* [PATCH v2 net-next 00/15] udp: Retire UDP-Lite.
@ 2026-03-05 21:49 Kuniyuki Iwashima
2026-03-05 21:49 ` [PATCH v2 net-next 01/15] udp: Make udp[46]_seq_show() static Kuniyuki Iwashima
` (14 more replies)
0 siblings, 15 replies; 19+ messages in thread
From: Kuniyuki Iwashima @ 2026-03-05 21:49 UTC (permalink / raw)
To: Willem de Bruijn, David Ahern, David S . Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni
Cc: Simon Horman, Florian Westphal, Kuniyuki Iwashima,
Kuniyuki Iwashima, netdev
In 2023, syzbot found a null-ptr-deref bug triggered when UDP-Lite
attempted to charge an skb after the total memory usage for UDP-Lite
_and_ UDP exceeded a system-wide threshold, net.ipv4.udp_mem[1].
Since this threshold is shared with UDP, the bug would have been
easy to trigger if any real-world applications were using UDP-Lite;
however, only syzbot ever stumbled upon it.
The bug had persisted since 2016, suggesting that UDP-Lite had
remained unused for 7 years at that point.
The bug was fixed in commit ad42a35bdfc6 ("udplite: Fix NULL pointer
dereference in __sk_mem_raise_allocated()."), and we added another
commit be28c14ac8bb ("udplite: Print deprecation notice.") to
announce the deprecation plan.
Since then, no one has complained, so it is time to officially
retire UDP-Lite.
This series first removes IPv6 and IPv4 UDP-Lite sockets, then
gradually cleans up the remaining dead/unnecessary code within
the UDP stack.
By removing a bunch of conditionals for UDP-Lite from the fast
path, udp_rr with 20,000 flows sees a 10% increase in pps
(13.3 Mpps -> 14.7 Mpps) on an AMD EPYC 7B12 (Zen 2) 64-Core
Processor platform.
[ With FDO, the baseline is much higher and the delta was ~3%,
20.1 Mpps -> 20.7 Mpps ]
Before:
$ nstat > /dev/null; sleep 1; nstat | grep Udp
Udp6InDatagrams 14013408 0.0
Udp6OutDatagrams 14013128 0.0
After:
$ nstat > /dev/null; sleep 1; nstat | grep Udp
Udp6InDatagrams 15491971 0.0
Udp6OutDatagrams 15491671 0.0
$ ./scripts/bloat-o-meter vmlinux.before vmlinux.after
add/remove: 13/75 grow/shrink: 11/75 up/down: 13777/-18401 (-4624)
Function old new delta
udp4_gro_receive 872 866 -6
udp6_gro_receive 910 903 -7
udp_rcv 32 1727 +1695
udpv6_rcv 32 1450 +1418
__udp4_lib_rcv 2045 - -2045
__udp6_lib_rcv 2084 - -2084
udp_unicast_rcv_skb 160 149 -11
udp6_unicast_rcv_skb 196 181 -15
__udp4_lib_mcast_deliver 925 846 -79
__udp6_lib_mcast_deliver 922 810 -112
__udp4_lib_lookup 973 969 -4
__udp6_lib_lookup 940 929 -11
__udp4_lib_lookup_skb 106 100 -6
__udp6_lib_lookup_skb 71 66 -5
udp4_lib_lookup_skb 132 127 -5
udp6_lib_lookup_skb 87 81 -6
udp_queue_rcv_skb 326 356 +30
udpv6_queue_rcv_skb 331 361 +30
udp_queue_rcv_one_skb 1233 914 -319
udpv6_queue_rcv_one_skb 1250 930 -320
__udp_enqueue_schedule_skb 1067 995 -72
udp_rcv_segment 520 480 -40
udp_post_segment_fix_csum 120 - -120
udp_lib_checksum_complete 200 84 -116
udp_err 27 1103 +1076
udpv6_err 36 1417 +1381
__udp4_lib_err 1112 - -1112
__udp6_lib_err 1448 - -1448
udp_recvmsg 1149 994 -155
udpv6_recvmsg 1349 1294 -55
udp_sendmsg 2730 2648 -82
udp_send_skb 909 681 -228
udpv6_sendmsg 3022 2861 -161
udp_v6_send_skb 1214 952 -262
...
Total: Before=18446744073748075501, After=18446744073748070877, chg -0.00%
Changes:
v2:
Patch 4 : Make udp_seq_ops static
Patch 9 : Add a detailed reason in include/uapi/linux/udp.h
v1: https://lore.kernel.org/netdev/20260304193034.1870586-1-kuniyu@google.com/
Kuniyuki Iwashima (15):
udp: Make udp[46]_seq_show() static.
ipv6: Retire UDP-Lite.
ipv6: Remove UDP-Lite support for IPV6_ADDRFORM.
ipv4: Retire UDP-Lite.
udp: Remove UDP-Lite SNMP stats.
smack: Remove IPPROTO_UDPLITE support in security_sock_rcv_skb().
udp: Remove partial csum code in RX.
udp: Remove partial csum code in TX.
udp: Remove UDPLITE_SEND_CSCOV and UDPLITE_RECV_CSCOV.
udp: Remove struct proto.h.udp_table.
udp: Remove udp_table in struct udp_seq_afinfo.
udp: Remove dead check in __udp[46]_lib_lookup() for BPF.
udp: Don't pass udptable to IPv6 socket lookup functions.
udp: Don't pass udptable to IPv4 socket lookup functions.
udp: Don't pass proto to __udp4_lib_rcv() and __udp6_lib_rcv().
include/linux/udp.h | 10 +-
include/net/ipv6.h | 2 -
include/net/ipv6_stubs.h | 7 +-
include/net/netns/mib.h | 5 -
include/net/sock.h | 5 +-
include/net/transp_v6.h | 3 -
include/net/udp.h | 86 +++-----
include/net/udplite.h | 88 --------
include/uapi/linux/udp.h | 2 +
net/core/filter.c | 5 +-
net/ipv4/Makefile | 2 +-
net/ipv4/af_inet.c | 12 --
net/ipv4/proc.c | 16 --
net/ipv4/udp.c | 428 ++++++++++++-------------------------
net/ipv4/udp_bpf.c | 2 -
net/ipv4/udp_diag.c | 128 +++--------
net/ipv4/udp_impl.h | 27 ---
net/ipv4/udp_offload.c | 3 +-
net/ipv4/udplite.c | 135 ------------
net/ipv6/Makefile | 2 +-
net/ipv6/af_inet6.c | 32 +--
net/ipv6/ip6_checksum.c | 2 +-
net/ipv6/ipv6_sockglue.c | 17 +-
net/ipv6/proc.c | 16 --
net/ipv6/udp.c | 305 ++++++++++----------------
net/ipv6/udp_impl.h | 31 ---
net/ipv6/udp_offload.c | 3 +-
net/ipv6/udplite.c | 139 ------------
net/rxrpc/output.c | 2 -
security/smack/smack_lsm.c | 4 +-
30 files changed, 341 insertions(+), 1178 deletions(-)
delete mode 100644 include/net/udplite.h
delete mode 100644 net/ipv4/udp_impl.h
delete mode 100644 net/ipv4/udplite.c
delete mode 100644 net/ipv6/udp_impl.h
delete mode 100644 net/ipv6/udplite.c
--
2.53.0.473.g4a7958ca14-goog
^ permalink raw reply [flat|nested] 19+ messages in thread
* [PATCH v2 net-next 01/15] udp: Make udp[46]_seq_show() static.
2026-03-05 21:49 [PATCH v2 net-next 00/15] udp: Retire UDP-Lite Kuniyuki Iwashima
@ 2026-03-05 21:49 ` Kuniyuki Iwashima
2026-03-05 21:49 ` [PATCH v2 net-next 02/15] ipv6: Retire UDP-Lite Kuniyuki Iwashima
` (13 subsequent siblings)
14 siblings, 0 replies; 19+ messages in thread
From: Kuniyuki Iwashima @ 2026-03-05 21:49 UTC (permalink / raw)
To: Willem de Bruijn, David Ahern, David S . Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni
Cc: Simon Horman, Florian Westphal, Kuniyuki Iwashima,
Kuniyuki Iwashima, netdev, Willem de Bruijn
Since commit a3d2599b2446 ("ipv{4,6}/udp{,lite}: simplify proc
registration"), udp4_seq_show() and udp6_seq_show() are not
used in net/ipv4/udplite.c and net/ipv6/udplite.c.
Instead, udp_seq_ops and udp6_seq_ops are exposed to UDP-Lite.
Let's make udp4_seq_show() and udp6_seq_show() static.
udp_seq_ops and udp6_seq_ops are moved to udp_impl.h so that
we can make them static when the header is removed.
Signed-off-by: Kuniyuki Iwashima <kuniyu@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
---
include/net/udp.h | 3 ---
net/ipv4/udp.c | 3 +--
net/ipv4/udp_impl.h | 2 +-
net/ipv6/udp.c | 3 +--
net/ipv6/udp_impl.h | 2 +-
5 files changed, 4 insertions(+), 9 deletions(-)
diff --git a/include/net/udp.h b/include/net/udp.h
index b648003e5792..f51a51c0e468 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -576,9 +576,6 @@ void *udp_seq_start(struct seq_file *seq, loff_t *pos);
void *udp_seq_next(struct seq_file *seq, void *v, loff_t *pos);
void udp_seq_stop(struct seq_file *seq, void *v);
-extern const struct seq_operations udp_seq_ops;
-extern const struct seq_operations udp6_seq_ops;
-
int udp4_proc_init(void);
void udp4_proc_exit(void);
#endif /* CONFIG_PROC_FS */
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 27384024ebc0..f0e7e2e919ad 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -3438,7 +3438,7 @@ static void udp4_format_sock(struct sock *sp, struct seq_file *f,
sk_drops_read(sp));
}
-int udp4_seq_show(struct seq_file *seq, void *v)
+static int udp4_seq_show(struct seq_file *seq, void *v)
{
seq_setwidth(seq, 127);
if (v == SEQ_START_TOKEN)
@@ -3748,7 +3748,6 @@ const struct seq_operations udp_seq_ops = {
.stop = udp_seq_stop,
.show = udp4_seq_show,
};
-EXPORT_IPV6_MOD(udp_seq_ops);
static struct udp_seq_afinfo udp4_seq_afinfo = {
.family = AF_INET,
diff --git a/net/ipv4/udp_impl.h b/net/ipv4/udp_impl.h
index 17a6fa8b1409..0ca4384f9afa 100644
--- a/net/ipv4/udp_impl.h
+++ b/net/ipv4/udp_impl.h
@@ -22,6 +22,6 @@ int udp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int flags);
void udp_destroy_sock(struct sock *sk);
#ifdef CONFIG_PROC_FS
-int udp4_seq_show(struct seq_file *seq, void *v);
+extern const struct seq_operations udp_seq_ops;
#endif
#endif /* _UDP4_IMPL_H */
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 5a3984e59c90..5fef1c226697 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -1903,7 +1903,7 @@ int udpv6_getsockopt(struct sock *sk, int level, int optname,
/* ------------------------------------------------------------------------ */
#ifdef CONFIG_PROC_FS
-int udp6_seq_show(struct seq_file *seq, void *v)
+static int udp6_seq_show(struct seq_file *seq, void *v)
{
if (v == SEQ_START_TOKEN) {
seq_puts(seq, IPV6_SEQ_DGRAM_HEADER);
@@ -1924,7 +1924,6 @@ const struct seq_operations udp6_seq_ops = {
.stop = udp_seq_stop,
.show = udp6_seq_show,
};
-EXPORT_SYMBOL(udp6_seq_ops);
static struct udp_seq_afinfo udp6_seq_afinfo = {
.family = AF_INET6,
diff --git a/net/ipv6/udp_impl.h b/net/ipv6/udp_impl.h
index 1bd4a573e1bb..525ea600228a 100644
--- a/net/ipv6/udp_impl.h
+++ b/net/ipv6/udp_impl.h
@@ -26,6 +26,6 @@ int udpv6_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int flags);
void udpv6_destroy_sock(struct sock *sk);
#ifdef CONFIG_PROC_FS
-int udp6_seq_show(struct seq_file *seq, void *v);
+extern const struct seq_operations udp6_seq_ops;
#endif
#endif /* _UDP6_IMPL_H */
--
2.53.0.473.g4a7958ca14-goog
^ permalink raw reply related [flat|nested] 19+ messages in thread
* [PATCH v2 net-next 02/15] ipv6: Retire UDP-Lite.
2026-03-05 21:49 [PATCH v2 net-next 00/15] udp: Retire UDP-Lite Kuniyuki Iwashima
2026-03-05 21:49 ` [PATCH v2 net-next 01/15] udp: Make udp[46]_seq_show() static Kuniyuki Iwashima
@ 2026-03-05 21:49 ` Kuniyuki Iwashima
2026-03-05 21:49 ` [PATCH v2 net-next 03/15] ipv6: Remove UDP-Lite support for IPV6_ADDRFORM Kuniyuki Iwashima
` (12 subsequent siblings)
14 siblings, 0 replies; 19+ messages in thread
From: Kuniyuki Iwashima @ 2026-03-05 21:49 UTC (permalink / raw)
To: Willem de Bruijn, David Ahern, David S . Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni
Cc: Simon Horman, Florian Westphal, Kuniyuki Iwashima,
Kuniyuki Iwashima, netdev, Willem de Bruijn
As announced in commit be28c14ac8bb ("udplite: Print deprecation
notice."), it's time to deprecate UDP-Lite.
As a first step, let's drop support for IPv6 UDP-Lite sockets.
We will remove the remaining dead code gradually.
Along with the removal of udplite.c, most of the functions exposed
via udp_impl.h are made static.
The prototypes of udpv6_sendmsg() and udpv6_recvmsg() are moved
to udp.h, but only udpv6_recvmsg() has INDIRECT_CALLABLE_DECLARE()
because udpv6_sendmsg() is exported for rxrpc since commit ed472b0c8783
("rxrpc: Call udp_sendmsg() directly").
Also, udpv6_recvmsg() needs INDIRECT_CALLABLE_SCOPE for
CONFIG_MITIGATION_RETPOLINE=n.
Note that udplite.h is included temporarily for udplite_csum().
Signed-off-by: Kuniyuki Iwashima <kuniyu@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
---
include/net/ipv6.h | 2 -
include/net/transp_v6.h | 3 -
include/net/udp.h | 4 ++
net/ipv6/Makefile | 2 +-
net/ipv6/af_inet6.c | 23 +------
net/ipv6/proc.c | 2 -
net/ipv6/udp.c | 32 ++++-----
net/ipv6/udp_impl.h | 31 ---------
net/ipv6/udplite.c | 139 ----------------------------------------
net/rxrpc/output.c | 2 -
10 files changed, 23 insertions(+), 217 deletions(-)
delete mode 100644 net/ipv6/udp_impl.h
delete mode 100644 net/ipv6/udplite.c
diff --git a/include/net/ipv6.h b/include/net/ipv6.h
index 1c0ce5151275..0958cc5c6ec3 100644
--- a/include/net/ipv6.h
+++ b/include/net/ipv6.h
@@ -1179,8 +1179,6 @@ int tcp6_proc_init(struct net *net);
void tcp6_proc_exit(struct net *net);
int udp6_proc_init(struct net *net);
void udp6_proc_exit(struct net *net);
-int udplite6_proc_init(void);
-void udplite6_proc_exit(void);
int ipv6_misc_proc_init(void);
void ipv6_misc_proc_exit(void);
int snmp6_register_dev(struct inet6_dev *idev);
diff --git a/include/net/transp_v6.h b/include/net/transp_v6.h
index 1a97e3f32029..c0a421fe0c2a 100644
--- a/include/net/transp_v6.h
+++ b/include/net/transp_v6.h
@@ -8,7 +8,6 @@
/* IPv6 transport protocols */
extern struct proto rawv6_prot;
extern struct proto udpv6_prot;
-extern struct proto udplitev6_prot;
extern struct proto tcpv6_prot;
extern struct proto pingv6_prot;
@@ -28,8 +27,6 @@ int rawv6_init(void);
void rawv6_exit(void);
int udpv6_init(void);
void udpv6_exit(void);
-int udplitev6_init(void);
-void udplitev6_exit(void);
int tcpv6_init(void);
void tcpv6_exit(void);
diff --git a/include/net/udp.h b/include/net/udp.h
index f51a51c0e468..05f63e9e00a7 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -282,6 +282,10 @@ typedef struct sock *(*udp_lookup_t)(const struct sk_buff *skb, __be16 sport,
void udp_v6_early_demux(struct sk_buff *skb);
INDIRECT_CALLABLE_DECLARE(int udpv6_rcv(struct sk_buff *));
+int udpv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len);
+INDIRECT_CALLABLE_DECLARE(int udpv6_recvmsg(struct sock *sk, struct msghdr *msg,
+ size_t len, int flags));
+
struct sk_buff *__udp_gso_segment(struct sk_buff *gso_skb,
netdev_features_t features, bool is_ipv6);
diff --git a/net/ipv6/Makefile b/net/ipv6/Makefile
index 0492f1a0b491..2c9ce2ccbde1 100644
--- a/net/ipv6/Makefile
+++ b/net/ipv6/Makefile
@@ -7,7 +7,7 @@ obj-$(CONFIG_IPV6) += ipv6.o
ipv6-y := af_inet6.o anycast.o ip6_output.o ip6_input.o addrconf.o \
addrlabel.o \
- route.o ip6_fib.o ipv6_sockglue.o ndisc.o udp.o udplite.o \
+ route.o ip6_fib.o ipv6_sockglue.o ndisc.o udp.o \
raw.o icmp.o mcast.o reassembly.o tcp_ipv6.o ping.o \
exthdrs.o datagram.o ip6_flowlabel.o inet6_connection_sock.o \
udp_offload.o seg6.o fib6_notifier.o rpl.o ioam6.o
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 0b995a961359..9ad907a5a093 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -43,7 +43,6 @@
#include <net/ip.h>
#include <net/ipv6.h>
#include <net/udp.h>
-#include <net/udplite.h>
#include <net/tcp.h>
#include <net/ping.h>
#include <net/protocol.h>
@@ -644,8 +643,6 @@ int inet6_compat_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
EXPORT_SYMBOL_GPL(inet6_compat_ioctl);
#endif /* CONFIG_COMPAT */
-INDIRECT_CALLABLE_DECLARE(int udpv6_sendmsg(struct sock *, struct msghdr *,
- size_t));
int inet6_sendmsg(struct socket *sock, struct msghdr *msg, size_t size)
{
struct sock *sk = sock->sk;
@@ -660,8 +657,6 @@ int inet6_sendmsg(struct socket *sock, struct msghdr *msg, size_t size)
sk, msg, size);
}
-INDIRECT_CALLABLE_DECLARE(int udpv6_recvmsg(struct sock *, struct msghdr *,
- size_t, int));
int inet6_recvmsg(struct socket *sock, struct msghdr *msg, size_t size,
int flags)
{
@@ -1088,13 +1083,9 @@ static int __init inet6_init(void)
if (err)
goto out_unregister_tcp_proto;
- err = proto_register(&udplitev6_prot, 1);
- if (err)
- goto out_unregister_udp_proto;
-
err = proto_register(&rawv6_prot, 1);
if (err)
- goto out_unregister_udplite_proto;
+ goto out_unregister_udp_proto;
err = proto_register(&pingv6_prot, 1);
if (err)
@@ -1145,8 +1136,6 @@ static int __init inet6_init(void)
err = -ENOMEM;
if (raw6_proc_init())
goto proc_raw6_fail;
- if (udplite6_proc_init())
- goto proc_udplite6_fail;
if (ipv6_misc_proc_init())
goto proc_misc6_fail;
if (if6_proc_init())
@@ -1182,10 +1171,6 @@ static int __init inet6_init(void)
if (err)
goto udpv6_fail;
- err = udplitev6_init();
- if (err)
- goto udplitev6_fail;
-
err = udpv6_offload_init();
if (err)
goto udpv6_offload_fail;
@@ -1256,8 +1241,6 @@ static int __init inet6_init(void)
tcpv6_fail:
udpv6_offload_exit();
udpv6_offload_fail:
- udplitev6_exit();
-udplitev6_fail:
udpv6_exit();
udpv6_fail:
ipv6_frag_exit();
@@ -1279,8 +1262,6 @@ static int __init inet6_init(void)
proc_if6_fail:
ipv6_misc_proc_exit();
proc_misc6_fail:
- udplite6_proc_exit();
-proc_udplite6_fail:
raw6_proc_exit();
proc_raw6_fail:
#endif
@@ -1304,8 +1285,6 @@ static int __init inet6_init(void)
proto_unregister(&pingv6_prot);
out_unregister_raw_proto:
proto_unregister(&rawv6_prot);
-out_unregister_udplite_proto:
- proto_unregister(&udplitev6_prot);
out_unregister_udp_proto:
proto_unregister(&udpv6_prot);
out_unregister_tcp_proto:
diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c
index 73296f38c252..21bfc73152f0 100644
--- a/net/ipv6/proc.c
+++ b/net/ipv6/proc.c
@@ -39,8 +39,6 @@ static int sockstat6_seq_show(struct seq_file *seq, void *v)
sock_prot_inuse_get(net, &tcpv6_prot));
seq_printf(seq, "UDP6: inuse %d\n",
sock_prot_inuse_get(net, &udpv6_prot));
- seq_printf(seq, "UDPLITE6: inuse %d\n",
- sock_prot_inuse_get(net, &udplitev6_prot));
seq_printf(seq, "RAW6: inuse %d\n",
sock_prot_inuse_get(net, &rawv6_prot));
seq_printf(seq, "FRAG6: inuse %u memory %lu\n",
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 5fef1c226697..aa859bb0527d 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -37,6 +37,7 @@
#include <trace/events/udp.h>
#include <net/addrconf.h>
+#include <net/aligned_data.h>
#include <net/ndisc.h>
#include <net/protocol.h>
#include <net/transp_v6.h>
@@ -57,7 +58,7 @@
#include <linux/proc_fs.h>
#include <linux/seq_file.h>
#include <trace/events/skb.h>
-#include "udp_impl.h"
+#include <net/udplite.h>
static void udpv6_destruct_sock(struct sock *sk)
{
@@ -65,7 +66,7 @@ static void udpv6_destruct_sock(struct sock *sk)
inet6_sock_destruct(sk);
}
-int udpv6_init_sock(struct sock *sk)
+static int udpv6_init_sock(struct sock *sk)
{
int res = udp_lib_init_sock(sk);
@@ -95,7 +96,7 @@ u32 udp6_ehashfn(const struct net *net,
udp6_ehash_secret + net_hash_mix(net));
}
-int udp_v6_get_port(struct sock *sk, unsigned short snum)
+static int udp_v6_get_port(struct sock *sk, unsigned short snum)
{
unsigned int hash2_nulladdr =
ipv6_portaddr_hash(sock_net(sk), &in6addr_any, snum);
@@ -107,7 +108,7 @@ int udp_v6_get_port(struct sock *sk, unsigned short snum)
return udp_lib_get_port(sk, snum, hash2_nulladdr);
}
-void udp_v6_rehash(struct sock *sk)
+static void udp_v6_rehash(struct sock *sk)
{
u16 new_hash = ipv6_portaddr_hash(sock_net(sk),
&sk->sk_v6_rcv_saddr,
@@ -464,6 +465,7 @@ static int udp6_skb_len(struct sk_buff *skb)
* return it, otherwise we block.
*/
+INDIRECT_CALLABLE_SCOPE
int udpv6_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
int flags)
{
@@ -700,9 +702,9 @@ static struct sock *__udp6_lib_err_encap(struct net *net,
return sk;
}
-int __udp6_lib_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
- u8 type, u8 code, int offset, __be32 info,
- struct udp_table *udptable)
+static int __udp6_lib_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
+ u8 type, u8 code, int offset, __be32 info,
+ struct udp_table *udptable)
{
struct ipv6_pinfo *np;
const struct ipv6hdr *hdr = (const struct ipv6hdr *)skb->data;
@@ -1115,8 +1117,8 @@ static int udp6_csum_init(struct sk_buff *skb, struct udphdr *uh, int proto)
return 0;
}
-int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,
- int proto)
+static int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,
+ int proto)
{
enum skb_drop_reason reason = SKB_DROP_REASON_NOT_SPECIFIED;
const struct in6_addr *saddr, *daddr;
@@ -1854,7 +1856,7 @@ static void udpv6_splice_eof(struct socket *sock)
release_sock(sk);
}
-void udpv6_destroy_sock(struct sock *sk)
+static void udpv6_destroy_sock(struct sock *sk)
{
struct udp_sock *up = udp_sk(sk);
lock_sock(sk);
@@ -1882,8 +1884,8 @@ void udpv6_destroy_sock(struct sock *sk)
/*
* Socket option code for UDP
*/
-int udpv6_setsockopt(struct sock *sk, int level, int optname, sockptr_t optval,
- unsigned int optlen)
+static int udpv6_setsockopt(struct sock *sk, int level, int optname,
+ sockptr_t optval, unsigned int optlen)
{
if (level == SOL_UDP || level == SOL_UDPLITE || level == SOL_SOCKET)
return udp_lib_setsockopt(sk, level, optname,
@@ -1892,8 +1894,8 @@ int udpv6_setsockopt(struct sock *sk, int level, int optname, sockptr_t optval,
return ipv6_setsockopt(sk, level, optname, optval, optlen);
}
-int udpv6_getsockopt(struct sock *sk, int level, int optname,
- char __user *optval, int __user *optlen)
+static int udpv6_getsockopt(struct sock *sk, int level, int optname,
+ char __user *optval, int __user *optlen)
{
if (level == SOL_UDP || level == SOL_UDPLITE)
return udp_lib_getsockopt(sk, level, optname, optval, optlen);
@@ -1918,7 +1920,7 @@ static int udp6_seq_show(struct seq_file *seq, void *v)
return 0;
}
-const struct seq_operations udp6_seq_ops = {
+static const struct seq_operations udp6_seq_ops = {
.start = udp_seq_start,
.next = udp_seq_next,
.stop = udp_seq_stop,
diff --git a/net/ipv6/udp_impl.h b/net/ipv6/udp_impl.h
deleted file mode 100644
index 525ea600228a..000000000000
--- a/net/ipv6/udp_impl.h
+++ /dev/null
@@ -1,31 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef _UDP6_IMPL_H
-#define _UDP6_IMPL_H
-#include <net/aligned_data.h>
-#include <net/udp.h>
-#include <net/udplite.h>
-#include <net/protocol.h>
-#include <net/addrconf.h>
-#include <net/inet_common.h>
-#include <net/transp_v6.h>
-
-int __udp6_lib_rcv(struct sk_buff *, struct udp_table *, int);
-int __udp6_lib_err(struct sk_buff *, struct inet6_skb_parm *, u8, u8, int,
- __be32, struct udp_table *);
-
-int udpv6_init_sock(struct sock *sk);
-int udp_v6_get_port(struct sock *sk, unsigned short snum);
-void udp_v6_rehash(struct sock *sk);
-
-int udpv6_getsockopt(struct sock *sk, int level, int optname,
- char __user *optval, int __user *optlen);
-int udpv6_setsockopt(struct sock *sk, int level, int optname, sockptr_t optval,
- unsigned int optlen);
-int udpv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len);
-int udpv6_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int flags);
-void udpv6_destroy_sock(struct sock *sk);
-
-#ifdef CONFIG_PROC_FS
-extern const struct seq_operations udp6_seq_ops;
-#endif
-#endif /* _UDP6_IMPL_H */
diff --git a/net/ipv6/udplite.c b/net/ipv6/udplite.c
deleted file mode 100644
index e867721cda4d..000000000000
--- a/net/ipv6/udplite.c
+++ /dev/null
@@ -1,139 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0-or-later
-/*
- * UDPLITEv6 An implementation of the UDP-Lite protocol over IPv6.
