diff options
Diffstat (limited to 'arch/x86/kernel/apic/apic.c')
| -rw-r--r-- | arch/x86/kernel/apic/apic.c | 1412 |
1 files changed, 663 insertions, 749 deletions
diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c index b7bcdd781651..d93f87f29d03 100644 --- a/arch/x86/kernel/apic/apic.c +++ b/arch/x86/kernel/apic/apic.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0-only /* * Local APIC handling, local APIC timers * @@ -18,6 +19,7 @@ #include <linux/kernel_stat.h> #include <linux/mc146818rtc.h> #include <linux/acpi_pmtmr.h> +#include <linux/bitmap.h> #include <linux/clockchips.h> #include <linux/interrupt.h> #include <linux/memblock.h> @@ -34,18 +36,23 @@ #include <linux/dmi.h> #include <linux/smp.h> #include <linux/mm.h> +#include <linux/kvm_types.h> + +#include <xen/xen.h> #include <asm/trace/irq_vectors.h> #include <asm/irq_remapping.h> +#include <asm/pc-conf-reg.h> #include <asm/perf_event.h> #include <asm/x86_init.h> -#include <asm/pgalloc.h> #include <linux/atomic.h> +#include <asm/barrier.h> #include <asm/mpspec.h> #include <asm/i8259.h> #include <asm/proto.h> #include <asm/traps.h> #include <asm/apic.h> +#include <asm/acpi.h> #include <asm/io_apic.h> #include <asm/desc.h> #include <asm/hpet.h> @@ -53,67 +60,44 @@ #include <asm/time.h> #include <asm/smp.h> #include <asm/mce.h> +#include <asm/msr.h> #include <asm/tsc.h> #include <asm/hypervisor.h> #include <asm/cpu_device_id.h> #include <asm/intel-family.h> #include <asm/irq_regs.h> +#include <asm/cpu.h> -unsigned int num_processors; - -unsigned disabled_cpus; +#include "local.h" /* Processor that is doing the boot up */ -unsigned int boot_cpu_physical_apicid = -1U; +u32 boot_cpu_physical_apicid __ro_after_init = BAD_APICID; EXPORT_SYMBOL_GPL(boot_cpu_physical_apicid); -u8 boot_cpu_apic_version; - -/* - * The highest APIC ID seen during enumeration. - */ -static unsigned int max_physical_apicid; - -/* - * Bitmask of physically existing CPUs: - */ -physid_mask_t phys_cpu_present_map; - -/* - * Processor to be disabled specified by kernel parameter - * disable_cpu_apicid=<int>, mostly used for the kdump 2nd kernel to - * avoid undefined behaviour caused by sending INIT from AP to BSP. - */ -static unsigned int disabled_cpu_apicid __read_mostly = BAD_APICID; +u8 boot_cpu_apic_version __ro_after_init; /* * This variable controls which CPUs receive external NMIs. By default, * external NMIs are delivered only to the BSP. */ -static int apic_extnmi = APIC_EXTNMI_BSP; +static int apic_extnmi __ro_after_init = APIC_EXTNMI_BSP; /* - * Map cpu index to physical APIC ID + * Hypervisor supports 15 bits of APIC ID in MSI Extended Destination ID */ -DEFINE_EARLY_PER_CPU_READ_MOSTLY(u16, x86_cpu_to_apicid, BAD_APICID); -DEFINE_EARLY_PER_CPU_READ_MOSTLY(u16, x86_bios_cpu_apicid, BAD_APICID); -DEFINE_EARLY_PER_CPU_READ_MOSTLY(u32, x86_cpu_to_acpiid, U32_MAX); -EXPORT_EARLY_PER_CPU_SYMBOL(x86_cpu_to_apicid); -EXPORT_EARLY_PER_CPU_SYMBOL(x86_bios_cpu_apicid); -EXPORT_EARLY_PER_CPU_SYMBOL(x86_cpu_to_acpiid); +static bool virt_ext_dest_id __ro_after_init; -#ifdef CONFIG_X86_32 +/* For parallel bootup. */ +unsigned long apic_mmio_base __ro_after_init; -/* - * On x86_32, the mapping between cpu and logical apicid may vary - * depending on apic in use. The following early percpu variable is - * used for the mapping. This is where the behaviors of x86_64 and 32 - * actually diverge. Let's keep it ugly for now. - */ -DEFINE_EARLY_PER_CPU_READ_MOSTLY(int, x86_cpu_to_logical_apicid, BAD_APICID); +static inline bool apic_accessible(void) +{ + return x2apic_mode || apic_mmio_base; +} +#ifdef CONFIG_X86_32 /* Local APIC was disabled by the BIOS and enabled by the kernel */ -static int enabled_via_apicbase; +static int enabled_via_apicbase __ro_after_init; /* * Handle interrupt mode configuration register (IMCR). @@ -125,18 +109,14 @@ static int enabled_via_apicbase; */ static inline void imcr_pic_to_apic(void) { - /* select IMCR register */ - outb(0x70, 0x22); /* NMI and 8259 INTR go through APIC */ - outb(0x01, 0x23); + pc_conf_set(PC_CONF_MPS_IMCR, 0x01); } static inline void imcr_apic_to_pic(void) { - /* select IMCR register */ - outb(0x70, 0x22); /* NMI and 8259 INTR go directly to BSP */ - outb(0x00, 0x23); + pc_conf_set(PC_CONF_MPS_IMCR, 0x00); } #endif @@ -166,40 +146,39 @@ static __init int setup_apicpmtimer(char *s) { apic_calibrate_pmtmr = 1; notsc_setup(NULL); - return 0; + return 1; } __setup("apicpmtimer", setup_apicpmtimer); #endif -unsigned long mp_lapic_addr; -int disable_apic; +static unsigned long mp_lapic_addr __ro_after_init; +bool apic_is_disabled __ro_after_init; /* Disable local APIC timer from the kernel commandline or via dmi quirk */ static int disable_apic_timer __initdata; /* Local APIC timer works in C2 */ -int local_apic_timer_c2_ok; +int local_apic_timer_c2_ok __ro_after_init; EXPORT_SYMBOL_GPL(local_apic_timer_c2_ok); /* * Debug level, exported for io_apic.c */ -unsigned int apic_verbosity; +int apic_verbosity __ro_after_init; -int pic_mode; +int pic_mode __ro_after_init; /* Have we found an MP table */ -int smp_found_config; +int smp_found_config __ro_after_init; static struct resource lapic_resource = { .name = "Local APIC", .flags = IORESOURCE_MEM | IORESOURCE_BUSY, }; -unsigned int lapic_timer_frequency = 0; +/* Measured in ticks per HZ. */ +unsigned int lapic_timer_period = 0; static void apic_pm_activate(void); -static unsigned long apic_phys; - /* * Get the LAPIC version */ @@ -239,31 +218,7 @@ static int modern_apic(void) */ static void __init apic_disable(void) { - pr_info("APIC: switched to apic NOOP\n"); - apic = &apic_noop; -} - -void native_apic_wait_icr_idle(void) -{ - while (apic_read(APIC_ICR) & APIC_ICR_BUSY) - cpu_relax(); -} - -u32 native_safe_apic_wait_icr_idle(void) -{ - u32 send_status; - int timeout; - - timeout = 0; - do { - send_status = apic_read(APIC_ICR) & APIC_ICR_BUSY; - if (!send_status) - break; - inc_irq_stat(icr_read_retry_count); - udelay(100); - } while (timeout++ < 1000); - - return send_status; + apic_install_driver(&apic_noop); } void native_apic_icr_write(u32 low, u32 id) @@ -271,7 +226,7 @@ void native_apic_icr_write(u32 low, u32 id) unsigned long flags; local_irq_save(flags); - apic_write(APIC_ICR2, SET_APIC_DEST_FIELD(id)); + apic_write(APIC_ICR2, SET_XAPIC_DEST_FIELD(id)); apic_write(APIC_ICR, low); local_irq_restore(flags); } @@ -286,16 +241,6 @@ u64 native_apic_icr_read(void) return icr1 | ((u64)icr2 << 32); } -#ifdef CONFIG_X86_32 -/** - * get_physical_broadcast - Get number of physical broadcast IDs - */ -int get_physical_broadcast(void) -{ - return modern_apic() ? 0xff : 0xf; -} -#endif - /** * lapic_get_maxlvt - get the maximum number of local vector table entries */ @@ -316,6 +261,9 @@ int lapic_get_maxlvt(void) #define APIC_DIVISOR 16 #define TSC_DIVISOR 8 +/* i82489DX specific */ +#define I82489DX_BASE_DIVIDER (((0x2) << 18)) + /* * This function sets up the local APIC timer, with a timeout of * 'clocks' APIC bus clock. During calibration we actually call @@ -336,8 +284,14 @@ static void __setup_APIC_LVTT(unsigned int clocks, int oneshot, int irqen) else if (boot_cpu_has(X86_FEATURE_TSC_DEADLINE_TIMER)) lvtt_value |= APIC_LVT_TIMER_TSCDEADLINE; + /* + * The i82489DX APIC uses bit 18 and 19 for the base divider. This + * overlaps with bit 18 on integrated APICs, but is not documented + * in the SDM. No problem though. i82489DX equipped systems do not + * have TSC deadline timer. + */ if (!lapic_is_integrated()) - lvtt_value |= SET_APIC_TIMER_BASE(APIC_TIMER_BASE_DIV); + lvtt_value |= I82489DX_BASE_DIVIDER; if (!irqen) lvtt_value |= APIC_LVT_MASKED; @@ -351,8 +305,6 @@ static void __setup_APIC_LVTT(unsigned int clocks, int oneshot, int irqen) * According to Intel, MFENCE can do the serialization here. */ asm volatile("mfence" : : : "memory"); - - printk_once(KERN_DEBUG "TSC deadline timer enabled\n"); return; } @@ -410,10 +362,9 @@ static unsigned int reserve_eilvt_offset(int offset, unsigned int new) if (vector && !eilvt_entry_is_changeable(vector, new)) /* may not change if vectors are different */ return rsvd; - rsvd = atomic_cmpxchg(&eilvt_offsets[offset], rsvd, new); - } while (rsvd != new); + } while (!atomic_try_cmpxchg(&eilvt_offsets[offset], &rsvd, new)); - rsvd &= ~APIC_EILVT_MASKED; + rsvd = new & ~APIC_EILVT_MASKED; if (rsvd && rsvd != vector) pr_info("LVT offset %d assigned for