1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
/*
* Copyright 2010,2011,2012 Reality Jockey, Ltd.
* info@rjdj.me
* http://rjdj.me/
*
* This file is part of ZenGarden.
*
* ZenGarden 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 3 of the License, or
* (at your option) any later version.
*
* ZenGarden 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 ZenGarden. If not, see <http://www.gnu.org/licenses/>.
*
*/
#include "ArrayArithmetic.h"
#include "DspWrap.h"
message::Object *DspWrap::new_object(pd::Message *init_message, PdGraph *graph) {
return new DspWrap(init_message, graph);
}
DspWrap::DspWrap(pd::Message *init_message, PdGraph *graph) : DspObject(0, 1, 0, 1, graph) {
process_function = &processSignal;
}
DspWrap::~DspWrap() {
// nothing to do
}
void DspWrap::processSignal(DspObject *dspObject, int fromIndex, int n4) {
DspWrap *d = reinterpret_cast<DspWrap *>(dspObject);
// as no messages are received and there is only one inlet, processDsp does not need much of the
// infrastructure provided by DspObject
#if __SSE4__
float *input = d->dspBufferAtInlet[0];
float *output = d->dspBufferAtOutlet[0];
while (n4) {
__m128 in = _mm_load_ps(input);
_mm_store_ps(output, _mm_sub_ps(in, mm_floor_ps(in)));
input += 4; output += 4; n4 -= 4;
}
#else
// NOTE(mhroth): amazingly, this seemingly ghetto code is pretty fast. Compiler autovectorises it.
float *input = d->dspBufferAtInlet[0];
float *output = d->dspBufferAtOutlet[0];
for (int i = 0; i < n4; i+=4) {
float a = input[i]; float b = input[i+1]; float c = input[i+2]; float d = input[i+3];
output[i]=a-floorf(a); output[i+1]=b-floorf(b); output[i+2]=c-floorf(c); output[i+3]=d-floorf(d);
}
#endif
}