Merge git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6

5 files changed, 108 insertions, 57 deletions

diff --git a/arch/x86/kernel/cpu/mcheck/mce-inject.c b/arch/x86/kernel/cpu/mcheck/mce-inject.c
index 7029f0e2aca..472763d9209 100644
--- a/arch/x86/kernel/cpu/mcheck/mce-inject.c
+++ b/arch/x86/kernel/cpu/mcheck/mce-inject.c
@@ -98,8 +98,9 @@ static struct notifier_block mce_raise_nb = {
 };
 
 /* Inject mce on current CPU */
-static int raise_local(struct mce *m)
+static int raise_local(void)
 {
+	struct mce *m = &__get_cpu_var(injectm);
 	int context = MCJ_CTX(m->inject_flags);
 	int ret = 0;
 	int cpu = m->extcpu;
@@ -167,12 +168,12 @@ static void raise_mce(struct mce *m)
 			}
 			cpu_relax();
 		}
-		raise_local(m);
+		raise_local();
 		put_cpu();
 		put_online_cpus();
 	} else
 #endif
-		raise_local(m);
+		raise_local();
 }
 
 /* Error injection interface */
diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c
index 2f5aab26320..721a77ca811 100644
--- a/arch/x86/kernel/cpu/mcheck/mce.c
+++ b/arch/x86/kernel/cpu/mcheck/mce.c
@@ -85,6 +85,18 @@ static DECLARE_WAIT_QUEUE_HEAD(mce_wait);
 static DEFINE_PER_CPU(struct mce, mces_seen);
 static int			cpu_missing;
 
+static void default_decode_mce(struct mce *m)
+{
+	pr_emerg("No human readable MCE decoding support on this CPU type.\n");
+	pr_emerg("Run the message through 'mcelog --ascii' to decode.\n");
+}
+
+/*
+ * CPU/chipset specific EDAC code can register a callback here to print
+ * MCE errors in a human-readable form:
+ */
+void (*x86_mce_decode_callback)(struct mce *m) = default_decode_mce;
+EXPORT_SYMBOL(x86_mce_decode_callback);
 
 /* MCA banks polled by the period polling timer for corrected events */
 DEFINE_PER_CPU(mce_banks_t, mce_poll_banks) = {
@@ -165,49 +177,46 @@ void mce_log(struct mce *mce)
 	set_bit(0, &mce_need_notify);
 }
 
-void __weak decode_mce(struct mce *m)
-{
-	return;
-}
-
 static void print_mce(struct mce *m)
 {
-	printk(KERN_EMERG
-	       "CPU %d: Machine Check Exception: %16Lx Bank %d: %016Lx\n",
+	pr_emerg("CPU %d: Machine Check Exception: %16Lx Bank %d: %016Lx\n",
 	       m->extcpu, m->mcgstatus, m->bank, m->status);
+
 	if (m->ip) {
-		printk(KERN_EMERG "RIP%s %02x:<%016Lx> ",
-		       !(m->mcgstatus & MCG_STATUS_EIPV) ? " !INEXACT!" : "",
-		       m->cs, m->ip);
+		pr_emerg("RIP%s %02x:<%016Lx> ",
+			!(m->mcgstatus & MCG_STATUS_EIPV) ? " !INEXACT!" : "",
+				m->cs, m->ip);
+
 		if (m->cs == __KERNEL_CS)
 			print_symbol("{%s}", m->ip);
-		printk(KERN_CONT "\n");
+		pr_cont("\n");
 	}
-	printk(KERN_EMERG "TSC %llx ", m->tsc);
+
+	pr_emerg("TSC %llx ", m->tsc);
 	if (m->addr)
-		printk(KERN_CONT "ADDR %llx ", m->addr);
+		pr_cont("ADDR %llx ", m->addr);
 	if (m->misc)
-		printk(KERN_CONT "MISC %llx ", m->misc);
-	printk(KERN_CONT "\n");
-	printk(KERN_EMERG "PROCESSOR %u:%x TIME %llu SOCKET %u APIC %x\n",
-			m->cpuvendor, m->cpuid, m->time, m->socketid,
-			m->apicid);
+		pr_cont("MISC %llx ", m->misc);
 
