Re: Centrino question

Re: Centrino question

[Dino Klein]

I have an Acer C302, with a Pentium M 1.6, and the Intel 855GME chipset. 
The CPU/chipset supports C1 - C4.
My hypothesis as to why you don't have C3 is because it is disabled thru 
the value of the C3 latency in the FADT. The DSDT code probably has a 
_CST field/method for the processor, which specifies the available 
C-States. I've used a patch (from here 
http://bugzilla.kernel.org/show_bug.cgi?id=1958), that reads the _CST 
object.
Give it a try, and see if it solves your issue.


Cyril Wattebled wrote:
> Hi there,
> 
> I'm wondering if all Pentium M processor support C1 C2 and C3 ?
> Because I have a Sony Vaio Z1 with a Pentium M 1.7ghz and linux only detects C1
> and C2. I have a friend who has a Toshiba Tecra with a Pentium M 1.4ghz and it
> supports C3.
> I have heard that even C3 and C4 are supported by Centrinos.
> I'm using kernel 2.6.6-rc3. 
> Is there a logical answer why the kernel does only detect C1 and C3 ?
> Thanks
> 
> Cyril


-------------------------------------------------------
This SF.Net email is sponsored by Sleepycat Software
Learn developer strategies Cisco, Motorola, Ericsson & Lucent use to deliver
higher performing products faster, at low TCO.
http://www.sleepycat.com/telcomwpreg.php?From=osdnemail3

Centrino question

[Cyril Wattebled]

Hi there,

I'm wondering if all Pentium M processor support C1 C2 and C3 ?
Because I have a Sony Vaio Z1 with a Pentium M 1.7ghz and linux only detects C1
and C2. I have a friend who has a Toshiba Tecra with a Pentium M 1.4ghz and it
supports C3.
I have heard that even C3 and C4 are supported by Centrinos.
I'm using kernel 2.6.6-rc3. 
Is there a logical answer why the kernel does only detect C1 and C3 ?
Thanks

Cyril


-------------------------------------------------------
This SF.Net email is sponsored by Sleepycat Software
Learn developer strategies Cisco, Motorola, Ericsson & Lucent use to deliver
higher performing products faster, at low TCO.
http://www.sleepycat.com/telcomwpreg.php?From=osdnemail3

Re: Centrino question

[Cyril Wattebled]

Thanks for the link, I applied the patch but I still only have C2 max ...
Any other clues ?

On Sun, 09 May 2004 13:51:13 -0400
Dino Klein <zagzag-lists-zY4eFNvK5D9If6P1QZMOBw@public.gmane.org> wrote:

> I have an Acer C302, with a Pentium M 1.6, and the Intel 855GME chipset. 
> The CPU/chipset supports C1 - C4.
> My hypothesis as to why you don't have C3 is because it is disabled thru 
> the value of the C3 latency in the FADT. The DSDT code probably has a 
> _CST field/method for the processor, which specifies the available 
> C-States. I've used a patch (from here 
> http://bugzilla.kernel.org/show_bug.cgi?id=1958), that reads the _CST 
> object.
> Give it a try, and see if it solves your issue.
> 
> 
> Cyril Wattebled wrote:
> > Hi there,
> > 
> > I'm wondering if all Pentium M processor support C1 C2 and C3 ?
> > Because I have a Sony Vaio Z1 with a Pentium M 1.7ghz and linux only
> > detects C1 and C2. I have a friend who has a Toshiba Tecra with a Pentium M
> > 1.4ghz and it supports C3.
> > I have heard that even C3 and C4 are supported by Centrinos.
> > I'm using kernel 2.6.6-rc3. 
> > Is there a logical answer why the kernel does only detect C1 and C3 ?
> > Thanks
> > 
> > Cyril
> 
> 
> -------------------------------------------------------
> This SF.Net email is sponsored by Sleepycat Software
> Learn developer strategies Cisco, Motorola, Ericsson & Lucent use to deliver
> higher performing products faster, at low TCO.
> http://www.sleepycat.com/telcomwpreg.php?From=osdnemail3
> _______________________________________________
> Acpi-devel mailing list
> Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
> https://lists.sourceforge.net/lists/listinfo/acpi-devel
> 


-------------------------------------------------------
This SF.Net email is sponsored by Sleepycat Software
Learn developer strategies Cisco, Motorola, Ericsson & Lucent use to deliver
higher performing products faster, at low TCO.
http://www.sleepycat.com/telcomwpreg.php?From=osdnemail3

RE: Centrino question

[Yu, Luming]

Need output of acpidmp to verify .

Thanks,
Luming
>-----Original Message-----
>From: acpi-devel-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org 
>[mailto:acpi-devel-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org] On Behalf Of 
>Cyril Wattebled
>Sent: Monday, May 10, 2004 8:28 AM
>To: acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
>Subject: Re: [ACPI] Centrino question
>
>Thanks for the link, I applied the patch but I still only have 
>C2 max ...
>Any other clues ?
>
>On Sun, 09 May 2004 13:51:13 -0400
>Dino Klein <zagzag-lists-zY4eFNvK5D9If6P1QZMOBw@public.gmane.org> wrote:
>
>> I have an Acer C302, with a Pentium M 1.6, and the Intel 
>855GME chipset. 
>> The CPU/chipset supports C1 - C4.
>> My hypothesis as to why you don't have C3 is because it is 
>disabled thru 
>> the value of the C3 latency in the FADT. The DSDT code 
>probably has a 
>> _CST field/method for the processor, which specifies the available 
>> C-States. I've used a patch (from here 
>> http://bugzilla.kernel.org/show_bug.cgi?id=1958), that reads 
>the _CST 
>> object.
>> Give it a try, and see if it solves your issue.
>> 
>> 
>> Cyril Wattebled wrote:
>> > Hi there,
>> > 
>> > I'm wondering if all Pentium M processor support C1 C2 and C3 ?
>> > Because I have a Sony Vaio Z1 with a Pentium M 1.7ghz and 
>linux only
>> > detects C1 and C2. I have a friend who has a Toshiba Tecra 
>with a Pentium M
>> > 1.4ghz and it supports C3.
>> > I have heard that even C3 and C4 are supported by Centrinos.
>> > I'm using kernel 2.6.6-rc3. 
>> > Is there a logical answer why the kernel does only detect 
>C1 and C3 ?
>> > Thanks
>> > 
>> > Cyril
>> 
>> 
>> -------------------------------------------------------
>> This SF.Net email is sponsored by Sleepycat Software
>> Learn developer strategies Cisco, Motorola, Ericsson & 
>Lucent use to deliver
>> higher performing products faster, at low TCO.
>> http://www.sleepycat.com/telcomwpreg.php?From=osdnemail3
>> _______________________________________________
>> Acpi-devel mailing list
>> Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
>> https://lists.sourceforge.net/lists/listinfo/acpi-devel
>> 
>
>
>-------------------------------------------------------
>This SF.Net email is sponsored by Sleepycat Software
>Learn developer strategies Cisco, Motorola, Ericsson & Lucent 
>use to deliver
>higher performing products faster, at low TCO.
>http://www.sleepycat.com/telcomwpreg.php?From=osdnemail3
>_______________________________________________
>Acpi-devel mailing list
>Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
>https://lists.sourceforge.net/lists/listinfo/acpi-devel
>


