Commit | Line | Data |
---|---|---|
1da177e4 LT |
1 | <!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook V4.1//EN"> |
2 | ||
3 | <book> | |
4 | <?dbhtml filename="index.html"> | |
5 | ||
6 | <!-- ****************************************************** --> | |
7 | <!-- Header --> | |
8 | <!-- ****************************************************** --> | |
9 | <bookinfo> | |
10 | <title>Writing an ALSA Driver</title> | |
11 | <author> | |
12 | <firstname>Takashi</firstname> | |
13 | <surname>Iwai</surname> | |
14 | <affiliation> | |
15 | <address> | |
16 | <email>tiwai@suse.de</email> | |
17 | </address> | |
18 | </affiliation> | |
19 | </author> | |
20 | ||
5fe76e4d TI |
21 | <date>November 17, 2005</date> |
22 | <edition>0.3.6</edition> | |
1da177e4 LT |
23 | |
24 | <abstract> | |
25 | <para> | |
26 | This document describes how to write an ALSA (Advanced Linux | |
27 | Sound Architecture) driver. | |
28 | </para> | |
29 | </abstract> | |
30 | ||
31 | <legalnotice> | |
32 | <para> | |
7c22f1aa | 33 | Copyright (c) 2002-2005 Takashi Iwai <email>tiwai@suse.de</email> |
1da177e4 LT |
34 | </para> |
35 | ||
36 | <para> | |
37 | This document is free; you can redistribute it and/or modify it | |
38 | under the terms of the GNU General Public License as published by | |
39 | the Free Software Foundation; either version 2 of the License, or | |
40 | (at your option) any later version. | |
41 | </para> | |
42 | ||
43 | <para> | |
44 | This document is distributed in the hope that it will be useful, | |
45 | but <emphasis>WITHOUT ANY WARRANTY</emphasis>; without even the | |
46 | implied warranty of <emphasis>MERCHANTABILITY or FITNESS FOR A | |
47 | PARTICULAR PURPOSE</emphasis>. See the GNU General Public License | |
48 | for more details. | |
49 | </para> | |
50 | ||
51 | <para> | |
52 | You should have received a copy of the GNU General Public | |
53 | License along with this program; if not, write to the Free | |
54 | Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, | |
55 | MA 02111-1307 USA | |
56 | </para> | |
57 | </legalnotice> | |
58 | ||
59 | </bookinfo> | |
60 | ||
61 | <!-- ****************************************************** --> | |
62 | <!-- Preface --> | |
63 | <!-- ****************************************************** --> | |
64 | <preface id="preface"> | |
65 | <title>Preface</title> | |
66 | <para> | |
67 | This document describes how to write an | |
68 | <ulink url="http://www.alsa-project.org/"><citetitle> | |
69 | ALSA (Advanced Linux Sound Architecture)</citetitle></ulink> | |
70 | driver. The document focuses mainly on the PCI soundcard. | |
71 | In the case of other device types, the API might | |
72 | be different, too. However, at least the ALSA kernel API is | |
73 | consistent, and therefore it would be still a bit help for | |
74 | writing them. | |
75 | </para> | |
76 | ||
77 | <para> | |
78 | The target of this document is ones who already have enough | |
79 | skill of C language and have the basic knowledge of linux | |
80 | kernel programming. This document doesn't explain the general | |
81 | topics of linux kernel codes and doesn't cover the detail of | |
82 | implementation of each low-level driver. It describes only how is | |
83 | the standard way to write a PCI sound driver on ALSA. | |
84 | </para> | |
85 | ||
86 | <para> | |
87 | If you are already familiar with the older ALSA ver.0.5.x, you | |
88 | can check the drivers such as <filename>es1938.c</filename> or | |
89 | <filename>maestro3.c</filename> which have also almost the same | |
90 | code-base in the ALSA 0.5.x tree, so you can compare the differences. | |
91 | </para> | |
92 | ||
93 | <para> | |
94 | This document is still a draft version. Any feedbacks and | |
95 | corrections, please!! | |
96 | </para> | |
97 | </preface> | |
98 | ||
99 | ||
100 | <!-- ****************************************************** --> | |
101 | <!-- File Tree Structure --> | |
102 | <!-- ****************************************************** --> | |
103 | <chapter id="file-tree"> | |
104 | <title>File Tree Structure</title> | |
105 | ||
106 | <section id="file-tree-general"> | |
107 | <title>General</title> | |
108 | <para> | |
109 | The ALSA drivers are provided in the two ways. | |
110 | </para> | |
111 | ||
112 | <para> | |
113 | One is the trees provided as a tarball or via cvs from the | |
114 | ALSA's ftp site, and another is the 2.6 (or later) Linux kernel | |
115 | tree. To synchronize both, the ALSA driver tree is split into | |
116 | two different trees: alsa-kernel and alsa-driver. The former | |
117 | contains purely the source codes for the Linux 2.6 (or later) | |
118 | tree. This tree is designed only for compilation on 2.6 or | |
119 | later environment. The latter, alsa-driver, contains many subtle | |
120 | files for compiling the ALSA driver on the outside of Linux | |
121 | kernel like configure script, the wrapper functions for older, | |
122 | 2.2 and 2.4 kernels, to adapt the latest kernel API, | |
123 | and additional drivers which are still in development or in | |
124 | tests. The drivers in alsa-driver tree will be moved to | |
125 | alsa-kernel (eventually 2.6 kernel tree) once when they are | |
126 | finished and confirmed to work fine. | |
127 | </para> | |
128 | ||
129 | <para> | |
130 | The file tree structure of ALSA driver is depicted below. Both | |
131 | alsa-kernel and alsa-driver have almost the same file | |
132 | structure, except for <quote>core</quote> directory. It's | |
133 | named as <quote>acore</quote> in alsa-driver tree. | |
134 | ||
135 | <example> | |
136 | <title>ALSA File Tree Structure</title> | |
137 | <literallayout> | |
138 | sound | |
139 | /core | |
140 | /oss | |
141 | /seq | |
142 | /oss | |
143 | /instr | |
144 | /ioctl32 | |
145 | /include | |
146 | /drivers | |
147 | /mpu401 | |
148 | /opl3 | |
149 | /i2c | |
150 | /l3 | |
151 | /synth | |
152 | /emux | |
153 | /pci | |
154 | /(cards) | |
155 | /isa | |
156 | /(cards) | |
157 | /arm | |
158 | /ppc | |
159 | /sparc | |
160 | /usb | |
161 | /pcmcia /(cards) | |
162 | /oss | |
163 | </literallayout> | |
164 | </example> | |
165 | </para> | |
166 | </section> | |
167 | ||
168 | <section id="file-tree-core-directory"> | |
169 | <title>core directory</title> | |
170 | <para> | |
171 | This directory contains the middle layer, that is, the heart | |
172 | of ALSA drivers. In this directory, the native ALSA modules are | |
173 | stored. The sub-directories contain different modules and are | |
174 | dependent upon the kernel config. | |
175 | </para> | |
176 | ||
177 | <section id="file-tree-core-directory-oss"> | |
178 | <title>core/oss</title> | |
179 | ||
180 | <para> | |
181 | The codes for PCM and mixer OSS emulation modules are stored | |
182 | in this directory. The rawmidi OSS emulation is included in | |
183 | the ALSA rawmidi code since it's quite small. The sequencer | |
184 | code is stored in core/seq/oss directory (see | |
185 | <link linkend="file-tree-core-directory-seq-oss"><citetitle> | |
186 | below</citetitle></link>). | |
187 | </para> | |
188 | </section> | |
189 | ||
190 | <section id="file-tree-core-directory-ioctl32"> | |
191 | <title>core/ioctl32</title> | |
192 | ||
193 | <para> | |
194 | This directory contains the 32bit-ioctl wrappers for 64bit | |
195 | architectures such like x86-64, ppc64 and sparc64. For 32bit | |
196 | and alpha architectures, these are not compiled. | |
197 | </para> | |
198 | </section> | |
199 | ||
200 | <section id="file-tree-core-directory-seq"> | |
201 | <title>core/seq</title> | |
202 | <para> | |
203 | This and its sub-directories are for the ALSA | |
204 | sequencer. This directory contains the sequencer core and | |
205 | primary sequencer modules such like snd-seq-midi, | |
206 | snd-seq-virmidi, etc. They are compiled only when | |
207 | <constant>CONFIG_SND_SEQUENCER</constant> is set in the kernel | |
208 | config. | |
209 | </para> | |
210 | </section> | |
211 | ||
212 | <section id="file-tree-core-directory-seq-oss"> | |
213 | <title>core/seq/oss</title> | |
214 | <para> | |
215 | This contains the OSS sequencer emulation codes. | |
216 | </para> | |
217 | </section> | |
218 | ||
219 | <section id="file-tree-core-directory-deq-instr"> | |
220 | <title>core/seq/instr</title> | |
221 | <para> | |
222 | This directory contains the modules for the sequencer | |
223 | instrument layer. | |
224 | </para> | |
225 | </section> | |
226 | </section> | |
227 | ||
228 | <section id="file-tree-include-directory"> | |
229 | <title>include directory</title> | |
230 | <para> | |
231 | This is the place for the public header files of ALSA drivers, | |
232 | which are to be exported to the user-space, or included by | |
233 | several files at different directories. Basically, the private | |
234 | header files should not be placed in this directory, but you may | |
235 | still find files there, due to historical reason :) | |
236 | </para> | |
237 | </section> | |
238 | ||
239 | <section id="file-tree-drivers-directory"> | |
240 | <title>drivers directory</title> | |
241 | <para> | |
242 | This directory contains the codes shared among different drivers | |
243 | on the different architectures. They are hence supposed not to be | |
244 | architecture-specific. | |
245 | For example, the dummy pcm driver and the serial MIDI | |
246 | driver are found in this directory. In the sub-directories, | |
247 | there are the codes for components which are independent from | |
248 | bus and cpu architectures. | |
249 | </para> | |
250 | ||
251 | <section id="file-tree-drivers-directory-mpu401"> | |
252 | <title>drivers/mpu401</title> | |
253 | <para> | |
254 | The MPU401 and MPU401-UART modules are stored here. | |
255 | </para> | |
256 | </section> | |
257 | ||
258 | <section id="file-tree-drivers-directory-opl3"> | |
259 | <title>drivers/opl3 and opl4</title> | |
260 | <para> | |
261 | The OPL3 and OPL4 FM-synth stuff is found here. | |
262 | </para> | |
263 | </section> | |
264 | </section> | |
265 | ||
266 | <section id="file-tree-i2c-directory"> | |
267 | <title>i2c directory</title> | |
268 | <para> | |
269 | This contains the ALSA i2c components. | |
270 | </para> | |
271 | ||
272 | <para> | |
273 | Although there is a standard i2c layer on Linux, ALSA has its | |
274 | own i2c codes for some cards, because the soundcard needs only a | |
275 | simple operation and the standard i2c API is too complicated for | |
276 | such a purpose. | |
277 | </para> | |
278 | ||
279 | <section id="file-tree-i2c-directory-l3"> | |
280 | <title>i2c/l3</title> | |
281 | <para> | |
282 | This is a sub-directory for ARM L3 i2c. | |
283 | </para> | |
284 | </section> | |
285 | </section> | |
286 | ||
287 | <section id="file-tree-synth-directory"> | |
288 | <title>synth directory</title> | |
289 | <para> | |
290 | This contains the synth middle-level modules. | |
291 | </para> | |
292 | ||
293 | <para> | |
294 | So far, there is only Emu8000/Emu10k1 synth driver under | |
295 | synth/emux sub-directory. | |
296 | </para> | |
297 | </section> | |
298 | ||
299 | <section id="file-tree-pci-directory"> | |
300 | <title>pci directory</title> | |
301 | <para> | |
302 | This and its sub-directories hold the top-level card modules | |
303 | for PCI soundcards and the codes specific to the PCI BUS. | |
304 | </para> | |
305 | ||
306 | <para> | |
307 | The drivers compiled from a single file is stored directly on | |
308 | pci directory, while the drivers with several source files are | |
309 | stored on its own sub-directory (e.g. emu10k1, ice1712). | |
310 | </para> | |
311 | </section> | |
312 | ||
313 | <section id="file-tree-isa-directory"> | |
314 | <title>isa directory</title> | |
315 | <para> | |
316 | This and its sub-directories hold the top-level card modules | |
317 | for ISA soundcards. | |
318 | </para> | |
319 | </section> | |
320 | ||
321 | <section id="file-tree-arm-ppc-sparc-directories"> | |
322 | <title>arm, ppc, and sparc directories</title> | |
323 | <para> | |
324 | These are for the top-level card modules which are | |
325 | specific to each given architecture. | |
326 | </para> | |
327 | </section> | |
328 | ||
329 | <section id="file-tree-usb-directory"> | |
330 | <title>usb directory</title> | |
331 | <para> | |
332 | This contains the USB-audio driver. On the latest version, the | |
333 | USB MIDI driver is integrated together with usb-audio driver. | |
334 | </para> | |
335 | </section> | |
336 | ||
337 | <section id="file-tree-pcmcia-directory"> | |
338 | <title>pcmcia directory</title> | |
339 | <para> | |
340 | The PCMCIA, especially PCCard drivers will go here. CardBus | |
341 | drivers will be on pci directory, because its API is identical | |
342 | with the standard PCI cards. | |
343 | </para> | |
344 | </section> | |
345 | ||
346 | <section id="file-tree-oss-directory"> | |
347 | <title>oss directory</title> | |
348 | <para> | |
349 | The OSS/Lite source files are stored here on Linux 2.6 (or | |
350 | later) tree. (In the ALSA driver tarball, it's empty, of course :) | |
351 | </para> | |
352 | </section> | |
353 | </chapter> | |
354 | ||
355 | ||
356 | <!-- ****************************************************** --> | |
357 | <!-- Basic Flow for PCI Drivers --> | |
358 | <!-- ****************************************************** --> | |
359 | <chapter id="basic-flow"> | |
360 | <title>Basic Flow for PCI Drivers</title> | |
361 | ||
362 | <section id="basic-flow-outline"> | |
363 | <title>Outline</title> | |
364 | <para> | |
365 | The minimum flow of PCI soundcard is like the following: | |
366 | ||
367 | <itemizedlist> | |
368 | <listitem><para>define the PCI ID table (see the section | |
369 | <link linkend="pci-resource-entries"><citetitle>PCI Entries | |
370 | </citetitle></link>).</para></listitem> | |
371 | <listitem><para>create <function>probe()</function> callback.</para></listitem> | |
372 | <listitem><para>create <function>remove()</function> callback.</para></listitem> | |
373 | <listitem><para>create pci_driver table which contains the three pointers above.</para></listitem> | |
01d25d46 | 374 | <listitem><para>create <function>init()</function> function just calling <function>pci_register_driver()</function> to register the pci_driver table defined above.</para></listitem> |
1da177e4 LT |
375 | <listitem><para>create <function>exit()</function> function to call <function>pci_unregister_driver()</function> function.</para></listitem> |
376 | </itemizedlist> | |
377 | </para> | |
378 | </section> | |
379 | ||
380 | <section id="basic-flow-example"> | |
381 | <title>Full Code Example</title> | |
382 | <para> | |
383 | The code example is shown below. Some parts are kept | |
384 | unimplemented at this moment but will be filled in the | |
385 | succeeding sections. The numbers in comment lines of | |
386 | <function>snd_mychip_probe()</function> function are the | |
387 | markers. | |
388 | ||
389 | <example> | |
390 | <title>Basic Flow for PCI Drivers Example</title> | |
391 | <programlisting> | |
392 | <![CDATA[ | |
393 | #include <sound/driver.h> | |
394 | #include <linux/init.h> | |
395 | #include <linux/pci.h> | |
396 | #include <linux/slab.h> | |
397 | #include <sound/core.h> | |
398 | #include <sound/initval.h> | |
399 | ||
400 | /* module parameters (see "Module Parameters") */ | |
401 | static int index[SNDRV_CARDS] = SNDRV_DEFAULT_IDX; | |
402 | static char *id[SNDRV_CARDS] = SNDRV_DEFAULT_STR; | |
403 | static int enable[SNDRV_CARDS] = SNDRV_DEFAULT_ENABLE_PNP; | |
404 | ||
405 | /* definition of the chip-specific record */ | |
446ab5f5 TI |
406 | struct mychip { |
407 | struct snd_card *card; | |
1da177e4 LT |
408 | // rest of implementation will be in the section |
409 | // "PCI Resource Managements" | |
410 | }; | |
411 | ||
412 | /* chip-specific destructor | |
413 | * (see "PCI Resource Managements") | |
414 | */ | |
446ab5f5 | 415 | static int snd_mychip_free(struct mychip *chip) |
1da177e4 LT |
416 | { |
417 | .... // will be implemented later... | |
418 | } | |
419 | ||
420 | /* component-destructor | |
421 | * (see "Management of Cards and Components") | |
422 | */ | |
446ab5f5 | 423 | static int snd_mychip_dev_free(struct snd_device *device) |
1da177e4 | 424 | { |
446ab5f5 | 425 | return snd_mychip_free(device->device_data); |
1da177e4 LT |
426 | } |
427 | ||
428 | /* chip-specific constructor | |
429 | * (see "Management of Cards and Components") | |
430 | */ | |
446ab5f5 | 431 | static int __devinit snd_mychip_create(struct snd_card *card, |
1da177e4 | 432 | struct pci_dev *pci, |
446ab5f5 | 433 | struct mychip **rchip) |
1da177e4 | 434 | { |
446ab5f5 | 435 | struct mychip *chip; |
1da177e4 | 436 | int err; |
446ab5f5 | 437 | static struct snd_device_ops ops = { |
1da177e4 LT |
438 | .dev_free = snd_mychip_dev_free, |
439 | }; | |
440 | ||
441 | *rchip = NULL; | |
442 | ||
443 | // check PCI availability here | |
444 | // (see "PCI Resource Managements") | |
445 | .... | |
446 | ||
447 | /* allocate a chip-specific data with zero filled */ | |
561b220a | 448 | chip = kzalloc(sizeof(*chip), GFP_KERNEL); |
1da177e4 LT |
449 | if (chip == NULL) |
450 | return -ENOMEM; | |
451 | ||
452 | chip->card = card; | |
453 | ||
454 | // rest of initialization here; will be implemented | |
455 | // later, see "PCI Resource Managements" | |
456 | .... | |
457 | ||
458 | if ((err = snd_device_new(card, SNDRV_DEV_LOWLEVEL, | |
459 | chip, &ops)) < 0) { | |
460 | snd_mychip_free(chip); | |
461 | return err; | |
462 | } | |
463 | ||
464 | snd_card_set_dev(card, &pci->dev); | |
465 | ||
466 | *rchip = chip; | |
467 | return 0; | |
468 | } | |
469 | ||
470 | /* constructor -- see "Constructor" sub-section */ | |
471 | static int __devinit snd_mychip_probe(struct pci_dev *pci, | |
472 | const struct pci_device_id *pci_id) | |
473 | { | |
474 | static int dev; | |
446ab5f5 TI |
475 | struct snd_card *card; |
476 | struct mychip *chip; | |
1da177e4 LT |
477 | int err; |
478 | ||
479 | /* (1) */ | |
480 | if (dev >= SNDRV_CARDS) | |
481 | return -ENODEV; | |
482 | if (!enable[dev]) { | |
483 | dev++; | |
484 | return -ENOENT; | |
485 | } | |
486 | ||
487 | /* (2) */ | |
488 | card = snd_card_new(index[dev], id[dev], THIS_MODULE, 0); | |
489 | if (card == NULL) | |
490 | return -ENOMEM; | |
491 | ||
492 | /* (3) */ | |
493 | if ((err = snd_mychip_create(card, pci, &chip)) < 0) { | |
494 | snd_card_free(card); | |
495 | return err; | |
496 | } | |
497 | ||
498 | /* (4) */ | |
499 | strcpy(card->driver, "My Chip"); | |
500 | strcpy(card->shortname, "My Own Chip 123"); | |
501 | sprintf(card->longname, "%s at 0x%lx irq %i", | |
502 | card->shortname, chip->ioport, chip->irq); | |
503 | ||
504 | /* (5) */ | |
505 | .... // implemented later | |
506 | ||
507 | /* (6) */ | |
508 | if ((err = snd_card_register(card)) < 0) { | |
509 | snd_card_free(card); | |
510 | return err; | |
511 | } | |
512 | ||
513 | /* (7) */ | |
514 | pci_set_drvdata(pci, card); | |
515 | dev++; | |
516 | return 0; | |
517 | } | |
518 | ||
519 | /* destructor -- see "Destructor" sub-section */ | |
520 | static void __devexit snd_mychip_remove(struct pci_dev *pci) | |
521 | { | |
522 | snd_card_free(pci_get_drvdata(pci)); | |
523 | pci_set_drvdata(pci, NULL); | |
524 | } | |
525 | ]]> | |
526 | </programlisting> | |
527 | </example> | |
528 | </para> | |
529 | </section> | |
530 | ||
531 | <section id="basic-flow-constructor"> | |
532 | <title>Constructor</title> | |
533 | <para> | |
534 | The real constructor of PCI drivers is probe callback. The | |
535 | probe callback and other component-constructors which are called | |
536 | from probe callback should be defined with | |
537 | <parameter>__devinit</parameter> prefix. You | |
538 | cannot use <parameter>__init</parameter> prefix for them, | |
539 | because any PCI device could be a hotplug device. | |
540 | </para> | |
541 | ||
542 | <para> | |
543 | In the probe callback, the following scheme is often used. | |
544 | </para> | |
545 | ||
546 | <section id="basic-flow-constructor-device-index"> | |
547 | <title>1) Check and increment the device index.</title> | |
548 | <para> | |
549 | <informalexample> | |
550 | <programlisting> | |
551 | <![CDATA[ | |
552 | static int dev; | |
553 | .... | |
554 | if (dev >= SNDRV_CARDS) | |
555 | return -ENODEV; | |
556 | if (!enable[dev]) { | |
557 | dev++; | |
558 | return -ENOENT; | |
559 | } | |
560 | ]]> | |
561 | </programlisting> | |
562 | </informalexample> | |
563 | ||
564 | where enable[dev] is the module option. | |
565 | </para> | |
566 | ||
567 | <para> | |
568 | At each time probe callback is called, check the | |
569 | availability of the device. If not available, simply increment | |
570 | the device index and returns. dev will be incremented also | |
571 | later (<link | |
572 | linkend="basic-flow-constructor-set-pci"><citetitle>step | |
573 | 7</citetitle></link>). | |
574 | </para> | |
575 | </section> | |
576 | ||
577 | <section id="basic-flow-constructor-create-card"> | |
578 | <title>2) Create a card instance</title> | |
579 | <para> | |
580 | <informalexample> | |
581 | <programlisting> | |
582 | <![CDATA[ | |
446ab5f5 | 583 | struct snd_card *card; |
1da177e4 LT |
584 | .... |
585 | card = snd_card_new(index[dev], id[dev], THIS_MODULE, 0); | |
586 | ]]> | |
587 | </programlisting> | |
588 | </informalexample> | |
589 | </para> | |
590 | ||
591 | <para> | |
592 | The detail will be explained in the section | |
593 | <link linkend="card-management-card-instance"><citetitle> | |
594 | Management of Cards and Components</citetitle></link>. | |
595 | </para> | |
596 | </section> | |
597 | ||
598 | <section id="basic-flow-constructor-create-main"> | |
599 | <title>3) Create a main component</title> | |
600 | <para> | |
601 | In this part, the PCI resources are allocated. | |
602 | ||
603 | <informalexample> | |
604 | <programlisting> | |
605 | <![CDATA[ | |
446ab5f5 | 606 | struct mychip *chip; |
1da177e4 LT |
607 | .... |
608 | if ((err = snd_mychip_create(card, pci, &chip)) < 0) { | |
609 | snd_card_free(card); | |
610 | return err; | |
611 | } | |
612 | ]]> | |
613 | </programlisting> | |
614 | </informalexample> | |
615 | ||
616 | The detail will be explained in the section <link | |
617 | linkend="pci-resource"><citetitle>PCI Resource | |
618 | Managements</citetitle></link>. | |
619 | </para> | |
620 | </section> | |
621 | ||
622 | <section id="basic-flow-constructor-main-component"> | |
623 | <title>4) Set the driver ID and name strings.</title> | |
624 | <para> | |
625 | <informalexample> | |
626 | <programlisting> | |
627 | <![CDATA[ | |
628 | strcpy(card->driver, "My Chip"); | |
629 | strcpy(card->shortname, "My Own Chip 123"); | |
630 | sprintf(card->longname, "%s at 0x%lx irq %i", | |
631 | card->shortname, chip->ioport, chip->irq); | |
632 | ]]> | |
633 | </programlisting> | |
634 | </informalexample> | |
635 | ||
636 | The driver field holds the minimal ID string of the | |
637 | chip. This is referred by alsa-lib's configurator, so keep it | |
638 | simple but unique. | |
639 | Even the same driver can have different driver IDs to | |
640 | distinguish the functionality of each chip type. | |
641 | </para> | |
642 | ||
643 | <para> | |
644 | The shortname field is a string shown as more verbose | |
645 | name. The longname field contains the information which is | |
646 | shown in <filename>/proc/asound/cards</filename>. | |
647 | </para> | |
648 | </section> | |
649 | ||
650 | <section id="basic-flow-constructor-create-other"> | |
651 | <title>5) Create other components, such as mixer, MIDI, etc.</title> | |
652 | <para> | |
653 | Here you define the basic components such as | |
654 | <link linkend="pcm-interface"><citetitle>PCM</citetitle></link>, | |
655 | mixer (e.g. <link linkend="api-ac97"><citetitle>AC97</citetitle></link>), | |
656 | MIDI (e.g. <link linkend="midi-interface"><citetitle>MPU-401</citetitle></link>), | |
657 | and other interfaces. | |
658 | Also, if you want a <link linkend="proc-interface"><citetitle>proc | |
659 | file</citetitle></link>, define it here, too. | |
660 | </para> | |
661 | </section> | |
662 | ||
663 | <section id="basic-flow-constructor-register-card"> | |
664 | <title>6) Register the card instance.</title> | |
665 | <para> | |
666 | <informalexample> | |
667 | <programlisting> | |
668 | <![CDATA[ | |
669 | if ((err = snd_card_register(card)) < 0) { | |
670 | snd_card_free(card); | |
671 | return err; | |
672 | } | |
673 | ]]> | |
674 | </programlisting> | |
675 | </informalexample> | |
676 | </para> | |
677 | ||
678 | <para> | |
679 | Will be explained in the section <link | |
680 | linkend="card-management-registration"><citetitle>Management | |
681 | of Cards and Components</citetitle></link>, too. | |
682 | </para> | |
683 | </section> | |
684 | ||
685 | <section id="basic-flow-constructor-set-pci"> | |
686 | <title>7) Set the PCI driver data and return zero.</title> | |
687 | <para> | |
688 | <informalexample> | |
689 | <programlisting> | |
690 | <![CDATA[ | |
691 | pci_set_drvdata(pci, card); | |
692 | dev++; | |
693 | return 0; | |
694 | ]]> | |
695 | </programlisting> | |
696 | </informalexample> | |
697 | ||
698 | In the above, the card record is stored. This pointer is | |
699 | referred in the remove callback and power-management | |
700 | callbacks, too. | |
701 | </para> | |
702 | </section> | |
703 | </section> | |
704 | ||
705 | <section id="basic-flow-destructor"> | |
706 | <title>Destructor</title> | |
707 | <para> | |
708 | The destructor, remove callback, simply releases the card | |
709 | instance. Then the ALSA middle layer will release all the | |
710 | attached components automatically. | |
711 | </para> | |
712 | ||
713 | <para> | |
714 | It would be typically like the following: | |
715 | ||
716 | <informalexample> | |
717 | <programlisting> | |
718 | <![CDATA[ | |
719 | static void __devexit snd_mychip_remove(struct pci_dev *pci) | |
720 | { | |
721 | snd_card_free(pci_get_drvdata(pci)); | |
722 | pci_set_drvdata(pci, NULL); | |
723 | } | |
724 | ]]> | |
725 | </programlisting> | |
726 | </informalexample> | |
727 | ||
728 | The above code assumes that the card pointer is set to the PCI | |
729 | driver data. | |
730 | </para> | |
731 | </section> | |
732 | ||
733 | <section id="basic-flow-header-files"> | |
734 | <title>Header Files</title> | |
735 | <para> | |
736 | For the above example, at least the following include files | |
737 | are necessary. | |
738 | ||
739 | <informalexample> | |
740 | <programlisting> | |
741 | <![CDATA[ | |
742 | #include <sound/driver.h> | |
743 | #include <linux/init.h> | |
744 | #include <linux/pci.h> | |
745 | #include <linux/slab.h> | |
746 | #include <sound/core.h> | |
747 | #include <sound/initval.h> | |
748 | ]]> | |
749 | </programlisting> | |
750 | </informalexample> | |
751 | ||
752 | where the last one is necessary only when module options are | |
753 | defined in the source file. If the codes are split to several | |
754 | files, the file without module options don't need them. | |
755 | </para> | |
756 | ||
757 | <para> | |
758 | In addition to them, you'll need | |
759 | <filename><linux/interrupt.h></filename> for the interrupt | |
760 | handling, and <filename><asm/io.h></filename> for the i/o | |
761 | access. If you use <function>mdelay()</function> or | |
762 | <function>udelay()</function> functions, you'll need to include | |
763 | <filename><linux/delay.h></filename>, too. | |
764 | </para> | |
765 | ||
766 | <para> | |
767 | The ALSA interfaces like PCM or control API are defined in other | |
768 | header files as <filename><sound/xxx.h></filename>. | |
769 | They have to be included after | |
770 | <filename><sound/core.h></filename>. | |
771 | </para> | |
772 | ||
773 | </section> | |
774 | </chapter> | |
775 | ||
776 | ||
777 | <!-- ****************************************************** --> | |
778 | <!-- Management of Cards and Components --> | |
779 | <!-- ****************************************************** --> | |
780 | <chapter id="card-management"> | |
781 | <title>Management of Cards and Components</title> | |
782 | ||
783 | <section id="card-management-card-instance"> | |
784 | <title>Card Instance</title> | |
785 | <para> | |
786 | For each soundcard, a <quote>card</quote> record must be allocated. | |
787 | </para> | |
788 | ||
789 | <para> | |
790 | A card record is the headquarters of the soundcard. It manages | |
791 | the list of whole devices (components) on the soundcard, such as | |
792 | PCM, mixers, MIDI, synthesizer, and so on. Also, the card | |
793 | record holds the ID and the name strings of the card, manages | |
794 | the root of proc files, and controls the power-management states | |
795 | and hotplug disconnections. The component list on the card | |
796 | record is used to manage the proper releases of resources at | |
797 | destruction. | |
798 | </para> | |
799 | ||
800 | <para> | |
801 | As mentioned above, to create a card instance, call | |
802 | <function>snd_card_new()</function>. | |
803 | ||
804 | <informalexample> | |
805 | <programlisting> | |
806 | <![CDATA[ | |
446ab5f5 | 807 | struct snd_card *card; |
1da177e4 LT |
808 | card = snd_card_new(index, id, module, extra_size); |
809 | ]]> | |
810 | </programlisting> | |
811 | </informalexample> | |
812 | </para> | |
813 | ||
814 | <para> | |
815 | The function takes four arguments, the card-index number, the | |
816 | id string, the module pointer (usually | |
817 | <constant>THIS_MODULE</constant>), | |
