[PATCH tip/core/rcu 2/4] rcu-tasks: Add block comment laying out RCU Rude design

[PATCH tip/core/rcu 2/4] rcu-tasks: Add block comment laying out RCU Rude design

[Paul E. McKenney]

This commit adds a block comment that gives a high-level overview of
how RCU Rude grace periods progress.  It also gives an overview of the
memory ordering.

Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
---
 kernel/rcu/tasks.h | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/kernel/rcu/tasks.h b/kernel/rcu/tasks.h
index 94d2c2c7f0ab..d6aa352cd705 100644
--- a/kernel/rcu/tasks.h
+++ b/kernel/rcu/tasks.h
@@ -645,8 +645,13 @@ void exit_tasks_rcu_finish(void) { exit_tasks_rcu_finish_trace(current); }
 // passing an empty function to schedule_on_each_cpu().  This approach
 // provides an asynchronous call_rcu_tasks_rude() API and batching
 // of concurrent calls to the synchronous synchronize_rcu_rude() API.
-// This sends IPIs far and wide and induces otherwise unnecessary context
-// switches on all online CPUs, whether idle or not.
+// This invokes schedule_on_each_cpu() in order to send IPIs far and wide
+// and induces otherwise unnecessary context switches on all online CPUs,
+// whether idle or not.
+//
+// Callback handling is provided by the rcu_tasks_kthread() function.
+//
+// Ordering is provided by the scheduler's context-switch code.
 
 // Empty function to allow workqueues to force a context switch.
 static void rcu_tasks_be_rude(struct work_struct *work)
-- 
2.31.1.189.g2e36527f23

[PATCH tip/core/rcu 0/6] SRCU updates for v5.14

[Paul E. McKenney]

Hello!

This series provides some updates for SRCU, all courtesy of Frederic
Weisbecker.

1.	Remove superfluous sdp->srcu_lock_count zero filling.

2.	Remove superfluous ssp initialization for early callbacks.

3.	Unconditionally embed struct lockdep_map.

4.	Initialize SRCU after timers.

5.	Fix broken node geometry after early ssp init.

6.	Early test SRCU polling start.

						Thanx, Paul

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

 b/include/linux/srcu.h     |    6 ++++++
 b/include/linux/srcutree.h |    2 --
 b/init/main.c              |    2 ++
 b/kernel/rcu/rcu.h         |    6 ------
 b/kernel/rcu/srcutree.c    |   12 ++----------
 b/kernel/rcu/tiny.c        |    1 -
 b/kernel/rcu/tree.c        |    1 -
 b/kernel/rcu/update.c      |    6 +++++-
 kernel/rcu/rcu.h           |    2 ++
 kernel/rcu/srcutree.c      |    9 ++++++++-
 kernel/rcu/tree.c          |   16 +++++++++++++++-
 11 files changed, 40 insertions(+), 23 deletions(-)

[PATCH tip/core/rcu 2/4] rcu-tasks: Add block comment laying out RCU Rude design

[Paul E. McKenney]

This commit adds a block comment that gives a high-level overview of
how RCU Rude grace periods progress.  It also gives an overview of the
memory ordering.

Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
---
 kernel/rcu/tasks.h | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/kernel/rcu/tasks.h b/kernel/rcu/tasks.h
index 94d2c2c7f0ab..d6aa352cd705 100644
--- a/kernel/rcu/tasks.h
+++ b/kernel/rcu/tasks.h
@@ -645,8 +645,13 @@ void exit_tasks_rcu_finish(void) { exit_tasks_rcu_finish_trace(current); }
 // passing an empty function to schedule_on_each_cpu().  This approach
 // provides an asynchronous call_rcu_tasks_rude() API and batching
 // of concurrent calls to the synchronous synchronize_rcu_rude() API.
-// This sends IPIs far and wide and induces otherwise unnecessary context
-// switches on all online CPUs, whether idle or not.
+// This invokes schedule_on_each_cpu() in order to send IPIs far and wide
+// and induces otherwise unnecessary context switches on all online CPUs,
+// whether idle or not.
+//
+// Callback handling is provided by the rcu_tasks_kthread() function.
+//
+// Ordering is provided by the scheduler's context-switch code.
 
 // Empty function to allow workqueues to force a context switch.
 static void rcu_tasks_be_rude(struct work_struct *work)
-- 
2.31.1.189.g2e36527f23