* [patch 0/3] KVM virtual PTP driver
@ 2017-01-13 12:01 Marcelo Tosatti
2017-01-13 12:01 ` [patch 1/3] KVM: x86: provide realtime host clock via vsyscall notifiers Marcelo Tosatti
` (2 more replies)
0 siblings, 3 replies; 54+ messages in thread
From: Marcelo Tosatti @ 2017-01-13 12:01 UTC (permalink / raw)
To: kvm, linux-kernel
Cc: Paolo Bonzini, Radim Krcmar, Richard Cochran, Miroslav Lichvar
This patchset implements a virtual PTP driver which allows guest to sync
its clock to the host clock with high precision ( error is < 10ns).
^ permalink raw reply [flat|nested] 54+ messages in thread* [patch 1/3] KVM: x86: provide realtime host clock via vsyscall notifiers 2017-01-13 12:01 [patch 0/3] KVM virtual PTP driver Marcelo Tosatti @ 2017-01-13 12:01 ` Marcelo Tosatti 2017-01-13 15:18 ` Radim Krcmar 2017-01-13 12:01 ` [patch 2/3] KVM: x86: add KVM_HC_CLOCK_OFFSET hypercall Marcelo Tosatti 2017-01-13 12:01 ` [patch 3/3] PTP: add kvm PTP driver Marcelo Tosatti 2 siblings, 1 reply; 54+ messages in thread From: Marcelo Tosatti @ 2017-01-13 12:01 UTC (permalink / raw) To: kvm, linux-kernel Cc: Paolo Bonzini, Radim Krcmar, Richard Cochran, Miroslav Lichvar, Marcelo Tosatti [-- Attachment #1: kvm-do-realtime --] [-- Type: text/plain, Size: 2074 bytes --] Expose the realtime host clock and save the TSC value used for the clock calculation. Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com> --- arch/x86/kvm/x86.c | 38 ++++++++++++++++++++++++++++++++++++++ 1 file changed, 38 insertions(+) Index: kvm-ptpdriver/arch/x86/kvm/x86.c =================================================================== --- kvm-ptpdriver.orig/arch/x86/kvm/x86.c 2017-01-13 08:59:03.015895353 -0200 +++ kvm-ptpdriver/arch/x86/kvm/x86.c 2017-01-13 09:04:46.581415259 -0200 @@ -1139,6 +1139,8 @@ u64 boot_ns; u64 nsec_base; + u64 wall_time_sec; + u64 wall_time_snsec; }; static struct pvclock_gtod_data pvclock_gtod_data; @@ -1162,6 +1164,9 @@ vdata->boot_ns = boot_ns; vdata->nsec_base = tk->tkr_mono.xtime_nsec; + vdata->wall_time_sec = tk->xtime_sec; + vdata->wall_time_snsec = tk->tkr_mono.xtime_nsec; + write_seqcount_end(&vdata->seq); } #endif @@ -1623,6 +1628,28 @@ 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->wall_time_snsec; + 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 +1659,17 @@ 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 /* ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 1/3] KVM: x86: provide realtime host clock via vsyscall notifiers 2017-01-13 12:01 ` [patch 1/3] KVM: x86: provide realtime host clock via vsyscall notifiers Marcelo Tosatti @ 2017-01-13 15:18 ` Radim Krcmar 2017-01-13 15:34 ` Marcelo Tosatti 2017-01-13 15:41 ` Konrad Rzeszutek Wilk 0 siblings, 2 replies; 54+ messages in thread From: Radim Krcmar @ 2017-01-13 15:18 UTC (permalink / raw) To: Marcelo Tosatti Cc: kvm, linux-kernel, Paolo Bonzini, Richard Cochran, Miroslav Lichvar 2017-01-13 10:01-0200, Marcelo Tosatti: > Expose the realtime host clock and save the TSC value > used for the clock calculation. > > Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com> > > --- > arch/x86/kvm/x86.c | 38 ++++++++++++++++++++++++++++++++++++++ > 1 file changed, 38 insertions(+) > > Index: kvm-ptpdriver/arch/x86/kvm/x86.c > =================================================================== > --- kvm-ptpdriver.orig/arch/x86/kvm/x86.c 2017-01-13 08:59:03.015895353 -0200 > +++ kvm-ptpdriver/arch/x86/kvm/x86.c 2017-01-13 09:04:46.581415259 -0200 > @@ -1139,6 +1139,8 @@ > > u64 boot_ns; > u64 nsec_base; > + u64 wall_time_sec; > + u64 wall_time_snsec; The leading "s" in "snsec" looks like a copy-paste residue. > }; > > static struct pvclock_gtod_data pvclock_gtod_data; > @@ -1162,6 +1164,9 @@ > vdata->boot_ns = boot_ns; > vdata->nsec_base = tk->tkr_mono.xtime_nsec; > > + vdata->wall_time_sec = tk->xtime_sec; > + vdata->wall_time_snsec = tk->tkr_mono.xtime_nsec; Using tk->tkr_mono offsets for real time seems wrong -- what happens if the real time is half a second shifted from monotonic time? If it's ok, then vdata->nsec_base == vdata->wall_time_snsec, so we don't need it. > + > write_seqcount_end(&vdata->seq); > } > #endif > @@ -1623,6 +1628,28 @@ > return mode; > } > > +static int do_realtime(struct timespec *ts, cycle_t *cycle_now) This is too similar to do_monotonic_boot(), but I don't see a solution that is both nice and efficient. :( (It usually means macros or copying pvclock_gtod_data.) > +{ > + 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->wall_time_snsec; > + 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 +1659,17 @@ > > 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 > > /* > > ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 1/3] KVM: x86: provide realtime host clock via vsyscall notifiers 2017-01-13 15:18 ` Radim Krcmar @ 2017-01-13 15:34 ` Marcelo Tosatti 2017-01-13 16:28 ` Radim Krcmar 2017-01-13 15:41 ` Konrad Rzeszutek Wilk 1 sibling, 1 reply; 54+ messages in thread From: Marcelo Tosatti @ 2017-01-13 15:34 UTC (permalink / raw) To: Radim Krcmar Cc: kvm, linux-kernel, Paolo Bonzini, Richard Cochran, Miroslav Lichvar On Fri, Jan 13, 2017 at 04:18:04PM +0100, Radim Krcmar wrote: > 2017-01-13 10:01-0200, Marcelo Tosatti: > > Expose the realtime host clock and save the TSC value > > used for the clock calculation. > > > > Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com> > > > > --- > > arch/x86/kvm/x86.c | 38 ++++++++++++++++++++++++++++++++++++++ > > 1 file changed, 38 insertions(+) > > > > Index: kvm-ptpdriver/arch/x86/kvm/x86.c > > =================================================================== > > --- kvm-ptpdriver.orig/arch/x86/kvm/x86.c 2017-01-13 08:59:03.015895353 -0200 > > +++ kvm-ptpdriver/arch/x86/kvm/x86.c 2017-01-13 09:04:46.581415259 -0200 > > @@ -1139,6 +1139,8 @@ > > > > u64 boot_ns; > > u64 nsec_base; > > + u64 wall_time_sec; > > + u64 wall_time_snsec; > > The leading "s" in "snsec" looks like a copy-paste residue. Just copying the userspace vsyscall interface. > > }; > > > > static struct pvclock_gtod_data pvclock_gtod_data; > > @@ -1162,6 +1164,9 @@ > > vdata->boot_ns = boot_ns; > > vdata->nsec_base = tk->tkr_mono.xtime_nsec; > > > > + vdata->wall_time_sec = tk->xtime_sec; > > + vdata->wall_time_snsec = tk->tkr_mono.xtime_nsec; > > Using tk->tkr_mono offsets for real time seems wrong -- what happens if > the real time is half a second shifted from monotonic time? Both the userspace vsyscall interface and getnstimeofday use it for realtime clock. Monotonic clock adds the offset: vdata->monotonic_time_snsec = tk->tkr_mono.xtime_nsec + ((u64)tk->wall_to_monotonic.tv_nsec << tk->tkr_mono.shift); > If it's ok, then vdata->nsec_base == vdata->wall_time_snsec, so we don't > need it. Just copying the userspace vsyscall interface. Do you actually want to change the "s" and unify wall_time_snsec with nsec_base? > > + > > write_seqcount_end(&vdata->seq); > > } > > #endif > > @@ -1623,6 +1628,28 @@ > > return mode; > > } > > > > +static int do_realtime(struct timespec *ts, cycle_t *cycle_now) > > This is too similar to do_monotonic_boot(), but I don't see a solution > that is both nice and efficient. :( > > (It usually means macros or copying pvclock_gtod_data.) Yep. ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 1/3] KVM: x86: provide realtime host clock via vsyscall notifiers 2017-01-13 15:34 ` Marcelo Tosatti @ 2017-01-13 16:28 ` Radim Krcmar 2017-01-13 17:51 ` Marcelo Tosatti 0 siblings, 1 reply; 54+ messages in thread From: Radim Krcmar @ 2017-01-13 16:28 UTC (permalink / raw) To: Marcelo Tosatti Cc: kvm, linux-kernel, Paolo Bonzini, Richard Cochran, Miroslav Lichvar 2017-01-13 13:34-0200, Marcelo Tosatti: > On Fri, Jan 13, 2017 at 04:18:04PM +0100, Radim Krcmar wrote: >> 2017-01-13 10:01-0200, Marcelo Tosatti: >> > Expose the realtime host clock and save the TSC value >> > used for the clock calculation. >> > >> > Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com> >> > >> > --- >> > arch/x86/kvm/x86.c | 38 ++++++++++++++++++++++++++++++++++++++ >> > 1 file changed, 38 insertions(+) >> > >> > Index: kvm-ptpdriver/arch/x86/kvm/x86.c >> > =================================================================== >> > --- kvm-ptpdriver.orig/arch/x86/kvm/x86.c 2017-01-13 08:59:03.015895353 -0200 >> > +++ kvm-ptpdriver/arch/x86/kvm/x86.c 2017-01-13 09:04:46.581415259 -0200 >> > @@ -1139,6 +1139,8 @@ >> > >> > u64 boot_ns; >> > u64 nsec_base; >> > + u64 wall_time_sec; >> > + u64 wall_time_snsec; >> >> The leading "s" in "snsec" looks like a copy-paste residue. > > Just copying the userspace vsyscall interface. Oh, so the "s" means "sub-" for sub-nanosecond precision. >> > }; >> > >> > static struct pvclock_gtod_data pvclock_gtod_data; >> > @@ -1162,6 +1164,9 @@ >> > vdata->boot_ns = boot_ns; >> > vdata->nsec_base = tk->tkr_mono.xtime_nsec; >> > >> > + vdata->wall_time_sec = tk->xtime_sec; >> > + vdata->wall_time_snsec = tk->tkr_mono.xtime_nsec; >> >> Using tk->tkr_mono offsets for real time seems wrong -- what happens if >> the real time is half a second shifted from monotonic time? > > Both the userspace vsyscall interface and getnstimeofday > use it for realtime clock. > > Monotonic clock adds the offset: > > vdata->monotonic_time_snsec = tk->tkr_mono.xtime_nsec > + > ((u64)tk->wall_to_monotonic.tv_nsec > << tk->tkr_mono.shift); I see, thanks. Makes me wonder why our monotonic time is correct then, but that is problably thanks to boot_ns. >> If it's ok, then vdata->nsec_base == vdata->wall_time_snsec, so we don't >> need it. > > Just copying the userspace vsyscall interface. > > Do you actually want to change the "s" and unify wall_time_snsec with > nsec_base? The "s" isn't important, even though I don't think we do anything that would justify it, but make use just 8 bytes for both. Renaming nsec_base is ok, but I'm not sure what tk->tkr_mono.xtime_nsec is anymore. ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 1/3] KVM: x86: provide realtime host clock via vsyscall notifiers 2017-01-13 16:28 ` Radim Krcmar @ 2017-01-13 17:51 ` Marcelo Tosatti 2017-01-16 15:40 ` Radim Krcmar 0 siblings, 1 reply; 54+ messages in thread From: Marcelo Tosatti @ 2017-01-13 17:51 UTC (permalink / raw) To: Radim Krcmar Cc: kvm, linux-kernel, Paolo Bonzini, Richard Cochran, Miroslav Lichvar On Fri, Jan 13, 2017 at 05:28:09PM +0100, Radim Krcmar wrote: > 2017-01-13 13:34-0200, Marcelo Tosatti: > > On Fri, Jan 13, 2017 at 04:18:04PM +0100, Radim Krcmar wrote: > >> 2017-01-13 10:01-0200, Marcelo Tosatti: > >> > Expose the realtime host clock and save the TSC value > >> > used for the clock calculation. > >> > > >> > Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com> > >> > > >> > --- > >> > arch/x86/kvm/x86.c | 38 ++++++++++++++++++++++++++++++++++++++ > >> > 1 file changed, 38 insertions(+) > >> > > >> > Index: kvm-ptpdriver/arch/x86/kvm/x86.c > >> > =================================================================== > >> > --- kvm-ptpdriver.orig/arch/x86/kvm/x86.c 2017-01-13 08:59:03.015895353 -0200 > >> > +++ kvm-ptpdriver/arch/x86/kvm/x86.c 2017-01-13 09:04:46.581415259 -0200 > >> > @@ -1139,6 +1139,8 @@ > >> > > >> > u64 boot_ns; > >> > u64 nsec_base; > >> > + u64 wall_time_sec; > >> > + u64 wall_time_snsec; > >> > >> The leading "s" in "snsec" looks like a copy-paste residue. > > > > Just copying the userspace vsyscall interface. > > Oh, so the "s" means "sub-" for sub-nanosecond precision. It only counts nanoseconds, how can it be sub nanosecond precise? > >> > }; > >> > > >> > static struct pvclock_gtod_data pvclock_gtod_data; > >> > @@ -1162,6 +1164,9 @@ > >> > vdata->boot_ns = boot_ns; > >> > vdata->nsec_base = tk->tkr_mono.xtime_nsec; > >> > > >> > + vdata->wall_time_sec = tk->xtime_sec; > >> > + vdata->wall_time_snsec = tk->tkr_mono.xtime_nsec; > >> > >> Using tk->tkr_mono offsets for real time seems wrong -- what happens if > >> the real time is half a second shifted from monotonic time? > > > > Both the userspace vsyscall interface and getnstimeofday > > use it for realtime clock. > > > > Monotonic clock adds the offset: > > > > vdata->monotonic_time_snsec = tk->tkr_mono.xtime_nsec > > + > > ((u64)tk->wall_to_monotonic.tv_nsec > > << tk->tkr_mono.shift); > > I see, thanks. Makes me wonder why our monotonic time is correct then, > but that is problably thanks to boot_ns. The actual starting point of the system_timestamp part of kvmclock does not matter, all it matters is that it counts in nanoseconds. > >> If it's ok, then vdata->nsec_base == vdata->wall_time_snsec, so we don't > >> need it. > > > > Just copying the userspace vsyscall interface. > > > > Do you actually want to change the "s" and unify wall_time_snsec with > > nsec_base? > > The "s" isn't important, even though I don't think we do anything that > would justify it, but make use just 8 bytes for both. Unified. > Renaming nsec_base is ok, but I'm not sure what tk->tkr_mono.xtime_nsec > is anymore. Is the nsec part of tk->xtime_sec. See accumulate_nsecs_to_secs (which is called from the timer interrupt handler). /** * struct timekeeper - Structure holding internal timekeeping values. * @tkr_mono: The readout base structure for CLOCK_MONOTONIC * @tkr_raw: The readout base structure for * CLOCK_MONOTONIC_RAW * @xtime_sec: Current CLOCK_REALTIME time in seconds * @ktime_sec: Current CLOCK_MONOTONIC time in seconds * @wall_to_monotonic: CLOCK_REALTIME to CLOCK_MONOTONIC offset * @offs_real: Offset clock monotonic -> clock realtime * @offs_boot: Offset clock monotonic -> clock boottime * @offs_tai: Offset clock monotonic -> clock tai * @tai_offset: The current UTC to TAI offset in seconds ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 1/3] KVM: x86: provide realtime host clock via vsyscall notifiers 2017-01-13 17:51 ` Marcelo Tosatti @ 2017-01-16 15:40 ` Radim Krcmar 0 siblings, 0 replies; 54+ messages in thread From: Radim Krcmar @ 2017-01-16 15:40 UTC (permalink / raw) To: Marcelo Tosatti Cc: kvm, linux-kernel, Paolo Bonzini, Richard Cochran, Miroslav Lichvar 2017-01-13 15:51-0200, Marcelo Tosatti: > On Fri, Jan 13, 2017 at 05:28:09PM +0100, Radim Krcmar wrote: >> 2017-01-13 13:34-0200, Marcelo Tosatti: >> > On Fri, Jan 13, 2017 at 04:18:04PM +0100, Radim Krcmar wrote: >> >> 2017-01-13 10:01-0200, Marcelo Tosatti: >> >> > Expose the realtime host clock and save the TSC value >> >> > used for the clock calculation. >> >> > >> >> > Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com> >> >> > >> >> > --- >> >> > arch/x86/kvm/x86.c | 38 ++++++++++++++++++++++++++++++++++++++ >> >> > 1 file changed, 38 insertions(+) >> >> > >> >> > Index: kvm-ptpdriver/arch/x86/kvm/x86.c >> >> > =================================================================== >> >> > --- kvm-ptpdriver.orig/arch/x86/kvm/x86.c 2017-01-13 08:59:03.015895353 -0200 >> >> > +++ kvm-ptpdriver/arch/x86/kvm/x86.c 2017-01-13 09:04:46.581415259 -0200 >> >> > @@ -1139,6 +1139,8 @@ >> >> > >> >> > u64 boot_ns; >> >> > u64 nsec_base; >> >> > + u64 wall_time_sec; >> >> > + u64 wall_time_snsec; >> >> >> >> The leading "s" in "snsec" looks like a copy-paste residue. >> > >> > Just copying the userspace vsyscall interface. >> >> Oh, so the "s" means "sub-" for sub-nanosecond precision. > > It only counts nanoseconds, how can it be sub nanosecond precise? Because it doesn't count nanoseconds. In update_vsyscall(), the *_snsec are shifted by tk->tkr_mono.shift bits to the left and that precision goes to sub-nanoseconds. 64 bit value makes sense then -- would be nice if we could pass that extra precision to the guest. >> >> > }; >> >> > >> >> > static struct pvclock_gtod_data pvclock_gtod_data; >> >> > @@ -1162,6 +1164,9 @@ >> >> > vdata->boot_ns = boot_ns; >> >> > vdata->nsec_base = tk->tkr_mono.xtime_nsec; >> >> > >> >> > + vdata->wall_time_sec = tk->xtime_sec; >> >> > + vdata->wall_time_snsec = tk->tkr_mono.xtime_nsec; >> >> >> >> Using tk->tkr_mono offsets for real time seems wrong -- what happens if >> >> the real time is half a second shifted from monotonic time? >> > >> > Both the userspace vsyscall interface and getnstimeofday >> > use it for realtime clock. >> > >> > Monotonic clock adds the offset: >> > >> > vdata->monotonic_time_snsec = tk->tkr_mono.xtime_nsec >> > + >> > ((u64)tk->wall_to_monotonic.tv_nsec >> > << tk->tkr_mono.shift); >> >> I see, thanks. Makes me wonder why our monotonic time is correct then, >> but that is problably thanks to boot_ns. > > The actual starting point of the system_timestamp part of kvmclock > does not matter, all it matters is that it counts in nanoseconds. True. >> >> If it's ok, then vdata->nsec_base == vdata->wall_time_snsec, so we don't >> >> need it. >> > >> > Just copying the userspace vsyscall interface. >> > >> > Do you actually want to change the "s" and unify wall_time_snsec with >> > nsec_base? >> >> The "s" isn't important, even though I don't think we do anything that >> would justify it, but make use just 8 bytes for both. > > Unified. > >> Renaming nsec_base is ok, but I'm not sure what tk->tkr_mono.xtime_nsec >> is anymore. > > Is the nsec part of tk->xtime_sec. See accumulate_nsecs_to_secs > (which is called from the timer interrupt handler). I see, thanks. They are not nanoseconds as the sub-nanosecond shift is there as well: u64 nsecps = (u64)NSEC_PER_SEC << tk->tkr_mono.shift; while (tk->tkr_mono.xtime_nsec >= nsecps) { tk->tkr_mono.xtime_nsec -= nsecps; tk->xtime_sec++; } ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 1/3] KVM: x86: provide realtime host clock via vsyscall notifiers 2017-01-13 15:18 ` Radim Krcmar 2017-01-13 15:34 ` Marcelo Tosatti @ 2017-01-13 15:41 ` Konrad Rzeszutek Wilk 2017-01-13 15:46 ` Marcelo Tosatti 1 sibling, 1 reply; 54+ messages in thread From: Konrad Rzeszutek Wilk @ 2017-01-13 15:41 UTC (permalink / raw) To: Radim Krcmar, xen-devel, Joao Martins Cc: Marcelo Tosatti, kvm, linux-kernel, Paolo Bonzini, Richard Cochran, Miroslav Lichvar On Fri, Jan 13, 2017 at 04:18:04PM +0100, Radim Krcmar wrote: > 2017-01-13 10:01-0200, Marcelo Tosatti: > > Expose the realtime host clock and save the TSC value > > used for the clock calculation. > > > > Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com> > > > > --- > > arch/x86/kvm/x86.c | 38 ++++++++++++++++++++++++++++++++++++++ > > 1 file changed, 38 insertions(+) > > > > Index: kvm-ptpdriver/arch/x86/kvm/x86.c > > =================================================================== > > --- kvm-ptpdriver.orig/arch/x86/kvm/x86.c 2017-01-13 08:59:03.015895353 -0200 > > +++ kvm-ptpdriver/arch/x86/kvm/x86.c 2017-01-13 09:04:46.581415259 -0200 > > @@ -1139,6 +1139,8 @@ > > > > u64 boot_ns; > > u64 nsec_base; > > + u64 wall_time_sec; > > + u64 wall_time_snsec; > > The leading "s" in "snsec" looks like a copy-paste residue. > > > }; > > > > static struct pvclock_gtod_data pvclock_gtod_data; > > @@ -1162,6 +1164,9 @@ > > vdata->boot_ns = boot_ns; > > vdata->nsec_base = tk->tkr_mono.xtime_nsec; > > > > + vdata->wall_time_sec = tk->xtime_sec; > > + vdata->wall_time_snsec = tk->tkr_mono.xtime_nsec; > > Using tk->tkr_mono offsets for real time seems wrong -- what happens if > the real time is half a second shifted from monotonic time? > > If it's ok, then vdata->nsec_base == vdata->wall_time_snsec, so we don't > need it. > > > + > > write_seqcount_end(&vdata->seq); > > } > > #endif > > @@ -1623,6 +1628,28 @@ > > return mode; > > } > > > > +static int do_realtime(struct timespec *ts, cycle_t *cycle_now) > > This is too similar to do_monotonic_boot(), but I don't see a solution > that is both nice and efficient. :( > > (It usually means macros or copying pvclock_gtod_data.) PV Clock is hypervisor agnostic so both KVM and Xen can use it. Is this clock interface suppose to follow that? Thanks. > > > +{ > > + 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->wall_time_snsec; > > + 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 +1659,17 @@ > > > > 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 > > > > /* > > > > > -- > To unsubscribe from this list: send the line "unsubscribe kvm" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 1/3] KVM: x86: provide realtime host clock via vsyscall notifiers 2017-01-13 15:41 ` Konrad Rzeszutek Wilk @ 2017-01-13 15:46 ` Marcelo Tosatti 0 siblings, 0 replies; 54+ messages in thread From: Marcelo Tosatti @ 2017-01-13 15:46 UTC (permalink / raw) To: Konrad Rzeszutek Wilk Cc: Radim Krcmar, xen-devel, Joao Martins, kvm, linux-kernel, Paolo Bonzini, Richard Cochran, Miroslav Lichvar On Fri, Jan 13, 2017 at 10:41:10AM -0500, Konrad Rzeszutek Wilk wrote: > On Fri, Jan 13, 2017 at 04:18:04PM +0100, Radim Krcmar wrote: > > 2017-01-13 10:01-0200, Marcelo Tosatti: > > > Expose the realtime host clock and save the TSC value > > > used for the clock calculation. > > > > > > Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com> > > > > > > --- > > > arch/x86/kvm/x86.c | 38 ++++++++++++++++++++++++++++++++++++++ > > > 1 file changed, 38 insertions(+) > > > > > > Index: kvm-ptpdriver/arch/x86/kvm/x86.c > > > =================================================================== > > > --- kvm-ptpdriver.orig/arch/x86/kvm/x86.c 2017-01-13 08:59:03.015895353 -0200 > > > +++ kvm-ptpdriver/arch/x86/kvm/x86.c 2017-01-13 09:04:46.581415259 -0200 > > > @@ -1139,6 +1139,8 @@ > > > > > > u64 boot_ns; > > > u64 nsec_base; > > > + u64 wall_time_sec; > > > + u64 wall_time_snsec; > > > > The leading "s" in "snsec" looks like a copy-paste residue. > > > > > }; > > > > > > static struct pvclock_gtod_data pvclock_gtod_data; > > > @@ -1162,6 +1164,9 @@ > > > vdata->boot_ns = boot_ns; > > > vdata->nsec_base = tk->tkr_mono.xtime_nsec; > > > > > > + vdata->wall_time_sec = tk->xtime_sec; > > > + vdata->wall_time_snsec = tk->tkr_mono.xtime_nsec; > > > > Using tk->tkr_mono offsets for real time seems wrong -- what happens if > > the real time is half a second shifted from monotonic time? > > > > If it's ok, then vdata->nsec_base == vdata->wall_time_snsec, so we don't > > need it. > > > > > + > > > write_seqcount_end(&vdata->seq); > > > } > > > #endif > > > @@ -1623,6 +1628,28 @@ > > > return mode; > > > } > > > > > > +static int do_realtime(struct timespec *ts, cycle_t *cycle_now) > > > > This is too similar to do_monotonic_boot(), but I don't see a solution > > that is both nice and efficient. :( > > > > (It usually means macros or copying pvclock_gtod_data.) > > > PV Clock is hypervisor agnostic so both KVM and Xen can use it. Is this clock > interface suppose to follow that? If Xen implements the KVM_HC_CLOCK_OFFSET hypercall, Xen guests can use the kvm ptp driver, yes. ^ permalink raw reply [flat|nested] 54+ messages in thread
* [patch 2/3] KVM: x86: add KVM_HC_CLOCK_OFFSET hypercall 2017-01-13 12:01 [patch 0/3] KVM virtual PTP driver Marcelo Tosatti 2017-01-13 12:01 ` [patch 1/3] KVM: x86: provide realtime host clock via vsyscall notifiers Marcelo Tosatti @ 2017-01-13 12:01 ` Marcelo Tosatti 2017-01-13 15:31 ` Radim Krcmar 2017-01-13 12:01 ` [patch 3/3] PTP: add kvm PTP driver Marcelo Tosatti 2 siblings, 1 reply; 54+ messages in thread From: Marcelo Tosatti @ 2017-01-13 12:01 UTC (permalink / raw) To: kvm, linux-kernel Cc: Paolo Bonzini, Radim Krcmar, Richard Cochran, Miroslav Lichvar, Marcelo Tosatti [-- Attachment #1: kvm-hv-clock-offset --] [-- Type: text/plain, Size: 4788 bytes --] Add a hypercall to retrieve the host realtime clock and the TSC value used to calculate that clock read. Used to implement clock synchronization between host and guest. Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com> --- Documentation/virtual/kvm/hypercalls.txt | 30 ++++++++++++++++++++++ arch/x86/include/uapi/asm/kvm_para.h | 9 ++++++ arch/x86/kvm/x86.c | 41 +++++++++++++++++++++++++++++++ include/uapi/linux/kvm_para.h | 3 ++ 4 files changed, 83 insertions(+) Index: kvm-ptpdriver/arch/x86/include/uapi/asm/kvm_para.h =================================================================== --- kvm-ptpdriver.orig/arch/x86/include/uapi/asm/kvm_para.h 2017-01-13 09:03:33.823305270 -0200 +++ kvm-ptpdriver/arch/x86/include/uapi/asm/kvm_para.h 2017-01-13 09:07:54.633699270 -0200 @@ -50,6 +50,15 @@ __u32 pad[11]; }; +#define KVM_CLOCK_OFFSET_WALLCLOCK 0 +struct kvm_clock_offset { + __s64 sec; + __s64 nsec; + __u64 tsc; + __u32 flags; + __u32 pad[1]; +}; + #define KVM_STEAL_ALIGNMENT_BITS 5 #define KVM_STEAL_VALID_BITS ((-1ULL << (KVM_STEAL_ALIGNMENT_BITS + 1))) #define KVM_STEAL_RESERVED_MASK (((1 << KVM_STEAL_ALIGNMENT_BITS) - 1 ) << 1) Index: kvm-ptpdriver/Documentation/virtual/kvm/hypercalls.txt =================================================================== --- kvm-ptpdriver.orig/Documentation/virtual/kvm/hypercalls.txt 2017-01-13 09:03:33.823305270 -0200 +++ kvm-ptpdriver/Documentation/virtual/kvm/hypercalls.txt 2017-01-13 09:15:32.971389949 -0200 @@ -81,3 +81,33 @@ same guest can wakeup the sleeping vcpu by issuing KVM_HC_KICK_CPU hypercall, specifying APIC ID (a1) of the vcpu to be woken up. An additional argument (a0) is used in the hypercall for future use. + + +6. KVM_HC_CLOCK_OFFSET +------------------------ +Architecture: x86 +Status: active +Purpose: Hypercall used to synchronize host and guest clocks. +Usage: + +a0: guest physical address where host copies +"struct kvm_clock_offset" structure. + +a1: clock_type, ATM only KVM_CLOCK_OFFSET_WALLCLOCK (0) +is supported (hosts CLOCK_REALTIME clock). + + struct kvm_clock_offset { + __s64 sec; + __s64 nsec; + __u64 tsc; + __u32 flags; + __u32 pad; + }; + + Where: + * sec: seconds from clock_type clock. + * nsec: nanoseconds from clock_type clock. + * tsc: TSC value used to calculate sec/nsec pair + (this hypercall only works when host uses TSC clocksource). + * flags: flags, unused at the moment. + Index: kvm-ptpdriver/arch/x86/kvm/x86.c =================================================================== --- kvm-ptpdriver.orig/arch/x86/kvm/x86.c 2017-01-13 09:04:46.581415259 -0200 +++ kvm-ptpdriver/arch/x86/kvm/x86.c 2017-01-13 09:08:51.557785166 -0200 @@ -6121,6 +6121,44 @@ } EXPORT_SYMBOL_GPL(kvm_emulate_halt); +static int kvm_pv_clock_offset(struct kvm_vcpu *vcpu, gpa_t paddr, + unsigned long clock_type) +{ + struct kvm_clock_offset *clock_offset; + struct timespec ts; + cycle_t cycle; + int ret; + + if (clock_type != KVM_CLOCK_OFFSET_WALLCLOCK) + return -KVM_EOPNOTSUPP; + + if (pvclock_gtod_data.clock.vclock_mode != VCLOCK_TSC) + return -KVM_EOPNOTSUPP; + + clock_offset = kzalloc(sizeof(struct kvm_clock_offset), GFP_KERNEL); + if (clock_offset == NULL) + return -KVM_ENOMEM; + + if (kvm_get_walltime_and_clockread(&ts, &cycle) == false) { + kfree(clock_offset); + return -KVM_EOPNOTSUPP; + } + + clock_offset->sec = ts.tv_sec; + clock_offset->nsec = ts.tv_nsec; + clock_offset->tsc = kvm_read_l1_tsc(vcpu, cycle); + clock_offset->flags = 0; + + ret = 0; + if (kvm_write_guest(vcpu->kvm, paddr, clock_offset, + sizeof(struct kvm_clock_offset))) + ret = -KVM_EFAULT; + + kfree(clock_offset); + + return ret; +} + /* * kvm_pv_kick_cpu_op: Kick a vcpu. * @@ -6185,6 +6223,9 @@ kvm_pv_kick_cpu_op(vcpu->kvm, a0, a1); ret = 0; break; + case KVM_HC_CLOCK_OFFSET: + ret = kvm_pv_clock_offset(vcpu, a0, a1); + break; default: ret = -KVM_ENOSYS; break; Index: kvm-ptpdriver/include/uapi/linux/kvm_para.h =================================================================== --- kvm-ptpdriver.orig/include/uapi/linux/kvm_para.h 2017-01-13 09:03:33.823305270 -0200 +++ kvm-ptpdriver/include/uapi/linux/kvm_para.h 2017-01-13 09:07:54.636699274 -0200 @@ -14,6 +14,8 @@ #define KVM_EFAULT EFAULT #define KVM_E2BIG E2BIG #define KVM_EPERM EPERM +#define KVM_EOPNOTSUPP EOPNOTSUPP +#define KVM_ENOMEM ENOMEM #define KVM_HC_VAPIC_POLL_IRQ 1 #define KVM_HC_MMU_OP 2 @@ -23,6 +25,7 @@ #define KVM_HC_MIPS_GET_CLOCK_FREQ 6 #define KVM_HC_MIPS_EXIT_VM 7 #define KVM_HC_MIPS_CONSOLE_OUTPUT 8 +#define KVM_HC_CLOCK_OFFSET 9 /* * hypercalls use architecture specific ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 2/3] KVM: x86: add KVM_HC_CLOCK_OFFSET hypercall 2017-01-13 12:01 ` [patch 2/3] KVM: x86: add KVM_HC_CLOCK_OFFSET hypercall Marcelo Tosatti @ 2017-01-13 15:31 ` Radim Krcmar 2017-01-13 15:43 ` Marcelo Tosatti 0 siblings, 1 reply; 54+ messages in thread From: Radim Krcmar @ 2017-01-13 15:31 UTC (permalink / raw) To: Marcelo Tosatti Cc: kvm, linux-kernel, Paolo Bonzini, Richard Cochran, Miroslav Lichvar 2017-01-13 10:01-0200, Marcelo Tosatti: > Add a hypercall to retrieve the host realtime clock > and the TSC value used to calculate that clock read. > > Used to implement clock synchronization between > host and guest. > > Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com> > > --- > Documentation/virtual/kvm/hypercalls.txt | 30 ++++++++++++++++++++++ > arch/x86/include/uapi/asm/kvm_para.h | 9 ++++++ > arch/x86/kvm/x86.c | 41 +++++++++++++++++++++++++++++++ > include/uapi/linux/kvm_para.h | 3 ++ > 4 files changed, 83 insertions(+) > > Index: kvm-ptpdriver/arch/x86/include/uapi/asm/kvm_para.h > =================================================================== > --- kvm-ptpdriver.orig/arch/x86/include/uapi/asm/kvm_para.h 2017-01-13 09:03:33.823305270 -0200 > +++ kvm-ptpdriver/arch/x86/include/uapi/asm/kvm_para.h 2017-01-13 09:07:54.633699270 -0200 > @@ -50,6 +50,15 @@ > __u32 pad[11]; > }; > > +#define KVM_CLOCK_OFFSET_WALLCLOCK 0 > +struct kvm_clock_offset { > + __s64 sec; > + __s64 nsec; > + __u64 tsc; > + __u32 flags; > + __u32 pad[1]; > +}; > + > #define KVM_STEAL_ALIGNMENT_BITS 5 > #define KVM_STEAL_VALID_BITS ((-1ULL << (KVM_STEAL_ALIGNMENT_BITS + 1))) > #define KVM_STEAL_RESERVED_MASK (((1 << KVM_STEAL_ALIGNMENT_BITS) - 1 ) << 1) > Index: kvm-ptpdriver/Documentation/virtual/kvm/hypercalls.txt > =================================================================== > --- kvm-ptpdriver.orig/Documentation/virtual/kvm/hypercalls.txt 2017-01-13 09:03:33.823305270 -0200 > +++ kvm-ptpdriver/Documentation/virtual/kvm/hypercalls.txt 2017-01-13 09:15:32.971389949 -0200 > @@ -81,3 +81,33 @@ > same guest can wakeup the sleeping vcpu by issuing KVM_HC_KICK_CPU hypercall, > specifying APIC ID (a1) of the vcpu to be woken up. An additional argument (a0) > is used in the hypercall for future use. > + > + > +6. KVM_HC_CLOCK_OFFSET > +------------------------ > +Architecture: x86 > +Status: active > +Purpose: Hypercall used to synchronize host and guest clocks. > +Usage: > + > +a0: guest physical address where host copies > +"struct kvm_clock_offset" structure. > + > +a1: clock_type, ATM only KVM_CLOCK_OFFSET_WALLCLOCK (0) > +is supported (hosts CLOCK_REALTIME clock). > + > + struct kvm_clock_offset { > + __s64 sec; > + __s64 nsec; Why is nsec: 1) signed -- it is a remainder after division by NSEC_PER_SEC 2) bigger than 32 bit -- NSEC_PER_SEC < 2^32 ? > + __u64 tsc; > + __u32 flags; > + __u32 pad; > + }; > + > + Where: > + * sec: seconds from clock_type clock. > + * nsec: nanoseconds from clock_type clock. The important part of an offset is the starting point -- I assume it is the the usual one, but documentation better be explicit. > + * tsc: TSC value used to calculate sec/nsec pair This hypercall could in theory work even when computing the offset with a different clocksource on the host and the estimating TSC at that time. > + (this hypercall only works when host uses TSC clocksource). Describe that error as a return value. > + * flags: flags, unused at the moment. unused = 0. > + > Index: kvm-ptpdriver/arch/x86/kvm/x86.c > =================================================================== > --- kvm-ptpdriver.orig/arch/x86/kvm/x86.c 2017-01-13 09:04:46.581415259 -0200 > +++ kvm-ptpdriver/arch/x86/kvm/x86.c 2017-01-13 09:08:51.557785166 -0200 > @@ -6121,6 +6121,44 @@ > } > EXPORT_SYMBOL_GPL(kvm_emulate_halt); > > +static int kvm_pv_clock_offset(struct kvm_vcpu *vcpu, gpa_t paddr, > + unsigned long clock_type) > +{ > + struct kvm_clock_offset *clock_offset; > + struct timespec ts; > + cycle_t cycle; > + int ret; > + > + if (clock_type != KVM_CLOCK_OFFSET_WALLCLOCK) > + return -KVM_EOPNOTSUPP; > + > + if (pvclock_gtod_data.clock.vclock_mode != VCLOCK_TSC) > + return -KVM_EOPNOTSUPP; > + > + clock_offset = kzalloc(sizeof(struct kvm_clock_offset), GFP_KERNEL); Can't clock_offset be on the stack? > + if (clock_offset == NULL) > + return -KVM_ENOMEM; > + > + if (kvm_get_walltime_and_clockread(&ts, &cycle) == false) { > + kfree(clock_offset); > + return -KVM_EOPNOTSUPP; > + } > + > + clock_offset->sec = ts.tv_sec; > + clock_offset->nsec = ts.tv_nsec; > + clock_offset->tsc = kvm_read_l1_tsc(vcpu, cycle); > + clock_offset->flags = 0; > + > + ret = 0; > + if (kvm_write_guest(vcpu->kvm, paddr, clock_offset, > + sizeof(struct kvm_clock_offset))) > + ret = -KVM_EFAULT; > + > + kfree(clock_offset); > + > + return ret; > +} > + > /* > * kvm_pv_kick_cpu_op: Kick a vcpu. > * > @@ -6185,6 +6223,9 @@ > kvm_pv_kick_cpu_op(vcpu->kvm, a0, a1); > ret = 0; > break; > + case KVM_HC_CLOCK_OFFSET: > + ret = kvm_pv_clock_offset(vcpu, a0, a1); > + break; > default: > ret = -KVM_ENOSYS; > break; > Index: kvm-ptpdriver/include/uapi/linux/kvm_para.h > =================================================================== > --- kvm-ptpdriver.orig/include/uapi/linux/kvm_para.h 2017-01-13 09:03:33.823305270 -0200 > +++ kvm-ptpdriver/include/uapi/linux/kvm_para.h 2017-01-13 09:07:54.636699274 -0200 > @@ -14,6 +14,8 @@ > #define KVM_EFAULT EFAULT > #define KVM_E2BIG E2BIG > #define KVM_EPERM EPERM > +#define KVM_EOPNOTSUPP EOPNOTSUPP > +#define KVM_ENOMEM ENOMEM > > #define KVM_HC_VAPIC_POLL_IRQ 1 > #define KVM_HC_MMU_OP 2 > @@ -23,6 +25,7 @@ > #define KVM_HC_MIPS_GET_CLOCK_FREQ 6 > #define KVM_HC_MIPS_EXIT_VM 7 > #define KVM_HC_MIPS_CONSOLE_OUTPUT 8 > +#define KVM_HC_CLOCK_OFFSET 9 > > /* > * hypercalls use architecture specific > > ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 2/3] KVM: x86: add KVM_HC_CLOCK_OFFSET hypercall 2017-01-13 15:31 ` Radim Krcmar @ 2017-01-13 15:43 ` Marcelo Tosatti 2017-01-13 17:07 ` Radim Krcmar 0 siblings, 1 reply; 54+ messages in thread From: Marcelo Tosatti @ 2017-01-13 15:43 UTC (permalink / raw) To: Radim Krcmar Cc: kvm, linux-kernel, Paolo Bonzini, Richard Cochran, Miroslav Lichvar On Fri, Jan 13, 2017 at 04:31:58PM +0100, Radim Krcmar wrote: > 2017-01-13 10:01-0200, Marcelo Tosatti: > > Add a hypercall to retrieve the host realtime clock > > and the TSC value used to calculate that clock read. > > > > Used to implement clock synchronization between > > host and guest. > > > > Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com> > > > > --- > > Documentation/virtual/kvm/hypercalls.txt | 30 ++++++++++++++++++++++ > > arch/x86/include/uapi/asm/kvm_para.h | 9 ++++++ > > arch/x86/kvm/x86.c | 41 +++++++++++++++++++++++++++++++ > > include/uapi/linux/kvm_para.h | 3 ++ > > 4 files changed, 83 insertions(+) > > > > Index: kvm-ptpdriver/arch/x86/include/uapi/asm/kvm_para.h > > =================================================================== > > --- kvm-ptpdriver.orig/arch/x86/include/uapi/asm/kvm_para.h 2017-01-13 09:03:33.823305270 -0200 > > +++ kvm-ptpdriver/arch/x86/include/uapi/asm/kvm_para.h 2017-01-13 09:07:54.633699270 -0200 > > @@ -50,6 +50,15 @@ > > __u32 pad[11]; > > }; > > > > +#define KVM_CLOCK_OFFSET_WALLCLOCK 0 > > +struct kvm_clock_offset { > > + __s64 sec; > > + __s64 nsec; > > + __u64 tsc; > > + __u32 flags; > > + __u32 pad[1]; > > +}; > > + > > #define KVM_STEAL_ALIGNMENT_BITS 5 > > #define KVM_STEAL_VALID_BITS ((-1ULL << (KVM_STEAL_ALIGNMENT_BITS + 1))) > > #define KVM_STEAL_RESERVED_MASK (((1 << KVM_STEAL_ALIGNMENT_BITS) - 1 ) << 1) > > Index: kvm-ptpdriver/Documentation/virtual/kvm/hypercalls.txt > > =================================================================== > > --- kvm-ptpdriver.orig/Documentation/virtual/kvm/hypercalls.txt 2017-01-13 09:03:33.823305270 -0200 > > +++ kvm-ptpdriver/Documentation/virtual/kvm/hypercalls.txt 2017-01-13 09:15:32.971389949 -0200 > > @@ -81,3 +81,33 @@ > > same guest can wakeup the sleeping vcpu by issuing KVM_HC_KICK_CPU hypercall, > > specifying APIC ID (a1) of the vcpu to be woken up. An additional argument (a0) > > is used in the hypercall for future use. > > + > > + > > +6. KVM_HC_CLOCK_OFFSET > > +------------------------ > > +Architecture: x86 > > +Status: active > > +Purpose: Hypercall used to synchronize host and guest clocks. > > +Usage: > > + > > +a0: guest physical address where host copies > > +"struct kvm_clock_offset" structure. > > + > > +a1: clock_type, ATM only KVM_CLOCK_OFFSET_WALLCLOCK (0) > > +is supported (hosts CLOCK_REALTIME clock). > > + > > + struct kvm_clock_offset { > > + __s64 sec; > > + __s64 nsec; > > Why is nsec: > 1) signed -- it is a remainder after division by NSEC_PER_SEC Because "struct timespec" is signed because it can be used for time deltas (you won't actually get signed values for kvm_get_walltime_and_clockread). Just wanted to match "struct timespec". > 2) bigger than 32 bit -- NSEC_PER_SEC < 2^32 > ? Again matching struct timespec. > > + __u64 tsc; > > + __u32 flags; > > + __u32 pad; > > + }; > > + > > + Where: > > + * sec: seconds from clock_type clock. > > + * nsec: nanoseconds from clock_type clock. > > The important part of an offset is the starting point -- I assume it is > the the usual one, but documentation better be explicit. Don't get what you mean? (the values have same meaning as hosts clock_gettime(CLOCK_REALTIME), supposedly that is clear). > > + * tsc: TSC value used to calculate sec/nsec pair > > This hypercall could in theory work even when computing the offset with > a different clocksource on the host and the estimating TSC at that time. Yes it could. > > + (this hypercall only works when host uses TSC clocksource). > > Describe that error as a return value. Fixed. + + if (kvm_get_walltime_and_clockread(&ts, &cycle) == false) { + kfree(clock_offset); + return -KVM_EOPNOTSUPP; + } > > > + * flags: flags, unused at the moment. > > unused = 0. Fixed. > > + > > Index: kvm-ptpdriver/arch/x86/kvm/x86.c > > =================================================================== > > --- kvm-ptpdriver.orig/arch/x86/kvm/x86.c 2017-01-13 09:04:46.581415259 -0200 > > +++ kvm-ptpdriver/arch/x86/kvm/x86.c 2017-01-13 09:08:51.557785166 -0200 > > @@ -6121,6 +6121,44 @@ > > } > > EXPORT_SYMBOL_GPL(kvm_emulate_halt); > > > > +static int kvm_pv_clock_offset(struct kvm_vcpu *vcpu, gpa_t paddr, > > + unsigned long clock_type) > > +{ > > + struct kvm_clock_offset *clock_offset; > > + struct timespec ts; > > + cycle_t cycle; > > + int ret; > > + > > + if (clock_type != KVM_CLOCK_OFFSET_WALLCLOCK) > > + return -KVM_EOPNOTSUPP; > > + > > + if (pvclock_gtod_data.clock.vclock_mode != VCLOCK_TSC) > > + return -KVM_EOPNOTSUPP; > > + > > + clock_offset = kzalloc(sizeof(struct kvm_clock_offset), GFP_KERNEL); > > Can't clock_offset be on the stack? Well since the kernel stack is limited, thought it would be safer to allocate it. > > + if (clock_offset == NULL) > > + return -KVM_ENOMEM; > > + > > + if (kvm_get_walltime_and_clockread(&ts, &cycle) == false) { > > + kfree(clock_offset); > > + return -KVM_EOPNOTSUPP; > > + } > > + > > + clock_offset->sec = ts.tv_sec; > > + clock_offset->nsec = ts.tv_nsec; > > + clock_offset->tsc = kvm_read_l1_tsc(vcpu, cycle); > > + clock_offset->flags = 0; > > + > > + ret = 0; > > + if (kvm_write_guest(vcpu->kvm, paddr, clock_offset, > > + sizeof(struct kvm_clock_offset))) > > + ret = -KVM_EFAULT; > > + > > + kfree(clock_offset); > > + > > + return ret; > > +} > > + > > /* > > * kvm_pv_kick_cpu_op: Kick a vcpu. > > * > > @@ -6185,6 +6223,9 @@ > > kvm_pv_kick_cpu_op(vcpu->kvm, a0, a1); > > ret = 0; > > break; > > + case KVM_HC_CLOCK_OFFSET: > > + ret = kvm_pv_clock_offset(vcpu, a0, a1); > > + break; > > default: > > ret = -KVM_ENOSYS; > > break; > > Index: kvm-ptpdriver/include/uapi/linux/kvm_para.h > > =================================================================== > > --- kvm-ptpdriver.orig/include/uapi/linux/kvm_para.h 2017-01-13 09:03:33.823305270 -0200 > > +++ kvm-ptpdriver/include/uapi/linux/kvm_para.h 2017-01-13 09:07:54.636699274 -0200 > > @@ -14,6 +14,8 @@ > > #define KVM_EFAULT EFAULT > > #define KVM_E2BIG E2BIG > > #define KVM_EPERM EPERM > > +#define KVM_EOPNOTSUPP EOPNOTSUPP > > +#define KVM_ENOMEM ENOMEM > > > > #define KVM_HC_VAPIC_POLL_IRQ 1 > > #define KVM_HC_MMU_OP 2 > > @@ -23,6 +25,7 @@ > > #define KVM_HC_MIPS_GET_CLOCK_FREQ 6 > > #define KVM_HC_MIPS_EXIT_VM 7 > > #define KVM_HC_MIPS_CONSOLE_OUTPUT 8 > > +#define KVM_HC_CLOCK_OFFSET 9 > > > > /* > > * hypercalls use architecture specific > > > > ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 2/3] KVM: x86: add KVM_HC_CLOCK_OFFSET hypercall 2017-01-13 15:43 ` Marcelo Tosatti @ 2017-01-13 17:07 ` Radim Krcmar 2017-01-13 17:57 ` Marcelo Tosatti 0 siblings, 1 reply; 54+ messages in thread From: Radim Krcmar @ 2017-01-13 17:07 UTC (permalink / raw) To: Marcelo Tosatti Cc: kvm, linux-kernel, Paolo Bonzini, Richard Cochran, Miroslav Lichvar 2017-01-13 13:43-0200, Marcelo Tosatti: > On Fri, Jan 13, 2017 at 04:31:58PM +0100, Radim Krcmar wrote: >> 2017-01-13 10:01-0200, Marcelo Tosatti: >> > Add a hypercall to retrieve the host realtime clock >> > and the TSC value used to calculate that clock read. >> > >> > Used to implement clock synchronization between >> > host and guest. >> > >> > Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com> >> > >> > --- >> > Index: kvm-ptpdriver/Documentation/virtual/kvm/hypercalls.txt >> > @@ -81,3 +81,33 @@ >> > +6. KVM_HC_CLOCK_OFFSET >> > +------------------------ >> > +Architecture: x86 >> > +Status: active >> > +Purpose: Hypercall used to synchronize host and guest clocks. >> > +Usage: >> > + >> > +a0: guest physical address where host copies >> > +"struct kvm_clock_offset" structure. >> > + >> > +a1: clock_type, ATM only KVM_CLOCK_OFFSET_WALLCLOCK (0) >> > +is supported (hosts CLOCK_REALTIME clock). >> > + >> > + struct kvm_clock_offset { >> > + __s64 sec; >> > + __s64 nsec; >> >> Why is nsec: >> 1) signed -- it is a remainder after division by NSEC_PER_SEC > > Because "struct timespec" is signed because it can be used for > time deltas (you won't actually get signed values for > kvm_get_walltime_and_clockread). > > Just wanted to match "struct timespec". > >> 2) bigger than 32 bit -- NSEC_PER_SEC < 2^32 >> ? > > Again matching struct timespec. It is "long" in struct timespec, which could also be "s32" ... I'd rather waste those 8 bytes inside padding -- its purpose is clear there. :) >> > + __u64 tsc; >> > + __u32 flags; >> > + __u32 pad; >> > + }; >> > + >> > + Where: >> > + * sec: seconds from clock_type clock. >> > + * nsec: nanoseconds from clock_type clock. >> >> The important part of an offset is the starting point -- I assume it is >> the the usual one, but documentation better be explicit. > > Don't get what you mean? (the values have same meaning as hosts > clock_gettime(CLOCK_REALTIME), supposedly that is clear). Ah, I didn't understand that clock_type was refering to CLOCK_REALTIME. I'd drop offset from the hypercall name. IIUC, all various clock types would be compared to TSC, so we could name the hypercall somewhat like KVM_HC_CLOCK_AT_TSC -- we want to know what was the time on the selected clock when TSC was at value __u64 tsc. The only offset is between sec+nsec and the beginning of time (and tsc and 0), but we don't care about that offset by itself -- we care about relation of sec+nsec to tsc, which isn't a simple offset as they flow differently. If we wanted to be more generic, then KVM_HC_CLOCK_PAIRING/SNAPSHOT/... and argument 1 would contain clock_types, here REALTIME and TSC. ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 2/3] KVM: x86: add KVM_HC_CLOCK_OFFSET hypercall 2017-01-13 17:07 ` Radim Krcmar @ 2017-01-13 17:57 ` Marcelo Tosatti 0 siblings, 0 replies; 54+ messages in thread From: Marcelo Tosatti @ 2017-01-13 17:57 UTC (permalink / raw) To: Radim Krcmar Cc: kvm, linux-kernel, Paolo Bonzini, Richard Cochran, Miroslav Lichvar On Fri, Jan 13, 2017 at 06:07:40PM +0100, Radim Krcmar wrote: > 2017-01-13 13:43-0200, Marcelo Tosatti: > > On Fri, Jan 13, 2017 at 04:31:58PM +0100, Radim Krcmar wrote: > >> 2017-01-13 10:01-0200, Marcelo Tosatti: > >> > Add a hypercall to retrieve the host realtime clock > >> > and the TSC value used to calculate that clock read. > >> > > >> > Used to implement clock synchronization between > >> > host and guest. > >> > > >> > Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com> > >> > > >> > --- > >> > Index: kvm-ptpdriver/Documentation/virtual/kvm/hypercalls.txt > >> > @@ -81,3 +81,33 @@ > >> > +6. KVM_HC_CLOCK_OFFSET > >> > +------------------------ > >> > +Architecture: x86 > >> > +Status: active > >> > +Purpose: Hypercall used to synchronize host and guest clocks. > >> > +Usage: > >> > + > >> > +a0: guest physical address where host copies > >> > +"struct kvm_clock_offset" structure. > >> > + > >> > +a1: clock_type, ATM only KVM_CLOCK_OFFSET_WALLCLOCK (0) > >> > +is supported (hosts CLOCK_REALTIME clock). > >> > + > >> > + struct kvm_clock_offset { > >> > + __s64 sec; > >> > + __s64 nsec; > >> > >> Why is nsec: > >> 1) signed -- it is a remainder after division by NSEC_PER_SEC > > > > Because "struct timespec" is signed because it can be used for > > time deltas (you won't actually get signed values for > > kvm_get_walltime_and_clockread). > > > > Just wanted to match "struct timespec". > > > >> 2) bigger than 32 bit -- NSEC_PER_SEC < 2^32 > >> ? > > > > Again matching struct timespec. > > It is "long" in struct timespec, which could also be "s32" ... Which fits in a s64. > I'd rather waste those 8 bytes inside padding -- its purpose is clear > there. :) The purpose of the s64 is to match timespec. > >> > + __u64 tsc; > >> > + __u32 flags; > >> > + __u32 pad; > >> > + }; > >> > + > >> > + Where: > >> > + * sec: seconds from clock_type clock. > >> > + * nsec: nanoseconds from clock_type clock. > >> > >> The important part of an offset is the starting point -- I assume it is > >> the the usual one, but documentation better be explicit. > > > > Don't get what you mean? (the values have same meaning as hosts > > clock_gettime(CLOCK_REALTIME), supposedly that is clear). > > Ah, I didn't understand that clock_type was refering to CLOCK_REALTIME. > > I'd drop offset from the hypercall name. IIUC, all various clock types > would be compared to TSC, so we could name the hypercall somewhat like > KVM_HC_CLOCK_AT_TSC -- we want to know what was the time on the selected > clock when TSC was at value __u64 tsc. Well its the offset from TSC, so thats why "offset" in the name. And as you say, the interface could be extended to support other clocks (using the available data in pad, i'll make it larger BTW). Which would render the "TSC" from the hypercall name invalid. > The only offset is between sec+nsec and the beginning of time (and tsc > and 0), but we don't care about that offset by itself -- we care about > relation of sec+nsec to tsc, which isn't a simple offset as they flow > differently. > > If we wanted to be more generic, then KVM_HC_CLOCK_PAIRING/SNAPSHOT/... > and argument 1 would contain clock_types, here REALTIME and TSC. Pairing is fine, changed. ^ permalink raw reply [flat|nested] 54+ messages in thread
* [patch 3/3] PTP: add kvm PTP driver 2017-01-13 12:01 [patch 0/3] KVM virtual PTP driver Marcelo Tosatti 2017-01-13 12:01 ` [patch 1/3] KVM: x86: provide realtime host clock via vsyscall notifiers Marcelo Tosatti 2017-01-13 12:01 ` [patch 2/3] KVM: x86: add KVM_HC_CLOCK_OFFSET hypercall Marcelo Tosatti @ 2017-01-13 12:01 ` Marcelo Tosatti 2017-01-13 15:56 ` Radim Krcmar 2 siblings, 1 reply; 54+ messages in thread From: Marcelo Tosatti @ 2017-01-13 12:01 UTC (permalink / raw) To: kvm, linux-kernel Cc: Paolo Bonzini, Radim Krcmar, Richard Cochran, Miroslav Lichvar [-- Attachment #1: kvm-ptpdriver --] [-- Type: text/plain, Size: 6517 bytes --] Add a driver with gettime method returning hosts realtime clock. This allows Chrony to synchronize host and guest clocks with high precision (see results below). chronyc> sources MS Name/IP address Stratum Poll Reach LastRx Last sample =============================================================================== #* PHC0 0 3 377 6 +4ns[ +4ns] +/- 3ns To configure Chronyd to use PHC refclock, add the following line to its configuration file: refclock PHC /dev/ptpX poll 3 dpoll -2 offset 0 Where /dev/ptpX is the kvmclock PTP clock. --- drivers/ptp/Kconfig | 12 +++ drivers/ptp/Makefile | 1 drivers/ptp/ptp_kvm.c | 180 ++++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 193 insertions(+) Index: kvm-ptpdriver/drivers/ptp/Kconfig =================================================================== --- kvm-ptpdriver.orig/drivers/ptp/Kconfig 2017-01-13 09:17:31.724568567 -0200 +++ kvm-ptpdriver/drivers/ptp/Kconfig 2017-01-13 09:55:33.344208894 -0200 @@ -90,4 +90,16 @@ To compile this driver as a module, choose M here: the module will be called ptp_pch. +config PTP_1588_CLOCK_KVM + tristate "KVM virtual PTP clock" + depends on PTP_1588_CLOCK + depends on KVM_GUEST + default y + help + This driver adds support for using kvm infrastructure as a PTP + clock. This clock is only useful if you are using KVM guests. + + To compile this driver as a module, choose M here: the module + will be called ptp_kvm. + endmenu Index: kvm-ptpdriver/drivers/ptp/ptp_kvm.c =================================================================== --- /dev/null 1970-01-01 00:00:00.000000000 +0000 +++ kvm-ptpdriver/drivers/ptp/ptp_kvm.c 2017-01-13 09:57:55.013440645 -0200 @@ -0,0 +1,180 @@ +/* + * Virtual PTP 1588 clock for use with KVM guests + * + * Copyright (C) 2017 Red Hat Inc. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + */ +#include <linux/device.h> +#include <linux/err.h> +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <uapi/linux/kvm_para.h> +#include <asm/kvm_para.h> +#include <asm/pvclock.h> +#include <uapi/asm/kvm_para.h> + +#include <linux/ptp_clock_kernel.h> + +struct kvm_ptp_clock { + struct ptp_clock *ptp_clock; + struct ptp_clock_info caps; +}; + +DEFINE_SPINLOCK(kvm_ptp_lock); + +static struct pvclock_vsyscall_time_info *hv_clock; + +/* + * PTP clock operations + */ + +static int ptp_kvm_adjfreq(struct ptp_clock_info *ptp, s32 ppb) +{ + return -EOPNOTSUPP; +} + +static int ptp_kvm_adjtime(struct ptp_clock_info *ptp, s64 delta) +{ + return -EOPNOTSUPP; +} + +static struct kvm_clock_offset clock_off; +static phys_addr_t clock_off_gpa; + +static int ptp_kvm_gettime(struct ptp_clock_info *ptp, struct timespec64 *ts) +{ + unsigned long ret; + struct timespec64 tspec; + u64 delta; + cycle_t offset; + unsigned version; + int cpu; + struct pvclock_vcpu_time_info *src; + + preempt_disable_notrace(); + cpu = smp_processor_id(); + src = &hv_clock[cpu].pvti; + + spin_lock(&kvm_ptp_lock); + + do { + /* + * We are measuring the delay between + * kvm_hypercall and rdtsc using TSC, + * and converting that delta to + * tsc_to_system_mul and tsc_shift + * So any changes to tsc_to_system_mul + * and tsc_shift in this region + * invalidate the measurement. + */ + version = pvclock_read_begin(src); + + ret = kvm_hypercall2(KVM_HC_CLOCK_OFFSET, + clock_off_gpa, + KVM_CLOCK_OFFSET_WALLCLOCK); + if (ret != 0) { + pr_err("clock offset hypercall ret %lu\n", ret); + spin_unlock(&kvm_ptp_lock); + preempt_enable_notrace(); + return -EOPNOTSUPP; + } + + tspec.tv_sec = clock_off.sec; + tspec.tv_nsec = clock_off.nsec; + + delta = rdtsc_ordered() - clock_off.tsc; + + offset = pvclock_scale_delta(delta, src->tsc_to_system_mul, + src->tsc_shift); + + } while (pvclock_read_retry(src, version)); + + preempt_enable_notrace(); + + tspec.tv_nsec = tspec.tv_nsec + offset; + + spin_unlock(&kvm_ptp_lock); + + if (tspec.tv_nsec >= NSEC_PER_SEC) { + u64 secs = tspec.tv_nsec; + + tspec.tv_nsec = do_div(secs, NSEC_PER_SEC); + tspec.tv_sec += secs; + } + + memcpy(ts, &tspec, sizeof(struct timespec64)); + + return 0; +} + +static int ptp_kvm_settime(struct ptp_clock_info *ptp, + const struct timespec64 *ts) +{ + return -EOPNOTSUPP; +} + +static int ptp_kvm_enable(struct ptp_clock_info *ptp, + struct ptp_clock_request *rq, int on) +{ + return -EOPNOTSUPP; +} + +static struct ptp_clock_info ptp_kvm_caps = { + .owner = THIS_MODULE, + .name = "KVM virtual PTP", + .max_adj = 0, + .n_ext_ts = 0, + .n_pins = 0, + .pps = 0, + .adjfreq = ptp_kvm_adjfreq, + .adjtime = ptp_kvm_adjtime, + .gettime64 = ptp_kvm_gettime, + .settime64 = ptp_kvm_settime, + .enable = ptp_kvm_enable, +}; + +/* module operations */ + +static struct kvm_ptp_clock kvm_ptp_clock; + +static void __exit ptp_kvm_exit(void) +{ + ptp_clock_unregister(kvm_ptp_clock.ptp_clock); +} + +static int __init ptp_kvm_init(void) +{ + if (!kvm_para_available()) + return -ENODEV; + + kvm_ptp_clock.caps = ptp_kvm_caps; + + kvm_ptp_clock.ptp_clock = ptp_clock_register(&kvm_ptp_clock.caps, NULL); + + if (IS_ERR(kvm_ptp_clock.ptp_clock)) + return PTR_ERR(kvm_ptp_clock.ptp_clock); + + clock_off_gpa = slow_virt_to_phys(&clock_off); + + hv_clock = pvclock_pvti_cpu0_va(); + + return 0; +} + +module_init(ptp_kvm_init); +module_exit(ptp_kvm_exit); + +MODULE_AUTHOR("Marcelo Tosatti <mtosatti@redhat.com>"); +MODULE_DESCRIPTION("PTP clock using KVMCLOCK"); +MODULE_LICENSE("GPL"); Index: kvm-ptpdriver/drivers/ptp/Makefile =================================================================== --- kvm-ptpdriver.orig/drivers/ptp/Makefile 2017-01-13 09:17:31.724568567 -0200 +++ kvm-ptpdriver/drivers/ptp/Makefile 2017-01-13 09:17:58.997609570 -0200 @@ -6,3 +6,4 @@ obj-$(CONFIG_PTP_1588_CLOCK) += ptp.o obj-$(CONFIG_PTP_1588_CLOCK_IXP46X) += ptp_ixp46x.o obj-$(CONFIG_PTP_1588_CLOCK_PCH) += ptp_pch.o +obj-$(CONFIG_PTP_1588_CLOCK_KVM) += ptp_kvm.o ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-13 12:01 ` [patch 3/3] PTP: add kvm PTP driver Marcelo Tosatti @ 2017-01-13 15:56 ` Radim Krcmar 2017-01-13 17:40 ` Marcelo Tosatti 2017-01-14 15:26 ` Richard Cochran 0 siblings, 2 replies; 54+ messages in thread From: Radim Krcmar @ 2017-01-13 15:56 UTC (permalink / raw) To: Marcelo Tosatti Cc: kvm, linux-kernel, Paolo Bonzini, Richard Cochran, Miroslav Lichvar 2017-01-13 10:01-0200, Marcelo Tosatti: > Add a driver with gettime method returning hosts realtime clock. > This allows Chrony to synchronize host and guest clocks with > high precision (see results below). > > chronyc> sources > MS Name/IP address Stratum Poll Reach LastRx Last sample > =============================================================================== > #* PHC0 0 3 377 6 +4ns[ +4ns] +/- 3ns > > To configure Chronyd to use PHC refclock, add the > following line to its configuration file: > > refclock PHC /dev/ptpX poll 3 dpoll -2 offset 0 > > Where /dev/ptpX is the kvmclock PTP clock. > > > --- > drivers/ptp/Kconfig | 12 +++ > drivers/ptp/Makefile | 1 > drivers/ptp/ptp_kvm.c | 180 ++++++++++++++++++++++++++++++++++++++++++++++++++ > 3 files changed, 193 insertions(+) > > Index: kvm-ptpdriver/drivers/ptp/Kconfig > =================================================================== > --- kvm-ptpdriver.orig/drivers/ptp/Kconfig 2017-01-13 09:17:31.724568567 -0200 > +++ kvm-ptpdriver/drivers/ptp/Kconfig 2017-01-13 09:55:33.344208894 -0200 > @@ -90,4 +90,16 @@ > To compile this driver as a module, choose M here: the module > will be called ptp_pch. > > +config PTP_1588_CLOCK_KVM > + tristate "KVM virtual PTP clock" > + depends on PTP_1588_CLOCK > + depends on KVM_GUEST > + default y > + help > + This driver adds support for using kvm infrastructure as a PTP > + clock. This clock is only useful if you are using KVM guests. > + > + To compile this driver as a module, choose M here: the module > + will be called ptp_kvm. > + > endmenu > Index: kvm-ptpdriver/drivers/ptp/ptp_kvm.c > =================================================================== > --- /dev/null 1970-01-01 00:00:00.000000000 +0000 > +++ kvm-ptpdriver/drivers/ptp/ptp_kvm.c 2017-01-13 09:57:55.013440645 -0200 > @@ -0,0 +1,180 @@ > +/* > + * Virtual PTP 1588 clock for use with KVM guests > + * > + * Copyright (C) 2017 Red Hat Inc. > + * > + * This program is free software; you can redistribute it and/or modify > + * it under the terms of the GNU General Public License as published by > + * the Free Software Foundation; either version 2 of the License, or > + * (at your option) any later version. > + * > + * This program is distributed in the hope that it will be useful, > + * but WITHOUT ANY WARRANTY; without even the implied warranty of > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the > + * GNU General Public License for more details. > + * > + */ > +#include <linux/device.h> > +#include <linux/err.h> > +#include <linux/init.h> > +#include <linux/kernel.h> > +#include <linux/module.h> > +#include <uapi/linux/kvm_para.h> > +#include <asm/kvm_para.h> > +#include <asm/pvclock.h> > +#include <uapi/asm/kvm_para.h> > + > +#include <linux/ptp_clock_kernel.h> > + > +struct kvm_ptp_clock { > + struct ptp_clock *ptp_clock; > + struct ptp_clock_info caps; > +}; > + > +DEFINE_SPINLOCK(kvm_ptp_lock); > + > +static struct pvclock_vsyscall_time_info *hv_clock; > + > +/* > + * PTP clock operations > + */ > + > +static int ptp_kvm_adjfreq(struct ptp_clock_info *ptp, s32 ppb) > +{ > + return -EOPNOTSUPP; > +} > + > +static int ptp_kvm_adjtime(struct ptp_clock_info *ptp, s64 delta) > +{ > + return -EOPNOTSUPP; > +} > + > +static struct kvm_clock_offset clock_off; > +static phys_addr_t clock_off_gpa; > + > +static int ptp_kvm_gettime(struct ptp_clock_info *ptp, struct timespec64 *ts) > +{ > + unsigned long ret; > + struct timespec64 tspec; > + u64 delta; > + cycle_t offset; > + unsigned version; > + int cpu; > + struct pvclock_vcpu_time_info *src; > + > + preempt_disable_notrace(); > + cpu = smp_processor_id(); > + src = &hv_clock[cpu].pvti; > + > + spin_lock(&kvm_ptp_lock); > + > + do { > + /* > + * We are measuring the delay between > + * kvm_hypercall and rdtsc using TSC, > + * and converting that delta to > + * tsc_to_system_mul and tsc_shift > + * So any changes to tsc_to_system_mul > + * and tsc_shift in this region > + * invalidate the measurement. > + */ This assumes that the host uses kvmclock, but the guest can be using just TSC and even have kvmclock disabled. This code should at least check that kvmclock enabled. (If we made kvmclock mandatory, then we could fix pvclock_wall_clock interface, because it is already defined to provide real time based on kvmclock ...) > + version = pvclock_read_begin(src); > + > + ret = kvm_hypercall2(KVM_HC_CLOCK_OFFSET, > + clock_off_gpa, > + KVM_CLOCK_OFFSET_WALLCLOCK); > + if (ret != 0) { > + pr_err("clock offset hypercall ret %lu\n", ret); > + spin_unlock(&kvm_ptp_lock); > + preempt_enable_notrace(); > + return -EOPNOTSUPP; > + } > + > + tspec.tv_sec = clock_off.sec; > + tspec.tv_nsec = clock_off.nsec; > + > + delta = rdtsc_ordered() - clock_off.tsc; > + > + offset = pvclock_scale_delta(delta, src->tsc_to_system_mul, > + src->tsc_shift); > + > + } while (pvclock_read_retry(src, version)); > + > + preempt_enable_notrace(); > + > + tspec.tv_nsec = tspec.tv_nsec + offset; > + > + spin_unlock(&kvm_ptp_lock); > + > + if (tspec.tv_nsec >= NSEC_PER_SEC) { > + u64 secs = tspec.tv_nsec; > + > + tspec.tv_nsec = do_div(secs, NSEC_PER_SEC); > + tspec.tv_sec += secs; > + } > + > + memcpy(ts, &tspec, sizeof(struct timespec64)); But the whole idea is of improving the time by reading tsc a bit later is just weird ... why is it better to provide tsc + x, time + tsc_delta_to_time(x) than just tsc, time ? Because we'll always be quering the time at tsc + y, where y >> x, and we'd likely have other problems if shifting the time base by few thousand cycles made a difference. > + > + return 0; > +} > + > +static int ptp_kvm_settime(struct ptp_clock_info *ptp, > + const struct timespec64 *ts) > +{ > + return -EOPNOTSUPP; > +} > + > +static int ptp_kvm_enable(struct ptp_clock_info *ptp, > + struct ptp_clock_request *rq, int on) > +{ > + return -EOPNOTSUPP; > +} > + > +static struct ptp_clock_info ptp_kvm_caps = { > + .owner = THIS_MODULE, > + .name = "KVM virtual PTP", > + .max_adj = 0, > + .n_ext_ts = 0, > + .n_pins = 0, > + .pps = 0, > + .adjfreq = ptp_kvm_adjfreq, > + .adjtime = ptp_kvm_adjtime, > + .gettime64 = ptp_kvm_gettime, > + .settime64 = ptp_kvm_settime, > + .enable = ptp_kvm_enable, > +}; > + > +/* module operations */ > + > +static struct kvm_ptp_clock kvm_ptp_clock; > + > +static void __exit ptp_kvm_exit(void) > +{ > + ptp_clock_unregister(kvm_ptp_clock.ptp_clock); > +} > + > +static int __init ptp_kvm_init(void) > +{ > + if (!kvm_para_available()) > + return -ENODEV; > + > + kvm_ptp_clock.caps = ptp_kvm_caps; > + > + kvm_ptp_clock.ptp_clock = ptp_clock_register(&kvm_ptp_clock.caps, NULL); It is a shame that the infrastructure uses polling when the guest could be notified on every host real time change, but this should be good enough. > + > + if (IS_ERR(kvm_ptp_clock.ptp_clock)) > + return PTR_ERR(kvm_ptp_clock.ptp_clock); > + > + clock_off_gpa = slow_virt_to_phys(&clock_off); > + > + hv_clock = pvclock_pvti_cpu0_va(); Would safer to assign required globals before the registration -- races could get ugly. Thanks. > + > + return 0; > +} > + > +module_init(ptp_kvm_init); > +module_exit(ptp_kvm_exit); > + > +MODULE_AUTHOR("Marcelo Tosatti <mtosatti@redhat.com>"); > +MODULE_DESCRIPTION("PTP clock using KVMCLOCK"); > +MODULE_LICENSE("GPL"); > Index: kvm-ptpdriver/drivers/ptp/Makefile > =================================================================== > --- kvm-ptpdriver.orig/drivers/ptp/Makefile 2017-01-13 09:17:31.724568567 -0200 > +++ kvm-ptpdriver/drivers/ptp/Makefile 2017-01-13 09:17:58.997609570 -0200 > @@ -6,3 +6,4 @@ > obj-$(CONFIG_PTP_1588_CLOCK) += ptp.o > obj-$(CONFIG_PTP_1588_CLOCK_IXP46X) += ptp_ixp46x.o > obj-$(CONFIG_PTP_1588_CLOCK_PCH) += ptp_pch.o > +obj-$(CONFIG_PTP_1588_CLOCK_KVM) += ptp_kvm.o > > ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-13 15:56 ` Radim Krcmar @ 2017-01-13 17:40 ` Marcelo Tosatti 2017-01-16 16:26 ` Radim Krcmar 2017-01-14 15:26 ` Richard Cochran 1 sibling, 1 reply; 54+ messages in thread From: Marcelo Tosatti @ 2017-01-13 17:40 UTC (permalink / raw) To: Radim Krcmar Cc: kvm, linux-kernel, Paolo Bonzini, Richard Cochran, Miroslav Lichvar On Fri, Jan 13, 2017 at 04:56:58PM +0100, Radim Krcmar wrote: > 2017-01-13 10:01-0200, Marcelo Tosatti: > > Add a driver with gettime method returning hosts realtime clock. > > This allows Chrony to synchronize host and guest clocks with > > high precision (see results below). > > > > chronyc> sources > > MS Name/IP address Stratum Poll Reach LastRx Last sample > > =============================================================================== > > #* PHC0 0 3 377 6 +4ns[ +4ns] +/- 3ns > > > > To configure Chronyd to use PHC refclock, add the > > following line to its configuration file: > > > > refclock PHC /dev/ptpX poll 3 dpoll -2 offset 0 > > > > Where /dev/ptpX is the kvmclock PTP clock. > > > > > > --- > > drivers/ptp/Kconfig | 12 +++ > > drivers/ptp/Makefile | 1 > > drivers/ptp/ptp_kvm.c | 180 ++++++++++++++++++++++++++++++++++++++++++++++++++ > > 3 files changed, 193 insertions(+) > > > > Index: kvm-ptpdriver/drivers/ptp/Kconfig > > =================================================================== > > --- kvm-ptpdriver.orig/drivers/ptp/Kconfig 2017-01-13 09:17:31.724568567 -0200 > > +++ kvm-ptpdriver/drivers/ptp/Kconfig 2017-01-13 09:55:33.344208894 -0200 > > @@ -90,4 +90,16 @@ > > To compile this driver as a module, choose M here: the module > > will be called ptp_pch. > > > > +config PTP_1588_CLOCK_KVM > > + tristate "KVM virtual PTP clock" > > + depends on PTP_1588_CLOCK > > + depends on KVM_GUEST > > + default y > > + help > > + This driver adds support for using kvm infrastructure as a PTP > > + clock. This clock is only useful if you are using KVM guests. > > + > > + To compile this driver as a module, choose M here: the module > > + will be called ptp_kvm. > > + > > endmenu > > Index: kvm-ptpdriver/drivers/ptp/ptp_kvm.c > > =================================================================== > > --- /dev/null 1970-01-01 00:00:00.000000000 +0000 > > +++ kvm-ptpdriver/drivers/ptp/ptp_kvm.c 2017-01-13 09:57:55.013440645 -0200 > > @@ -0,0 +1,180 @@ > > +/* > > + * Virtual PTP 1588 clock for use with KVM guests > > + * > > + * Copyright (C) 2017 Red Hat Inc. > > + * > > + * This program is free software; you can redistribute it and/or modify > > + * it under the terms of the GNU General Public License as published by > > + * the Free Software Foundation; either version 2 of the License, or > > + * (at your option) any later version. > > + * > > + * This program is distributed in the hope that it will be useful, > > + * but WITHOUT ANY WARRANTY; without even the implied warranty of > > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the > > + * GNU General Public License for more details. > > + * > > + */ > > +#include <linux/device.h> > > +#include <linux/err.h> > > +#include <linux/init.h> > > +#include <linux/kernel.h> > > +#include <linux/module.h> > > +#include <uapi/linux/kvm_para.h> > > +#include <asm/kvm_para.h> > > +#include <asm/pvclock.h> > > +#include <uapi/asm/kvm_para.h> > > + > > +#include <linux/ptp_clock_kernel.h> > > + > > +struct kvm_ptp_clock { > > + struct ptp_clock *ptp_clock; > > + struct ptp_clock_info caps; > > +}; > > + > > +DEFINE_SPINLOCK(kvm_ptp_lock); > > + > > +static struct pvclock_vsyscall_time_info *hv_clock; > > + > > +/* > > + * PTP clock operations > > + */ > > + > > +static int ptp_kvm_adjfreq(struct ptp_clock_info *ptp, s32 ppb) > > +{ > > + return -EOPNOTSUPP; > > +} > > + > > +static int ptp_kvm_adjtime(struct ptp_clock_info *ptp, s64 delta) > > +{ > > + return -EOPNOTSUPP; > > +} > > + > > +static struct kvm_clock_offset clock_off; > > +static phys_addr_t clock_off_gpa; > > + > > +static int ptp_kvm_gettime(struct ptp_clock_info *ptp, struct timespec64 *ts) > > +{ > > + unsigned long ret; > > + struct timespec64 tspec; > > + u64 delta; > > + cycle_t offset; > > + unsigned version; > > + int cpu; > > + struct pvclock_vcpu_time_info *src; > > + > > + preempt_disable_notrace(); > > + cpu = smp_processor_id(); > > + src = &hv_clock[cpu].pvti; > > + > > + spin_lock(&kvm_ptp_lock); > > + > > + do { > > + /* > > + * We are measuring the delay between > > + * kvm_hypercall and rdtsc using TSC, > > + * and converting that delta to > > + * tsc_to_system_mul and tsc_shift > > + * So any changes to tsc_to_system_mul > > + * and tsc_shift in this region > > + * invalidate the measurement. > > + */ > > This assumes that the host uses kvmclock, but the guest can be using > just TSC and even have kvmclock disabled. This code should at least > check that kvmclock enabled. Fixed. > (If we made kvmclock mandatory, then we could fix pvclock_wall_clock > interface, because it is already defined to provide real time based on > kvmclock ...) > > > + version = pvclock_read_begin(src); > > + > > + ret = kvm_hypercall2(KVM_HC_CLOCK_OFFSET, > > + clock_off_gpa, > > + KVM_CLOCK_OFFSET_WALLCLOCK); > > + if (ret != 0) { > > + pr_err("clock offset hypercall ret %lu\n", ret); > > + spin_unlock(&kvm_ptp_lock); > > + preempt_enable_notrace(); > > + return -EOPNOTSUPP; > > + } > > + > > + tspec.tv_sec = clock_off.sec; > > + tspec.tv_nsec = clock_off.nsec; > > + > > + delta = rdtsc_ordered() - clock_off.tsc; > > + > > + offset = pvclock_scale_delta(delta, src->tsc_to_system_mul, > > + src->tsc_shift); > > + > > + } while (pvclock_read_retry(src, version)); > > + > > + preempt_enable_notrace(); > > + > > + tspec.tv_nsec = tspec.tv_nsec + offset; > > + > > + spin_unlock(&kvm_ptp_lock); > > + > > + if (tspec.tv_nsec >= NSEC_PER_SEC) { > > + u64 secs = tspec.tv_nsec; > > + > > + tspec.tv_nsec = do_div(secs, NSEC_PER_SEC); > > + tspec.tv_sec += secs; > > + } > > + > > + memcpy(ts, &tspec, sizeof(struct timespec64)); > > But the whole idea is of improving the time by reading tsc a bit later > is just weird ... why is it better to provide > > tsc + x, time + tsc_delta_to_time(x) > > than just > > tsc, time > > ? Because you want to calculate the value of the host realtime clock at the moment of ptp_kvm_gettime. We do: 1. kvm_hypercall. 2. get {sec, nsec, guest_tsc}. 3. kvm_hypercall returns. 4. delay = rdtsc() - guest_tsc. Where delay is the delta (measured with the TSC) between points 2 and 4. > Because we'll always be quering the time at tsc + y, where y >> x, and > we'd likely have other problems if shifting the time base by few > thousand cycles made a difference. Radim, i didnt get your "tsc + x", "time + tsc_delta_to_time(x)" formulas above. Can you be more verbose please? > > + > > + return 0; > > +} > > + > > +static int ptp_kvm_settime(struct ptp_clock_info *ptp, > > + const struct timespec64 *ts) > > +{ > > + return -EOPNOTSUPP; > > +} > > + > > +static int ptp_kvm_enable(struct ptp_clock_info *ptp, > > + struct ptp_clock_request *rq, int on) > > +{ > > + return -EOPNOTSUPP; > > +} > > + > > +static struct ptp_clock_info ptp_kvm_caps = { > > + .owner = THIS_MODULE, > > + .name = "KVM virtual PTP", > > + .max_adj = 0, > > + .n_ext_ts = 0, > > + .n_pins = 0, > > + .pps = 0, > > + .adjfreq = ptp_kvm_adjfreq, > > + .adjtime = ptp_kvm_adjtime, > > + .gettime64 = ptp_kvm_gettime, > > + .settime64 = ptp_kvm_settime, > > + .enable = ptp_kvm_enable, > > +}; > > + > > +/* module operations */ > > + > > +static struct kvm_ptp_clock kvm_ptp_clock; > > + > > +static void __exit ptp_kvm_exit(void) > > +{ > > + ptp_clock_unregister(kvm_ptp_clock.ptp_clock); > > +} > > + > > +static int __init ptp_kvm_init(void) > > +{ > > + if (!kvm_para_available()) > > + return -ENODEV; > > + > > + kvm_ptp_clock.caps = ptp_kvm_caps; > > + > > + kvm_ptp_clock.ptp_clock = ptp_clock_register(&kvm_ptp_clock.caps, NULL); > > It is a shame that the infrastructure uses polling when the guest could > be notified on every host real time change, but this should be good > enough. Well, the whole NTP scheme is a feedback loop, so it wants to query the value of the clock at the moments the feedback loop logic decides it. > > + if (IS_ERR(kvm_ptp_clock.ptp_clock)) > > + return PTR_ERR(kvm_ptp_clock.ptp_clock); > > + > > + clock_off_gpa = slow_virt_to_phys(&clock_off); > > + > > + hv_clock = pvclock_pvti_cpu0_va(); > > Would safer to assign required globals before the registration -- races > could get ugly. Fixed, thanks. ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-13 17:40 ` Marcelo Tosatti @ 2017-01-16 16:26 ` Radim Krcmar 2017-01-16 16:54 ` Radim Krcmar 2017-01-16 17:04 ` Marcelo Tosatti 0 siblings, 2 replies; 54+ messages in thread From: Radim Krcmar @ 2017-01-16 16:26 UTC (permalink / raw) To: Marcelo Tosatti Cc: kvm, linux-kernel, Paolo Bonzini, Richard Cochran, Miroslav Lichvar 2017-01-13 15:40-0200, Marcelo Tosatti: > On Fri, Jan 13, 2017 at 04:56:58PM +0100, Radim Krcmar wrote: > > 2017-01-13 10:01-0200, Marcelo Tosatti: >> > + version = pvclock_read_begin(src); >> > + >> > + ret = kvm_hypercall2(KVM_HC_CLOCK_OFFSET, >> > + clock_off_gpa, >> > + KVM_CLOCK_OFFSET_WALLCLOCK); >> > + if (ret != 0) { >> > + pr_err("clock offset hypercall ret %lu\n", ret); >> > + spin_unlock(&kvm_ptp_lock); >> > + preempt_enable_notrace(); >> > + return -EOPNOTSUPP; >> > + } >> > + >> > + tspec.tv_sec = clock_off.sec; >> > + tspec.tv_nsec = clock_off.nsec; >> > + >> > + delta = rdtsc_ordered() - clock_off.tsc; >> > + >> > + offset = pvclock_scale_delta(delta, src->tsc_to_system_mul, >> > + src->tsc_shift); >> > + >> > + } while (pvclock_read_retry(src, version)); >> > + >> > + preempt_enable_notrace(); >> > + >> > + tspec.tv_nsec = tspec.tv_nsec + offset; >> > + >> > + spin_unlock(&kvm_ptp_lock); >> > + >> > + if (tspec.tv_nsec >= NSEC_PER_SEC) { >> > + u64 secs = tspec.tv_nsec; >> > + >> > + tspec.tv_nsec = do_div(secs, NSEC_PER_SEC); >> > + tspec.tv_sec += secs; >> > + } >> > + >> > + memcpy(ts, &tspec, sizeof(struct timespec64)); >> >> But the whole idea is of improving the time by reading tsc a bit later >> is just weird ... why is it better to provide >> >> tsc + x, time + tsc_delta_to_time(x) >> >> than just >> >> tsc, time >> >> ? > > Because you want to calculate the value of the host realtime clock > at the moment of ptp_kvm_gettime. > > We do: > > 1. kvm_hypercall. > 2. get {sec, nsec, guest_tsc}. > 3. kvm_hypercall returns. > 4. delay = rdtsc() - guest_tsc. > > Where delay is the delta (measured with the TSC) between points 2 and 4. I see now ... the PTP interface is just not good for our purposes. We don't return {sec, nsec, guest_tsc}, we just return {sec, nsec} at some random time in the past. And to make it a bit more accurate, you add a best-effort delta before returning, which makes sense. When we have to depend on pvclock, what are the advantages of not using the existing pvclock API for wall clock? (You mentioned some extensions.) struct pvclock_wall_clock { u32 version; u32 sec; u32 nsec; } __attribute__((__packed__)); It gives the wall clock when pvclock was 0, so you just add current kvmclock and get the host wall clock. Without a VM exit. And how often is ptp_kvm_gettime() usually called? Thanks. >> Because we'll always be quering the time at tsc + y, where y >> x, and >> we'd likely have other problems if shifting the time base by few >> thousand cycles made a difference. > > Radim, i didnt get your "tsc + x", "time + tsc_delta_to_time(x)" > formulas above. Can you be more verbose please? x is the delta, tsc_delta_to_time() is what pvclock_scale_delta() does. I assumed that we set precise time with TSC, so the delta wouldn't matter, because PTP would either get {sec, nsec, guest_tsc}, or the same, but just shifted by delta, hence {sec + tsc_delta_to_time(x) / NSEC_PER_SEC, nsec + tsc_delta_to_time(x) % NSEC_PER_SEC, guest_tsc + x}. ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-16 16:26 ` Radim Krcmar @ 2017-01-16 16:54 ` Radim Krcmar 2017-01-16 17:08 ` Marcelo Tosatti 2017-01-16 17:04 ` Marcelo Tosatti 1 sibling, 1 reply; 54+ messages in thread From: Radim Krcmar @ 2017-01-16 16:54 UTC (permalink / raw) To: Marcelo Tosatti Cc: kvm, linux-kernel, Paolo Bonzini, Richard Cochran, Miroslav Lichvar 2017-01-16 17:26+0100, Radim Krcmar: > 2017-01-13 15:40-0200, Marcelo Tosatti: >> On Fri, Jan 13, 2017 at 04:56:58PM +0100, Radim Krcmar wrote: >> > 2017-01-13 10:01-0200, Marcelo Tosatti: >>> > + version = pvclock_read_begin(src); >>> > + >>> > + ret = kvm_hypercall2(KVM_HC_CLOCK_OFFSET, >>> > + clock_off_gpa, >>> > + KVM_CLOCK_OFFSET_WALLCLOCK); >>> > + if (ret != 0) { >>> > + pr_err("clock offset hypercall ret %lu\n", ret); >>> > + spin_unlock(&kvm_ptp_lock); >>> > + preempt_enable_notrace(); >>> > + return -EOPNOTSUPP; >>> > + } >>> > + >>> > + tspec.tv_sec = clock_off.sec; >>> > + tspec.tv_nsec = clock_off.nsec; >>> > + >>> > + delta = rdtsc_ordered() - clock_off.tsc; >>> > + >>> > + offset = pvclock_scale_delta(delta, src->tsc_to_system_mul, >>> > + src->tsc_shift); >>> > + >>> > + } while (pvclock_read_retry(src, version)); >>> > + >>> > + preempt_enable_notrace(); >>> > + >>> > + tspec.tv_nsec = tspec.tv_nsec + offset; >>> > + >>> > + spin_unlock(&kvm_ptp_lock); >>> > + >>> > + if (tspec.tv_nsec >= NSEC_PER_SEC) { >>> > + u64 secs = tspec.tv_nsec; >>> > + >>> > + tspec.tv_nsec = do_div(secs, NSEC_PER_SEC); >>> > + tspec.tv_sec += secs; >>> > + } >>> > + >>> > + memcpy(ts, &tspec, sizeof(struct timespec64)); >>> >>> But the whole idea is of improving the time by reading tsc a bit later >>> is just weird ... why is it better to provide >>> >>> tsc + x, time + tsc_delta_to_time(x) >>> >>> than just >>> >>> tsc, time >>> >>> ? >> >> Because you want to calculate the value of the host realtime clock >> at the moment of ptp_kvm_gettime. >> >> We do: >> >> 1. kvm_hypercall. >> 2. get {sec, nsec, guest_tsc}. >> 3. kvm_hypercall returns. >> 4. delay = rdtsc() - guest_tsc. >> >> Where delay is the delta (measured with the TSC) between points 2 and 4. > > I see now ... the PTP interface is just not good for our purposes. There is getcrosststamp() callback in PTP, which seems to be exactly what we want when pairing with TSC, so the pvclock delay fixup can be dropped when using it. ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-16 16:54 ` Radim Krcmar @ 2017-01-16 17:08 ` Marcelo Tosatti 2017-01-16 17:27 ` Radim Krcmar 0 siblings, 1 reply; 54+ messages in thread From: Marcelo Tosatti @ 2017-01-16 17:08 UTC (permalink / raw) To: Radim Krcmar Cc: kvm, linux-kernel, Paolo Bonzini, Richard Cochran, Miroslav Lichvar On Mon, Jan 16, 2017 at 05:54:11PM +0100, Radim Krcmar wrote: > 2017-01-16 17:26+0100, Radim Krcmar: > > 2017-01-13 15:40-0200, Marcelo Tosatti: > >> On Fri, Jan 13, 2017 at 04:56:58PM +0100, Radim Krcmar wrote: > >> > 2017-01-13 10:01-0200, Marcelo Tosatti: > >>> > + version = pvclock_read_begin(src); > >>> > + > >>> > + ret = kvm_hypercall2(KVM_HC_CLOCK_OFFSET, > >>> > + clock_off_gpa, > >>> > + KVM_CLOCK_OFFSET_WALLCLOCK); > >>> > + if (ret != 0) { > >>> > + pr_err("clock offset hypercall ret %lu\n", ret); > >>> > + spin_unlock(&kvm_ptp_lock); > >>> > + preempt_enable_notrace(); > >>> > + return -EOPNOTSUPP; > >>> > + } > >>> > + > >>> > + tspec.tv_sec = clock_off.sec; > >>> > + tspec.tv_nsec = clock_off.nsec; > >>> > + > >>> > + delta = rdtsc_ordered() - clock_off.tsc; > >>> > + > >>> > + offset = pvclock_scale_delta(delta, src->tsc_to_system_mul, > >>> > + src->tsc_shift); > >>> > + > >>> > + } while (pvclock_read_retry(src, version)); > >>> > + > >>> > + preempt_enable_notrace(); > >>> > + > >>> > + tspec.tv_nsec = tspec.tv_nsec + offset; > >>> > + > >>> > + spin_unlock(&kvm_ptp_lock); > >>> > + > >>> > + if (tspec.tv_nsec >= NSEC_PER_SEC) { > >>> > + u64 secs = tspec.tv_nsec; > >>> > + > >>> > + tspec.tv_nsec = do_div(secs, NSEC_PER_SEC); > >>> > + tspec.tv_sec += secs; > >>> > + } > >>> > + > >>> > + memcpy(ts, &tspec, sizeof(struct timespec64)); > >>> > >>> But the whole idea is of improving the time by reading tsc a bit later > >>> is just weird ... why is it better to provide > >>> > >>> tsc + x, time + tsc_delta_to_time(x) > >>> > >>> than just > >>> > >>> tsc, time > >>> > >>> ? > >> > >> Because you want to calculate the value of the host realtime clock > >> at the moment of ptp_kvm_gettime. > >> > >> We do: > >> > >> 1. kvm_hypercall. > >> 2. get {sec, nsec, guest_tsc}. > >> 3. kvm_hypercall returns. > >> 4. delay = rdtsc() - guest_tsc. > >> > >> Where delay is the delta (measured with the TSC) between points 2 and 4. > > > > I see now ... the PTP interface is just not good for our purposes. > > There is getcrosststamp() callback in PTP, which seems to be exactly > what we want when pairing with TSC, so the pvclock delay fixup can be > dropped when using it. What pvclock delay fixup you refer to? The "rdtsc() - clock_offset.tsc" part? You can't drop it, because if you do then your "host realtime clock read" will be behind by "rdtsc() - clock_offset.tsc" TSC cycles. We want the highest precision as possible. ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-16 17:08 ` Marcelo Tosatti @ 2017-01-16 17:27 ` Radim Krcmar 2017-01-16 17:39 ` Marcelo Tosatti 0 siblings, 1 reply; 54+ messages in thread From: Radim Krcmar @ 2017-01-16 17:27 UTC (permalink / raw) To: Marcelo Tosatti Cc: kvm, linux-kernel, Paolo Bonzini, Richard Cochran, Miroslav Lichvar 2017-01-16 15:08-0200, Marcelo Tosatti: > On Mon, Jan 16, 2017 at 05:54:11PM +0100, Radim Krcmar wrote: >> 2017-01-16 17:26+0100, Radim Krcmar: >> > 2017-01-13 15:40-0200, Marcelo Tosatti: >> >> On Fri, Jan 13, 2017 at 04:56:58PM +0100, Radim Krcmar wrote: >> >> > 2017-01-13 10:01-0200, Marcelo Tosatti: >> >>> > + version = pvclock_read_begin(src); >> >>> > + >> >>> > + ret = kvm_hypercall2(KVM_HC_CLOCK_OFFSET, >> >>> > + clock_off_gpa, >> >>> > + KVM_CLOCK_OFFSET_WALLCLOCK); >> >>> > + if (ret != 0) { >> >>> > + pr_err("clock offset hypercall ret %lu\n", ret); >> >>> > + spin_unlock(&kvm_ptp_lock); >> >>> > + preempt_enable_notrace(); >> >>> > + return -EOPNOTSUPP; >> >>> > + } >> >>> > + >> >>> > + tspec.tv_sec = clock_off.sec; >> >>> > + tspec.tv_nsec = clock_off.nsec; >> >>> > + >> >>> > + delta = rdtsc_ordered() - clock_off.tsc; >> >>> > + >> >>> > + offset = pvclock_scale_delta(delta, src->tsc_to_system_mul, >> >>> > + src->tsc_shift); >> >>> > + >> >>> > + } while (pvclock_read_retry(src, version)); >> >>> > + >> >>> > + preempt_enable_notrace(); >> >>> > + >> >>> > + tspec.tv_nsec = tspec.tv_nsec + offset; >> >>> > + >> >>> > + spin_unlock(&kvm_ptp_lock); >> >>> > + >> >>> > + if (tspec.tv_nsec >= NSEC_PER_SEC) { >> >>> > + u64 secs = tspec.tv_nsec; >> >>> > + >> >>> > + tspec.tv_nsec = do_div(secs, NSEC_PER_SEC); >> >>> > + tspec.tv_sec += secs; >> >>> > + } >> >>> > + >> >>> > + memcpy(ts, &tspec, sizeof(struct timespec64)); >> >>> >> >>> But the whole idea is of improving the time by reading tsc a bit later >> >>> is just weird ... why is it better to provide >> >>> >> >>> tsc + x, time + tsc_delta_to_time(x) >> >>> >> >>> than just >> >>> >> >>> tsc, time >> >>> >> >>> ? >> >> >> >> Because you want to calculate the value of the host realtime clock >> >> at the moment of ptp_kvm_gettime. >> >> >> >> We do: >> >> >> >> 1. kvm_hypercall. >> >> 2. get {sec, nsec, guest_tsc}. >> >> 3. kvm_hypercall returns. >> >> 4. delay = rdtsc() - guest_tsc. >> >> >> >> Where delay is the delta (measured with the TSC) between points 2 and 4. >> > >> > I see now ... the PTP interface is just not good for our purposes. >> >> There is getcrosststamp() callback in PTP, which seems to be exactly >> what we want when pairing with TSC, so the pvclock delay fixup can be >> dropped when using it. > > What pvclock delay fixup you refer to? The "rdtsc() - clock_offset.tsc" > part? Yes. > You can't drop it, because if you do then your "host realtime > clock read" will be behind by "rdtsc() - clock_offset.tsc" TSC cycles. The TSC read will be some cycles old when the hypercall ends, but that doesn't matter, because we will pass {sec, nsec, guest_tsc} to PTP and PTP should plug them into kernel's realtime clock roughly like this: sec/nsec + (rdtsc() - guest_tsc) * tsc_freq Adding delay to guest_tsc and sec/nsec cannot improve precision. (And will likely degrade it as kvmclock's frequency is incorrect.) > We want the highest precision as possible. I agree, which is why we don't want to lose precision in the delay guesswork because of gettime64(). ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-16 17:27 ` Radim Krcmar @ 2017-01-16 17:39 ` Marcelo Tosatti 2017-01-16 18:01 ` Radim Krcmar 0 siblings, 1 reply; 54+ messages in thread From: Marcelo Tosatti @ 2017-01-16 17:39 UTC (permalink / raw) To: Radim Krcmar Cc: kvm, linux-kernel, Paolo Bonzini, Richard Cochran, Miroslav Lichvar On Mon, Jan 16, 2017 at 06:27:58PM +0100, Radim Krcmar wrote: > 2017-01-16 15:08-0200, Marcelo Tosatti: > > On Mon, Jan 16, 2017 at 05:54:11PM +0100, Radim Krcmar wrote: > >> 2017-01-16 17:26+0100, Radim Krcmar: > >> > 2017-01-13 15:40-0200, Marcelo Tosatti: > >> >> On Fri, Jan 13, 2017 at 04:56:58PM +0100, Radim Krcmar wrote: > >> >> > 2017-01-13 10:01-0200, Marcelo Tosatti: > >> >>> > + version = pvclock_read_begin(src); > >> >>> > + > >> >>> > + ret = kvm_hypercall2(KVM_HC_CLOCK_OFFSET, > >> >>> > + clock_off_gpa, > >> >>> > + KVM_CLOCK_OFFSET_WALLCLOCK); > >> >>> > + if (ret != 0) { > >> >>> > + pr_err("clock offset hypercall ret %lu\n", ret); > >> >>> > + spin_unlock(&kvm_ptp_lock); > >> >>> > + preempt_enable_notrace(); > >> >>> > + return -EOPNOTSUPP; > >> >>> > + } > >> >>> > + > >> >>> > + tspec.tv_sec = clock_off.sec; > >> >>> > + tspec.tv_nsec = clock_off.nsec; > >> >>> > + > >> >>> > + delta = rdtsc_ordered() - clock_off.tsc; > >> >>> > + > >> >>> > + offset = pvclock_scale_delta(delta, src->tsc_to_system_mul, > >> >>> > + src->tsc_shift); > >> >>> > + > >> >>> > + } while (pvclock_read_retry(src, version)); > >> >>> > + > >> >>> > + preempt_enable_notrace(); > >> >>> > + > >> >>> > + tspec.tv_nsec = tspec.tv_nsec + offset; > >> >>> > + > >> >>> > + spin_unlock(&kvm_ptp_lock); > >> >>> > + > >> >>> > + if (tspec.tv_nsec >= NSEC_PER_SEC) { > >> >>> > + u64 secs = tspec.tv_nsec; > >> >>> > + > >> >>> > + tspec.tv_nsec = do_div(secs, NSEC_PER_SEC); > >> >>> > + tspec.tv_sec += secs; > >> >>> > + } > >> >>> > + > >> >>> > + memcpy(ts, &tspec, sizeof(struct timespec64)); > >> >>> > >> >>> But the whole idea is of improving the time by reading tsc a bit later > >> >>> is just weird ... why is it better to provide > >> >>> > >> >>> tsc + x, time + tsc_delta_to_time(x) > >> >>> > >> >>> than just > >> >>> > >> >>> tsc, time > >> >>> > >> >>> ? > >> >> > >> >> Because you want to calculate the value of the host realtime clock > >> >> at the moment of ptp_kvm_gettime. > >> >> > >> >> We do: > >> >> > >> >> 1. kvm_hypercall. > >> >> 2. get {sec, nsec, guest_tsc}. > >> >> 3. kvm_hypercall returns. > >> >> 4. delay = rdtsc() - guest_tsc. > >> >> > >> >> Where delay is the delta (measured with the TSC) between points 2 and 4. > >> > > >> > I see now ... the PTP interface is just not good for our purposes. > >> > >> There is getcrosststamp() callback in PTP, which seems to be exactly > >> what we want when pairing with TSC, so the pvclock delay fixup can be > >> dropped when using it. > > > > What pvclock delay fixup you refer to? The "rdtsc() - clock_offset.tsc" > > part? > > Yes. > > > You can't drop it, because if you do then your "host realtime > > clock read" will be behind by "rdtsc() - clock_offset.tsc" TSC cycles. > > The TSC read will be some cycles old when the hypercall ends, but that > doesn't matter, because we will pass {sec, nsec, guest_tsc} to PTP and > PTP should plug them into kernel's realtime clock roughly like this: > > sec/nsec + (rdtsc() - guest_tsc) * tsc_freq > > Adding delay to guest_tsc and sec/nsec cannot improve precision. > (And will likely degrade it as kvmclock's frequency is incorrect.) > > > We want the highest precision as possible. > > I agree, which is why we don't want to lose precision in the delay > guesswork because of gettime64(). Sorry the clock difference is 10ns now. So the guest clock is off by _10 ns_ of the host clock. You are suggesting to use getcrosststamp instead, to drop the (rdtsc() - guest_tsc) part ? Please be more verbose. ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-16 17:39 ` Marcelo Tosatti @ 2017-01-16 18:01 ` Radim Krcmar 2017-01-16 19:36 ` Marcelo Tosatti 0 siblings, 1 reply; 54+ messages in thread From: Radim Krcmar @ 2017-01-16 18:01 UTC (permalink / raw) To: Marcelo Tosatti Cc: kvm, linux-kernel, Paolo Bonzini, Richard Cochran, Miroslav Lichvar 2017-01-16 15:39-0200, Marcelo Tosatti: > On Mon, Jan 16, 2017 at 06:27:58PM +0100, Radim Krcmar wrote: >> 2017-01-16 15:08-0200, Marcelo Tosatti: >> > On Mon, Jan 16, 2017 at 05:54:11PM +0100, Radim Krcmar wrote: >> >> 2017-01-16 17:26+0100, Radim Krcmar: >> >> > 2017-01-13 15:40-0200, Marcelo Tosatti: >> >> >> On Fri, Jan 13, 2017 at 04:56:58PM +0100, Radim Krcmar wrote: >> >> >> > 2017-01-13 10:01-0200, Marcelo Tosatti: >> >> >>> > + version = pvclock_read_begin(src); >> >> >>> > + >> >> >>> > + ret = kvm_hypercall2(KVM_HC_CLOCK_OFFSET, >> >> >>> > + clock_off_gpa, >> >> >>> > + KVM_CLOCK_OFFSET_WALLCLOCK); >> >> >>> > + if (ret != 0) { >> >> >>> > + pr_err("clock offset hypercall ret %lu\n", ret); >> >> >>> > + spin_unlock(&kvm_ptp_lock); >> >> >>> > + preempt_enable_notrace(); >> >> >>> > + return -EOPNOTSUPP; >> >> >>> > + } >> >> >>> > + >> >> >>> > + tspec.tv_sec = clock_off.sec; >> >> >>> > + tspec.tv_nsec = clock_off.nsec; >> >> >>> > + >> >> >>> > + delta = rdtsc_ordered() - clock_off.tsc; >> >> >>> > + >> >> >>> > + offset = pvclock_scale_delta(delta, src->tsc_to_system_mul, >> >> >>> > + src->tsc_shift); >> >> >>> > + >> >> >>> > + } while (pvclock_read_retry(src, version)); >> >> >>> > + >> >> >>> > + preempt_enable_notrace(); >> >> >>> > + >> >> >>> > + tspec.tv_nsec = tspec.tv_nsec + offset; >> >> >>> > + >> >> >>> > + spin_unlock(&kvm_ptp_lock); >> >> >>> > + >> >> >>> > + if (tspec.tv_nsec >= NSEC_PER_SEC) { >> >> >>> > + u64 secs = tspec.tv_nsec; >> >> >>> > + >> >> >>> > + tspec.tv_nsec = do_div(secs, NSEC_PER_SEC); >> >> >>> > + tspec.tv_sec += secs; >> >> >>> > + } >> >> >>> > + >> >> >>> > + memcpy(ts, &tspec, sizeof(struct timespec64)); >> >> >>> >> >> >>> But the whole idea is of improving the time by reading tsc a bit later >> >> >>> is just weird ... why is it better to provide >> >> >>> >> >> >>> tsc + x, time + tsc_delta_to_time(x) >> >> >>> >> >> >>> than just >> >> >>> >> >> >>> tsc, time >> >> >>> >> >> >>> ? >> >> >> >> >> >> Because you want to calculate the value of the host realtime clock >> >> >> at the moment of ptp_kvm_gettime. >> >> >> >> >> >> We do: >> >> >> >> >> >> 1. kvm_hypercall. >> >> >> 2. get {sec, nsec, guest_tsc}. >> >> >> 3. kvm_hypercall returns. >> >> >> 4. delay = rdtsc() - guest_tsc. >> >> >> >> >> >> Where delay is the delta (measured with the TSC) between points 2 and 4. >> >> > >> >> > I see now ... the PTP interface is just not good for our purposes. >> >> >> >> There is getcrosststamp() callback in PTP, which seems to be exactly >> >> what we want when pairing with TSC, so the pvclock delay fixup can be >> >> dropped when using it. >> > >> > What pvclock delay fixup you refer to? The "rdtsc() - clock_offset.tsc" >> > part? >> >> Yes. >> >> > You can't drop it, because if you do then your "host realtime >> > clock read" will be behind by "rdtsc() - clock_offset.tsc" TSC cycles. >> >> The TSC read will be some cycles old when the hypercall ends, but that >> doesn't matter, because we will pass {sec, nsec, guest_tsc} to PTP and >> PTP should plug them into kernel's realtime clock roughly like this: >> >> sec/nsec + (rdtsc() - guest_tsc) * tsc_freq >> >> Adding delay to guest_tsc and sec/nsec cannot improve precision. >> (And will likely degrade it as kvmclock's frequency is incorrect.) >> >> > We want the highest precision as possible. >> >> I agree, which is why we don't want to lose precision in the delay >> guesswork because of gettime64(). > > Sorry the clock difference is 10ns now. So the guest clock is off by _10 ns_ > of the host clock. That is pretty good. > You are suggesting to use getcrosststamp instead, to drop the (rdtsc() - > guest_tsc) part ? Yes, it results in simpler code, doesn't create dependency on the dreaded kvmclock, and is the best we can currently do wrt. precision. Thanks. ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-16 18:01 ` Radim Krcmar @ 2017-01-16 19:36 ` Marcelo Tosatti 2017-01-16 19:47 ` Marcelo Tosatti 0 siblings, 1 reply; 54+ messages in thread From: Marcelo Tosatti @ 2017-01-16 19:36 UTC (permalink / raw) To: Radim Krcmar, Miroslav Lichvar Cc: kvm, linux-kernel, Paolo Bonzini, Richard Cochran, Miroslav Lichvar On Mon, Jan 16, 2017 at 07:01:48PM +0100, Radim Krcmar wrote: > > Sorry the clock difference is 10ns now. So the guest clock is off by _10 ns_ > > of the host clock. > > That is pretty good. Yes. > > You are suggesting to use getcrosststamp instead, to drop the (rdtsc() - > > guest_tsc) part ? > > Yes, it results in simpler code, doesn't create dependency on the > dreaded kvmclock, and is the best we can currently do wrt. precision. Sorry, unless i am misunderstanding how this works, it'll get the guest clock 2us behind, which is something not wanted. Miroslav, if ->gettime64 returns the host realtime at 2us in the past, this means Chrony will sync the guest clock to host realtime - 2us Is that correct? ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-16 19:36 ` Marcelo Tosatti @ 2017-01-16 19:47 ` Marcelo Tosatti 2017-01-16 20:01 ` Marcelo Tosatti 0 siblings, 1 reply; 54+ messages in thread From: Marcelo Tosatti @ 2017-01-16 19:47 UTC (permalink / raw) To: Radim Krcmar, Miroslav Lichvar Cc: kvm, linux-kernel, Paolo Bonzini, Richard Cochran On Mon, Jan 16, 2017 at 05:36:55PM -0200, Marcelo Tosatti wrote: > On Mon, Jan 16, 2017 at 07:01:48PM +0100, Radim Krcmar wrote: > > > Sorry the clock difference is 10ns now. So the guest clock is off by _10 ns_ > > > of the host clock. > > > > That is pretty good. > > Yes. > > > > You are suggesting to use getcrosststamp instead, to drop the (rdtsc() - > > > guest_tsc) part ? > > > > Yes, it results in simpler code, doesn't create dependency on the > > dreaded kvmclock, and is the best we can currently do wrt. precision. Even if the PHC sync algorithm manages to detect that the clock read is incorrect, consider the following: Variability in the VM-entry code path, such as cache effects and interrupts would cause certain readings to be longer then the average (assuming an average where cache is hot). Using the TSC removes this variability, which can be large in case of non realtime guests, where you do: 1. kvm_hypercall. 2. read host realtime clock. 3. schedule out qemu-kvm vcpu. 4. schedule in qemu-kvm vcpu. So using the delta between read host realtime and ->gettime64 increases precision and decreases variability. > Sorry, unless i am misunderstanding how this works, it'll get the guest clock > 2us behind, which is something not wanted. > > Miroslav, if ->gettime64 returns the host realtime at 2us in the past, > this means Chrony will sync the guest clock to > > host realtime - 2us > > Is that correct? > ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-16 19:47 ` Marcelo Tosatti @ 2017-01-16 20:01 ` Marcelo Tosatti 2017-01-17 8:03 ` Miroslav Lichvar 0 siblings, 1 reply; 54+ messages in thread From: Marcelo Tosatti @ 2017-01-16 20:01 UTC (permalink / raw) To: Radim Krcmar, Miroslav Lichvar Cc: kvm, linux-kernel, Paolo Bonzini, Richard Cochran On Mon, Jan 16, 2017 at 05:47:15PM -0200, Marcelo Tosatti wrote: > On Mon, Jan 16, 2017 at 05:36:55PM -0200, Marcelo Tosatti wrote: > > On Mon, Jan 16, 2017 at 07:01:48PM +0100, Radim Krcmar wrote: > > > > Sorry the clock difference is 10ns now. So the guest clock is off by _10 ns_ > > > > of the host clock. > > > > > > That is pretty good. > > > > Yes. > > > > > > You are suggesting to use getcrosststamp instead, to drop the (rdtsc() - > > > > guest_tsc) part ? > > > > > > Yes, it results in simpler code, doesn't create dependency on the > > > dreaded kvmclock, and is the best we can currently do wrt. precision. > > Even if the PHC sync algorithm manages to detect that the clock read is > incorrect, consider the following: > > Variability in the VM-entry code path, such as cache effects and interrupts would cause > certain readings to be longer then the average (assuming an average > where cache is hot). > > Using the TSC removes this variability, which can be large in case of > non realtime guests, where you do: > > 1. kvm_hypercall. > 2. read host realtime clock. > 3. schedule out qemu-kvm vcpu. > 4. schedule in qemu-kvm vcpu. > > So using the delta between read host realtime and > ->gettime64 increases precision and decreases variability. > > > Sorry, unless i am misunderstanding how this works, it'll get the guest clock > > 2us behind, which is something not wanted. > > > > Miroslav, if ->gettime64 returns the host realtime at 2us in the past, > > this means Chrony will sync the guest clock to > > > > host realtime - 2us > > > > Is that correct? Drop the offset correction and the following happens: Clock offset seems to vary between negative hundreds of ns: 210 Number of sources = 1 MS Name/IP address Stratum Poll Reach LastRx Last sample =============================================================================== #* PHC0 0 3 377 11 -131ns[ -309ns] +/- 3ns And positive: 210 Number of sources = 1 MS Name/IP address Stratum Poll Reach LastRx Last sample =============================================================================== #* PHC0 0 3 377 4 +79ns[ +155ns] +/- 3ns ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-16 20:01 ` Marcelo Tosatti @ 2017-01-17 8:03 ` Miroslav Lichvar 2017-01-17 11:30 ` Marcelo Tosatti 0 siblings, 1 reply; 54+ messages in thread From: Miroslav Lichvar @ 2017-01-17 8:03 UTC (permalink / raw) To: Marcelo Tosatti Cc: Radim Krcmar, kvm, linux-kernel, Paolo Bonzini, Richard Cochran On Mon, Jan 16, 2017 at 06:01:14PM -0200, Marcelo Tosatti wrote: > On Mon, Jan 16, 2017 at 05:47:15PM -0200, Marcelo Tosatti wrote: > > On Mon, Jan 16, 2017 at 05:36:55PM -0200, Marcelo Tosatti wrote: > > > Sorry, unless i am misunderstanding how this works, it'll get the guest clock > > > 2us behind, which is something not wanted. > > > > > > Miroslav, if ->gettime64 returns the host realtime at 2us in the past, > > > this means Chrony will sync the guest clock to > > > > > > host realtime - 2us > > > > > > Is that correct? Probably. It depends on the error of both host and guest timestamps. If the error is the same on both sides, it will cancel out. An occasional spike in the delay shouldn't be a problem as the reading will be filtered out, but for best accuracy it's necessary that the host's timestamp is taken in the middle between the guest's timestamps. Users of the PTP_SYS_OFFSET ioctl assume that (ts[0]+ts[2])/2 corresponds to ts[1], (ts[2]+ts[4])/2 corresponds to ts[3], and so on. ts[1] ts[3] Host time ---------+---------+........ | | | | Guest time ----+---------+---------+...... ts[0] ts[2] ts[4] -- Miroslav Lichvar ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-17 8:03 ` Miroslav Lichvar @ 2017-01-17 11:30 ` Marcelo Tosatti 2017-01-17 15:36 ` Radim Krcmar 0 siblings, 1 reply; 54+ messages in thread From: Marcelo Tosatti @ 2017-01-17 11:30 UTC (permalink / raw) To: Miroslav Lichvar Cc: Radim Krcmar, kvm, linux-kernel, Paolo Bonzini, Richard Cochran On Tue, Jan 17, 2017 at 09:03:27AM +0100, Miroslav Lichvar wrote: > On Mon, Jan 16, 2017 at 06:01:14PM -0200, Marcelo Tosatti wrote: > > On Mon, Jan 16, 2017 at 05:47:15PM -0200, Marcelo Tosatti wrote: > > > On Mon, Jan 16, 2017 at 05:36:55PM -0200, Marcelo Tosatti wrote: > > > > Sorry, unless i am misunderstanding how this works, it'll get the guest clock > > > > 2us behind, which is something not wanted. > > > > > > > > Miroslav, if ->gettime64 returns the host realtime at 2us in the past, > > > > this means Chrony will sync the guest clock to > > > > > > > > host realtime - 2us > > > > > > > > Is that correct? > > Probably. It depends on the error of both host and guest timestamps. > If the error is the same on both sides, it will cancel out. An > occasional spike in the delay shouldn't be a problem as the reading > will be filtered out, but for best accuracy it's necessary that the > host's timestamp is taken in the middle between the guest's > timestamps. The problem is that spikes can be far from occasional: it depends on activity of the host CPU and interrupts. Whose delay can be "intermittent": as long as interrupts are being sent to the host CPU, for example, the delay will be high (which can last minutes). The TSC reading in the guest KVM PTP driver corrects for that delay. > Users of the PTP_SYS_OFFSET ioctl assume that (ts[0]+ts[2])/2 > corresponds to ts[1], (ts[2]+ts[4])/2 corresponds to ts[3], and so on. > > ts[1] ts[3] > Host time ---------+---------+........ > | | > | | > Guest time ----+---------+---------+...... > ts[0] ts[2] ts[4] > > -- > Miroslav Lichvar ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-17 11:30 ` Marcelo Tosatti @ 2017-01-17 15:36 ` Radim Krcmar 2017-01-18 12:17 ` Marcelo Tosatti 0 siblings, 1 reply; 54+ messages in thread From: Radim Krcmar @ 2017-01-17 15:36 UTC (permalink / raw) To: Marcelo Tosatti Cc: Miroslav Lichvar, kvm, linux-kernel, Paolo Bonzini, Richard Cochran 2017-01-17 09:30-0200, Marcelo Tosatti: > On Tue, Jan 17, 2017 at 09:03:27AM +0100, Miroslav Lichvar wrote: >> On Mon, Jan 16, 2017 at 06:01:14PM -0200, Marcelo Tosatti wrote: >> > On Mon, Jan 16, 2017 at 05:47:15PM -0200, Marcelo Tosatti wrote: >> > > On Mon, Jan 16, 2017 at 05:36:55PM -0200, Marcelo Tosatti wrote: >> > > > Sorry, unless i am misunderstanding how this works, it'll get the guest clock >> > > > 2us behind, which is something not wanted. >> > > > >> > > > Miroslav, if ->gettime64 returns the host realtime at 2us in the past, >> > > > this means Chrony will sync the guest clock to >> > > > >> > > > host realtime - 2us >> > > > >> > > > Is that correct? >> >> Probably. It depends on the error of both host and guest timestamps. >> If the error is the same on both sides, it will cancel out. An >> occasional spike in the delay shouldn't be a problem as the reading >> will be filtered out, but for best accuracy it's necessary that the >> host's timestamp is taken in the middle between the guest's >> timestamps. > > The problem is that spikes can be far from occasional: it depends on activity of > the host CPU and interrupts. Whose delay can be "intermittent": as long > as interrupts are being sent to the host CPU, for example, the delay > will be high (which can last minutes). > > The TSC reading in the guest KVM PTP driver corrects for that delay. > >> Users of the PTP_SYS_OFFSET ioctl assume that (ts[0]+ts[2])/2 >> corresponds to ts[1], (ts[2]+ts[4])/2 corresponds to ts[3], and so on. >> >> ts[1] ts[3] >> Host time ---------+---------+........ >> | | >> | | >> Guest time ----+---------+---------+...... >> ts[0] ts[2] ts[4] KVM PTP delay moves host ts[i] to be close to guest ts[i+1] and makes the offset very consistent, so the graph would look like: ts[1] ts[3] Host time -------------+---------+........ | | | | Guest time ----+---------+---------+...... ts[0] ts[2] ts[4] which doesn't sound good if users assume that the host reading is in the middle -- the guest time would be ahead of the host time. I'm wondering why is the PTP precision around 10ns, when the hypercall takes around 2-3k cycles. Have you measured the guest<->host offset by getting the output of the hypercall, i.e. {host_sec @ tsc, host_nsec @ tsc, tsc} and comparing it with guest time computed from the same tsc, i.e. {guest_sec @ tsc, guest_nsec @ tsc} ? Thanks. ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-17 15:36 ` Radim Krcmar @ 2017-01-18 12:17 ` Marcelo Tosatti 2017-01-18 12:24 ` Marcelo Tosatti 0 siblings, 1 reply; 54+ messages in thread From: Marcelo Tosatti @ 2017-01-18 12:17 UTC (permalink / raw) To: Radim Krcmar Cc: Miroslav Lichvar, kvm, linux-kernel, Paolo Bonzini, Richard Cochran On Tue, Jan 17, 2017 at 04:36:21PM +0100, Radim Krcmar wrote: > 2017-01-17 09:30-0200, Marcelo Tosatti: > > On Tue, Jan 17, 2017 at 09:03:27AM +0100, Miroslav Lichvar wrote: > >> On Mon, Jan 16, 2017 at 06:01:14PM -0200, Marcelo Tosatti wrote: > >> > On Mon, Jan 16, 2017 at 05:47:15PM -0200, Marcelo Tosatti wrote: > >> > > On Mon, Jan 16, 2017 at 05:36:55PM -0200, Marcelo Tosatti wrote: > >> > > > Sorry, unless i am misunderstanding how this works, it'll get the guest clock > >> > > > 2us behind, which is something not wanted. > >> > > > > >> > > > Miroslav, if ->gettime64 returns the host realtime at 2us in the past, > >> > > > this means Chrony will sync the guest clock to > >> > > > > >> > > > host realtime - 2us > >> > > > > >> > > > Is that correct? > >> > >> Probably. It depends on the error of both host and guest timestamps. > >> If the error is the same on both sides, it will cancel out. An > >> occasional spike in the delay shouldn't be a problem as the reading > >> will be filtered out, but for best accuracy it's necessary that the > >> host's timestamp is taken in the middle between the guest's > >> timestamps. > > > > The problem is that spikes can be far from occasional: it depends on activity of > > the host CPU and interrupts. Whose delay can be "intermittent": as long > > as interrupts are being sent to the host CPU, for example, the delay > > will be high (which can last minutes). > > > > The TSC reading in the guest KVM PTP driver corrects for that delay. > > > >> Users of the PTP_SYS_OFFSET ioctl assume that (ts[0]+ts[2])/2 > >> corresponds to ts[1], (ts[2]+ts[4])/2 corresponds to ts[3], and so on. > >> > >> ts[1] ts[3] > >> Host time ---------+---------+........ > >> | | > >> | | > >> Guest time ----+---------+---------+...... > >> ts[0] ts[2] ts[4] > > KVM PTP delay moves host ts[i] to be close to guest ts[i+1] and makes > the offset very consistent, so the graph would look like: > > ts[1] ts[3] > Host time -------------+---------+........ > | | > | | > Guest time ----+---------+---------+...... > ts[0] ts[2] ts[4] > > which doesn't sound good if users assume that the host reading is in the > middle -- the guest time would be ahead of the host time. > > I'm wondering why is the PTP precision around 10ns, when the hypercall > takes around 2-3k cycles. Have you measured the guest<->host offset by > getting the output of the hypercall, i.e. > {host_sec @ tsc, host_nsec @ tsc, tsc} > and comparing it with guest time computed from the same tsc, i.e. > {guest_sec @ tsc, guest_nsec @ tsc} > ? > > Thanks. Testcase: run a guest and a loop sending SIGUSR1 to vcpu0 (emulating intense interrupts). Follows results: Without TSC delta calculation: ============================= #* PHC0 0 3 377 2 -99ns[ +206ns] +/- 116ns #* PHC0 0 3 377 8 +202ns[ +249ns] +/- 111ns #* PHC0 0 3 377 8 -213ns[ +683ns] +/- 88ns #* PHC0 0 3 377 6 +77ns[ +319ns] +/- 56ns #* PHC0 0 3 377 4 -771ns[-1029ns] +/- 93ns #* PHC0 0 3 377 10 -49ns[ -58ns] +/- 121ns #* PHC0 0 3 377 9 +562ns[ +703ns] +/- 107ns #* PHC0 0 3 377 6 -2ns[ -3ns] +/- 94ns #* PHC0 0 3 377 4 +451ns[ +494ns] +/- 138ns #* PHC0 0 3 377 11 -67ns[ -74ns] +/- 113ns #* PHC0 0 3 377 8 +244ns[ +264ns] +/- 119ns #* PHC0 0 3 377 7 -696ns[ -890ns] +/- 89ns #* PHC0 0 3 377 4 +468ns[ +560ns] +/- 110ns #* PHC0 0 3 377 11 -310ns[ -430ns] +/- 72ns #* PHC0 0 3 377 9 +189ns[ +298ns] +/- 54ns #* PHC0 0 3 377 7 +594ns[ +473ns] +/- 96ns #* PHC0 0 3 377 5 +151ns[ +280ns] +/- 71ns #* PHC0 0 3 377 10 -590ns[ -696ns] +/- 94ns #* PHC0 0 3 377 8 +415ns[ +526ns] +/- 74ns #* PHC0 0 3 377 6 +1381ns[+1469ns] +/- 101ns #* PHC0 0 3 377 4 +571ns[+1304ns] +/- 54ns #* PHC0 0 3 377 8 -5ns[ +71ns] +/- 139ns #* PHC0 0 3 377 7 -247ns[ -502ns] +/- 69ns #* PHC0 0 3 377 5 -283ns[ +879ns] +/- 73ns #* PHC0 0 3 377 3 +148ns[ -109ns] +/- 61ns With TSC delta calculation: ============================ #* PHC0 0 3 377 7 +379ns[ +432ns] +/- 53ns #* PHC0 0 3 377 9 +106ns[ +420ns] +/- 42ns #* PHC0 0 3 377 7 -58ns[ -136ns] +/- 62ns #* PHC0 0 3 377 12 +93ns[ -38ns] +/- 64ns #* PHC0 0 3 377 8 +84ns[ +107ns] +/- 69ns #* PHC0 0 3 377 3 -76ns[ -103ns] +/- 52ns #* PHC0 0 3 377 7 +52ns[ +63ns] +/- 50ns #* PHC0 0 3 377 11 +29ns[ +31ns] +/- 70ns #* PHC0 0 3 377 7 -47ns[ -56ns] +/- 42ns #* PHC0 0 3 377 10 -35ns[ -42ns] +/- 33ns #* PHC0 0 3 377 7 -32ns[ -34ns] +/- 42ns #* PHC0 0 3 377 11 -172ns[ -173ns] +/- 118ns #* PHC0 0 3 377 6 +65ns[ +76ns] +/- 23ns #* PHC0 0 3 377 9 +18ns[ +23ns] +/- 37ns #* PHC0 0 3 377 6 +41ns[ -60ns] +/- 30ns #* PHC0 0 3 377 10 +39ns[ +183ns] +/- 42ns #* PHC0 0 3 377 6 +50ns[ +102ns] +/- 86ns #* PHC0 0 3 377 11 +50ns[ +75ns] +/- 52ns #* PHC0 0 3 377 6 +50ns[ +116ns] +/- 100ns #* PHC0 0 3 377 10 +46ns[ +65ns] +/- 79ns #* PHC0 0 3 377 7 -38ns[ -51ns] +/- 29ns #* PHC0 0 3 377 10 -11ns[ -12ns] +/- 32ns #* PHC0 0 3 377 7 -31ns[ -32ns] +/- 99ns #* PHC0 0 3 377 10 +222ns[ +238ns] +/- 58ns #* PHC0 0 3 377 6 +185ns[ +207ns] +/- 39ns #* PHC0 0 3 377 10 -392ns[ -394ns] +/- 118ns #* PHC0 0 3 377 6 -9ns[ -50ns] +/- 35ns #* PHC0 0 3 377 10 -346ns[ -355ns] +/- 111ns Do you still want to drop it in favour of simplicity? ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-18 12:17 ` Marcelo Tosatti @ 2017-01-18 12:24 ` Marcelo Tosatti 2017-01-18 12:46 ` Paolo Bonzini 0 siblings, 1 reply; 54+ messages in thread From: Marcelo Tosatti @ 2017-01-18 12:24 UTC (permalink / raw) To: Radim Krcmar Cc: Miroslav Lichvar, kvm, linux-kernel, Paolo Bonzini, Richard Cochran On Wed, Jan 18, 2017 at 10:17:38AM -0200, Marcelo Tosatti wrote: > On Tue, Jan 17, 2017 at 04:36:21PM +0100, Radim Krcmar wrote: > > 2017-01-17 09:30-0200, Marcelo Tosatti: > > > On Tue, Jan 17, 2017 at 09:03:27AM +0100, Miroslav Lichvar wrote: > > >> On Mon, Jan 16, 2017 at 06:01:14PM -0200, Marcelo Tosatti wrote: > > >> > On Mon, Jan 16, 2017 at 05:47:15PM -0200, Marcelo Tosatti wrote: > > >> > > On Mon, Jan 16, 2017 at 05:36:55PM -0200, Marcelo Tosatti wrote: > > >> > > > Sorry, unless i am misunderstanding how this works, it'll get the guest clock > > >> > > > 2us behind, which is something not wanted. > > >> > > > > > >> > > > Miroslav, if ->gettime64 returns the host realtime at 2us in the past, > > >> > > > this means Chrony will sync the guest clock to > > >> > > > > > >> > > > host realtime - 2us > > >> > > > > > >> > > > Is that correct? > > >> > > >> Probably. It depends on the error of both host and guest timestamps. > > >> If the error is the same on both sides, it will cancel out. An > > >> occasional spike in the delay shouldn't be a problem as the reading > > >> will be filtered out, but for best accuracy it's necessary that the > > >> host's timestamp is taken in the middle between the guest's > > >> timestamps. > > > > > > The problem is that spikes can be far from occasional: it depends on activity of > > > the host CPU and interrupts. Whose delay can be "intermittent": as long > > > as interrupts are being sent to the host CPU, for example, the delay > > > will be high (which can last minutes). > > > > > > The TSC reading in the guest KVM PTP driver corrects for that delay. > > > > > >> Users of the PTP_SYS_OFFSET ioctl assume that (ts[0]+ts[2])/2 > > >> corresponds to ts[1], (ts[2]+ts[4])/2 corresponds to ts[3], and so on. > > >> > > >> ts[1] ts[3] > > >> Host time ---------+---------+........ > > >> | | > > >> | | > > >> Guest time ----+---------+---------+...... > > >> ts[0] ts[2] ts[4] > > > > KVM PTP delay moves host ts[i] to be close to guest ts[i+1] and makes > > the offset very consistent, so the graph would look like: > > > > ts[1] ts[3] > > Host time -------------+---------+........ > > | | > > | | > > Guest time ----+---------+---------+...... > > ts[0] ts[2] ts[4] > > > > which doesn't sound good if users assume that the host reading is in the > > middle -- the guest time would be ahead of the host time. > > > > I'm wondering why is the PTP precision around 10ns, when the hypercall > > takes around 2-3k cycles. Have you measured the guest<->host offset by > > getting the output of the hypercall, i.e. > > {host_sec @ tsc, host_nsec @ tsc, tsc} > > and comparing it with guest time computed from the same tsc, i.e. > > {guest_sec @ tsc, guest_nsec @ tsc} > > ? > > > > Thanks. > > Testcase: run a guest and a loop sending SIGUSR1 to vcpu0 (emulating > intense interrupts). Follows results: > > Without TSC delta calculation: > ============================= > > #* PHC0 0 3 377 2 -99ns[ +206ns] +/- 116ns > #* PHC0 0 3 377 8 +202ns[ +249ns] +/- 111ns > #* PHC0 0 3 377 8 -213ns[ +683ns] +/- 88ns > #* PHC0 0 3 377 6 +77ns[ +319ns] +/- 56ns > #* PHC0 0 3 377 4 -771ns[-1029ns] +/- 93ns > #* PHC0 0 3 377 10 -49ns[ -58ns] +/- 121ns > #* PHC0 0 3 377 9 +562ns[ +703ns] +/- 107ns > #* PHC0 0 3 377 6 -2ns[ -3ns] +/- 94ns > #* PHC0 0 3 377 4 +451ns[ +494ns] +/- 138ns > #* PHC0 0 3 377 11 -67ns[ -74ns] +/- 113ns > #* PHC0 0 3 377 8 +244ns[ +264ns] +/- 119ns > #* PHC0 0 3 377 7 -696ns[ -890ns] +/- 89ns > #* PHC0 0 3 377 4 +468ns[ +560ns] +/- 110ns > #* PHC0 0 3 377 11 -310ns[ -430ns] +/- 72ns > #* PHC0 0 3 377 9 +189ns[ +298ns] +/- 54ns > #* PHC0 0 3 377 7 +594ns[ +473ns] +/- 96ns > #* PHC0 0 3 377 5 +151ns[ +280ns] +/- 71ns > #* PHC0 0 3 377 10 -590ns[ -696ns] +/- 94ns > #* PHC0 0 3 377 8 +415ns[ +526ns] +/- 74ns > #* PHC0 0 3 377 6 +1381ns[+1469ns] +/- 101ns > #* PHC0 0 3 377 4 +571ns[+1304ns] +/- 54ns > #* PHC0 0 3 377 8 -5ns[ +71ns] +/- 139ns > #* PHC0 0 3 377 7 -247ns[ -502ns] +/- 69ns > #* PHC0 0 3 377 5 -283ns[ +879ns] +/- 73ns > #* PHC0 0 3 377 3 +148ns[ -109ns] +/- 61ns > > With TSC delta calculation: > ============================ > > #* PHC0 0 3 377 7 +379ns[ +432ns] +/- 53ns > #* PHC0 0 3 377 9 +106ns[ +420ns] +/- 42ns > #* PHC0 0 3 377 7 -58ns[ -136ns] +/- 62ns > #* PHC0 0 3 377 12 +93ns[ -38ns] +/- 64ns > #* PHC0 0 3 377 8 +84ns[ +107ns] +/- 69ns > #* PHC0 0 3 377 3 -76ns[ -103ns] +/- 52ns > #* PHC0 0 3 377 7 +52ns[ +63ns] +/- 50ns > #* PHC0 0 3 377 11 +29ns[ +31ns] +/- 70ns > #* PHC0 0 3 377 7 -47ns[ -56ns] +/- 42ns > #* PHC0 0 3 377 10 -35ns[ -42ns] +/- 33ns > #* PHC0 0 3 377 7 -32ns[ -34ns] +/- 42ns > #* PHC0 0 3 377 11 -172ns[ -173ns] +/- 118ns > #* PHC0 0 3 377 6 +65ns[ +76ns] +/- 23ns > #* PHC0 0 3 377 9 +18ns[ +23ns] +/- 37ns > #* PHC0 0 3 377 6 +41ns[ -60ns] +/- 30ns > #* PHC0 0 3 377 10 +39ns[ +183ns] +/- 42ns > #* PHC0 0 3 377 6 +50ns[ +102ns] +/- 86ns > #* PHC0 0 3 377 11 +50ns[ +75ns] +/- 52ns > #* PHC0 0 3 377 6 +50ns[ +116ns] +/- 100ns > #* PHC0 0 3 377 10 +46ns[ +65ns] +/- 79ns > #* PHC0 0 3 377 7 -38ns[ -51ns] +/- 29ns > #* PHC0 0 3 377 10 -11ns[ -12ns] +/- 32ns > #* PHC0 0 3 377 7 -31ns[ -32ns] +/- 99ns > #* PHC0 0 3 377 10 +222ns[ +238ns] +/- 58ns > #* PHC0 0 3 377 6 +185ns[ +207ns] +/- 39ns > #* PHC0 0 3 377 10 -392ns[ -394ns] +/- 118ns > #* PHC0 0 3 377 6 -9ns[ -50ns] +/- 35ns > #* PHC0 0 3 377 10 -346ns[ -355ns] +/- 111ns > > > Do you still want to drop it in favour of simplicity? > This is the output of "chronyc sources". See section "Time sources" of https://chrony.tuxfamily.org/doc/2.4/chronyc.html. ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-18 12:24 ` Marcelo Tosatti @ 2017-01-18 12:46 ` Paolo Bonzini 2017-01-18 13:36 ` Miroslav Lichvar ` (3 more replies) 0 siblings, 4 replies; 54+ messages in thread From: Paolo Bonzini @ 2017-01-18 12:46 UTC (permalink / raw) To: Marcelo Tosatti, Radim Krcmar Cc: Miroslav Lichvar, kvm, linux-kernel, Richard Cochran On 18/01/2017 13:24, Marcelo Tosatti wrote: > On Wed, Jan 18, 2017 at 10:17:38AM -0200, Marcelo Tosatti wrote: >> On Tue, Jan 17, 2017 at 04:36:21PM +0100, Radim Krcmar wrote: >>> 2017-01-17 09:30-0200, Marcelo Tosatti: >>>> On Tue, Jan 17, 2017 at 09:03:27AM +0100, Miroslav Lichvar wrote: >>>>> Users of the PTP_SYS_OFFSET ioctl assume that (ts[0]+ts[2])/2 >>>>> corresponds to ts[1], (ts[2]+ts[4])/2 corresponds to ts[3], and so on. >>>>> >>>>> ts[1] ts[3] >>>>> Host time ---------+---------+........ >>>>> | | >>>>> | | >>>>> Guest time ----+---------+---------+...... >>>>> ts[0] ts[2] ts[4] >>> >>> KVM PTP delay moves host ts[i] to be close to guest ts[i+1] and makes >>> the offset very consistent, so the graph would look like: >>> >>> ts[1] ts[3] >>> Host time -------------+---------+........ >>> | | >>> | | >>> Guest time ----+---------+---------+...... >>> ts[0] ts[2] ts[4] >>> >>> which doesn't sound good if users assume that the host reading is in the >>> middle -- the guest time would be ahead of the host time. >> >> Testcase: run a guest and a loop sending SIGUSR1 to vcpu0 (emulating >> intense interrupts). Follows results: >> >> Without TSC delta calculation: >> ============================= >> >> #* PHC0 0 3 377 2 -99ns[ +206ns] +/- 116ns >> #* PHC0 0 3 377 8 +202ns[ +249ns] +/- 111ns >> #* PHC0 0 3 377 8 -213ns[ +683ns] +/- 88ns >> #* PHC0 0 3 377 6 +77ns[ +319ns] +/- 56ns >> #* PHC0 0 3 377 4 -771ns[-1029ns] +/- 93ns >> #* PHC0 0 3 377 10 -49ns[ -58ns] +/- 121ns >> #* PHC0 0 3 377 9 +562ns[ +703ns] +/- 107ns >> #* PHC0 0 3 377 6 -2ns[ -3ns] +/- 94ns >> #* PHC0 0 3 377 4 +451ns[ +494ns] +/- 138ns >> #* PHC0 0 3 377 11 -67ns[ -74ns] +/- 113ns >> #* PHC0 0 3 377 8 +244ns[ +264ns] +/- 119ns >> #* PHC0 0 3 377 7 -696ns[ -890ns] +/- 89ns >> #* PHC0 0 3 377 4 +468ns[ +560ns] +/- 110ns >> #* PHC0 0 3 377 11 -310ns[ -430ns] +/- 72ns >> #* PHC0 0 3 377 9 +189ns[ +298ns] +/- 54ns >> #* PHC0 0 3 377 7 +594ns[ +473ns] +/- 96ns >> #* PHC0 0 3 377 5 +151ns[ +280ns] +/- 71ns >> #* PHC0 0 3 377 10 -590ns[ -696ns] +/- 94ns >> #* PHC0 0 3 377 8 +415ns[ +526ns] +/- 74ns >> #* PHC0 0 3 377 6 +1381ns[+1469ns] +/- 101ns >> #* PHC0 0 3 377 4 +571ns[+1304ns] +/- 54ns >> #* PHC0 0 3 377 8 -5ns[ +71ns] +/- 139ns >> #* PHC0 0 3 377 7 -247ns[ -502ns] +/- 69ns >> #* PHC0 0 3 377 5 -283ns[ +879ns] +/- 73ns >> #* PHC0 0 3 377 3 +148ns[ -109ns] +/- 61ns >> >> With TSC delta calculation: >> ============================ >> >> #* PHC0 0 3 377 7 +379ns[ +432ns] +/- 53ns >> #* PHC0 0 3 377 9 +106ns[ +420ns] +/- 42ns >> #* PHC0 0 3 377 7 -58ns[ -136ns] +/- 62ns >> #* PHC0 0 3 377 12 +93ns[ -38ns] +/- 64ns >> #* PHC0 0 3 377 8 +84ns[ +107ns] +/- 69ns >> #* PHC0 0 3 377 3 -76ns[ -103ns] +/- 52ns >> #* PHC0 0 3 377 7 +52ns[ +63ns] +/- 50ns >> #* PHC0 0 3 377 11 +29ns[ +31ns] +/- 70ns >> #* PHC0 0 3 377 7 -47ns[ -56ns] +/- 42ns >> #* PHC0 0 3 377 10 -35ns[ -42ns] +/- 33ns >> #* PHC0 0 3 377 7 -32ns[ -34ns] +/- 42ns >> #* PHC0 0 3 377 11 -172ns[ -173ns] +/- 118ns >> #* PHC0 0 3 377 6 +65ns[ +76ns] +/- 23ns >> #* PHC0 0 3 377 9 +18ns[ +23ns] +/- 37ns >> #* PHC0 0 3 377 6 +41ns[ -60ns] +/- 30ns >> #* PHC0 0 3 377 10 +39ns[ +183ns] +/- 42ns >> #* PHC0 0 3 377 6 +50ns[ +102ns] +/- 86ns >> #* PHC0 0 3 377 11 +50ns[ +75ns] +/- 52ns >> #* PHC0 0 3 377 6 +50ns[ +116ns] +/- 100ns >> #* PHC0 0 3 377 10 +46ns[ +65ns] +/- 79ns >> #* PHC0 0 3 377 7 -38ns[ -51ns] +/- 29ns >> #* PHC0 0 3 377 10 -11ns[ -12ns] +/- 32ns >> #* PHC0 0 3 377 7 -31ns[ -32ns] +/- 99ns >> #* PHC0 0 3 377 10 +222ns[ +238ns] +/- 58ns >> #* PHC0 0 3 377 6 +185ns[ +207ns] +/- 39ns >> #* PHC0 0 3 377 10 -392ns[ -394ns] +/- 118ns >> #* PHC0 0 3 377 6 -9ns[ -50ns] +/- 35ns >> #* PHC0 0 3 377 10 -346ns[ -355ns] +/- 111ns >> >> >> Do you still want to drop it in favour of simplicity? > > This is the output of "chronyc sources". See section "Time sources" > of https://chrony.tuxfamily.org/doc/2.4/chronyc.html. It's just that it's not obvious why you get better results with biased host timestamps. What makes the biased host timestamp more precise? I'd rather use PTP_SYS_OFFSET_PRECISE instead, but unfortunately chrony does not support it---but I would still prefer you to support PTP_SYS_OFFSET_PRECISE as well. Paolo ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-18 12:46 ` Paolo Bonzini @ 2017-01-18 13:36 ` Miroslav Lichvar 2017-01-18 14:02 ` Paolo Bonzini 2017-01-18 14:24 ` Marcelo Tosatti ` (2 subsequent siblings) 3 siblings, 1 reply; 54+ messages in thread From: Miroslav Lichvar @ 2017-01-18 13:36 UTC (permalink / raw) To: Paolo Bonzini Cc: Marcelo Tosatti, Radim Krcmar, kvm, linux-kernel, Richard Cochran On Wed, Jan 18, 2017 at 01:46:58PM +0100, Paolo Bonzini wrote: > On 18/01/2017 13:24, Marcelo Tosatti wrote: > >> Testcase: run a guest and a loop sending SIGUSR1 to vcpu0 (emulating > >> intense interrupts). Follows results: > >> Do you still want to drop it in favour of simplicity? > It's just that it's not obvious why you get better results with biased > host timestamps. What makes the biased host timestamp more precise? > > I'd rather use PTP_SYS_OFFSET_PRECISE instead, but unfortunately chrony > does not support it---but I would still prefer you to support > PTP_SYS_OFFSET_PRECISE as well. Interesting. I wasn't aware that there is a new ioctl for measuring the HW-sys offset. Adding support to chrony shouldn't be difficult. If I understand it correctly, PTP_SYS_OFFSET can be emulated on top of PTP_SYS_OFFSET_PRECISE simply by copying the sys_realtime and device fields to corresponding ts slots. The apparent delay will be zero, but that's ok if the conversion is really accurate. I'm not sure if trying to do that in the opposite direction is a good idea. An application using PTP_SYS_OFFSET_PRECISE may assume the conversion is accurate and not include any delay/dispersion in an estimate of the maximum error, which is needed in NTP for instance. If we know the host timestamp ts[1] is not in the middle between the guests timestamps ts[0] and ts[2], but rather closer to ts[2], why not simply shift ts[1] by (ts[2]-ts[0])/2 ? -- Miroslav Lichvar ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-18 13:36 ` Miroslav Lichvar @ 2017-01-18 14:02 ` Paolo Bonzini 2017-01-18 14:50 ` Marcelo Tosatti 0 siblings, 1 reply; 54+ messages in thread From: Paolo Bonzini @ 2017-01-18 14:02 UTC (permalink / raw) To: Miroslav Lichvar Cc: Marcelo Tosatti, Radim Krcmar, kvm, linux-kernel, Richard Cochran On 18/01/2017 14:36, Miroslav Lichvar wrote: > On Wed, Jan 18, 2017 at 01:46:58PM +0100, Paolo Bonzini wrote: >> On 18/01/2017 13:24, Marcelo Tosatti wrote: >>>> Testcase: run a guest and a loop sending SIGUSR1 to vcpu0 (emulating >>>> intense interrupts). Follows results: > >>>> Do you still want to drop it in favour of simplicity? > >> It's just that it's not obvious why you get better results with biased >> host timestamps. What makes the biased host timestamp more precise? >> >> I'd rather use PTP_SYS_OFFSET_PRECISE instead, but unfortunately chrony >> does not support it---but I would still prefer you to support >> PTP_SYS_OFFSET_PRECISE as well. > > Interesting. I wasn't aware that there is a new ioctl for measuring > the HW-sys offset. Adding support to chrony shouldn't be difficult. > > If I understand it correctly, PTP_SYS_OFFSET can be emulated on top of > PTP_SYS_OFFSET_PRECISE simply by copying the sys_realtime and device > fields to corresponding ts slots. The apparent delay will be zero, but > that's ok if the conversion is really accurate. Yes, for 1 sample only. Otherwise you'd have the same issue as in Marcelo's driver (the device aka guest timestamp from PTP_SYS_OFFSET_PRECISE would not be halfway between the system aka host timestamps), and your idea below could be applied. > I'm not sure if trying to do that in the opposite direction is a good > idea. An application using PTP_SYS_OFFSET_PRECISE may assume the > conversion is accurate and not include any delay/dispersion in an > estimate of the maximum error, which is needed in NTP for instance. > > If we know the host timestamp ts[1] is not in the middle between the > guests timestamps ts[0] and ts[2], but rather closer to ts[2], why not > simply shift ts[1] by (ts[2]-ts[0])/2 ? Interesting idea! For this to work, KVM needs to implement getcrosstimestamp and ptp_chardev.c can then add an alternative implementation of PTP_SYS_OFFSET, based on precise cross timestamps. Something like for (i = 0; i <= sysoff->n_samples; i++) { // ... call getcrosststamp ... sysns = ktime_to_ns(xtstamp.sys_realtime); if (i > 0) { devns = ktime_to_ns(xtstamp.device); devns -= (sysns - prev_sysns) / 2; devts = ns_to_timespec(devns); pct->sec = devts.tv_sec; pct->nsec = devts.tv_nsec; pct++; } systs = ns_to_timespec(sysns); pct->sec = ts.tv_sec; pct->nsec = ts.tv_nsec; pct++; prev_sysns = sysns; } Marcelo, can you give it a try? Thanks, Paolo ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-18 14:02 ` Paolo Bonzini @ 2017-01-18 14:50 ` Marcelo Tosatti 2017-01-18 15:35 ` Radim Krcmar 2017-01-18 15:45 ` Paolo Bonzini 0 siblings, 2 replies; 54+ messages in thread From: Marcelo Tosatti @ 2017-01-18 14:50 UTC (permalink / raw) To: Paolo Bonzini Cc: Miroslav Lichvar, Radim Krcmar, kvm, linux-kernel, Richard Cochran On Wed, Jan 18, 2017 at 03:02:23PM +0100, Paolo Bonzini wrote: > > > On 18/01/2017 14:36, Miroslav Lichvar wrote: > > On Wed, Jan 18, 2017 at 01:46:58PM +0100, Paolo Bonzini wrote: > >> On 18/01/2017 13:24, Marcelo Tosatti wrote: > >>>> Testcase: run a guest and a loop sending SIGUSR1 to vcpu0 (emulating > >>>> intense interrupts). Follows results: > > > >>>> Do you still want to drop it in favour of simplicity? > > > >> It's just that it's not obvious why you get better results with biased > >> host timestamps. What makes the biased host timestamp more precise? > >> > >> I'd rather use PTP_SYS_OFFSET_PRECISE instead, but unfortunately chrony > >> does not support it---but I would still prefer you to support > >> PTP_SYS_OFFSET_PRECISE as well. > > > > Interesting. I wasn't aware that there is a new ioctl for measuring > > the HW-sys offset. Adding support to chrony shouldn't be difficult. > > > > If I understand it correctly, PTP_SYS_OFFSET can be emulated on top of > > PTP_SYS_OFFSET_PRECISE simply by copying the sys_realtime and device > > fields to corresponding ts slots. The apparent delay will be zero, but > > that's ok if the conversion is really accurate. > > Yes, for 1 sample only. Otherwise you'd have the same issue as in > Marcelo's driver (the device aka guest timestamp from > PTP_SYS_OFFSET_PRECISE would not be halfway between the system aka host > timestamps), and your idea below could be applied. > > > I'm not sure if trying to do that in the opposite direction is a good > > idea. An application using PTP_SYS_OFFSET_PRECISE may assume the > > conversion is accurate and not include any delay/dispersion in an > > estimate of the maximum error, which is needed in NTP for instance. > > > > If we know the host timestamp ts[1] is not in the middle between the > > guests timestamps ts[0] and ts[2], but rather closer to ts[2], why not > > simply shift ts[1] by (ts[2]-ts[0])/2 ? > > Interesting idea! For this to work, KVM needs to implement > getcrosstimestamp and ptp_chardev.c can then add an alternative > implementation of PTP_SYS_OFFSET, based on precise cross timestamps. > > Something like > > for (i = 0; i <= sysoff->n_samples; i++) { > // ... call getcrosststamp ... > sysns = ktime_to_ns(xtstamp.sys_realtime); > if (i > 0) { > devns = ktime_to_ns(xtstamp.device); > devns -= (sysns - prev_sysns) / 2; > devts = ns_to_timespec(devns); > pct->sec = devts.tv_sec; > pct->nsec = devts.tv_nsec; > pct++; > } > systs = ns_to_timespec(sysns); > pct->sec = ts.tv_sec; > pct->nsec = ts.tv_nsec; > pct++; > prev_sysns = sysns; > } > > Marcelo, can you give it a try? Can convert fine, but problem is the simultaneous read of host and guest clocks. > Thanks, > > Paolo It seems to me anything else other than using a single TSC read (for both host and guest clocks) is a poor PTP_SYS_OFFSET_PRECISE implementation (because it would claim to be similar to ART, where the timestamps are simultaneous), but not be. ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-18 14:50 ` Marcelo Tosatti @ 2017-01-18 15:35 ` Radim Krcmar 2017-01-18 15:45 ` Paolo Bonzini 1 sibling, 0 replies; 54+ messages in thread From: Radim Krcmar @ 2017-01-18 15:35 UTC (permalink / raw) To: Marcelo Tosatti Cc: Paolo Bonzini, Miroslav Lichvar, kvm, linux-kernel, Richard Cochran 2017-01-18 12:50-0200, Marcelo Tosatti: > On Wed, Jan 18, 2017 at 03:02:23PM +0100, Paolo Bonzini wrote: >> >> >> On 18/01/2017 14:36, Miroslav Lichvar wrote: >> > On Wed, Jan 18, 2017 at 01:46:58PM +0100, Paolo Bonzini wrote: >> >> On 18/01/2017 13:24, Marcelo Tosatti wrote: >> >>>> Testcase: run a guest and a loop sending SIGUSR1 to vcpu0 (emulating >> >>>> intense interrupts). Follows results: >> > >> >>>> Do you still want to drop it in favour of simplicity? >> > >> >> It's just that it's not obvious why you get better results with biased >> >> host timestamps. What makes the biased host timestamp more precise? >> >> >> >> I'd rather use PTP_SYS_OFFSET_PRECISE instead, but unfortunately chrony >> >> does not support it---but I would still prefer you to support >> >> PTP_SYS_OFFSET_PRECISE as well. >> > >> > Interesting. I wasn't aware that there is a new ioctl for measuring >> > the HW-sys offset. Adding support to chrony shouldn't be difficult. >> > >> > If I understand it correctly, PTP_SYS_OFFSET can be emulated on top of >> > PTP_SYS_OFFSET_PRECISE simply by copying the sys_realtime and device >> > fields to corresponding ts slots. The apparent delay will be zero, but >> > that's ok if the conversion is really accurate. >> >> Yes, for 1 sample only. Otherwise you'd have the same issue as in >> Marcelo's driver (the device aka guest timestamp from >> PTP_SYS_OFFSET_PRECISE would not be halfway between the system aka host >> timestamps), and your idea below could be applied. >> >> > I'm not sure if trying to do that in the opposite direction is a good >> > idea. An application using PTP_SYS_OFFSET_PRECISE may assume the >> > conversion is accurate and not include any delay/dispersion in an >> > estimate of the maximum error, which is needed in NTP for instance. >> > >> > If we know the host timestamp ts[1] is not in the middle between the >> > guests timestamps ts[0] and ts[2], but rather closer to ts[2], why not >> > simply shift ts[1] by (ts[2]-ts[0])/2 ? > > > >> >> Interesting idea! For this to work, KVM needs to implement >> getcrosstimestamp and ptp_chardev.c can then add an alternative >> implementation of PTP_SYS_OFFSET, based on precise cross timestamps. >> >> Something like >> >> for (i = 0; i <= sysoff->n_samples; i++) { >> // ... call getcrosststamp ... >> sysns = ktime_to_ns(xtstamp.sys_realtime); >> if (i > 0) { >> devns = ktime_to_ns(xtstamp.device); >> devns -= (sysns - prev_sysns) / 2; >> devts = ns_to_timespec(devns); >> pct->sec = devts.tv_sec; >> pct->nsec = devts.tv_nsec; >> pct++; >> } >> systs = ns_to_timespec(sysns); >> pct->sec = ts.tv_sec; >> pct->nsec = ts.tv_nsec; >> pct++; >> prev_sysns = sysns; >> } Nice. PTP_SYS_OFFSET seems to be mandatory, so this hunk could be upstreamable as a fallback for PTP devices that have getcrosststamp and not gettime64. KVM PTP would be the only driver using it so far. >> Marcelo, can you give it a try? > > Can convert fine, but problem is the simultaneous read > of host and guest clocks. > >> Thanks, >> >> Paolo > > It seems to me anything else other than using a single TSC read > (for both host and guest clocks) is a poor PTP_SYS_OFFSET_PRECISE > implementation (because it would claim to be similar to ART, where > the timestamps are simultaneous), but not be. Yes, the guest should use the TSC returned by the hypercall to compute the corresponding guest time, which will allow us to know the host<->guest offset. Is this impossible to do with the current API? Thanks. ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-18 14:50 ` Marcelo Tosatti 2017-01-18 15:35 ` Radim Krcmar @ 2017-01-18 15:45 ` Paolo Bonzini 2017-01-18 15:57 ` Marcelo Tosatti 1 sibling, 1 reply; 54+ messages in thread From: Paolo Bonzini @ 2017-01-18 15:45 UTC (permalink / raw) To: Marcelo Tosatti Cc: Miroslav Lichvar, Radim Krcmar, kvm, linux-kernel, Richard Cochran On 18/01/2017 15:50, Marcelo Tosatti wrote: >> Interesting idea! For this to work, KVM needs to implement >> getcrosstimestamp and ptp_chardev.c can then add an alternative >> implementation of PTP_SYS_OFFSET, based on precise cross timestamps. >> >> Something like >> >> for (i = 0; i <= sysoff->n_samples; i++) { >> // ... call getcrosststamp ... >> sysns = ktime_to_ns(xtstamp.sys_realtime); >> if (i > 0) { >> devns = ktime_to_ns(xtstamp.device); >> devns -= (sysns - prev_sysns) / 2; >> devts = ns_to_timespec(devns); >> pct->sec = devts.tv_sec; >> pct->nsec = devts.tv_nsec; >> pct++; >> } >> systs = ns_to_timespec(sysns); >> pct->sec = ts.tv_sec; >> pct->nsec = ts.tv_nsec; >> pct++; >> prev_sysns = sysns; >> } >> >> Marcelo, can you give it a try? > > Can convert fine, but problem is the simultaneous read > of host and guest clocks. Could the TSC from the hypercall be applied to kvmclock to do this? My understanding is that get_device_system_crosststamp (which is used in the sole in-tree implementation of getcrosststamp) already contains all the logic to do that. Paolo ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-18 15:45 ` Paolo Bonzini @ 2017-01-18 15:57 ` Marcelo Tosatti 0 siblings, 0 replies; 54+ messages in thread From: Marcelo Tosatti @ 2017-01-18 15:57 UTC (permalink / raw) To: Paolo Bonzini Cc: Miroslav Lichvar, Radim Krcmar, kvm, linux-kernel, Richard Cochran On Wed, Jan 18, 2017 at 04:45:07PM +0100, Paolo Bonzini wrote: > > > On 18/01/2017 15:50, Marcelo Tosatti wrote: > >> Interesting idea! For this to work, KVM needs to implement > >> getcrosstimestamp and ptp_chardev.c can then add an alternative > >> implementation of PTP_SYS_OFFSET, based on precise cross timestamps. > >> > >> Something like > >> > >> for (i = 0; i <= sysoff->n_samples; i++) { > >> // ... call getcrosststamp ... > >> sysns = ktime_to_ns(xtstamp.sys_realtime); > >> if (i > 0) { > >> devns = ktime_to_ns(xtstamp.device); > >> devns -= (sysns - prev_sysns) / 2; > >> devts = ns_to_timespec(devns); > >> pct->sec = devts.tv_sec; > >> pct->nsec = devts.tv_nsec; > >> pct++; > >> } > >> systs = ns_to_timespec(sysns); > >> pct->sec = ts.tv_sec; > >> pct->nsec = ts.tv_nsec; > >> pct++; > >> prev_sysns = sysns; > >> } > >> > >> Marcelo, can you give it a try? > > > > Can convert fine, but problem is the simultaneous read > > of host and guest clocks. > > Could the TSC from the hypercall be applied to kvmclock to do this? My > understanding is that get_device_system_crosststamp (which is used in > the sole in-tree implementation of getcrosststamp) already contains all > the logic to do that. > > Paolo Yeah feed that TSC to pvclock_clocksource_read. Cool will take a look at that function thanks. ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-18 12:46 ` Paolo Bonzini 2017-01-18 13:36 ` Miroslav Lichvar @ 2017-01-18 14:24 ` Marcelo Tosatti 2017-01-18 15:54 ` Miroslav Lichvar 2017-01-18 14:37 ` Marcelo Tosatti 2017-01-18 15:59 ` Radim Krcmar 3 siblings, 1 reply; 54+ messages in thread From: Marcelo Tosatti @ 2017-01-18 14:24 UTC (permalink / raw) To: Paolo Bonzini Cc: Radim Krcmar, Miroslav Lichvar, kvm, linux-kernel, Richard Cochran On Wed, Jan 18, 2017 at 01:46:58PM +0100, Paolo Bonzini wrote: > On 18/01/2017 13:24, Marcelo Tosatti wrote: > > On Wed, Jan 18, 2017 at 10:17:38AM -0200, Marcelo Tosatti wrote: > >> On Tue, Jan 17, 2017 at 04:36:21PM +0100, Radim Krcmar wrote: > >>> 2017-01-17 09:30-0200, Marcelo Tosatti: > >>>> On Tue, Jan 17, 2017 at 09:03:27AM +0100, Miroslav Lichvar wrote: > >>>>> Users of the PTP_SYS_OFFSET ioctl assume that (ts[0]+ts[2])/2 > >>>>> corresponds to ts[1], (ts[2]+ts[4])/2 corresponds to ts[3], and so on. > >>>>> > >>>>> ts[1] ts[3] > >>>>> Host time ---------+---------+........ > >>>>> | | > >>>>> | | > >>>>> Guest time ----+---------+---------+...... > >>>>> ts[0] ts[2] ts[4] > >>> > >>> KVM PTP delay moves host ts[i] to be close to guest ts[i+1] and makes > >>> the offset very consistent, so the graph would look like: > >>> > >>> ts[1] ts[3] > >>> Host time -------------+---------+........ > >>> | | > >>> | | > >>> Guest time ----+---------+---------+...... > >>> ts[0] ts[2] ts[4] > >>> > >>> which doesn't sound good if users assume that the host reading is in the > >>> middle -- the guest time would be ahead of the host time. > >> > >> Testcase: run a guest and a loop sending SIGUSR1 to vcpu0 (emulating > >> intense interrupts). Follows results: > >> > >> Without TSC delta calculation: > >> ============================= > >> > >> #* PHC0 0 3 377 2 -99ns[ +206ns] +/- 116ns > >> #* PHC0 0 3 377 8 +202ns[ +249ns] +/- 111ns > >> #* PHC0 0 3 377 8 -213ns[ +683ns] +/- 88ns > >> #* PHC0 0 3 377 6 +77ns[ +319ns] +/- 56ns > >> #* PHC0 0 3 377 4 -771ns[-1029ns] +/- 93ns > >> #* PHC0 0 3 377 10 -49ns[ -58ns] +/- 121ns > >> #* PHC0 0 3 377 9 +562ns[ +703ns] +/- 107ns > >> #* PHC0 0 3 377 6 -2ns[ -3ns] +/- 94ns > >> #* PHC0 0 3 377 4 +451ns[ +494ns] +/- 138ns > >> #* PHC0 0 3 377 11 -67ns[ -74ns] +/- 113ns > >> #* PHC0 0 3 377 8 +244ns[ +264ns] +/- 119ns > >> #* PHC0 0 3 377 7 -696ns[ -890ns] +/- 89ns > >> #* PHC0 0 3 377 4 +468ns[ +560ns] +/- 110ns > >> #* PHC0 0 3 377 11 -310ns[ -430ns] +/- 72ns > >> #* PHC0 0 3 377 9 +189ns[ +298ns] +/- 54ns > >> #* PHC0 0 3 377 7 +594ns[ +473ns] +/- 96ns > >> #* PHC0 0 3 377 5 +151ns[ +280ns] +/- 71ns > >> #* PHC0 0 3 377 10 -590ns[ -696ns] +/- 94ns > >> #* PHC0 0 3 377 8 +415ns[ +526ns] +/- 74ns > >> #* PHC0 0 3 377 6 +1381ns[+1469ns] +/- 101ns > >> #* PHC0 0 3 377 4 +571ns[+1304ns] +/- 54ns > >> #* PHC0 0 3 377 8 -5ns[ +71ns] +/- 139ns > >> #* PHC0 0 3 377 7 -247ns[ -502ns] +/- 69ns > >> #* PHC0 0 3 377 5 -283ns[ +879ns] +/- 73ns > >> #* PHC0 0 3 377 3 +148ns[ -109ns] +/- 61ns > >> > >> With TSC delta calculation: > >> ============================ > >> > >> #* PHC0 0 3 377 7 +379ns[ +432ns] +/- 53ns > >> #* PHC0 0 3 377 9 +106ns[ +420ns] +/- 42ns > >> #* PHC0 0 3 377 7 -58ns[ -136ns] +/- 62ns > >> #* PHC0 0 3 377 12 +93ns[ -38ns] +/- 64ns > >> #* PHC0 0 3 377 8 +84ns[ +107ns] +/- 69ns > >> #* PHC0 0 3 377 3 -76ns[ -103ns] +/- 52ns > >> #* PHC0 0 3 377 7 +52ns[ +63ns] +/- 50ns > >> #* PHC0 0 3 377 11 +29ns[ +31ns] +/- 70ns > >> #* PHC0 0 3 377 7 -47ns[ -56ns] +/- 42ns > >> #* PHC0 0 3 377 10 -35ns[ -42ns] +/- 33ns > >> #* PHC0 0 3 377 7 -32ns[ -34ns] +/- 42ns > >> #* PHC0 0 3 377 11 -172ns[ -173ns] +/- 118ns > >> #* PHC0 0 3 377 6 +65ns[ +76ns] +/- 23ns > >> #* PHC0 0 3 377 9 +18ns[ +23ns] +/- 37ns > >> #* PHC0 0 3 377 6 +41ns[ -60ns] +/- 30ns > >> #* PHC0 0 3 377 10 +39ns[ +183ns] +/- 42ns > >> #* PHC0 0 3 377 6 +50ns[ +102ns] +/- 86ns > >> #* PHC0 0 3 377 11 +50ns[ +75ns] +/- 52ns > >> #* PHC0 0 3 377 6 +50ns[ +116ns] +/- 100ns > >> #* PHC0 0 3 377 10 +46ns[ +65ns] +/- 79ns > >> #* PHC0 0 3 377 7 -38ns[ -51ns] +/- 29ns > >> #* PHC0 0 3 377 10 -11ns[ -12ns] +/- 32ns > >> #* PHC0 0 3 377 7 -31ns[ -32ns] +/- 99ns > >> #* PHC0 0 3 377 10 +222ns[ +238ns] +/- 58ns > >> #* PHC0 0 3 377 6 +185ns[ +207ns] +/- 39ns > >> #* PHC0 0 3 377 10 -392ns[ -394ns] +/- 118ns > >> #* PHC0 0 3 377 6 -9ns[ -50ns] +/- 35ns > >> #* PHC0 0 3 377 10 -346ns[ -355ns] +/- 111ns > >> > >> > >> Do you still want to drop it in favour of simplicity? > > > > This is the output of "chronyc sources". See section "Time sources" > > of https://chrony.tuxfamily.org/doc/2.4/chronyc.html. > > It's just that it's not obvious why you get better results with biased > host timestamps. What makes the biased host timestamp more precise? The issue is that, without it you have a larger window for interruptions to take place and therefore the read value when ->gettime64 return is from longer time in the past. > I'd rather use PTP_SYS_OFFSET_PRECISE instead, but unfortunately chrony > does not support it---but I would still prefer you to support > PTP_SYS_OFFSET_PRECISE as well. Sure, I'll check if it makes sense to implement PTP_SYS_OFFSET_PRECISE for KVM case. ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-18 14:24 ` Marcelo Tosatti @ 2017-01-18 15:54 ` Miroslav Lichvar 2017-01-18 16:07 ` Paolo Bonzini 2017-01-18 16:14 ` Radim Krcmar 0 siblings, 2 replies; 54+ messages in thread From: Miroslav Lichvar @ 2017-01-18 15:54 UTC (permalink / raw) To: Marcelo Tosatti Cc: Paolo Bonzini, Radim Krcmar, kvm, linux-kernel, Richard Cochran On Wed, Jan 18, 2017 at 12:24:09PM -0200, Marcelo Tosatti wrote: > On Wed, Jan 18, 2017 at 01:46:58PM +0100, Paolo Bonzini wrote: > > I'd rather use PTP_SYS_OFFSET_PRECISE instead, but unfortunately chrony > > does not support it---but I would still prefer you to support > > PTP_SYS_OFFSET_PRECISE as well. > > Sure, I'll check if it makes sense to implement PTP_SYS_OFFSET_PRECISE for > KVM case. But is it really so precise that the application can safely assume there are no errors due to asymmetric delays, etc? I think PTP_SYS_OFFSET_PRECISE should be supported only if the accuracy of the offset measured between the HW and system clock is not worse than the precision of the system clock (typically few tens of nanoseconds). It would be good to verify the accuracy of the offset when the host and guest clocks are synchronised to each other over PTP using two NICs with HW timestamping. -- Miroslav Lichvar ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-18 15:54 ` Miroslav Lichvar @ 2017-01-18 16:07 ` Paolo Bonzini 2017-01-18 16:14 ` Radim Krcmar 1 sibling, 0 replies; 54+ messages in thread From: Paolo Bonzini @ 2017-01-18 16:07 UTC (permalink / raw) To: Miroslav Lichvar, Marcelo Tosatti Cc: Radim Krcmar, kvm, linux-kernel, Richard Cochran On 18/01/2017 16:54, Miroslav Lichvar wrote: >> Sure, I'll check if it makes sense to implement PTP_SYS_OFFSET_PRECISE for >> KVM case. > But is it really so precise that the application can safely assume > there are no errors due to asymmetric delays, etc? I think > PTP_SYS_OFFSET_PRECISE should be supported only if the accuracy of > the offset measured between the HW and system clock is not worse than > the precision of the system clock (typically few tens of nanoseconds). > > It would be good to verify the accuracy of the offset when the host > and guest clocks are synchronised to each other over PTP using two > NICs with HW timestamping. PTP_SYS_OFFSET_PRECISE works if the guest can compute its own timestamp based on the same source as the device. On bare metal you have: - the source for system clock is the TSC (with clocksource_tsc) - device provides a (time, ART) tuple You can convert ART->TSC and then ask clocksource_tsc for a system time based on the device-provided ART value. Likewise for KVM: - the source for system clock is the guest TSC (with kvmclock) - host can provide a (time, guest TSC) tuple The PTP driver can take the host-provided guest TSC, and ask kvmclock for a system time based on that TSC value. It's even simpler because there's no ART->TSC conversion involved, and it will always be precise and independent of any vmexit or interrupt delay. Paolo ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-18 15:54 ` Miroslav Lichvar 2017-01-18 16:07 ` Paolo Bonzini @ 2017-01-18 16:14 ` Radim Krcmar 1 sibling, 0 replies; 54+ messages in thread From: Radim Krcmar @ 2017-01-18 16:14 UTC (permalink / raw) To: Miroslav Lichvar Cc: Marcelo Tosatti, Paolo Bonzini, kvm, linux-kernel, Richard Cochran 2017-01-18 16:54+0100, Miroslav Lichvar: > On Wed, Jan 18, 2017 at 12:24:09PM -0200, Marcelo Tosatti wrote: >> On Wed, Jan 18, 2017 at 01:46:58PM +0100, Paolo Bonzini wrote: >> > I'd rather use PTP_SYS_OFFSET_PRECISE instead, but unfortunately chrony >> > does not support it---but I would still prefer you to support >> > PTP_SYS_OFFSET_PRECISE as well. >> >> Sure, I'll check if it makes sense to implement PTP_SYS_OFFSET_PRECISE for >> KVM case. > > But is it really so precise that the application can safely assume > there are no errors due to asymmetric delays, etc? I think > PTP_SYS_OFFSET_PRECISE should be supported only if the accuracy of > the offset measured between the HW and system clock is not worse than > the precision of the system clock (typically few tens of nanoseconds). KVM is actually the perfect user of PTP_SYS_OFFSET_PRECISE: the host and guest system clocks use the same hardware clock. (We want to copy the host data about the clock to the guest, and we just happen to use PTP for that.) > It would be good to verify the accuracy of the offset when the host > and guest clocks are synchronised to each other over PTP using two > NICs with HW timestamping. We can verify the accuracy just by reading the host time and tsc computing guest time from that tsc -- they should be equal. Well, we don't give host frequency to the guest, but I hope that the guest can compute it accurately after polling the host few times. (Still, a room for improvement.) NICs with HW timestamping would be a bit more complicated to set up. :) ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-18 12:46 ` Paolo Bonzini 2017-01-18 13:36 ` Miroslav Lichvar 2017-01-18 14:24 ` Marcelo Tosatti @ 2017-01-18 14:37 ` Marcelo Tosatti 2017-01-18 14:53 ` Marcelo Tosatti 2017-01-18 15:59 ` Radim Krcmar 3 siblings, 1 reply; 54+ messages in thread From: Marcelo Tosatti @ 2017-01-18 14:37 UTC (permalink / raw) To: Paolo Bonzini Cc: Radim Krcmar, Miroslav Lichvar, kvm, linux-kernel, Richard Cochran On Wed, Jan 18, 2017 at 01:46:58PM +0100, Paolo Bonzini wrote: > > > On 18/01/2017 13:24, Marcelo Tosatti wrote: > > On Wed, Jan 18, 2017 at 10:17:38AM -0200, Marcelo Tosatti wrote: > >> On Tue, Jan 17, 2017 at 04:36:21PM +0100, Radim Krcmar wrote: > >>> 2017-01-17 09:30-0200, Marcelo Tosatti: > >>>> On Tue, Jan 17, 2017 at 09:03:27AM +0100, Miroslav Lichvar wrote: > >>>>> Users of the PTP_SYS_OFFSET ioctl assume that (ts[0]+ts[2])/2 > >>>>> corresponds to ts[1], (ts[2]+ts[4])/2 corresponds to ts[3], and so on. > >>>>> > >>>>> ts[1] ts[3] > >>>>> Host time ---------+---------+........ > >>>>> | | > >>>>> | | > >>>>> Guest time ----+---------+---------+...... > >>>>> ts[0] ts[2] ts[4] > >>> > >>> KVM PTP delay moves host ts[i] to be close to guest ts[i+1] and makes > >>> the offset very consistent, so the graph would look like: > >>> > >>> ts[1] ts[3] > >>> Host time -------------+---------+........ > >>> | | > >>> | | > >>> Guest time ----+---------+---------+...... > >>> ts[0] ts[2] ts[4] > >>> > >>> which doesn't sound good if users assume that the host reading is in the > >>> middle -- the guest time would be ahead of the host time. > >> > >> Testcase: run a guest and a loop sending SIGUSR1 to vcpu0 (emulating > >> intense interrupts). Follows results: > >> > >> Without TSC delta calculation: > >> ============================= > >> > >> #* PHC0 0 3 377 2 -99ns[ +206ns] +/- 116ns > >> #* PHC0 0 3 377 8 +202ns[ +249ns] +/- 111ns > >> #* PHC0 0 3 377 8 -213ns[ +683ns] +/- 88ns > >> #* PHC0 0 3 377 6 +77ns[ +319ns] +/- 56ns > >> #* PHC0 0 3 377 4 -771ns[-1029ns] +/- 93ns > >> #* PHC0 0 3 377 10 -49ns[ -58ns] +/- 121ns > >> #* PHC0 0 3 377 9 +562ns[ +703ns] +/- 107ns > >> #* PHC0 0 3 377 6 -2ns[ -3ns] +/- 94ns > >> #* PHC0 0 3 377 4 +451ns[ +494ns] +/- 138ns > >> #* PHC0 0 3 377 11 -67ns[ -74ns] +/- 113ns > >> #* PHC0 0 3 377 8 +244ns[ +264ns] +/- 119ns > >> #* PHC0 0 3 377 7 -696ns[ -890ns] +/- 89ns > >> #* PHC0 0 3 377 4 +468ns[ +560ns] +/- 110ns > >> #* PHC0 0 3 377 11 -310ns[ -430ns] +/- 72ns > >> #* PHC0 0 3 377 9 +189ns[ +298ns] +/- 54ns > >> #* PHC0 0 3 377 7 +594ns[ +473ns] +/- 96ns > >> #* PHC0 0 3 377 5 +151ns[ +280ns] +/- 71ns > >> #* PHC0 0 3 377 10 -590ns[ -696ns] +/- 94ns > >> #* PHC0 0 3 377 8 +415ns[ +526ns] +/- 74ns > >> #* PHC0 0 3 377 6 +1381ns[+1469ns] +/- 101ns > >> #* PHC0 0 3 377 4 +571ns[+1304ns] +/- 54ns > >> #* PHC0 0 3 377 8 -5ns[ +71ns] +/- 139ns > >> #* PHC0 0 3 377 7 -247ns[ -502ns] +/- 69ns > >> #* PHC0 0 3 377 5 -283ns[ +879ns] +/- 73ns > >> #* PHC0 0 3 377 3 +148ns[ -109ns] +/- 61ns > >> > >> With TSC delta calculation: > >> ============================ > >> > >> #* PHC0 0 3 377 7 +379ns[ +432ns] +/- 53ns > >> #* PHC0 0 3 377 9 +106ns[ +420ns] +/- 42ns > >> #* PHC0 0 3 377 7 -58ns[ -136ns] +/- 62ns > >> #* PHC0 0 3 377 12 +93ns[ -38ns] +/- 64ns > >> #* PHC0 0 3 377 8 +84ns[ +107ns] +/- 69ns > >> #* PHC0 0 3 377 3 -76ns[ -103ns] +/- 52ns > >> #* PHC0 0 3 377 7 +52ns[ +63ns] +/- 50ns > >> #* PHC0 0 3 377 11 +29ns[ +31ns] +/- 70ns > >> #* PHC0 0 3 377 7 -47ns[ -56ns] +/- 42ns > >> #* PHC0 0 3 377 10 -35ns[ -42ns] +/- 33ns > >> #* PHC0 0 3 377 7 -32ns[ -34ns] +/- 42ns > >> #* PHC0 0 3 377 11 -172ns[ -173ns] +/- 118ns > >> #* PHC0 0 3 377 6 +65ns[ +76ns] +/- 23ns > >> #* PHC0 0 3 377 9 +18ns[ +23ns] +/- 37ns > >> #* PHC0 0 3 377 6 +41ns[ -60ns] +/- 30ns > >> #* PHC0 0 3 377 10 +39ns[ +183ns] +/- 42ns > >> #* PHC0 0 3 377 6 +50ns[ +102ns] +/- 86ns > >> #* PHC0 0 3 377 11 +50ns[ +75ns] +/- 52ns > >> #* PHC0 0 3 377 6 +50ns[ +116ns] +/- 100ns > >> #* PHC0 0 3 377 10 +46ns[ +65ns] +/- 79ns > >> #* PHC0 0 3 377 7 -38ns[ -51ns] +/- 29ns > >> #* PHC0 0 3 377 10 -11ns[ -12ns] +/- 32ns > >> #* PHC0 0 3 377 7 -31ns[ -32ns] +/- 99ns > >> #* PHC0 0 3 377 10 +222ns[ +238ns] +/- 58ns > >> #* PHC0 0 3 377 6 +185ns[ +207ns] +/- 39ns > >> #* PHC0 0 3 377 10 -392ns[ -394ns] +/- 118ns > >> #* PHC0 0 3 377 6 -9ns[ -50ns] +/- 35ns > >> #* PHC0 0 3 377 10 -346ns[ -355ns] +/- 111ns > >> > >> > >> Do you still want to drop it in favour of simplicity? > > > > This is the output of "chronyc sources". See section "Time sources" > > of https://chrony.tuxfamily.org/doc/2.4/chronyc.html. > > It's just that it's not obvious why you get better results with biased > host timestamps. What makes the biased host timestamp more precise? > > I'd rather use PTP_SYS_OFFSET_PRECISE instead, but unfortunately chrony > does not support it---but I would still prefer you to support > PTP_SYS_OFFSET_PRECISE as well. A single TSC read could be used to implement the PRECISE ioctl, but if a timer interrupt takes place on either the host or the guest, and that timer interrupt "adds" the TSC delta to xtime.nsec/xtime.sec, then that single TSC read cannot be used. So you would have to stop timer interrupts (in guest and host) for the duration of the PRECISE ioctl in the guest to avoid that situation, which seems a bit overkill to me. Any other ideas? ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-18 14:37 ` Marcelo Tosatti @ 2017-01-18 14:53 ` Marcelo Tosatti 2017-01-18 15:20 ` Radim Krcmar 0 siblings, 1 reply; 54+ messages in thread From: Marcelo Tosatti @ 2017-01-18 14:53 UTC (permalink / raw) To: Paolo Bonzini Cc: Radim Krcmar, Miroslav Lichvar, kvm, linux-kernel, Richard Cochran GOn Wed, Jan 18, 2017 at 12:37:25PM -0200, Marcelo Tosatti wrote: > On Wed, Jan 18, 2017 at 01:46:58PM +0100, Paolo Bonzini wrote: > > > > > > On 18/01/2017 13:24, Marcelo Tosatti wrote: > > > On Wed, Jan 18, 2017 at 10:17:38AM -0200, Marcelo Tosatti wrote: > > >> On Tue, Jan 17, 2017 at 04:36:21PM +0100, Radim Krcmar wrote: > > >>> 2017-01-17 09:30-0200, Marcelo Tosatti: > > >>>> On Tue, Jan 17, 2017 at 09:03:27AM +0100, Miroslav Lichvar wrote: > > >>>>> Users of the PTP_SYS_OFFSET ioctl assume that (ts[0]+ts[2])/2 > > >>>>> corresponds to ts[1], (ts[2]+ts[4])/2 corresponds to ts[3], and so on. > > >>>>> > > >>>>> ts[1] ts[3] > > >>>>> Host time ---------+---------+........ > > >>>>> | | > > >>>>> | | > > >>>>> Guest time ----+---------+---------+...... > > >>>>> ts[0] ts[2] ts[4] > > >>> > > >>> KVM PTP delay moves host ts[i] to be close to guest ts[i+1] and makes > > >>> the offset very consistent, so the graph would look like: > > >>> > > >>> ts[1] ts[3] > > >>> Host time -------------+---------+........ > > >>> | | > > >>> | | > > >>> Guest time ----+---------+---------+...... > > >>> ts[0] ts[2] ts[4] > > >>> > > >>> which doesn't sound good if users assume that the host reading is in the > > >>> middle -- the guest time would be ahead of the host time. > > >> > > >> Testcase: run a guest and a loop sending SIGUSR1 to vcpu0 (emulating > > >> intense interrupts). Follows results: > > >> > > >> Without TSC delta calculation: > > >> ============================= > > >> > > >> #* PHC0 0 3 377 2 -99ns[ +206ns] +/- 116ns > > >> #* PHC0 0 3 377 8 +202ns[ +249ns] +/- 111ns > > >> #* PHC0 0 3 377 8 -213ns[ +683ns] +/- 88ns > > >> #* PHC0 0 3 377 6 +77ns[ +319ns] +/- 56ns > > >> #* PHC0 0 3 377 4 -771ns[-1029ns] +/- 93ns > > >> #* PHC0 0 3 377 10 -49ns[ -58ns] +/- 121ns > > >> #* PHC0 0 3 377 9 +562ns[ +703ns] +/- 107ns > > >> #* PHC0 0 3 377 6 -2ns[ -3ns] +/- 94ns > > >> #* PHC0 0 3 377 4 +451ns[ +494ns] +/- 138ns > > >> #* PHC0 0 3 377 11 -67ns[ -74ns] +/- 113ns > > >> #* PHC0 0 3 377 8 +244ns[ +264ns] +/- 119ns > > >> #* PHC0 0 3 377 7 -696ns[ -890ns] +/- 89ns > > >> #* PHC0 0 3 377 4 +468ns[ +560ns] +/- 110ns > > >> #* PHC0 0 3 377 11 -310ns[ -430ns] +/- 72ns > > >> #* PHC0 0 3 377 9 +189ns[ +298ns] +/- 54ns > > >> #* PHC0 0 3 377 7 +594ns[ +473ns] +/- 96ns > > >> #* PHC0 0 3 377 5 +151ns[ +280ns] +/- 71ns > > >> #* PHC0 0 3 377 10 -590ns[ -696ns] +/- 94ns > > >> #* PHC0 0 3 377 8 +415ns[ +526ns] +/- 74ns > > >> #* PHC0 0 3 377 6 +1381ns[+1469ns] +/- 101ns > > >> #* PHC0 0 3 377 4 +571ns[+1304ns] +/- 54ns > > >> #* PHC0 0 3 377 8 -5ns[ +71ns] +/- 139ns > > >> #* PHC0 0 3 377 7 -247ns[ -502ns] +/- 69ns > > >> #* PHC0 0 3 377 5 -283ns[ +879ns] +/- 73ns > > >> #* PHC0 0 3 377 3 +148ns[ -109ns] +/- 61ns > > >> > > >> With TSC delta calculation: > > >> ============================ > > >> > > >> #* PHC0 0 3 377 7 +379ns[ +432ns] +/- 53ns > > >> #* PHC0 0 3 377 9 +106ns[ +420ns] +/- 42ns > > >> #* PHC0 0 3 377 7 -58ns[ -136ns] +/- 62ns > > >> #* PHC0 0 3 377 12 +93ns[ -38ns] +/- 64ns > > >> #* PHC0 0 3 377 8 +84ns[ +107ns] +/- 69ns > > >> #* PHC0 0 3 377 3 -76ns[ -103ns] +/- 52ns > > >> #* PHC0 0 3 377 7 +52ns[ +63ns] +/- 50ns > > >> #* PHC0 0 3 377 11 +29ns[ +31ns] +/- 70ns > > >> #* PHC0 0 3 377 7 -47ns[ -56ns] +/- 42ns > > >> #* PHC0 0 3 377 10 -35ns[ -42ns] +/- 33ns > > >> #* PHC0 0 3 377 7 -32ns[ -34ns] +/- 42ns > > >> #* PHC0 0 3 377 11 -172ns[ -173ns] +/- 118ns > > >> #* PHC0 0 3 377 6 +65ns[ +76ns] +/- 23ns > > >> #* PHC0 0 3 377 9 +18ns[ +23ns] +/- 37ns > > >> #* PHC0 0 3 377 6 +41ns[ -60ns] +/- 30ns > > >> #* PHC0 0 3 377 10 +39ns[ +183ns] +/- 42ns > > >> #* PHC0 0 3 377 6 +50ns[ +102ns] +/- 86ns > > >> #* PHC0 0 3 377 11 +50ns[ +75ns] +/- 52ns > > >> #* PHC0 0 3 377 6 +50ns[ +116ns] +/- 100ns > > >> #* PHC0 0 3 377 10 +46ns[ +65ns] +/- 79ns > > >> #* PHC0 0 3 377 7 -38ns[ -51ns] +/- 29ns > > >> #* PHC0 0 3 377 10 -11ns[ -12ns] +/- 32ns > > >> #* PHC0 0 3 377 7 -31ns[ -32ns] +/- 99ns > > >> #* PHC0 0 3 377 10 +222ns[ +238ns] +/- 58ns > > >> #* PHC0 0 3 377 6 +185ns[ +207ns] +/- 39ns > > >> #* PHC0 0 3 377 10 -392ns[ -394ns] +/- 118ns > > >> #* PHC0 0 3 377 6 -9ns[ -50ns] +/- 35ns > > >> #* PHC0 0 3 377 10 -346ns[ -355ns] +/- 111ns > > >> > > >> > > >> Do you still want to drop it in favour of simplicity? > > > > > > This is the output of "chronyc sources". See section "Time sources" > > > of https://chrony.tuxfamily.org/doc/2.4/chronyc.html. > > > > It's just that it's not obvious why you get better results with biased > > host timestamps. What makes the biased host timestamp more precise? > > > > I'd rather use PTP_SYS_OFFSET_PRECISE instead, but unfortunately chrony > > does not support it---but I would still prefer you to support > > PTP_SYS_OFFSET_PRECISE as well. > > A single TSC read could be used to implement the PRECISE ioctl, but if > a timer interrupt takes place on either the host or the guest, and that > timer interrupt "adds" the TSC delta to xtime.nsec/xtime.sec, then that > single TSC read cannot be used. > > So you would have to stop timer interrupts (in guest and host) for the duration of the > PRECISE ioctl in the guest to avoid that situation, which seems a bit > overkill to me. > > Any other ideas? Could have a hypercall that disables host timer interrupts for a specified amount of time... But that does not scale with multiple VMs. ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-18 14:53 ` Marcelo Tosatti @ 2017-01-18 15:20 ` Radim Krcmar 2017-01-18 15:28 ` Marcelo Tosatti 0 siblings, 1 reply; 54+ messages in thread From: Radim Krcmar @ 2017-01-18 15:20 UTC (permalink / raw) To: Marcelo Tosatti Cc: Paolo Bonzini, Miroslav Lichvar, kvm, linux-kernel, Richard Cochran 2017-01-18 12:53-0200, Marcelo Tosatti: > GOn Wed, Jan 18, 2017 at 12:37:25PM -0200, Marcelo Tosatti wrote: > > On Wed, Jan 18, 2017 at 01:46:58PM +0100, Paolo Bonzini wrote: > > > > > > > > > On 18/01/2017 13:24, Marcelo Tosatti wrote: > > > > On Wed, Jan 18, 2017 at 10:17:38AM -0200, Marcelo Tosatti wrote: > > > >> On Tue, Jan 17, 2017 at 04:36:21PM +0100, Radim Krcmar wrote: > > > >>> 2017-01-17 09:30-0200, Marcelo Tosatti: > > > >>>> On Tue, Jan 17, 2017 at 09:03:27AM +0100, Miroslav Lichvar wrote: > > > >>>>> Users of the PTP_SYS_OFFSET ioctl assume that (ts[0]+ts[2])/2 > > > >>>>> corresponds to ts[1], (ts[2]+ts[4])/2 corresponds to ts[3], and so on. > > > >>>>> > > > >>>>> ts[1] ts[3] > > > >>>>> Host time ---------+---------+........ > > > >>>>> | | > > > >>>>> | | > > > >>>>> Guest time ----+---------+---------+...... > > > >>>>> ts[0] ts[2] ts[4] > > > >>> > > > >>> KVM PTP delay moves host ts[i] to be close to guest ts[i+1] and makes > > > >>> the offset very consistent, so the graph would look like: > > > >>> > > > >>> ts[1] ts[3] > > > >>> Host time -------------+---------+........ > > > >>> | | > > > >>> | | > > > >>> Guest time ----+---------+---------+...... > > > >>> ts[0] ts[2] ts[4] > > > >>> > > > >>> which doesn't sound good if users assume that the host reading is in the > > > >>> middle -- the guest time would be ahead of the host time. > > > >> > > > >> Testcase: run a guest and a loop sending SIGUSR1 to vcpu0 (emulating > > > >> intense interrupts). Follows results: > > > >> > > > >> Without TSC delta calculation: > > > >> ============================= > > > >> > > > >> #* PHC0 0 3 377 2 -99ns[ +206ns] +/- 116ns > > > >> #* PHC0 0 3 377 8 +202ns[ +249ns] +/- 111ns > > > >> #* PHC0 0 3 377 8 -213ns[ +683ns] +/- 88ns > > > >> #* PHC0 0 3 377 6 +77ns[ +319ns] +/- 56ns > > > >> #* PHC0 0 3 377 4 -771ns[-1029ns] +/- 93ns > > > >> #* PHC0 0 3 377 10 -49ns[ -58ns] +/- 121ns > > > >> #* PHC0 0 3 377 9 +562ns[ +703ns] +/- 107ns > > > >> #* PHC0 0 3 377 6 -2ns[ -3ns] +/- 94ns > > > >> #* PHC0 0 3 377 4 +451ns[ +494ns] +/- 138ns > > > >> #* PHC0 0 3 377 11 -67ns[ -74ns] +/- 113ns > > > >> #* PHC0 0 3 377 8 +244ns[ +264ns] +/- 119ns > > > >> #* PHC0 0 3 377 7 -696ns[ -890ns] +/- 89ns > > > >> #* PHC0 0 3 377 4 +468ns[ +560ns] +/- 110ns > > > >> #* PHC0 0 3 377 11 -310ns[ -430ns] +/- 72ns > > > >> #* PHC0 0 3 377 9 +189ns[ +298ns] +/- 54ns > > > >> #* PHC0 0 3 377 7 +594ns[ +473ns] +/- 96ns > > > >> #* PHC0 0 3 377 5 +151ns[ +280ns] +/- 71ns > > > >> #* PHC0 0 3 377 10 -590ns[ -696ns] +/- 94ns > > > >> #* PHC0 0 3 377 8 +415ns[ +526ns] +/- 74ns > > > >> #* PHC0 0 3 377 6 +1381ns[+1469ns] +/- 101ns > > > >> #* PHC0 0 3 377 4 +571ns[+1304ns] +/- 54ns > > > >> #* PHC0 0 3 377 8 -5ns[ +71ns] +/- 139ns > > > >> #* PHC0 0 3 377 7 -247ns[ -502ns] +/- 69ns > > > >> #* PHC0 0 3 377 5 -283ns[ +879ns] +/- 73ns > > > >> #* PHC0 0 3 377 3 +148ns[ -109ns] +/- 61ns > > > >> > > > >> With TSC delta calculation: > > > >> ============================ > > > >> > > > >> #* PHC0 0 3 377 7 +379ns[ +432ns] +/- 53ns > > > >> #* PHC0 0 3 377 9 +106ns[ +420ns] +/- 42ns > > > >> #* PHC0 0 3 377 7 -58ns[ -136ns] +/- 62ns > > > >> #* PHC0 0 3 377 12 +93ns[ -38ns] +/- 64ns > > > >> #* PHC0 0 3 377 8 +84ns[ +107ns] +/- 69ns > > > >> #* PHC0 0 3 377 3 -76ns[ -103ns] +/- 52ns > > > >> #* PHC0 0 3 377 7 +52ns[ +63ns] +/- 50ns > > > >> #* PHC0 0 3 377 11 +29ns[ +31ns] +/- 70ns > > > >> #* PHC0 0 3 377 7 -47ns[ -56ns] +/- 42ns > > > >> #* PHC0 0 3 377 10 -35ns[ -42ns] +/- 33ns > > > >> #* PHC0 0 3 377 7 -32ns[ -34ns] +/- 42ns > > > >> #* PHC0 0 3 377 11 -172ns[ -173ns] +/- 118ns > > > >> #* PHC0 0 3 377 6 +65ns[ +76ns] +/- 23ns > > > >> #* PHC0 0 3 377 9 +18ns[ +23ns] +/- 37ns > > > >> #* PHC0 0 3 377 6 +41ns[ -60ns] +/- 30ns > > > >> #* PHC0 0 3 377 10 +39ns[ +183ns] +/- 42ns > > > >> #* PHC0 0 3 377 6 +50ns[ +102ns] +/- 86ns > > > >> #* PHC0 0 3 377 11 +50ns[ +75ns] +/- 52ns > > > >> #* PHC0 0 3 377 6 +50ns[ +116ns] +/- 100ns > > > >> #* PHC0 0 3 377 10 +46ns[ +65ns] +/- 79ns > > > >> #* PHC0 0 3 377 7 -38ns[ -51ns] +/- 29ns > > > >> #* PHC0 0 3 377 10 -11ns[ -12ns] +/- 32ns > > > >> #* PHC0 0 3 377 7 -31ns[ -32ns] +/- 99ns > > > >> #* PHC0 0 3 377 10 +222ns[ +238ns] +/- 58ns > > > >> #* PHC0 0 3 377 6 +185ns[ +207ns] +/- 39ns > > > >> #* PHC0 0 3 377 10 -392ns[ -394ns] +/- 118ns > > > >> #* PHC0 0 3 377 6 -9ns[ -50ns] +/- 35ns > > > >> #* PHC0 0 3 377 10 -346ns[ -355ns] +/- 111ns > > > >> > > > >> > > > >> Do you still want to drop it in favour of simplicity? > > > > > > > > This is the output of "chronyc sources". See section "Time sources" > > > > of https://chrony.tuxfamily.org/doc/2.4/chronyc.html. > > > > > > It's just that it's not obvious why you get better results with biased > > > host timestamps. What makes the biased host timestamp more precise? > > > > > > I'd rather use PTP_SYS_OFFSET_PRECISE instead, but unfortunately chrony > > > does not support it---but I would still prefer you to support > > > PTP_SYS_OFFSET_PRECISE as well. > > > > A single TSC read could be used to implement the PRECISE ioctl, but if > > a timer interrupt takes place on either the host or the guest, and that > > timer interrupt "adds" the TSC delta to xtime.nsec/xtime.sec, then that > > single TSC read cannot be used. > > > > So you would have to stop timer interrupts (in guest and host) for the duration of the > > PRECISE ioctl in the guest to avoid that situation, which seems a bit > > overkill to me. > > > > Any other ideas? > > Could have a hypercall that disables host timer interrupts for > a specified amount of time... But that does not scale with multiple VMs. No need to disable interrupts on guest nor host as both protect the time by a seqlock. Do timer interrupts always advance the time base? (We could disable guest interrupts if they were too disruptive because of the long hypercall, but the host is fine in any case.) The guest protects tsc->sec/nsec by tk_core.seq, so we can make sure that the guest computed sec/nsec from the host tsc read from hypercall is consistent. The {sec @ tsc, nsec @ tsc, tsc} triple provided by the host is also correct. (PTP_SYS_OFFSET_PRECISE can't have any problems with the host that PTP_SYS_OFFSET wouldn't also have.) ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-18 15:20 ` Radim Krcmar @ 2017-01-18 15:28 ` Marcelo Tosatti 2017-01-20 14:18 ` Radim Krcmar 0 siblings, 1 reply; 54+ messages in thread From: Marcelo Tosatti @ 2017-01-18 15:28 UTC (permalink / raw) To: Radim Krcmar Cc: Paolo Bonzini, Miroslav Lichvar, kvm, linux-kernel, Richard Cochran On Wed, Jan 18, 2017 at 04:20:33PM +0100, Radim Krcmar wrote: > 2017-01-18 12:53-0200, Marcelo Tosatti: > > GOn Wed, Jan 18, 2017 at 12:37:25PM -0200, Marcelo Tosatti wrote: > > > On Wed, Jan 18, 2017 at 01:46:58PM +0100, Paolo Bonzini wrote: > > > > > > > > > > > > On 18/01/2017 13:24, Marcelo Tosatti wrote: > > > > > On Wed, Jan 18, 2017 at 10:17:38AM -0200, Marcelo Tosatti wrote: > > > > >> On Tue, Jan 17, 2017 at 04:36:21PM +0100, Radim Krcmar wrote: > > > > >>> 2017-01-17 09:30-0200, Marcelo Tosatti: > > > > >>>> On Tue, Jan 17, 2017 at 09:03:27AM +0100, Miroslav Lichvar wrote: > > > > >>>>> Users of the PTP_SYS_OFFSET ioctl assume that (ts[0]+ts[2])/2 > > > > >>>>> corresponds to ts[1], (ts[2]+ts[4])/2 corresponds to ts[3], and so on. > > > > >>>>> > > > > >>>>> ts[1] ts[3] > > > > >>>>> Host time ---------+---------+........ > > > > >>>>> | | > > > > >>>>> | | > > > > >>>>> Guest time ----+---------+---------+...... > > > > >>>>> ts[0] ts[2] ts[4] > > > > >>> > > > > >>> KVM PTP delay moves host ts[i] to be close to guest ts[i+1] and makes > > > > >>> the offset very consistent, so the graph would look like: > > > > >>> > > > > >>> ts[1] ts[3] > > > > >>> Host time -------------+---------+........ > > > > >>> | | > > > > >>> | | > > > > >>> Guest time ----+---------+---------+...... > > > > >>> ts[0] ts[2] ts[4] > > > > >>> > > > > >>> which doesn't sound good if users assume that the host reading is in the > > > > >>> middle -- the guest time would be ahead of the host time. > > > > >> > > > > >> Testcase: run a guest and a loop sending SIGUSR1 to vcpu0 (emulating > > > > >> intense interrupts). Follows results: > > > > >> > > > > >> Without TSC delta calculation: > > > > >> ============================= > > > > >> > > > > >> #* PHC0 0 3 377 2 -99ns[ +206ns] +/- 116ns > > > > >> #* PHC0 0 3 377 8 +202ns[ +249ns] +/- 111ns > > > > >> #* PHC0 0 3 377 8 -213ns[ +683ns] +/- 88ns > > > > >> #* PHC0 0 3 377 6 +77ns[ +319ns] +/- 56ns > > > > >> #* PHC0 0 3 377 4 -771ns[-1029ns] +/- 93ns > > > > >> #* PHC0 0 3 377 10 -49ns[ -58ns] +/- 121ns > > > > >> #* PHC0 0 3 377 9 +562ns[ +703ns] +/- 107ns > > > > >> #* PHC0 0 3 377 6 -2ns[ -3ns] +/- 94ns > > > > >> #* PHC0 0 3 377 4 +451ns[ +494ns] +/- 138ns > > > > >> #* PHC0 0 3 377 11 -67ns[ -74ns] +/- 113ns > > > > >> #* PHC0 0 3 377 8 +244ns[ +264ns] +/- 119ns > > > > >> #* PHC0 0 3 377 7 -696ns[ -890ns] +/- 89ns > > > > >> #* PHC0 0 3 377 4 +468ns[ +560ns] +/- 110ns > > > > >> #* PHC0 0 3 377 11 -310ns[ -430ns] +/- 72ns > > > > >> #* PHC0 0 3 377 9 +189ns[ +298ns] +/- 54ns > > > > >> #* PHC0 0 3 377 7 +594ns[ +473ns] +/- 96ns > > > > >> #* PHC0 0 3 377 5 +151ns[ +280ns] +/- 71ns > > > > >> #* PHC0 0 3 377 10 -590ns[ -696ns] +/- 94ns > > > > >> #* PHC0 0 3 377 8 +415ns[ +526ns] +/- 74ns > > > > >> #* PHC0 0 3 377 6 +1381ns[+1469ns] +/- 101ns > > > > >> #* PHC0 0 3 377 4 +571ns[+1304ns] +/- 54ns > > > > >> #* PHC0 0 3 377 8 -5ns[ +71ns] +/- 139ns > > > > >> #* PHC0 0 3 377 7 -247ns[ -502ns] +/- 69ns > > > > >> #* PHC0 0 3 377 5 -283ns[ +879ns] +/- 73ns > > > > >> #* PHC0 0 3 377 3 +148ns[ -109ns] +/- 61ns > > > > >> > > > > >> With TSC delta calculation: > > > > >> ============================ > > > > >> > > > > >> #* PHC0 0 3 377 7 +379ns[ +432ns] +/- 53ns > > > > >> #* PHC0 0 3 377 9 +106ns[ +420ns] +/- 42ns > > > > >> #* PHC0 0 3 377 7 -58ns[ -136ns] +/- 62ns > > > > >> #* PHC0 0 3 377 12 +93ns[ -38ns] +/- 64ns > > > > >> #* PHC0 0 3 377 8 +84ns[ +107ns] +/- 69ns > > > > >> #* PHC0 0 3 377 3 -76ns[ -103ns] +/- 52ns > > > > >> #* PHC0 0 3 377 7 +52ns[ +63ns] +/- 50ns > > > > >> #* PHC0 0 3 377 11 +29ns[ +31ns] +/- 70ns > > > > >> #* PHC0 0 3 377 7 -47ns[ -56ns] +/- 42ns > > > > >> #* PHC0 0 3 377 10 -35ns[ -42ns] +/- 33ns > > > > >> #* PHC0 0 3 377 7 -32ns[ -34ns] +/- 42ns > > > > >> #* PHC0 0 3 377 11 -172ns[ -173ns] +/- 118ns > > > > >> #* PHC0 0 3 377 6 +65ns[ +76ns] +/- 23ns > > > > >> #* PHC0 0 3 377 9 +18ns[ +23ns] +/- 37ns > > > > >> #* PHC0 0 3 377 6 +41ns[ -60ns] +/- 30ns > > > > >> #* PHC0 0 3 377 10 +39ns[ +183ns] +/- 42ns > > > > >> #* PHC0 0 3 377 6 +50ns[ +102ns] +/- 86ns > > > > >> #* PHC0 0 3 377 11 +50ns[ +75ns] +/- 52ns > > > > >> #* PHC0 0 3 377 6 +50ns[ +116ns] +/- 100ns > > > > >> #* PHC0 0 3 377 10 +46ns[ +65ns] +/- 79ns > > > > >> #* PHC0 0 3 377 7 -38ns[ -51ns] +/- 29ns > > > > >> #* PHC0 0 3 377 10 -11ns[ -12ns] +/- 32ns > > > > >> #* PHC0 0 3 377 7 -31ns[ -32ns] +/- 99ns > > > > >> #* PHC0 0 3 377 10 +222ns[ +238ns] +/- 58ns > > > > >> #* PHC0 0 3 377 6 +185ns[ +207ns] +/- 39ns > > > > >> #* PHC0 0 3 377 10 -392ns[ -394ns] +/- 118ns > > > > >> #* PHC0 0 3 377 6 -9ns[ -50ns] +/- 35ns > > > > >> #* PHC0 0 3 377 10 -346ns[ -355ns] +/- 111ns > > > > >> > > > > >> > > > > >> Do you still want to drop it in favour of simplicity? > > > > > > > > > > This is the output of "chronyc sources". See section "Time sources" > > > > > of https://chrony.tuxfamily.org/doc/2.4/chronyc.html. > > > > > > > > It's just that it's not obvious why you get better results with biased > > > > host timestamps. What makes the biased host timestamp more precise? > > > > > > > > I'd rather use PTP_SYS_OFFSET_PRECISE instead, but unfortunately chrony > > > > does not support it---but I would still prefer you to support > > > > PTP_SYS_OFFSET_PRECISE as well. > > > > > > A single TSC read could be used to implement the PRECISE ioctl, but if > > > a timer interrupt takes place on either the host or the guest, and that > > > timer interrupt "adds" the TSC delta to xtime.nsec/xtime.sec, then that > > > single TSC read cannot be used. > > > > > > So you would have to stop timer interrupts (in guest and host) for the duration of the > > > PRECISE ioctl in the guest to avoid that situation, which seems a bit > > > overkill to me. > > > > > > Any other ideas? > > > > Could have a hypercall that disables host timer interrupts for > > a specified amount of time... But that does not scale with multiple VMs. > > No need to disable interrupts on guest nor host as both protect the time > by a seqlock. Still not scalable with multiple VMs... so need a different solution. > Do timer interrupts always advance the time base? No only when a full "cycle_interval" has been accumulated. > (We could disable guest interrupts if they were too disruptive because > of the long hypercall, but the host is fine in any case.) VMs can harm host interrupt processing, don't see that as being OK (even when rate limited does scale to high number of VMs). > The guest protects tsc->sec/nsec by tk_core.seq, so we can make sure > that the guest computed sec/nsec from the host tsc read from hypercall > is consistent. Doh, of course. > The {sec @ tsc, nsec @ tsc, tsc} triple provided by the host is also > correct. (PTP_SYS_OFFSET_PRECISE can't have any problems with the host > that PTP_SYS_OFFSET wouldn't also have.) Yep, should work, lets try. ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-18 15:28 ` Marcelo Tosatti @ 2017-01-20 14:18 ` Radim Krcmar 0 siblings, 0 replies; 54+ messages in thread From: Radim Krcmar @ 2017-01-20 14:18 UTC (permalink / raw) To: Marcelo Tosatti Cc: Paolo Bonzini, Miroslav Lichvar, kvm, linux-kernel, Richard Cochran 2017-01-18 13:28-0200, Marcelo Tosatti: > On Wed, Jan 18, 2017 at 04:20:33PM +0100, Radim Krcmar wrote: >> 2017-01-18 12:53-0200, Marcelo Tosatti: >> > GOn Wed, Jan 18, 2017 at 12:37:25PM -0200, Marcelo Tosatti wrote: >> > > On Wed, Jan 18, 2017 at 01:46:58PM +0100, Paolo Bonzini wrote: >> > > > >> > > > >> > > > On 18/01/2017 13:24, Marcelo Tosatti wrote: >> > > > > On Wed, Jan 18, 2017 at 10:17:38AM -0200, Marcelo Tosatti wrote: >> > > > >> On Tue, Jan 17, 2017 at 04:36:21PM +0100, Radim Krcmar wrote: >> > > > >>> 2017-01-17 09:30-0200, Marcelo Tosatti: >> > > > >>>> On Tue, Jan 17, 2017 at 09:03:27AM +0100, Miroslav Lichvar wrote: >> > > > >>>>> Users of the PTP_SYS_OFFSET ioctl assume that (ts[0]+ts[2])/2 >> > > > >>>>> corresponds to ts[1], (ts[2]+ts[4])/2 corresponds to ts[3], and so on. >> > > > >>>>> >> > > > >>>>> ts[1] ts[3] >> > > > >>>>> Host time ---------+---------+........ >> > > > >>>>> | | >> > > > >>>>> | | >> > > > >>>>> Guest time ----+---------+---------+...... >> > > > >>>>> ts[0] ts[2] ts[4] >> > > > >>> >> > > > >>> KVM PTP delay moves host ts[i] to be close to guest ts[i+1] and makes >> > > > >>> the offset very consistent, so the graph would look like: >> > > > >>> >> > > > >>> ts[1] ts[3] >> > > > >>> Host time -------------+---------+........ >> > > > >>> | | >> > > > >>> | | >> > > > >>> Guest time ----+---------+---------+...... >> > > > >>> ts[0] ts[2] ts[4] >> > > > >>> >> > > > >>> which doesn't sound good if users assume that the host reading is in the >> > > > >>> middle -- the guest time would be ahead of the host time. >> > > > >> >> > > > >> Testcase: run a guest and a loop sending SIGUSR1 to vcpu0 (emulating >> > > > >> intense interrupts). Follows results: >> > > > >> >> > > > >> Without TSC delta calculation: >> > > > >> ============================= >> > > > >> >> > > > >> #* PHC0 0 3 377 2 -99ns[ +206ns] +/- 116ns >> > > > >> #* PHC0 0 3 377 8 +202ns[ +249ns] +/- 111ns >> > > > >> #* PHC0 0 3 377 8 -213ns[ +683ns] +/- 88ns >> > > > >> #* PHC0 0 3 377 6 +77ns[ +319ns] +/- 56ns >> > > > >> #* PHC0 0 3 377 4 -771ns[-1029ns] +/- 93ns >> > > > >> #* PHC0 0 3 377 10 -49ns[ -58ns] +/- 121ns >> > > > >> #* PHC0 0 3 377 9 +562ns[ +703ns] +/- 107ns >> > > > >> #* PHC0 0 3 377 6 -2ns[ -3ns] +/- 94ns >> > > > >> #* PHC0 0 3 377 4 +451ns[ +494ns] +/- 138ns >> > > > >> #* PHC0 0 3 377 11 -67ns[ -74ns] +/- 113ns >> > > > >> #* PHC0 0 3 377 8 +244ns[ +264ns] +/- 119ns >> > > > >> #* PHC0 0 3 377 7 -696ns[ -890ns] +/- 89ns >> > > > >> #* PHC0 0 3 377 4 +468ns[ +560ns] +/- 110ns >> > > > >> #* PHC0 0 3 377 11 -310ns[ -430ns] +/- 72ns >> > > > >> #* PHC0 0 3 377 9 +189ns[ +298ns] +/- 54ns >> > > > >> #* PHC0 0 3 377 7 +594ns[ +473ns] +/- 96ns >> > > > >> #* PHC0 0 3 377 5 +151ns[ +280ns] +/- 71ns >> > > > >> #* PHC0 0 3 377 10 -590ns[ -696ns] +/- 94ns >> > > > >> #* PHC0 0 3 377 8 +415ns[ +526ns] +/- 74ns >> > > > >> #* PHC0 0 3 377 6 +1381ns[+1469ns] +/- 101ns >> > > > >> #* PHC0 0 3 377 4 +571ns[+1304ns] +/- 54ns >> > > > >> #* PHC0 0 3 377 8 -5ns[ +71ns] +/- 139ns >> > > > >> #* PHC0 0 3 377 7 -247ns[ -502ns] +/- 69ns >> > > > >> #* PHC0 0 3 377 5 -283ns[ +879ns] +/- 73ns >> > > > >> #* PHC0 0 3 377 3 +148ns[ -109ns] +/- 61ns >> > > > >> >> > > > >> With TSC delta calculation: >> > > > >> ============================ >> > > > >> >> > > > >> #* PHC0 0 3 377 7 +379ns[ +432ns] +/- 53ns >> > > > >> #* PHC0 0 3 377 9 +106ns[ +420ns] +/- 42ns >> > > > >> #* PHC0 0 3 377 7 -58ns[ -136ns] +/- 62ns >> > > > >> #* PHC0 0 3 377 12 +93ns[ -38ns] +/- 64ns >> > > > >> #* PHC0 0 3 377 8 +84ns[ +107ns] +/- 69ns >> > > > >> #* PHC0 0 3 377 3 -76ns[ -103ns] +/- 52ns >> > > > >> #* PHC0 0 3 377 7 +52ns[ +63ns] +/- 50ns >> > > > >> #* PHC0 0 3 377 11 +29ns[ +31ns] +/- 70ns >> > > > >> #* PHC0 0 3 377 7 -47ns[ -56ns] +/- 42ns >> > > > >> #* PHC0 0 3 377 10 -35ns[ -42ns] +/- 33ns >> > > > >> #* PHC0 0 3 377 7 -32ns[ -34ns] +/- 42ns >> > > > >> #* PHC0 0 3 377 11 -172ns[ -173ns] +/- 118ns >> > > > >> #* PHC0 0 3 377 6 +65ns[ +76ns] +/- 23ns >> > > > >> #* PHC0 0 3 377 9 +18ns[ +23ns] +/- 37ns >> > > > >> #* PHC0 0 3 377 6 +41ns[ -60ns] +/- 30ns >> > > > >> #* PHC0 0 3 377 10 +39ns[ +183ns] +/- 42ns >> > > > >> #* PHC0 0 3 377 6 +50ns[ +102ns] +/- 86ns >> > > > >> #* PHC0 0 3 377 11 +50ns[ +75ns] +/- 52ns >> > > > >> #* PHC0 0 3 377 6 +50ns[ +116ns] +/- 100ns >> > > > >> #* PHC0 0 3 377 10 +46ns[ +65ns] +/- 79ns >> > > > >> #* PHC0 0 3 377 7 -38ns[ -51ns] +/- 29ns >> > > > >> #* PHC0 0 3 377 10 -11ns[ -12ns] +/- 32ns >> > > > >> #* PHC0 0 3 377 7 -31ns[ -32ns] +/- 99ns >> > > > >> #* PHC0 0 3 377 10 +222ns[ +238ns] +/- 58ns >> > > > >> #* PHC0 0 3 377 6 +185ns[ +207ns] +/- 39ns >> > > > >> #* PHC0 0 3 377 10 -392ns[ -394ns] +/- 118ns >> > > > >> #* PHC0 0 3 377 6 -9ns[ -50ns] +/- 35ns >> > > > >> #* PHC0 0 3 377 10 -346ns[ -355ns] +/- 111ns >> > > > >> >> > > > >> >> > > > >> Do you still want to drop it in favour of simplicity? >> > > > > >> > > > > This is the output of "chronyc sources". See section "Time sources" >> > > > > of https://chrony.tuxfamily.org/doc/2.4/chronyc.html. >> > > > >> > > > It's just that it's not obvious why you get better results with biased >> > > > host timestamps. What makes the biased host timestamp more precise? >> > > > >> > > > I'd rather use PTP_SYS_OFFSET_PRECISE instead, but unfortunately chrony >> > > > does not support it---but I would still prefer you to support >> > > > PTP_SYS_OFFSET_PRECISE as well. >> > > >> > > A single TSC read could be used to implement the PRECISE ioctl, but if >> > > a timer interrupt takes place on either the host or the guest, and that >> > > timer interrupt "adds" the TSC delta to xtime.nsec/xtime.sec, then that >> > > single TSC read cannot be used. >> > > >> > > So you would have to stop timer interrupts (in guest and host) for the duration of the >> > > PRECISE ioctl in the guest to avoid that situation, which seems a bit >> > > overkill to me. >> > > >> > > Any other ideas? >> > >> > Could have a hypercall that disables host timer interrupts for >> > a specified amount of time... But that does not scale with multiple VMs. >> >> No need to disable interrupts on guest nor host as both protect the time >> by a seqlock. > > Still not scalable with multiple VMs... so need a different solution. What doesn't scale? The VM hypercall takes read on the tk_core.seq, which doesn't block other VMs from doing the same. ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-18 12:46 ` Paolo Bonzini ` (2 preceding siblings ...) 2017-01-18 14:37 ` Marcelo Tosatti @ 2017-01-18 15:59 ` Radim Krcmar 3 siblings, 0 replies; 54+ messages in thread From: Radim Krcmar @ 2017-01-18 15:59 UTC (permalink / raw) To: Paolo Bonzini Cc: Marcelo Tosatti, Miroslav Lichvar, kvm, linux-kernel, Richard Cochran 2017-01-18 13:46+0100, Paolo Bonzini: > > > On 18/01/2017 13:24, Marcelo Tosatti wrote: > > On Wed, Jan 18, 2017 at 10:17:38AM -0200, Marcelo Tosatti wrote: > >> On Tue, Jan 17, 2017 at 04:36:21PM +0100, Radim Krcmar wrote: > >>> 2017-01-17 09:30-0200, Marcelo Tosatti: > >>>> On Tue, Jan 17, 2017 at 09:03:27AM +0100, Miroslav Lichvar wrote: > >>>>> Users of the PTP_SYS_OFFSET ioctl assume that (ts[0]+ts[2])/2 > >>>>> corresponds to ts[1], (ts[2]+ts[4])/2 corresponds to ts[3], and so on. > >>>>> > >>>>> ts[1] ts[3] > >>>>> Host time ---------+---------+........ > >>>>> | | > >>>>> | | > >>>>> Guest time ----+---------+---------+...... > >>>>> ts[0] ts[2] ts[4] > >>> > >>> KVM PTP delay moves host ts[i] to be close to guest ts[i+1] and makes > >>> the offset very consistent, so the graph would look like: > >>> > >>> ts[1] ts[3] > >>> Host time -------------+---------+........ > >>> | | > >>> | | > >>> Guest time ----+---------+---------+...... > >>> ts[0] ts[2] ts[4] > >>> > >>> which doesn't sound good if users assume that the host reading is in the > >>> middle -- the guest time would be ahead of the host time. > >> > >> Testcase: run a guest and a loop sending SIGUSR1 to vcpu0 (emulating > >> intense interrupts). Follows results: > >> > >> Without TSC delta calculation: > >> ============================= > >> > >> #* PHC0 0 3 377 2 -99ns[ +206ns] +/- 116ns > >> #* PHC0 0 3 377 8 +202ns[ +249ns] +/- 111ns > >> #* PHC0 0 3 377 8 -213ns[ +683ns] +/- 88ns > >> #* PHC0 0 3 377 6 +77ns[ +319ns] +/- 56ns > >> #* PHC0 0 3 377 4 -771ns[-1029ns] +/- 93ns > >> #* PHC0 0 3 377 10 -49ns[ -58ns] +/- 121ns > >> #* PHC0 0 3 377 9 +562ns[ +703ns] +/- 107ns > >> #* PHC0 0 3 377 6 -2ns[ -3ns] +/- 94ns > >> #* PHC0 0 3 377 4 +451ns[ +494ns] +/- 138ns > >> #* PHC0 0 3 377 11 -67ns[ -74ns] +/- 113ns > >> #* PHC0 0 3 377 8 +244ns[ +264ns] +/- 119ns > >> #* PHC0 0 3 377 7 -696ns[ -890ns] +/- 89ns > >> #* PHC0 0 3 377 4 +468ns[ +560ns] +/- 110ns > >> #* PHC0 0 3 377 11 -310ns[ -430ns] +/- 72ns > >> #* PHC0 0 3 377 9 +189ns[ +298ns] +/- 54ns > >> #* PHC0 0 3 377 7 +594ns[ +473ns] +/- 96ns > >> #* PHC0 0 3 377 5 +151ns[ +280ns] +/- 71ns > >> #* PHC0 0 3 377 10 -590ns[ -696ns] +/- 94ns > >> #* PHC0 0 3 377 8 +415ns[ +526ns] +/- 74ns > >> #* PHC0 0 3 377 6 +1381ns[+1469ns] +/- 101ns > >> #* PHC0 0 3 377 4 +571ns[+1304ns] +/- 54ns > >> #* PHC0 0 3 377 8 -5ns[ +71ns] +/- 139ns > >> #* PHC0 0 3 377 7 -247ns[ -502ns] +/- 69ns > >> #* PHC0 0 3 377 5 -283ns[ +879ns] +/- 73ns > >> #* PHC0 0 3 377 3 +148ns[ -109ns] +/- 61ns > >> > >> With TSC delta calculation: > >> ============================ > >> > >> #* PHC0 0 3 377 7 +379ns[ +432ns] +/- 53ns > >> #* PHC0 0 3 377 9 +106ns[ +420ns] +/- 42ns > >> #* PHC0 0 3 377 7 -58ns[ -136ns] +/- 62ns > >> #* PHC0 0 3 377 12 +93ns[ -38ns] +/- 64ns > >> #* PHC0 0 3 377 8 +84ns[ +107ns] +/- 69ns > >> #* PHC0 0 3 377 3 -76ns[ -103ns] +/- 52ns > >> #* PHC0 0 3 377 7 +52ns[ +63ns] +/- 50ns > >> #* PHC0 0 3 377 11 +29ns[ +31ns] +/- 70ns > >> #* PHC0 0 3 377 7 -47ns[ -56ns] +/- 42ns > >> #* PHC0 0 3 377 10 -35ns[ -42ns] +/- 33ns > >> #* PHC0 0 3 377 7 -32ns[ -34ns] +/- 42ns > >> #* PHC0 0 3 377 11 -172ns[ -173ns] +/- 118ns > >> #* PHC0 0 3 377 6 +65ns[ +76ns] +/- 23ns > >> #* PHC0 0 3 377 9 +18ns[ +23ns] +/- 37ns > >> #* PHC0 0 3 377 6 +41ns[ -60ns] +/- 30ns > >> #* PHC0 0 3 377 10 +39ns[ +183ns] +/- 42ns > >> #* PHC0 0 3 377 6 +50ns[ +102ns] +/- 86ns > >> #* PHC0 0 3 377 11 +50ns[ +75ns] +/- 52ns > >> #* PHC0 0 3 377 6 +50ns[ +116ns] +/- 100ns > >> #* PHC0 0 3 377 10 +46ns[ +65ns] +/- 79ns > >> #* PHC0 0 3 377 7 -38ns[ -51ns] +/- 29ns > >> #* PHC0 0 3 377 10 -11ns[ -12ns] +/- 32ns > >> #* PHC0 0 3 377 7 -31ns[ -32ns] +/- 99ns > >> #* PHC0 0 3 377 10 +222ns[ +238ns] +/- 58ns > >> #* PHC0 0 3 377 6 +185ns[ +207ns] +/- 39ns > >> #* PHC0 0 3 377 10 -392ns[ -394ns] +/- 118ns > >> #* PHC0 0 3 377 6 -9ns[ -50ns] +/- 35ns > >> #* PHC0 0 3 377 10 -346ns[ -355ns] +/- 111ns > >> > >> > >> Do you still want to drop it in favour of simplicity? > > > > This is the output of "chronyc sources". See section "Time sources" > > of https://chrony.tuxfamily.org/doc/2.4/chronyc.html. > > It's just that it's not obvious why you get better results with biased > host timestamps. What makes the biased host timestamp more precise? I think that TSC offsetting is more precise (the difference between two samples smaller), but less accurate (the difference between guest and host clock is bigger). And chronyc can't tell how much off from the host clock our guest clock is (that would be measured how I wrote above). The main reason why it is more precise is because the TSC offset erases the effect of many latency sources that can happen before or after the host time read. The precision with TSC offsetting is almost only affected by the duration between two guest time measurements, because the host time is shifted to be very close to the latter guest time measurement. Without TSC offsetting, the host time will be closer to the middle between two guest reads (= better accuracy), but the moment of the tsc read on the host has more variables -- host interrupts and whatnot, which will result in lower precision. > I'd rather use PTP_SYS_OFFSET_PRECISE instead, but unfortunately chrony > does not support it---but I would still prefer you to support > PTP_SYS_OFFSET_PRECISE as well. Oh, I didn't expect that userspace can't use PTP_SYS_OFFSET_PRECISE ... ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-16 16:26 ` Radim Krcmar 2017-01-16 16:54 ` Radim Krcmar @ 2017-01-16 17:04 ` Marcelo Tosatti 2017-01-16 17:46 ` Radim Krcmar 1 sibling, 1 reply; 54+ messages in thread From: Marcelo Tosatti @ 2017-01-16 17:04 UTC (permalink / raw) To: Radim Krcmar Cc: kvm, linux-kernel, Paolo Bonzini, Richard Cochran, Miroslav Lichvar On Mon, Jan 16, 2017 at 05:26:53PM +0100, Radim Krcmar wrote: > 2017-01-13 15:40-0200, Marcelo Tosatti: > > On Fri, Jan 13, 2017 at 04:56:58PM +0100, Radim Krcmar wrote: > > > 2017-01-13 10:01-0200, Marcelo Tosatti: > >> > + version = pvclock_read_begin(src); > >> > + > >> > + ret = kvm_hypercall2(KVM_HC_CLOCK_OFFSET, > >> > + clock_off_gpa, > >> > + KVM_CLOCK_OFFSET_WALLCLOCK); > >> > + if (ret != 0) { > >> > + pr_err("clock offset hypercall ret %lu\n", ret); > >> > + spin_unlock(&kvm_ptp_lock); > >> > + preempt_enable_notrace(); > >> > + return -EOPNOTSUPP; > >> > + } > >> > + > >> > + tspec.tv_sec = clock_off.sec; > >> > + tspec.tv_nsec = clock_off.nsec; > >> > + > >> > + delta = rdtsc_ordered() - clock_off.tsc; > >> > + > >> > + offset = pvclock_scale_delta(delta, src->tsc_to_system_mul, > >> > + src->tsc_shift); > >> > + > >> > + } while (pvclock_read_retry(src, version)); > >> > + > >> > + preempt_enable_notrace(); > >> > + > >> > + tspec.tv_nsec = tspec.tv_nsec + offset; > >> > + > >> > + spin_unlock(&kvm_ptp_lock); > >> > + > >> > + if (tspec.tv_nsec >= NSEC_PER_SEC) { > >> > + u64 secs = tspec.tv_nsec; > >> > + > >> > + tspec.tv_nsec = do_div(secs, NSEC_PER_SEC); > >> > + tspec.tv_sec += secs; > >> > + } > >> > + > >> > + memcpy(ts, &tspec, sizeof(struct timespec64)); > >> > >> But the whole idea is of improving the time by reading tsc a bit later > >> is just weird ... why is it better to provide > >> > >> tsc + x, time + tsc_delta_to_time(x) > >> > >> than just > >> > >> tsc, time > >> > >> ? > > > > Because you want to calculate the value of the host realtime clock > > at the moment of ptp_kvm_gettime. > > > > We do: > > > > 1. kvm_hypercall. > > 2. get {sec, nsec, guest_tsc}. > > 3. kvm_hypercall returns. > > 4. delay = rdtsc() - guest_tsc. > > > > Where delay is the delta (measured with the TSC) between points 2 and 4. > > I see now ... the PTP interface is just not good for our purposes. > We don't return {sec, nsec, guest_tsc}, we just return {sec, nsec} at > some random time in the past. And to make it a bit more accurate, you > add a best-effort delta before returning, which makes sense. Not random time in the past. We return {sec, nsec} from the host realtime at the moment the user ran the hypercall. Since PTP is very accurate, that "a bit more" counts, yes. > When we have to depend on pvclock, what are the advantages of not using > the existing pvclock API for wall clock? > (You mentioned some extensions.) > > struct pvclock_wall_clock { > u32 version; > u32 sec; > u32 nsec; > } __attribute__((__packed__)); > It gives the wall clock when pvclock was 0, so you just add current > kvmclock and get the host wall clock. Well, no. For one, the TSC part of kvmclock: kvmclock-read = system_timestamp + convert-to-1GHz(rdtsc() - tsc_timestamp) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Drifts relative to UTC. This part can be large. The guests NTP is responsible for fixing that drift of the guests realtime clock (talking about current setup, without KVM PTP driver). Now, we want very high precision (less than 1us) for this driver. Very small TSC drifts on a large delta defeat the purpose. > Without a VM exit. Huge performance is not an issue. Accuracy (how different from the host realtime clock our "approximation" of the host realtime clock) is. > And how often is ptp_kvm_gettime() usually called? The PTP_SYS_OFFSET ioctl calls the following code in a loop: struct ptp_sys_offset { unsigned int n_samples; /* Desired number of measurements. */ unsigned int rsv[3]; /* Reserved for future use. */ /* * Array of interleaved system/phc time stamps. The kernel * will provide 2*n_samples + 1 time stamps, with the last * one as a system time stamp. */ struct ptp_clock_time ts[2 * PTP_MAX_SAMPLES + 1]; }; #define PTP_MAX_SAMPLES 25 /* Maximum allowed offset measurement samples. */ case PTP_SYS_OFFSET: sysoff = memdup_user((void __user *)arg, sizeof(*sysoff)); if (IS_ERR(sysoff)) { err = PTR_ERR(sysoff); sysoff = NULL; break; } if (sysoff->n_samples > PTP_MAX_SAMPLES) { err = -EINVAL; break; } pct = &sysoff->ts[0]; for (i = 0; i < sysoff->n_samples; i++) { getnstimeofday64(&ts); pct->sec = ts.tv_sec; pct->nsec = ts.tv_nsec; pct++; ptp->info->gettime64(ptp->info, &ts); pct->sec = ts.tv_sec; pct->nsec = ts.tv_nsec; pct++; } getnstimeofday64(&ts); pct->sec = ts.tv_sec; pct->nsec = ts.tv_nsec; How often that ioctl is called depends on the parameters of the Chrony PHC code. Initially (to determine the clock difference Chrony should call it more frequently, later on it should call it less frequency). Perhaps once every second initially (the ioctl). I'll confirm with the exact value for my setup and reply to this email. > > Thanks. > > >> Because we'll always be quering the time at tsc + y, where y >> x, and > >> we'd likely have other problems if shifting the time base by few > >> thousand cycles made a difference. > > > > Radim, i didnt get your "tsc + x", "time + tsc_delta_to_time(x)" > > formulas above. Can you be more verbose please? > > x is the delta, tsc_delta_to_time() is what pvclock_scale_delta() does. > > I assumed that we set precise time with TSC, so the delta wouldn't > matter, because PTP would either get {sec, nsec, guest_tsc}, or the > same, but just shifted by delta, hence > {sec + tsc_delta_to_time(x) / NSEC_PER_SEC, > nsec + tsc_delta_to_time(x) % NSEC_PER_SEC, > guest_tsc + x}. Ah, OK. I understand you now understood the meaning of "tsc" part of the {sec, nsec, guest_tsc} triple. ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-16 17:04 ` Marcelo Tosatti @ 2017-01-16 17:46 ` Radim Krcmar 2017-01-16 19:33 ` Marcelo Tosatti 0 siblings, 1 reply; 54+ messages in thread From: Radim Krcmar @ 2017-01-16 17:46 UTC (permalink / raw) To: Marcelo Tosatti Cc: kvm, linux-kernel, Paolo Bonzini, Richard Cochran, Miroslav Lichvar 2017-01-16 15:04-0200, Marcelo Tosatti: > On Mon, Jan 16, 2017 at 05:26:53PM +0100, Radim Krcmar wrote: >> 2017-01-13 15:40-0200, Marcelo Tosatti: >> > On Fri, Jan 13, 2017 at 04:56:58PM +0100, Radim Krcmar wrote: >> > > 2017-01-13 10:01-0200, Marcelo Tosatti: >> >> > + version = pvclock_read_begin(src); >> >> > + >> >> > + ret = kvm_hypercall2(KVM_HC_CLOCK_OFFSET, >> >> > + clock_off_gpa, >> >> > + KVM_CLOCK_OFFSET_WALLCLOCK); >> >> > + if (ret != 0) { >> >> > + pr_err("clock offset hypercall ret %lu\n", ret); >> >> > + spin_unlock(&kvm_ptp_lock); >> >> > + preempt_enable_notrace(); >> >> > + return -EOPNOTSUPP; >> >> > + } >> >> > + >> >> > + tspec.tv_sec = clock_off.sec; >> >> > + tspec.tv_nsec = clock_off.nsec; >> >> > + >> >> > + delta = rdtsc_ordered() - clock_off.tsc; >> >> > + >> >> > + offset = pvclock_scale_delta(delta, src->tsc_to_system_mul, >> >> > + src->tsc_shift); >> >> > + >> >> > + } while (pvclock_read_retry(src, version)); >> >> > + >> >> > + preempt_enable_notrace(); >> >> > + >> >> > + tspec.tv_nsec = tspec.tv_nsec + offset; >> >> > + >> >> > + spin_unlock(&kvm_ptp_lock); >> >> > + >> >> > + if (tspec.tv_nsec >= NSEC_PER_SEC) { >> >> > + u64 secs = tspec.tv_nsec; >> >> > + >> >> > + tspec.tv_nsec = do_div(secs, NSEC_PER_SEC); >> >> > + tspec.tv_sec += secs; >> >> > + } >> >> > + >> >> > + memcpy(ts, &tspec, sizeof(struct timespec64)); >> >> >> >> But the whole idea is of improving the time by reading tsc a bit later >> >> is just weird ... why is it better to provide >> >> >> >> tsc + x, time + tsc_delta_to_time(x) >> >> >> >> than just >> >> >> >> tsc, time >> >> >> >> ? >> > >> > Because you want to calculate the value of the host realtime clock >> > at the moment of ptp_kvm_gettime. >> > >> > We do: >> > >> > 1. kvm_hypercall. >> > 2. get {sec, nsec, guest_tsc}. >> > 3. kvm_hypercall returns. >> > 4. delay = rdtsc() - guest_tsc. >> > >> > Where delay is the delta (measured with the TSC) between points 2 and 4. >> >> I see now ... the PTP interface is just not good for our purposes. >> We don't return {sec, nsec, guest_tsc}, we just return {sec, nsec} at >> some random time in the past. And to make it a bit more accurate, you >> add a best-effort delta before returning, which makes sense. > > Not random time in the past. We return {sec, nsec} from the host > realtime at the moment the user ran the hypercall. That is what we want, but {sec, nsec} is not anchored to any running time, so it is unavoidably late when we return it. > Since PTP is very accurate, that "a bit more" counts, yes. > >> When we have to depend on pvclock, what are the advantages of not using >> the existing pvclock API for wall clock? >> (You mentioned some extensions.) >> >> struct pvclock_wall_clock { >> u32 version; >> u32 sec; >> u32 nsec; >> } __attribute__((__packed__)); > >> It gives the wall clock when pvclock was 0, so you just add current >> kvmclock and get the host wall clock. > > Well, no. For one, the TSC part of kvmclock: > > kvmclock-read = system_timestamp + convert-to-1GHz(rdtsc() - tsc_timestamp) > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > Drifts relative to UTC. This part can be large. > The guests NTP is responsible for fixing > that drift of the guests realtime clock (talking about current setup, > without KVM PTP driver). > > Now, we want very high precision (less than 1us) for this > driver. Very small TSC drifts on a large delta defeat the purpose. True. >> Without a VM exit. > > Huge performance is not an issue. Accuracy (how different from the host > realtime clock our "approximation" of the host realtime clock) is. Ah, ok. >> And how often is ptp_kvm_gettime() usually called? > > The PTP_SYS_OFFSET ioctl calls the following code in a loop: > > struct ptp_sys_offset { > unsigned int n_samples; /* Desired number of measurements. */ > unsigned int rsv[3]; /* Reserved for future use. */ > /* > * Array of interleaved system/phc time stamps. The kernel > * will provide 2*n_samples + 1 time stamps, with the last > * one as a system time stamp. > */ > struct ptp_clock_time ts[2 * PTP_MAX_SAMPLES + 1]; > }; > > #define PTP_MAX_SAMPLES 25 /* Maximum allowed offset measurement > samples. */ > > case PTP_SYS_OFFSET: > sysoff = memdup_user((void __user *)arg, > sizeof(*sysoff)); > if (IS_ERR(sysoff)) { > err = PTR_ERR(sysoff); > sysoff = NULL; > break; > } > if (sysoff->n_samples > PTP_MAX_SAMPLES) { > err = -EINVAL; > break; > } > pct = &sysoff->ts[0]; > for (i = 0; i < sysoff->n_samples; i++) { > getnstimeofday64(&ts); > pct->sec = ts.tv_sec; > pct->nsec = ts.tv_nsec; > pct++; > ptp->info->gettime64(ptp->info, &ts); > pct->sec = ts.tv_sec; > pct->nsec = ts.tv_nsec; > pct++; Hm, this loop alternates between system time and ptp time and I'd guess that userspace then manipulates deltas of these two readings and applies the result to system time. I'm not even sure that pvclock delta improves the situation anymore -- it adds an offset to all gettime64() reads, but that can be computed from multiple datapoints. (And the delta also adds uncertainty to results.) > } > getnstimeofday64(&ts); > pct->sec = ts.tv_sec; > pct->nsec = ts.tv_nsec; > > How often that ioctl is called depends on the parameters of the Chrony > PHC code. Initially (to determine the clock difference Chrony should call it > more frequently, later on it should call it less frequency). > > Perhaps once every second initially (the ioctl). I'll confirm with the > exact value for my setup and reply to this email. Thanks. ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-16 17:46 ` Radim Krcmar @ 2017-01-16 19:33 ` Marcelo Tosatti 0 siblings, 0 replies; 54+ messages in thread From: Marcelo Tosatti @ 2017-01-16 19:33 UTC (permalink / raw) To: Radim Krcmar Cc: kvm, linux-kernel, Paolo Bonzini, Richard Cochran, Miroslav Lichvar [-- Attachment #1: Type: text/plain, Size: 7008 bytes --] On Mon, Jan 16, 2017 at 06:46:49PM +0100, Radim Krcmar wrote: > 2017-01-16 15:04-0200, Marcelo Tosatti: > > On Mon, Jan 16, 2017 at 05:26:53PM +0100, Radim Krcmar wrote: > >> 2017-01-13 15:40-0200, Marcelo Tosatti: > >> > On Fri, Jan 13, 2017 at 04:56:58PM +0100, Radim Krcmar wrote: > >> > > 2017-01-13 10:01-0200, Marcelo Tosatti: > >> >> > + version = pvclock_read_begin(src); > >> >> > + > >> >> > + ret = kvm_hypercall2(KVM_HC_CLOCK_OFFSET, > >> >> > + clock_off_gpa, > >> >> > + KVM_CLOCK_OFFSET_WALLCLOCK); > >> >> > + if (ret != 0) { > >> >> > + pr_err("clock offset hypercall ret %lu\n", ret); > >> >> > + spin_unlock(&kvm_ptp_lock); > >> >> > + preempt_enable_notrace(); > >> >> > + return -EOPNOTSUPP; > >> >> > + } > >> >> > + > >> >> > + tspec.tv_sec = clock_off.sec; > >> >> > + tspec.tv_nsec = clock_off.nsec; > >> >> > + > >> >> > + delta = rdtsc_ordered() - clock_off.tsc; > >> >> > + > >> >> > + offset = pvclock_scale_delta(delta, src->tsc_to_system_mul, > >> >> > + src->tsc_shift); > >> >> > + > >> >> > + } while (pvclock_read_retry(src, version)); > >> >> > + > >> >> > + preempt_enable_notrace(); > >> >> > + > >> >> > + tspec.tv_nsec = tspec.tv_nsec + offset; > >> >> > + > >> >> > + spin_unlock(&kvm_ptp_lock); > >> >> > + > >> >> > + if (tspec.tv_nsec >= NSEC_PER_SEC) { > >> >> > + u64 secs = tspec.tv_nsec; > >> >> > + > >> >> > + tspec.tv_nsec = do_div(secs, NSEC_PER_SEC); > >> >> > + tspec.tv_sec += secs; > >> >> > + } > >> >> > + > >> >> > + memcpy(ts, &tspec, sizeof(struct timespec64)); > >> >> > >> >> But the whole idea is of improving the time by reading tsc a bit later > >> >> is just weird ... why is it better to provide > >> >> > >> >> tsc + x, time + tsc_delta_to_time(x) > >> >> > >> >> than just > >> >> > >> >> tsc, time > >> >> > >> >> ? > >> > > >> > Because you want to calculate the value of the host realtime clock > >> > at the moment of ptp_kvm_gettime. > >> > > >> > We do: > >> > > >> > 1. kvm_hypercall. > >> > 2. get {sec, nsec, guest_tsc}. > >> > 3. kvm_hypercall returns. > >> > 4. delay = rdtsc() - guest_tsc. > >> > > >> > Where delay is the delta (measured with the TSC) between points 2 and 4. > >> > >> I see now ... the PTP interface is just not good for our purposes. > >> We don't return {sec, nsec, guest_tsc}, we just return {sec, nsec} at > >> some random time in the past. And to make it a bit more accurate, you > >> add a best-effort delta before returning, which makes sense. > > > > Not random time in the past. We return {sec, nsec} from the host > > realtime at the moment the user ran the hypercall. > > That is what we want, but {sec, nsec} is not anchored to any running > time, so it is unavoidably late when we return it. > > > Since PTP is very accurate, that "a bit more" counts, yes. > > > >> When we have to depend on pvclock, what are the advantages of not using > >> the existing pvclock API for wall clock? > >> (You mentioned some extensions.) > >> > >> struct pvclock_wall_clock { > >> u32 version; > >> u32 sec; > >> u32 nsec; > >> } __attribute__((__packed__)); > > > >> It gives the wall clock when pvclock was 0, so you just add current > >> kvmclock and get the host wall clock. > > > > Well, no. For one, the TSC part of kvmclock: > > > > kvmclock-read = system_timestamp + convert-to-1GHz(rdtsc() - tsc_timestamp) > > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > > Drifts relative to UTC. This part can be large. > > The guests NTP is responsible for fixing > > that drift of the guests realtime clock (talking about current setup, > > without KVM PTP driver). > > > > Now, we want very high precision (less than 1us) for this > > driver. Very small TSC drifts on a large delta defeat the purpose. > > True. > > >> Without a VM exit. > > > > Huge performance is not an issue. Accuracy (how different from the host > > realtime clock our "approximation" of the host realtime clock) is. > > Ah, ok. > > >> And how often is ptp_kvm_gettime() usually called? > > > > The PTP_SYS_OFFSET ioctl calls the following code in a loop: > > > > struct ptp_sys_offset { > > unsigned int n_samples; /* Desired number of measurements. */ > > unsigned int rsv[3]; /* Reserved for future use. */ > > /* > > * Array of interleaved system/phc time stamps. The kernel > > * will provide 2*n_samples + 1 time stamps, with the last > > * one as a system time stamp. > > */ > > struct ptp_clock_time ts[2 * PTP_MAX_SAMPLES + 1]; > > }; > > > > #define PTP_MAX_SAMPLES 25 /* Maximum allowed offset measurement > > samples. */ > > > > case PTP_SYS_OFFSET: > > sysoff = memdup_user((void __user *)arg, > > sizeof(*sysoff)); > > if (IS_ERR(sysoff)) { > > err = PTR_ERR(sysoff); > > sysoff = NULL; > > break; > > } > > if (sysoff->n_samples > PTP_MAX_SAMPLES) { > > err = -EINVAL; > > break; > > } > > pct = &sysoff->ts[0]; > > for (i = 0; i < sysoff->n_samples; i++) { > > getnstimeofday64(&ts); > > pct->sec = ts.tv_sec; > > pct->nsec = ts.tv_nsec; > > pct++; > > ptp->info->gettime64(ptp->info, &ts); > > pct->sec = ts.tv_sec; > > pct->nsec = ts.tv_nsec; > > pct++; > > Hm, this loop alternates between system time and ptp time and I'd guess > that userspace then manipulates deltas of these two readings and applies > the result to system time. Yes. > I'm not even sure that pvclock delta improves the situation anymore -- > it adds an offset to all gettime64() reads, but that can be computed > from multiple datapoints. (And the delta also adds uncertainty to > results.) Sorry: the value returned from ->gettime64 should be the value of the clock at that instant. If you remove the pvclock delta (lets assume its 2000ns), then you'll sync getnstimeofday64 to a clock thats 2000ns behind the host clock. Does that make sense? > > } > > getnstimeofday64(&ts); > > pct->sec = ts.tv_sec; > > pct->nsec = ts.tv_nsec; > > > > How often that ioctl is called depends on the parameters of the Chrony > > PHC code. Initially (to determine the clock difference Chrony should call it > > more frequently, later on it should call it less frequency). > > > > Perhaps once every second initially (the ioctl). I'll confirm with the > > exact value for my setup and reply to this email. > > Thanks. Attached. 1551 hypercalls in 38 secs is about 40 times per second. The ioctl is calling ->gettime64 10 per ioctl, so thats about 4 ioctls per sec. [-- Attachment #2: trace.txt --] [-- Type: text/plain, Size: 185238 bytes --] # tracer: nop # # entries-in-buffer/entries-written: 1540/1540 #P:4 # # _-----=> irqs-off # / _----=> need-resched # | / _---=> hardirq/softirq # || / _--=> preempt-depth # ||| / delay # TASK-PID CPU# |||| TIMESTAMP FUNCTION # | | | |||| | | qemu-system-x86-1630 [000] .... 612.087050: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.087059: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.087061: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.087064: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.087066: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.087068: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.087070: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.087073: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.087075: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.087077: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.337454: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.337461: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.337463: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.337465: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.337467: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.337469: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.337471: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.337473: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.337474: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.337476: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.587820: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.587825: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.587828: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.587829: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.587831: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.587833: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.587835: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.587836: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.587838: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.587840: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.838183: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.838189: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.838191: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.838193: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.838195: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.838197: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.838198: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.838200: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.838202: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 612.838204: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.088535: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.088541: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.088543: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.088545: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.088547: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.088549: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.088551: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.088553: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.088554: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.088579: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.338897: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.338904: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.338907: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.338909: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.338911: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.338913: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.338915: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.338916: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.338918: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.338920: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.589255: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.589260: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.589262: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.589264: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.589266: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.589268: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.589270: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.589272: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.589274: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.589276: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.839626: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.839633: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.839635: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.839637: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.839639: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.839641: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.839642: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.839645: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.839647: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 613.839649: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 614.089987: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 614.089993: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 614.089995: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 614.089997: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 614.089999: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 614.090001: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 614.090003: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 614.090005: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 614.090007: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 614.090009: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 614.340343: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 614.340348: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 614.340351: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 614.340352: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 614.340354: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 614.340356: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 614.340358: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 614.340360: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 614.340362: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 614.340364: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 614.590713: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 614.590719: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 614.590721: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 614.590722: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 614.590724: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 614.590726: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 614.590728: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 614.590730: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 614.590731: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 614.590733: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 614.841063: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 614.841068: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 614.841070: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 614.841072: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 614.841074: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 614.841076: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 614.841078: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 614.841080: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 614.841082: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 614.841084: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.091415: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.091420: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.091423: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.091425: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.091426: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.091428: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.091430: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.091432: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.091434: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.091436: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.341680: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.341685: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.341687: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.341689: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.341691: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.341693: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.341695: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.341697: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.341699: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.341701: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.592040: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.592045: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.592047: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.592049: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.592051: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.592052: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.592054: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.592056: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.592058: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.592060: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.842388: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.842393: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.842395: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.842397: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.842399: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.842401: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.842403: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.842405: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.842407: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 615.842409: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.092734: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.092739: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.092741: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.092743: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.092745: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.092746: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.092748: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.092750: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.092752: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.092754: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.343097: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.343104: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.343106: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.343108: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.343110: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.343112: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.343114: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.343116: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.343118: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.343120: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.593453: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.593459: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.593461: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.593463: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.593465: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.593467: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.593469: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.593470: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.593472: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.593474: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.843828: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.843835: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.843837: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.843839: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.843841: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.843843: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.843845: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.843847: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.843849: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 616.843850: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.094184: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.094190: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.094192: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.094194: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.094196: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.094198: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.094200: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.094201: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.094203: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.094205: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.344560: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.344565: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.344568: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.344570: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.344572: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.344574: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.344575: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.344577: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.344579: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.344581: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.594934: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.594939: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.594942: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.594944: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.594945: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.594947: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.594949: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.594951: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.594953: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.594955: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.845286: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.845291: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.845294: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.845296: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.845298: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.845299: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.845301: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.845303: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.845305: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 617.845307: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.095669: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.095674: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.095677: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.095678: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.095680: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.095682: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.095684: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.095685: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.095687: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.095689: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.346029: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.346035: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.346037: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.346038: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.346041: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.346043: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.346045: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.346047: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.346048: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.346050: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.596384: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.596389: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.596391: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.596393: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.596395: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.596397: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.596398: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.596400: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.596402: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.596404: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.846733: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.846739: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.846741: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.846743: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.846745: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.846747: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.846749: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.846751: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.846753: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 618.846755: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.097087: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.097092: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.097095: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.097096: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.097098: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.097100: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.097102: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.097104: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.097106: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.097108: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.347439: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.347445: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.347447: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.347449: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.347451: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.347452: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.347454: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.347456: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.347458: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.347460: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.597810: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.597817: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.597820: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.597823: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.597826: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.597829: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.597832: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.597835: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.597837: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.597840: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.848175: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.848180: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.848182: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.848184: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.848186: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.848188: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.848190: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.848192: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.848193: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 619.848195: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.098579: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.098586: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.098588: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.098590: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.098592: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.098594: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.098595: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.098597: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.098599: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.098601: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.348942: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.348948: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.348950: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.348952: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.348953: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.348955: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.348957: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.348959: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.348960: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.348962: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.599294: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.599300: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.599302: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.599304: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.599306: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.599308: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.599310: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.599312: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.599313: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.599315: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.849646: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.849652: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.849654: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.849656: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.849658: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.849660: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.849662: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.849664: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.849666: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 620.849667: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.100009: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.100016: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.100019: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.100021: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.100022: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.100024: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.100026: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.100028: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.100030: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.100032: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.350363: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.350369: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.350371: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.350373: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.350375: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.350377: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.350379: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.350381: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.350382: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.350384: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.600744: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.600749: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.600752: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.600753: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.600755: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.600757: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.600759: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.600760: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.600762: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.600764: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.851104: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.851110: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.851112: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.851114: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.851116: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.851118: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.851120: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.851122: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.851124: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 621.851126: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.101458: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.101464: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.101466: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.101468: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.101470: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.101472: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.101474: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.101476: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.101477: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.101479: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.351816: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.351822: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.351824: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.351826: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.351828: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.351830: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.351832: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.351834: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.351836: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.351838: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.602175: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.602181: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.602183: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.602185: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.602187: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.602188: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.602190: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.602192: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.602194: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.602196: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.852527: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.852533: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.852535: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.852537: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.852539: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.852541: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.852544: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.852547: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.852550: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 622.852553: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.102887: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.102892: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.102895: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.102897: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.102898: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.102900: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.102902: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.102904: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.102906: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.102908: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.353282: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.353289: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.353291: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.353293: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.353295: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.353297: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.353299: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.353300: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.353302: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.353304: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.603663: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.603669: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.603671: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.603673: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.603675: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.603677: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.603678: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.603680: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.603682: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.603684: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.854015: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.854021: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.854023: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.854025: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.854027: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.854029: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.854031: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.854033: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.854035: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 623.854037: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 624.104363: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 624.104369: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 624.104371: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 624.104373: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 624.104374: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 624.104376: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 624.104378: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 624.104380: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 624.104382: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 624.104383: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 624.354724: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 624.354731: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 624.354733: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 624.354735: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 624.354737: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 624.354739: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 624.354741: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 624.354742: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 624.354744: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 624.354746: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 624.605113: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 624.605120: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 624.605122: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 624.605124: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 624.605126: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 624.605128: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 624.605130: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 624.605132: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 624.605134: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 624.605136: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 624.855482: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 624.855488: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 624.855490: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 624.855491: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 624.855493: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 624.855495: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 624.855497: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 624.855498: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 624.855500: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 624.855502: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.105843: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.105848: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.105850: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.105852: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.105854: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.105856: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.105857: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.105859: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.105861: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.105863: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.356218: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.356225: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.356229: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.356232: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.356235: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.356238: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.356241: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.356244: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.356247: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.356250: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.606599: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.606606: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.606608: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.606610: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.606612: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.606614: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.606616: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.606618: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.606620: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.606621: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.856960: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.856966: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.856968: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.856970: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.856972: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.856974: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.856976: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.856978: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.856979: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 625.856981: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.107314: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.107319: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.107321: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.107323: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.107326: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.107328: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.107330: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.107332: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.107334: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.107335: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.357688: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.357693: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.357695: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.357697: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.357699: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.357701: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.357703: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.357705: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.357707: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.357708: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.608042: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.608048: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.608050: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.608052: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.608054: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.608056: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.608058: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.608059: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.608061: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.608063: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.858408: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.858417: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.858421: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.858423: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.858435: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.858437: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.858440: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.858442: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.858445: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 626.858447: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.108799: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.108829: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.108831: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.108833: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.108835: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.108837: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.108839: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.108840: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.108842: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.108844: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.359180: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.359186: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.359188: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.359190: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.359192: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.359194: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.359195: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.359197: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.359199: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.359201: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.609531: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.609537: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.609539: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.609541: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.609543: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.609545: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.609547: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.609549: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.609551: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.609553: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.859886: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.859891: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.859894: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.859895: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.859897: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.859899: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.859901: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.859903: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.859905: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 627.859907: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.110254: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.110260: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.110262: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.110264: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.110266: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.110268: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.110270: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.110271: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.110273: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.110275: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.360604: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.360609: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.360611: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.360613: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.360615: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.360617: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.360619: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.360621: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.360623: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.360625: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.610959: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.610965: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.610967: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.610969: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.610971: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.610973: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.610975: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.610977: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.610978: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.610980: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.861312: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.861317: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.861319: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.861321: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.861323: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.861325: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.861336: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.861338: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.861340: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 628.861341: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.111675: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.111689: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.111692: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.111693: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.111695: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.111697: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.111699: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.111701: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.111703: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.111705: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.362076: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.362082: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.362084: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.362086: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.362088: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.362090: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.362092: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.362094: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.362095: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.362097: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.612427: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.612433: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.612435: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.612437: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.612439: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.612441: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.612443: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.612444: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.612446: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.612448: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.862800: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.862806: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.862808: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.862810: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.862812: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.862814: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.862816: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.862818: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.862820: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 629.862822: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.113174: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.113180: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.113182: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.113184: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.113186: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.113188: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.113190: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.113191: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.113193: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.113195: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.363529: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.363535: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.363537: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.363539: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.363541: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.363543: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.363544: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.363546: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.363548: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.363550: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.613888: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.613893: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.613895: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.613897: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.613899: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.613901: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.613902: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.613904: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.613906: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.613908: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.864240: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.864246: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.864248: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.864250: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.864252: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.864254: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.864256: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.864257: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.864259: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 630.864261: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.114595: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.114601: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.114604: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.114606: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.114607: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.114609: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.114611: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.114613: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.114615: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.114617: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.364951: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.364956: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.364958: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.364960: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.364962: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.364964: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.364966: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.364968: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.364970: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.364972: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.615311: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.615317: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.615319: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.615321: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.615323: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.615325: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.615326: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.615328: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.615330: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.615332: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.865673: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.865679: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.865681: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.865683: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.865684: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.865686: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.865688: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.865690: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.865692: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 631.865693: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.116033: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.116039: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.116041: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.116043: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.116044: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.116046: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.116048: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.116050: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.116052: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.116053: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.366386: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.366401: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.366404: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.366407: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.366409: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.366412: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.366414: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.366417: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.366419: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.366423: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.616752: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.616758: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.616760: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.616762: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.616764: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.616765: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.616767: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.616769: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.616771: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.616797: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.867134: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.867140: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.867142: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.867144: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.867146: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.867148: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.867150: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.867152: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.867153: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 632.867155: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.117488: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.117494: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.117496: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.117498: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.117500: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.117502: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.117504: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.117506: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.117508: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.117510: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.367842: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.367847: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.367849: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.367851: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.367853: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.367855: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.367857: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.367859: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.367861: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.367863: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.618195: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.618201: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.618203: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.618205: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.618207: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.618208: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.618210: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.618212: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.618214: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.618216: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.868549: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.868555: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.868557: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.868559: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.868561: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.868563: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.868565: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.868566: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.868568: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 633.868570: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.118902: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.118908: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.118910: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.118912: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.118914: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.118916: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.118917: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.118919: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.118921: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.118939: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.369273: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.369279: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.369281: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.369283: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.369285: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.369287: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.369288: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.369290: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.369292: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.369294: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.619630: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.619636: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.619638: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.619640: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.619641: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.619643: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.619645: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.619647: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.619648: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.619650: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.869986: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.869991: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.869993: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.869995: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.869997: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.869999: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.870001: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.870002: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.870004: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 634.870006: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.120333: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.120338: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.120341: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.120343: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.120345: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.120347: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.120349: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.120351: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.120353: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.120354: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.370689: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.370694: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.370696: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.370698: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.370699: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.370701: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.370703: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.370704: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.370706: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.370708: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.621042: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.621048: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.621050: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.621052: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.621054: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.621056: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.621057: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.621059: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.621061: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.621063: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.871399: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.871406: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.871408: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.871410: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.871412: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.871414: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.871416: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.871417: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.871419: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 635.871421: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.121877: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.121882: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.121884: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.121886: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.121888: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.121890: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.121892: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.121894: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.121896: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.121898: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.372233: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.372239: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.372241: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.372243: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.372245: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.372247: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.372249: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.372250: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.372252: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.372254: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.622596: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.622602: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.622605: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.622607: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.622608: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.622610: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.622612: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.622614: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.622616: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.622618: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.872756: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.872763: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.872765: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.872767: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.872768: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.872770: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.872772: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.872774: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.872775: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 636.872778: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.123108: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.123114: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.123116: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.123118: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.123120: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.123122: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.123124: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.123125: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.123127: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.123129: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.373466: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.373471: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.373474: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.373476: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.373477: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.373479: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.373481: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.373483: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.373485: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.373487: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.623820: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.623825: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.623828: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.623830: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.623831: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.623833: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.623835: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.623837: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.623839: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.623841: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.873968: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.873974: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.873976: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.873978: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.873980: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.873982: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.873984: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.873986: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.873987: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 637.873989: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.124348: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.124355: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.124357: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.124359: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.124361: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.124363: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.124364: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.124366: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.124368: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.124370: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.374708: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.374713: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.374715: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.374717: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.374719: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.374721: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.374723: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.374724: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.374726: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.374728: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.625083: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.625089: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.625091: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.625093: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.625095: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.625096: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.625098: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.625100: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.625102: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.625103: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.875444: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.875450: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.875452: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.875454: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.875456: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.875458: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.875460: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.875462: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.875464: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 638.875466: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.125797: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.125803: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.125805: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.125807: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.125809: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.125810: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.125812: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.125814: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.125816: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.125818: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.376152: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.376158: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.376160: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.376162: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.376164: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.376166: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.376167: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.376169: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.376171: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.376173: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.626513: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.626519: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.626522: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.626524: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.626526: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.626527: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.626529: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.626531: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.626533: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.626535: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.876872: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.876879: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.876881: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.876883: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.876885: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.876887: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.876889: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.876891: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.876893: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 