Re: [lm-sensors] [PATCH] coretemp driver and Celeron 430 - works

Re: [lm-sensors] [PATCH] coretemp driver and Celeron 430 - works

[Rudolf Marek]

[-- Attachment #1: Type: text/plain, Size: 2189 bytes --]

Hello Jan,

Sorry for the delay, I'm busy with other stuff too.

Thank you for the report, I check the revision guide and everything seems ok. I
would like you to test attach patch. I think it does only what you did + updates
the documentation.

Please can you test that patch?

Signed-off-by: Rudolf Marek <r.marek@assembler.cz>

This patch adds support for the Celeron 4xx based on Core 2 core.

Thanks,

Rudolf




Jan Richling wrote:
> Hi,
> 
> I just tested your coretemp driver on a Celeron 430 (Core 2 based) and
> got unknown CPU model 16 which is triggered by the condition that you
> only accept model 0xe or 0xf as valid devices. The Celeron 4xx is model
> 22 (0x16) so I patched the driver by adding (|| (c->x86_model == 0x16)).
> This works out of the box with no further modifications and shows
> temperatures that are equal to those displayed by a Windows utility on
> the same machine.
> 
> So I can tell you that your driver works for family 6, model 22 also.
> 
> Just to be complete here is the cpuinfo:
> 
> cat /proc/cpuinfo
> processor       : 0
> vendor_id       : GenuineIntel
> cpu family      : 6
> model           : 22
> model name      : Intel(R) Celeron(R) CPU          430  @ 1.80GHz
> stepping        : 1
> cpu MHz         : 2400.147
> cache size      : 512 KB
> fpu             : yes
> fpu_exception   : yes
> cpuid level     : 10
> wp              : yes
> flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge
> mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss tm syscall lm
> constant_tsc pni monitor ds_cpl tm2 ssse3 cx16 xtpr lahf_lm
> bogomips        : 4802.79
> clflush size    : 64
> cache_alignment : 64
> address sizes   : 36 bits physical, 48 bits virtual
> power management:
> 
> (it shows 2400 MHz due to FSB-mod (200->266) and the resulting
> overclock, stock is 9x200=1,8 GHz)
> 
> Greetings and many thanks for the driver,
> 
> Jan
> 
>    ------------------------------------------------------------------
>   / Dr. Jan Richling - http://www.richling.de                      /
>  / Member of Computer Architecture and Communication Group -      /
> / Department of Computer Science - Humboldt University Berlin    /

[-- Attachment #2: add_support_for_sc_celerons.patch --]
[-- Type: text/x-diff, Size: 1367 bytes --]

Index: linux-2.6.23-rc7-rc9/drivers/hwmon/coretemp.c
===================================================================
--- linux-2.6.23-rc7-rc9.orig/drivers/hwmon/coretemp.c	2007-10-07 13:18:51.650651170 +0200
+++ linux-2.6.23-rc7-rc9/drivers/hwmon/coretemp.c	2007-10-07 13:27:05.086652886 +0200
@@ -371,9 +371,10 @@
 	for_each_online_cpu(i) {
 		struct cpuinfo_x86 *c = &(cpu_data)[i];
 
-		/* check if family 6, models e, f */
+		/* check if family 6, models e, f, 16 */
 		if ((c->cpuid_level < 0) || (c->x86 != 0x6) ||
-		    !((c->x86_model == 0xe) || (c->x86_model == 0xf))) {
+		    !((c->x86_model == 0xe) || (c->x86_model == 0xf) ||
+			(c->x86_model == 0x16))) {
 
 			/* supported CPU not found, but report the unknown
 			   family 6 CPU */
Index: linux-2.6.23-rc7-rc9/Documentation/hwmon/coretemp
===================================================================
--- linux-2.6.23-rc7-rc9.orig/Documentation/hwmon/coretemp	2007-10-07 13:27:18.682652097 +0200
+++ linux-2.6.23-rc7-rc9/Documentation/hwmon/coretemp	2007-10-07 13:27:39.226648400 +0200
@@ -4,7 +4,7 @@
 Supported chips:
   * All Intel Core family
     Prefix: 'coretemp'
-    CPUID: family 0x6, models 0xe, 0xf
+    CPUID: family 0x6, models 0xe, 0xf, 0x16
     Datasheet: Intel 64 and IA-32 Architectures Software Developer's Manual
                Volume 3A: System Programming Guide
 

