netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: ebiederm@xmission.com (Eric W. Biederman)
To: Eric Dumazet <eric.dumazet@gmail.com>
Cc: "David S. Miller" <davem@davemloft.net>, Greg KH <greg@kroah.com>,
	Benjamin LaHaise <bcrl@lhnet.ca>,
	Octavian Purdila <opurdila@ixiacom.com>,
	netdev@vger.kernel.org, Cosmin Ratiu <cratiu@ixiacom.com>,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH] sysctl: reduce ram usage by 40 %
Date: Tue, 03 Nov 2009 02:23:08 -0800	[thread overview]
Message-ID: <m1bpjjewkz.fsf@fess.ebiederm.org> (raw)
In-Reply-To: <4AEFD544.6040602@gmail.com> (Eric Dumazet's message of "Tue\, 03 Nov 2009 08\:01\:24 +0100")

Eric Dumazet <eric.dumazet@gmail.com> writes:

> Eric Dumazet a écrit :
>
>> Its curious because in my tests the biggest problems come from
>> kernel/sysctl.c (__register_sysctl_paths) consuming 80% of cpu
>> in following attempt to create 20.000 devices

I bet that is Al's cute glue all the sysctl data structures together
patch.  It improves readdir and lookup at a small cost at registration
time.

>> (disable hotplug before trying this, and ipv6 too !)
>> modprobe dummy numdummies=20000


>> I believe we should address __register_sysctl_paths() scalability
>> problems too.

Agreed.

>> I dont know what is the 'sentinel' we allocate after each struct ctl_table
>> But I suspect we could reduce size requirement of the 'sentinel' to include
>> only needed fields for the sentinel (and move them at start of ctl_table)

The sentinel is just a NULL terminator.

> Here is the patch to reduce ram usage of sysctl :
>
> [PATCH] sysctl: reduce ram usage by 40 %
>
> We currently reserve space for a so called sentinel, a full struct ctl_table
> for each ctl_table. We can cheat a bit since only needed fields of a sentinel
> are ctl_name and procname. Add a new structure (struct ctl_table_sentinel)
> that includes a full ctl_table and only required part of a sentinel.

Before we address sysctl I would like to get out my patchset that
makes sys_sysctl a wrapper around the ascii version of
/proc/sys/net. Once that goes in it becomes much easier to do things
and perform radical surgery on sysctl.  Little things like .ctl_name and
.strategy go away.

Have you happened to look at the other cost of /proc proper?  Hmm.
Except for /proc/net/dev_snmp6 it doesn't look like we keep per
interface directories in proc so without ivp6 you won't see the proc
generic code at all.

The practical consequence is if /proc/net/dev_snmp6 is not painful during
registration right now we can probably convert all of /proc/sys/net to proc
generic after my other changes are in.

Eric

  reply	other threads:[~2009-11-03 10:23 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-11-01 16:31 [PATCH 1/3] sysfs directory scaling: rbtree for dirent name lookups Benjamin LaHaise
2009-11-01 16:32 ` [PATCH 2/3] sysfs directory scaling: doubly linked list for dirents Benjamin LaHaise
2009-11-01 16:33   ` [PATCH 2/3] sysfs directory scaling: count number of children dirs Benjamin LaHaise
2009-11-03  3:50 ` [PATCH 1/3] sysfs directory scaling: rbtree for dirent name lookups Greg KH
2009-11-03  6:14   ` Eric Dumazet
2009-11-03  7:01     ` [PATCH] sysctl: reduce ram usage by 40 % Eric Dumazet
2009-11-03 10:23       ` Eric W. Biederman [this message]
2009-11-03 16:07     ` [PATCH 1/3] sysfs directory scaling: rbtree for dirent name lookups Greg KH
2009-11-03 16:38       ` Octavian Purdila
2009-11-03 16:45       ` Benjamin LaHaise
2009-11-03 17:56         ` Greg KH
2009-11-03 22:28       ` Eric W. Biederman
2009-11-03 20:01   ` Benjamin LaHaise
2009-11-03 21:32     ` Eric W. Biederman
2009-11-03 21:43       ` Eric W. Biederman
2009-11-03 21:56         ` Benjamin LaHaise
2009-11-03 22:06           ` Eric Dumazet
2009-11-03 21:52       ` Benjamin LaHaise
2009-11-03 22:18         ` Eric W. Biederman
2009-11-03 10:41 ` Eric W. Biederman

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=m1bpjjewkz.fsf@fess.ebiederm.org \
    --to=ebiederm@xmission.com \
    --cc=bcrl@lhnet.ca \
    --cc=cratiu@ixiacom.com \
    --cc=davem@davemloft.net \
    --cc=eric.dumazet@gmail.com \
    --cc=greg@kroah.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=opurdila@ixiacom.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;
as well as URLs for NNTP newsgroup(s).