4 * Copyright (c) 2004-2012 Michael Niedermayer <michaelni@gmx.at>
6 * This file is part of FFmpeg.
8 * FFmpeg is free software; you can redistribute it and/or
9 * modify it under the terms of the GNU Lesser General Public
10 * License as published by the Free Software Foundation; either
11 * version 2.1 of the License, or (at your option) any later version.
13 * FFmpeg is distributed in the hope that it will be useful,
14 * but WITHOUT ANY WARRANTY; without even the implied warranty of
15 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
16 * Lesser General Public License for more details.
18 * You should have received a copy of the GNU Lesser General Public
19 * License along with FFmpeg; if not, write to the Free Software
20 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
25 #include "libavutil/cpu.h"
26 #include "libavutil/avassert.h"
28 #include "libavutil/aarch64/cpu.h"
29 #include "libswresample/resample.h"
31 #define DECLARE_RESAMPLE_COMMON_TEMPLATE(TYPE, DELEM, FELEM, FELEM2, OUT) \
33 void ff_resample_common_apply_filter_x4_##TYPE##_neon(FELEM2 *acc, const DELEM *src, \
34 const FELEM *filter, int length); \
36 void ff_resample_common_apply_filter_x8_##TYPE##_neon(FELEM2 *acc, const DELEM *src, \
37 const FELEM *filter, int length); \
39 static int ff_resample_common_##TYPE##_neon(ResampleContext *c, void *dest, const void *source, \
40 int n, int update_ctx) \
43 const DELEM *src = source; \
45 int index = c->index; \
47 int sample_index = 0; \
48 int x4_aligned_filter_length = c->filter_length & ~3; \
49 int x8_aligned_filter_length = c->filter_length & ~7; \
51 while (index >= c->phase_count) { \
53 index -= c->phase_count; \
56 for (dst_index = 0; dst_index < n; dst_index++) { \
57 FELEM *filter = ((FELEM *) c->filter_bank) + c->filter_alloc * index; \
61 if (x8_aligned_filter_length >= 8) { \
62 ff_resample_common_apply_filter_x8_##TYPE##_neon(&val, &src[sample_index], \
63 filter, x8_aligned_filter_length); \
64 i += x8_aligned_filter_length; \
66 } else if (x4_aligned_filter_length >= 4) { \
67 ff_resample_common_apply_filter_x4_##TYPE##_neon(&val, &src[sample_index], \
68 filter, x4_aligned_filter_length); \
69 i += x4_aligned_filter_length; \
71 for (; i < c->filter_length; i++) { \
72 val += src[sample_index + i] * (FELEM2)filter[i]; \
74 OUT(dst[dst_index], val); \
76 frac += c->dst_incr_mod; \
77 index += c->dst_incr_div; \
78 if (frac >= c->src_incr) { \
79 frac -= c->src_incr; \
83 while (index >= c->phase_count) { \
85 index -= c->phase_count; \
94 return sample_index; \
97 #define OUT(d, v) d = v
98 DECLARE_RESAMPLE_COMMON_TEMPLATE(float, float, float, float, OUT)
101 #define OUT(d, v) (v) = ((v) + (1<<(14)))>>15; (d) = av_clip_int16(v)
102 DECLARE_RESAMPLE_COMMON_TEMPLATE(s16, int16_t, int16_t, int32_t, OUT)
105 av_cold void swri_resample_dsp_aarch64_init(ResampleContext *c)
107 int cpu_flags = av_get_cpu_flags();
109 if (!have_neon(cpu_flags))
113 case AV_SAMPLE_FMT_FLTP:
114 c->dsp.resample_common = ff_resample_common_float_neon;
116 case AV_SAMPLE_FMT_S16P:
117 c->dsp.resample_common = ff_resample_common_s16_neon;