On Thu, Jan 15, 2015 at 05:16:46PM -0800, Andrew Morton wrote:
> On Thu, 15 Jan 2015 16:40:33 +0900 Joonsoo Kim <iamjoonsoo.kim@xxxxxxx> wrote:
>
> > compound_head() is implemented with assumption that there would be
> > race condition when checking tail flag. This assumption is only true
> > when we try to access arbitrary positioned struct page.
> >
> > The situation that virt_to_head_page() is called is different case.
> > We call virt_to_head_page() only in the range of allocated pages,
> > so there is no race condition on tail flag. In this case, we don't
> > need to handle race condition and we can reduce overhead slightly.
> > This patch implements compound_head_fast() which is similar with
> > compound_head() except tail flag race handling. And then,
> > virt_to_head_page() uses this optimized function to improve performance.
> >
> > I saw 1.8% win in a fast-path loop over kmem_cache_alloc/free,
> > (14.063 ns -> 13.810 ns) if target object is on tail page.
> >
> > ...
> >
> > --- a/include/linux/mm.h
> > +++ b/include/linux/mm.h
> > @@ -453,6 +453,13 @@ static inline struct page *compound_head(struct page *page)
> > return page;
> > }
> >
> > +static inline struct page *compound_head_fast(struct page *page)
> > +{
> > + if (unlikely(PageTail(page)))
> > + return page->first_page;
> > + return page;
> > +}
>
> Can we please have some code comments which let people know when they
> should and shouldn't use compound_head_fast()? I shouldn't have to say
> this :(
Okay.
>
> > /*
> > * The atomic page->_mapcount, starts from -1: so that transitions
> > * both from it and to it can be tracked, using atomic_inc_and_test
> > @@ -531,7 +538,8 @@ static inline void get_page(struct page *page)
> > static inline struct page *virt_to_head_page(const void *x)
> > {
> > struct page *page = virt_to_page(x);
> > - return compound_head(page);
> > +
> > + return compound_head_fast(page);
>
> And perhaps some explanation here as to why virt_to_head_page() can
> safely use compound_head_fast(). There's an assumption here that
> nobody will be dismantling the compound page while virt_to_head_page()
> is in progress, yes? And this assumption also holds for the calling
> code, because otherwise the virt_to_head_page() return value is kinda
> meaningless.
>
> This is tricky stuff - let's spell it out carefully.
Okay.
I already sent v3 and it would have proper code comments.
Thanks.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>