Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Fri, 15 Nov 2019 15:55:28 +0100
From:   Jiri Olsa <jolsa@redhat.com>
To:     Andi Kleen <andi@firstfloor.org>
Cc:     jolsa@kernel.org, acme@kernel.org, linux-kernel@vger.kernel.org,
        Andi Kleen <ak@linux.intel.com>
Subject: Re: [PATCH v6 10/12] perf stat: Use affinity for reading
Message-ID: <20191115145528.GB4255@krava>
References: <20191112005941.649137-1-andi@firstfloor.org>
 <20191112005941.649137-11-andi@firstfloor.org>
MIME-Version: 1.0
In-Reply-To: <20191112005941.649137-11-andi@firstfloor.org>
User-Agent: Mutt/1.12.1 (2019-06-15)
Content-Type: text/plain; charset=WINDOWS-1252
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Mon, Nov 11, 2019 at 04:59:39PM -0800, Andi Kleen wrote:
> From: Andi Kleen <ak@linux.intel.com>
>=20
> Restructure event reading to use affinity to minimize the number
> of IPIs needed.
>=20
> Before on a large test case with 94 CPUs:
>=20
> % time     seconds  usecs/call     calls    errors syscall
> ------ ----------- ----------- --------- --------- ----------------
>   3.16    0.106079           4     22082           read
>=20
> After:
>=20
>   3.43    0.081295           3     22082           read
>=20
> Signed-off-by: Andi Kleen <ak@linux.intel.com>
>=20
> ---
>=20
> v2: Use new iterator macros
> v3: Use new iterator macros
> v4: Change iterator macros even more
> v5: Preserve counter->err in all cases
> ---
>  tools/perf/builtin-stat.c | 95 ++++++++++++++++++++++-----------------
>  tools/perf/util/evsel.h   |  1 +
>  2 files changed, 55 insertions(+), 41 deletions(-)
>=20
> diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
> index 039aefb07777..7784f5a93944 100644
> --- a/tools/perf/builtin-stat.c
> +++ b/tools/perf/builtin-stat.c
> @@ -266,15 +266,10 @@ static int read_single_counter(struct evsel *counte=
r, int cpu,
>   * Read out the results of a single counter:
>   * do not aggregate counts across CPUs in system-wide mode
>   */
> -static int read_counter(struct evsel *counter, struct timespec *rs)
> +static int read_counter(struct evsel *counter, struct timespec *rs, int =
cpu)

please rename this to read_counter_cpu

thanks,
jirka