[FFmpeg-devel] [PATCH] mdct15: simplify the fft15 x86 SIMD

[Rostislav Pehlivanov]

On 6 May 2018 at 23:19, Rostislav Pehlivanov <atomnuker at gmail.com> wrote:

> Saves 1 gpr and 2 instructions and simplifies the macros a bit.
>
> Signed-off-by: Rostislav Pehlivanov <atomnuker at gmail.com>
> ---
>  libavcodec/x86/mdct15.asm | 37 +++++++++++++++++--------------------
>  1 file changed, 17 insertions(+), 20 deletions(-)
>
> diff --git a/libavcodec/x86/mdct15.asm b/libavcodec/x86/mdct15.asm
> index 0309112538..2a2cdbd21b 100644
> --- a/libavcodec/x86/mdct15.asm
> +++ b/libavcodec/x86/mdct15.asm
> @@ -76,7 +76,7 @@ SECTION .text
>      addps       m%3,  m%3,  m0          ; Finally offset with DCs
>  %endmacro
>
> -%macro BUTTERFLIES_DC 2 ; %1 - exptab_offset, %2 - out
> +%macro BUTTERFLIES_DC 1 ; %1 - exptab_offset
>      mulps xm0,  xm9, [exptabq + %1 + 16*0]
>      mulps xm1, xm10, [exptabq + %1 + 16*1]
>
> @@ -86,10 +86,10 @@ SECTION .text
>      addps   xm0,  xm1
>      addps   xm0,  xm8
>
> -    movsd [%2q], xm0
> +    movsd [outq], xm0
>  %endmacro
>
> -%macro BUTTERFLIES_AC 2 ; exptab, exptab_offset, src1, src2, src3, out
> (uses m0-m3)
> +%macro BUTTERFLIES_AC 1 ; %1 - exptab_offset
>      mulps  m0, m12, [exptabq + 64*0 + 0*mmsize + %1]
>      mulps  m1, m12, [exptabq + 64*0 + 1*mmsize + %1]
>      mulps  m2, m13, [exptabq + 64*1 + 0*mmsize + %1]
> @@ -104,15 +104,14 @@ SECTION .text
>
>      vextractf128 xm1, m0, 1
>
> -    movlps [%2q + strideq*1], xm0
> -    movhps [%2q + strideq*2], xm0
> -    movlps [%2q +  stride3q], xm1
> -    movhps [%2q + strideq*4], xm1
> +    movlps [outq + strideq*1], xm0
> +    movhps [outq + strideq*2], xm0
> +    movlps [outq +  stride3q], xm1
> +    movhps [outq + strideq*4], xm1
>  %endmacro
>
>  INIT_YMM avx
> -cglobal fft15, 4, 6, 14, out, in, exptab, stride, stride3, stride5
> -%define out0q inq
> +cglobal fft15, 4, 5, 14, out, in, exptab, stride, stride5
>      shl strideq, 3
>
>      movaps xm5, [exptabq + 480 + 16*0]
> @@ -123,22 +122,20 @@ cglobal fft15, 4, 6, 14, out, in, exptab, stride,
> stride3, stride5
>      FFT5  8,  xm9, 12
>      FFT5 16, xm10, 13
>
> +%define stride3q inq
>      lea stride3q, [strideq + strideq*2]
>      lea stride5q, [strideq + strideq*4]
>
> -    mov out0q, outq
> +    BUTTERFLIES_DC (8*6 + 4*0)*2*4
> +    BUTTERFLIES_AC (8*0 + 0*0)*2*4
>
> -    BUTTERFLIES_DC (8*6 + 4*0)*2*4, out0
> -    lea outq, [out0q + stride5q*1]
> -    BUTTERFLIES_DC (8*6 + 4*1)*2*4, out
> -    lea outq, [out0q + stride5q*2]
> -    BUTTERFLIES_DC (8*6 + 4*2)*2*4, out
> +    add outq, stride5q
> +    BUTTERFLIES_DC (8*6 + 4*1)*2*4
> +    BUTTERFLIES_AC (8*2 + 0*0)*2*4
>
> -    BUTTERFLIES_AC (8*0)*2*4, out0
> -    lea outq, [out0q + stride5q*1]
> -    BUTTERFLIES_AC (8*2)*2*4, out
> -    lea outq, [out0q + stride5q*2]
> -    BUTTERFLIES_AC (8*4)*2*4, out
> +    add outq, stride5q
> +    BUTTERFLIES_DC (8*6 + 4*2)*2*4
> +    BUTTERFLIES_AC (8*4 + 0*0)*2*4
>
>      RET
>
> --
> 2.17.0
>
>
Pushed alongside a patch to simplify the init function