[PATCH -next] myri10ge: clear NETIF_F_LRO bit directly

[PATCH -next] myri10ge: clear NETIF_F_LRO bit directly

[Stanislaw Gruszka]

Do not use ethtool_op_set_flags() to clear one bit in ->features.
Inform user about disabling LRO.

Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
---
 drivers/net/myri10ge/myri10ge.c |    8 +++-----
 1 files changed, 3 insertions(+), 5 deletions(-)

diff --git a/drivers/net/myri10ge/myri10ge.c b/drivers/net/myri10ge/myri10ge.c
index e0b47cc..2259168 100644
--- a/drivers/net/myri10ge/myri10ge.c
+++ b/drivers/net/myri10ge/myri10ge.c
@@ -1725,17 +1725,15 @@ static u32 myri10ge_get_rx_csum(struct net_device *netdev)
 static int myri10ge_set_rx_csum(struct net_device *netdev, u32 csum_enabled)
 {
 	struct myri10ge_priv *mgp = netdev_priv(netdev);
-	int err = 0;
 
 	if (csum_enabled)
 		mgp->csum_flag = MXGEFW_FLAGS_CKSUM;
 	else {
-		u32 flags = ethtool_op_get_flags(netdev);
-		err = ethtool_op_set_flags(netdev, (flags & ~ETH_FLAG_LRO));
 		mgp->csum_flag = 0;
-
+		netdev->features &= ~NETIF_F_LRO;
+		netdev_info(netdev, "RX checksumming set off, disabling LRO\n");
 	}
-	return err;
+	return 0;
 }
 
 static int myri10ge_set_tso(struct net_device *netdev, u32 tso_enabled)
-- 
1.5.5.6

Re: [PATCH -next] myri10ge: clear NETIF_F_LRO bit directly

[Andrew Gallatin]

Stanislaw Gruszka wrote:
> Do not use ethtool_op_set_flags() to clear one bit in ->features.
> Inform user about disabling LRO.

Thanks.. That simplifies things nicely.  But was
direct manipulation of netdev->features ever discouraged,
or has my use of ethtool_op_{get,set}_flags() to manipulate
the features always been complete overkill?

> Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>

Acked-by: Andrew J. Gallatin <gallatin@myri.com>


Thanks,

Drew

Re: [PATCH -next] myri10ge: clear NETIF_F_LRO bit directly

[Stanislaw Gruszka]

On Tue, 29 Jun 2010 11:44:14 -0400
Andrew Gallatin <gallatin@myri.com> wrote:

> Stanislaw Gruszka wrote:
> > Do not use ethtool_op_set_flags() to clear one bit in ->features.
> > Inform user about disabling LRO.
> 
> Thanks.. That simplifies things nicely.  But was
> direct manipulation of netdev->features ever discouraged,
> or has my use of ethtool_op_{get,set}_flags() to manipulate
> the features always been complete overkill?

It was a bit overkill. Manipulate netdev->features is fine,
many drivers do it.

> > Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
> 
> Acked-by: Andrew J. Gallatin <gallatin@myri.com>

Thanks. FYI: Ben Hutchings posted other patch spread over
all drivers: http://patchwork.ozlabs.org/patch/57286/
we most likely will use it to put this myri10ge bits in.

Stanislaw