From mboxrd@z Thu Jan  1 00:00:00 1970
From: Jan Sacha <Jan.Sacha-cfy2TCaE7SFv+uJa97DSA9BPR1lH4CV8@public.gmane.org>
Subject: rte_lpm_add in librte_lpm
Date: Fri, 5 Sep 2014 17:14:50 -0700
Message-ID: <540A51FA.3070905@alcatel-lucent.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="ISO-8859-1"; format=flowed
Content-Transfer-Encoding: 7bit
To: <dev-VfR2kkLFssw@public.gmane.org>
Return-path: <dev-bounces-VfR2kkLFssw@public.gmane.org>
List-Id: patches and discussions about DPDK <dev.dpdk.org>
List-Unsubscribe: <http://dpdk.org/ml/options/dev>,
 <mailto:dev-request-VfR2kkLFssw@public.gmane.org?subject=unsubscribe>
List-Archive: <http://dpdk.org/ml/archives/dev/>
List-Post: <mailto:dev-VfR2kkLFssw@public.gmane.org>
List-Help: <mailto:dev-request-VfR2kkLFssw@public.gmane.org?subject=help>
List-Subscribe: <http://dpdk.org/ml/listinfo/dev>,
 <mailto:dev-request-VfR2kkLFssw@public.gmane.org?subject=subscribe>
Errors-To: dev-bounces-VfR2kkLFssw@public.gmane.org
Sender: "dev" <dev-bounces-VfR2kkLFssw@public.gmane.org>


Hi,

I've been playing with DPDK and I got some crashes in the LPM module. As 
far as I know, this problem has not been reported yet. These crashes 
happened in DPDK 1.7.0 and 1.6.0r2 (I didn't check other versions) when 
I added new LPM routes.

In the lib/librte_lpm/rte_lpm.c file, line 399 (add_depth_small 
function), there is the following condition:

if (!lpm->tbl24[i].valid || (lpm->tbl24[i].ext_entry == 0 && 
lpm->tbl24[i].depth <= depth))

If the new route has a lower depth than an existing non-extended entry, 
this condition is not met. We then skip to line 418, where we assume the 
entry is extended and use an uninitialized index. Is this a bug?

Thanks,
Jan