All of lore.kernel.org
 help / color / mirror / Atom feed
From: Chris Metcalf <cmetcalf@ezchip.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: Chris Metcalf <cmetcalf@ezchip.com>,
	"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>,
	Manfred Spraul <manfred@colorfullife.com>,
	"Oleg Nesterov" <oleg@redhat.com>,
	Kirill Tkhai <ktkhai@parallels.com>,
	<linux-kernel@vger.kernel.org>, Ingo Molnar <mingo@redhat.com>,
	"Josh Poimboeuf" <jpoimboe@redhat.com>
Subject: [PATCH 1/2] tile: modify arch_spin_unlock_wait() semantics
Date: Tue, 28 Apr 2015 13:33:56 -0400	[thread overview]
Message-ID: <1430242437-19938-1-git-send-email-cmetcalf@ezchip.com> (raw)
In-Reply-To: <20150428162458.GI5029@twins.programming.kicks-ass.net>

Rather than trying to wait until all possible lockers have
unlocked the lock, we now only wait until the current locker
(if any) has released the lock.

The old code was correct, but the new code works more like the x86
code and thus hopefully is more appropriate under contention.
See commit 78bff1c8684f ("x86/ticketlock: Fix spin_unlock_wait()
livelock") for x86.

Signed-off-by: Chris Metcalf <cmetcalf@ezchip.com>
---
 arch/tile/lib/spinlock_32.c | 11 ++++++++++-
 arch/tile/lib/spinlock_64.c | 11 ++++++++++-
 2 files changed, 20 insertions(+), 2 deletions(-)

diff --git a/arch/tile/lib/spinlock_32.c b/arch/tile/lib/spinlock_32.c
index b34f79aada48..c3df3c237496 100644
--- a/arch/tile/lib/spinlock_32.c
+++ b/arch/tile/lib/spinlock_32.c
@@ -65,8 +65,17 @@ EXPORT_SYMBOL(arch_spin_trylock);
 void arch_spin_unlock_wait(arch_spinlock_t *lock)
 {
 	u32 iterations = 0;
-	while (arch_spin_is_locked(lock))
+	int current = READ_ONCE(lock->current_ticket);
+	int next = READ_ONCE(lock->next_ticket);
+
+	/* Return immediately if unlocked. */
+	if (next == current)
+		return;
+
+	/* Wait until the current locker has released the lock. */
+	do {
 		delay_backoff(iterations++);
+	} while (READ_ONCE(lock->current_ticket) == current)
 }
 EXPORT_SYMBOL(arch_spin_unlock_wait);
 
diff --git a/arch/tile/lib/spinlock_64.c b/arch/tile/lib/spinlock_64.c
index d6fb9581e980..aaf12813d4db 100644
--- a/arch/tile/lib/spinlock_64.c
+++ b/arch/tile/lib/spinlock_64.c
@@ -65,8 +65,17 @@ EXPORT_SYMBOL(arch_spin_trylock);
 void arch_spin_unlock_wait(arch_spinlock_t *lock)
 {
 	u32 iterations = 0;
-	while (arch_spin_is_locked(lock))
+	u32 val = READ_ONCE(lock->lock);
+	u32 current = arch_spin_current(val);
+
+	/* Return immediately if unlocked. */
+	if (arch_spin_next(val) == current)
+		return;
+
+	/* Wait until the current locker has released the lock. */
+	do {
 		delay_backoff(iterations++);
+	} while (arch_spin_current(READ_ONCE(lock->lock)) == current);
 }
 EXPORT_SYMBOL(arch_spin_unlock_wait);
 
-- 
2.1.2


  parent reply	other threads:[~2015-04-28 17:34 UTC|newest]

Thread overview: 46+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <20150217104516.12144.85911.stgit@tkhai>
2015-02-17 10:47 ` [PATCH 2/2] [PATCH] sched: Add smp_rmb() in task rq locking cycles Kirill Tkhai
2015-02-17 12:12   ` Peter Zijlstra
2015-02-17 12:36     ` Kirill Tkhai
2015-02-17 12:45       ` Peter Zijlstra
2015-02-17 13:05     ` Peter Zijlstra
2015-02-17 16:05       ` Paul E. McKenney
2015-02-17 18:01         ` Paul E. McKenney
2015-02-17 18:23           ` Peter Zijlstra
2015-02-17 21:45             ` Paul E. McKenney
2015-02-18 13:41               ` Peter Zijlstra
2015-02-17 18:36         ` Peter Zijlstra
2015-02-17 21:52           ` Paul E. McKenney
2015-02-18 13:47             ` Peter Zijlstra
2015-02-18 18:43               ` Paul E. McKenney
2015-02-18 15:53             ` Oleg Nesterov
2015-02-18 16:11               ` Peter Zijlstra
2015-02-18 16:32                 ` Oleg Nesterov
2015-02-18 19:23                   ` Paul E. McKenney
2015-02-18 15:59             ` Oleg Nesterov
2015-02-18 19:14               ` Manfred Spraul
2015-02-18 22:43                 ` Peter Zijlstra
2015-02-19 14:19                   ` Oleg Nesterov
2015-02-20 18:28                     ` Manfred Spraul
2015-02-20 18:45                       ` Peter Zijlstra
2015-02-20 20:23                         ` Oleg Nesterov
2015-02-21 12:54                           ` Peter Zijlstra
2015-04-25 19:56                         ` Paul E. McKenney
2015-04-26 10:52                           ` Paul E. McKenney
2015-04-28 14:33                             ` Peter Zijlstra
2015-04-28 15:53                               ` Chris Metcalf
2015-04-28 16:24                                 ` Peter Zijlstra
2015-04-28 16:44                                   ` [PATCH] spinlock: clarify doc for raw_spin_unlock_wait() Chris Metcalf
2015-04-29 17:34                                     ` Manfred Spraul
2015-04-28 17:33                                   ` Chris Metcalf [this message]
2015-04-28 17:33                                     ` [PATCH 2/2] tile: use READ_ONCE() in arch_spin_is_locked() Chris Metcalf
2015-04-28 16:40                                 ` [PATCH 2/2] [PATCH] sched: Add smp_rmb() in task rq locking cycles Peter Zijlstra
2015-04-28 16:58                                   ` Chris Metcalf
2015-04-28 17:43                                     ` Peter Zijlstra
2015-04-28 18:00                                       ` Chris Metcalf
2015-04-28 18:24                                         ` Peter Zijlstra
2015-04-28 18:38                                           ` Chris Metcalf
2015-04-28 14:32                           ` Peter Zijlstra
2015-04-28 20:33                             ` Paul E. McKenney
2015-02-21  3:26                       ` Paul E. McKenney
2015-02-23 18:29                         ` Paul E. McKenney
2015-02-18 17:05     ` [tip:sched/core] sched: Clarify ordering between task_rq_lock() and move_queued_task() tip-bot for Peter Zijlstra

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=1430242437-19938-1-git-send-email-cmetcalf@ezchip.com \
    --to=cmetcalf@ezchip.com \
    --cc=jpoimboe@redhat.com \
    --cc=ktkhai@parallels.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=manfred@colorfullife.com \
    --cc=mingo@redhat.com \
    --cc=oleg@redhat.com \
    --cc=paulmck@linux.vnet.ibm.com \
    --cc=peterz@infradead.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.