[ntp:hackers] odd behaviour (one more time)

David L. Mills mills at udel.edu
Fri Oct 3 18:31:17 UTC 2008


For clock discipline loop engineering, see Chapter 4 in my book.

I don't knoe what Linux has done to the kernel code that left here. The 
code that left here (nanokernel.tar.gz) has a variable hz presumed equal 
to the timer interrupt frequency. The discipline loop closely simulates 
the math, so no simulation is necessary.


Reg Clemens wrote:

>>> Another nit with Linux is that some configurations come up with the
>>> timer frequency of 1000 Hz rather that the design parameter of 100 Hz.
>>> It is possible, and indeed is the case for Ultrix (256 Hz) and OSF/1
>>> (1024 Hz) to change the loop parameters to match any frequency, but
>>> Linux hasn't done that.
>> Thats probably it.
>> The kernel in question was built with a 1000 Hz Clock.
> Could you tell me (actually, point me at a previous report) how you 
> compute
> the loop parameters for a given frequency, and which ones need to be 
> changed.
> Is this a straightforward calculation, or do you iterate with the 
> simulator?

More information about the hackers mailing list