vector 0x%02x\n", offset, rsvd); @@ -473,8 +424,11 @@ static int lapic_next_deadline(unsigned long delta, { u64 tsc; + /* This MSR is special and need a special fence: */ + weak_wrmsr_fence(); + tsc = rdtsc(); - wrmsrl(MSR_IA32_TSC_DEADLINE, tsc + (((u64) delta) * TSC_DIVISOR)); + wrmsrq(MSR_IA32_TSC_DEADLINE, tsc + (((u64) delta) * TSC_DIVISOR)); return 0; } @@ -489,7 +443,19 @@ static int lapic_timer_shutdown(struct clock_event_device *evt) v = apic_read(APIC_LVTT); v |= (APIC_LVT_MASKED | LOCAL_TIMER_VECTOR); apic_write(APIC_LVTT, v); - apic_write(APIC_TMICT, 0); + + /* + * Setting APIC_LVT_MASKED (above) should be enough to tell + * the hardware that this timer will never fire. But AMD + * erratum 411 and some Intel CPU behavior circa 2024 say + * otherwise. Time for belt and suspenders programming: mask + * the timer _and_ zero the counter registers: + */ + if (v & APIC_LVT_TIMER_TSCDEADLINE) + wrmsrq(MSR_IA32_TSC_DEADLINE, 0); + else + apic_write(APIC_TMICT, 0); + return 0; } @@ -500,7 +466,7 @@ lapic_timer_set_periodic_oneshot(struct clock_event_device *evt, bool oneshot) if (evt->features & CLOCK_EVT_FEAT_DUMMY) return 0; - __setup_APIC_LVTT(lapic_timer_frequency, oneshot, 1); + __setup_APIC_LVTT(lapic_timer_period, oneshot, 1); return 0; } @@ -520,7 +486,7 @@ static int lapic_timer_set_oneshot(struct clock_event_device *evt) static void lapic_timer_broadcast(const struct cpumask *mask) { #ifdef CONFIG_SMP - apic->send_IPI_mask(mask, LOCAL_TIMER_VECTOR); + __apic_send_IPI_mask(mask, LOCAL_TIMER_VECTOR); #endif } @@ -545,97 +511,60 @@ static struct clock_event_device lapic_clockevent = { }; static DEFINE_PER_CPU(struct clock_event_device, lapic_events); -#define DEADLINE_MODEL_MATCH_FUNC(model, func) \ - { X86_VENDOR_INTEL, 6, model, X86_FEATURE_ANY, (unsigned long)&func } - -#define DEADLINE_MODEL_MATCH_REV(model, rev) \ - { X86_VENDOR_INTEL, 6, model, X86_FEATURE_ANY, (unsigned long)rev } - -static u32 hsx_deadline_rev(void) -{ - switch (boot_cpu_data.x86_stepping) { - case 0x02: return 0x3a; /* EP */ - case 0x04: return 0x0f; /* EX */ - } - - return ~0U; -} - -static u32 bdx_deadline_rev(void) -{ - switch (boot_cpu_data.x86_stepping) { - case 0x02: return 0x00000011; - case 0x03: return 0x0700000e; - case 0x04: return 0x0f00000c; - case 0x05: return 0x0e000003; - } - - return ~0U; -} - -static u32 skx_deadline_rev(void) -{ - switch (boot_cpu_data.x86_stepping) { - case 0x03: return 0x01000136; - case 0x04: return 0x02000014; - } +static const struct x86_cpu_id deadline_match[] __initconst = { + X86_MATCH_VFM_STEPS(INTEL_HASWELL_X, 0x2, 0x2, 0x3a), /* EP */ + X86_MATCH_VFM_STEPS(INTEL_HASWELL_X, 0x4, 0x4, 0x0f), /* EX */ - if (boot_cpu_data.x86_stepping > 4) - return 0; + X86_MATCH_VFM(INTEL_BROADWELL_X, 0x0b000020), - return ~0U; -} + X86_MATCH_VFM_STEPS(INTEL_BROADWELL_D, 0x2, 0x2, 0x00000011), + X86_MATCH_VFM_STEPS(INTEL_BROADWELL_D, 0x3, 0x3, 0x0700000e), + X86_MATCH_VFM_STEPS(INTEL_BROADWELL_D, 0x4, 0x4, 0x0f00000c), + X86_MATCH_VFM_STEPS(INTEL_BROADWELL_D, 0x5, 0x5, 0x0e000003), -static const struct x86_cpu_id deadline_match[] = { - DEADLINE_MODEL_MATCH_FUNC( INTEL_FAM6_HASWELL_X, hsx_deadline_rev), - DEADLINE_MODEL_MATCH_REV ( INTEL_FAM6_BROADWELL_X, 0x0b000020), - DEADLINE_MODEL_MATCH_FUNC( INTEL_FAM6_BROADWELL_XEON_D, bdx_deadline_rev), - DEADLINE_MODEL_MATCH_FUNC( INTEL_FAM6_SKYLAKE_X, skx_deadline_rev), + X86_MATCH_VFM_STEPS(INTEL_SKYLAKE_X, 0x3, 0x3, 0x01000136), + X86_MATCH_VFM_STEPS(INTEL_SKYLAKE_X, 0x4, 0x4, 0x02000014), + X86_MATCH_VFM_STEPS(INTEL_SKYLAKE_X, 0x5, 0xf, 0), - DEADLINE_MODEL_MATCH_REV ( INTEL_FAM6_HASWELL_CORE, 0x22), - DEADLINE_MODEL_MATCH_REV ( INTEL_FAM6_HASWELL_ULT, 0x20), - DEADLINE_MODEL_MATCH_REV ( INTEL_FAM6_HASWELL_GT3E, 0x17), + X86_MATCH_VFM(INTEL_HASWELL, 0x22), + X86_MATCH_VFM(INTEL_HASWELL_L, 0x20), + X86_MATCH_VFM(INTEL_HASWELL_G, 0x17), - DEADLINE_MODEL_MATCH_REV ( INTEL_FAM6_BROADWELL_CORE, 0x25), - DEADLINE_MODEL_MATCH_REV ( INTEL_FAM6_BROADWELL_GT3E, 0x17), + X86_MATCH_VFM(INTEL_BROADWELL, 0x25), + X86_MATCH_VFM(INTEL_BROADWELL_G, 0x17), - DEADLINE_MODEL_MATCH_REV ( INTEL_FAM6_SKYLAKE_MOBILE, 0xb2), - DEADLINE_MODEL_MATCH_REV ( INTEL_FAM6_SKYLAKE_DESKTOP, 0xb2), + X86_MATCH_VFM(INTEL_SKYLAKE_L, 0xb2), + X86_MATCH_VFM(INTEL_SKYLAKE, 0xb2), - DEADLINE_MODEL_MATCH_REV ( INTEL_FAM6_KABYLAKE_MOBILE, 0x52), - DEADLINE_MODEL_MATCH_REV ( INTEL_FAM6_KABYLAKE_DESKTOP, 0x52), + X86_MATCH_VFM(INTEL_KABYLAKE_L, 0x52), + X86_MATCH_VFM(INTEL_KABYLAKE, 0x52), {}, }; -static void apic_check_deadline_errata(void) +static __init bool apic_validate_deadline_timer(void) { const struct x86_cpu_id *m; u32 rev; - if (!boot_cpu_has(X86_FEATURE_TSC_DEADLINE_TIMER) || - boot_cpu_has(X86_FEATURE_HYPERVISOR)) - return; + if (!boot_cpu_has(X86_FEATURE_TSC_DEADLINE_TIMER)) + return false; + if (boot_cpu_has(X86_FEATURE_HYPERVISOR)) + return true; m = x86_match_cpu(deadline_match); if (!m) - return; + return true; - /* - * Function pointers will have the MSB set due to address layout, - * immediate revisions will not. - */ - if ((long)m->driver_data < 0) - rev = ((u32 (*)(void))(m->driver_data))(); - else - rev = (u32)m->driver_data; + rev = (u32)m->driver_data; if (boot_cpu_data.microcode >= rev) - return; + return true; setup_clear_cpu_cap(X86_FEATURE_TSC_DEADLINE_TIMER); pr_err(FW_BUG "TSC_DEADLINE disabled due to Errata; " "please update microcode to version: 0x%x (or later)\n", rev); + return false; } /* @@ -648,7 +577,7 @@ static void setup_APIC_timer(void) if (this_cpu_has(X86_FEATURE_ARAT)) { lapic_clockevent.features &= ~CLOCK_EVT_FEAT_C3STOP; - /* Make LAPIC timer preferrable over percpu HPET */ + /* Make LAPIC timer preferable over percpu HPET */ lapic_clockevent.rating = 150; } @@ -665,6 +594,8 @@ static void setup_APIC_timer(void) 0xF, ~0UL); } else clockevents_register_device(levt); + + apic_update_vector(smp_processor_id(), LOCAL_TIMER_VECTOR, true); } /* @@ -695,7 +626,7 @@ void lapic_update_tsc_freq(void) * In this functions we calibrate APIC bus clocks to the external timer. * * We want to do the calibration only once since we want to have local timer - * irqs syncron. CPUs connected by the same APIC bus have the very same bus + * irqs synchronous. CPUs connected by the same APIC bus have the very same bus * frequency. * * This was previously done by reading the PIT/HPET and waiting for a wrap @@ -717,17 +648,17 @@ void lapic_update_tsc_freq(void) static __initdata int lapic_cal_loops = -1; static __initdata long lapic_cal_t1, lapic_cal_t2; static __initdata unsigned long long lapic_cal_tsc1, lapic_cal_tsc2; -static __initdata unsigned long lapic_cal_pm1, lapic_cal_pm2; +static __initdata u32 lapic_cal_pm1, lapic_cal_pm2; static __initdata unsigned long lapic_cal_j1, lapic_cal_j2; /* - * Temporary interrupt handler. + * Temporary interrupt handler and polled calibration function. */ static void __init lapic_cal_handler(struct clock_event_device *dev) { unsigned long long tsc = 0; long tapic = apic_read(APIC_TMCCT); - unsigned long pm = acpi_pm_read_early(); + u32 pm = acpi_pm_read_early(); if (boot_cpu_has(X86_FEATURE_TSC)) tsc = rdtsc(); @@ -752,7 +683,7 @@ static void __init lapic_cal_handler(struct clock_event_device *dev) } static int __init -calibrate_by_pmtimer(long deltapm, long *delta, long *deltatsc) +calibrate_by_pmtimer(u32 deltapm, long *delta, long *deltatsc) { const long pm_100ms = PMTMR_TICKS_PER_SEC / 10; const long pm_thresh = pm_100ms / 100; @@ -763,7 +694,7 @@ calibrate_by_pmtimer(long deltapm, long *delta, long *deltatsc) return -1; #endif - apic_printk(APIC_VERBOSE, "... PM-Timer delta = %ld\n", deltapm); + apic_pr_verbose("... PM-Timer delta = %u\n", deltapm); /* Check, if the PM timer is available */ if (!deltapm) @@ -773,14 +704,14 @@ calibrate_by_pmtimer(long deltapm, long *delta, long *deltatsc) if (deltapm > (pm_100ms - pm_thresh) && deltapm < (pm_100ms + pm_thresh)) { - apic_printk(APIC_VERBOSE, "... PM-Timer result ok\n"); + apic_pr_verbose("... PM-Timer result ok\n"); return 0; } res = (((u64)deltapm) * mult) >> 22; do_div(res, 1000000); - pr_warning("APIC calibration not consistent " - "with PM-Timer: %ldms instead of 100ms\n",(long)res); + pr_warn("APIC calibration not consistent with PM-Timer: %ldms instead of 100ms\n", + (long)res); /* Correct the lapic counter value */ res = (((u64)(*delta)) * pm_100ms); @@ -793,75 +724,164 @@ calibrate_by_pmtimer(long deltapm, long *delta, long *deltatsc) if (boot_cpu_has(X86_FEATURE_TSC)) { res = (((u64)(*deltatsc)) * pm_100ms); do_div(res, deltapm); - apic_printk(APIC_VERBOSE, "TSC delta adjusted to " - "PM-Timer: %lu (%ld)\n", - (unsigned long)res, *deltatsc); + apic_pr_verbose("TSC delta adjusted to PM-Timer: %lu (%ld)\n", + (unsigned long)res, *deltatsc); *deltatsc = (long)res; } return 0; } +static int __init lapic_init_clockevent(void) +{ + if (!lapic_timer_period) + return -1; + + /* Calculate the scaled math multiplication factor */ + lapic_clockevent.mult = div_sc(lapic_timer_period/APIC_DIVISOR, + TICK_NSEC, lapic_clockevent.shift); + lapic_clockevent.max_delta_ns = + clockevent_delta2ns(0x7FFFFFFF, &lapic_clockevent); + lapic_clockevent.max_delta_ticks = 0x7FFFFFFF; + lapic_clockevent.min_delta_ns = + clockevent_delta2ns(0xF, &lapic_clockevent); + lapic_clockevent.min_delta_ticks = 0xF; + + return 0; +} + +bool __init apic_needs_pit(void) +{ + /* + * If the frequencies are not known, PIT is required for both TSC + * and apic timer calibration. + */ + if (!tsc_khz || !cpu_khz) + return true; + + /* Is there an APIC at all or is it disabled? */ + if (!boot_cpu_has(X86_FEATURE_APIC) || apic_is_disabled) + return true; + + /* + * If interrupt delivery mode is legacy PIC or virtual wire without + * configuration, the local APIC timer won't be set up. Make sure + * that the PIT is initialized. + */ + if (apic_intr_mode == APIC_PIC || + apic_intr_mode == APIC_VIRTUAL_WIRE_NO_CONFIG) + return true; + + /* Virt guests may lack ARAT, but still have DEADLINE */ + if (!boot_cpu_has(X86_FEATURE_ARAT)) + return true; + + /* Deadline timer is based on TSC so no further PIT action required */ + if (boot_cpu_has(X86_FEATURE_TSC_DEADLINE_TIMER)) + return false; + + /* APIC timer disabled? */ + if (disable_apic_timer) + return true; + /* + * The APIC timer frequency is known already, no PIT calibration + * required. If unknown, let the PIT be initialized. + */ + return lapic_timer_period == 0; +} + static int __init calibrate_APIC_clock(void) { struct clock_event_device *levt = this_cpu_ptr(&lapic_events); - void (*real_handler)(struct clock_event_device *dev); + u64 tsc_perj = 0, tsc_start = 0; + long delta_tsc_khz, bus_khz; + unsigned long jif_start; unsigned long deltaj; long delta, deltatsc; int pm_referenced = 0; - /** - * check if lapic timer has already been calibrated by platform - * specific routine, such as tsc calibration code. if so, we just fill + if (boot_cpu_has(X86_FEATURE_TSC_DEADLINE_TIMER)) + return 0; + + /* + * Check if lapic timer has already been calibrated by platform + * specific routine, such as tsc calibration code. If so just fill * in the clockevent structure and return. */ - - if (boot_cpu_has(X86_FEATURE_TSC_DEADLINE_TIMER)) { - return 0; - } else if (lapic_timer_frequency) { - apic_printk(APIC_VERBOSE, "lapic timer already calibrated %d\n", - lapic_timer_frequency); - lapic_clockevent.mult = div_sc(lapic_timer_frequency/APIC_DIVISOR, - TICK_NSEC, lapic_clockevent.shift); - lapic_clockevent.max_delta_ns = - clockevent_delta2ns(0x7FFFFF, &lapic_clockevent); - lapic_clockevent.max_delta_ticks = 0x7FFFFF; - lapic_clockevent.min_delta_ns = - clockevent_delta2ns(0xF, &lapic_clockevent); - lapic_clockevent.min_delta_ticks = 0xF; + if (!lapic_init_clockevent()) { + apic_pr_verbose("lapic timer already calibrated %d\n", lapic_timer_period); + /* + * Direct calibration methods must have an always running + * local APIC timer, no need for broadcast timer. + */ lapic_clockevent.features &= ~CLOCK_EVT_FEAT_DUMMY; return 0; } - apic_printk(APIC_VERBOSE, "Using local APIC timer interrupts.\n" - "calibrating APIC timer ...\n"); + apic_pr_verbose("Using local APIC timer interrupts. Calibrating APIC timer ...\n"); + /* + * There are platforms w/o global clockevent devices. Instead of + * making the calibration conditional on that, use a polling based + * approach everywhere. + */ local_irq_disable(); - /* Replace the global interrupt handler */ - real_handler = global_clock_event->event_handler; - global_clock_event->event_handler = lapic_cal_handler; - /* * Setup the APIC counter to maximum. There is no way the lapic * can underflow in the 100ms detection time frame */ __setup_APIC_LVTT(0xffffffff, 0, 0); - /* Let the interrupts run */ + /* + * Methods to terminate the calibration loop: + * 1) Global clockevent if available (jiffies) + * 2) TSC if available and frequency is known + */ + jif_start = READ_ONCE(jiffies); + + if (tsc_khz) { + tsc_start = rdtsc(); + tsc_perj = div_u64((u64)tsc_khz * 1000, HZ); + } + + /* + * Enable interrupts so the tick can fire, if a global + * clockevent device is available + */ local_irq_enable(); - while (lapic_cal_loops <= LAPIC_CAL_LOOPS) - cpu_relax(); + while (lapic_cal_loops <= LAPIC_CAL_LOOPS) { + /* Wait for a tick to elapse */ + while (1) { + if (tsc_khz) { + u64 tsc_now = rdtsc(); + if ((tsc_now - tsc_start) >= tsc_perj) { + tsc_start += tsc_perj; + break; + } + } else { + unsigned long jif_now = READ_ONCE(jiffies); - local_irq_disable(); + if (time_after(jif_now, jif_start)) { + jif_start = jif_now; + break; + } + } + cpu_relax(); + } - /* Restore the real event handler */ - global_clock_event->event_handler = real_handler; + /* Invoke the calibration routine */ + local_irq_disable(); + lapic_cal_handler(NULL); + local_irq_enable(); + } + + local_irq_disable(); /* Build delta t1-t2 as apic timer counts down */ delta = lapic_cal_t1 - lapic_cal_t2; - apic_printk(APIC_VERBOSE, "... lapic delta = %ld\n", delta); + apic_pr_verbose("... lapic delta = %ld\n", delta); deltatsc = (long)(lapic_cal_tsc2 - lapic_cal_tsc1); @@ -869,52 +889,42 @@ static int __init calibrate_APIC_clock(void) pm_referenced = !calibrate_by_pmtimer(lapic_cal_pm2 - lapic_cal_pm1, &delta, &deltatsc); - /* Calculate the scaled math multiplication factor */ - lapic_clockevent.mult = div_sc(delta, TICK_NSEC * LAPIC_CAL_LOOPS, - lapic_clockevent.shift); - lapic_clockevent.max_delta_ns = - clockevent_delta2ns(0x7FFFFFFF, &lapic_clockevent); - lapic_clockevent.max_delta_ticks = 0x7FFFFFFF; - lapic_clockevent.min_delta_ns = - clockevent_delta2ns(0xF, &lapic_clockevent); - lapic_clockevent.min_delta_ticks = 0xF; - - lapic_timer_frequency = (delta * APIC_DIVISOR) / LAPIC_CAL_LOOPS; + lapic_timer_period = (delta * APIC_DIVISOR) / LAPIC_CAL_LOOPS; + lapic_init_clockevent(); - apic_printk(APIC_VERBOSE, "..... delta %ld\n", delta); - apic_printk(APIC_VERBOSE, "..... mult: %u\n", lapic_clockevent.mult); - apic_printk(APIC_VERBOSE, "..... calibration result: %u\n", - lapic_timer_frequency); + apic_pr_verbose("..... delta %ld\n", delta); + apic_pr_verbose("..... mult: %u\n", lapic_clockevent.mult); + apic_pr_verbose("..... calibration result: %u\n", lapic_timer_period); if (boot_cpu_has(X86_FEATURE_TSC)) { - apic_printk(APIC_VERBOSE, "..... CPU clock speed is " - "%ld.%04ld MHz.\n", - (deltatsc / LAPIC_CAL_LOOPS) / (1000000 / HZ), - (deltatsc / LAPIC_CAL_LOOPS) % (1000000 / HZ)); + delta_tsc_khz = (deltatsc * HZ) / (1000 * LAPIC_CAL_LOOPS); + + apic_pr_verbose("..... CPU clock speed is %ld.%03ld MHz.\n", + delta_tsc_khz / 1000, delta_tsc_khz % 1000); } - apic_printk(APIC_VERBOSE, "..... host bus clock speed is " - "%u.%04u MHz.\n", - lapic_timer_frequency / (1000000 / HZ), - lapic_timer_frequency % (1000000 / HZ)); + bus_khz = (long)lapic_timer_period * HZ / 1000; + apic_pr_verbose("..... host bus clock speed is %ld.%03ld MHz.