* [PATCH v11 net-next 00/15] AccECN protocol patch series
@ 2025-07-03 5:17 chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 01/15] tcp: reorganize SYN ECN code chia-yu.chang
` (14 more replies)
0 siblings, 15 replies; 16+ messages in thread
From: chia-yu.chang @ 2025-07-03 5:17 UTC (permalink / raw)
To: pabeni, edumazet, linux-doc, corbet, horms, dsahern, kuniyu, bpf,
netdev, dave.taht, jhs, kuba, stephen, xiyou.wangcong, jiri,
davem, andrew+netdev, donald.hunter, ast, liuhangbin, shuah,
linux-kselftest, ij, ncardwell, koen.de_schepper, g.white,
ingemar.s.johansson, mirja.kuehlewind, cheshire, rs.ietf,
Jason_Livingood, vidhi_goel
Cc: Chia-Yu Chang
From: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
Hello,
Please find the v10 AccECN protocol patch series, which covers the core
functionality of Accurate ECN, AccECN negotiation, AccECN TCP options,
and AccECN failure handling. The Accurate ECN draft can be found in
https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-accurate-ecn-28
This patch series is part of the full AccECN patch series, which is available at
https://github.com/L4STeam/linux-net-next/commits/upstream_l4steam/
Best Regards,
Chia-Yu
---
v11 (04-Jul-2025)
- Fix compilation issue of some intermediate patches in v10
v10 (03-Jul-2025)
- Add new patch of separated header file include/net/tcp_ecn.h to include ECN and AccECN functions (Eric Dumazet <edumazet@google.com>)
- Add comments on the AccECN helper functions in tcp_ecn.h (Eric Dumazet <edumazet@google.com>)
- Add documentation of tcp_ecn, tcp_ecn_option, tcp_ecn_beacon in ip-sysctl.rst to the corresponding patch (Eric Dumazet <edumazet@google.com>)
- Split wait third ACK functionality into a separated patch from AccECN negotiation patch (Eric Dumazet <edumazet@google.com>)
- Add READ_ONCE() over every reads of sysctl for all patches in the series (Eric Dumazet <edumazet@google.com>)
- Merge heuristics of AccECN option ceb/cep and ACE field multi-wrap into a single patch
- Add a table of SACK block reduction and required AccECN field in patch #15 commit message (Eric Dumazet <edumazet@google.com>)
v9 (21-Jun-2025)
- Use tcp_data_ecn_check() to set TCP_ECN_SEE flag only for RFC3168 ECN (Paolo Abeni <pabeni@redhat.com>)
- Add comments about setting TCP_ECN_SEEN flag for RFC3168 and Accruate ECN (Paolo Abeni <pabeni@redhat.com>)
- Restruct the code in the for loop of tcp_accecn_process_option() (Paolo Abeni <pabeni@redhat.com>)
- Remove ecn_bytes and add use_synack_ecn_bytes flag to identify whether syn_ack_bytes or received_ecn_bytes is used (Paolo Abeni <pabeni@redhat.com>)
- Replace leftover_bytes and leftover_size with leftover_highbyte and leftover_lowbyte and add comments in tcp_options_write() (Paolo Abeni <pabeni@redhat.com>)
- Add comments and commit message about the 1st retx SYN still attempt AccECN negotiation (Paolo Abeni <pabeni@redhat.com>)
v8 (10-Jun-2025)
- Add new helper function tcp_ecn_received_counters_payload() in #6 (Paolo Abeni <pabeni@redhat.com>)
- Set opts->num_sack_blocks=0 to avoid potential undefined value in #8 (Paolo Abeni <pabeni@redhat.com>)
- Reset leftover_size to 2 once leftover_bytes is used in #9 (Paolo Abeni <pabeni@redhat.com>)
- Add new helper function tcp_accecn_opt_demand_min() in #10 (Paolo Abeni <pabeni@redhat.com>)
- Add new helper function tcp_accecn_saw_opt_fail_recv() in #11 (Paolo Abeni <pabeni@redhat.com>)
- Update tcp_options_fit_accecn() to avoid using recursion in #14 (Paolo Abeni <pabeni@redhat.com>)
v7 (14-May-2025)
- Modify group sizes of tcp_sock_write_txrx and tcp_sock_write_rx in #3 based on pahole results (Paolo Abeni <pabeni@redhat.com>)
- Fix the issue in #4 and #5 where the RFC3168 ECN behavior in tcp_ecn_send() is changed (Paolo Abeni <pabeni@redhat.com>)
- Modify group size of tcp_sock_write_txrx in #4 and #6 based on pahole results (Paolo Abeni <pabeni@redhat.com>)
- Update commit message for #9 to explain the increase in tcp_sock_write_rx group size
- Modify group size of tcp_sock_write_tx in #10 based on pahole results
v6 (09-May-2025)
- Add #3 to utilize exisintg holes of tcp_sock_write_txrx group for later patches (#4, #9, #10) with new u8 members (Paolo Abeni <pabeni@redhat.com>)
- Add pahole outcomes before and after commit in #4, #5, #6, #9, #10, #15 (Paolo Abeni <pabeni@redhat.com>)
- Define new helper function tcp_send_ack_reflect_ect() for sending ACK with reflected ECT in #5 (Paolo Abeni <pabeni@redhat.com>)
- Add comments for function tcp_ecn_rcv_synack() in #5 (Paolo Abeni <pabeni@redhat.com>)
- Add enum/define to be used by sysctl_tcp_ecn in #5, sysctl_tcp_ecn_option in #9, and sysctl_tcp_ecn_option_beacon in #10 (Paolo Abeni <pabeni@redhat.com>)
- Move accecn_fail_mode and saw_accecn_opt in #5 and #11 to use exisintg holes of tcp_sock (Paolo Abeni <pabeni@redhat.com>)
- Change data type of new members of tcp_request_sock and move them to the end of struct in #5 and #11 (Paolo Abeni <pabeni@redhat.com>)
- Move new members of tcp_info to the end of struct in #6 (Paolo Abeni <pabeni@redhat.com>)
- Merge previous #7 into #9 (Paolo Abeni <pabeni@redhat.com>)
- Mask ecnfield with INET_ECN_MASK to remove WARN_ONCE in #9 (Paolo Abeni <pabeni@redhat.com>)
- Reduce the indentation levels for reabability in #9 and #10 (Paolo Abeni <pabeni@redhat.com>)
- Move delivered_ecn_bytes to the RX group in #9, accecn_opt_tstamp to the TX group in #10, pkts_acked_ewma to the RX group in #15 (Paolo Abeni <pabeni@redhat.com>)
- Add changes in Documentation/networking/net_cachelines/tcp_sock.rst for new tcp_sock members in #3, #5, #6, #9, #10, #15
v5 (22-Apr-2025)
- Further fix for 32-bit ARM alignment in tcp.c (Simon Horman <horms@kernel.org>)
v4 (18-Apr-2025)
- Fix 32-bit ARM assertion for alignment requirement (Simon Horman <horms@kernel.org>)
v3 (14-Apr-2025)
- Fix patch apply issue in v2 (Jakub Kicinski <kuba@kernel.org>)
v2 (18-Mar-2025)
- Add one missing patch from the previous AccECN protocol preparation patch series to this patch series.
---
Chia-Yu Chang (6):
tcp: reorganize tcp_sock_write_txrx group for variables later
tcp: ecn functions in separated include file
tcp: Add wait_third_ack flag for ECN negotiation in simultaneous
connect
tcp: accecn: AccECN option send control
tcp: accecn: AccECN option failure handling
tcp: accecn: try to fit AccECN option with SACK
Ilpo Järvinen (9):
tcp: reorganize SYN ECN code
tcp: fast path functions later
tcp: AccECN core
tcp: accecn: AccECN negotiation
tcp: accecn: add AccECN rx byte counters
tcp: accecn: AccECN needs to know delivered bytes
tcp: sack option handling improvements
tcp: accecn: AccECN option
tcp: accecn: AccECN option ceb/cep and ACE field multi-wrap heuristics
Documentation/networking/ip-sysctl.rst | 55 +-
.../networking/net_cachelines/tcp_sock.rst | 13 +
include/linux/tcp.h | 33 +-
include/net/netns/ipv4.h | 2 +
include/net/tcp.h | 87 ++-
include/net/tcp_ecn.h | 648 ++++++++++++++++++
include/uapi/linux/tcp.h | 7 +
net/ipv4/syncookies.c | 4 +
net/ipv4/sysctl_net_ipv4.c | 19 +
net/ipv4/tcp.c | 29 +-
net/ipv4/tcp_input.c | 371 ++++++++--
net/ipv4/tcp_ipv4.c | 8 +-
net/ipv4/tcp_minisocks.c | 40 +-
net/ipv4/tcp_output.c | 297 ++++++--
net/ipv6/syncookies.c | 2 +
net/ipv6/tcp_ipv6.c | 1 +
16 files changed, 1429 insertions(+), 187 deletions(-)
create mode 100644 include/net/tcp_ecn.h
--
2.34.1
^ permalink raw reply [flat|nested] 16+ messages in thread
* [PATCH v11 net-next 01/15] tcp: reorganize SYN ECN code
2025-07-03 5:17 [PATCH v11 net-next 00/15] AccECN protocol patch series chia-yu.chang
@ 2025-07-03 5:17 ` chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 02/15] tcp: fast path functions later chia-yu.chang
` (13 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: chia-yu.chang @ 2025-07-03 5:17 UTC (permalink / raw)
To: pabeni, edumazet, linux-doc, corbet, horms, dsahern, kuniyu, bpf,
netdev, dave.taht, jhs, kuba, stephen, xiyou.wangcong, jiri,
davem, andrew+netdev, donald.hunter, ast, liuhangbin, shuah,
linux-kselftest, ij, ncardwell, koen.de_schepper, g.white,
ingemar.s.johansson, mirja.kuehlewind, cheshire, rs.ietf,
Jason_Livingood, vidhi_goel
Cc: Chia-Yu Chang
From: Ilpo Järvinen <ij@kernel.org>
Prepare for AccECN that needs to have access here on IP ECN
field value which is only available after INET_ECN_xmit().
No functional changes.
Signed-off-by: Ilpo Järvinen <ij@kernel.org>
Signed-off-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
---
net/ipv4/tcp_output.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index b616776e3354..caa980c3f25a 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -350,10 +350,11 @@ static void tcp_ecn_send_syn(struct sock *sk, struct sk_buff *skb)
tp->ecn_flags = 0;
if (use_ecn) {
- TCP_SKB_CB(skb)->tcp_flags |= TCPHDR_ECE | TCPHDR_CWR;
- tcp_ecn_mode_set(tp, TCP_ECN_MODE_RFC3168);
if (tcp_ca_needs_ecn(sk) || bpf_needs_ecn)
INET_ECN_xmit(sk);
+
+ TCP_SKB_CB(skb)->tcp_flags |= TCPHDR_ECE | TCPHDR_CWR;
+ tcp_ecn_mode_set(tp, TCP_ECN_MODE_RFC3168);
}
}
--
2.34.1
^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH v11 net-next 02/15] tcp: fast path functions later
2025-07-03 5:17 [PATCH v11 net-next 00/15] AccECN protocol patch series chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 01/15] tcp: reorganize SYN ECN code chia-yu.chang
@ 2025-07-03 5:17 ` chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 03/15] tcp: reorganize tcp_sock_write_txrx group for variables later chia-yu.chang
` (12 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: chia-yu.chang @ 2025-07-03 5:17 UTC (permalink / raw)
To: pabeni, edumazet, linux-doc, corbet, horms, dsahern, kuniyu, bpf,
netdev, dave.taht, jhs, kuba, stephen, xiyou.wangcong, jiri,
davem, andrew+netdev, donald.hunter, ast, liuhangbin, shuah,
linux-kselftest, ij, ncardwell, koen.de_schepper, g.white,
ingemar.s.johansson, mirja.kuehlewind, cheshire, rs.ietf,
Jason_Livingood, vidhi_goel
Cc: Chia-Yu Chang
From: Ilpo Järvinen <ij@kernel.org>
The following patch will use tcp_ecn_mode_accecn(),
TCP_ACCECN_CEP_INIT_OFFSET, TCP_ACCECN_CEP_ACE_MASK in
__tcp_fast_path_on() to make new flag for AccECN.
No functional changes.
Signed-off-by: Ilpo Järvinen <ij@kernel.org>
Signed-off-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
---
include/net/tcp.h | 54 +++++++++++++++++++++++------------------------
1 file changed, 27 insertions(+), 27 deletions(-)
diff --git a/include/net/tcp.h b/include/net/tcp.h
index bc08de49805c..b462f8c3e85b 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -811,33 +811,6 @@ static inline u32 __tcp_set_rto(const struct tcp_sock *tp)
return usecs_to_jiffies((tp->srtt_us >> 3) + tp->rttvar_us);
}
-static inline void __tcp_fast_path_on(struct tcp_sock *tp, u32 snd_wnd)
-{
- /* mptcp hooks are only on the slow path */
- if (sk_is_mptcp((struct sock *)tp))
- return;
-
- tp->pred_flags = htonl((tp->tcp_header_len << 26) |
- ntohl(TCP_FLAG_ACK) |
- snd_wnd);
-}
-
-static inline void tcp_fast_path_on(struct tcp_sock *tp)
-{
- __tcp_fast_path_on(tp, tp->snd_wnd >> tp->rx_opt.snd_wscale);
-}
-
-static inline void tcp_fast_path_check(struct sock *sk)
-{
- struct tcp_sock *tp = tcp_sk(sk);
-
- if (RB_EMPTY_ROOT(&tp->out_of_order_queue) &&
- tp->rcv_wnd &&
- atomic_read(&sk->sk_rmem_alloc) < sk->sk_rcvbuf &&
- !tp->urg_data)
- tcp_fast_path_on(tp);
-}
-
u32 tcp_delack_max(const struct sock *sk);
/* Compute the actual rto_min value */
@@ -1797,6 +1770,33 @@ static inline bool tcp_paws_reject(const struct tcp_options_received *rx_opt,
return true;
}
+static inline void __tcp_fast_path_on(struct tcp_sock *tp, u32 snd_wnd)
+{
+ /* mptcp hooks are only on the slow path */
+ if (sk_is_mptcp((struct sock *)tp))
+ return;
+
+ tp->pred_flags = htonl((tp->tcp_header_len << 26) |
+ ntohl(TCP_FLAG_ACK) |
+ snd_wnd);
+}
+
+static inline void tcp_fast_path_on(struct tcp_sock *tp)
+{
+ __tcp_fast_path_on(tp, tp->snd_wnd >> tp->rx_opt.snd_wscale);
+}
+
+static inline void tcp_fast_path_check(struct sock *sk)
+{
+ struct tcp_sock *tp = tcp_sk(sk);
+
+ if (RB_EMPTY_ROOT(&tp->out_of_order_queue) &&
+ tp->rcv_wnd &&
+ atomic_read(&sk->sk_rmem_alloc) < sk->sk_rcvbuf &&
+ !tp->urg_data)
+ tcp_fast_path_on(tp);
+}
+
bool tcp_oow_rate_limited(struct net *net, const struct sk_buff *skb,
int mib_idx, u32 *last_oow_ack_time);
--
2.34.1
^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH v11 net-next 03/15] tcp: reorganize tcp_sock_write_txrx group for variables later
2025-07-03 5:17 [PATCH v11 net-next 00/15] AccECN protocol patch series chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 01/15] tcp: reorganize SYN ECN code chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 02/15] tcp: fast path functions later chia-yu.chang
@ 2025-07-03 5:17 ` chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 04/15] tcp: ecn functions in separated include file chia-yu.chang
` (11 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: chia-yu.chang @ 2025-07-03 5:17 UTC (permalink / raw)
To: pabeni, edumazet, linux-doc, corbet, horms, dsahern, kuniyu, bpf,
netdev, dave.taht, jhs, kuba, stephen, xiyou.wangcong, jiri,
davem, andrew+netdev, donald.hunter, ast, liuhangbin, shuah,
linux-kselftest, ij, ncardwell, koen.de_schepper, g.white,
ingemar.s.johansson, mirja.kuehlewind, cheshire, rs.ietf,
Jason_Livingood, vidhi_goel
Cc: Chia-Yu Chang
From: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
Use the first 3-byte hole at the beginning of the tcp_sock_write_txrx
group for 'noneagle'/'rate_app_limited' to fill in the existing hole
in later patches. Therefore, the group size of tcp_sock_write_txrx is
reduced from 92 + 4 to 91 + 4. In addition, the group size of
tcp_sock_write_rx is changed to 96 to fit in the pahole outcome.
Below are the trimmed pahole outcomes before and after this patch:
[BEFORE THIS PATCH]
struct tcp_sock {
[...]
__cacheline_group_begin__tcp_sock_write_txrx[0]; /* 2585 0 */
/* XXX 3 bytes hole, try to pack */
[...]
struct tcp_options_received rx_opt; /* 2652 24 */
u8 nonagle:4; /* 2676: 0 1 */
u8 rate_app_limited:1; /* 2676: 4 1 */
/* XXX 3 bits hole, try to pack */
__cacheline_group_end__tcp_sock_write_txrx[0]; /* 2677 0 */
/* XXX 3 bytes hole, try to pack */
__cacheline_group_begin__tcp_sock_write_rx[0] __attribute__((__aligned__(8))); /* 2680 0 */
[...]
__cacheline_group_end__tcp_sock_write_rx[0]; /* 2776 0 */
[...]
/* size: 3264, cachelines: 51, members: 163 */
}
[AFTER THIS PATCH]
struct tcp_sock {
[...]
__cacheline_group_begin__tcp_sock_write_txrx[0]; /* 2585 0 */
u8 nonagle:4; /* 2585: 0 1 */
u8 rate_app_limited:1; /* 2585: 4 1 */
/* XXX 3 bits hole, try to pack */
/* XXX 2 bytes hole, try to pack */
[...]
struct tcp_options_received rx_opt; /* 2652 24 */
__cacheline_group_end__tcp_sock_write_txrx[0]; /* 2676 0 */
/* XXX 4 bytes hole, try to pack */
__cacheline_group_begin__tcp_sock_write_rx[0] __attribute__((__aligned__(8))); /* 2680 0 */
[...]
__cacheline_group_end__tcp_sock_write_rx[0]; /* 2776 0 */
[...]
/* size: 3264, cachelines: 51, members: 163 */
}
Signed-off-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
---
include/linux/tcp.h | 4 ++--
net/ipv4/tcp.c | 4 ++--
2 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 1a5737b3753d..bd692764d039 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -285,6 +285,8 @@ struct tcp_sock {
* Header prediction flags
* 0x5?10 << 16 + snd_wnd in net byte order
*/
+ u8 nonagle : 4,/* Disable Nagle algorithm? */
+ rate_app_limited:1; /* rate_{delivered,interval_us} limited? */
__be32 pred_flags;
u64 tcp_clock_cache; /* cache last tcp_clock_ns() (see tcp_mstamp_refresh()) */
u64 tcp_mstamp; /* most recent packet received/sent */
@@ -303,8 +305,6 @@ struct tcp_sock {
* Options received (usually on last packet, some only on SYN packets).
*/
struct tcp_options_received rx_opt;
- u8 nonagle : 4,/* Disable Nagle algorithm? */
- rate_app_limited:1; /* rate_{delivered,interval_us} limited? */
__cacheline_group_end(tcp_sock_write_txrx);
/* RX read-write hotpath cache lines */
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 925b2c572ca2..ef5574795174 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -5123,7 +5123,7 @@ static void __init tcp_struct_check(void)
/* 32bit arches with 8byte alignment on u64 fields might need padding
* before tcp_clock_cache.
*/
- CACHELINE_ASSERT_GROUP_SIZE(struct tcp_sock, tcp_sock_write_txrx, 92 + 4);
+ CACHELINE_ASSERT_GROUP_SIZE(struct tcp_sock, tcp_sock_write_txrx, 91 + 4);
/* RX read-write hotpath cache lines */
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_rx, bytes_received);
@@ -5140,7 +5140,7 @@ static void __init tcp_struct_check(void)
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_rx, bytes_acked);
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_rx, rcv_rtt_est);
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_rx, rcvq_space);
- CACHELINE_ASSERT_GROUP_SIZE(struct tcp_sock, tcp_sock_write_rx, 99);
+ CACHELINE_ASSERT_GROUP_SIZE(struct tcp_sock, tcp_sock_write_rx, 96);
}
void __init tcp_init(void)
--
2.34.1
^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH v11 net-next 04/15] tcp: ecn functions in separated include file
2025-07-03 5:17 [PATCH v11 net-next 00/15] AccECN protocol patch series chia-yu.chang
` (2 preceding siblings ...)
2025-07-03 5:17 ` [PATCH v11 net-next 03/15] tcp: reorganize tcp_sock_write_txrx group for variables later chia-yu.chang
@ 2025-07-03 5:17 ` chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 05/15] tcp: AccECN core chia-yu.chang
` (10 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: chia-yu.chang @ 2025-07-03 5:17 UTC (permalink / raw)
To: pabeni, edumazet, linux-doc, corbet, horms, dsahern, kuniyu, bpf,
netdev, dave.taht, jhs, kuba, stephen, xiyou.wangcong, jiri,
davem, andrew+netdev, donald.hunter, ast, liuhangbin, shuah,
linux-kselftest, ij, ncardwell, koen.de_schepper, g.white,
ingemar.s.johansson, mirja.kuehlewind, cheshire, rs.ietf,
Jason_Livingood, vidhi_goel
Cc: Chia-Yu Chang
From: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
The following patches will modify ECN helpers and add AccECN herlpers,
and this patch moves the existing ones into a separated include file.
No functional changes.
Signed-off-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
---
include/net/tcp_ecn.h | 116 ++++++++++++++++++++++++++++++++++++++++++
net/ipv4/tcp_input.c | 45 +---------------
net/ipv4/tcp_output.c | 56 +-------------------
3 files changed, 118 insertions(+), 99 deletions(-)
create mode 100644 include/net/tcp_ecn.h
diff --git a/include/net/tcp_ecn.h b/include/net/tcp_ecn.h
new file mode 100644
index 000000000000..b3430557676b
--- /dev/null
+++ b/include/net/tcp_ecn.h
@@ -0,0 +1,116 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+#ifndef _TCP_ECN_H
+#define _TCP_ECN_H
+
+#include <linux/tcp.h>
+#include <linux/skbuff.h>
+
+#include <net/inet_connection_sock.h>
+#include <net/sock.h>
+#include <net/tcp.h>
+#include <net/inet_ecn.h>
+
+static inline void tcp_ecn_queue_cwr(struct tcp_sock *tp)
+{
+ if (tcp_ecn_mode_rfc3168(tp))
+ tp->ecn_flags |= TCP_ECN_QUEUE_CWR;
+}
+
+static inline void tcp_ecn_accept_cwr(struct sock *sk,
+ const struct sk_buff *skb)
+{
+ if (tcp_hdr(skb)->cwr) {
+ tcp_sk(sk)->ecn_flags &= ~TCP_ECN_DEMAND_CWR;
+
+ /* If the sender is telling us it has entered CWR, then its
+ * cwnd may be very low (even just 1 packet), so we should ACK
+ * immediately.
+ */
+ if (TCP_SKB_CB(skb)->seq != TCP_SKB_CB(skb)->end_seq)
+ inet_csk(sk)->icsk_ack.pending |= ICSK_ACK_NOW;
+ }
+}
+
+static inline void tcp_ecn_withdraw_cwr(struct tcp_sock *tp)
+{
+ tp->ecn_flags &= ~TCP_ECN_QUEUE_CWR;
+}
+
+static inline void tcp_ecn_rcv_synack(struct tcp_sock *tp,
+ const struct tcphdr *th)
+{
+ if (tcp_ecn_mode_rfc3168(tp) && (!th->ece || th->cwr))
+ tcp_ecn_mode_set(tp, TCP_ECN_DISABLED);
+}
+
+static inline void tcp_ecn_rcv_syn(struct tcp_sock *tp,
+ const struct tcphdr *th)
+{
+ if (tcp_ecn_mode_rfc3168(tp) && (!th->ece || !th->cwr))
+ tcp_ecn_mode_set(tp, TCP_ECN_DISABLED);
+}
+
+static inline bool tcp_ecn_rcv_ecn_echo(const struct tcp_sock *tp,
+ const struct tcphdr *th)
+{
+ if (th->ece && !th->syn && tcp_ecn_mode_rfc3168(tp))
+ return true;
+ return false;
+}
+
+/* Packet ECN state for a SYN-ACK */
+static inline void tcp_ecn_send_synack(struct sock *sk, struct sk_buff *skb)
+{
+ const struct tcp_sock *tp = tcp_sk(sk);
+
+ TCP_SKB_CB(skb)->tcp_flags &= ~TCPHDR_CWR;
+ if (tcp_ecn_disabled(tp))
+ TCP_SKB_CB(skb)->tcp_flags &= ~TCPHDR_ECE;
+ else if (tcp_ca_needs_ecn(sk) ||
+ tcp_bpf_ca_needs_ecn(sk))
+ INET_ECN_xmit(sk);
+}
+
+/* Packet ECN state for a SYN. */
+static inline void tcp_ecn_send_syn(struct sock *sk, struct sk_buff *skb)
+{
+ struct tcp_sock *tp = tcp_sk(sk);
+ bool bpf_needs_ecn = tcp_bpf_ca_needs_ecn(sk);
+ bool use_ecn = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_ecn) == 1 ||
+ tcp_ca_needs_ecn(sk) || bpf_needs_ecn;
+
+ if (!use_ecn) {
+ const struct dst_entry *dst = __sk_dst_get(sk);
+
+ if (dst && dst_feature(dst, RTAX_FEATURE_ECN))
+ use_ecn = true;
+ }
+
+ tp->ecn_flags = 0;
+
+ if (use_ecn) {
+ if (tcp_ca_needs_ecn(sk) || bpf_needs_ecn)
+ INET_ECN_xmit(sk);
+
+ TCP_SKB_CB(skb)->tcp_flags |= TCPHDR_ECE | TCPHDR_CWR;
+ tcp_ecn_mode_set(tp, TCP_ECN_MODE_RFC3168);
+ }
+}
+
+static inline void tcp_ecn_clear_syn(struct sock *sk, struct sk_buff *skb)
+{
+ if (READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_ecn_fallback))
+ /* tp->ecn_flags are cleared at a later point in time when
+ * SYN ACK is ultimatively being received.
