Re: [Qemu-devel] [qemu patch V2 2/2] kvmclock: reduce kvmclock difference on migration

[Eduardo Habkost]

On Thu, Nov 17, 2016 at 03:15:35PM -0200, Marcelo Tosatti wrote:
> On Thu, Nov 17, 2016 at 12:11:26PM -0200, Eduardo Habkost wrote:
> > On Thu, Nov 17, 2016 at 11:24:13AM -0200, Marcelo Tosatti wrote:
> > > Check for KVM_CAP_ADJUST_CLOCK capability KVM_CLOCK_TSC_STABLE, which
> > > indicates that KVM_GET_CLOCK returns a value as seen by the guest at
> > > that moment.
> > > 
> > > For new machine types, use this value rather than reading 
> > > from guest memory.
> > > 
> > > This reduces kvmclock difference on migration from 5s to 0.1s
> > > (when max_downtime == 5s).
> > > 
> > > Signed-off-by: Marcelo Tosatti <address@hidden>
> > > 
> > > ---
> > >  hw/i386/kvm/clock.c    |  108 
> > > ++++++++++++++++++++++++++++++++++++++++++-------
> > >  include/hw/i386/pc.h   |    5 ++
> > >  target-i386/kvm.c      |    7 +++
> > >  target-i386/kvm_i386.h |    1 
> > >  4 files changed, 107 insertions(+), 14 deletions(-)
> > > 
> > > v2: 
> > > - improve variable names (Juan)
> > > - consolidate code on kvm_get_clock function (Paolo)
> > > - return mach_use_reliable_get_clock from needed function (Paolo)
> > > 
> > > Index: qemu-mig-advance-clock/hw/i386/kvm/clock.c
> > > ===================================================================
> > > --- qemu-mig-advance-clock.orig/hw/i386/kvm/clock.c       2016-11-14 
> > > 10:40:39.748116312 -0200
> > > +++ qemu-mig-advance-clock/hw/i386/kvm/clock.c    2016-11-14 
> > > 13:38:29.299955042 -0200
> > > @@ -36,6 +36,12 @@
> > >  
> > >      uint64_t clock;
> > >      bool clock_valid;
> > > +
> > > +    /* whether machine supports reliable KVM_GET_CLOCK */
> > > +    bool mach_use_reliable_get_clock;
> > > +
> > > +    /* whether source host supported reliable KVM_GET_CLOCK */
> > > +    bool src_use_reliable_get_clock;
> > >  } KVMClockState;
> > >  
> > >  struct pvclock_vcpu_time_info {
> > > @@ -81,6 +87,19 @@
> > >      return nsec + time.system_time;
> > >  }
> > >  
> > > +static uint64_t kvm_get_clock(void)
> > > +{
> > > +    struct kvm_clock_data data;
> > > +    int ret;
> > > +
> > > +    ret = kvm_vm_ioctl(kvm_state, KVM_GET_CLOCK, &data);
> > > +    if (ret < 0) {
> > > +        fprintf(stderr, "KVM_GET_CLOCK failed: %s\n", strerror(ret));
> > > +                abort();
> > > +    }
> > > +    return data.clock;
> > > +}
> > > +
> > >  static void kvmclock_vm_state_change(void *opaque, int running,
> > >                                       RunState state)
> > >  {
> > > @@ -91,15 +110,37 @@
> > >  
> > >      if (running) {
> > >          struct kvm_clock_data data = {};
> > > -        uint64_t time_at_migration = kvmclock_current_nsec(s);
> > > +        uint64_t pvclock_via_mem = 0;
> > >  
> > > -        s->clock_valid = false;
> > > +        /* local (running VM) restore */
> > > +        if (s->clock_valid) {
> > > +            /*
> > > +             * if host does not support reliable KVM_GET_CLOCK,
> > > +             * read kvmclock value from memory
> > > +             */
> > > +            if (!kvm_has_adjust_clock_stable()) {
> > > +                pvclock_via_mem = kvmclock_current_nsec(s);
> > > +            }
> > > +        /* migration/savevm/init restore */
> > > +        } else {
> > > +            /*
> > > +             * use s->clock in case machine uses reliable
> > > +             * get clock and source host supported
> > > +             * reliable get clock
> > > +             */
> > > +            if (!(s->mach_use_reliable_get_clock &&
> > > +                  s->src_use_reliable_get_clock)) {
> > > +                pvclock_via_mem = kvmclock_current_nsec(s);
> > > +            }
> > 
> > The s->mach_use_reliable_get_clock check seems redundant.
> > src_use_reliable_get_clock is set only if
> > mach_use_reliable_get_clock is true.
> 
> Done.
> 
> > > +        }
> > >  
> > > -        /* We can't rely on the migrated clock value, just discard it */
> > > -        if (time_at_migration) {
> > > -            s->clock = time_at_migration;
> > > +        /* We can't rely on the saved clock value, just discard it */
> > > +        if (pvclock_via_mem) {
> > > +            s->clock = pvclock_via_mem;
> > >          }
> > >  
> > > +        s->clock_valid = false;
> > > +
> > >          data.clock = s->clock;
> > >          ret = kvm_vm_ioctl(kvm_state, KVM_SET_CLOCK, &data);
> > >          if (ret < 0) {
> > > @@ -120,8 +161,6 @@
> > >              }
> > >          }
> > >      } else {
> > > -        struct kvm_clock_data data;
> > > -        int ret;
> > >  
> > >          if (s->clock_valid) {
> > >              return;
> > > @@ -129,13 +168,7 @@
> > >  
> > >          kvm_synchronize_all_tsc();
> > >  
> > > -        ret = kvm_vm_ioctl(kvm_state, KVM_GET_CLOCK, &data);
> > > -        if (ret < 0) {
> > > -            fprintf(stderr, "KVM_GET_CLOCK failed: %s\n", strerror(ret));
> > > -            abort();
> > > -        }
> > > -        s->clock = data.clock;
> > > -
> > > +        s->clock = kvm_get_clock();
> > >          /*
> > >           * If the VM is stopped, declare the clock state valid to
> > >           * avoid re-reading it on next vmsave (which would return
> > > @@ -152,22 +185,69 @@
> > >      qemu_add_vm_change_state_handler(kvmclock_vm_state_change, s);
> > >  }
> > >  
> > > +static bool kvmclock_src_use_reliable_get_clock(void *opaque)
> > > +{
> > > +    KVMClockState *s = opaque;
> > > +
> > > +    /*
> > > +     * On machine types that support reliable KVM_GET_CLOCK,
> > > +     * if host kernel does provide reliable KVM_GET_CLOCK,
> > > +     * set src_use_reliable_get_clock=true so that destination
> > > +     * avoids reading kvmclock from memory.
> > > +     */
> > > +    if (s->mach_use_reliable_get_clock && kvm_has_adjust_clock_stable()) 
> > > {
> > > +        s->src_use_reliable_get_clock = true;
> > > +    }
> > 
> > It feels fragile to change device state inside the .needed
> > function. Better to initialize src_use_reliable_get_clock on
> > kvmclock_realize()?
> > 
> > What exactly ensures src_use_reliable_get_clock is correctly
> > initialized on the migration destination as well?
> 
> Its initialized to false (because its part of device state).

