[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20140715111210.GB9918@twins.programming.kicks-ass.net>
Date: Tue, 15 Jul 2014 13:12:10 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: "Yan, Zheng" <zheng.z.yan@...el.com>
Cc: linux-kernel@...r.kernel.org, mingo@...nel.org, acme@...radead.org,
eranian@...gle.com, andi@...stfloor.org
Subject: Re: [PATCH v2 6/7] perf, x86: enable large PEBS interrupt threshold
for SNB/IVB/HSW
On Tue, Jul 15, 2014 at 04:58:58PM +0800, Yan, Zheng wrote:
> Signed-off-by: Yan, Zheng <zheng.z.yan@...el.com>
> ---
> arch/x86/kernel/cpu/perf_event_intel.c | 4 ++++
> 1 file changed, 4 insertions(+)
>
> diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c
> index cb5a838..dba03b3 100644
> --- a/arch/x86/kernel/cpu/perf_event_intel.c
> +++ b/arch/x86/kernel/cpu/perf_event_intel.c
> @@ -2456,6 +2456,7 @@ __init int intel_pmu_init(void)
> intel_perfmon_event_map[PERF_COUNT_HW_STALLED_CYCLES_BACKEND] =
> X86_CONFIG(.event=0xb1, .umask=0x01, .inv=1, .cmask=1);
>
> + x86_pmu.multi_pebs = true;
> pr_cont("SandyBridge events, ");
> break;
> case 58: /* IvyBridge */
> @@ -2484,6 +2485,7 @@ __init int intel_pmu_init(void)
> intel_perfmon_event_map[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND] =
> X86_CONFIG(.event=0x0e, .umask=0x01, .inv=1, .cmask=1);
>
> + x86_pmu.multi_pebs = true;
> pr_cont("IvyBridge events, ");
> break;
>
> @@ -2511,6 +2513,8 @@ __init int intel_pmu_init(void)
> x86_pmu.get_event_constraints = hsw_get_event_constraints;
> x86_pmu.cpu_events = hsw_events_attrs;
> x86_pmu.lbr_double_abort = true;
> +
> + x86_pmu.multi_pebs = true;
> pr_cont("Haswell events, ");
> break;
NAK for the very same reason. All (PEBS capable) hardware supports this.
Content of type "application/pgp-signature" skipped
Powered by blists - more mailing lists