From owner-freebsd-hackers Wed Feb 10 19:10:06 1999 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id TAA07770 for freebsd-hackers-outgoing; Wed, 10 Feb 1999 19:10:06 -0800 (PST) (envelope-from owner-freebsd-hackers@FreeBSD.ORG) Received: from spinner.netplex.com.au (spinner.netplex.com.au [202.12.86.3]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id TAA07718 for ; Wed, 10 Feb 1999 19:10:01 -0800 (PST) (envelope-from peter@netplex.com.au) Received: from spinner.netplex.com.au (localhost [127.0.0.1]) by spinner.netplex.com.au (8.9.2/8.9.2/Netplex) with ESMTP id LAA60544; Thu, 11 Feb 1999 11:08:47 +0800 (WST) (envelope-from peter@spinner.netplex.com.au) Message-Id: <199902110308.LAA60544@spinner.netplex.com.au> X-Mailer: exmh version 2.0.2 2/24/98 To: Matthew Dillon cc: Dag-Erling Smorgrav , Christoph Kukulies , Terry Lambert , hackers@FreeBSD.ORG Subject: Re: portability of shm, mmap, pipes and socket IPC In-reply-to: Your message of "Wed, 10 Feb 1999 12:14:45 PST." <199902102014.MAA85946@apollo.backplane.com> Date: Thu, 11 Feb 1999 11:08:44 +0800 From: Peter Wemm Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Matthew Dillon wrote: > > : > :Matthew Dillon writes: > :> The problem is that linux updates the timeval structure on return, > :> telling you how much time is left. > : > :Yup. I wish FreeBSD did that - the man page already states that one > :shouldn't rely on tv not being modified, so it shouldn't break POLA. > : > :DES > :-- > :Dag-Erling Smorgrav - des@flood.ping.uio.no > > Ick. It was a disaster.... and the feature is overrated anyway. I > was actually heavily involved with linux back in those days and I used > this select() feature myself, but the disadvantages outweighed the > advantages by an order of magnitude. It really isn't all that expensive > to do a separate gettimeofday() system call. I implemented it on FreeBSD back in 1996 or so but gave up in the end. The biggest offender was the libc RPC code, but there were a constant supply of things that mysteriously failed. It was a real nightmare trying to track down and locate them. There were things using timeouts of a day or a week or so, and would gradually reduce the timeout as select chipped it away and after a week or so things would mysteriously start running at a 100% CPU tight loop around select(). I don't think I have the code anymore, I lost it during an accident while working on poll() and never bothered to restore it from backup. > -Matt > Matthew Dillon > Cheers, -Peter To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message