From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([140.186.70.92]:47368) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1RDcDG-0000YR-GX for qemu-devel@nongnu.org; Tue, 11 Oct 2011 09:17:43 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1RDcDE-0004Ru-P3 for qemu-devel@nongnu.org; Tue, 11 Oct 2011 09:17:42 -0400 Received: from mx1.redhat.com ([209.132.183.28]:11725) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1RDcDE-0004RQ-HR for qemu-devel@nongnu.org; Tue, 11 Oct 2011 09:17:40 -0400 Date: Tue, 11 Oct 2011 15:17:36 +0200 From: Gleb Natapov Message-ID: <20111011131736.GY30105@redhat.com> References: <20111010170803.GV9408@redhat.com> <4E933F2D.7090703@codemonkey.ws> <20111011082315.GI14627@redhat.com> <4E940919.7010901@redhat.com> <5C80782F-C30A-4F35-93FD-0397A1040AFF@suse.de> <4E940BB6.2000400@redhat.com> <050FFBD4-BF45-4425-865B-24E7C228B592@suse.de> <4E9440B6.3060201@codemonkey.ws> <2DB3FD25-4BF7-4F82-9A62-49A6891316B3@suse.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <2DB3FD25-4BF7-4F82-9A62-49A6891316B3@suse.de> Subject: Re: [Qemu-devel] Slow kernel/initrd loading via fw_cfg; Was Re: Hack integrating SeaBios / LinuxBoot option rom with QEMU trace backends List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Alexander Graf Cc: "Richard W.M. Jones" , Avi Kivity , qemu-devel On Tue, Oct 11, 2011 at 03:14:52PM +0200, Alexander Graf wrote: > > rep/ins is effectively equivalent to DMA except in how it's handled within QEMU. > > No, DMA has a lot bigger granularities in kvm/user interaction. We can easily DMA a 50MB region with a single kvm/user exit. For PIO we can at most do page granularity. > Not only granularity, but double copy too. May be Anthony is referring to real HW in which case I also do not see how it can be true since one operation is synchronous and another is not. -- Gleb.