Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
From:   Daniel Bristot de Oliveira <bristot@redhat.com>
To:     linux-kernel@vger.kernel.org
Cc:     williams@redhat.com, daniel@bristot.me,
        "Steven Rostedt (VMware)" <rostedt@goodmis.org>,
        Ingo Molnar <mingo@redhat.com>,
        Peter Zijlstra <peterz@infradead.org>,
        Thomas Gleixner <tglx@linutronix.de>,
        "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>,
        Matthias Kaehlcke <mka@chromium.org>,
        "Joel Fernandes (Google)" <joel@joelfernandes.org>,
        Frederic Weisbecker <frederic@kernel.org>,
        Yangtao Li <tiny.windzz@gmail.com>,
        Tommaso Cucinotta <tommaso.cucinotta@santannapisa.it>
Subject: [RFC 0/3] preempt_tracer: Fix preempt_disable tracepoint
Date:   Tue, 28 May 2019 17:16:21 +0200
Message-Id: <cover.1559051152.git.bristot@redhat.com>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

While playing with the model + working in a fix for the task
context & trace recursion, I ended up hitting two cases in which the
preempt_disable/enable tracepoint was supposed to happen, but didn't.

There is an explanation for each case in the log message.

This is an RFC exposing the problem, with possible ways to fix it.
But I bet there might be better solutions as well, so this is a real
RFC.

Daniel Bristot de Oliveira (3):
  softirq: Use preempt_latency_stop/start to trace preemption
  preempt_tracer: Disable IRQ while starting/stopping due to a
    preempt_counter change
  preempt_tracer: Use a percpu variable to control traceble calls

 kernel/sched/core.c | 66 ++++++++++++++++++++++++++++++++++-----------
 kernel/softirq.c    | 13 ++++-----
 2 files changed, 55 insertions(+), 24 deletions(-)

-- 
2.20.1