From: Tanaka Akira <akr@...>
Date: 2013-07-24T10:50:31+09:00
Subject: [ruby-core:56135] Re: [ruby-trunk - Feature #8658] Process.clock_gettime

2013/7/24 KOSAKI Motohiro <kosaki.motohiro@gmail.com>:

> So, Why do you choice Process.clock_gettime() instead of
> Time.clock_gettime()?

I don't like the result value of clock_gettime(CLOCK_REALTIME) because
the value is interpreted differently between systems which use
leapseconds and not.
Time.now should be used instead.

The patch defines Process::CLOCK_REALTIME but it is just for consistency.
I felt definining CLOCK_* constants except CLOCK_REALTIME is too inconsistent.

The expected my main usecase would be CLOCK_PROCESS_CPUTIME_ID (or
CLOCK_THREAD_CPUTIME_ID).
For example, I use Process.times for measure Bignum speed but
Process.times cannot measure under 10ms on my environment.
Repeating target operation (as I do) improves precision but
high resolution clocks can be used to obtain similar precision
with less repeatation.
-- 
Tanaka Akira