Do you mean it is always going to be false on the destination
host? What if .needed is called by other code? .needed shouldn't
have any side-effects.

Also, the difference between the variables on the destination and
the source is very confusing, see other comments below:

> 
> > 
> > > +
> > > +    return s->mach_use_reliable_get_clock;
> > 
> > If if kvm_has_adjust_clock_stable() is false, isn't it simpler to
> > simply skip the section?
> > 
> > It looks like mach_use_reliable_get_clock and
> > src_use_reliable_get_clock could become a single field:
> > 
> > * use_reliable_get_clock ("x-use-reliable-get-clock" property)
> >   set to true by default (set on DEFINE_PROP_BOOL parameter)
> > * "x-use-reliable-get-clock" set to false by default on older
> >   machine-types
> > * use_reliable_get_clock forced to false on kvmclock_realize() if
> >   !kvm_has_adjust_clock_stable()
> > * kvmclock_reliable_get_clock.needed return s->use_reliable_get_clock
> 
> No because use_reliable_get_clock can be false because the destination host
> does not support kvm_has_adjust_clock_stable(), but the source host
> supports it (so we want to use the value in the first incoming
> migration).
> 

Are you talking about the value of the fields on the source host,
or on the destination host? In this case, you mean that
use_reliable_get_clock should be false on the destination, right?

> So the variable indicates whether the source host supported
> migration, not whether the destination supports it.

Good, that's the whole point of the field, isn't it? I will try
to summarize all cases below. Please correct me if something is
incorrect.

--------+------------+---------------+-----------+------------+
        | kvm_has_adj_clock_stable() | use_reliable_get_clock |
machine | src        | dst           | src       | dst        |
--------+------------+---------------+-----------+------------+
pc-2.8  | false      | false         | false     | false      |
pc-2.8  | false      | true          | false     | false (*)  |
pc-2.8  | true       | false         | true      | true (**)  |
pc-2.8  | true       | true          | true      | true       |
--------+------------+---------------+-----------+------------+
pc-2.7  | (any)      | (any)         | false     | false      |
--------+------------+---------------+-----------+------------+

(*) I guess this is where things would break if we skip the
    subsection (and use a single field). If the subsection is
    skipped, use_reliable_get_clock would be set to true on the
    destination. (See my comment below on the new patch you sent)

(**) Destination can still do the right thing here, right?



