The biggest spiral submatrix program written in C doesn't return what's expected

My program should return the biggest spiral submatrix. For 2x2 matrix with values [[1,2], [4,3]] it returns 2x2 submatrix with same values (as it should), but when I enter 2x3 matrix with values [[1,2,5], [4,3,9]] it doesn't return anything, just prints that there are no spiral submatrices.

#include<stdio.h>
#include<stdlib.h>
#include<string.h>
#include<string.h>

void printMat(int n, int m, int mat[n][m]);
int isSpiral(int n, int m, int mat[n][m]);
void printSpiral(int n, int m, int mat[n][m]);
void copyMat(int n, int m, int mat[n][m], int mat2[n][m]);

int main(void) {
    int n, m;
    printf("Enter number of rows and columns: \n");
    scanf("%d %d", &n, &m);
    int mat[n][m];
    printf("Enter %d elements of matrix: \n", n * m);
    for(int i = 0; i < n; i++) {
        for(int j = 0; j < m; j++) {
            scanf("%d", &mat[i][j]);
        }
    }
    printMat(n, m, mat);
    printSpiral(n, m, mat);
    return 0;
}

void printMat(int n, int m, int mat[n][m]) {
    printf("\n");
    for(int i = 0; i < n; i++) {
        for(int j = 0; j < m; j++) {
            printf("%5d", mat[i][j]);
        }
        printf("\n");
    }
}

int isSpiral(int n, int m, int mat[n][m]) {
    int k = 0;

        for(int i = 0; i <= n / 2 && k < n * m; i++) {

             if((i > 0) && (mat[i][i] - mat[i][i - 1] != 1))
                return 0;
            else
                k++;

            for(int j = i; j < m - 1 - i && k < n * m; j++) {
                k++;
                if(mat[i][j + 1] - mat[i][j] != 1)
                    return 0;
            }

            for(int j = i; j < n - 1 - i && k < n * m; j++) {
                k++;
                if(mat[j + 1][m - 1 - i] - mat[j][m - 1 - i] != 1)
                    return 0;
            }

            for(int j = i; j < m - 1 - i && k < n * m; j++) {
                k++;
                if(mat[n - 1 - i][j] - mat[n - 1 - i][j + 1] != 1)
                    return 0;
            }

            for(int j = i + 1; j < n - 1 - i && k < n * m; j++) {
                k++;
                if(mat[j][i] - mat[j + 1][i] != 1)
                    return 0;
            }
        }
    return 1;
}

void copyMat(int n, int m, int mat[n][m], int mat2[n][m]) {
    for(int i = 0; i < n; i++) {
        for(int j = 0; j < m; j++) {
            mat2[i][j] = mat[i][j];
        }
    }
}

void printSpiral(int n, int m, int mat[n][m]) {
    int mat3[n][m], i2 = 0, j2 = 0;
    int maxi2 = -1, maxj2 = -1;
    int mat2[n][m];

    for(int g = n; g > 1; g--) {

        for(int f = m; f > 1; f--) {

            for(int k = 0; k < g - 1; k++) {

                for(int l = 0; l < f - 1; l++) {

                    i2 = 0;
                    for(int i = k; i < g; i++) {
                        j2 = 0;
                        for(int j = l; j < f; j++) {
                            mat2[i2][j2] = mat[i][j];
                            j2++;
                        }
                        i2++;
                    }

                if(isSpiral(i2, j2, mat2)) {
                    if(i2 >= maxi2 && j2 >= maxj2) {
                        maxi2 = i2;
                        maxj2 = j2;
                        copyMat(maxi2, maxj2, mat2, mat3);
                    }
                }
                }
            }
        }
    }

    if(maxi2 > 0 && maxj2 > 0)
        printMat(maxi2, maxj2, mat3);
    else
        printf("\nNo spiral submatrix");
}

I tried debugging it but it didn't help much. I noticed that when it calls isSpiral function from printSpiral function it returns 0 for [[1,2], [4,3]] submatrix and it returns it when variable k is equal to 3 at second nested for, which shouldn't do.

This my first time asking question here so I apologize in advance if my question doesn't have proper form.

The problem was the way I was declaring my functions parameters.

Here, I created isSpiral(int n, int m, int mat[n][m]), and the problem was int mat[n][m], because n and m would change the size of my mat variable (matrix).

If the size was changed, the function would get wrong indexes, so I just added two new parameters, int i2 and int j2:

isSpiral(int n, int m, int i2, int j2, int mat[i2][j2]).

That way, the size of my matrix remains the same, constant, as it should, and limits for submatrices are variable.