Re: [PATCH] eventfd: support delayed wakeup for non-semaphore eventfd to reduce cpu utilization

[Wen Yang]

在 2023/4/19 17:12, Christian Brauner 写道:
> On Tue, Apr 18, 2023 at 08:15:03PM -0600, Jens Axboe wrote:
> > On 4/17/23 10:32?AM, Wen Yang wrote:
> > > ? 2023/4/17 22:38, Jens Axboe ??:
> > > > On 4/16/23 5:31?AM, wenyang.linux@xxxxxxxxxxx wrote:
> > > > > From: Wen Yang <wenyang.linux@xxxxxxxxxxx>
> > > > >
> > > > > For the NON SEMAPHORE eventfd, if it's counter has a nonzero value,
> > > > > then a read(2) returns 8 bytes containing that value, and the counter's
> > > > > value is reset to zero. Therefore, in the NON SEMAPHORE scenario,
> > > > > N event_writes vs ONE event_read is possible.
> > > > >
> > > > > However, the current implementation wakes up the read thread immediately
> > > > > in eventfd_write so that the cpu utilization increases unnecessarily.
> > > > >
> > > > > By adding a configurable delay after eventfd_write, these unnecessary
> > > > > wakeup operations are avoided, thereby reducing cpu utilization.
> > > > What's the real world use case of this, and what would the expected
> > > > delay be there? With using a delayed work item for this, there's
> > > > certainly a pretty wide grey zone in terms of delay where this would
> > > > perform considerably worse than not doing any delayed wakeups at all.
> > >
> > > Thanks for your comments.
> > >
> > > We have found that the CPU usage of the message middleware is high in
> > > our environment, because sensor messages from MCU are very frequent
> > > and constantly reported, possibly several hundred thousand times per
> > > second. As a result, the message receiving thread is frequently
> > > awakened to process short messages.
> > >
> > > The following is the simplified test code:
> > > https://github.com/w-simon/tests/blob/master/src/test.c
> > >
> > > And the test code in this patch is further simplified.
> > >
> > > Finally, only a configuration item has been added here, allowing users
> > > to make more choices.
> > I think you'd have a higher chance of getting this in if the delay
> > setting was per eventfd context, rather than a global thing.

Thank you.
We will follow your suggestion to change the global configuration to per eventfd.

> That patch seems really weird. Is that an established paradigm to
> address problems like this through a configured wakeup delay? Because
> naively this looks like a pretty brutal hack.

Thanks.

Well, what you are concerned about may be that the rough delay may cause 
additional problems, which is indeed worth considering.

Meanwhile, prolonged and frequent write_eventfd calls are actually 
another type of attack.

If we change it to this:

When a continuous write_eventfd reaches a certain threshold in a short 
period of time, a delay is added as a penalty.

Do you think this is acceptable?


--

Best wishes,

Wen