Re: Large inlines in include/linux/skbuff.h

[Andi Kleen]

On Thu, Apr 22, 2004 at 03:25:25PM -0700, David S. Miller wrote:
> On 23 Apr 2004 00:23:26 +0200
> Andi Kleen <ak@xxxxxx> wrote:
> 
> > > Of course, but it would be good to see some measurements.
> > 
> > It's useless in this case. networking is dominated by cache misses
> > and locks and a modern CPU can do hundreds of function calls in a 
> > single cache miss.
> 
> Indeed, this change does influence the I-cache footprint of the
> networking, so I conclude that it is relevant in this case.

All it does it is to lower the cache foot print, not enlarge it.
So even if it made a difference it could only get better.
It's extremly unlikely to make it detectable worse.

-Andi
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/