.\" $NetBSD: audio.9,v 1.56 2019/06/12 13:53:25 wiz Exp $
.\"
.\" Copyright (c) 1999, 2000 The NetBSD Foundation, Inc.
.\" All rights reserved.
.\"
.\" This code is derived from software contributed to The NetBSD Foundation
.\" by Lennart Augustsson.
.\"
.\" Redistribution and use in source and binary forms, with or without
.\" modification, are permitted provided that the following conditions
.\" are met:
.\" 1. Redistributions of source code must retain the above copyright
.\" notice, this list of conditions and the following disclaimer.
.\" 2. Redistributions in binary form must reproduce the above copyright
.\" notice, this list of conditions and the following disclaimer in the
.\" documentation and/or other materials provided with the distribution.
.\"
.\" THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS
.\" ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
.\" TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
.\" PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS
.\" BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
.\" CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
.\" SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
.\" INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
.\" CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
.\" ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
.\" POSSIBILITY OF SUCH DAMAGE.
.\"
.Dd May 9, 2019
.Dt AUDIO 9
.Os
.Sh NAME
.Nm audio
.Nd interface between low and high level audio drivers
.Sh DESCRIPTION
The audio device driver is divided into a high level,
hardware independent layer, and a low level hardware
dependent layer.
The interface between these is the
.Va audio_hw_if
structure.
.Bd -literal
struct audio_hw_if {
int (*open)(void *, int);
void (*close)(void *);
int (*query_format)(void *, audio_format_query_t *);
int (*set_format)(void *, int,
const audio_params_t *, const audio_params_t *,
audio_filter_reg_t *, audio_filter_reg_t *);
int (*round_blocksize)(void *, int, int, const audio_params_t *);
int (*commit_settings)(void *);
int (*init_output)(void *, void *, int);
int (*init_input)(void *, void *, int);
int (*start_output)(void *, void *, int, void (*)(void *),
void *);
int (*start_input)(void *, void *, int, void (*)(void *),
void *);
int (*halt_output)(void *);
int (*halt_input)(void *);
int (*speaker_ctl)(void *, int);
#define SPKR_ON 1
#define SPKR_OFF 0
int (*getdev)(void *, struct audio_device *);
int (*set_port)(void *, mixer_ctrl_t *);
int (*get_port)(void *, mixer_ctrl_t *);
int (*query_devinfo)(void *, mixer_devinfo_t *);
void *(*allocm)(void *, int, size_t);
void (*freem)(void *, void *, size_t);
size_t (*round_buffersize)(void *, int, size_t);
int (*get_props)(void *);
int (*trigger_output)(void *, void *, void *, int,
void (*)(void *), void *, const audio_params_t *);
int (*trigger_input)(void *, void *, void *, int,
void (*)(void *), void *, const audio_params_t *);
int (*dev_ioctl)(void *, u_long, void *, int, struct lwp *);
void (*get_locks)(void *, kmutex_t **, kmutex_t **);
};
typedef struct audio_params {
u_int sample_rate; /* sample rate */
u_int encoding; /* e.g. mu-law, linear, etc */
u_int precision; /* bits/subframe */
u_int validbits; /* valid bits in a subframe */
u_int channels; /* mono(1), stereo(2) */
} audio_params_t;
.Ed
.Pp
The high level audio driver attaches to the low level driver
when the latter calls
.Va audio_attach_mi .
This call should be
.Bd -literal
device_t
audio_attach_mi(const struct audio_hw_if *ahwp, void *hdl, device_t dev);
.Ed
.Pp
The
.Va audio_hw_if
struct is as shown above.
The
.Va hdl
argument is a handle to some low level data structure.
It is sent as the first argument to all the functions
in
.Va audio_hw_if
when the high level driver calls them.
.Va dev
is the device struct for the hardware device.
.Pp
The upper layer of the audio driver allocates one buffer for playing
and one for recording.
It handles the buffering of data from the user processes in these.
The data is presented to the lower level in smaller chunks, called blocks.