[-- Attachment #3: Type: text/plain, Size: 153 bytes --]

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

Re: [lm-sensors] [PATCH] coretemp driver and Celeron 430 - works

[Jan Richling]

Hi Rudolf,

thanks for your answer and the patch. I just reversed my changes (which
were, by the way, equal to yours except one linebreak) and applied your
patch. Compiles and loads with no problems and gives...

hotaru # sensors
coretemp-isa-0000
Adapter: ISA adapter
Core 0:      +45 C  (high =  +100 C)

-> works perfectly.

Greetings,

Jan

> Sorry for the delay, I'm busy with other stuff too.
> 
> Thank you for the report, I check the revision guide and everything seems ok. I
> would like you to test attach patch. I think it does only what you did + updates
> the documentation.
> 
> Please can you test that patch?
> 
> Signed-off-by: Rudolf Marek <r.marek@assembler.cz>
> 
> This patch adds support for the Celeron 4xx based on Core 2 core.
> 
> Thanks,
> 
> Rudolf
> 
> 
> 
> 
> Jan Richling wrote:
>> Hi,
>>
>> I just tested your coretemp driver on a Celeron 430 (Core 2 based) and
>> got unknown CPU model 16 which is triggered by the condition that you
>> only accept model 0xe or 0xf as valid devices. The Celeron 4xx is model
>> 22 (0x16) so I patched the driver by adding (|| (c->x86_model = 0x16)).
>> This works out of the box with no further modifications and shows
>> temperatures that are equal to those displayed by a Windows utility on
>> the same machine.
>>
>> So I can tell you that your driver works for family 6, model 22 also.
>>
>> Just to be complete here is the cpuinfo:
>>
>> cat /proc/cpuinfo
>> processor       : 0
>> vendor_id       : GenuineIntel
>> cpu family      : 6
>> model           : 22
>> model name      : Intel(R) Celeron(R) CPU          430  @ 1.80GHz
>> stepping        : 1
>> cpu MHz         : 2400.147
>> cache size      : 512 KB
>> fpu             : yes
>> fpu_exception   : yes
>> cpuid level     : 10
>> wp              : yes
>> flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge
>> mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss tm syscall lm
>> constant_tsc pni monitor ds_cpl tm2 ssse3 cx16 xtpr lahf_lm
>> bogomips        : 4802.79
>> clflush size    : 64
>> cache_alignment : 64
>> address sizes   : 36 bits physical, 48 bits virtual
>> power management:
>>
>> (it shows 2400 MHz due to FSB-mod (200->266) and the resulting
>> overclock, stock is 9x200=1,8 GHz)
>>
>> Greetings and many thanks for the driver,
>>
>> Jan
>>
>>    ------------------------------------------------------------------
>>   / Dr. Jan Richling - http://www.richling.de                      /
>>  / Member of Computer Architecture and Communication Group -      /
>> / Department of Computer Science - Humboldt University Berlin    /


-- 

Jan

   ------------------------------------------------------------------
  / Dr. Jan Richling - http://www.richling.de                      /
 / Member of Computer Architecture and Communication Group -      /
/ Department of Computer Science - Humboldt University Berlin    /

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

Re: [lm-sensors] [PATCH] coretemp driver and Celeron 430 - works

[Rudolf Marek]

Jan,

Cool thanks. Please note that the temperature might be more non-physical. So in
your case you have still 55C to overheat. The processor reports temp relative to
trip point, which is "100C" in your case. If it is 85C or 100C works fine on
mobile variants, should not on desktop CPUs.

To sum it up. Even if your temperature is not a real one, you have still 55C to
critical level.

