[PATCH] x86/mce: Lower throttling MCE messages to warnings

[PATCH] x86/mce: Lower throttling MCE messages to warnings

[Benjamin Berg]

On modern CPUs it is quite normal that the temperature limits are
reached and the CPU is throttled. In fact, often the thermal design is
not sufficient to cool the CPU at full load and limits can quickly be
reached when a burst in load happens. This will even happen with
technologies like RAPL limitting the long term power consumption of
the package.

So these messages do not usually indicate a hardware issue (e.g.
insufficient cooling). Log them as warnings to avoid confusion about
their severity.

Signed-off-by: Benjamin Berg <bberg@redhat.com>
Tested-by: Christian Kellner <ckellner@redhat.com>
---
 arch/x86/kernel/cpu/mce/therm_throt.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86/kernel/cpu/mce/therm_throt.c b/arch/x86/kernel/cpu/mce/therm_throt.c
index 6e2becf547c5..bc441d68d060 100644
--- a/arch/x86/kernel/cpu/mce/therm_throt.c
+++ b/arch/x86/kernel/cpu/mce/therm_throt.c
@@ -188,7 +188,7 @@ static void therm_throt_process(bool new_event, int event, int level)
 	/* if we just entered the thermal event */
 	if (new_event) {
 		if (event == THERMAL_THROTTLING_EVENT)
-			pr_crit("CPU%d: %s temperature above threshold, cpu clock throttled (total events = %lu)\n",
+			pr_warn("CPU%d: %s temperature above threshold, cpu clock throttled (total events = %lu)\n",
 				this_cpu,
 				level == CORE_LEVEL ? "Core" : "Package",
 				state->count);
-- 
2.23.0

Re: [PATCH] x86/mce: Lower throttling MCE messages to warnings

[Hans de Goede]

Hi,

On 09-10-2019 17:54, Benjamin Berg wrote:
> On modern CPUs it is quite normal that the temperature limits are
> reached and the CPU is throttled. In fact, often the thermal design is
> not sufficient to cool the CPU at full load and limits can quickly be
> reached when a burst in load happens. This will even happen with
> technologies like RAPL limitting the long term power consumption of
> the package.
> 
> So these messages do not usually indicate a hardware issue (e.g.
> insufficient cooling). Log them as warnings to avoid confusion about
> their severity.
> 
> Signed-off-by: Benjamin Berg <bberg@redhat.com>
> Tested-by: Christian Kellner <ckellner@redhat.com>

Ah, yes lets please lower the log-prio of these messages:

Reviewed-by: Hans de Goede <hdegoede@redhat.com>

Regards,

Hans



> ---
>   arch/x86/kernel/cpu/mce/therm_throt.c | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/arch/x86/kernel/cpu/mce/therm_throt.c b/arch/x86/kernel/cpu/mce/therm_throt.c
> index 6e2becf547c5..bc441d68d060 100644
> --- a/arch/x86/kernel/cpu/mce/therm_throt.c
> +++ b/arch/x86/kernel/cpu/mce/therm_throt.c
> @@ -188,7 +188,7 @@ static void therm_throt_process(bool new_event, int event, int level)
>   	/* if we just entered the thermal event */
>   	if (new_event) {
>   		if (event == THERMAL_THROTTLING_EVENT)
> -			pr_crit("CPU%d: %s temperature above threshold, cpu clock throttled (total events = %lu)\n",
> +			pr_warn("CPU%d: %s temperature above threshold, cpu clock throttled (total events = %lu)\n",
>   				this_cpu,
>   				level == CORE_LEVEL ? "Core" : "Package",
>   				state->count);
> 

Re: [PATCH] x86/mce: Lower throttling MCE messages to warnings

[Borislav Petkov]

