NASM - 循环中的分段错误

Question

i got two similar loops, in which i write an address to eax and then from eax to a data-segment. The first first loop is working, the second loop returns a segfault. Why is the second loop wrong and the first not!?

section .data
n1: db 1
n2: db 1

[...]

n_1_1:
mov eax,one
mov [n1],eax
jmp DIG2

 n_2_1:
 mov eax,one
 mov [n2],eax        ; segfault
 jmp DISP2

Answer 1

db declares a byte (8 bits), which isn't enough to hold eax (32 bits). Declare every variable so that it has enough space to hold the largest value you will try to write to / read from it.

For example:

; declare n1 and n2 as doublewords with initial values of 1
n1: dd 1
n2: dd 1   

---

As for why one of them crashes and the other doesn't; it's hard to say from the code you've shown and not knowing the execution environment. But the second write overwrites 1 byte further into unallocated space (by the looks of it), which could be enough to make it crash even though the first write didn't cause a crash.