linux-stable.git/include/net/ip6_route.h, branch v6.8

ipv6: lockless IPV6_ADDR_PREFERENCES implementation

2023-09-19T16:21:44+00:00

We have data-races while reading np->srcprefs

Switch the field to a plain byte, add READ_ONCE()
and WRITE_ONCE() annotations where needed,
and IPV6_ADDR_PREFERENCES setsockopt() can now be lockless.

Signed-off-by: Eric Dumazet 
Reviewed-by: David Ahern 
Link: https://lore.kernel.org/r/20230918142321.1794107-1-edumazet@google.com
Signed-off-by: Paolo Abeni

ipv6: lockless IPV6_MTU_DISCOVER implementation

2023-09-15T09:33:48+00:00

Most np->pmtudisc reads are racy.

Move this 3bit field on a full byte, add annotations
and make IPV6_MTU_DISCOVER setsockopt() lockless.

Signed-off-by: Eric Dumazet 
Reviewed-by: David Ahern 
Signed-off-by: David S. Miller

IPv6: add extack info for IPv6 address add/delete

2023-07-28T10:01:56+00:00

Add extack info for IPv6 address add/delete, which would be useful for
users to understand the problem without having to read kernel code.

Suggested-by: Beniamino Galvani 
Reviewed-by: Ido Schimmel 
Reviewed-by: David Ahern 
Signed-off-by: Hangbin Liu 
Reviewed-by: Simon Horman 
Signed-off-by: David S. Miller

net: dst: Prevent false sharing vs. dst_entry:: __refcnt

2023-03-29T01:52:22+00:00

dst_entry::__refcnt is highly contended in scenarios where many connections
happen from and to the same IP. The reference count is an atomic_t, so the
reference count operations have to take the cache-line exclusive.

Aside of the unavoidable reference count contention there is another
significant problem which is caused by that: False sharing.

perf top identified two affected read accesses. dst_entry::lwtstate and
rtable::rt_genid.

dst_entry:__refcnt is located at offset 64 of dst_entry, which puts it into
a seperate cacheline vs. the read mostly members located at the beginning
of the struct.

That prevents false sharing vs. the struct members in the first 64
bytes of the structure, but there is also

  dst_entry::lwtstate

which is located after the reference count and in the same cache line. This
member is read after a reference count has been acquired.

struct rtable embeds a struct dst_entry at offset 0. struct dst_entry has a
size of 112 bytes, which means that the struct members of rtable which
follow the dst member share the same cache line as dst_entry::__refcnt.
Especially

  rtable::rt_genid

is also read by the contexts which have a reference count acquired
already.

When dst_entry:__refcnt is incremented or decremented via an atomic
operation these read accesses stall. This was found when analysing the
memtier benchmark in 1:100 mode, which amplifies the problem extremly.

Move the rt[6i]_uncached[_list] members out of struct rtable and struct
rt6_info into struct dst_entry to provide padding and move the lwtstate
member after that so it ends up in the same cache line.

The resulting improvement depends on the micro-architecture and the number
of CPUs. It ranges from +20% to +120% with a localhost memtier/memcached
benchmark.

[ tglx: Rearrange struct ]

Signed-off-by: Wangyang Guo 
Signed-off-by: Arjan van de Ven 
Signed-off-by: Thomas Gleixner 
Reviewed-by: Eric Dumazet 
Reviewed-by: David Ahern 
Link: https://lore.kernel.org/r/20230323102800.042297517@linutronix.de
Signed-off-by: Jakub Kicinski

ipv6: Make ip6_route_output_flags_noref() static.

2023-01-25T02:12:52+00:00

This function is only used in net/ipv6/route.c and has no reason to be
visible outside of it.

Signed-off-by: Guillaume Nault 
Reviewed-by: David Ahern 
Link: https://lore.kernel.org/r/50706db7f675e40b3594d62011d9363dce32b92e.1674495822.git.gnault@redhat.com
Signed-off-by: Jakub Kicinski

net: add missing includes and forward declarations under net/

2022-07-22T11:53:22+00:00

This patch adds missing includes to headers under include/net.
All these problems are currently masked by the existing users
including the missing dependency before the broken header.

Signed-off-by: Jakub Kicinski 
Signed-off-by: David S. Miller

ipv6: ip6_skb_dst_mtu() cleanups

2021-11-20T04:09:55+00:00

Use const attribute where we can, and cache skb_dst()

Signed-off-by: Eric Dumazet 
Link: https://lore.kernel.org/r/20211119022355.2985984-1-eric.dumazet@gmail.com
Signed-off-by: Jakub Kicinski

Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

2021-08-05T22:08:47+00:00

Build failure in drivers/net/wwan/mhi_wwan_mbim.c:
add missing parameter (0, assuming we don't want buffer pre-alloc).

Conflict in drivers/net/dsa/sja1105/sja1105_main.c between:
  589918df9322 ("net: dsa: sja1105: be stateless with FDB entries on SJA1105P/Q/R/S/SJA1110 too")
  0fac6aa098ed ("net: dsa: sja1105: delete the best_effort_vlan_filtering mode")

Follow the instructions from the commit message of the former commit
- removed the if conditions. When looking at commit 589918df9322 ("net:
dsa: sja1105: be stateless with FDB entries on SJA1105P/Q/R/S/SJA1110 too")
note that the mask_iotag fields get removed by the following patch.

Signed-off-by: Jakub Kicinski

net: ipv6: fix returned variable type in ip6_skb_dst_mtu

2021-08-03T11:56:17+00:00

The patch fixing the returned value of ip6_skb_dst_mtu (int -> unsigned
int) was rebased between its initial review and the version applied. In
the meantime fade56410c22 was applied, which added a new variable (int)
used as the returned value. This lead to a mismatch between the function
prototype and the variable used as the return value.

Fixes: 40fc3054b458 ("net: ipv6: fix return value of ip6_skb_dst_mtu")
Cc: Vadim Fedorenko 
Signed-off-by: Antoine Tenart 
Signed-off-by: David S. Miller

net: ipv6: introduce ip6_dst_mtu_maybe_forward

2021-07-21T15:22:02+00:00

Replace ip6_dst_mtu_forward with ip6_dst_mtu_maybe_forward and
reuse this code in ip6_mtu. Actually these two functions were
almost duplicates, this change will simplify the maintaince of
mtu calculation code.

Signed-off-by: Vadim Fedorenko 
Reviewed-by: David Ahern 
Signed-off-by: David S. Miller