Re: [PATCH RFC 1/2] mm, hugetlb: unify core page allocation accounting and initialization

2017-11-29 Thread Mike Kravetz
On 11/28/2017 10:57 PM, Michal Hocko wrote: > On Tue 28-11-17 13:34:53, Mike Kravetz wrote: >> On 11/28/2017 06:12 AM, Michal Hocko wrote: > [...] >>> +/* >>> + * Allocates a fresh page to the hugetlb allocator pool in the node >>> interleaved >>> + * manner. >>> + */ >>> static int

Re: [PATCH RFC 1/2] mm, hugetlb: unify core page allocation accounting and initialization

2017-11-29 Thread Mike Kravetz
On 11/28/2017 10:57 PM, Michal Hocko wrote: > On Tue 28-11-17 13:34:53, Mike Kravetz wrote: >> On 11/28/2017 06:12 AM, Michal Hocko wrote: > [...] >>> +/* >>> + * Allocates a fresh page to the hugetlb allocator pool in the node >>> interleaved >>> + * manner. >>> + */ >>> static int

Re: [PATCH RFC 1/2] mm, hugetlb: unify core page allocation accounting and initialization

2017-11-28 Thread Michal Hocko
On Tue 28-11-17 13:34:53, Mike Kravetz wrote: > On 11/28/2017 06:12 AM, Michal Hocko wrote: [...] > > +/* > > + * Allocates a fresh page to the hugetlb allocator pool in the node > > interleaved > > + * manner. > > + */ > > static int alloc_fresh_huge_page(struct hstate *h, nodemask_t > >

Re: [PATCH RFC 1/2] mm, hugetlb: unify core page allocation accounting and initialization

2017-11-28 Thread Michal Hocko
On Tue 28-11-17 13:34:53, Mike Kravetz wrote: > On 11/28/2017 06:12 AM, Michal Hocko wrote: [...] > > +/* > > + * Allocates a fresh page to the hugetlb allocator pool in the node > > interleaved > > + * manner. > > + */ > > static int alloc_fresh_huge_page(struct hstate *h, nodemask_t > >

Re: [PATCH RFC 1/2] mm, hugetlb: unify core page allocation accounting and initialization

2017-11-28 Thread Mike Kravetz
On 11/28/2017 06:12 AM, Michal Hocko wrote: > From: Michal Hocko > > hugetlb allocator has two entry points to the page allocator > - alloc_fresh_huge_page_node > - __hugetlb_alloc_buddy_huge_page > > The two differ very subtly in two aspects. The first one doesn't care > about

Re: [PATCH RFC 1/2] mm, hugetlb: unify core page allocation accounting and initialization

2017-11-28 Thread Mike Kravetz
On 11/28/2017 06:12 AM, Michal Hocko wrote: > From: Michal Hocko > > hugetlb allocator has two entry points to the page allocator > - alloc_fresh_huge_page_node > - __hugetlb_alloc_buddy_huge_page > > The two differ very subtly in two aspects. The first one doesn't care > about HTLB_BUDDY_*

[PATCH RFC 1/2] mm, hugetlb: unify core page allocation accounting and initialization

2017-11-28 Thread Michal Hocko
From: Michal Hocko hugetlb allocator has two entry points to the page allocator - alloc_fresh_huge_page_node - __hugetlb_alloc_buddy_huge_page The two differ very subtly in two aspects. The first one doesn't care about HTLB_BUDDY_* stats and it doesn't initialize the huge page.

[PATCH RFC 1/2] mm, hugetlb: unify core page allocation accounting and initialization

2017-11-28 Thread Michal Hocko
From: Michal Hocko hugetlb allocator has two entry points to the page allocator - alloc_fresh_huge_page_node - __hugetlb_alloc_buddy_huge_page The two differ very subtly in two aspects. The first one doesn't care about HTLB_BUDDY_* stats and it doesn't initialize the huge page.