<html dir="ltr">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<style id="owaParaStyle" type="text/css">P {margin-top:0;margin-bottom:0;}</style>
</head>
<body ocsi="0" fpstyle="1">
<div style="direction: ltr;font-family: Tahoma;color: #000000;font-size: 10pt;"><br>
<div style="font-family: Times New Roman; color: #000000; font-size: 16px">
<div></div>
<br>
<div>
<div dir="ltr">
<div><span style="font-size:13px">Jeff Squyres wrote:</span></div>
<div><span style="font-size:13px"><br>
</span></div>
<span style="font-size:13px">>Hmm. Ok, that's the opposite advice we got for MPI-3.0 (which was now admittedly several years ago).</span><br style="font-size:13px">
><br style="font-size:13px">
<span style="font-size:13px">>Soo... which way should it go these days?</span><br style="font-size:13px">
><br style="font-size:13px">
<span style="font-size:13px">>- use native Fortran datatypes</span><br style="font-size:13px">
<span style="font-size:13px">>- use C datatypes (via BIND(C))<br>
<br>
<br style="font-size:13px">
</span>I think that MPI Fortran Interface should support native Fortran datatypes in pair types !<br>
<br>
(*) We (NEC) don't know any relevant Fortran application which doesn't use native Fortran datatypes.<br>
If you switch for C Datatypes in MINLOC/MAXLOC, all application programmers must solve<br>
the portability problems (possibly loosing accuracy in reduce operations).<br>
<br>
(*) Nearly all Fortran compilers support compile time options such as "-r8" or "-size real64"<br>
which would cause problems with the C datatypes.<br>
And this options are used by application programmers !<br>
<br>
(*) C datatypes would be inconsistent with functions such as MPI_Type_create_f90_real.<br>
<br>
(*) Supporting Fortran subarrays without supporting native Fortran datatypes doesn't make sense for me.<br>
<br>
Hubert<br>
<br style="font-size:13px">
<br>
<font style="background-color:#ffffff"></font></div>
</div>
</div>
</div>
</body>
</html>