[openib-general] CM header file

[Roland Dreier]

    Sean> I think it makes sense in the case of a timeout for the CM
    Sean> to optimize for the retry case.  (Such as keeping the last
    Sean> sent MAD around for retransmission.)  The argument used
    Sean> against putting retries into the MAD layer was to allow the
    Sean> consumer to set the retry policy, such as increasing the
    Sean> timeout value.

    Sean> A major difference between the MAD API and the CM API is
    Sean> that the CM abstracts the MAD from the user, making the
    Sean> retry issue a more difficult to do efficiently.  So, while I
    Sean> think it would be nice to let the user initiate retries, I
    Sean> want to make sure that it doesn't have a substantial impact
    Sean> on performance.

The CM retry policy is specified much more tightly than for general
MADs.  The total number of retries is limited by the "Max CM Retries"
field and the timeout waiting for each response is also part of the CM
protocol.

 - Roland