If, during playback, there is no data available from the user process when
the hardware request another block a block of silence will be used instead.
Furthermore, if the user process does not read data quickly enough during
recording data will be thrown away.
.Pp
The phase that these functions are called is classified into three.
Attach phase, Closed phase and Opened phase.
Attach phase is during device attach and
it transits to the Closed phase when the attach succeeded.
Closed phase is when no sampling device is opened and
it transits to the Opened phase when open succeeded.
Opened phase is when any sampling device is opened and
it transits to the Closed phase when close succeeded.
.Pp
The fields of
.Va audio_hw_if
are described in some more detail below.
Some fields are optional and can be set to
.Dv NULL
if not needed.
.Bl -tag -width indent
.It Dv int open(void *hdl, int flags)
optional, is called when the first device combining playback and recording
is opened.
On a full duplex hardware,
.Dv ( FREAD | FWRITE )
is passed to flags.
On a half duplex hardware,
.Dv FWRITE
is passed for playback, or
.Dv FREAD
for recording.
Every successful call to
.Va open
is matched by a call to
.Va close .
Return 0 on success, otherwise an error code.
It is called in the Closed phase.
.It Dv void close(void *hdl)
optional, is called when the last audio device combining
playback and recording is closed.
It is called in the Opened phase.
.It Dv int query_format(void *hdl, audio_format_query_t *afp)
is called to enumerate formats supported by the hardware.
It should fill the
.Vt audio_format_t
structure according to given number
.Va afp->index .
If there is no format with given number, return
.Er EINVAL .
It is called at any time.
.Bd -literal
typedef struct audio_format_query {
u_int index;
struct audio_format fmt;
} audio_format_query_t;
.Ed
.Pp
It is also used to determine the default format.
The upper layer chooses the most preferred one as default format by following:
.Bl -enum
.It
Higher priority is preferred (normally 0, the highest is 3, the lowest is 0).
.It
.Dv AUDIO_ENCODING_SLINEAR_NE:16
is preferred if exists.
.It
.Dv AUDIO_ENCODING_SLINEAR_OE:16
is preferred if exists.
.It
More channels is preferred.
.El
.Pp
If the driver supports
.Dv SLINEAR_NE:16
and the upper layer chooses it,
the driver does not need to provide conversion function in
.Va set_format .
Similarly, if the driver supports
.Dv SLINEAR_OE:16
and the upper layer chooses it,
the driver does not need to provide a conversion function.
Because the upper layer only supports conversion between
.Dv SLINEAR_NE:16
and
.Dv SLINEAR_OE:16
for convenience.
If the upper layer chooses another format,
the driver needs to provide a conversion function in
.Va set_format .
See also
.Va set_format .
If the driver can not provide the conversion from/to
.Dv SLINEAR_NE:16 ,
set priority to \-1.
It means that the hardware supports this format but the driver does not
(e.g. AC3), and it will never be chosen.
.It Dv int set_foramt(void *hdl, int setmode,
.Dv "const audio_params_t *play, const audio_params_t *rec,"
.Dv "audio_filter_reg_t *pfil, audio_filter_reg_t *rfil)"
.Pp
is called to set specified format to the hardware,
when the device is attached or the hardware format is changed.
.Va setmode
is a combination of the
.Dv AUMODE_RECORD
and
.Dv AUMODE_PLAY
flags to indicate which modes are to be set.
.Pp
The
.Va play
and
.Va rec
structures contain the encoding parameters that should be set to the hardware.
If the driver has query_format interface,
all parameters on
.Va play
and/or
.Va rec
are chosen from formats returned by query_format.
Therefore
.Va play
and/or
.Va rec
are always settable.
If the driver does not have query_format interface,
the driver has to validate the format.
If the hardware does not support
.Dv AUDIO_ENCODING_SLINEAR_{NE,OE}:16 ,
conversion information should be filled the
.Va pfil
for playing or
.Va rfil
for recording.
