All of lore.kernel.org
 help / color / mirror / Atom feed
From: Oleg Nesterov <oleg@redhat.com>
To: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Ingo Molnar <mingo@elte.hu>,
	Andrew Morton <akpm@linux-foundation.org>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH] workqueue: don't alloc_percpu for single workqueue
Date: Wed, 21 Jan 2009 13:17:56 +0100	[thread overview]
Message-ID: <20090121121756.GA11942@redhat.com> (raw)
In-Reply-To: <4976EE0B.5090200@cn.fujitsu.com>

On 01/21, Lai Jiangshan wrote:
>
> allocating memory for every cpu for single workqueue is waste.

Yes, perhaps this makes sense, we can save a bit of per-cpu memory
for each single-threaded wq, and the patch looks correct.

> -static struct cpu_workqueue_struct *
> -init_cpu_workqueue(struct workqueue_struct *wq, int cpu)
> +static void init_cpu_workqueue(struct workqueue_struct *wq,
> +		struct cpu_workqueue_struct *cwq)
>  {
> -	struct cpu_workqueue_struct *cwq = per_cpu_ptr(wq->cpu_wq, cpu);
> -
>  	cwq->wq = wq;
>  	spin_lock_init(&cwq->lock);
>  	INIT_LIST_HEAD(&cwq->worklist);
>  	init_waitqueue_head(&cwq->more_work);
> -
> -	return cwq;
>  }

Do we really need to change the prototype of init_cpu_workqueue()
and change then change __create_workqueue_key() accordingly?
Afaics, the only change in init_cpu_workqueue() we need is

	-	struct cpu_workqueue_struct *cwq = per_cpu_ptr(wq->cpu_wq, cpu);
	+	struct cpu_workqueue_struct *cwq = wq_per_cpu(wq, cpu);

no?

> @@ -906,6 +907,13 @@ void destroy_workqueue(struct workqueue_struct *wq)
>  	const struct cpumask *cpu_map = wq_cpu_map(wq);
>  	int cpu;
>
> +	if (is_wq_single_threaded(wq)) {
> +		cleanup_workqueue_thread(wq->cpu_wq);
> +		kfree(wq->cpu_wq);
> +		kfree(wq);
> +		return;
> +	}

again, not sure I understand why this change is needed. Afaics we
only need to use kfree(wq->cpu_wq) instead of free_percpu() if
it is single-threaded.

Oleg.


  parent reply	other threads:[~2009-01-21 12:20 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-01-21  9:42 [PATCH] workqueue: don't alloc_percpu for single workqueue Lai Jiangshan
2009-01-21 10:29 ` Frédéric Weisbecker
2009-01-21 10:48 ` Ingo Molnar
2009-01-22  1:05   ` Lai Jiangshan
2009-01-21 12:17 ` Oleg Nesterov [this message]
2009-01-22  3:42   ` Lai Jiangshan
2009-01-22 16:11     ` Oleg Nesterov
2009-02-16  2:39       ` Lai Jiangshan
2009-01-21 13:01 ` Eric Dumazet

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=20090121121756.GA11942@redhat.com \
    --to=oleg@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=laijs@cn.fujitsu.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    /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.