Re: [lm-sensors] Coretemp goes up since 2.6.31.4

Re: [lm-sensors] Coretemp goes up since 2.6.31.4

[Jean Delvare]

On Thu, 08 Apr 2010 18:36:34 +0200, dienet wrote:
> I'll try to ask You first then I'll write do LKML.

Neither is correct. Please check MAINTAINERS, it says:

CORETEMP HARDWARE MONITORING DRIVER
M:      Rudolf Marek <r.marek@assembler.cz>
M:      Huaxu Wan <huaxu.wan@intel.com>
L:      lm-sensors@lm-sensors.org
S:      Maintained
F:      Documentation/hwmon/coretemp
F:      drivers/hwmon/coretemp.c

I'm adding the lm-sensors list to Cc now.

> I was using 2.6.31.4 for quite some time - it works OK.
> Command 'sensor' that gives me temperature on my C2D, was saying that  
> cores have around 28-45 deg (min and maxs on normal day work)
> Now on 2.6.33.1 sensors gives temperatures around 32-48 deg. I never seen  
> 29 or below. No room temperature is changing (got two kernels installed,  
> so I can change it quickly)
> Who is telling the truth here? Was there any works on coretemp module  
> since 2.6.31.4 that made the change in read-out?

There were some changes to the coretemp driver in 2.6.32 and 2.6.33,
but these shouldn't affect older CPU models and would also not change
the readout by a few degrees. You can check if the high temperature
limit is the same - then you can assume the driver changes did not
affect your system. If you want to double-check, feel free to download
the standalone version of the coretemp driver from:
  http://khali.linux-fr.org/devel/misc/coretemp/
You can use that driver with both kernels. If the temperature readings
are still different, then this means that the coretemp driver is
innocent.

> I think that because that temp. change fan is working more often.
> 
> I'm using Lenovo ThinkPad T61.

This suggests that the temperature increase is real and this isn't just
a reporting issue. I suggest that you check
in /proc/acpi/processor/*/power, the amount of time spent in C2 and C3
states. Also install powertop and run it on both kernels, see if 2.6.33
has more frequent wake-ups than 2.6.31.

-- 
Jean Delvare
http://khali.linux-fr.org/wishlist.html

_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors

Re: [lm-sensors] Coretemp goes up since 2.6.31.4

[dienet]

On Fri, 09 Apr 2010 11:17:33 +0200, Jean Delvare <khali@linux-fr.org>
wrote:

> On Thu, 08 Apr 2010 18:36:34 +0200, dienet wrote:
>> I was using 2.6.31.4 for quite some time - it works OK.
>> Command 'sensor' that gives me temperature on my C2D, was saying that
>> cores have around 28-45 deg (min and maxs on normal day work)
>> Now on 2.6.33.1 sensors gives temperatures around 32-48 deg. I never  
>> seen
>> 29 or below. No room temperature is changing (got two kernels installed,
>> so I can change it quickly)
>> Who is telling the truth here? Was there any works on coretemp module
>> since 2.6.31.4 that made the change in read-out?
>
> There were some changes to the coretemp driver in 2.6.32 and 2.6.33,
> but these shouldn't affect older CPU models and would also not change
> the readout by a few degrees. You can check if the high temperature
> limit is the same - then you can assume the driver changes did not
> affect your system.

On 2.6.31.4:

coretemp-isa-0000
Adapter: ISA adapter
Core 0:      +33.0°C  (high = +100.0°C, crit = +100.0°C)

coretemp-isa-0001
Adapter: ISA adapter
Core 1:      +33.0°C  (high = +100.0°C, crit = +100.0°C)

On 2.6.33.1:

coretemp-isa-0000
Adapter: ISA adapter
Core 0:      +34.0°C  (high = +105.0°C, crit = +105.0°C)

coretemp-isa-0001
Adapter: ISA adapter
Core 1:      +34.0°C  (high = +105.0°C, crit = +105.0°C)

So there is a 5 deg diffrance. I did not change any config files.

>> I think that because that temp. change fan is working more often.
>>
>> I'm using Lenovo ThinkPad T61.
>
> This suggests that the temperature increase is real and this isn't just
> a reporting issue. I suggest that you check
> in /proc/acpi/processor/*/power, the amount of time spent in C2 and C3
> states. Also install powertop and run it on both kernels, see if 2.6.33
> has more frequent wake-ups than 2.6.31.