- * See also net/ipv4/udplite.c
- *
- * Authors: Gerrit Renker <gerrit@erg.abdn.ac.uk>
- *
- * Changes:
- * Fixes:
- */
-#define pr_fmt(fmt) "UDPLite6: " fmt
-
-#include <linux/export.h>
-#include <linux/proc_fs.h>
-#include "udp_impl.h"
-
-static int udplitev6_sk_init(struct sock *sk)
-{
- pr_warn_once("UDP-Lite is deprecated and scheduled to be removed in 2025, "
- "please contact the netdev mailing list\n");
- return udpv6_init_sock(sk);
-}
-
-static int udplitev6_rcv(struct sk_buff *skb)
-{
- return __udp6_lib_rcv(skb, &udplite_table, IPPROTO_UDPLITE);
-}
-
-static int udplitev6_err(struct sk_buff *skb,
- struct inet6_skb_parm *opt,
- u8 type, u8 code, int offset, __be32 info)
-{
- return __udp6_lib_err(skb, opt, type, code, offset, info,
- &udplite_table);
-}
-
-static const struct inet6_protocol udplitev6_protocol = {
- .handler = udplitev6_rcv,
- .err_handler = udplitev6_err,
- .flags = INET6_PROTO_NOPOLICY|INET6_PROTO_FINAL,
-};
-
-struct proto udplitev6_prot = {
- .name = "UDPLITEv6",
- .owner = THIS_MODULE,
- .close = udp_lib_close,
- .connect = ip6_datagram_connect,
- .disconnect = udp_disconnect,
- .ioctl = udp_ioctl,
- .init = udplitev6_sk_init,
- .destroy = udpv6_destroy_sock,
- .setsockopt = udpv6_setsockopt,
- .getsockopt = udpv6_getsockopt,
- .sendmsg = udpv6_sendmsg,
- .recvmsg = udpv6_recvmsg,
- .hash = udp_lib_hash,
- .unhash = udp_lib_unhash,
- .rehash = udp_v6_rehash,
- .get_port = udp_v6_get_port,
-
- .memory_allocated = &net_aligned_data.udp_memory_allocated,
- .per_cpu_fw_alloc = &udp_memory_per_cpu_fw_alloc,
-
- .sysctl_mem = sysctl_udp_mem,
- .sysctl_wmem_offset = offsetof(struct net, ipv4.sysctl_udp_wmem_min),
- .sysctl_rmem_offset = offsetof(struct net, ipv4.sysctl_udp_rmem_min),
- .obj_size = sizeof(struct udp6_sock),
- .ipv6_pinfo_offset = offsetof(struct udp6_sock, inet6),
- .h.udp_table = &udplite_table,
-};
-
-static struct inet_protosw udplite6_protosw = {
- .type = SOCK_DGRAM,
- .protocol = IPPROTO_UDPLITE,
- .prot = &udplitev6_prot,
- .ops = &inet6_dgram_ops,
- .flags = INET_PROTOSW_PERMANENT,
-};
-
-int __init udplitev6_init(void)
-{
- int ret;
-
- ret = inet6_add_protocol(&udplitev6_protocol, IPPROTO_UDPLITE);
- if (ret)
- goto out;
-
- ret = inet6_register_protosw(&udplite6_protosw);
- if (ret)
- goto out_udplitev6_protocol;
-out:
- return ret;
-
-out_udplitev6_protocol:
- inet6_del_protocol(&udplitev6_protocol, IPPROTO_UDPLITE);
- goto out;
-}
-
-void udplitev6_exit(void)
-{
- inet6_unregister_protosw(&udplite6_protosw);
- inet6_del_protocol(&udplitev6_protocol, IPPROTO_UDPLITE);
-}
-
-#ifdef CONFIG_PROC_FS
-static struct udp_seq_afinfo udplite6_seq_afinfo = {
- .family = AF_INET6,
- .udp_table = &udplite_table,
-};
-
-static int __net_init udplite6_proc_init_net(struct net *net)
-{
- if (!proc_create_net_data("udplite6", 0444, net->proc_net,
- &udp6_seq_ops, sizeof(struct udp_iter_state),
- &udplite6_seq_afinfo))
- return -ENOMEM;
- return 0;
-}
-
-static void __net_exit udplite6_proc_exit_net(struct net *net)
-{
- remove_proc_entry("udplite6", net->proc_net);
-}
-
-static struct pernet_operations udplite6_net_ops = {
- .init = udplite6_proc_init_net,
- .exit = udplite6_proc_exit_net,
-};
-
-int __init udplite6_proc_init(void)
-{
- return register_pernet_subsys(&udplite6_net_ops);
-}
-
-void udplite6_proc_exit(void)
-{
- unregister_pernet_subsys(&udplite6_net_ops);
-}
-#endif
diff --git a/net/rxrpc/output.c b/net/rxrpc/output.c
index d70db367e358..e5880116e087 100644
--- a/net/rxrpc/output.c
+++ b/net/rxrpc/output.c
@@ -16,8 +16,6 @@
#include <net/udp.h>
#include "ar-internal.h"
-extern int udpv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len);
-
ssize_t do_udp_sendmsg(struct socket *socket, struct msghdr *msg, size_t len)
{
struct sockaddr *sa = msg->msg_name;
--
2.53.0.473.g4a7958ca14-goog
^ permalink raw reply related [flat|nested] 19+ messages in thread
* [PATCH v2 net-next 03/15] ipv6: Remove UDP-Lite support for IPV6_ADDRFORM.
2026-03-05 21:49 [PATCH v2 net-next 00/15] udp: Retire UDP-Lite Kuniyuki Iwashima
2026-03-05 21:49 ` [PATCH v2 net-next 01/15] udp: Make udp[46]_seq_show() static Kuniyuki Iwashima
2026-03-05 21:49 ` [PATCH v2 net-next 02/15] ipv6: Retire UDP-Lite Kuniyuki Iwashima
@ 2026-03-05 21:49 ` Kuniyuki Iwashima
2026-03-05 21:49 ` [PATCH v2 net-next 04/15] ipv4: Retire UDP-Lite Kuniyuki Iwashima
` (11 subsequent siblings)
14 siblings, 0 replies; 19+ messages in thread
From: Kuniyuki Iwashima @ 2026-03-05 21:49 UTC (permalink / raw)
To: Willem de Bruijn, David Ahern, David S . Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni
Cc: Simon Horman, Florian Westphal, Kuniyuki Iwashima,
Kuniyuki Iwashima, netdev, Willem de Bruijn
We cannot create IPv6 UDP-Lite sockets anymore.
Let's remove dead code in IPV6_ADDRFORM.
Signed-off-by: Kuniyuki Iwashima <kuniyu@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
---
net/ipv6/ipv6_sockglue.c | 17 ++++-------------
1 file changed, 4 insertions(+), 13 deletions(-)
diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c
index 02c4cab60c69..b4c977434c2e 100644
--- a/net/ipv6/ipv6_sockglue.c
+++ b/net/ipv6/ipv6_sockglue.c
@@ -45,7 +45,6 @@
#include <net/inet_common.h>
#include <net/tcp.h>
#include <net/udp.h>
-#include <net/udplite.h>
#include <net/xfrm.h>
#include <net/compat.h>
#include <net/seg6.h>
@@ -563,10 +562,8 @@ int do_ipv6_setsockopt(struct sock *sk, int level, int optname,
if (sk->sk_type == SOCK_RAW)
break;
- if (sk->sk_protocol == IPPROTO_UDP ||
- sk->sk_protocol == IPPROTO_UDPLITE) {
- struct udp_sock *up = udp_sk(sk);
- if (up->pending == AF_INET6) {
+ if (sk->sk_protocol == IPPROTO_UDP) {
+ if (udp_sk(sk)->pending == AF_INET6) {
retv = -EBUSY;
break;
}
@@ -607,16 +604,11 @@ int do_ipv6_setsockopt(struct sock *sk, int level, int optname,
WRITE_ONCE(sk->sk_family, PF_INET);
tcp_sync_mss(sk, icsk->icsk_pmtu_cookie);
} else {
- struct proto *prot = &udp_prot;
-
- if (sk->sk_protocol == IPPROTO_UDPLITE)
- prot = &udplite_prot;
-
sock_prot_inuse_add(net, sk->sk_prot, -1);
- sock_prot_inuse_add(net, prot, 1);
+ sock_prot_inuse_add(net, &udp_prot, 1);
/* Paired with READ_ONCE(sk->sk_prot) in inet6_dgram_ops */
- WRITE_ONCE(sk->sk_prot, prot);
+ WRITE_ONCE(sk->sk_prot, &udp_prot);
WRITE_ONCE(sk->sk_socket->ops, &inet_dgram_ops);
WRITE_ONCE(sk->sk_family, PF_INET);
}
@@ -1098,7 +1090,6 @@ int do_ipv6_getsockopt(struct sock *sk, int level, int optname,
switch (optname) {
case IPV6_ADDRFORM:
if (sk->sk_protocol != IPPROTO_UDP &&
- sk->sk_protocol != IPPROTO_UDPLITE &&
sk->sk_protocol != IPPROTO_TCP)
return -ENOPROTOOPT;
if (sk->sk_state != TCP_ESTABLISHED)
--
2.53.0.473.g4a7958ca14-goog
^ permalink raw reply related [flat|nested] 19+ messages in thread
* [PATCH v2 net-next 04/15] ipv4: Retire UDP-Lite.
2026-03-05 21:49 [PATCH v2 net-next 00/15] udp: Retire UDP-Lite Kuniyuki Iwashima
` (2 preceding siblings ...)
2026-03-05 21:49 ` [PATCH v2 net-next 03/15] ipv6: Remove UDP-Lite support for IPV6_ADDRFORM Kuniyuki Iwashima
@ 2026-03-05 21:49 ` Kuniyuki Iwashima
2026-03-05 21:49 ` [PATCH v2 net-next 05/15] udp: Remove UDP-Lite SNMP stats Kuniyuki Iwashima
` (10 subsequent siblings)
14 siblings, 0 replies; 19+ messages in thread
From: Kuniyuki Iwashima @ 2026-03-05 21:49 UTC (permalink / raw)
To: Willem de Bruijn, David Ahern, David S . Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni
Cc: Simon Horman, Florian Westphal, Kuniyuki Iwashima,
Kuniyuki Iwashima, netdev, Willem de Bruijn
We have deprecated IPv6 UDP-Lite sockets.
Let's drop support for IPv4 UDP-Lite sockets as well.
Most of the changes are similar to the IPv6 patch: removing
udplite.c and udp_impl.h, marking most functions in udp_impl.h
as static, moving the prototype for udp_recvmsg() to udp.h, and
adding INDIRECT_CALLABLE_SCOPE for it.
In addition, the INET_DIAG support for UDP-Lite is dropped.
We will remove the remaining dead code in the following patches.
Signed-off-by: Kuniyuki Iwashima <kuniyu@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
---
v2: Make udp_seq_ops static
---
include/net/sock.h | 4 +-
include/net/udp.h | 7 ++-
include/net/udplite.h | 4 --
net/ipv4/Makefile | 2 +-
net/ipv4/af_inet.c | 6 --
net/ipv4/proc.c | 3 -
net/ipv4/udp.c | 33 ++++++-----
net/ipv4/udp_bpf.c | 2 -
net/ipv4/udp_diag.c | 47 +--------------
net/ipv4/udp_impl.h | 27 ---------
net/ipv4/udplite.c | 135 ------------------------------------------
11 files changed, 26 insertions(+), 244 deletions(-)
delete mode 100644 net/ipv4/udp_impl.h
delete mode 100644 net/ipv4/udplite.c
diff --git a/include/net/sock.h b/include/net/sock.h
index 6c3f1340e8ef..16a1b8895206 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -126,14 +126,14 @@ typedef __u64 __bitwise __addrpair;
* @skc_bypass_prot_mem: bypass the per-protocol memory accounting for skb
* @skc_bound_dev_if: bound device index if != 0
* @skc_bind_node: bind hash linkage for various protocol lookup tables
- * @skc_portaddr_node: second hash linkage for UDP/UDP-Lite protocol
+ * @skc_portaddr_node: second hash linkage for UDP
* @skc_prot: protocol handlers inside a network family
* @skc_net: reference to the network namespace of this socket
* @skc_v6_daddr: IPV6 destination address
* @skc_v6_rcv_saddr: IPV6 source address
* @skc_cookie: socket's cookie value
* @skc_node: main hash linkage for various protocol lookup tables
- * @skc_nulls_node: main hash linkage for TCP/UDP/UDP-Lite protocol
+ * @skc_nulls_node: main hash linkage for TCP
* @skc_tx_queue_mapping: tx queue number for this connection
* @skc_rx_queue_mapping: rx queue number for this connection
* @skc_flags: place holder for sk_flags
diff --git a/include/net/udp.h b/include/net/udp.h
index 05f63e9e00a7..39223e2692e9 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -105,7 +105,7 @@ struct udp_table {
unsigned int log;
};
extern struct udp_table udp_table;
-void udp_table_init(struct udp_table *, const char *);
+
static inline struct udp_hslot *udp_hashslot(struct udp_table *table,
const struct net *net,
unsigned int num)
@@ -312,7 +312,7 @@ static inline void udp_drops_inc(struct sock *sk)
numa_drop_add(&udp_sk(sk)->drop_counters, 1);
}
-/* hash routines shared between UDPv4/6 and UDP-Litev4/6 */
+/* hash routines shared between UDPv4/6 */
static inline int udp_lib_hash(struct sock *sk)
{
BUG();
@@ -420,6 +420,8 @@ bool udp_sk_rx_dst_set(struct sock *sk, struct dst_entry *dst);
int udp_err(struct sk_buff *, u32);
int udp_abort(struct sock *sk, int err);
int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len);
+INDIRECT_CALLABLE_DECLARE(int udp_recvmsg(struct sock *sk, struct msghdr *msg,
+ size_t len, int flags));
void udp_splice_eof(struct socket *sock);
int udp_push_pending_frames(struct sock *sk);
void udp_flush_pending_frames(struct sock *sk);
@@ -427,7 +429,6 @@ int udp_cmsg_send(struct sock *sk, struct msghdr *msg, u16 *gso_size);
void udp4_hwcsum(struct sk_buff *skb, __be32 src, __be32 dst);
int udp_rcv(struct sk_buff *skb);
int udp_ioctl(struct sock *sk, int cmd, int *karg);
-int udp_init_sock(struct sock *sk);
int udp_pre_connect(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len);
int __udp_disconnect(struct sock *sk, int flags);
int udp_disconnect(struct sock *sk, int flags);
diff --git a/include/net/udplite.h b/include/net/udplite.h
index 786919d29f8d..fdd769745ac4 100644
--- a/include/net/udplite.h
+++ b/include/net/udplite.h
@@ -12,9 +12,6 @@
#define UDPLITE_SEND_CSCOV 10 /* sender partial coverage (as sent) */
#define UDPLITE_RECV_CSCOV 11 /* receiver partial coverage (threshold ) */
-extern struct proto udplite_prot;
-extern struct udp_table udplite_table;
-
/*
* Checksum computation is all in software, hence simpler getfrag.
*/
@@ -84,5 +81,4 @@ static inline __wsum udplite_csum(struct sk_buff *skb)
return skb_checksum(skb, off, len, 0);
}
-void udplite4_register(void);
#endif /* _UDPLITE_H */
diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile
index 18108a6f0499..7f9f98813986 100644
--- a/net/ipv4/Makefile
+++ b/net/ipv4/Makefile
@@ -10,7 +10,7 @@ obj-y := route.o inetpeer.o protocol.o \
tcp.o tcp_input.o tcp_output.o tcp_timer.o tcp_ipv4.o \
tcp_minisocks.o tcp_cong.o tcp_metrics.o tcp_fastopen.o \
tcp_recovery.o tcp_ulp.o \
- tcp_offload.o tcp_plb.o datagram.o raw.o udp.o udplite.o \
+ tcp_offload.o tcp_plb.o datagram.o raw.o udp.o \
udp_offload.o arp.o icmp.o devinet.o af_inet.o igmp.o \
fib_frontend.o fib_semantics.o fib_trie.o fib_notifier.o \
inet_fragment.o ping.o ip_tunnel_core.o gre_offload.o \
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index babcd75a08e2..429d7150a19b 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -104,7 +104,6 @@
#include <net/tcp.h>
#include <net/psp.h>
#include <net/udp.h>
-#include <net/udplite.h>
#include <net/ping.h>
#include <linux/skbuff.h>
#include <net/sock.h>
@@ -876,8 +875,6 @@ void inet_splice_eof(struct socket *sock)
}
EXPORT_SYMBOL_GPL(inet_splice_eof);
-INDIRECT_CALLABLE_DECLARE(int udp_recvmsg(struct sock *, struct msghdr *,
- size_t, int));
int inet_recvmsg(struct socket *sock, struct msghdr *msg, size_t size,
int flags)
{
@@ -1975,9 +1972,6 @@ static int __init inet_init(void)
/* Setup UDP memory threshold */
udp_init();
- /* Add UDP-Lite (RFC 3828) */
- udplite4_register();
-
raw_init();
ping_init();
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index 974afc4ecbe2..cf51f8fcf34b 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -35,7 +35,6 @@
#include <net/mptcp.h>
#include <net/proto_memory.h>
#include <net/udp.h>
-#include <net/udplite.h>
#include <linux/bottom_half.h>
#include <linux/inetdevice.h>
#include <linux/proc_fs.h>
@@ -65,8 +64,6 @@ static int sockstat_seq_show(struct seq_file *seq, void *v)
seq_printf(seq, "UDP: inuse %d mem %ld\n",
sock_prot_inuse_get(net, &udp_prot),
proto_memory_allocated(&udp_prot));
- seq_printf(seq, "UDPLITE: inuse %d\n",
- sock_prot_inuse_get(net, &udplite_prot));
seq_printf(seq, "RAW: inuse %d\n",
sock_prot_inuse_get(net, &raw_prot));
seq_printf(seq, "FRAG: inuse %u memory %lu\n",
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index f0e7e2e919ad..ff6237749eb1 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -98,8 +98,10 @@
#include <linux/skbuff.h>
#include <linux/proc_fs.h>
#include <linux/seq_file.h>
+#include <net/aligned_data.h>
#include <net/net_namespace.h>
#include <net/icmp.h>
+#include <net/inet_common.h>
#include <net/inet_hashtables.h>
#include <net/ip.h>
#include <net/ip_tunnels.h>
@@ -112,10 +114,10 @@
#include <linux/btf_ids.h>
#include <trace/events/skb.h>
#include <net/busy_poll.h>
-#include "udp_impl.h"
#include <net/sock_reuseport.h>
#include <net/addrconf.h>
#include <net/udp_tunnel.h>
+#include <net/udplite.h>
#include <net/gro.h>
#if IS_ENABLED(CONFIG_IPV6)
#include <net/ipv6_stubs.h>
@@ -229,7 +231,7 @@ static int udp_reuseport_add_sock(struct sock *sk, struct udp_hslot *hslot)
}
/**
- * udp_lib_get_port - UDP/-Lite port lookup for IPv4 and IPv6
+ * udp_lib_get_port - UDP port lookup for IPv4 and IPv6
*
* @sk: socket struct in question
* @snum: port number to look up
@@ -353,7 +355,7 @@ int udp_lib_get_port(struct sock *sk, unsigned short snum,
}
EXPORT_IPV6_MOD(udp_lib_get_port);
-int udp_v4_get_port(struct sock *sk, unsigned short snum)
+static int udp_v4_get_port(struct sock *sk, unsigned short snum)
{
unsigned int hash2_nulladdr =
ipv4_portaddr_hash(sock_net(sk), htonl(INADDR_ANY), snum);
@@ -928,7 +930,7 @@ static struct sock *__udp4_lib_err_encap(struct net *net,
* to find the appropriate port.
*/
-int __udp4_lib_err(struct sk_buff *skb, u32 info, struct udp_table *udptable)
+static int __udp4_lib_err(struct sk_buff *skb, u32 info, struct udp_table *udptable)
{
struct inet_sock *inet;
const struct iphdr *iph = (const struct iphdr *)skb->data;
@@ -1855,7 +1857,7 @@ static void udp_destruct_sock(struct sock *sk)
inet_sock_destruct(sk);
}
-int udp_init_sock(struct sock *sk)
+static int udp_init_sock(struct sock *sk)
{
int res = udp_lib_init_sock(sk);
@@ -2070,6 +2072,7 @@ EXPORT_IPV6_MOD(udp_read_skb);
* return it, otherwise we block.
*/
+INDIRECT_CALLABLE_SCOPE
int udp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int flags)
{
struct inet_sock *inet = inet_sk(sk);
@@ -2337,7 +2340,7 @@ void udp_lib_rehash(struct sock *sk, u16 newhash, u16 newhash4)
}
EXPORT_IPV6_MOD(udp_lib_rehash);
-void udp_v4_rehash(struct sock *sk)
+static void udp_v4_rehash(struct sock *sk)
{
u16 new_hash = ipv4_portaddr_hash(sock_net(sk),
inet_sk(sk)->inet_rcv_saddr,
@@ -2683,8 +2686,8 @@ static int udp_unicast_rcv_skb(struct sock *sk, struct sk_buff *skb,
* All we need to do is get the socket, and then do a checksum.
