Example run output (what “success” looks like)

This page shows representative stdout/stderr for common entry points so you can compare your terminal to a known-good shape. Exact numbers and formatting depend on OpenPFC version, MPI, and log settings; use this as a shape check, not a byte-for-byte diff.

examples/05_simulator

Source: examples/05_simulator.cpp. Rank 0 prints:

  1. Create initial condition — once, when the Gaussian IC is applied.

  2. n = …, t = …, min = …, max = … — one line after initialization and one per timestep until the run finishes (the Time range is fixed in code: t1 and 42 steps).

  3. Test pass! — if the final psi maximum is within tolerance of 0.5; otherwise Test failed! on stderr.

Example shape (middle lines omitted):

Create initial condition
n = 0, t = 0.000000000000, min = ..., max = ...
n = 1, t = ..., min = ..., max = ...
...
n = 42, t = ..., min = ..., max = ...
Test pass!

Run from the build tree, e.g. mpirun -n 4 ./examples/05_simulator. Other ranks stay quiet for the std::cout paths above.

apps/tungsten/tungsten (CPU)

The App path logs with the [app] prefix on rank 0 (see include/openpfc/frontend/ui/app.hpp). You should see, among others:

  • [app] Reading JSON configuration from … or the TOML equivalent

  • [app] Effective configuration (JSON): followed by an indented dump of the parsed settings

  • [app] Domain: … summarizing the grid

  • [app] Initializing model...

  • [app] Starting time integration (Simulator integrator API)

If model.params validation is enabled in your binary, you may also see a Configuration Validation Summary block (root README.md, parameter_validation.md) before the time loop.

Exit code 0 from mpirun and completion without an exception mean the driver finished; output files appear only when saveat and fields / writers are configured (see io_results.md).

GPU binaries (tungsten_cuda, tungsten_hip, …)

Log shape matches the CPU App flow. Additional lines may appear when GPU-aware MPI or device backends are compiled in (see tutorials/gpu_app_quickstart.md and INSTALL.LUMI.md). Your JSON/TOML should set plan_options / backend consistently with the build (examples/fft_backend_selection.toml).

See also