From: Peter Zijlstra <peterz@infradead.org>
To: rjw@rjwysocki.net, bp@alien8.de
Cc: x86@kernel.org, tony.luck@intel.com, lenb@kernel.org,
daniel.lezcano@linaro.org, linux-kernel@vger.kernel.org,
linux-acpi@vger.kernel.org, linux-pm@vger.kernel.org,
ulf.hansson@linaro.org, paulmck@kernel.org, tglx@linutronix.de,
naresh.kamboju@linaro.org, peterz@infradead.org
Subject: [RFC][PATCH 3/4] cpuidle: Allow cpuidle drivers to take over RCU-idle
Date: Tue, 15 Sep 2020 12:32:00 +0200 [thread overview]
Message-ID: <20200915103806.411374792@infradead.org> (raw)
In-Reply-To: 20200915103157.345404192@infradead.org
Some drivers have to do significant work, some of which relies on RCU
still being active. Instead of using RCU_NONIDLE in the drivers and
flipping RCU back on, allow drivers to take over RCU-idle duty.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
---
drivers/cpuidle/cpuidle.c | 15 ++++++++++-----
include/linux/cpuidle.h | 1 +
2 files changed, 11 insertions(+), 5 deletions(-)
--- a/drivers/cpuidle/cpuidle.c
+++ b/drivers/cpuidle/cpuidle.c
@@ -138,6 +138,7 @@ static void enter_s2idle_proper(struct c
struct cpuidle_device *dev, int index)
{
ktime_t time_start, time_end;
+ struct cpuidle_state *target_state = &drv->states[index];
time_start = ns_to_ktime(local_clock());
@@ -153,8 +154,9 @@ static void enter_s2idle_proper(struct c
* suspended is generally unsafe.
*/
stop_critical_timings();
- rcu_idle_enter();
- drv->states[index].enter_s2idle(dev, drv, index);
+ if (!(target_state->flags & CPUIDLE_FLAG_RCU_IDLE))
+ rcu_idle_enter();
+ target_state->enter_s2idle(dev, drv, index);
if (WARN_ON_ONCE(!irqs_disabled()))
local_irq_disable();
/*
@@ -162,7 +164,8 @@ static void enter_s2idle_proper(struct c
* first CPU executing it calls functions containing RCU read-side
* critical sections, so tell RCU about that.
*/
- rcu_idle_exit();
+ if (!(target_state->flags & CPUIDLE_FLAG_RCU_IDLE))
+ rcu_idle_exit();
tick_unfreeze();
start_critical_timings();
@@ -239,9 +242,11 @@ int cpuidle_enter_state(struct cpuidle_d
time_start = ns_to_ktime(local_clock());
stop_critical_timings();
- rcu_idle_enter();
+ if (!(target_state->flags & CPUIDLE_FLAG_RCU_IDLE))
+ rcu_idle_enter();
entered_state = target_state->enter(dev, drv, index);
- rcu_idle_exit();
+ if (!(target_state->flags & CPUIDLE_FLAG_RCU_IDLE))
+ rcu_idle_exit();
start_critical_timings();
sched_clock_idle_wakeup_event();
--- a/include/linux/cpuidle.h
+++ b/include/linux/cpuidle.h
@@ -82,6 +82,7 @@ struct cpuidle_state {
#define CPUIDLE_FLAG_UNUSABLE BIT(3) /* avoid using this state */
#define CPUIDLE_FLAG_OFF BIT(4) /* disable this state by default */
#define CPUIDLE_FLAG_TLB_FLUSHED BIT(5) /* idle-state flushes TLBs */
+#define CPUIDLE_FLAG_RCU_IDLE BIT(6) /* idle-state takes care of RCU */
struct cpuidle_device_kobj;
struct cpuidle_state_kobj;
next prev parent reply other threads:[~2020-09-15 10:44 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-09-15 10:31 [RFC][PATCH 0/4] Fix up ACPI processor idle vs RCU Peter Zijlstra
2020-09-15 10:31 ` [RFC][PATCH 1/4] acpi: Use CPUIDLE_FLAG_TIMER_STOP Peter Zijlstra
2020-09-15 16:26 ` Rafael J. Wysocki
2020-09-16 15:42 ` peterz
2020-09-16 16:01 ` Borislav Petkov
2020-09-16 17:38 ` Rafael J. Wysocki
2020-09-16 17:38 ` Rafael J. Wysocki
2020-09-22 3:26 ` Guenter Roeck
2020-09-22 19:03 ` Rafael J. Wysocki
2020-09-15 10:31 ` [RFC][PATCH 2/4] acpi: Use CPUIDLE_FLAG_TLB_FLUSHED Peter Zijlstra
2020-09-15 10:32 ` Peter Zijlstra [this message]
2020-09-15 13:20 ` [RFC][PATCH 3/4] cpuidle: Allow cpuidle drivers to take over RCU-idle Ulf Hansson
2020-09-15 10:32 ` [RFC][PATCH 4/4] acpi: Take over RCU-idle for C3-BM idle Peter Zijlstra
2020-09-21 9:12 ` Sven Joachim
2020-09-21 10:37 ` [PATCH] rcu/tree: Export rcu_idle_{enter,exit} to module Borislav Petkov
2020-09-21 12:15 ` Rafael J. Wysocki
2020-09-21 13:32 ` Paul E. McKenney
2020-09-21 13:38 ` Rafael J. Wysocki
2020-09-25 15:20 ` [RFC][PATCH 4/4] acpi: Take over RCU-idle for C3-BM idle Guenter Roeck
2020-09-25 15:24 ` Rafael J. Wysocki
2020-09-25 15:29 ` Paul E. McKenney
2020-09-15 18:31 ` [RFC][PATCH 0/4] Fix up ACPI processor idle vs RCU Borislav Petkov
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=20200915103806.411374792@infradead.org \
--to=peterz@infradead.org \
--cc=bp@alien8.de \
--cc=daniel.lezcano@linaro.org \
--cc=lenb@kernel.org \
--cc=linux-acpi@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pm@vger.kernel.org \
--cc=naresh.kamboju@linaro.org \
--cc=paulmck@kernel.org \
--cc=rjw@rjwysocki.net \
--cc=tglx@linutronix.de \
--cc=tony.luck@intel.com \
--cc=ulf.hansson@linaro.org \
--cc=x86@kernel.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;
as well as URLs for NNTP newsgroup(s).