On Tue, 2009-04-28 at 11:15 -0500, Ambrose, Martin wrote:
> On Wed, Apr 22, 2009 at 12:58:12, Ambrose, Martin wrote:
> 
> > > On Tue, Apr 21, 2009 at 05:05:14PM -0700, Mark A. Greer wrote:
> > > > From: Mark A. Greer <mgr...@mvista.com>
> > 
> > Thanks for this. It builds for me but hangs after 
> > uncompressing the kernel. Do you see anything wrong with this 
> > sequence?
> 
> I did some further digging. Note that my problem persists with the v4 patches 
> -- no surprise here I think since there were no functional changes. But I can 
> reply on that thread if you like.
> 
> The are no log messages after "uncompressing kernel". So used md in uboot 
> to dump the log messages. 
> 
> <1>Unable to handle kernel paging request at virtual address 34767068.
> <1>pgd = c0004000.
> <1>[34767068] *pgd=00000000.
> <4>Internal error: Oops: 0 [#1] PREEMPT.
> <4>Modules linked in:.
> <4>CPU: 0    Nottainted  (2.6.30-rc2-davinci1-05906-ged400fa-dirty #49).
> <4>PC is at 0x34767068.
> <4>LR is at register_netdevice_notifier+0x4c/0x190.
> 
> So added a print in register_netdevice_notifier [net/core/dev.c], rebuilt, 
> and rebooted. Still no log messages so agained dumped the buffer.
> 
> <4>nb = 0xc029ebfc.
> <4>nb->notifier_call = 0x34767069.
> 
> So the page fault is coming from the notifier_call. Then commented out this
> print and put equivalent in arp_init [net/ipv4/arp.c], rebuilt, and rebooted.
> Oddly now it gets passed the previous point (there are log messages in the 
> console), but it fails later with a page fault at different location.
> 
> &arp_netdev_notifier = 0xc02ed7cc
> arp_netdev_notifier.notifier_call = 0xc01e0f74
> 
> So by moving the printk the address of notifier call is different/valid.
> 
> I'm not sure how to proceed and any debugging tips are appreciated.

The PC invalid (0x34767068 is not a valid RAM address).  It is likely
you jump to that location via uninitialized or corrupt function pointer.
You can try to disable Ethernet in kernel config to see if you get pass
this point.  If you don't get this error, that means the problem is in
Ethernet initialization/driver.

Regards,

Steve

> 
> Regards, 
> Martin
> _______________________________________________
> Davinci-linux-open-source mailing list
> Davinci-linux-open-source@linux.davincidsp.com
> http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source


_______________________________________________
Davinci-linux-open-source mailing list
Davinci-linux-open-source@linux.davincidsp.com
http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source

Reply via email to