xref: /linux/arch/x86/kernel/cpu/mce/core.c (revision d9c43b6e43643e0c884179fd4598e2003f1a04a3)

// SPDX-License-Identifier: GPL-2.0-only
/*
 * Machine check handler.
 *
 * K8 parts Copyright 2002,2003 Andi Kleen, SuSE Labs.
 * Rest from unknown author(s).
 * 2004 Andi Kleen. Rewrote most of it.
 * Copyright 2008 Intel Corporation
 * Author: Andi Kleen
 */

#include <linux/thread_info.h>
#include <linux/capability.h>
#include <linux/miscdevice.h>
#include <linux/ratelimit.h>
#include <linux/rcupdate.h>
#include <linux/kobject.h>
#include <linux/uaccess.h>
#include <linux/kdebug.h>
#include <linux/kernel.h>
#include <linux/percpu.h>
#include <linux/string.h>
#include <linux/device.h>
#include <linux/syscore_ops.h>
#include <linux/delay.h>
#include <linux/ctype.h>
#include <linux/sched.h>
#include <linux/sysfs.h>
#include <linux/types.h>
#include <linux/slab.h>
#include <linux/init.h>
#include <linux/kmod.h>
#include <linux/poll.h>
#include <linux/nmi.h>
#include <linux/cpu.h>
#include <linux/ras.h>
#include <linux/smp.h>
#include <linux/fs.h>
#include <linux/mm.h>
#include <linux/debugfs.h>
#include <linux/irq_work.h>
#include <linux/export.h>
#include <linux/set_memory.h>
#include <linux/sync_core.h>
#include <linux/task_work.h>
#include <linux/hardirq.h>
#include <linux/kexec.h>

#include <asm/fred.h>
#include <asm/cpu_device_id.h>
#include <asm/processor.h>
#include <asm/traps.h>
#include <asm/tlbflush.h>
#include <asm/mce.h>
#include <asm/msr.h>
#include <asm/reboot.h>
#include <asm/tdx.h>

#include "internal.h"

/* sysfs synchronization */
static DEFINE_MUTEX(mce_sysfs_mutex);

#define CREATE_TRACE_POINTS
#include <trace/events/mce.h>

#define SPINUNIT		100	/* 100ns */

DEFINE_PER_CPU(unsigned, mce_exception_count);

DEFINE_PER_CPU_READ_MOSTLY(unsigned int, mce_num_banks);

DEFINE_PER_CPU_READ_MOSTLY(struct mce_bank[MAX_NR_BANKS], mce_banks_array);

#define ATTR_LEN               16
/* One object for each MCE bank, shared by all CPUs */
struct mce_bank_dev {
	struct device_attribute	attr;			/* device attribute */
	char			attrname[ATTR_LEN];	/* attribute name */
	u8			bank;			/* bank number */
};
static struct mce_bank_dev mce_bank_devs[MAX_NR_BANKS];

struct mce_vendor_flags mce_flags __read_mostly;

struct mca_config mca_cfg __read_mostly = {
	.bootlog  = -1,
	.monarch_timeout = -1
};

static DEFINE_PER_CPU(struct mce_hw_err, hw_errs_seen);
static unsigned long mce_need_notify;

/*
 * 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).
 */
DEFINE_PER_CPU(mce_banks_t, mce_poll_banks) = {
	[0 ... BITS_TO_LONGS(MAX_NR_BANKS)-1] = ~0UL
};

/*
 * MCA banks controlled through firmware first for corrected errors.
 * This is a global list of banks for which we won't enable CMCI and we
 * won't poll. Firmware controls these banks and is responsible for
 * reporting corrected errors through GHES. Uncorrected/recoverable
 * errors are still notified through a machine check.
 */
mce_banks_t mce_banks_ce_disabled;

static struct work_struct mce_work;
static struct irq_work mce_irq_work;

/*
 * CPU/chipset specific EDAC code can register a notifier call here to print
 * MCE errors in a human-readable form.
 */
BLOCKING_NOTIFIER_HEAD(x86_mce_decoder_chain);

void mce_prep_record_common(struct mce *m)
{
	m->cpuid	= cpuid_eax(1);
	m->cpuvendor	= boot_cpu_data.x86_vendor;
	m->mcgcap	= native_rdmsrq(MSR_IA32_MCG_CAP);
	/* need the internal __ version to avoid deadlocks */
	m->time		= __ktime_get_real_seconds();
}

void mce_prep_record_per_cpu(unsigned int cpu, struct mce *m)
{
	m->cpu		= cpu;
	m->extcpu	= cpu;
	m->apicid	= cpu_data(cpu).topo.initial_apicid;
	m->microcode	= cpu_data(cpu).microcode;
	m->ppin		= topology_ppin(cpu);
	m->socketid	= topology_physical_package_id(cpu);
}

/* Do initial initialization of struct mce_hw_err */
void mce_prep_record(struct mce_hw_err *err)
{
	struct mce *m = &err->m;

	memset(err, 0, sizeof(struct mce_hw_err));
	mce_prep_record_common(m);
	mce_prep_record_per_cpu(smp_processor_id(), m);
}

DEFINE_PER_CPU(struct mce, injectm);
EXPORT_PER_CPU_SYMBOL_GPL(injectm);

void mce_log(struct mce_hw_err *err)
{
	if (mce_gen_pool_add(err))
		irq_work_queue(&mce_irq_work);
}
EXPORT_SYMBOL_GPL(mce_log);

void mce_register_decode_chain(struct notifier_block *nb)
{
	if (WARN_ON(nb->priority < MCE_PRIO_LOWEST ||
		    nb->priority > MCE_PRIO_HIGHEST))
		return;

	blocking_notifier_chain_register(&x86_mce_decoder_chain, nb);
}
EXPORT_SYMBOL_GPL(mce_register_decode_chain);

void mce_unregister_decode_chain(struct notifier_block *nb)
{
	blocking_notifier_chain_unregister(&x86_mce_decoder_chain, nb);
}
EXPORT_SYMBOL_GPL(mce_unregister_decode_chain);

static void __print_mce(struct mce_hw_err *err)
{
	struct mce *m = &err->m;

	pr_emerg(HW_ERR "CPU %d: Machine Check%s: %Lx Bank %d: %016Lx\n",
		 m->extcpu,
		 (m->mcgstatus & MCG_STATUS_MCIP ? " Exception" : ""),
		 m->mcgstatus, m->bank, m->status);

	if (m->ip) {
		pr_emerg(HW_ERR "RIP%s %02x:<%016Lx> ",
			!(m->mcgstatus & MCG_STATUS_EIPV) ? " !INEXACT!" : "",
			m->cs, m->ip);

		if (m->cs == __KERNEL_CS)
			pr_cont("{%pS}", (void *)(unsigned long)m->ip);
		pr_cont("\n");
	}

	pr_emerg(HW_ERR "TSC %llx ", m->tsc);
	if (m->addr)
		pr_cont("ADDR %llx ", m->addr);
	if (m->misc)
		pr_cont("MISC %llx ", m->misc);
	if (m->ppin)
		pr_cont("PPIN %llx ", m->ppin);

	if (mce_flags.smca) {
		if (m->synd)
			pr_cont("SYND %llx ", m->synd);
		if (err->vendor.amd.synd1)
			pr_cont("SYND1 %llx ", err->vendor.amd.synd1);
		if (err->vendor.amd.synd2)
			pr_cont("SYND2 %llx ", err->vendor.amd.synd2);
		if (m->ipid)
			pr_cont("IPID %llx ", m->ipid);
	}

	pr_cont("\n");

	/*
	 * Note this output is parsed by external tools and old fields
	 * should not be changed.
	 */
	pr_emerg(HW_ERR "PROCESSOR %u:%x TIME %llu SOCKET %u APIC %x microcode %x\n",
		m->cpuvendor, m->cpuid, m->time, m->socketid, m->apicid,
		m->microcode);
}

static void print_mce(struct mce_hw_err *err)
{
	struct mce *m = &err->m;

	__print_mce(err);

	if (m->cpuvendor != X86_VENDOR_AMD && m->cpuvendor != X86_VENDOR_HYGON)
		pr_emerg_ratelimited(HW_ERR "Run the above through 'mcelog --ascii'\n");
}

#define PANIC_TIMEOUT 5 /* 5 seconds */

static atomic_t mce_panicked;

static int fake_panic;
static atomic_t mce_fake_panicked;

/* Panic in progress. Enable interrupts and wait for final IPI */
static void wait_for_panic(void)
{
	long timeout = PANIC_TIMEOUT*USEC_PER_SEC;

	preempt_disable();
	local_irq_enable();
	while (timeout-- > 0)
		udelay(1);
	if (panic_timeout == 0)
		panic_timeout = mca_cfg.panic_timeout;
	panic("Panicing machine check CPU died");
}

static const char *mce_dump_aux_info(struct mce *m)
{
	if (boot_cpu_has_bug(X86_BUG_TDX_PW_MCE))
		return tdx_dump_mce_info(m);

	return NULL;
}

static noinstr void mce_panic(const char *msg, struct mce_hw_err *final, char *exp)
{
	struct llist_node *pending;
	struct mce_evt_llist *l;
	int apei_err = 0;
	const char *memmsg;

	/*
	 * Allow instrumentation around external facilities usage. Not that it
	 * matters a whole lot since the machine is going to panic anyway.
	 */
	instrumentation_begin();

	if (!fake_panic) {
		/*
		 * Make sure only one CPU runs in machine check panic
		 */
		if (atomic_inc_return(&mce_panicked) > 1)
			wait_for_panic();
		barrier();

		bust_spinlocks(1);
		console_verbose();
	} else {
		/* Don't log too much for fake panic */
		if (atomic_inc_return(&mce_fake_panicked) > 1)
			goto out;
	}
	pending = mce_gen_pool_prepare_records();
	/* First print corrected ones that are still unlogged */
	llist_for_each_entry(l, pending, llnode) {
		struct mce_hw_err *err = &l->err;
		struct mce *m = &err->m;
		if (!(m->status & MCI_STATUS_UC)) {
			print_mce(err);
			if (!apei_err)
				apei_err = apei_write_mce(m);
		}
	}
	/* Now print uncorrected but with the final one last */
	llist_for_each_entry(l, pending, llnode) {
		struct mce_hw_err *err = &l->err;
		struct mce *m = &err->m;
		if (!(m->status & MCI_STATUS_UC))
			continue;
		if (!final || mce_cmp(m, &final->m)) {
			print_mce(err);
			if (!apei_err)
				apei_err = apei_write_mce(m);
		}
	}
	if (final) {
		print_mce(final);
		if (!apei_err)
			apei_err = apei_write_mce(&final->m);
	}
	if (exp)
		pr_emerg(HW_ERR "Machine check: %s\n", exp);

	memmsg = mce_dump_aux_info(&final->m);
	if (memmsg)
		pr_emerg(HW_ERR "Machine check: %s\n", memmsg);

	if (!fake_panic) {
		if (panic_timeout == 0)
			panic_timeout = mca_cfg.panic_timeout;

