ALSA busted in 2.5.69

ALSA busted in 2.5.69

[Torrey Hoffman]

ALSA isn't working for me in 2.5.69.  It appears to be because
/proc/asound/dev is missing the control devices.

lsmod says I have:

Module                  Size  Used by
snd_cmipci             22168  -
snd_pcm                59968  -
snd_page_alloc          4140  -
snd_opl3_lib            5896  -
snd_timer              13888  -
snd_hwdep               4096  -
snd_mpu401_uart         3144  -
snd_rawmidi            12768  -
snd                    29156  -
soundcore               3296  -

and no errors were reported while loading those modules,
but "ls -l /proc/asound/dev" shows only:
total 0
crw-rw-rw-    1 root     root     116,  33 May  8 16:37 timer

and all the alsa utilities die trying to open other entries under
/dev/snd, which is a symlink to /proc/asound/dev (as set up by the
ALSA script.)

For instance, strace alsamixer shows this at the end:

access("/etc/asound.conf", R_OK)        = -1 ENOENT (No such file or
directory)
access("/root/.asoundrc", R_OK)         = -1 ENOENT (No such file or
directory)
open("/dev/snd/controlC0", O_RDONLY)    = -1 ENOENT (No such file or
directory)
open("/dev/aloadC0", O_RDONLY)          = -1 ENODEV (No such device)
open("/dev/snd/controlC0", O_RDWR)      = -1 ENOENT (No such file or
directory)
open("/dev/snd/controlC0", O_RDONLY)    = -1 ENOENT (No such file or
directory)
open("/dev/aloadC0", O_RDONLY)          = -1 ENODEV (No such device)
open("/dev/snd/controlC0", O_RDWR)      = -1 ENOENT (No such file or
directory)
fstat64(1, {st_mode=S_IFCHR|0620, st_rdev=makedev(136, 0), ...}) = 0
mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1,
0) = 0x40017000
write(1, "\n", 1
)                       = 1
write(2, "alsamixer: function snd_ctl_open"..., 79alsamixer: function
snd_ctl_open failed for default: No such file or directory
) = 79
munmap(0x40017000, 4096)                = 0
exit_group(1)                           = ?

Suggestions???


-- 
Torrey Hoffman <thoffman@arnor.net>

RE: ALSA busted in 2.5.69

[Giuliano Pochini]

On 08-May-2003 Torrey Hoffman wrote:
> ALSA isn't working for me in 2.5.69.  It appears to be because
> /proc/asound/dev is missing the control devices.
> open("/dev/snd/controlC0", O_RDWR)      = -1 ENOENT (No such file or
> directory)
> [...]

If you are not using devfs, you need to create the devices. There is a
script in the ALSA-driver package to do that. Otherwise I can't help
you because I never tried devfs and linux 2.5.x.


Bye.

Re: ALSA busted in 2.5.69

[Jan Dittmer]

Giuliano Pochini wrote:
> On 08-May-2003 Torrey Hoffman wrote:
> 
>>ALSA isn't working for me in 2.5.69.  It appears to be because
>>/proc/asound/dev is missing the control devices.
>>open("/dev/snd/controlC0", O_RDWR)      = -1 ENOENT (No such file or
>>directory)
>>[...]
> 
> 
> If you are not using devfs, you need to create the devices. There is a
> script in the ALSA-driver package to do that. Otherwise I can't help
> you because I never tried devfs and linux 2.5.x.
> 

I've the same problem, using devfs: (.69-mm3)

ds666:/dev/snd# ls -l
total 0
crw-rw----    1 root     audio    116,   0 Jan  1  1970 controlC0
crw-rw----    1 root     audio    116,  32 Jan  1  1970 controlC1
crw-rw----    1 root     audio    116,  64 Jan  1  1970 controlC2
crw-rw----    1 root     audio    116,  96 Jan  1  1970 controlC3
crw-rw----    1 root     audio    116, 128 Jan  1  1970 controlC4
crw-rw----    1 root     audio    116, 160 Jan  1  1970 controlC5
crw-rw----    1 root     audio    116, 192 Jan  1  1970 controlC6
crw-rw----    1 root     audio    116, 224 Jan  1  1970 controlC7

