RCU Archive on lore.kernel.org
 help / color / Atom feed
* need_heavy_qs flag for PREEMPT=y kernels
@ 2019-08-11 18:08 Joel Fernandes
  2019-08-11 18:34 ` Joel Fernandes
  2019-08-11 21:13 ` Paul E. McKenney
  0 siblings, 2 replies; 31+ messages in thread
From: Joel Fernandes @ 2019-08-11 18:08 UTC (permalink / raw)
  To: rcu, paulmck

Hi Paul, everyone,

I noticed on reading code that the need_heavy_qs check and
rcu_momentary_dyntick_idle() is only called for !PREEMPT kernels. Don't we
need to call this for PREEMPT kernels for the benefit of nohz_full CPUs?

Consider the following events:
1. Kernel is PREEMPT=y configuration.
2. CPU 2 is a nohz_full CPU running only a single task and the tick is off.
3. CPU 2 is running only in kernel mode and does not enter user mode or idle.
4. Grace period thread running on CPU 3 enter the fqs loop.
5. Enough time passes and it sets the need_heavy_qs for CPU2.
6. CPU 2 is still in kernel mode but does cond_resched().
7. cond_resched() does not call rcu_momentary_dyntick_idle() because PREEMPT=y.

Is 7. not calling rcu_momentary_dyntick_idle() a lost opportunity for the FQS
loop to detect that the CPU has crossed a quiescent point?

Is this done so that cond_resched() is fast for PREEMPT=y kernels?

thanks,

 - Joel



^ permalink raw reply	[flat|nested] 31+ messages in thread

end of thread, back to index

Thread overview: 31+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-08-11 18:08 need_heavy_qs flag for PREEMPT=y kernels Joel Fernandes
2019-08-11 18:34 ` Joel Fernandes
2019-08-11 21:16   ` Paul E. McKenney
2019-08-11 21:25     ` Joel Fernandes
2019-08-11 23:30       ` Paul E. McKenney
2019-08-12  1:24         ` Joel Fernandes
2019-08-12  1:40           ` Joel Fernandes
2019-08-12  3:57             ` Paul E. McKenney
2019-08-11 21:13 ` Paul E. McKenney
2019-08-12  3:21   ` Joel Fernandes
2019-08-12  3:53     ` Paul E. McKenney
2019-08-12 21:20       ` Joel Fernandes
2019-08-12 23:01         ` Paul E. McKenney
2019-08-13  1:02           ` Joel Fernandes
2019-08-13  1:05             ` Joel Fernandes
2019-08-13  2:28               ` Paul E. McKenney
2019-08-13  2:27             ` Paul E. McKenney
2019-08-13  2:50               ` Paul E. McKenney
2019-08-15 17:17             ` Paul E. McKenney
2019-08-15 20:04               ` Joel Fernandes
2019-08-15 20:31                 ` Paul E. McKenney
2019-08-15 21:22                   ` Joel Fernandes
2019-08-15 21:27                     ` Joel Fernandes
2019-08-15 21:34                       ` Joel Fernandes
2019-08-15 21:57                         ` Paul E. McKenney
2019-08-15 21:45                     ` Paul E. McKenney
2019-08-16  0:02                       ` Joel Fernandes
2019-08-19 12:34                         ` Frederic Weisbecker
2019-08-19 12:09                   ` Frederic Weisbecker
2019-08-19 16:57                   ` Frederic Weisbecker
2019-08-19 22:31                     ` Paul E. McKenney

RCU Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/rcu/0 rcu/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 rcu rcu/ https://lore.kernel.org/rcu \
		rcu@vger.kernel.org
	public-inbox-index rcu

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.rcu


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git