According to the host node map given on the command line the VCPUs are pinned to the respective node (allowing at least scheduling between the cores belonging to this node). The mmap'ed guest memory will be bound to the correct host nodes (this will of course not take effect until the memory actually faults in). Since libnuma is not that widespread (in default installations), I chose 'enable via configure' by now: --enable-numa will compile the parts in. Signed-off-by: Andre Przywara -- Andre Przywara AMD-Operating System Research Center (OSRC), Dresden, Germany Tel: +49 351 277-84917 ----to satisfy European Law for business letters: AMD Saxony Limited Liability Company & Co. KG, Wilschdorfer Landstr. 101, 01109 Dresden, Germany Register Court Dresden: HRA 4896, General Partner authorized to represent: AMD Saxony LLC (Wilmington, Delaware, US) General Manager of AMD Saxony LLC: Dr. Hans-R. Deppe, Thomas McCoy