		/*
		 * Kdump skips the poisoned page in order to avoid
		 * touching the error bits again. Poison the page even
		 * if the error is fatal and the machine is about to
		 * panic.
		 */
		if (kexec_crash_loaded()) {
			if (final && (final->m.status & MCI_STATUS_ADDRV)) {
				struct page *p;
				p = pfn_to_online_page(final->m.addr >> PAGE_SHIFT);
				if (p)
					SetPageHWPoison(p);
			}
		}
		panic(msg);
	} else
		pr_emerg(HW_ERR "Fake kernel panic: %s\n", msg);

out:
	instrumentation_end();
}

/* Support code for software error injection */

static int msr_to_offset(u32 msr)
{
	unsigned bank = __this_cpu_read(injectm.bank);

	if (msr == mca_cfg.rip_msr)
		return offsetof(struct mce, ip);
	if (msr == mca_msr_reg(bank, MCA_STATUS))
		return offsetof(struct mce, status);
	if (msr == mca_msr_reg(bank, MCA_ADDR))
		return offsetof(struct mce, addr);
	if (msr == mca_msr_reg(bank, MCA_MISC))
		return offsetof(struct mce, misc);
	if (msr == MSR_IA32_MCG_STATUS)
		return offsetof(struct mce, mcgstatus);
	return -1;
}

void ex_handler_msr_mce(struct pt_regs *regs, bool wrmsr)
{
	if (wrmsr) {
		pr_emerg("MSR access error: WRMSR to 0x%x (tried to write 0x%08x%08x) at rIP: 0x%lx (%pS)\n",
			 (unsigned int)regs->cx, (unsigned int)regs->dx, (unsigned int)regs->ax,
			 regs->ip, (void *)regs->ip);
	} else {
		pr_emerg("MSR access error: RDMSR from 0x%x at rIP: 0x%lx (%pS)\n",
			 (unsigned int)regs->cx, regs->ip, (void *)regs->ip);
	}

	show_stack_regs(regs);

	panic("MCA architectural violation!\n");

	while (true)
		cpu_relax();
}

/* MSR access wrappers used for error injection */
noinstr u64 mce_rdmsrq(u32 msr)
{
	EAX_EDX_DECLARE_ARGS(val, low, high);

	if (__this_cpu_read(injectm.finished)) {
		int offset;
		u64 ret;

		instrumentation_begin();

		offset = msr_to_offset(msr);
		if (offset < 0)
			ret = 0;
		else
			ret = *(u64 *)((char *)this_cpu_ptr(&injectm) + offset);

		instrumentation_end();

		return ret;
	}

	/*
	 * RDMSR on MCA MSRs should not fault. If they do, this is very much an
	 * architectural violation and needs to be reported to hw vendor. Panic
	 * the box to not allow any further progress.
	 */
	asm volatile("1: rdmsr\n"
		     "2:\n"
		     _ASM_EXTABLE_TYPE(1b, 2b, EX_TYPE_RDMSR_IN_MCE)
		     : EAX_EDX_RET(val, low, high) : "c" (msr));


	return EAX_EDX_VAL(val, low, high);
}

noinstr void mce_wrmsrq(u32 msr, u64 v)
{
	u32 low, high;

	if (__this_cpu_read(injectm.finished)) {
		int offset;

		instrumentation_begin();

		offset = msr_to_offset(msr);
		if (offset >= 0)
			*(u64 *)((char *)this_cpu_ptr(&injectm) + offset) = v;

		instrumentation_end();

		return;
	}

	low  = (u32)v;
	high = (u32)(v >> 32);

	/* See comment in mce_rdmsrq() */
	asm volatile("1: wrmsr\n"
		     "2:\n"
		     _ASM_EXTABLE_TYPE(1b, 2b, EX_TYPE_WRMSR_IN_MCE)
		     : : "c" (msr), "a"(low), "d" (high) : "memory");
}

/*
 * Collect all global (w.r.t. this processor) status about this machine
 * check into our "mce" struct so that we can use it later to assess
 * the severity of the problem as we read per-bank specific details.
 */
static noinstr void mce_gather_info(struct mce_hw_err *err, struct pt_regs *regs)
{
	struct mce *m;
	/*
	 * Enable instrumentation around mce_prep_record() which calls external
	 * facilities.
	 */
	instrumentation_begin();
	mce_prep_record(err);
	instrumentation_end();

	m = &err->m;
	m->mcgstatus = mce_rdmsrq(MSR_IA32_MCG_STATUS);
	if (regs) {
		/*
		 * Get the address of the instruction at the time of
		 * the machine check error.
		 */
		if (m->mcgstatus & (MCG_STATUS_RIPV|MCG_STATUS_EIPV)) {
			m->ip = regs->ip;
			m->cs = regs->cs;

			/*
			 * When in VM86 mode make the cs look like ring 3
			 * always. This is a lie, but it's better than passing
			 * the additional vm86 bit around everywhere.
			 */
			if (v8086_mode(regs))
				m->cs |= 3;
		}
		/* Use accurate RIP reporting if available. */
		if (mca_cfg.rip_msr)
			m->ip = mce_rdmsrq(mca_cfg.rip_msr);
	}
}

bool mce_available(struct cpuinfo_x86 *c)
{
	if (mca_cfg.disabled)
		return false;
	return cpu_has(c, X86_FEATURE_MCE) && cpu_has(c, X86_FEATURE_MCA);
}

static void mce_schedule_work(void)
{
	if (!mce_gen_pool_empty())
		schedule_work(&mce_work);
}

static void mce_irq_work_cb(struct irq_work *entry)
{
	mce_schedule_work();
}

bool mce_usable_address(struct mce *m)
{
	if (!(m->status & MCI_STATUS_ADDRV))
		return false;

	switch (m->cpuvendor) {
	case X86_VENDOR_AMD:
		return amd_mce_usable_address(m);

	case X86_VENDOR_INTEL:
	case X86_VENDOR_ZHAOXIN:
		return intel_mce_usable_address(m);

	default:
		return true;
	}
}
EXPORT_SYMBOL_GPL(mce_usable_address);

bool mce_is_memory_error(struct mce *m)
{
	switch (m->cpuvendor) {
	case X86_VENDOR_AMD:
	case X86_VENDOR_HYGON:
		return amd_mce_is_memory_error(m);

	case X86_VENDOR_INTEL:
	case X86_VENDOR_ZHAOXIN:
		/*
		 * Intel SDM Volume 3B - 15.9.2 Compound Error Codes
		 *
		 * Bit 7 of the MCACOD field of IA32_MCi_STATUS is used for
		 * indicating a memory error. Bit 8 is used for indicating a
		 * cache hierarchy error. The combination of bit 2 and bit 3
		 * is used for indicating a `generic' cache hierarchy error
		 * But we can't just blindly check the above bits, because if
		 * bit 11 is set, then it is a bus/interconnect error - and
		 * either way the above bits just gives more detail on what
		 * bus/interconnect error happened. Note that bit 12 can be
		 * ignored, as it's the "filter" bit.
		 */
		return (m->status & 0xef80) == BIT(7) ||
		       (m->status & 0xef00) == BIT(8) ||
		       (m->status & 0xeffc) == 0xc;

	default:
		return false;
	}
}
EXPORT_SYMBOL_GPL(mce_is_memory_error);

static bool whole_page(struct mce *m)
{
	if (!mca_cfg.ser || !(m->status & MCI_STATUS_MISCV))
		return true;

	return MCI_MISC_ADDR_LSB(m->misc) >= PAGE_SHIFT;
}

bool mce_is_correctable(struct mce *m)
{
	if (m->cpuvendor == X86_VENDOR_AMD && m->status & MCI_STATUS_DEFERRED)
		return false;

	if (m->cpuvendor == X86_VENDOR_HYGON && m->status & MCI_STATUS_DEFERRED)
		return false;

	if (m->status & MCI_STATUS_UC)
		return false;

	return true;
}
EXPORT_SYMBOL_GPL(mce_is_correctable);

/*
 * Notify the user(s) about new machine check events.
 * Can be called from interrupt context, but not from machine check/NMI
 * context.
 */
static bool mce_notify_irq(void)
{
	/* Not more than two messages every minute */
	static DEFINE_RATELIMIT_STATE(ratelimit, 60*HZ, 2);

	if (test_and_clear_bit(0, &mce_need_notify)) {
		mce_work_trigger();

		if (__ratelimit(&ratelimit))
			pr_info(HW_ERR "Machine check events logged\n");

		return true;
	}

	return false;
}

static int mce_early_notifier(struct notifier_block *nb, unsigned long val,
			      void *data)
{
	struct mce_hw_err *err = to_mce_hw_err(data);

	if (!err)
		return NOTIFY_DONE;

	/* Emit the trace record: */
	trace_mce_record(err);

	set_bit(0, &mce_need_notify);

	mce_notify_irq();

	return NOTIFY_DONE;
}

static struct notifier_block early_nb = {
	.notifier_call	= mce_early_notifier,
	.priority	= MCE_PRIO_EARLY,
};

static int uc_decode_notifier(struct notifier_block *nb, unsigned long val,
			      void *data)
{
	struct mce *mce = (struct mce *)data;
	unsigned long pfn;

	if (!mce || !mce_usable_address(mce))
		return NOTIFY_DONE;

	if (mce->severity != MCE_AO_SEVERITY &&
	    mce->severity != MCE_DEFERRED_SEVERITY)
		return NOTIFY_DONE;

	pfn = (mce->addr & MCI_ADDR_PHYSADDR) >> PAGE_SHIFT;
	if (!memory_failure(pfn, 0)) {
		set_mce_nospec(pfn);
		mce->kflags |= MCE_HANDLED_UC;
	}

	return NOTIFY_OK;
}

static struct notifier_block mce_uc_nb = {
	.notifier_call	= uc_decode_notifier,
	.priority	= MCE_PRIO_UC,
};

static int mce_default_notifier(struct notifier_block *nb, unsigned long val,
				void *data)
{
	struct mce_hw_err *err = to_mce_hw_err(data);

	if (!err)
		return NOTIFY_DONE;

	if (mca_cfg.print_all || !(err->m.kflags))
		__print_mce(err);

	return NOTIFY_DONE;
}

static struct notifier_block mce_default_nb = {
	.notifier_call	= mce_default_notifier,
	/* lowest prio, we want it to run last. */
	.priority	= MCE_PRIO_LOWEST,
};

/*
 * Read ADDR and MISC registers.
 */
static noinstr void mce_read_aux(struct mce_hw_err *err, int i)
{
	struct mce *m = &err->m;

	if (m->status & MCI_STATUS_MISCV)
		m->misc = mce_rdmsrq(mca_msr_reg(i, MCA_MISC));

	if (m->status & MCI_STATUS_ADDRV) {
		m->addr = mce_rdmsrq(mca_msr_reg(i, MCA_ADDR));

		/*
		 * Mask the reported address by the reported granularity.
		 */
		if (mca_cfg.ser && (m->status & MCI_STATUS_MISCV)) {
			u8 shift = MCI_MISC_ADDR_LSB(m->misc);
			m->addr >>= shift;
			m->addr <<= shift;
		}

		smca_extract_err_addr(m);
	}

	if (mce_flags.smca) {
		m->ipid = mce_rdmsrq(MSR_AMD64_SMCA_MCx_IPID(i));

		if (m->status & MCI_STATUS_SYNDV) {
			m->synd = mce_rdmsrq(MSR_AMD64_SMCA_MCx_SYND(i));
			err->vendor.amd.synd1 = mce_rdmsrq(MSR_AMD64_SMCA_MCx_SYND1(i));
			err->vendor.amd.synd2 = mce_rdmsrq(MSR_AMD64_SMCA_MCx_SYND2(i));
		}
	}
}

DEFINE_PER_CPU(unsigned, mce_poll_count);

/*
 * Newer Intel systems that support software error
 * recovery need to make additional checks. Other
 * CPUs should skip over uncorrected errors, but log
 * everything else.
 */
static bool ser_should_log_poll_error(struct mce *m)
{
	/* Log "not enabled" (speculative) errors */
	if (!(m->status & MCI_STATUS_EN))
		return true;

	/*
	 * Log UCNA (SDM: 15.6.3 "UCR Error Classification")
	 * UC == 1 && PCC == 0 && S == 0
	 */
	if (!(m->status & MCI_STATUS_PCC) && !(m->status & MCI_STATUS_S))
		return true;

	return false;
}

static bool should_log_poll_error(enum mcp_flags flags, struct mce_hw_err *err)
{
	struct mce *m = &err->m;

	/* If this entry is not valid, ignore it. */
	if (!(m->status & MCI_STATUS_VAL))
		return false;

	/*
	 * If we are logging everything (at CPU online) or this
	 * is a corrected error, then we must log it.
	 */
	if ((flags & MCP_UC) || !(m->status & MCI_STATUS_UC))
		return true;

	if (mca_cfg.ser)
		return ser_should_log_poll_error(m);

	if (m->status & MCI_STATUS_UC)
		return false;

	return true;
}

static void clear_bank(struct mce *m)
{
	if (m->cpuvendor == X86_VENDOR_AMD)
		return amd_clear_bank(m);

	mce_wrmsrq(mca_msr_reg(m->bank, MCA_STATUS), 0);
}

/*
 * Poll for corrected events or events that happened before reset.
 * Those are just logged through /dev/mcelog.
 *
 * This is executed in standard interrupt context.
 *
 * Note: spec recommends to panic for fatal unsignalled
 * errors here. However this would be quite problematic --
 * we would need to reimplement the Monarch handling and
 * it would mess up the exclusion between exception handler
 * and poll handler -- * so we skip this for now.
 * These cases should not happen anyways, or only when the CPU
 * is already totally * confused. In this case it's likely it will
 * not fully execute the machine check handler either.
 */
void machine_check_poll(enum mcp_flags flags, mce_banks_t *b)
{
	struct mce_bank *mce_banks = this_cpu_ptr(mce_banks_array);
	struct mce_hw_err err;
	struct mce *m;
	int i;

	this_cpu_inc(mce_poll_count);

	mce_gather_info(&err, NULL);
	m = &err.m;

	if (flags & MCP_TIMESTAMP)
		m->tsc = rdtsc();

	for (i = 0; i < this_cpu_read(mce_num_banks); i++) {
		if (!mce_banks[i].ctl || !test_bit(i, *b))
			continue;

		m->misc = 0;
		m->addr = 0;
		m->bank = i;

		barrier();
		m->status = mce_rdmsrq(mca_msr_reg(i, MCA_STATUS));

		/*
		 * Update storm tracking here, before checking for the
		 * MCI_STATUS_VAL bit. Valid corrected errors count
		 * towards declaring, or maintaining, storm status. No
		 * error in a bank counts towards avoiding, or ending,
		 * storm status.
		 */
		if (!mca_cfg.cmci_disabled)
			mce_track_storm(m);

		/* Verify that the error should be logged based on hardware conditions. */
		if (!should_log_poll_error(flags, &err))
			continue;

		mce_read_aux(&err, i);
		m->severity = mce_severity(m, NULL, NULL, false);
		/*
		 * Don't get the IP here because it's unlikely to
		 * have anything to do with the actual error location.
		 */

		if (mca_cfg.dont_log_ce && !mce_usable_address(m))
			goto clear_it;

		if (flags & MCP_QUEUE_LOG)
			mce_gen_pool_add(&err);
		else
			mce_log(&err);

clear_it:
		clear_bank(m);
	}

	/*
	 * Don't clear MCG_STATUS here because it's only defined for
	 * exceptions.
	 */

	sync_core();
}
EXPORT_SYMBOL_GPL(machine_check_poll);

/*
 * During IFU recovery Sandy Bridge -EP4S processors set the RIPV and
 * EIPV bits in MCG_STATUS to zero on the affected logical processor (SDM
 * Vol 3B Table 15-20). But this confuses both the code that determines
 * whether the machine check occurred in kernel or user mode, and also
 * the severity assessment code. Pretend that EIPV was set, and take the
 * ip/cs values from the pt_regs that mce_gather_info() ignored earlier.
 */
static __always_inline void
quirk_sandybridge_ifu(int bank, struct mce *m, struct pt_regs *regs)
{
	if (bank != 0)
		return;
	if ((m->mcgstatus & (MCG_STATUS_EIPV|MCG_STATUS_RIPV)) != 0)
		return;
	if ((m->status & (MCI_STATUS_OVER|MCI_STATUS_UC|
		          MCI_STATUS_EN|MCI_STATUS_MISCV|MCI_STATUS_ADDRV|
			  MCI_STATUS_PCC|MCI_STATUS_S|MCI_STATUS_AR|
			  MCACOD)) !=
			 (MCI_STATUS_UC|MCI_STATUS_EN|
			  MCI_STATUS_MISCV|MCI_STATUS_ADDRV|MCI_STATUS_S|
			  MCI_STATUS_AR|MCACOD_INSTR))
		return;

	m->mcgstatus |= MCG_STATUS_EIPV;
	m->ip = regs->ip;
	m->cs = regs->cs;
}

/*
 * Disable fast string copy and return from the MCE handler upon the first SRAR
 * MCE on bank 1 due to a CPU erratum on Intel Skylake/Cascade Lake/Cooper Lake
 * CPUs.
 * The fast string copy instructions ("REP; MOVS*") could consume an
 * uncorrectable memory error in the cache line _right after_ the desired region
 * to copy and raise an MCE with RIP pointing to the instruction _after_ the
 * "REP; MOVS*".
 * This mitigation addresses the issue completely with the caveat of performance
 * degradation on the CPU affected. This is still better than the OS crashing on
 * MCEs raised on an irrelevant process due to "REP; MOVS*" accesses from a
 * kernel context (e.g., copy_page).
 *
 * Returns true when fast string copy on CPU has been disabled.
 */
static noinstr bool quirk_skylake_repmov(void)
{
	u64 mcgstatus   = mce_rdmsrq(MSR_IA32_MCG_STATUS);
	u64 misc_enable = mce_rdmsrq(MSR_IA32_MISC_ENABLE);
	u64 mc1_status;

	/*
	 * Apply the quirk only to local machine checks, i.e., no broadcast
	 * sync is needed.
	 */
	if (!(mcgstatus & MCG_STATUS_LMCES) ||
	    !(misc_enable & MSR_IA32_MISC_ENABLE_FAST_STRING))
		return false;

	mc1_status = mce_rdmsrq(MSR_IA32_MCx_STATUS(1));

	/* Check for a software-recoverable data fetch error. */
	if ((mc1_status &
	     (MCI_STATUS_VAL | MCI_STATUS_OVER | MCI_STATUS_UC | MCI_STATUS_EN |
	      MCI_STATUS_ADDRV | MCI_STATUS_MISCV | MCI_STATUS_PCC |
	      MCI_STATUS_AR | MCI_STATUS_S)) ==
	     (MCI_STATUS_VAL |                   MCI_STATUS_UC | MCI_STATUS_EN |
	      MCI_STATUS_ADDRV | MCI_STATUS_MISCV |
	      MCI_STATUS_AR | MCI_STATUS_S)) {
		misc_enable &= ~MSR_IA32_MISC_ENABLE_FAST_STRING;
		mce_wrmsrq(MSR_IA32_MISC_ENABLE, misc_enable);
		mce_wrmsrq(MSR_IA32_MCx_STATUS(1), 0);

		instrumentation_begin();
		pr_err_once("Erratum detected, disable fast string copy instructions.\n");
		instrumentation_end();

		return true;
	}

	return false;
}

/*
 * Some Zen-based Instruction Fetch Units set EIPV=RIPV=0 on poison consumption
 * errors. This means mce_gather_info() will not save the "ip" and "cs" registers.
 *
 * However, the context is still valid, so save the "cs" register for later use.
 *
 * The "ip" register is truly unknown, so don't save it or fixup EIPV/RIPV.
 *
 * The Instruction Fetch Unit is at MCA bank 1 for all affected systems.
 */
static __always_inline void quirk_zen_ifu(int bank, struct mce *m, struct pt_regs *regs)
{
	if (bank != 1)
		return;
	if (!(m->status & MCI_STATUS_POISON))
		return;

	m->cs = regs->cs;
}

/*
 * Do a quick check if any of the events requires a panic.
 * This decides if we keep the events around or clear them.
 */
static __always_inline int mce_no_way_out(struct mce_hw_err *err, char **msg, unsigned long *validp,
					  struct pt_regs *regs)
{
	struct mce *m = &err->m;
	char *tmp = *msg;
	int i;

	for (i = 0; i < this_cpu_read(mce_num_banks); i++) {
		m->status = mce_rdmsrq(mca_msr_reg(i, MCA_STATUS));
		if (!(m->status & MCI_STATUS_VAL))
			continue;

		arch___set_bit(i, validp);
		if (mce_flags.snb_ifu_quirk)
			quirk_sandybridge_ifu(i, m, regs);

		if (mce_flags.zen_ifu_quirk)
			quirk_zen_ifu(i, m, regs);

		m->bank = i;
		if (mce_severity(m, regs, &tmp, true) >= MCE_PANIC_SEVERITY) {
			mce_read_aux(err, i);
			*msg = tmp;
			return 1;
		}
	}
	return 0;
}

/*
 * Variable to establish order between CPUs while scanning.
 * Each CPU spins initially until executing is equal its number.
 */
static atomic_t mce_executing;

/*
 * Defines order of CPUs on entry. First CPU becomes Monarch.
 */
static atomic_t mce_callin;

/*
 * Track which CPUs entered the MCA broadcast synchronization and which not in
 * order to print holdouts.
 */
static cpumask_t mce_missing_cpus = CPU_MASK_ALL;

/*
 * Check if a timeout waiting for other CPUs happened.
 */
static noinstr int mce_timed_out(u64 *t, const char *msg)
{
	int ret = 0;

	/* Enable instrumentation around calls to external facilities */
	instrumentation_begin();

