partition minor numbering wrong with GPT

[Richard Hirst]

Hi,
  If I "mklabel gpt", then "mkpart .." 4 times, then parted and the
kernel agree that the partition minor numbers are 1,2,3 and 4.

If I then "rm 2", parted thinks the minors are 1,3 and 4, but the kernel
thinks they are 1,2 and 3.  That is real bad news, I think.

I did some experiments with msdos labels, and there parted and the
kernel always agree.  If you delete a primary partition the numbering of
the others doesn't change.  If you delete a logical partition the higher
ones renumber so that there is no gap.

So, for GPT the kernel emulates msdos logical partition behaviour.
parted emulates msdos primary partition behaviour.

Which is "right"?  I favour parted's way of doing things, because minor
numbers will not change when you delete a partition or add a new one in
a gap between two others.  Effectively, a minor number becomes tied to a
partition table slot.

I'm assuming everything orders partitions by the order on the partition
table entries, and not the phyiscal order of th partitions on disk, but
I havn't verified that.

I'm sure we need to fix this, and I guess fixing parted is the safest
way to go, even though it yields the worst solution, IMHO.

Comments?

Richard