linux-acpi.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* Preventing ACPI from Damaging Your CPU
@ 2006-10-12 14:26 Mathew Brown
  2006-10-12 14:46 ` Rafael J. Wysocki
  2006-10-13  1:24 ` Matthew Garrett
  0 siblings, 2 replies; 10+ messages in thread
From: Mathew Brown @ 2006-10-12 14:26 UTC (permalink / raw)
  To: linux-acpi

Hi,
  I've been reading an alarming article on getting Gentoo to work on the
  HP Compaq nx6325
  (http://gentoo-wiki.com/HARDWARE_Gentoo_on_HP_Compaq_nx6325) and the
  serious ACPI problems that were encountered.  The author states after
  mentioning how to apply certain patches, "Without that, the CPU might
  overheat and get damaged during heavy load (such as compiling a Gentoo
  stage ...)!"

  What would be the safest way to ensure that ACPI doesn't do this to
  your machine / laptop without disabling ACPI?  Would recompiling the
  DSDT using the Intel compiler and re-inserting it be enough to ensure
  that ACPI doesn't cause your CPU to overheat or get damaged?  Thanks
  for your help.
-- 
  Mathew Brown
  mathewbrown@fastmail.fm

-- 
http://www.fastmail.fm - A no graphics, no pop-ups email service


^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: Preventing ACPI from Damaging Your CPU
  2006-10-12 14:26 Mathew Brown
@ 2006-10-12 14:46 ` Rafael J. Wysocki
  2006-10-12 16:51   ` Mathew Brown
  2006-10-13  1:24 ` Matthew Garrett
  1 sibling, 1 reply; 10+ messages in thread
From: Rafael J. Wysocki @ 2006-10-12 14:46 UTC (permalink / raw)
  To: Mathew Brown; +Cc: linux-acpi

On Thursday, 12 October 2006 16:26, Mathew Brown wrote:
> Hi,
>   I've been reading an alarming article on getting Gentoo to work on the
>   HP Compaq nx6325
>   (http://gentoo-wiki.com/HARDWARE_Gentoo_on_HP_Compaq_nx6325) and the
>   serious ACPI problems that were encountered.  The author states after
>   mentioning how to apply certain patches, "Without that, the CPU might
>   overheat and get damaged during heavy load (such as compiling a Gentoo
>   stage ...)!"
> 
>   What would be the safest way to ensure that ACPI doesn't do this to
>   your machine / laptop without disabling ACPI?  Would recompiling the
>   DSDT using the Intel compiler and re-inserting it be enough to ensure
>   that ACPI doesn't cause your CPU to overheat or get damaged?  Thanks
>   for your help.

Please see http://bugzilla.kernel.org/show_bug.cgi?id=7122, especially
http://bugzilla.kernel.org/show_bug.cgi?id=7122#c27 and below.

Greetings,
Rafael


-- 
You never change things by fighting the existing reality.
		R. Buckminster Fuller

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: Preventing ACPI from Damaging Your CPU
  2006-10-12 14:46 ` Rafael J. Wysocki
@ 2006-10-12 16:51   ` Mathew Brown
  2006-10-12 19:34     ` Rafael J. Wysocki
  0 siblings, 1 reply; 10+ messages in thread
From: Mathew Brown @ 2006-10-12 16:51 UTC (permalink / raw)
  To: Rafael J. Wysocki; +Cc: linux-acpi

Thanks for your reply Rafael.  However, I was looking for a generic
solution.  I don't have an nx6325 but was thinking of purchasing a
laptop and was afraid that ACPI could damage / overheat by CPUs.  I
think the link you provided is specific to the nx6325 unless I'm
mistaken.  Is there any generic solution?  Would recompiling the DSDT
and re-inserting it guarantee that no overheat / damage happens
regardless of the laptop / machine?  Thanks.


On Thu, 12 Oct 2006 16:46:19 +0200, "Rafael J. Wysocki" <rjw@sisk.pl>
said:
> On Thursday, 12 October 2006 16:26, Mathew Brown wrote:
> > Hi,
> >   I've been reading an alarming article on getting Gentoo to work on the
> >   HP Compaq nx6325
> >   (http://gentoo-wiki.com/HARDWARE_Gentoo_on_HP_Compaq_nx6325) and the
> >   serious ACPI problems that were encountered.  The author states after
> >   mentioning how to apply certain patches, "Without that, the CPU might
> >   overheat and get damaged during heavy load (such as compiling a Gentoo
> >   stage ...)!"
> > 
> >   What would be the safest way to ensure that ACPI doesn't do this to
> >   your machine / laptop without disabling ACPI?  Would recompiling the
> >   DSDT using the Intel compiler and re-inserting it be enough to ensure
> >   that ACPI doesn't cause your CPU to overheat or get damaged?  Thanks
> >   for your help.
> 
> Please see http://bugzilla.kernel.org/show_bug.cgi?id=7122, especially
> http://bugzilla.kernel.org/show_bug.cgi?id=7122#c27 and below.
> 
> Greetings,
> Rafael
> 
> 
> -- 
> You never change things by fighting the existing reality.
> 		R. Buckminster Fuller
-- 
  Mathew Brown
  mathewbrown@fastmail.fm

-- 
http://www.fastmail.fm - A fast, anti-spam email service.