*/
-int __udp4_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,
- int proto)
+static int __udp4_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,
+ int proto)
{
struct sock *sk = NULL;
struct udphdr *uh;
@@ -2930,7 +2933,7 @@ int udp_rcv(struct sk_buff *skb)
return __udp4_lib_rcv(skb, dev_net(skb->dev)->ipv4.udp_table, IPPROTO_UDP);
}
-void udp_destroy_sock(struct sock *sk)
+static void udp_destroy_sock(struct sock *sk)
{
struct udp_sock *up = udp_sk(sk);
bool slow = lock_sock_fast(sk);
@@ -3120,8 +3123,8 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname,
}
EXPORT_IPV6_MOD(udp_lib_setsockopt);
-int udp_setsockopt(struct sock *sk, int level, int optname, sockptr_t optval,
- unsigned int optlen)
+static int udp_setsockopt(struct sock *sk, int level, int optname, sockptr_t optval,
+ unsigned int optlen)
{
if (level == SOL_UDP || level == SOL_UDPLITE || level == SOL_SOCKET)
return udp_lib_setsockopt(sk, level, optname,
@@ -3191,8 +3194,8 @@ int udp_lib_getsockopt(struct sock *sk, int level, int optname,
}
EXPORT_IPV6_MOD(udp_lib_getsockopt);
-int udp_getsockopt(struct sock *sk, int level, int optname,
- char __user *optval, int __user *optlen)
+static int udp_getsockopt(struct sock *sk, int level, int optname,
+ char __user *optval, int __user *optlen)
{
if (level == SOL_UDP || level == SOL_UDPLITE)
return udp_lib_getsockopt(sk, level, optname, optval, optlen);
@@ -3742,7 +3745,7 @@ static unsigned short seq_file_family(const struct seq_file *seq)
return afinfo->family;
}
-const struct seq_operations udp_seq_ops = {
+static const struct seq_operations udp_seq_ops = {
.start = udp_seq_start,
.next = udp_seq_next,
.stop = udp_seq_stop,
@@ -3801,7 +3804,7 @@ static int __init set_uhash_entries(char *str)
}
__setup("uhash_entries=", set_uhash_entries);
-void __init udp_table_init(struct udp_table *table, const char *name)
+static void __init udp_table_init(struct udp_table *table, const char *name)
{
unsigned int i, slot_size;
diff --git a/net/ipv4/udp_bpf.c b/net/ipv4/udp_bpf.c
index 912f0bfef4af..fc0e4f2e1085 100644
--- a/net/ipv4/udp_bpf.c
+++ b/net/ipv4/udp_bpf.c
@@ -7,8 +7,6 @@
#include <net/inet_common.h>
#include <asm/ioctls.h>
-#include "udp_impl.h"
-
static struct proto *udpv6_prot_saved __read_mostly;
static int sk_udp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
diff --git a/net/ipv4/udp_diag.c b/net/ipv4/udp_diag.c
index 6e491c720c90..a010d05062a0 100644
--- a/net/ipv4/udp_diag.c
+++ b/net/ipv4/udp_diag.c
@@ -10,7 +10,6 @@
#include <linux/inet_diag.h>
#include <linux/udp.h>
#include <net/udp.h>
-#include <net/udplite.h>
#include <linux/sock_diag.h>
static int sk_diag_dump(struct sock *sk, struct sk_buff *skb,
@@ -224,12 +223,6 @@ static int udp_diag_destroy(struct sk_buff *in_skb,
return __udp_diag_destroy(in_skb, req, sock_net(in_skb->sk)->ipv4.udp_table);
}
-static int udplite_diag_destroy(struct sk_buff *in_skb,
- const struct inet_diag_req_v2 *req)
-{
- return __udp_diag_destroy(in_skb, req, &udplite_table);
-}
-
#endif
static const struct inet_diag_handler udp_diag_handler = {
@@ -244,50 +237,13 @@ static const struct inet_diag_handler udp_diag_handler = {
#endif
};
-static void udplite_diag_dump(struct sk_buff *skb, struct netlink_callback *cb,
- const struct inet_diag_req_v2 *r)
-{
- udp_dump(&udplite_table, skb, cb, r);
-}
-
-static int udplite_diag_dump_one(struct netlink_callback *cb,
- const struct inet_diag_req_v2 *req)
-{
- return udp_dump_one(&udplite_table, cb, req);
-}
-
-static const struct inet_diag_handler udplite_diag_handler = {
- .owner = THIS_MODULE,
- .dump = udplite_diag_dump,
- .dump_one = udplite_diag_dump_one,
- .idiag_get_info = udp_diag_get_info,
- .idiag_type = IPPROTO_UDPLITE,
- .idiag_info_size = 0,
-#ifdef CONFIG_INET_DIAG_DESTROY
- .destroy = udplite_diag_destroy,
-#endif
-};
-
static int __init udp_diag_init(void)
{
- int err;
-
- err = inet_diag_register(&udp_diag_handler);
- if (err)
- goto out;
- err = inet_diag_register(&udplite_diag_handler);
- if (err)
- goto out_lite;
-out:
- return err;
-out_lite:
- inet_diag_unregister(&udp_diag_handler);
- goto out;
+ return inet_diag_register(&udp_diag_handler);
}
static void __exit udp_diag_exit(void)
{
- inet_diag_unregister(&udplite_diag_handler);
inet_diag_unregister(&udp_diag_handler);
}
@@ -296,4 +252,3 @@ module_exit(udp_diag_exit);
MODULE_LICENSE("GPL");
MODULE_DESCRIPTION("UDP socket monitoring via SOCK_DIAG");
MODULE_ALIAS_NET_PF_PROTO_TYPE(PF_NETLINK, NETLINK_SOCK_DIAG, 2-17 /* AF_INET - IPPROTO_UDP */);
-MODULE_ALIAS_NET_PF_PROTO_TYPE(PF_NETLINK, NETLINK_SOCK_DIAG, 2-136 /* AF_INET - IPPROTO_UDPLITE */);
diff --git a/net/ipv4/udp_impl.h b/net/ipv4/udp_impl.h
deleted file mode 100644
index 0ca4384f9afa..000000000000
--- a/net/ipv4/udp_impl.h
+++ /dev/null
@@ -1,27 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef _UDP4_IMPL_H
-#define _UDP4_IMPL_H
-#include <net/aligned_data.h>
-#include <net/udp.h>
-#include <net/udplite.h>
-#include <net/protocol.h>
-#include <net/inet_common.h>
-
-int __udp4_lib_rcv(struct sk_buff *, struct udp_table *, int);
-int __udp4_lib_err(struct sk_buff *, u32, struct udp_table *);
-
-int udp_v4_get_port(struct sock *sk, unsigned short snum);
-void udp_v4_rehash(struct sock *sk);
-
-int udp_setsockopt(struct sock *sk, int level, int optname, sockptr_t optval,
- unsigned int optlen);
-int udp_getsockopt(struct sock *sk, int level, int optname,
- char __user *optval, int __user *optlen);
-
-int udp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int flags);
-void udp_destroy_sock(struct sock *sk);
-
-#ifdef CONFIG_PROC_FS
-extern const struct seq_operations udp_seq_ops;
-#endif
-#endif /* _UDP4_IMPL_H */
diff --git a/net/ipv4/udplite.c b/net/ipv4/udplite.c
deleted file mode 100644
index 826e9e79eb19..000000000000
--- a/net/ipv4/udplite.c
+++ /dev/null
@@ -1,135 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0-or-later
-/*
- * UDPLITE An implementation of the UDP-Lite protocol (RFC 3828).
- *
- * Authors: Gerrit Renker <gerrit@erg.abdn.ac.uk>
- *
- * Changes:
- * Fixes:
- */
-
-#define pr_fmt(fmt) "UDPLite: " fmt
-
-#include <linux/export.h>
-#include <linux/proc_fs.h>
-#include "udp_impl.h"
-
-struct udp_table udplite_table __read_mostly;
-EXPORT_SYMBOL(udplite_table);
-
-/* Designate sk as UDP-Lite socket */
-static int udplite_sk_init(struct sock *sk)
-{
- pr_warn_once("UDP-Lite is deprecated and scheduled to be removed in 2025, "
- "please contact the netdev mailing list\n");
- return udp_init_sock(sk);
-}
-
-static int udplite_rcv(struct sk_buff *skb)
-{
- return __udp4_lib_rcv(skb, &udplite_table, IPPROTO_UDPLITE);
-}
-
-static int udplite_err(struct sk_buff *skb, u32 info)
-{
- return __udp4_lib_err(skb, info, &udplite_table);
-}
-
-static const struct net_protocol udplite_protocol = {
- .handler = udplite_rcv,
- .err_handler = udplite_err,
- .no_policy = 1,
-};
-
-struct proto udplite_prot = {
- .name = "UDP-Lite",
- .owner = THIS_MODULE,
- .close = udp_lib_close,
- .connect = ip4_datagram_connect,
- .disconnect = udp_disconnect,
- .ioctl = udp_ioctl,
- .init = udplite_sk_init,
- .destroy = udp_destroy_sock,
- .setsockopt = udp_setsockopt,
- .getsockopt = udp_getsockopt,
- .sendmsg = udp_sendmsg,
- .recvmsg = udp_recvmsg,
- .hash = udp_lib_hash,
- .unhash = udp_lib_unhash,
- .rehash = udp_v4_rehash,
- .get_port = udp_v4_get_port,
-
- .memory_allocated = &net_aligned_data.udp_memory_allocated,
- .per_cpu_fw_alloc = &udp_memory_per_cpu_fw_alloc,
-
- .sysctl_mem = sysctl_udp_mem,
- .sysctl_wmem_offset = offsetof(struct net, ipv4.sysctl_udp_wmem_min),
- .sysctl_rmem_offset = offsetof(struct net, ipv4.sysctl_udp_rmem_min),
- .obj_size = sizeof(struct udp_sock),
- .h.udp_table = &udplite_table,
-};
-EXPORT_SYMBOL(udplite_prot);
-
-static struct inet_protosw udplite4_protosw = {
- .type = SOCK_DGRAM,
- .protocol = IPPROTO_UDPLITE,
- .prot = &udplite_prot,
- .ops = &inet_dgram_ops,
- .flags = INET_PROTOSW_PERMANENT,
-};
-
-#ifdef CONFIG_PROC_FS
-static struct udp_seq_afinfo udplite4_seq_afinfo = {
- .family = AF_INET,
- .udp_table = &udplite_table,
-};
-
-static int __net_init udplite4_proc_init_net(struct net *net)
-{
- if (!proc_create_net_data("udplite", 0444, net->proc_net, &udp_seq_ops,
- sizeof(struct udp_iter_state), &udplite4_seq_afinfo))
- return -ENOMEM;
- return 0;
-}
-
-static void __net_exit udplite4_proc_exit_net(struct net *net)
-{
- remove_proc_entry("udplite", net->proc_net);
-}
-
-static struct pernet_operations udplite4_net_ops = {
- .init = udplite4_proc_init_net,
- .exit = udplite4_proc_exit_net,
-};
-
-static __init int udplite4_proc_init(void)
-{
- return register_pernet_subsys(&udplite4_net_ops);
-}
-#else
-static inline int udplite4_proc_init(void)
-{
- return 0;
-}
-#endif
-
-void __init udplite4_register(void)
-{
- udp_table_init(&udplite_table, "UDP-Lite");
- if (proto_register(&udplite_prot, 1))
- goto out_register_err;
-
- if (inet_add_protocol(&udplite_protocol, IPPROTO_UDPLITE) < 0)
- goto out_unregister_proto;
-
- inet_register_protosw(&udplite4_protosw);
-
- if (udplite4_proc_init())
- pr_err("%s: Cannot register /proc!\n", __func__);
- return;
-
-out_unregister_proto:
- proto_unregister(&udplite_prot);
-out_register_err:
- pr_crit("%s: Cannot add UDP-Lite protocol\n", __func__);
-}
--
2.53.0.473.g4a7958ca14-goog
^ permalink raw reply related [flat|nested] 19+ messages in thread
* [PATCH v2 net-next 05/15] udp: Remove UDP-Lite SNMP stats.
2026-03-05 21:49 [PATCH v2 net-next 00/15] udp: Retire UDP-Lite Kuniyuki Iwashima
` (3 preceding siblings ...)
2026-03-05 21:49 ` [PATCH v2 net-next 04/15] ipv4: Retire UDP-Lite Kuniyuki Iwashima
@ 2026-03-05 21:49 ` Kuniyuki Iwashima
2026-03-05 21:49 ` [PATCH v2 net-next 06/15] smack: Remove IPPROTO_UDPLITE support in security_sock_rcv_skb() Kuniyuki Iwashima
` (9 subsequent siblings)
14 siblings, 0 replies; 19+ messages in thread
From: Kuniyuki Iwashima @ 2026-03-05 21:49 UTC (permalink / raw)
To: Willem de Bruijn, David Ahern, David S . Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni
Cc: Simon Horman, Florian Westphal, Kuniyuki Iwashima,
Kuniyuki Iwashima, netdev, Willem de Bruijn
Since UDP and UDP-Lite shared most of the code, we have had
to check the protocol every time we increment SNMP stats.
Now that the UDP-Lite paths are dead, let's remove UDP-Lite
SNMP stats.
Signed-off-by: Kuniyuki Iwashima <kuniyu@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
---
include/net/netns/mib.h | 5 ---
include/net/udp.h | 46 ++++++++++---------------
net/ipv4/af_inet.c | 6 ----
net/ipv4/proc.c | 13 -------
net/ipv4/udp.c | 75 ++++++++++++++++++-----------------------
net/ipv6/af_inet6.c | 9 ++---
net/ipv6/proc.c | 14 --------
net/ipv6/udp.c | 48 +++++++++++---------------
8 files changed, 71 insertions(+), 145 deletions(-)
diff --git a/include/net/netns/mib.h b/include/net/netns/mib.h
index 7e373664b1e7..dce05f8e6a33 100644
--- a/include/net/netns/mib.h
+++ b/include/net/netns/mib.h
@@ -28,11 +28,6 @@ struct netns_mib {
DEFINE_SNMP_STAT(struct mptcp_mib, mptcp_statistics);
#endif
- DEFINE_SNMP_STAT(struct udp_mib, udplite_statistics);
-#if IS_ENABLED(CONFIG_IPV6)
- DEFINE_SNMP_STAT(struct udp_mib, udplite_stats_in6);
-#endif
-
DEFINE_SNMP_STAT(struct icmp_mib, icmp_statistics);
DEFINE_SNMP_STAT_ATOMIC(struct icmpmsg_mib, icmpmsg_statistics);
#if IS_ENABLED(CONFIG_IPV6)
diff --git a/include/net/udp.h b/include/net/udp.h
index 39223e2692e9..264c10607d2e 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -529,38 +529,28 @@ static inline int copy_linear_skb(struct sk_buff *skb, int len, int off,
}
/*
- * SNMP statistics for UDP and UDP-Lite
+ * SNMP statistics for UDP
*/
-#define UDP_INC_STATS(net, field, is_udplite) do { \
- if (unlikely(is_udplite)) SNMP_INC_STATS((net)->mib.udplite_statistics, field); \
- else SNMP_INC_STATS((net)->mib.udp_statistics, field); } while(0)
-#define __UDP_INC_STATS(net, field, is_udplite) do { \
- if (unlikely(is_udplite)) __SNMP_INC_STATS((net)->mib.udplite_statistics, field); \
- else __SNMP_INC_STATS((net)->mib.udp_statistics, field); } while(0)
-
-#define __UDP6_INC_STATS(net, field, is_udplite) do { \
- if (unlikely(is_udplite)) __SNMP_INC_STATS((net)->mib.udplite_stats_in6, field); \
- else __SNMP_INC_STATS((net)->mib.udp_stats_in6, field); \
-} while(0)
-#define UDP6_INC_STATS(net, field, __lite) do { \
- if (unlikely(__lite)) SNMP_INC_STATS((net)->mib.udplite_stats_in6, field); \
- else SNMP_INC_STATS((net)->mib.udp_stats_in6, field); \
-} while(0)
+#define __UDP_INC_STATS(net, field) \
+ __SNMP_INC_STATS((net)->mib.udp_statistics, field)
+#define UDP_INC_STATS(net, field) \
+ SNMP_INC_STATS((net)->mib.udp_statistics, field)
+#define __UDP6_INC_STATS(net, field) \
+ __SNMP_INC_STATS((net)->mib.udp_stats_in6, field)
+#define UDP6_INC_STATS(net, field) \
+ SNMP_INC_STATS((net)->mib.udp_stats_in6, field)
#if IS_ENABLED(CONFIG_IPV6)
-#define __UDPX_MIB(sk, ipv4) \
-({ \
- ipv4 ? (IS_UDPLITE(sk) ? sock_net(sk)->mib.udplite_statistics : \
- sock_net(sk)->mib.udp_statistics) : \
- (IS_UDPLITE(sk) ? sock_net(sk)->mib.udplite_stats_in6 : \
- sock_net(sk)->mib.udp_stats_in6); \
-})
+#define __UDPX_MIB(sk, ipv4) \
+ ({ \
+ ipv4 ? sock_net(sk)->mib.udp_statistics : \
+ sock_net(sk)->mib.udp_stats_in6; \
+ })
#else
-#define __UDPX_MIB(sk, ipv4) \
-({ \
- IS_UDPLITE(sk) ? sock_net(sk)->mib.udplite_statistics : \
- sock_net(sk)->mib.udp_statistics; \
-})
+#define __UDPX_MIB(sk, ipv4) \
+ ({ \
+ sock_net(sk)->mib.udp_statistics; \
+ })
#endif
#define __UDPX_INC_STATS(sk, field) \
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 429d7150a19b..68a4d2296ab4 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1723,9 +1723,6 @@ static __net_init int ipv4_mib_init_net(struct net *net)
net->mib.udp_statistics = alloc_percpu(struct udp_mib);
if (!net->mib.udp_statistics)
goto err_udp_mib;
- net->mib.udplite_statistics = alloc_percpu(struct udp_mib);
- if (!net->mib.udplite_statistics)
- goto err_udplite_mib;
net->mib.icmp_statistics = alloc_percpu(struct icmp_mib);
if (!net->mib.icmp_statistics)
goto err_icmp_mib;
@@ -1739,8 +1736,6 @@ static __net_init int ipv4_mib_init_net(struct net *net)
err_icmpmsg_mib:
free_percpu(net->mib.icmp_statistics);
err_icmp_mib:
- free_percpu(net->mib.udplite_statistics);
-err_udplite_mib:
free_percpu(net->mib.udp_statistics);
err_udp_mib:
free_percpu(net->mib.net_statistics);
@@ -1756,7 +1751,6 @@ static __net_exit void ipv4_mib_exit_net(struct net *net)
{
kfree(net->mib.icmpmsg_statistics);
free_percpu(net->mib.icmp_statistics);
- free_percpu(net->mib.udplite_statistics);
free_percpu(net->mib.udp_statistics);
free_percpu(net->mib.net_statistics);
free_percpu(net->mib.ip_statistics);
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index cf51f8fcf34b..bfc06d1713ec 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -444,19 +444,6 @@ static int snmp_seq_show_tcp_udp(struct seq_file *seq, void *v)
for (i = 0; i < udp_cnt; i++)
seq_printf(seq, " %lu", buff[i]);
- memset(buff, 0, udp_cnt * sizeof(unsigned long));
-
- /* the UDP and UDP-Lite MIBs are the same */
- seq_puts(seq, "\nUdpLite:");
- snmp_get_cpu_field_batch_cnt(buff, snmp4_udp_list,
- udp_cnt,
- net->mib.udplite_statistics);
- for (i = 0; i < udp_cnt; i++)
- seq_printf(seq, " %s", snmp4_udp_list[i].name);
- seq_puts(seq, "\nUdpLite:");
- for (i = 0; i < udp_cnt; i++)
- seq_printf(seq, " %lu", buff[i]);
-
seq_putc(seq, '\n');
return 0;
}
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index ff6237749eb1..38ce5d887878 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1198,13 +1198,12 @@ static int udp_send_skb(struct sk_buff *skb, struct flowi4 *fl4,
if (unlikely(err)) {
if (err == -ENOBUFS &&
!inet_test_bit(RECVERR, sk)) {
- UDP_INC_STATS(sock_net(sk),
- UDP_MIB_SNDBUFERRORS, is_udplite);
+ UDP_INC_STATS(sock_net(sk), UDP_MIB_SNDBUFERRORS);
err = 0;
}
- } else
- UDP_INC_STATS(sock_net(sk),
- UDP_MIB_OUTDATAGRAMS, is_udplite);
+ } else {
+ UDP_INC_STATS(sock_net(sk), UDP_MIB_OUTDATAGRAMS);
+ }
return err;
}
@@ -1535,10 +1534,9 @@ int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
* things). We could add another new stat but at least for now that
* seems like overkill.
*/
- if (err == -ENOBUFS || test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) {
- UDP_INC_STATS(sock_net(sk),
- UDP_MIB_SNDBUFERRORS, is_udplite);
- }
+ if (err == -ENOBUFS || test_bit(SOCK_NOSPACE, &sk->sk_socket->flags))
+ UDP_INC_STATS(sock_net(sk), UDP_MIB_SNDBUFERRORS);
+
return err;
do_confirm:
@@ -1897,10 +1895,10 @@ static struct sk_buff *__first_packet_length(struct sock *sk,
while ((skb = skb_peek(rcvq)) != NULL) {
if (udp_lib_checksum_complete(skb)) {
- __UDP_INC_STATS(sock_net(sk), UDP_MIB_CSUMERRORS,
- IS_UDPLITE(sk));
- __UDP_INC_STATS(sock_net(sk), UDP_MIB_INERRORS,
- IS_UDPLITE(sk));
+ struct net *net = sock_net(sk);
+
+ __UDP_INC_STATS(net, UDP_MIB_CSUMERRORS);
+ __UDP_INC_STATS(net, UDP_MIB_INERRORS);
udp_drops_inc(sk);
__skb_unlink(skb, rcvq);
*total += skb->truesize;
@@ -2052,11 +2050,10 @@ int udp_read_skb(struct sock *sk, skb_read_actor_t recv_actor)
return err;
if (udp_lib_checksum_complete(skb)) {
- int is_udplite = IS_UDPLITE(sk);
struct net *net = sock_net(sk);
- __UDP_INC_STATS(net, UDP_MIB_CSUMERRORS, is_udplite);
- __UDP_INC_STATS(net, UDP_MIB_INERRORS, is_udplite);
+ __UDP_INC_STATS(net, UDP_MIB_CSUMERRORS);
+ __UDP_INC_STATS(net, UDP_MIB_INERRORS);
udp_drops_inc(sk);
kfree_skb_reason(skb, SKB_DROP_REASON_UDP_CSUM);
goto try_again;
@@ -2081,6 +2078,7 @@ int udp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int flags)
unsigned int ulen, copied;
int off, err, peeking = flags & MSG_PEEK;
int is_udplite = IS_UDPLITE(sk);
+ struct net *net = sock_net(sk);
bool checksum_valid = false;
if (flags & MSG_ERRQUEUE)
@@ -2128,16 +2126,14 @@ int udp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int flags)
if (unlikely(err)) {
if (!peeking) {
udp_drops_inc(sk);
- UDP_INC_STATS(sock_net(sk),
- UDP_MIB_INERRORS, is_udplite);
+ UDP_INC_STATS(net, UDP_MIB_INERRORS);
}
kfree_skb(skb);
return err;
}
if (!peeking)
- UDP_INC_STATS(sock_net(sk),
- UDP_MIB_INDATAGRAMS, is_udplite);
+ UDP_INC_STATS(net, UDP_MIB_INDATAGRAMS);
sock_recv_cmsgs(msg, sk, skb);
@@ -2170,8 +2166,8 @@ int udp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int flags)
csum_copy_err:
if (!__sk_queue_drop_skb(sk, &udp_sk(sk)->reader_queue, skb, flags,
udp_skb_destructor)) {
- UDP_INC_STATS(sock_net(sk), UDP_MIB_CSUMERRORS, is_udplite);
- UDP_INC_STATS(sock_net(sk), UDP_MIB_INERRORS, is_udplite);
+ UDP_INC_STATS(net, UDP_MIB_CSUMERRORS);
+ UDP_INC_STATS(net, UDP_MIB_INERRORS);
}
kfree_skb_reason(skb, SKB_DROP_REASON_UDP_CSUM);
@@ -2366,20 +2362,18 @@ static int __udp_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
rc = __udp_enqueue_schedule_skb(sk, skb);
if (rc < 0) {
- int is_udplite = IS_UDPLITE(sk);
+ struct net *net = sock_net(sk);
int drop_reason;
/* Note that an ENOMEM error is charged twice */
if (rc == -ENOMEM) {
- UDP_INC_STATS(sock_net(sk), UDP_MIB_RCVBUFERRORS,
- is_udplite);
+ UDP_INC_STATS(net, UDP_MIB_RCVBUFERRORS);
drop_reason = SKB_DROP_REASON_SOCKET_RCVBUFF;
} else {
- UDP_INC_STATS(sock_net(sk), UDP_MIB_MEMERRORS,
- is_udplite);
+ UDP_INC_STATS(net, UDP_MIB_MEMERRORS);
drop_reason = SKB_DROP_REASON_PROTO_MEM;
}
- UDP_INC_STATS(sock_net(sk), UDP_MIB_INERRORS, is_udplite);
+ UDP_INC_STATS(net, UDP_MIB_INERRORS);
trace_udp_fail_queue_rcv_skb(rc, sk, skb);
sk_skb_reason_drop(sk, skb, drop_reason);
return -1;
@@ -2400,7 +2394,7 @@ static int udp_queue_rcv_one_skb(struct sock *sk, struct sk_buff *skb)
{
enum skb_drop_reason drop_reason = SKB_DROP_REASON_NOT_SPECIFIED;
struct udp_sock *up = udp_sk(sk);
- int is_udplite = IS_UDPLITE(sk);
+ struct net *net = sock_net(sk);
/*
* Charge it to the socket, dropping if the queue is full.
