Conversion of argument 2 from "real_T [2]" to "const int []" not possible;

Question

Bhushan Ravindra Attarde 2020-5-9

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/524218-conversion-of-argument-2-from-real_t-2-to-const-int-not-possible

评论： Geoff Hayes 2020-5-11

Hi I am trying to call my function in coder.ceval method. The function declaration of my code in header file is :

STATE_SPACE_EQN_DLL_EXPORT extern void State_Space_Eqn(const double a_data[],
  const int a_size[2], const double b_data[], const int b_size[2], const double
  c_data[], const int c_size[2], const double d_data[], const int d_size[2],
  double x_data[], int x_size[2], const double u_data[], const int u_size[2],
  double out_data[], int out_size[2]);

And I am calling it by using coder.ceval method as follows:

function output = State_Space_Eqn_X(A,B,C,D,X,U)
    % Running in generated code, call library function.
    coder.cinclude('State_Space_Eqn.h');
    
    output = coder.nullcopy(U);
     a = size(A);
     b = size(B);
     c = size(C);
     d = size(D);
     x = size(X);
     u = size(U);
     out = size(output);
              
  
    coder.ceval('State_Space_Eqn', coder.rref(A),a,coder.rref(B),b,coder.rref(C),c,coder.rref(D),d,coder.rref(X),x,coder.rref(U),u,coder.wref(output),out);
end

When I call this function in Matlab function block on Simulink, I am getting this error:

Conversion of argument 2 from "real_T [2]" to "const int []" not possible;
The types referenced are not linked; the conversion requires a reinterpret_cast operator or a type conversion in C or function format.

Can someone please help me in this code?

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Geoff Hayes 2020-5-9

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/524218-conversion-of-argument-2-from-real_t-2-to-const-int-not-possible#answer_431373

在 MATLAB Online 中打开

Bhushan - I don't use Simulink so don't know for sure if this will work but if the error message is telling you that it can't convert from a real (double) to an integer, then maybe you have to do the conversion yourself. Remember that the class type of the array returned by size is still a double so try to do the conversion there

function output = State_Space_Eqn_X(A,B,C,D,X,U)
    % Running in generated code, call library function.
    coder.cinclude('State_Space_Eqn.h');
    
    output = coder.nullcopy(U);
    a = int32(size(A));
    b = int32(size(B));
    c = int32(size(C));
    d = int32(size(D));
    x = int32(size(X));
    u = int32(size(U));
    out = int32(size(output));
    
    coder.ceval('State_Space_Eqn', coder.rref(A),a,coder.rref(B),b,coder.rref(C),c,coder.rref(D),d,coder.rref(X),x,coder.rref(U),u,coder.wref(output),out);
end

I'm assuming that an int32 is equivalent in C to an int.

2 个评论
显示无隐藏无

Bhushan Ravindra Attarde 2020-5-9

Yes. This is working perfectly fine for me. I tried using int64 but was showing same error. Can you please explain why it was not working with int64?

Thank you for your valuable time.

Geoff Hayes 2020-5-11

Bhushan - in C/C++, an integer, int, is typically 32 bits so that is why you need to convert your size dimensions from the (probably) 64-bit doubles to a 32-bit integer via int32. When you use int64, this will convert the doubles to 64-bit integers which is not the expected/required 32 bits as given in the function signature.

请先登录，再进行评论。