From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:41742) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Vfuqq-0006Wl-VM for qemu-devel@nongnu.org; Mon, 11 Nov 2013 12:00:42 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1Vfuqk-00008m-Ur for qemu-devel@nongnu.org; Mon, 11 Nov 2013 12:00:36 -0500 Received: from mx1.redhat.com ([209.132.183.28]:52042) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Vfuqk-00008V-Ml for qemu-devel@nongnu.org; Mon, 11 Nov 2013 12:00:30 -0500 Received: from int-mx02.intmail.prod.int.phx2.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) by mx1.redhat.com (8.14.4/8.14.4) with ESMTP id rABH0SQt014796 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK) for ; Mon, 11 Nov 2013 12:00:29 -0500 Message-ID: <52810D0F.50500@redhat.com> Date: Mon, 11 Nov 2013 17:59:59 +0100 From: Paolo Bonzini MIME-Version: 1.0 References: <1384187560-20557-1-git-send-email-mst@redhat.com> In-Reply-To: <1384187560-20557-1-git-send-email-mst@redhat.com> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH RFC 0/5] making address spaces 64 bit wide List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: "Michael S. Tsirkin" Cc: qemu-devel@nongnu.org, Luiz Capitulino Il 11/11/2013 17:40, Michael S. Tsirkin ha scritto: > At the moment, exec ignores high bits in each address, > for efficiency. > This is incorrect: devices can do full 64 bit DMA, it's > only the CPU that is limited by target address space. > Using full 64 bit addresses was clocked at 12% performance > hit on a microbenchmark. > To solve, teach pagetables to skip bits at any level > and not just the lowest level. > > This should solve the performance problem (only one line > of code changed on the data path). > I'm still trying to figure out how to measure speed > properly with TCG, sending this out for early feedback and flames. I used this: x86_64-softmmu/qemu-system-x86_64 -kernel ../../kvm-unit-tests/x86/vmexit.flat -serial stdio -device isa-debug-exit,iobase=0xf4 with only one test enabled (I tried both inl_from_qemu and inl_from_pmtimer) and with roughly the same inlining of the "inb %dx, %al" instruction that you suggested earlier on the mailing list. Paolo