On Fri, 2008-04-18 at 11:02 +0200, Holger Schurig wrote:
> NACK.
Thanks for testing.
> When I apply this patch and do "ethtool -S eth1", I get this:
>
> $ ethtool -S eth1
> NIC statistics:
> drop_duplicate_bcast: 7
> drop_ttl_zero: 0
> drop_no_fwd_route: 0
> drop_no_buffers: 9223372036854775808
> fwded_unicast_cnt: 0
> fwded_bcast_cnt: 0
> drop_blind_table: 0
> tx_failed_cnt: 9223372036854775808
You were getting that anyway.
> But more severely, I get this in dmesg:
>
> libertas enter: lbs_ethtool_get_stats()
> libertas: PREP_CMD: command 0x009b failed: 2
Now at least it's _trying_ to get the stats :)
> My firmware simply doesn't support the 009b command. It seems
> also wrong to call an MESH related command on an struct
> net_device *dev which is not priv->mesh_dev.
>
>
> That ethtool returns junk like "9223372036854775808" is another
> error. lbs_mesh_access() returned 2, and the code does "if (ret)
> return ret". Maybe it should be like "if (ret) return -ENOSYS
> (or some other ENOxxx) instead?
The code does 'if (ret) return;', because it has no option to return an
error. I think the way round it would be something like this:
diff --git a/drivers/net/wireless/libertas/ethtool.c b/drivers/net/wireless/libertas/ethtool.c
index 21e6f98..2ca3f4f 100644
--- a/drivers/net/wireless/libertas/ethtool.c
+++ b/drivers/net/wireless/libertas/ethtool.c
@@ -146,12 +144,12 @@ static void lbs_ethtool_get_stats(struct net_device * dev,
static int lbs_ethtool_get_sset_count(struct net_device * dev, int sset)
{
- switch (sset) {
- case ETH_SS_STATS:
+ struct lbs_private *priv = dev->priv;
+
+ if (sset == ETH_SS_STATS && dev == priv->mesh_dev)
return MESH_STATS_NUM;
- default:
+ else
return -EOPNOTSUPP;
- }
}
static void lbs_ethtool_get_strings(struct net_device *dev,
--
dwmw2
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
