Re: [RFC] csum experts, csum_replace2() is too expensive
From: Andi Kleen
Date: Fri Mar 21 2014 - 09:28:35 EST
On Fri, Mar 21, 2014 at 05:50:50AM -0700, Eric Dumazet wrote:
> On Thu, 2014-03-20 at 18:56 -0700, Andi Kleen wrote:
> > Eric Dumazet <eric.dumazet@xxxxxxxxx> writes:
> > >
> > > I saw csum_partial() consuming 1% of cpu cycles in a GRO workload, that
> > > is insane...
> > Couldn't it just be the cache miss?
> Or the fact that we mix 16 bit stores and 32bit loads ?
It should cause a small stall from not doing load-store
forwarding, but 1% of a serious workload would be surprising.
Are you sure it's not some skid effect?
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/