* doubt about qla3xxx checksum "fix"
@ 2007-06-04 1:00 David Miller
2007-06-04 1:50 ` Stephen Hemminger
0 siblings, 1 reply; 3+ messages in thread
From: David Miller @ 2007-06-04 1:00 UTC (permalink / raw)
To: netdev; +Cc: shemminger, jeff
I don't think it's %100 right, I'm referring specifically to
e68a8c10c4c5daf363e946d10c1a5cba77d7f92c
Sure, qla3xxx is only handling ipv4 in that ql_hw_csum_setup()
function, but if you check the call site it really wants
CHECKSUM_PARTIAL skb's to work on. And CHECKSUM_PARTIAL skbs are only
setup for paths that advertise the NETIF_F_HW_CSUM capability.
With Stephen's change it now won't checksum offload anything at all.
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: doubt about qla3xxx checksum "fix"
2007-06-04 1:00 doubt about qla3xxx checksum "fix" David Miller
@ 2007-06-04 1:50 ` Stephen Hemminger
2007-06-04 2:03 ` David Miller
0 siblings, 1 reply; 3+ messages in thread
From: Stephen Hemminger @ 2007-06-04 1:50 UTC (permalink / raw)
To: David Miller; +Cc: netdev, jeff
On Sun, 03 Jun 2007 18:00:51 -0700 (PDT)
David Miller <davem@davemloft.net> wrote:
>
> I don't think it's %100 right, I'm referring specifically to
>
> e68a8c10c4c5daf363e946d10c1a5cba77d7f92c
>
> Sure, qla3xxx is only handling ipv4 in that ql_hw_csum_setup()
> function, but if you check the call site it really wants
> CHECKSUM_PARTIAL skb's to work on. And CHECKSUM_PARTIAL skbs are only
> setup for paths that advertise the NETIF_F_HW_CSUM capability.
CHECKSUM_PARTIAL is used on xmit for both NETIF_F_HWCSUM and NETIF_F_IPCSUM
See acenic, 3c59x, ...
--
Stephen Hemminger <shemminger@linux-foundation.org>
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: doubt about qla3xxx checksum "fix"
2007-06-04 1:50 ` Stephen Hemminger
@ 2007-06-04 2:03 ` David Miller
0 siblings, 0 replies; 3+ messages in thread
From: David Miller @ 2007-06-04 2:03 UTC (permalink / raw)
To: shemminger; +Cc: netdev, jeff
From: Stephen Hemminger <shemminger@linux-foundation.org>
Date: Sun, 3 Jun 2007 18:50:59 -0700
> On Sun, 03 Jun 2007 18:00:51 -0700 (PDT)
> David Miller <davem@davemloft.net> wrote:
>
> >
> > I don't think it's %100 right, I'm referring specifically to
> >
> > e68a8c10c4c5daf363e946d10c1a5cba77d7f92c
> >
> > Sure, qla3xxx is only handling ipv4 in that ql_hw_csum_setup()
> > function, but if you check the call site it really wants
> > CHECKSUM_PARTIAL skb's to work on. And CHECKSUM_PARTIAL skbs are only
> > setup for paths that advertise the NETIF_F_HW_CSUM capability.
>
> CHECKSUM_PARTIAL is used on xmit for both NETIF_F_HWCSUM and NETIF_F_IPCSUM
> See acenic, 3c59x, ...
I missed that, thanks a lot Stephen.
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2007-06-04 2:03 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2007-06-04 1:00 doubt about qla3xxx checksum "fix" David Miller
2007-06-04 1:50 ` Stephen Hemminger
2007-06-04 2:03 ` David Miller
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).