Re: GFS requirements (was: Non-GPL export of invalidate_mmap_range)

[Daniel Phillips <phillips@arcor.de>]

On Saturday 21 February 2004 09:17, Lars Marowsky-Bree wrote:
> On 2004-02-20T22:16:09, I said:
> > Each DFS is free to implement its own infrastructure, possibly involving
> > kernel extensions.
>
> Yes. Though I do reserve the right to find this highly silly, that we
> might end up with multiple hooks for clustering infrastructure in the
> kernel...

But the one true clustering infrastructure hasn't been developed yet.  The 
upcoming crop of designs need a chance to evolve before a framework is cast 
in stone.  Perhaps we will eventually end up with a generic harness, 
something like a vfs for cluster infrastructure, but in my opinion, we're far 
from being able to define that sensibly now.  It's better to implement 
exactly what a given DFS needs for the time being.

> So, how does OpenGFS/GFS achieve the communication? How does it interact
> with the infrastructure (which, I infere from your above comments, is
> meant to reside in user-space)?

It's done both ways, actually.  No new kernel hooks are used in either case.

Regards,

Daniel