Re: [PATCH RFC v3 4/6] Documentation: Add three sysctls for smart idle poll

[quan.xu04@xxxxxxxxx]

On 2017å11æ14æ 15:44, Ingo Molnar wrote:
> * Quan Xu <quan.xu0@xxxxxxxxx> wrote:
>
> > On 2017/11/13 23:08, Ingo Molnar wrote:
> > > * Quan Xu <quan.xu04@xxxxxxxxx> wrote:
> > >
> > > > From: Quan Xu <quan.xu0@xxxxxxxxx>
> > > >
> > > > To reduce the cost of poll, we introduce three sysctl to control the
> > > > poll time when running as a virtual machine with paravirt.
> > > >
> > > > Signed-off-by: Yang Zhang <yang.zhang.wz@xxxxxxxxx>
> > > > Signed-off-by: Quan Xu <quan.xu0@xxxxxxxxx>
> > > > ---
> > > >    Documentation/sysctl/kernel.txt |   35 +++++++++++++++++++++++++++++++++++
> > > >    arch/x86/kernel/paravirt.c      |    4 ++++
> > > >    include/linux/kernel.h          |    6 ++++++
> > > >    kernel/sysctl.c                 |   34 ++++++++++++++++++++++++++++++++++
> > > >    4 files changed, 79 insertions(+), 0 deletions(-)
> > > >
> > > > diff --git a/Documentation/sysctl/kernel.txt b/Documentation/sysctl/kernel.txt
> > > > index 694968c..30c25fb 100644
> > > > --- a/Documentation/sysctl/kernel.txt
> > > > +++ b/Documentation/sysctl/kernel.txt
> > > > @@ -714,6 +714,41 @@ kernel tries to allocate a number starting from this one.
> > > >    ==============================================================
> > > > +paravirt_poll_grow: (X86 only)
> > > > +
> > > > +Multiplied value to increase the poll time. This is expected to take
> > > > +effect only when running as a virtual machine with CONFIG_PARAVIRT
> > > > +enabled. This can't bring any benifit on bare mental even with
> > > > +CONFIG_PARAVIRT enabled.
> > > > +
> > > > +By default this value is 2. Possible values to set are in range {2..16}.
> > > > +
> > > > +==============================================================
> > > > +
> > > > +paravirt_poll_shrink: (X86 only)
> > > > +
> > > > +Divided value to reduce the poll time. This is expected to take effect
> > > > +only when running as a virtual machine with CONFIG_PARAVIRT enabled.
> > > > +This can't bring any benifit on bare mental even with CONFIG_PARAVIRT
> > > > +enabled.
> > > > +
> > > > +By default this value is 2. Possible values to set are in range {2..16}.
> > > > +
> > > > +==============================================================
> > > > +
> > > > +paravirt_poll_threshold_ns: (X86 only)
> > > > +
> > > > +Controls the maximum poll time before entering real idle path. This is
> > > > +expected to take effect only when running as a virtual machine with
> > > > +CONFIG_PARAVIRT enabled. This can't bring any benifit on bare mental
> > > > +even with CONFIG_PARAVIRT enabled.
> > > > +
> > > > +By default, this value is 0 means not to poll. Possible values to set
> > > > +are in range {0..500000}. Change the value to non-zero if running
> > > > +latency-bound workloads in a virtual machine.
> > > I absolutely hate it how this hybrid idle loop polling mechanism is not
> > > self-tuning!
> > Ingo, actually it is self-tuning..
> Then why the hell does it touch the syscall ABI?


just for more data about performance and CPU utilization with different
the maximum poll time.

there are 3 parameters, paravirt_poll_{grow|shrink|threshold_ns}..
we didn't touch paravirt_poll_{grow|shrink} since we sent out v1.

We tested it based on benchmark contextswitch / netperf with different
paravirt_poll_threshold_ns.

Here is the data we get when running benchmark contextswitch to measure
the latency(lower is better):
ÂÂÂÂÂ halt_poll_threshold=0ÂÂÂÂÂ -- 3402.9 ns/ctxsw -- 199.8 %CPU
ÂÂÂÂÂ halt_poll_threshold=10000Â -- 1151.4 ns/ctxsw -- 200.1 %CPU
ÂÂÂÂÂ halt_poll_threshold=20000Â -- 1149.7 ns/ctxsw -- 199.9 %CPU
ÂÂÂÂÂ halt_poll_threshold=30000Â -- 1151.0 ns/ctxsw -- 199.9 %CPU
ÂÂÂÂÂ halt_poll_threshold=40000Â -- 1155.4 ns/ctxsw -- 199.3 %CPU
ÂÂÂÂÂ halt_poll_threshold=50000Â -- 1161.0 ns/ctxsw -- 200.0 %CPU
ÂÂÂÂÂ halt_poll_threshold=100000 -- 1163.8 ns/ctxsw -- 200.4 %CPU
ÂÂÂÂÂ halt_poll_threshold=200000 -- 1163.8 ns/ctxsw -- 201.4 %CPU
ÂÂÂÂÂ halt_poll_threshold=300000 -- 1159.4 ns/ctxsw -- 201.9 %CPU
ÂÂÂÂÂ halt_poll_threshold=500000 -- 1163.5 ns/ctxsw -- 205.5 %CPU


Here is the data we get when running benchmark netperf:
ÂÂÂÂÂ halt_poll_threshold=0ÂÂÂÂÂ -- 29031.6 bit/s -- 76.1Â %CPU
ÂÂÂÂÂ halt_poll_threshold=10000Â -- 29021.7 bit/s -- 105.1 %CPU
ÂÂÂÂÂ halt_poll_threshold=20000Â -- 33463.5 bit/s -- 128.2 %CPU
ÂÂÂÂÂ halt_poll_threshold=30000Â -- 34436.4 bit/s -- 127.8 %CPU
ÂÂÂÂÂ halt_poll_threshold=40000Â -- 35563.3 bit/s -- 129.6 %CPU
ÂÂÂÂÂ halt_poll_threshold=50000Â -- 35787.7 bit/s -- 129.4 %CPU
ÂÂÂÂÂ halt_poll_threshold=100000 -- 35477.7 bit/s -- 130.0 %CPU
ÂÂÂÂÂ halt_poll_threshold=200000 -- 35877.7 bit/s -- 131.0 %CPU
ÂÂÂÂÂ halt_poll_threshold=300000 -- 35730.0 bit/s -- 132.4 %CPU
ÂÂÂÂÂ halt_poll_threshold=500000 -- 34978.4 bit/s -- 134.2 %CPU


and think of the default value(200000, for x86) of kvm dynamic poll,
I'll set it as the same as kvm dynamic poll.

I also test idle VM with diffrent halt_poll_threshold, which doesn't
make CPU utilization fluctuated..


> > could I only leave paravirt_poll_threshold_ns parameter (the maximum poll time),
> > which is as similar as "adaptive halt-polling" Wanpeng mentioned.. then user can
> > turn it off, or find an appropriate threshold for some odd scenario..
> That way lies utter madness. Maybe add it as a debugfs knob, but exposing it to
> userspace: NAK.
.. so, I will make these 3 parameters by default in next v4.
ÂÂÂÂ paravirt_poll_threshold_ns = 200000
ÂÂÂÂ paravirt_poll_shrink = 2
ÂÂÂÂ paravirt_poll_grow = 2

neither touch the syscal ABI nor expose it to userspace again.


Quan