> 
[...]
> 
> How about the following:
> 
> Index: qemu-mig-advance-clock/hw/i386/kvm/clock.c
> ===================================================================
> --- qemu-mig-advance-clock.orig/hw/i386/kvm/clock.c   2016-11-17 
> 15:07:11.220632761 -0200
> +++ qemu-mig-advance-clock/hw/i386/kvm/clock.c        2016-11-17 
> 15:11:51.372048640 -0200
> @@ -36,6 +36,12 @@
>  
>      uint64_t clock;
>      bool clock_valid;
> +
> +    /* whether machine supports reliable KVM_GET_CLOCK */
> +    bool mach_use_reliable_get_clock;
> +
> +    /* whether source host supported reliable KVM_GET_CLOCK */
> +    bool src_use_reliable_get_clock;
>  } KVMClockState;
>  
>  struct pvclock_vcpu_time_info {
> @@ -81,6 +87,19 @@
>      return nsec + time.system_time;
>  }
>  
> +static uint64_t kvm_get_clock(void)
> +{
> +    struct kvm_clock_data data;
> +    int ret;
> +
> +    ret = kvm_vm_ioctl(kvm_state, KVM_GET_CLOCK, &data);
> +    if (ret < 0) {
> +        fprintf(stderr, "KVM_GET_CLOCK failed: %s\n", strerror(ret));
> +                abort();
> +    }
> +    return data.clock;
> +}
> +
>  static void kvmclock_vm_state_change(void *opaque, int running,
>                                       RunState state)
>  {
> @@ -91,15 +110,36 @@
>  
>      if (running) {
>          struct kvm_clock_data data = {};
> -        uint64_t time_at_migration = kvmclock_current_nsec(s);
> +        uint64_t pvclock_via_mem = 0;
>  
> -        s->clock_valid = false;
> +        /* local (running VM) restore */
> +        if (s->clock_valid) {
> +            /*
> +             * if host does not support reliable KVM_GET_CLOCK,
> +             * read kvmclock value from memory
> +             */
> +            if (!kvm_has_adjust_clock_stable()) {
> +                pvclock_via_mem = kvmclock_current_nsec(s);
> +            }
> +        /* migration/savevm/init restore */
> +        } else {
> +            /*
> +             * use s->clock in case machine uses reliable
> +             * get clock and source host supported
> +             * reliable get clock
> +             */
> +            if (!s->src_use_reliable_get_clock) {
> +                pvclock_via_mem = kvmclock_current_nsec(s);
> +            }
> +        }
>  
> -        /* We can't rely on the migrated clock value, just discard it */
> -        if (time_at_migration) {
> -            s->clock = time_at_migration;
> +        /* We can't rely on the saved clock value, just discard it */
> +        if (pvclock_via_mem) {
> +            s->clock = pvclock_via_mem;
>          }
>  
> +        s->clock_valid = false;
> +
>          data.clock = s->clock;
>          ret = kvm_vm_ioctl(kvm_state, KVM_SET_CLOCK, &data);
>          if (ret < 0) {
> @@ -120,8 +160,6 @@
>              }
>          }
>      } else {
> -        struct kvm_clock_data data;
> -        int ret;
>  
>          if (s->clock_valid) {
>              return;
> @@ -129,13 +167,7 @@
>  
>          kvm_synchronize_all_tsc();
>  
> -        ret = kvm_vm_ioctl(kvm_state, KVM_GET_CLOCK, &data);
> -        if (ret < 0) {
> -            fprintf(stderr, "KVM_GET_CLOCK failed: %s\n", strerror(ret));
> -            abort();
> -        }
> -        s->clock = data.clock;
> -
> +        s->clock = kvm_get_clock();
>          /*
>           * If the VM is stopped, declare the clock state valid to
>           * avoid re-reading it on next vmsave (which would return
> @@ -149,25 +181,72 @@
>  {
>      KVMClockState *s = KVM_CLOCK(dev);
>  
> +    /*
> +     * On machine types that support reliable KVM_GET_CLOCK,
> +     * if host kernel does provide reliable KVM_GET_CLOCK,
> +     * set src_use_reliable_get_clock=true so that destination
> +     * avoids reading kvmclock from memory.
> +     */
> +    if (s->mach_use_reliable_get_clock && kvm_has_adjust_clock_stable()) {
> +        s->src_use_reliable_get_clock = true;
> +    }
> +
>      qemu_add_vm_change_state_handler(kvmclock_vm_state_change, s);
>  }
>  
> +static bool kvmclock_src_use_reliable_get_clock(void *opaque)
> +{
> +    KVMClockState *s = opaque;
> +
> +    return s->mach_use_reliable_get_clock;
> +}

I didn't review kvmclock_vm_state_change() and
kvm_has_adjust_clock_stable(), but the rules to migrate
src_use_reliable_get_clock look good now, and seem to match the
table above.

