From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:34357) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1VDnZ3-0003gF-Gy for qemu-devel@nongnu.org; Sun, 25 Aug 2013 23:34:09 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1VDnYv-000399-PV for qemu-devel@nongnu.org; Sun, 25 Aug 2013 23:34:01 -0400 Received: from e28smtp04.in.ibm.com ([122.248.162.4]:49491) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1VDnYv-00038j-3F for qemu-devel@nongnu.org; Sun, 25 Aug 2013 23:33:53 -0400 Received: from /spool/local by e28smtp04.in.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Mon, 26 Aug 2013 08:55:41 +0530 From: "Aneesh Kumar K.V" In-Reply-To: <1377465202.3819.25.camel@pasglop> References: <1377235210-27093-1-git-send-email-aneesh.kumar@linux.vnet.ibm.com> <1377235210-27093-3-git-send-email-aneesh.kumar@linux.vnet.ibm.com> <31707B5A-C56F-4343-8851-1DD516A099AD@suse.de> <1377465202.3819.25.camel@pasglop> Date: Mon, 26 Aug 2013 09:03:42 +0530 Message-ID: <87txiddt2h.fsf@linux.vnet.ibm.com> MIME-Version: 1.0 Content-Type: text/plain Subject: Re: [Qemu-devel] [Qemu-ppc] [PATCH -V3 2/4] target-ppc: Fix page table lookup with kvm enabled List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Benjamin Herrenschmidt , Alexander Graf Cc: Paul Mackerras , "qemu-ppc@nongnu.org list:PowerPC" , "qemu-devel@nongnu.org qemu-devel" Benjamin Herrenschmidt writes: > On Sun, 2013-08-25 at 19:32 +0100, Alexander Graf wrote: >> > + * At this point we are only interested in reading only bolted >> entries >> > + */ >> > + ghf.flags = KVM_GET_HTAB_BOLTED_ONLY; >> > + ghf.start_index = index; >> > + htab_fd = kvm_vm_ioctl(kvm_state, KVM_PPC_GET_HTAB_FD, &ghf); >> >> We should cache this. >> > Also why bolted only ? because non bolted entries could be invalidated and reused by the time we look at the returned hpte values. I am not sure, whether it is ok or we need to make sure such a thing doesn't happen. For the use case i am looking at, ie, to dump the kernel address looking at bolted entries was enough. -aneesh