@@ -2437,9 +2431,7 @@ static int udp_queue_rcv_one_skb(struct sock *sk, struct sk_buff *skb)
ret = encap_rcv(sk, skb);
if (ret <= 0) {
- __UDP_INC_STATS(sock_net(sk),
- UDP_MIB_INDATAGRAMS,
- is_udplite);
+ __UDP_INC_STATS(net, UDP_MIB_INDATAGRAMS);
return -ret;
}
}
@@ -2498,9 +2490,9 @@ static int udp_queue_rcv_one_skb(struct sock *sk, struct sk_buff *skb)
csum_error:
drop_reason = SKB_DROP_REASON_UDP_CSUM;
- __UDP_INC_STATS(sock_net(sk), UDP_MIB_CSUMERRORS, is_udplite);
+ __UDP_INC_STATS(net, UDP_MIB_CSUMERRORS);
drop:
- __UDP_INC_STATS(sock_net(sk), UDP_MIB_INERRORS, is_udplite);
+ __UDP_INC_STATS(net, UDP_MIB_INERRORS);
udp_drops_inc(sk);
sk_skb_reason_drop(sk, skb, drop_reason);
return -1;
@@ -2587,10 +2579,8 @@ static int __udp4_lib_mcast_deliver(struct net *net, struct sk_buff *skb,
if (unlikely(!nskb)) {
udp_drops_inc(sk);
- __UDP_INC_STATS(net, UDP_MIB_RCVBUFERRORS,
- IS_UDPLITE(sk));
- __UDP_INC_STATS(net, UDP_MIB_INERRORS,
- IS_UDPLITE(sk));
+ __UDP_INC_STATS(net, UDP_MIB_RCVBUFERRORS);
+ __UDP_INC_STATS(net, UDP_MIB_INERRORS);
continue;
}
if (udp_queue_rcv_skb(sk, nskb) > 0)
@@ -2608,8 +2598,7 @@ static int __udp4_lib_mcast_deliver(struct net *net, struct sk_buff *skb,
consume_skb(skb);
} else {
kfree_skb(skb);
- __UDP_INC_STATS(net, UDP_MIB_IGNOREDMULTI,
- proto == IPPROTO_UDPLITE);
+ __UDP_INC_STATS(net, UDP_MIB_IGNOREDMULTI);
}
return 0;
}
@@ -2759,7 +2748,7 @@ static int __udp4_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,
goto csum_error;
drop_reason = SKB_DROP_REASON_NO_SOCKET;
- __UDP_INC_STATS(net, UDP_MIB_NOPORTS, proto == IPPROTO_UDPLITE);
+ __UDP_INC_STATS(net, UDP_MIB_NOPORTS);
icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0);
/*
@@ -2788,9 +2777,9 @@ static int __udp4_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,
proto == IPPROTO_UDPLITE ? "Lite" : "",
&saddr, ntohs(uh->source), &daddr, ntohs(uh->dest),
ulen);
- __UDP_INC_STATS(net, UDP_MIB_CSUMERRORS, proto == IPPROTO_UDPLITE);
+ __UDP_INC_STATS(net, UDP_MIB_CSUMERRORS);
drop:
- __UDP_INC_STATS(net, UDP_MIB_INERRORS, proto == IPPROTO_UDPLITE);
+ __UDP_INC_STATS(net, UDP_MIB_INERRORS);
sk_skb_reason_drop(sk, skb, drop_reason);
return 0;
}
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 9ad907a5a093..97d0bbce1720 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -894,9 +894,7 @@ static int __net_init ipv6_init_mibs(struct net *net)
net->mib.udp_stats_in6 = alloc_percpu(struct udp_mib);
if (!net->mib.udp_stats_in6)
return -ENOMEM;
- net->mib.udplite_stats_in6 = alloc_percpu(struct udp_mib);
- if (!net->mib.udplite_stats_in6)
- goto err_udplite_mib;
+
net->mib.ipv6_statistics = alloc_percpu(struct ipstats_mib);
if (!net->mib.ipv6_statistics)
goto err_ip_mib;
@@ -907,10 +905,10 @@ static int __net_init ipv6_init_mibs(struct net *net)
u64_stats_init(&af_inet6_stats->syncp);
}
-
net->mib.icmpv6_statistics = alloc_percpu(struct icmpv6_mib);
if (!net->mib.icmpv6_statistics)
goto err_icmp_mib;
+
net->mib.icmpv6msg_statistics = kzalloc_obj(struct icmpv6msg_mib);
if (!net->mib.icmpv6msg_statistics)
goto err_icmpmsg_mib;
@@ -921,8 +919,6 @@ static int __net_init ipv6_init_mibs(struct net *net)
err_icmp_mib:
free_percpu(net->mib.ipv6_statistics);
err_ip_mib:
- free_percpu(net->mib.udplite_stats_in6);
-err_udplite_mib:
free_percpu(net->mib.udp_stats_in6);
return -ENOMEM;
}
@@ -930,7 +926,6 @@ static int __net_init ipv6_init_mibs(struct net *net)
static void ipv6_cleanup_mibs(struct net *net)
{
free_percpu(net->mib.udp_stats_in6);
- free_percpu(net->mib.udplite_stats_in6);
free_percpu(net->mib.ipv6_statistics);
free_percpu(net->mib.icmpv6_statistics);
kfree(net->mib.icmpv6msg_statistics);
diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c
index 21bfc73152f0..813013ca4e75 100644
--- a/net/ipv6/proc.c
+++ b/net/ipv6/proc.c
@@ -108,17 +108,6 @@ static const struct snmp_mib snmp6_udp6_list[] = {
SNMP_MIB_ITEM("Udp6MemErrors", UDP_MIB_MEMERRORS),
};
-static const struct snmp_mib snmp6_udplite6_list[] = {
- SNMP_MIB_ITEM("UdpLite6InDatagrams", UDP_MIB_INDATAGRAMS),
- SNMP_MIB_ITEM("UdpLite6NoPorts", UDP_MIB_NOPORTS),
- SNMP_MIB_ITEM("UdpLite6InErrors", UDP_MIB_INERRORS),
- SNMP_MIB_ITEM("UdpLite6OutDatagrams", UDP_MIB_OUTDATAGRAMS),
- SNMP_MIB_ITEM("UdpLite6RcvbufErrors", UDP_MIB_RCVBUFERRORS),
- SNMP_MIB_ITEM("UdpLite6SndbufErrors", UDP_MIB_SNDBUFERRORS),
- SNMP_MIB_ITEM("UdpLite6InCsumErrors", UDP_MIB_CSUMERRORS),
- SNMP_MIB_ITEM("UdpLite6MemErrors", UDP_MIB_MEMERRORS),
-};
-
static void snmp6_seq_show_icmpv6msg(struct seq_file *seq, atomic_long_t *smib)
{
char name[32];
@@ -226,9 +215,6 @@ static int snmp6_seq_show(struct seq_file *seq, void *v)
snmp6_seq_show_item(seq, net->mib.udp_stats_in6,
NULL, snmp6_udp6_list,
ARRAY_SIZE(snmp6_udp6_list));
- snmp6_seq_show_item(seq, net->mib.udplite_stats_in6,
- NULL, snmp6_udplite6_list,
- ARRAY_SIZE(snmp6_udplite6_list));
return 0;
}
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index aa859bb0527d..07308b7156a6 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -796,20 +796,18 @@ static int __udpv6_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
rc = __udp_enqueue_schedule_skb(sk, skb);
if (rc < 0) {
- int is_udplite = IS_UDPLITE(sk);
enum skb_drop_reason drop_reason;
+ struct net *net = sock_net(sk);
/* Note that an ENOMEM error is charged twice */
if (rc == -ENOMEM) {
- UDP6_INC_STATS(sock_net(sk),
- UDP_MIB_RCVBUFERRORS, is_udplite);
+ UDP6_INC_STATS(net, UDP_MIB_RCVBUFERRORS);
drop_reason = SKB_DROP_REASON_SOCKET_RCVBUFF;
} else {
- UDP6_INC_STATS(sock_net(sk),
- UDP_MIB_MEMERRORS, is_udplite);
+ UDP6_INC_STATS(net, UDP_MIB_MEMERRORS);
drop_reason = SKB_DROP_REASON_PROTO_MEM;
}
- UDP6_INC_STATS(sock_net(sk), UDP_MIB_INERRORS, is_udplite);
+ UDP6_INC_STATS(net, UDP_MIB_INERRORS);
trace_udp_fail_queue_rcv_skb(rc, sk, skb);
sk_skb_reason_drop(sk, skb, drop_reason);
return -1;
@@ -830,7 +828,7 @@ static int udpv6_queue_rcv_one_skb(struct sock *sk, struct sk_buff *skb)
{
enum skb_drop_reason drop_reason = SKB_DROP_REASON_NOT_SPECIFIED;
struct udp_sock *up = udp_sk(sk);
- int is_udplite = IS_UDPLITE(sk);
+ struct net *net = sock_net(sk);
if (!xfrm6_policy_check(sk, XFRM_POLICY_IN, skb)) {
drop_reason = SKB_DROP_REASON_XFRM_POLICY;
@@ -864,9 +862,7 @@ static int udpv6_queue_rcv_one_skb(struct sock *sk, struct sk_buff *skb)
ret = encap_rcv(sk, skb);
if (ret <= 0) {
- __UDP6_INC_STATS(sock_net(sk),
- UDP_MIB_INDATAGRAMS,
- is_udplite);
+ __UDP6_INC_STATS(net, UDP_MIB_INDATAGRAMS);
return -ret;
}
}
@@ -909,9 +905,9 @@ static int udpv6_queue_rcv_one_skb(struct sock *sk, struct sk_buff *skb)
csum_error:
drop_reason = SKB_DROP_REASON_UDP_CSUM;
- __UDP6_INC_STATS(sock_net(sk), UDP_MIB_CSUMERRORS, is_udplite);
+ __UDP6_INC_STATS(net, UDP_MIB_CSUMERRORS);
drop:
- __UDP6_INC_STATS(sock_net(sk), UDP_MIB_INERRORS, is_udplite);
+ __UDP6_INC_STATS(net, UDP_MIB_INERRORS);
udp_drops_inc(sk);
sk_skb_reason_drop(sk, skb, drop_reason);
return -1;
@@ -1018,10 +1014,8 @@ static int __udp6_lib_mcast_deliver(struct net *net, struct sk_buff *skb,
nskb = skb_clone(skb, GFP_ATOMIC);
if (unlikely(!nskb)) {
udp_drops_inc(sk);
- __UDP6_INC_STATS(net, UDP_MIB_RCVBUFERRORS,
- IS_UDPLITE(sk));
- __UDP6_INC_STATS(net, UDP_MIB_INERRORS,
- IS_UDPLITE(sk));
+ __UDP6_INC_STATS(net, UDP_MIB_RCVBUFERRORS);
+ __UDP6_INC_STATS(net, UDP_MIB_INERRORS);
continue;
}
@@ -1040,8 +1034,7 @@ static int __udp6_lib_mcast_deliver(struct net *net, struct sk_buff *skb,
consume_skb(skb);
} else {
kfree_skb(skb);
- __UDP6_INC_STATS(net, UDP_MIB_IGNOREDMULTI,
- proto == IPPROTO_UDPLITE);
+ __UDP6_INC_STATS(net, UDP_MIB_IGNOREDMULTI);
}
return 0;
}
@@ -1213,7 +1206,7 @@ static int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,
if (udp_lib_checksum_complete(skb))
goto csum_error;
- __UDP6_INC_STATS(net, UDP_MIB_NOPORTS, proto == IPPROTO_UDPLITE);
+ __UDP6_INC_STATS(net, UDP_MIB_NOPORTS);
icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, 0);
sk_skb_reason_drop(sk, skb, reason);
@@ -1234,9 +1227,9 @@ static int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,
csum_error:
if (reason == SKB_DROP_REASON_NOT_SPECIFIED)
reason = SKB_DROP_REASON_UDP_CSUM;
- __UDP6_INC_STATS(net, UDP_MIB_CSUMERRORS, proto == IPPROTO_UDPLITE);
+ __UDP6_INC_STATS(net, UDP_MIB_CSUMERRORS);
discard:
- __UDP6_INC_STATS(net, UDP_MIB_INERRORS, proto == IPPROTO_UDPLITE);
+ __UDP6_INC_STATS(net, UDP_MIB_INERRORS);
sk_skb_reason_drop(sk, skb, reason);
return 0;
}
@@ -1490,13 +1483,11 @@ static int udp_v6_send_skb(struct sk_buff *skb, struct flowi6 *fl6,
err = ip6_send_skb(skb);
if (unlikely(err)) {
if (err == -ENOBUFS && !inet6_test_bit(RECVERR6, sk)) {
- UDP6_INC_STATS(sock_net(sk),
- UDP_MIB_SNDBUFERRORS, is_udplite);
+ UDP6_INC_STATS(sock_net(sk), UDP_MIB_SNDBUFERRORS);
err = 0;
}
} else {
- UDP6_INC_STATS(sock_net(sk),
- UDP_MIB_OUTDATAGRAMS, is_udplite);
+ UDP6_INC_STATS(sock_net(sk), UDP_MIB_OUTDATAGRAMS);
}
return err;
}
@@ -1826,10 +1817,9 @@ int udpv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
* things). We could add another new stat but at least for now that
* seems like overkill.
*/
- if (err == -ENOBUFS || test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) {
- UDP6_INC_STATS(sock_net(sk),
- UDP_MIB_SNDBUFERRORS, is_udplite);
- }
+ if (err == -ENOBUFS || test_bit(SOCK_NOSPACE, &sk->sk_socket->flags))
+ UDP6_INC_STATS(sock_net(sk), UDP_MIB_SNDBUFERRORS);
+
return err;
do_confirm:
--
2.53.0.473.g4a7958ca14-goog
^ permalink raw reply related [flat|nested] 19+ messages in thread
* [PATCH v2 net-next 06/15] smack: Remove IPPROTO_UDPLITE support in security_sock_rcv_skb().
2026-03-05 21:49 [PATCH v2 net-next 00/15] udp: Retire UDP-Lite Kuniyuki Iwashima
` (4 preceding siblings ...)
2026-03-05 21:49 ` [PATCH v2 net-next 05/15] udp: Remove UDP-Lite SNMP stats Kuniyuki Iwashima
@ 2026-03-05 21:49 ` Kuniyuki Iwashima
2026-03-05 21:49 ` [PATCH v2 net-next 07/15] udp: Remove partial csum code in RX Kuniyuki Iwashima
` (8 subsequent siblings)
14 siblings, 0 replies; 19+ messages in thread
From: Kuniyuki Iwashima @ 2026-03-05 21:49 UTC (permalink / raw)
To: Willem de Bruijn, David Ahern, David S . Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni
Cc: Simon Horman, Florian Westphal, Kuniyuki Iwashima,
Kuniyuki Iwashima, netdev, Willem de Bruijn, Casey Schaufler,
Paul Moore, James Morris, Serge E. Hallyn, linux-security-module
smack_socket_sock_rcv_skb() is registered as socket_sock_rcv_skb,
which is called as security_sock_rcv_skb() in sk_filter_trim_cap().
Now that UDP-Lite is gone, let's remove the IPPROTO_UDPLITE support
in smack_socket_sock_rcv_skb().
Signed-off-by: Kuniyuki Iwashima <kuniyu@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Acked-by: Casey Schaufler <casey@schaufler-ca.com>
---
Cc: Paul Moore <paul@paul-moore.com>
Cc: James Morris <jmorris@namei.org>
Cc: "Serge E. Hallyn" <serge@hallyn.com>
Cc: linux-security-module@vger.kernel.org
---
security/smack/smack_lsm.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/security/smack/smack_lsm.c b/security/smack/smack_lsm.c
index 98af9d7b9434..e581d6465946 100644
--- a/security/smack/smack_lsm.c
+++ b/security/smack/smack_lsm.c
@@ -4176,7 +4176,6 @@ static int smk_skb_to_addr_ipv6(struct sk_buff *skb, struct sockaddr_in6 *sip)
sip->sin6_port = th->source;
break;
case IPPROTO_UDP:
- case IPPROTO_UDPLITE:
uh = skb_header_pointer(skb, offset, sizeof(_udph), &_udph);
if (uh != NULL)
sip->sin6_port = uh->source;
@@ -4301,8 +4300,7 @@ static int smack_socket_sock_rcv_skb(struct sock *sk, struct sk_buff *skb)
#if IS_ENABLED(CONFIG_IPV6)
case PF_INET6:
proto = smk_skb_to_addr_ipv6(skb, &sadd);
- if (proto != IPPROTO_UDP && proto != IPPROTO_UDPLITE &&
- proto != IPPROTO_TCP)
+ if (proto != IPPROTO_UDP && proto != IPPROTO_TCP)
break;
#ifdef SMACK_IPV6_SECMARK_LABELING
skp = smack_from_skb(skb);
--
2.53.0.473.g4a7958ca14-goog
^ permalink raw reply related [flat|nested] 19+ messages in thread
* [PATCH v2 net-next 07/15] udp: Remove partial csum code in RX.
2026-03-05 21:49 [PATCH v2 net-next 00/15] udp: Retire UDP-Lite Kuniyuki Iwashima
` (5 preceding siblings ...)
2026-03-05 21:49 ` [PATCH v2 net-next 06/15] smack: Remove IPPROTO_UDPLITE support in security_sock_rcv_skb() Kuniyuki Iwashima
@ 2026-03-05 21:49 ` Kuniyuki Iwashima
2026-03-10 10:25 ` [v2,net-next,07/15] " Paolo Abeni
2026-03-10 20:00 ` [PATCH v2 net-next 07/15] " David Laight
2026-03-05 21:49 ` [PATCH v2 net-next 08/15] udp: Remove partial csum code in TX Kuniyuki Iwashima
` (7 subsequent siblings)
14 siblings, 2 replies; 19+ messages in thread
From: Kuniyuki Iwashima @ 2026-03-05 21:49 UTC (permalink / raw)
To: Willem de Bruijn, David Ahern, David S . Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni
Cc: Simon Horman, Florian Westphal, Kuniyuki Iwashima,
Kuniyuki Iwashima, netdev, Willem de Bruijn
UDP-Lite supports the partial checksum and the coverage is
stored in the position of the length field of struct udphdr.
In RX paths, udp4_csum_init() / udp6_csum_init() save the value
in UDP_SKB_CB(skb)->cscov and set UDP_SKB_CB(skb)->partial_cov
to 1 if the coverage is not full.
The subsequent processing diverges depending on the value,
but such paths are now dead.
Also, these functions have some code guarded for UDP:
* udp_unicast_rcv_skb / udp6_unicast_rcv_skb
* __udp4_lib_rcv() and __udp6_lib_rcv().
Let's remove the partial csum code and the unnecessary
guard for UDP-Lite in RX.
Signed-off-by: Kuniyuki Iwashima <kuniyu@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
---
include/net/udp.h | 17 ++-------
include/net/udplite.h | 34 ------------------
net/ipv4/udp.c | 83 ++++++++-----------------------------------
net/ipv6/udp.c | 83 +++++++++++--------------------------------
4 files changed, 38 insertions(+), 179 deletions(-)
diff --git a/include/net/udp.h b/include/net/udp.h
index 264c10607d2e..bc275cda9f8c 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -32,11 +32,9 @@
#include <linux/math.h>
/**
- * struct udp_skb_cb - UDP(-Lite) private variables
+ * struct udp_skb_cb - UDP private variables
*
* @header: private variables used by IPv4/IPv6
- * @cscov: checksum coverage length (UDP-Lite only)
- * @partial_cov: if set indicates partial csum coverage
*/
struct udp_skb_cb {
union {
@@ -45,8 +43,6 @@ struct udp_skb_cb {
struct inet6_skb_parm h6;
#endif
} header;
- __u16 cscov;
- __u8 partial_cov;
};
#define UDP_SKB_CB(__skb) ((struct udp_skb_cb *)((__skb)->cb))
@@ -216,13 +212,11 @@ extern int sysctl_udp_wmem_min;
struct sk_buff;
/*
- * Generic checksumming routines for UDP(-Lite) v4 and v6
+ * Generic checksumming routines for UDP v4 and v6
*/
static inline __sum16 __udp_lib_checksum_complete(struct sk_buff *skb)
{
- return (UDP_SKB_CB(skb)->cscov == skb->len ?
- __skb_checksum_complete(skb) :
- __skb_checksum_complete_head(skb, UDP_SKB_CB(skb)->cscov));
+ return __skb_checksum_complete(skb);
}
static inline int udp_lib_checksum_complete(struct sk_buff *skb)
@@ -273,7 +267,6 @@ static inline void udp_csum_pull_header(struct sk_buff *skb)
skb->csum = csum_partial(skb->data, sizeof(struct udphdr),
skb->csum);
skb_pull_rcsum(skb, sizeof(struct udphdr));
- UDP_SKB_CB(skb)->cscov -= sizeof(struct udphdr);
}
typedef struct sock *(*udp_lookup_t)(const struct sk_buff *skb, __be16 sport,
@@ -641,9 +634,6 @@ static inline struct sk_buff *udp_rcv_segment(struct sock *sk,
static inline void udp_post_segment_fix_csum(struct sk_buff *skb)
{
- /* UDP-lite can't land here - no GRO */
- WARN_ON_ONCE(UDP_SKB_CB(skb)->partial_cov);
-
/* UDP packets generated with UDP_SEGMENT and traversing:
*
* UDP tunnel(xmit) -> veth (segmentation) -> veth (gro) -> UDP tunnel (rx)
@@ -657,7 +647,6 @@ static inline void udp_post_segment_fix_csum(struct sk_buff *skb)
* a valid csum after the segmentation.
* Additionally fixup the UDP CB.
*/
- UDP_SKB_CB(skb)->cscov = skb->len;
if (skb->ip_summed == CHECKSUM_NONE && !skb->csum_valid)
skb->csum_valid = 1;
}
diff --git a/include/net/udplite.h b/include/net/udplite.h
index fdd769745ac4..0456a14c993b 100644
--- a/include/net/udplite.h
+++ b/include/net/udplite.h
@@ -25,40 +25,6 @@ static __inline__ int udplite_getfrag(void *from, char *to, int offset,
/*
* Checksumming routines
*/
-static inline int udplite_checksum_init(struct sk_buff *skb, struct udphdr *uh)
-{
- u16 cscov;
-
- /* In UDPv4 a zero checksum means that the transmitter generated no
- * checksum. UDP-Lite (like IPv6) mandates checksums, hence packets
- * with a zero checksum field are illegal. */
- if (uh->check == 0) {
- net_dbg_ratelimited("UDPLite: zeroed checksum field\n");
- return 1;
- }
-
- cscov = ntohs(uh->len);
-
- if (cscov == 0) /* Indicates that full coverage is required. */
- ;
- else if (cscov < 8 || cscov > skb->len) {
- /*
- * Coverage length violates RFC 3828: log and discard silently.
- */
- net_dbg_ratelimited("UDPLite: bad csum coverage %d/%d\n",
- cscov, skb->len);
- return 1;
-
- } else if (cscov < skb->len) {
- UDP_SKB_CB(skb)->partial_cov = 1;
- UDP_SKB_CB(skb)->cscov = cscov;
- if (skb->ip_summed == CHECKSUM_COMPLETE)
- skb->ip_summed = CHECKSUM_NONE;
- skb->csum_valid = 0;
- }
-
- return 0;
-}
/* Fast-path computation of checksum. Socket may not be locked. */
static inline __wsum udplite_csum(struct sk_buff *skb)
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 38ce5d887878..51a3e3d3d4a9 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -2072,14 +2072,13 @@ EXPORT_IPV6_MOD(udp_read_skb);
INDIRECT_CALLABLE_SCOPE
int udp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int flags)
{
- struct inet_sock *inet = inet_sk(sk);
DECLARE_SOCKADDR(struct sockaddr_in *, sin, msg->msg_name);
- struct sk_buff *skb;
- unsigned int ulen, copied;
int off, err, peeking = flags & MSG_PEEK;
- int is_udplite = IS_UDPLITE(sk);
+ struct inet_sock *inet = inet_sk(sk);
struct net *net = sock_net(sk);
bool checksum_valid = false;
+ unsigned int ulen, copied;
+ struct sk_buff *skb;
if (flags & MSG_ERRQUEUE)
return ip_recv_error(sk, msg, len);
@@ -2097,14 +2096,10 @@ int udp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int flags)
else if (copied < ulen)
msg->msg_flags |= MSG_TRUNC;
- /*
- * If checksum is needed at all, try to do it while copying the
- * data. If the data is truncated, or if we only want a partial
- * coverage checksum (UDP-Lite), do it before the copy.
+ /* If checksum is needed at all, try to do it while copying the
+ * data. If the data is truncated, do it before the copy.
*/
-
- if (copied < ulen || peeking ||
- (is_udplite && UDP_SKB_CB(skb)->partial_cov)) {
+ if (copied < ulen || peeking) {
checksum_valid = udp_skb_csum_unnecessary(skb) ||
!__udp_lib_checksum_complete(skb);
if (!checksum_valid)
@@ -2439,42 +2434,6 @@ static int udp_queue_rcv_one_skb(struct sock *sk, struct sk_buff *skb)
/* FALLTHROUGH -- it's a UDP Packet */
}
- /*
- * UDP-Lite specific tests, ignored on UDP sockets
- */
- if (unlikely(udp_test_bit(UDPLITE_RECV_CC, sk) &&
- UDP_SKB_CB(skb)->partial_cov)) {
- u16 pcrlen = READ_ONCE(up->pcrlen);
-
- /*
- * MIB statistics other than incrementing the error count are
- * disabled for the following two types of errors: these depend
- * on the application settings, not on the functioning of the
- * protocol stack as such.
- *
- * RFC 3828 here recommends (sec 3.3): "There should also be a
- * way ... to ... at least let the receiving application block
- * delivery of packets with coverage values less than a value
- * provided by the application."
- */
- if (pcrlen == 0) { /* full coverage was set */
- net_dbg_ratelimited("UDPLite: partial coverage %d while full coverage %d requested\n",
- UDP_SKB_CB(skb)->cscov, skb->len);
- goto drop;
- }
- /* The next case involves violating the min. coverage requested
- * by the receiver. This is subtle: if receiver wants x and x is
- * greater than the buffersize/MTU then receiver will complain
- * that it wants x while sender emits packets of smaller size y.
- * Therefore the above ...()->partial_cov statement is essential.
- */
- if (UDP_SKB_CB(skb)->cscov < pcrlen) {
- net_dbg_ratelimited("UDPLite: coverage %d too small, need min %d\n",
- UDP_SKB_CB(skb)->cscov, pcrlen);
- goto drop;
- }
- }
-
prefetch(&sk->sk_rmem_alloc);
if (rcu_access_pointer(sk->sk_filter) &&
udp_lib_checksum_complete(skb))
@@ -2608,29 +2567,14 @@ static int __udp4_lib_mcast_deliver(struct net *net, struct sk_buff *skb,
* Otherwise, csum completion requires checksumming packet body,
* including udp header and folding it to skb->csum.
*/
-static inline int udp4_csum_init(struct sk_buff *skb, struct udphdr *uh,
- int proto)
+static inline int udp4_csum_init(struct sk_buff *skb, struct udphdr *uh)
{
int err;
- UDP_SKB_CB(skb)->partial_cov = 0;
- UDP_SKB_CB(skb)->cscov = skb->len;
-
- if (proto == IPPROTO_UDPLITE) {
- err = udplite_checksum_init(skb, uh);
- if (err)
- return err;
-
- if (UDP_SKB_CB(skb)->partial_cov) {
- skb->csum = inet_compute_pseudo(skb, proto);
- return 0;
- }
- }
-
/* Note, we are only interested in != 0 or == 0, thus the
* force to int.
