* [PATCH/RFC] e1000: Check for hw type before enabling ipv6 TSO
@ 2007-03-26 23:19 Sridhar Samudrala
2007-03-26 23:48 ` Kok, Auke
0 siblings, 1 reply; 2+ messages in thread
From: Sridhar Samudrala @ 2007-03-26 23:19 UTC (permalink / raw)
To: auke-jan.h.kok; +Cc: netdev
I think the check for hw type is missing in e1000_set_tso()
before enabling ipv6 TSO.
e1000_probe() makes sure that NETIF_F_TSO6 is set only if
hw.mac_type > e1000_82547_rev_2. But this is not done in
e1000_set_tso() that gets called when trying to enable TSO
using ethtool.
I noticed this problem when i tried to do an ipv6 netperf
test with TSO enabled on 82541 based e1000 card.
The following patch adds this check to e1000_set_tso()
Thanks
Sridhar
Signed-off-by: Sridhar Samudrala <sri@us.ibm.com>
diff --git a/drivers/net/e1000/e1000_ethtool.c b/drivers/net/e1000/e1000_ethtool.c
--- a/drivers/net/e1000/e1000_ethtool.c
+++ b/drivers/net/e1000/e1000_ethtool.c
@@ -351,10 +351,12 @@ e1000_set_tso(struct net_device *netdev, uint32_t data)
else
netdev->features &= ~NETIF_F_TSO;
- if (data)
- netdev->features |= NETIF_F_TSO6;
- else
- netdev->features &= ~NETIF_F_TSO6;
+ if (adapter->hw.mac_type > e1000_82547_rev_2) {
+ if (data)
+ netdev->features |= NETIF_F_TSO6;
+ else
+ netdev->features &= ~NETIF_F_TSO6;
+ }
DPRINTK(PROBE, INFO, "TSO is %s\n", data ? "Enabled" : "Disabled");
adapter->tso_force = TRUE;
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: [PATCH/RFC] e1000: Check for hw type before enabling ipv6 TSO
2007-03-26 23:19 [PATCH/RFC] e1000: Check for hw type before enabling ipv6 TSO Sridhar Samudrala
@ 2007-03-26 23:48 ` Kok, Auke
0 siblings, 0 replies; 2+ messages in thread
From: Kok, Auke @ 2007-03-26 23:48 UTC (permalink / raw)
To: Sridhar Samudrala; +Cc: netdev, Jesse Brandeburg, Tantilov, Emil S
Sridhar Samudrala wrote:
> I think the check for hw type is missing in e1000_set_tso()
> before enabling ipv6 TSO.
>
> e1000_probe() makes sure that NETIF_F_TSO6 is set only if
> hw.mac_type > e1000_82547_rev_2. But this is not done in
> e1000_set_tso() that gets called when trying to enable TSO
> using ethtool.
>
> I noticed this problem when i tried to do an ipv6 netperf
> test with TSO enabled on 82541 based e1000 card.
>
> The following patch adds this check to e1000_set_tso()
that looks indeed like an oversight. I'll make sure this is verified and will
include this patch in my upstream submission to the kernel maintainer if it
checks out,
thanks for posting!
Auke
>
> Thanks
> Sridhar
>
> Signed-off-by: Sridhar Samudrala <sri@us.ibm.com>
>
> diff --git a/drivers/net/e1000/e1000_ethtool.c b/drivers/net/e1000/e1000_ethtool.c
> --- a/drivers/net/e1000/e1000_ethtool.c
> +++ b/drivers/net/e1000/e1000_ethtool.c
> @@ -351,10 +351,12 @@ e1000_set_tso(struct net_device *netdev, uint32_t data)
> else
> netdev->features &= ~NETIF_F_TSO;
>
> - if (data)
> - netdev->features |= NETIF_F_TSO6;
> - else
> - netdev->features &= ~NETIF_F_TSO6;
> + if (adapter->hw.mac_type > e1000_82547_rev_2) {
> + if (data)
> + netdev->features |= NETIF_F_TSO6;
> + else
> + netdev->features &= ~NETIF_F_TSO6;
> + }
>
> DPRINTK(PROBE, INFO, "TSO is %s\n", data ? "Enabled" : "Disabled");
> adapter->tso_force = TRUE;
>
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2007-03-26 23:48 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2007-03-26 23:19 [PATCH/RFC] e1000: Check for hw type before enabling ipv6 TSO Sridhar Samudrala
2007-03-26 23:48 ` Kok, Auke
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).