From owner-freebsd-hackers Wed Feb 10 22:25:44 1999 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id WAA29594 for freebsd-hackers-outgoing; Wed, 10 Feb 1999 22:25:44 -0800 (PST) (envelope-from owner-freebsd-hackers@FreeBSD.ORG) Received: from apollo.backplane.com (apollo.backplane.com [209.157.86.2]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id WAA29589 for ; Wed, 10 Feb 1999 22:25:43 -0800 (PST) (envelope-from dillon@apollo.backplane.com) Received: (from dillon@localhost) by apollo.backplane.com (8.9.2/8.9.1) id WAA88612; Wed, 10 Feb 1999 22:25:29 -0800 (PST) (envelope-from dillon) Date: Wed, 10 Feb 1999 22:25:29 -0800 (PST) From: Matthew Dillon Message-Id: <199902110625.WAA88612@apollo.backplane.com> To: Terry Lambert Cc: dyson@iquest.net, dg@root.com, kuku@gilberto.physik.RWTH-Aachen.DE, hackers@FreeBSD.ORG Subject: Re: portability of shm, mmap, pipes and socket IPC References: <199902110220.TAA08932@usr06.primenet.com> Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG :> :> ... and has absolutely nothing to do with the question that John was :> replying to, which was related to the fault overhead/expense of using :> mmap() verses a SYS-V shared memory segment. : :But has everything to do with his statement to the effect that :"There is no effective difference", contained in the answer. The :rest was quoted for context. I am not so stupid that I cannot :fathom his answer, as you would see if you read the rest of the :messages in this thread, where I said "Mea culpa" ("My Fault") to :David's quotation of John's reply in a different context. : : : Terry Lambert : terry@lambert.org The issue of file-backed storage verses anon mmap is certainly real, but it's divergent enough from the original discussion that it should have been brought up as a separate item rather then as a response, That's all. This brings up a good point, though... I think it might make sense to be able to specify a MAP_ flag to mmap to indicate that the file's dirty data backing the map ( for non-anonymous maps ) does not have to be synced by the syncer. That would make mmap() a much more useful tool for sharing working sets. The file data would be synced on the last close of the vnode and if the program really doesn't want it to be synced at all, the program could madvise() it free ( and we could fix madvise() to actually support throw-away on a vnode ). Another possibility would be adding a memory reference passing capability to unix-domain socket messaging. We can already pass descriptors and access rights, why not memory too? ( ala mach ). -Matt Matthew Dillon To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message