[PATCH tip/core/rcu 0/12] RCU flavor consolidation cleanups for v5.1

[PATCH tip/core/rcu 0/12] RCU flavor consolidation cleanups for v5.1

[Paul E. McKenney]

Hello!

This series contains additional cleanups for RCU flavor consolidation:

1-2.	Replace scheduler _sched() APIs with _rcu() counterparts.

3.	Renaming and commenting changes due to only one rcuo kthread
	per CPU.

4.	Make expedited IPI handler return after handling critical section.
	This is a maintainability and performance change.

5.	Inline force_quiescent_state() into rcu_force_quiescent_state().

6.	Eliminate RCU_BH_FLAVOR and RCU_SCHED_FLAVOR.

7.	Inline rcu_kthread_do_work() into its sole remaining caller.

8.	Determine expedited-GP IPI handler at build time.

9.	Consolidate PREEMPT and !PREEMPT synchronize_rcu_expedited().

10.	Consolidate PREEMPT and !PREEMPT synchronize_rcu().

11.	Inline _synchronize_rcu_expedited() into synchronize_rcu_expedited().

12.	Discard separate per-CPU callback counts.

							Thanx, Paul

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

 Documentation/RCU/Design/Expedited-Grace-Periods/ExpSchedFlow.svg             |   18 
 Documentation/RCU/Design/Expedited-Grace-Periods/Expedited-Grace-Periods.html |   26 -
 Documentation/RCU/stallwarn.txt                                               |   15 
 kernel/rcu/rcu.h                                                              |    2 
 kernel/rcu/tree.c                                                             |  123 +++--
 kernel/rcu/tree.h                                                             |    9 
 kernel/rcu/tree_exp.h                                                         |  244 ++++------
 kernel/rcu/tree_plugin.h                                                      |  129 -----
 kernel/sched/cpufreq.c                                                        |    4 
 kernel/sched/cpufreq_schedutil.c                                              |    2 
 kernel/sched/sched.h                                                          |    2 
 kernel/sched/topology.c                                                       |    4 
 12 files changed, 237 insertions(+), 341 deletions(-)

[PATCH tip/core/rcu 01/12] sched: Replace call_rcu_sched() with call_rcu()

[Paul E. McKenney]

Now that call_rcu()'s callback is not invoked until after all
preempt-disable regions of code have completed (in addition to explicitly
marked RCU read-side critical sections), call_rcu() can be used in place
of call_rcu_sched().  This commit therefore makes that change.

While in the area, this commit also updates an outdated header comment
for for_each_domain().

Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
---
 kernel/sched/sched.h    | 2 +-
 kernel/sched/topology.c | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index d04530bf251f..6665b9c02e2f 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -1260,7 +1260,7 @@ extern void sched_ttwu_pending(void);
 
 /*
  * The domain tree (rq->sd) is protected by RCU's quiescent state transition.
- * See detach_destroy_domains: synchronize_sched for details.
+ * See destroy_sched_domains: call_rcu for details.
  *
  * The domain tree of any CPU may only be accessed from within
  * preempt-disabled sections.
diff --git a/kernel/sched/topology.c b/kernel/sched/topology.c
index 3f35ba1d8fde..7d905f55e7fa 100644
--- a/kernel/sched/topology.c
+++ b/kernel/sched/topology.c
@@ -442,7 +442,7 @@ void rq_attach_root(struct rq *rq, struct root_domain *rd)
 	raw_spin_unlock_irqrestore(&rq->lock, flags);
 
 	if (old_rd)
-		call_rcu_sched(&old_rd->rcu, free_rootdomain);
+		call_rcu(&old_rd->rcu, free_rootdomain);
 }
 
 void sched_get_rd(struct root_domain *rd)
@@ -455,7 +455,7 @@ void sched_put_rd(struct root_domain *rd)
 	if (!atomic_dec_and_test(&rd->refcount))
 		return;
 
-	call_rcu_sched(&rd->rcu, free_rootdomain);
+	call_rcu(&rd->rcu, free_rootdomain);
 }
 
 static int init_rootdomain(struct root_domain *rd)
-- 
2.17.1

[PATCH tip/core/rcu 02/12] sched: Replace synchronize_sched() with synchronize_rcu()

[Paul E. McKenney]

Now that synchronize_rcu() waits for preempt-disable regions of
code as well as RCU read-side critical sections, synchronize_sched()
can be replaced by synchronize_rcu(), in fact, synchronize_sched()
is now completely equivalent to synchronize_rcu().  This commit
therefore replaces synchronize_sched() with synchronize_rcu() so that
synchronize_sched() can eventually be removed entirely.

Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
---
 kernel/sched/cpufreq.c           | 4 ++--
 kernel/sched/cpufreq_schedutil.c | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/kernel/sched/cpufreq.c b/kernel/sched/cpufreq.c
index 22bd8980f32f..835671f0f917 100644
--- a/kernel/sched/cpufreq.c
+++ b/kernel/sched/cpufreq.c
@@ -48,8 +48,8 @@ EXPORT_SYMBOL_GPL(cpufreq_add_update_util_hook);
  *
  * Clear the update_util_data pointer for the given CPU.
  *
- * Callers must use RCU-sched callbacks to free any memory that might be
- * accessed via the old update_util_data pointer or invoke synchronize_sched()
+ * Callers must use RCU callbacks to free any memory that might be
+ * accessed via the old update_util_data pointer or invoke synchronize_rcu()
  * right after this function to avoid use-after-free.
  */
 void cpufreq_remove_update_util_hook(int cpu)
diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c
index 033ec7c45f13..2efe629425be 100644
--- a/kernel/sched/cpufreq_schedutil.c
+++ b/kernel/sched/cpufreq_schedutil.c
@@ -859,7 +859,7 @@ static void sugov_stop(struct cpufreq_policy *policy)
 	for_each_cpu(cpu, policy->cpus)
 		cpufreq_remove_update_util_hook(cpu);
 
-	synchronize_sched();
+	synchronize_rcu();
 
 	if (!policy->fast_switch_enabled) {
 		irq_work_sync(&sg_policy->irq_work);
-- 
2.17.1

[PATCH tip/core/rcu 03/12] rcu: Rename and comment changes due to only one rcuo kthread per CPU

[Paul E. McKenney]

Given RCU flavor consolidation, the name rcu_spawn_all_nocb_kthreads()
is quite misleading.  It no longer ever creates more than one kthread,
and it does so only for the specified CPU.  This commit therefore changes
this name to the more descriptive rcu_spawn_cpu_nocb_kthread(), and also
fixes up a similar issue in its header comment while in the area.

Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
---
 kernel/rcu/tree.c        | 2 +-
 kernel/rcu/tree.h        | 2 +-
 kernel/rcu/tree_plugin.h | 8 ++++----
 3 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
index 9180158756d2..f4edc664fb65 100644
--- a/kernel/rcu/tree.c
+++ b/kernel/rcu/tree.c
@@ -3299,7 +3299,7 @@ int rcutree_prepare_cpu(unsigned int cpu)
 	trace_rcu_grace_period(rcu_state.name, rdp->gp_seq, TPS("cpuonl"));
 	raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 	rcu_prepare_kthreads(cpu);
-	rcu_spawn_all_nocb_kthreads(cpu);
+	rcu_spawn_cpu_nocb_kthread(cpu);
 
 	return 0;
 }
diff --git a/kernel/rcu/tree.h b/kernel/rcu/tree.h
index d90b02b53c0e..bcfd684a5c57 100644
--- a/kernel/rcu/tree.h
+++ b/kernel/rcu/tree.h
@@ -451,7 +451,7 @@ static bool rcu_nocb_adopt_orphan_cbs(struct rcu_data *my_rdp,
 static int rcu_nocb_need_deferred_wakeup(struct rcu_data *rdp);
 static void do_nocb_deferred_wakeup(struct rcu_data *rdp);
 static void rcu_boot_init_nocb_percpu_data(struct rcu_data *rdp);
-static void rcu_spawn_all_nocb_kthreads(int cpu);
+static void rcu_spawn_cpu_nocb_kthread(int cpu);
 static void __init rcu_spawn_nocb_kthreads(void);
 #ifdef CONFIG_RCU_NOCB_CPU
 static void __init rcu_organize_nocb_kthreads(void);
diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h
index 1b3dd2fc0cd6..4d4091565a2c 100644
--- a/kernel/rcu/tree_plugin.h
+++ b/kernel/rcu/tree_plugin.h
@@ -2517,9 +2517,9 @@ static void rcu_spawn_one_nocb_kthread(int cpu)
 
