linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Davidlohr Bueso <dave@stgolabs.net>
To: Peter Zijlstra <peterz@infradead.org>, Ingo Molnar <mingo@kernel.org>
Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>,
	Jason Low <jason.low2@hp.com>,
	Michel Lespinasse <walken@google.com>,
	Tim Chen <tim.c.chen@linux.intel.com>,
	linux-kernel@vger.kernel.org, Davidlohr Bueso <dave@stgolabs.net>,
	Davidlohr Bueso <dbueso@suse.de>
Subject: [PATCH 3/5] locking/rwsem: Set lock ownership ASAP
Date: Fri, 30 Jan 2015 01:14:25 -0800	[thread overview]
Message-ID: <1422609267-15102-4-git-send-email-dave@stgolabs.net> (raw)
In-Reply-To: <1422609267-15102-1-git-send-email-dave@stgolabs.net>

In order to optimize the spinning step, we need to set the lock
owner as soon as the lock is acquired; after a successful counter
cmpxchg operation, that is. This is particularly useful as rwsems
need to set the owner to nil for readers, so there is a greater
chance of falling out of the spinning. Currently we only set the
owner much later in the game, in the more generic level -- latency
can be specially bad when waiting for a node->next pointer when
releasing the osq in up_write calls.

As such, update the owner inside rwsem_try_write_lock (when the
lock is obtained after blocking) and rwsem_try_write_lock_unqueued
(when the lock is obtained while spinning). This requires creating
a new internal rwsem.h header to share the owner related calls.

Also cleanup some headers for mutex and rwsem.

Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Davidlohr Bueso <dbueso@suse.de>
---
 kernel/locking/mutex.c      |  2 +-
 kernel/locking/rwsem-xadd.c |  8 ++++++--
 kernel/locking/rwsem.c      | 22 +---------------------
 kernel/locking/rwsem.h      | 20 ++++++++++++++++++++
 4 files changed, 28 insertions(+), 24 deletions(-)
 create mode 100644 kernel/locking/rwsem.h

diff --git a/kernel/locking/mutex.c b/kernel/locking/mutex.c
index c67a60b..166355e 100644
--- a/kernel/locking/mutex.c
+++ b/kernel/locking/mutex.c
@@ -25,7 +25,7 @@
 #include <linux/spinlock.h>
 #include <linux/interrupt.h>
 #include <linux/debug_locks.h>
-#include "mcs_spinlock.h"
+#include <linux/osq_lock.h>
 
 /*
  * In the DEBUG case we are using the "NULL fastpath" for mutexes,
diff --git a/kernel/locking/rwsem-xadd.c b/kernel/locking/rwsem-xadd.c
index 82aba46..07713e5 100644
--- a/kernel/locking/rwsem-xadd.c
+++ b/kernel/locking/rwsem-xadd.c
@@ -14,8 +14,9 @@
 #include <linux/init.h>
 #include <linux/export.h>
 #include <linux/sched/rt.h>
+#include <linux/osq_lock.h>
 
-#include "mcs_spinlock.h"
+#include "rwsem.h"
 
 /*
  * Guide to the rw_semaphore's count field for common values.
@@ -265,6 +266,7 @@ static inline bool rwsem_try_write_lock(long count, struct rw_semaphore *sem)
 		    RWSEM_ACTIVE_WRITE_BIAS) == RWSEM_WAITING_BIAS) {
 		if (!list_is_singular(&sem->wait_list))
 			rwsem_atomic_update(RWSEM_WAITING_BIAS, sem);
+		rwsem_set_owner(sem);
 		return true;
 	}
 
@@ -284,8 +286,10 @@ static inline bool rwsem_try_write_lock_unqueued(struct rw_semaphore *sem)
 			return false;
 
 		old = cmpxchg(&sem->count, count, count + RWSEM_ACTIVE_WRITE_BIAS);
-		if (old == count)
+		if (old == count) {
+			rwsem_set_owner(sem);
 			return true;
+		}
 
 		count = old;
 	}
diff --git a/kernel/locking/rwsem.c b/kernel/locking/rwsem.c
index e2d3bc7..205be0c 100644
--- a/kernel/locking/rwsem.c
+++ b/kernel/locking/rwsem.c
@@ -9,29 +9,9 @@
 #include <linux/sched.h>
 #include <linux/export.h>
 #include <linux/rwsem.h>
-
 #include <linux/atomic.h>
 
