[petsc-users] Code break when PETSc not debugging

[Jed Brown]

Manuel Valera <mvalera-w at sdsu.edu> writes:

> Hello everyone,
>
> I am working on running part of my code in a GPU, recently i was able to
> run the whole model using one P100 GPU and one processor with good timing
> but using  --with-debugging=1 as configure argument,
>
> With this in mind i compiled PETSc in a separate folder with the same exact
> flags except for --with-debugging=no instead to do some profiling, but this
> was enough to give segfault as an error after running the code, it looks
> the error happens just after solving the linear system,
>
> Any idea on why this may be happening?

Run in a debugger and send a stack trace.

> My configure options:
>
>  ./configure PETSC_ARCH=cuda  --with-mpi-dir=/usr/lib64/openmpi
> --COPTFLAGS='-O2' --CXXOPTFLAGS='-O2' --FOPTFLAGS='-O2'
> --with-shared-libraries=1 --with-debugging=no --with-cuda=1
> --CUDAFLAGS=-arch=sm_60  --with-blaslapack-dir=/usr/lib64
> --download-viennacl
>
> My running arguments:
>
> mpirun -n 1 ./gcmLEP tc=TestCases/LockRelease/LE_401x6x101/ jid=cuda_dt0.1
> -dm_vec_type viennacl -dm_mat_type aijviennacl -pc_type saviennacl
> -log_view
>
> Thanks.