Bill Studenmund wrote:

>Do you still have code to re-read the disklabel?
>
Yes. When I was relying on the kernel mod all I did was issue a disk 
label write ioctl. The kernel mod just added a disklabelread at the end 
of the dummy disklabelwrite code.  That way if anyone eventually filled 
in the kernel code to actually write BSD labels the code in sysinst would 
still work without change.  Anyway, the code to do this is in the 
original version of sysinst that is filed with the PR I wrote when I 
submitted the code.  There are a number of other changes to the rest of 
sysinst that I made for the new version I have that I really think should 
be included though.  I cleaned up a number of things, added some helpful 
debugging stuff, supplied some default partition size calculations like 
the other ports have, added some additional message screens and such.

>> >Scott, weren't you going to make a new ioctl to do that?
>> 
>> This was also my understanding that with a new MI approach to the 
>> disklabel we would have a new ioctl to re-read in the on-disk label into 
>> the in-core one.  Basically that's the only remaining hold-up for having 
>> a functional sysinst for the mac68k port.
>
>Cool. Let's get this done soon.
>
>I'm in the process of getting installboot on the macppc to play nice with
>an existing Apple Partition Map, so that we can get 1.5 to coexist nicely
>with other OS's. :-)
>
Ah!  I was dinking with some of the same stuff on the macppc port.  Got 
side-tracked on the whole OF and bootable CD stuff though.  I have some 
utilites that might be of use to help debug this and they work on both 
mac68k and macppc ports.  One is a modification to the pdisk utility from 
MkLinux.  Basically I added a formatted dump option for decoding the 
Partition Map into something readable.  It runs under MacOS on both 
platforms.  The other utility I did was a program that does a similar 
thing under NetBSD.

>> There is one other minor problem. Sysinst creates new FFS file systems 
>> which are incompatible with the old Installer and semi-incompatible with 
>> the current Booter.  It's not a major problem if one never has to go back 
>> and use the Installer.  The Booter is a slighlty different problem, but 
>> since sysinst forces the creation of a separate root and usr the root 
>> tends to be small enough that the Booter almost always succeeds.
>
>Can we get sysinst to just build an old-style filesystem? We still have
>the code to do it in newfs.

Yes we can do this.  The newfs command in sysinst is in the MI portion of 
the code and I tried to stay away from changes there as they might affect 
all other ports.  One thing I thought about was formatting the root 
partition in old style for the Booter, and use the new style for all the 
other partitions.  This wouldn't help the Installer, but if sysinst works 
as it should there won't be much of a need for the Installer or Mkfs.

-bob