+ */
+ TCP_SKB_CB(skb)->tcp_flags &= ~(TCPHDR_ECE | TCPHDR_CWR);
+}
+
+static inline void
+tcp_ecn_make_synack(const struct request_sock *req, struct tcphdr *th)
+{
+ if (inet_rsk(req)->ecn_ok)
+ th->ece = 1;
+}
+
+#endif /* _LINUX_TCP_ECN_H */
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 79e3bfb0108f..da4245718e63 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -72,6 +72,7 @@
#include <linux/prefetch.h>
#include <net/dst.h>
#include <net/tcp.h>
+#include <net/tcp_ecn.h>
#include <net/proto_memory.h>
#include <net/inet_common.h>
#include <linux/ipsec.h>
@@ -339,31 +340,6 @@ static bool tcp_in_quickack_mode(struct sock *sk)
(icsk->icsk_ack.quick && !inet_csk_in_pingpong_mode(sk));
}
-static void tcp_ecn_queue_cwr(struct tcp_sock *tp)
-{
- if (tcp_ecn_mode_rfc3168(tp))
- tp->ecn_flags |= TCP_ECN_QUEUE_CWR;
-}
-
-static void tcp_ecn_accept_cwr(struct sock *sk, const struct sk_buff *skb)
-{
- if (tcp_hdr(skb)->cwr) {
- tcp_sk(sk)->ecn_flags &= ~TCP_ECN_DEMAND_CWR;
-
- /* If the sender is telling us it has entered CWR, then its
- * cwnd may be very low (even just 1 packet), so we should ACK
- * immediately.
- */
- if (TCP_SKB_CB(skb)->seq != TCP_SKB_CB(skb)->end_seq)
- inet_csk(sk)->icsk_ack.pending |= ICSK_ACK_NOW;
- }
-}
-
-static void tcp_ecn_withdraw_cwr(struct tcp_sock *tp)
-{
- tp->ecn_flags &= ~TCP_ECN_QUEUE_CWR;
-}
-
static void tcp_data_ecn_check(struct sock *sk, const struct sk_buff *skb)
{
struct tcp_sock *tp = tcp_sk(sk);
@@ -399,25 +375,6 @@ static void tcp_data_ecn_check(struct sock *sk, const struct sk_buff *skb)
}
}
-static void tcp_ecn_rcv_synack(struct tcp_sock *tp, const struct tcphdr *th)
-{
- if (tcp_ecn_mode_rfc3168(tp) && (!th->ece || th->cwr))
- tcp_ecn_mode_set(tp, TCP_ECN_DISABLED);
-}
-
-static void tcp_ecn_rcv_syn(struct tcp_sock *tp, const struct tcphdr *th)
-{
- if (tcp_ecn_mode_rfc3168(tp) && (!th->ece || !th->cwr))
- tcp_ecn_mode_set(tp, TCP_ECN_DISABLED);
-}
-
-static bool tcp_ecn_rcv_ecn_echo(const struct tcp_sock *tp, const struct tcphdr *th)
-{
- if (th->ece && !th->syn && tcp_ecn_mode_rfc3168(tp))
- return true;
- return false;
-}
-
static void tcp_count_delivered_ce(struct tcp_sock *tp, u32 ecn_count)
{
tp->delivered_ce += ecn_count;
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index caa980c3f25a..2f1102616024 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -38,6 +38,7 @@
#define pr_fmt(fmt) "TCP: " fmt
#include <net/tcp.h>
+#include <net/tcp_ecn.h>
#include <net/mptcp.h>
#include <net/proto_memory.h>
@@ -319,61 +320,6 @@ static u16 tcp_select_window(struct sock *sk)
return new_win;
}
-/* Packet ECN state for a SYN-ACK */
-static void tcp_ecn_send_synack(struct sock *sk, struct sk_buff *skb)
-{
- const struct tcp_sock *tp = tcp_sk(sk);
-
- TCP_SKB_CB(skb)->tcp_flags &= ~TCPHDR_CWR;
- if (tcp_ecn_disabled(tp))
- TCP_SKB_CB(skb)->tcp_flags &= ~TCPHDR_ECE;
- else if (tcp_ca_needs_ecn(sk) ||
- tcp_bpf_ca_needs_ecn(sk))
- INET_ECN_xmit(sk);
-}
-
-/* Packet ECN state for a SYN. */
-static void tcp_ecn_send_syn(struct sock *sk, struct sk_buff *skb)
-{
- struct tcp_sock *tp = tcp_sk(sk);
- bool bpf_needs_ecn = tcp_bpf_ca_needs_ecn(sk);
- bool use_ecn = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_ecn) == 1 ||
- tcp_ca_needs_ecn(sk) || bpf_needs_ecn;
-
- if (!use_ecn) {
- const struct dst_entry *dst = __sk_dst_get(sk);
-
- if (dst && dst_feature(dst, RTAX_FEATURE_ECN))
- use_ecn = true;
- }
-
- tp->ecn_flags = 0;
-
- if (use_ecn) {
- if (tcp_ca_needs_ecn(sk) || bpf_needs_ecn)
- INET_ECN_xmit(sk);
-
- TCP_SKB_CB(skb)->tcp_flags |= TCPHDR_ECE | TCPHDR_CWR;
- tcp_ecn_mode_set(tp, TCP_ECN_MODE_RFC3168);
- }
-}
-
-static void tcp_ecn_clear_syn(struct sock *sk, struct sk_buff *skb)
-{
- if (READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_ecn_fallback))
- /* tp->ecn_flags are cleared at a later point in time when
- * SYN ACK is ultimatively being received.
- */
- TCP_SKB_CB(skb)->tcp_flags &= ~(TCPHDR_ECE | TCPHDR_CWR);
-}
-
-static void
-tcp_ecn_make_synack(const struct request_sock *req, struct tcphdr *th)
-{
- if (inet_rsk(req)->ecn_ok)
- th->ece = 1;
-}
-
/* Set up ECN state for a packet on a ESTABLISHED socket that is about to
* be sent.
*/
--
2.34.1
^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH v11 net-next 05/15] tcp: AccECN core
2025-07-03 5:17 [PATCH v11 net-next 00/15] AccECN protocol patch series chia-yu.chang
` (3 preceding siblings ...)
2025-07-03 5:17 ` [PATCH v11 net-next 04/15] tcp: ecn functions in separated include file chia-yu.chang
@ 2025-07-03 5:17 ` chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 06/15] tcp: accecn: AccECN negotiation chia-yu.chang
` (9 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: chia-yu.chang @ 2025-07-03 5:17 UTC (permalink / raw)
To: pabeni, edumazet, linux-doc, corbet, horms, dsahern, kuniyu, bpf,
netdev, dave.taht, jhs, kuba, stephen, xiyou.wangcong, jiri,
davem, andrew+netdev, donald.hunter, ast, liuhangbin, shuah,
linux-kselftest, ij, ncardwell, koen.de_schepper, g.white,
ingemar.s.johansson, mirja.kuehlewind, cheshire, rs.ietf,
Jason_Livingood, vidhi_goel
Cc: Olivier Tilmans, Chia-Yu Chang
From: Ilpo Järvinen <ij@kernel.org>
This change implements Accurate ECN without negotiation and
AccECN Option (that will be added by later changes). Based on
AccECN specifications:
https://tools.ietf.org/id/draft-ietf-tcpm-accurate-ecn-28.txt
Accurate ECN allows feeding back the number of CE (congestion
experienced) marks accurately to the sender in contrast to
RFC3168 ECN that can only signal one marks-seen-yes/no per RTT.
Congestion control algorithms can take advantage of the accurate
ECN information to fine-tune their congestion response to avoid
drastic rate reduction when only mild congestion is encountered.
With Accurate ECN, tp->received_ce (r.cep in AccECN spec) keeps
track of how many segments have arrived with a CE mark. Accurate
ECN uses ACE field (ECE, CWR, AE) to communicate the value back
to the sender which updates tp->delivered_ce (s.cep) based on the
feedback. This signalling channel is lossy when ACE field overflow
occurs.
Conservative strategy is selected here to deal with the ACE
overflow, however, some strategies using the AccECN option later
in the overall patchset mitigate against false overflows detected.
The ACE field values on the wire are offset by
TCP_ACCECN_CEP_INIT_OFFSET. Delivered_ce/received_ce count the
real CE marks rather than forcing all downstream users to adapt
to the wire offset.
This patch uses the first 1-byte hole and the last 4-byte hole of
the tcp_sock_write_txrx for 'received_ce_pending' and 'received_ce'.
Also, the group size of tcp_sock_write_txrx is increased from
91 + 4 to 95 + 4 due to the new u32 received_ce member. Below are
the trimmed pahole outcomes before and after this patch.
[BEFORE THIS PATCH]
struct tcp_sock {
[...]
__cacheline_group_begin__tcp_sock_write_txrx[0]; /* 2585 0 */
u8 nonagle:4; /* 2585: 0 1 */
u8 rate_app_limited:1; /* 2585: 4 1 */
/* XXX 3 bits hole, try to pack */
/* XXX 2 bytes hole, try to pack */
[...]
u32 delivered_ce; /* 2640 4 */
u32 app_limited; /* 2644 4 */
u32 rcv_wnd; /* 2648 4 */
struct tcp_options_received rx_opt; /* 2652 24 */
__cacheline_group_end__tcp_sock_write_txrx[0]; /* 2676 0 */
/* XXX 4 bytes hole, try to pack */
[...]
/* size: 3264, cachelines: 51, members: 163 */
}
[AFTER THIS PATCH]
struct tcp_sock {
[...]
__cacheline_group_begin__tcp_sock_write_txrx[0]; /* 2585 0 */
u8 nonagle:4; /* 2585: 0 1 */
u8 rate_app_limited:1; /* 2585: 4 1 */
/* XXX 3 bits hole, try to pack */
/* Force alignment to the next boundary: */
u8 :0;
u8 received_ce_pending:4;/* 2586: 0 1 */
u8 unused2:4; /* 2586: 4 1 */
/* XXX 1 byte hole, try to pack */
[...]
u32 delivered_ce; /* 2640 4 */
u32 received_ce; /* 2644 4 */
u32 app_limited; /* 2648 4 */
u32 rcv_wnd; /* 2652 4 */
struct tcp_options_received rx_opt; /* 2656 24 */
__cacheline_group_end__tcp_sock_write_txrx[0]; /* 2680 0 */
[...]
/* size: 3264, cachelines: 51, members: 166 */
}
Signed-off-by: Ilpo Järvinen <ij@kernel.org>
Co-developed-by: Olivier Tilmans <olivier.tilmans@nokia.com>
Signed-off-by: Olivier Tilmans <olivier.tilmans@nokia.com>
Co-developed-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
Signed-off-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
---
v9:
- Use tcp_data_ecn_check() to set TCP_ECN_SEE flag only for RFC3168 ECN
- Add comments about setting TCP_ECN_SEEN flag for RFC3168 and Accruate ECN
---
.../networking/net_cachelines/tcp_sock.rst | 2 +
include/linux/tcp.h | 3 +
include/net/tcp.h | 15 +++
include/net/tcp_ecn.h | 53 +++++++++-
net/ipv4/tcp.c | 5 +-
net/ipv4/tcp_input.c | 100 ++++++++++++++++--
net/ipv4/tcp_output.c | 9 +-
7 files changed, 175 insertions(+), 12 deletions(-)
diff --git a/Documentation/networking/net_cachelines/tcp_sock.rst b/Documentation/networking/net_cachelines/tcp_sock.rst
index 7bbda5944ee2..31313a9adccc 100644
--- a/Documentation/networking/net_cachelines/tcp_sock.rst
+++ b/Documentation/networking/net_cachelines/tcp_sock.rst
@@ -101,6 +101,8 @@ u32 prr_delivered
u32 prr_out read_mostly read_mostly tcp_rate_skb_sent,tcp_newly_delivered(tx);tcp_ack,tcp_rate_gen,tcp_clean_rtx_queue(rx)
u32 delivered read_mostly read_write tcp_rate_skb_sent, tcp_newly_delivered(tx);tcp_ack, tcp_rate_gen, tcp_clean_rtx_queue (rx)
u32 delivered_ce read_mostly read_write tcp_rate_skb_sent(tx);tcp_rate_gen(rx)
+u32 received_ce read_mostly read_write
+u8:4 received_ce_pending read_mostly read_write
u32 lost read_mostly tcp_ack
u32 app_limited read_write read_mostly tcp_rate_check_app_limited,tcp_rate_skb_sent(tx);tcp_rate_gen(rx)
u64 first_tx_mstamp read_write tcp_rate_skb_sent
diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index bd692764d039..874e0e45dfad 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -287,6 +287,8 @@ struct tcp_sock {
*/
u8 nonagle : 4,/* Disable Nagle algorithm? */
rate_app_limited:1; /* rate_{delivered,interval_us} limited? */
+ u8 received_ce_pending:4, /* Not yet transmit cnt of received_ce */
+ unused2:4;
__be32 pred_flags;
u64 tcp_clock_cache; /* cache last tcp_clock_ns() (see tcp_mstamp_refresh()) */
u64 tcp_mstamp; /* most recent packet received/sent */
@@ -299,6 +301,7 @@ struct tcp_sock {
u32 snd_up; /* Urgent pointer */
u32 delivered; /* Total data packets delivered incl. rexmits */
u32 delivered_ce; /* Like the above but only ECE marked packets */
+ u32 received_ce; /* Like the above but for rcvd CE marked pkts */
u32 app_limited; /* limited until "delivered" reaches this val */
u32 rcv_wnd; /* Current receiver window */
/*
diff --git a/include/net/tcp.h b/include/net/tcp.h
index b462f8c3e85b..30b04cd8df60 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -963,6 +963,14 @@ static inline u32 tcp_rsk_tsval(const struct tcp_request_sock *treq)
#define TCPHDR_ACE (TCPHDR_ECE | TCPHDR_CWR | TCPHDR_AE)
#define TCPHDR_SYN_ECN (TCPHDR_SYN | TCPHDR_ECE | TCPHDR_CWR)
+#define TCP_ACCECN_CEP_ACE_MASK 0x7
+#define TCP_ACCECN_ACE_MAX_DELTA 6
+
+/* To avoid/detect middlebox interference, not all counters start at 0.
+ * See draft-ietf-tcpm-accurate-ecn for the latest values.
+ */
+#define TCP_ACCECN_CEP_INIT_OFFSET 5
+
/* State flags for sacked in struct tcp_skb_cb */
enum tcp_skb_cb_sacked_flags {
TCPCB_SACKED_ACKED = (1 << 0), /* SKB ACK'd by a SACK block */
@@ -1772,11 +1780,18 @@ static inline bool tcp_paws_reject(const struct tcp_options_received *rx_opt,
static inline void __tcp_fast_path_on(struct tcp_sock *tp, u32 snd_wnd)
{
+ u32 ace;
+
/* mptcp hooks are only on the slow path */
if (sk_is_mptcp((struct sock *)tp))
return;
+ ace = tcp_ecn_mode_accecn(tp) ?
+ ((tp->delivered_ce + TCP_ACCECN_CEP_INIT_OFFSET) &
+ TCP_ACCECN_CEP_ACE_MASK) : 0;
+
tp->pred_flags = htonl((tp->tcp_header_len << 26) |
+ (ace << 22) |
ntohl(TCP_FLAG_ACK) |
snd_wnd);
}
diff --git a/include/net/tcp_ecn.h b/include/net/tcp_ecn.h
index b3430557676b..b0ed89dbad41 100644
--- a/include/net/tcp_ecn.h
+++ b/include/net/tcp_ecn.h
@@ -12,6 +12,7 @@
static inline void tcp_ecn_queue_cwr(struct tcp_sock *tp)
{
+ /* Do not set CWR if in AccECN mode! */
if (tcp_ecn_mode_rfc3168(tp))
tp->ecn_flags |= TCP_ECN_QUEUE_CWR;
}
@@ -19,8 +20,10 @@ static inline void tcp_ecn_queue_cwr(struct tcp_sock *tp)
static inline void tcp_ecn_accept_cwr(struct sock *sk,
const struct sk_buff *skb)
{
- if (tcp_hdr(skb)->cwr) {
- tcp_sk(sk)->ecn_flags &= ~TCP_ECN_DEMAND_CWR;
+ struct tcp_sock *tp = tcp_sk(sk);
+
+ if (tcp_ecn_mode_rfc3168(tp) && tcp_hdr(skb)->cwr) {
+ tp->ecn_flags &= ~TCP_ECN_DEMAND_CWR;
/* If the sender is telling us it has entered CWR, then its
* cwnd may be very low (even just 1 packet), so we should ACK
@@ -36,6 +39,52 @@ static inline void tcp_ecn_withdraw_cwr(struct tcp_sock *tp)
tp->ecn_flags &= ~TCP_ECN_QUEUE_CWR;
}
+static inline u8 tcp_accecn_ace(const struct tcphdr *th)
+{
+ return (th->ae << 2) | (th->cwr << 1) | th->ece;
+}
+
+static inline void tcp_accecn_init_counters(struct tcp_sock *tp)
+{
+ tp->received_ce = 0;
+ tp->received_ce_pending = 0;
+}
+
+/* Updates Accurate ECN received counters from the received IP ECN field */
+static inline void tcp_ecn_received_counters(struct sock *sk, const struct sk_buff *skb)
+{
+ u8 ecnfield = TCP_SKB_CB(skb)->ip_dsfield & INET_ECN_MASK;
+ u8 is_ce = INET_ECN_is_ce(ecnfield);
+ struct tcp_sock *tp = tcp_sk(sk);
+
+ if (!INET_ECN_is_not_ect(ecnfield)) {
+ u32 pcount = is_ce * max_t(u16, 1, skb_shinfo(skb)->gso_segs);
+
+ /* As for accurate ECN, the TCP_ECN_SEEN flag is set by
+ * tcp_ecn_received_counters() when the ECN codepoint of
+ * received TCP data or ACK contains ECT(0), ECT(1), or CE.
+ */
+ if (!tcp_ecn_mode_rfc3168(tp))
+ tp->ecn_flags |= TCP_ECN_SEEN;
+
+ /* ACE counter tracks *all* segments including pure ACKs */
+ tp->received_ce += pcount;
+ tp->received_ce_pending = min(tp->received_ce_pending + pcount,
+ 0xfU);
+ }
+}
+
+static inline void tcp_accecn_set_ace(struct tcphdr *th, struct tcp_sock *tp)
+{
+ u32 wire_ace;
+
+ wire_ace = tp->received_ce + TCP_ACCECN_CEP_INIT_OFFSET;
+ th->ece = !!(wire_ace & 0x1);
+ th->cwr = !!(wire_ace & 0x2);
+ th->ae = !!(wire_ace & 0x4);
+ tp->received_ce_pending = 0;
+}
+
static inline void tcp_ecn_rcv_synack(struct tcp_sock *tp,
const struct tcphdr *th)
{
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index ef5574795174..c5e128d6ecc7 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -271,6 +271,7 @@
#include <net/icmp.h>
#include <net/inet_common.h>
#include <net/tcp.h>
+#include <net/tcp_ecn.h>
#include <net/mptcp.h>
#include <net/proto_memory.h>
#include <net/xfrm.h>
@@ -3390,6 +3391,7 @@ int tcp_disconnect(struct sock *sk, int flags)
tp->window_clamp = 0;
tp->delivered = 0;
tp->delivered_ce = 0;
+ tcp_accecn_init_counters(tp);
if (icsk->icsk_ca_initialized && icsk->icsk_ca_ops->release)
icsk->icsk_ca_ops->release(sk);
memset(icsk->icsk_ca_priv, 0, sizeof(icsk->icsk_ca_priv));
@@ -5116,6 +5118,7 @@ static void __init tcp_struct_check(void)
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_txrx, snd_up);
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_txrx, delivered);
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_txrx, delivered_ce);
+ CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_txrx, received_ce);
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_txrx, app_limited);
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_txrx, rcv_wnd);
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_txrx, rx_opt);
@@ -5123,7 +5126,7 @@ static void __init tcp_struct_check(void)
/* 32bit arches with 8byte alignment on u64 fields might need padding
* before tcp_clock_cache.
*/
- CACHELINE_ASSERT_GROUP_SIZE(struct tcp_sock, tcp_sock_write_txrx, 91 + 4);
+ CACHELINE_ASSERT_GROUP_SIZE(struct tcp_sock, tcp_sock_write_txrx, 95 + 4);
/* RX read-write hotpath cache lines */
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_rx, bytes_received);
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index da4245718e63..56c8514ee139 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -360,16 +360,25 @@ static void tcp_data_ecn_check(struct sock *sk, const struct sk_buff *skb)
if (tcp_ca_needs_ecn(sk))
tcp_ca_event(sk, CA_EVENT_ECN_IS_CE);
- if (!(tp->ecn_flags & TCP_ECN_DEMAND_CWR)) {
+ if (!(tp->ecn_flags & TCP_ECN_DEMAND_CWR) &&
+ tcp_ecn_mode_rfc3168(tp)) {
/* Better not delay acks, sender can have a very low cwnd */
tcp_enter_quickack_mode(sk, 2);
tp->ecn_flags |= TCP_ECN_DEMAND_CWR;
}
+ /* As for RFC3168 ECN, the TCP_ECN_SEEN flag is set by
+ * tcp_data_ecn_check() when the ECN codepoint of
+ * received TCP data contains ECT(0), ECT(1), or CE.
+ */
+ if (!tcp_ecn_mode_rfc3168(tp))
+ break;
tp->ecn_flags |= TCP_ECN_SEEN;
break;
default:
if (tcp_ca_needs_ecn(sk))
tcp_ca_event(sk, CA_EVENT_ECN_NO_CE);
+ if (!tcp_ecn_mode_rfc3168(tp))
+ break;
tp->ecn_flags |= TCP_ECN_SEEN;
break;
}
@@ -385,10 +394,64 @@ static void tcp_count_delivered(struct tcp_sock *tp, u32 delivered,
bool ece_ack)
{
tp->delivered += delivered;
- if (ece_ack)
+ if (tcp_ecn_mode_rfc3168(tp) && ece_ack)
tcp_count_delivered_ce(tp, delivered);
}
+/* Returns the ECN CE delta */
+static u32 __tcp_accecn_process(struct sock *sk, const struct sk_buff *skb,
+ u32 delivered_pkts, int flag)
+{
+ const struct tcphdr *th = tcp_hdr(skb);
+ struct tcp_sock *tp = tcp_sk(sk);
+ u32 delta, safe_delta;
+ u32 corrected_ace;
+
+ /* Reordered ACK or uncertain due to lack of data to send and ts */
+ if (!(flag & (FLAG_FORWARD_PROGRESS | FLAG_TS_PROGRESS)))
+ return 0;
+
+ if (!(flag & FLAG_SLOWPATH)) {
+ /* AccECN counter might overflow on large ACKs */
+ if (delivered_pkts <= TCP_ACCECN_CEP_ACE_MASK)
+ return 0;
+ }
+
+ /* ACE field is not available during handshake */
+ if (flag & FLAG_SYN_ACKED)
+ return 0;
+
+ if (tp->received_ce_pending >= TCP_ACCECN_ACE_MAX_DELTA)
+ inet_csk(sk)->icsk_ack.pending |= ICSK_ACK_NOW;
+
+ corrected_ace = tcp_accecn_ace(th) - TCP_ACCECN_CEP_INIT_OFFSET;
+ delta = (corrected_ace - tp->delivered_ce) & TCP_ACCECN_CEP_ACE_MASK;
+ if (delivered_pkts <= TCP_ACCECN_CEP_ACE_MASK)
+ return delta;
+
+ safe_delta = delivered_pkts -
+ ((delivered_pkts - delta) & TCP_ACCECN_CEP_ACE_MASK);
+
+ return safe_delta;
+}
+
+static u32 tcp_accecn_process(struct sock *sk, const struct sk_buff *skb,
+ u32 delivered_pkts, int *flag)
+{
+ struct tcp_sock *tp = tcp_sk(sk);
+ u32 delta;
+
+ delta = __tcp_accecn_process(sk, skb, delivered_pkts, *flag);
+ if (delta > 0) {
+ tcp_count_delivered_ce(tp, delta);
+ *flag |= FLAG_ECE;
+ /* Recalculate header predictor */
+ if (tp->pred_flags)
+ tcp_fast_path_on(tp);
+ }
+ return delta;
+}
+
/* Buffer size and advertised window tuning.
*
* 1. Tuning sk->sk_sndbuf, when connection enters established state.
@@ -3744,7 +3807,8 @@ static void tcp_xmit_recovery(struct sock *sk, int rexmit)
}
/* Returns the number of packets newly acked or sacked by the current ACK */
-static u32 tcp_newly_delivered(struct sock *sk, u32 prior_delivered, int flag)
+static u32 tcp_newly_delivered(struct sock *sk, u32 prior_delivered,
+ u32 ecn_count, int flag)
{
const struct net *net = sock_net(sk);
struct tcp_sock *tp = tcp_sk(sk);
@@ -3752,8 +3816,12 @@ static u32 tcp_newly_delivered(struct sock *sk, u32 prior_delivered, int flag)
delivered = tp->delivered - prior_delivered;
NET_ADD_STATS(net, LINUX_MIB_TCPDELIVERED, delivered);
- if (flag & FLAG_ECE)
- NET_ADD_STATS(net, LINUX_MIB_TCPDELIVEREDCE, delivered);
+
+ if (flag & FLAG_ECE) {
+ if (tcp_ecn_mode_rfc3168(tp))
+ ecn_count = delivered;
+ NET_ADD_STATS(net, LINUX_MIB_TCPDELIVEREDCE, ecn_count);
+ }
return delivered;
}
@@ -3774,6 +3842,7 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
u32 delivered = tp->delivered;
u32 lost = tp->lost;
int rexmit = REXMIT_NONE; /* Flag to (re)transmit to recover losses */
+ u32 ecn_count = 0; /* Did we receive ECE/an AccECN ACE update? */
u32 prior_fack;
sack_state.first_sackt = 0;
@@ -3881,6 +3950,11 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
tcp_rack_update_reo_wnd(sk, &rs);
+ if (tcp_ecn_mode_accecn(tp))
+ ecn_count = tcp_accecn_process(sk, skb,
+ tp->delivered - delivered,
+ &flag);
+
tcp_in_ack_event(sk, flag);
if (tp->tlp_high_seq)
@@ -3905,7 +3979,8 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
if ((flag & FLAG_FORWARD_PROGRESS) || !(flag & FLAG_NOT_DUP))
sk_dst_confirm(sk);
- delivered = tcp_newly_delivered(sk, delivered, flag);
+ delivered = tcp_newly_delivered(sk, delivered, ecn_count, flag);
+
lost = tp->lost - lost; /* freshly marked lost */
rs.is_ack_delayed = !!(flag & FLAG_ACK_MAYBE_DELAYED);
tcp_rate_gen(sk, delivered, lost, is_sack_reneg, sack_state.rate);
@@ -3914,12 +3989,16 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
return 1;
no_queue:
+ if (tcp_ecn_mode_accecn(tp))
+ ecn_count = tcp_accecn_process(sk, skb,
+ tp->delivered - delivered,
+ &flag);
tcp_in_ack_event(sk, flag);
/* If data was DSACKed, see if we can undo a cwnd reduction. */
if (flag & FLAG_DSACKING_ACK) {
tcp_fastretrans_alert(sk, prior_snd_una, num_dupack, &flag,
&rexmit);
- tcp_newly_delivered(sk, delivered, flag);
+ tcp_newly_delivered(sk, delivered, ecn_count, flag);
}
/* If this ack opens up a zero window, clear backoff. It was
* being used to time the probes, and is probably far higher than
@@ -3940,7 +4019,7 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
&sack_state);
tcp_fastretrans_alert(sk, prior_snd_una, num_dupack, &flag,
&rexmit);
- tcp_newly_delivered(sk, delivered, flag);
+ tcp_newly_delivered(sk, delivered, ecn_count, flag);
tcp_xmit_recovery(sk, rexmit);
}
@@ -6041,6 +6120,8 @@ void tcp_rcv_established(struct sock *sk, struct sk_buff *skb)
flag |= __tcp_replace_ts_recent(tp,
delta);
+ tcp_ecn_received_counters(sk, skb);
+
/* We know that such packets are checksummed
* on entry.
