From mboxrd@z Thu Jan  1 00:00:00 1970
From: erik quanstrom <quanstro@quanstro.net>
To: 9fans@cse.psu.edu, Russ Cox <rsc@swtch.com>
References: <20060108010536.A53891439F9@dexter-peak.quanstro.net>
	<ee9e417a0601071710v37714f35he1a439744cf5ef2d@mail.gmail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8
In-Reply-To: <ee9e417a0601071710v37714f35he1a439744cf5ef2d@mail.gmail.com>
Subject: Re: [9fans] p9p procrfork
Message-Id: <20060108022412.8239A1F6C94@dexter-peak.quanstro.net>
Date: Sat,  7 Jan 2006 20:24:12 -0600
Cc: 
Topicbox-Message-UUID: d3edc3ce-ead0-11e9-9d60-3106f5b1d025

all the worlds a vax, i mean, a linux box. ;-)
i misread the source. i was reading Linux.c; 
i didn't realize that that is only for nptl.

actually, i was thinking about ndb/dns is doing.

- erik

Russ Cox <rsc@swtch.com> writes

| 
| > what is the reason that p9p doesn't support procrfork?
| > linux might be able to support procrfork via clone(2), but
| > of course, pthreads might have something else to say about
| > that.
| 
| you answered the question yourself.
| we're not programming to linux.
| we're programming to what we can do
| portably across a range of unixes.
| file descriptor and name space groups
| (the only things you ever use procrfork for)
| are not part of that.
| 
| if you need procrfork for rffdg before procexec, usually
| you can get by with threadexec etc.
| 
| russ