Skip to contentSkip to conversation

perf/x86/intel: ignore CondChgd bit to avoid false NMI handling

PATCH Superseded HATAYAMA Daisuke 4 participants 7 messages Jun 13, 2014 – Jul 2, 2014

Subsystems (2)

v2 of 3 Superseded Latest is v3 →

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 perf/x86/intel: ignore CondChgd bit to avoid false NMI handling

Currently, a NMI handler for NMI watchdog may falsely handle any NMI signaled for different purpose if CondChgd bit in MSR_CORE_PERF_GLOBAL_STATUS MSR is set.

This commit deals with the issue simply by ignoring CondChgd bit.

Here is explanation in detail.

On x86 NMI watchdog uses performance monitoring feature to periodically signal NMI each time performance counter gets overflowed.

intel_pmu_handle_irq() is called as a NMI_LOCAL handler from a NMI handler of NMI watchdog, perf_event_nmi_handler(). It identifies an owner of a given NMI by looking at overflow status bits in MSR_CORE_PERF_GLOBAL_STATUS MSR. If some of the bits are set, then it handles the given NMI as its own NMI.

The problem is that the intel_pmu_handle_irq() doesn't distinguish CondChgd bit from other bits. Unlike the other status bits, CondChgd bit doesn't represent overflow status for performance counters. Thus, CondChgd bit cannot be thought of as a mark indicating a given NMI is NMI watchdog's. As a result, if CondChgd bit is set, any NMI is falsely handled by the NMI handler of NMI watchdog. Also, if type of the falsely handled NMI is either NMI_UNKNOWN, NMI_SERR or NMI_IO_CHECK, the corresponding action is never performed until CondChgd bit is cleared.

I noticed this behavior on systems with Ivy Bridge processors: Intel Xeon CPU E5-2630 v2 and Intel Xeon CPU E7-8890 v2. On both systems, CondChgd bit in MSR_CORE_PERF_GLOBAL_STATUS MSR has already been set in the beginning at boot. Then the CondChgd bit is immediately cleared by next wrmsr to MSR_CORE_PERF_GLOBAL_CTRL MSR and appears to remain 0.

On the other hand, on older processors such as Nehalem, Xeon E7540, CondChgd bit is not set in the beginning at boot.

I'm not sure about exact behavior of CondChgd bit, in particular when this bit is set. Although I read Intel System Programmer's Manual to figure out that, the descriptions I found are:

In 18.9.1:

"The MSR_PERF_GLOBAL_STATUS MSR also provides a ‘sticky bit’ to
indicate changes to the state of performancmonitoring hardware"

In Table 35-2 IA-32 Architectural MSRs

63 CondChg: status bits of this register has changed.

These are different from the bahviour I see on the actual system as I explained above.

At least, I think ignoring CondChgd bit should be enough for NMI watchdog perspective.

---
 arch/x86/kernel/cpu/perf_event_intel.c |    9 +++++++++
 1 file changed, 9 insertions(+)
arch/x86/kernel/cpu/perf_event_intel.c +9 -0
index adb02aa..07846d7 100644
@@ -1382,6 +1382,15 @@ again:
 	intel_pmu_lbr_read();
 
 	/*
+	 * CondChgd bit 63 doesn't mean any overflow status. Ignore
+	 * and clear the bit.
+	 */
+	if (__test_and_clear_bit(63, (unsigned long *)&status)) {
+		if (!status)
+			goto done;
+	}
+
+	/*
 	 * PEBS overflow sets bit 62 in the global status register
 	 */
 	if (__test_and_clear_bit(62, (unsigned long *)&status)) {
Acked-by on v2 1/1
#1 lore raw
Quoted 48 lines · HATAYAMA Daisuke

Currently, a NMI handler for NMI watchdog may falsely handle any NMI signaled for different purpose if CondChgd bit in MSR_CORE_PERF_GLOBAL_STATUS MSR is set.

This commit deals with the issue simply by ignoring CondChgd bit.

Here is explanation in detail.

On x86 NMI watchdog uses performance monitoring feature to periodically signal NMI each time performance counter gets overflowed.

intel_pmu_handle_irq() is called as a NMI_LOCAL handler from a NMI handler of NMI watchdog, perf_event_nmi_handler(). It identifies an owner of a given NMI by looking at overflow status bits in MSR_CORE_PERF_GLOBAL_STATUS MSR. If some of the bits are set, then it handles the given NMI as its own NMI.