*/
@@ -6085,6 +6166,7 @@ void tcp_rcv_established(struct sock *sk, struct sk_buff *skb)
/* Bulk data transfer: receiver */
tcp_cleanup_skb(skb);
__skb_pull(skb, tcp_header_len);
+ tcp_ecn_received_counters(sk, skb);
eaten = tcp_queue_rcv(sk, skb, &fragstolen);
tcp_event_data_recv(sk, skb);
@@ -6125,6 +6207,8 @@ void tcp_rcv_established(struct sock *sk, struct sk_buff *skb)
return;
step5:
+ tcp_ecn_received_counters(sk, skb);
+
reason = tcp_ack(sk, skb, FLAG_SLOWPATH | FLAG_UPDATE_TS_RECENT);
if ((int)reason < 0) {
reason = -reason;
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 2f1102616024..78018870a389 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -328,7 +328,14 @@ static void tcp_ecn_send(struct sock *sk, struct sk_buff *skb,
{
struct tcp_sock *tp = tcp_sk(sk);
- if (tcp_ecn_mode_rfc3168(tp)) {
+ if (!tcp_ecn_mode_any(tp))
+ return;
+
+ if (tcp_ecn_mode_accecn(tp)) {
+ INET_ECN_xmit(sk);
+ tcp_accecn_set_ace(th, tp);
+ skb_shinfo(skb)->gso_type |= SKB_GSO_TCP_ACCECN;
+ } else {
/* Not-retransmitted data segment: set ECT and inject CWR. */
if (skb->len != tcp_header_len &&
!before(TCP_SKB_CB(skb)->seq, tp->snd_nxt)) {
--
2.34.1
^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH v11 net-next 06/15] tcp: accecn: AccECN negotiation
2025-07-03 5:17 [PATCH v11 net-next 00/15] AccECN protocol patch series chia-yu.chang
` (4 preceding siblings ...)
2025-07-03 5:17 ` [PATCH v11 net-next 05/15] tcp: AccECN core chia-yu.chang
@ 2025-07-03 5:17 ` chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 07/15] tcp: Add wait_third_ack flag for ECN negotiation in simultaneous connect chia-yu.chang
` (8 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: chia-yu.chang @ 2025-07-03 5:17 UTC (permalink / raw)
To: pabeni, edumazet, linux-doc, corbet, horms, dsahern, kuniyu, bpf,
netdev, dave.taht, jhs, kuba, stephen, xiyou.wangcong, jiri,
davem, andrew+netdev, donald.hunter, ast, liuhangbin, shuah,
linux-kselftest, ij, ncardwell, koen.de_schepper, g.white,
ingemar.s.johansson, mirja.kuehlewind, cheshire, rs.ietf,
Jason_Livingood, vidhi_goel
Cc: Olivier Tilmans, Chia-Yu Chang
From: Ilpo Järvinen <ij@kernel.org>
Accurate ECN negotiation parts based on the specification:
https://tools.ietf.org/id/draft-ietf-tcpm-accurate-ecn-28.txt
Accurate ECN is negotiated using ECE, CWR and AE flags in the
TCP header. TCP falls back into using RFC3168 ECN if one of the
ends supports only RFC3168-style ECN.
The AccECN negotiation includes reflecting IP ECN field value
seen in SYN and SYNACK back using the same bits as negotiation
to allow responding to SYN CE marks and to detect ECN field
mangling. CE marks should not occur currently because SYN=1
segments are sent with Non-ECT in IP ECN field (but proposal
exists to remove this restriction).
Reflecting SYN IP ECN field in SYNACK is relatively simple.
Reflecting SYNACK IP ECN field in the final/third ACK of
the handshake is more challenging. Linux TCP code is not well
prepared for using the final/third ACK a signalling channel
which makes things somewhat complicated here.
tcp_ecn sysctl can be used to select the highest ECN variant
(Accurate ECN, ECN, No ECN) that is attemped to be negotiated and
requested for incoming connection and outgoing connection:
TCP_ECN_IN_NOECN_OUT_NOECN, TCP_ECN_IN_ECN_OUT_ECN,
TCP_ECN_IN_ECN_OUT_NOECN, TCP_ECN_IN_ACCECN_OUT_ACCECN,
TCP_ECN_IN_ACCECN_OUT_ECN, and TCP_ECN_IN_ACCECN_OUT_NOECN.
After this patch, the size of tcp_request_sock remains unchanged
and no new holes are added. Below are the pahole outcomes before
and after this patch:
[BEFORE THIS PATCH]
struct tcp_request_sock {
[...]
u32 rcv_nxt; /* 352 4 */
u8 syn_tos; /* 356 1 */
/* size: 360, cachelines: 6, members: 16 */
}
[AFTER THIS PATCH]
struct tcp_request_sock {
[...]
u32 rcv_nxt; /* 352 4 */
u8 syn_tos; /* 356 1 */
bool accecn_ok; /* 357 1 */
u8 syn_ect_snt:2; /* 358: 0 1 */
u8 syn_ect_rcv:2; /* 358: 2 1 */
u8 accecn_fail_mode:4; /* 358: 4 1 */
/* size: 360, cachelines: 6, members: 20 */
}
After this patch, the size of tcp_sock remains unchanged and no new
holes are added. Also, 4 bits of the existing 2-byte hole are exploited.
Below are the pahole outcomes before and after this patch:
[BEFORE THIS PATCH]
struct tcp_sock {
[...]
u8 dup_ack_counter:2; /* 2825: 0 1 */
u8 tlp_retrans:1; /* 2825: 2 1 */
u8 unused:5; /* 2825: 3 1 */
u8 thin_lto:1; /* 2826: 0 1 */
u8 fastopen_connect:1; /* 2826: 1 1 */
u8 fastopen_no_cookie:1; /* 2826: 2 1 */
u8 fastopen_client_fail:2; /* 2826: 3 1 */
u8 frto:1; /* 2826: 5 1 */
/* XXX 2 bits hole, try to pack */
[...]
u8 keepalive_probes; /* 2829 1 */
/* XXX 2 bytes hole, try to pack */
[...]
/* size: 3264, cachelines: 51, members: 166 */
}
[AFTER THIS PATCH]
struct tcp_sock {
[...]
u8 dup_ack_counter:2; /* 2825: 0 1 */
u8 tlp_retrans:1; /* 2825: 2 1 */
u8 syn_ect_snt:2; /* 2825: 3 1 */
u8 syn_ect_rcv:2; /* 2825: 5 1 */
/* XXX 1 bit hole, try to pack */
u8 thin_lto:1; /* 2826: 0 1 */
u8 fastopen_connect:1; /* 2826: 1 1 */
u8 fastopen_no_cookie:1; /* 2826: 2 1 */
u8 fastopen_client_fail:2; /* 2826: 3 1 */
u8 frto:1; /* 2826: 5 1 */
/* XXX 2 bits hole, try to pack */
[...]
u8 keepalive_probes; /* 2829 1 */
u8 accecn_fail_mode:4; /* 2830: 0 1 */
/* XXX 4 bits hole, try to pack */
/* XXX 1 byte hole, try to pack */
[...]
/* size: 3264, cachelines: 51, members: 168 */
}
Signed-off-by: Ilpo Järvinen <ij@kernel.org>
Co-developed-by: Olivier Tilmans <olivier.tilmans@nokia.com>
Signed-off-by: Olivier Tilmans <olivier.tilmans@nokia.com>
Co-developed-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
Signed-off-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
---
v10
- Add documentation of tcp_ecn in ip-sysctl.rst to this patch
- Move wait third ACK functionality into a later patch
- Add comments on the AccECN helper functions in tcp_ecn.h
---
---
Documentation/networking/ip-sysctl.rst | 36 +-
.../networking/net_cachelines/tcp_sock.rst | 3 +
include/linux/tcp.h | 8 +-
include/net/tcp.h | 1 +
include/net/tcp_ecn.h | 327 ++++++++++++++++--
net/ipv4/syncookies.c | 3 +
net/ipv4/sysctl_net_ipv4.c | 3 +-
net/ipv4/tcp.c | 1 +
net/ipv4/tcp_input.c | 50 ++-
net/ipv4/tcp_ipv4.c | 6 +-
net/ipv4/tcp_minisocks.c | 24 +-
net/ipv4/tcp_output.c | 10 +-
net/ipv6/syncookies.c | 1 +
net/ipv6/tcp_ipv6.c | 1 +
14 files changed, 415 insertions(+), 59 deletions(-)
diff --git a/Documentation/networking/ip-sysctl.rst b/Documentation/networking/ip-sysctl.rst
index 9af5a8935d57..ebd391a1a3af 100644
--- a/Documentation/networking/ip-sysctl.rst
+++ b/Documentation/networking/ip-sysctl.rst
@@ -443,20 +443,28 @@ tcp_early_retrans - INTEGER
tcp_ecn - INTEGER
Control use of Explicit Congestion Notification (ECN) by TCP.
- ECN is used only when both ends of the TCP connection indicate
- support for it. This feature is useful in avoiding losses due
- to congestion by allowing supporting routers to signal
- congestion before having to drop packets.
-
- Possible values are:
-
- = =====================================================
- 0 Disable ECN. Neither initiate nor accept ECN.
- 1 Enable ECN when requested by incoming connections and
- also request ECN on outgoing connection attempts.
- 2 Enable ECN when requested by incoming connections
- but do not request ECN on outgoing connections.
- = =====================================================
+ ECN is used only when both ends of the TCP connection indicate support
+ for it. This feature is useful in avoiding losses due to congestion by
+ allowing supporting routers to signal congestion before having to drop
+ packets. A host that supports ECN both sends ECN at the IP layer and
+ feeds back ECN at the TCP layer. The highest variant of ECN feedback
+ that both peers support is chosen by the ECN negotiation (Accurate ECN,
+ ECN, or no ECN).
+
+ The highest negotiated variant for incoming connection requests
+ and the highest variant requested by outgoing connection
+ attempts:
+
+ ===== ==================== ====================
+ Value Incoming connections Outgoing connections
+ ===== ==================== ====================
+ 0 No ECN No ECN
+ 1 ECN ECN
+ 2 ECN No ECN
+ 3 AccECN AccECN
+ 4 AccECN ECN
+ 5 AccECN No ECN
+ ===== ==================== ====================
Default: 2
diff --git a/Documentation/networking/net_cachelines/tcp_sock.rst b/Documentation/networking/net_cachelines/tcp_sock.rst
index 31313a9adccc..4f71ece7c655 100644
--- a/Documentation/networking/net_cachelines/tcp_sock.rst
+++ b/Documentation/networking/net_cachelines/tcp_sock.rst
@@ -103,6 +103,9 @@ u32 delivered read_mostly read_w
u32 delivered_ce read_mostly read_write tcp_rate_skb_sent(tx);tcp_rate_gen(rx)
u32 received_ce read_mostly read_write
u8:4 received_ce_pending read_mostly read_write
+u8:2 syn_ect_snt write_mostly read_write
+u8:2 syn_ect_rcv read_mostly read_write
+u8:4 accecn_fail_mode
u32 lost read_mostly tcp_ack
u32 app_limited read_write read_mostly tcp_rate_check_app_limited,tcp_rate_skb_sent(tx);tcp_rate_gen(rx)
u64 first_tx_mstamp read_write tcp_rate_skb_sent
diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 874e0e45dfad..994a6e7256e1 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -168,6 +168,10 @@ struct tcp_request_sock {
* after data-in-SYN.
*/
u8 syn_tos;
+ bool accecn_ok;
+ u8 syn_ect_snt: 2,
+ syn_ect_rcv: 2,
+ accecn_fail_mode:4;
#ifdef CONFIG_TCP_AO
u8 ao_keyid;
u8 ao_rcv_next;
@@ -375,7 +379,8 @@ struct tcp_sock {
u8 compressed_ack;
u8 dup_ack_counter:2,
tlp_retrans:1, /* TLP is a retransmission */
- unused:5;
+ syn_ect_snt:2, /* AccECN ECT memory, only */
+ syn_ect_rcv:2; /* ... needed durign 3WHS + first seqno */
u8 thin_lto : 1,/* Use linear timeouts for thin streams */
fastopen_connect:1, /* FASTOPEN_CONNECT sockopt */
fastopen_no_cookie:1, /* Allow send/recv SYN+data without a cookie */
@@ -391,6 +396,7 @@ struct tcp_sock {
syn_fastopen_child:1; /* created TFO passive child socket */
u8 keepalive_probes; /* num of allowed keep alive probes */
+ u8 accecn_fail_mode:4; /* AccECN failure handling */
u32 tcp_tx_delay; /* delay (in usec) added to TX packets */
/* RTT measurement */
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 30b04cd8df60..61b103633da4 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -962,6 +962,7 @@ static inline u32 tcp_rsk_tsval(const struct tcp_request_sock *treq)
#define TCPHDR_ACE (TCPHDR_ECE | TCPHDR_CWR | TCPHDR_AE)
#define TCPHDR_SYN_ECN (TCPHDR_SYN | TCPHDR_ECE | TCPHDR_CWR)
+#define TCPHDR_SYNACK_ACCECN (TCPHDR_SYN | TCPHDR_ACK | TCPHDR_CWR)
#define TCP_ACCECN_CEP_ACE_MASK 0x7
#define TCP_ACCECN_ACE_MAX_DELTA 6
diff --git a/include/net/tcp_ecn.h b/include/net/tcp_ecn.h
index b0ed89dbad41..7ac68cb70ef3 100644
--- a/include/net/tcp_ecn.h
+++ b/include/net/tcp_ecn.h
@@ -4,12 +4,26 @@
#include <linux/tcp.h>
#include <linux/skbuff.h>
+#include <linux/bitfield.h>
#include <net/inet_connection_sock.h>
#include <net/sock.h>
#include <net/tcp.h>
#include <net/inet_ecn.h>
+/* The highest ECN variant (Accurate ECN, ECN, or no ECN) that is
+ * attemped to be negotiated and requested for incoming connection
+ * and outgoing connection, respectively.
+ */
+enum tcp_ecn_mode {
+ TCP_ECN_IN_NOECN_OUT_NOECN = 0,
+ TCP_ECN_IN_ECN_OUT_ECN = 1,
+ TCP_ECN_IN_ECN_OUT_NOECN = 2,
+ TCP_ECN_IN_ACCECN_OUT_ACCECN = 3,
+ TCP_ECN_IN_ACCECN_OUT_ECN = 4,
+ TCP_ECN_IN_ACCECN_OUT_NOECN = 5,
+};
+
static inline void tcp_ecn_queue_cwr(struct tcp_sock *tp)
{
/* Do not set CWR if in AccECN mode! */
@@ -39,19 +53,130 @@ static inline void tcp_ecn_withdraw_cwr(struct tcp_sock *tp)
tp->ecn_flags &= ~TCP_ECN_QUEUE_CWR;
}
+/* tp->accecn_fail_mode */
+#define TCP_ACCECN_ACE_FAIL_SEND BIT(0)
+#define TCP_ACCECN_ACE_FAIL_RECV BIT(1)
+#define TCP_ACCECN_OPT_FAIL_SEND BIT(2)
+#define TCP_ACCECN_OPT_FAIL_RECV BIT(3)
+
+static inline bool tcp_accecn_ace_fail_send(const struct tcp_sock *tp)
+{
+ return tp->accecn_fail_mode & TCP_ACCECN_ACE_FAIL_SEND;
+}
+
+static inline bool tcp_accecn_ace_fail_recv(const struct tcp_sock *tp)
+{
+ return tp->accecn_fail_mode & TCP_ACCECN_ACE_FAIL_RECV;
+}
+
+static inline bool tcp_accecn_opt_fail_send(const struct tcp_sock *tp)
+{
+ return tp->accecn_fail_mode & TCP_ACCECN_OPT_FAIL_SEND;
+}
+
+static inline bool tcp_accecn_opt_fail_recv(const struct tcp_sock *tp)
+{
+ return tp->accecn_fail_mode & TCP_ACCECN_OPT_FAIL_RECV;
+}
+
+static inline void tcp_accecn_fail_mode_set(struct tcp_sock *tp, u8 mode)
+{
+ tp->accecn_fail_mode |= mode;
+}
+
static inline u8 tcp_accecn_ace(const struct tcphdr *th)
{
return (th->ae << 2) | (th->cwr << 1) | th->ece;
}
-static inline void tcp_accecn_init_counters(struct tcp_sock *tp)
+/* Infer the ECT value our SYN arrived with from the echoed ACE field */
+static inline int tcp_accecn_extract_syn_ect(u8 ace)
{
- tp->received_ce = 0;
- tp->received_ce_pending = 0;
+ /* Below is an excerpt from Tables 2 of the AccECN spec:
+ * +================================+========================+
+ * | Echoed ACE field | Received ECT values of |
+ * | AE CWR ECE | our SYN arrived with |
+ * +================================+========================+
+ * | 0 1 0 | Not-ECT |
+ * | 0 1 1 | ECT(1) |
+ * | 1 0 0 | ECT(0) |
+ * | 1 1 0 | CE |
+ * +================================+========================+
+ */
+ if (ace & 0x1)
+ return INET_ECN_ECT_1;
+ if (!(ace & 0x2))
+ return INET_ECN_ECT_0;
+ if (ace & 0x4)
+ return INET_ECN_CE;
+ return INET_ECN_NOT_ECT;
+}
+
+/* Check ECN field transition to detect invalid transitions */
+static inline bool tcp_ect_transition_valid(u8 snt, u8 rcv)
+{
+ if (rcv == snt)
+ return true;
+
+ /* Non-ECT altered to something or something became non-ECT */
+ if (snt == INET_ECN_NOT_ECT || rcv == INET_ECN_NOT_ECT)
+ return false;
+ /* CE -> ECT(0/1)? */
+ if (snt == INET_ECN_CE)
+ return false;
+ return true;
+}
+
+static inline bool tcp_accecn_validate_syn_feedback(struct sock *sk, u8 ace,
+ u8 sent_ect)
+{
+ u8 ect = tcp_accecn_extract_syn_ect(ace);
+ struct tcp_sock *tp = tcp_sk(sk);
+
+ if (!READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_ecn_fallback))
+ return true;
+
+ if (!tcp_ect_transition_valid(sent_ect, ect)) {
+ tcp_accecn_fail_mode_set(tp, TCP_ACCECN_ACE_FAIL_RECV);
+ return false;
+ }
+
+ return true;
+}
+
+/* Validate the 3rd ACK based on the ACE field, see Table 4 of AccECN spec */
+static inline void tcp_accecn_third_ack(struct sock *sk,
+ const struct sk_buff *skb, u8 sent_ect)
+{
+ u8 ace = tcp_accecn_ace(tcp_hdr(skb));
+ struct tcp_sock *tp = tcp_sk(sk);
+
+ switch (ace) {
+ case 0x0:
+ /* Invalid value */
+ tcp_accecn_fail_mode_set(tp, TCP_ACCECN_ACE_FAIL_RECV);
+ break;
+ case 0x7:
+ case 0x5:
+ case 0x1:
+ /* Unused but legal values */
+ break;
+ default:
+ /* Validation only applies to first non-data packet */
+ if (TCP_SKB_CB(skb)->seq == TCP_SKB_CB(skb)->end_seq &&
+ !TCP_SKB_CB(skb)->sacked &&
+ tcp_accecn_validate_syn_feedback(sk, ace, sent_ect)) {
+ if ((tcp_accecn_extract_syn_ect(ace) == INET_ECN_CE) &&
+ !tp->delivered_ce)
+ tp->delivered_ce++;
+ }
+ break;
+ }
}
/* Updates Accurate ECN received counters from the received IP ECN field */
-static inline void tcp_ecn_received_counters(struct sock *sk, const struct sk_buff *skb)
+static inline void tcp_ecn_received_counters(struct sock *sk,
+ const struct sk_buff *skb)
{
u8 ecnfield = TCP_SKB_CB(skb)->ip_dsfield & INET_ECN_MASK;
u8 is_ce = INET_ECN_is_ce(ecnfield);
@@ -74,27 +199,157 @@ static inline void tcp_ecn_received_counters(struct sock *sk, const struct sk_bu
}
}
-static inline void tcp_accecn_set_ace(struct tcphdr *th, struct tcp_sock *tp)
+/* AccECN specification, 5.1: [...] a server can determine that it
+ * negotiated AccECN as [...] if the ACK contains an ACE field with
+ * the value 0b010 to 0b111 (decimal 2 to 7).
+ */
+static inline bool cookie_accecn_ok(const struct tcphdr *th)
{
- u32 wire_ace;
+ return tcp_accecn_ace(th) > 0x1;
+}
+
+/* Used to form the ACE flags for SYN/ACK */
+static inline u16 tcp_accecn_reflector_flags(u8 ect)
+{
+ /* TCP ACE flags of SYN/ACK are set based on IP-ECN codepoint received
+ * from SYN. Below is an excerpt from Table 2 of the AccECN spec:
+ * +====================+====================================+
+ * | IP-ECN codepoint | Respective ACE falgs on SYN/ACK |
+ * | received on SYN | AE CWR ECE |
+ * +====================+====================================+
+ * | Not-ECT | 0 1 0 |
+ * | ECT(1) | 0 1 1 |
+ * | ECT(0) | 1 0 0 |
+ * | CE | 1 1 0 |
+ * +====================+====================================+
+ */
+ u32 flags = ect + 2;
+
+ if (ect == 3)
+ flags++;
+ return FIELD_PREP(TCPHDR_ACE, flags);
+}
- wire_ace = tp->received_ce + TCP_ACCECN_CEP_INIT_OFFSET;
- th->ece = !!(wire_ace & 0x1);
- th->cwr = !!(wire_ace & 0x2);
- th->ae = !!(wire_ace & 0x4);
+/* AccECN specificaiton, 3.1.2: If a TCP server that implements AccECN
+ * receives a SYN with the three TCP header flags (AE, CWR and ECE) set
+ * to any combination other than 000, 011 or 111, it MUST negotiate the
+ * use of AccECN as if they had been set to 111.
+ */
+static inline bool tcp_accecn_syn_requested(const struct tcphdr *th)
+{
+ u8 ace = tcp_accecn_ace(th);
+
+ return ace && ace != 0x3;
+}
+
+static inline void tcp_accecn_init_counters(struct tcp_sock *tp)
+{
+ tp->received_ce = 0;
tp->received_ce_pending = 0;
}
-static inline void tcp_ecn_rcv_synack(struct tcp_sock *tp,
- const struct tcphdr *th)
+/* Used for make_synack to form the ACE flags */
+static inline void tcp_accecn_echo_syn_ect(struct tcphdr *th, u8 ect)
+{
+ /* TCP ACE flags of SYN/ACK are set based on IP-ECN codepoint received
+ * from SYN. Below is an excerpt from Table 2 of the AccECN spec:
+ * +====================+====================================+
+ * | IP-ECN codepoint | Respective ACE falgs on SYN/ACK |
+ * | received on SYN | AE CWR ECE |
+ * +====================+====================================+
+ * | Not-ECT | 0 1 0 |
+ * | ECT(1) | 0 1 1 |
+ * | ECT(0) | 1 0 0 |
+ * | CE | 1 1 0 |
+ * +====================+====================================+
+ */
+ th->ae = !!(ect & INET_ECN_ECT_0);
+ th->cwr = ect != INET_ECN_ECT_0;
+ th->ece = ect == INET_ECN_ECT_1;
+}
+
+static inline void tcp_accecn_set_ace(struct tcp_sock *tp, struct sk_buff *skb,
+ struct tcphdr *th)
{
- if (tcp_ecn_mode_rfc3168(tp) && (!th->ece || th->cwr))
+ u32 wire_ace;
+
+ /* The final packet of the 3WHS or anything like it must reflect
+ * the SYN/ACK ECT instead of putting CEP into ACE field, such
+ * case show up in tcp_flags.
