Difference between revisions of "Advanced Linux Sound Architecture"

From ArchWiki
Jump to navigation Jump to search
(This is mentioned at least 4 times in the article - Undo revision 346751 by Grakker (talk))
m (→‎HDA analyzer: The listed workaround in the "note" section is not entirely correct. The script will not run with "python run.py" after the edits are made; instead it will run with "python2 run.py". This was only tested after the edits are made.)
Line 1,156: Line 1,156:
* The script is done by such way that it is incompatible with Python 3 (which is now shipped with Arch Linux) but tries to use it. The workaround is: open "run.py", find all occurrences of "python" (2 occurrences - one on the first line, and the second on the last line) and replace them all by "python2". The script requires root privileges.
* The script is done by such way that it is incompatible with Python 3 (which is now shipped with Arch Linux) but tries to use it. The workaround is: open "run.py", find all occurrences of "python" (2 occurrences - one on the first line, and the second on the last line) and replace them all by "python2". Then "python2 run.py" The script requires root privileges.

Revision as of 17:48, 10 December 2014

zh-CN:Advanced Linux Sound Architecture

The Advanced Linux Sound Architecture (ALSA) is a Linux kernel component which replaced the original Open Sound System (OSSv3) for providing device drivers for sound cards. Besides the sound device drivers, ALSA also bundles a user space library for application developers who want to use driver features with a higher level API than direct interaction with the kernel drivers.

Note: For an alternative sound environment, see the Open Sound System page.



ALSA is included in the default Arch Linux kernel (linux) as a set of modules, so manual installation is not necessary.

udev will automatically probe your hardware at boot, loading the corresponding kernel module for your audio card. Therefore, your sound should already be working, but muted (as is the upstream default).

Users with a local login (at a virtual terminal or a display manager) have permission to play audio and change mixer levels. To allow this for a remote login, the user has to be added to the audio group. Membership in the audio group also allows direct access to devices, which can lead to applications grabbing exclusive output (breaking software mixing) and breaks fast-user-switching, and multiseat. Therefore, adding a user to the audio group is not recommended, unless you specifically need to[1].

User space utilities

Install the alsa-utils package from the official repositories, which (among other utilities) contains the programs alsamixer and amixer. alsamixer provides an ncurses based interface for shell based configuration of audio devices. Also install the alsa-plugins package if you want high quality resampling, upmixing/downmixing and other advanced features.

If you want OSS applications to work with dmix install the alsa-oss package as well. Load the kernel modules snd_seq_oss, snd_pcm_oss and snd_mixer_oss to enable OSS emulation.

Unmuting the channels

ALSA by default has all channels muted, all of which will need to be unmuted manually. This can be done using amixer:

$ amixer sset Master unmute

This can alternatively be done using alsamixer:

$ alsamixer

The MM label below a channel indicates that the channel is muted, and 00 indicates that it is open.

Scroll to the Master and PCM channels with the and keys and unmute them by pressing the m key. Use the key to increase the volume and obtain a value of 0 dB gain. The gain can be found in the upper left next to the Item: field.

Note: If gain is set above 0 dB audible distortion can become present.

To get full 5.1 or 7.1 surround sound you will likely need to unmute other channels such as Front, Surround, Center, LFE (subwoofer) and Side (these are the names of the channels with Intel HD Audio, they may vary with different hardware). Please take note that this will not automatically upmix stereo sources (like most music). In order to accomplish that, see #Upmixing/downmixing.

To enable your microphone, switch to the Capture tab with F4 and enable a channel with Space.

To exit alsamixer, press Esc.

  • Some cards need to have digital output muted or disabled in order to hear analog sound. For the Soundblaster Audigy LS mute the channel labeled IEC958.
  • Some machines, (like the Thinkpad T61), have a Speaker channel which must be unmuted and adjusted as well.
  • Some machines, (like the Dell E6400) may also require the Front and Headphone channels to be unmuted and adjusted.
  • If your volume adjustments seem to be lost after you reboot, try running alsamixer as root.

Next, test to see if sound works:

$ speaker-test -c 2

Change -c to fit your speaker setup. Use -c 8 for 7.1, for instance:

$ speaker-test -c 8

If audio is being outputted to the wrong device, try manually specifying it with the argument -D.

$ speaker-test -D default -c 8

-D accepts PCM channel names as values, which can be retrieved by running the following:

$ aplay -L | grep :CARD
default:CARD=PCH  # 'default' is the PCM channel name

If that does not work, consult the Configuration and then Troubleshooting sections.

The alsa-utils package includes the systemd unit configuration files alsa-restore.service and alsa-store.service, which are activated during boot and before shutdown respectively. They do not need to be manually enabled using systemctl. For reference, ALSA stores its settings in /var/lib/alsa/asound.state.


Basic syntax

ALSA configuration files follow a simple syntax consisting of hierarchical value to parameter (key) assignments. Below are (modified) excerpts from asoundrc.txt, which is usually found in alsa-lib package but can be also reached here.

Assignments and Separators

Assignments define a value of a given key. There are different assignment types and styles available.

Simple assignment
# This is a comment. Everything after the '#' symbol to the end of the line will be ignored by ALSA.
key = value # Equal signs are usually left out, since space can also be used as an separator.

key value # Equivalent to the example above.

Separators are used to indicate the start and end of an assignment, but using commas or whitespace is also possible.

# The following three assignments are equivalent.
key value0; key valueN;
key value0, key valueN,
key value0 key valueN


Compound assignments use braces as separators.

Compound assignment
key {	subkey0 value0;
	subkeyN valueN;	}

key.subkey0 value0; # Equivalent to the example above.
key.subkeyN valueN;

For easier reading, it is recommended to use first style for definitions including more then three keys.

Array definitions use brackets as separators.

Single array
key [	"value0";
	"valueN";	]

key.0 "value0"; # Equivalent to the example above
key.N "valueN";

Everything depends on user preferences when it comes to different styles of configuration, however one should avoid mixing different styles. Further information on basic configuration can be found here.