^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: Preventing ACPI from Damaging Your CPU
  2006-10-12 16:51   ` Mathew Brown
@ 2006-10-12 19:34     ` Rafael J. Wysocki
  0 siblings, 0 replies; 10+ messages in thread
From: Rafael J. Wysocki @ 2006-10-12 19:34 UTC (permalink / raw)
  To: Mathew Brown; +Cc: linux-acpi

On Thursday, 12 October 2006 18:51, Mathew Brown wrote:
> Thanks for your reply Rafael.  However, I was looking for a generic
> solution.  I don't have an nx6325 but was thinking of purchasing a
> laptop and was afraid that ACPI could damage / overheat by CPUs.  I
> think the link you provided is specific to the nx6325 unless I'm
> mistaken.  Is there any generic solution?  Would recompiling the DSDT
> and re-inserting it guarantee that no overheat / damage happens
> regardless of the laptop / machine?

No, it wouldn't, AFAICT.

Greetings,
Rafael


> On Thu, 12 Oct 2006 16:46:19 +0200, "Rafael J. Wysocki" <rjw@sisk.pl>
> said:
> > On Thursday, 12 October 2006 16:26, Mathew Brown wrote:
> > > Hi,
> > >   I've been reading an alarming article on getting Gentoo to work on the
> > >   HP Compaq nx6325
> > >   (http://gentoo-wiki.com/HARDWARE_Gentoo_on_HP_Compaq_nx6325) and the
> > >   serious ACPI problems that were encountered.  The author states after
> > >   mentioning how to apply certain patches, "Without that, the CPU might
> > >   overheat and get damaged during heavy load (such as compiling a Gentoo
> > >   stage ...)!"
> > > 
> > >   What would be the safest way to ensure that ACPI doesn't do this to
> > >   your machine / laptop without disabling ACPI?  Would recompiling the
> > >   DSDT using the Intel compiler and re-inserting it be enough to ensure
> > >   that ACPI doesn't cause your CPU to overheat or get damaged?  Thanks
> > >   for your help.
> > 
> > Please see http://bugzilla.kernel.org/show_bug.cgi?id=7122, especially
> > http://bugzilla.kernel.org/show_bug.cgi?id=7122#c27 and below.
> > 
> > Greetings,
> > Rafael
> > 
> > 
> > -- 
> > You never change things by fighting the existing reality.
> > 		R. Buckminster Fuller
> -- 
>   Mathew Brown
>   mathewbrown@fastmail.fm
> 

-- 
You never change things by fighting the existing reality.
		R. Buckminster Fuller

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: Preventing ACPI from Damaging Your CPU
  2006-10-12 14:26 Mathew Brown
  2006-10-12 14:46 ` Rafael J. Wysocki
@ 2006-10-13  1:24 ` Matthew Garrett
  2006-10-13  4:41   ` Mathew Brown
  1 sibling, 1 reply; 10+ messages in thread
From: Matthew Garrett @ 2006-10-13  1:24 UTC (permalink / raw)
  To: Mathew Brown; +Cc: linux-acpi

On Thu, Oct 12, 2006 at 07:26:00AM -0700, Mathew Brown wrote:

>   I've been reading an alarming article on getting Gentoo to work on the
>   HP Compaq nx6325
>   (http://gentoo-wiki.com/HARDWARE_Gentoo_on_HP_Compaq_nx6325) and the
>   serious ACPI problems that were encountered.  The author states after
>   mentioning how to apply certain patches, "Without that, the CPU might
>   overheat and get damaged during heavy load (such as compiling a Gentoo
>   stage ...)!"

The author is wrong. Recent HP laptops all have hardcoded support in the 
fan control hardware, which will force the fan on if the temperature 
rises above 80 degrees.
-- 
Matthew Garrett | mjg59@srcf.ucam.org

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: Preventing ACPI from Damaging Your CPU
  2006-10-13  1:24 ` Matthew Garrett
@ 2006-10-13  4:41   ` Mathew Brown
  2006-10-13  4:53     ` Matthew Garrett
  0 siblings, 1 reply; 10+ messages in thread
From: Mathew Brown @ 2006-10-13  4:41 UTC (permalink / raw)
  To: Matthew Garrett; +Cc: linux-acpi

Thanks Matthew.  Just out of curiosity, how did you find out about this
feature (forcing the fan at 80C)?  Also, is this specific to HP or do
other vendors also support this?  Finally, is 80C safe for the CPU
(Intel or AMD?)?  I read that Opteron chips could be damaged if their
temperature reaches 70C.  Thanks for your help.

