Date: Sun, 27 Sep 1998 21:09:58 +0000 (GMT) From: Terry Lambert <tlambert@primenet.com> To: dholland@cs.toronto.edu (David Holland) Cc: rotel@indigo.ie, tlambert@primenet.com, current@FreeBSD.ORG Subject: Re: DEVFS & SLICE? Message-ID: <199809272109.OAA29425@usr05.primenet.com> In-Reply-To: <98Sep26.231232edt.37814-5346@qew.cs.toronto.edu> from "David Holland" at Sep 26, 98 11:12:28 pm
next in thread | previous in thread | raw e-mail | index | archive | help
> Major and minor device numbers are a projection of internal kernel > guts into user space. As such, they are a problem, not a feature. Here here. > The DEVFS approach, when taken to its conclusion, ultimately leads to > abolishing the concept entirely. You'd still need dev_t for stat, but > you don't need or want to interpret the contents. For what it's worth, on Solaris, the dev_t returned by stat is the address of the device descriptor in kernel space, and not an index. > > These two issues seem more or less the same, using DEVFS simply > > moves the checking from mknod code to DEVFS code. Any general > > framework implemented in DEVFS for controlling device visibility > > in chroot environments could just as easily be provided for mknod. > > Huh? How do you tell mknod(2) that it can only create nodes in (say) > /dev, /home/ftp/dev, and /usr/test/chroot/dev? Or better yet: # mknod wd0 c 21 1 mknod: Command not found. # (Yes, I know you will still need "mknod" to build an NFS "/dev" for primitive OS's like DEC UNIX and Linux; don't take everything so seriously!). Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199809272109.OAA29425>