On Wed, Oct 09, 2019 at 05:54:24PM +0200, Benjamin Berg wrote:
> On modern CPUs it is quite normal that the temperature limits are
> reached and the CPU is throttled. In fact, often the thermal design is
> not sufficient to cool the CPU at full load and limits can quickly be
> reached when a burst in load happens. This will even happen with
> technologies like RAPL limitting the long term power consumption of
> the package.
> 
> So these messages do not usually indicate a hardware issue (e.g.
> insufficient cooling). Log them as warnings to avoid confusion about
> their severity.
> 
> Signed-off-by: Benjamin Berg <bberg@redhat.com>
> Tested-by: Christian Kellner <ckellner@redhat.com>
> ---
>  arch/x86/kernel/cpu/mce/therm_throt.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/arch/x86/kernel/cpu/mce/therm_throt.c b/arch/x86/kernel/cpu/mce/therm_throt.c
> index 6e2becf547c5..bc441d68d060 100644
> --- a/arch/x86/kernel/cpu/mce/therm_throt.c
> +++ b/arch/x86/kernel/cpu/mce/therm_throt.c
> @@ -188,7 +188,7 @@ static void therm_throt_process(bool new_event, int event, int level)
>  	/* if we just entered the thermal event */
>  	if (new_event) {
>  		if (event == THERMAL_THROTTLING_EVENT)
> -			pr_crit("CPU%d: %s temperature above threshold, cpu clock throttled (total events = %lu)\n",
> +			pr_warn("CPU%d: %s temperature above threshold, cpu clock throttled (total events = %lu)\n",
>  				this_cpu,
>  				level == CORE_LEVEL ? "Core" : "Package",
>  				state->count);
> -- 

This has carried over since its very first addition in

commit 3867eb75b9279c7b0f6840d2ad9f27694ba6c4e4
Author: Dave Jones <davej@suse.de>
Date:   Tue Apr 2 20:02:27 2002 -0800

    [PATCH] x86 bluesmoke update.
    
    o  Make MCE compile time optional       (Paul Gortmaker)
    o  P4 thermal trip monitoring.          (Zwane Mwaikambo)
    o  Non-fatal MCE logging.               (Me)


It used to be KERN_EMERG back then, though.

And yes, this issue has come up in the past already so I think I'll take
it. I'll just give Intel folks a couple of days to object should there
be anything to object to.

Thx.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette

Re: [PATCH] x86/mce: Lower throttling MCE messages to warnings

[Joe Perches]

On Wed, 2019-10-09 at 19:56 +0200, Borislav Petkov wrote:
> On Wed, Oct 09, 2019 at 05:54:24PM +0200, Benjamin Berg wrote:
> > On modern CPUs it is quite normal that the temperature limits are
> > reached and the CPU is throttled. In fact, often the thermal design is
> > not sufficient to cool the CPU at full load and limits can quickly be
> > reached when a burst in load happens. This will even happen with
> > technologies like RAPL limitting the long term power consumption of
> > the package.
> > 
> > So these messages do not usually indicate a hardware issue (e.g.
> > insufficient cooling). Log them as warnings to avoid confusion about
> > their severity.
[]
> > diff --git a/arch/x86/kernel/cpu/mce/therm_throt.c b/arch/x86/kernel/cpu/mce/therm_throt.c
[]
> > @@ -188,7 +188,7 @@ static void therm_throt_process(bool new_event, int event, int level)
> >  	/* if we just entered the thermal event */
> >  	if (new_event) {
> >  		if (event == THERMAL_THROTTLING_EVENT)
> > -			pr_crit("CPU%d: %s temperature above threshold, cpu clock throttled (total events = %lu)\n",
> > +			pr_warn("CPU%d: %s temperature above threshold, cpu clock throttled (total events = %lu)\n",
> >  				this_cpu,
> >  				level == CORE_LEVEL ? "Core" : "Package",
> >  				state->count);
> > -- 
> 
> This has carried over since its very first addition in
> 
> commit 3867eb75b9279c7b0f6840d2ad9f27694ba6c4e4
> Author: Dave Jones <davej@suse.de>
> Date:   Tue Apr 2 20:02:27 2002 -0800
> 
>     [PATCH] x86 bluesmoke update.
>     
>     o  Make MCE compile time optional       (Paul Gortmaker)
>     o  P4 thermal trip monitoring.          (Zwane Mwaikambo)
>     o  Non-fatal MCE logging.               (Me)
> 
> 
> It used to be KERN_EMERG back then, though.
> 
> And yes, this issue has come up in the past already so I think I'll take
> it. I'll just give Intel folks a couple of days to object should there
> be anything to object to.

