Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 13 Jan 1997 10:40:09 -0700 (MST)
From:      Terry Lambert <terry@lambert.org>
To:        julian@whistle.com (Julian Elischer)
Cc:        terry@lambert.org, davidn@unique.usn.blaze.net.au, bde@zeta.org.au, joerg_wunsch@uriah.heep.sax.de, hackers@FreeBSD.org
Subject:   Re: DEVFS permissions &c.
Message-ID:  <199701131740.KAA27986@phaeton.artisoft.com>
In-Reply-To: <32D98774.59E2B600@whistle.com> from "Julian Elischer" at Jan 12, 97 04:53:08 pm

next in thread | previous in thread | raw e-mail | index | archive | help
> > I mean, shouldn't pty's be handled by:
> > 
> [description of cloning devices deleted]
> 
> yes, but it could be even simpler..
> 
> make the server side appear when you open and are assigned a 
> client side pty.

OK, I don't understand this... I think you want the master open
before the slave (doesn't closing the master EOF the slave,
potentially SIGHUP'ing the process group if -clocal?).

Why do you want an FS object, rather than a post-open ioctl, to
derive the slave node?

I don't need a namespace entry in the devfs to provide a means
of creating an fd reference to the slave vnode ...my example used
an ioctl() on the master to get there.

> julian
> (patches accepted)

Let's explore what a "correct" implementation should look like,
first, and generate patches after that.


					Regards,
					Terry Lambert
					terry@lambert.org
---
Any opinions in this posting are my own and not those of my present
or previous employers.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199701131740.KAA27986>