> The reason is that these function pointers are only accessed through an indirection layer, i.e., memory objects on the heap contain a pointer to a virtual table which is located in the code or data section of the application and contains a number of pointers to virtual functions. Since the attackers can only disclose the virtual table pointer, but not the virtual table itself, as it is not on the heap, they cannot disclose gadget addresses. Note that, when only ASLR is applied, the address of the virtual table is randomized with the same offset as the ROP gadgets. Therefore, such an attack can bypass ASLR but not selfrando.

> We therefore conclude that selfrando can thwart most real-world exploits. Attackers can only succeed in rare cases where they can disclose the complete heap and data section.

[1] : https://people.torproject.org/~gk/misc/Selfrando-Tor-Browser...