+ */
+ if (likely(!(TCP_SKB_CB(skb)->tcp_flags & TCPHDR_ACE))) {
+ wire_ace = tp->received_ce + TCP_ACCECN_CEP_INIT_OFFSET;
+ th->ece = !!(wire_ace & 0x1);
+ th->cwr = !!(wire_ace & 0x2);
+ th->ae = !!(wire_ace & 0x4);
+ tp->received_ce_pending = 0;
+ }
+}
+
+/* See Table 2 of the AccECN draft */
+static inline void tcp_ecn_rcv_synack(struct sock *sk, const struct tcphdr *th,
+ u8 ip_dsfield)
+{
+ struct tcp_sock *tp = tcp_sk(sk);
+ u8 ace = tcp_accecn_ace(th);
+
+ switch (ace) {
+ case 0x0:
+ case 0x7:
+ /* +========+========+============+=============+
+ * | A | B | SYN/ACK | Feedback |
+ * | | | B->A | Mode of A |
+ * | | | AE CWR ECE | |
+ * +========+========+============+=============+
+ * | AccECN | No ECN | 0 0 0 | Not ECN |
+ * | AccECN | Broken | 1 1 1 | Not ECN |
+ * +========+========+============+=============+
+ */
tcp_ecn_mode_set(tp, TCP_ECN_DISABLED);
+ break;
+ case 0x1:
+ case 0x5:
+ /* +========+========+============+=============+
+ * | A | B | SYN/ACK | Feedback |
+ * | | | B->A | Mode of A |
+ * | | | AE CWR ECE | |
+ * +========+========+============+=============+
+ * | AccECN | Nonce | 1 0 1 | (Reserved) |
+ * | AccECN | ECN | 0 0 1 | Classic ECN |
+ * | Nonce | AccECN | 0 0 1 | Classic ECN |
+ * | ECN | AccECN | 0 0 1 | Classic ECN |
+ * +========+========+============+=============+
+ */
+ if (tcp_ecn_mode_pending(tp))
+ /* Downgrade from AccECN, or requested initially */
+ tcp_ecn_mode_set(tp, TCP_ECN_MODE_RFC3168);
+ break;
+ default:
+ tcp_ecn_mode_set(tp, TCP_ECN_MODE_ACCECN);
+ tp->syn_ect_rcv = ip_dsfield & INET_ECN_MASK;
+ if (INET_ECN_is_ce(ip_dsfield) &&
+ tcp_accecn_validate_syn_feedback(sk, ace,
+ tp->syn_ect_snt)) {
+ tp->received_ce++;
+ tp->received_ce_pending++;
+ }
+ break;
+ }
}
-static inline void tcp_ecn_rcv_syn(struct tcp_sock *tp,
- const struct tcphdr *th)
+static inline void tcp_ecn_rcv_syn(struct tcp_sock *tp, const struct tcphdr *th,
+ const struct sk_buff *skb)
{
+ if (tcp_ecn_mode_pending(tp)) {
+ if (!tcp_accecn_syn_requested(th)) {
+ /* Downgrade to classic ECN feedback */
+ tcp_ecn_mode_set(tp, TCP_ECN_MODE_RFC3168);
+ } else {
+ tp->syn_ect_rcv = TCP_SKB_CB(skb)->ip_dsfield &
+ INET_ECN_MASK;
+ tcp_ecn_mode_set(tp, TCP_ECN_MODE_ACCECN);
+ }
+ }
if (tcp_ecn_mode_rfc3168(tp) && (!th->ece || !th->cwr))
tcp_ecn_mode_set(tp, TCP_ECN_DISABLED);
}
@@ -110,7 +365,7 @@ static inline bool tcp_ecn_rcv_ecn_echo(const struct tcp_sock *tp,
/* Packet ECN state for a SYN-ACK */
static inline void tcp_ecn_send_synack(struct sock *sk, struct sk_buff *skb)
{
- const struct tcp_sock *tp = tcp_sk(sk);
+ struct tcp_sock *tp = tcp_sk(sk);
TCP_SKB_CB(skb)->tcp_flags &= ~TCPHDR_CWR;
if (tcp_ecn_disabled(tp))
@@ -118,6 +373,13 @@ static inline void tcp_ecn_send_synack(struct sock *sk, struct sk_buff *skb)
else if (tcp_ca_needs_ecn(sk) ||
tcp_bpf_ca_needs_ecn(sk))
INET_ECN_xmit(sk);
+
+ if (tp->ecn_flags & TCP_ECN_MODE_ACCECN) {
+ TCP_SKB_CB(skb)->tcp_flags &= ~TCPHDR_ACE;
+ TCP_SKB_CB(skb)->tcp_flags |=
+ tcp_accecn_reflector_flags(tp->syn_ect_rcv);
+ tp->syn_ect_snt = inet_sk(sk)->tos & INET_ECN_MASK;
+ }
}
/* Packet ECN state for a SYN. */
@@ -125,8 +387,20 @@ static inline void tcp_ecn_send_syn(struct sock *sk, struct sk_buff *skb)
{
struct tcp_sock *tp = tcp_sk(sk);
bool bpf_needs_ecn = tcp_bpf_ca_needs_ecn(sk);
- bool use_ecn = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_ecn) == 1 ||
- tcp_ca_needs_ecn(sk) || bpf_needs_ecn;
+ bool use_ecn, use_accecn;
+ u8 tcp_ecn = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_ecn);
+
+ /* +================+===========================+
+ * | tcp_ecn values | Outgoing connections |
+ * +================+===========================+
+ * | 0,2,5 | Do not request ECN |
+ * | 1,4 | Request ECN connection |
+ * | 3 | Request AccECN connection |
+ * +================+===========================+
+ */
+ use_accecn = tcp_ecn == 3;
+ use_ecn = tcp_ecn == 1 || tcp_ecn == 4 ||
+ tcp_ca_needs_ecn(sk) || bpf_needs_ecn || use_accecn;
if (!use_ecn) {
const struct dst_entry *dst = __sk_dst_get(sk);
@@ -142,23 +416,32 @@ static inline void tcp_ecn_send_syn(struct sock *sk, struct sk_buff *skb)
INET_ECN_xmit(sk);
TCP_SKB_CB(skb)->tcp_flags |= TCPHDR_ECE | TCPHDR_CWR;
- tcp_ecn_mode_set(tp, TCP_ECN_MODE_RFC3168);
+ if (use_accecn) {
+ TCP_SKB_CB(skb)->tcp_flags |= TCPHDR_AE;
+ tcp_ecn_mode_set(tp, TCP_ECN_MODE_PENDING);
+ tp->syn_ect_snt = inet_sk(sk)->tos & INET_ECN_MASK;
+ } else {
+ tcp_ecn_mode_set(tp, TCP_ECN_MODE_RFC3168);
+ }
}
}
static inline void tcp_ecn_clear_syn(struct sock *sk, struct sk_buff *skb)
{
- if (READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_ecn_fallback))
+ if (READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_ecn_fallback)) {
/* tp->ecn_flags are cleared at a later point in time when
* SYN ACK is ultimatively being received.
*/
- TCP_SKB_CB(skb)->tcp_flags &= ~(TCPHDR_ECE | TCPHDR_CWR);
+ TCP_SKB_CB(skb)->tcp_flags &= ~TCPHDR_ACE;
+ }
}
static inline void
tcp_ecn_make_synack(const struct request_sock *req, struct tcphdr *th)
{
- if (inet_rsk(req)->ecn_ok)
+ if (tcp_rsk(req)->accecn_ok)
+ tcp_accecn_echo_syn_ect(th, tcp_rsk(req)->syn_ect_rcv);
+ else if (inet_rsk(req)->ecn_ok)
th->ece = 1;
}
diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c
index eb0819463fae..e0c1e951c53c 100644
--- a/net/ipv4/syncookies.c
+++ b/net/ipv4/syncookies.c
@@ -403,6 +403,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
struct tcp_sock *tp = tcp_sk(sk);
struct inet_request_sock *ireq;
struct net *net = sock_net(sk);
+ struct tcp_request_sock *treq;
struct request_sock *req;
struct sock *ret = sk;
struct flowi4 fl4;
@@ -428,6 +429,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
}
ireq = inet_rsk(req);
+ treq = tcp_rsk(req);
sk_rcv_saddr_set(req_to_sk(req), ip_hdr(skb)->daddr);
sk_daddr_set(req_to_sk(req), ip_hdr(skb)->saddr);
@@ -483,6 +485,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
if (!req->syncookie)
ireq->rcv_wscale = rcv_wscale;
ireq->ecn_ok &= cookie_ecn_ok(net, &rt->dst);
+ treq->accecn_ok = ireq->ecn_ok && cookie_accecn_ok(th);
ret = tcp_get_cookie_sock(sk, skb, req, &rt->dst);
/* ip_queue_xmit() depends on our flow being setup
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index 3a43010d726f..75ec1a599b52 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -47,6 +47,7 @@ static unsigned int udp_child_hash_entries_max = UDP_HTABLE_SIZE_MAX;
static int tcp_plb_max_rounds = 31;
static int tcp_plb_max_cong_thresh = 256;
static unsigned int tcp_tw_reuse_delay_max = TCP_PAWS_MSL * MSEC_PER_SEC;
+static int tcp_ecn_mode_max = 5;
/* obsolete */
static int sysctl_tcp_low_latency __read_mostly;
@@ -728,7 +729,7 @@ static struct ctl_table ipv4_net_table[] = {
.mode = 0644,
.proc_handler = proc_dou8vec_minmax,
.extra1 = SYSCTL_ZERO,
- .extra2 = SYSCTL_TWO,
+ .extra2 = &tcp_ecn_mode_max,
},
{
.procname = "tcp_ecn_fallback",
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index c5e128d6ecc7..45c85c788def 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -3391,6 +3391,7 @@ int tcp_disconnect(struct sock *sk, int flags)
tp->window_clamp = 0;
tp->delivered = 0;
tp->delivered_ce = 0;
+ tp->accecn_fail_mode = 0;
tcp_accecn_init_counters(tp);
if (icsk->icsk_ca_initialized && icsk->icsk_ca_ops->release)
icsk->icsk_ca_ops->release(sk);
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 56c8514ee139..ea5d47354978 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -3665,8 +3665,18 @@ bool tcp_oow_rate_limited(struct net *net, const struct sk_buff *skb,
return __tcp_oow_rate_limited(net, mib_idx, last_oow_ack_time);
}
+static void tcp_send_ack_reflect_ect(struct sock *sk, bool accecn_reflector)
+{
+ struct tcp_sock *tp = tcp_sk(sk);
+ u16 flags = 0;
+
+ if (accecn_reflector)
+ flags = tcp_accecn_reflector_flags(tp->syn_ect_rcv);
+ __tcp_send_ack(sk, tp->rcv_nxt, flags);
+}
+
/* RFC 5961 7 [ACK Throttling] */
-static void tcp_send_challenge_ack(struct sock *sk)
+static void tcp_send_challenge_ack(struct sock *sk, bool accecn_reflector)
{
struct tcp_sock *tp = tcp_sk(sk);
struct net *net = sock_net(sk);
@@ -3696,7 +3706,7 @@ static void tcp_send_challenge_ack(struct sock *sk)
WRITE_ONCE(net->ipv4.tcp_challenge_count, count - 1);
send_ack:
NET_INC_STATS(net, LINUX_MIB_TCPCHALLENGEACK);
- tcp_send_ack(sk);
+ tcp_send_ack_reflect_ect(sk, accecn_reflector);
}
}
@@ -3863,7 +3873,7 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
/* RFC 5961 5.2 [Blind Data Injection Attack].[Mitigation] */
if (before(ack, prior_snd_una - max_window)) {
if (!(flag & FLAG_NO_CHALLENGE_ACK))
- tcp_send_challenge_ack(sk);
+ tcp_send_challenge_ack(sk, false);
return -SKB_DROP_REASON_TCP_TOO_OLD_ACK;
}
goto old_ack;
@@ -5882,6 +5892,7 @@ static bool tcp_validate_incoming(struct sock *sk, struct sk_buff *skb,
const struct tcphdr *th, int syn_inerr)
{
struct tcp_sock *tp = tcp_sk(sk);
+ bool accecn_reflector = false;
SKB_DR(reason);
/* RFC1323: H1. Apply PAWS check first. */
@@ -5974,7 +5985,7 @@ static bool tcp_validate_incoming(struct sock *sk, struct sk_buff *skb,
if (tp->syn_fastopen && !tp->data_segs_in &&
sk->sk_state == TCP_ESTABLISHED)
tcp_fastopen_active_disable(sk);
- tcp_send_challenge_ack(sk);
+ tcp_send_challenge_ack(sk, false);
SKB_DR_SET(reason, TCP_RESET);
goto discard;
}
@@ -5985,6 +5996,8 @@ static bool tcp_validate_incoming(struct sock *sk, struct sk_buff *skb,
* RFC 5961 4.2 : Send a challenge ack
*/
if (th->syn) {
+ if (tcp_ecn_mode_accecn(tp))
+ accecn_reflector = true;
if (sk->sk_state == TCP_SYN_RECV && sk->sk_socket && th->ack &&
TCP_SKB_CB(skb)->seq + 1 == TCP_SKB_CB(skb)->end_seq &&
TCP_SKB_CB(skb)->seq + 1 == tp->rcv_nxt &&
@@ -5994,7 +6007,7 @@ static bool tcp_validate_incoming(struct sock *sk, struct sk_buff *skb,
if (syn_inerr)
TCP_INC_STATS(sock_net(sk), TCP_MIB_INERRS);
NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPSYNCHALLENGE);
- tcp_send_challenge_ack(sk);
+ tcp_send_challenge_ack(sk, accecn_reflector);
SKB_DR_SET(reason, TCP_INVALID_SYN);
goto discard;
}
@@ -6459,7 +6472,8 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb,
* state to ESTABLISHED..."
*/
- tcp_ecn_rcv_synack(tp, th);
+ if (tcp_ecn_mode_any(tp))
+ tcp_ecn_rcv_synack(sk, th, TCP_SKB_CB(skb)->ip_dsfield);
tcp_init_wl(tp, TCP_SKB_CB(skb)->seq);
tcp_try_undo_spurious_syn(sk);
@@ -6531,7 +6545,7 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb,
TCP_DELACK_MAX, false);
goto consume;
}
- tcp_send_ack(sk);
+ tcp_send_ack_reflect_ect(sk, tcp_ecn_mode_accecn(tp));
return -1;
}
@@ -6590,7 +6604,7 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb,
tp->snd_wl1 = TCP_SKB_CB(skb)->seq;
tp->max_window = tp->snd_wnd;
- tcp_ecn_rcv_syn(tp, th);
+ tcp_ecn_rcv_syn(tp, th, skb);
tcp_mtup_init(sk);
tcp_sync_mss(sk, icsk->icsk_pmtu_cookie);
@@ -6772,7 +6786,7 @@ tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb)
}
/* accept old ack during closing */
if ((int)reason < 0) {
- tcp_send_challenge_ack(sk);
+ tcp_send_challenge_ack(sk, false);
reason = -reason;
goto discard;
}
@@ -6819,9 +6833,12 @@ tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb)
tp->lsndtime = tcp_jiffies32;
tcp_initialize_rcv_mss(sk);
+ if (tcp_ecn_mode_accecn(tp))
+ tcp_accecn_third_ack(sk, skb, tp->syn_ect_snt);
tcp_fast_path_on(tp);
if (sk->sk_shutdown & SEND_SHUTDOWN)
tcp_shutdown(sk, SEND_SHUTDOWN);
+
break;
case TCP_FIN_WAIT1: {
@@ -6991,6 +7008,15 @@ static void tcp_ecn_create_request(struct request_sock *req,
bool ect, ecn_ok;
u32 ecn_ok_dst;
+ if (tcp_accecn_syn_requested(th) &&
+ READ_ONCE(net->ipv4.sysctl_tcp_ecn) >= 3) {
+ inet_rsk(req)->ecn_ok = 1;
+ tcp_rsk(req)->accecn_ok = 1;
+ tcp_rsk(req)->syn_ect_rcv = TCP_SKB_CB(skb)->ip_dsfield &
+ INET_ECN_MASK;
+ return;
+ }
+
if (!th_ecn)
return;
@@ -6998,7 +7024,8 @@ static void tcp_ecn_create_request(struct request_sock *req,
ecn_ok_dst = dst_feature(dst, DST_FEATURE_ECN_MASK);
ecn_ok = READ_ONCE(net->ipv4.sysctl_tcp_ecn) || ecn_ok_dst;
- if (((!ect || th->res1) && ecn_ok) || tcp_ca_needs_ecn(listen_sk) ||
+ if (((!ect || th->res1 || th->ae) && ecn_ok) ||
+ tcp_ca_needs_ecn(listen_sk) ||
(ecn_ok_dst & DST_FEATURE_ECN_CA) ||
tcp_bpf_ca_needs_ecn((struct sock *)req))
inet_rsk(req)->ecn_ok = 1;
@@ -7016,6 +7043,9 @@ static void tcp_openreq_init(struct request_sock *req,
tcp_rsk(req)->snt_synack = 0;
tcp_rsk(req)->snt_tsval_first = 0;
tcp_rsk(req)->last_oow_ack_time = 0;
+ tcp_rsk(req)->accecn_ok = 0;
+ tcp_rsk(req)->syn_ect_rcv = 0;
+ tcp_rsk(req)->syn_ect_snt = 0;
req->mss = rx_opt->mss_clamp;
req->ts_recent = rx_opt->saw_tstamp ? rx_opt->rcv_tsval : 0;
ireq->tstamp_ok = rx_opt->tstamp_ok;
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index a847d894ace3..b95e4ed227cb 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -64,6 +64,7 @@
#include <net/icmp.h>
#include <net/inet_hashtables.h>
#include <net/tcp.h>
+#include <net/tcp_ecn.h>
#include <net/transp_v6.h>
#include <net/ipv6.h>
#include <net/inet_common.h>
@@ -1190,7 +1191,7 @@ static int tcp_v4_send_synack(const struct sock *sk, struct dst_entry *dst,
enum tcp_synack_type synack_type,
struct sk_buff *syn_skb)
{
- const struct inet_request_sock *ireq = inet_rsk(req);
+ struct inet_request_sock *ireq = inet_rsk(req);
struct flowi4 fl4;
int err = -1;
struct sk_buff *skb;
@@ -1203,6 +1204,7 @@ static int tcp_v4_send_synack(const struct sock *sk, struct dst_entry *dst,
skb = tcp_make_synack(sk, dst, req, foc, synack_type, syn_skb);
if (skb) {
+ tcp_rsk(req)->syn_ect_snt = inet_sk(sk)->tos & INET_ECN_MASK;
__tcp_v4_send_check(skb, ireq->ir_loc_addr, ireq->ir_rmt_addr);
tos = READ_ONCE(inet_sk(sk)->tos);
@@ -3448,7 +3450,7 @@ static void __net_init tcp_set_hashinfo(struct net *net)
static int __net_init tcp_sk_init(struct net *net)
{
- net->ipv4.sysctl_tcp_ecn = 2;
+ net->ipv4.sysctl_tcp_ecn = TCP_ECN_IN_ECN_OUT_NOECN;
net->ipv4.sysctl_tcp_ecn_fallback = 1;
net->ipv4.sysctl_tcp_base_mss = TCP_BASE_MSS;
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index 2994c9222c9c..845b1c81b3b0 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -20,6 +20,7 @@
*/
#include <net/tcp.h>
+#include <net/tcp_ecn.h>
#include <net/xfrm.h>
#include <net/busy_poll.h>
#include <net/rstreason.h>
@@ -461,12 +462,23 @@ void tcp_openreq_init_rwin(struct request_sock *req,
ireq->rcv_wscale = rcv_wscale;
}
-static void tcp_ecn_openreq_child(struct tcp_sock *tp,
- const struct request_sock *req)
+static void tcp_ecn_openreq_child(struct sock *sk,
+ const struct request_sock *req,
+ const struct sk_buff *skb)
{
- tcp_ecn_mode_set(tp, inet_rsk(req)->ecn_ok ?
- TCP_ECN_MODE_RFC3168 :
- TCP_ECN_DISABLED);
+ const struct tcp_request_sock *treq = tcp_rsk(req);
+ struct tcp_sock *tp = tcp_sk(sk);
+
+ if (treq->accecn_ok) {
+ tcp_ecn_mode_set(tp, TCP_ECN_MODE_ACCECN);
+ tp->syn_ect_snt = treq->syn_ect_snt;
+ tcp_accecn_third_ack(sk, skb, treq->syn_ect_snt);
+ tcp_ecn_received_counters(sk, skb);
+ } else {
+ tcp_ecn_mode_set(tp, inet_rsk(req)->ecn_ok ?
+ TCP_ECN_MODE_RFC3168 :
+ TCP_ECN_DISABLED);
+ }
}
void tcp_ca_openreq_child(struct sock *sk, const struct dst_entry *dst)
@@ -631,7 +643,7 @@ struct sock *tcp_create_openreq_child(const struct sock *sk,
if (skb->len >= TCP_MSS_DEFAULT + newtp->tcp_header_len)
newicsk->icsk_ack.last_seg_size = skb->len - newtp->tcp_header_len;
newtp->rx_opt.mss_clamp = req->mss;
- tcp_ecn_openreq_child(newtp, req);
+ tcp_ecn_openreq_child(newsk, req, skb);
newtp->fastopen_req = NULL;
RCU_INIT_POINTER(newtp->fastopen_rsk, NULL);
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 78018870a389..debcb1d431e6 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -332,8 +332,9 @@ static void tcp_ecn_send(struct sock *sk, struct sk_buff *skb,
return;
if (tcp_ecn_mode_accecn(tp)) {
- INET_ECN_xmit(sk);
- tcp_accecn_set_ace(th, tp);
+ if (!tcp_accecn_ace_fail_recv(tp))
+ INET_ECN_xmit(sk);
+ tcp_accecn_set_ace(tp, skb, th);
skb_shinfo(skb)->gso_type |= SKB_GSO_TCP_ACCECN;
} else {
/* Not-retransmitted data segment: set ECT and inject CWR. */
@@ -3343,7 +3344,10 @@ int __tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb, int segs)
tcp_retrans_try_collapse(sk, skb, avail_wnd);
}
- /* RFC3168, section 6.1.1.1. ECN fallback */
+ /* RFC3168, section 6.1.1.1. ECN fallback
+ * As AccECN uses the same SYN flags (+ AE), this check covers both
+ * cases.
+ */
if ((TCP_SKB_CB(skb)->tcp_flags & TCPHDR_SYN_ECN) == TCPHDR_SYN_ECN)
tcp_ecn_clear_syn(sk, skb);
diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c
index f0ee1a909771..f0b87a3268ae 100644
--- a/net/ipv6/syncookies.c
+++ b/net/ipv6/syncookies.c
@@ -264,6 +264,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
if (!req->syncookie)
ireq->rcv_wscale = rcv_wscale;
ireq->ecn_ok &= cookie_ecn_ok(net, dst);
+ tcp_rsk(req)->accecn_ok = ireq->ecn_ok && cookie_accecn_ok(th);
ret = tcp_get_cookie_sock(sk, skb, req, dst);
if (!ret) {
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 8f2c3cba1f1f..1356fbc87fa2 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -545,6 +545,7 @@ static int tcp_v6_send_synack(const struct sock *sk, struct dst_entry *dst,
skb = tcp_make_synack(sk, dst, req, foc, synack_type, syn_skb);
if (skb) {
+ tcp_rsk(req)->syn_ect_snt = np->tclass & INET_ECN_MASK;
__tcp_v6_send_check(skb, &ireq->ir_v6_loc_addr,
&ireq->ir_v6_rmt_addr);
--
2.34.1
^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH v11 net-next 07/15] tcp: Add wait_third_ack flag for ECN negotiation in simultaneous connect
2025-07-03 5:17 [PATCH v11 net-next 00/15] AccECN protocol patch series chia-yu.chang
` (5 preceding siblings ...)
2025-07-03 5:17 ` [PATCH v11 net-next 06/15] tcp: accecn: AccECN negotiation chia-yu.chang
@ 2025-07-03 5:17 ` chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 08/15] tcp: accecn: add AccECN rx byte counters chia-yu.chang
` (7 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: chia-yu.chang @ 2025-07-03 5:17 UTC (permalink / raw)
To: pabeni, edumazet, linux-doc, corbet, horms, dsahern, kuniyu, bpf,
netdev, dave.taht, jhs, kuba, stephen, xiyou.wangcong, jiri,
davem, andrew+netdev, donald.hunter, ast, liuhangbin, shuah,
linux-kselftest, ij, ncardwell, koen.de_schepper, g.white,
ingemar.s.johansson, mirja.kuehlewind, cheshire, rs.ietf,
Jason_Livingood, vidhi_goel
Cc: Chia-Yu Chang
From: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
In simultaneous connect scenario, the connection will be established
when SYN/ACK is received after patch 23e89e8ee7be7. However, the
third ACK is still anticipated to complete the negotiation for either
RFC3168 ECN and Accurate ECN. In this sense, an additional flag
wait_third_ack is introduced to identify that the 3rd ACK is still
anticipated to ensure ECN or AccECN negotiation will be done in the
ESTABLISHED state.
Signed-off-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
Co-developed-by: Ilpo Järvinen <ij@kernel.org>
Signed-off-by: Ilpo Järvinen <ij@kernel.org>
---
.../networking/net_cachelines/tcp_sock.rst | 1 +
include/linux/tcp.h | 3 ++-
net/ipv4/tcp.c | 1 +
net/ipv4/tcp_input.c | 23 +++++++++++++++----
4 files changed, 23 insertions(+), 5 deletions(-)
diff --git a/Documentation/networking/net_cachelines/tcp_sock.rst b/Documentation/networking/net_cachelines/tcp_sock.rst
index 4f71ece7c655..22ac668fe6c7 100644
--- a/Documentation/networking/net_cachelines/tcp_sock.rst
+++ b/Documentation/networking/net_cachelines/tcp_sock.rst
@@ -105,6 +105,7 @@ u32 received_ce read_mostly read_w
u8:4 received_ce_pending read_mostly read_write
u8:2 syn_ect_snt write_mostly read_write
u8:2 syn_ect_rcv read_mostly read_write
+u8:1 wait_third_ack read_write
u8:4 accecn_fail_mode
u32 lost read_mostly tcp_ack
u32 app_limited read_write read_mostly tcp_rate_check_app_limited,tcp_rate_skb_sent(tx);tcp_rate_gen(rx)
diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 994a6e7256e1..1d8301f2883c 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -380,7 +380,8 @@ struct tcp_sock {
u8 dup_ack_counter:2,
tlp_retrans:1, /* TLP is a retransmission */
syn_ect_snt:2, /* AccECN ECT memory, only */
- syn_ect_rcv:2; /* ... needed durign 3WHS + first seqno */
+ syn_ect_rcv:2, /* ... needed durign 3WHS + first seqno */
+ wait_third_ack:1; /* Wait 3rd ACK in simultaneous open */
u8 thin_lto : 1,/* Use linear timeouts for thin streams */
fastopen_connect:1, /* FASTOPEN_CONNECT sockopt */
fastopen_no_cookie:1, /* Allow send/recv SYN+data without a cookie */
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 45c85c788def..0548c74458c6 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -3391,6 +3391,7 @@ int tcp_disconnect(struct sock *sk, int flags)
tp->window_clamp = 0;
tp->delivered = 0;
tp->delivered_ce = 0;
+ tp->wait_third_ack = 0;
tp->accecn_fail_mode = 0;
tcp_accecn_init_counters(tp);
if (icsk->icsk_ca_initialized && icsk->icsk_ca_ops->release)
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index ea5d47354978..7ccfde9bcfda 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -6001,8 +6001,13 @@ static bool tcp_validate_incoming(struct sock *sk, struct sk_buff *skb,
if (sk->sk_state == TCP_SYN_RECV && sk->sk_socket && th->ack &&
TCP_SKB_CB(skb)->seq + 1 == TCP_SKB_CB(skb)->end_seq &&
TCP_SKB_CB(skb)->seq + 1 == tp->rcv_nxt &&
- TCP_SKB_CB(skb)->ack_seq == tp->snd_nxt)
+ TCP_SKB_CB(skb)->ack_seq == tp->snd_nxt) {
+ if (!tcp_ecn_disabled(tp)) {
+ tp->wait_third_ack = true;
+ tcp_send_ack_reflect_ect(sk, accecn_reflector);
+ }
goto pass;
+ }
syn_challenge:
if (syn_inerr)
TCP_INC_STATS(sock_net(sk), TCP_MIB_INERRS);
@@ -6220,6 +6225,12 @@ void tcp_rcv_established(struct sock *sk, struct sk_buff *skb)
return;
step5:
+ if (unlikely(tp->wait_third_ack)) {
+ tp->wait_third_ack = 0;
+ if (tcp_ecn_mode_accecn(tp))
+ tcp_accecn_third_ack(sk, skb, tp->syn_ect_snt);
+ tcp_fast_path_on(tp);
+ }
tcp_ecn_received_counters(sk, skb);
reason = tcp_ack(sk, skb, FLAG_SLOWPATH | FLAG_UPDATE_TS_RECENT);
@@ -6833,12 +6844,16 @@ tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb)
tp->lsndtime = tcp_jiffies32;
tcp_initialize_rcv_mss(sk);
- if (tcp_ecn_mode_accecn(tp))
- tcp_accecn_third_ack(sk, skb, tp->syn_ect_snt);
- tcp_fast_path_on(tp);
+ if (likely(!tp->wait_third_ack)) {
+ if (tcp_ecn_mode_accecn(tp))
+ tcp_accecn_third_ack(sk, skb, tp->syn_ect_snt);
+ tcp_fast_path_on(tp);
+ }
if (sk->sk_shutdown & SEND_SHUTDOWN)
tcp_shutdown(sk, SEND_SHUTDOWN);
+ if (sk->sk_socket && tp->wait_third_ack)
+ goto consume;
break;
case TCP_FIN_WAIT1: {
--
2.34.1
^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH v11 net-next 08/15] tcp: accecn: add AccECN rx byte counters
2025-07-03 5:17 [PATCH v11 net-next 00/15] AccECN protocol patch series chia-yu.chang
` (6 preceding siblings ...)
