diff options
Diffstat (limited to 'kernel/arch/x86/kernel')
40 files changed, 533 insertions, 172 deletions
diff --git a/kernel/arch/x86/kernel/acpi/boot.c b/kernel/arch/x86/kernel/acpi/boot.c index e75907601..a29fc4f84 100644 --- a/kernel/arch/x86/kernel/acpi/boot.c +++ b/kernel/arch/x86/kernel/acpi/boot.c @@ -87,7 +87,9 @@ static u64 acpi_lapic_addr __initdata = APIC_DEFAULT_PHYS_BASE; * ->ioapic_mutex * ->ioapic_lock */ +#ifdef CONFIG_X86_IO_APIC static DEFINE_MUTEX(acpi_ioapic_lock); +#endif /* -------------------------------------------------------------------------- Boot-time Configuration diff --git a/kernel/arch/x86/kernel/amd_nb.c b/kernel/arch/x86/kernel/amd_nb.c index 29fa475ec..c986d0b3b 100644 --- a/kernel/arch/x86/kernel/amd_nb.c +++ b/kernel/arch/x86/kernel/amd_nb.c @@ -71,8 +71,8 @@ int amd_cache_northbridges(void) while ((misc = next_northbridge(misc, amd_nb_misc_ids)) != NULL) i++; - if (i == 0) - return 0; + if (!i) + return -ENODEV; nb = kzalloc(i * sizeof(struct amd_northbridge), GFP_KERNEL); if (!nb) diff --git a/kernel/arch/x86/kernel/apic/apic.c b/kernel/arch/x86/kernel/apic/apic.c index 58e7cdb25..042e744ef 100644 --- a/kernel/arch/x86/kernel/apic/apic.c +++ b/kernel/arch/x86/kernel/apic/apic.c @@ -1611,6 +1611,9 @@ void __init enable_IR_x2apic(void) unsigned long flags; int ret, ir_stat; + if (skip_ioapic_setup) + return; + ir_stat = irq_remapping_prepare(); if (ir_stat < 0 && !x2apic_supported()) return; diff --git a/kernel/arch/x86/kernel/apic/io_apic.c b/kernel/arch/x86/kernel/apic/io_apic.c index 678c711e2..7bfb53808 100644 --- a/kernel/arch/x86/kernel/apic/io_apic.c +++ b/kernel/arch/x86/kernel/apic/io_apic.c @@ -2116,6 +2116,7 @@ static inline void __init check_timer(void) if (idx != -1 && irq_trigger(idx)) unmask_ioapic_irq(irq_get_chip_data(0)); } + irq_domain_deactivate_irq(irq_data); irq_domain_activate_irq(irq_data); if (timer_irq_works()) { if (disable_timer_pin_1 > 0) @@ -2137,6 +2138,7 @@ static inline void __init check_timer(void) * legacy devices should be connected to IO APIC #0 */ replace_pin_at_irq_node(data, node, apic1, pin1, apic2, pin2); + irq_domain_deactivate_irq(irq_data); irq_domain_activate_irq(irq_data); legacy_pic->unmask(0); if (timer_irq_works()) { diff --git a/kernel/arch/x86/kernel/apic/vector.c b/kernel/arch/x86/kernel/apic/vector.c index a35f6b547..0988e204f 100644 --- a/kernel/arch/x86/kernel/apic/vector.c +++ b/kernel/arch/x86/kernel/apic/vector.c @@ -211,6 +211,7 @@ update: */ cpumask_and(d->old_domain, d->old_domain, cpu_online_mask); d->move_in_progress = !cpumask_empty(d->old_domain); + d->cfg.old_vector = d->move_in_progress ? d->cfg.vector : 0; d->cfg.vector = vector; cpumask_copy(d->domain, vector_cpumask); success: @@ -253,7 +254,8 @@ static void clear_irq_vector(int irq, struct apic_chip_data *data) struct irq_desc *desc; int cpu, vector; - BUG_ON(!data->cfg.vector); + if (!data->cfg.vector) + return; vector = data->cfg.vector; for_each_cpu_and(cpu, data->domain, cpu_online_mask) @@ -653,46 +655,114 @@ void irq_complete_move(struct irq_cfg *cfg) } /* - * Called with @desc->lock held and interrupts disabled. + * Called from fixup_irqs() with @desc->lock held and interrupts disabled. */ void irq_force_complete_move(struct irq_desc *desc) { - struct irq_data *irqdata = irq_desc_get_irq_data(desc); - struct apic_chip_data *data = apic_chip_data(irqdata); - struct irq_cfg *cfg = data ? &data->cfg : NULL; + struct irq_data *irqdata; + struct apic_chip_data *data; + struct irq_cfg *cfg; + unsigned int cpu; - if (!cfg) + /* + * The function is called for all descriptors regardless of which + * irqdomain they belong to. For example if an IRQ is provided by + * an irq_chip as part of a GPIO driver, the chip data for that + * descriptor is specific to the irq_chip in question. + * + * Check first that the chip_data is what we expect + * (apic_chip_data) before touching it any further. + */ + irqdata = irq_domain_get_irq_data(x86_vector_domain, + irq_desc_get_irq(desc)); + if (!irqdata) return; - __irq_complete_move(cfg, cfg->vector); + data = apic_chip_data(irqdata); + cfg = data ? &data->cfg : NULL; + + if (!cfg) + return; /* * This is tricky. If the cleanup of @data->old_domain has not been * done yet, then the following setaffinity call will fail with * -EBUSY. This can leave the interrupt in a stale state. * - * The cleanup cannot make progress because we hold @desc->lock. So in - * case @data->old_domain is not yet cleaned up, we need to drop the - * lock and acquire it again. @desc cannot go away, because the - * hotplug code holds the sparse irq lock. + * All CPUs are stuck in stop machine with interrupts disabled so + * calling __irq_complete_move() would be completely pointless. */ raw_spin_lock(&vector_lock); - /* Clean out all offline cpus (including ourself) first. */ + /* + * Clean out all offline cpus (including the outgoing one) from the + * old_domain mask. + */ cpumask_and(data->old_domain, data->old_domain, cpu_online_mask); - while (!cpumask_empty(data->old_domain)) { + + /* + * If move_in_progress is cleared and the old_domain mask is empty, + * then there is nothing to cleanup. fixup_irqs() will take care of + * the stale vectors on the outgoing cpu. + */ + if (!data->move_in_progress && cpumask_empty(data->old_domain)) { raw_spin_unlock(&vector_lock); - raw_spin_unlock(&desc->lock); - cpu_relax(); - raw_spin_lock(&desc->lock); + return; + } + + /* + * 1) The interrupt is in move_in_progress state. That means that we + * have not seen an interrupt since the io_apic was reprogrammed to + * the new vector. + * + * 2) The interrupt has fired on the new vector, but the cleanup IPIs + * have not been processed yet. + */ + if (data->move_in_progress) { /* - * Reevaluate apic_chip_data. It might have been cleared after - * we dropped @desc->lock. + * In theory there is a race: + * + * set_ioapic(new_vector) <-- Interrupt is raised before update + * is effective, i.e. it's raised on + * the old vector. + * + * So if the target cpu cannot handle that interrupt before + * the old vector is cleaned up, we get a spurious interrupt + * and in the worst case the ioapic irq line becomes stale. + * + * But in case of cpu hotplug this should be a non issue + * because if the affinity update happens right before all + * cpus rendevouz in stop machine, there is no way that the + * interrupt can be blocked on the target cpu because all cpus + * loops first with interrupts enabled in stop machine, so the + * old vector is not yet cleaned up when the interrupt fires. + * + * So the only way to run into this issue is if the delivery + * of the interrupt on the apic/system bus would be delayed + * beyond the point where the target cpu disables interrupts + * in stop machine. I doubt that it can happen, but at least + * there is a theroretical chance. Virtualization might be + * able to expose this, but AFAICT the IOAPIC emulation is not + * as stupid as the real hardware. + * + * Anyway, there is nothing we can do about that at this point + * w/o refactoring the whole fixup_irq() business completely. + * We print at least the irq number and the old vector number, + * so we have the necessary information when a problem in that + * area arises. */ - data = apic_chip_data(irqdata); - if (!data) - return; - raw_spin_lock(&vector_lock); + pr_warn("IRQ fixup: irq %d move in progress, old vector %d\n", + irqdata->irq, cfg->old_vector); } + /* + * If old_domain is not empty, then other cpus still have the irq + * descriptor set in their vector array. Clean it up. + */ + for_each_cpu(cpu, data->old_domain) + per_cpu(vector_irq, cpu)[cfg->old_vector] = VECTOR_UNUSED; + + /* Cleanup the left overs of the (half finished) move */ + cpumask_clear(data->old_domain); + data->move_in_progress = 0; raw_spin_unlock(&vector_lock); } #endif diff --git a/kernel/arch/x86/kernel/cpu/amd.c b/kernel/arch/x86/kernel/cpu/amd.c index a8816b325..e2defc759 100644 --- a/kernel/arch/x86/kernel/cpu/amd.c +++ b/kernel/arch/x86/kernel/cpu/amd.c @@ -352,7 +352,6 @@ static void amd_detect_cmp(struct cpuinfo_x86 *c) #ifdef CONFIG_SMP unsigned bits; int cpu = smp_processor_id(); - unsigned int socket_id, core_complex_id; bits = c->x86_coreid_bits; /* Low order bits define the core id (index of core in socket) */ @@ -370,10 +369,7 @@ static void amd_detect_cmp(struct cpuinfo_x86 *c) if (c->x86 != 0x17 || !cpuid_edx(0x80000006)) return; - socket_id = (c->apicid >> bits) - 1; - core_complex_id = (c->apicid & ((1 << bits) - 1)) >> 3; - - per_cpu(cpu_llc_id, cpu) = (socket_id << 3) | core_complex_id; + per_cpu(cpu_llc_id, cpu) = c->apicid >> 3; #endif } @@ -656,6 +652,17 @@ static void init_amd_gh(struct cpuinfo_x86 *c) set_cpu_bug(c, X86_BUG_AMD_TLB_MMATCH); } +#define MSR_AMD64_DE_CFG 0xC0011029 + +static void init_amd_ln(struct cpuinfo_x86 *c) +{ + /* + * Apply erratum 665 fix unconditionally so machines without a BIOS + * fix work. + */ + msr_set_bit(MSR_AMD64_DE_CFG, 31); +} + static void init_amd_bd(struct cpuinfo_x86 *c) { u64 value; @@ -713,6 +720,7 @@ static void init_amd(struct cpuinfo_x86 *c) case 6: init_amd_k7(c); break; case 0xf: init_amd_k8(c); break; case 0x10: init_amd_gh(c); break; + case 0x12: init_amd_ln(c); break; case 0x15: init_amd_bd(c); break; } diff --git a/kernel/arch/x86/kernel/cpu/common.c b/kernel/arch/x86/kernel/cpu/common.c index c2b7522cb..637ca414d 100644 --- a/kernel/arch/x86/kernel/cpu/common.c +++ b/kernel/arch/x86/kernel/cpu/common.c @@ -737,21 +737,20 @@ static void __init early_identify_cpu(struct cpuinfo_x86 *c) identify_cpu_without_cpuid(c); /* cyrix could have cpuid enabled via c_identify()*/ - if (!have_cpuid_p()) - return; + if (have_cpuid_p()) { + cpu_detect(c); + get_cpu_vendor(c); + get_cpu_cap(c); - cpu_detect(c); - get_cpu_vendor(c); - get_cpu_cap(c); - - if (this_cpu->c_early_init) - this_cpu->c_early_init(c); + if (this_cpu->c_early_init) + this_cpu->c_early_init(c); - c->cpu_index = 0; - filter_cpuid_features(c, false); + c->cpu_index = 0; + filter_cpuid_features(c, false); - if (this_cpu->c_bsp_init) - this_cpu->c_bsp_init(c); + if (this_cpu->c_bsp_init) + this_cpu->c_bsp_init(c); + } setup_force_cpu_cap(X86_FEATURE_ALWAYS); fpu__init_system(c); @@ -1130,7 +1129,7 @@ static __init int setup_disablecpuid(char *arg) { int bit; - if (get_option(&arg, &bit) && bit < NCAPINTS*32) + if (get_option(&arg, &bit) && bit >= 0 && bit < NCAPINTS * 32) setup_clear_cpu_cap(bit); else return 0; diff --git a/kernel/arch/x86/kernel/cpu/mcheck/mce-genpool.c b/kernel/arch/x86/kernel/cpu/mcheck/mce-genpool.c index 0a850100c..2658e2af7 100644 --- a/kernel/arch/x86/kernel/cpu/mcheck/mce-genpool.c +++ b/kernel/arch/x86/kernel/cpu/mcheck/mce-genpool.c @@ -29,7 +29,7 @@ static char gen_pool_buf[MCE_POOLSZ]; void mce_gen_pool_process(void) { struct llist_node *head; - struct mce_evt_llist *node; + struct mce_evt_llist *node, *tmp; struct mce *mce; head = llist_del_all(&mce_event_llist); @@ -37,7 +37,7 @@ void mce_gen_pool_process(void) return; head = llist_reverse_order(head); - llist_for_each_entry(node, head, llnode) { + llist_for_each_entry_safe(node, tmp, head, llnode) { mce = &node->mce; atomic_notifier_call_chain(&x86_mce_decoder_chain, 0, mce); gen_pool_free(mce_evt_pool, (unsigned long)node, sizeof(*node)); diff --git a/kernel/arch/x86/kernel/cpu/mcheck/mce.c b/kernel/arch/x86/kernel/cpu/mcheck/mce.c index a080b4939..430a4ec07 100644 --- a/kernel/arch/x86/kernel/cpu/mcheck/mce.c +++ b/kernel/arch/x86/kernel/cpu/mcheck/mce.c @@ -42,7 +42,7 @@ #include <linux/irq_work.h> #include <linux/export.h> #include <linux/jiffies.h> -#include <linux/work-simple.h> +#include <linux/swork.h> #include <asm/processor.h> #include <asm/traps.h> diff --git a/kernel/arch/x86/kernel/cpu/mcheck/therm_throt.c b/kernel/arch/x86/kernel/cpu/mcheck/therm_throt.c index 2c5aaf8c2..05538582a 100644 --- a/kernel/arch/x86/kernel/cpu/mcheck/therm_throt.c +++ b/kernel/arch/x86/kernel/cpu/mcheck/therm_throt.c @@ -385,6 +385,9 @@ static void intel_thermal_interrupt(void) { __u64 msr_val; + if (static_cpu_has(X86_FEATURE_HWP)) + wrmsrl_safe(MSR_HWP_STATUS, 0); + rdmsrl(MSR_IA32_THERM_STATUS, msr_val); /* Check for violation of core thermal thresholds*/ diff --git a/kernel/arch/x86/kernel/cpu/microcode/intel.c b/kernel/arch/x86/kernel/cpu/microcode/intel.c index ce47402eb..ac8975a65 100644 --- a/kernel/arch/x86/kernel/cpu/microcode/intel.c +++ b/kernel/arch/x86/kernel/cpu/microcode/intel.c @@ -555,10 +555,14 @@ scan_microcode(struct mc_saved_data *mc_saved_data, unsigned long *initrd, cd.data = NULL; cd.size = 0; - cd = find_cpio_data(p, (void *)start, size, &offset); - if (!cd.data) { + /* try built-in microcode if no initrd */ + if (!size) { if (!load_builtin_intel_microcode(&cd)) return UCODE_ERROR; + } else { + cd = find_cpio_data(p, (void *)start, size, &offset); + if (!cd.data) + return UCODE_ERROR; } return get_matching_model_microcode(0, start, cd.data, cd.size, @@ -694,7 +698,7 @@ int __init save_microcode_in_initrd_intel(void) if (count == 0) return ret; - copy_initrd_ptrs(mc_saved, mc_saved_in_initrd, initrd_start, count); + copy_initrd_ptrs(mc_saved, mc_saved_in_initrd, get_initrd_start(), count); ret = save_microcode(&mc_saved_data, mc_saved, count); if (ret) pr_err("Cannot save microcode patches from initrd.