linux-kernel - Re: [PATCH net-next v2 2/2] net: stmmac: Add a devlink attribute to control timestamping mode

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <71310577-7cea-42ce-8442-49e09e0b958a@bootlin.com>
Date: Wed, 29 Oct 2025 07:59:10 +0100
From: Maxime Chevallier <maxime.chevallier@...tlin.com>
To: Jakub Kicinski <kuba@...nel.org>
Cc: Alexandre Torgue <alexandre.torgue@...s.st.com>,
 Jose Abreu <joabreu@...opsys.com>, Andrew Lunn <andrew+netdev@...n.ch>,
 davem@...emloft.net, Eric Dumazet <edumazet@...gle.com>,
 Paolo Abeni <pabeni@...hat.com>, Maxime Coquelin
 <mcoquelin.stm32@...il.com>, Richard Cochran <richardcochran@...il.com>,
 Russell King <linux@...linux.org.uk>,
 Köry Maincent <kory.maincent@...tlin.com>,
 Vadim Fedorenko <vadim.fedorenko@...ux.dev>,
 Alexis Lothoré <alexis.lothore@...tlin.com>,
 Thomas Petazzoni <thomas.petazzoni@...tlin.com>, netdev@...r.kernel.org,
 linux-stm32@...md-mailman.stormreply.com,
 linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH net-next v2 2/2] net: stmmac: Add a devlink attribute to
 control timestamping mode

Hi Jakub,

On 28/10/2025 23:19, Jakub Kicinski wrote:
> Sorry didn't get to review this in time.
> 
> On Fri, 24 Oct 2025 09:07:18 +0200 Maxime Chevallier wrote:
>> +   * - ``ts_coarse``
> 
> This is not a great name IMHO. Is "coarse" from the PRM?

Yes, it uses "fine/coarse"

> It's the increment that's coarse, right? Not the timestamp
> This naming confuses me greatly.

That is true, the ts_ was added as this configuration is done based
on the timestamping control registers, and is refered to as
"timestamping control fine update" in the register defs :(

So you are correct that in the end the clock frequency is coarsely
adjusted.

The patch was applied, should we revert or add another patch to rename
that parameter ?

> 
>> +     - Boolean
>> +     - runtime
>> +     - Enable the Coarse timestamping mode. In Coarse mode, the ptp clock is
>> +       expected to be updated through an external PPS input, but the subsecond
> 
> I guess the definition of "PPS input" got diluted but technically it
> means Pulse Per Second, right? Here IIUC we need an actual 50MHz clock
> fed in?

For GM, yes indeed. I can update the doc accordingly.

> 
>> +       increment used for timestamping is set to 1/ptp_clock_rate. In Fine mode
>> +       (i.e. Coarse mode == false), the ptp clock frequency is adjusted more
>> +       frequently, but the subsecond increment is set to 2/ptp_clock_rate.
>> +       Coarse mode is suitable for PTP Grand Master operation. If unsure, leave
>> +       the parameter to False.
> 
> My understanding based on your previous explanation is that basically
> in one of the modes the frequency cannot be adjusted. It's only usable
> if a very stable reference clock is fed into the device (or otherwise
> we "trust" the clock that's fed in). So that's why Grand Master.
> 
> In the other mode we can tweak the frequency more accurately.
> But it comes at a cost of the HW time incrementing 2x larger step.
> 
> If that's the case I think we should update the documentation and
> rename the knob to indicate that it's the frequency adjustment that's
> coarse.

That's fine by me, just let me know abut the exact process, I can followup
on that :)

Maxime