DPDK patches and discussions
 help / color / mirror / Atom feed
From: Ferruh Yigit <ferruh.yigit@intel.com>
To: Xiaoyun Li <xiaoyun.li@intel.com>, <Aman.Deep.Singh@intel.com>,
	<olivier.matz@6wind.com>, <mb@smartsharesystems.com>,
	<konstantin.ananyev@intel.com>, <stephen@networkplumber.org>,
	<vladimir.medvedkin@intel.com>
Cc: <dev@dpdk.org>, Sunil Pai G <sunil.pai.g@intel.com>
Subject: Re: [PATCH v5 2/2] testpmd: fix l4 sw csum over multi segments
Date: Fri, 21 Jan 2022 15:16:40 +0000	[thread overview]
Message-ID: <1c366b66-4e1a-94ba-e1af-161a1dded713@intel.com> (raw)
In-Reply-To: <20220106160333.762686-3-xiaoyun.li@intel.com>

On 1/6/2022 4:03 PM, Xiaoyun Li wrote:
> In csum forwarding mode, software UDP/TCP csum calculation only takes
> the first segment into account while using the whole packet length so
> the calculation will read invalid memory region with multi-segments
> packets and will get wrong value.
> This patch fixes this issue.
> 
> Signed-off-by: Xiaoyun Li <xiaoyun.li@intel.com>
> Tested-by: Sunil Pai G <sunil.pai.g@intel.com>

Can you please check following check-git-log.sh warnings:

./devtools/check-git-log.sh -2
Wrong headline label:
         testpmd: fix l4 sw csum over multi segments
Wrong headline case:
                         "testpmd: fix l4 sw csum over multi segments": l4 --> L4
Wrong headline case:
                         "testpmd: fix l4 sw csum over multi segments": sw --> SW
Missing 'Fixes' tag:
         testpmd: fix l4 sw csum over multi segments



