Skip to contentSkip to conversation

nohz: Remove tick_nohz_idle_enter_norcu() / tick_nohz_idle_exit_norcu()

PATCH In review Frederic Weisbecker 3 participants 8 messages Nov 17, 2011 – Nov 21, 2011

Subsystems (11)
Arm (32-bit) x86 PowerPC MIPS +7 more subsystems
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 nohz: Remove tick_nohz_idle_enter_norcu() / tick_nohz_idle_exit_norcu()

Those two APIs were provided to optimize the calls of tick_nohz_idle_enter() and rcu_idle_enter() into a single irq disabled section. This way no interrupt happening in-between would needlessly process any RCU job.

Now we are talking about an optimization for which benefits have yet to be measured. Let's start simple and completely decouple idle rcu and dyntick idle logics to simplify.

Cc Ingo Molnar <[email protected]>
Cc Thomas Gleixner <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Josh Triplett <[email protected]>
---
 arch/arm/kernel/process.c              |    6 +++-
 arch/avr32/kernel/process.c            |    6 +++-
 arch/blackfin/kernel/process.c         |    6 +++-
 arch/microblaze/kernel/process.c       |    6 +++-
 arch/mips/kernel/process.c             |    6 +++-
 arch/openrisc/kernel/idle.c            |    6 +++-
 arch/powerpc/kernel/idle.c             |   15 +++++-----
 arch/powerpc/platforms/iseries/setup.c |   12 +++++---
 arch/s390/kernel/process.c             |    6 +++-
 arch/sh/kernel/idle.c                  |    6 +++-
 arch/sparc/kernel/process_64.c         |    6 +++-
 arch/tile/kernel/process.c             |    6 +++-
 arch/um/kernel/process.c               |    6 +++-
 arch/unicore32/kernel/process.c        |    6 +++-
 arch/x86/kernel/process_32.c           |    6 +++-
 include/linux/tick.h                   |   47 +-------------------------------
 kernel/time/tick-sched.c               |   15 +++++-----
 17 files changed, 76 insertions(+), 91 deletions(-)
arch/arm/kernel/process.c +4 -2
index 4f83362..0e42a9c 100644
@@ -183,7 +183,8 @@ void cpu_idle(void)
 
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		leds_event(led_idle_start);
 		while (!need_resched()) {
 #ifdef CONFIG_HOTPLUG_CPU
@@ -210,7 +211,8 @@ void cpu_idle(void)
 			}
 		}
 		leds_event(led_idle_end);
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/avr32/kernel/process.c +4 -2
index 34c8c70..ea33957 100644
@@ -34,10 +34,12 @@ void cpu_idle(void)
 {
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		while (!need_resched())
 			cpu_idle_sleep();
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/blackfin/kernel/process.c +4 -2
index 57e0749..8dd0416 100644
@@ -88,10 +88,12 @@ void cpu_idle(void)
 #endif
 		if (!idle)
 			idle = default_idle;
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		while (!need_resched())
 			idle();
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/microblaze/kernel/process.c +4 -2
index c6ece38..37ed945 100644
@@ -103,10 +103,12 @@ void cpu_idle(void)
 		if (!idle)
 			idle = default_idle;
 
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		while (!need_resched())
 			idle();
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 
 		preempt_enable_no_resched();
 		schedule();
arch/mips/kernel/process.c +4 -2
index 7df2ffc..7937367 100644
@@ -56,7 +56,8 @@ void __noreturn cpu_idle(void)
 
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		while (!need_resched() && cpu_online(cpu)) {
 #ifdef CONFIG_MIPS_MT_SMTC
 			extern void smtc_idle_loop_hook(void);
@@ -77,7 +78,8 @@ void __noreturn cpu_idle(void)
 		     system_state == SYSTEM_BOOTING))
 			play_dead();
 #endif
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/openrisc/kernel/idle.c +4 -2
index 2e82cd0..e5fc7887 100644
@@ -51,7 +51,8 @@ void cpu_idle(void)
 
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 
 		while (!need_resched()) {
 			check_pgt_cache();
@@ -69,7 +70,8 @@ void cpu_idle(void)
 			set_thread_flag(TIF_POLLING_NRFLAG);
 		}
 
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/powerpc/kernel/idle.c +7 -8
index 3cd73d1..9c3cd49 100644
@@ -62,10 +62,10 @@ void cpu_idle(void)
 
 	set_thread_flag(TIF_POLLING_NRFLAG);
 	while (1) {
-		if (idle_uses_rcu)
-			tick_nohz_idle_enter();
-		else
-			tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		if (!idle_uses_rcu)
+			rcu_idle_enter();
+
 		while (!need_resched() && !cpu_should_die()) {
 			ppc64_runlatch_off();
 
@@ -102,10 +102,9 @@ void cpu_idle(void)
 
 		HMT_medium();
 		ppc64_runlatch_on();
-		if (idle_uses_rcu)
-			tick_nohz_idle_exit();
-		else
-			tick_nohz_idle_exit_norcu();
+		if (!idle_uses_rcu)
+			rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		if (cpu_should_die())
 			cpu_die();
arch/powerpc/platforms/iseries/setup.c +8 -4
index 77ff6eb..097f7d5 100644
@@ -562,7 +562,8 @@ static void yield_shared_processor(void)
 static void iseries_shared_idle(void)
 {
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		while (!need_resched() && !hvlpevent_is_pending()) {
 			local_irq_disable();
 			ppc64_runlatch_off();
@@ -576,7 +577,8 @@ static void iseries_shared_idle(void)
 		}
 
 		ppc64_runlatch_on();
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 
 		if (hvlpevent_is_pending())
 			process_iSeries_events();
@@ -592,7 +594,8 @@ static void iseries_dedicated_idle(void)
 	set_thread_flag(TIF_POLLING_NRFLAG);
 
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		if (!need_resched()) {
 			while (!need_resched()) {
 				ppc64_runlatch_off();
@@ -609,7 +612,8 @@ static void iseries_dedicated_idle(void)
 		}
 
 		ppc64_runlatch_on();
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/s390/kernel/process.c +4 -2
index 44028ae..bf2bc31 100644
@@ -90,10 +90,12 @@ static void default_idle(void)
 void cpu_idle(void)
 {
 	for (;;) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		while (!need_resched())
 			default_idle();
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/sh/kernel/idle.c +4 -2
index ad58e75..406508d 100644
@@ -89,7 +89,8 @@ void cpu_idle(void)
 
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 
 		while (!need_resched()) {
 			check_pgt_cache();
@@ -111,7 +112,8 @@ void cpu_idle(void)
 			start_critical_timings();
 		}
 
