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 AAD08436CE; Tue, 12 Dec 2023 09:10:08 +0100 (CET) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 8385542DEC; Tue, 12 Dec 2023 09:10:08 +0100 (CET) Received: from dkmailrelay1.smartsharesystems.com (smartserver.smartsharesystems.com [77.243.40.215]) by mails.dpdk.org (Postfix) with ESMTP id BE23342DCE; Tue, 12 Dec 2023 09:10:07 +0100 (CET) Received: from smartserver.smartsharesystems.com (smartserver.smartsharesys.local [192.168.4.10]) by dkmailrelay1.smartsharesystems.com (Postfix) with ESMTP id 90E0922D2B; Tue, 12 Dec 2023 09:10:07 +0100 (CET) Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Subject: RE: [PATCH v3] lib/net: fix tcp/udp cksum with padding data X-MimeOLE: Produced By Microsoft Exchange V6.5 Date: Tue, 12 Dec 2023 09:10:04 +0100 Message-ID: <98CBD80474FA8B44BF855DF32C47DC35E9F0B0@smartserver.smartshare.dk> In-Reply-To: <20231212021619.2038881-1-kaiwenx.deng@intel.com> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: [PATCH v3] lib/net: fix tcp/udp cksum with padding data Thread-Index: Adosp4H+XWo4RYj3QE6rrO1S/vzwIgAKGCBA References: <20231207085303.1924630-1-kaiwenx.deng@intel.com> <20231212021619.2038881-1-kaiwenx.deng@intel.com> From: =?iso-8859-1?Q?Morten_Br=F8rup?= To: "Kaiwen Deng" , Cc: , , , "Xiaoyun Li" , "Aman Singh" , "Ferruh Yigit" 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 > From: Kaiwen Deng [mailto:kaiwenx.deng@intel.com] > Sent: Tuesday, 12 December 2023 03.16 >=20 > IEEE 802 packets may have a minimum size limit. The data fields > should be padded when necessary. In some cases, the padding data > is not zero. >=20 > In 'rte_ipv4_udptcp_cksum_mbuf()', as payload length > "mbuf->pkt_len - l4_off" is used, which includes padding and if > padding is not zero it will end up producing wrong checksum. >=20 > This patch will use IP header to get the payload size to calculate > tcp/udp checksum. >=20 > Fixes: d178f693bbfe ("net: add UDP/TCP checksum in mbuf segments") > Cc: stable@dpdk.org >=20 > Signed-off-by: Kaiwen Deng > --- > lib/net/rte_ip.h | 9 +++++++-- > 1 file changed, 7 insertions(+), 2 deletions(-) >=20 > diff --git a/lib/net/rte_ip.h b/lib/net/rte_ip.h > index 6fa98a5a0f..c503a2b57f 100644 > --- a/lib/net/rte_ip.h > +++ b/lib/net/rte_ip.h > @@ -423,7 +423,10 @@ __rte_ipv4_udptcp_cksum_mbuf(const struct = rte_mbuf > *m, > if (l4_off > m->pkt_len) > return 0; It's not directly related, but while you are at it, please also add = unlikely to the above check: - if (l4_off > m->pkt_len) - return 0; + if (unlikely(l4_off > m->pkt_len)) + return 0; /* invalid params, return a dummy value */ >=20 > - if (rte_raw_cksum_mbuf(m, l4_off, m->pkt_len - l4_off, > &raw_cksum)) > + uint16_t len =3D rte_be_to_cpu_16(ipv4_hdr->total_length) - > + (uint16_t)rte_ipv4_hdr_len(ipv4_hdr); > + > + if (rte_raw_cksum_mbuf(m, l4_off, len, &raw_cksum)) Please declare "uint16_t len;" with the other variables at the top of = the function, and only set its value here. (It's allowed to declare = here, but please follow this function's existing convention of where = variables are declared.) > return 0; >=20 > cksum =3D raw_cksum + rte_ipv4_phdr_cksum(ipv4_hdr, 0); > @@ -666,7 +669,9 @@ __rte_ipv6_udptcp_cksum_mbuf(const struct rte_mbuf > *m, > if (l4_off > m->pkt_len) > return 0; Again not directly related, but please also add unlikely to this = comparison: - if (l4_off > m->pkt_len) - return 0; + if (unlikely(l4_off > m->pkt_len)) + return 0; /* invalid params, return a dummy value */ >=20 > - if (rte_raw_cksum_mbuf(m, l4_off, m->pkt_len - l4_off, > &raw_cksum)) > + uint16_t len =3D rte_be_to_cpu_16(ipv6_hdr->payload_len); > + > + if (rte_raw_cksum_mbuf(m, l4_off, len, &raw_cksum)) No need for "len" variable here, just use = rte_be_to_cpu_16(ipv6_hdr->payload_len) directly: - if (rte_raw_cksum_mbuf(m, l4_off, m->pkt_len - l4_off, &raw_cksum)) + if (rte_raw_cksum_mbuf(m, l4_off, = rte_be_to_cpu_16(ipv6_hdr->payload_len), &raw_cksum)) > return 0; >=20 > cksum =3D raw_cksum + rte_ipv6_phdr_cksum(ipv6_hdr, 0); > -- > 2.34.1