Coda File System

Re: Coda development (FUSE, pioctl)

From: <u-myfx_at_aetey.se>
Date: Fri, 6 May 2016 08:28:27 +0200
On Thu, May 05, 2016 at 11:08:34AM -0400, Jan Harkes wrote:
> On Thu, May 05, 2016 at 01:13:53PM +0200, u-myfx_at_aetey.se wrote:
> > 
> > As Jan commented and I agree, FUSE is unfortunately hardly viable.
> 
> I never said it was hardly viable, just that the pioctls would have to
> be handled like a a virtual FS instead of using an ioctl interface.

My impression is that the needed amount of work is too big compared to
the available resources.

Surely I will appload if we can skip the pioctl interface (we do not
have to change or throw away the existing kernel module for that reason).

AFAICT on Linux, FreeBSD and NetBSD we could instead talk over a local
socket and reliably inquire the identity of the other party, but probably
not on every platform.

Wouldn't such a change be a step in the right direction?
It would not introduce any more platform limitations than we have today
and make a transition to FUSE possible. It would in any case reduce
the amount of the in-kernel code we have to care about.

Rune
Received on 2016-05-06 02:29:01