Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Tue, 27 Oct 2020 11:28:51 +0100
From:   Sebastian Andrzej Siewior <bigeasy@linutronix.de>
To:     Daniel Wagner <wagi@monom.org>
Cc:     Thomas Gleixner <tglx@linutronix.de>,
        LKML <linux-kernel@vger.kernel.org>,
        linux-rt-users <linux-rt-users@vger.kernel.org>,
        Steven Rostedt <rostedt@goodmis.org>
Subject: Re: [ANNOUNCE] v5.9.1-rt19
Message-ID: <20201027102851.gizepjlu4opensqb@linutronix.de>
References: <20201024091838.gzhn2dlx2j7xnixg@linutronix.de>
 <20201027093616.5vn6xinmthxulhvx@beryllium.lan>
 <20201027100049.xtkmjqdwkn7zec2f@linutronix.de>
 <20201027102547.y6wop7j2ovzg2tyx@beryllium.lan>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <20201027102547.y6wop7j2ovzg2tyx@beryllium.lan>
Precedence: bulk

On 2020-10-27 11:25:47 [+0100], Daniel Wagner wrote:
> Well, one thing I need to figure out is how to get pi_stress working
> correctly on my machines (that is for all -rt trees, so it's nothing
> new). It consistently triggers RCU stall warnings with hackbench as
> workload. Clark told me the test only works if there are enough CPUs to
> run on. Need to look into it.

Is it running as a RT task? Do you have RCU-boosting enabled?
Otherwise it looks that if you throw enough non-RT load on the system,
RCU can not catch up which does not sound good.

Sebastian