linux-sctp.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Xin Long <lucien.xin@gmail.com>
To: network dev <netdev@vger.kernel.org>, linux-sctp@vger.kernel.org
Cc: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>,
	Neil Horman <nhorman@tuxdriver.com>,
	Michael Tuexen <tuexen@fh-muenster.de>,
	davem@davemloft.net
Subject: [PATCHv2 net-next 03/17] udp: do checksum properly in skb_udp_tunnel_segment
Date: Thu, 08 Oct 2020 09:47:59 +0000	[thread overview]
Message-ID: <483d9eec159b22172fe04dacd58d7f88dfc2f301.1602150362.git.lucien.xin@gmail.com> (raw)
In-Reply-To: <c36b016ee429980b9585144f4f9af31bcda467ee.1602150362.git.lucien.xin@gmail.com>

This patch fixes two things:

  When skb->ip_summed = CHECKSUM_PARTIAL, skb_checksum_help() should be
  called do the checksum, instead of gso_make_checksum(), which is used
  to do the checksum for current proto after calling skb_segment(), not
  after the inner proto's gso_segment().

  When offload_csum is disabled, the hardware will not do the checksum
  for the current proto, udp. So instead of calling gso_make_checksum(),
  it should calculate checksum for udp itself.

Cc: Tom Herbert <tom@herbertland.com>
Signed-off-by: Xin Long <lucien.xin@gmail.com>
---
 net/ipv4/udp_offload.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c
index e67a66f..c0b010b 100644
--- a/net/ipv4/udp_offload.c
+++ b/net/ipv4/udp_offload.c
@@ -131,14 +131,15 @@ static struct sk_buff *__skb_udp_tunnel_segment(struct sk_buff *skb,
 		uh->check = ~csum_fold(csum_add(partial,
 				       (__force __wsum)htonl(len)));
 
-		if (skb->encapsulation || !offload_csum) {
-			uh->check = gso_make_checksum(skb, ~uh->check);
-			if (uh->check = 0)
-				uh->check = CSUM_MANGLED_0;
-		} else {
+		if (skb->encapsulation)
+			skb_checksum_help(skb);
+
+		if (offload_csum) {
 			skb->ip_summed = CHECKSUM_PARTIAL;
 			skb->csum_start = skb_transport_header(skb) - skb->head;
 			skb->csum_offset = offsetof(struct udphdr, check);
+		} else {
+			uh->check = csum_fold(skb_checksum(skb, udp_offset, len, 0));
 		}
 	} while ((skb = skb->next));
 out:
-- 
2.1.0

WARNING: multiple messages have this Message-ID (diff)
From: Xin Long <lucien.xin@gmail.com>
To: network dev <netdev@vger.kernel.org>, linux-sctp@vger.kernel.org
Cc: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>,
	Neil Horman <nhorman@tuxdriver.com>,
	Michael Tuexen <tuexen@fh-muenster.de>,
	davem@davemloft.net
Subject: [PATCHv2 net-next 03/17] udp: do checksum properly in skb_udp_tunnel_segment
Date: Thu,  8 Oct 2020 17:47:59 +0800	[thread overview]
Message-ID: <483d9eec159b22172fe04dacd58d7f88dfc2f301.1602150362.git.lucien.xin@gmail.com> (raw)
Message-ID: <20201008094759.P20cJbrkHzWa3O8yXqVqKDlcTCAAsbMDNDsVOQr_cJs@z> (raw)
In-Reply-To: <c36b016ee429980b9585144f4f9af31bcda467ee.1602150362.git.lucien.xin@gmail.com>
In-Reply-To: <cover.1602150362.git.lucien.xin@gmail.com>

This patch fixes two things:

  When skb->ip_summed == CHECKSUM_PARTIAL, skb_checksum_help() should be
  called do the checksum, instead of gso_make_checksum(), which is used
  to do the checksum for current proto after calling skb_segment(), not
  after the inner proto's gso_segment().

  When offload_csum is disabled, the hardware will not do the checksum
  for the current proto, udp. So instead of calling gso_make_checksum(),
  it should calculate checksum for udp itself.

Cc: Tom Herbert <tom@herbertland.com>
Signed-off-by: Xin Long <lucien.xin@gmail.com>
---
 net/ipv4/udp_offload.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c
index e67a66f..c0b010b 100644
--- a/net/ipv4/udp_offload.c
+++ b/net/ipv4/udp_offload.c
@@ -131,14 +131,15 @@ static struct sk_buff *__skb_udp_tunnel_segment(struct sk_buff *skb,
 		uh->check = ~csum_fold(csum_add(partial,
 				       (__force __wsum)htonl(len)));
 
-		if (skb->encapsulation || !offload_csum) {
-			uh->check = gso_make_checksum(skb, ~uh->check);
-			if (uh->check == 0)
-				uh->check = CSUM_MANGLED_0;
-		} else {
+		if (skb->encapsulation)
+			skb_checksum_help(skb);
+
+		if (offload_csum) {
 			skb->ip_summed = CHECKSUM_PARTIAL;
 			skb->csum_start = skb_transport_header(skb) - skb->head;
 			skb->csum_offset = offsetof(struct udphdr, check);
+		} else {
+			uh->check = csum_fold(skb_checksum(skb, udp_offset, len, 0));
 		}
 	} while ((skb = skb->next));
 out:
-- 
2.1.0


  parent reply	other threads:[~2020-10-08  9:47 UTC|newest]

