From: Andrew Morton <akpm@linux-foundation.org>
To: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org,
cl@linux-foundation.org, tokunaga.keiich@jp.fujitsu.com,
kosaki.motohiro@jp.fujitsu.com
Subject: Re: [RFC][PATCH 1/2] Show quicklist at meminfo
Date: Wed, 20 Aug 2008 11:35:59 -0700 [thread overview]
Message-ID: <20080820113559.f559a411.akpm@linux-foundation.org> (raw)
In-Reply-To: <20080820200607.12ED.KOSAKI.MOTOHIRO@jp.fujitsu.com>
On Wed, 20 Aug 2008 20:07:06 +0900
KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> wrote:
> Now, Quicklist can spent several GB memory.
> So, if end user can't hou much spent memory, he misunderstand to memory leak happend.
>
>
> after this patch applied, /proc/meminfo output following.
>
> % cat /proc/meminfo
>
> MemTotal: 7701504 kB
> MemFree: 5159040 kB
> Buffers: 112960 kB
> Cached: 337536 kB
> SwapCached: 0 kB
> Active: 218944 kB
> Inactive: 350848 kB
> Active(anon): 120832 kB
> Inactive(anon): 0 kB
> Active(file): 98112 kB
> Inactive(file): 350848 kB
> Unevictable: 0 kB
> Mlocked: 0 kB
> SwapTotal: 2031488 kB
> SwapFree: 2031488 kB
> Dirty: 320 kB
> Writeback: 0 kB
> AnonPages: 119488 kB
> Mapped: 38528 kB
> Slab: 1595712 kB
> SReclaimable: 23744 kB
> SUnreclaim: 1571968 kB
> PageTables: 14336 kB
> NFS_Unstable: 0 kB
> Bounce: 0 kB
> WritebackTmp: 0 kB
> CommitLimit: 5882240 kB
> Committed_AS: 356672 kB
> VmallocTotal: 17592177655808 kB
> VmallocUsed: 29056 kB
> VmallocChunk: 17592177626304 kB
> Quicklists: 283776 kB
> HugePages_Total: 0
> HugePages_Free: 0
> HugePages_Rsvd: 0
> HugePages_Surp: 0
> Hugepagesize: 262144 kB
>
> ...
>
> K(committed),
> (unsigned long)VMALLOC_TOTAL >> 10,
> vmi.used >> 10,
> - vmi.largest_chunk >> 10
> + vmi.largest_chunk >> 10,
> + K(quicklist_total_size())
> );
quicklist_total_size() is racy against cpu hotplug. That's OK for
/proc/meminfo purposes (occasional transient inaccuracy?), but will it
crash? Not in the current implementation of per_cpu() afaict, but it
might crash if we ever teach cpu hotunplug to free up the percpu
resources.
I see no cpu hotplug handling in the quicklist code. Do we leak all
the hot-unplugged CPU's pages?
WARNING: multiple messages have this Message-ID (diff)
From: Andrew Morton <akpm@linux-foundation.org>
To: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org,
cl@linux-foundation.org, tokunaga.keiich@jp.fujitsu.com
Subject: Re: [RFC][PATCH 1/2] Show quicklist at meminfo
Date: Wed, 20 Aug 2008 11:35:59 -0700 [thread overview]
Message-ID: <20080820113559.f559a411.akpm@linux-foundation.org> (raw)
In-Reply-To: <20080820200607.12ED.KOSAKI.MOTOHIRO@jp.fujitsu.com>
On Wed, 20 Aug 2008 20:07:06 +0900
KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> wrote:
> Now, Quicklist can spent several GB memory.
> So, if end user can't hou much spent memory, he misunderstand to memory leak happend.
>
>
> after this patch applied, /proc/meminfo output following.
>
> % cat /proc/meminfo
>
> MemTotal: 7701504 kB
> MemFree: 5159040 kB
> Buffers: 112960 kB
> Cached: 337536 kB
> SwapCached: 0 kB
> Active: 218944 kB
> Inactive: 350848 kB
> Active(anon): 120832 kB
> Inactive(anon): 0 kB
> Active(file): 98112 kB
> Inactive(file): 350848 kB
> Unevictable: 0 kB
> Mlocked: 0 kB
> SwapTotal: 2031488 kB
> SwapFree: 2031488 kB
> Dirty: 320 kB
> Writeback: 0 kB
> AnonPages: 119488 kB
> Mapped: 38528 kB
> Slab: 1595712 kB
> SReclaimable: 23744 kB
> SUnreclaim: 1571968 kB
> PageTables: 14336 kB
> NFS_Unstable: 0 kB
> Bounce: 0 kB
> WritebackTmp: 0 kB
> CommitLimit: 5882240 kB
> Committed_AS: 356672 kB
> VmallocTotal: 17592177655808 kB
> VmallocUsed: 29056 kB
> VmallocChunk: 17592177626304 kB
> Quicklists: 283776 kB
> HugePages_Total: 0
> HugePages_Free: 0
> HugePages_Rsvd: 0
> HugePages_Surp: 0
> Hugepagesize: 262144 kB
>
> ...
>
> K(committed),
> (unsigned long)VMALLOC_TOTAL >> 10,
> vmi.used >> 10,
> - vmi.largest_chunk >> 10
> + vmi.largest_chunk >> 10,
> + K(quicklist_total_size())
> );
quicklist_total_size() is racy against cpu hotplug. That's OK for
/proc/meminfo purposes (occasional transient inaccuracy?), but will it
crash? Not in the current implementation of per_cpu() afaict, but it
might crash if we ever teach cpu hotunplug to free up the percpu
resources.
