diff options
Diffstat (limited to 'arch/x86/kernel/apic')
| -rw-r--r-- | arch/x86/kernel/apic/Makefile | 15 | ||||
| -rw-r--r-- | arch/x86/kernel/apic/apic.c | 2212 | ||||
| -rw-r--r-- | arch/x86/kernel/apic/apic_64.c | 89 | ||||
| -rw-r--r-- | arch/x86/kernel/apic/apic_flat_64.c | 389 | ||||
| -rw-r--r-- | arch/x86/kernel/apic/io_apic.c | 4160 | ||||
| -rw-r--r-- | arch/x86/kernel/apic/ipi.c | 164 | ||||
| -rw-r--r-- | arch/x86/kernel/apic/nmi.c | 564 | ||||
| -rw-r--r-- | arch/x86/kernel/apic/x2apic_cluster.c | 243 | ||||
| -rw-r--r-- | arch/x86/kernel/apic/x2apic_phys.c | 229 | ||||
| -rw-r--r-- | arch/x86/kernel/apic/x2apic_uv_x.c | 643 |
10 files changed, 8708 insertions, 0 deletions
diff --git a/arch/x86/kernel/apic/Makefile b/arch/x86/kernel/apic/Makefile new file mode 100644 index 000000000000..da20b70c4000 --- /dev/null +++ b/arch/x86/kernel/apic/Makefile @@ -0,0 +1,15 @@ +# +# Makefile for local APIC drivers and for the IO-APIC code +# + +obj-y := apic.o ipi.o nmi.o +obj-$(CONFIG_X86_IO_APIC) += io_apic.o +obj-$(CONFIG_SMP) += ipi.o + +ifeq ($(CONFIG_X86_64),y) +obj-y += apic_64.o apic_flat_64.o +obj-$(CONFIG_X86_X2APIC) += x2apic_cluster.o +obj-$(CONFIG_X86_X2APIC) += x2apic_phys.o +obj-$(CONFIG_X86_UV) += x2apic_uv_x.o +endif + diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c new file mode 100644 index 000000000000..c12823eb55b5 --- /dev/null +++ b/arch/x86/kernel/apic/apic.c @@ -0,0 +1,2212 @@ +/* + * Local APIC handling, local APIC timers + * + * (c) 1999, 2000, 2009 Ingo Molnar <mingo@redhat.com> + * + * Fixes + * Maciej W. Rozycki : Bits for genuine 82489DX APICs; + * thanks to Eric Gilmore + * and Rolf G. Tews + * for testing these extensively. + * Maciej W. Rozycki : Various updates and fixes. + * Mikael Pettersson : Power Management for UP-APIC. + * Pavel Machek and + * Mikael Pettersson : PM converted to driver model. + */ + +#include <linux/kernel_stat.h> +#include <linux/mc146818rtc.h> +#include <linux/acpi_pmtmr.h> +#include <linux/clockchips.h> +#include <linux/interrupt.h> +#include <linux/bootmem.h> +#include <linux/ftrace.h> +#include <linux/ioport.h> +#include <linux/module.h> +#include <linux/sysdev.h> +#include <linux/delay.h> +#include <linux/timex.h> +#include <linux/dmar.h> +#include <linux/init.h> +#include <linux/cpu.h> +#include <linux/dmi.h> +#include <linux/nmi.h> +#include <linux/smp.h> +#include <linux/mm.h> + +#include <asm/arch_hooks.h> +#include <asm/pgalloc.h> +#include <asm/atomic.h> +#include <asm/mpspec.h> +#include <asm/i8253.h> +#include <asm/i8259.h> +#include <asm/proto.h> +#include <asm/apic.h> +#include <asm/desc.h> +#include <asm/hpet.h> +#include <asm/idle.h> +#include <asm/mtrr.h> +#include <asm/smp.h> + +unsigned int num_processors; + +unsigned disabled_cpus __cpuinitdata; + +/* Processor that is doing the boot up */ +unsigned int boot_cpu_physical_apicid = -1U; + +/* + * The highest APIC ID seen during enumeration. + * + * This determines the messaging protocol we can use: if all APIC IDs + * are in the 0 ... 7 range, then we can use logical addressing which + * has some performance advantages (better broadcasting). + * + * If there's an APIC ID above 8, we use physical addressing. + */ +unsigned int max_physical_apicid; + +/* + * Bitmask of physically existing CPUs: + */ +physid_mask_t phys_cpu_present_map; + +/* + * Map cpu index to physical APIC ID + */ +DEFINE_EARLY_PER_CPU(u16, x86_cpu_to_apicid, BAD_APICID); +DEFINE_EARLY_PER_CPU(u16, x86_bios_cpu_apicid, BAD_APICID); +EXPORT_EARLY_PER_CPU_SYMBOL(x86_cpu_to_apicid); +EXPORT_EARLY_PER_CPU_SYMBOL(x86_bios_cpu_apicid); + +#ifdef CONFIG_X86_32 +/* + * Knob to control our willingness to enable the local APIC. + * + * +1=force-enable + */ +static int force_enable_local_apic; +/* + * APIC command line parameters + */ +static int __init parse_lapic(char *arg) +{ + force_enable_local_apic = 1; + return 0; +} +early_param("lapic", parse_lapic); +/* Local APIC was disabled by the BIOS and enabled by the kernel */ +static int enabled_via_apicbase; + +#endif + +#ifdef CONFIG_X86_64 +static