From owner-freebsd-smp Sun Sep 29 17:32:22 1996 Return-Path: owner-smp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id RAA23298 for smp-outgoing; Sun, 29 Sep 1996 17:32:22 -0700 (PDT) Received: from uruk.org (uruk.org [198.145.95.253]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id RAA21228 for ; Sun, 29 Sep 1996 17:29:49 -0700 (PDT) From: erich@uruk.org Received: from loopback (loopback [127.0.0.1]) by uruk.org (8.7.4/8.7.3) with SMTP id RAA29056; Sun, 29 Sep 1996 17:29:59 -0700 (PDT) Message-Id: <199609300029.RAA29056@uruk.org> X-Authentication-Warning: uruk.org: Host loopback [127.0.0.1] didn't use HELO protocol To: freebsd-smp@freebsd.org, linux-smp@vger.rutgers.edu cc: alan@cymru.net, peter@spinner.dialix.com, smp@csn.net, terry@lambert.org Subject: Generic SMP startup sequence and some docs Date: Sun, 29 Sep 1996 17:29:02 -0700 Sender: owner-smp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I finally have most of it almost done (complaints about work, other projects, etc. appropriate here ;-). The main part missing is the inclusion of most of the the CPU startup sequence itself, a generic IPI sender routine of some sort (still thinking on that one a bit), and one more look over the "default" configuration startup (there's some things missing right now, though they are only relevant to using the APIC in Symmetric I/O modes). Look on my MPS webpage in the "Example Code/Clarifications" section for details: http://www.uruk.org/~erich/mps.html Several more details about how to support integrated vs. discrete APICs are in this code and docs. When I get more of the Symmetic I/O stuff done, yet more will come out. One interesting detail is to note that in general, the other CPUs are already set up for generic symmetic delivery... all you have to do is to enable the LINTIN0 (ISA interrupts) and LINTIN1 (NMI interrupt) entries on each CPU, as they should already be programmed. The "local interrupt" entries of the MP Configuration Table, if set to "0xFF" APIC destination, are already set up on all the CPUs. The code I have parses all these, plus sets up a "fake" table which calls all the appropriate callbacks for the default configurations (the debugging I mentioned is making sure all the entries are correct). Anyway... I'll send more soon. -- Erich Stefan Boleyn \_ E-mail (preferred): Mad Genius wanna-be, CyberMuffin \__ (finger me for other stats) Web: http://www.uruk.org/~erich/ Motto: "I'll live forever or die trying"