\n"); @@ -732,16 +736,20 @@ void __init load_ucode_intel_bsp(void) struct boot_params *p; p = (struct boot_params *)__pa_nodebug(&boot_params); - start = p->hdr.ramdisk_image; size = p->hdr.ramdisk_size; - _load_ucode_intel_bsp( - (struct mc_saved_data *)__pa_nodebug(&mc_saved_data), - (unsigned long *)__pa_nodebug(&mc_saved_in_initrd), - start, size); + /* + * Set start only if we have an initrd image. We cannot use initrd_start + * because it is not set that early yet. + */ + start = (size ? p->hdr.ramdisk_image : 0); + + _load_ucode_intel_bsp((struct mc_saved_data *)__pa_nodebug(&mc_saved_data), + (unsigned long *)__pa_nodebug(&mc_saved_in_initrd), + start, size); #else - start = boot_params.hdr.ramdisk_image + PAGE_OFFSET; size = boot_params.hdr.ramdisk_size; + start = (size ? boot_params.hdr.ramdisk_image + PAGE_OFFSET : 0); _load_ucode_intel_bsp(&mc_saved_data, mc_saved_in_initrd, start, size); #endif @@ -752,20 +760,14 @@ void load_ucode_intel_ap(void) struct mc_saved_data *mc_saved_data_p; struct ucode_cpu_info uci; unsigned long *mc_saved_in_initrd_p; - unsigned long initrd_start_addr; enum ucode_state ret; #ifdef CONFIG_X86_32 - unsigned long *initrd_start_p; - mc_saved_in_initrd_p = - (unsigned long *)__pa_nodebug(mc_saved_in_initrd); + mc_saved_in_initrd_p = (unsigned long *)__pa_nodebug(mc_saved_in_initrd); mc_saved_data_p = (struct mc_saved_data *)__pa_nodebug(&mc_saved_data); - initrd_start_p = (unsigned long *)__pa_nodebug(&initrd_start); - initrd_start_addr = (unsigned long)__pa_nodebug(*initrd_start_p); #else - mc_saved_data_p = &mc_saved_data; mc_saved_in_initrd_p = mc_saved_in_initrd; - initrd_start_addr = initrd_start; + mc_saved_data_p = &mc_saved_data; #endif /* @@ -777,7 +779,7 @@ void load_ucode_intel_ap(void) collect_cpu_info_early(&uci); ret = load_microcode(mc_saved_data_p, mc_saved_in_initrd_p, - initrd_start_addr, &uci); + get_initrd_start_addr(), &uci); if (ret != UCODE_OK) return; diff --git a/kernel/arch/x86/kernel/cpu/mshyperv.c b/kernel/arch/x86/kernel/cpu/mshyperv.c index 20e242ea1..cfc4a966e 100644 --- a/kernel/arch/x86/kernel/cpu/mshyperv.c +++ b/kernel/arch/x86/kernel/cpu/mshyperv.c @@ -152,6 +152,11 @@ static struct clocksource hyperv_cs = { .flags = CLOCK_SOURCE_IS_CONTINUOUS, }; +static unsigned char hv_get_nmi_reason(void) +{ + return 0; +} + static void __init ms_hyperv_init_platform(void) { /* @@ -191,6 +196,13 @@ static void __init ms_hyperv_init_platform(void) machine_ops.crash_shutdown = hv_machine_crash_shutdown; #endif mark_tsc_unstable("running on Hyper-V"); + + /* + * Generation 2 instances don't support reading the NMI status from + * 0x61 port. + */ + if (efi_enabled(EFI_BOOT)) + x86_platform.get_nmi_reason = hv_get_nmi_reason; } const __refconst struct hypervisor_x86 x86_hyper_ms_hyperv = { diff --git a/kernel/arch/x86/kernel/cpu/mtrr/generic.c b/kernel/arch/x86/kernel/cpu/mtrr/generic.c index 3b533cf37..b5624fafa 100644 --- a/kernel/arch/x86/kernel/cpu/mtrr/generic.c +++ b/kernel/arch/x86/kernel/cpu/mtrr/generic.c @@ -444,11 +444,24 @@ static void __init print_mtrr_state(void) pr_debug("TOM2: %016llx aka %lldM\n", mtrr_tom2, mtrr_tom2>>20); } +/* PAT setup for BP. We need to go through sync steps here */ +void __init mtrr_bp_pat_init(void) +{ + unsigned long flags; + + local_irq_save(flags); + prepare_set(); + + pat_init(); + + post_set(); + local_irq_restore(flags); +} + /* Grab all of the MTRR state for this CPU into *state */ bool __init get_mtrr_state(void) { struct mtrr_var_range *vrs; - unsigned long flags; unsigned lo, dummy; unsigned int i; @@ -481,15 +494,6 @@ bool __init get_mtrr_state(void) mtrr_state_set = 1; - /* PAT setup for BP. We need to go through sync steps here */ - local_irq_save(flags); - prepare_set(); - - pat_init(); - - post_set(); - local_irq_restore(flags); - return !!(mtrr_state.enabled & MTRR_STATE_MTRR_ENABLED); } diff --git a/kernel/arch/x86/kernel/cpu/mtrr/main.c b/kernel/arch/x86/kernel/cpu/mtrr/main.c index f891b4750..fa77ac829 100644 --- a/kernel/arch/x86/kernel/cpu/mtrr/main.c +++ b/kernel/arch/x86/kernel/cpu/mtrr/main.c @@ -752,6 +752,9 @@ void __init mtrr_bp_init(void) /* BIOS may override */ __mtrr_enabled = get_mtrr_state(); + if (mtrr_enabled()) + mtrr_bp_pat_init(); + if (mtrr_cleanup(phys_addr)) { changed_by_mtrr_cleanup = 1; mtrr_if->set_all(); @@ -759,8 +762,16 @@ void __init mtrr_bp_init(void) } } - if (!mtrr_enabled()) + if (!mtrr_enabled()) { pr_info("MTRR: Disabled\n"); + + /* + * PAT initialization relies on MTRR's rendezvous handler. + * Skip PAT init until the handler can initialize both + * features independently. + */ + pat_disable("MTRRs disabled, skipping PAT initialization too."); + } } void mtrr_ap_init(void) diff --git a/kernel/arch/x86/kernel/cpu/mtrr/mtrr.h b/kernel/arch/x86/kernel/cpu/mtrr/mtrr.h index 951884dcc..6c7ced07d 100644 --- a/kernel/arch/x86/kernel/cpu/mtrr/mtrr.h +++ b/kernel/arch/x86/kernel/cpu/mtrr/mtrr.h @@ -52,6 +52,7 @@ void set_mtrr_prepare_save(struct set_mtrr_context *ctxt); void fill_mtrr_var_range(unsigned int index, u32 base_lo, u32 base_hi, u32 mask_lo, u32 mask_hi); bool get_mtrr_state(void); +void mtrr_bp_pat_init(void); extern void set_mtrr_ops(const struct mtrr_ops *ops); diff --git a/kernel/arch/x86/kernel/cpu/perf_event.c b/kernel/arch/x86/kernel/cpu/perf_event.c index 2bf79d7c9..1a8256dd6 100644 --- a/kernel/arch/x86/kernel/cpu/perf_event.c +++ b/kernel/arch/x86/kernel/cpu/perf_event.c @@ -67,7 +67,7 @@ u64 x86_perf_event_update(struct perf_event *event) int shift = 64 - x86_pmu.cntval_bits; u64 prev_raw_count, new_raw_count; int idx = hwc->idx; - s64 delta; + u64 delta; if (idx == INTEL_PMC_IDX_FIXED_BTS) return 0; @@ -593,6 +593,19 @@ void x86_pmu_disable_all(void) } } +/* + * There may be PMI landing after enabled=0. The PMI hitting could be before or + * after disable_all. + * + * If PMI hits before disable_all, the PMU will be disabled in the NMI handler. + * It will not be re-enabled in the NMI handler again, because enabled=0. After + * handling the NMI, disable_all will be called, which will not change the + * state either. If PMI hits after disable_all, the PMU is already disabled + * before entering NMI handler. The NMI handler will not change the state + * either. + * + * So either situation is harmless. + */ static void x86_pmu_disable(struct pmu *pmu) { struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events); diff --git a/kernel/arch/x86/kernel/cpu/perf_event.h b/kernel/arch/x86/kernel/cpu/perf_event.h index d0e35ebb2..ee70445fb 100644 --- a/kernel/arch/x86/kernel/cpu/perf_event.h +++ b/kernel/arch/x86/kernel/cpu/perf_event.h @@ -591,6 +591,7 @@ struct x86_pmu { pebs_active :1, pebs_broken :1; int pebs_record_size; + int pebs_buffer_size; void (*drain_pebs)(struct pt_regs *regs); struct event_constraint *pebs_constraints; void (*pebs_aliases)(struct perf_event *event); @@ -907,6 +908,8 @@ void intel_pmu_lbr_init_hsw(void); void