Storing GUIDs in linux [filesystems]

[Valdis.Kletnieks at vt.edu]

On Thu, 27 Mar 2014 16:55:53 +0530, Sankar P said:

> The size of the "uid" in UNIX is just 16 bits. So what is the way used
> by the Linux [filesystems] to store the GUID in linux to uniquely
> identify an user ?

The filesystem only stores UIDs, not GUIDS for users.

> Were there any proposals in the past, to increase the size of the unix
> uid to match that of a GUID ?

There's been proposals.  They usually end up dying off because the person
making the proposal can't make the business case.  What *actual* benefits do we get
from doing this?  Wnat problem does it solve that (for instance) aren't solved
just as well by doing uid-guid mapping in the nfsd or cifs code as things go
out on the wire?  How do we do conversion of all the existing filesystems that
only store 32-bit values?  What's the performance impact of having to do
128-bit comparisons rather than 32-bit?
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 848 bytes
Desc: not available
Url : http://lists.kernelnewbies.org/pipermail/kernelnewbies/attachments/20140327/091f60a9/attachment.bin