问题
my nasm x86 assembly code contains the following:
; The code should mimic the following C-code:
; int a[10];
; for (int i = 0; i < 10; i++){
; a[i] = i;
; }
SECTION .data
arraylen dd 10
SECTION .bss
array RESD 10
SECTION .text
global main
main:
mov ecx, 0
mov eax, 0
loop:
inc ecx
mov dword [array+eax*4], ecx
inc eax
cmp ecx, arraylen
jl loop
end:
mov ebx, 0
mov eax, 1
int 0x80
Now what i want is to check whether this code works in gdb.
However, how do i print array
?
print array
only returns the $1 = 1
.
print array + X
unfortunately is an arithmetical operation, i.e.
e.g. print array + 50
actually prints 1+50 = 51
and not the non-existent 51st array element.
回答1:
You can do:
(gdb) x/10 &array
0x8049618: 1 2 3 4
0x8049628: 5 6 7 8
0x8049638: 9 10
PS: Your code is broken, you need cmp ecx, [arraylen]
.
回答2:
; The code should mimic the following C-code:
In addition to wrong boundary noted by Jester, you also have the wrong initialization: your code is equivalent to:
for (int i = 0; i < 10; i++) {
a[i] = i + 1; // different from stated goal of "a[i] = i;"
}
However, how do i print
array
?
This is no different from printing array in C
, when the source is compiled without debug info:
(gdb) p array
$1 = 0
(gdb) p {int[10]}&array
$2 = {0, 1, 2, 3, 4, 5, 6, 7, 8, 9}
print array + X unfortunately is an arithmetical operation
You can then use:
(gdb) p $2[4]
$3 = 4
回答3:
ARM example
x86 should be analogous:
.data:
a1:
.float 0.0, 0.1, 0.2, 0.3
a2:
.word 1, 2, 3, 4
.text
/* Register r1 contains the address of a1. */
ldr r1, =a1
ldr r2, =a2
GDB session:
(gdb) p (float[4])a1
$1 = {0, 0.100000001, 0.200000003, 0.300000012}
(gdb) p (int[4])a2
$2 = {1, 2, 3, 4}
(gdb) p (float[4])*$r1
$5 = {0, 0.100000001, 0.200000003, 0.300000012}
(gdb) p (int[4])*$r2
$7 = {1, 2, 3, 4}
Tested on GDB 8.1, Ubuntu 18.04.
来源:https://stackoverflow.com/questions/32300718/printing-array-from-bss-in-gdb