my system won't work with it either. I used to be able to run the  
i386 version on
my amd 64fx2 but this also crashes with no error messages, just a  
lock up after
a while, seeming to be due to disk access. I can't get anywhere at  
all with the
64 bit version on this machine. I am trying the same 20070328 and a  
couple
of prior versions also.


On Mar 31, 2007, at 5:19 AM, M Graff wrote:

> I'm running vmware workstation on a dual core AMD 64 CPU, and am  
> getting a crash on 4.0_BETA2 kernel compiled from the auto build  
> site dated 200703280002Z.  This is fairly reproducable since  
> building a -current or 4.x branch kernel will trigger it pretty  
> much every time.
>
> Here's what I see on the screen just prior to the machine rebooting:
>
> uvm_fault(0xffffffff809e3b40, 0xffff8001110d9000, 1) -> e
> fatal page fault in supervisor mode
> trap type 6 code 0 rip ffffffff80461070 cs 8 rflags 10202 cr2  
> ffff8001110d9000
> cpl 5 rsp ffff800013a6d6f0
> panic: trap
> syncing disks...
>
> <and the machine reboots.>
>
> vmware could be the culprit here I suppose as the 64-bit support is  
> new and experimental.  I have two virtual machines (one will become  
> -current if it compiles a kernel ever...) and the other will remain  
> 4.x.  Both will be used for compile and integration testing.
>
> The source tree is on an NFS partition (which happens to be on  
> another vmware virtual machine, running i386 4.0_BETA2 kernel) so  
> nfs client could also be the problem.
>
> Unfortunately it does not drop into DDB so I can't trace it further  
> at this point.