An Alternative Method for Installation of FEAP on Windows

[K.Li]

Dear FEAP users,

I have added an alternative method for installation of FEAP on Windows machines on FEAPwiki at

http://feap.berkeley.edu/wiki/index.php/Installation

This method has been tested with FEAP 8.5 source code and Intel Fortran Compiler 2013 in Microsoft Visual Studio Pro 2012 on a 64-bit Windows machine. Other versions of Intel Fortran Compiler or Visual Studio may also work. The advantage of this new method is that you can build FEAP main program and all the subroutines at once. So you do not need to build the lib and then program.  In addition, you can run FEAP in debug mode in Visual Studio just like other programs.

If you have questions or comments about this new method, please reply to this post.

[neel]

Dear Mr. Li,

Thank you for giving detailed procedure.

I could build with your procedure in Release mode (windows) but getting some errors in building debug mode.

Below are the errors

Error   1    error #7983: The storage extent of the dummy argument exceeds that of the actual argument.   [VT]   D:\xxxx\feappv-4.1i\plot\pltqfl.f   40   
Error   2   Compilation Aborted (code 1)   D:\xxxxxx\feappv-4.1i\plot\pltqfl.f   1   

If I comment  call pltefl(3,ilt,xt,vt,vc,nc) statement and then build below are the errors -

Error   1    error LNK2005: "private: __thiscall type_info::type_info(class type_info const &)" (??0type_info@@AAE@ABV0@@Z) already defined in LIBCMTD.lib(typinfo.obj)   MSVCRT.lib(ti_inst.obj)   
Warning   3    warning LNK4098: defaultlib 'MSVCRT' conflicts with use of other libs; use /NODEFAULTLIB:library   LINK   
Error   4    fatal error LNK1169: one or more multiply defined symbols found   Debug\feappv.exe

[K.Li]

Hi, thank you for your feedback! Glad that it worked with FEAPpv in the release mode too.

For this error, it may be caused by the way FEAPpv was coded in an old fashion, see here

http://feap.berkeley.edu/forum/index.php?topic=241.msg892

So try to make sure that the option Check Routine Interfaces is off.

Right-click lib85 project in the Solution Explorer and the click Properties > Fortran > Diagnostics > Check Routine Interfaces = NO

See also here

https://software.intel.com/en-us/forums/intel-visual-fortran-compiler-for-windows/topic/299025

[Prof. R.L. Taylor]

This is a dimensioning error somewhere in the call to the subroutine.  Are the current dimensions in the routine pltqfl.f these
      logical       :: cont
      integer       :: i1,i2, i, nc,ilq(*),ilt(3)
      real (kind=8) :: xl(3,*),vl(*),vc(nc)
      real (kind=8) :: xt(3,4),vt(4)

[K.Li]

In the current version of FEAPpv, I found this,

Code: [Select]

     
      logical     cont
      integer    i1,i2, i, nc,ilq(*),ilt(3)
      real*8     xl(3,*),vl(*),vc(nc)
      real*8     xt(3,3),vt(3)


[neel]

Dear Mr. Li,

Thank you so much for the suggestion. It has worked.

[Prof. R.L. Taylor]

Check futher int he routine.  If the element is a quad it will go to 4 -- so the dimension is wrong.  I would think the compiler would complain?

[K.Li]

Hi, Prof Taylor:

You are right. The array vt(3) was defined in pltqfl.f with a dimension of 3. But in the subroutine "pltefl", it was defined to be an array of 4 elements. The dimension of vt(*) and xt(*,*) in pltqfl.f should be corrected in FEAPpv.

[Prof. R.L. Taylor]

The dimension should also be corrected in recent versions of FEAP.

[Prof. R.L. Taylor]

While the dimensions on use of xl go to 4, those for the xt(3,3) and vt(3) only go to 3 -- so it does not need to be redimensioned.

[Ren]

Mr. Li,

Thank you for your suggestion. I have done it successfully. However, when I input the input file, which is one of the patch tests in Example Manual, an error exists as,

Code: [Select]

*ERROR* FILNAM: Specified input file: Input_2D_Truss_Problem
does not exist. Reinput filename
I put the Input file in the same folder with the project. I am not sure where I should put this input file.

Thank you in advance.

Ren

[K.Li]

Hi, Ren:

It looks like FEAP cannot find the input file you specified. So try to enter the full name of the file such as "Input_2D_Truss_Problem.txt" if your file has an extension of ".txt".

On Windows machine, you need to enter that part of the file name too.

Normally, I would not put the input file in the project folder. I would copy the FEAP executable file "feap.exe" to the folder where the input file is and then run FEAP from there. 

[Ren]

Hi, Li:

Thank you for you information! It works as your suggestion!

Best Regards,
Ren