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/sparc/kernel/process_64.c +4 -2
index 78b1bc0..fde8d72 100644
@@ -95,12 +95,14 @@ void cpu_idle(void)
 	set_thread_flag(TIF_POLLING_NRFLAG);
 
 	while(1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 
 		while (!need_resched() && !cpu_is_offline(cpu))
 			sparc64_yield(cpu);
 
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 
 		preempt_enable_no_resched();
 
arch/tile/kernel/process.c +4 -2
index 53ac895..4c1ac6e 100644
@@ -85,7 +85,8 @@ void cpu_idle(void)
 
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		while (!need_resched()) {
 			if (cpu_is_offline(cpu))
 				BUG();  /* no HOTPLUG_CPU */
@@ -105,7 +106,8 @@ void cpu_idle(void)
 				local_irq_enable();
 			current_thread_info()->status |= TS_POLLING;
 		}
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/um/kernel/process.c +4 -2
index 9e7176b..b652ff1 100644
@@ -245,10 +245,12 @@ void default_idle(void)
 		if (need_resched())
 			schedule();
 
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		nsecs = disable_timer();
 		idle_sleep(nsecs);
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 	}
 }
 
arch/unicore32/kernel/process.c +4 -2
index 095ff5a..52edc2b 100644
@@ -55,7 +55,8 @@ void cpu_idle(void)
 {
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		while (!need_resched()) {
 			local_irq_disable();
 			stop_critical_timings();
@@ -63,7 +64,8 @@ void cpu_idle(void)
 			local_irq_enable();
 			start_critical_timings();
 		}
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/x86/kernel/process_32.c +4 -2
index f311d096..44e3384 100644
@@ -98,7 +98,8 @@ void cpu_idle(void)
 
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		while (!need_resched()) {
 
 			check_pgt_cache();
@@ -114,7 +115,8 @@ void cpu_idle(void)
 				pm_idle();
 			start_critical_timings();
 		}
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
include/linux/tick.h +1 -46

60 lines not on this page · raw message

kernel/time/tick-sched.c +8 -7
index 360d028..0d887e8 100644
@@ -425,21 +425,20 @@ out:
  * When the next event is more than a tick into the future, stop the idle tick
  * Called when we start the idle loop.
  *
- * If no use of RCU is made in the idle loop between
- * tick_nohz_idle_enter() and tick_nohz_idle_exit() calls, then
- * tick_nohz_idle_enter_norcu() should be called instead and the arch
- * doesn't need to call rcu_idle_enter() and rcu_idle_exit() explicitly.
- *
- * Otherwise the arch is responsible of calling:
+ * The arch is responsible of calling:
  *
  * - rcu_idle_enter() after its last use of RCU before the CPU is put
  *  to sleep.
  * - rcu_idle_exit() before the first use of RCU after the CPU is woken up.
  */
-void __tick_nohz_idle_enter(void)
+void tick_nohz_idle_enter(void)
 {
 	struct tick_sched *ts;
 
+	WARN_ON_ONCE(irqs_disabled());
+
+	local_irq_disable();
+
 	ts = &__get_cpu_var(tick_cpu_sched);
 	/*
 	 * set ts->inidle unconditionally. even if the system did not
@@ -448,6 +447,8 @@ void __tick_nohz_idle_enter(void)
 	 */
 	ts->inidle = 1;
 	tick_nohz_stop_sched_tick(ts);
+
+	local_irq_enable();
 }
 
 /**
Signature · 1 line
1.7.5.4
Reviewed-by on v1 1/1
#1 lore raw
Quoted 9 lines · Frederic Weisbecker

Those two APIs were provided to optimize the calls of tick_nohz_idle_enter() and rcu_idle_enter() into a single irq disabled section. This way no interrupt happening in-between would needlessly process any RCU job.

Now we are talking about an optimization for which benefits have yet to be measured. Let's start simple and completely decouple idle rcu and dyntick idle logics to simplify.

Quoted 5 lines
Cc Ingo Molnar <[email protected]>
Cc Thomas Gleixner <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Josh Triplett <[email protected]>
Quoted 471 lines
arch/arm/kernel/process.c +4 -2
index 4f83362..0e42a9c 100644
@@ -183,7 +183,8 @@ void cpu_idle(void)
 
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		leds_event(led_idle_start);
 		while (!need_resched()) {
 #ifdef CONFIG_HOTPLUG_CPU
@@ -210,7 +211,8 @@ void cpu_idle(void)
 			}
 		}
 		leds_event(led_idle_end);
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/avr32/kernel/process.c +4 -2
index 34c8c70..ea33957 100644
@@ -34,10 +34,12 @@ void cpu_idle(void)
 {
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		while (!need_resched())
 			cpu_idle_sleep();
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/blackfin/kernel/process.c +4 -2
index 57e0749..8dd0416 100644
@@ -88,10 +88,12 @@ void cpu_idle(void)
 #endif
 		if (!idle)
 			idle = default_idle;
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		while (!need_resched())
 			idle();
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/microblaze/kernel/process.c +4 -2
index c6ece38..37ed945 100644
@@ -103,10 +103,12 @@ void cpu_idle(void)
 		if (!idle)
 			idle = default_idle;
 
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		while (!need_resched())
 			idle();
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 
 		preempt_enable_no_resched();
 		schedule();
arch/mips/kernel/process.c +4 -2
index 7df2ffc..7937367 100644
@@ -56,7 +56,8 @@ void __noreturn cpu_idle(void)
 
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		while (!need_resched() && cpu_online(cpu)) {
 #ifdef CONFIG_MIPS_MT_SMTC
 			extern void smtc_idle_loop_hook(void);
@@ -77,7 +78,8 @@ void __noreturn cpu_idle(void)
 		     system_state == SYSTEM_BOOTING))
 			play_dead();
 #endif
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/openrisc/kernel/idle.c +4 -2
index 2e82cd0..e5fc7887 100644
@@ -51,7 +51,8 @@ void cpu_idle(void)
 
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 
 		while (!need_resched()) {
 			check_pgt_cache();
@@ -69,7 +70,8 @@ void cpu_idle(void)
 			set_thread_flag(TIF_POLLING_NRFLAG);
 		}
 
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/powerpc/kernel/idle.c +7 -8
index 3cd73d1..9c3cd49 100644
@@ -62,10 +62,10 @@ void cpu_idle(void)
 
 	set_thread_flag(TIF_POLLING_NRFLAG);
 	while (1) {
-		if (idle_uses_rcu)
-			tick_nohz_idle_enter();
-		else
-			tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		if (!idle_uses_rcu)
+			rcu_idle_enter();
+
 		while (!need_resched() && !cpu_should_die()) {
 			ppc64_runlatch_off();
 
@@ -102,10 +102,9 @@ void cpu_idle(void)
 
