[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <AANLkTikjfM=8xa8smgT6D_WTYC_S-JZ69iiqu8gD21sK@mail.gmail.com>
Date: Fri, 26 Nov 2010 11:57:51 +0100
From: Stephane Eranian <eranian@...gle.com>
To: Cyrill Gorcunov <gorcunov@...nvz.org>
Cc: Ingo Molnar <mingo@...e.hu>, LKML <linux-kernel@...r.kernel.org>,
ming.m.lin@...el.com, peterz@...radead.org
Subject: Re: [rfc 1/3] perf, x86: P4 PMU - describe config format
On Tue, Nov 23, 2010 at 11:46 PM, Cyrill Gorcunov <gorcunov@...nvz.org> wrote:
> Add description of .config in a sake of RAW events.
> At least this should bring some light to those who
> will be reading this code.
>
> Signed-off-by: Cyrill Gorcunov <gorcunov@...nvz.org>
> CC: Lin Ming <ming.m.lin@...el.com>
> CC: Stephane Eranian <eranian@...gle.com>
> CC: Peter Zijlstra <peterz@...radead.org>
> ---
> arch/x86/include/asm/perf_event_p4.h | 62 ++++++++++++++++++++++++++++++-----
> 1 file changed, 54 insertions(+), 8 deletions(-)
>
> Index: linux-2.6.git/arch/x86/include/asm/perf_event_p4.h
> =====================================================================
> --- linux-2.6.git.orig/arch/x86/include/asm/perf_event_p4.h
> +++ linux-2.6.git/arch/x86/include/asm/perf_event_p4.h
> @@ -744,14 +744,6 @@ enum P4_ESCR_EMASKS {
> };
>
> /*
> - * P4 PEBS specifics (Replay Event only)
> - *
> - * Format (bits):
> - * 0-6: metric from P4_PEBS_METRIC enum
> - * 7 : reserved
> - * 8 : reserved
> - * 9-11 : reserved
> - *
> * Note we have UOP and PEBS bits reserved for now
> * just in case if we will need them once
> */
> @@ -788,5 +780,59 @@ enum P4_PEBS_METRIC {
> P4_PEBS_METRIC__max
> };
>
> +/*
> + * Notes on internal configuration of ESCR+CCCR tuples
> + *
> + * Since P4 has quite the different architecture of
> + * performance registers in compare with "architectural"
> + * once and we have on 64 bits to keep configuration
> + * of performance event, the following trick is used.
> + *
> + * 1) Since both ESCR and CCCR registers have only low
> + * 32 bits valuable, we pack them into a single 64 bit
> + * configuration. Low 32 bits of such config correspond
> + * to low 32 bits of CCCR register and high 32 bits
> + * correspond to low 32 bits of ESCR register.
> + *
> + * 2) The meaning of every bit of such config field can
> + * be found in Intel SDM but it should be noted that
> + * we "borrow" some reserved bits for own usage and
> + * clean them or set to a proper value when we do
> + * a real write to hardware registers.
> + *
> + * 3) The format of bits of config is the following
> + * and should be either 0 or set to some predefined
> + * values:
> + *
> + * Low 32 bits
> + * -----------
> + * 0-6: P4_PEBS_METRIC enum
> + * 7-11: reserved
> + * 12: Active thread
I don't understand bit 12. In the actual register, it
corresponds to the enable bit. Seems you're overriding
its usage. Do I interpret this as saying: 0 = enable when
running on thread0, 1=monitoring when running on thread1?
And if I don't care?
> + * 13-15: reserved (ESCR select)
> + * 16-17: Compare
> + * 18: Complement
> + * 20-23: Threshold
> + * 24: Edge
> + * 25: reserved (FORCE_OVF)
> + * 26: reserved (OVF_PMI_T0)
> + * 27: reserved (OVF_PMI_T1)
> + * 28-29: reserved
> + * 30: reserved (Cascade)
> + * 31: reserved (OVF)
> + *
> + * High 32 bits
> + * ------------
> + * 0: reserved (T1_USR)
> + * 1: reserved (T1_OS)
> + * 2: reserved (T0_USR)
> + * 3: reserved (T0_OS)
> + * 4: Tag Enable
> + * 5-8: Tag Value
> + * 9-24: Event Mask (may use P4_ESCR_EMASK_BIT helper)
> + * 25-30: enum P4_EVENTS
> + * 31: reserved (HT thread)
> + */
> +
> #endif /* PERF_EVENT_P4_H */
>
>
>
--
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/
Powered by blists - more mailing lists