From: Jesper Dangaard Brouer <brouer@redhat.com>
To: Eric Dumazet <eric.dumazet@gmail.com>
Cc: "David S. Miller" <davem@davemloft.net>,
Hannes Frederic Sowa <hannes@stressinduktion.org>,
netdev@vger.kernel.org
Subject: Re: [net-next PATCH 4/4] net: frag LRU list per CPU
Date: Thu, 25 Apr 2013 16:06:19 +0200 [thread overview]
Message-ID: <1366898779.26911.609.camel@localhost> (raw)
In-Reply-To: <1366855546.8964.125.camel@edumazet-glaptop>
On Wed, 2013-04-24 at 19:05 -0700, Eric Dumazet wrote:
> By the way, the frag_evictor() idea of cleaning 20% or 30% of the
> frags simply doesn't scale to thousands of fragments.
Yes I know, that's why I changed the mem limit in this patch to "only"
clean 128K (one max frag mem acct size).
> It adds huge latencies in softirq context.
Yes, it sucks. This is one thing I would really like to get rid of.
Does the "direct-hash-cleaning" solve this?
> If we really want to evict old fragments before expiration timer, then
> we can introduce a garbage collector in a work queue, and remove the
> need of a timer per fragment.
I like this idea (just don't know how to implement it). As my perf
results with "direct-hash-cleaning", show that we can hit a very
unfortunate situation, where the add/remove timer spinlock gets
hot/congested.
--Jesper
next prev parent reply other threads:[~2013-04-25 14:06 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-04-24 15:47 [net-next PATCH 0/4] net: frag patchset for fixing LRU scalability issue Jesper Dangaard Brouer
2013-04-24 15:48 ` [net-next PATCH 1/4] Revert "inet: limit length of fragment queue hash table bucket lists" Jesper Dangaard Brouer
2013-04-25 0:00 ` Eric Dumazet
2013-04-25 13:10 ` Jesper Dangaard Brouer
2013-04-25 13:58 ` David Laight
2013-05-02 7:59 ` Jesper Dangaard Brouer
2013-05-02 15:16 ` Eric Dumazet
2013-05-03 9:15 ` Jesper Dangaard Brouer
2013-04-24 15:48 ` [net-next PATCH 2/4] net: increase frag hash size Jesper Dangaard Brouer
2013-04-24 22:09 ` Sergei Shtylyov
2013-04-25 10:13 ` Jesper Dangaard Brouer
2013-04-25 12:13 ` Sergei Shtylyov
2013-04-25 19:11 ` David Miller
2013-04-24 23:48 ` Eric Dumazet
2013-04-25 3:26 ` Hannes Frederic Sowa
2013-04-25 19:52 ` [net-next PATCH V2] " Jesper Dangaard Brouer
2013-04-29 17:44 ` David Miller
2013-04-24 15:48 ` [net-next PATCH 3/4] net: avoid false perf interpretations in frag code Jesper Dangaard Brouer
2013-04-24 23:48 ` Eric Dumazet
2013-04-24 23:54 ` David Miller
2013-04-25 10:57 ` Jesper Dangaard Brouer
2013-04-25 19:13 ` David Miller
2013-04-24 15:48 ` [net-next PATCH 4/4] net: frag LRU list per CPU Jesper Dangaard Brouer
2013-04-25 0:25 ` Eric Dumazet
2013-04-25 2:05 ` Eric Dumazet
2013-04-25 14:06 ` Jesper Dangaard Brouer [this message]
2013-04-25 14:37 ` Eric Dumazet
2013-04-25 13:59 ` Jesper Dangaard Brouer
2013-04-25 14:10 ` Eric Dumazet
2013-04-25 14:18 ` Eric Dumazet
2013-04-25 19:15 ` Jesper Dangaard Brouer
2013-04-25 19:22 ` Eric Dumazet
2013-04-24 16:21 ` [net-next PATCH 0/4] net: frag patchset for fixing LRU scalabilityissue David Laight
2013-04-25 11:39 ` Jesper Dangaard Brouer
2013-04-25 12:57 ` David Laight
2013-04-24 17:27 ` [net-next PATCH 0/4] net: frag patchset for fixing LRU scalability issue Hannes Frederic Sowa
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=1366898779.26911.609.camel@localhost \
--to=brouer@redhat.com \
--cc=davem@davemloft.net \
--cc=eric.dumazet@gmail.com \
--cc=hannes@stressinduktion.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).