<div dir="ltr"><div class="gmail_extra"><br><div class="gmail_quote">On Mon, Mar 17, 2014 at 11:52 AM, Craig Rasmussen <span dir="ltr"><<a href="mailto:rasmus@cas.uoregon.edu" target="_blank">rasmus@cas.uoregon.edu</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div style="word-wrap:break-word">Jeff and I had a similar conversation last week. It is my viewpoint that the MPI-3 interfaces are formal interfaces that can be adapted by an implementation in a way that formally meets the specification.</div>
</blockquote><div><br></div><div>+1 for this view.</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div style="word-wrap:break-word"><div></div><div>
For Open MPI I believe that Jeff will add configuration magic to check is INTEGER(C_INT) and default INTEGER have the same KIND value. This should be true for 100% of the compilers. If not, then that compiler will not be able to build Open MPI.</div>
</div></blockquote><div><br></div><div>Supporting different INTEGER and c_int 1) is useless; 2) adds code complexity; 3) adds extra runtime overhead (in MPICH implementation) even when INTEGER and c_int are the same. I prefer stopping the support and doing configuration time check.</div>
<div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div style="word-wrap:break-word"><div><br></div><div>The use of INTEGER(C_INT) to replace the default INTEGER in the MPI types is needed to stopped compiler warning from GNU. We did this work (adding BIND(C) to gfortran) at Los Alamos and I think we perhaps got carried away with the warning messages.<span class="HOEnZb"><font color="#888888"><br>
<div><br></div></font></span></div></div></blockquote></div></div></div>