generic-library/ffmpeg
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
aa1f38015c
ffmpeg/libavcodec/x86/dcadsp.asm
James Almer aa1f38015c x86/synth_filter: improve FMA version 
Replace mulps+subps with fnmaddps, resulting in two less instructions inside the
inner loops.
About 1% faster FMA3 performance.

Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-03-17 21:04:15 +01:00

416 lines
11 KiB
NASM
Raw Blame History

;******************************************************************************
;* SSE-optimized functions for the DCA decoder
;* Copyright (C) 2012-2014 Christophe Gisquet <christophe.gisquet@gmail.com>
;*
;* This file is part of FFmpeg.
;*
;* FFmpeg is free software; you can redistribute it and/or
;* modify it under the terms of the GNU Lesser General Public
;* License as published by the Free Software Foundation; either
;* version 2.1 of the License, or (at your option) any later version.
;*
;* FFmpeg is distributed in the hope that it will be useful,
;* but WITHOUT ANY WARRANTY; without even the implied warranty of
;* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
;* Lesser General Public License for more details.
;*
;* You should have received a copy of the GNU Lesser General Public
;* License along with FFmpeg; if not, write to the Free Software
;* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
;******************************************************************************
%include "libavutil/x86/x86util.asm"
SECTION_RODATA
pf_inv16: times 4 dd 0x3D800000 ; 1/16
SECTION_TEXT
; void decode_hf(float dst[DCA_SUBBANDS][8], const int32_t vq_num[DCA_SUBBANDS],
; const int8_t hf_vq[1024][32], intptr_t vq_offset,
; int32_t scale[DCA_SUBBANDS][2], intptr_t start, intptr_t end)
%macro DECODE_HF 0
cglobal decode_hf, 6,6,5, dst, num, src, offset, scale, start, end
lea srcq, [srcq + offsetq]
shl startq, 2
mov offsetd, endm
%define DICT offsetq
shl offsetq, 2
mov endm, offsetq
.loop:
%if ARCH_X86_64
mov offsetd, [scaleq + 2 * startq]
cvtsi2ss m0, offsetd
%else
cvtsi2ss m0, [scaleq + 2 * startq]
%endif
mov offsetd, [numq + startq]
mulss m0, [pf_inv16]
shl DICT, 5
shufps m0, m0, 0
%if cpuflag(sse2)
%if cpuflag(sse4)
pmovsxbd m1, [srcq + DICT + 0]
pmovsxbd m2, [srcq + DICT + 4]
%else
movq m1, [srcq + DICT]
punpcklbw m1, m1
mova m2, m1
punpcklwd m1, m1
punpckhwd m2, m2
psrad m1, 24
psrad m2, 24
%endif
cvtdq2ps m1, m1
cvtdq2ps m2, m2
%else
movd mm0, [srcq + DICT + 0]
movd mm1, [srcq + DICT + 4]
punpcklbw mm0, mm0
punpcklbw mm1, mm1
movq mm2, mm0
movq mm3, mm1
punpcklwd mm0, mm0
punpcklwd mm1, mm1
punpckhwd mm2, mm2
punpckhwd mm3, mm3
psrad mm0, 24
psrad mm1, 24
psrad mm2, 24
psrad mm3, 24
cvtpi2ps m1, mm0
cvtpi2ps m2, mm1
cvtpi2ps m3, mm2
cvtpi2ps m4, mm3
shufps m0, m0, 0
shufps m1, m3, q1010
shufps m2, m4, q1010
%endif
mulps m1, m0
mulps m2, m0
mova [dstq + 8 * startq + 0], m1
mova [dstq + 8 * startq + 16], m2
add startq, 4
cmp startq, endm
jl .loop
.end:
%if notcpuflag(sse2)
emms
%endif
REP_RET
%endmacro
%if ARCH_X86_32
INIT_XMM sse
DECODE_HF
%endif
INIT_XMM sse2
DECODE_HF
INIT_XMM sse4
DECODE_HF
; %1=v0/v1 %2=in1 %3=in2
