From mboxrd@z Thu Jan 1 00:00:00 1970 From: Marcelo Tosatti Subject: Re: [PATCH v2 5/6] KVM: MMU: remove 'clear_unsync' parameter Date: Thu, 18 Nov 2010 13:43:49 -0200 Message-ID: <20101118154349.GF14327@amt.cnet> References: <4CE355A0.6020907@cn.fujitsu.com> <4CE3565D.4080004@cn.fujitsu.com> <20101117164903.GE18959@amt.cnet> <4CE4D8C9.90302@cn.fujitsu.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Avi Kivity , LKML , KVM To: Xiao Guangrong Return-path: Received: from mx1.redhat.com ([209.132.183.28]:57692 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1759487Ab0KRQP7 (ORCPT ); Thu, 18 Nov 2010 11:15:59 -0500 Content-Disposition: inline In-Reply-To: <4CE4D8C9.90302@cn.fujitsu.com> Sender: kvm-owner@vger.kernel.org List-ID: On Thu, Nov 18, 2010 at 03:42:01PM +0800, Xiao Guangrong wrote: > On 11/18/2010 12:49 AM, Marcelo Tosatti wrote: > bool clear_unsync) > >> +static int FNAME(sync_page)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) > >> { > >> int i, offset, nr_present; > >> bool host_writable; > >> @@ -781,7 +780,7 @@ static int FNAME(sync_page)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, > >> u64 nonpresent; > >> > >> if (rsvd_bits_set || is_present_gpte(gpte) || > >> - !clear_unsync) > >> + sp->unsync) > >> nonpresent = shadow_trap_nonpresent_pte; > >> else > >> nonpresent = shadow_notrap_nonpresent_pte; > > > > Its better to keep this explicit as a parameter. > > > > But after patch 6 (KVM: MMU: cleanup update_pte, pte_prefetch and sync_page functions), > this parameter is not used anymore... i don't have strong opinion on it :-) On a second thought, using sp->unsync is fine.