Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::3:7 as permitted sender) client-ip=2620:137:e000::3:7;
From:   Valentin Schneider <vschneid@redhat.com>
To:     dccp@vger.kernel.org, netdev@vger.kernel.org,
        linux-kernel@vger.kernel.org, linux-rt-users@vger.kernel.org
Cc:     "David S. Miller" <davem@davemloft.net>,
        Eric Dumazet <edumazet@google.com>,
        Jakub Kicinski <kuba@kernel.org>,
        Paolo Abeni <pabeni@redhat.com>,
        David Ahern <dsahern@kernel.org>,
        Juri Lelli <juri.lelli@redhat.com>,
        Tomas Glozar <tglozar@redhat.com>,
        Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
        Thomas Gleixner <tglx@linutronix.de>
Subject: [PATCH v2 0/2] tcp/dcpp: tw_timer tweaks for nohz_full and PREEMPT_RT
Date:   Wed, 15 Nov 2023 16:05:07 -0500
Message-ID: <20231115210509.481514-1-vschneid@redhat.com>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Precedence: bulk

Hi,

This is v2 of [1] where the tw_timer is un-pinned to get rid of interferences in
isolated CPUs setups.

Patch 1 is pretty much the same as v1, just got an extra comment in
inet_twsk_deschedule_put() to highlight the race.

Patch 2 was added as AFAICT the bh_disable is no longer needed after patch 1,
and Sebastian mentioned during LPC the he had been looking at getting rid of it
for removing softirq_ctrl.lock in PREEMPT_RT.

Eric mentionned rsk_timer needs looking into, but I haven't had the time to do
that. It doesn't show up in our testing, which might be due to its relatively
low timeout (IIUC 3s).

[1]: https://lore.kernel.org/all/20231016125934.1970789-1-vschneid@redhat.com/

Valentin Schneider (2):
  tcp/dcpp: Un-pin tw_timer
  tcp/dcpp: Don't disable bh around timewait_sock initialization

 net/dccp/minisocks.c          | 14 ++++----------
 net/ipv4/inet_timewait_sock.c | 20 +++++++++++++++-----
 net/ipv4/tcp_minisocks.c      | 14 ++++----------
 3 files changed, 23 insertions(+), 25 deletions(-)

--
2.41.0