Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757986Ab2EVDET (ORCPT ); Mon, 21 May 2012 23:04:19 -0400 Received: from haggis.pcug.org.au ([203.10.76.10]:46409 "EHLO members.tip.net.au" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757589Ab2EVDEA (ORCPT ); Mon, 21 May 2012 23:04:00 -0400 Date: Tue, 22 May 2012 13:03:54 +1000 From: Stephen Rothwell To: David Rientjes Cc: LKML , linux-next@vger.kernel.org, ppc-dev , Thomas Gleixner , Ingo Molnar , "H. Peter Anvin" , Peter Zijlstra , Lee Schermerhorn , Linus Subject: Re: linux-next: PowerPC boot failures in next-20120521 Message-Id: <20120522130354.fdb335eb294f4206b4b2fed5@canb.auug.org.au> In-Reply-To: References: <20120522114051.0c9db9a7c2d660bc9e0e1be2@canb.auug.org.au> X-Mailer: Sylpheed 3.2.0beta7 (GTK+ 2.24.10; i486-pc-linux-gnu) Mime-Version: 1.0 Content-Type: multipart/signed; protocol="application/pgp-signature"; micalg="PGP-SHA256"; boundary="Signature=_Tue__22_May_2012_13_03_54_+1000_haxAE.AciyWA7TxE" Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2517 Lines: 67 --Signature=_Tue__22_May_2012_13_03_54_+1000_haxAE.AciyWA7TxE Content-Type: text/plain; charset=US-ASCII Content-Disposition: inline Content-Transfer-Encoding: quoted-printable Hi David, On Mon, 21 May 2012 18:53:37 -0700 (PDT) David Rientjes wrote: > > Yeah, it's sched/numa since that's what introduced numa_init(). It does= =20 > for_each_node() for each node and does a kmalloc_node() even though that= =20 > node may not be online. Slub ends up passing this node to the page=20 > allocator through alloc_pages_exact_node(). CONFIG_DEBUG_VM would have=20 > caught this and your config confirms its not enabled. >=20 > sched/numa either needs a memory hotplug notifier or it needs to pass=20 > NUMA_NO_NODE for nodes that aren't online. Until we get the former, the= =20 > following should fix it. >=20 >=20 > sched, numa: Allocate node_queue on any node for offline nodes >=20 > struct node_queue must be allocated with NUMA_NO_NODE for nodes that are= =20 > not (yet) online, otherwise the page allocator has a bad zonelist. >=20 > Signed-off-by: David Rientjes Thanks, that fixes it. Tested-by: Stephen Rothwell --=20 Cheers, Stephen Rothwell sfr@canb.auug.org.au --Signature=_Tue__22_May_2012_13_03_54_+1000_haxAE.AciyWA7TxE Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.12 (GNU/Linux) iQIcBAEBCAAGBQJPuwIaAAoJEECxmPOUX5FEoj8P/izwzpFzKBIxDEt4t+Nx8b5a KEHdgnB0XA7jd6W5O5cr+QikpYBqoZso1iSzO72bLcDgG3hSiX2yuXexFNCerVfI nKFZtsbyZ2G5NhZKaQFdglMtTpAraUBVoJevEF3bB/miijBgK4cnDgwVpX0mKlFd QL3SBgEoUVBF7Qdy8HQD204fqO015ykQyTn6flsWA/S7eVEQ07eOhsFCY5NhwDHb V8h++WCvdrdSd/HyiIlgJMM3SOIjcd0SrbklIaR0wk2mAXyj7Tvz96WOwad4WR4Z SMt3DityCY6oS/QR1Bd5zcNTDz8FKd8adPM8JyyGSx9AwLu6tSAf8toEadncRLTp LdMm+aoysJCX9fEJNwC9n8w14ubTSyvP0135MZythF3RW27yKuOgaJ5P3BgGhn/I 8ZpukpHLfMlWWT2jrSCJt63fjyCi4++SSFk7lZPHZTvgNns/LQjBeVkwZH3IJUux NXzqpGFlgrjpiS/mYRZyL6tY9Z71cI141PUPPfPUwBLi8F2So9dAHRM6krsKjvY4 9FMzW+ax5Slnt1Q/7+ajcODC+UFfYDAtanwszAdKYfTRQxTIj9Gnmx8WhdBkFdz9 UtKsxXe5yyo2t4s/kyPX3aOIRmzZYp6xgcRpkLblcXf6mNVd2vGj/ozR7KNTSH/2 JinIuO9RyoOq4VwvtALM =BcqL -----END PGP SIGNATURE----- --Signature=_Tue__22_May_2012_13_03_54_+1000_haxAE.AciyWA7TxE-- -- 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/