From: Frederic Weisbecker <frederic@kernel.org>
To: "Paul E . McKenney" <paulmck@kernel.org>
Cc: LKML <linux-kernel@vger.kernel.org>,
Frederic Weisbecker <frederic@kernel.org>,
Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
Valentin Schneider <Valentin.Schneider@arm.com>,
Peter Zijlstra <peterz@infradead.org>,
Uladzislau Rezki <urezki@gmail.com>,
Thomas Gleixner <tglx@linutronix.de>,
Valentin Schneider <valentin.schneider@arm.com>,
Boqun Feng <boqun.feng@gmail.com>,
Neeraj Upadhyay <neeraju@codeaurora.org>,
Josh Triplett <josh@joshtriplett.org>,
Joel Fernandes <joel@joelfernandes.org>,
rcu@vger.kernel.org
Subject: [PATCH 06/11] rcu/nocb: Check a stable offloaded state to manipulate qlen_last_fqs_check
Date: Mon, 11 Oct 2021 16:51:35 +0200 [thread overview]
Message-ID: <20211011145140.359412-7-frederic@kernel.org> (raw)
In-Reply-To: <20211011145140.359412-1-frederic@kernel.org>
It's not entirely clear why rdp->qlen_last_fqs_check is updated before
processing the queue only on offloaded rdp. There can be different
effect to that, either in favour of triggering the force quiescent state
path or not. For example:
1) If the number of callbacks has decreased since the last
rdp->qlen_last_fqs_check update (because we recently called
rcu_do_batch() and we executed below qhimark callbacks) and the number
of processed callbacks on a subsequent do_batch() arranges for
exceeding qhimark on non-offloaded but not on offloaded setup, then we
may spare a later run to the force quiescent state
slow path on __call_rcu_nocb_wake(), as compared to the non-offloaded
counterpart scenario.
Here is such an offloaded scenario instance:
qhimark = 1000
rdp->last_qlen_last_fqs_check = 3000
rcu_segcblist_n_cbs(rdp) = 2000
rcu_do_batch() {
if (offloaded)
rdp->last_qlen_fqs_check = rcu_segcblist_n_cbs(rdp) // 2000
// run 1000 callback
rcu_segcblist_n_cbs(rdp) = 1000
// Not updating rdp->qlen_last_fqs_check
if (count < rdp->qlen_last_fqs_check - qhimark)
rdp->qlen_last_fqs_check = count;
}
call_rcu() * 1001 {
__call_rcu_nocb_wake() {
// not taking the fqs slowpath:
// rcu_segcblist_n_cbs(rdp) == 2001
// rdp->qlen_last_fqs_check == 2000
// qhimark == 1000
if (len > rdp->qlen_last_fqs_check + qhimark)
...
}
In the case of a non-offloaded scenario, rdp->qlen_last_fqs_check
would be 1000 and the fqs slowpath would have executed.
2) If the number of callbacks has increased since the last
rdp->qlen_last_fqs_check update (because we recently queued below
qhimark callbacks) and the number of callbacks executed in rcu_do_batch()
doesn't exceed qhimark for either offloaded or non-offloaded setup,
then it's possible that the offloaded scenario later run the force
quiescent state slow path on __call_rcu_nocb_wake() while the
non-offloaded doesn't.
qhimark = 1000
rdp->last_qlen_last_fqs_check = 3000
rcu_segcblist_n_cbs(rdp) = 2000
rcu_do_batch() {
if (offloaded)
rdp->last_qlen_last_fqs_check = rcu_segcblist_n_cbs(rdp) // 2000
// run 100 callbacks
// concurrent queued 100
rcu_segcblist_n_cbs(rdp) = 2000
// Not updating rdp->qlen_last_fqs_check
if (count < rdp->qlen_last_fqs_check - qhimark)
rdp->qlen_last_fqs_check = count;
}
call_rcu() * 1001 {
__call_rcu_nocb_wake() {
// Taking the fqs slowpath:
// rcu_segcblist_n_cbs(rdp) == 3001
// rdp->qlen_last_fqs_check == 2000
// qhimark == 1000
if (len > rdp->qlen_last_fqs_check + qhimark)
...
}
In the case of a non-offloaded scenario, rdp->qlen_last_fqs_check
would be 3000 and the fqs slowpath would have executed.