> ---
>   app/test-pmd/csumonly.c | 41 ++++++++++++++++++++++++++---------------
>   1 file changed, 26 insertions(+), 15 deletions(-)
> 
> diff --git a/app/test-pmd/csumonly.c b/app/test-pmd/csumonly.c
> index 2aeea243b6..0fbe1f1be7 100644
> --- a/app/test-pmd/csumonly.c
> +++ b/app/test-pmd/csumonly.c
> @@ -96,12 +96,13 @@ struct simple_gre_hdr {
>   } __rte_packed;
>   
>   static uint16_t
> -get_udptcp_checksum(void *l3_hdr, void *l4_hdr, uint16_t ethertype)
> +get_udptcp_checksum(struct rte_mbuf *m, void *l3_hdr, uint16_t l4_off,
> +		    uint16_t ethertype)
>   {
>   	if (ethertype == _htons(RTE_ETHER_TYPE_IPV4))
> -		return rte_ipv4_udptcp_cksum(l3_hdr, l4_hdr);
> +		return rte_ipv4_udptcp_cksum_mbuf(m, l3_hdr, l4_off);
>   	else /* assume ethertype == RTE_ETHER_TYPE_IPV6 */
> -		return rte_ipv6_udptcp_cksum(l3_hdr, l4_hdr);
> +		return rte_ipv6_udptcp_cksum_mbuf(m, l3_hdr, l4_off);
>   }
>   
>   /* Parse an IPv4 header to fill l3_len, l4_len, and l4_proto */
> @@ -460,7 +461,7 @@ parse_encap_ip(void *encap_ip, struct testpmd_offload_info *info)
>    * depending on the testpmd command line configuration */
>   static uint64_t
>   process_inner_cksums(void *l3_hdr, const struct testpmd_offload_info *info,
> -	uint64_t tx_offloads)
> +	uint64_t tx_offloads, struct rte_mbuf *m)
>   {
>   	struct rte_ipv4_hdr *ipv4_hdr = l3_hdr;
>   	struct rte_udp_hdr *udp_hdr;
> @@ -468,6 +469,7 @@ process_inner_cksums(void *l3_hdr, const struct testpmd_offload_info *info,
>   	struct rte_sctp_hdr *sctp_hdr;
>   	uint64_t ol_flags = 0;
>   	uint32_t max_pkt_len, tso_segsz = 0;
> +	uint16_t l4_off;
>   
>   	/* ensure packet is large enough to require tso */
>   	if (!info->is_tunnel) {
> @@ -510,9 +512,15 @@ process_inner_cksums(void *l3_hdr, const struct testpmd_offload_info *info,
>   			if (tx_offloads & RTE_ETH_TX_OFFLOAD_UDP_CKSUM) {
>   				ol_flags |= RTE_MBUF_F_TX_UDP_CKSUM;
>   			} else {
> +				if (info->is_tunnel)
> +					l4_off = info->l2_len +
> +						 info->outer_l3_len +
> +						 info->l2_len + info->l3_len;
> +				else
> +					l4_off = info->l2_len +	info->l3_len;
>   				udp_hdr->dgram_cksum = 0;
>   				udp_hdr->dgram_cksum =
> -					get_udptcp_checksum(l3_hdr, udp_hdr,
> +					get_udptcp_checksum(m, l3_hdr, l4_off,
>   						info->ethertype);
>   			}
>   		}
> @@ -527,9 +535,14 @@ process_inner_cksums(void *l3_hdr, const struct testpmd_offload_info *info,
>   		else if (tx_offloads & RTE_ETH_TX_OFFLOAD_TCP_CKSUM) {
>   			ol_flags |= RTE_MBUF_F_TX_TCP_CKSUM;
>   		} else {
> +			if (info->is_tunnel)
> +				l4_off = info->l2_len + info->outer_l3_len +
> +					 info->l2_len + info->l3_len;
> +			else
> +				l4_off = info->l2_len + info->l3_len;
>   			tcp_hdr->cksum = 0;
>   			tcp_hdr->cksum =
> -				get_udptcp_checksum(l3_hdr, tcp_hdr,
> +				get_udptcp_checksum(m, l3_hdr, l4_off,
>   					info->ethertype);
>   		}
>   #ifdef RTE_LIB_GSO
> @@ -557,7 +570,7 @@ process_inner_cksums(void *l3_hdr, const struct testpmd_offload_info *info,
>   /* Calculate the checksum of outer header */
>   static uint64_t
>   process_outer_cksums(void *outer_l3_hdr, struct testpmd_offload_info *info,
> -	uint64_t tx_offloads, int tso_enabled)
> +	uint64_t tx_offloads, int tso_enabled, struct rte_mbuf *m)
>   {
>   	struct rte_ipv4_hdr *ipv4_hdr = outer_l3_hdr;
>   	struct rte_ipv6_hdr *ipv6_hdr = outer_l3_hdr;
> @@ -611,12 +624,9 @@ process_outer_cksums(void *outer_l3_hdr, struct testpmd_offload_info *info,
>   	/* do not recalculate udp cksum if it was 0 */
>   	if (udp_hdr->dgram_cksum != 0) {
>   		udp_hdr->dgram_cksum = 0;
> -		if (info->outer_ethertype == _htons(RTE_ETHER_TYPE_IPV4))
> -			udp_hdr->dgram_cksum =
> -				rte_ipv4_udptcp_cksum(ipv4_hdr, udp_hdr);
> -		else
> -			udp_hdr->dgram_cksum =
> -				rte_ipv6_udptcp_cksum(ipv6_hdr, udp_hdr);
> +		udp_hdr->dgram_cksum = get_udptcp_checksum(m, outer_l3_hdr,
> +					info->l2_len + info->outer_l3_len,
> +					info->outer_ethertype);
>   	}
>   
>   	return ol_flags;
> @@ -957,7 +967,7 @@ pkt_burst_checksum_forward(struct fwd_stream *fs)
>   
>   		/* process checksums of inner headers first */
>   		tx_ol_flags |= process_inner_cksums(l3_hdr, &info,
> -			tx_offloads);
> +			tx_offloads, m);
>   
>   		/* Then process outer headers if any. Note that the software
>   		 * checksum will be wrong if one of the inner checksums is
> @@ -965,7 +975,8 @@ pkt_burst_checksum_forward(struct fwd_stream *fs)
>   		if (info.is_tunnel == 1) {
>   			tx_ol_flags |= process_outer_cksums(outer_l3_hdr, &info,
>   					tx_offloads,
> -					!!(tx_ol_flags & RTE_MBUF_F_TX_TCP_SEG));
> +					!!(tx_ol_flags & RTE_MBUF_F_TX_TCP_SEG),
> +					m);
>   		}
>   
>   		/* step 3: fill the mbuf meta data (flags and header lengths) */