818 | and the size of extra-data space. The last argument is used to | |
819 | allocate card->private_data for the | |
820 | chip-specific data. Note that this data | |
821 | <emphasis>is</emphasis> allocated by | |
822 | <function>snd_card_new()</function>. | |
823 | </para> | |
824 | </section> | |
825 | ||
826 | <section id="card-management-component"> | |
827 | <title>Components</title> | |
828 | <para> | |
829 | After the card is created, you can attach the components | |
830 | (devices) to the card instance. On ALSA driver, a component is | |
446ab5f5 | 831 | represented as a struct <structname>snd_device</structname> object. |
1da177e4 LT |
832 | A component can be a PCM instance, a control interface, a raw |
833 | MIDI interface, etc. Each of such instances has one component | |
834 | entry. | |
835 | </para> | |
836 | ||
837 | <para> | |
838 | A component can be created via | |
839 | <function>snd_device_new()</function> function. | |
840 | ||
841 | <informalexample> | |
842 | <programlisting> | |
843 | <![CDATA[ | |
844 | snd_device_new(card, SNDRV_DEV_XXX, chip, &ops); | |
845 | ]]> | |
846 | </programlisting> | |
847 | </informalexample> | |
848 | </para> | |
849 | ||
850 | <para> | |
851 | This takes the card pointer, the device-level | |
852 | (<constant>SNDRV_DEV_XXX</constant>), the data pointer, and the | |
853 | callback pointers (<parameter>&ops</parameter>). The | |
854 | device-level defines the type of components and the order of | |
855 | registration and de-registration. For most of components, the | |
856 | device-level is already defined. For a user-defined component, | |
857 | you can use <constant>SNDRV_DEV_LOWLEVEL</constant>. | |
858 | </para> | |
859 | ||
860 | <para> | |
861 | This function itself doesn't allocate the data space. The data | |
862 | must be allocated manually beforehand, and its pointer is passed | |
863 | as the argument. This pointer is used as the identifier | |
864 | (<parameter>chip</parameter> in the above example) for the | |
865 | instance. | |
866 | </para> | |
867 | ||
868 | <para> | |
869 | Each ALSA pre-defined component such as ac97 or pcm calls | |
870 | <function>snd_device_new()</function> inside its | |
871 | constructor. The destructor for each component is defined in the | |
872 | callback pointers. Hence, you don't need to take care of | |
873 | calling a destructor for such a component. | |
874 | </para> | |
875 | ||
876 | <para> | |
877 | If you would like to create your own component, you need to | |
878 | set the destructor function to dev_free callback in | |
879 | <parameter>ops</parameter>, so that it can be released | |
880 | automatically via <function>snd_card_free()</function>. The | |
881 | example will be shown later as an implementation of a | |
882 | chip-specific data. | |
883 | </para> | |
884 | </section> | |
885 | ||
886 | <section id="card-management-chip-specific"> | |
887 | <title>Chip-Specific Data</title> | |
888 | <para> | |
889 | The chip-specific information, e.g. the i/o port address, its | |
890 | resource pointer, or the irq number, is stored in the | |
891 | chip-specific record. | |
1da177e4 LT |
892 | |
893 | <informalexample> | |
894 | <programlisting> | |
895 | <![CDATA[ | |
446ab5f5 | 896 | struct mychip { |
1da177e4 LT |
897 | .... |
898 | }; | |
899 | ]]> | |
900 | </programlisting> | |
901 | </informalexample> | |
902 | </para> | |
903 | ||
904 | <para> | |
905 | In general, there are two ways to allocate the chip record. | |
906 | </para> | |
907 | ||
908 | <section id="card-management-chip-specific-snd-card-new"> | |
909 | <title>1. Allocating via <function>snd_card_new()</function>.</title> | |
910 | <para> | |
911 | As mentioned above, you can pass the extra-data-length to the 4th argument of <function>snd_card_new()</function>, i.e. | |
912 | ||
913 | <informalexample> | |
914 | <programlisting> | |
915 | <![CDATA[ | |
446ab5f5 | 916 | card = snd_card_new(index[dev], id[dev], THIS_MODULE, sizeof(struct mychip)); |
1da177e4 LT |
917 | ]]> |
918 | </programlisting> | |
919 | </informalexample> | |
920 | ||
446ab5f5 | 921 | whether struct <structname>mychip</structname> is the type of the chip record. |
1da177e4 LT |
922 | </para> |
923 | ||
924 | <para> | |
925 | In return, the allocated record can be accessed as | |
926 | ||
927 | <informalexample> | |
928 | <programlisting> | |
929 | <![CDATA[ | |
446ab5f5 | 930 | struct mychip *chip = (struct mychip *)card->private_data; |
1da177e4 LT |
931 | ]]> |
932 | </programlisting> | |
933 | </informalexample> | |
934 | ||
935 | With this method, you don't have to allocate twice. | |
936 | The record is released together with the card instance. | |
937 | </para> | |
938 | </section> | |
939 | ||
940 | <section id="card-management-chip-specific-allocate-extra"> | |
941 | <title>2. Allocating an extra device.</title> | |
942 | ||
943 | <para> | |
944 | After allocating a card instance via | |
945 | <function>snd_card_new()</function> (with | |
946 | <constant>NULL</constant> on the 4th arg), call | |
561b220a | 947 | <function>kzalloc()</function>. |
1da177e4 LT |
948 | |
949 | <informalexample> | |
950 | <programlisting> | |
951 | <![CDATA[ | |
446ab5f5 TI |
952 | struct snd_card *card; |
953 | struct mychip *chip; | |
1da177e4 LT |
954 | card = snd_card_new(index[dev], id[dev], THIS_MODULE, NULL); |
955 | ..... | |
561b220a | 956 | chip = kzalloc(sizeof(*chip), GFP_KERNEL); |
1da177e4 LT |
957 | ]]> |
958 | </programlisting> | |
959 | </informalexample> | |
960 | </para> | |
961 | ||
962 | <para> | |
963 | The chip record should have the field to hold the card | |
964 | pointer at least, | |
965 | ||
966 | <informalexample> | |
967 | <programlisting> | |
968 | <![CDATA[ | |
446ab5f5 TI |
969 | struct mychip { |
970 | struct snd_card *card; | |
1da177e4 LT |
971 | .... |
972 | }; | |
973 | ]]> | |
974 | </programlisting> | |
975 | </informalexample> | |
976 | </para> | |
977 | ||
978 | <para> | |
979 | Then, set the card pointer in the returned chip instance. | |
980 | ||
981 | <informalexample> | |
982 | <programlisting> | |
983 | <![CDATA[ | |
984 | chip->card = card; | |
985 | ]]> | |
986 | </programlisting> | |
987 | </informalexample> | |
988 | </para> | |
989 | ||
990 | <para> | |
991 | Next, initialize the fields, and register this chip | |
992 | record as a low-level device with a specified | |
993 | <parameter>ops</parameter>, | |
994 | ||
995 | <informalexample> | |
996 | <programlisting> | |
997 | <![CDATA[ | |
446ab5f5 | 998 | static struct snd_device_ops ops = { |
1da177e4 LT |
999 | .dev_free = snd_mychip_dev_free, |
1000 | }; | |
1001 | .... | |
1002 | snd_device_new(card, SNDRV_DEV_LOWLEVEL, chip, &ops); | |
1003 | ]]> | |
1004 | </programlisting> | |
1005 | </informalexample> | |
1006 | ||
1007 | <function>snd_mychip_dev_free()</function> is the | |
1008 | device-destructor function, which will call the real | |
1009 | destructor. | |
1010 | </para> | |
1011 | ||
1012 | <para> | |
1013 | <informalexample> | |
1014 | <programlisting> | |
1015 | <![CDATA[ | |
446ab5f5 | 1016 | static int snd_mychip_dev_free(struct snd_device *device) |
1da177e4 | 1017 | { |
446ab5f5 | 1018 | return snd_mychip_free(device->device_data); |
1da177e4 LT |
1019 | } |
1020 | ]]> | |
1021 | </programlisting> | |
1022 | </informalexample> | |
1023 | ||
1024 | where <function>snd_mychip_free()</function> is the real destructor. | |
1025 | </para> | |
1026 | </section> | |
1027 | </section> | |
1028 | ||
1029 | <section id="card-management-registration"> | |
1030 | <title>Registration and Release</title> | |
1031 | <para> | |
1032 | After all components are assigned, register the card instance | |
1033 | by calling <function>snd_card_register()</function>. The access | |
1034 | to the device files are enabled at this point. That is, before | |
1035 | <function>snd_card_register()</function> is called, the | |
1036 | components are safely inaccessible from external side. If this | |
1037 | call fails, exit the probe function after releasing the card via | |
1038 | <function>snd_card_free()</function>. | |
1039 | </para> | |
1040 | ||
1041 | <para> | |
1042 | For releasing the card instance, you can call simply | |
1043 | <function>snd_card_free()</function>. As already mentioned, all | |
1044 | components are released automatically by this call. | |
1045 | </para> | |
1046 | ||
1047 | <para> | |
1048 | As further notes, the destructors (both | |
1049 | <function>snd_mychip_dev_free</function> and | |
1050 | <function>snd_mychip_free</function>) cannot be defined with | |
1051 | <parameter>__devexit</parameter> prefix, because they may be | |
1052 | called from the constructor, too, at the false path. | |
1053 | </para> | |
1054 | ||
1055 | <para> | |
1056 | For a device which allows hotplugging, you can use | |
1057 | <function>snd_card_free_in_thread</function>. This one will | |
1058 | postpone the destruction and wait in a kernel-thread until all | |
1059 | devices are closed. | |
1060 | </para> | |
1061 | ||
1062 | </section> | |
1063 | ||
1064 | </chapter> | |
1065 | ||
1066 | ||
1067 | <!-- ****************************************************** --> | |
1068 | <!-- PCI Resource Managements --> | |
1069 | <!-- ****************************************************** --> | |
1070 | <chapter id="pci-resource"> | |
1071 | <title>PCI Resource Managements</title> | |
1072 | ||
1073 | <section id="pci-resource-example"> | |
1074 | <title>Full Code Example</title> | |
1075 | <para> | |
1076 | In this section, we'll finish the chip-specific constructor, | |
1077 | destructor and PCI entries. The example code is shown first, | |
1078 | below. | |
1079 | ||
1080 | <example> | |
1081 | <title>PCI Resource Managements Example</title> | |
1082 | <programlisting> | |
1083 | <![CDATA[ | |
446ab5f5 TI |
1084 | struct mychip { |
1085 | struct snd_card *card; | |
1da177e4 LT |
1086 | struct pci_dev *pci; |
1087 | ||
1088 | unsigned long port; | |
1089 | int irq; | |
1090 | }; | |
1091 | ||
446ab5f5 | 1092 | static int snd_mychip_free(struct mychip *chip) |
1da177e4 LT |
1093 | { |
1094 | /* disable hardware here if any */ | |
1095 | .... // (not implemented in this document) | |
1096 | ||
1097 | /* release the irq */ | |
1098 | if (chip->irq >= 0) | |
1099 | free_irq(chip->irq, (void *)chip); | |
1100 | /* release the i/o ports & memory */ | |
1101 | pci_release_regions(chip->pci); | |
1102 | /* disable the PCI entry */ | |
1103 | pci_disable_device(chip->pci); | |
1104 | /* release the data */ | |
1105 | kfree(chip); | |
1106 | return 0; | |
1107 | } | |
1108 | ||
1109 | /* chip-specific constructor */ | |
446ab5f5 | 1110 | static int __devinit snd_mychip_create(struct snd_card *card, |
1da177e4 | 1111 | struct pci_dev *pci, |
446ab5f5 | 1112 | struct mychip **rchip) |
1da177e4 | 1113 | { |
446ab5f5 | 1114 | struct mychip *chip; |
1da177e4 | 1115 | int err; |
446ab5f5 | 1116 | static struct snd_device_ops ops = { |
1da177e4 LT |
1117 | .dev_free = snd_mychip_dev_free, |
1118 | }; | |
1119 | ||
1120 | *rchip = NULL; | |
1121 | ||
1122 | /* initialize the PCI entry */ | |
1123 | if ((err = pci_enable_device(pci)) < 0) | |
1124 | return err; | |
1125 | /* check PCI availability (28bit DMA) */ | |
1126 | if (pci_set_dma_mask(pci, 0x0fffffff) < 0 || | |
1127 | pci_set_consistent_dma_mask(pci, 0x0fffffff) < 0) { | |
1128 | printk(KERN_ERR "error to set 28bit mask DMA\n"); | |
1129 | pci_disable_device(pci); | |
1130 | return -ENXIO; | |
1131 | } | |
1132 | ||
561b220a | 1133 | chip = kzalloc(sizeof(*chip), GFP_KERNEL); |
1da177e4 LT |
1134 | if (chip == NULL) { |
1135 | pci_disable_device(pci); | |
1136 | return -ENOMEM; | |
1137 | } | |
1138 | ||
1139 | /* initialize the stuff */ | |
1140 | chip->card = card; | |
1141 | chip->pci = pci; | |
1142 | chip->irq = -1; | |
1143 | ||
1144 | /* (1) PCI resource allocation */ | |
1145 | if ((err = pci_request_regions(pci, "My Chip")) < 0) { | |
1146 | kfree(chip); | |
1147 | pci_disable_device(pci); | |
1148 | return err; | |
1149 | } | |
1150 | chip->port = pci_resource_start(pci, 0); | |
1151 | if (request_irq(pci->irq, snd_mychip_interrupt, | |
446ab5f5 | 1152 | SA_INTERRUPT|SA_SHIRQ, "My Chip", chip)) { |
1da177e4 LT |
1153 | printk(KERN_ERR "cannot grab irq %d\n", pci->irq); |
1154 | snd_mychip_free(chip); | |
1155 | return -EBUSY; | |
1156 | } | |
1157 | chip->irq = pci->irq; | |
1158 | ||
1159 | /* (2) initialization of the chip hardware */ | |
1160 | .... // (not implemented in this document) | |
1161 | ||
1162 | if ((err = snd_device_new(card, SNDRV_DEV_LOWLEVEL, | |
1163 | chip, &ops)) < 0) { | |
1164 | snd_mychip_free(chip); | |
1165 | return err; | |
1166 | } | |
1167 | ||
1168 | snd_card_set_dev(card, &pci->dev); | |
1169 | ||
1170 | *rchip = chip; | |
1171 | return 0; | |
1172 | } | |
1173 | ||
1174 | /* PCI IDs */ | |
1175 | static struct pci_device_id snd_mychip_ids[] = { | |
1176 | { PCI_VENDOR_ID_FOO, PCI_DEVICE_ID_BAR, | |
1177 | PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, }, | |
1178 | .... | |
1179 | { 0, } | |
1180 | }; | |
1181 | MODULE_DEVICE_TABLE(pci, snd_mychip_ids); | |
1182 | ||
1183 | /* pci_driver definition */ | |
1184 | static struct pci_driver driver = { | |
1185 | .name = "My Own Chip", | |
1186 | .id_table = snd_mychip_ids, | |
1187 | .probe = snd_mychip_probe, | |
1188 | .remove = __devexit_p(snd_mychip_remove), | |
1189 | }; | |
1190 | ||
1191 | /* initialization of the module */ | |
1192 | static int __init alsa_card_mychip_init(void) | |
1193 | { | |
01d25d46 | 1194 | return pci_register_driver(&driver); |
1da177e4 LT |
1195 | } |
1196 | ||
1197 | /* clean up the module */ | |
1198 | static void __exit alsa_card_mychip_exit(void) | |
1199 | { | |
1200 | pci_unregister_driver(&driver); | |
1201 | } | |
1202 | ||
1203 | module_init(alsa_card_mychip_init) | |
1204 | module_exit(alsa_card_mychip_exit) | |
1205 | ||
1206 | EXPORT_NO_SYMBOLS; /* for old kernels only */ | |
1207 | ]]> | |
1208 | </programlisting> | |
1209 | </example> | |
1210 | </para> | |
1211 | </section> | |
1212 | ||
1213 | <section id="pci-resource-some-haftas"> | |
1214 | <title>Some Hafta's</title> | |
1215 | <para> | |
1216 | The allocation of PCI resources is done in the | |
1217 | <function>probe()</function> function, and usually an extra | |
1218 | <function>xxx_create()</function> function is written for this | |
1219 | purpose. | |
1220 | </para> | |
1221 | ||
1222 | <para> | |
1223 | In the case of PCI devices, you have to call at first | |
1224 | <function>pci_enable_device()</function> function before | |
1225 | allocating resources. Also, you need to set the proper PCI DMA | |
1226 | mask to limit the accessed i/o range. In some cases, you might | |
1227 | need to call <function>pci_set_master()</function> function, | |
1228 | too. | |
1229 | </para> | |
1230 | ||
1231 | <para> | |
1232 | Suppose the 28bit mask, and the code to be added would be like: | |
1233 | ||
1234 | <informalexample> | |
1235 | <programlisting> | |
1236 | <![CDATA[ | |
1237 | if ((err = pci_enable_device(pci)) < 0) | |
1238 | return err; | |
1239 | if (pci_set_dma_mask(pci, 0x0fffffff) < 0 || | |
1240 | pci_set_consistent_dma_mask(pci, 0x0fffffff) < 0) { | |
1241 | printk(KERN_ERR "error to set 28bit mask DMA\n"); | |
1242 | pci_disable_device(pci); | |
1243 | return -ENXIO; | |
1244 | } | |
1245 | ||
1246 | ]]> | |
1247 | </programlisting> | |
1248 | </informalexample> | |
1249 | </para> | |
1250 | </section> | |
1251 | ||
1252 | <section id="pci-resource-resource-allocation"> | |
1253 | <title>Resource Allocation</title> | |
1254 | <para> | |
1255 | The allocation of I/O ports and irqs are done via standard kernel | |
1256 | functions. Unlike ALSA ver.0.5.x., there are no helpers for | |
1257 | that. And these resources must be released in the destructor | |
1258 | function (see below). Also, on ALSA 0.9.x, you don't need to | |
1259 | allocate (pseudo-)DMA for PCI like ALSA 0.5.x. | |
1260 | </para> | |
1261 | ||
1262 | <para> | |
1263 | Now assume that this PCI device has an I/O port with 8 bytes | |
446ab5f5 | 1264 | and an interrupt. Then struct <structname>mychip</structname> will have the |
1da177e4 LT |
1265 | following fields: |
1266 | ||
1267 | <informalexample> | |
1268 | <programlisting> | |
1269 | <![CDATA[ | |
446ab5f5 TI |
1270 | struct mychip { |
1271 | struct snd_card *card; | |
1da177e4 LT |
1272 | |
1273 | unsigned long port; | |
1274 | int irq; | |
1275 | }; | |
1276 | ]]> | |
1277 | </programlisting> | |
1278 | </informalexample> | |
1279 | </para> | |
1280 | ||
1281 | <para> | |
1282 | For an i/o port (and also a memory region), you need to have | |
1283 | the resource pointer for the standard resource management. For | |
1284 | an irq, you have to keep only the irq number (integer). But you | |
1285 | need to initialize this number as -1 before actual allocation, | |
1286 | since irq 0 is valid. The port address and its resource pointer | |
1287 | can be initialized as null by | |
561b220a | 1288 | <function>kzalloc()</function> automatically, so you |
1da177e4 LT |
1289 | don't have to take care of resetting them. |
1290 | </para> | |
1291 | ||
1292 | <para> | |
1293 | The allocation of an i/o port is done like this: | |
1294 | ||
1295 | <informalexample> | |
1296 | <programlisting> | |
1297 | <![CDATA[ | |
1298 | if ((err = pci_request_regions(pci, "My Chip")) < 0) { | |
1299 | kfree(chip); | |
1300 | pci_disable_device(pci); | |
1301 | return err; | |
1302 | } | |
1303 | chip->port = pci_resource_start(pci, 0); | |
1304 | ]]> | |
1305 | </programlisting> | |
1306 | </informalexample> | |
1307 | </para> | |
1308 | ||
1309 | <para> | |
1310 | <!-- obsolete --> | |
1311 | It will reserve the i/o port region of 8 bytes of the given | |
1312 | PCI device. The returned value, chip->res_port, is allocated | |
1313 | via <function>kmalloc()</function> by | |
1314 | <function>request_region()</function>. The pointer must be | |
1315 | released via <function>kfree()</function>, but there is some | |
1316 | problem regarding this. This issue will be explained more below. | |
1317 | </para> | |
1318 | ||
1319 | <para> | |
1320 | The allocation of an interrupt source is done like this: | |
1321 | ||
1322 | <informalexample> | |
1323 | <programlisting> | |
1324 | <![CDATA[ | |
1325 | if (request_irq(pci->irq, snd_mychip_interrupt, | |
446ab5f5 | 1326 | SA_INTERRUPT|SA_SHIRQ, "My Chip", chip)) { |
1da177e4 LT |
1327 | printk(KERN_ERR "cannot grab irq %d\n", pci->irq); |
1328 | snd_mychip_free(chip); | |
1329 | return -EBUSY; | |
1330 | } | |
1331 | chip->irq = pci->irq; | |
1332 | ]]> | |
1333 | </programlisting> | |
1334 | </informalexample> | |
1335 | ||
1336 | where <function>snd_mychip_interrupt()</function> is the | |
1337 | interrupt handler defined <link | |
1338 | linkend="pcm-interface-interrupt-handler"><citetitle>later</citetitle></link>. | |
1339 | Note that chip->irq should be defined | |
1340 | only when <function>request_irq()</function> succeeded. | |
1341 | </para> | |
1342 | ||
1343 | <para> | |
1344 | On the PCI bus, the interrupts can be shared. Thus, | |
1345 | <constant>SA_SHIRQ</constant> is given as the interrupt flag of | |
1346 | <function>request_irq()</function>. | |
1347 | </para> | |
1348 | ||
1349 | <para> | |
1350 | The last argument of <function>request_irq()</function> is the | |
1351 | data pointer passed to the interrupt handler. Usually, the | |
1352 | chip-specific record is used for that, but you can use what you | |
1353 | like, too. | |
1354 | </para> | |
1355 | ||
1356 | <para> | |
1357 | I won't define the detail of the interrupt handler at this | |
1358 | point, but at least its appearance can be explained now. The | |
1359 | interrupt handler looks usually like the following: | |
1360 | ||
1361 | <informalexample> | |
1362 | <programlisting> | |
1363 | <![CDATA[ | |
1364 | static irqreturn_t snd_mychip_interrupt(int irq, void *dev_id, | |
1365 | struct pt_regs *regs) | |
1366 | { | |
446ab5f5 | 1367 | struct mychip *chip = dev_id; |
1da177e4 LT |
1368 | .... |
1369 | return IRQ_HANDLED; | |
1370 | } | |
1371 | ]]> | |
1372 | </programlisting> | |
1373 | </informalexample> | |
1374 | </para> | |
1375 | ||
1376 | <para> | |
1377 | Now let's write the corresponding destructor for the resources | |
1378 | above. The role of destructor is simple: disable the hardware | |
1379 | (if already activated) and release the resources. So far, we | |
1380 | have no hardware part, so the disabling is not written here. | |
1381 | </para> | |
1382 | ||
1383 | <para> | |
1384 | For releasing the resources, <quote>check-and-release</quote> | |
1385 | method is a safer way. For the interrupt, do like this: | |
1386 | ||
1387 | <informalexample> | |
1388 | <programlisting> | |
1389 | <![CDATA[ | |
1390 | if (chip->irq >= 0) | |
1391 | free_irq(chip->irq, (void *)chip); | |
1392 | ]]> | |
1393 | </programlisting> | |
1394 | </informalexample> | |
1395 | ||
1396 | Since the irq number can start from 0, you should initialize | |
1397 | chip->irq with a negative value (e.g. -1), so that you can | |
1398 | check the validity of the irq number as above. | |
1399 | </para> | |
1400 | ||
1401 | <para> | |
1402 | When you requested I/O ports or memory regions via | |
1403 | <function>pci_request_region()</function> or | |
1404 | <function>pci_request_regions()</function> like this example, | |
1405 | release the resource(s) using the corresponding function, | |
1406 | <function>pci_release_region()</function> or | |
1407 | <function>pci_release_regions()</function>. | |
1408 | ||
1409 | <informalexample> | |
1410 | <programlisting> | |
1411 | <![CDATA[ | |
1412 | pci_release_regions(chip->pci); | |
1413 | ]]> | |
1414 | </programlisting> | |
1415 | </informalexample> | |
1416 | </para> | |
1417 | ||
1418 | <para> | |
1419 | When you requested manually via <function>request_region()</function> | |
1420 | or <function>request_mem_region</function>, you can release it via | |
1421 | <function>release_resource()</function>. Suppose that you keep | |
1422 | the resource pointer returned from <function>request_region()</function> | |
1423 | in chip->res_port, the release procedure looks like below: | |
1424 | ||
1425 | <informalexample> | |
1426 | <programlisting> | |
1427 | <![CDATA[ | |
b1d5776d | 1428 | release_and_free_resource(chip->res_port); |
1da177e4 LT |
1429 | ]]> |
1430 | </programlisting> | |
1431 | </informalexample> | |
1da177e4 LT |
1432 | </para> |
1433 | ||
1434 | <para> | |
1435 | Don't forget to call <function>pci_disable_device()</function> | |
1436 | before all finished. | |
1437 | </para> | |
1438 | ||
1439 | <para> | |
1440 | And finally, release the chip-specific record. | |
1441 | ||
1442 | <informalexample> | |
1443 | <programlisting> | |
1444 | <![CDATA[ | |
1445 | kfree(chip); | |
1446 | ]]> | |
1447 | </programlisting> | |
1448 | </informalexample> | |
1449 | </para> | |
1450 | ||
1451 | <para> | |
1452 | Again, remember that you cannot | |
1453 | set <parameter>__devexit</parameter> prefix for this destructor. | |
1454 | </para> | |
1455 | ||
1456 | <para> | |
1457 | We didn't implement the hardware-disabling part in the above. | |
1458 | If you need to do this, please note that the destructor may be | |
1459 | called even before the initialization of the chip is completed. | |
1460 | It would be better to have a flag to skip the hardware-disabling | |
1461 | if the hardware was not initialized yet. | |
1462 | </para> | |
1463 | ||
1464 | <para> | |
1465 | When the chip-data is assigned to the card using | |
1466 | <function>snd_device_new()</function> with | |
1467 | <constant>SNDRV_DEV_LOWLELVEL</constant> , its destructor is | |
1468 | called at the last. That is, it is assured that all other | |
1469 | components like PCMs and controls have been already released. | |
1470 | You don't have to call stopping PCMs, etc. explicitly, but just | |
1471 | stop the hardware in the low-level. | |
1472 | </para> | |
1473 | ||
1474 | <para> | |
1475 | The management of a memory-mapped region is almost as same as | |
1476 | the management of an i/o port. You'll need three fields like | |
1477 | the following: | |
1478 | ||
1479 | <informalexample> | |
1480 | <programlisting> | |
1481 | <![CDATA[ | |
446ab5f5 | 1482 | struct mychip { |
1da177e4 LT |
1483 | .... |
1484 | unsigned long iobase_phys; | |
1485 | void __iomem *iobase_virt; | |
1486 | }; | |
1487 | ]]> | |
1488 | </programlisting> | |
1489 | </informalexample> | |
1490 | ||
1491 | and the allocation would be like below: | |
1492 | ||
1493 | <informalexample> | |
1494 | <programlisting> | |
1495 | <![CDATA[ | |
1496 | if ((err = pci_request_regions(pci, "My Chip")) < 0) { | |
1497 | kfree(chip); | |
1498 | return err; | |
1499 | } | |
1500 | chip->iobase_phys = pci_resource_start(pci, 0); | |
1501 | chip->iobase_virt = ioremap_nocache(chip->iobase_phys, | |
1502 | pci_resource_len(pci, 0)); | |
1503 | ]]> | |
1504 | </programlisting> | |
1505 | </informalexample> | |
1506 | ||
1507 | and the corresponding destructor would be: | |
1508 | ||
1509 | <informalexample> | |
1510 | <programlisting> | |
1511 | <![CDATA[ | |
446ab5f5 | 1512 | static int snd_mychip_free(struct mychip *chip) |
1da177e4 LT |
1513 | { |
1514 | .... | |
1515 | if (chip->iobase_virt) | |
1516 | iounmap(chip->iobase_virt); | |
1517 | .... | |
1518 | pci_release_regions(chip->pci); | |
1519 | .... | |
1520 | } | |
1521 | ]]> | |
1522 | </programlisting> | |
1523 | </informalexample> | |
1524 | </para> | |
1525 | ||
1526 | </section> | |
1527 | ||
1528 | <section id="pci-resource-device-struct"> | |
1529 | <title>Registration of Device Struct</title> | |
1530 | <para> | |
1531 | At some point, typically after calling <function>snd_device_new()</function>, | |
446ab5f5 | 1532 | you need to register the struct <structname>device</structname> of the chip |
1da177e4 LT |
1533 | you're handling for udev and co. ALSA provides a macro for compatibility with |
1534 | older kernels. Simply call like the following: | |
1535 | <informalexample> | |
1536 | <programlisting> | |
1537 | <![CDATA[ | |
1538 | snd_card_set_dev(card, &pci->dev); | |
1539 | ]]> | |
1540 | </programlisting> | |
1541 | </informalexample> | |
1542 | so that it stores the PCI's device pointer to the card. This will be | |
1543 | referred by ALSA core functions later when the devices are registered. | |
1544 | </para> | |
1545 | <para> | |
1546 | In the case of non-PCI, pass the proper device struct pointer of the BUS | |
1547 | instead. (In the case of legacy ISA without PnP, you don't have to do | |
1548 | anything.) | |
1549 | </para> | |
1550 | </section> | |
1551 | ||
1552 | <section id="pci-resource-entries"> | |
1553 | <title>PCI Entries</title> | |
1554 | <para> | |
1555 | So far, so good. Let's finish the rest of missing PCI | |
1556 | stuffs. At first, we need a | |
1557 | <structname>pci_device_id</structname> table for this | |
1558 | chipset. It's a table of PCI vendor/device ID number, and some | |
1559 | masks. | |
1560 | </para> | |
1561 | ||
1562 | <para> | |
1563 | For example, | |
1564 | ||
1565 | <informalexample> | |
1566 | <programlisting> | |
1567 | <![CDATA[ | |
1568 | static struct pci_device_id snd_mychip_ids[] = { | |
1569 | { PCI_VENDOR_ID_FOO, PCI_DEVICE_ID_BAR, | |
1570 | PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, }, | |
1571 | .... | |
1572 | { 0, } | |
1573 | }; | |
1574 | MODULE_DEVICE_TABLE(pci, snd_mychip_ids); | |
1575 | ]]> | |
1576 | </programlisting> | |
1577 | </informalexample> | |
1578 | </para> | |
1579 | ||
1580 | <para> | |
1581 | The first and second fields of | |
1582 | <structname>pci_device_id</structname> struct are the vendor and | |
1583 | device IDs. If you have nothing special to filter the matching | |
1584 | devices, you can use the rest of fields like above. The last | |
1585 | field of <structname>pci_device_id</structname> struct is a | |
1586 | private data for this entry. You can specify any value here, for | |
1587 | example, to tell the type of different operations per each | |
1588 | device IDs. Such an example is found in intel8x0 driver. | |
1589 | </para> | |
1590 | ||
1591 | <para> | |
1592 | The last entry of this list is the terminator. You must | |
1593 | specify this all-zero entry. | |
1594 | </para> | |
1595 | ||
1596 | <para> | |
1597 | Then, prepare the <structname>pci_driver</structname> record: | |
1598 | ||
1599 | <informalexample> | |
1600 | <programlisting> | |
1601 | <![CDATA[ | |
1602 | static struct pci_driver driver = { | |
1603 | .name = "My Own Chip", | |
1604 | .id_table = snd_mychip_ids, | |
1605 | .probe = snd_mychip_probe, | |
1606 | .remove = __devexit_p(snd_mychip_remove), | |
1607 | }; | |
1608 | ]]> | |
1609 | </programlisting> | |
1610 | </informalexample> | |
1611 | </para> | |
1612 | ||
1613 | <para> | |
1614 | The <structfield>probe</structfield> and | |
1615 | <structfield>remove</structfield> functions are what we already | |
1616 | defined in | |
1617 | the previous sections. The <structfield>remove</structfield> should | |
1618 | be defined with | |
1619 | <function>__devexit_p()</function> macro, so that it's not | |
1620 | defined for built-in (and non-hot-pluggable) case. The | |
1621 | <structfield>name</structfield> | |
1622 | field is the name string of this device. Note that you must not | |
1623 | use a slash <quote>/</quote> in this string. | |
1624 | </para> | |
1625 | ||
1626 | <para> | |
