New blog draft on the TESTFRAME debacle

[Hal Murray]

>> Using the term PLL.
> But that's how the code is organized.  In the absence of a PLL it doesn't
> slew.  Or am I missing something here? 

(at least) One of us is confused.

The text from the blog:
> There are two kinds of NTP hosts; one uses a kernel facility known as the
> "PLL" for doing fine adjustments to the tick speed of the system clock, the
> other kind does not.

That just seems wrong.  There is no PLL involved with slewing a clock.

Do you have a reference for that terminology?

There may be confusion introduced from gpsd.  I think they have misused the 
term PLL in that sort of way.

If you had said something like:
  There are two kinds of NTP hosts; one which makes small adjustments by 
slewing the
  clock, the other kind does not.
Then I would have been happy.

I'd like to understand why slew or not has significant impact on the code 
path and/or TESTFRAME.

------

There is a PLL in the kernel that runs off PPS.  It's not in the default 
kernels shipped with most Linux distros because they all use noHz and the 
PPS/PLL doesn't work with that.

That would be a major shift in the code path.  When the kernel PPS/PLL is 
active, ntpd mostly just sits off to the side and watches.  That's an 
important path if you want super-good time, but TESTFRAME would have been a 
success if it didn't support that.

-- 
These are my opinions.  I hate spam.