[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20220907134320.01d1693387bde4af5d3100fb@linux-foundation.org>
Date: Wed, 7 Sep 2022 13:43:20 -0700
From: Andrew Morton <akpm@...ux-foundation.org>
To: "Sun, Jiebin" <jiebin.sun@...el.com>
Cc: Tim Chen <tim.c.chen@...ux.intel.com>,
Shakeel Butt <shakeelb@...gle.com>, vasily.averin@...ux.dev,
Dennis Zhou <dennis@...nel.org>, Tejun Heo <tj@...nel.org>,
Christoph Lameter <cl@...ux.com>,
"Eric W. Biederman" <ebiederm@...ssion.com>,
Alexey Gladkov <legion@...nel.org>,
Manfred Spraul <manfred@...orfullife.com>,
alexander.mikhalitsyn@...tuozzo.com, Linux MM <linux-mm@...ck.org>,
LKML <linux-kernel@...r.kernel.org>,
"Chen, Tim C" <tim.c.chen@...el.com>,
Feng Tang <feng.tang@...el.com>,
Huang Ying <ying.huang@...el.com>, tianyou.li@...el.com,
wangyang.guo@...el.com
Subject: Re: [PATCH] ipc/msg.c: mitigate the lock contention with percpu
counter
On Wed, 7 Sep 2022 17:39:47 +0800 "Sun, Jiebin" <jiebin.sun@...el.com> wrote:
> I have sent out the patch v4 which use percpu_counter_add_batch. If we use
> a tuned large batch size (1024),
Oh. Why not simply use a batch size of INT_MAX?
> the performance gain is 3.17x (patch v4)
> vs 3.38x (patch v3) previously in stress-ng -- message. It still has
> significant performance improvement and also good balance between
> performance gain and overflow issue.
Powered by blists - more mailing lists