On Tue, 14 May 2024 at 20:13, Willy Tarreau <w@xxxxxx> wrote:
>
> Wouldn't it be more effective to have a non-resettable prctl() allowing
> the application to prefer to be killed upon such an munmap() failure in
> order to stay consistent and more robust against such class of attacks?
This whole argument is based on a castle of sand, and some notion that
this is a problem in the first place.
Guys, if you let untrusted code execute random system calls, the whole
"look, now unmap() acts oddly" IS THE LEAST OF YOUR ISSUES.
This whole "problem" is made-up. It's not real. Theo is literally
upset about something that Linux has done forever, and that has never
been an issue.
Stop inventing make-believe problems - there are enough *real* bugs
people can look at that you really don't need to.
Linus
