Discussion: HW-M5

32-bit mode or 64-bit mode?

The homework says that the programming exercises can be completed in 32 or 64 bit mode.  I know we are focused on 32 bit only but is it acceptable to submit 64 bit mode for these programming exercises?

"4.10 Programming Exercises

The following exercises may be completed in either 32-bit mode or 64-bit mode."

The professor said a while ago that no matter what the homework said, you should only focus on writing your code on 32 bit. 

This entire class is in 32-bit mode.

I agree that the hw should be done in 32 bit mode. But also wanted to ask if this means that it is ok to skip the sections in the book that specifically talk about 64 bit mode?

I don't believe we need to worry about the 64-bit content.

32 bit mode is what we are learning so far. We might have some opportunity to learn 64 bit mode at the end.

Yep, 32 bit is what we have been learning and is going to be in 32 - bit mode for the entire class.

I was wondering this as well because of the professor's emphasis on only using 32-bit.

Should be 32 bit mode

Just a note from the class tonight on programming homework submission for:

Only submit .code section in a text file, plus .data section if you have declared any data. The purpose is that the professor can then copy your code and execute it in a non-Microsoft assembler.  Thanks!

I submitted way more than this, but I'll keep it in mind from now on.

Thanks for making note of this, I also have been submitting the entire .asm file, so knowing that we only need to include the .data & .code is better.

Thanks a bunch! Great to know this,

Thank you so much!

Also, and this has been said a million times, you can leave out the main proc, exitprocess, and dumpregs/other irvine stuff.

Thanks for the reminder! Good to know this.

Thank you Amy. Glad to know about this.

It just makes it easier for the professor, as the other parts are not implemented by us and is just part of the template to make sure it compiles.

Thank you, would have totally missed this!

Will we be marked off if we submitted the whole thing already? (for former homeworks)

Thanks, appreciate it! Totally would've missed this.

Fibonacci numbers in Assembly?

how do we implement it?

Here are some examples from stack overflow:

https://stackoverflow.com/questions/32659715/assembly-language-x86-how-to-create-a-loop-to-calculate-fibonacci-sequence

Hi there! Summer gave a nice resource. Here is another useful one: https://chevythunder82.tripod.com/coursework/COMP235/fibonacci.html

Hi! Here is a good resource that also provides other helpful links:https://www.programmersought.com/article/20332992916/ (Links to an external site.)

Hey!  I also found another link that might be helpful!

https://titanwolf.org/Network/Articles/Article?AID=fb26d35b-eba3-4dee-afcc-0e64ac5cad86#gsc.tab=0

I had the same difficulty. So I found a good material while searching.

https://stackoverflow.com/questions/32659715/assembly-language-x86-how-to-create-a-loop-to-calculate-fibonacci-sequence

Thank you for the resources. They are helping me out!

This is what I used:https://blog.seso.io/posts/creating-a-fibonacci-generator-in-assembly/ (Links to an external site.)

Hi Darren,

This is one of the tough questions in my opinion. This is where I went for further materials

https://assembly.happycodings.com/code1.html (Links to an external site.)

I hope this helps.

Converting from Big Endian to Little Endian

In problem 1 of section 4.10, it is my understanding from the question that the order of the bytes in bigEndian should be reversed in littleEndian, i.e, it should be 78h, 56h, 34h, 12h. However, to me it looks like the chapter 4 solutions posted maintains the original order of bigEndian in littleEndian as well. Did I get this wrong? Please explain this to me. Thank you.

The chapter 4 solutions change it to little endian. It looks like it kept the byte ordering as big endian since DumpRegs prints out the bytes in big endian order. Try moving 1 to EAX and calling DumpRegs to see this.

In the book, section 4.5.3 shows how you can see the elements of an array while debugging, which could be helpful to see if the array becomes littleEndian.

Thank you for your replies! This makes a bit more sense. 

Thanks this is very helpful while I am also struggling with HW. 

https://stackoverflow.com/questions/35713272/big-endian-to-little-endian (Links to an external site.)has pretty good explanation of this problem. 

mov al,DWORD PTR bigEndian+0
mov ah,DWORD PTR bigEndian+1
mov bl,DWORD PTR bigEndian+2
mov bh,DWORD PTR bigEndian+3

mov littleEndian+0,bh
mov littleEndian+1,bl
mov littleEndian+2,ah
mov littleEndian+3,al

There is a x86 instruction to do it for you :)

Fibonacci Numbers question

I have some question with the problem 5 of section 4.10. I'm wondering how should I start this program with? Any other method than loop should I use?

This might be a good resource to get started with pseudocode and diagram given.

• What is a pseudocode for fibonnaic series? - Quora (Links to an external site.)

Hello Yen-Jung!

Here is my approach:

The first and second Fibonacci numbers are given: both are 1. We need to find out the next 5 F numbers by summing the previous two Fibonacci numbers.

Declare data array of 7 elements, first two values known;

Declare count = 5, move to ECX. ECX is used the LOOP internally. 

Get the OFFSET of the data array

Move the pointer to the 3rd number (ADD ESI, TYPE array twice)

Inside the loop:

read the two numbers before the current location by using indexed addressing:

MOV eax, [ESI-4] -> move backwards 4 bytes to get the value immediately before current memory location;

MOV eax, [esi -8] - > move backwards 8 bytes,

Add above two values, assign to the memeory address of the current element memory. 

Move the pointer to next in the array. 

Hope this helps and good luck!

These replies are really helpful as I was going to ask this same question. Thanks, everyone!

PTR Using

I think I have to use PTR to get access to the array's item. Is my understanding correct?

For probs1, can I move the bigEndian array's item to bx(bh & bl) temporarily instead of ax(al & ah)?

According tohttps://stackoverflow.com/questions/2864011/accessing-arrays-element-in-assembly-language-windows (Links to an external site.)

I don't think you have to use a PTR 

Problem 3 Summing the Gaps between Array Values

For question 3 what does it mean by the array {0, 2, 5, 9, 10} has gaps of 2, 3, 4, and 1 and whats the reason we need to know this?

When it says gaps, it refers to the difference between consecutive array elements.

I think this question is a practice for us to loop through an array and do some calculations around it.

The gap means the difference between two consecutive array elements, which can be calculated by subtracting the left element by the right element. For example, 2-0=2, 5-2=3, 9-5=4, 10-9=1.

In addition to what Sarah stated, it's simply just an algebra problem that is solved using array functionalities so that's really all you need to know to figure it out.

I'm confused on question 2:

If the input array is values 1,2,3,4,5,6 etc are they saying the output should be 2,1,4,3,6,5? This is also the midterm question so I need to make sure I don't get messed up by the wording lol. 

Hi Aaron,

That is also how I interpreted the question. It says that item i needs to be swapped with item i+1, and then item i+2 with item i+3. So if i is 0, then item 0 is swapped with item 1, and item 2 is swapped with item 3. I believe that you must also assume that the array contains an even number of elements so you do not need to worry about the condition where there's an odd number.

Problem #2 - Exchanging Pairs of Array Values

Can someone further explain the instructions of this problem, I am not quite sure what I am suppose to be doing.

Problem #2 - Exchanging Pairs of Array Values

Can someone further explain the instructions of this problem, I am not quite sure what I am suppose to be doing.

I think all you have to do is just use a loop and loop through moving and exchanging the register values.

Hey Aman, 

I found this on stackoverflow to initiate the question. Hope that helps.

assembly - X86 Fibonacci program - Stack Overflow (Links to an external site.)

Thank you Ajay. I'm stuck in this question too.