Re: [PATCH v2 6/7] net: dsa: b53: Add logic for TX timestamping

On Wed, Nov 10, 2021 at 07:08:55AM -0800, Richard Cochran wrote:

In any case, I agree that the driver should enable the capabilities of
the HW and not impose artificial limitations.

Which was not my actual point. We were discussing whether the kernel
should attempt to take a 2-step TX timestamp if a different one is
outstanding, be it from the same socket or from a different one.
Whether the hardware supports multiple concurrent TX timestamps is
orthogonal to that.