int apic_calibrate_pmtmr __initdata; +static __init int setup_apicpmtimer(char *s) +{ + apic_calibrate_pmtmr = 1; + notsc_setup(NULL); + return 0; +} +__setup("apicpmtimer", setup_apicpmtimer); +#endif + +#ifdef CONFIG_X86_X2APIC +int x2apic; +/* x2apic enabled before OS handover */ +static int x2apic_preenabled; +static int disable_x2apic; +static __init int setup_nox2apic(char *str) +{ + disable_x2apic = 1; + setup_clear_cpu_cap(X86_FEATURE_X2APIC); + return 0; +} +early_param("nox2apic", setup_nox2apic); +#endif + +unsigned long mp_lapic_addr; +int disable_apic; +/* Disable local APIC timer from the kernel commandline or via dmi quirk */ +static int disable_apic_timer __cpuinitdata; +/* Local APIC timer works in C2 */ +int local_apic_timer_c2_ok; +EXPORT_SYMBOL_GPL(local_apic_timer_c2_ok); + +int first_system_vector = 0xfe; + +/* + * Debug level, exported for io_apic.c + */ +unsigned int apic_verbosity; + +int pic_mode; + +/* Have we found an MP table */ +int smp_found_config; + +static struct resource lapic_resource = { + .name = "Local APIC", + .flags = IORESOURCE_MEM | IORESOURCE_BUSY, +}; + +static unsigned int calibration_result; + +static int lapic_next_event(unsigned long delta, + struct clock_event_device *evt); +static void lapic_timer_setup(enum clock_event_mode mode, + struct clock_event_device *evt); +static void lapic_timer_broadcast(const struct cpumask *mask); +static void apic_pm_activate(void); + +/* + * The local apic timer can be used for any function which is CPU local. + */ +static struct clock_event_device lapic_clockevent = { + .name = "lapic", + .features = CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT + | CLOCK_EVT_FEAT_C3STOP | CLOCK_EVT_FEAT_DUMMY, + .shift = 32, + .set_mode = lapic_timer_setup, + .set_next_event = lapic_next_event, + .broadcast = lapic_timer_broadcast, + .rating = 100, + .irq = -1, +}; +static DEFINE_PER_CPU(struct clock_event_device, lapic_events); + +static unsigned long apic_phys; + +/* + * Get the LAPIC version + */ +static inline int lapic_get_version(void) +{ + return GET_APIC_VERSION(apic_read(APIC_LVR)); +} + +/* + * Check, if the APIC is integrated or a separate chip + */ +static inline int lapic_is_integrated(void) +{ +#ifdef CONFIG_X86_64 + return 1; +#else + return APIC_INTEGRATED(lapic_get_version()); +#endif +} + +/* + * Check, whether this is a modern or a first generation APIC + */ +static int modern_apic(void) +{ + /* AMD systems use old APIC versions, so check the CPU */ + if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD && + boot_cpu_data.x86 >= 0xf) + return 1; + return lapic_get_version() >= 0x14; +} + +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; + udelay(100); + } while (timeout++ < 1000); + + return send_status; +} + +void native_apic_icr_write(u32 low, u32 id) +{ + apic_write(APIC_ICR2, SET_APIC_DEST_FIELD(id)); + apic_write(APIC_ICR, low); +} + +u64 native_apic_icr_read(void) +{ + u32 icr1, icr2; + + icr2 = apic_read(APIC_ICR2); + icr1 = apic_read(APIC_ICR); + + return icr1 | ((u64)icr2 << 32); +} + +/** + * enable_NMI_through_LVT0 - enable NMI through local vector table 0 + */ +void __cpuinit enable_NMI_through_LVT0(void) +{ + unsigned int v; + + /* unmask and set to NMI */ + v = APIC_DM_NMI; + + /* Level triggered for 82489DX (32bit mode) */ + if (!lapic_is_integrated()) + v |= APIC_LVT_LEVEL_TRIGGER; + + apic_write(APIC_LVT0, v); +} + +#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 + */ +int lapic_get_maxlvt(void) +{ + unsigned int v; + + v = apic_read(APIC_LVR); + /* + * - we always have APIC integrated on 64bit mode + * - 82489DXs do not report # of LVT entries + */ + return APIC_INTEGRATED(GET_APIC_VERSION(v)) ? GET_APIC_MAXLVT(v) : 2; +} + +/* + * Local APIC timer + */ + +/* Clock divisor */ +#define APIC_DIVISOR 16 + +/* + * This function sets up the local APIC timer, with a timeout of + * 'clocks' APIC bus clock. During calibration we actually call + * this function twice on the boot CPU, once with a bogus timeout + * value, second time for real. The other (noncalibrating) CPUs + * call this function only once, with the real, calibrated value. + * + * We do reads before writes even if unnecessary, to get around the + * P5 APIC double write bug. + */ +static void __setup_APIC_LVTT(unsigned int clocks, int oneshot, int irqen) +{ + unsigned int lvtt_value, tmp_value; + + lvtt_value = LOCAL_TIMER_VECTOR; + if (!oneshot) + lvtt_value |= APIC_LVT_TIMER_PERIODIC; + if (!lapic_is_integrated()) + lvtt_value |= SET_APIC_TIMER_BASE(APIC_TIMER_BASE_DIV); + + if (!irqen) + lvtt_value |= APIC_LVT_MASKED; + + apic_write(APIC_LVTT, lvtt_value); + + /* + * Divide PICLK by 16 + */ + tmp_value = apic_read(APIC_TDCR); + apic_write(APIC_TDCR, + (tmp_value & ~(APIC_TDR_DIV_1 | APIC_TDR_DIV_TMBASE)) | + APIC_TDR_DIV_16); + + if (!oneshot) + apic_write(APIC_TMICT, clocks / APIC_DIVISOR); +} + +/* + * Setup extended LVT, AMD specific (K8, family 10h) + * + * Vector mappings are hard coded. On K8 only offset 0 (APIC500) and + * MCE interrupts are supported. Thus MCE offset must be set to 0. + * + * If mask=1, the LVT entry does not generate interrupts while mask=0 + * enables the vector. See also the BKDGs. + */ + +#define APIC_EILVT_LVTOFF_MCE 0 +#define APIC_EILVT_LVTOFF_IBS 1 + +static void setup_APIC_eilvt(u8 lvt_off, u8 vector, u8 msg_type, u8 mask) +{ + unsigned long reg = (lvt_off << 4) + APIC_EILVT0; + unsigned int v = (mask << 16) | (msg_type << 8) | vector; + + apic_write(reg, v); +} + +u8 setup_APIC_eilvt_mce(u8 vector, u8 msg_type, u8 mask) +{ + setup_APIC_eilvt(APIC_EILVT_LVTOFF_MCE, vector, msg_type, mask); + return APIC_EILVT_LVTOFF_MCE; +} + +u8 setup_APIC_eilvt_ibs(u8 vector, u8 msg_type, u8 mask) +{ + setup_APIC_eilvt(APIC_EILVT_LVTOFF_IBS, vector, msg_type, mask); + return APIC_EILVT_LVTOFF_IBS; +} +EXPORT_SYMBOL_GPL(setup_APIC_eilvt_ibs); + +/* + * Program the next event, relative to now + */ +static int lapic_next_event(unsigned long delta, + struct clock_event_device *evt) +{ + apic_write(APIC_TMICT, delta); + return 0; +} + +/* + * Setup the lapic timer in periodic or oneshot mode + */ +static void lapic_timer_setup(enum clock_event_mode mode, + struct clock_event_device *evt) +{ + unsigned long flags; + unsigned int v; + + /* Lapic used as dummy for broadcast ? */ + if (evt->features & CLOCK_EVT_FEAT_DUMMY) + return; + + local_irq_save(flags); + + switch (mode) { + case CLOCK_EVT_MODE_PERIODIC: + case CLOCK_EVT_MODE_ONESHOT: + __setup_APIC_LVTT(calibration_result, + mode != CLOCK_EVT_MODE_PERIODIC, 1); + break; + case CLOCK_EVT_MODE_UNUSED: + case CLOCK_EVT_MODE_SHUTDOWN: + v = apic_read(APIC_LVTT); + v |= (APIC_LVT_MASKED | LOCAL_TIMER_VECTOR); + apic_write(APIC_LVTT, v); + apic_write(APIC_TMICT, 0xffffffff); + break; + case CLOCK_EVT_MODE_RESUME: + /* Nothing to do here */ + break; + } + + local_irq_restore(flags); +} + +/* + * Local APIC timer broadcast function + */ +static void lapic_timer_broadcast(const struct cpumask *mask) +{ +#ifdef CONFIG_SMP + apic->send_IPI_mask(mask, LOCAL_TIMER_VECTOR); +#endif +} + +/* + * Setup the local APIC timer for this CPU. Copy the initilized values + * of the boot CPU and register the clock event in the framework. + */ +static void __cpuinit setup_APIC_timer(void) +{ + struct clock_event_device *levt = &__get_cpu_var(lapic_events); + + memcpy(levt, &lapic_clockevent, sizeof(*levt)); + levt->cpumask = cpumask_of(smp_processor_id()); + + clockevents_register_device(levt); +} + +/* + * 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 + * frequency. + * + * This was previously done by reading the PIT/HPET and waiting for a wrap + * around to find out, that a tick has elapsed. I have a box, where the PIT + * readout is broken, so it never gets out of the wait loop again. This was + * also reported by others. + * + * Monitoring the jiffies value is inaccurate and the clockevents + * infrastructure allows us to do a simple substitution of the interrupt + * handler. + * + * The calibration routine also uses the pm_timer when possible, as the PIT + * happens to run way too slow (factor 2.3 on my VAIO CoreDuo, which goes + * back to normal