1627 | And at last, the module entries: | |
1628 | ||
1629 | <informalexample> | |
1630 | <programlisting> | |
1631 | <![CDATA[ | |
1632 | static int __init alsa_card_mychip_init(void) | |
1633 | { | |
01d25d46 | 1634 | return pci_register_driver(&driver); |
1da177e4 LT |
1635 | } |
1636 | ||
1637 | static void __exit alsa_card_mychip_exit(void) | |
1638 | { | |
1639 | pci_unregister_driver(&driver); | |
1640 | } | |
1641 | ||
1642 | module_init(alsa_card_mychip_init) | |
1643 | module_exit(alsa_card_mychip_exit) | |
1644 | ]]> | |
1645 | </programlisting> | |
1646 | </informalexample> | |
1647 | </para> | |
1648 | ||
1649 | <para> | |
1650 | Note that these module entries are tagged with | |
1651 | <parameter>__init</parameter> and | |
1652 | <parameter>__exit</parameter> prefixes, not | |
1653 | <parameter>__devinit</parameter> nor | |
1654 | <parameter>__devexit</parameter>. | |
1655 | </para> | |
1656 | ||
1657 | <para> | |
1658 | Oh, one thing was forgotten. If you have no exported symbols, | |
1659 | you need to declare it on 2.2 or 2.4 kernels (on 2.6 kernels | |
1660 | it's not necessary, though). | |
1661 | ||
1662 | <informalexample> | |
1663 | <programlisting> | |
1664 | <![CDATA[ | |
1665 | EXPORT_NO_SYMBOLS; | |
1666 | ]]> | |
1667 | </programlisting> | |
1668 | </informalexample> | |
1669 | ||
1670 | That's all! | |
1671 | </para> | |
1672 | </section> | |
1673 | </chapter> | |
1674 | ||
1675 | ||
1676 | <!-- ****************************************************** --> | |
1677 | <!-- PCM Interface --> | |
1678 | <!-- ****************************************************** --> | |
1679 | <chapter id="pcm-interface"> | |
1680 | <title>PCM Interface</title> | |
1681 | ||
1682 | <section id="pcm-interface-general"> | |
1683 | <title>General</title> | |
1684 | <para> | |
1685 | The PCM middle layer of ALSA is quite powerful and it is only | |
1686 | necessary for each driver to implement the low-level functions | |
1687 | to access its hardware. | |
1688 | </para> | |
1689 | ||
1690 | <para> | |
1691 | For accessing to the PCM layer, you need to include | |
1692 | <filename><sound/pcm.h></filename> above all. In addition, | |
1693 | <filename><sound/pcm_params.h></filename> might be needed | |
1694 | if you access to some functions related with hw_param. | |
1695 | </para> | |
1696 | ||
1697 | <para> | |
1698 | Each card device can have up to four pcm instances. A pcm | |
1699 | instance corresponds to a pcm device file. The limitation of | |
1700 | number of instances comes only from the available bit size of | |
1701 | the linux's device number. Once when 64bit device number is | |
1702 | used, we'll have more available pcm instances. | |
1703 | </para> | |
1704 | ||
1705 | <para> | |
1706 | A pcm instance consists of pcm playback and capture streams, | |
1707 | and each pcm stream consists of one or more pcm substreams. Some | |
1708 | soundcard supports the multiple-playback function. For example, | |
1709 | emu10k1 has a PCM playback of 32 stereo substreams. In this case, at | |
1710 | each open, a free substream is (usually) automatically chosen | |
1711 | and opened. Meanwhile, when only one substream exists and it was | |
1712 | already opened, the succeeding open will result in the blocking | |
1713 | or the error with <constant>EAGAIN</constant> according to the | |
1714 | file open mode. But you don't have to know the detail in your | |
1715 | driver. The PCM middle layer will take all such jobs. | |
1716 | </para> | |
1717 | </section> | |
1718 | ||
1719 | <section id="pcm-interface-example"> | |
1720 | <title>Full Code Example</title> | |
1721 | <para> | |
1722 | The example code below does not include any hardware access | |
1723 | routines but shows only the skeleton, how to build up the PCM | |
1724 | interfaces. | |
1725 | ||
1726 | <example> | |
1727 | <title>PCM Example Code</title> | |
1728 | <programlisting> | |
1729 | <![CDATA[ | |
1730 | #include <sound/pcm.h> | |
1731 | .... | |
1732 | ||
1733 | /* hardware definition */ | |
446ab5f5 | 1734 | static struct snd_pcm_hardware snd_mychip_playback_hw = { |
1da177e4 LT |
1735 | .info = (SNDRV_PCM_INFO_MMAP | |
1736 | SNDRV_PCM_INFO_INTERLEAVED | | |
1737 | SNDRV_PCM_INFO_BLOCK_TRANSFER | | |
1738 | SNDRV_PCM_INFO_MMAP_VALID), | |
1739 | .formats = SNDRV_PCM_FMTBIT_S16_LE, | |
1740 | .rates = SNDRV_PCM_RATE_8000_48000, | |
1741 | .rate_min = 8000, | |
1742 | .rate_max = 48000, | |
1743 | .channels_min = 2, | |
1744 | .channels_max = 2, | |
1745 | .buffer_bytes_max = 32768, | |
1746 | .period_bytes_min = 4096, | |
1747 | .period_bytes_max = 32768, | |
1748 | .periods_min = 1, | |
1749 | .periods_max = 1024, | |
1750 | }; | |
1751 | ||
1752 | /* hardware definition */ | |
446ab5f5 | 1753 | static struct snd_pcm_hardware snd_mychip_capture_hw = { |
1da177e4 LT |
1754 | .info = (SNDRV_PCM_INFO_MMAP | |
1755 | SNDRV_PCM_INFO_INTERLEAVED | | |
1756 | SNDRV_PCM_INFO_BLOCK_TRANSFER | | |
1757 | SNDRV_PCM_INFO_MMAP_VALID), | |
1758 | .formats = SNDRV_PCM_FMTBIT_S16_LE, | |
1759 | .rates = SNDRV_PCM_RATE_8000_48000, | |
1760 | .rate_min = 8000, | |
1761 | .rate_max = 48000, | |
1762 | .channels_min = 2, | |
1763 | .channels_max = 2, | |
1764 | .buffer_bytes_max = 32768, | |
1765 | .period_bytes_min = 4096, | |
1766 | .period_bytes_max = 32768, | |
1767 | .periods_min = 1, | |
1768 | .periods_max = 1024, | |
1769 | }; | |
1770 | ||
1771 | /* open callback */ | |
446ab5f5 | 1772 | static int snd_mychip_playback_open(struct snd_pcm_substream *substream) |
1da177e4 | 1773 | { |
446ab5f5 TI |
1774 | struct mychip *chip = snd_pcm_substream_chip(substream); |
1775 | struct snd_pcm_runtime *runtime = substream->runtime; | |
1da177e4 LT |
1776 | |
1777 | runtime->hw = snd_mychip_playback_hw; | |
1778 | // more hardware-initialization will be done here | |
1779 | return 0; | |
1780 | } | |
1781 | ||
1782 | /* close callback */ | |
446ab5f5 | 1783 | static int snd_mychip_playback_close(struct snd_pcm_substream *substream) |
1da177e4 | 1784 | { |
446ab5f5 | 1785 | struct mychip *chip = snd_pcm_substream_chip(substream); |
1da177e4 LT |
1786 | // the hardware-specific codes will be here |
1787 | return 0; | |
1788 | ||
1789 | } | |
1790 | ||
1791 | /* open callback */ | |
446ab5f5 | 1792 | static int snd_mychip_capture_open(struct snd_pcm_substream *substream) |
1da177e4 | 1793 | { |
446ab5f5 TI |
1794 | struct mychip *chip = snd_pcm_substream_chip(substream); |
1795 | struct snd_pcm_runtime *runtime = substream->runtime; | |
1da177e4 LT |
1796 | |
1797 | runtime->hw = snd_mychip_capture_hw; | |
1798 | // more hardware-initialization will be done here | |
1799 | return 0; | |
1800 | } | |
1801 | ||
1802 | /* close callback */ | |
446ab5f5 | 1803 | static int snd_mychip_capture_close(struct snd_pcm_substream *substream) |
1da177e4 | 1804 | { |
446ab5f5 | 1805 | struct mychip *chip = snd_pcm_substream_chip(substream); |
1da177e4 LT |
1806 | // the hardware-specific codes will be here |
1807 | return 0; | |
1808 | ||
1809 | } | |
1810 | ||
1811 | /* hw_params callback */ | |
446ab5f5 TI |
1812 | static int snd_mychip_pcm_hw_params(struct snd_pcm_substream *substream, |
1813 | struct snd_pcm_hw_params *hw_params) | |
1da177e4 LT |
1814 | { |
1815 | return snd_pcm_lib_malloc_pages(substream, | |
1816 | params_buffer_bytes(hw_params)); | |
1817 | } | |
1818 | ||
1819 | /* hw_free callback */ | |
446ab5f5 | 1820 | static int snd_mychip_pcm_hw_free(struct snd_pcm_substream *substream) |
1da177e4 LT |
1821 | { |
1822 | return snd_pcm_lib_free_pages(substream); | |
1823 | } | |
1824 | ||
1825 | /* prepare callback */ | |
446ab5f5 | 1826 | static int snd_mychip_pcm_prepare(struct snd_pcm_substream *substream) |
1da177e4 | 1827 | { |
446ab5f5 TI |
1828 | struct mychip *chip = snd_pcm_substream_chip(substream); |
1829 | struct snd_pcm_runtime *runtime = substream->runtime; | |
1da177e4 LT |
1830 | |
1831 | /* set up the hardware with the current configuration | |
1832 | * for example... | |
1833 | */ | |
1834 | mychip_set_sample_format(chip, runtime->format); | |
1835 | mychip_set_sample_rate(chip, runtime->rate); | |
1836 | mychip_set_channels(chip, runtime->channels); | |
1837 | mychip_set_dma_setup(chip, runtime->dma_area, | |
1838 | chip->buffer_size, | |
1839 | chip->period_size); | |
1840 | return 0; | |
1841 | } | |
1842 | ||
1843 | /* trigger callback */ | |
446ab5f5 | 1844 | static int snd_mychip_pcm_trigger(struct snd_pcm_substream *substream, |
1da177e4 LT |
1845 | int cmd) |
1846 | { | |
1847 | switch (cmd) { | |
1848 | case SNDRV_PCM_TRIGGER_START: | |
1849 | // do something to start the PCM engine | |
1850 | break; | |
1851 | case SNDRV_PCM_TRIGGER_STOP: | |
1852 | // do something to stop the PCM engine | |
1853 | break; | |
1854 | default: | |
1855 | return -EINVAL; | |
1856 | } | |
1857 | } | |
1858 | ||
1859 | /* pointer callback */ | |
1860 | static snd_pcm_uframes_t | |
446ab5f5 | 1861 | snd_mychip_pcm_pointer(struct snd_pcm_substream *substream) |
1da177e4 | 1862 | { |
446ab5f5 | 1863 | struct mychip *chip = snd_pcm_substream_chip(substream); |
1da177e4 LT |
1864 | unsigned int current_ptr; |
1865 | ||
1866 | /* get the current hardware pointer */ | |
1867 | current_ptr = mychip_get_hw_pointer(chip); | |
1868 | return current_ptr; | |
1869 | } | |
1870 | ||
1871 | /* operators */ | |
446ab5f5 | 1872 | static struct snd_pcm_ops snd_mychip_playback_ops = { |
1da177e4 LT |
1873 | .open = snd_mychip_playback_open, |
1874 | .close = snd_mychip_playback_close, | |
1875 | .ioctl = snd_pcm_lib_ioctl, | |
1876 | .hw_params = snd_mychip_pcm_hw_params, | |
1877 | .hw_free = snd_mychip_pcm_hw_free, | |
1878 | .prepare = snd_mychip_pcm_prepare, | |
1879 | .trigger = snd_mychip_pcm_trigger, | |
1880 | .pointer = snd_mychip_pcm_pointer, | |
1881 | }; | |
1882 | ||
1883 | /* operators */ | |
446ab5f5 | 1884 | static struct snd_pcm_ops snd_mychip_capture_ops = { |
1da177e4 LT |
1885 | .open = snd_mychip_capture_open, |
1886 | .close = snd_mychip_capture_close, | |
1887 | .ioctl = snd_pcm_lib_ioctl, | |
1888 | .hw_params = snd_mychip_pcm_hw_params, | |
1889 | .hw_free = snd_mychip_pcm_hw_free, | |
1890 | .prepare = snd_mychip_pcm_prepare, | |
1891 | .trigger = snd_mychip_pcm_trigger, | |
1892 | .pointer = snd_mychip_pcm_pointer, | |
1893 | }; | |
1894 | ||
1895 | /* | |
1896 | * definitions of capture are omitted here... | |
1897 | */ | |
1898 | ||
1899 | /* create a pcm device */ | |
446ab5f5 | 1900 | static int __devinit snd_mychip_new_pcm(struct mychip *chip) |
1da177e4 | 1901 | { |
446ab5f5 | 1902 | struct snd_pcm *pcm; |
1da177e4 LT |
1903 | int err; |
1904 | ||
1905 | if ((err = snd_pcm_new(chip->card, "My Chip", 0, 1, 1, | |
1906 | &pcm)) < 0) | |
1907 | return err; | |
1908 | pcm->private_data = chip; | |
1909 | strcpy(pcm->name, "My Chip"); | |
1910 | chip->pcm = pcm; | |
1911 | /* set operators */ | |
1912 | snd_pcm_set_ops(pcm, SNDRV_PCM_STREAM_PLAYBACK, | |
1913 | &snd_mychip_playback_ops); | |
1914 | snd_pcm_set_ops(pcm, SNDRV_PCM_STREAM_CAPTURE, | |
1915 | &snd_mychip_capture_ops); | |
1916 | /* pre-allocation of buffers */ | |
1917 | /* NOTE: this may fail */ | |
1918 | snd_pcm_lib_preallocate_pages_for_all(pcm, SNDRV_DMA_TYPE_DEV, | |
1919 | snd_dma_pci_data(chip->pci), | |
1920 | 64*1024, 64*1024); | |
1921 | return 0; | |
1922 | } | |
1923 | ]]> | |
1924 | </programlisting> | |
1925 | </example> | |
1926 | </para> | |
1927 | </section> | |
1928 | ||
1929 | <section id="pcm-interface-constructor"> | |
1930 | <title>Constructor</title> | |
1931 | <para> | |
1932 | A pcm instance is allocated by <function>snd_pcm_new()</function> | |
1933 | function. It would be better to create a constructor for pcm, | |
1934 | namely, | |
1935 | ||
1936 | <informalexample> | |
1937 | <programlisting> | |
1938 | <![CDATA[ | |
446ab5f5 | 1939 | static int __devinit snd_mychip_new_pcm(struct mychip *chip) |
1da177e4 | 1940 | { |
446ab5f5 | 1941 | struct snd_pcm *pcm; |
1da177e4 LT |
1942 | int err; |
1943 | ||
1944 | if ((err = snd_pcm_new(chip->card, "My Chip", 0, 1, 1, | |
1945 | &pcm)) < 0) | |
1946 | return err; | |
1947 | pcm->private_data = chip; | |
1948 | strcpy(pcm->name, "My Chip"); | |
1949 | chip->pcm = pcm; | |
1950 | .... | |
1951 | return 0; | |
1952 | } | |
1953 | ]]> | |
1954 | </programlisting> | |
1955 | </informalexample> | |
1956 | </para> | |
1957 | ||
1958 | <para> | |
1959 | The <function>snd_pcm_new()</function> function takes the four | |
1960 | arguments. The first argument is the card pointer to which this | |
1961 | pcm is assigned, and the second is the ID string. | |
1962 | </para> | |
1963 | ||
1964 | <para> | |
1965 | The third argument (<parameter>index</parameter>, 0 in the | |
1966 | above) is the index of this new pcm. It begins from zero. When | |
1967 | you will create more than one pcm instances, specify the | |
1968 | different numbers in this argument. For example, | |
1969 | <parameter>index</parameter> = 1 for the second PCM device. | |
1970 | </para> | |
1971 | ||
1972 | <para> | |
1973 | The fourth and fifth arguments are the number of substreams | |
1974 | for playback and capture, respectively. Here both 1 are given in | |
1975 | the above example. When no playback or no capture is available, | |
1976 | pass 0 to the corresponding argument. | |
1977 | </para> | |
1978 | ||
1979 | <para> | |
1980 | If a chip supports multiple playbacks or captures, you can | |
1981 | specify more numbers, but they must be handled properly in | |
1982 | open/close, etc. callbacks. When you need to know which | |
1983 | substream you are referring to, then it can be obtained from | |
446ab5f5 | 1984 | struct <structname>snd_pcm_substream</structname> data passed to each callback |
1da177e4 LT |
1985 | as follows: |
1986 | ||
1987 | <informalexample> | |
1988 | <programlisting> | |
1989 | <![CDATA[ | |
446ab5f5 | 1990 | struct snd_pcm_substream *substream; |
1da177e4 LT |
1991 | int index = substream->number; |
1992 | ]]> | |
1993 | </programlisting> | |
1994 | </informalexample> | |
1995 | </para> | |
1996 | ||
1997 | <para> | |
1998 | After the pcm is created, you need to set operators for each | |
1999 | pcm stream. | |
2000 | ||
2001 | <informalexample> | |
2002 | <programlisting> | |
2003 | <![CDATA[ | |
2004 | snd_pcm_set_ops(pcm, SNDRV_PCM_STREAM_PLAYBACK, | |
2005 | &snd_mychip_playback_ops); | |
2006 | snd_pcm_set_ops(pcm, SNDRV_PCM_STREAM_CAPTURE, | |
2007 | &snd_mychip_capture_ops); | |
2008 | ]]> | |
2009 | </programlisting> | |
2010 | </informalexample> | |
2011 | </para> | |
2012 | ||
2013 | <para> | |
2014 | The operators are defined typically like this: | |
2015 | ||
2016 | <informalexample> | |
2017 | <programlisting> | |
2018 | <![CDATA[ | |
446ab5f5 | 2019 | static struct snd_pcm_ops snd_mychip_playback_ops = { |
1da177e4 LT |
2020 | .open = snd_mychip_pcm_open, |
2021 | .close = snd_mychip_pcm_close, | |
2022 | .ioctl = snd_pcm_lib_ioctl, | |
2023 | .hw_params = snd_mychip_pcm_hw_params, | |
2024 | .hw_free = snd_mychip_pcm_hw_free, | |
2025 | .prepare = snd_mychip_pcm_prepare, | |
2026 | .trigger = snd_mychip_pcm_trigger, | |
2027 | .pointer = snd_mychip_pcm_pointer, | |
2028 | }; | |
2029 | ]]> | |
2030 | </programlisting> | |
2031 | </informalexample> | |
2032 | ||
2033 | Each of callbacks is explained in the subsection | |
2034 | <link linkend="pcm-interface-operators"><citetitle> | |
2035 | Operators</citetitle></link>. | |
2036 | </para> | |
2037 | ||
2038 | <para> | |
2039 | After setting the operators, most likely you'd like to | |
2040 | pre-allocate the buffer. For the pre-allocation, simply call | |
2041 | the following: | |
2042 | ||
2043 | <informalexample> | |
2044 | <programlisting> | |
2045 | <![CDATA[ | |
2046 | snd_pcm_lib_preallocate_pages_for_all(pcm, SNDRV_DMA_TYPE_DEV, | |
2047 | snd_dma_pci_data(chip->pci), | |
2048 | 64*1024, 64*1024); | |
2049 | ]]> | |
2050 | </programlisting> | |
2051 | </informalexample> | |
2052 | ||
2053 | It will allocate up to 64kB buffer as default. The details of | |
2054 | buffer management will be described in the later section <link | |
2055 | linkend="buffer-and-memory"><citetitle>Buffer and Memory | |
2056 | Management</citetitle></link>. | |
2057 | </para> | |
2058 | ||
2059 | <para> | |
2060 | Additionally, you can set some extra information for this pcm | |
2061 | in pcm->info_flags. | |
2062 | The available values are defined as | |
2063 | <constant>SNDRV_PCM_INFO_XXX</constant> in | |
2064 | <filename><sound/asound.h></filename>, which is used for | |
2065 | the hardware definition (described later). When your soundchip | |
2066 | supports only half-duplex, specify like this: | |
2067 | ||
2068 | <informalexample> | |
2069 | <programlisting> | |
2070 | <![CDATA[ | |
2071 | pcm->info_flags = SNDRV_PCM_INFO_HALF_DUPLEX; | |
2072 | ]]> | |
2073 | </programlisting> | |
2074 | </informalexample> | |
2075 | </para> | |
2076 | </section> | |
2077 | ||
2078 | <section id="pcm-interface-destructor"> | |
2079 | <title>... And the Destructor?</title> | |
2080 | <para> | |
2081 | The destructor for a pcm instance is not always | |
2082 | necessary. Since the pcm device will be released by the middle | |
2083 | layer code automatically, you don't have to call destructor | |
2084 | explicitly. | |
2085 | </para> | |
2086 | ||
2087 | <para> | |
2088 | The destructor would be necessary when you created some | |
2089 | special records internally and need to release them. In such a | |
2090 | case, set the destructor function to | |
2091 | pcm->private_free: | |
2092 | ||
2093 | <example> | |
2094 | <title>PCM Instance with a Destructor</title> | |
2095 | <programlisting> | |
2096 | <![CDATA[ | |
446ab5f5 | 2097 | static void mychip_pcm_free(struct snd_pcm *pcm) |
1da177e4 | 2098 | { |
446ab5f5 | 2099 | struct mychip *chip = snd_pcm_chip(pcm); |
1da177e4 LT |
2100 | /* free your own data */ |
2101 | kfree(chip->my_private_pcm_data); | |
2102 | // do what you like else | |
2103 | .... | |
2104 | } | |
2105 | ||
446ab5f5 | 2106 | static int __devinit snd_mychip_new_pcm(struct mychip *chip) |
1da177e4 | 2107 | { |
446ab5f5 | 2108 | struct snd_pcm *pcm; |
1da177e4 LT |
2109 | .... |
2110 | /* allocate your own data */ | |
2111 | chip->my_private_pcm_data = kmalloc(...); | |
2112 | /* set the destructor */ | |
2113 | pcm->private_data = chip; | |
2114 | pcm->private_free = mychip_pcm_free; | |
2115 | .... | |
2116 | } | |
2117 | ]]> | |
2118 | </programlisting> | |
2119 | </example> | |
2120 | </para> | |
2121 | </section> | |
2122 | ||
2123 | <section id="pcm-interface-runtime"> | |
2124 | <title>Runtime Pointer - The Chest of PCM Information</title> | |
2125 | <para> | |
2126 | When the PCM substream is opened, a PCM runtime instance is | |
2127 | allocated and assigned to the substream. This pointer is | |
2128 | accessible via <constant>substream->runtime</constant>. | |
2129 | This runtime pointer holds the various information; it holds | |
2130 | the copy of hw_params and sw_params configurations, the buffer | |
2131 | pointers, mmap records, spinlocks, etc. Almost everyhing you | |
2132 | need for controlling the PCM can be found there. | |
2133 | </para> | |
2134 | ||
2135 | <para> | |
2136 | The definition of runtime instance is found in | |
2137 | <filename><sound/pcm.h></filename>. Here is the | |
2138 | copy from the file. | |
2139 | <informalexample> | |
2140 | <programlisting> | |
2141 | <![CDATA[ | |
2142 | struct _snd_pcm_runtime { | |
2143 | /* -- Status -- */ | |
446ab5f5 | 2144 | struct snd_pcm_substream *trigger_master; |
1da177e4 LT |
2145 | snd_timestamp_t trigger_tstamp; /* trigger timestamp */ |
2146 | int overrange; | |
2147 | snd_pcm_uframes_t avail_max; | |
2148 | snd_pcm_uframes_t hw_ptr_base; /* Position at buffer restart */ | |
2149 | snd_pcm_uframes_t hw_ptr_interrupt; /* Position at interrupt time*/ | |
2150 | ||
2151 | /* -- HW params -- */ | |
2152 | snd_pcm_access_t access; /* access mode */ | |
2153 | snd_pcm_format_t format; /* SNDRV_PCM_FORMAT_* */ | |
2154 | snd_pcm_subformat_t subformat; /* subformat */ | |
2155 | unsigned int rate; /* rate in Hz */ | |
2156 | unsigned int channels; /* channels */ | |
2157 | snd_pcm_uframes_t period_size; /* period size */ | |
2158 | unsigned int periods; /* periods */ | |
2159 | snd_pcm_uframes_t buffer_size; /* buffer size */ | |
2160 | unsigned int tick_time; /* tick time */ | |
2161 | snd_pcm_uframes_t min_align; /* Min alignment for the format */ | |
2162 | size_t byte_align; | |
2163 | unsigned int frame_bits; | |
2164 | unsigned int sample_bits; | |
2165 | unsigned int info; | |
2166 | unsigned int rate_num; | |
2167 | unsigned int rate_den; | |
2168 | ||
2169 | /* -- SW params -- */ | |
07799e75 | 2170 | struct timespec tstamp_mode; /* mmap timestamp is updated */ |
1da177e4 LT |
2171 | unsigned int period_step; |
2172 | unsigned int sleep_min; /* min ticks to sleep */ | |
2173 | snd_pcm_uframes_t xfer_align; /* xfer size need to be a multiple */ | |
2174 | snd_pcm_uframes_t start_threshold; | |
2175 | snd_pcm_uframes_t stop_threshold; | |
2176 | snd_pcm_uframes_t silence_threshold; /* Silence filling happens when | |
2177 | noise is nearest than this */ | |
2178 | snd_pcm_uframes_t silence_size; /* Silence filling size */ | |
2179 | snd_pcm_uframes_t boundary; /* pointers wrap point */ | |
2180 | ||
2181 | snd_pcm_uframes_t silenced_start; | |
2182 | snd_pcm_uframes_t silenced_size; | |
2183 | ||
2184 | snd_pcm_sync_id_t sync; /* hardware synchronization ID */ | |
2185 | ||
2186 | /* -- mmap -- */ | |
446ab5f5 TI |
2187 | volatile struct snd_pcm_mmap_status *status; |
2188 | volatile struct snd_pcm_mmap_control *control; | |
1da177e4 LT |
2189 | atomic_t mmap_count; |
2190 | ||
2191 | /* -- locking / scheduling -- */ | |
2192 | spinlock_t lock; | |
2193 | wait_queue_head_t sleep; | |
2194 | struct timer_list tick_timer; | |
2195 | struct fasync_struct *fasync; | |
2196 | ||
2197 | /* -- private section -- */ | |
2198 | void *private_data; | |
446ab5f5 | 2199 | void (*private_free)(struct snd_pcm_runtime *runtime); |
1da177e4 LT |
2200 | |
2201 | /* -- hardware description -- */ | |
446ab5f5 TI |
2202 | struct snd_pcm_hardware hw; |
2203 | struct snd_pcm_hw_constraints hw_constraints; | |
1da177e4 LT |
2204 | |
2205 | /* -- interrupt callbacks -- */ | |
446ab5f5 TI |
2206 | void (*transfer_ack_begin)(struct snd_pcm_substream *substream); |
2207 | void (*transfer_ack_end)(struct snd_pcm_substream *substream); | |
1da177e4 LT |
2208 | |
2209 | /* -- timer -- */ | |
2210 | unsigned int timer_resolution; /* timer resolution */ | |
2211 | ||
2212 | /* -- DMA -- */ | |
2213 | unsigned char *dma_area; /* DMA area */ | |
2214 | dma_addr_t dma_addr; /* physical bus address (not accessible from main CPU) */ | |
2215 | size_t dma_bytes; /* size of DMA area */ | |
2216 | ||
2217 | struct snd_dma_buffer *dma_buffer_p; /* allocated buffer */ | |
2218 | ||
2219 | #if defined(CONFIG_SND_PCM_OSS) || defined(CONFIG_SND_PCM_OSS_MODULE) | |
2220 | /* -- OSS things -- */ | |
446ab5f5 | 2221 | struct snd_pcm_oss_runtime oss; |
1da177e4 LT |
2222 | #endif |
2223 | }; | |
2224 | ]]> | |
2225 | </programlisting> | |
2226 | </informalexample> | |
2227 | </para> | |
2228 | ||
2229 | <para> | |
2230 | For the operators (callbacks) of each sound driver, most of | |
2231 | these records are supposed to be read-only. Only the PCM | |
2232 | middle-layer changes / updates these info. The exceptions are | |
2233 | the hardware description (hw), interrupt callbacks | |
2234 | (transfer_ack_xxx), DMA buffer information, and the private | |
2235 | data. Besides, if you use the standard buffer allocation | |
2236 | method via <function>snd_pcm_lib_malloc_pages()</function>, | |
2237 | you don't need to set the DMA buffer information by yourself. | |
2238 | </para> | |
2239 | ||
2240 | <para> | |
2241 | In the sections below, important records are explained. | |
2242 | </para> | |
2243 | ||
2244 | <section id="pcm-interface-runtime-hw"> | |
2245 | <title>Hardware Description</title> | |
2246 | <para> | |
446ab5f5 | 2247 | The hardware descriptor (struct <structname>snd_pcm_hardware</structname>) |
1da177e4 LT |
2248 | contains the definitions of the fundamental hardware |
2249 | configuration. Above all, you'll need to define this in | |
2250 | <link linkend="pcm-interface-operators-open-callback"><citetitle> | |
2251 | the open callback</citetitle></link>. | |
2252 | Note that the runtime instance holds the copy of the | |
2253 | descriptor, not the pointer to the existing descriptor. That | |
2254 | is, in the open callback, you can modify the copied descriptor | |
2255 | (<constant>runtime->hw</constant>) as you need. For example, if the maximum | |
2256 | number of channels is 1 only on some chip models, you can | |
2257 | still use the same hardware descriptor and change the | |
2258 | channels_max later: | |
2259 | <informalexample> | |
2260 | <programlisting> | |
2261 | <![CDATA[ | |
446ab5f5 | 2262 | struct snd_pcm_runtime *runtime = substream->runtime; |
1da177e4 LT |
2263 | ... |
2264 | runtime->hw = snd_mychip_playback_hw; /* common definition */ | |
2265 | if (chip->model == VERY_OLD_ONE) | |
2266 | runtime->hw.channels_max = 1; | |
2267 | ]]> | |
2268 | </programlisting> | |
2269 | </informalexample> | |
2270 | </para> | |
2271 | ||
2272 | <para> | |
2273 | Typically, you'll have a hardware descriptor like below: | |
2274 | <informalexample> | |
2275 | <programlisting> | |
2276 | <![CDATA[ | |
446ab5f5 | 2277 | static struct snd_pcm_hardware snd_mychip_playback_hw = { |
1da177e4 LT |
2278 | .info = (SNDRV_PCM_INFO_MMAP | |
2279 | SNDRV_PCM_INFO_INTERLEAVED | | |
2280 | SNDRV_PCM_INFO_BLOCK_TRANSFER | | |
2281 | SNDRV_PCM_INFO_MMAP_VALID), | |
2282 | .formats = SNDRV_PCM_FMTBIT_S16_LE, | |
2283 | .rates = SNDRV_PCM_RATE_8000_48000, | |
2284 | .rate_min = 8000, | |
2285 | .rate_max = 48000, | |
2286 | .channels_min = 2, | |
2287 | .channels_max = 2, | |
2288 | .buffer_bytes_max = 32768, | |
2289 | .period_bytes_min = 4096, | |
2290 | .period_bytes_max = 32768, | |
2291 | .periods_min = 1, | |
2292 | .periods_max = 1024, | |
2293 | }; | |
2294 | ]]> | |
2295 | </programlisting> | |
2296 | </informalexample> | |
2297 | </para> | |
2298 | ||
2299 | <para> | |
2300 | <itemizedlist> | |
2301 | <listitem><para> | |
2302 | The <structfield>info</structfield> field contains the type and | |
2303 | capabilities of this pcm. The bit flags are defined in | |
2304 | <filename><sound/asound.h></filename> as | |
2305 | <constant>SNDRV_PCM_INFO_XXX</constant>. Here, at least, you | |
2306 | have to specify whether the mmap is supported and which | |
2307 | interleaved format is supported. | |
2308 | When the mmap is supported, add | |
2309 | <constant>SNDRV_PCM_INFO_MMAP</constant> flag here. When the | |
2310 | hardware supports the interleaved or the non-interleaved | |
2311 | format, <constant>SNDRV_PCM_INFO_INTERLEAVED</constant> or | |
2312 | <constant>SNDRV_PCM_INFO_NONINTERLEAVED</constant> flag must | |
2313 | be set, respectively. If both are supported, you can set both, | |
2314 | too. | |
2315 | </para> | |
2316 | ||
2317 | <para> | |
2318 | In the above example, <constant>MMAP_VALID</constant> and | |
2319 | <constant>BLOCK_TRANSFER</constant> are specified for OSS mmap | |
2320 | mode. Usually both are set. Of course, | |
2321 | <constant>MMAP_VALID</constant> is set only if the mmap is | |
2322 | really supported. | |
2323 | </para> | |
2324 | ||
2325 | <para> | |
2326 | The other possible flags are | |
2327 | <constant>SNDRV_PCM_INFO_PAUSE</constant> and | |
2328 | <constant>SNDRV_PCM_INFO_RESUME</constant>. The | |
2329 | <constant>PAUSE</constant> bit means that the pcm supports the | |
2330 | <quote>pause</quote> operation, while the | |
2331 | <constant>RESUME</constant> bit means that the pcm supports | |
5fe76e4d TI |
2332 | the full <quote>suspend/resume</quote> operation. |
2333 | If <constant>PAUSE</constant> flag is set, | |
2334 | the <structfield>trigger</structfield> callback below | |
2335 | must handle the corresponding (pause push/release) commands. | |
2336 | The suspend/resume trigger commands can be defined even without | |
2337 | <constant>RESUME</constant> flag. See <link | |
2338 | linkend="power-management"><citetitle> | |
2339 | Power Management</citetitle></link> section for details. | |
1da177e4 LT |
2340 | </para> |
2341 | ||
2342 | <para> | |
2343 | When the PCM substreams can be synchronized (typically, | |
2344 | synchorinized start/stop of a playback and a capture streams), | |
2345 | you can give <constant>SNDRV_PCM_INFO_SYNC_START</constant>, | |
2346 | too. In this case, you'll need to check the linked-list of | |
2347 | PCM substreams in the trigger callback. This will be | |
2348 | described in the later section. | |
2349 | </para> | |
2350 | </listitem> | |
2351 | ||
2352 | <listitem> | |
2353 | <para> | |
2354 | <structfield>formats</structfield> field contains the bit-flags | |
2355 | of supported formats (<constant>SNDRV_PCM_FMTBIT_XXX</constant>). | |
2356 | If the hardware supports more than one format, give all or'ed | |
2357 | bits. In the example above, the signed 16bit little-endian | |
2358 | format is specified. | |
2359 | </para> | |
2360 | </listitem> | |
2361 | ||
2362 | <listitem> | |
2363 | <para> | |
2364 | <structfield>rates</structfield> field contains the bit-flags of | |
2365 | supported rates (<constant>SNDRV_PCM_RATE_XXX</constant>). | |
2366 | When the chip supports continuous rates, pass | |
