aboutsummaryrefslogtreecommitdiff
path: root/arch/x86/kernel/cpu/mcheck
diff options
context:
space:
mode:
Diffstat (limited to 'arch/x86/kernel/cpu/mcheck')
-rw-r--r--arch/x86/kernel/cpu/mcheck/mce.c90
-rw-r--r--arch/x86/kernel/cpu/mcheck/therm_throt.c45
2 files changed, 94 insertions, 41 deletions
diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c
index 15ba9c972d7..a8aacd4b513 100644
--- a/arch/x86/kernel/cpu/mcheck/mce.c
+++ b/arch/x86/kernel/cpu/mcheck/mce.c
@@ -46,6 +46,9 @@
#include "mce-internal.h"
+#define CREATE_TRACE_POINTS
+#include <trace/events/mce.h>
+
int mce_disabled __read_mostly;
#define MISC_MCELOG_MINOR 227
@@ -149,6 +152,9 @@ void mce_log(struct mce *mce)
{
unsigned next, entry;
+ /* Emit the trace record: */
+ trace_mce_record(mce);
+
mce->finished = 0;
wmb();
for (;;) {
@@ -1130,7 +1136,7 @@ static int check_interval = 5 * 60; /* 5 minutes */
static DEFINE_PER_CPU(int, mce_next_interval); /* in jiffies */
static DEFINE_PER_CPU(struct timer_list, mce_timer);
-static void mcheck_timer(unsigned long data)
+static void mce_start_timer(unsigned long data)
{
struct timer_list *t = &per_cpu(mce_timer, data);
int *n;
@@ -1195,7 +1201,7 @@ int mce_notify_irq(void)
}
EXPORT_SYMBOL_GPL(mce_notify_irq);
-static int mce_banks_init(void)
+static int __cpuinit __mcheck_cpu_mce_banks_init(void)
{
int i;
@@ -1214,7 +1220,7 @@ static int mce_banks_init(void)
/*
* Initialize Machine Checks for a CPU.
*/
-static int __cpuinit mce_cap_init(void)
+static int __cpuinit __mcheck_cpu_cap_init(void)
{
unsigned b;
u64 cap;
@@ -1222,7 +1228,8 @@ 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
@@ -1235,7 +1242,7 @@ static int __cpuinit mce_cap_init(void)
WARN_ON(banks != 0 && b != banks);
banks = b;
if (!mce_banks) {
- int err = mce_banks_init();
+ int err = __mcheck_cpu_mce_banks_init();
if (err)
return err;
@@ -1251,7 +1258,7 @@ static int __cpuinit mce_cap_init(void)
return 0;
}
-static void mce_init(void)
+static void __mcheck_cpu_init_generic(void)
{
mce_banks_t all_banks;
u64 cap;
@@ -1280,7 +1287,7 @@ static void mce_init(void)
}
/* Add per CPU specific workarounds here */
-static int __cpuinit mce_cpu_quirks(struct cpuinfo_x86 *c)
+static int __cpuinit __mcheck_cpu_apply_quirks(struct cpuinfo_x86 *c)
{
if (c->x86_vendor == X86_VENDOR_UNKNOWN) {
pr_info("MCE: unknown CPU type - not enabling MCE support.\n");
@@ -1348,7 +1355,7 @@ static int __cpuinit mce_cpu_quirks(struct cpuinfo_x86 *c)
return 0;
}
-static void __cpuinit mce_ancient_init(struct cpuinfo_x86 *c)
+static void __cpuinit __mcheck_cpu_ancient_init(struct cpuinfo_x86 *c)
{
if (c->x86 != 5)
return;
@@ -1362,7 +1369,7 @@ static void __cpuinit mce_ancient_init(struct cpuinfo_x86 *c)
}
}
-static void mce_cpu_features(struct cpuinfo_x86 *c)
+static void __mcheck_cpu_init_vendor(struct cpuinfo_x86 *c)
{
switch (c->x86_vendor) {
case X86_VENDOR_INTEL:
@@ -1376,18 +1383,19 @@ static void mce_cpu_features(struct cpuinfo_x86 *c)
}
}
-static void mce_init_timer(void)
+static void __mcheck_cpu_init_timer(void)
{
struct timer_list *t = &__get_cpu_var(mce_timer);
int *n = &__get_cpu_var(mce_next_interval);
+ setup_timer(t, mce_start_timer, smp_processor_id());
+
if (mce_ignore_ce)
return;
*n = check_interval * HZ;
if (!*n)
return;
- setup_timer(t, mcheck_timer, smp_processor_id());
t->expires = round_jiffies(jiffies + *n);
add_timer_on(t, smp_processor_id());
}
@@ -1407,30 +1415,28 @@ void (*machine_check_vector)(struct pt_regs *, long error_code) =
* Called for each booted CPU to set up machine checks.
