前几日淘了块创新的Audigy2 ZS Notebook声卡,声卡的驱动还算不错,只要内核中开启pcmcia声卡支持为加上emu10k1的模块就好了.但之后设置的问题也来了,目标非常简单,希望声卡插进、拔出的时候,alsa的设置能相应自动变化.

我们可以利用udev来帮我们实现这个目标.

我们需要两个脚本和两个内容略为不同的asound.conf.

先看看两个声卡是啥名字:

代码:

ls -l /proc/asound/

我的机子上是:

代码:

lrwxrwxrwx 1 root root 5 2009-03-16 17:27 Audigy2 -> card2

dr-xr-xr-x 9 root root 0 2009-03-16 17:27 card0

dr-xr-xr-x 5 root root 0 2009-03-16 17:27 card1

dr-xr-xr-x 8 root root 0 2009-03-16 17:27 card2

-r--r--r-- 1 root root 0 2009-03-16 17:27 cards

-r--r--r-- 1 root root 0 2009-03-16 17:27 devices

-r--r--r-- 1 root root 0 2009-03-16 17:27 hwdep

lrwxrwxrwx 1 root root 5 2009-03-16 17:27 I82801DBICH4 -> card0

lrwxrwxrwx 1 root root 5 2009-03-16 17:27 Modem -> card1

dr-xr-xr-x 2 root root 0 2009-03-16 17:27 oss

-r--r--r-- 1 root root 0 2009-03-16 17:27 pcm

dr-xr-xr-x 2 root root 0 2009-03-16 17:27 seq

-r--r--r-- 1 root root 0 2009-03-16 17:27 timers

-r--r--r-- 1 root root 0 2009-03-16 17:27 version

其中板载声卡是 I82801DBICH4,外置的pcmcia声卡是Audigy2.要把名字记下,待会我们需要它们.

接下来要写两个asound.conf来指定外置声卡插拔情况下的alsa设置.先

代码:

mkdir /etc/alsa

cp /etc/asound.conf /etc/alsa/asound.ac97

mv /etc/asound.conf /etc/alsa/asound.audigy2

,asound.ac97是没有插外置声卡时用的,asound.audigy2就是外置声卡插上时用的.

然后在asound.ac97中加入:

代码:

# onboard

pcm.ac97 {

type plug

slave.pcm "dmix"

}

ctl.ac97 {

type hw

card I82801DBICH4

}

pcm.!default pcm.ac97

ctl.!default ctl.ac97

在asound.audigy2中加入

代码:

# pcmcia

pcm.audigy2hw {

type hw

card Audigy2

}

pcm.audigy2 {

type plug

slave.pcm "audigy2hw"

}

ctl.audigy2 {

type hw

card Audigy2

}

# onboard

pcm.ac97 {

type plug

slave.pcm "dmix"

}

ctl.ac97 {

type hw

card I82801DBICH4

}

pcm.!default pcm.audigy2

ctl.!default ctl.audigy2

要注意的是,xxx.ac97中声卡名是我板载声卡的名字,xxx.audigy2则是我外置声卡的名字,你需要替换成你自己的声卡.

这样在不同情况下,我们只要 ln -sf /etc/alsa/asound.xxx /etc/asound.conf就OK了.

接下来我们需要一个脚本在适当的时候改变上面的链接,脚本是从ALSA Wiki上处理usb外接声卡转换的脚本改写而来的,脚本switch_default_sound:

代码:

#!/bin/bash

# Name: switch_default_sound

# Last modified: 2009.03.16

# use to switch the asound.conf file

#

# because we hope to restore the sound card levels as well, but udev

# will hang when waiting for this script to finish,

# so we call restore_alsa_sound as daemon, which wait and try to restore

# sound level after udev finish, and alsa make the card ready.

usage(){

echo usage: `basename $0` [`ls /etc/alsa/ -1 | sed -e 's:/etc/alsa/::g' -e 's/asound.//g'`]

}

if [ -n $1 ] && [ -e /etc/alsa/asound."$1" ];

then

ln -sf /etc/alsa/asound."$1" /etc/asound.conf

# we need to call restore_alsa_sound as daemon, so that, we can

# return to udev, and leave rest to restore_alsa_sound

restore_alsa_sound "$1" &

else

usage

fi

这个脚本除了完成asound.conf的链接工作外,还将调用restore_alsa_sound完成音量重置的工作（我还用它重启mpd）.udev在调用switch_default_sound的时候是停在那里等它返回的,而在udev完成设备添加前alsa是得不到声卡信息的,音量恢复就无法完成,而我们在switch_default_sound中用while来等,这样udev就死等在那里了.为了避免这样的问题,我们需要用 daemon的形式调用restore_alsa_sound,把剩下的工作交给它.