2367 | <constant>CONTINUOUS</constant> bit additionally. | |
2368 | The pre-defined rate bits are provided only for typical | |
2369 | rates. If your chip supports unconventional rates, you need to add | |
2370 | <constant>KNOT</constant> bit and set up the hardware | |
2371 | constraint manually (explained later). | |
2372 | </para> | |
2373 | </listitem> | |
2374 | ||
2375 | <listitem> | |
2376 | <para> | |
2377 | <structfield>rate_min</structfield> and | |
2378 | <structfield>rate_max</structfield> define the minimal and | |
2379 | maximal sample rate. This should correspond somehow to | |
2380 | <structfield>rates</structfield> bits. | |
2381 | </para> | |
2382 | </listitem> | |
2383 | ||
2384 | <listitem> | |
2385 | <para> | |
2386 | <structfield>channel_min</structfield> and | |
2387 | <structfield>channel_max</structfield> | |
2388 | define, as you might already expected, the minimal and maximal | |
2389 | number of channels. | |
2390 | </para> | |
2391 | </listitem> | |
2392 | ||
2393 | <listitem> | |
2394 | <para> | |
2395 | <structfield>buffer_bytes_max</structfield> defines the | |
2396 | maximal buffer size in bytes. There is no | |
2397 | <structfield>buffer_bytes_min</structfield> field, since | |
2398 | it can be calculated from the minimal period size and the | |
2399 | minimal number of periods. | |
2400 | Meanwhile, <structfield>period_bytes_min</structfield> and | |
2401 | define the minimal and maximal size of the period in bytes. | |
2402 | <structfield>periods_max</structfield> and | |
2403 | <structfield>periods_min</structfield> define the maximal and | |
2404 | minimal number of periods in the buffer. | |
2405 | </para> | |
2406 | ||
2407 | <para> | |
2408 | The <quote>period</quote> is a term, that corresponds to | |
2409 | fragment in the OSS world. The period defines the size at | |
2410 | which the PCM interrupt is generated. This size strongly | |
2411 | depends on the hardware. | |
2412 | Generally, the smaller period size will give you more | |
2413 | interrupts, that is, more controls. | |
2414 | In the case of capture, this size defines the input latency. | |
2415 | On the other hand, the whole buffer size defines the | |
2416 | output latency for the playback direction. | |
2417 | </para> | |
2418 | </listitem> | |
2419 | ||
2420 | <listitem> | |
2421 | <para> | |
2422 | There is also a field <structfield>fifo_size</structfield>. | |
2423 | This specifies the size of the hardware FIFO, but it's not | |
2424 | used currently in the driver nor in the alsa-lib. So, you | |
2425 | can ignore this field. | |
2426 | </para> | |
2427 | </listitem> | |
2428 | </itemizedlist> | |
2429 | </para> | |
2430 | </section> | |
2431 | ||
2432 | <section id="pcm-interface-runtime-config"> | |
2433 | <title>PCM Configurations</title> | |
2434 | <para> | |
2435 | Ok, let's go back again to the PCM runtime records. | |
2436 | The most frequently referred records in the runtime instance are | |
2437 | the PCM configurations. | |
2438 | The PCM configurations are stored on runtime instance | |
2439 | after the application sends <type>hw_params</type> data via | |
2440 | alsa-lib. There are many fields copied from hw_params and | |
2441 | sw_params structs. For example, | |
2442 | <structfield>format</structfield> holds the format type | |
2443 | chosen by the application. This field contains the enum value | |
2444 | <constant>SNDRV_PCM_FORMAT_XXX</constant>. | |
2445 | </para> | |
2446 | ||
2447 | <para> | |
2448 | One thing to be noted is that the configured buffer and period | |
2449 | sizes are stored in <quote>frames</quote> in the runtime | |
2450 | In the ALSA world, 1 frame = channels * samples-size. | |
2451 | For conversion between frames and bytes, you can use the | |
2452 | helper functions, <function>frames_to_bytes()</function> and | |
2453 | <function>bytes_to_frames()</function>. | |
2454 | <informalexample> | |
2455 | <programlisting> | |
2456 | <![CDATA[ | |
2457 | period_bytes = frames_to_bytes(runtime, runtime->period_size); | |
2458 | ]]> | |
2459 | </programlisting> | |
2460 | </informalexample> | |
2461 | </para> | |
2462 | ||
2463 | <para> | |
2464 | Also, many software parameters (sw_params) are | |
2465 | stored in frames, too. Please check the type of the field. | |
2466 | <type>snd_pcm_uframes_t</type> is for the frames as unsigned | |
2467 | integer while <type>snd_pcm_sframes_t</type> is for the frames | |
2468 | as signed integer. | |
2469 | </para> | |
2470 | </section> | |
2471 | ||
2472 | <section id="pcm-interface-runtime-dma"> | |
2473 | <title>DMA Buffer Information</title> | |
2474 | <para> | |
2475 | The DMA buffer is defined by the following four fields, | |
2476 | <structfield>dma_area</structfield>, | |
2477 | <structfield>dma_addr</structfield>, | |
2478 | <structfield>dma_bytes</structfield> and | |
2479 | <structfield>dma_private</structfield>. | |
2480 | The <structfield>dma_area</structfield> holds the buffer | |
2481 | pointer (the logical address). You can call | |
2482 | <function>memcpy</function> from/to | |
2483 | this pointer. Meanwhile, <structfield>dma_addr</structfield> | |
2484 | holds the physical address of the buffer. This field is | |
2485 | specified only when the buffer is a linear buffer. | |
2486 | <structfield>dma_bytes</structfield> holds the size of buffer | |
2487 | in bytes. <structfield>dma_private</structfield> is used for | |
2488 | the ALSA DMA allocator. | |
2489 | </para> | |
2490 | ||
2491 | <para> | |
2492 | If you use a standard ALSA function, | |
2493 | <function>snd_pcm_lib_malloc_pages()</function>, for | |
2494 | allocating the buffer, these fields are set by the ALSA middle | |
2495 | layer, and you should <emphasis>not</emphasis> change them by | |
2496 | yourself. You can read them but not write them. | |
2497 | On the other hand, if you want to allocate the buffer by | |
2498 | yourself, you'll need to manage it in hw_params callback. | |
2499 | At least, <structfield>dma_bytes</structfield> is mandatory. | |
2500 | <structfield>dma_area</structfield> is necessary when the | |
2501 | buffer is mmapped. If your driver doesn't support mmap, this | |
2502 | field is not necessary. <structfield>dma_addr</structfield> | |
2503 | is also not mandatory. You can use | |
2504 | <structfield>dma_private</structfield> as you like, too. | |
2505 | </para> | |
2506 | </section> | |
2507 | ||
2508 | <section id="pcm-interface-runtime-status"> | |
2509 | <title>Running Status</title> | |
2510 | <para> | |
2511 | The running status can be referred via <constant>runtime->status</constant>. | |
446ab5f5 | 2512 | This is the pointer to struct <structname>snd_pcm_mmap_status</structname> |
1da177e4 LT |
2513 | record. For example, you can get the current DMA hardware |
2514 | pointer via <constant>runtime->status->hw_ptr</constant>. | |
2515 | </para> | |
2516 | ||
2517 | <para> | |
2518 | The DMA application pointer can be referred via | |
2519 | <constant>runtime->control</constant>, which points | |
446ab5f5 | 2520 | struct <structname>snd_pcm_mmap_control</structname> record. |
1da177e4 LT |
2521 | However, accessing directly to this value is not recommended. |
2522 | </para> | |
2523 | </section> | |
2524 | ||
2525 | <section id="pcm-interface-runtime-private"> | |
2526 | <title>Private Data</title> | |
2527 | <para> | |
2528 | You can allocate a record for the substream and store it in | |
2529 | <constant>runtime->private_data</constant>. Usually, this | |
2530 | done in | |
2531 | <link linkend="pcm-interface-operators-open-callback"><citetitle> | |
2532 | the open callback</citetitle></link>. | |
2533 | Don't mix this with <constant>pcm->private_data</constant>. | |
2534 | The <constant>pcm->private_data</constant> usually points the | |
2535 | chip instance assigned statically at the creation of PCM, while the | |
2536 | <constant>runtime->private_data</constant> points a dynamic | |
2537 | data created at the PCM open callback. | |
2538 | ||
2539 | <informalexample> | |
2540 | <programlisting> | |
2541 | <![CDATA[ | |
446ab5f5 | 2542 | static int snd_xxx_open(struct snd_pcm_substream *substream) |
1da177e4 | 2543 | { |
446ab5f5 | 2544 | struct my_pcm_data *data; |
1da177e4 LT |
2545 | .... |
2546 | data = kmalloc(sizeof(*data), GFP_KERNEL); | |
2547 | substream->runtime->private_data = data; | |
2548 | .... | |
2549 | } | |
2550 | ]]> | |
2551 | </programlisting> | |
2552 | </informalexample> | |
2553 | </para> | |
2554 | ||
2555 | <para> | |
2556 | The allocated object must be released in | |
2557 | <link linkend="pcm-interface-operators-open-callback"><citetitle> | |
2558 | the close callback</citetitle></link>. | |
2559 | </para> | |
2560 | </section> | |
2561 | ||
2562 | <section id="pcm-interface-runtime-intr"> | |
2563 | <title>Interrupt Callbacks</title> | |
2564 | <para> | |
2565 | The field <structfield>transfer_ack_begin</structfield> and | |
2566 | <structfield>transfer_ack_end</structfield> are called at | |
2567 | the beginning and the end of | |
2568 | <function>snd_pcm_period_elapsed()</function>, respectively. | |
2569 | </para> | |
2570 | </section> | |
2571 | ||
2572 | </section> | |
2573 | ||
2574 | <section id="pcm-interface-operators"> | |
2575 | <title>Operators</title> | |
2576 | <para> | |
2577 | OK, now let me explain the detail of each pcm callback | |
2578 | (<parameter>ops</parameter>). In general, every callback must | |
2579 | return 0 if successful, or a negative number with the error | |
2580 | number such as <constant>-EINVAL</constant> at any | |
2581 | error. | |
2582 | </para> | |
2583 | ||
2584 | <para> | |
2585 | The callback function takes at least the argument with | |
446ab5f5 | 2586 | <structname>snd_pcm_substream</structname> pointer. For retrieving the |
1da177e4 LT |
2587 | chip record from the given substream instance, you can use the |
2588 | following macro. | |
2589 | ||
2590 | <informalexample> | |
2591 | <programlisting> | |
2592 | <![CDATA[ | |
2593 | int xxx() { | |
446ab5f5 | 2594 | struct mychip *chip = snd_pcm_substream_chip(substream); |
1da177e4 LT |
2595 | .... |
2596 | } | |
2597 | ]]> | |
2598 | </programlisting> | |
2599 | </informalexample> | |
2600 | ||
2601 | The macro reads <constant>substream->private_data</constant>, | |
2602 | which is a copy of <constant>pcm->private_data</constant>. | |
2603 | You can override the former if you need to assign different data | |
2604 | records per PCM substream. For example, cmi8330 driver assigns | |
2605 | different private_data for playback and capture directions, | |
2606 | because it uses two different codecs (SB- and AD-compatible) for | |
2607 | different directions. | |
2608 | </para> | |
2609 | ||
2610 | <section id="pcm-interface-operators-open-callback"> | |
2611 | <title>open callback</title> | |
2612 | <para> | |
2613 | <informalexample> | |
2614 | <programlisting> | |
2615 | <![CDATA[ | |
446ab5f5 | 2616 | static int snd_xxx_open(struct snd_pcm_substream *substream); |
1da177e4 LT |
2617 | ]]> |
2618 | </programlisting> | |
2619 | </informalexample> | |
2620 | ||
2621 | This is called when a pcm substream is opened. | |
2622 | </para> | |
2623 | ||
2624 | <para> | |
2625 | At least, here you have to initialize the runtime->hw | |
2626 | record. Typically, this is done by like this: | |
2627 | ||
2628 | <informalexample> | |
2629 | <programlisting> | |
2630 | <![CDATA[ | |
446ab5f5 | 2631 | static int snd_xxx_open(struct snd_pcm_substream *substream) |
1da177e4 | 2632 | { |
446ab5f5 TI |
2633 | struct mychip *chip = snd_pcm_substream_chip(substream); |
2634 | struct snd_pcm_runtime *runtime = substream->runtime; | |
1da177e4 LT |
2635 | |
2636 | runtime->hw = snd_mychip_playback_hw; | |
2637 | return 0; | |
2638 | } | |
2639 | ]]> | |
2640 | </programlisting> | |
2641 | </informalexample> | |
2642 | ||
2643 | where <parameter>snd_mychip_playback_hw</parameter> is the | |
2644 | pre-defined hardware description. | |
2645 | </para> | |
2646 | ||
2647 | <para> | |
2648 | You can allocate a private data in this callback, as described | |
2649 | in <link linkend="pcm-interface-runtime-private"><citetitle> | |
2650 | Private Data</citetitle></link> section. | |
2651 | </para> | |
2652 | ||
2653 | <para> | |
2654 | If the hardware configuration needs more constraints, set the | |
2655 | hardware constraints here, too. | |
2656 | See <link linkend="pcm-interface-constraints"><citetitle> | |
2657 | Constraints</citetitle></link> for more details. | |
2658 | </para> | |
2659 | </section> | |
2660 | ||
2661 | <section id="pcm-interface-operators-close-callback"> | |
2662 | <title>close callback</title> | |
2663 | <para> | |
2664 | <informalexample> | |
2665 | <programlisting> | |
2666 | <![CDATA[ | |
446ab5f5 | 2667 | static int snd_xxx_close(struct snd_pcm_substream *substream); |
1da177e4 LT |
2668 | ]]> |
2669 | </programlisting> | |
2670 | </informalexample> | |
2671 | ||
2672 | Obviously, this is called when a pcm substream is closed. | |
2673 | </para> | |
2674 | ||
2675 | <para> | |
2676 | Any private instance for a pcm substream allocated in the | |
2677 | open callback will be released here. | |
2678 | ||
2679 | <informalexample> | |
2680 | <programlisting> | |
2681 | <![CDATA[ | |
446ab5f5 | 2682 | static int snd_xxx_close(struct snd_pcm_substream *substream) |
1da177e4 LT |
2683 | { |
2684 | .... | |
2685 | kfree(substream->runtime->private_data); | |
2686 | .... | |
2687 | } | |
2688 | ]]> | |
2689 | </programlisting> | |
2690 | </informalexample> | |
2691 | </para> | |
2692 | </section> | |
2693 | ||
2694 | <section id="pcm-interface-operators-ioctl-callback"> | |
2695 | <title>ioctl callback</title> | |
2696 | <para> | |
2697 | This is used for any special action to pcm ioctls. But | |
2698 | usually you can pass a generic ioctl callback, | |
2699 | <function>snd_pcm_lib_ioctl</function>. | |
2700 | </para> | |
2701 | </section> | |
2702 | ||
2703 | <section id="pcm-interface-operators-hw-params-callback"> | |
2704 | <title>hw_params callback</title> | |
2705 | <para> | |
2706 | <informalexample> | |
2707 | <programlisting> | |
2708 | <![CDATA[ | |
446ab5f5 TI |
2709 | static int snd_xxx_hw_params(struct snd_pcm_substream *substream, |
2710 | struct snd_pcm_hw_params *hw_params); | |
1da177e4 LT |
2711 | ]]> |
2712 | </programlisting> | |
2713 | </informalexample> | |
2714 | ||
2715 | This and <structfield>hw_free</structfield> callbacks exist | |
2716 | only on ALSA 0.9.x. | |
2717 | </para> | |
2718 | ||
2719 | <para> | |
2720 | This is called when the hardware parameter | |
2721 | (<structfield>hw_params</structfield>) is set | |
2722 | up by the application, | |
2723 | that is, once when the buffer size, the period size, the | |
2724 | format, etc. are defined for the pcm substream. | |
2725 | </para> | |
2726 | ||
2727 | <para> | |
2728 | Many hardware set-up should be done in this callback, | |
2729 | including the allocation of buffers. | |
2730 | </para> | |
2731 | ||
2732 | <para> | |
2733 | Parameters to be initialized are retrieved by | |
2734 | <function>params_xxx()</function> macros. For allocating a | |
2735 | buffer, you can call a helper function, | |
2736 | ||
2737 | <informalexample> | |
2738 | <programlisting> | |
2739 | <![CDATA[ | |
2740 | snd_pcm_lib_malloc_pages(substream, params_buffer_bytes(hw_params)); | |
2741 | ]]> | |
2742 | </programlisting> | |
2743 | </informalexample> | |
2744 | ||
2745 | <function>snd_pcm_lib_malloc_pages()</function> is available | |
2746 | only when the DMA buffers have been pre-allocated. | |
2747 | See the section <link | |
2748 | linkend="buffer-and-memory-buffer-types"><citetitle> | |
2749 | Buffer Types</citetitle></link> for more details. | |
2750 | </para> | |
2751 | ||
2752 | <para> | |
2753 | Note that this and <structfield>prepare</structfield> callbacks | |
2754 | may be called multiple times per initialization. | |
2755 | For example, the OSS emulation may | |
2756 | call these callbacks at each change via its ioctl. | |
2757 | </para> | |
2758 | ||
2759 | <para> | |
2760 | Thus, you need to take care not to allocate the same buffers | |
2761 | many times, which will lead to memory leak! Calling the | |
2762 | helper function above many times is OK. It will release the | |
2763 | previous buffer automatically when it was already allocated. | |
2764 | </para> | |
2765 | ||
2766 | <para> | |
2767 | Another note is that this callback is non-atomic | |
2768 | (schedulable). This is important, because the | |
2769 | <structfield>trigger</structfield> callback | |
2770 | is atomic (non-schedulable). That is, mutex or any | |
2771 | schedule-related functions are not available in | |
2772 | <structfield>trigger</structfield> callback. | |
2773 | Please see the subsection | |
2774 | <link linkend="pcm-interface-atomicity"><citetitle> | |
2775 | Atomicity</citetitle></link> for details. | |
2776 | </para> | |
2777 | </section> | |
2778 | ||
2779 | <section id="pcm-interface-operators-hw-free-callback"> | |
2780 | <title>hw_free callback</title> | |
2781 | <para> | |
2782 | <informalexample> | |
2783 | <programlisting> | |
2784 | <![CDATA[ | |
446ab5f5 | 2785 | static int snd_xxx_hw_free(struct snd_pcm_substream *substream); |
1da177e4 LT |
2786 | ]]> |
2787 | </programlisting> | |
2788 | </informalexample> | |
2789 | </para> | |
2790 | ||
2791 | <para> | |
2792 | This is called to release the resources allocated via | |
2793 | <structfield>hw_params</structfield>. For example, releasing the | |
2794 | buffer via | |
2795 | <function>snd_pcm_lib_malloc_pages()</function> is done by | |
2796 | calling the following: | |
2797 | ||
2798 | <informalexample> | |
2799 | <programlisting> | |
2800 | <![CDATA[ | |
2801 | snd_pcm_lib_free_pages(substream); | |
2802 | ]]> | |
2803 | </programlisting> | |
2804 | </informalexample> | |
2805 | </para> | |
2806 | ||
2807 | <para> | |
2808 | This function is always called before the close callback is called. | |
2809 | Also, the callback may be called multiple times, too. | |
2810 | Keep track whether the resource was already released. | |
2811 | </para> | |
2812 | </section> | |
2813 | ||
2814 | <section id="pcm-interface-operators-prepare-callback"> | |
2815 | <title>prepare callback</title> | |
2816 | <para> | |
2817 | <informalexample> | |
2818 | <programlisting> | |
2819 | <![CDATA[ | |
446ab5f5 | 2820 | static int snd_xxx_prepare(struct snd_pcm_substream *substream); |
1da177e4 LT |
2821 | ]]> |
2822 | </programlisting> | |
2823 | </informalexample> | |
2824 | </para> | |
2825 | ||
2826 | <para> | |
2827 | This callback is called when the pcm is | |
2828 | <quote>prepared</quote>. You can set the format type, sample | |
2829 | rate, etc. here. The difference from | |
2830 | <structfield>hw_params</structfield> is that the | |
2831 | <structfield>prepare</structfield> callback will be called at each | |
2832 | time | |
2833 | <function>snd_pcm_prepare()</function> is called, i.e. when | |
2834 | recovered after underruns, etc. | |
2835 | </para> | |
2836 | ||
2837 | <para> | |
2838 | Note that this callback became non-atomic since the recent version. | |
2839 | You can use schedule-related fucntions safely in this callback now. | |
2840 | </para> | |
2841 | ||
2842 | <para> | |
2843 | In this and the following callbacks, you can refer to the | |
2844 | values via the runtime record, | |
2845 | substream->runtime. | |
2846 | For example, to get the current | |
2847 | rate, format or channels, access to | |
2848 | runtime->rate, | |
2849 | runtime->format or | |
2850 | runtime->channels, respectively. | |
2851 | The physical address of the allocated buffer is set to | |
2852 | runtime->dma_area. The buffer and period sizes are | |
2853 | in runtime->buffer_size and runtime->period_size, | |
2854 | respectively. | |
2855 | </para> | |
2856 | ||
2857 | <para> | |
2858 | Be careful that this callback will be called many times at | |
2859 | each set up, too. | |
2860 | </para> | |
2861 | </section> | |
2862 | ||
2863 | <section id="pcm-interface-operators-trigger-callback"> | |
2864 | <title>trigger callback</title> | |
2865 | <para> | |
2866 | <informalexample> | |
2867 | <programlisting> | |
2868 | <![CDATA[ | |
446ab5f5 | 2869 | static int snd_xxx_trigger(struct snd_pcm_substream *substream, int cmd); |
1da177e4 LT |
2870 | ]]> |
2871 | </programlisting> | |
2872 | </informalexample> | |
2873 | ||
2874 | This is called when the pcm is started, stopped or paused. | |
2875 | </para> | |
2876 | ||
2877 | <para> | |
2878 | Which action is specified in the second argument, | |
2879 | <constant>SNDRV_PCM_TRIGGER_XXX</constant> in | |
2880 | <filename><sound/pcm.h></filename>. At least, | |
2881 | <constant>START</constant> and <constant>STOP</constant> | |
2882 | commands must be defined in this callback. | |
2883 | ||
2884 | <informalexample> | |
2885 | <programlisting> | |
2886 | <![CDATA[ | |
2887 | switch (cmd) { | |
2888 | case SNDRV_PCM_TRIGGER_START: | |
2889 | // do something to start the PCM engine | |
2890 | break; | |
2891 | case SNDRV_PCM_TRIGGER_STOP: | |
2892 | // do something to stop the PCM engine | |
2893 | break; | |
2894 | default: | |
2895 | return -EINVAL; | |
2896 | } | |
2897 | ]]> | |
2898 | </programlisting> | |
2899 | </informalexample> | |
2900 | </para> | |
2901 | ||
2902 | <para> | |
2903 | When the pcm supports the pause operation (given in info | |
2904 | field of the hardware table), <constant>PAUSE_PUSE</constant> | |
2905 | and <constant>PAUSE_RELEASE</constant> commands must be | |
2906 | handled here, too. The former is the command to pause the pcm, | |
2907 | and the latter to restart the pcm again. | |
2908 | </para> | |
2909 | ||
2910 | <para> | |
5fe76e4d TI |
2911 | When the pcm supports the suspend/resume operation, |
2912 | regardless of full or partial suspend/resume support, | |
1da177e4 LT |
2913 | <constant>SUSPEND</constant> and <constant>RESUME</constant> |
2914 | commands must be handled, too. | |
2915 | These commands are issued when the power-management status is | |
2916 | changed. Obviously, the <constant>SUSPEND</constant> and | |
2917 | <constant>RESUME</constant> | |
2918 | do suspend and resume of the pcm substream, and usually, they | |
2919 | are identical with <constant>STOP</constant> and | |
2920 | <constant>START</constant> commands, respectively. | |
5fe76e4d TI |
2921 | See <link linkend="power-management"><citetitle> |
2922 | Power Management</citetitle></link> section for details. | |
1da177e4 LT |
2923 | </para> |
2924 | ||
2925 | <para> | |
2926 | As mentioned, this callback is atomic. You cannot call | |
2927 | the function going to sleep. | |
2928 | The trigger callback should be as minimal as possible, | |
2929 | just really triggering the DMA. The other stuff should be | |
2930 | initialized hw_params and prepare callbacks properly | |
2931 | beforehand. | |
2932 | </para> | |
2933 | </section> | |
2934 | ||
2935 | <section id="pcm-interface-operators-pointer-callback"> | |
2936 | <title>pointer callback</title> | |
2937 | <para> | |
2938 | <informalexample> | |
2939 | <programlisting> | |
2940 | <![CDATA[ | |
446ab5f5 | 2941 | static snd_pcm_uframes_t snd_xxx_pointer(struct snd_pcm_substream *substream) |
1da177e4 LT |
2942 | ]]> |
2943 | </programlisting> | |
2944 | </informalexample> | |
2945 | ||
2946 | This callback is called when the PCM middle layer inquires | |
2947 | the current hardware position on the buffer. The position must | |
2948 | be returned in frames (which was in bytes on ALSA 0.5.x), | |
2949 | ranged from 0 to buffer_size - 1. | |
2950 | </para> | |
2951 | ||
2952 | <para> | |
2953 | This is called usually from the buffer-update routine in the | |
2954 | pcm middle layer, which is invoked when | |
2955 | <function>snd_pcm_period_elapsed()</function> is called in the | |
2956 | interrupt routine. Then the pcm middle layer updates the | |
2957 | position and calculates the available space, and wakes up the | |
2958 | sleeping poll threads, etc. | |
2959 | </para> | |
2960 | ||
2961 | <para> | |
2962 | This callback is also atomic. | |
2963 | </para> | |
2964 | </section> | |
2965 | ||
2966 | <section id="pcm-interface-operators-copy-silence"> | |
2967 | <title>copy and silence callbacks</title> | |
2968 | <para> | |
2969 | These callbacks are not mandatory, and can be omitted in | |
2970 | most cases. These callbacks are used when the hardware buffer | |
2971 | cannot be on the normal memory space. Some chips have their | |
2972 | own buffer on the hardware which is not mappable. In such a | |
2973 | case, you have to transfer the data manually from the memory | |
2974 | buffer to the hardware buffer. Or, if the buffer is | |
2975 | non-contiguous on both physical and virtual memory spaces, | |
2976 | these callbacks must be defined, too. | |
2977 | </para> | |
2978 | ||
2979 | <para> | |
2980 | If these two callbacks are defined, copy and set-silence | |
2981 | operations are done by them. The detailed will be described in | |
2982 | the later section <link | |
2983 | linkend="buffer-and-memory"><citetitle>Buffer and Memory | |
2984 | Management</citetitle></link>. | |
2985 | </para> | |
2986 | </section> | |
2987 | ||
2988 | <section id="pcm-interface-operators-ack"> | |
2989 | <title>ack callback</title> | |
2990 | <para> | |
2991 | This callback is also not mandatory. This callback is called | |
2992 | when the appl_ptr is updated in read or write operations. | |
2993 | Some drivers like emu10k1-fx and cs46xx need to track the | |
2994 | current appl_ptr for the internal buffer, and this callback | |
2995 | is useful only for such a purpose. | |
2996 | </para> | |
2997 | <para> | |
2998 | This callback is atomic. | |
2999 | </para> | |
3000 | </section> | |
3001 | ||
3002 | <section id="pcm-interface-operators-page-callback"> | |
3003 | <title>page callback</title> | |
3004 | ||
3005 | <para> | |
3006 | This callback is also not mandatory. This callback is used | |
3007 | mainly for the non-contiguous buffer. The mmap calls this | |
3008 | callback to get the page address. Some examples will be | |
3009 | explained in the later section <link | |
3010 | linkend="buffer-and-memory"><citetitle>Buffer and Memory | |
3011 | Management</citetitle></link>, too. | |
3012 | </para> | |
3013 | </section> | |
3014 | </section> | |
3015 | ||
3016 | <section id="pcm-interface-interrupt-handler"> | |
3017 | <title>Interrupt Handler</title> | |
3018 | <para> | |
3019 | The rest of pcm stuff is the PCM interrupt handler. The | |
3020 | role of PCM interrupt handler in the sound driver is to update | |
3021 | the buffer position and to tell the PCM middle layer when the | |
3022 | buffer position goes across the prescribed period size. To | |
3023 | inform this, call <function>snd_pcm_period_elapsed()</function> | |
3024 | function. | |
3025 | </para> | |
3026 | ||
3027 | <para> | |
3028 | There are several types of sound chips to generate the interrupts. | |
3029 | </para> | |
3030 | ||
3031 | <section id="pcm-interface-interrupt-handler-boundary"> | |
3032 | <title>Interrupts at the period (fragment) boundary</title> | |
3033 | <para> | |
3034 | This is the most frequently found type: the hardware | |
3035 | generates an interrupt at each period boundary. | |
3036 | In this case, you can call | |
3037 | <function>snd_pcm_period_elapsed()</function> at each | |
3038 | interrupt. | |
3039 | </para> | |
3040 | ||
3041 | <para> | |
3042 | <function>snd_pcm_period_elapsed()</function> takes the | |
3043 | substream pointer as its argument. Thus, you need to keep the | |
3044 | substream pointer accessible from the chip instance. For | |
3045 | example, define substream field in the chip record to hold the | |
3046 | current running substream pointer, and set the pointer value | |
3047 | at open callback (and reset at close callback). | |
3048 | </para> | |
3049 | ||
3050 | <para> | |
3051 | If you aquire a spinlock in the interrupt handler, and the | |
3052 | lock is used in other pcm callbacks, too, then you have to | |
3053 | release the lock before calling | |
3054 | <function>snd_pcm_period_elapsed()</function>, because | |
3055 | <function>snd_pcm_period_elapsed()</function> calls other pcm | |
3056 | callbacks inside. | |
3057 | </para> | |
3058 | ||
3059 | <para> | |
3060 | A typical coding would be like: | |
3061 | ||
3062 | <example> | |
3063 | <title>Interrupt Handler Case #1</title> | |
3064 | <programlisting> | |
3065 | <![CDATA[ | |
3066 | static irqreturn_t snd_mychip_interrupt(int irq, void *dev_id, | |
3067 | struct pt_regs *regs) | |
3068 | { | |
446ab5f5 | 3069 | struct mychip *chip = dev_id; |
1da177e4 LT |
3070 | spin_lock(&chip->lock); |
3071 | .... | |
3072 | if (pcm_irq_invoked(chip)) { | |
3073 | /* call updater, unlock before it */ | |
3074 | spin_unlock(&chip->lock); | |
3075 | snd_pcm_period_elapsed(chip->substream); | |
3076 | spin_lock(&chip->lock); | |
3077 | // acknowledge the interrupt if necessary | |
3078 | } | |
3079 | .... | |
3080 | spin_unlock(&chip->lock); | |
3081 | return IRQ_HANDLED; | |
3082 | } | |
3083 | ]]> | |
3084 | </programlisting> | |
3085 | </example> | |
3086 | </para> | |
3087 | </section> | |
3088 | ||
3089 | <section id="pcm-interface-interrupt-handler-timer"> | |
3090 | <title>High-frequent timer interrupts</title> | |
3091 | <para> | |
3092 | This is the case when the hardware doesn't generate interrupts | |
3093 | at the period boundary but do timer-interrupts at the fixed | |
3094 | timer rate (e.g. es1968 or ymfpci drivers). | |