On 2.6.31.4 powertop says:

C3 - 90, 7% of the time
P (800MHz) - 98%
Wakeups: 627

On 2.6.33.1 powertop says:

C3 - 93% of the time
P (800MHz) - 98%
Wakeups: 556

Was running the same programs. Kernel 2.6.33.1 was running for about 1h.

I'm asuming that /proc/acpi/processor/*/power is used by powertop so I'm
not paseting it. But it also says that C3 is more often used then others.

-- 
pozdr0
dienet

"Old C programmers never die. They're just cast into void."

------------------------------------------------------
Porozmawiaj z wrozka!
Sprawdz >> http://link.interia.pl/f2661


_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors

Re: [lm-sensors] Coretemp goes up since 2.6.31.4

[Jean Delvare]

On Fri, 09 Apr 2010 20:08:18 +0200, dienet wrote:
> On Fri, 09 Apr 2010 11:17:33 +0200, Jean Delvare <khali@linux-fr.org>
> wrote:
> 
> > On Thu, 08 Apr 2010 18:36:34 +0200, dienet wrote:
> >> I was using 2.6.31.4 for quite some time - it works OK.
> >> Command 'sensor' that gives me temperature on my C2D, was saying that
> >> cores have around 28-45 deg (min and maxs on normal day work)
> >> Now on 2.6.33.1 sensors gives temperatures around 32-48 deg. I never  
> >> seen
> >> 29 or below. No room temperature is changing (got two kernels installed,
> >> so I can change it quickly)
> >> Who is telling the truth here? Was there any works on coretemp module
> >> since 2.6.31.4 that made the change in read-out?
> >
> > There were some changes to the coretemp driver in 2.6.32 and 2.6.33,
> > but these shouldn't affect older CPU models and would also not change
> > the readout by a few degrees. You can check if the high temperature
> > limit is the same - then you can assume the driver changes did not
> > affect your system.
> 
> On 2.6.31.4:
> 
> coretemp-isa-0000
> Adapter: ISA adapter
> Core 0:      +33.0°C  (high = +100.0°C, crit = +100.0°C)
> 
> coretemp-isa-0001
> Adapter: ISA adapter
> Core 1:      +33.0°C  (high = +100.0°C, crit = +100.0°C)
> 
> On 2.6.33.1:
> 
> coretemp-isa-0000
> Adapter: ISA adapter
> Core 0:      +34.0°C  (high = +105.0°C, crit = +105.0°C)
> 
> coretemp-isa-0001
> Adapter: ISA adapter
> Core 1:      +34.0°C  (high = +105.0°C, crit = +105.0°C)
> 
> So there is a 5 deg diffrance. I did not change any config files.

I assume these results are using the original coretemp driver of each
kernel. So, you are in one of these cases where the heuristic changes
in 2.6. I can't say whether this is correct in your case or not, this
heuristic is a horrible mess. But the relevant thing here is that your
CPU is actually running _cooler_ in 2.6.33 than in 2.6.31: 71 degrees
below the critical limit, instead of 67 degrees below the limit
previously. In both cases, you have a huge thermal margin, so it's
alright.

The fact that the high limit has the same value as the critical limit
is certainly a bug, as it doesn't make any sense physically.

> >> I think that because that temp. change fan is working more often.
> >>
> >> I'm using Lenovo ThinkPad T61.
> >
> > This suggests that the temperature increase is real and this isn't just
> > a reporting issue. I suggest that you check
> > in /proc/acpi/processor/*/power, the amount of time spent in C2 and C3
> > states. Also install powertop and run it on both kernels, see if 2.6.33
> > has more frequent wake-ups than 2.6.31.
> 
> On 2.6.31.4 powertop says:
> 
> C3 - 90, 7% of the time
> P (800MHz) - 98%
> Wakeups: 627
> 
> On 2.6.33.1 powertop says:
> 
> C3 - 93% of the time
> P (800MHz) - 98%
> Wakeups: 556

So 2.6.33 is indeed better, which is good news.

> Was running the same programs. Kernel 2.6.33.1 was running for about 1h.
> 
> I'm asuming that /proc/acpi/processor/*/power is used by powertop so I'm
> not paseting it. But it also says that C3 is more often used then others.


