From: Avi Kivity Subject: Re: One (possible) x86 get_user_pages bug Date: Sun, 30 Jan 2011 15:01:45 +0200 Message-ID: <4D456139.4090508@redhat.com> References: <4D416D9A.9010603@huawei.com> <4D419416020000780002ECB7@vpn.id2.novell.com> <4D41B90D.5000305@goop.org> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <4D41B90D.5000305@goop.org> Sender: linux-kernel-owner@vger.kernel.org To: Jeremy Fitzhardinge Cc: Jan Beulich , Xiaowei Yang , Nick Piggin , Peter Zijlstra , fanhenglong@huawei.com, Kaushik Barde , Kenneth Lee , linqaingmin , wangzhenguo@huawei.com, Wu Fengguang , "xen-devel@lists.xensource.com" , linux-kernel@vger.kernel.org, Marcelo Tosatti List-Id: xen-devel@lists.xenproject.org On 01/27/2011 08:27 PM, Jeremy Fitzhardinge wrote: > And even just considering virtualization, having non-IPI-based tlb > shootdown is a measurable performance win, since a hypervisor can > optimise away a cross-VCPU shootdown if it knows no physical TLB > contains the target VCPU's entries. I can imagine the KVM folks could > get some benefit from that as well. It's nice to avoid the IPI (and waking up a cpu if it happens to be asleep) but I think the risk of deviating too much from the baremetal arch is too large, as demonstrated by this bug. (well, async page faults is a counterexample, I wonder if/when it will bite us) -- error compiling committee.c: too many arguments to function