From: Frederic Weisbecker <fweisbec@gmail.com>
To: LKML <linux-kernel@vger.kernel.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>,
Ingo Molnar <mingo@kernel.org>,
"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>,
Peter Zijlstra <peterz@infradead.org>,
Steven Rostedt <rostedt@goodmis.org>,
Thomas Gleixner <tglx@linutronix.de>,
Viresh Kumar <viresh.kumar@linaro.org>
Subject: [PATCH 05/10] smp: Fast path check on IPI list
Date: Sat, 19 Jul 2014 02:44:16 +0200 [thread overview]
Message-ID: <1405730661-9355-6-git-send-email-fweisbec@gmail.com> (raw)
In-Reply-To: <1405730661-9355-1-git-send-email-fweisbec@gmail.com>
When we enqueue a remote irq work, we trigger the same IPI as those
raised by smp_call_function_*() family.
So when we receive such IPI, we check both irq_work and smp_call_function
queues. Thus if we trigger a remote irq work, we'll likely find the
smp_call_function queue empty unless we collide with concurrent enqueuers
but the probability is low.
Meanwhile, checking the smp_call_function queue can be costly because
we use llist_del_all() which relies on cmpxchg().
We can reduce this overhead by doing a fast path check with llist_empty().
Given the implicit IPI ordering:
Enqueuer Dequeuer
--------- --------
llist_add(csd, queue) get_IPI() {
send_IPI() if (llist_empty(queue)
...
When the IPI is sent, we are guaranteed that the IPI receiver will
see the new csd.
So lets do the fast path check to optimize non smp_call_function() related
jobs.
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
---
kernel/smp.c | 11 ++++++++++-
1 file changed, 10 insertions(+), 1 deletion(-)
diff --git a/kernel/smp.c b/kernel/smp.c
index a1812d1..34378d4 100644
--- a/kernel/smp.c
+++ b/kernel/smp.c
@@ -184,11 +184,19 @@ static int generic_exec_single(int cpu, struct call_single_data *csd,
*/
void generic_smp_call_function_single_interrupt(void)
{
+ struct llist_head *head = &__get_cpu_var(call_single_queue);
struct llist_node *entry;
struct call_single_data *csd, *csd_next;
static bool warned;
- entry = llist_del_all(&__get_cpu_var(call_single_queue));
+ /*
+ * Fast check: in case of irq work remote queue, the IPI list
+ * is likely empty. We can spare the expensive llist_del_all().
+ */
+ if (llist_empty(head))
+ goto irq_work;
+
+ entry = llist_del_all(head);
entry = llist_reverse_order(entry);
/*
@@ -212,6 +220,7 @@ void generic_smp_call_function_single_interrupt(void)
csd_unlock(csd);
}
+irq_work:
/*
* Handle irq works queued remotely by irq_work_queue_on().
* Smp functions above are typically synchronous so they
--
1.8.3.1
next prev parent reply other threads:[~2014-07-19 0:47 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-07-19 0:44 [RFC PATCH 00/10] nohz: Support sysidle (and some more cleanups) Frederic Weisbecker
2014-07-19 0:44 ` [PATCH 01/10] irq_work: Introduce void irq work Frederic Weisbecker
2014-07-21 18:16 ` Paul E. McKenney
2014-07-19 0:44 ` [PATCH 02/10] nohz: Kick full dynticks timer targets with an empty IPI Frederic Weisbecker
2014-07-19 7:19 ` Peter Zijlstra
2014-07-19 13:18 ` Frederic Weisbecker
2014-07-19 13:47 ` Peter Zijlstra
2014-07-19 13:54 ` Frederic Weisbecker
2014-07-19 0:44 ` [PATCH 03/10] rcu: Kick full dynticks CPU on extended grace period with a void IRQ Frederic Weisbecker
2014-07-21 18:16 ` Paul E. McKenney
2014-07-19 0:44 ` [PATCH 04/10] nohz: Appropriate timekeeper kick on sysidle break Frederic Weisbecker
2014-07-21 18:15 ` Paul E. McKenney
2014-07-19 0:44 ` Frederic Weisbecker [this message]
2014-07-19 0:44 ` [PATCH 06/10] nohz: Define meaningful symbol for nohz full timekeeper Frederic Weisbecker
2014-07-21 18:11 ` Paul E. McKenney
2014-07-21 18:12 ` Paul E. McKenney
2014-07-25 21:27 ` Frederic Weisbecker
2014-07-19 0:44 ` [PATCH 07/10] nohz: Enforce timekeeping on CPU 0 Frederic Weisbecker
2014-07-19 17:31 ` Nicolas Pitre
2014-07-19 18:31 ` Peter Zijlstra
2014-07-19 18:46 ` Nicolas Pitre
2014-07-19 19:45 ` Peter Zijlstra
2014-07-20 1:07 ` Frederic Weisbecker
2014-07-19 0:44 ` [PATCH 08/10] nohz: Fetch timekeeping max deferment only for timekeeper Frederic Weisbecker
2014-07-19 0:44 ` [PATCH 09/10] nohz: Switch nohz full timekeeper to dynticks idle on top of sysidle detection Frederic Weisbecker
2014-07-21 17:50 ` Paul E. McKenney
2014-07-19 0:44 ` [PATCH 10/10] nohz: Warn on illegal timekeeper switch in nohz full Frederic Weisbecker
2014-07-21 17:51 ` Paul E. McKenney
2014-07-28 17:37 [PATCH 00/10] nohz: Support sysidle (+ some more nohz kick cleanups) Frederic Weisbecker
2014-07-28 17:37 ` [PATCH 05/10] smp: Fast path check on IPI list Frederic Weisbecker
2014-07-29 12:07 ` Peter Zijlstra
2014-07-29 21:54 ` Frederic Weisbecker
2014-07-29 21:55 ` 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=1405730661-9355-6-git-send-email-fweisbec@gmail.com \
--to=fweisbec@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=paulmck@linux.vnet.ibm.com \
--cc=peterz@infradead.org \
--cc=rostedt@goodmis.org \
--cc=tglx@linutronix.de \
--cc=viresh.kumar@linaro.org \
/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
all inboxes | Powered by JetHome®