  reply	other threads:[~2022-01-21 15:16 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-10-15  5:13 [dpdk-dev] [PATCH] app/testpmd: " Xiaoyun Li
2021-10-15  8:09 ` David Marchand
2021-10-18  2:02   ` Li, Xiaoyun
2021-10-18  2:16 ` [dpdk-dev] [PATCH v2] " Xiaoyun Li
2021-10-18  3:00 ` [dpdk-dev] [PATCH] " Stephen Hemminger
2021-10-18  3:16   ` Li, Xiaoyun
2021-10-18  4:40     ` Li, Xiaoyun
2021-10-18 10:15     ` Ananyev, Konstantin
2021-10-19  1:54       ` Li, Xiaoyun
2021-10-20 10:12 ` [dpdk-dev] [PATCH v3] " Xiaoyun Li
2021-10-27 10:48   ` Ferruh Yigit
2021-10-27 11:29     ` Morten Brørup
2021-10-29  8:29       ` Olivier Matz
2021-12-03 11:31         ` Li, Xiaoyun
2021-12-03 11:38 ` [PATCH v4 0/2] Add functions to calculate UDP/TCP cksum in mbuf Xiaoyun Li
2021-12-03 11:38   ` [PATCH v4 1/2] net: add " Xiaoyun Li
2021-12-15 11:33     ` Singh, Aman Deep
2022-01-04 15:18       ` Li, Xiaoyun
2022-01-04 15:40         ` Li, Xiaoyun
2022-01-06 12:56           ` Singh, Aman Deep
2021-12-03 11:38   ` [PATCH v4 2/2] testpmd: fix l4 sw csum over multi segments Xiaoyun Li
2021-12-08  6:10   ` [PATCH v4 0/2] Add functions to calculate UDP/TCP cksum in mbuf Pai G, Sunil
2022-01-06 16:03 ` [PATCH v5 " Xiaoyun Li
2022-01-06 16:03   ` [PATCH v5 1/2] net: add " Xiaoyun Li
2022-01-21 15:16     ` Ferruh Yigit
2022-01-06 16:03   ` [PATCH v5 2/2] testpmd: fix l4 sw csum over multi segments Xiaoyun Li
2022-01-21 15:16     ` Ferruh Yigit [this message]
2022-01-24  9:43       ` Li, Xiaoyun
2022-01-24 10:16         ` Ferruh Yigit
2022-01-21 17:09     ` Kevin Traynor
2022-01-24  9:16       ` Ferruh Yigit
2022-01-24 10:30         ` Kevin Traynor
2022-01-24 11:02           ` Ferruh Yigit
2022-01-24 12:28 ` [PATCH v6 0/2] Add functions to calculate UDP/TCP cksum in mbuf Xiaoyun Li
2022-01-24 12:28   ` [PATCH v6 1/2] net: add " Xiaoyun Li
2022-02-03 12:41     ` Ferruh Yigit
2022-01-24 12:28   ` [PATCH v6 2/2] app/testpmd: enable L4 SW csum over multi segments Xiaoyun Li
2022-02-04 13:12     ` Ferruh Yigit
2022-02-04 13:12   ` [PATCH v6 0/2] Add functions to calculate UDP/TCP cksum in mbuf Ferruh Yigit

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1c366b66-4e1a-94ba-e1af-161a1dded713@intel.com \
    --to=ferruh.yigit@intel.com \
    --cc=Aman.Deep.Singh@intel.com \
    --cc=dev@dpdk.org \
    --cc=konstantin.ananyev@intel.com \
    --cc=mb@smartsharesystems.com \
    --cc=olivier.matz@6wind.com \
    --cc=stephen@networkplumber.org \
    --cc=sunil.pai.g@intel.com \
    --cc=vladimir.medvedkin@intel.com \
    --cc=xiaoyun.li@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).