* Must be called with preempt off:
*/
-void __cpuinit mcheck_init(struct cpuinfo_x86 *c)
+void __cpuinit mcheck_cpu_init(struct cpuinfo_x86 *c)
{
if (mce_disabled)
return;
- mce_ancient_init(c);
+ __mcheck_cpu_ancient_init(c);
if (!mce_available(c))
return;
- if (mce_cap_init() < 0 || mce_cpu_quirks(c) < 0) {
+ if (__mcheck_cpu_cap_init() < 0 || __mcheck_cpu_apply_quirks(c) < 0) {
mce_disabled = 1;
return;
}
machine_check_vector = do_machine_check;
- mce_init();
- mce_cpu_features(c);
- mce_init_timer();
+ __mcheck_cpu_init_generic();
+ __mcheck_cpu_init_vendor(c);
+ __mcheck_cpu_init_timer();
INIT_WORK(&__get_cpu_var(mce_work), mce_process_work);
- if (raw_smp_processor_id() == 0)
- atomic_notifier_chain_register(&x86_mce_decoder_chain, &mce_dec_nb);
}
/*
@@ -1650,6 +1656,15 @@ static int __init mcheck_enable(char *str)
}
__setup("mce", mcheck_enable);
+int __init mcheck_init(void)
+{
+ atomic_notifier_chain_register(&x86_mce_decoder_chain, &mce_dec_nb);
+
+ mcheck_intel_therm_init();
+
+ return 0;
+}
+
/*
* Sysfs support
*/
@@ -1658,7 +1673,7 @@ __setup("mce", mcheck_enable);
* Disable machine checks on suspend and shutdown. We can't really handle
* them later.
*/
-static int mce_disable(void)
+static int mce_disable_error_reporting(void)
{
int i;
@@ -1673,12 +1688,12 @@ static int mce_disable(void)
static int mce_suspend(struct sys_device *dev, pm_message_t state)
{
- return mce_disable();
+ return mce_disable_error_reporting();
}
static int mce_shutdown(struct sys_device *dev)
{
- return mce_disable();
+ return mce_disable_error_reporting();
}
/*
@@ -1688,8 +1703,8 @@ static int mce_shutdown(struct sys_device *dev)
*/
static int mce_resume(struct sys_device *dev)
{
- mce_init();
- mce_cpu_features(&current_cpu_data);
+ __mcheck_cpu_init_generic();
+ __mcheck_cpu_init_vendor(&current_cpu_data);
return 0;
}
@@ -1699,8 +1714,8 @@ static void mce_cpu_restart(void *data)
del_timer_sync(&__get_cpu_var(mce_timer));
if (!mce_available(&current_cpu_data))
return;
- mce_init();
- mce_init_timer();
+ __mcheck_cpu_init_generic();
+ __mcheck_cpu_init_timer();
}
/* Reinit MCEs after user configuration changes */
@@ -1726,7 +1741,7 @@ static void mce_enable_ce(void *all)
cmci_reenable();
cmci_recheck();
if (all)
- mce_init_timer();
+ __mcheck_cpu_init_timer();
}
static struct sysdev_class mce_sysclass = {
@@ -1914,7 +1929,7 @@ error2:
sysdev_remove_file(&per_cpu(mce_dev, cpu), &mce_banks[j].attr);
error:
while (--i >= 0)
- sysdev_remove_file(&per_cpu(mce_dev, cpu), &mce_banks[i].attr);
+ sysdev_remove_file(&per_cpu(mce_dev, cpu), mce_attrs[i]);
sysdev_unregister(&per_cpu(mce_dev, cpu));
@@ -1939,13 +1954,14 @@ static __cpuinit void mce_remove_device(unsigned int cpu)
}
/* Make sure there are no machine checks on offlined CPUs. */
-static void mce_disable_cpu(void *h)
+static void __cpuinit mce_disable_cpu(void *h)
{
unsigned long action = *(unsigned long *)h;
int i;
if (!mce_available(&current_cpu_data))
return;
+
if (!(action & CPU_TASKS_FROZEN))
cmci_clear();
for (i = 0; i < banks; i++) {
@@ -1956,7 +1972,7 @@ static void mce_disable_cpu(void *h)
}
}
-static void mce_reenable_cpu(void *h)
+static void __cpuinit mce_reenable_cpu(void *h)
{
unsigned long action = *(unsigned long *)h;
int i;
@@ -2001,9 +2017,11 @@ mce_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu)
break;
case CPU_DOWN_FAILED:
case CPU_DOWN_FAILED_FROZEN:
- t->expires = round_jiffies(jiffies +
+ if (!mce_ignore_ce && check_interval) {
+ t->expires = round_jiffies(jiffies +
__get_cpu_var(mce_next_interval));
- add_timer_on(t, cpu);
+ add_timer_on(t, cpu);
+ }
smp_call_function_single(cpu, mce_reenable_cpu, &action, 1);
break;
case CPU_POST_DEAD:
@@ -2035,7 +2053,7 @@ static __init void mce_init_banks(void)
}
}
-static __init int mce_init_device(void)
+static __init int mcheck_init_device(void)
{
int err;
int i = 0;
@@ -2063,7 +2081,7 @@ static __init int mce_init_device(void)
return err;
}
-device_initcall(mce_init_device);
+device_initcall(mcheck_init_device);
/*
* Old style boot options parsing. Only for compatibility.
@@ -2111,7 +2129,7 @@ static int fake_panic_set(void *data, u64 val)
DEFINE_SIMPLE_ATTRIBUTE(fake_panic_fops, fake_panic_get,
fake_panic_set, "%llu\n");
-static int __init mce_debugfs_init(void)
+static int __init mcheck_debugfs_init(void)
{
struct dentry *dmce, *ffake_panic;
@@ -2125,5 +2143,5 @@ static int __init mce_debugfs_init(void)
return 0;
}
-late_initcall(mce_debugfs_init);
+late_initcall(mcheck_debugfs_init);
#endif
diff --git a/arch/x86/kernel/cpu/mcheck/therm_throt.c b/arch/x86/kernel/cpu/mcheck/therm_throt.c
index b3a1dba7533..81c499eceb2 100644
--- a/arch/x86/kernel/cpu/mcheck/therm_throt.c
+++ b/arch/x86/kernel/cpu/mcheck/therm_throt.c
@@ -49,6 +49,8 @@ static DEFINE_PER_CPU(struct thermal_state, thermal_state);
static atomic_t therm_throt_en = ATOMIC_INIT(0);
+static u32 lvtthmr_init __read_mostly;
+
#ifdef CONFIG_SYSFS
#define define_therm_throt_sysdev_one_ro(_name) \
static SYSDEV_ATTR(_name, 0444, therm_throt_sysdev_show_##_name, NULL)
@@ -254,14 +256,34 @@ asmlinkage void smp_thermal_interrupt(struct pt_regs *regs)
ack_APIC_irq();
}
+/* Thermal monitoring depends on APIC, ACPI and clock modulation */
+static int intel_thermal_supported(struct cpuinfo_x86 *c)
+{
+ if (!cpu_has_apic)
+ return 0;
+ if (!cpu_has(c, X86_FEATURE_ACPI) || !cpu_has(c, X86_FEATURE_ACC))
+ return 0;
+ return 1;
+}
+
+void __init mcheck_intel_therm_init(void)
+{
+ /*
+ * This function is only called on boot CPU. Save the init thermal
+ * LVT value on BSP and use that value to restore APs' thermal LVT
+ * entry BIOS programmed later
+ */
+ if (intel_thermal_supported(&boot_cpu_data))
+ lvtthmr_init = apic_read(APIC_LVTTHMR);
+}
+
void intel_init_thermal(struct cpuinfo_x86 *c)
{
unsigned int cpu = smp_processor_id();
int tm2 = 0;
u32 l, h;
- /* Thermal monitoring depends on ACPI and clock modulation*/
- if (!cpu_has(c, X86_FEATURE_ACPI) || !cpu_has(c, X86_FEATURE_ACC))
+ if (!intel_thermal_supported(c))
return;
/*
@@ -270,7 +292,20 @@ void intel_init_thermal(struct cpuinfo_x86 *c)
* since it might be delivered via SMI already:
*/
rdmsr(MSR_IA32_MISC_ENABLE, l, h);
- h = apic_read(APIC_LVTTHMR);
+
+ /*
+ * The initial value of thermal LVT entries on all APs always reads
+ * 0x10000 because APs are woken up by BSP issuing INIT-SIPI-SIPI
+ * sequence to them and LVT registers are reset to 0s except for
+ * the mask bits which are set to 1s when APs receive INIT IPI.
+ * Always restore the value that BIOS has programmed on AP based on
+ * BSP's info we saved since BIOS is always setting the same value
+ * for all threads/cores
+ */
+ apic_write(APIC_LVTTHMR, lvtthmr_init);
+
+ h = lvtthmr_init;
+
if ((l & MSR_IA32_MISC_ENABLE_TM1) && (h & APIC_DM_SMI)) {
printk(KERN_DEBUG
"CPU%d: Thermal monitoring handled by SMI\n", cpu);
@@ -312,8 +347,8 @@ void intel_init_thermal(struct cpuinfo_x86 *c)
l = apic_read(APIC_LVTTHMR);
apic_write(APIC_LVTTHMR, l & ~APIC_LVT_MASKED);
- printk(KERN_INFO "CPU%d: Thermal monitoring enabled (%s)\n",
- cpu, tm2 ? "TM2" : "TM1");
+ printk_once(KERN_INFO "CPU0: Thermal monitoring enabled (%s)\n",
+ tm2 ? "TM2" : "TM1");
/* enable thermal throttle processing */
atomic_set(&therm_throt_en, 1);