Hi all,
How can I declear a 64-bit integer in linux and unix enviroment? I tried to ues long long on redhat linux but did not work

64bit integer in Linux?

What language are we talking here? bash script, c, c++, python...

i dont think there is such a thing. ints are 32-bits. However, when programming with MCUs, we would use extra variables, take 2 8-bit registers to make the equiv. of a 16-bit register. But that's low level stuff, the same could be done with char's if you wanted to use 8-bit chunks and make then unsigned (0-255).

why do you need a 64-bit int by the way.

You should be able to use "long long" with gcc, in what way did it not work?

Alex

theres some sort of bug in gcc 3.3, but it works fine for me in gcc 3.2

there is such a thing as a 64-bit int. I was downloading the latest version of BIND and found this:



there is porbably a typedef for it already pre-defined somewhere (if you are using C).

I think 64 bit integer is the integer type of the 64 bit compiler for the 64 bit machines.
64 bit processors have 64 bit registers.
So I think, if you are working on a 64 bit machine, if you declare an integer, it's 64 bit automatically.

Well, I don't know if this will help you or not, but I've got a "huge_int" class that I wrote in C++. It can represent an arbitrary length integer. You specify how many bits the register will use, and it behaves like a normal integer from that point on. I used overloading operators to allow addition by unsigned integers and string representations of integers. In fact, that's how the class is implemented: a string of 1's and 0's. It allows for display in decimal, hex, or binary formats. Most math operators are implemented: addition, subtraction, multiplication, division, +=, -=, etc. I can dig it up if anyone is interested.

Note: I never got around to making it a signed integer. So if you need negative values, then you'll have to modify the class, or add the appropriate twos-complement adjustment when it's displayed.

And yes, traditionally, the base size of an integer is the width of a processor's internal registers. So 32-bit machines have 32-bit integers, a 64-bit has 64-bit integers, etc.

I don't think this is true for the (quasi) 64-bit processors of the x86 architecture, you have to use "long long" to get a 64-bit int. It would be helpful if the original poster would say exactly what problem he had with this solution in Redhat.

Alex

This might be out of the question for your situation, but you may want to look into the gmp library.... you can go as big (or small) as you want

... just a thought

Enjoy.