From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([140.186.70.92]:55831) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1RAKRo-0004Aa-0i for qemu-devel@nongnu.org; Sun, 02 Oct 2011 07:43:08 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1RAKRm-0007Bc-LE for qemu-devel@nongnu.org; Sun, 02 Oct 2011 07:43:07 -0400 Received: from mx1.redhat.com ([209.132.183.28]:31419) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1RAKRm-0007BQ-CY for qemu-devel@nongnu.org; Sun, 02 Oct 2011 07:43:06 -0400 Date: Sun, 2 Oct 2011 13:43:56 +0200 From: "Michael S. Tsirkin" Message-ID: <20111002114356.GI30747@redhat.com> References: <1315197304-22469-1-git-send-email-david@gibson.dropbear.id.au> <1315197304-22469-2-git-send-email-david@gibson.dropbear.id.au> <20111002102547.GC30747@redhat.com> <4E883CF4.6060606@redhat.com> <20111002105238.GE30747@redhat.com> <4E8843DB.1020404@redhat.com> <20111002111700.GF30747@redhat.com> <5707D1AA-7185-4934-B2EB-91131853ABCA@suse.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <5707D1AA-7185-4934-B2EB-91131853ABCA@suse.de> Subject: Re: [Qemu-devel] [PATCH 1/9] Add stub functions for PCI device models to do PCI DMA List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Alexander Graf Cc: aliguori@us.ibm.com, kraxel@redhat.com, joerg.roedel@amd.com, qemu-devel@nongnu.org, Avi Kivity , eduard.munteanu@linux360.ro, David Gibson , rth@twiddle.net On Sun, Oct 02, 2011 at 01:28:37PM +0200, Alexander Graf wrote: > >> Good point. Though this is really a virtio specific issue since > >> other devices have explicit endianness (not guest dependent). > > > > Hmm, not entirely virtio specific, some devices use stX macros to do the > > conversion. E.g. stw_be_phys and stl_le_phys are used in several > > places. > > Yes, explicit endianness. Virtio is the only device type we support in QEMU that changes its endianness depending on the guest CPU. All other devices are independent of the guest CPU we're targeting. > > > Alex True I think, for pci devices. And virtio bypasses the iommu anyway, so we don't need to worry about it. But the point is that it makes sense to support endian-ness handling in the core. -- MST