Re: [PATCH v2] Revert "mm/page_alloc: fix memmap_init_zone pageblock alignment"

[Michal Hocko]

On Thu 15-03-18 15:48:47, Ard Biesheuvel wrote:
> On 15 March 2018 at 15:34, Daniel Vacek <neelx@xxxxxxxxxx> wrote:
[...]
> > Hence I was asking what is the abstract meaning
> > of it. As I see two *way_different* implementations so I am not sure
> > how I should understand that.
> >
> 
> My interpretation is that it has a struct page associated with it, but
> it seems the semantics of pfn_valid() aren't well defined.

Well, pfn_valid says that a given pfn is backed by a real memory and it
has a valid struct page backing it.
-- 
Michal Hocko
SUSE Labs