Re: [PATCH] Fix: disable sys_membarrier when nohz_full is enabled

From	"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Newsgroups	linux.kernel
Subject	Re: [PATCH] Fix: disable sys_membarrier when nohz_full is enabled
Date	2016-11-17 19:20 +0100
Message-ID	<sEzrX-4eD-9@gated-at.bofh.it> (permalink)
References	(3 earlier) <sEw15-1T0-69@gated-at.bofh.it> <sEw15-1T0-67@gated-at.bofh.it> <sEyvT-3yz-1@gated-at.bofh.it> <sEyFz-3CT-43@gated-at.bofh.it> <sEyYV-3Mb-1@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

On Thu, Nov 17, 2016 at 10:17:25AM -0500, Steven Rostedt wrote:
> On Thu, 17 Nov 2016 15:02:18 +0000 (UTC)
> Mathieu Desnoyers <mathieu.desnoyers@efficios.com> wrote:
> 
> 
> > That's an interesting approach. I would be tempted to give it a
> > per-thread (rather than per-process) scope.
> 
> Sure, per thread, but have it inherit to child processes.
> 
> > 
> > E.g., a thread could do the following to ask to be
> > interrupted by IPIs:
> > 
> > membarrier(MEMBARRIER_CMD_REGISTER_EXPEDITED, 0)
> > 
> > and could unregister with:
> > 
> > membarrier(MEMBARRIER_CMD_UNREGISTER_EXPEDITED, 0)
> 
> Sure why not ;-)

Makes a lot of sense to me!

> > We can then keep a per-thread refcount internally.
> > (not sure the "EXPEDITED" is the right word there...
> > do we want it to be "NOHZ_FULL" instead ?)
> 
> No, it shouldn't mention NOHZ_FULL. Perhaps have all tasks do this
> regardless, even though it will only affect nohz full ones. But in the
> future it may be other tasks as well.
> 
> > 
> > Then in membarrier(MEMBARRIER_CMD_SHARED, 0), for each
> > nohz_full cpu, we grab the rq lock, and only send an IPI
> > if the running thread is registered as "expedited".
> 
> Yeah, something like that. That way it wont interrupt tasks that are
> running in no-hz-full and don't care about this syscall.

And this as well!

							Thanx, Paul

Thread

Re: [PATCH] Fix: disable sys_membarrier when nohz_full is enabled Lai Jiangshan <jiangshanlai@gmail.com> - 2016-11-17 08:00 +0100
  Re: [PATCH] Fix: disable sys_membarrier when nohz_full is enabled Mathieu Desnoyers <mathieu.desnoyers@efficios.com> - 2016-11-17 12:50 +0100
    Re: [PATCH] Fix: disable sys_membarrier when nohz_full is enabled Mathieu Desnoyers <mathieu.desnoyers@efficios.com> - 2016-11-17 15:40 +0100
      Re: [PATCH] Fix: disable sys_membarrier when nohz_full is enabled Steven Rostedt <rostedt@goodmis.org> - 2016-11-17 18:20 +0100
        Re: [PATCH] Fix: disable sys_membarrier when nohz_full is enabled Mathieu Desnoyers <mathieu.desnoyers@efficios.com> - 2016-11-17 18:30 +0100
          Re: [PATCH] Fix: disable sys_membarrier when nohz_full is enabled Steven Rostedt <rostedt@goodmis.org> - 2016-11-17 18:50 +0100
            Re: [PATCH] Fix: disable sys_membarrier when nohz_full is enabled "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2016-11-17 19:20 +0100
      Re: [PATCH] Fix: disable sys_membarrier when nohz_full is enabled "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2016-11-17 19:50 +0100
    Re: [PATCH] Fix: disable sys_membarrier when nohz_full is enabled "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2016-11-17 20:30 +0100
  Re: [PATCH] Fix: disable sys_membarrier when nohz_full is enabled "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> - 2016-11-17 18:30 +0100

csiph-web