Chiebot-Mirror
/
FFmpeg
mirror of https://github.com/FFmpeg/FFmpeg.git
8
0
Fork 0
Code Issues Projects Releases Wiki Activity
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
113601 Commits
37 Branches
400 Tags
465 MiB
Tag: Branch: Tree: ce7c90ff82
Branches Tags
${ item.name }
Create tag ${ searchTerm }
Create branch ${ searchTerm }
from 'ce7c90ff82'
${ noResults }
FFmpeg/libavcodec/riscv/pixblockdsp_rvv.S

80 lines
2.3 KiB
Raw Normal View History Unescape

lavc/pixblockdsp: RISC-V V 8-bit get_pixels & get_pixels_unaligned
2 years ago
/*
* 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"
lavc/pixblockdsp: aligned R-V V 8-bit functions If the scan lines are aligned, we can load each row as a 64-bit value, thus avoiding segmentation. And then we can factor the conversion or subtraction. In principle, the same optimisation should be possible for high depth, but would require 128-bit elements, for which no FFmpeg CPU flag exists.
1 year ago
func ff_get_pixels_8_rvv, zve64x
vsetivli zero, 8, e8, mf2, ta, ma
li t0, 8 * 8
lavc/pixblockdsp: rework R-V V get_pixels_unaligned As in the aligned case, we can use VLSE64.V, though the way of doing so gets more convoluted, so the performance gains are more modest: get_pixels_unaligned_c: 126.7 get_pixels_unaligned_rvv_i32: 145.5 (before) get_pixels_unaligned_rvv_i64: 62.2 (after) For the reference, those are the aligned benchmarks (unchanged) on the same T-Head C908 hardware: get_pixels_c: 126.7 get_pixels_rvi: 85.7 get_pixels_rvv_i64: 33.2
1 year ago
1:
lavc/pixblockdsp: aligned R-V V 8-bit functions If the scan lines are aligned, we can load each row as a 64-bit value, thus avoiding segmentation. And then we can factor the conversion or subtraction. In principle, the same optimisation should be possible for high depth, but would require 128-bit elements, for which no FFmpeg CPU flag exists.
1 year ago
vlse64.v v16, (a1), a2
vsetvli zero, t0, e8, m4, ta, ma
vwcvtu.x.x.v v8, v16
vse16.v v8, (a0)
ret
endfunc
lavc/pixblockdsp: rework R-V V get_pixels_unaligned As in the aligned case, we can use VLSE64.V, though the way of doing so gets more convoluted, so the performance gains are more modest: get_pixels_unaligned_c: 126.7 get_pixels_unaligned_rvv_i32: 145.5 (before) get_pixels_unaligned_rvv_i64: 62.2 (after) For the reference, those are the aligned benchmarks (unchanged) on the same T-Head C908 hardware: get_pixels_c: 126.7 get_pixels_rvi: 85.7 get_pixels_rvv_i64: 33.2
1 year ago
func ff_get_pixels_unaligned_8_rvv, zve64x
andi t1, a1, 7
vsetivli zero, 8, e64, m4, ta, ma
li t0, 8 * 8
beqz t1, 1b
andi a1, a1, -8
slli t2, t1, 3
addi t1, a1, 8
sub t3, t0, t2
vlse64.v v16, (a1), a2
vlse64.v v24, (t1), a2
vsrl.vx v16, v16, t2
vsll.vx v24, v24, t3
vor.vv v16, v16, v24
vsetvli zero, t0, e8, m4, ta, ma
lavc/pixblockdsp: RISC-V V 8-bit get_pixels & get_pixels_unaligned
2 years ago
vwcvtu.x.x.v v8, v16
lavc/pixblockdsp: rework R-V V get_pixels_unaligned As in the aligned case, we can use VLSE64.V, though the way of doing so gets more convoluted, so the performance gains are more modest: get_pixels_unaligned_c: 126.7 get_pixels_unaligned_rvv_i32: 145.5 (before) get_pixels_unaligned_rvv_i64: 62.2 (after) For the reference, those are the aligned benchmarks (unchanged) on the same T-Head C908 hardware: get_pixels_c: 126.7 get_pixels_rvi: 85.7 get_pixels_rvv_i64: 33.2
1 year ago
vse16.v v8, (a0)
lavc/pixblockdsp: RISC-V V 8-bit get_pixels & get_pixels_unaligned
2 years ago
ret
endfunc
lavc/pixblockdsp: RISC-V V 16-bit get_pixels & get_pixels_unaligned
2 years ago
lavc/pixblockdsp: aligned R-V V 8-bit functions If the scan lines are aligned, we can load each row as a 64-bit value, thus avoiding segmentation. And then we can factor the conversion or subtraction. In principle, the same optimisation should be possible for high depth, but would require 128-bit elements, for which no FFmpeg CPU flag exists.
1 year ago
func ff_diff_pixels_rvv, zve64x
vsetivli zero, 8, e8, mf2, ta, ma
li t0, 8 * 8
vlse64.v v16, (a1), a3
vlse64.v v24, (a2), a3
vsetvli zero, t0, e8, m4, ta, ma
vwsubu.vv v8, v16, v24
vse16.v v8, (a0)
ret
endfunc
lavc/pixblockdsp: rename unaligned R-V V functions
1 year ago
func ff_diff_pixels_unaligned_rvv, zve32x
lavc/pixblockdsp: RISC-V diff_pixels & diff_pixels_unaligned
2 years ago
vsetivli zero, 8, e8, mf2, ta, ma
vlsseg8e8.v v16, (a1), a3
vlsseg8e8.v v24, (a2), a3
vwsubu.vv v8, v16, v24
vwsubu.vv v9, v17, v25
vwsubu.vv v10, v18, v26
vwsubu.vv v11, v19, v27
vwsubu.vv v12, v20, v28
vwsubu.vv v13, v21, v29
vwsubu.vv v14, v22, v30
vwsubu.vv v15, v23, v31
vsseg8e16.v v8, (a0)
ret
endfunc