coretemp: Support for Intel Atom E6XX CPU (TunnelCreek)?
Hello,
I have a patch (for v2.6.39) which adds support for Intel Atom E6XX (TunnelCreek) to coretemp. It's merely only adding x86_model 0x26. But there are models (e.g. E660 and E660T) with different TjMax, namely 90 degrees C and 110 degrees C. But these different model can't be detected by reading from hardware.
IMO there should be some support to adjust the temperature from userspace. Reading Documentation/hwmon/sysfs-interface only temp1_offset seems to be useable. But I think it is somewhat misleading (especially on multicores), because there must only be one offset.
What do you think about this?
Regards,
Alexander
_______________________________________________
lm-sensors mailing list
[email protected]
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
Quoted 13 lines · Alexander Stein
Hello,
I have a patch (for v2.6.39) which adds support for Intel Atom E6XX (TunnelCreek) to coretemp. It's merely only adding
x86_model0x26. But there are models (e.g. E660 and E660T) with different TjMax, namely 90 degrees C and 110 degrees C. But these different model can't be detected by reading from hardware.
IMO there should be some support to adjust the temperature from userspace. ReadingDocumentation/hwmon/sysfs-interfaceonlytemp1_offsetseems to be useable. But I think it is somewhat misleading (especially on multicores), because there must only be one offset.
What do you think about this?
You can adjust the reported temperature with an entry in sensors3.conf.
Guenter
_______________________________________________
lm-sensors mailing list
[email protected]
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
Hi Alexander,
Quoted 2 lines · Alexander Stein
I have a patch (for v2.6.39) which adds support for Intel Atom E6XX (TunnelCreek) to coretemp. It's merely only adding
x86_model0x26.
You have a patch, great for you. What do you expect if you don't share it with us?
I'm not quite sure what your patch would be doing anyway. Since kernel 2.6.35, supported CPU models are detected using the DTS feature flag rather than the family and model numbers, so your Atom E6XX should be detected just fine.
Note that there was a bug in kernels 2.6.35 to 2.6.39 with regards to TjMax guessing, which was fixed by Gunter Roeck with:
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;hO5f71a7abe329bdad81ee6a8e4545054a7cc30a
You'll have to update to kernel version 2.6.39.2 to get this fix.
Do you happen to know what CPUs model number 0x26 covers? Do you know if this model supports MSR_IA32_TEMPERATURE_TARGET or not? The original Atom (model 0x1c) did not.
Quoted 3 lines
But there are models (e.g. E660 and E660T) with different TjMax, namely 90 degrees C and 110 degrees C. But these different model can't be detected by reading from hardware.
I would appreciate a patch to Documentation/hwmon/coretemp adding the known TjMax for these new Atom models.
BTW, is it really impossible to identify these models with a different TjMax? Don't the strings "E660" and "E660T" appear in the respective "model name" entries in /proc/cpuinfo?
Quoted 4 lines
IMO there should be some support to adjust the temperature from userspace. Reading
Documentation/hwmon/sysfs-interfaceonlytemp1_offsetseems to be useable. But I think it is somewhat misleading (especially on multicores), because there must only be one offset.
No, tempN_offset isn't suitable for this case, as it would only shift the current temperature and not the limits.
Instead, we could detect the specific CPUs using the model name string and adjust TjMax accordingly. And/or we could let the user override TjMax through a module parameter (I doubt anyone runs a system with CPUs with different TjMax values.)
Signature · 6 lines
Jean Delvare _______________________________________________ lm-sensors mailing list [email protected] http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
Quoted 8 lines · On Monday 27 June 2011 16:41:08 Jean Delvare wrote:
Hi Alexander,
Quoted 2 lines · Alexander Stein
I have a patch (for v2.6.39) which adds support for Intel Atom E6XX (TunnelCreek) to coretemp. It's merely only adding
x86_model0x26.You have a patch, great for you. What do you expect if you don't share it with us?