-------------------------------------------------------
This SF.Net email is sponsored by Sleepycat Software
Learn developer strategies Cisco, Motorola, Ericsson & Lucent use to deliver
higher performing products faster, at low TCO.
http://www.sleepycat.com/telcomwpreg.php?From=osdnemail3

RE: Centrino question

[Yu, Luming]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset="us-ascii", Size: 677 bytes --]

>I so frustrated because I can last up to 6h unplugged on the 
>windows platform

How did you run windows ? 

>but only 2h30 on linux. There must be something wrong.

How did you run Linux?

>Do you guys have any ideas?

As the first step to optimize battery life, you need to 
make sure S3 works,  then compare the battery 
life in S3.  

Thanks,
Luming





-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g.
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id149&alloc_id66&op=click

Re: Centrino question

[Karol Kozimor]

Thus wrote Cyril Wattebled:
> However, I don't understand, I was told that the centrino processor and chipset
> supports C3 and even C4.
> Why would they intend C3 to be disabled ?
> By the way, I tried to change the value of ACPI_PROCESSOR_MAX_C3_LATENCY to
> 1002 so that C3 is enabled. However, it doesn't change anything, C3 state
> is never called.

You'd go for below 1000, that is 999 and less. But anyway, as Len already
wrote, your OEM has probably explicitly disabled C3 (on a Pentium-M laptop?
come on?), so be prepared to face problems such as data corruption.
Best regards,

-- 
Karol 'sziwan' Kozimor
sziwan-DETuoxkZsSqrDJvtcaxF/A@public.gmane.org


-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

Re: Centrino question

[Cyril Wattebled]

Thanks for your answer,

However, I don't understand, I was told that the centrino processor and chipset
supports C3 and even C4.
Why would they intend C3 to be disabled ?
By the way, I tried to change the value of ACPI_PROCESSOR_MAX_C3_LATENCY to
1002 so that C3 is enabled. However, it doesn't change anything, C3 state
is never called.
Actually, I'm looking for a way to make my laptop last longer
unplugged because I think and I also have been told that 2h30 of battery life
is really disappointing for the centrino. Some guys last 3h30 with a P4
3.06ghz, with almost the same battery capacity. I thought that C3 would help a
bit on that but since it's disabled, I don't know what to do. I already enabled
Speedstep, tweaked hdparm spin down value, disabled devices like Bluetooth and
pcmcia.
I so frustrated because I can last up to 6h unplugged on the windows platform
but only 2h30 on linux. There must be something wrong.
Do you guys have any ideas?
Thanks anyway.

--
Cyril Wattebled
neurowork-GANU6spQydw@public.gmane.org


On 20 May 2004 23:50:29 -0400
Len Brown <len.brown-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org> wrote:

> On Wed, 2004-05-19 at 21:57, Cyril Wattebled wrote:
> > Thanks for your answer.
> > 
> > I have checked why exactly the C3 state is disabled thanks to the
> > information you provided.
> > It seems that it's because the C3 latency is too large. The value is 1001.
> > As the max value is 1000, my question is:
> > Is it safe to change ACPI_PROCESSOR_MAX_C3_LATENCY ?
> > I don't have the knowledge to understand what this latency is about, so I
> > don't really know if 1 microsecond really changes something.
> > So is it the FADT table value that is wrong, if not, what should I do ?
> > Could you give me a hint on that if you have time.
> > Thank you.
> 
> 1000 is from the ACPI spec.  If the BIOS put 1001 there, then they
> intended for C3 to be disabled.  You don't want to touch this.
> 
> Indeed, if you were just under 1000, you'd have trouble on 2.6,
> since clock ticks are 1ms apart...
> 
> -Len
> 
> 
> 


-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

Re: Centrino question

[Bruno Ducrot]

On Mon, May 10, 2004 at 12:28:14AM +0000, Cyril Wattebled wrote:
> 
> On Sun, 09 May 2004 13:51:13 -0400
> Dino Klein <zagzag-lists-zY4eFNvK5D9If6P1QZMOBw@public.gmane.org> wrote:
> 
> > http://bugzilla.kernel.org/show_bug.cgi?id=1958), that reads the _CST 
> > object.
> > Give it a try, and see if it solves your issue.

> Thanks for the link, I applied the patch but I still only have C2 max ...
> Any other clues ?


Could you send me the output of acpidmp, please ?

wget ftp://ftp.kernel.org//pub/linux/kernel/people/lenb/acpi/utils/pmtools-20031210.tar.bz2
tar xjvfp pmtools-20031210.tar.bz2
cd pmtools-20031210/acpidmp
make
sudo ./acpidmp > acpidmp.out
bzip2 acpidmp.out 

Then mail me the acpidmp.out.bz2

Cheers,

-- 
Bruno Ducrot

--  Which is worse:  ignorance or apathy?
--  Don't know.  Don't care.


-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

Re: Centrino question

[Ow Mun Heng]

On Fri, 2004-05-21 at 04:52, Karol Kozimor wrote:
> Thus wrote Cyril Wattebled:
> > However, I don't understand, I was told that the centrino processor and chipset
> > supports C3 and even C4.

