Re: [patch] futex-2.5.40-B5

[Linus Torvalds]

In article <[e-mail]>,
Oleg Nesterov  <[e-mail]> wrote:
>Hello.
>
>Ingo Molnar wrote:
>>   the new lookup code first does a lightweight follow_page(), then if no
>>   page is present we do the get_user_pages() thing.
>
>What if futex placed in VM_HUGETLB area?
>Then follow_page() return garbage.
>
>I beleive in i386 case it can be fixed something like this:

This will fix it for i386, but will break it for all other
architectures, as you're now depending on a x86-specific implementation
detail in an arch-independent file using a generic CONFIG test..

>Then follow_hugetlb_page() hook can be killed.

I'd much rather have a hook that works, than having #ifdef's in code
that don't..

		Linus