Re: [Linuxptp-devel] [PATCH] Add scaledLastGmFreqChange computation

[Erez <ere...@gm...>]

Hi,

I did not argue whether it is an error or not.

Once an enterprise management TLV is published (by a linuxptp official
version).
The TLV is a public API.
We should avoid changing them.
If the case is really bad, we may create a new enterprise management TLV
with the corrections.

TIME_STATUS_NP was added in september 2012 and published in version 1.1.
You can fix calculations of values in the TLV, but not sizes, types, order
or *NAMES*.

Wireshark should follow IEEE 1588 and linuxptp enterprise management TLVs,
not the other way around.

Erez




On Wed, 29 Nov 2023 at 07:52, Choong, Chwee Lin <chw...@in...>
wrote:

> Hi Erez, Richard,
>
> There's a potential misunderstanding among users who may believe that both
> lastGmPhaseChange and scaledLastGmPhaseChange are derived from phase
> measurements, with the latter representing a scaled version of the former.
> This assumption, particularly evident when they are displayed
> simultaneously in the PMC readout
>
> Additionally, it's worth noting that Wireshark has already addressed a
> similar naming discrepancy:
> https://www.wireshark.org/lists/wireshark-bugs//201602/msg00662.html
>
> Thanks
> CL
>
> -----Original Message-----
> From: Richard Cochran <ric...@gm...>
> Sent: Tuesday, November 28, 2023 7:42 AM
> To: Erez <ere...@gm...>
> Cc: Choong, Chwee Lin <chw...@in...>;
> lin...@li...
> Subject: Re: [Linuxptp-devel] [PATCH] Add scaledLastGmFreqChange
> computation
>
> On Mon, Nov 27, 2023 at 10:20:56AM +0100, Erez wrote:
> > It is public if you change the printed name in PMC.
>
> You mean scripts that invoke pmc?
>
> Yeah, that could cause users' setups to break.
>
> Thanks,
> Richard
>