Re: (x * x) % y

Eric Young (eay@uq.net.au)
Fri, 23 Apr 1999 08:33:59 +1000 (GMT+1000)

• Next message: Derek Atkins: "Re: (x * x) % y"
• Previous message: Olivier Langlois: "(x * x) % y"
• In reply to: Mail System Internal Data: "DON'T DELETE THIS MESSAGE -- FOLDER INTERNAL DATA"
• Next in thread: Derek Atkins: "Re: (x * x) % y"

On Thu, 22 Apr 1999, Olivier Langlois wrote:
> unsigned test( unsigned i, unsigned j )
> {
> return (i*i)%j;
> }
> my compiler generate the following assembler code
...
> The problem is that an overflow can occur during the multiplication
> while with this code
...
> My question is : Is it possible to write a legal C construction to would
> generate the correct assembler code ?

The thing is that the correct asm is being generated according to the C
syntax.
 32bit * 32bit -> 32bit
You need to hope that you C compiler supports the long long data type
(_int64 for Visual C) and do some casting.

        return(((unsigned long long)i*i)%j);

should give the correct result.

eric

• Next message: Derek Atkins: "Re: (x * x) % y"
• Previous message: Olivier Langlois: "(x * x) % y"
• In reply to: Mail System Internal Data: "DON'T DELETE THIS MESSAGE -- FOLDER INTERNAL DATA"
• Next in thread: Derek Atkins: "Re: (x * x) % y"