 		HMT_medium();
 		ppc64_runlatch_on();
-		if (idle_uses_rcu)
-			tick_nohz_idle_exit();
-		else
-			tick_nohz_idle_exit_norcu();
+		if (!idle_uses_rcu)
+			rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		if (cpu_should_die())
 			cpu_die();
arch/powerpc/platforms/iseries/setup.c +8 -4
index 77ff6eb..097f7d5 100644
@@ -562,7 +562,8 @@ static void yield_shared_processor(void)
 static void iseries_shared_idle(void)
 {
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		while (!need_resched() && !hvlpevent_is_pending()) {
 			local_irq_disable();
 			ppc64_runlatch_off();
@@ -576,7 +577,8 @@ static void iseries_shared_idle(void)
 		}
 
 		ppc64_runlatch_on();
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 
 		if (hvlpevent_is_pending())
 			process_iSeries_events();
@@ -592,7 +594,8 @@ static void iseries_dedicated_idle(void)
 	set_thread_flag(TIF_POLLING_NRFLAG);
 
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		if (!need_resched()) {
 			while (!need_resched()) {
 				ppc64_runlatch_off();
@@ -609,7 +612,8 @@ static void iseries_dedicated_idle(void)
 		}
 
 		ppc64_runlatch_on();
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/s390/kernel/process.c +4 -2
index 44028ae..bf2bc31 100644
@@ -90,10 +90,12 @@ static void default_idle(void)
 void cpu_idle(void)
 {
 	for (;;) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		while (!need_resched())
 			default_idle();
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/sh/kernel/idle.c +4 -2
index ad58e75..406508d 100644
@@ -89,7 +89,8 @@ void cpu_idle(void)
 
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 
 		while (!need_resched()) {
 			check_pgt_cache();
@@ -111,7 +112,8 @@ void cpu_idle(void)
 			start_critical_timings();
 		}
 
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/sparc/kernel/process_64.c +4 -2
index 78b1bc0..fde8d72 100644
@@ -95,12 +95,14 @@ void cpu_idle(void)
 	set_thread_flag(TIF_POLLING_NRFLAG);
 
 	while(1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 
 		while (!need_resched() && !cpu_is_offline(cpu))
 			sparc64_yield(cpu);
 
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 
 		preempt_enable_no_resched();
 
arch/tile/kernel/process.c +4 -2
index 53ac895..4c1ac6e 100644
@@ -85,7 +85,8 @@ void cpu_idle(void)
 
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		while (!need_resched()) {
 			if (cpu_is_offline(cpu))
 				BUG();  /* no HOTPLUG_CPU */
@@ -105,7 +106,8 @@ void cpu_idle(void)
 				local_irq_enable();
 			current_thread_info()->status |= TS_POLLING;
 		}
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/um/kernel/process.c +4 -2
index 9e7176b..b652ff1 100644
@@ -245,10 +245,12 @@ void default_idle(void)
 		if (need_resched())
 			schedule();
 
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		nsecs = disable_timer();
 		idle_sleep(nsecs);
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 	}
 }
 
arch/unicore32/kernel/process.c +4 -2
index 095ff5a..52edc2b 100644
@@ -55,7 +55,8 @@ void cpu_idle(void)
 {
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		while (!need_resched()) {
 			local_irq_disable();
 			stop_critical_timings();
@@ -63,7 +64,8 @@ void cpu_idle(void)
 			local_irq_enable();
 			start_critical_timings();
 		}
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/x86/kernel/process_32.c +4 -2
index f311d096..44e3384 100644
@@ -98,7 +98,8 @@ void cpu_idle(void)
 
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		while (!need_resched()) {
 
 			check_pgt_cache();
@@ -114,7 +115,8 @@ void cpu_idle(void)
 				pm_idle();
 			start_critical_timings();
 		}
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
include/linux/tick.h +1 -46
index 327434a..ab8be90 100644
@@ -122,45 +122,8 @@ static inline int tick_oneshot_mode_active(void) { return 0; }
 #endif /* !CONFIG_GENERIC_CLOCKEVENTS */
 
 # ifdef CONFIG_NO_HZ
-extern void __tick_nohz_idle_enter(void);
-static inline void tick_nohz_idle_enter(void)
-{
-	local_irq_disable();
-	__tick_nohz_idle_enter();
-	local_irq_enable();
-}
+extern void tick_nohz_idle_enter(void);
 extern void tick_nohz_idle_exit(void);
-
-/*
- * Call this pair of function if the arch doesn't make any use
- * of RCU in-between. You won't need to call rcu_idle_enter() and
- * rcu_idle_exit().
- * Otherwise you need to call tick_nohz_idle_enter() and tick_nohz_idle_exit()
- * and explicitly tell RCU about the window around the place the CPU enters low
- * power mode where no RCU use is made. This is done by calling rcu_idle_enter()
- * after the last use of RCU before the CPU is put to sleep and by calling
- * rcu_idle_exit() before the first use of RCU after the CPU woke up.
- */
-static inline void tick_nohz_idle_enter_norcu(void)
-{
-	/*
-	 * Also call rcu_idle_enter() in the irq disabled section even
-	 * if it disables irq itself.
-	 * Just an optimization that prevents from an interrupt happening
-	 * between it and __tick_nohz_idle_enter() to lose time to help
-	 * completing a grace period while we could be in extended grace
-	 * period already.
-	 */
-	local_irq_disable();
-	__tick_nohz_idle_enter();
-	rcu_idle_enter();
-	local_irq_enable();
-}
-static inline void tick_nohz_idle_exit_norcu(void)
-{
-	rcu_idle_exit();
-	tick_nohz_idle_exit();
-}
 extern void tick_nohz_irq_exit(void);
 extern ktime_t tick_nohz_get_sleep_length(void);
 extern u64 get_cpu_idle_time_us(int cpu, u64 *last_update_time);
@@ -168,14 +131,6 @@ extern u64 get_cpu_iowait_time_us(int cpu, u64 *last_update_time);
 # else
 static inline void tick_nohz_idle_enter(void) { }
 static inline void tick_nohz_idle_exit(void) { }
-static inline void tick_nohz_idle_enter_norcu(void)
-{
-	rcu_idle_enter();
-}
-static inline void tick_nohz_idle_exit_norcu(void)
-{
-	rcu_idle_exit();
-}
 
 static inline ktime_t tick_nohz_get_sleep_length(void)
 {
kernel/time/tick-sched.c +8 -7
index 360d028..0d887e8 100644
@@ -425,21 +425,20 @@ out:
  * When the next event is more than a tick into the future, stop the idle tick
  * Called when we start the idle loop.
  *
- * If no use of RCU is made in the idle loop between
- * tick_nohz_idle_enter() and tick_nohz_idle_exit() calls, then
- * tick_nohz_idle_enter_norcu() should be called instead and the arch
- * doesn't need to call rcu_idle_enter() and rcu_idle_exit() explicitly.
- *
- * Otherwise the arch is responsible of calling:
+ * The arch is responsible of calling:
  *
  * - rcu_idle_enter() after its last use of RCU before the CPU is put
  *  to sleep.
  * - rcu_idle_exit() before the first use of RCU after the CPU is woken up.
  */
-void __tick_nohz_idle_enter(void)
+void tick_nohz_idle_enter(void)
 {
 	struct tick_sched *ts;
 
