diff options
| author | Paul E. McKenney <paulmck@linux.vnet.ibm.com> | 2018-04-14 10:40:57 -0700 |
|---|---|---|
| committer | Paul E. McKenney <paulmck@linux.vnet.ibm.com> | 2018-05-15 10:29:33 -0700 |
| commit | 51af970d19f395fc57b82514022126de6c5420cd (patch) | |
| tree | 20637fb321eddd24f16483bb1ac44b9f8eb5eb57 /kernel/rcu/tree.h | |
| parent | fb31340f8a43a6f2e871164822ef4979b36232ae (diff) | |
rcu: Avoid losing ->need_future_gp[] values due to GP start/end races
The rcu_cbs_completed() function provides the value of ->completed
at which new callbacks can safely be invoked. This is recorded in
two-element ->need_future_gp[] arrays in the rcu_node structure, and
the elements of these arrays corresponding to the just-completed grace
period are zeroed at the end of that grace period. However, the
rcu_cbs_completed() function can return the current ->completed value
plus either one or two, so it is possible for the corresponding
->need_future_gp[] entry to be cleared just after it was set, thus
losing a request for a future grace period.
This commit avoids this race by expanding ->need_future_gp[] to four
elements.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Tested-by: Nicholas Piggin <npiggin@gmail.com>
Diffstat (limited to 'kernel/rcu/tree.h')
| -rw-r--r-- | kernel/rcu/tree.h | 3 |
1 files changed, 1 insertions, 2 deletions
diff --git a/kernel/rcu/tree.h b/kernel/rcu/tree.h index bd1103763551..952cd0c223fe 100644 --- a/kernel/rcu/tree.h +++ b/kernel/rcu/tree.h @@ -150,8 +150,7 @@ struct rcu_node { struct swait_queue_head nocb_gp_wq[2]; /* Place for rcu_nocb_kthread() to wait GP. */ #endif /* #ifdef CONFIG_RCU_NOCB_CPU */ - int need_future_gp[2]; - /* Counts of upcoming no-CB GP requests. */ + int need_future_gp[4]; /* Counts of upcoming no-CB GP requests. */ raw_spinlock_t fqslock ____cacheline_internodealigned_in_smp; spinlock_t exp_lock ____cacheline_internodealigned_in_smp; |