Compiler error: unmatched call to pow(...)

My company has a piece of software, sporting a rather large codebase. Recently I was assigned the task of checking wether the code would compile on an x86_64 target using gcc 4.1.2. I've gotten pretty far in the compilation with very minor modifications to the code but just this morning I got a somewhat confusing compile error.

The code is trying, and failing, to call powfrom <cmath> using int, unsigned int& as parameters. The compiler spits out an error because it can't find a suitable match to call. The overloads for pow in <cmath> are as follows:

double pow(double base, double exponent)
long double pow(long double base, long double exponent)
float pow(float base, float exponent)
double pow(double base, int exponent)
long double pow(long double base, int expon开发者_开发知识库ent)

I'm not quite shure as to why this builds on our 32-bit environments but that's beside the point right now.

My question is: how should I cast the parameters, which pow should I use? Thanks.

P.S. I can't change the datatype of the parameters as doing so would require too much work. My assignment is to get the code to compile, detailing any hacks I make so that later, we can go over those hacks and find proper ways do deal with them.

---

If you are making many calls to pow(int, unsigned int) why don't you just code it by yourself? If execution speed is not an issue, it's not much work.

Otherwise, I'd use a pow() overload whose input parameters are guaranteed to contain your expected values, such as pow(float, float) or pow(double, double). Anyway, I feel that making your own version could prevent problems with conversion between floating point and integer.

---

The result will always be integer, with these types of arguments.

Depending on the expected range of the arguments, especially the exponent, you should choose for the float or double or long double version.

So that would become

pow( (float) i, (int)ui );

You can find the allowed range of arguments by solving the equation pow(i,ui) < max_double.