From mboxrd@z Thu Jan 1 00:00:00 1970 From: Julien Grall Subject: Re: [PATCH v2 00/41] Add ACPI support for arm64 on Xen Date: Mon, 18 May 2015 13:46:04 +0100 Message-ID: <5559DF0C.5090904@citrix.com> References: <1431893048-5214-1-git-send-email-parth.dixit@linaro.org> <555903E9.6030005@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: Parth Dixit , Ian Campbell , Stefano Stabellini , Tim Deegan Cc: xen-devel List-Id: xen-devel@lists.xenproject.org Hi Parth (You dropped xen-devel, re-cc it) On 18/05/15 10:59, Parth Dixit wrote: > yes i tested with linux-next of 15th may and it is working fine with > it,except mounting of root partiton because kernel is not able to > detetct the partition > i did not looked further into it but its presumably configuration of > right driver. > Linux mainline will need 3 additional patches > 1. Adding STAO and XENV > 2. Hiding of UART > 3. Enabling xen compilation with ACPI which was disabled by Hanjun. > I'll wait for comment's on this series and if there are no major changes > i'll cleanup linux patches and send them for review as well. I think you can send an RFC now. The STA0 and XENV table are well-defined and the rest of the patches doesn't seem to be tight with your Xen series. > > > Few workarounds have been made to get it working, these are as > follows > 1. In Xen interrupts are routed at the boot time with > edge/trigger level set to 0 > because this information is not available at the time of booting. > > > edge/trigger is only a bit. Can't you just avoid to set it rather > than using a potentially incorrect value? > > That would require disabling the checks for edge/trigger at multiple > places for acpi, if you are ok with the approach > i'll go ahead and send the patch. IIRC, during the last Linaro connect we agreed to differ edge/trigger setting for IRQ assigned to a Domain for both ACPI and DT. > > 2. EFI runtime services are disaled in linux but proper solution > has to come from linux side. > > > Can you details a bit more? What is missing? > > We only create EFI stub for passing RSDP and memory data in xen, we are > not implementing full fledged runtime service support > for EFI. At present linux expects that if EFI interface is present > runtime services are also available. Thanks for the explanation. Please add it in the next cover letter. It's useful for people who weren't present at the meeting or forgot what was said. Regards, -- Julien Grall