FMU/FMI not working neither for local machine nor for cross compilation

[ritesh.sharma@…]

Explaining the scenarios with building and compiling FMI models:
CASE 1: OUTPUT If OMSimulator is used with an FMU created on ARM processor specifying platform as "arm-linux-gnueabihf": (why do we need this when it is not cross-platform compilation?)
OMSimulator works but the output is incorrect when plotted

CASE 2: OUTPUT If OMSimulator is used with an FMU created on ARM processor but without specifying platform: (with the assumption that it will create FMU for the native machine)
execution fails with the following message

error: module FMILIB: Could not change to the DLL directory /tmp/temp_ZaD1mT/binaries/arm-linux-gnueabihf/

error: module FMILIB: The FMU contains no binary for this platform.
error: Could not create the DLL loading mechanism (C-API). Error: The FMU contains no binary for this platform.
info: Result file: model_res.mat
info: Simulation finished.
info: Removed model from scope: model

CASE 3: OUTPUT If OMShell-terminal is used for importing FMU and then simulating:(irrespective of specifying platform)
importFMU() works fine and generates a .mo file but simulation fails with following message

simulate(RLC_me_FMU, stopTime=30.0)

record SimulationResult

resultFile = "",
messages = "Simulation execution failed for model: RLC_me_FMU

module = FMILIB, log level = FATAL: Could not change to the DLL directory /home/pi/Desktop/NPCIL/OMS/binaries/linux32/
module = FMILIB, log level = FATAL: The FMU contains no binary for this platform.
assert | debug | Loading of FMU dynamic link library failed with status : UNEXPECTED
assert | info | simulation terminated by an assertion at initialization
Limited backtrace at point of segmentation fault
Segmentation fault
"
end SimulationResult;

CASE 4: Output when Cross-compiling on a different machine for ARM:
Same output as CASE 2 and CASE 3

[Francesco Casella]

@ritesh, please provide test models (you can attach them to this ticket) so that we can reproduce the issues, otherwise it is hard to fix them.

[ritesh.sharma@…]

Problem is generic and not for just one code but I tried for a simple RLC circuit example. Here is the code. It showed the behavior as described in the 4 cases in the ticket:

model RLC
  type Voltage=Real(unit="V");
  type Current=Real(unit="A");
  type Resistance=Real(unit="Ohm");
  type Capacitance=Real(unit="F");
  type Inductance=Real(unit="H");
  
  parameter Voltage Vb=24 "Battery voltage";
  parameter Inductance L = 1;
  parameter Resistance R = 100;
  parameter Capacitance C = 1e-3;
  
  Voltage V;
  Current i_L;
  Current i_R;
  Current i_C;

equation
  V = i_R*R;
  C*der(V) = i_C;
  L*der(i_L) = (Vb-V);
  i_L=i_R+i_C;
annotation(
    experiment(StartTime = 0, StopTime = 30, Interval = 0.01),
    __OpenModelica_simulationFlags(rt = "1.0")
    );

end RLC;