[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <20251008104612.1824200-1-edumazet@google.com>
Date: Wed, 8 Oct 2025 10:46:07 +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>, Kuniyuki Iwashima <kuniyu@...gle.com>,
Willem de Bruijn <willemb@...gle.com>, netdev@...r.kernel.org, eric.dumazet@...il.com,
Eric Dumazet <edumazet@...gle.com>
Subject: [PATCH RFC net-next 0/4] net: deal with strange attractors tx queues
Back 2010, Tom Herbert added skb->ooo_okay to TCP flows.
Extend the feature to connected flows for other protocols like UDP.
skb->ooo_okay might never be set for bulk flows that always
have at least one skb in a qdisc queue of NIC queue,
especially if TX completion is delayed because of a stressed cpu.
The so-called "strange attractors" has caused many performance
issues, we need to do better now that TCP deals better with
potential reorders.
Add new net.core.txq_reselection_ms sysctl to let
flows follow XPS and select a more efficient queue.
After this series, we no longer have to make sure threads
are pinned to cpus, they now can be migrated without
adding too much spinlock/qdisc/TX completion pressure anymore.
Eric Dumazet (4):
net: add SK_WMEM_ALLOC_BIAS constant
net: control skb->ooo_okay from skb_set_owner_w()
net: add /proc/sys/net/core/txq_reselection_ms control
net: allow busy connected flows to switch tx queues
Documentation/admin-guide/sysctl/net.rst | 15 +++++++++
include/net/netns/core.h | 1 +
include/net/sock.h | 43 ++++++++++++------------
net/atm/common.c | 2 +-
net/core/dev.c | 27 +++++++++++++--
net/core/net_namespace.c | 1 +
net/core/sock.c | 17 +++++++---
net/core/sysctl_net_core.c | 7 ++++
8 files changed, 84 insertions(+), 29 deletions(-)
--
2.51.0.710.ga91ca5db03-goog
Powered by blists - more mailing lists