--- zzzz-none-000/linux-3.10.107/arch/x86/kernel/cpu/mcheck/mce.c 2017-06-27 09:49:32.000000000 +0000 +++ scorpion-7490-727/linux-3.10.107/arch/x86/kernel/cpu/mcheck/mce.c 2021-02-04 17:41:59.000000000 +0000 @@ -43,6 +43,8 @@ #include #include +#include +#include #include #include @@ -50,21 +52,23 @@ static DEFINE_MUTEX(mce_chrdev_read_mutex); -#define rcu_dereference_check_mce(p) \ - rcu_dereference_index_check((p), \ - rcu_read_lock_sched_held() || \ - lockdep_is_held(&mce_chrdev_read_mutex)) +#define mce_log_get_idx_check(p) \ +({ \ + RCU_LOCKDEP_WARN(!rcu_read_lock_sched_held() && \ + !lockdep_is_held(&mce_chrdev_read_mutex), \ + "suspicious mce_log_get_idx_check() usage"); \ + smp_load_acquire(&(p)); \ +}) #define CREATE_TRACE_POINTS #include -#define SPINUNIT 100 /* 100ns */ - -atomic_t mce_entry; +#define SPINUNIT 100 /* 100ns */ DEFINE_PER_CPU(unsigned, mce_exception_count); struct mce_bank *mce_banks __read_mostly; +struct mce_vendor_flags mce_flags __read_mostly; struct mca_config mca_cfg __read_mostly = { .bootlog = -1, @@ -89,14 +93,28 @@ static DEFINE_PER_CPU(struct mce, mces_seen); static int cpu_missing; -/* MCA banks polled by the period polling timer for corrected events */ +/* + * 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 }; -static DEFINE_PER_CPU(struct work_struct, mce_work); +/* + * 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; static void (*quirk_no_way_out)(int bank, struct mce *m, struct pt_regs *regs); +static int mce_usable_address(struct mce *m); /* * CPU/chipset specific EDAC code can register a notifier call here to print @@ -109,7 +127,7 @@ { memset(m, 0, sizeof(struct mce)); m->cpu = m->extcpu = smp_processor_id(); - rdtscll(m->tsc); + m->tsc = rdtsc(); /* We hope get_seconds stays lockless */ m->time = get_seconds(); m->cpuvendor = boot_cpu_data.x86_vendor; @@ -137,19 +155,17 @@ void mce_log(struct mce *mce) { unsigned next, entry; - int ret = 0; /* Emit the trace record: */ trace_mce_record(mce); - ret = atomic_notifier_call_chain(&x86_mce_decoder_chain, 0, mce); - if (ret == NOTIFY_STOP) - return; + if (!mce_gen_pool_add(mce)) + irq_work_queue(&mce_irq_work); mce->finished = 0; wmb(); for (;;) { - entry = rcu_dereference_check_mce(mcelog.next); + entry = mce_log_get_idx_check(mcelog.next); for (;;) { /* @@ -183,48 +199,23 @@ set_bit(0, &mce_need_notify); } -static void drain_mcelog_buffer(void) +void mce_inject_log(struct mce *m) { - unsigned int next, i, prev = 0; - - next = ACCESS_ONCE(mcelog.next); - - do { - struct mce *m; - - /* drain what was logged during boot */ - for (i = prev; i < next; i++) { - unsigned long start = jiffies; - unsigned retries = 1; - - m = &mcelog.entry[i]; - - while (!m->finished) { - if (time_after_eq(jiffies, start + 2*retries)) - retries++; - - cpu_relax(); - - if (!m->finished && retries >= 4) { - pr_err("skipping error being logged currently!\n"); - break; - } - } - smp_rmb(); - atomic_notifier_call_chain(&x86_mce_decoder_chain, 0, m); - } - - memset(mcelog.entry + prev, 0, (next - prev) * sizeof(*m)); - prev = next; - next = cmpxchg(&mcelog.next, prev, 0); - } while (next != prev); + mutex_lock(&mce_chrdev_read_mutex); + mce_log(m); + mutex_unlock(&mce_chrdev_read_mutex); } +EXPORT_SYMBOL_GPL(mce_inject_log); +static struct notifier_block mce_srao_nb; void mce_register_decode_chain(struct notifier_block *nb) { + /* Ensure SRAO notifier has the highest priority in the decode chain. */ + if (nb != &mce_srao_nb && nb->priority == INT_MAX) + nb->priority -= 1; + atomic_notifier_chain_register(&x86_mce_decoder_chain, nb); - drain_mcelog_buffer(); } EXPORT_SYMBOL_GPL(mce_register_decode_chain); @@ -279,10 +270,10 @@ #define PANIC_TIMEOUT 5 /* 5 seconds */ -static atomic_t mce_paniced; +static atomic_t mce_panicked; static int fake_panic; -static atomic_t mce_fake_paniced; +static atomic_t mce_fake_panicked; /* Panic in progress. Enable interrupts and wait for final IPI */ static void wait_for_panic(void) @@ -298,7 +289,7 @@ panic("Panicing machine check CPU died"); } -static void mce_panic(char *msg, struct mce *final, char *exp) +static void mce_panic(const char *msg, struct mce *final, char *exp) { int i, apei_err = 0; @@ -306,7 +297,7 @@ /* * Make sure only one CPU runs in machine check panic */ - if (atomic_inc_return(&mce_paniced) > 1) + if (atomic_inc_return(&mce_panicked) > 1) wait_for_panic(); barrier(); @@ -314,7 +305,7 @@ console_verbose(); } else { /* Don't log too much for fake panic */ - if (atomic_inc_return(&mce_fake_paniced) > 1) + if (atomic_inc_return(&mce_fake_panicked) > 1) return; } /* First print corrected ones that are still unlogged */ @@ -387,7 +378,7 @@ if (offset < 0) return 0; - return *(u64 *)((char *)&__get_cpu_var(injectm) + offset); + return *(u64 *)((char *)this_cpu_ptr(&injectm) + offset); } if (rdmsrl_safe(msr, &v)) { @@ -409,7 +400,7 @@ int offset = msr_to_offset(msr); if (offset >= 0) - *(u64 *)((char *)&__get_cpu_var(injectm) + offset) = v; + *(u64 *)((char *)this_cpu_ptr(&injectm) + offset) = v; return; } wrmsrl(msr, v); @@ -448,61 +439,6 @@ } } -/* - * Simple lockless ring to communicate PFNs from the exception handler with the - * process context work function. This is vastly simplified because there's - * only a single reader and a single writer. - */ -#define MCE_RING_SIZE 16 /* we use one entry less */ - -struct mce_ring { - unsigned short start; - unsigned short end; - unsigned long ring[MCE_RING_SIZE]; -}; -static DEFINE_PER_CPU(struct mce_ring, mce_ring); - -/* Runs with CPU affinity in workqueue */ -static int mce_ring_empty(void) -{ - struct mce_ring *r = &__get_cpu_var(mce_ring); - - return r->start == r->end; -} - -static int mce_ring_get(unsigned long *pfn) -{ - struct mce_ring *r; - int ret = 0; - - *pfn = 0; - get_cpu(); - r = &__get_cpu_var(mce_ring); - if (r->start == r->end) - goto out; - *pfn = r->ring[r->start]; - r->start = (r->start + 1) % MCE_RING_SIZE; - ret = 1; -out: - put_cpu(); - return ret; -} - -/* Always runs in MCE context with preempt off */ -static int mce_ring_add(unsigned long pfn) -{ - struct mce_ring *r = &__get_cpu_var(mce_ring); - unsigned next; - - next = (r->end + 1) % MCE_RING_SIZE; - if (next == r->start) - return -1; - r->ring[r->end] = pfn; - wmb(); - r->end = next; - return 0; -} - int mce_available(struct cpuinfo_x86 *c) { if (mca_cfg.disabled) @@ -512,12 +448,10 @@ static void mce_schedule_work(void) { - if (!mce_ring_empty()) - schedule_work(&__get_cpu_var(mce_work)); + if (!mce_gen_pool_empty() && keventd_up()) + schedule_work(&mce_work); } -DEFINE_PER_CPU(struct irq_work, mce_irq_work); - static void mce_irq_work_cb(struct irq_work *entry) { mce_notify_irq(); @@ -538,9 +472,30 @@ return; } - irq_work_queue(&__get_cpu_var(mce_irq_work)); + irq_work_queue(&mce_irq_work); } +static int srao_decode_notifier(struct notifier_block *nb, unsigned long val, + void *data) +{ + struct mce *mce = (struct mce *)data; + unsigned long pfn; + + if (!mce) + return NOTIFY_DONE; + + if (mce->usable_addr && (mce->severity == MCE_AO_SEVERITY)) { + pfn = mce->addr >> PAGE_SHIFT; + memory_failure(pfn, MCE_VECTOR, 0); + } + + return NOTIFY_OK; +} +static struct notifier_block mce_srao_nb = { + .notifier_call = srao_decode_notifier, + .priority = INT_MAX, +}; + /* * Read ADDR and MISC registers. */ @@ -562,6 +517,37 @@ } } +static bool memory_error(struct mce *m) +{ + struct cpuinfo_x86 *c = &boot_cpu_data; + + if (c->x86_vendor == X86_VENDOR_AMD) { + /* + * coming soon + */ + return false; + } else if (c->x86_vendor == X86_VENDOR_INTEL) { + /* + * 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; + } + + return false; +} + DEFINE_PER_CPU(unsigned, mce_poll_count); /* @@ -579,9 +565,11 @@ * 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) +bool machine_check_poll(enum mcp_flags flags, mce_banks_t *b) { + bool error_logged = false; struct mce m; + int severity; int i; this_cpu_inc(mce_poll_count); @@ -602,6 +590,7 @@ if (!(m.status & MCI_STATUS_VAL)) continue; + /* * Uncorrected or signalled events are handled by the exception * handler when it is enabled, so don't process those here. @@ -616,12 +605,31 @@ if (!(flags & MCP_TIMESTAMP)) m.tsc = 0; + + severity = mce_severity(&m, mca_cfg.tolerant, NULL, false); + + /* + * In the cases where we don't have a valid address after all, + * do not add it into the ring buffer. + */ + if (severity == MCE_DEFERRED_SEVERITY && memory_error(&m)) { + if (m.status & MCI_STATUS_ADDRV) { + m.severity = severity; + m.usable_addr = mce_usable_address(&m); + + if (!mce_gen_pool_add(&m)) + mce_schedule_work(); + } + } + /* * Don't get the IP here because it's unlikely to * have anything to do with the actual error location. */ - if (!(flags & MCP_DONTLOG) && !mca_cfg.dont_log_ce) + if (!