From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S934133AbcLARcx (ORCPT ); Thu, 1 Dec 2016 12:32:53 -0500 Received: from foss.arm.com ([217.140.101.70]:50180 "EHLO foss.arm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753027AbcLARcv (ORCPT ); Thu, 1 Dec 2016 12:32:51 -0500 Subject: Re: Kernel Panics on Xen ARM64 for Domain0 and Guest To: "tj@kernel.org" References: <420a44c0-f86f-e6ab-44af-93ada7e01b58@arm.com> <20161128152915.GA7806@htj.duckdns.org> Cc: Wei Chen , "zijun_hu@htc.com" , "cl@linux.com" , "linux-mm@kvack.org" , "linux-kernel@vger.kernel.org" , "xen-devel@lists.xenproject.org" , Kaly Xin , Steve Capper , Stefano Stabellini From: Julien Grall Message-ID: Date: Thu, 1 Dec 2016 17:31:57 +0000 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101 Thunderbird/45.4.0 MIME-Version: 1.0 In-Reply-To: <20161128152915.GA7806@htj.duckdns.org> Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 28/11/16 15:29, tj@kernel.org wrote: > Hello, Hello, > On Mon, Nov 28, 2016 at 11:59:15AM +0000, Julien Grall wrote: >>> commit 3ca45a46f8af8c4a92dd8a08eac57787242d5021 >>> percpu: ensure the requested alignment is power of two >> >> It would have been useful to specify the tree used. In this case, >> this commit comes from linux-next. > > I'm surprised this actually triggered. > >> diff --git a/arch/arm/xen/enlighten.c b/arch/arm/xen/enlighten.c >> index f193414..4986dc0 100644 >> --- a/arch/arm/xen/enlighten.c >> +++ b/arch/arm/xen/enlighten.c >> @@ -372,8 +372,7 @@ static int __init xen_guest_init(void) >> * for secondary CPUs as they are brought up. >> * For uniformity we use VCPUOP_register_vcpu_info even on cpu0. >> */ >> - xen_vcpu_info = __alloc_percpu(sizeof(struct vcpu_info), >> - sizeof(struct vcpu_info)); >> + xen_vcpu_info = alloc_percpu(struct vcpu_info); >> if (xen_vcpu_info == NULL) >> return -ENOMEM; > > Yes, this looks correct. Can you please cc stable too? percpu > allocator never supported alignments which aren't power of two and has > always behaved incorrectly with alignments which aren't power of two. I will send the patch soon with stable CCed. Regards, -- Julien Grall