<html><head></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; "><div style="font-size: 16px; ">Hello,</div><div style="font-size: 16px; "><br></div><div style="font-size: 16px; "><div>Le 12 janv. 2012 à 18:07, Darius Buntinas a écrit :</div><br class="Apple-interchange-newline"><blockquote type="cite">Great, I'm glad it worked. BTW, the kernel.shmall et.al. parameters are used for System V shared memory, so they'll have no effect when using mmap-ed shared memory, which is how MPICH is configured by default.<br><br>-d<br><font class="Apple-style-span" color="#000000"><font class="Apple-style-span" color="#144FAE"><br></font></font></blockquote><div style="font-size: 15px; "><br></div><div style="font-size: 15px; ">Now that your patch has solved the shared memory failure, I encountered a new failure when specifying -iface option</div><div style="font-size: 15px; "><br></div><div style="font-size: 15px; ">With a minimal piece of code (*) , </div><div style="font-size: 15px; "> with -iface, I always get assertion error like this one :</div><div style="font-size: 17px; "><font class="Apple-style-span" size="4" style="font-size: 21px; "><span class="Apple-style-span" style="font-size: 15px; "><div><i>>mpiexec -iface eth0 -n 150 bin/basic_test</i></div><div style="font-size: 14px; "><i>[mpiexec@ccwpge0061] control_cb (./pm/pmiserv/pmiserv_cb.c:215): assert (!closed) failed</i></div><div style="font-size: 14px; "><i>[mpiexec@ccwpge0061] HYDT_dmxu_poll_wait_for_event (./tools/demux/demux_poll.c:77): callback returned error status</i></div><div style="font-size: 14px; "><i>[mpiexec@ccwpge0061] HYD_pmci_wait_for_completion (./pm/pmiserv/pmiserv_pmci.c:181): error waiting for event</i></div><div style="font-size: 14px; "><i>[mpiexec@ccwpge0061] main (./ui/mpich/mpiexec.c:405): process manager error waiting for completion</i></div><div><br></div><div><span class="Apple-style-span" style="font-size: 16px; "><div style="font-size: 15px; "> without -iface, it works <b>always</b></div></span><div style="font-size: 17px; "><font class="Apple-style-span" size="4" style="font-size: 21px; "><span class="Apple-style-span" style="font-size: 15px; ">> mpiexec -n 150 bin/basic_test : </span></font><span class="Apple-style-span" style="font-size: 15px; ">it's OK 100 % of time (</span><font class="Apple-style-span" size="4" style="font-size: 21px; "><span class="Apple-style-span" style="font-size: 15px; "> </span></font><span class="Apple-style-span" style="font-size: 15px; ">without -iface)</span></div></div><div><br></div><div><div>I'm confused with that because I can't provide the 10Gb/s (<i>-iface eth2</i>) to our customers </div></div><div><br></div><div>Best regards</div><div><br></div><div><br></div><div>(*) minimal code :</div></span></font></div><div style="font-size: 16px; "><div style="font-size: 13px; "><i> if (MPI_Init(NULL, NULL) != MPI_SUCCESS ) {</i></div><div style="font-size: 13px; "><i> printf("Error calling MPI_Init !!, => exiting \n") ; fflush(stdout);</i></div><div style="font-size: 13px; "><i> return(1);</i></div><div style="font-size: 13px; "><i> } else {</i></div><div style="font-size: 13px; "><i> MPI_Finalize(); </i></div><div style="font-size: 13px; "><i> printf("It's OK \n") ; fflush(stdout);</i></div><div style="font-size: 13px; "><i> return(0);</i></div><div style="font-size: 13px; "><i> }</i></div><div><i><br></i></div><div><i><br></i></div></div><blockquote type="cite">On Jan 12, 2012, at 5:31 AM, Bernard Chambon wrote:<br><br><blockquote type="cite">Hello,<br></blockquote><blockquote type="cite"><br></blockquote><blockquote type="cite">Good news, It works !<br></blockquote><blockquote type="cite">Whith your patch I can run 255 tasks and perhaps more, without any special<br></blockquote><blockquote type="cite">configuration on the machine (*)<br></blockquote><blockquote type="cite"><br></blockquote><blockquote type="cite"><br></blockquote><br class="Apple-interchange-newline"></blockquote></div><br style="font-size: 16px; "><div style="font-size: 16px; ">
<div><div><div><div><div>---------------<br>Bernard CHAMBON<br>IN2P3 / CNRS<br>04 72 69 42 18<br></div></div></div></div></div>
</div>
<br style="font-size: 16px; "></body></html>