I uploaded a program on the online compiler mbed that uses the functions log2 and pow of the library math.h and gives me the following error: Identifier "log2" is undefined "num_bit=1+log2(decimale);" Expression must have integral or enum type "temp[num_bit-1-i]=decimale%2+'0';" More than once instance of overloaded function "pow" matches the argument list "dec=dec+(binario[j]-'0' *pow(2, 5-j);" Where is the problem?

#include "mbed.h"
#include "math.h"

#define _M_LN2  0.693147180559945309417 // Natural log of 2
#define log2(x) (log(x)/_M_LN2)

int main(void) {
    
    float   decimale = 12;
    int     dec;
    int     temp[100];
    int     binario[100];
    uint8_t i = 0;
    uint8_t j = 0;

    int num_bit = 1 + log2(decimale);
    temp[num_bit - 1 - i] = (int)decimale % 2 + '0';
    dec = dec + (binario[j] - '0') * pow((float)2, (float)(5-j));

    while(1) {}
}

Hello Simone, *There is no ##log2## function defined in the math library linked by the online compiler, so you have to define one by yourself. *The ##pow## function is overloaded, i.e. it is defined with various argument lists. As a consequence, it must be called with unambiguous arguments (signature) - see [[https://developer.mbed.org/questions/77726/mathh-function-pow-error]] for more explanation. For example the following code doesn't do anything useful but compiles with no errors: <<code>> #include "mbed.h" #include "math.h" #define _M_LN2 0.693147180559945309417 // Natural log of 2 #define log2(x) (log(x)/_M_LN2) int main(void) { float decimale = 12; int dec; int temp[100]; int binario[100]; uint8_t i = 0; uint8_t j = 0; int num_bit = 1 + log2(decimale); temp[num_bit - 1 - i] = (int)decimale % 2 + '0'; dec = dec + (binario[j] - '0') * pow((float)2, (float)(5-j)); while(1) {} } <</code>>