From mboxrd@z Thu Jan 1 00:00:00 1970 From: Marcelo Tosatti Subject: Re: [patch 00/13] RFC: out of sync shadow Date: Mon, 8 Sep 2008 04:23:55 -0300 Message-ID: <20080908072355.GD1014@dmt.cnet> References: <20080906184822.560099087@localhost.localdomain> <48C3B987.6020803@qumranet.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: kvm@vger.kernel.org To: Avi Kivity Return-path: Received: from mx1.redhat.com ([66.187.233.31]:37758 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752031AbYIHHYz (ORCPT ); Mon, 8 Sep 2008 03:24:55 -0400 Content-Disposition: inline In-Reply-To: <48C3B987.6020803@qumranet.com> Sender: kvm-owner@vger.kernel.org List-ID: On Sun, Sep 07, 2008 at 02:22:47PM +0300, Avi Kivity wrote: >> VMX "bypass_guest_pf" feature on prefetch_page breaks it (since new >> PTE writes need no TLB flush, I assume). Not sure if its worthwhile to >> convert notrap_nonpresent -> trap_nonpresent on unshadow or just go >> for unconditional nonpaging_prefetch_page. >> >> > > Doesn't it kill bypass_guest_pf completely? As soon as we unsync a > page, we can't have nontrapping nonpresent ptes in it. Yes, thats what I meant. > We can try convertion on unsync, it does speed up demand paging. > >> And I think other optimizations are possible now, for example the guest >> can be responsible for remote TLB flushing on kvm_mmu_pte_write(). >> > > But kvm_mmu_pte_write() is no longer called, since we unsync? On higher level pagetables it still is. The IPI functions show high up in profiling (probably more due to write protection, but still).