From mboxrd@z Thu Jan 1 00:00:00 1970 From: Julien Grall Subject: Re: [RFC PATCH] xen/arm: GICv3 support for domU Date: Mon, 22 Sep 2014 13:44:17 +0100 Message-ID: <542019A1.1010009@linaro.org> References: <1408529366-21176-1-git-send-email-vijay.kilari@gmail.com> <54136F2B.8010501@linaro.org> 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: Vijay Kilari Cc: Ian Campbell , Stefano Stabellini , Prasun Kapoor , Vijaya Kumar K , Tim Deegan , "xen-devel@lists.xen.org" , Stefano Stabellini , Jan Beulich , manish.jaggi@caviumnetworks.com List-Id: xen-devel@lists.xenproject.org Hello Vijay, On 09/22/2014 12:53 PM, Vijay Kilari wrote: > On Mon, Sep 15, 2014 at 8:43 PM, Vijay Kilari wrote: >> Hi Julien, >> >> On Sat, Sep 13, 2014 at 3:39 AM, Julien Grall wrote: >>> Hi Vijay, >>> >>> On 20/08/14 03:09, vijay.kilari@gmail.com wrote: >>>> >>>> From: Vijaya Kumar K >>>> >>>> Update libxl tool for arm64 to generate GICv3 device >>>> tree node for domU. >>>> >>>> This patch is only for verifying DomU boot on hardware >>>> with GICv3 support and hardcodes GICv3 node information >>>> for DomU. Clean implementation is required using >>>> domctl >>> >>> >>> Thank you for this patch. Do you plan to send a clean implementation soon? >> >> I plan to implemented generic gicv2/gicv3 dt node >> based on your patch "xen/dts: Add hypercalls to retrieve device node >> information" >> to read gic device compatibility. > > Your patch "xen/dts: Add hypercalls to retrieve device node information" > always relies on full node patch to read device node information. > > But in some cases, like GIC we don't have full patch. Here we need > to fetch node information like compatibility and other properties like number of > re-distributor regions, re-distributor offset etc., from node name. > > So new set of hypercalls which take node name instead of path is required. I guess when you said "patch", you meant "path", right? Why do you need to get those information? You should define yourself the regions of the GICv3 in the guest layout and not relying on the host DT. In anycase, using the name of the node would be be broken as it could be duplicated and it's not standardize. Lastly, I don't plan to keep the patch "xen/dts: Add hypercalls to retrieve device node information" in my series. See the discussion on the cover letter. Regards, -- Julien Grall