virtualenv/FFmpeg
1
0
Fork 0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2024-12-18 03:19:31 +02:00
Code Issues Releases Activity
740ce93fae
FFmpeg/libavcodec/riscv/opusdsp_rvv.S

76 lines
2.2 KiB
ArmAsm
Raw Normal View History

lavc/opusdsp: RISC-V V (128-bit) postfilter This is implemented for a vector size of 128-bit. Since the scalar product in the inner loop covers 5 samples or 160 bits, we need a group multipler of 2. To avoid reconfiguring the vector type, the outer loop, which loads multiple input samples sticks to the same multipler. Consequently, the outer loop loads 8 samples per iteration. This is safe since the minimum period of the CELT codec is 15 samples. The same code would also work, albeit needlessly inefficiently with a vector length of 256 bits. A proper implementation will follow instead.
2022-10-05 18:12:53 +02:00
/*
* 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"
func ff_opus_postfilter_rvv_128, zve32f
lavc/opusdsp: RISC-V V (256-bit) postfilter This adds a variant of the postfilter for use with 256-bit vectors. As a single vector is then large enough to perform the scalar product, the group multipler is reduced to just one at run-time. The different vector type is passed via register. Unfortunately, there is no VSETIVL instruction, so the constant vector size (5) also needs to be passed via a register.
2022-10-05 18:12:55 +02:00
lvtypei a5, e32, m2, ta, ma
j 1f
endfunc
lavc/opusdsp: RISC-V V (512-bit) postfilter This adds a variant of the postfilter for use with 512-bit vectors. Half a vector is enough to perform the scalar product. Normally a whole vector would be used anyhow. Indeed fractional multiplers are no faster than the unit multipler. But in this particular function, a full vector makes up 16 samples, which would be loaded at each iteration of the outer loop. The minimum guaranteed CELT postfilter period is only 15. Accounting for the edges, we can only safely preload up to 13 samples. The fractional multipler is thus used to cap the selected vector length to a safe value of 8 elements or 256 bits. Likewise, we have the 1024-bit variant with the quarter multipler. In theory, a 2048-bit one would be possible with the eigth multipler, but that length is not even defined in the specifications as of yet, nor is it supported by any emulator - forget actual hardware.
2022-10-05 18:12:56 +02:00
func ff_opus_postfilter_rvv_512, zve32f
lvtypei a5, e32, mf2, ta, ma
j 1f
endfunc
func ff_opus_postfilter_rvv_1024, zve32f
lvtypei a5, e32, mf4, ta, ma
j 1f
endfunc
lavc/opusdsp: RISC-V V (256-bit) postfilter This adds a variant of the postfilter for use with 256-bit vectors. As a single vector is then large enough to perform the scalar product, the group multipler is reduced to just one at run-time. The different vector type is passed via register. Unfortunately, there is no VSETIVL instruction, so the constant vector size (5) also needs to be passed via a register.
2022-10-05 18:12:55 +02:00
func ff_opus_postfilter_rvv_256, zve32f
lvtypei a5, e32, m1, ta, ma
1:
li a4, 5
lavc/opusdsp: RISC-V V (128-bit) postfilter This is implemented for a vector size of 128-bit. Since the scalar product in the inner loop covers 5 samples or 160 bits, we need a group multipler of 2. To avoid reconfiguring the vector type, the outer loop, which loads multiple input samples sticks to the same multipler. Consequently, the outer loop loads 8 samples per iteration. This is safe since the minimum period of the CELT codec is 15 samples. The same code would also work, albeit needlessly inefficiently with a vector length of 256 bits. A proper implementation will follow instead.
2022-10-05 18:12:53 +02:00
addi a1, a1, 2
slli a1, a1, 2
lw t1, 4(a2)
vsetivli zero, 3, e32, m1, ta, ma
vle32.v v24, (a2)
sub a1, a0, a1 // a1 = &x4 = &data[-(period + 2)]
lavc/opusdsp: RISC-V V (256-bit) postfilter This adds a variant of the postfilter for use with 256-bit vectors. As a single vector is then large enough to perform the scalar product, the group multipler is reduced to just one at run-time. The different vector type is passed via register. Unfortunately, there is no VSETIVL instruction, so the constant vector size (5) also needs to be passed via a register.
