Re: [PATCH RFC v3 1/6] x86/paravirt: Add pv_idle_ops to paravirt ops

[Quan Xu]

On 2017/11/13 18:53, Juergen Gross wrote:
> On 13/11/17 11:06, Quan Xu wrote:
> > From: Quan Xu <quan.xu0@xxxxxxxxx>
> >
> > So far, pv_idle_ops.poll is the only ops for pv_idle. .poll is called
> > in idle path which will poll for a while before we enter the real idle
> > state.
> >
> > In virtualization, idle path includes several heavy operations
> > includes timer access(LAPIC timer or TSC deadline timer) which will
> > hurt performance especially for latency intensive workload like message
> > passing task. The cost is mainly from the vmexit which is a hardware
> > context switch between virtual machine and hypervisor. Our solution is
> > to poll for a while and do not enter real idle path if we can get the
> > schedule event during polling.
> >
> > Poll may cause the CPU waste so we adopt a smart polling mechanism to
> > reduce the useless poll.
> >
> > Signed-off-by: Yang Zhang <yang.zhang.wz@xxxxxxxxx>
> > Signed-off-by: Quan Xu <quan.xu0@xxxxxxxxx>
> > Cc: Juergen Gross <jgross@xxxxxxxx>
> > Cc: Alok Kataria <akataria@xxxxxxxxxx>
> > Cc: Rusty Russell <rusty@xxxxxxxxxxxxxxx>
> > Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
> > Cc: Ingo Molnar <mingo@xxxxxxxxxx>
> > Cc: "H. Peter Anvin" <hpa@xxxxxxxxx>
> > Cc: x86@xxxxxxxxxx
> > Cc: virtualization@xxxxxxxxxxxxxxxxxxxxxxxxxx
> > Cc: linux-kernel@xxxxxxxxxxxxxxx
> > Cc: xen-devel@xxxxxxxxxxxxxxxxxxxx
> Hmm, is the idle entry path really so critical to performance that a new
> pvops function is necessary?
Juergen, Here is the data we get when running benchmark netperf:
Â1. w/o patch and disable kvm dynamic poll (halt_poll_ns=0):
ÂÂÂ 29031.6 bit/s -- 76.1 %CPU

Â2. w/ patch and disable kvm dynamic poll (halt_poll_ns=0):
ÂÂÂ 35787.7 bit/s -- 129.4 %CPU

Â3. w/ kvm dynamic poll:
ÂÂÂ 35735.6 bit/s -- 200.0 %CPU

Â4. w/patch and w/ kvm dynamic poll:
ÂÂÂ 42225.3 bit/s -- 198.7 %CPU

Â5. idle=poll
ÂÂÂ 37081.7 bit/s -- 998.1 %CPU



Âw/ this patch, we will improve performance by 23%.. even we could improve
Âperformance by 45.4%, if we use w/patch and w/ kvm dynamic poll. also the
Âcost of CPU is much lower than 'idle=poll' case..

> Wouldn't a function pointer, maybe guarded
> by a static key, be enough? A further advantage would be that this would
> work on other architectures, too.

I assume this feature will be ported to other archs.. a new pvops makes code
clean and easy to maintain. also I tried to add it into existed pvops, 
but it
doesn't match.



Quan
Alibaba Cloud
> Juergen