*/
- err = (__force int)skb_checksum_init_zero_check(skb, proto, uh->check,
+ err = (__force int)skb_checksum_init_zero_check(skb, IPPROTO_UDP, uh->check,
inet_compute_pseudo);
if (err)
return err;
@@ -2658,7 +2602,7 @@ static int udp_unicast_rcv_skb(struct sock *sk, struct sk_buff *skb,
{
int ret;
- if (inet_get_convert_csum(sk) && uh->check && !IS_UDPLITE(sk))
+ if (inet_get_convert_csum(sk) && uh->check)
skb_checksum_try_convert(skb, IPPROTO_UDP, inet_compute_pseudo);
ret = udp_queue_rcv_skb(sk, skb);
@@ -2703,14 +2647,15 @@ static int __udp4_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,
if (ulen > skb->len)
goto short_packet;
- if (proto == IPPROTO_UDP) {
- /* UDP validates ulen. */
- if (ulen < sizeof(*uh) || pskb_trim_rcsum(skb, ulen))
+ /* UDP validates ulen. */
+ if (ulen < sizeof(*uh)) {
+ if (pskb_trim_rcsum(skb, ulen))
goto short_packet;
+
uh = udp_hdr(skb);
}
- if (udp4_csum_init(skb, uh, proto))
+ if (udp4_csum_init(skb, uh))
goto csum_error;
sk = inet_steal_sock(net, skb, sizeof(struct udphdr), saddr, uh->source, daddr, uh->dest,
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 07308b7156a6..49100b9e78cd 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -469,15 +469,13 @@ INDIRECT_CALLABLE_SCOPE
int udpv6_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
int flags)
{
+ int off, is_udp4, err, peeking = flags & MSG_PEEK;
struct ipv6_pinfo *np = inet6_sk(sk);
struct inet_sock *inet = inet_sk(sk);
- struct sk_buff *skb;
- unsigned int ulen, copied;
- int off, err, peeking = flags & MSG_PEEK;
- int is_udplite = IS_UDPLITE(sk);
struct udp_mib __percpu *mib;
bool checksum_valid = false;
- int is_udp4;
+ unsigned int ulen, copied;
+ struct sk_buff *skb;
if (flags & MSG_ERRQUEUE)
return ipv6_recv_error(sk, msg, len);
@@ -501,14 +499,10 @@ int udpv6_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
is_udp4 = (skb->protocol == htons(ETH_P_IP));
mib = __UDPX_MIB(sk, is_udp4);
- /*
- * If checksum is needed at all, try to do it while copying the
- * data. If the data is truncated, or if we only want a partial
- * coverage checksum (UDP-Lite), do it before the copy.
+ /* If checksum is needed at all, try to do it while copying the
+ * data. If the data is truncated, do it before the copy.
*/
-
- if (copied < ulen || peeking ||
- (is_udplite && UDP_SKB_CB(skb)->partial_cov)) {
+ if (copied < ulen || peeking) {
checksum_valid = udp_skb_csum_unnecessary(skb) ||
!__udp_lib_checksum_complete(skb);
if (!checksum_valid)
@@ -870,25 +864,6 @@ static int udpv6_queue_rcv_one_skb(struct sock *sk, struct sk_buff *skb)
/* FALLTHROUGH -- it's a UDP Packet */
}
- /*
- * UDP-Lite specific tests, ignored on UDP sockets (see net/ipv4/udp.c).
- */
- if (unlikely(udp_test_bit(UDPLITE_RECV_CC, sk) &&
- UDP_SKB_CB(skb)->partial_cov)) {
- u16 pcrlen = READ_ONCE(up->pcrlen);
-
- if (pcrlen == 0) { /* full coverage was set */
- net_dbg_ratelimited("UDPLITE6: partial coverage %d while full coverage %d requested\n",
- UDP_SKB_CB(skb)->cscov, skb->len);
- goto drop;
- }
- if (UDP_SKB_CB(skb)->cscov < pcrlen) {
- net_dbg_ratelimited("UDPLITE6: coverage %d too small, need min %d\n",
- UDP_SKB_CB(skb)->cscov, pcrlen);
- goto drop;
- }
- }
-
prefetch(&sk->sk_rmem_alloc);
if (rcu_access_pointer(sk->sk_filter) &&
udp_lib_checksum_complete(skb))
@@ -1053,7 +1028,7 @@ static int udp6_unicast_rcv_skb(struct sock *sk, struct sk_buff *skb,
{
int ret;
- if (inet_get_convert_csum(sk) && uh->check && !IS_UDPLITE(sk))
+ if (inet_get_convert_csum(sk) && uh->check)
skb_checksum_try_convert(skb, IPPROTO_UDP, ip6_compute_pseudo);
ret = udpv6_queue_rcv_skb(sk, skb);
@@ -1064,24 +1039,10 @@ static int udp6_unicast_rcv_skb(struct sock *sk, struct sk_buff *skb,
return 0;
}
-static int udp6_csum_init(struct sk_buff *skb, struct udphdr *uh, int proto)
+static int udp6_csum_init(struct sk_buff *skb, struct udphdr *uh)
{
int err;
- UDP_SKB_CB(skb)->partial_cov = 0;
- UDP_SKB_CB(skb)->cscov = skb->len;
-
- if (proto == IPPROTO_UDPLITE) {
- err = udplite_checksum_init(skb, uh);
- if (err)
- return err;
-
- if (UDP_SKB_CB(skb)->partial_cov) {
- skb->csum = ip6_compute_pseudo(skb, proto);
- return 0;
- }
- }
-
/* To support RFC 6936 (allow zero checksum in UDP/IPV6 for tunnels)
* we accept a checksum of zero here. When we find the socket
* for the UDP packet we'll check if that socket allows zero checksum
@@ -1090,7 +1051,7 @@ static int udp6_csum_init(struct sk_buff *skb, struct udphdr *uh, int proto)
* Note, we are only interested in != 0 or == 0, thus the
* force to int.
*/
- err = (__force int)skb_checksum_init_zero_check(skb, proto, uh->check,
+ err = (__force int)skb_checksum_init_zero_check(skb, IPPROTO_UDP, uh->check,
ip6_compute_pseudo);
if (err)
return err;
@@ -1132,26 +1093,24 @@ static int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,
if (ulen > skb->len)
goto short_packet;
- if (proto == IPPROTO_UDP) {
- /* UDP validates ulen. */
+ /* Check for jumbo payload */
+ if (ulen == 0)
+ ulen = skb->len;
- /* Check for jumbo payload */
- if (ulen == 0)
- ulen = skb->len;
+ /* UDP validates ulen. */
+ if (ulen < sizeof(*uh))
+ goto short_packet;
- if (ulen < sizeof(*uh))
+ if (ulen < skb->len) {
+ if (pskb_trim_rcsum(skb, ulen))
goto short_packet;
- if (ulen < skb->len) {
- if (pskb_trim_rcsum(skb, ulen))
- goto short_packet;
- saddr = &ipv6_hdr(skb)->saddr;
- daddr = &ipv6_hdr(skb)->daddr;
- uh = udp_hdr(skb);
- }
+ saddr = &ipv6_hdr(skb)->saddr;
+ daddr = &ipv6_hdr(skb)->daddr;
+ uh = udp_hdr(skb);
}
- if (udp6_csum_init(skb, uh, proto))
+ if (udp6_csum_init(skb, uh))
goto csum_error;
/* Check if the socket is already available, e.g. due to early demux */
--
2.53.0.473.g4a7958ca14-goog
^ permalink raw reply related [flat|nested] 19+ messages in thread
* [PATCH v2 net-next 08/15] udp: Remove partial csum code in TX.
2026-03-05 21:49 [PATCH v2 net-next 00/15] udp: Retire UDP-Lite Kuniyuki Iwashima
` (6 preceding siblings ...)
2026-03-05 21:49 ` [PATCH v2 net-next 07/15] udp: Remove partial csum code in RX Kuniyuki Iwashima
@ 2026-03-05 21:49 ` Kuniyuki Iwashima
2026-03-05 21:49 ` [PATCH v2 net-next 09/15] udp: Remove UDPLITE_SEND_CSCOV and UDPLITE_RECV_CSCOV Kuniyuki Iwashima
` (6 subsequent siblings)
14 siblings, 0 replies; 19+ messages in thread
From: Kuniyuki Iwashima @ 2026-03-05 21:49 UTC (permalink / raw)
To: Willem de Bruijn, David Ahern, David S . Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni
Cc: Simon Horman, Florian Westphal, Kuniyuki Iwashima,
Kuniyuki Iwashima, netdev, Willem de Bruijn
UDP TX paths also have some code for UDP-Lite partial
checksum:
* udplite_csum() in udp_send_skb() and udp_v6_send_skb()
* udplite_getfrag() in udp_sendmsg() and udpv6_sendmsg()
Let's remove such code.
Now, we can use IPPROTO_UDP directly instead of sk->sk_protocol
or fl6->flowi6_proto for csum_tcpudp_magic() and csum_ipv6_magic().
Signed-off-by: Kuniyuki Iwashima <kuniyu@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
---
include/net/udplite.h | 35 --------------------------
net/ipv4/udp.c | 58 ++++++++++++++++---------------------------
net/ipv6/udp.c | 54 +++++++++++++++++++---------------------
3 files changed, 47 insertions(+), 100 deletions(-)
diff --git a/include/net/udplite.h b/include/net/udplite.h
index 0456a14c993b..6bfa1d6833d1 100644
--- a/include/net/udplite.h
+++ b/include/net/udplite.h
@@ -12,39 +12,4 @@
#define UDPLITE_SEND_CSCOV 10 /* sender partial coverage (as sent) */
#define UDPLITE_RECV_CSCOV 11 /* receiver partial coverage (threshold ) */
-/*
- * Checksum computation is all in software, hence simpler getfrag.
- */
-static __inline__ int udplite_getfrag(void *from, char *to, int offset,
- int len, int odd, struct sk_buff *skb)
-{
- struct msghdr *msg = from;
- return copy_from_iter_full(to, len, &msg->msg_iter) ? 0 : -EFAULT;
-}
-
-/*
- * Checksumming routines
- */
-
-/* Fast-path computation of checksum. Socket may not be locked. */
-static inline __wsum udplite_csum(struct sk_buff *skb)
-{
- const int off = skb_transport_offset(skb);
- const struct sock *sk = skb->sk;
- int len = skb->len - off;
-
- if (udp_test_bit(UDPLITE_SEND_CC, sk)) {
- u16 pcslen = READ_ONCE(udp_sk(sk)->pcslen);
-
- if (pcslen < len) {
- if (pcslen > 0)
- len = pcslen;
- udp_hdr(skb)->len = htons(pcslen);
- }
- }
- skb->ip_summed = CHECKSUM_NONE; /* no HW support for checksumming */
-
- return skb_checksum(skb, off, len, 0);
-}
-
#endif /* _UDPLITE_H */
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 51a3e3d3d4a9..7d83e694115c 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1120,20 +1120,19 @@ static int udp_send_skb(struct sk_buff *skb, struct flowi4 *fl4,
struct inet_cork *cork)
{
struct sock *sk = skb->sk;
- struct inet_sock *inet = inet_sk(sk);
+ int offset, len, datalen;
struct udphdr *uh;
int err;
- int is_udplite = IS_UDPLITE(sk);
- int offset = skb_transport_offset(skb);
- int len = skb->len - offset;
- int datalen = len - sizeof(*uh);
- __wsum csum = 0;
+
+ offset = skb_transport_offset(skb);
+ len = skb->len - offset;
+ datalen = len - sizeof(*uh);
/*
* Create a UDP header
*/
uh = udp_hdr(skb);
- uh->source = inet->inet_sport;
+ uh->source = inet_sk(sk)->inet_sport;
uh->dest = fl4->fl4_dport;
uh->len = htons(len);
uh->check = 0;
@@ -1154,7 +1153,7 @@ static int udp_send_skb(struct sk_buff *skb, struct flowi4 *fl4,
kfree_skb(skb);
return -EINVAL;
}
- if (is_udplite || dst_xfrm(skb_dst(skb))) {
+ if (dst_xfrm(skb_dst(skb))) {
kfree_skb(skb);
return -EIO;
}
@@ -1170,26 +1169,18 @@ static int udp_send_skb(struct sk_buff *skb, struct flowi4 *fl4,
}
}
- if (is_udplite) /* UDP-Lite */
- csum = udplite_csum(skb);
-
- else if (sk->sk_no_check_tx) { /* UDP csum off */
-
+ if (sk->sk_no_check_tx) { /* UDP csum off */
skb->ip_summed = CHECKSUM_NONE;
goto send;
-
} else if (skb->ip_summed == CHECKSUM_PARTIAL) { /* UDP hardware csum */
csum_partial:
-
udp4_hwcsum(skb, fl4->saddr, fl4->daddr);
goto send;
-
- } else
- csum = udp_csum(skb);
+ }
/* add protocol-dependent pseudo-header */
uh->check = csum_tcpudp_magic(fl4->saddr, fl4->daddr, len,
- sk->sk_protocol, csum);
+ IPPROTO_UDP, udp_csum(skb));
if (uh->check == 0)
uh->check = CSUM_MANGLED_0;
@@ -1270,26 +1261,23 @@ EXPORT_IPV6_MOD_GPL(udp_cmsg_send);
int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
{
+ int corkreq = udp_test_bit(CORK, sk) || msg->msg_flags & MSG_MORE;
DEFINE_RAW_FLEX(struct ip_options_rcu, opt_copy, opt.__data,
IP_OPTIONS_DATA_FIXED_SIZE);
+ DECLARE_SOCKADDR(struct sockaddr_in *, usin, msg->msg_name);
+ int ulen = len, free = 0, connected = 0;
struct inet_sock *inet = inet_sk(sk);
struct udp_sock *up = udp_sk(sk);
- DECLARE_SOCKADDR(struct sockaddr_in *, usin, msg->msg_name);
+ __be32 daddr, faddr, saddr;
+ struct rtable *rt = NULL;
struct flowi4 fl4_stack;
- struct flowi4 *fl4;
- int ulen = len;
struct ipcm_cookie ipc;
- struct rtable *rt = NULL;
- int free = 0;
- int connected = 0;
- __be32 daddr, faddr, saddr;
- u8 scope;
- __be16 dport;
- int err, is_udplite = IS_UDPLITE(sk);
- int corkreq = udp_test_bit(CORK, sk) || msg->msg_flags & MSG_MORE;
- int (*getfrag)(void *, char *, int, int, int, struct sk_buff *);
struct sk_buff *skb;
+ struct flowi4 *fl4;
+ __be16 dport;
int uc_index;
+ u8 scope;
+ int err;
if (len > 0xFFFF)
return -EMSGSIZE;
@@ -1301,8 +1289,6 @@ int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
if (msg->msg_flags & MSG_OOB) /* Mirror BSD error message compatibility */
return -EOPNOTSUPP;
- getfrag = is_udplite ? udplite_getfrag : ip_generic_getfrag;
-
fl4 = &inet->cork.fl.u.ip4;
if (READ_ONCE(up->pending)) {
/*
@@ -1444,7 +1430,7 @@ int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
flowi4_init_output(fl4, ipc.oif, ipc.sockc.mark,
ipc.tos & INET_DSCP_MASK, scope,
- sk->sk_protocol, flow_flags, faddr, saddr,
+ IPPROTO_UDP, flow_flags, faddr, saddr,
dport, inet->inet_sport,
sk_uid(sk));
@@ -1478,7 +1464,7 @@ int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
if (!corkreq) {
struct inet_cork cork;
- skb = ip_make_skb(sk, fl4, getfrag, msg, ulen,
+ skb = ip_make_skb(sk, fl4, ip_generic_getfrag, msg, ulen,
sizeof(struct udphdr), &ipc, &rt,
&cork, msg->msg_flags);
err = PTR_ERR(skb);
@@ -1509,7 +1495,7 @@ int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
do_append_data:
up->len += ulen;
- err = ip_append_data(sk, fl4, getfrag, msg, ulen,
+ err = ip_append_data(sk, fl4, ip_generic_getfrag, msg, ulen,
sizeof(struct udphdr), &ipc, &rt,
corkreq ? msg->msg_flags|MSG_MORE : msg->msg_flags);
if (err)
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 49100b9e78cd..402b3b821480 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -1371,13 +1371,13 @@ static int udp_v6_send_skb(struct sk_buff *skb, struct flowi6 *fl6,
struct inet_cork *cork)
{
struct sock *sk = skb->sk;
+ int offset, len, datalen;
struct udphdr *uh;
int err = 0;
- int is_udplite = IS_UDPLITE(sk);
- __wsum csum = 0;
- int offset = skb_transport_offset(skb);
- int len = skb->len - offset;
- int datalen = len - sizeof(*uh);
+
+ offset = skb_transport_offset(skb);
+ len = skb->len - offset;
+ datalen = len - sizeof(*uh);
/*
* Create a UDP header
@@ -1404,7 +1404,7 @@ static int udp_v6_send_skb(struct sk_buff *skb, struct flowi6 *fl6,
kfree_skb(skb);
return -EINVAL;
}
- if (is_udplite || dst_xfrm(skb_dst(skb))) {
+ if (dst_xfrm(skb_dst(skb))) {
kfree_skb(skb);
return -EIO;
}
@@ -1420,21 +1420,18 @@ static int udp_v6_send_skb(struct sk_buff *skb, struct flowi6 *fl6,
}
}
- if (is_udplite)
- csum = udplite_csum(skb);
- else if (udp_get_no_check6_tx(sk)) { /* UDP csum disabled */
+ if (udp_get_no_check6_tx(sk)) { /* UDP csum disabled */
skb->ip_summed = CHECKSUM_NONE;
goto send;
} else if (skb->ip_summed == CHECKSUM_PARTIAL) { /* UDP hardware csum */
csum_partial:
udp6_hwcsum_outgoing(sk, skb, &fl6->saddr, &fl6->daddr, len);
goto send;
- } else
- csum = udp_csum(skb);
+ }
/* add protocol-dependent pseudo-header */
uh->check = csum_ipv6_magic(&fl6->saddr, &fl6->daddr,
- len, fl6->flowi6_proto, csum);
+ len, IPPROTO_UDP, udp_csum(skb));
if (uh->check == 0)
uh->check = CSUM_MANGLED_0;
@@ -1474,27 +1471,26 @@ static int udp_v6_push_pending_frames(struct sock *sk)
int udpv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
{
- struct ipv6_txoptions opt_space;
- struct udp_sock *up = udp_sk(sk);
- struct inet_sock *inet = inet_sk(sk);
- struct ipv6_pinfo *np = inet6_sk(sk);
+ int corkreq = udp_test_bit(CORK, sk) || msg->msg_flags & MSG_MORE;
DECLARE_SOCKADDR(struct sockaddr_in6 *, sin6, msg->msg_name);
- struct in6_addr *daddr, *final_p, final;
- struct ipv6_txoptions *opt = NULL;
struct ipv6_txoptions *opt_to_free = NULL;
+ struct in6_addr *daddr, *final_p, final;
struct ip6_flowlabel *flowlabel = NULL;
+ struct inet_sock *inet = inet_sk(sk);
+ struct ipv6_pinfo *np = inet6_sk(sk);
+ struct ipv6_txoptions *opt = NULL;
+ struct udp_sock *up = udp_sk(sk);
+ struct ipv6_txoptions opt_space;
+ int addr_len = msg->msg_namelen;
struct inet_cork_full cork;
- struct flowi6 *fl6 = &cork.fl.u.ip6;
- struct dst_entry *dst;
struct ipcm6_cookie ipc6;
- int addr_len = msg->msg_namelen;
bool connected = false;
+ struct dst_entry *dst;
+ struct flowi6 *fl6;
int ulen = len;
- int corkreq = udp_test_bit(CORK, sk) || msg->msg_flags & MSG_MORE;
int err;
- int is_udplite = IS_UDPLITE(sk);
- int (*getfrag)(void *, char *, int, int, int, struct sk_buff *);
+ fl6 = &cork.fl.u.ip6;
ipcm6_init_sk(&ipc6, sk);
ipc6.gso_size = READ_ONCE(up->gso_size);
@@ -1553,7 +1549,6 @@ int udpv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
if (len > INT_MAX - sizeof(struct udphdr))
return -EMSGSIZE;
- getfrag = is_udplite ? udplite_getfrag : ip_generic_getfrag;
if (READ_ONCE(up->pending)) {
if (READ_ONCE(up->pending) == AF_INET)
return udp_sendmsg(sk, msg, len);
@@ -1655,7 +1650,7 @@ int udpv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
opt = ipv6_fixup_options(&opt_space, opt);
ipc6.opt = opt;
- fl6->flowi6_proto = sk->sk_protocol;
+ fl6->flowi6_proto = IPPROTO_UDP;
fl6->flowi6_mark = ipc6.sockc.mark;
fl6->daddr = *daddr;
if (ipv6_addr_any(&fl6->saddr) && !ipv6_addr_any(&np->saddr))
@@ -1722,7 +1717,7 @@ int udpv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
if (!corkreq) {
struct sk_buff *skb;
- skb = ip6_make_skb(sk, getfrag, msg, ulen,
+ skb = ip6_make_skb(sk, ip_generic_getfrag, msg, ulen,
sizeof(struct udphdr), &ipc6,
dst_rt6_info(dst),
msg->msg_flags, &cork);
@@ -1748,8 +1743,9 @@ int udpv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
do_append_data:
up->len += ulen;
- err = ip6_append_data(sk, getfrag, msg, ulen, sizeof(struct udphdr),
- &ipc6, fl6, dst_rt6_info(dst),
+ err = ip6_append_data(sk, ip_generic_getfrag, msg, ulen,
+ sizeof(struct udphdr), &ipc6, fl6,
+ dst_rt6_info(dst),
corkreq ? msg->msg_flags|MSG_MORE : msg->msg_flags);
if (err)
udp_v6_flush_pending_frames(sk);
--
2.53.0.473.g4a7958ca14-goog
^ permalink raw reply related [flat|nested] 19+ messages in thread
* [PATCH v2 net-next 09/15] udp: Remove UDPLITE_SEND_CSCOV and UDPLITE_RECV_CSCOV.
2026-03-05 21:49 [PATCH v2 net-next 00/15] udp: Retire UDP-Lite Kuniyuki Iwashima
` (7 preceding siblings ...)
2026-03-05 21:49 ` [PATCH v2 net-next 08/15] udp: Remove partial csum code in TX Kuniyuki Iwashima
@ 2026-03-05 21:49 ` Kuniyuki Iwashima
2026-03-05 21:49 ` [PATCH v2 net-next 10/15] udp: Remove struct proto.h.udp_table Kuniyuki Iwashima
` (5 subsequent siblings)
14 siblings, 0 replies; 19+ messages in thread
From: Kuniyuki Iwashima @ 2026-03-05 21:49 UTC (permalink / raw)
To: Willem de Bruijn, David Ahern, David S . Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni
Cc: Simon Horman, Florian Westphal, Kuniyuki Iwashima,
Kuniyuki Iwashima, netdev, Willem de Bruijn
UDP-Lite supports variable-length checksum and has two socket
options, UDPLITE_SEND_CSCOV and UDPLITE_RECV_CSCOV, to control
the checksum coverage.
Let's remove the support.
setsockopt(UDPLITE_SEND_CSCOV / UDPLITE_RECV_CSCOV) was only
available for UDP-Lite and returned -ENOPROTOOPT for UDP.
Now, the options are handled in ip_setsockopt() and
ipv6_setsockopt(), which still return the same error.
getsockopt(UDPLITE_SEND_CSCOV / UDPLITE_RECV_CSCOV) was available
for UDP and always returned 0, meaning full checksum, but now
-ENOPROTOOPT is returned.
Given that getsockopt() is meaningless for UDP and even the options
are not defined under include/uapi/, this should not be a problem.
$ man 7 udplite
...
BUGS
Where glibc support is missing, the following definitions
are needed:
#define IPPROTO_UDPLITE 136
#define UDPLITE_SEND_CSCOV 10
#define UDPLITE_RECV_CSCOV 11
Signed-off-by: Kuniyuki Iwashima <kuniyu@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
---
v2: Add a detailed reason in include/uapi/linux/udp.h
---
include/linux/udp.h | 10 +--------
include/net/udplite.h | 15 -------------
include/uapi/linux/udp.h | 2 ++
net/ipv4/udp.c | 46 ++--------------------------------------
net/ipv6/ip6_checksum.c | 2 +-
net/ipv6/udp.c | 5 ++---
6 files changed, 8 insertions(+), 72 deletions(-)
delete mode 100644 include/net/udplite.h
diff --git a/include/linux/udp.h b/include/linux/udp.h
index 1cbf6b4d3aab..ce56ebcee5cb 100644
--- a/include/linux/udp.h
+++ b/include/linux/udp.h
@@ -40,8 +40,6 @@ enum {
UDP_FLAGS_ACCEPT_FRAGLIST,
UDP_FLAGS_ACCEPT_L4,
UDP_FLAGS_ENCAP_ENABLED, /* This socket enabled encap */
- UDP_FLAGS_UDPLITE_SEND_CC, /* set via udplite setsockopt */
- UDP_FLAGS_UDPLITE_RECV_CC, /* set via udplite setsockopt */
};
/* per NUMA structure for lockless producer usage. */
@@ -74,11 +72,7 @@ struct udp_sock {
*/
__u16 len; /* total length of pending frames */
__u16 gso_size;
- /*
- * Fields specific to UDP-Lite.
- */
- __u16 pcslen;
- __u16 pcrlen;
+
/*
* For encapsulation sockets.
*/
@@ -236,8 +230,6 @@ static inline void udp_allow_gso(struct sock *sk)
hlist_nulls_for_each_entry_rcu(__up, node, list, udp_lrpa_node)
#endif
-#define IS_UDPLITE(__sk) (unlikely(__sk->sk_protocol == IPPROTO_UDPLITE))
-
static inline struct sock *udp_tunnel_sk(const struct net *net, bool is_ipv6)
{
#if IS_ENABLED(CONFIG_NET_UDP_TUNNEL)
diff --git a/include/net/udplite.h b/include/net/udplite.h
deleted file mode 100644
index 6bfa1d6833d1..000000000000
--- a/include/net/udplite.h
+++ /dev/null
@@ -1,15 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-/*
- * Definitions for the UDP-Lite (RFC 3828) code.
