[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20180709204248.GF3593@linux.vnet.ibm.com>
Date: Mon, 9 Jul 2018 13:42:48 -0700
From: "Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>
To: David Woodhouse <dwmw2@...radead.org>
Cc: Peter Zijlstra <peterz@...radead.org>, mhillenb@...zon.de,
linux-kernel <linux-kernel@...r.kernel.org>
Subject: Re: [RFC] Make need_resched() return true when rcu_urgent_qs
requested
On Mon, Jul 09, 2018 at 09:35:38PM +0100, David Woodhouse wrote:
>
>
> On Mon, 2018-07-09 at 13:34 -0700, Paul E. McKenney wrote:
> >
> > So here are the possible code paths when .rcu_urgent_qs is set to true:
> >
> > 1. A context switch will record the quiescent state and clear
> > .rcu_urgent_qs. (The failure to do the clearing in current -rcu
> > for PREEMPT builds is a performance bug that I need to fix.)
>
> What if there's nothing else runnable and there is no actual context
> switch?
The scheduler invokes rcu_note_context_switch() before looking to see
if there really will or won't be a context switch.
I am sure that Peter will correct me if I am confused on this point. ;-)
Thanx, Paul
> > 2. A cond_resched() will cause rcu_all_qs() to be invoked, which
> > will record a quiescent state and clear .rcu_urgent_qs.
> >
> > 3. With the patch below, a scheduling-clock interrupt of a
> > non-idle non-userspace task will force a reschedule, which
> > will result in #1 above happening.
Powered by blists - more mailing lists