From mboxrd@z Thu Jan 1 00:00:00 1970 From: Gleb Natapov Subject: Re: [PATCH v4 2/2] KVM: VMX: Add Posted Interrupt supporting Date: Sat, 23 Feb 2013 21:59:36 +0200 Message-ID: <20130223195936.GE25155@redhat.com> References: <1361540552-2016-1-git-send-email-yang.z.zhang@intel.com> <1361540552-2016-3-git-send-email-yang.z.zhang@intel.com> <20130223134309.GA2632@amt.cnet> <20130223143530.GA25155@redhat.com> <20130223144854.GA5618@amt.cnet> <20130223153144.GB25155@redhat.com> <20130223170513.GB8236@amt.cnet> <20130223194214.GD25155@redhat.com> <20130223195259.GA24760@amt.cnet> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: "Zhang, Yang Z" , "kvm@vger.kernel.org" , "avi.kivity@gmail.com" , "Zhang, Xiantao" To: Marcelo Tosatti Return-path: Received: from mx1.redhat.com ([209.132.183.28]:3072 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757539Ab3BWT7k (ORCPT ); Sat, 23 Feb 2013 14:59:40 -0500 Content-Disposition: inline In-Reply-To: <20130223195259.GA24760@amt.cnet> Sender: kvm-owner@vger.kernel.org List-ID: On Sat, Feb 23, 2013 at 04:52:59PM -0300, Marcelo Tosatti wrote: > > > OTOH spinlock is not the end of the world, can figure out something later > > > (we've tried without success so far). > > It serializes all injections into vcpu. I do not believe now that even > > with lock we are safe for the reason I mention above. We can use pir->on > > bit as a lock, but that only emphasise how ridiculous serialization of > > injections becomes. > > Please review the 2nd iteration of pseudocode in patchset v4 thread, it > should be good. Can you point me to what exactly I should look at? All I see is a discussion on how to no miss an IPI, but this is not the problem I am talking about here. -- Gleb.