--- On Tue, 21/6/11, Eric Van Hensbergen <eri...@gmail.com> wrote: ... >> Um, does v9fs remote Linux devices? I find it hard to >> imagine it would remote ioctls but it makes sense *nix to >> *nix. >> > > Depends on how you configure it. There is a nodevmap > option to the v9fs mount which will instruct it to just access > the remote devices directly instead of just mapping their major/minor > numbers to local devices. You are correct in your imagining that we > don't go anywhere near ioctls with a 10 foot pole. However, many things > "just work" without ioctls these days. > Thanks for the info, but the devices encumbered with ioctls are the tricky ones and even if they can be sorted out I'm sure there are some other traps out there. Too bad there are no RFS gurus lurking here to offer their wisdom on remoting devices. I have a sneaking suspicion few people would have bothered, since the few devices worth remoting back then were easily handled by rsh/rcmd. Did RFS make it beyond SVR4? > > > > That just leaves my issues with X. > > > > Actually, its a bit worse than that. The physical > network devices aren't file system accessible anymore, Actually, I'd blotted out of my mind all knowledges of STREAMs devices and the related horror of TLI programming until you reminded me. > so you'd need to remote them as a service (via Inferno or something) or > use the tap device and remote that and hope that it doesn't require ioctls > (and I think it might). >
Oh, I'm not worried about remoting network interfaces. I'm fine with packet forwarding and can live with NAT for now. However, it reminds me of another point in Plan 9's favour that the introductory papers should be updated to be more explicit about - NATs are unnecessary in a pure Plan 9 deployment. Unfortunately, neither Plan 9 routers or decent alternatives to NAT such as RSIP widely available. > -eric > Andrew P.S. There's spammers subscribed to this list. Hi there, friends of Khalifa.
