From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx1.evs.tv (mx1.evs.tv [195.190.86.126]) by dpdk.org (Postfix) with ESMTP id 3A65E6CB3 for ; Mon, 17 Oct 2016 16:46:00 +0200 (CEST) From: "VERDOUX, Sylvain" To: "users@dpdk.org" Thread-Topic: How to debug I40E PMD RX/TX frees Thread-Index: AdIog+rB4Xjvfdu4T921BZ+cNRc3SA== Date: Mon, 17 Oct 2016 14:45:59 +0000 Message-ID: <294AEAFC4B0A3E498038549EC65E9D1B018AB03E87@BELGSMBX01.EVS.TV> Accept-Language: fr-FR, fr-BE, en-US Content-Language: fr-FR X-MS-Has-Attach: X-MS-TNEF-Correlator: x-tm-as-product-ver: SMEX-11.0.0.4255-8.000.1202-22644.000 x-tm-as-result: No--46.881200-0.000000-31 x-tm-as-user-approved-sender: Yes x-tm-as-user-blocked-sender: No MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable X-Content-Filtered-By: Mailman/MimeDel 2.1.15 Subject: [dpdk-users] How to debug I40E PMD RX/TX frees X-BeenThere: users@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: usage discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 17 Oct 2016 14:46:00 -0000 Hello, I am currently trying a small igmp sample test using dpdk I40E PMD (NIC is = X710). The code is quite simple and uses 2 lcores to handle a recv loop on = RX0 and a send loop on TX0. When the recv loop detects a query, it notifies= the send loop to emit a report. I use at startup a fdir filter to move the= mcast flow to another queue to not be polluted with this flow in my recv l= oop. For a while it works fine but after some time, the recv loop just stop= retrieving queries, and I have difficulties debugging the problem. It seem= s like a mbuf free problem but even if I have activated I40E debug and mbuf= debug in my dpdk config file then recompile it, I can't see any alarming m= essage. For now I think that I free received mbufs properly and sent mbufs = should be released by the tx_burst function, so I don't really understand t= he issue. So I'm asking if tools are available to debug deeply the problem = ? Thanks, Best Regards Sylvain