All of lore.kernel.org
 help / color / mirror / Atom feed
From: Nick Piggin <nickpiggin@yahoo.com.au>
To: nagar@watson.ibm.com
Cc: linux-kernel <linux-kernel@vger.kernel.org>,
	lse-tech <lse-tech@lists.sourceforge.net>
Subject: Re: [Patch 2/7] Add sysctl for schedstats
Date: Mon, 27 Feb 2006 20:17:47 +1100	[thread overview]
Message-ID: <4402C3BB.7010705@yahoo.com.au> (raw)
In-Reply-To: <1141027923.5785.50.camel@elinux04.optonline.net>

Shailabh Nagar wrote:
> schedstats-sysctl.patch
> 
> Add sysctl option for controlling schedstats collection
> dynamically. Delay accounting leverages schedstats for
> cpu delay statistics.
> 

I'd sort of rather not tie this in with schedstats if possible.
Schedstats adds a reasonable amount of cache footprint and
branches in hot paths. Most of schedstats stuff is something that
hardly anyone will use.

Sure you can share common code though...

>  
> Index: linux-2.6.16-rc4/kernel/sched.c
> ===================================================================
> --- linux-2.6.16-rc4.orig/kernel/sched.c	2006-02-27 01:20:04.000000000 -0500
> +++ linux-2.6.16-rc4/kernel/sched.c	2006-02-27 01:52:52.000000000 -0500
> @@ -382,11 +382,56 @@ static inline void task_rq_unlock(runque
>  }
>  
>  #ifdef CONFIG_SCHEDSTATS
> +
> +int schedstats_sysctl = 0;		/* schedstats turned off by default */

Should be read mostly.

> +static DEFINE_PER_CPU(int, schedstats) = 0;
> +

When the above is in the read mostly section, you won't need this at all.

You don't intend to switch the sysctl with great frequency, do you?

> +static void schedstats_set(int val)
> +{
> +	int i;
> +	static spinlock_t schedstats_lock = SPIN_LOCK_UNLOCKED;
> +
> +	spin_lock(&schedstats_lock);
> +	schedstats_sysctl = val;
> +	for (i = 0; i < NR_CPUS; i++)
> +		per_cpu(schedstats, i) = val;
> +	spin_unlock(&schedstats_lock);
> +}
> +
> +static int __init schedstats_setup_enable(char *str)
> +{
> +	schedstats_sysctl = 1;
> +	schedstats_set(schedstats_sysctl);
> +	return 1;
> +}
> +
> +__setup("schedstats", schedstats_setup_enable);
> +
> +int schedstats_sysctl_handler(ctl_table *table, int write, struct file *filp,
> +			void __user *buffer, size_t *lenp, loff_t *ppos)
> +{
> +	int ret, prev = schedstats_sysctl;
> +	struct task_struct *g, *t;
> +
> +	ret = proc_dointvec(table, write, filp, buffer, lenp, ppos);
> +	if ((ret != 0) || (prev == schedstats_sysctl))
> +		return ret;
> +	if (schedstats_sysctl) {
> +		read_lock(&tasklist_lock);
> +		do_each_thread(g, t) {
> +			memset(&t->sched_info, 0, sizeof(t->sched_info));
> +		} while_each_thread(g, t);
> +		read_unlock(&tasklist_lock);
> +	}
> +	schedstats_set(schedstats_sysctl);

You don't clear the rq's schedstats stuff here.

And clearing this at all is not really needed for the schedstats interface.
You have a timestamp and a set of accumulated values, so it is easy to work
out deltas. So do you need this?

-- 
SUSE Labs, Novell Inc.
Send instant messages to your online friends http://au.messenger.yahoo.com 

  parent reply	other threads:[~2006-02-27  9:17 UTC|newest]

