From owner-freebsd-current Sun Mar 26 4:31: 3 2000 Delivered-To: freebsd-current@freebsd.org Received: from finch-post-12.mail.demon.net (finch-post-12.mail.demon.net [194.217.242.41]) by hub.freebsd.org (Postfix) with ESMTP id 3073C37B7E7 for ; Sun, 26 Mar 2000 04:30:58 -0800 (PST) (envelope-from dfr@nlsystems.com) Received: from nlsys.demon.co.uk ([158.152.125.33] helo=herring.nlsystems.com) by finch-post-12.mail.demon.net with esmtp (Exim 2.12 #1) id 12ZCC0-00050r-0C; Sun, 26 Mar 2000 12:30:56 +0000 Received: from salmon.nlsystems.com (salmon.nlsystems.com [10.0.0.3]) by herring.nlsystems.com (8.9.3/8.8.8) with ESMTP id NAA98838; Sun, 26 Mar 2000 13:33:57 +0100 (BST) (envelope-from dfr@nlsystems.com) Date: Sun, 26 Mar 2000 13:37:50 +0100 (BST) From: Doug Rabson To: Matthew Dillon Cc: Alfred Perlstein , freebsd-current@freebsd.org Subject: Re: SMP/BGL patch 04 In-Reply-To: <200003250216.SAA18264@apollo.backplane.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Fri, 24 Mar 2000, Matthew Dillon wrote: > Patch 04 is ready. > > http://www.backplane.com/FreeBSD4/ > http://www.backplane.com/FreeBSD4/smp-patch-04.diff > > Contains lots of cleanup of stale SMP code. There are still a few places > where get_mplock is being called with interrupts disabled which I haven't > found, so I put the sti test back in. > > I also removed the FAST_SIMPLELOCK optimization, which had been turned > on - mainly because it's not useful if we are moving to an interrupt > thread scheme (as in it's still a global interrupt lock, whereas the > thread scheme will allow concurrent interrupt execution), but also > because it makes too many assumptions about what can run outside the > MP lock. > > I'm going to let people bang on this for a few days, and then I think > it should be committed into -CURRENT (5.x) in order to allow people > to start banging on optimizing the BGL/syscall paths. All comments > welcome (but not necessary acted upon) :-). I haven't looked at the patch so I can't comment on it (but it sounds very promising). I have been writing some of the infrastructure for SMP on alpha and as an experiment I've been using the mutex_t primitive from BSD/OS as an experiment. This is a nice simple api for a counting mutex which BSD/OS 4.1 uses for its BGL. I have not yet seen the new BSD/OS smp code but it seemed that using mutex_t might make it marginally easier to work with that. -- Doug Rabson Mail: dfr@nlsystems.com Nonlinear Systems Ltd. Phone: +44 181 442 9037 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message