later in the boot process). + */ + +#define LAPIC_CAL_LOOPS (HZ/10) + +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 unsigned long lapic_cal_j1, lapic_cal_j2; + +/* + * Temporary interrupt handler. + */ +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(); + + if (cpu_has_tsc) + rdtscll(tsc); + + switch (lapic_cal_loops++) { + case 0: + lapic_cal_t1 = tapic; + lapic_cal_tsc1 = tsc; + lapic_cal_pm1 = pm; + lapic_cal_j1 = jiffies; + break; + + case LAPIC_CAL_LOOPS: + lapic_cal_t2 = tapic; + lapic_cal_tsc2 = tsc; + if (pm < lapic_cal_pm1) + pm += ACPI_PM_OVRRUN; + lapic_cal_pm2 = pm; + lapic_cal_j2 = jiffies; + break; + } +} + +static int __init +calibrate_by_pmtimer(long deltapm, long *delta, long *deltatsc) +{ + const long pm_100ms = PMTMR_TICKS_PER_SEC / 10; + const long pm_thresh = pm_100ms / 100; + unsigned long mult; + u64 res; + +#ifndef CONFIG_X86_PM_TIMER + return -1; +#endif + + apic_printk(APIC_VERBOSE, "... PM-Timer delta = %ld\n", deltapm); + + /* Check, if the PM timer is available */ + if (!deltapm) + return -1; + + mult = clocksource_hz2mult(PMTMR_TICKS_PER_SEC, 22); + + if (deltapm > (pm_100ms - pm_thresh) && + deltapm < (pm_100ms + pm_thresh)) { + apic_printk(APIC_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); + + /* Correct the lapic counter value */ + res = (((u64)(*delta)) * pm_100ms); + do_div(res, deltapm); + pr_info("APIC delta adjusted to PM-Timer: " + "%lu (%ld)\n", (unsigned long)res, *delta); + *delta = (long)res; + + /* Correct the tsc counter value */ + if (cpu_has_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); + *deltatsc = (long)res; + } + + return 0; +} + +static int __init calibrate_APIC_clock(void) +{ + struct clock_event_device *levt = &__get_cpu_var(lapic_events); + void (*real_handler)(struct clock_event_device *dev); + unsigned long deltaj; + long delta, deltatsc; + int pm_referenced = 0; + + 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 */ + local_irq_enable(); + + while (lapic_cal_loops <= LAPIC_CAL_LOOPS) + cpu_relax(); + + local_irq_disable(); + + /* Restore the real event handler */ + global_clock_event->event_handler = real_handler; + + /* 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); + + deltatsc = (long)(lapic_cal_tsc2 - lapic_cal_tsc1); + + /* we trust the PM based calibration if possible */ + 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(0x7FFFFF, &lapic_clockevent); + lapic_clockevent.min_delta_ns = + clockevent_delta2ns(0xF, &lapic_clockevent); + + calibration_result = (delta * APIC_DIVISOR) / LAPIC_CAL_LOOPS; + + apic_printk(APIC_VERBOSE, "..... delta %ld\n", delta); + apic_printk(APIC_VERBOSE, "..... mult: %ld\n", lapic_clockevent.mult); + apic_printk(APIC_VERBOSE, "..... calibration result: %u\n", + calibration_result); + + if (cpu_has_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)); + } + + apic_printk(APIC_VERBOSE, "..... host bus clock speed is " + "%u.%04u MHz.\n", + calibration_result / (1000000 / HZ), + calibration_result % (1000000 / HZ)); + + /* + * Do a sanity check on the APIC calibration result + */ + if (calibration_result < (1000000 / HZ)) { + local_irq_enable(); + pr_warning("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 + */ + if (!pm_referenced) { + apic_printk(APIC_VERBOSE, "... verify APIC timer\n"); + + /* + * Setup the apic timer manually + */ + levt->event_handler = lapic_cal_handler; + lapic_timer_setup(CLOCK_EVT_MODE_PERIODIC, levt); + lapic_cal_loops = -1; + + /* Let the interrupts run */ + local_irq_enable(); + + while (lapic_cal_loops <= LAPIC_CAL_LOOPS) + cpu_relax(); + + /* Stop the lapic timer */ + lapic_timer_setup(CLOCK_EVT_MODE_SHUTDOWN, levt); + + /* Jiffies delta */ + deltaj = lapic_cal_j2 - lapic_cal_j1; + apic_printk(APIC_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"); + else + levt->features |= CLOCK_EVT_FEAT_DUMMY; + } else + local_irq_enable(); + + if (levt->features & CLOCK_EVT_FEAT_DUMMY) { + pr_warning("APIC timer disabled due to verification failure\n"); + return -1; + } + + return 0; +} + +/* + * Setup the boot APIC + * + * Calibrate and verify the result. + */ +void __init setup_boot_APIC_clock(void) +{ + /* + * The local apic timer can be disabled via the kernel + * commandline or from the CPU detection code. Register the lapic + * timer as a dummy clock event source on SMP systems, so the + * broadcast mechanism is used. On UP systems simply ignore it. + */ + if (disable_apic_timer) { + pr_info("Disabling APIC timer\n"); + /* No broadcast on UP ! */ + if (num_possible_cpus() > 1) { + lapic_clockevent.mult = 1; + setup_APIC_timer(); + } + return; + } + + apic_printk(APIC_VERBOSE, "Using local APIC timer interrupts.