 /*
  * If the specified CPU is a no-CBs CPU that does not already have its
- * rcuo kthreads, spawn them.
+ * rcuo kthread, spawn it.
  */
-static void rcu_spawn_all_nocb_kthreads(int cpu)
+static void rcu_spawn_cpu_nocb_kthread(int cpu)
 {
 	if (rcu_scheduler_fully_active)
 		rcu_spawn_one_nocb_kthread(cpu);
@@ -2536,7 +2536,7 @@ static void __init rcu_spawn_nocb_kthreads(void)
 	int cpu;
 
 	for_each_online_cpu(cpu)
-		rcu_spawn_all_nocb_kthreads(cpu);
+		rcu_spawn_cpu_nocb_kthread(cpu);
 }
 
 /* How many follower CPU IDs per leader?  Default of -1 for sqrt(nr_cpu_ids). */
@@ -2670,7 +2670,7 @@ static void do_nocb_deferred_wakeup(struct rcu_data *rdp)
 {
 }
 
-static void rcu_spawn_all_nocb_kthreads(int cpu)
+static void rcu_spawn_cpu_nocb_kthread(int cpu)
 {
 }
 
-- 
2.17.1

[PATCH tip/core/rcu 04/12] rcu: Make expedited IPI handler return after handling critical section

[Paul E. McKenney]

During expedited RCU grace-period initialization, IPIs are sent to
all non-idle online CPUs.  The IPI handler checks to see if the CPU is
in quiescent state, reporting one if so.  This handler looks at three
different cases: (1) The CPU is not in an rcu_read_lock()-based critical
section, (2) The CPU is in the process of exiting an rcu_read_lock()-based
critical section, and (3) The CPU is in an rcu_read_lock()-based critical
section.  In case (2), execution falls through into case (3).

This is harmless from a functionality viewpoint, but can result in
needless overhead during an improbable corner case.  This commit therefore
adds the "return" statement needed to prevent fall-through.

Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
---
 kernel/rcu/tree_exp.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/kernel/rcu/tree_exp.h b/kernel/rcu/tree_exp.h
index 928fe5893a57..6d4eb4694b6f 100644
--- a/kernel/rcu/tree_exp.h
+++ b/kernel/rcu/tree_exp.h
@@ -697,6 +697,7 @@ static void sync_rcu_exp_handler(void *unused)
 			WRITE_ONCE(t->rcu_read_unlock_special.b.exp_hint, true);
 		}
 		raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
+		return;
 	}
 
 	/*
-- 
2.17.1

[PATCH tip/core/rcu 05/12] rcu: Inline force_quiescent_state() into rcu_force_quiescent_state()

[Paul E. McKenney]

Given that rcu_force_quiescent_state() is a simple wrapper around
force_quiescent_state(), this commit saves a few lines of code by
inlining force_quiescent_state() into rcu_force_quiescent_state(),
and changing all references to force_quiescent_state() to instead
invoke rcu_force_quiescent_state().

Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
---
 kernel/rcu/tree.c | 21 ++++++---------------
 1 file changed, 6 insertions(+), 15 deletions(-)

diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
index f4edc664fb65..e56a46444775 100644
--- a/kernel/rcu/tree.c
+++ b/kernel/rcu/tree.c
@@ -479,7 +479,6 @@ module_param_cb(jiffies_till_next_fqs, &next_fqs_jiffies_ops, &jiffies_till_next
 module_param(rcu_kick_kthreads, bool, 0644);
 
 static void force_qs_rnp(int (*f)(struct rcu_data *rdp));
-static void force_quiescent_state(void);
 static int rcu_pending(void);
 
 /*
@@ -503,15 +502,6 @@ unsigned long rcu_exp_batches_completed(void)
 }
 EXPORT_SYMBOL_GPL(rcu_exp_batches_completed);
 
-/*
- * Force a quiescent state.
- */
-void rcu_force_quiescent_state(void)
-{
-	force_quiescent_state();
-}
-EXPORT_SYMBOL_GPL(rcu_force_quiescent_state);
-
 /*
  * Convert a ->gp_state value to a character string.
  */
@@ -1310,7 +1300,7 @@ static void print_other_cpu_stall(unsigned long gp_seq)
 
 	panic_on_rcu_stall();
 
-	force_quiescent_state();  /* Kick them all. */
+	rcu_force_quiescent_state();  /* Kick them all. */
 }
 
 static void print_cpu_stall(void)
@@ -2578,7 +2568,7 @@ static void force_qs_rnp(int (*f)(struct rcu_data *rdp))
  * Force quiescent states on reluctant CPUs, and also detect which
  * CPUs are in dyntick-idle mode.
  */
-static void force_quiescent_state(void)
+void rcu_force_quiescent_state(void)
 {
 	unsigned long flags;
 	bool ret;
@@ -2610,6 +2600,7 @@ static void force_quiescent_state(void)
 	raw_spin_unlock_irqrestore_rcu_node(rnp_old, flags);
 	rcu_gp_kthread_wake();
 }
+EXPORT_SYMBOL_GPL(rcu_force_quiescent_state);
 
 /*
  * This function checks for grace-period requests that fail to motivate
@@ -2801,9 +2792,9 @@ static void __call_rcu_core(struct rcu_data *rdp, struct rcu_head *head,
 
 	/*
 	 * Force the grace period if too many callbacks or too long waiting.
-	 * Enforce hysteresis, and don't invoke force_quiescent_state()
+	 * Enforce hysteresis, and don't invoke rcu_force_quiescent_state()
 	 * if some other CPU has recently done so.  Also, don't bother
-	 * invoking force_quiescent_state() if the newly enqueued callback
+	 * invoking rcu_force_quiescent_state() if the newly enqueued callback
 	 * is the only one waiting for a grace period to complete.
 	 */
 	if (unlikely(rcu_segcblist_n_cbs(&rdp->cblist) >
@@ -2820,7 +2811,7 @@ static void __call_rcu_core(struct rcu_data *rdp, struct rcu_head *head,
 			rdp->blimit = LONG_MAX;
 			if (rcu_state.n_force_qs == rdp->n_force_qs_snap &&
 			    rcu_segcblist_first_pend_cb(&rdp->cblist) != head)
-				force_quiescent_state();
+				rcu_force_quiescent_state();
 			rdp->n_force_qs_snap = rcu_state.n_force_qs;
 			rdp->qlen_last_fqs_check = rcu_segcblist_n_cbs(&rdp->cblist);
 		}
-- 
2.17.1

[PATCH tip/core/rcu 06/12] rcu: Eliminate RCU_BH_FLAVOR and RCU_SCHED_FLAVOR

[Paul E. McKenney]

Now that the RCU flavors have been consolidated, RCU_BH_FLAVOR and
RCU_SCHED_FLAVOR are no longer used.  This commit therefore saves a
few lines by removing them.

Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
---
 kernel/rcu/rcu.h  | 2 --
 kernel/rcu/tree.c | 2 --
 2 files changed, 4 deletions(-)

diff --git a/kernel/rcu/rcu.h b/kernel/rcu/rcu.h
index a393e24a9195..75787186bd4f 100644
--- a/kernel/rcu/rcu.h
+++ b/kernel/rcu/rcu.h
@@ -462,8 +462,6 @@ void rcu_request_urgent_qs_task(struct task_struct *t);
 
 enum rcutorture_type {
 	RCU_FLAVOR,
-	RCU_BH_FLAVOR,
-	RCU_SCHED_FLAVOR,
 	RCU_TASKS_FLAVOR,
 	SRCU_FLAVOR,
 	INVALID_RCU_FLAVOR
diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
index e56a46444775..fc37bec32731 100644
--- a/kernel/rcu/tree.c
+++ b/kernel/rcu/tree.c
@@ -556,8 +556,6 @@ void rcutorture_get_gp_data(enum rcutorture_type test_type, int *flags,
 {
 	switch (test_type) {
 	case RCU_FLAVOR:
-	case RCU_BH_FLAVOR:
-	case RCU_SCHED_FLAVOR:
 		*flags = READ_ONCE(rcu_state.gp_flags);
 		*gp_seq = rcu_seq_current(&rcu_state.gp_seq);
 		break;
-- 
2.17.1

[PATCH tip/core/rcu 07/12] rcu: Inline rcu_kthread_do_work() into its sole remaining caller

[Paul E. McKenney]

The rcu_kthread_do_work() function has a single-line body and only one
remaining caller.  This commit therefore saves a few lines of code by
inlining rcu_kthread_do_work() into its sole remaining caller.

Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
---
 kernel/rcu/tree_plugin.h | 7 +------
 1 file changed, 1 insertion(+), 6 deletions(-)

diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h
index 4d4091565a2c..bcf3e7366a28 100644
--- a/kernel/rcu/tree_plugin.h
+++ b/kernel/rcu/tree_plugin.h
@@ -1369,11 +1369,6 @@ static int rcu_spawn_one_boost_kthread(struct rcu_node *rnp)
 	return 0;
 }
 
-static void rcu_kthread_do_work(void)
-{
-	rcu_do_batch(this_cpu_ptr(&rcu_data));
-}
-
 static void rcu_cpu_kthread_setup(unsigned int cpu)
 {
 	struct sched_param sp;
@@ -1413,7 +1408,7 @@ static void rcu_cpu_kthread(unsigned int cpu)
 		*workp = 0;
 		local_irq_enable();
 		if (work)
-			rcu_kthread_do_work();
+			rcu_do_batch(this_cpu_ptr(&rcu_data));
 		local_bh_enable();
 		if (*workp == 0) {
 			trace_rcu_utilization(TPS("End CPU kthread@rcu_wait"));
-- 
2.17.1

[PATCH tip/core/rcu 08/12] rcu: Determine expedited-GP IPI handler at build time

[Paul E. McKenney]

Back when there could be multiple RCU flavors running in the same kernel
at the same time, it was necessary to specify the expedited grace-period
IPI handler at runtime.  Now that there is only one RCU flavor, the
IPI handler can be determined at build time.  There is therefore no
longer any reason for the RCU-preempt and RCU-sched IPI handlers to
have different names, nor is there any reason to pass these handlers in
function arguments and in the data structures enclosing workqueues.

This commit therefore makes all these changes, pushing the specification
of the expedited grace-period IPI handler down to the point of use.

Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
---
 .../Expedited-Grace-Periods/ExpSchedFlow.svg  | 18 ++++++-----
 .../Expedited-Grace-Periods.html              | 26 ++++++++--------
 kernel/rcu/tree.h                             |  1 -
 kernel/rcu/tree_exp.h                         | 30 +++++++++----------
 4 files changed, 38 insertions(+), 37 deletions(-)

diff --git a/Documentation/RCU/Design/Expedited-Grace-Periods/ExpSchedFlow.svg b/Documentation/RCU/Design/Expedited-Grace-Periods/ExpSchedFlow.svg
index e4233ac93c2b..6189ffcc6aff 100644
--- a/Documentation/RCU/Design/Expedited-Grace-Periods/ExpSchedFlow.svg
+++ b/Documentation/RCU/Design/Expedited-Grace-Periods/ExpSchedFlow.svg
@@ -328,13 +328,13 @@
      inkscape:window-height="1148"
      id="namedview90"
      showgrid="true"
-     inkscape:zoom="0.80021373"
-     inkscape:cx="462.49289"
-     inkscape:cy="473.6718"
+     inkscape:zoom="0.69092787"
+     inkscape:cx="476.34085"
+     inkscape:cy="712.80957"
      inkscape:window-x="770"
      inkscape:window-y="24"
      inkscape:window-maximized="0"
-     inkscape:current-layer="g4114-9-3-9"
+     inkscape:current-layer="g4"
      inkscape:snap-grids="false"
      fit-margin-top="5"
      fit-margin-right="5"
@@ -813,14 +813,18 @@
       <text
          sodipodi:linespacing="125%"
          id="text4110-5-7-6-2-4-0"
-         y="841.88086"
+         y="670.74316"
          x="1460.1007"
          style="font-size:267.24359131px;font-style:normal;font-weight:normal;text-align:center;line-height:125%;letter-spacing:0px;word-spacing:0px;text-anchor:middle;fill:#000000;fill-opacity:1;stroke:none;font-family:Sans"
          xml:space="preserve"><tspan
-           y="841.88086"
+           y="670.74316"
+           x="1460.1007"
+           sodipodi:role="line"
+           id="tspan4925-1-2-4-5">Request</tspan><tspan
+           y="1004.7976"
            x="1460.1007"
            sodipodi:role="line"
-           id="tspan4925-1-2-4-5">reched_cpu()</tspan></text>
+           id="tspan3100">context switch</tspan></text>
     </g>
   </g>
 </svg>
diff --git a/Documentation/RCU/Design/Expedited-Grace-Periods/Expedited-Grace-Periods.html b/Documentation/RCU/Design/Expedited-Grace-Periods/Expedited-Grace-Periods.html
index 8e4f873b979f..19e7a5fb6b73 100644
--- a/Documentation/RCU/Design/Expedited-Grace-Periods/Expedited-Grace-Periods.html
+++ b/Documentation/RCU/Design/Expedited-Grace-Periods/Expedited-Grace-Periods.html
@@ -72,10 +72,10 @@ will ignore it because idle and offline CPUs are already residing
 in quiescent states.
 Otherwise, the expedited grace period will use
 <tt>smp_call_function_single()</tt> to send the CPU an IPI, which
-is handled by <tt>sync_rcu_exp_handler()</tt>.
+is handled by <tt>rcu_exp_handler()</tt>.
 
 <p>
-However, because this is preemptible RCU, <tt>sync_rcu_exp_handler()</tt>
+However, because this is preemptible RCU, <tt>rcu_exp_handler()</tt>
 can check to see if the CPU is currently running in an RCU read-side
 critical section.
 If not, the handler can immediately report a quiescent state.
@@ -145,19 +145,18 @@ expedited grace period is shown in the following diagram:
 <p><img src="ExpSchedFlow.svg" alt="ExpSchedFlow.svg" width="55%">
 
 <p>
-As with RCU-preempt's <tt>synchronize_rcu_expedited()</tt>,
+As with RCU-preempt, RCU-sched's
 <tt>synchronize_sched_expedited()</tt> ignores offline and
 idle CPUs, again because they are in remotely detectable
 quiescent states.
-However, the <tt>synchronize_rcu_expedited()</tt> handler
-is <tt>sync_sched_exp_handler()</tt>, and because the
+However, because the
 <tt>rcu_read_lock_sched()</tt> and <tt>rcu_read_unlock_sched()</tt>
 leave no trace of their invocation, in general it is not possible to tell
 whether or not the current CPU is in an RCU read-side critical section.
-The best that <tt>sync_sched_exp_handler()</tt> can do is to check
+The best that RCU-sched's <tt>rcu_exp_handler()</tt> can do is to check
 for idle, on the off-chance that the CPU went idle while the IPI
 was in flight.
-If the CPU is idle, then <tt>sync_sched_exp_handler()</tt> reports
+If the CPU is idle, then <tt>rcu_exp_handler()</tt> reports
 the quiescent state.
 
 <p> Otherwise, the handler forces a future context switch by setting the
@@ -298,19 +297,18 @@ Instead, the task pushing the grace period forward will include the
 idle CPUs in the mask passed to <tt>rcu_report_exp_cpu_mult()</tt>.
 
 <p>
-For RCU-sched, there is an additional check for idle in the IPI
-handler, <tt>sync_sched_exp_handler()</tt>.
+For RCU-sched, there is an additional check:
 If the IPI has interrupted the idle loop, then
-<tt>sync_sched_exp_handler()</tt> invokes <tt>rcu_report_exp_rdp()</tt>
+<tt>rcu_exp_handler()</tt> invokes <tt>rcu_report_exp_rdp()</tt>
 to report the corresponding quiescent state.
 
 <p>
 For RCU-preempt, there is no specific check for idle in the
-IPI handler (<tt>sync_rcu_exp_handler()</tt>), but because
+IPI handler (<tt>rcu_exp_handler()</tt>), but because
 RCU read-side critical sections are not permitted within the
-idle loop, if <tt>sync_rcu_exp_handler()</tt> sees that the CPU is within
+idle loop, if <tt>rcu_exp_handler()</tt> sees that the CPU is within
 RCU read-side critical section, the CPU cannot possibly be idle.
-Otherwise, <tt>sync_rcu_exp_handler()</tt> invokes
+Otherwise, <tt>rcu_exp_handler()</tt> invokes
 <tt>rcu_report_exp_rdp()</tt> to report the corresponding quiescent
 state, regardless of whether or not that quiescent state was due to
 the CPU being idle.
@@ -625,6 +623,8 @@ checks, but only during the mid-boot dead zone.
 <p>
 With this refinement, synchronous grace periods can now be used from
 task context pretty much any time during the life of the kernel.
+That is, aside from some points in the suspend, hibernate, or shutdown
+code path.
 