CONFIG_SND=y
CONFIG_SND_SEQUENCER=y
CONFIG_SND_OSSEMUL=y
CONFIG_SND_MIXER_OSS=y
CONFIG_SND_PCM_OSS=y
CONFIG_SND_SEQUENCER_OSS=y
CONFIG_SND_RTCTIMER=y
CONFIG_SND_EMU10K1=y
CONFIG_SND_VIA82XX=m

CONFIG_DEVFS_FS=y
CONFIG_DEVFS_MOUNT=y
CONFIG_DEVPTS_FS=y

RE: ALSA busted in 2.5.69

[Torrey Hoffman]

On Fri, 2003-05-09 at 01:09, Giuliano Pochini wrote:
> On 08-May-2003 Torrey Hoffman wrote:
> > ALSA isn't working for me in 2.5.69.  It appears to be because
> > /proc/asound/dev is missing the control devices.
...
> If you are not using devfs, you need to create the devices. There is a
> script in the ALSA-driver package to do that. Otherwise I can't help
> you because I never tried devfs and linux 2.5.x.

No.  /dev/snd is a symbolic link to /proc/asound/dev,
and that symbolic link was created by the script you mention.
(I am not using devfs.)

So the missing "/dev/snd/controlC0" should actually be created
by the ALSA modules in the proc filesystem as
"/proc/asound/dev/controlC0".  But only the timer device is there.

(This works under Red Hat 9's standard kernel with the ALSA 0.9.3
drivers, by the way.)

Torrey

Re: ALSA busted in 2.5.69

[Jose Luis Domingo Lopez]

On Thursday, 08 May 2003, at 16:39:51 -0700,
Torrey Hoffman wrote:

> ALSA isn't working for me in 2.5.69.  It appears to be because
> /proc/asound/dev is missing the control devices.
> 
I have been using ALSA for my soundcards for some days, and today I
booted into 2.5.69, and everything still works.

I used the same .config file in both kernel versions, and did a "make
oldconfig" to adapt the older to the newer version.

> lsmod says I have:
> 
Module                  Size  Used by
cls_u32                 6596  1 [unsafe]
sch_ingress             3264  1 [unsafe]
ipt_LOG                 4416  1 
ipt_limit               1728  1 
ipt_state               1408  1 
iptable_filter          2112  1 
ipt_MASQUERADE          2816  1 
iptable_nat            20052  2 ipt_MASQUERADE
ip_conntrack           24012  3 ipt_state,ipt_MASQUERADE,iptable_nat
ip_tables              14208  6 ipt_LOG,ipt_limit,ipt_state,iptable_filter,ipt_MASQUERADE,iptable_nat
ppp_deflate             5056  0 [unsafe]
zlib_deflate           21240  1 ppp_deflate
zlib_inflate           21056  1 ppp_deflate
bsd_comp                5376  0 [unsafe]
ppp_async               9280  1 [unsafe]
tuner                  14340  1 [unsafe]
tvaudio                19264  0 
bttv                   92384  1 
video_buf              15616  1 bttv
i2c_algo_bit            9288  1 bttv
v4l2_common             3776  1 bttv
videodev                8192  1 bttv
ppp_generic            24200  5 ppp_deflate,bsd_comp,ppp_async
slhc                    6336  1 ppp_generic
parport_pc             18784  1 
lp                      8480  0 
parport                35456  2 parport_pc,lp
snd_seq_oss            33408  0 
snd_seq_midi_event      5632  1 snd_seq_oss
snd_seq                54544  4 snd_seq_oss,snd_seq_midi_event
snd_pcm_oss            57220  0 
snd_mixer_oss          18880  4 snd_pcm_oss
snd_via82xx            16260  1 
snd_ymfpci             47296  2 
snd_pcm                89536  3 snd_pcm_oss,snd_via82xx,snd_ymfpci
snd_ac97_codec         46532  2 snd_via82xx,snd_ymfpci
snd_opl3_lib            9728  1 snd_ymfpci
snd_timer              21632  3 snd_seq,snd_pcm,snd_opl3_lib
snd_hwdep               7360  1 snd_opl3_lib
snd_page_alloc          8388  3 snd_via82xx,snd_ymfpci,snd_pcm
snd_mpu401_uart         5952  2 snd_via82xx,snd_ymfpci
snd_rawmidi            20736  1 snd_mpu401_uart
snd_seq_device          6916  4 snd_seq_oss,snd_seq,snd_opl3_lib,snd_rawmidi
md5                     3712  1 
snd                    49220  15 snd_seq_oss,snd_seq_midi_event,snd_seq,snd_pcm_oss,snd_mixer_oss,snd_via82xx,snd_ymfpci,snd_pcm,snd_ac97_codec,snd_opl3_lib,snd_timer,snd_hwdep,snd_mpu401_uart,snd_rawmidi,snd_seq_device
soundcore               6528  2 bttv,snd
ipv6                  197792  7 
8139too                18176  1 
mii                     4096  1 8139too
crc32                   3840  1 8139too
xfs                   541108  1 
dm_mod                 31680  6 
md                     40576  0 
i2c_isa                 1600  0 
i2c_core               18820  5 tuner,tvaudio,bttv,i2c_algo_bit,i2c_isa
rtc                     9572  0 

