why not make everything partitionable?

why not make everything partitionable?

[martin f krafft]

[-- Attachment #1: Type: text/plain, Size: 649 bytes --]

Hi folks,

you cannot create partitions within partitions, but you can well use
whole disks for a filesystem without any partitions.

Along the same lines, I wonder why md/mdadm distinguish between
partitionable and non-partitionable in the first place. Why isn't
everything partitionable?

Thanks for any explanation(s)!

-- 
martin;              (greetings from the heart of the sun.)
  \____ echo mailto: !#^."<*>"|tr "<*> mailto:" net@madduck
 
spamtraps: madduck.bogus@madduck.net
 
the reason that every major university
maintains a department of mathematics
is that it's cheaper than
institutionalizing all those people.

[-- Attachment #2: Digital signature (GPG/PGP) --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: why not make everything partitionable?

[Michael Tokarev]

martin f krafft wrote:
> Hi folks,
> 
> you cannot create partitions within partitions, but you can well use
> whole disks for a filesystem without any partitions.

It's usually better to have a partition table in place, at least on x86.
Just to stop possible confusion - be it from kernel, or from inability
to identify disks properly (think [c]fdisk displaying labels) or from
anything else.  But ok.

> Along the same lines, I wonder why md/mdadm distinguish between
> partitionable and non-partitionable in the first place. Why isn't
> everything partitionable?

It's both historic (before, there was no partitionable md arrays),
and due to the fact that the number of partitions is limited by
only single major number (ie, 256 (sub)partitions max).

Maybe there are other reasons - I don't have a defite answer.

/mjt