 <h3><a name="Summary">
 Summary</a></h3>
diff --git a/kernel/rcu/tree.h b/kernel/rcu/tree.h
index bcfd684a5c57..50bb41cdc5fb 100644
--- a/kernel/rcu/tree.h
+++ b/kernel/rcu/tree.h
@@ -36,7 +36,6 @@
 
 /* Communicate arguments to a workqueue handler. */
 struct rcu_exp_work {
-	smp_call_func_t rew_func;
 	unsigned long rew_s;
 	struct work_struct rew_work;
 };
diff --git a/kernel/rcu/tree_exp.h b/kernel/rcu/tree_exp.h
index 6d4eb4694b6f..7f5cb4228b59 100644
--- a/kernel/rcu/tree_exp.h
+++ b/kernel/rcu/tree_exp.h
@@ -22,6 +22,8 @@
 
 #include <linux/lockdep.h>
 
+static void rcu_exp_handler(void *unused);
+
 /*
  * Record the start of an expedited grace period.
  */
@@ -344,7 +346,6 @@ static void sync_rcu_exp_select_node_cpus(struct work_struct *wp)
 {
 	int cpu;
 	unsigned long flags;
-	smp_call_func_t func;
 	unsigned long mask_ofl_test;
 	unsigned long mask_ofl_ipi;
 	int ret;
@@ -352,7 +353,6 @@ static void sync_rcu_exp_select_node_cpus(struct work_struct *wp)
 		container_of(wp, struct rcu_exp_work, rew_work);
 	struct rcu_node *rnp = container_of(rewp, struct rcu_node, rew);
 
-	func = rewp->rew_func;
 	raw_spin_lock_irqsave_rcu_node(rnp, flags);
 
 	/* Each pass checks a CPU for identity, offline, and idle. */
@@ -396,7 +396,7 @@ static void sync_rcu_exp_select_node_cpus(struct work_struct *wp)
 			mask_ofl_test |= mask;
 			continue;
 		}
-		ret = smp_call_function_single(cpu, func, NULL, 0);
+		ret = smp_call_function_single(cpu, rcu_exp_handler, NULL, 0);
 		if (!ret) {
 			mask_ofl_ipi &= ~mask;
 			continue;
@@ -426,7 +426,7 @@ static void sync_rcu_exp_select_node_cpus(struct work_struct *wp)
  * Select the nodes that the upcoming expedited grace period needs
  * to wait for.
  */
-static void sync_rcu_exp_select_cpus(smp_call_func_t func)
+static void sync_rcu_exp_select_cpus(void)
 {
 	int cpu;
 	struct rcu_node *rnp;
@@ -440,7 +440,6 @@ static void sync_rcu_exp_select_cpus(smp_call_func_t func)
 		rnp->exp_need_flush = false;
 		if (!READ_ONCE(rnp->expmask))
 			continue; /* Avoid early boot non-existent wq. */
-		rnp->rew.rew_func = func;
 		if (!READ_ONCE(rcu_par_gp_wq) ||
 		    rcu_scheduler_active != RCU_SCHEDULER_RUNNING ||
 		    rcu_is_last_leaf_node(rnp)) {
@@ -580,10 +579,10 @@ static void rcu_exp_wait_wake(unsigned long s)
  * Common code to drive an expedited grace period forward, used by
  * workqueues and mid-boot-time tasks.
  */
-static void rcu_exp_sel_wait_wake(smp_call_func_t func, unsigned long s)
+static void rcu_exp_sel_wait_wake(unsigned long s)
 {
 	/* Initialize the rcu_node tree in preparation for the wait. */
-	sync_rcu_exp_select_cpus(func);
+	sync_rcu_exp_select_cpus();
 
 	/* Wait and clean up, including waking everyone. */
 	rcu_exp_wait_wake(s);
@@ -597,14 +596,14 @@ static void wait_rcu_exp_gp(struct work_struct *wp)
 	struct rcu_exp_work *rewp;
 
 	rewp = container_of(wp, struct rcu_exp_work, rew_work);
-	rcu_exp_sel_wait_wake(rewp->rew_func, rewp->rew_s);
+	rcu_exp_sel_wait_wake(rewp->rew_s);
 }
 
 /*
  * Given a smp_call_function() handler, kick off the specified
  * implementation of expedited grace period.
  */
-static void _synchronize_rcu_expedited(smp_call_func_t func)
+static void _synchronize_rcu_expedited(void)
 {
 	struct rcu_data *rdp;
 	struct rcu_exp_work rew;
@@ -625,10 +624,9 @@ static void _synchronize_rcu_expedited(smp_call_func_t func)
 	/* Ensure that load happens before action based on it. */
 	if (unlikely(rcu_scheduler_active == RCU_SCHEDULER_INIT)) {
 		/* Direct call during scheduler init and early_initcalls(). */
-		rcu_exp_sel_wait_wake(func, s);
+		rcu_exp_sel_wait_wake(s);
 	} else {
 		/* Marshall arguments & schedule the expedited grace period. */
-		rew.rew_func = func;
 		rew.rew_s = s;
 		INIT_WORK_ONSTACK(&rew.rew_work, wait_rcu_exp_gp);
 		queue_work(rcu_gp_wq, &rew.rew_work);
@@ -654,7 +652,7 @@ static void _synchronize_rcu_expedited(smp_call_func_t func)
  * ->expmask fields in the rcu_node tree.  Otherwise, immediately
  * report the quiescent state.
  */
-static void sync_rcu_exp_handler(void *unused)
+static void rcu_exp_handler(void *unused)
 {
 	unsigned long flags;
 	struct rcu_data *rdp = this_cpu_ptr(&rcu_data);
@@ -760,14 +758,14 @@ void synchronize_rcu_expedited(void)
 
 	if (rcu_scheduler_active == RCU_SCHEDULER_INACTIVE)
 		return;
-	_synchronize_rcu_expedited(sync_rcu_exp_handler);
+	_synchronize_rcu_expedited();
 }
 EXPORT_SYMBOL_GPL(synchronize_rcu_expedited);
 
 #else /* #ifdef CONFIG_PREEMPT_RCU */
 
 /* Invoked on each online non-idle CPU for expedited quiescent state. */
-static void sync_sched_exp_handler(void *unused)
+static void rcu_exp_handler(void *unused)
 {
 	struct rcu_data *rdp;
 	struct rcu_node *rnp;
@@ -799,7 +797,7 @@ static void sync_sched_exp_online_cleanup(int cpu)
 	rnp = rdp->mynode;
 	if (!(READ_ONCE(rnp->expmask) & rdp->grpmask))
 		return;
-	ret = smp_call_function_single(cpu, sync_sched_exp_handler, NULL, 0);
+	ret = smp_call_function_single(cpu, rcu_exp_handler, NULL, 0);
 	WARN_ON_ONCE(ret);
 }
 
@@ -835,7 +833,7 @@ void synchronize_rcu_expedited(void)
 	if (rcu_blocking_is_gp())
 		return;
 
-	_synchronize_rcu_expedited(sync_sched_exp_handler);
+	_synchronize_rcu_expedited();
 }
 EXPORT_SYMBOL_GPL(synchronize_rcu_expedited);
 
-- 
2.17.1

[PATCH tip/core/rcu 09/12] rcu: Consolidate PREEMPT and !PREEMPT synchronize_rcu_expedited()

[Paul E. McKenney]

The CONFIG_PREEMPT=n and CONFIG_PREEMPT=y implementations of
synchronize_rcu_expedited() are quite similar, and with small
modifications to rcu_blocking_is_gp() can be made identical.  This commit
therefore makes this change in order to save a few lines of code and to
reduce the amount of duplicate code.

Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
---
 kernel/rcu/tree_exp.h | 105 ++++++++++++++++++++----------------------
 1 file changed, 49 insertions(+), 56 deletions(-)

diff --git a/kernel/rcu/tree_exp.h b/kernel/rcu/tree_exp.h
index 7f5cb4228b59..b800bdfe74b3 100644
--- a/kernel/rcu/tree_exp.h
+++ b/kernel/rcu/tree_exp.h
@@ -643,6 +643,33 @@ static void _synchronize_rcu_expedited(void)
 	mutex_unlock(&rcu_state.exp_mutex);
 }
 
+/*
+ * During early boot, any blocking grace-period wait automatically
+ * implies a grace period.  Later on, this is never the case for PREEMPT.
+ *
+ * Howevr, because a context switch is a grace period for !PREEMPT, any
+ * blocking grace-period wait automatically implies a grace period if
+ * there is only one CPU online at any point time during execution of
+ * either synchronize_rcu() or synchronize_rcu_expedited().  It is OK to
+ * occasionally incorrectly indicate that there are multiple CPUs online
+ * when there was in fact only one the whole time, as this just adds some
+ * overhead: RCU still operates correctly.
+ */
+static int rcu_blocking_is_gp(void)
+{
+	int ret;
+
+	if (rcu_scheduler_active == RCU_SCHEDULER_INACTIVE)
+		return true;
+	if (IS_ENABLED(CONFIG_PREEMPT))
+		return false;
+	might_sleep();  /* Check for RCU read-side critical section. */
+	preempt_disable();
+	ret = num_online_cpus() <= 1;
+	preempt_enable();
+	return ret;
+}
+
 #ifdef CONFIG_PREEMPT_RCU
 
 /*
@@ -729,39 +756,6 @@ static void sync_sched_exp_online_cleanup(int cpu)
 {
 }
 
-/**
- * synchronize_rcu_expedited - Brute-force RCU grace period
- *
- * Wait for an RCU-preempt grace period, but expedite it.  The basic
- * idea is to IPI all non-idle non-nohz online CPUs.  The IPI handler
- * checks whether the CPU is in an RCU-preempt critical section, and
- * if so, it sets a flag that causes the outermost rcu_read_unlock()
- * to report the quiescent state.  On the other hand, if the CPU is
- * not in an RCU read-side critical section, the IPI handler reports
- * the quiescent state immediately.
- *
- * Although this is a greate improvement over previous expedited
- * implementations, it is still unfriendly to real-time workloads, so is
- * thus not recommended for any sort of common-case code.  In fact, if
- * you are using synchronize_rcu_expedited() in a loop, please restructure
- * your code to batch your updates, and then Use a single synchronize_rcu()
- * instead.
- *
- * This has the same semantics as (but is more brutal than) synchronize_rcu().
- */
-void synchronize_rcu_expedited(void)
-{
-	RCU_LOCKDEP_WARN(lock_is_held(&rcu_bh_lock_map) ||
-			 lock_is_held(&rcu_lock_map) ||
-			 lock_is_held(&rcu_sched_lock_map),
-			 "Illegal synchronize_rcu_expedited() in RCU read-side critical section");
-
-	if (rcu_scheduler_active == RCU_SCHEDULER_INACTIVE)
-		return;
-	_synchronize_rcu_expedited();
-}
-EXPORT_SYMBOL_GPL(synchronize_rcu_expedited);
-
 #else /* #ifdef CONFIG_PREEMPT_RCU */
 
 /* Invoked on each online non-idle CPU for expedited quiescent state. */
@@ -801,27 +795,28 @@ static void sync_sched_exp_online_cleanup(int cpu)
 	WARN_ON_ONCE(ret);
 }
 
-/*
- * Because a context switch is a grace period for !PREEMPT, any
- * blocking grace-period wait automatically implies a grace period if
- * there is only one CPU online at any point time during execution of
- * either synchronize_rcu() or synchronize_rcu_expedited().  It is OK to
- * occasionally incorrectly indicate that there are multiple CPUs online
- * when there was in fact only one the whole time, as this just adds some
- * overhead: RCU still operates correctly.
- */
-static int rcu_blocking_is_gp(void)
-{
-	int ret;
-
-	might_sleep();  /* Check for RCU read-side critical section. */
-	preempt_disable();
-	ret = num_online_cpus() <= 1;
-	preempt_enable();
-	return ret;
-}
+#endif /* #else #ifdef CONFIG_PREEMPT_RCU */
 
-/* PREEMPT=n implementation of synchronize_rcu_expedited(). */
+/**
+ * synchronize_rcu_expedited - Brute-force RCU grace period
+ *
+ * Wait for an RCU grace period, but expedite it.  The basic idea is to
+ * IPI all non-idle non-nohz online CPUs.  The IPI handler checks whether
+ * the CPU is in an RCU critical section, and if so, it sets a flag that
+ * causes the outermost rcu_read_unlock() to report the quiescent state
+ * for RCU-preempt or asks the scheduler for help for RCU-sched.  On the
+ * other hand, if the CPU is not in an RCU read-side critical section,
+ * the IPI handler reports the quiescent state immediately.
+ *
+ * Although this is a greate improvement over previous expedited
+ * implementations, it is still unfriendly to real-time workloads, so is
+ * thus not recommended for any sort of common-case code.  In fact, if
+ * you are using synchronize_rcu_expedited() in a loop, please restructure
+ * your code to batch your updates, and then Use a single synchronize_rcu()
+ * instead.
+ *
+ * This has the same semantics as (but is more brutal than) synchronize_rcu().
+ */
 void synchronize_rcu_expedited(void)
 {
 	RCU_LOCKDEP_WARN(lock_is_held(&rcu_bh_lock_map) ||
@@ -829,12 +824,10 @@ void synchronize_rcu_expedited(void)
 			 lock_is_held(&rcu_sched_lock_map),
 			 "Illegal synchronize_rcu_expedited() in RCU read-side critical section");
 
-	/* If only one CPU, this is automatically a grace period. */
+	/* Is the state is such that the call is a grace period? */
 	if (rcu_blocking_is_gp())
 		return;
 
 	_synchronize_rcu_expedited();
 }
 EXPORT_SYMBOL_GPL(synchronize_rcu_expedited);
-
-#endif /* #else #ifdef CONFIG_PREEMPT_RCU */
-- 
2.17.1

[PATCH tip/core/rcu 10/12] rcu: Consolidate PREEMPT and !PREEMPT synchronize_rcu()

[Paul E. McKenney]

Now that rcu_blocking_is_gp() makes the correct immediate-return
decision for both PREEMPT and !PREEMPT, a single implementation of
synchronize_rcu() will work correctly under both configurations.
This commit therefore eliminates a few lines of code by consolidating
the two implementations of synchronize_rcu().

Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
---
 kernel/rcu/tree.c        | 73 ++++++++++++++++++++++++++++++++++++++++
 kernel/rcu/tree_exp.h    | 27 ---------------
 kernel/rcu/tree_plugin.h | 64 -----------------------------------
 3 files changed, 73 insertions(+), 91 deletions(-)

diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
index fc37bec32731..e2bd42b2b563 100644
--- a/kernel/rcu/tree.c
+++ b/kernel/rcu/tree.c
@@ -2950,6 +2950,79 @@ void kfree_call_rcu(struct rcu_head *head, rcu_callback_t func)
 }
 EXPORT_SYMBOL_GPL(kfree_call_rcu);
 
+/*
+ * During early boot, any blocking grace-period wait automatically
+ * implies a grace period.  Later on, this is never the case for PREEMPT.
+ *
+ * Howevr, because a context switch is a grace period for !PREEMPT, any
+ * blocking grace-period wait automatically implies a grace period if
+ * there is only one CPU online at any point time during execution of
+ * either synchronize_rcu() or synchronize_rcu_expedited().  It is OK to
+ * occasionally incorrectly indicate that there are multiple CPUs online
+ * when there was in fact only one the whole time, as this just adds some
+ * overhead: RCU still operates correctly.
+ */
+static int rcu_blocking_is_gp(void)
+{
+	int ret;
+
+	if (IS_ENABLED(CONFIG_PREEMPT))
+		return rcu_scheduler_active == RCU_SCHEDULER_INACTIVE;
+	might_sleep();  /* Check for RCU read-side critical section. */
+	preempt_disable();
+	ret = num_online_cpus() <= 1;
+	preempt_enable();
+	return ret;
+}
+
+/**
+ * synchronize_rcu - wait until a grace period has elapsed.
+ *
+ * Control will return to the caller some time after a full grace
+ * period has elapsed, in other words after all currently executing RCU
+ * read-side critical sections have completed.  Note, however, that
+ * upon return from synchronize_rcu(), the caller might well be executing
+ * concurrently with new RCU read-side critical sections that began while
+ * synchronize_rcu() was waiting.  RCU read-side critical sections are
+ * delimited by rcu_read_lock() and rcu_read_unlock(), and may be nested.
+ * In addition, regions of code across which interrupts, preemption, or
+ * softirqs have been disabled also serve as RCU read-side critical
+ * sections.  This includes hardware interrupt handlers, softirq handlers,
+ * and NMI handlers.
+ *
+ * Note that this guarantee implies further memory-ordering guarantees.
+ * On systems with more than one CPU, when synchronize_rcu() returns,
+ * each CPU is guaranteed to have executed a full memory barrier since
+ * the end of its last RCU read-side critical section whose beginning
+ * preceded the call to synchronize_rcu().  In addition, each CPU having
+ * an RCU read-side critical section that extends beyond the return from
+ * synchronize_rcu() is guaranteed to have executed a full memory barrier
+ * after the beginning of synchronize_rcu() and before the beginning of
+ * that RCU read-side critical section.  Note that these guarantees include
+ * CPUs that are offline, idle, or executing in user mode, as well as CPUs
+ * that are executing in the kernel.
+ *
+ * Furthermore, if CPU A invoked synchronize_rcu(), which returned
+ * to its caller on CPU B, then both CPU A and CPU B are guaranteed
+ * to have executed a full memory barrier during the execution of
+ * synchronize_rcu() -- even if CPU A and CPU B are the same CPU (but
+ * again only if the system has more than one CPU).
+ */
+void synchronize_rcu(void)
+{
+	RCU_LOCKDEP_WARN(lock_is_held(&rcu_bh_lock_map) ||
+			 lock_is_held(&rcu_lock_map) ||
+			 lock_is_held(&rcu_sched_lock_map),
+			 "Illegal synchronize_rcu() in RCU read-side critical section");
+	if (rcu_blocking_is_gp())
+		return;
+	if (rcu_gp_is_expedited())
+		synchronize_rcu_expedited();
+	else
+		wait_rcu_gp(call_rcu);
+}
+EXPORT_SYMBOL_GPL(synchronize_rcu);
+
 /**
  * get_state_synchronize_rcu - Snapshot current RCU state
  *
diff --git a/kernel/rcu/tree_exp.h b/kernel/rcu/tree_exp.h
index b800bdfe74b3..353d113c0cd4 100644
--- a/kernel/rcu/tree_exp.h
+++ b/kernel/rcu/tree_exp.h
@@ -643,33 +643,6 @@ static void _synchronize_rcu_expedited(void)
 	mutex_unlock(&rcu_state.exp_mutex);
 }
 
