Re: [PATCH v2 2/2] perf/x86/intel: force resched when TFA sysctl is modified

From: Stephane Eranian <eranian@google.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: LKML <linux-kernel@vger.kernel.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Andi Kleen <ak@linux.intel.com>,
	"Liang, Kan" <kan.liang@intel.com>,
	mingo@elte.hu, nelson.dsouza@intel.com,
	Jiri Olsa <jolsa@redhat.com>,
	tonyj@suse.com
Subject: Re: [PATCH v2 2/2] perf/x86/intel: force resched when TFA sysctl is modified
Date: Mon, 15 Apr 2019 10:20:57 -0700	[thread overview]
Message-ID: <CABPqkBQtJ81TMvjExJ3+hm1f+E1zkbTi-au-GkJRTg9ovZgEsQ@mail.gmail.com> (raw)
In-Reply-To: <20190415155742.GI12232@hirez.programming.kicks-ass.net>

On Mon, Apr 15, 2019 at 8:57 AM Peter Zijlstra <peterz@infradead.org> wrote:
>
> On Mon, Apr 08, 2019 at 10:32:52AM -0700, Stephane Eranian wrote:
> > +static ssize_t set_sysctl_tfa(struct device *cdev,
> > +                           struct device_attribute *attr,
> > +                           const char *buf, size_t count)
> > +{
> > +     bool val;
> > +     ssize_t ret;
> > +
> > +     ret = kstrtobool(buf, &val);
> > +     if (ret)
> > +             return ret;
> > +
> > +     /* no change */
> > +     if (val == allow_tsx_force_abort)
> > +             return count;
> > +
> > +     allow_tsx_force_abort = val;
> > +
> > +     get_online_cpus();
> > +     on_each_cpu(update_tfa_sched, NULL, 1);
> > +     put_online_cpus();
> > +
> > +     return count;
> > +}
>
> So we care about concurrent writing to that file?
Not likely but we care about seeing the effects on event scheduling
before the sysctl returns.