Re: [PATCH -mm v6 5/6] proc: add kpageidle file

[Vladimir Davydov <vdavydov@parallels.com>]

On Thu, Jul 09, 2015 at 04:19:00PM +0300, Vladimir Davydov wrote:
> On Wed, Jul 08, 2015 at 04:01:13PM -0700, Andres Lagar-Cavilla wrote:
> > On Fri, Jun 12, 2015 at 2:52 AM, Vladimir Davydov
> > > +#ifdef CONFIG_IDLE_PAGE_TRACKING
> > > +/*
> > > + * Idle page tracking only considers user memory pages, for other types of
> > > + * pages the idle flag is always unset and an attempt to set it is silently
> > > + * ignored.
> > > + *
> > > + * We treat a page as a user memory page if it is on an LRU list, because it is
> > > + * always safe to pass such a page to page_referenced(), which is essential for
> > > + * idle page tracking. With such an indicator of user pages we can skip
> > > + * isolated pages, but since there are not usually many of them, it will hardly
> > > + * affect the overall result.
> > > + *
> > > + * This function tries to get a user memory page by pfn as described above.
> > > + */
> > > +static struct page *kpageidle_get_page(unsigned long pfn)
> > > +{
> > > +       struct page *page;
> > > +       struct zone *zone;
> > > +
> > > +       if (!pfn_valid(pfn))
> > > +               return NULL;
> > > +
> > > +       page = pfn_to_page(pfn);
> > > +       if (!page || !PageLRU(page))
> > 
> > Isolation can race in while you're processing the page, after these
> > checks. This is ok, but worth a small comment.
> 
> Agree, will add one.

Oh, the comment is already present - it's in the description to this
function. Minchan asked me to add it long time ago, and so I did.
Completely forgot about it.

Thanks,
Vladimir