<div class="gmail_extra">On Mon, Apr 23, 2012 at 5:26 PM, Satish Balay <span dir="ltr"><<a href="mailto:balay@mcs.anl.gov" target="_blank">balay@mcs.anl.gov</a>></span> wrote:<br><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
'make libfast' can't handle different file types [but 'make lib' can].<br>
<br>
For libfast to work - you can split up the sources into multiple dirs.<br>
<br>
i.e<br>
<br>
src/mat/impls/dia/seq/dia.c<br>
src/mat/impls/dia/seq/seqcusp/<a href="http://diakerns.cu" target="_blank">diakerns.cu</a></blockquote><div><br></div><div>The Python build can handle this.</div><div><br></div><div> Matt</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<span class="HOEnZb"><font color="#888888"><br>
Satish<br>
</font></span><div class="HOEnZb"><div class="h5"><br>
On Mon, 23 Apr 2012, Daniel Lowell wrote:<br>
<br>
> Hi all,<br>
><br>
> I'm building a matrix type with two separate files dai.c and <a href="http://diakerns.cu" target="_blank">diakerns.cu</a>.<br>
> If I "make all" in the PETSc root dir, it will be unable to find the object<br>
> file from the .cu file; i.e., I get this:<br>
><br>
> /usr/bin/ar: diakerns.o: No such file or directory<br>
><br>
> However if I merely do a "make lib" inside the matrix type directory I have<br>
> no problem. This has happened to other types I've create, I end up just<br>
> shoving everything into one large .cu file to quickly deal with it.<br>
> Here what the folder level makefile looks like:<br>
><br>
><br>
><br>
> SOURCECU = <a href="http://diakerns.cu" target="_blank">diakerns.cu</a><br>
> SOURCEC = dia.c<br>
> SOURCEF =<br>
> SOURCEH = dia.h<br>
> OBJSC = dia.o diakerns.o<br>
> OBJSF =<br>
> LIBBASE = libpetscmat<br>
> DIRS =<br>
> MANSEC = Mat<br>
> LOCDIR = src/mat/impls/dia/seq/<br>
><br>
><br>
><br>
><br>
> This is what my configuration script for petsc root looks like:<br>
><br>
><br>
> MPI_DIR="--with-mpi-dir=$MPICH2_HOME"<br>
> LAD="--download-f-blas-lapack=yes"<br>
> CUD="--with-cuda-dir=/soft/cuda-4.2/cuda"<br>
> ./config/configure.py $MPI_DIR $LAD $CUD \<br>
> --with-debugging=1 \<br>
> --with-cudac="nvcc -m64 -g -G -v" \<br>
> --with-precision=double \<br>
> --with-clanguage=c \<br>
> --with-cuda-arch=sm_20 \<br>
><br>
><br>
><br>
><br>
><br>
><br>
<br>
</div></div></blockquote></div><br><br clear="all"><div><br></div>-- <br>What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead.<br>
-- Norbert Wiener<br>
</div>