From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
        id S1166971AbdDXInQ (ORCPT <rfc822;w@1wt.eu>);
        Mon, 24 Apr 2017 04:43:16 -0400
Received: from mout.gmx.net ([212.227.17.20]:56996 "EHLO mout.gmx.net"
        rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
        id S1166939AbdDXInI (ORCPT <rfc822;linux-kernel@vger.kernel.org>);
        Mon, 24 Apr 2017 04:43:08 -0400
Message-ID: <1493023382.6266.9.camel@gmx.de>
Subject: Re: TREE_SRCU slows hotplug by factor ~16
From: Mike Galbraith <efault@gmx.de>
To: paulmck@linux.vnet.ibm.com
Cc: LKML <linux-kernel@vger.kernel.org>
Date: Mon, 24 Apr 2017 10:43:02 +0200
In-Reply-To: <1493019303.4617.6.camel@gmx.de>
References: <1493002089.4145.7.camel@gmx.de>
         <20170424033250.GX3956@linux.vnet.ibm.com>
         <1493011484.4166.13.camel@gmx.de>
         <20170424062220.GZ3956@linux.vnet.ibm.com> <1493019303.4617.6.camel@gmx.de>
Content-Type: text/plain; charset="UTF-8"
X-Mailer: Evolution 3.16.5 
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
X-Provags-ID: V03:K0:1ydAMYGJXyGiE3BQpgC0RpIbGBKZ7OTWzldXUyt19mKMBOQ0x1L
 keKGudPrNwG87NnAJTm/Fqyf1FKqZOXxq8uSWK237VRyGdJMbIwhGDpZtNJJqStFv1iyAab
 hs/KzZSHmGj+TlQoQo1a+pvYu+Kd7OrHuJocARULlUvRHB9Fm8y4uJ+CgUqDVtRKFsRgnAF
 XsyAWmZvSWnQSo/2AGTvA==
X-UI-Out-Filterresults: notjunk:1;V01:K0:haA2s6EQuok=:uVOLA0A8/Wp5HoIfowkLJf
 ow71/daDzXYayhzLrzxhTtsb+yc5df/Fh9Ii5yEuqMOQfQyFIKMYCsuWShzFPIC+RfERkCxrQ
 qA1XqKvyBOUygVyLRuVvB6DZGbPPEBErToSVUVRc57NuSM0RwBwRmN5/XrMdvHWdzDYXEBxLO
 m79+FU0bou+T8reWzru47vv/slfDSVnMgwBG5e4ct8kEzfvls9u+abxMeEX0QOfIfDNNGFK0h
 VygsWVW+PU+JL65zpwb91dEUqrJqxbZ4LnjJYez04QkQSD3uZz0VaTNGz9abwTHzmjK44ULxK
 2v0xff2VAo64OHWUOSjkQnUFE598k3psfhWB8DFbH9igUW1gaEWOPFdjxuuJWJtc/RgjanOOC
 TG5+tWAOidZvvCDEYMxzZvcSKeicK+6WpSZdMNJT3GDvLvxBsb60Uan87blfZg/kg9rqgSvdp
 KvqhB/51Y4k8QfE4RXeb4PXtAqeBV+xH0T+jw5TRmEdlYpHclbPe/qz7NOhrSIIOxw3AjChQp
 bBCsG01pqNBXwCxAjZ8vh6bsKX6JK+dULSm8OMVw69z+cUPB5EtF/ZO4czoZ3v4S2IpolK6iP
 cKLa+Fs3VKlGVglgSexA5T54YSUHHpWf4P5mOfre28MlPkcrDcOo9j/unTPDIMY0dKj77jdQ9
 z0C1Wjoxo5wJ5qsQVCzyITzuuoUr+lzDB1S6c7NwFZ356YQ1L/Y69ERAEHl4vOMFb/BgFob/V
 8czsHKQQ4oZtdm2FRk+0IkNnl99rfvEe8myr/BdCtxCRKMv7mHUTdBAgmteBFKSp28UNxmJSc
 b+/nPBJ
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On Mon, 2017-04-24 at 09:35 +0200, Mike Galbraith wrote:

> # tracer: nop
> #
> # entries-in-buffer/entries-written: 229332/229332   #P:8
> #
> #                              _-----=> irqs-off
> #                             / _----=> need-resched
> #                            | / _---=> hardirq/softirq
> #                            || / _--=> preempt-depth
> #                            ||| /     delay
> #           TASK-PID   CPU#  ||||    TIMESTAMP  FUNCTION
> #              | |       |   ||||       |         |
>  stress-cpu-hotp-2443  [004] ....   144.123608: synchronize_srcu: enter
>  stress-cpu-hotp-2443  [004] ....   144.123610: call_srcu: enter
>  stress-cpu-hotp-2443  [004] ....   144.123612: call_srcu: exit
>  stress-cpu-hotp-2443  [005] ....   144.139528: synchronize_srcu: exit

I probably could have sent a LOT less trace.

classic:

# tracer: nop
#
# entries-in-buffer/entries-written: 114688/114688   #P:8
#
#                              _-----=> irqs-off
#                             / _----=> need-resched
#                            | / _---=> hardirq/softirq
#                            || / _--=> preempt-depth
#                            ||| /     delay
#           TASK-PID   CPU#  ||||    TIMESTAMP  FUNCTION
#              | |       |   ||||       |         |
 stress-cpu-hotp-4337  [007] ....   174.082801: synchronize_srcu: enter
 stress-cpu-hotp-4337  [007] ....   174.082802: synchronize_srcu: exit
 stress-cpu-hotp-4337  [007] ....   174.082896: synchronize_srcu: enter
 stress-cpu-hotp-4337  [007] ....   174.082896: synchronize_srcu: exit
 stress-cpu-hotp-4337  [007] ....   174.082975: synchronize_srcu: enter
 stress-cpu-hotp-4337  [007] ....   174.082975: synchronize_srcu: exit
 stress-cpu-hotp-4337  [007] ....   174.083051: synchronize_srcu: enter
 stress-cpu-hotp-4337  [007] ....   174.083051: synchronize_srcu: exit
....