public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Peter Hurley <peter@hurleysoftware.com>
To: Fengguang Wu <fengguang.wu@intel.com>
Cc: Greg KH <gregkh@linuxfoundation.org>,
	LKML <linux-kernel@vger.kernel.org>,
	lkp@01.org, Tejun Heo <tj@kernel.org>
Subject: Re: increased vmap_area_lock contentions on "n_tty: Move buffers into n_tty_data"
Date: Tue, 17 Sep 2013 20:49:51 -0400	[thread overview]
Message-ID: <5238F8AF.4050309@hurleysoftware.com> (raw)
In-Reply-To: <20130913010936.GA1291@localhost>

On 09/12/2013 09:09 PM, Fengguang Wu wrote:
> On Fri, Sep 13, 2013 at 08:51:33AM +0800, Fengguang Wu wrote:
>> Hi Peter,
>>
>> FYI, we noticed much increased vmap_area_lock contentions since this
>> commit:
>>
>> commit 20bafb3d23d108bc0a896eb8b7c1501f4f649b77
>> Author: Peter Hurley <peter@hurleysoftware.com>
>> Date:   Sat Jun 15 10:21:19 2013 -0400
>>
>>      n_tty: Move buffers into n_tty_data
>>
>>      Reduce pointer reloading and improve locality-of-reference;
>>      allocate read_buf and echo_buf within struct n_tty_data.
>
> Here are some comparison between this commit [o] with its parent commit [*].

[...]

>      8cb06c983822103da1cf      20bafb3d23d108bc0a89
> ------------------------  ------------------------
>                    976.67      +108.3%      2034.67  lkp-a04/micro/netperf/120s-200%-TCP_STREAM
>                   8971.36       +11.4%      9997.05  nhm-white/micro/aim7/exec_test
>                   9948.03       +20.9%     12031.72  TOTAL slabinfo.kmalloc-128.active_objs
>
>      8cb06c983822103da1cf      20bafb3d23d108bc0a89
> ------------------------  ------------------------
>                    976.67      +108.3%      2034.67  lkp-a04/micro/netperf/120s-200%-TCP_STREAM
>                   9127.64       +11.4%     10164.15  nhm-white/micro/aim7/exec_test
>                  10104.31       +20.7%     12198.82  TOTAL slabinfo.kmalloc-128.num_objs

The dramatic increase in 128-byte kmalloc blocks is from vmalloc overhead
with associated with each allocation. On a x86_64, struct vmap_area is 104 bytes,
rounded to 128, allocated with every vmalloc allocation. This is approx 1% overhead
(which seems high to me).

The reason this is still visible after the test completes is the vmap area is
lazily reclaimed (see mm/vmalloc.c:__purge_vmap_area_lazy()).

1% memory overhead coupled with the unwanted vmap_area_lock contention (even though it is
test-induced) -- I might revert this anyway.

Regards,
Peter Hurley

  parent reply	other threads:[~2013-09-18  0:49 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-09-13  0:51 increased vmap_area_lock contentions on "n_tty: Move buffers into n_tty_data" Fengguang Wu
2013-09-13  1:09 ` Fengguang Wu
2013-09-17 15:34   ` Peter Hurley
2013-09-17 23:22     ` Fengguang Wu
2013-09-18  0:22       ` Peter Hurley
2013-09-25  9:04         ` Lin Ming
2013-09-25 11:30           ` Peter Hurley
2013-09-25 14:53             ` Lin Ming
2013-09-25 16:02             ` Lin Ming
2013-09-26  3:20               ` Andi Kleen
2013-09-26 11:52                 ` Peter Hurley
2013-09-26 15:32                   ` Andi Kleen
2013-09-26 17:22                     ` Peter Hurley
2013-09-26  7:33         ` Andrew Morton
2013-09-26 11:31           ` Peter Hurley
2013-09-26 15:04             ` Greg KH
2013-09-26 17:35               ` Peter Hurley
2013-09-26 18:05                 ` Andrew Morton
2013-09-26 21:42                   ` Peter Hurley
2013-09-26 21:58                     ` Andrew Morton
2013-09-26 22:21                       ` Peter Hurley
2013-09-18  0:49   ` Peter Hurley [this message]
2013-09-13  3:17 ` Greg KH
2013-09-13  3:38   ` Fengguang Wu
2013-09-13  3:44     ` Greg KH
2013-09-13  9:55       ` Peter Hurley
2013-09-13 12:34         ` Greg KH
2013-09-17  2:42     ` Peter Hurley
2013-09-17  2:56       ` Fengguang Wu

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=5238F8AF.4050309@hurleysoftware.com \
    --to=peter@hurleysoftware.com \
    --cc=fengguang.wu@intel.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lkp@01.org \
    --cc=tj@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox