You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
At first glance that check seems reasonable; negative addresses doesn't exist. But what if my information leak takes addresses relative to some offset like, say, the stack pointer.
I see two obvious solutions: 1) just remove the check or 2) add a flag to the MemLeak constructor where you can specify whether addresses are relative or absolute and then only enforce the check in case of absolute addresses.
Ps. I know that I was the one putting in the check in the first place. I now think that I was wrong.
The text was updated successfully, but these errors were encountered:
See #601 for why this was put in place. If you can make everything work while allowing negative values, sounds good to me. It hadn't ever occurred to me to use MemLeak for relative leaks before 😮
Leaking memory at a negative address with
MemLeak
always returnsNone
because of the check at https://github.com/Gallopsled/pwntools/blob/dev/pwnlib/memleak.py#L156.At first glance that check seems reasonable; negative addresses doesn't exist. But what if my information leak takes addresses relative to some offset like, say, the stack pointer.
I see two obvious solutions: 1) just remove the check or 2) add a flag to the
MemLeak
constructor where you can specify whether addresses are relative or absolute and then only enforce the check in case of absolute addresses.Ps. I know that I was the one putting in the check in the first place. I now think that I was wrong.
The text was updated successfully, but these errors were encountered: