linux-kernel - Re: [PATCH 2/4] perf_event: add PERF_COUNT_HW_REF_CPU

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Mon, 12 Dec 2011 22:45:58 -0800
From:	Stephane Eranian <eranian@...gle.com>
To:	Ingo Molnar <mingo@...e.hu>
Cc:	Peter Zijlstra <peterz@...radead.org>,
	linux-kernel@...r.kernel.org, acme@...radead.org,
	robert.richter@....com, ming.m.lin@...el.com, andi@...stfloor.org,
	asharma@...com
Subject: Re: [PATCH 2/4] perf_event: add PERF_COUNT_HW_REF_CPU_CYCLES generic
 PMU event

On Mon, Dec 12, 2011 at 9:32 AM, Ingo Molnar <mingo@...e.hu> wrote:
>
> * Peter Zijlstra <peterz@...radead.org> wrote:
>
>> On Sun, 2011-12-11 at 00:28 +0100, Stephane Eranian wrote:
>> > This event counts the number of reference core cpu cycles.
>> > Reference means that the event increments at a constant rate which
>> > is not subject to core CPU frequency adjustments. The event may
>> > not count when the processor is in halted (low power) state.
>> > As such, it may not be equivalent to wall clock time. However,
>> > when the processor is not halted state, the event keeps
>> > a constant correlation with wall clock time.
>> >
>> > Signed-off-by: Stephane Eranian <eranian@...gle.com>
>> > ---
>> >  include/linux/perf_event.h |    1 +
>> >  1 files changed, 1 insertions(+), 0 deletions(-)
>> >
>> > diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
>> > index 564769c..0885561 100644
>> > --- a/include/linux/perf_event.h
>> > +++ b/include/linux/perf_event.h
>> > @@ -54,6 +54,7 @@ enum perf_hw_id {
>> >     PERF_COUNT_HW_BUS_CYCLES                = 6,
>> >     PERF_COUNT_HW_STALLED_CYCLES_FRONTEND   = 7,
>> >     PERF_COUNT_HW_STALLED_CYCLES_BACKEND    = 8,
>> > +   PERF_COUNT_HW_REF_CPU_CYCLES            = 9,
>> >
>> >     PERF_COUNT_HW_MAX,                      /* non-ABI */
>> >  };
>>
>> Does it make sense to add this to the 'generic' events? Are
>> other archs going to use this?
>>
>> That is, I already queued patch 1, I'm just wondering if the
>> generic bit makes sense, Even BUS_CYCLES seems to be a
>> questionable 'generic' event, but that's history and we can't
>> fix it.
>
> If we named it in a generic way, with a generic,
> platform-independent meaning behind it, then it shouldn't be a
> problem. This is why i suggested naming it 'constant CPU cycles'
> - or 'constant freq cycles' or a variant of that.
>
Isn't CPU_REF_CYCLES good enough?
Should we speel out ref completely to 'REFERENCE'.
In the changelog, I gave a generic definition of what it is supposed to
measure. If most platforms don't have such events, then that's fine too.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/