- */
-#ifndef _UDPLITE_H
-#define _UDPLITE_H
-
-#include <net/ip6_checksum.h>
-#include <net/udp.h>
-
-/* UDP-Lite socket options */
-#define UDPLITE_SEND_CSCOV 10 /* sender partial coverage (as sent) */
-#define UDPLITE_RECV_CSCOV 11 /* receiver partial coverage (threshold ) */
-
-#endif /* _UDPLITE_H */
diff --git a/include/uapi/linux/udp.h b/include/uapi/linux/udp.h
index edca3e430305..877fb02df8fb 100644
--- a/include/uapi/linux/udp.h
+++ b/include/uapi/linux/udp.h
@@ -29,6 +29,8 @@ struct udphdr {
/* UDP socket options */
#define UDP_CORK 1 /* Never send partially complete segments */
+/* Deprecated, reserved for UDPLITE_SEND_CSCOV 10 */
+/* Deprecated, reserved for UDPLITE_RECV_CSCOV 11 */
#define UDP_ENCAP 100 /* Set the socket to accept encapsulated packets */
#define UDP_NO_CHECK6_TX 101 /* Disable sending checksum for UDP6X */
#define UDP_NO_CHECK6_RX 102 /* Disable accepting checksum for UDP6 */
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 7d83e694115c..99abcc1e690a 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -117,7 +117,6 @@
#include <net/sock_reuseport.h>
#include <net/addrconf.h>
#include <net/udp_tunnel.h>
-#include <net/udplite.h>
#include <net/gro.h>
#if IS_ENABLED(CONFIG_IPV6)
#include <net/ipv6_stubs.h>
@@ -2917,7 +2916,6 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname,
struct udp_sock *up = udp_sk(sk);
int val, valbool;
int err = 0;
- int is_udplite = IS_UDPLITE(sk);
if (level == SOL_SOCKET) {
err = sk_setsockopt(sk, level, optname, optval, optlen);
@@ -3004,36 +3002,6 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname,
sockopt_release_sock(sk);
break;
- /*
- * UDP-Lite's partial checksum coverage (RFC 3828).
- */
- /* The sender sets actual checksum coverage length via this option.
- * The case coverage > packet length is handled by send module. */
- case UDPLITE_SEND_CSCOV:
- if (!is_udplite) /* Disable the option on UDP sockets */
- return -ENOPROTOOPT;
- if (val != 0 && val < 8) /* Illegal coverage: use default (8) */
- val = 8;
- else if (val > USHRT_MAX)
- val = USHRT_MAX;
- WRITE_ONCE(up->pcslen, val);
- udp_set_bit(UDPLITE_SEND_CC, sk);
- break;
-
- /* The receiver specifies a minimum checksum coverage value. To make
- * sense, this should be set to at least 8 (as done below). If zero is
- * used, this again means full checksum coverage. */
- case UDPLITE_RECV_CSCOV:
- if (!is_udplite) /* Disable the option on UDP sockets */
- return -ENOPROTOOPT;
- if (val != 0 && val < 8) /* Avoid silly minimal values. */
- val = 8;
- else if (val > USHRT_MAX)
- val = USHRT_MAX;
- WRITE_ONCE(up->pcrlen, val);
- udp_set_bit(UDPLITE_RECV_CC, sk);
- break;
-
default:
err = -ENOPROTOOPT;
break;
@@ -3046,7 +3014,7 @@ EXPORT_IPV6_MOD(udp_lib_setsockopt);
static int udp_setsockopt(struct sock *sk, int level, int optname, sockptr_t optval,
unsigned int optlen)
{
- if (level == SOL_UDP || level == SOL_UDPLITE || level == SOL_SOCKET)
+ if (level == SOL_UDP || level == SOL_SOCKET)
return udp_lib_setsockopt(sk, level, optname,
optval, optlen,
udp_push_pending_frames);
@@ -3092,16 +3060,6 @@ int udp_lib_getsockopt(struct sock *sk, int level, int optname,
val = udp_test_bit(GRO_ENABLED, sk);
break;
- /* The following two cannot be changed on UDP sockets, the return is
- * always 0 (which corresponds to the full checksum coverage of UDP). */
- case UDPLITE_SEND_CSCOV:
- val = READ_ONCE(up->pcslen);
- break;
-
- case UDPLITE_RECV_CSCOV:
- val = READ_ONCE(up->pcrlen);
- break;
-
default:
return -ENOPROTOOPT;
}
@@ -3117,7 +3075,7 @@ EXPORT_IPV6_MOD(udp_lib_getsockopt);
static int udp_getsockopt(struct sock *sk, int level, int optname,
char __user *optval, int __user *optlen)
{
- if (level == SOL_UDP || level == SOL_UDPLITE)
+ if (level == SOL_UDP)
return udp_lib_getsockopt(sk, level, optname, optval, optlen);
return ip_getsockopt(sk, level, optname, optval, optlen);
}
diff --git a/net/ipv6/ip6_checksum.c b/net/ipv6/ip6_checksum.c
index 8bb68a0cdfd6..e1a594873675 100644
--- a/net/ipv6/ip6_checksum.c
+++ b/net/ipv6/ip6_checksum.c
@@ -1,7 +1,7 @@
// SPDX-License-Identifier: GPL-2.0
#include <net/ip.h>
+#include <net/ip6_checksum.h>
#include <net/udp.h>
-#include <net/udplite.h>
#include <asm/checksum.h>
#ifndef _HAVE_ARCH_IPV6_CSUM
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 402b3b821480..49c315d0154d 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -58,7 +58,6 @@
#include <linux/proc_fs.h>
#include <linux/seq_file.h>
#include <trace/events/skb.h>
-#include <net/udplite.h>
static void udpv6_destruct_sock(struct sock *sk)
{
@@ -1832,7 +1831,7 @@ static void udpv6_destroy_sock(struct sock *sk)
static int udpv6_setsockopt(struct sock *sk, int level, int optname,
sockptr_t optval, unsigned int optlen)
{
- if (level == SOL_UDP || level == SOL_UDPLITE || level == SOL_SOCKET)
+ if (level == SOL_UDP || level == SOL_SOCKET)
return udp_lib_setsockopt(sk, level, optname,
optval, optlen,
udp_v6_push_pending_frames);
@@ -1842,7 +1841,7 @@ static int udpv6_setsockopt(struct sock *sk, int level, int optname,
static int udpv6_getsockopt(struct sock *sk, int level, int optname,
char __user *optval, int __user *optlen)
{
- if (level == SOL_UDP || level == SOL_UDPLITE)
+ if (level == SOL_UDP)
return udp_lib_getsockopt(sk, level, optname, optval, optlen);
return ipv6_getsockopt(sk, level, optname, optval, optlen);
}
--
2.53.0.473.g4a7958ca14-goog
^ permalink raw reply related [flat|nested] 19+ messages in thread
* [PATCH v2 net-next 10/15] udp: Remove struct proto.h.udp_table.
2026-03-05 21:49 [PATCH v2 net-next 00/15] udp: Retire UDP-Lite Kuniyuki Iwashima
` (8 preceding siblings ...)
2026-03-05 21:49 ` [PATCH v2 net-next 09/15] udp: Remove UDPLITE_SEND_CSCOV and UDPLITE_RECV_CSCOV Kuniyuki Iwashima
@ 2026-03-05 21:49 ` Kuniyuki Iwashima
2026-03-05 21:49 ` [PATCH v2 net-next 11/15] udp: Remove udp_table in struct udp_seq_afinfo Kuniyuki Iwashima
` (4 subsequent siblings)
14 siblings, 0 replies; 19+ messages in thread
From: Kuniyuki Iwashima @ 2026-03-05 21:49 UTC (permalink / raw)
To: Willem de Bruijn, David Ahern, David S . Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni
Cc: Simon Horman, Florian Westphal, Kuniyuki Iwashima,
Kuniyuki Iwashima, netdev, Willem de Bruijn
Since UDP and UDP-Lite had dedicated socket hash tables for
each, we have had to fetch them from different pointers.
UDP always has its global or per-netns table in
net->ipv4.udp_table and struct proto.h.udp_table is NULL.
OTOH, UDP-Lite had only one global table in the pointer.
We no longer use the field.
Let's remove it and udp_get_table_prot().
Signed-off-by: Kuniyuki Iwashima <kuniyu@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
---
include/net/sock.h | 1 -
net/ipv4/udp.c | 26 ++++++++++++--------------
net/ipv6/udp.c | 1 -
3 files changed, 12 insertions(+), 16 deletions(-)
diff --git a/include/net/sock.h b/include/net/sock.h
index 16a1b8895206..7d51ac9e7d9a 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -1392,7 +1392,6 @@ struct proto {
union {
struct inet_hashinfo *hashinfo;
- struct udp_table *udp_table;
struct raw_hashinfo *raw_hash;
struct smc_hashinfo *smc_hash;
} h;
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 99abcc1e690a..432b4c5e7ad4 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -134,11 +134,6 @@ EXPORT_PER_CPU_SYMBOL_GPL(udp_memory_per_cpu_fw_alloc);
#define MAX_UDP_PORTS 65536
#define PORTS_PER_CHAIN (MAX_UDP_PORTS / UDP_HTABLE_SIZE_MIN_PERNET)
-static struct udp_table *udp_get_table_prot(struct sock *sk)
-{
- return sk->sk_prot->h.udp_table ? : sock_net(sk)->ipv4.udp_table;
-}
-
static int udp_lib_lport_inuse(struct net *net, __u16 num,
const struct udp_hslot *hslot,
unsigned long *bitmap,
@@ -240,11 +235,13 @@ static int udp_reuseport_add_sock(struct sock *sk, struct udp_hslot *hslot)
int udp_lib_get_port(struct sock *sk, unsigned short snum,
unsigned int hash2_nulladdr)
{
- struct udp_table *udptable = udp_get_table_prot(sk);
struct udp_hslot *hslot, *hslot2;
struct net *net = sock_net(sk);
+ struct udp_table *udptable;
int error = -EADDRINUSE;
+ udptable = net->ipv4.udp_table;
+
if (!snum) {
DECLARE_BITMAP(bitmap, PORTS_PER_CHAIN);
unsigned short first, last;
@@ -2224,12 +2221,13 @@ EXPORT_IPV6_MOD(udp_disconnect);
void udp_lib_unhash(struct sock *sk)
{
if (sk_hashed(sk)) {
- struct udp_table *udptable = udp_get_table_prot(sk);
struct udp_hslot *hslot, *hslot2;
+ struct net *net = sock_net(sk);
+ struct udp_table *udptable;
sock_rps_delete_flow(sk);
- hslot = udp_hashslot(udptable, sock_net(sk),
- udp_sk(sk)->udp_port_hash);
+ udptable = net->ipv4.udp_table;
+ hslot = udp_hashslot(udptable, net, udp_sk(sk)->udp_port_hash);
hslot2 = udp_hashslot2(udptable, udp_sk(sk)->udp_portaddr_hash);
spin_lock_bh(&hslot->lock);
@@ -2238,7 +2236,7 @@ void udp_lib_unhash(struct sock *sk)
if (sk_del_node_init_rcu(sk)) {
hslot->count--;
inet_sk(sk)->inet_num = 0;
- sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1);
+ sock_prot_inuse_add(net, sk->sk_prot, -1);
spin_lock(&hslot2->lock);
hlist_del_init_rcu(&udp_sk(sk)->udp_portaddr_node);
@@ -2258,11 +2256,12 @@ EXPORT_IPV6_MOD(udp_lib_unhash);
void udp_lib_rehash(struct sock *sk, u16 newhash, u16 newhash4)
{
if (sk_hashed(sk)) {
- struct udp_table *udptable = udp_get_table_prot(sk);
struct udp_hslot *hslot, *hslot2, *nhslot2;
+ struct net *net = sock_net(sk);
+ struct udp_table *udptable;
- hslot = udp_hashslot(udptable, sock_net(sk),
- udp_sk(sk)->udp_port_hash);
+ udptable = net->ipv4.udp_table;
+ hslot = udp_hashslot(udptable, net, udp_sk(sk)->udp_port_hash);
hslot2 = udp_hashslot2(udptable, udp_sk(sk)->udp_portaddr_hash);
nhslot2 = udp_hashslot2(udptable, newhash);
udp_sk(sk)->udp_portaddr_hash = newhash;
@@ -3168,7 +3167,6 @@ struct proto udp_prot = {
.sysctl_wmem_offset = offsetof(struct net, ipv4.sysctl_udp_wmem_min),
.sysctl_rmem_offset = offsetof(struct net, ipv4.sysctl_udp_rmem_min),
.obj_size = sizeof(struct udp_sock),
- .h.udp_table = NULL,
.diag_destroy = udp_abort,
};
EXPORT_SYMBOL(udp_prot);
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 49c315d0154d..abbea5bcbd6f 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -1925,7 +1925,6 @@ struct proto udpv6_prot = {
.sysctl_rmem_offset = offsetof(struct net, ipv4.sysctl_udp_rmem_min),
.obj_size = sizeof(struct udp6_sock),
.ipv6_pinfo_offset = offsetof(struct udp6_sock, inet6),
- .h.udp_table = NULL,
.diag_destroy = udp_abort,
};
--
2.53.0.473.g4a7958ca14-goog
^ permalink raw reply related [flat|nested] 19+ messages in thread
* [PATCH v2 net-next 11/15] udp: Remove udp_table in struct udp_seq_afinfo.
2026-03-05 21:49 [PATCH v2 net-next 00/15] udp: Retire UDP-Lite Kuniyuki Iwashima
` (9 preceding siblings ...)
2026-03-05 21:49 ` [PATCH v2 net-next 10/15] udp: Remove struct proto.h.udp_table Kuniyuki Iwashima
@ 2026-03-05 21:49 ` Kuniyuki Iwashima
2026-03-05 21:49 ` [PATCH v2 net-next 12/15] udp: Remove dead check in __udp[46]_lib_lookup() for BPF Kuniyuki Iwashima
` (3 subsequent siblings)
14 siblings, 0 replies; 19+ messages in thread
From: Kuniyuki Iwashima @ 2026-03-05 21:49 UTC (permalink / raw)
To: Willem de Bruijn, David Ahern, David S . Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni
Cc: Simon Horman, Florian Westphal, Kuniyuki Iwashima,
Kuniyuki Iwashima, netdev, Willem de Bruijn
Since UDP and UDP-Lite had dedicated socket hash tables for
each, we have had to fetch them from different pointers for
procfs or bpf iterator.
UDP always has its global or per-netns table in
net->ipv4.udp_table and struct udp_seq_afinfo.udp_table
is NULL.
OTOH, UDP-Lite had only one global table in the pointer.
We no longer use the field.
Let's remove it and udp_get_table_seq().
Signed-off-by: Kuniyuki Iwashima <kuniyu@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
---
include/net/udp.h | 1 -
net/ipv4/udp.c | 22 ++++------------------
net/ipv6/udp.c | 1 -
3 files changed, 4 insertions(+), 20 deletions(-)
diff --git a/include/net/udp.h b/include/net/udp.h
index bc275cda9f8c..76f401988353 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -552,7 +552,6 @@ static inline int copy_linear_skb(struct sk_buff *skb, int len, int off,
#ifdef CONFIG_PROC_FS
struct udp_seq_afinfo {
sa_family_t family;
- struct udp_table *udp_table;
};
struct udp_iter_state {
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 432b4c5e7ad4..5ae0d88e3cca 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -3186,21 +3186,8 @@ static bool seq_sk_match(struct seq_file *seq, const struct sock *sk)
#ifdef CONFIG_BPF_SYSCALL
static const struct seq_operations bpf_iter_udp_seq_ops;
-#endif
-static struct udp_table *udp_get_table_seq(struct seq_file *seq,
- struct net *net)
-{
- const struct udp_seq_afinfo *afinfo;
-
-#ifdef CONFIG_BPF_SYSCALL
- if (seq->op == &bpf_iter_udp_seq_ops)
- return net->ipv4.udp_table;
#endif
- afinfo = pde_data(file_inode(seq->file));
- return afinfo->udp_table ? : net->ipv4.udp_table;
-}
-
static struct sock *udp_get_first(struct seq_file *seq, int start)
{
struct udp_iter_state *state = seq->private;
@@ -3208,7 +3195,7 @@ static struct sock *udp_get_first(struct seq_file *seq, int start)
struct udp_table *udptable;
struct sock *sk;
- udptable = udp_get_table_seq(seq, net);
+ udptable = net->ipv4.udp_table;
for (state->bucket = start; state->bucket <= udptable->mask;
++state->bucket) {
@@ -3240,7 +3227,7 @@ static struct sock *udp_get_next(struct seq_file *seq, struct sock *sk)
} while (sk && !seq_sk_match(seq, sk));
if (!sk) {
- udptable = udp_get_table_seq(seq, net);
+ udptable = net->ipv4.udp_table;
if (state->bucket <= udptable->mask)
spin_unlock_bh(&udptable->hash[state->bucket].lock);
@@ -3288,7 +3275,7 @@ void udp_seq_stop(struct seq_file *seq, void *v)
struct udp_iter_state *state = seq->private;
struct udp_table *udptable;
- udptable = udp_get_table_seq(seq, seq_file_net(seq));
+ udptable = seq_file_net(seq)->ipv4.udp_table;
if (state->bucket <= udptable->mask)
spin_unlock_bh(&udptable->hash[state->bucket].lock);
@@ -3392,7 +3379,7 @@ static struct sock *bpf_iter_udp_batch(struct seq_file *seq)
if (iter->cur_sk == iter->end_sk)
state->bucket++;
- udptable = udp_get_table_seq(seq, net);
+ udptable = net->ipv4.udp_table;
again:
/* New batch for the next bucket.
@@ -3630,7 +3617,6 @@ static const struct seq_operations udp_seq_ops = {
static struct udp_seq_afinfo udp4_seq_afinfo = {
.family = AF_INET,
- .udp_table = NULL,
};
static int __net_init udp4_proc_init_net(struct net *net)
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index abbea5bcbd6f..54b5a92b04d0 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -1873,7 +1873,6 @@ static const struct seq_operations udp6_seq_ops = {
static struct udp_seq_afinfo udp6_seq_afinfo = {
.family = AF_INET6,
- .udp_table = NULL,
};
int __net_init udp6_proc_init(struct net *net)
--
2.53.0.473.g4a7958ca14-goog
^ permalink raw reply related [flat|nested] 19+ messages in thread
* [PATCH v2 net-next 12/15] udp: Remove dead check in __udp[46]_lib_lookup() for BPF.
2026-03-05 21:49 [PATCH v2 net-next 00/15] udp: Retire UDP-Lite Kuniyuki Iwashima
` (10 preceding siblings ...)
2026-03-05 21:49 ` [PATCH v2 net-next 11/15] udp: Remove udp_table in struct udp_seq_afinfo Kuniyuki Iwashima
@ 2026-03-05 21:49 ` Kuniyuki Iwashima
2026-03-05 21:49 ` [PATCH v2 net-next 13/15] udp: Don't pass udptable to IPv6 socket lookup functions Kuniyuki Iwashima
` (2 subsequent siblings)
14 siblings, 0 replies; 19+ messages in thread
From: Kuniyuki Iwashima @ 2026-03-05 21:49 UTC (permalink / raw)
To: Willem de Bruijn, David Ahern, David S . Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni
Cc: Simon Horman, Florian Westphal, Kuniyuki Iwashima,
Kuniyuki Iwashima, netdev, Willem de Bruijn
BPF socket lookup for SO_REUSEPORT does not support UDP-Lite.
In __udp4_lib_lookup() and __udp6_lib_lookup(), it checks if
the passed udptable pointer is the same as net->ipv4.udp_table,
which is only true for UDP.
Now, the condition is always true.
Let's remove the check.
Signed-off-by: Kuniyuki Iwashima <kuniyu@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
---
net/ipv4/udp.c | 3 +--
net/ipv6/udp.c | 3 +--
2 files changed, 2 insertions(+), 4 deletions(-)
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 5ae0d88e3cca..86765a17de67 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -699,8 +699,7 @@ struct sock *__udp4_lib_lookup(const struct net *net, __be32 saddr,
goto done;
/* Lookup redirect from BPF */
- if (static_branch_unlikely(&bpf_sk_lookup_enabled) &&
- udptable == net->ipv4.udp_table) {
+ if (static_branch_unlikely(&bpf_sk_lookup_enabled)) {
sk = inet_lookup_run_sk_lookup(net, IPPROTO_UDP, skb, sizeof(struct udphdr),
saddr, sport, daddr, hnum, dif,
udp_ehashfn);
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 54b5a92b04d0..fb50dcf8a384 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -370,8 +370,7 @@ struct sock *__udp6_lib_lookup(const struct net *net,
goto done;
/* Lookup redirect from BPF */
- if (static_branch_unlikely(&bpf_sk_lookup_enabled) &&
- udptable == net->ipv4.udp_table) {
+ if (static_branch_unlikely(&bpf_sk_lookup_enabled)) {
sk = inet6_lookup_run_sk_lookup(net, IPPROTO_UDP, skb, sizeof(struct udphdr),
saddr, sport, daddr, hnum, dif,
udp6_ehashfn);
--
2.53.0.473.g4a7958ca14-goog
^ permalink raw reply related [flat|nested] 19+ messages in thread
* [PATCH v2 net-next 13/15] udp: Don't pass udptable to IPv6 socket lookup functions.
2026-03-05 21:49 [PATCH v2 net-next 00/15] udp: Retire UDP-Lite Kuniyuki Iwashima
` (11 preceding siblings ...)
2026-03-05 21:49 ` [PATCH v2 net-next 12/15] udp: Remove dead check in __udp[46]_lib_lookup() for BPF Kuniyuki Iwashima
@ 2026-03-05 21:49 ` Kuniyuki Iwashima
2026-03-05 21:50 ` [PATCH v2 net-next 14/15] udp: Don't pass udptable to IPv4 " Kuniyuki Iwashima
2026-03-05 21:50 ` [PATCH v2 net-next 15/15] udp: Don't pass proto to __udp4_lib_rcv() and __udp6_lib_rcv() Kuniyuki Iwashima
14 siblings, 0 replies; 19+ messages in thread
From: Kuniyuki Iwashima @ 2026-03-05 21:49 UTC (permalink / raw)
To: Willem de Bruijn, David Ahern, David S . Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni
Cc: Simon Horman, Florian Westphal, Kuniyuki Iwashima,
Kuniyuki Iwashima, netdev, Willem de Bruijn
Since UDP and UDP-Lite had dedicated socket hash tables for
each, we have had to pass the pointer down to many socket
lookup functions.
UDP-Lite gone, and we do not need to do that.
Let's fetch net->ipv4.udp_table only where needed in IPv6
stack: __udp6_lib_lookup() and __udp6_lib_mcast_deliver().
__udp6_lib_err() is renamed to udpv6_err() as its wrapper
is no longer needed.
Signed-off-by: Kuniyuki Iwashima <kuniyu@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
---
include/net/ipv6_stubs.h | 7 ++--
include/net/udp.h | 3 +-
net/core/filter.c | 3 +-
net/ipv4/udp_diag.c | 20 +++++-----
net/ipv6/udp.c | 82 ++++++++++++++++++----------------------
net/ipv6/udp_offload.c | 3 +-
6 files changed, 53 insertions(+), 65 deletions(-)
diff --git a/include/net/ipv6_stubs.h b/include/net/ipv6_stubs.h
index d3013e721b14..907681cecde8 100644
--- a/include/net/ipv6_stubs.h
+++ b/include/net/ipv6_stubs.h
@@ -83,10 +83,9 @@ struct ipv6_bpf_stub {
int (*inet6_bind)(struct sock *sk, struct sockaddr_unsized *uaddr, int addr_len,
u32 flags);
struct sock *(*udp6_lib_lookup)(const struct net *net,
- const struct in6_addr *saddr, __be16 sport,
- const struct in6_addr *daddr, __be16 dport,
- int dif, int sdif, struct udp_table *tbl,
- struct sk_buff *skb);
+ const struct in6_addr *saddr, __be16 sport,
+ const struct in6_addr *daddr, __be16 dport,
+ int dif, int sdif, struct sk_buff *skb);
int (*ipv6_setsockopt)(struct sock *sk, int level, int optname,
sockptr_t optval, unsigned int optlen);
int (*ipv6_getsockopt)(struct sock *sk, int level, int optname,
diff --git a/include/net/udp.h b/include/net/udp.h
index 76f401988353..adec74531ee1 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -449,8 +449,7 @@ struct sock *udp6_lib_lookup(const struct net *net,
struct sock *__udp6_lib_lookup(const struct net *net,
const struct in6_addr *saddr, __be16 sport,
const struct in6_addr *daddr, __be16 dport,
- int dif, int sdif, struct udp_table *tbl,
- struct sk_buff *skb);
+ int dif, int sdif, struct sk_buff *skb);
struct sock *udp6_lib_lookup_skb(const struct sk_buff *skb,
__be16 sport, __be16 dport);
int udp_read_skb(struct sock *sk, skb_read_actor_t recv_actor);
diff --git a/net/core/filter.c b/net/core/filter.c
index 0d5d5a17acb2..7a3440d009a8 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -6895,8 +6895,7 @@ static struct sock *sk_lookup(struct net *net, struct bpf_sock_tuple *tuple,
sk = ipv6_bpf_stub->udp6_lib_lookup(net,
src6, tuple->ipv6.sport,
dst6, tuple->ipv6.dport,
- dif, sdif,
- net->ipv4.udp_table, NULL);
+ dif, sdif, NULL);
#endif
}
diff --git a/net/ipv4/udp_diag.c b/net/ipv4/udp_diag.c
index a010d05062a0..0899c60cce53 100644
--- a/net/ipv4/udp_diag.c
+++ b/net/ipv4/udp_diag.c
@@ -44,11 +44,11 @@ static int udp_dump_one(struct udp_table *tbl,
#if IS_ENABLED(CONFIG_IPV6)
else if (req->sdiag_family == AF_INET6)
sk = __udp6_lib_lookup(net,
- (struct in6_addr *)req->id.idiag_src,
- req->id.idiag_sport,
- (struct in6_addr *)req->id.idiag_dst,
- req->id.idiag_dport,
- req->id.idiag_if, 0, tbl, NULL);
+ (struct in6_addr *)req->id.idiag_src,
+ req->id.idiag_sport,
+ (struct in6_addr *)req->id.idiag_dst,
+ req->id.idiag_dport,
+ req->id.idiag_if, 0, NULL);
#endif
if (sk && !refcount_inc_not_zero(&sk->sk_refcnt))
sk = NULL;
@@ -185,11 +185,11 @@ static int __udp_diag_destroy(struct sk_buff *in_skb,
else
sk = __udp6_lib_lookup(net,
- (struct in6_addr *)req->id.idiag_dst,
- req->id.idiag_dport,
- (struct in6_addr *)req->id.idiag_src,
- req->id.idiag_sport,
- req->id.idiag_if, 0, tbl, NULL);
+ (struct in6_addr *)req->id.idiag_dst,
+ req->id.idiag_dport,
+ (struct in6_addr *)req->id.idiag_src,
+ req->id.idiag_sport,
+ req->id.idiag_if, 0, NULL);
}
#endif
else {
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index fb50dcf8a384..0de28cbfe793 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -344,9 +344,9 @@ static void udp6_hash4(struct sock *sk)
struct sock *__udp6_lib_lookup(const struct net *net,
const struct in6_addr *saddr, __be16 sport,
const struct in6_addr *daddr, __be16 dport,
- int dif, int sdif, struct udp_table *udptable,
- struct sk_buff *skb)
+ int dif, int sdif, struct sk_buff *skb)
{
+ struct udp_table *udptable = net->ipv4.udp_table;
unsigned short hnum = ntohs(dport);
struct udp_hslot *hslot2;
struct sock *result, *sk;
@@ -406,14 +406,13 @@ struct sock *__udp6_lib_lookup(const struct net *net,
EXPORT_SYMBOL_GPL(__udp6_lib_lookup);
static struct sock *__udp6_lib_lookup_skb(struct sk_buff *skb,
- __be16 sport, __be16 dport,
- struct udp_table *udptable)
+ __be16 sport, __be16 dport)
{
const struct ipv6hdr *iph = ipv6_hdr(skb);
return __udp6_lib_lookup(dev_net(skb->dev), &iph->saddr, sport,
&iph->daddr, dport, inet6_iif(skb),
- inet6_sdif(skb), udptable, skb);
+ inet6_sdif(skb), skb);
}
struct sock *udp6_lib_lookup_skb(const struct sk_buff *skb,
@@ -421,14 +420,12 @@ struct sock *udp6_lib_lookup_skb(const struct sk_buff *skb,
{
const u16 offset = NAPI_GRO_CB(skb)->network_offsets[skb->encapsulation];
const struct ipv6hdr *iph = (struct ipv6hdr *)(skb->data + offset);
- struct net *net = dev_net(skb->dev);
int iif, sdif;
inet6_get_iif_sdif(skb, &iif, &sdif);
- return __udp6_lib_lookup(net, &iph->saddr, sport,
- &iph->daddr, dport, iif,
- sdif, net->ipv4.udp_table, NULL);
+ return __udp6_lib_lookup(dev_net(skb->dev), &iph->saddr, sport,
+ &iph->daddr, dport, iif, sdif, NULL);
}
/* Must be called under rcu_read_lock().
