summaryrefslogtreecommitdiff
path: root/arch/x86/kernel/cpu/mce
diff options
context:
space:
mode:
Diffstat (limited to 'arch/x86/kernel/cpu/mce')
-rw-r--r--arch/x86/kernel/cpu/mce/core.c16
-rw-r--r--arch/x86/kernel/cpu/mce/dev-mcelog.c47
-rw-r--r--arch/x86/kernel/cpu/mce/intel.c26
-rw-r--r--arch/x86/kernel/cpu/mce/internal.h5
-rw-r--r--arch/x86/kernel/cpu/mce/therm_throt.c9
5 files changed, 75 insertions, 28 deletions
diff --git a/arch/x86/kernel/cpu/mce/core.c b/arch/x86/kernel/cpu/mce/core.c
index 2c4f949611e4..54165f3569e8 100644
--- a/arch/x86/kernel/cpu/mce/core.c
+++ b/arch/x86/kernel/cpu/mce/core.c
@@ -142,6 +142,8 @@ void mce_setup(struct mce *m)
if (this_cpu_has(X86_FEATURE_INTEL_PPIN))
rdmsrl(MSR_PPIN, m->ppin);
+ else if (this_cpu_has(X86_FEATURE_AMD_PPIN))
+ rdmsrl(MSR_AMD_PPIN, m->ppin);
m->microcode = boot_cpu_data.microcode;
}
@@ -1213,8 +1215,14 @@ static void __mc_scan_banks(struct mce *m, struct mce *final,
* On Intel systems this is entered on all CPUs in parallel through
* MCE broadcast. However some CPUs might be broken beyond repair,
* so be always careful when synchronizing with others.
+ *
+ * Tracing and kprobes are disabled: if we interrupted a kernel context
+ * with IF=1, we need to minimize stack usage. There are also recursion
+ * issues: if the machine check was due to a failure of the memory
+ * backing the user stack, tracing that reads the user stack will cause
+ * potentially infinite recursion.
*/
-void do_machine_check(struct pt_regs *regs, long error_code)
+void notrace do_machine_check(struct pt_regs *regs, long error_code)
{
DECLARE_BITMAP(valid_banks, MAX_NR_BANKS);
DECLARE_BITMAP(toclear, MAX_NR_BANKS);
@@ -1360,6 +1368,7 @@ out_ist:
ist_exit(regs);
}
EXPORT_SYMBOL_GPL(do_machine_check);
+NOKPROBE_SYMBOL(do_machine_check);
#ifndef CONFIG_MEMORY_FAILURE
int memory_failure(unsigned long pfn, int flags)
@@ -1877,6 +1886,8 @@ bool filter_mce(struct mce *m)
{
if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD)
return amd_filter_mce(m);
+ if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL)
+ return intel_filter_mce(m);
return false;
}
@@ -1892,10 +1903,11 @@ static void unexpected_machine_check(struct pt_regs *regs, long error_code)
void (*machine_check_vector)(struct pt_regs *, long error_code) =
unexpected_machine_check;
-dotraplinkage void do_mce(struct pt_regs *regs, long error_code)
+dotraplinkage notrace void do_mce(struct pt_regs *regs, long error_code)
{
machine_check_vector(regs, error_code);
}
+NOKPROBE_SYMBOL(do_mce);
/*
* Called for each booted CPU to set up machine checks.
diff --git a/arch/x86/kernel/cpu/mce/dev-mcelog.c b/arch/x86/kernel/cpu/mce/dev-mcelog.c
index 7c8958dee103..d089567a9ce8 100644
--- a/arch/x86/kernel/cpu/mce/dev-mcelog.c
+++ b/arch/x86/kernel/cpu/mce/dev-mcelog.c
@@ -29,11 +29,7 @@ static char *mce_helper_argv[2] = { mce_helper, NULL };
* separate MCEs from kernel messages to avoid bogus bug reports.