2025-07-03 5:17 ` [PATCH v11 net-next 07/15] tcp: Add wait_third_ack flag for ECN negotiation in simultaneous connect chia-yu.chang
@ 2025-07-03 5:17 ` chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 09/15] tcp: accecn: AccECN needs to know delivered bytes chia-yu.chang
` (6 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: chia-yu.chang @ 2025-07-03 5:17 UTC (permalink / raw)
To: pabeni, edumazet, linux-doc, corbet, horms, dsahern, kuniyu, bpf,
netdev, dave.taht, jhs, kuba, stephen, xiyou.wangcong, jiri,
davem, andrew+netdev, donald.hunter, ast, liuhangbin, shuah,
linux-kselftest, ij, ncardwell, koen.de_schepper, g.white,
ingemar.s.johansson, mirja.kuehlewind, cheshire, rs.ietf,
Jason_Livingood, vidhi_goel
Cc: Chia-Yu Chang
From: Ilpo Järvinen <ij@kernel.org>
These three byte counters track IP ECN field payload byte sums for
all arriving (acceptable) packets for ECT0, ECT1, and CE. The
AccECN option (added by a later patch in the series) echoes these
counters back to sender side; therefore, it is placed within the
group of tcp_sock_write_txrx.
Below are the pahole outcomes before and after this patch, in which
the group size of tcp_sock_write_txrx is increased from 95 + 4 to
107 + 4 and an extra 4-byte hole is created but will be exploited
in later patches:
[BEFORE THIS PATCH]
struct tcp_sock {
[...]
u32 delivered_ce; /* 2640 4 */
u32 received_ce; /* 2644 4 */
u32 app_limited; /* 2648 4 */
u32 rcv_wnd; /* 2652 4 */
struct tcp_options_received rx_opt; /* 2656 24 */
__cacheline_group_end__tcp_sock_write_txrx[0]; /* 2680 0 */
[...]
/* size: 3264, cachelines: 51, members: 169 */
}
[AFTER THIS PATCH]
struct tcp_sock {
[...]
u32 delivered_ce; /* 2640 4 */
u32 received_ce; /* 2644 4 */
u32 received_ecn_bytes[3];/* 2648 12 */
u32 app_limited; /* 2660 4 */
u32 rcv_wnd; /* 2664 4 */
struct tcp_options_received rx_opt; /* 2668 24 */
__cacheline_group_end__tcp_sock_write_txrx[0]; /* 2692 0 */
/* XXX 4 bytes hole, try to pack */
[...]
/* size: 3264, cachelines: 51, members: 170 */
}
Signed-off-by: Ilpo Järvinen <ij@kernel.org>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Co-developed-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
Signed-off-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
---
v8:
- Add new helper function tcp_ecn_received_counters_payload()
---
.../networking/net_cachelines/tcp_sock.rst | 1 +
include/linux/tcp.h | 4 +++
include/net/tcp.h | 3 +++
include/net/tcp_ecn.h | 26 ++++++++++++++++++-
net/ipv4/syncookies.c | 1 +
net/ipv4/tcp.c | 3 ++-
net/ipv4/tcp_input.c | 7 ++---
net/ipv4/tcp_minisocks.c | 2 +-
net/ipv6/syncookies.c | 1 +
9 files changed, 42 insertions(+), 6 deletions(-)
diff --git a/Documentation/networking/net_cachelines/tcp_sock.rst b/Documentation/networking/net_cachelines/tcp_sock.rst
index 22ac668fe6c7..804480d39132 100644
--- a/Documentation/networking/net_cachelines/tcp_sock.rst
+++ b/Documentation/networking/net_cachelines/tcp_sock.rst
@@ -102,6 +102,7 @@ u32 prr_out read_mostly read_m
u32 delivered read_mostly read_write tcp_rate_skb_sent, tcp_newly_delivered(tx);tcp_ack, tcp_rate_gen, tcp_clean_rtx_queue (rx)
u32 delivered_ce read_mostly read_write tcp_rate_skb_sent(tx);tcp_rate_gen(rx)
u32 received_ce read_mostly read_write
+u32[3] received_ecn_bytes read_mostly read_write
u8:4 received_ce_pending read_mostly read_write
u8:2 syn_ect_snt write_mostly read_write
u8:2 syn_ect_rcv read_mostly read_write
diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 1d8301f2883c..0c2331e186e8 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -306,6 +306,10 @@ struct tcp_sock {
u32 delivered; /* Total data packets delivered incl. rexmits */
u32 delivered_ce; /* Like the above but only ECE marked packets */
u32 received_ce; /* Like the above but for rcvd CE marked pkts */
+ u32 received_ecn_bytes[3]; /* received byte counters for three ECN
+ * types: INET_ECN_ECT_1, INET_ECN_ECT_0,
+ * and INET_ECN_CE
+ */
u32 app_limited; /* limited until "delivered" reaches this val */
u32 rcv_wnd; /* Current receiver window */
/*
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 61b103633da4..0d8e1a676dad 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -971,6 +971,9 @@ static inline u32 tcp_rsk_tsval(const struct tcp_request_sock *treq)
* See draft-ietf-tcpm-accurate-ecn for the latest values.
*/
#define TCP_ACCECN_CEP_INIT_OFFSET 5
+#define TCP_ACCECN_E1B_INIT_OFFSET 1
+#define TCP_ACCECN_E0B_INIT_OFFSET 1
+#define TCP_ACCECN_CEB_INIT_OFFSET 0
/* State flags for sacked in struct tcp_skb_cb */
enum tcp_skb_cb_sacked_flags {
diff --git a/include/net/tcp_ecn.h b/include/net/tcp_ecn.h
index 7ac68cb70ef3..68e392ba2f9e 100644
--- a/include/net/tcp_ecn.h
+++ b/include/net/tcp_ecn.h
@@ -176,7 +176,7 @@ static inline void tcp_accecn_third_ack(struct sock *sk,
/* Updates Accurate ECN received counters from the received IP ECN field */
static inline void tcp_ecn_received_counters(struct sock *sk,
- const struct sk_buff *skb)
+ const struct sk_buff *skb, u32 len)
{
u8 ecnfield = TCP_SKB_CB(skb)->ip_dsfield & INET_ECN_MASK;
u8 is_ce = INET_ECN_is_ce(ecnfield);
@@ -196,9 +196,21 @@ static inline void tcp_ecn_received_counters(struct sock *sk,
tp->received_ce += pcount;
tp->received_ce_pending = min(tp->received_ce_pending + pcount,
0xfU);
+
+ if (len > 0)
+ tp->received_ecn_bytes[ecnfield - 1] += len;
}
}
+
+static inline void tcp_ecn_received_counters_payload(struct sock *sk,
+ const struct sk_buff *skb)
+{
+ const struct tcphdr *th = (const struct tcphdr *)skb->data;
+
+ tcp_ecn_received_counters(sk, skb, skb->len - th->doff * 4);
+}
+
/* AccECN specification, 5.1: [...] a server can determine that it
* negotiated AccECN as [...] if the ACK contains an ACE field with
* the value 0b010 to 0b111 (decimal 2 to 7).
@@ -242,10 +254,22 @@ static inline bool tcp_accecn_syn_requested(const struct tcphdr *th)
return ace && ace != 0x3;
}
+static inline void __tcp_accecn_init_bytes_counters(int *counter_array)
+{
+ BUILD_BUG_ON(INET_ECN_ECT_1 != 0x1);
+ BUILD_BUG_ON(INET_ECN_ECT_0 != 0x2);
+ BUILD_BUG_ON(INET_ECN_CE != 0x3);
+
+ counter_array[INET_ECN_ECT_1 - 1] = 0;
+ counter_array[INET_ECN_ECT_0 - 1] = 0;
+ counter_array[INET_ECN_CE - 1] = 0;
+}
+
static inline void tcp_accecn_init_counters(struct tcp_sock *tp)
{
tp->received_ce = 0;
tp->received_ce_pending = 0;
+ __tcp_accecn_init_bytes_counters(tp->received_ecn_bytes);
}
/* Used for make_synack to form the ACE flags */
diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c
index e0c1e951c53c..569befcf021b 100644
--- a/net/ipv4/syncookies.c
+++ b/net/ipv4/syncookies.c
@@ -12,6 +12,7 @@
#include <linux/export.h>
#include <net/secure_seq.h>
#include <net/tcp.h>
+#include <net/tcp_ecn.h>
#include <net/route.h>
static siphash_aligned_key_t syncookie_secret[2];
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 0548c74458c6..237435927b9a 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -5121,6 +5121,7 @@ static void __init tcp_struct_check(void)
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_txrx, delivered);
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_txrx, delivered_ce);
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_txrx, received_ce);
+ CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_txrx, received_ecn_bytes);
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_txrx, app_limited);
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_txrx, rcv_wnd);
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_txrx, rx_opt);
@@ -5128,7 +5129,7 @@ static void __init tcp_struct_check(void)
/* 32bit arches with 8byte alignment on u64 fields might need padding
* before tcp_clock_cache.
*/
- CACHELINE_ASSERT_GROUP_SIZE(struct tcp_sock, tcp_sock_write_txrx, 95 + 4);
+ CACHELINE_ASSERT_GROUP_SIZE(struct tcp_sock, tcp_sock_write_txrx, 107 + 4);
/* RX read-write hotpath cache lines */
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_rx, bytes_received);
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 7ccfde9bcfda..eea790295e54 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -6138,7 +6138,7 @@ void tcp_rcv_established(struct sock *sk, struct sk_buff *skb)
flag |= __tcp_replace_ts_recent(tp,
delta);
- tcp_ecn_received_counters(sk, skb);
+ tcp_ecn_received_counters(sk, skb, 0);
/* We know that such packets are checksummed
* on entry.
@@ -6184,7 +6184,8 @@ void tcp_rcv_established(struct sock *sk, struct sk_buff *skb)
/* Bulk data transfer: receiver */
tcp_cleanup_skb(skb);
__skb_pull(skb, tcp_header_len);
- tcp_ecn_received_counters(sk, skb);
+ tcp_ecn_received_counters(sk, skb,
+ len - tcp_header_len);
eaten = tcp_queue_rcv(sk, skb, &fragstolen);
tcp_event_data_recv(sk, skb);
@@ -6231,7 +6232,7 @@ void tcp_rcv_established(struct sock *sk, struct sk_buff *skb)
tcp_accecn_third_ack(sk, skb, tp->syn_ect_snt);
tcp_fast_path_on(tp);
}
- tcp_ecn_received_counters(sk, skb);
+ tcp_ecn_received_counters_payload(sk, skb);
reason = tcp_ack(sk, skb, FLAG_SLOWPATH | FLAG_UPDATE_TS_RECENT);
if ((int)reason < 0) {
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index 845b1c81b3b0..0a0ee33e38ad 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -473,7 +473,7 @@ static void tcp_ecn_openreq_child(struct sock *sk,
tcp_ecn_mode_set(tp, TCP_ECN_MODE_ACCECN);
tp->syn_ect_snt = treq->syn_ect_snt;
tcp_accecn_third_ack(sk, skb, treq->syn_ect_snt);
- tcp_ecn_received_counters(sk, skb);
+ tcp_ecn_received_counters_payload(sk, skb);
} else {
tcp_ecn_mode_set(tp, inet_rsk(req)->ecn_ok ?
TCP_ECN_MODE_RFC3168 :
diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c
index f0b87a3268ae..7e007f013ec8 100644
--- a/net/ipv6/syncookies.c
+++ b/net/ipv6/syncookies.c
@@ -16,6 +16,7 @@
#include <net/secure_seq.h>
#include <net/ipv6.h>
#include <net/tcp.h>
+#include <net/tcp_ecn.h>
#define COOKIEBITS 24 /* Upper bits store count */
#define COOKIEMASK (((__u32)1 << COOKIEBITS) - 1)
--
2.34.1
^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH v11 net-next 09/15] tcp: accecn: AccECN needs to know delivered bytes
2025-07-03 5:17 [PATCH v11 net-next 00/15] AccECN protocol patch series chia-yu.chang
` (7 preceding siblings ...)
2025-07-03 5:17 ` [PATCH v11 net-next 08/15] tcp: accecn: add AccECN rx byte counters chia-yu.chang
@ 2025-07-03 5:17 ` chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 10/15] tcp: sack option handling improvements chia-yu.chang
` (5 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: chia-yu.chang @ 2025-07-03 5:17 UTC (permalink / raw)
To: pabeni, edumazet, linux-doc, corbet, horms, dsahern, kuniyu, bpf,
netdev, dave.taht, jhs, kuba, stephen, xiyou.wangcong, jiri,
davem, andrew+netdev, donald.hunter, ast, liuhangbin, shuah,
linux-kselftest, ij, ncardwell, koen.de_schepper, g.white,
ingemar.s.johansson, mirja.kuehlewind, cheshire, rs.ietf,
Jason_Livingood, vidhi_goel
Cc: Chia-Yu Chang
From: Ilpo Järvinen <ij@kernel.org>
AccECN byte counter estimation requires delivered bytes
which can be calculated while processing SACK blocks and
cumulative ACK. The delivered bytes will be used to estimate
the byte counters between AccECN option (on ACKs w/o the
option).
Non-SACK calculation is quite annoying, inaccurate, and
likely bogus.
Signed-off-by: Ilpo Järvinen <ij@kernel.org>
Signed-off-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
---
net/ipv4/tcp_input.c | 14 ++++++++++++--
1 file changed, 12 insertions(+), 2 deletions(-)
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index eea790295e54..f7d7649612a2 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -1050,6 +1050,7 @@ struct tcp_sacktag_state {
u64 last_sackt;
u32 reord;
u32 sack_delivered;
+ u32 delivered_bytes;
int flag;
unsigned int mss_now;
struct rate_sample *rate;
@@ -1411,7 +1412,7 @@ static int tcp_match_skb_to_sack(struct sock *sk, struct sk_buff *skb,
static u8 tcp_sacktag_one(struct sock *sk,
struct tcp_sacktag_state *state, u8 sacked,
u32 start_seq, u32 end_seq,
- int dup_sack, int pcount,
+ int dup_sack, int pcount, u32 plen,
u64 xmit_time)
{
struct tcp_sock *tp = tcp_sk(sk);
@@ -1471,6 +1472,7 @@ static u8 tcp_sacktag_one(struct sock *sk,
tp->sacked_out += pcount;
/* Out-of-order packets delivered */
state->sack_delivered += pcount;
+ state->delivered_bytes += plen;
}
/* D-SACK. We can detect redundant retransmission in S|R and plain R
@@ -1507,7 +1509,7 @@ static bool tcp_shifted_skb(struct sock *sk, struct sk_buff *prev,
* tcp_highest_sack_seq() when skb is highest_sack.
*/
tcp_sacktag_one(sk, state, TCP_SKB_CB(skb)->sacked,
- start_seq, end_seq, dup_sack, pcount,
+ start_seq, end_seq, dup_sack, pcount, skb->len,
tcp_skb_timestamp_us(skb));
tcp_rate_skb_delivered(sk, skb, state->rate);
@@ -1792,6 +1794,7 @@ static struct sk_buff *tcp_sacktag_walk(struct sk_buff *skb, struct sock *sk,
TCP_SKB_CB(skb)->end_seq,
dup_sack,
tcp_skb_pcount(skb),
+ skb->len,
tcp_skb_timestamp_us(skb));
tcp_rate_skb_delivered(sk, skb, state->rate);
if (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED)
@@ -3300,6 +3303,8 @@ static int tcp_clean_rtx_queue(struct sock *sk, const struct sk_buff *ack_skb,
if (sacked & TCPCB_SACKED_ACKED) {
tp->sacked_out -= acked_pcount;
+ /* snd_una delta covers these skbs */
+ sack->delivered_bytes -= skb->len;
} else if (tcp_is_sack(tp)) {
tcp_count_delivered(tp, acked_pcount, ece_ack);
if (!tcp_skb_spurious_retrans(tp, skb))
@@ -3396,6 +3401,10 @@ static int tcp_clean_rtx_queue(struct sock *sk, const struct sk_buff *ack_skb,
if (before(reord, prior_fack))
tcp_check_sack_reordering(sk, reord, 0);
}
+
+ sack->delivered_bytes = (skb ?
+ TCP_SKB_CB(skb)->seq : tp->snd_una) -
+ prior_snd_una;
} else if (skb && rtt_update && sack_rtt_us >= 0 &&
sack_rtt_us > tcp_stamp_us_delta(tp->tcp_mstamp,
tcp_skb_timestamp_us(skb))) {
@@ -3858,6 +3867,7 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
sack_state.first_sackt = 0;
sack_state.rate = &rs;
sack_state.sack_delivered = 0;
+ sack_state.delivered_bytes = 0;
/* We very likely will need to access rtx queue. */
prefetch(sk->tcp_rtx_queue.rb_node);
--
2.34.1
^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH v11 net-next 10/15] tcp: sack option handling improvements
2025-07-03 5:17 [PATCH v11 net-next 00/15] AccECN protocol patch series chia-yu.chang
` (8 preceding siblings ...)
2025-07-03 5:17 ` [PATCH v11 net-next 09/15] tcp: accecn: AccECN needs to know delivered bytes chia-yu.chang
@ 2025-07-03 5:17 ` chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 11/15] tcp: accecn: AccECN option chia-yu.chang
` (4 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: chia-yu.chang @ 2025-07-03 5:17 UTC (permalink / raw)
To: pabeni, edumazet, linux-doc, corbet, horms, dsahern, kuniyu, bpf,
netdev, dave.taht, jhs, kuba, stephen, xiyou.wangcong, jiri,
davem, andrew+netdev, donald.hunter, ast, liuhangbin, shuah,
linux-kselftest, ij, ncardwell, koen.de_schepper, g.white,
ingemar.s.johansson, mirja.kuehlewind, cheshire, rs.ietf,
Jason_Livingood, vidhi_goel
Cc: Chia-Yu Chang
From: Ilpo Järvinen <ij@kernel.org>
1) Don't early return when sack doesn't fit. AccECN code will be
placed after this fragment so no early returns please.
2) Make sure opts->num_sack_blocks is not left undefined. E.g.,
tcp_current_mss() does not memset its opts struct to zero.
AccECN code checks if SACK option is present and may even
alter it to make room for AccECN option when many SACK blocks
are present. Thus, num_sack_blocks needs to be always valid.
Signed-off-by: Ilpo Järvinen <ij@kernel.org>
Signed-off-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
---
v8:
- Set opts->num_sack_blocks=0 to avoid potential undefined value
---
net/ipv4/tcp_output.c | 25 ++++++++++++++-----------
1 file changed, 14 insertions(+), 11 deletions(-)
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index debcb1d431e6..d98a1a17eb52 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -985,17 +985,20 @@ static unsigned int tcp_established_options(struct sock *sk, struct sk_buff *skb
eff_sacks = tp->rx_opt.num_sacks + tp->rx_opt.dsack;
if (unlikely(eff_sacks)) {
const unsigned int remaining = MAX_TCP_OPTION_SPACE - size;
- if (unlikely(remaining < TCPOLEN_SACK_BASE_ALIGNED +
- TCPOLEN_SACK_PERBLOCK))
- return size;
-
- opts->num_sack_blocks =
- min_t(unsigned int, eff_sacks,
- (remaining - TCPOLEN_SACK_BASE_ALIGNED) /
- TCPOLEN_SACK_PERBLOCK);
-
- size += TCPOLEN_SACK_BASE_ALIGNED +
- opts->num_sack_blocks * TCPOLEN_SACK_PERBLOCK;
+ if (likely(remaining >= TCPOLEN_SACK_BASE_ALIGNED +
+ TCPOLEN_SACK_PERBLOCK)) {
+ opts->num_sack_blocks =
+ min_t(unsigned int, eff_sacks,
+ (remaining - TCPOLEN_SACK_BASE_ALIGNED) /
+ TCPOLEN_SACK_PERBLOCK);
+
+ size += TCPOLEN_SACK_BASE_ALIGNED +
+ opts->num_sack_blocks * TCPOLEN_SACK_PERBLOCK;
+ } else {
+ opts->num_sack_blocks = 0;
+ }
+ } else {
+ opts->num_sack_blocks = 0;
}
if (unlikely(BPF_SOCK_OPS_TEST_FLAG(tp,
--
2.34.1
^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH v11 net-next 11/15] tcp: accecn: AccECN option
2025-07-03 5:17 [PATCH v11 net-next 00/15] AccECN protocol patch series chia-yu.chang
` (9 preceding siblings ...)
2025-07-03 5:17 ` [PATCH v11 net-next 10/15] tcp: sack option handling improvements chia-yu.chang
@ 2025-07-03 5:17 ` chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 12/15] tcp: accecn: AccECN option send control chia-yu.chang
` (3 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: chia-yu.chang @ 2025-07-03 5:17 UTC (permalink / raw)
To: pabeni, edumazet, linux-doc, corbet, horms, dsahern, kuniyu, bpf,
netdev, dave.taht, jhs, kuba, stephen, xiyou.wangcong, jiri,
davem, andrew+netdev, donald.hunter, ast, liuhangbin, shuah,
linux-kselftest, ij, ncardwell, koen.de_schepper, g.white,
ingemar.s.johansson, mirja.kuehlewind, cheshire, rs.ietf,
Jason_Livingood, vidhi_goel
Cc: Chia-Yu Chang
From: Ilpo Järvinen <ij@kernel.org>
The Accurate ECN allows echoing back the sum of bytes for
each IP ECN field value in the received packets using
AccECN option. This change implements AccECN option tx & rx
side processing without option send control related features
that are added by a later change.
Based on specification:
https://tools.ietf.org/id/draft-ietf-tcpm-accurate-ecn-28.txt
(Some features of the spec will be added in the later changes
rather than in this one).
A full-length AccECN option is always attempted but if it does
not fit, the minimum length is selected based on the counters
that have changed since the last update. The AccECN option
(with 24-bit fields) often ends in odd sizes so the option
write code tries to take advantage of some nop used to pad
the other TCP options.
The delivered_ecn_bytes pairs with received_ecn_bytes similar
to how delivered_ce pairs with received_ce. In contrast to
ACE field, however, the option is not always available to update
delivered_ecn_bytes. For ACK w/o AccECN option, the delivered
bytes calculated based on the cumulative ACK+SACK information
are assigned to one of the counters using an estimation
heuristic to select the most likely ECN byte counter. Any
estimation error is corrected when the next AccECN option
arrives. It may occur that the heuristic gets too confused
when there are enough different byte counter deltas between
ACKs with the AccECN option in which case the heuristic just
gives up on updating the counters for a while.
tcp_ecn_option sysctl can be used to select option sending
mode for AccECN: TCP_ECN_OPTION_DISABLED, TCP_ECN_OPTION_MINIMUM,
and TCP_ECN_OPTION_FULL.
This patch increases the size of tcp_info struct, as there is
no existing holes for new u32 variables. Below are the pahole
outcomes before and after this patch:
[BEFORE THIS PATCH]
struct tcp_info {
[...]
__u32 tcpi_total_rto_time; /* 244 4 */
/* size: 248, cachelines: 4, members: 61 */
}
[AFTER THIS PATCH]
struct tcp_info {
[...]
__u32 tcpi_total_rto_time; /* 244 4 */
__u32 tcpi_received_ce; /* 248 4 */
__u32 tcpi_delivered_e1_bytes; /* 252 4 */
__u32 tcpi_delivered_e0_bytes; /* 256 4 */
__u32 tcpi_delivered_ce_bytes; /* 260 4 */
__u32 tcpi_received_e1_bytes; /* 264 4 */
__u32 tcpi_received_e0_bytes; /* 268 4 */
__u32 tcpi_received_ce_bytes; /* 272 4 */
/* size: 280, cachelines: 5, members: 68 */
}
This patch uses the existing 1-byte holes in the tcp_sock_write_txrx
group for new u8 members, but adds a 4-byte hole in tcp_sock_write_rx
group after the new u32 delivered_ecn_bytes[3] member. Therefore, the
group size of tcp_sock_write_rx is increased from 96 to 112. Below
are the pahole outcomes before and after this patch:
[BEFORE THIS PATCH]
struct tcp_sock {
[...]
u8 received_ce_pending:4; /* 2586: 0 1 */
u8 unused2:4; /* 2586: 4 1 */
/* XXX 1 byte hole, try to pack */
[...]
u32 rcv_rtt_last_tsecr; /* 2732 4 */
[...]
__cacheline_group_end__tcp_sock_write_rx[0]; /* 2792 0 */
[...]
/* size: 3264, cachelines: 51, members: 170 */
}
[AFTER THIS PATCH]
struct tcp_sock {
[...]
u8 received_ce_pending:4;/* 2586: 0 1 */
u8 unused2:4; /* 2586: 4 1 */
u8 accecn_minlen:2; /* 2587: 0 1 */
u8 est_ecnfield:2; /* 2587: 2 1 */
u8 unused3:4; /* 2587: 4 1 */
[...]
u32 rcv_rtt_last_tsecr; /* 2732 4 */
u32 delivered_ecn_bytes[3];/* 2736 12 */
/* XXX 4 bytes hole, try to pack */
[...]
__cacheline_group_end__tcp_sock_write_rx[0]; /* 2808 0 */
[...]
/* size: 3264, cachelines: 51, members: 174 */
}
Signed-off-by: Ilpo Järvinen <ij@kernel.org>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
Co-developed-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
Signed-off-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
---
v10
- Add documentation of tcp_ecn_option in ip-sysctl.rst to this patch
- Remove the global variable u32 synack_ecn_bytes[3]
- Add READ_ONCE() over every reads of sysctl
v9:
- Restruct the code in the for loop of tcp_accecn_process_option()
- Remove ecn_bytes and add use_synack_ecn_bytes flag in tcp_out_options
struct to identify whether syn_ack_bytes or received_ecn_bytes is used
- Replace leftover_bytes and leftover_size with leftover_highbyte and
leftover_lowbyte and add comments in tcp_options_write()
v8:
- Reset leftover_size to 2 once leftover_bytes is used
---
Documentation/networking/ip-sysctl.rst | 19 ++
.../networking/net_cachelines/tcp_sock.rst | 3 +
include/linux/tcp.h | 9 +-
include/net/netns/ipv4.h | 1 +
include/net/tcp.h | 10 ++
include/net/tcp_ecn.h | 84 ++++++++-
include/uapi/linux/tcp.h | 7 +
net/ipv4/sysctl_net_ipv4.c | 9 +
net/ipv4/tcp.c | 15 +-
net/ipv4/tcp_input.c | 93 +++++++++-
net/ipv4/tcp_ipv4.c | 1 +
net/ipv4/tcp_output.c | 166 +++++++++++++++++-
12 files changed, 404 insertions(+), 13 deletions(-)
diff --git a/Documentation/networking/ip-sysctl.rst b/Documentation/networking/ip-sysctl.rst
index ebd391a1a3af..9b63466bcbf5 100644
--- a/Documentation/networking/ip-sysctl.rst
+++ b/Documentation/networking/ip-sysctl.rst
@@ -468,6 +468,25 @@ tcp_ecn - INTEGER
Default: 2
+tcp_ecn_option - INTEGER
+ Control Accurate ECN (AccECN) option sending when AccECN has been
+ successfully negotiated during handshake. Send logic inhibits
+ sending AccECN options regarless of this setting when no AccECN
+ option has been seen for the reverse direction.
