linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Peter Zijlstra <peterz@infradead.org>
To: Frederic Weisbecker <fweisbec@gmail.com>
Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>,
	linux-kernel@vger.kernel.org, mingo@elte.hu,
	laijs@cn.fujitsu.com, dipankar@in.ibm.com,
	akpm@linux-foundation.org, mathieu.desnoyers@polymtl.ca,
	josh@joshtriplett.org, niv@us.ibm.com, tglx@linutronix.de,
	rostedt@goodmis.org, Valdis.Kletnieks@vt.edu,
	dhowells@redhat.com, eric.dumazet@gmail.com, darren@dvhart.com,
	sbw@mit.edu, patches@linaro.org,
	Alessio Igor Bogani <abogani@kernel.org>,
	Avi Kivity <avi@redhat.com>, Chris Metcalf <cmetcalf@tilera.com>,
	Christoph Lameter <cl@linux.com>,
	Geoff Levand <geoff@infradead.org>,
	Gilad Ben Yossef <gilad@benyossef.com>,
	Hakan Akkan <hakanakkan@gmail.com>,
	"H. Peter Anvin" <hpa@zytor.com>, Ingo Molnar <mingo@kernel.org>,
	Kevin Hilman <khilman@ti.com>,
	Max Krasnyansky <maxk@qualcomm.com>,
	Stephen Hemminger <shemminger@vyatta.com>,
	Sven-Thorsten Dietrich <thebigcorporation@gmail.com>
Subject: Re: [PATCH tip/core/rcu 11/26] rcu: Exit RCU extended QS on user preemption
Date: Wed, 12 Sep 2012 14:41:36 +0200	[thread overview]
Message-ID: <1347453696.15764.24.camel@twins> (raw)
In-Reply-To: <20120912120627.GA14705@somewhere>

On Wed, 2012-09-12 at 14:06 +0200, Frederic Weisbecker wrote:
> 
> 1) This can happen if something calls set_need_resched() while no other task is
> on the runqueue.

People really shouldn't be doing that... I think I know why RCU does
this, but yuck. I also think RCU can avoid doing this, but its a toss up
if that's worth the trouble.

> 2) Remote wake up done but we haven't yet received the schedule IPI.
> 
> 3) Non IPI remote wakeup you're referring above, I'm not sure
> what you mean though.

Well there's two ways of doing remote wakeups, one is doing the wakeup
from the waking cpu and sending an IPI over to reschedule iff you need
wakeup-preemption, the other is queueing the task remotely and sending
an IPI to do the wakeup on the remote cpu.

The former has the problem, the latter not. 

See ttwu_queue().

We could of course mandate that all remote wakeups to special nohz cpus
get queued. That would just leave us with RCU and it would simply not
send resched IPIs to extended quiescent CPUs anyway, right?

So at that point all return to user schedule() calls have nr_running > 1
and the tick is running and RCU is not in extended quiescent state.
Since either we had nr_running > 1 and pre and post state are the same,
or we had nr_running == 1 and we just got a fresh wakeup pushing it to
2, the wakeup will have executed on our cpu and have re-started the tick
and kicked RCU into active gear again.

We cannot hit return to user schedule() with nr_running == 0, simply
because in that case there's no userspace to return to, only the idle
thread and that's very much not userspace :-)

Hmm ?

  reply	other threads:[~2012-09-12 12:48 UTC|newest]

