Re:[PATCH] Mobil Pentium 4 HT and the NMI

Re:[PATCH] Mobil Pentium 4 HT and the NMI

[Steven Rostedt]

Hi,

I'm resending this since I don't see it in git yet, and I'm wondering if
there is a problem with this patch.  I have a IBM ThinkPad G41 with a
Mobile Pentium 4 HT.  Without this patch, the NMI won't be setup.  Is
there a reason that if the x86_model is greater than 0x3 it will return.
Since my processor has a 0x4 x86_model, I upped it to that. Otherwise my
laptop won't be able to use the NMI.

Thanks,

-- Steve

Description:
  This patch is to allow the Mobile Penitum 4 HT to use the NMI.

Signed-off-by: Steven Rostedt <rostedt@goodmis.org>

--- linux-2.6.13-rc6-git10/arch/i386/kernel/nmi.c.orig	2005-08-18 21:51:11.000000000 -0400
+++ linux-2.6.13-rc6-git10/arch/i386/kernel/nmi.c	2005-08-18 21:52:03.000000000 -0400
@@ -195,7 +195,7 @@ static void disable_lapic_nmi_watchdog(v
 			wrmsr(MSR_P6_EVNTSEL0, 0, 0);
 			break;
 		case 15:
-			if (boot_cpu_data.x86_model > 0x3)
+			if (boot_cpu_data.x86_model > 0x4)
 				break;
 
 			wrmsr(MSR_P4_IQ_CCCR0, 0, 0);
@@ -432,7 +432,7 @@ void setup_apic_nmi_watchdog (void)
 			setup_p6_watchdog();
 			break;
 		case 15:
-			if (boot_cpu_data.x86_model > 0x3)
+			if (boot_cpu_data.x86_model > 0x4)
 				return;
 
 			if (!setup_p4_watchdog())

Re: [PATCH] Mobil Pentium 4 HT and the NMI

[Andrew Morton]

Steven Rostedt <rostedt@goodmis.org> wrote:
>
> Hi,
> 
> I'm resending this since I don't see it in git yet, and I'm wondering if
> there is a problem with this patch.  I have a IBM ThinkPad G41 with a
> Mobile Pentium 4 HT.  Without this patch, the NMI won't be setup.  Is
> there a reason that if the x86_model is greater than 0x3 it will return.
> Since my processor has a 0x4 x86_model, I upped it to that. Otherwise my
> laptop won't be able to use the NMI.
> 

Well I was hoping that someone with knowledge of the low-level Intel model
differences would pipe up, but they all seem to be in hiding.  (Wildly
bcc's lots of x86 people).

> 
> Description:
>   This patch is to allow the Mobile Penitum 4 HT to use the NMI.
> 
> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
> 
> --- linux-2.6.13-rc6-git10/arch/i386/kernel/nmi.c.orig	2005-08-18 21:51:11.000000000 -0400
> +++ linux-2.6.13-rc6-git10/arch/i386/kernel/nmi.c	2005-08-18 21:52:03.000000000 -0400
> @@ -195,7 +195,7 @@ static void disable_lapic_nmi_watchdog(v
>  			wrmsr(MSR_P6_EVNTSEL0, 0, 0);
>  			break;
>  		case 15:
> -			if (boot_cpu_data.x86_model > 0x3)
> +			if (boot_cpu_data.x86_model > 0x4)
>  				break;
>  
>  			wrmsr(MSR_P4_IQ_CCCR0, 0, 0);
> @@ -432,7 +432,7 @@ void setup_apic_nmi_watchdog (void)
>  			setup_p6_watchdog();
>  			break;
>  		case 15:
> -			if (boot_cpu_data.x86_model > 0x3)
> +			if (boot_cpu_data.x86_model > 0x4)
>  				return;
>  
>  			if (!setup_p4_watchdog())

Re: [PATCH] Mobil Pentium 4 HT and the NMI

[Steven Rostedt]

On Thu, 2005-08-18 at 20:23 -0700, Andrew Morton wrote:
> Steven Rostedt <rostedt@goodmis.org> wrote:
> >
> > Hi,
> > 
> > I'm resending this since I don't see it in git yet, and I'm wondering if
> > there is a problem with this patch.  I have a IBM ThinkPad G41 with a
> > Mobile Pentium 4 HT.  Without this patch, the NMI won't be setup.  Is
> > there a reason that if the x86_model is greater than 0x3 it will return.
> > Since my processor has a 0x4 x86_model, I upped it to that. Otherwise my
> > laptop won't be able to use the NMI.
> > 
> 
> Well I was hoping that someone with knowledge of the low-level Intel model
> differences would pipe up, but they all seem to be in hiding.  (Wildly
> bcc's lots of x86 people).
> 

If this is any consolation, I've been using this patch all day today
debugging a deadlock in Ingo's RT patch.  It seems to work fine with me.

Who knows, maybe in two days my Laptop will be dead because this set up
some self destruct register. (knocks on wood).

-- Steve

Re: [PATCH] Mobil Pentium 4 HT and the NMI

[Zwane Mwaikambo]

On Thu, 18 Aug 2005, Andrew Morton wrote:

> Steven Rostedt <rostedt@goodmis.org> wrote:
> >
> > Hi,
> > 
> > I'm resending this since I don't see it in git yet, and I'm wondering if
> > there is a problem with this patch.  I have a IBM ThinkPad G41 with a
> > Mobile Pentium 4 HT.  Without this patch, the NMI won't be setup.  Is
> > there a reason that if the x86_model is greater than 0x3 it will return.
> > Since my processor has a 0x4 x86_model, I upped it to that. Otherwise my
> > laptop won't be able to use the NMI.
> > 
> 
> Well I was hoping that someone with knowledge of the low-level Intel model
> differences would pipe up, but they all seem to be in hiding.  (Wildly
> bcc's lots of x86 people).

Looks ok to me, they haven't changed the performance counter setup on 
those processors.

Acked-by: Zwane Mwaikambo <zwane@arm.linux.org.uk>

Re: [PATCH] Mobil Pentium 4 HT and the NMI

[Philippe Elie]

On Thu, 18 Aug 2005 at 20:23 +0000, Andrew Morton wrote:

> Steven Rostedt <rostedt@goodmis.org> wrote:
> >
> > Hi,
> > 
> > I'm resending this since I don't see it in git yet, and I'm wondering if
> > there is a problem with this patch.  I have a IBM ThinkPad G41 with a
> > Mobile Pentium 4 HT.  Without this patch, the NMI won't be setup.  Is
> > there a reason that if the x86_model is greater than 0x3 it will return.
> > Since my processor has a 0x4 x86_model, I upped it to that. Otherwise my
> > laptop won't be able to use the NMI.
> > 
> 
> Well I was hoping that someone with knowledge of the low-level Intel model
> differences would pipe up, but they all seem to be in hiding.  (Wildly
> bcc's lots of x86 people).

There is no documented change regarding to the used msr for P4 model 4,
Zwane do this for oprofile the 2005-02-02 and nobody complained.

-- 
Philippe Elie