3095 | In this case, you need to check the current hardware | |
3096 | position and accumulates the processed sample length at each | |
3097 | interrupt. When the accumulated size overcomes the period | |
3098 | size, call | |
3099 | <function>snd_pcm_period_elapsed()</function> and reset the | |
3100 | accumulator. | |
3101 | </para> | |
3102 | ||
3103 | <para> | |
3104 | A typical coding would be like the following. | |
3105 | ||
3106 | <example> | |
3107 | <title>Interrupt Handler Case #2</title> | |
3108 | <programlisting> | |
3109 | <![CDATA[ | |
3110 | static irqreturn_t snd_mychip_interrupt(int irq, void *dev_id, | |
3111 | struct pt_regs *regs) | |
3112 | { | |
446ab5f5 | 3113 | struct mychip *chip = dev_id; |
1da177e4 LT |
3114 | spin_lock(&chip->lock); |
3115 | .... | |
3116 | if (pcm_irq_invoked(chip)) { | |
3117 | unsigned int last_ptr, size; | |
3118 | /* get the current hardware pointer (in frames) */ | |
3119 | last_ptr = get_hw_ptr(chip); | |
3120 | /* calculate the processed frames since the | |
3121 | * last update | |
3122 | */ | |
3123 | if (last_ptr < chip->last_ptr) | |
3124 | size = runtime->buffer_size + last_ptr | |
3125 | - chip->last_ptr; | |
3126 | else | |
3127 | size = last_ptr - chip->last_ptr; | |
3128 | /* remember the last updated point */ | |
3129 | chip->last_ptr = last_ptr; | |
3130 | /* accumulate the size */ | |
3131 | chip->size += size; | |
3132 | /* over the period boundary? */ | |
3133 | if (chip->size >= runtime->period_size) { | |
3134 | /* reset the accumulator */ | |
3135 | chip->size %= runtime->period_size; | |
3136 | /* call updater */ | |
3137 | spin_unlock(&chip->lock); | |
3138 | snd_pcm_period_elapsed(substream); | |
3139 | spin_lock(&chip->lock); | |
3140 | } | |
3141 | // acknowledge the interrupt if necessary | |
3142 | } | |
3143 | .... | |
3144 | spin_unlock(&chip->lock); | |
3145 | return IRQ_HANDLED; | |
3146 | } | |
3147 | ]]> | |
3148 | </programlisting> | |
3149 | </example> | |
3150 | </para> | |
3151 | </section> | |
3152 | ||
3153 | <section id="pcm-interface-interrupt-handler-both"> | |
3154 | <title>On calling <function>snd_pcm_period_elapsed()</function></title> | |
3155 | <para> | |
3156 | In both cases, even if more than one period are elapsed, you | |
3157 | don't have to call | |
3158 | <function>snd_pcm_period_elapsed()</function> many times. Call | |
3159 | only once. And the pcm layer will check the current hardware | |
3160 | pointer and update to the latest status. | |
3161 | </para> | |
3162 | </section> | |
3163 | </section> | |
3164 | ||
3165 | <section id="pcm-interface-atomicity"> | |
3166 | <title>Atomicity</title> | |
3167 | <para> | |
3168 | One of the most important (and thus difficult to debug) problem | |
3169 | on the kernel programming is the race condition. | |
3170 | On linux kernel, usually it's solved via spin-locks or | |
3171 | semaphores. In general, if the race condition may | |
3172 | happen in the interrupt handler, it's handled as atomic, and you | |
3173 | have to use spinlock for protecting the critical session. If it | |
3174 | never happens in the interrupt and it may take relatively long | |
3175 | time, you should use semaphore. | |
3176 | </para> | |
3177 | ||
3178 | <para> | |
3179 | As already seen, some pcm callbacks are atomic and some are | |
3180 | not. For example, <parameter>hw_params</parameter> callback is | |
3181 | non-atomic, while <parameter>trigger</parameter> callback is | |
3182 | atomic. This means, the latter is called already in a spinlock | |
3183 | held by the PCM middle layer. Please take this atomicity into | |
3184 | account when you use a spinlock or a semaphore in the callbacks. | |
3185 | </para> | |
3186 | ||
3187 | <para> | |
3188 | In the atomic callbacks, you cannot use functions which may call | |
3189 | <function>schedule</function> or go to | |
3190 | <function>sleep</function>. The semaphore and mutex do sleep, | |
3191 | and hence they cannot be used inside the atomic callbacks | |
3192 | (e.g. <parameter>trigger</parameter> callback). | |
3193 | For taking a certain delay in such a callback, please use | |
3194 | <function>udelay()</function> or <function>mdelay()</function>. | |
3195 | </para> | |
3196 | ||
3197 | <para> | |
3198 | All three atomic callbacks (trigger, pointer, and ack) are | |
3199 | called with local interrupts disabled. | |
3200 | </para> | |
3201 | ||
3202 | </section> | |
3203 | <section id="pcm-interface-constraints"> | |
3204 | <title>Constraints</title> | |
3205 | <para> | |
3206 | If your chip supports unconventional sample rates, or only the | |
3207 | limited samples, you need to set a constraint for the | |
3208 | condition. | |
3209 | </para> | |
3210 | ||
3211 | <para> | |
3212 | For example, in order to restrict the sample rates in the some | |
3213 | supported values, use | |
3214 | <function>snd_pcm_hw_constraint_list()</function>. | |
3215 | You need to call this function in the open callback. | |
3216 | ||
3217 | <example> | |
3218 | <title>Example of Hardware Constraints</title> | |
3219 | <programlisting> | |
3220 | <![CDATA[ | |
3221 | static unsigned int rates[] = | |
3222 | {4000, 10000, 22050, 44100}; | |
446ab5f5 | 3223 | static struct snd_pcm_hw_constraint_list constraints_rates = { |
1da177e4 LT |
3224 | .count = ARRAY_SIZE(rates), |
3225 | .list = rates, | |
3226 | .mask = 0, | |
3227 | }; | |
3228 | ||
446ab5f5 | 3229 | static int snd_mychip_pcm_open(struct snd_pcm_substream *substream) |
1da177e4 LT |
3230 | { |
3231 | int err; | |
3232 | .... | |
3233 | err = snd_pcm_hw_constraint_list(substream->runtime, 0, | |
3234 | SNDRV_PCM_HW_PARAM_RATE, | |
3235 | &constraints_rates); | |
3236 | if (err < 0) | |
3237 | return err; | |
3238 | .... | |
3239 | } | |
3240 | ]]> | |
3241 | </programlisting> | |
3242 | </example> | |
3243 | </para> | |
3244 | ||
3245 | <para> | |
3246 | There are many different constraints. | |
3247 | Look in <filename>sound/pcm.h</filename> for a complete list. | |
3248 | You can even define your own constraint rules. | |
3249 | For example, let's suppose my_chip can manage a substream of 1 channel | |
3250 | if and only if the format is S16_LE, otherwise it supports any format | |
446ab5f5 | 3251 | specified in the <structname>snd_pcm_hardware</structname> stucture (or in any |
1da177e4 LT |
3252 | other constraint_list). You can build a rule like this: |
3253 | ||
3254 | <example> | |
3255 | <title>Example of Hardware Constraints for Channels</title> | |
3256 | <programlisting> | |
3257 | <![CDATA[ | |
446ab5f5 TI |
3258 | static int hw_rule_format_by_channels(struct snd_pcm_hw_params *params, |
3259 | struct snd_pcm_hw_rule *rule) | |
1da177e4 | 3260 | { |
446ab5f5 TI |
3261 | struct snd_interval *c = hw_param_interval(params, |
3262 | SNDRV_PCM_HW_PARAM_CHANNELS); | |
3263 | struct snd_mask *f = hw_param_mask(params, SNDRV_PCM_HW_PARAM_FORMAT); | |
3264 | struct snd_mask fmt; | |
1da177e4 LT |
3265 | |
3266 | snd_mask_any(&fmt); /* Init the struct */ | |
3267 | if (c->min < 2) { | |
3268 | fmt.bits[0] &= SNDRV_PCM_FMTBIT_S16_LE; | |
3269 | return snd_mask_refine(f, &fmt); | |
3270 | } | |
3271 | return 0; | |
3272 | } | |
3273 | ]]> | |
3274 | </programlisting> | |
3275 | </example> | |
3276 | </para> | |
3277 | ||
3278 | <para> | |
3279 | Then you need to call this function to add your rule: | |
3280 | ||
3281 | <informalexample> | |
3282 | <programlisting> | |
3283 | <![CDATA[ | |
3284 | snd_pcm_hw_rule_add(substream->runtime, 0, SNDRV_PCM_HW_PARAM_CHANNELS, | |
3285 | hw_rule_channels_by_format, 0, SNDRV_PCM_HW_PARAM_FORMAT, | |
3286 | -1); | |
3287 | ]]> | |
3288 | </programlisting> | |
3289 | </informalexample> | |
3290 | </para> | |
3291 | ||
3292 | <para> | |
3293 | The rule function is called when an application sets the number of | |
3294 | channels. But an application can set the format before the number of | |
3295 | channels. Thus you also need to define the inverse rule: | |
3296 | ||
3297 | <example> | |
3298 | <title>Example of Hardware Constraints for Channels</title> | |
3299 | <programlisting> | |
3300 | <![CDATA[ | |
446ab5f5 TI |
3301 | static int hw_rule_channels_by_format(struct snd_pcm_hw_params *params, |
3302 | struct snd_pcm_hw_rule *rule) | |
1da177e4 | 3303 | { |
446ab5f5 TI |
3304 | struct snd_interval *c = hw_param_interval(params, |
3305 | SNDRV_PCM_HW_PARAM_CHANNELS); | |
3306 | struct snd_mask *f = hw_param_mask(params, SNDRV_PCM_HW_PARAM_FORMAT); | |
3307 | struct snd_interval ch; | |
1da177e4 LT |
3308 | |
3309 | snd_interval_any(&ch); | |
3310 | if (f->bits[0] == SNDRV_PCM_FMTBIT_S16_LE) { | |
3311 | ch.min = ch.max = 1; | |
3312 | ch.integer = 1; | |
3313 | return snd_interval_refine(c, &ch); | |
3314 | } | |
3315 | return 0; | |
3316 | } | |
3317 | ]]> | |
3318 | </programlisting> | |
3319 | </example> | |
3320 | </para> | |
3321 | ||
3322 | <para> | |
3323 | ...and in the open callback: | |
3324 | <informalexample> | |
3325 | <programlisting> | |
3326 | <![CDATA[ | |
3327 | snd_pcm_hw_rule_add(substream->runtime, 0, SNDRV_PCM_HW_PARAM_FORMAT, | |
3328 | hw_rule_format_by_channels, 0, SNDRV_PCM_HW_PARAM_CHANNELS, | |
3329 | -1); | |
3330 | ]]> | |
3331 | </programlisting> | |
3332 | </informalexample> | |
3333 | </para> | |
3334 | ||
3335 | <para> | |
3336 | I won't explain more details here, rather I | |
3337 | would like to say, <quote>Luke, use the source.</quote> | |
3338 | </para> | |
3339 | </section> | |
3340 | ||
3341 | </chapter> | |
3342 | ||
3343 | ||
3344 | <!-- ****************************************************** --> | |
3345 | <!-- Control Interface --> | |
3346 | <!-- ****************************************************** --> | |
3347 | <chapter id="control-interface"> | |
3348 | <title>Control Interface</title> | |
3349 | ||
3350 | <section id="control-interface-general"> | |
3351 | <title>General</title> | |
3352 | <para> | |
3353 | The control interface is used widely for many switches, | |
3354 | sliders, etc. which are accessed from the user-space. Its most | |
3355 | important use is the mixer interface. In other words, on ALSA | |
3356 | 0.9.x, all the mixer stuff is implemented on the control kernel | |
3357 | API (while there was an independent mixer kernel API on 0.5.x). | |
3358 | </para> | |
3359 | ||
3360 | <para> | |
3361 | ALSA has a well-defined AC97 control module. If your chip | |
3362 | supports only the AC97 and nothing else, you can skip this | |
3363 | section. | |
3364 | </para> | |
3365 | ||
3366 | <para> | |
3367 | The control API is defined in | |
3368 | <filename><sound/control.h></filename>. | |
3369 | Include this file if you add your own controls. | |
3370 | </para> | |
3371 | </section> | |
3372 | ||
3373 | <section id="control-interface-definition"> | |
3374 | <title>Definition of Controls</title> | |
3375 | <para> | |
3376 | For creating a new control, you need to define the three | |
3377 | callbacks: <structfield>info</structfield>, | |
3378 | <structfield>get</structfield> and | |
3379 | <structfield>put</structfield>. Then, define a | |
446ab5f5 | 3380 | struct <structname>snd_kcontrol_new</structname> record, such as: |
1da177e4 LT |
3381 | |
3382 | <example> | |
3383 | <title>Definition of a Control</title> | |
3384 | <programlisting> | |
3385 | <![CDATA[ | |
446ab5f5 | 3386 | static struct snd_kcontrol_new my_control __devinitdata = { |
1da177e4 LT |
3387 | .iface = SNDRV_CTL_ELEM_IFACE_MIXER, |
3388 | .name = "PCM Playback Switch", | |
3389 | .index = 0, | |
3390 | .access = SNDRV_CTL_ELEM_ACCESS_READWRITE, | |
3391 | .private_values = 0xffff, | |
3392 | .info = my_control_info, | |
3393 | .get = my_control_get, | |
3394 | .put = my_control_put | |
3395 | }; | |
3396 | ]]> | |
3397 | </programlisting> | |
3398 | </example> | |
3399 | </para> | |
3400 | ||
3401 | <para> | |
3402 | Most likely the control is created via | |
3403 | <function>snd_ctl_new1()</function>, and in such a case, you can | |
3404 | add <parameter>__devinitdata</parameter> prefix to the | |
3405 | definition like above. | |
3406 | </para> | |
3407 | ||
3408 | <para> | |
3409 | The <structfield>iface</structfield> field specifies the type of | |
67ed4161 CL |
3410 | the control, <constant>SNDRV_CTL_ELEM_IFACE_XXX</constant>, which |
3411 | is usually <constant>MIXER</constant>. | |
3412 | Use <constant>CARD</constant> for global controls that are not | |
3413 | logically part of the mixer. | |
3414 | If the control is closely associated with some specific device on | |
3415 | the sound card, use <constant>HWDEP</constant>, | |
3416 | <constant>PCM</constant>, <constant>RAWMIDI</constant>, | |
3417 | <constant>TIMER</constant>, or <constant>SEQUENCER</constant>, and | |
3418 | specify the device number with the | |
3419 | <structfield>device</structfield> and | |
3420 | <structfield>subdevice</structfield> fields. | |
1da177e4 LT |
3421 | </para> |
3422 | ||
3423 | <para> | |
3424 | The <structfield>name</structfield> is the name identifier | |
3425 | string. On ALSA 0.9.x, the control name is very important, | |
3426 | because its role is classified from its name. There are | |
3427 | pre-defined standard control names. The details are described in | |
3428 | the subsection | |
3429 | <link linkend="control-interface-control-names"><citetitle> | |
3430 | Control Names</citetitle></link>. | |
3431 | </para> | |
3432 | ||
3433 | <para> | |
3434 | The <structfield>index</structfield> field holds the index number | |
3435 | of this control. If there are several different controls with | |
3436 | the same name, they can be distinguished by the index | |
3437 | number. This is the case when | |
3438 | several codecs exist on the card. If the index is zero, you can | |
3439 | omit the definition above. | |
3440 | </para> | |
3441 | ||
3442 | <para> | |
3443 | The <structfield>access</structfield> field contains the access | |
3444 | type of this control. Give the combination of bit masks, | |
3445 | <constant>SNDRV_CTL_ELEM_ACCESS_XXX</constant>, there. | |
3446 | The detailed will be explained in the subsection | |
3447 | <link linkend="control-interface-access-flags"><citetitle> | |
3448 | Access Flags</citetitle></link>. | |
3449 | </para> | |
3450 | ||
3451 | <para> | |
3452 | The <structfield>private_values</structfield> field contains | |
3453 | an arbitrary long integer value for this record. When using | |
3454 | generic <structfield>info</structfield>, | |
3455 | <structfield>get</structfield> and | |
3456 | <structfield>put</structfield> callbacks, you can pass a value | |
3457 | through this field. If several small numbers are necessary, you can | |
3458 | combine them in bitwise. Or, it's possible to give a pointer | |
3459 | (casted to unsigned long) of some record to this field, too. | |
3460 | </para> | |
3461 | ||
3462 | <para> | |
3463 | The other three are | |
3464 | <link linkend="control-interface-callbacks"><citetitle> | |
3465 | callback functions</citetitle></link>. | |
3466 | </para> | |
3467 | </section> | |
3468 | ||
3469 | <section id="control-interface-control-names"> | |
3470 | <title>Control Names</title> | |
3471 | <para> | |
3472 | There are some standards for defining the control names. A | |
3473 | control is usually defined from the three parts as | |
3474 | <quote>SOURCE DIRECTION FUNCTION</quote>. | |
3475 | </para> | |
3476 | ||
3477 | <para> | |
3478 | The first, <constant>SOURCE</constant>, specifies the source | |
3479 | of the control, and is a string such as <quote>Master</quote>, | |
3480 | <quote>PCM</quote>, <quote>CD</quote> or | |
3481 | <quote>Line</quote>. There are many pre-defined sources. | |
3482 | </para> | |
3483 | ||
3484 | <para> | |
3485 | The second, <constant>DIRECTION</constant>, is one of the | |
3486 | following strings according to the direction of the control: | |
3487 | <quote>Playback</quote>, <quote>Capture</quote>, <quote>Bypass | |
3488 | Playback</quote> and <quote>Bypass Capture</quote>. Or, it can | |
3489 | be omitted, meaning both playback and capture directions. | |
3490 | </para> | |
3491 | ||
3492 | <para> | |
3493 | The third, <constant>FUNCTION</constant>, is one of the | |
3494 | following strings according to the function of the control: | |
3495 | <quote>Switch</quote>, <quote>Volume</quote> and | |
3496 | <quote>Route</quote>. | |
3497 | </para> | |
3498 | ||
3499 | <para> | |
3500 | The example of control names are, thus, <quote>Master Capture | |
3501 | Switch</quote> or <quote>PCM Playback Volume</quote>. | |
3502 | </para> | |
3503 | ||
3504 | <para> | |
3505 | There are some exceptions: | |
3506 | </para> | |
3507 | ||
3508 | <section id="control-interface-control-names-global"> | |
3509 | <title>Global capture and playback</title> | |
3510 | <para> | |
3511 | <quote>Capture Source</quote>, <quote>Capture Switch</quote> | |
3512 | and <quote>Capture Volume</quote> are used for the global | |
3513 | capture (input) source, switch and volume. Similarly, | |
3514 | <quote>Playback Switch</quote> and <quote>Playback | |
3515 | Volume</quote> are used for the global output gain switch and | |
3516 | volume. | |
3517 | </para> | |
3518 | </section> | |
3519 | ||
3520 | <section id="control-interface-control-names-tone"> | |
3521 | <title>Tone-controls</title> | |
3522 | <para> | |
3523 | tone-control switch and volumes are specified like | |
3524 | <quote>Tone Control - XXX</quote>, e.g. <quote>Tone Control - | |
3525 | Switch</quote>, <quote>Tone Control - Bass</quote>, | |
3526 | <quote>Tone Control - Center</quote>. | |
3527 | </para> | |
3528 | </section> | |
3529 | ||
3530 | <section id="control-interface-control-names-3d"> | |
3531 | <title>3D controls</title> | |
3532 | <para> | |
3533 | 3D-control switches and volumes are specified like <quote>3D | |
3534 | Control - XXX</quote>, e.g. <quote>3D Control - | |
3535 | Switch</quote>, <quote>3D Control - Center</quote>, <quote>3D | |
3536 | Control - Space</quote>. | |
3537 | </para> | |
3538 | </section> | |
3539 | ||
3540 | <section id="control-interface-control-names-mic"> | |
3541 | <title>Mic boost</title> | |
3542 | <para> | |
3543 | Mic-boost switch is set as <quote>Mic Boost</quote> or | |
3544 | <quote>Mic Boost (6dB)</quote>. | |
3545 | </para> | |
3546 | ||
3547 | <para> | |
3548 | More precise information can be found in | |
3549 | <filename>Documentation/sound/alsa/ControlNames.txt</filename>. | |
3550 | </para> | |
3551 | </section> | |
3552 | </section> | |
3553 | ||
3554 | <section id="control-interface-access-flags"> | |
3555 | <title>Access Flags</title> | |
3556 | ||
3557 | <para> | |
3558 | The access flag is the bit-flags which specifies the access type | |
3559 | of the given control. The default access type is | |
3560 | <constant>SNDRV_CTL_ELEM_ACCESS_READWRITE</constant>, | |
3561 | which means both read and write are allowed to this control. | |
3562 | When the access flag is omitted (i.e. = 0), it is | |
3563 | regarded as <constant>READWRITE</constant> access as default. | |
3564 | </para> | |
3565 | ||
3566 | <para> | |
3567 | When the control is read-only, pass | |
3568 | <constant>SNDRV_CTL_ELEM_ACCESS_READ</constant> instead. | |
3569 | In this case, you don't have to define | |
3570 | <structfield>put</structfield> callback. | |
3571 | Similarly, when the control is write-only (although it's a rare | |
3572 | case), you can use <constant>WRITE</constant> flag instead, and | |
3573 | you don't need <structfield>get</structfield> callback. | |
3574 | </para> | |
3575 | ||
3576 | <para> | |
3577 | If the control value changes frequently (e.g. the VU meter), | |
3578 | <constant>VOLATILE</constant> flag should be given. This means | |
3579 | that the control may be changed without | |
3580 | <link linkend="control-interface-change-notification"><citetitle> | |
3581 | notification</citetitle></link>. Applications should poll such | |
3582 | a control constantly. | |
3583 | </para> | |
3584 | ||
3585 | <para> | |
3586 | When the control is inactive, set | |
3587 | <constant>INACTIVE</constant> flag, too. | |
3588 | There are <constant>LOCK</constant> and | |
3589 | <constant>OWNER</constant> flags for changing the write | |
3590 | permissions. | |
3591 | </para> | |
3592 | ||
3593 | </section> | |
3594 | ||
3595 | <section id="control-interface-callbacks"> | |
3596 | <title>Callbacks</title> | |
3597 | ||
3598 | <section id="control-interface-callbacks-info"> | |
3599 | <title>info callback</title> | |
3600 | <para> | |
3601 | The <structfield>info</structfield> callback is used to get | |
3602 | the detailed information of this control. This must store the | |
446ab5f5 | 3603 | values of the given struct <structname>snd_ctl_elem_info</structname> |
1da177e4 LT |
3604 | object. For example, for a boolean control with a single |
3605 | element will be: | |
3606 | ||
3607 | <example> | |
3608 | <title>Example of info callback</title> | |
3609 | <programlisting> | |
3610 | <![CDATA[ | |
446ab5f5 TI |
3611 | static int snd_myctl_info(struct snd_kcontrol *kcontrol, |
3612 | struct snd_ctl_elem_info *uinfo) | |
1da177e4 LT |
3613 | { |
3614 | uinfo->type = SNDRV_CTL_ELEM_TYPE_BOOLEAN; | |
3615 | uinfo->count = 1; | |
3616 | uinfo->value.integer.min = 0; | |
3617 | uinfo->value.integer.max = 1; | |
3618 | return 0; | |
3619 | } | |
3620 | ]]> | |
3621 | </programlisting> | |
3622 | </example> | |
3623 | </para> | |
3624 | ||
3625 | <para> | |
3626 | The <structfield>type</structfield> field specifies the type | |
3627 | of the control. There are <constant>BOOLEAN</constant>, | |
3628 | <constant>INTEGER</constant>, <constant>ENUMERATED</constant>, | |
3629 | <constant>BYTES</constant>, <constant>IEC958</constant> and | |
3630 | <constant>INTEGER64</constant>. The | |
3631 | <structfield>count</structfield> field specifies the | |
3632 | number of elements in this control. For example, a stereo | |
3633 | volume would have count = 2. The | |
3634 | <structfield>value</structfield> field is a union, and | |
3635 | the values stored are depending on the type. The boolean and | |
3636 | integer are identical. | |
3637 | </para> | |
3638 | ||
3639 | <para> | |
3640 | The enumerated type is a bit different from others. You'll | |
3641 | need to set the string for the currently given item index. | |
3642 | ||
3643 | <informalexample> | |
3644 | <programlisting> | |
3645 | <![CDATA[ | |
446ab5f5 TI |
3646 | static int snd_myctl_info(struct snd_kcontrol *kcontrol, |
3647 | struct snd_ctl_elem_info *uinfo) | |
1da177e4 LT |
3648 | { |
3649 | static char *texts[4] = { | |
3650 | "First", "Second", "Third", "Fourth" | |
3651 | }; | |
3652 | uinfo->type = SNDRV_CTL_ELEM_TYPE_ENUMERATED; | |
3653 | uinfo->count = 1; | |
3654 | uinfo->value.enumerated.items = 4; | |
3655 | if (uinfo->value.enumerated.item > 3) | |
3656 | uinfo->value.enumerated.item = 3; | |
3657 | strcpy(uinfo->value.enumerated.name, | |
3658 | texts[uinfo->value.enumerated.item]); | |
3659 | return 0; | |
3660 | } | |
3661 | ]]> | |
3662 | </programlisting> | |
3663 | </informalexample> | |
3664 | </para> | |
3665 | </section> | |
3666 | ||
3667 | <section id="control-interface-callbacks-get"> | |
3668 | <title>get callback</title> | |
3669 | ||
3670 | <para> | |
3671 | This callback is used to read the current value of the | |
3672 | control and to return to the user-space. | |
3673 | </para> | |
3674 | ||
3675 | <para> | |
3676 | For example, | |
3677 | ||
3678 | <example> | |
3679 | <title>Example of get callback</title> | |
3680 | <programlisting> | |
3681 | <![CDATA[ | |
446ab5f5 TI |
3682 | static int snd_myctl_get(struct snd_kcontrol *kcontrol, |
3683 | struct snd_ctl_elem_value *ucontrol) | |
1da177e4 | 3684 | { |
446ab5f5 | 3685 | struct mychip *chip = snd_kcontrol_chip(kcontrol); |
1da177e4 LT |
3686 | ucontrol->value.integer.value[0] = get_some_value(chip); |
3687 | return 0; | |
3688 | } | |
3689 | ]]> | |
3690 | </programlisting> | |
3691 | </example> | |
3692 | </para> | |
3693 | ||
3694 | <para> | |
3695 | Here, the chip instance is retrieved via | |
3696 | <function>snd_kcontrol_chip()</function> macro. This macro | |
063859c8 | 3697 | just accesses to kcontrol->private_data. The |
1da177e4 LT |
3698 | kcontrol->private_data field is |
3699 | given as the argument of <function>snd_ctl_new()</function> | |
3700 | (see the later subsection | |
3701 | <link linkend="control-interface-constructor"><citetitle>Constructor</citetitle></link>). | |
3702 | </para> | |
3703 | ||
3704 | <para> | |
3705 | The <structfield>value</structfield> field is depending on | |
3706 | the type of control as well as on info callback. For example, | |
3707 | the sb driver uses this field to store the register offset, | |
3708 | the bit-shift and the bit-mask. The | |
3709 | <structfield>private_value</structfield> is set like | |
3710 | <informalexample> | |
3711 | <programlisting> | |
3712 | <![CDATA[ | |
3713 | .private_value = reg | (shift << 16) | (mask << 24) | |
3714 | ]]> | |
3715 | </programlisting> | |
3716 | </informalexample> | |
3717 | and is retrieved in callbacks like | |
3718 | <informalexample> | |
3719 | <programlisting> | |
3720 | <![CDATA[ | |
446ab5f5 TI |
3721 | static int snd_sbmixer_get_single(struct snd_kcontrol *kcontrol, |
3722 | struct snd_ctl_elem_value *ucontrol) | |
1da177e4 LT |
3723 | { |
3724 | int reg = kcontrol->private_value & 0xff; | |
3725 | int shift = (kcontrol->private_value >> 16) & 0xff; | |
3726 | int mask = (kcontrol->private_value >> 24) & 0xff; | |
3727 | .... | |
3728 | } | |
3729 | ]]> | |
3730 | </programlisting> | |
3731 | </informalexample> | |
3732 | </para> | |
3733 | ||
3734 | <para> | |
3735 | In <structfield>get</structfield> callback, you have to fill all the elements if the | |
3736 | control has more than one elements, | |
3737 | i.e. <structfield>count</structfield> > 1. | |
3738 | In the example above, we filled only one element | |
3739 | (<structfield>value.integer.value[0]</structfield>) since it's | |
3740 | assumed as <structfield>count</structfield> = 1. | |
3741 | </para> | |
3742 | </section> | |
3743 | ||
3744 | <section id="control-interface-callbacks-put"> | |
3745 | <title>put callback</title> | |
3746 | ||
3747 | <para> | |
3748 | This callback is used to write a value from the user-space. | |
3749 | </para> | |
3750 | ||
3751 | <para> | |
3752 | For example, | |
3753 | ||
3754 | <example> | |
3755 | <title>Example of put callback</title> | |
3756 | <programlisting> | |
3757 | <![CDATA[ | |
446ab5f5 TI |
3758 | static int snd_myctl_put(struct snd_kcontrol *kcontrol, |
3759 | struct snd_ctl_elem_value *ucontrol) | |
1da177e4 | 3760 | { |
446ab5f5 | 3761 | struct mychip *chip = snd_kcontrol_chip(kcontrol); |
1da177e4 LT |
3762 | int changed = 0; |
3763 | if (chip->current_value != | |
3764 | ucontrol->value.integer.value[0]) { | |
3765 | change_current_value(chip, | |
3766 | ucontrol->value.integer.value[0]); | |
3767 | changed = 1; | |
3768 | } | |
3769 | return changed; | |
3770 | } | |
3771 | ]]> | |
3772 | </programlisting> | |
3773 | </example> | |
3774 | ||
3775 | As seen above, you have to return 1 if the value is | |
3776 | changed. If the value is not changed, return 0 instead. | |
3777 | If any fatal error happens, return a negative error code as | |
3778 | usual. | |
3779 | </para> | |
3780 | ||
3781 | <para> | |
3782 | Like <structfield>get</structfield> callback, | |
3783 | when the control has more than one elements, | |
3784 | all elemehts must be evaluated in this callback, too. | |
3785 | </para> | |
3786 | </section> | |
3787 | ||
3788 | <section id="control-interface-callbacks-all"> | |
3789 | <title>Callbacks are not atomic</title> | |
3790 | <para> | |
3791 | All these three callbacks are basically not atomic. | |
3792 | </para> | |
3793 | </section> | |
3794 | </section> | |
3795 | ||
3796 | <section id="control-interface-constructor"> | |
3797 | <title>Constructor</title> | |
3798 | <para> | |
3799 | When everything is ready, finally we can create a new | |
3800 | control. For creating a control, there are two functions to be | |
3801 | called, <function>snd_ctl_new1()</function> and | |
3802 | <function>snd_ctl_add()</function>. | |
3803 | </para> | |
3804 | ||
3805 | <para> | |
3806 | In the simplest way, you can do like this: | |
3807 | ||
3808 | <informalexample> | |
3809 | <programlisting> | |
3810 | <![CDATA[ | |
3811 | if ((err = snd_ctl_add(card, snd_ctl_new1(&my_control, chip))) < 0) | |
3812 | return err; | |
3813 | ]]> | |
3814 | </programlisting> | |
3815 | </informalexample> | |
3816 | ||
3817 | where <parameter>my_control</parameter> is the | |
446ab5f5 | 3818 | struct <structname>snd_kcontrol_new</structname> object defined above, and chip |
1da177e4 LT |
3819 | is the object pointer to be passed to |
3820 | kcontrol->private_data | |
3821 | which can be referred in callbacks. | |
3822 | </para> | |
3823 | ||
3824 | <para> | |
3825 | <function>snd_ctl_new1()</function> allocates a new | |
446ab5f5 | 3826 | <structname>snd_kcontrol</structname> instance (that's why the definition |
1da177e4 LT |
3827 | of <parameter>my_control</parameter> can be with |
3828 | <parameter>__devinitdata</parameter> | |
3829 | prefix), and <function>snd_ctl_add</function> assigns the given | |
3830 | control component to the card. | |
3831 | </para> | |
3832 | </section> | |
3833 | ||
3834 | <section id="control-interface-change-notification"> | |
3835 | <title>Change Notification</title> | |
3836 | <para> | |
3837 | If you need to change and update a control in the interrupt | |
