python - 从python调用静态链接的C库在Windows上永久挂起，但在OSX或Linux上永远挂起

我有以下c库tridiag.c：

#include <stdio.h>
#include <stdlib.h>
#include <time.h>
#include <math.h>
#define PI 3.14159265358979323846
typedef struct tridiag_matrix{
    double *a;
    double *b;
    double *c;
} Tridiag_M;

inline
double interpl_4(double a,double b,double c,double d,double frac){
    return b*(1-frac)+c*frac;
};
double cubic_interpolate( float y0, float y1, float y2, float y3, float mu ) {

   double a0, a1, a2, a3, mu2;

   mu2 = mu*mu;
   a0 = y3 - y2 - y0 + y1; //p
   a1 = y0 - y1 - a0;
   a2 = y2 - y0;
   a3 = y1;

   return ( a0*mu*mu2 + a1*mu2 + a2*mu + a3 );
}
inline double cos_interpl(double a,double b,double frac){
    double mu2=(1-cos(frac*PI))/2;
    return a*(1-mu2)+b*mu2;
};

void solve_tridiagonal(Tridiag_M *t, double *r,double *x,int N) {
    int in;
    double *cprime=(double*) malloc(N*sizeof(double));
    cprime[0] = t->c[0] / t->b[0];
    x[0] = r[0] / t->b[0];

    /* loop from 1 to N - 1 inclusive */
    for (in = 1; in < N; in++) {
        double m = 1.0 / (t->b[in] - t->a[in] * cprime[in - 1]);
        cprime[in] = t->c[in] * m;
        x[in] = (r[in] - t->a[in] * x[in - 1]) * m;
    }
    /* loop from N - 2 to 0 inclusive, safely testing loop end condition */
    for (in = N - 1; in-- > 0; ){
        x[in] = x[in] - cprime[in] * x[in + 1]; /*wrong cprime[in] ebasta!*/
    }
        /* free scratch space */
    free(cprime);
}

void delay_line(double *Y, int *delay0,int *delay1,int *delay2,int *delay3,double *dev,double *out,int M,int N){
    int i;
    for(i=0;i<N;i++){
        int k=M*i;
        if(dev[i]<1){
            out[i]=cubic_interpolate(Y[k+delay0[i]],Y[k+delay1[i]],Y[k+delay2[i]],Y[k+delay3[i]],dev[i]);
        }
        else{
            out[i]=cubic_interpolate(Y[k+(delay0[i]+1)%M],Y[k+(delay1[i]+1)%M],Y[k+(delay2[i]+1)%M],Y[k+(delay3[i]+1)%M],dev[i]-1);
        }

    }
}

void calculate_g(double *V,double *Y,double *sherad_factor,double *sheraD,double *sheraRho,double *Yzweig,double *omega,double *g,double d_m_factor,const int n){
    int i;
    g[0]=d_m_factor*V[0];
    for(i=1;i<n;i++){
        g[i]=sherad_factor[i]*sheraD[i]*V[i]+omega[i]*omega[i]*(Y[i]+sheraRho[i]*Yzweig[i]);
    }
}

我正在Windows 7 64位上使用gcc 4.5.0 / mingw32进行编译。

gcc -shared -m32 -o -Wall tridiag.so_windows tridiag.c。

干净地退出，没有任何输出。

在python 3.5.x中，我正在加载库：

# load C library
libtrisolv = np.ctypeslib.load_library(tridiagName, resources_root)

# load tridiagonal solver function and defines input
libtrisolv.solve_tridiagonal.restype = None
libtrisolv.solve_tridiagonal.argtypes = [ctypes.POINTER(TridiagMatrix),  # aa
                                         PDOUBLE,  # vv
                                         PDOUBLE,  # solution
                                         INT,  # nrows
                                         ]

libtrisolv.delay_line.restype = None  # TODO SPEEDUP W POINTERS!
libtrisolv.delay_line.argtypes = [PDOUBLE,  # in_matrix
                                  PINT,  # delay1
                                  PINT,  # delay2
                                  PINT,  # delay1
                                  PINT,  # delay1
                                  PDOUBLE,  # dev
                                  PDOUBLE,  # YZweig
                                  INT,  # delay_buffer_length
                                  INT  # n
                                  ]

这很好。

但是，当我调用任何成员函数时，控制永远不会返回到python：

libtrisolv.delay_line(
        model.Ybuffer_pointer, model.Zrp_pointer, model.Zrp1_pointer,
        model.Zrp2_pointer, model.Zrp3_pointer, model.Dev_pointer,
        model.YZweig_pointer, ctypes.c_int(model.YbufferLgt),
        ctypes.c_int(model.n + 1))

在Mac / Linux上，大约需要3秒钟才能返回，而且确实如此。在Windows上，我让它运行了一整夜而没有返回。

这是因为我在64位系统上使用mingw32，还是发生了其他情况？

最佳答案

得到它的工作。我安装了Mingw-64与
gcc (x86_64-posix-seh-rev0, Built by MinGW-W64 project) 5.3.0

并用

gcc -Wall -shared -m64 -o tridiag.so_windows tridiag.c

现在运行。所以我认为这只是一个愚蠢的体系结构问题。