	/*
	 * The others already did panic for some reason.
	 * Bail out like in a timeout.
	 * rmb() to tell the compiler that system_state
	 * might have been modified by someone else.
	 */
	rmb();
	if (atomic_read(&mce_panicked))
		wait_for_panic();
	if (!mca_cfg.monarch_timeout)
		goto out;
	if ((s64)*t < SPINUNIT) {
		if (cpumask_and(&mce_missing_cpus, cpu_online_mask, &mce_missing_cpus))
			pr_emerg("CPUs not responding to MCE broadcast (may include false positives): %*pbl\n",
				 cpumask_pr_args(&mce_missing_cpus));
		mce_panic(msg, NULL, NULL);

		ret = 1;
		goto out;
	}
	*t -= SPINUNIT;

out:
	touch_nmi_watchdog();

	instrumentation_end();

	return ret;
}

/*
 * The Monarch's reign.  The Monarch is the CPU who entered
 * the machine check handler first. It waits for the others to
 * raise the exception too and then grades them. When any
 * error is fatal panic. Only then let the others continue.
 *
 * The other CPUs entering the MCE handler will be controlled by the
 * Monarch. They are called Subjects.
 *
 * This way we prevent any potential data corruption in a unrecoverable case
 * and also makes sure always all CPU's errors are examined.
 *
 * Also this detects the case of a machine check event coming from outer
 * space (not detected by any CPUs) In this case some external agent wants
 * us to shut down, so panic too.
 *
 * The other CPUs might still decide to panic if the handler happens
 * in a unrecoverable place, but in this case the system is in a semi-stable
 * state and won't corrupt anything by itself. It's ok to let the others
 * continue for a bit first.
 *
 * All the spin loops have timeouts; when a timeout happens a CPU
 * typically elects itself to be Monarch.
 */
static void mce_reign(void)
{
	struct mce_hw_err *err = NULL;
	struct mce *m = NULL;
	int global_worst = 0;
	char *msg = NULL;
	int cpu;

	/*
	 * This CPU is the Monarch and the other CPUs have run
	 * through their handlers.
	 * Grade the severity of the errors of all the CPUs.
	 */
	for_each_possible_cpu(cpu) {
		struct mce_hw_err *etmp = &per_cpu(hw_errs_seen, cpu);
		struct mce *mtmp = &etmp->m;

		if (mtmp->severity > global_worst) {
			global_worst = mtmp->severity;
			err = &per_cpu(hw_errs_seen, cpu);
			m = &err->m;
		}
	}

	/*
	 * Cannot recover? Panic here then.
	 * This dumps all the mces in the log buffer and stops the
	 * other CPUs.
	 */
	if (m && global_worst >= MCE_PANIC_SEVERITY) {
		/* call mce_severity() to get "msg" for panic */
		mce_severity(m, NULL, &msg, true);
		mce_panic("Fatal machine check", err, msg);
	}

	/*
	 * For UC somewhere we let the CPU who detects it handle it.
	 * Also must let continue the others, otherwise the handling
	 * CPU could deadlock on a lock.
	 */

	/*
	 * No machine check event found. Must be some external
	 * source or one CPU is hung. Panic.
	 */
	if (global_worst <= MCE_KEEP_SEVERITY)
		mce_panic("Fatal machine check from unknown source", NULL, NULL);

	/*
	 * Now clear all the hw_errs_seen so that they don't reappear on
	 * the next mce.
	 */
	for_each_possible_cpu(cpu)
		memset(&per_cpu(hw_errs_seen, cpu), 0, sizeof(struct mce_hw_err));
}

static atomic_t global_nwo;

/*
 * Start of Monarch synchronization. This waits until all CPUs have
 * entered the exception handler and then determines if any of them
 * saw a fatal event that requires panic. Then it executes them
 * in the entry order.
 * TBD double check parallel CPU hotunplug
 */
static noinstr int mce_start(int *no_way_out)
{
	u64 timeout = (u64)mca_cfg.monarch_timeout * NSEC_PER_USEC;
	int order, ret = -1;

	if (!timeout)
		return ret;

	raw_atomic_add(*no_way_out, &global_nwo);
	/*
	 * Rely on the implied barrier below, such that global_nwo
	 * is updated before mce_callin.
	 */
	order = raw_atomic_inc_return(&mce_callin);
	arch_cpumask_clear_cpu(smp_processor_id(), &mce_missing_cpus);

	/* Enable instrumentation around calls to external facilities */
	instrumentation_begin();

	/*
	 * Wait for everyone.
	 */
	while (raw_atomic_read(&mce_callin) != num_online_cpus()) {
		if (mce_timed_out(&timeout,
				  "Timeout: Not all CPUs entered broadcast exception handler")) {
			raw_atomic_set(&global_nwo, 0);
			goto out;
		}
		ndelay(SPINUNIT);
	}

	/*
	 * mce_callin should be read before global_nwo
	 */
	smp_rmb();

	if (order == 1) {
		/*
		 * Monarch: Starts executing now, the others wait.
		 */
		raw_atomic_set(&mce_executing, 1);
	} else {
		/*
		 * Subject: Now start the scanning loop one by one in
		 * the original callin order.
		 * This way when there are any shared banks it will be
		 * only seen by one CPU before cleared, avoiding duplicates.
		 */
		while (raw_atomic_read(&mce_executing) < order) {
			if (mce_timed_out(&timeout,
					  "Timeout: Subject CPUs unable to finish machine check processing")) {
				raw_atomic_set(&global_nwo, 0);
				goto out;
			}
			ndelay(SPINUNIT);
		}
	}

	/*
	 * Cache the global no_way_out state.
	 */
	*no_way_out = raw_atomic_read(&global_nwo);

	ret = order;

out:
	instrumentation_end();

	return ret;
}

/*
 * Synchronize between CPUs after main scanning loop.
 * This invokes the bulk of the Monarch processing.
 */
static noinstr int mce_end(int order)
{
	u64 timeout = (u64)mca_cfg.monarch_timeout * NSEC_PER_USEC;
	int ret = -1;

	/* Allow instrumentation around external facilities. */
	instrumentation_begin();

	if (!timeout)
		goto reset;
	if (order < 0)
		goto reset;

	/*
	 * Allow others to run.
	 */
	atomic_inc(&mce_executing);

	if (order == 1) {
		/*
		 * Monarch: Wait for everyone to go through their scanning
		 * loops.
		 */
		while (atomic_read(&mce_executing) <= num_online_cpus()) {
			if (mce_timed_out(&timeout,
					  "Timeout: Monarch CPU unable to finish machine check processing"))
				goto reset;
			ndelay(SPINUNIT);
		}

		mce_reign();
		barrier();
		ret = 0;
	} else {
		/*
		 * Subject: Wait for Monarch to finish.
		 */
		while (atomic_read(&mce_executing) != 0) {
			if (mce_timed_out(&timeout,
					  "Timeout: Monarch CPU did not finish machine check processing"))
				goto reset;
			ndelay(SPINUNIT);
		}

		/*
		 * Don't reset anything. That's done by the Monarch.
		 */
		ret = 0;
		goto out;
	}

	/*
	 * Reset all global state.
	 */
reset:
	atomic_set(&global_nwo, 0);
	atomic_set(&mce_callin, 0);
	cpumask_setall(&mce_missing_cpus);
	barrier();

	/*
	 * Let others run again.
	 */
	atomic_set(&mce_executing, 0);

out:
	instrumentation_end();

	return ret;
}

static __always_inline void mce_clear_state(unsigned long *toclear)
{
	int i;

	for (i = 0; i < this_cpu_read(mce_num_banks); i++) {
		if (arch_test_bit(i, toclear))
			mce_wrmsrq(mca_msr_reg(i, MCA_STATUS), 0);
	}
}

/*
 * Cases where we avoid rendezvous handler timeout:
 * 1) If this CPU is offline.
 *
 * 2) If crashing_cpu was set, e.g. we're entering kdump and we need to
 *  skip those CPUs which remain looping in the 1st kernel - see
 *  crash_nmi_callback().
 *
 * Note: there still is a small window between kexec-ing and the new,
 * kdump kernel establishing a new #MC handler where a broadcasted MCE
 * might not get handled properly.
 */
static noinstr bool mce_check_crashing_cpu(void)
{
	unsigned int cpu = smp_processor_id();

	if (arch_cpu_is_offline(cpu) ||
	    (crashing_cpu != -1 && crashing_cpu != cpu)) {
		u64 mcgstatus;

		mcgstatus = native_rdmsrq(MSR_IA32_MCG_STATUS);

		if (boot_cpu_data.x86_vendor == X86_VENDOR_ZHAOXIN) {
			if (mcgstatus & MCG_STATUS_LMCES)
				return false;
		}

		if (mcgstatus & MCG_STATUS_RIPV) {
			native_wrmsrq(MSR_IA32_MCG_STATUS, 0);
			return true;
		}
	}
	return false;
}

static __always_inline int
__mc_scan_banks(struct mce_hw_err *err, struct pt_regs *regs,
		struct mce_hw_err *final, unsigned long *toclear,
		unsigned long *valid_banks, int no_way_out, int *worst)
{
	struct mce_bank *mce_banks = this_cpu_ptr(mce_banks_array);
	struct mca_config *cfg = &mca_cfg;
	int severity, i, taint = 0;
	struct mce *m = &err->m;

	for (i = 0; i < this_cpu_read(mce_num_banks); i++) {
		arch___clear_bit(i, toclear);
		if (!arch_test_bit(i, valid_banks))
			continue;

		if (!mce_banks[i].ctl)
			continue;

		m->misc = 0;
		m->addr = 0;
		m->bank = i;

		m->status = mce_rdmsrq(mca_msr_reg(i, MCA_STATUS));
		if (!(m->status & MCI_STATUS_VAL))
			continue;

		/*
		 * Corrected or non-signaled errors are handled by
		 * machine_check_poll(). Leave them alone, unless this panics.
		 */
		if (!(m->status & (cfg->ser ? MCI_STATUS_S : MCI_STATUS_UC)) &&
			!no_way_out)
			continue;

		/* Set taint even when machine check was not enabled. */
		taint++;

		severity = mce_severity(m, regs, NULL, true);

		/*
		 * When machine check was for corrected/deferred handler don't
		 * touch, unless we're panicking.
		 */
		if ((severity == MCE_KEEP_SEVERITY ||
		     severity == MCE_UCNA_SEVERITY) && !no_way_out)
			continue;

		arch___set_bit(i, toclear);

		/* Machine check event was not enabled. Clear, but ignore. */
		if (severity == MCE_NO_SEVERITY)
			continue;

		mce_read_aux(err, i);

		/* assuming valid severity level != 0 */
		m->severity = severity;

		/*
		 * Enable instrumentation around the mce_log() call which is
		 * done in #MC context, where instrumentation is disabled.
		 */
		instrumentation_begin();
		mce_log(err);
		instrumentation_end();

		if (severity > *worst) {
			*final = *err;
			*worst = severity;
		}
	}

	/* mce_clear_state will clear *final, save locally for use later */
	*err = *final;

	return taint;
}

static void kill_me_now(struct callback_head *ch)
{
	struct task_struct *p = container_of(ch, struct task_struct, mce_kill_me);

	p->mce_count = 0;
	force_sig(SIGBUS);
}

static void kill_me_maybe(struct callback_head *cb)
{
	struct task_struct *p = container_of(cb, struct task_struct, mce_kill_me);
	int flags = MF_ACTION_REQUIRED;
	unsigned long pfn;
	int ret;

	p->mce_count = 0;
	pr_err("Uncorrected hardware memory error in user-access at %llx", p->mce_addr);

	if (!p->mce_ripv)
		flags |= MF_MUST_KILL;

	pfn = (p->mce_addr & MCI_ADDR_PHYSADDR) >> PAGE_SHIFT;
	ret = memory_failure(pfn, flags);
	if (!ret) {
		set_mce_nospec(pfn);
		sync_core();
		return;
	}