-	decode_mce(m);
+	pr_cont("\n");
+	pr_emerg("PROCESSOR %u:%x TIME %llu SOCKET %u APIC %x\n",
+		m->cpuvendor, m->cpuid, m->time, m->socketid, m->apicid);
+
+	/*
+	 * Print out human-readable details about the MCE error,
+	 * (if the CPU has an implementation for that):
+	 */
+	x86_mce_decode_callback(m);
 }
 
 static void print_mce_head(void)
 {
-	printk(KERN_EMERG "\nHARDWARE ERROR\n");
+	pr_emerg("\nHARDWARE ERROR\n");
 }
 
 static void print_mce_tail(void)
 {
-	printk(KERN_EMERG "This is not a software problem!\n"
-#if (!defined(CONFIG_EDAC) || !defined(CONFIG_CPU_SUP_AMD))
-	       "Run through mcelog --ascii to decode and contact your hardware vendor\n"
-#endif
-	       );
+	pr_emerg("This is not a software problem!\n");
 }
 
 #define PANIC_TIMEOUT 5 /* 5 seconds */
@@ -221,6 +230,7 @@ static atomic_t mce_fake_paniced;
 static void wait_for_panic(void)
 {
 	long timeout = PANIC_TIMEOUT*USEC_PER_SEC;
+
 	preempt_disable();
 	local_irq_enable();
 	while (timeout-- > 0)
@@ -288,6 +298,7 @@ static void mce_panic(char *msg, struct mce *final, char *exp)
 static int msr_to_offset(u32 msr)
 {
 	unsigned bank = __get_cpu_var(injectm.bank);
+
 	if (msr == rip_msr)
 		return offsetof(struct mce, ip);
 	if (msr == MSR_IA32_MCx_STATUS(bank))
@@ -305,13 +316,25 @@ static int msr_to_offset(u32 msr)
 static u64 mce_rdmsrl(u32 msr)
 {
 	u64 v;
+
 	if (__get_cpu_var(injectm).finished) {
 		int offset = msr_to_offset(msr);
+
 		if (offset < 0)
 			return 0;
 		return *(u64 *)((char *)&__get_cpu_var(injectm) + offset);
 	}
-	rdmsrl(msr, v);
+
+	if (rdmsrl_safe(msr, &v)) {
+		WARN_ONCE(1, "mce: Unable to read msr %d!\n", msr);
+		/*
+		 * Return zero in case the access faulted. This should
+		 * not happen normally but can happen if the CPU does
+		 * something weird, or if the code is buggy.
+		 */
+		v = 0;
+	}
+
 	return v;
 }
 
@@ -319,6 +342,7 @@ static void mce_wrmsrl(u32 msr, u64 v)
 {
 	if (__get_cpu_var(injectm).finished) {
 		int offset = msr_to_offset(msr);
+
 		if (offset >= 0)
 			*(u64 *)((char *)&__get_cpu_var(injectm) + offset) = v;
 		return;
@@ -415,7 +439,7 @@ static inline void mce_get_rip(struct mce *m, struct pt_regs *regs)
 		m->ip = mce_rdmsrl(rip_msr);
 }
 
-#ifdef CONFIG_X86_LOCAL_APIC 
+#ifdef CONFIG_X86_LOCAL_APIC
 /*
  * Called after interrupts have been reenabled again
  * when a MCE happened during an interrupts off region
@@ -1172,6 +1196,7 @@ static int mce_banks_init(void)
 		return -ENOMEM;
 	for (i = 0; i < banks; i++) {
 		struct mce_bank *b = &mce_banks[i];
+
 		b->ctl = -1ULL;
 		b->init = 1;
 	}
@@ -1189,7 +1214,8 @@ static int __cpuinit mce_cap_init(void)
 	rdmsrl(MSR_IA32_MCG_CAP, cap);
 
 	b = cap & MCG_BANKCNT_MASK;
-	printk(KERN_INFO "mce: CPU supports %d MCE banks\n", b);
+	if (!banks)
+		printk(KERN_INFO "mce: CPU supports %d MCE banks\n", b);
 