The definition of
.Vt audio_filter_reg_t
and a related structure follow:
.Bd -literal
typedef struct {
const void *src;
const audio_format2_t *srcfmt;
void *dst;
const audio_format2_t *dstfmt;
int count;
void *context;
} audio_filter_arg_t;
typedef void(*audio_filter_t)(audio_filter_arg_t *arg);
typedef struct {
audio_filter_t codec;
void *context;
} audio_filter_reg_t;
.Ed
.Pp
.Va codec
is a conversion function and
.Va context
is an optional opaque pointer passed to
.Va codec .
.Pp
When
.Va codec
is called, all parameters required by
.Va codec
are contained in
.Va arg .
.Va src
points to the input buffer block,
.Va srcfmt
contains the input encoding parameters,
.Va dst
points to the output buffer block and
.Va dstfmt
contains the output encoding parameters.
.Va count
represents the number of frames to process on this call.
.Va src
and
.Va dst
are guaranteed to be able to consecutively access number of frames
specified by
.Va count.
.Va codec
must fill entire
.Va dst .
For example, let count = 100, srcfmt is { precision = 16, channels = 3 },
dstfmt is { precision = 8, channels = 4 },
in this case,
src block length = 2(bytes) * 3(channels) * 100(frames) = 600 bytes,
The length to be written to
.Va dst
block is 1(byte) * 4(channels) * 100(frames) = 400 bytes.
.Va codec
cannot abort the conversion halfway and there is no error reporting mechanism.
.Va context
is a opaque pointer that can be used by
.Va codec
if necessary.
.Pp
If the device does not have the
.Dv AUDIO_PROP_INDEPENDENT
property the same value is passed in both
.Va play
and
.Va rec .
Return 0 on success, otherwise an error code.
It is called in the Attach or Closed phases.
.It Dv int round_blocksize(void *hdl, int bs, int mode,
.Dv "const audio_params_t *param)"
.Pp
optional, is called with the block size,
.Va bs ,
that has been computed by the upper layer,
.Va mode ,
.Dv AUMODE_PLAY
or
.Dv AUMODE_RECORD ,
and
.Va param ,
encoding parameters for the hardware.
It should return a block size, possibly changed according to the needs
of the hardware driver.
It is called in the Attach or Closed phases.
.It Dv int commit_settings(void *hdl)
optional, is called after all calls to
.Va set_params ,
and
.Va set_port ,
are done.
A hardware driver that needs to get the hardware in and out of command
mode for each change can save all the changes during previous calls and
do them all here.
Return 0 on success, otherwise an error code.
It is called in the Attach or Closed phases.
.It Dv int init_output(void *hdl, void *buffer, int size)
optional, is called before any output starts, but when the total
.Va size
of the output
.Va buffer
has been determined.
It can be used to initialize looping DMA for hardware that needs that.
Return 0 on success, otherwise an error code.
It is called in the Attach or Closed phases.
.It Dv int init_input(void *hdl, void *buffer, int size)
optional, is called before any input starts, but when the total
.Va size
of the input
.Va buffer
has been determined.
It can be used to initialize looping DMA for hardware that needs that.
Return 0 on success, otherwise an error code.
It is called in the Attach or Closed phases.
.It Dv int start_output(void *hdl, void *block, int blksize,
.Dv "void (*intr)(void*), void *intrarg)"
.Pp
is called to start the transfer of
.Va blksize
bytes from
.Va block
to the audio hardware.
The call should return when the data transfer has been initiated
(normally with DMA).
When the hardware is ready to accept more samples the function
.Va intr
should be called with the argument
.Va intrarg .
Calling
.Va intr
will normally initiate another call to
.Va start_output .
Return 0 on success, otherwise an error code.
It is called in the Opened phase.
.It Dv int start_input(void *hdl, void *block, int blksize,
.Dv "void (*intr)(void*), void *intrarg)"
.Pp
is called to start the transfer of
.Va blksize
bytes to
.Va block
from the audio hardware.
The call should return when the data transfer has been initiated
(normally with DMA).