C3 is Supported


active state:            C2
default state:           C1
bus master activity:     ffffffff
states:
    C1:  promotion[C2] demotion[--] latency[000] usage[01028830]
   *C2: promotion[C3] demotion[C1] latency[050] usage[04785987]
    C3:  promotion[--] demotion[C2] latency[050] usage[00001327]

This is a Dell D600 1.4g. c3 values are low because I have the USB mouse
on right now. (note the fffffff??)

I've tested it with and w/o C3. The difference in the battery lifetime
was only very minmal. I got like maybe 10 mins extra.

I last 2:30 on Linux and 3h on Windows.

Batt Life Sucks..



-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

RE: Centrino question

[Nakajima, Jun]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset="us-ascii", Size: 2126 bytes --]

To get longer battery life, you need to use P-state transitions via
Enhanced SpeedStep as well.

Jun
>-----Original Message-----
>From: acpi-devel-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org [mailto:acpi-devel-
>admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org] On Behalf Of Ow Mun Heng
>Sent: Friday, May 21, 2004 10:39 AM
>To: acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
>Subject: Re: [ACPI] Centrino question
>
>On Fri, 2004-05-21 at 04:52, Karol Kozimor wrote:
>> Thus wrote Cyril Wattebled:
>> > However, I don't understand, I was told that the centrino processor
and
>chipset
>> > supports C3 and even C4.
>
>C3 is Supported
>
>
>active state:            C2
>default state:           C1
>bus master activity:     ffffffff
>states:
>    C1:  promotion[C2] demotion[--] latency[000] usage[01028830]
>   *C2: promotion[C3] demotion[C1] latency[050] usage[04785987]
>    C3:  promotion[--] demotion[C2] latency[050] usage[00001327]
>
>This is a Dell D600 1.4g. c3 values are low because I have the USB
mouse
>on right now. (note the fffffff??)
>
>I've tested it with and w/o C3. The difference in the battery lifetime
>was only very minmal. I got like maybe 10 mins extra.
>
>I last 2:30 on Linux and 3h on Windows.
>
>Batt Life Sucks..
>
>
>
>-------------------------------------------------------
>This SF.Net email is sponsored by: Oracle 10g
>Get certified on the hottest thing ever to hit the market... Oracle
10g.
>Take an Oracle 10g class now, and we'll give you the exam FREE.
>http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click
>_______________________________________________
>Acpi-devel mailing list
>Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
>https://lists.sourceforge.net/lists/listinfo/acpi-devel



-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g.
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id149&alloc_id66&op=click

RE: Centrino question

[Ow Mun Heng]

On Fri, 2004-05-21 at 11:08, Nakajima, Jun wrote:
> To get longer battery life, you need to use P-state transitions via
> Enhanced SpeedStep as well.

Which means we have to use the drivers provided by Intel??
Or do you just mean the speedstep functions?

I have cpuspeedfreq installed and when I'm quoting the number
of hours, I actually am quoting it with cpufreq turned on and running 
at 600Mhz only with nothing plugged into the USB, backlight at it's dimmest
and all that. (Even the fan is turned off!)

No JOY!

So.. Centrino gives a better/longer Batt life?? MY foot! It actually depends
on the batt capacity the OEM gives you! The extra 1MB cache in Pentium-M 
is nice though. :)

What I wouldn't give to get a fuel cell laptop



-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

Re: Centrino question

[Bruno Ducrot]

On Fri, May 21, 2004 at 11:17:54AM -0700, Ow Mun Heng wrote:
> On Fri, 2004-05-21 at 11:08, Nakajima, Jun wrote:
> > To get longer battery life, you need to use P-state transitions via
> > Enhanced SpeedStep as well.
> 
> Which means we have to use the drivers provided by Intel??

Those are deprecated.

> Or do you just mean the speedstep functions?

I guess Jun means the cpufreq centrino speedstep driver.

> I have cpuspeedfreq installed and when I'm quoting the number
> of hours, I actually am quoting it with cpufreq turned on and running 
> at 600Mhz only with nothing plugged into the USB, backlight at it's dimmest
> and all that. (Even the fan is turned off!)
> 
> No JOY!
> 
> So.. Centrino gives a better/longer Batt life?? MY foot! It actually depends
> on the batt capacity the OEM gives you! The extra 1MB cache in Pentium-M 
> is nice though. :)

That's strange.  Are you sure you are not running a kind of throttling
control stuff instead?

-- 
Bruno Ducrot

--  Which is worse:  ignorance or apathy?
--  Don't know.  Don't care.


-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

Re: Centrino question

[Ow Mun Heng]

On Fri, 2004-05-21 at 11:31, Bruno Ducrot wrote:
> > 
> > Which means we have to use the drivers provided by Intel??
> 
> Those are deprecated.

Ah..

> > Or do you just mean the speedstep functions?
> 
> I guess Jun means the cpufreq centrino speedstep driver.

That's what I'm using

> > I have cpuspeedfreq installed and when I'm quoting the number
> > of hours, I actually am quoting it with cpufreq turned on and running 
> > at 600Mhz only with nothing plugged into the USB, backlight at it's dimmest
> > and all that. (Even the fan is turned off!


> That's strange.  Are you sure you are not running a kind of throttling
> control stuff instead?

What do you mean? When I'm on battery, I also run the laptop_mode patch.
(though now, I have trouble with XFS not understanding the commit=600 or
something like that)

I don't have the P-states anymore since upgrading to 2.6 kernel on RH9
(2.6.3)

cat /proc/acpi/info                
version:                 20040211


 cat /proc/acpi/processor/CPU0/throttling
state count:             8
active state:            T0
states:
   *T0:                  00%
    T1:                  12%
    T2:                  25%
    T3:                  37%
    T4:                  50%
    T5:                  62%
    T6:                  75%
    T7:                  87%

cat /proc/acpi/processor/CPU0/info
processor id:            0
acpi id:                 0
bus mastering control:   yes
power management:        yes
throttling control:      yes
limit interface:         yes

cat /proc/acpi/processor/CPU0/limit
active limit:            P0:T0
user limit:              P0:T0
thermal limit:           P0:T0


-- 


-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

Re: Centrino question

[Bruno Ducrot]

On Fri, May 21, 2004 at 11:36:25AM -0700, Ow Mun Heng wrote:
> > That's strange.  Are you sure you are not running a kind of throttling
> > control stuff instead?
> 
> What do you mean? When I'm on battery, I also run the laptop_mode patch.
> (though now, I have trouble with XFS not understanding the commit=600 or
> something like that)

There are people (even OEM btw) that sometimes think that throttling a
CPU is the same as for P-state.

