"Stephen C. Tweedie" wrote:
>
> Hi,
>
> On Fri, Feb 23, 2001 at 02:22:34PM -0500, Robert Watson wrote:
> >
> > 1) I'd like it if we moved to namespace partitioning not inside the name
> > string.
>
> Agreed.
I personally like the name string notion as you can add
new partition types without changing header files. The
counter argument, which is that applications really ought
to be aware of what they're dealing with does have merit,
however.
> > 2) I'd actually like to propose an additional namespace, the "internal" or
> > "filesystem" namespace, to be used by the file system to manage
> > internal attributes, as distinguished from "system" attributes.
>
> I don't see any reason for this. If the filesystem happens to
> implement such an internal namespace, then it can maintain that
> internally and only expose access to it via the system namespace. The
> API doesn't have to give access to it (if it did, it wouldn't be an
> internal namespace any more!)
Yeah, like he said.
If they're exposed to the system, they're system attributes.
If they're not, you can treat them any way you want in your
file system code, and no one will be the wiser.
--
Casey Schaufler Manager, Trust Technology, SGI
[EMAIL PROTECTED] voice: 650.933.1634
[EMAIL PROTECTED] Pager: 888.220.0607
-
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
the body of a message to [EMAIL PROTECTED]