+	WARN_ON_ONCE(irqs_disabled());
+
+	local_irq_disable();
+
 	ts = &__get_cpu_var(tick_cpu_sched);
 	/*
 	 * set ts->inidle unconditionally. even if the system did not
@@ -448,6 +447,8 @@ void __tick_nohz_idle_enter(void)
 	 */
 	ts->inidle = 1;
 	tick_nohz_stop_sched_tick(ts);
+
+	local_irq_enable();
 }
 
 /**
Signature · 1 line
1.7.5.4
#2 lore raw
Quoted 10 lines · Josh Triplett
Quoted 9 lines · Frederic Weisbecker

Those two APIs were provided to optimize the calls of tick_nohz_idle_enter() and rcu_idle_enter() into a single irq disabled section. This way no interrupt happening in-between would needlessly process any RCU job.

Now we are talking about an optimization for which benefits have yet to be measured. Let's start simple and completely decouple idle rcu and dyntick idle logics to simplify.

Quoted 7 lines
Quoted 5 lines
Cc Ingo Molnar <[email protected]>
Cc Thomas Gleixner <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Josh Triplett <[email protected]>

Merged, thank you both!

Thanx, Paul

Quoted 472 lines
Quoted 471 lines
arch/arm/kernel/process.c +4 -2
index 4f83362..0e42a9c 100644
@@ -183,7 +183,8 @@ void cpu_idle(void)
 
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		leds_event(led_idle_start);
 		while (!need_resched()) {
 #ifdef CONFIG_HOTPLUG_CPU
@@ -210,7 +211,8 @@ void cpu_idle(void)
 			}
 		}
 		leds_event(led_idle_end);
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/avr32/kernel/process.c +4 -2
index 34c8c70..ea33957 100644
@@ -34,10 +34,12 @@ void cpu_idle(void)
 {
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		while (!need_resched())
 			cpu_idle_sleep();
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/blackfin/kernel/process.c +4 -2
index 57e0749..8dd0416 100644
@@ -88,10 +88,12 @@ void cpu_idle(void)
 #endif
 		if (!idle)
 			idle = default_idle;
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		while (!need_resched())
 			idle();
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/microblaze/kernel/process.c +4 -2
index c6ece38..37ed945 100644
@@ -103,10 +103,12 @@ void cpu_idle(void)
 		if (!idle)
 			idle = default_idle;
 
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		while (!need_resched())
 			idle();
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 
 		preempt_enable_no_resched();
 		schedule();
arch/mips/kernel/process.c +4 -2
index 7df2ffc..7937367 100644
@@ -56,7 +56,8 @@ void __noreturn cpu_idle(void)
 
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		while (!need_resched() && cpu_online(cpu)) {
 #ifdef CONFIG_MIPS_MT_SMTC
 			extern void smtc_idle_loop_hook(void);
@@ -77,7 +78,8 @@ void __noreturn cpu_idle(void)
 		     system_state == SYSTEM_BOOTING))
 			play_dead();
 #endif
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/openrisc/kernel/idle.c +4 -2
index 2e82cd0..e5fc7887 100644
@@ -51,7 +51,8 @@ void cpu_idle(void)
 
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 
 		while (!need_resched()) {
 			check_pgt_cache();
@@ -69,7 +70,8 @@ void cpu_idle(void)
 			set_thread_flag(TIF_POLLING_NRFLAG);
 		}
 
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/powerpc/kernel/idle.c +7 -8
index 3cd73d1..9c3cd49 100644
@@ -62,10 +62,10 @@ void cpu_idle(void)
 
 	set_thread_flag(TIF_POLLING_NRFLAG);
 	while (1) {
-		if (idle_uses_rcu)
-			tick_nohz_idle_enter();
-		else
-			tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		if (!idle_uses_rcu)
+			rcu_idle_enter();
+
 		while (!need_resched() && !cpu_should_die()) {
 			ppc64_runlatch_off();
 
@@ -102,10 +102,9 @@ void cpu_idle(void)
 
 		HMT_medium();
 		ppc64_runlatch_on();
-		if (idle_uses_rcu)
-			tick_nohz_idle_exit();
-		else
-			tick_nohz_idle_exit_norcu();
+		if (!idle_uses_rcu)
+			rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		if (cpu_should_die())
 			cpu_die();
arch/powerpc/platforms/iseries/setup.c +8 -4
index 77ff6eb..097f7d5 100644
@@ -562,7 +562,8 @@ static void yield_shared_processor(void)
 static void iseries_shared_idle(void)
 {
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		while (!need_resched() && !hvlpevent_is_pending()) {
 			local_irq_disable();
 			ppc64_runlatch_off();
@@ -576,7 +577,8 @@ static void iseries_shared_idle(void)
 		}
 
 		ppc64_runlatch_on();
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 
 		if (hvlpevent_is_pending())
 			process_iSeries_events();
@@ -592,7 +594,8 @@ static void iseries_dedicated_idle(void)
 	set_thread_flag(TIF_POLLING_NRFLAG);
 
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		if (!need_resched()) {
 			while (!need_resched()) {
 				ppc64_runlatch_off();
@@ -609,7 +612,8 @@ static void iseries_dedicated_idle(void)
 		}
 
 		ppc64_runlatch_on();
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/s390/kernel/process.c +4 -2
index 44028ae..bf2bc31 100644
@@ -90,10 +90,12 @@ static void default_idle(void)
 void cpu_idle(void)
 {
 	for (;;) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		while (!need_resched())
 			default_idle();
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/sh/kernel/idle.c +4 -2
index ad58e75..406508d 100644
@@ -89,7 +89,8 @@ void cpu_idle(void)
 
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 
 		while (!need_resched()) {
 			check_pgt_cache();
@@ -111,7 +112,8 @@ void cpu_idle(void)
 			start_critical_timings();
 		}
 
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/sparc/kernel/process_64.c +4 -2
index 78b1bc0..fde8d72 100644
@@ -95,12 +95,14 @@ void cpu_idle(void)
 	set_thread_flag(TIF_POLLING_NRFLAG);
 
 	while(1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 
 		while (!need_resched() && !cpu_is_offline(cpu))
 			sparc64_yield(cpu);
 
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 
 		preempt_enable_no_resched();
 
arch/tile/kernel/process.c +4 -2
index 53ac895..4c1ac6e 100644
@@ -85,7 +85,8 @@ void cpu_idle(void)
 