-- 
Jean Delvare

_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors

Re: [lm-sensors] Coretemp goes up since 2.6.31.4

[dienet]

On Sat, 10 Apr 2010 10:07:30 +0200, Jean Delvare <khali@linux-fr.org>  
wrote:


> I assume these results are using the original coretemp driver of each
> kernel. So, you are in one of these cases where the heuristic changes
> in 2.6. I can't say whether this is correct in your case or not, this
> heuristic is a horrible mess. But the relevant thing here is that your
> CPU is actually running _cooler_ in 2.6.33 than in 2.6.31: 71 degrees
> below the critical limit, instead of 67 degrees below the limit
> previously. In both cases, you have a huge thermal margin, so it's
> alright.

I'm just saying what I see. And I see that sensors show higher temp. on  
2.6.33 then on 2.6.31 - that's all I can say.
Now I'm on 2.6.33.2 and I never see 30 deg. or below. On 2.6.31 30 deg. or  
below was quite normal (but I don't know if that was the *real* temp).

> The fact that the high limit has the same value as the critical limit
> is certainly a bug, as it doesn't make any sense physically.

Any way to fix it? I can say that this bug is quite old or my system isn't  
configured for a long time.

> So 2.6.33 is indeed better, which is good news.

I'll say that after longger work. 1h is nothing, and there is a big margin  
of error.

-- 
pozdr0
dienet

"Old C programmers never die. They're just cast into void."

_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors

Re: [lm-sensors] Coretemp goes up since 2.6.31.4

[Jean Delvare]

On Mon, 12 Apr 2010 19:22:22 +0200, dienet wrote:
> On Sat, 10 Apr 2010 10:07:30 +0200, Jean Delvare <khali@linux-fr.org>  
> wrote:
> 
> 
> > I assume these results are using the original coretemp driver of each
> > kernel. So, you are in one of these cases where the heuristic changes
> > in 2.6. I can't say whether this is correct in your case or not, this
> > heuristic is a horrible mess. But the relevant thing here is that your
> > CPU is actually running _cooler_ in 2.6.33 than in 2.6.31: 71 degrees
> > below the critical limit, instead of 67 degrees below the limit
> > previously. In both cases, you have a huge thermal margin, so it's
> > alright.
> 
> I'm just saying what I see. And I see that sensors show higher temp. on  
> 2.6.33 then on 2.6.31 - that's all I can say.
> Now I'm on 2.6.33.2 and I never see 30 deg. or below. On 2.6.31 30 deg. or  
> below was quite normal (but I don't know if that was the *real* temp).

Again, coretemp is never reporting a real temperature. It is reporting
a margin to a critical, arbitrary limit, which we don't know for sure.
Even if we knew it for sure, the sensor is only accurate when it isn't
too far from the limit. Your CPUs are running cool, way below the
limit, so outside of the sensor's good accuracy range. Neither 2.6.31
nor 2.6.33 is giving you a real temperature.

> > The fact that the high limit has the same value as the critical limit
> > is certainly a bug, as it doesn't make any sense physically.
> 
> Any way to fix it? I can say that this bug is quite old or my system isn't  
> configured for a long time.

No idea, sorry. There are bug reports and patches flying around for the
coretemp driver, but I do not have the time to look into them right now.

-- 
Jean Delvare
http://khali.linux-fr.org/wishlist.html

_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors

Re: [lm-sensors] Coretemp goes up since 2.6.31.4

[dienet]

On Mon, 12 Apr 2010 21:11:52 +0200, Jean Delvare <khali@linux-fr.org>  
wrote:

> Again, coretemp is never reporting a real temperature. It is reporting
> a margin to a critical, arbitrary limit, which we don't know for sure.
> Even if we knew it for sure, the sensor is only accurate when it isn't
> too far from the limit. Your CPUs are running cool, way below the
> limit, so outside of the sensor's good accuracy range. Neither 2.6.31
> nor 2.6.33 is giving you a real temperature.

Ah OK now I understand. But let me assume this: if fan is cool more often  
then on 2.6.31, it means that there is something going on on 2.6.33 -  
something wrong. Am I right?

-- 
pozdr0
dienet

"Old C programmers never die. They're just cast into void."

_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors