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;
From:   Vineet Gupta <vineet.gupta1@synopsys.com>
To:     Alexey Brodkin <alexey.brodkin@synopsys.com>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
CC:     "linux-snps-arc@lists.infradead.org" 
        <linux-snps-arc@lists.infradead.org>,
        Daniel Lezcano <daniel.lezcano@linaro.org>,
        Thomas Gleixner <tglx@linutronix.de>
Subject: Re: [PATCH v3] clocksource/drivers/arc_timer: Utilize generic
 sched_clock
Thread-Topic: [PATCH v3] clocksource/drivers/arc_timer: Utilize generic
 sched_clock
Thread-Index: AQHUf/s5h49b24gHWUWfthbIv7yKtg==
Date:   Mon, 19 Nov 2018 17:33:58 +0000
Message-ID: <C2D7FE5348E1B147BCA15975FBA2307501464139C4@US01WEMBX2.internal.synopsys.com>
References: <20181119112917.4214-1-abrodkin@synopsys.com>
Accept-Language: en-US
Content-Language: en-US
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 11/19/18 3:30 AM, Alexey Brodkin wrote:=0A=
> It turned out we used to use default implementation of sched_clock()=0A=
> from kernel/sched/clock.c which was as precise as 1/HZ, i.e.=0A=
> by default we had 10 msec granularity of time measurement.=0A=
>=0A=
> Now given ARC built-in timers are clocked with the same frequency as=0A=
> CPU cores we may get much higher precision of time tracking.=0A=
>=0A=
> Thus we switch to generic sched_clock which really reads ARC hardware=0A=
> counters.=0A=
>=0A=
> This is especially helpful for measuring short events.=0A=
> That's what we used to have:=0A=
> ------------------------------>8------------------------=0A=
> $ perf stat /bin/sh -c /root/lmbench-master/bin/arc/hello > /dev/null=0A=
>=0A=
>  Performance counter stats for '/bin/sh -c /root/lmbench-master/bin/arc/h=
ello':=0A=
>=0A=
>          10.000000      task-clock (msec)         #    2.832 CPUs utilize=
d=0A=
>                  1      context-switches          #    0.100 K/sec=0A=
>                  1      cpu-migrations            #    0.100 K/sec=0A=
>                 63      page-faults               #    0.006 M/sec=0A=
>            3049480      cycles                    #    0.305 GHz=0A=
>            1091259      instructions              #    0.36  insn per cyc=
le=0A=
>             256828      branches                  #   25.683 M/sec=0A=
>              27026      branch-misses             #   10.52% of all branc=
hes=0A=
>=0A=
>        0.003530687 seconds time elapsed=0A=
>=0A=
>        0.000000000 seconds user=0A=
>        0.010000000 seconds sys=0A=
> ------------------------------>8------------------------=0A=
>=0A=
> And now we'll see:=0A=
> ------------------------------>8------------------------=0A=
> $ perf stat /bin/sh -c /root/lmbench-master/bin/arc/hello > /dev/null=0A=
>=0A=
>  Performance counter stats for '/bin/sh -c /root/lmbench-master/bin/arc/h=
ello':=0A=
>=0A=
>           3.004322      task-clock (msec)         #    0.865 CPUs utilize=
d=0A=
>                  1      context-switches          #    0.333 K/sec=0A=
>                  1      cpu-migrations            #    0.333 K/sec=0A=
>                 63      page-faults               #    0.021 M/sec=0A=
>            2986734      cycles                    #    0.994 GHz=0A=
>            1087466      instructions              #    0.36  insn per cyc=
le=0A=
>             255209      branches                  #   84.947 M/sec=0A=
>              26002      branch-misses             #   10.19% of all branc=
hes=0A=
>=0A=
>        0.003474829 seconds time elapsed=0A=
>=0A=
>        0.003519000 seconds user=0A=
>        0.000000000 seconds sys=0A=
> ------------------------------>8------------------------=0A=
>=0A=
> Note how much more meaningful is the second output - time spent for=0A=
> execution pretty much matches number of cycles spent (we're runnign=0A=
> @ 1GHz here).=0A=
>=0A=
> Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>=0A=
> Cc: Daniel Lezcano <daniel.lezcano@linaro.org>=0A=
> Cc: Vineet Gupta <vgupta@synopsys.com>=0A=
> Cc: Thomas Gleixner <tglx@linutronix.de>=0A=
=0A=
Acked-by: Vineet Gupta <vgupta@synopsys.com>=0A=
=0A=
Thx,=0A=
-Vineet=0A=