Until we sort this out, keep the current behaviour, whatever the
original intent is, but make sure we check a stable and not volatile
offloading state in order not to raise a useless alarm on -rt
Reported-and-tested-by: Valentin Schneider <valentin.schneider@arm.com>
Tested-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Original-patch-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Cc: Valentin Schneider <valentin.schneider@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: Josh Triplett <josh@joshtriplett.org>
Cc: Joel Fernandes <joel@joelfernandes.org>
Cc: Boqun Feng <boqun.feng@gmail.com>
Cc: Neeraj Upadhyay <neeraju@codeaurora.org>
Cc: Uladzislau Rezki <urezki@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
---
kernel/rcu/tree.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
index a43924244000..27500981d4b1 100644
--- a/kernel/rcu/tree.c
+++ b/kernel/rcu/tree.c
@@ -2508,7 +2508,7 @@ static void rcu_do_batch(struct rcu_data *rdp)
trace_rcu_batch_start(rcu_state.name,
rcu_segcblist_n_cbs(&rdp->cblist), bl);
rcu_segcblist_extract_done_cbs(&rdp->cblist, &rcl);
- if (offloaded)
+ if (rcu_rdp_is_offloaded(rdp))
rdp->qlen_last_fqs_check = rcu_segcblist_n_cbs(&rdp->cblist);
trace_rcu_segcb_stats(&rdp->cblist, TPS("SegCbDequeued"));
--
2.25.1
next prev parent reply other threads:[~2021-10-11 14:52 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-10-11 14:51 [PATCH 00/11] rcu: Make rcu_core() safe in PREEMPT_RT with NOCB + a few other fixes v2 Frederic Weisbecker
2021-10-11 14:51 ` [PATCH 01/11] rcu/nocb: Make local rcu_nocb_lock_irqsave() safe against concurrent deoffloading Frederic Weisbecker
2021-10-11 14:51 ` [PATCH 02/11] rcu/nocb: Prepare state machine for a new step Frederic Weisbecker
2021-10-11 14:51 ` [PATCH 03/11] rcu/nocb: Invoke rcu_core() at the start of deoffloading Frederic Weisbecker
2021-10-13 16:07 ` Boqun Feng
2021-10-14 11:07 ` Frederic Weisbecker
2021-10-14 11:42 ` Valentin Schneider
2021-10-14 13:57 ` Boqun Feng
2021-10-11 14:51 ` [PATCH 04/11] rcu/nocb: Make rcu_core() callbacks acceleration preempt-safe Frederic Weisbecker
2021-10-11 14:51 ` [PATCH 05/11] rcu/nocb: Make rcu_core() callbacks acceleration (de-)offloading safe Frederic Weisbecker
2021-10-11 14:51 ` Frederic Weisbecker [this message]
2021-10-11 14:51 ` [PATCH 07/11] rcu/nocb: Use appropriate rcu_nocb_lock_irqsave() Frederic Weisbecker
2021-10-11 14:51 ` [PATCH 08/11] rcu/nocb: Limit number of softirq callbacks only on softirq Frederic Weisbecker
2021-10-11 14:51 ` [PATCH 09/11] rcu: Fix callbacks processing time limit retaining cond_resched() Frederic Weisbecker
2021-10-11 14:51 ` [PATCH 10/11] rcu: Apply callbacks processing time limit only on softirq Frederic Weisbecker
2021-10-11 14:51 ` [PATCH 11/11] rcu/nocb: Don't invoke local rcu core on callback overload from nocb kthread Frederic Weisbecker
2021-10-13 0:32 ` [PATCH 00/11] rcu: Make rcu_core() safe in PREEMPT_RT with NOCB + a few other fixes v2 Paul E. McKenney
2021-10-13 3:28 ` Paul E. McKenney
2021-10-13 10:01 ` Frederic Weisbecker
2021-10-13 11:43 ` Frederic Weisbecker
2021-10-13 16:27 ` Paul E. McKenney
2021-10-14 10:43 ` Frederic Weisbecker
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=20211011145140.359412-7-frederic@kernel.org \
--to=frederic@kernel.org \
--cc=Valentin.Schneider@arm.com \
--cc=bigeasy@linutronix.de \
--cc=boqun.feng@gmail.com \
--cc=joel@joelfernandes.org \
--cc=josh@joshtriplett.org \
--cc=linux-kernel@vger.kernel.org \
--cc=neeraju@codeaurora.org \
--cc=paulmck@kernel.org \
--cc=peterz@infradead.org \
--cc=rcu@vger.kernel.org \
--cc=tglx@linutronix.de \
--cc=urezki@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.