2022-10-05 18:12:55 +02:00
vsetvl zero, a4, a5
lavc/opusdsp: RISC-V V (128-bit) postfilter This is implemented for a vector size of 128-bit. Since the scalar product in the inner loop covers 5 samples or 160 bits, we need a group multipler of 2. To avoid reconfiguring the vector type, the outer loop, which loads multiple input samples sticks to the same multipler. Consequently, the outer loop loads 8 samples per iteration. This is safe since the minimum period of the CELT codec is 15 samples. The same code would also work, albeit needlessly inefficiently with a vector length of 256 bits. A proper implementation will follow instead.
2022-10-05 18:12:53 +02:00
vslide1up.vx v8, v24, t1
lw t2, 8(a2)
vle32.v v16, (a1)
vslide1up.vx v24, v8, t2 // v24 = { g[2], g[1], g[0], g[1], g[2] }
2:
lavc/opusdsp: RISC-V V (256-bit) postfilter This adds a variant of the postfilter for use with 256-bit vectors. As a single vector is then large enough to perform the scalar product, the group multipler is reduced to just one at run-time. The different vector type is passed via register. Unfortunately, there is no VSETIVL instruction, so the constant vector size (5) also needs to be passed via a register.
2022-10-05 18:12:55 +02:00
vsetvl t0, a3, a5
lavc/opusdsp: RISC-V V (128-bit) postfilter This is implemented for a vector size of 128-bit. Since the scalar product in the inner loop covers 5 samples or 160 bits, we need a group multipler of 2. To avoid reconfiguring the vector type, the outer loop, which loads multiple input samples sticks to the same multipler. Consequently, the outer loop loads 8 samples per iteration. This is safe since the minimum period of the CELT codec is 15 samples. The same code would also work, albeit needlessly inefficiently with a vector length of 256 bits. A proper implementation will follow instead.
2022-10-05 18:12:53 +02:00
vle32.v v0, (a0)
sub a3, a3, t0
3:
lavc/opusdsp: RISC-V V (256-bit) postfilter This adds a variant of the postfilter for use with 256-bit vectors. As a single vector is then large enough to perform the scalar product, the group multipler is reduced to just one at run-time. The different vector type is passed via register. Unfortunately, there is no VSETIVL instruction, so the constant vector size (5) also needs to be passed via a register.
2022-10-05 18:12:55 +02:00
vsetvl zero, a4, a5
lavc/opusdsp: RISC-V V (128-bit) postfilter This is implemented for a vector size of 128-bit. Since the scalar product in the inner loop covers 5 samples or 160 bits, we need a group multipler of 2. To avoid reconfiguring the vector type, the outer loop, which loads multiple input samples sticks to the same multipler. Consequently, the outer loop loads 8 samples per iteration. This is safe since the minimum period of the CELT codec is 15 samples. The same code would also work, albeit needlessly inefficiently with a vector length of 256 bits. A proper implementation will follow instead.
2022-10-05 18:12:53 +02:00
lw t2, 20(a1)
vfmul.vv v8, v24, v16
addi a0, a0, 4
vslide1down.vx v16, v16, t2
addi a1, a1, 4
vfredusum.vs v0, v8, v0
lavc/opusdsp: RISC-V V (256-bit) postfilter This adds a variant of the postfilter for use with 256-bit vectors. As a single vector is then large enough to perform the scalar product, the group multipler is reduced to just one at run-time. The different vector type is passed via register. Unfortunately, there is no VSETIVL instruction, so the constant vector size (5) also needs to be passed via a register.
2022-10-05 18:12:55 +02:00
vsetvl zero, t0, a5
lavc/opusdsp: RISC-V V (128-bit) postfilter This is implemented for a vector size of 128-bit. Since the scalar product in the inner loop covers 5 samples or 160 bits, we need a group multipler of 2. To avoid reconfiguring the vector type, the outer loop, which loads multiple input samples sticks to the same multipler. Consequently, the outer loop loads 8 samples per iteration. This is safe since the minimum period of the CELT codec is 15 samples. The same code would also work, albeit needlessly inefficiently with a vector length of 256 bits. A proper implementation will follow instead.
2022-10-05 18:12:53 +02:00
vmv.x.s t1, v0
addi t0, t0, -1
vslide1down.vx v0, v0, zero
sw t1, -4(a0)
bnez t0, 3b
bnez a3, 2b
ret
endfunc
Reference in New Issue Copy Permalink