> I don't have the P-states anymore since upgrading to 2.6 kernel on RH9
> (2.6.3)

And?  You are using the cpufreq infrastructure with the enhanced
speedstep driver, which is the same thing anyway.

Well, if really you last only 2 hours and a half on battery, even if
running at 600MHz via cpufreq, and C3 is also enabled, with laptop mode
enabled and so on, either you are using the cdrom continously (which I doubt)
or your battery is failing imho.

-- 
Bruno Ducrot

--  Which is worse:  ignorance or apathy?
--  Don't know.  Don't care.


-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

Re: Centrino question

[Martin Hermanowski]

[-- Attachment #1: Type: text/plain, Size: 1009 bytes --]

On Fri, May 21, 2004 at 10:38:57AM -0700, Ow Mun Heng wrote:
> On Fri, 2004-05-21 at 04:52, Karol Kozimor wrote:
> > Thus wrote Cyril Wattebled:
> > > However, I don't understand, I was told that the centrino processor and chipset
> > > supports C3 and even C4.
> 
> C3 is Supported
> 
> 
> active state:            C2
> default state:           C1
> bus master activity:     ffffffff
> states:
>     C1:  promotion[C2] demotion[--] latency[000] usage[01028830]
>    *C2: promotion[C3] demotion[C1] latency[050] usage[04785987]
>     C3:  promotion[--] demotion[C2] latency[050] usage[00001327]
> 
> This is a Dell D600 1.4g. c3 values are low because I have the USB mouse
> on right now. (note the fffffff??)

The same here (ibm t41p, 1.7GHz). Using USB forces c2. But c1 seems to be used
very seldom (usage count of 10).

> I've tested it with and w/o C3. The difference in the battery lifetime
> was only very minmal. I got like maybe 10 mins extra.

15min here.

LLAP, Martin

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: Centrino question

[Ow Mun Heng]

On Fri, 2004-05-21 at 11:47, Bruno Ducrot wrote:
> On Fri, May 21, 2004 at 11:36:25AM -0700, Ow Mun Heng wrote:
> > > That's strange.  Are you sure you are not running a kind of throttling
> > > control stuff instead?
> > 
> > What do you mean? When I'm on battery, I also run the laptop_mode patch.
> > (though now, I have trouble with XFS not understanding the commit=600 or
> > something like that)
> 
> There are people (even OEM btw) that sometimes think that throttling a
> CPU is the same as for P-state.

Oh.. OK..

> 
> > I don't have the P-states anymore since upgrading to 2.6 kernel on RH9
> > (2.6.3)
> 
> And?  You are using the cpufreq infrastructure with the enhanced
> speedstep driver, which is the same thing anyway.

Cool.. So... nothing to worry about there.

> Well, if really you last only 2 hours and a half on battery, even if
> running at 600MHz via cpufreq, and C3 is also enabled, with laptop mode
> enabled and so on, either you are using the cdrom continously (which I doubt)
> or your battery is failing imho.
hehe.. It has already failed once. Called up DELL for a replacement.
Still Sucks.

Nope.. When I try to detemine the max of my batter life, I don't use the
CDROM at all, I disable Wifi, 600Mhz, laptop_mode, no usb, no sound,
(not using the speakers, but drivers still on) screen to dimmest(14.1
SXGA screen 1400x1050) etc..

2:30 hours.

so... I've stopped hoping / looking for ways to get my battery life
better. It just plainly _sucks_.

Where are you Art Fuel_cell?

(Unless of course I go get a N-Charge battery which will cause this D600
to feel like lugging around a piece of brick!)

-- 


-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

Re: Centrino question

[Ow Mun Heng]

On Fri, 2004-05-21 at 11:48, Martin Hermanowski wrote:
> On Fri, May 21, 2004 at 10:38:57AM -0700, Ow Mun Heng wrote:
> > On Fri, 2004-05-21 at 04:52, Karol Kozimor wrote:
> > > Thus wrote Cyril Wattebled:
> > > > However, I don't understand, I was told that the centrino processor and chipset
> > > > supports C3 and even C4.
> > 
> > C3 is Supported
> > states:
> >     C1:  promotion[C2] demotion[--] latency[000] usage[01028830]
> >    *C2: promotion[C3] demotion[C1] latency[050] usage[04785987]
> >     C3:  promotion[--] demotion[C2] latency[050] usage[00001327]
> 
> The same here (ibm t41p, 1.7GHz). Using USB forces c2. But c1 seems to be used
> very seldom (usage count of 10).
I need to mention that those values are plugged in values. so.. C1 is
there a lot

-- 


-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

RE: Centrino question

[Nakajima, Jun]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset="us-ascii", Size: 1028 bytes --]

>From: Ow Mun Heng [mailto:Ow.Mun.Heng-Sjgp3cTcYWE@public.gmane.org]
>Sent: Friday, May 21, 2004 12:05 PM
>To: Bruno Ducrot
>Cc: Nakajima, Jun; acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
>Subject: Re: [ACPI] Centrino question
>
>Nope.. When I try to detemine the max of my batter life, I don't use
the
>CDROM at all, I disable Wifi, 600Mhz, laptop_mode, no usb, no sound,
>(not using the speakers, but drivers still on) screen to dimmest(14.1
>SXGA screen 1400x1050) etc..
>
>2:30 hours.
>
>so... I've stopped hoping / looking for ways to get my battery life
>better. It just plainly _sucks_.

So how long does it last when you run the processor at full speed,
rather than 600Hhz?

Jun




-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g.
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id149&alloc_id66&op=click

RE: Centrino question

[Ow Mun Heng]

On Fri, 2004-05-21 at 14:27, Nakajima, Jun wrote:
> >From: Ow Mun Heng [mailto:Ow.Mun.Heng-Sjgp3cTcYWE@public.gmane.org]
> >
> >so... I've stopped hoping / looking for ways to get my battery life
> >better. It just plainly _sucks_.
> 
> So how long does it last when you run the processor at full speed,
> rather than 600Hhz?

Hmm.. frankly I have no idea.. I might just try that..
-- 


-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

Re: Centrino question

[Cyril Wattebled]

On Fri, 21 May 2004 19:42:58
"Yu, Luming" <luming.yu-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org> wrote:

> How did you run windows ? 
Well, the first thing to say is that, on Windows, when the laptop is unplugged,
the processor slows down to 200 mhz instead of 600 mhz when plugged. That
says a lot of heat and, I guess power. When I say up to 6h, I mean when the
laptop is idle of course.

> How did you run Linux?

Same here, I tried to let it idle and it hold 2h30 ..

> As the first step to optimize battery life, you need to 
> make sure S3 works,  then compare the battery 
> life in S3.
Yeah, I've been trying for some time now to make it work.
It suspends very well but never ever woke up. Even worse, I can't even hard
reboot. I have to remove the battery and plug it back to make it work.
By the way, if anyone has an idea about what's wrong ... please share =]

> Thanks,
> Luming
No no no, thank you ! =]


-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

Re: Centrino question

[Len Brown]

On Fri, 2004-05-21 at 14:48, Martin Hermanowski wrote:
> On Fri, May 21, 2004 at 10:38:57AM -0700, Ow Mun Heng wrote:
> > On Fri, 2004-05-21 at 04:52, Karol Kozimor wrote:
> > > Thus wrote Cyril Wattebled:
> > > > However, I don't understand, I was told that the centrino processor and chipset
> > > > supports C3 and even C4.
> > 
> > C3 is Supported
> > 
> > 
> > active state:            C2
> > default state:           C1
> > bus master activity:     ffffffff
> > states:
> >     C1:  promotion[C2] demotion[--] latency[000] usage[01028830]
> >    *C2: promotion[C3] demotion[C1] latency[050] usage[04785987]
> >     C3:  promotion[--] demotion[C2] latency[050] usage[00001327]
> > 
> > This is a Dell D600 1.4g. c3 values are low because I have the USB mouse
> > on right now. (note the fffffff??)
> 
> The same here (ibm t41p, 1.7GHz). Using USB forces c2. But c1 seems to be used
> very seldom (usage count of 10).

Known issue with C1 and C2...
When C2 has very low latency, say 1, then after we're in C1
10 times and promote to C2, we never demote to C1 again, ever.

The processor driver is tracking how successful we are at
being idle, but is not tracking how non-idle the system is,
so our current demotion policy is somewhat simplistic.

-Len




-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

RE: Centrino question

[Len Brown]

On Fri, 2004-05-21 at 19:45, Ow Mun Heng wrote:
> On Fri, 2004-05-21 at 14:27, Nakajima, Jun wrote:

> > So how long does it last when you run the processor at full speed,
> > rather than 600Hhz?
> 
> Hmm.. frankly I have no idea.. I might just try that..

Curious if you've also got numbers for the same hardware
running 2.4.  I'd like to know if 2.4's 100HZ clock saved
you anything over 2.6's 1000HZ clock.  C3 wouuld be more
effective on an idle system with the slower clock.

-Len




-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

Re: Centrino question

[Len Brown]

On Fri, 2004-05-21 at 20:56, Cyril Wattebled wrote:
> On Fri, 21 May 2004 19:42:58
> "Yu, Luming" <luming.yu-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org> wrote:
> 
> > How did you run windows ? 
> Well, the first thing to say is that, on Windows, when the laptop is unplugged,
> the processor slows down to 200 mhz instead of 600 mhz when plugged. That
> says a lot of heat and, I guess power. When I say up to 6h, I mean when the
> laptop is idle of course.
> 
> > How did you run Linux?
> 
> Same here, I tried to let it idle and it hold 2h30 ..

Linux slows the processor to 600MHz when Windows gets down to 200MHz?

How about the display -- is the brightness comparable on Linux vs
Windows?

thanks,
-Len




-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

Re: Centrino question

[Cyril Wattebled]

On 22 May 2004 23:24:10 -0400
Len Brown <len.brown-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org> wrote:

> On Fri, 2004-05-21 at 20:56, Cyril Wattebled wrote:
> > On Fri, 21 May 2004 19:42:58
> > "Yu, Luming" <luming.yu-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org> wrote:
> > 
> > > How did you run windows ? 
> > Well, the first thing to say is that, on Windows, when the laptop is
> > unplugged, the processor slows down to 200 mhz instead of 600 mhz when
> > plugged. That says a lot of heat and, I guess power. When I say up to 6h, I
> > mean when the laptop is idle of course.
> > 
> > > How did you run Linux?
> > 
> > Same here, I tried to let it idle and it hold 2h30 ..
> 
> Linux slows the processor to 600MHz when Windows gets down to 200MHz?

Yes, I have already asked around why this happends. Windows slows the processor
down to 200 mhz when unplugged and down to 600 mhz when plugged. Windows
behaves just like linux with the AC power on but when on battery, it manages
to slow it down to 200mhz. This allows the laptop to last more than on linux
and also, the temperature of the overall laptop is cooler (which is really more
confortable).
It seems that Windows is using some kind of clock modulation along with the
enhanced speedstep. I'm not sure about that but, I was told that it should be
something like it. (cf cpufreq mailing list, thread called "Pentium M minimum
clock modulation").

> How about the display -- is the brightness comparable on Linux vs
> Windows?

yes, The screen brightness is set to the same value.

> thanks,
> -Len

Thanks,

--
Cyril Wattebled
neurowork-GANU6spQydw@public.gmane.org


-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

Re: Centrino question

[Dino Klein]

Windows will use throttling to save on battery, when you are using the Max 
Battery policy, and power is unplugged.
Check the following link, where MS explains the policies in XP:
http://www.microsoft.com/whdc/system/pnppwr/powermgmt/ProcPerfCtrl.mspx


>From: Cyril Wattebled <neurowork-GANU6spQydw@public.gmane.org>
>....
>Yes, I have already asked around why this happends. Windows slows the 
>processor
>down to 200 mhz when unplugged and down to 600 mhz when plugged. Windows
>behaves just like linux with the AC power on but when on battery, it 
>manages
>to slow it down to 200mhz. This allows the laptop to last more than on 
>linux
>and also, the temperature of the overall laptop is cooler (which is really 
>more
>confortable).
>It seems that Windows is using some kind of clock modulation along with the
>enhanced speedstep. I'm not sure about that but, I was told that it should 
>be
>something like it. (cf cpufreq mailing list, thread called "Pentium M 
>minimum
>clock modulation").

