Source-Changes archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: CVS commit: syssrc/sys/sys



On Wed, May 15, 2002 at 09:24:31AM -0700, Jason Thorpe wrote:
  | On Thu, May 16, 2002 at 01:35:34AM +1000, Luke Mewburn wrote:
  | 
  |  > I suppose we will just end up in semantic arguments about whether the
  |  > Apple Partition Map and related stuff or PC MBR related stuff is a
  |  > "bootblock" related issue (such as the alpha/vax/pmax stuff already in
  |  > <sys/bootblock.h>), or a "disklabel" related issue. 
  |  > My gut feel is that it's more bootblock-y than disklabel-y...
  | 
  | Let's see.. the MBR defines partitions on the disk...
  | 
  | ...the Apple Partition Map defines partitions on the disk...
  | 
  | "Seems pretty partition-related to me!"

As I said, "semantic arguments".

As we know, "BIOS/PROM partition" != "BSD disklabel".

The BSD disklabel is traditionally frobbed with disklabel(8),
but the "BIOS/PROM partitions" are frobbed with fdisk(8) or
other tools.

So; <sys/partition.h> ?  <sys/partition_mbr.h> ?  <sys/disklabel_mbr.h> ?
<dev/isa/mbr.h>  ?  (we had <dev/dec/dec_boot.h>)   Any other
variations?

I felt it was easier for installboot(8)'s portability and for
searching where the various boot-block related goop was (when
trying to document this a while ago), and having all of the
"boot block" related stuff (including BIOS/PROM partitions)
for the various platforms in one place made sense.

Is this really that big an issue for you?  Do we really need to break
up <sys/bootblock.h> into N different files, one for each different
type of MD bootblock/BIOS-partition stuff ?  I personally don't see
the win.



Home | Main Index | Thread Index | Old Index