blob: 7a2e0e98986a744230a032970d9a072b6a9b8422 [file] [log] [blame]
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook XML V4.1.2//EN"
"http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd" []>
<!-- ****************************************************** -->
<!-- Header -->
<!-- ****************************************************** -->
<book id="Writing-an-ALSA-Driver">
<bookinfo>
<title>Writing an ALSA Driver</title>
<author>
<firstname>Takashi</firstname>
<surname>Iwai</surname>
<affiliation>
<address>
<email>tiwai@suse.de</email>
</address>
</affiliation>
</author>
<date>Oct 15, 2007</date>
<edition>0.3.7</edition>
<abstract>
<para>
This document describes how to write an ALSA (Advanced Linux
Sound Architecture) driver.
</para>
</abstract>
<legalnotice>
<para>
Copyright (c) 2002-2005 Takashi Iwai <email>tiwai@suse.de</email>
</para>
<para>
This document is free; you can redistribute it and/or modify it
under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or
(at your option) any later version.
</para>
<para>
This document is distributed in the hope that it will be useful,
but <emphasis>WITHOUT ANY WARRANTY</emphasis>; without even the
implied warranty of <emphasis>MERCHANTABILITY or FITNESS FOR A
PARTICULAR PURPOSE</emphasis>. See the GNU General Public License
for more details.
</para>
<para>
You should have received a copy of the GNU General Public
License along with this program; if not, write to the Free
Software Foundation, Inc., 59 Temple Place, Suite 330, Boston,
MA 02111-1307 USA
</para>
</legalnotice>
</bookinfo>
<!-- ****************************************************** -->
<!-- Preface -->
<!-- ****************************************************** -->
<preface id="preface">
<title>Preface</title>
<para>
This document describes how to write an
<ulink url="http://www.alsa-project.org/"><citetitle>
ALSA (Advanced Linux Sound Architecture)</citetitle></ulink>
driver. The document focuses mainly on PCI soundcards.
In the case of other device types, the API might
be different, too. However, at least the ALSA kernel API is
consistent, and therefore it would be still a bit help for
writing them.
</para>
<para>
This document targets people who already have enough
C language skills and have basic linux kernel programming
knowledge. This document doesn't explain the general
topic of linux kernel coding and doesn't cover low-level
driver implementation details. It only describes
the standard way to write a PCI sound driver on ALSA.
</para>
<para>
If you are already familiar with the older ALSA ver.0.5.x API, you
can check the drivers such as <filename>sound/pci/es1938.c</filename> or
<filename>sound/pci/maestro3.c</filename> which have also almost the same
code-base in the ALSA 0.5.x tree, so you can compare the differences.
</para>
<para>
This document is still a draft version. Any feedback and
corrections, please!!
</para>
</preface>
<!-- ****************************************************** -->
<!-- File Tree Structure -->
<!-- ****************************************************** -->
<chapter id="file-tree">
<title>File Tree Structure</title>
<section id="file-tree-general">
<title>General</title>
<para>
The ALSA drivers are provided in two ways.
</para>
<para>
One is the trees provided as a tarball or via cvs from the
ALSA's ftp site, and another is the 2.6 (or later) Linux kernel
tree. To synchronize both, the ALSA driver tree is split into
two different trees: alsa-kernel and alsa-driver. The former
contains purely the source code for the Linux 2.6 (or later)
tree. This tree is designed only for compilation on 2.6 or
later environment. The latter, alsa-driver, contains many subtle
files for compiling ALSA drivers outside of the Linux kernel tree,
wrapper functions for older 2.2 and 2.4 kernels, to adapt the latest kernel API,
and additional drivers which are still in development or in
tests. The drivers in alsa-driver tree will be moved to
alsa-kernel (and eventually to the 2.6 kernel tree) when they are
finished and confirmed to work fine.
</para>
<para>
The file tree structure of ALSA driver is depicted below. Both
alsa-kernel and alsa-driver have almost the same file
structure, except for <quote>core</quote> directory. It's
named as <quote>acore</quote> in alsa-driver tree.
<example>
<title>ALSA File Tree Structure</title>
<literallayout>
sound
/core
/oss
/seq
/oss
/instr
/ioctl32
/include
/drivers
/mpu401
/opl3
/i2c
/l3
/synth
/emux
/pci
/(cards)
/isa
/(cards)
/arm
/ppc
/sparc
/usb
/pcmcia /(cards)
/oss
</literallayout>
</example>
</para>
</section>
<section id="file-tree-core-directory">
<title>core directory</title>
<para>
This directory contains the middle layer which is the heart
of ALSA drivers. In this directory, the native ALSA modules are
stored. The sub-directories contain different modules and are
dependent upon the kernel config.
</para>
<section id="file-tree-core-directory-oss">
<title>core/oss</title>
<para>
The codes for PCM and mixer OSS emulation modules are stored
in this directory. The rawmidi OSS emulation is included in
the ALSA rawmidi code since it's quite small. The sequencer
code is stored in <filename>core/seq/oss</filename> directory (see
<link linkend="file-tree-core-directory-seq-oss"><citetitle>
below</citetitle></link>).
</para>
</section>
<section id="file-tree-core-directory-ioctl32">
<title>core/ioctl32</title>
<para>
This directory contains the 32bit-ioctl wrappers for 64bit
architectures such like x86-64, ppc64 and sparc64. For 32bit
and alpha architectures, these are not compiled.
</para>
</section>
<section id="file-tree-core-directory-seq">
<title>core/seq</title>
<para>
This directory and its sub-directories are for the ALSA
sequencer. This directory contains the sequencer core and
primary sequencer modules such like snd-seq-midi,
snd-seq-virmidi, etc. They are compiled only when
<constant>CONFIG_SND_SEQUENCER</constant> is set in the kernel
config.
</para>
</section>
<section id="file-tree-core-directory-seq-oss">
<title>core/seq/oss</title>
<para>
This contains the OSS sequencer emulation codes.
</para>
</section>
<section id="file-tree-core-directory-deq-instr">
<title>core/seq/instr</title>
<para>
This directory contains the modules for the sequencer
instrument layer.
</para>
</section>
</section>
<section id="file-tree-include-directory">
<title>include directory</title>
<para>
This is the place for the public header files of ALSA drivers,
which are to be exported to user-space, or included by
several files at different directories. Basically, the private
header files should not be placed in this directory, but you may
still find files there, due to historical reasons :)
</para>
</section>
<section id="file-tree-drivers-directory">
<title>drivers directory</title>
<para>
This directory contains code shared among different drivers
on different architectures. They are hence supposed not to be
architecture-specific.
For example, the dummy pcm driver and the serial MIDI
driver are found in this directory. In the sub-directories,
there is code for components which are independent from
bus and cpu architectures.
</para>
<section id="file-tree-drivers-directory-mpu401">
<title>drivers/mpu401</title>
<para>
The MPU401 and MPU401-UART modules are stored here.
</para>
</section>
<section id="file-tree-drivers-directory-opl3">
<title>drivers/opl3 and opl4</title>
<para>
The OPL3 and OPL4 FM-synth stuff is found here.
</para>
</section>
</section>
<section id="file-tree-i2c-directory">
<title>i2c directory</title>
<para>
This contains the ALSA i2c components.
</para>
<para>
Although there is a standard i2c layer on Linux, ALSA has its
own i2c code for some cards, because the soundcard needs only a
simple operation and the standard i2c API is too complicated for
such a purpose.
</para>
<section id="file-tree-i2c-directory-l3">
<title>i2c/l3</title>
<para>
This is a sub-directory for ARM L3 i2c.
</para>
</section>
</section>
<section id="file-tree-synth-directory">
<title>synth directory</title>
<para>
This contains the synth middle-level modules.
</para>
<para>
So far, there is only Emu8000/Emu10k1 synth driver under
the <filename>synth/emux</filename> sub-directory.
</para>
</section>
<section id="file-tree-pci-directory">
<title>pci directory</title>
<para>
This directory and its sub-directories hold the top-level card modules
for PCI soundcards and the code specific to the PCI BUS.
</para>
<para>
The drivers compiled from a single file are stored directly
in the pci directory, while the drivers with several source files are
stored on their own sub-directory (e.g. emu10k1, ice1712).
</para>
</section>
<section id="file-tree-isa-directory">
<title>isa directory</title>
<para>
This directory and its sub-directories hold the top-level card modules
for ISA soundcards.
</para>
</section>
<section id="file-tree-arm-ppc-sparc-directories">
<title>arm, ppc, and sparc directories</title>
<para>
They are used for top-level card modules which are
specific to one of these architectures.
</para>
</section>
<section id="file-tree-usb-directory">
<title>usb directory</title>
<para>
This directory contains the USB-audio driver. In the latest version, the
USB MIDI driver is integrated in the usb-audio driver.
</para>
</section>
<section id="file-tree-pcmcia-directory">
<title>pcmcia directory</title>
<para>
The PCMCIA, especially PCCard drivers will go here. CardBus
drivers will be in the pci directory, because their API is identical
to that of standard PCI cards.
</para>
</section>
<section id="file-tree-oss-directory">
<title>oss directory</title>
<para>
The OSS/Lite source files are stored here in Linux 2.6 (or
later) tree. In the ALSA driver tarball, this directory is empty,
of course :)
</para>
</section>
</chapter>
<!-- ****************************************************** -->
<!-- Basic Flow for PCI Drivers -->
<!-- ****************************************************** -->
<chapter id="basic-flow">
<title>Basic Flow for PCI Drivers</title>
<section id="basic-flow-outline">
<title>Outline</title>
<para>
The minimum flow for PCI soundcards is as follows:
<itemizedlist>
<listitem><para>define the PCI ID table (see the section
<link linkend="pci-resource-entries"><citetitle>PCI Entries
</citetitle></link>).</para></listitem>
<listitem><para>create <function>probe()</function> callback.</para></listitem>
<listitem><para>create <function>remove()</function> callback.</para></listitem>
<listitem><para>create a <structname>pci_driver</structname> structure
containing the three pointers above.</para></listitem>
<listitem><para>create an <function>init()</function> function just calling
the <function>pci_register_driver()</function> to register the pci_driver table
defined above.</para></listitem>
<listitem><para>create an <function>exit()</function> function to call
the <function>pci_unregister_driver()</function> function.</para></listitem>
</itemizedlist>
</para>
</section>
<section id="basic-flow-example">
<title>Full Code Example</title>
<para>
The code example is shown below. Some parts are kept
unimplemented at this moment but will be filled in the
next sections. The numbers in the comment lines of the
<function>snd_mychip_probe()</function> function
refer to details explained in the following section.
<example>
<title>Basic Flow for PCI Drivers - Example</title>
<programlisting>
<![CDATA[
#include <linux/init.h>
#include <linux/pci.h>
#include <linux/slab.h>
#include <sound/core.h>
#include <sound/initval.h>
/* module parameters (see "Module Parameters") */
/* SNDRV_CARDS: maximum number of cards supported by this module */
static int index[SNDRV_CARDS] = SNDRV_DEFAULT_IDX;
static char *id[SNDRV_CARDS] = SNDRV_DEFAULT_STR;
static int enable[SNDRV_CARDS] = SNDRV_DEFAULT_ENABLE_PNP;
/* definition of the chip-specific record */
struct mychip {
struct snd_card *card;
/* the rest of the implementation will be in section
* "PCI Resource Management"
*/
};
/* chip-specific destructor
* (see "PCI Resource Management")
*/
static int snd_mychip_free(struct mychip *chip)
{
.... /* will be implemented later... */
}
/* component-destructor
* (see "Management of Cards and Components")
*/
static int snd_mychip_dev_free(struct snd_device *device)
{
return snd_mychip_free(device->device_data);
}
/* chip-specific constructor
* (see "Management of Cards and Components")
*/
static int __devinit snd_mychip_create(struct snd_card *card,
struct pci_dev *pci,
struct mychip **rchip)
{
struct mychip *chip;
int err;
static struct snd_device_ops ops = {
.dev_free = snd_mychip_dev_free,
};
*rchip = NULL;
/* check PCI availability here
* (see "PCI Resource Management")
*/
....
