From: Xin Long <lucien.xin@gmail.com> To: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Cc: network dev <netdev@vger.kernel.org>, linux-sctp@vger.kernel.org, Neil Horman <nhorman@tuxdriver.com>, Michael Tuexen <tuexen@fh-muenster.de>, davem <davem@davemloft.net>, tom@herbertland.com Subject: Re: [PATCH net-next 03/15] udp: do checksum properly in skb_udp_tunnel_segment Date: Sat, 03 Oct 2020 07:40:05 +0000 [thread overview] Message-ID: <CADvbK_c4cStJcjLQr_fA=zo2d9GzRFy42LkeeY3_CVYu7RxGug@mail.gmail.com> (raw) In-Reply-To: <20201003040428.GC70998@localhost.localdomain> On Sat, Oct 3, 2020 at 12:04 PM Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> wrote: > > On Tue, Sep 29, 2020 at 09:48:55PM +0800, Xin Long wrote: > > 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. > > Gotta say, this is odd. It is really flipping the two around. What > about other users of this function, did you test them too? Not yet, I couldn't found other cases to trigger this. But I think gso_make_checksum() is not correct to be used here, as it's trying to calculate the checksum for inner protocol instead of UDP's. It should be skb_checksum_help(), like on the xmit path. > > It makes sense to be, but would be nice if someone else could review > this. Fix the mail of Tom Herbert, and he is the right person to review this.
WARNING: multiple messages have this Message-ID (diff)
From: Xin Long <lucien.xin@gmail.com> To: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Cc: network dev <netdev@vger.kernel.org>, linux-sctp@vger.kernel.org, Neil Horman <nhorman@tuxdriver.com>, Michael Tuexen <tuexen@fh-muenster.de>, davem <davem@davemloft.net>, tom@herbertland.com Subject: Re: [PATCH net-next 03/15] udp: do checksum properly in skb_udp_tunnel_segment Date: Sat, 3 Oct 2020 15:40:05 +0800 [thread overview] Message-ID: <CADvbK_c4cStJcjLQr_fA=zo2d9GzRFy42LkeeY3_CVYu7RxGug@mail.gmail.com> (raw) Message-ID: <20201003074005.NXkRrjLaTa5MAi2-kSBZakF15O1xupCHPGQkRp3PlvQ@z> (raw) In-Reply-To: <20201003040428.GC70998@localhost.localdomain> On Sat, Oct 3, 2020 at 12:04 PM Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> wrote: > > On Tue, Sep 29, 2020 at 09:48:55PM +0800, Xin Long wrote: > > 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. > > Gotta say, this is odd. It is really flipping the two around. What > about other users of this function, did you test them too? Not yet, I couldn't found other cases to trigger this. But I think gso_make_checksum() is not correct to be used here, as it's trying to calculate the checksum for inner protocol instead of UDP's. It should be skb_checksum_help(), like on the xmit path. > > It makes sense to be, but would be nice if someone else could review > this. Fix the mail of Tom Herbert, and he is the right person to review this.
next prev parent reply other threads:[~2020-10-03 7:40 UTC|newest] Thread overview: 82+ messages / expand[flat|nested] mbox.gz Atom feed top 2020-09-29 13:48 [PATCH net-next 00/15] sctp: Implement RFC6951: UDP Encapsulation of SCTP Xin Long 2020-09-29 13:48 ` Xin Long 2020-09-29 13:48 ` [PATCH net-next 01/15] udp: check udp sock encap_type in __udp_lib_err Xin Long 2020-09-29 13:48 ` Xin Long 2020-09-29 13:48 ` [PATCH net-next 02/15] udp6: move the mss check after udp gso tunnel processing Xin Long 2020-09-29 13:48 ` Xin Long 2020-09-29 13:48 ` [PATCH net-next 03/15] udp: do checksum properly in skb_udp_tunnel_segment Xin Long 2020-09-29 13:48 ` Xin Long 2020-09-29 13:48 ` [PATCH net-next 04/15] udp: support sctp over udp " Xin Long 2020-09-29 13:48 ` Xin Long 2020-09-29 13:48 ` [PATCH net-next 05/15] sctp: create udp4 sock and add its encap_rcv Xin Long 2020-09-29 13:48 ` Xin Long 2020-09-29 13:48 ` [PATCH net-next 06/15] sctp: create udp6 