[ofa-general] Bonding and hw_csum

[Or Gerlitz]

Koen Segers wrote:
> I just saw some patches on the mailing list concerning csum offloading.
> Are these applied in RC3? Or are they going to be introduced in the
> daily build of tomorrow?
> Is it correct to state that these patches replace the hw_csum parameter
> by offloading the csum computation to the mthca? This would mean that
> the results should be similar also.

no and no, best if you take a look on the presentation @ 
http://openfabrics.org/archives/nov2007sc/IPoIB-UD%20SO.pdf

Basically the "checksum offloading" patches are for the datagram mode 
and is the standard offload as in the Ethernet world, where the 
"hw_csum" patch was for the connected mode.

> Does the new offload patch depend on the type of hca being used?
> According to lspci, we have the "InfiniBand: Mellanox Technologies
> MT25208 InfiniHost III Ex (rev a0)" card. Do these patches work on a
> sles 10 sp1 installed on x3755 and x3655 machines of IBM that have this
> card inserted?

checksum offloading is supported by the connectx and some of the other 
Mellanox devices, I am quite sure that 25208 is one of them, but you 
have to clarify this with Mellanox

> Is bonding going to work with this type of offloading?

sure! and if not, we will fix it.

Or.