<html><head></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">Ok. Still having problems. <div>I defaulted everything to private and explicitly declared my shared's. Now what happens is that the outer "k" loop never gets incremented. <div>Even if I run with only one thread, "k" always equals 1.<br><div><br></div><div>So the snippet of code follows. Where m_ex and m_ib are parameters in Fortran and are hard-coded as numbers by the compiler.</div><div>If I compile with -fopenmp it works fine, but at the -fopenmp and "setenv OMP_NUM_THREADS 1" and it won't increment.</div><div><br></div><div>Any new ideas? Thanks in advance.</div><div>-Scott</div><div><br></div><div><br></div><div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; color: rgb(0, 132, 38); ">!$OMP PARALLEL DO DEFAULT(PRIVATE) SHARED(mask,ibonly,ax,ay,az,</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; color: rgb(0, 132, 38); ">!$OMP& agxx,agxy,agxz,agyy,agyz,agzz,</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; color: rgb(0, 132, 38); ">!$OMP& aKxx,aKxy,aKxz,aKyy,aKyz,aKzz,nx,ny,nz) </div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; color: rgb(0, 132, 38); ">!$OMP& SCHEDULE(STATIC,chunk) </div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> <span style="color: #c22a9c">do</span> k = <span style="color: #2f2fcf">1</span>, nz </div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; color: rgb(0, 132, 38); "><span class="Apple-style-span" style="color: rgb(0, 0, 0); "> <span style="color: #c22a9c">write</span>(msg,*)<span style="color: #d92823">' setbkgrnd: '</span> //</span></div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> & <span style="color: #d92823">'nz = '</span>,nz,<span style="color: #d92823">', % done = '</span>,int(k*<span style="color: #2f2fcf">1.0</span>d<span style="color: #2f2fcf">2</span>/nz),<span style="color: #d92823">' '</span></div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> <span style="color: #c22a9c">call</span> writemessage(msg)</div><p style="margin: 0.0px 0.0px 0.0px 0.0px; font: 11.0px Menlo; min-height: 13.0px"> <br class="webkit-block-placeholder"></p><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> <span style="color: #c22a9c">do</span> j = <span style="color: #2f2fcf">1</span>, ny</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> <span style="color: #c22a9c">do</span> i = <span style="color: #2f2fcf">1</span>, nx</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; min-height: 13px; "> </div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> <span style="color: #c22a9c">if</span> (mask(i,j,k) <span style="color: #c22a9c">.ne.</span> m_ex <span style="color: #c22a9c">.and.</span></div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> & (ibonly <span style="color: #c22a9c">.eq.</span> <span style="color: #2f2fcf">0</span> <span style="color: #c22a9c">.or.</span> </div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> & mask(i,j,k) <span style="color: #c22a9c">.eq.</span> m_ib)) <span style="color: #c22a9c">then</span></div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; min-height: 13px; "><br></div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; min-height: 13px; "><br></div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> x = ax(i)</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> y = ay(j)</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> z = az(k)</div></div><div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> <span style="color: #c22a9c">include</span> <span style="color: #d92823">'gd.inc'</span></div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> <span style="color: #c22a9c">include</span> <span style="color: #d92823">'kd.inc'</span></div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; min-height: 13px; "><br></div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; color: rgb(0, 132, 38); "><span class="Apple-style-span" style="color: rgb(0, 0, 0); "> agxx(i,j,k) = gxx</span></div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> agxy(i,j,k) = gxy</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> agxz(i,j,k) = gxz</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> agyy(i,j,k) = gyy</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> agyz(i,j,k) = gyz</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> agzz(i,j,k) = gzz</div><p style="margin: 0.0px 0.0px 0.0px 0.0px; font: 11.0px Menlo; min-height: 13.0px"> <br class="webkit-block-placeholder"></p><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> aKxx(i,j,k) = Kxx</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> aKxy(i,j,k) = Kxy</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> aKxz(i,j,k) = Kxz</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> aKyy(i,j,k) = Kyy</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> aKyz(i,j,k) = Kyz</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> aKzz(i,j,k) = Kzz</div></div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> <span style="color: #c22a9c">else</span> <span style="color: #c22a9c">if</span> (mask(i,j,k) <span style="color: #c22a9c">.eq.</span> m_ex) <span style="color: #c22a9c">then</span></div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; color: rgb(0, 132, 38); ">c Excised points</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> agxx(i,j,k) = exval</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> agxy(i,j,k) = exval</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> agxz(i,j,k) = exval</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> agyy(i,j,k) = exval</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> agyz(i,j,k) = exval</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> agzz(i,j,k) = exval</div><p style="margin: 0.0px 0.0px 0.0px 0.0px; font: 11.0px Menlo; min-height: 13.0px"> <br class="webkit-block-placeholder"></p><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> aKxx(i,j,k) = exval</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> aKxy(i,j,k) = exval</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> aKxz(i,j,k) = exval</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> aKyy(i,j,k) = exval</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> aKyz(i,j,k) = exval</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> aKzz(i,j,k) = exval</div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> <span style="color: #c22a9c">endif</span></div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; min-height: 13px; "><br></div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> <span style="color: #c22a9c">enddo</span></div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> <span style="color: #c22a9c">enddo</span></div><div style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0px; font: normal normal normal 11px/normal Menlo; "> <span style="color: #c22a9c">enddo</span></div><div><span style="color: #c22a9c"><br></span></div></div><div><br></div><div>There is no explicit OMP END DO statement because it's optional.</div><div><br></div><div><br></div><div><br></div><div><br></div><div><br><div>