_________________________________________________________________
MSN Messenger: instale grátis e converse com seus amigos. 
http://messenger.msn.com.br



-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

Re: Centrino question

[Martin Hermanowski]

[-- Attachment #1: Type: text/plain, Size: 1688 bytes --]

On Sun, May 23, 2004 at 05:33:04PM -0300, Dino Klein wrote:
> Windows will use throttling to save on battery, when you are using the Max 
> Battery policy, and power is unplugged.
> Check the following link, where MS explains the policies in XP:
> http://www.microsoft.com/whdc/system/pnppwr/powermgmt/ProcPerfCtrl.mspx
> 
> 
> >From: Cyril Wattebled <neurowork-GANU6spQydw@public.gmane.org>
> >....
> >Yes, I have already asked around why this happends. Windows slows the 
> >processor
> >down to 200 mhz when unplugged and down to 600 mhz when plugged. Windows
> >behaves just like linux with the AC power on but when on battery, it 
> >manages
> >to slow it down to 200mhz. This allows the laptop to last more than on 
> >linux
> >and also, the temperature of the overall laptop is cooler (which is really 
> >more
> >confortable).
> >It seems that Windows is using some kind of clock modulation along with the
> >enhanced speedstep. I'm not sure about that but, I was told that it should 
> >be
> >something like it. (cf cpufreq mailing list, thread called "Pentium M 
> >minimum
> >clock modulation").

I just tried `echo 7 > throttling', and I noticed that the system is
quite underpowered at 75 MHz, so this seems to work. Byt it did not
change the power consumption. I am getting 3.5h at throttling = 0 and
the same at 4, but this seems quite ok with the 6 cell battery in the
t41p. I can get up to 4h when I turn off the hard disk.

I never tried the pre-installed windows, but I think that the advertised
runtime with this battery is about 4 hours.

Should I be able to reduce the power consumption by activating
throttling?

LLAP, Martin

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: Centrino question

[Karol Kozimor]

Thus wrote Martin Hermanowski:
> Should I be able to reduce the power consumption by activating
> throttling?

Not directly. When under load, your processor will generate less heat. If
that allows you to turn active cooling off (i.e. the fans), you'll save
some power. When idle, the consumption should be about the same.
Best regards,

-- 
Karol 'sziwan' Kozimor
sziwan-DETuoxkZsSqrDJvtcaxF/A@public.gmane.org


-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

RE: Centrino question

[Yu, Luming]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset="us-ascii", Size: 731 bytes --]

>> Linux slows the processor to 600MHz when Windows gets down to 200MHz?
>

Hmm, could you post dmesg, and let us to see the lowest frequency
supported 
in your CPU freq driver.Maybe this is a bug!

>
>> How about the display -- is the brightness comparable on Linux vs
>> Windows?
>
>yes, The screen brightness is set to the same value.
>

How about turning off disk, and unplugging USB devices. 

Thanks,
Luming


-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g.
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id149&alloc_id66&op=click

Re: Centrino question

[Ow Mun Heng]

On Sun, 2004-05-23 at 06:21, Cyril Wattebled wrote:
> On 22 May 2004 23:24:10 -0400
> Len Brown <len.brown-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org> wrote:
> > Linux slows the processor to 600MHz when Windows gets down to 200MHz?

Len,

	On my system, either 2.4 or 2.6 kernel, running on battery, cpufreq is at 600Mhz but
if i were running gkrellm, the x86info plugin will show that the cpufreq is actually fluactuating 
between 99 to 200 mHz. I'm not sure where this plugin gets it's data but..


-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

RE: Centrino question

[Ow Mun Heng]

On Sat, 2004-05-22 at 20:22, Len Brown wrote:
> On Fri, 2004-05-21 at 19:45, Ow Mun Heng wrote:
> > On Fri, 2004-05-21 at 14:27, Nakajima, Jun wrote:
> 
> > > So how long does it last when you run the processor at full speed,
> > > rather than 600Hhz?
> > 
> > Hmm.. frankly I have no idea.. I might just try that..
> 
> Curious if you've also got numbers for the same hardware
> running 2.4.  I'd like to know if 2.4's 100HZ clock saved
> you anything over 2.6's 1000HZ clock.  C3 wouuld be more
> effective on an idle system with the slower clock.

Well, running 2.4 or 2.6 kernel, I really didn't 'feel' any difference
or increase in battery usage time. 

All I know is, C3 saved me like only 10 minutes. (regardless of 2.4 or
2.6 I would think. I can't remember exactly but...)

-- 


-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

Re: Centrino question

[Ow Mun Heng]

On Sun, 2004-05-23 at 13:33, Dino Klein wrote:
> Windows will use throttling to save on battery, when you are using the Max 
> Battery policy, and power is unplugged.
> Check the following link, where MS explains the policies in XP:
> http://www.microsoft.com/whdc/system/pnppwr/powermgmt/ProcPerfCtrl.mspx

D@@@! The whitepaper is a ms Word file packaged as an .exe file?? What
gives? How to view that whitepaper without dishing out a copy of
Windows??!!



-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

Re: Centrino question

[Arjen Verweij]

Maybe try antiword or similar?

On Mon, 24 May 2004, Ow Mun Heng wrote:

> On Sun, 2004-05-23 at 13:33, Dino Klein wrote:
> > Windows will use throttling to save on battery, when you are using the Max
> > Battery policy, and power is unplugged.
> > Check the following link, where MS explains the policies in XP:
> > http://www.microsoft.com/whdc/system/pnppwr/powermgmt/ProcPerfCtrl.mspx
>
> D@@@! The whitepaper is a ms Word file packaged as an .exe file?? What
> gives? How to view that whitepaper without dishing out a copy of
> Windows??!!
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by: Oracle 10g
> Get certified on the hottest thing ever to hit the market... Oracle 10g.
> Take an Oracle 10g class now, and we'll give you the exam FREE.
> http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click
> _______________________________________________
> Acpi-devel mailing list
> Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
> https://lists.sourceforge.net/lists/listinfo/acpi-devel
>



-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

Re: Centrino question

[David Eriksson]

On Mon, 2004-05-24 at 18:46, Ow Mun Heng wrote:
> On Sun, 2004-05-23 at 13:33, Dino Klein wrote:
> > Windows will use throttling to save on battery, when you are using the Max 
> > Battery policy, and power is unplugged.
> > Check the following link, where MS explains the policies in XP:
> > http://www.microsoft.com/whdc/system/pnppwr/powermgmt/ProcPerfCtrl.mspx
> 
> D@@@! The whitepaper is a ms Word file packaged as an .exe file?? What
> gives? How to view that whitepaper without dishing out a copy of
> Windows??!!