When the hardware is ready to deliver more samples the function
.Va intr
should be called with the argument
.Va intrarg .
Calling
.Va intr
will normally initiate another call to
.Va start_input .
Return 0 on success, otherwise an error code.
It is called in the Opened phase.
.It Dv int halt_output(void *hdl)
is called to abort the output transfer (started by
.Va start_output )
in progress.
Return 0 on success, otherwise an error code.
It is called in the Opened phase.
.It Dv int halt_input(void *hdl)
is called to abort the input transfer (started by
.Va start_input )
in progress.
Return 0 on success, otherwise an error code.
It is called in the Opened phase.
.It Dv int speaker_ctl(void *hdl, int on)
optional, is called when a half duplex device changes between
playing and recording.
It can, e.g., be used to turn on
and off the speaker.
Return 0 on success, otherwise an error code.
It is called in the Opened phase.
.It Dv int getdev(void *hdl, struct audio_device *ret)
Should fill the
.Va audio_device
struct with relevant information about the driver.
Return 0 on success, otherwise an error code.
It is called in the Opened phase.
.It Dv int set_port(void *hdl, mixer_ctrl_t *mc)
is called in when
.Dv AUDIO_MIXER_WRITE
is used.
It should take data from the
.Va mixer_ctrl_t
struct at set the corresponding mixer values.
Return 0 on success, otherwise an error code.
It is called in the Opened or Closed phases.
.It Dv int get_port(void *hdl, mixer_ctrl_t *mc)
is called in when
.Dv AUDIO_MIXER_READ
is used.
It should fill the
.Va mixer_ctrl_t
struct.
Return 0 on success, otherwise an error code.
It is called in the Opened or Closed phases.
.It Dv int query_devinfo(void *hdl, mixer_devinfo_t *di)
is called in when
.Dv AUDIO_MIXER_DEVINFO
is used.
It should fill the
.Va mixer_devinfo_t
struct.
Return 0 on success, otherwise an error code.
It is called at any time.
.It Dv "void *allocm(void *hdl, int direction, size_t size)"
optional, is called to allocate the device buffers.
If not present
.Xr malloc 9
is used instead (with the same arguments but the first two).
The reason for using a device dependent routine instead of
.Xr malloc 9
is that some buses need special allocation to do DMA.
Returns the address of the buffer, or
.Dv NULL
on failure.
It is called in the Attached or Closed phases.
.It Dv void freem(void *hdl, void *addr, size_t size)
optional, is called to free memory allocated by
.Va allocm .
If not supplied
.Xr free 9
is used.
It is called in the Attached or Closed phases.
.It Dv size_t round_buffersize(void *hdl, int direction, size_t bufsize)
optional, is called at startup to determine the audio
buffer size.
The upper layer supplies the suggested size in
.Va bufsize ,
which the hardware driver can then change if needed.
E.g., DMA on the ISA bus cannot exceed 65536 bytes.
It is called in the Attached or Closed phases.
.It Dv int get_props(void *hdl)
Should return the device properties in a combination of following flags:
.Pp
.Bl -tag -width AUDIO_PROP_INDEPENDENT -compact
.It Dv AUDIO_PROP_PLAYBACK
the device is capable of audio playback.
.It Dv AUDIO_PROP_CAPTURE
the device is capable of audio capture.
.It Dv AUDIO_PROP_FULLDUPLEX
the device admits full duplex operation.
Don't set it if the device is unidirectional.
.It Dv AUDIO_PROP_INDEPENDENT
the device can set the playing and recording encoding parameters
independently.
Don't set it if the device is unidirectional.
.It Dv AUDIO_PROP_MMAP
is handled in the upper layer, so new drivers should not return this property.
.El
It is called in the Attach phase.
.It Dv int trigger_output(void *hdl, void *start, void *end,
.Dv "int blksize, void (*intr)(void*), void *intrarg,"
.Pp
.Dv "const audio_params_t *param)"
.Pp
optional, is called to start the transfer of data from the circular buffer
delimited by
.Va start
and
.Va end
to the audio hardware, parameterized as in
.Va param .