Data types

ALSA uses different data types for parameter values, which must be set in the users respective configuration file. Some keys accept multiple data types, while most do not. A list of configuration options and their respective type requirements for PCM plugins can be found here

Operation modes

There are different operation modes for parsing nodes, the default mode is merge and create. If operation mode is either merge/create or merge type checking is done. Only same type assignments can be merged, so strings cannot be merged with integers. Trying to define a simple assignment in default operation mode to a compound (and vice versa) will also not work.

Prefixes of operation modes:

  • "+" -- merge and create
  • "-" -- merge
  • "?" -- do not override
  • "!" -- override
Operation modes
# Merge/create - If a node does not exist, it is created. If it does exist and types match,
# subkeyN is merged into key.
key.subkeyN valueN;

# Merge/create - Equivalent to above
key.+subkeyN valueN;

# Merge - Node key.subkeyN must already exist and must have same data type
key.-subkeyN valueN;

# No override - Ignore new assignment if key.subkeyN node already exists
key.?subkeyN valueN;

# Override - Removes subkeyN and all keys below it, then creates node key.subkeyN
key.!subkeyN valueN;

Using override operation mode, when done correctly, is usually safe, however one should bear in mind, that there might be other necessary keys in a node for proper functioning.

Warning: Overriding pcm node itself will most definitely make alsa unusable, since every plugin definition will be deleted. Therefore do not use !pcm.key unless you are making a configuration from scratch.
An example of setting default device using "defaults" node

Assuming that "defaults" node is set in /usr/share/alsa/alsa.conf, where defaults.pcm.card and its ctl counterpart have assignment values "0" (type integer), user wants to set default pcm and control device to (third) sound card "2" or "SB" for an Azalia sound card.

Defaults node
defaults.ctl.card 2; # Sets default device and control to third card (counting begins with 0).
defaults.pcm.card 2; # This does not change the data type.

defaults.ctl.+card 2; # Equivalent to above.
defaults.pcm.+card 2;

defaults.ctl.-card 2; # Same effect on a default setup, however if defaults node was removed or
defaults.pcm.-card 2; # type has been changed merge operation mode will result in no changes.

defaults.pcm.?card 2; # This does nothing, since this assignment already exists.
defaults.ctl.?card 2;

defaults.pcm.!card "SB"; # The override operation mode is necessary here, because of
defaults.ctl.!card "SB"; # different value types.

Using double quotes here automatically sets values data type to string, so in the above example setting defaults.pcm.!card "2" would result in retaining last default device, in this case card 0. Using double quotes for strings is not mandatory as long as no special characters are used, which ideally should never be the case. This may be irrelevant in other assignments.

Note: From a configuration point of view those are not equivalent to setting a compound "default" pcm device, since most users specify addressing type in there also, which actually may be the same, but the assignment itself still differs. Also defaults.pcm.card is referred to multiple times in alsa configuration files, usually as a fallback assignment, where different environment variables take precedence.


Sometimes it may be useful and even easier to read using nesting in configuration.

Nesting PCM plugins
pcm.azalia {	type hw; card 0	}
pcm.!default {	type plug; slave.pcm "azalia"	}

# is equivalent to

pcm.!default {	type plug; slave.pcm {	type hw; card 0;	}	}

# which is also equivalent to

pcm.!default.type plug;
pcm.default.slave.pcm.type hw;
pcm.default.slave.pcm.card 0;

Including configuration files

Include other configuration files
</path/to/configuration-file> # Include a configuration file
<confdir:/path/to/configuration-file> # Reference to a global configuration directory

Set the default sound card

Tango-view-fullscreen.pngThis article or section needs expansion.Tango-view-fullscreen.png

Reason: Explain the difference between a "card" and a "device." amixer --help suggests the two are not synonymous (Discuss in Talk:Advanced Linux Sound Architecture#)

If your sound card order changes on boot, you can specify their order in any file ending with .conf in /etc/modprobe.d (/etc/modprobe.d/alsa-base.conf is suggested). For example, if you want your mia sound card to be #0:

options snd_mia index=0
options snd_hda_intel index=1

Use $ cat /proc/asound/modules to get the loaded sound modules and their order. This list is usually all that is needed for the loading order. Use $ lsmod | grep snd to get a devices & modules list. This configuration assumes you have one mia sound card using snd_mia and one (e.g. onboard) card using snd_hda_intel.

You can also provide an index of -2 to instruct ALSA to never use a card as the primary one. Distributions such as Linux Mint and Ubuntu use the following settings to avoid USB and other "abnormal" drivers from getting index 0:

options bt87x index=-2
options cx88_alsa index=-2
options saa7134-alsa index=-2
options snd-atiixp-modem index=-2
options snd-intel8x0m index=-2
options snd-via82xx-modem index=-2
options snd-usb-audio index=-2
options snd-usb-caiaq index=-2
options snd-usb-ua101 index=-2
options snd-usb-us122l index=-2
options snd-usb-usx2y index=-2
options snd-pcsp index=-2
options snd-usb-audio index=-2

These changes require a system reboot.

See also [2]

Select the default PCM via environment variable

In your configuration file, preferably global, add:

pcm.!default {
    type plug
    slave.pcm {
        @func getenv
        vars [ ALSAPCM ]
        default "hw:Audigy2"

You need to replace the default line with the name of your card (in the example is Audigy2). You can get the names with aplay -l or you can also use PCMs like surround51. But if you need to use the microphone it is a good idea to select full-duplex PCM as default.

Now you can start programs selecting the sound card just changing the environment variable ALSAPCM. It works fine for all program that do not allow to select the card, for the others ensure you keep the default card. For example, assuming you wrote a downmix PCM called mix51to20 you can use it with mplayer using the commandline ALSAPCM=mix51to20 mplayer example_6_channel.wav

Instead of using new variables, you could set one of those mentioned in default global configuration.

