bonding: ensure that the TSO being set on bond master

PATCH Superseded Ding Tianhong 2 participants 3 messages Jan 8, 2014 – Jan 14, 2014

Subsystems (2)

Networking Network drivers

lore

Share & tools

lore

Ding Tianhong

author

2014-01-08 07:28 UTC #0 lore raw

cover bonding: ensure that the TSO being set on bond master

The commit b0ce3508(bonding: allow TSO being set on bonding master) has make the TSO being set for bond dev, but in some situation, if the slave did not have the NETIF_F_SG features, the bond master will miss the TSO features in netdev_fix_features because the TSO is depended on SG. So I have to add SG and TSO features on bond master together.

The function netdev_add_tso_features() was only be used for bonding, so no need to export it in netdevice.h, remove it and add it to bonding.

Ding Tianhong (2):
bonding: ensure that the TSO being set on bond master
net: remove the unused function netdev_add_tso_features()

 drivers/net/bonding/bond_main.c | 16 +++++++++++++++-
 include/linux/netdevice.h       | 10 ----------
 2 files changed, 15 insertions(+), 11 deletions(-)

Signature · 1 line

1.8.0

David Miller

maintainer

2014-01-13 19:15 UTC #1 lore raw

From: Ding Tianhong <[email protected]>
Date: Wed, 8 Jan 2014 15:28:21 +0800

Quoted 9 lines

The commit b0ce3508(bonding: allow TSO being set on bonding master) has make the TSO being set for bond dev, but in some situation, if the slave did not have the NETIF_F_SG features, the bond master will miss the TSO features in netdev_fix_features because the TSO is depended on SG. So I have to add SG and TSO features on bond master together.
The function netdev_add_tso_features() was only be used for bonding, so no need to export it in netdevice.h, remove it and add it to bonding.

As far as I can tell from the discussion, there is some issue wrt. TSO about what happens if SG is not supported by some of the slaves.

Quoted 1 line

From my perspective it appears that some changes to these patches are

necessary to handle that correctly.

So I am going to mark them as "Changes Requested" in patchwork.

If this is not the case, please resubmit these changes with appropriate explanations added to the commit message(s).

Thanks.

Ding Tianhong

author

2014-01-14 02:06 UTC #2 lore raw

Quoted 17 lines · David Miller

From: Ding Tianhong <[email protected]>
Date: Wed, 8 Jan 2014 15:28:21 +0800
Quoted 9 lines
The commit b0ce3508(bonding: allow TSO being set on bonding master) has make the TSO being set for bond dev, but in some situation, if the slave did not have the NETIF_F_SG features, the bond master will miss the TSO features in netdev_fix_features because the TSO is depended on SG. So I have to add SG and TSO features on bond master together.
The function netdev_add_tso_features() was only be used for bonding, so no need to export it in netdevice.h, remove it and add it to bonding.
As far as I can tell from the discussion, there is some issue wrt. TSO about what happens if SG is not supported by some of the slaves.
Quoted 1 line
From my perspective it appears that some changes to these patches are

Quoted 10 lines

necessary to handle that correctly.
So I am going to mark them as "Changes Requested" in patchwork.
If this is not the case, please resubmit these changes with appropriate explanations added to the commit message(s).
Thanks.

Ok, thanks.

Regards
Ding