> but "ls -l /proc/asound/dev" shows only:
>

total 0
crw-rw-rw-    1 root     root     116,   0 2003-05-09 21:36 controlC0
crw-rw-rw-    1 root     root     116,  32 2003-05-09 21:36 controlC1
crw-rw-rw-    1 root     root     116,  24 2003-05-09 21:36 pcmC0D0c
crw-rw-rw-    1 root     root     116,  16 2003-05-09 21:36 pcmC0D0p
crw-rw-rw-    1 root     root     116,  17 2003-05-09 21:36 pcmC0D1p
crw-rw-rw-    1 root     root     116,  18 2003-05-09 21:36 pcmC0D2p
crw-rw-rw-    1 root     root     116,  27 2003-05-09 21:36 pcmC0D3c
crw-rw-rw-    1 root     root     116,  56 2003-05-09 21:36 pcmC1D0c
crw-rw-rw-    1 root     root     116,  48 2003-05-09 21:36 pcmC1D0p
crw-rw-rw-    1 root     root     116,  49 2003-05-09 21:36 pcmC1D1p
crw-rw-rw-    1 root     root     116,   1 2003-05-09 21:36 seq
crw-rw-rw-    1 root     root     116,  33 2003-05-09 21:36 timer

Hope this helps.

-- 
Jose Luis Domingo Lopez
Linux Registered User #189436     Debian Linux Sid (Linux 2.5.69)

Re: ALSA busted in 2.5.69

[David van Hoose]

Norbert Wolff wrote:
> On 09 May 2003 12:12:10 -0700
> Andy Pfiffer <andyp@osdl.org> wrote:
> 
> 
>>I'm not using devfs, and I've had no luck getting ALSA to work on my
>>i810-audio system.  OSS works fine.
>>
>>Is there a step-by-step writeup available for morons like me that
>>haven't gotten ALSA working?
> 
> 
> Hi Andy !
> 
> The Problem seems to be that ALSA has moved their devices some weeks ago.
> In the alsa-driver-0.9.3a-Package (ftp://ftp.alsa-project.org) is a script
> called snddevices (attached) which creates the needed devices and links.
> 
> Execute it as root and all should be fine ...
> 
> Maybe this Script should be distributed with the Kernel too ?

Will I have any problems if I run this script even though I'm using ALSA 
0.9.2 with 2.4.21-rc2 and whatever version is in 2.5.69-bk4?