Variable name # Definition
ALSA_CARD # pcm.default pcm.hw pcm.plughw ctl.sysdefault ctl.hw rawmidi.default rawmidi.hw hwdep.hw
ALSA_CTL_CARD # ctl.sysdefault ctl.hw
ALSA_HWDEP_CARD # hwdep.default hwdep.hw
ALSA_HWDEP_DEVICE # hwdep.default hwdep.hw
ALSA_PCM_CARD # pcm.default pcm.hw pcm.plughw
ALSA_PCM_DEVICE # pcm.hw pcm.plughw
ALSA_RAWMIDI_CARD # rawmidi.default rawmidi.hw
ALSA_RAWMIDI_DEVICE # rawmidi.default rawmidi.hw
Note: Pay attention to default addressing type.

Alternative method

Tip: This process can be partly automated using asoundconfAUR from the Arch User Repository.

First you will have to find out the card and device id that you want to set as the default:

$ aplay -l
**** List of PLAYBACK Hardware Devices ****
card 0: Intel [HDA Intel], device 0: CONEXANT Analog [CONEXANT Analog]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 0: Intel [HDA Intel], device 1: Conexant Digital [Conexant Digital]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 1: JamLab [JamLab], device 0: USB Audio [USB Audio]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 2: Audio [Altec Lansing XT1 - USB Audio], device 0: USB Audio [USB Audio]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
Warning: Simply setting a type hw as default card is equivalent to addressing hardware directly, which leaves the device unavailable to other applications. This method is only recommended if it is a part of a more sophisticated setup ~/.asoundrc or if user deliberately wants to address sound card directly (digital output through eic958 or dedicated music server for example).

For example, the last entry in this list has the card ID 2 and the device ID 0. To set this card as the default, you can either use the system-wide file /etc/asound.conf or the user-specific file ~/.asoundrc. You may have to create the file if it does not exist. Then insert the following options with the corresponding card.

pcm.!default {
    type hw
    card 2

ctl.!default {
    type hw
    card 2
Note: For the Asus U32U serie it seems that card should be set to 1 for both pcm and ctl.

In most cases it is recommended to use sound card names instead of number references, which also solves boot order problem. Therefore the following would be correct for the above example.

pcm.!default {
    type hw
    card Audio

ctl.!default {
    type hw
    card Audio

To get valid ALSA card names, use aplay:

$ aplay -l | awk -F \: '/,/{print $2}' | awk '{print $1}' | uniq

Alternatively use cat, which might return unused devices:

$ cat /proc/asound/card*/id
Note: This method could be problematic if your system has several cards of the same (ALSA)name.

The 'pcm' options affect which card and device will be used for audio playback while the 'ctl' option affects which card is used by control utilities like alsamixer .

The changes should take effect as soon as you (re-)start an application (MPlayer etc.). You can also test with a command like aplay.

$ aplay -D default your_favourite_sound.wav

If you receive an error regarding your asound configuration, check the upstream documentation for possible changes to the config file format.

Verifying correct sound modules are loaded

You can assume that udev will autodetect your sound properly. You can check this with the command

$ lsmod | grep '^snd' | column -t
snd_hda_codec_hdmi     22378   4
snd_hda_codec_realtek  294191  1
snd_hda_intel          21738   1
snd_hda_codec          73739   3  snd_hda_codec_hdmi,snd_hda_codec_realtek,snd_hda_intel
snd_hwdep              6134    1  snd_hda_codec
snd_pcm                71032   3  snd_hda_codec_hdmi,snd_hda_intel,snd_hda_codec
snd_timer              18992   1  snd_pcm
snd                    55132   9  snd_hda_codec_hdmi,snd_hda_codec_realtek,snd_hda_intel,snd_hda_codec,snd_hwdep,snd_pcm,snd_timer
snd_page_alloc         7017    2  snd_hda_intel,snd_pcm

If the output looks similar, your sound drivers have been successfully autodetected.

Note: Since udev>=171, the OSS emulation modules (snd_seq_oss, snd_pcm_oss, snd_mixer_oss) are not loaded by default: Load them manually if they are needed.

You might also want to check the directory /dev/snd/ for the right device files:

$ ls -l /dev/snd
total 0
crw-rw----  1 root audio 116,  0 Apr  8 14:17 controlC0
crw-rw----  1 root audio 116, 32 Apr  8 14:17 controlC1
crw-rw----  1 root audio 116, 24 Apr  8 14:17 pcmC0D0c
crw-rw----  1 root audio 116, 16 Apr  8 14:17 pcmC0D0p
crw-rw----  1 root audio 116, 25 Apr  8 14:17 pcmC0D1c
crw-rw----  1 root audio 116, 56 Apr  8 14:17 pcmC1D0c
crw-rw----  1 root audio 116, 48 Apr  8 14:17 pcmC1D0p
crw-rw----  1 root audio 116,  1 Apr  8 14:17 seq
crw-rw----  1 root audio 116, 33 Apr  8 14:17 timer
Note: If requesting help on IRC or the forums, please post the output of the above commands if requested.

If you have at least the devices controlC0 and pcmC0D0p or similar, then your sound modules have been detected and loaded properly.

If this is not the case, your sound modules have not been detected properly. To solve this, you can try loading the modules manually:

  • Locate the module for your sound card: ALSA Soundcard Matrix The module will be prefixed with 'snd-' (for example: snd-via82xx).
  • Load the module.
  • Check for the device files in /dev/snd (see above) and/or try if alsamixer or amixer have reasonable output.
  • Configure snd-NAME-OF-MODULE and snd-pcm-oss to load at boot.

Getting S/PDIF output

S/PDIF is a digital audio interface often used to connect a computer to a digital amplifier (such as a home theatre with 5.1/7.1 surround sound).

Depending on what shell you use, add the following line to your shell's configuration file:

amixer -c 0 cset name='IEC958 Playback Switch' on

You can see the name of your card's digital output with:

$ amixer scontrols

System-wide equalizer

Using ALSAEqual (provides UI)

Install the alsaequalAUR package from the Arch User Repository.