Thread overview: 66+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-02-27  7:56 [Patch 0/7] Per-task delay accounting Shailabh Nagar
2006-02-27  8:02 ` [Patch 1/7] timespec diff utility Shailabh Nagar
2006-02-27  8:12   ` [Patch 2/7] Add sysctl for schedstats Shailabh Nagar
2006-02-27  8:52     ` Ingo Molnar
2006-02-27 10:46       ` [Lse-tech] " Balbir Singh
2006-02-27 12:18         ` Arjan van de Ven
2006-02-27 12:29           ` Balbir Singh
2006-02-27 13:06             ` Arjan van de Ven
2006-02-27 16:16               ` Balbir Singh
2006-02-27  9:17     ` Nick Piggin [this message]
2006-02-27  9:41       ` Shailabh Nagar
2006-02-27 12:28         ` Nick Piggin
2006-02-27 19:09       ` [Lse-tech] " Chandra Seetharaman
2006-02-28  0:25         ` Nick Piggin
2006-02-28  1:42           ` chandra seetharaman
2006-03-07 17:26       ` schedstats refinement (was Re: [Lse-tech] Re: [Patch 2/7] Add sysctl for schedstats) Balbir Singh
2006-02-27 17:05     ` [Lse-tech] [Patch 2/7] Add sysctl for schedstats Bryan O'Sullivan
2006-02-27 20:55       ` Shailabh Nagar
2006-02-27 22:26         ` Bryan O'Sullivan
2006-02-27  8:22   ` [Patch 1/7] timespec diff utility Arjan van de Ven
2006-02-27  8:34     ` Shailabh Nagar
2006-02-27  8:37       ` Arjan van de Ven
2006-02-27  8:15 ` [Patch 3/7] delay accounting initial setup Shailabh Nagar
2006-02-27  8:18 ` [Patch 4/7] Add sysctl for delay accounting Shailabh Nagar
2006-02-27  8:26   ` Arjan van de Ven
2006-02-27  8:38     ` Shailabh Nagar
2006-02-27  8:42       ` Arjan van de Ven
2006-02-27  8:59         ` Shailabh Nagar
2006-02-27 11:18           ` [Lse-tech] " Balbir Singh
2006-02-27  9:04         ` Dipankar Sarma
2006-02-27  9:13           ` Arjan van de Ven
2006-02-27 10:11             ` [Lse-tech] " Balbir Singh
2006-02-27 11:24               ` Arjan van de Ven
2006-02-27 12:00                 ` Balbir Singh
2006-02-27 12:23                 ` Srivatsa Vaddagiri
2006-02-27  8:20 ` [Patch 5/7] synchronous block I/O delays Shailabh Nagar
2006-02-27  8:29   ` Arjan van de Ven
2006-02-27  9:13     ` Shailabh Nagar
2006-02-27  9:18       ` Arjan van de Ven
2006-02-27  9:24       ` Arjan van de Ven
2006-02-27 14:18   ` Andi Kleen
2006-02-27 21:31     ` Shailabh Nagar
2006-02-27 22:09     ` Shailabh Nagar
2006-02-27 22:20       ` Andi Kleen
2006-02-28  8:10       ` Arjan van de Ven
2006-02-27  8:22 ` [Patch 6/7] Swapin page fault delays Shailabh Nagar
2006-02-27  8:30   ` Arjan van de Ven
2006-02-27 22:16     ` Shailabh Nagar
2006-02-27  8:31 ` [Patch 7/7] Generic netlink interface (delay accounting) Shailabh Nagar
     [not found]   ` <1141045194.5363.12.camel@localhost.localdomain>
     [not found]     ` <4403608E.1050304@watson.ibm.com>
     [not found]       ` <1141652556.5185.64.camel@localhost.localdomain>
2006-03-06 17:00         ` Shailabh Nagar
2006-03-07 14:38           ` [Lse-tech] " jamal
2006-03-08 21:56             ` Shailabh Nagar
2006-03-08 21:56               ` Shailabh Nagar
2006-03-09 14:37               ` [UPDATED PATCH] Re: [Lse-tech] " Balbir Singh
2006-03-09 14:37                 ` [UPDATED PATCH] " Balbir Singh
2006-03-09 16:06                 ` [UPDATED PATCH] Re: [Lse-tech] " Shailabh Nagar
2006-03-09 16:06                   ` [UPDATED PATCH] " Shailabh Nagar
2006-03-10 14:53                 ` [UPDATED PATCH] Re: [Lse-tech] " jamal
2006-03-10 14:53                   ` [UPDATED PATCH] " jamal
2006-03-10 15:35                   ` [UPDATED PATCH] Re: [Lse-tech] " jamal
2006-03-10 16:39                   ` Balbir Singh
2006-03-10 16:39                     ` [UPDATED PATCH] " Balbir Singh
2006-03-11 13:30                     ` [UPDATED PATCH] Re: [Lse-tech] " jamal
2006-03-11 13:30                       ` [UPDATED PATCH] " jamal
2006-03-13 16:21                       ` [UPDATED PATCH] Re: [Lse-tech] " Balbir Singh
2006-02-27  9:10 ` [Patch 0/7] Per-task delay accounting Nick Piggin

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=4402C3BB.7010705@yahoo.com.au \
    --to=nickpiggin@yahoo.com.au \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lse-tech@lists.sourceforge.net \
    --cc=nagar@watson.ibm.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 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.