kernel/linux.git/net/ipv6/udp.c, branch linux-7.1.y

udp: Force compute_score to always inline

2026-04-13T22:44:42+00:00

Back in 2024 I reported a 7-12% regression on an iperf3 UDP loopback thoughput test that we traced to the extra overhead of calling compute_score on two places, introduced by commit f0ea27e7bfe1 ("udp: re-score reuseport groups when connected sockets are present"). At the time, I pointed out the overhead was caused by the multiple calls, associated with cpu-specific mitigations, and merged commit 50aee97d1511 ("udp: Avoid call to compute_score on multiple sites") to jump back explicitly, to force the rescore call in a single place. Recently though, we got another regression report against a newer distro version, which a team colleague traced back to the same root-cause. Turns out that once we updated to gcc-13, the compiler got smart enough to unroll the loop, undoing my previous mitigation. Let's bite the bullet and __always_inline compute_score on both ipv4 and ipv6 to prevent gcc from de-optimizing it again in the future. These functions are only called in two places each, udpX_lib_lookup1 and udpX_lib_lookup2, so the extra size shouldn't be a problem and it is hot enough to be very visible in profilings. In fact, with gcc13, forcing the inline will prevent gcc from unrolling the fix from commit 50aee97d1511, so we don't end up increasing udpX_lib_lookup2 at all. I haven't recollected the results myself, as I don't have access to the machine at the moment. But the same colleague reported 4.67% inprovement with this patch in the loopback benchmark, solving the regression report within noise margins. Eric Dumazet reported no size change to vmlinux when built with clang. I report the same also with gcc-13: scripts/bloat-o-meter vmlinux vmlinux-inline add/remove: 0/2 grow/shrink: 4/0 up/down: 616/-416 (200) Function old new delta udp6_lib_lookup2 762 949 +187 __udp6_lib_lookup 810 975 +165 udp4_lib_lookup2 757 906 +149 __udp4_lib_lookup 871 986 +115 __pfx_compute_score 32 - -32 compute_score 384 - -384 Total: Before=35011784, After=35011984, chg +0.00% Fixes: 50aee97d1511 ("udp: Avoid call to compute_score on multiple sites") Reviewed-by: Eric Dumazet Acked-by: Willem de Bruijn Signed-off-by: Gabriel Krisman Bertazi Link: https://patch.msgid.link/20260410155936.654915-1-krisman@suse.de Signed-off-by: Jakub Kicinski

net: change sk_filter_trim_cap() to return a drop_reason by value

2026-04-12T21:30:25+00:00

Current return value can be replaced with the drop_reason, reducing kernel bloat: $ scripts/bloat-o-meter -t vmlinux.old vmlinux.new add/remove: 0/2 grow/shrink: 1/11 up/down: 32/-603 (-571) Function old new delta tcp_v6_rcv 3135 3167 +32 unix_dgram_sendmsg 1731 1726 -5 netlink_unicast 957 945 -12 netlink_dump 1372 1359 -13 sk_filter_trim_cap 882 858 -24 tcp_v4_rcv 3143 3111 -32 __pfx_tcp_filter 32 - -32 netlink_broadcast_filtered 1633 1595 -38 sock_queue_rcv_skb_reason 126 76 -50 tun_net_xmit 1127 1074 -53 __sk_receive_skb 690 632 -58 udpv6_queue_rcv_one_skb 935 869 -66 udp_queue_rcv_one_skb 919 853 -66 tcp_filter 154 - -154 Total: Before=29722783, After=29722212, chg -0.00% Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20260409145625.2306224-6-edumazet@google.com Signed-off-by: Jakub Kicinski

udp: Don't pass proto to __udp4_lib_rcv() and __udp6_lib_rcv().

2026-03-14T01:57:46+00:00

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 Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20260311052020.1213705-16-kuniyu@google.com Signed-off-by: Jakub Kicinski

udp: Don't pass udptable to IPv6 socket lookup functions.

2026-03-14T01:57:46+00:00

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 Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20260311052020.1213705-14-kuniyu@google.com Signed-off-by: Jakub Kicinski

udp: Remove dead check in __udp[46]_lib_lookup() for BPF.

2026-03-14T01:57:45+00:00

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 Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20260311052020.1213705-13-kuniyu@google.com Signed-off-by: Jakub Kicinski

udp: Remove udp_table in struct udp_seq_afinfo.

2026-03-14T01:57:45+00:00

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 Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20260311052020.1213705-12-kuniyu@google.com Signed-off-by: Jakub Kicinski

udp: Remove struct proto.h.udp_table.

2026-03-14T01:57:45+00:00

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 Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20260311052020.1213705-11-kuniyu@google.com Signed-off-by: Jakub Kicinski

udp: Remove UDPLITE_SEND_CSCOV and UDPLITE_RECV_CSCOV.

2026-03-14T01:57:45+00:00

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 Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20260311052020.1213705-10-kuniyu@google.com Signed-off-by: Jakub Kicinski

udp: Remove partial csum code in TX.

2026-03-14T01:57:45+00:00

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 Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20260311052020.1213705-9-kuniyu@google.com Signed-off-by: Jakub Kicinski

udp: Remove partial csum code in RX.

2026-03-14T01:57:45+00:00

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 Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20260311052020.1213705-8-kuniyu@google.com Signed-off-by: Jakub Kicinski