639.876894: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.127054: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.127061: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.127063: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.127065: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.127067: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.127069: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.127071: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.127073: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.127075: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.127076: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.377413: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.377420: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.377422: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.377424: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.377426: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.377428: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.377430: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.377432: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.377434: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.377436: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.627783: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.627790: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.627792: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.627794: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.627795: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.627797: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.627799: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.627801: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.627803: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.627805: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.878141: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.878148: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.878150: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.878153: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.878155: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.878157: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.878159: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.878160: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.878162: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 640.878164: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.128511: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.128518: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.128520: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.128522: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.128523: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.128525: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.128527: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.128529: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.128531: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.128533: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.378877: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.378884: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.378886: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.378888: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.378890: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.378892: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.378894: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.378896: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.378898: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.378899: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.629247: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.629254: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.629256: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.629258: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.629260: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.629262: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.629264: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.629266: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.629268: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.629269: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.879608: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.879615: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.879617: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.879619: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.879621: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.879622: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.879624: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.879626: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.879628: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 641.879630: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.129989: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.129996: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.129999: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.130000: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.130002: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.130004: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.130006: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.130008: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.130010: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.130012: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.380341: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.380347: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.380350: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.380352: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.380353: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.380355: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.380358: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.380361: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.380364: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.380367: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.630714: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.630720: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.630722: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.630724: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.630726: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.630728: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.630730: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.630732: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.630733: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.630735: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.881079: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.881099: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.881102: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.881104: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.881105: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.881107: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.881109: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.881111: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.881113: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 642.881115: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 643.131458: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 643.131466: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 643.131468: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 643.131469: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 643.131471: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 643.131473: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 643.131475: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 643.131477: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 643.131478: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 643.131480: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 643.381826: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 643.381832: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 643.381834: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 643.381836: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 643.381838: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 643.381840: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 643.381842: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 643.381843: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 643.381845: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 643.381847: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 643.632194: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 643.632201: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 643.632203: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 643.632205: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 643.632207: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 643.632209: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 643.632211: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 643.632212: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 643.632214: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 643.632216: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 643.882559: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 643.882566: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 643.882568: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 643.882570: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 643.882572: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 643.882573: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 643.882575: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 643.882577: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 643.882579: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 643.882581: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.132948: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.132955: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.132957: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.132959: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.132961: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.132963: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.132965: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.132967: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.132969: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.132971: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.383329: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.383335: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.383337: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.383339: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.383341: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.383343: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.383345: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.383347: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.383349: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.383351: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.633692: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.633698: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.633701: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.633702: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.633704: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.633706: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.633708: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.633710: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.633712: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.633714: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.883841: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.883850: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.883853: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.883856: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.883859: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.883862: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.883865: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.883875: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.883878: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 644.883880: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.134215: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.134222: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.134224: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.134226: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.134228: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.134229: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.134231: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.134234: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.134236: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.134238: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.384588: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.384594: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.384597: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.384598: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.384600: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.384602: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.384604: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.384605: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.384607: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.384609: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.634936: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.634942: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.634944: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.634946: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.634948: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.634950: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.634952: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.634954: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.634956: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.634958: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.885302: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.885309: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.885311: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.885313: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.885315: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.885317: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.885318: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.885320: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.885322: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 645.885324: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.135660: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.135666: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.135668: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.135670: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.135672: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.135674: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.135676: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.135678: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.135679: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.135681: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.386038: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.386044: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.386046: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.386048: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.386050: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.386052: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.386054: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.386056: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.386058: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.386059: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.636394: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.636400: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.636402: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.636404: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.636406: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.636407: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.636409: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.636411: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.636413: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.636415: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.886757: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.886763: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.886765: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.886767: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.886769: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.886771: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.886772: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.886774: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.886776: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 646.886778: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 647.137115: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 647.137121: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 647.137124: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 647.137125: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 647.137128: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 647.137130: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 647.137132: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 647.137134: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 647.137136: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [003] .... 647.137137: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 647.387460: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 647.387466: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 647.387468: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 647.387470: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 647.387472: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 647.387473: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 647.387475: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 647.387477: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 647.387479: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [000] .... 647.387480: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 647.637820: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 647.637826: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 647.637828: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 647.637829: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 647.637831: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 647.637833: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 647.637835: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 647.637837: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 647.637838: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 647.637840: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 647.888197: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 647.888203: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 647.888205: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 647.888207: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 647.888209: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 647.888211: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 647.888212: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 647.888214: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 647.888216: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 647.888218: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.138555: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.138560: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.138562: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.138564: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.138566: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.138568: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.138570: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.138571: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.138573: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.138575: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.388911: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.388918: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.388920: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.388922: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.388924: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.388926: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.388928: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.388930: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.388931: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.388933: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.639263: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.639269: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.639271: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.639273: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.639274: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.639276: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.639278: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.639280: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.639282: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.639284: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.889619: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.889625: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.889627: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.889629: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.889631: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.889633: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.889635: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.889637: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.889638: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 648.889640: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.139973: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.139979: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.139981: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.139983: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.139985: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.139987: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.139989: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.139990: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.139992: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.139994: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.390329: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.390334: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.390336: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.390338: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.390340: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.390342: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.390344: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.390346: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.390348: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.390350: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.640699: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.640705: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.640710: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.640712: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.640714: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.640715: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.640717: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.640719: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.640730: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.640732: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.891064: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.891069: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.891072: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.891093: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.891096: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.891098: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.891100: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.891102: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.891103: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [001] .... 649.891105: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 650.141454: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 650.141460: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 650.141462: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 650.141464: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 650.141466: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 650.141468: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 650.141469: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 650.141471: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 650.141473: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 650.141475: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 650.391813: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 650.391818: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 650.391820: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 650.391822: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 650.391823: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 650.391825: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 650.391827: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 650.391829: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 650.391830: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 qemu-system-x86-1630 [002] .... 650.391832: kvm_hypercall: nr 0x9 a0 0x3725ff20 a1 0x0 a2 0x6 a3 0xffffb75e40e7bca8 ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-13 15:56 ` Radim Krcmar 2017-01-13 17:40 ` Marcelo Tosatti @ 2017-01-14 15:26 ` Richard Cochran 2017-01-16 15:48 ` Radim Krcmar 1 sibling, 1 reply; 54+ messages in thread From: Richard Cochran @ 2017-01-14 15:26 UTC (permalink / raw) To: Radim Krcmar Cc: Marcelo Tosatti, kvm, linux-kernel, Paolo Bonzini, Miroslav Lichvar On Fri, Jan 13, 2017 at 04:56:58PM +0100, Radim Krcmar wrote: > > +static int __init ptp_kvm_init(void) > > +{ > > + if (!kvm_para_available()) > > + return -ENODEV; > > + > > + kvm_ptp_clock.caps = ptp_kvm_caps; > > + > > + kvm_ptp_clock.ptp_clock = ptp_clock_register(&kvm_ptp_clock.caps, NULL); > > It is a shame that the infrastructure uses polling when the guest could > be notified on every host real time change, but this should be good > enough. This comment makes no sense at all. What do you mean by "host real time change"? Thanks, Richard ^ permalink raw reply [flat|nested] 54+ messages in thread
* Re: [patch 3/3] PTP: add kvm PTP driver 2017-01-14 15:26 ` Richard Cochran @ 2017-01-16 15:48 ` Radim Krcmar 0 siblings, 0 replies; 54+ messages in thread From: Radim Krcmar @ 2017-01-16 15:48 UTC (permalink / raw) To: Richard Cochran Cc: Marcelo Tosatti, kvm, linux-kernel, Paolo Bonzini, Miroslav Lichvar 2017-01-14 16:26+0100, Richard Cochran: > On Fri, Jan 13, 2017 at 04:56:58PM +0100, Radim Krcmar wrote: >> > +static int __init ptp_kvm_init(void) >> > +{ >> > + if (!kvm_para_available()) >> > + return -ENODEV; >> > + >> > + kvm_ptp_clock.caps = ptp_kvm_caps; >> > + >> > + kvm_ptp_clock.ptp_clock = ptp_clock_register(&kvm_ptp_clock.caps, NULL); >> >> It is a shame that the infrastructure uses polling when the guest could >> be notified on every host real time change, but this should be good >> enough. > > This comment makes no sense at all. What do you mean by "host real > time change"? Real time in the sense of wall clock, as perceived by the host, and the change of that time. Unlike other PTP drivers, host (source) and guest (destination) share the same hardware clock, so they cannot shift or drift unless one of them changes its "TSC to real time" conversion (the host is most likely using NTP/PTP to keep its own real time). I meant that the host could notify the guest when a change happens, which would be more efficient. ^ permalink raw reply [flat|nested] 54+ messages in thread
* [patch 0/3] KVM virtual PTP driver (v2)
@ 2017-01-13 18:45 Marcelo Tosatti
2017-01-13 18:46 ` [patch 2/3] KVM: x86: add KVM_HC_CLOCK_OFFSET hypercall Marcelo Tosatti
0 siblings, 1 reply; 54+ messages in thread
From: Marcelo Tosatti @ 2017-01-13 18:45 UTC (permalink / raw)
To: kvm, linux-kernel
Cc: Paolo Bonzini, Radim Krcmar, Richard Cochran, Miroslav Lichvar
This patchset implements a virtual PTP driver which allows guest to sync
its clock to the host clock with high precision ( error is < 10ns).
Changelog from v1
Patch1:
v2: unify nsec_base (Radim)
Patch2:
v2: improve documentation (Radim)
change hypercall name to KVM_HC_CLOCK_PAIRING (Radim)
increase padding size
Patch3:
v2: check for kvmclock (Radim)
initialize global variables before device registration (Radim)
^ permalink raw reply [flat|nested] 54+ messages in thread* [patch 2/3] KVM: x86: add KVM_HC_CLOCK_OFFSET hypercall 2017-01-13 18:45 [patch 0/3] KVM virtual PTP driver (v2) Marcelo Tosatti @ 2017-01-13 18:46 ` Marcelo Tosatti 0 siblings, 0 replies; 54+ messages in thread From: Marcelo Tosatti @ 2017-01-13 18:46 UTC (permalink / raw) To: kvm, linux-kernel Cc: Paolo Bonzini, Radim Krcmar, Richard Cochran, Miroslav Lichvar, Marcelo Tosatti [-- Attachment #1: kvm-hv-clock-offset --] [-- Type: text/plain, Size: 5057 bytes --] Add a hypercall to retrieve the host realtime clock and the TSC value used to calculate that clock read. Used to implement clock synchronization between host and guest. Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com> --- Documentation/virtual/kvm/hypercalls.txt | 33 ++++++++++++++++++++++++ arch/x86/include/uapi/asm/kvm_para.h | 9 ++++++ arch/x86/kvm/x86.c | 41 +++++++++++++++++++++++++++++++ include/uapi/linux/kvm_para.h | 3 ++ 4 files changed, 86 insertions(+) v2: improve documentation (Radim) change hypercall name to KVM_HC_CLOCK_PAIRING (Radim) increase padding size Index: kvm-ptpdriver/arch/x86/include/uapi/asm/kvm_para.h =================================================================== --- kvm-ptpdriver.orig/arch/x86/include/uapi/asm/kvm_para.h 2017-01-13 15:43:47.196867691 -0200 +++ kvm-ptpdriver/arch/x86/include/uapi/asm/kvm_para.h 2017-01-13 15:59:39.596193853 -0200 @@ -50,6 +50,15 @@ __u32 pad[11]; }; +#define KVM_CLOCK_PAIRING_WALLCLOCK 0 +struct kvm_clock_offset { + __s64 sec; + __s64 nsec; + __u64 tsc; + __u32 flags; + __u32 pad[9]; +}; + #define KVM_STEAL_ALIGNMENT_BITS 5 #define KVM_STEAL_VALID_BITS ((-1ULL << (KVM_STEAL_ALIGNMENT_BITS + 1))) #define KVM_STEAL_RESERVED_MASK (((1 << KVM_STEAL_ALIGNMENT_BITS) - 1 ) << 1) Index: kvm-ptpdriver/Documentation/virtual/kvm/hypercalls.txt =================================================================== --- kvm-ptpdriver.orig/Documentation/virtual/kvm/hypercalls.txt 2017-01-13 15:43:47.196867691 -0200 +++ kvm-ptpdriver/Documentation/virtual/kvm/hypercalls.txt 2017-01-13 15:59:26.676163231 -0200 @@ -81,3 +81,36 @@ same guest can wakeup the sleeping vcpu by issuing KVM_HC_KICK_CPU hypercall, specifying APIC ID (a1) of the vcpu to be woken up. An additional argument (a0) is used in the hypercall for future use. + + +6. KVM_HC_CLOCK_PAIRING +------------------------ +Architecture: x86 +Status: active +Purpose: Hypercall used to synchronize host and guest clocks. +Usage: + +a0: guest physical address where host copies +"struct kvm_clock_offset" structure. + +a1: clock_type, ATM only KVM_CLOCK_PAIRING_WALLCLOCK (0) +is supported (hosts CLOCK_REALTIME clock). + + struct kvm_clock_offset { + __s64 sec; + __s64 nsec; + __u64 tsc; + __u32 flags; + __u32 pad[9]; + }; + + Where: + * sec: seconds from clock_type clock. + * nsec: nanoseconds from clock_type clock. + * tsc: TSC value used to calculate sec/nsec pair + (this hypercall only works when host uses TSC clocksource). + * flags: flags, unused (0) at the moment. + +Returns KVM_EOPNOTSUPP if the host does not use TSC clocksource, +or if clock type is different than KVM_CLOCK_PAIRING_WALLCLOCK. + Index: kvm-ptpdriver/arch/x86/kvm/x86.c =================================================================== --- kvm-ptpdriver.orig/arch/x86/kvm/x86.c 2017-01-13 15:49:14.058347822 -0200 +++ kvm-ptpdriver/arch/x86/kvm/x86.c 2017-01-13 16:00:56.964373208 -0200 @@ -6119,6 +6119,44 @@ } EXPORT_SYMBOL_GPL(kvm_emulate_halt); +static int kvm_pv_clock_pairing(struct kvm_vcpu *vcpu, gpa_t paddr, + unsigned long clock_type) +{ + struct kvm_clock_offset *clock_offset; + struct timespec ts; + cycle_t cycle; + int ret; + + if (clock_type != KVM_CLOCK_PAIRING_WALLCLOCK) + return -KVM_EOPNOTSUPP; + + if (pvclock_gtod_data.clock.vclock_mode != VCLOCK_TSC) + return -KVM_EOPNOTSUPP; + + clock_offset = kzalloc(sizeof(struct kvm_clock_offset), GFP_KERNEL); + if (clock_offset == NULL) + return -KVM_ENOMEM; + + if (kvm_get_walltime_and_clockread(&ts, &cycle) == false) { + kfree(clock_offset); + return -KVM_EOPNOTSUPP; + } + + clock_offset->sec = ts.tv_sec; + clock_offset->nsec = ts.tv_nsec; + clock_offset->tsc = kvm_read_l1_tsc(vcpu, cycle); + clock_offset->flags = 0; + + ret = 0; + if (kvm_write_guest(vcpu->kvm, paddr, clock_offset, + sizeof(struct kvm_clock_offset))) + ret = -KVM_EFAULT; + + kfree(clock_offset); + + return ret; +} + /* * kvm_pv_kick_cpu_op: Kick a vcpu. * @@ -6183,6 +6221,9 @@ kvm_pv_kick_cpu_op(vcpu->kvm, a0, a1); ret = 0; break; + case KVM_HC_CLOCK_PAIRING: + ret = kvm_pv_clock_pairing(vcpu, a0, a1); + break; default: ret = -KVM_ENOSYS; break; Index: kvm-ptpdriver/include/uapi/linux/kvm_para.h =================================================================== --- kvm-ptpdriver.orig/include/uapi/linux/kvm_para.h 2017-01-13 15:43:47.196867691 -0200 +++ kvm-ptpdriver/include/uapi/linux/kvm_para.h 2017-01-13 16:01:09.290401187 -0200 @@ -14,6 +14,8 @@ #define KVM_EFAULT EFAULT #define KVM_E2BIG E2BIG #define KVM_EPERM EPERM +#define KVM_EOPNOTSUPP EOPNOTSUPP +#define KVM_ENOMEM ENOMEM #define KVM_HC_VAPIC_POLL_IRQ 1 #define KVM_HC_MMU_OP 2 @@ -23,6 +25,7 @@ #define KVM_HC_MIPS_GET_CLOCK_FREQ 6 #define KVM_HC_MIPS_EXIT_VM 7 #define KVM_HC_MIPS_CONSOLE_OUTPUT 8 +#define KVM_HC_CLOCK_PAIRING 9 /* * hypercalls use architecture specific ^ permalink raw reply [flat|nested] 54+ messages in thread
end of thread, other threads:[~2017-01-20 14:27 UTC | newest] Thread overview: 54+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2017-01-13 12:01 [patch 0/3] KVM virtual PTP driver Marcelo Tosatti 2017-01-13 12:01 ` [patch 1/3] KVM: x86: provide realtime host clock via vsyscall notifiers Marcelo Tosatti 2017-01-13 15:18 ` Radim Krcmar 2017-01-13 15:34 ` Marcelo Tosatti 2017-01-13 16:28 ` Radim Krcmar 2017-01-13 17:51 ` Marcelo Tosatti 2017-01-16 15:40 ` Radim Krcmar 2017-01-13 15:41 ` Konrad Rzeszutek Wilk 2017-01-13 15:46 ` Marcelo Tosatti 2017-01-13 12:01 ` [patch 2/3] KVM: x86: add KVM_HC_CLOCK_OFFSET hypercall Marcelo Tosatti 2017-01-13 15:31 ` Radim Krcmar 2017-01-13 15:43 ` Marcelo Tosatti 2017-01-13 17:07 ` Radim Krcmar 2017-01-13 17:57 ` Marcelo Tosatti 2017-01-13 12:01 ` [patch 3/3] PTP: add kvm PTP driver Marcelo Tosatti 2017-01-13 15:56 ` Radim Krcmar 2017-01-13 17:40 ` Marcelo Tosatti 2017-01-16 16:26 ` Radim Krcmar 2017-01-16 16:54 ` Radim Krcmar 2017-01-16 17:08 ` Marcelo Tosatti 2017-01-16 17:27 ` Radim Krcmar 2017-01-16 17:39 ` Marcelo Tosatti 2017-01-16 18:01 ` Radim Krcmar 2017-01-16 19:36 ` Marcelo Tosatti 2017-01-16 19:47 ` Marcelo Tosatti 2017-01-16 20:01 ` Marcelo Tosatti 2017-01-17 8:03 ` Miroslav Lichvar 2017-01-17 11:30 ` Marcelo Tosatti 2017-01-17 15:36 ` Radim Krcmar 2017-01-18 12:17 ` Marcelo Tosatti 2017-01-18 12:24 ` Marcelo Tosatti 2017-01-18 12:46 ` Paolo Bonzini 2017-01-18 13:36 ` Miroslav Lichvar 2017-01-18 14:02 ` Paolo Bonzini 2017-01-18 14:50 ` Marcelo Tosatti 2017-01-18 15:35 ` Radim Krcmar 2017-01-18 15:45 ` Paolo Bonzini 2017-01-18 15:57 ` Marcelo Tosatti 2017-01-18 14:24 ` Marcelo Tosatti 2017-01-18 15:54 ` Miroslav Lichvar 2017-01-18 16:07 ` Paolo Bonzini 2017-01-18 16:14 ` Radim Krcmar 2017-01-18 14:37 ` Marcelo Tosatti 2017-01-18 14:53 ` Marcelo Tosatti 2017-01-18 15:20 ` Radim Krcmar 2017-01-18 15:28 ` Marcelo Tosatti 2017-01-20 14:18 ` Radim Krcmar 2017-01-18 15:59 ` Radim Krcmar 2017-01-16 17:04 ` Marcelo Tosatti 2017-01-16 17:46 ` Radim Krcmar 2017-01-16 19:33 ` Marcelo Tosatti 2017-01-14 15:26 ` Richard Cochran 2017-01-16 15:48 ` Radim Krcmar 2017-01-13 18:45 [patch 0/3] KVM virtual PTP driver (v2) Marcelo Tosatti 2017-01-13 18:46 ` [patch 2/3] KVM: x86: add KVM_HC_CLOCK_OFFSET hypercall Marcelo Tosatti
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).