This patchset introduces 4-tuple hash for connected udp sockets, to make
connected udp lookup faster.

Patch1: Add a new counter for hslot2 named hash4_cnt, to avoid cache line
        miss when lookup.
Patch2 and 3: Implement 4-tuple hash for ipv4.

The detailed motivation is described in Patch 3.

AFAICS the patchset can be further improved by:
(a) Better interact with hash2/reuseport. Now hash4 hardly affects other
mechanisms, but maintaining sockets in both hash4 and hash2 lists seems
unnecessary.
(b) Support early demux and ipv6.

changelogs:
RFCv2 -> v3 (Gur Stavi):
- minor fix in udp_hashslot2() and udp_table_init()
- add rcu sync in rehash4()

RFCv1 -> RFCv2:
- add a new struct for hslot2
- remove the sockopt UDP_HASH4 because it has little side effect for
  unconnected sockets
- add rehash in connect()
- re-organize the patch into 3 smaller ones
- other minor fix

RFCv2:
https://lore.kernel.org/all/20240924110414.52618-1-lulie@linux.alibaba.com/ (local)
RFCv1:
https://lore.kernel.org/all/20240913100941.8565-1-lulie@linux.alibaba.com/ (local)

Philo Lu (3):
  net/udp: Add a new struct for hash2 slot
  net/udp: Add 4-tuple hash list basis
  ipv4/udp: Add 4-tuple hash for connected socket

 include/linux/udp.h |   7 ++
 include/net/udp.h   |  44 ++++++++--
 net/ipv4/udp.c      | 199 ++++++++++++++++++++++++++++++++++++++------
 net/ipv6/udp.c      |  17 ++--
 4 files changed, 225 insertions(+), 42 deletions(-)

--
2.32.0.3.g01195cf9f