--<br>Scott H. Hawley, Ph.D. <span class="Apple-tab-span" style="white-space: pre; "> </span> <span class="Apple-tab-span" style="white-space: pre; "> </span>Asst. Prof. of Physics <br>Chemistry & Physics Dept <span class="Apple-tab-span" style="white-space: pre; "> </span>Office: Hitch 100D <br>Belmont University <span class="Apple-tab-span" style="white-space: pre; "> </span>Tel: +1-615-460-6206<br>Nashville, TN 37212 USA <span class="Apple-tab-span" style="white-space: pre; "> </span>Fax: +1-615-460-5458<br>PGP Key at <a href="http://sks-keyservers.net">http://sks-keyservers.net</a><br><br><br>
</div>
<br><div><div>On May 18, 2011, at 4:37 PM, Scott Hawley wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite"><div>Erik, Frank, Peter: Thanks guys. I will pursue your suggestions. <br><br><br>--<br>Scott H. Hawley, Ph.D. <span class="Apple-tab-span" style="white-space:pre"> </span> <span class="Apple-tab-span" style="white-space:pre"> </span><span class="Apple-tab-span" style="white-space:pre"> </span>Asst. Prof. of Physics <br>Chemistry & Physics Dept <span class="Apple-tab-span" style="white-space:pre"> </span><span class="Apple-tab-span" style="white-space:pre"> </span>Office: Hitch 100D <br>Belmont University <span class="Apple-tab-span" style="white-space:pre"> </span>Tel: +1-615-460-6206<br>Nashville, TN 37212 USA <span class="Apple-tab-span" style="white-space:pre"> </span>Fax: +1-615-460-5458<br>PGP Key at <a href="http://sks-keyservers.net">http://sks-keyservers.net</a><br><br><br><br>On May 18, 2011, at 12:14 AM, Peter Diener wrote:<br><br><blockquote type="cite">Hi Scott,<br></blockquote><blockquote type="cite"><br></blockquote><blockquote type="cite"><br></blockquote><blockquote type="cite">On Tue, 17 May 2011, Frank Loeffler wrote:<br></blockquote><blockquote type="cite"><br></blockquote><blockquote type="cite"><blockquote type="cite">Hi,<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite">On Tue, May 17, 2011 at 03:02:00PM -0700, Scott Hawley wrote:<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">Do these all be need to be declared as private?<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite">If the temporary variables are declared only inside the loop they are<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite">automatically thread-local. Oh wait, that is Fortran. Well - in that<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite">case you should either declare all private, or (maybe easier) put the<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite">include files into separate functions, declare the temporary variables<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite">only there and call the functions from within the loop, in which case<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite">they also don't have to be specified for openmp (as long as they are not<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite">static).<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">i certainly don't want the various processors overwriting each others'<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">work, which might be what they're doing. -- maybe they're even<br></blockquote></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><blockquote type="cite">generating NaNs which would slow things down a bit!<br></blockquote></blockquote></blockquote><blockquote type="cite"><br></blockquote><blockquote type="cite">Alternatively you may use the DEFAULT(PRIVATE) clause, so that you only <br></blockquote><blockquote type="cite">have to specify the shared variables. However, in that case you have to <br></blockquote><blockquote type="cite">make sure to really declare all the shared variables as shared, since <br></blockquote><blockquote type="cite">otherwise all processors will have to allocate storage and if they are 3d <br></blockquote><blockquote type="cite">variables this will slow down the code and increase memory consumption. <br></blockquote><blockquote type="cite">Also private variables have undefined values on entry to the parallel <br></blockquote><blockquote type="cite">region so not declaring all shared variables properly can also adversely<br></blockquote><blockquote type="cite">affect the result. So be careful.<br></blockquote><blockquote type="cite"><br></blockquote><blockquote type="cite"><blockquote type="cite">You should see that in the results though. It might make sense to first<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite">make sure that the results with different numbers of threads are the<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite">same (depending on the problem you might actually get bit-by-bit<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite">identical results), and work on optimization later. I agree that your<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite">slow-down actually points towards some bug.<br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite"><br></blockquote></blockquote><blockquote type="cite"><blockquote type="cite">Frank<br></blockquote></blockquote><blockquote type="cite"><br></blockquote><blockquote type="cite">Cheers,<br></blockquote><blockquote type="cite"><br></blockquote><blockquote type="cite"> Peter<br></blockquote><blockquote type="cite"><br></blockquote><br><span><PGP.sig></span><span><ATT00001..txt></span></div></blockquote></div><br></div></div></div></body></html>