intel_pmu_lbr_init_skl(void); +void intel_pmu_pebs_data_source_nhm(void); + int intel_pmu_setup_lbr_filter(struct perf_event *event); void intel_pt_interrupt(void); diff --git a/kernel/arch/x86/kernel/cpu/perf_event_intel.c b/kernel/arch/x86/kernel/cpu/perf_event_intel.c index e2a430021..5cc2242d7 100644 --- a/kernel/arch/x86/kernel/cpu/perf_event_intel.c +++ b/kernel/arch/x86/kernel/cpu/perf_event_intel.c @@ -1458,7 +1458,15 @@ static __initconst const u64 slm_hw_cache_event_ids }; /* - * Use from PMIs where the LBRs are already disabled. + * Used from PMIs where the LBRs are already disabled. + * + * This function could be called consecutively. It is required to remain in + * disabled state if called consecutively. + * + * During consecutive calls, the same disable value will be written to related + * registers, so the PMU state remains unchanged. hw.state in + * intel_bts_disable_local will remain PERF_HES_STOPPED too in consecutive + * calls. */ static void __intel_pmu_disable_all(void) { @@ -1840,6 +1848,16 @@ again: if (__test_and_clear_bit(62, (unsigned long *)&status)) { handled++; x86_pmu.drain_pebs(regs); + /* + * There are cases where, even though, the PEBS ovfl bit is set + * in GLOBAL_OVF_STATUS, the PEBS events may also have their + * overflow bits set for their counters. We must clear them + * here because they have been processed as exact samples in + * the drain_pebs() routine. They must not be processed again + * in the for_each_bit_set() loop for regular samples below. + */ + status &= ~cpuc->pebs_enabled; + status &= x86_pmu.intel_ctrl | GLOBAL_STATUS_TRACE_TOPAPMI; } /* @@ -1885,7 +1903,10 @@ again: goto again; done: - __intel_pmu_enable_all(0, true); + /* Only restore PMU state when it's active. See x86_pmu_disable(). */ + if (cpuc->enabled) + __intel_pmu_enable_all(0, true); + /* * Only unmask the NMI after the overflow counters * have been reset. This avoids spurious NMIs on @@ -3315,6 +3336,7 @@ __init int intel_pmu_init(void) intel_perfmon_event_map[PERF_COUNT_HW_STALLED_CYCLES_BACKEND] = X86_CONFIG(.event=0xb1, .umask=0x3f, .inv=1, .cmask=1); + intel_pmu_pebs_data_source_nhm(); x86_add_quirk(intel_nehalem_quirk); pr_cont("Nehalem events, "); @@ -3377,6 +3399,7 @@ __init int intel_pmu_init(void) intel_perfmon_event_map[PERF_COUNT_HW_STALLED_CYCLES_BACKEND] = X86_CONFIG(.event=0xb1, .umask=0x3f, .inv=1, .cmask=1); + intel_pmu_pebs_data_source_nhm(); pr_cont("Westmere events, "); break; @@ -3578,7 +3601,7 @@ __init int intel_pmu_init(void) c->idxmsk64 |= (1ULL << x86_pmu.num_counters) - 1; } c->idxmsk64 &= - ~(~0UL << (INTEL_PMC_IDX_FIXED + x86_pmu.num_counters_fixed)); + ~(~0ULL << (INTEL_PMC_IDX_FIXED + x86_pmu.num_counters_fixed)); c->weight = hweight64(c->idxmsk64); } } @@ -3613,7 +3636,7 @@ __init int intel_pmu_init(void) /* Support full width counters using alternative MSR range */ if (x86_pmu.intel_cap.full_width_write) { - x86_pmu.max_period = x86_pmu.cntval_mask; + x86_pmu.max_period = x86_pmu.cntval_mask >> 1; x86_pmu.perfctr = MSR_IA32_PMC0; pr_cont("full-width counters, "); } diff --git a/kernel/arch/x86/kernel/cpu/perf_event_intel_cqm.c b/kernel/arch/x86/kernel/cpu/perf_event_intel_cqm.c index a316ca96f..fc704ed58 100644 --- a/kernel/arch/x86/kernel/cpu/perf_event_intel_cqm.c +++ b/kernel/arch/x86/kernel/cpu/perf_event_intel_cqm.c @@ -211,6 +211,20 @@ static void __put_rmid(u32 rmid) list_add_tail(&entry->list, &cqm_rmid_limbo_lru); } +static void cqm_cleanup(void) +{ + int i; + + if (!cqm_rmid_ptrs) + return; + + for (i = 0; i < cqm_max_rmid; i++) + kfree(cqm_rmid_ptrs[i]); + + kfree(cqm_rmid_ptrs); + cqm_rmid_ptrs = NULL; +} + static int intel_cqm_setup_rmid_cache(void) { struct cqm_rmid_entry *entry; @@ -218,7 +232,7 @@ static int intel_cqm_setup_rmid_cache(void) int r = 0; nr_rmids = cqm_max_rmid + 1; - cqm_rmid_ptrs = kmalloc(sizeof(struct cqm_rmid_entry *) * + cqm_rmid_ptrs = kzalloc(sizeof(struct cqm_rmid_entry *) * nr_rmids, GFP_KERNEL); if (!cqm_rmid_ptrs) return -ENOMEM; @@ -249,11 +263,9 @@ static int intel_cqm_setup_rmid_cache(void) mutex_unlock(&cache_mutex); return 0; -fail: - while (r--) - kfree(cqm_rmid_ptrs[r]); - kfree(cqm_rmid_ptrs); +fail: + cqm_cleanup(); return -ENOMEM; } @@ -281,9 +293,13 @@ static bool __match_event(struct perf_event *a, struct perf_event *b) /* * Events that target same task are placed into the same cache group. + * Mark it as a multi event group, so that we update ->count + * for every event rather than just the group leader later. */ - if (a->hw.target == b->hw.target) + if (a->hw.target == b->hw.target) { + b->hw.is_group_event = true; return true; + } /* * Are we an inherited event? @@ -849,6 +865,7 @@ static void intel_cqm_setup_event(struct perf_event *event, bool conflict = false; u32 rmid; + event->hw.is_group_event = false; list_for_each_entry(iter, &cache_groups, hw.cqm_groups_entry) { rmid = iter->hw.cqm_rmid; @@ -940,7 +957,9 @@ static u64 intel_cqm_event_count(struct perf_event *event) return __perf_event_count(event); /* - * Only the group leader gets to report values. This stops us + * Only the group leader gets to report values except in case of + * multiple events in the same group, we still need to read the + * other events.This stops us * reporting duplicate values to userspace, and gives us a clear * rule for which task gets to report the values. * @@ -948,7 +967,7 @@ static u64 intel_cqm_event_count(struct perf_event *event) * specific packages - we forfeit that ability when we create * task events. */ - if (!cqm_group_leader(event)) + if (!cqm_group_leader(event) && !event->hw.is_group_event) return 0; /* @@ -1315,7 +1334,7 @@ static const struct x86_cpu_id intel_cqm_match[] = { static int __init intel_cqm_init(void) { - char *str, scale[20]; + char *str = NULL, scale[20]; int i, cpu, ret; if (!x86_match_cpu(intel_cqm_match)) @@ -1375,16 +1394,25 @@ static int __init intel_cqm_init(void) cqm_pick_event_reader(i); } - __perf_cpu_notifier(intel_cqm_cpu_notifier); - ret = perf_pmu_register(&intel_cqm_pmu, "intel_cqm", -1); - if (ret) + if (ret) { pr_err("Intel CQM perf registration failed: %d\n", ret); - else - pr_info("Intel CQM monitoring enabled\n"); + goto out; + } + + pr_info("Intel CQM monitoring enabled\n"); + /* + * Register the hot cpu notifier once we are sure cqm + * is enabled to avoid notifier leak. + */ + __perf_cpu_notifier(intel_cqm_cpu_notifier); out: cpu_notifier_register_done(); + if (ret) { + kfree(str); + cqm_cleanup(); + } return ret; } diff --git a/kernel/arch/x86/kernel/cpu/perf_event_intel_ds.c b/kernel/arch/x86/kernel/cpu/perf_event_intel_ds.c index 5db1c7755..1e7de3cef 100644 --- a/kernel/arch/x86/kernel/cpu/perf_event_intel_ds.c +++ b/kernel/arch/x86/kernel/cpu/perf_event_intel_ds.c @@ -51,7 +51,8 @@ union intel_x86_pebs_dse { #define OP_LH (P(OP, LOAD) | P(LVL, HIT)) #define SNOOP_NONE_MISS (P(SNOOP, NONE) | P(SNOOP, MISS)) -static const u64 pebs_data_source[] = { +/* Version for Sandy Bridge and later */ +static u64 pebs_data_source[] = { P(OP, LOAD) | P(LVL, MISS) | P(LVL, L3) | P(SNOOP, NA),/* 0x00:ukn L3 */ OP_LH | P(LVL, L1) | P(SNOOP, NONE), /* 0x01: L1 local */ OP_LH | P(LVL, LFB) | P(SNOOP, NONE), /* 0x02: LFB hit */ @@ -70,6 +71,14 @@ static const u64 pebs_data_source[] = { OP_LH | P(LVL, UNC) | P(SNOOP, NONE), /* 0x0f: uncached */ }; +/* Patch up minor differences in the bits */ +void __init intel_pmu_pebs_data_source_nhm(void) +{ + pebs_data_source[0x05] = OP_LH | P(LVL, L3) | P(SNOOP, HIT); + pebs_data_source[0x06] = OP_LH | P(LVL, L3) | P(SNOOP, HITM); + pebs_data_source[0x07] = OP_LH | P(LVL, L3) | P(SNOOP, HITM); +} + static u64 precise_store_data(u64 status) { union intel_x86_pebs_dse dse; @@ -269,7 +278,7 @@ static int alloc_pebs_buffer(int cpu) if (!x86_pmu.pebs) return 0; - buffer = kzalloc_node(PEBS_BUFFER_SIZE, GFP_KERNEL, node); + buffer = kzalloc_node(x86_pmu.pebs_buffer_size, GFP_KERNEL, node); if (unlikely(!buffer)) return -ENOMEM; @@ -286,7 +295,7 @@ static int alloc_pebs_buffer(int cpu) per_cpu(insn_buffer, cpu) = ibuffer; } - max = PEBS_BUFFER_SIZE / x86_pmu.pebs_record_size; + max = x86_pmu.pebs_buffer_size / x86_pmu.pebs_record_size; ds->pebs_buffer_base = (u64)(unsigned long)buffer; ds->pebs_index = ds->pebs_buffer_base; @@ -1101,6 +1110,13 @@ get_next_pebs_record_by_bit(void *base, void *top, int bit) void *at; u64 pebs_status; + /* + * fmt0 does not have a status bitfield (does not use + * perf_record_nhm format) + */ + if (x86_pmu.intel_cap.pebs_format < 1) + return base; + if (base == NULL) return NULL; @@ -1186,7 +1202,7 @@ static void intel_pmu_drain_pebs_core(struct pt_regs *iregs) if (!event->attr.precise_ip) return; - n = (top - at) / x86_pmu.pebs_record_size; + n = top - at; if (n <= 0) return; @@ -1296,6 +1312,7 @@ void __init intel_ds_init(void) x86_pmu.bts = boot_cpu_has(X86_FEATURE_BTS); x86_pmu.pebs = boot_cpu_has(X86_FEATURE_PEBS); + x86_pmu.pebs_buffer_size = PEBS_BUFFER_SIZE; if (x86_pmu.pebs) { char pebs_type = x86_pmu.intel_cap.pebs_trap ? '+' : '-'; int format = x86_pmu.intel_cap.pebs_format; @@ -1304,6 +1321,14 @@ void __init intel_ds_init(void) case 0: printk(KERN_CONT "PEBS fmt0%c, ", pebs_type); x86_pmu.pebs_record_size = sizeof(struct pebs_record_core); + /* + * Using >PAGE_SIZE buffers makes the WRMSR to + * PERF_GLOBAL_CTRL in intel_pmu_enable_all() + * mysteriously hang on Core2. + * + * As a workaround, we don't do this. + */ + x86_pmu.pebs_buffer_size = PAGE_SIZE; x86_pmu.drain_pebs = intel_pmu_drain_pebs_core; break; diff --git a/kernel/arch/x86/kernel/cpu/perf_event_intel_pt.c b/kernel/arch/x86/kernel/cpu/perf_event_intel_pt.c index 868e11943..49e35d003 100644 --- a/kernel/arch/x86/kernel/cpu/perf_event_intel_pt.c +++ b/kernel/arch/x86/kernel/cpu/perf_event_intel_pt.c @@ -694,6 +694,7 @@ static int pt_buffer_reset_markers(struct pt_buffer *buf, /* clear STOP and INT from current entry */ buf->topa_index[buf->stop_pos]->stop = 0; + buf->topa_index[buf->stop_pos]->intr = 0; buf->topa_index[buf->intr_pos]->intr = 0; /* how many pages till the STOP marker */ @@ -718,6 +719,7 @@ static int pt_buffer_reset_markers(struct pt_buffer *buf, buf->intr_pos = idx; buf->topa_index[buf->stop_pos]->stop = 1; + buf->topa_index[buf->stop_pos]->intr = 1; buf->topa_index[buf->intr_pos]->intr = 1; return 0; diff --git a/kernel/arch/x86/kernel/cpu/perf_event_intel_rapl.c b/kernel/arch/x86/kernel/cpu/perf_event_intel_rapl.c index ed446bdcb..d2ac364e2 100644 --- a/kernel/arch/x86/kernel/cpu/perf_event_intel_rapl.c +++ b/kernel/arch/x86/kernel/cpu/perf_event_intel_rapl.c @@ -117,7 +117,7 @@ static struct perf_pmu_events_attr event_attr_##v = { \ }; struct rapl_pmu { - spinlock_t lock; + raw_spinlock_t lock; int n_active; /* number of active events */ struct list_head active_list; struct pmu *pmu; /* pointer to rapl_pmu_class */ @@ -220,13 +220,13 @@ static enum hrtimer_restart rapl_hrtimer_handle(struct hrtimer *hrtimer) if (!pmu->n_active) return HRTIMER_NORESTART; - spin_lock_irqsave(&pmu->lock, flags); + raw_spin_lock_irqsave(&pmu->lock, flags); list_for_each_entry(event, &pmu->active_list, active_entry) { rapl_event_update(event); } - spin_unlock_irqrestore(&pmu->lock, flags); + raw_spin_unlock_irqrestore(&pmu->lock, flags); hrtimer_forward_now(hrtimer, pmu->timer_interval); @@ -263,9 +263,9 @@ static void rapl_pmu_event_start(struct perf_event *event, int mode) struct rapl_pmu *pmu = __this_cpu_read(rapl_pmu); unsigned long flags; - spin_lock_irqsave(&pmu->lock, flags); + raw_spin_lock_irqsave(&pmu->lock, flags); __rapl_pmu_event_start(pmu, event); - spin_unlock_irqrestore(&pmu->lock, flags); + raw_spin_unlock_irqrestore(&pmu->lock, flags); } static void rapl_pmu_event_stop(struct perf_event *event, int mode) @@ -274,7 +274,7 @@ static void rapl_pmu_event_stop(struct perf_event *event, int mode) struct hw_perf_event *hwc = &event->hw; unsigned long flags; - spin_lock_irqsave(&pmu->lock, flags); + raw_spin_lock_irqsave(&pmu->lock, flags); /* mark event as deactivated and stopped */ if (!(hwc->state & PERF_HES_STOPPED)) { @@ -299,7 +299,7 @@ static void rapl_pmu_event_stop(struct perf_event *event, int mode) hwc->state |= PERF_HES_UPTODATE; } - spin_unlock_irqrestore(&pmu->lock, flags); + raw_spin_unlock_irqrestore(&pmu->lock, flags); } static int rapl_pmu_event_add(struct perf_event *event, int mode) @@ -308,14 +308,14 @@ static int rapl_pmu_event_add(struct perf_event *event, int mode) struct hw_perf_event *hwc = &event->hw; unsigned long flags; - spin_lock_irqsave(&pmu->lock, flags); + raw_spin_lock_irqsave(&pmu->lock, flags); hwc->state = PERF_HES_UPTODATE | PERF_HES_STOPPED; if (mode & PERF_EF_START) __rapl_pmu_event_start(pmu, event); - spin_unlock_irqrestore(&pmu->lock, flags); + raw_spin_unlock_irqrestore(&pmu->lock, flags); return 0; } @@ -603,7 +603,7 @@ static int rapl_cpu_prepare(int cpu) pmu = kzalloc_node(sizeof(*pmu), GFP_KERNEL, cpu_to_node(cpu)); if (!pmu) return -1; - spin_lock_init(&pmu->lock); + raw_spin_lock_init(&pmu->lock); INIT_LIST_HEAD(&pmu->active_list); diff --git a/kernel/arch/x86/kernel/cpu/perf_event_knc.c b/kernel/arch/x86/kernel/cpu/perf_event_knc.c index 5b0c232d1..b931095e8 100644 --- a/kernel/arch/x86/kernel/cpu/perf_event_knc.c +++ b/kernel/arch/x86/kernel/cpu/perf_event_knc.c @@ -263,7 +263,9 @@ again: goto again; done: - knc_pmu_enable_all(0); + /* Only restore PMU state when it's active. See x86_pmu_disable(). */ + if (cpuc->enabled) + knc_pmu_enable_all(0); return handled; } diff --git a/kernel/arch/x86/kernel/e820.c