On Thu, 30 Oct 2003, Jeff Squyres wrote:
> ...this allows LAM to block in the underlying write() instead of
> traversing the entire communications stack (MPI_Send all the way down to
> the underlying write()) multiple times in a non-blocking fashion, each
> time writing out a few more bytes (as OS buffering allows).
Oops -- that should read:
"...(MPI_Test all the way down to the underlyng write())..."
Sorry for any confusion...
--
{+} Jeff Squyres
{+} jsquyres_at_[hidden]
{+} http://www.lam-mpi.org/
|
