From: Peter Zijlstra <peterz@infradead.org>
To: linux-kernel@vger.kernel.org, mingo@kernel.org, will@kernel.org
Cc: npiggin@gmail.com, elver@google.com, jgross@suse.com,
paulmck@kernel.org, rostedt@goodmis.org, rjw@rjwysocki.net,
joel@joelfernandes.org, svens@linux.ibm.com, tglx@linutronix.de,
peterz@infradead.org
Subject: [PATCH 2/9] sched,idle,rcu: Push rcu_idle deeper into the idle path
Date: Thu, 20 Aug 2020 09:30:33 +0200 [thread overview]
Message-ID: <20200820074640.401636694@infradead.org> (raw)
In-Reply-To: 20200820073031.886217423@infradead.org
Lots of things take locks, due to a wee bug, rcu_lockdep didn't notice
that the locking tracepoints were using RCU.
Push rcu_idle_{enter,exit}() as deep as possible into the idle paths,
this also resolves a lot of _rcuidle()/RCU_NONIDLE() usage.
Specifically, sched_clock_idle_wakeup_event() will use ktime which
will use seqlocks which will tickle lockdep.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
---
drivers/cpuidle/cpuidle.c | 15 +++++++++++----
kernel/sched/idle.c | 18 ++++++------------
kernel/time/tick-broadcast-hrtimer.c | 29 ++++++++++++-----------------
3 files changed, 29 insertions(+), 33 deletions(-)
--- a/drivers/cpuidle/cpuidle.c
+++ b/drivers/cpuidle/cpuidle.c
@@ -145,22 +145,27 @@ static void enter_s2idle_proper(struct c
* executing it contains RCU usage regarded as invalid in the idle
* context, so tell RCU about that.
*/
- RCU_NONIDLE(tick_freeze());
+ tick_freeze();
/*
* The state used here cannot be a "coupled" one, because the "coupled"
* cpuidle mechanism enables interrupts and doing that with timekeeping
* suspended is generally unsafe.
*/
+ rcu_idle_enter();
stop_critical_timings();
drv->states[index].enter_s2idle(dev, drv, index);
- WARN_ON(!irqs_disabled());
+ if (WARN_ON(!irqs_disabled()))
+ local_irq_disable();
/*
* timekeeping_resume() that will be called by tick_unfreeze() for the
* first CPU executing it calls functions containing RCU read-side
* critical sections, so tell RCU about that.
+ *
+ * FIXME, order against rcu_idle_exit ?
*/
RCU_NONIDLE(tick_unfreeze());
start_critical_timings();
+ rcu_idle_exit();
time_end = ns_to_ktime(local_clock());
@@ -228,16 +233,18 @@ int cpuidle_enter_state(struct cpuidle_d
/* Take note of the planned idle state. */
sched_idle_set_state(target_state);
- trace_cpu_idle_rcuidle(index, dev->cpu);
+ trace_cpu_idle(index, dev->cpu);
time_start = ns_to_ktime(local_clock());
+ rcu_idle_enter();
stop_critical_timings();
entered_state = target_state->enter(dev, drv, index);
start_critical_timings();
+ rcu_idle_exit();
sched_clock_idle_wakeup_event();
time_end = ns_to_ktime(local_clock());
- trace_cpu_idle_rcuidle(PWR_EVENT_EXIT, dev->cpu);
+ trace_cpu_idle(PWR_EVENT_EXIT, dev->cpu);
/* The cpu is no longer idle or about to enter idle. */
sched_idle_set_state(NULL);
--- a/kernel/sched/idle.c
+++ b/kernel/sched/idle.c
@@ -54,17 +54,18 @@ __setup("hlt", cpu_idle_nopoll_setup);
static noinline int __cpuidle cpu_idle_poll(void)
{
+ trace_cpu_idle(0, smp_processor_id());
rcu_idle_enter();
- trace_cpu_idle_rcuidle(0, smp_processor_id());
local_irq_enable();
stop_critical_timings();
while (!tif_need_resched() &&
- (cpu_idle_force_poll || tick_check_broadcast_expired()))
+ (cpu_idle_force_poll || tick_check_broadcast_expired()))
cpu_relax();
+
start_critical_timings();
