From: Russell King <rmk+lkml@arm.linux.org.uk>
To: "David S. Miller" <davem@davemloft.net>,
Robert.Olsson@data.slu.se, akpm@osdl.org, torvalds@osdl.org,
alexn@dsv.su.se, kas@fi.muni.cz, linux-kernel@vger.kernel.org,
netdev@oss.sgi.com
Subject: Re: Memory leak in 2.6.11-rc1?
Date: Sun, 30 Jan 2005 15:34:49 +0000 [thread overview]
Message-ID: <20050130153449.B25000@flint.arm.linux.org.uk> (raw)
In-Reply-To: <20050130132343.A25000@flint.arm.linux.org.uk>; from rmk+lkml@arm.linux.org.uk on Sun, Jan 30, 2005 at 01:23:43PM +0000
On Sun, Jan 30, 2005 at 01:23:43PM +0000, Russell King wrote:
> Anyway, I've produced some code which keeps a record of the __refcnt
> increments and decrements, and I think it's produced some interesting
> results. Essentially, I'm seeing the odd dst entry with a __refcnt of
> 14000 or so (which is still in active use, so probably ok), and a number
> with 4, 7, and 13 which haven't had the refcount touched for at least 14
> minutes.
An hour or so goes by. I now have 14 dst cache entries with non-zero
refcounts, and these have the following properties:
* The five from before (with counts 13, 14473, 4, 4, 7 respectively):
+ all remain unfreed.
+ show precisely no change in the refcounts.
+ the refcount has not been touched for more than an hour.
* They have all been touched by ip_copy_metadata.
* Their remaining refcounts are precisely half the number of
ip_copy_metadata calls in every instance.
No entries with a refcount of zero contain ip_copy_metadata() and do
appear in /proc/net/rt_cache.
The following may also be a pointer - from /proc/net/snmp:
Ip: Forwarding DefaultTTL InReceives InHdrErrors InAddrErrors ForwDatagrams InUnknownProtos InDiscards InDelivers OutRequests OutDiscards OutNoRoutes ReasmTimeout ReasmReqds ReasmOKs ReasmFails FragOKs FragFails FragCreates
Ip: 1 64 140510 0 0 36861 0 0 93549 131703 485 0 21 46622 15695 21 21950 0 0
Since FragCreates is 0, this means that we are using the frag_lists
rather than creating our own fragments (and indeed the first
ip_copy_metadata() call rather than the second in ip_fragment()).
I think the case against the IPv4 fragmentation code is mounting.
However, without knowing what the expected conditions for this code,
(eg, are skbs on the fraglist supposed to have NULL skb->dst?) I'm
unable to progress this any further. However, I think it's quite
clear that there is something bad going on here.
Why many more people aren't seeing this I've no idea.
--
Russell King
Linux kernel 2.6 ARM Linux - http://www.arm.linux.org.uk/
maintainer of: 2.6 PCMCIA - http://pcmcia.arm.linux.org.uk/
2.6 Serial core
next prev parent reply other threads:[~2005-01-30 15:34 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20050121161959.GO3922@fi.muni.cz>
[not found] ` <1106360639.15804.1.camel@boxen>
[not found] ` <20050123091154.GC16648@suse.de>
[not found] ` <20050123011918.295db8e8.akpm@osdl.org>
[not found] ` <20050123095608.GD16648@suse.de>
[not found] ` <20050123023248.263daca9.akpm@osdl.org>
2005-01-23 20:03 ` Memory leak in 2.6.11-rc1? Russell King
2005-01-24 11:48 ` Russell King
2005-01-25 19:32 ` Russell King
2005-01-27 8:28 ` Russell King
2005-01-27 8:47 ` Andrew Morton
2005-01-27 10:19 ` Alessandro Suardi
2005-01-27 12:17 ` Martin Josefsson
2005-01-27 12:56 ` Robert Olsson
2005-01-27 13:03 ` Robert Olsson
2005-01-27 16:49 ` Russell King
2005-01-27 18:37 ` Phil Oester
2005-01-27 19:25 ` Russell King
2005-01-27 20:40 ` Phil Oester
2005-01-28 9:32 ` Russell King
2005-01-27 20:33 ` David S. Miller
2005-01-28 0:17 ` Russell King
2005-01-28 0:34 ` David S. Miller
2005-01-28 8:58 ` Russell King
2005-01-30 13:23 ` Russell King
2005-01-30 15:34 ` Russell King [this message]
2005-01-30 16:57 ` Phil Oester
2005-01-30 17:23 ` Patrick McHardy
2005-01-30 17:26 ` Patrick McHardy
2005-01-30 17:58 ` Patrick McHardy
2005-01-30 18:45 ` Russell King
2005-01-31 2:48 ` David S. Miller
2005-01-31 4:11 ` Herbert Xu
2005-01-31 4:45 ` YOSHIFUJI Hideaki / 吉藤英明
2005-01-31 5:00 ` Patrick McHardy
2005-01-31 5:11 ` David S. Miller
2005-01-31 5:40 ` Herbert Xu
2005-01-31 5:16 ` YOSHIFUJI Hideaki / 吉藤英明
2005-01-31 5:42 ` Yasuyuki KOZAKAI
2005-01-30 18:01 ` Russell King
2005-01-30 18:19 ` Phil Oester
2005-01-28 1:41 ` Phil Oester
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=20050130153449.B25000@flint.arm.linux.org.uk \
--to=rmk+lkml@arm.linux.org.uk \
--cc=Robert.Olsson@data.slu.se \
--cc=akpm@osdl.org \
--cc=alexn@dsv.su.se \
--cc=davem@davemloft.net \
--cc=kas@fi.muni.cz \
--cc=linux-kernel@vger.kernel.org \
--cc=netdev@oss.sgi.com \
--cc=torvalds@osdl.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;
as well as URLs for NNTP newsgroup(s).