On 10/30/06, Joco Salvatti <[EMAIL PROTECTED]> wrote:
Hi all,
Reading Theo de Raadt's presentation about exploit mitigation
techniques, I could not understand the advantages of using randomized
malloc and randomized mmap techniques. I've searched on the Internet
about this subject but I couldn't find a thing, maybe because this
subject is too technical.
I'm sorry for the lack of abilities to understand the presentation.
Could anyone, please, give me an example or point me any website in
which I could find informations in order to understand why randomized
malloc and randomized mmap are more secure than the traditional ones?
Thanks in advance for the time wasted reading this e-mail.
If memory locations are predictable and you can access program memory
arbitrarily (e.g. kernel mode or a bug in something) then you can
easily calculate where to find certain pieces of data. This is bad. It
lets you read private data or maybe even write it.
At least, I think that's about right, right?
-Nick
