From owner-cvs-all Thu Nov 12 13:17:57 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id NAA22841 for cvs-all-outgoing; Thu, 12 Nov 1998 13:17:57 -0800 (PST) (envelope-from owner-cvs-all@FreeBSD.ORG) Received: from dingo.cdrom.com (dingo.cdrom.com [204.216.28.145]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id NAA22836 for ; Thu, 12 Nov 1998 13:17:56 -0800 (PST) (envelope-from mike@dingo.cdrom.com) Received: from dingo.cdrom.com (localhost.cdrom.com [127.0.0.1]) by dingo.cdrom.com (8.9.1/8.8.8) with ESMTP id NAA00637; Thu, 12 Nov 1998 13:13:38 -0800 (PST) (envelope-from mike@dingo.cdrom.com) Message-Id: <199811122113.NAA00637@dingo.cdrom.com> X-Mailer: exmh version 2.0.2 2/24/98 To: Wayne Scott cc: cvs-all@FreeBSD.ORG, mike@smith.net.au Subject: Re: sendfile.2 (was Re: cvs commit: ...) In-reply-to: Your message of "Thu, 12 Nov 1998 11:46:46 PST." <199811121946.LAA02237@pdxlx008.pdx.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 12 Nov 1998 13:13:38 -0800 From: Mike Smith Sender: owner-cvs-all@FreeBSD.ORG Precedence: bulk > > That's about consistent with expectations; context switch from one > > process context to another is about as fast as it can be. Crossing the > > user:kernel protection boundary is not so easy - one of the major > > shortcomings of the i386 architecture it seems. > > > > Read the info on the SYSENTER instruction. > > Intel is finally documenting it properly. > > http://developer.intel.com/design/pentiumii/manuals/243689.htm > > It is an ugly hack, but you can switch from user to kernel twice as > fast as other methods. (PII only unfortunatly) Urk. This means recompiling all your user-space binaries, right? That's not going to be very popular. 8( -- \\ Sometimes you're ahead, \\ Mike Smith \\ sometimes you're behind. \\ mike@smith.net.au \\ The race is long, and in the \\ msmith@freebsd.org \\ end it's only with yourself. \\ msmith@cdrom.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message