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		while (!need_resched()) {
 			if (cpu_is_offline(cpu))
 				BUG();  /* no HOTPLUG_CPU */
@@ -105,7 +106,8 @@ void cpu_idle(void)
 				local_irq_enable();
 			current_thread_info()->status |= TS_POLLING;
 		}
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/um/kernel/process.c +4 -2
index 9e7176b..b652ff1 100644
@@ -245,10 +245,12 @@ void default_idle(void)
 		if (need_resched())
 			schedule();
 
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		nsecs = disable_timer();
 		idle_sleep(nsecs);
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 	}
 }
 
arch/unicore32/kernel/process.c +4 -2
index 095ff5a..52edc2b 100644
@@ -55,7 +55,8 @@ void cpu_idle(void)
 {
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		while (!need_resched()) {
 			local_irq_disable();
 			stop_critical_timings();
@@ -63,7 +64,8 @@ void cpu_idle(void)
 			local_irq_enable();
 			start_critical_timings();
 		}
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
arch/x86/kernel/process_32.c +4 -2
index f311d096..44e3384 100644
@@ -98,7 +98,8 @@ void cpu_idle(void)
 
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_idle_enter_norcu();
+		tick_nohz_idle_enter();
+		rcu_idle_enter();
 		while (!need_resched()) {
 
 			check_pgt_cache();
@@ -114,7 +115,8 @@ void cpu_idle(void)
 				pm_idle();
 			start_critical_timings();
 		}
-		tick_nohz_idle_exit_norcu();
+		rcu_idle_exit();
+		tick_nohz_idle_exit();
 		preempt_enable_no_resched();
 		schedule();
 		preempt_disable();
include/linux/tick.h +1 -46
index 327434a..ab8be90 100644
@@ -122,45 +122,8 @@ static inline int tick_oneshot_mode_active(void) { return 0; }
 #endif /* !CONFIG_GENERIC_CLOCKEVENTS */
 
 # ifdef CONFIG_NO_HZ
-extern void __tick_nohz_idle_enter(void);
-static inline void tick_nohz_idle_enter(void)
-{
-	local_irq_disable();
-	__tick_nohz_idle_enter();
-	local_irq_enable();
-}
+extern void tick_nohz_idle_enter(void);
 extern void tick_nohz_idle_exit(void);
-
-/*
- * Call this pair of function if the arch doesn't make any use
- * of RCU in-between. You won't need to call rcu_idle_enter() and
- * rcu_idle_exit().
- * Otherwise you need to call tick_nohz_idle_enter() and tick_nohz_idle_exit()
- * and explicitly tell RCU about the window around the place the CPU enters low
- * power mode where no RCU use is made. This is done by calling rcu_idle_enter()
- * after the last use of RCU before the CPU is put to sleep and by calling
- * rcu_idle_exit() before the first use of RCU after the CPU woke up.
- */
-static inline void tick_nohz_idle_enter_norcu(void)
-{
-	/*
-	 * Also call rcu_idle_enter() in the irq disabled section even
-	 * if it disables irq itself.
-	 * Just an optimization that prevents from an interrupt happening
-	 * between it and __tick_nohz_idle_enter() to lose time to help
-	 * completing a grace period while we could be in extended grace
-	 * period already.
-	 */
-	local_irq_disable();
-	__tick_nohz_idle_enter();
-	rcu_idle_enter();
-	local_irq_enable();
-}
-static inline void tick_nohz_idle_exit_norcu(void)
-{
-	rcu_idle_exit();
-	tick_nohz_idle_exit();
-}
 extern void tick_nohz_irq_exit(void);
 extern ktime_t tick_nohz_get_sleep_length(void);
 extern u64 get_cpu_idle_time_us(int cpu, u64 *last_update_time);
@@ -168,14 +131,6 @@ extern u64 get_cpu_iowait_time_us(int cpu, u64 *last_update_time);
 # else
 static inline void tick_nohz_idle_enter(void) { }
 static inline void tick_nohz_idle_exit(void) { }
-static inline void tick_nohz_idle_enter_norcu(void)
-{
-	rcu_idle_enter();
-}
-static inline void tick_nohz_idle_exit_norcu(void)
-{
-	rcu_idle_exit();
-}
 
 static inline ktime_t tick_nohz_get_sleep_length(void)
 {
kernel/time/tick-sched.c +8 -7
index 360d028..0d887e8 100644
@@ -425,21 +425,20 @@ out:
  * When the next event is more than a tick into the future, stop the idle tick
  * Called when we start the idle loop.
  *
- * If no use of RCU is made in the idle loop between
- * tick_nohz_idle_enter() and tick_nohz_idle_exit() calls, then
- * tick_nohz_idle_enter_norcu() should be called instead and the arch
- * doesn't need to call rcu_idle_enter() and rcu_idle_exit() explicitly.
- *
- * Otherwise the arch is responsible of calling:
+ * The arch is responsible of calling:
  *
  * - rcu_idle_enter() after its last use of RCU before the CPU is put
  *  to sleep.
  * - rcu_idle_exit() before the first use of RCU after the CPU is woken up.
  */
-void __tick_nohz_idle_enter(void)
+void tick_nohz_idle_enter(void)
 {
 	struct tick_sched *ts;
 
+	WARN_ON_ONCE(irqs_disabled());
+
+	local_irq_disable();
+
 	ts = &__get_cpu_var(tick_cpu_sched);
 	/*
 	 * set ts->inidle unconditionally. even if the system did not
@@ -448,6 +447,8 @@ void __tick_nohz_idle_enter(void)
 	 */
 	ts->inidle = 1;
 	tick_nohz_stop_sched_tick(ts);
+
+	local_irq_enable();
 }
 
 /**
Signature · 1 line
1.7.5.4
#3 lore raw
Quoted 17 lines · Paul E. McKenney
Quoted 16 lines · Josh Triplett
Quoted 14 lines · Frederic Weisbecker

Those two APIs were provided to optimize the calls of tick_nohz_idle_enter() and rcu_idle_enter() into a single irq disabled section. This way no interrupt happening in-between would needlessly process any RCU job.

Now we are talking about an optimization for which benefits have yet to be measured. Let's start simple and completely decouple idle rcu and dyntick idle logics to simplify.

Cc Ingo Molnar <[email protected]>
Cc Thomas Gleixner <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Josh Triplett <[email protected]>
Quoted 3 lines
Quoted 1 line

Merged, thank you both!

And here is a patch on top of yours to allow nesting of rcu_idle_enter() and rcu_idle_exit(). Thoughts?

