Re: [Qemu-devel] [PATCH v1 00/17] dataplane: optimization and multi virtqueue support

[Stefan Hajnoczi]

On Tue, Aug 05, 2014 at 06:00:22PM +0800, Ming Lei wrote:
> On Tue, Aug 5, 2014 at 5:48 PM, Kevin Wolf <address@hidden> wrote:
> > Am 05.08.2014 um 05:33 hat Ming Lei geschrieben:
> >> Hi,
> >>
> >> These patches bring up below 4 changes:
> >>         - introduce object allocation pool and apply it to
> >>         virtio-blk dataplane for improving its performance
> >>
> >>         - introduce selective coroutine bypass mechanism
> >>         for improving performance of virtio-blk dataplane with
> >>         raw format image
> >
> > Before applying any bypassing patches, I think we should understand in
> > detail where we are losing performance with coroutines enabled.
> 
> From the below profiling data, CPU becomes slow to run instructions
> with coroutine, and CPU dcache miss is increased so it is very
> likely caused by switching stack frequently.
> 
> http://marc.info/?l=qemu-devel&m=140679721126306&w=2
> 
> http://pastebin.com/ae0vnQ6V

I have been wondering how to prove that the root cause is the ucontext
coroutine mechanism (stack switching).  Here is an idea:

Hack your "bypass" code path to run the request inside a coroutine.
That way you can compare "bypass without coroutine" against "bypass with
coroutine".

Right now I think there are doubts because the bypass code path is
indeed a different (and not 100% correct) code path.  So this approach
might prove that the coroutines are adding the overhead and not
something that you bypassed.

Stefan

pgp3AudxGHZ1j.pgp
Description: PGP signature