Note: If you have a x86_64-system and are using a 32bit-flashplugin the sound in flash will not work. You have to disable alsaequal or build alsaequal for 32bit.

After installing the package, add the following to your ALSA configuration file:

ctl.equal {
    type equal;

pcm.plugequal {
    type equal;
    # Modify the line below if you do not
    # want to use sound card 0.
    #slave.pcm "plughw:0,0";
    # by default we want to play from more sources at time:
    slave.pcm "plug:dmix";

# pcm.equal {
# If you do not want the equalizer to be your
# default soundcard comment the following
# line and uncomment the above line. (You can
# choose it as the output device by addressing
# it with specific apps,eg mpg123 -a equal 06.Back_In_Black.mp3)
pcm.!default {
    type plug;
    slave.pcm plugequal;

And you are ready to change your equalizer using command

$ alsamixer -D equal

Note that configuration file is different for each user (until not specified else) it is saved in ~/.alsaequal.bin. so if you want to use ALSAEqual with mpd or another software running under different user, you can configure it using

$ su mpd -c 'alsamixer -D equal'

or for example, you can make a symlink to your .alsaequal.bin in his home...

Managing ALSAEqual states

Install alsaequal-mgrAUR from the Arch User Repository.

Configure the equalizer as usual with

$ alsamixer -D equal

When you are satisfied with the state, you may give it a name ("foo" in this example) and save it:

$ alsaequal-mgr save foo

The state "foo" can then be restored at a later time with

$ alsaequal-mgr load foo

You can thus create different equalizer states for games, movies, music genres, VoIP apps, etc. and reload them as necessary.

See the project page and the help message for more options.

Using mbeq

Note: This method requires the use of a LADSPA plugin which might be CPU intensive during playback. In addition, this was made with stereophonic sound (e.g. headphones) in mind.

Install the alsa-plugins, ladspa and swh-plugins packages if you do not already have them.

If you have not already created either an ~/.asoundrc or a /etc/asound.conf file, then create either one and insert the following:

pcm.eq {
    type ladspa

    # The output from the EQ can either go direct to a hardware device
    # (if you have a hardware mixer, e.g. SBLive/Audigy) or it can go
    # to the software mixer shown here.
    #slave.pcm "plughw:0,0"
    slave.pcm "plug:dmix"

    # Sometimes you may need to specify the path to the plugins,
    # especially if you have just installed them.  Once you have logged
    # out/restarted this should not be necessary, but if you get errors
    # about being unable to find plugins, try uncommenting this.
    #path "/usr/lib/ladspa"

    plugins [
        label mbeq
        id 1197
        input {
            # The following setting is just an example, edit to your own taste:
            # bands: 50hz, 100hz, 156hz, 220hz, 311hz, 440hz, 622hz, 880hz, 1250hz, 1750hz, 25000hz,
            # 50000hz, 10000hz, 20000hz
            controls [ -5 -5 -5 -5 -5 -10 -20 -15 -10 -10 -10 -10 -10 -3 -2 ]

# Redirect the default device to go via the EQ - you may want to do
# this last, once you are sure everything is working.  Otherwise all
# your audio programs will break/crash if something has gone wrong.
pcm.!default {
    type plug
    slave.pcm "eq"

# Redirect the OSS emulation through the EQ too (when programs are running through "aoss")
pcm.dsp0 {
    type plug
    slave.pcm "eq"

High quality resampling

When software mixing is enabled, ALSA is forced to resample everything to the same frequency (48 kHz by default when supported). By default, it will try to use the speexrate converter to do so, and fallback to low-quality linear interpolation if it is not available[3]. Thus, if you are getting poor sound quality due to bad resampling, the problem can be solved by simply installing alsa-plugins.

For even higher quality resampling, you can change the default rate converter to speexrate_medium or speexrate_best. Both perform well enough that in practice it does not matter which one you choose, so using the best converter is usually not worth the extra CPU cycles it requires.

To change the default converter place the following contents in your ~/.asoundrc or /etc/asound.conf:

defaults.pcm.rate_converter "speexrate_medium"
Note: It is also possible to use libsamplerate converters, which are only about half as fast as the speexrate converters but do not achieve much greater performance. See discussion.
Note: Some applications (like MPlayer and its forks) do their own resampling by default because some ALSA drivers have incorrect delay reporting when resampling is enabled (hence leading to AV desynchronization), so changing this setting will not have any effect unless you configure them to use ALSA resampling.



In order for stereo sources like music to be able to saturate a 5.1 or 7.1 sound system, you need to use upmixing. In darker days this used to be tricky and error prone but nowadays plugins exist to easily take care of this task. We will use the upmix plugin, included in alsa-plugins.

Then add the following to your ALSA configuration file of choice (either /etc/asound.conf or ~/.asoundrc):

pcm.upmix71 {
    type upmix
    slave.pcm "surround71"
    delay 15
    channels 8

You can easily change this example for 7.1 upmixing to 5.1 or 4.0.

The following example adds a new PCM channel that you can use for upmixing. If you want all sound sources to go through this channel, add it as a default below the previous definition like so:

pcm.!default "plug:upmix71"

The plugin automatically allows multiple sources to play through it without problems so setting is as a default is actually a safe choice. If this is not working, you have to setup your own dmixer for the upmixing PCM like this:

pcm.dmix6 {
    type asym
    playback.pcm {
        type dmix
        ipc_key 567829
        slave {
            pcm "hw:0,0"
            channels 6

and use "dmix6" instead of "surround71". If you experience skipping or distorted sound, consider increasing the buffer_size (to 32768, for example) or use a high quality resampler.


If you want to downmix sources to stereo because you, for instance, want to watch a movie with 5.1 sound on a stereo system, use the vdownmix plugin, included in alsa-plugins..

