Re: [PATCH 3/3] mm: tail page refcounting optimization for slab andhugetlbfs

[Andrea Arcangeli]

Hi Andrew!

On Tue, Nov 19, 2013 at 03:14:16PM -0800, Andrew Morton wrote:
> This looks like it was attacked by Lindent.  How's this look?

Indeed, I didn't try to optimize as 3/3 was mostly a test patch so
far.

> +	struct page *page_head;
> +
> +	if (likely(PageTail(page))) {

!PageTail here.

The current version would look like below, which is less obviously
horrible than before :). We may also consider to keep your indent
cleanup incremental as it's much easier to review the actual change of
code logic the below way.