Re: [PATCH] mm: limit direct reclaim for higher order allocations

[Rik van Riel]

On Wed, 2016-02-24 at 14:15 -0800, David Rientjes wrote:
> On Wed, 24 Feb 2016, Rik van Riel wrote:
> 
> > For multi page allocations smaller than PAGE_ALLOC_COSTLY_ORDER,
> > the kernel will do direct reclaim if compaction failed for any
> > reason. This worked fine when Linux systems had 128MB RAM, but
> > on my 24GB system I frequently see higher order allocations
> > free up over 3GB of memory, pushing all kinds of things into
> > swap, and slowing down applications.
> >Â
> 
> Just curious, are these higher order allocations typically done by
> theÂ
> slub allocator or where are they coming from?

These are slab allocator ones, indeed.

The allocations seem to be order 2 and 3, mostly
on behalf of the inode cache and alloc_skb.

> > It would be much better to limit the amount of reclaim done,
> > rather than cause excessive pageout activity.
> >Â
> > When enough memory is free to do compaction for the highest order
> > allocation possible, bail out of the direct page reclaim code.
> >Â
> > On smaller systems, this may be enough to obtain contiguous
> > free memory areas to satisfy small allocations, continuing our
> > strategy of relying on luck occasionally. On larger systems,
> > relying on luck like that has not been working for years.
> >Â
> > Signed-off-by: Rik van Riel <riel@xxxxxxxxxx>
> > ---
> >Â mm/vmscan.c | 19 ++++++++-----------
> >Â 1 file changed, 8 insertions(+), 11 deletions(-)
> >Â
> > diff --git a/mm/vmscan.c b/mm/vmscan.c
> > index fc62546096f9..8dd15d514761 100644
> > --- a/mm/vmscan.c
> > +++ b/mm/vmscan.c
> > @@ -2584,20 +2584,17 @@ static bool shrink_zones(struct zonelist
> *zonelist, struct scan_control *sc)
> >ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂcontinue;ÂÂÂÂÂÂÂ/* Let kswapd poll it
> */
> >ÂÂ
> >ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ/*
> > -ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ * If we already have plenty of memory free
> for
> > -ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ * compaction in this zone, don't free any
> more.
> > -ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ * Even though compaction is invoked for any
> > -ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ * non-zero order, only frequent costly order
> > -ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ * reclamation is disruptive enough to become
> a
> > -ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ * noticeable problem, like transparent huge
> > -ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ * page allocations.
> > +ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ * For higher order allocations, free enough
> memory
> > +ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ * to be able to do compaction for the
> largest possible
> > +ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ * allocation. On smaller systems, this may
> be enough
> > +ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ * that smaller allocations can skip
> compaction, if
> > +ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ * enough adjacent pages get freed.
> >ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ */
> > -ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂif (IS_ENABLED(CONFIG_COMPACTION) &&
> > -ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ sc->order > PAGE_ALLOC_COSTLY_ORDER &&
> > +ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂif (IS_ENABLED(CONFIG_COMPACTION) && sc-
> >order &&
> >ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ zonelist_zone_idx(z) <= requested_highidx
> &&
> > -ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ compaction_ready(zone, sc->order)) {
> > +ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ compaction_ready(zone, MAX_ORDER)) {
> >ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂsc->compaction_ready = true;
> > -ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂcontinue;
> > +ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂreturn true;
> >ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ}
> >ÂÂ
> >ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ/*
> >Â
-- 
All Rights Reversed.

Attachment: signature.asc
Description: This is a digitally signed message part