-/*
- * During early boot, any blocking grace-period wait automatically
- * implies a grace period.  Later on, this is never the case for PREEMPT.
- *
- * Howevr, because a context switch is a grace period for !PREEMPT, any
- * blocking grace-period wait automatically implies a grace period if
- * there is only one CPU online at any point time during execution of
- * either synchronize_rcu() or synchronize_rcu_expedited().  It is OK to
- * occasionally incorrectly indicate that there are multiple CPUs online
- * when there was in fact only one the whole time, as this just adds some
- * overhead: RCU still operates correctly.
- */
-static int rcu_blocking_is_gp(void)
-{
-	int ret;
-
-	if (rcu_scheduler_active == RCU_SCHEDULER_INACTIVE)
-		return true;
-	if (IS_ENABLED(CONFIG_PREEMPT))
-		return false;
-	might_sleep();  /* Check for RCU read-side critical section. */
-	preempt_disable();
-	ret = num_online_cpus() <= 1;
-	preempt_enable();
-	return ret;
-}
-
 #ifdef CONFIG_PREEMPT_RCU
 
 /*
diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h
index bcf3e7366a28..43f3f2ee9d63 100644
--- a/kernel/rcu/tree_plugin.h
+++ b/kernel/rcu/tree_plugin.h
@@ -825,54 +825,6 @@ static void rcu_flavor_check_callbacks(int user)
 		t->rcu_read_unlock_special.b.need_qs = true;
 }
 
-/**
- * synchronize_rcu - wait until a grace period has elapsed.
- *
- * Control will return to the caller some time after a full grace
- * period has elapsed, in other words after all currently executing RCU
- * read-side critical sections have completed.  Note, however, that
- * upon return from synchronize_rcu(), the caller might well be executing
- * concurrently with new RCU read-side critical sections that began while
- * synchronize_rcu() was waiting.  RCU read-side critical sections are
- * delimited by rcu_read_lock() and rcu_read_unlock(), and may be nested.
- * In addition, regions of code across which interrupts, preemption, or
- * softirqs have been disabled also serve as RCU read-side critical
- * sections.  This includes hardware interrupt handlers, softirq handlers,
- * and NMI handlers.
- *
- * Note that this guarantee implies further memory-ordering guarantees.
- * On systems with more than one CPU, when synchronize_rcu() returns,
- * each CPU is guaranteed to have executed a full memory barrier since
- * the end of its last RCU read-side critical section whose beginning
- * preceded the call to synchronize_rcu().  In addition, each CPU having
- * an RCU read-side critical section that extends beyond the return from
- * synchronize_rcu() is guaranteed to have executed a full memory barrier
- * after the beginning of synchronize_rcu() and before the beginning of
- * that RCU read-side critical section.  Note that these guarantees include
- * CPUs that are offline, idle, or executing in user mode, as well as CPUs
- * that are executing in the kernel.
- *
- * Furthermore, if CPU A invoked synchronize_rcu(), which returned
- * to its caller on CPU B, then both CPU A and CPU B are guaranteed
- * to have executed a full memory barrier during the execution of
- * synchronize_rcu() -- even if CPU A and CPU B are the same CPU (but
- * again only if the system has more than one CPU).
- */
-void synchronize_rcu(void)
-{
-	RCU_LOCKDEP_WARN(lock_is_held(&rcu_bh_lock_map) ||
-			 lock_is_held(&rcu_lock_map) ||
-			 lock_is_held(&rcu_sched_lock_map),
-			 "Illegal synchronize_rcu() in RCU read-side critical section");
-	if (rcu_scheduler_active == RCU_SCHEDULER_INACTIVE)
-		return;
-	if (rcu_gp_is_expedited())
-		synchronize_rcu_expedited();
-	else
-		wait_rcu_gp(call_rcu);
-}
-EXPORT_SYMBOL_GPL(synchronize_rcu);
-
 /*
  * Check for a task exiting while in a preemptible-RCU read-side
  * critical section, clean up if so.  No need to issue warnings,
@@ -1115,22 +1067,6 @@ static void rcu_flavor_check_callbacks(int user)
 	}
 }
 
-/* PREEMPT=n implementation of synchronize_rcu(). */
-void synchronize_rcu(void)
-{
-	RCU_LOCKDEP_WARN(lock_is_held(&rcu_bh_lock_map) ||
-			 lock_is_held(&rcu_lock_map) ||
-			 lock_is_held(&rcu_sched_lock_map),
-			 "Illegal synchronize_rcu() in RCU read-side critical section");
-	if (rcu_blocking_is_gp())
-		return;
-	if (rcu_gp_is_expedited())
-		synchronize_rcu_expedited();
-	else
-		wait_rcu_gp(call_rcu);
-}
-EXPORT_SYMBOL_GPL(synchronize_rcu);
-
 /*
  * Because preemptible RCU does not exist, tasks cannot possibly exit
  * while in preemptible RCU read-side critical sections.
-- 
2.17.1

[PATCH tip/core/rcu 11/12] rcu: Inline _synchronize_rcu_expedited() into synchronize_rcu_expedited()

[Paul E. McKenney]

Now that _synchronize_rcu_expedited() has only one caller, and given that
this is a tail call, this commit inlines _synchronize_rcu_expedited()
into synchronize_rcu_expedited().

Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
---
 kernel/rcu/tree_exp.h | 81 +++++++++++++++++++------------------------
 1 file changed, 36 insertions(+), 45 deletions(-)

diff --git a/kernel/rcu/tree_exp.h b/kernel/rcu/tree_exp.h
index 353d113c0cd4..d882ca0cd01b 100644
--- a/kernel/rcu/tree_exp.h
+++ b/kernel/rcu/tree_exp.h
@@ -599,50 +599,6 @@ static void wait_rcu_exp_gp(struct work_struct *wp)
 	rcu_exp_sel_wait_wake(rewp->rew_s);
 }
 
-/*
- * Given a smp_call_function() handler, kick off the specified
- * implementation of expedited grace period.
- */
-static void _synchronize_rcu_expedited(void)
-{
-	struct rcu_data *rdp;
-	struct rcu_exp_work rew;
-	struct rcu_node *rnp;
-	unsigned long s;
-
-	/* If expedited grace periods are prohibited, fall back to normal. */
-	if (rcu_gp_is_normal()) {
-		wait_rcu_gp(call_rcu);
-		return;
-	}
-
-	/* Take a snapshot of the sequence number.  */
-	s = rcu_exp_gp_seq_snap();
-	if (exp_funnel_lock(s))
-		return;  /* Someone else did our work for us. */
-
-	/* Ensure that load happens before action based on it. */
-	if (unlikely(rcu_scheduler_active == RCU_SCHEDULER_INIT)) {
-		/* Direct call during scheduler init and early_initcalls(). */
-		rcu_exp_sel_wait_wake(s);
-	} else {
-		/* Marshall arguments & schedule the expedited grace period. */
-		rew.rew_s = s;
-		INIT_WORK_ONSTACK(&rew.rew_work, wait_rcu_exp_gp);
-		queue_work(rcu_gp_wq, &rew.rew_work);
-	}
-
-	/* Wait for expedited grace period to complete. */
-	rdp = per_cpu_ptr(&rcu_data, raw_smp_processor_id());
-	rnp = rcu_get_root();
-	wait_event(rnp->exp_wq[rcu_seq_ctr(s) & 0x3],
-		   sync_exp_work_done(s));
-	smp_mb(); /* Workqueue actions happen before return. */
-
-	/* Let the next expedited grace period start. */
-	mutex_unlock(&rcu_state.exp_mutex);
-}
-
 #ifdef CONFIG_PREEMPT_RCU
 