*/
-static struct mce_log_buffer mcelog = {
- .signature = MCE_LOG_SIGNATURE,
- .len = MCE_LOG_LEN,
- .recordlen = sizeof(struct mce),
-};
+static struct mce_log_buffer *mcelog;
static DECLARE_WAIT_QUEUE_HEAD(mce_chrdev_wait);
@@ -45,21 +41,21 @@ static int dev_mce_log(struct notifier_block *nb, unsigned long val,
mutex_lock(&mce_chrdev_read_mutex);
- entry = mcelog.next;
+ entry = mcelog->next;
/*
* When the buffer fills up discard new entries. Assume that the
* earlier errors are the more interesting ones:
*/
- if (entry >= MCE_LOG_LEN) {
- set_bit(MCE_OVERFLOW, (unsigned long *)&mcelog.flags);
+ if (entry >= mcelog->len) {
+ set_bit(MCE_OVERFLOW, (unsigned long *)&mcelog->flags);
goto unlock;
}
- mcelog.next = entry + 1;
+ mcelog->next = entry + 1;
- memcpy(mcelog.entry + entry, mce, sizeof(struct mce));
- mcelog.entry[entry].finished = 1;
+ memcpy(mcelog->entry + entry, mce, sizeof(struct mce));
+ mcelog->entry[entry].finished = 1;
/* wake processes polling /dev/mcelog */
wake_up_interruptible(&mce_chrdev_wait);
@@ -214,21 +210,21 @@ static ssize_t mce_chrdev_read(struct file *filp, char __user *ubuf,
/* Only supports full reads right now */
err = -EINVAL;
- if (*off != 0 || usize < MCE_LOG_LEN*sizeof(struct mce))
+ if (*off != 0 || usize < mcelog->len * sizeof(struct mce))
goto out;
- next = mcelog.next;
+ next = mcelog->next;
err = 0;
for (i = 0; i < next; i++) {
- struct mce *m = &mcelog.entry[i];
+ struct mce *m = &mcelog->entry[i];
err |= copy_to_user(buf, m, sizeof(*m));
buf += sizeof(*m);
}
- memset(mcelog.entry, 0, next * sizeof(struct mce));
- mcelog.next = 0;
+ memset(mcelog->entry, 0, next * sizeof(struct mce));
+ mcelog->next = 0;
if (err)
err = -EFAULT;
@@ -242,7 +238,7 @@ out:
static __poll_t mce_chrdev_poll(struct file *file, poll_table *wait)
{
poll_wait(file, &mce_chrdev_wait, wait);
- if (READ_ONCE(mcelog.next))
+ if (READ_ONCE(mcelog->next))
return EPOLLIN | EPOLLRDNORM;
if (!mce_apei_read_done && apei_check_mce())
return EPOLLIN | EPOLLRDNORM;
@@ -261,13 +257,13 @@ static long mce_chrdev_ioctl(struct file *f, unsigned int cmd,
case MCE_GET_RECORD_LEN:
return put_user(sizeof(struct mce), p);
case MCE_GET_LOG_LEN:
- return put_user(MCE_LOG_LEN, p);
+ return put_user(mcelog->len, p);
case MCE_GETCLEAR_FLAGS: {
unsigned flags;
do {
- flags = mcelog.flags;
- } while (cmpxchg(&mcelog.flags, flags, 0) != flags);
+ flags = mcelog->flags;
+ } while (cmpxchg(&mcelog->flags, flags, 0) != flags);
return put_user(flags, p);
}
@@ -339,8 +335,18 @@ static struct miscdevice mce_chrdev_device = {
static __init int dev_mcelog_init_device(void)
{
+ int mce_log_len;
int err;
+ mce_log_len = max(MCE_LOG_MIN_LEN, num_online_cpus());
+ mcelog = kzalloc(sizeof(*mcelog) + mce_log_len * sizeof(struct mce), GFP_KERNEL);
+ if (!mcelog)
+ return -ENOMEM;
+
+ strncpy(mcelog->signature, MCE_LOG_SIGNATURE, sizeof(mcelog->signature));
+ mcelog->len = mce_log_len;
+ mcelog->recordlen = sizeof(struct mce);
+
/* register character device /dev/mcelog */
err = misc_register(&mce_chrdev_device);
if (err) {
@@ -350,6 +356,7 @@ static __init int dev_mcelog_init_device(void)
else
pr_err("Unable to init device /dev/mcelog (rc: %d)\n", err);
+ kfree(mcelog);
return err;
}
diff --git a/arch/x86/kernel/cpu/mce/intel.c b/arch/x86/kernel/cpu/mce/intel.c
