86_INLINE_ASM
>
> #define PREDICT_16x16_P_CORE\
> int H = 0;\
> diff --git a/encoder/cavlc.c b/encoder/cavlc.c
> index 5d826ff..1d393fd 100644
> --- a/encoder/cavlc.c
> +++ b/encoder/cavlc.c
> @@ -289,6 +289,7 @@ static ALWAYS_INLINE void x264_cavlc_macroblock_luma_residual( x264_t *h, int pl
> x264_cavlc_block_residual( h, DCT_LUMA_4x4, i4+i8*4+p*16, h->dct.luma4x4[i4+i8*4+p*16] );
> }
>
> +#if RDO_SKIP_BS
> static ALWAYS_INLINE void x264_cavlc_partition_luma_residual( x264_t *h, int i8, int p )
> {
> if( h->mb.b_transform_8x8 && h->mb.cache.non_zero_count[x264_scan8[i8*4]] )
> @@ -299,6 +300,7 @@ static ALWAYS_INLINE void x264_cavlc_partition_luma_residual( x264_t *h, int i8,
> for( int i4 = 0; i4 < 4; i4++ )
> x264_cavlc_block_residual( h, DCT_LUMA_4x4, i4+i8*4+p*16, h->dct.luma4x4[i4+i8*4+p*16] );
> }
> +#endif
>
> static void x264_cavlc_mb_header_i( x264_t *h, int i_mb_type, int i_mb_i_offset, int chroma )
> {
>
>
>
> ------------------------------
>
> Subject: Digest Footer
>
> _______________________________________________
> x264-devel mailing list
> x264-devel at videolan.org
> https://mailman.videolan.org/listinfo/x264-devel
>
>
> ------------------------------
>
> End of x264-devel Digest, Vol 90, Issue 8
> *****************************************
More information about the x264-devel
mailing list