From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeremy Fitzhardinge Subject: Re: Re: [PATCH 2/2] Linux support for hugepages as a Xen PV guest Date: Wed, 15 Oct 2008 14:48:22 -0700 Message-ID: <48F66526.9050807@goop.org> References: <200810101006.44389.dcm@mccr.org> <200810151201.06367.dcm@mccr.org> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <200810151201.06367.dcm@mccr.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: Dave McCracken Cc: xen-devel@lists.xensource.com, Keir Fraser List-Id: xen-devel@lists.xenproject.org Dave McCracken wrote: > For starters, Xen currently does remove PSE from the CPUID it returns to the > guest. I'll add some code to allow PSE when hugepages are enabled. > Yep. > Second, in Linux for the x86_64 architecture the macro to check for PSE is > hard-wired to always be true. > Not in current kernels. > Third, hugepages in Linux makes no check whatsoever for the presence or > absence of PSE. It assumes it's there. > Not in current kernels. But it does assume that if PSE is present, then physical memory can always be mapped with large pages. > I'll look into the simplest way to make the various Linux parts obey the PSE > flag properly. > No, use a current mainline git linux kernel. J