[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20240506070403.181b574d@kernel.org>
Date: Mon, 6 May 2024 07:04:03 -0700
From: Jakub Kicinski <kuba@...nel.org>
To: fuyuanli <fuyuanli@...iglobal.com>
Cc: <mingo@...hat.com>, <peterz@...radead.org>, <juri.lelli@...hat.com>,
<vincent.guittot@...aro.org>, <dietmar.eggemann@....com>,
<rostedt@...dmis.org>, <bsegall@...gle.com>, <mgorman@...e.de>,
<bristot@...hat.com>, <vschneid@...hat.com>, <kolyshkin@...il.com>,
<akpm@...ux-foundation.org>, <kent.overstreet@...ux.dev>,
<frederic@...nel.org>, <oleg@...hat.com>, <hca@...ux.ibm.com>,
<paulmck@...nel.org>, <tj@...nel.org>, <bigeasy@...utronix.de>,
<CruzZhao@...ux.alibaba.com>, <pabeni@...hat.com>,
<linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] sched: Introduce task_struct::latency_sensi_flag.
On Sun, 5 May 2024 11:06:15 +0800 fuyuanli wrote:
> A test has been made in two hosts named A and B. In A, several clients
> sent udp packets to a single server in B concurrently as fast as
> possible. In B, the IRQs of these flows were bound to CPU 0 by flow
> director, so there was always a triggered net_rx softirq on CPU 0. Then
> a test program was started in B, which was also bound to CPU 0, and
> keeped calling sendto() in a loop. Sampling with perf, results showed
> that about 25% of running time of test program was spent executing
> local_bh_enable() contained in syscall sendto(), but after setting
> latency_sensi_flag to 1, this proportion had been reduced to 0.5%.
Enable threaded NAPI, it will have the same effect.
Powered by blists - more mailing lists