terryl@sail.LABS.TEK.COM (04/11/91)
We have some hardware based on the Motorola Hypermodule (multiple 88ks on a card), and we're trying to use Motorola's MVME188BUG debugger to do disk reads(using the .DSKRD trap function), and it appears that the blocks we ask for are not necessarily the blocks that are read. Specifically, we are trying to modify the secondary boot code to recognize a Berkeley Fast File system, and we have to "fudge" the disk block numbers fed to the .DSKRD trap function; unfortunately, the "fudge" factor is not constant for the whole disk, and through empirical methods (read: brute force method) varies depending where on the disk we try to read from. So, the question(s) is(are) this: Is there anyone out there who has done something like this?? Is there some environment that we need to set up in order for this to work??? About the only configuration thing we can see is the ROMs look like they are identified as "Version 4.1", and yes, we do have the docomentation for MVME188BUG, but there's a LOT of assumptions that are made in the original Motorola secondary boot code that aren't mentioned in the manual (like disk block numbers are in terms of 256 byte sectors, instead of 512 byte sectors; ditto for counts...). Terry Laskodi of Tektronix terryl@sail.labs.tek.com (If your mailer groks domains) OR Choose one of the below: ucbvax!\ \ uw-beaver!tektronix!sail.labs!terryl / decvax!/