[PATCH 2.6.32] x86: enable lapic nmi watchdog on AMD Family 11h

[PATCH 2.6.32] x86: enable lapic nmi watchdog on AMD Family 11h

[Mikael Pettersson]

The x86 lapic nmi watchdog does not recognize AMD Family 11h,
resulting in:

NMI watchdog: CPU not supported

As far as I can see from available documentation (the BKDM),
family 11h looks identical to family 10h as far as the PMU
is concerned.

Extending the check to accept family 11h results in:

Testing NMI watchdog ... OK.

I've been running with this change on a Turion X2 Ultra ZM-82
laptop for a couple of weeks now without problems.

Signed-off-by: Mikael Pettersson <mikpe@it.uu.se>
---

diff -rupN linux-2.6.32/arch/x86/kernel/cpu/perfctr-watchdog.c linux-2.6.32.x86-amd-fam11-nmi/arch/x86/kernel/cpu/perfctr-watchdog.c
--- linux-2.6.32/arch/x86/kernel/cpu/perfctr-watchdog.c	2009-12-03 12:38:32.000000000 +0100
+++ linux-2.6.32.x86-amd-fam11-nmi/arch/x86/kernel/cpu/perfctr-watchdog.c	2009-12-03 12:47:59.000000000 +0100
@@ -712,7 +712,7 @@ static void probe_nmi_watchdog(void)
 	switch (boot_cpu_data.x86_vendor) {
 	case X86_VENDOR_AMD:
 		if (boot_cpu_data.x86 != 6 && boot_cpu_data.x86 != 15 &&
-		    boot_cpu_data.x86 != 16)
+		    boot_cpu_data.x86 != 16 && boot_cpu_data.x86 != 17)
 			return;
 		wd_ops = &k7_wd_ops;
 		break;

Re: [PATCH 2.6.32] x86: enable lapic nmi watchdog on AMD Family 11h

[Ingo Molnar]

* Mikael Pettersson <mikpe@it.uu.se> wrote:

> The x86 lapic nmi watchdog does not recognize AMD Family 11h, 
> resulting in:
> 
> NMI watchdog: CPU not supported
> 
> As far as I can see from available documentation (the BKDM), family 
> 11h looks identical to family 10h as far as the PMU is concerned.
> 
> Extending the check to accept family 11h results in:
> 
> Testing NMI watchdog ... OK.
> 
> I've been running with this change on a Turion X2 Ultra ZM-82 laptop 
> for a couple of weeks now without problems.

Applied, thanks Mikael!

Long term i think we want to use the PMU driver to support the NMI tick, 
that way all new CPUs that have PMU support will get NMI watchdog 
support too, automatically.

That's a more complex change though and there's no reason not to apply 
this simple patch of yours. Btw., do you [or anyone else] have 
interest/time to pursue the more complex angle as well? It would be a 
nice feature.

	Ingo

[tip:x86/apic] x86, apic: Enable lapic nmi watchdog on AMD Family 11h

[tip-bot for Mikael Pettersson]

Commit-ID:  7d1849aff6687a135a8da3a75e32a00e3137a5e2
Gitweb:     http://git.kernel.org/tip/7d1849aff6687a135a8da3a75e32a00e3137a5e2
Author:     Mikael Pettersson <mikpe@it.uu.se>
AuthorDate: Thu, 3 Dec 2009 15:52:44 +0100
Committer:  Ingo Molnar <mingo@elte.hu>
CommitDate: Thu, 3 Dec 2009 16:25:15 +0100

x86, apic: Enable lapic nmi watchdog on AMD Family 11h

The x86 lapic nmi watchdog does not recognize AMD Family 11h,
resulting in:

  NMI watchdog: CPU not supported

As far as I can see from available documentation (the BKDM),
family 11h looks identical to family 10h as far as the PMU
is concerned.

Extending the check to accept family 11h results in:

  Testing NMI watchdog ... OK.

I've been running with this change on a Turion X2 Ultra ZM-82
laptop for a couple of weeks now without problems.

Signed-off-by: Mikael Pettersson <mikpe@it.uu.se>
Cc: Andreas Herrmann <andreas.herrmann3@amd.com>
Cc: Joerg Roedel <joerg.roedel@amd.com>
Cc: <stable@kernel.org>
LKML-Reference: <19223.53436.931768.278021@pilspetsen.it.uu.se>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/cpu/perfctr-watchdog.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/arch/x86/kernel/cpu/perfctr-watchdog.c b/arch/x86/kernel/cpu/perfctr-watchdog.c
index fab786f..898df97 100644
--- a/arch/x86/kernel/cpu/perfctr-watchdog.c
+++ b/arch/x86/kernel/cpu/perfctr-watchdog.c
@@ -712,7 +712,7 @@ static void probe_nmi_watchdog(void)
 	switch (boot_cpu_data.x86_vendor) {
 	case X86_VENDOR_AMD:
 		if (boot_cpu_data.x86 != 6 && boot_cpu_data.x86 != 15 &&
-		    boot_cpu_data.x86 != 16)
+		    boot_cpu_data.x86 != 16 && boot_cpu_data.x86 != 17)
 			return;
 		wd_ops = &k7_wd_ops;
 		break;