From mboxrd@z Thu Jan 1 00:00:00 1970 From: Julien Grall Subject: Re: [PATCH] xen: arm: flush TLB on all CPUs when setting or clearing fixmaps Date: Wed, 15 Jan 2014 14:58:09 +0000 Message-ID: <52D6A201.6030708@linaro.org> References: <1389718513-1638-1-git-send-email-ian.campbell@citrix.com> <52D5880B.30506@linaro.org> <1389778658.12434.120.camel@kazak.uk.xensource.com> <52D6921F.1030307@linaro.org> <1389794748.3793.48.camel@kazak.uk.xensource.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1389794748.3793.48.camel@kazak.uk.xensource.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: Ian Campbell Cc: George Dunlap , stefano.stabellini@eu.citrix.com, tim@xen.org, xen-devel@lists.xen.org List-Id: xen-devel@lists.xenproject.org On 01/15/2014 02:05 PM, Ian Campbell wrote: > On Wed, 2014-01-15 at 13:50 +0000, Julien Grall wrote: >> On 01/15/2014 09:37 AM, Ian Campbell wrote: >>> On Tue, 2014-01-14 at 18:55 +0000, Julien Grall wrote: >>>> On 01/14/2014 04:55 PM, Ian Campbell wrote: >>>>> These mappings are global and therefore need flushing on all processors. Add >>>>> flush_all_xen_data_tlb_range_va which accomplishes this. >>>> >>>> Can we make name consistent across every *tlb* function call? On >>>> flushtlb.h we use *_local for maintenance on the current processor only. >>>> If the suffix is not present then the maintenance will be done on every >>>> processor. >>> >>> I was trying to avoid a massive renaming of the existing flush_xen_*. I >>> suppose I should just go ahead and do it. >> >> If it's too big for 4.4, > > With my temporary-RM hat on I've struggled with this a few times this > week -- that is, larger, mostly mechanical, textual changes which come > about because it is the correct/cleanest thing to do as part of a > smaller change which on their own would be pretty clear candidates for > an exception. Chen's change "xen/arm{32, 64}: fix section shift when > mapping 2MB block in boot page table" is in a similar boat. > > I'm not sure where the balance should lie really. The "issue" I see is backporting patch from Xen 4.5 to Xen 4.4 will be less trivial. We will have to think about the function name. -- Julien Grall