[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Xen-devel] small page physmap population for HVM guests vs POD

  • To: "xen-devel" <xen-devel@xxxxxxxxxxxxx>
  • From: "Jan Beulich" <JBeulich@xxxxxxxx>
  • Date: Wed, 25 Jul 2012 14:11:04 +0100
  • Delivery-date: Wed, 25 Jul 2012 13:11:34 +0000
  • List-id: Xen developer discussion <xen-devel.lists.xen.org>

Is there a reason why for 1G and 2M requests setup_guest() does
pass XENMEMF_populate_on_demand, yet in the event of falling
back to 4k allocations it doesn't? A POD guest that can't get large
pages allocated seems penalized enough already, so I don't see
why its startup gets even further slowed down by requiring the
pages to be reclaimed (if that's possible at all, considering that
the pages don't get cleared explicitly) rather than added to the
POD cache from the beginning.

Thanks, Jan

Xen-devel mailing list



Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.