All of lore.kernel.org
 help / color / mirror / Atom feed
From: Lai Jiangshan <laijs-BthXqXjhjHXQFUHtdCDX3A@public.gmane.org>
To: Christoph Lameter <cl-vYTEC60ixJUAvxtiuMwx3w@public.gmane.org>
Cc: Jiri Kosina <jkosina-AlSwsSmVLrQ@public.gmane.org>,
	Dan Magenheimer
	<dan.magenheimer-QHcLZuEGTsvQT0dZR+AlfA@public.gmane.org>,
	linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	Michal Hocko <mhocko-AlSwsSmVLrQ@public.gmane.org>,
	Paul Gortmaker
	<paul.gortmaker-CWA4WttNNZF54TAoqtyWWQ@public.gmane.org>,
	Konstantin Khlebnikov
	<khlebnikov-GEFAQzZX7r8dnm+yROfE0A@public.gmane.org>,
	"H. Peter Anvin" <hpa-YMNOUZJC4hwAvxtiuMwx3w@public.gmane.org>,
	Sam Ravnborg <sam-uyr5N9Q2VtJg9hUCZPvPmw@public.gmane.org>,
	Gavin Shan
	<shangw-23VcF4HTsmIX0ybBhKVfKdBPR1lH4CV8@public.gmane.org>,
	Rik van Riel <riel-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
	cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	x86-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org,
	Hugh Dickins <hughd-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>,
	Ingo Molnar <mingo-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
	Mel Gorman <mgorman-l3A5Bk7waGM@public.gmane.org>,
	KOSAKI Motohiro
	<kosaki.motohiro-+CUm20s59erQFUHtdCDX3A@public.gmane.org>,
	David Rientjes <rientjes-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>,
	Petr Holasek <pholasek-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
	linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org,
	Wanlong Gao <gaowanlong-BthXqXjhjHXQFUHtdCDX3A@public.gmane.org>,
	Djalal Harouni <tixxdz-Umm1ozX2/EEdnm+yROfE0A@public.gmane.org>,
	Rusty Russell <rusty-8n+1lVoiYb80n/F98K4Iww@public.gmane.org>,
	Wen Congyang <wency-BthXqXjhjHXQFUHtdCDX3A@public.gmane.org>,
	Peter Zijlstra
	<a.p.zijlstra-/NLkJaSkS4VmR6Xm/wNWPw@public.gmane.org>,
	Pawel Moll <pawel.moll-5wv7dgnIgG8@public.gmane.org>,
	Me
Subject: Re: [RFC PATCH 14/23 V2] slub, hotplug: ignore unrelated node's hot-adding and hot-removing
Date: Mon, 06 Aug 2012 13:39:36 +0800	[thread overview]
Message-ID: <501F5898.8070502@cn.fujitsu.com> (raw)
In-Reply-To: <alpine.DEB.2.00.1208021102560.23049-sBS69tsa9Uj/9pzu0YdTqQ@public.gmane.org>

On 08/03/2012 12:04 AM, Christoph Lameter wrote:
> On Thu, 2 Aug 2012, Lai Jiangshan wrote:
> 
>> SLUB only fucus on the nodes which has normal memory, so ignore the other
>> node's hot-adding and hot-removing.
> 
> You would need to do the same for SLAB. SLAB has an easier time with
> falling back to other nodes (and therefore does not show up in your
> tests) but as a result SLAB will be quite ineffective
> because it created bogus structures that are never used yet constantly
> traversed.
> 
> 

Because SLAB creates bogus structures, so it can care less things when in hotplug.
It is just ugly but it is not bug currently, so I want to delay to fix it.
And SLAB is not so friendly with hot-remove as SLUB in my view, we can fix it altogether in future.

