Implicit casting of the gcc compiler

[Curt Nowak]

Dear Sir or Madam,

I am a student from the German university of hildesheim 
(http://www.uni-hildesheim.de) and I am trying to find out how the gcc 
compiler handles implicit casting.
For more a better understanding, I would like to know the rules that are 
applied when handling something like this:
double d = 1.4;
float fa = 0.4f;
float fb = 0.4f;

int i = d + fa + fb;  // ignoring the warnings

Is there any documentation available that covers this issue? Or can one 
generally say that gcc will do the casting as late as possible?
Thank you in advance,

Curt Nowak