(flags & MCP_DONTLOG) && !mca_cfg.dont_log_ce) { + error_logged = true; mce_log(&m); + } /* * Clear state for this bank. @@ -635,6 +643,8 @@ */ sync_core(); + + return error_logged; } EXPORT_SYMBOL_GPL(machine_check_poll); @@ -646,6 +656,7 @@ struct pt_regs *regs) { int i, ret = 0; + char *tmp; for (i = 0; i < mca_cfg.banks; i++) { m->status = mce_rdmsrl(MSR_IA32_MCx_STATUS(i)); @@ -654,8 +665,11 @@ if (quirk_no_way_out) quirk_no_way_out(i, m, regs); } - if (mce_severity(m, mca_cfg.tolerant, msg) >= MCE_PANIC_SEVERITY) + + if (mce_severity(m, mca_cfg.tolerant, &tmp, true) >= MCE_PANIC_SEVERITY) { + *msg = tmp; ret = 1; + } } return ret; } @@ -674,7 +688,7 @@ /* * Check if a timeout waiting for other CPUs happened. */ -static int mce_timed_out(u64 *t) +static int mce_timed_out(u64 *t, const char *msg) { /* * The others already did panic for some reason. @@ -683,15 +697,13 @@ * might have been modified by someone else. */ rmb(); - if (atomic_read(&mce_paniced)) + if (atomic_read(&mce_panicked)) wait_for_panic(); if (!mca_cfg.monarch_timeout) goto out; if ((s64)*t < SPINUNIT) { - /* CHECKME: Make panic default for 1 too? */ - if (mca_cfg.tolerant < 1) - mce_panic("Timeout synchronizing machine check over CPUs", - NULL, NULL); + if (mca_cfg.tolerant <= 1) + mce_panic(msg, NULL, NULL); cpu_missing = 1; return 1; } @@ -741,7 +753,7 @@ for_each_possible_cpu(cpu) { int severity = mce_severity(&per_cpu(mces_seen, cpu), mca_cfg.tolerant, - &nmsg); + &nmsg, true); if (severity > global_worst) { msg = nmsg; global_worst = severity; @@ -755,7 +767,7 @@ * other CPUs. */ if (m && global_worst >= MCE_PANIC_SEVERITY && mca_cfg.tolerant < 3) - mce_panic("Fatal Machine check", m, msg); + mce_panic("Fatal machine check", m, msg); /* * For UC somewhere we let the CPU who detects it handle it. @@ -768,7 +780,7 @@ * source or one CPU is hung. Panic. */ if (global_worst <= MCE_KEEP_SEVERITY && mca_cfg.tolerant < 3) - mce_panic("Machine check from unknown source", NULL, NULL); + mce_panic("Fatal machine check from unknown source", NULL, NULL); /* * Now clear all the mces_seen so that they don't reappear on @@ -807,7 +819,8 @@ * Wait for everyone. */ while (atomic_read(&mce_callin) != cpus) { - if (mce_timed_out(&timeout)) { + if (mce_timed_out(&timeout, + "Timeout: Not all CPUs entered broadcast exception handler")) { atomic_set(&global_nwo, 0); return -1; } @@ -832,7 +845,8 @@ * only seen by one CPU before cleared, avoiding duplicates. */ while (atomic_read(&mce_executing) < order) { - if (mce_timed_out(&timeout)) { + if (mce_timed_out(&timeout, + "Timeout: Subject CPUs unable to finish machine check processing")) { atomic_set(&global_nwo, 0); return -1; } @@ -876,7 +890,8 @@ * loops. */ while (atomic_read(&mce_executing) <= cpus) { - if (mce_timed_out(&timeout)) + if (mce_timed_out(&timeout, + "Timeout: Monarch CPU unable to finish machine check processing")) goto reset; ndelay(SPINUNIT); } @@ -889,7 +904,8 @@ * Subject: Wait for Monarch to finish. */ while (atomic_read(&mce_executing) != 0) { - if (mce_timed_out(&timeout)) + if (mce_timed_out(&timeout, + "Timeout: Monarch CPU did not finish machine check processing")) goto reset; ndelay(SPINUNIT); } @@ -943,51 +959,6 @@ } /* - * Need to save faulting physical address associated with a process - * in the machine check handler some place where we can grab it back - * later in mce_notify_process() - */ -#define MCE_INFO_MAX 16 - -struct mce_info { - atomic_t inuse; - struct task_struct *t; - __u64 paddr; - int restartable; -} mce_info[MCE_INFO_MAX]; - -static void mce_save_info(__u64 addr, int c) -{ - struct mce_info *mi; - - for (mi = mce_info; mi < &mce_info[MCE_INFO_MAX]; mi++) { - if (atomic_cmpxchg(&mi->inuse, 0, 1) == 0) { - mi->t = current; - mi->paddr = addr; - mi->restartable = c; - return; - } - } - - mce_panic("Too many concurrent recoverable errors", NULL, NULL); -} - -static struct mce_info *mce_find_info(void) -{ - struct mce_info *mi; - - for (mi = mce_info; mi < &mce_info[MCE_INFO_MAX]; mi++) - if (atomic_read(&mi->inuse) && mi->t == current) - return mi; - return NULL; -} - -static void mce_clear_info(struct mce_info *mi) -{ - atomic_set(&mi->inuse, 0); -} - -/* * The actual machine check handler. This only handles real * exceptions when something got corrupted coming in through int 18. * @@ -1024,8 +995,22 @@ DECLARE_BITMAP(toclear, MAX_NR_BANKS); DECLARE_BITMAP(valid_banks, MAX_NR_BANKS); char *msg = "Unknown"; + u64 recover_paddr = ~0ull; + int flags = MF_ACTION_REQUIRED; + int lmce = 0; - atomic_inc(&mce_entry); + /* If this CPU is offline, just bail out. */ + if (cpu_is_offline(smp_processor_id())) { + u64 mcgstatus; + + mcgstatus = mce_rdmsrl(MSR_IA32_MCG_STATUS); + if (mcgstatus & MCG_STATUS_RIPV) { + mce_wrmsrl(MSR_IA32_MCG_STATUS, 0); + return; + } + } + + ist_enter(regs); this_cpu_inc(mce_exception_count); @@ -1034,7 +1019,7 @@ mce_gather_info(&m, regs); - final = &__get_cpu_var(mces_seen); + final = this_cpu_ptr(&mces_seen); *final = m; memset(valid_banks, 0, sizeof(valid_banks)); @@ -1051,11 +1036,20 @@ kill_it = 1; /* - * Go through all the 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. + * Check if this MCE is signaled to only this logical processor */ - order = mce_start(&no_way_out); + if (m.mcgstatus & MCG_STATUS_LMCES) + lmce = 1; + else { + /* + * Go through all the 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 this is a Local MCE, then no need to perform rendezvous. + */ + order = mce_start(&no_way_out); + } + for (i = 0; i < cfg->banks; i++) { __clear_bit(i, toclear); if (!test_bit(i, valid_banks)) @@ -1084,13 +1078,14 @@ */ add_taint(TAINT_MACHINE_CHECK, LOCKDEP_NOW_UNRELIABLE); - severity = mce_severity(&m, cfg->tolerant, NULL); + severity = mce_severity(&m, cfg->tolerant, NULL, true); /* - * When machine check was for corrected handler don't touch, - * unless we're panicing. + * When machine check was for corrected/deferred handler don't + * touch, unless we're panicing. */ - if (severity == MCE_KEEP_SEVERITY && !no_way_out) + if ((severity == MCE_KEEP_SEVERITY || + severity == MCE_UCNA_SEVERITY) && !no_way_out) continue; __set_bit(i, toclear); if (severity == MCE_NO_SEVERITY) { @@ -1103,15 +1098,9 @@ mce_read_aux(&m, i); - /* - * Action optional error. Queue address for later processing. - * When the ring overflows we just ignore the AO error. - * RED-PEN add some logging mechanism when - * usable_address or mce_add_ring fails. - * RED-PEN don't ignore overflow for mca_cfg.tolerant == 0 - */ - if (severity == MCE_AO_SEVERITY && mce_usable_address(&m)) - mce_ring_add(m.addr >> PAGE_SHIFT); + /* assuming valid severity level != 0 */ + m.severity = severity; + m.usable_addr = mce_usable_address(&m); mce_log(&m); @@ -1131,8 +1120,18 @@ * Do most of the synchronization with other CPUs. * When there's any problem use only local no_way_out state. */ - if (mce_end(order) < 0) - no_way_out = worst >= MCE_PANIC_SEVERITY; + if (!lmce) { + if (mce_end(order) < 0) + no_way_out = worst >= MCE_PANIC_SEVERITY; + } else { + /* + * Local MCE skipped calling mce_reign() + * If we found a fatal error, we need to panic here. + */ + if (worst >= MCE_PANIC_SEVERITY && mca_cfg.tolerant < 3) + mce_panic("Machine check from unknown source", + NULL, NULL); + } /* * At insane "tolerant" levels we take no action. Otherwise @@ -1144,9 +1143,9 @@ if (no_way_out) mce_panic("Fatal machine check on current CPU", &m, msg); if (worst == MCE_AR_SEVERITY) { - /* schedule action before return to userland */ - mce_save_info(m.addr, m.mcgstatus & MCG_STATUS_RIPV); - set_thread_flag(TIF_MCE_NOTIFY); + recover_paddr = m.addr; + if (!(m.mcgstatus & MCG_STATUS_RIPV)) + flags |= MF_MUST_KILL; } else if (kill_it) { force_sig(SIGBUS, current); } @@ -1156,8 +1155,28 @@ mce_report_event(regs); mce_wrmsrl(MSR_IA32_MCG_STATUS, 0); out: - atomic_dec(&mce_entry); sync_core(); + + if (recover_paddr == ~0ull) + goto done; + + pr_err("Uncorrected hardware memory error in user-access at %llx", + recover_paddr); + /* + * We must call memory_failure() here even if the current process is + * doomed. We still need to mark the page as poisoned and alert any + * other users of the page. + */ + ist_begin_non_atomic(regs); + local_irq_enable(); + if (memory_failure(recover_paddr >> PAGE_SHIFT, MCE_VECTOR, flags) < 0) { + pr_err("Memory error not recovered"); + force_sig(SIGBUS, current); + } + local_irq_disable(); + ist_end_non_atomic(); +done: + ist_exit(regs); } EXPORT_SYMBOL_GPL(do_machine_check); @@ -1175,52 +1194,13 @@ #endif /* - * Called in process context that interrupted by MCE and marked with - * TIF_MCE_NOTIFY, just before returning to erroneous userland. - * This code is allowed to sleep. - * Attempt possible recovery such as calling the high level VM handler to - * process any corrupted pages, and kill/signal current process if required. - * Action required errors