Oops: iMic, alsa cvs, 2.4.23-pre2

Oops: iMic, alsa cvs, 2.4.23-pre2

[Andrew Burgess]

NULL pointer dereference

Trace; f09b16bf <[usb-ohci]dl_del_list+40f/460>
Trace; c0119619 <process_timeout+19/20>
Trace; f09b255b <[usb-ohci]hc_interrupt+10b/1a0>
Trace; c0109080 <handle_IRQ_event+60/a0>
Trace; c010929b <do_IRQ+8b/e0>
Trace; c0105420 <default_idle+0/40>
Trace; c010bd18 <call_do_IRQ+5/d>
Trace; c0105420 <default_idle+0/40>
Trace; c010544c <default_idle+2c/40>
Trace; c01054c2 <cpu_idle+42/60>
Trace; c011d42b <call_console_drivers+5b/120>
Trace; c011d607 <printk+a7/170>

I run:

  arecord -D imic --rate 44100 -f S16_LE - | aplay -

and then press ^C

I get:

A few URB timeouts the first time but the kernel stayed up.
Then the oops the 2nd time I ran it and a hard hang (had to use
a pencil for the backtrace :-).

kernel 2.4.23-pre2 (because it has a new batch of USB patches)
alsa-cvs
dual athlon mb
Griffin iMic
Creative Soundblaster-Live!

PS Any chance the alsa cvs could set its version to something other
than 0.9.6? At least 0.9.6-cvs or maybe some actual cvs version
that would make sense to developers? Maybe the build script could
detect it was running in a cvs tree and grab the version from the cvs info?
If I were a cvs guru I'd send a patch...




-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf

Re: Oops: iMic, alsa cvs, 2.4.23-pre2

[Takashi Iwai]

At Sun, 31 Aug 2003 14:30:38 -0700,
Andrew Burgess wrote:
> 
> NULL pointer dereference
> 
> Trace; f09b16bf <[usb-ohci]dl_del_list+40f/460>
> Trace; c0119619 <process_timeout+19/20>
> Trace; f09b255b <[usb-ohci]hc_interrupt+10b/1a0>
> Trace; c0109080 <handle_IRQ_event+60/a0>
> Trace; c010929b <do_IRQ+8b/e0>
> Trace; c0105420 <default_idle+0/40>
> Trace; c010bd18 <call_do_IRQ+5/d>
> Trace; c0105420 <default_idle+0/40>
> Trace; c010544c <default_idle+2c/40>
> Trace; c01054c2 <cpu_idle+42/60>
> Trace; c011d42b <call_console_drivers+5b/120>
> Trace; c011d607 <printk+a7/170>
> 
> I run:
> 
>   arecord -D imic --rate 44100 -f S16_LE - | aplay -
> 
> and then press ^C
> 
> I get:
> 
> A few URB timeouts the first time but the kernel stayed up.
> Then the oops the 2nd time I ran it and a hard hang (had to use
> a pencil for the backtrace :-).

it looks rather like a bug of usb-ohci driver...


> kernel 2.4.23-pre2 (because it has a new batch of USB patches)
> alsa-cvs
> dual athlon mb
> Griffin iMic
> Creative Soundblaster-Live!
> 
> PS Any chance the alsa cvs could set its version to something other
> than 0.9.6? At least 0.9.6-cvs or maybe some actual cvs version
> that would make sense to developers? Maybe the build script could
> detect it was running in a cvs tree and grab the version from the cvs info?
> If I were a cvs guru I'd send a patch...

you can see the compilation date in /proc/asound/version.


Takashi


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf