Richard Treumann wrote:

On a single OS under AIX, IBM MPI does map the portion of send side memory that holds the send buffer into the receivers address space. We do this for both contiguous and non-contiguous buffers. The mapping lasts just long enough for the receive side CPU to do a memory copy from send buffer to receive buffer. (see patent *7,392,256*)

This optimization does not have any effect on the addressability of the send buffer by the sending task. In our case, at least, this optimization does not argue against the proposal.

This reference to the patent would be a clear indication for other implementers (such as me) to implement it using Alexander's approach (in order to avoid any problem with IBM's patent lawyers). And the change of the actual standard might limit implementers to implement Alexander's approach and might decrease the performance of their MPI implementation. (OK; 2 might's, not really a strong argument since it is not already implemented in this way; but the mapping approach seems to be quite reasonable). Or does IBM have also a patent for Alexander's approach ? Or does Intel have already a patent on this approach ? ;-) Hubert Hubert Ritzdorf NEC Europe *