[openib-general] CMA: compliancy issue?

[Or Gerlitz]

Michael S. Tsirkin wrote:
> Quoting r. Or Gerlitz <ogerlitz at voltaire.com>:
>> You are suggesting a design change in the CMA which would effect also 
>> the current CMA ULP consumers: iSER, RDS, NFSoRDMA and Lustre.

> I don't see how it will affect these ULPs: don't see how ULPs that don't check
> private data care whether RTU is sent before, or after, the handler.

I ment to say that today they just get an ESTABLISHED event, and with 
the change you suggest they will get CONNECT_RESPONSE event and will 
need to call either rdma_accept or rdma_reject, that's a little change 
in the ULP state machine.

> In any case, CMA is still in early stages of development so its natural to
> expect API changes.

The CMA is pushed upstream for 2.6.18 so i am not sure what do you mean 
by "early" here, its been under work for many months and its you that 
started to look/use it only lately. The CMA has very solid design and 
implementation, its not that down the road, design/API changes are not 
possible but saying it is in "early stages of development" is insulting, 
no less.

Or.