+
+ Possible values are:
+
+ = ============================================================
+ 0 Never send AccECN option. This also disables sending AccECN
+ option in SYN/ACK during handshake.
+ 1 Send AccECN option sparingly according to the minimum option
+ rules outlined in draft-ietf-tcpm-accurate-ecn.
+ 2 Send AccECN option on every packet whenever it fits into TCP
+ option space.
+ = ============================================================
+
+ Default: 2
+
tcp_ecn_fallback - BOOLEAN
If the kernel detects that ECN connection misbehaves, enable fall
back to non-ECN. Currently, this knob implements the fallback
diff --git a/Documentation/networking/net_cachelines/tcp_sock.rst b/Documentation/networking/net_cachelines/tcp_sock.rst
index 804480d39132..ed6af7d0110c 100644
--- a/Documentation/networking/net_cachelines/tcp_sock.rst
+++ b/Documentation/networking/net_cachelines/tcp_sock.rst
@@ -104,9 +104,12 @@ u32 delivered_ce read_mostly read_w
u32 received_ce read_mostly read_write
u32[3] received_ecn_bytes read_mostly read_write
u8:4 received_ce_pending read_mostly read_write
+u32[3] delivered_ecn_bytes read_write
u8:2 syn_ect_snt write_mostly read_write
u8:2 syn_ect_rcv read_mostly read_write
u8:1 wait_third_ack read_write
+u8:2 accecn_minlen write_mostly read_write
+u8:2 est_ecnfield read_write
u8:4 accecn_fail_mode
u32 lost read_mostly tcp_ack
u32 app_limited read_write read_mostly tcp_rate_check_app_limited,tcp_rate_skb_sent(tx);tcp_rate_gen(rx)
diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 0c2331e186e8..0008b8190ea0 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -122,8 +122,9 @@ struct tcp_options_received {
smc_ok : 1, /* SMC seen on SYN packet */
snd_wscale : 4, /* Window scaling received from sender */
rcv_wscale : 4; /* Window scaling to send to receiver */
- u8 saw_unknown:1, /* Received unknown option */
- unused:7;
+ u8 accecn:6, /* AccECN index in header, 0=no options */
+ saw_unknown:1, /* Received unknown option */
+ unused:1;
u8 num_sacks; /* Number of SACK blocks */
u16 user_mss; /* mss requested by user in ioctl */
u16 mss_clamp; /* Maximal mss, negotiated at connection setup */
@@ -293,6 +294,9 @@ struct tcp_sock {
rate_app_limited:1; /* rate_{delivered,interval_us} limited? */
u8 received_ce_pending:4, /* Not yet transmit cnt of received_ce */
unused2:4;
+ u8 accecn_minlen:2,/* Minimum length of AccECN option sent */
+ est_ecnfield:2,/* ECN field for AccECN delivered estimates */
+ unused3:4;
__be32 pred_flags;
u64 tcp_clock_cache; /* cache last tcp_clock_ns() (see tcp_mstamp_refresh()) */
u64 tcp_mstamp; /* most recent packet received/sent */
@@ -337,6 +341,7 @@ struct tcp_sock {
u32 rate_delivered; /* saved rate sample: packets delivered */
u32 rate_interval_us; /* saved rate sample: time elapsed */
u32 rcv_rtt_last_tsecr;
+ u32 delivered_ecn_bytes[3];
u64 first_tx_mstamp; /* start of window send phase */
u64 delivered_mstamp; /* time we reached "delivered" */
u64 bytes_acked; /* RFC4898 tcpEStatsAppHCThruOctetsAcked
diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index 6373e3f17da8..4569a9ef4fb8 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -148,6 +148,7 @@ struct netns_ipv4 {
struct local_ports ip_local_ports;
u8 sysctl_tcp_ecn;
+ u8 sysctl_tcp_ecn_option;
u8 sysctl_tcp_ecn_fallback;
u8 sysctl_ip_default_ttl;
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 0d8e1a676dad..4e8d6a0a4171 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -203,6 +203,8 @@ static_assert((1 << ATO_BITS) > TCP_DELACK_MAX);
#define TCPOPT_AO 29 /* Authentication Option (RFC5925) */
#define TCPOPT_MPTCP 30 /* Multipath TCP (RFC6824) */
#define TCPOPT_FASTOPEN 34 /* Fast open (RFC7413) */
+#define TCPOPT_ACCECN0 172 /* 0xAC: Accurate ECN Order 0 */
+#define TCPOPT_ACCECN1 174 /* 0xAE: Accurate ECN Order 1 */
#define TCPOPT_EXP 254 /* Experimental */
/* Magic number to be after the option value for sharing TCP
* experimental options. See draft-ietf-tcpm-experimental-options-00.txt
@@ -220,6 +222,7 @@ static_assert((1 << ATO_BITS) > TCP_DELACK_MAX);
#define TCPOLEN_TIMESTAMP 10
#define TCPOLEN_MD5SIG 18
#define TCPOLEN_FASTOPEN_BASE 2
+#define TCPOLEN_ACCECN_BASE 2
#define TCPOLEN_EXP_FASTOPEN_BASE 4
#define TCPOLEN_EXP_SMC_BASE 6
@@ -233,6 +236,13 @@ static_assert((1 << ATO_BITS) > TCP_DELACK_MAX);
#define TCPOLEN_MD5SIG_ALIGNED 20
#define TCPOLEN_MSS_ALIGNED 4
#define TCPOLEN_EXP_SMC_BASE_ALIGNED 8
+#define TCPOLEN_ACCECN_PERFIELD 3
+
+/* Maximum number of byte counters in AccECN option + size */
+#define TCP_ACCECN_NUMFIELDS 3
+#define TCP_ACCECN_MAXSIZE (TCPOLEN_ACCECN_BASE + \
+ TCPOLEN_ACCECN_PERFIELD * \
+ TCP_ACCECN_NUMFIELDS)
/* Flags in tp->nonagle */
#define TCP_NAGLE_OFF 1 /* Nagle's algo is disabled */
diff --git a/include/net/tcp_ecn.h b/include/net/tcp_ecn.h
index 68e392ba2f9e..945daecff082 100644
--- a/include/net/tcp_ecn.h
+++ b/include/net/tcp_ecn.h
@@ -24,6 +24,13 @@ enum tcp_ecn_mode {
TCP_ECN_IN_ACCECN_OUT_NOECN = 5,
};
+/* AccECN option sending when AccECN has been successfully negotiated */
+enum tcp_accecn_option {
+ TCP_ACCECN_OPTION_DISABLED = 0,
+ TCP_ACCECN_OPTION_MINIMUM = 1,
+ TCP_ACCECN_OPTION_FULL = 2,
+};
+
static inline void tcp_ecn_queue_cwr(struct tcp_sock *tp)
{
/* Do not set CWR if in AccECN mode! */
@@ -174,6 +181,74 @@ static inline void tcp_accecn_third_ack(struct sock *sk,
}
}
+/* Maps IP ECN field ECT/CE code point to AccECN option field number, given
+ * we are sending fields with Accurate ECN Order 1: ECT(1), CE, ECT(0).
+ */
+static inline u8 tcp_ecnfield_to_accecn_optfield(u8 ecnfield)
+{
+ switch (ecnfield & INET_ECN_MASK) {
+ case INET_ECN_NOT_ECT:
+ return 0; /* AccECN does not send counts of NOT_ECT */
+ case INET_ECN_ECT_1:
+ return 1;
+ case INET_ECN_CE:
+ return 2;
+ case INET_ECN_ECT_0:
+ return 3;
+ }
+ return 0;
+}
+
+/* Maps IP ECN field ECT/CE code point to AccECN option field value offset.
+ * Some fields do not start from zero, to detect zeroing by middleboxes.
+ */
+static inline u32 tcp_accecn_field_init_offset(u8 ecnfield)
+{
+ switch (ecnfield & INET_ECN_MASK) {
+ case INET_ECN_NOT_ECT:
+ return 0; /* AccECN does not send counts of NOT_ECT */
+ case INET_ECN_ECT_1:
+ return TCP_ACCECN_E1B_INIT_OFFSET;
+ case INET_ECN_CE:
+ return TCP_ACCECN_CEB_INIT_OFFSET;
+ case INET_ECN_ECT_0:
+ return TCP_ACCECN_E0B_INIT_OFFSET;
+ }
+ return 0;
+}
+
+/* Maps AccECN option field #nr to IP ECN field ECT/CE bits */
+static inline unsigned int tcp_accecn_optfield_to_ecnfield(unsigned int option,
+ bool order)
+{
+ u8 tmp;
+
+ option = order ? 2 - option : option;
+ tmp = option + 2;
+
+ return (tmp + (tmp >> 2)) & INET_ECN_MASK;
+}
+
+/* Handles AccECN option ECT and CE 24-bit byte counters update into
+ * the u32 value in tcp_sock. As we're processing TCP options, it is
+ * safe to access from - 1.
+ */
+static inline s32 tcp_update_ecn_bytes(u32 *cnt, const char *from,
+ u32 init_offset)
+{
+ u32 truncated = (get_unaligned_be32(from - 1) - init_offset) &
+ 0xFFFFFFU;
+ u32 delta = (truncated - *cnt) & 0xFFFFFFU;
+
+ /* If delta has the highest bit set (24th bit) indicating
+ * negative, sign extend to correct an estimation using
+ * sign_extend32(delta, 24 - 1)
+ */
+ delta = sign_extend32(delta, 23);
+ *cnt += delta;
+ return (s32)delta;
+}
+
/* Updates Accurate ECN received counters from the received IP ECN field */
static inline void tcp_ecn_received_counters(struct sock *sk,
const struct sk_buff *skb, u32 len)
@@ -197,8 +272,12 @@ static inline void tcp_ecn_received_counters(struct sock *sk,
tp->received_ce_pending = min(tp->received_ce_pending + pcount,
0xfU);
- if (len > 0)
+ if (len > 0) {
+ u8 minlen = tcp_ecnfield_to_accecn_optfield(ecnfield);
tp->received_ecn_bytes[ecnfield - 1] += len;
+ tp->accecn_minlen = max_t(u8, tp->accecn_minlen,
+ minlen);
+ }
}
}
@@ -270,6 +349,9 @@ static inline void tcp_accecn_init_counters(struct tcp_sock *tp)
tp->received_ce = 0;
tp->received_ce_pending = 0;
__tcp_accecn_init_bytes_counters(tp->received_ecn_bytes);
+ __tcp_accecn_init_bytes_counters(tp->delivered_ecn_bytes);
+ tp->accecn_minlen = 0;
+ tp->est_ecnfield = 0;
}
/* Used for make_synack to form the ACE flags */
diff --git a/include/uapi/linux/tcp.h b/include/uapi/linux/tcp.h
index bdac8c42fa82..53e0e85b52be 100644
--- a/include/uapi/linux/tcp.h
+++ b/include/uapi/linux/tcp.h
@@ -316,6 +316,13 @@ struct tcp_info {
* in milliseconds, including any
* unfinished recovery.
*/
+ __u32 tcpi_received_ce; /* # of CE marks received */
+ __u32 tcpi_delivered_e1_bytes; /* Accurate ECN byte counters */
+ __u32 tcpi_delivered_e0_bytes;
+ __u32 tcpi_delivered_ce_bytes;
+ __u32 tcpi_received_e1_bytes;
+ __u32 tcpi_received_e0_bytes;
+ __u32 tcpi_received_ce_bytes;
};
/* netlink attributes types for SCM_TIMESTAMPING_OPT_STATS */
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index 75ec1a599b52..1d7fd86ca7b9 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -731,6 +731,15 @@ static struct ctl_table ipv4_net_table[] = {
.extra1 = SYSCTL_ZERO,
.extra2 = &tcp_ecn_mode_max,
},
+ {
+ .procname = "tcp_ecn_option",
+ .data = &init_net.ipv4.sysctl_tcp_ecn_option,
+ .maxlen = sizeof(u8),
+ .mode = 0644,
+ .proc_handler = proc_dou8vec_minmax,
+ .extra1 = SYSCTL_ZERO,
+ .extra2 = SYSCTL_TWO,
+ },
{
.procname = "tcp_ecn_fallback",
.data = &init_net.ipv4.sysctl_tcp_ecn_fallback,
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 237435927b9a..a2b83acf5b94 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -270,6 +270,7 @@
#include <net/icmp.h>
#include <net/inet_common.h>
+#include <net/inet_ecn.h>
#include <net/tcp.h>
#include <net/tcp_ecn.h>
#include <net/mptcp.h>
@@ -4137,6 +4138,9 @@ void tcp_get_info(struct sock *sk, struct tcp_info *info)
{
const struct tcp_sock *tp = tcp_sk(sk); /* iff sk_type == SOCK_STREAM */
const struct inet_connection_sock *icsk = inet_csk(sk);
+ const u8 ect1_idx = INET_ECN_ECT_1 - 1;
+ const u8 ect0_idx = INET_ECN_ECT_0 - 1;
+ const u8 ce_idx = INET_ECN_CE - 1;
unsigned long rate;
u32 now;
u64 rate64;
@@ -4263,6 +4267,14 @@ void tcp_get_info(struct sock *sk, struct tcp_info *info)
if (tp->rto_stamp)
info->tcpi_total_rto_time += tcp_clock_ms() - tp->rto_stamp;
+ info->tcpi_received_ce = tp->received_ce;
+ info->tcpi_delivered_e1_bytes = tp->delivered_ecn_bytes[ect1_idx];
+ info->tcpi_delivered_e0_bytes = tp->delivered_ecn_bytes[ect0_idx];
+ info->tcpi_delivered_ce_bytes = tp->delivered_ecn_bytes[ce_idx];
+ info->tcpi_received_e1_bytes = tp->received_ecn_bytes[ect1_idx];
+ info->tcpi_received_e0_bytes = tp->received_ecn_bytes[ect0_idx];
+ info->tcpi_received_ce_bytes = tp->received_ecn_bytes[ce_idx];
+
unlock_sock_fast(sk, slow);
}
EXPORT_SYMBOL_GPL(tcp_get_info);
@@ -5141,12 +5153,13 @@ static void __init tcp_struct_check(void)
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_rx, rate_delivered);
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_rx, rate_interval_us);
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_rx, rcv_rtt_last_tsecr);
+ CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_rx, delivered_ecn_bytes);
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_rx, first_tx_mstamp);
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_rx, delivered_mstamp);
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_rx, bytes_acked);
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_rx, rcv_rtt_est);
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_rx, rcvq_space);
- CACHELINE_ASSERT_GROUP_SIZE(struct tcp_sock, tcp_sock_write_rx, 96);
+ CACHELINE_ASSERT_GROUP_SIZE(struct tcp_sock, tcp_sock_write_rx, 112);
}
void __init tcp_init(void)
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index f7d7649612a2..107c20ae0f5b 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -70,6 +70,7 @@
#include <linux/sysctl.h>
#include <linux/kernel.h>
#include <linux/prefetch.h>
+#include <linux/bitops.h>
#include <net/dst.h>
#include <net/tcp.h>
#include <net/tcp_ecn.h>
@@ -384,6 +385,72 @@ static void tcp_data_ecn_check(struct sock *sk, const struct sk_buff *skb)
}
}
+/* Returns true if the byte counters can be used */
+static bool tcp_accecn_process_option(struct tcp_sock *tp,
+ const struct sk_buff *skb,
+ u32 delivered_bytes, int flag)
+{
+ u8 estimate_ecnfield = tp->est_ecnfield;
+ bool ambiguous_ecn_bytes_incr = false;
+ bool first_changed = false;
+ unsigned int optlen;
+ bool order1, res;
+ unsigned int i;
+ u8 *ptr;
+
+ if (!(flag & FLAG_SLOWPATH) || !tp->rx_opt.accecn) {
+ if (estimate_ecnfield) {
+ u8 ecnfield = estimate_ecnfield - 1;
+
+ tp->delivered_ecn_bytes[ecnfield] += delivered_bytes;
+ return true;
+ }
+ return false;
+ }
+
+ ptr = skb_transport_header(skb) + tp->rx_opt.accecn;
+ optlen = ptr[1] - 2;
+ WARN_ON_ONCE(ptr[0] != TCPOPT_ACCECN0 && ptr[0] != TCPOPT_ACCECN1);
+ order1 = (ptr[0] == TCPOPT_ACCECN1);
+ ptr += 2;
+
+ res = !!estimate_ecnfield;
+ for (i = 0; i < 3; i++) {
+ u32 init_offset;
+ u8 ecnfield;
+ s32 delta;
+ u32 *cnt;
+
+ if (optlen < TCPOLEN_ACCECN_PERFIELD)
+ break;
+
+ ecnfield = tcp_accecn_optfield_to_ecnfield(i, order1);
+ init_offset = tcp_accecn_field_init_offset(ecnfield);
+ cnt = &tp->delivered_ecn_bytes[ecnfield - 1];
+ delta = tcp_update_ecn_bytes(cnt, ptr, init_offset);
+ if (delta && delta < 0) {
+ res = false;
+ ambiguous_ecn_bytes_incr = true;
+ }
+ if (delta && ecnfield != estimate_ecnfield) {
+ if (!first_changed) {
+ tp->est_ecnfield = ecnfield;
+ first_changed = true;
+ } else {
+ res = false;
+ ambiguous_ecn_bytes_incr = true;
+ }
+ }
+
+ optlen -= TCPOLEN_ACCECN_PERFIELD;
+ ptr += TCPOLEN_ACCECN_PERFIELD;
+ }
+ if (ambiguous_ecn_bytes_incr)
+ tp->est_ecnfield = 0;
+
+ return res;
+}
+
static void tcp_count_delivered_ce(struct tcp_sock *tp, u32 ecn_count)
{
tp->delivered_ce += ecn_count;
@@ -400,7 +467,8 @@ static void tcp_count_delivered(struct tcp_sock *tp, u32 delivered,
/* Returns the ECN CE delta */
static u32 __tcp_accecn_process(struct sock *sk, const struct sk_buff *skb,
- u32 delivered_pkts, int flag)
+ u32 delivered_pkts, u32 delivered_bytes,
+ int flag)
{
const struct tcphdr *th = tcp_hdr(skb);
struct tcp_sock *tp = tcp_sk(sk);
@@ -411,6 +479,8 @@ static u32 __tcp_accecn_process(struct sock *sk, const struct sk_buff *skb,
if (!(flag & (FLAG_FORWARD_PROGRESS | FLAG_TS_PROGRESS)))
return 0;
+ tcp_accecn_process_option(tp, skb, delivered_bytes, flag);
+
if (!(flag & FLAG_SLOWPATH)) {
/* AccECN counter might overflow on large ACKs */
if (delivered_pkts <= TCP_ACCECN_CEP_ACE_MASK)
@@ -436,12 +506,14 @@ static u32 __tcp_accecn_process(struct sock *sk, const struct sk_buff *skb,
}
static u32 tcp_accecn_process(struct sock *sk, const struct sk_buff *skb,
- u32 delivered_pkts, int *flag)
+ u32 delivered_pkts, u32 delivered_bytes,
+ int *flag)
{
struct tcp_sock *tp = tcp_sk(sk);
u32 delta;
- delta = __tcp_accecn_process(sk, skb, delivered_pkts, *flag);
+ delta = __tcp_accecn_process(sk, skb, delivered_pkts,
+ delivered_bytes, *flag);
if (delta > 0) {
tcp_count_delivered_ce(tp, delta);
*flag |= FLAG_ECE;
@@ -3973,6 +4045,7 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
if (tcp_ecn_mode_accecn(tp))
ecn_count = tcp_accecn_process(sk, skb,
tp->delivered - delivered,
+ sack_state.delivered_bytes,
&flag);
tcp_in_ack_event(sk, flag);
@@ -4012,6 +4085,7 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag)
if (tcp_ecn_mode_accecn(tp))
ecn_count = tcp_accecn_process(sk, skb,
tp->delivered - delivered,
+ sack_state.delivered_bytes,
&flag);
tcp_in_ack_event(sk, flag);
/* If data was DSACKed, see if we can undo a cwnd reduction. */
@@ -4139,6 +4213,7 @@ void tcp_parse_options(const struct net *net,
ptr = (const unsigned char *)(th + 1);
opt_rx->saw_tstamp = 0;
+ opt_rx->accecn = 0;
opt_rx->saw_unknown = 0;
while (length > 0) {
@@ -4230,6 +4305,12 @@ void tcp_parse_options(const struct net *net,
ptr, th->syn, foc, false);
break;
+ case TCPOPT_ACCECN0:
+ case TCPOPT_ACCECN1:
+ /* Save offset of AccECN option in TCP header */
+ opt_rx->accecn = (ptr - 2) - (__u8 *)th;
+ break;
+
case TCPOPT_EXP:
/* Fast Open option shares code 254 using a
* 16 bits magic number.
@@ -4290,11 +4371,14 @@ static bool tcp_fast_parse_options(const struct net *net,
*/
if (th->doff == (sizeof(*th) / 4)) {
tp->rx_opt.saw_tstamp = 0;
+ tp->rx_opt.accecn = 0;
return false;
} else if (tp->rx_opt.tstamp_ok &&
th->doff == ((sizeof(*th) + TCPOLEN_TSTAMP_ALIGNED) / 4)) {
- if (tcp_parse_aligned_timestamp(tp, th))
+ if (tcp_parse_aligned_timestamp(tp, th)) {
+ tp->rx_opt.accecn = 0;
return true;
+ }
}
tcp_parse_options(net, skb, &tp->rx_opt, 1, NULL);
@@ -6094,6 +6178,7 @@ void tcp_rcv_established(struct sock *sk, struct sk_buff *skb)
*/
tp->rx_opt.saw_tstamp = 0;
+ tp->rx_opt.accecn = 0;
/* pred_flags is 0xS?10 << 16 + snd_wnd
* if header_prediction is to be made
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index b95e4ed227cb..1da7c5e4da32 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -3451,6 +3451,7 @@ static void __net_init tcp_set_hashinfo(struct net *net)
static int __net_init tcp_sk_init(struct net *net)
{
net->ipv4.sysctl_tcp_ecn = TCP_ECN_IN_ECN_OUT_NOECN;
+ net->ipv4.sysctl_tcp_ecn_option = TCP_ACCECN_OPTION_FULL;
net->ipv4.sysctl_tcp_ecn_fallback = 1;
net->ipv4.sysctl_tcp_base_mss = TCP_BASE_MSS;
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index d98a1a17eb52..2169fd28594e 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -385,6 +385,7 @@ static inline bool tcp_urg_mode(const struct tcp_sock *tp)
#define OPTION_SMC BIT(9)
#define OPTION_MPTCP BIT(10)
#define OPTION_AO BIT(11)
+#define OPTION_ACCECN BIT(12)
static void smc_options_write(__be32 *ptr, u16 *options)
{
@@ -406,6 +407,8 @@ struct tcp_out_options {
u16 mss; /* 0 to disable */
u8 ws; /* window scale, 0 to disable */
u8 num_sack_blocks; /* number of SACK blocks to include */
+ u8 num_accecn_fields:7, /* number of AccECN fields needed */
+ use_synack_ecn_bytes:1; /* Use synack_ecn_bytes or not */
u8 hash_size; /* bytes in hash_location */
u8 bpf_opt_len; /* length of BPF hdr option */
__u8 *hash_location; /* temporary pointer, overloaded */
@@ -621,6 +624,8 @@ static void tcp_options_write(struct tcphdr *th, struct tcp_sock *tp,
struct tcp_out_options *opts,
struct tcp_key *key)
{
+ u8 leftover_highbyte = TCPOPT_NOP; /* replace 1st NOP if avail */
+ u8 leftover_lowbyte = TCPOPT_NOP; /* replace 2nd NOP in succession */
__be32 *ptr = (__be32 *)(th + 1);
u16 options = opts->options; /* mungable copy */
@@ -656,15 +661,79 @@ static void tcp_options_write(struct tcphdr *th, struct tcp_sock *tp,
*ptr++ = htonl(opts->tsecr);
}
+ if (OPTION_ACCECN & options) {
+ /* Initial values for AccECN option, ordered is based on ECN field bits
+ * similar to received_ecn_bytes. Used for SYN/ACK AccECN option.
+ */
+ static u32 synack_ecn_bytes[3] = { 0, 0, 0 };
+ const u8 ect0_idx = INET_ECN_ECT_0 - 1;
+ const u8 ect1_idx = INET_ECN_ECT_1 - 1;
+ const u8 ce_idx = INET_ECN_CE - 1;
+ u32 e0b;
+ u32 e1b;
+ u32 ceb;
+ u8 len;
+
+ if (opts->use_synack_ecn_bytes) {
+ e0b = synack_ecn_bytes[ect0_idx] + TCP_ACCECN_E0B_INIT_OFFSET;
+ e1b = synack_ecn_bytes[ect1_idx] + TCP_ACCECN_E1B_INIT_OFFSET;
+ ceb = synack_ecn_bytes[ce_idx] + TCP_ACCECN_CEB_INIT_OFFSET;
+ } else {
+ e0b = tp->received_ecn_bytes[ect0_idx] + TCP_ACCECN_E0B_INIT_OFFSET;
+ e1b = tp->received_ecn_bytes[ect1_idx] + TCP_ACCECN_E1B_INIT_OFFSET;
+ ceb = tp->received_ecn_bytes[ce_idx] + TCP_ACCECN_CEB_INIT_OFFSET;
+ }
+
+ len = TCPOLEN_ACCECN_BASE +
+ opts->num_accecn_fields * TCPOLEN_ACCECN_PERFIELD;
+
+ if (opts->num_accecn_fields == 2) {
+ *ptr++ = htonl((TCPOPT_ACCECN1 << 24) | (len << 16) |
+ ((e1b >> 8) & 0xffff));
+ *ptr++ = htonl(((e1b & 0xff) << 24) |
+ (ceb & 0xffffff));
+ } else if (opts->num_accecn_fields == 1) {
+ *ptr++ = htonl((TCPOPT_ACCECN1 << 24) | (len << 16) |
+ ((e1b >> 8) & 0xffff));
+ leftover_highbyte = e1b & 0xff;
+ leftover_lowbyte = TCPOPT_NOP;
+ } else if (opts->num_accecn_fields == 0) {
+ leftover_highbyte = TCPOPT_ACCECN1;
+ leftover_lowbyte = len;
+ } else if (opts->num_accecn_fields == 3) {
+ *ptr++ = htonl((TCPOPT_ACCECN1 << 24) | (len << 16) |
+ ((e1b >> 8) & 0xffff));
+ *ptr++ = htonl(((e1b & 0xff) << 24) |
+ (ceb & 0xffffff));
+ *ptr++ = htonl(((e0b & 0xffffff) << 8) |
+ TCPOPT_NOP);
+ }
+ if (tp)
+ tp->accecn_minlen = 0;
+ }
+
if (unlikely(OPTION_SACK_ADVERTISE & options)) {
- *ptr++ = htonl((TCPOPT_NOP << 24) |
- (TCPOPT_NOP << 16) |
+ *ptr++ = htonl((leftover_highbyte << 24) |
+ (leftover_lowbyte << 16) |
(TCPOPT_SACK_PERM << 8) |
TCPOLEN_SACK_PERM);
+ leftover_highbyte = TCPOPT_NOP;
+ leftover_lowbyte = TCPOPT_NOP;
}
if (unlikely(OPTION_WSCALE & options)) {
- *ptr++ = htonl((TCPOPT_NOP << 24) |
+ u8 highbyte = TCPOPT_NOP;
+
+ /* Do not split the leftover 2-byte to fit into a single
+ * NOP, i.e., replace this NOP only when 1 byte is leftover
+ * within leftover_highbyte.