\n", + bus_khz / 1000, bus_khz % 1000); /* * Do a sanity check on the APIC calibration result */ - if (lapic_timer_frequency < (1000000 / HZ)) { + if (lapic_timer_period < (1000000 / HZ)) { local_irq_enable(); - pr_warning("APIC frequency too slow, disabling apic timer\n"); + pr_warn("APIC frequency too slow, disabling apic timer\n"); return -1; } levt->features &= ~CLOCK_EVT_FEAT_DUMMY; /* - * PM timer calibration failed or not turned on - * so lets try APIC timer based calibration + * PM timer calibration failed or not turned on so lets try APIC + * timer based calibration, if a global clockevent device is + * available. */ - if (!pm_referenced) { - apic_printk(APIC_VERBOSE, "... verify APIC timer\n"); + if (!pm_referenced && global_clock_event) { + apic_pr_verbose("... verify APIC timer\n"); /* * Setup the apic timer manually @@ -935,18 +945,18 @@ static int __init calibrate_APIC_clock(void) /* Jiffies delta */ deltaj = lapic_cal_j2 - lapic_cal_j1; - apic_printk(APIC_VERBOSE, "... jiffies delta = %lu\n", deltaj); + apic_pr_verbose("... jiffies delta = %lu\n", deltaj); /* Check, if the jiffies result is consistent */ if (deltaj >= LAPIC_CAL_LOOPS-2 && deltaj <= LAPIC_CAL_LOOPS+2) - apic_printk(APIC_VERBOSE, "... jiffies result ok\n"); + apic_pr_verbose("... jiffies result ok\n"); else levt->features |= CLOCK_EVT_FEAT_DUMMY; } local_irq_enable(); if (levt->features & CLOCK_EVT_FEAT_DUMMY) { - pr_warning("APIC timer disabled due to verification failure\n"); + pr_warn("APIC timer disabled due to verification failure\n"); return -1; } @@ -1020,8 +1030,8 @@ static void local_apic_timer_interrupt(void) * spurious. */ if (!evt->event_handler) { - pr_warning("Spurious LAPIC timer interrupt on cpu %d\n", - smp_processor_id()); + pr_warn("Spurious LAPIC timer interrupt on cpu %d\n", + smp_processor_id()); /* Switch it off */ lapic_timer_shutdown(evt); return; @@ -1043,32 +1053,18 @@ static void local_apic_timer_interrupt(void) * [ if a single-CPU system runs an SMP kernel then we call the local * interrupt as well. Thus we cannot inline the local irq ... ] */ -__visible void __irq_entry smp_apic_timer_interrupt(struct pt_regs *regs) +DEFINE_IDTENTRY_SYSVEC(sysvec_apic_timer_interrupt) { struct pt_regs *old_regs = set_irq_regs(regs); - /* - * NOTE! We'd better ACK the irq immediately, - * because timer handling can be slow. - * - * update_process_times() expects us to have done irq_enter(). - * Besides, if we don't timer interrupts ignore the global - * interrupt lock, which is the WrongThing (tm) to do. - */ - entering_ack_irq(); + apic_eoi(); trace_local_timer_entry(LOCAL_TIMER_VECTOR); local_apic_timer_interrupt(); trace_local_timer_exit(LOCAL_TIMER_VECTOR); - exiting_irq(); set_irq_regs(old_regs); } -int setup_profiling_timer(unsigned int multiplier) -{ - return -EINVAL; -} - /* * Local APIC start and shutdown */ @@ -1085,8 +1081,7 @@ void clear_local_APIC(void) int maxlvt; u32 v; - /* APIC hasn't been mapped yet */ - if (!x2apic_mode && !apic_phys) + if (!apic_accessible()) return; maxlvt = lapic_get_maxlvt(); @@ -1149,25 +1144,40 @@ void clear_local_APIC(void) } /** - * disable_local_APIC - clear and disable the local APIC + * apic_soft_disable - Clears and software disables the local APIC on hotplug + * + * Contrary to disable_local_APIC() this does not touch the enable bit in + * MSR_IA32_APICBASE. Clearing that bit on systems based on the 3 wire APIC + * bus would require a hardware reset as the APIC would lose track of bus + * arbitration. On systems with FSB delivery APICBASE could be disabled, + * but it has to be guaranteed that no interrupt is sent to the APIC while + * in that state and it's not clear from the SDM whether it still responds + * to INIT/SIPI messages. Stay on the safe side and use software disable. */ -void disable_local_APIC(void) +void apic_soft_disable(void) { - unsigned int value; - - /* APIC hasn't been mapped yet */ - if (!x2apic_mode && !apic_phys) - return; + u32 value; clear_local_APIC(); - /* - * Disable APIC (implies clearing of registers - * for 82489DX!). - */ + /* Soft disable APIC (implies clearing of registers for 82489DX!). */ value = apic_read(APIC_SPIV); value &= ~APIC_SPIV_APIC_ENABLED; apic_write(APIC_SPIV, value); +} + +/** + * disable_local_APIC - clear and disable the local APIC + */ +void disable_local_APIC(void) +{ + if (!apic_accessible()) + return; + + if (apic->teardown) + apic->teardown(); + + apic_soft_disable(); #ifdef CONFIG_X86_32 /* @@ -1227,17 +1237,16 @@ void __init sync_Arb_IDs(void) */ apic_wait_icr_idle(); - apic_printk(APIC_DEBUG, "Synchronizing Arb IDs.\n"); - apic_write(APIC_ICR, APIC_DEST_ALLINC | - APIC_INT_LEVELTRIG | APIC_DM_INIT); + apic_pr_debug("Synchronizing Arb IDs.\n"); + apic_write(APIC_ICR, APIC_DEST_ALLINC | APIC_INT_LEVELTRIG | APIC_DM_INIT); } -enum apic_intr_mode_id apic_intr_mode; +enum apic_intr_mode_id apic_intr_mode __ro_after_init; -static int __init apic_intr_mode_select(void) +static int __init __apic_intr_mode_select(void) { /* Check kernel option */ - if (disable_apic) { + if (apic_is_disabled) { pr_info("APIC disabled via kernel command line\n"); return APIC_PIC; } @@ -1246,7 +1255,7 @@ static int __init apic_intr_mode_select(void) #ifdef CONFIG_X86_64 /* On 64-bit, the APIC must be integrated, Check local APIC only */ if (!boot_cpu_has(X86_FEATURE_APIC)) { - disable_apic = 1; + apic_is_disabled = true; pr_info("APIC disabled by BIOS\n"); return APIC_PIC; } @@ -1255,16 +1264,15 @@ static int __init apic_intr_mode_select(void) /* Neither 82489DX nor integrated APIC ? */ if (!boot_cpu_has(X86_FEATURE_APIC) && !smp_found_config) { - disable_apic = 1; + apic_is_disabled = true; return APIC_PIC; } /* If the BIOS pretends there is an integrated APIC ? */ if (!boot_cpu_has(X86_FEATURE_APIC) && APIC_INTEGRATED(boot_cpu_apic_version)) { - disable_apic = 1; - pr_err(FW_BUG "Local APIC %d not detected, force emulation\n", - boot_cpu_physical_apicid); + apic_is_disabled = true; + pr_err(FW_BUG "Local APIC not detected, force emulation\n"); return APIC_PIC; } #endif @@ -1285,17 +1293,17 @@ static int __init apic_intr_mode_select(void) pr_info("APIC: SMP mode deactivated\n"); return APIC_SYMMETRIC_IO_NO_ROUTING; } - - if (read_apic_id() != boot_cpu_physical_apicid) { - panic("Boot APIC ID in local APIC unexpected (%d vs %d)", - read_apic_id(), boot_cpu_physical_apicid); - /* Or can we switch back to PIC here? */ - } #endif return APIC_SYMMETRIC_IO; } +/* Select the interrupt delivery mode for the BSP */ +void __init apic_intr_mode_select(void) +{ + apic_intr_mode = __apic_intr_mode_select(); +} + /* * An initial setup of the virtual wire mode. */ @@ -1345,35 +1353,37 @@ void __init init_bsp_APIC(void) apic_write(APIC_LVT1, value); } +static void __init apic_bsp_setup(bool upmode); + /* Init the interrupt delivery mode for the BSP */ void __init apic_intr_mode_init(void) { bool upmode = IS_ENABLED(CONFIG_UP_LATE_INIT); - apic_intr_mode = apic_intr_mode_select(); - switch (apic_intr_mode) { case APIC_PIC: pr_info("APIC: Keep in PIC mode(8259)\n"); return; case APIC_VIRTUAL_WIRE: pr_info("APIC: Switch to virtual wire mode setup\n"); - default_setup_apic_routing(); break; case APIC_VIRTUAL_WIRE_NO_CONFIG: pr_info("APIC: Switch to virtual wire mode setup with no configuration\n"); upmode = true; - default_setup_apic_routing(); break; case APIC_SYMMETRIC_IO: pr_info("APIC: Switch to symmetric I/O mode setup\n"); - default_setup_apic_routing(); break; case APIC_SYMMETRIC_IO_NO_ROUTING: pr_info("APIC: Switch to symmetric I/O mode setup in no SMP routine\n"); break; } + x86_64_probe_apic(); + + if (x86_platform.apic_post_init) + x86_platform.apic_post_init(); + apic_bsp_setup(upmode); } @@ -1412,59 +1422,76 @@ static void lapic_setup_esr(void) if (maxlvt > 3) apic_write(APIC_ESR, 0); value = apic_read(APIC_ESR); - if (value != oldvalue) - apic_printk(APIC_VERBOSE, "ESR value before enabling " - "vector: 0x%08x after: 0x%08x\n", - oldvalue, value); + if (value != oldvalue) { + apic_pr_verbose("ESR value before enabling vector: 0x%08x after: 0x%08x\n", + oldvalue, value); + } } -static void apic_pending_intr_clear(void) +#define APIC_IR_REGS APIC_ISR_NR +#define APIC_IR_BITS (APIC_IR_REGS * 32) +#define APIC_IR_MAPSIZE (APIC_IR_BITS / BITS_PER_LONG) + +union apic_ir { + unsigned long map[APIC_IR_MAPSIZE]; + u32 regs[APIC_IR_REGS]; +}; + +static bool apic_check_and_eoi_isr(union apic_ir *isr) { - long long max_loops = cpu_khz ? cpu_khz : 1000000; - unsigned long long tsc = 0, ntsc; - unsigned int queued; - unsigned long value; - int i, j, acked = 0; + int i, bit; + + /* Read the ISRs */ + for (i = 0; i < APIC_IR_REGS; i++) + isr->regs[i] = apic_read(APIC_ISR + i * 0x10); + + /* If the ISR map empty, nothing to do here. */ + if (bitmap_empty(isr->map, APIC_IR_BITS)) + return true; - if (boot_cpu_has(X86_FEATURE_TSC)) - tsc = rdtsc(); /* - * After a crash, we no longer service the interrupts and a pending - * interrupt from previous kernel might still have ISR bit set. - * - * Most probably by now CPU has serviced that pending interrupt and - * it might not have done the ack_APIC_irq() because it thought, - * interrupt came from i8259 as ExtInt. LAPIC did not get EOI so it - * does not clear the ISR bit and cpu thinks it has already serivced - * the interrupt. Hence a vector might get locked. It was noticed - * for timer irq (vector 0x31). Issue an extra EOI to clear ISR. + * There can be multiple ISR bits set when a high priority + * interrupt preempted a lower priority one. Issue an EOI for each + * set bit. The priority traversal order does not matter as there + * can't be new ISR bits raised at this point. What matters is that + * an EOI is issued for each ISR bit. */ - do { - queued = 0; - for (i = APIC_ISR_NR - 1; i >= 0; i--) - queued |= apic_read(APIC_IRR + i*0x10); - - for (i = APIC_ISR_NR - 1; i >= 0; i--) { - value = apic_read(APIC_ISR + i*0x10); - for_each_set_bit(j, &value, 32) { - ack_APIC_irq(); - acked++; - } - } - if (acked > 256) { - pr_err("LAPIC pending interrupts after %d EOI\n", acked); - break; - } - if (queued) { - if (boot_cpu_has(X86_FEATURE_TSC) && cpu_khz) { - ntsc = rdtsc(); - max_loops = (cpu_khz << 10) - (ntsc - tsc); - } else { - max_loops--; - } - } - } while (queued && max_loops > 0); - WARN_ON(max_loops <= 0); + for_each_set_bit(bit, isr->map, APIC_IR_BITS) + apic_eoi(); + + /* Reread the ISRs, they should be empty now */ + for (i = 0; i < APIC_IR_REGS; i++) + isr->regs[i] = apic_read(APIC_ISR + i * 0x10); + + return bitmap_empty(isr->map, APIC_IR_BITS); +} + +/* + * If a CPU services an interrupt and crashes before issuing EOI to the + * local APIC, the corresponding ISR bit is still set when the crashing CPU + * jumps into a crash kernel. Read the ISR and issue an EOI for each set + * bit to acknowledge it as otherwise these slots would be locked forever + * waiting for an EOI. + * + * If there are pending bits in the IRR, then they won't be converted into + * ISR bits as the CPU has interrupts disabled. They will be delivered once + * the CPU enables interrupts and there is nothing which can prevent that. + * + * In the worst case this results in spurious interrupt warnings. + */ +static void apic_clear_isr(void) +{ + union apic_ir ir; + unsigned int i; + + if (!apic_check_and_eoi_isr(&ir)) + pr_warn("APIC: Stale ISR: %256pb\n", ir.map); + + for (i = 0; i < APIC_IR_REGS; i++) + ir.regs[i] = apic_read(APIC_IRR + i * 0x10); + + if (!bitmap_empty(ir.map, APIC_IR_BITS)) + pr_warn("APIC: Stale IRR: %256pb\n", ir.map); } /** @@ -1477,16 +1504,23 @@ static void setup_local_APIC(void) { int cpu = smp_processor_id(); unsigned int value; -#ifdef CONFIG_X86_32 - int logical_apicid, ldr_apicid; -#endif - - if (disable_apic) { + if (apic_is_disabled) { disable_ioapic_support(); return; } + if (apic->setup) + apic->setup(); + + /* + * If this comes from kexec/kcrash the APIC might be enabled in + * SPIV. Soft disable it before doing further initialization. + */ + value = apic_read(APIC_SPIV); + value &= ~APIC_SPIV_APIC_ENABLED; + apic_write(APIC_SPIV, value); + #ifdef CONFIG_X86_32 /* Pound the ESR really hard over the head with a big hammer - mbligh */ if (lapic_is_integrated() && apic->disable_esr) { @@ -1496,43 +1530,28 @@ static void setup_local_APIC(void) apic_write(APIC_ESR, 0); } #endif - perf_events_lapic_init(); - - /* - * Double-check whether this APIC is really registered. - * This is meaningless in clustered apic mode, so we skip it. - */ - BUG_ON(!apic->apic_id_registered()); - /* * Intel recommends to set DFR, LDR and TPR before enabling * an APIC. See e.g. "AP-388 82489DX User's Manual" (Intel - * document number 292116). So here it goes... + * document number 292116). + * + * Except for APICs which operate in physical destination mode. */ - apic->init_apic_ldr(); + if (apic->init_apic_ldr) + apic->init_apic_ldr(); -#ifdef CONFIG_X86_32 /* - * APIC LDR is initialized. If logical_apicid mapping was - * initialized during get_smp_config(), make sure it matches the - * actual value. - */ - logical_apicid = early_per_cpu(x86_cpu_to_logical_apicid, cpu); - ldr_apicid = GET_APIC_LOGICAL_ID(apic_read(APIC_LDR)); - WARN_ON(logical_apicid != BAD_APICID && logical_apicid != ldr_apicid); - /* always use the value from LDR */ - early_per_cpu(x86_cpu_to_logical_apicid, cpu) = ldr_apicid; -#endif - - /* - * Set Task Priority to 'accept all'. We never change this - * later on. + * Set Task Priority to 'accept all except vectors 0-31'. An APIC + * vector in the 16-31 range could be delivered if TPR == 0, but we + * would think it's an exception and terrible things will happen. We + * never change this later on. */ value = apic_read(APIC_TASKPRI); value &= ~APIC_TPRI_MASK; + value |= 0x10; apic_write(APIC_TASKPRI, value); - apic_pending_intr_clear(); + apic_clear_isr(); /* * Now that we are all set up, enable the APIC @@ -1560,7 +1579,7 @@ static void setup_local_APIC(void) */ /* * Actually disabling the focus CPU check just makes the hang less - * frequent as it makes the interrupt distributon model be more + * frequent as it makes the interrupt distribution model be more * like LRU than MRU (the short-term load is more even across CPUs). */ @@ -1578,6 +1597,8 @@ static void setup_local_APIC(void) value |= SPURIOUS_APIC_VECTOR; apic_write(APIC_SPIV, value); + perf_events_lapic_init(); + /* * Set up LVT0, LVT1: * @@ -1589,12 +1610,12 @@ static void setup_local_APIC(void) * TODO: set up through-local-APIC from through-I/O-APIC? --macro */ value = apic_read(APIC_LVT0) & APIC_LVT_MASKED; - if (!cpu && (pic_mode || !value || skip_ioapic_setup)) { + if (!cpu && (pic_mode || !value || ioapic_is_disabled)) { value = APIC_DM_EXTINT; - apic_printk(APIC_VERBOSE, "enabled ExtINT on CPU#%d\n", cpu); + apic_pr_verbose("Enabled ExtINT on CPU#%d\n", cpu); } else { value = APIC_DM_EXTINT | APIC_LVT_MASKED; - apic_printk(APIC_VERBOSE, "masked ExtINT on CPU#%d\n", cpu); + apic_pr_verbose("Masked ExtINT on CPU#%d\n", cpu); } apic_write(APIC_LVT0, value); @@ -1646,16 +1667,49 @@ void apic_ap_setup(void) end_local_APIC_setup(); } +static __init void apic_read_boot_cpu_id(bool x2apic) +{ + /* + * This can be invoked from check_x2apic() before the APIC has been + * selected. But that code knows for sure that the BIOS enabled + * X2APIC. + */ + if (x2apic) { + boot_cpu_physical_apicid = native_apic_msr_read(APIC_ID); + boot_cpu_apic_version = GET_APIC_VERSION(native_apic_msr_read(APIC_LVR)); + } else { + boot_cpu_physical_apicid = read_apic_id(); + boot_cpu_apic_version = GET_APIC_VERSION(apic_read(APIC_LVR)); + } + topology_register_boot_apic(boot_cpu_physical_apicid); +} + #ifdef CONFIG_X86_X2APIC int x2apic_mode; +EXPORT_SYMBOL_GPL(x2apic_mode); enum { X2APIC_OFF, - X2APIC_ON, X2APIC_DISABLED, + /* All states below here have X2APIC enabled */ + X2APIC_ON, + X2APIC_ON_LOCKED }; static int x2apic_state; +static bool x2apic_hw_locked(void) +{ + u64 x86_arch_cap_msr; + u64 msr; + + x86_arch_cap_msr = x86_read_arch_cap_msr(); + if (x86_arch_cap_msr & ARCH_CAP_XAPIC_DISABLE) { + rdmsrq(MSR_IA32_XAPIC_DISABLE_STATUS, msr); + return (msr & LEGACY_XAPIC_DISABLED); + } + return false; +} + static void __x2apic_disable(void) { u64 msr; @@ -1663,12 +1717,12 @@ static void __x2apic_disable(void) if (!boot_cpu_has(X86_FEATURE_APIC)) return; - rdmsrl(MSR_IA32_APICBASE, msr); + rdmsrq(MSR_IA32_APICBASE, msr); if (!(msr & X2APIC_ENABLE)) return; /* Disable xapic and x2apic first and then reenable xapic mode */ - wrmsrl(MSR_IA32_APICBASE, msr & ~(X2APIC_ENABLE | XAPIC_ENABLE)); - wrmsrl(MSR_IA32_APICBASE, msr & ~X2APIC_ENABLE); + wrmsrq(MSR_IA32_APICBASE, msr & ~(X2APIC_ENABLE | XAPIC_ENABLE)); + wrmsrq(MSR_IA32_APICBASE, msr & ~X2APIC_ENABLE); printk_once(KERN_INFO "x2apic disabled\n"); } @@ -1676,24 +1730,28 @@ static void __x2apic_enable(void) { u64 msr; - rdmsrl(MSR_IA32_APICBASE, msr); + rdmsrq(MSR_IA32_APICBASE, msr); if (msr & X2APIC_ENABLE) return; - wrmsrl(MSR_IA32_APICBASE, msr | X2APIC_ENABLE); + wrmsrq(MSR_IA32_APICBASE, msr | X2APIC_ENABLE); printk_once(KERN_INFO "x2apic enabled\n"); } static int __init setup_nox2apic(char *str) { if (x2apic_enabled()) { - int apicid = native_apic_msr_read(APIC_ID); + u32 apicid = native_apic_msr_read(APIC_ID); if (apicid >= 255) { - pr_warning("Apicid: %08x, cannot enforce nox2apic\n", - apicid); + pr_warn("Apicid: %08x, cannot enforce nox2apic\n", + apicid); return 0; } - pr_warning("x2apic already enabled.\n"); + if (x2apic_hw_locked()) { + pr_warn("APIC locked in x2apic mode, can't disable\n"); + return 0; + } + pr_warn("x2apic already enabled.