Re: Algoritmic Complexity Attacks and 2.4.20 the dcache code

[Scott A Crosby <scrosby@cs.rice.edu>]

On Fri, 30 May 2003 02:00:40 -0700 (PDT), "David S. Miller" <davem@redhat.com> writes:

>    From: Alex Riesen <alexander.riesen@synopsys.COM>
>    Date: Fri, 30 May 2003 10:59:01 +0200
> 
>        static
>        int hash_3(int hi, int c)
>        {
>    	return (hi + (c << 4) + (c >> 4)) * 11;
>        }
>    
>    gcc-3.2.1 -O2 -march=pentium
>  ...   
>    It is not guaranteed to be this way on all architectures, of course.
>    But still - no multiplications.
> 
> Indeed, I'd missed this.  GCC will emit the constant multiply
> expansion unless the multiply cost is set VERY low.

It may still be a win. This does a bit under a dozen instructions per
byte. However, jenkin's does many bytes at a time.

Scott