From owner-freebsd-net@FreeBSD.ORG  Wed Nov 30 02:50:31 2005
Return-Path: <owner-freebsd-net@FreeBSD.ORG>
X-Original-To: freebsd-net@freebsd.org
Delivered-To: freebsd-net@freebsd.org
Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125])
	by hub.freebsd.org (Postfix) with ESMTP id E032A16A420;
	Wed, 30 Nov 2005 02:50:31 +0000 (GMT)
	(envelope-from nielsen-list@memberwebs.com)
Received: from mail.npubs.com (mail.zoneseven.net [209.66.100.224])
	by mx1.FreeBSD.org (Postfix) with ESMTP id CE38043D72;
	Wed, 30 Nov 2005 02:50:13 +0000 (GMT)
	(envelope-from nielsen-list@memberwebs.com)
Message-ID: <438D0896.1070808@memberwebs.com>
From: Nate Nielsen <nielsen-list@memberwebs.com>
User-Agent: Mozilla Thunderbird 1.0.7 (X11/20051013)
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: freebsd-hackers@freebsd.org, freebsd-net@freebsd.org,
	freebsd-current@freebsd.org
References: <438CE0A8.4010205@memberwebs.com>
In-Reply-To: <438CE0A8.4010205@memberwebs.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
X-Virus-Scanned: ClamAV using ClamSMTP
Date: Wed, 30 Nov 2005 03:04:44 +0000 (GMT)
Cc: 
Subject: Re: Memory leak in net80211 on FBSD 6.0
X-BeenThere: freebsd-net@freebsd.org
X-Mailman-Version: 2.1.5
Precedence: list
Reply-To: nielsen@memberwebs.com
List-Id: Networking and TCP/IP with FreeBSD <freebsd-net.freebsd.org>
List-Unsubscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-net>,
	<mailto:freebsd-net-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/freebsd-net>
List-Post: <mailto:freebsd-net@freebsd.org>
List-Help: <mailto:freebsd-net-request@freebsd.org?subject=help>
List-Subscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-net>,
	<mailto:freebsd-net-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Wed, 30 Nov 2005 02:50:32 -0000

Nate Nielsen wrote:
> I'm experiencing a memory leak in the net80211 code. I have two atheros
> 5213-A cards on two embedded systems running FreeBSD 6.0. They are setup
> as IBSS (adhoc) stations. After roughly 15 seconds of ~14Mbps TCP
> traffic (single stream) I promptly run out of memory:
> 
>>login: panic: kmem_malloc(4096): kmem_map too small: 25165824 total allocated
>>Uptime: 3m0s
>>Cannot dump. No dump device defined.
>>Automatic reboot in 5 seconds - press a key on the console to abort
>>Rebooting...
> 
> The boxes have 64MB of memory each.
> vmstat -m reports the following just before the panic:
> 
>>         Type InUse MemUse HighUse Requests  Size(s)
>><snip>
>>    80211node 20554 20554K       -    20559  512,1024

I've put debugging code in node_alloc() and node_free() (in
net80211/ieee80211_node.c) and can confirm that thousands of
ieee80211_node structures are being allocated, at a cost of 1K a piece.

I'll continue to look into this (stumbling about), but if anyone has any
advice or ideas of where to look, I'd be eternally grateful.

Cheers,
Nate