Re: [RFC PATCH v2 0/7] x86/idle: add halt poll support

[Quan Xu]

on 2017/9/13 19:56, Yang Zhang wrote:
> On 2017/8/29 22:56, Michael S. Tsirkin wrote:
> > On Tue, Aug 29, 2017 at 11:46:34AM +0000, Yang Zhang wrote:
> > > Some latency-intensive workload will see obviously performance
> > > drop when running inside VM.
> >
> > But are we trading a lot of CPU for a bit of lower latency?
> >
> > > The main reason is that the overhead
> > > is amplified when running inside VM. The most cost i have seen is
> > > inside idle path.
> > >
> > > This patch introduces a new mechanism to poll for a while before
> > > entering idle state. If schedule is needed during poll, then we
> > > don't need to goes through the heavy overhead path.
> >
> > Isn't it the job of an idle driver to find the best way to
> > halt the CPU?
> >
> > It looks like just by adding a cstate we can make it
> > halt at higher latencies only. And at lower latencies,
> > if it's doing a good job we can hopefully use mwait to
> > stop the CPU.
> >
> > In fact I have been experimenting with exactly that.
> > Some initial results are encouraging but I could use help
> > with testing and especially tuning. If you can help
> > pls let me know!
>
> Quan, Can you help to test it and give result? Thanks.

Hi, MST

I have tested the patch "intel_idle: add pv cstates when running on 
kvm"Â on a recent host that allows guests
to execute mwait without an exit. also I have tested our patch "[RFC 
PATCH v2 0/7] x86/idle: add halt poll support",
upstream linux, and idle=poll.

the following is the result (which seems better than ever berfore, as I 
ran test case on a more powerful machine):

for __netperf__, the first column is trans. rate per sec, the second 
column is CPU utilzation.

1. upstream linux

ÂÂÂÂÂ 28371.7 bits/s -- 76.6 %CPU

2. idle=poll

ÂÂÂÂÂ 34372 bit/s -- 999.3 %CPU

3. "[RFC PATCH v2 0/7] x86/idle: add halt poll support", with different 
values of parameter 'halt_poll_threshold':

ÂÂÂÂÂ 28362.7 bits/s -- 74.7Â %CPU (halt_poll_threshold=10000)
ÂÂÂÂÂ 32949.5 bits/s -- 82.5Â %CPU (halt_poll_threshold=20000)
ÂÂÂÂÂ 39717.9 bits/s -- 104.1 %CPU (halt_poll_threshold=30000)
ÂÂÂÂÂ 40137.9 bits/s -- 104.4 %CPU (halt_poll_threshold=40000)
ÂÂÂÂÂ 40079.8 bits/s -- 105.6 %CPU (halt_poll_threshold=50000)


4. "intel_idle: add pv cstates when running on kvm"

 33041.8 bits/s -- 999.4 %CPU





for __ctxsw__, the first column is the time per process context 
switches, the second column is CPU utilzation..

1. upstream linux

ÂÂÂÂÂ 3624.19 ns/ctxsw -- 191.9 %CPU

2. idle=poll

ÂÂÂÂÂ 3419.66 ns/ctxsw -- 999.2 %CPU

3. "[RFC PATCH v2 0/7] x86/idle: add halt poll support", with different 
values of parameter 'halt_poll_threshold':

ÂÂÂÂÂ 1123.40 ns/ctxsw -- 199.6 %CPU (halt_poll_threshold=10000)
ÂÂÂÂÂ 1127.38 ns/ctxsw -- 199.7 %CPU (halt_poll_threshold=20000)
ÂÂÂÂÂ 1113.58 ns/ctxsw -- 199.6 %CPU (halt_poll_threshold=30000)
ÂÂÂÂÂ 1117.12 ns/ctxsw -- 199.6 %CPU (halt_poll_threshold=40000)
ÂÂÂÂÂ 1121.62 ns/ctxsw -- 199.6 %CPU (halt_poll_threshold=50000)

Â4. "intel_idle: add pv cstates when running on kvm"

ÂÂÂÂÂ 3427.59 ns/ctxsw -- 999.4 %CPU

-Quan