Mark Kettenis <[email protected]> wrote:

> I did not add the LFENCE to tc_get_timecount() itself.  We probably
> should, but in the past some networking folks complained about slow
> timecounters affecting network performance, so I wanted confirmation
> that this didn't cause problems.

I don't see how a more precise time can cause such a problem.  If
we are worried about any possibility of negative deltas, which might
be put into unsigned variables, we should use lfence so it can't
appear to go backwards..

Reply via email to