The call should return when the data transfer has been initiated
(normally with DMA).
When the hardware is finished transferring each
.Va blksize
sized block, the function
.Va intr
should be called with the argument
.Va intrarg
(typically from the audio hardware interrupt service routine).
Once started the transfer may be stopped using
.Va halt_output .
Return 0 on success, otherwise an error code.
It is called in the Opened phase.
.It Dv int trigger_input(void *hdl, void *start, void *end,
.Dv "int blksize, void (*intr)(void*), void *intrarg,"
.Pp
.Dv "const audio_params_t *param)"
.Pp
optional, is called to start the transfer of data from the audio hardware,
parameterized as in
.Va param ,
to the circular buffer delimited by
.Va start
and
.Va end .
The call should return when the data transfer has been initiated
(normally with DMA).
When the hardware is finished transferring each
.Va blksize
sized block, the function
.Va intr
should be called with the argument
.Va intrarg
(typically from the audio hardware interrupt service routine).
Once started the transfer may be stopped using
.Va halt_input .
Return 0 on success, otherwise an error code.
It is called in the Opened phase.
.It Dv int dev_ioctl(void *hdl, u_long cmd, void *addr,
.Pp
.Dv "int flag, struct lwp *l)"
.Pp
optional, is called when an
.Xr ioctl 2
is not recognized by the generic audio driver.
Return 0 on success, otherwise an error code.
It is called in the Opened phase.
.It Dv void get_locks(void *hdl, kmutex_t **intr, kmutex_t **thread)
Returns the interrupt and thread locks to the common audio layer.
It is called in the Attach phase.
.El
.Pp
The
.Va query_devinfo
method should define certain mixer controls for
.Dv AUDIO_SETINFO
to be able to change the port and gain,
and
.Dv AUDIO_GETINFO
to read them, as follows.
.Pp
If the record mixer is capable of input from more than one source,
it should define
.Dv AudioNsource
in class
.Dv AudioCrecord .
This mixer control should be of type
.Dv AUDIO_MIXER_ENUM
or
.Dv AUDIO_MIXER_SET
and enumerate the possible input sources.
Each of the named sources for which the recording level can be set
should have a control in the
.Dv AudioCrecord
class of type
.Dv AUDIO_MIXER_VALUE ,
except the
.Qq mixerout
source is special,
and will never have its own control.
Its selection signifies,
rather,
that various sources in class
.Dv AudioCrecord
will be combined and presented to the single recording output
in the same fashion that the sources of class
.Dv AudioCinputs
are combined and presented to the playback output(s).
If the overall recording level can be changed,
regardless of the input source,
then this control should be named
.Dv AudioNmaster
and be of class
.Dv AudioCrecord .
.Pp
Controls for various sources that affect only the playback output,
as opposed to recording,
should be in the
.Dv AudioCinputs
class,
as of course should any controls that affect both playback and recording.
.Pp
If the play
mixer is capable of output to more than one destination,
it should define
.Dv AudioNselect
in class
.Dv AudioCoutputs .
This mixer control should be of type
.Dv AUDIO_MIXER_ENUM
or
.Dv AUDIO_MIXER_SET
and enumerate the possible destinations.
For each of the named destinations for which the output level can be set,
there should be
a control in the
.Dv AudioCoutputs
class of type
.Dv AUDIO_MIXER_VALUE .
If the overall output level can be changed,
which is invariably the case,
then this control should be named
.Dv AudioNmaster
and be of class
.Dv AudioCoutputs .
.Pp
There's one additional source recognized specially by
.Dv AUDIO_SETINFO
and
.Dv AUDIO_GETINFO ,
to be presented as monitor_gain,
and that is a control named
.Dv AudioNmonitor ,
of class
.Dv AudioCmonitor .
.Sh SEE ALSO
.Xr audio 4
.Sh HISTORY
This
.Nm
interface first appeared in
.Nx 1.3 .