Re: [PATCH -mm 4/7] mm: thp: turn put_compound_page() into__put_page_tail()

[Oleg Nesterov]

On 12/18, Peter Zijlstra wrote:
>
> On Wed, Dec 18, 2013 at 08:19:58PM +0100, Oleg Nesterov wrote:
> > @@ -247,7 +242,7 @@ bool __get_page_tail(struct page *page)
> >  	bool got;
> >  	struct page *page_head = compound_trans_head(page);
> >
> > -	/* Ref to put_compound_page() comment. */
> > +	/* Ref to __put_page_tail() comment. */
> >  	if (!__compound_tail_refcounted(page_head)) {
> >  		smp_rmb();
> >  		if (likely(PageTail(page))) {
>
> What code is this against, my local tree doesn't have that smp_rmb().

This is against -mm code.

> This suggests its a recent patch; which is good since then we can still
> drop it and wait for people to send one with a proper comment in.

put_compound_page() explains this barrier, this is what "Ref" above means,
I guess.

To remind, I think we can do more cleanups here.

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/