Again, in your configuration file, add this:

pcm.!surround51 {
    type vdownmix
    slave.pcm "default"
pcm.!surround40 {
    type vdownmix
    slave.pcm "default"
Note: This might not be enough to make downmixing working, see [4]. So, you might also need to add pcm.!default "plug:surround51" or pcm.!default "plug:surround40". Only one vdownmix plug can be used; if you have 7.1 channels, you will need to use surround71 instead the configuration above. A good example, which includes a configuration that makes both vdownmix and dmix working, can be found here.


Mixing enables multiple applications to output sound at the same time. Most discrete sound cards support hardware mixing, which is enabled by default if available. Integrated motherboard sound cards (such as Intel HD Audio), usually do not support hardware mixing. On such cards, software mixing is done by an ALSA plugin called dmix. This feature is enabled automatically if hardware mixing is unavailable.

Note: Dmix is enabled by default for soundcards which do not support hardware mixing. Dmix is not enabled by default for digital output (S/PDIF) and will require the configuration snippet below.

To manually enable dmix, add the following to your ALSA configuration file:

pcm.dsp {
    type plug
    slave.pcm "dmix"


No sound in VirtualBox

If you experience problems with VirtualBox, the following command might be helpful:

$ alsactl init
Found hardware: "ICH" "SigmaTel STAC9700,83,84" "AC97a:83847600" "0x8086" "0x0000"
Hardware is initialized using a generic method

You might need to activate the ALSA output in your audio software as well.

Sound skipping while using dynamic CPU frequency scaling

Some combinations of ALSA drivers and chipsets may cause audio from all sources to skip when used in combination with a dynamic frequency scaling governor such as ondemand or conservative. Currently, the solution is to switch back to the performance governor.

Refer to the CPU frequency scaling for more information.

Problems with availability to only one user at a time

You might find that only one user can use the dmixer at a time. This is probably ok for most, but for those who run mpd as a separate user this poses a problem. When mpd is playing a normal user cannot play sounds though the dmixer. While it is quite possible to just run mpd under a user's login account, another solution has been found. Adding the line ipc_key_add_uid 0 to the pcm.dmixer block disables this locking. The following is a snippet from asound.conf, the rest is the same as above.

pcm.dmixer {
    type dmix
    ipc_key 1024
    ipc_key_add_uid 0
    ipc_perm 0660
slave {

Simultaneous playback problems

If you are having problems with simultaneous playback, and if PulseAudio is installed, its default configuration is set to "hijack" the soundcard. Some users of ALSA may not want to use PulseAudio and are quite content with their current ALSA settings. One fix is to edit /etc/asound.conf and comment out the following lines:

# Use PulseAudio by default
pcm.!default {
    type pulse
    fallback "sysdefault"
    hint {
        show on
        description "Default ALSA Output (currently PulseAudio Sound Server)"

Commenting the following out also may help:

ctl.!default {
    type pulse
    fallback "sysdefault"

This may be a much simpler solution than completely uninstalling PulseAudio.

Effectively, here is an example of a working /etc/asound.conf:

pcm.dmixer {
    type dmix
    ipc_key 1024
    ipc_key_add_uid 0
    ipc_perm 0660
pcm.dsp {
    type plug
    slave.pcm "dmix"
Note: This /etc/asound.conf file was intended for and used successfully with a global MPD configuration. See this section on multiple users

Random lack of sound on startup

You can quickly test sound by running speaker-test. If there is no sound, the error message might look something like

 ALSA lib pcm_dmix.c:1022:(snd_pcm_dmix_open) unable to open slave
 Playback open error: -16
 Device or resource busy

If you have no sound on startup, this may be because your system has multiple sound cards, and their order may sometimes change on startup. If this is the case, try setting the default sound card.

If you use mpd and the configuration tips above do not work for you, try reading this instead.


Timidity can be the cause of missing audio. Try running:

$ systemctl status timidity

If it failed, try # killall -9 timidity. If this solves the issue, then you should disable the timidity daemon to be started at boot.

Specific program problems

For other programs who insist on their own audio setup, e.g., XMMS or MPlayer, you would need to set their specific options.

For MPlayer or mpv, add the following line to the respective configuration file:


For XMMS2, go into their options and make sure the sound driver is set to ALSA, not oss.

To do this in XMMS:

  • Open XMMS
    • Options > Preferences.
    • Choose the ALSA output plugin.

For applications which do not provide a ALSA output, you can use aoss from the alsa-oss package. To use aoss, when you run the program, prefix it with aoss, e.g.:

aoss realplay

pcm.!default{ ... } doesnt work for me anymore. but this does:

pcm.default pcm.dmixer

Model settings

Although ALSA detects your soundcard through the BIOS, at times ALSA may not be able to recognize your model type. The soundcard chip can be found in alsamixer (e.g. ALC662) and the model can be set in /etc/modprobe.d/modprobe.conf or /etc/modprobe.d/sound.conf. For example:

options snd-hda-intel model=MODEL

There are other model settings too. For most cases ALSA defaults will do. If you want to look at more specific settings for your soundcard take a look at the ALSA Soundcard List find your model, then Details, then look at the "Setting up modprobe..." section. Enter these values in /etc/modprobe.d/modprobe.conf. For example, for an Intel AC97 audio:

# ALSA portion
alias char-major-116 snd
alias snd-card-0 snd-intel8x0
# module options should go here

# OSS/Free portion
alias char-major-14 soundcore
alias sound-slot-0 snd-card-0

# card #1
alias sound-service-0-0 snd-mixer-oss
alias sound-service-0-1 snd-seq-oss
alias sound-service-0-3 snd-pcm-oss
alias sound-service-0-8 snd-seq-oss
alias sound-service-0-12 snd-pcm-oss