 	if (b > MAX_NR_BANKS) {
 		printk(KERN_WARNING
@@ -1203,6 +1229,7 @@ static int __cpuinit mce_cap_init(void)
 	banks = b;
 	if (!mce_banks) {
 		int err = mce_banks_init();
+
 		if (err)
 			return err;
 	}
@@ -1237,6 +1264,7 @@ static void mce_init(void)
 
 	for (i = 0; i < banks; i++) {
 		struct mce_bank *b = &mce_banks[i];
+
 		if (!b->init)
 			continue;
 		wrmsrl(MSR_IA32_MCx_CTL(i), b->ctl);
@@ -1626,6 +1654,7 @@ static int mce_disable(void)
 
 	for (i = 0; i < banks; i++) {
 		struct mce_bank *b = &mce_banks[i];
+
 		if (b->init)
 			wrmsrl(MSR_IA32_MCx_CTL(i), 0);
 	}
@@ -1911,6 +1940,7 @@ static void mce_disable_cpu(void *h)
 		cmci_clear();
 	for (i = 0; i < banks; i++) {
 		struct mce_bank *b = &mce_banks[i];
+
 		if (b->init)
 			wrmsrl(MSR_IA32_MCx_CTL(i), 0);
 	}
@@ -1928,6 +1958,7 @@ static void mce_reenable_cpu(void *h)
 		cmci_reenable();
 	for (i = 0; i < banks; i++) {
 		struct mce_bank *b = &mce_banks[i];
+
 		if (b->init)
 			wrmsrl(MSR_IA32_MCx_CTL(i), b->ctl);
 	}
diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd.c b/arch/x86/kernel/cpu/mcheck/mce_amd.c
index 8cd5224943b..83a3d1f4efc 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_amd.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_amd.c
@@ -489,8 +489,9 @@ static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank)
 	int i, err = 0;
 	struct threshold_bank *b = NULL;
 	char name[32];
+#ifdef CONFIG_SMP
 	struct cpuinfo_x86 *c = &cpu_data(cpu);
-
+#endif
 
 	sprintf(name, "threshold_bank%i", bank);
 
diff --git a/arch/x86/kernel/cpu/mcheck/mce_intel.c b/arch/x86/kernel/cpu/mcheck/mce_intel.c
index 889f665fe93..7c785634af2 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_intel.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_intel.c
@@ -8,6 +8,7 @@
 #include <linux/init.h>
 #include <linux/interrupt.h>
 #include <linux/percpu.h>
+#include <linux/sched.h>
 #include <asm/apic.h>
 #include <asm/processor.h>
 #include <asm/msr.h>
diff --git a/arch/x86/kernel/cpu/mcheck/therm_throt.c b/arch/x86/kernel/cpu/mcheck/therm_throt.c
index 63a56d147e4..b3a1dba7533 100644
--- a/arch/x86/kernel/cpu/mcheck/therm_throt.c
+++ b/arch/x86/kernel/cpu/mcheck/therm_throt.c
@@ -34,20 +34,31 @@
 /* How long to wait between reporting thermal events */
 #define CHECK_INTERVAL		(300 * HZ)
 
-static DEFINE_PER_CPU(__u64, next_check) = INITIAL_JIFFIES;
-static DEFINE_PER_CPU(unsigned long, thermal_throttle_count);
-static DEFINE_PER_CPU(bool, thermal_throttle_active);
+/*
+ * Current thermal throttling state:
+ */
+struct thermal_state {
+	bool			is_throttled;
+
+	u64			next_check;
+	unsigned long		throttle_count;
+	unsigned long		last_throttle_count;
+};
+
+static DEFINE_PER_CPU(struct thermal_state, thermal_state);
 
-static atomic_t therm_throt_en		= ATOMIC_INIT(0);
+static atomic_t therm_throt_en	= ATOMIC_INIT(0);
 
 #ifdef CONFIG_SYSFS
 #define define_therm_throt_sysdev_one_ro(_name)				\
 	static SYSDEV_ATTR(_name, 0444, therm_throt_sysdev_show_##_name, NULL)
 
