The commit is pushed to "branch-rh7-3.10.0-229.7.2-ovz" and will appear at https://src.openvz.org/scm/ovz/vzkernel.git after rh7-3.10.0-229.7.2.vz7.6.3 ------> commit 932bf29b63b1e7c74669a8847d7c69cc8b8ba919 Author: Vladimir Davydov <vdavy...@parallels.com> Date: Fri Aug 28 14:49:25 2015 +0400
ms/percpu-refcount: use RCU-sched insted of normal RCU Patchset description: Pulling upstream patches converting css refcnt to percpu_ref. https://jira.sw.ru/browse/PSBM-34174 Kent Overstreet (2): percpu: implement generic percpu refcounting percpu-refcount: Don't use silly cmpxchg() Tejun Heo (9): percpu-refcount: consistently use plain (non-sched) RCU percpu-refcount: cosmetic updates percpu-refcount: add __must_check to percpu_ref_init() and don't use ACCESS_ONCE() in percpu_ref_kill_rcu() percpu-refcount: implement percpu_ref_cancel_init() percpu-refcount: implement percpu_tryget() along with percpu_ref_kill_and_confirm() percpu-refcount: use RCU-sched insted of normal RCU cgroup: reorder the operations in cgroup_destroy_locked() cgroup: split cgroup destruction into two steps cgroup: use percpu refcnt for cgroup_subsys_states === This patch description: From: Tejun Heo <t...@kernel.org> percpu-refcount was incorrectly using preempt_disable/enable() for RCU critical sections against call_rcu(). 6a24474da8 ("percpu-refcount: consistently use plain (non-sched) RCU") fixed it by converting the preepmtion operations with rcu_read_[un]lock() citing that there isn't any advantage in using sched-RCU over using the usual one; however, rcu_read_[un]lock() for the preemptible RCU implementation - CONFIG_TREE_PREEMPT_RCU, chosen when CONFIG_PREEMPT - are slightly more expensive than preempt_disable/enable(). In a contrived microbench which repeats the followings, - percpu_ref_get() - copy 32 bytes of data into percpu buffer - percpu_put_get() - copy 32 bytes of data into percpu buffer rcu_read_[un]lock() used in percpu_ref_get/put() makes it go slower by about 15% when compared to using sched-RCU. As the RCU critical sections are extremely short, using sched-RCU shouldn't have any latency implications. Convert to RCU-sched. Signed-off-by: Tejun Heo <t...@kernel.org> Acked-by: Kent Overstreet <koverstr...@google.com> Acked-by: "Paul E. McKenney" <paul...@linux.vnet.ibm.com> Cc: Michal Hocko <mho...@suse.cz> Cc: Rusty Russell <ru...@rustcorp.com.au> (cherry picked from commit a4244454df1296e90cc961c1b636b1176ef0d9a0) Signed-off-by: Vladimir Davydov <vdavy...@parallels.com> --- include/linux/percpu-refcount.h | 12 ++++++------ lib/percpu-refcount.c | 2 +- 2 files changed, 7 insertions(+), 7 deletions(-) diff --git a/include/linux/percpu-refcount.h b/include/linux/percpu-refcount.h index dd2a086..95961f0 100644 --- a/include/linux/percpu-refcount.h +++ b/include/linux/percpu-refcount.h @@ -105,7 +105,7 @@ static inline void percpu_ref_get(struct percpu_ref *ref) { unsigned __percpu *pcpu_count; - rcu_read_lock(); + rcu_read_lock_sched(); pcpu_count = ACCESS_ONCE(ref->pcpu_count); @@ -114,7 +114,7 @@ static inline void percpu_ref_get(struct percpu_ref *ref) else atomic_inc(&ref->count); - rcu_read_unlock(); + rcu_read_unlock_sched(); } /** @@ -134,7 +134,7 @@ static inline bool percpu_ref_tryget(struct percpu_ref *ref) unsigned __percpu *pcpu_count; int ret = false; - rcu_read_lock(); + rcu_read_lock_sched(); pcpu_count = ACCESS_ONCE(ref->pcpu_count); @@ -143,7 +143,7 @@ static inline bool percpu_ref_tryget(struct percpu_ref *ref) ret = true; } - rcu_read_unlock(); + rcu_read_unlock_sched(); return ret; } @@ -159,7 +159,7 @@ static inline void percpu_ref_put(struct percpu_ref *ref) { unsigned __percpu *pcpu_count; - rcu_read_lock(); + rcu_read_lock_sched(); pcpu_count = ACCESS_ONCE(ref->pcpu_count); @@ -168,7 +168,7 @@ static inline void percpu_ref_put(struct percpu_ref *ref) else if (unlikely(atomic_dec_and_test(&ref->count))) ref->release(ref); - rcu_read_unlock(); + rcu_read_unlock_sched(); } #endif diff --git a/lib/percpu-refcount.c b/lib/percpu-refcount.c index 8bf9e71..7deeb62 100644 --- a/lib/percpu-refcount.c +++ b/lib/percpu-refcount.c @@ -154,5 +154,5 @@ void percpu_ref_kill_and_confirm(struct percpu_ref *ref, (((unsigned long) ref->pcpu_count)|PCPU_REF_DEAD); ref->confirm_kill = confirm_kill; - call_rcu(&ref->rcu, percpu_ref_kill_rcu); + call_rcu_sched(&ref->rcu, percpu_ref_kill_rcu); } _______________________________________________ Devel mailing list Devel@openvz.org https://lists.openvz.org/mailman/listinfo/devel