GMAC Precision time protocol

Discussion around product based on ARM Cortex M4 core.

Moderators: nferre, ncollot

Location: Auckland, New Zealand
Posts: 14
Joined: Mon Jun 04, 2007 6:57 pm

GMAC Precision time protocol

Sun Mar 13, 2016 10:56 pm

First off, is anyone aware of any minimal implementations of the precision time protocol (PTP) for this IC? Otherwise I'll continue with my own...

Secondly, it appears as though the datasheet in section 42.6.14 "Time Stamp Unit" is referring to a non-existent register GMAC_TISUBN. The description all sounds very impressive, giving 15 femtosecond resolution, etc but the register is not listed in section 42.8 and does not appear to exist in the API.

In the past I have frequently tried to use various documented GMAC functions only to find that they don't actually exist (statistics is one example). I now just hope that all of the necessary functionality for PTP really does exist and that working on this doesn't lead to a dead end.
Posts: 3
Joined: Thu Jul 21, 2016 4:30 pm

Re: GMAC Precision time protocol

Fri Jul 22, 2016 4:08 pm


The latest documentation (02/2016) mentions the GMAC_TISUBN.
The board I have it pretty 'old' (older than that) and I confirm that the GMAC_TISUBN
does not have any effect.

Assuming the clock rate of GMAC is 132Mhz, I should have an increment of 7 nsecs
+ a delta... As the delta in subns register is ignored, I measure a late jitter of about 4-5 secs per minute ...
Same when writing 0xffff.

I have tried to used the (legacy ?) "alternate" NIT and ACNS fields in the increment register.
(1000 = 7*132 + 76;
132 = 0x84; 76 = 0xc4 so the increment register is set to 0x844c07)

The result is somehow better, but I am still jittering , about 1 sec per 3 min, which is
still very bad, and make ptp4l unusable.

Did you make any progress on your side ? I am expecting a more recent board, hopefully
it has the new 'femto' feature.

Best regards

Return to “SAM4 Cortex-M4 MCU”

Who is online

Users browsing this forum: No registered users and 3 guests