Thread overview: 73+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-08-30 21:05 [PATCH tip/core/rcu 0/26] idle-related changes Paul E. McKenney
2012-08-30 21:05 ` [PATCH tip/core/rcu 01/26] rcu: New rcu_user_enter() and rcu_user_exit() APIs Paul E. McKenney
2012-08-30 21:05   ` [PATCH tip/core/rcu 02/26] rcu: New rcu_user_enter_irq() and rcu_user_exit_irq() APIs Paul E. McKenney
2012-08-31 19:13     ` Josh Triplett
2012-08-31 19:54       ` Frederic Weisbecker
2012-08-31 21:38         ` Josh Triplett
2012-08-30 21:05   ` [PATCH tip/core/rcu 03/26] rcu: Make RCU_FAST_NO_HZ handle adaptive ticks Paul E. McKenney
2012-08-31 23:40     ` Josh Triplett
2012-08-30 21:05   ` [PATCH tip/core/rcu 04/26] rcu: Settle config for userspace extended quiescent state Paul E. McKenney
2012-08-31 23:44     ` Josh Triplett
2012-09-05  0:34       ` Paul E. McKenney
2012-09-05  0:46         ` Josh Triplett
2012-09-10 21:35           ` Frederic Weisbecker
2012-08-30 21:05   ` [PATCH tip/core/rcu 05/26] rcu: Allow rcu_user_enter()/exit() to nest Paul E. McKenney
2012-08-31 23:45     ` Josh Triplett
2012-08-30 21:05   ` [PATCH tip/core/rcu 06/26] rcu: Ignore userspace extended quiescent state by default Paul E. McKenney
2012-08-31 23:46     ` Josh Triplett
2012-08-30 21:05   ` [PATCH tip/core/rcu 07/26] rcu: Switch task's syscall hooks on context switch Paul E. McKenney
2012-08-31 23:48     ` Josh Triplett
2012-08-30 21:05   ` [PATCH tip/core/rcu 08/26] x86: Syscall hooks for userspace RCU extended QS Paul E. McKenney
2012-08-31 23:59     ` Josh Triplett
2012-09-10 21:30       ` Frederic Weisbecker
2012-08-30 21:05   ` [PATCH tip/core/rcu 09/26] x86: Exception " Paul E. McKenney
2012-08-31 23:51     ` Josh Triplett
2012-08-30 21:05   ` [PATCH tip/core/rcu 10/26] rcu: Exit RCU extended QS on kernel preemption after irq/exception Paul E. McKenney
2012-08-31 23:51     ` Josh Triplett
2012-09-06 16:52     ` Peter Zijlstra
2012-09-10 20:31       ` Frederic Weisbecker
2012-08-30 21:05   ` [PATCH tip/core/rcu 11/26] rcu: Exit RCU extended QS on user preemption Paul E. McKenney
2012-08-31 23:52     ` Josh Triplett
2012-09-06 17:02     ` Peter Zijlstra
2012-09-06 17:13       ` Peter Zijlstra
2012-09-10 20:26         ` Frederic Weisbecker
2012-09-12  9:33           ` Peter Zijlstra
2012-09-12 12:06             ` Frederic Weisbecker
2012-09-12 12:41               ` Peter Zijlstra [this message]
2012-09-12 12:52                 ` Peter Zijlstra
2012-09-12 13:54                   ` Frederic Weisbecker
2012-09-12 14:09                     ` Peter Zijlstra
2012-09-12 13:52                 ` Frederic Weisbecker
2012-09-12 15:21                 ` Paul E. McKenney
2012-08-30 21:05   ` [PATCH tip/core/rcu 12/26] x86: Use the new schedule_user API on userspace preemption Paul E. McKenney
2012-08-31 23:53     ` Josh Triplett
2012-08-30 21:05   ` [PATCH tip/core/rcu 13/26] x86: Exit RCU extended QS on notify resume Paul E. McKenney
2012-08-31 23:54     ` Josh Triplett
2012-08-30 21:05   ` [PATCH tip/core/rcu 14/26] rcu: Userspace RCU extended QS selftest Paul E. McKenney
2012-08-31 23:54     ` Josh Triplett
2012-08-30 21:05   ` [PATCH tip/core/rcu 15/26] alpha: Fix preemption handling in idle loop Paul E. McKenney
2012-08-31 23:55     ` Josh Triplett
2012-08-30 21:05   ` [PATCH tip/core/rcu 16/26] alpha: Add missing RCU idle APIs on " Paul E. McKenney
2012-09-01  0:00     ` Josh Triplett
2012-08-30 21:05   ` [PATCH tip/core/rcu 17/26] cris: " Paul E. McKenney
2012-09-01  0:01     ` Josh Triplett
2012-08-30 21:05   ` [PATCH tip/core/rcu 18/26] frv: " Paul E. McKenney
2012-09-01  0:01     ` Josh Triplett
2012-08-30 21:05   ` [PATCH tip/core/rcu 19/26] h8300: " Paul E. McKenney
2012-09-01  0:02     ` Josh Triplett
2012-08-30 21:05   ` [PATCH tip/core/rcu 20/26] m32r: " Paul E. McKenney
2012-09-01  0:02     ` Josh Triplett
2012-08-30 21:05   ` [PATCH tip/core/rcu 21/26] m68k: " Paul E. McKenney
2012-09-01  0:02     ` Josh Triplett
2012-08-30 21:05   ` [PATCH tip/core/rcu 22/26] mn10300: " Paul E. McKenney
2012-09-01  0:03     ` Josh Triplett
2012-08-30 21:05   ` [PATCH tip/core/rcu 23/26] parisc: " Paul E. McKenney
2012-09-01  0:03     ` Josh Triplett
2012-08-30 21:05   ` [PATCH tip/core/rcu 24/26] score: " Paul E. McKenney
2012-09-01  0:04     ` Josh Triplett
2012-08-30 21:05   ` [PATCH tip/core/rcu 25/26] xtensa: " Paul E. McKenney
2012-09-01  0:05     ` Josh Triplett
2012-08-30 21:05   ` [PATCH tip/core/rcu 26/26] ia64: " Paul E. McKenney
2012-09-01  0:05     ` Josh Triplett
2012-08-31 19:07   ` [PATCH tip/core/rcu 01/26] rcu: New rcu_user_enter() and rcu_user_exit() APIs Josh Triplett
2012-09-05  1:04     ` Paul E. McKenney

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1347453696.15764.24.camel@twins \
    --to=peterz@infradead.org \
    --cc=Valdis.Kletnieks@vt.edu \
    --cc=abogani@kernel.org \
    --cc=akpm@linux-foundation.org \
    --cc=avi@redhat.com \
    --cc=cl@linux.com \
    --cc=cmetcalf@tilera.com \
    --cc=darren@dvhart.com \
    --cc=dhowells@redhat.com \
    --cc=dipankar@in.ibm.com \
    --cc=eric.dumazet@gmail.com \
    --cc=fweisbec@gmail.com \
    --cc=geoff@infradead.org \
    --cc=gilad@benyossef.com \
    --cc=hakanakkan@gmail.com \
    --cc=hpa@zytor.com \
    --cc=josh@joshtriplett.org \
    --cc=khilman@ti.com \
    --cc=laijs@cn.fujitsu.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mathieu.desnoyers@polymtl.ca \
    --cc=maxk@qualcomm.com \
    --cc=mingo@elte.hu \
    --cc=mingo@kernel.org \
    --cc=niv@us.ibm.com \
    --cc=patches@linaro.org \
    --cc=paulmck@linux.vnet.ibm.com \
    --cc=rostedt@goodmis.org \
    --cc=sbw@mit.edu \
    --cc=shemminger@vyatta.com \
    --cc=tglx@linutronix.de \
    --cc=thebigcorporation@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).