my bubblesort for 100 ints

Hi I have been trying to implent this sorting algorithm for 100 ints.. I used the bubble sort one as well as quicksort...
I have spent nearly two weeks trying to code quicksort and gladly changed to bubble seen that qsort nearly gave me a nervous breakdown!

bubble sort is this:
Bubblesort (int data[],int n) {
   int tmp,i,j;

   for (i=0; i<n-1; i++) {
       for (j=0; j<n-i-1; j++)
            if (data[j] > data[j+1]) {
               tmp = data[j];
               data[j] = data[j+1];
               data[j+1] = tmp;

and here is my implementation in bubblesort........

array:      .word 0x11,0x111,0x1111111,0x11111,0x1,0x111111111,0x111111,0x111111111,0x111,0x11,0x1111,0x11111111,0x111,0x111111,0x111,0x11111111,0x111111111,0,0x1,0x1111111


      dadd r1, r0, r0  ;i
      dadd r2, r0, r0  ;j
      daddi r6, r2, 8  ;j+1
      ld r3, array(r0)  
      dsll r3, r3, 3  ;n
      daddi r7, r3, -8   ;n-1
label:      slt r12 ,r1, r7  ;check i< n-1
      beqz r12, done ;if not halt
      dsub r13, r3, r1  ; else set up for j loop....(ie) n-i
      daddi r14, r13, -8  ; n-i-1
here:      slt r15, r2, r14  ; if !r2 < n-i-1
      beqz r15,inci    ; go to next i
      ld r19, array(r2)  ; else do the swap
      ld r20, array(r6)
      slt r21, r19, r20
      bnez r21, incj
      ld r18, array(r6)
      ld r16, array(r2)
      sd r16, array(r6)
      sd r18, array(r2)
incj:      daddi r2, r2, 8   ; next iteration of j
      daddi r6, r2, 8
      j here
inci:      dadd r2, r0, r0  ; inc i and reset j and j+ 1
      daddi r1, r1, 8
      daddi r6, r2, 8
        j label ; go again
done:   halt

the thing is i have an application from that shows the stats of the  reg and mem etc.. a kind of simulator!   can be downloaded from here   installs in seconds
but this sort is taking way to many clock cycles maybe it can be optimized better also it will not sort the last four of the array...please help
Who is Participating?
I wear a lot of hats...

"The solutions and answers provided on Experts Exchange have been extremely helpful to me over the last few years. I wear a lot of hats - Developer, Database Administrator, Help Desk, etc., so I know a lot of things but not a lot about one thing. Experts Exchange gives me answers from people who do know a lot about one thing, in a easy to use platform." -Todd S.

Hi pablo23,

 >>  for (i=0; i<n-1; i++) {
You dont need to stop i at n-1 since j is stopping at n-i-1. Use:
   for (i=0; i<n; i++) {

I'll take a look at the rest in a bit.

Hi pablo23,

I'm not familliar with this language so forgive me if these are not of use.

There's some optimisations you can do:

1. You use -8 and 8 as values. Have you got two more registers you can use to hold them?
2. You load both array elements into registers before comparison. Can you compare a register with a memory location?
3. ld r18, array(r6). Surely, that value is in r20 already.

Are there any instructions to read a number of values into sequential registers in one hit?

pablo23Author Commented:
No probs Paul-. Maybe I will try that with the r20 op (use a dadd)instead of ld, but the cc is still very poor and it won't sort the last five data regs.... and thats when its just sort 20 ints as you can see. Its up around 2000 and more! I want it down well past 1000 ;(!
The problem of not sorting the last few should be covered by my first comment. Have you tried that?

You cannot judge the algorithms efficiency on one set of data. Bubblesort is least efficient if the data is initially in reverse order and most efficient if the data is already sorted. You must test with random data and average the results.

There will also be a minimum number of cycles for any sort, and that is to compare every item with at least one other. How many cycles would that take?


Experts Exchange Solution brought to you by

Your issues matter to us.

Facing a tech roadblock? Get the help and guidance you need from experienced professionals who care. Ask your question anytime, anywhere, with no hassle.

Start your 7-day free trial
Bubblesort is one of the worst sorting algorithms it has an order of magnitude of n^2 for large n.

If you have an ASM quicksort algorithm, I'd spend time working out the bugs in that as the order of magnitude for a properly written quick sort algorithm becomes closer to n log n.

It's more than this solution.Get answers and train to solve all your tech problems - anytime, anywhere.Try it for free Edge Out The Competitionfor your dream job with proven skills and certifications.Get started today Stand Outas the employee with proven skills.Start learning today for free Move Your Career Forwardwith certification training in the latest technologies.Start your trial today

From novice to tech pro — start learning today.

Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.