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
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
|
#include "libaudio.h"
#include "synthInternals.h"
#include "stddef.h"
#ifndef MIN
#define MIN(a, b) (((a) < (b)) ? (a) : (b))
#endif
static s32 __nextSampleTime(ALSynth* drvr, ALPlayer** client);
static s32 _timeToSamplesNoRound(ALSynth* synth, s32 micros);
void alSynNew(ALSynth* drvr, ALSynConfig* c) {
s32 i;
ALVoice* vv;
PVoice* pv;
ALVoice* vvoices;
PVoice* pvoices;
ALHeap* hp = c->heap;
ALSave* save;
ALFilter* sources;
ALParam* params;
ALParam* paramPtr;
drvr->head = NULL;
drvr->numPVoices = c->maxPVoices;
drvr->curSamples = 0;
drvr->paramSamples = 0;
drvr->outputRate = c->outputRate;
drvr->maxOutSamples = AL_MAX_RSP_SAMPLES;
drvr->dma = (ALDMANew)c->dmaproc;
save = alHeapAlloc(hp, 1, sizeof(ALSave));
alSaveNew(save);
drvr->outputFilter = &save->filter;
// allocate and initialize the auxilliary effects bus. at present we only support 1 effects bus.
drvr->auxBus = alHeapAlloc(hp, 1, sizeof(ALAuxBus));
drvr->maxAuxBusses = 1;
sources = alHeapAlloc(hp, c->maxPVoices, sizeof(ALFilter*));
alAuxBusNew(drvr->auxBus, sources, c->maxPVoices);
// allocate and initialize the main bus.
drvr->mainBus = alHeapAlloc(hp, 1, sizeof(ALMainBus));
sources = alHeapAlloc(hp, c->maxPVoices, sizeof(ALFilter*));
alMainBusNew(drvr->mainBus, sources, c->maxPVoices);
if (c->fxType != AL_FX_NONE) {
// Allocate an effect and set parameters
alSynAllocFX(drvr, 0, c, hp);
} else {
// Connect the aux bus to the main bus
alMainBusParam(drvr->mainBus, AL_FILTER_ADD_SOURCE, &drvr->auxBus[0]);
}
// Build the physical voice lists
drvr->pFreeList.next = NULL;
drvr->pFreeList.prev = NULL;
drvr->pLameList.next = NULL;
drvr->pLameList.prev = NULL;
drvr->pAllocList.next = NULL;
drvr->pAllocList.prev = NULL;
pvoices = alHeapAlloc(hp, c->maxPVoices, sizeof(PVoice));
for (i = 0; i < c->maxPVoices; i++) {
pv = &pvoices[i];
alLink(&pv->node, &drvr->pFreeList);
pv->vvoice = NULL;
alLoadNew(&pv->decoder, drvr->dma, hp);
alLoadParam(&pv->decoder, AL_FILTER_SET_SOURCE, NULL);
alResampleNew(&pv->resampler, hp);
alResampleParam(&pv->resampler, AL_FILTER_SET_SOURCE, &pv->decoder);
alEnvmixerNew(&pv->envmixer, hp);
alEnvmixerParam(&pv->envmixer, AL_FILTER_SET_SOURCE, &pv->resampler);
alAuxBusParam(drvr->auxBus, AL_FILTER_ADD_SOURCE, &pv->envmixer);
pv->channelKnob = &pv->envmixer.filter;
}
alSaveParam(save, AL_FILTER_SET_SOURCE, drvr->mainBus);
// build the parameter update list
params = alHeapAlloc(hp, c->maxUpdates, sizeof(ALParam));
drvr->paramList = NULL;
for (i = 0; i < c->maxUpdates; i++) {
paramPtr = ¶ms[i];
paramPtr->next = drvr->paramList;
drvr->paramList = paramPtr;
}
drvr->heap = hp;
}
/**
* slAudioFrame() is called every video frame, and is based on the video
* frame interrupt. It is assumed to be an accurate time source for the
* clients.
