I actually agree that it might not be a bad thing. However, as we've seen with lots of things that touch vfs it's pretty easy to get to 80 or 90 percent functionality and then the last 10% is a royal red pain in the butt, with possibly awful crashing bugs.
So I'm certainly not averse to someone working on it, but it would have to be solid and with people to love it before it ever made it into the tree - which might not ever happen without that last 10%. I do not want anything else messing around with the midlayer in the tree that is "mostly complete" - because that's "partly wrong" - and the partly wrongs tend to become Vietnam/Iraq/Afghanistan issues[1] by those who imported the code in the first place. So, with that in mind, if you're committed, please keep hacking on it. I might even have time to look occasionally iif you send me a diff. -Bob [1] Withdraw and leave the mess for the natives to deal with. On Fri, May 30, 2014 at 12:59 PM, Stefan Fritsch <s...@sfritsch.de> wrote: > On Thursday 29 May 2014 19:17:25, M Farkas-Dyck wrote: > > On 29/05/2014, Ted Unangst <t...@tedunangst.com> wrote: > > > The first question is why not use fuse? I think it's better to > > > have one userland filesystem interface than two. > > > > We already have 2: fuse, nfs. > > > > • 9p can operate over arbitrary transports, such as virtio and tcp. > > Fuse can't. > > • To my knowledge one can't have root as fuse. > > • Fuse requires a copy to/from a memory buffer on read/write. I have > > usage cases in which fuse is the performance-limiting factor. > > I agree that 9p support would be nice, especially with virtio support. > I don't think I will have time to help with it in the near future, > though. > > I have noticed that there is also o9fs [1]. I haven't tried it, and I > don't know what the plans of its author are. But in case you haven't, > you may want to look at it and decide if you want to merge your > projects, cherry-pick a few things, or just ignore it. > > Cheers, > Stefan > > > [1] https://bitbucket.org/iru/o9fs/overview > > >