Mark please will you take the patch?

Thanks,

Rudolf

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

Re: [lm-sensors] [PATCH] coretemp driver and Celeron 430 - works

[Jan Richling]

Hi Rudolf,

> Cool thanks. Please note that the temperature might be more non-physical. So in
> your case you have still 55C to overheat. The processor reports temp relative to
> trip point, which is "100C" in your case. If it is 85C or 100C works fine on
> mobile variants, should not on desktop CPUs.
> 
> To sum it up. Even if your temperature is not a real one, you have still 55C to
> critical level.

I know... I did a kind of "test" of this. Running burnP6 with overclock
to 3 GHz and the stock cooler triggered auto shutdown of the CPU in less
than 30 seconds. Now I have a Pentium D820 boxed cooler and the core
goes up to 69 which is still 31 below the critical level.

Regarding the patch - I guess you can add 0x17 soon. This is the new
Penryn core Intel will bring to market in november. Question is only if
there is somebody to test.

Greetings,

Jan

   ------------------------------------------------------------------
  / Dr. Jan Richling - http://www.richling.de                      /
 / Member of Computer Architecture and Communication Group -      /
/ Department of Computer Science - Humboldt University Berlin    /

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

Re: [lm-sensors] [PATCH] coretemp driver and Celeron 430 - works

[Rudolf Marek]

Hello,

Thanks for the update.

> Regarding the patch - I guess you can add 0x17 soon. This is the new
> Penryn core Intel will bring to market in november. Question is only if
> there is somebody to test.
> 

The early revisions seems to have the sensor broken somehow. Need to wait for
someone with released CPU.

Rudolf

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

Re: [lm-sensors] [PATCH] coretemp driver and Celeron 430 - works

[Mark M. Hoffman]

Hi:

* Rudolf Marek <r.marek@assembler.cz> [2007-10-07 13:42:09 +0200]:
> Signed-off-by: Rudolf Marek <r.marek@assembler.cz>
> 
> This patch adds support for the Celeron 4xx based on Core 2 core.

> Index: linux-2.6.23-rc7-rc9/drivers/hwmon/coretemp.c
> =================================> --- linux-2.6.23-rc7-rc9.orig/drivers/hwmon/coretemp.c	2007-10-07 13:18:51.650651170 +0200
> +++ linux-2.6.23-rc7-rc9/drivers/hwmon/coretemp.c	2007-10-07 13:27:05.086652886 +0200
> @@ -371,9 +371,10 @@
>  	for_each_online_cpu(i) {
>  		struct cpuinfo_x86 *c = &(cpu_data)[i];
>  
> -		/* check if family 6, models e, f */
> +		/* check if family 6, models e, f, 16 */
>  		if ((c->cpuid_level < 0) || (c->x86 != 0x6) ||
> -		    !((c->x86_model = 0xe) || (c->x86_model = 0xf))) {
> +		    !((c->x86_model = 0xe) || (c->x86_model = 0xf) ||
> +			(c->x86_model = 0x16))) {
>  
>  			/* supported CPU not found, but report the unknown
>  			   family 6 CPU */
> Index: linux-2.6.23-rc7-rc9/Documentation/hwmon/coretemp
> =================================> --- linux-2.6.23-rc7-rc9.orig/Documentation/hwmon/coretemp	2007-10-07 13:27:18.682652097 +0200
> +++ linux-2.6.23-rc7-rc9/Documentation/hwmon/coretemp	2007-10-07 13:27:39.226648400 +0200
> @@ -4,7 +4,7 @@
>  Supported chips:
>    * All Intel Core family
>      Prefix: 'coretemp'
> -    CPUID: family 0x6, models 0xe, 0xf
> +    CPUID: family 0x6, models 0xe, 0xf, 0x16
>      Datasheet: Intel 64 and IA-32 Architectures Software Developer's Manual
>                 Volume 3A: System Programming Guide
>  

Applied to hwmon-2.6.git/testing, thanks.

-- 
Mark M. Hoffman
mhoffman@lightlink.com


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