[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CA+2MQi9KHbw5fJZ4pZJYbcKU545rhM3bPddCJjupmJZCGqHqaw@mail.gmail.com>
Date: Thu, 16 Apr 2020 09:30:04 +0800
From: Liang Li <liliang324@...il.com>
To: Alexander Duyck <alexander.h.duyck@...ux.intel.com>
Cc: Mel Gorman <mgorman@...hsingularity.net>, linux-mm@...ck.org,
linux-kernel@...r.kernel.org,
Andrea Arcangeli <aarcange@...hat.com>,
Dan Williams <dan.j.williams@...el.com>,
Dave Hansen <dave.hansen@...el.com>,
David Hildenbrand <david@...hat.com>,
Michal Hocko <mhocko@...nel.org>,
Andrew Morton <akpm@...ux-foundation.org>,
Alex Williamson <alex.williamson@...hat.com>
Subject: Re: [RFC PATCH 3/4] mm: add sys fs configuration for page reporting
On Mon, Apr 13, 2020 at 11:02 PM Alexander Duyck
<alexander.h.duyck@...ux.intel.com> wrote:
>
> On 4/12/2020 2:09 AM, liliangleo wrote:
> > This patch add 'delay_millisecs', 'mini_order', 'batch_size',
> > in '/sys/kernel/mm/page_report/'. Usage:
> >
> > "delay_millisecs":
> > Time delay interval between page free and work start to run.
> >
> > "mini_order":
> > Only pages with order equal or greater than mini_order will be
> > reported.
> >
> > "batch_size"
> > Wake up the worker only when free pages total size are greater
> > than 'batch_size'.
> >
> > Cc: Alexander Duyck <alexander.h.duyck@...ux.intel.com>
> > Cc: Mel Gorman <mgorman@...hsingularity.net>
> > Cc: Andrea Arcangeli <aarcange@...hat.com>
> > Cc: Dan Williams <dan.j.williams@...el.com>
> > Cc: Dave Hansen <dave.hansen@...el.com>
> > Cc: David Hildenbrand <david@...hat.com>
> > Cc: Michal Hocko <mhocko@...nel.org>
> > Cc: Andrew Morton <akpm@...ux-foundation.org>
> > Cc: Alex Williamson <alex.williamson@...hat.com>
> > Signed-off-by: liliangleo <liliangleo@...iglobal.com>
>
> I am not really a fan of making these configurable globally. Especially
> since the existing virtio-balloon is relying on some of this being
> configured the way it is.
>
> It would make much more sense to push these configuration options out to
> the registration interface so that the thing that is registering for
> page reporting can configure them when it is registered.
>
Agree, that's better. Thanks!
Powered by blists - more mailing lists