[LEAPSECS] Leap second smearing test results

[Steve Summit]

zefram wrote:
> Steve Summit wrote:
>> which keep true UTC in the kernel (accepting it from NTP), and do
>> the smearing when user mode code asks for a time_t.
>
> Even better.  Presumably user-mode code calling adjtimex(2) would see the
> real UTC parameters,

That's one path.  (adjtimex has to remain unsmeared because it's
what ntpd looks at.)

> There should be some way for sufficiently aware user-mode code
> to detect this and recover real UTC.

Of course.  clock_gettime(CLOCK_UTC), as Markus Kuhn described
almost 20 years ago (and Warnecke and Kempen before that).