On Tue, Aug 13, 2024 at 3:00 PM Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> wrote:
>
> On Mon, 12 Aug 2024 16:48:23 -0600 Yu Zhao <yuzhao@xxxxxxxxxx> wrote:
>
> > Batch the HVO work, including de-HVO of the source and HVO of the
> > destination hugeTLB folios, to speed up demotion.
> >
> > After commit bd225530a4c7 ("mm/hugetlb_vmemmap: fix race with
> > speculative PFN walkers"), each request of HVO or de-HVO, batched or
> > not, invokes synchronize_rcu() once. For example, when not batched,
> > demoting one 1GB hugeTLB folio to 512 2MB hugeTLB folios invokes
> > synchronize_rcu() 513 times (1 de-HVO plus 512 HVO requests), whereas
> > when batched, only twice (1 de-HVO plus 1 HVO request). And the
> > performance difference between the two cases is significant, e.g.,
> > echo 2048kB >/sys/kernel/mm/hugepages/hugepages-1048576kB/demote_size
> > time echo 100 >/sys/kernel/mm/hugepages/hugepages-1048576kB/demote
> >
> > Before this patch:
> > real 8m58.158s
> > user 0m0.009s
> > sys 0m5.900s
> >
> > After this patch:
> > real 0m0.900s
> > user 0m0.000s
> > sys 0m0.851s
>
> That's a large change. I assume the now-fixed regression was of
> similar magnitude?
Correct, and only the `real` time was regressed, due to
synchronize_rcu(); the `sys` time is an improvement, since it's not
affected by synchronize_rcu() (or not I could measure).
> > Note that this patch changes the behavior of the `demote` interface
> > when de-HVO fails. Before, the interface aborts immediately upon
> > failure; now, it tries to finish an entire batch, meaning it can make
> > extra progress if the rest of the batch contains folios that do not
> > need to de-HVO.
> >
> > Fixes: bd225530a4c7 ("mm/hugetlb_vmemmap: fix race with speculative PFN walkers")
>
> Do we think we should add this to 6.10.x? I do.
Agreed.
