From mboxrd@z Thu Jan 1 00:00:00 1970 From: Paolo Bonzini Subject: Re: [PATCH] KVM: x86: Avoid zapping mmio sptes twice for generation wraparound Date: Wed, 03 Jul 2013 10:53:51 +0200 Message-ID: <51D3E69F.2050104@redhat.com> References: <20130703171804.89d6cc2c.yoshikawa_takuya_b1@lab.ntt.co.jp> <51D3E093.3020408@redhat.com> <51D3E33D.1090704@linux.vnet.ibm.com> <51D3E5DC.5020902@linux.vnet.ibm.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Cc: Takuya Yoshikawa , gleb@redhat.com, kvm@vger.kernel.org To: Xiao Guangrong Return-path: Received: from mx1.redhat.com ([209.132.183.28]:13849 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932215Ab3GCIyD (ORCPT ); Wed, 3 Jul 2013 04:54:03 -0400 In-Reply-To: <51D3E5DC.5020902@linux.vnet.ibm.com> Sender: kvm-owner@vger.kernel.org List-ID: Il 03/07/2013 10:50, Xiao Guangrong ha scritto: >> > Please wait a while. I can not understand it very clearly. >> > >> > This conditional check will cause caching a overflow value into mmio spte. >> > The simple case is that kvm adds new slots for many times, the mmio-gen is easily >> > more than MMIO_MAX_GEN. >> > > Actually, the double zapping can be avoided by moving kvm_mmu_invalidate_mmio_sptes to > the end of install_new_memslots(). > > Yes, the actual operation would be the same as this patch. You can rename kvm_mmu_invalidate_mmio_sptes to kvm_arch_memslots_installed, or something like that. But it would have to touch all architectures. Paolo