On Fri, 13 Oct 2006 02:24:30 +0100, "Matthew Garrett"
<mjg59@srcf.ucam.org> said:
> On Thu, Oct 12, 2006 at 07:26:00AM -0700, Mathew Brown wrote:
> 
> >   I've been reading an alarming article on getting Gentoo to work on the
> >   HP Compaq nx6325
> >   (http://gentoo-wiki.com/HARDWARE_Gentoo_on_HP_Compaq_nx6325) and the
> >   serious ACPI problems that were encountered.  The author states after
> >   mentioning how to apply certain patches, "Without that, the CPU might
> >   overheat and get damaged during heavy load (such as compiling a Gentoo
> >   stage ...)!"
> 
> The author is wrong. Recent HP laptops all have hardcoded support in the 
> fan control hardware, which will force the fan on if the temperature 
> rises above 80 degrees.
> -- 
> Matthew Garrett | mjg59@srcf.ucam.org
-- 
  Mathew Brown
  mathewbrown@fastmail.fm

-- 
http://www.fastmail.fm - mmm... Fastmail...


^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: Preventing ACPI from Damaging Your CPU
  2006-10-13  4:41   ` Mathew Brown
@ 2006-10-13  4:53     ` Matthew Garrett
  2006-10-13  4:56       ` Mathew Brown
  0 siblings, 1 reply; 10+ messages in thread
From: Matthew Garrett @ 2006-10-13  4:53 UTC (permalink / raw)
  To: Mathew Brown; +Cc: linux-acpi

On Thu, Oct 12, 2006 at 09:41:15PM -0700, Mathew Brown wrote:
> Thanks Matthew.  Just out of curiosity, how did you find out about this
> feature (forcing the fan at 80C)?  Also, is this specific to HP or do
> other vendors also support this?  Finally, is 80C safe for the CPU
> (Intel or AMD?)?  I read that Opteron chips could be damaged if their
> temperature reaches 70C.  Thanks for your help.

A certain amount of reverse engineering of HP hardware while I was 
trying to diagnose some fan problems before. The fan control chip that 
HP use is pretty standard, so I think there are docs around somewhere.

-- 
Matthew Garrett | mjg59@srcf.ucam.org

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: Preventing ACPI from Damaging Your CPU
  2006-10-13  4:53     ` Matthew Garrett
@ 2006-10-13  4:56       ` Mathew Brown
  2006-10-13 11:31         ` Rafael J. Wysocki
  0 siblings, 1 reply; 10+ messages in thread
From: Mathew Brown @ 2006-10-13  4:56 UTC (permalink / raw)
  To: Matthew Garrett; +Cc: linux-acpi

Thanks.  But is 80C safe for a CPU?  As I mentioned, 70C is considered
dangerous for an Opteron chip.  What type of temperatures can the
Turions and the C2D support?

On Fri, 13 Oct 2006 05:53:39 +0100, "Matthew Garrett"
<mjg59@srcf.ucam.org> said:
> On Thu, Oct 12, 2006 at 09:41:15PM -0700, Mathew Brown wrote:
> > Thanks Matthew.  Just out of curiosity, how did you find out about this
> > feature (forcing the fan at 80C)?  Also, is this specific to HP or do
> > other vendors also support this?  Finally, is 80C safe for the CPU
> > (Intel or AMD?)?  I read that Opteron chips could be damaged if their
> > temperature reaches 70C.  Thanks for your help.
> 
> A certain amount of reverse engineering of HP hardware while I was 
> trying to diagnose some fan problems before. The fan control chip that 
> HP use is pretty standard, so I think there are docs around somewhere.
> 
> -- 
> Matthew Garrett | mjg59@srcf.ucam.org
-- 
  Mathew Brown
  mathewbrown@fastmail.fm

-- 
http://www.fastmail.fm - Send your email first class


^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: Preventing ACPI from Damaging Your CPU
  2006-10-13  4:56       ` Mathew Brown
@ 2006-10-13 11:31         ` Rafael J. Wysocki
  0 siblings, 0 replies; 10+ messages in thread
From: Rafael J. Wysocki @ 2006-10-13 11:31 UTC (permalink / raw)
  To: Mathew Brown; +Cc: Matthew Garrett, linux-acpi

On Friday, 13 October 2006 06:56, Mathew Brown wrote:
> Thanks.  But is 80C safe for a CPU?  As I mentioned, 70C is considered
> dangerous for an Opteron chip.  What type of temperatures can the
> Turions and the C2D support?

