Re: [RFC -tip] perf, x86: Add PERF_COUNT_HW_NMI_WATCHDOG event v2

[Cyrill Gorcunov <gorcunov@openvz.org>]

Ok, I'll take a look (sorry for top posting, ISP problem)

On Tuesday, June 21, 2011, Peter Zijlstra <a.p.zijlstra@chello.nl> wrote:
> On Tue, 2011-06-21 at 19:40 +0400, Cyrill Gorcunov wrote:
>> On Tue, Jun 21, 2011 at 7:23 PM, Don Zickus <dzickus@redhat.com> wrote:
>> > On Thu, Apr 28, 2011 at 07:37:25PM +0400, Cyrill Gorcunov wrote:
>> >> Due to restriction and specifics of Netburst PMU we need a separated
>> >> event for NMI watchdog. In particular every Netburst event consume not
>> >> just a counter and config register, but also an additional ESCR register.
>> >> Since ESCR registers are grouped upon counters (i.e. if ESCR is occupied
>> >> for some event there is no room for another event to enter the room until
>> >> it's released) we need to pick up "least" used ESCR (or most available)
>> >> for nmi-watchdog purpose -- MSR_P4_CRU_ESCR2/3 was chosen.
>> >>
>> >> Note that on all other PMUs which support relocation of events between
>> >> counters this event is a simple alias for PERF_COUNT_HW_CPU_CYCLES.
>> >
>> > Cyril,
>> >
>> > What ever happened with this patch?
>> >
>> > Cheers,
>> > Don
>> >
>>
>> Well, it's flowing around, I didn't get any more Ack's other than your. Since it
>> introduces kind of abi I presume more ack's would be welcome before it
>> (possibly) get merged.
>
> we could avoid that abi thing by simply not making it an exposed event
>
> Its not like userspace ever should be using it anyway
>