Re: [RFC PATCH 0/4] Reduce tree_lock contention during swap and reclaim of a single file v1

[Linus Torvalds]

On Fri, Sep 9, 2016 at 2:59 AM, Mel Gorman <mgorman@xxxxxxxxxxxxxxxxxxx> wrote:
>
> The progression of this series has been unsatisfactory.

Yeah, I have to say that I particularly don't like patch #1. It's some
rather nasty complexity for dubious gains, and holding the lock for
longer times might have downsides.

And the numbers seem to not necessarily be in favor of patch #3
either, which I would have otherwise been predisposed to like (ie it
looks fairly targeted and not very complex).

#2 seems trivially correct but largely irrelevant.

So I think this series is one of those "we need to find that it makes
a big positive impact" to make sense.

              Linus