Use unzip to get the Word document out of the .exe file.

-- 
Regards,
               -\- David Eriksson -/-

        SynCE - http://synce.sourceforge.net
       CalcEm - http://calcem.sourceforge.net
      ScummVM - http://scummvm.sourceforge.net
     Desquirr - http://desquirr.sourceforge.net
  SetiWrapper - http://setiwrapper.sourceforge.net



-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

Re: Centrino question

[Ow Mun Heng]

On Mon, 2004-05-24 at 09:53, Arjen Verweij wrote:
> Maybe try antiword or similar?

Maybe you didn't read the prob fully. It's packaged as an ".exe" file.

I have openoffice installed so that's not a prob.
> 
> On Mon, 24 May 2004, Ow Mun Heng wrote:
> 
> > On Sun, 2004-05-23 at 13:33, Dino Klein wrote:
> > > Windows will use throttling to save on battery, when you are using the Max
> > > Battery policy, and power is unplugged.
> > > Check the following link, where MS explains the policies in XP:
> > > http://www.microsoft.com/whdc/system/pnppwr/powermgmt/ProcPerfCtrl.mspx
> >
> > D@@@! The whitepaper is a ms Word file packaged as an .exe file?? What
> > gives? How to view that whitepaper without dishing out a copy of
> > Windows??!!
> >
> >
> >
> > -------------------------------------------------------
> > This SF.Net email is sponsored by: Oracle 10g
> > Get certified on the hottest thing ever to hit the market... Oracle 10g.
> > Take an Oracle 10g class now, and we'll give you the exam FREE.
> > http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click
> > _______________________________________________
> > Acpi-devel mailing list
> > Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
> > https://lists.sourceforge.net/lists/listinfo/acpi-devel
> >
> 
> 
> 
> -------------------------------------------------------
> This SF.Net email is sponsored by: Oracle 10g
> Get certified on the hottest thing ever to hit the market... Oracle 10g. 
> Take an Oracle 10g class now, and we'll give you the exam FREE.
> http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click
> _______________________________________________
> Acpi-devel mailing list
> Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
> https://lists.sourceforge.net/lists/listinfo/acpi-devel
-- 


-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

Re: Centrino question

[Dino Klein]

easy... :)
I just used unzip on solaris & linux and it decompressed it.


>From: Ow Mun Heng <Ow.Mun.Heng-Sjgp3cTcYWE@public.gmane.org>
>To: acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
>Subject: Re: [ACPI] Centrino question
>Date: Mon, 24 May 2004 09:46:58 -0700
>
>On Sun, 2004-05-23 at 13:33, Dino Klein wrote:
> > Windows will use throttling to save on battery, when you are using the 
>Max
> > Battery policy, and power is unplugged.
> > Check the following link, where MS explains the policies in XP:
> > http://www.microsoft.com/whdc/system/pnppwr/powermgmt/ProcPerfCtrl.mspx
>
>D@@@! The whitepaper is a ms Word file packaged as an .exe file?? What
>gives? How to view that whitepaper without dishing out a copy of
>Windows??!!
>
>
>
>-------------------------------------------------------
>This SF.Net email is sponsored by: Oracle 10g
>Get certified on the hottest thing ever to hit the market... Oracle 10g.
>Take an Oracle 10g class now, and we'll give you the exam FREE.
>http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click
>_______________________________________________
>Acpi-devel mailing list
>Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
>https://lists.sourceforge.net/lists/listinfo/acpi-devel

_________________________________________________________________
MSN Messenger: instale grátis e converse com seus amigos. 
http://messenger.msn.com.br



-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

Re: Centrino question

[Stefan Seyfried]

On Mon, May 24, 2004 at 09:11:18AM -0700, Ow Mun Heng wrote:
> On Sun, 2004-05-23 at 06:21, Cyril Wattebled wrote:
> > On 22 May 2004 23:24:10 -0400
> > Len Brown <len.brown-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org> wrote:
> > > Linux slows the processor to 600MHz when Windows gets down to 200MHz?
> 
> Len,
> 
> 	On my system, either 2.4 or 2.6 kernel, running on battery, cpufreq is at 600Mhz but
> if i were running gkrellm, the x86info plugin will show that the cpufreq is actually fluactuating 
> between 99 to 200 mHz. I'm not sure where this plugin gets it's data but..

I am quite sure that this is due to the tool you are using and due to C3 usage.
I don't know about the gkrellm x86info plugin, but the suse 9.1 powersaved
has a function to get the "real" CPU speed. It goes something like this:

float getRealProcessorSpeed() {

        /* get this function in cached memory */
        gettimeofday(&tvstart, NULL);
        __asm__ volatile ("rdtsc" : "=A" (x));
        cycles[0] = x;
        gettimeofday(&tvstart, NULL);

        /* we don't trust that this is any specific length of time */
        sleep_left = sleep(1);
        if (sleep_left || errno == EINTR)
                return -1.0;
        __asm__ volatile ("rdtsc" : "=A" (x));
        cycles[1] = x;
        gettimeofday(&tvstop, NULL);
        microseconds = ((tvstop.tv_sec-tvstart.tv_sec)*1000000) +
                (tvstop.tv_usec-tvstart.tv_usec);

        ret = (float)(cycles[1]-cycles[0])/microseconds;
        return ret;
}

There is of course some more checking around, it is only working on i386, 
not on ia64.  This shows me:

seife-4N56LNXDvWs@public.gmane.org:~> powersave -r
82.559082 MHz

but from /sys/.../scaling_setspeed i see that my cpu is running on 600mhz.

