From: Mat Martineau <mathew.j.martineau@linux.intel.com>
To: netdev@vger.kernel.org
Cc: Geliang Tang <geliangtang@gmail.com>,
davem@davemloft.net, kuba@kernel.org,
matthieu.baerts@tessares.net, mptcp@lists.linux.dev,
pabeni@redhat.com,
Mat Martineau <mathew.j.martineau@linux.intel.com>
Subject: [PATCH net-next 04/16] mptcp: send out checksum for MP_CAPABLE with data
Date: Thu, 17 Jun 2021 16:46:10 -0700 [thread overview]
Message-ID: <20210617234622.472030-5-mathew.j.martineau@linux.intel.com> (raw)
In-Reply-To: <20210617234622.472030-1-mathew.j.martineau@linux.intel.com>
From: Geliang Tang <geliangtang@gmail.com>
If the checksum is enabled, send out the data checksum with the
MP_CAPABLE suboption with data.
In mptcp_established_options_mp, save the data checksum in
opts->ext_copy.csum. In mptcp_write_options, adjust the option length and
send it out with the MP_CAPABLE suboption.
Co-developed-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Geliang Tang <geliangtang@gmail.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
---
net/mptcp/options.c | 52 +++++++++++++++++++++++++++++++++++++--------
1 file changed, 43 insertions(+), 9 deletions(-)
diff --git a/net/mptcp/options.c b/net/mptcp/options.c
index bb3a1f3b6e99..b4da08db1221 100644
--- a/net/mptcp/options.c
+++ b/net/mptcp/options.c
@@ -439,6 +439,7 @@ static bool mptcp_established_options_mp(struct sock *sk, struct sk_buff *skb,
struct mptcp_sock *msk = mptcp_sk(subflow->conn);
struct mptcp_ext *mpext;
unsigned int data_len;
+ u8 len;
/* When skb is not available, we better over-estimate the emitted
* options len. A full DSS option (28 bytes) is longer than
@@ -474,10 +475,16 @@ static bool mptcp_established_options_mp(struct sock *sk, struct sk_buff *skb,
* packets that start the first subflow of an MPTCP connection,
* as well as the first packet that carries data
*/
- if (data_len > 0)
- *size = ALIGN(TCPOLEN_MPTCP_MPC_ACK_DATA, 4);
- else
+ if (data_len > 0) {
+ len = TCPOLEN_MPTCP_MPC_ACK_DATA;
+ if (opts->csum_reqd) {
+ opts->ext_copy.csum = mpext->csum;
+ len += TCPOLEN_MPTCP_DSS_CHECKSUM;
+ }
+ *size = ALIGN(len, 4);
+ } else {
*size = TCPOLEN_MPTCP_MPC_ACK;
+ }
pr_debug("subflow=%p, local_key=%llu, remote_key=%llu map_len=%d",
subflow, subflow->local_key, subflow->remote_key,
@@ -1122,6 +1129,25 @@ static void mptcp_set_rwin(const struct tcp_sock *tp)
WRITE_ONCE(msk->rcv_wnd_sent, ack_seq);
}
+static u16 mptcp_make_csum(const struct mptcp_ext *mpext)
+{
+ struct csum_pseudo_header header;
+ __wsum csum;
+
+ /* cfr RFC 8684 3.3.1.:
+ * the data sequence number used in the pseudo-header is
+ * always the 64-bit value, irrespective of what length is used in the
+ * DSS option itself.
