public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Daniel Wagner <daniel.wagner@bmw-carit.de>
To: Peter Zijlstra <peterz@infradead.org>
Cc: <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v1 3/8] sched/completion: convert completions to use simple wait queues
Date: Mon, 12 Oct 2015 12:03:06 +0200	[thread overview]
Message-ID: <561B855A.2060403@bmw-carit.de> (raw)
In-Reply-To: <561B7A9A.3020904@bmw-carit.de>

On 10/12/2015 11:17 AM, Daniel Wagner wrote:
> On 09/09/2015 04:26 PM, Peter Zijlstra wrote:
>> On Wed, Sep 09, 2015 at 02:05:29PM +0200, Daniel Wagner wrote:
>>> @@ -50,10 +50,10 @@ void complete_all(struct completion *x)
>>>  {
>>>  	unsigned long flags;
>>>  
>>> -	spin_lock_irqsave(&x->wait.lock, flags);
>>> +	raw_spin_lock_irqsave(&x->wait.lock, flags);
>>>  	x->done += UINT_MAX/2;
>>> -	__wake_up_locked(&x->wait, TASK_NORMAL, 0);
>>> -	spin_unlock_irqrestore(&x->wait.lock, flags);
>>> +	swake_up_locked(&x->wait);
>>> +	raw_spin_unlock_irqrestore(&x->wait.lock, flags);
>>>  }
>>>  EXPORT_SYMBOL(complete_all);
>>
>> I don't think that's correct; __wake_up_locked(.nr=0) would wake all
>> waiters, where swake_up_locked() will only wake one.
> 
> I read that x->done should be protected via wait.lock during the whole
> operation. swake_up_all() will release and reacquire the lock while
> processing the all waiters. So we need to get
> 
> Could we play a trick like setting the highest bit in done for
> indicating the complete_all() operation. The UINT_MAX/2 update looks
> like do this by setting a value which has the biggest offset from 0 (but
> why adding instead of just going for assigning...).


I had something like this here in mind:

diff --git a/kernel/sched/completion.c b/kernel/sched/completion.c
index b020159..c03d4de 100644
--- a/kernel/sched/completion.c
+++ b/kernel/sched/completion.c
@@ -14,6 +14,9 @@
 #include <linux/sched.h>
 #include <linux/completion.h>

+#define COMPLETION_DONE_ALL (1UL << 31)
+#define COMPLETION_DONE_MASK (COMPLETION_DONE_ALL - 1)
+
 /**
  * complete: - signals a single thread waiting on this completion
  * @x:  holds the state of this particular completion
@@ -31,7 +34,7 @@ void complete(struct completion *x)
 	unsigned long flags;

 	raw_spin_lock_irqsave(&x->wait.lock, flags);
-	x->done++;
+	x->done = (x->done + 1) & COMPLETION_DONE_MASK;
 	swake_up_locked(&x->wait);
 	raw_spin_unlock_irqrestore(&x->wait.lock, flags);
 }
@@ -51,9 +54,9 @@ void complete_all(struct completion *x)
 	unsigned long flags;

 	raw_spin_lock_irqsave(&x->wait.lock, flags);
-	x->done += UINT_MAX/2;
-	swake_up_locked(&x->wait);
+	x->done |= COMPLETION_DONE_ALL;
 	raw_spin_unlock_irqrestore(&x->wait.lock, flags);
+	swake_up_all(&x->wait);
 }
 EXPORT_SYMBOL(complete_all);

@@ -79,7 +82,7 @@ do_wait_for_common(struct completion *x,
 		if (!x->done)
 			return timeout;
 	}
-	x->done--;
+	x->done = (x->done - 1) & COMPLETION_DONE_MASK;
 	return timeout ?: 1;
 }

@@ -281,7 +284,7 @@ bool try_wait_for_completion(struct completion *x)
 	if (!x->done)
 		ret = 0;
 	else
-		x->done--;
+		x->done = (x->done - 1) & COMPLETION_DONE_MASK;
 	raw_spin_unlock_irqrestore(&x->wait.lock, flags);
 	return ret;
 }


  reply	other threads:[~2015-10-12 10:03 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-09-09 12:05 [PATCH v0 0/8] Simple wait queue support Daniel Wagner
2015-09-09 12:05 ` [PATCH v1 1/8] wait.[ch]: Introduce the simple waitqueue (swait) implementation Daniel Wagner
2015-09-09 12:05 ` [PATCH v1 2/8] KVM: use simple waitqueue for vcpu->wq Daniel Wagner
2015-09-09 12:05 ` [PATCH v1 3/8] sched/completion: convert completions to use simple wait queues Daniel Wagner
2015-09-09 14:26   ` Peter Zijlstra
2015-10-12  9:17     ` Daniel Wagner
2015-10-12 10:03       ` Daniel Wagner [this message]
2015-10-12 11:58         ` Peter Zijlstra
2015-10-12 12:44           ` Daniel Wagner
2015-09-09 12:05 ` [PATCH v1 4/8] rcu: use simple wait queues where possible in rcutree Daniel Wagner
2015-09-09 12:05 ` [PATCH v1 5/8] rcu: Do not call swake_up_all with rnp->lock holding Daniel Wagner
2015-09-09 12:05 ` [PATCH v1 6/8] gadgetfs: Fix fallout of wait to swait completion change Daniel Wagner
2015-09-09 12:05 ` [PATCH v1 7/8] usb: gadget: f_fs: " Daniel Wagner
2015-09-09 12:05 ` [PATCH v1 8/8] orinoco_usb: " Daniel Wagner

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=561B855A.2060403@bmw-carit.de \
    --to=daniel.wagner@bmw-carit.de \
    --cc=linux-kernel@vger.kernel.org \
    --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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox