From mboxrd@z Thu Jan 1 00:00:00 1970 From: Denis Plotnikov Subject: [PATCH v2 10/11] pvclock: add clocksource change notification on changing of tsc stable bit Date: Fri, 21 Jul 2017 18:45:17 +0300 Message-ID: <1500651918-14156-11-git-send-email-dplotnikov@virtuozzo.com> References: <1500651918-14156-1-git-send-email-dplotnikov@virtuozzo.com> Cc: pbonzini@redhat.com, den@virtuozzo.com, rkagan@virtuozzo.com To: kvm@vger.kernel.org, rkrcmar@redhat.com Return-path: Received: from mailhub.sw.ru ([195.214.232.25]:15007 "EHLO relay.sw.ru" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754488AbdGUPpf (ORCPT ); Fri, 21 Jul 2017 11:45:35 -0400 In-Reply-To: <1500651918-14156-1-git-send-email-dplotnikov@virtuozzo.com> Sender: kvm-owner@vger.kernel.org List-ID: We going to allow L2 guest to use L1's kvm-clock clocksource, which is a pvclock clocksource, as a stable one if it's stable in L1. Therefore, we need to know when L1's kvm-clock becomes stable/unstable to make it stable/unstable in L2 as well. Do it, by adding tracking stability flag in pvclock Signed-off-by: Denis Plotnikov --- arch/x86/kernel/pvclock.c | 19 ++++++++++++++++--- 1 file changed, 16 insertions(+), 3 deletions(-) diff --git a/arch/x86/kernel/pvclock.c b/arch/x86/kernel/pvclock.c index ab54c92..c73e5a5 100644 --- a/arch/x86/kernel/pvclock.c +++ b/arch/x86/kernel/pvclock.c @@ -22,6 +22,7 @@ #include #include #include +#include #include #include @@ -73,6 +74,8 @@ u8 pvclock_read_flags(struct pvclock_vcpu_time_info *src) return flags & valid_flags; } +static atomic64_t clocksource_stable = ATOMIC64_INIT(0); + u64 pvclock_clocksource_read(struct pvclock_vcpu_time_info *src, u64 *cycles_stamp) { @@ -96,10 +99,20 @@ u64 pvclock_clocksource_read(struct pvclock_vcpu_time_info *src, pvclock_touch_watchdogs(); } - if ((valid_flags & PVCLOCK_TSC_STABLE_BIT) && - (flags & PVCLOCK_TSC_STABLE_BIT)) - return ret; + if (likely(valid_flags & PVCLOCK_TSC_STABLE_BIT)) { + bool stable_now = !!(flags & PVCLOCK_TSC_STABLE_BIT); + bool stable_last = (bool) atomic64_read(&clocksource_stable); + + if (unlikely(stable_now != stable_last)) { + /* send notification once */ + if (stable_last == atomic64_cmpxchg( + &clocksource_stable, stable_last, stable_now)) + clocksource_changes_notify(); + } + if (stable_now) + return ret; + } /* * Assumption here is that last_value, a global accumulator, always goes * forward. If we are less than that, we should not be much smaller. -- 2.7.4