	/*
	 * -EHWPOISON from memory_failure() means that it already sent SIGBUS
	 * to the current process with the proper error info,
	 * -EOPNOTSUPP means hwpoison_filter() filtered the error event,
	 *
	 * In both cases, no further processing is required.
	 */
	if (ret == -EHWPOISON || ret == -EOPNOTSUPP)
		return;

	pr_err("Memory error not recovered");
	kill_me_now(cb);
}

static void kill_me_never(struct callback_head *cb)
{
	struct task_struct *p = container_of(cb, struct task_struct, mce_kill_me);
	unsigned long pfn;

	p->mce_count = 0;
	pr_err("Kernel accessed poison in user space at %llx\n", p->mce_addr);
	pfn = (p->mce_addr & MCI_ADDR_PHYSADDR) >> PAGE_SHIFT;
	if (!memory_failure(pfn, 0))
		set_mce_nospec(pfn);
}

static void queue_task_work(struct mce_hw_err *err, char *msg, void (*func)(struct callback_head *))
{
	int count = ++current->mce_count;
	struct mce *m = &err->m;

	/* First call, save all the details */
	if (count == 1) {
		current->mce_addr = m->addr;
		current->mce_kflags = m->kflags;
		current->mce_ripv = !!(m->mcgstatus & MCG_STATUS_RIPV);
		current->mce_whole_page = whole_page(m);
		current->mce_kill_me.func = func;
	}

	/* Ten is likely overkill. Don't expect more than two faults before task_work() */
	if (count > 10)
		mce_panic("Too many consecutive machine checks while accessing user data",
			  err, msg);

	/* Second or later call, make sure page address matches the one from first call */
	if (count > 1 && (current->mce_addr >> PAGE_SHIFT) != (m->addr >> PAGE_SHIFT))
		mce_panic("Consecutive machine checks to different user pages", err, msg);

	/* Do not call task_work_add() more than once */
	if (count > 1)
		return;

	task_work_add(current, &current->mce_kill_me, TWA_RESUME);
}

/* Handle unconfigured int18 (should never happen) */
static noinstr void unexpected_machine_check(struct pt_regs *regs)
{
	instrumentation_begin();
	pr_err("CPU#%d: Unexpected int18 (Machine Check)\n",
	       smp_processor_id());
	instrumentation_end();
}

/*
 * The actual machine check handler. This only handles real exceptions when
 * something got corrupted coming in through int 18.
 *
 * This is executed in #MC context not subject to normal locking rules.
 * This implies that most kernel services cannot be safely used. Don't even
 * think about putting a printk in there!
 *
 * 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.
 *
 * Currently, the #MC handler calls out to a number of external facilities
 * and, therefore, allows instrumentation around them. The optimal thing to
 * have would be to do the absolutely minimal work required in #MC context
 * and have instrumentation disabled only around that. Further processing can
 * then happen in process context where instrumentation is allowed. Achieving
 * that requires careful auditing and modifications. Until then, the code
 * allows instrumentation temporarily, where required. *
 */
noinstr void do_machine_check(struct pt_regs *regs)
{
	int worst = 0, order, no_way_out, kill_current_task, lmce, taint = 0;
	DECLARE_BITMAP(valid_banks, MAX_NR_BANKS) = { 0 };
	DECLARE_BITMAP(toclear, MAX_NR_BANKS) = { 0 };
	struct mce_hw_err *final;
	struct mce_hw_err err;
	char *msg = NULL;
	struct mce *m;

	if (unlikely(mce_flags.p5))
		return pentium_machine_check(regs);
	else if (unlikely(mce_flags.winchip))
		return winchip_machine_check(regs);
	else if (unlikely(!mca_cfg.initialized))
		return unexpected_machine_check(regs);

	if (mce_flags.skx_repmov_quirk && quirk_skylake_repmov())
		goto clear;

	/*
	 * Establish sequential order between the CPUs entering the machine
	 * check handler.
	 */
	order = -1;

	/*
	 * If no_way_out gets set, there is no safe way to recover from this
	 * MCE.
	 */
	no_way_out = 0;

	/*
	 * If kill_current_task is not set, there might be a way to recover from this
	 * error.
	 */
	kill_current_task = 0;

	/*
	 * MCEs are always local on AMD. Same is determined by MCG_STATUS_LMCES
	 * on Intel.
	 */
	lmce = 1;

	this_cpu_inc(mce_exception_count);

	mce_gather_info(&err, regs);
	m = &err.m;
	m->tsc = rdtsc();

	final = this_cpu_ptr(&hw_errs_seen);
	*final = err;

	no_way_out = mce_no_way_out(&err, &msg, valid_banks, regs);

	barrier();

	/*
	 * When no restart IP might need to kill or panic.
	 * Assume the worst for now, but if we find the
	 * severity is MCE_AR_SEVERITY we have other options.
	 */
	if (!(m->mcgstatus & MCG_STATUS_RIPV))
		kill_current_task = 1;
	/*
	 * Check if this MCE is signaled to only this logical processor,
	 * on Intel, Zhaoxin only.
	 */
	if (m->cpuvendor == X86_VENDOR_INTEL ||
	    m->cpuvendor == X86_VENDOR_ZHAOXIN)
		lmce = m->mcgstatus & MCG_STATUS_LMCES;

	/*
	 * Local machine check may already know that we have to panic.
	 * Broadcast machine check begins rendezvous in mce_start()
	 * Go through all banks in exclusion of the other CPUs. This way we
	 * don't report duplicated events on shared banks because the first one
	 * to see it will clear it.
	 */
	if (lmce) {
		if (no_way_out)
			mce_panic("Fatal local machine check", &err, msg);
	} else {
		order = mce_start(&no_way_out);
	}

	taint = __mc_scan_banks(&err, regs, final, toclear, valid_banks, no_way_out, &worst);

	if (!no_way_out)
		mce_clear_state(toclear);

	/*
	 * Do most of the synchronization with other CPUs.
	 * When there's any problem use only local no_way_out state.
	 */
	if (!lmce) {
		if (mce_end(order) < 0) {
			if (!no_way_out)
				no_way_out = worst >= MCE_PANIC_SEVERITY;

			if (no_way_out)
				mce_panic("Fatal machine check on current CPU", &err, msg);
		}
	} else {
		/*
		 * If there was a fatal machine check we should have
		 * already called mce_panic earlier in this function.
		 * Since we re-read the banks, we might have found
		 * something new. Check again to see if we found a
		 * fatal error. We call "mce_severity()" again to
		 * make sure we have the right "msg".
		 */
		if (worst >= MCE_PANIC_SEVERITY) {
			mce_severity(m, regs, &msg, true);
			mce_panic("Local fatal machine check!", &err, msg);
		}
	}

	/*
	 * Enable instrumentation around the external facilities like task_work_add()
	 * (via queue_task_work()), fixup_exception() etc. For now, that is. Fixing this
	 * properly would need a lot more involved reorganization.
	 */
	instrumentation_begin();

	if (taint)
		add_taint(TAINT_MACHINE_CHECK, LOCKDEP_NOW_UNRELIABLE);

	if (worst != MCE_AR_SEVERITY && !kill_current_task)
		goto out;

	/* Fault was in user mode and we need to take some action */
	if ((m->cs & 3) == 3) {
		/* If this triggers there is no way to recover. Die hard. */
		BUG_ON(!on_thread_stack() || !user_mode(regs));

		if (!mce_usable_address(m))
			queue_task_work(&err, msg, kill_me_now);
		else
			queue_task_work(&err, msg, kill_me_maybe);

	} else if (m->mcgstatus & MCG_STATUS_SEAM_NR) {
		/*
		 * Saved RIP on stack makes it look like the machine check
		 * was taken in the kernel on the instruction following
		 * the entry to SEAM mode. But MCG_STATUS_SEAM_NR indicates
		 * that the machine check was taken inside SEAM non-root
		 * mode.  CPU core has already marked that guest as dead.
		 * It is OK for the kernel to resume execution at the
		 * apparent point of the machine check as the fault did
		 * not occur there. Mark the page as poisoned so it won't
		 * be added to free list when the guest is terminated.
		 */
		if (mce_usable_address(m)) {
			struct page *p = pfn_to_online_page(m->addr >> PAGE_SHIFT);

			if (p)
				SetPageHWPoison(p);
		}
	} else {
		/*
		 * Handle an MCE which has happened in kernel space but from
		 * which the kernel can recover: ex_has_fault_handler() has
		 * already verified that the rIP at which the error happened is
		 * a rIP from which the kernel can recover (by jumping to
		 * recovery code specified in _ASM_EXTABLE_FAULT()) and the
		 * corresponding exception handler which would do that is the
		 * proper one.
		 */
		if (m->kflags & MCE_IN_KERNEL_RECOV) {
			if (!fixup_exception(regs, X86_TRAP_MC, 0, 0))
				mce_panic("Failed kernel mode recovery", &err, msg);
		}

		if (m->kflags & MCE_IN_KERNEL_COPYIN)
			queue_task_work(&err, msg, kill_me_never);
	}

out:
	instrumentation_end();

clear:
	mce_wrmsrq(MSR_IA32_MCG_STATUS, 0);
}
EXPORT_SYMBOL_GPL(do_machine_check);

#ifndef CONFIG_MEMORY_FAILURE
int memory_failure(unsigned long pfn, int flags)
{
	/* mce_severity() should not hand us an ACTION_REQUIRED error */
	BUG_ON(flags & MF_ACTION_REQUIRED);
	pr_err("Uncorrected memory error in page 0x%lx ignored\n"
	       "Rebuild kernel with CONFIG_MEMORY_FAILURE=y for smarter handling\n",
	       pfn);

	return 0;
}
#endif

/*
 * Periodic polling timer for "silent" machine check errors.  If the
 * poller finds an MCE, poll 2x faster.  When the poller finds no more
 * errors, poll 2x slower (up to check_interval seconds).
 */
static unsigned long check_interval = INITIAL_CHECK_INTERVAL;

static DEFINE_PER_CPU(unsigned long, mce_next_interval); /* in jiffies */
static DEFINE_PER_CPU(struct timer_list, mce_timer);

static void __start_timer(struct timer_list *t, unsigned long interval)
{
	unsigned long when = jiffies + interval;
	unsigned long flags;

	local_irq_save(flags);

	if (!timer_pending(t) || time_before(when, t->expires))
		mod_timer(t, round_jiffies(when));

	local_irq_restore(flags);
}

static void mc_poll_banks_default(void)
{
	machine_check_poll(0, this_cpu_ptr(&mce_poll_banks));
}

void (*mc_poll_banks)(void) = mc_poll_banks_default;

static bool should_enable_timer(unsigned long iv)
{
	return !mca_cfg.ignore_ce && iv;
}

static void mce_timer_fn(struct timer_list *t)
{
	struct timer_list *cpu_t = this_cpu_ptr(&mce_timer);
	unsigned long iv;

