2 * Copyright (c) 2016 Matthieu Bouron <matthieu.bouron stupeflix.com>
4 * This file is part of FFmpeg.
6 * FFmpeg is free software; you can redistribute it and/or
7 * modify it under the terms of the GNU Lesser General Public
8 * License as published by the Free Software Foundation; either
9 * version 2.1 of the License, or (at your option) any later version.
11 * FFmpeg is distributed in the hope that it will be useful,
12 * but WITHOUT ANY WARRANTY; without even the implied warranty of
13 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
14 * Lesser General Public License for more details.
16 * You should have received a copy of the GNU Lesser General Public
17 * License along with FFmpeg; if not, write to the Free Software
18 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
21 #include "libavutil/arm/asm.S"
23 function ff_resample_common_apply_filter_x4_float_neon, export=1
24 vmov.f32 q0, #0.0 @ accumulator
25 1: vld1.32 {q1}, [r1]! @ src[0..3]
26 vld1.32 {q2}, [r2]! @ filter[0..3]
27 vmla.f32 q0, q1, q2 @ accumulator += src[0..3] * filter[0..3]
28 subs r3, #4 @ filter_length -= 4
29 bgt 1b @ loop until filter_length
30 vpadd.f32 d0, d0, d1 @ pair adding of the 4x32-bit accumulated values
31 vpadd.f32 d0, d0, d0 @ pair adding of the 4x32-bit accumulator values
32 vst1.32 {d0[0]}, [r0] @ write accumulator
36 function ff_resample_common_apply_filter_x8_float_neon, export=1
37 vmov.f32 q0, #0.0 @ accumulator
38 1: vld1.32 {q1}, [r1]! @ src[0..3]
39 vld1.32 {q2}, [r2]! @ filter[0..3]
40 vld1.32 {q8}, [r1]! @ src[4..7]
41 vld1.32 {q9}, [r2]! @ filter[4..7]
42 vmla.f32 q0, q1, q2 @ accumulator += src[0..3] * filter[0..3]
43 vmla.f32 q0, q8, q9 @ accumulator += src[4..7] * filter[4..7]
44 subs r3, #8 @ filter_length -= 8
45 bgt 1b @ loop until filter_length
46 vpadd.f32 d0, d0, d1 @ pair adding of the 4x32-bit accumulated values
47 vpadd.f32 d0, d0, d0 @ pair adding of the 4x32-bit accumulator values
48 vst1.32 {d0[0]}, [r0] @ write accumulator
52 function ff_resample_common_apply_filter_x4_s16_neon, export=1
53 vmov.s32 q0, #0 @ accumulator
54 1: vld1.16 {d2}, [r1]! @ src[0..3]
55 vld1.16 {d4}, [r2]! @ filter[0..3]
56 vmlal.s16 q0, d2, d4 @ accumulator += src[0..3] * filter[0..3]
57 subs r3, #4 @ filter_length -= 4
58 bgt 1b @ loop until filter_length
59 vpadd.s32 d0, d0, d1 @ pair adding of the 4x32-bit accumulated values
60 vpadd.s32 d0, d0, d0 @ pair adding of the 4x32-bit accumulator values
61 vst1.32 {d0[0]}, [r0] @ write accumulator
65 function ff_resample_common_apply_filter_x8_s16_neon, export=1
66 vmov.s32 q0, #0 @ accumulator
67 1: vld1.16 {q1}, [r1]! @ src[0..7]
68 vld1.16 {q2}, [r2]! @ filter[0..7]
69 vmlal.s16 q0, d2, d4 @ accumulator += src[0..3] * filter[0..3]
70 vmlal.s16 q0, d3, d5 @ accumulator += src[4..7] * filter[4..7]
71 subs r3, #8 @ filter_length -= 8
72 bgt 1b @ loop until filter_length
73 vpadd.s32 d0, d0, d1 @ pair adding of the 4x32-bit accumulated values
74 vpadd.s32 d0, d0, d0 @ pair adding of the 4x32-bit accumulator values
75 vst1.32 {d0[0]}, [r0] @ write accumulator