If you insist on seeing that patch (for 2.6.39 btw):
Documentation/hwmon/coretemp +2 -1
@@ -6,7 +6,8 @@ Supported chips: Prefix: 'coretemp' CPUID: family 0x6, models 0xe (Pentium M DC), 0xf (Core 2 DC 65nm), 0x16 (Core 2 SC 65nm), 0x17 (Penryn 45nm), - 0x1a (Nehalem), 0x1c (Atom), 0x1e (Lynnfield) + 0x1a (Nehalem), 0x1c (Atom), 0x1e (Lynnfield), + 0x26 (Tunnelcreek) Datasheet: Intel 64 and IA-32 Architectures Software Developer's Manual Volume 3A: System Programming Guide http://softwarecommunity.intel.com/Wiki/Mobility/720.htm
drivers/hwmon/coretemp.c +0 -0
@@ -173,9 +173,9 @@ static int __devinit adjust_tjmax(struct cpuinfo_x86 *c,
u32 id, struct device *
usemsr_ee = 0;
}
- /* Atom CPUs */
+ /* Atom and TunnelCreek CPUs */
- if (c->x86_model = 0x1c) {
+ if ((c->x86_model = 0x1c) || (c->x86_model = 0x26)) {
usemsr_ee = 0;
host_bridge = pci_get_bus_and_slot(0, PCI_DEVFN(0, 0));
@@ -283,6 +283,7 @@ static int __devinit get_tjmax(struct cpuinfo_x86 *c, u32 id,
return 100000;
case 0x17:
case 0x1c: /* Atom CPUs */
+ case 0x26: /* TunnelCreek CPU */
return adjust_tjmax(c, id, dev);
default:
dev_warn(dev, "CPU (model=0x%x) is not supported yet,"
@@ -361,7 +362,8 @@ static int __devinit coretemp_probe(struct platform_device *pdev)
* Atoms don't have it either.
*/
- if ((c->x86_model > 0xe) && (c->x86_model != 0x1c)) {
+ if ((c->x86_model > 0xe) && (c->x86_model != 0x1c)
+ && (c->x86_model != 0x26)) {
err = rdmsr_safe_on_cpu(data->id, MSR_IA32_TEMPERATURE_TARGET,
&eax, &edx);
if (err) {
The patch without hunk 2 still stays valid to me for current git master.
Quoted 4 lines
I'm not quite sure what your patch would be doing anyway. Since kernel 2.6.35, supported CPU models are detected using the DTS feature flag rather than the family and model numbers, so your Atom E6XX should be detected just fine.
Those 3 output lines don't seem like the mode is detected. coretemp coretemp.0: Unable to read TjMax from CPU.
coretemp coretemp.0: CPU (model=0x26) is not supported yet, using default TjMax of 100C.
coretemp coretemp.0: Unable to read IA32_TEMPERATURE_TARGET MSR
Quoted 5 lines
Note that there was a bug in kernels 2.6.35 to 2.6.39 with regards to TjMax guessing, which was fixed by Gunter Roeck with:
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h
O5f71a7abe329bdad81ee6a8e4545054a7cc30a You'll have to update to kernel version 2.6.39.2 to get this fix.
Cherry-picking this patch, only reduces the output to those two lines:
coretemp coretemp.0: Unable to read TjMax from CPU. coretemp coretemp.0: Unable to read IA32_TEMPERATURE_TARGET MSR
Quoted 3 lines
Do you happen to know what CPUs model number
0x26covers? Do you know if this model supportsMSR_IA32_TEMPERATURE_TARGETor not? The original Atom (model0x1c) did not.
Both the linux kernel (see above) and the bootloader can't use MSR_IA32_TEMPERATURE_TARGET, so I guess this model does not support this feature.
I don't know which models have number 0x26, despite the E6xx.
Quoted 10 lines
Quoted 3 lines
But there are models (e.g. E660 and E660T) with different TjMax, namely 90 degrees C and 110 degrees C. But these different model can't be detected by reading from hardware.
I would appreciate a patch to
Documentation/hwmon/coretempadding the known TjMax for these new Atom models.BTW, is it really impossible to identify these models with a different TjMax? Don't the strings "E660" and "E660T" appear in the respective "model name" entries in
/proc/cpuinfo?
Here's the cpuinfo output. I didnt find anything which would allow this distinction.
# cat /proc/cpuinfo processor : 0 vendor_id : GenuineIntel cpu family : 6 model : 38 model name : Genuine Intel(R) CPU @ 1.30GHz stepping : 1 cpu MHz : 1299.936 cache size : 512 KB physical id : 0 siblings : 2 core id : 0 cpu cores : 1 apicid : 0 initial apicid : 0 fdiv_bug : no hlt_bug : no f00f_bug : no coma_bug : no 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 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx lm constant_tsc arch_perfmon pebs bts aperfmperf pni dtes64 monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr pdcm movbe lahf_lm dts tpr_shadow vnmi bogomips : 2599.87 clflush size : 64 cache_alignment : 64 address sizes : 32 bits physical, 48 bits virtual power management:
processor : 1vendor_id : GenuineIntel
cpu family : 6
model : 38
model name : Genuine Intel(R) CPU @ 1.30GHz
stepping : 1
cpu MHz : 1299.936
cache size : 512 KB
physical id : 0
siblings : 2
core id : 0
cpu cores : 1
apicid : 1
initial apicid : 1fdiv_bug : nohlt_bug : nof00f_bug : nocoma_bug : no
fpu : yesfpu_exception : yes
cpuid level : 10
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca
cmov pat clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx lm constant_tsc arch_perfmon pebs bts aperfmperf pni dtes64 monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr pdcm movbe lahf_lm dts tpr_shadow vnmi
bogomips : 2599.54
clflush size : 64cache_alignment : 64
address sizes : 32 bits physical, 48 bits virtual
power management:
Quoted 5 lines
Quoted 4 lines
IMO there should be some support to adjust the temperature from userspace. Reading
Documentation/hwmon/sysfs-interfaceonlytemp1_offsetseems to be useable. But I think it is somewhat misleading (especially on multicores), because there must only be one offset.