Thanx, Paul

------------------------------------------------------------------------

rcu: Allow nesting of rcu_idle_enter() and rcu_idle_exit()

Running user tasks in dyntick-idle mode requires RCU to undergo an idle-to-non-idle transition on each entry into the kernel, and vice versa on each exit from the kernel. However, situations where user tasks cannot run in dyntick-idle mode (for example, when there is more than one runnable task on the CPU in question) also require RCU to undergo an idle-to-non-idle transition when coming out of the idle loop (and vice versa when entering the idle loop). In this case, RCU would see one idle-to-non-idle transition when the task became runnable, and another when the task executed a system call.

Therefore, rcu_idle_enter() and rcu_idle_exit() must handle nested calls, which this commit provides for.

kernel/rcu.h +17 -1
index aa88baa..8a76a5b 100644
@@ -33,8 +33,24 @@
  * Process-level increment to ->dynticks_nesting field.  This allows for
  * architectures that use half-interrupts and half-exceptions from
  * process context.
+ *
+ * DYNTICK_TASK_NESTING_MASK is a three-bit field that counts the number
+ * of process-based reasons why RCU cannot consider the corresponding CPU
+ * to be idle, and DYNTICK_TASK_NESTING_VALUE is the value used to increment
+ * or decrement this three-bit field.  The rest of the bits could in
+ * principle be used to count interrupts, but this would mean that a
+ * negative-one value in the interrupt field could incorrectly zero out
+ * the DYNTICK_TASK_NESTING_MASK field.  We therefore provide a two-bit
+ * guard field defined by DYNTICK_TASK_MASK that is set to DYNTICK_TASK_FLAG
+ * upon initial exit from idle.  The DYNTICK_TASK_EXIT_IDLE value is
+ * thus the combined value used upon initial exit from idle.
  */
-#define DYNTICK_TASK_NESTING (LLONG_MAX / 2 - 1)
+#define DYNTICK_TASK_NESTING_VALUE (LLONG_MAX / 8 + 1)
+#define DYNTICK_TASK_NESTING_MASK  (LLONG_MAX - DYNTICK_TASK_NESTING_VALUE + 1)
+#define DYNTICK_TASK_FLAG	   ((DYNTICK_TASK_NESTING_VALUE / 8) * 2)
+#define DYNTICK_TASK_MASK	   ((DYNTICK_TASK_NESTING_VALUE / 8) * 3)
+#define DYNTICK_TASK_EXIT_IDLE	   (DYNTICK_TASK_NESTING_VALUE + \
+				    DYNTICK_TASK_FLAG)
 
 /*
  * debug_rcu_head_queue()/debug_rcu_head_unqueue() are used internally
kernel/rcutiny.c +12 -4
index e5bd949..10523d6 100644
@@ -53,7 +53,7 @@ static void __call_rcu(struct rcu_head *head,
 
 #include "rcutiny_plugin.h"
 
-static long long rcu_dynticks_nesting = DYNTICK_TASK_NESTING;
+static long long rcu_dynticks_nesting = DYNTICK_TASK_NESTING_VALUE;
 
 /* Common code for rcu_idle_enter() and rcu_irq_exit(), see kernel/rcutree.c. */
 static void rcu_idle_enter_common(long long oldval)
@@ -88,7 +88,12 @@ void rcu_idle_enter(void)
 
 	local_irq_save(flags);
 	oldval = rcu_dynticks_nesting;
-	rcu_dynticks_nesting = 0;
+	WARN_ON_ONCE((rcu_dynticks_nesting & DYNTICK_TASK_NESTING_MASK) == 0);
+	if ((rcu_dynticks_nesting & DYNTICK_TASK_NESTING_MASK) ==
+	    DYNTICK_TASK_NESTING_VALUE)
+		rcu_dynticks_nesting = 0;
+	else
+		rcu_dynticks_nesting  -= DYNTICK_TASK_NESTING_VALUE;
 	rcu_idle_enter_common(oldval);
 	local_irq_restore(flags);
 }
@@ -140,8 +145,11 @@ void rcu_idle_exit(void)
 
 	local_irq_save(flags);
 	oldval = rcu_dynticks_nesting;
-	WARN_ON_ONCE(oldval != 0);
-	rcu_dynticks_nesting = DYNTICK_TASK_NESTING;
+	WARN_ON_ONCE(rcu_dynticks_nesting < 0);
+	if (rcu_dynticks_nesting & DYNTICK_TASK_NESTING_MASK)
+		rcu_dynticks_nesting += DYNTICK_TASK_NESTING_VALUE;
+	else
+		rcu_dynticks_nesting = DYNTICK_TASK_EXIT_IDLE;
 	rcu_idle_exit_common(oldval);
 	local_irq_restore(flags);
 }
kernel/rcutree.c +16 -6
index 7fb8b0e..f1a3379 100644
@@ -196,7 +196,7 @@ void rcu_note_context_switch(int cpu)
 EXPORT_SYMBOL_GPL(rcu_note_context_switch);
 
 DEFINE_PER_CPU(struct rcu_dynticks, rcu_dynticks) = {
-	.dynticks_nesting = DYNTICK_TASK_NESTING,
+	.dynticks_nesting = DYNTICK_TASK_NESTING_VALUE,
 	.dynticks = ATOMIC_INIT(1),
 };
 
@@ -394,7 +394,11 @@ void rcu_idle_enter(void)
 	local_irq_save(flags);
 	rdtp = &__get_cpu_var(rcu_dynticks);
 	oldval = rdtp->dynticks_nesting;
-	rdtp->dynticks_nesting = 0;
+	WARN_ON_ONCE((oldval & DYNTICK_TASK_NESTING_MASK) == 0);
+	if ((oldval & DYNTICK_TASK_NESTING_MASK) == DYNTICK_TASK_NESTING_VALUE)
+		rdtp->dynticks_nesting = 0;
+	else
+		rdtp->dynticks_nesting -= DYNTICK_TASK_NESTING_VALUE;
 	rcu_idle_enter_common(rdtp, oldval);
 	local_irq_restore(flags);
 }
@@ -481,8 +485,11 @@ void rcu_idle_exit(void)
 	local_irq_save(flags);
 	rdtp = &__get_cpu_var(rcu_dynticks);
 	oldval = rdtp->dynticks_nesting;
-	WARN_ON_ONCE(oldval != 0);
-	rdtp->dynticks_nesting = DYNTICK_TASK_NESTING;
+	WARN_ON_ONCE(oldval < 0);
+	if (oldval & DYNTICK_TASK_NESTING_MASK)
+		rdtp->dynticks_nesting += DYNTICK_TASK_NESTING_VALUE;
+	else
+		rdtp->dynticks_nesting = DYNTICK_TASK_EXIT_IDLE;
 	rcu_idle_exit_common(rdtp, oldval);
 	local_irq_restore(flags);
 }
@@ -2028,7 +2035,8 @@ rcu_boot_init_percpu_data(int cpu, struct rcu_state *rsp)
 		rdp->nxttail[i] = &rdp->nxtlist;
 	rdp->qlen = 0;
 	rdp->dynticks = &per_cpu(rcu_dynticks, cpu);
-	WARN_ON_ONCE(rdp->dynticks->dynticks_nesting != DYNTICK_TASK_NESTING);
+	WARN_ON_ONCE(rdp->dynticks->dynticks_nesting !=
+		     DYNTICK_TASK_NESTING_VALUE);
 	WARN_ON_ONCE(atomic_read(&rdp->dynticks->dynticks) != 1);
 	rdp->cpu = cpu;
 	rdp->rsp = rsp;
@@ -2056,8 +2064,10 @@ rcu_init_percpu_data(int cpu, struct rcu_state *rsp, int preemptible)
 	rdp->qlen_last_fqs_check = 0;
 	rdp->n_force_qs_snap = rsp->n_force_qs;
 	rdp->blimit = blimit;
-	WARN_ON_ONCE(rdp->dynticks->dynticks_nesting != DYNTICK_TASK_NESTING);
+	rdp->dynticks->dynticks_nesting = DYNTICK_TASK_NESTING_VALUE;
 	WARN_ON_ONCE((atomic_read(&rdp->dynticks->dynticks) & 0x1) != 1);
+	atomic_set(&rdp->dynticks->dynticks,
+		   (atomic_read(&rdp->dynticks->dynticks) & ~0x1) + 1);
 	raw_spin_unlock(&rnp->lock);		/* irqs remain disabled. */
 
