[parted-devel] [PATCH] GPT & BIOS Boot partition

[H. Peter Anvin]

Robert Millan wrote:
>>>
>> Sectors 1-62 were never a good idea to use, simply because there was no 
>> management of it.  Plus, you couldn't assume they were there.
> 
> Please could you ellaborate on that?  GRUB 2 currently uses them (I think GRUB
> Legacy as well) as long as its bootstrap code fits, and I don't recall someone
> reporting trouble.
> 

There are a number of users of this space, and NONE of them check that 
anything else steps on anything else... for example, I know a network OS 
installer that unconditionally clobbers sector 4.

It's unmanaged space, and has all the problems of unmanaged space. 
Furthermore, it's not 63 sectors, it's one track, however little that 
happens to be; PLUS there are partitioning systems in the field that 
don't reserve ANY space after the MBR (all existing versions of Grub 
fails on disks thus partitioned, although my understanding it was more 
because of the utterly broken handing of disk write consistency in the 
Grub installer, rather than it occupying these sectors.)

>>>> especially with Redmond OSes seemingly going a different way.
>>> What do you mean?  Last I heard, their way was not supporting GPT on BIOS 
>>> at
>>> all :-)
>> They do; you have to have MBR entries for your boot partitions, but then 
>> they pick up other partitions from the GPT.
>>
>> Boot partition is obviously limited to 2 TB.
> 
> Ah, the hybrid MBR-GPT.  An horrible hack, if you ask me...

Perhaps, it's the closest thing there is to a standard; certainly the 
only multi-OS standard in existence.  Your thing is so far apart from a 
multi-anything standard that can be; I would recommend relabelling the 
proposed UUID  "Grub 2" (since that's what it is, no more, no less), and 
use the standard methods for setting the partition type UUID.

At least that way the space is managed (reserved for Grub 2) and nothing 
else.

	-hpa