index 5627b1091b85..d8f9230d2034 100644
--- a/arch/x86/kernel/cpu/mce/intel.c
+++ b/arch/x86/kernel/cpu/mce/intel.c
@@ -493,17 +493,18 @@ static void intel_ppin_init(struct cpuinfo_x86 *c)
return;
if ((val & 3UL) == 1UL) {
- /* PPIN available but disabled: */
+ /* PPIN locked in disabled mode */
return;
}
- /* If PPIN is disabled, but not locked, try to enable: */
- if (!(val & 3UL)) {
+ /* If PPIN is disabled, try to enable */
+ if (!(val & 2UL)) {
wrmsrl_safe(MSR_PPIN_CTL, val | 2UL);
rdmsrl_safe(MSR_PPIN_CTL, &val);
}
- if ((val & 3UL) == 2UL)
+ /* Is the enable bit set? */
+ if (val & 2UL)
set_cpu_cap(c, X86_FEATURE_INTEL_PPIN);
}
}
@@ -520,3 +521,20 @@ void mce_intel_feature_clear(struct cpuinfo_x86 *c)
{
intel_clear_lmce();
}
+
+bool intel_filter_mce(struct mce *m)
+{
+ struct cpuinfo_x86 *c = &boot_cpu_data;
+
+ /* MCE errata HSD131, HSM142, HSW131, BDM48, and HSM142 */
+ if ((c->x86 == 6) &&
+ ((c->x86_model == INTEL_FAM6_HASWELL) ||
+ (c->x86_model == INTEL_FAM6_HASWELL_L) ||
+ (c->x86_model == INTEL_FAM6_BROADWELL) ||
+ (c->x86_model == INTEL_FAM6_HASWELL_G)) &&
+ (m->bank == 0) &&
+ ((m->status & 0xa0000000ffffffff) == 0x80000000000f0005))
+ return true;
+
+ return false;
+}
diff --git a/arch/x86/kernel/cpu/mce/internal.h b/arch/x86/kernel/cpu/mce/internal.h
index b785c0d0b590..3b008172ad73 100644
--- a/arch/x86/kernel/cpu/mce/internal.h
+++ b/arch/x86/kernel/cpu/mce/internal.h
@@ -8,6 +8,9 @@
#include <linux/device.h>
#include <asm/mce.h>
+/* Pointer to the installed machine check handler for this CPU setup. */
+extern void (*machine_check_vector)(struct pt_regs *, long error_code);
+
enum severity_level {
MCE_NO_SEVERITY,
MCE_DEFERRED_SEVERITY,
@@ -48,6 +51,7 @@ void cmci_disable_bank(int bank);
void intel_init_cmci(void);
void intel_init_lmce(void);
void intel_clear_lmce(void);
+bool intel_filter_mce(struct mce *m);
#else
# define cmci_intel_adjust_timer mce_adjust_timer_default
static inline bool mce_intel_cmci_poll(void) { return false; }
@@ -56,6 +60,7 @@ static inline void cmci_disable_bank(int bank) { }
static inline void intel_init_cmci(void) { }
static inline void intel_init_lmce(void) { }
static inline void intel_clear_lmce(void) { }
+static inline bool intel_filter_mce(struct mce *m) { return false; };
#endif
void mce_timer_kick(unsigned long interval);
diff --git a/arch/x86/kernel/cpu/mce/therm_throt.c b/arch/x86/kernel/cpu/mce/therm_throt.c
index 58b4ee3cda77..f36dc0742085 100644
--- a/arch/x86/kernel/cpu/mce/therm_throt.c
+++ b/arch/x86/kernel/cpu/mce/therm_throt.c
@@ -486,9 +486,14 @@ static int thermal_throttle_offline(unsigned int cpu)
{
struct thermal_state *state = &per_cpu(thermal_state, cpu);
struct device *dev = get_cpu_device(cpu);
+ u32 l;
+
+ /* Mask the thermal vector before draining evtl. pending work */
+ l = apic_read(APIC_LVTTHMR);
+ apic_write(APIC_LVTTHMR, l | APIC_LVT_MASKED);
- cancel_delayed_work(&state->package_throttle.therm_work);
- cancel_delayed_work(&state->core_throttle.therm_work);
+ cancel_delayed_work_sync(&state->package_throttle.therm_work);
+ cancel_delayed_work_sync(&state->core_throttle.therm_work);
state->package_throttle.rate_control_active = false;
state->core_throttle.rate_control_active = false;
generated by cgit v1.2.3 (git 2.39.0) at 2024-09-11 20:03:53 +0300