are handled here. - */ -void mce_notify_process(void) -{ - unsigned long pfn; - struct mce_info *mi = mce_find_info(); - int flags = MF_ACTION_REQUIRED; - - if (!mi) - mce_panic("Lost physical address for unconsumed uncorrectable error", NULL, NULL); - pfn = mi->paddr >> PAGE_SHIFT; - - clear_thread_flag(TIF_MCE_NOTIFY); - - pr_err("Uncorrected hardware memory error in user-access at %llx", - mi->paddr); - /* - * We must call memory_failure() here even if the current process is - * doomed. We still need to mark the page as poisoned and alert any - * other users of the page. - */ - if (!mi->restartable) - flags |= MF_MUST_KILL; - if (memory_failure(pfn, MCE_VECTOR, flags) < 0) { - pr_err("Memory error not recovered"); - force_sig(SIGBUS, current); - } - mce_clear_info(mi); -} - -/* * Action optional processing happens here (picking up * from the list of faulting pages that do_machine_check() - * placed into the "ring"). + * placed into the genpool). */ static void mce_process_work(struct work_struct *dummy) { - unsigned long pfn; - - while (mce_ring_get(&pfn)) - memory_failure(pfn, MCE_VECTOR, 0); + mce_gen_pool_process(); } #ifdef CONFIG_X86_MCE_INTEL @@ -1253,7 +1233,7 @@ * 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 = 5 * 60; /* 5 minutes */ +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); @@ -1263,39 +1243,57 @@ return interval; } -static unsigned long (*mce_adjust_timer)(unsigned long interval) = - mce_adjust_timer_default; +static unsigned long (*mce_adjust_timer)(unsigned long interval) = mce_adjust_timer_default; + +static void __restart_timer(struct timer_list *t, unsigned long interval) +{ + unsigned long when = jiffies + interval; + unsigned long flags; + + local_irq_save(flags); + + if (timer_pending(t)) { + if (time_before(when, t->expires)) + mod_timer_pinned(t, when); + } else { + t->expires = round_jiffies(when); + add_timer_on(t, smp_processor_id()); + } + + local_irq_restore(flags); +} static void mce_timer_fn(unsigned long data) { - struct timer_list *t = &__get_cpu_var(mce_timer); + struct timer_list *t = this_cpu_ptr(&mce_timer); + int cpu = smp_processor_id(); unsigned long iv; - WARN_ON(smp_processor_id() != data); + WARN_ON(cpu != data); - if (mce_available(__this_cpu_ptr(&cpu_info))) { - machine_check_poll(MCP_TIMESTAMP, - &__get_cpu_var(mce_poll_banks)); - mce_intel_cmci_poll(); + iv = __this_cpu_read(mce_next_interval); + + if (mce_available(this_cpu_ptr(&cpu_info))) { + machine_check_poll(MCP_TIMESTAMP, this_cpu_ptr(&mce_poll_banks)); + + if (mce_intel_cmci_poll()) { + iv = mce_adjust_timer(iv); + goto done; + } } /* - * Alert userspace if needed. If we logged an MCE, reduce the - * polling interval, otherwise increase the polling interval. + * Alert userspace if needed. If we logged an MCE, reduce the polling + * interval, otherwise increase the polling interval. */ - iv = __this_cpu_read(mce_next_interval); - if (mce_notify_irq()) { + if (mce_notify_irq()) iv = max(iv / 2, (unsigned long) HZ/100); - } else { + else iv = min(iv * 2, round_jiffies_relative(check_interval * HZ)); - iv = mce_adjust_timer(iv); - } + +done: __this_cpu_write(mce_next_interval, iv); - /* Might have become 0 after CMCI storm subsided */ - if (iv) { - t->expires = jiffies + iv; - add_timer_on(t, smp_processor_id()); - } + __restart_timer(t, iv); } /* @@ -1303,17 +1301,11 @@ */ void mce_timer_kick(unsigned long interval) { - struct timer_list *t = &__get_cpu_var(mce_timer); - unsigned long when = jiffies + interval; + struct timer_list *t = this_cpu_ptr(&mce_timer); unsigned long iv = __this_cpu_read(mce_next_interval); - if (timer_pending(t)) { - if (time_before(when, t->expires)) - mod_timer_pinned(t, when); - } else { - t->expires = round_jiffies(when); - add_timer_on(t, smp_processor_id()); - } + __restart_timer(t, interval); + if (interval < iv) __this_cpu_write(mce_next_interval, interval); } @@ -1360,7 +1352,7 @@ } EXPORT_SYMBOL_GPL(mce_notify_irq); -static int __cpuinit __mcheck_cpu_mce_banks_init(void) +static int __mcheck_cpu_mce_banks_init(void) { int i; u8 num_banks = mca_cfg.banks; @@ -1381,7 +1373,7 @@ /* * Initialize Machine Checks for a CPU. */ -static int __cpuinit __mcheck_cpu_cap_init(void) +static int __mcheck_cpu_cap_init(void) { unsigned b; u64 cap; @@ -1435,7 +1427,7 @@ bitmap_fill(all_banks, MAX_NR_BANKS); machine_check_poll(MCP_UC | m_fl, &all_banks); - set_in_cr4(X86_CR4_MCE); + cr4_set_bits(X86_CR4_MCE); rdmsrl(MSR_IA32_MCG_CAP, cap); if (cap & MCG_CTL_P) @@ -1480,7 +1472,7 @@ } /* Add per CPU specific workarounds here */ -static int __cpuinit __mcheck_cpu_apply_quirks(struct cpuinfo_x86 *c) +static int __mcheck_cpu_apply_quirks(struct cpuinfo_x86 *c) { struct mca_config *cfg = &mca_cfg; @@ -1510,45 +1502,46 @@ * Various K7s with broken bank 0 around. Always disable * by default. */ - if (c->x86 == 6 && cfg->banks > 0) + if (c->x86 == 6 && cfg->banks > 0) mce_banks[0].ctl = 0; - /* - * Turn off MC4_MISC thresholding banks on those models since - * they're not supported there. - */ - if (c->x86 == 0x15 && - (c->x86_model >= 0x10 && c->x86_model <= 0x1f)) { - int i; - u64 val, hwcr; - bool need_toggle; - u32 msrs[] = { + /* + * 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; + + /* + * Turn off MC4_MISC thresholding banks on those models since + * they're not supported there. + */ + if (c->x86 == 0x15 && + (c->x86_model >= 0x10 && c->x86_model <= 0x1f)) { + int i; + u64 hwcr; + bool need_toggle; + u32 msrs[] = { 0x00000413, /* MC4_MISC0 */ 0xc0000408, /* MC4_MISC1 */ - }; + }; + + rdmsrl(MSR_K7_HWCR, hwcr); + + /* McStatusWrEn has to be set */ + need_toggle = !(hwcr & BIT(18)); - rdmsrl(MSR_K7_HWCR, hwcr); + if (need_toggle) + wrmsrl(MSR_K7_HWCR, hwcr | BIT(18)); - /* McStatusWrEn has to be set */ - need_toggle = !(hwcr & BIT(18)); + /* Clear CntP bit safely */ + for (i = 0; i < ARRAY_SIZE(msrs); i++) + msr_clear_bit(msrs[i], 62); - if (need_toggle) - wrmsrl(MSR_K7_HWCR, hwcr | BIT(18)); - - for (i = 0; i < ARRAY_SIZE(msrs); i++) { - rdmsrl(msrs[i], val); - - /* CntP bit set? */ - if (val & BIT_64(62)) { - val &= ~BIT_64(62); - wrmsrl(msrs[i], val); - } - } - - /* restore old settings */ - if (need_toggle) - wrmsrl(MSR_K7_HWCR, hwcr); - } + /* restore old settings */ + if (need_toggle) + wrmsrl(MSR_K7_HWCR, hwcr); + } } if (c->x86_vendor == X86_VENDOR_INTEL) { @@ -1590,7 +1583,7 @@ return 0; } -static int __cpuinit __mcheck_cpu_ancient_init(struct cpuinfo_x86 *c) +static int __mcheck_cpu_ancient_init(struct cpuinfo_x86 *c) { if (c->x86 != 5) return 0; @@ -1604,6 +1597,8 @@ winchip_mcheck_init(c); return 1; break; + default: + return 0; } return 0; @@ -1614,32 +1609,52 @@ switch (c->x86_vendor) { case X86_VENDOR_INTEL: mce_intel_feature_init(c); - mce_adjust_timer = mce_intel_adjust_timer; + mce_adjust_timer = cmci_intel_adjust_timer; break; - case X86_VENDOR_AMD: + + case X86_VENDOR_AMD: { + u32 ebx = cpuid_ebx(0x80000007); + mce_amd_feature_init(c); + mce_flags.overflow_recov = !!(ebx & BIT(0)); + mce_flags.succor = !!(ebx & BIT(1)); + mce_flags.smca = !!(ebx & BIT(3)); + break; + } + default: break; } } -static void mce_start_timer(unsigned int cpu, struct timer_list *t) +static void __mcheck_cpu_clear_vendor(struct cpuinfo_x86 *c) { - unsigned long iv = mce_adjust_timer(check_interval * HZ); + switch (c->x86_vendor) { + case X86_VENDOR_INTEL: + mce_intel_feature_clear(c); + break; + default: + break; + } +} - __this_cpu_write(mce_next_interval, iv); +static void mce_start_timer(unsigned int cpu, struct timer_list *t) +{ + unsigned long iv = check_interval * HZ; if (mca_cfg.ignore_ce || !iv) return; + per_cpu(mce_next_interval, cpu) = iv; + t->expires = round_jiffies(jiffies + iv); - add_timer_on(t, smp_processor_id()); + add_timer_on(t, cpu); } static void __mcheck_cpu_init_timer(void) { - struct timer_list *t = &__get_cpu_var(mce_timer); + struct timer_list *t = this_cpu_ptr(&mce_timer); unsigned int cpu = smp_processor_id(); setup_timer(t, mce_timer_fn, cpu); @@ -1661,7 +1676,7 @@ * Called for each booted CPU to set up machine checks. * Must be called with preempt off: */ -void __cpuinit mcheck_cpu_init(struct cpuinfo_x86 *c) +void mcheck_cpu_init(struct cpuinfo_x86 *c) { if (mca_cfg.disabled) return; @@ -1677,13 +1692,36 @@ return; } + if (mce_gen_pool_init()) { + mca_cfg.disabled = true; + pr_emerg("Couldn't allocate MCE records pool!