Re: [PATCH v2 for v5.9] mm/page_alloc: handle a missing case for memalloc_nocma_{save/restore} APIs

[Joonsoo Kim <js1304@gmail.com>]

2020년 9월 29일 (화) 오후 5:08, Michal Hocko <mhocko@suse.com>님이 작성:
>
> On Mon 28-09-20 17:50:46, Joonsoo Kim wrote:
> > From: Joonsoo Kim <iamjoonsoo.kim@lge.com>
> >
> > memalloc_nocma_{save/restore} APIs can be used to skip page allocation
> > on CMA area, but, there is a missing case and the page on CMA area could
> > be allocated even if APIs are used. This patch handles this case to fix
> > the potential issue.
> >
> > Missing case is an allocation from the pcplist. MIGRATE_MOVABLE pcplist
> > could have the pages on CMA area so we need to skip it if ALLOC_CMA isn't
> > specified.
> >
> > Fixes: 8510e69c8efe (mm/page_alloc: fix memalloc_nocma_{save/restore} APIs)
> > Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
> > ---
> >  mm/page_alloc.c | 13 ++++++++++---
> >  1 file changed, 10 insertions(+), 3 deletions(-)
> >
> > diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> > index fab5e97..104d2e1 100644
> > --- a/mm/page_alloc.c
> > +++ b/mm/page_alloc.c
> > @@ -3367,9 +3367,16 @@ struct page *rmqueue(struct zone *preferred_zone,
> >       struct page *page;
> >
> >       if (likely(order == 0)) {
> > -             page = rmqueue_pcplist(preferred_zone, zone, gfp_flags,
> > +             /*
> > +              * MIGRATE_MOVABLE pcplist could have the pages on CMA area and
> > +              * we need to skip it when CMA area isn't allowed.
> > +              */
> > +             if (!IS_ENABLED(CONFIG_CMA) || alloc_flags & ALLOC_CMA ||
> > +                             migratetype != MIGRATE_MOVABLE) {
> > +                     page = rmqueue_pcplist(preferred_zone, zone, gfp_flags,
> >                                       migratetype, alloc_flags);
> > -             goto out;
> > +                     goto out;
> > +             }
> >       }
>
> This approach looks definitely better than the previous version.

Thanks!

> >
> >       /*
> > @@ -3381,7 +3388,7 @@ struct page *rmqueue(struct zone *preferred_zone,
> >
> >       do {
> >               page = NULL;
> > -             if (alloc_flags & ALLOC_HARDER) {
> > +             if (order > 0 && alloc_flags & ALLOC_HARDER) {
> >                       page = __rmqueue_smallest(zone, order, MIGRATE_HIGHATOMIC);
> >                       if (page)
> >                               trace_mm_page_alloc_zone_locked(page, order, migratetype);
>
> But this condition is not clear to me. __rmqueue_smallest doesn't access
> pcp lists. Maybe I have missed the point in the original discussion but
> this deserves a comment at least.

Before the pcplist skipping is applied, order-0 request can not reach here.
But, now, an order-0 request can reach here. Free memory on
MIGRATE_HIGHATOMIC is reserved for high-order atomic allocation
so an order-0 request should skip it.

I will add a code comment on the next version.

Thanks.