	WARN_ON(cpu_t != t);

	iv = __this_cpu_read(mce_next_interval);

	if (mce_available(this_cpu_ptr(&cpu_info)))
		mc_poll_banks();

	/*
	 * Alert userspace if needed. If we logged an MCE, reduce the polling
	 * interval, otherwise increase the polling interval.
	 */
	if (mce_notify_irq())
		iv = max(iv / 2, (unsigned long) HZ/100);
	else
		iv = min(iv * 2, round_jiffies_relative(check_interval * HZ));

	if (mce_get_storm_mode()) {
		__start_timer(t, HZ);
	} else if (should_enable_timer(iv)) {
		__this_cpu_write(mce_next_interval, iv);
		__start_timer(t, iv);
	}
}

/*
 * When a storm starts on any bank on this CPU, switch to polling
 * once per second. When the storm ends, revert to the default
 * polling interval.
 */
void mce_timer_kick(bool storm)
{
	struct timer_list *t = this_cpu_ptr(&mce_timer);

	mce_set_storm_mode(storm);

	if (storm)
		__start_timer(t, HZ);
	else
		__this_cpu_write(mce_next_interval, check_interval * HZ);
}

/* Must not be called in IRQ context where timer_delete_sync() can deadlock */
static void mce_timer_delete_all(void)
{
	int cpu;

	for_each_online_cpu(cpu)
		timer_delete_sync(&per_cpu(mce_timer, cpu));
}

static void __mcheck_cpu_mce_banks_init(void)
{
	struct mce_bank *mce_banks = this_cpu_ptr(mce_banks_array);
	u8 n_banks = this_cpu_read(mce_num_banks);
	int i;

	for (i = 0; i < n_banks; i++) {
		struct mce_bank *b = &mce_banks[i];

		/*
		 * Init them all by default.
		 *
		 * The required vendor quirks will be applied before
		 * __mcheck_cpu_init_prepare_banks() does the final bank setup.
		 */
		b->ctl = -1ULL;
		b->init = true;
	}
}

/*
 * Initialize Machine Checks for a CPU.
 */
static void __mcheck_cpu_cap_init(void)
{
	u64 cap;
	u8 b;

	rdmsrq(MSR_IA32_MCG_CAP, cap);

	b = cap & MCG_BANKCNT_MASK;

	if (b > MAX_NR_BANKS) {
		pr_warn("CPU%d: Using only %u machine check banks out of %u\n",
			smp_processor_id(), MAX_NR_BANKS, b);
		b = MAX_NR_BANKS;
	}

	this_cpu_write(mce_num_banks, b);

	__mcheck_cpu_mce_banks_init();
}

static void __mcheck_cpu_init_generic(void)
{
	u64 cap;

	rdmsrq(MSR_IA32_MCG_CAP, cap);
	if (cap & MCG_CTL_P)
		wrmsr(MSR_IA32_MCG_CTL, 0xffffffff, 0xffffffff);
}

static void __mcheck_cpu_init_prepare_banks(void)
{
	struct mce_bank *mce_banks = this_cpu_ptr(mce_banks_array);
	u64 msrval;
	int i;

	/*
	 * Log the machine checks left over from the previous reset. Log them
	 * only, do not start processing them. That will happen in mcheck_late_init()
	 * when all consumers have been registered on the notifier chain.
	 */
	if (mca_cfg.bootlog) {
		mce_banks_t all_banks;

		bitmap_fill(all_banks, MAX_NR_BANKS);
		machine_check_poll(MCP_UC | MCP_QUEUE_LOG, &all_banks);
	}

	for (i = 0; i < this_cpu_read(mce_num_banks); i++) {
		struct mce_bank *b = &mce_banks[i];

		if (!b->init)
			continue;

		wrmsrq(mca_msr_reg(i, MCA_CTL), b->ctl);
		wrmsrq(mca_msr_reg(i, MCA_STATUS), 0);

		rdmsrq(mca_msr_reg(i, MCA_CTL), msrval);
		b->init = !!msrval;
	}
}

static void amd_apply_global_quirks(struct cpuinfo_x86 *c)
{
	if (c->x86 < 0x11 && mca_cfg.bootlog < 0) {
		/*
		 * Lots of broken BIOS around that don't clear them
		 * by default and leave crap in there. Don't log:
		 */
		mca_cfg.bootlog = 0;
	}

	/*
	 * overflow_recov is supported for F15h Models 00h-0fh
	 * even though we don't have a CPUID bit for it.
	 */
	if (c->x86 == 0x15 && c->x86_model <= 0xf)
		mce_flags.overflow_recov = 1;

	if (c->x86 >= 0x17 && c->x86 <= 0x1A)
		mce_flags.zen_ifu_quirk = 1;
}

static void intel_apply_global_quirks(struct cpuinfo_x86 *c)
{
	/* Older CPUs (prior to family 6) don't need quirks. */
	if (c->x86_vfm < INTEL_PENTIUM_PRO)
		return;

	/*
	 * All newer Intel systems support MCE broadcasting. Enable
	 * synchronization with a one second timeout.
	 */
	if (c->x86_vfm >= INTEL_CORE_YONAH && mca_cfg.monarch_timeout < 0)
		mca_cfg.monarch_timeout = USEC_PER_SEC;

	/*
	 * There are also broken BIOSes on some Pentium M and
	 * earlier systems:
	 */
	if (c->x86_vfm < INTEL_CORE_YONAH && mca_cfg.bootlog < 0)
		mca_cfg.bootlog = 0;

	if (c->x86_vfm == INTEL_SANDYBRIDGE_X)
		mce_flags.snb_ifu_quirk = 1;

	/*
	 * Skylake, Cascacde Lake and Cooper Lake require a quirk on
	 * rep movs.
	 */
	if (c->x86_vfm == INTEL_SKYLAKE_X)
		mce_flags.skx_repmov_quirk = 1;
}

static void zhaoxin_apply_global_quirks(struct cpuinfo_x86 *c)
{
	/*
	 * All newer Zhaoxin CPUs support MCE broadcasting. Enable
	 * synchronization with a one second timeout.
	 */
	if (c->x86 > 6 || (c->x86_model == 0x19 || c->x86_model == 0x1f)) {
		if (mca_cfg.monarch_timeout < 0)
			mca_cfg.monarch_timeout = USEC_PER_SEC;
	}
}

static bool __mcheck_cpu_ancient_init(struct cpuinfo_x86 *c)
{
	if (c->x86 != 5)
		return false;

	switch (c->x86_vendor) {
	case X86_VENDOR_INTEL:
		intel_p5_mcheck_init(c);
		mce_flags.p5 = 1;
		return true;
	case X86_VENDOR_CENTAUR:
		winchip_mcheck_init(c);
		mce_flags.winchip = 1;
		return true;
	default:
		return false;
	}

	return false;
}

static void mce_centaur_feature_init(struct cpuinfo_x86 *c)
{
	struct mca_config *cfg = &mca_cfg;

	 /*
	  * All newer Centaur CPUs support MCE broadcasting. Enable
	  * synchronization with a one second timeout.
	  */
	if ((c->x86 == 6 && c->x86_model == 0xf && c->x86_stepping >= 0xe) ||
	     c->x86 > 6) {
		if (cfg->monarch_timeout < 0)
			cfg->monarch_timeout = USEC_PER_SEC;
	}
}

static void mce_zhaoxin_feature_init(struct cpuinfo_x86 *c)
{
	struct mce_bank *mce_banks = this_cpu_ptr(mce_banks_array);

	/*
	 * These CPUs have MCA bank 8 which reports only one error type called
	 * SVAD (System View Address Decoder). The reporting of that error is
	 * controlled by IA32_MC8.CTL.0.
	 *
	 * If enabled, prefetching on these CPUs will cause SVAD MCE when
	 * virtual machines start and result in a system  panic. Always disable
	 * bank 8 SVAD error by default.
	 */
	if ((c->x86 == 7 && c->x86_model == 0x1b) ||
	    (c->x86_model == 0x19 || c->x86_model == 0x1f)) {
		if (this_cpu_read(mce_num_banks) > 8)
			mce_banks[8].ctl = 0;
	}

	intel_init_cmci();
	intel_init_lmce();
}

static void mce_zhaoxin_feature_clear(struct cpuinfo_x86 *c)
{
	intel_clear_lmce();
}

static void __mcheck_cpu_init_vendor(struct cpuinfo_x86 *c)
{
	switch (c->x86_vendor) {
	case X86_VENDOR_INTEL:
		mce_intel_feature_init(c);
		break;

	case X86_VENDOR_AMD:
	case X86_VENDOR_HYGON:
		mce_amd_feature_init(c);
		break;

	case X86_VENDOR_CENTAUR:
		mce_centaur_feature_init(c);
		break;

	case X86_VENDOR_ZHAOXIN:
		mce_zhaoxin_feature_init(c);
		break;

	default:
		break;
	}
}

static void __mcheck_cpu_clear_vendor(struct cpuinfo_x86 *c)
{
	switch (c->x86_vendor) {
	case X86_VENDOR_INTEL:
		mce_intel_feature_clear(c);
		break;

	case X86_VENDOR_ZHAOXIN:
		mce_zhaoxin_feature_clear(c);
		break;

	default:
		break;
	}
}

static void mce_start_timer(struct timer_list *t)
{
	unsigned long iv = check_interval * HZ;

	if (should_enable_timer(iv)) {
		this_cpu_write(mce_next_interval, iv);
		__start_timer(t, iv);
	}
}

static void __mcheck_cpu_setup_timer(void)
{
	struct timer_list *t = this_cpu_ptr(&mce_timer);

	timer_setup(t, mce_timer_fn, TIMER_PINNED);
}

static void __mcheck_cpu_init_timer(void)
{
	struct timer_list *t = this_cpu_ptr(&mce_timer);

	timer_setup(t, mce_timer_fn, TIMER_PINNED);
	mce_start_timer(t);
}

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;
}

static __always_inline void exc_machine_check_kernel(struct pt_regs *regs)
{
	irqentry_state_t irq_state;

	WARN_ON_ONCE(user_mode(regs));

	/*
	 * Only required when from kernel mode. See
	 * mce_check_crashing_cpu() for details.
	 */
	if (mca_cfg.initialized && mce_check_crashing_cpu())
		return;

	irq_state = irqentry_nmi_enter(regs);

	do_machine_check(regs);

	irqentry_nmi_exit(regs, irq_state);
}

static __always_inline void exc_machine_check_user(struct pt_regs *regs)
{
	irqentry_enter_from_user_mode(regs);

	do_machine_check(regs);

	irqentry_exit_to_user_mode(regs);
}

#ifdef CONFIG_X86_64
/* MCE hit kernel mode */
DEFINE_IDTENTRY_MCE(exc_machine_check)
{
	unsigned long dr7;

	dr7 = local_db_save();
	exc_machine_check_kernel(regs);
	local_db_restore(dr7);
}

