[ALSA - utils 0001779]: alsaconf creates incorrect configuration file for snd-cs4236

[bugtrack@alsa-project.org]

A NOTE has been added to this issue.
======================================================================
<https://bugtrack.alsa-project.org/alsa-bug/view.php?id=1779> 
======================================================================
Reported By:                briand
Assigned To:                
======================================================================
Project:                    ALSA - utils
Issue ID:                   1779
Category:                   alsaconf
Reproducibility:            always
Severity:                   block
Priority:                   normal
Status:                     new
======================================================================
Date Submitted:             01-25-2006 05:01 CET
Last Modified:              01-29-2006 15:15 CET
======================================================================
Summary:                    alsaconf creates incorrect configuration file for
snd-cs4236
Description: 
(see also 0001748)

Fresh compile/install of alsa-driver, alsa-utils, alsa-tools (1.0.11rc2).

When I run `alsaconf` on this machine, it (eventually) detects the CS4237B
sound chip, then claims to be testing playback/capture, then offers to
write /etc/modprobe.conf with the "discovered" settings.  During the
"tests", there is no sound emitted from the machine.

----[ alsaconf.log ]-----------------------------------------------------
 Starting alsaconf: Wed Jan 11 22:04:07 EST 2006
 cs4236 port=0x530 cport=0x538 isapnp=0 dma1=0 dma2=1 irq=5
 playback OK
 Now checking capture...
 cs4236 yes port=0x530 cport=0x538 isapnp=0 dma1=0 dma2=1 irq=5
 capture OK
-------------------------------------------------------------------------

----[ /etc/modprobe.conf ]-----------------------------------------------
 # --- BEGIN: Generated by ALSACONF, do not edit. ---
 # --- ALSACONF version 1.0.11rc2 ---
 alias snd-card-0 snd-cs4236
 alias sound-slot-0 snd-cs4236
 options snd-cs4236 port=0x530 cport=0x538 isapnp=0 dma1=1 dma2=0 irq=5
 # --- END: Generated by ALSACONF, do not edit. ---
-------------------------------------------------------------------------

When I `modprobe cs4236` I hear a faint 'click'.  Sound driver is -not-
loaded (but everything it wanted to have loaded [snd_opl3_lib,
snd_cs4231_lib, snd_pcm, et al] is loaded).   dmesg includes the line:

"ALSA
/home/brian/alsa/alsa-driver-1.0.11rc2/isa/cs423x/../../alsa-kernel/isa/cs423x/cs4236_lib.c:305:
CS4236+ chip detected, but control port 0x538 is not valid"

not specifying a cport= parameter (on command line or in
/etc/modprobe.conf) produces a different message, stating that one must be
supplied.

I loaded another kernel, compiled with CONFIG_PNPBIOS, and read the /sys
directory structure to discover that the actual control port for this chip
is at 0xF00.   Made the appropriate change to /etc/modprobe.conf,
rebooted, and the sound driver now loads, with no messages generated in
dmesg.

======================================================================
Relationships       ID      Summary
----------------------------------------------------------------------
related to          0001748 sound driver will not load (CS4237B chip)
======================================================================

----------------------------------------------------------------------
 briand - 01-27-06 23:19 
----------------------------------------------------------------------
This bug concerns alsaconf, not the cs4236 driver.  I've (finally) gotten
the driver loaded, in non-PnP mode, despite what alsaconf attempted.

The cs4236 module is already detected via PnP, and loads...  the problem
there is, with CONFIG_PNPBIOS enabled in the kernel, which is necessary
for PnP to work for this card, no other peripherals (my pcmcia network
card, most importantly) will work. This seems to be a widely reported
error, and the popular "solution" is to disable CONFIG_PNPBIOS in the
kernel.

   (Endless Loop -- see Loop, Endless)

The point is, when I am not using PnP, alsaconf does not pick the correct
values for the sound chip.  It picks values, claims they work, claims that
it "tested" everything, and yet ... with the values it picks, the driver
does not even load.  Without the driver loaded, the "test" should fail,
not pass... and it shouldn't be offering to modify my /etc/modprobe.conf
with values it really didn't test, that really don't work, and then claim
everything is "good to go"...

----------------------------------------------------------------------
 perex - 01-29-06 15:15 
----------------------------------------------------------------------
Yes, there is a comment in
https://bugtrack.alsa-project.org/alsa-bug/view.php?id=1748 that cport works
with 0xf00 but
alsaconf selects for some reason 0x538 . It seems like a bug in alsaconf
script, because 0xf00 cport is there (probed) for cs4236 .

Issue History
Date Modified  Username       Field                    Change              
======================================================================
01-25-06 05:01 briand         New Issue                                    
01-27-06 15:13 tiwai          Note Added: 0007803                          
01-27-06 23:19 briand         Note Added: 0007809                          
01-29-06 15:13 perex          Relationship added       related to 0001748  
01-29-06 15:15 perex          Note Added: 0007831                          
======================================================================




-------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems?  Stop!  Download the new AJAX search engine that makes
searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=103432&bid=230486&dat=121642