<div dir="ltr">BTW, I seemed to have gotten a debug build to work yesterday. I have not idea what I did different but I have tried every permutation of building the optimized version, without luck.</div><div class="gmail_extra"><br><div class="gmail_quote">On Thu, Jul 13, 2017 at 2:31 PM, Barry Smith <span dir="ltr"><<a href="mailto:bsmith@mcs.anl.gov" target="_blank">bsmith@mcs.anl.gov</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><span class=""><br>
> On Jul 13, 2017, at 1:28 PM, Tobin Isaac <<a href="mailto:tisaac@gatech.edu">tisaac@gatech.edu</a>> wrote:<br>
><br>
> p4est is a pretty standard autotools project, I'll check if there's any better cross-compiling support than what we're using.<br>
<br>
</span> The problem with most cross-compiling support in autotools is it assumes it "somehow" knows about all systems you are cross compiling for; which won't be true for new systems that didn't exist when the cross compiling support was implemented. This is why I don't have much respect for auto tools cross compiling support.<br>
<div class="HOEnZb"><div class="h5"><br>
<br>
><br>
> On July 13, 2017 2:18:51 PM EDT, Barry Smith <<a href="mailto:bsmith@mcs.anl.gov">bsmith@mcs.anl.gov</a>> wrote:<br>
>><br>
>> p4est configure should provide an option to not run compiled code and<br>
>> instead have needed values passed in as configure arguments.<br>
>><br>
>>> On Jul 13, 2017, at 1:07 PM, Matthew Knepley <<a href="mailto:knepley@gmail.com">knepley@gmail.com</a>><br>
>> wrote:<br>
>>><br>
>>> On Thu, Jul 13, 2017 at 12:32 PM, Mark Adams <<a href="mailto:mfadams@lbl.gov">mfadams@lbl.gov</a>> wrote:<br>
>>><br>
>>> Are you cross-compiling there?<br>
>>><br>
>>> configure:4469: checking whether we are cross compiling<br>
>>> configure:4477: ftn -o conftest -g -O -fp-model fast -qopt-report=5<br>
>> -qopenmp-simd conftest.f<br>
>> -Wl,-rpath,/global/homes/m/<wbr>madams/petsc/arch-cori-knl-<wbr>opt-intel/lib<br>
>> -L/global/homes/m/madams/<wbr>petsc/arch-cori-knl-opt-intel/<wbr>lib -lflapack<br>
>> -Wl,-rpath,/global/homes/m/<wbr>madams/petsc/arch-cori-knl-<wbr>opt-intel/lib<br>
>> -L/global/homes/m/madams/<wbr>petsc/arch-cori-knl-opt-intel/<wbr>lib -lfblas<br>
>> -L/opt/cray/pe/mpt/7.4.4/gni/<wbr>sma/lib64<br>
>> -L/opt/cray/pe/libsci/16.09.1/<wbr>INTEL/15.0/mic_knl/lib<br>
>> -L/opt/cray/pe/hdf5-parallel/<wbr>1.10.0/INTEL/15.0/lib<br>
>> -L/opt/cray/dmapp/default/<wbr>lib64<br>
>> -L/opt/cray/pe/mpt/7.4.4/gni/<wbr>mpich-intel/16.0/lib<br>
>> -L/usr/common/software/<wbr>darshan/3.1.4/lib<br>
>> -L/opt/cray/rca/2.1.6_<wbr>g2c60fbf-2.265/lib64<br>
>> -L/opt/cray/alps/6.3.4-2.21/<wbr>lib64<br>
>> -L/opt/cray/xpmem/2.1.1_<wbr>gf9c9084-2.38/lib64<br>
>> -L/opt/cray/dmapp/7.1.1-39.37/<wbr>lib64<br>
>> -L/opt/cray/pe/pmi/5.0.10-1.<wbr>0000.11050.0.0.ari/lib64<br>
>> -L/opt/cray/ugni/6.0.15-2.2/<wbr>lib64 -L/opt/cray/udreg/2.3.2-7.54/<wbr>lib64<br>
>> -L/opt/cray/pe/atp/2.1.0/<wbr>libApp -L/opt/cray/wlm_detect/1.2.1-<wbr>3.10/lib64<br>
>> -L/usr/common/software/h5part/<wbr>1.6.6/lib<br>
>> -L/opt/intel/compilers_and_<wbr>libraries_2017.2.174/linux/<wbr>compiler/lib/intel64<br>
>> -L/opt/intel/compilers_and_<wbr>libraries_2017.2.174/linux/<wbr>mkl/lib/intel64<br>
