As long as there's no floating point arithmetic, 32 bit integer
types are computationally equivalent to 32 bit floating point.

Change-Id: I5ab19da440f52617a9bc64dff6fcaf8eff796aeb
Reviewed-on: https://swiftshader-review.googlesource.com/4717Tested-by: Alexis Hétu <sugoi@google.com>
Reviewed-by: Nicolas Capens <capn@google.com>