+ */
+ if (unlikely(leftover_highbyte != TCPOPT_NOP &&
+ leftover_lowbyte == TCPOPT_NOP)) {
+ highbyte = leftover_highbyte;
+ leftover_highbyte = TCPOPT_NOP;
+ }
+ *ptr++ = htonl((highbyte << 24) |
(TCPOPT_WINDOW << 16) |
(TCPOLEN_WINDOW << 8) |
opts->ws);
@@ -675,11 +744,13 @@ static void tcp_options_write(struct tcphdr *th, struct tcp_sock *tp,
tp->duplicate_sack : tp->selective_acks;
int this_sack;
- *ptr++ = htonl((TCPOPT_NOP << 24) |
- (TCPOPT_NOP << 16) |
+ *ptr++ = htonl((leftover_highbyte << 24) |
+ (leftover_lowbyte << 16) |
(TCPOPT_SACK << 8) |
(TCPOLEN_SACK_BASE + (opts->num_sack_blocks *
TCPOLEN_SACK_PERBLOCK)));
+ leftover_highbyte = TCPOPT_NOP;
+ leftover_lowbyte = TCPOPT_NOP;
for (this_sack = 0; this_sack < opts->num_sack_blocks;
++this_sack) {
@@ -688,6 +759,14 @@ static void tcp_options_write(struct tcphdr *th, struct tcp_sock *tp,
}
tp->rx_opt.dsack = 0;
+ } else if (unlikely(leftover_highbyte != TCPOPT_NOP ||
+ leftover_lowbyte != TCPOPT_NOP)) {
+ *ptr++ = htonl((leftover_highbyte << 24) |
+ (leftover_lowbyte << 16) |
+ (TCPOPT_NOP << 8) |
+ TCPOPT_NOP);
+ leftover_highbyte = TCPOPT_NOP;
+ leftover_lowbyte = TCPOPT_NOP;
}
if (unlikely(OPTION_FAST_OPEN_COOKIE & options)) {
@@ -768,6 +847,59 @@ static void mptcp_set_option_cond(const struct request_sock *req,
}
}
+static u32 tcp_synack_options_combine_saving(struct tcp_out_options *opts)
+{
+ /* How much there's room for combining with the alignment padding? */
+ if ((opts->options & (OPTION_SACK_ADVERTISE | OPTION_TS)) ==
+ OPTION_SACK_ADVERTISE)
+ return 2;
+ else if (opts->options & OPTION_WSCALE)
+ return 1;
+ return 0;
+}
+
+/* Calculates how long AccECN option will fit to @remaining option space.
+ *
+ * AccECN option can sometimes replace NOPs used for alignment of other
+ * TCP options (up to @max_combine_saving available).
+ *
+ * Only solutions with at least @required AccECN fields are accepted.
+ *
+ * Returns: The size of the AccECN option excluding space repurposed from
+ * the alignment of the other options.
+ */
+static int tcp_options_fit_accecn(struct tcp_out_options *opts, int required,
+ int remaining)
+{
+ int size = TCP_ACCECN_MAXSIZE;
+ int max_combine_saving;
+
+ if (opts->use_synack_ecn_bytes)
+ max_combine_saving = tcp_synack_options_combine_saving(opts);
+ else
+ max_combine_saving = opts->num_sack_blocks > 0 ? 2 : 0;
+ opts->num_accecn_fields = TCP_ACCECN_NUMFIELDS;
+ while (opts->num_accecn_fields >= required) {
+ int leftover_size = size & 0x3;
+ /* Pad to dword if cannot combine */
+ if (leftover_size > max_combine_saving)
+ leftover_size = -((4 - leftover_size) & 0x3);
+
+ if (remaining >= size - leftover_size) {
+ size -= leftover_size;
+ break;
+ }
+
+ opts->num_accecn_fields--;
+ size -= TCPOLEN_ACCECN_PERFIELD;
+ }
+ if (opts->num_accecn_fields < required)
+ return 0;
+
+ opts->options |= OPTION_ACCECN;
+ return size;
+}
+
/* Compute TCP options for SYN packets. This is not the final
* network wire format yet.
*/
@@ -850,6 +982,15 @@ static unsigned int tcp_syn_options(struct sock *sk, struct sk_buff *skb,
}
}
+ /* Simultaneous open SYN/ACK needs AccECN option but not SYN */
+ if (unlikely((TCP_SKB_CB(skb)->tcp_flags & TCPHDR_ACK) &&
+ tcp_ecn_mode_accecn(tp) &&
+ READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_ecn_option) &&
+ remaining >= TCPOLEN_ACCECN_BASE)) {
+ opts->use_synack_ecn_bytes = 1;
+ remaining -= tcp_options_fit_accecn(opts, 0, remaining);
+ }
+
bpf_skops_hdr_opt_len(sk, skb, NULL, NULL, 0, opts, &remaining);
return MAX_TCP_OPTION_SPACE - remaining;
@@ -867,6 +1008,7 @@ static unsigned int tcp_synack_options(const struct sock *sk,
{
struct inet_request_sock *ireq = inet_rsk(req);
unsigned int remaining = MAX_TCP_OPTION_SPACE;
+ struct tcp_request_sock *treq = tcp_rsk(req);
if (tcp_key_is_md5(key)) {
opts->options |= OPTION_MD5;
@@ -929,6 +1071,13 @@ static unsigned int tcp_synack_options(const struct sock *sk,
smc_set_option_cond(tcp_sk(sk), ireq, opts, &remaining);
+ if (treq->accecn_ok &&
+ READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_ecn_option) &&
+ remaining >= TCPOLEN_ACCECN_BASE) {
+ opts->use_synack_ecn_bytes = 1;
+ remaining -= tcp_options_fit_accecn(opts, 0, remaining);
+ }
+
bpf_skops_hdr_opt_len((struct sock *)sk, skb, req, syn_skb,
synack_type, opts, &remaining);
@@ -1001,6 +1150,13 @@ static unsigned int tcp_established_options(struct sock *sk, struct sk_buff *skb
opts->num_sack_blocks = 0;
}
+ if (tcp_ecn_mode_accecn(tp) &&
+ READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_ecn_option)) {
+ opts->use_synack_ecn_bytes = 0;
+ size += tcp_options_fit_accecn(opts, tp->accecn_minlen,
+ MAX_TCP_OPTION_SPACE - size);
+ }
+
if (unlikely(BPF_SOCK_OPS_TEST_FLAG(tp,
BPF_SOCK_OPS_WRITE_HDR_OPT_CB_FLAG))) {
unsigned int remaining = MAX_TCP_OPTION_SPACE - size;
--
2.34.1
^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH v11 net-next 12/15] tcp: accecn: AccECN option send control
2025-07-03 5:17 [PATCH v11 net-next 00/15] AccECN protocol patch series chia-yu.chang
` (10 preceding siblings ...)
2025-07-03 5:17 ` [PATCH v11 net-next 11/15] tcp: accecn: AccECN option chia-yu.chang
@ 2025-07-03 5:17 ` chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 13/15] tcp: accecn: AccECN option failure handling chia-yu.chang
` (2 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: chia-yu.chang @ 2025-07-03 5:17 UTC (permalink / raw)
To: pabeni, edumazet, linux-doc, corbet, horms, dsahern, kuniyu, bpf,
netdev, dave.taht, jhs, kuba, stephen, xiyou.wangcong, jiri,
davem, andrew+netdev, donald.hunter, ast, liuhangbin, shuah,
linux-kselftest, ij, ncardwell, koen.de_schepper, g.white,
ingemar.s.johansson, mirja.kuehlewind, cheshire, rs.ietf,
Jason_Livingood, vidhi_goel
Cc: Chia-Yu Chang
From: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
Instead of sending the option in every ACK, limit sending to
those ACKs where the option is necessary:
- Handshake
- "Change-triggered ACK" + the ACK following it. The
2nd ACK is necessary to unambiguously indicate which
of the ECN byte counters in increasing. The first
ACK has two counters increasing due to the ecnfield
edge.
- ACKs with CE to allow CEP delta validations to take
advantage of the option.
- Force option to be sent every at least once per 2^22
bytes. The check is done using the bit edges of the
byte counters (avoids need for extra variables).
- AccECN option beacon to send a few times per RTT even if
nothing in the ECN state requires that. The default is 3
times per RTT, and its period can be set via
sysctl_tcp_ecn_option_beacon.
Below are the pahole outcomes before and after this patch,
in which the group size of tcp_sock_write_tx is increased
from 89 to 97 due to the new u64 accecn_opt_tstamp member:
[BEFORE THIS PATCH]
struct tcp_sock {
[...]
u64 tcp_wstamp_ns; /* 2552 8 */
struct list_head tsorted_sent_queue; /* 2560 16 */
[...]
__cacheline_group_end__tcp_sock_write_tx[0]; /* 2585 0 */
__cacheline_group_begin__tcp_sock_write_txrx[0]; /* 2585 0 */
u8 nonagle:4; /* 2585: 0 1 */
u8 rate_app_limited:1; /* 2585: 4 1 */
/* XXX 3 bits hole, try to pack */
/* Force alignment to the next boundary: */
u8 :0;
u8 received_ce_pending:4;/* 2586: 0 1 */
u8 unused2:4; /* 2586: 4 1 */
u8 accecn_minlen:2; /* 2587: 0 1 */
u8 est_ecnfield:2; /* 2587: 2 1 */
u8 unused3:4; /* 2587: 4 1 */
[...]
__cacheline_group_end__tcp_sock_write_txrx[0]; /* 2692 0 */
[...]
/* size: 3264, cachelines: 51, members: 174 */
}
[AFTER THIS PATCH]
struct tcp_sock {
[...]
u64 tcp_wstamp_ns; /* 2552 8 */
u64 accecn_opt_tstamp; /* 2560 8 */
struct list_head tsorted_sent_queue; /* 2568 16 */
[...]
__cacheline_group_end__tcp_sock_write_tx[0]; /* 2593 0 */
__cacheline_group_begin__tcp_sock_write_txrx[0]; /* 2593 0 */
u8 nonagle:4; /* 2593: 0 1 */
u8 rate_app_limited:1; /* 2593: 4 1 */
/* XXX 3 bits hole, try to pack */
/* Force alignment to the next boundary: */
u8 :0;
u8 received_ce_pending:4;/* 2594: 0 1 */
u8 unused2:4; /* 2594: 4 1 */
u8 accecn_minlen:2; /* 2595: 0 1 */
u8 est_ecnfield:2; /* 2595: 2 1 */
u8 accecn_opt_demand:2; /* 2595: 4 1 */
u8 prev_ecnfield:2; /* 2595: 6 1 */
[...]
__cacheline_group_end__tcp_sock_write_txrx[0]; /* 2700 0 */
[...]
/* size: 3264, cachelines: 51, members: 176 */
}
Signed-off-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
Co-developed-by: Ilpo Järvinen <ij@kernel.org>
Signed-off-by: Ilpo Järvinen <ij@kernel.org>
---
v10
- Add documentation of tcp_ecn_option_beacon in ip-sysctl.rst to this patch
v8:
- Add new helper function tcp_accecn_opt_demand_min()
- Remove white space only change
---
Documentation/networking/ip-sysctl.rst | 6 +++
.../networking/net_cachelines/tcp_sock.rst | 3 ++
include/linux/tcp.h | 4 +-
include/net/netns/ipv4.h | 1 +
include/net/tcp.h | 3 ++
include/net/tcp_ecn.h | 50 +++++++++++++++++++
net/ipv4/sysctl_net_ipv4.c | 9 ++++
net/ipv4/tcp.c | 5 +-
net/ipv4/tcp_input.c | 4 +-
net/ipv4/tcp_ipv4.c | 1 +
net/ipv4/tcp_minisocks.c | 2 +
net/ipv4/tcp_output.c | 16 +++++-
12 files changed, 99 insertions(+), 5 deletions(-)
diff --git a/Documentation/networking/ip-sysctl.rst b/Documentation/networking/ip-sysctl.rst
index 9b63466bcbf5..5698dab3a7b0 100644
--- a/Documentation/networking/ip-sysctl.rst
+++ b/Documentation/networking/ip-sysctl.rst
@@ -487,6 +487,12 @@ tcp_ecn_option - INTEGER
Default: 2
+tcp_ecn_option_beacon - INTEGER
+ Control Accurate ECN (AccECN) option sending frequency per RTT and it
+ takes effect only when tcp_ecn_option is set to 2.
+
+ Default: 3 (AccECN will be send at least 3 times per RTT)
+
tcp_ecn_fallback - BOOLEAN
If the kernel detects that ECN connection misbehaves, enable fall
back to non-ECN. Currently, this knob implements the fallback
diff --git a/Documentation/networking/net_cachelines/tcp_sock.rst b/Documentation/networking/net_cachelines/tcp_sock.rst
index ed6af7d0110c..88c668f9c68b 100644
--- a/Documentation/networking/net_cachelines/tcp_sock.rst
+++ b/Documentation/networking/net_cachelines/tcp_sock.rst
@@ -110,6 +110,9 @@ u8:2 syn_ect_rcv read_mostly read_w
u8:1 wait_third_ack read_write
u8:2 accecn_minlen write_mostly read_write
u8:2 est_ecnfield read_write
+u8:2 accecn_opt_demand read_mostly read_write
+u8:2 prev_ecnfield read_write
+u64 accecn_opt_tstamp read_write
u8:4 accecn_fail_mode
u32 lost read_mostly tcp_ack
u32 app_limited read_write read_mostly tcp_rate_check_app_limited,tcp_rate_skb_sent(tx);tcp_rate_gen(rx)
diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 0008b8190ea0..2c0a2db6e6dd 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -275,6 +275,7 @@ struct tcp_sock {
u32 mdev_us; /* medium deviation */
u32 rtt_seq; /* sequence number to update rttvar */
u64 tcp_wstamp_ns; /* departure time for next sent data packet */
+ u64 accecn_opt_tstamp; /* Last AccECN option sent timestamp */
struct list_head tsorted_sent_queue; /* time-sorted sent but un-SACKed skbs */
struct sk_buff *highest_sack; /* skb just after the highest
* skb with SACKed bit set
@@ -296,7 +297,8 @@ struct tcp_sock {
unused2:4;
u8 accecn_minlen:2,/* Minimum length of AccECN option sent */
est_ecnfield:2,/* ECN field for AccECN delivered estimates */
- unused3:4;
+ accecn_opt_demand:2,/* Demand AccECN option for n next ACKs */
+ prev_ecnfield:2; /* ECN bits from the previous segment */
__be32 pred_flags;
u64 tcp_clock_cache; /* cache last tcp_clock_ns() (see tcp_mstamp_refresh()) */
u64 tcp_mstamp; /* most recent packet received/sent */
diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index 4569a9ef4fb8..ff8b5b56ad00 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -149,6 +149,7 @@ struct netns_ipv4 {
u8 sysctl_tcp_ecn;
u8 sysctl_tcp_ecn_option;
+ u8 sysctl_tcp_ecn_option_beacon;
u8 sysctl_tcp_ecn_fallback;
u8 sysctl_ip_default_ttl;
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 4e8d6a0a4171..4516b4910fae 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -90,6 +90,9 @@ void tcp_time_wait(struct sock *sk, int state, int timeo);
/* Maximal number of window scale according to RFC1323 */
#define TCP_MAX_WSCALE 14U
+/* Default sending frequency of accurate ECN option per RTT */
+#define TCP_ACCECN_OPTION_BEACON 3
+
/* urg_data states */
#define TCP_URG_VALID 0x0100
#define TCP_URG_NOTYET 0x0200
diff --git a/include/net/tcp_ecn.h b/include/net/tcp_ecn.h
index 945daecff082..667f4df961b3 100644
--- a/include/net/tcp_ecn.h
+++ b/include/net/tcp_ecn.h
@@ -181,6 +181,16 @@ static inline void tcp_accecn_third_ack(struct sock *sk,
}
}
+static inline void tcp_accecn_opt_demand_min(struct sock *sk,
+ u8 opt_demand_min)
+{
+ struct tcp_sock *tp = tcp_sk(sk);
+ u8 opt_demand;
+
+ opt_demand = max_t(u8, opt_demand_min, tp->accecn_opt_demand);
+ tp->accecn_opt_demand = opt_demand;
+}
+
/* Maps IP ECN field ECT/CE code point to AccECN option field number, given
* we are sending fields with Accurate ECN Order 1: ECT(1), CE, ECT(0).
*/
@@ -256,6 +266,7 @@ static inline void tcp_ecn_received_counters(struct sock *sk,
u8 ecnfield = TCP_SKB_CB(skb)->ip_dsfield & INET_ECN_MASK;
u8 is_ce = INET_ECN_is_ce(ecnfield);
struct tcp_sock *tp = tcp_sk(sk);
+ bool ecn_edge;
if (!INET_ECN_is_not_ect(ecnfield)) {
u32 pcount = is_ce * max_t(u16, 1, skb_shinfo(skb)->gso_segs);
@@ -274,9 +285,33 @@ static inline void tcp_ecn_received_counters(struct sock *sk,
if (len > 0) {
u8 minlen = tcp_ecnfield_to_accecn_optfield(ecnfield);
+ u32 oldbytes = tp->received_ecn_bytes[ecnfield - 1];
+
tp->received_ecn_bytes[ecnfield - 1] += len;
tp->accecn_minlen = max_t(u8, tp->accecn_minlen,
minlen);
+
+ /* Demand AccECN option at least every 2^22 bytes to
+ * avoid overflowing the ECN byte counters.
+ */
+ if ((tp->received_ecn_bytes[ecnfield - 1] ^ oldbytes) &
+ ~((1 << 22) - 1)) {
+ tcp_accecn_opt_demand_min(sk, 1);
+ }
+ }
+ }
+
+ ecn_edge = tp->prev_ecnfield != ecnfield;
+ if (ecn_edge || is_ce) {
+ tp->prev_ecnfield = ecnfield;
+ /* Demand Accurate ECN change-triggered ACKs. Two ACK are
+ * demanded to indicate unambiguously the ecnfield value
+ * in the latter ACK.
+ */
+ if (tcp_ecn_mode_accecn(tp)) {
+ if (ecn_edge)
+ inet_csk(sk)->icsk_ack.pending |= ICSK_ACK_NOW;
+ tp->accecn_opt_demand = 2;
}
}
}
@@ -351,6 +386,7 @@ static inline void tcp_accecn_init_counters(struct tcp_sock *tp)
__tcp_accecn_init_bytes_counters(tp->received_ecn_bytes);
__tcp_accecn_init_bytes_counters(tp->delivered_ecn_bytes);
tp->accecn_minlen = 0;
+ tp->accecn_opt_demand = 0;
tp->est_ecnfield = 0;
}
@@ -433,6 +469,7 @@ static inline void tcp_ecn_rcv_synack(struct sock *sk, const struct tcphdr *th,
default:
tcp_ecn_mode_set(tp, TCP_ECN_MODE_ACCECN);
tp->syn_ect_rcv = ip_dsfield & INET_ECN_MASK;
+ tp->accecn_opt_demand = 2;
if (INET_ECN_is_ce(ip_dsfield) &&
tcp_accecn_validate_syn_feedback(sk, ace,
tp->syn_ect_snt)) {
@@ -453,6 +490,7 @@ static inline void tcp_ecn_rcv_syn(struct tcp_sock *tp, const struct tcphdr *th,
} else {
tp->syn_ect_rcv = TCP_SKB_CB(skb)->ip_dsfield &
INET_ECN_MASK;
+ tp->prev_ecnfield = tp->syn_ect_rcv;
tcp_ecn_mode_set(tp, TCP_ECN_MODE_ACCECN);
}
}
@@ -551,4 +589,16 @@ tcp_ecn_make_synack(const struct request_sock *req, struct tcphdr *th)
th->ece = 1;
}
+static inline bool tcp_accecn_option_beacon_check(const struct sock *sk)
+{
+ const struct tcp_sock *tp = tcp_sk(sk);
+
+ if (!READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_ecn_option_beacon))
+ return false;
+
+ return tcp_stamp_us_delta(tp->tcp_mstamp, tp->accecn_opt_tstamp) *
+ READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_ecn_option_beacon) >=
+ (tp->srtt_us >> 3);
+}
+
#endif /* _LINUX_TCP_ECN_H */
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index 1d7fd86ca7b9..169a393374b3 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -740,6 +740,15 @@ static struct ctl_table ipv4_net_table[] = {
.extra1 = SYSCTL_ZERO,
.extra2 = SYSCTL_TWO,
},
+ {
+ .procname = "tcp_ecn_option_beacon",
+ .data = &init_net.ipv4.sysctl_tcp_ecn_option_beacon,
+ .maxlen = sizeof(u8),
+ .mode = 0644,
+ .proc_handler = proc_dou8vec_minmax,
+ .extra1 = SYSCTL_ZERO,
+ .extra2 = SYSCTL_THREE,
+ },
{
.procname = "tcp_ecn_fallback",
.data = &init_net.ipv4.sysctl_tcp_ecn_fallback,
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index a2b83acf5b94..7bfe763da9f4 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -3395,6 +3395,8 @@ int tcp_disconnect(struct sock *sk, int flags)
tp->wait_third_ack = 0;
tp->accecn_fail_mode = 0;
tcp_accecn_init_counters(tp);
+ tp->prev_ecnfield = 0;
+ tp->accecn_opt_tstamp = 0;
if (icsk->icsk_ca_initialized && icsk->icsk_ca_ops->release)
icsk->icsk_ca_ops->release(sk);
memset(icsk->icsk_ca_priv, 0, sizeof(icsk->icsk_ca_priv));
@@ -5113,11 +5115,12 @@ static void __init tcp_struct_check(void)
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_tx, lsndtime);
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_tx, mdev_us);
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_tx, tcp_wstamp_ns);
+ CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_tx, accecn_opt_tstamp);
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_tx, rtt_seq);
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_tx, tsorted_sent_queue);
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_tx, highest_sack);
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_tx, ecn_flags);
- CACHELINE_ASSERT_GROUP_SIZE(struct tcp_sock, tcp_sock_write_tx, 89);
+ CACHELINE_ASSERT_GROUP_SIZE(struct tcp_sock, tcp_sock_write_tx, 97);
/* TXRX read-write hotpath cache lines */
CACHELINE_ASSERT_GROUP_MEMBER(struct tcp_sock, tcp_sock_write_txrx, pred_flags);
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 107c20ae0f5b..103260fc9130 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -6090,8 +6090,10 @@ static bool tcp_validate_incoming(struct sock *sk, struct sk_buff *skb,
* RFC 5961 4.2 : Send a challenge ack
*/
if (th->syn) {
- if (tcp_ecn_mode_accecn(tp))
+ if (tcp_ecn_mode_accecn(tp)) {
accecn_reflector = true;
+ tcp_accecn_opt_demand_min(sk, 1);
+ }
if (sk->sk_state == TCP_SYN_RECV && sk->sk_socket && th->ack &&
TCP_SKB_CB(skb)->seq + 1 == TCP_SKB_CB(skb)->end_seq &&
TCP_SKB_CB(skb)->seq + 1 == tp->rcv_nxt &&
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 1da7c5e4da32..c32a965e8cd6 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -3452,6 +3452,7 @@ static int __net_init tcp_sk_init(struct net *net)
{
net->ipv4.sysctl_tcp_ecn = TCP_ECN_IN_ECN_OUT_NOECN;
net->ipv4.sysctl_tcp_ecn_option = TCP_ACCECN_OPTION_FULL;
+ net->ipv4.sysctl_tcp_ecn_option_beacon = TCP_ACCECN_OPTION_BEACON;
net->ipv4.sysctl_tcp_ecn_fallback = 1;
net->ipv4.sysctl_tcp_base_mss = TCP_BASE_MSS;
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index 0a0ee33e38ad..848c756a37b8 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -473,6 +473,8 @@ static void tcp_ecn_openreq_child(struct sock *sk,
tcp_ecn_mode_set(tp, TCP_ECN_MODE_ACCECN);
tp->syn_ect_snt = treq->syn_ect_snt;
tcp_accecn_third_ack(sk, skb, treq->syn_ect_snt);
+ tp->prev_ecnfield = treq->syn_ect_rcv;
+ tp->accecn_opt_demand = 1;
tcp_ecn_received_counters_payload(sk, skb);
} else {
tcp_ecn_mode_set(tp, inet_rsk(req)->ecn_ok ?
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 2169fd28594e..65ef1c3c97c2 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -708,8 +708,12 @@ static void tcp_options_write(struct tcphdr *th, struct tcp_sock *tp,
*ptr++ = htonl(((e0b & 0xffffff) << 8) |
TCPOPT_NOP);
}
- if (tp)
+ if (tp) {
tp->accecn_minlen = 0;
+ tp->accecn_opt_tstamp = tp->tcp_mstamp;
+ if (tp->accecn_opt_demand)
+ tp->accecn_opt_demand--;
+ }
}
if (unlikely(OPTION_SACK_ADVERTISE & options)) {
@@ -1151,7 +1155,10 @@ static unsigned int tcp_established_options(struct sock *sk, struct sk_buff *skb
}
if (tcp_ecn_mode_accecn(tp) &&
- READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_ecn_option)) {
+ READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_ecn_option) &&
+ (READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_ecn_option) >= TCP_ACCECN_OPTION_FULL ||
+ tp->accecn_opt_demand ||
+ tcp_accecn_option_beacon_check(sk))) {
opts->use_synack_ecn_bytes = 0;
size += tcp_options_fit_accecn(opts, tp->accecn_minlen,
MAX_TCP_OPTION_SPACE - size);
@@ -2864,6 +2871,11 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle,
sent_pkts = 0;
tcp_mstamp_refresh(tp);
+
+ /* AccECN option beacon depends on mstamp, it may change mss */
+ if (tcp_ecn_mode_accecn(tp) && tcp_accecn_option_beacon_check(sk))
+ mss_now = tcp_current_mss(sk);
+
if (!push_one) {
/* Do MTU probing. */
result = tcp_mtu_probe(sk);
--
2.34.1
^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH v11 net-next 13/15] tcp: accecn: AccECN option failure handling
2025-07-03 5:17 [PATCH v11 net-next 00/15] AccECN protocol patch series chia-yu.chang
` (11 preceding siblings ...)