/* The user mode variant. */
DEFINE_IDTENTRY_MCE_USER(exc_machine_check)
{
	unsigned long dr7;

	dr7 = local_db_save();
	exc_machine_check_user(regs);
	local_db_restore(dr7);
}

#ifdef CONFIG_X86_FRED
/*
 * When occurred on different ring level, i.e., from user or kernel
 * context, #MCE needs to be handled on different stack: User #MCE
 * on current task stack, while kernel #MCE on a dedicated stack.
 *
 * This is exactly how FRED event delivery invokes an exception
 * handler: ring 3 event on level 0 stack, i.e., current task stack;
 * ring 0 event on the #MCE dedicated stack specified in the
 * IA32_FRED_STKLVLS MSR. So unlike IDT, the FRED machine check entry
 * stub doesn't do stack switch.
 */
DEFINE_FREDENTRY_MCE(exc_machine_check)
{
	unsigned long dr7;

	dr7 = local_db_save();
	if (user_mode(regs))
		exc_machine_check_user(regs);
	else
		exc_machine_check_kernel(regs);
	local_db_restore(dr7);
}
#endif
#else
/* 32bit unified entry point */
DEFINE_IDTENTRY_RAW(exc_machine_check)
{
	unsigned long dr7;

	dr7 = local_db_save();
	if (user_mode(regs))
		exc_machine_check_user(regs);
	else
		exc_machine_check_kernel(regs);
	local_db_restore(dr7);
}
#endif

void mca_bsp_init(struct cpuinfo_x86 *c)
{
	u64 cap;

	if (!mce_available(c))
		return;

	if (c->x86_vendor == X86_VENDOR_UNKNOWN) {
		mca_cfg.disabled = 1;
		pr_info("unknown CPU type - not enabling MCE support\n");
		return;
	}

	mce_flags.overflow_recov = cpu_feature_enabled(X86_FEATURE_OVERFLOW_RECOV);
	mce_flags.succor	 = cpu_feature_enabled(X86_FEATURE_SUCCOR);
	mce_flags.smca		 = cpu_feature_enabled(X86_FEATURE_SMCA);

	if (mce_flags.smca)
		smca_bsp_init();

	rdmsrq(MSR_IA32_MCG_CAP, cap);

	/* Use accurate RIP reporting if available. */
	if ((cap & MCG_EXT_P) && MCG_EXT_CNT(cap) >= 9)
		mca_cfg.rip_msr = MSR_IA32_MCG_EIP;

	if (cap & MCG_SER_P)
		mca_cfg.ser = 1;

	switch (c->x86_vendor) {
	case X86_VENDOR_AMD:
		amd_apply_global_quirks(c);
		break;
	case X86_VENDOR_INTEL:
		intel_apply_global_quirks(c);
		break;
	case X86_VENDOR_ZHAOXIN:
		zhaoxin_apply_global_quirks(c);
		break;
	}

	if (mca_cfg.monarch_timeout < 0)
		mca_cfg.monarch_timeout = 0;
	if (mca_cfg.bootlog != 0)
		mca_cfg.panic_timeout = 30;
}

/*
 * Called for each booted CPU to set up machine checks.
 * Must be called with preempt off:
 */
void mcheck_cpu_init(struct cpuinfo_x86 *c)
{
	if (mca_cfg.disabled)
		return;

	if (__mcheck_cpu_ancient_init(c))
		return;

	if (!mce_available(c))
		return;

	__mcheck_cpu_cap_init();

	if (!mce_gen_pool_init()) {
		mca_cfg.disabled = 1;
		pr_emerg("Couldn't allocate MCE records pool!\n");
		return;
	}

	mca_cfg.initialized = 1;

	__mcheck_cpu_init_generic();
	__mcheck_cpu_init_vendor(c);
	__mcheck_cpu_init_prepare_banks();
	__mcheck_cpu_setup_timer();
	cr4_set_bits(X86_CR4_MCE);
}

/*
 * Called for each booted CPU to clear some machine checks opt-ins
 */
void mcheck_cpu_clear(struct cpuinfo_x86 *c)
{
	if (mca_cfg.disabled)
		return;

	if (!mce_available(c))
		return;

	/*
	 * Possibly to clear general settings generic to x86
	 * __mcheck_cpu_clear_generic(c);
	 */
	__mcheck_cpu_clear_vendor(c);

}

static void __mce_disable_bank(void *arg)
{
	int bank = *((int *)arg);
	__clear_bit(bank, this_cpu_ptr(mce_poll_banks));
	cmci_disable_bank(bank);
}

void mce_disable_bank(int bank)
{
	if (bank >= this_cpu_read(mce_num_banks)) {
		pr_warn(FW_BUG
			"Ignoring request to disable invalid MCA bank %d.\n",
			bank);
		return;
	}
	set_bit(bank, mce_banks_ce_disabled);
	on_each_cpu(__mce_disable_bank, &bank, 1);
}

/*
 * mce=off Disables machine check
 * mce=no_cmci Disables CMCI
 * mce=no_lmce Disables LMCE
 * mce=dont_log_ce Clears corrected events silently, no log created for CEs.
 * mce=print_all Print all machine check logs to console
 * mce=ignore_ce Disables polling and CMCI, corrected events are not cleared.
 * mce=TOLERANCELEVEL[,monarchtimeout] (number, see above)
 *	monarchtimeout is how long to wait for other CPUs on machine
 *	check, or 0 to not wait
 * mce=bootlog Log MCEs from before booting. Disabled by default on AMD Fam10h
	and older.
 * mce=nobootlog Don't log MCEs from before booting.
 * mce=bios_cmci_threshold Don't program the CMCI threshold
 * mce=recovery force enable copy_mc_fragile()
 */
static int __init mcheck_enable(char *str)
{
	struct mca_config *cfg = &mca_cfg;

	if (*str == 0) {
		enable_p5_mce();
		return 1;
	}
	if (*str == '=')
		str++;
	if (!strcmp(str, "off"))
		cfg->disabled = 1;
	else if (!strcmp(str, "no_cmci"))
		cfg->cmci_disabled = true;
	else if (!strcmp(str, "no_lmce"))
		cfg->lmce_disabled = 1;
	else if (!strcmp(str, "dont_log_ce"))
		cfg->dont_log_ce = true;
	else if (!strcmp(str, "print_all"))
		cfg->print_all = true;
	else if (!strcmp(str, "ignore_ce"))
		cfg->ignore_ce = true;
	else if (!strcmp(str, "bootlog") || !strcmp(str, "nobootlog"))
		cfg->bootlog = (str[0] == 'b');
	else if (!strcmp(str, "bios_cmci_threshold"))
		cfg->bios_cmci_threshold = 1;
	else if (!strcmp(str, "recovery"))
		cfg->recovery = 1;
	else if (isdigit(str[0]))
		get_option(&str, &(cfg->monarch_timeout));
	else {
		pr_info("mce argument %s ignored. Please use /sys\n", str);
		return 0;
	}
	return 1;
}
__setup("mce", mcheck_enable);

int __init mcheck_init(void)
{
	mce_register_decode_chain(&early_nb);
	mce_register_decode_chain(&mce_uc_nb);
	mce_register_decode_chain(&mce_default_nb);

	INIT_WORK(&mce_work, mce_gen_pool_process);
	init_irq_work(&mce_irq_work, mce_irq_work_cb);

	return 0;
}

/*
 * mce_syscore: PM support
 */

/*
 * Disable machine checks on suspend and shutdown. We can't really handle
 * them later.
 */
static void mce_disable_error_reporting(void)
{
	struct mce_bank *mce_banks = this_cpu_ptr(mce_banks_array);
	int i;

	for (i = 0; i < this_cpu_read(mce_num_banks); i++) {
		struct mce_bank *b = &mce_banks[i];

		if (b->init)
			wrmsrq(mca_msr_reg(i, MCA_CTL), 0);
	}
	return;
}

static void vendor_disable_error_reporting(void)
{
	/*
	 * Don't clear on Intel or AMD or Hygon or Zhaoxin CPUs. Some of these
	 * MSRs are socket-wide. Disabling them for just a single offlined CPU
	 * is bad, since it will inhibit reporting for all shared resources on
	 * the socket like the last level cache (LLC), the integrated memory
	 * controller (iMC), etc.
	 */
	if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL ||
	    boot_cpu_data.x86_vendor == X86_VENDOR_HYGON ||
	    boot_cpu_data.x86_vendor == X86_VENDOR_AMD ||
	    boot_cpu_data.x86_vendor == X86_VENDOR_ZHAOXIN)
		return;

	mce_disable_error_reporting();
}

static int mce_syscore_suspend(void)
{
	vendor_disable_error_reporting();
	return 0;
}

static void mce_syscore_shutdown(void)
{
	vendor_disable_error_reporting();
}

/*
 * On resume clear all MCE state. Don't want to see leftovers from the BIOS.
 * Only one CPU is active at this time, the others get re-added later using
 * CPU hotplug:
 */
static void mce_syscore_resume(void)
{
	__mcheck_cpu_init_generic();
	__mcheck_cpu_init_vendor(raw_cpu_ptr(&cpu_info));
	__mcheck_cpu_init_prepare_banks();
	cr4_set_bits(X86_CR4_MCE);
}

static struct syscore_ops mce_syscore_ops = {
	.suspend	= mce_syscore_suspend,
	.shutdown	= mce_syscore_shutdown,
	.resume		= mce_syscore_resume,
};

/*
 * mce_device: Sysfs support
 */

static void mce_cpu_restart(void *data)
{
	if (!mce_available(raw_cpu_ptr(&cpu_info)))
		return;
	__mcheck_cpu_init_generic();
	__mcheck_cpu_init_prepare_banks();
	__mcheck_cpu_init_timer();
	cr4_set_bits(X86_CR4_MCE);
}

/* Reinit MCEs after user configuration changes */
static void mce_restart(void)
{
	mce_timer_delete_all();
	on_each_cpu(mce_cpu_restart, NULL, 1);
	mce_schedule_work();
}

/* Toggle features for corrected errors */
static void mce_disable_cmci(void *data)
{
	if (!mce_available(raw_cpu_ptr(&cpu_info)))
		return;
	cmci_clear();
}

static void mce_enable_ce(void *all)
{
	if (!mce_available(raw_cpu_ptr(&cpu_info)))
		return;
	cmci_reenable();
	cmci_recheck();
	if (all)
		__mcheck_cpu_init_timer();
}

static const struct bus_type mce_subsys = {
	.name		= "machinecheck",
	.dev_name	= "machinecheck",
};

