mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Avi Norowitz <anorowitz@fortressitx.com>
To: linux-kernel@vger.kernel.org
Subject: High Memory Results in Poor Performance on Asus P4P800-MX Motherboard
Date: Tue, 14 Sep 2004 21:29:44 -0400	[thread overview]
Message-ID: <009f01c49ac3$7b04aa20$0100000a@FIRE> (raw)

Greetings,

At the dedicated hosting company I work for we have been having some serious
performance problems with Linux 2.6.8.1 on Pentium 4 servers running the
Asus P4P800-MX motherboard. Some of these servers run as if they were
Pentium 1s, while others run as if they were 386s. (I am not exaggerating.)

Eventually I found that this problem is resolved with a compiled kernel with
high memory disabled. Of course, this is not an attractive solution, since
many of our servers have over 1GB of memory. However, I came across this
solution on linux-kernel, archived at Google:

http://groups.google.com/groups?selm=1bD5H-1aI-9%40gated-at.bofh.it

This solution worked for servers with 1GB of memory, but not servers with
1.5GB. However, I was able to modify the commands so it worked with servers
with 1.5GB of memory, and presumably servers up to 4GB:

echo "disable=9" >| /proc/mtrr
echo "disable=8" >| /proc/mtrr
echo "disable=7" >| /proc/mtrr
echo "disable=6" >| /proc/mtrr
echo "disable=5" >| /proc/mtrr
echo "disable=4" >| /proc/mtrr
echo "disable=3" >| /proc/mtrr
echo "disable=2" >| /proc/mtrr
echo "disable=1" >| /proc/mtrr
echo "disable=0" >| /proc/mtrr
echo "base=0x00000000 size=0x100000000 type=write-back" > /proc/mtrr

I am very glad that I was able to resolve the problem, but I have three
further questions.

1. Is there a better fix to use than to have the above commands run at boot?

2. Which is at fault: the motherboard or the Linux kernel?

3. Should I expect this problem to be resolved in future kernels?

If any developers would like root access to a test server having this
problem to investigate the problem, please let me know.

Thank you for your time.

Sincerely,
Avi Norowitz
Unix Technical Support
Fortress ITX / DedicatedNOW / Pegasus Web Technologies


                 reply	other threads:[~2004-09-15  1:29 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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='009f01c49ac3$7b04aa20$0100000a@FIRE' \
    --to=anorowitz@fortressitx.com \
    --cc=linux-kernel@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

Powered by JetHome