Saurabh Mishra
2013-Dec-17 18:59 UTC
NUMA memory allocation differences between XM and XL toolchain
Hi, We are using Xen 4.2.2_06 on SLES SP3 Updates and wanted to know if there is a simple way to gather information about physical pages allocated for a HVM guest. We are trying to figure whether XL is better off in allocating contiguous huge/large pages for a guest or XM. I guess it does not matter since Xen''s hypervisor would be implementing page allocation polices. With xl debug-key u, we know how much memory was allocated from each NUMA node, but we would also like to know whether how much of them were huge pages and were they contiguous or not. Basically we need to retrieve machine pfn and VM''s pfn to do some comparison. (XEN) Memory location of each domain: (XEN) Domain 0 (total: 603765): (XEN) Node 0: 363652 (XEN) Node 1: 240113 (XEN) Domain 1 (total: 2096119): (XEN) Node 0: 1047804 (XEN) Node 1: 1048315 (XEN) Domain 2 (total: 25164798): (XEN) Node 0: 12582143 (XEN) Node 1: 12582655 We would like Xen to allocate as many as continuous huge/large pages for a HVM guest. So if there is a tunable, then please let me know. The HW is x86-64. Thanks, /Saurabh _______________________________________________ Xen-users mailing list Xen-users@lists.xen.org http://lists.xen.org/xen-users
Dario Faggioli
2013-Dec-18 10:06 UTC
Re: NUMA memory allocation differences between XM and XL toolchain
On mar, 2013-12-17 at 10:59 -0800, Saurabh Mishra wrote:> Hi, >Mmm... I replied to this same e-mail on xen-devel already. For anyone interested, the answer is here: http://bugs.xenproject.org/xen/mid/%3C1387330709.3880.52.camel@Solace%3E That being said, please, avoid cross posting and/or posting the same e-mail to both -devel and -users, this will fragment the discussion and make it a lot less useful than it could be. I appreciate that this can be seen as something in the middle between a user and developer question, but that''s not uncommon, and yet people usually post in either one ml _or_ the other. What you should do is try to guess where the message is better suited and send it only there... If, during the thread, we realize it would be better to move the discussion to the other mailing list, we''ll do that at that time. Regards, Dario -- <<This happens because I choose it to happen!>> (Raistlin Majere) ----------------------------------------------------------------- Dario Faggioli, Ph.D, http://about.me/dario.faggioli Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK) _______________________________________________ Xen-users mailing list Xen-users@lists.xen.org http://lists.xen.org/xen-users