Subject: Re: Interface traffic timestamps
To: Martin Husemann <>
From: Nathan J. Williams <>
List: tech-kern
Date: 06/11/2004 18:55:45
Martin Husemann <> writes:

> On Fri, Jun 11, 2004 at 11:35:07AM -0700, Jason Thorpe wrote:
>> Yah, perhaps seconds is good enough ... then you pretty much definitely 
>> do not have to worry about spl frobbing for a decent "ballpark" 
>> timestamp.
> Yes - seconds is good enough and I would just not do any locking - but
> overall this all sounds hackish and I'm open to suggestions for better
> ways.

It sounds like what you want is something that could say "there
{was,was not}" a packet in the last N seconds, which the timeout
mechanism could check when it fires, and then clear. Writing a single
flag value repeatedly into a word will be much cheaper (and
cache-friendly) than using the actual time value.

                - Nathan