[ofa-general] Re: [PATCH V4 8/10] net/bonding: Handlle wrong assumptions that slave is always an Ethernet device
Moni Shoua
monisonlists at gmail.com
Wed Aug 29 06:37:34 PDT 2007
Jay Vosburgh wrote:
> Moni Shoua <monis at voltaire.com> wrote:
>
>> bonding sometimes uses Ethernet constants (such as MTU and address length) which
>> are not good when it enslaves non Ethernet devices (such as InfiniBand).
>>
>> Signed-off-by: Moni Shoua <monis at voltaire.com>
>> ---
>> drivers/net/bonding/bond_main.c | 3 ++-
>> drivers/net/bonding/bond_sysfs.c | 19 +++++++++++++------
>> drivers/net/bonding/bonding.h | 1 +
>> 3 files changed, 16 insertions(+), 7 deletions(-)
>>
>> Index: net-2.6/drivers/net/bonding/bond_main.c
>> ===================================================================
>> --- net-2.6.orig/drivers/net/bonding/bond_main.c 2007-08-15 10:55:48.000000000 +0300
>> +++ net-2.6/drivers/net/bonding/bond_main.c 2007-08-20 14:29:11.911298577 +0300
>> @@ -1224,7 +1224,8 @@ static int bond_compute_features(struct
>> struct slave *slave;
>> struct net_device *bond_dev = bond->dev;
>> unsigned long features = bond_dev->features;
>> - unsigned short max_hard_header_len = ETH_HLEN;
>> + unsigned short max_hard_header_len = max((u16)ETH_HLEN,
>> + bond_dev->hard_header_len);
>
> Since non-IB bonding masters are run through ether_setup, which
> sets hard_header_len to ETH_HLEN, the max() is probably unnecessary, and
> I think this could just be bond_dev->hard_header_len.
>
This is true except for the case where there are no slaves left.
In that case max_hard_header_len has equals to the initialization value.
bond_for_each_slave(bond, slave, i) {
features = netdev_compute_features(features,
slave->dev->features);
if (slave->dev->hard_header_len > max_hard_header_len)
max_hard_header_len = slave->dev->hard_header_len;
}
features |= (bond_dev->features & BOND_VLAN_FEATURES);
bond_dev->features = features;
bond_dev->hard_header_len = max_hard_header_len;
More information about the general
mailing list