\n"); + return; + } + machine_check_vector = do_machine_check; __mcheck_cpu_init_generic(); __mcheck_cpu_init_vendor(c); __mcheck_cpu_init_timer(); - INIT_WORK(&__get_cpu_var(mce_work), mce_process_work); - init_irq_work(&__get_cpu_var(mce_irq_work), &mce_irq_work_cb); +} + +/* + * 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); + } /* @@ -1730,7 +1768,7 @@ { unsigned long *cpu_tsc = (unsigned long *)data; - rdtscll(cpu_tsc[smp_processor_id()]); + cpu_tsc[smp_processor_id()] = rdtsc(); } static int mce_apei_read_done; @@ -1796,7 +1834,7 @@ goto out; } - next = rcu_dereference_check_mce(mcelog.next); + next = mce_log_get_idx_check(mcelog.next); /* Only supports full reads right now */ err = -EINVAL; @@ -1862,7 +1900,7 @@ static unsigned int mce_chrdev_poll(struct file *file, poll_table *wait) { poll_wait(file, &mce_chrdev_wait, wait); - if (rcu_access_index(mcelog.next)) + if (READ_ONCE(mcelog.next)) return POLLIN | POLLRDNORM; if (!mce_apei_read_done && apei_check_mce()) return POLLIN | POLLRDNORM; @@ -1907,8 +1945,8 @@ } EXPORT_SYMBOL_GPL(register_mce_write_callback); -ssize_t mce_chrdev_write(struct file *filp, const char __user *ubuf, - size_t usize, loff_t *off) +static ssize_t mce_chrdev_write(struct file *filp, const char __user *ubuf, + size_t usize, loff_t *off) { if (mce_write) return mce_write(filp, ubuf, usize, off); @@ -1932,9 +1970,29 @@ &mce_chrdev_ops, }; +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 >= mca_cfg.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=ignore_ce Disables polling and CMCI, corrected events are not cleared. * mce=TOLERANCELEVEL[,monarchtimeout] (number, see above) @@ -1958,6 +2016,8 @@ cfg->disabled = true; else if (!strcmp(str, "no_cmci")) cfg->cmci_disabled = true; + else if (!strcmp(str, "no_lmce")) + cfg->lmce_disabled = true; else if (!strcmp(str, "dont_log_ce")) cfg->dont_log_ce = true; else if (!strcmp(str, "ignore_ce")) @@ -1967,11 +2027,8 @@ else if (!strcmp(str, "bios_cmci_threshold")) cfg->bios_cmci_threshold = true; else if (isdigit(str[0])) { - get_option(&str, &(cfg->tolerant)); - if (*str == ',') { - ++str; + if (get_option(&str, &cfg->tolerant) == 2) get_option(&str, &(cfg->monarch_timeout)); - } } else { pr_info("mce argument %s ignored. Please use /sys\n", str); return 0; @@ -1983,6 +2040,11 @@ int __init mcheck_init(void) { mcheck_intel_therm_init(); + mce_register_decode_chain(&mce_srao_nb); + mcheck_vendor_init_severity(); + + INIT_WORK(&mce_work, mce_process_work); + init_irq_work(&mce_irq_work, mce_irq_work_cb); return 0; } @@ -1995,7 +2057,7 @@ * Disable machine checks on suspend and shutdown. We can't really handle * them later. */ -static int mce_disable_error_reporting(void) +static void mce_disable_error_reporting(void) { int i; @@ -2005,17 +2067,32 @@ if (b->init) wrmsrl(MSR_IA32_MCx_CTL(i), 0); } - return 0; + return; +} + +static void vendor_disable_error_reporting(void) +{ + /* + * Don't clear on Intel 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) + return; + + mce_disable_error_reporting(); } static int mce_syscore_suspend(void) { - return mce_disable_error_reporting(); + vendor_disable_error_reporting(); + return 0; } static void mce_syscore_shutdown(void) { - mce_disable_error_reporting(); + vendor_disable_error_reporting(); } /* @@ -2026,7 +2103,7 @@ static void mce_syscore_resume(void) { __mcheck_cpu_init_generic(); - __mcheck_cpu_init_vendor(__this_cpu_ptr(&cpu_info)); + __mcheck_cpu_init_vendor(raw_cpu_ptr(&cpu_info)); } static struct syscore_ops mce_syscore_ops = { @@ -2041,7 +2118,7 @@ static void mce_cpu_restart(void *data) { - if (!mce_available(__this_cpu_ptr(&cpu_info))) + if (!mce_available(raw_cpu_ptr(&cpu_info))) return; __mcheck_cpu_init_generic(); __mcheck_cpu_init_timer(); @@ -2057,14 +2134,14 @@ /* Toggle features for corrected errors */ static void mce_disable_cmci(void *data) { - if (!mce_available(__this_cpu_ptr(&cpu_info))) + if (!mce_available(raw_cpu_ptr(&cpu_info))) return; cmci_clear(); } static void mce_enable_ce(void *all) { - if (!mce_available(__this_cpu_ptr(&cpu_info))) + if (!mce_available(raw_cpu_ptr(&cpu_info))) return; cmci_reenable(); cmci_recheck(); @@ -2079,7 +2156,6 @@ DEFINE_PER_CPU(struct device *, mce_device); -__cpuinitdata void (*threshold_cpu_callback)(unsigned long action, unsigned int cpu); static inline struct mce_bank *attr_to_bank(struct device_attribute *attr) @@ -2098,7 +2174,7 @@ { u64 new; - if (strict_strtoull(buf, 0, &new) < 0) + if (kstrtou64(buf, 0, &new) < 0) return -EINVAL; attr_to_bank(attr)->ctl = new; @@ -2136,7 +2212,7 @@ { u64 new; - if (strict_strtoull(buf, 0, &new) < 0) + if (kstrtou64(buf, 0, &new) < 0) return -EINVAL; if (mca_cfg.ignore_ce ^ !!new) { @@ -2160,7 +2236,7 @@ { u64 new; - if (strict_strtoull(buf, 0, &new) < 0) + if (kstrtou64(buf, 0, &new) < 0) return -EINVAL; if (mca_cfg.cmci_disabled ^ !!new) { @@ -2225,7 +2301,7 @@ } /* Per cpu device init. All of the cpus still share the same ctrl bank: */ -static __cpuinit int mce_device_create(unsigned int cpu) +static int mce_device_create(unsigned int cpu) { struct device *dev; int err; @@ -2242,8 +2318,10 @@ dev->release = &mce_device_release; err = device_register(dev); - if (err) + if (err) { + put_device(dev); return err; + } for (i = 0; mce_device_attrs[i]; i++) { err = device_create_file(dev, mce_device_attrs[i]); @@ -2271,7 +2349,7 @@ return err; } -static __cpuinit void mce_device_remove(unsigned int cpu) +static void mce_device_remove(unsigned int cpu) { struct device *dev = per_cpu(mce_device, cpu); int i; @@ -2291,30 +2369,25 @@ } /* Make sure there are no machine checks on offlined CPUs. */ -static void __cpuinit mce_disable_cpu(void *h) +static void mce_disable_cpu(void *h) { unsigned long action = *(unsigned long *)h; - int i; - if (!mce_available(__this_cpu_ptr(&cpu_info))) + if (!mce_available(raw_cpu_ptr(&cpu_info))) return; if (!(action & CPU_TASKS_FROZEN)) cmci_clear(); - for (i = 0; i < mca_cfg.banks; i++) { - struct mce_bank *b = &mce_banks[i]; - if (b->init) - wrmsrl(MSR_IA32_MCx_CTL(i), 0); - } + vendor_disable_error_reporting(); } -static void __cpuinit mce_reenable_cpu(void *h) +static void mce_reenable_cpu(void *h) { unsigned long action = *(unsigned long *)h; int i; - if (!mce_available(__this_cpu_ptr(&cpu_info))) + if (!mce_available(raw_cpu_ptr(&cpu_info))) return; if (!(action & CPU_TASKS_FROZEN)) @@ -2328,7 +2401,7 @@ } /* Get notified when a cpu comes on/off. Be hotplug friendly. */ -static int __cpuinit +static int mce_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { unsigned int cpu = (unsigned long)hcpu; @@ -2345,6 +2418,10 @@ threshold_cpu_callback(action, cpu); mce_device_remove(cpu); mce_intel_hcpu_update(cpu); + + /* intentionally ignoring frozen here */ + if (!(action & CPU_TASKS_FROZEN)) + cmci_rediscover(); break; case CPU_DOWN_PREPARE: smp_call_function_single(cpu, mce_disable_cpu, &action, 1); @@ -2356,15 +2433,10 @@ break; } - if (action == CPU_POST_DEAD) { - /* intentionally ignoring frozen here */ - cmci_rediscover(); - } - return NOTIFY_OK; } -static struct notifier_block mce_cpu_notifier __cpuinitdata = { +static struct notifier_block mce_cpu_notifier = { .notifier_call = mce_cpu_callback, }; @@ -2391,28 +2463,67 @@ int err; int i = 0; - if (!mce_available(&boot_cpu_data)) - return -EIO; + if (!mce_available(&boot_cpu_data)) { + err = -EIO; + goto err_out; + } - zalloc_cpumask_var(&mce_device_initialized, GFP_KERNEL); + 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) - return err; + goto err_out_mem; + cpu_notifier_register_begin(); for_each_online_cpu(i) { err = mce_device_create(i); - if (err) - return err; + if (err) { + /* + * Register notifier anyway (and do not unreg it) so + * that we don't leave undeleted timers, see notifier + * callback above. + */ + __register_hotcpu_notifier(&mce_cpu_notifier); + cpu_notifier_register_done(); + goto err_device_create; + } } + __register_hotcpu_notifier(&mce_cpu_notifier); + cpu_notifier_register_done(); + register_syscore_ops(&mce_syscore_ops); - register_hotcpu_notifier(&mce_cpu_notifier); /* register character device /dev/mcelog */ - misc_register(&mce_chrdev_device); + err = misc_register(&mce_chrdev_device); + if (err) + goto err_register; + + return 0; + +err_register: + unregister_syscore_ops(&mce_syscore_ops); + +err_device_create: + /* + * We didn't keep track of which devices were created above, but + * even if we had, the set of online cpus might have changed. + * Play safe and remove for every possible cpu, since + * mce_device_remove() will do the right thing. + */ + for_each_possible_cpu(i) + mce_device_remove(i); + +err_out_mem: + free_cpumask_var(mce_device_initialized); + +err_out: + pr_err("Unable to init device /dev/mcelog (rc: %d)\n", err); return err; } @@ -2442,7 +2553,7 @@ static void mce_reset(void) { cpu_missing = 0; - atomic_set(&mce_fake_paniced, 0); + atomic_set(&mce_fake_panicked, 0); atomic_set(&mce_executing, 0); atomic_set(&mce_callin, 0); atomic_set(&global_nwo, 0); @@ -2478,5 +2589,20 @@ return 0; } -late_initcall(mcheck_debugfs_init); +#else +static int __init mcheck_debugfs_init(void) { return -EINVAL; } #endif + +static int __init mcheck_late_init(void) +{ + 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);