From: Eric Wong <normalperson@...>
Date: 2017-05-28T02:02:12+00:00
Subject: [ruby-core:81430] Re: Fwd: [ruby-changes:46809] normal:r58924 (trunk): test for IO.copy_stream CPU usage (r58534)

SASADA Koichi <ko1@atdot.net> wrote:
> On 2017/05/28 10:47, Eric Wong wrote:
> > Can you check the CONFIG_HZ on the Linux machines?
> > It's in /proc/config.gz if enabled by the kernel; and some distros
> > have it in /boot/config-$KERNEL_VERSION
> 
> ko1@ruby-sky3:~$ cat /boot/config-4.4.0-72-generic  | grep CONFIG_HZ
> # CONFIG_HZ_PERIODIC is not set
> # CONFIG_HZ_100 is not set
> CONFIG_HZ_250=y
> # CONFIG_HZ_300 is not set
> # CONFIG_HZ_1000 is not set
> CONFIG_HZ=250
> 
> 10:51:01 ko1@sasada-8core:~$ cat /boot/config-3.2.0-4-amd64 |grep CONFIG_HZ
> # CONFIG_HZ_100 is not set
> CONFIG_HZ_250=y
> # CONFIG_HZ_300 is not set
> # CONFIG_HZ_1000 is not set
> CONFIG_HZ=250
> 
> Do they help you?

Yes, thank you. I think 250Hz may also be OK with lower
threshold after r58935 (Process.times using getrusage);
but I guess there can still be machines with 100Hz where
neither times() nor getrusage() is accurate to <millisecond.

Unsubscribe: <mailto:ruby-core-request@ruby-lang.org?subject=unsubscribe>
<http://lists.ruby-lang.org/cgi-bin/mailman/options/ruby-core>