seife-4N56LNXDvWs@public.gmane.org:~> cat /proc/acpi/processor/CPU0/power 
active state:            C3
default state:           C1
bus master activity:     00000000
states:
    C1:                  promotion[C2] demotion[--] latency[000] usage[00020000]
    C2:                  promotion[C3] demotion[C1] latency[002] usage[00737909]
   *C3:                  promotion[--] demotion[C2] latency[205] usage[35772579]

if i force C2 state (by plugging in a usb mouse or playing sound), it shows
the frequency correct:

seife-4N56LNXDvWs@public.gmane.org:~> powersave -r
598.058594 MHz

seife-4N56LNXDvWs@public.gmane.org:~> cat /proc/acpi/processor/CPU0/power 
active state:            C2
default state:           C1
bus master activity:     ffffffff
states:
    C1:                  promotion[C2] demotion[--] latency[000] usage[00020850]
   *C2:                  promotion[C3] demotion[C1] latency[002] usage[00770669]
    C3:                  promotion[--] demotion[C2] latency[205] usage[35825285]

So the effects you are seeing are clearly related to a tool that is measuring
wrong.
-- 
Stefan Seyfried



-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

Re: Centrino question

[Alexander Hoogerhuis]

On Fri, 2004-05-21 at 10:38 -0700, Ow Mun Heng wrote:
> On Fri, 2004-05-21 at 04:52, Karol Kozimor wrote:
> > Thus wrote Cyril Wattebled:
> > > However, I don't understand, I was told that the centrino processor and chipset
> > > supports C3 and even C4.
> 
> C3 is Supported
> 
> 
> active state:            C2
> default state:           C1
> bus master activity:     ffffffff
> states:
>     C1:  promotion[C2] demotion[--] latency[000] usage[01028830]
>    *C2: promotion[C3] demotion[C1] latency[050] usage[04785987]
>     C3:  promotion[--] demotion[C2] latency[050] usage[00001327]
> 
> This is a Dell D600 1.4g. c3 values are low because I have the USB mouse
> on right now. (note the fffffff??)
> 
> I've tested it with and w/o C3. The difference in the battery lifetime
> was only very minmal. I got like maybe 10 mins extra.
> 
> I last 2:30 on Linux and 3h on Windows.
> 
> Batt Life Sucks..
> 

I've been through tracking down what disables C3 on my laptop (HP
nc6000). It all happens when the uhci-hci USB drviver goes in. When I
queried about it on lkml a while back it was due tro some support
lacking in the UHCI dirvers as of yet.

mvh,
A



-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

Re: Centrino question

[Karol Kozimor]

Thus wrote Alexander Hoogerhuis:
> I've been through tracking down what disables C3 on my laptop (HP
> nc6000). It all happens when the uhci-hci USB drviver goes in. When I
> queried about it on lkml a while back it was due tro some support
> lacking in the UHCI dirvers as of yet.

Whenever something is plugged into the controller, bus mastering kicks in
and C3 is effectively disabled, also on Windows. This is, to my best
knowledge, a limitation of the USB interface itself. As far as I know, at
least uhci-hcd driver in 2.6 suspends the controller when nothing is
plugged in, thus allowing to use C3. The situation might be different under
2.4.
Best regards,

-- 
Karol 'sziwan' Kozimor
sziwan-DETuoxkZsSqrDJvtcaxF/A@public.gmane.org


-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

Re: Centrino question

[Len Brown]

On Tue, 2004-05-25 at 19:28, Alexander Hoogerhuis wrote:
> On Fri, 2004-05-21 at 10:38 -0700, Ow Mun Heng wrote:
> > On Fri, 2004-05-21 at 04:52, Karol Kozimor wrote:
> > > Thus wrote Cyril Wattebled:
> > > > However, I don't understand, I was told that the centrino processor and chipset
> > > > supports C3 and even C4.
> > 
> > C3 is Supported
> > 
> > 
> > active state:            C2
> > default state:           C1
> > bus master activity:     ffffffff
> > states:
> >     C1:  promotion[C2] demotion[--] latency[000] usage[01028830]
> >    *C2: promotion[C3] demotion[C1] latency[050] usage[04785987]
> >     C3:  promotion[--] demotion[C2] latency[050] usage[00001327]
> > 
> > This is a Dell D600 1.4g. c3 values are low because I have the USB mouse
> > on right now. (note the fffffff??)
> > 
> > I've tested it with and w/o C3. The difference in the battery lifetime
> > was only very minmal. I got like maybe 10 mins extra.

> > I last 2:30 on Linux and 3h on Windows.
> > 
> > Batt Life Sucks..

try dimming the display.

> 
> I've been through tracking down what disables C3 on my laptop (HP
> nc6000). It all happens when the uhci-hci USB drviver goes in. When I
> queried about it on lkml a while back it was due tro some support
> lacking in the UHCI dirvers as of yet.

USB needs to do a better job of recognizing that it is idle.
It isn't really a USB bug, or a missing feature -- more of
a "system integration" issue...

-Len




-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

Re: Centrino question

[Pavel Machek]

Hi!

> Whenever something is plugged into the controller, bus mastering kicks in
> and C3 is effectively disabled, also on Windows. This is, to my best
> knowledge, a limitation of the USB interface itself. As far as I know, at

OHCI does not have that problem; unfortunately UHCI is
misdesigned. (See also FSBR)
				Pavel
-- 
64 bytes from 195.113.31.123: icmp_seq=28 ttl=51 time=448769.1 ms         



-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click

Re: Centrino question

[Ow Mun Heng]

On Wed, 2004-05-26 at 23:22, Len Brown wrote:
> > > I've tested it with and w/o C3. The difference in the battery lifetime
> > > was only very minmal. I got like maybe 10 mins extra.
> 
> > > I last 2:30 on Linux and 3h on Windows.
> > > 
> > > Batt Life Sucks..
> 
> try dimming the display.

The quoted time _is_ with the dimmest display blacklight.


> -Len
-- 


-------------------------------------------------------
This SF.Net email is sponsored by: Oracle 10g
Get certified on the hottest thing ever to hit the market... Oracle 10g. 
Take an Oracle 10g class now, and we'll give you the exam FREE.
http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click