3838 | routine, you can call <function>snd_ctl_notify()</function>. For | |
3839 | example, | |
3840 | ||
3841 | <informalexample> | |
3842 | <programlisting> | |
3843 | <![CDATA[ | |
3844 | snd_ctl_notify(card, SNDRV_CTL_EVENT_MASK_VALUE, id_pointer); | |
3845 | ]]> | |
3846 | </programlisting> | |
3847 | </informalexample> | |
3848 | ||
3849 | This function takes the card pointer, the event-mask, and the | |
3850 | control id pointer for the notification. The event-mask | |
3851 | specifies the types of notification, for example, in the above | |
3852 | example, the change of control values is notified. | |
446ab5f5 | 3853 | The id pointer is the pointer of struct <structname>snd_ctl_elem_id</structname> |
1da177e4 LT |
3854 | to be notified. |
3855 | You can find some examples in <filename>es1938.c</filename> or | |
3856 | <filename>es1968.c</filename> for hardware volume interrupts. | |
3857 | </para> | |
3858 | </section> | |
3859 | ||
3860 | </chapter> | |
3861 | ||
3862 | ||
3863 | <!-- ****************************************************** --> | |
3864 | <!-- API for AC97 Codec --> | |
3865 | <!-- ****************************************************** --> | |
3866 | <chapter id="api-ac97"> | |
3867 | <title>API for AC97 Codec</title> | |
3868 | ||
3869 | <section> | |
3870 | <title>General</title> | |
3871 | <para> | |
3872 | The ALSA AC97 codec layer is a well-defined one, and you don't | |
3873 | have to write many codes to control it. Only low-level control | |
3874 | routines are necessary. The AC97 codec API is defined in | |
3875 | <filename><sound/ac97_codec.h></filename>. | |
3876 | </para> | |
3877 | </section> | |
3878 | ||
3879 | <section id="api-ac97-example"> | |
3880 | <title>Full Code Example</title> | |
3881 | <para> | |
3882 | <example> | |
3883 | <title>Example of AC97 Interface</title> | |
3884 | <programlisting> | |
3885 | <![CDATA[ | |
446ab5f5 | 3886 | struct mychip { |
1da177e4 | 3887 | .... |
446ab5f5 | 3888 | struct snd_ac97 *ac97; |
1da177e4 LT |
3889 | .... |
3890 | }; | |
3891 | ||
446ab5f5 | 3892 | static unsigned short snd_mychip_ac97_read(struct snd_ac97 *ac97, |
1da177e4 LT |
3893 | unsigned short reg) |
3894 | { | |
446ab5f5 | 3895 | struct mychip *chip = ac97->private_data; |
1da177e4 LT |
3896 | .... |
3897 | // read a register value here from the codec | |
3898 | return the_register_value; | |
3899 | } | |
3900 | ||
446ab5f5 | 3901 | static void snd_mychip_ac97_write(struct snd_ac97 *ac97, |
1da177e4 LT |
3902 | unsigned short reg, unsigned short val) |
3903 | { | |
446ab5f5 | 3904 | struct mychip *chip = ac97->private_data; |
1da177e4 LT |
3905 | .... |
3906 | // write the given register value to the codec | |
3907 | } | |
3908 | ||
446ab5f5 | 3909 | static int snd_mychip_ac97(struct mychip *chip) |
1da177e4 | 3910 | { |
446ab5f5 TI |
3911 | struct snd_ac97_bus *bus; |
3912 | struct snd_ac97_template ac97; | |
1da177e4 | 3913 | int err; |
446ab5f5 | 3914 | static struct snd_ac97_bus_ops ops = { |
1da177e4 LT |
3915 | .write = snd_mychip_ac97_write, |
3916 | .read = snd_mychip_ac97_read, | |
3917 | }; | |
3918 | ||
3919 | if ((err = snd_ac97_bus(chip->card, 0, &ops, NULL, &bus)) < 0) | |
3920 | return err; | |
3921 | memset(&ac97, 0, sizeof(ac97)); | |
3922 | ac97.private_data = chip; | |
3923 | return snd_ac97_mixer(bus, &ac97, &chip->ac97); | |
3924 | } | |
3925 | ||
3926 | ]]> | |
3927 | </programlisting> | |
3928 | </example> | |
3929 | </para> | |
3930 | </section> | |
3931 | ||
3932 | <section id="api-ac97-constructor"> | |
3933 | <title>Constructor</title> | |
3934 | <para> | |
3935 | For creating an ac97 instance, first call <function>snd_ac97_bus</function> | |
3936 | with an <type>ac97_bus_ops_t</type> record with callback functions. | |
3937 | ||
3938 | <informalexample> | |
3939 | <programlisting> | |
3940 | <![CDATA[ | |
446ab5f5 TI |
3941 | struct snd_ac97_bus *bus; |
3942 | static struct snd_ac97_bus_ops ops = { | |
1da177e4 LT |
3943 | .write = snd_mychip_ac97_write, |
3944 | .read = snd_mychip_ac97_read, | |
3945 | }; | |
3946 | ||
3947 | snd_ac97_bus(card, 0, &ops, NULL, &pbus); | |
3948 | ]]> | |
3949 | </programlisting> | |
3950 | </informalexample> | |
3951 | ||
3952 | The bus record is shared among all belonging ac97 instances. | |
3953 | </para> | |
3954 | ||
3955 | <para> | |
446ab5f5 TI |
3956 | And then call <function>snd_ac97_mixer()</function> with an |
3957 | struct <structname>snd_ac97_template</structname> | |
1da177e4 LT |
3958 | record together with the bus pointer created above. |
3959 | ||
3960 | <informalexample> | |
3961 | <programlisting> | |
3962 | <![CDATA[ | |
446ab5f5 | 3963 | struct snd_ac97_template ac97; |
1da177e4 LT |
3964 | int err; |
3965 | ||
3966 | memset(&ac97, 0, sizeof(ac97)); | |
3967 | ac97.private_data = chip; | |
3968 | snd_ac97_mixer(bus, &ac97, &chip->ac97); | |
3969 | ]]> | |
3970 | </programlisting> | |
3971 | </informalexample> | |
3972 | ||
3973 | where chip->ac97 is the pointer of a newly created | |
3974 | <type>ac97_t</type> instance. | |
3975 | In this case, the chip pointer is set as the private data, so that | |
3976 | the read/write callback functions can refer to this chip instance. | |
3977 | This instance is not necessarily stored in the chip | |
3978 | record. When you need to change the register values from the | |
3979 | driver, or need the suspend/resume of ac97 codecs, keep this | |
3980 | pointer to pass to the corresponding functions. | |
3981 | </para> | |
3982 | </section> | |
3983 | ||
3984 | <section id="api-ac97-callbacks"> | |
3985 | <title>Callbacks</title> | |
3986 | <para> | |
3987 | The standard callbacks are <structfield>read</structfield> and | |
3988 | <structfield>write</structfield>. Obviously they | |
3989 | correspond to the functions for read and write accesses to the | |
3990 | hardware low-level codes. | |
3991 | </para> | |
3992 | ||
3993 | <para> | |
3994 | The <structfield>read</structfield> callback returns the | |
3995 | register value specified in the argument. | |
3996 | ||
3997 | <informalexample> | |
3998 | <programlisting> | |
3999 | <![CDATA[ | |
446ab5f5 | 4000 | static unsigned short snd_mychip_ac97_read(struct snd_ac97 *ac97, |
1da177e4 LT |
4001 | unsigned short reg) |
4002 | { | |
446ab5f5 | 4003 | struct mychip *chip = ac97->private_data; |
1da177e4 LT |
4004 | .... |
4005 | return the_register_value; | |
4006 | } | |
4007 | ]]> | |
4008 | </programlisting> | |
4009 | </informalexample> | |
4010 | ||
4011 | Here, the chip can be cast from ac97->private_data. | |
4012 | </para> | |
4013 | ||
4014 | <para> | |
4015 | Meanwhile, the <structfield>write</structfield> callback is | |
4016 | used to set the register value. | |
4017 | ||
4018 | <informalexample> | |
4019 | <programlisting> | |
4020 | <![CDATA[ | |
446ab5f5 | 4021 | static void snd_mychip_ac97_write(struct snd_ac97 *ac97, |
1da177e4 LT |
4022 | unsigned short reg, unsigned short val) |
4023 | ]]> | |
4024 | </programlisting> | |
4025 | </informalexample> | |
4026 | </para> | |
4027 | ||
4028 | <para> | |
4029 | These callbacks are non-atomic like the callbacks of control API. | |
4030 | </para> | |
4031 | ||
4032 | <para> | |
4033 | There are also other callbacks: | |
4034 | <structfield>reset</structfield>, | |
4035 | <structfield>wait</structfield> and | |
4036 | <structfield>init</structfield>. | |
4037 | </para> | |
4038 | ||
4039 | <para> | |
4040 | The <structfield>reset</structfield> callback is used to reset | |
4041 | the codec. If the chip requires a special way of reset, you can | |
4042 | define this callback. | |
4043 | </para> | |
4044 | ||
4045 | <para> | |
4046 | The <structfield>wait</structfield> callback is used for a | |
4047 | certain wait at the standard initialization of the codec. If the | |
4048 | chip requires the extra wait-time, define this callback. | |
4049 | </para> | |
4050 | ||
4051 | <para> | |
4052 | The <structfield>init</structfield> callback is used for | |
4053 | additional initialization of the codec. | |
4054 | </para> | |
4055 | </section> | |
4056 | ||
4057 | <section id="api-ac97-updating-registers"> | |
4058 | <title>Updating Registers in The Driver</title> | |
4059 | <para> | |
4060 | If you need to access to the codec from the driver, you can | |
4061 | call the following functions: | |
4062 | <function>snd_ac97_write()</function>, | |
4063 | <function>snd_ac97_read()</function>, | |
4064 | <function>snd_ac97_update()</function> and | |
4065 | <function>snd_ac97_update_bits()</function>. | |
4066 | </para> | |
4067 | ||
4068 | <para> | |
4069 | Both <function>snd_ac97_write()</function> and | |
4070 | <function>snd_ac97_update()</function> functions are used to | |
4071 | set a value to the given register | |
4072 | (<constant>AC97_XXX</constant>). The difference between them is | |
4073 | that <function>snd_ac97_update()</function> doesn't write a | |
4074 | value if the given value has been already set, while | |
4075 | <function>snd_ac97_write()</function> always rewrites the | |
4076 | value. | |
4077 | ||
4078 | <informalexample> | |
4079 | <programlisting> | |
4080 | <![CDATA[ | |
4081 | snd_ac97_write(ac97, AC97_MASTER, 0x8080); | |
4082 | snd_ac97_update(ac97, AC97_MASTER, 0x8080); | |
4083 | ]]> | |
4084 | </programlisting> | |
4085 | </informalexample> | |
4086 | </para> | |
4087 | ||
4088 | <para> | |
4089 | <function>snd_ac97_read()</function> is used to read the value | |
4090 | of the given register. For example, | |
4091 | ||
4092 | <informalexample> | |
4093 | <programlisting> | |
4094 | <![CDATA[ | |
4095 | value = snd_ac97_read(ac97, AC97_MASTER); | |
4096 | ]]> | |
4097 | </programlisting> | |
4098 | </informalexample> | |
4099 | </para> | |
4100 | ||
4101 | <para> | |
4102 | <function>snd_ac97_update_bits()</function> is used to update | |
4103 | some bits of the given register. | |
4104 | ||
4105 | <informalexample> | |
4106 | <programlisting> | |
4107 | <![CDATA[ | |
4108 | snd_ac97_update_bits(ac97, reg, mask, value); | |
4109 | ]]> | |
4110 | </programlisting> | |
4111 | </informalexample> | |
4112 | </para> | |
4113 | ||
4114 | <para> | |
4115 | Also, there is a function to change the sample rate (of a | |
4116 | certain register such as | |
4117 | <constant>AC97_PCM_FRONT_DAC_RATE</constant>) when VRA or | |
4118 | DRA is supported by the codec: | |
4119 | <function>snd_ac97_set_rate()</function>. | |
4120 | ||
4121 | <informalexample> | |
4122 | <programlisting> | |
4123 | <![CDATA[ | |
4124 | snd_ac97_set_rate(ac97, AC97_PCM_FRONT_DAC_RATE, 44100); | |
4125 | ]]> | |
4126 | </programlisting> | |
4127 | </informalexample> | |
4128 | </para> | |
4129 | ||
4130 | <para> | |
4131 | The following registers are available for setting the rate: | |
4132 | <constant>AC97_PCM_MIC_ADC_RATE</constant>, | |
4133 | <constant>AC97_PCM_FRONT_DAC_RATE</constant>, | |
4134 | <constant>AC97_PCM_LR_ADC_RATE</constant>, | |
4135 | <constant>AC97_SPDIF</constant>. When the | |
4136 | <constant>AC97_SPDIF</constant> is specified, the register is | |
4137 | not really changed but the corresponding IEC958 status bits will | |
4138 | be updated. | |
4139 | </para> | |
4140 | </section> | |
4141 | ||
4142 | <section id="api-ac97-clock-adjustment"> | |
4143 | <title>Clock Adjustment</title> | |
4144 | <para> | |
4145 | On some chip, the clock of the codec isn't 48000 but using a | |
4146 | PCI clock (to save a quartz!). In this case, change the field | |
4147 | bus->clock to the corresponding | |
4148 | value. For example, intel8x0 | |
4149 | and es1968 drivers have the auto-measurement function of the | |
4150 | clock. | |
4151 | </para> | |
4152 | </section> | |
4153 | ||
4154 | <section id="api-ac97-proc-files"> | |
4155 | <title>Proc Files</title> | |
4156 | <para> | |
4157 | The ALSA AC97 interface will create a proc file such as | |
4158 | <filename>/proc/asound/card0/codec97#0/ac97#0-0</filename> and | |
4159 | <filename>ac97#0-0+regs</filename>. You can refer to these files to | |
4160 | see the current status and registers of the codec. | |
4161 | </para> | |
4162 | </section> | |
4163 | ||
4164 | <section id="api-ac97-multiple-codecs"> | |
4165 | <title>Multiple Codecs</title> | |
4166 | <para> | |
4167 | When there are several codecs on the same card, you need to | |
446ab5f5 | 4168 | call <function>snd_ac97_mixer()</function> multiple times with |
1da177e4 LT |
4169 | ac97.num=1 or greater. The <structfield>num</structfield> field |
4170 | specifies the codec | |
4171 | number. | |
4172 | </para> | |
4173 | ||
4174 | <para> | |
4175 | If you have set up multiple codecs, you need to either write | |
4176 | different callbacks for each codec or check | |
4177 | ac97->num in the | |
4178 | callback routines. | |
4179 | </para> | |
4180 | </section> | |
4181 | ||
4182 | </chapter> | |
4183 | ||
4184 | ||
4185 | <!-- ****************************************************** --> | |
4186 | <!-- MIDI (MPU401-UART) Interface --> | |
4187 | <!-- ****************************************************** --> | |
4188 | <chapter id="midi-interface"> | |
4189 | <title>MIDI (MPU401-UART) Interface</title> | |
4190 | ||
4191 | <section id="midi-interface-general"> | |
4192 | <title>General</title> | |
4193 | <para> | |
4194 | Many soundcards have built-in MIDI (MPU401-UART) | |
4195 | interfaces. When the soundcard supports the standard MPU401-UART | |
4196 | interface, most likely you can use the ALSA MPU401-UART API. The | |
4197 | MPU401-UART API is defined in | |
4198 | <filename><sound/mpu401.h></filename>. | |
4199 | </para> | |
4200 | ||
4201 | <para> | |
4202 | Some soundchips have similar but a little bit different | |
4203 | implementation of mpu401 stuff. For example, emu10k1 has its own | |
4204 | mpu401 routines. | |
4205 | </para> | |
4206 | </section> | |
4207 | ||
4208 | <section id="midi-interface-constructor"> | |
4209 | <title>Constructor</title> | |
4210 | <para> | |
4211 | For creating a rawmidi object, call | |
4212 | <function>snd_mpu401_uart_new()</function>. | |
4213 | ||
4214 | <informalexample> | |
4215 | <programlisting> | |
4216 | <![CDATA[ | |
446ab5f5 | 4217 | struct snd_rawmidi *rmidi; |
1da177e4 LT |
4218 | snd_mpu401_uart_new(card, 0, MPU401_HW_MPU401, port, integrated, |
4219 | irq, irq_flags, &rmidi); | |
4220 | ]]> | |
4221 | </programlisting> | |
4222 | </informalexample> | |
4223 | </para> | |
4224 | ||
4225 | <para> | |
4226 | The first argument is the card pointer, and the second is the | |
4227 | index of this component. You can create up to 8 rawmidi | |
4228 | devices. | |
4229 | </para> | |
4230 | ||
4231 | <para> | |
4232 | The third argument is the type of the hardware, | |
4233 | <constant>MPU401_HW_XXX</constant>. If it's not a special one, | |
4234 | you can use <constant>MPU401_HW_MPU401</constant>. | |
4235 | </para> | |
4236 | ||
4237 | <para> | |
4238 | The 4th argument is the i/o port address. Many | |
4239 | backward-compatible MPU401 has an i/o port such as 0x330. Or, it | |
4240 | might be a part of its own PCI i/o region. It depends on the | |
4241 | chip design. | |
4242 | </para> | |
4243 | ||
4244 | <para> | |
4245 | When the i/o port address above is a part of the PCI i/o | |
4246 | region, the MPU401 i/o port might have been already allocated | |
4247 | (reserved) by the driver itself. In such a case, pass non-zero | |
4248 | to the 5th argument | |
4249 | (<parameter>integrated</parameter>). Otherwise, pass 0 to it, | |
4250 | and | |
4251 | the mpu401-uart layer will allocate the i/o ports by itself. | |
4252 | </para> | |
4253 | ||
4254 | <para> | |
4255 | Usually, the port address corresponds to the command port and | |
4256 | port + 1 corresponds to the data port. If not, you may change | |
4257 | the <structfield>cport</structfield> field of | |
446ab5f5 TI |
4258 | struct <structname>snd_mpu401</structname> manually |
4259 | afterward. However, <structname>snd_mpu401</structname> pointer is not | |
1da177e4 LT |
4260 | returned explicitly by |
4261 | <function>snd_mpu401_uart_new()</function>. You need to cast | |
4262 | rmidi->private_data to | |
446ab5f5 | 4263 | <structname>snd_mpu401</structname> explicitly, |
1da177e4 LT |
4264 | |
4265 | <informalexample> | |
4266 | <programlisting> | |
4267 | <![CDATA[ | |
446ab5f5 | 4268 | struct snd_mpu401 *mpu; |
1da177e4 LT |
4269 | mpu = rmidi->private_data; |
4270 | ]]> | |
4271 | </programlisting> | |
4272 | </informalexample> | |
4273 | ||
4274 | and reset the cport as you like: | |
4275 | ||
4276 | <informalexample> | |
4277 | <programlisting> | |
4278 | <![CDATA[ | |
4279 | mpu->cport = my_own_control_port; | |
4280 | ]]> | |
4281 | </programlisting> | |
4282 | </informalexample> | |
4283 | </para> | |
4284 | ||
4285 | <para> | |
4286 | The 6th argument specifies the irq number for UART. If the irq | |
4287 | is already allocated, pass 0 to the 7th argument | |
4288 | (<parameter>irq_flags</parameter>). Otherwise, pass the flags | |
4289 | for irq allocation | |
4290 | (<constant>SA_XXX</constant> bits) to it, and the irq will be | |
4291 | reserved by the mpu401-uart layer. If the card doesn't generates | |
4292 | UART interrupts, pass -1 as the irq number. Then a timer | |
4293 | interrupt will be invoked for polling. | |
4294 | </para> | |
4295 | </section> | |
4296 | ||
4297 | <section id="midi-interface-interrupt-handler"> | |
4298 | <title>Interrupt Handler</title> | |
4299 | <para> | |
4300 | When the interrupt is allocated in | |
4301 | <function>snd_mpu401_uart_new()</function>, the private | |
4302 | interrupt handler is used, hence you don't have to do nothing | |
4303 | else than creating the mpu401 stuff. Otherwise, you have to call | |
4304 | <function>snd_mpu401_uart_interrupt()</function> explicitly when | |
4305 | a UART interrupt is invoked and checked in your own interrupt | |
4306 | handler. | |
4307 | </para> | |
4308 | ||
4309 | <para> | |
4310 | In this case, you need to pass the private_data of the | |
4311 | returned rawmidi object from | |
4312 | <function>snd_mpu401_uart_new()</function> as the second | |
4313 | argument of <function>snd_mpu401_uart_interrupt()</function>. | |
4314 | ||
4315 | <informalexample> | |
4316 | <programlisting> | |
4317 | <![CDATA[ | |
4318 | snd_mpu401_uart_interrupt(irq, rmidi->private_data, regs); | |
4319 | ]]> | |
4320 | </programlisting> | |
4321 | </informalexample> | |
4322 | </para> | |
4323 | </section> | |
4324 | ||
4325 | </chapter> | |
4326 | ||
4327 | ||
4328 | <!-- ****************************************************** --> | |
4329 | <!-- RawMIDI Interface --> | |
4330 | <!-- ****************************************************** --> | |
4331 | <chapter id="rawmidi-interface"> | |
4332 | <title>RawMIDI Interface</title> | |
4333 | ||
4334 | <section id="rawmidi-interface-overview"> | |
4335 | <title>Overview</title> | |
4336 | ||
4337 | <para> | |
4338 | The raw MIDI interface is used for hardware MIDI ports that can | |
4339 | be accessed as a byte stream. It is not used for synthesizer | |
4340 | chips that do not directly understand MIDI. | |
4341 | </para> | |
4342 | ||
4343 | <para> | |
4344 | ALSA handles file and buffer management. All you have to do is | |
4345 | to write some code to move data between the buffer and the | |
4346 | hardware. | |
4347 | </para> | |
4348 | ||
4349 | <para> | |
4350 | The rawmidi API is defined in | |
4351 | <filename><sound/rawmidi.h></filename>. | |
4352 | </para> | |
4353 | </section> | |
4354 | ||
4355 | <section id="rawmidi-interface-constructor"> | |
4356 | <title>Constructor</title> | |
4357 | ||
4358 | <para> | |
4359 | To create a rawmidi device, call the | |
4360 | <function>snd_rawmidi_new</function> function: | |
4361 | <informalexample> | |
4362 | <programlisting> | |
4363 | <![CDATA[ | |
446ab5f5 | 4364 | struct snd_rawmidi *rmidi; |
1da177e4 LT |
4365 | err = snd_rawmidi_new(chip->card, "MyMIDI", 0, outs, ins, &rmidi); |
4366 | if (err < 0) | |
4367 | return err; | |
4368 | rmidi->private_data = chip; | |
4369 | strcpy(rmidi->name, "My MIDI"); | |
4370 | rmidi->info_flags = SNDRV_RAWMIDI_INFO_OUTPUT | | |
4371 | SNDRV_RAWMIDI_INFO_INPUT | | |
4372 | SNDRV_RAWMIDI_INFO_DUPLEX; | |
4373 | ]]> | |
4374 | </programlisting> | |
4375 | </informalexample> | |
4376 | </para> | |
4377 | ||
4378 | <para> | |
4379 | The first argument is the card pointer, the second argument is | |
4380 | the ID string. | |
4381 | </para> | |
4382 | ||
4383 | <para> | |
4384 | The third argument is the index of this component. You can | |
4385 | create up to 8 rawmidi devices. | |
4386 | </para> | |
4387 | ||
4388 | <para> | |
4389 | The fourth and fifth arguments are the number of output and | |
4390 | input substreams, respectively, of this device. (A substream is | |
4391 | the equivalent of a MIDI port.) | |
4392 | </para> | |
4393 | ||
4394 | <para> | |
4395 | Set the <structfield>info_flags</structfield> field to specify | |
4396 | the capabilities of the device. | |
4397 | Set <constant>SNDRV_RAWMIDI_INFO_OUTPUT</constant> if there is | |
4398 | at least one output port, | |
4399 | <constant>SNDRV_RAWMIDI_INFO_INPUT</constant> if there is at | |
4400 | least one input port, | |
4401 | and <constant>SNDRV_RAWMIDI_INFO_DUPLEX</constant> if the device | |
4402 | can handle output and input at the same time. | |
4403 | </para> | |
4404 | ||
4405 | <para> | |
4406 | After the rawmidi device is created, you need to set the | |
4407 | operators (callbacks) for each substream. There are helper | |
4408 | functions to set the operators for all substream of a device: | |
4409 | <informalexample> | |
4410 | <programlisting> | |
4411 | <![CDATA[ | |
4412 | snd_rawmidi_set_ops(rmidi, SNDRV_RAWMIDI_STREAM_OUTPUT, &snd_mymidi_output_ops); | |
4413 | snd_rawmidi_set_ops(rmidi, SNDRV_RAWMIDI_STREAM_INPUT, &snd_mymidi_input_ops); | |
4414 | ]]> | |
4415 | </programlisting> | |
4416 | </informalexample> | |
4417 | </para> | |
4418 | ||
4419 | <para> | |
4420 | The operators are usually defined like this: | |
4421 | <informalexample> | |
4422 | <programlisting> | |
4423 | <![CDATA[ | |
446ab5f5 | 4424 | static struct snd_rawmidi_ops snd_mymidi_output_ops = { |
1da177e4 LT |
4425 | .open = snd_mymidi_output_open, |
4426 | .close = snd_mymidi_output_close, | |
4427 | .trigger = snd_mymidi_output_trigger, | |
4428 | }; | |
4429 | ]]> | |
4430 | </programlisting> | |
4431 | </informalexample> | |
4432 | These callbacks are explained in the <link | |
4433 | linkend="rawmidi-interface-callbacks"><citetitle>Callbacks</citetitle></link> | |
4434 | section. | |
4435 | </para> | |
4436 | ||
4437 | <para> | |
4438 | If there is more than one substream, you should give each one a | |
4439 | unique name: | |
4440 | <informalexample> | |
4441 | <programlisting> | |
4442 | <![CDATA[ | |
4443 | struct list_head *list; | |
446ab5f5 | 4444 | struct snd_rawmidi_substream *substream; |
1da177e4 | 4445 | list_for_each(list, &rmidi->streams[SNDRV_RAWMIDI_STREAM_OUTPUT].substreams) { |
446ab5f5 | 4446 | substream = list_entry(list, struct snd_rawmidi_substream, list); |
1da177e4 LT |
4447 | sprintf(substream->name, "My MIDI Port %d", substream->number + 1); |
4448 | } | |
4449 | /* same for SNDRV_RAWMIDI_STREAM_INPUT */ | |
4450 | ]]> | |
4451 | </programlisting> | |
4452 | </informalexample> | |
4453 | </para> | |
4454 | </section> | |
4455 | ||
4456 | <section id="rawmidi-interface-callbacks"> | |
4457 | <title>Callbacks</title> | |
4458 | ||
4459 | <para> | |
4460 | In all callbacks, the private data that you've set for the | |
4461 | rawmidi device can be accessed as | |
4462 | substream->rmidi->private_data. | |
4463 | <!-- <code> isn't available before DocBook 4.3 --> | |
4464 | </para> | |
4465 | ||
4466 | <para> | |
4467 | If there is more than one port, your callbacks can determine the | |
446ab5f5 | 4468 | port index from the struct snd_rawmidi_substream data passed to each |
1da177e4 LT |
4469 | callback: |
4470 | <informalexample> | |
4471 | <programlisting> | |
4472 | <![CDATA[ | |
446ab5f5 | 4473 | struct snd_rawmidi_substream *substream; |
1da177e4 LT |
4474 | int index = substream->number; |
4475 | ]]> | |
4476 | </programlisting> | |
4477 | </informalexample> | |
4478 | </para> | |
4479 | ||
4480 | <section id="rawmidi-interface-op-open"> | |
4481 | <title><function>open</function> callback</title> | |
4482 | ||
4483 | <informalexample> | |
4484 | <programlisting> | |
4485 | <![CDATA[ | |
446ab5f5 | 4486 | static int snd_xxx_open(struct snd_rawmidi_substream *substream); |
1da177e4 LT |
4487 | ]]> |
4488 | </programlisting> | |
4489 | </informalexample> | |
4490 | ||
4491 | <para> | |
4492 | This is called when a substream is opened. | |
4493 | You can initialize the hardware here, but you should not yet | |
4494 | start transmitting/receiving data. | |
4495 | </para> | |
4496 | </section> | |
4497 | ||
4498 | <section id="rawmidi-interface-op-close"> | |
4499 | <title><function>close</function> callback</title> | |
4500 | ||
4501 | <informalexample> | |
4502 | <programlisting> | |
4503 | <![CDATA[ | |
446ab5f5 | 4504 | static int snd_xxx_close(struct snd_rawmidi_substream *substream); |
1da177e4 LT |
4505 | ]]> |
4506 | </programlisting> | |
4507 | </informalexample> | |
4508 | ||
4509 | <para> | |
4510 | Guess what. | |
4511 | </para> | |
4512 | ||
4513 | <para> | |
4514 | The <function>open</function> and <function>close</function> | |
4515 | callbacks of a rawmidi device are serialized with a mutex, | |
4516 | and can sleep. | |
4517 | </para> | |
4518 | </section> | |
4519 | ||
4520 | <section id="rawmidi-interface-op-trigger-out"> | |
4521 | <title><function>trigger</function> callback for output | |
4522 | substreams</title> | |
4523 | ||
4524 | <informalexample> | |
4525 | <programlisting> | |
4526 | <![CDATA[ | |
446ab5f5 | 4527 | static void snd_xxx_output_trigger(struct snd_rawmidi_substream *substream, int up); |
1da177e4 LT |
4528 | ]]> |
4529 | </programlisting> | |
4530 | </informalexample> | |
4531 | ||
4532 | <para> | |
4533 | This is called with a nonzero <parameter>up</parameter> | |
4534 | parameter when there is some data in the substream buffer that | |
4535 | must be transmitted. | |
4536 | </para> | |
4537 | ||
4538 | <para> | |
4539 | To read data from the buffer, call | |
4540 | <function>snd_rawmidi_transmit_peek</function>. It will | |
4541 | return the number of bytes that have been read; this will be | |
4542 | less than the number of bytes requested when there is no more | |
4543 | data in the buffer. | |
4544 | After the data has been transmitted successfully, call | |
4545 | <function>snd_rawmidi_transmit_ack</function> to remove the | |
4546 | data from the substream buffer: | |
4547 | <informalexample> | |
4548 | <programlisting> | |
4549 | <![CDATA[ | |
4550 | unsigned char data; | |
4551 | while (snd_rawmidi_transmit_peek(substream, &data, 1) == 1) { | |
446ab5f5 | 4552 | if (snd_mychip_try_to_transmit(data)) |
1da177e4 LT |
4553 | snd_rawmidi_transmit_ack(substream, 1); |
4554 | else | |
4555 | break; /* hardware FIFO full */ | |
4556 | } | |
4557 | ]]> | |
4558 | </programlisting> | |
4559 | </informalexample> | |
4560 | </para> | |
4561 | ||
4562 | <para> | |
4563 | If you know beforehand that the hardware will accept data, you | |
4564 | can use the <function>snd_rawmidi_transmit</function> function | |
4565 | which reads some data and removes it from the buffer at once: | |
4566 | <informalexample> | |
4567 | <programlisting> | |
4568 | <![CDATA[ | |
446ab5f5 | 4569 | while (snd_mychip_transmit_possible()) { |
1da177e4 LT |
4570 | unsigned char data; |
4571 | if (snd_rawmidi_transmit(substream, &data, 1) != 1) | |
4572 | break; /* no more data */ | |
446ab5f5 | 4573 | snd_mychip_transmit(data); |
1da177e4 LT |
4574 | } |
4575 | ]]> | |
4576 | </programlisting> | |
4577 | </informalexample> | |
4578 | </para> | |
4579 | ||
4580 | <para> | |
4581 | If you know beforehand how many bytes you can accept, you can | |
4582 | use a buffer size greater than one with the | |
4583 | <function>snd_rawmidi_transmit*</function> functions. | |
4584 | </para> | |
4585 | ||
4586 | <para> | |
4587 | The <function>trigger</function> callback must not sleep. If | |
4588 | the hardware FIFO is full before the substream buffer has been | |
4589 | emptied, you have to continue transmitting data later, either | |
4590 | in an interrupt handler, or with a timer if the hardware | |
4591 | doesn't have a MIDI transmit interrupt. | |
4592 | </para> | |
4593 | ||
4594 | <para> | |
4595 | The <function>trigger</function> callback is called with a | |
4596 | zero <parameter>up</parameter> parameter when the transmission | |
4597 | of data should be aborted. | |
4598 | </para> | |
4599 | </section> | |
4600 | ||
4601 | <section id="rawmidi-interface-op-trigger-in"> | |
4602 | <title><function>trigger</function> callback for input | |
4603 | substreams</title> | |
4604 | ||
4605 | <informalexample> | |
4606 | <programlisting> | |
4607 | <![CDATA[ | |
446ab5f5 | 4608 | static void snd_xxx_input_trigger(struct snd_rawmidi_substream *substream, int up); |
1da177e4 LT |
4609 | ]]> |
4610 | </programlisting> | |
4611 | </informalexample> | |
4612 | ||
4613 | <para> | |
4614 | This is called with a nonzero <parameter>up</parameter> | |
4615 | parameter to enable receiving data, or with a zero | |
4616 | <parameter>up</parameter> parameter do disable receiving data. | |
4617 | </para> | |
4618 | ||
4619 | <para> | |
4620 | The <function>trigger</function> callback must not sleep; the | |
