Date: Sat, 14 Jan 2006 23:26:42 +0000 (GMT) From: Robert Watson <rwatson@FreeBSD.org> To: Brooks Davis <brooks@one-eyed-alien.net> Cc: freebsd-net@freebsd.org, David Leimbach <leimy2k@gmail.com> Subject: Re: tap interfaces on FreeBSD 6.0 Message-ID: <20060114232422.C38874@fledge.watson.org> In-Reply-To: <20060114223029.GA10293@odin.ac.hmc.edu> References: <3e1162e60601141157k2d89204eg379d39edeb13af0b@mail.gmail.com> <20060114223029.GA10293@odin.ac.hmc.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, 14 Jan 2006, Brooks Davis wrote: > IIRC, you don't need to create the device nodes at all. I think qemu just > tries to open /dev/tap# from 0-9. In my local code to use if_tap, I just iterate through /dev/tapX nodes until I find one that isn't busy. That causes the creation of the interface tapX, which can then be ioctl'd (etc). I've never run into the symptoms described, so it could well be it's an odd result of auto-loading by ifconfig -- however, normally you'd expect ENOENT if the module wasn't loaded, since the device node simply wouldn't exist. I'll run some local tests tomorrow and see if I can reproduce it with a recent tree version. Robert N M Watson
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060114232422.C38874>