Thread overview: 62+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-10-08  9:47 [PATCHv2 net-next 00/17] sctp: Implement RFC6951: UDP Encapsulation of SCTP Xin Long
2020-10-08  9:47 ` Xin Long
2020-10-08  9:47 ` [PATCHv2 net-next 01/17] udp: check udp sock encap_type in __udp_lib_err Xin Long
2020-10-08  9:47   ` Xin Long
2020-10-08  9:47   ` [PATCHv2 net-next 02/17] udp6: move the mss check after udp gso tunnel processing Xin Long
2020-10-08  9:47     ` Xin Long
2020-10-08  9:47     ` Xin Long [this message]
2020-10-08  9:47       ` [PATCHv2 net-next 03/17] udp: do checksum properly in skb_udp_tunnel_segment Xin Long
2020-10-08  9:48       ` [PATCHv2 net-next 04/17] udp: support sctp over udp " Xin Long
2020-10-08  9:48         ` Xin Long
2020-10-08  9:48         ` [PATCHv2 net-next 05/17] sctp: create udp4 sock and add its encap_rcv Xin Long
2020-10-08  9:48           ` Xin Long
2020-10-08  9:48           ` [PATCHv2 net-next 06/17] sctp: create udp6 sock and set " Xin Long
2020-10-08  9:48             ` Xin Long
2020-10-08  9:48             ` [PATCHv2 net-next 07/17] sctp: add encap_err_lookup for udp encap socks Xin Long
2020-10-08  9:48               ` Xin Long
2020-10-08  9:48               ` [PATCHv2 net-next 08/17] sctp: add encap_port for netns sock asoc and transport Xin Long
2020-10-08  9:48                 ` Xin Long
2020-10-08  9:48                 ` [PATCHv2 net-next 09/17] sctp: add SCTP_REMOTE_UDP_ENCAPS_PORT sockopt Xin Long
2020-10-08  9:48                   ` Xin Long
2020-10-08  9:48                   ` [PATCHv2 net-next 10/17] sctp: allow changing transport encap_port by peer packets Xin Long
2020-10-08  9:48                     ` Xin Long
2020-10-08  9:48                     ` [PATCHv2 net-next 11/17] sctp: add udphdr to overhead when udp_port is set Xin Long
2020-10-08  9:48                       ` Xin Long
2020-10-08  9:48                       ` [PATCHv2 net-next 12/17] sctp: call sk_setup_caps in sctp_packet_transmit instead Xin Long
2020-10-08  9:48                         ` Xin Long
2020-10-08  9:48                         ` [PATCHv2 net-next 13/17] sctp: support for sending packet over udp4 sock Xin Long
2020-10-08  9:48                           ` Xin Long
2020-10-08  9:48                           ` [PATCHv2 net-next 14/17] sctp: support for sending packet over udp6 sock Xin Long
2020-10-08  9:48                             ` Xin Long
2020-10-08  9:48                             ` [PATCHv2 net-next 15/17] sctp: add the error cause for new encapsulation port restart Xin Long
2020-10-08  9:48                               ` Xin Long
2020-10-08  9:48                               ` [PATCHv2 net-next 16/17] sctp: handle the init chunk matching an existing asoc Xin Long
2020-10-08  9:48                                 ` Xin Long
2020-10-08  9:48                                 ` [PATCHv2 net-next 17/17] sctp: enable udp tunneling socks Xin Long
2020-10-08  9:48                                   ` Xin Long
2020-10-08 15:46                                   ` kernel test robot
2020-10-08 15:46                                     ` kernel test robot
2020-10-09  2:02                                     ` Xin Long
2020-10-09  2:02                                       ` Xin Long
2020-10-08 14:33                               ` [PATCHv2 net-next 15/17] sctp: add the error cause for new encapsulation port restart kernel test robot
2020-10-08 14:33                                 ` kernel test robot
2020-10-08 13:10                   ` [PATCHv2 net-next 09/17] sctp: add SCTP_REMOTE_UDP_ENCAPS_PORT sockopt kernel test robot
2020-10-08 13:10                     ` kernel test robot
2020-10-08 21:18             ` [PATCHv2 net-next 06/17] sctp: create udp6 sock and set its encap_rcv kernel test robot
2020-10-08 21:18               ` kernel test robot
2020-10-09  1:59               ` Xin Long
2020-10-09  1:59                 ` Xin Long
2020-10-08 18:29           ` [PATCHv2 net-next 05/17] sctp: create udp4 sock and add " kernel test robot
2020-10-08 18:29             ` kernel test robot
2020-10-09  1:59             ` Xin Long
2020-10-09  1:59               ` Xin Long
2020-10-08 10:01         ` [PATCHv2 net-next 04/17] udp: support sctp over udp in skb_udp_tunnel_segment Xin Long
2020-10-08 10:01           ` Xin Long
2020-10-08 10:01       ` [PATCHv2 net-next 03/17] udp: do checksum properly " Xin Long
2020-10-08 10:01         ` Xin Long
2020-10-08 12:44     ` [PATCHv2 net-next 02/17] udp6: move the mss check after udp gso tunnel processing Willem de Bruijn
2020-10-08 12:44       ` Willem de Bruijn
2020-10-09  1:48       ` Xin Long
2020-10-09  1:48         ` Xin Long
2020-10-09 13:59         ` Willem de Bruijn
2020-10-09 13:59           ` Willem de Bruijn

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=483d9eec159b22172fe04dacd58d7f88dfc2f301.1602150362.git.lucien.xin@gmail.com \
    --to=lucien.xin@gmail.com \
    --cc=davem@davemloft.net \
    --cc=linux-sctp@vger.kernel.org \
    --cc=marcelo.leitner@gmail.com \
    --cc=netdev@vger.kernel.org \
    --cc=nhorman@tuxdriver.com \
    --cc=tuexen@fh-muenster.de \
    /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).