\n"); __x2apic_disable(); } setup_clear_cpu_cap(X86_FEATURE_X2APIC); @@ -1707,32 +1765,53 @@ early_param("nox2apic", setup_nox2apic); void x2apic_setup(void) { /* - * If x2apic is not in ON state, disable it if already enabled + * Try to make the AP's APIC state match that of the BSP, but if the + * BSP is unlocked and the AP is locked then there is a state mismatch. + * Warn about the mismatch in case a GP fault occurs due to a locked AP + * trying to be turned off. + */ + if (x2apic_state != X2APIC_ON_LOCKED && x2apic_hw_locked()) + pr_warn("x2apic lock mismatch between BSP and AP.\n"); + /* + * If x2apic is not in ON or LOCKED state, disable it if already enabled * from BIOS. */ - if (x2apic_state != X2APIC_ON) { + if (x2apic_state < X2APIC_ON) { __x2apic_disable(); return; } __x2apic_enable(); } +static __init void apic_set_fixmap(bool read_apic); + static __init void x2apic_disable(void) { - u32 x2apic_id, state = x2apic_state; - - x2apic_mode = 0; - x2apic_state = X2APIC_DISABLED; + u32 x2apic_id; - if (state != X2APIC_ON) + if (x2apic_state < X2APIC_ON) return; x2apic_id = read_apic_id(); if (x2apic_id >= 255) panic("Cannot disable x2apic, id: %08x\n", x2apic_id); + if (x2apic_hw_locked()) { + pr_warn("Cannot disable locked x2apic, id: %08x\n", x2apic_id); + return; + } + __x2apic_disable(); - register_lapic_address(mp_lapic_addr); + + x2apic_mode = 0; + x2apic_state = X2APIC_DISABLED; + + /* + * Don't reread the APIC ID as it was already done from + * check_x2apic() and the APIC driver still is a x2APIC variant, + * which fails to do the read after x2APIC was disabled. + */ + apic_set_fixmap(false); } static __init void x2apic_enable(void) @@ -1751,20 +1830,34 @@ static __init void try_to_enable_x2apic(int remap_mode) return; if (remap_mode != IRQ_REMAP_X2APIC_MODE) { - /* IR is required if there is APIC ID > 255 even when running - * under KVM + u32 apic_limit = 255; + + /* + * Using X2APIC without IR is not architecturally supported + * on bare metal but may be supported in guests. */ - if (max_physical_apicid > 255 || - !x86_init.hyper.x2apic_available()) { + if (!x86_init.hyper.x2apic_available()) { pr_info("x2apic: IRQ remapping doesn't support X2APIC mode\n"); x2apic_disable(); return; } /* - * without IR all CPUs can be addressed by IOAPIC/MSI - * only in physical mode + * If the hypervisor supports extended destination ID in + * MSI, that increases the maximum APIC ID that can be + * used for non-remapped IRQ domains. */ + if (x86_init.hyper.msi_ext_dest_id()) { + virt_ext_dest_id = 1; + apic_limit = 32767; + } + + /* + * Without IR, all CPUs can be addressed by IOAPIC/MSI only + * in physical mode, and CPUs with an APIC ID that cannot + * be addressed must not be brought online. + */ + x2apic_set_max_apicid(apic_limit); x2apic_phys = 1; } x2apic_enable(); @@ -1775,22 +1868,29 @@ void __init check_x2apic(void) if (x2apic_enabled()) { pr_info("x2apic: enabled by BIOS, switching to x2apic ops\n"); x2apic_mode = 1; - x2apic_state = X2APIC_ON; + if (x2apic_hw_locked()) + x2apic_state = X2APIC_ON_LOCKED; + else + x2apic_state = X2APIC_ON; + apic_read_boot_cpu_id(true); } else if (!boot_cpu_has(X86_FEATURE_X2APIC)) { x2apic_state = X2APIC_DISABLED; } } #else /* CONFIG_X86_X2APIC */ -static int __init validate_x2apic(void) +void __init check_x2apic(void) { if (!apic_is_x2apic_enabled()) - return 0; + return; /* - * Checkme: Can we simply turn off x2apic here instead of panic? + * Checkme: Can we simply turn off x2APIC here instead of disabling the APIC? */ - panic("BIOS has enabled x2apic but kernel doesn't support x2apic, please disable x2apic in BIOS.\n"); + pr_err("Kernel does not support x2APIC, please recompile with CONFIG_X86_X2APIC.\n"); + pr_err("Disabling APIC, expect reduced performance and functionality.\n"); + + apic_is_disabled = true; + setup_clear_cpu_cap(X86_FEATURE_APIC); } -early_initcall(validate_x2apic); static inline void try_to_enable_x2apic(int remap_mode) { } static inline void __x2apic_enable(void) { } @@ -1801,7 +1901,7 @@ void __init enable_IR_x2apic(void) unsigned long flags; int ret, ir_stat; - if (skip_ioapic_setup) { + if (ioapic_is_disabled) { pr_info("Not enabling interrupt remapping due to skipped IO-APIC setup\n"); return; } @@ -1839,19 +1939,19 @@ void __init enable_IR_x2apic(void) * On AMD64 we trust the BIOS - if it says no APIC it is likely * not correctly set up (usually the APIC timer won't work etc.) */ -static int __init detect_init_APIC(void) +static bool __init detect_init_APIC(void) { if (!boot_cpu_has(X86_FEATURE_APIC)) { pr_info("No local APIC present\n"); - return -1; + return false; } - mp_lapic_addr = APIC_DEFAULT_PHYS_BASE; - return 0; + register_lapic_address(APIC_DEFAULT_PHYS_BASE); + return true; } #else -static int __init apic_verify(void) +static bool __init apic_verify(unsigned long addr) { u32 features, h, l; @@ -1861,29 +1961,29 @@ static int __init apic_verify(void) */ features = cpuid_edx(1); if (!(features & (1 << X86_FEATURE_APIC))) { - pr_warning("Could not enable APIC!\n"); - return -1; + pr_warn("Could not enable APIC!\n"); + return false; } set_cpu_cap(&boot_cpu_data, X86_FEATURE_APIC); - mp_lapic_addr = APIC_DEFAULT_PHYS_BASE; /* The BIOS may have set up the APIC at some other address */ if (boot_cpu_data.x86 >= 6) { rdmsr(MSR_IA32_APICBASE, l, h); if (l & MSR_IA32_APICBASE_ENABLE) - mp_lapic_addr = l & MSR_IA32_APICBASE_BASE; + addr = l & MSR_IA32_APICBASE_BASE; } + register_lapic_address(addr); pr_info("Found and enabled local APIC!\n"); - return 0; + return true; } -int __init apic_force_enable(unsigned long addr) +bool __init apic_force_enable(unsigned long addr) { u32 h, l; - if (disable_apic) - return -1; + if (apic_is_disabled) + return false; /* * Some BIOSes disable the local APIC in the APIC_BASE @@ -1900,17 +2000,17 @@ int __init apic_force_enable(unsigned long addr) enabled_via_apicbase = 1; } } - return apic_verify(); + return apic_verify(addr); } /* * Detect and initialize APIC */ -static int __init detect_init_APIC(void) +static bool __init detect_init_APIC(void) { /* Disabled by kernel option? */ - if (disable_apic) - return -1; + if (apic_is_disabled) + return false; switch (boot_cpu_data.x86_vendor) { case X86_VENDOR_AMD: @@ -1921,8 +2021,8 @@ static int __init detect_init_APIC(void) case X86_VENDOR_HYGON: break; case X86_VENDOR_INTEL: - if (boot_cpu_data.x86 == 6 || boot_cpu_data.x86 == 15 || - (boot_cpu_data.x86 == 5 && boot_cpu_has(X86_FEATURE_APIC))) + if ((boot_cpu_data.x86 == 5 && boot_cpu_has(X86_FEATURE_APIC)) || + boot_cpu_data.x86_vfm >= INTEL_PENTIUM_PRO) break; goto no_apic; default: @@ -1937,22 +2037,22 @@ static int __init detect_init_APIC(void) if (!force_enable_local_apic) { pr_info("Local APIC disabled by BIOS -- " "you can enable it with \"lapic\"\n"); - return -1; + return false; } - if (apic_force_enable(APIC_DEFAULT_PHYS_BASE)) - return -1; + if (!apic_force_enable(APIC_DEFAULT_PHYS_BASE)) + return false; } else { - if (apic_verify()) - return -1; + if (!apic_verify(APIC_DEFAULT_PHYS_BASE)) + return false; } apic_pm_activate(); - return 0; + return true; no_apic: pr_info("No local APIC present or hardware disabled\n"); - return -1; + return false; } #endif @@ -1961,62 +2061,37 @@ no_apic: */ void __init init_apic_mappings(void) { - unsigned int new_apicid; + if (apic_validate_deadline_timer()) + pr_info("TSC deadline timer available\n"); - apic_check_deadline_errata(); - - if (x2apic_mode) { - boot_cpu_physical_apicid = read_apic_id(); + if (x2apic_mode) return; - } - - /* If no local APIC can be found return early */ - if (!smp_found_config && detect_init_APIC()) { - /* lets NOP'ify apic operations */ - pr_info("APIC: disable apic facility\n"); - apic_disable(); - } else { - apic_phys = mp_lapic_addr; - /* - * If the system has ACPI MADT tables or MP info, the LAPIC - * address is already registered. - */ - if (!acpi_lapic && !smp_found_config) - register_lapic_address(apic_phys); + if (!smp_found_config) { + if (!detect_init_APIC()) { + pr_info("APIC: disable apic facility\n"); + apic_disable(); + } } +} - /* - * Fetch the APIC ID of the BSP in case we have a - * default configuration (or the MP table is broken). - */ - new_apicid = read_apic_id(); - if (boot_cpu_physical_apicid != new_apicid) { - boot_cpu_physical_apicid = new_apicid; - /* - * yeah -- we lie about apic_version - * in case if apic was disabled via boot option - * but it's not a problem for SMP compiled