[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <20250905165813.1470708-1-edumazet@google.com>
Date: Fri, 5 Sep 2025 16:58:04 +0000
From: Eric Dumazet <edumazet@...gle.com>
To: "David S . Miller" <davem@...emloft.net>, Jakub Kicinski <kuba@...nel.org>,
Paolo Abeni <pabeni@...hat.com>
Cc: Simon Horman <horms@...nel.org>, David Ahern <dsahern@...nel.org>,
Jamie Bainbridge <jamie.bainbridge@...il.com>, Abhishek Rawal <rawal.abhishek92@...il.com>,
netdev@...r.kernel.org, eric.dumazet@...il.com,
Eric Dumazet <edumazet@...gle.com>
Subject: [PATCH v2 net-next 0/9] ipv6: snmp: avoid performance issue with RATELIMITHOST
Addition of ICMP6_MIB_RATELIMITHOST in commit d0941130c9351
("icmp: Add counters for rate limits") introduced a performance
drop in case of DOS (like receiving UDP packets
to closed ports).
Per netns ICMP6_MIB_RATELIMITHOST tracking uses per-cpu
storage and is enough, we do not need per-device and slow tracking
for this metric.
In v2 of this series, I completed the removal of SNMP_MIB_SENTINEL
in all the kernel for consistency.
v1: had an issue : https://lore.kernel.org/netdev/20250904092432.113c4940@kernel.org/
Eric Dumazet (9):
ipv6: snmp: remove icmp6type2name[]
ipv6: snmp: do not use SNMP_MIB_SENTINEL anymore
ipv6: snmp: do not track per idev ICMP6_MIB_RATELIMITHOST
ipv4: snmp: do not use SNMP_MIB_SENTINEL anymore
mptcp: snmp: do not use SNMP_MIB_SENTINEL anymore
sctp: snmp: do not use SNMP_MIB_SENTINEL anymore
tls: snmp: do not use SNMP_MIB_SENTINEL anymore
xfrm: snmp: do not use SNMP_MIB_SENTINEL anymore
net: snmp: remove SNMP_MIB_SENTINEL
include/net/ip.h | 9 +++--
include/net/snmp.h | 5 ---
net/ipv4/proc.c | 65 +++++++++++++++----------------
net/ipv6/icmp.c | 3 +-
net/ipv6/proc.c | 91 ++++++++++++++++++++++++--------------------
net/mptcp/mib.c | 12 +++---
net/sctp/proc.c | 12 +++---
net/tls/tls_proc.c | 10 +++--
net/xfrm/xfrm_proc.c | 12 +++---
9 files changed, 113 insertions(+), 106 deletions(-)
--
2.51.0.355.g5224444f11-goog
Powered by blists - more mailing lists