/* allocate a chip-specific data with zero filled */
chip = kzalloc(sizeof(*chip), GFP_KERNEL);
if (chip == NULL)
return -ENOMEM;
chip->card = card;
/* rest of initialization here; will be implemented
* later, see "PCI Resource Management"
*/
....
err = snd_device_new(card, SNDRV_DEV_LOWLEVEL, chip, &ops);
if (err < 0) {
snd_mychip_free(chip);
return err;
}
snd_card_set_dev(card, &pci->dev);
*rchip = chip;
return 0;
}
/* constructor -- see "Constructor" sub-section */
static int __devinit snd_mychip_probe(struct pci_dev *pci,
const struct pci_device_id *pci_id)
{
static int dev;
struct snd_card *card;
struct mychip *chip;
int err;
/* (1) */
if (dev >= SNDRV_CARDS)
return -ENODEV;
if (!enable[dev]) {
dev++;
return -ENOENT;
}
/* (2) */
err = snd_card_create(index[dev], id[dev], THIS_MODULE, 0, &card);
if (err < 0)
return err;
/* (3) */
err = snd_mychip_create(card, pci, &chip);
if (err < 0) {
snd_card_free(card);
return err;
}
/* (4) */
strcpy(card->driver, "My Chip");
strcpy(card->shortname, "My Own Chip 123");
sprintf(card->longname, "%s at 0x%lx irq %i",
card->shortname, chip->ioport, chip->irq);
/* (5) */
.... /* implemented later */
/* (6) */
err = snd_card_register(card);
if (err < 0) {
snd_card_free(card);
return err;
}
/* (7) */
pci_set_drvdata(pci, card);
dev++;
return 0;
}
/* destructor -- see the "Destructor" sub-section */
static void __devexit snd_mychip_remove(struct pci_dev *pci)
{
snd_card_free(pci_get_drvdata(pci));
pci_set_drvdata(pci, NULL);
}
]]>
</programlisting>
</example>
</para>
</section>
<section id="basic-flow-constructor">
<title>Constructor</title>
<para>
The real constructor of PCI drivers is the <function>probe</function> callback.
The <function>probe</function> callback and other component-constructors which are called
from the <function>probe</function> callback should be defined with
the <parameter>__devinit</parameter> prefix. You
cannot use the <parameter>__init</parameter> prefix for them,
because any PCI device could be a hotplug device.
</para>
<para>
In the <function>probe</function> callback, the following scheme is often used.
</para>
<section id="basic-flow-constructor-device-index">
<title>1) Check and increment the device index.</title>
<para>
<informalexample>
<programlisting>
<![CDATA[
static int dev;
....
if (dev >= SNDRV_CARDS)
return -ENODEV;
if (!enable[dev]) {
dev++;
return -ENOENT;
}
]]>
</programlisting>
</informalexample>
where enable[dev] is the module option.
</para>
<para>
Each time the <function>probe</function> callback is called, check the
availability of the device. If not available, simply increment
the device index and returns. dev will be incremented also
later (<link
linkend="basic-flow-constructor-set-pci"><citetitle>step
7</citetitle></link>).
</para>
</section>
<section id="basic-flow-constructor-create-card">
<title>2) Create a card instance</title>
<para>
<informalexample>
<programlisting>
<![CDATA[
struct snd_card *card;
int err;
....
err = snd_card_create(index[dev], id[dev], THIS_MODULE, 0, &card);
]]>
</programlisting>
</informalexample>
</para>
<para>
The details will be explained in the section
<link linkend="card-management-card-instance"><citetitle>
Management of Cards and Components</citetitle></link>.
</para>
</section>
<section id="basic-flow-constructor-create-main">
<title>3) Create a main component</title>
<para>
In this part, the PCI resources are allocated.
<informalexample>
<programlisting>
<![CDATA[
struct mychip *chip;
....
err = snd_mychip_create(card, pci, &chip);
if (err < 0) {
snd_card_free(card);
return err;
}
]]>
</programlisting>
</informalexample>
The details will be explained in the section <link
linkend="pci-resource"><citetitle>PCI Resource
Management</citetitle></link>.
</para>
</section>
<section id="basic-flow-constructor-main-component">
<title>4) Set the driver ID and name strings.</title>
<para>
<informalexample>
<programlisting>
<![CDATA[
strcpy(card->driver, "My Chip");
strcpy(card->shortname, "My Own Chip 123");
sprintf(card->longname, "%s at 0x%lx irq %i",
card->shortname, chip->ioport, chip->irq);
]]>
</programlisting>
</informalexample>
The driver field holds the minimal ID string of the
chip. This is used by alsa-lib's configurator, so keep it
simple but unique.
Even the same driver can have different driver IDs to
distinguish the functionality of each chip type.
</para>
<para>
The shortname field is a string shown as more verbose
name. The longname field contains the information
shown in <filename>/proc/asound/cards</filename>.
</para>
</section>
<section id="basic-flow-constructor-create-other">
<title>5) Create other components, such as mixer, MIDI, etc.</title>
<para>
Here you define the basic components such as
<link linkend="pcm-interface"><citetitle>PCM</citetitle></link>,
mixer (e.g. <link linkend="api-ac97"><citetitle>AC97</citetitle></link>),
MIDI (e.g. <link linkend="midi-interface"><citetitle>MPU-401</citetitle></link>),
and other interfaces.
Also, if you want a <link linkend="proc-interface"><citetitle>proc
file</citetitle></link>, define it here, too.
</para>
</section>
<section id="basic-flow-constructor-register-card">
<title>6) Register the card instance.</title>
<para>
<informalexample>
<programlisting>
<![CDATA[
err = snd_card_register(card);
if (err < 0) {
snd_card_free(card);
return err;
}
]]>
</programlisting>
</informalexample>
</para>
<para>
Will be explained in the section <link
linkend="card-management-registration"><citetitle>Management
of Cards and Components</citetitle></link>, too.
</para>
</section>
<section id="basic-flow-constructor-set-pci">
<title>7) Set the PCI driver data and return zero.</title>
<para>
<informalexample>
<programlisting>
<![CDATA[
pci_set_drvdata(pci, card);
dev++;
return 0;
]]>
</programlisting>
</informalexample>
In the above, the card record is stored. This pointer is
used in the remove callback and power-management
callbacks, too.
</para>
</section>
</section>
<section id="basic-flow-destructor">
<title>Destructor</title>
<para>
The destructor, remove callback, simply releases the card
instance. Then the ALSA middle layer will release all the
attached components automatically.
</para>
<para>
It would be typically like the following:
<informalexample>
<programlisting>
<![CDATA[
static void __devexit snd_mychip_remove(struct pci_dev *pci)
{
snd_card_free(pci_get_drvdata(pci));
pci_set_drvdata(pci, NULL);
}
]]>
</programlisting>
</informalexample>
The above code assumes that the card pointer is set to the PCI
driver data.
</para>
</section>
<section id="basic-flow-header-files">
<title>Header Files</title>
<para>
For the above example, at least the following include files
are necessary.
<informalexample>
<programlisting>
<![CDATA[
#include <linux/init.h>
#include <linux/pci.h>
#include <linux/slab.h>
#include <sound/core.h>
#include <sound/initval.h>
]]>
</programlisting>
</informalexample>
where the last one is necessary only when module options are
defined in the source file. If the code is split into several
files, the files without module options don't need them.
</para>
<para>
In addition to these headers, you'll need
<filename>&lt;linux/interrupt.h&gt;</filename> for interrupt
handling, and <filename>&lt;asm/io.h&gt;</filename> for I/O
access. If you use the <function>mdelay()</function> or
<function>udelay()</function> functions, you'll need to include
<filename>&lt;linux/delay.h&gt;</filename> too.
</para>
<para>
The ALSA interfaces like the PCM and control APIs are defined in other
<filename>&lt;sound/xxx.h&gt;</filename> header files.
They have to be included after
<filename>&lt;sound/core.h&gt;</filename>.
</para>
</section>
</chapter>
<!-- ****************************************************** -->
<!-- Management of Cards and Components -->
<!-- ****************************************************** -->
<chapter id="card-management">
<title>Management of Cards and Components</title>
<section id="card-management-card-instance">
<title>Card Instance</title>
<para>
For each soundcard, a <quote>card</quote> record must be allocated.
</para>
<para>
A card record is the headquarters of the soundcard. It manages
the whole list of devices (components) on the soundcard, such as
PCM, mixers, MIDI, synthesizer, and so on. Also, the card
record holds the ID and the name strings of the card, manages
the root of proc files, and controls the power-management states
and hotplug disconnections. The component list on the card
record is used to manage the correct release of resources at
destruction.
</para>
<para>
As mentioned above, to create a card instance, call
<function>snd_card_create()</function>.
<informalexample>
<programlisting>
<![CDATA[
struct snd_card *card;
int err;
err = snd_card_create(index, id, module, extra_size, &card);
]]>
</programlisting>
</informalexample>
</para>
<para>
The function takes five arguments, the card-index number, the
id string, the module pointer (usually
<constant>THIS_MODULE</constant>),
the size of extra-data space, and the pointer to return the
card instance. The extra_size argument is used to
allocate card-&gt;private_data for the
chip-specific data. Note that these data
are allocated by <function>snd_card_create()</function>.
</para>
</section>
<section id="card-management-component">
<title>Components</title>
<para>
After the card is created, you can attach the components
(devices) to the card instance. In an ALSA driver, a component is
represented as a struct <structname>snd_device</structname> object.
A component can be a PCM instance, a control interface, a raw
MIDI interface, etc. Each such instance has one component
entry.
</para>
<para>
A component can be created via
<function>snd_device_new()</function> function.
<informalexample>
<programlisting>
<![CDATA[
snd_device_new(card, SNDRV_DEV_XXX, chip, &ops);
]]>
</programlisting>
</informalexample>
</para>
<para>
This takes the card pointer, the device-level
(<constant>SNDRV_DEV_XXX</constant>), the data pointer, and the
callback pointers (<parameter>&amp;ops</parameter>). The
device-level defines the type of components and the order of
registration and de-registration. For most components, the
device-level is already defined. For a user-defined component,
you can use <constant>SNDRV_DEV_LOWLEVEL</constant>.
</para>
<para>
This function itself doesn't allocate the data space. The data
must be allocated manually beforehand, and its pointer is passed
as the argument. This pointer is used as the
(<parameter>chip</parameter> identifier in the above example)
for the instance.
</para>
<para>
Each pre-defined ALSA component such as ac97 and pcm calls
<function>snd_device_new()</function> inside its
constructor. The destructor for each component is defined in the
callback pointers. Hence, you don't need to take care of
calling a destructor for such a component.
</para>
<para>
If you wish to create your own component, you need to
set the destructor function to the dev_free callback in
the <parameter>ops</parameter>, so that it can be released
automatically via <function>snd_card_free()</function>.
The next example will show an implementation of chip-specific
data.
</para>
</section>
<section id="card-management-chip-specific">
<title>Chip-Specific Data</title>
<para>
Chip-specific information, e.g. the I/O port address, its
resource pointer, or the irq number, is stored in the
chip-specific record.
<informalexample>
<programlisting>
<![CDATA[
struct mychip {
....
};
]]>
</programlisting>
</informalexample>
</para>
<para>
In general, there are two ways of allocating the chip record.
</para>
<section id="card-management-chip-specific-snd-card-new">
<title>1. Allocating via <function>snd_card_create()</function>.</title>
<para>
As mentioned above, you can pass the extra-data-length
to the 4th argument of <function>snd_card_create()</function>, i.e.
<informalexample>
<programlisting>
<![CDATA[
err = snd_card_create(index[dev], id[dev], THIS_MODULE,
sizeof(struct mychip), &card);
]]>
</programlisting>
</informalexample>
struct <structname>mychip</structname> is the type of the chip record.
</para>
<para>
In return, the allocated record can be accessed as
<informalexample>
<programlisting>
<![CDATA[
struct mychip *chip = card->private_data;
]]>
</programlisting>
</informalexample>
With this method, you don't have to allocate twice.
The record is released together with the card instance.
</para>
</section>
<section id="card-management-chip-specific-allocate-extra">
<title>2. Allocating an extra device.</title>
<para>
After allocating a card instance via
<function>snd_card_create()</function> (with
<constant>0</constant> on the 4th arg), call
<function>kzalloc()</function>.
<informalexample>
<programlisting>
<![CDATA[
struct snd_card *card;
struct mychip *chip;
err = snd_card_create(index[dev], id[dev], THIS_MODULE, 0, &card);
.....
chip = kzalloc(sizeof(*chip), GFP_KERNEL);
]]>
</programlisting>
</informalexample>
</para>
<para>
The chip record should have the field to hold the card
pointer at least,
<informalexample>
<programlisting>
<![CDATA[
struct mychip {
struct snd_card *card;
....
};
]]>
</programlisting>
</informalexample>
</para>
<para>
Then, set the card pointer in the returned chip instance.
<informalexample>
<programlisting>
<![CDATA[
chip->card = card;
]]>
</programlisting>
</informalexample>
</para>
<para>
Next, initialize the fields, and register this chip
record as a low-level device with a specified
<parameter>ops</parameter>,
<informalexample>
<programlisting>
<![CDATA[
static struct snd_device_ops ops = {
.dev_free = snd_mychip_dev_free,
};
....
snd_device_new(card, SNDRV_DEV_LOWLEVEL, chip, &ops);
]]>
</programlisting>
</informalexample>
<function>snd_mychip_dev_free()</function> is the
device-destructor function, which will call the real
destructor.
</para>
<para>
<informalexample>
<programlisting>
<![CDATA[
static int snd_mychip_dev_free(struct snd_device *device)
{
return snd_mychip_free(device->device_data);
}
]]>
</programlisting>
</informalexample>
where <function>snd_mychip_free()</function> is the real destructor.
</para>
</section>
</section>
<section id="card-management-registration">
<title>Registration and Release</title>
<para>
After all components are assigned, register the card instance
by calling <function>snd_card_register()</function>. Access
to the device files is enabled at this point. That is, before
<function>snd_card_register()</function> is called, the
components are safely inaccessible from external side. If this
call fails, exit the probe function after releasing the card via
<function>snd_card_free()</function>.
</para>
<para>
For releasing the card instance, you can call simply
<function>snd_card_free()</function>. As mentioned earlier, all
components are released automatically by this call.
</para>
<para>
As further notes, the destructors (both
<function>snd_mychip_dev_free</function> and
<function>snd_mychip_free</function>) cannot be defined with
the <parameter>__devexit</parameter> prefix, because they may be
called from the constructor, too, at the false path.
</para>
<para>
For a device which allows hotplugging, you can use
<function>snd_card_free_when_closed</function>. This one will
postpone the destruction until all devices are closed.
</para>
</section>
</chapter>
<!-- ****************************************************** -->
<!-- PCI Resource Management -->
<!-- ****************************************************** -->
<chapter id="pci-resource">
<title>PCI Resource Management</title>
<section id="pci-resource-example">
<title>Full Code Example</title>
<para>
In this section, we'll complete the chip-specific constructor,
destructor and PCI entries. Example code is shown first,
below.
<example>
<title>PCI Resource Management Example</title>
<programlisting>
<![CDATA[
struct mychip {
struct snd_card *card;
struct pci_dev *pci;
unsigned long port;
int irq;
};
static int snd_mychip_free(struct mychip *chip)
{
/* disable hardware here if any */
.... /* (not implemented in this document) */
/* release the irq */
if (chip->irq >= 0)
free_irq(chip->irq, chip);
/* release the I/O ports & memory */
pci_release_regions(chip->pci);
/* disable the PCI entry */
pci_disable_device(chip->pci);
/* release the data */
kfree(chip);
return 0;
}
/* chip-specific constructor */
static int __devinit snd_mychip_create(struct snd_card *card,
struct pci_dev *pci,
struct mychip **rchip)
{
struct mychip *chip;
int err;
static struct snd_device_ops ops = {
.dev_free = snd_mychip_dev_free,
};
*rchip = NULL;
/* initialize the PCI entry */
err = pci_enable_device(pci);
if (err < 0)
return err;
/* check PCI availability (28bit DMA) */
if (pci_set_dma_mask(pci, DMA_BIT_MASK(28)) < 0 ||
pci_set_consistent_dma_mask(pci, DMA_BIT_MASK(28)) < 0) {
printk(KERN_ERR "error to set 28bit mask DMA\n");
pci_disable_device(pci);
return -ENXIO;
}
chip = kzalloc(sizeof(*chip), GFP_KERNEL);
if (chip == NULL) {
pci_disable_device(pci);
return -ENOMEM;
}
/* initialize the stuff */
chip->card = card;
chip->pci = pci;
chip->irq = -1;
/* (1) PCI resource allocation */
err = pci_request_regions(pci, "My Chip");
if (err < 0) {
kfree(chip);
pci_disable_device(pci);
return err;
}
chip->port = pci_resource_start(pci, 0);
if (request_irq(pci->irq, snd_mychip_interrupt,
IRQF_SHARED, "My Chip", chip)) {
printk(KERN_ERR "cannot grab irq %d\n", pci->irq);
snd_mychip_free(chip);
return -EBUSY;
}
chip->irq = pci->irq;
/* (2) initialization of the chip hardware */
.... /* (not implemented in this document) */
err = snd_device_new(card, SNDRV_DEV_LOWLEVEL, chip, &ops);
if (err < 0) {
snd_mychip_free(chip);
return err;
}
snd_card_set_dev(card, &pci->dev);
*rchip = chip;
return 0;
}
/* PCI IDs */
static struct pci_device_id snd_mychip_ids[] = {
{ PCI_VENDOR_ID_FOO, PCI_DEVICE_ID_BAR,
PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, },
....
{ 0, }
};
MODULE_DEVICE_TABLE(pci, snd_mychip_ids);
/* pci_driver definition */
static struct pci_driver driver = {
.name = "My Own Chip",
.id_table = snd_mychip_ids,
.probe = snd_mychip_probe,
.remove = __devexit_p(snd_mychip_remove),
};
/* module initialization */
static int __init alsa_card_mychip_init(void)
{
return pci_register_driver(&driver);
}
/* module clean up */
static void __exit alsa_card_mychip_exit(void)
{
pci_unregister_driver(&driver);
}
module_init(alsa_card_mychip_init)
module_exit(alsa_card_mychip_exit)
EXPORT_NO_SYMBOLS; /* for old kernels only */
]]>
</programlisting>
</example>
</para>
</section>
<section id="pci-resource-some-haftas">
<title>Some Hafta's</title>
<para>
The allocation of PCI resources is done in the
<function>probe()</function> function, and usually an extra
<function>xxx_create()</function> function is written for this
purpose.
</para>
<para>
In the case of PCI devices, you first have to call
the <function>pci_enable_device()</function> function before
allocating resources. Also, you need to set the proper PCI DMA
mask to limit the accessed I/O range. In some cases, you might
need to call <function>pci_set_master()</function> function,
too.
</para>
<para>
Suppose the 28bit mask, and the code to be added would be like:
<informalexample>
<programlisting>
<![CDATA[
err = pci_enable_device(pci);
if (err < 0)
return err;
if (pci_set_dma_mask(pci, DMA_BIT_MASK(28)) < 0 ||
pci_set_consistent_dma_mask(pci, DMA_BIT_MASK(28)) < 0) {
printk(KERN_ERR "error to set 28bit mask DMA\n");
pci_disable_device(pci);
return -ENXIO;
}
]]>
</programlisting>
</informalexample>
</para>
</section>
<section id="pci-resource-resource-allocation">
<title>Resource Allocation</title>
<para>
The allocation of I/O ports and irqs is done via standard kernel
functions. Unlike ALSA ver.0.5.x., there are no helpers for
that. And these resources must be released in the destructor
function (see below). Also, on ALSA 0.9.x, you don't need to
allocate (pseudo-)DMA for PCI like in ALSA 0.5.x.
</para>
<para>
Now assume that the PCI device has an I/O port with 8 bytes
and an interrupt. Then struct <structname>mychip</structname> will have the
following fields:
<informalexample>
<programlisting>
<![CDATA[
struct mychip {
struct snd_card *card;
unsigned long port;
int irq;
};
]]>
</programlisting>
</informalexample>
</para>
<para>
For an I/O port (and also a memory region), you need to have
the resource pointer for the standard resource management. For
an irq, you have to keep only the irq number (integer). But you
need to initialize this number as -1 before actual allocation,
since irq 0 is valid. The port address and its resource pointer
can be initialized as null by
<function>kzalloc()</function> automatically, so you
don't have to take care of resetting them.
</para>
<para>
The allocation of an I/O port is done like this:
<informalexample>
<programlisting>
<![CDATA[
err = pci_request_regions(pci, "My Chip");
if (err < 0) {
kfree(chip);
pci_disable_device(pci);
return err;
}
chip->port = pci_resource_start(pci, 0);
]]>
</programlisting>
</informalexample>
</para>
<para>
<!-- obsolete -->
It will reserve the I/O port region of 8 bytes of the given
PCI device. The returned value, chip-&gt;res_port, is allocated
via <function>kmalloc()</function> by
<function>request_region()</function>. The pointer must be
released via <function>kfree()</function>, but there is a
problem with this. This issue will be explained later.
</para>
<para>
The allocation of an interrupt source is done like this:
<informalexample>
<programlisting>
<![CDATA[
if (request_irq(pci->irq, snd_mychip_interrupt,
IRQF_SHARED, "My Chip", chip)) {
printk(KERN_ERR "cannot grab irq %d\n", pci->irq);
snd_mychip_free(chip);
return -EBUSY;
}
chip->irq = pci->irq;
]]>
</programlisting>
</informalexample>
where <function>snd_mychip_interrupt()</function> is the
interrupt handler defined <link
linkend="pcm-interface-interrupt-handler"><citetitle>later</citetitle></link>.
Note that chip-&gt;irq should be defined
only when <function>request_irq()</function> succeeded.
</para>
<para>
On the PCI bus, interrupts can be shared. Thus,
<constant>IRQF_SHARED</constant> is used as the interrupt flag of
<function>request_irq()</function>.
</para>
<para>
The last argument of <function>request_irq()</function> is the
data pointer passed to the interrupt handler. Usually, the
chip-specific record is used for that, but you can use what you
like, too.
