Date:   Mon, 6 Mar 2023 20:58:18 -0500
From:   Steven Rostedt <rostedt@goodmis.org>
To:     Hillf Danton <hdanton@sina.com>
Cc:     John Stultz <jstultz@google.com>,
        LKML <linux-kernel@vger.kernel.org>, Wei Wang <wvw@google.com>,
        Midas Chien <midaschieh@google.com>,
        "Chunhui Li (=?UTF-8?B?5p2O5pil6L6J?=)" <chunhui.li@mediatek.com>,
        Kees Cook <keescook@chromium.org>,
        Anton Vorontsov <anton@enomsg.org>,
        "Guilherme G. Piccoli" <gpiccoli@igalia.com>,
        Tony Luck <tony.luck@intel.com>, linux-mm@kvack.org
Subject: Re: [PATCH v2] pstore: Revert pmsg_lock back to a normal mutex
Message-ID: <20230306205818.5a2bc51a@gandalf.local.home>
In-Reply-To: <20230307003106.1768-1-hdanton@sina.com>
References: <20230302062741.483079-1-jstultz@google.com>
        <20230304031029.3037914-1-jstultz@google.com>
        <20230306010323.2909-1-hdanton@sina.com>
        <20230307003106.1768-1-hdanton@sina.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Precedence: bulk

On Tue,  7 Mar 2023 08:31:06 +0800
Hillf Danton <hdanton@sina.com> wrote:

> On Mon, 6 Mar 2023 10:28:44 -0500 Steven Rostedt <rostedt@goodmis.org>
> > On Mon, 6 Mar 2023 09:03:23 +0800 Hillf Danton <hdanton@sina.com> wrote:  
> > >
> > > PS what sense made by spinning on owner until need_resched() with preempt
> > > disabled in the non-rt context?  
> > 
> > Not sure what the question you have here is? If need_resched() is set, we
> > want to schedule out.  
> 
> Given the critical section under mutex could be preempted, what is hard to
> understand is the wakeup of a ten-minute sleeper could not preempt a nice-10
> mutex spinner for instance.

But it can. Most wakeups are done by interrupts or softirqs. Both of which
will happen even if a task is running with preemption disabled.

-- Steve