I see no cpu hotplug handling in the quicklist code. Do we leak all
the hot-unplugged CPU's pages?
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2008-08-20 18:36 UTC|newest]
Thread overview: 86+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-08-20 11:05 [RFC][PATCH 0/2] Quicklist is slighly problematic KOSAKI Motohiro
2008-08-20 11:05 ` KOSAKI Motohiro
2008-08-20 11:07 ` [RFC][PATCH 1/2] Show quicklist at meminfo KOSAKI Motohiro
2008-08-20 11:07 ` KOSAKI Motohiro
2008-08-20 18:35 ` Andrew Morton [this message]
2008-08-20 18:35 ` Andrew Morton
2008-08-21 7:36 ` KOSAKI Motohiro
2008-08-21 7:36 ` KOSAKI Motohiro
2008-08-22 1:05 ` KOSAKI Motohiro
2008-08-22 1:05 ` KOSAKI Motohiro
2008-08-22 4:28 ` Andrew Morton
2008-08-22 4:28 ` Andrew Morton
2008-08-22 13:23 ` Robin Holt
2008-08-22 13:23 ` Robin Holt
2008-08-22 13:56 ` Christoph Lameter
2008-08-22 13:56 ` Christoph Lameter
2008-08-23 8:24 ` KOSAKI Motohiro
2008-08-23 8:24 ` KOSAKI Motohiro
2008-08-24 5:29 ` Andrew Morton
2008-08-24 5:29 ` Andrew Morton
2008-08-20 11:08 ` [RFC][PATCH 2/2] quicklist shouldn't be proportional to # of CPUs KOSAKI Motohiro
2008-08-20 11:08 ` KOSAKI Motohiro
2008-08-20 15:27 ` Christoph Lameter
2008-08-20 15:27 ` Christoph Lameter
2008-08-21 6:46 ` Andrew Morton
2008-08-21 6:46 ` Andrew Morton
2008-08-21 7:13 ` David Miller
2008-08-21 7:13 ` David Miller, Andrew Morton
2008-08-21 7:18 ` KOSAKI Motohiro
2008-08-21 7:18 ` KOSAKI Motohiro
2008-08-21 7:27 ` Andrew Morton
2008-08-21 7:27 ` Andrew Morton
2008-08-21 7:31 ` KOSAKI Motohiro
2008-08-21 7:31 ` KOSAKI Motohiro
2008-08-21 9:32 ` Peter Zijlstra
2008-08-21 9:32 ` Peter Zijlstra
2008-08-21 10:04 ` KOSAKI Motohiro
2008-08-21 10:04 ` KOSAKI Motohiro
2008-08-21 10:09 ` David Miller
2008-08-21 10:09 ` David Miller, KOSAKI Motohiro
2008-08-21 10:13 ` KOSAKI Motohiro
2008-08-21 10:13 ` KOSAKI Motohiro
2008-08-21 10:26 ` David Miller
2008-08-21 10:26 ` David Miller, KOSAKI Motohiro
2008-08-21 10:22 ` KOSAKI Motohiro
2008-08-21 10:22 ` KOSAKI Motohiro
2008-08-21 12:02 ` KOSAKI Motohiro
2008-08-21 12:02 ` KOSAKI Motohiro
2008-08-25 18:48 ` Mike Travis
2008-08-25 18:48 ` Mike Travis
2008-08-25 23:33 ` KOSAKI Motohiro
2008-08-25 23:33 ` KOSAKI Motohiro
2008-08-26 20:35 ` Mike Travis
2008-08-26 20:35 ` Mike Travis
2008-08-25 18:44 ` Mike Travis
2008-08-25 18:44 ` Mike Travis
2008-08-25 18:40 ` Mike Travis
2008-08-25 18:40 ` Mike Travis
2008-08-25 23:31 ` KOSAKI Motohiro
2008-08-25 23:31 ` KOSAKI Motohiro
2008-08-20 14:10 ` [RFC][PATCH 0/2] Quicklist is slighly problematic Christoph Lameter
2008-08-20 14:10 ` Christoph Lameter
2008-08-20 14:49 ` KOSAKI Motohiro
2008-08-20 14:49 ` KOSAKI Motohiro
2008-08-20 15:26 ` Christoph Lameter
2008-08-20 15:26 ` Christoph Lameter
2008-08-21 2:13 ` Robin Holt
2008-08-21 2:13 ` Robin Holt
2008-08-21 2:16 ` Robin Holt
2008-08-21 2:16 ` Robin Holt
2008-08-21 3:08 ` David Miller
2008-08-21 3:08 ` David Miller, Robin Holt
2008-08-21 13:10 ` Christoph Lameter
2008-08-21 13:10 ` Christoph Lameter
2008-08-20 18:31 ` Andrew Morton
2008-08-20 18:31 ` Andrew Morton
2008-08-21 2:42 ` Robin Holt
2008-08-21 2:42 ` Robin Holt
2008-08-21 13:07 ` Christoph Lameter
2008-08-21 13:07 ` Christoph Lameter
2008-08-21 13:14 ` Robin Holt
2008-08-21 13:14 ` Robin Holt
2008-08-21 13:18 ` Christoph Lameter
2008-08-21 13:18 ` Christoph Lameter
2008-08-21 13:45 ` Robin Holt
2008-08-21 13:45 ` Robin Holt
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=20080820113559.f559a411.akpm@linux-foundation.org \
--to=akpm@linux-foundation.org \
--cc=cl@linux-foundation.org \
--cc=kosaki.motohiro@jp.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=tokunaga.keiich@jp.fujitsu.com \
/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.