All of lore.kernel.org
 help / color / mirror / Atom feed
From: Nebojsa Trpkovic <trx.lists@gmail.com>
To: cpufreq@www.linux.org.uk
Subject: acpi-cpufreq dissapires when FSB is changed
Date: Thu, 18 Oct 2007 17:45:51 +0200	[thread overview]
Message-ID: <47177FAF.2070206@gmail.com> (raw)

kernel 2.6.23
intel p35 board
core2duo e6750 (6x333=2000MHz, 8x333=2667MHz)

when it runs on default FSB (333MHz) it looks like this:
uname -a
Linux tnt 2.6.23-gentoo #3 SMP Sun Oct 21 15:37:59 CEST 2007 x86_64
Intel(R) Core(TM)2 Duo CPU E6750 @ 2.66GHz GenuineIntel GNU/Linux
ls -lah /sys/devices/system/cpu/cpu0/cpufreq/
total 0
drwxr-xr-x 4 root root    0 Oct 18 17:19 .
drwxr-xr-x 6 root root    0 Oct 18 17:19 ..
-r--r--r-- 1 root root 4.0K Oct 18 17:24 affected_cpus
-r--r--r-- 1 root root 4.0K Oct 18 17:24 cpuinfo_max_freq
-r--r--r-- 1 root root 4.0K Oct 18 17:24 cpuinfo_min_freq
drwxr-xr-x 2 root root    0 Oct 18 17:24 ondemand
-r--r--r-- 1 root root 4.0K Oct 18 17:24 scaling_available_frequencies
-r--r--r-- 1 root root 4.0K Oct 18 17:24 scaling_available_governors
-r--r--r-- 1 root root 4.0K Oct 18 17:24 scaling_cur_freq
-r--r--r-- 1 root root 4.0K Oct 18 17:24 scaling_driver
-rw-r--r-- 1 root root    0 Oct 18 17:19 scaling_governor
-rw-r--r-- 1 root root 4.0K Oct 18 17:24 scaling_max_freq
-rw-r--r-- 1 root root 4.0K Oct 18 17:24 scaling_min_freq
drwxr-xr-x 2 root root    0 Oct 18 17:24 stats
cat /proc/cpuinfo |grep MHz
cpu MHz         : 2000.000
cpu MHz         : 2000.000

when FSB is changed (400MHz) it looks like this:
uname -a
Linux tnt 2.6.23-gentoo #3 SMP Sun Oct 21 15:37:59 CEST 2007 x86_64
Intel(R) Core(TM)2 Duo CPU E6750 @ 2.66GHz GenuineIntel GNU/Linux
ls -lah /sys/devices/system/cpu/cpu0/cpufreq/
ls: cannot access /sys/devices/system/cpu/cpu0/cpufreq/: No such file or
directory
cat /proc/cpuinfo |grep MHz
cpu MHz         : 3200.000
cpu MHz         : 3200.000

the only thing that was changed in BIOS is FSB freq. no power-saving
options were changed.

so, in other operating systems frequency scaling continue to work using
the same multiplier-voltage data (for e6750 it's 1.15V for 6x and 1.35V
for 8x). it leads to 2.4GHz @1.15V and 3.2GHz @1.35V and that is far
from unstable for this CPU (the one I own is rock stable at 2.4GHz
@1.01V and 3.2GHz @1.26V).

before this Intel CPU, I had experiance with Athlon64 CPUs and
powernow-k8 cpufreq driver. it has been using multiplier-voltage data no
matter what FSB freq was specified, just like windows does with core2duo
CPUs. that was much more flexibile and usefull.

here, I guess, acpi-cpufreq can not locate data for 2.4GHz and 3.2GHz
freqs and fails to work.

is there any way to make acpi-cpufreq use multiplier-voltage instead of
frequency-voltage pairs?

             reply	other threads:[~2007-10-18 15:45 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-10-18 15:45 Nebojsa Trpkovic [this message]
2007-10-19 18:45 ` acpi-cpufreq dissapires when FSB is changed Dave Jones

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=47177FAF.2070206@gmail.com \
    --to=trx.lists@gmail.com \
    --cc=cpufreq@www.linux.org.uk \
    /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.