[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20180423130837.GA25183@kroah.com>
Date: Mon, 23 Apr 2018 15:08:37 +0200
From: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
To: NeilBrown <neilb@...e.com>
Cc: Oleg Drokin <oleg.drokin@...el.com>,
James Simmons <jsimmons@...radead.org>,
Andreas Dilger <andreas.dilger@...el.com>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
Lustre Development List <lustre-devel@...ts.lustre.org>
Subject: Re: [PATCH 00/20] staging: lustre: convert to rhashtable
On Thu, Apr 12, 2018 at 07:54:48AM +1000, NeilBrown wrote:
> libcfs in lustre has a resizeable hashtable.
> Linux already has a resizeable hashtable, rhashtable, which is better
> is most metrics. See https://lwn.net/Articles/751374/ in a few days
> for an introduction to rhashtable.
>
> This series converts lustre to use rhashtable. This affects several
> different tables, and each is different is various ways.
>
> There are two outstanding issues. One is that a bug in rhashtable
> means that we cannot enable auto-shrinking in one of the tables. That
> is documented as appropriate and should be fixed soon.
>
> The other is that rhashtable has an atomic_t which counts the elements
> in a hash table. At least one table in lustre went to some trouble to
> avoid any table-wide atomics, so that could lead to a regression.
> I'm hoping that rhashtable can be enhanced with the option of a
> per-cpu counter, or similar.
>
>
> I have enabled automatic shrinking on all tables where it makes sense
> and doesn't trigger the bug. I have also removed all hints concerning
> min/max size - I cannot see how these could be useful.
>
> The dump_pgcache debugfs file provided some interesting challenges. I
> think I have cleaned it up enough so that it all makes sense. An
> extra pair of eyes examining that code in particular would be
> appreciated.
>
> This series passes all the same tests that pass before the patches are
> applied.
I've taken the first 4 patches of this series, as they were "obviously
correct". I'll let you and James argue about the rest. Feel free to
resend when there's some sort of agreement.
thanks,
greg k-h
Powered by blists - more mailing lists