kernel - * since apic_intr_mode_select is prepared for such - * a case and disable smp mode - */ - boot_cpu_apic_version = GET_APIC_VERSION(apic_read(APIC_LVR)); - } +static __init void apic_set_fixmap(bool read_apic) +{ + set_fixmap_nocache(FIX_APIC_BASE, mp_lapic_addr); + apic_mmio_base = APIC_BASE; + apic_pr_verbose("Mapped APIC to %16lx (%16lx)\n", apic_mmio_base, mp_lapic_addr); + if (read_apic) + apic_read_boot_cpu_id(false); } void __init register_lapic_address(unsigned long address) { + /* This should only happen once */ + WARN_ON_ONCE(mp_lapic_addr); mp_lapic_addr = address; - if (!x2apic_mode) { - set_fixmap_nocache(FIX_APIC_BASE, address); - apic_printk(APIC_VERBOSE, "mapped APIC to %16lx (%16lx)\n", - APIC_BASE, address); - } - if (boot_cpu_physical_apicid == -1U) { - boot_cpu_physical_apicid = read_apic_id(); - boot_cpu_apic_version = GET_APIC_VERSION(apic_read(APIC_LVR)); - } + if (!x2apic_mode) + apic_set_fixmap(true); } /* @@ -2024,39 +2099,67 @@ void __init register_lapic_address(unsigned long address) */ /* - * This interrupt should _never_ happen with our APIC/SMP architecture + * Common handling code for spurious_interrupt and spurious_vector entry + * points below. No point in allowing the compiler to inline it twice. */ -__visible void __irq_entry smp_spurious_interrupt(struct pt_regs *regs) +static noinline void handle_spurious_interrupt(u8 vector) { - u8 vector = ~regs->orig_ax; u32 v; - entering_irq(); trace_spurious_apic_entry(vector); + inc_irq_stat(irq_spurious_count); + /* - * Check if this really is a spurious interrupt and ACK it - * if it is a vectored one. Just in case... - * Spurious interrupts should not be ACKed. + * If this is a spurious interrupt then do not acknowledge */ - v = apic_read(APIC_ISR + ((vector & ~0x1f) >> 1)); - if (v & (1 << (vector & 0x1f))) - ack_APIC_irq(); + if (vector == SPURIOUS_APIC_VECTOR) { + /* See SDM vol 3 */ + pr_info("Spurious APIC interrupt (vector 0xFF) on CPU#%d, should never happen.\n", + smp_processor_id()); + goto out; + } - inc_irq_stat(irq_spurious_count); + /* + * If it is a vectored one, verify it's set in the ISR. If set, + * acknowledge it. + */ + v = apic_read(APIC_ISR + ((vector & ~0x1f) >> 1)); + if (v & (1 << (vector & 0x1f))) { + pr_info("Spurious interrupt (vector 0x%02x) on CPU#%d. Acked\n", + vector, smp_processor_id()); + apic_eoi(); + } else { + pr_info("Spurious interrupt (vector 0x%02x) on CPU#%d. Not pending!\n", + vector, smp_processor_id()); + } +out: + trace_spurious_apic_exit(vector); +} - /* see sw-dev-man vol 3, chapter 7.4.13.5 */ - pr_info("spurious APIC interrupt through vector %02x on CPU#%d, " - "should never happen.\n", vector, smp_processor_id()); +/** + * spurious_interrupt - Catch all for interrupts raised on unused vectors + * @regs: Pointer to pt_regs on stack + * @vector: The vector number + * + * This is invoked from ASM entry code to catch all interrupts which + * trigger on an entry which is routed to the common_spurious idtentry + * point. + */ +DEFINE_IDTENTRY_IRQ(spurious_interrupt) +{ + handle_spurious_interrupt(vector); +} - trace_spurious_apic_exit(vector); - exiting_irq(); +DEFINE_IDTENTRY_SYSVEC(sysvec_spurious_apic_interrupt) +{ + handle_spurious_interrupt(SPURIOUS_APIC_VECTOR); } /* * This interrupt should never happen with our APIC/SMP architecture */ -__visible void __irq_entry smp_error_interrupt(struct pt_regs *regs) +DEFINE_IDTENTRY_SYSVEC(sysvec_error_interrupt) { static const char * const error_interrupt_reason[] = { "Send CS error", /* APIC Error Bit 0 */ @@ -2070,31 +2173,28 @@ __visible void __irq_entry smp_error_interrupt(struct pt_regs *regs) }; u32 v, i = 0; - entering_irq(); trace_error_apic_entry(ERROR_APIC_VECTOR); /* First tickle the hardware, only then report what went on. -- REW */ if (lapic_get_maxlvt() > 3) /* Due to the Pentium erratum 3AP. */ apic_write(APIC_ESR, 0); v = apic_read(APIC_ESR); - ack_APIC_irq(); + apic_eoi(); atomic_inc(&irq_err_count); - apic_printk(APIC_DEBUG, KERN_DEBUG "APIC error on CPU%d: %02x", - smp_processor_id(), v); + apic_pr_debug("APIC error on CPU%d: %02x", smp_processor_id(), v); v &= 0xff; while (v) { if (v & 0x1) - apic_printk(APIC_DEBUG, KERN_CONT " : %s", error_interrupt_reason[i]); + apic_pr_debug_cont(" : %s", error_interrupt_reason[i]); i++; v >>= 1; } - apic_printk(APIC_DEBUG, KERN_CONT "\n"); + apic_pr_debug_cont("\n"); trace_error_apic_exit(ERROR_APIC_VECTOR); - exiting_irq(); } /** @@ -2112,8 +2212,7 @@ static void __init connect_bsp_APIC(void) * PIC mode, enable APIC mode in the IMCR, i.e. connect BSP's * local APIC to INT and NMI lines. */ - apic_printk(APIC_VERBOSE, "leaving PIC mode, " - "enabling APIC mode.\n"); + apic_pr_verbose("Leaving PIC mode, enabling APIC mode.\n"); imcr_pic_to_apic(); } #endif @@ -2138,8 +2237,7 @@ void disconnect_bsp_APIC(int virt_wire_setup) * IPIs, won't work beyond this point! The only exception are * INIT IPIs. */ - apic_printk(APIC_VERBOSE, "disabling APIC mode, " - "entering PIC mode.\n"); + apic_pr_verbose("Disabling APIC mode, entering PIC mode.\n"); imcr_apic_to_pic(); return; } @@ -2184,236 +2282,56 @@ void disconnect_bsp_APIC(int virt_wire_setup) apic_write(APIC_LVT1, value); } -/* - * The number of allocated logical CPU IDs. Since logical CPU IDs are allocated - * contiguously, it equals to current allocated max logical CPU ID plus 1. - * All allocated CPU IDs should be in the [0, nr_logical_cpuids) range, - * so the maximum of nr_logical_cpuids is nr_cpu_ids. - * - * NOTE: Reserve 0 for BSP. - */ -static int nr_logical_cpuids = 1; - -/* - * Used to store mapping between logical CPU IDs and APIC IDs. - */ -static int cpuid_to_apicid[] = { - [0 ... NR_CPUS - 1] = -1, -}; - -#ifdef CONFIG_SMP -/** - * apic_id_is_primary_thread - Check whether APIC ID belongs to a primary thread - * @id: APIC ID to check - */ -bool apic_id_is_primary_thread(unsigned int apicid) +void __irq_msi_compose_msg(struct irq_cfg *cfg, struct msi_msg *msg, + bool dmar) { - u32 mask; - - if (smp_num_siblings == 1) - return true; - /* Isolate the SMT bit(s) in the APICID and check for 0 */ - mask = (1U << (fls(smp_num_siblings) - 1)) - 1; - return !(apicid & mask); -} -#endif - -/* - * Should use this API to allocate logical CPU IDs to keep nr_logical_cpuids - * and cpuid_to_apicid[] synchronized. - */ -static int allocate_logical_cpuid(int apicid) -{ - int i; - - /* - * cpuid <-> apicid mapping is persistent, so when a cpu is up, - * check if the kernel has allocated a cpuid for it. - */ - for (i = 0; i < nr_logical_cpuids; i++) { - if (cpuid_to_apicid[i] == apicid) - return i; - } - - /* Allocate a new cpuid. */ - if (nr_logical_cpuids >= nr_cpu_ids) { - WARN_ONCE(1, "APIC: NR_CPUS/possible_cpus limit of %u reached. " - "Processor %d/0x%x and the rest are ignored.\n", - nr_cpu_ids, nr_logical_cpuids, apicid); - return -EINVAL; - } - - cpuid_to_apicid[nr_logical_cpuids] = apicid; - return nr_logical_cpuids++; -} - -int generic_processor_info(int apicid, int version) -{ - int cpu, max = nr_cpu_ids; - bool boot_cpu_detected = physid_isset(boot_cpu_physical_apicid, - phys_cpu_present_map); - - /* - * boot_cpu_physical_apicid is designed to have the apicid - * returned by read_apic_id(), i.e, the apicid of the - * currently booting-up processor. However, on some platforms, - * it is temporarily modified by the apicid reported as BSP - * through MP table. Concretely: - * - * - arch/x86/kernel/mpparse.c: MP_processor_info() - * - arch/x86/mm/amdtopology.c: amd_numa_init() - * - * This function is executed with the modified - * boot_cpu_physical_apicid. So, disabled_cpu_apicid kernel - * parameter doesn't work to disable APs on kdump 2nd kernel. - * - * Since fixing handling of boot_cpu_physical_apicid requires - * another discussion and tests on each platform, we leave it - * for now and here we use read_apic_id() directly in this - * function, generic_processor_info(). - */ - if (disabled_cpu_apicid != BAD_APICID && - disabled_cpu_apicid != read_apic_id() && - disabled_cpu_apicid == apicid) { - int thiscpu = num_processors + disabled_cpus; - - pr_warning("APIC: Disabling requested cpu." - " Processor %d/0x%x ignored.