Conflicting PC speaker

If you are sure nothing is muted, that your drivers are installed correctly, and that your volume is right, but you still do not hear anything, then try adding the following line to /etc/modprobe.d/modprobe.conf:

options snd-NAME-OF-MODULE ac97_quirk=0

The above fix has been observed to work with via82xx

options snd-NAME-OF-MODULE ac97_quirk=1

The above fix has been reported to work with snd-intel8x0

Tango-edit-clear.pngThis article or section needs language, wiki syntax or style improvements. See Help:Style for reference.Tango-edit-clear.png

Reason: Duplicated sections (e.g. for microphone) should be merged with eachother (Discuss in Talk:Advanced Linux Sound Architecture#)

No microphone input

In alsamixer, make sure that all the volume levels are up under recording, and that CAPTURE is toggled active on the microphone (e.g. Mic, Internal Mic) and/or on Capture (in alsamixer, select these items and press space). Try making positive Mic Boost and raising Capture and Digital levels higher; this make make static or distortion, but then you can adjust them back down once you are hearing something when you record

As the pulseaudio wrapper is shown as "default" in alsamixer, you may have to press F6 to select your actual soundcard first. You may also need to enable and increase the volume of Line-in in the Playback section.

To test the microphone, run these commands (see arecord's man page for further information):

$ arecord -d 5 test-mic.wav
$ aplay test-mic.wav

If all fails, you may want to eliminate hardware failure by testing the microphone with a different device.

For at least some computers, muting a microphone (MM) simply means its input does not go immediately to the speakers. It still receives input.

Many Dell laptops need "-dmic" to be appended to the model name in /etc/modprobe.d/modprobe.conf:

 options snd-hda-intel model=dell-m6-dmic

Some programs use try to use OSS as the main input software. If you have enabled the snd_pcm_oss, snd_mixer_oss or snd_seq_oss kernel modules previously (they are not loaded by default), try unloading them.

See also:

Crackling in microphone

If you are getting a crackling or popping from your microphone that cannot be resolved with ALSA settings or cleaning your microphone jack, try adding the following line to /etc/modprobe.d/modprobe.conf:

options snd-hda-intel model=MODEL position_fix=3

This option will fix crackling on pure ALSA, but will cause issues to pulseaudio. To let Pulse use these settings effectively, edit /etc/pulse/default.pa and find this line:

load-module module-udev-detect

And change it to this:

load-module module-udev-detect tsched=0

See the DMA-Position Problem in the kernel docs.

Setting the default microphone/capture device

Some applications (Pidgin, Adobe Flash) do not provide an option to change the capture device. It becomes a problem if your microphone is on a separate device (e.g. USB webcam or microphone) than your internal sound card. To change only the default capture device, leaving the default playback device as is, you can modify your ~/.asoundrc file to include the following:

    type hw
    card U0x46d0x81d

    type asym
        type plug
        slave.pcm "dmix"
        type plug
        slave.pcm "usb"

Replace "U0x46d0x81d" with your capture device's card name in ALSA. You can use arecord -L to list all the capture devices detected by ALSA.

Internal microphone not working

First make sure the volume is enabled under the Capture view in alsamixer. Add the following to /etc/modprobe.d/snd-hda-intel.conf:

options snd-hda-intel enable_msi=1

Then reload the module:

# rmmod snd-hda-intel && modprobe snd-hda-intel

Now there should be an additional input under the previously mentioned Capture view.

No sound with onboard Intel sound card

There may be a problem with two conflicting modules loaded, namely snd-intel8x0 and snd-intel8x0m. In this case, blacklist snd-intel8x0m:

blacklist snd-intel8x0m

Muting the "External Amplifier" in alsamixer or amixer may also help. See the ALSA wiki.

Unmuting the "Mix" setting in the mixer might help, also.

No headphone sound with onboard Intel sound card

With Intel Corporation 82801 I (ICH9 Family) HD Audio Controller on laptop, you may need to add this line to modprobe or sound.conf:

options snd-hda-intel model=model

Where model is any one of the following:

  • dell-m6
  • dell-vostro
  • generic
  • laptop
  • laptop-hpsense
  • olpc-xo-1_5
Note: It may be necessary to put this "options" line below (after) any "alias" lines about your card.

You can see all the available models in the kernel documentation. For example here, but check that it is the correct version of that document for your kernel version.

A list of available models is also available here. To know your chip name type the following command (with * being corrected to match your files). Note that some chips could have been renamed and do not directly match the available ones in the file.

$ grep Codec /proc/asound/card*/codec*

Note that there is a high chance none of the input devices (all internal and external mics) will work if you choose to do this, so it is either your headphones or your mic. Please report to ALSA if you are affected by this bug.

And also, if you have problems getting beeps to work (pcspkr):

options snd-hda-intel model=$model enable=1 index=0

No sound when S/PDIF video card is installed

Discover available modules and their order:

$ cat /proc/asound/modules
 0 snd_hda_intel
 1 snd_ca0106

Disable the undesired video card audio codec in /etc/modprobe.d/modprobe.conf:

install snd_hda_intel /bin/false

If both devices use the same module then we can use the enable parameter from snd-hda-intel module; it is an array of booleans that can enable/disable the desired sound card.

options snd-hda-intel enable=1,0

Poor sound quality or clipping

If you experience poor sound quality, try setting the PCM volume (in alsamixer) to a level such that gain is 0.

If snd-usb-audio driver has been loaded, you could try to enable softvol in /etc/asound.conf file. Example configuration for the first audio device:

pcm.!default {
    type plug
    slave.pcm "softvol"
pcm.dmixer {
    type dmix
    ipc_key 1024
    slave {
        pcm "hw:0"
        period_size 4096
        buffer_size 131072
        rate 50000
    bindings {
        0 0
        1 1
pcm.dsnooper {
    type dsnoop
    ipc_key 1024
    slave {
        pcm "hw:0"
        channels 2
        period_size 4096
        buffer_size 131072
        rate 50000
    bindings {
        0 0
        1 1
pcm.softvol {
  type softvol
  slave { pcm "dmixer" }
  control {
    name "Master"
    card 0
ctl.!default {
  type hw
  card 0
ctl.softvol {
  type hw
  card 0
ctl.dmixer {
  type hw
  card 0

Pops when starting and stopping playback

Some modules (e.g. snd_ac97_codec and snd_hda_intel) can power off your sound card when not in use. This can make an audible noise (like a crack/pop/scratch) when turning on/off your sound card. Sometimes even when move the slider volume, or open and close windows (KDE4). If you find this annoying try modinfo snd_MY_MODULE, and look for a module option that adjusts or disables this feature.