Quoted 7 lines
No,
tempN_offsetisn't suitable for this case, as it would only shift the current temperature and not the limits.Instead, we could detect the specific CPUs using the model name string and adjust TjMax accordingly. And/or we could let the user override TjMax through a module parameter (I doubt anyone runs a system with CPUs with different TjMax values.)
As there seem to be no model string to distinct, only the module parameter seems valid to do this change.
Regards,
Alexander
_______________________________________________
lm-sensors mailing list
[email protected]
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
Quoted 6 lines · Jean Delvare
Hi Alexander,
Quoted 2 lines · Alexander Stein
I have a patch (for v2.6.39) which adds support for Intel Atom E6XX (TunnelCreek) to coretemp. It's merely only adding
x86_model0x26.
Quoted 8 lines
You have a patch, great for you. What do you expect if you don't share it with us?
I'm not quite sure what your patch would be doing anyway. Since kernel 2.6.35, supported CPU models are detected using the DTS feature flag rather than the family and model numbers, so your Atom E6XX should be detected just fine.
Maybe it is for Tjmax detection ?
Quoted 13 lines
Note that there was a bug in kernels 2.6.35 to 2.6.39 with regards to TjMax guessing, which was fixed by Gunter Roeck with:
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;hO5f71a7abe329bdad81ee6a8e4545054a7cc30a
You'll have to update to kernel version 2.6.39.2 to get this fix.Do you happen to know what CPUs model number
0x26covers? Do you know if this model supportsMSR_IA32_TEMPERATURE_TARGETor not? The original Atom (model0x1c) did not.Quoted 3 lines
But there are models (e.g. E660 and E660T) with different TjMax, namely 90 degrees C and 110 degrees C. But these different model can't be detected by reading from hardware.
Quoted 6 lines
I would appreciate a patch to
Documentation/hwmon/coretempadding the known TjMax for these new Atom models.BTW, is it really impossible to identify these models with a different TjMax? Don't the strings "E660" and "E660T" appear in the respective "model name" entries in
/proc/cpuinfo?
I thought about replacing the manual TjMax detection code with code using the model string to take care of problems like this - essentially by providing a table with entries { model string, Tjmax } for each CPU with Tjmax other than 100 degrees C. This way we could get rid of some of the odd code we have today. Would that make sense ?
Thanks,
Guenter
_______________________________________________
lm-sensors mailing list
[email protected]
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
Hi Guenter,
Quoted 5 lines
I thought about replacing the manual TjMax detection code with code using the model string to take care of problems like this - essentially by providing a table with entries { model string, Tjmax } for each CPU with Tjmax other than 100 degrees C. This way we could get rid of some of the odd code we have today. Would that make sense ?
I completely agree with you here.
This way, a lot of code in get_tjmax will be reduced, by just a search through some array.
Thanks,
Durga
Quoted 8 lines · Durgadoss
Hi Guenter,
Quoted 5 lines
I thought about replacing the manual TjMax detection code with code using the model string to take care of problems like this - essentially by providing a table with entries { model string, Tjmax } for each CPU with Tjmax other than 100 degrees C. This way we could get rid of some of the odd code we have today. Would that make sense ?
Quoted 4 lines
I completely agree with you here.
This way, a lot of code inget_tjmaxwill be reduced, by just a search through some array.
Unfortunately, looking at Alexander's /proc/cpuinfo output, it looks like the E6xx chips have no mention of E6xx in the CPU ID string, which unfortunately kills the idea.
Do you know if there is _any_ means to distinguish E6xx from E6xxT ?
