public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
* wfcqueue (in Userspace RCU) for Linux kernel (for epoll)
@ 2013-03-11 19:35 Eric Wong
  2013-03-11 21:25 ` Mathieu Desnoyers
  0 siblings, 1 reply; 2+ messages in thread
From: Eric Wong @ 2013-03-11 19:35 UTC (permalink / raw)
  To: Mathieu Desnoyers, Lai Jiangshan; +Cc: linux-kernel, Davide Libenzi

Hi, I'm looking to reduce contention for the ep->lock spin lock in epoll.

I came across wfcqueue in Userspace RCU and am wondering if there's any
reason (other that lack of developer time/users) it hasn't been adapted
for the Linux kernel.

I'd be happy to do the work if it's suitable (and omit parts which
may not be, such as busy wait).

Using a mutex lock (while in process context) for dequeue works for
epoll.  I mainly want a lock-free enqueue for ep_poll_callback since
that may be called in IRQ context.

I suppose I can also use the llist LIFO in ep_poll_callback and
reverse that while in process context, too...


ref: http://git.lttng.org/userspace-rcu.git

^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: wfcqueue (in Userspace RCU) for Linux kernel (for epoll)
  2013-03-11 19:35 wfcqueue (in Userspace RCU) for Linux kernel (for epoll) Eric Wong
@ 2013-03-11 21:25 ` Mathieu Desnoyers
  0 siblings, 0 replies; 2+ messages in thread
From: Mathieu Desnoyers @ 2013-03-11 21:25 UTC (permalink / raw)
  To: Eric Wong; +Cc: Lai Jiangshan, linux-kernel, Davide Libenzi

* Eric Wong (normalperson@yhbt.net) wrote:
> Hi, I'm looking to reduce contention for the ep->lock spin lock in epoll.
> 
> I came across wfcqueue in Userspace RCU and am wondering if there's any
> reason (other that lack of developer time/users) it hasn't been adapted
> for the Linux kernel.
> 
> I'd be happy to do the work if it's suitable (and omit parts which
> may not be, such as busy wait).

Hi Eric,

I'm glad you're interested in our data structure!

I actually have a port ready, I just lacked the time for thorough
testing and also lacked users of the API at kernel-level. It looks like
now would be a good time to send the patch your way as a RFC!

> Using a mutex lock (while in process context) for dequeue works for
> epoll.  I mainly want a lock-free enqueue for ep_poll_callback since
> that may be called in IRQ context.

You'd get even better than this: wait-free enqueue, with a simple xchg
and store.

> 
> I suppose I can also use the llist LIFO in ep_poll_callback and
> reverse that while in process context, too...

Yeah, but that seems more expensive than just having the list in the
right order already. ;)

Patch incoming in a separate post,

Thanks,

Mathieu

> 
> 
> ref: http://git.lttng.org/userspace-rcu.git

-- 
Mathieu Desnoyers
EfficiOS Inc.
http://www.efficios.com

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2013-03-11 21:25 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2013-03-11 19:35 wfcqueue (in Userspace RCU) for Linux kernel (for epoll) Eric Wong
2013-03-11 21:25 ` Mathieu Desnoyers

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox