[Linuxptp-devel] More fun with i210 EXTTS: measure some PPS against a reference PPS

2020-03-05 Thread Frantisek Rysanek
Dear gentlemen, after two years of silence, I've played some more with my multiport i210 test rig... I now have 4 ports of i210, two of them metallic, two of them fiber optic (with Gigabit SERDES SFP slots). And I've hacked the optical cards to have SDP0 and SDP3 inputs, as well as the obvious

Re: [Linuxptp-devel] More fun with i210 EXTTS: measure some PPS against a reference PPS

2020-03-05 Thread Jacob Keller
On 3/5/2020 1:11 AM, Frantisek Rysanek wrote: > And another sideways question is: in the i210 hardware, there's a > register called SYSTIMR, supposedly holding the fraction of a > nanosecond (= sub-nanosecond resolution). And this register is > ignored by the igb driver in Linux - first and

Re: [Linuxptp-devel] [PATCH RFC 06/30] interface: Introduce an access method for the name field.

2020-03-05 Thread Jacob Keller
On 3/4/2020 8:56 AM, Richard Cochran wrote: > On Tue, Feb 18, 2020 at 11:33:32AM -0800, Jacob Keller wrote: >> So the interface.o isn't being added to something like $(CONFIG)? > > I like the idea, but I'll leave that as a follow-on task. > > Thanks, > Richard > Yea makes sense. I don't see a

Re: [Linuxptp-devel] [PATCH RFC 07/30] Convert call sites to the proper method for getting interface names.

2020-03-05 Thread Jacob Keller
On 3/4/2020 8:59 AM, Richard Cochran wrote: > On Tue, Feb 18, 2020 at 11:38:43AM -0800, Jacob Keller wrote: >> >> Is there a way to generate the network of how interconnected the various >> object files are? > > The .d files have the includes. I once wrote a script to turn the .d > files into a

Re: [Linuxptp-devel] More fun with i210 EXTTS: measure some PPS against a reference PPS

2020-03-05 Thread Jacob Keller
On 3/5/2020 8:19 AM, Frantisek Rysanek wrote: > during the initial settling of the PHC frequency / servo loop, > I can see offsets in low units of ns, not aligned in any way. > But: once the PHC settles to offset==0 && ppb==0, > any measured edges captured via EXTTS channel 0 or 1 > will be

Re: [Linuxptp-devel] [PATCH RFC 28/30] interface: Hide the implementation details.

2020-03-05 Thread Jacob Keller
On 3/4/2020 9:24 AM, Richard Cochran wrote: > On Tue, Feb 18, 2020 at 01:32:01PM -0800, Jacob Keller wrote: >> I'd appreciate if there was some way to ensure we catch the interface >> structure layout changing such that the definitions in clock.c and >> config.c aren't compatible with interface.c

Re: [Linuxptp-devel] [PATCH RFC 20/30] interface: Introduce a method to test the time stamping information validity.

2020-03-05 Thread Jacob Keller
On 3/4/2020 9:13 AM, Richard Cochran wrote: > On Tue, Feb 18, 2020 at 01:19:09PM -0800, Jacob Keller wrote: >>> +bool interface_tsinfo_valid(struct interface *iface) >>> +{ >>> + return iface->ts_info.valid ? true : false; >>> +} >> >> Do you actually need the ternary here? shouldn't

Re: [Linuxptp-devel] [PATCH RFC 24/30] interface: Introduce methods to create and destroy instances.

2020-03-05 Thread Jacob Keller
On 3/4/2020 9:19 AM, Richard Cochran wrote: > On Tue, Feb 18, 2020 at 01:25:00PM -0800, Jacob Keller wrote: >> >> I still think it would be good to have the functions guarantee the NULL >> by manually assigning or using one of the string copy implementations >> that will guarantee it. That way

Re: [Linuxptp-devel] [PATCH RFC 14/30] interface: Introduce a method to set the name.

2020-03-05 Thread Jacob Keller
On 3/4/2020 9:06 AM, Richard Cochran wrote: > On Tue, Feb 18, 2020 at 01:07:52PM -0800, Jacob Keller wrote: >> >> Good, the name is marked as constant. Side note, for those interface_* >> functions that don't modify the interface, does it make sense to mark >> them as taking a const interface

Re: [Linuxptp-devel] More fun with i210 EXTTS: measure some PPS against a reference PPS

2020-03-05 Thread Frantisek Rysanek
...so I've tried myselfs. I've re-written my proggie to first settle the PHC using EXTTS CH0 to the external reference PPS, and then switch EXTTS CH0 to the measured PPS signal (reconfigure the CH0 to take input from the second SDP input). Guess what: the measured deviations are still quantised