Re: [External] Re: [PATCH v2 1/6] mm: migrate: do not migrate HugeTLB page whose refcount is one

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Thu, Jan 7, 2021 at 7:16 PM Michal Hocko <mhocko@xxxxxxxx> wrote:
>
> On Thu 07-01-21 10:52:21, Muchun Song wrote:
> > On Thu, Jan 7, 2021 at 12:13 AM Michal Hocko <mhocko@xxxxxxxx> wrote:
> > >
> > > On Wed 06-01-21 16:47:34, Muchun Song wrote:
> > > > If the refcount is one when it is migrated, it means that the page
> > > > was freed from under us. So we are done and do not need to migrate.
> > >
> > > Is this common enough that it would warrant the explicit check for each
> > > migration?
> >
> > Are you worried about the overhead caused by the check? Thanks.
>
> I am not as worried as I would like to see some actual justification for
> this optimization.

OK. The HugeTLB page can be freed after it was isolated but before
being migrated. Right? If we catch this case, it is an optimization.
I do this just like unmap_and_move() does for a base page.

> --
> Michal Hocko
> SUSE Labs




[Index of Archives]     [Linux ARM Kernel]     [Linux ARM]     [Linux Omap]     [Fedora ARM]     [IETF Annouce]     [Bugtraq]     [Linux OMAP]     [Linux MIPS]     [eCos]     [Asterisk Internet PBX]     [Linux API]

  Powered by Linux