I am working on a project that relies on the standard C datatype uint64_t to multiply 32 bit integers. But I am concerned about portability issues, as I am already re-implementing many parts of my source code in C due to portability reasons. I realize that it is difficult to provide an exhaustive list, but due to lack of hardware resources, available to me, I know the specs of my systems (PC (Windows, x86) / Android phones based on ARM etc. ) only. My target project, needs to run on a wide variety of platforms like Arduino, embedded systems etc. And I was wondering whether I should replace, uint64_t with an alternative like two uint32_tvariables, to store the high and low bits of my 64 bit number...
Sorry, for this direct question. But I intend to strike up a detailed discussion of implementing my multiplication faster and in constant time.
Thankyou very much in advance!
To deal with some 32 bit systems, that approach can make sense.