From mboxrd@z Thu Jan 1 00:00:00 1970 From: Keir Fraser Subject: Re: [PATCH] x86-64: don't use xmalloc_array() for allocation of the (per-CPU) IDTs Date: Thu, 13 Oct 2011 16:23:32 +0100 Message-ID: References: <4E971D06020000780005B352@nat28.tlf.novell.com> Mime-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <4E971D06020000780005B352@nat28.tlf.novell.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Jan Beulich Cc: "xen-devel@lists.xensource.com" List-Id: xen-devel@lists.xenproject.org On 13/10/2011 16:16, "Jan Beulich" wrote: >>> Looks good to me - I was actually considering to convert the x86-64 >>> code back to alloc_xenheap_pages() too (for we'll need to do that >>> eventually anyway when we want to support more than 5Tb of memory) >>> when I put together that earlier patch, but then refrained from doing so >>> to keep the patch size down. >> >> You mean there's a 5TB limit for alloc_domheap_pages() allocations?? > > No, I mean that currently we can't use more than 5Tb on any system. > Due to address space limitations, going beyond that will require to > introduce split domain and Xen heaps (including map_domain_page() etc) > and hence the consistent use of alloc_xenheap_pages() vs. > alloc_domheap_pages() (which currently isn't the case - you fixed just > one group of them). Remember talking about that on the summit? Ah yes, I do remember now! K.