Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm List-Subscribe: List-Archive: List-Post: List-Help: , Sender: cygwin-owner AT cygwin DOT com Mail-Followup-To: cygwin AT cygwin DOT com Delivered-To: mailing list cygwin AT cygwin DOT com X-Authentication-Warning: localhost.localdomain: ronald owned process doing -bs Date: Wed, 19 Feb 2003 12:15:59 +0100 (CET) From: Ronald Landheer-Cieslak X-X-Sender: ronald AT localhost DOT localdomain To: Max Bowsher cc: Ben Clewett , Subject: Re: Duplicate CygWin In-Reply-To: <001801c2d78b$7a8ef390$78d96f83@pomello> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Tue, 18 Feb 2003, Max Bowsher wrote: > Ronald Landheer-Cieslak wrote: >> On Tue, 18 Feb 2003, Ben Clewett wrote: >>> - Get the different root '/' mount points to be respected. >> yes. > No. You need a different *set* of mounts. Not just one... therefore... Ehm.. I think this is what he meant, right? >>> With the mount points, I was hoping a *hack* as simple as: >>> >>> char *pRootMount; pRootMount = getenv("ROOT_MOUNT"); >>> if (pRootMount) { ....; return; } > ... this is no good. .. unless he can get the mount points to work relative to a root mount. His fstabs idea might work here: you mount a single root (under which you *must* have an etc directory with at least an fstabs file), read the fstabs file and mount the rest, using only POSIX paths as reference. That's why I told him to "write and contribute". I think the idea might be a good one - especially if he develops it and tests it on his Cygwin fork. >>> I'll look for the memory naming following your instructions. >>> In both cases, I would prefer any options which didn't mean forking >>> the code. So I would be interested in your opinions... >> Options *not* forking the code but allowing you to have two different >> Cygwin installations with different mountpoints on the same system?? > Huh?!? If the second Cygwin *isn't* a fork already, then why do you need a > second installation at all? Well.. it might be possible to introduce a configuration option at compile time, setting the defines as Igor suggested yesterday. If he could write a patch and get it accepted he would no longer need a fork. As Cliff Hones suggested (also yesterday), it may well be a good idea to start supporting this kind of thing. If Ben wants to put some time and effort into it, forking off a Cygwin, developing the capability to have two run at the same time, that could lead to a merge with the current Cygwin to let Cygwin have that capability (and if he can get the mount points out of the registry, I think Chris and Corinna may be interested in that as well). Anything in that direction would have to start off with a fork, though.. rlc -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/