mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Peter Osterlund <petero2@telia.com>
To: Dominik Brodowski <devel@brodo.de>
Cc: zwane@linuxpower.ca, alan@lxorguk.ukuu.org.uk,
	linux-kernel@vger.kernel.org
Subject: Re: Linux 2.4.19-rc1-ac3
Date: 14 Jul 2002 17:19:54 +0200	[thread overview]
Message-ID: <m2k7nyqqud.fsf@best.localdomain> (raw)
In-Reply-To: <20020714150912.A1148@brodo.de>

Dominik Brodowski <devel@brodo.de> writes:

> On Sun, Jul 14, 2002 at 03:03:08PM +0200, Alan Cox wrote:
> > > 3. The cpu voltage is automatically reduced when the frequency is
> > >    reduced.
> > 
> > True for some x86 processors, either automatically, or on some
> > controlled by us.
> 
> The p4-clockmod driver you seem to using does not scale the voltage. 
> In case you own a P4-M and a ICH2-M or ICH3-M southbridge: the 
> speedstep.c driver in the 2.5.-cpufreq patchset (backport to 2.4. 
> will be available soon) should do the job and adjust the processor 
> voltage.

I tried speedstep but it didn't work because of this check in
speedstep_detect_processor:

		/* Intel Pentium 4 Mobile P4-M */
		if (c->x86_model != 2)
			return 0;

		if (c->x86_mask != 4)
			return 0;     /* all those seem to support Enhanced
					 SpeedStep */

My cpu has model == 1 and mask == 2. I deleted that code just to see
what would happen, but then I tripped over this code in
pentium4_get_frequency:

	/* Don't trust unseen values yet, except in the MHz field 
	 */
	if (msr_hi || ((msr_lo & 0x00FFFFFF) != 0x300511)) {
		printk(KERN_INFO "cpufreq: Due to incomplete documentation, please send a mail to devel@brodo.de\n");
		printk(KERN_INFO "with a dmesg of a boot while on ac-power, and one of a boot on battery-power.\n");
		printk(KERN_INFO "Thanks in advance.\n");
		return 0;
	}

I removed the return statement, and then got:

        apm: BIOS version 1.2 Flags 0x03 (Driver version 1.16)
        cpufreq: Intel(R) SpeedStep(TM) support $Revision: 1.34 $
        model:1 mask:2
        cpufreq: chipset 0x3 - processor 0x3
        cpufreq: activating SpeedStep (TM) registers
        cpufreq: read at pmbase 0x1000 + 0x50 returned 0x1
        cpufreq: P4 - MSR_EBC_FREQUENCY_ID: 0x8a01fe00 0x0
        cpufreq: Due to incomplete documentation, please send a mail to devel@brodo.de
        with a dmesg of a boot while on ac-power, and one of a boot on battery-power.
        Thanks in advance.
        cpufreq: read at pmbase 0x1000 + 0x50 returned 0x1
        cpufreq: writing 0x0 to pmbase 0x1000 + 0x50
        cpufreq: read at pmbase 0x1000 + 0x50 returned 0x0
        cpufreq: change to 0 MHz succeded
        cpufreq: read at pmbase 0x1000 + 0x50 returned 0x0
        cpufreq: P4 - MSR_EBC_FREQUENCY_ID: 0x8a01fe00 0x0
        cpufreq: Due to incomplete documentation, please send a mail to devel@brodo.de
        with a dmesg of a boot while on ac-power, and one of a boot on battery-power.
        Thanks in advance.
        cpufreq: read at pmbase 0x1000 + 0x50 returned 0x0
        cpufreq: read at pmbase 0x1000 + 0x50 returned 0x0
        cpufreq: currently at low speed setting - 13800 MHz
        CPU clock: 13800.000 MHz (13800.000-13800.000 MHz)
        Starting kswapd

There was no difference in dmesg output when booting on battery power
instead of ac power.

So what can I do to make speedstep work? According to the notebook
manual, speedstep is supported on this computer.

-- 
Peter Osterlund - petero2@telia.com
http://w1.894.telia.com/~u89404340

  reply	other threads:[~2002-07-14 15:18 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2002-07-12 19:14 Alan Cox
2002-07-12 23:42 ` Peter Osterlund
2002-07-13  7:41   ` Zwane Mwaikambo
2002-07-13 10:34     ` Peter Osterlund
2002-07-13 12:38       ` Zwane Mwaikambo
2002-07-13 16:22         ` Peter Osterlund
2002-07-13 18:27           ` Alan Cox
2002-07-13 19:43             ` Peter Osterlund
2002-07-13 19:54               ` Russell King
2002-07-13 20:56                 ` Peter Osterlund
2002-07-15 10:13                 ` Thomas Sailer
2002-07-15 10:16                   ` Russell King
2002-07-14  1:50               ` Alan Cox
     [not found]               ` <6010.1026651788@www53.gmx.net>
2002-07-14 13:09                 ` Dominik Brodowski
2002-07-14 15:19                   ` Peter Osterlund [this message]
2002-07-14 15:50                     ` Dominik Brodowski
2002-07-14 18:52                       ` Peter Osterlund
2002-07-13 10:25   ` Peter Osterlund
2002-07-13 14:15 ` Pavel Machek
2002-07-13 17:09   ` Alan Cox
2002-07-13 15:59 ` fchabaud

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=m2k7nyqqud.fsf@best.localdomain \
    --to=petero2@telia.com \
    --cc=alan@lxorguk.ukuu.org.uk \
    --cc=devel@brodo.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=zwane@linuxpower.ca \
    /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

all inboxes | Powered by JetHome®