Linus Torvalds writes: (Summary) wrote:
But when we register the rseq, we could easily check that the top bits of the IP is clear, no?
of the IP is clear, no?
Sure, user space can change it after the fact, but at that point it's literally "user space is being intentionally stupid". The real worry is that 32-bit compat code never initializes those bits at all, no?
at all, no?
Linus
Linus
Linus
[...]
interrupt/trap after a preemption.But when we register the rseq, we could easily check that the top bits of the IP is clear, no?
of the IP is clear, no?
Sure, user space can change it after the fact, but at that point it's literally "user space is being intentionally stupid". The real worry is that 32-bit compat code never initializes those bits at all, no?
at all, no?
Linus
Linus
Linus