From: Stephen Hemminger <shemminger@vyatta.com>
To: Vladislav Bolkhovitin <vst@vlnb.net>
Cc: "James Bottomley" <James.Bottomley@suse.de>,
"Lukas Kolbe" <lkolbe@techfak.uni-bielefeld.de>,
"Kai Mäkisara" <kai.makisara@kolumbus.fi>,
"FUJITA Tomonori" <fujita.tomonori@lab.ntt.co.jp>,
linux-scsi@vger.kernel.org,
"Kashyap Desai" <Kashyap.Desai@lsi.com>,
netdev@vger.kernel.org
Subject: Re: After memory pressure: can't read from tape anymore
Date: Tue, 14 Dec 2010 14:23:20 -0800 [thread overview]
Message-ID: <20101214142320.27d911d5@nehalam> (raw)
In-Reply-To: <4D07D519.70800@vlnb.net>
On Tue, 14 Dec 2010 23:35:37 +0300
Vladislav Bolkhovitin <vst@vlnb.net> wrote:
> What is interesting to me in this regard is how networking with 9K jumbo
> frames manages to work acceptably reliable? Jumbo frames used
> sufficiently often, including under high memory pressure.
>
> I'm not a deep networking guru, but network drivers need to allocate
> physically continual memory for skbs, which means 16K per 9K packet,
> which means order 2 allocations per skb.
Good network drivers support fragmentation and allocate a small portion
for the header and allocate pages for the rest. This requires no higher
order allocation. The networking stack takes fragmented data coming
in and does the necessary copy/merging to access contiguous headers.
There are still some crap network drivers that require large contiguous
allocation. These should not be used with jumbo frames in real
environments.
--
next prev parent reply other threads:[~2010-12-14 22:23 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-11-28 19:15 After memory pressure: can't read from tape anymore Lukas Kolbe
2010-11-29 17:09 ` Kai Makisara
2010-11-30 13:31 ` Lukas Kolbe
2010-11-30 16:10 ` Boaz Harrosh
2010-11-30 16:23 ` Kai Makisara
2010-11-30 16:44 ` Boaz Harrosh
2010-11-30 17:04 ` Kai Makisara
2010-11-30 17:24 ` Boaz Harrosh
2010-11-30 19:53 ` Kai Makisara
2010-12-01 9:40 ` Lukas Kolbe
2010-12-02 11:17 ` Desai, Kashyap
2010-12-02 16:22 ` Kai Makisara
2010-12-02 18:14 ` Desai, Kashyap
2010-12-02 20:25 ` Kai Makisara
2010-12-05 10:44 ` Lukas Kolbe
2010-12-03 10:13 ` FUJITA Tomonori
2010-12-03 10:45 ` Desai, Kashyap
2010-12-03 11:11 ` FUJITA Tomonori
2010-12-02 10:01 ` Lukas Kolbe
2010-12-03 9:44 ` FUJITA Tomonori
2010-11-30 16:20 ` Kai Makisara
2010-12-01 17:06 ` Lukas Kolbe
2010-12-02 16:41 ` Kai Makisara
2010-12-06 7:59 ` Kai Makisara
2010-12-06 8:50 ` FUJITA Tomonori
2010-12-06 9:36 ` Lukas Kolbe
2010-12-06 11:34 ` Bjørn Mork
2010-12-08 14:19 ` Lukas Kolbe
2010-12-03 12:27 ` FUJITA Tomonori
2010-12-03 14:59 ` Kai Mäkisara
2010-12-03 15:06 ` James Bottomley
2010-12-03 17:03 ` Lukas Kolbe
2010-12-03 18:10 ` James Bottomley
2010-12-05 10:53 ` Lukas Kolbe
2010-12-05 12:16 ` FUJITA Tomonori
2010-12-14 20:35 ` Vladislav Bolkhovitin
2010-12-14 22:23 ` Stephen Hemminger [this message]
2010-12-15 16:27 ` Vladislav Bolkhovitin
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=20101214142320.27d911d5@nehalam \
--to=shemminger@vyatta.com \
--cc=James.Bottomley@suse.de \
--cc=Kashyap.Desai@lsi.com \
--cc=fujita.tomonori@lab.ntt.co.jp \
--cc=kai.makisara@kolumbus.fi \
--cc=linux-scsi@vger.kernel.org \
--cc=lkolbe@techfak.uni-bielefeld.de \
--cc=netdev@vger.kernel.org \
--cc=vst@vlnb.net \
/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.