I think HP would force the fan at lower temperatures if 80C were considered
as dangerous ....

Greetings,
Rafael


> On Fri, 13 Oct 2006 05:53:39 +0100, "Matthew Garrett"
> <mjg59@srcf.ucam.org> said:
> > On Thu, Oct 12, 2006 at 09:41:15PM -0700, Mathew Brown wrote:
> > > Thanks Matthew.  Just out of curiosity, how did you find out about this
> > > feature (forcing the fan at 80C)?  Also, is this specific to HP or do
> > > other vendors also support this?  Finally, is 80C safe for the CPU
> > > (Intel or AMD?)?  I read that Opteron chips could be damaged if their
> > > temperature reaches 70C.  Thanks for your help.
> > 
> > A certain amount of reverse engineering of HP hardware while I was 
> > trying to diagnose some fan problems before. The fan control chip that 
> > HP use is pretty standard, so I think there are docs around somewhere.
> > 
> > -- 
> > Matthew Garrett | mjg59@srcf.ucam.org
> -- 
>   Mathew Brown
>   mathewbrown@fastmail.fm
> 

-- 
You never change things by fighting the existing reality.
		R. Buckminster Fuller

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: Preventing ACPI from Damaging Your CPU
@ 2006-10-15 14:16 Mathew Brown
  0 siblings, 0 replies; 10+ messages in thread
From: Mathew Brown @ 2006-10-15 14:16 UTC (permalink / raw)
  To: Rafael J. Wysocki; +Cc: Matthew Garrett, linux-acpi

So just to re-iterate, even if you recompile the DSDT so that it
compiles without any errors, there's no guarantee that ACPI won't
kill/damage your processor unless the vendor has something special that
can't be over-ridden by ACPI (such as is the case with HP's newer
motherboards/processors).  So how does Windows manage to get around this
issue?  From my limited understanding of ACPI, the problems seem to be
due to the compilation of the DSDT with the Microsoft Compiler which
Linux would have trouble dealing with because it's too lack on what gets
through.  Am I understanding this issue properly?  Thanks for your help.

Rafael J. Wysocki wrote:
> On Friday, 13 October 2006 06:56, Mathew Brown wrote:
>> Thanks.  But is 80C safe for a CPU?  As I mentioned, 70C is considered
>> dangerous for an Opteron chip.  What type of temperatures can the
>> Turions and the C2D support?
>
> I think HP would force the fan at lower temperatures if 80C were considered
> as dangerous ....
>
> Greetings,
> Rafael
>
>
>> On Fri, 13 Oct 2006 05:53:39 +0100, "Matthew Garrett"
>> <mjg59@srcf.ucam.org> said:
>>> On Thu, Oct 12, 2006 at 09:41:15PM -0700, Mathew Brown wrote:
>>>> Thanks Matthew.  Just out of curiosity, how did you find out about this
>>>> feature (forcing the fan at 80C)?  Also, is this specific to HP or do
>>>> other vendors also support this?  Finally, is 80C safe for the CPU
>>>> (Intel or AMD?)?  I read that Opteron chips could be damaged if their
>>>> temperature reaches 70C.  Thanks for your help.
>>> A certain amount of reverse engineering of HP hardware while I was 
>>> trying to diagnose some fan problems before. The fan control chip that 
>>> HP use is pretty standard, so I think there are docs around somewhere.
>>>
>>> -- 
>>> Matthew Garrett | mjg59@srcf.ucam.org
>> -- 
>>   Mathew Brown
>>   mathewbrown@fastmail.fm
>>
>

-- 
  Mathew Brown
  mathewbrown@fastmail.fm

-- 
http://www.fastmail.fm - Faster than the air-speed velocity of an
                          unladen european swallow


^ permalink raw reply	[flat|nested] 10+ messages in thread

end of thread, other threads:[~2006-10-15 14:16 UTC | newest]

Thread overview: 10+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2006-10-15 14:16 Preventing ACPI from Damaging Your CPU Mathew Brown
  -- strict thread matches above, loose matches on Subject: below --
2006-10-12 14:26 Mathew Brown
2006-10-12 14:46 ` Rafael J. Wysocki
2006-10-12 16:51   ` Mathew Brown
2006-10-12 19:34     ` Rafael J. Wysocki
2006-10-13  1:24 ` Matthew Garrett
2006-10-13  4:41   ` Mathew Brown
2006-10-13  4:53     ` Matthew Garrett
2006-10-13  4:56       ` Mathew Brown
2006-10-13 11:31         ` Rafael J. Wysocki

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).