From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dave McCracken Subject: Re: [Linux PATCH] Mark hugepages to the hypervisor to speed up mapping (with signoff this time) Date: Wed, 9 Jun 2010 13:20:33 -0500 Message-ID: <201006091320.34136.dcm@mccr.org> References: <20100609172016.19083.9014.sendpatchset@magnum.int.mccr.org> <4C0FD790.3070603@goop.org> Mime-Version: 1.0 Content-Type: Text/Plain; charset="iso-8859-15" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <4C0FD790.3070603@goop.org> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Jeremy Fitzhardinge Cc: Xen Developers List List-Id: xen-devel@lists.xenproject.org On Wednesday, June 09, 2010, Jeremy Fitzhardinge wrote: > > + return 1; > > + } > > + for (i = 0, m = mfn; i < HPAGE_SIZE/PAGE_SIZE; i++, pfn++, m++) { > > + if (pfn_to_mfn(pfn) != m) { > > + printk("Guest pages are not properly aligned to use > > hugepages\n"); > > > > > > Isn't this a superset of the previous test? The code is doublechecking to make sure this hugepage is actually contiguous at the machine level. Dave McCracken Oracle Corp.