From: "Paul E. McKenney" <paulmck@linux.ibm.com>
To: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: linux-kernel@vger.kernel.org,
Josh Triplett <josh@joshtriplett.org>,
Steven Rostedt <rostedt@goodmis.org>,
Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
Lai Jiangshan <jiangshanlai@gmail.com>,
Joel Fernandes <joel@joelfernandes.org>,
tglx@linutronix.de, Mike Galbraith <efault@gmx.de>
Subject: Re: [PATCH v2] rcu: Allow to eliminate softirq processing from rcutree
Date: Wed, 20 Mar 2019 10:30:01 -0700 [thread overview]
Message-ID: <20190320173001.GM4102@linux.ibm.com> (raw)
In-Reply-To: <20190320163532.mr32oi53iaueuizw@linutronix.de>
On Wed, Mar 20, 2019 at 05:35:32PM +0100, Sebastian Andrzej Siewior wrote:
> On 2019-03-20 09:15:00 [-0700], Paul E. McKenney wrote:
> > I am considering making it a module_param() to avoid namespace pollution,
> > as it would become something like rcutree.nosoftirq.
> >
> > Thoughts?
>
> nope, perfect.
Please see below for an untested patch. Thoughts?
Thanx, Paul
------------------------------------------------------------------------
diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
index d377a2166b79..767cdea30a1c 100644
--- a/Documentation/admin-guide/kernel-parameters.txt
+++ b/Documentation/admin-guide/kernel-parameters.txt
@@ -3672,6 +3672,10 @@
the propagation of recent CPU-hotplug changes up
the rcu_node combining tree.
+ rcutree.nosoftirq= [KNL]
+ If set, move all RCU_SOFTIRQ processing to per-CPU
+ rcuc kthreads. Defaults to using RCU_SOFTIRQ.
+
rcutree.rcu_fanout_exact= [KNL]
Disable autobalancing of the rcu_node combining
tree. This is used by rcutorture, and might
diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
index 961dbc7b8949..e4baba8800f3 100644
--- a/kernel/rcu/tree.c
+++ b/kernel/rcu/tree.c
@@ -98,6 +98,9 @@ struct rcu_state rcu_state = {
/* Dump rcu_node combining tree at boot to verify correct setup. */
static bool dump_tree;
module_param(dump_tree, bool, 0444);
+/* Move RCU_SOFTIRQ to rcuc kthreads. */
+static bool nosoftirq;
+module_param(nosoftirq, bool, 0444);
/* Control rcu_node-tree auto-balancing at boot time. */
static bool rcu_fanout_exact;
module_param(rcu_fanout_exact, bool, 0444);
@@ -2330,15 +2333,6 @@ static void rcu_wake_cond(struct task_struct *t, int status)
wake_up_process(t);
}
-static bool rcu_softirq_enabled = true;
-
-static int __init rcunosoftirq_setup(char *str)
-{
- rcu_softirq_enabled = false;
- return 0;
-}
-__setup("rcunosoftirq", rcunosoftirq_setup);
-
/*
* Wake up this CPU's rcuc kthread to do RCU core processing.
