From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261325AbUBYUYI (ORCPT ); Wed, 25 Feb 2004 15:24:08 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261395AbUBYUYI (ORCPT ); Wed, 25 Feb 2004 15:24:08 -0500 Received: from fmr04.intel.com ([143.183.121.6]:25307 "EHLO caduceus.sc.intel.com") by vger.kernel.org with ESMTP id S261325AbUBYUYE (ORCPT ); Wed, 25 Feb 2004 15:24:04 -0500 Subject: Re: 2.6.3 doesn't see my 2nd CPU From: Len Brown To: Phil White Cc: linux-kernel@vger.kernel.org In-Reply-To: References: Content-Type: text/plain Organization: Message-Id: <1077740640.5913.433.camel@dhcppc4> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.2.3 Date: 25 Feb 2004 15:24:00 -0500 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org > CONFIG_HIGHMEM64G=y > CONFIG_X86_PAE=y > ACPI: S3 and PAE do not like each other for now, S3 disabled. Try disabling PAE mode -- it only slows down your 1GB system anyway and who knows -- maybe that is what broke. > Booting processor 1/1 eip 2000 > Unable to handle kernel paging request<1>Unable to handle kernel > paging request at virtual address 3f83ec0d re: acpi=force ACPI and MPS both recognize your 2nd processor as LAPIC ID 1, and the smpboot code that tries to start it up and fails is common to both modes -- so I don't see an MPS or ACPI-specific issue here. > Another update. The second processor still isn't working, > but with args noapic, noirqbalance, and > maxcpus=4 (even though there's only 2 cpus), > the APIC problems go away. They're replaced with > spurious interrupts however. I don't see these options having any effect on the root problem, which is that the kernel faults when starting the 2nd processor. Note that "noapic" technically isn't supported on an MPS compliant SMP system. -Len