Re: NetBSD 5 STABLE wm0 panic

[Chris Ross <cross+netbsd%distal.com@localhost>]

  Okay.  I was able to compile a GENERIC netbsd-5-0 kernel, and it  
fails in the same way.  The addresses are slightly different than  
mentioned in my earlier post, but the same error clearly in the same  
place occurs.  As there have been no changes in the netbsd-5-0 branch  
in the wm driver since netbsd-5-0-RELEASE, this should apply to that  
release as well.

  Please, someone let me know if there's any more information I can  
provide.  I am anxious to get this problem located and resolved so  
that I can update this machine, as the first of many 4.0 machines I  
plan to upgrade.

  Thank you.

                             - Chris


On May 17, 2009, at 23:07, Chris Ross wrote:
>  From sources updated today, I built the world on an earlier NetBSD  
> 5.0_STABLE i386 system, for i386.  I built a custom kernel for my  
> i386 (MP) router, and it doesn't boot successfully.
>
>  In further testing, it appears that it's the act of simply bringing  
> "up" the wm0 interface[...]
> I can bring online all of the filesystems, and then simply executing:
>
>  ifconfig wm0 up
>
>  causes a panic.  Handwritten/copied here it is:
>
> uvm_fault(0xcadffd40, 0, 1) -> 0xe
> fatal page fault in supervisor mode
> trap type 6 code 0 eip c061788b cs 8 eflags 10286 cr2 0 ilevel 6
> kernel: supervisor trap page fault, code=0
> Stopped in pid 29.1 (ifconfig) at netbsd:wm_tbi_mediachange+0xb:  
> movl 0
> (%eax),%esi
> db{0}>
>
>  I know there have been changes singe 5.0 release that affect wm,  
> but I don't know if 5.0 release works either.  I'm going to build a  
> netbsd-5-0 kernel next, but that will take some time to fetch that  
> source tree.
>
>  NetBSD 4.0_STABLE reports my wm as:
>
> wm0 at pci0 dev 13 function 0: Intel i82543GC 1000BASE-X Ethernet,  
> rev. 2
> wm0: interrupting at ioapic0 pin 17 (irq 11)
> wm0: 32-bit 33MHz PCI bus
> wm0: 64 word (6 address bits) MicroWire EEPROM
> wm0: Ethernet address 00:03:47:0d:eb:86
> wm0: 1000baseSX, 1000baseSX-FDX, auto
>
>  ..and the pcictl dump for this device is attached.
>
>  Anyone have any idea?  Thanks.