<html><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">
Notes inline.<div><br><div><div>On Jul 18, 2008, at 1:02 PM, Jeff Squyres wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite"><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">On Jul 17, 2008, at 12:41 PM, William Gropp wrote:</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; min-height: 14px; "><br></div> <blockquote type="cite"><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">There's another option.<span class="Apple-converted-space"> </span>Drop the requirement that all of the</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">predefined datatypes are predefined in all languages, but specify</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">that they may be used (with the appropriate handle conversion) as</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">input to any routine where it makes sense.</div> </blockquote><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; min-height: 14px; "><br></div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">Just curious: if you drop the requirement for direct access to all the</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">predefined datatypes in all languages, why do you need to specify</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">anything else?<span class="Apple-converted-space"> </span>MPI-2.1:16.3.4 (transfer of handles -- the *f2c/*c2f</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">stuff) seems to be sufficient.<span class="Apple-converted-space"> </span>Is there a need to say something more?</div></blockquote><div><br></div>The implicit assumption you are making is that the existence of the handle transfer functions means that the handles can be used everywhere. You can certainly argue that nothing else is needed, but MPI 2.0 felt that it was important to make it clear that this applied to the datatype routines, even if those datatypes do not exist in the calling language. In other words, you can't have different versions of MPI_Type_xxx that only know about the predefined datatypes in the calling language. Without this clarification, we could be having that argument instead :)</div><div><br><blockquote type="cite"><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; min-height: 14px; "><br></div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">Further, if we drop the requirement, do we need a "high quality</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">implementation" note about making all predefined handles available in</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">all languages for MPI programmer convenience?<span class="Apple-converted-space"> </span>I.e., what reasonable</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">expectation should an MPI programmer have (particularly for already-</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">existing codes) about accessing language A's predefined handles in</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">language B?<span class="Apple-converted-space"> </span>Should they *always* call C to do the translation?</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; min-height: 14px; "><br></div> <blockquote type="cite"><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">For datatypes, this includes the datatype construction routines but</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">not the communication routines (don't use MPI_INT when sending</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">INTEGERs with Fortran).<span class="Apple-converted-space"> </span>This is what the Opaque Objects text expects.</div> </blockquote><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; min-height: 14px; "><br></div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">To me, dropping the requirement sounds like we're just making the</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">user's life hard -- a Fortran library may have to introduce C code</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">into their project, and then figure out all the cross-language</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">compiling and linking stuff (which is not for the meek).</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; min-height: 14px; "><br></div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">It might be a whole lot simpler to specifically define which</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">predefined handles need to be available in each language.<span class="Apple-converted-space"> </span>The</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">spectrum ranges from "all languages' predefined handles must be</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">available in every language" to "only that language A's predefined</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">handles must be available in language A".<span class="Apple-converted-space"> </span>Suggestions?</div></blockquote><div><br></div>I'm in favor of explicitly specifying the values and names (because of the need to use alternate names when their might be a conflict, as in the MPI::F_COMPLEX case), and that these have the exact same value as if the handle transfer function was used.</div><div><br></div><div>Bill</div><div><br><blockquote type="cite"><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; min-height: 14px; "><br></div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">--</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">Jeff Squyres</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">Cisco Systems</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; min-height: 14px; "><br></div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">_______________________________________________</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; ">mpi-22 mailing list</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; "><a href="mailto:mpi-22@lists.mpi-forum.org">mpi-22@lists.mpi-forum.org</a></div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; "><a href="http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-22">http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-22</a></div> </blockquote></div><br><div> <span class="Apple-style-span" style="border-collapse: separate; border-spacing: 0px 0px; color: rgb(0, 0, 0); font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; text-align: auto; -khtml-text-decorations-in-effect: none; text-indent: 0px; -apple-text-size-adjust: auto; text-transform: none; orphans: 2; white-space: normal; widows: 2; word-spacing: 0px; "><div>William Gropp</div><div>Paul and Cynthia Saylor Professor of Computer Science</div><div>University of Illinois Urbana-Champaign</div><div><br class="khtml-block-placeholder"></div><br class="Apple-interchange-newline"></span> </div><br></div></body></html>