Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Sat, 1 Mar 2003 03:26:54 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Sat, 1 Mar 2003 03:26:54 -0500 Received: from modemcable092.130-200-24.mtl.mc.videotron.ca ([24.200.130.92]:64090 "EHLO montezuma.mastecende.com") by vger.kernel.org with ESMTP id ; Sat, 1 Mar 2003 03:26:54 -0500 Date: Sat, 1 Mar 2003 03:35:24 -0500 (EST) From: Zwane Mwaikambo X-X-Sender: zwane@montezuma.mastecende.com To: Linux Kernel Subject: [PATCH][RFC][0/2] per-CPU IDTs aka booting high irq count ia32 boxes Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1034 Lines: 25 I'm currently tackling booting a box with over 8 IOAPICs which results in 320 interrupt sources. We won't be able to take all 320 obviously and stuck with 255 but i have a scheme for trying to 'increase' the global vector space by only installing vectors for irqs which the cpus on that node will be servicing. To achieve this i've first setup per cpu IDTs and then splitting up vector allocation to allocate per node. I'm currently booting on a 16way but will be wiring up a 32way over the weekend. dmesg: http://function.linuxpower.ca/patches/numaq/dmesg-4quad-percpu_idt mptable: http://function.linuxpower.ca/patches/numaq/mptables-numaq-4quad I'd appreciate any comments on the following patches. Boot and run on UP, SMP, NUMAQ/SMP Thanks, Zwane -- function.linuxpower.ca - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/