1
0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2024-12-28 20:53:54 +02:00
FFmpeg/libavcodec/x86/alacdsp.asm
Lynne bbe95f7353
x86: replace explicit REP_RETs with RETs
From x86inc:
> On AMD cpus <=K10, an ordinary ret is slow if it immediately follows either
> a branch or a branch target. So switch to a 2-byte form of ret in that case.
> We can automatically detect "follows a branch", but not a branch target.
> (SSSE3 is a sufficient condition to know that your cpu doesn't have this problem.)

x86inc can automatically determine whether to use REP_RET rather than
REP in most of these cases, so impact is minimal. Additionally, a few
REP_RETs were used unnecessary, despite the return being nowhere near a
branch.

The only CPUs affected were AMD K10s, made between 2007 and 2011, 16
years ago and 12 years ago, respectively.

In the future, everyone involved with x86inc should consider dropping
REP_RETs altogether.
2023-02-01 04:23:55 +01:00

134 lines
3.5 KiB
NASM

;******************************************************************************
;* ALAC DSP SIMD optimizations
;*
;* Copyright (C) 2015 James Almer
;*
;* 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 .text
INIT_XMM sse4
%if ARCH_X86_64
cglobal alac_decorrelate_stereo, 2, 5, 8, buf0, len, shift, weight, buf1
%else
cglobal alac_decorrelate_stereo, 2, 3, 8, buf0, len, shift, weight
%define buf1q r2q
%endif
movd m6, shiftm
movd m7, weightm
SPLATD m7
shl lend, 2
mov buf1q, [buf0q + gprsize]
mov buf0q, [buf0q]
add buf1q, lenq
add buf0q, lenq
neg lenq
align 16
.loop:
mova m0, [buf0q + lenq]
mova m1, [buf0q + lenq + mmsize]
mova m2, [buf1q + lenq]
mova m3, [buf1q + lenq + mmsize]
pmulld m4, m2, m7
pmulld m5, m3, m7
psrad m4, m6
psrad m5, m6
psubd m0, m4
psubd m1, m5
paddd m2, m0
paddd m3, m1
mova [buf1q + lenq], m0
mova [buf1q + lenq + mmsize], m1
mova [buf0q + lenq], m2
mova [buf0q + lenq + mmsize], m3
add lenq, mmsize*2
jl .loop
RET
INIT_XMM sse2
cglobal alac_append_extra_bits_stereo, 2, 5, 5, buf0, exbuf0, buf1, exbuf1, len
movifnidn lend, lenm
movd m4, r2m ; exbits
shl lend, 2
mov buf1q, [buf0q + gprsize]
mov buf0q, [buf0q]
mov exbuf1q, [exbuf0q + gprsize]
mov exbuf0q, [exbuf0q]
add buf1q, lenq
add buf0q, lenq
add exbuf1q, lenq
add exbuf0q, lenq
neg lenq
align 16
.loop:
mova m0, [buf0q + lenq]
mova m1, [buf0q + lenq + mmsize]
pslld m0, m4
pslld m1, m4
mova m2, [buf1q + lenq]
mova m3, [buf1q + lenq + mmsize]
pslld m2, m4
pslld m3, m4
por m0, [exbuf0q + lenq]
por m1, [exbuf0q + lenq + mmsize]
por m2, [exbuf1q + lenq]
por m3, [exbuf1q + lenq + mmsize]
mova [buf0q + lenq ], m0
mova [buf0q + lenq + mmsize], m1
mova [buf1q + lenq ], m2
mova [buf1q + lenq + mmsize], m3
add lenq, mmsize*2
jl .loop
RET
%if ARCH_X86_64
cglobal alac_append_extra_bits_mono, 2, 5, 3, buf, exbuf, exbits, ch, len
%else
cglobal alac_append_extra_bits_mono, 2, 3, 3, buf, exbuf, len
%define exbitsm r2m
%endif
movifnidn lend, r4m
movd m2, exbitsm
shl lend, 2
mov bufq, [bufq]
mov exbufq, [exbufq]
add bufq, lenq
add exbufq, lenq
neg lenq
align 16
.loop:
mova m0, [bufq + lenq]
mova m1, [bufq + lenq + mmsize]
pslld m0, m2
pslld m1, m2
por m0, [exbufq + lenq]
por m1, [exbufq + lenq + mmsize]
mova [bufq + lenq], m0
mova [bufq + lenq + mmsize], m1
add lenq, mmsize*2
jl .loop
RET