[PATCH] x86: Recover timer_ack lost in the merge of the NMI watchdog

[PATCH] x86: Recover timer_ack lost in the merge of the NMI watchdog

[Maciej W. Rozycki]

 In the course of the recent unification of the NMI watchdog an assignment
to timer_ack to switch off unnecesary POLL commands to the 8259A in the
case of a watchdog failure has been accidentally removed.  The statement
used to be limited to the 32-bit variation as since the rewrite of the
timer code it has been relevant for the 82489DX only.  This change brings
it back.

Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org>
---
patch-next-2.6.26-rc9-20080710-nmi-timer_ack-2
diff -up --recursive --new-file linux-next-2.6.26-rc9-20080710.macro/arch/x86/kernel/nmi.c linux-next-2.6.26-rc9-20080710/arch/x86/kernel/nmi.c
--- linux-next-2.6.26-rc9-20080710.macro/arch/x86/kernel/nmi.c	2008-07-11 02:57:31.000000000 +0000
+++ linux-next-2.6.26-rc9-20080710/arch/x86/kernel/nmi.c	2008-07-11 03:20:33.000000000 +0000
@@ -171,6 +171,9 @@ int __init check_nmi_watchdog(void)
 error:
 	if (nmi_watchdog == NMI_IO_APIC && !timer_through_8259)
 		disable_8259A_irq(0);
+#ifdef CONFIG_X86_32
+	timer_ack = 0;
+#endif
 	return -1;
 }
 

Re: [PATCH] x86: Recover timer_ack lost in the merge of the NMI watchdog

[Ingo Molnar]

* Maciej W. Rozycki <macro@linux-mips.org> wrote:

>  In the course of the recent unification of the NMI watchdog an 
> assignment to timer_ack to switch off unnecesary POLL commands to the 
> 8259A in the case of a watchdog failure has been accidentally removed.  
> The statement used to be limited to the 32-bit variation as since the 
> rewrite of the timer code it has been relevant for the 82489DX only.  
> This change brings it back.

i've applied your fixes to tip/x86/core:

 Maciej W. Rozycki (4):
      x86: L-APIC: Set IRQ0 as edge-triggered
      x86: L-APIC: Always fully configure IRQ0
      x86: I/O APIC: Never configure IRQ2
      x86: Recover timer_ack lost in the merge of the NMI watchdog

thanks Maciej,

	Ingo

Re: [PATCH] x86: Recover timer_ack lost in the merge of the NMI watchdog

[Maciej W. Rozycki]

On Fri, 11 Jul 2008, Ingo Molnar wrote:

> i've applied your fixes to tip/x86/core:
> 
>  Maciej W. Rozycki (4):
>       x86: L-APIC: Set IRQ0 as edge-triggered
>       x86: L-APIC: Always fully configure IRQ0
>       x86: I/O APIC: Never configure IRQ2
>       x86: Recover timer_ack lost in the merge of the NMI watchdog

 Thanks, please apply "x86: I/O APIC: Remove an IRQ2-mask hack" on top of
these.  Then what currently comprises dmi_ignore_irq0_timer_override() can
be done more properly as soon as we know how to get information about
INTIN0/2 configurations from the SB400.  Furthermore, following
considerations in the other mail, I think the HPET driver should use the
notion of ISA IRQ numbers by mapping them appropriately from I/O APIC pin
numbers.  Reusing pin_2_irq() for this purpose seems a reasonable way of
doing this to me.

  Maciej

Re: [PATCH] x86: Recover timer_ack lost in the merge of the NMI watchdog

[Ingo Molnar]

* Maciej W. Rozycki <macro@linux-mips.org> wrote:

> On Fri, 11 Jul 2008, Ingo Molnar wrote:
> 
> > i've applied your fixes to tip/x86/core:
> > 
> >  Maciej W. Rozycki (4):
> >       x86: L-APIC: Set IRQ0 as edge-triggered
> >       x86: L-APIC: Always fully configure IRQ0
> >       x86: I/O APIC: Never configure IRQ2
> >       x86: Recover timer_ack lost in the merge of the NMI watchdog
> 
> Thanks, please apply "x86: I/O APIC: Remove an IRQ2-mask hack" on top 
> of these.  Then what currently comprises 
> dmi_ignore_irq0_timer_override() can be done more properly as soon as 
> we know how to get information about INTIN0/2 configurations from the 
> SB400.  Furthermore, following considerations in the other mail, I 
> think the HPET driver should use the notion of ISA IRQ numbers by 
> mapping them appropriately from I/O APIC pin numbers.  Reusing 
> pin_2_irq() for this purpose seems a reasonable way of doing this to 
> me.

applied to tip/x86/core, thanks Maciej.

	Ingo