1
0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2024-12-23 12:43:46 +02:00
FFmpeg/libavcodec/riscv/alacdsp_rvv.S
Rémi Denis-Courmont c541ecf0dc lavc/alacdsp: unroll RISC-V V loops
This increases the group multiplier as per T-Head C910 benchmarks:

alac_append_extra_bits_mono_c: 803.0
alac_append_extra_bits_stereo_c: 1604.2
alac_decorrelate_stereo_c: 1077.5

LMUL=1
alac_append_extra_bits_mono_rvv_i32: 418.2
alac_append_extra_bits_stereo_rvv_i32: 693.2
alac_decorrelate_stereo_rvv_i32: 673.5

LMUL=2
alac_append_extra_bits_mono_rvv_i32: 382.2
alac_append_extra_bits_stereo_rvv_i32: 648.2
alac_decorrelate_stereo_rvv_i32: 542.7

LMUL=4
alac_append_extra_bits_mono_rvv_i32: 241.5
alac_append_extra_bits_stereo_rvv_i32: 512.7
alac_decorrelate_stereo_rvv_i32: 364.2

LMUL=8
alac_append_extra_bits_mono_rvv_i32: 239.7
alac_append_extra_bits_stereo_rvv_i32: 497.2
alac_decorrelate_stereo_rvv_i32: 426.7
2023-07-16 23:24:00 +03:00

91 lines
2.4 KiB
ArmAsm

/*
* Copyright © 2022 Rémi Denis-Courmont.
*
* 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/riscv/asm.S"
#if (__riscv_xlen == 64)
func ff_alac_decorrelate_stereo_rvv, zve32x
ld a4, 8(a0)
ld a0, 0(a0)
1:
vsetvli t0, a1, e32, m4, ta, ma
vle32.v v24, (a4)
sub a1, a1, t0
vle32.v v16, (a0)
vmul.vx v8, v24, a3
vadd.vv v24, v24, v16
vsra.vx v8, v8, a2
vsub.vv v16, v16, v8
vsub.vv v24, v24, v8
vse32.v v16, (a4)
sh2add a4, t0, a4
vse32.v v24, (a0)
sh2add a0, t0, a0
bnez a1, 1b
ret
endfunc
func ff_alac_append_extra_bits_mono_rvv, zve32x
ld a0, (a0)
ld a1, (a1)
1:
vsetvli t0, a4, e32, m8, ta, ma
vle32.v v16, (a0)
sub a4, a4, t0
vle32.v v24, (a1)
sh2add a1, t0, a1
vsll.vx v16, v16, a2
vor.vv v16, v24, v16
vse32.v v16, (a0)
sh2add a0, t0, a0
bnez a4, 1b
ret
endfunc
func ff_alac_append_extra_bits_stereo_rvv, zve32x
ld a6, 8(a0)
ld a0, (a0)
ld a7, 8(a1)
ld a1, (a1)
1:
vsetvli t0, a4, e32, m8, ta, ma
vle32.v v16, (a0)
sub a4, a4, t0
vle32.v v0, (a6)
vsll.vx v16, v16, a2
vsll.vx v0, v0, a2
vle32.v v24, (a1)
sh2add a1, t0, a1
vle32.v v8, (a7)
sh2add a7, t0, a7
vor.vv v16, v24, v16
vor.vv v0, v8, v0
vse32.v v16, (a0)
sh2add a0, t0, a0
vse32.v v0, (a6)
sh2add a6, t0, a6
bnez a4, 1b
ret
endfunc
#endif