All of lore.kernel.org
 help / color / mirror / Atom feed
From: Thomas Renninger <acpi@renninger.de>
To: Minty <mintywalker@gmail.com>
Cc: linux-acpi@vger.kernel.org
Subject: Re: acpi on the sony tx1xp vaio laptop
Date: Fri, 20 Jan 2006 09:12:51 +0100	[thread overview]
Message-ID: <43D09B83.1030905@renninger.de> (raw)
In-Reply-To: <e1868cfe0601181717m47dbb220l92d2b70033ca4655@mail.gmail.com>

Minty wrote:
> hello,
> 
> I've got a Sony TX1XP laptop running Kubuntu [1].
> 
> There are a couple of acpi things that I'd love to be improved on this
> machine, so I'm offering to help out if I can - but I could use some
> direction from someone who has a clue about acpi and it's workings. 
> I'm happy enough (although not that great at) kernel compiling if need
> be.  Fwiw, I know of two others running some linux variant on this
> machine/hardware.
> 
> Main issue:
> /proc/acpi/fan exists, but is empty.
> Fan comes on for about 20 sec, off for 5 secs, repeat.
> Regardless of cpu temp.
> 
> cpu temp is accurately reported, normally sitting at 61C, but I've
> seen it go higher when the machine is under load.  thermal_zone
> reports passive cooling, but aside from reporting the temp, anything
> that might be set is reported as "<setting not supported>".  It
Passive/active cooling resets new trip point values exported by BIOS.
If this is not supported(nearly none machine does) override them
manually(see below).
> appears that cpu throttling works via
> /proc/acpi/processor/CPU0/throttling but this doesn't appear to alter
> the behaviour of the fan.
Lower the passive trip point(s). You need to play arround a bit to
find your preferred performance/accoustic balance.
The kernel will lower cpufreq as soon as you reach passive temp.
As fans are mostly controlled by BIOS, this is a good and mostly only
solution to get your machine quiet.
Not sure whether kubuntu already includes powersave/kpowersave packages,
AFAIK they exist.
There you can define schemes and it will override the trip points
as defined (you still need to define them manually as a perfect value 
needs a bit of watching the temps and listening to the fan activity 
of a machine). You can switch the schemes as you wish then...
This makes only sense if your machine supports cpufreq! Throttling is bad
and makes your machine unusable slow.
(test -d /sys/devices/system/cpu/cpu0/cpufreq && echo "cpufreq supported").
You also can simply override the values by:
echo "CRITICAL:HOT:PASSIVE:ACTIVE[0]:ACTIVE[1]" >/proc/acpi/thermal_zone/X/trip_points
You need to pass all five values, not defined values (do a cat before) will be
ignored. Be careful with the critical trip point value, you shouldn't use 
another value than the one defined by BIOS!

    Thomas

      parent reply	other threads:[~2006-01-20  8:12 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-01-19  1:17 acpi on the sony tx1xp vaio laptop Minty
2006-01-19  9:47 ` Matthew Garrett
2006-01-20  8:12 ` Thomas Renninger [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=43D09B83.1030905@renninger.de \
    --to=acpi@renninger.de \
    --cc=linux-acpi@vger.kernel.org \
    --cc=mintywalker@gmail.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.