The problem is that the intel_pmu_handle_irq() doesn't distinguish CondChgd bit from other bits. Unlike the other status bits, CondChgd bit doesn't represent overflow status for performance counters. Thus, CondChgd bit cannot be thought of as a mark indicating a given NMI is NMI watchdog's. As a result, if CondChgd bit is set, any NMI is falsely handled by the NMI handler of NMI watchdog. Also, if type of the falsely handled NMI is either NMI_UNKNOWN, NMI_SERR or NMI_IO_CHECK, the corresponding action is never performed until CondChgd bit is cleared.

I noticed this behavior on systems with Ivy Bridge processors: Intel Xeon CPU E5-2630 v2 and Intel Xeon CPU E7-8890 v2. On both systems, CondChgd bit in MSR_CORE_PERF_GLOBAL_STATUS MSR has already been set in the beginning at boot. Then the CondChgd bit is immediately cleared by next wrmsr to MSR_CORE_PERF_GLOBAL_CTRL MSR and appears to remain 0.

On the other hand, on older processors such as Nehalem, Xeon E7540, CondChgd bit is not set in the beginning at boot.

I'm not sure about exact behavior of CondChgd bit, in particular when this bit is set. Although I read Intel System Programmer's Manual to figure out that, the descriptions I found are:

In 18.9.1:

"The MSR_PERF_GLOBAL_STATUS MSR also provides a ‘sticky bit’ to
indicate changes to the state of performancmonitoring hardware"

In Table 35-2 IA-32 Architectural MSRs

Quoted 7 lines

63 CondChg: status bits of this register has changed.

These are different from the bahviour I see on the actual system as I explained above.

At least, I think ignoring CondChgd bit should be enough for NMI watchdog perspective.

As I said in a previous email, I ran into a similar problem and was going to solve it by zeroing out all the registers on init (which probably would have upset Peter :-) ). This is a smaller solution and seems ok. The only downside is it is called in the nmi handler.

I am working with our customer to try and talk with Intel why this bit is set to begin with. Our customer says their BIOS doesn't use the PMU during boot so it wasn't clear why this is now set on IVBs (though I don't see them on Intel whitebox IVBs).

I am for this patch as it solves our problem too. But it makes me wonder if this is just yet another workaround for something broken deeper.

Quoted 32 lines
---
 arch/x86/kernel/cpu/perf_event_intel.c |    9 +++++++++
 1 file changed, 9 insertions(+)
arch/x86/kernel/cpu/perf_event_intel.c +9 -1
index adb02aa..07846d7 100644
@@ -1382,6 +1382,15 @@ again:
 	intel_pmu_lbr_read();
 
 	/*
+	 * CondChgd bit 63 doesn't mean any overflow status. Ignore
+	 * and clear the bit.
+	 */
+	if (__test_and_clear_bit(63, (unsigned long *)&status)) {
+		if (!status)
+			goto done;
+	}
+
+	/*
 	 * PEBS overflow sets bit 62 in the global status register
 	 */
 	if (__test_and_clear_bit(62, (unsigned long *)&status)) {
 
--

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

#2 lore raw

Hello,

Quoted 63 lines · Don Zickus
Quoted 48 lines · HATAYAMA Daisuke

Currently, a NMI handler for NMI watchdog may falsely handle any NMI signaled for different purpose if CondChgd bit in MSR_CORE_PERF_GLOBAL_STATUS MSR is set.

This commit deals with the issue simply by ignoring CondChgd bit.

Here is explanation in detail.

On x86 NMI watchdog uses performance monitoring feature to periodically signal NMI each time performance counter gets overflowed.

intel_pmu_handle_irq() is called as a NMI_LOCAL handler from a NMI handler of NMI watchdog, perf_event_nmi_handler(). It identifies an owner of a given NMI by looking at overflow status bits in MSR_CORE_PERF_GLOBAL_STATUS MSR. If some of the bits are set, then it handles the given NMI as its own NMI.

The problem is that the intel_pmu_handle_irq() doesn't distinguish CondChgd bit from other bits. Unlike the other status bits, CondChgd bit doesn't represent overflow status for performance counters. Thus, CondChgd bit cannot be thought of as a mark indicating a given NMI is NMI watchdog's. As a result, if CondChgd bit is set, any NMI is falsely handled by the NMI handler of NMI watchdog. Also, if type of the falsely handled NMI is either NMI_UNKNOWN, NMI_SERR or NMI_IO_CHECK, the corresponding action is never performed until CondChgd bit is cleared.

