login
Header Space

 
 

RE: What if a TLB flush needed to sleep?

Score:
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
To: Matthew Wilcox <matthew@...>
Cc: <linux-arch@...>, <linux-mm@...>, <linux-kernel@...>
Date: Wednesday, March 26, 2008 - 4:29 pm

> Of course, someone who wrote it could do better ;-)

Here is Willy's code in patch format (against linux-next
tree tag next-20080326 which includes his re-write of the
semaphore code).

This looks a lot cleaner than my ia64 specific code that
used cmpxchg() for the down() operation and fetchadd for
the up() ... using a brand new semaphore_spin data type.

It appears to work ... I tried to do some timing comparisons
of this generic version against my arch specific one, but the
hackbench test case has a run to run variation of a factor of
three (from 1min9sec to 3min44sec) so it is hopeless to try
and see some small percentage difference.

commit 0359fbb64297d44328f26ec5fda3a3c26f1c5ba7
Author: Matthew Wilcox <matthew@wil.cx>
Date:   Wed Mar 26 11:08:18 2008 -0700

    Add "down_spin()" API for semaphores
    
    For those places that need semaphore semantics but cannot sleep
    
    Signed-off-by: Tony Luck <tony.luck@intel.com>

diff --git a/include/linux/semaphore.h b/include/linux/semaphore.h
index a7125da..3404ce5 100644
--- a/include/linux/semaphore.h
+++ b/include/linux/semaphore.h
@@ -53,6 +53,12 @@ static inline void sema_init(struct semaphore *sem, int val)
 extern void down(struct semaphore *sem);
 
 /*
+ * Attempt to acquire the semaphore.  If another task is already holding the
+ * semaphore, spin until the semaphore is released.
+ */
+extern void down_spin(struct semaphore *sem);
+
+/*
  * As down(), except the sleep may be interrupted by a signal.  If it is,
  * this function will return -EINTR.
  */
diff --git a/kernel/semaphore.c b/kernel/semaphore.c
index bef977b..d3eb559 100644
--- a/kernel/semaphore.c
+++ b/kernel/semaphore.c
@@ -23,6 +23,7 @@
  */
 
 static noinline void __down(struct semaphore *sem);
+static noinline void __down_spin(struct semaphore *sem);
 static noinline int __down_interruptible(struct semaphore *sem);
 static noinline int __down_killable(struct semaphore *sem);
 static noinline int __down_timeout(struct semaphore *sem, long jiffies);
@@ -41,6 +42,21 @@ void down(struct semaphore *sem)
 }
 EXPORT_SYMBOL(down);
 
+void down_spin(struct semaphore *sem)
+{
+	unsigned long flags;
+	int count;
+
+	spin_lock_irqsave(&sem->lock, flags);
+	count = sem->count - 1;
+	if (likely(count >= 0))
+		sem->count = count;
+	else
+		__down_spin(sem);
+	spin_unlock_irqrestore(&sem->lock, flags);
+}
+EXPORT_SYMBOL(down_spin);
+
 int down_interruptible(struct semaphore *sem)
 {
 	unsigned long flags;
@@ -197,6 +213,20 @@ static noinline int __sched __down_timeout(struct semaphore *sem, long jiffies)
 	return __down_common(sem, TASK_UNINTERRUPTIBLE, jiffies);
 }
 
+static noinline void __sched __down_spin(struct semaphore *sem)
+{
+	struct semaphore_waiter waiter;
+
+	list_add_tail(&waiter.list, &sem->wait_list);
+	waiter.task = current;
+	waiter.up = 0;
+
+	spin_unlock_irq(&sem->lock);
+	while (!waiter.up)
+		cpu_relax();
+	spin_lock_irq(&sem->lock);
+}
+
 static noinline void __sched __up(struct semaphore *sem)
 {
 	struct semaphore_waiter *waiter = list_first_entry(&sem->wait_list,
--
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:
What if a TLB flush needed to sleep?, Luck, Tony, (Tue Mar 25, 4:49 pm)
Re: What if a TLB flush needed to sleep?, Christoph Lameter, (Wed Mar 26, 3:25 pm)
Re: What if a TLB flush needed to sleep?, Thomas Gleixner, (Wed Mar 26, 4:29 pm)
Re: What if a TLB flush needed to sleep?, Christoph Lameter, (Wed Mar 26, 9:19 pm)
Re: What if a TLB flush needed to sleep?, Peter Zijlstra, (Thu Mar 27, 9:20 am)
Re: What if a TLB flush needed to sleep?, Christoph Lameter, (Thu Mar 27, 2:44 pm)
Re: What if a TLB flush needed to sleep?, Peter Zijlstra, (Fri Mar 28, 5:59 am)
Re: What if a TLB flush needed to sleep?, Matthew Wilcox, (Wed Mar 26, 8:32 am)
RE: What if a TLB flush needed to sleep?, Luck, Tony, (Wed Mar 26, 4:29 pm)
down_spin() implementation, Matthew Wilcox, (Thu Mar 27, 10:15 am)
Re: down_spin() implementation, Jens Axboe, (Fri Mar 28, 8:51 am)
Re: down_spin() implementation, Matthew Wilcox, (Fri Mar 28, 9:17 am)
Re: down_spin() implementation, Jens Axboe, (Fri Mar 28, 9:24 am)
Re: down_spin() implementation, Nick Piggin, (Thu Mar 27, 8:01 pm)
Re: down_spin() implementation, Matthew Wilcox, (Fri Mar 28, 8:45 am)
Re: down_spin() implementation, Nick Piggin, (Fri Mar 28, 9:04 pm)
RE: down_spin() implementation, Luck, Tony, (Fri Mar 28, 5:16 pm)
Re: down_spin() implementation, Arnd Bergmann, (Fri Mar 28, 7:48 pm)
Re: down_spin() implementation, Stephen Rothwell, (Fri Mar 28, 12:51 am)
Re: down_spin() implementation, Nick Piggin, (Fri Mar 28, 1:03 am)
Re: down_spin() implementation, Matthew Wilcox, (Fri Mar 28, 8:46 am)
Re: What if a TLB flush needed to sleep?, Jens Axboe, (Thu Mar 27, 4:09 am)
Re: What if a TLB flush needed to sleep?, Alan Cox, (Tue Mar 25, 5:47 pm)
speck-geostationary