From mboxrd@z Thu Jan 1 00:00:00 1970 From: Xiao Guangrong Subject: Re: [PATCH 03/11] KVM: x86: fast emulate repeat string write instructions Date: Wed, 27 Jul 2011 09:47:52 +0800 Message-ID: <4E2F6E48.7030802@cn.fujitsu.com> References: <4E2EA3DB.7040403@cn.fujitsu.com> <4E2EA476.9070607@cn.fujitsu.com> <20110726122710.GM4404@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Cc: Avi Kivity , Marcelo Tosatti , LKML , KVM To: Gleb Natapov Return-path: In-Reply-To: <20110726122710.GM4404@redhat.com> Sender: linux-kernel-owner@vger.kernel.org List-Id: kvm.vger.kernel.org On 07/26/2011 08:27 PM, Gleb Natapov wrote: > On Tue, Jul 26, 2011 at 07:26:46PM +0800, Xiao Guangrong wrote: >> We usually use repeat string instructions to clear the page, for example, > By "we" do you mean Linux guest? > I do not know other guests except linux, but, generally rep instruction is not used to update a page table which is been using. >> we call memset to clear a page table, stosb is used in this function, and >> repeated for 1024 times, that means we should occupy mmu lock for 1024 times >> and walking shadow page cache for 1024 times, it is terrible >> >> In fact, if it is the repeat string instructions emulated and it is not a >> IO/MMIO access, we can zap all the corresponding shadow pages and return to the >> guest, then the mapping can became writable and directly write the page >> > So this patch does two independent things as far as I can see. First it > stops reentering guest if rep instruction is done on memory and second No. Oppositely, it enters guest as soon as possible if rep instruction is done on memory ;-) After this patch, we only need to emulate one time for a rep instruction. > it drops shadow pages if access to shadowed page table is rep. Why not > separate those in different patches? Umm, i will zap shadow page firstly and stop emulation rep instruction in the second patch. > BTW not entering guest periodically > increases interrupt latencies. Oppositely, It reduces the latencies. :-) > Why not zap shadow, make page writable > and reenter the guest instead of emulation, it should be much faster (do we > care to optimize for old cpus by complicating the code anyway?). > We do better lazily update the mapping to writable, it can be done by the later access.