[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <alpine.LRH.2.02.1806131446110.26196@file01.intranet.prod.int.rdu2.redhat.com>
Date: Wed, 13 Jun 2018 14:53:47 -0400 (EDT)
From: Mikulas Patocka <mpatocka@...hat.com>
To: Christoph Hellwig <hch@...radead.org>
cc: Christopher Lameter <cl@...ux.com>,
Mike Snitzer <snitzer@...hat.com>,
Vlastimil Babka <vbabka@...e.cz>,
Matthew Wilcox <willy@...radead.org>,
Pekka Enberg <penberg@...nel.org>, linux-mm@...ck.org,
dm-devel@...hat.com, David Rientjes <rientjes@...gle.com>,
Joonsoo Kim <iamjoonsoo.kim@....com>,
Andrew Morton <akpm@...ux-foundation.org>,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH RESEND] slab: introduce the flag SLAB_MINIMIZE_WASTE
On Wed, 13 Jun 2018, Christoph Hellwig wrote:
> On Wed, Jun 13, 2018 at 01:01:22PM -0400, Mikulas Patocka wrote:
> > Hi
> >
> > I'd like to ask about this patch - will you commit it, or do you want to
> > make some more changes to it?
>
> How about you resend it with the series adding an actual user once
> ready? I haven't actually seen patches using it posted on any list yet.
dm-bufio is already using it. Starting with the kernel 4.17 (f51f2e0a7fb1
- "dm bufio: support non-power-of-two block sizes"), dm-bufio has the
capability to use non-power-of-two buffers. It uses slab cache for its
buffers - so we would like to have this slab optimization - to avoid
excessive memory wasting.
Originally, the slub patch used a new flag SLAB_MINIMIZE_WASTE, but after
a suggestion from others, I reworked the patch so that it minimizes waste
of all slub caches and doesn't need an extra flag to activate.
Mikulas
Powered by blists - more mailing lists