From owner-freebsd-hackers Fri Jun 28 18:43:14 1996 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id SAA23749 for hackers-outgoing; Fri, 28 Jun 1996 18:43:14 -0700 (PDT) Received: from terra.Sarnoff.COM (terra.sarnoff.com [130.33.11.203]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id SAA23740 for ; Fri, 28 Jun 1996 18:43:10 -0700 (PDT) Received: (from rminnich@localhost) by terra.Sarnoff.COM (8.6.12/8.6.12) id VAA20442; Fri, 28 Jun 1996 21:40:52 -0400 Date: Fri, 28 Jun 1996 21:40:52 -0400 (EDT) From: "Ron G. Minnich" X-Sender: rminnich@terra To: Jukka Ukkonen cc: hackers@freebsd.com Subject: Re: physical memory addresses & memory locking... In-Reply-To: <199606282219.BAA23320@jau.csc.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk I have a simple lock LKM which in essence allows processes to 1) test a lock 2) drop into kernel sleep() if it's set 3) if it's not set, lock it using pentium cswap 4) when clearing it, if there was contention, syscall that does a wakeup() it's very efficient, much more so than sysv semaphores. If you want it let me know. It's called fastlock. ron Ron Minnich |"Inferno runs on MIPS ..., Intel ..., and AMD's rminnich@sarnoff.com |29-kilobit-per-second chip-based architectures ..." (609)-734-3120 | Comm. week, may 13, pg. 4. ftp://ftp.sarnoff.com/pub/mnfs/www/docs/cluster.html