\n", - thiscpu, apicid); - - disabled_cpus++; - return -ENODEV; - } - - /* - * If boot cpu has not been detected yet, then only allow upto - * nr_cpu_ids - 1 processors and keep one slot free for boot cpu - */ - if (!boot_cpu_detected && num_processors >= nr_cpu_ids - 1 && - apicid != boot_cpu_physical_apicid) { - int thiscpu = max + disabled_cpus - 1; - - pr_warning( - "APIC: NR_CPUS/possible_cpus limit of %i almost" - " reached. Keeping one slot for boot cpu." - " Processor %d/0x%x ignored.\n", max, thiscpu, apicid); - - disabled_cpus++; - return -ENODEV; - } - - if (num_processors >= nr_cpu_ids) { - int thiscpu = max + disabled_cpus; - - pr_warning("APIC: NR_CPUS/possible_cpus limit of %i " - "reached. Processor %d/0x%x ignored.\n", - max, thiscpu, apicid); - - disabled_cpus++; - return -EINVAL; - } + memset(msg, 0, sizeof(*msg)); - if (apicid == boot_cpu_physical_apicid) { - /* - * x86_bios_cpu_apicid is required to have processors listed - * in same order as logical cpu numbers. Hence the first - * entry is BSP, and so on. - * boot_cpu_init() already hold bit 0 in cpu_present_mask - * for BSP. - */ - cpu = 0; + msg->arch_addr_lo.base_address = X86_MSI_BASE_ADDRESS_LOW; + msg->arch_addr_lo.dest_mode_logical = apic->dest_mode_logical; + msg->arch_addr_lo.destid_0_7 = cfg->dest_apicid & 0xFF; - /* Logical cpuid 0 is reserved for BSP. */ - cpuid_to_apicid[0] = apicid; - } else { - cpu = allocate_logical_cpuid(apicid); - if (cpu < 0) { - disabled_cpus++; - return -EINVAL; - } - } + msg->arch_data.delivery_mode = APIC_DELIVERY_MODE_FIXED; + msg->arch_data.vector = cfg->vector; + msg->address_hi = X86_MSI_BASE_ADDRESS_HIGH; /* - * Validate version + * Only the IOMMU itself can use the trick of putting destination + * APIC ID into the high bits of the address. Anything else would + * just be writing to memory if it tried that, and needs IR to + * address APICs which can't be addressed in the normal 32-bit + * address range at 0xFFExxxxx. That is typically just 8 bits, but + * some hypervisors allow the extended destination ID field in bits + * 5-11 to be used, giving support for 15 bits of APIC IDs in total. */ - if (version == 0x0) { - pr_warning("BIOS bug: APIC version is 0 for CPU %d/0x%x, fixing up to 0x10\n", - cpu, apicid); - version = 0x10; - } - - if (version != boot_cpu_apic_version) { - pr_warning("BIOS bug: APIC version mismatch, boot CPU: %x, CPU %d: version %x\n", - boot_cpu_apic_version, cpu, version); - } - - if (apicid > max_physical_apicid) - max_physical_apicid = apicid; - -#if defined(CONFIG_SMP) || defined(CONFIG_X86_64) - early_per_cpu(x86_cpu_to_apicid, cpu) = apicid; - early_per_cpu(x86_bios_cpu_apicid, cpu) = apicid; -#endif -#ifdef CONFIG_X86_32 - early_per_cpu(x86_cpu_to_logical_apicid, cpu) = - apic->x86_32_early_logical_apicid(cpu); -#endif - set_cpu_possible(cpu, true); - physid_set(apicid, phys_cpu_present_map); - set_cpu_present(cpu, true); - num_processors++; - - return cpu; + if (dmar) + msg->arch_addr_hi.destid_8_31 = cfg->dest_apicid >> 8; + else if (virt_ext_dest_id && cfg->dest_apicid < 0x8000) + msg->arch_addr_lo.virt_destid_8_14 = cfg->dest_apicid >> 8; + else + WARN_ON_ONCE(cfg->dest_apicid > 0xFF); } -int hard_smp_processor_id(void) +u32 x86_msi_msg_get_destid(struct msi_msg *msg, bool extid) { - return read_apic_id(); -} + u32 dest = msg->arch_addr_lo.destid_0_7; -/* - * Override the generic EOI implementation with an optimized version. - * Only called during early boot when only one CPU is active and with - * interrupts disabled, so we know this does not race with actual APIC driver - * use. - */ -void __init apic_set_eoi_write(void (*eoi_write)(u32 reg, u32 v)) -{ - struct apic **drv; - - for (drv = __apicdrivers; drv < __apicdrivers_end; drv++) { - /* Should happen once for each apic */ - WARN_ON((*drv)->eoi_write == eoi_write); - (*drv)->native_eoi_write = (*drv)->eoi_write; - (*drv)->eoi_write = eoi_write; - } + if (extid) + dest |= msg->arch_addr_hi.destid_8_31 << 8; + return dest; } +EXPORT_SYMBOL_FOR_KVM(x86_msi_msg_get_destid); static void __init apic_bsp_up_setup(void) { -#ifdef CONFIG_X86_64 - apic_write(APIC_ID, apic->set_apic_id(boot_cpu_physical_apicid)); -#else - /* - * Hack: In case of kdump, after a crash, kernel might be booting - * on a cpu with non-zero lapic id. But boot_cpu_physical_apicid - * might be zero if read from MP tables. Get it from LAPIC. - */ -# ifdef CONFIG_CRASH_DUMP - boot_cpu_physical_apicid = read_apic_id(); -# endif -#endif - physid_set_mask_of_physid(boot_cpu_physical_apicid, &phys_cpu_present_map); + reset_phys_cpu_present_map(boot_cpu_physical_apicid); } /** * apic_bsp_setup - Setup function for local apic and io-apic * @upmode: Force UP mode (for APIC_init_uniprocessor) - * - * Returns: - * apic_id of BSP APIC */ -void __init apic_bsp_setup(bool upmode) +static void __init apic_bsp_setup(bool upmode) { connect_bsp_APIC(); if (upmode) @@ -2424,6 +2342,7 @@ void __init apic_bsp_setup(bool upmode) end_local_APIC_setup(); irq_remap_enable_fault_handling(); setup_IO_APIC(); + lapic_update_legacy_vectors(); } #ifdef CONFIG_UP_LATE_INIT @@ -2450,7 +2369,7 @@ static struct { */ int active; /* r/w apic fields */ - unsigned int apic_id; + u32 apic_id; unsigned int apic_taskpri; unsigned int apic_ldr; unsigned int apic_dfr; @@ -2466,7 +2385,7 @@ static struct { unsigned int apic_cmci; } apic_pm_state; -static int lapic_suspend(void) +static int lapic_suspend(void *data) { unsigned long flags; int maxlvt; @@ -2499,6 +2418,13 @@ static int lapic_suspend(void) #endif local_irq_save(flags); + + /* + * Mask IOAPIC before disabling the local APIC to prevent stale IRR + * entries on some implementations. + */ + mask_ioapic_entries(); + disable_local_APIC(); irq_remapping_disable(); @@ -2507,7 +2433,7 @@ static int lapic_suspend(void) return 0; } -static void lapic_resume(void) +static void lapic_resume(void *data) { unsigned int l, h; unsigned long flags; @@ -2582,11 +2508,15 @@ static void lapic_resume(void) * are needed on every CPU up until machine_halt/restart/poweroff. */ -static struct syscore_ops lapic_syscore_ops = { +static const struct syscore_ops lapic_syscore_ops = { .resume = lapic_resume, .suspend = lapic_suspend, }; +static struct syscore lapic_syscore = { + .ops = &lapic_syscore_ops, +}; + static void apic_pm_activate(void) { apic_pm_state.active = 1; @@ -2596,7 +2526,7 @@ static int __init init_lapic_sysfs(void) { /* XXX: remove suspend/resume procs if !apic_pm_state.active? */ if (boot_cpu_has(X86_FEATURE_APIC)) - register_syscore_ops(&lapic_syscore_ops); + register_syscore(&lapic_syscore); return 0; } @@ -2663,19 +2593,12 @@ int apic_is_clustered_box(void) /* * APIC command line parameters */ -static int __init setup_disableapic(char *arg) +static int __init setup_nolapic(char *arg) { - disable_apic = 1; + apic_is_disabled = true; setup_clear_cpu_cap(X86_FEATURE_APIC); return 0; } -early_param("disableapic", setup_disableapic); - -/* same as disableapic, for compatibility */ -static int __init setup_nolapic(char *arg) -{ - return setup_disableapic(arg); -} early_param("nolapic", setup_nolapic); static int __init parse_lapic_timer_c2_ok(char *arg) @@ -2702,11 +2625,11 @@ early_param("nolapic_timer", parse_nolapic_timer); static int __init apic_set_verbosity(char *arg) { if (!arg) { -#ifdef CONFIG_X86_64 - skip_ioapic_setup = 0; + if (IS_ENABLED(CONFIG_X86_32)) + return -EINVAL; + + ioapic_is_disabled = false; return 0; -#endif - return -EINVAL; } if (strcmp("debug", arg) == 0) @@ -2715,7 +2638,7 @@ static int __init apic_set_verbosity(char *arg) apic_verbosity = APIC_VERBOSE; #ifdef CONFIG_X86_64 else { - pr_warning("APIC Verbosity level %s not recognised" + pr_warn("APIC Verbosity level %s not recognised" " use apic=verbose or apic=debug\n", arg); return -EINVAL; } @@ -2727,11 +2650,11 @@ early_param("apic", apic_set_verbosity); static int __init lapic_insert_resource(void) { - if (!apic_phys) + if (!apic_mmio_base) return -1; /* Put local APIC into the resource map. */ - lapic_resource.start = apic_phys; + lapic_resource.start = apic_mmio_base; lapic_resource.end = lapic_resource.start + PAGE_SIZE - 1; insert_resource(&iomem_resource, &lapic_resource); @@ -2744,15 +2667,6 @@ static int __init lapic_insert_resource(void) */ late_initcall(lapic_insert_resource); -static int __init apic_set_disabled_cpu_apicid(char *arg) -{ - if (!arg || !get_option(&arg, &disabled_cpu_apicid)) - return -EINVAL; - - return 0; -} -early_param("disable_cpu_apicid", apic_set_disabled_cpu_apicid); - static int __init apic_set_extnmi(char *arg) { if (!arg) |