Thanks,
Guenter
_______________________________________________
lm-sensors mailing list
[email protected]
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
Hi Guenter,
Quoted 6 lines
Unfortunately, looking at Alexander's
/proc/cpuinfooutput, it looks like the E6xx chips have no mention of E6xx in the CPU ID string, which unfortunately kills the idea.Do you know if there is _any_ means to distinguish E6xx from E6xxT ?
Sorry I do not know. But I can ask some folks here and get back to you guys. Should not take more time, once I identify the right person to ask :-)
Thanks,
Durga
���{.n�+�������+%�����ݶ��w��{.n�+����{��G�����{ay�ʇڙ�,j��f���h��������z_�(�階�ݢj"���m������G������?����&���~��iO��z��v�^�m����������?�I�
Hello,
Quoted 9 lines · Durgadoss
Quoted 5 lines
Unfortunately, looking at Alexander's
/proc/cpuinfooutput, it looks like the E6xx chips have no mention of E6xx in the CPU ID string, which unfortunately kills the idea.Do you know if there is _any_ means to distinguish E6xx from E6xxT ?
Sorry I do not know. But I can ask some folks here and get back to you guys. Should not take more time, once I identify the right person to ask :-)
Any news on this topic?
Regards,
Alexander
_______________________________________________
lm-sensors mailing list
[email protected]
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
Hi All,
Yes I checked with some Atom Guys here. Apparently there is no way to know it from the OS Level.
Sorry,
Durga
Quoted 18 lines
-----Original Message-----
From: Alexander Stein [mailto:[email protected]]
Sent: Wednesday, August 03, 2011 6:28 PM
To: R, Durgadoss
Cc: [email protected]; Jean Delvare; Yu, Fenghua; lm-sensors@lm- sensors.org; [email protected]
Subject: Re: [lm-sensors] coretemp: Support for Intel Atom E6XX CPU (TunnelCreek)?Hello,
Quoted 6 lines · Durgadoss
Quoted 5 lines
Unfortunately, looking at Alexander's
/proc/cpuinfooutput, it looks like the E6xx chips have no mention of E6xx in the CPU ID string, which unfortunately kills the idea.Do you know if there is _any_ means to distinguish E6xx from E6xxT ?
Quoted 8 lines
Quoted 3 lines
Sorry I do not know. But I can ask some folks here and get back to you guys. Should not take more time, once I identify the right person to ask :-)
Any news on this topic?
Regards,
Alexander
���{.n�+�������+%�����ݶ��w��{.n�+����{��G�����{ay�ʇڙ�,j��f���h��������z_�(�階�ݢj"���m������G������?����&���~��iO��z��v�^�m����������?�I�
Hi,
Quoted 13 lines · Durgadoss
Yes I checked with some Atom Guys here. Apparently there is no way to know it from the OS Level.
Quoted 10 lines · Durgadoss
Quoted 9 lines
Quoted 5 lines
Unfortunately, looking at Alexander's
/proc/cpuinfooutput, it looks like the E6xx chips have no mention of E6xx in the CPU ID string, which unfortunately kills the idea.Do you know if there is _any_ means to distinguish E6xx from E6xxT ?
Sorry I do not know. But I can ask some folks here and get back to you guys. Should not take more time, once I identify the right person to ask :-)
So, is there a possibility to add a module parameter to force a either specific tjmax or the T model for E6xx?
Regards,
Alexander
_______________________________________________
lm-sensors mailing list
[email protected]
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
Hi Alexander,
Quoted 9 lines · Alexander Stein
Quoted 8 lines · Durgadoss
Yes I checked with some Atom Guys here. Apparently there is no way to know it from the OS Level.
Quoted 5 lines · Durgadoss
Quoted 4 lines
Quoted 4 lines
Unfortunately, looking at Alexander's
/proc/cpuinfooutput, it looks like the E6xx chips have no mention of E6xx in the CPU ID string, which unfortunately kills the idea.
Quoted 8 lines
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
Quoted 1 line
Do you know if there is _any_ means to distinguish E6xx from E6xxT ?
Sorry I do not know. But I can ask some folks here and get back to you guys. Should not take more time, once I identify the right person to ask :-)
So, is there a possibility to add a module parameter to force a either specific tjmax or the T model for E6xx?
A parameter to force the T model for E6xx would be very very specific, while the problem at hand is much broader: the heuristics in adjust_tjmax() are best effort but largely insufficient in many (most?) cases. So, as we have been enable to provide a relative temperature scale interface in the past 2 years, a module parameter letting the user set TjMax as needed seems the best approach to me. I'll submit a patch doing this in a moment.
Signature · 6 lines
Jean Delvare _______________________________________________ lm-sensors mailing list [email protected] http://lists.lm-sensors.org/mailman/listinfo/lm-sensors