<meta http-equiv="Content-Type" content="text/html; charset=utf-8"><div dir="ltr">I guess this is a bug but MPICH 3.1.x isn't the basis for the supported MPI on BGQ, so I doubt you will get much traction by reporting it.  IBM made an effort to support MPI-3 with PAMID but it was as an open-source, best effort project, and I recall there were some issues with it, including deadlock in certain asynchronous operations.<div><br></div><div>You should try the supported MPI-2.2 implementation on BGQ or try the unsupported OFI-based implementation of MPI-3.</div><div><br class="gmail-Apple-interchange-newline">Disclaimer: The comments above are spoken in my capacity as the person who used to support MPI at ALCF, not my current job role.<br></div><div><br></div><div>Best,</div><div><br></div><div>Jeff<br><div class="gmail_extra"><br><div class="gmail_quote">On Thu, Mar 30, 2017 at 3:21 AM, Sebastian Rinke <span dir="ltr"><<a href="mailto:rinke@cs.tu-darmstadt.de" target="_blank">rinke@cs.tu-darmstadt.de</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div style="word-wrap:break-word"><div>Same window, i.e.:</div><div><br></div><div><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Process 0:<br class="gmail-m_5168060242488893594gmail_msg"><br class="gmail-m_5168060242488893594gmail_msg"></blockquote></div><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Win_lock(MPI_LOCK_SHARED, target=A, window=win)<br class="gmail-m_5168060242488893594gmail_msg"></blockquote></div><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Win_lock(MPI_LOCK_SHARED, target=B, window=win)</blockquote></div></div><span class="gmail-HOEnZb"><font color="#888888"><div><br></div><div>Sebastian</div></font></span><div><div class="gmail-h5"><div><br></div><br><div><div>On 30 Mar 2017, at 06:24, Jeff Hammond <<a href="mailto:jeff.science@gmail.com" target="_blank">jeff.science@gmail.com</a>> wrote:</div><br class="gmail-m_5168060242488893594Apple-interchange-newline"><blockquote type="cite"><div>Same window or different windows?</div><div><br></div><div>Jeff</div><div><br><div class="gmail_quote"><div>On Wed, Mar 29, 2017 at 5:59 PM Sebastian Rinke <<a href="mailto:rinke@cs.tu-darmstadt.de" target="_blank">rinke@cs.tu-darmstadt.de</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Dear all,<br class="gmail-m_5168060242488893594gmail_msg">
<br class="gmail-m_5168060242488893594gmail_msg">
I have some issue with MPI_Win_lock in MPICH-3.1.4 on Blue Gene/Q.<br class="gmail-m_5168060242488893594gmail_msg">
<br class="gmail-m_5168060242488893594gmail_msg">
Here is my example:<br class="gmail-m_5168060242488893594gmail_msg">
<br class="gmail-m_5168060242488893594gmail_msg">
Process 0:<br class="gmail-m_5168060242488893594gmail_msg">
<br class="gmail-m_5168060242488893594gmail_msg">
Win_lock(MPI_LOCK_SHARED, target=A)<br class="gmail-m_5168060242488893594gmail_msg">
Win_lock(MPI_LOCK_SHARED, target=B)<br class="gmail-m_5168060242488893594gmail_msg">
<br class="gmail-m_5168060242488893594gmail_msg">
No matter what I use for A and B (given A != B), a process cannot acquire more than one lock<br class="gmail-m_5168060242488893594gmail_msg">
at a time.<br class="gmail-m_5168060242488893594gmail_msg">
<br class="gmail-m_5168060242488893594gmail_msg">
To my understanding, it should be possible to acquire more than one lock.<br class="gmail-m_5168060242488893594gmail_msg">
<br class="gmail-m_5168060242488893594gmail_msg">
Can you confirm this issue?<br class="gmail-m_5168060242488893594gmail_msg">
<br class="gmail-m_5168060242488893594gmail_msg">
Thanks,<br class="gmail-m_5168060242488893594gmail_msg">
Sebastian<br class="gmail-m_5168060242488893594gmail_msg">
______________________________<wbr>_________________<br class="gmail-m_5168060242488893594gmail_msg">
discuss mailing list     <a href="mailto:discuss@mpich.org" class="gmail-m_5168060242488893594gmail_msg" target="_blank">discuss@mpich.org</a><br class="gmail-m_5168060242488893594gmail_msg">
To manage subscription options or unsubscribe:<br class="gmail-m_5168060242488893594gmail_msg">
<a href="https://lists.mpich.org/mailman/listinfo/discuss" rel="noreferrer" class="gmail-m_5168060242488893594gmail_msg" target="_blank">https://lists.mpich.org/<wbr>mailman/listinfo/discuss</a><br class="gmail-m_5168060242488893594gmail_msg">
</blockquote></div></div><div dir="ltr">-- <br></div><div>Jeff Hammond<br><a href="mailto:jeff.science@gmail.com" target="_blank">jeff.science@gmail.com</a><br><a href="http://jeffhammond.github.io/" target="_blank">http://jeffhammond.github.io/</a></div>
______________________________<wbr>_________________<br>discuss mailing list     <a href="mailto:discuss@mpich.org" target="_blank">discuss@mpich.org</a><br>To manage subscription options or unsubscribe:<br><a href="https://lists.mpich.org/mailman/listinfo/discuss" target="_blank">https://lists.mpich.org/<wbr>mailman/listinfo/discuss</a></blockquote></div><br></div></div></div><br>______________________________<wbr>_________________<br>
discuss mailing list     <a href="mailto:discuss@mpich.org">discuss@mpich.org</a><br>
To manage subscription options or unsubscribe:<br>
<a href="https://lists.mpich.org/mailman/listinfo/discuss" rel="noreferrer" target="_blank">https://lists.mpich.org/<wbr>mailman/listinfo/discuss</a><br></blockquote></div><br><br clear="all"><div><br></div>-- <br><div class="gmail_signature">Jeff Hammond<br><a href="mailto:jeff.science@gmail.com" target="_blank">jeff.science@gmail.com</a><br><a href="http://jeffhammond.github.io/" target="_blank">http://jeffhammond.github.io/</a></div>
</div></div></div>