rcu: Yet another fix for preemption and CPU hotplug

As noted earlier, the following sequence of events can occur when
running PREEMPT_RCU and HOTPLUG_CPU on a system with a multi-level
rcu_node combining tree:

1.	A group of tasks block on CPUs corresponding to a given leaf
	rcu_node structure while within RCU read-side critical sections.
2.	All CPUs corrsponding to that rcu_node structure go offline.
3.	The next grace period starts, but because there are still tasks
	blocked, the upper-level bits corresponding to this leaf rcu_node
	structure remain set.
4.	All the tasks exit their RCU read-side critical sections and
	remove themselves from the leaf rcu_node structure's list,
	leaving it empty.
5.	But because there now is code to check for this condition at
	force-quiescent-state time, the upper bits are cleared and the
	grace period completes.

However, there is another complication that can occur following step 4 above:

4a.	The grace period starts, and the leaf rcu_node structure's
	gp_tasks pointer is set to NULL because there are no tasks
	blocked on this structure.
4b.	One of the CPUs corresponding to the leaf rcu_node structure
	comes back online.
4b.	An endless stream of tasks are preempted within RCU read-side
	critical sections on this CPU, such that the ->blkd_tasks
	list is always non-empty.

The grace period will never end.

This commit therefore makes the force-quiescent-state processing check only
for absence of tasks blocking the current grace period rather than absence
of tasks altogether.  This will cause a quiescent state to be reported if
the current leaf rcu_node structure is not blocking the current grace period
and its parent thinks that it is, regardless of how RCU managed to get
itself into this state.

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: <stable@vger.kernel.org> # 4.0.x
Tested-by: Sasha Levin <sasha.levin@oracle.com>
This commit is contained in:
Paul E. McKenney 2015-03-08 14:52:27 -07:00
parent 5c60d25fa1
commit a77da14ce9

View file

@ -2199,8 +2199,8 @@ static void rcu_report_unblock_qs_rnp(struct rcu_state *rsp,
unsigned long mask;
struct rcu_node *rnp_p;
WARN_ON_ONCE(rsp == &rcu_bh_state || rsp == &rcu_sched_state);
if (rnp->qsmask != 0 || rcu_preempt_blocked_readers_cgp(rnp)) {
if (rcu_state_p == &rcu_sched_state || rsp != rcu_state_p ||
rnp->qsmask != 0 || rcu_preempt_blocked_readers_cgp(rnp)) {
raw_spin_unlock_irqrestore(&rnp->lock, flags);
return; /* Still need more quiescent states! */
}
@ -2208,9 +2208,8 @@ static void rcu_report_unblock_qs_rnp(struct rcu_state *rsp,
rnp_p = rnp->parent;
if (rnp_p == NULL) {
/*
* Either there is only one rcu_node in the tree,
* or tasks were kicked up to root rcu_node due to
* CPUs going offline.
* Only one rcu_node structure in the tree, so don't
* try to report up to its nonexistent parent!
*/
rcu_report_qs_rsp(rsp, flags);
return;
@ -2713,8 +2712,29 @@ static void force_qs_rnp(struct rcu_state *rsp,
return;
}
if (rnp->qsmask == 0) {
rcu_initiate_boost(rnp, flags); /* releases rnp->lock */
continue;
if (rcu_state_p == &rcu_sched_state ||
rsp != rcu_state_p ||
rcu_preempt_blocked_readers_cgp(rnp)) {
/*
* No point in scanning bits because they
* are all zero. But we might need to
* priority-boost blocked readers.
*/
rcu_initiate_boost(rnp, flags);
/* rcu_initiate_boost() releases rnp->lock */
continue;
}
if (rnp->parent &&
(rnp->parent->qsmask & rnp->grpmask)) {
/*
* Race between grace-period
* initialization and task exiting RCU
* read-side critical section: Report.
*/
rcu_report_unblock_qs_rnp(rsp, rnp, flags);
/* rcu_report_unblock_qs_rnp() rlses ->lock */
continue;
}
}
cpu = rnp->grplo;
bit = 1;
@ -2729,15 +2749,6 @@ static void force_qs_rnp(struct rcu_state *rsp,
if (mask != 0) {
/* Idle/offline CPUs, report. */
rcu_report_qs_rnp(mask, rsp, rnp, flags);
} else if (rnp->parent &&
list_empty(&rnp->blkd_tasks) &&
!rnp->qsmask &&
(rnp->parent->qsmask & rnp->grpmask)) {
/*
* Race between grace-period initialization and task
* existing RCU read-side critical section, report.
*/
rcu_report_unblock_qs_rnp(rsp, rnp, flags);
} else {
/* Nothing to do here, so just drop the lock. */
raw_spin_unlock_irqrestore(&rnp->lock, flags);