Re: [chrony-dev] PPSAPI: kernel consumer

[Alexander Gordeev <lasaine@xxxxxxxxxxxxx>]

On Mon, 8 Feb 2010 15:47:43 -0800 (PST)
Bill Unruh <unruh@xxxxxxxxxxxxxx> wrote:

> I had a quick look at the code you have suggested adding to the
> kernel, and am really bothered by a bunch of hard loops you have in
> there waiting for some time to occur or some event to happen. That
> stikes me as a real waste of computer resources. The cpu is spinning
> its wheels for up to a second waiting for an event to happen. Surely
> you should be using interrupts (times interrupts or parallel port
> interrupts) not busy loops doing nothing but tying up the computer so
> nothing else can be accomplished either.

Do you think I'm crazy?
Please have a deeper look and you will see that both the loops run for
about 30us by default. This is required to not suffer from interrupt
handling delays. It spends only 30us per second busy looping but
instead I can be sure that each timestamp is within 1-1.5us (port
read/write time) from the actual clear edge.

> I also am confused about the philosophy. Why not just run a line from
> the primary GPS ( with a buffer amplifier) to the various other
> computers, rather than  having one computer strobe the parallel line,
> and the others responding to that?

We do not have the primary GPS and do not want to have one. There is a
number of reasons for that.

> What kind of accuracy are you hoping to get and why? I strongly doubt
> that you can control/stamp anything on your system to better than 1us
> accuracy, just because of the kernel overhead, etc.

We haven't yet done tests with oscilloscope but I'm pretty sure that
time divergence is within 1-3us.

But this is off-topic...

-- 
  Alexander

Attachment: signature.asc
Description: PGP signature