Use uppercase for all Fortran instructions, variable names

Use lowercase for pre-compiler directives

Comments should be written in English only (uppercase & lowercase accepted)

Tabulations are forbidden (to be replaced by spaces at the beginning of line)

Use meaningful names for variables, in English, keep the same name throughout the code

Systematically indent your blocks (IF/END IF, DO/END DO, ...) in a homogeneous way for clarity of the code

In case of long IF/ELSE/END IF block, it is recommended to add comment like below: 

By default, one file contains only one subroutine or function, except when the understanding of the code is facilitated by grouping few procedures

The same rule applies for modules: one module per file, same name

For module several subroutines can be defined under the CONTAINS close

Each source file should have the copyright notice

Each procedure needs to have a conforming header, containing by order of appearance:

• A minimal description of the routine

• The list of caller subroutines/functions

• The list of called subroutines/functions

• The SUBROUTINE/FUNCTION declaration with the list of dummy arguments, 5 per line, matching the call declaration

• The used modules

• The implicit_f.inc (IMPLICIT NONE)

• The optional remaining global parameters and commons if any (for compatibility with the past)

• The list of dummy arguments, one per line with its type, INTENT attribute, explicit bounds in case of array, a short description of the variable

• The local variables

• The source code

It is important to comment algorithms, especially when non straightforward coding is used

Comments are written in English

Comments respect Fortran90 standard. The use of "!" at beginning of line is preferred to "C" or "*"

Except for preprocessor directives, the following characters ',", \, /*, */, # are forbidden, even inside comments. Especially "\" is dangerous as it is interpreted as a continuation line by the preprocessor

 Example:

Module Format

Generic format of a Fortran90 module is as follows:     

Naming Convention 

Module name is defined as follows: MODULENAME_MOD

With module file name: modulename_mod.F

Module file is placed at the same location as other files used by the option

Restart Variables

All the variables communicated between Starter and Engine are declared in module RESTART_MOD, used by subroutine ARRALLOC for their allocation, by RDRESB for their reading from RESTART file, then by RESOL_HEAD in which they are used as argument for RESOL subroutine. All these argument variables are then passed by argument to procedures called from RESOL, ...)

Interface definition

Fortran90 interface allows the compiler to do additional checks like coherency between argument types, attributes and number between calling and callee routines. It is required in some cases like when a procedure has a dummy argument with attribute ALLOCATABLE, POINTER, TARGET

In practice, it was introduced in few places of the code for routines which were called at several different places

Such coherency is automatically tested by QA static analysis tools (Forcheck).

And for pointer, good practice is to use derived data types instead of pointer directly

Therefore, the remaining use of interface is regarding routine with optional arguments

This feature should be spread in the code instead of adding additional “dummy” arguments

Dynamic memory allocation mechanism

Large arrays should be allocatable arrays: Avoid using pointer when possible, and do not use automatic arrays. Arrays should be explicitly deallocated as soon as possible

Developers are required to check the success of the allocation 

For large arrays it is preferred to print a specific message, with some advice for the user, or at least the option concerned by this failure   

Nevertheless, there are macros that will check and print a generic error message

• MY_ALLOCATE needs the file my_allocate.inc to be included and the module MESSAGE_MOD to be used

• It is still mandatory to call DEALLOCATE;

• MY_ALLOCATE2D and MY_ALLOCATE3D to be used for respectively 2 and 3 dimensional arrays.

See my_allocate.inc in GitHub for the definitions

Global Memory

Memory allocation of global data structures, arrays and derived data types, should be done at the highest level, in LECTUR for Starter, in RADIOSS2 or RESOL for Engine

It is advised to use derived data type with structure of arrays. This way it is possible to declare the variable at the upper level, gather the allocation of array members in a dedicated subroutine, then use the variable in procedures called at lower level without losing traceability

Local Memory

In a procedure, local variable allocation method depends on its size:

Shared Memory Programming (SMP) and memory allocation

For OpenRadioss Engine, OpenMP programming model is used for second level parallelization

By default any memory allocation done outside of a parallel section is shared between threads

Most of the parallel sections are started from RESOL. So for the need of a shared memory array, the simplest way is to declare it at the level of RESOL. It will be shared inside the different parallel sections started from RESOL, possibly passed by argument to routines called from RESOL

The same way, any variable defined in a common or module is shared by default

For pointer, notice that a single thread needs to allocate and deallocate it. Synchronization may be needed after the allocation and before the deallocation

The !$OMP THREADPRIVATE directive overrides default behavior by creating thread local storage variables

Array Aliasing

Example:

CALL MYROUTINE(A(1:8), A(7:10)) is not accepted because A(7) and A(8) are common to two dummy arguments