Thanks,
David

Re: ALSA busted in 2.5.69

[David Ford]

Shrug :)

I use devfs, all is magic.  All is [nearly always] correct.

David

Norbert Wolff wrote:

>On 09 May 2003 12:12:10 -0700
>Andy Pfiffer <andyp@osdl.org> wrote:
>
>  
>
>>I'm not using devfs, and I've had no luck getting ALSA to work on my
>>i810-audio system.  OSS works fine.
>>
>>Is there a step-by-step writeup available for morons like me that
>>haven't gotten ALSA working?
>>    
>>
>
>Hi Andy !
>
>The Problem seems to be that ALSA has moved their devices some weeks ago.
>In the alsa-driver-0.9.3a-Package (ftp://ftp.alsa-project.org) is a script
>called snddevices (attached) which creates the needed devices and links.
>
>Execute it as root and all should be fine ...
>
>Maybe this Script should be distributed with the Kernel too ?
>
>Regards,
>
>	Norbert
>
>
>  
>

Re: ALSA busted in 2.5.69

[Norbert Wolff]

[-- Attachment #1: Type: text/plain, Size: 743 bytes --]

On 09 May 2003 12:12:10 -0700
Andy Pfiffer <andyp@osdl.org> wrote:

> I'm not using devfs, and I've had no luck getting ALSA to work on my
> i810-audio system.  OSS works fine.
> 
> Is there a step-by-step writeup available for morons like me that
> haven't gotten ALSA working?

Hi Andy !

The Problem seems to be that ALSA has moved their devices some weeks ago.
In the alsa-driver-0.9.3a-Package (ftp://ftp.alsa-project.org) is a script
called snddevices (attached) which creates the needed devices and links.

Execute it as root and all should be fine ...

Maybe this Script should be distributed with the Kernel too ?

Regards,

	Norbert


-- 
 Norbert Wolff
 OpenPGP-Key:
   http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xF13BD6F6


[-- Attachment #2: snddevices --]
[-- Type: application/octet-stream, Size: 3611 bytes --]

#!/bin/bash

MAJOR=116
OSSMAJOR=14
MAX_CARDS=4
PERM=666
STATIC=0
OWNER=root.root

if [ "`grep -w -E "^audio" /etc/group`x" != x ]; then
  PERM=660
  OWNER=root.audio
fi

function create_odevice () {
  rm -f $1
  echo -n "Creating $1..."
  mknod -m $PERM $1 c $OSSMAJOR $2
  chown $OWNER $1
  echo " done"
}

function create_odevices () {
  tmp=0
  tmp1=0
  rm -f $1 $1?
  echo -n "Creating $1?..."
  while [ $tmp1 -lt $MAX_CARDS ]; do
    minor=$[ $2 + $tmp ]
    mknod -m $PERM $1$tmp1 c $OSSMAJOR $minor
    chown $OWNER $1$tmp1
    tmp=$[ $tmp + 16 ]
    tmp1=$[ $tmp1 + 1 ]
  done
  echo " done"
}

function create_device1 () {
  rm -f $1
  minor=$2
  echo -n "Creating $1..."
  mknod -m $PERM $1 c $MAJOR $minor
  chown $OWNER $1
  echo " done"
}

function create_devices () {
  tmp=0
  rm -f $1 $1?
  echo -n "Creating $1?..."
  while [ $tmp -lt $MAX_CARDS ]; do
    minor=$[ $tmp * 32 ]
    minor=$[ $2 + $minor ]
    mknod -m $PERM "${1}C${tmp}" c $MAJOR $minor
    chown $OWNER "${1}C${tmp}"
    tmp=$[ $tmp + 1 ]
  done
  echo " done"
}