- trace_cpu_idle_rcuidle(PWR_EVENT_EXIT, smp_processor_id());
rcu_idle_exit();
+ trace_cpu_idle(PWR_EVENT_EXIT, smp_processor_id());
return 1;
}
@@ -90,9 +91,11 @@ void __cpuidle default_idle_call(void)
if (current_clr_polling_and_test()) {
local_irq_enable();
} else {
+ rcu_idle_enter();
stop_critical_timings();
arch_cpu_idle();
start_critical_timings();
+ rcu_idle_exit();
}
}
@@ -158,7 +161,6 @@ static void cpuidle_idle_call(void)
if (cpuidle_not_available(drv, dev)) {
tick_nohz_idle_stop_tick();
- rcu_idle_enter();
default_idle_call();
goto exit_idle;
@@ -178,21 +180,17 @@ static void cpuidle_idle_call(void)
u64 max_latency_ns;
if (idle_should_enter_s2idle()) {
- rcu_idle_enter();
entered_state = call_cpuidle_s2idle(drv, dev);
if (entered_state > 0)
goto exit_idle;
- rcu_idle_exit();
-
max_latency_ns = U64_MAX;
} else {
max_latency_ns = dev->forced_idle_latency_limit_ns;
}
tick_nohz_idle_stop_tick();
- rcu_idle_enter();
next_state = cpuidle_find_deepest_state(drv, dev, max_latency_ns);
call_cpuidle(drv, dev, next_state);
@@ -209,8 +207,6 @@ static void cpuidle_idle_call(void)
else
tick_nohz_idle_retain_tick();
- rcu_idle_enter();
-
entered_state = call_cpuidle(drv, dev, next_state);
/*
* Give the governor an opportunity to reflect on the outcome
@@ -226,8 +222,6 @@ static void cpuidle_idle_call(void)
*/
if (WARN_ON_ONCE(irqs_disabled()))
local_irq_enable();
-
- rcu_idle_exit();
}
/*
next prev parent reply other threads:[~2020-08-20 7:50 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-08-20 7:30 [PATCH 0/9] TRACE_IRQFLAGS wreckage Peter Zijlstra
2020-08-20 7:30 ` [PATCH 1/9] lockdep: Use raw_cpu_*() for per-cpu variables Peter Zijlstra
2020-08-20 13:54 ` Steven Rostedt
2020-08-20 7:30 ` Peter Zijlstra [this message]
2020-08-20 13:58 ` [PATCH 2/9] sched,idle,rcu: Push rcu_idle deeper into the idle path Steven Rostedt
2020-08-20 7:30 ` [PATCH 3/9] cpuidle: Make CPUIDLE_FLAG_TLB_FLUSHED generic Peter Zijlstra
2020-08-20 7:30 ` [PATCH 4/9] cpuidle: Move trace_cpu_idle() into generic code Peter Zijlstra
2020-08-20 7:30 ` [PATCH 5/9] x86/entry: Remove unused THUNKs Peter Zijlstra
2020-08-20 7:30 ` [PATCH 6/9] locking/lockdep: Cleanup Peter Zijlstra
2020-08-20 7:30 ` [PATCH 7/9] nds32: Implement arch_irqs_disabled() Peter Zijlstra
2020-08-20 7:30 ` [PATCH 8/9] lockdep: Only trace IRQ edges Peter Zijlstra
2020-08-20 7:30 ` [PATCH 9/9] lockdep,trace: Expose tracepoints Peter Zijlstra
2020-08-20 14:36 ` [PATCH 0/9] TRACE_IRQFLAGS wreckage Steven Rostedt
2020-08-20 14:39 ` Steven Rostedt
2020-08-20 14:49 ` Marco Elver
2020-08-20 14:58 ` peterz
2020-08-20 16:53 ` Steven Rostedt
2020-08-20 17:20 ` Marco Elver
2020-08-20 19:59 ` Steven Rostedt
2020-08-21 6:37 ` Marco Elver
2020-08-21 6:54 ` peterz
2020-08-21 7:05 ` Marco Elver
2020-08-27 7:54 ` [tip: sched/urgent] sched: Use __always_inline on is_idle_task() tip-bot2 for Marco Elver
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=20200820074640.401636694@infradead.org \
--to=peterz@infradead.org \
--cc=elver@google.com \
--cc=jgross@suse.com \
--cc=joel@joelfernandes.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=npiggin@gmail.com \
--cc=paulmck@kernel.org \
--cc=rjw@rjwysocki.net \
--cc=rostedt@goodmis.org \
--cc=svens@linux.ibm.com \
--cc=tglx@linutronix.de \
--cc=will@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