b/kernel/arch/x86/kernel/e820.c index 569c1e4f9..52a2526c3 100644 --- a/kernel/arch/x86/kernel/e820.c +++ b/kernel/arch/x86/kernel/e820.c @@ -753,7 +753,7 @@ u64 __init early_reserve_e820(u64 size, u64 align) /* * Find the highest page frame number we have available */ -static unsigned long __init e820_end_pfn(unsigned long limit_pfn) +static unsigned long __init e820_end_pfn(unsigned long limit_pfn, unsigned type) { int i; unsigned long last_pfn = 0; @@ -764,11 +764,7 @@ static unsigned long __init e820_end_pfn(unsigned long limit_pfn) unsigned long start_pfn; unsigned long end_pfn; - /* - * Persistent memory is accounted as ram for purposes of - * establishing max_pfn and mem_map. - */ - if (ei->type != E820_RAM && ei->type != E820_PRAM) + if (ei->type != type) continue; start_pfn = ei->addr >> PAGE_SHIFT; @@ -793,12 +789,12 @@ static unsigned long __init e820_end_pfn(unsigned long limit_pfn) } unsigned long __init e820_end_of_ram_pfn(void) { - return e820_end_pfn(MAX_ARCH_PFN); + return e820_end_pfn(MAX_ARCH_PFN, E820_RAM); } unsigned long __init e820_end_of_low_ram_pfn(void) { - return e820_end_pfn(1UL << (32-PAGE_SHIFT)); + return e820_end_pfn(1UL << (32 - PAGE_SHIFT), E820_RAM); } static void early_panic(char *msg) diff --git a/kernel/arch/x86/kernel/early-quirks.c b/kernel/arch/x86/kernel/early-quirks.c index db9a675e7..a257d6077 100644 --- a/kernel/arch/x86/kernel/early-quirks.c +++ b/kernel/arch/x86/kernel/early-quirks.c @@ -11,7 +11,11 @@ #include <linux/pci.h> #include <linux/acpi.h> +#include <linux/delay.h> +#include <linux/dmi.h> #include <linux/pci_ids.h> +#include <linux/bcma/bcma.h> +#include <linux/bcma/bcma_regs.h> #include <drm/i915_drm.h> #include <asm/pci-direct.h> #include <asm/dma.h> @@ -21,6 +25,9 @@ #include <asm/iommu.h> #include <asm/gart.h> #include <asm/irq_remapping.h> +#include <asm/early_ioremap.h> + +#define dev_err(msg) pr_err("pci 0000:%02x:%02x.%d: %s", bus, slot, func, msg) static void __init fix_hypertransport_config(int num, int slot, int func) { @@ -76,6 +83,13 @@ static void __init nvidia_bugs(int num, int slot, int func) #ifdef CONFIG_ACPI #ifdef CONFIG_X86_IO_APIC /* + * Only applies to Nvidia root ports (bus 0) and not to + * Nvidia graphics cards with PCI ports on secondary buses. + */ + if (num) + return; + + /* * All timer overrides on Nvidia are * wrong unless HPET is enabled. * Unfortunately that's not true on many Asus boards. @@ -317,12 +331,11 @@ static u32 __init i85x_stolen_base(int num, int slot, int func, size_t stolen_si static u32 __init i865_stolen_base(int num, int slot, int func, size_t stolen_size) { - /* - * FIXME is the graphics stolen memory region - * always at TOUD? Ie. is it always the last - * one to be allocated by the BIOS? - */ - return read_pci_config_16(0, 0, 0, I865_TOUD) << 16; + u16 toud = 0; + + toud = read_pci_config_16(0, 0, 0, I865_TOUD); + + return (phys_addr_t)(toud << 16) + i845_tseg_size(); } static size_t __init i830_stolen_size(int num, int slot, int func) @@ -589,6 +602,61 @@ static void __init force_disable_hpet(int num, int slot, int func) #endif } +#define BCM4331_MMIO_SIZE 16384 +#define BCM4331_PM_CAP 0x40 +#define bcma_aread32(reg) ioread32(mmio + 1 * BCMA_CORE_SIZE + reg) +#define bcma_awrite32(reg, val) iowrite32(val, mmio + 1 * BCMA_CORE_SIZE + reg) + +static void __init apple_airport_reset(int bus, int slot, int func) +{ + void __iomem *mmio; + u16 pmcsr; + u64 addr; + int i; + + if (!dmi_match(DMI_SYS_VENDOR, "Apple Inc.")) + return; + + /* Card may have been put into PCI_D3hot by grub quirk */ + pmcsr = read_pci_config_16(bus, slot, func, BCM4331_PM_CAP + PCI_PM_CTRL); + + if ((pmcsr & PCI_PM_CTRL_STATE_MASK) != PCI_D0) { + pmcsr &= ~PCI_PM_CTRL_STATE_MASK; + write_pci_config_16(bus, slot, func, BCM4331_PM_CAP + PCI_PM_CTRL, pmcsr); + mdelay(10); + + pmcsr = read_pci_config_16(bus, slot, func, BCM4331_PM_CAP + PCI_PM_CTRL); + if ((pmcsr & PCI_PM_CTRL_STATE_MASK) != PCI_D0) { + dev_err("Cannot power up Apple AirPort card\n"); + return; + } + } + + addr = read_pci_config(bus, slot, func, PCI_BASE_ADDRESS_0); + addr |= (u64)read_pci_config(bus, slot, func, PCI_BASE_ADDRESS_1) << 32; + addr &= PCI_BASE_ADDRESS_MEM_MASK; + + mmio = early_ioremap(addr, BCM4331_MMIO_SIZE); + if (!mmio) { + dev_err("Cannot iomap Apple AirPort card\n"); + return; + } + + pr_info("Resetting Apple AirPort card (left enabled by EFI)\n"); + + for (i = 0; bcma_aread32(BCMA_RESET_ST) && i < 30; i++) + udelay(10); + + bcma_awrite32(BCMA_RESET_CTL, BCMA_RESET_CTL_RESET); + bcma_aread32(BCMA_RESET_CTL); + udelay(1); + + bcma_awrite32(BCMA_RESET_CTL, 0); + bcma_aread32(BCMA_RESET_CTL); + udelay(10); + + early_iounmap(mmio, BCM4331_MMIO_SIZE); +} #define QFLAG_APPLY_ONCE 0x1 #define QFLAG_APPLIED 0x2 @@ -602,12 +670,6 @@ struct chipset { void (*f)(int num, int slot, int func); }; -/* - * Only works for devices on the root bus. If you add any devices - * not on bus 0 readd another loop level in early_quirks(). But - * be careful because at least the Nvidia quirk here relies on - * only matching on bus 0. - */ static struct chipset early_qrk[] __initdata = { { PCI_VENDOR_ID_NVIDIA, PCI_ANY_ID, PCI_CLASS_BRIDGE_PCI, PCI_ANY_ID, QFLAG_APPLY_ONCE, nvidia_bugs }, @@ -637,9 +699,13 @@ static struct chipset early_qrk[] __initdata = { */ { PCI_VENDOR_ID_INTEL, 0x0f00, PCI_CLASS_BRIDGE_HOST, PCI_ANY_ID, 0, force_disable_hpet}, + { PCI_VENDOR_ID_BROADCOM, 0x4331, + PCI_CLASS_NETWORK_OTHER, PCI_ANY_ID, 0, apple_airport_reset}, {} }; +static void __init early_pci_scan_bus(int bus); + /** * check_dev_quirk - apply early quirks to a given PCI device * @num: bus number @@ -648,7 +714,7 @@ static struct chipset early_qrk[] __initdata = { * * Check the vendor & device ID against the early quirks table. * - * If the device is single function, let early_quirks() know so we don't + * If the device is single function, let early_pci_scan_bus() know so we don't * poke at this device again. */ static int __init check_dev_quirk(int num, int slot, int func) @@ -657,6 +723,7 @@ static int __init check_dev_quirk(int num, int slot, int func) u16 vendor; u16 device; u8 type; + u8 sec; int i; class = read_pci_config_16(num, slot, func, PCI_CLASS_DEVICE); @@ -684,25 +751,36 @@ static int __init check_dev_quirk(int num, int slot, int func) type = read_pci_config_byte(num, slot, func, PCI_HEADER_TYPE); + + if ((type & 0x7f) == PCI_HEADER_TYPE_BRIDGE) { + sec = read_pci_config_byte(num, slot, func, PCI_SECONDARY_BUS); + if (sec > num) + early_pci_scan_bus(sec); + } + if (!