From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1NuqVC-0005W2-33 for qemu-devel@nongnu.org; Thu, 25 Mar 2010 13:05:50 -0400 Received: from [140.186.70.92] (port=33295 helo=eggs.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1NuqV5-0004vH-83 for qemu-devel@nongnu.org; Thu, 25 Mar 2010 13:05:49 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.69) (envelope-from ) id 1NuqGY-0006GU-3y for qemu-devel@nongnu.org; Thu, 25 Mar 2010 12:50:43 -0400 Received: from mx1.redhat.com ([209.132.183.28]:45168) by eggs.gnu.org with esmtp (Exim 4.69) (envelope-from ) id 1NuqGX-0006GK-Sw for qemu-devel@nongnu.org; Thu, 25 Mar 2010 12:50:42 -0400 Message-ID: <4BAB9456.2010503@redhat.com> Date: Thu, 25 Mar 2010 18:50:30 +0200 From: Avi Kivity MIME-Version: 1.0 References: <1269497376-21903-1-git-send-email-cam@cs.ualberta.ca> <20100325091552.GB11153@redhat.com> <8286e4ee1003250918x659e87ai9e5ca7c68c23a7c3@mail.gmail.com> <4BAB8DE9.9080603@codemonkey.ws> <4BAB900F.5030104@redhat.com> <20100325164052.GB16506@redhat.com> In-Reply-To: <20100325164052.GB16506@redhat.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Subject: [Qemu-devel] Re: [PATCH v3 1/1] Shared memory uio_pci driver List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: "Michael S. Tsirkin" Cc: Cam Macdonell , virtualization@lists.linux-foundation.org, kvm@vger.kernel.org, qemu-devel@nongnu.org On 03/25/2010 06:40 PM, Michael S. Tsirkin wrote: > On Thu, Mar 25, 2010 at 06:32:15PM +0200, Avi Kivity wrote: > >> On 03/25/2010 06:23 PM, Anthony Liguori wrote: >> >>>> There has been previous discussion of virtio, however while virtio is >>>> good for exporting guest memory, it's not ideal for importing memory >>>> into a guest. >>>> >>> virtio is a DMA-based API which means that it doesn't assume cache >>> coherent shared memory. The PCI transport takes advantage of cache >>> coherent shared memory but it's not strictly required. >>> >> Aren't we violating this by not using dma_alloc_coherent() for the queues? >> > I don't see what changing this would buys us though, unless > a non-cache coherent architecture implements kvm. > We're preventing people from implementing virtio devices in hardware, not that anyone's doing that. -- error compiling committee.c: too many arguments to function