</para>
<para>
I won't give details about the interrupt handler at this
point, but at least its appearance can be explained now. The
interrupt handler looks usually like the following:
<informalexample>
<programlisting>
<![CDATA[
static irqreturn_t snd_mychip_interrupt(int irq, void *dev_id)
{
struct mychip *chip = dev_id;
....
return IRQ_HANDLED;
}
]]>
</programlisting>
</informalexample>
</para>
<para>
Now let's write the corresponding destructor for the resources
above. The role of destructor is simple: disable the hardware
(if already activated) and release the resources. So far, we
have no hardware part, so the disabling code is not written here.
</para>
<para>
To release the resources, the <quote>check-and-release</quote>
method is a safer way. For the interrupt, do like this:
<informalexample>
<programlisting>
<![CDATA[
if (chip->irq >= 0)
free_irq(chip->irq, chip);
]]>
</programlisting>
</informalexample>
Since the irq number can start from 0, you should initialize
chip-&gt;irq with a negative value (e.g. -1), so that you can
check the validity of the irq number as above.
</para>
<para>
When you requested I/O ports or memory regions via
<function>pci_request_region()</function> or
<function>pci_request_regions()</function> like in this example,
release the resource(s) using the corresponding function,
<function>pci_release_region()</function> or
<function>pci_release_regions()</function>.
<informalexample>
<programlisting>
<![CDATA[
pci_release_regions(chip->pci);
]]>
</programlisting>
</informalexample>
</para>
<para>
When you requested manually via <function>request_region()</function>
or <function>request_mem_region</function>, you can release it via
<function>release_resource()</function>. Suppose that you keep
the resource pointer returned from <function>request_region()</function>
in chip-&gt;res_port, the release procedure looks like:
<informalexample>
<programlisting>
<![CDATA[
release_and_free_resource(chip->res_port);
]]>
</programlisting>
</informalexample>
</para>
<para>
Don't forget to call <function>pci_disable_device()</function>
before the end.
</para>
<para>
And finally, release the chip-specific record.
<informalexample>
<programlisting>
<![CDATA[
kfree(chip);
]]>
</programlisting>
</informalexample>
</para>
<para>
Again, remember that you cannot
use the <parameter>__devexit</parameter> prefix for this destructor.
</para>
<para>
We didn't implement the hardware disabling part in the above.
If you need to do this, please note that the destructor may be
called even before the initialization of the chip is completed.
It would be better to have a flag to skip hardware disabling
if the hardware was not initialized yet.
</para>
<para>
When the chip-data is assigned to the card using
<function>snd_device_new()</function> with
<constant>SNDRV_DEV_LOWLELVEL</constant> , its destructor is
called at the last. That is, it is assured that all other
components like PCMs and controls have already been released.
You don't have to stop PCMs, etc. explicitly, but just
call low-level hardware stopping.
</para>
<para>
The management of a memory-mapped region is almost as same as
the management of an I/O port. You'll need three fields like
the following:
<informalexample>
<programlisting>
<![CDATA[
struct mychip {
....
unsigned long iobase_phys;
void __iomem *iobase_virt;
};
]]>
</programlisting>
</informalexample>
and the allocation would be like below:
<informalexample>
<programlisting>
<![CDATA[
if ((err = pci_request_regions(pci, "My Chip")) < 0) {
kfree(chip);
return err;
}
chip->iobase_phys = pci_resource_start(pci, 0);
chip->iobase_virt = ioremap_nocache(chip->iobase_phys,
pci_resource_len(pci, 0));
]]>
</programlisting>
</informalexample>
and the corresponding destructor would be:
<informalexample>
<programlisting>
<![CDATA[
static int snd_mychip_free(struct mychip *chip)
{
....
if (chip->iobase_virt)
iounmap(chip->iobase_virt);
....
pci_release_regions(chip->pci);
....
}
]]>
</programlisting>
</informalexample>
</para>
</section>
<section id="pci-resource-device-struct">
<title>Registration of Device Struct</title>
<para>
At some point, typically after calling <function>snd_device_new()</function>,
you need to register the struct <structname>device</structname> of the chip
you're handling for udev and co. ALSA provides a macro for compatibility with
older kernels. Simply call like the following:
<informalexample>
<programlisting>
<![CDATA[
snd_card_set_dev(card, &pci->dev);
]]>
</programlisting>
</informalexample>
so that it stores the PCI's device pointer to the card. This will be
referred by ALSA core functions later when the devices are registered.
</para>
<para>
In the case of non-PCI, pass the proper device struct pointer of the BUS
instead. (In the case of legacy ISA without PnP, you don't have to do
anything.)
</para>
</section>
<section id="pci-resource-entries">
<title>PCI Entries</title>
<para>
So far, so good. Let's finish the missing PCI
stuff. At first, we need a
<structname>pci_device_id</structname> table for this
chipset. It's a table of PCI vendor/device ID number, and some
masks.
</para>
<para>
For example,
<informalexample>
<programlisting>
<![CDATA[
static struct pci_device_id snd_mychip_ids[] = {
{ PCI_VENDOR_ID_FOO, PCI_DEVICE_ID_BAR,
PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, },
....
{ 0, }
};
MODULE_DEVICE_TABLE(pci, snd_mychip_ids);
]]>
</programlisting>
</informalexample>
</para>
<para>
The first and second fields of
the <structname>pci_device_id</structname> structure are the vendor and
device IDs. If you have no reason to filter the matching
devices, you can leave the remaining fields as above. The last
field of the <structname>pci_device_id</structname> struct contains
private data for this entry. You can specify any value here, for
example, to define specific operations for supported device IDs.
Such an example is found in the intel8x0 driver.
</para>
<para>
The last entry of this list is the terminator. You must
specify this all-zero entry.
</para>
<para>
Then, prepare the <structname>pci_driver</structname> record:
<informalexample>
<programlisting>
<![CDATA[
static struct pci_driver driver = {
.name = "My Own Chip",
.id_table = snd_mychip_ids,
.probe = snd_mychip_probe,
.remove = __devexit_p(snd_mychip_remove),
};
]]>
</programlisting>
</informalexample>
</para>
<para>
The <structfield>probe</structfield> and
<structfield>remove</structfield> functions have already
been defined in the previous sections.
The <structfield>remove</structfield> function should
be defined with the
<function>__devexit_p()</function> macro, so that it's not
defined for built-in (and non-hot-pluggable) case. The
<structfield>name</structfield>
field is the name string of this device. Note that you must not
use a slash <quote>/</quote> in this string.
</para>
<para>
And at last, the module entries:
<informalexample>
<programlisting>
<![CDATA[
static int __init alsa_card_mychip_init(void)
{
return pci_register_driver(&driver);
}
static void __exit alsa_card_mychip_exit(void)
{
pci_unregister_driver(&driver);
}
module_init(alsa_card_mychip_init)
module_exit(alsa_card_mychip_exit)
]]>
</programlisting>
</informalexample>
</para>
<para>
Note that these module entries are tagged with
<parameter>__init</parameter> and
<parameter>__exit</parameter> prefixes, not
<parameter>__devinit</parameter> nor
<parameter>__devexit</parameter>.
</para>
<para>
Oh, one thing was forgotten. If you have no exported symbols,
you need to declare it in 2.2 or 2.4 kernels (it's not necessary in 2.6 kernels).
<informalexample>
<programlisting>
<![CDATA[
EXPORT_NO_SYMBOLS;
]]>
</programlisting>
</informalexample>
That's all!
</para>
</section>
</chapter>
<!-- ****************************************************** -->
<!-- PCM Interface -->
<!-- ****************************************************** -->
<chapter id="pcm-interface">
<title>PCM Interface</title>
<section id="pcm-interface-general">
<title>General</title>
<para>
The PCM middle layer of ALSA is quite powerful and it is only
necessary for each driver to implement the low-level functions
to access its hardware.
</para>
<para>
For accessing to the PCM layer, you need to include
<filename>&lt;sound/pcm.h&gt;</filename> first. In addition,
<filename>&lt;sound/pcm_params.h&gt;</filename> might be needed
if you access to some functions related with hw_param.
</para>
<para>
Each card device can have up to four pcm instances. A pcm
instance corresponds to a pcm device file. The limitation of
number of instances comes only from the available bit size of
the Linux's device numbers. Once when 64bit device number is
used, we'll have more pcm instances available.
</para>
<para>
A pcm instance consists of pcm playback and capture streams,
and each pcm stream consists of one or more pcm substreams. Some
soundcards support multiple playback functions. For example,
emu10k1 has a PCM playback of 32 stereo substreams. In this case, at
each open, a free substream is (usually) automatically chosen
and opened. Meanwhile, when only one substream exists and it was
already opened, the successful open will either block
or error with <constant>EAGAIN</constant> according to the
file open mode. But you don't have to care about such details in your
driver. The PCM middle layer will take care of such work.
</para>
</section>
<section id="pcm-interface-example">
<title>Full Code Example</title>
<para>
The example code below does not include any hardware access
routines but shows only the skeleton, how to build up the PCM
interfaces.
<example>
<title>PCM Example Code</title>
<programlisting>
<![CDATA[
#include <sound/pcm.h>
....
/* hardware definition */
static struct snd_pcm_hardware snd_mychip_playback_hw = {
.info = (SNDRV_PCM_INFO_MMAP |
SNDRV_PCM_INFO_INTERLEAVED |
SNDRV_PCM_INFO_BLOCK_TRANSFER |
SNDRV_PCM_INFO_MMAP_VALID),
.formats = SNDRV_PCM_FMTBIT_S16_LE,
.rates = SNDRV_PCM_RATE_8000_48000,
.rate_min = 8000,
.rate_max = 48000,
.channels_min = 2,
.channels_max = 2,
.buffer_bytes_max = 32768,
.period_bytes_min = 4096,
.period_bytes_max = 32768,
.periods_min = 1,
.periods_max = 1024,
};
/* hardware definition */
static struct snd_pcm_hardware snd_mychip_capture_hw = {
.info = (SNDRV_PCM_INFO_MMAP |
SNDRV_PCM_INFO_INTERLEAVED |
SNDRV_PCM_INFO_BLOCK_TRANSFER |
SNDRV_PCM_INFO_MMAP_VALID),
.formats = SNDRV_PCM_FMTBIT_S16_LE,
.rates = SNDRV_PCM_RATE_8000_48000,
.rate_min = 8000,
.rate_max = 48000,
.channels_min = 2,
.channels_max = 2,
.buffer_bytes_max = 32768,
.period_bytes_min = 4096,
.period_bytes_max = 32768,
.periods_min = 1,
.periods_max = 1024,
};
/* open callback */
static int snd_mychip_playback_open(struct snd_pcm_substream *substream)
{
struct mychip *chip = snd_pcm_substream_chip(substream);
struct snd_pcm_runtime *runtime = substream->runtime;
runtime->hw = snd_mychip_playback_hw;
/* more hardware-initialization will be done here */
....
return 0;
}
/* close callback */
static int snd_mychip_playback_close(struct snd_pcm_substream *substream)
{
struct mychip *chip = snd_pcm_substream_chip(substream);
/* the hardware-specific codes will be here */
....
return 0;
}
/* open callback */
static int snd_mychip_capture_open(struct snd_pcm_substream *substream)
{
struct mychip *chip = snd_pcm_substream_chip(substream);
struct snd_pcm_runtime *runtime = substream->runtime;
runtime->hw = snd_mychip_capture_hw;
/* more hardware-initialization will be done here */
....
return 0;
}
/* close callback */
static int snd_mychip_capture_close(struct snd_pcm_substream *substream)
{
struct mychip *chip = snd_pcm_substream_chip(substream);
/* the hardware-specific codes will be here */
....
return 0;
}
/* hw_params callback */
static int snd_mychip_pcm_hw_params(struct snd_pcm_substream *substream,
struct snd_pcm_hw_params *hw_params)
{
return snd_pcm_lib_malloc_pages(substream,
params_buffer_bytes(hw_params));
}
/* hw_free callback */
static int snd_mychip_pcm_hw_free(struct snd_pcm_substream *substream)
{
return snd_pcm_lib_free_pages(substream);
}
/* prepare callback */
static int snd_mychip_pcm_prepare(struct snd_pcm_substream *substream)
{
struct mychip *chip = snd_pcm_substream_chip(substream);
struct snd_pcm_runtime *runtime = substream->runtime;
/* set up the hardware with the current configuration
* for example...