2025-07-03 5:17 ` [PATCH v11 net-next 12/15] tcp: accecn: AccECN option send control chia-yu.chang
@ 2025-07-03 5:17 ` chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 14/15] tcp: accecn: AccECN option ceb/cep and ACE field multi-wrap heuristics chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 15/15] tcp: accecn: try to fit AccECN option with SACK chia-yu.chang
14 siblings, 0 replies; 16+ messages in thread
From: chia-yu.chang @ 2025-07-03 5:17 UTC (permalink / raw)
To: pabeni, edumazet, linux-doc, corbet, horms, dsahern, kuniyu, bpf,
netdev, dave.taht, jhs, kuba, stephen, xiyou.wangcong, jiri,
davem, andrew+netdev, donald.hunter, ast, liuhangbin, shuah,
linux-kselftest, ij, ncardwell, koen.de_schepper, g.white,
ingemar.s.johansson, mirja.kuehlewind, cheshire, rs.ietf,
Jason_Livingood, vidhi_goel
Cc: Chia-Yu Chang
From: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
AccECN option may fail in various way, handle these:
- Attempt to negotiate the use of AccECN on the 1st retransmitted SYN
- From the 2nd retransmitted SYN, stop AccECN negotiation
- Remove option from SYN/ACK rexmits to handle blackholes
- If no option arrives in SYN/ACK, assume Option is not usable
- If an option arrives later, re-enabled
- If option is zeroed, disable AccECN option processing
This patch use existing padding bits in tcp_request_sock and
holes in tcp_sock without increasing the size.
Signed-off-by: Ilpo Järvinen <ij@kernel.org>
Signed-off-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
---
v9:
- Add comments about the 1st retx SYN still attempt AccECN negotiation
v8:
- Add new helper function tcp_accecn_saw_opt_fail_recv()
---
include/linux/tcp.h | 4 +++-
include/net/tcp_ecn.h | 50 +++++++++++++++++++++++++++++++++++++---
net/ipv4/tcp.c | 1 +
net/ipv4/tcp_input.c | 35 ++++++++++++++++++++++++++--
net/ipv4/tcp_minisocks.c | 14 +++++++++++
net/ipv4/tcp_output.c | 11 +++++++--
6 files changed, 107 insertions(+), 8 deletions(-)
diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index 2c0a2db6e6dd..8609fb138108 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -173,6 +173,7 @@ struct tcp_request_sock {
u8 syn_ect_snt: 2,
syn_ect_rcv: 2,
accecn_fail_mode:4;
+ u8 saw_accecn_opt :2;
#ifdef CONFIG_TCP_AO
u8 ao_keyid;
u8 ao_rcv_next;
@@ -408,7 +409,8 @@ struct tcp_sock {
syn_fastopen_child:1; /* created TFO passive child socket */
u8 keepalive_probes; /* num of allowed keep alive probes */
- u8 accecn_fail_mode:4; /* AccECN failure handling */
+ u8 accecn_fail_mode:4, /* AccECN failure handling */
+ saw_accecn_opt:2; /* An AccECN option was seen */
u32 tcp_tx_delay; /* delay (in usec) added to TX packets */
/* RTT measurement */
diff --git a/include/net/tcp_ecn.h b/include/net/tcp_ecn.h
index 667f4df961b3..1f6394405c4e 100644
--- a/include/net/tcp_ecn.h
+++ b/include/net/tcp_ecn.h
@@ -91,6 +91,11 @@ static inline void tcp_accecn_fail_mode_set(struct tcp_sock *tp, u8 mode)
tp->accecn_fail_mode |= mode;
}
+#define TCP_ACCECN_OPT_NOT_SEEN 0x0
+#define TCP_ACCECN_OPT_EMPTY_SEEN 0x1
+#define TCP_ACCECN_OPT_COUNTER_SEEN 0x2
+#define TCP_ACCECN_OPT_FAIL_SEEN 0x3
+
static inline u8 tcp_accecn_ace(const struct tcphdr *th)
{
return (th->ae << 2) | (th->cwr << 1) | th->ece;
@@ -151,6 +156,14 @@ static inline bool tcp_accecn_validate_syn_feedback(struct sock *sk, u8 ace,
return true;
}
+static inline void tcp_accecn_saw_opt_fail_recv(struct tcp_sock *tp,
+ u8 saw_opt)
+{
+ tp->saw_accecn_opt = saw_opt;
+ if (tp->saw_accecn_opt == TCP_ACCECN_OPT_FAIL_SEEN)
+ tcp_accecn_fail_mode_set(tp, TCP_ACCECN_OPT_FAIL_RECV);
+}
+
/* Validate the 3rd ACK based on the ACE field, see Table 4 of AccECN spec */
static inline void tcp_accecn_third_ack(struct sock *sk,
const struct sk_buff *skb, u8 sent_ect)
@@ -428,9 +441,34 @@ static inline void tcp_accecn_set_ace(struct tcp_sock *tp, struct sk_buff *skb,
}
}
+static inline u8 tcp_accecn_option_init(const struct sk_buff *skb,
+ u8 opt_offset)
+{
+ u8 *ptr = skb_transport_header(skb) + opt_offset;
+ unsigned int optlen = ptr[1] - 2;
+
+ WARN_ON_ONCE(ptr[0] != TCPOPT_ACCECN0 && ptr[0] != TCPOPT_ACCECN1);
+ ptr += 2;
+
+ /* Detect option zeroing: an AccECN connection "MAY check that the
+ * initial value of the EE0B field or the EE1B field is non-zero"
+ */
+ if (optlen < TCPOLEN_ACCECN_PERFIELD)
+ return TCP_ACCECN_OPT_EMPTY_SEEN;
+ if (get_unaligned_be24(ptr) == 0)
+ return TCP_ACCECN_OPT_FAIL_SEEN;
+ if (optlen < TCPOLEN_ACCECN_PERFIELD * 3)
+ return TCP_ACCECN_OPT_COUNTER_SEEN;
+ ptr += TCPOLEN_ACCECN_PERFIELD * 2;
+ if (get_unaligned_be24(ptr) == 0)
+ return TCP_ACCECN_OPT_FAIL_SEEN;
+
+ return TCP_ACCECN_OPT_COUNTER_SEEN;
+}
+
/* See Table 2 of the AccECN draft */
-static inline void tcp_ecn_rcv_synack(struct sock *sk, const struct tcphdr *th,
- u8 ip_dsfield)
+static inline void tcp_ecn_rcv_synack(struct sock *sk, const struct sk_buff *skb,
+ const struct tcphdr *th, u8 ip_dsfield)
{
struct tcp_sock *tp = tcp_sk(sk);
u8 ace = tcp_accecn_ace(th);
@@ -469,7 +507,13 @@ static inline void tcp_ecn_rcv_synack(struct sock *sk, const struct tcphdr *th,
default:
tcp_ecn_mode_set(tp, TCP_ECN_MODE_ACCECN);
tp->syn_ect_rcv = ip_dsfield & INET_ECN_MASK;
- tp->accecn_opt_demand = 2;
+ if (tp->rx_opt.accecn &&
+ tp->saw_accecn_opt < TCP_ACCECN_OPT_COUNTER_SEEN) {
+ u8 saw_opt = tcp_accecn_option_init(skb, tp->rx_opt.accecn);
+
+ tcp_accecn_saw_opt_fail_recv(tp, saw_opt);
+ tp->accecn_opt_demand = 2;
+ }
if (INET_ECN_is_ce(ip_dsfield) &&
tcp_accecn_validate_syn_feedback(sk, ace,
tp->syn_ect_snt)) {
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 7bfe763da9f4..ec001f40307f 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -3394,6 +3394,7 @@ int tcp_disconnect(struct sock *sk, int flags)
tp->delivered_ce = 0;
tp->wait_third_ack = 0;
tp->accecn_fail_mode = 0;
+ tp->saw_accecn_opt = TCP_ACCECN_OPT_NOT_SEEN;
tcp_accecn_init_counters(tp);
tp->prev_ecnfield = 0;
tp->accecn_opt_tstamp = 0;
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 103260fc9130..5a8f5ca9efde 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -398,7 +398,22 @@ static bool tcp_accecn_process_option(struct tcp_sock *tp,
unsigned int i;
u8 *ptr;
+ if (tcp_accecn_opt_fail_recv(tp))
+ return false;
+
if (!(flag & FLAG_SLOWPATH) || !tp->rx_opt.accecn) {
+ if (!tp->saw_accecn_opt) {
+ /* Too late to enable after this point due to
+ * potential counter wraps
+ */
+ if (tp->bytes_sent >= (1 << 23) - 1) {
+ u8 saw_opt = TCP_ACCECN_OPT_FAIL_SEEN;
+
+ tcp_accecn_saw_opt_fail_recv(tp, saw_opt);
+ }
+ return false;
+ }
+
if (estimate_ecnfield) {
u8 ecnfield = estimate_ecnfield - 1;
@@ -414,6 +429,13 @@ static bool tcp_accecn_process_option(struct tcp_sock *tp,
order1 = (ptr[0] == TCPOPT_ACCECN1);
ptr += 2;
+ if (tp->saw_accecn_opt < TCP_ACCECN_OPT_COUNTER_SEEN) {
+ tp->saw_accecn_opt = tcp_accecn_option_init(skb,
+ tp->rx_opt.accecn);
+ if (tp->saw_accecn_opt == TCP_ACCECN_OPT_FAIL_SEEN)
+ tcp_accecn_fail_mode_set(tp, TCP_ACCECN_OPT_FAIL_RECV);
+ }
+
res = !!estimate_ecnfield;
for (i = 0; i < 3; i++) {
u32 init_offset;
@@ -6092,7 +6114,13 @@ static bool tcp_validate_incoming(struct sock *sk, struct sk_buff *skb,
if (th->syn) {
if (tcp_ecn_mode_accecn(tp)) {
accecn_reflector = true;
- tcp_accecn_opt_demand_min(sk, 1);
+ if (tp->rx_opt.accecn &&
+ tp->saw_accecn_opt < TCP_ACCECN_OPT_COUNTER_SEEN) {
+ u8 saw_opt = tcp_accecn_option_init(skb, tp->rx_opt.accecn);
+
+ tcp_accecn_saw_opt_fail_recv(tp, saw_opt);
+ tcp_accecn_opt_demand_min(sk, 1);
+ }
}
if (sk->sk_state == TCP_SYN_RECV && sk->sk_socket && th->ack &&
TCP_SKB_CB(skb)->seq + 1 == TCP_SKB_CB(skb)->end_seq &&
@@ -6582,7 +6610,8 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb,
*/
if (tcp_ecn_mode_any(tp))
- tcp_ecn_rcv_synack(sk, th, TCP_SKB_CB(skb)->ip_dsfield);
+ tcp_ecn_rcv_synack(sk, skb, th,
+ TCP_SKB_CB(skb)->ip_dsfield);
tcp_init_wl(tp, TCP_SKB_CB(skb)->seq);
tcp_try_undo_spurious_syn(sk);
@@ -7157,6 +7186,8 @@ static void tcp_openreq_init(struct request_sock *req,
tcp_rsk(req)->snt_tsval_first = 0;
tcp_rsk(req)->last_oow_ack_time = 0;
tcp_rsk(req)->accecn_ok = 0;
+ tcp_rsk(req)->saw_accecn_opt = TCP_ACCECN_OPT_NOT_SEEN;
+ tcp_rsk(req)->accecn_fail_mode = 0;
tcp_rsk(req)->syn_ect_rcv = 0;
tcp_rsk(req)->syn_ect_snt = 0;
req->mss = rx_opt->mss_clamp;
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index 848c756a37b8..fd80e24ea30b 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -473,6 +473,7 @@ static void tcp_ecn_openreq_child(struct sock *sk,
tcp_ecn_mode_set(tp, TCP_ECN_MODE_ACCECN);
tp->syn_ect_snt = treq->syn_ect_snt;
tcp_accecn_third_ack(sk, skb, treq->syn_ect_snt);
+ tp->saw_accecn_opt = treq->saw_accecn_opt;
tp->prev_ecnfield = treq->syn_ect_rcv;
tp->accecn_opt_demand = 1;
tcp_ecn_received_counters_payload(sk, skb);
@@ -688,6 +689,7 @@ struct sock *tcp_check_req(struct sock *sk, struct sk_buff *skb,
bool own_req;
tmp_opt.saw_tstamp = 0;
+ tmp_opt.accecn = 0;
if (th->doff > (sizeof(struct tcphdr)>>2)) {
tcp_parse_options(sock_net(sk), skb, &tmp_opt, 0, NULL);
@@ -865,6 +867,18 @@ struct sock *tcp_check_req(struct sock *sk, struct sk_buff *skb,
if (!(flg & TCP_FLAG_ACK))
return NULL;
+ if (tcp_rsk(req)->accecn_ok && tmp_opt.accecn &&
+ tcp_rsk(req)->saw_accecn_opt < TCP_ACCECN_OPT_COUNTER_SEEN) {
+ u8 saw_opt = tcp_accecn_option_init(skb, tmp_opt.accecn);
+
+ tcp_rsk(req)->saw_accecn_opt = saw_opt;
+ if (tcp_rsk(req)->saw_accecn_opt == TCP_ACCECN_OPT_FAIL_SEEN) {
+ u8 fail_mode = TCP_ACCECN_OPT_FAIL_RECV;
+
+ tcp_rsk(req)->accecn_fail_mode |= fail_mode;
+ }
+ }
+
/* For Fast Open no more processing is needed (sk is the
* child socket).
*/
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 65ef1c3c97c2..560b0ca54bb8 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -986,9 +986,14 @@ static unsigned int tcp_syn_options(struct sock *sk, struct sk_buff *skb,
}
}
- /* Simultaneous open SYN/ACK needs AccECN option but not SYN */
+ /* Simultaneous open SYN/ACK needs AccECN option but not SYN.
+ * It is attempted to negotiate the use of AccECN also on the first
+ * retransmitted SYN, as mentioned in "3.1.4.1. Retransmitted SYNs"
+ * of AccECN draft.
+ */
if (unlikely((TCP_SKB_CB(skb)->tcp_flags & TCPHDR_ACK) &&
tcp_ecn_mode_accecn(tp) &&
+ inet_csk(sk)->icsk_retransmits < 2 &&
READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_ecn_option) &&
remaining >= TCPOLEN_ACCECN_BASE)) {
opts->use_synack_ecn_bytes = 1;
@@ -1077,7 +1082,7 @@ static unsigned int tcp_synack_options(const struct sock *sk,
if (treq->accecn_ok &&
READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_ecn_option) &&
- remaining >= TCPOLEN_ACCECN_BASE) {
+ req->num_timeout < 1 && remaining >= TCPOLEN_ACCECN_BASE) {
opts->use_synack_ecn_bytes = 1;
remaining -= tcp_options_fit_accecn(opts, 0, remaining);
}
@@ -1156,6 +1161,8 @@ static unsigned int tcp_established_options(struct sock *sk, struct sk_buff *skb
if (tcp_ecn_mode_accecn(tp) &&
READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_ecn_option) &&
+ tp->saw_accecn_opt &&
+ !tcp_accecn_opt_fail_send(tp) &&
(READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_ecn_option) >= TCP_ACCECN_OPTION_FULL ||
tp->accecn_opt_demand ||
tcp_accecn_option_beacon_check(sk))) {
--
2.34.1
^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH v11 net-next 14/15] tcp: accecn: AccECN option ceb/cep and ACE field multi-wrap heuristics
2025-07-03 5:17 [PATCH v11 net-next 00/15] AccECN protocol patch series chia-yu.chang
` (12 preceding siblings ...)
2025-07-03 5:17 ` [PATCH v11 net-next 13/15] tcp: accecn: AccECN option failure handling chia-yu.chang
@ 2025-07-03 5:17 ` chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 15/15] tcp: accecn: try to fit AccECN option with SACK chia-yu.chang
14 siblings, 0 replies; 16+ messages in thread
From: chia-yu.chang @ 2025-07-03 5:17 UTC (permalink / raw)
To: pabeni, edumazet, linux-doc, corbet, horms, dsahern, kuniyu, bpf,
netdev, dave.taht, jhs, kuba, stephen, xiyou.wangcong, jiri,
davem, andrew+netdev, donald.hunter, ast, liuhangbin, shuah,
linux-kselftest, ij, ncardwell, koen.de_schepper, g.white,
ingemar.s.johansson, mirja.kuehlewind, cheshire, rs.ietf,
Jason_Livingood, vidhi_goel
Cc: Chia-Yu Chang
From: Ilpo Järvinen <ij@kernel.org>
The AccECN option ceb/cep heuristic algorithm is from AccECN spec
Appendix A.2.2 to mitigate against false ACE field overflows. Armed
with ceb delta from option, delivered bytes, and delivered packets it
is possible to estimate how many times ACE field wrapped.
This calculation is necessary only if more than one wrap is possible.
Without SACK, delivered bytes and packets are not always trustworthy in
which case TCP falls back to the simpler no-or-all wraps ceb algorithm.
Signed-off-by: Ilpo Järvinen <ij@kernel.org>
Signed-off-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
---
include/net/tcp.h | 1 +
net/ipv4/tcp_input.c | 36 ++++++++++++++++++++++++++++++++++--
2 files changed, 35 insertions(+), 2 deletions(-)
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 4516b4910fae..ad96bd2040dd 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -246,6 +246,7 @@ static_assert((1 << ATO_BITS) > TCP_DELACK_MAX);
#define TCP_ACCECN_MAXSIZE (TCPOLEN_ACCECN_BASE + \
TCPOLEN_ACCECN_PERFIELD * \
TCP_ACCECN_NUMFIELDS)
+#define TCP_ACCECN_SAFETY_SHIFT 1 /* SAFETY_FACTOR in accecn draft */
/* Flags in tp->nonagle */
#define TCP_NAGLE_OFF 1 /* Nagle's algo is disabled */
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 5a8f5ca9efde..094645e0d9fc 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -492,16 +492,19 @@ static u32 __tcp_accecn_process(struct sock *sk, const struct sk_buff *skb,
u32 delivered_pkts, u32 delivered_bytes,
int flag)
{
+ u32 old_ceb = tcp_sk(sk)->delivered_ecn_bytes[INET_ECN_CE - 1];
const struct tcphdr *th = tcp_hdr(skb);
struct tcp_sock *tp = tcp_sk(sk);
- u32 delta, safe_delta;
+ u32 delta, safe_delta, d_ceb;
+ bool opt_deltas_valid;
u32 corrected_ace;
/* Reordered ACK or uncertain due to lack of data to send and ts */
if (!(flag & (FLAG_FORWARD_PROGRESS | FLAG_TS_PROGRESS)))
return 0;
- tcp_accecn_process_option(tp, skb, delivered_bytes, flag);
+ opt_deltas_valid = tcp_accecn_process_option(tp, skb,
+ delivered_bytes, flag);
if (!(flag & FLAG_SLOWPATH)) {
/* AccECN counter might overflow on large ACKs */
@@ -524,6 +527,35 @@ static u32 __tcp_accecn_process(struct sock *sk, const struct sk_buff *skb,
safe_delta = delivered_pkts -
((delivered_pkts - delta) & TCP_ACCECN_CEP_ACE_MASK);
+ if (opt_deltas_valid) {
+ d_ceb = tp->delivered_ecn_bytes[INET_ECN_CE - 1] - old_ceb;
+ if (!d_ceb)
+ return delta;
+
+ if ((delivered_pkts >= (TCP_ACCECN_CEP_ACE_MASK + 1) * 2) &&
+ (tcp_is_sack(tp) ||
+ ((1 << inet_csk(sk)->icsk_ca_state) &
+ (TCPF_CA_Open | TCPF_CA_CWR)))) {
+ u32 est_d_cep;
+
+ if (delivered_bytes <= d_ceb)
+ return safe_delta;
+
+ est_d_cep = DIV_ROUND_UP_ULL((u64)d_ceb *
+ delivered_pkts,
+ delivered_bytes);
+ return min(safe_delta,
+ delta +
+ (est_d_cep & ~TCP_ACCECN_CEP_ACE_MASK));
+ }
+
+ if (d_ceb > delta * tp->mss_cache)
+ return safe_delta;
+ if (d_ceb <
+ safe_delta * tp->mss_cache >> TCP_ACCECN_SAFETY_SHIFT)
+ return delta;
+ }
+
return safe_delta;
}
--
2.34.1
^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH v11 net-next 15/15] tcp: accecn: try to fit AccECN option with SACK
2025-07-03 5:17 [PATCH v11 net-next 00/15] AccECN protocol patch series chia-yu.chang
` (13 preceding siblings ...)
2025-07-03 5:17 ` [PATCH v11 net-next 14/15] tcp: accecn: AccECN option ceb/cep and ACE field multi-wrap heuristics chia-yu.chang
@ 2025-07-03 5:17 ` chia-yu.chang
14 siblings, 0 replies; 16+ messages in thread
From: chia-yu.chang @ 2025-07-03 5:17 UTC (permalink / raw)
To: pabeni, edumazet, linux-doc, corbet, horms, dsahern, kuniyu, bpf,
netdev, dave.taht, jhs, kuba, stephen, xiyou.wangcong, jiri,
davem, andrew+netdev, donald.hunter, ast, liuhangbin, shuah,
linux-kselftest, ij, ncardwell, koen.de_schepper, g.white,
ingemar.s.johansson, mirja.kuehlewind, cheshire, rs.ietf,
Jason_Livingood, vidhi_goel
Cc: Chia-Yu Chang
From: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
As SACK blocks tend to eat all option space when there are
many holes, it is useful to compromise on sending many SACK
blocks in every ACK and attempt to fit the AccECN option
there by reducing the number of SACK blocks. However, it will
never go below two SACK blocks because of the AccECN option.
As the AccECN option is often not put to every ACK, the space
hijack is usually only temporary. Depending on the reuqired
AccECN fields (can be either 3, 2, 1, or 0, cf. Table 5 in
AccECN spec) and the NOPs used for alignment of other
TCP options, up to two SACK blocks will be reduced. Please
find below tables for more details:
+====================+=========================================+
| Number of | Required | Remaining | Number of | Final |
| SACK | AccECN | option | reduced | number of |
| blocks | fields | spaces | SACK blocks | SACK blocks |
+===========+==========+===========+=============+=============+
| x (<=2) | 0 to 3 | any | 0 | x |
+-----------+----------+-----------+-------------+-------------+
| 3 | 0 | any | 0 | 3 |
| 3 | 1 | <4 | 1 | 2 |
| 3 | 1 | >=4 | 0 | 3 |
| 3 | 2 | <8 | 1 | 2 |
| 3 | 2 | >=8 | 0 | 3 |
| 3 | 3 | <12 | 1 | 2 |
| 3 | 3 | >=12 | 0 | 3 |
+-----------+----------+-----------+-------------+-------------+
| y (>=4) | 0 | any | 0 | y |
| y (>=4) | 1 | <4 | 1 | y-1 |
| y (>=4) | 1 | >=4 | 0 | y |
| y (>=4) | 2 | <8 | 1 | y-1 |
| y (>=4) | 2 | >=8 | 0 | y |
| y (>=4) | 3 | <4 | 2 | y-2 |
| y (>=4) | 3 | <12 | 1 | y-1 |
| y (>=4) | 3 | >=12 | 0 | y |
+===========+==========+===========+=============+=============+
Signed-off-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com>
Co-developed-by: Ilpo Järvinen <ij@kernel.org>
Signed-off-by: Ilpo Järvinen <ij@kernel.org>
---
v8:
- Update tcp_options_fit_accecn() to avoid using recursion
---
net/ipv4/tcp_output.c | 21 ++++++++++++++++++++-
1 file changed, 20 insertions(+), 1 deletion(-)
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 560b0ca54bb8..cf1d40e9c0ed 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -876,7 +876,9 @@ static int tcp_options_fit_accecn(struct tcp_out_options *opts, int required,
int remaining)
{
int size = TCP_ACCECN_MAXSIZE;
+ int sack_blocks_reduce = 0;
int max_combine_saving;
+ int rem = remaining;
if (opts->use_synack_ecn_bytes)
max_combine_saving = tcp_synack_options_combine_saving(opts);
@@ -889,14 +891,31 @@ static int tcp_options_fit_accecn(struct tcp_out_options *opts, int required,
if (leftover_size > max_combine_saving)
leftover_size = -((4 - leftover_size) & 0x3);
- if (remaining >= size - leftover_size) {
+ if (rem >= size - leftover_size) {
size -= leftover_size;
break;
+ } else if (opts->num_accecn_fields == required &&
+ opts->num_sack_blocks > 2 &&
+ required > 0) {
+ /* Try to fit the option by removing one SACK block */
+ opts->num_sack_blocks--;
+ sack_blocks_reduce++;
+ rem = rem + TCPOLEN_SACK_PERBLOCK;
+
+ opts->num_accecn_fields = TCP_ACCECN_NUMFIELDS;
+ size = TCP_ACCECN_MAXSIZE;
+ continue;
}
opts->num_accecn_fields--;
size -= TCPOLEN_ACCECN_PERFIELD;
}
+ if (sack_blocks_reduce > 0) {
+ if (opts->num_accecn_fields >= required)
+ size -= sack_blocks_reduce * TCPOLEN_SACK_PERBLOCK;
+ else
+ opts->num_sack_blocks += sack_blocks_reduce;
+ }
if (opts->num_accecn_fields < required)
return 0;
--
2.34.1
^ permalink raw reply related [flat|nested] 16+ messages in thread
end of thread, other threads:[~2025-07-03 5:19 UTC | newest]
Thread overview: 16+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-07-03 5:17 [PATCH v11 net-next 00/15] AccECN protocol patch series chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 01/15] tcp: reorganize SYN ECN code chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 02/15] tcp: fast path functions later chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 03/15] tcp: reorganize tcp_sock_write_txrx group for variables later chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 04/15] tcp: ecn functions in separated include file chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 05/15] tcp: AccECN core chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 06/15] tcp: accecn: AccECN negotiation chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 07/15] tcp: Add wait_third_ack flag for ECN negotiation in simultaneous connect chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 08/15] tcp: accecn: add AccECN rx byte counters chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 09/15] tcp: accecn: AccECN needs to know delivered bytes chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 10/15] tcp: sack option handling improvements chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 11/15] tcp: accecn: AccECN option chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 12/15] tcp: accecn: AccECN option send control chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 13/15] tcp: accecn: AccECN option failure handling chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 14/15] tcp: accecn: AccECN option ceb/cep and ACE field multi-wrap heuristics chia-yu.chang
2025-07-03 5:17 ` [PATCH v11 net-next 15/15] tcp: accecn: try to fit AccECN option with SACK chia-yu.chang
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).