sock and set " Xin Long 2020-09-29 13:48 ` Xin Long 2020-09-29 13:48 ` [PATCH net-next 07/15] sctp: add encap_err_lookup for udp encap socks Xin Long 2020-09-29 13:48 ` Xin Long 2020-09-29 13:49 ` [PATCH net-next 08/15] sctp: add encap_port for netns sock asoc and transport Xin Long 2020-09-29 13:49 ` Xin Long 2020-09-29 13:49 ` [PATCH net-next 09/15] sctp: add SCTP_REMOTE_UDP_ENCAPS_PORT sockopt Xin Long 2020-09-29 13:49 ` Xin Long 2020-09-29 13:49 ` [PATCH net-next 10/15] sctp: allow changing transport encap_port by peer packets Xin Long 2020-09-29 13:49 ` Xin Long 2020-09-29 13:49 ` [PATCH net-next 11/15] sctp: add udphdr to overhead when udp_port is set Xin Long 2020-09-29 13:49 ` Xin Long 2020-09-29 13:49 ` [PATCH net-next 12/15] sctp: call sk_setup_caps in sctp_packet_transmit instead Xin Long 2020-09-29 13:49 ` Xin Long 2020-09-29 13:49 ` [PATCH net-next 13/15] sctp: support for sending packet over udp4 sock Xin Long 2020-09-29 13:49 ` Xin Long 2020-09-29 13:49 ` [PATCH net-next 14/15] sctp: support for sending packet over udp6 sock Xin Long 2020-09-29 13:49 ` Xin Long 2020-09-29 13:49 ` [PATCH net-next 15/15] sctp: enable udp tunneling socks Xin Long 2020-09-29 13:49 ` Xin Long 2020-10-03 4:12 ` Marcelo Ricardo Leitner 2020-10-03 4:12 ` Marcelo Ricardo Leitner 2020-10-03 8:20 ` Xin Long 2020-10-03 8:20 ` Xin Long 2020-09-29 16:25 ` [PATCH net-next 13/15] sctp: support for sending packet over udp4 sock kernel test robot 2020-09-29 16:25 ` kernel test robot 2020-09-30 6:26 ` Xin Long 2020-09-30 6:26 ` Xin Long 2020-09-29 19:19 ` kernel test robot 2020-09-29 19:19 ` kernel test robot 2020-10-03 4:09 ` [PATCH net-next 12/15] sctp: call sk_setup_caps in sctp_packet_transmit instead Marcelo Ricardo Leitner 2020-10-03 4:09 ` Marcelo Ricardo Leitner 2020-10-03 7:45 ` Xin Long 2020-10-03 7:45 ` Xin Long 2020-09-29 19:00 ` [PATCH net-next 11/15] sctp: add udphdr to overhead when udp_port is set kernel test robot 2020-09-29 19:00 ` kernel test robot 2020-10-03 4:08 ` Marcelo Ricardo Leitner 2020-10-03 4:08 ` Marcelo Ricardo Leitner 2020-10-03 7:57 ` Xin Long 2020-10-03 8:12 ` Xin Long 2020-10-03 11:23 ` Xin Long 2020-10-03 11:23 ` Xin Long 2020-10-03 12:24 ` Xin Long 2020-10-03 12:24 ` Xin Long 2020-10-05 19:01 ` Marcelo Ricardo Leitner 2020-10-05 19:01 ` Marcelo Ricardo Leitner 2020-10-05 20:08 ` Michael Tuexen 2020-10-05 20:08 ` Michael Tuexen 2020-10-08 9:37 ` Xin Long 2020-10-08 9:37 ` Xin Long 2020-10-03 4:07 ` Marcelo Ricardo Leitner 2020-10-03 4:07 ` Marcelo Ricardo Leitner 2020-10-03 7:54 ` Xin Long 2020-10-03 7:54 ` Xin Long 2020-10-03 4:06 ` [PATCH net-next 10/15] sctp: allow changing transport encap_port by peer packets Marcelo Ricardo Leitner 2020-10-03 4:06 ` Marcelo Ricardo Leitner 2020-10-03 4:05 ` [PATCH net-next 09/15] sctp: add SCTP_REMOTE_UDP_ENCAPS_PORT sockopt Marcelo Ricardo Leitner 2020-10-03 4:05 ` Marcelo Ricardo Leitner 2020-10-03 7:41 ` Xin Long 2020-10-03 7:41 ` Xin Long 2020-10-03 4:04 ` [PATCH net-next 03/15] udp: do checksum properly in skb_udp_tunnel_segment Marcelo Ricardo Leitner 2020-10-03 4:04 ` Marcelo Ricardo Leitner 2020-10-03 7:40 ` Xin Long [this message] 2020-10-03 7:40 ` Xin Long 2020-09-29 16:39 ` [PATCH net-next 00/15] sctp: Implement RFC6951: UDP Encapsulation of SCTP Michael Tuexen 2020-09-29 16:39 ` Michael Tuexen 2020-09-29 17:49 ` Xin Long 2020-09-29 18:04 ` Xin Long 2020-10-01 12:41 ` Marcelo Ricardo Leitner 2020-10-01 12:41 ` Marcelo Ricardo Leitner
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='CADvbK_c4cStJcjLQr_fA=zo2d9GzRFy42LkeeY3_CVYu7RxGug@mail.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=tom@herbertland.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: linkBe 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).