diff options
author | David S. Miller <davem@davemloft.net> | 2015-02-09 13:20:53 -0800 |
---|---|---|
committer | David S. Miller <davem@davemloft.net> | 2015-02-09 13:20:53 -0800 |
commit | 9dce285b70c157754d753203112cfef22770b1f9 (patch) | |
tree | 4859799a8311ecd637e2a582600af1057a78e08b /include/net | |
parent | c8ac18f2006b2926ce375c01646b2f487d1c33b2 (diff) | |
parent | 941787b82982b3f33ac398c8c00035ddd0f8c514 (diff) | |
download | op-kernel-dev-9dce285b70c157754d753203112cfef22770b1f9.zip op-kernel-dev-9dce285b70c157754d753203112cfef22770b1f9.tar.gz |
Merge branch 'tipc-next'
Richard Alpe says:
====================
tipc: new compat layer for the legacy NL API
This is a compatibility / transcoding layer for the old netlink API.
It relies on the new netlink API to collect data or perform actions
(dumpit / doit).
The main benefit of this compat layer is that it removes a lot of
complex code from the tipc core as only the new API needs to be able
harness data or perform actions. I.e. the compat layer isn't concerned
with locking or how the internal data-structures look. As long as the
new API stays relatively intact the compat layer should be fine.
The main challenge in this compat layer is the randomness of the legacy
API. Some commands send binary data and some send ASCII data, some are
very picky in optimizing there buffer sizes and some just don't care.
Most legacy commands put there data in a single TLV (data container) but some
segment the data into multiple TLV's. This list of randomness goes on and on..
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'include/net')
0 files changed, 0 insertions, 0 deletions