Thanks,
Lai

WARNING: multiple messages have this Message-ID (diff)
From: Lai Jiangshan <laijs-BthXqXjhjHXQFUHtdCDX3A@public.gmane.org>
To: Christoph Lameter <cl-vYTEC60ixJUAvxtiuMwx3w@public.gmane.org>
Cc: Jiri Kosina <jkosina-AlSwsSmVLrQ@public.gmane.org>,
	Dan Magenheimer
	<dan.magenheimer-QHcLZuEGTsvQT0dZR+AlfA@public.gmane.org>,
	linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	Michal Hocko <mhocko-AlSwsSmVLrQ@public.gmane.org>,
	Paul Gortmaker
	<paul.gortmaker-CWA4WttNNZF54TAoqtyWWQ@public.gmane.org>,
	Konstantin Khlebnikov
	<khlebnikov-GEFAQzZX7r8dnm+yROfE0A@public.gmane.org>,
	"H. Peter Anvin" <hpa-YMNOUZJC4hwAvxtiuMwx3w@public.gmane.org>,
	Sam Ravnborg <sam-uyr5N9Q2VtJg9hUCZPvPmw@public.gmane.org>,
	Gavin Shan
	<shangw-23VcF4HTsmIX0ybBhKVfKdBPR1lH4CV8@public.gmane.org>,
	Rik van Riel <riel-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
	cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	x86-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org,
	Hugh Dickins <hughd-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>,
	Ingo Molnar <mingo-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
	Mel Gorman <mgorman-l3A5Bk7waGM@public.gmane.org>,
	KOSAKI Motohiro
	<kosaki.motohiro-+CUm20s59erQFUHtdCDX3A@public.gmane.org>,
	David Rientjes <rientjes-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>,
	Petr Holasek <pholasek-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
	linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org,
	Wanlong Gao <gaowanlong-BthXqXjhjHXQFUHtdCDX3A@public.gmane.org>,
	Djalal Harouni <tixxdz-Umm1ozX2/EEdnm+yROfE0A@public.gmane.org>,
	Rusty Russell <rusty-8n+1lVoiYb80n/F98K4Iww@public.gmane.org>,
	Wen Congyang <wency-BthXqXjhjHXQFUHtdCDX3A@public.gmane.org>,
	Peter Zijlstra
	<a.p.zijlstra-/NLkJaSkS4VmR6Xm/wNWPw@public.gmane.org>,
	Pawel Moll <pawel.moll-5wv7dgnIgG8@public.gmane.org>Me
Subject: Re: [RFC PATCH 14/23 V2] slub, hotplug: ignore unrelated node's hot-adding and hot-removing
Date: Mon, 06 Aug 2012 13:39:36 +0800	[thread overview]
Message-ID: <501F5898.8070502@cn.fujitsu.com> (raw)
In-Reply-To: <alpine.DEB.2.00.1208021102560.23049-sBS69tsa9Uj/9pzu0YdTqQ@public.gmane.org>

On 08/03/2012 12:04 AM, Christoph Lameter wrote:
> On Thu, 2 Aug 2012, Lai Jiangshan wrote:
> 
>> SLUB only fucus on the nodes which has normal memory, so ignore the other
>> node's hot-adding and hot-removing.
> 
> You would need to do the same for SLAB. SLAB has an easier time with
> falling back to other nodes (and therefore does not show up in your
> tests) but as a result SLAB will be quite ineffective
> because it created bogus structures that are never used yet constantly
> traversed.
> 
> 

Because SLAB creates bogus structures, so it can care less things when in hotplug.
It is just ugly but it is not bug currently, so I want to delay to fix it.
And SLAB is not so friendly with hot-remove as SLUB in my view, we can fix it altogether in future.