I noticed this behavior on systems with Ivy Bridge processors: Intel Xeon CPU E5-2630 v2 and Intel Xeon CPU E7-8890 v2. On both systems, CondChgd bit in MSR_CORE_PERF_GLOBAL_STATUS MSR has already been set in the beginning at boot. Then the CondChgd bit is immediately cleared by next wrmsr to MSR_CORE_PERF_GLOBAL_CTRL MSR and appears to remain 0.

On the other hand, on older processors such as Nehalem, Xeon E7540, CondChgd bit is not set in the beginning at boot.

I'm not sure about exact behavior of CondChgd bit, in particular when this bit is set. Although I read Intel System Programmer's Manual to figure out that, the descriptions I found are:

In 18.9.1:

"The MSR_PERF_GLOBAL_STATUS MSR also provides a ‘sticky bit’ to
indicate changes to the state of performancmonitoring hardware"

In Table 35-2 IA-32 Architectural MSRs

Quoted 7 lines

63 CondChg: status bits of this register has changed.

These are different from the bahviour I see on the actual system as I explained above.

At least, I think ignoring CondChgd bit should be enough for NMI watchdog perspective.

As I said in a previous email, I ran into a similar problem and was going to solve it by zeroing out all the registers on init (which probably would have upset Peter :-) ). This is a smaller solution and seems ok. The only downside is it is called in the nmi handler.

Quoted 5 lines

I am working with our customer to try and talk with Intel why this bit is set to begin with. Our customer says their BIOS doesn't use the PMU during boot so it wasn't clear why this is now set on IVBs (though I don't see them on Intel whitebox IVBs).

I'm also interested in the behaviour of CondChgd bit on Ivy Bridge processors.

Do you know something about this behaviour?

Signature · 2 lines
Thanks.
HATAYAMA, Daisuke
#3 lore raw
Quoted 2 lines

I'm also interested in the behaviour of CondChgd bit on Ivy Bridge processors.

The intended meaning of CondChgd is that a hardware debugger has taken over the PMU. It shouldn't really be set in other circumstances.

I think right now for perf it would be best to just ignore it.

In theory could stop using the PMU, but if some BIOS set it it would completely disable perf there. So better to just ignore it.

-Andi

Signature · 1 line
[email protected] -- Speaking for myself only
maintainer
#4 lore raw
Quoted 5 lines · Andi Kleen
Quoted 2 lines

I'm also interested in the behaviour of CondChgd bit on Ivy Bridge processors.

The intended meaning of CondChgd is that a hardware debugger has taken over the PMU. It shouldn't really be set in other circumstances.

The SDM utterly and totally fails to mention this, please fix that.

Quoted 4 lines

I think right now for perf it would be best to just ignore it.

In theory could stop using the PMU, but if some BIOS set it it would completely disable perf there. So better to just ignore it.

The even better option would be to kill off SMM mode, all that ever does is create problems.

#5 lore raw
In reply to Andi Kleen
Quoted 5 lines · Andi Kleen
Quoted 2 lines

I'm also interested in the behaviour of CondChgd bit on Ivy Bridge processors.

The intended meaning of CondChgd is that a hardware debugger has taken over the PMU. It shouldn't really be set in other circumstances.

Interesting. My concern is we have customers that can soft reboot a machine and have this bit accidentally be enabled. I am pretty sure they do not have a hardware debugger attached (I would have to double check). Is there other ways this can be set?

Cheers,
Don

Quoted 10 lines

I think right now for perf it would be best to just ignore it.

In theory could stop using the PMU, but if some BIOS set it it would completely disable perf there. So better to just ignore it.

-Andi

Signature · 1 line
[email protected] -- Speaking for myself only
#6 lore raw
In reply to Andi Kleen

From: Andi Kleen <[email protected]>
Subject: Re: [PATCH v2] perf/x86/intel: ignore CondChgd bit to avoid false NMI handling
Date: Mon, 30 Jun 2014 15:22:24 -0700

Quoted 6 lines
Quoted 2 lines

I'm also interested in the behaviour of CondChgd bit on Ivy Bridge processors.

The intended meaning of CondChgd is that a hardware debugger has taken over the PMU. It shouldn't really be set in other circumstances.

Thanks for your explanation.

The hardware debugger you mean is a kind of ITP?

Quoted 5 lines

I think right now for perf it would be best to just ignore it.

In theory could stop using the PMU, but if some BIOS set it it would completely disable perf there. So better to just ignore it.

I'll reflect this information in the patch description.

--
Thanks.
HATAYAMA, Daisuke