From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 75674A0C43; Thu, 5 Aug 2021 08:08:19 +0200 (CEST) Received: from [217.70.189.124] (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 3F8D34014F; Thu, 5 Aug 2021 08:08:19 +0200 (CEST) Received: from smartserver.smartsharesystems.com (smartserver.smartsharesystems.com [77.243.40.215]) by mails.dpdk.org (Postfix) with ESMTP id 865E040143 for ; Thu, 5 Aug 2021 08:08:18 +0200 (CEST) Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Date: Thu, 5 Aug 2021 08:08:15 +0200 Message-ID: <98CBD80474FA8B44BF855DF32C47DC35C6195B@smartserver.smartshare.dk> X-MimeOLE: Produced By Microsoft Exchange V6.5 In-Reply-To: <20210804132952.2674942-1-thomas@monjalon.net> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: [dpdk-dev] [PATCH] doc: add known issue with mbuf segment Thread-Index: AdeJNOaO3NAsJjLuRjqS8bO1RO2WZgAia9Sg References: <20210804132952.2674942-1-thomas@monjalon.net> From: =?iso-8859-1?Q?Morten_Br=F8rup?= To: "Thomas Monjalon" , Cc: , , , , , , , , "Olivier Matz" Subject: Re: [dpdk-dev] [PATCH] doc: add known issue with mbuf segment X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Sender: "dev" > From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Thomas Monjalon > Sent: Wednesday, 4 August 2021 15.30 >=20 > A bug with segmented packets has been discovered but the agreement > to apply the fix is not concluded at the time of DPDK 21.08 release. > This bug seems to be in DPDK for many years and should be fixed in > 21.11. >=20 > Suggested-by: Olivier Matz > Signed-off-by: Thomas Monjalon > --- > doc/guides/rel_notes/release_21_08.rst | 12 ++++++++++++ > 1 file changed, 12 insertions(+) >=20 > diff --git a/doc/guides/rel_notes/release_21_08.rst > b/doc/guides/rel_notes/release_21_08.rst > index d7559ec6bf..14e32585b8 100644 > --- a/doc/guides/rel_notes/release_21_08.rst > +++ b/doc/guides/rel_notes/release_21_08.rst > @@ -251,6 +251,18 @@ Known Issues > Also, make sure to start the actual text at the margin. > = =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D >=20 > +* **Last mbuf segment not implicitly reset.** > + > + It is expected that free mbufs have their field ``nb_seg`` set to = 1, > + so that when it is allocated, the user does not need to set its > value. > + The mbuf free functions are responsible of resetting this field to = 1 > + before returning the mbuf to the pool. > + > + When a multi-segment mbuf is freed, the field ``nb_seg`` is not > reset > + to 1 for the last segment of the chain. On next allocation of this > segment, > + if the field is not explicitly reset by the user, > + an invalid mbuf can be created, and can cause an undefined = behavior. > + >=20 I have a couple of suggestions: 1. Add that this bug was present in previous DPDK releases for many = years. If possible, mention which release introduced the bug. We don't want users to choose to stay on older release, thinking that it = does not have this bug. Also, if a user experiences unexplainable problems with segmented = packets an older DPDK release, he might get lucky and find these release = notes, which could reveal the root cause of the problems he is = experiencing. 2. Add a link to the patch that fixes the bug. It will let users = affected by the bug easily fix it (possibly at the expense of degraded = performance). With those two additions, Acked-by: Morten Br=F8rup -Morten