\n" + "calibrating APIC timer ...\n"); + + if (calibrate_APIC_clock()) { + /* No broadcast on UP ! */ + if (num_possible_cpus() > 1) + setup_APIC_timer(); + return; + } + + /* + * If nmi_watchdog is set to IO_APIC, we need the + * PIT/HPET going. Otherwise register lapic as a dummy + * device. + */ + if (nmi_watchdog != NMI_IO_APIC) + lapic_clockevent.features &= ~CLOCK_EVT_FEAT_DUMMY; + else + pr_warning("APIC timer registered as dummy," + " due to nmi_watchdog=%d!\n", nmi_watchdog); + + /* Setup the lapic or request the broadcast */ + setup_APIC_timer(); +} + +void __cpuinit setup_secondary_APIC_clock(void) +{ + setup_APIC_timer(); +} + +/* + * The guts of the apic timer interrupt + */ +static void local_apic_timer_interrupt(void) +{ + int cpu = smp_processor_id(); + struct clock_event_device *evt = &per_cpu(lapic_events, cpu); + + /* + * Normally we should not be here till LAPIC has been initialized but + * in some cases like kdump, its possible that there is a pending LAPIC + * timer interrupt from previous kernel's context and is delivered in + * new kernel the moment interrupts are enabled. + * + * Interrupts are enabled early and LAPIC is setup much later, hence + * its possible that when we get here evt->event_handler is NULL. + * Check for event_handler being NULL and discard the interrupt as + * spurious. + */ + if (!evt->event_handler) { + pr_warning("Spurious LAPIC timer interrupt on cpu %d\n", cpu); + /* Switch it off */ + lapic_timer_setup(CLOCK_EVT_MODE_SHUTDOWN, evt); + return; + } + + /* + * the NMI deadlock-detector uses this. + */ + inc_irq_stat(apic_timer_irqs); + + evt->event_handler(evt); +} + +/* + * Local APIC timer interrupt. This is the most natural way for doing + * local interrupts, but local timer interrupts can be emulated by + * broadcast interrupts too. [in case the hw doesn't support APIC timers] + * + * [ if a single-CPU system runs an SMP kernel then we call the local + * interrupt as well. Thus we cannot inline the local irq ... ] + */ +void __irq_entry smp_apic_timer_interrupt(struct pt_regs *regs) +{ + struct pt_regs *old_regs = set_irq_regs(regs); + + /* + * NOTE! We'd better ACK the irq immediately, + * because timer handling can be slow. + */ + ack_APIC_irq(); + /* + * 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. + */ + exit_idle(); + irq_enter(); + local_apic_timer_interrupt(); + irq_exit(); + + set_irq_regs(old_regs); +} + +int setup_profiling_timer(unsigned int multiplier) +{ + return -EINVAL; +} + +/* + * Local APIC start and shutdown + */ + +/** + * clear_local_APIC - shutdown the local APIC + * + * This is called, when a CPU is disabled and before rebooting, so the state of + * the local APIC has no dangling leftovers. Also used to cleanout any BIOS + * leftovers during boot. + */ +void clear_local_APIC(void) +{ + int maxlvt; + u32 v; + + /* APIC hasn't been mapped yet */ + if (!apic_phys) + return; + + maxlvt = lapic_get_maxlvt(); + /* + * Masking an LVT entry can trigger a local APIC error + * if the vector is zero. Mask LVTERR first to prevent this. + */ + if (maxlvt >= 3) { + v = ERROR_APIC_VECTOR; /* any non-zero vector will do */ + apic_write(APIC_LVTERR, v | APIC_LVT_MASKED); + } + /* + * Careful: we have to set masks only first to deassert + * any level-triggered sources. + */ + v = apic_read(APIC_LVTT); + apic_write(APIC_LVTT, v | APIC_LVT_MASKED); + v = apic_read(APIC_LVT0); + apic_write(APIC_LVT0, v | APIC_LVT_MASKED); + v = apic_read(APIC_LVT1); + apic_write(APIC_LVT1, v | APIC_LVT_MASKED); + if (maxlvt >= 4) { + v = apic_read(APIC_LVTPC); + apic_write(APIC_LVTPC, v | APIC_LVT_MASKED); + } + + /* lets not touch this if we didn't frob it */ +#if defined(CONFIG_X86_MCE_P4THERMAL) || defined(X86_MCE_INTEL) + if (maxlvt >= 5) { + v = apic_read(APIC_LVTTHMR); + apic_write(APIC_LVTTHMR, v | APIC_LVT_MASKED); + } +#endif + /* + * Clean APIC state for other OSs: + */ + apic_write(APIC_LVTT, APIC_LVT_MASKED); + apic_write(APIC_LVT0, APIC_LVT_MASKED); + apic_write(APIC_LVT1, APIC_LVT_MASKED); + if (maxlvt >= 3) + apic_write(APIC_LVTERR, APIC_LVT_MASKED); + if (maxlvt >= 4) + apic_write(APIC_LVTPC, APIC_LVT_MASKED); + + /* Integrated APIC (!82489DX) ? */ + if (lapic_is_integrated()) { + if (maxlvt > 3) + /* Clear ESR due to Pentium errata 3AP and 11AP */ + apic_write(APIC_ESR, 0); + apic_read(APIC_ESR); + } +} + +/** + * disable_local_APIC - clear and disable the local APIC + */ +void disable_local_APIC(void) +{ + unsigned int value; + + /* APIC hasn't been mapped yet */ + if (!apic_phys) + return; + + clear_local_APIC(); + + /* + * Disable APIC (implies clearing of registers + * for 82489DX!). + */ + value = apic_read(APIC_SPIV); + value &= ~APIC_SPIV_APIC_ENABLED; + apic_write(APIC_SPIV, value); + +#ifdef CONFIG_X86_32 + /* + * When LAPIC was disabled by the BIOS and enabled by the kernel, + * restore the disabled state. + */ + if (enabled_via_apicbase) { + unsigned int l, h; + + rdmsr(MSR_IA32_APICBASE, l, h); + l &= ~MSR_IA32_APICBASE_ENABLE; + wrmsr(MSR_IA32_APICBASE, l, h); + } +#endif +} + +/* + * If Linux enabled the LAPIC against the BIOS default disable it down before + * re-entering the BIOS on shutdown. Otherwise the BIOS may get confused and + * not power-off. Additionally clear all LVT entries before disable_local_APIC + * for the case where Linux didn't enable the LAPIC. + */ +void lapic_shutdown(void) +{ + unsigned long flags; + + if (!cpu_has_apic) + return; + + local_irq_save(flags); + +#ifdef CONFIG_X86_32 + if (!enabled_via_apicbase) + clear_local_APIC(); + else +#endif + disable_local_APIC(); + + + local_irq_restore(flags); +} + +/* + * This is to verify that we're looking at a real local APIC. + * Check these against your board if the CPUs aren't getting + * started for no apparent reason. + */ +int __init verify_local_APIC(void) +{ + unsigned int reg0, reg1; + + /* + * The version register is read-only in a real APIC. + */ + reg0 = apic_read(APIC_LVR); + apic_printk(APIC_DEBUG, "Getting VERSION: %x\n", reg0); + apic_write(APIC_LVR, reg0 ^ APIC_LVR_MASK); + reg1 = apic_read(APIC_LVR); + apic_printk(APIC_DEBUG, "Getting VERSION: %x\n", reg1); + + /* + * The two version reads above should print the same + * numbers. If the second one is different, then we + * poke at a non-APIC. + */ + if (reg1 != reg0) + return 0; + + /* + * Check if the version looks reasonably. + */ + reg1 = GET_APIC_VERSION(reg0); + if (reg1 == 0x00 || reg1 == 0xff) + return 0; + reg1 = lapic_get_maxlvt(); + if (reg1 < 0x02 || reg1 == 0xff) + return 0; + + /* + * The ID register is read/write in a real APIC. + */ + reg0 = apic_read(APIC_ID); + apic_printk(APIC_DEBUG, "Getting ID: %x\n", reg0); + apic_write(APIC_ID, reg0 ^ apic->apic_id_mask); + reg1 = apic_read(APIC_ID); + apic_printk(APIC_DEBUG, "Getting ID: %x\n", reg1); + apic_write(APIC_ID, reg0); + if (reg1 != (reg0 ^ apic->apic_id_mask)) + return 0; + + /* + * The next two are just to see if we have sane values. + * They're only really relevant if we're in Virtual Wire + * compatibility mode, but most boxes are anymore. + */ + reg0 = apic_read(APIC_LVT0); + apic_printk(APIC_DEBUG, "Getting LVT0: %x\n", reg0); + reg1 = apic_read(APIC_LVT1); + apic_printk(APIC_DEBUG, "Getting LVT1: %x\n", reg1); + + return 1; +} + +/** + * sync_Arb_IDs - synchronize APIC bus arbitration IDs + */ +void __init sync_Arb_IDs(void) +{ + /* + * Unsupported on P4 - see Intel Dev. Manual Vol. 3, Ch. 8.6.1 And not + * needed on AMD. + */ + if (modern_apic() || boot_cpu_data.x86_vendor == X86_VENDOR_AMD) + return; + + /* + * Wait