 	/*
#4 lore raw

2011/11/19 Paul E. McKenney <[email protected]>:

Quoted 31 lines · Paul E. McKenney
Quoted 17 lines · Josh Triplett
Quoted 16 lines · Frederic Weisbecker
Quoted 14 lines

Those two APIs were provided to optimize the calls of tick_nohz_idle_enter() and rcu_idle_enter() into a single irq disabled section. This way no interrupt happening in-between would needlessly process any RCU job.

Now we are talking about an optimization for which benefits have yet to be measured. Let's start simple and completely decouple idle rcu and dyntick idle logics to simplify.

Cc Ingo Molnar <[email protected]>
Cc Thomas Gleixner <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Josh Triplett <[email protected]>
Quoted 3 lines
Quoted 1 line

Merged, thank you both!

And here is a patch on top of yours to allow nesting of rcu_idle_enter() and rcu_idle_exit().  Thoughts?

                                                       Thanx, Paul

------------------------------------------------------------------------

rcu: Allow nesting of rcu_idle_enter() and rcu_idle_exit()

Quoted 7 lines

Running user tasks in dyntick-idle mode requires RCU to undergo an idle-to-non-idle transition on each entry into the kernel, and vice versa on each exit from the kernel.  However, situations where user tasks cannot run in dyntick-idle mode (for example, when there is more than one runnable task on the CPU in question) also require RCU to undergo an idle-to-non-idle transition when coming out of the idle loop (and vice versa when entering the idle loop).

Not sure what you mean about the idle loop with the dyntick-idle mode we can't enter when we resume to userspace with more than one task in the runqueue.

Quoted 3 lines

 In this case,
RCU would see one idle-to-non-idle transition when the task became runnable, and another when the task executed a system call.

I'm a bit confused with this changelog.

What can happen with the adaptive tickless thing is:

- When we resume to userspace after a syscall/irq/exception and we are not in RCU extended quiescent state, then switch to it. We may call it RCU idle mode I guess but that may start to be confusing.
So this may involve several kind of nesting. From a single rcu_idle_enter() to more complicated scenario if we switch to RCU extended qs from an an interrupt: rcu_idle_exit() is called on entry of the irq, rcu_idle_enter() is called in the middle then finally a last call to rcu_idle_enter() in the irq exit at which point only we want the RCU extended qs to be effective.

- We may also exit that RCU extended qs state by involving other funny nesting. We have the simple syscall enter that just calls rcu_idle_exit() if we were in userspace in RCU extended qs. We may also receive an IPI that enqueues a new task, in which case we may exit the RCU extended quiescent from the irq with the following nesting:
rcu_idle_exit() on irq entry, then another call to rcu_idle_exit() to prevent from resuming the RCU extended quiescent state when we come back to userspace and finally the rcu_idle_enter() in the irq exit.

Is that what you had in mind?

Thanks.

#5 lore raw
Quoted 60 lines · Frederic Weisbecker

2011/11/19 Paul E. McKenney <[email protected]>:

Quoted 31 lines · Paul E. McKenney
Quoted 17 lines · Josh Triplett
Quoted 16 lines · Frederic Weisbecker
Quoted 14 lines

Those two APIs were provided to optimize the calls of tick_nohz_idle_enter() and rcu_idle_enter() into a single irq disabled section. This way no interrupt happening in-between would needlessly process any RCU job.

Now we are talking about an optimization for which benefits have yet to be measured. Let's start simple and completely decouple idle rcu and dyntick idle logics to simplify.

Cc Ingo Molnar <[email protected]>
Cc Thomas Gleixner <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Josh Triplett <[email protected]>
Quoted 3 lines
Quoted 1 line

Merged, thank you both!

And here is a patch on top of yours to allow nesting of rcu_idle_enter() and rcu_idle_exit().  Thoughts?

                                                       Thanx, Paul

------------------------------------------------------------------------

rcu: Allow nesting of rcu_idle_enter() and rcu_idle_exit()

Quoted 7 lines

Running user tasks in dyntick-idle mode requires RCU to undergo an idle-to-non-idle transition on each entry into the kernel, and vice versa on each exit from the kernel.  However, situations where user tasks cannot run in dyntick-idle mode (for example, when there is more than one runnable task on the CPU in question) also require RCU to undergo an idle-to-non-idle transition when coming out of the idle loop (and vice versa when entering the idle loop).

Not sure what you mean about the idle loop with the dyntick-idle mode we can't enter when we resume to userspace with more than one task in the runqueue.

Quoted 3 lines

 In this case,
RCU would see one idle-to-non-idle transition when the task became runnable, and another when the task executed a system call.

I'm a bit confused with this changelog.

What can happen with the adaptive tickless thing is:

- When we resume to userspace after a syscall/irq/exception and we are not in RCU extended quiescent state, then switch to it. We may call it RCU idle mode I guess but that may start to be confusing.
So this may involve several kind of nesting. From a single rcu_idle_enter() to more complicated scenario if we switch to RCU extended qs from an an interrupt: rcu_idle_exit() is called on entry of the irq, rcu_idle_enter() is called in the middle then finally a last call to rcu_idle_enter() in the irq exit at which point only we want the RCU extended qs to be effective.

Quoted 3 lines

- We may also exit that RCU extended qs state by involving other funny nesting. We have the simple syscall enter that just calls rcu_idle_exit() if we were in userspace in RCU extended qs.