-#ifdef CONFIG_RWSEM_SPIN_ON_OWNER
-static inline void rwsem_set_owner(struct rw_semaphore *sem)
-{
-	sem->owner = current;
-}
-
-static inline void rwsem_clear_owner(struct rw_semaphore *sem)
-{
-	sem->owner = NULL;
-}
-
-#else
-static inline void rwsem_set_owner(struct rw_semaphore *sem)
-{
-}
-
-static inline void rwsem_clear_owner(struct rw_semaphore *sem)
-{
-}
-#endif
+#include "rwsem.h"
 
 /*
  * lock for reading
diff --git a/kernel/locking/rwsem.h b/kernel/locking/rwsem.h
new file mode 100644
index 0000000..870ed9a
--- /dev/null
+++ b/kernel/locking/rwsem.h
@@ -0,0 +1,20 @@
+#ifdef CONFIG_RWSEM_SPIN_ON_OWNER
+static inline void rwsem_set_owner(struct rw_semaphore *sem)
+{
+	sem->owner = current;
+}
+
+static inline void rwsem_clear_owner(struct rw_semaphore *sem)
+{
+	sem->owner = NULL;
+}
+
+#else
+static inline void rwsem_set_owner(struct rw_semaphore *sem)
+{
+}
+
+static inline void rwsem_clear_owner(struct rw_semaphore *sem)
+{
+}
+#endif
-- 
2.1.4


  parent reply	other threads:[~2015-01-30  9:15 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-01-30  9:14 [PATCH -tip v2 0/5] rwsem: Fine tuning Davidlohr Bueso
2015-01-30  9:14 ` [PATCH 1/5] locking/rwsem: Use task->state helpers Davidlohr Bueso
2015-01-30  9:14 ` [PATCH 2/5] locking/rwsem: Document barrier need when waking tasks Davidlohr Bueso
2015-02-18 17:11   ` [tip:locking/core] " tip-bot for Davidlohr Bueso
2015-01-30  9:14 ` Davidlohr Bueso [this message]
2015-02-18 17:11   ` [tip:locking/core] locking/rwsem: Set lock ownership ASAP tip-bot for Davidlohr Bueso
2015-01-30  9:14 ` [PATCH 4/5] locking/rwsem: Avoid deceiving lock spinners Davidlohr Bueso
2015-01-31  1:51   ` Tim Chen
2015-01-31  2:28     ` Davidlohr Bueso
2015-02-03 17:16       ` Tim Chen
2015-02-03 17:54         ` Jason Low
2015-02-03 19:43           ` Tim Chen
2015-02-03 21:04             ` Jason Low
2015-02-03 21:48               ` Tim Chen
2015-02-04 12:06             ` Peter Zijlstra
2015-02-04 17:39               ` Tim Chen
2015-01-31  9:29   ` Peter Zijlstra
2015-01-31 21:14     ` Davidlohr Bueso
2015-01-31 21:17       ` Davidlohr Bueso
2015-02-18 17:12   ` [tip:locking/core] " tip-bot for Davidlohr Bueso
2015-01-30  9:14 ` [PATCH 5/5] locking/rwsem: Check for active lock before bailing on spinning Davidlohr Bueso
2015-02-18 17:12   ` [tip:locking/core] " tip-bot for Davidlohr Bueso

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1422609267-15102-4-git-send-email-dave@stgolabs.net \
    --to=dave@stgolabs.net \
    --cc=dbueso@suse.de \
    --cc=jason.low2@hp.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@kernel.org \
    --cc=paulmck@linux.vnet.ibm.com \
    --cc=peterz@infradead.org \
    --cc=tim.c.chen@linux.intel.com \
    --cc=walken@google.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).