%macro FIR_LOOP 2-3
.loop%1:
%define va m1
%define vb m2
%if %1
%define OFFSET 0
%else
%define OFFSET NUM_COEF*count
%endif
; for v0, incrementing and for v1, decrementing
mova va, [cf0q + OFFSET]
mova vb, [cf0q + OFFSET + 4*NUM_COEF]
%if %0 == 3
mova m4, [cf0q + OFFSET + mmsize]
mova m0, [cf0q + OFFSET + 4*NUM_COEF + mmsize]
%endif
mulps va, %2
mulps vb, %2
%if %0 == 3
mulps m4, %3
mulps m0, %3
addps va, m4
addps vb, m0
%endif
; va = va1 va2 va3 va4
; vb = vb1 vb2 vb3 vb4
%if %1
SWAP va, vb
%endif
mova m4, va
unpcklps va, vb ; va3 vb3 va4 vb4
unpckhps m4, vb ; va1 vb1 va2 vb2
addps m4, va ; va1+3 vb1+3 va2+4 vb2+4
movhlps vb, m4 ; va1+3 vb1+3
addps vb, m4 ; va0..4 vb0..4
movh [outq + count], vb
%if %1
sub cf0q, 8*NUM_COEF
%endif
add count, 8
jl .loop%1
%endmacro
; void dca_lfe_fir(float *out, float *in, float *coefs)
%macro DCA_LFE_FIR 1
cglobal dca_lfe_fir%1, 3,3,6-%1, out, in, cf0
%define IN1 m3
%define IN2 m5
%define count inq
%define NUM_COEF 4*(2-%1)
%define NUM_OUT 32*(%1+1)
movu IN1, [inq + 4 - 1*mmsize]
shufps IN1, IN1, q0123
%if %1 == 0
movu IN2, [inq + 4 - 2*mmsize]
shufps IN2, IN2, q0123
%endif
mov count, -4*NUM_OUT
add cf0q, 4*NUM_COEF*NUM_OUT
add outq, 4*NUM_OUT
; compute v0 first
%if %1 == 0
FIR_LOOP 0, IN1, IN2
%else
FIR_LOOP 0, IN1
%endif
shufps IN1, IN1, q0123
mov count, -4*NUM_OUT
; cf1 already correctly positioned
add outq, 4*NUM_OUT ; outq now at out2
sub cf0q, 8*NUM_COEF
%if %1 == 0
shufps IN2, IN2, q0123
FIR_LOOP 1, IN2, IN1
%else
FIR_LOOP 1, IN1
%endif
RET
%endmacro
INIT_XMM sse
DCA_LFE_FIR 0
DCA_LFE_FIR 1
%macro SETZERO 1
%if cpuflag(sse2) && notcpuflag(avx)
pxor %1, %1
%else
xorps %1, %1, %1
%endif
%endmacro
%macro SHUF 3
%if cpuflag(avx)
mova %3, [%2 - 16]
vperm2f128 %1, %3, %3, 1
vshufps %1, %1, %1, q0123
%elif cpuflag(sse2)
pshufd %1, [%2], q0123
%else
mova %1, [%2]
shufps %1, %1, q0123
%endif
%endmacro
%macro INNER_LOOP 1
; reading backwards: ptr1 = synth_buf + j + i; ptr2 = synth_buf + j - i
;~ a += window[i + j] * (-synth_buf[15 - i + j])
;~ b += window[i + j + 16] * (synth_buf[i + j])
SHUF m5, ptr2 + j + (15 - 3) * 4, m6
mova m6, [ptr1 + j]
%if ARCH_X86_64
SHUF m11, ptr2 + j + (15 - 3) * 4 - mmsize, m12
mova m12, [ptr1 + j + mmsize]
%endif
%if cpuflag(fma3)
fmaddps m2, m6, [win + %1 + j + 16 * 4], m2
fnmaddps m1, m5, [win + %1 + j], m1
%if ARCH_X86_64
fmaddps m8, m12, [win + %1 + j + mmsize + 16 * 4], m8
fnmaddps m7, m11, [win + %1 + j + mmsize], m7
%endif
%else ; non-FMA
mulps m6, m6, [win + %1 + j + 16 * 4]
mulps m5, m5, [win + %1 + j]
%if ARCH_X86_64
mulps m12, m12, [win + %1 + j + mmsize + 16 * 4]
mulps m11, m11, [win + %1 + j + mmsize]
%endif
addps m2, m2, m6
subps m1, m1, m5
%if ARCH_X86_64
addps m8, m8, m12
subps m7, m7, m11
%endif
%endif ; cpuflag(fma3)
;~ c += window[i + j + 32] * (synth_buf[16 + i + j])
;~ d += window[i + j + 48] * (synth_buf[31 - i + j])
SHUF m6, ptr2 + j + (31 - 3) * 4, m5
mova m5, [ptr1 + j + 16 * 4]
%if ARCH_X86_64
SHUF m12, ptr2 + j + (31 - 3) * 4 - mmsize, m11
mova m11, [ptr1 + j + mmsize + 16 * 4]
%endif
%if cpuflag(fma3)
fmaddps m3, m5, [win + %1 + j + 32 * 4], m3
fmaddps m4, m6, [win + %1 + j + 48 * 4], m4
