Stefano:<div>Thanks for sending us the patch. Is this for petsc-3.3 or petsc-dev?</div><div>I'll test the patch first, then push to the repository.</div><div><br><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Yesterday it was the first time I try to build a single-precision version of PETSc with some external packages.<br>I noticed that SuperLU and MUMPS, which are currently fully compiled (i.e. with scalar types s,d,c and z) by the configure, cannot be used when requiring single precision at configure time. So I decided to wrote a patch to enable their use when PETSC_USE_REAL_SINGLE is defined.<br>
<br>Attached please find a patch. Is there any reason for which you did not already provide this functionality?<br></blockquote><div>No user has ever requested single precision for these external packages.</div><div> </div>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">I tested the patch on BlueGene/Q and it works for scalar types real. Currently, I don't have a complex version of PETSc installed. Can someone test it?<br clear="all">
</blockquote><div>I'll test it. </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>I also noticed that UMFPACK (which itself does not have support for single precision), is accepted by the configure by requiring --with-precision=single --download-umfpack=1. Is it normal? It breaks the final linking of the executables. <br>
</blockquote><div>This might be a bug in petsc. I'll check it.</div><div><br></div><div>Hong </div></div><br></div>