*/
mychip_set_sample_format(chip, runtime->format);
mychip_set_sample_rate(chip, runtime->rate);
mychip_set_channels(chip, runtime->channels);
mychip_set_dma_setup(chip, runtime->dma_addr,
chip->buffer_size,
chip->period_size);
return 0;
}
/* trigger callback */
static int snd_mychip_pcm_trigger(struct snd_pcm_substream *substream,
int cmd)
{
switch (cmd) {
case SNDRV_PCM_TRIGGER_START:
/* do something to start the PCM engine */
....
break;
case SNDRV_PCM_TRIGGER_STOP:
/* do something to stop the PCM engine */
....
break;
default:
return -EINVAL;
}
}
/* pointer callback */
static snd_pcm_uframes_t
snd_mychip_pcm_pointer(struct snd_pcm_substream *substream)
{
struct mychip *chip = snd_pcm_substream_chip(substream);
unsigned int current_ptr;
/* get the current hardware pointer */
current_ptr = mychip_get_hw_pointer(chip);
return current_ptr;
}
/* operators */
static struct snd_pcm_ops snd_mychip_playback_ops = {
.open = snd_mychip_playback_open,
.close = snd_mychip_playback_close,
.ioctl = snd_pcm_lib_ioctl,
.hw_params = snd_mychip_pcm_hw_params,
.hw_free = snd_mychip_pcm_hw_free,
.prepare = snd_mychip_pcm_prepare,
.trigger = snd_mychip_pcm_trigger,
.pointer = snd_mychip_pcm_pointer,
};
/* operators */
static struct snd_pcm_ops snd_mychip_capture_ops = {
.open = snd_mychip_capture_open,
.close = snd_mychip_capture_close,
.ioctl = snd_pcm_lib_ioctl,
.hw_params = snd_mychip_pcm_hw_params,
.hw_free = snd_mychip_pcm_hw_free,
.prepare = snd_mychip_pcm_prepare,
.trigger = snd_mychip_pcm_trigger,
.pointer = snd_mychip_pcm_pointer,
};
/*
* definitions of capture are omitted here...
*/
/* create a pcm device */
static int __devinit snd_mychip_new_pcm(struct mychip *chip)
{
struct snd_pcm *pcm;
int err;
err = snd_pcm_new(chip->card, "My Chip", 0, 1, 1, &pcm);
if (err < 0)
return err;
pcm->private_data = chip;
strcpy(pcm->name, "My Chip");
chip->pcm = pcm;
/* set operators */
snd_pcm_set_ops(pcm, SNDRV_PCM_STREAM_PLAYBACK,
&snd_mychip_playback_ops);
snd_pcm_set_ops(pcm, SNDRV_PCM_STREAM_CAPTURE,
&snd_mychip_capture_ops);
/* pre-allocation of buffers */
/* NOTE: this may fail */
snd_pcm_lib_preallocate_pages_for_all(pcm, SNDRV_DMA_TYPE_DEV,
snd_dma_pci_data(chip->pci),
64*1024, 64*1024);
return 0;
}
]]>
</programlisting>
</example>
</para>
</section>
<section id="pcm-interface-constructor">
<title>Constructor</title>
<para>
A pcm instance is allocated by the <function>snd_pcm_new()</function>
function. It would be better to create a constructor for pcm,
namely,
<informalexample>
<programlisting>
<![CDATA[
static int __devinit snd_mychip_new_pcm(struct mychip *chip)
{
struct snd_pcm *pcm;
int err;
err = snd_pcm_new(chip->card, "My Chip", 0, 1, 1, &pcm);
if (err < 0)
return err;
pcm->private_data = chip;
strcpy(pcm->name, "My Chip");
chip->pcm = pcm;
....
return 0;
}
]]>
</programlisting>
</informalexample>
</para>
<para>
The <function>snd_pcm_new()</function> function takes four
arguments. The first argument is the card pointer to which this
pcm is assigned, and the second is the ID string.
</para>
<para>
The third argument (<parameter>index</parameter>, 0 in the
above) is the index of this new pcm. It begins from zero. If
you create more than one pcm instances, specify the
different numbers in this argument. For example,
<parameter>index</parameter> = 1 for the second PCM device.
</para>
<para>
The fourth and fifth arguments are the number of substreams
for playback and capture, respectively. Here 1 is used for
both arguments. When no playback or capture substreams are available,
pass 0 to the corresponding argument.
</para>
<para>
If a chip supports multiple playbacks or captures, you can
specify more numbers, but they must be handled properly in
open/close, etc. callbacks. When you need to know which
substream you are referring to, then it can be obtained from
struct <structname>snd_pcm_substream</structname> data passed to each callback
as follows:
<informalexample>
<programlisting>
<![CDATA[
struct snd_pcm_substream *substream;
int index = substream->number;
]]>
</programlisting>
</informalexample>
</para>
<para>
After the pcm is created, you need to set operators for each
pcm stream.
<informalexample>
<programlisting>
<![CDATA[
snd_pcm_set_ops(pcm, SNDRV_PCM_STREAM_PLAYBACK,
&snd_mychip_playback_ops);
snd_pcm_set_ops(pcm, SNDRV_PCM_STREAM_CAPTURE,
&snd_mychip_capture_ops);
]]>
</programlisting>
</informalexample>
</para>
<para>
The operators are defined typically like this:
<informalexample>
<programlisting>
<![CDATA[
static struct snd_pcm_ops snd_mychip_playback_ops = {
.open = snd_mychip_pcm_open,
.close = snd_mychip_pcm_close,
.ioctl = snd_pcm_lib_ioctl,
.hw_params = snd_mychip_pcm_hw_params,
.hw_free = snd_mychip_pcm_hw_free,
.prepare = snd_mychip_pcm_prepare,
.trigger = snd_mychip_pcm_trigger,
.pointer = snd_mychip_pcm_pointer,
};
]]>
</programlisting>
</informalexample>
All the callbacks are described in the
<link linkend="pcm-interface-operators"><citetitle>
Operators</citetitle></link> subsection.
</para>
<para>
After setting the operators, you probably will want to
pre-allocate the buffer. For the pre-allocation, simply call
the following:
<informalexample>
<programlisting>
<![CDATA[
snd_pcm_lib_preallocate_pages_for_all(pcm, SNDRV_DMA_TYPE_DEV,
snd_dma_pci_data(chip->pci),
64*1024, 64*1024);
]]>
</programlisting>
</informalexample>
It will allocate a buffer up to 64kB as default.
Buffer management details will be described in the later section <link
linkend="buffer-and-memory"><citetitle>Buffer and Memory
Management</citetitle></link>.
</para>
<para>
Additionally, you can set some extra information for this pcm
in pcm-&gt;info_flags.
The available values are defined as
<constant>SNDRV_PCM_INFO_XXX</constant> in
<filename>&lt;sound/asound.h&gt;</filename>, which is used for
the hardware definition (described later). When your soundchip
supports only half-duplex, specify like this:
<informalexample>
<programlisting>
<![CDATA[
pcm->info_flags = SNDRV_PCM_INFO_HALF_DUPLEX;
]]>
</programlisting>
</informalexample>
</para>
</section>
<section id="pcm-interface-destructor">
<title>... And the Destructor?</title>
<para>
The destructor for a pcm instance is not always
necessary. Since the pcm device will be released by the middle
layer code automatically, you don't have to call the destructor
explicitly.
</para>
<para>
The destructor would be necessary if you created
special records internally and needed to release them. In such a
case, set the destructor function to
pcm-&gt;private_free:
<example>
<title>PCM Instance with a Destructor</title>
<programlisting>
<![CDATA[
static void mychip_pcm_free(struct snd_pcm *pcm)
{
struct mychip *chip = snd_pcm_chip(pcm);
/* free your own data */
kfree(chip->my_private_pcm_data);
/* do what you like else */
....
}
static int __devinit snd_mychip_new_pcm(struct mychip *chip)
{
struct snd_pcm *pcm;
....
/* allocate your own data */
chip->my_private_pcm_data = kmalloc(...);
/* set the destructor */
pcm->private_data = chip;
pcm->private_free = mychip_pcm_free;
....
}
]]>
</programlisting>
</example>
</para>
</section>
<section id="pcm-interface-runtime">
<title>Runtime Pointer - The Chest of PCM Information</title>
<para>
When the PCM substream is opened, a PCM runtime instance is
allocated and assigned to the substream. This pointer is
accessible via <constant>substream-&gt;runtime</constant>.
This runtime pointer holds most information you need
to control the PCM: the copy of hw_params and sw_params configurations, the buffer
pointers, mmap records, spinlocks, etc.
</para>
<para>
The definition of runtime instance is found in
<filename>&lt;sound/pcm.h&gt;</filename>. Here are
the contents of this file:
<informalexample>
<programlisting>
<![CDATA[
struct _snd_pcm_runtime {
/* -- Status -- */
struct snd_pcm_substream *trigger_master;
snd_timestamp_t trigger_tstamp; /* trigger timestamp */
int overrange;
snd_pcm_uframes_t avail_max;
snd_pcm_uframes_t hw_ptr_base; /* Position at buffer restart */
snd_pcm_uframes_t hw_ptr_interrupt; /* Position at interrupt time*/
/* -- HW params -- */
snd_pcm_access_t access; /* access mode */
snd_pcm_format_t format; /* SNDRV_PCM_FORMAT_* */
snd_pcm_subformat_t subformat; /* subformat */
unsigned int rate; /* rate in Hz */
unsigned int channels; /* channels */
snd_pcm_uframes_t period_size; /* period size */
unsigned int periods; /* periods */
snd_pcm_uframes_t buffer_size; /* buffer size */
unsigned int tick_time; /* tick time */
snd_pcm_uframes_t min_align; /* Min alignment for the format */
size_t byte_align;
unsigned int frame_bits;
unsigned int sample_bits;
unsigned int info;
unsigned int rate_num;
unsigned int rate_den;
/* -- SW params -- */
struct timespec tstamp_mode; /* mmap timestamp is updated */
unsigned int period_step;
unsigned int sleep_min; /* min ticks to sleep */
snd_pcm_uframes_t start_threshold;
snd_pcm_uframes_t stop_threshold;
snd_pcm_uframes_t silence_threshold; /* Silence filling happens when
noise is nearest than this */
snd_pcm_uframes_t silence_size; /* Silence filling size */
snd_pcm_uframes_t boundary; /* pointers wrap point */
snd_pcm_uframes_t silenced_start;
snd_pcm_uframes_t silenced_size;
snd_pcm_sync_id_t sync; /* hardware synchronization ID */
/* -- mmap -- */
volatile struct snd_pcm_mmap_status *status;
volatile struct snd_pcm_mmap_control *control;
atomic_t mmap_count;
/* -- locking / scheduling -- */
spinlock_t lock;
wait_queue_head_t sleep;
struct timer_list tick_timer;
struct fasync_struct *fasync;
/* -- private section -- */
void *private_data;
void (*private_free)(struct snd_pcm_runtime *runtime);
/* -- hardware description -- */
struct snd_pcm_hardware hw;
struct snd_pcm_hw_constraints hw_constraints;
/* -- interrupt callbacks -- */
void (*transfer_ack_begin)(struct snd_pcm_substream *substream);
void (*transfer_ack_end)(struct snd_pcm_substream *substream);
/* -- timer -- */
unsigned int timer_resolution; /* timer resolution */
/* -- DMA -- */
unsigned char *dma_area; /* DMA area */
dma_addr_t dma_addr; /* physical bus address (not accessible from main CPU) */
size_t dma_bytes; /* size of DMA area */
struct snd_dma_buffer *dma_buffer_p; /* allocated buffer */
#if defined(CONFIG_SND_PCM_OSS) || defined(CONFIG_SND_PCM_OSS_MODULE)
/* -- OSS things -- */
struct snd_pcm_oss_runtime oss;
#endif
};
]]>
</programlisting>
</informalexample>
</para>
<para>
For the operators (callbacks) of each sound driver, most of
these records are supposed to be read-only. Only the PCM
middle-layer changes / updates them. The exceptions are
the hardware description (hw), interrupt callbacks
(transfer_ack_xxx), DMA buffer information, and the private
data. Besides, if you use the standard buffer allocation
method via <function>snd_pcm_lib_malloc_pages()</function>,
you don't need to set the DMA buffer information by yourself.
