[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20070510192814.5b71d01c@freepuppy>
Date: Thu, 10 May 2007 19:28:14 -0700
From: Stephen Hemminger <shemminger@...ux-foundation.org>
To: Gagan Arneja <gaagaan@...il.com>
Cc: David Miller <davem@...emloft.net>, rick.jones2@...com,
vladislav.yasevich@...com, ian.mcdonald@...di.co.nz,
krkumar2@...ibm.com, johnpol@....mipt.ru, netdev@...r.kernel.org
Subject: Re: [RFC] New driver API to speed up small packets xmits
On Thu, 10 May 2007 14:14:05 -0700
Gagan Arneja <gaagaan@...il.com> wrote:
> David Miller wrote:
> > From: Rick Jones <rick.jones2@...com>
> > Date: Thu, 10 May 2007 13:49:44 -0700
> >
> >> I'd think one would only do this in those situations/places where a
> >> natural "out of driver" queue develops in the first place wouldn't
> >> one?
> >
> > Indeed.
>
> And one builds in qdisc because your device sink is slow. There's
> nothing inherently there in the kernel encouraging Qdisc->q to grow.
> It's precisely these broken devices that can take advantage of cluster
> transmits.
If you have braindead slow hardware,
there is nothing that says your start_xmit routine can't do its own
coalescing. The cost of calling the transmit routine is the responsibility
of the driver, not the core network code.
--
Stephen Hemminger <shemminger@...ux-foundation.org>
-
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists