Re: [RFC PATCH V2] tcp: introduce raw access to experimental options

On 01/07/2013 05:44 PM, Eric Dumazet wrote:

Thats a big addition, and for example doesn't help if the SYNACK should
include an option that is depending on the content of SYN message.

For TCP fastopen for example, the cookie we send to the client is not a
constant cookie.

Also TCP coalescing of TCP collapse will merge several skbs, so storing
"the last received options" in the socket is kind of not well defined
semantic.

It looks like you need to add hooks and kernel modules to fully use
experimental options, like congestion control modules.

Thanks for the fast feedback, Eric. Let me try to make sure that I
understand your suggestions correctly:
You suggest to have a set of hooks that allow to move all code that
exploits experimental options into dedicated modules (e.g. one module
for TCP fastopen and one for cookies). Extensibility then comes on
a per module granularity. Is my understanding correct?

Thx,
Einar.