[FFmpeg-devel] [PATCH 4/7] x86: sbrdsp: implement SSE hf_apply_noise

[Michael Niedermayer]

On Sat, Apr 06, 2013 at 10:52:11AM +0000, Christophe Gisquet wrote:
> 233 to 115(sse)/110(sse2) cycles on Arrandale and Win64.
> Replacing the multiplication by s_m[m] by an andps and an xorps with
> appropriate vectors is slower. Unrolling is a 15 cycles win.
> ---
>  libavcodec/x86/sbrdsp.asm    | 145 +++++++++++++++++++++++++++++++++++++++++++
>  libavcodec/x86/sbrdsp_init.c |  32 ++++++++++
>  2 files changed, 177 insertions(+)
> 
> diff --git a/libavcodec/x86/sbrdsp.asm b/libavcodec/x86/sbrdsp.asm
> index 65c972e..a7998fa 100644
> --- a/libavcodec/x86/sbrdsp.asm
> +++ b/libavcodec/x86/sbrdsp.asm
> @@ -26,6 +26,12 @@ SECTION_RODATA
>  ps_mask         times 2 dd 1<<31, 0
>  ps_mask2        times 2 dd 0, 1<<31
>  ps_neg          times 4 dd 1<<31
> +ps_noise0       times 2 dd  1.0,  0.0,
> +ps_noise2       times 2 dd -1.0,  0.0
> +ps_noise13      dd  0.0,  1.0, 0.0, -1.0
> +                dd  0.0, -1.0, 0.0,  1.0
> +                dd  0.0,  1.0, 0.0, -1.0
> +cextern         sbr_noise_table
>  
>  SECTION_TEXT
>  
> @@ -358,3 +364,142 @@ SBR_QMF_DEINT_BFLY
>      
>  INIT_XMM sse2
>  SBR_QMF_DEINT_BFLY
> +
> +%if WIN64
> +%define NREGS 0
> +%else
> +%ifndef PIC
> +%define NREGS 1
> +%else
> +%define NREGS 0
> +%endif
> +%endif
> +
> +%macro SBUTTERFLY_F128 3
> +    vperm2i128  m%3, m%1, m%2, q0301 ; punpckh
> +    vinserti128 m%1, m%1, xm%2, 1    ; punpckl
> +    SWAP %2, %3
> +%endmacro
> +
> +%macro SBR_HF_APPLY_NOISE 0
> +; sbr_hf_apply_noise_0(float (*Y)[2], const float *s_m,
> +;                      const float *q_filt, int noise,
> +;                      int kx, int m_max)
> +cglobal sbr_hf_apply_noise_0, 5,5+NREGS,8, Y,s_m,q_filt,noise,kx,m_max
> +    mova       m0, [ps_noise0]
> +    jmp apply_noise_main %+ SUFFIX
> +
> +; sbr_hf_apply_noise_1(float (*Y)[2], const float *s_m,
> +;                      const float *q_filt, int noise,
> +;                      int kx, int m_max)
> +cglobal sbr_hf_apply_noise_1, 5,5+NREGS,8, Y,s_m,q_filt,noise,kx,m_max
> +    and       kxq, 1
> +    shl       kxq, 4

> +%if NREGS
> +    lea       r5q, [ps_noise13]
> +    mova       m0, [kxq + r5q]
> +%else
> +    mova       m0, [kxq + ps_noise13]
> +%endif

this could probably be simplified with some kind of generic macro
like
FOO mova, r5q, m0, ps_noise13, kxq
or something similar ...

[...]
-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

During times of universal deceit, telling the truth becomes a
revolutionary act. -- George Orwell
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: Digital signature
URL: <http://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20130406/7d826d50/attachment.asc>