From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754667Ab0D1C4T (ORCPT ); Tue, 27 Apr 2010 22:56:19 -0400 Received: from mga11.intel.com ([192.55.52.93]:57108 "EHLO mga11.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753683Ab0D1C4R (ORCPT ); Tue, 27 Apr 2010 22:56:17 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.52,285,1270450800"; d="scan'208";a="793411662" Subject: Re: [PATCH 2/2] KVM, Fix QEMU-KVM is killed by guest SRAO MCE From: Huang Ying To: Avi Kivity Cc: "linux-kernel@vger.kernel.org" , Andi Kleen , Andrew Morton , "masbock@linux.vnet.ibm.com" , "kvm@vger.kernel.org" In-Reply-To: <4BD6AEC7.5020609@redhat.com> References: <1272351860.24125.15.camel@yhuang-dev.sh.intel.com> <4BD69680.10402@redhat.com> <1272360341.24125.116.camel@yhuang-dev.sh.intel.com> <4BD6AEC7.5020609@redhat.com> Content-Type: text/plain; charset="UTF-8" Date: Wed, 28 Apr 2010 10:56:15 +0800 Message-ID: <1272423375.24125.189.camel@yhuang-dev.sh.intel.com> Mime-Version: 1.0 X-Mailer: Evolution 2.28.3 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, 2010-04-27 at 17:30 +0800, Avi Kivity wrote: > On 04/27/2010 12:25 PM, Huang Ying wrote: > > > > > >> On 04/27/2010 10:04 AM, Huang Ying wrote: > >> > >>> +static void kvm_send_hwpoison_signal(struct kvm *kvm, gfn_t gfn) > >>> +{ > >>> + char buf[1]; > >>> + void __user *hva; > >>> + int r; > >>> + > >>> + /* Touch the page, so send SIGBUS */ > >>> + hva = (void __user *)gfn_to_hva(kvm, gfn); > >>> + r = copy_from_user(buf, hva, 1); > >>> > >>> > >> No error check? What will a copy_from_user() of poisoned page expected > >> to return? > >> > >> Best to return -EFAULT on failure for consistency. > >> > > Just want to use the side effect of copy_from_user, SIGBUS will be sent > > to current process because the page touched is marked as poisoned. That > > is, failure is expected, so the return value is not checked. > > > > What if the failure doesn't happen? Say, someone mmap()ed over the page. Sorry, not get your idea clearly. hva is re-mmap()ed? We just read the hva, not write, so I think it should be OK here. > btw, better to use (void)copy_from_user(...) instead to avoid the > initialized but not used warning the compiler may generate. OK. Will do that. Best Regards, Huang Ying