*/
Acmd* alAudioFrame(Acmd* cmdList, s32* cmdLen, s16* outBuf, s32 outLen) {
ALPlayer* client;
ALFilter* output;
ALSynth* drvr = &alGlobals->drvr;
s16 tmp = 0; // Starting buffer in DMEM
Acmd* cmdlEnd = cmdList;
Acmd* cmdPtr;
s32 nOut;
s16* lOutBuf = outBuf;
if (drvr->head == NULL) {
// nothing to do
*cmdLen = 0;
return cmdList;
}
// run down list of clients and execute callback if needed this
// subframe. Here we do all the work for the frame at the
// start. Time offsets that occur before the next frame are
// executed "early".
//
// paramSamples = time of next parameter change.
// curSamples = current sample time.
// so paramSamples - curSamples is the time until the next parameter change.
// if the next parameter change occurs within this frame time (outLen),
// then call back the client that contains the parameter change.
// Note, paramSamples must be rounded down to 16 sample boundary for use
// during the client handler.
for (drvr->paramSamples = __nextSampleTime(drvr, &client); drvr->paramSamples - drvr->curSamples < outLen;
drvr->paramSamples = __nextSampleTime(drvr, &client)) {
drvr->paramSamples &= ~0xF;
client->samplesLeft += _timeToSamplesNoRound(drvr, (*client->handler)(client));
}
// for safety's sake, always store paramSamples aligned to 16 sample boundary.
// this way, if an voice handler routine gets called outside the ALVoiceHandler
// routine (alSynAllocVoice) it will get timestamped with an aligned value and
// will be processed immediately next audio frame.
drvr->paramSamples &= ~0xF;
// Now build the command list in small chunks
while (outLen > 0) {
nOut = MIN(drvr->maxOutSamples, outLen);
// construct the command list for each physical voice by calling the head of the filter chain.
cmdPtr = cmdlEnd;
aSegment(cmdPtr++, 0, 0);
output = drvr->outputFilter;
(*output->setParam)(output, AL_FILTER_SET_DRAM, lOutBuf);
cmdlEnd = (*output->handler)(output, &tmp, nOut, drvr->curSamples, cmdPtr);
outLen -= nOut;
lOutBuf += nOut << 1; // For Stereo
drvr->curSamples += nOut;
}
*cmdLen = (s32)(cmdlEnd - cmdList);
_collectPVoices(drvr); // collect free physical voices
return cmdlEnd;
}
ALParam* __allocParam(void) {
ALParam* update = NULL;
ALSynth* drvr = &alGlobals->drvr;
if (drvr->paramList != NULL) {
update = drvr->paramList;
drvr->paramList = drvr->paramList->next;
update->next = NULL;
}
return update;
}
void __freeParam(ALParam* param) {
ALSynth* drvr = &alGlobals->drvr;
param->next = drvr->paramList;
drvr->paramList = param;
}
void _collectPVoices(ALSynth* drvr) {
ALLink* dl;
PVoice* pv;
while ((dl = drvr->pLameList.next) != NULL) {
pv = (PVoice*)dl;
alUnlink(&pv->node);
alLink(&pv->node, &drvr->pFreeList);
}
}
void _freePVoice(ALSynth* drvr, PVoice* pvoice) {
// move the voice from the allocated list to the lame list
alUnlink(&pvoice->node);
alLink(&pvoice->node, &drvr->pLameList);
}
static s32 _timeToSamplesNoRound(ALSynth* synth, s32 micros) {
// Add 0.5 to adjust the average affect of the truncation error produced by casting a float to an int.
f32 tmp = ((f32)micros) * synth->outputRate / 1000000.0 + 0.5;
return (s32)tmp;
}
s32 _timeToSamples(ALSynth* synth, s32 micros) {
return _timeToSamplesNoRound(synth, micros) & ~0xF;
}
static s32 __nextSampleTime(ALSynth* drvr, ALPlayer** client) {
ALMicroTime delta = 0x7FFFFFFF; // max delta for s32
ALPlayer* cl;
*client = NULL;
for (cl = drvr->head; cl != NULL; cl = cl->next) {
if (cl->samplesLeft - drvr->curSamples < delta) {
*client = cl;
delta = cl->samplesLeft - drvr->curSamples;
}
}
return (*client)->samplesLeft;
}
|