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

[Xen-devel] [PATCH 00/11] x86: support up to 16Tb

  • To: "xen-devel" <xen-devel@xxxxxxxxxxxxx>
  • From: "Jan Beulich" <JBeulich@xxxxxxxx>
  • Date: Tue, 22 Jan 2013 10:45:57 +0000
  • Delivery-date: Tue, 22 Jan 2013 10:46:22 +0000
  • List-id: Xen developer discussion <xen-devel.lists.xen.org>

This series enables Xen to support up to 16Tb.

01: x86: introduce virt_to_xen_l1e()
02: x86: extend frame table virtual space
03: x86: re-introduce map_domain_page() et al
04: x86: properly use map_domain_page() when building Dom0
05: x86: consolidate initialization of PV guest L4 page tables
06: x86: properly use map_domain_page() during domain creation/destruction
07: x86: properly use map_domain_page() during page table manipulation
08: x86: properly use map_domain_page() in nested HVM code
09: x86: properly use map_domain_page() in miscellaneous places
10: tmem: partial adjustments for x86 16Tb support
11: x86: support up to 16Tb

As I don't have a 16Tb system around, I used the following
debugging patch to simulate the most critical aspect the changes
above would have on a system with this much memory: Not all of
the 1:1 mapping being accessible when in PV guest context. To do
so, a command line option to pull the split point down is being
added. The patch is being provided in the raw form I used it, but
has pieces properly formatted and not marked "//temp" which I
would think might be worth considering to add. The other pieces
are likely less worthwhile, but if others think differently I could
certainly also put them into "normal" shape.

12: x86: debugging code for testing 16Tb support on smaller memory systems

Signed-off-by: Jan Beulich <jbeulich@xxxxxxxx>

Xen-devel mailing list



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