4621 | actual reading of data from the device is usually done in an | |
4622 | interrupt handler. | |
4623 | </para> | |
4624 | ||
4625 | <para> | |
4626 | When data reception is enabled, your interrupt handler should | |
4627 | call <function>snd_rawmidi_receive</function> for all received | |
4628 | data: | |
4629 | <informalexample> | |
4630 | <programlisting> | |
4631 | <![CDATA[ | |
4632 | void snd_mychip_midi_interrupt(...) | |
4633 | { | |
4634 | while (mychip_midi_available()) { | |
4635 | unsigned char data; | |
4636 | data = mychip_midi_read(); | |
4637 | snd_rawmidi_receive(substream, &data, 1); | |
4638 | } | |
4639 | } | |
4640 | ]]> | |
4641 | </programlisting> | |
4642 | </informalexample> | |
4643 | </para> | |
4644 | </section> | |
4645 | ||
4646 | <section id="rawmidi-interface-op-drain"> | |
4647 | <title><function>drain</function> callback</title> | |
4648 | ||
4649 | <informalexample> | |
4650 | <programlisting> | |
4651 | <![CDATA[ | |
446ab5f5 | 4652 | static void snd_xxx_drain(struct snd_rawmidi_substream *substream); |
1da177e4 LT |
4653 | ]]> |
4654 | </programlisting> | |
4655 | </informalexample> | |
4656 | ||
4657 | <para> | |
4658 | This is only used with output substreams. This function should wait | |
4659 | until all data read from the substream buffer has been transmitted. | |
4660 | This ensures that the device can be closed and the driver unloaded | |
4661 | without losing data. | |
4662 | </para> | |
4663 | ||
4664 | <para> | |
4665 | This callback is optional. If you do not set | |
446ab5f5 | 4666 | <structfield>drain</structfield> in the struct snd_rawmidi_ops |
1da177e4 LT |
4667 | structure, ALSA will simply wait for 50 milliseconds |
4668 | instead. | |
4669 | </para> | |
4670 | </section> | |
4671 | </section> | |
4672 | ||
4673 | </chapter> | |
4674 | ||
4675 | ||
4676 | <!-- ****************************************************** --> | |
4677 | <!-- Miscellaneous Devices --> | |
4678 | <!-- ****************************************************** --> | |
4679 | <chapter id="misc-devices"> | |
4680 | <title>Miscellaneous Devices</title> | |
4681 | ||
4682 | <section id="misc-devices-opl3"> | |
4683 | <title>FM OPL3</title> | |
4684 | <para> | |
4685 | The FM OPL3 is still used on many chips (mainly for backward | |
4686 | compatibility). ALSA has a nice OPL3 FM control layer, too. The | |
4687 | OPL3 API is defined in | |
4688 | <filename><sound/opl3.h></filename>. | |
4689 | </para> | |
4690 | ||
4691 | <para> | |
4692 | FM registers can be directly accessed through direct-FM API, | |
4693 | defined in <filename><sound/asound_fm.h></filename>. In | |
4694 | ALSA native mode, FM registers are accessed through | |
4695 | Hardware-Dependant Device direct-FM extension API, whereas in | |
4696 | OSS compatible mode, FM registers can be accessed with OSS | |
4697 | direct-FM compatible API on <filename>/dev/dmfmX</filename> device. | |
4698 | </para> | |
4699 | ||
4700 | <para> | |
4701 | For creating the OPL3 component, you have two functions to | |
4702 | call. The first one is a constructor for <type>opl3_t</type> | |
4703 | instance. | |
4704 | ||
4705 | <informalexample> | |
4706 | <programlisting> | |
4707 | <![CDATA[ | |
446ab5f5 | 4708 | struct snd_opl3 *opl3; |
1da177e4 LT |
4709 | snd_opl3_create(card, lport, rport, OPL3_HW_OPL3_XXX, |
4710 | integrated, &opl3); | |
4711 | ]]> | |
4712 | </programlisting> | |
4713 | </informalexample> | |
4714 | </para> | |
4715 | ||
4716 | <para> | |
4717 | The first argument is the card pointer, the second one is the | |
4718 | left port address, and the third is the right port address. In | |
4719 | most cases, the right port is placed at the left port + 2. | |
4720 | </para> | |
4721 | ||
4722 | <para> | |
4723 | The fourth argument is the hardware type. | |
4724 | </para> | |
4725 | ||
4726 | <para> | |
4727 | When the left and right ports have been already allocated by | |
4728 | the card driver, pass non-zero to the fifth argument | |
4729 | (<parameter>integrated</parameter>). Otherwise, opl3 module will | |
4730 | allocate the specified ports by itself. | |
4731 | </para> | |
4732 | ||
4733 | <para> | |
4734 | When the accessing to the hardware requires special method | |
4735 | instead of the standard I/O access, you can create opl3 instance | |
4736 | separately with <function>snd_opl3_new()</function>. | |
4737 | ||
4738 | <informalexample> | |
4739 | <programlisting> | |
4740 | <![CDATA[ | |
446ab5f5 | 4741 | struct snd_opl3 *opl3; |
1da177e4 LT |
4742 | snd_opl3_new(card, OPL3_HW_OPL3_XXX, &opl3); |
4743 | ]]> | |
4744 | </programlisting> | |
4745 | </informalexample> | |
4746 | </para> | |
4747 | ||
4748 | <para> | |
4749 | Then set <structfield>command</structfield>, | |
4750 | <structfield>private_data</structfield> and | |
4751 | <structfield>private_free</structfield> for the private | |
4752 | access function, the private data and the destructor. | |
4753 | The l_port and r_port are not necessarily set. Only the | |
4754 | command must be set properly. You can retrieve the data | |
4755 | from opl3->private_data field. | |
4756 | </para> | |
4757 | ||
4758 | <para> | |
4759 | After creating the opl3 instance via <function>snd_opl3_new()</function>, | |
4760 | call <function>snd_opl3_init()</function> to initialize the chip to the | |
4761 | proper state. Note that <function>snd_opl3_create()</function> always | |
4762 | calls it internally. | |
4763 | </para> | |
4764 | ||
4765 | <para> | |
4766 | If the opl3 instance is created successfully, then create a | |
4767 | hwdep device for this opl3. | |
4768 | ||
4769 | <informalexample> | |
4770 | <programlisting> | |
4771 | <![CDATA[ | |
446ab5f5 | 4772 | struct snd_hwdep *opl3hwdep; |
1da177e4 LT |
4773 | snd_opl3_hwdep_new(opl3, 0, 1, &opl3hwdep); |
4774 | ]]> | |
4775 | </programlisting> | |
4776 | </informalexample> | |
4777 | </para> | |
4778 | ||
4779 | <para> | |
4780 | The first argument is the <type>opl3_t</type> instance you | |
4781 | created, and the second is the index number, usually 0. | |
4782 | </para> | |
4783 | ||
4784 | <para> | |
4785 | The third argument is the index-offset for the sequencer | |
4786 | client assigned to the OPL3 port. When there is an MPU401-UART, | |
4787 | give 1 for here (UART always takes 0). | |
4788 | </para> | |
4789 | </section> | |
4790 | ||
4791 | <section id="misc-devices-hardware-dependent"> | |
4792 | <title>Hardware-Dependent Devices</title> | |
4793 | <para> | |
4794 | Some chips need the access from the user-space for special | |
4795 | controls or for loading the micro code. In such a case, you can | |
4796 | create a hwdep (hardware-dependent) device. The hwdep API is | |
4797 | defined in <filename><sound/hwdep.h></filename>. You can | |
4798 | find examples in opl3 driver or | |
4799 | <filename>isa/sb/sb16_csp.c</filename>. | |
4800 | </para> | |
4801 | ||
4802 | <para> | |
4803 | Creation of the <type>hwdep</type> instance is done via | |
4804 | <function>snd_hwdep_new()</function>. | |
4805 | ||
4806 | <informalexample> | |
4807 | <programlisting> | |
4808 | <![CDATA[ | |
446ab5f5 | 4809 | struct snd_hwdep *hw; |
1da177e4 LT |
4810 | snd_hwdep_new(card, "My HWDEP", 0, &hw); |
4811 | ]]> | |
4812 | </programlisting> | |
4813 | </informalexample> | |
4814 | ||
4815 | where the third argument is the index number. | |
4816 | </para> | |
4817 | ||
4818 | <para> | |
4819 | You can then pass any pointer value to the | |
4820 | <parameter>private_data</parameter>. | |
4821 | If you assign a private data, you should define the | |
4822 | destructor, too. The destructor function is set to | |
4823 | <structfield>private_free</structfield> field. | |
4824 | ||
4825 | <informalexample> | |
4826 | <programlisting> | |
4827 | <![CDATA[ | |
446ab5f5 | 4828 | struct mydata *p = kmalloc(sizeof(*p), GFP_KERNEL); |
1da177e4 LT |
4829 | hw->private_data = p; |
4830 | hw->private_free = mydata_free; | |
4831 | ]]> | |
4832 | </programlisting> | |
4833 | </informalexample> | |
4834 | ||
4835 | and the implementation of destructor would be: | |
4836 | ||
4837 | <informalexample> | |
4838 | <programlisting> | |
4839 | <![CDATA[ | |
446ab5f5 | 4840 | static void mydata_free(struct snd_hwdep *hw) |
1da177e4 | 4841 | { |
446ab5f5 | 4842 | struct mydata *p = hw->private_data; |
1da177e4 LT |
4843 | kfree(p); |
4844 | } | |
4845 | ]]> | |
4846 | </programlisting> | |
4847 | </informalexample> | |
4848 | </para> | |
4849 | ||
4850 | <para> | |
4851 | The arbitrary file operations can be defined for this | |
4852 | instance. The file operators are defined in | |
4853 | <parameter>ops</parameter> table. For example, assume that | |
4854 | this chip needs an ioctl. | |
4855 | ||
4856 | <informalexample> | |
4857 | <programlisting> | |
4858 | <![CDATA[ | |
4859 | hw->ops.open = mydata_open; | |
4860 | hw->ops.ioctl = mydata_ioctl; | |
4861 | hw->ops.release = mydata_release; | |
4862 | ]]> | |
4863 | </programlisting> | |
4864 | </informalexample> | |
4865 | ||
4866 | And implement the callback functions as you like. | |
4867 | </para> | |
4868 | </section> | |
4869 | ||
4870 | <section id="misc-devices-IEC958"> | |
4871 | <title>IEC958 (S/PDIF)</title> | |
4872 | <para> | |
4873 | Usually the controls for IEC958 devices are implemented via | |
4874 | control interface. There is a macro to compose a name string for | |
4875 | IEC958 controls, <function>SNDRV_CTL_NAME_IEC958()</function> | |
4876 | defined in <filename><include/asound.h></filename>. | |
4877 | </para> | |
4878 | ||
4879 | <para> | |
4880 | There are some standard controls for IEC958 status bits. These | |
4881 | controls use the type <type>SNDRV_CTL_ELEM_TYPE_IEC958</type>, | |
4882 | and the size of element is fixed as 4 bytes array | |
4883 | (value.iec958.status[x]). For <structfield>info</structfield> | |
4884 | callback, you don't specify | |
4885 | the value field for this type (the count field must be set, | |
4886 | though). | |
4887 | </para> | |
4888 | ||
4889 | <para> | |
4890 | <quote>IEC958 Playback Con Mask</quote> is used to return the | |
4891 | bit-mask for the IEC958 status bits of consumer mode. Similarly, | |
4892 | <quote>IEC958 Playback Pro Mask</quote> returns the bitmask for | |
4893 | professional mode. They are read-only controls, and are defined | |
4894 | as MIXER controls (iface = | |
4895 | <constant>SNDRV_CTL_ELEM_IFACE_MIXER</constant>). | |
4896 | </para> | |
4897 | ||
4898 | <para> | |
4899 | Meanwhile, <quote>IEC958 Playback Default</quote> control is | |
4900 | defined for getting and setting the current default IEC958 | |
4901 | bits. Note that this one is usually defined as a PCM control | |
4902 | (iface = <constant>SNDRV_CTL_ELEM_IFACE_PCM</constant>), | |
4903 | although in some places it's defined as a MIXER control. | |
4904 | </para> | |
4905 | ||
4906 | <para> | |
4907 | In addition, you can define the control switches to | |
4908 | enable/disable or to set the raw bit mode. The implementation | |
4909 | will depend on the chip, but the control should be named as | |
4910 | <quote>IEC958 xxx</quote>, preferably using | |
4911 | <function>SNDRV_CTL_NAME_IEC958()</function> macro. | |
4912 | </para> | |
4913 | ||
4914 | <para> | |
4915 | You can find several cases, for example, | |
4916 | <filename>pci/emu10k1</filename>, | |
4917 | <filename>pci/ice1712</filename>, or | |
4918 | <filename>pci/cmipci.c</filename>. | |
4919 | </para> | |
4920 | </section> | |
4921 | ||
4922 | </chapter> | |
4923 | ||
4924 | ||
4925 | <!-- ****************************************************** --> | |
4926 | <!-- Buffer and Memory Management --> | |
4927 | <!-- ****************************************************** --> | |
4928 | <chapter id="buffer-and-memory"> | |
4929 | <title>Buffer and Memory Management</title> | |
4930 | ||
4931 | <section id="buffer-and-memory-buffer-types"> | |
4932 | <title>Buffer Types</title> | |
4933 | <para> | |
4934 | ALSA provides several different buffer allocation functions | |
4935 | depending on the bus and the architecture. All these have a | |
4936 | consistent API. The allocation of physically-contiguous pages is | |
4937 | done via | |
4938 | <function>snd_malloc_xxx_pages()</function> function, where xxx | |
4939 | is the bus type. | |
4940 | </para> | |
4941 | ||
4942 | <para> | |
4943 | The allocation of pages with fallback is | |
4944 | <function>snd_malloc_xxx_pages_fallback()</function>. This | |
4945 | function tries to allocate the specified pages but if the pages | |
4946 | are not available, it tries to reduce the page sizes until the | |
4947 | enough space is found. | |
4948 | </para> | |
4949 | ||
4950 | <para> | |
4951 | For releasing the space, call | |
4952 | <function>snd_free_xxx_pages()</function> function. | |
4953 | </para> | |
4954 | ||
4955 | <para> | |
4956 | Usually, ALSA drivers try to allocate and reserve | |
4957 | a large contiguous physical space | |
4958 | at the time the module is loaded for the later use. | |
4959 | This is called <quote>pre-allocation</quote>. | |
4960 | As already written, you can call the following function at the | |
4961 | construction of pcm instance (in the case of PCI bus). | |
4962 | ||
4963 | <informalexample> | |
4964 | <programlisting> | |
4965 | <![CDATA[ | |
4966 | snd_pcm_lib_preallocate_pages_for_all(pcm, SNDRV_DMA_TYPE_DEV, | |
4967 | snd_dma_pci_data(pci), size, max); | |
4968 | ]]> | |
4969 | </programlisting> | |
4970 | </informalexample> | |
4971 | ||
4972 | where <parameter>size</parameter> is the byte size to be | |
4973 | pre-allocated and the <parameter>max</parameter> is the maximal | |
4974 | size to be changed via <filename>prealloc</filename> proc file. | |
4975 | The allocator will try to get as large area as possible | |
4976 | within the given size. | |
4977 | </para> | |
4978 | ||
4979 | <para> | |
4980 | The second argument (type) and the third argument (device pointer) | |
4981 | are dependent on the bus. | |
4982 | In the case of ISA bus, pass <function>snd_dma_isa_data()</function> | |
4983 | as the third argument with <constant>SNDRV_DMA_TYPE_DEV</constant> type. | |
4984 | For the continuous buffer unrelated to the bus can be pre-allocated | |
4985 | with <constant>SNDRV_DMA_TYPE_CONTINUOUS</constant> type and the | |
4986 | <function>snd_dma_continuous_data(GFP_KERNEL)</function> device pointer, | |
4987 | whereh <constant>GFP_KERNEL</constant> is the kernel allocation flag to | |
4988 | use. For the SBUS, <constant>SNDRV_DMA_TYPE_SBUS</constant> and | |
4989 | <function>snd_dma_sbus_data(sbus_dev)</function> are used instead. | |
4990 | For the PCI scatter-gather buffers, use | |
4991 | <constant>SNDRV_DMA_TYPE_DEV_SG</constant> with | |
4992 | <function>snd_dma_pci_data(pci)</function> | |
4993 | (see the section | |
4994 | <link linkend="buffer-and-memory-non-contiguous"><citetitle>Non-Contiguous Buffers | |
4995 | </citetitle></link>). | |
4996 | </para> | |
4997 | ||
4998 | <para> | |
4999 | Once when the buffer is pre-allocated, you can use the | |
5000 | allocator in the <structfield>hw_params</structfield> callback | |
5001 | ||
5002 | <informalexample> | |
5003 | <programlisting> | |
5004 | <![CDATA[ | |
5005 | snd_pcm_lib_malloc_pages(substream, size); | |
5006 | ]]> | |
5007 | </programlisting> | |
5008 | </informalexample> | |
5009 | ||
5010 | Note that you have to pre-allocate to use this function. | |
5011 | </para> | |
5012 | </section> | |
5013 | ||
5014 | <section id="buffer-and-memory-external-hardware"> | |
5015 | <title>External Hardware Buffers</title> | |
5016 | <para> | |
5017 | Some chips have their own hardware buffers and the DMA | |
5018 | transfer from the host memory is not available. In such a case, | |
5019 | you need to either 1) copy/set the audio data directly to the | |
5020 | external hardware buffer, or 2) make an intermediate buffer and | |
5021 | copy/set the data from it to the external hardware buffer in | |
5022 | interrupts (or in tasklets, preferably). | |
5023 | </para> | |
5024 | ||
5025 | <para> | |
5026 | The first case works fine if the external hardware buffer is enough | |
5027 | large. This method doesn't need any extra buffers and thus is | |
5028 | more effective. You need to define the | |
5029 | <structfield>copy</structfield> and | |
5030 | <structfield>silence</structfield> callbacks for | |
5031 | the data transfer. However, there is a drawback: it cannot | |
5032 | be mmapped. The examples are GUS's GF1 PCM or emu8000's | |
5033 | wavetable PCM. | |
5034 | </para> | |
5035 | ||
5036 | <para> | |
5037 | The second case allows the mmap of the buffer, although you have | |
5038 | to handle an interrupt or a tasklet for transferring the data | |
5039 | from the intermediate buffer to the hardware buffer. You can find an | |
5040 | example in vxpocket driver. | |
5041 | </para> | |
5042 | ||
5043 | <para> | |
5044 | Another case is that the chip uses a PCI memory-map | |
5045 | region for the buffer instead of the host memory. In this case, | |
5046 | mmap is available only on certain architectures like intel. In | |
5047 | non-mmap mode, the data cannot be transferred as the normal | |
5048 | way. Thus you need to define <structfield>copy</structfield> and | |
5049 | <structfield>silence</structfield> callbacks as well | |
5050 | as in the cases above. The examples are found in | |
5051 | <filename>rme32.c</filename> and <filename>rme96.c</filename>. | |
5052 | </para> | |
5053 | ||
5054 | <para> | |
5055 | The implementation of <structfield>copy</structfield> and | |
5056 | <structfield>silence</structfield> callbacks depends upon | |
5057 | whether the hardware supports interleaved or non-interleaved | |
5058 | samples. The <structfield>copy</structfield> callback is | |
5059 | defined like below, a bit | |
5060 | differently depending whether the direction is playback or | |
5061 | capture: | |
5062 | ||
5063 | <informalexample> | |
5064 | <programlisting> | |
5065 | <![CDATA[ | |
446ab5f5 | 5066 | static int playback_copy(struct snd_pcm_substream *substream, int channel, |
1da177e4 | 5067 | snd_pcm_uframes_t pos, void *src, snd_pcm_uframes_t count); |
446ab5f5 | 5068 | static int capture_copy(struct snd_pcm_substream *substream, int channel, |
1da177e4 LT |
5069 | snd_pcm_uframes_t pos, void *dst, snd_pcm_uframes_t count); |
5070 | ]]> | |
5071 | </programlisting> | |
5072 | </informalexample> | |
5073 | </para> | |
5074 | ||
5075 | <para> | |
5076 | In the case of interleaved samples, the second argument | |
5077 | (<parameter>channel</parameter>) is not used. The third argument | |
5078 | (<parameter>pos</parameter>) points the | |
5079 | current position offset in frames. | |
5080 | </para> | |
5081 | ||
5082 | <para> | |
5083 | The meaning of the fourth argument is different between | |
5084 | playback and capture. For playback, it holds the source data | |
5085 | pointer, and for capture, it's the destination data pointer. | |
5086 | </para> | |
5087 | ||
5088 | <para> | |
5089 | The last argument is the number of frames to be copied. | |
5090 | </para> | |
5091 | ||
5092 | <para> | |
5093 | What you have to do in this callback is again different | |
5094 | between playback and capture directions. In the case of | |
5095 | playback, you do: copy the given amount of data | |
5096 | (<parameter>count</parameter>) at the specified pointer | |
5097 | (<parameter>src</parameter>) to the specified offset | |
5098 | (<parameter>pos</parameter>) on the hardware buffer. When | |
5099 | coded like memcpy-like way, the copy would be like: | |
5100 | ||
5101 | <informalexample> | |
5102 | <programlisting> | |
5103 | <![CDATA[ | |
5104 | my_memcpy(my_buffer + frames_to_bytes(runtime, pos), src, | |
5105 | frames_to_bytes(runtime, count)); | |
5106 | ]]> | |
5107 | </programlisting> | |
5108 | </informalexample> | |
5109 | </para> | |
5110 | ||
5111 | <para> | |
5112 | For the capture direction, you do: copy the given amount of | |
5113 | data (<parameter>count</parameter>) at the specified offset | |
5114 | (<parameter>pos</parameter>) on the hardware buffer to the | |
5115 | specified pointer (<parameter>dst</parameter>). | |
5116 | ||
5117 | <informalexample> | |
5118 | <programlisting> | |
5119 | <![CDATA[ | |
5120 | my_memcpy(dst, my_buffer + frames_to_bytes(runtime, pos), | |
5121 | frames_to_bytes(runtime, count)); | |
5122 | ]]> | |
5123 | </programlisting> | |
5124 | </informalexample> | |
5125 | ||
5126 | Note that both of the position and the data amount are given | |
5127 | in frames. | |
5128 | </para> | |
5129 | ||
5130 | <para> | |
5131 | In the case of non-interleaved samples, the implementation | |
5132 | will be a bit more complicated. | |
5133 | </para> | |
5134 | ||
5135 | <para> | |
5136 | You need to check the channel argument, and if it's -1, copy | |
5137 | the whole channels. Otherwise, you have to copy only the | |
5138 | specified channel. Please check | |
5139 | <filename>isa/gus/gus_pcm.c</filename> as an example. | |
5140 | </para> | |
5141 | ||
5142 | <para> | |
5143 | The <structfield>silence</structfield> callback is also | |
5144 | implemented in a similar way. | |
5145 | ||
5146 | <informalexample> | |
5147 | <programlisting> | |
5148 | <![CDATA[ | |
446ab5f5 | 5149 | static int silence(struct snd_pcm_substream *substream, int channel, |
1da177e4 LT |
5150 | snd_pcm_uframes_t pos, snd_pcm_uframes_t count); |
5151 | ]]> | |
5152 | </programlisting> | |
5153 | </informalexample> | |
5154 | </para> | |
5155 | ||
5156 | <para> | |
5157 | The meanings of arguments are identical with the | |
5158 | <structfield>copy</structfield> | |
5159 | callback, although there is no <parameter>src/dst</parameter> | |
5160 | argument. In the case of interleaved samples, the channel | |
5161 | argument has no meaning, as well as on | |
5162 | <structfield>copy</structfield> callback. | |
5163 | </para> | |
5164 | ||
5165 | <para> | |
5166 | The role of <structfield>silence</structfield> callback is to | |
5167 | set the given amount | |
5168 | (<parameter>count</parameter>) of silence data at the | |
5169 | specified offset (<parameter>pos</parameter>) on the hardware | |
5170 | buffer. Suppose that the data format is signed (that is, the | |
5171 | silent-data is 0), and the implementation using a memset-like | |
5172 | function would be like: | |
5173 | ||
5174 | <informalexample> | |
5175 | <programlisting> | |
5176 | <![CDATA[ | |
5177 | my_memcpy(my_buffer + frames_to_bytes(runtime, pos), 0, | |
5178 | frames_to_bytes(runtime, count)); | |
5179 | ]]> | |
5180 | </programlisting> | |
5181 | </informalexample> | |
5182 | </para> | |
5183 | ||
5184 | <para> | |
5185 | In the case of non-interleaved samples, again, the | |
5186 | implementation becomes a bit more complicated. See, for example, | |
5187 | <filename>isa/gus/gus_pcm.c</filename>. | |
5188 | </para> | |
5189 | </section> | |
5190 | ||
5191 | <section id="buffer-and-memory-non-contiguous"> | |
5192 | <title>Non-Contiguous Buffers</title> | |
5193 | <para> | |
5194 | If your hardware supports the page table like emu10k1 or the | |
5195 | buffer descriptors like via82xx, you can use the scatter-gather | |
5196 | (SG) DMA. ALSA provides an interface for handling SG-buffers. | |
5197 | The API is provided in <filename><sound/pcm.h></filename>. | |
5198 | </para> | |
5199 | ||
5200 | <para> | |
5201 | For creating the SG-buffer handler, call | |
5202 | <function>snd_pcm_lib_preallocate_pages()</function> or | |
5203 | <function>snd_pcm_lib_preallocate_pages_for_all()</function> | |
5204 | with <constant>SNDRV_DMA_TYPE_DEV_SG</constant> | |
5205 | in the PCM constructor like other PCI pre-allocator. | |
5206 | You need to pass the <function>snd_dma_pci_data(pci)</function>, | |
5207 | where pci is the struct <structname>pci_dev</structname> pointer | |
5208 | of the chip as well. | |
5209 | The <type>snd_sg_buf_t</type> instance is created as | |
5210 | substream->dma_private. You can cast | |
5211 | the pointer like: | |
5212 | ||
5213 | <informalexample> | |
5214 | <programlisting> | |
5215 | <![CDATA[ | |
446ab5f5 | 5216 | struct snd_sg_buf *sgbuf = (struct snd_sg_buf_t*)substream->dma_private; |
1da177e4 LT |
5217 | ]]> |
5218 | </programlisting> | |
5219 | </informalexample> | |
5220 | </para> | |
5221 | ||
5222 | <para> | |
5223 | Then call <function>snd_pcm_lib_malloc_pages()</function> | |
5224 | in <structfield>hw_params</structfield> callback | |
5225 | as well as in the case of normal PCI buffer. | |
5226 | The SG-buffer handler will allocate the non-contiguous kernel | |
5227 | pages of the given size and map them onto the virtually contiguous | |
5228 | memory. The virtual pointer is addressed in runtime->dma_area. | |
5229 | The physical address (runtime->dma_addr) is set to zero, | |
5230 | because the buffer is physically non-contigous. | |
5231 | The physical address table is set up in sgbuf->table. | |
5232 | You can get the physical address at a certain offset via | |
5233 | <function>snd_pcm_sgbuf_get_addr()</function>. | |
5234 | </para> | |
5235 | ||
5236 | <para> | |
5237 | When a SG-handler is used, you need to set | |
5238 | <function>snd_pcm_sgbuf_ops_page</function> as | |
5239 | the <structfield>page</structfield> callback. | |
5240 | (See <link linkend="pcm-interface-operators-page-callback"> | |
5241 | <citetitle>page callback section</citetitle></link>.) | |
5242 | </para> | |
5243 | ||
5244 | <para> | |
5245 | For releasing the data, call | |
5246 | <function>snd_pcm_lib_free_pages()</function> in the | |
5247 | <structfield>hw_free</structfield> callback as usual. | |
5248 | </para> | |
5249 | </section> | |
5250 | ||
5251 | <section id="buffer-and-memory-vmalloced"> | |
5252 | <title>Vmalloc'ed Buffers</title> | |
5253 | <para> | |
5254 | It's possible to use a buffer allocated via | |
5255 | <function>vmalloc</function>, for example, for an intermediate | |
5256 | buffer. Since the allocated pages are not contiguous, you need | |
5257 | to set the <structfield>page</structfield> callback to obtain | |
5258 | the physical address at every offset. | |
5259 | </para> | |
5260 | ||
5261 | <para> | |
5262 | The implementation of <structfield>page</structfield> callback | |
5263 | would be like this: | |
5264 | ||
5265 | <informalexample> | |
5266 | <programlisting> | |
5267 | <![CDATA[ | |
5268 | #include <linux/vmalloc.h> | |
5269 | ||
5270 | /* get the physical page pointer on the given offset */ | |
446ab5f5 | 5271 | static struct page *mychip_page(struct snd_pcm_substream *substream, |
1da177e4 LT |
5272 | unsigned long offset) |
5273 | { | |
5274 | void *pageptr = substream->runtime->dma_area + offset; | |
5275 | return vmalloc_to_page(pageptr); | |
5276 | } | |
5277 | ]]> | |
5278 | </programlisting> | |
5279 | </informalexample> | |
5280 | </para> | |
5281 | </section> | |
5282 | ||
5283 | </chapter> | |
5284 | ||
5285 | ||
5286 | <!-- ****************************************************** --> | |
5287 | <!-- Proc Interface --> | |
5288 | <!-- ****************************************************** --> | |
5289 | <chapter id="proc-interface"> | |
5290 | <title>Proc Interface</title> | |
5291 | <para> | |
5292 | ALSA provides an easy interface for procfs. The proc files are | |
5293 | very useful for debugging. I recommend you set up proc files if | |
5294 | you write a driver and want to get a running status or register | |
5295 | dumps. The API is found in | |
5296 | <filename><sound/info.h></filename>. | |
5297 | </para> | |
5298 | ||
5299 | <para> | |
5300 | For creating a proc file, call | |
5301 | <function>snd_card_proc_new()</function>. | |
5302 | ||
5303 | <informalexample> | |
5304 | <programlisting> | |
5305 | <![CDATA[ | |
446ab5f5 | 5306 | struct snd_info_entry *entry; |
1da177e4 LT |
5307 | int err = snd_card_proc_new(card, "my-file", &entry); |
5308 | ]]> | |
5309 | </programlisting> | |
5310 | </informalexample> | |
5311 | ||
5312 | where the second argument specifies the proc-file name to be | |
5313 | created. The above example will create a file | |
5314 | <filename>my-file</filename> under the card directory, | |
5315 | e.g. <filename>/proc/asound/card0/my-file</filename>. | |
5316 | </para> | |
5317 | ||
5318 | <para> | |
5319 | Like other components, the proc entry created via | |
5320 | <function>snd_card_proc_new()</function> will be registered and | |
5321 | released automatically in the card registration and release | |
5322 | functions. | |
5323 | </para> | |
5324 | ||
5325 | <para> | |
5326 | When the creation is successful, the function stores a new | |
5327 | instance at the pointer given in the third argument. | |
5328 | It is initialized as a text proc file for read only. For using | |
5329 | this proc file as a read-only text file as it is, set the read | |
5330 | callback with a private data via | |
5331 | <function>snd_info_set_text_ops()</function>. | |
5332 | ||
5333 | <informalexample> | |
5334 | <programlisting> | |
5335 | <![CDATA[ | |
5336 | snd_info_set_text_ops(entry, chip, read_size, my_proc_read); | |
5337 | ]]> | |
5338 | </programlisting> | |
5339 | </informalexample> | |
5340 | ||
5341 | where the second argument (<parameter>chip</parameter>) is the | |
5342 | private data to be used in the callbacks. The third parameter | |
5343 | specifies the read buffer size and the fourth | |
5344 | (<parameter>my_proc_read</parameter>) is the callback function, which | |
5345 | is defined like | |
5346 | ||
5347 | <informalexample> | |
