From mboxrd@z Thu Jan 1 00:00:00 1970 From: Julien Grall Subject: Re: [PATCH v2 6/8] tools: arm: refactor code to setup guest p2m and fill it with RAM Date: Fri, 25 Apr 2014 14:28:45 +0100 Message-ID: <535A630D.8010809@linaro.org> References: <1398424945.18537.424.camel@kazak.uk.xensource.com> <1398424967-9306-6-git-send-email-ian.campbell@citrix.com> <535A5A61.1070706@linaro.org> <1398430786.18537.459.camel@kazak.uk.xensource.com> <535A5F4F.3030102@linaro.org> <1398432122.18537.464.camel@kazak.uk.xensource.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; Format="flowed" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1398432122.18537.464.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: ian.jackson@eu.citrix.com, stefano.stabellini@eu.citrix.com, tim@xen.org, xen-devel@lists.xen.org List-Id: xen-devel@lists.xenproject.org On 25/04/14 14:22, Ian Campbell wrote: > On Fri, 2014-04-25 at 14:12 +0100, Julien Grall wrote: >> >> On 25/04/14 13:59, Ian Campbell wrote: >> >> It think we can avoid looping twice by making the two banks contiguous >>>> in the memory (i.e starting the second bank at 4GB instead of 8GB). >>> >>> As explained in the commit message I have deliberately left a hole so >>> that we can see that such configurations actually work. >> >> IHMO, the code path doesn't seem very complicate. Adding this overhead >> (the two loop + the 1GB hole) just for it seems pointless. > > I disagree. It is always worth exercising these things, and this will > make sure we don't bake in assumptions about using a single contiguous > bank anywhere by mistake. > > I think you are overestimating the overhead and underestimating the > benefit.0 > >> BTW, is there any issue to create one big bank rather than 2? > > I expect it would work just fine, but I am not going to do that either > for the reasons already discussed. I'm not entirely convince of this hole... but it might help to also avoid assumption during guest migration: Acked-by: Julien Grall -- Julien Grall