</para>
<para>
In the sections below, important records are explained.
</para>
<section id="pcm-interface-runtime-hw">
<title>Hardware Description</title>
<para>
The hardware descriptor (struct <structname>snd_pcm_hardware</structname>)
contains the definitions of the fundamental hardware
configuration. Above all, you'll need to define this in
<link linkend="pcm-interface-operators-open-callback"><citetitle>
the open callback</citetitle></link>.
Note that the runtime instance holds the copy of the
descriptor, not the pointer to the existing descriptor. That
is, in the open callback, you can modify the copied descriptor
(<constant>runtime-&gt;hw</constant>) as you need. For example, if the maximum
number of channels is 1 only on some chip models, you can
still use the same hardware descriptor and change the
channels_max later:
<informalexample>
<programlisting>
<![CDATA[
struct snd_pcm_runtime *runtime = substream->runtime;
...
runtime->hw = snd_mychip_playback_hw; /* common definition */
if (chip->model == VERY_OLD_ONE)
runtime->hw.channels_max = 1;
]]>
</programlisting>
</informalexample>
</para>
<para>
Typically, you'll have a hardware descriptor as below:
<informalexample>
<programlisting>
<![CDATA[
static struct snd_pcm_hardware snd_mychip_playback_hw = {
.info = (SNDRV_PCM_INFO_MMAP |
SNDRV_PCM_INFO_INTERLEAVED |
SNDRV_PCM_INFO_BLOCK_TRANSFER |
SNDRV_PCM_INFO_MMAP_VALID),
.formats = SNDRV_PCM_FMTBIT_S16_LE,
.rates = SNDRV_PCM_RATE_8000_48000,
.rate_min = 8000,
.rate_max = 48000,
.channels_min = 2,
.channels_max = 2,
.buffer_bytes_max = 32768,
.period_bytes_min = 4096,
.period_bytes_max = 32768,
.periods_min = 1,
.periods_max = 1024,
};
]]>
</programlisting>
</informalexample>
</para>
<para>
<itemizedlist>
<listitem><para>
The <structfield>info</structfield> field contains the type and
capabilities of this pcm. The bit flags are defined in
<filename>&lt;sound/asound.h&gt;</filename> as
<constant>SNDRV_PCM_INFO_XXX</constant>. Here, at least, you
have to specify whether the mmap is supported and which
interleaved format is supported.
When the is supported, add the
<constant>SNDRV_PCM_INFO_MMAP</constant> flag here. When the
hardware supports the interleaved or the non-interleaved
formats, <constant>SNDRV_PCM_INFO_INTERLEAVED</constant> or
<constant>SNDRV_PCM_INFO_NONINTERLEAVED</constant> flag must
be set, respectively. If both are supported, you can set both,
too.
</para>
<para>
In the above example, <constant>MMAP_VALID</constant> and
<constant>BLOCK_TRANSFER</constant> are specified for the OSS mmap
mode. Usually both are set. Of course,
<constant>MMAP_VALID</constant> is set only if the mmap is
really supported.
</para>
<para>
The other possible flags are
<constant>SNDRV_PCM_INFO_PAUSE</constant> and
<constant>SNDRV_PCM_INFO_RESUME</constant>. The
<constant>PAUSE</constant> bit means that the pcm supports the
<quote>pause</quote> operation, while the
<constant>RESUME</constant> bit means that the pcm supports
the full <quote>suspend/resume</quote> operation.
If the <constant>PAUSE</constant> flag is set,
the <structfield>trigger</structfield> callback below
must handle the corresponding (pause push/release) commands.
The suspend/resume trigger commands can be defined even without
the <constant>RESUME</constant> flag. See <link
linkend="power-management"><citetitle>
Power Management</citetitle></link> section for details.
</para>
<para>
When the PCM substreams can be synchronized (typically,
synchronized start/stop of a playback and a capture streams),
you can give <constant>SNDRV_PCM_INFO_SYNC_START</constant>,
too. In this case, you'll need to check the linked-list of
PCM substreams in the trigger callback. This will be
described in the later section.
</para>
</listitem>
<listitem>
<para>
<structfield>formats</structfield> field contains the bit-flags
of supported formats (<constant>SNDRV_PCM_FMTBIT_XXX</constant>).
If the hardware supports more than one format, give all or'ed
bits. In the example above, the signed 16bit little-endian
format is specified.
</para>
</listitem>
<listitem>
<para>
<structfield>rates</structfield> field contains the bit-flags of
supported rates (<constant>SNDRV_PCM_RATE_XXX</constant>).
When the chip supports continuous rates, pass
<constant>CONTINUOUS</constant> bit additionally.
The pre-defined rate bits are provided only for typical
rates. If your chip supports unconventional rates, you need to add
the <constant>KNOT</constant> bit and set up the hardware
constraint manually (explained later).
</para>
</listitem>
<listitem>
<para>
<structfield>rate_min</structfield> and
<structfield>rate_max</structfield> define the minimum and
maximum sample rate. This should correspond somehow to
<structfield>rates</structfield> bits.
</para>
</listitem>
<listitem>
<para>
<structfield>channel_min</structfield> and
<structfield>channel_max</structfield>
define, as you might already expected, the minimum and maximum
number of channels.
</para>
</listitem>
<listitem>
<para>
<structfield>buffer_bytes_max</structfield> defines the
maximum buffer size in bytes. There is no
<structfield>buffer_bytes_min</structfield> field, since
it can be calculated from the minimum period size and the
minimum number of periods.
Meanwhile, <structfield>period_bytes_min</structfield> and
define the minimum and maximum size of the period in bytes.
<structfield>periods_max</structfield> and
<structfield>periods_min</structfield> define the maximum and
minimum number of periods in the buffer.
</para>
<para>
The <quote>period</quote> is a term that corresponds to
a fragment in the OSS world. The period defines the size at
which a PCM interrupt is generated. This size strongly
depends on the hardware.
Generally, the smaller period size will give you more
interrupts, that is, more controls.
In the case of capture, this size defines the input latency.
On the other hand, the whole buffer size defines the
output latency for the playback direction.
</para>
</listitem>
<listitem>
<para>
There is also a field <structfield>fifo_size</structfield>.
This specifies the size of the hardware FIFO, but currently it
is neither used in the driver nor in the alsa-lib. So, you
can ignore this field.
</para>
</listitem>
</itemizedlist>
</para>
</section>
<section id="pcm-interface-runtime-config">
<title>PCM Configurations</title>
<para>
Ok, let's go back again to the PCM runtime records.
The most frequently referred records in the runtime instance are
the PCM configurations.
The PCM configurations are stored in the runtime instance
after the application sends <type>hw_params</type> data via
alsa-lib. There are many fields copied from hw_params and
sw_params structs. For example,
<structfield>format</structfield> holds the format type
chosen by the application. This field contains the enum value
<constant>SNDRV_PCM_FORMAT_XXX</constant>.
</para>
<para>
One thing to be noted is that the configured buffer and period
sizes are stored in <quote>frames</quote> in the runtime.
In the ALSA world, 1 frame = channels * samples-size.
For conversion between frames and bytes, you can use the
<function>frames_to_bytes()</function> and
<function>bytes_to_frames()</function> helper functions.
<informalexample>
<programlisting>
<![CDATA[
period_bytes = frames_to_bytes(runtime, runtime->period_size);
]]>
</programlisting>
</informalexample>
</para>
<para>
Also, many software parameters (sw_params) are
stored in frames, too. Please check the type of the field.
<type>snd_pcm_uframes_t</type> is for the frames as unsigned
integer while <type>snd_pcm_sframes_t</type> is for the frames
as signed integer.
</para>
</section>
<section id="pcm-interface-runtime-dma">
<title>DMA Buffer Information</title>
<para>
The DMA buffer is defined by the following four fields,
<structfield>dma_area</structfield>,
<structfield>dma_addr</structfield>,
<structfield>dma_bytes</structfield> and
<structfield>dma_private</structfield>.
The <structfield>dma_area</structfield> holds the buffer
pointer (the logical address). You can call
<function>memcpy</function> from/to
this pointer. Meanwhile, <structfield>dma_addr</structfield>
holds the physical address of the buffer. This field is
specified only when the buffer is a linear buffer.
<structfield>dma_bytes</structfield> holds the size of buffer
in bytes. <structfield>dma_private</structfield> is used for
the ALSA DMA allocator.
</para>
<para>
If you use a standard ALSA function,
<function>snd_pcm_lib_malloc_pages()</function>, for
allocating the buffer, these fields are set by the ALSA middle
layer, and you should <emphasis>not</emphasis> change them by
yourself. You can read them but not write them.
On the other hand, if you want to allocate the buffer by
yourself, you'll need to manage it in hw_params callback.
At least, <structfield>dma_bytes</structfield> is mandatory.
<structfield>dma_area</structfield> is necessary when the
buffer is mmapped. If your driver doesn't support mmap, this
field is not necessary. <structfield>dma_addr</structfield>
is also optional. You can use
<structfield>dma_private</structfield> as you like, too.
</para>
</section>
<section id="pcm-interface-runtime-status">
<title>Running Status</title>
<para>
The running status can be referred via <constant>runtime-&gt;status</constant>.
This is the pointer to the struct <structname>snd_pcm_mmap_status</structname>
record. For example, you can get the current DMA hardware
pointer via <constant>runtime-&gt;status-&gt;hw_ptr</constant>.
</para>
<para>
The DMA application pointer can be referred via
<constant>runtime-&gt;control</constant>, which points to the
struct <structname>snd_pcm_mmap_control</structname> record.
However, accessing directly to this value is not recommended.
</para>
</section>
<section id="pcm-interface-runtime-private">
<title>Private Data</title>
<para>
You can allocate a record for the substream and store it in
<constant>runtime-&gt;private_data</constant>. Usually, this
is done in
<link linkend="pcm-interface-operators-open-callback"><citetitle>
the open callback</citetitle></link>.
Don't mix this with <constant>pcm-&gt;private_data</constant>.
The <constant>pcm-&gt;private_data</constant> usually points to the
chip instance assigned statically at the creation of PCM, while the
<constant>runtime-&gt;private_data</constant> points to a dynamic
data structure created at the PCM open callback.