@@ -440,8 +437,7 @@ struct sock *udp6_lib_lookup(const struct net *net, const struct in6_addr *saddr
{
struct sock *sk;
- sk = __udp6_lib_lookup(net, saddr, sport, daddr, dport,
- dif, 0, net->ipv4.udp_table, NULL);
+ sk = __udp6_lib_lookup(net, saddr, sport, daddr, dport, dif, 0, NULL);
if (sk && !refcount_inc_not_zero(&sk->sk_refcnt))
sk = NULL;
return sk;
@@ -642,7 +638,6 @@ static int __udp6_lib_err_encap_no_sk(struct sk_buff *skb,
static struct sock *__udp6_lib_err_encap(struct net *net,
const struct ipv6hdr *hdr, int offset,
struct udphdr *uh,
- struct udp_table *udptable,
struct sock *sk,
struct sk_buff *skb,
struct inet6_skb_parm *opt,
@@ -673,7 +668,7 @@ static struct sock *__udp6_lib_err_encap(struct net *net,
sk = __udp6_lib_lookup(net, &hdr->daddr, uh->source,
&hdr->saddr, uh->dest,
- inet6_iif(skb), 0, udptable, skb);
+ inet6_iif(skb), 0, skb);
if (sk) {
up = udp_sk(sk);
@@ -694,29 +689,28 @@ static struct sock *__udp6_lib_err_encap(struct net *net,
return sk;
}
-static int __udp6_lib_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
- u8 type, u8 code, int offset, __be32 info,
- struct udp_table *udptable)
+static int udpv6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
+ u8 type, u8 code, int offset, __be32 info)
{
- struct ipv6_pinfo *np;
const struct ipv6hdr *hdr = (const struct ipv6hdr *)skb->data;
- const struct in6_addr *saddr = &hdr->saddr;
- const struct in6_addr *daddr = seg6_get_daddr(skb, opt) ? : &hdr->daddr;
- struct udphdr *uh = (struct udphdr *)(skb->data+offset);
+ struct udphdr *uh = (struct udphdr *)(skb->data + offset);
+ const struct in6_addr *saddr, *daddr;
+ struct net *net = dev_net(skb->dev);
+ struct ipv6_pinfo *np;
bool tunnel = false;
struct sock *sk;
int harderr;
int err;
- struct net *net = dev_net(skb->dev);
+ daddr = seg6_get_daddr(skb, opt) ? : &hdr->daddr;
+ saddr = &hdr->saddr;
sk = __udp6_lib_lookup(net, daddr, uh->dest, saddr, uh->source,
- inet6_iif(skb), inet6_sdif(skb), udptable, NULL);
+ inet6_iif(skb), inet6_sdif(skb), NULL);
if (!sk || READ_ONCE(udp_sk(sk)->encap_type)) {
/* No socket for error: try tunnels before discarding */
if (static_branch_unlikely(&udpv6_encap_needed_key)) {
- sk = __udp6_lib_err_encap(net, hdr, offset, uh,
- udptable, sk, skb,
+ sk = __udp6_lib_err_encap(net, hdr, offset, uh, sk, skb,
opt, type, code, info);
if (!sk)
return 0;
@@ -808,14 +802,6 @@ static int __udpv6_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
return 0;
}
-static __inline__ int udpv6_err(struct sk_buff *skb,
- struct inet6_skb_parm *opt, u8 type,
- u8 code, int offset, __be32 info)
-{
- return __udp6_lib_err(skb, opt, type, code, offset, info,
- dev_net(skb->dev)->ipv4.udp_table);
-}
-
static int udpv6_queue_rcv_one_skb(struct sock *sk, struct sk_buff *skb)
{
enum skb_drop_reason drop_reason = SKB_DROP_REASON_NOT_SPECIFIED;
@@ -947,19 +933,27 @@ static void udp6_csum_zero_error(struct sk_buff *skb)
* so we don't need to lock the hashes.
*/
static int __udp6_lib_mcast_deliver(struct net *net, struct sk_buff *skb,
- const struct in6_addr *saddr, const struct in6_addr *daddr,
- struct udp_table *udptable, int proto)
+ const struct in6_addr *saddr,
+ const struct in6_addr *daddr,
+ int proto)
{
- struct sock *sk, *first = NULL;
+ struct udp_table *udptable = net->ipv4.udp_table;
const struct udphdr *uh = udp_hdr(skb);
+ unsigned int hash2, hash2_any, offset;
unsigned short hnum = ntohs(uh->dest);
- struct udp_hslot *hslot = udp_hashslot(udptable, net, hnum);
- unsigned int offset = offsetof(typeof(*sk), sk_node);
- unsigned int hash2 = 0, hash2_any = 0, use_hash2 = (hslot->count > 10);
- int dif = inet6_iif(skb);
+ struct sock *sk, *first = NULL;
int sdif = inet6_sdif(skb);
+ int dif = inet6_iif(skb);
struct hlist_node *node;
+ struct udp_hslot *hslot;
struct sk_buff *nskb;
+ bool use_hash2;
+
+ hash2_any = 0;
+ hash2 = 0;
+ hslot = udp_hashslot(udptable, net, hnum);
+ use_hash2 = hslot->count > 10;
+ offset = offsetof(typeof(*sk), sk_node);
if (use_hash2) {
hash2_any = ipv6_portaddr_hash(net, &in6addr_any, hnum) &
@@ -1069,8 +1063,7 @@ static int udp6_csum_init(struct sk_buff *skb, struct udphdr *uh)
return 0;
}
-static int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,
- int proto)
+static int __udp6_lib_rcv(struct sk_buff *skb, int proto)
{
enum skb_drop_reason reason = SKB_DROP_REASON_NOT_SPECIFIED;
const struct in6_addr *saddr, *daddr;
@@ -1140,11 +1133,10 @@ static int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,
* Multicast receive code
*/
if (ipv6_addr_is_multicast(daddr))
- return __udp6_lib_mcast_deliver(net, skb,
- saddr, daddr, udptable, proto);
+ return __udp6_lib_mcast_deliver(net, skb, saddr, daddr, proto);
/* Unicast */
- sk = __udp6_lib_lookup_skb(skb, uh->source, uh->dest, udptable);
+ sk = __udp6_lib_lookup_skb(skb, uh->source, uh->dest);
if (sk) {
if (!uh->check && !udp_get_no_check6_rx(sk))
goto report_csum_error;
@@ -1262,7 +1254,7 @@ void udp_v6_early_demux(struct sk_buff *skb)
INDIRECT_CALLABLE_SCOPE int udpv6_rcv(struct sk_buff *skb)
{
- return __udp6_lib_rcv(skb, dev_net(skb->dev)->ipv4.udp_table, IPPROTO_UDP);
+ return __udp6_lib_rcv(skb, IPPROTO_UDP);
}
/*
diff --git a/net/ipv6/udp_offload.c b/net/ipv6/udp_offload.c
index e003b8494dc0..778afc7453ce 100644
--- a/net/ipv6/udp_offload.c
+++ b/net/ipv6/udp_offload.c
@@ -128,8 +128,7 @@ static struct sock *udp6_gro_lookup_skb(struct sk_buff *skb, __be16 sport,
inet6_get_iif_sdif(skb, &iif, &sdif);
return __udp6_lib_lookup(net, &iph->saddr, sport,
- &iph->daddr, dport, iif,
- sdif, net->ipv4.udp_table, NULL);
+ &iph->daddr, dport, iif, sdif, NULL);
}
struct sk_buff *udp6_gro_receive(struct list_head *head, struct sk_buff *skb)
--
2.53.0.473.g4a7958ca14-goog
^ permalink raw reply related [flat|nested] 19+ messages in thread
* [PATCH v2 net-next 14/15] udp: Don't pass udptable to IPv4 socket lookup functions.
2026-03-05 21:49 [PATCH v2 net-next 00/15] udp: Retire UDP-Lite Kuniyuki Iwashima
` (12 preceding siblings ...)
2026-03-05 21:49 ` [PATCH v2 net-next 13/15] udp: Don't pass udptable to IPv6 socket lookup functions Kuniyuki Iwashima
@ 2026-03-05 21:50 ` Kuniyuki Iwashima
2026-03-05 21:50 ` [PATCH v2 net-next 15/15] udp: Don't pass proto to __udp4_lib_rcv() and __udp6_lib_rcv() Kuniyuki Iwashima
14 siblings, 0 replies; 19+ messages in thread
From: Kuniyuki Iwashima @ 2026-03-05 21:50 UTC (permalink / raw)
To: Willem de Bruijn, David Ahern, David S . Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni
Cc: Simon Horman, Florian Westphal, Kuniyuki Iwashima,
Kuniyuki Iwashima, netdev, Willem de Bruijn
Since UDP and UDP-Lite had dedicated socket hash tables for
each, we have had to pass the pointer down to many socket
lookup functions.
UDP-Lite gone, and we do not need to do that.
Let's fetch net->ipv4.udp_table only where needed in IPv4
stack: __udp4_lib_lookup(), __udp4_lib_mcast_deliver(),
and udp_diag_dump().
Some functions are renamed as the wrapper functions are no
longer needed.
__udp4_lib_err() -> udp_err()
__udp_diag_destroy() -> udp_diag_destroy()
udp_dump_one() -> udp_diag_dump_one()
udp_dump() -> udp_diag_dump()
Signed-off-by: Kuniyuki Iwashima <kuniyu@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
---
include/net/udp.h | 5 ++-
net/core/filter.c | 2 +-
net/ipv4/udp.c | 75 +++++++++++++++++++-----------------------
net/ipv4/udp_diag.c | 61 ++++++++++++----------------------
net/ipv4/udp_offload.c | 3 +-
5 files changed, 59 insertions(+), 87 deletions(-)
diff --git a/include/net/udp.h b/include/net/udp.h
index adec74531ee1..8262e2b215b4 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -437,9 +437,8 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname,
struct sock *udp4_lib_lookup(const struct net *net, __be32 saddr, __be16 sport,
__be32 daddr, __be16 dport, int dif);
struct sock *__udp4_lib_lookup(const struct net *net, __be32 saddr,
- __be16 sport,
- __be32 daddr, __be16 dport, int dif, int sdif,
- struct udp_table *tbl, struct sk_buff *skb);
+ __be16 sport, __be32 daddr, __be16 dport,
+ int dif, int sdif, struct sk_buff *skb);
struct sock *udp4_lib_lookup_skb(const struct sk_buff *skb,
__be16 sport, __be16 dport);
struct sock *udp6_lib_lookup(const struct net *net,
diff --git a/net/core/filter.c b/net/core/filter.c
index 7a3440d009a8..de6ef1915af3 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -6880,7 +6880,7 @@ static struct sock *sk_lookup(struct net *net, struct bpf_sock_tuple *tuple,
else
sk = __udp4_lib_lookup(net, src4, tuple->ipv4.sport,
dst4, tuple->ipv4.dport,
- dif, sdif, net->ipv4.udp_table, NULL);
+ dif, sdif, NULL);
#if IS_ENABLED(CONFIG_IPV6)
} else {
struct in6_addr *src6 = (struct in6_addr *)&tuple->ipv6.saddr;
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 86765a17de67..0436462cb9fd 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -673,9 +673,10 @@ EXPORT_IPV6_MOD(udp4_hash4);
* harder than this. -DaveM
*/
struct sock *__udp4_lib_lookup(const struct net *net, __be32 saddr,
- __be16 sport, __be32 daddr, __be16 dport, int dif,
- int sdif, struct udp_table *udptable, struct sk_buff *skb)
+ __be16 sport, __be32 daddr, __be16 dport,
+ int dif, int sdif, struct sk_buff *skb)
{
+ struct udp_table *udptable = net->ipv4.udp_table;
unsigned short hnum = ntohs(dport);
struct udp_hslot *hslot2;
struct sock *result, *sk;
@@ -741,14 +742,13 @@ struct sock *__udp4_lib_lookup(const struct net *net, __be32 saddr,
EXPORT_SYMBOL_GPL(__udp4_lib_lookup);
static inline struct sock *__udp4_lib_lookup_skb(struct sk_buff *skb,
- __be16 sport, __be16 dport,
- struct udp_table *udptable)
+ __be16 sport, __be16 dport)
{
const struct iphdr *iph = ip_hdr(skb);
return __udp4_lib_lookup(dev_net(skb->dev), iph->saddr, sport,
iph->daddr, dport, inet_iif(skb),
- inet_sdif(skb), udptable, skb);
+ inet_sdif(skb), skb);
}
struct sock *udp4_lib_lookup_skb(const struct sk_buff *skb,
@@ -756,14 +756,12 @@ struct sock *udp4_lib_lookup_skb(const struct sk_buff *skb,
{
const u16 offset = NAPI_GRO_CB(skb)->network_offsets[skb->encapsulation];
const struct iphdr *iph = (struct iphdr *)(skb->data + offset);
- struct net *net = dev_net(skb->dev);
int iif, sdif;
inet_get_iif_sdif(skb, &iif, &sdif);
- return __udp4_lib_lookup(net, iph->saddr, sport,
- iph->daddr, dport, iif,
- sdif, net->ipv4.udp_table, NULL);
+ return __udp4_lib_lookup(dev_net(skb->dev), iph->saddr, sport,
+ iph->daddr, dport, iif, sdif, NULL);
}
/* Must be called under rcu_read_lock().
@@ -775,8 +773,7 @@ struct sock *udp4_lib_lookup(const struct net *net, __be32 saddr, __be16 sport,
{
struct sock *sk;
- sk = __udp4_lib_lookup(net, saddr, sport, daddr, dport,
- dif, 0, net->ipv4.udp_table, NULL);
+ sk = __udp4_lib_lookup(net, saddr, sport, daddr, dport, dif, 0, NULL);
if (sk && !refcount_inc_not_zero(&sk->sk_refcnt))
sk = NULL;
return sk;
@@ -866,7 +863,6 @@ static int __udp4_lib_err_encap_no_sk(struct sk_buff *skb, u32 info)
static struct sock *__udp4_lib_err_encap(struct net *net,
const struct iphdr *iph,
struct udphdr *uh,
- struct udp_table *udptable,
struct sock *sk,
struct sk_buff *skb, u32 info)
{
@@ -894,8 +890,7 @@ static struct sock *__udp4_lib_err_encap(struct net *net,
}
sk = __udp4_lib_lookup(net, iph->daddr, uh->source,
- iph->saddr, uh->dest, skb->dev->ifindex, 0,
- udptable, NULL);
+ iph->saddr, uh->dest, skb->dev->ifindex, 0, NULL);
if (sk) {
up = udp_sk(sk);
@@ -924,29 +919,28 @@ static struct sock *__udp4_lib_err_encap(struct net *net,
* header points to the first 8 bytes of the udp header. We need
* to find the appropriate port.
*/
-
-static int __udp4_lib_err(struct sk_buff *skb, u32 info, struct udp_table *udptable)
+int udp_err(struct sk_buff *skb, u32 info)
{
- struct inet_sock *inet;
const struct iphdr *iph = (const struct iphdr *)skb->data;
- struct udphdr *uh = (struct udphdr *)(skb->data+(iph->ihl<<2));
const int type = icmp_hdr(skb)->type;
const int code = icmp_hdr(skb)->code;
+ struct net *net = dev_net(skb->dev);
+ struct inet_sock *inet;
bool tunnel = false;
+ struct udphdr *uh;
struct sock *sk;
int harderr;
int err;
- struct net *net = dev_net(skb->dev);
+ uh = (struct udphdr *)(skb->data + (iph->ihl << 2));
sk = __udp4_lib_lookup(net, iph->daddr, uh->dest,
iph->saddr, uh->source, skb->dev->ifindex,
- inet_sdif(skb), udptable, NULL);
+ inet_sdif(skb), NULL);
if (!sk || READ_ONCE(udp_sk(sk)->encap_type)) {
/* No socket for error: try tunnels before discarding */
if (static_branch_unlikely(&udp_encap_needed_key)) {
- sk = __udp4_lib_err_encap(net, iph, uh, udptable, sk, skb,
- info);
+ sk = __udp4_lib_err_encap(net, iph, uh, sk, skb, info);
if (!sk)
return 0;
} else
@@ -1019,11 +1013,6 @@ static int __udp4_lib_err(struct sk_buff *skb, u32 info, struct udp_table *udpta
return 0;
}
-int udp_err(struct sk_buff *skb, u32 info)
-{
- return __udp4_lib_err(skb, info, dev_net(skb->dev)->ipv4.udp_table);
-}
-
/*
* Throw away all pending data and cancel the corking. Socket is locked.
*/
@@ -2486,18 +2475,24 @@ EXPORT_IPV6_MOD(udp_sk_rx_dst_set);
static int __udp4_lib_mcast_deliver(struct net *net, struct sk_buff *skb,
struct udphdr *uh,
__be32 saddr, __be32 daddr,
- struct udp_table *udptable,
int proto)
{
- struct sock *sk, *first = NULL;
+ struct udp_table *udptable = net->ipv4.udp_table;
+ unsigned int hash2, hash2_any, offset;
unsigned short hnum = ntohs(uh->dest);
- struct udp_hslot *hslot = udp_hashslot(udptable, net, hnum);
- unsigned int hash2 = 0, hash2_any = 0, use_hash2 = (hslot->count > 10);
- unsigned int offset = offsetof(typeof(*sk), sk_node);
+ struct sock *sk, *first = NULL;
int dif = skb->dev->ifindex;
int sdif = inet_sdif(skb);
struct hlist_node *node;
+ struct udp_hslot *hslot;
struct sk_buff *nskb;
+ bool use_hash2;
+
+ hash2_any = 0;
+ hash2 = 0;
+ hslot = udp_hashslot(udptable, net, hnum);
+ use_hash2 = hslot->count > 10;
+ offset = offsetof(typeof(*sk), sk_node);
if (use_hash2) {
hash2_any = ipv4_portaddr_hash(net, htonl(INADDR_ANY), hnum) &
@@ -2602,15 +2597,14 @@ static int udp_unicast_rcv_skb(struct sock *sk, struct sk_buff *skb,
* All we need to do is get the socket, and then do a checksum.
*/
-static int __udp4_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,
- int proto)
+static int __udp4_lib_rcv(struct sk_buff *skb, int proto)
{
+ struct rtable *rt = skb_rtable(skb);
+ struct net *net = dev_net(skb->dev);
struct sock *sk = NULL;
- struct udphdr *uh;
unsigned short ulen;
- struct rtable *rt = skb_rtable(skb);
__be32 saddr, daddr;
- struct net *net = dev_net(skb->dev);
+ struct udphdr *uh;
bool refcounted;
int drop_reason;
@@ -2660,10 +2654,9 @@ static int __udp4_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,
}
if (rt->rt_flags & (RTCF_BROADCAST|RTCF_MULTICAST))
- return __udp4_lib_mcast_deliver(net, skb, uh,
- saddr, daddr, udptable, proto);
+ return __udp4_lib_mcast_deliver(net, skb, uh, saddr, daddr, proto);
- sk = __udp4_lib_lookup_skb(skb, uh->source, uh->dest, udptable);
+ sk = __udp4_lib_lookup_skb(skb, uh->source, uh->dest);
if (sk)
return udp_unicast_rcv_skb(sk, skb, uh);
no_sk:
@@ -2847,7 +2840,7 @@ enum skb_drop_reason udp_v4_early_demux(struct sk_buff *skb)
int udp_rcv(struct sk_buff *skb)
{
- return __udp4_lib_rcv(skb, dev_net(skb->dev)->ipv4.udp_table, IPPROTO_UDP);
+ return __udp4_lib_rcv(skb, IPPROTO_UDP);
}
static void udp_destroy_sock(struct sock *sk)
diff --git a/net/ipv4/udp_diag.c b/net/ipv4/udp_diag.c
index 0899c60cce53..f4b24e628cf8 100644
--- a/net/ipv4/udp_diag.c
+++ b/net/ipv4/udp_diag.c
@@ -24,23 +24,24 @@ static int sk_diag_dump(struct sock *sk, struct sk_buff *skb,
net_admin);
}
-static int udp_dump_one(struct udp_table *tbl,
- struct netlink_callback *cb,
- const struct inet_diag_req_v2 *req)
+static int udp_diag_dump_one(struct netlink_callback *cb,
+ const struct inet_diag_req_v2 *req)
{
struct sk_buff *in_skb = cb->skb;
- int err;
struct sock *sk = NULL;
struct sk_buff *rep;
- struct net *net = sock_net(in_skb->sk);
+ struct net *net;
+ int err;
+
+ net = sock_net(in_skb->sk);
rcu_read_lock();
if (req->sdiag_family == AF_INET)
/* src and dst are swapped for historical reasons */
sk = __udp4_lib_lookup(net,
- req->id.idiag_src[0], req->id.idiag_sport,
- req->id.idiag_dst[0], req->id.idiag_dport,
- req->id.idiag_if, 0, tbl, NULL);
+ req->id.idiag_src[0], req->id.idiag_sport,
+ req->id.idiag_dst[0], req->id.idiag_dport,
+ req->id.idiag_if, 0, NULL);
#if IS_ENABLED(CONFIG_IPV6)
else if (req->sdiag_family == AF_INET6)
sk = __udp6_lib_lookup(net,
@@ -85,14 +86,15 @@ static int udp_dump_one(struct udp_table *tbl,
return err;
}
-static void udp_dump(struct udp_table *table, struct sk_buff *skb,
- struct netlink_callback *cb,
- const struct inet_diag_req_v2 *r)
+static void udp_diag_dump(struct sk_buff *skb, struct netlink_callback *cb,
+ const struct inet_diag_req_v2 *r)
{
bool net_admin = netlink_net_capable(cb->skb, CAP_NET_ADMIN);
struct net *net = sock_net(skb->sk);
int num, s_num, slot, s_slot;
+ struct udp_table *table;
+ table = net->ipv4.udp_table;
s_slot = cb->args[0];
num = s_num = cb->args[1];
@@ -139,18 +141,6 @@ static void udp_dump(struct udp_table *table, struct sk_buff *skb,
cb->args[1] = num;
}
-static void udp_diag_dump(struct sk_buff *skb, struct netlink_callback *cb,
- const struct inet_diag_req_v2 *r)
-{
- udp_dump(sock_net(cb->skb->sk)->ipv4.udp_table, skb, cb, r);
-}
-
-static int udp_diag_dump_one(struct netlink_callback *cb,
- const struct inet_diag_req_v2 *req)
-{
- return udp_dump_one(sock_net(cb->skb->sk)->ipv4.udp_table, cb, req);
-}
-
static void udp_diag_get_info(struct sock *sk, struct inet_diag_msg *r,
void *info)
{
@@ -159,9 +149,8 @@ static void udp_diag_get_info(struct sock *sk, struct inet_diag_msg *r,
}
#ifdef CONFIG_INET_DIAG_DESTROY
-static int __udp_diag_destroy(struct sk_buff *in_skb,
- const struct inet_diag_req_v2 *req,
- struct udp_table *tbl)
+static int udp_diag_destroy(struct sk_buff *in_skb,
+ const struct inet_diag_req_v2 *req)
{
struct net *net = sock_net(in_skb->sk);
struct sock *sk;
@@ -171,18 +160,17 @@ static int __udp_diag_destroy(struct sk_buff *in_skb,
if (req->sdiag_family == AF_INET)
sk = __udp4_lib_lookup(net,
- req->id.idiag_dst[0], req->id.idiag_dport,
- req->id.idiag_src[0], req->id.idiag_sport,
- req->id.idiag_if, 0, tbl, NULL);
+ req->id.idiag_dst[0], req->id.idiag_dport,
+ req->id.idiag_src[0], req->id.idiag_sport,
+ req->id.idiag_if, 0, NULL);
#if IS_ENABLED(CONFIG_IPV6)
else if (req->sdiag_family == AF_INET6) {
if (ipv6_addr_v4mapped((struct in6_addr *)req->id.idiag_dst) &&
ipv6_addr_v4mapped((struct in6_addr *)req->id.idiag_src))
sk = __udp4_lib_lookup(net,
- req->id.idiag_dst[3], req->id.idiag_dport,
- req->id.idiag_src[3], req->id.idiag_sport,
- req->id.idiag_if, 0, tbl, NULL);
-
+ req->id.idiag_dst[3], req->id.idiag_dport,
+ req->id.idiag_src[3], req->id.idiag_sport,
+ req->id.idiag_if, 0, NULL);
else
sk = __udp6_lib_lookup(net,
(struct in6_addr *)req->id.idiag_dst,
@@ -216,13 +204,6 @@ static int __udp_diag_destroy(struct sk_buff *in_skb,
return err;
}
-
-static int udp_diag_destroy(struct sk_buff *in_skb,
- const struct inet_diag_req_v2 *req)
-{
- return __udp_diag_destroy(in_skb, req, sock_net(in_skb->sk)->ipv4.udp_table);
-}
-
#endif
static const struct inet_diag_handler udp_diag_handler = {
diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c
index 6b1654c1ad4a..98e92da726b5 100644
--- a/net/ipv4/udp_offload.c
+++ b/net/ipv4/udp_offload.c
@@ -869,8 +869,7 @@ static struct sock *udp4_gro_lookup_skb(struct sk_buff *skb, __be16 sport,
inet_get_iif_sdif(skb, &iif, &sdif);
return __udp4_lib_lookup(net, iph->saddr, sport,
- iph->daddr, dport, iif,
- sdif, net->ipv4.udp_table, NULL);
+ iph->daddr, dport, iif, sdif, NULL);
}
INDIRECT_CALLABLE_SCOPE
--
2.53.0.473.g4a7958ca14-goog
^ permalink raw reply related [flat|nested] 19+ messages in thread
* [PATCH v2 net-next 15/15] udp: Don't pass proto to __udp4_lib_rcv() and __udp6_lib_rcv().