 /*
@@ -792,6 +748,11 @@ static void sync_sched_exp_online_cleanup(int cpu)
  */
 void synchronize_rcu_expedited(void)
 {
+	struct rcu_data *rdp;
+	struct rcu_exp_work rew;
+	struct rcu_node *rnp;
+	unsigned long s;
+
 	RCU_LOCKDEP_WARN(lock_is_held(&rcu_bh_lock_map) ||
 			 lock_is_held(&rcu_lock_map) ||
 			 lock_is_held(&rcu_sched_lock_map),
@@ -801,6 +762,36 @@ void synchronize_rcu_expedited(void)
 	if (rcu_blocking_is_gp())
 		return;
 
-	_synchronize_rcu_expedited();
+	/* If expedited grace periods are prohibited, fall back to normal. */
+	if (rcu_gp_is_normal()) {
+		wait_rcu_gp(call_rcu);
+		return;
+	}
+
+	/* Take a snapshot of the sequence number.  */
+	s = rcu_exp_gp_seq_snap();
+	if (exp_funnel_lock(s))
+		return;  /* Someone else did our work for us. */
+
+	/* Ensure that load happens before action based on it. */
+	if (unlikely(rcu_scheduler_active == RCU_SCHEDULER_INIT)) {
+		/* Direct call during scheduler init and early_initcalls(). */
+		rcu_exp_sel_wait_wake(s);
+	} else {
+		/* Marshall arguments & schedule the expedited grace period. */
+		rew.rew_s = s;
+		INIT_WORK_ONSTACK(&rew.rew_work, wait_rcu_exp_gp);
+		queue_work(rcu_gp_wq, &rew.rew_work);
+	}
+
+	/* Wait for expedited grace period to complete. */
+	rdp = per_cpu_ptr(&rcu_data, raw_smp_processor_id());
+	rnp = rcu_get_root();
+	wait_event(rnp->exp_wq[rcu_seq_ctr(s) & 0x3],
+		   sync_exp_work_done(s));
+	smp_mb(); /* Workqueue actions happen before return. */
+
+	/* Let the next expedited grace period start. */
+	mutex_unlock(&rcu_state.exp_mutex);
 }
 EXPORT_SYMBOL_GPL(synchronize_rcu_expedited);
-- 
2.17.1

[PATCH tip/core/rcu 12/12] rcu: Discard separate per-CPU callback counts

[Paul E. McKenney]

Back when there were multiple flavors of RCU, it was necessary to
separately count lazy and non-lazy callbacks for each CPU.  These counts
were used in CONFIG_RCU_FAST_NO_HZ kernels to determine how long a newly
idle CPU should be allowed to sleep before handling its RCU callbacks.
But now that there is only one flavor, the callback counts for a given
CPU's sole rcu_data structure are the counts for that CPU.

This commit therefore removes the rcu_data structure's ->nonlazy_posted
and ->nonlazy_posted_snap fields, the rcu_idle_count_callbacks_posted()
and rcu_cpu_has_callbacks() functions, repurposes the rcu_data structure's
->all_lazy field to record the laziness state at the beginning of the
latest idle sojourn, and modifies CONFIG_RCU_FAST_NO_HZ RCU CPU stall
warnings accordingly.

Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
---
 Documentation/RCU/stallwarn.txt | 15 +++++-----
 kernel/rcu/tree.c               | 25 -----------------
 kernel/rcu/tree.h               |  6 +---
 kernel/rcu/tree_plugin.h        | 50 ++++++++-------------------------
 4 files changed, 21 insertions(+), 75 deletions(-)

diff --git a/Documentation/RCU/stallwarn.txt b/Documentation/RCU/stallwarn.txt
index 073dbc12d1ea..1ab70c37921f 100644
--- a/Documentation/RCU/stallwarn.txt
+++ b/Documentation/RCU/stallwarn.txt
@@ -219,17 +219,18 @@ an estimate of the total number of RCU callbacks queued across all CPUs
 In kernels with CONFIG_RCU_FAST_NO_HZ, more information is printed
 for each CPU:
 
-	0: (64628 ticks this GP) idle=dd5/3fffffffffffffff/0 softirq=82/543 last_accelerate: a345/d342 nonlazy_posted: 25 .D
+	0: (64628 ticks this GP) idle=dd5/3fffffffffffffff/0 softirq=82/543 last_accelerate: a345/d342 Nonlazy posted: ..D
 
 The "last_accelerate:" prints the low-order 16 bits (in hex) of the
 jiffies counter when this CPU last invoked rcu_try_advance_all_cbs()
 from rcu_needs_cpu() or last invoked rcu_accelerate_cbs() from
-rcu_prepare_for_idle().  The "nonlazy_posted:" prints the number
-of non-lazy callbacks posted since the last call to rcu_needs_cpu().
-Finally, an "L" indicates that there are currently no non-lazy callbacks
-("." is printed otherwise, as shown above) and "D" indicates that
-dyntick-idle processing is enabled ("." is printed otherwise, for example,
-if disabled via the "nohz=" kernel boot parameter).
+rcu_prepare_for_idle().  The "Nonlazy posted:" indicates lazy-callback
+status, so that an "l" indicates that all callbacks were lazy at the start
+of the last idle period and an "L" indicates that there are currently
+no non-lazy callbacks (in both cases, "." is printed otherwise, as
+shown above) and "D" indicates that dyntick-idle processing is enabled
+("." is printed otherwise, for example, if disabled via the "nohz="
+kernel boot parameter).
 
 If the grace period ends just as the stall warning starts printing,
 there will be a spurious stall-warning message, which will include
diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
index e2bd42b2b563..e53a586b397b 100644
--- a/kernel/rcu/tree.c
+++ b/kernel/rcu/tree.c
@@ -2878,9 +2878,6 @@ __call_rcu(struct rcu_head *head, rcu_callback_t func, int cpu, bool lazy)
 			rcu_segcblist_init(&rdp->cblist);
 	}
 	rcu_segcblist_enqueue(&rdp->cblist, head, lazy);
-	if (!lazy)
-		rcu_idle_count_callbacks_posted();
-
 	if (__is_kfree_rcu_offset((unsigned long)func))
 		trace_rcu_kfree_callback(rcu_state.name, head,
 					 (unsigned long)func,
@@ -3110,28 +3107,6 @@ static int rcu_pending(void)
 	return 0;
 }
 
-/*
- * Return true if the specified CPU has any callback.  If all_lazy is
- * non-NULL, store an indication of whether all callbacks are lazy.
- * (If there are no callbacks, all of them are deemed to be lazy.)
- */
-static bool rcu_cpu_has_callbacks(bool *all_lazy)
-{
-	bool al = true;
-	bool hc = false;
-	struct rcu_data *rdp;
-
-	rdp = this_cpu_ptr(&rcu_data);
-	if (!rcu_segcblist_empty(&rdp->cblist)) {
-		hc = true;
-		if (rcu_segcblist_n_nonlazy_cbs(&rdp->cblist))
-			al = false;
-	}
-	if (all_lazy)
-		*all_lazy = al;
-	return hc;
-}
-
 /*
  * Helper function for rcu_barrier() tracing.  If tracing is disabled,
  * the compiler is expected to optimize this away.
diff --git a/kernel/rcu/tree.h b/kernel/rcu/tree.h
index 50bb41cdc5fb..20feecbb0ab6 100644
--- a/kernel/rcu/tree.h
+++ b/kernel/rcu/tree.h
@@ -193,10 +193,7 @@ struct rcu_data {
 	bool rcu_need_heavy_qs;		/* GP old, so heavy quiescent state! */
 	bool rcu_urgent_qs;		/* GP old need light quiescent state. */
 #ifdef CONFIG_RCU_FAST_NO_HZ
-	bool all_lazy;			/* Are all CPU's CBs lazy? */
-	unsigned long nonlazy_posted;	/* # times non-lazy CB posted to CPU. */
-	unsigned long nonlazy_posted_snap;
-					/* Nonlazy_posted snapshot. */
+	bool all_lazy;			/* All CPU's CBs lazy at idle start? */
 	unsigned long last_accelerate;	/* Last jiffy CBs were accelerated. */
 	unsigned long last_advance_all;	/* Last jiffy CBs were all advanced. */
 	int tick_nohz_enabled_snap;	/* Previously seen value from sysfs. */
@@ -430,7 +427,6 @@ static void __init rcu_spawn_boost_kthreads(void);
 static void rcu_prepare_kthreads(int cpu);
 static void rcu_cleanup_after_idle(void);
 static void rcu_prepare_for_idle(void);
-static void rcu_idle_count_callbacks_posted(void);
 static bool rcu_preempt_has_tasks(struct rcu_node *rnp);
 static bool rcu_preempt_need_deferred_qs(struct task_struct *t);
 static void rcu_preempt_deferred_qs(struct task_struct *t);
diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h
index 43f3f2ee9d63..abd238c70900 100644
--- a/kernel/rcu/tree_plugin.h
+++ b/kernel/rcu/tree_plugin.h
@@ -1474,7 +1474,7 @@ static void rcu_prepare_kthreads(int cpu)
 int rcu_needs_cpu(u64 basemono, u64 *nextevt)
 {
 	*nextevt = KTIME_MAX;
-	return rcu_cpu_has_callbacks(NULL);
+	return !rcu_segcblist_empty(&this_cpu_ptr(&rcu_data)->cblist);
 }
 
 /*
@@ -1493,14 +1493,6 @@ static void rcu_prepare_for_idle(void)
 {
 }
 
-/*
- * Don't bother keeping a running count of the number of RCU callbacks
- * posted because CONFIG_RCU_FAST_NO_HZ=n.
- */
-static void rcu_idle_count_callbacks_posted(void)
-{
-}
-
 #else /* #if !defined(CONFIG_RCU_FAST_NO_HZ) */
 
 /*
@@ -1583,11 +1575,8 @@ int rcu_needs_cpu(u64 basemono, u64 *nextevt)
 
 	lockdep_assert_irqs_disabled();
 
-	/* Snapshot to detect later posting of non-lazy callback. */
-	rdp->nonlazy_posted_snap = rdp->nonlazy_posted;
-
 	/* If no callbacks, RCU doesn't need the CPU. */
-	if (!rcu_cpu_has_callbacks(&rdp->all_lazy)) {
+	if (rcu_segcblist_empty(&rdp->cblist)) {
 		*nextevt = KTIME_MAX;
 		return 0;
 	}
@@ -1601,11 +1590,12 @@ int rcu_needs_cpu(u64 basemono, u64 *nextevt)
 	rdp->last_accelerate = jiffies;
 
 	/* Request timer delay depending on laziness, and round. */
-	if (!rdp->all_lazy) {
+	rdp->all_lazy = !rcu_segcblist_n_nonlazy_cbs(&rdp->cblist);
+	if (rdp->all_lazy) {
+		dj = round_jiffies(rcu_idle_lazy_gp_delay + jiffies) - jiffies;
+	} else {
 		dj = round_up(rcu_idle_gp_delay + jiffies,
 			       rcu_idle_gp_delay) - jiffies;
-	} else {
-		dj = round_jiffies(rcu_idle_lazy_gp_delay + jiffies) - jiffies;
 	}
 	*nextevt = basemono + dj * TICK_NSEC;
 	return 0;
@@ -1635,7 +1625,7 @@ static void rcu_prepare_for_idle(void)
 	/* Handle nohz enablement switches conservatively. */
 	tne = READ_ONCE(tick_nohz_active);
 	if (tne != rdp->tick_nohz_enabled_snap) {
-		if (rcu_cpu_has_callbacks(NULL))
+		if (!rcu_segcblist_empty(&rdp->cblist))
 			invoke_rcu_core(); /* force nohz to see update. */
 		rdp->tick_nohz_enabled_snap = tne;
 		return;
@@ -1648,10 +1638,8 @@ static void rcu_prepare_for_idle(void)
 	 * callbacks, invoke RCU core for the side-effect of recalculating
 	 * idle duration on re-entry to idle.
 	 */
-	if (rdp->all_lazy &&
-	    rdp->nonlazy_posted != rdp->nonlazy_posted_snap) {
+	if (rdp->all_lazy && rcu_segcblist_n_nonlazy_cbs(&rdp->cblist)) {
 		rdp->all_lazy = false;
-		rdp->nonlazy_posted_snap = rdp->nonlazy_posted;
 		invoke_rcu_core();
 		return;
 	}
@@ -1687,19 +1675,6 @@ static void rcu_cleanup_after_idle(void)
 		invoke_rcu_core();
 }
 
-/*
- * Keep a running count of the number of non-lazy callbacks posted
- * on this CPU.  This running counter (which is never decremented) allows
- * rcu_prepare_for_idle() to detect when something out of the idle loop
- * posts a callback, even if an equal number of callbacks are invoked.
- * Of course, callbacks should only be posted from within a trace event
- * designed to be called from idle or from within RCU_NONIDLE().
- */
-static void rcu_idle_count_callbacks_posted(void)
-{
-	__this_cpu_add(rcu_data.nonlazy_posted, 1);
-}
-
 #endif /* #else #if !defined(CONFIG_RCU_FAST_NO_HZ) */
 
 #ifdef CONFIG_RCU_FAST_NO_HZ
@@ -1707,13 +1682,12 @@ static void rcu_idle_count_callbacks_posted(void)
 static void print_cpu_stall_fast_no_hz(char *cp, int cpu)
 {
 	struct rcu_data *rdp = &per_cpu(rcu_data, cpu);
-	unsigned long nlpd = rdp->nonlazy_posted - rdp->nonlazy_posted_snap;
 
-	sprintf(cp, "last_accelerate: %04lx/%04lx, nonlazy_posted: %ld, %c%c",
+	sprintf(cp, "last_accelerate: %04lx/%04lx, Nonlazy posted: %c%c%c",
 		rdp->last_accelerate & 0xffff, jiffies & 0xffff,
-		ulong2long(nlpd),
-		rdp->all_lazy ? 'L' : '.',
-		rdp->tick_nohz_enabled_snap ? '.' : 'D');
+		".l"[rdp->all_lazy],
+		".L"[!rcu_segcblist_n_nonlazy_cbs(&rdp->cblist)],
+		".D"[!rdp->tick_nohz_enabled_snap]);
 }
 
 #else /* #ifdef CONFIG_RCU_FAST_NO_HZ */
-- 
2.17.1

[PATCH tip/core/rcu 13/12] Remove wrapper definitions for obsolete RCU update functions

[Paul E. McKenney]

None of synchronize_rcu_bh, synchronize_rcu_bh_expedited, call_rcu_bh,
rcu_barrier_bh, synchronize_sched, synchronize_sched_expedited,
call_rcu_sched, rcu_barrier_sched, get_state_synchronize_sched, and
cond_synchronize_sched are actually used.  This commit therefore removes
their trivial wrapper-function definitions.

Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
[ paulmck: The intent was to defer this to the following release, but
  there are no more uses of these functions, so full speed ahead!!! ]

diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index 4db8bcacc51a..0e39e0d2629e 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -896,57 +896,4 @@ rcu_head_after_call_rcu(struct rcu_head *rhp, rcu_callback_t f)
 	return false;
 }
 
-
-/* Transitional pre-consolidation compatibility definitions. */
-
-static inline void synchronize_rcu_bh(void)
-{
-	synchronize_rcu();
-}
-
-static inline void synchronize_rcu_bh_expedited(void)
-{
-	synchronize_rcu_expedited();
-}
-
-static inline void call_rcu_bh(struct rcu_head *head, rcu_callback_t func)
-{
-	call_rcu(head, func);
-}
-
-static inline void rcu_barrier_bh(void)
-{
-	rcu_barrier();
-}
-
-static inline void synchronize_sched(void)
-{
-	synchronize_rcu();
-}
-
-static inline void synchronize_sched_expedited(void)
-{
-	synchronize_rcu_expedited();
-}
-
-static inline void call_rcu_sched(struct rcu_head *head, rcu_callback_t func)
-{
-	call_rcu(head, func);
-}
-
-static inline void rcu_barrier_sched(void)
-{
-	rcu_barrier();
-}
-
-static inline unsigned long get_state_synchronize_sched(void)
-{
-	return get_state_synchronize_rcu();
-}
-
-static inline void cond_synchronize_sched(unsigned long oldstate)
-{
-	cond_synchronize_rcu(oldstate);
-}
-
 #endif /* __LINUX_RCUPDATE_H */