On Wed, 21 Aug 2024 at 20:41, Yafang Shao <laoar.shao@xxxxxxxxx> wrote:
>
> One potential approach could be to rename GFP_NOFAIL to
> GFP_NOFAIL_FOR_SMALL_ALLOC, specifically for smaller allocations, and
> to clear this flag for larger allocations.
Yes, that sounds like a good way to make sure people don't blame the
MM layer when they themselves were the cause of problems.
> However, the challenge lies
> in determining what constitutes a 'small' allocation.
I think we could easily just stick to the historical "order <
PAGE_ALLOC_COSTLY_ORDER":
* PAGE_ALLOC_COSTLY_ORDER is the order at which allocations are deemed
* costly to service.
(And the value for that is 3 - orders 0-2 are considered "cheap")
Linus