OK, so perhaps this is what I am missing. Do you avoid calling rcu_idle_exit() in the case where the user-mode execution was not an RCU extended quiescent state? If so, then my patch is not needed, and I can revert it.

Quoted 8 lines

We may also receive an IPI
that enqueues a new task, in which case we may exit the RCU extended quiescent from the irq with the following nesting:
rcu_idle_exit() on irq entry, then another call to rcu_idle_exit() to prevent from resuming the RCU extended quiescent state when we come back to userspace and finally the rcu_idle_enter() in the irq exit.

Is that what you had in mind?

I was concerned about the following scenario:

1. A CPU is initially idle.

2. Task A wakes up on that CPU, enters user-mode execution
in an RCU extended quiescent state.

3. Task B wakes up on that CPU, forcing the CPU out of its
RCU extended quiescent state. However, Task A is higher priority than is Task B, so Task A continues running.

4. Task A invokes a system call. If the system-call entry
code were to again invoke rcu_idle_enter(), then my patch is required. If you check and avoid invoking rcu_idle_enter() in this case, then my patch is not required.

Thanx, Paul

#6 lore raw
Quoted 61 lines · Paul E. McKenney
Quoted 60 lines · Frederic Weisbecker

2011/11/19 Paul E. McKenney <[email protected]>:

Quoted 31 lines · Paul E. McKenney
Quoted 17 lines · Josh Triplett
Quoted 16 lines · Frederic Weisbecker
Quoted 14 lines

Those two APIs were provided to optimize the calls of tick_nohz_idle_enter() and rcu_idle_enter() into a single irq disabled section. This way no interrupt happening in-between would needlessly process any RCU job.

Now we are talking about an optimization for which benefits have yet to be measured. Let's start simple and completely decouple idle rcu and dyntick idle logics to simplify.

Cc Ingo Molnar <[email protected]>
Cc Thomas Gleixner <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Josh Triplett <[email protected]>
Quoted 3 lines
Quoted 1 line

Merged, thank you both!

And here is a patch on top of yours to allow nesting of rcu_idle_enter() and rcu_idle_exit().  Thoughts?

                                                       Thanx, Paul

------------------------------------------------------------------------

rcu: Allow nesting of rcu_idle_enter() and rcu_idle_exit()

Quoted 7 lines

Running user tasks in dyntick-idle mode requires RCU to undergo an idle-to-non-idle transition on each entry into the kernel, and vice versa on each exit from the kernel.  However, situations where user tasks cannot run in dyntick-idle mode (for example, when there is more than one runnable task on the CPU in question) also require RCU to undergo an idle-to-non-idle transition when coming out of the idle loop (and vice versa when entering the idle loop).

Not sure what you mean about the idle loop with the dyntick-idle mode we can't enter when we resume to userspace with more than one task in the runqueue.

Quoted 3 lines

 In this case,
RCU would see one idle-to-non-idle transition when the task became runnable, and another when the task executed a system call.

I'm a bit confused with this changelog.

What can happen with the adaptive tickless thing is:

- When we resume to userspace after a syscall/irq/exception and we are not in RCU extended quiescent state, then switch to it. We may call it RCU idle mode I guess but that may start to be confusing.
So this may involve several kind of nesting. From a single rcu_idle_enter() to more complicated scenario if we switch to RCU extended qs from an an interrupt: rcu_idle_exit() is called on entry of the irq, rcu_idle_enter() is called in the middle then finally a last call to rcu_idle_enter() in the irq exit at which point only we want the RCU extended qs to be effective.

Quoted 8 lines
Quoted 3 lines

- We may also exit that RCU extended qs state by involving other funny nesting. We have the simple syscall enter that just calls rcu_idle_exit() if we were in userspace in RCU extended qs.

OK, so perhaps this is what I am missing. Do you avoid calling rcu_idle_exit() in the case where the user-mode execution was not an RCU extended quiescent state? If so, then my patch is not needed, and I can revert it.

Yes, if we resume to userspace after a syscall but we have more than one task in the runqueue, then we don't switch to RCU extended qs: we don't call rcu_idle_exit() on syscall return in this case.

Quoted 8 lines
Quoted 8 lines

We may also receive an IPI
that enqueues a new task, in which case we may exit the RCU extended quiescent from the irq with the following nesting:
rcu_idle_exit() on irq entry, then another call to rcu_idle_exit() to prevent from resuming the RCU extended quiescent state when we come back to userspace and finally the rcu_idle_enter() in the irq exit.

Quoted 8 lines
Quoted 1 line

Is that what you had in mind?

I was concerned about the following scenario:

1. A CPU is initially idle.

2. Task A wakes up on that CPU, enters user-mode execution
in an RCU extended quiescent state.

Just in case, I would like to note what happens in detail here:

- Idle notices the need to resched, goes out of its idle loop and calls rcu_idle_exit(). The scheduler context switching may need RCU and we don't know where the next task will resume. If it's in the kernel it may need RCU as well. So we need this unconditional rcu_idle_exit() that re-enables RCU.

- We also re-enable the tick unconditionally on idle exit time. So when the user task resumes, the tick is there and may decide to shut down again, in which case we may call rcu_idle_enter() if we are in userspace. Otherwise this is done later when we resume userspace (syscall or exception).

Quoted 4 lines

3. Task B wakes up on that CPU, forcing the CPU out of its
RCU extended quiescent state. However, Task A is higher priority than is Task B, so Task A continues running.

Right but we have two tasks in the runqueue then, so we restart the tick and call rcu_idle_exit().

Quoted 5 lines

4. Task A invokes a system call. If the system-call entry
code were to again invoke rcu_idle_enter(), then my patch is required. If you check and avoid invoking rcu_idle_enter() in this case, then my patch is not required.

You mean rcu_idle_exit()? So yeah, since we have the tick running and thus RCU not in extended QS, we won't call rcu_idle_exit() on syscall entry.

#7 lore raw
Quoted 5 lines · Frederic Weisbecker
Quoted 4 lines · Paul E. McKenney
Quoted 3 lines · Frederic Weisbecker

2011/11/19 Paul E. McKenney <[email protected]>:

Quoted 2 lines · Paul E. McKenney
Quoted 1 line · Josh Triplett

[ . . . ]

Quoted 8 lines
Quoted 4 lines

4. Task A invokes a system call. If the system-call entry
code were to again invoke rcu_idle_enter(), then my patch is required. If you check and avoid invoking rcu_idle_enter() in this case, then my patch is not required.

You mean rcu_idle_exit()? So yeah, since we have the tick running and thus RCU not in extended QS, we won't call rcu_idle_exit() on syscall entry.

OK, then I will drop my patch. ;-)

Thanx, Paul