Thanks,
Lai

  parent reply	other threads:[~2012-08-06  5:39 UTC|newest]

Thread overview: 49+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-08-02  2:52 [RFC PATCH 00/23 V2] memory, numa: introduce MOVABLE-dedicated node and online_movable for hotplug Lai Jiangshan
     [not found] ` <1343875991-7533-1-git-send-email-laijs-BthXqXjhjHXQFUHtdCDX3A@public.gmane.org>
2012-08-02  2:52   ` [RFC PATCH 01/23 V2] node_states: introduce N_MEMORY Lai Jiangshan
2012-08-02 16:01     ` Christoph Lameter
2012-08-02 16:01       ` Christoph Lameter
     [not found]     ` <1343875991-7533-2-git-send-email-laijs-BthXqXjhjHXQFUHtdCDX3A@public.gmane.org>
2012-08-02 16:01       ` Christoph Lameter
2012-08-04 13:38       ` Hillf Danton
2012-08-04 13:38     ` Hillf Danton
2012-08-04 13:38       ` Hillf Danton
2012-08-02  2:52   ` [RFC PATCH 02/23 V2] cpuset: use N_MEMORY instead N_HIGH_MEMORY Lai Jiangshan
     [not found]     ` <1343875991-7533-3-git-send-email-laijs-BthXqXjhjHXQFUHtdCDX3A@public.gmane.org>
2012-08-04 13:51       ` Hillf Danton
2012-08-04 13:51         ` Hillf Danton
2012-08-02  2:52   ` [RFC PATCH 03/23 V2] procfs: " Lai Jiangshan
     [not found]     ` <1343875991-7533-4-git-send-email-laijs-BthXqXjhjHXQFUHtdCDX3A@public.gmane.org>
2012-08-04 13:53       ` Hillf Danton
2012-08-04 13:53         ` Hillf Danton
2012-08-02  2:52   ` [RFC PATCH 04/23 V2] oom: " Lai Jiangshan
     [not found]     ` <1343875991-7533-5-git-send-email-laijs-BthXqXjhjHXQFUHtdCDX3A@public.gmane.org>
2012-08-04 13:54       ` Hillf Danton
2012-08-04 13:54     ` Hillf Danton
2012-08-04 13:54       ` Hillf Danton
2012-08-02  2:52   ` [RFC PATCH 05/23 V2] mm,migrate: " Lai Jiangshan
2012-08-02  2:52   ` [RFC PATCH 06/23 V2] mempolicy: " Lai Jiangshan
2012-08-02  2:52   ` [RFC PATCH 07/23 V2] memcontrol: " Lai Jiangshan
2012-08-02  2:52   ` [RFC PATCH 08/23 V2] hugetlb: " Lai Jiangshan
2012-08-02  2:52   ` [RFC PATCH 09/23 V2] vmstat: " Lai Jiangshan
2012-08-02  2:52   ` [RFC PATCH 10/23 V2] kthread: " Lai Jiangshan
2012-08-02  2:52   ` [RFC PATCH 11/23 V2] init: " Lai Jiangshan
2012-08-02  2:53   ` [RFC PATCH 12/23 V2] vmscan: " Lai Jiangshan
     [not found]     ` <1343875991-7533-13-git-send-email-laijs-BthXqXjhjHXQFUHtdCDX3A@public.gmane.org>
2012-08-04 13:55       ` Hillf Danton
2012-08-04 13:55         ` Hillf Danton
2012-08-02  2:53   ` [RFC PATCH 13/23 V2] page_alloc: use N_MEMORY instead N_HIGH_MEMORY change the node_states initialization Lai Jiangshan
2012-08-02  2:53   ` [RFC PATCH 14/23 V2] slub, hotplug: ignore unrelated node's hot-adding and hot-removing Lai Jiangshan
     [not found]     ` <1343875991-7533-15-git-send-email-laijs-BthXqXjhjHXQFUHtdCDX3A@public.gmane.org>
2012-08-02 16:04       ` Christoph Lameter
2012-08-02 16:04         ` Christoph Lameter
2012-08-02 16:04         ` Christoph Lameter
     [not found]         ` <alpine.DEB.2.00.1208021102560.23049-sBS69tsa9Uj/9pzu0YdTqQ@public.gmane.org>
2012-08-06  5:39           ` Lai Jiangshan [this message]
2012-08-06  5:39             ` Lai Jiangshan
2012-08-02  2:53   ` [RFC PATCH 15/23 V2] memory_hotplug: fix missing nodemask management Lai Jiangshan
2012-08-02  2:53   ` [RFC PATCH 16/23 V2] numa: add CONFIG_MOVABLE_NODE for movable-dedicated node Lai Jiangshan
     [not found]     ` <1343875991-7533-17-git-send-email-laijs-BthXqXjhjHXQFUHtdCDX3A@public.gmane.org>
2012-08-02 16:06       ` Christoph Lameter
2012-08-02 16:06     ` Christoph Lameter
2012-08-02 16:06       ` Christoph Lameter
2012-08-02  2:53   ` [RFC PATCH 17/23 V2] page_alloc.c: don't subtract unrelated memmap from zone's present pages Lai Jiangshan
2012-08-02  2:53   ` [RFC PATCH 18/23 V2] page_alloc: add kernelcore_max_addr Lai Jiangshan
2012-08-02  2:53   ` [RFC PATCH 19/23 V2] x86: get pg_data_t's memory from other node Lai Jiangshan
2012-08-02  2:53   ` [RFC PATCH 20/23 V2] x86: use memblock_set_current_limit() to set memblock.current_limit Lai Jiangshan
2012-08-02  2:53   ` [RFC PATCH 21/23 V2] memblock: limit memory address from memblock Lai Jiangshan
2012-08-02  2:53   ` [RFC PATCH 22/23 V2] memblock: compare current_limit with end variable at memblock_find_in_range_node() Lai Jiangshan
2012-08-02  2:53   ` [RFC PATCH 23/23 V2] mm, memory-hotplug: add online_movable Lai Jiangshan
  -- strict thread matches above, loose matches on Subject: below --
2012-08-02  6:01 [RFC PATCH 00/23 V2] memory,numa: introduce MOVABLE-dedicated node and online_movable for hotplug Lai Jiangshan
2012-08-02  6:01 ` [RFC PATCH 14/23 V2] slub, hotplug: ignore unrelated node's hot-adding and hot-removing Lai Jiangshan
2012-08-02  6:01   ` Lai Jiangshan