function create_devices2 () {
  tmp=0
  rm -f $1 $1?
  echo -n "Creating $1??..."
  while [ $tmp -lt $MAX_CARDS ]; do
    tmp1=0
    while [ $tmp1 -lt $3 ]; do
      minor=$[ $tmp * 32 ]
      minor=$[ $2 + $minor + $tmp1 ]
      mknod -m $PERM "${1}C${tmp}D${tmp1}" c $MAJOR $minor
      chown $OWNER "${1}C${tmp}D${tmp1}"
      tmp1=$[ $tmp1 + 1 ]
    done
    tmp=$[ $tmp + 1 ]
  done
  echo " done"
}

function create_devices3 () {
  tmp=0
  rm -f $1 $1?
  echo -n "Creating $1??$4..."
  while [ $tmp -lt $MAX_CARDS ]; do
    tmp1=0
    while [ $tmp1 -lt $3 ]; do
      minor=$[ $tmp * 32 ]
      minor=$[ $2 + $minor + $tmp1 ]
      mknod -m $PERM "${1}C${tmp}D${tmp1}${4}" c $MAJOR $minor
      chown $OWNER "${1}C${tmp}D${tmp1}${4}"
      tmp1=$[ $tmp1 + 1 ]
    done
    tmp=$[ $tmp + 1 ]
  done
  echo " done"
}

if test "$1" = "-?" || test "$1" = "-h" || test "$1" = "--help"; then
  echo "Usage: snddevices [max]"
  exit
fi

if test "$1" = "max"; then
  DSP_MINOR=19
fi

# OSS (Lite) compatible devices...

if test $OSSMAJOR -eq 14; then
  create_odevices /dev/mixer		0
  create_odevice /dev/sequencer		1
  create_odevices /dev/midi		2
  create_odevices /dev/dsp		3
  create_odevices /dev/audio		4
  create_odevice /dev/sndstat		6
  create_odevice /dev/music		8
  create_odevices /dev/dmmidi		9
  create_odevices /dev/dmfm		10
  create_odevices /dev/amixer		11	# alternate mixer
  create_odevices /dev/adsp		12	# alternate dsp
  create_odevices /dev/amidi		13	# alternate midi
  create_odevices /dev/admmidi		14	# alternate direct midi
  # create symlinks
  ln -svf /dev/mixer0 /dev/mixer
  ln -svf /dev/midi0 /dev/midi
  ln -svf /dev/dsp0 /dev/dsp
  ln -svf /dev/audio0 /dev/audio
  ln -svf /dev/music /dev/sequencer2
  ln -svf /dev/adsp0 /dev/adsp
  ln -svf /dev/amidi0 /dev/amidi
fi

# New devices (obsolete now)...

mv -f /dev/sndstat /dev/1sndstat
rm -f /dev/snd*
mv -f /dev/1sndstat /dev/sndstat
if [ -d /dev/snd ]; then
  rm -f /dev/snd/*
  rmdir /dev/snd
fi
if [ $STATIC -gt 0 ]; then
  mkdir -p /dev/snd
  create_devices  /dev/snd/control	0
  create_device1  /dev/snd/seq		1
  create_device1  /dev/snd/timer	33
  create_devices2 /dev/snd/hw		4	4
  create_devices2 /dev/snd/midi		8	8
  create_devices3 /dev/snd/pcm		16	8	p
  create_devices3 /dev/snd/pcm		24	8	c
fi

# New dynamic sound device filesystem

if [ $STATIC -le 0 ]; then
  echo "ALSA dynamic sound device filesystem"
  ln -sfv /proc/asound/dev /dev/snd
fi

# Loader devices

echo "ALSA loader devices"
rm -f /dev/aload*
create_devices  /dev/aload		0
create_device1  /dev/aloadSEQ		1

RE: ALSA busted in 2.5.69

[Downing, Thomas]

