Re: Panic when loading snd_ich.

From: Ariff Abdullah <ariff_at_FreeBSD.org>
Date: Mon, 26 Mar 2007 01:28:26 +0800
On Sun, 25 Mar 2007 13:17:09 -0400
Sean Bryant <sean_at_cyberwang.net> wrote:
[....]
> >>>>>   
> >>>>>       
> >>>>>           
> >>>> That was it! Sorry.
> >>>>
> >>>> Ariff, ohci_add_done is on the actual panic line. that's how I
> >>>>         
> >>> come > up  with that.
> >>>       
> >>>>     
> >>>>         
> >>> So where does the "loading snd_ich cause panic" fits in? Are you
> >>> saying that it is all about "kldload snd_ich" or panic during
> >>> boot?
> >>>
> >>>
> >>>       
> >> panic during boot. not when loading the kernel module:
> >>
> >> pcm0: <nVidia nForce4> port 0xdc00-0xdcff mem
> >0xd5003000-0xd5003fff > irq  23 at device 4.0 on pci0
> >> pcm0: [ITHREAD]
> >> pcm0: cannot reset channel 0
> >> pcm0: unable to initialize the card
> >> device_attach: pcm0 attach returned 6
> >>
> >>
> >> that's what happens when I load the module after everything has
> >> started.  snd_ich worked as of 3/06/07. I decided to give my
> >machine > an update and  the panic is what happened.
> >>
> >>
> >>     
> >
> > I'm suspecting something more generic. Virtually, nothing is
> > changed within snd_ich land within that timeframe, except few bug
> > fixes mostly during detach procedure.
> >
> > Besides, your panic source is from USB land, not snd_ich.
> >
> That's what I suspected. So I just installed the latest HPS usb
> stack  and it's the same.
> 
> pcm0: <nVidia nForce4> port 0xdc00-0xdcff,0x4400-0x44ff mem 
> 0xd5003000-0xd5003fff irq 23 at device 4.0 on pci0
> pcm0: [ITHREAD]
> pcm0: cannot reset channel 0
> pcm0: unable to initialize the card
> device_attach: pcm0 attach returned 6
> 

Try this one: (sys/dev/sound/pci/ich.c)

  http://people.freebsd.org/~ariff/test/ich.c

(no, this has nothing to do with the panic at all)

> 
> With the latest HPS USB stack not sure if that's the same one in
> HEAD or  not.
> Maybe something is conflicting or some other change has caused the 
> resources to be tied up. That's a really rough guess with no
> evidence to  support it.
> 

You should reproduce the panic, generate the dump, submit the
backtrace so things will become much clear for us. Will you?


--
Ariff Abdullah
FreeBSD

... Recording in stereo is obviously too advanced
    and confusing for us idiot ***** users :P ........

Received on Sun Mar 25 2007 - 15:28:55 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:39:07 UTC