MIME-Version: 1.0
In-Reply-To: <20170616175020.GF4789@tassilo.jf.intel.com>
References: <20170615172320.GB4789@tassilo.jf.intel.com> <CABPqkBTYjTTquE8S4rTqe+Ps7c3BzhX_W7ZJRaeXhYRK97esFQ@mail.gmail.com>
 <20170615200205.GC4789@tassilo.jf.intel.com> <CABPqkBTKS3xJSB5daB8G8BmTOEgQeZ=du6+sZTkhiB3i4dTK7g@mail.gmail.com>
 <CABPqkBQCNqp_Lze6Va5RgFL_sQYEBVPmeLj5WcCOF99PeS=P4w@mail.gmail.com>
 <20170615231816.GD4789@tassilo.jf.intel.com> <CABPqkBQCbiW=b0kXXCUwh_e1FFTNLgHrA8en2Q1_Ov-W9-NBcQ@mail.gmail.com>
 <20170616160630.GE4789@tassilo.jf.intel.com> <CABPqkBQY+be3Ly5TS096gNi_fKhQWG=Ep-ue=BL6Sh3t7sjv8g@mail.gmail.com>
 <CABPqkBRB5sB5_OY0UNmKUsamqHMFnTkS+Xe9fjSDzBdi2rxU5Q@mail.gmail.com> <20170616175020.GF4789@tassilo.jf.intel.com>
From: Stephane Eranian <eranian@google.com>
Date: Fri, 16 Jun 2017 12:15:14 -0700
Message-ID: <CABPqkBTuyDByke_uN_qOZhvdv5vLKtfFK5qxyFU9-mV=nakcRQ@mail.gmail.com>
Subject: Re: [PATCH 0/5] perf: add support for capturing skid IP
To: Andi Kleen <ak@linux.intel.com>
Cc: LKML <linux-kernel@vger.kernel.org>,
        Arnaldo Carvalho de Melo <acme@redhat.com>,
        Peter Zijlstra <peterz@infradead.org>, "mingo@elte.hu" <mingo@elte.hu>,
        "Liang, Kan" <kan.liang@intel.com>, Jiri Olsa <jolsa@redhat.com>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 553
Lines: 14

On Fri, Jun 16, 2017 at 10:50 AM, Andi Kleen <ak@linux.intel.com> wrote:
>> > Yeah, I think it is easier and more portable, especially on hardware with a
>> > PEBS-like mechanism but no branch buffer (like LBR). FYI, I did do a test
>> > implementation yesterday to evaluate the difficulty.
>> >
>> A more generalized usage of the feature is to evaluate the amount of skid
>> for any precise event.
>
> It should be always the same (one instruction), except for the control flow
> change case.
>
That's on Intel X86. What about the other arch?

> -Andi