Il 28/02/2014 15:52, Marcelo Tosatti ha scritto:
On Fri, Feb 28, 2014 at 12:52:53PM +0100, Andrew Jones wrote:
This patch series addresses two issues with global clock updates.
The first fixes a bug found on hosts that have a tsc marked as
unstable. As global clock updates get triggered on every vcpu load
in these cases, guests with a large number of vcpus have their
progress nearly halted. The fix for that bug should also go to
stable. The second patch in this series ensures that NTP corrections
on the host, as well as on guests with all vcpus pinned, will be
propagated periodically. That patch improves things, but doesn't
fix a bug, thus it can be merged at a later time than the first.
I've posted them together as a series, as the second one builds
on the first.
Andrew Jones (2):
x86: kvm: rate-limit global clock updates
x86: kvm: introduce periodic global clock updates
arch/x86/include/asm/kvm_host.h | 2 +
arch/x86/kvm/x86.c | 92 +++++++++++++++++++++++++++++++++++++++--
2 files changed, 90 insertions(+), 4 deletions(-)
--
1.8.1.4
Looks good.
Applied to kvm/queue, thanks to both.
Paolo
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
