[LibreQoS] Fwd: [PATCH v5 net 0/3] Report RCU QS for busy network kthreads

[Dave Taht]

---------- Forwarded message ---------
From: Yan Zhai <yan at cloudflare.com>
Date: Tue, Mar 19, 2024 at 4:44 PM
Subject: [PATCH v5 net 0/3] Report RCU QS for busy network kthreads
To: <netdev at vger.kernel.org>
Cc: David S. Miller <davem at davemloft.net>, Eric Dumazet
<edumazet at google.com>, Jakub Kicinski <kuba at kernel.org>, Paolo Abeni
<pabeni at redhat.com>, Jiri Pirko <jiri at resnulli.us>, Simon Horman
<horms at kernel.org>, Daniel Borkmann <daniel at iogearbox.net>, Lorenzo
Bianconi <lorenzo at kernel.org>, Coco Li <lixiaoyan at google.com>, Wei
Wang <weiwan at google.com>, Alexander Duyck <alexanderduyck at fb.com>,
<linux-kernel at vger.kernel.org>, <rcu at vger.kernel.org>,
<bpf at vger.kernel.org>, <kernel-team at cloudflare.com>, Joel Fernandes
<joel at joelfernandes.org>, Paul E. McKenney <paulmck at kernel.org>, Toke
Høiland-Jørgensen <toke at redhat.com>, Alexei Starovoitov
<alexei.starovoitov at gmail.com>, Steven Rostedt <rostedt at goodmis.org>,
<mark.rutland at arm.com>, Jesper Dangaard Brouer <hawk at kernel.org>,
Sebastian Andrzej Siewior <bigeasy at linutronix.de>


This changeset fixes a common problem for busy networking kthreads.
These threads, e.g. NAPI threads, typically will do:

* polling a batch of packets
* if there are more work, call cond_resched() to allow scheduling
* continue to poll more packets when rx queue is not empty

We observed this being a problem in production, since it can block RCU
tasks from making progress under heavy load. Investigation indicates
that just calling cond_resched() is insufficient for RCU tasks to reach
quiescent states. This also has the side effect of frequently clearing
the TIF_NEED_RESCHED flag on voluntary preempt kernels. As a result,
schedule() will not be called in these circumstances, despite schedule()
in fact provides required quiescent states. This at least affects NAPI
threads, napi_busy_loop, and also cpumap kthread.

By reporting RCU QSes in these kthreads periodically before cond_resched, the
blocked RCU waiters can correctly progress. Instead of just reporting QS for
RCU tasks, these code share the same concern as noted in the commit
d28139c4e967 ("rcu: Apply RCU-bh QSes to RCU-sched and RCU-preempt when safe").
So report a consolidated QS for safety.

It is worth noting that, although this problem is reproducible in
napi_busy_loop, it only shows up when setting the polling interval to as high
as 2ms, which is far larger than recommended 50us-100us in the documentation.
So napi_busy_loop is left untouched.

Lastly, this does not affect RT kernels, which does not enter the scheduler
through cond_resched(). Without the mentioned side effect, schedule() will
be called time by time, and clear the RCU task holdouts.

V4: https://lore.kernel.org/bpf/cover.1710525524.git.yan@cloudflare.com/
V3: https://lore.kernel.org/lkml/20240314145459.7b3aedf1@kernel.org/t/
V2: https://lore.kernel.org/bpf/ZeFPz4D121TgvCje@debian.debian/
V1: https://lore.kernel.org/lkml/Zd4DXTyCf17lcTfq@debian.debian/#t

changes since v4:
 * polished comments and docs for the RCU helper as Paul McKenney suggested

changes since v3:
 * fixed kernel-doc errors

changes since v2:
 * created a helper in rcu header to abstract the behavior
 * fixed cpumap kthread in addition

changes since v1:
 * disable preemption first as Paul McKenney suggested

Yan Zhai (3):
  rcu: add a helper to report consolidated flavor QS
  net: report RCU QS on threaded NAPI repolling
  bpf: report RCU QS in cpumap kthread

 include/linux/rcupdate.h | 31 +++++++++++++++++++++++++++++++
 kernel/bpf/cpumap.c      |  3 +++
 net/core/dev.c           |  3 +++
 3 files changed, 37 insertions(+)

--
2.30.2





-- 
https://www.youtube.com/watch?v=N0Tmvv5jJKs Epik Mellon Podcast
Dave Täht CSO, LibreQos