<informalexample>
<programlisting>
<![CDATA[
static int snd_xxx_open(struct snd_pcm_substream *substream)
{
struct my_pcm_data *data;
....
data = kmalloc(sizeof(*data), GFP_KERNEL);
substream->runtime->private_data = data;
....
}
]]>
</programlisting>
</informalexample>
</para>
<para>
The allocated object must be released in
<link linkend="pcm-interface-operators-open-callback"><citetitle>
the close callback</citetitle></link>.
</para>
</section>
<section id="pcm-interface-runtime-intr">
<title>Interrupt Callbacks</title>
<para>
The field <structfield>transfer_ack_begin</structfield> and
<structfield>transfer_ack_end</structfield> are called at
the beginning and at the end of
<function>snd_pcm_period_elapsed()</function>, respectively.
</para>
</section>
</section>
<section id="pcm-interface-operators">
<title>Operators</title>
<para>
OK, now let me give details about each pcm callback
(<parameter>ops</parameter>). In general, every callback must
return 0 if successful, or a negative error number
such as <constant>-EINVAL</constant>. To choose an appropriate
error number, it is advised to check what value other parts of
the kernel return when the same kind of request fails.
</para>
<para>
The callback function takes at least the argument with
<structname>snd_pcm_substream</structname> pointer. To retrieve
the chip record from the given substream instance, you can use the
following macro.
<informalexample>
<programlisting>
<![CDATA[
int xxx() {
struct mychip *chip = snd_pcm_substream_chip(substream);
....
}
]]>
</programlisting>
</informalexample>
The macro reads <constant>substream-&gt;private_data</constant>,
which is a copy of <constant>pcm-&gt;private_data</constant>.
You can override the former if you need to assign different data
records per PCM substream. For example, the cmi8330 driver assigns
different private_data for playback and capture directions,
because it uses two different codecs (SB- and AD-compatible) for
different directions.
</para>
<section id="pcm-interface-operators-open-callback">
<title>open callback</title>
<para>
<informalexample>
<programlisting>
<![CDATA[
static int snd_xxx_open(struct snd_pcm_substream *substream);
]]>
</programlisting>
</informalexample>
This is called when a pcm substream is opened.
</para>
<para>
At least, here you have to initialize the runtime-&gt;hw
record. Typically, this is done by like this:
<informalexample>
<programlisting>
<![CDATA[
static int snd_xxx_open(struct snd_pcm_substream *substream)
{
struct mychip *chip = snd_pcm_substream_chip(substream);
struct snd_pcm_runtime *runtime = substream->runtime;
runtime->hw = snd_mychip_playback_hw;
return 0;
}
]]>
</programlisting>
</informalexample>
where <parameter>snd_mychip_playback_hw</parameter> is the
pre-defined hardware description.
</para>
<para>
You can allocate a private data in this callback, as described
in <link linkend="pcm-interface-runtime-private"><citetitle>
Private Data</citetitle></link> section.
</para>
<para>
If the hardware configuration needs more constraints, set the
hardware constraints here, too.
See <link linkend="pcm-interface-constraints"><citetitle>
Constraints</citetitle></link> for more details.
</para>
</section>
<section id="pcm-interface-operators-close-callback">
<title>close callback</title>
<para>
<informalexample>
<programlisting>
<![CDATA[
static int snd_xxx_close(struct snd_pcm_substream *substream);
]]>
</programlisting>
</informalexample>
Obviously, this is called when a pcm substream is closed.
</para>
<para>
Any private instance for a pcm substream allocated in the
open callback will be released here.
<informalexample>
<programlisting>
<![CDATA[
static int snd_xxx_close(struct snd_pcm_substream *substream)
{
....
kfree(substream->runtime->private_data);
....
}
]]>
</programlisting>
</informalexample>
</para>
</section>
<section id="pcm-interface-operators-ioctl-callback">
<title>ioctl callback</title>
<para>
This is used for any special call to pcm ioctls. But
usually you can pass a generic ioctl callback,
<function>snd_pcm_lib_ioctl</function>.
</para>
</section>
<section id="pcm-interface-operators-hw-params-callback">
<title>hw_params callback</title>
<para>
<informalexample>
<programlisting>
<![CDATA[
static int snd_xxx_hw_params(struct snd_pcm_substream *substream,
struct snd_pcm_hw_params *hw_params);
]]>
</programlisting>
</informalexample>
</para>
<para>
This is called when the hardware parameter
(<structfield>hw_params</structfield>) is set
up by the application,
that is, once when the buffer size, the period size, the
format, etc. are defined for the pcm substream.
</para>
<para>
Many hardware setups should be done in this callback,
including the allocation of buffers.
</para>
<para>
Parameters to be initialized are retrieved by
<function>params_xxx()</function> macros. To allocate
buffer, you can call a helper function,
<informalexample>
<programlisting>
<![CDATA[
snd_pcm_lib_malloc_pages(substream, params_buffer_bytes(hw_params));
]]>
</programlisting>
</informalexample>
<function>snd_pcm_lib_malloc_pages()</function> is available
only when the DMA buffers have been pre-allocated.
See the section <link
linkend="buffer-and-memory-buffer-types"><citetitle>
Buffer Types</citetitle></link> for more details.
</para>
<para>
Note that this and <structfield>prepare</structfield> callbacks
may be called multiple times per initialization.
For example, the OSS emulation may
call these callbacks at each change via its ioctl.
</para>
<para>
Thus, you need to be careful not to allocate the same buffers
many times, which will lead to memory leaks! Calling the
helper function above many times is OK. It will release the
previous buffer automatically when it was already allocated.
</para>
<para>
Another note is that this callback is non-atomic
(schedulable). This is important, because the
<structfield>trigger</structfield> callback
is atomic (non-schedulable). That is, mutexes or any
schedule-related functions are not available in
<structfield>trigger</structfield> callback.
Please see the subsection
<link linkend="pcm-interface-atomicity"><citetitle>
Atomicity</citetitle></link> for details.
</para>
</section>
<section id="pcm-interface-operators-hw-free-callback">
<title>hw_free callback</title>
<para>
<informalexample>
<programlisting>
<![CDATA[
static int snd_xxx_hw_free(struct snd_pcm_substream *substream);
]]>
</programlisting>
</informalexample>
</para>
<para>
This is called to release the resources allocated via
<structfield>hw_params</structfield>. For example, releasing the
buffer via
<function>snd_pcm_lib_malloc_pages()</function> is done by
calling the following:
<informalexample>
<programlisting>
<![CDATA[
snd_pcm_lib_free_pages(substream);
]]>
</programlisting>
</informalexample>
</para>
<para>
This function is always called before the close callback is called.
Also, the callback may be called multiple times, too.
Keep track whether the resource was already released.
</para>
</section>
<section id="pcm-interface-operators-prepare-callback">
<title>prepare callback</title>
<para>
<informalexample>
<programlisting>
<![CDATA[
static int snd_xxx_prepare(struct snd_pcm_substream *substream);
]]>
</programlisting>
</informalexample>
</para>
<para>
This callback is called when the pcm is
<quote>prepared</quote>. You can set the format type, sample
rate, etc. here. The difference from
<structfield>hw_params</structfield> is that the
<structfield>prepare</structfield> callback will be called each
time
<function>snd_pcm_prepare()</function> is called, i.e. when
recovering after underruns, etc.
</para>
<para>
Note that this callback is now non-atomic.
You can use schedule-related functions safely in this callback.
</para>
<para>
In this and the following callbacks, you can refer to the
values via the runtime record,
substream-&gt;runtime.
For example, to get the current
rate, format or channels, access to
runtime-&gt;rate,
runtime-&gt;format or
runtime-&gt;channels, respectively.
The physical address of the allocated buffer is set to
runtime-&gt;dma_area. The buffer and period sizes are
in runtime-&gt;buffer_size and runtime-&gt;period_size,
respectively.
</para>
<para>
Be careful that this callback will be called many times at
each setup, too.
</para>
</section>
<section id="pcm-interface-operators-trigger-callback">
<title>trigger callback</title>
<para>
<informalexample>
<programlisting>
<![CDATA[
static int snd_xxx_trigger(struct snd_pcm_substream *substream, int cmd);
]]>
</programlisting>
</informalexample>
This is called when the pcm is started, stopped or paused.
</para>
<para>
Which action is specified in the second argument,
<constant>SNDRV_PCM_TRIGGER_XXX</constant> in
<filename>&lt;sound/pcm.h&gt;</filename>. At least,
the <constant>START</constant> and <constant>STOP</constant>
commands must be defined in this callback.
<informalexample>
<programlisting>
<![CDATA[
switch (cmd) {
case SNDRV_PCM_TRIGGER_START:
/* do something to start the PCM engine */
break;
case SNDRV_PCM_TRIGGER_STOP:
/* do something to stop the PCM engine */
break;
default:
return -EINVAL;
}
]]>
</programlisting>
</informalexample>
</para>
<para>
When the pcm supports the pause operation (given in the info
field of the hardware table), the <constant>PAUSE_PUSE</constant>
and <constant>PAUSE_RELEASE</constant> commands must be
handled here, too. The former is the command to pause the pcm,
and the latter to restart the pcm again.
</para>
<para>
When the pcm supports the suspend/resume operation,
regardless of full or partial suspend/resume support,
the <constant>SUSPEND</constant> and <constant>RESUME</constant>
commands must be handled, too.
These commands are issued when the power-management status is
changed. Obviously, the <constant>SUSPEND</constant> and
<constant>RESUME</constant> commands
suspend and resume the pcm substream, and usually, they
are identical to the <constant>STOP</constant> and
<constant>START</constant> commands, respectively.
See the <link linkend="power-management"><citetitle>
Power Management</citetitle></link> section for details.
</para>
<para>
As mentioned, this callback is atomic. You cannot call
functions which may sleep.
The trigger callback should be as minimal as possible,
just really triggering the DMA. The other stuff should be
initialized hw_params and prepare callbacks properly
beforehand.
</para>
</section>
<section id="pcm-interface-operators-pointer-callback">
<title>pointer callback</title>
<para>
<informalexample>
<programlisting>
<![CDATA[
static snd_pcm_uframes_t snd_xxx_pointer(struct snd_pcm_substream *substream)
]]>
</programlisting>
</informalexample>
This callback is called when the PCM middle layer inquires
the current hardware position on the buffer. The position must
be returned in frames,
ranging from 0 to buffer_size - 1.
</para>
<para>
This is called usually from the buffer-update routine in the
pcm middle layer, which is invoked when
<function>snd_pcm_period_elapsed()</function> is called in the
interrupt routine. Then the pcm middle layer updates the
position and calculates the available space, and wakes up the
sleeping poll threads, etc.
</para>
<para>
This callback is also atomic.
</para>
</section>
<section id="pcm-interface-operators-copy-silence">
<title>copy and silence callbacks</title>
<para>
These callbacks are not mandatory, and can be omitted in
most cases. These callbacks are used when the hardware buffer
cannot be in the normal memory space. Some chips have their
own buffer on the hardware which is not mappable. In such a
case, you have to transfer the data manually from the memory
buffer to the hardware buffer. Or, if the buffer is
non-contiguous on both physical and virtual memory spaces,
these callbacks must be defined, too.
</para>
<para>
If these two callbacks are defined, copy and set-silence
operations are done by them. The detailed will be described in
the later section <link
linkend="buffer-and-memory"><citetitle>Buffer and Memory
Management</citetitle></link>.
</para>
</section>
<section id="pcm-interface-operators-ack">
<title>ack callback</title>
<para>
This callback is also not mandatory. This callback is called
when the appl_ptr is updated in read or write operations.
Some drivers like emu10k1-fx and cs46xx need to track the
current appl_ptr for the internal buffer, and this callback
is useful only for such a purpose.
</para>
<para>
This callback is atomic.
</para>
</section>
<section id="pcm-interface-operators-page-callback">
<title>page callback</title>
<para>
This callback is optional too. This callback is used
mainly for non-contiguous buffers. The mmap calls this
callback to get the page address. Some examples will be
explained in the later section <link
linkend="buffer-and-memory"><citetitle>Buffer and Memory
Management</citetitle></link>, too.
</para>
</section>
</section>
<section id="pcm-interface-interrupt-handler">
<title>Interrupt Handler</title>
<para>
The rest of pcm stuff is the PCM interrupt handler. The
role of PCM interrupt handler in the sound driver is to update
the buffer position and to tell the PCM middle layer when the
buffer position goes across the prescribed period size. To
inform this, call the <function>snd_pcm_period_elapsed()</function>
function.
</para>
<para>
There are several types of sound chips to generate the interrupts.
</para>
<section id="pcm-interface-interrupt-handler-boundary">
<title>Interrupts at the period (fragment) boundary</title>
<para>
This is the most frequently found type: the hardware
generates an interrupt at each period boundary.
In this case, you can call
<function>snd_pcm_period_elapsed()</function> at each
interrupt.
</para>
<para>
<function>snd_pcm_period_elapsed()</function> takes the
substream pointer as its argument. Thus, you need to keep the
substream pointer accessible from the chip instance. For
example, define substream field in the chip record to hold the
current running substream pointer, and set the pointer value
at open callback (and reset at close callback).
</para>
<para>
If you acquire a spinlock in the interrupt handler, and the
lock is used in other pcm callbacks, too, then you have to
release the lock before calling
<function>snd_pcm_period_elapsed()</function>, because
<function>snd_pcm_period_elapsed()</function> calls other pcm
callbacks inside.
</para>
<para>
Typical code would be like:
<example>
<title>Interrupt Handler Case #1</title>
<programlisting>
<![CDATA[
static irqreturn_t snd_mychip_interrupt(int irq, void *dev_id)
{
struct mychip *chip = dev_id;
spin_lock(&chip->lock);
....
if (pcm_irq_invoked(chip)) {
/* call updater, unlock before it */
spin_unlock(&chip->lock);
snd_pcm_period_elapsed(chip->substream);
spin_lock(&chip->lock);
/* acknowledge the interrupt if necessary */
}
....
spin_unlock(&chip->lock);
return IRQ_HANDLED;
}
]]>
</programlisting>
</example>
</para>
</section>
<section id="pcm-interface-interrupt-handler-timer">
<title>High frequency timer interrupts</title>
<para>
This happense when the hardware doesn't generate interrupts
at the period boundary but issues timer interrupts at a fixed
timer rate (e.g. es1968 or ymfpci drivers).
In this case, you need to check the current hardware
position and accumulate the processed sample length at each
interrupt. When the accumulated size exceeds the period
size, call
<function>snd_pcm_period_elapsed()</function> and reset the
accumulator.
</para>
<para>
Typical code would be like the following.
<example>
<title>Interrupt Handler Case #2</title>
<programlisting>
<![CDATA[
static irqreturn_t snd_mychip_interrupt(int irq, void *dev_id)
{
struct mychip *chip = dev_id;
spin_lock(&chip->lock);
....
if (pcm_irq_invoked(chip)) {
unsigned int last_ptr, size;
/* get the current hardware pointer (in frames) */
last_ptr = get_hw_ptr(chip);
/* calculate the processed frames since the
* last update
*/
if (last_ptr < chip->last_ptr)
size = runtime->buffer_size + last_ptr
- chip->last_ptr;
else
size = last_ptr - chip->last_ptr;
/* remember the last updated point */
chip->last_ptr = last_ptr;
/* accumulate the size */
chip->size += size;
/* over the period boundary? */
if (chip->size >= runtime->period_size) {
/* reset the accumulator */
chip->size %= runtime->period_size;
/* call updater */
spin_unlock(&chip->lock);
snd_pcm_period_elapsed(substream);
spin_lock(&chip->lock);
}
/* acknowledge the interrupt if necessary */
}
....
spin_unlock(&chip->lock);
return IRQ_HANDLED;
}
]]>
</programlisting>
</example>
</para>
</section>
<section id="pcm-interface-interrupt-handler-both">
<title>On calling <function>snd_pcm_period_elapsed()</function></title>
<para>
In both cases, even if more than one period are elapsed, you
don't have to call
<function>snd_pcm_period_elapsed()</function> many times. Call
only once. And the pcm layer will check the current hardware
pointer and update to the latest status.
</para>
</section>
</section>
<section id="pcm-interface-atomicity">
<title>Atomicity</title>
<para>
One of the most important (and thus difficult to debug) problems
in kernel programming are race conditions.
In the Linux kernel, they are usually avoided via spin-locks, mutexes
or semaphores. In general, if a race condition can happen
in an interrupt handler, it has to be managed atomically, and you
have to use a spinlock to protect the critical session. If the
critical section is not in interrupt handler code and
if taking a relatively long time to execute is acceptable, you
should use mutexes or semaphores instead.
</para>
<para>
As already seen, some pcm callbacks are atomic and some are
not. For example, the <parameter>hw_params</parameter> callback is
non-atomic, while <parameter>trigger</parameter> callback is
atomic. This means, the latter is called already in a spinlock
held by the PCM middle layer. Please take this atomicity into
account when you choose a locking scheme in the callbacks.
</para>
<para>
In the atomic callbacks, you cannot use functions which may call
<function>schedule</function> or go to
<function>sleep</function>. Semaphores and mutexes can sleep,
and hence they cannot be used inside the atomic callbacks
(e.g. <parameter>trigger</parameter> callback).
To implement some delay in such a callback, please use
<function>udelay()</function> or <function>mdelay()</function>.
</para>
<para>
All three atomic callbacks (trigger, pointer, and ack) are
called with local interrupts disabled.
</para>
</section>
<section id="pcm-interface-constraints">
<title>Constraints</title>
<para>
If your chip supports unconventional sample rates, or only the
limited samples, you need to set a constraint for the
condition.
</para>
<para>
For example, in order to restrict the sample rates in the some
supported values, use
<function>snd_pcm_hw_constraint_list()</function>.
You need to call this function in the open callback.
<example>
<title>Example of Hardware Constraints</title>
<programlisting>
<![CDATA[
static unsigned int rates[] =
{4000, 10000, 22050, 44100};
static struct snd_pcm_hw_constraint_list constraints_rates = {
.count = ARRAY_SIZE(rates),
.list = rates,
.mask = 0,
};
static int snd_mychip_pcm_open(struct snd_pcm_substream *substream)
{
int err;
....
err = snd_pcm_hw_constraint_list(substream->runtime, 0,
SNDRV_PCM_HW_PARAM_RATE,
&constraints_rates);
if (err < 0)
return err;
....
}
]]>
</programlisting>
</example>
</para>
<para>
There are many different constraints.
Look at <filename>sound/pcm.h</filename> for a complete list.
You can even define your own constraint rules.
For example, let's suppose my_chip can manage a substream of 1 channel
if and only if the format is S16_LE, otherwise it supports any format
specified in the <structname>snd_pcm_hardware</structname> structure (or in any
other constraint_list). You can build a rule like this:
<example>
<title>Example of Hardware Constraints for Channels</title>
<programlisting>
<![CDATA[
static int hw_rule_format_by_channels(struct snd_pcm_hw_params *params,
struct snd_pcm_hw_rule *rule)
{
struct snd_interval *c = hw_param_interval(params,
SNDRV_PCM_HW_PARAM_CHANNELS);
struct snd_mask *f = hw_param_mask(params, SNDRV_PCM_HW_PARAM_FORMAT);
struct snd_mask fmt;
snd_mask_any(&fmt); /* Init the struct */
if (c->min < 2) {
fmt.bits[0] &= SNDRV_PCM_FMTBIT_S16_LE;
return snd_mask_refine(f, &fmt);
}
return 0;
}
]]>
</programlisting>
</example>
</para>
<para>
Then you need to call this function to add your rule:
<informalexample>
<programlisting>
<![CDATA[
snd_pcm_hw_rule_add(substream->runtime, 0, SNDRV_PCM_HW_PARAM_CHANNELS,
hw_rule_channels_by_format, 0, SNDRV_PCM_HW_PARAM_FORMAT,
-1);
]]>
</programlisting>
</informalexample>
</para>
<para>
The rule function is called when an application sets the number of
channels. But an application can set the format before the number of
channels. Thus you also need to define the inverse rule:
<example>
<title>Example of Hardware Constraints for Channels</title>
<programlisting>
<![CDATA[
static int hw_rule_channels_by_format(struct snd_pcm_hw_params *params,
struct snd_pcm_hw_rule *rule)
{
struct snd_interval *c = hw_param_interval(params,
SNDRV_PCM_HW_PARAM_CHANNELS);
struct snd_mask *f = hw_param_mask(params, SNDRV_PCM_HW_PARAM_FORMAT);
struct snd_interval ch;
snd_interval_any(&ch);
if (f->bits[0] == SNDRV_PCM_FMTBIT_S16_LE) {
ch.min = ch.max = 1;
ch.integer = 1;
return snd_interval_refine(c, &ch);
}
return 0;
}
]]>
</programlisting>
</example>
</para>
<para>
...and in the open callback:
<informalexample>
<programlisting>
<![CDATA[
snd_pcm_hw_rule_add(substream->runtime, 0, SNDRV_PCM_HW_PARAM_FORMAT,
hw_rule_format_by_channels, 0, SNDRV_PCM_HW_PARAM_CHANNELS,
-1);
]]>
</programlisting>
</informalexample>
</para>
<para>
I won't give more details here, rather I
would like to say, <quote>Luke, use the source.</quote>
</para>
</section>
</chapter>
<!-- ****************************************************** -->
<!-- Control Interface -->
<!-- ****************************************************** -->
<chapter id="control-interface">
<title>Control Interface</title>
<section id="control-interface-general">
<title>General</title>
<para>
The control interface is used widely for many switches,
sliders, etc. which are accessed from user-space. Its most
important use is the mixer interface. In other words, since ALSA
0.9.x, all the mixer stuff is implemented on the control kernel API.
</para>
<para>
ALSA has a well-defined AC97 control module. If your chip
supports only the AC97 and nothing else, you can skip this
section.
</para>
<para>
The control API is defined in
<filename>&lt;sound/control.h&gt;</filename>.
Include this file if you want to add your own controls.
</para>
</section>
<section id="control-interface-definition">
<title>Definition of Controls</title>
<para>
To create a new control, you need to define the
following three
callbacks: <structfield>info</structfield>,
<structfield>get</structfield> and
<structfield>put</structfield>. Then, define a
struct <structname>snd_kcontrol_new</structname> record, such as:
<example>
<title>Definition of a Control</title>
<programlisting>
<![CDATA[
static struct snd_kcontrol_new my_control __devinitdata = {
.iface = SNDRV_CTL_ELEM_IFACE_MIXER,
.name = "PCM Playback Switch",
.index = 0,
.access = SNDRV_CTL_ELEM_ACCESS_READWRITE,
.private_value = 0xffff,
.info = my_control_info,
.get = my_control_get,
.put = my_control_put
};
]]>
</programlisting>
</example>
</para>