5348 | <programlisting> | |
5349 | <![CDATA[ | |
446ab5f5 TI |
5350 | static void my_proc_read(struct snd_info_entry *entry, |
5351 | struct snd_info_buffer *buffer); | |
1da177e4 LT |
5352 | ]]> |
5353 | </programlisting> | |
5354 | </informalexample> | |
5355 | ||
5356 | </para> | |
5357 | ||
5358 | <para> | |
5359 | In the read callback, use <function>snd_iprintf()</function> for | |
5360 | output strings, which works just like normal | |
5361 | <function>printf()</function>. For example, | |
5362 | ||
5363 | <informalexample> | |
5364 | <programlisting> | |
5365 | <![CDATA[ | |
446ab5f5 TI |
5366 | static void my_proc_read(struct snd_info_entry *entry, |
5367 | struct snd_info_buffer *buffer) | |
1da177e4 | 5368 | { |
446ab5f5 | 5369 | struct my_chip *chip = entry->private_data; |
1da177e4 LT |
5370 | |
5371 | snd_iprintf(buffer, "This is my chip!\n"); | |
5372 | snd_iprintf(buffer, "Port = %ld\n", chip->port); | |
5373 | } | |
5374 | ]]> | |
5375 | </programlisting> | |
5376 | </informalexample> | |
5377 | </para> | |
5378 | ||
5379 | <para> | |
5380 | The file permission can be changed afterwards. As default, it's | |
5381 | set as read only for all users. If you want to add the write | |
5382 | permission to the user (root as default), set like below: | |
5383 | ||
5384 | <informalexample> | |
5385 | <programlisting> | |
5386 | <![CDATA[ | |
5387 | entry->mode = S_IFREG | S_IRUGO | S_IWUSR; | |
5388 | ]]> | |
5389 | </programlisting> | |
5390 | </informalexample> | |
5391 | ||
5392 | and set the write buffer size and the callback | |
5393 | ||
5394 | <informalexample> | |
5395 | <programlisting> | |
5396 | <![CDATA[ | |
5397 | entry->c.text.write_size = 256; | |
5398 | entry->c.text.write = my_proc_write; | |
5399 | ]]> | |
5400 | </programlisting> | |
5401 | </informalexample> | |
5402 | </para> | |
5403 | ||
5404 | <para> | |
5405 | The buffer size for read is set to 1024 implicitly by | |
5406 | <function>snd_info_set_text_ops()</function>. It should suffice | |
5407 | in most cases (the size will be aligned to | |
5408 | <constant>PAGE_SIZE</constant> anyway), but if you need to handle | |
5409 | very large text files, you can set it explicitly, too. | |
5410 | ||
5411 | <informalexample> | |
5412 | <programlisting> | |
5413 | <![CDATA[ | |
5414 | entry->c.text.read_size = 65536; | |
5415 | ]]> | |
5416 | </programlisting> | |
5417 | </informalexample> | |
5418 | </para> | |
5419 | ||
5420 | <para> | |
5421 | For the write callback, you can use | |
5422 | <function>snd_info_get_line()</function> to get a text line, and | |
5423 | <function>snd_info_get_str()</function> to retrieve a string from | |
5424 | the line. Some examples are found in | |
5425 | <filename>core/oss/mixer_oss.c</filename>, core/oss/and | |
5426 | <filename>pcm_oss.c</filename>. | |
5427 | </para> | |
5428 | ||
5429 | <para> | |
5430 | For a raw-data proc-file, set the attributes like the following: | |
5431 | ||
5432 | <informalexample> | |
5433 | <programlisting> | |
5434 | <![CDATA[ | |
5435 | static struct snd_info_entry_ops my_file_io_ops = { | |
5436 | .read = my_file_io_read, | |
5437 | }; | |
5438 | ||
5439 | entry->content = SNDRV_INFO_CONTENT_DATA; | |
5440 | entry->private_data = chip; | |
5441 | entry->c.ops = &my_file_io_ops; | |
5442 | entry->size = 4096; | |
5443 | entry->mode = S_IFREG | S_IRUGO; | |
5444 | ]]> | |
5445 | </programlisting> | |
5446 | </informalexample> | |
5447 | </para> | |
5448 | ||
5449 | <para> | |
5450 | The callback is much more complicated than the text-file | |
5451 | version. You need to use a low-level i/o functions such as | |
5452 | <function>copy_from/to_user()</function> to transfer the | |
5453 | data. | |
5454 | ||
5455 | <informalexample> | |
5456 | <programlisting> | |
5457 | <![CDATA[ | |
446ab5f5 | 5458 | static long my_file_io_read(struct snd_info_entry *entry, |
1da177e4 LT |
5459 | void *file_private_data, |
5460 | struct file *file, | |
5461 | char *buf, | |
5462 | unsigned long count, | |
5463 | unsigned long pos) | |
5464 | { | |
5465 | long size = count; | |
5466 | if (pos + size > local_max_size) | |
5467 | size = local_max_size - pos; | |
5468 | if (copy_to_user(buf, local_data + pos, size)) | |
5469 | return -EFAULT; | |
5470 | return size; | |
5471 | } | |
5472 | ]]> | |
5473 | </programlisting> | |
5474 | </informalexample> | |
5475 | </para> | |
5476 | ||
5477 | </chapter> | |
5478 | ||
5479 | ||
5480 | <!-- ****************************************************** --> | |
5481 | <!-- Power Management --> | |
5482 | <!-- ****************************************************** --> | |
5483 | <chapter id="power-management"> | |
5484 | <title>Power Management</title> | |
5485 | <para> | |
5486 | If the chip is supposed to work with with suspend/resume | |
5487 | functions, you need to add the power-management codes to the | |
5488 | driver. The additional codes for the power-management should be | |
5489 | <function>ifdef</function>'ed with | |
5490 | <constant>CONFIG_PM</constant>. | |
5491 | </para> | |
5492 | ||
5fe76e4d TI |
5493 | <para> |
5494 | If the driver supports the suspend/resume | |
5495 | <emphasis>fully</emphasis>, that is, the device can be | |
5496 | properly resumed to the status at the suspend is called, | |
5497 | you can set <constant>SNDRV_PCM_INFO_RESUME</constant> flag | |
5498 | to pcm info field. Usually, this is possible when the | |
5499 | registers of ths chip can be safely saved and restored to the | |
5500 | RAM. If this is set, the trigger callback is called with | |
5501 | <constant>SNDRV_PCM_TRIGGER_RESUME</constant> after resume | |
5502 | callback is finished. | |
5503 | </para> | |
5504 | ||
5505 | <para> | |
5506 | Even if the driver doesn't support PM fully but only the | |
5507 | partial suspend/resume is possible, it's still worthy to | |
5508 | implement suspend/resume callbacks. In such a case, applications | |
5509 | would reset the status by calling | |
5510 | <function>snd_pcm_prepare()</function> and restart the stream | |
5511 | appropriately. Hence, you can define suspend/resume callbacks | |
5512 | below but don't set <constant>SNDRV_PCM_INFO_RESUME</constant> | |
5513 | info flag to the PCM. | |
5514 | </para> | |
5515 | ||
5516 | <para> | |
5517 | Note that the trigger with SUSPEND can be always called when | |
5518 | <function>snd_pcm_suspend_all</function> is called, | |
5519 | regardless of <constant>SNDRV_PCM_INFO_RESUME</constant> flag. | |
5520 | The <constant>RESUME</constant> flag affects only the behavior | |
5521 | of <function>snd_pcm_resume()</function>. | |
5522 | (Thus, in theory, | |
5523 | <constant>SNDRV_PCM_TRIGGER_RESUME</constant> isn't needed | |
5524 | to be handled in the trigger callback when no | |
5525 | <constant>SNDRV_PCM_INFO_RESUME</constant> flag is set. But, | |
5526 | it's better to keep it for compatibility reason.) | |
5527 | </para> | |
1da177e4 | 5528 | <para> |
5fe76e4d TI |
5529 | In the earlier version of ALSA drivers, a common |
5530 | power-management layer was provided, but it has been removed. | |
5531 | The driver needs to define the suspend/resume hooks according to | |
5532 | the bus the device is assigned. In the case of PCI driver, the | |
5533 | callbacks look like below: | |
1da177e4 LT |
5534 | |
5535 | <informalexample> | |
5536 | <programlisting> | |
5537 | <![CDATA[ | |
5538 | #ifdef CONFIG_PM | |
5fe76e4d | 5539 | static int snd_my_suspend(struct pci_dev *pci, pm_message_t state) |
1da177e4 | 5540 | { |
5fe76e4d | 5541 | .... /* do things for suspsend */ |
1da177e4 LT |
5542 | return 0; |
5543 | } | |
5fe76e4d | 5544 | static int snd_my_resume(struct pci_dev *pci) |
1da177e4 | 5545 | { |
5fe76e4d | 5546 | .... /* do things for suspsend */ |
1da177e4 LT |
5547 | return 0; |
5548 | } | |
5549 | #endif | |
5550 | ]]> | |
5551 | </programlisting> | |
5552 | </informalexample> | |
5553 | </para> | |
5554 | ||
5555 | <para> | |
5556 | The scheme of the real suspend job is as following. | |
5557 | ||
5558 | <orderedlist> | |
5fe76e4d TI |
5559 | <listitem><para>Retrieve the card and the chip data.</para></listitem> |
5560 | <listitem><para>Call <function>snd_power_change_state()</function> with | |
5561 | <constant>SNDRV_CTL_POWER_D3hot</constant> to change the | |
5562 | power status.</para></listitem> | |
1da177e4 | 5563 | <listitem><para>Call <function>snd_pcm_suspend_all()</function> to suspend the running PCM streams.</para></listitem> |
5fe76e4d TI |
5564 | <listitem><para>If AC97 codecs are used, call |
5565 | <function>snd_ac97_resume()</function> for each codec.</para></listitem> | |
1da177e4 LT |
5566 | <listitem><para>Save the register values if necessary.</para></listitem> |
5567 | <listitem><para>Stop the hardware if necessary.</para></listitem> | |
5fe76e4d TI |
5568 | <listitem><para>Disable the PCI device by calling |
5569 | <function>pci_disable_device()</function>. Then, call | |
5570 | <function>pci_save_state()</function> at last.</para></listitem> | |
1da177e4 LT |
5571 | </orderedlist> |
5572 | </para> | |
5573 | ||
5574 | <para> | |
5575 | A typical code would be like: | |
5576 | ||
5577 | <informalexample> | |
5578 | <programlisting> | |
5579 | <![CDATA[ | |
32357988 | 5580 | static int mychip_suspend(struct pci_dev *pci, pm_message_t state) |
1da177e4 LT |
5581 | { |
5582 | /* (1) */ | |
5fe76e4d TI |
5583 | struct snd_card *card = pci_get_drvdata(pci); |
5584 | struct mychip *chip = card->private_data; | |
1da177e4 | 5585 | /* (2) */ |
5fe76e4d | 5586 | snd_power_change_state(card, SNDRV_CTL_POWER_D3hot); |
1da177e4 | 5587 | /* (3) */ |
5fe76e4d | 5588 | snd_pcm_suspend_all(chip->pcm); |
1da177e4 | 5589 | /* (4) */ |
5fe76e4d | 5590 | snd_ac97_suspend(chip->ac97); |
1da177e4 | 5591 | /* (5) */ |
5fe76e4d TI |
5592 | snd_mychip_save_registers(chip); |
5593 | /* (6) */ | |
5594 | snd_mychip_stop_hardware(chip); | |
5595 | /* (7) */ | |
5596 | pci_disable_device(pci); | |
5597 | pci_save_state(pci); | |
1da177e4 LT |
5598 | return 0; |
5599 | } | |
5600 | ]]> | |
5601 | </programlisting> | |
5602 | </informalexample> | |
5603 | </para> | |
5604 | ||
5605 | <para> | |
5606 | The scheme of the real resume job is as following. | |
5607 | ||
5608 | <orderedlist> | |
5fe76e4d TI |
5609 | <listitem><para>Retrieve the card and the chip data.</para></listitem> |
5610 | <listitem><para>Set up PCI. First, call <function>pci_restore_state()</function>. | |
5611 | Then enable the pci device again by calling <function>pci_enable_device()</function>. | |
5612 | Call <function>pci_set_master()</function> if necessary, too.</para></listitem> | |
1da177e4 LT |
5613 | <listitem><para>Re-initialize the chip.</para></listitem> |
5614 | <listitem><para>Restore the saved registers if necessary.</para></listitem> | |
5615 | <listitem><para>Resume the mixer, e.g. calling | |
5616 | <function>snd_ac97_resume()</function>.</para></listitem> | |
5617 | <listitem><para>Restart the hardware (if any).</para></listitem> | |
5fe76e4d TI |
5618 | <listitem><para>Call <function>snd_power_change_state()</function> with |
5619 | <constant>SNDRV_CTL_POWER_D0</constant> to notify the processes.</para></listitem> | |
1da177e4 LT |
5620 | </orderedlist> |
5621 | </para> | |
5622 | ||
5623 | <para> | |
5624 | A typical code would be like: | |
5625 | ||
5626 | <informalexample> | |
5627 | <programlisting> | |
5628 | <![CDATA[ | |
5fe76e4d | 5629 | static int mychip_resume(struct pci_dev *pci) |
1da177e4 LT |
5630 | { |
5631 | /* (1) */ | |
5fe76e4d TI |
5632 | struct snd_card *card = pci_get_drvdata(pci); |
5633 | struct mychip *chip = card->private_data; | |
1da177e4 | 5634 | /* (2) */ |
5fe76e4d TI |
5635 | pci_restore_state(pci); |
5636 | pci_enable_device(pci); | |
5637 | pci_set_master(pci); | |
1da177e4 LT |
5638 | /* (3) */ |
5639 | snd_mychip_reinit_chip(chip); | |
5640 | /* (4) */ | |
5641 | snd_mychip_restore_registers(chip); | |
5642 | /* (5) */ | |
5643 | snd_ac97_resume(chip->ac97); | |
5644 | /* (6) */ | |
5645 | snd_mychip_restart_chip(chip); | |
5fe76e4d TI |
5646 | /* (7) */ |
5647 | snd_power_change_state(card, SNDRV_CTL_POWER_D0); | |
1da177e4 LT |
5648 | return 0; |
5649 | } | |
5650 | ]]> | |
5651 | </programlisting> | |
5652 | </informalexample> | |
5653 | </para> | |
5654 | ||
5655 | <para> | |
5fe76e4d TI |
5656 | As shown in the above, it's better to save registers after |
5657 | suspending the PCM operations via | |
5658 | <function>snd_pcm_suspend_all()</function> or | |
5659 | <function>snd_pcm_suspend()</function>. It means that the PCM | |
5660 | streams are already stoppped when the register snapshot is | |
5661 | taken. But, remind that you don't have to restart the PCM | |
5662 | stream in the resume callback. It'll be restarted via | |
5663 | trigger call with <constant>SNDRV_PCM_TRIGGER_RESUME</constant> | |
5664 | when necessary. | |
5665 | </para> | |
5666 | ||
5667 | <para> | |
5668 | OK, we have all callbacks now. Let's set them up. In the | |
5669 | initialization of the card, make sure that you can get the chip | |
5670 | data from the card instance, typically via | |
5671 | <structfield>private_data</structfield> field, in case you | |
5672 | created the chip data individually. | |
5673 | ||
5674 | <informalexample> | |
5675 | <programlisting> | |
5676 | <![CDATA[ | |
5677 | static int __devinit snd_mychip_probe(struct pci_dev *pci, | |
5678 | const struct pci_device_id *pci_id) | |
5679 | { | |
5680 | .... | |
5681 | struct snd_card *card; | |
5682 | struct mychip *chip; | |
5683 | .... | |
5684 | card = snd_card_new(index[dev], id[dev], THIS_MODULE, NULL); | |
5685 | .... | |
5686 | chip = kzalloc(sizeof(*chip), GFP_KERNEL); | |
5687 | .... | |
5688 | card->private_data = chip; | |
5689 | .... | |
5690 | } | |
5691 | ]]> | |
5692 | </programlisting> | |
5693 | </informalexample> | |
5694 | ||
5695 | When you created the chip data with | |
5696 | <function>snd_card_new()</function>, it's anyway accessible | |
5697 | via <structfield>private_data</structfield> field. | |
1da177e4 LT |
5698 | |
5699 | <informalexample> | |
5700 | <programlisting> | |
5701 | <![CDATA[ | |
5702 | static int __devinit snd_mychip_probe(struct pci_dev *pci, | |
5703 | const struct pci_device_id *pci_id) | |
5704 | { | |
5705 | .... | |
446ab5f5 TI |
5706 | struct snd_card *card; |
5707 | struct mychip *chip; | |
1da177e4 | 5708 | .... |
5fe76e4d TI |
5709 | card = snd_card_new(index[dev], id[dev], THIS_MODULE, |
5710 | sizeof(struct mychip)); | |
5711 | .... | |
5712 | chip = card->private_data; | |
1da177e4 LT |
5713 | .... |
5714 | } | |
5715 | ]]> | |
5716 | </programlisting> | |
5717 | </informalexample> | |
5718 | ||
1da177e4 LT |
5719 | </para> |
5720 | ||
5721 | <para> | |
5fe76e4d TI |
5722 | If you need a space for saving the registers, allocate the |
5723 | buffer for it here, too, since it would be fatal | |
1da177e4 LT |
5724 | if you cannot allocate a memory in the suspend phase. |
5725 | The allocated buffer should be released in the corresponding | |
5726 | destructor. | |
5727 | </para> | |
5728 | ||
5729 | <para> | |
5fe76e4d | 5730 | And next, set suspend/resume callbacks to the pci_driver. |
1da177e4 LT |
5731 | |
5732 | <informalexample> | |
5733 | <programlisting> | |
5734 | <![CDATA[ | |
5735 | static struct pci_driver driver = { | |
5736 | .name = "My Chip", | |
5737 | .id_table = snd_my_ids, | |
5738 | .probe = snd_my_probe, | |
5739 | .remove = __devexit_p(snd_my_remove), | |
5fe76e4d TI |
5740 | #ifdef CONFIG_PM |
5741 | .suspend = snd_my_suspend, | |
5742 | .resume = snd_my_resume, | |
5743 | #endif | |
1da177e4 LT |
5744 | }; |
5745 | ]]> | |
5746 | </programlisting> | |
5747 | </informalexample> | |
5748 | </para> | |
5749 | ||
5750 | </chapter> | |
5751 | ||
5752 | ||
5753 | <!-- ****************************************************** --> | |
5754 | <!-- Module Parameters --> | |
5755 | <!-- ****************************************************** --> | |
5756 | <chapter id="module-parameters"> | |
5757 | <title>Module Parameters</title> | |
5758 | <para> | |
5759 | There are standard module options for ALSA. At least, each | |
5760 | module should have <parameter>index</parameter>, | |
5761 | <parameter>id</parameter> and <parameter>enable</parameter> | |
5762 | options. | |
5763 | </para> | |
5764 | ||
5765 | <para> | |
5766 | If the module supports multiple cards (usually up to | |
5767 | 8 = <constant>SNDRV_CARDS</constant> cards), they should be | |
5768 | arrays. The default initial values are defined already as | |
5769 | constants for ease of programming: | |
5770 | ||
5771 | <informalexample> | |
5772 | <programlisting> | |
5773 | <![CDATA[ | |
5774 | static int index[SNDRV_CARDS] = SNDRV_DEFAULT_IDX; | |
5775 | static char *id[SNDRV_CARDS] = SNDRV_DEFAULT_STR; | |
5776 | static int enable[SNDRV_CARDS] = SNDRV_DEFAULT_ENABLE_PNP; | |
5777 | ]]> | |
5778 | </programlisting> | |
5779 | </informalexample> | |
5780 | </para> | |
5781 | ||
5782 | <para> | |
5783 | If the module supports only a single card, they could be single | |
5784 | variables, instead. <parameter>enable</parameter> option is not | |
5785 | always necessary in this case, but it wouldn't be so bad to have a | |
5786 | dummy option for compatibility. | |
5787 | </para> | |
5788 | ||
5789 | <para> | |
5790 | The module parameters must be declared with the standard | |
5791 | <function>module_param()()</function>, | |
5792 | <function>module_param_array()()</function> and | |
5793 | <function>MODULE_PARM_DESC()</function> macros. | |
5794 | </para> | |
5795 | ||
5796 | <para> | |
5797 | The typical coding would be like below: | |
5798 | ||
5799 | <informalexample> | |
5800 | <programlisting> | |
5801 | <![CDATA[ | |
5802 | #define CARD_NAME "My Chip" | |
5803 | ||
5804 | module_param_array(index, int, NULL, 0444); | |
5805 | MODULE_PARM_DESC(index, "Index value for " CARD_NAME " soundcard."); | |
5806 | module_param_array(id, charp, NULL, 0444); | |
5807 | MODULE_PARM_DESC(id, "ID string for " CARD_NAME " soundcard."); | |
5808 | module_param_array(enable, bool, NULL, 0444); | |
5809 | MODULE_PARM_DESC(enable, "Enable " CARD_NAME " soundcard."); | |
5810 | ]]> | |
5811 | </programlisting> | |
5812 | </informalexample> | |
5813 | </para> | |
5814 | ||
5815 | <para> | |
5816 | Also, don't forget to define the module description, classes, | |
5817 | license and devices. Especially, the recent modprobe requires to | |
5818 | define the module license as GPL, etc., otherwise the system is | |
5819 | shown as <quote>tainted</quote>. | |
5820 | ||
5821 | <informalexample> | |
5822 | <programlisting> | |
5823 | <![CDATA[ | |
5824 | MODULE_DESCRIPTION("My Chip"); | |
5825 | MODULE_LICENSE("GPL"); | |
5826 | MODULE_SUPPORTED_DEVICE("{{Vendor,My Chip Name}}"); | |
5827 | ]]> | |
5828 | </programlisting> | |
5829 | </informalexample> | |
5830 | </para> | |
5831 | ||
5832 | </chapter> | |
5833 | ||
5834 | ||
5835 | <!-- ****************************************************** --> | |
5836 | <!-- How To Put Your Driver --> | |
5837 | <!-- ****************************************************** --> | |
5838 | <chapter id="how-to-put-your-driver"> | |
5839 | <title>How To Put Your Driver Into ALSA Tree</title> | |
5840 | <section> | |
5841 | <title>General</title> | |
5842 | <para> | |
5843 | So far, you've learned how to write the driver codes. | |
5844 | And you might have a question now: how to put my own | |
5845 | driver into the ALSA driver tree? | |
5846 | Here (finally :) the standard procedure is described briefly. | |
5847 | </para> | |
5848 | ||
5849 | <para> | |
5850 | Suppose that you'll create a new PCI driver for the card | |
5851 | <quote>xyz</quote>. The card module name would be | |
5852 | snd-xyz. The new driver is usually put into alsa-driver | |
5853 | tree, <filename>alsa-driver/pci</filename> directory in | |
5854 | the case of PCI cards. | |
5855 | Then the driver is evaluated, audited and tested | |
5856 | by developers and users. After a certain time, the driver | |
5857 | will go to alsa-kernel tree (to the corresponding directory, | |
5858 | such as <filename>alsa-kernel/pci</filename>) and eventually | |
5859 | integrated into Linux 2.6 tree (the directory would be | |
5860 | <filename>linux/sound/pci</filename>). | |
5861 | </para> | |
5862 | ||
5863 | <para> | |
5864 | In the following sections, the driver code is supposed | |
5865 | to be put into alsa-driver tree. The two cases are assumed: | |
5866 | a driver consisting of a single source file and one consisting | |
5867 | of several source files. | |
5868 | </para> | |
5869 | </section> | |
5870 | ||
5871 | <section> | |
5872 | <title>Driver with A Single Source File</title> | |
5873 | <para> | |
5874 | <orderedlist> | |
5875 | <listitem> | |
5876 | <para> | |
5877 | Modify alsa-driver/pci/Makefile | |
5878 | </para> | |
5879 | ||
5880 | <para> | |
5881 | Suppose you have a file xyz.c. Add the following | |
5882 | two lines | |
5883 | <informalexample> | |
5884 | <programlisting> | |
5885 | <![CDATA[ | |
5886 | snd-xyz-objs := xyz.o | |
5887 | obj-$(CONFIG_SND_XYZ) += snd-xyz.o | |
5888 | ]]> | |
5889 | </programlisting> | |
5890 | </informalexample> | |
5891 | </para> | |
5892 | </listitem> | |
5893 | ||
5894 | <listitem> | |
5895 | <para> | |
5896 | Create the Kconfig entry | |
5897 | </para> | |
5898 | ||
5899 | <para> | |
5900 | Add the new entry of Kconfig for your xyz driver. | |
5901 | <informalexample> | |
5902 | <programlisting> | |
5903 | <![CDATA[ | |
5904 | config SND_XYZ | |
5905 | tristate "Foobar XYZ" | |
5906 | depends on SND | |
5907 | select SND_PCM | |
5908 | help | |
5909 | Say Y here to include support for Foobar XYZ soundcard. | |
5910 | ||
5911 | To compile this driver as a module, choose M here: the module | |
5912 | will be called snd-xyz. | |
5913 | ]]> | |
5914 | </programlisting> | |
5915 | </informalexample> | |
5916 | ||
5917 | the line, select SND_PCM, specifies that the driver xyz supports | |
5918 | PCM. In addition to SND_PCM, the following components are | |
5919 | supported for select command: | |
5920 | SND_RAWMIDI, SND_TIMER, SND_HWDEP, SND_MPU401_UART, | |
5921 | SND_OPL3_LIB, SND_OPL4_LIB, SND_VX_LIB, SND_AC97_CODEC. | |
5922 | Add the select command for each supported component. | |
5923 | </para> | |
5924 | ||
5925 | <para> | |
5926 | Note that some selections imply the lowlevel selections. | |
5927 | For example, PCM includes TIMER, MPU401_UART includes RAWMIDI, | |
5928 | AC97_CODEC includes PCM, and OPL3_LIB includes HWDEP. | |
5929 | You don't need to give the lowlevel selections again. | |
5930 | </para> | |
5931 | ||
5932 | <para> | |
5933 | For the details of Kconfig script, refer to the kbuild | |
5934 | documentation. | |
5935 | </para> | |
5936 | ||
5937 | </listitem> | |
5938 | ||
5939 | <listitem> | |
5940 | <para> | |
5941 | Run cvscompile script to re-generate the configure script and | |
5942 | build the whole stuff again. | |
5943 | </para> | |
5944 | </listitem> | |
5945 | </orderedlist> | |
5946 | </para> | |
5947 | </section> | |
5948 | ||
5949 | <section> | |
5950 | <title>Drivers with Several Source Files</title> | |
5951 | <para> | |
5952 | Suppose that the driver snd-xyz have several source files. | |
5953 | They are located in the new subdirectory, | |
5954 | pci/xyz. | |
5955 | ||
5956 | <orderedlist> | |
5957 | <listitem> | |
5958 | <para> | |
5959 | Add a new directory (<filename>xyz</filename>) in | |
5960 | <filename>alsa-driver/pci/Makefile</filename> like below | |
5961 | ||
5962 | <informalexample> | |
5963 | <programlisting> | |
5964 | <![CDATA[ | |
5965 | obj-$(CONFIG_SND) += xyz/ | |
5966 | ]]> | |
5967 | </programlisting> | |
5968 | </informalexample> | |
5969 | </para> | |
5970 | </listitem> | |
5971 | ||
5972 | <listitem> | |
5973 | <para> | |
5974 | Under the directory <filename>xyz</filename>, create a Makefile | |
5975 | ||
5976 | <example> | |
5977 | <title>Sample Makefile for a driver xyz</title> | |
5978 | <programlisting> | |
5979 | <![CDATA[ | |
5980 | ifndef SND_TOPDIR | |
5981 | SND_TOPDIR=../.. | |
5982 | endif | |
5983 | ||
5984 | include $(SND_TOPDIR)/toplevel.config | |
5985 | include $(SND_TOPDIR)/Makefile.conf | |
5986 | ||
5987 | snd-xyz-objs := xyz.o abc.o def.o | |
5988 | ||
5989 | obj-$(CONFIG_SND_XYZ) += snd-xyz.o | |
5990 | ||
5991 | include $(SND_TOPDIR)/Rules.make | |
5992 | ]]> | |
5993 | </programlisting> | |
5994 | </example> | |
5995 | </para> | |
5996 | </listitem> | |
5997 | ||
5998 | <listitem> | |
5999 | <para> | |
6000 | Create the Kconfig entry | |
6001 | </para> | |
6002 | ||
6003 | <para> | |
6004 | This procedure is as same as in the last section. | |
6005 | </para> | |
6006 | </listitem> | |
6007 | ||
6008 | <listitem> | |
6009 | <para> | |
6010 | Run cvscompile script to re-generate the configure script and | |
6011 | build the whole stuff again. | |
6012 | </para> | |
6013 | </listitem> | |
6014 | </orderedlist> | |
6015 | </para> | |
6016 | </section> | |
6017 | ||
6018 | </chapter> | |
6019 | ||
6020 | <!-- ****************************************************** --> | |
6021 | <!-- Useful Functions --> | |
6022 | <!-- ****************************************************** --> | |
6023 | <chapter id="useful-functions"> | |
6024 | <title>Useful Functions</title> | |
6025 | ||
6026 | <section id="useful-functions-snd-printk"> | |
6027 | <title><function>snd_printk()</function> and friends</title> | |
6028 | <para> | |
6029 | ALSA provides a verbose version of | |
6030 | <function>printk()</function> function. If a kernel config | |
6031 | <constant>CONFIG_SND_VERBOSE_PRINTK</constant> is set, this | |
6032 | function prints the given message together with the file name | |
6033 | and the line of the caller. The <constant>KERN_XXX</constant> | |
6034 | prefix is processed as | |
6035 | well as the original <function>printk()</function> does, so it's | |
6036 | recommended to add this prefix, e.g. | |
6037 | ||
6038 | <informalexample> | |
6039 | <programlisting> | |
6040 | <![CDATA[ | |
6041 | snd_printk(KERN_ERR "Oh my, sorry, it's extremely bad!\n"); | |
6042 | ]]> | |
6043 | </programlisting> | |
6044 | </informalexample> | |
6045 | </para> | |
6046 | ||
6047 | <para> | |
6048 | There are also <function>printk()</function>'s for | |
6049 | debugging. <function>snd_printd()</function> can be used for | |
6050 | general debugging purposes. If | |
6051 | <constant>CONFIG_SND_DEBUG</constant> is set, this function is | |
6052 | compiled, and works just like | |
6053 | <function>snd_printk()</function>. If the ALSA is compiled | |
6054 | without the debugging flag, it's ignored. | |
6055 | </para> | |
6056 | ||
6057 | <para> | |
6058 | <function>snd_printdd()</function> is compiled in only when | |
6059 | <constant>CONFIG_SND_DEBUG_DETECT</constant> is set. Please note | |
6060 | that <constant>DEBUG_DETECT</constant> is not set as default | |
6061 | even if you configure the alsa-driver with | |
6062 | <option>--with-debug=full</option> option. You need to give | |
6063 | explicitly <option>--with-debug=detect</option> option instead. | |
6064 | </para> | |
6065 | </section> | |
6066 | ||
6067 | <section id="useful-functions-snd-assert"> | |
6068 | <title><function>snd_assert()</function></title> | |
6069 | <para> | |
6070 | <function>snd_assert()</function> macro is similar with the | |
6071 | normal <function>assert()</function> macro. For example, | |
6072 | ||
6073 | <informalexample> | |
6074 | <programlisting> | |
6075 | <![CDATA[ | |
6076 | snd_assert(pointer != NULL, return -EINVAL); | |
6077 | ]]> | |
6078 | </programlisting> | |
6079 | </informalexample> | |
6080 | </para> | |
6081 | ||
6082 | <para> | |
6083 | The first argument is the expression to evaluate, and the | |
6084 | second argument is the action if it fails. When | |
6085 | <constant>CONFIG_SND_DEBUG</constant>, is set, it will show an | |
7c22f1aa TI |
6086 | error message such as <computeroutput>BUG? (xxx)</computeroutput> |
6087 | together with stack trace. | |
1da177e4 | 6088 | </para> |
1da177e4 | 6089 | <para> |
7c22f1aa | 6090 | When no debug flag is set, this macro is ignored. |
1da177e4 LT |
6091 | </para> |
6092 | </section> | |
6093 | ||
6094 | <section id="useful-functions-snd-bug"> | |
6095 | <title><function>snd_BUG()</function></title> | |
6096 | <para> | |
7c22f1aa TI |
6097 | It shows <computeroutput>BUG?</computeroutput> message and |
6098 | stack trace as well as <function>snd_assert</function> at the point. | |
6099 | It's useful to show that a fatal error happens there. | |
6100 | </para> | |
6101 | <para> | |
6102 | When no debug flag is set, this macro is ignored. | |
1da177e4 LT |
6103 | </para> |
6104 | </section> | |
6105 | </chapter> | |
6106 | ||
6107 | ||
6108 | <!-- ****************************************************** --> | |
6109 | <!-- Acknowledgments --> | |
6110 | <!-- ****************************************************** --> | |
6111 | <chapter id="acknowledments"> | |
6112 | <title>Acknowledgments</title> | |
6113 | <para> | |
6114 | I would like to thank Phil Kerr for his help for improvement and | |
6115 | corrections of this document. | |
6116 | </para> | |
6117 | <para> | |
6118 | Kevin Conder reformatted the original plain-text to the | |
6119 | DocBook format. | |
6120 | </para> | |
6121 | <para> | |
6122 | Giuliano Pochini corrected typos and contributed the example codes | |
6123 | in the hardware constraints section. | |
6124 | </para> | |
6125 | </chapter> | |
6126 | ||
6127 | ||
6128 | </book> |