Matthew Dillon wrote:
quoted text
> :    The compatibility slice (s0) will no longer point at the first BSD slice
> :    in the upcoming commit.  This should not effect anyone as nobody uses
> :    the compatibility slice in that manner any more.
> :
> :    This change is necessary because to implement GPT partitions, which I've
> :    decided will be abstracted as slices, I want GPT partition #0 to 
> :    be s0, GPT partition #1 to be s1, and so forth to avoid confusion.
> 
>     Hmm.  I need some help.  The documentation for EFI says the partition
>     numbers start at 0, but the FreeBSD GPT program seems to start numbering
>     them at 1.
> 
>     Does anyone know how Linux interprets the partition numbers for a GPT?
> 
> 					-Matt
> 					Matthew Dillon 
> 					<dillon@backplane.com>

I'm not where I can look it up, but ISTR Linux starts with the 'zeroth' 
partition - GPT or otherwise.

Your previous post w/r 'reserved' compatibility slice 0 explains (finally) why 
FreeBSD has always appeared to 'start' with 1. Seems there is/was a method to 
their apparant inconsistency after all?

'Spose I'd have known that if I ran anything that neeed a 'compatibility' slice.

W/R going GPT-way or legacy FreeBSD way - perhaps use of zeroth should be 
weighed vis whatever FreeBSD 7.X is doing?

I mean - with whom/what/(anything?) should DFLY try for compatibility?

HTH,

Bill