(type & 0x80)) return -1; return 0; } -void __init early_quirks(void) +static void __init early_pci_scan_bus(int bus) { int slot, func; - if (!early_pci_allowed()) - return; - /* Poor man's PCI discovery */ - /* Only scan the root bus */ for (slot = 0; slot < 32; slot++) for (func = 0; func < 8; func++) { /* Only probe function 0 on single fn devices */ - if (check_dev_quirk(0, slot, func)) + if (check_dev_quirk(bus, slot, func)) break; } } + +void __init early_quirks(void) +{ + if (!early_pci_allowed()) + return; + + early_pci_scan_bus(0); +} diff --git a/kernel/arch/x86/kernel/head_32.S b/kernel/arch/x86/kernel/head_32.S index 6bc9ae24b..8f1a3f443 100644 --- a/kernel/arch/x86/kernel/head_32.S +++ b/kernel/arch/x86/kernel/head_32.S @@ -571,7 +571,7 @@ early_idt_handler_common: movl %eax,%ds movl %eax,%es - cmpl $(__KERNEL_CS),32(%esp) + cmpw $(__KERNEL_CS),32(%esp) jne 10f leal 28(%esp),%eax # Pointer to %eip diff --git a/kernel/arch/x86/kernel/hpet.c b/kernel/arch/x86/kernel/hpet.c index b8e6ff5cd..acc9b8f19 100644 --- a/kernel/arch/x86/kernel/hpet.c +++ b/kernel/arch/x86/kernel/hpet.c @@ -351,6 +351,7 @@ static int hpet_resume(struct clock_event_device *evt, int timer) } else { struct hpet_dev *hdev = EVT_TO_HPET_DEV(evt); + irq_domain_deactivate_irq(irq_get_irq_data(hdev->irq)); irq_domain_activate_irq(irq_get_irq_data(hdev->irq)); disable_irq(hdev->irq); irq_set_affinity(hdev->irq, cpumask_of(hdev->cpu)); diff --git a/kernel/arch/x86/kernel/ioport.c b/kernel/arch/x86/kernel/ioport.c index 37dae792d..589b3193f 100644 --- a/kernel/arch/x86/kernel/ioport.c +++ b/kernel/arch/x86/kernel/ioport.c @@ -96,9 +96,14 @@ asmlinkage long sys_ioperm(unsigned long from, unsigned long num, int turn_on) SYSCALL_DEFINE1(iopl, unsigned int, level) { struct pt_regs *regs = current_pt_regs(); - unsigned int old = (regs->flags >> 12) & 3; struct thread_struct *t = ¤t->thread; + /* + * Careful: the IOPL bits in regs->flags are undefined under Xen PV + * and changing them has no effect. + */ + unsigned int old = t->iopl >> X86_EFLAGS_IOPL_BIT; + if (level > 3) return -EINVAL; /* Trying to gain more privileges? */ @@ -106,8 +111,9 @@ SYSCALL_DEFINE1(iopl, unsigned int, level) if (!capable(CAP_SYS_RAWIO)) return -EPERM; } - regs->flags = (regs->flags & ~X86_EFLAGS_IOPL) | (level << 12); - t->iopl = level << 12; + regs->flags = (regs->flags & ~X86_EFLAGS_IOPL) | + (level << X86_EFLAGS_IOPL_BIT); + t->iopl = level << X86_EFLAGS_IOPL_BIT; set_iopl_mask(t->iopl); return 0; diff --git a/kernel/arch/x86/kernel/kprobes/core.c b/kernel/arch/x86/kernel/kprobes/core.c index 1deffe6cc..023c442c3 100644 --- a/kernel/arch/x86/kernel/kprobes/core.c +++ b/kernel/arch/x86/kernel/kprobes/core.c @@ -959,7 +959,19 @@ int kprobe_fault_handler(struct pt_regs *regs, int trapnr) * normal page fault. */ regs->ip = (unsigned long)cur->addr; + /* + * Trap flag (TF) has been set here because this fault + * happened where the single stepping will be done. + * So clear it by resetting the current kprobe: + */ + regs->flags &= ~X86_EFLAGS_TF; + + /* + * If the TF flag was set before the kprobe hit, + * don't touch it: + */ regs->flags |= kcb->kprobe_old_flags; + if (kcb->kprobe_status == KPROBE_REENTER) restore_previous_kprobe(kcb); else diff --git a/kernel/arch/x86/kernel/mcount_64.S b/kernel/arch/x86/kernel/mcount_64.S index 87e1762e2..5d9afbcb6 100644 --- a/kernel/arch/x86/kernel/mcount_64.S +++ b/kernel/arch/x86/kernel/mcount_64.S @@ -180,7 +180,8 @@ GLOBAL(ftrace_graph_call) jmp ftrace_stub #endif -GLOBAL(ftrace_stub) +/* This is weak to keep gas from relaxing the jumps */ +WEAK(ftrace_stub) retq END(ftrace_caller) diff --git a/kernel/arch/x86/kernel/nmi.c b/kernel/arch/x86/kernel/nmi.c index 697f90db0..424aec4a4 100644 --- a/kernel/arch/x86/kernel/nmi.c +++ b/kernel/arch/x86/kernel/nmi.c @@ -231,7 +231,7 @@ pci_serr_error(unsigned char reason, struct pt_regs *regs) #endif if (panic_on_unrecovered_nmi) - panic("NMI: Not continuing"); + nmi_panic(regs, "NMI: Not continuing"); pr_emerg("Dazed and confused, but trying to continue\n"); @@ -255,8 +255,16 @@ io_check_error(unsigned char reason, struct pt_regs *regs) reason, smp_processor_id()); show_regs(regs); - if (panic_on_io_nmi) - panic("NMI IOCK error: Not continuing"); + if (panic_on_io_nmi) { + nmi_panic(regs, "NMI IOCK error: Not continuing"); + + /* + * If we end up here, it means we have received an NMI while + * processing panic(). Simply return without delaying and + * re-enabling NMIs. + */ + return; + } /* Re-enable the IOCK line, wait for a few seconds */ reason = (reason & NMI_REASON_CLEAR_MASK) | NMI_REASON_CLEAR_IOCHK; @@ -297,7 +305,7 @@ unknown_nmi_error(unsigned char reason, struct pt_regs *regs) pr_emerg("Do you have a strange power saving mode enabled?\n"); if (unknown_nmi_panic || panic_on_unrecovered_nmi) - panic("NMI: Not continuing"); + nmi_panic(regs, "NMI: Not continuing"); pr_emerg("Dazed and confused, but trying to continue\n"); } diff --git a/kernel/arch/x86/kernel/paravirt.c b/kernel/arch/x86/kernel/paravirt.c index c2130aef3..f534a0e3a 100644 --- a/kernel/arch/x86/kernel/paravirt.c +++ b/kernel/arch/x86/kernel/paravirt.c @@ -55,12 +55,12 @@ asm (".pushsection .entry.text, \"ax\"\n" ".popsection"); /* identity function, which can be inlined */ -u32 _paravirt_ident_32(u32 x) +u32 notrace _paravirt_ident_32(u32 x) { return x; } -u64 _paravirt_ident_64(u64 x) +u64 notrace _paravirt_ident_64(u64 x) { return x; } diff --git a/kernel/arch/x86/kernel/process_64.c b/kernel/arch/x86/kernel/process_64.c index e835d263a..4cbb60fbf 100644 --- a/kernel/arch/x86/kernel/process_64.c +++ b/kernel/arch/x86/kernel/process_64.c @@ -48,6 +48,7 @@ #include <asm/syscalls.h> #include <asm/debugreg.h> #include <asm/switch_to.h> +#include <asm/xen/hypervisor.h> asmlinkage extern void ret_from_fork(void); @@ -411,6 +412,17 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) task_thread_info(prev_p)->flags & _TIF_WORK_CTXSW_PREV)) __switch_to_xtra(prev_p, next_p, tss); +#ifdef CONFIG_XEN + /* + * On Xen PV, IOPL bits in pt_regs->flags have no effect, and + * current_pt_regs()->flags may not match the current task's + * intended IOPL. We need to switch it manually. + */ + if (unlikely(static_cpu_has(X86_FEATURE_XENPV) && + prev->iopl != next->iopl)) + xen_set_iopl_mask(next->iopl); +#endif + if (static_cpu_has_bug(X86_BUG_SYSRET_SS_ATTRS)) { /* * AMD CPUs have a misfeature: SYSRET sets the SS selector but diff --git a/kernel/arch/x86/kernel/ptrace.c b/kernel/arch/x86/kernel/ptrace.c index 558f50ede..479a409dd 100644 --- a/kernel/arch/x86/kernel/ptrace.c +++ b/kernel/arch/x86/kernel/ptrace.c @@ -188,8 +188,8 @@ unsigned long kernel_stack_pointer(struct pt_regs *regs) return sp; prev_esp = (u32 *)(context); - if (prev_esp) - return (unsigned long)prev_esp; + if (*prev_esp) + return (unsigned long)*prev_esp; return (unsigned long)regs; } diff --git a/kernel/arch/x86/kernel/pvclock.c b/kernel/arch/x86/kernel/pvclock.c index 