<html><body>
<p>Sorry - I do not see how that solves the problem. Compiled application code allocates some number of bytes based on a structure declaration in C or a constant (MPI_STATUS_SIZE) in Fortran. Each byte in the structure is required because the MPI library has a use for it. There will not be spare bytes that can be used to cache the extra baggage.<br>
<br>
Dick Treumann - MPI Team <br>
IBM Systems & Technology Group<br>
Dept X2ZA / MS P963 -- 2455 South Road -- Poughkeepsie, NY 12601<br>
Tele (845) 433-7846 Fax (845) 433-8363<br>
<br>
<br>
<img width="16" height="16" src="cid:1__=0ABBFC63DFCDB38B8f9e8a93df938@us.ibm.com" border="0" alt="Inactive hide details for Greg Bronevetsky ---03/24/2010 12:36:17 PM--->I agree with Dick. We're getting very strong ISV feedba"><font color="#424282">Greg Bronevetsky ---03/24/2010 12:36:17 PM--->I agree with Dick. We're getting very strong ISV feedback that an >MPI-3 compatibility break will</font><br>
<br>
<table width="100%" border="0" cellspacing="0" cellpadding="0">
<tr valign="top"><td width="1%"><img width="96" height="1" src="cid:2__=0ABBFC63DFCDB38B8f9e8a93df938@us.ibm.com" border="0" alt=""><br>
<font size="2" color="#5F5F5F">From:</font></td><td width="100%"><img width="1" height="1" src="cid:2__=0ABBFC63DFCDB38B8f9e8a93df938@us.ibm.com" border="0" alt=""><br>
<font size="2">Greg Bronevetsky <bronevetsky1@llnl.gov></font></td></tr>
<tr valign="top"><td width="1%"><img width="96" height="1" src="cid:2__=0ABBFC63DFCDB38B8f9e8a93df938@us.ibm.com" border="0" alt=""><br>
<font size="2" color="#5F5F5F">To:</font></td><td width="100%"><img width="1" height="1" src="cid:2__=0ABBFC63DFCDB38B8f9e8a93df938@us.ibm.com" border="0" alt=""><br>
<font size="2">"MPI 3.0 Fault Tolerance and Dynamic Process Control working Group" <mpi3-ft@lists.mpi-forum.org>, "MPI 3.0 Fault Tolerance and Dynamic Process Control working Group" <mpi3-ft@lists.mpi-forum.org></font></td></tr>
<tr valign="top"><td width="1%"><img width="96" height="1" src="cid:2__=0ABBFC63DFCDB38B8f9e8a93df938@us.ibm.com" border="0" alt=""><br>
<font size="2" color="#5F5F5F">Date:</font></td><td width="100%"><img width="1" height="1" src="cid:2__=0ABBFC63DFCDB38B8f9e8a93df938@us.ibm.com" border="0" alt=""><br>
<font size="2">03/24/2010 12:36 PM</font></td></tr>
<tr valign="top"><td width="1%"><img width="96" height="1" src="cid:2__=0ABBFC63DFCDB38B8f9e8a93df938@us.ibm.com" border="0" alt=""><br>
<font size="2" color="#5F5F5F">Subject:</font></td><td width="100%"><img width="1" height="1" src="cid:2__=0ABBFC63DFCDB38B8f9e8a93df938@us.ibm.com" border="0" alt=""><br>
<font size="2">Re: [Mpi3-ft] FT Working Group Teleconf (today)</font></td></tr>
<tr valign="top"><td width="1%"><img width="96" height="1" src="cid:2__=0ABBFC63DFCDB38B8f9e8a93df938@us.ibm.com" border="0" alt=""><br>
<font size="2" color="#5F5F5F">Sent by:</font></td><td width="100%"><img width="1" height="1" src="cid:2__=0ABBFC63DFCDB38B8f9e8a93df938@us.ibm.com" border="0" alt=""><br>
<font size="2">mpi3-ft-bounces@lists.mpi-forum.org</font></td></tr>
</table>
<hr width="100%" size="2" align="left" noshade style="color:#8091A5; "><br>
<br>
<br>
<tt><br>
>I agree with Dick. We're getting very strong ISV feedback that an <br>
>MPI-3 compatibility break will constitute a serious adoption barrier for them.<br>
<br>
The other option would be to have an extra function like <br>
MPI_Get_count that extracts internal information from MPI_Status <br>
objects. Can that be implemented without breaking compatibility?<br>
<br>
Greg Bronevetsky<br>
Lawrence Livermore National Lab<br>
(925) 424-5756<br>
bronevetsky@llnl.gov<br>
</tt><tt><a href="http://greg.bronevetsky.com">http://greg.bronevetsky.com</a></tt><tt> <br>
<br>
_______________________________________________<br>
mpi3-ft mailing list<br>
mpi3-ft@lists.mpi-forum.org<br>
</tt><tt><a href="http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-ft">http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-ft</a></tt><tt><br>
</tt><br>
<br>
</body></html>