Question: "backwards" adressing

[Ulf Zibis]

Hi Christian,

I tried a test (see below).

Now I'm wondering, that loop1 is slightly faster than loop2, as ...
loop1 has:
    1. load constant
    2. add to offset
    3. store indirected 2nd constant
loop2 has:
    1. increment offset
    2. store indirected 2nd constant

Do you know any reason for this ?



/**
 *
 * @author Ulf Zibis <Ulf.Zibis at CoSoCo.de>
 */
public class IncRandomSpeed {

    static void loop1(int off, char in1, char in2, char[] out) {
        out[off+3] = in1;
        out[off+5] = in2;
        out[off+0] = in1;
        out[off+4] = in2;
        out[off+9] = in1;
        out[off+8] = in2;
        out[off+6] = in1;
        out[off+1] = in2;
        out[off+7] = in1;
        out[off+2] = in2;
    }

    static void loop2(int off, char in1, char in2, char[] out) {
        out[off++] = in1;
        out[off++] = in2;
        out[off++] = in1;
        out[off++] = in2;
        out[off++] = in1;
        out[off++] = in2;
        out[off++] = in1;
        out[off++] = in2;
        out[off++] = in1;
        out[off++] = in2;
    }

    static final int SIZE = 1000;
    static char[] out = new char[SIZE+10];

    /**
     * @param args the command line arguments
     */
    public static void main(String[] args) {
        long time = System.nanoTime();
        long time1 = 0;
        long time2 = 0;
        for (char i=0; i<100; i++)
            for (char j=0; j<1000; j++) {
                time = System.nanoTime();
                for (int k=0; k<SIZE; k++)
                    loop1(k, i, j, out);
                time1 += System.nanoTime() - time;
                time = System.nanoTime();
                for (int k=0; k<SIZE; k++)
                    loop2(k, i, j, out);
                time2 += System.nanoTime() - time;
            }
        System.out.println("time1: "+time1);
        System.out.println("time2: "+time2);
    }

}



Am 24.08.2009 17:53, Christian Thalinger schrieb:
>
> Well, this was only pseudo-code, as was yours.  Let's get a little deeper...
>
> base+index is calculated only once into a temporary register and used
> afterwards.  But if that happens depends on the register allocator and
> register pressure.
>   

Hm, that's what I thought as first, so in forward case cpu should do:
   1. base+index -> index register
   2. move highSurrogate(accumulator), [index register]
   3. increment index register
   4. move lowSurrogate(accumulator), [index register]

Backward case cpu should do:
   1. base+index -> index register
   2. move -> temp
   3. increment index register
   4. move lowSurrogate(accumulator), [index register]
   5. load temp -> index register
   6. move highSurrogate(accumulator), [index register]

Please excuse, that I'm insisting, but I really don't understand why 
both should run in same time.
Can you explain once more?

-Ulf


>   
>>  > a register machine here, not a stack machine
>> I'm too wondering, how method parameters were passed by register, 
>> especially if they were more.
>>     
>
> On IA32?  The first two integer arguments are passed in registers, the
> rest on the stack, on SPARC it's different.  But again, we are only
> talking in pseudo-code here.
>   

Yes, of course.