GDB Buffer Overflow Analysis

(Simon Bohlander) #1

Hello Hackers,

I was trying myself at the Gnu Debugger (GDB) analysis while trying to execute a buffer overflow on a very easy C program. See: YouTube Buffer Overflow in C

On a VM and on a normal PC on Debian, I get 0x41414141 at Buffer Overflow (print ‘A’*264 on char x[256]). But on Bananian RPi 1 I get 0x41414140. Can you explain that please?

Yours Simon

(David Coles-Dobay) #2

Let me know if you find this. I saw something similar looking at the buffer under tests in PHP compile. I am only guessing at this point but the math for 64Long to 32Long is wrong for the A83T it may be related. I will be going over the libraries over the next few weeks to find the correct one for this chipset. One thing is sure the current kernel source is lacking a bunch of commits from the chip designer. Scanning the sunnxi email archives I have found a few instances where the chip manufacturer tried to get the code for this chip included in the repository and it was rebuked.