Re: [PATCH 4/4] mm: numa: Slow PTE scan rate if migration failures occur

[Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>]

On Sun, Mar 8, 2015 at 11:35 AM, Linus Torvalds
<torvalds@xxxxxxxxxxxxxxxxxxxx> wrote:
>> As a second hack (not to be applied), could we change:
>>
>>  #define _PAGE_BIT_PROTNONE      _PAGE_BIT_GLOBAL
>>
>> to:
>>
>>  #define _PAGE_BIT_PROTNONE      (_PAGE_BIT_GLOBAL+1)
>>
>> to double check that the position of the bit does not matter?
>
> Agreed. We should definitely try that.

There's a second reason to do that, actually: the __supported_pte_mask
thing, _and_ the pageattr stuff in __split_large_page() etc play games
with _PAGE_GLOBAL. As does drivers/lguest for some reason.

So looking at this all, there's a lot of room for confusion with _PAGE_GLOBAL.

That kind of confusion would certainly explain the whole "the changes
_look_ like they do the same thing, but don't" - because of silly
semantic conflicts with PROTNONE vs GLOBAL.

                                Linus