[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-Id: <1474285153-4216-1-git-send-email-pbonzini@redhat.com>
Date: Mon, 19 Sep 2016 13:39:09 +0200
From: Paolo Bonzini <pbonzini@...hat.com>
To: linux-kernel@...r.kernel.org, kvm@...r.kernel.org
Cc: rkrcmar@...hat.com, den@...nvz.org, rkagan@...tuozzo.com,
peterhornyack@...gle.com
Subject: [PATCH v2 0/4] KVM: x86: kvmclock cleanups for Hyper-V TSC page
This implements the plan that we discussed at KVM Forum. Patch 3
implements a new function that returns the current value of kvmclock as
read by the guest. The function is then used to provide the Hyper-V
time reference counter. Now that the reference counter is synchronized
with kvmclock, the TSC page can be supported nicely.
Paolo
v2->v3: use TSC page information to compute time reference counter value
fixed case where TSC seq is 0xFFFFFFFE
Paolo Bonzini (4):
KVM: x86: always fill in vcpu->arch.hv_clock
KVM: x86: initialize kvmclock_offset
KVM: x86: introduce get_kvmclock_ns
KVM: x86: Hyper-V tsc page setup
arch/x86/entry/vdso/vclock_gettime.c | 2 +-
arch/x86/include/asm/kvm_host.h | 2 +
arch/x86/include/asm/pvclock.h | 5 +-
arch/x86/kernel/pvclock.c | 2 +-
arch/x86/kvm/hyperv.c | 162 +++++++++++++++++++++++++++++++----
arch/x86/kvm/hyperv.h | 3 +
arch/x86/kvm/x86.c | 158 ++++++++++++++++++++--------------
arch/x86/kvm/x86.h | 6 +-
8 files changed, 253 insertions(+), 87 deletions(-)
--
1.8.3.1
Powered by blists - more mailing lists