Skip to content

RFC net and/or udp optimisations

Pavel Begunkov netdev 1 version In review

Get series with b4
b4 am [email protected]

Versions newest first

  1. v1 RFC Latest27 patches 2 repliesIn review

Patches in v1 No applies recorded yet

#Subject and filesSizeRepliesApplied
1/27sock: deduplicate ->sk_wmem_alloc checknet/core/sock.c+2 -30 replies
2/27sock: optimise sock_def_write_space send refcountingnet/core/sock.c+14 -00 replies
3/27sock: optimise sock_def_write_space barriersnet/core/sock.c+25 -10 replies
4/27skbuff: drop zero check from skb_zcopy_setinclude/linux/skbuff.h net/ipv4/ip_output.c net/ipv6/ip6_output.c+5 -30 replies
5/27skbuff: drop null check from skb_zcopyinclude/linux/skbuff.h net/core/dev.c net/core/skbuff.c +3 more+18 -80 replies
6/27net: xen: set zc flags only when there is ubufdrivers/net/xen-netback/interface.c+2 -10 replies
7/27skbuff: introduce skb_is_zcopy()include/linux/skbuff.h net/core/skbuff.c+22 -180 replies
8/27skbuff: optimise alloc_skb_with_frags()include/linux/skbuff.h net/core/skbuff.c+48 -240 replies
9/27net: inline sock_alloc_send_skbinclude/net/sock.h net/core/sock.c+8 -90 replies
10/27net: inline part of skb_csum_hwoffload_helpinclude/linux/netdevice.h net/core/dev.c+14 -100 replies
11/27net: inline skb_zerocopy_iter_dgraminclude/linux/skbuff.h net/core/datagram.c net/core/datagram.h +1 more+22 -380 replies
12/27ipv6: inline ip6_local_out()include/net/ipv6.h net/ipv6/output_core.c+12 -130 replies
13/27ipv6: help __ip6_finish_output() inliningnet/ipv6/ip6_output.c+0 -10 replies
14/27ipv6: refactor ip6_finish_output2()net/ipv6/ip6_output.c+13 -110 replies
15/27net: inline dev_queue_xmit()include/linux/netdevice.h net/core/dev.c+14 -150 replies
16/27ipv6: partially inline fl6_update_dst()include/net/ipv6.h net/ipv6/exthdrs.c+18 -120 replies
17/27tcp: optimise skb_zerocopy_iter_stream()net/core/skbuff.c+1 -20 replies
18/27net: optimise ipcm6 cookie initinclude/net/ipv6.h net/ipv6/datagram.c net/ipv6/ip6_output.c +3 more+5 -330 replies
19/27udp/ipv6: refactor udpv6_sendmsg udplite checksnet/ipv6/udp.c+3 -30 replies
20/27udp/ipv6: move pending section of udpv6_sendmsgnet/ipv6/udp.c+40 -270 replies
21/27udp/ipv6: prioritise the ip6 path over ip4 checksnet/ipv6/udp.c+17 -200 replies
22/27udp/ipv6: optimise udpv6_sendmsg() daddr checksnet/ipv6/udp.c+11 -120 replies
23/27udp/ipv6: optimise out daddr reassignmentnet/ipv6/udp.c+0 -80 replies
24/27udp/ipv6: clean up udpv6_sendmsg's saddr initnet/ipv6/udp.c+4 -60 replies
25/27ipv6: refactor opts push in __ip6_make_skb()net/ipv6/ip6_output.c+8 -50 replies
26/27ipv6: improve opt-less __ip6_make_skb()net/ipv6/ip6_output.c+3 -50 replies
27/27ipv6: clean up ip6_setup_corknet/ipv6/ip6_output.c+13 -170 replies

Reviews trailers per patch

No Reviewed-by, Acked-by or Tested-by trailers for v1 yet.

Read the v1 thread

Cover letter [RFC net-next 00/27] net and/or udp optimisations

A mix of various net optimisations, which were mostly discovered during UDP testing. Benchmarked with an io_uring test using 16B UDP/IPv6 over dummy netdev:
2090K vs 2229K tx/s, +6.6%, or in a 4-8% range if not averaging across reboots.

1-3 removes extra atomics and barriers from sock_wfree() mainly benefitting UDP. 4-7 cleans up some zerocopy helpers
8-16 do inlining of ipv6 and generic net pathes
17 is a small nice performance improvement for TCP zerocopy
18-27 refactors UDP to shed some more overhead

Pavel Begunkov (27):
sock: deduplicate ->sk_wmem_alloc check
sock: optimise sock_def_write_space send refcounting
sock: optimise sock_def_write_space barriers
skbuff: drop zero check from skb_zcopy_set
skbuff: drop null check from skb_zcopy
net: xen: set zc flags only when there is ubuf
skbuff: introduce skb_is_zcopy()
skbuff: optimise alloc_skb_with_frags()
net: inline sock_alloc_send_skb
net: inline part of skb_csum_hwoffload_help
net: inline skb_zerocopy_iter_dgram
ipv6: inline ip6_local_out()
ipv6: help __ip6_finish_output() inlining
ipv6: refactor ip6_finish_output2()
net: inline dev_queue_xmit()
ipv6: partially inline fl6_update_dst()
tcp: optimise skb_zerocopy_iter_stream()
net: optimise ipcm6 cookie init
udp/ipv6: refactor udpv6_sendmsg udplite checks udp/ipv6: move pending section of udpv6_sendmsg udp/ipv6: prioritise the ip6 path over ip4 checks udp/ipv6: optimise udpv6_sendmsg() daddr checks udp/ipv6: optimise out daddr reassignment udp/ipv6: clean up udpv6_sendmsg's saddr init
ipv6: refactor opts push in __ip6_make_skb()
ipv6: improve opt-less __ip6_make_skb()
ipv6: clean up ip6_setup_cork

 drivers/net/xen-netback/interface.c |   3 +-
 include/linux/netdevice.h           |  27 ++++-
 include/linux/skbuff.h              | 102 +++++++++++++-----
 include/net/ipv6.h                  |  37 ++++---
 include/net/sock.h                  |  10 +-
 net/core/datagram.c                 |   2 -
 net/core/datagram.h                 |  15 ---
 net/core/dev.c                      |  28 ++---
 net/core/skbuff.c                   |  59 ++++-------
 net/core/sock.c                     |  50 +++++++--
 net/ipv4/ip_output.c                |  10 +-
 net/ipv4/tcp.c                      |   5 +-
 net/ipv6/datagram.c                 |   4 +-
 net/ipv6/exthdrs.c                  |  15 ++-
 net/ipv6/ip6_output.c               |  88 ++++++++--------
 net/ipv6/output_core.c              |  12 ---
 net/ipv6/raw.c                      |   8 +-
 net/ipv6/udp.c                      | 158 +++++++++++++---------------
 net/l2tp/l2tp_ip6.c                 |   8 +-
 19 files changed, 339 insertions(+), 302 deletions(-)
 delete mode 100644 net/core/datagram.h
Signature · 1 line
2.35.1