x86, CMCI: Add proper detection of end of CMCI storms

When CMCI storm persists for a long time(at least beyond predefined
threshold. It's 30 seconds for now), we can watch CMCI storm is
detected immediately after it subsides.

...
Dec 10 22:04:29 kernel: CMCI storm detected: switching to poll mode
Dec 10 22:04:59 kernel: CMCI storm subsided: switching to interrupt mode
Dec 10 22:04:59 kernel: CMCI storm detected: switching to poll mode
Dec 10 22:05:29 kernel: CMCI storm subsided: switching to interrupt mode
...

The problem is that our logic that determines that the storm has
ended is incorrect. We announce the end, re-enable interrupts and
realize that the storm is still going on, so we switch back to
polling mode. Rinse, repeat.

When a storm happens we disable signaling of errors via CMCI and begin
polling machine check banks instead. If we find any logged errors,
then we need to set a per-cpu flag so that our per-cpu tests that
check whether the storm is ongoing will see that errors are still
being logged independently of whether mce_notify_irq() says that the
error has been fully processed.

cmci_clear() is not the right tool to disable a bank. It disables the
interrupt for the bank as desired, but it also clears the bit for
this bank in "mce_banks_owned" so we will skip the bank when polling
(so we fail to see that the storm continues because we stop looking).
New cmci_storm_disable_banks() just disables the interrupt while
allowing polling to continue.

Reported-by: William Dauchy <wdauchy@gmail.com>
Signed-off-by: Chen, Gong <gong.chen@linux.intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>

2 files changed, 35 insertions, 2 deletions

diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c
index 4d5419b249da..78c92125db8a 100644
--- a/arch/x86/kernel/cpu/mcheck/mce.c
+++ b/arch/x86/kernel/cpu/mcheck/mce.c
@@ -89,6 +89,9 @@ static DECLARE_WAIT_QUEUE_HEAD(mce_chrdev_wait);
 static DEFINE_PER_CPU(struct mce, mces_seen);
 static int			cpu_missing;
 
+/* CMCI storm detection filter */
+static DEFINE_PER_CPU(unsigned long, mce_polled_error);
+
 /*
  * MCA banks polled by the period polling timer for corrected events.
  * With Intel CMCI, this only has MCA banks which do not support CMCI (if any).
@@ -595,6 +598,7 @@ void machine_check_poll(enum mcp_flags flags, mce_banks_t *b)
 {
 	struct mce m;
 	int i;
+	unsigned long *v;
 
 	this_cpu_inc(mce_poll_count);
 
@@ -614,6 +618,8 @@ void machine_check_poll(enum mcp_flags flags, mce_banks_t *b)
 		if (!(m.status & MCI_STATUS_VAL))
 			continue;
 
+		v = &get_cpu_var(mce_polled_error);
+		set_bit(0, v);
 		/*
 		 * Uncorrected or signalled events are handled by the exception
 		 * handler when it is enabled, so don't process those here.
@@ -1278,10 +1284,18 @@ static unsigned long mce_adjust_timer_default(unsigned long interval)
 static unsigned long (*mce_adjust_timer)(unsigned long interval) =
 	mce_adjust_timer_default;
 
+static int cmc_error_seen(void)
+{
+	unsigned long *v = &__get_cpu_var(mce_polled_error);
+
+	return test_and_clear_bit(0, v);
+}
+
 static void mce_timer_fn(unsigned long data)
 {
 	struct timer_list *t = &__get_cpu_var(mce_timer);
 	unsigned long iv;
+	int notify;
 
 	WARN_ON(smp_processor_id() != data);
 
@@ -1296,7 +1310,9 @@ static void mce_timer_fn(unsigned long data)
 	 * polling interval, otherwise increase the polling interval.
 	 */
 	iv = __this_cpu_read(mce_next_interval);
-	if (mce_notify_irq()) {
+	notify = mce_notify_irq();
+	notify |= cmc_error_seen();
+	if (notify) {
 		iv = max(iv / 2, (unsigned long) HZ/100);
 	} else {
 		iv = min(iv * 2, round_jiffies_relative(check_interval * HZ));
diff --git a/arch/x86/kernel/cpu/mcheck/mce_intel.c b/arch/x86/kernel/cpu/mcheck/mce_intel.c
index fb6156fee6f7..3bdb95ae8c43 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_intel.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_intel.c
@@ -9,6 +9,7 @@
 #include <linux/interrupt.h>
 #include <linux/percpu.h>
 #include <linux/sched.h>
+#include <linux/cpumask.h>
 #include <asm/apic.h>
 #include <asm/processor.h>
 #include <asm/msr.h>
@@ -137,6 +138,22 @@ unsigned long mce_intel_adjust_timer(unsigned long interval)
 	}
 }
 
+static void cmci_storm_disable_banks(void)
+{
+	unsigned long flags, *owned;
+	int bank;
+	u64 val;
+
+	raw_spin_lock_irqsave(&cmci_discover_lock, flags);
+	owned = __get_cpu_var(mce_banks_owned);
+	for_each_set_bit(bank, owned, MAX_NR_BANKS) {
+		rdmsrl(MSR_IA32_MCx_CTL2(bank), val);
+		val &= ~MCI_CTL2_CMCI_EN;
+		wrmsrl(MSR_IA32_MCx_CTL2(bank), val);
+	}
+	raw_spin_unlock_irqrestore(&cmci_discover_lock, flags);
+}
+
 static bool cmci_storm_detect(void)
 {
 	unsigned int cnt = __this_cpu_read(cmci_storm_cnt);
@@ -158,7 +175,7 @@ static bool cmci_storm_detect(void)
 	if (cnt <= CMCI_STORM_THRESHOLD)
 		return false;
 
-	cmci_clear();
+	cmci_storm_disable_banks();
 	__this_cpu_write(cmci_storm_state, CMCI_STORM_ACTIVE);
 	r = atomic_add_return(1, &cmci_storm_on_cpus);
 	mce_timer_kick(CMCI_POLL_INTERVAL);