-----Original Message-----
From: Andy Pfiffer [mailto:andyp@osdl.org]
> On Fri, 2003-05-09 at 11:14, walt wrote:
>> Torrey Hoffman wrote:
>> > On Fri, 2003-05-09 at 01:09, Giuliano Pochini wrote:
>> > 
>> >>On 08-May-2003 Torrey Hoffman wrote:
>> >>
>> >>>ALSA isn't working for me in 2.5.69.  It appears to be because
>> >>>/proc/asound/dev is missing the control devices.
>> > ...
>> >>If you are not using devfs, you need to create the devices. There is a
>> >>script in the ALSA-driver package to do that. Otherwise I can't help
>> >>you because I never tried devfs and linux 2.5.x.
>> > 
>> > No.  /dev/snd is a symbolic link to /proc/asound/dev,
>> > and that symbolic link was created by the script you mention.
>> > (I am not using devfs.)
>
> I'm not using devfs, and I've had no luck getting ALSA to work on my
> i810-audio system.  OSS works fine.
>
> Is there a step-by-step writeup available for morons like me that
> haven't gotten ALSA working?
>
> Thanks,
> Andy

I have i810 working on my machine -

1.) You may have already done this - but here it is:  Go to
www.alsa-project.org, under "New Users" click on "Supported Soundcars"
On this page, find the Intel i810 listing, and click on "Details".
Follow all instructions _carefully_.  (I created the devices by
hand - this script didn't work correctly for me.)  Make sure that
both the char-major-14 and char-major-116 devices exist. (14 is OSS
and 116 is ALSA.)  Make sure the aliases are correct in modprobe.conf.

2.) After that, (or if you've already done that), use amixer to look
at and set the sound controls for your card.  NOTE:  the i810 driver
has a "gotcha".  To set the volume for the line out jack on you PC
you will probably have to use the "Headphone" mixer control.  That
little curve ball stumped me till somebody else dropped me the hint.

3.) If all that still doesn'te work, post again.

td

Re: ALSA busted in 2.5.69

[David van Hoose]

Andy Pfiffer wrote:
> On Fri, 2003-05-09 at 11:14, walt wrote:
> 
>>Torrey Hoffman wrote:
>>
>>>On Fri, 2003-05-09 at 01:09, Giuliano Pochini wrote:
>>>
>>>
>>>>On 08-May-2003 Torrey Hoffman wrote:
>>>>
>>>>
>>>>>ALSA isn't working for me in 2.5.69.  It appears to be because
>>>>>/proc/asound/dev is missing the control devices.
>>>
>>>...
>>>
>>>>If you are not using devfs, you need to create the devices. There is a
>>>>script in the ALSA-driver package to do that. Otherwise I can't help
>>>>you because I never tried devfs and linux 2.5.x.
>>>
>>>No.  /dev/snd is a symbolic link to /proc/asound/dev,
>>>and that symbolic link was created by the script you mention.
>>>(I am not using devfs.)
> 
> 
> I'm not using devfs, and I've had no luck getting ALSA to work on my
> i810-audio system.  OSS works fine.
> 
> Is there a step-by-step writeup available for morons like me that
> haven't gotten ALSA working?

I have ALSA working under 2.4.21-rc1 using Erik's kernel patch, but I 
cannot get it to work under 2.5.69. I don't know what's wrong. I get 
Audio from my CD-ROM since it is connected to my sound card, but I get 
no sound from the Arts sound server in KDE. I get no errors and no warnings.
I don't understand how it can't work. I'm thinking that the OSS 
emulation is broken. The ALSA version I have for 2.4 is 0.9.2. 0.9.3a is 
in 2.5.69, right? Have any bugs been reported to the ALSA people?

Regards,
David

Re: ALSA busted in 2.5.69

[Andy Pfiffer]

On Fri, 2003-05-09 at 11:14, walt wrote:
> Torrey Hoffman wrote:
> > On Fri, 2003-05-09 at 01:09, Giuliano Pochini wrote:
> > 
> >>On 08-May-2003 Torrey Hoffman wrote:
> >>
> >>>ALSA isn't working for me in 2.5.69.  It appears to be because
> >>>/proc/asound/dev is missing the control devices.
> > ...
> >>If you are not using devfs, you need to create the devices. There is a
> >>script in the ALSA-driver package to do that. Otherwise I can't help
> >>you because I never tried devfs and linux 2.5.x.
> > 
> > No.  /dev/snd is a symbolic link to /proc/asound/dev,
> > and that symbolic link was created by the script you mention.
> > (I am not using devfs.)