Perhaps this should be

	pr_warn_ratelimited(...)

as the temperature changes can be relatively quick.

Re: [PATCH] x86/mce: Lower throttling MCE messages to warnings

[Borislav Petkov]

On Wed, Oct 09, 2019 at 11:05:37AM -0700, Joe Perches wrote:
> Perhaps this should be
> 
> 	pr_warn_ratelimited(...)
> 
> as the temperature changes can be relatively quick.

There's already ratelimiting machinery a bit above in the same function.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette

Re: [PATCH] x86/mce: Lower throttling MCE messages to warnings

[Joe Perches]

On Wed, 2019-10-09 at 20:22 +0200, Borislav Petkov wrote:
> On Wed, Oct 09, 2019 at 11:05:37AM -0700, Joe Perches wrote:
> > Perhaps this should be
> > 
> > 	pr_warn_ratelimited(...)
> > 
> > as the temperature changes can be relatively quick.
> 
> There's already ratelimiting machinery a bit above in the same function.

right, thanks, nevermind...

Re: [PATCH] x86/mce: Lower throttling MCE messages to warnings

[Srinivas Pandruvada]

Hi Benjamin,

On Wed, 2019-10-09 at 19:56 +0200, Borislav Petkov wrote:
> On Wed, Oct 09, 2019 at 05:54:24PM +0200, Benjamin Berg wrote:
> > On modern CPUs it is quite normal that the temperature limits are
> > reached and the CPU is throttled. In fact, often the thermal design
> > is
> > not sufficient to cool the CPU at full load and limits can quickly
> > be
> > reached when a burst in load happens. This will even happen with
> > technologies like RAPL limitting the long term power consumption of
> > the package.
> > 
> > So these messages do not usually indicate a hardware issue (e.g.
> > insufficient cooling). Log them as warnings to avoid confusion
> > about
> > their severity.
> > 
I have a patch to address this. Instead of avoiding any critical
warnings or wait for 300 seconds for next one, the warning is based on
how long the system is working on throttled condition. If for example
the fan broke, then the throttling is extended for a long time. Then we
better warn.
I am waiting for internal review, and hope to post by tomorrow.

Thanks
Srinivas

> > Signed-off-by: Benjamin Berg <bberg@redhat.com>
> > Tested-by: Christian Kellner <ckellner@redhat.com>
> > ---
> >  arch/x86/kernel/cpu/mce/therm_throt.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/arch/x86/kernel/cpu/mce/therm_throt.c
> > b/arch/x86/kernel/cpu/mce/therm_throt.c
> > index 6e2becf547c5..bc441d68d060 100644
> > --- a/arch/x86/kernel/cpu/mce/therm_throt.c
> > +++ b/arch/x86/kernel/cpu/mce/therm_throt.c
> > @@ -188,7 +188,7 @@ static void therm_throt_process(bool new_event,
> > int event, int level)
> >  	/* if we just entered the thermal event */
> >  	if (new_event) {
> >  		if (event == THERMAL_THROTTLING_EVENT)
> > -			pr_crit("CPU%d: %s temperature above threshold,
> > cpu clock throttled (total events = %lu)\n",
> > +			pr_warn("CPU%d: %s temperature above threshold,
> > cpu clock throttled (total events = %lu)\n",
> >  				this_cpu,
> >  				level == CORE_LEVEL ? "Core" :
> > "Package",
> >  				state->count);
> > -- 
> 
> This has carried over since its very first addition in
> 
> commit 3867eb75b9279c7b0f6840d2ad9f27694ba6c4e4
> Author: Dave Jones <davej@suse.de>
> Date:   Tue Apr 2 20:02:27 2002 -0800
> 
>     [PATCH] x86 bluesmoke update.
>     
>     o  Make MCE compile time optional       (Paul Gortmaker)
>     o  P4 thermal trip monitoring.          (Zwane Mwaikambo)
>     o  Non-fatal MCE logging.               (Me)
> 
> 
> It used to be KERN_EMERG back then, though.
> 
> And yes, this issue has come up in the past already so I think I'll
> take
> it. I'll just give Intel folks a couple of days to object should
> there
> be anything to object to.
> 
> Thx.
> 

