> I think that suggests there is a difference in the HW. If the two
> boards performed exactly the same, there should be no difference in
> the measured offset.
>
> I'm not sure what that could be. Maybe the CPU is throttling due to
> high temperature? On x86_64 frequency scaling (power saving, tu
On Sun, Jun 09, 2019 at 03:31:32AM +0200, Mino Sharkhawy wrote:
> > > What happens when you swap the master/slave roles? Does the offset
> > > change? Only the sign, or also the absolute value?
> > Good idea! I'll give it a try as soon as I can.
>
> So I swapped the master/slave roles and let it r
> > What happens when you swap the master/slave roles? Does the offset
> > change? Only the sign, or also the absolute value?
> Good idea! I'll give it a try as soon as I can.
So I swapped the master/slave roles and let it run for another 48 hours.
Both sign and absolute value changed...
I have a
> Hm. Interesting. If the HW, SW and PPS are identical, I'm not sure
> what else it could be.
Yeah, that's my problem, I'd at least like to understand where this is
coming from. But without being able to measure the PHCs directly it's
tricky to locate the cause.
> What happens when you swap the ma
On Wed, Jun 05, 2019 at 03:46:38PM +0200, Mino Sharkhawy wrote:
> > One explanation would be an asymmetry in phc2sys. Synchronizing the
> > system clock to the PHC uses a more accurate method for measuring the
> > offset than synchronizing the PHC to the system clock.
> >
> > This asymmetry was fi
> One explanation would be an asymmetry in phc2sys. Synchronizing the
> system clock to the PHC uses a more accurate method for measuring the
> offset than synchronizing the PHC to the system clock.
>
> This asymmetry was fixed only recently after linuxptp-2.0 was
> released. Have you tried phc2sy
On Tue, Jun 04, 2019 at 01:42:35AM +0200, Mino Sharkhawy wrote:
> Using this setup and plotting a histogram of the offsets measured on the
> slave (between system clock and GPS reference) over two days, I noticed
> that while the offsets roughly follow a normal distribution, they are
> centered at
Hi,
I've been experimenting with two Beaglebone Blacks and Linuxptp to measure
the accuracy of the clock synchronization. I use a GPS receiver with a 1PPS
Signal as a reference clock. On my "master" Beaglebone, the system clock is
synchronized to the GPS/PPS with NTP and the kernel PPS driver. I t
