<html><head><meta http-equiv="Content-Type" content="text/html charset=windows-1252"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;">I’ve submitted an issue on this. Thanks for noting it.<div><br></div><div>Bill</div><div><br><div>
<div style="color: rgb(0, 0, 0); letter-spacing: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px; word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;">William Gropp<br>Acting Director and Chief Scientist, NCSA</div><div style="color: rgb(0, 0, 0); letter-spacing: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px; word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;">Director, Parallel Computing Institute<br>Thomas M. Siebel Chair in Computer Science<br>University of Illinois Urbana-Champaign<br><br><br><br><br></div>
</div>
<br><div><div>On Sep 13, 2016, at 8:17 AM, William Gropp <<a href="mailto:wgropp@illinois.edu">wgropp@illinois.edu</a>> wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite">
<meta http-equiv="Content-Type" content="text/html; charset=us-ascii"><div style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;">I agree with Jeff - this is a simple errata item to be fixed.<div><br></div><div>Bill</div><div><br><div>
<div style="letter-spacing: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px; word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;">William Gropp<br>Acting Director and Chief Scientist, NCSA</div><div style="letter-spacing: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px; word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;">Director, Parallel Computing Institute<br>Thomas M. Siebel Chair in Computer Science<br>University of Illinois Urbana-Champaign<br><br><br><br><br></div>
</div>
<br><div><div>On Sep 13, 2016, at 8:09 AM, Jeff Hammond <<a href="mailto:jeff.science@gmail.com">jeff.science@gmail.com</a>> wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite">This is a bug in the text that should be fixed by an erratum. <br><br>Jeff <br><br>Sent from my iPhone<br><br><blockquote type="cite">On Sep 13, 2016, at 5:36 AM, Rolf Rabenseifner <<a href="mailto:rabenseifner@hlrs.de">rabenseifner@hlrs.de</a>> wrote:<br><br>Hello,<br><br>my colleague Joseph detected that an MPI implementation is allowed <br>to not allow passive target communication/synchronization (i.e.,<br>all forms of MPI_Lock and Unlock) on MPI shared memory windows.<br><br>Was it intented or is it only a historical review-bug that <br>MPI_WIN_ALLOCATE_SHARED was never added to the list of <br>MPI_ALLOC_MEM (since MPI-2.0), and MPI_WIN_ALLOCATE & <br>MPI_WIN_ATTACH (since MPI-3.0) in the following text.<br><br><br>MPI-3.1 Sect. 11.5.3 on page 448 lines 1-12 say<br><br>"Implementors may restrict the use of RMA communication <br>that is synchronized by lock calls to windows in <br>memory allocated by MPI_ALLOC_MEM (Section 8.2),<br>MPI_WIN_ALLOCATE (Section 11.2.2), or attached with <br>MPI_WIN_ATTACH (Section 11.2.4). <br>Locks can be used portably only in such memory.<br><br>Rationale. The implementation of passive target <br>communication when Memory is not shared may require <br>an asynchronous software agent. Such an agent can be<br>implemented more easily, and can achieve better <br>performance, if restricted to specially allocated memory. <br>It can be avoided altogether if shared memory is used. <br>It seems natural to impose restrictions that allows <br>one to use shared memory for third party communication <br>in shared memory machines.<br>(End of rationale.)"<br><br><br>The rationale not really fits to the current state<br>that MPI_WIN_ALLOCATE_SHARED is missing in the list. <br>It would fit if MPI_WIN_ALLOCATE_SHARED would be added<br>to the list.<br><br>Best regards<br>Rolf<br><br>-- <br>Dr. Rolf Rabenseifner . . . . . . . . . .. email <a href="mailto:rabenseifner@hlrs.de">rabenseifner@hlrs.de</a><br>High Performance Computing Center (HLRS) . phone ++49(0)711/685-65530<br>University of Stuttgart . . . . . . . . .. fax ++49(0)711 / 685-65832<br>Head of Dpmt Parallel Computing . . . <a href="https://urldefense.proofpoint.com/v2/url?u=http-3A__www.hlrs.de_people_rabenseifner&d=CwMFAg&c=8hUWFZcy2Z-Za5rBPlktOQ&r=195OD1c98cxYlIPBUDzl00KueizThEKs34a9DvL37eY&m=wfzy8pX-gARAFABeatfHPyO20a14S4gT3TTvPbzdGoU&s=i9Tg5cbQSt3jN7pvWec2C-mUvw8kfWEoY4mN4eTAbl4&e=">www.hlrs.de/people/rabenseifner</a><br>Nobelstr. 19, D-70550 Stuttgart, Germany . . . . (Office: Room 1.307)<br>_______________________________________________<br>mpiwg-rma mailing list<br><a href="mailto:mpiwg-rma@lists.mpi-forum.org">mpiwg-rma@lists.mpi-forum.org</a><br><a href="http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-rma">http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-rma</a><br></blockquote>_______________________________________________<br>mpiwg-rma mailing list<br><a href="mailto:mpiwg-rma@lists.mpi-forum.org">mpiwg-rma@lists.mpi-forum.org</a><br><a href="http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-rma">http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-rma</a><br></blockquote></div><br></div></div>_______________________________________________<br>mpiwg-rma mailing list<br><a href="mailto:mpiwg-rma@lists.mpi-forum.org">mpiwg-rma@lists.mpi-forum.org</a><br>http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-rma</blockquote></div><br></div></body></html>