Re: [PATCH -mm 7/7] mm: thp: introduce compound_head_put_tail(),change get_futex_key() to use it

[Oleg Nesterov]

On 12/18, Peter Zijlstra wrote:
>
> On Wed, Dec 18, 2013 at 08:20:08PM +0100, Oleg Nesterov wrote:
> > +struct page *compound_head_put_tail(struct page *page)
> > +{
> > +	struct page *page_head;
> > +	unsigned long flags;
> > +
> > +	if (!PageTail(page))
> > +		return page;
> > +
> > +	page_head = compound_trans_head(page);
> > +
> > +	if (!__compound_tail_refcounted(page_head)) {
>
> This barrier is missing a comment describing the ordering and the
> pairing.

Yes, and this function lacks other comments it needs, they should be
copy-and-past'ed from the very similar __put_page_tail().

But I am going to unify them later if this series passes the review,
can't the comments wait till then?

OTOH, I won't mind to send v2 with the additional comment about this
rmb() at least. Or perhaps I should simply add /* See the comments in
__put_page_tail() */ into the new helper.

OK. But I'll wait for other reviews first.

Oleg.

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/