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=501F5898.8070502@cn.fujitsu.com \
    --to=laijs-bthxqxjhjhxqfuhtdcdx3a@public.gmane.org \
    --cc=a.p.zijlstra-/NLkJaSkS4VmR6Xm/wNWPw@public.gmane.org \
    --cc=cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=cl-vYTEC60ixJUAvxtiuMwx3w@public.gmane.org \
    --cc=dan.magenheimer-QHcLZuEGTsvQT0dZR+AlfA@public.gmane.org \
    --cc=gaowanlong-BthXqXjhjHXQFUHtdCDX3A@public.gmane.org \
    --cc=hpa-YMNOUZJC4hwAvxtiuMwx3w@public.gmane.org \
    --cc=hughd-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org \
    --cc=jkosina-AlSwsSmVLrQ@public.gmane.org \
    --cc=khlebnikov-GEFAQzZX7r8dnm+yROfE0A@public.gmane.org \
    --cc=kosaki.motohiro-+CUm20s59erQFUHtdCDX3A@public.gmane.org \
    --cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org \
    --cc=mgorman-l3A5Bk7waGM@public.gmane.org \
    --cc=mhocko-AlSwsSmVLrQ@public.gmane.org \
    --cc=mingo-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
    --cc=paul.gortmaker-CWA4WttNNZF54TAoqtyWWQ@public.gmane.org \
    --cc=pawel.moll-5wv7dgnIgG8@public.gmane.org \
    --cc=pholasek-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
    --cc=riel-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
    --cc=rientjes-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org \
    --cc=rusty-8n+1lVoiYb80n/F98K4Iww@public.gmane.org \
    --cc=sam-uyr5N9Q2VtJg9hUCZPvPmw@public.gmane.org \
    --cc=shangw-23VcF4HTsmIX0ybBhKVfKdBPR1lH4CV8@public.gmane.org \
    --cc=tixxdz-Umm1ozX2/EEdnm+yROfE0A@public.gmane.org \
    --cc=wency-BthXqXjhjHXQFUHtdCDX3A@public.gmane.org \
    --cc=x86-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.