Re: [PATCH 17/17] mm: filemap: Avoid unnecessary barries and waitqueue lookup in unlock_page fastpath

[Peter Zijlstra]

On Thu, May 01, 2014 at 09:44:48AM +0100, Mel Gorman wrote:
> +/*
> + * If PageWaiters was found to be set at unlock time, __wake_page_waiters
> + * should be called to actually perform the wakeup of waiters.
> + */
> +static inline void __wake_page_waiters(struct page *page)
> +{
> +	ClearPageWaiters(page);

-ENOCOMMENT

barriers should always come with a comment that explain the memory
ordering and reference the pairing barrier.

Also, FWIW, there's a mass rename queued for .16 that'll make this:

  smp_mb__after_atomic();

but for now it will also still provide the old names with a __deprecated
tag on, so no real harm.

> +	smp_mb__after_clear_bit();
> +	wake_up_page(page, PG_locked);
> +}
--
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/