Speed difference between using int and unsigned int when mixed with doubles

Question

I have an application where part of the inner loop was basically:

double sum = 0;
for (int i = 0; i != N; ++i, ++data, ++x) sum += *data * x;

Answer 1

Here's some code produced by VC++ 6.0 - no optimisation:

4:        int x = 12345;
0040E6D8   mov         dword ptr [ebp-4],3039h
5:        double d1 = x;
0040E6DF   fild        dword ptr [ebp-4]
0040E6E2   fstp        qword ptr [ebp-0Ch]
6:        unsigned int y = 12345;
0040E6E5   mov         dword ptr [ebp-10h],3039h
7:        double d2 = y;
0040E6EC   mov         eax,dword ptr [ebp-10h]
0040E6EF   mov         dword ptr [ebp-20h],eax
0040E6F2   mov         dword ptr [ebp-1Ch],0
0040E6F9   fild        qword ptr [ebp-20h]
0040E6FC   fstp        qword ptr [ebp-18h]

As you can see, converting the unsigned does quite a bit more work.