[petsc-dev] KNC build?

[Satish Balay]

Also:

>>>>>>>>>

Configure Options: --configModules=PETSc.Configure --optionsModule=PETSc.compilerOptions --download-parmetis --download-metis --download-hypre --download-superlu_dist --with-batch=1 --CPPFLAGS=-mmic --CFLAGS="${CEE_OPTIONS}"


Executing: cc -c -o /global/scratch2/sd/madams/petsc-VGa0sP/config.setCompilers/conftest.o -I/global/scratch2/sd/madams/petsc-VGa0sP/config.setCompilers "${CEE_OPTIONS}" -mmic /global/scratch2/sd/madams/petsc-VGa0sP/config.setCompilers/conftest.c
Possible ERROR while running compiler: exit code 256
stderr:
cc1: error: unrecognized command line option "-mmic -mkl -fp-model precise"
cc1: error: unrecognized command line option "-mmic"

<<<<<<

Try expanding ${CEE_OPTIONS} env variable before configure script is
called.. [or not use this env variable and pass in these options
directly - assuming they are valid for this compiler]

Satish

On Tue, 4 Nov 2014, Satish Balay wrote:

> There is no message in configure.log. Looks like you are getting this
> on the terminal.
> 
> It appears to be an icc bug. [bug report to Cray/Intel?]
> 
> You can try:
> 
> export MALLOC_CHECK_=0
> ./configure ...
> 
> http://stackoverflow.com/questions/151268/how-to-force-abort-on-glibc-detected-free-invalid-pointer
> 
> Satish
> 
> On Tue, 4 Nov 2014, Mark Adams wrote:
> 
> > I sent it in the previous email.  here it is again (it is small).
> > 
> > On Tue, Nov 4, 2014 at 5:47 PM, Barry Smith <bsmith at mcs.anl.gov> wrote:
> > 
> > >
> > >   Cool, we need configure.log
> > >
> > > > On Nov 4, 2014, at 4:35 PM, Mark Adams <mfadams at lbl.gov> wrote:
> > > >
> > > > Additionally I get this output:
> > > >
> > > >
> > > ===============================================================================
> > > >              Configuring PETSc to compile on your system
> > > >
> > > ===============================================================================
> > > > TESTING: checkCCompiler from
> > > config.setCompilers(config/BuildSystem/config/setCompilers.py:534)
> > >                               *** glibc detected *** icc: double free or
> > > corruption (top): 0x00000000021a2240 ***
> > > > ======= Backtrace: =========
> > > > /lib64/libc.so.6(+0x75e76)[0x2ae7076cce76]
> > > > /lib64/libc.so.6(+0x789b3)[0x2ae7076cf9b3]
> > > > /lib64/libc.so.6(fclose+0x14d)[0x2ae7076bd4dd]
> > > > icc[0x4fc06c]
> > > > icc[0x4f3177]
> > > > icc[0x4fc151]
> > > > icc[0x4fc05e]
> > > >    .....
> > > >
> > > >
> > > > On Tue, Nov 4, 2014 at 3:59 PM, Mark Adams <mfadams at lbl.gov> wrote:
> > > > Thanks Victor,
> > > >
> > > > I am getting cc error.  Any ideas?
> > > > Mark
> > > >
> > > >
> > > > On Tue, Nov 4, 2014 at 12:38 PM, Victor Eijkhout <
> > > eijkhout at tacc.utexas.edu> wrote:
> > > >
> > > > > On Nov 4, 2014, at 11:32 AM, Mark Adams <mfadams at lbl.gov> wrote:
> > > > >
> > > > > Has anyone built PETSc on KNC (Babbage at NERSC to be specific)?
> > > >
> > > > At TACC under intel 14:
> > > >
> > > > CEE_OPTIONS="-mmic -mkl -fp-model precise"
> > > > ./configure --PETSC_ARCH=mic --with-fc=0 --with-debug=0 \
> > > >     --with-batch=1 --CPPFLAGS=-mmic \
> > > >     --CFLAGS="${CEE_OPTIONS}" --CXXFLAGS="${CEE_OPTIONS}"
> > > --FFLAGS="${CEE_OPTIONS}" \
> > > >     --with-mpi=1 --known-mpi-shared-libraries=1 \
> > > >         --with-mpi-include=${MPICH_HOME}/mic/include \
> > > >         --with-mpi-lib=${MPICH_HOME}/mic/lib \
> > > >         --with-mpicc=/opt/apps/intel13/impi/4.1.0.030/intel64/bin/mpicc
> > > \
> > > >         --with-mpicxx=/opt/apps/intel13/impi/
> > > 4.1.0.030/intel64/bin/mpicxx \
> > > >         --with-mpif90=/opt/apps/intel13/impi/
> > > 4.1.0.030/intel64/bin/mpif90
> > > >
> > > > That first line is still giving me problems. Just nix the fp-model.
> > > >
> > > > Victor.
> > > >
> > > >
> > > >
> > >
> > >
> > 
> 
>