linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Jens Axboe <jens.axboe@oracle.com>
To: "Luck, Tony" <tony.luck@intel.com>
Cc: Matthew Wilcox <matthew@wil.cx>,
	linux-arch@vger.kernel.org, linux-mm@kvack.org,
	linux-kernel@vger.kernel.org
Subject: Re: What if a TLB flush needed to sleep?
Date: Thu, 27 Mar 2008 09:09:26 +0100	[thread overview]
Message-ID: <20080327080926.GE12346@kernel.dk> (raw)
In-Reply-To: <1FE6DD409037234FAB833C420AA843ECE9EB1C@orsmsx424.amr.corp.intel.com>

On Wed, Mar 26 2008, Luck, Tony wrote:
> > 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);
> +}

This doesn't look very nice - if down_spin() is called with interrupts
disabled, __down_spin() enables them.

-- 
Jens Axboe

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  reply	other threads:[~2008-03-27  8:09 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-03-25 20:49 Luck, Tony
2008-03-25 21:47 ` Alan Cox
2008-03-26 12:32 ` Matthew Wilcox
2008-03-26 20:29   ` Luck, Tony
2008-03-27  8:09     ` Jens Axboe [this message]
2008-03-27 14:15     ` down_spin() implementation Matthew Wilcox
2008-03-28  0:01       ` Nick Piggin
2008-03-28 12:45         ` Matthew Wilcox
2008-03-28 21:16           ` Luck, Tony
2008-03-28 23:48             ` Arnd Bergmann
     [not found]           ` <20080328124517.GQ16721-6jwH94ZQLHl74goWV3ctuw@public.gmane.org>
2008-03-29  1:04             ` Nick Piggin
2008-03-28  4:51       ` Stephen Rothwell
2008-03-28  5:03         ` Nick Piggin
2008-03-28 12:46           ` Matthew Wilcox
2008-03-28 12:51       ` Jens Axboe
2008-03-28 13:17         ` Matthew Wilcox
2008-03-28 13:24           ` Jens Axboe
2008-03-26 19:25 ` What if a TLB flush needed to sleep? Christoph Lameter
2008-03-26 20:29   ` Thomas Gleixner
2008-03-27  1:19     ` Christoph Lameter
2008-03-27 13:20       ` Peter Zijlstra
2008-03-27 18:44         ` Christoph Lameter
2008-03-28  9:59           ` 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=20080327080926.GE12346@kernel.dk \
    --to=jens.axboe@oracle.com \
    --cc=linux-arch@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=matthew@wil.cx \
    --cc=tony.luck@intel.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