*/
@@ -2349,7 +2343,7 @@ static void invoke_rcu_core(void)
if (!cpu_online(smp_processor_id()))
return;
- if (rcu_softirq_enabled) {
+ if (!nosoftirq) {
raise_softirq(RCU_SOFTIRQ);
} else {
local_irq_save(flags);
@@ -2424,7 +2418,7 @@ static int __init rcu_spawn_core_kthreads(void)
for_each_possible_cpu(cpu)
per_cpu(rcu_data.rcu_cpu_has_work, cpu) = 0;
- if (!IS_ENABLED(CONFIG_RCU_BOOST) && rcu_softirq_enabled)
+ if (!IS_ENABLED(CONFIG_RCU_BOOST) && !nosoftirq)
return 0;
WARN_ONCE(smpboot_register_percpu_thread(&rcu_cpu_thread_spec), "%s: Could not start rcub kthread, OOM is now expected behavior\n", __func__);
return 0;
@@ -3467,7 +3461,7 @@ void __init rcu_init(void)
rcu_init_one();
if (dump_tree)
rcu_dump_rcu_node_tree();
- if (rcu_softirq_enabled)
+ if (!nosoftirq)
open_softirq(RCU_SOFTIRQ, rcu_core_si);
/*
diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h
index eb99e750a930..c5a2acb2c7af 100644
--- a/kernel/rcu/tree_plugin.h
+++ b/kernel/rcu/tree_plugin.h
@@ -72,6 +72,8 @@ static void __init rcu_bootup_announce_oddness(void)
pr_info("\tRCU debug GP init slowdown %d jiffies.\n", gp_init_delay);
if (gp_cleanup_delay)
pr_info("\tRCU debug GP init slowdown %d jiffies.\n", gp_cleanup_delay);
+ if (nosoftirq)
+ pr_info("\tRCU_SOFTIRQ processing moved to rcuc kthreads.\n");
if (IS_ENABLED(CONFIG_RCU_EQS_DEBUG))
pr_info("\tRCU debug extended QS entry/exit.\n");
rcupdate_announce_bootup_oddness();
next prev parent reply other threads:[~2019-03-20 17:29 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-03-15 11:11 [PATCH] rcu: Allow to eliminate softirq processing from rcutree Sebastian Andrzej Siewior
2019-03-15 13:35 ` Steven Rostedt
2019-03-15 13:57 ` Sebastian Andrzej Siewior
2019-03-18 2:24 ` Paul E. McKenney
2019-03-19 11:44 ` [PATCH v2] " Sebastian Andrzej Siewior
2019-03-19 15:59 ` Paul E. McKenney
2019-03-19 16:24 ` Sebastian Andrzej Siewior
2019-03-19 16:50 ` Paul E. McKenney
2019-03-19 17:02 ` Sebastian Andrzej Siewior
2019-03-20 11:32 ` Sebastian Andrzej Siewior
2019-03-20 15:21 ` Paul E. McKenney
2019-03-20 15:44 ` Paul E. McKenney
2019-03-20 16:05 ` Sebastian Andrzej Siewior
2019-03-20 16:15 ` Paul E. McKenney
2019-03-20 16:35 ` Sebastian Andrzej Siewior
2019-03-20 17:30 ` Paul E. McKenney [this message]
2019-03-20 17:59 ` Sebastian Andrzej Siewior
2019-03-20 18:12 ` Paul E. McKenney
2019-03-20 18:14 ` Sebastian Andrzej Siewior
2019-03-20 21:13 ` [PATCH v3] " Sebastian Andrzej Siewior
2019-03-20 23:46 ` Paul E. McKenney
2019-03-21 8:27 ` Sebastian Andrzej Siewior
2019-03-21 13:26 ` Paul E. McKenney
2019-03-21 23:32 ` Paul E. McKenney
2019-03-22 7:35 ` Paul E. McKenney
2019-03-22 12:43 ` Paul E. McKenney
2019-03-22 13:42 ` Joel Fernandes
2019-03-22 14:58 ` Paul E. McKenney
2019-03-22 15:50 ` Joel Fernandes
2019-03-22 16:26 ` Paul E. McKenney
2019-03-22 18:07 ` Paul E. McKenney
2019-03-22 23:48 ` Joel Fernandes
2019-03-23 0:25 ` Paul E. McKenney
2019-03-23 1:04 ` Joel Fernandes
2019-03-23 16:10 ` Paul E. McKenney
2019-03-24 23:42 ` Paul E. McKenney
2019-03-25 13:41 ` Joel Fernandes
2019-03-25 15:08 ` Paul E. McKenney
2019-03-25 15:52 ` Paul E. McKenney
2019-03-20 0:26 ` [PATCH] " Joel Fernandes
2019-03-20 11:28 ` Sebastian Andrzej Siewior
2019-03-21 12:06 ` Joel Fernandes
2019-03-21 13:52 ` Paul E. McKenney
2019-03-20 15:24 ` 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=20190320173001.GM4102@linux.ibm.com \
--to=paulmck@linux.ibm.com \
--cc=bigeasy@linutronix.de \
--cc=efault@gmx.de \
--cc=jiangshanlai@gmail.com \
--cc=joel@joelfernandes.org \
--cc=josh@joshtriplett.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mathieu.desnoyers@efficios.com \
--cc=rostedt@goodmis.org \
--cc=tglx@linutronix.de \
/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).