%if ARCH_X86_64
fmaddps m9, m11, [win + %1 + j + mmsize + 32 * 4], m9
fmaddps m10, m12, [win + %1 + j + mmsize + 48 * 4], m10
%endif
%else ; non-FMA
mulps m5, m5, [win + %1 + j + 32 * 4]
mulps m6, m6, [win + %1 + j + 48 * 4]
%if ARCH_X86_64
mulps m11, m11, [win + %1 + j + mmsize + 32 * 4]
mulps m12, m12, [win + %1 + j + mmsize + 48 * 4]
%endif
addps m3, m3, m5
addps m4, m4, m6
%if ARCH_X86_64
addps m9, m9, m11
addps m10, m10, m12
%endif
%endif ; cpuflag(fma3)
sub j, 64 * 4
%endmacro
; void ff_synth_filter_inner_<opt>(float *synth_buf, float synth_buf2[32],
; const float window[512], float out[32],
; intptr_t offset, float scale)
%macro SYNTH_FILTER 0
cglobal synth_filter_inner, 0, 6 + 4 * ARCH_X86_64, 7 + 6 * ARCH_X86_64, \
synth_buf, synth_buf2, window, out, off, scale
%define scale m0
%if ARCH_X86_32 || WIN64
%if cpuflag(sse2) && notcpuflag(avx)
movd m0, scalem
SPLATD m0
%else
VBROADCASTSS m0, scalem
%endif
; Make sure offset is in a register and not on the stack
%define OFFQ r4q
%else
SPLATD xmm0
%if cpuflag(avx)
vinsertf128 m0, m0, xmm0, 1
%endif
%define OFFQ offq
%endif
; prepare inner counter limit 1
mov r5q, 480
sub r5q, offmp
and r5q, -64
shl r5q, 2
mov OFFQ, r5q
%define i r5q
mov i, 16 * 4 - (ARCH_X86_64 + 1) * mmsize ; main loop counter
%define buf2 synth_buf2q
%if ARCH_X86_32
mov buf2, synth_buf2mp
%endif
.mainloop
; m1 = a m2 = b m3 = c m4 = d
SETZERO m3
SETZERO m4
mova m1, [buf2 + i]
mova m2, [buf2 + i + 16 * 4]
%if ARCH_X86_32
%define ptr1 r0q
%define ptr2 r1q
%define win r2q
%define j r3q
mov win, windowm
mov ptr1, synth_bufm
add win, i
add ptr1, i
%else ; ARCH_X86_64
%define ptr1 r6q
%define ptr2 r7q ; must be loaded
%define win r8q
%define j r9q
SETZERO m9
SETZERO m10
mova m7, [buf2 + i + mmsize]
mova m8, [buf2 + i + mmsize + 16 * 4]
lea win, [windowq + i]
lea ptr1, [synth_bufq + i]
%endif
mov ptr2, synth_bufmp
; prepare the inner loop counter
mov j, OFFQ
sub ptr2, i
.loop1:
INNER_LOOP 0
jge .loop1
mov j, 448 * 4
sub j, OFFQ
jz .end
sub ptr1, j
sub ptr2, j
add win, OFFQ ; now at j-64, so define OFFSET
sub j, 64 * 4
.loop2:
INNER_LOOP 64 * 4
jge .loop2
.end:
%if ARCH_X86_32
mov buf2, synth_buf2m ; needed for next iteration anyway
mov outq, outmp ; j, which will be set again during it
%endif
;~ out[i] = a * scale;
;~ out[i + 16] = b * scale;
mulps m1, m1, scale
mulps m2, m2, scale
%if ARCH_X86_64
mulps m7, m7, scale
mulps m8, m8, scale
%endif
;~ synth_buf2[i] = c;
;~ synth_buf2[i + 16] = d;
mova [buf2 + i + 0 * 4], m3
mova [buf2 + i + 16 * 4], m4
%if ARCH_X86_64
mova [buf2 + i + 0 * 4 + mmsize], m9
mova [buf2 + i + 16 * 4 + mmsize], m10
%endif
;~ out[i] = a;
;~ out[i + 16] = a;
mova [outq + i + 0 * 4], m1
mova [outq + i + 16 * 4], m2
%if ARCH_X86_64
mova [outq + i + 0 * 4 + mmsize], m7
mova [outq + i + 16 * 4 + mmsize], m8
%endif
sub i, (ARCH_X86_64 + 1) * mmsize
jge .mainloop
RET
%endmacro
%if ARCH_X86_32
INIT_XMM sse
SYNTH_FILTER
%endif
INIT_XMM sse2
SYNTH_FILTER
%if HAVE_AVX_EXTERNAL
INIT_YMM avx
SYNTH_FILTER
%endif
%if HAVE_FMA3_EXTERNAL
INIT_YMM fma3
SYNTH_FILTER
%endif
Reference in New Issue View Git Blame Copy Permalink