From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1759432Ab1CDNtN (ORCPT ); Fri, 4 Mar 2011 08:49:13 -0500 Received: from mail-fx0-f46.google.com ([209.85.161.46]:41847 "EHLO mail-fx0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1758521Ab1CDNtM (ORCPT ); Fri, 4 Mar 2011 08:49:12 -0500 DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=sender:date:from:to:cc:subject:message-id:references:mime-version :content-type:content-disposition:in-reply-to:user-agent; b=SRMMfXQmi1jcncJKjzlwGObalp/4QKEmra7xoWZ1GhbdA71vlcRKu/CU0RK8tCUa/i neHcgvqcnQL63hHo2SwQbsbbCCwnM1IaOtPDg/Wk0Y/juE6pFIsqv76X74mwxZJ6T9gK tvQziilvas7Y2BD4AhvBC4ViPZiYoKmhY0ZBg= Date: Fri, 4 Mar 2011 14:47:36 +0100 From: Tejun Heo To: Yinghai Lu Cc: Thomas Gleixner , Ingo Molnar , "H. Peter Anvin" , David Rientjes , Andrew Morton , "linux-kernel@vger.kernel.org" Subject: Re: [PATCH for.38] x86, numa: Fix numa_emulation code with node0 without RAM Message-ID: <20110304134736.GN20499@htj.dyndns.org> References: <4D6ECF72.5010308@kernel.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4D6ECF72.5010308@kernel.org> User-Agent: Mutt/1.5.20 (2009-06-14) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Mar 02, 2011 at 03:14:58PM -0800, Yinghai Lu wrote: > > On one system that does not have RAM on node0. > > when numa_emulation is compiled in, and > 1. boot system without numa=fake... > 2. or boot system with numa=fake=128 to make emulation fail > > will get: > > [ 0.076025] ------------[ cut here ]------------ > [ 0.080004] kernel BUG at arch/x86/mm/numa_64.c:788! > [ 0.080004] invalid opcode: 0000 [#1] SMP > [ 0.080004] last sysfs file: > [ 0.080004] CPU 0 > [ 0.080004] Modules linked in: > [ 0.080004] > [ 0.080004] Pid: 0, comm: swapper Not tainted 2.6.38-rc6-yh-00212-g3e1f235 #1 Sun Microsystems Sun Fire X4240/Sun Fire X4240 > [ 0.080004] RIP: 0010:[] [] numa_add_cpu+0x8a/0x156 > ... > [ 0.080004] Call Trace: > [ 0.080004] [] identify_cpu+0x2d7/0x2df > [ 0.080004] [] identify_boot_cpu+0x10/0x30 > [ 0.080004] [] check_bugs+0x9/0x2d > [ 0.080004] [] start_kernel+0x3bf/0x3d9 > [ 0.080004] [] x86_64_start_reservations+0xb6/0xba > [ 0.080004] [] x86_64_start_kernel+0xf7/0xfe > [ 0.080004] Code: 74 06 48 8d 04 90 eb 0f 48 c7 c0 10 d9 00 00 48 03 04 d5 50 e1 5f 82 8b 00 83 f8 ff 74 0d 0f a3 05 7a f9 92 00 19 d2 85 d2 75 02 <0f> 0b 48 98 31 db 48 8b 04 c5 d0 ec 5f 82 48 8b 90 90 af 00 00 > > need to used early_cpu_to_node directly, because cpu_to_apicid and apicid_to_node > will return node0 that is not onlined. > > Signed-off-by: Yinghai Lu Acked-by: Tejun Heo Thanks. -- tejun