Re: [PATCH] x86/mce: Lower throttling MCE messages to warnings

[Benjamin Berg]

Hi Srinivas,

On Thu, 2019-10-10 at 14:08 -0700, Srinivas Pandruvada wrote:
> I have a patch to address this. Instead of avoiding any critical
> warnings or wait for 300 seconds for next one, the warning is based on
> how long the system is working on throttled condition. If for example
> the fan broke, then the throttling is extended for a long time. Then we
> better warn.
> I am waiting for internal review, and hope to post by tomorrow.

Nice! I agree that a heuristic seems better than the very simple
approach taken in this patch.

Thanks,
Benjamin

> Thanks
> Srinivas
> 
> > > Signed-off-by: Benjamin Berg <bberg@redhat.com>
> > > Tested-by: Christian Kellner <ckellner@redhat.com>
> > > ---
> > >  arch/x86/kernel/cpu/mce/therm_throt.c | 2 +-
> > >  1 file changed, 1 insertion(+), 1 deletion(-)
> > > 
> > > diff --git a/arch/x86/kernel/cpu/mce/therm_throt.c
> > > b/arch/x86/kernel/cpu/mce/therm_throt.c
> > > index 6e2becf547c5..bc441d68d060 100644
> > > --- a/arch/x86/kernel/cpu/mce/therm_throt.c
> > > +++ b/arch/x86/kernel/cpu/mce/therm_throt.c
> > > @@ -188,7 +188,7 @@ static void therm_throt_process(bool
> > > new_event,
> > > int event, int level)
> > >  	/* if we just entered the thermal event */
> > >  	if (new_event) {
> > >  		if (event == THERMAL_THROTTLING_EVENT)
> > > -			pr_crit("CPU%d: %s temperature above threshold,
> > > cpu clock throttled (total events = %lu)\n",
> > > +			pr_warn("CPU%d: %s temperature above threshold,
> > > cpu clock throttled (total events = %lu)\n",
> > >  				this_cpu,
> > >  				level == CORE_LEVEL ? "Core" :
> > > "Package",
> > >  				state->count);
> > > -- 
> > 
> > This has carried over since its very first addition in
> > 
> > commit 3867eb75b9279c7b0f6840d2ad9f27694ba6c4e4
> > Author: Dave Jones <davej@suse.de>
> > Date:   Tue Apr 2 20:02:27 2002 -0800
> > 
> >     [PATCH] x86 bluesmoke update.
> >     
> >     o  Make MCE compile time optional       (Paul Gortmaker)
> >     o  P4 thermal trip monitoring.          (Zwane Mwaikambo)
> >     o  Non-fatal MCE logging.               (Me)
> > 
> > 
> > It used to be KERN_EMERG back then, though.
> > 
> > And yes, this issue has come up in the past already so I think I'll
> > take
> > it. I'll just give Intel folks a couple of days to object should
> > there
> > be anything to object to.
> > 
> > Thx.
> > 

[tip: ras/core] x86/mce: Lower throttling MCE messages' priority to warning

[tip-bot2 for Benjamin Berg]

The following commit has been merged into the ras/core branch of tip:

Commit-ID:     9c3bafaa1fd88e4dd2dba3735a1f1abb0f2c7bb7
Gitweb:        https://git.kernel.org/tip/9c3bafaa1fd88e4dd2dba3735a1f1abb0f2c7bb7
Author:        Benjamin Berg <bberg@redhat.com>
AuthorDate:    Wed, 09 Oct 2019 17:54:24 +02:00
Committer:     Borislav Petkov <bp@suse.de>
CommitterDate: Thu, 17 Oct 2019 09:07:09 +02:00

x86/mce: Lower throttling MCE messages' priority to warning

On modern CPUs it is quite normal that the temperature limits are
reached and the CPU is throttled. In fact, often the thermal design is
not sufficient to cool the CPU at full load and limits can quickly be
reached when a burst in load happens. This will even happen with
technologies like RAPL limitting the long term power consumption of
the package.