 #define define_therm_throt_sysdev_show_func(name)			\
-static ssize_t therm_throt_sysdev_show_##name(struct sys_device *dev,	\
-					struct sysdev_attribute *attr,	\
-					      char *buf)		\
+									\
+static ssize_t therm_throt_sysdev_show_##name(				\
+			struct sys_device *dev,				\
+			struct sysdev_attribute *attr,			\
+			char *buf)					\
 {									\
 	unsigned int cpu = dev->id;					\
 	ssize_t ret;							\
@@ -55,7 +66,7 @@ static ssize_t therm_throt_sysdev_show_##name(struct sys_device *dev,	\
 	preempt_disable();	/* CPU hotplug */			\
 	if (cpu_online(cpu))						\
 		ret = sprintf(buf, "%lu\n",				\
-			      per_cpu(thermal_throttle_##name, cpu));	\
+			      per_cpu(thermal_state, cpu).name);	\
 	else								\
 		ret = 0;						\
 	preempt_enable();						\
@@ -63,11 +74,11 @@ static ssize_t therm_throt_sysdev_show_##name(struct sys_device *dev,	\
 	return ret;							\
 }
 
-define_therm_throt_sysdev_show_func(count);
-define_therm_throt_sysdev_one_ro(count);
+define_therm_throt_sysdev_show_func(throttle_count);
+define_therm_throt_sysdev_one_ro(throttle_count);
 
 static struct attribute *thermal_throttle_attrs[] = {
-	&attr_count.attr,
+	&attr_throttle_count.attr,
 	NULL
 };
 
@@ -93,33 +104,39 @@ static struct attribute_group thermal_throttle_attr_group = {
  *          1 : Event should be logged further, and a message has been
  *              printed to the syslog.
  */
-static int therm_throt_process(int curr)
+static int therm_throt_process(bool is_throttled)
 {
-	unsigned int cpu = smp_processor_id();
-	__u64 tmp_jiffs = get_jiffies_64();
-	bool was_throttled = __get_cpu_var(thermal_throttle_active);
-	bool is_throttled = __get_cpu_var(thermal_throttle_active) = curr;
+	struct thermal_state *state;
+	unsigned int this_cpu;
+	bool was_throttled;
+	u64 now;
+
+	this_cpu = smp_processor_id();
+	now = get_jiffies_64();
+	state = &per_cpu(thermal_state, this_cpu);
+
+	was_throttled = state->is_throttled;
+	state->is_throttled = is_throttled;
 
 	if (is_throttled)
-		__get_cpu_var(thermal_throttle_count)++;
+		state->throttle_count++;
 
-	if (!(was_throttled ^ is_throttled) &&
-	    time_before64(tmp_jiffs, __get_cpu_var(next_check)))
+	if (time_before64(now, state->next_check) &&
+			state->throttle_count != state->last_throttle_count)
 		return 0;
 
-	__get_cpu_var(next_check) = tmp_jiffs + CHECK_INTERVAL;
+	state->next_check = now + CHECK_INTERVAL;
+	state->last_throttle_count = state->throttle_count;
 
 	/* if we just entered the thermal event */
 	if (is_throttled) {
-		printk(KERN_CRIT "CPU%d: Temperature above threshold, "
-		       "cpu clock throttled (total events = %lu)\n",
-		       cpu, __get_cpu_var(thermal_throttle_count));
+		printk(KERN_CRIT "CPU%d: Temperature above threshold, cpu clock throttled (total events = %lu)\n", this_cpu, state->throttle_count);
 
 		add_taint(TAINT_MACHINE_CHECK);
 		return 1;
 	}
 	if (was_throttled) {
-		printk(KERN_INFO "CPU%d: Temperature/speed normal\n", cpu);
+		printk(KERN_INFO "CPU%d: Temperature/speed normal\n", this_cpu);
 		return 1;
 	}
 
@@ -213,7 +230,7 @@ static void intel_thermal_interrupt(void)
 	__u64 msr_val;
 
 	rdmsrl(MSR_IA32_THERM_STATUS, msr_val);
-	if (therm_throt_process(msr_val & THERM_STATUS_PROCHOT))
+	if (therm_throt_process((msr_val & THERM_STATUS_PROCHOT) != 0))
 		mce_log_therm_throt_event(msr_val);
 }