From mboxrd@z Thu Jan 1 00:00:00 1970 From: Chen Baozi Subject: Re: [PATCH] xen/arm64: Avoid sending SGI when kicking secondary cpus with spin_table Date: Sat, 9 May 2015 20:52:56 +0800 Message-ID: <20150509125255.GA11093@cbz-thinkpad> References: <1428392032-11551-1-git-send-email-cbz@baozis.org> <554CE697.1070605@citrix.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Content-Disposition: inline In-Reply-To: <554CE697.1070605@citrix.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Julien Grall Cc: julien.grall@linaro.org, ian.campbell@citrix.com, xen-devel@lists.xen.org List-Id: xen-devel@lists.xenproject.org Hi Julien, On Fri, May 08, 2015 at 05:38:47PM +0100, Julien Grall wrote: > Hi Chen, > > On 07/04/15 08:33, Chen Baozi wrote: > > From: Chen Baozi > > > > On arm64, either firmware or xen's smp_up_cpu gate uses WFE on secondary > > cpus to stand-by when booting. Thus, using SEV is enough for the boot > > cpu to kick other secondaries. Further more, the current implementation > > of cpu_up_send_sgi would pass a NULL cpumask pointer to send_SGI, which > > then lead a data fault on GICv3 send_SGI implementation. > > I've sent a patch to fix send_SGI on GICv3 [1]. You should not see the > segfault anymore. Yes, I have. And sorry for not being able to test&reply it in time. I was busy for other stuffs last few weeks. I'll try the patch as soon as possible. > > Although, you were saying that the SGI is not necessary for the spin > table. Right? Yes. I have checked the firmware, linux and xen. They all use WFE on secondary CPUs to stand-by when booting. Cheers, Baozi.