From: Eliezer Tamir <eliezer.tamir@linux.intel.com>
To: Eric Dumazet <eric.dumazet@gmail.com>
Cc: David Miller <davem@davemloft.net>,
linux-kernel@vger.kernel.org, netdev@vger.kernel.org,
Jesse Brandeburg <jesse.brandeburg@intel.com>,
Don Skidmore <donald.c.skidmore@intel.com>,
e1000-devel@lists.sourceforge.net,
Willem de Bruijn <willemb@google.com>,
Ben Hutchings <bhutchings@solarflare.com>,
Andi Kleen <andi@firstfloor.org>, HPA <hpa@zytor.com>,
Eilon Greenstien <eilong@broadcom.com>,
Or Gerlitz <or.gerlitz@gmail.com>,
Amir Vadai <amirv@mellanox.com>,
Eliezer Tamir <eliezer@tamir.org.il>
Subject: Re: [PATCH v9 net-next 5/7] net: simple poll/select low latency socket poll
Date: Wed, 05 Jun 2013 18:47:14 +0300 [thread overview]
Message-ID: <51AF5D82.1090308@linux.intel.com> (raw)
In-Reply-To: <1370445606.24311.265.camel@edumazet-glaptop>
On 05/06/2013 18:20, Eric Dumazet wrote:
> On Wed, 2013-06-05 at 16:41 +0300, Eliezer Tamir wrote:
>> On 05/06/2013 16:30, Eric Dumazet wrote:
>
>>> I am a bit uneasy with this one, because an applicatio polling() on one
>>> thousand file descriptors using select()/poll(), will call sk_poll_ll()
>>> one thousand times.
>>
>> But we call sk_poll_ll() with nonblock set, so it will only test once
>> for each socket and not loop.
>>
>> I think this is not as bad as it sounds.
>> We still honor the time limit on how long to poll.
>
> We still call ndo_ll_poll() a thousand times, and probably do a
> spinlock/unlock a thousand times in the driver.
>
> I would definitely be convinced if you give us some performance numbers
> of a poll() on a thousand tcp sockets for example.
So with 1000 sockets this is defiantly not a win
sockperf with 1000 udp sockets
sysctl 50 0
select 178.5 us / 130.0 us
poll 188.6 us / 130.0 us
next prev parent reply other threads:[~2013-06-05 15:47 UTC|newest]
Thread overview: 51+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-06-05 10:34 [PATCH v9 net-next 0/7] net: low latency Ethernet device polling Eliezer Tamir
2013-06-05 10:34 ` [PATCH v9 net-next 1/7] net: add napi_id and hash Eliezer Tamir
2013-06-05 10:34 ` Eliezer Tamir
2013-06-05 13:18 ` Eric Dumazet
2013-06-05 10:34 ` [PATCH v9 net-next 2/7] net: add low latency socket poll Eliezer Tamir
2013-06-05 10:34 ` Eliezer Tamir
2013-06-05 13:23 ` Eric Dumazet
2013-06-05 13:23 ` Eric Dumazet
2013-06-05 15:28 ` Willem de Bruijn
2013-06-05 15:31 ` Eliezer Tamir
2013-06-05 15:21 ` Eric Dumazet
2013-06-05 15:30 ` Eliezer Tamir
2013-06-05 15:39 ` Eric Dumazet
2013-06-05 15:39 ` Eric Dumazet
2013-06-05 15:46 ` Eliezer Tamir
2013-06-05 15:46 ` Eliezer Tamir
2013-06-05 15:59 ` Eric Dumazet
2013-06-05 15:59 ` Eric Dumazet
2013-06-06 12:50 ` Eliezer Tamir
2013-06-05 10:34 ` [PATCH v9 net-next 3/7] udp: add low latency socket poll support Eliezer Tamir
2013-06-05 10:34 ` Eliezer Tamir
2013-06-05 13:25 ` Eric Dumazet
2013-06-05 13:25 ` Eric Dumazet
2013-06-05 10:34 ` [PATCH v9 net-next 4/7] tcp: " Eliezer Tamir
2013-06-05 10:34 ` Eliezer Tamir
2013-06-05 13:25 ` Eric Dumazet
2013-06-05 13:25 ` Eric Dumazet
2013-06-05 10:34 ` [PATCH v9 net-next 5/7] net: simple poll/select low latency socket poll Eliezer Tamir
2013-06-05 10:34 ` Eliezer Tamir
2013-06-05 13:30 ` Eric Dumazet
2013-06-05 13:30 ` Eric Dumazet
2013-06-05 13:41 ` Eliezer Tamir
2013-06-05 13:41 ` Eliezer Tamir
2013-06-05 13:56 ` Eric Dumazet
2013-06-05 13:56 ` Eric Dumazet
2013-06-05 14:17 ` Eric Dumazet
2013-06-05 14:17 ` Eric Dumazet
2013-06-05 14:56 ` Eliezer Tamir
2013-06-05 15:20 ` Eric Dumazet
2013-06-05 15:20 ` Eric Dumazet
2013-06-05 15:47 ` Eliezer Tamir [this message]
2013-06-05 13:49 ` David Laight
2013-06-05 13:49 ` David Laight
2013-06-05 14:00 ` Eric Dumazet
2013-06-05 14:00 ` Eric Dumazet
2013-06-05 10:35 ` [PATCH v9 net-next 6/7] ixgbe: add support for ndo_ll_poll Eliezer Tamir
2013-06-05 10:35 ` Eliezer Tamir
2013-06-05 10:35 ` [PATCH v9 net-next 7/7] ixgbe: add extra stats " Eliezer Tamir
2013-06-05 10:35 ` Eliezer Tamir
2013-06-07 21:48 ` [PATCH v9 net-next 0/7] net: low latency Ethernet device polling David Miller
2013-06-08 18:06 ` Eliezer Tamir
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=51AF5D82.1090308@linux.intel.com \
--to=eliezer.tamir@linux.intel.com \
--cc=amirv@mellanox.com \
--cc=andi@firstfloor.org \
--cc=bhutchings@solarflare.com \
--cc=davem@davemloft.net \
--cc=donald.c.skidmore@intel.com \
--cc=e1000-devel@lists.sourceforge.net \
--cc=eilong@broadcom.com \
--cc=eliezer@tamir.org.il \
--cc=eric.dumazet@gmail.com \
--cc=hpa@zytor.com \
--cc=jesse.brandeburg@intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=or.gerlitz@gmail.com \
--cc=willemb@google.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.