A C function which accepts a large integer type?

Project Infrastructure LLVM Dev List Archives
1

Is there anyway in C I can write a function which accepts a very large
integer? That is, say in LLVM I'm using a 256bit integer, is it possible
to write a C function which accepts that value?

I understand it wouldn't be standard C, or may not even be pretty, but
that doesn't matter.

2

edA-qa mort-ora-y <eda-qa@disemia.com> writes:

Is there anyway in C I can write a function which accepts a very large
integer? That is, say in LLVM I'm using a 256bit integer, is it possible
to write a C function which accepts that value?

I understand it wouldn't be standard C, or may not even be pretty, but
that doesn't matter.

What's wrong with passing the address of the integer?

3

Yes, that idea came to me also shortly after pressing "Send". :wink:

I assume the data is simply ordered like a native integer. So on
little-endian it is lowest byte first.

4

Hi edA-qa mort-ora-y,

5

Actually I would also like to know that case (as I support odd sizes).
If this is too complicated though I would just extend them to the next
power of 2 before calling the function.

6

IIRC, they are extended to the nearest multiple of 8 bits by adding bits with
an unspecified value, then the extended integer is stored.

Ciao, Duncan.