From mboxrd@z Thu Jan 1 00:00:00 1970 From: Paolo Bonzini Subject: Re: [PATCH 5/7] KVM: MMU: document fast page fault in Date: Wed, 19 Jun 2013 14:13:55 +0200 Message-ID: <51C1A083.9050209@redhat.com> References: <1371632965-20077-1-git-send-email-xiaoguangrong@linux.vnet.ibm.com> <1371632965-20077-6-git-send-email-xiaoguangrong@linux.vnet.ibm.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-15 Content-Transfer-Encoding: 7bit Cc: gleb@redhat.com, avi.kivity@gmail.com, mtosatti@redhat.com, linux-kernel@vger.kernel.org, kvm@vger.kernel.org To: Xiao Guangrong Return-path: In-Reply-To: <1371632965-20077-6-git-send-email-xiaoguangrong@linux.vnet.ibm.com> Sender: linux-kernel-owner@vger.kernel.org List-Id: kvm.vger.kernel.org Il 19/06/2013 11:09, Xiao Guangrong ha scritto: > Document fast page fault to Documentation/virtual/kvm/mmu.txt > > Signed-off-by: Xiao Guangrong > --- > Documentation/virtual/kvm/mmu.txt | 3 +++ > 1 file changed, 3 insertions(+) > > diff --git a/Documentation/virtual/kvm/mmu.txt b/Documentation/virtual/kvm/mmu.txt > index 4fb442b..b5ce7dd 100644 > --- a/Documentation/virtual/kvm/mmu.txt > +++ b/Documentation/virtual/kvm/mmu.txt > @@ -275,6 +275,9 @@ Handling a page fault is performed as follows: > information is stored and cache the information to vcpu->arch.mmio_gva, > vcpu->arch.access and vcpu->arch.mmio_gfn then call the emulator to emulate > the instruction who will get the benefit from the cached mmio info > + - If both P bit and R/W bit of error code are set, there has a chance to > + fast fix the page fault, please see the description of "Fast page fault" in > + Documentation/virtual/kvm/locking.txt - If both P bit and R/W bit of error code are set, this could possibly be handled as a "fast page fault" (fixed without taking the MMU lock). See the description in Documentation/virtual/kvm/locking.txt. Paolo > - if needed, walk the guest page tables to determine the guest translation > (gva->gpa or ngpa->gpa) > - if permissions are insufficient, reflect the fault back to the guest >