Example: disable the power saving mode and solve cracking sound trough speakers problem, using snd_hda_intel add in /etc/modprobe.d/modprobe.conf:

options snd_hda_intel power_save=0


options snd_hda_intel power_save=0 power_save_controller=N

You can also try it with modprobe snd_hda_intel power_save=0 before.

You may also have to unmute the 'Line' ALSA channel for this to work. Any value will do (other than '0' or something too high).

Example: on an onboard VIA VT1708S (using the snd_hda_intel module) these cracks occured even though 'power_save' was set to 0. Unmuting the 'Line' channel and setting a value of '1' solved the problem.

Source: https://www.kernel.org/doc/Documentation/sound/alsa/powersave.txt

If you use a laptop, pm-utils will change power_save back to 1 when you go onto battery power even if you disable power saving in /etc/modprobe.d. Disable this for pm-utils by disabling the script that makes the change (see Disabling a hook for more information):

# touch /etc/pm/power.d/intel-audio-powersave

S/PDIF output does not work

If the optical/coaxial digital output of your motherboard/sound card is not working or stopped working, and have already enabled and unmuted it in alsamixer, try running the following:

$ iecset audio on

You can also put this command in an enabled systemd service as it sometimes it may stop working after a reboot.


Output does not work

The procedure described below can be used to test HDMI audio. Before proceeding, make sure you have enabled and unmuted the output with alsamixer.

Note: If you are using an AMD card a necessary kernel module is disabled by default. See ATI#HDMI_audio.

Connect your PC to the Display via HDMI cable and enable the display with xrandr.

Use aplay -l to get the discover the card and device number. For example:

$ aplay -l
**** List of PLAYBACK Hardware Devices ****
card 0: SB [HDA ATI SB], device 0: ALC892 Analog [ALC892 Analog]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 0: SB [HDA ATI SB], device 1: ALC892 Digital [ALC892 Digital]
  Subdevices: 1/1
  Subdevice #0: subdevice #0
card 1: Generic [HD-Audio Generic], device 3: HDMI 0 [HDMI 0]
  Subdevices: 1/1
  Subdevice #0: subdevice #0

Send sound to the device. Following the example in the previous step, you would send sound to card 1, device 3:

$ aplay -D plughw:1,3 /usr/share/sounds/alsa/Front_Center.wav

If aplay does not output any errors, but still no sound is heared, "reboot" the receiver, monitor or tv set. Since the HDMI interface executes a handshake on connection, it might have noticed before that there was no audio stream embedded, and disabled audio decoding. If you are using a standalone window manager, you may need to have sound playing while plugging in the HDMI cable.

mplay and other application could be configured to use special HDMI device as audio output. But flashplugin could only use default device. The following method is used to override default device. But you need to change it back when your TV is disconnected from HDMI port.

If the test is successful, create or edit your ~/.asoundrc file to set HDMI as the default audio device.

pcm.!default {
    type hw
    card 0
    device 3

Or if the above configuration does not work try:

defaults.pcm.card 0
defaults.pcm.device 3
defaults.ctl.card 0

Multi-channel PCM output does not work (Intel)

As of Linux 3.1 multi-channel PCM output through HDMI with a Intel card (Intel Eaglelake, IbexPeak/Ironlake,SandyBridge/CougarPoint and IvyBridge/PantherPoint) is not yet supported. Support for it has been recently added and expected to be available in Linux 3.2. To make it work in Linux 3.1 you need to apply the following patches:

5.1 sound goes to wrong speakers

Sound can be redirected to the intended speakers using ALSA's remap function. To do so, add the following to /etc/asound.conf:

pcm.!hdmi-remap {
    type asym
    playback.pcm {
        type plug
        slave.pcm "remap-surround51"

pcm.!remap-surround51 {
    type route
    slave.pcm "hw:0,3"
    ttable {
        0.0= 1
        1.1= 1
        2.4= 1
        3.5= 1
        4.2= 1
        5.3= 1

HP TX2500

Add these 2 lines into /etc/modprobe.d/modprobe.conf:

options snd-cmipci mpu_port=0x330 fm_port=0x388
options snd-hda-intel index=0 model=toshiba position_fix=1
options snd-hda-intel model=hp (works for tx2000cto)

Sound skipping during playback

Run alsamixer, and if channels exist for nonexistent output devices then disable them (e.g. alsamixer showing a center speaker but you not having one).

Crackling sound with USB devices

Try modifying the parameters of the snd-usb-audio kernel_module for minimal latency via modprobe.conf.

options snd-usb-audio nrpacks=1

For more information, see the following: Tuning USB devices for minimal latency

Error 'Unknown hardware' appears after a kernel update

The following messages may be displayed during ALSA's initialization:

Unknown hardware "foo" "bar" ...
Hardware is initialized using a guess method
/usr/bin/alsactl: set_control:nnnn:failed to obtain info for control #mm (No such file or directory)


Found hardware: "HDA-Intel" "VIA VT1705" "HDA:11064397,18490397,00100000" "0x1849" "0x0397"
Hardware is initialized using a generic method
/usr/bin/alsactl: set_control:1328: failed to obtain info for control #1 (No such file or directory)
/usr/bin/alsactl: set_control:1328: failed to obtain info for control #2 (No such file or directory)
/usr/bin/alsactl: set_control:1328: failed to obtain info for control #25 (No such file or directory)
/usr/bin/alsactl: set_control:1328: failed to obtain info for control #26 (No such file or directory)

Simply store ALSA mixer settings again:

# alsactl -f /var/lib/alsa/asound.state store

It may be necessary configure ALSA again with alsamixer

HDA analyzer

If the mappings to your audio pins(plugs) do not correspond but ALSA works fine, you could try HDA Analyzer -- a pyGTK2 GUI for HD-audio control can be found at the ALSA wiki. Try tweaking the Widget Control section of the PIN nodes, to make microphones IN and headphone jacks OUT. Referring to the Config Defaults heading is a good idea.

