Solved: Strange stack error causes program terminated without an error - Page 2

yang__xiaodong · ‎09-24-2021

I'm an experienced fortran programmer, yet a really strange problem confused me.

As mentioned, the program terminates occasionally after the output

"============================ start performing inversion ..."

It seems a stack smash occurs.

The visual studio 2019 project (with oneapi 2021.3 installed) and needed files are attached. You could run the program with

DCFI3D -a mod1

MKL libraries are needed. Could anyone help find the error? Thanks very much.

yang__xiaodong · ‎09-25-2021

Okay, now I find where the problem is.

It is because that when I use the MKL subroutine mkl_sparse_d_mm that has a proto of

stat = mkl_sparse_d_mm (operation, alpha,A, descr, layout, B, columns, ldb,beta, C,ldc)

I declare the variable columns larger than what I need. Thus, it results in a C which takes more memory space than expected. This further causes heap corruption, and no error message could be generated by the Fortran compiler.

View solution in original post

jimdempseyatthecove · ‎09-26-2021

Yang, I wish to thank you for replying with the solution to your problem. This is valuable information for others reading this forum thread.

Jim Dempsey

JohnNichols · ‎09-26-2021

@yang__xiaodong - jolly well done.

Next question, what are you using the mesh for?

yang__xiaodong · ‎09-28-2021

HAPPY solving the problem.

The mesh is used for finite element solving of a large size 3-D direct current electrostatic problem. In which geophysicists eject electric current into the ground, and voltages are measured elsewhere.

As a boundary condition V=0 is applied, the mesh must be large enough to ensure the voltage shrinks.

I think, why the error is hard to find, is because that the MKL interface, including the mkl_sparse_d_mm subroutine, may have a C proto and then be converted to Fortran. Thus, not enough info could be captured by the Fortran compiler.