From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: [PATCH 1/2] KVM: X86: remove read buffer for mmio read Date: Tue, 10 Jul 2012 19:04:29 +0300 Message-ID: <4FFC528D.8040900@redhat.com> References: <4FFA9E16.10001@linux.vnet.ibm.com> <4FFABD59.2070108@redhat.com> <20120709112346.GA23898@redhat.com> <4FFAD341.3080508@redhat.com> <4FFADB37.4090705@linux.vnet.ibm.com> <4FFADDFA.5000603@redhat.com> <20120710103629.GC23898@redhat.com> <4FFC07BB.8080802@redhat.com> <20120710104847.GD23898@redhat.com> <4FFC251F.4050504@redhat.com> <20120710130149.GE23898@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: Xiao Guangrong , Marcelo Tosatti , LKML , KVM To: Gleb Natapov Return-path: In-Reply-To: <20120710130149.GE23898@redhat.com> Sender: linux-kernel-owner@vger.kernel.org List-Id: kvm.vger.kernel.org On 07/10/2012 04:01 PM, Gleb Natapov wrote: > On Tue, Jul 10, 2012 at 03:50:39PM +0300, Avi Kivity wrote: > > On 07/10/2012 01:48 PM, Gleb Natapov wrote: > > > > > > > > > > > But the code is already here, why drop it? > > > > > > > > The read cache is not effective for multiple disjunct reads. > > > What do you mean? > > > > If an instruction reads from several sources in mmio, then the first > > read will be flushed from the cache by the second read. So if we need a > > third read, we'll have to exit for the first again. > > > The cache is flashed only before instruction decode, never during > emulation. Oh, I misread the code. -- I have a truly marvellous patch that fixes the bug which this signature is too narrow to contain.