Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756899Ab1F1JnQ (ORCPT ); Tue, 28 Jun 2011 05:43:16 -0400 Received: from mail-fx0-f52.google.com ([209.85.161.52]:58104 "EHLO mail-fx0-f52.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755518Ab1F1JlN (ORCPT ); Tue, 28 Jun 2011 05:41:13 -0400 Date: Tue, 28 Jun 2011 11:41:07 +0200 From: Tejun Heo To: Conny Seidel , Ingo Molnar , "H. Peter Anvin" , Thomas Gleixner Cc: hans.rosenfeld@amd.com, linux-kernel@vger.kernel.org, Christoph Lameter Subject: [PATCH tip:x86/urgent] x86-32, NUMA: Fix boot regression caused by NUMA init unification on highmem machines Message-ID: <20110628094107.GB3386@htj.dyndns.org> References: <20110621174131.054f0422.conny.seidel_amd.com@marah.osrc.amd.com> <20110626102235.GC12200@mtj.dyndns.org> <20110626223807.47cef5c6.conny.seidel_amd.com@marah.osrc.amd.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20110626223807.47cef5c6.conny.seidel_amd.com@marah.osrc.amd.com> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 3260 Lines: 79 During 32/64 NUMA init unification, commit 797390d855 "x86-32, NUMA: use sparse_memory_present_with_active_regions()" made 32bit mm init call memory_present() automatically from active_regions instead of leaving it to each NUMA init path. This commit description is inaccurate - memory_present() calls aren't the same for flat and numaq. After the commit, memory_present() is only called for the intersection of e820 and NUMA layout. Before, on flatmem, memory_present() would be called from 0 to max_pfn. After, it would be called only on the areas that e820 indicates to be populated. This is how x86_64 works and should be okay as memmap is allowed to contain holes; however, x86_32 DISCONTIGMEM is missing early_pfn_valid(), which makes memmap_init_zone() assume that memmap doesn't contain any hole. This leads to the following oops if e820 map contains holes as it often does on machine with near or more 4GiB of memory by calling pfn_to_page() on a pfn which isn't mapped to a NUMA node. BUG: unable to handle kernel paging request at 000012b0 IP: [] memmap_init_zone+0x6c/0xf2 *pdpt =3D 0000000000000000 *pde =3D f000eef3f000ee00 Oops: 0000 [#1] SMP last sysfs file: Modules linked in: Pid: 0, comm: swapper Not tainted 2.6.39-rc5-00164-g797390d #1 To Be Filled By O.E.M. To Be Filled By O.E.M./E350M1 EIP: 0060:[] EFLAGS: 00010012 CPU: 0 EIP is at memmap_init_zone+0x6c/0xf2 EAX: 00000000 EBX: 000a8000 ECX: 000a7fff EDX: f2c00b80 ESI: 000a8000 EDI: f2c00800 EBP: c19ffe54 ESP: c19ffe34 DS: 007b ES: 007b FS: 00d8 GS: 0000 SS: 0068 Process swapper (pid: 0, ti=3Dc19fe000 task=3Dc1a07f60 task.ti=3Dc19fe000) Stack: 00000002 00000000 0023f000 00000000 10000000 00000a00 f2c00000 f2c00b58 c19ffeb0 c1a80f24 000375fe 00000000 f2c00800 00000800 00000100 00000030 c1abb768 0000003c 00000000 00000000 00000004 00207a02 f2c00800 000375fe Call Trace: [] free_area_init_node+0x358/0x385 [] free_area_init_nodes+0x420/0x487 [] paging_init+0x114/0x11b [] setup_arch+0xb37/0xc0a [] start_kernel+0x76/0x316 [] i386_start_kernel+0xa8/0xb0 This patch fixes the bug by defining early_pfn_valid() to be the same as pfn_valid() when DISCONTIGMEM. Signed-off-by: Tejun Heo Reported-and-bisected-by: Conny Seidel LKML-Reference: <20110621174131.054f0422.conny.seidel_amd.com@marah.osrc.amd.com> --- Conny, can you please verify this fixes the boot problem you're seeing? Thanks. arch/x86/include/asm/mmzone_32.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/x86/include/asm/mmzone_32.h b/arch/x86/include/asm/mmzone_32.h index 5e83a41..756d2a7 100644 --- a/arch/x86/include/asm/mmzone_32.h +++ b/arch/x86/include/asm/mmzone_32.h @@ -68,6 +68,8 @@ static inline int pfn_valid(int pfn) return 0; } +#define early_pfn_valid(pfn) pfn_valid((pfn)) + #endif /* CONFIG_DISCONTIGMEM */ #ifdef CONFIG_NEED_MULTIPLE_NODES -- 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/