mirror of
https://github.com/xen-project/xen.git
synced 2026-10-28 10:11:27 +01:00
Frediano Ziglio
50baf2d9f7
Right now on migrate/resume, all guest memory is allocated with 4k small pages. This was an intended oversight in the original Migration v2 implementation. With some test memory program stressing TLB (many small random memory accesses) superpages can show a 15% performance improvement. On the first memory iteration the sender is currently sending memory in 4MB aligned chunks of GFN space which allows the receiver to allocate most pages as 2MB superpages instead of single 4KB pages. This change does not handle 1GB superpages as this will require changes to the migration protocol. Signed-off-by: Frediano Ziglio <frediano.ziglio@cloud.com> Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com> Release-Acked-by: Oleksii Kurochko <oleksii.kurochko@gmail.com>