Memory refrencing Assembly Language

1
votes

i am a beginner in assembly language i understand many things till now but for many days i stuck in one section it's confusing to me and what make it worst that through the many pages i visited over internet the information is twisted again ,this section is the Addressing Modes specifcally the memory refrence so for example in this Addressing Modes in Assembly Language (IA-32 NASM)

two instructions seems same but the comment is diffrent

  1. mov esi, var: Copies address of var (say 0x0040120e) into esi
  2. mov dword [eax], var1: copies value in var1 into the memory slot
    specified by eax

both instructions deals with var or var1 which are labels but in comments for one it's mentioned it's copy the content ,while for the other it's copy the address.

and so another question i ask what will be the cooment for this instruction:

  • mov bx, [var]

and what is the diffrence between these instructions also

so i need to know what is difference and is there is a confidence source that i can read from and be sure it's correct,by the way i'am using this tutorial,and NASM as assembler under windows.

1
assemblyx86nasmaddressing-mode

1 Answers

1
votes

mov dword [eax], var1: copies value in var1 into the memory slot specified by eax

Nope. If var1 is a label, this will store the address of var1 as a 32-bit value into the memory pointed to by eax. For example, if the address of var1 is 0x04000000 and eax contains 0x12345678 then this would write 0x04000000 to memory at 0x12345678.

what will be the comment for this instruction: mov bx, [var] ?

; Move the word located at var into bx

(note: "move" in assembly really means copy)

mov [var], bx is this only applicaple if var is an array? if not what it differ from mov var, bx?

Types aren't really enforced by the assembler. You can store anything anywhere (well, anywhere that your program is allowed to write to). So what you declared var as is largely irrelevant. There could be code located at var for all the assembler cares (that would probably not be such a good idea unless you really know what you're doing).

The difference to mov var,bx is that mov var,bx isn't a valid instruction. The address of var is an immediate, so this would be kind of like saying mov 5, bx, which obviously doesn't work.

lea eax, [var] — the value in var is placed in EAX

No. The address of var is placed in eax. LEA means Load Effective Address. You give it a memory operand the same way as if you were going to access memory, but instead of the value at that address you get the address itself.

LEA can also be used to perform some simple arithmetic. For example, lea ebx,[eax*4 + eax] would calculate eax*5 and put the result in ebx.

LEA EBX, [MY_TABLE] here its mean effective address

Yes. There's no difference between this and the previous example.