I'm not using devfs, and I've had no luck getting ALSA to work on my
i810-audio system.  OSS works fine.

Is there a step-by-step writeup available for morons like me that
haven't gotten ALSA working?

Thanks,
Andy

Re: ALSA busted in 2.5.69

[walt]

Torrey Hoffman wrote:
> On Fri, 2003-05-09 at 01:09, Giuliano Pochini wrote:
> 
>>On 08-May-2003 Torrey Hoffman wrote:
>>
>>>ALSA isn't working for me in 2.5.69.  It appears to be because
>>>/proc/asound/dev is missing the control devices.
> 
> ...
> 
>>If you are not using devfs, you need to create the devices. There is a
>>script in the ALSA-driver package to do that. Otherwise I can't help
>>you because I never tried devfs and linux 2.5.x.
> 
> 
> No.  /dev/snd is a symbolic link to /proc/asound/dev,
> and that symbolic link was created by the script you mention.
> (I am not using devfs.)
> 
> So the missing "/dev/snd/controlC0" should actually be created
> by the ALSA modules in the proc filesystem as
> "/proc/asound/dev/controlC0".  But only the timer device is there.

Warning:  I'm no expert on any of this -- but -- I'm listening to sound
right now using 2.5.69 (with devfs) and this is what I see:

There are NO symlinks from /dev to /proc.  None.

/dev/snd and /dev/sound are real directories -- neither is a symlink.

#ls -l /dev/snd
total 0
crw-------    1 wa1ter   audio    116,   0 Dec 31  1969 controlC0
crw-------    1 wa1ter   audio    116,  32 Dec 31  1969 controlC1
crw-------    1 wa1ter   audio    116,  64 Dec 31  1969 controlC2
crw-------    1 wa1ter   audio    116,  96 Dec 31  1969 controlC3
crw-------    1 wa1ter   audio    116, 128 Dec 31  1969 controlC4
crw-------    1 wa1ter   audio    116, 160 Dec 31  1969 controlC5
crw-------    1 wa1ter   audio    116, 192 Dec 31  1969 controlC6
crw-------    1 wa1ter   audio    116, 224 Dec 31  1969 controlC7
crw-------    1 wa1ter   audio    116,  24 Dec 31  1969 pcmC0D0c
crw-------    1 wa1ter   audio    116,  16 Dec 31  1969 pcmC0D0p
crw-------    1 wa1ter   audio    116,  25 Dec 31  1969 pcmC0D1c
crw-------    1 wa1ter   audio    116,  17 Dec 31  1969 pcmC0D1p
crw-------    1 wa1ter   audio    116,   1 Dec 31  1969 seq
crw-------    1 wa1ter   audio    116,  33 Dec 31  1969 timer

#ls -l /dev/sound
total 0
crw-------    1 wa1ter   audio     14,  12 Dec 31  1969 adsp
crw-------    1 wa1ter   audio     14,   4 Dec 31  1969 audio
crw-------    1 wa1ter   audio     14,   3 Dec 31  1969 dsp
crw-------    1 wa1ter   audio     14,   0 Dec 31  1969 mixer
crw-------    1 wa1ter   audio     14,   1 Dec 31  1969 sequencer
crw-------    1 wa1ter   audio     14,   8 Dec 31  1969 sequencer2

#ls -l /dev/dsp
lr-xr-xr-x    1 root     root            9 May  9 10:33 /dev/dsp -> sound/dsp
#ls -l /dev/adsp
lr-xr-xr-x    1 root     root           10 May  9 10:33 /dev/adsp -> sound/adsp


Seems to me that any symlinks to /proc don't really belong in /dev.
Any opinions to the contrary?  (Remember:  my sound is working just fine.)