On Fri, Feb 09, 2001 at 12:13:40PM +0200, Danny Braniss wrote:
> In message <[EMAIL PROTECTED]>you write:
> }It its the bios disk probe that is causing the machine to fault.
>
> I can now confirm that it's a divide by zero in the Adaptec BIOS GRRRRRR.
> (in biosdisk, when 'probing' the adaptec) and it happens on a AIC 7899, but
> works ok on some other ADAPTECs.
>
> the problem is nasty because it's not easy to zero out the first block of the
> disk if you can't get the os working :-)
>
> }I suppose you really dont need to probe the disks when you are netbooting.
>
> i got the pxeboot not to check the disks but then the kernel stepped on it :-(
>
> is there a way to prevent BTX to halt if the 'caller' expects a div/zero?
>
> (in the days of the PDP/11 i could trap such things, print a message, and
> continue).
>
> danny
>
>
>
>
> To Unsubscribe: send mail to [EMAIL PROTECTED]
> with "unsubscribe freebsd-stable" in the body of the message
>
> end of the original message
I had this problem a couple of days ago with an Adaptec AHA-2940UW SCSI
controller and an IBM DDRS-34560W SCSI disk. I created one large slice in
dangerously dedicated mode and I wasn't able to boot anymore: I got "BTX
halted" and the registers dump. I low-level formatted the disk using the
controller BIOS, recreated the slice (this time in coperative mode!) and
everything worked!
Francesco Casadei
P.S. The DDRS-34560W uses part of the buffer for the firmware so I could
safely low-level format the disk. Check if you can low-level format your
disk without destroying it!
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-hackers" in the body of the message