Also, these limits are "softer", as Srinivas explains:

"CPU temperature doesn't have to hit max(TjMax) to get these warnings.
OEMs ha[ve] an ability to program a threshold where a thermal interrupt
can be generated. In some systems the offset is 20C+ (Read only value).

In recent systems, there is another offset on top of it which can be
programmed by OS, once some agent can adjust power limits dynamically.
By default this is set to low by the firmware, which I guess the
prime motivation of Benjamin to submit the patch."

So these messages do not usually indicate a hardware issue (e.g.
insufficient cooling). Log them as warnings to avoid confusion about
their severity.

 [ bp: Massage commit mesage. ]

Signed-off-by: Benjamin Berg <bberg@redhat.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Tested-by: Christian Kellner <ckellner@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: linux-edac <linux-edac@vger.kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Tony Luck <tony.luck@intel.com>
Cc: x86-ml <x86@kernel.org>
Link: https://lkml.kernel.org/r/20191009155424.249277-1-bberg@redhat.com
---
 arch/x86/kernel/cpu/mce/therm_throt.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86/kernel/cpu/mce/therm_throt.c b/arch/x86/kernel/cpu/mce/therm_throt.c
index 6e2becf..bc441d6 100644
--- a/arch/x86/kernel/cpu/mce/therm_throt.c
+++ b/arch/x86/kernel/cpu/mce/therm_throt.c
@@ -188,7 +188,7 @@ static void therm_throt_process(bool new_event, int event, int level)
 	/* if we just entered the thermal event */
 	if (new_event) {
 		if (event == THERMAL_THROTTLING_EVENT)
-			pr_crit("CPU%d: %s temperature above threshold, cpu clock throttled (total events = %lu)\n",
+			pr_warn("CPU%d: %s temperature above threshold, cpu clock throttled (total events = %lu)\n",
 				this_cpu,
 				level == CORE_LEVEL ? "Core" : "Package",
 				state->count);

Re: [PATCH] x86/mce: Lower throttling MCE messages to warnings

[Hans de Goede]

Hi Benjamin,

On 09-10-2019 17:54, Benjamin Berg wrote:
> On modern CPUs it is quite normal that the temperature limits are
> reached and the CPU is throttled. In fact, often the thermal design is
> not sufficient to cool the CPU at full load and limits can quickly be
> reached when a burst in load happens. This will even happen with
> technologies like RAPL limitting the long term power consumption of
> the package.
> 
> So these messages do not usually indicate a hardware issue (e.g.
> insufficient cooling). Log them as warnings to avoid confusion about
> their severity.
> 
> Signed-off-by: Benjamin Berg <bberg@redhat.com>
> Tested-by: Christian Kellner <ckellner@redhat.com>

This seems like the exact same patch as you send before, is there
any reason for this resend ?

Regards,

Hans


> ---
>   arch/x86/kernel/cpu/mce/therm_throt.c | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/arch/x86/kernel/cpu/mce/therm_throt.c b/arch/x86/kernel/cpu/mce/therm_throt.c
> index 6e2becf547c5..bc441d68d060 100644
> --- a/arch/x86/kernel/cpu/mce/therm_throt.c
> +++ b/arch/x86/kernel/cpu/mce/therm_throt.c
> @@ -188,7 +188,7 @@ static void therm_throt_process(bool new_event, int event, int level)
>   	/* if we just entered the thermal event */
>   	if (new_event) {
>   		if (event == THERMAL_THROTTLING_EVENT)
> -			pr_crit("CPU%d: %s temperature above threshold, cpu clock throttled (total events = %lu)\n",
> +			pr_warn("CPU%d: %s temperature above threshold, cpu clock throttled (total events = %lu)\n",
>   				this_cpu,
>   				level == CORE_LEVEL ? "Core" : "Package",
>   				state->count);
>