Messages in this thread Patch in this message | | | From | Waiman Long <> | Subject | [PATCH tip/locking/core v10 3/7] locking/qspinlock: Avoid redundant read of next pointer | Date | Mon, 9 Nov 2015 19:09:23 -0500 |
| |
With optimistic prefetch of the next node cacheline, the next pointer may have been properly inititalized. As a result, the reading of node->next in the contended path may be redundant. This patch eliminates the redundant read if the next pointer value is not NULL.
Signed-off-by: Waiman Long <Waiman.Long@hpe.com> --- kernel/locking/qspinlock.c | 9 ++++++--- 1 files changed, 6 insertions(+), 3 deletions(-)
diff --git a/kernel/locking/qspinlock.c b/kernel/locking/qspinlock.c index 365b203..9862078 100644 --- a/kernel/locking/qspinlock.c +++ b/kernel/locking/qspinlock.c @@ -396,6 +396,7 @@ queue: * p,*,* -> n,*,* */ old = xchg_tail(lock, tail); + next = NULL; /* * if there was a previous node; link it and wait until reaching the @@ -463,10 +464,12 @@ queue: } /* - * contended path; wait for next, release. + * contended path; wait for next if not observed yet, release. */ - while (!(next = READ_ONCE(node->next))) - cpu_relax(); + if (!next) { + while (!(next = READ_ONCE(node->next))) + cpu_relax(); + } arch_mcs_spin_unlock_contended(&next->locked); pv_kick_node(lock, next); -- 1.7.1
| |