DEFINE_PER_CPU(struct device *, mce_device);

static inline struct mce_bank_dev *attr_to_bank(struct device_attribute *attr)
{
	return container_of(attr, struct mce_bank_dev, attr);
}

static ssize_t show_bank(struct device *s, struct device_attribute *attr,
			 char *buf)
{
	u8 bank = attr_to_bank(attr)->bank;
	struct mce_bank *b;

	if (bank >= per_cpu(mce_num_banks, s->id))
		return -EINVAL;

	b = &per_cpu(mce_banks_array, s->id)[bank];

	if (!b->init)
		return -ENODEV;

	return sprintf(buf, "%llx\n", b->ctl);
}

static ssize_t set_bank(struct device *s, struct device_attribute *attr,
			const char *buf, size_t size)
{
	u8 bank = attr_to_bank(attr)->bank;
	struct mce_bank *b;
	u64 new;

	if (kstrtou64(buf, 0, &new) < 0)
		return -EINVAL;

	if (bank >= per_cpu(mce_num_banks, s->id))
		return -EINVAL;

	b = &per_cpu(mce_banks_array, s->id)[bank];
	if (!b->init)
		return -ENODEV;

	b->ctl = new;

	mutex_lock(&mce_sysfs_mutex);
	mce_restart();
	mutex_unlock(&mce_sysfs_mutex);

	return size;
}

static ssize_t set_ignore_ce(struct device *s,
			     struct device_attribute *attr,
			     const char *buf, size_t size)
{
	u64 new;

	if (kstrtou64(buf, 0, &new) < 0)
		return -EINVAL;

	mutex_lock(&mce_sysfs_mutex);
	if (mca_cfg.ignore_ce ^ !!new) {
		if (new) {
			/* disable ce features */
			mce_timer_delete_all();
			on_each_cpu(mce_disable_cmci, NULL, 1);
			mca_cfg.ignore_ce = true;
		} else {
			/* enable ce features */
			mca_cfg.ignore_ce = false;
			on_each_cpu(mce_enable_ce, (void *)1, 1);
		}
	}
	mutex_unlock(&mce_sysfs_mutex);

	return size;
}

static ssize_t set_cmci_disabled(struct device *s,
				 struct device_attribute *attr,
				 const char *buf, size_t size)
{
	u64 new;

	if (kstrtou64(buf, 0, &new) < 0)
		return -EINVAL;

	mutex_lock(&mce_sysfs_mutex);
	if (mca_cfg.cmci_disabled ^ !!new) {
		if (new) {
			/* disable cmci */
			on_each_cpu(mce_disable_cmci, NULL, 1);
			mca_cfg.cmci_disabled = true;
		} else {
			/* enable cmci */
			mca_cfg.cmci_disabled = false;
			on_each_cpu(mce_enable_ce, NULL, 1);
		}
	}
	mutex_unlock(&mce_sysfs_mutex);

	return size;
}

static ssize_t store_int_with_restart(struct device *s,
				      struct device_attribute *attr,
				      const char *buf, size_t size)
{
	unsigned long old_check_interval = check_interval;
	ssize_t ret = device_store_ulong(s, attr, buf, size);

	if (check_interval == old_check_interval)
		return ret;

	mutex_lock(&mce_sysfs_mutex);
	mce_restart();
	mutex_unlock(&mce_sysfs_mutex);

	return ret;
}

static DEVICE_INT_ATTR(monarch_timeout, 0644, mca_cfg.monarch_timeout);
static DEVICE_BOOL_ATTR(dont_log_ce, 0644, mca_cfg.dont_log_ce);
static DEVICE_BOOL_ATTR(print_all, 0644, mca_cfg.print_all);

static struct dev_ext_attribute dev_attr_check_interval = {
	__ATTR(check_interval, 0644, device_show_int, store_int_with_restart),
	&check_interval
};

static struct dev_ext_attribute dev_attr_ignore_ce = {
	__ATTR(ignore_ce, 0644, device_show_bool, set_ignore_ce),
	&mca_cfg.ignore_ce
};

static struct dev_ext_attribute dev_attr_cmci_disabled = {
	__ATTR(cmci_disabled, 0644, device_show_bool, set_cmci_disabled),
	&mca_cfg.cmci_disabled
};

static struct device_attribute *mce_device_attrs[] = {
	&dev_attr_check_interval.attr,
#ifdef CONFIG_X86_MCELOG_LEGACY
	&dev_attr_trigger,
#endif
	&dev_attr_monarch_timeout.attr,
	&dev_attr_dont_log_ce.attr,
	&dev_attr_print_all.attr,
	&dev_attr_ignore_ce.attr,
	&dev_attr_cmci_disabled.attr,
	NULL
};

static cpumask_var_t mce_device_initialized;

static void mce_device_release(struct device *dev)
{
	kfree(dev);
}

/* Per CPU device init. All of the CPUs still share the same bank device: */
static int mce_device_create(unsigned int cpu)
{
	struct device *dev;
	int err;
	int i, j;

	dev = per_cpu(mce_device, cpu);
	if (dev)
		return 0;

	dev = kzalloc(sizeof(*dev), GFP_KERNEL);
	if (!dev)
		return -ENOMEM;
	dev->id  = cpu;
	dev->bus = &mce_subsys;
	dev->release = &mce_device_release;

	err = device_register(dev);
	if (err) {
		put_device(dev);
		return err;
	}

	for (i = 0; mce_device_attrs[i]; i++) {
		err = device_create_file(dev, mce_device_attrs[i]);
		if (err)
			goto error;
	}
	for (j = 0; j < per_cpu(mce_num_banks, cpu); j++) {
		err = device_create_file(dev, &mce_bank_devs[j].attr);
		if (err)
			goto error2;
	}
	cpumask_set_cpu(cpu, mce_device_initialized);
	per_cpu(mce_device, cpu) = dev;

	return 0;
error2:
	while (--j >= 0)
		device_remove_file(dev, &mce_bank_devs[j].attr);
error:
	while (--i >= 0)
		device_remove_file(dev, mce_device_attrs[i]);

	device_unregister(dev);

	return err;
}

static void mce_device_remove(unsigned int cpu)
{
	struct device *dev = per_cpu(mce_device, cpu);
	int i;

	if (!cpumask_test_cpu(cpu, mce_device_initialized))
		return;

	for (i = 0; mce_device_attrs[i]; i++)
		device_remove_file(dev, mce_device_attrs[i]);

	for (i = 0; i < per_cpu(mce_num_banks, cpu); i++)
		device_remove_file(dev, &mce_bank_devs[i].attr);

	device_unregister(dev);
	cpumask_clear_cpu(cpu, mce_device_initialized);
	per_cpu(mce_device, cpu) = NULL;
}

/* Make sure there are no machine checks on offlined CPUs. */
static void mce_disable_cpu(void)
{
	if (!mce_available(raw_cpu_ptr(&cpu_info)))
		return;

	if (!cpuhp_tasks_frozen)
		cmci_clear();

	vendor_disable_error_reporting();
}

static void mce_reenable_cpu(void)
{
	struct mce_bank *mce_banks = this_cpu_ptr(mce_banks_array);
	int i;

	if (!mce_available(raw_cpu_ptr(&cpu_info)))
		return;

	if (!cpuhp_tasks_frozen)
		cmci_reenable();
	for (i = 0; i < this_cpu_read(mce_num_banks); i++) {
		struct mce_bank *b = &mce_banks[i];

		if (b->init)
			wrmsrq(mca_msr_reg(i, MCA_CTL), b->ctl);
	}
}

static int mce_cpu_dead(unsigned int cpu)
{
	/* intentionally ignoring frozen here */
	if (!cpuhp_tasks_frozen)
		cmci_rediscover();
	return 0;
}

static int mce_cpu_online(unsigned int cpu)
{
	struct timer_list *t = this_cpu_ptr(&mce_timer);

	mce_device_create(cpu);
	mce_threshold_create_device(cpu);
	mce_reenable_cpu();
	mce_start_timer(t);
	return 0;
}

static int mce_cpu_pre_down(unsigned int cpu)
{
	struct timer_list *t = this_cpu_ptr(&mce_timer);

	mce_disable_cpu();
	timer_delete_sync(t);
	mce_threshold_remove_device(cpu);
	mce_device_remove(cpu);
	return 0;
}

static __init void mce_init_banks(void)
{
	int i;

	for (i = 0; i < MAX_NR_BANKS; i++) {
		struct mce_bank_dev *b = &mce_bank_devs[i];
		struct device_attribute *a = &b->attr;

		b->bank = i;

		sysfs_attr_init(&a->attr);
		a->attr.name	= b->attrname;
		snprintf(b->attrname, ATTR_LEN, "bank%d", i);

		a->attr.mode	= 0644;
		a->show		= show_bank;
		a->store	= set_bank;
	}
}

/*
 * When running on XEN, this initcall is ordered against the XEN mcelog
 * initcall:
 *
 *   device_initcall(xen_late_init_mcelog);
 *   device_initcall_sync(mcheck_init_device);
 */
static __init int mcheck_init_device(void)
{
	int err;

	/*
	 * Check if we have a spare virtual bit. This will only become
	 * a problem if/when we move beyond 5-level page tables.
	 */
	MAYBE_BUILD_BUG_ON(__VIRTUAL_MASK_SHIFT >= 63);

	if (!mce_available(&boot_cpu_data)) {
		err = -EIO;
		goto err_out;
	}

	if (!zalloc_cpumask_var(&mce_device_initialized, GFP_KERNEL)) {
		err = -ENOMEM;
		goto err_out;
	}

	mce_init_banks();

	err = subsys_system_register(&mce_subsys, NULL);
	if (err)
		goto err_out_mem;

	err = cpuhp_setup_state(CPUHP_X86_MCE_DEAD, "x86/mce:dead", NULL,
				mce_cpu_dead);
	if (err)
		goto err_out_mem;

	/*
	 * Invokes mce_cpu_online() on all CPUs which are online when
	 * the state is installed.
	 */
	err = cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, "x86/mce:online",
				mce_cpu_online, mce_cpu_pre_down);
	if (err < 0)
		goto err_out_online;

	register_syscore_ops(&mce_syscore_ops);

	return 0;

err_out_online:
	cpuhp_remove_state(CPUHP_X86_MCE_DEAD);

err_out_mem:
	free_cpumask_var(mce_device_initialized);

err_out:
	pr_err("Unable to init MCE device (rc: %d)\n", err);

	return err;
}
device_initcall_sync(mcheck_init_device);

/*
 * Old style boot options parsing. Only for compatibility.
 */
static int __init mcheck_disable(char *str)
{
	mca_cfg.disabled = 1;
	return 1;
}
__setup("nomce", mcheck_disable);

#ifdef CONFIG_DEBUG_FS
struct dentry *mce_get_debugfs_dir(void)
{
	static struct dentry *dmce;

	if (!dmce)
		dmce = debugfs_create_dir("mce", NULL);

	return dmce;
}

static void mce_reset(void)
{
	atomic_set(&mce_fake_panicked, 0);
	atomic_set(&mce_executing, 0);
	atomic_set(&mce_callin, 0);
	atomic_set(&global_nwo, 0);
	cpumask_setall(&mce_missing_cpus);
}

static int fake_panic_get(void *data, u64 *val)
{
	*val = fake_panic;
	return 0;
}

static int fake_panic_set(void *data, u64 val)
{
	mce_reset();
	fake_panic = val;
	return 0;
}

DEFINE_DEBUGFS_ATTRIBUTE(fake_panic_fops, fake_panic_get, fake_panic_set,
			 "%llu\n");

static void __init mcheck_debugfs_init(void)
{
	struct dentry *dmce;

	dmce = mce_get_debugfs_dir();
	debugfs_create_file_unsafe("fake_panic", 0444, dmce, NULL,
				   &fake_panic_fops);
}
#else
static void __init mcheck_debugfs_init(void) { }
#endif

static int __init mcheck_late_init(void)
{
	if (mca_cfg.recovery)
		enable_copy_mc_fragile();

	mcheck_debugfs_init();

	/*
	 * Flush out everything that has been logged during early boot, now that
	 * everything has been initialized (workqueues, decoders, ...).
	 */
	mce_schedule_work();

	return 0;
}
late_initcall(mcheck_late_init);