+ */
+ header.data_seq = cpu_to_be64(mpext->data_seq);
+ header.subflow_seq = htonl(mpext->subflow_seq);
+ header.data_len = htons(mpext->data_len);
+ header.csum = 0;
+
+ csum = csum_partial(&header, sizeof(header), ~csum_unfold(mpext->csum));
+ return (__force u16)csum_fold(csum);
+}
+
void mptcp_write_options(__be32 *ptr, const struct tcp_sock *tp,
struct mptcp_out_options *opts)
{
@@ -1129,14 +1155,17 @@ void mptcp_write_options(__be32 *ptr, const struct tcp_sock *tp,
OPTION_MPTCP_MPC_ACK) & opts->suboptions) {
u8 len, flag = MPTCP_CAP_HMAC_SHA256;
- if (OPTION_MPTCP_MPC_SYN & opts->suboptions)
+ if (OPTION_MPTCP_MPC_SYN & opts->suboptions) {
len = TCPOLEN_MPTCP_MPC_SYN;
- else if (OPTION_MPTCP_MPC_SYNACK & opts->suboptions)
+ } else if (OPTION_MPTCP_MPC_SYNACK & opts->suboptions) {
len = TCPOLEN_MPTCP_MPC_SYNACK;
- else if (opts->ext_copy.data_len)
+ } else if (opts->ext_copy.data_len) {
len = TCPOLEN_MPTCP_MPC_ACK_DATA;
- else
+ if (opts->csum_reqd)
+ len += TCPOLEN_MPTCP_DSS_CHECKSUM;
+ } else {
len = TCPOLEN_MPTCP_MPC_ACK;
+ }
if (opts->csum_reqd)
flag |= MPTCP_CAP_CHECKSUM_REQD;
@@ -1159,8 +1188,13 @@ void mptcp_write_options(__be32 *ptr, const struct tcp_sock *tp,
if (!opts->ext_copy.data_len)
goto mp_capable_done;
- put_unaligned_be32(opts->ext_copy.data_len << 16 |
- TCPOPT_NOP << 8 | TCPOPT_NOP, ptr);
+ if (opts->csum_reqd) {
+ put_unaligned_be32(opts->ext_copy.data_len << 16 |
+ mptcp_make_csum(&opts->ext_copy), ptr);
+ } else {
+ put_unaligned_be32(opts->ext_copy.data_len << 16 |
+ TCPOPT_NOP << 8 | TCPOPT_NOP, ptr);
+ }
ptr += 1;
}
--
2.32.0
next prev parent reply other threads:[~2021-06-17 23:46 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-06-17 23:46 [PATCH net-next 00/16] mptcp: DSS checksum support Mat Martineau
2021-06-17 23:46 ` [PATCH net-next 01/16] mptcp: add csum_enabled in mptcp_sock Mat Martineau
2021-06-17 23:46 ` [PATCH net-next 02/16] mptcp: generate the data checksum Mat Martineau
2021-06-17 23:46 ` [PATCH net-next 03/16] mptcp: add csum_reqd in mptcp_out_options Mat Martineau
2021-06-17 23:46 ` Mat Martineau [this message]
2021-06-17 23:46 ` [PATCH net-next 05/16] mptcp: send out checksum for DSS Mat Martineau
2021-06-17 23:46 ` [PATCH net-next 06/16] mptcp: add sk parameter for mptcp_get_options Mat Martineau
2021-06-17 23:46 ` [PATCH net-next 07/16] mptcp: add csum_reqd in mptcp_options_received Mat Martineau
2021-06-17 23:46 ` [PATCH net-next 08/16] mptcp: receive checksum for MP_CAPABLE with data Mat Martineau
2021-06-17 23:46 ` [PATCH net-next 09/16] mptcp: receive checksum for DSS Mat Martineau
2021-06-17 23:46 ` [PATCH net-next 10/16] mptcp: validate the data checksum Mat Martineau
2021-06-17 23:46 ` [PATCH net-next 11/16] mptcp: tune re-injections for csum enabled mode Mat Martineau
2021-06-17 23:46 ` [PATCH net-next 12/16] mptcp: add the mib for data checksum Mat Martineau
2021-06-17 23:46 ` [PATCH net-next 13/16] mptcp: add a new sysctl checksum_enabled Mat Martineau
2021-06-17 23:46 ` [PATCH net-next 14/16] mptcp: dump csum fields in mptcp_dump_mpext Mat Martineau
2021-06-17 23:46 ` [PATCH net-next 15/16] selftests: mptcp: enable checksum in mptcp_connect.sh Mat Martineau
2021-06-17 23:46 ` [PATCH net-next 16/16] selftests: mptcp: enable checksum in mptcp_join.sh Mat Martineau
2021-06-18 18:50 ` [PATCH net-next 00/16] mptcp: DSS checksum support patchwork-bot+netdevbpf
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=20210617234622.472030-5-mathew.j.martineau@linux.intel.com \
--to=mathew.j.martineau@linux.intel.com \
--cc=davem@davemloft.net \
--cc=geliangtang@gmail.com \
--cc=kuba@kernel.org \
--cc=matthieu.baerts@tessares.net \
--cc=mptcp@lists.linux.dev \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.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).