From: jamal <hadi@cyberus.ca>
To: Andi Kleen <ak@muc.de>
Cc: Greg Banks <gnb@sgi.com>, Arthur Kepner <akepner@sgi.com>,
"Brandeburg, Jesse" <jesse.brandeburg@intel.com>,
netdev@oss.sgi.com, davem@redhat.com
Subject: Re: NAPI, e100, and system performance problem
Date: Fri, 22 Apr 2005 08:33:15 -0400 [thread overview]
Message-ID: <1114173195.7679.30.camel@localhost.localdomain> (raw)
In-Reply-To: <m1hdhzyrdz.fsf@muc.de>
On Fri, 2005-22-04 at 13:36 +0200, Andi Kleen wrote:
> Greg Banks <gnb@sgi.com> writes:
> >
> > An inordinate amount of CPU is being spent running around polling the
> > device instead of dealing with the packets in IP, TCP and NFS land.
> > By inordinate, we mean twice as much or more cpu% than a MIPS/Irix
> > box with slower CPUs.
>
> We have seen similar behaviour. With NAPI some benchmarks run
> a lot slower than on a driver on the same hardware/NIC without NAPI.
They should not run slower - but they may consume more CPU.
> This can be even observed with simple tests like netperf single stream
> between two boxes.
>
Yes, slow traffic coming into the system would chew more CPU if you have
a fast CPU ;-> You should know this Andi, but let me explain the reason
for about the 100th time:
If your CPU is fast enough such that the interupt only causes 1-2
packets to be processed before reenabling, then you will have anywhere
between 1-2 extra IO calls (recall dis/reenabling interupts) - this will
be noticed in the CPU % chewed. It shouldnt be really a lot of CPU.
If you use non-NAPI then you dont have the extra IO happening and
therefore you will see less CPU used at those rates.
OTOH, Try more than 1 netperfs and it becomes beneficial to have NAPI
around.
Yes, for low rates - this will make some of those benchmarks look bad.
But for such cases you can use mitigation in addition to NAPi - except
your latency in the benchmark will now look bad ;->
> There seems to be also some problems with bidirectional traffic, although
> I have not fully tracked them down to NAPI yet.
>
doubt it has to do with NAPI.
> There is definitely some problem in NAPI land ...
>
this is a design choice - a solution could be created to "fix" this but
hasnt happened because there has not been a good reason to complicate
things. The people who are bitching about this are benchmarkers who want
to win at both high and low rates and they are not happy because while
they can win at high rates, they cant at low rates.
cheers,
jamal
next prev parent reply other threads:[~2005-04-22 12:33 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-04-18 6:11 NAPI, e100, and system performance problem Brandeburg, Jesse
2005-04-18 12:14 ` jamal
2005-04-18 15:36 ` Robert Olsson
2005-04-18 16:55 ` Arthur Kepner
2005-04-18 19:34 ` Robert Olsson
2005-04-18 20:26 ` jamal
2005-04-19 5:55 ` Greg Banks
2005-04-19 18:36 ` David S. Miller
2005-04-19 20:38 ` NAPI and CPU utilization [was: NAPI, e100, and system performance problem] Arthur Kepner
2005-04-19 20:52 ` Rick Jones
2005-04-19 21:09 ` David S. Miller
[not found] ` <20050420145629.GH19415@sgi.com>
2005-04-20 15:15 ` NAPI, e100, and system performance problem jamal
2005-04-22 11:36 ` Andi Kleen
2005-04-22 12:33 ` jamal [this message]
2005-04-22 17:21 ` Andi Kleen
2005-04-22 18:18 ` jamal
2005-04-22 18:30 ` Andi Kleen
2005-04-22 18:37 ` Arthur Kepner
2005-04-22 18:52 ` David S. Miller
[not found] ` <Pine.LNX.4.61.0504241845070.2934@linux.site>
2005-04-25 11:25 ` jamal
2005-04-25 18:51 ` David S. Miller
2005-04-25 11:41 ` jamal
2005-04-25 12:16 ` Jamal Hadi Salim
2005-04-22 19:01 ` jamal
2005-04-22 19:07 ` David S. Miller
2005-04-22 19:21 ` jamal
2005-04-23 20:50 ` Robert Olsson
2005-04-23 16:56 ` Robert Olsson
2005-04-22 23:28 ` Greg Banks
2005-04-22 23:40 ` Stephen Hemminger
2005-04-22 23:43 ` David S. Miller
2005-04-23 2:51 ` Stephen Hemminger
2005-04-23 17:54 ` Robert Olsson
2005-04-23 3:04 ` jamal
2005-04-23 17:14 ` Robert Olsson
2005-04-22 14:52 ` Robert Olsson
2005-04-22 15:37 ` jamal
2005-04-22 17:22 ` Andi Kleen
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=1114173195.7679.30.camel@localhost.localdomain \
--to=hadi@cyberus.ca \
--cc=ak@muc.de \
--cc=akepner@sgi.com \
--cc=davem@redhat.com \
--cc=gnb@sgi.com \
--cc=jesse.brandeburg@intel.com \
--cc=netdev@oss.sgi.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.