On Fri, 24 Apr 2015, Andi Kleen wrote:
> > There are better ways to do that than using heuristics. We have to
> > deal with 3 variants of the reference counter:
> >
> > 1) Core and Atom: counts bus cycles and we know that frequency already
> > from the local apic calibration
> >
> >
> There are better ways to do that than using heuristics. We have to
> deal with 3 variants of the reference counter:
>
> 1) Core and Atom: counts bus cycles and we know that frequency already
> from the local apic calibration
>
> 2) Nehalem, Westmere: Same as TSC
>
> 3) Sandybri
On Fri, 24 Apr 2015, Alexander Shishkin wrote:
> Thomas Gleixner writes:
>
> > On Fri, 24 Apr 2015, Thomas Gleixner wrote:
> >> On Thu, 23 Apr 2015, Andi Kleen wrote:
> >> > > We can just detect the deviation in the callback itself:
> >> > >
> >> > >u64 now = ktime_get_mono_fast_ns();
>
Thomas Gleixner writes:
> On Fri, 24 Apr 2015, Thomas Gleixner wrote:
>> On Thu, 23 Apr 2015, Andi Kleen wrote:
>> > > We can just detect the deviation in the callback itself:
>> > >
>> > >u64 now = ktime_get_mono_fast_ns();
>> > >
>> > >if (now - __this_cpu_read(nmi_timestamp)
On Fri, 24 Apr 2015, Thomas Gleixner wrote:
> On Thu, 23 Apr 2015, Andi Kleen wrote:
> > > We can just detect the deviation in the callback itself:
> > >
> > >u64 now = ktime_get_mono_fast_ns();
> > >
> > >if (now - __this_cpu_read(nmi_timestamp) < period)
> > > r
On Thu, Apr 23, 2015 at 05:51:33PM -0700, Andi Kleen wrote:
> There were systems in the past that ran TSC at a much slower
> frequency, such as the early AMD Barcelona systems.
You mean the eval boxes which were never sold as production systems?
--
Regards/Gruss,
Boris.
ECO tip #101: Trim y
On Thu, 23 Apr 2015, Andi Kleen wrote:
> > We can just detect the deviation in the callback itself:
> >
> >u64 now = ktime_get_mono_fast_ns();
> >
> >if (now - __this_cpu_read(nmi_timestamp) < period)
> >return;
> >
> >__this_cpu_write(nmi_timestamp, n
> We can just detect the deviation in the callback itself:
>
>u64 now = ktime_get_mono_fast_ns();
>
>if (now - __this_cpu_read(nmi_timestamp) < period)
> return;
>
>__this_cpu_write(nmi_timestamp, now);
>
> It's that simple.
It's a simple short term
On Thu, 23 Apr 2015, Andi Kleen wrote:
> On Thu, Apr 23, 2015 at 10:01:04PM +0200, Thomas Gleixner wrote:
> > On Thu, 23 Apr 2015, Alexander Shishkin wrote:
> >
> > > The problem with using cycle counter for NMI watchdog is that its
> > > frequency changes with the corresponding core's frequency.
On Thu, Apr 23, 2015 at 10:01:04PM +0200, Thomas Gleixner wrote:
> On Thu, 23 Apr 2015, Alexander Shishkin wrote:
>
> > The problem with using cycle counter for NMI watchdog is that its
> > frequency changes with the corresponding core's frequency. This means
> > that, in particular, if the core f
On Thu, 23 Apr 2015, Alexander Shishkin wrote:
> The problem with using cycle counter for NMI watchdog is that its
> frequency changes with the corresponding core's frequency. This means
> that, in particular, if the core frequency scales up, watchdog NMI will
> arrive more frequently than what us
The problem with using cycle counter for NMI watchdog is that its
frequency changes with the corresponding core's frequency. This means
that, in particular, if the core frequency scales up, watchdog NMI will
arrive more frequently than what user requested through watchdog_thresh
and also increasing
12 matches
Mail list logo
