[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20110131224032.GA23032@rere.qmqm.pl>
Date: Mon, 31 Jan 2011 23:40:32 +0100
From: Michał Mirosław <mirq-linux@...e.qmqm.pl>
To: Stephen Hemminger <shemminger@...tta.com>
Cc: netdev@...r.kernel.org, Ben Hutchings <bhutchings@...arflare.com>
Subject: Re: [PATCH v2 16/16] skge: convert to hw_features
On Mon, Jan 31, 2011 at 10:53:43PM +0100, Michał Mirosław wrote:
> On Mon, Jan 31, 2011 at 08:45:02AM -0800, Stephen Hemminger wrote:
> > On Sat, 22 Jan 2011 23:14:14 +0100 (CET)
> > Michał Mirosław <mirq-linux@...e.qmqm.pl> wrote:
> > > The hardware might do full HW_CSUM, not just IP_CSUM, but it's not tested
> > > and so not changed here.
> > > Signed-off-by: Michał Mirosław <mirq-linux@...e.qmqm.pl>
> > The skge hardware does not do full HW_CSUM. It looks at the IP header.
> Driver code suggests otherwise. Please look at skge_xmit_frame() and TX
> descriptor format.
>
> Besides, I'm locally using a patch that makes skge advertise HW_CSUM - TCP
> over IPv6 is correctly checksummed in hardware (haven't tried VLANs yet).
IPv4 UDP inside VLAN-tagged packet is also checksummed properly by hardware.
I don't have time now to make a proper HW_CSUM test (with arbitrary header
prepended). If someone has one (or patches?), I'd be glad to use it, though.
Best Regards,
Michał Mirosław
--
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