From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ian Campbell Subject: Re: [PATCH OSSTEST v5 00/13] support for ARM32 arndale and cubietruck platforms Date: Fri, 23 Jan 2015 12:57:40 +0000 Message-ID: <1422017860.19859.69.camel@citrix.com> References: <1421924192.23842.9.camel@citrix.com> <1421941958.23842.52.camel@citrix.com> <1422006409.19859.8.camel@citrix.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Stefano Stabellini Cc: Ian Jackson , xen-devel List-Id: xen-devel@lists.xenproject.org On Fri, 2015-01-23 at 12:03 +0000, Stefano Stabellini wrote: > You are right, I was mistaken. However keep in mind that only the > ubuntu stable branch linux-3.16.y-queue has the following important > backport: > > commit 342443e998bd2dd6406d46c5f7e62540a6d4a170 > Author: Stefano Stabellini > Date: Fri Nov 21 11:07:39 2014 +0000 > > xen/arm/arm64: introduce xen_arch_need_swiotlb > > commit a4dba130891271084344c12537731542ec77cb85 upstream. > > Introduce an arch specific function to find out whether a particular dma > mapping operation needs to bounce on the swiotlb buffer. > > the main linux-3.16.y branch (currently tagged v3.16.7-ckt4) doesn't have it yet. > > Without it, DMA is not safe with non-coherent devices. Hrm, how unfortunate. We expect that patch to land in ckt5 I suppose? I think we can either wait for ckt5 or you could apply it to your branch, which means future versions would need merging not fast-forwarding. Not horrendous I suppose.