From mboxrd@z Thu Jan  1 00:00:00 1970
From: Roland Dreier <rdreier@cisco.com>
Subject: [ofa-general] Re: mthca use of dma_sync_single is bogus
Date: Tue, 10 Jul 2007 12:08:43 -0700
Message-ID: <adak5t89iv8.fsf@cisco.com>
References: <adalkdpxopo.fsf@cisco.com>
	<20070709213913.GB20052@mellanox.co.il> <adamyy4vjo9.fsf@cisco.com>
	<20070710141409.GH3885@ics.muni.cz> <ada7ip8b0be.fsf@cisco.com>
	<20070710190018.GK3885@ics.muni.cz>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Return-path: <general-bounces@lists.openfabrics.org>
In-Reply-To: <20070710190018.GK3885@ics.muni.cz> (Lukas Hejtmanek's message of
	"Tue, 10 Jul 2007 21:00:19 +0200")
List-Unsubscribe: <http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general>,
	<mailto:general-request@lists.openfabrics.org?subject=unsubscribe>
List-Archive: <http://lists.openfabrics.org/pipermail/general>
List-Post: <mailto:general@lists.openfabrics.org>
List-Help: <mailto:general-request@lists.openfabrics.org?subject=help>
List-Subscribe: <http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general>,
	<mailto:general-request@lists.openfabrics.org?subject=subscribe>
Sender: general-bounces@lists.openfabrics.org
Errors-To: general-bounces@lists.openfabrics.org
To: Lukas Hejtmanek <xhejtman@ics.muni.cz>
Cc: Keir Fraser <keir@xensource.com>, xen-devel@lists.xensource.com, "Michael S. Tsirkin" <mst@dev.mellanox.co.il>, general@lists.openfabrics.org
List-Id: xen-devel@lists.xenproject.org

 > I think he spoke about coherent, didn't he? On i386/x86_64, the consistent and
 > coherent are the same but on some architectures they are not and I think that
 > using consistent (in particular pci_alloc_consistent) is exactly what should
 > be used. Keir also recommended to use this one.

coherent and consistent are synonyms.  It's confusing because there is
pci_alloc_consistent(), which is in general just a wrapper for
dma_alloc_coherent().

 > And moreover, it avoids using swiotlb and bounce buffers, I think. Am I right,
 > Keir?

Yes, but I'm not really willing to make things worse for standard i386
just to make Xen work a little better.

 - R.