diff options
author | Paul E. McKenney <paul.mckenney@linaro.org> | 2012-07-31 17:09:49 -0400 |
---|---|---|
committer | Paul E. McKenney <paulmck@linux.vnet.ibm.com> | 2012-09-23 10:42:50 -0400 |
commit | 5cc900cf55fe58aaad93767c5a526e2a69cbcbc6 (patch) | |
tree | d6eb08a2df0797927d5288e2fa0ba9f2b8b1d4b6 /kernel/rcutree_plugin.h | |
parent | b4270ee356e5ecef5394ab80c0a0301c1676b7f0 (diff) |
rcu: Improve boost selection when moving tasks to root rcu_node
The rcu_preempt_offline_tasks() moves all tasks queued on a given leaf
rcu_node structure to the root rcu_node, which is done when the last CPU
corresponding the the leaf rcu_node structure goes offline. Now that
RCU-preempt's synchronize_rcu_expedited() implementation blocks CPU-hotplug
operations during the initialization of each rcu_node structure's
->boost_tasks pointer, rcu_preempt_offline_tasks() can do a better job
of setting the root rcu_node's ->boost_tasks pointer.
The key point is that rcu_preempt_offline_tasks() runs as part of the
CPU-hotplug process, so that a concurrent synchronize_rcu_expedited()
is guaranteed to either have not started on the one hand (in which case
there is no boosting on behalf of the expedited grace period) or to be
completely initialized on the other (in which case, in the absence of
other priority boosting, all ->boost_tasks pointers will be initialized).
Therefore, if rcu_preempt_offline_tasks() finds that the ->boost_tasks
pointer is equal to the ->exp_tasks pointer, it can be sure that it is
correctly placed.
In the case where there was boosting ongoing at the time that the
synchronize_rcu_expedited() function started, different nodes might start
boosting the tasks blocking the expedited grace period at different times.
In this mixed case, the root node will either be boosting tasks for
the expedited grace period already, or it will start as soon as it gets
done boosting for the normal grace period -- but in this latter case,
the root node's tasks needed to be boosted in any case.
This commit therefore adds a check of the ->boost_tasks pointer against
the ->exp_tasks pointer to the list that prevents updating ->boost_tasks.
Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Diffstat (limited to 'kernel/rcutree_plugin.h')
-rw-r--r-- | kernel/rcutree_plugin.h | 9 |
1 files changed, 7 insertions, 2 deletions
diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index b1b485111321..15d28febbbd4 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h | |||
@@ -588,10 +588,15 @@ static int rcu_preempt_offline_tasks(struct rcu_state *rsp, | |||
588 | rnp->exp_tasks = NULL; | 588 | rnp->exp_tasks = NULL; |
589 | #ifdef CONFIG_RCU_BOOST | 589 | #ifdef CONFIG_RCU_BOOST |
590 | rnp->boost_tasks = NULL; | 590 | rnp->boost_tasks = NULL; |
591 | /* In case root is being boosted and leaf was not. */ | 591 | /* |
592 | * In case root is being boosted and leaf was not. Make sure | ||
593 | * that we boost the tasks blocking the current grace period | ||
594 | * in this case. | ||
595 | */ | ||
592 | raw_spin_lock(&rnp_root->lock); /* irqs already disabled */ | 596 | raw_spin_lock(&rnp_root->lock); /* irqs already disabled */ |
593 | if (rnp_root->boost_tasks != NULL && | 597 | if (rnp_root->boost_tasks != NULL && |
594 | rnp_root->boost_tasks != rnp_root->gp_tasks) | 598 | rnp_root->boost_tasks != rnp_root->gp_tasks && |
599 | rnp_root->boost_tasks != rnp_root->exp_tasks) | ||
595 | rnp_root->boost_tasks = rnp_root->gp_tasks; | 600 | rnp_root->boost_tasks = rnp_root->gp_tasks; |
596 | raw_spin_unlock(&rnp_root->lock); /* irqs still disabled */ | 601 | raw_spin_unlock(&rnp_root->lock); /* irqs still disabled */ |
597 | #endif /* #ifdef CONFIG_RCU_BOOST */ | 602 | #endif /* #ifdef CONFIG_RCU_BOOST */ |