Mailing-List: contact cygwin-developers-help AT sourceware DOT cygnus DOT com; run by ezmlm List-Subscribe: List-Archive: List-Post: List-Help: , Sender: cygwin-developers-owner AT sources DOT redhat DOT com Delivered-To: mailing list cygwin-developers AT sources DOT redhat DOT com Message-ID: <0a3d01c174c9$b7475b40$0200a8c0@lifelesswks> From: "Robert Collins" To: References: <20011122060643 DOT GA12017 AT redhat DOT com> <03cd01c1732b$08b28630$0200a8c0 AT lifelesswks> <20011124031453 DOT GA22066 AT redhat DOT com> Subject: daemon (was) Re: /cygdrive improvements Date: Sat, 24 Nov 2001 20:23:43 +1100 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 X-OriginalArrivalTime: 24 Nov 2001 09:23:34.0319 (UTC) FILETIME=[B0999BF0:01C174C9] ----- Original Message ----- From: "Christopher Faylor" > On Thu, Nov 22, 2001 at 06:55:20PM +1100, Robert Collins wrote: > >When do you think you'll have time to tell me how you want the files > >laid out for the daemon, so we can do that post-1.3.4 merge? > > I thought I'd asked for feedback on the daemon. I don't recall getting > any. Has someone besides you looked at it? I don't think so. Uhmm, (fill in Roberts memory blank here - may Jonathan?) was going to , but I've heard nothing back. Corinna was interested in sockets vs DDE vs named pipes speed differences. Last actual discussion of merging was this thread: http://sources.redhat.com/ml/cygwin-patches/2001-q3/msg00151.html (Which was just Corinna and I). Anyway, I'm kinda unmotivated to put time into this at this point. The core code is complete, and AFAICT bugfree. The callback stuff works after a fashion, but I'd rip that out before a merge. (The callbacks are needed to fire off events after a process quits, allowing use to do for cygwin process's what win32 does for handles etc for mingw programs. - Again, think emulated FIFO's as a primary example of when that is needed. (And no, socketpairs and named pipes under win32 have the wrong behaviour to emulate FIFO's)). I've recently been looking at libsigc++ and it's looks pretty nice, so I'd like to use that to implement the callbacks - which is of course another discussion. (libsigc++ is GPL'd)..