From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754607Ab0HCIzY (ORCPT ); Tue, 3 Aug 2010 04:55:24 -0400 Received: from cn.fujitsu.com ([222.73.24.84]:51974 "EHLO song.cn.fujitsu.com" rhost-flags-OK-FAIL-OK-OK) by vger.kernel.org with ESMTP id S1752837Ab0HCIzW (ORCPT ); Tue, 3 Aug 2010 04:55:22 -0400 Message-ID: <4C57DA5D.9060700@cn.fujitsu.com> Date: Tue, 03 Aug 2010 16:59:09 +0800 From: Lai Jiangshan User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.1.9) Gecko/20100423 Thunderbird/3.0.4 MIME-Version: 1.0 To: Avi Kivity CC: Marcelo Tosatti , LKML , kvm@vger.kernel.org Subject: Re: [PATCH] kvm cleanup: Introduce sibling_pte and do cleanup for reverse map and parent_pte References: <4C577F45.9030208@cn.fujitsu.com> <4C57BC6D.8060306@redhat.com> In-Reply-To: <4C57BC6D.8060306@redhat.com> Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 08/03/2010 02:51 PM, Avi Kivity wrote: > On 08/03/2010 05:30 AM, Lai Jiangshan wrote: >> This patch is just a big cleanup. it reduces 220 lines of code. >> >> It introduces sibling_pte array for tracking identical sptes, so the >> identical sptes can be linked as a single linked list by their >> corresponding sibling_pte. A reverse map or a parent_pte points at >> the head of this single linked list. So we can do cleanup for >> reverse map and parent_pte VERY LARGELY. >> >> BAD: >> If most rmap have only one entry or most sp have only one parent, >> this patch may use more memory than before. > > That is the case with NPT and EPT. Each page has exactly one spte > (except a few vga pages), and each sp has exactly one parent_pte (except > the root pages). > >> GOOD: >> 1) Reduce a lot of code, The functions which are in hot path becomes >> very very simple and terrifically fast. >> 2) rmap_next(): O(N) -> O(1). traveling a ramp: O(N*N) -> O(N) > > The existing rmap_next() is not O(N), it's O(RMAP_EXT), which is 4. The > data structure was chosen over a simple linked list to avoid extra cache > misses. > >> 3) Remove the ugly interlayer: struct kvm_rmap_desc, struct >> kvm_pte_chain > > kvm_rmap_desc and kvm_pte_chain are indeed ugly, but they do save a lot > of memory and cache misses. > >> 4) We don't need to allocate any thing when we change the mappings. >> So we can avoid allocation when we have held kvm mmu spin lock. >> (this feature is very helpful in future). >> 5) better readability. > > I agree the new code is more readable. Unfortunately it uses more > memory and is likely to be slower. You add a cache miss for every spte, > while kvm_rmap_desc amortizes the cache miss among 4 sptes, and special > cases 1 spte to have no cache misses (or extra memory requirements). > You are right, please omit this patch thanks, lai.