Re: [patch 1/2] x86_64 page fault NMI-safe

[Linus Torvalds]

On Thu, Jul 15, 2010 at 11:31 AM, Mathieu Desnoyers
<mathieu.desnoyers@xxxxxxxxxxxx> wrote:
>
> Hrm, we could probably get away with only keeping the nmi_stack_nested per-cpu
> variable. The nmi_stack_ptr could be known statically if we set it at a fixed
> offset from the bottom of stack rather than using an offset relative to the top
> (which can change depending if we are nested over the kernel or userspace).
> We just have to reserve enough space for the bottom of stack.

I thought about trying that, but I don't think it's true. At least not
for the 32-bit case.

The thing is, the 32-bit case will re-use the kernel stack if it
happens in kernel space, and will thus start from a random space (and
won't push all the information anyway). So a nested NMI really doesn't
know where the original NMI stack is to be found unless we save it
off.

In the case of x86-64, I think the stack will always be at a fixed
address, and push a fixed amount of data (because we use the IST
thing). So there you could probably just use the flag, but you'd still
have to handle the 32-bit case, and quite frankly, I think it would be
much nicer if the logic could be shared for the 32-bit and 64-bit
cases.

But maybe I'm missing something.

             Linus
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/