> +
> +static const VMStateDescription kvmclock_reliable_get_clock = {
> +    .name = "kvmclock/src_use_reliable_get_clock",
> +    .version_id = 1,
> +    .minimum_version_id = 1,
> +    .needed = kvmclock_src_use_reliable_get_clock,
> +    .fields = (VMStateField[]) {
> +        VMSTATE_BOOL(src_use_reliable_get_clock, KVMClockState),
> +        VMSTATE_END_OF_LIST()
> +    }
> +};
> +
> +static void kvmclock_pre_save(void *opaque)
> +{
> +    KVMClockState *s = opaque;
> +
> +    s->clock = kvm_get_clock();
> +}
> +
>  static const VMStateDescription kvmclock_vmsd = {
>      .name = "kvmclock",
>      .version_id = 1,
>      .minimum_version_id = 1,
> +    .pre_save = kvmclock_pre_save,
>      .fields = (VMStateField[]) {
>          VMSTATE_UINT64(clock, KVMClockState),
>          VMSTATE_END_OF_LIST()
> +    },
> +    .subsections = (const VMStateDescription * []) {
> +        &kvmclock_reliable_get_clock,
> +        NULL
>      }
>  };
>  
> +static Property kvmclock_properties[] = {
> +    DEFINE_PROP_BOOL("x-mach-use-reliable-get-clock", KVMClockState,
> +                      mach_use_reliable_get_clock, true),
> +    DEFINE_PROP_END_OF_LIST(),
> +};
> +
>  static void kvmclock_class_init(ObjectClass *klass, void *data)
>  {
>      DeviceClass *dc = DEVICE_CLASS(klass);
>  
>      dc->realize = kvmclock_realize;
>      dc->vmsd = &kvmclock_vmsd;
> +    dc->props = kvmclock_properties;
>  }
>  
>  static const TypeInfo kvmclock_info = {
> Index: qemu-mig-advance-clock/include/hw/i386/pc.h
> ===================================================================
> --- qemu-mig-advance-clock.orig/include/hw/i386/pc.h  2016-11-17 
> 15:07:11.220632761 -0200
> +++ qemu-mig-advance-clock/include/hw/i386/pc.h       2016-11-17 
> 15:08:58.096791416 -0200
> @@ -370,6 +370,11 @@
>  #define PC_COMPAT_2_7 \
>      HW_COMPAT_2_7 \
>      {\
> +        .driver   = "kvmclock",\
> +        .property = "x-mach-use-reliable-get-clock",\
> +        .value    = "off",\
> +    },\
> +    {\
>          .driver   = TYPE_X86_CPU,\
>          .property = "l3-cache",\
>          .value    = "off",\
> Index: qemu-mig-advance-clock/target-i386/kvm.c
> ===================================================================
> --- qemu-mig-advance-clock.orig/target-i386/kvm.c     2016-11-17 
> 15:07:11.221632762 -0200
> +++ qemu-mig-advance-clock/target-i386/kvm.c  2016-11-17 15:07:15.867639659 
> -0200
> @@ -117,6 +117,13 @@
>      return kvm_check_extension(kvm_state, KVM_CAP_X86_SMM);
>  }
>  
> +bool kvm_has_adjust_clock_stable(void)
> +{
> +    int ret = kvm_check_extension(kvm_state, KVM_CAP_ADJUST_CLOCK);
> +
> +    return (ret == KVM_CLOCK_TSC_STABLE);
> +}
> +
>  bool kvm_allows_irq0_override(void)
>  {
>      return !kvm_irqchip_in_kernel() || kvm_has_gsi_routing();
> Index: qemu-mig-advance-clock/target-i386/kvm_i386.h
> ===================================================================
> --- qemu-mig-advance-clock.orig/target-i386/kvm_i386.h        2016-11-17 
> 15:07:11.222632764 -0200
> +++ qemu-mig-advance-clock/target-i386/kvm_i386.h     2016-11-17 
> 15:07:15.867639659 -0200
> @@ -17,6 +17,7 @@
>  
>  bool kvm_allows_irq0_override(void);
>  bool kvm_has_smm(void);
> +bool kvm_has_adjust_clock_stable(void);
>  void kvm_synchronize_all_tsc(void);
>  void kvm_arch_reset_vcpu(X86CPU *cs);
>  void kvm_arch_do_init_vcpu(X86CPU *cs);

-- 
Eduardo