First stab at ea inode disk format doc updates

First stab at ea inode disk format doc updates

[Darrick J. Wong]

Hi all,

I've taken a first stab at capturing the on-disk format updates for ea
inodes.  Let me know about any inaccuracies (though being a wiki, you
can fix them yourself if you're so inclined).

Also -- what is the value of i_mtime (ea inode parent pointer) when the
refcount > 1?

--D

[1] https://ext4.wiki.kernel.org/index.php/Ext4_Disk_Layout

Re: First stab at ea inode disk format doc updates

[Tahsin Erdogan]

hi Darrick,

> Also -- what is the value of i_mtime (ea inode parent pointer) when the
> refcount > 1?

In the current implementation, xattr inode format is assumed to be in
the new format (refcount in i_ctime and l_i_version, hash in i_atime).
parent xattr e_hash field incorporates xattr inode hash to achieve
strong binding. If hash validation fails, we fallback to old binding
validation (xattr inode i_mtime == parent ino && xattr inode
i_generation == parent i_generation). The idea is to allow migration
from old lustre disks. So, if refcount is valid then value in i_mtime
is not defined.

Re: First stab at ea inode disk format doc updates

[Andreas Dilger]

[-- Attachment #1: Type: text/plain, Size: 697 bytes --]

On Jun 22, 2017, at 12:18 PM, Darrick J. Wong <darrick.wong@oracle.com> wrote:
> 
> I've taken a first stab at capturing the on-disk format updates for ea
> inodes.  Let me know about any inaccuracies (though being a wiki, you
> can fix them yourself if you're so inclined).
> 
> Also -- what is the value of i_mtime (ea inode parent pointer) when the
> refcount > 1?

Thanks for the update Darrick.  I clarified that i_mtime/i_generation is
only for refcount == 1.  I don't know if the current code will clear these
values if refcount > 1 or not, or if they are just unused in that case?

> [1] https://ext4.wiki.kernel.org/index.php/Ext4_Disk_Layout


Cheers, Andreas






[-- Attachment #2: Message signed with OpenPGP --]
[-- Type: application/pgp-signature, Size: 195 bytes --]