On Thu, Jul 22, 2010 at 09:08:43AM +0800, Minchan Kim wrote:
> On Thu, Jul 22, 2010 at 9:27 AM, Shaohua Li <shaohua.li@xxxxxxxxx> wrote:
> >> > But we did see some strange regression. The regression is small (usually < 2%)
> >> > and most are from multithread test and none heavily use activate_page(). For
> >> > example, in the same system, we create 64 threads. Each thread creates a private
> >> > mmap region and does read access. We measure the total time and saw about 2%
> >> > regression. But in such workload, 99% time is on page fault and activate_page()
> >> > takes no time. Very strange, we haven't a good explanation for this so far,
> >> > hopefully somebody can share a hint.
> >>
> >> Mabye it might be due to lru_add_drain.
> >> You are adding cost in lru_add_drain and it is called several place.
> >> So if we can't get the gain in there, it could make a bit of regression.
> >> I might be wrong and it's a just my guessing.
> > The workload with regression doesn't invoke too many activate_page, so
> > basically activate_page_drain_cpu() is a nop, it should not take too much.
>
> I think it's culprit. little call activate_page, many call lru_drain_all.
> It would make losing pagevec's benefit.
> But as your scenario, I think it doesn't call lru_drain_all frequently.
> That's because it is called when process call things related unmap
> operation or swapping.
> Do you have a such workload in test case?
Yes, I'm testing if activate_page_drain_cpu() causes the regression. This regression
is small (<2%) for a stress test and sometimes not stable.
> >> > Signed-off-by: Shaohua Li <shaohua.li@xxxxxxxxx>
> >> >
> >> > diff --git a/mm/swap.c b/mm/swap.c
> >> > index 3ce7bc3..4a3fd7f 100644
> >> > --- a/mm/swap.c
> >> > +++ b/mm/swap.c
> >> > @@ -39,6 +39,7 @@ int page_cluster;
> >> >
> >> > static DEFINE_PER_CPU(struct pagevec[NR_LRU_LISTS], lru_add_pvecs);
> >> > static DEFINE_PER_CPU(struct pagevec, lru_rotate_pvecs);
> >> > +static DEFINE_PER_CPU(struct pagevec, activate_page_pvecs);
> >> >
> >> > /*
> >> > * This path almost never happens for VM activity - pages are normally
> >> > @@ -175,11 +176,10 @@ static void update_page_reclaim_stat(struct zone *zone, struct page *page,
> >> > /*
> >> > * FIXME: speed this up?
> >> > */
> >> Couldn't we remove above comment by this patch?
> > ha, yes.
> >
> >> > -void activate_page(struct page *page)
> >> > +static void __activate_page(struct page *page)
> >> > {
> >> > struct zone *zone = page_zone(page);
> >> >
> >> > - spin_lock_irq(&zone->lru_lock);
> >> > if (PageLRU(page) && !PageActive(page) && !PageUnevictable(page)) {
> >> > int file = page_is_file_cache(page);
> >> > int lru = page_lru_base_type(page);
> >> > @@ -192,7 +192,46 @@ void activate_page(struct page *page)
> >> >
> >> > update_page_reclaim_stat(zone, page, file, 1);
> >> > }
> >> > - spin_unlock_irq(&zone->lru_lock);
> >> > +}
> >> > +
> >> > +static void activate_page_drain_cpu(int cpu)
> >> > +{
> >> > + struct pagevec *pvec = &per_cpu(activate_page_pvecs, cpu);
> >> > + struct zone *last_zone = NULL, *zone;
> >> > + int i, j;
> >> > +
> >> > + for (i = 0; i < pagevec_count(pvec); i++) {
> >> > + zone = page_zone(pvec->pages[i]);
> >> > + if (zone == last_zone)
> >> > + continue;
> >> > +
> >> > + if (last_zone)
> >> > + spin_unlock_irq(&last_zone->lru_lock);
> >> > + last_zone = zone;
> >> > + spin_lock_irq(&last_zone->lru_lock);
> >> > +
> >> > + for (j = i; j < pagevec_count(pvec); j++) {
> >> > + struct page *page = pvec->pages[j];
> >> > +
> >> > + if (last_zone != page_zone(page))
> >> > + continue;
> >> > + __activate_page(page);
> >> > + }
> >> > + }
> >> > + if (last_zone)
> >> > + spin_unlock_irq(&last_zone->lru_lock);
> >> > + release_pages(pvec->pages, pagevec_count(pvec), pvec->cold);
> >> > + pagevec_reinit(pvec);
> >>
> >> In worst case(DMA->NORMAL->HIGHMEM->DMA->NORMA->HIGHMEM->......),
> >> overhead would is big than old. how about following as?
> >> static DEFINE_PER_CPU(struct pagevec[MAX_NR_ZONES], activate_page_pvecs);
> >> Is it a overkill?
> > activate_page_drain_cpu is a two level loop. In you case, the drain order
> > will be DMA->DMA->NORMAL->NORMAL->HIGHMEM->HIGHMEM. Since pagevec size is
> > 14, the loop should finish quickly.
> Yes. so why do we separates lru pagevec with pagevec[NR_LRU_LISTS]?
> I think It can remove looping unnecessary looping overhead but of
> course we have to use more memory.
Each node has zones, so a pagevec[MAX_NR_ZONES] doesn't work here. And in my
test wich activate_page heavily used, activate_page_drain_cpu overhead is quite
small. This isn't worthy IMO.
Thanks,
Shaohua
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxxx For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>