Testing ntpd and/or timing from gpsd

[Gary E. Miller]

Yo Chris!

On Mon, 16 May 2016 10:30:37 +1000
Chris Johns <chrisj at ntpsec.org> wrote:

> > Not a factor at all.  In hardware PTP all the timestamping is done
> > in the ethernet chip itself.  The software jitter matters not at all
> > since the time was already captured in the hardware.  
> 
> The handling of the hardware time-stamp in the PTP stack is outside
> what I know but my naive view is the jitter you would see would be
> the distance in time from the point the hardware captures the counter
> to the point in time the software handles it.

Yup, naive.  On a PTP ethernet port there is acutally a clock in
the chip.  The hardware, on getting a PTP packet, adjusts that
local clock that is in the ethernet chip.

The software if not involved at all in the clock management, other
than time independent management duties.

At least whne it actuall works.


> The size of that jitter
> would be effected by any hardware delays and the overhead and
> performance of the operating system.

So, no.

> An RTOS with deterministic
> interrupt latency and thread dispatch times help bound that value.

In theory.  But sadly every RTOS I have played with has had compensating
flasws.  And since the PTP timestamp is captured in hardware the
latency is not relevant.

The only latency is in reading the ethernet chip clock and setting the
system clock from it.  No syscalls or interrupts in that path.

> I
> suspect all Xilinx is saying in the Zynq TRM is "with Linux your
> results may vary and if they do, do not call us, rather bound the
> latency some how, eg an RTOS".

FUD.  Tthey just don't understand, or want to support Linux.

RGDS
GARY
---------------------------------------------------------------------------
Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97703
	gem at rellim.com  Tel:+1 541 382 8588
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 473 bytes
Desc: OpenPGP digital signature
URL: <http://lists.ntpsec.org/pipermail/devel/attachments/20160515/10130e63/attachment.bin>