netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Elad Lahav <elahav@uwaterloo.ca>
To: "Brandeburg, Jesse" <jesse.brandeburg@intel.com>
Cc: e1000-devel@lists.sourceforge.net, netdev@vger.kernel.org,
	Elad Lahav <elad_lahav@users.sourceforge.net>
Subject: Re: Performance degradration 2.6.25->2.6.26
Date: Fri, 12 Sep 2008 10:58:37 -0400	[thread overview]
Message-ID: <48CA839D.308@uwaterloo.ca> (raw)
In-Reply-To: <alpine.WNT.1.10.0808080906590.5428@jbrandeb-MOBL1.amr.corp.intel.com>

>> HyperThreading, for a total of 4 logical processors, and 4 Intel Gigabit
>>   NICs on a PCI 64/66 bus (82546EB). In the experiments, I am pinning
>> four sending processes (one per NIC) to logical processors 0 and 2, and
>> the NIC interrupts to logical processors 1 and 3, such that interrupts
>> are serviced by the sibling logical processor of the one sending the
>> packets. I have used both the driver that comes with the Linux kernel
>> (7.3.20-k2-NAPI), as well as a more recent one (7.6.15.5).

I believe I have found the problem, and it has nothing to do with the driver.
Up to the latest kernel, logical processors on the same chip had sequential CPU IDs. So 
CPU 0 and CPU 1 were two logical processors on the first physical one, 2 and 3 on the 
second physical one, etc.
It seems that now this ordering has changed. I was running my experiments on logical 
processors 0 and 2, assuming they were on different physical processors, when, in fact, 
they were sharing the same one. When pinning the processes to logical processor 0 and 1, I 
get at least as good a performance as I do on previous kernels.

Thanks,
Elad


-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/

      parent reply	other threads:[~2008-09-12 14:58 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <489C0F2C.9040101@users.sourceforge.net>
2008-08-08 16:23 ` Performance degradration 2.6.25->2.6.26 Brandeburg, Jesse
2008-08-10  8:27   ` Elad Lahav
2008-08-10  9:05   ` Elad Lahav
2008-09-12 14:58   ` Elad Lahav [this message]

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=48CA839D.308@uwaterloo.ca \
    --to=elahav@uwaterloo.ca \
    --cc=e1000-devel@lists.sourceforge.net \
    --cc=elad_lahav@users.sourceforge.net \
    --cc=jesse.brandeburg@intel.com \
    --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).