I am looking at the assembly language code of a switch statement.
I understand how the code works and what the cases are. My question is how do I decide on the case names?
Below is the assembly language code, which will be followed with my interpretation of it. I basically just need to use the jump table and fill in the case names.
1 8048420: push %ebp
2 8048421: mov %esp, $ebp
3 8048423: mov 0x8(%ebp), %eax // x
4 8048426: mov 0xc(%ebp), %edx // n
5 8048429: sub $0x32, %edx // so least value of case is 32
6 804842c: cmp $0x5, %edx // max value is 37
7 804842f: ja 8048448 <switch+0x28> // if >37, go to default
8 8048431: jmp *0x80485d0(, %edx, 4) //THIS RIGHT HERE ?
9 8048438: shl $0x2, %eax // CASE A
10 804843b: jmp 804844b <switch+0x2b> //break;
11 804843d: sar $0x2, %eax //CASE B
12 8048440: jmp 804844b <switch+0x2b> //break
13 8048442: lea (%eax, %eax, 2), %eax //CASE C
14 8048445: imul %eax, %eax
15 8048448: add $0xa, %eax //fall through to default
16 804844b: pop %ebp //return
17 804844c: ret
The jump table that the gdb command creates: I am doing x/6w 0x80485d0
0x80485d0: 0x08048438 0x08048448 0x08048438 0x0804843d
0x80485e0: 0x08048442 0x08048445
My interpretation:
int result = x;
switch(n) {
case __:
x = x << 2;
break;
case __:
x = x >> 2
break;
case __:
x = 4*x;
x = x*x
case __: //default
x += 0xa
return x;
}
I just don't understand how to look up the jump table and decide which values of n between 32 and 37 fit in which of the case blanks.
Any help would be appreciated. Thank you.
As Oli says, there's not much more to do. n-50
is stored in %edx, then switch+0x11 jumps to the address stored in 0x80485d0 + %edx * 4
. Looking at the table, that's switch+0x18 when n==50 or 52, switch+0x28 when n==51, switch+0x1d when n==53, switch+0x22 when n==54 and switch+0x25 when n==55.