Subject: kern/10778: uaudio doesn't work, panics machine
To: None <gnats-bugs@gnats.netbsd.org>
From: None <tls@cs.stevens-tech.edu>
List: netbsd-bugs
Date: 08/07/2000 14:46:18
>Number:         10778
>Category:       kern
>Synopsis:       uaudio doesn't work.  mixer ineffectual, writing data panics.
>Confidential:   no
>Severity:       critical
>Priority:       medium
>Responsible:    kern-bug-people
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Mon Aug 07 14:47:00 PDT 2000
>Closed-Date:
>Last-Modified:
>Originator:     Charlie Root
>Release:        20000601
>Organization:
Stevens Institute of Technology
>Environment:
System: NetBSD cantillon 1.4Z NetBSD 1.4Z (LAMENITH) #0: Thu Jun 1 19:01:12 EDT 2000 root@amstel.cs.stevens-tech.edu:/usr/src/sys/arch/i386/compile/LAMENITH i386


>Description:
Sometime between 1.4T and 1.4Z, uaudio was broken.  Now, though it probes and
attaches correctly with my Labtec USB speakers, it's impossible to actually
set the output level with the mixer -- it shows up as 0,0, and attempts to
set it (it's fea2-i1-master on my machine) for example with 
mixerctl -w fea2-i1-master=230 result in it still being set to 0,0.

Bill Squier reports the same problem with his Altec-Lansing USB speakers
which also worked just fine with older kernels.

Also, attempting to *write* to /dev/audio results in a panic: kmem_free:
address 0x4 not in kmem_map.  I don't know if Bill has this problem as well.

I cannot easily test with 1.5_ALPHA but I see no reason to think this problem
will not impact the release as well (I see no specific change which looks to
be related; perhaps I misread the code, if there is one.)
	
>How-To-Repeat:
Hook up some USB speakers.  Prepare to lose!
>Fix:
>Release-Note:
>Audit-Trail:
>Unformatted: