Re: unsigned int multiply, x86-64

[Andrew Haley <aph at redhat dot com>]

Bob Plantz wrote:
> I wrote some C code to multiply two unsigned (32-bit) ints and store the
> result in a 64-bit unsigned int.
> 
> int main(void)
> {
>    unsigned int x, y;
>    unsigned long int z;
> 
>    printf("Enter two integers: ");
>    scanf("%u %u", &x, &y);
> 
>    z = (long int)(x * y);
>    
>    printf("%u * %u = %lu\n", x , y, z);
> 
>    exit(0);
> }
> 
> For the multiply, gcc produces (with -O0):
> 	movl	-4(%rbp), %edx
> 	movl	-8(%rbp), %eax
> 	imull	%edx, %eax
> 	mov	%eax, %eax
> 	movq	%rax, -16(%rbp)
> 
> First, it uses signed multiply (imull). Second, it only keeps the
> low-order 32 bits of the result.
> 
> I would do something like:
>         movl     -4(%rbp), %edx
>         movl     -8(%rbp), %eax   # zeros high-order 32 bits
>         mull     %edx                       # 64-bit result in edx:eax
>         shlq     $32, %rdx             # shift to high-order
>         addq     %rdx, %rax           # combine into 64-bit result
>         movq     %rax, -16(%rbp)
> 
> Am I missing something here?

Knowledge of C, I think.

try

   z = (long int)x * y;

Andrew.