From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751450AbdAYMax (ORCPT ); Wed, 25 Jan 2017 07:30:53 -0500 Received: from mail-wm0-f68.google.com ([74.125.82.68]:35990 "EHLO mail-wm0-f68.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751346AbdAYMav (ORCPT ); Wed, 25 Jan 2017 07:30:51 -0500 Subject: Re: [patch 1/4] KVM: x86: provide realtime host clock via vsyscall notifiers To: Marcelo Tosatti , kvm@vger.kernel.org, linux-kernel@vger.kernel.org References: <20170124170938.909652059@redhat.com> <20170124171020.310170535@redhat.com> Cc: Radim Krcmar , Richard Cochran , Miroslav Lichvar From: Paolo Bonzini Message-ID: <2aa4a341-b6e0-e82d-66d8-f596c123fe12@redhat.com> Date: Wed, 25 Jan 2017 13:30:47 +0100 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101 Thunderbird/45.5.1 MIME-Version: 1.0 In-Reply-To: <20170124171020.310170535@redhat.com> Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 24/01/2017 18:09, Marcelo Tosatti wrote: > Expose the realtime host clock and save the TSC value > used for the clock calculation. > > Signed-off-by: Marcelo Tosatti > > --- > arch/x86/kvm/x86.c | 36 ++++++++++++++++++++++++++++++++++++ > 1 file changed, 36 insertions(+) > > v2: unify nsec_base (Radim) > > Index: kvm-ptpdriver/arch/x86/kvm/x86.c > =================================================================== > --- kvm-ptpdriver.orig/arch/x86/kvm/x86.c 2017-01-13 16:43:15.013244682 -0200 > +++ kvm-ptpdriver/arch/x86/kvm/x86.c 2017-01-20 09:56:41.404534904 -0200 > @@ -1139,6 +1139,7 @@ > > u64 boot_ns; > u64 nsec_base; > + u64 wall_time_sec; > }; > > static struct pvclock_gtod_data pvclock_gtod_data; > @@ -1162,6 +1163,8 @@ > vdata->boot_ns = boot_ns; > vdata->nsec_base = tk->tkr_mono.xtime_nsec; > > + vdata->wall_time_sec = tk->xtime_sec; > + > write_seqcount_end(&vdata->seq); > } > #endif > @@ -1623,6 +1626,28 @@ > return mode; > } > > +static int do_realtime(struct timespec *ts, cycle_t *cycle_now) > +{ > + struct pvclock_gtod_data *gtod = &pvclock_gtod_data; > + unsigned long seq; > + int mode; > + u64 ns; > + > + do { > + seq = read_seqcount_begin(>od->seq); > + mode = gtod->clock.vclock_mode; > + ts->tv_sec = gtod->wall_time_sec; > + ns = gtod->nsec_base; > + ns += vgettsc(cycle_now); > + ns >>= gtod->clock.shift; > + } while (unlikely(read_seqcount_retry(>od->seq, seq))); > + > + ts->tv_sec += __iter_div_u64_rem(ns, NSEC_PER_SEC, &ns); > + ts->tv_nsec = ns; > + > + return mode; > +} > + > /* returns true if host is using tsc clocksource */ > static bool kvm_get_time_and_clockread(s64 *kernel_ns, cycle_t *cycle_now) > { > @@ -1632,6 +1657,17 @@ > > return do_monotonic_boot(kernel_ns, cycle_now) == VCLOCK_TSC; > } > + > +/* returns true if host is using tsc clocksource */ > +static bool kvm_get_walltime_and_clockread(struct timespec *ts, > + cycle_t *cycle_now) > +{ > + /* checked again under seqlock below */ > + if (pvclock_gtod_data.clock.vclock_mode != VCLOCK_TSC) > + return false; > + > + return do_realtime(ts, cycle_now) == VCLOCK_TSC; > +} > #endif > > /* > > > Only used in the next patch, will squash. Also, cycle_t is gone now so use u64 instead. Paolo