From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andre Przywara Subject: Re: Host Numa informtion in dom0 Date: Mon, 1 Feb 2010 11:23:04 +0100 Message-ID: <4B66AB88.6090208@amd.com> References: <8EA2C2C4116BF44AB370468FBF85A7770123904A29@orsmsx504.amr.corp.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset="windows-1252"; format=flowed Content-Transfer-Encoding: quoted-printable Return-path: In-Reply-To: <8EA2C2C4116BF44AB370468FBF85A7770123904A29@orsmsx504.amr.corp.intel.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: "Kamble, Nitin A" Cc: "xen-devel@lists.xensource.com" , Keir Fraser List-Id: xen-devel@lists.xenproject.org Kamble, Nitin A wrote: > Hi Keir, >=20 > Attached is the patch which exposes the host numa information to=20 > dom0. With the patch =93xm info=94 command now also gives the cpu topol= ogy &=20 > host numa information. This will be later used to build guest numa supp= ort. What information are you missing from the current physinfo? As far as I=20 can see, only the total amount of memory per node is not provided. But=20 one could get this info from parsing the SRAT table in Dom0, which is at=20 least mapped into Dom0's memory. Or do you want to provide NUMA information to all PV guests (but then it=20 cannot be a sysctl)? This would be helpful, as this would avoid to=20 enable ACPI parsing in PV Linux for NUMA guest support. Beside that I have to oppose the introduction of sockets_per_node again.=20 Future AMD processors will feature _two_ nodes on _one_ socket, so this=20 variable should hold 1/2, but this will be rounded to zero. I think this=20 information is pretty useless anyway, as the number of sockets is mostly=20 interesting for licensing purposes, where a single number is sufficient.=20 For scheduling purposes cache topology is more important. My NUMA guest patches (currently for HVM only) are doing fine, I will=20 try to send out a RFC patches this week. I think they don't interfere=20 with this patch, but if you have other patches in development, we should=20 sync on this. The scope of my patches is to let the user (or xend) describe a guest's=20 topology (either by specifying only the number of guest nodes in the=20 config file or by explicitly describing the whole NUMA topology). Some=20 code will assign host nodes to the guest nodes (I am not sure yet=20 whether this really belongs into xend as it currently does, or is better=20 done in libxc, where libxenlight would also benefit). Then libxc's hvm_build_* will pass that info into the hvm_info_table,=20 where code in the hvmloader will generate an appropriate SRAT table. An extension of this would be to let Xen automatically decide whether a=20 split of the resources is necessary (because there is not enough memory=20 available (anymore) on one node). Looking forward to comments... Regards, Andre. --=20 Andre Przywara AMD-Operating System Research Center (OSRC), Dresden, Germany Tel: +49 351 448 3567 12 ----to satisfy European Law for business letters: Advanced Micro Devices GmbH Karl-Hammerschmidt-Str. 34, 85609 Dornach b. Muenchen Geschaeftsfuehrer: Andrew Bowd; Thomas M. McCoy; Giuliano Meroni Sitz: Dornach, Gemeinde Aschheim, Landkreis Muenchen Registergericht Muenchen, HRB Nr. 43632