代码:

#!/usr/bin/env python

# Name: restore_alsa_sound

# Last Modified: 2009.03.16

# use to retore the alsa sound level

from sys import argv

from os import system as os_system

from time import sleep

if argv[1] == "audigy2":

retry = 0

while ( retry <= 10):

if os_system ("alsactl restore Audigy2") == 0:

break

sleep(0.5)

os_system ("/etc/init.d/mpd restart")

注意, alsactl restore

Audigy2中的Audigy2是我外置声卡的名字,你得换成你自己的.另外,如果你没有用mpd的话,就把一行去掉.

接着,把switch_default_sound和restore_alsa_sound放到 /usr/local/sbin下,并chmod 755 它.

,我们需要让udev在插入和拔出声卡的时候调用switch_default_sound,就大功告成了.,我们要写一个udev的rule为我们服务.

先用lspci找到声卡的PCI_ID

代码:

lspci -nn

比如我的是

代码:

02:00.0 Multimedia audio controller [0401]: Creative Labs SB0400 Audigy2 Value [1102:0008]

,其中1102:0008就是了.

接着,我们需要找出相应的udev信息.先把udev的信息导到文件udev.txt中

代码:

udevadm info --export-db > udev.txt

然后找出1102:0008

代码:

grep -C10 1102:0008 udev.txt

在我的电脑上结果是:

代码:

E: PCI_SLOT_NAME=0000:01:0a.0

E: MODALIAS=pci:v00001180d00000476sv000010CFsd00001254bc06sc07i00

P: /devices/pci0000:00/0000:00:1e.0/0000:01:0a.0/0000:02:00.0

E: UDEV_LOG=3

E: DEVPATH=/devices/pci0000:00/0000:00:1e.0/0000:01:0a.0/0000:02:00.0

E: DRIVER=EMU10K1_Audigy

E: PHYSDEVBUS=pci

E: PHYSDEVDRIVER=EMU10K1_Audigy

E: PCI_CLASS=40100

E: PCI_ID=1102:0008

E: PCI_SUBSYS_ID=1102:2001

E: PCI_SLOT_NAME=0000:02:00.0

E: MODALIAS=pci:v00001102d00000008sv00001102sd00002001bc04sc01i00

P: /devices/pci0000:00/0000:00:1e.0/0000:01:0a.1

E: UDEV_LOG=3

E: DEVPATH=/devices/pci0000:00/0000:00:1e.0/0000:01:0a.1

E: DRIVER=yenta_cardbus

E: PHYSDEVBUS=pci

E: PHYSDEVDRIVER=yenta_cardbus

从中可以看到声卡的PCI_SUBSYS_ID是1102:2001.现在我们就有了足够的信息来写udev的rules了.

在/etc/udev/rules.d下建立一个文件75-alsa_conf.rules

代码:

# udev rules file for my Audigy2 ZS Notebook devices (for udev 0.98 version)

# call /usr/local/sbin/switch_default_sound to change the link

# of /etc/asound.conf

#

#

ENV{PCI_ID}=="1102:0008", ENV{PCI_SUBSYS_ID}=="1102:2001",ACTION=="add", RUN ="/usr/local/sbin/switch_default_sound audigy2"

ENV{PCI_ID}=="1102:0008", ENV{PCI_SUBSYS_ID}=="1102:2001",ACTION=="remove", RUN ="/usr/local/sbin/switch_default_sound ac97"

在这个rules中,我们要求udev在PCI_ID为1102:0008且PCI_SUBSYS_ID为1102:2001的设备(就是处置的那块声卡)要被插入时,即ACTION为add时,调用"/usr/local/sbin/switch_default_sound audigy2",在声卡被移除时,即ACTION为remove时调用"/usr/local/sbin/switch_default_sound ac97".

但为了能有效地restore声音,你需要在第一次插入声卡并设置好音量后手动运行一下alsactl store Audigy2.

另外,插拔声卡的时候要确认你没有在放音乐哦

PS：

1.PCI_SUBSYS_ID查不到也不要紧,只要把75-alsa_conf.rules改为：

....................

ENV{PCI_ID}=="1102:0008",ACTION=="add", RUN ="/usr/local/sbin/switch_default_sound audigy2"

ENV{PCI_ID}=="1102:0008",ACTION=="remove", RUN ="/usr/local/sbin/switch_default_sound ac97"

.....................

2.Ubuntu 系统 －》首选项－》音效－》把设备页默认的设备全部改为外置声卡

3./etc/asound.conf不存在,自己创建一个即可.