  • The script is done by such way that it is incompatible with Python 3 (which is now shipped with Arch Linux) but tries to use it. The workaround is: open "run.py", find all occurrences of "python" (2 occurrences - one on the first line, and the second on the last line) and replace them all by "python2". Then "python2 run.py" The script requires root privileges.

No sound with SDL

If you get no sound using SDL based applications, try setting the environment variable SDL_AUDIODRIVER to alsa.

Low sound workaround

If you are facing low sound even after maxing out your speakers/headphones, you can give the softvol plugin a try. Add the following to /etc/asound.conf.

pcm.!default {
    type plug
    slave.pcm "softvol"

pcm.softvol {
    type softvol
    slave {
        pcm "dmix"
    control {
        name "Pre-Amp"
        card 0
    min_dB -5.0
    max_dB 20.0
    resolution 6
Note: You will probably have to restart the computer, as restarting the alsa daemon did not load the new configuration for me. Also, if the configuration does not work even after restarting, try changing plug with hw in the above configuration.

After the changes are loaded successfully, you will see a Pre-Amp section in alsamixer. You can adjust the levels there.

  • Setting a high value for Pre-Amp can cause sound distortion, so adjust it according to the level that suits you.
  • Some audio codecs may need to have settings adjusted in the HDA Analyzer (see above) in order to achieve proper volume without distortion. Checking the HP option under widget control in the Playback Switch (Node[0x14] PIN in the ALC892 codec, for instance) can sometimes improve audio quality and volume significantly.

Popping sound after resuming from suspension

You might hear a popping sound after resuming the computer from suspension. This can be fixed by editing /etc/pm/sleep.d/90alsa and removing the line that says aplay -d 1 /dev/zero

Output is muted after reboot

Run the following command:

# alsactl restore

If the problem persists, verify that the Auto-Mute option in alsamixer is set to Disabled.

Volume is too low

Run |alsamixer and try to increase the value of the sliders, unmuting channels if necessary. Note that if you have many sliders, you may have to scroll to the right to see any missing sliders.

If all the sliders are maxed out, and the volume is still too low, you can try running the script at http://www.alsa-project.org/hda-analyzer.py to reset your codec settings:

$ wget http://www.alsa-project.org/hda-analyzer.py
$ su -c 'python2 hda-analyzer.py'

The script assumes the python binary refers to Python 2, which is incorrect on Arch by default. To avoid this issue run the following command:

$ sed -i 's/python %s/python2 %s/' hda-analyzer_run.py

Close the analyzer, and when prompted as to whether you want to reset the codecs, say "yes".

If the volume is *still* too low, run alsamixer again: resetting the codecs may have caused new sliders to become enabled, and some of them may be set to a low value.

Verifying output parameters

Check the contents of /proc/asound/cardX/pcmYp/subZ/hw_params, where X, Y, and Z are system dependent. In order to find this file, execute the following command while outputting anything via ALSA:

$ find /proc/asound/ -name hw_params | xargs -I FILE grep -v -l "closed" FILE | grep '/proc/asound/card./pcm.p/sub./hw_params'

If nothing is playing there should be no results.

Here is an example output for audio with a bit depth of 24 bits and a sampling frequency of 44.1 kilohertz:

cat /proc/asound/card1/pcm0p/sub0/hw_params
format: S24_3LE
subformat: STD
channels: 2
rate: 44100 (44100/1)
period_size: 5513
buffer_size: 22050

More info is available in the ALSA documentation.

Removing old ALSA state file (asound.state)

The alsa-utils package provides alsa-store.service which automatically stores the current ALSA state to /var/lib/alsa/asound.state upon system shutdown. This can be problematic for users who are trying to reset their current ALSA state as the asound.state file will be recreated with the current state upon every shutdown (e.g., attempting to remove user-defined channels from the mixer). The alsa-store.service service may be temporarily disabled by creating the following empty file:

# mkdir -p /etc/alsa
# touch /etc/alsa/state-daemon.conf

The presence of state-daemon.conf prevents alsa-store.service from saving asound.state during shutdown. After disabling this service, the asound.state file may be removed as such:

# rm /var/lib/alsa/asound.state

After rebooting, the previous ALSA state should be lost and the current state should be reset to defaults. Re-enable alsa-store.service by deleting the condition file we created:

# rm /etc/alsa/state-daemon.conf

On the next shutdown, the asound.state file should be recreated with ALSA defaults. The file may also be generated immediately using:

# alsactl store

No sound in applications that use OpenAL

OpenAL defaults to pulseaudio, to change the order, add the following configuration to /etc/openal/alsoft.conf:


Tips and tricks

Hot-plugging a USB sound card

The following udev rule can be used to automatically make a USB sound card the primary output device when the card is plugged in. To do so, create the following file with the specified contents and.

KERNEL=="pcmC[D0-9cp]*", ACTION=="add", PROGRAM="/usr/bin/sh -c 'K=%k; K=$${K#pcmC}; K=$${K%%D*}; echo defaults.ctl.card $$K > /etc/asound.conf; echo defaults.pcm.card $$K >>/etc/asound.conf'"
KERNEL=="pcmC[D0-9cp]*", ACTION=="remove", PROGRAM="/usr/bin/sh -c 'echo defaults.ctl.card 0 > /etc/asound.conf; echo defaults.pcm.card 0 >>/etc/asound.conf'"

For information regarding loading udev rules see the following: Udev#Loading_new_rules

See also