2026-03-05 21:49 [PATCH v2 net-next 00/15] udp: Retire UDP-Lite Kuniyuki Iwashima
` (13 preceding siblings ...)
2026-03-05 21:50 ` [PATCH v2 net-next 14/15] udp: Don't pass udptable to IPv4 " Kuniyuki Iwashima
@ 2026-03-05 21:50 ` Kuniyuki Iwashima
14 siblings, 0 replies; 19+ messages in thread
From: Kuniyuki Iwashima @ 2026-03-05 21:50 UTC (permalink / raw)
To: Willem de Bruijn, David Ahern, David S . Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni
Cc: Simon Horman, Florian Westphal, Kuniyuki Iwashima,
Kuniyuki Iwashima, netdev, Willem de Bruijn
UDP and UDP-Lite shared __udp4_lib_rcv() and __udp6_lib_rcv()
by passing IPPROTO_UDP or IPPROTO_UDPLITE.
Now, @proto is always IPPROTO_UDP.
Let's not pass it and rename the functions accordingly.
With this series removing a bunch of conditionals for UDP-Lite
from the fast path, udp_rr with 20,000 flows sees a 10% increase
in pps (13.3 Mpps -> 14.7 Mpps) on an AMD EPYC 7B12 (Zen 2)
64-Core Processor platform.
[ With FDO, the baseline is much higher and the delta was ~3%,
20.1 Mpps -> 20.7 Mpps ]
Before:
$ nstat > /dev/null; sleep 1; nstat | grep Udp
Udp6InDatagrams 14013408 0.0
Udp6OutDatagrams 14013128 0.0
After:
$ nstat > /dev/null; sleep 1; nstat | grep Udp
Udp6InDatagrams 15491971 0.0
Udp6OutDatagrams 15491671 0.0
$ ./scripts/bloat-o-meter vmlinux.before vmlinux.after
add/remove: 13/75 grow/shrink: 11/75 up/down: 13777/-18401 (-4624)
Function old new delta
udp4_gro_receive 872 866 -6
udp6_gro_receive 910 903 -7
udp_rcv 32 1727 +1695
udpv6_rcv 32 1450 +1418
__udp4_lib_rcv 2045 - -2045
__udp6_lib_rcv 2084 - -2084
udp_unicast_rcv_skb 160 149 -11
udp6_unicast_rcv_skb 196 181 -15
__udp4_lib_mcast_deliver 925 846 -79
__udp6_lib_mcast_deliver 922 810 -112
__udp4_lib_lookup 973 969 -4
__udp6_lib_lookup 940 929 -11
__udp4_lib_lookup_skb 106 100 -6
__udp6_lib_lookup_skb 71 66 -5
udp4_lib_lookup_skb 132 127 -5
udp6_lib_lookup_skb 87 81 -6
udp_queue_rcv_skb 326 356 +30
udpv6_queue_rcv_skb 331 361 +30
udp_queue_rcv_one_skb 1233 914 -319
udpv6_queue_rcv_one_skb 1250 930 -320
__udp_enqueue_schedule_skb 1067 995 -72
udp_rcv_segment 520 480 -40
udp_post_segment_fix_csum 120 - -120
udp_lib_checksum_complete 200 84 -116
udp_err 27 1103 +1076
udpv6_err 36 1417 +1381
__udp4_lib_err 1112 - -1112
__udp6_lib_err 1448 - -1448
udp_recvmsg 1149 994 -155
udpv6_recvmsg 1349 1294 -55
udp_sendmsg 2730 2648 -82
udp_send_skb 909 681 -228
udpv6_sendmsg 3022 2861 -161
udp_v6_send_skb 1214 952 -262
...
Total: Before=18446744073748075501, After=18446744073748070877, chg -0.00%
Signed-off-by: Kuniyuki Iwashima <kuniyu@google.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
---
net/ipv4/udp.c | 18 +++++-------------
net/ipv6/udp.c | 15 ++++-----------
2 files changed, 9 insertions(+), 24 deletions(-)
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 0436462cb9fd..d39dcceb8ba7 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -2474,8 +2474,7 @@ EXPORT_IPV6_MOD(udp_sk_rx_dst_set);
*/
static int __udp4_lib_mcast_deliver(struct net *net, struct sk_buff *skb,
struct udphdr *uh,
- __be32 saddr, __be32 daddr,
- int proto)
+ __be32 saddr, __be32 daddr)
{
struct udp_table *udptable = net->ipv4.udp_table;
unsigned int hash2, hash2_any, offset;
@@ -2597,7 +2596,7 @@ static int udp_unicast_rcv_skb(struct sock *sk, struct sk_buff *skb,
* All we need to do is get the socket, and then do a checksum.
*/
-static int __udp4_lib_rcv(struct sk_buff *skb, int proto)
+int udp_rcv(struct sk_buff *skb)
{
struct rtable *rt = skb_rtable(skb);
struct net *net = dev_net(skb->dev);
@@ -2654,7 +2653,7 @@ static int __udp4_lib_rcv(struct sk_buff *skb, int proto)
}
if (rt->rt_flags & (RTCF_BROADCAST|RTCF_MULTICAST))
- return __udp4_lib_mcast_deliver(net, skb, uh, saddr, daddr, proto);
+ return __udp4_lib_mcast_deliver(net, skb, uh, saddr, daddr);
sk = __udp4_lib_lookup_skb(skb, uh->source, uh->dest);
if (sk)
@@ -2681,8 +2680,7 @@ static int __udp4_lib_rcv(struct sk_buff *skb, int proto)
short_packet:
drop_reason = SKB_DROP_REASON_PKT_TOO_SMALL;
- net_dbg_ratelimited("UDP%s: short packet: From %pI4:%u %d/%d to %pI4:%u\n",
- proto == IPPROTO_UDPLITE ? "Lite" : "",
+ net_dbg_ratelimited("UDP: short packet: From %pI4:%u %d/%d to %pI4:%u\n",
&saddr, ntohs(uh->source),
ulen, skb->len,
&daddr, ntohs(uh->dest));
@@ -2694,8 +2692,7 @@ static int __udp4_lib_rcv(struct sk_buff *skb, int proto)
* the network is concerned, anyway) as per 4.1.3.4 (MUST).
*/
drop_reason = SKB_DROP_REASON_UDP_CSUM;
- net_dbg_ratelimited("UDP%s: bad checksum. From %pI4:%u to %pI4:%u ulen %d\n",
- proto == IPPROTO_UDPLITE ? "Lite" : "",
+ net_dbg_ratelimited("UDP: bad checksum. From %pI4:%u to %pI4:%u ulen %d\n",
&saddr, ntohs(uh->source), &daddr, ntohs(uh->dest),
ulen);
__UDP_INC_STATS(net, UDP_MIB_CSUMERRORS);
@@ -2838,11 +2835,6 @@ enum skb_drop_reason udp_v4_early_demux(struct sk_buff *skb)
return SKB_NOT_DROPPED_YET;
}
-int udp_rcv(struct sk_buff *skb)
-{
- return __udp4_lib_rcv(skb, IPPROTO_UDP);
-}
-
static void udp_destroy_sock(struct sock *sk)
{
struct udp_sock *up = udp_sk(sk);
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 0de28cbfe793..12a1004c4d7f 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -934,8 +934,7 @@ static void udp6_csum_zero_error(struct sk_buff *skb)
*/
static int __udp6_lib_mcast_deliver(struct net *net, struct sk_buff *skb,
const struct in6_addr *saddr,
- const struct in6_addr *daddr,
- int proto)
+ const struct in6_addr *daddr)
{
struct udp_table *udptable = net->ipv4.udp_table;
const struct udphdr *uh = udp_hdr(skb);
@@ -1063,7 +1062,7 @@ static int udp6_csum_init(struct sk_buff *skb, struct udphdr *uh)
return 0;
}
-static int __udp6_lib_rcv(struct sk_buff *skb, int proto)
+INDIRECT_CALLABLE_SCOPE int udpv6_rcv(struct sk_buff *skb)
{
enum skb_drop_reason reason = SKB_DROP_REASON_NOT_SPECIFIED;
const struct in6_addr *saddr, *daddr;
@@ -1133,7 +1132,7 @@ static int __udp6_lib_rcv(struct sk_buff *skb, int proto)
* Multicast receive code
*/
if (ipv6_addr_is_multicast(daddr))
- return __udp6_lib_mcast_deliver(net, skb, saddr, daddr, proto);
+ return __udp6_lib_mcast_deliver(net, skb, saddr, daddr);
/* Unicast */
sk = __udp6_lib_lookup_skb(skb, uh->source, uh->dest);
@@ -1164,8 +1163,7 @@ static int __udp6_lib_rcv(struct sk_buff *skb, int proto)
short_packet:
if (reason == SKB_DROP_REASON_NOT_SPECIFIED)
reason = SKB_DROP_REASON_PKT_TOO_SMALL;
- net_dbg_ratelimited("UDP%sv6: short packet: From [%pI6c]:%u %d/%d to [%pI6c]:%u\n",
- proto == IPPROTO_UDPLITE ? "-Lite" : "",
+ net_dbg_ratelimited("UDPv6: short packet: From [%pI6c]:%u %d/%d to [%pI6c]:%u\n",
saddr, ntohs(uh->source),
ulen, skb->len,
daddr, ntohs(uh->dest));
@@ -1252,11 +1250,6 @@ void udp_v6_early_demux(struct sk_buff *skb)
}
}
-INDIRECT_CALLABLE_SCOPE int udpv6_rcv(struct sk_buff *skb)
-{
- return __udp6_lib_rcv(skb, IPPROTO_UDP);
-}
-
/*
* Throw away all pending data and cancel the corking. Socket is locked.
*/
--
2.53.0.473.g4a7958ca14-goog
^ permalink raw reply related [flat|nested] 19+ messages in thread
* Re: [v2,net-next,07/15] udp: Remove partial csum code in RX.
2026-03-05 21:49 ` [PATCH v2 net-next 07/15] udp: Remove partial csum code in RX Kuniyuki Iwashima
@ 2026-03-10 10:25 ` Paolo Abeni
2026-03-10 16:17 ` Kuniyuki Iwashima
2026-03-10 20:00 ` [PATCH v2 net-next 07/15] " David Laight
1 sibling, 1 reply; 19+ messages in thread
From: Paolo Abeni @ 2026-03-10 10:25 UTC (permalink / raw)
To: kuniyu
Cc: willemb, serge, linux-security-module, netdev, davem, edumazet,
kuba, pabeni, dsahern, jmorris, casey, paul, kuni1840, fw, horms,
willemdebruijn.kernel
This is an AI-generated review of your patch. The human sending this
email has considered the AI review valid, or at least plausible.
For local reproduction steps see https://netdev-ai.bots.linux.dev/ai-local.html
---
udp: Remove partial csum code in RX.
This commit removes UDP-Lite specific partial checksum handling from UDP
receive paths, including removal of UDP_SKB_CB fields, simplification of
checksum initialization functions, and elimination of UDP-Lite guards.
Link: https://lore.kernel.org/netdev/20260305215013.2984628-8-kuniyu@google.com/
> diff --git a/include/net/udp.h b/include/net/udp.h
> --- a/include/net/udp.h
> +++ b/include/net/udp.h
[ ... ]
> diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
> --- a/net/ipv4/udp.c
> +++ b/net/ipv4/udp.c
[ ... ]
> @@ -2703,14 +2647,15 @@ static int __udp4_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,
> if (ulen > skb->len)
> goto short_packet;
>
> - if (proto == IPPROTO_UDP) {
> - /* UDP validates ulen. */
> - if (ulen < sizeof(*uh) || pskb_trim_rcsum(skb, ulen))
> + /* UDP validates ulen. */
> + if (ulen < sizeof(*uh)) {
> + if (pskb_trim_rcsum(skb, ulen))
> goto short_packet;
> +
> uh = udp_hdr(skb);
> }
>
> - if (udp4_csum_init(skb, uh, proto))
> + if (udp4_csum_init(skb, uh))
> goto csum_error;
>
Does this refactoring introduce two bugs?
The old code checked both conditions with short-circuit OR:
if (ulen < sizeof(*uh) || pskb_trim_rcsum(skb, ulen))
goto short_packet;
This correctly (1) rejected packets where ulen < 8, and (2) trimmed the
SKB to the declared UDP length when ulen < skb->len.
The new code changes the logic structure:
if (ulen < sizeof(*uh)) {
if (pskb_trim_rcsum(skb, ulen))
goto short_packet;
uh = udp_hdr(skb);
}
First potential bug: When sizeof(*uh) <= ulen < skb->len, the entire block
is skipped. This means pskb_trim_rcsum() is never called, so the SKB
retains trailing data beyond the declared UDP length. Won't this cause
checksum calculations to include extraneous bytes, leading to valid packets
being incorrectly rejected?
This scenario occurs when UDP length < IP payload length, which can happen
with malformed or crafted packets.
Second potential bug: When ulen < sizeof(*uh), the old code would
immediately goto short_packet via the || short-circuit. The new code enters
the block and tries to trim the SKB to a sub-header-size length. If
pskb_trim_rcsum() succeeds, processing continues with skb->len <
sizeof(struct udphdr). Is this safe?
Looking at the IPv6 counterpart __udp6_lib_rcv(), it handles this
differently:
if (ulen < sizeof(*uh))
goto short_packet;
if (ulen < skb->len) {
if (pskb_trim_rcsum(skb, ulen))
goto short_packet;
saddr = &ipv6_hdr(skb)->saddr;
daddr = &ipv6_hdr(skb)->daddr;
uh = udp_hdr(skb);
}
Should the IPv4 code follow the same pattern, or should it preserve the
original short-circuit logic:
if (ulen < sizeof(*uh) || pskb_trim_rcsum(skb, ulen))
goto short_packet;
uh = udp_hdr(skb);
^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [v2,net-next,07/15] udp: Remove partial csum code in RX.
2026-03-10 10:25 ` [v2,net-next,07/15] " Paolo Abeni
@ 2026-03-10 16:17 ` Kuniyuki Iwashima
0 siblings, 0 replies; 19+ messages in thread
From: Kuniyuki Iwashima @ 2026-03-10 16:17 UTC (permalink / raw)
To: Paolo Abeni
Cc: willemb, serge, linux-security-module, netdev, davem, edumazet,
kuba, dsahern, jmorris, casey, paul, kuni1840, fw, horms,
willemdebruijn.kernel
On Tue, Mar 10, 2026 at 3:26 AM Paolo Abeni <pabeni@redhat.com> wrote:
>
> This is an AI-generated review of your patch. The human sending this
> email has considered the AI review valid, or at least plausible.
>
> For local reproduction steps see https://netdev-ai.bots.linux.dev/ai-local.html
> ---
> udp: Remove partial csum code in RX.
>
> This commit removes UDP-Lite specific partial checksum handling from UDP
> receive paths, including removal of UDP_SKB_CB fields, simplification of
> checksum initialization functions, and elimination of UDP-Lite guards.
>
> Link: https://lore.kernel.org/netdev/20260305215013.2984628-8-kuniyu@google.com/
>
> > diff --git a/include/net/udp.h b/include/net/udp.h
> > --- a/include/net/udp.h
> > +++ b/include/net/udp.h
>
> [ ... ]
>
> > diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
> > --- a/net/ipv4/udp.c
> > +++ b/net/ipv4/udp.c
>
> [ ... ]
>
> > @@ -2703,14 +2647,15 @@ static int __udp4_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,
> > if (ulen > skb->len)
> > goto short_packet;
> >
> > - if (proto == IPPROTO_UDP) {
> > - /* UDP validates ulen. */
> > - if (ulen < sizeof(*uh) || pskb_trim_rcsum(skb, ulen))
> > + /* UDP validates ulen. */
> > + if (ulen < sizeof(*uh)) {
> > + if (pskb_trim_rcsum(skb, ulen))
> > goto short_packet;
> > +
> > uh = udp_hdr(skb);
> > }
> >
> > - if (udp4_csum_init(skb, uh, proto))
> > + if (udp4_csum_init(skb, uh))
> > goto csum_error;
> >
>
> Does this refactoring introduce two bugs?
>
> The old code checked both conditions with short-circuit OR:
> if (ulen < sizeof(*uh) || pskb_trim_rcsum(skb, ulen))
> goto short_packet;
>
> This correctly (1) rejected packets where ulen < 8, and (2) trimmed the
> SKB to the declared UDP length when ulen < skb->len.
>
> The new code changes the logic structure:
> if (ulen < sizeof(*uh)) {
> if (pskb_trim_rcsum(skb, ulen))
> goto short_packet;
> uh = udp_hdr(skb);
> }
>
> First potential bug: When sizeof(*uh) <= ulen < skb->len, the entire block
> is skipped. This means pskb_trim_rcsum() is never called, so the SKB
> retains trailing data beyond the declared UDP length. Won't this cause
> checksum calculations to include extraneous bytes, leading to valid packets
> being incorrectly rejected?
>
> This scenario occurs when UDP length < IP payload length, which can happen
> with malformed or crafted packets.
>
> Second potential bug: When ulen < sizeof(*uh), the old code would
> immediately goto short_packet via the || short-circuit. The new code enters
> the block and tries to trim the SKB to a sub-header-size length. If
> pskb_trim_rcsum() succeeds, processing continues with skb->len <
> sizeof(struct udphdr). Is this safe?
>
> Looking at the IPv6 counterpart __udp6_lib_rcv(), it handles this
> differently:
>
> if (ulen < sizeof(*uh))
> goto short_packet;
>
> if (ulen < skb->len) {
> if (pskb_trim_rcsum(skb, ulen))
> goto short_packet;
> saddr = &ipv6_hdr(skb)->saddr;
> daddr = &ipv6_hdr(skb)->daddr;
> uh = udp_hdr(skb);
> }
>
> Should the IPv4 code follow the same pattern, or should it preserve the
> original short-circuit logic:
Indeed, I'll follow the IPv6 pattern there.
Thanks !
^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [PATCH v2 net-next 07/15] udp: Remove partial csum code in RX.
2026-03-05 21:49 ` [PATCH v2 net-next 07/15] udp: Remove partial csum code in RX Kuniyuki Iwashima
2026-03-10 10:25 ` [v2,net-next,07/15] " Paolo Abeni
@ 2026-03-10 20:00 ` David Laight
1 sibling, 0 replies; 19+ messages in thread
From: David Laight @ 2026-03-10 20:00 UTC (permalink / raw)
To: Kuniyuki Iwashima
Cc: Willem de Bruijn, David Ahern, David S . Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, Simon Horman, Florian Westphal,
Kuniyuki Iwashima, netdev, Willem de Bruijn
On Thu, 5 Mar 2026 21:49:53 +0000
Kuniyuki Iwashima <kuniyu@google.com> wrote:
> UDP-Lite supports the partial checksum and the coverage is
> stored in the position of the length field of struct udphdr.
>
> In RX paths, udp4_csum_init() / udp6_csum_init() save the value
> in UDP_SKB_CB(skb)->cscov and set UDP_SKB_CB(skb)->partial_cov
> to 1 if the coverage is not full.
>
Is it worth completely removing the code that defers the udp RX checksum
until the copy_to_user() and the corresponding TX side code that checksums
the data being sent.
I'm not sure what the 'killer application' was for the RX side.
I'd guess the userspace NFS daemon - long since dead.
With most modern 'performance' NICs doing checksum offload you don't
want to 'waste' cycles doing a checksum during transmit that won't
be needed - and you don't know the interface until much later.
I also recall that pretty much only x86 actually tries to do the
checksum and copy together - most architectures checksum the buffer
fragments (of the iov[]) as they are copied - even though the kernel
buffer is linear; so has to handle more fragments.
ISTR there is a bug where 0 and ~0 get swapped.
Even for x86-64 the fastest IP checksum does 12-16 bytes/clock (I think
you can get 12 on Intel and 16 on AMD), and the copy will do 64 bytes
per clock (on recent enough cpu and once it is started and with an
aligned destination).
So even there separate copy and checksum may be faster.
The only time copy+checksum really helps is for buffers that spill
the L1-cache - which is why I think NFS's 8k UDP datagrams are why
the code was added at all.
(Last time I asked about this Linus couldn't remember.)
David
^ permalink raw reply [flat|nested] 19+ messages in thread
end of thread, other threads:[~2026-03-10 20:00 UTC | newest]
Thread overview: 19+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-03-05 21:49 [PATCH v2 net-next 00/15] udp: Retire UDP-Lite Kuniyuki Iwashima
2026-03-05 21:49 ` [PATCH v2 net-next 01/15] udp: Make udp[46]_seq_show() static Kuniyuki Iwashima
2026-03-05 21:49 ` [PATCH v2 net-next 02/15] ipv6: Retire UDP-Lite Kuniyuki Iwashima
2026-03-05 21:49 ` [PATCH v2 net-next 03/15] ipv6: Remove UDP-Lite support for IPV6_ADDRFORM Kuniyuki Iwashima
2026-03-05 21:49 ` [PATCH v2 net-next 04/15] ipv4: Retire UDP-Lite Kuniyuki Iwashima
2026-03-05 21:49 ` [PATCH v2 net-next 05/15] udp: Remove UDP-Lite SNMP stats Kuniyuki Iwashima
2026-03-05 21:49 ` [PATCH v2 net-next 06/15] smack: Remove IPPROTO_UDPLITE support in security_sock_rcv_skb() Kuniyuki Iwashima
2026-03-05 21:49 ` [PATCH v2 net-next 07/15] udp: Remove partial csum code in RX Kuniyuki Iwashima
2026-03-10 10:25 ` [v2,net-next,07/15] " Paolo Abeni
2026-03-10 16:17 ` Kuniyuki Iwashima
2026-03-10 20:00 ` [PATCH v2 net-next 07/15] " David Laight
2026-03-05 21:49 ` [PATCH v2 net-next 08/15] udp: Remove partial csum code in TX Kuniyuki Iwashima
2026-03-05 21:49 ` [PATCH v2 net-next 09/15] udp: Remove UDPLITE_SEND_CSCOV and UDPLITE_RECV_CSCOV Kuniyuki Iwashima
2026-03-05 21:49 ` [PATCH v2 net-next 10/15] udp: Remove struct proto.h.udp_table Kuniyuki Iwashima
2026-03-05 21:49 ` [PATCH v2 net-next 11/15] udp: Remove udp_table in struct udp_seq_afinfo Kuniyuki Iwashima
2026-03-05 21:49 ` [PATCH v2 net-next 12/15] udp: Remove dead check in __udp[46]_lib_lookup() for BPF Kuniyuki Iwashima
2026-03-05 21:49 ` [PATCH v2 net-next 13/15] udp: Don't pass udptable to IPv6 socket lookup functions Kuniyuki Iwashima
2026-03-05 21:50 ` [PATCH v2 net-next 14/15] udp: Don't pass udptable to IPv4 " Kuniyuki Iwashima
2026-03-05 21:50 ` [PATCH v2 net-next 15/15] udp: Don't pass proto to __udp4_lib_rcv() and __udp6_lib_rcv() Kuniyuki Iwashima
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox