[PATCH tip/core/rcu 7/8] documentation: Update RCU requirements based on expedited changes
From: Paul E. McKenney
Date: Fri Dec 04 2015 - 18:51:34 EST
Because RCU-sched expedited grace periods now use IPIs and interact
with rcu_read_unlock(), it is no longer sufficient to disable preemption
across RCU read-side critical sections that acquire and hold scheduler
locks. It is now necessary to instead disable interrupts. This commit
documents this change.
Signed-off-by: Paul E. McKenney <paulmck@xxxxxxxxxxxxxxxxxx>
---
Documentation/RCU/Design/Requirements/Requirements.html | 12 ++++++++++--
Documentation/RCU/Design/Requirements/Requirements.htmlx | 10 +++++++++-
2 files changed, 19 insertions(+), 3 deletions(-)
diff --git a/Documentation/RCU/Design/Requirements/Requirements.html b/Documentation/RCU/Design/Requirements/Requirements.html
index f5bf27228192..9aa8f2c402f5 100644
--- a/Documentation/RCU/Design/Requirements/Requirements.html
+++ b/Documentation/RCU/Design/Requirements/Requirements.html
@@ -1,5 +1,5 @@
<!-- DO NOT HAND EDIT. -->
-<!-- Instead, edit Requirements.htmlx and run 'sh htmlqqz.sh Requirements' -->
+<!-- Instead, edit Documentation/RCU/Design/Requirements/Requirements.htmlx and run 'sh htmlqqz.sh Documentation/RCU/Design/Requirements/Requirements' -->
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
"http://www.w3.org/TR/html4/loose.dtd";>
<html>
@@ -1957,7 +1957,7 @@ the runqueue locks and the priority-inheritance locks
in the middle of an outermost RCU read-side critical section unless either
(1) it releases them before exiting that same
RCU read-side critical section, or
-(2) preemption is disabled across
+(2) interrupts are disabled across
that entire RCU read-side critical section.
This same prohibition also applies (recursively!) to any lock that is acquired
while holding any lock to which this prohibition applies.
@@ -1966,6 +1966,14 @@ Adhering to this rule prevents preemptible RCU from invoking
priority-inheritance locks are held, thus avoiding deadlock.
<p>
+Prior to v4.4, it was only necessary to disable preemption across
+RCU read-side critical sections that acquired scheduler locks.
+In v4.4, expedited grace periods started using IPIs, and these
+IPIs could force a <tt>rcu_read_unlock()</tt> to take the slowpath.
+Therefore, this expedited-grace-period change required disabling of
+interrupts, not just preemption.
+
+<p>
For RCU's part, the preemptible-RCU <tt>rcu_read_unlock()</tt>
implementation must be written carefully to avoid similar deadlocks.
In particular, <tt>rcu_read_unlock()</tt> must tolerate an
diff --git a/Documentation/RCU/Design/Requirements/Requirements.htmlx b/Documentation/RCU/Design/Requirements/Requirements.htmlx
index 4082e588baa7..de86311e638e 100644
--- a/Documentation/RCU/Design/Requirements/Requirements.htmlx
+++ b/Documentation/RCU/Design/Requirements/Requirements.htmlx
@@ -2124,7 +2124,7 @@ the runqueue locks and the priority-inheritance locks
in the middle of an outermost RCU read-side critical section unless either
(1) it releases them before exiting that same
RCU read-side critical section, or
-(2) preemption is disabled across
+(2) interrupts are disabled across
that entire RCU read-side critical section.
This same prohibition also applies (recursively!) to any lock that is acquired
while holding any lock to which this prohibition applies.
@@ -2133,6 +2133,14 @@ Adhering to this rule prevents preemptible RCU from invoking
priority-inheritance locks are held, thus avoiding deadlock.
<p>
+Prior to v4.4, it was only necessary to disable preemption across
+RCU read-side critical sections that acquired scheduler locks.
+In v4.4, expedited grace periods started using IPIs, and these
+IPIs could force a <tt>rcu_read_unlock()</tt> to take the slowpath.
+Therefore, this expedited-grace-period change required disabling of
+interrupts, not just preemption.
+
+<p>
For RCU's part, the preemptible-RCU <tt>rcu_read_unlock()</tt>
implementation must be written carefully to avoid similar deadlocks.
In particular, <tt>rcu_read_unlock()</tt> must tolerate an
--
2.5.2
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/