[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <56D76138.1010606@hpe.com>
Date: Wed, 2 Mar 2016 16:55:04 -0500
From: Waiman Long <waiman.long@....com>
To: Ingo Molnar <mingo@...nel.org>
CC: Alexander Viro <viro@...iv.linux.org.uk>, Jan Kara <jack@...e.com>,
Jeff Layton <jlayton@...chiereds.net>,
"J. Bruce Fields" <bfields@...ldses.org>,
Tejun Heo <tj@...nel.org>,
Christoph Lameter <cl@...ux-foundation.org>,
<linux-fsdevel@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
Ingo Molnar <mingo@...hat.com>,
Peter Zijlstra <peterz@...radead.org>,
Andi Kleen <andi@...stfloor.org>,
Dave Chinner <dchinner@...hat.com>,
Boqun Feng <boqun.feng@...il.com>,
Scott J Norton <scott.norton@...com>,
Douglas Hatch <doug.hatch@...com>,
Linus Torvalds <torvalds@...ux-foundation.org>,
Andrew Morton <akpm@...ux-foundation.org>,
Peter Zijlstra <a.p.zijlstra@...llo.nl>
Subject: Re: [PATCH v5 5/5] lib/percpu-list: Add a config parameter for disabling
per-cpu list
On 03/02/2016 03:41 AM, Ingo Molnar wrote:
> * Waiman Long<Waiman.Long@....com> wrote:
>
>> As there is concern that the larger pcpu_list_node structure and the
>> per-cpu overhead may be a waste of resource on small system. This patch
>> adds a config parameter CONFIG_PERCPU_LIST to disable the per-cpu list
>> if the kernel builder chooses to do so. With per-cpu list disabled,
>> all the different groups of per-cpu lists will be degenerated into
>> global lists for all the CPUs.
>>
>> The current default is to enable per-cpu list. A kernel builder needs
>> to explicitly turn it off.
>>
>> Signed-off-by: Waiman Long<Waiman.Long@....com>
>> ---
>> fs/inode.c | 2 +-
>> include/linux/percpu-list.h | 93 ++++++++++++++++++++++++++++++++++++++++++-
>> lib/Kconfig | 14 ++++++
>> lib/percpu-list.c | 24 +++++++++++-
>> 4 files changed, 130 insertions(+), 3 deletions(-)
> I think this kind of #ifdef complexity and the doubling of our Kconfig and testing
> space is counterproductive, and I think the per CPU locking is a win on as small
> as dual core CPUs, and on UP CPUs the per CPU list becomes a single global list
> automatically.
>
> I'm not against visible memory savings for overly clever scalability features, but
> this does not appear to be such a case, so:
>
> NAKed-by: Ingo Molnar<mingo@...nel.org>
>
> Thanks,
>
> Ingo
The last patch was there to answer a feedback from Jan. I am fine if
that is not merged as I prefer to have the per-cpu list capability as
the default anyway.
Cheers,
Longman
Powered by blists - more mailing lists