Bill Studenmund wrote:

>when making Apple partition maps. Thus on a CD, you're supposed to use 2 K
>blocks, not 512-byte blocks. But lots of tools, like the one you used,
>only know about 512-byte blocks. 
>
Yes, I just dump the data, I didn't try to make much sense of it except 
for the "error" you noted below.

>> Partition map (with 512 byte blocks) on '/dev/ata0.0' (/dev/hda)
>>   sbSig:       0x4552    sbBlkSize:  0x0800    sbBlkCount:  0x00050000
>                                         ^^^^ Not 0x0100 :-)

The "void" entries in the map seem to be a new approach in Apple CDs.  I 
don't recall seeing anything like that before on the Apple CDs I've 
dumped.

>>   pmBootCksum:  0x00000000  pmProcessor:   
>>     *Error* - Partition not adjacent to previous partition
>      ^^^^^^^^^^^^^^^^^^^^^^^^^^^
>
>Since when is this an error??? :-)

The code in this utility has the ability to sort the Map Entries by 
starting block number. Then if there is space no accounted for on the 
device it issues this error.  According to the Apple doumentation all 
space (other than Block Zero) should be covered by an entry in the Map.  
Sorting the entries made it easier to force an order that would match the 
way NetBSD/mac68k finds the partitions.  So the answer is it's not really 
an error, but if there are gaps that aren't mapped on the disk it is a 
potential error.

-bob