Date: Mon, 7 Jan 2008 20:47:01 -0500
From: Benjamin LaHaise <bcrl@kvack.org>
To: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: linux-kernel@vger.kernel.org, torvalds@linux-foundation.org,
       David Miller <davem@davemloft.net>
Subject: Re: regression: sysctl_check changes in 2.6.24 are O(n) resulting in slow creation of 10000 network interfaces
Message-ID: <20080108014701.GI28570@kvack.org>
References: <20080106220307.GU28570@kvack.org> <m1ir26ktqi.fsf@ebiederm.dsl.xmission.com> <20080106.231040.222338659.davem@davemloft.net> <m1tzlqj5kz.fsf@ebiederm.dsl.xmission.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <m1tzlqj5kz.fsf@ebiederm.dsl.xmission.com>
User-Agent: Mutt/1.4.1i
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 888
Lines: 21

On Mon, Jan 07, 2008 at 03:25:00AM -0700, Eric W. Biederman wrote:
> I think someone failed to notice that using /proc/sys slowed to a crawl
> in that event, and now that I am doing a lookup on register it seems to
> showing up in the benchmarks.

The directory that is problematic rarely gets accessed.  Fixing sysfs 
properly at some point would be a good idea, it's just one of those things 
that isn't quite high priority yet.

> Benjamin can you test the patch below and tell me if it also
> keeps the network device performance at acceptable levels.

I can confirm that this patch does indeed put the regression at bay for 
now.

		-ben
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/