From: Eric Dumazet <eric.dumazet@gmail.com>
To: Alexander Duyck <alexander.h.duyck@redhat.com>
Cc: David Laight <David.Laight@ACULAB.COM>,
"Jiafei.Pan@freescale.com" <Jiafei.Pan@freescale.com>,
David Miller <davem@davemloft.net>,
"jkosina@suse.cz" <jkosina@suse.cz>,
"netdev@vger.kernel.org" <netdev@vger.kernel.org>,
"LeoLi@freescale.com" <LeoLi@freescale.com>,
"linux-doc@vger.kernel.org" <linux-doc@vger.kernel.org>
Subject: Re: [PATCH] net: use hardware buffer pool to allocate skb
Date: Fri, 17 Oct 2014 12:02:19 -0700 [thread overview]
Message-ID: <1413572539.25949.17.camel@edumazet-glaptop2.roam.corp.google.com> (raw)
In-Reply-To: <54415FEA.5000705@redhat.com>
On Fri, 2014-10-17 at 11:28 -0700, Alexander Duyck wrote:
> So long as it doesn't impact performance significantly I am fine with
> it. My concern is that you are bringing up issues that none of the
> customers were brining up when I was at Intel, and the fixes you are
> proposing are likely to result in customers seeing things they will
> report as issues.
We regularly hit these issues at Google.
We have memory containers, and we detect quite easily if some layer is
lying, because we cant afford having 20% of headroom on our servers.
I am not claiming IGB is the only offender.
I am sorry if you believe it was an attack on IGB, or any network
driver.
truesize should really be the thing that protects us from OOM,
and apparently driver authors hitting TCP performance problems
thinks it is better to simply let TCP do no garbage collection.
It seems that nobody cares or even understand what I am saying,
so I should probably not care and suggest Google to buy PetaBytes of
memory, right ?
next prev parent reply other threads:[~2014-10-17 19:02 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-10-15 3:26 [PATCH] net: use hardware buffer pool to allocate skb Pan Jiafei
2014-10-15 4:15 ` Eric Dumazet
2014-10-15 4:26 ` David Miller
2014-10-15 5:43 ` Jiafei.Pan
2014-10-15 9:15 ` Eric Dumazet
2014-10-15 4:25 ` David Miller
2014-10-15 5:34 ` Jiafei.Pan
2014-10-15 9:15 ` Eric Dumazet
2014-10-16 2:17 ` Jiafei.Pan
2014-10-16 4:15 ` Eric Dumazet
2014-10-16 5:15 ` Jiafei.Pan
2014-10-16 15:28 ` Alexander Duyck
2014-10-16 16:57 ` Eric Dumazet
2014-10-16 17:10 ` Alexander Duyck
2014-10-16 17:45 ` Eric Dumazet
2014-10-16 18:20 ` Alexander Duyck
2014-10-16 21:40 ` Eric Dumazet
2014-10-16 22:12 ` Alexander Duyck
2014-10-17 9:11 ` David Laight
2014-10-17 14:40 ` Alexander Duyck
2014-10-17 16:55 ` Eric Dumazet
2014-10-17 18:28 ` Alexander Duyck
2014-10-17 18:53 ` Eric Dumazet
2014-10-18 0:26 ` Eric Dumazet
2014-10-17 19:02 ` Eric Dumazet [this message]
2014-10-17 19:38 ` Alexander Duyck
2014-10-17 19:51 ` Eric Dumazet
2014-10-17 22:13 ` Alexander Duyck
2014-10-17 2:35 ` Jiafei.Pan
2014-10-17 14:05 ` Eric Dumazet
2014-10-17 14:12 ` Alexander Duyck
2014-10-16 2:17 ` Jiafei.Pan
2014-10-15 15:51 ` David Miller
2014-10-15 4:59 ` Oliver Hartkopp
2014-10-15 5:47 ` Jiafei.Pan
2014-10-15 8:57 ` David Laight
2014-10-15 9:33 ` Stephen Hemminger
2014-10-16 2:30 ` Jiafei.Pan
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=1413572539.25949.17.camel@edumazet-glaptop2.roam.corp.google.com \
--to=eric.dumazet@gmail.com \
--cc=David.Laight@ACULAB.COM \
--cc=Jiafei.Pan@freescale.com \
--cc=LeoLi@freescale.com \
--cc=alexander.h.duyck@redhat.com \
--cc=davem@davemloft.net \
--cc=jkosina@suse.cz \
--cc=linux-doc@vger.kernel.org \
--cc=netdev@vger.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;
as well as URLs for NNTP newsgroup(s).