>> -L/opt/intel/compilers_and_<wbr>libraries_2017.2.174/linux/<wbr>compiler/lib/intel64_lin<br>
>> -L/usr/lib64/gcc/x86_64-suse-<wbr>linux/4.8 -L/usr/x86_64-suse-linux/lib<br>
>> -lhdf5hl_fortran_parallel -lhdf5_fortran_parallel >&5<br>
>>> ifort: remark #10397: optimization reports are generated in *.optrpt<br>
>> files in the output location<br>
>>> configure:4481: $? = 0<br>
>>> configure:4488: ./conftest<br>
>>> libhugetlbfs [cori10:124117]: WARNING: Hugepage size 2097152<br>
>> unavailable<br>
>>> Please verify that both the operating system and the processor<br>
>> support Intel(R) AVX512F, ADX, RDSEED, AVX512ER, AVX512PF and AVX512CD<br>
>> instructions.<br>
>>><br>
>>> Matt<br>
>>><br>
>>> On Thu, Jul 13, 2017 at 12:21 PM, Matthew Knepley <<a href="mailto:knepley@gmail.com">knepley@gmail.com</a>><br>
>> wrote:<br>
>>> Toby, this is from the p4est configure:<br>
>>><br>
>>> o-----------------------------<wbr>----------<br>
>>> | Checking MPI and related programs<br>
>>> o-----------------------------<wbr>----------<br>
>>> checking whether we are using MPI... yes<br>
>>> checking whether we are using MPI I/O... yes<br>
>>> checking whether we are using MPI_Init_thread... yes<br>
>>> checking for mpiexec... no<br>
>>> checking for mpirun... no<br>
>>> configure: F77 set to ftn<br>
>>> configure: FC set to ftn<br>
>>> configure: CC set to cc<br>
>>> configure: CXX set to CC<br>
>>> checking whether the Fortran 77 compiler works... yes<br>
>>> checking for Fortran 77 compiler default output file name... a.out<br>
>>> checking for suffix of executables...<br>
>>> checking whether we are cross compiling...configure: error: in<br>
>> `/global/u2/m/madams/petsc/<wbr>arch-cori-knl-opt-intel/<wbr>externalpackages/git.p4est':<br>
>>> configure: error: cannot run Fortran 77 compiled programs.<br>
>>> If you meant to cross compile, use `--host'.<br>
>>> See `config.log' for more details<br>
>>><br>
>>> Mark, gets us<br>
>> /global/u2/m/madams/petsc/<wbr>arch-cori-knl-opt-intel/<wbr>externalpackages/git.p4est/<wbr>config.log<br>
>>><br>
>>> Thanks,<br>
>>><br>
>>> Matt<br>
>>><br>
>>><br>
>>> On Thu, Jul 13, 2017 at 11:15 AM, Mark Adams <<a href="mailto:mfadams@lbl.gov">mfadams@lbl.gov</a>> wrote:<br>
>>> I always get this error on Cori at NERSC, KNL, but I can usually<br>
>> rerun configure and it runs the second time. Well, I can't seem to get<br>
>> this to work now.<br>
>>> Thanks,<br>
>>> Mark<br>
>>><br>
>>><br>
>>><br>
>>> --<br>
>>> What most experimenters take for granted before they begin their<br>
>> experiments is infinitely more interesting than any results to which<br>
>> their experiments lead.<br>
>>> -- Norbert Wiener<br>
>>><br>
>>> <a href="http://www.caam.rice.edu/~mk51/" rel="noreferrer" target="_blank">http://www.caam.rice.edu/~<wbr>mk51/</a><br>
>>><br>
>>><br>
>>><br>
>>><br>
>>> --<br>
>>> What most experimenters take for granted before they begin their<br>
>> experiments is infinitely more interesting than any results to which<br>
>> their experiments lead.<br>
>>> -- Norbert Wiener<br>
>>><br>
>>> <a href="http://www.caam.rice.edu/~mk51/" rel="noreferrer" target="_blank">http://www.caam.rice.edu/~<wbr>mk51/</a><br>
<br>
</div></div></blockquote></div><br></div>