[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <d11aa2c9b888941385cbbd99000b94c6@walle.cc>
Date: Wed, 12 Apr 2023 13:12:54 +0200
From: Michael Walle <michael@...le.cc>
To: Vladimir Oltean <vladimir.oltean@....com>
Cc: kuba@...nel.org, gerhard@...leder-embedded.com, glipus@...il.com,
kory.maincent@...tlin.com, krzysztof.kozlowski+dt@...aro.org,
linux@...linux.org.uk, maxime.chevallier@...tlin.com,
netdev@...r.kernel.org, richardcochran@...il.com,
robh+dt@...nel.org, thomas.petazzoni@...tlin.com,
vadim.fedorenko@...ux.dev
Subject: Re: [PATCH net-next RFC v4 2/5] net: Expose available time stamping
layers to user space.
Am 2023-04-12 13:08, schrieb Vladimir Oltean:
> On Wed, Apr 12, 2023 at 12:50:34PM +0200, Michael Walle wrote:
>> >> +/* Hardware layer of the SO_TIMESTAMPING provider */
>> >> +enum timestamping_layer {
>> >> + SOF_MAC_TIMESTAMPING =3D (1<<0),
>> >> + SOF_PHY_TIMESTAMPING =3D (1<<1),
>> >
>> > What does SOF_ stand for?
>>
>> I'd guess start of frame. The timestamp will be taken at the
>> beginning of the frame.
>
> I would suggest (with all due respect) that it was an inapt adaptation
> of the Socket Option Flags that can be seen in
> Documentation/networking/timestamping.rst.
Agreed. Noticing the other SOF_TIMESTAMPING_* names, your
suggestion makes way more sense. Sorry for the noise.
-michael
Powered by blists - more mailing lists