From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ben Greear Subject: Possible memory leak in 3.14.5+ (hacked). Date: Tue, 03 Jun 2014 11:42:48 -0700 Message-ID: <538E1728.8000804@candelatech.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit To: netdev Return-path: Received: from mail2.candelatech.com ([208.74.158.173]:43583 "EHLO mail2.candelatech.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754271AbaFCSmt (ORCPT ); Tue, 3 Jun 2014 14:42:49 -0400 Received: from [192.168.100.236] (firewall.candelatech.com [70.89.124.249]) (using TLSv1 with cipher DHE-RSA-CAMELLIA256-SHA (256/256 bits)) (No client certificate requested) by mail2.candelatech.com (Postfix) with ESMTPSA id A0E9A40C697 for ; Tue, 3 Jun 2014 11:42:48 -0700 (PDT) Sender: netdev-owner@vger.kernel.org List-ID: Our wireless stress tests are running OOM (on 8GB RAM system) after a few days. I turned on kmemleak, and see this, among others.. Kernel is hacked 3.14.5+ kernel, but we don't have many changes in the ipv6 code, at least. Anyone seen similar? [root@ct521-2n-4255 ~]# cat /debug/kmemleak unreferenced object 0xffff8802008f8c00 (size 1024): comm "softirq", pid 0, jiffies 4294932288 (age 1574.522s) hex dump (first 32 bytes): 00 00 00 00 00 00 00 00 b0 d7 17 a0 ff ff ff ff ................ 40 f5 f1 05 02 88 ff ff e0 8c fe ff 00 00 00 00 @............... backtrace: [] kmemleak_alloc+0x6e/0x93 [] slab_post_alloc_hook+0x28/0x2a [] __kmalloc+0xef/0x12f [] kzalloc+0xf/0x11 [] __neigh_create+0x68/0x561 [] ip6_finish_output2+0x233/0x3d8 [ipv6] [] ip6_finish_output+0x6e/0x73 [ipv6] [] ip6_output+0x7c/0xa8 [ipv6] [] dst_output+0x18/0x1c [ipv6] [] ndisc_send_skb+0x1dc/0x2b4 [ipv6] [] ndisc_send_rs+0xd0/0xe5 [ipv6] [] addrconf_dad_completed+0x167/0x1ed [ipv6] [] addrconf_dad_work+0x221/0x2bb [ipv6] [] process_one_work+0x1cb/0x2ee [] worker_thread+0x12f/0x1fc [] kthread+0xc4/0xcc Thanks, Ben -- Ben Greear Candela Technologies Inc http://www.candelatech.com