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;
Date:   Sat, 23 Jun 2018 11:47:55 -0700
From:   Fenghua Yu <fenghua.yu@intel.com>
To:     Peter Zijlstra <peterz@infradead.org>
Cc:     Fenghua Yu <fenghua.yu@intel.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        Ingo Molnar <mingo@elte.hu>,
        "H. Peter Anvin" <hpa@linux.intel.com>,
        Ashok Raj <ashok.raj@intel.com>,
        Dave Hansen <dave.hansen@intel.com>,
        Rafael Wysocki <rafael.j.wysocki@intel.com>,
        Tony Luck <tony.luck@intel.com>,
        Alan Cox <alan@linux.intel.com>,
        Ravi V Shankar <ravi.v.shankar@intel.com>,
        Arjan van de Ven <arjan@infradead.org>,
        linux-kernel <linux-kernel@vger.kernel.org>, x86 <x86@kernel.org>
Subject: Re: [RFC PATCH 00/16] x86/split_lock: Enable #AC exception for split
 locked accesses
Message-ID: <20180623184755.GH18979@romley-ivt3.sc.intel.com>
References: <1527435965-202085-1-git-send-email-fenghua.yu@intel.com>
 <20180621193738.GA13636@worktop.programming.kicks-ass.net>
 <20180621201851.GC114883@romley-ivt3.sc.intel.com>
 <alpine.DEB.2.21.1806212223480.1591@nanos.tec.linutronix.de>
 <20180621220003.GD114883@romley-ivt3.sc.intel.com>
 <20180621221337.GE2494@hirez.programming.kicks-ass.net>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20180621221337.GE2494@hirez.programming.kicks-ass.net>
User-Agent: Mutt/1.5.23 (2014-03-12)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Fri, Jun 22, 2018 at 12:13:37AM +0200, Peter Zijlstra wrote:
> On Thu, Jun 21, 2018 at 03:00:03PM -0700, Fenghua Yu wrote:
> > For example, on a consolidated real-time machine, some cores are running
> 
> > Another example, in a public cloud deployed in the field, a user process
> 
> In either case a single split-lock shouldn't be a real problem, if you
> program the event with a count of 1 and have the NMI handler kill the
> offending task, you should be good.
> 
> Not saying the #AC isn't nicer, just saying the PMU based thing can
> still work.

The difference between perf event and #AC for split lock is perf event
is after the fact while #AC is before the fact.

In the examples (esp. the first example), hard real time cannot afforrd
a split lock/bus lock. Perf captures the event and kill the task; but
the split lock has been issued and harm has been done. #AC for split lock
kills the task before the task generates split lock or bus lock to hurt
real time tasks.

Thanks.

-Fenghua