2f355d229..bf0ce7573 100644 --- a/kernel/arch/x86/kernel/pvclock.c +++ b/kernel/arch/x86/kernel/pvclock.c @@ -66,6 +66,8 @@ u8 pvclock_read_flags(struct pvclock_vcpu_time_info *src) do { version = __pvclock_read_cycles(src, &ret, &flags); + /* Make sure that the version double-check is last. */ + smp_rmb(); } while ((src->version & 1) || version != src->version); return flags & valid_flags; @@ -80,6 +82,8 @@ cycle_t pvclock_clocksource_read(struct pvclock_vcpu_time_info *src) do { version = __pvclock_read_cycles(src, &ret, &flags); + /* Make sure that the version double-check is last. */ + smp_rmb(); } while ((src->version & 1) || version != src->version); if (unlikely((flags & PVCLOCK_GUEST_STOPPED) != 0)) { diff --git a/kernel/arch/x86/kernel/reboot.c b/kernel/arch/x86/kernel/reboot.c index f660d63f4..8384207ad 100644 --- a/kernel/arch/x86/kernel/reboot.c +++ b/kernel/arch/x86/kernel/reboot.c @@ -726,6 +726,7 @@ static int crashing_cpu; static nmi_shootdown_cb shootdown_callback; static atomic_t waiting_for_crash_ipi; +static int crash_ipi_issued; static int crash_nmi_callback(unsigned int val, struct pt_regs *regs) { @@ -788,6 +789,9 @@ void nmi_shootdown_cpus(nmi_shootdown_cb callback) smp_send_nmi_allbutself(); + /* Kick CPUs looping in NMI context. */ + WRITE_ONCE(crash_ipi_issued, 1); + msecs = 1000; /* Wait at most a second for the other cpus to stop */ while ((atomic_read(&waiting_for_crash_ipi) > 0) && msecs) { mdelay(1); @@ -796,6 +800,22 @@ void nmi_shootdown_cpus(nmi_shootdown_cb callback) /* Leave the nmi callback set */ } + +/* Override the weak function in kernel/panic.c */ +void nmi_panic_self_stop(struct pt_regs *regs) +{ + while (1) { + /* + * Wait for the crash dumping IPI to be issued, and then + * call its callback directly. + */ + if (READ_ONCE(crash_ipi_issued)) + crash_nmi_callback(0, regs); /* Don't return */ + + cpu_relax(); + } +} + #else /* !CONFIG_SMP */ void nmi_shootdown_cpus(nmi_shootdown_cb callback) { diff --git a/kernel/arch/x86/kernel/sysfb_efi.c b/kernel/arch/x86/kernel/sysfb_efi.c index b285d4e8c..5da924bbf 100644 --- a/kernel/arch/x86/kernel/sysfb_efi.c +++ b/kernel/arch/x86/kernel/sysfb_efi.c @@ -106,14 +106,24 @@ static int __init efifb_set_system(const struct dmi_system_id *id) continue; for (i = 0; i < DEVICE_COUNT_RESOURCE; i++) { resource_size_t start, end; + unsigned long flags; + + flags = pci_resource_flags(dev, i); + if (!(flags & IORESOURCE_MEM)) + continue; + + if (flags & IORESOURCE_UNSET) + continue; + + if (pci_resource_len(dev, i) == 0) + continue; start = pci_resource_start(dev, i); - if (start == 0) - break; end = pci_resource_end(dev, i); if (screen_info.lfb_base >= start && screen_info.lfb_base < end) { found_bar = 1; + break; } } } diff --git a/kernel/arch/x86/kernel/traps.c b/kernel/arch/x86/kernel/traps.c index ade185a46..679302c31 100644 --- a/kernel/arch/x86/kernel/traps.c +++ b/kernel/arch/x86/kernel/traps.c @@ -109,6 +109,12 @@ static inline void preempt_conditional_cli(struct pt_regs *regs) preempt_count_dec(); } +/* + * In IST context, we explicitly disable preemption. This serves two + * purposes: it makes it much less likely that we would accidentally + * schedule in IST context and it will force a warning if we somehow + * manage to schedule by accident. + */ void ist_enter(struct pt_regs *regs) { if (user_mode(regs)) { @@ -123,13 +129,7 @@ void ist_enter(struct pt_regs *regs) rcu_nmi_enter(); } - /* - * We are atomic because we're on the IST stack; or we're on - * x86_32, in which case we still shouldn't schedule; or we're - * on x86_64 and entered from user mode, in which case we're - * still atomic unless ist_begin_non_atomic is called. - */ - preempt_count_add(HARDIRQ_OFFSET); + preempt_disable(); /* This code is a bit fragile. Test it. */ RCU_LOCKDEP_WARN(!rcu_is_watching(), "ist_enter didn't work"); @@ -137,7 +137,7 @@ void ist_enter(struct pt_regs *regs) void ist_exit(struct pt_regs *regs) { - preempt_count_sub(HARDIRQ_OFFSET); + preempt_enable_no_resched(); if (!user_mode(regs)) rcu_nmi_exit(); @@ -168,7 +168,7 @@ void ist_begin_non_atomic(struct pt_regs *regs) BUG_ON((unsigned long)(current_top_of_stack() - current_stack_pointer()) >= THREAD_SIZE); - preempt_count_sub(HARDIRQ_OFFSET); + preempt_enable_no_resched(); } /** @@ -178,7 +178,7 @@ void ist_begin_non_atomic(struct pt_regs *regs) */ void ist_end_non_atomic(void) { - preempt_count_add(HARDIRQ_OFFSET); + preempt_disable(); } static nokprobe_inline int diff --git a/kernel/arch/x86/kernel/tsc_msr.c b/kernel/arch/x86/kernel/tsc_msr.c index 92ae6acac..6aa0f4d9e 100644 --- a/kernel/arch/x86/kernel/tsc_msr.c +++ b/kernel/arch/x86/kernel/tsc_msr.c @@ -92,7 +92,7 @@ unsigned long try_msr_calibrate_tsc(void) if (freq_desc_tables[cpu_index].msr_plat) { rdmsr(MSR_PLATFORM_INFO, lo, hi); - ratio = (lo >> 8) & 0x1f; + ratio = (lo >> 8) & 0xff; } else { rdmsr(MSR_IA32_PERF_STATUS, lo, hi); ratio = (hi >> 8) & 0x1f; diff --git a/kernel/arch/x86/kernel/uprobes.c b/kernel/arch/x86/kernel/uprobes.c index bf4db6eae..c6aace2bb 100644 --- a/kernel/arch/x86/kernel/uprobes.c +++ b/kernel/arch/x86/kernel/uprobes.c @@ -357,20 +357,22 @@ static void riprel_analyze(struct arch_uprobe *auprobe, struct insn *insn) *cursor &= 0xfe; } /* - * Similar treatment for VEX3 prefix. - * TODO: add XOP/EVEX treatment when insn decoder supports them + * Similar treatment for VEX3/EVEX prefix. + * TODO: add XOP treatment when insn decoder supports them */ - if (insn->vex_prefix.nbytes == 3) { + if (insn->vex_prefix.nbytes >= 3) { /* * vex2: c5 rvvvvLpp (has no b bit) * vex3/xop: c4/8f rxbmmmmm wvvvvLpp * evex: 62 rxbR00mm wvvvv1pp zllBVaaa - * (evex will need setting of both b and x since - * in non-sib encoding evex.x is 4th bit of MODRM.rm) - * Setting VEX3.b (setting because it has inverted meaning): + * Setting VEX3.b (setting because it has inverted meaning). + * Setting EVEX.x since (in non-SIB encoding) EVEX.x + * is the 4th bit of MODRM.rm, and needs the same treatment. + * For VEX3-encoded insns, VEX3.x value has no effect in + * non-SIB encoding, the change is superfluous but harmless. */ cursor = auprobe->insn + insn_offset_vex_prefix(insn) + 1; - *cursor |= 0x20; + *cursor |= 0x60; } /* @@ -415,12 +417,10 @@ static void riprel_analyze(struct arch_uprobe *auprobe, struct insn *insn) reg = MODRM_REG(insn); /* Fetch modrm.reg */ reg2 = 0xff; /* Fetch vex.vvvv */ - if (insn->vex_prefix.nbytes == 2) - reg2 = insn->vex_prefix.bytes[1]; - else if (insn->vex_prefix.nbytes == 3) + if (insn->vex_prefix.nbytes) reg2 = insn->vex_prefix.bytes[2]; /* - * TODO: add XOP, EXEV vvvv reading. + * TODO: add XOP vvvv reading. * * vex.vvvv field is in bits 6-3, bits are inverted. * But in 32-bit mode, high-order bit may be ignored. |