for idle. + */ + 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); +} + +/* + * An initial setup of the virtual wire mode. + */ +void __init init_bsp_APIC(void) +{ + unsigned int value; + + /* + * Don't do the setup now if we have a SMP BIOS as the + * through-I/O-APIC virtual wire mode might be active. + */ + if (smp_found_config || !cpu_has_apic) + return; + + /* + * Do not trust the local APIC being empty at bootup. + */ + clear_local_APIC(); + + /* + * Enable APIC. + */ + value = apic_read(APIC_SPIV); + value &= ~APIC_VECTOR_MASK; + value |= APIC_SPIV_APIC_ENABLED; + +#ifdef CONFIG_X86_32 + /* This bit is reserved on P4/Xeon and should be cleared */ + if ((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) && + (boot_cpu_data.x86 == 15)) + value &= ~APIC_SPIV_FOCUS_DISABLED; + else +#endif + value |= APIC_SPIV_FOCUS_DISABLED; + value |= SPURIOUS_APIC_VECTOR; + apic_write(APIC_SPIV, value); + + /* + * Set up the virtual wire mode. + */ + apic_write(APIC_LVT0, APIC_DM_EXTINT); + value = APIC_DM_NMI; + if (!lapic_is_integrated()) /* 82489DX */ + value |= APIC_LVT_LEVEL_TRIGGER; + apic_write(APIC_LVT1, value); +} + +static void __cpuinit lapic_setup_esr(void) +{ + unsigned int oldvalue, value, maxlvt; + + if (!lapic_is_integrated()) { + pr_info("No ESR for 82489DX.\n"); + return; + } + + if (apic->disable_esr) { + /* + * Something untraceable is creating bad interrupts on + * secondary quads ... for the moment, just leave the + * ESR disabled - we can't do anything useful with the + * errors anyway - mbligh + */ + pr_info("Leaving ESR disabled.\n"); + return; + } + + maxlvt = lapic_get_maxlvt(); + if (maxlvt > 3) /* Due to the Pentium erratum 3AP. */ + apic_write(APIC_ESR, 0); + oldvalue = apic_read(APIC_ESR); + + /* enables sending errors */ + value = ERROR_APIC_VECTOR; + apic_write(APIC_LVTERR, value); + + /* + * spec says clear errors after enabling vector. + */ + 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); +} + + +/** + * setup_local_APIC - setup the local APIC + */ +void __cpuinit setup_local_APIC(void) +{ + unsigned int value; + int i, j; + + if (disable_apic) { + arch_disable_smp_support(); + return; + } + +#ifdef CONFIG_X86_32 + /* Pound the ESR really hard over the head with a big hammer - mbligh */ + if (lapic_is_integrated() && apic->disable_esr) { + apic_write(APIC_ESR, 0); + apic_write(APIC_ESR, 0); + apic_write(APIC_ESR, 0); + apic_write(APIC_ESR, 0); + } +#endif + + preempt_disable(); + + /* + * Double-check whether this APIC is really registered. + * This is meaningless in clustered apic mode, so we skip it. + */ + if (!apic->apic_id_registered()) + BUG(); + + /* + * 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... + */ + apic->init_apic_ldr(); + + /* + * Set Task Priority to 'accept all'. We never change this + * later on. + */ + value = apic_read(APIC_TASKPRI); + value &= ~APIC_TPRI_MASK; + apic_write(APIC_TASKPRI, value); + + /* + * 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. + */ + for (i = APIC_ISR_NR - 1; i >= 0; i--) { + value = apic_read(APIC_ISR + i*0x10); + for (j = 31; j >= 0; j--) { + if (value & (1<<j)) + ack_APIC_irq(); + } + } + + /* + * Now that we are all set up, enable the APIC + */ + value = apic_read(APIC_SPIV); + value &= ~APIC_VECTOR_MASK; + /* + * Enable APIC + */ + value |= APIC_SPIV_APIC_ENABLED; + +#ifdef CONFIG_X86_32 + /* + * Some unknown Intel IO/APIC (or APIC) errata is biting us with + * certain networking cards. If high frequency interrupts are + * happening on a particular IOAPIC pin, plus the IOAPIC routing + * entry is masked/unmasked at a high rate as well then sooner or + * later IOAPIC line gets 'stuck', no more interrupts are received + * from the device. If focus CPU is disabled then the hang goes + * away, oh well :-( + * + * [ This bug can be reproduced easily with a level-triggered + * PCI Ne2000 networking cards and PII/PIII processors, dual + * BX chipset. ] + */ + /* + * Actually disabling the focus CPU check just makes the hang less + * frequent as it makes the interrupt distributon model be more + * like LRU than MRU (the short-term load is more even across CPUs). + * See also the comment in end_level_ioapic_irq(). --macro + */ + + /* + * - enable focus processor (bit==0) + * - 64bit mode always use processor focus + * so no need to set it + */ + value &= ~APIC_SPIV_FOCUS_DISABLED; +#endif + + /* + * Set spurious IRQ vector + */ + value |= SPURIOUS_APIC_VECTOR; + apic_write(APIC_SPIV, value); + + /* + * Set up LVT0, LVT1: + * + * set up through-local-APIC on the BP's LINT0. This is not + * strictly necessary in pure symmetric-IO mode, but sometimes + * we delegate interrupts to the 8259A. + */ + /* + * TODO: set up through-local-APIC from through-I/O-APIC? --macro + */ + value = apic_read(APIC_LVT0) & APIC_LVT_MASKED; + if (!smp_processor_id() && (pic_mode || !value)) { + value = APIC_DM_EXTINT; + apic_printk(APIC_VERBOSE, "enabled ExtINT on CPU#%d\n", + smp_processor_id()); + } else { + value = APIC_DM_EXTINT | APIC_LVT_MASKED; + apic_printk(APIC_VERBOSE, "masked ExtINT on CPU#%d\n", + smp_processor_id()); + } + apic_write(APIC_LVT0, value); + + /* + * only the BP should see the LINT1 NMI signal, obviously. + */ + if (!smp_processor_id()) + value = APIC_DM_NMI; + else + value = APIC_DM_NMI | APIC_LVT_MASKED; + if (!lapic_is_integrated()) /* 82489DX */ + value |= APIC_LVT_LEVEL_TRIGGER; + apic_write(APIC_LVT1, value); + + preempt_enable(); +} + +void __cpuinit end_local_APIC_setup(void) +{ + lapic_setup_esr(); + +#ifdef CONFIG_X86_32 + { + unsigned int value; + /* Disable the local apic timer */ + value = apic_read(APIC_LVTT); + value |= (APIC_LVT_MASKED | LOCAL_TIMER_VECTOR); + apic_write(APIC_LVTT, value); + } +#endif + + setup_apic_nmi_watchdog(NULL); + apic_pm_activate(); +} + +#ifdef CONFIG_X86_X2APIC +void check_x2apic(void) +{ + int msr, msr2; + + if (!cpu_has_x2apic) + return; + + rdmsr(MSR_IA32_APICBASE, msr, msr2); + + if (msr & X2APIC_ENABLE) { + pr_info("x2apic enabled by BIOS, switching to x2apic ops\n"); + x2apic_preenabled = x2apic = 1; + } +} + +void enable_x2apic(void) +{ + int msr, msr2; + + if (!x2apic) + return; + + rdmsr(MSR_IA32_APICBASE, msr, msr2); + if (!(msr & X2APIC_ENABLE)) { + pr_info("Enabling x2apic\n"); + wrmsr(MSR_IA32_APICBASE, msr | X2APIC_ENABLE, 0); + } +} + +void __init enable_IR_x2apic(void) +{ +#ifdef CONFIG_INTR_REMAP + int ret; + unsigned long flags; + + if (!cpu_has_x2apic) + return; + + if (!x2apic_preenabled && disable_x2apic) { + pr_info("Skipped enabling x2apic and Interrupt-remapping " + "because of nox2apic\n"); + return; + } + + if (x2apic_preenabled && disable_x2apic) + panic("Bios already enabled x2apic, can't enforce nox2apic"); + + if (!x2apic_preenabled && skip_ioapic_setup) { + pr_info("Skipped enabling x2apic and Interrupt-remapping " + "because of skipping io-apic setup\n"); + return; + } + + ret = dmar_table_init(); + if (ret) { + pr_info("dmar_table_init() failed with %d:\n", ret); + + if (x2apic_preenabled) + panic("x2apic enabled by bios. But IR enabling failed"); + else + pr_info("Not enabling x2apic,Intr-remapping\n"); + return; + } + + local_irq_save(flags); + mask_8259A(); + + ret = save_mask_IO_APIC_setup(); + if (ret) { + pr_info("Saving IO-APIC state failed: %d\n", ret); + goto end; + } + + ret = enable_intr_remapping(1); + + if (ret && x2apic_preenabled) { + local_irq_restore(flags); + panic("x2apic enabled by bios. But IR enabling failed"); + } + + if (ret) + goto end_restore; + + if (!x2apic) { + x2apic = 1; + enable_x2apic(); + } + +end_restore: + if (ret) + /* + * IR enabling failed + */ + restore_IO_APIC_setup(); + else + reinit_intr_remapped_IO_APIC(x2apic_preenabled); + +end: + unmask_8259A(); + local_irq_restore(flags); + + if (!ret) { + if (!x2apic_preenabled) + pr_info("Enabled x2apic and interrupt-remapping\n"); + else + pr_info("Enabled Interrupt-remapping\n"); + } else + pr_err("Failed to enable Interrupt-remapping and x2apic\n"); +#else + if (!cpu_has_x2apic) + return; + + if (x2apic_preenabled) + panic("x2apic enabled prior OS handover," + " enable CONFIG_INTR_REMAP"); + + pr_info("Enable CONFIG_INTR_REMAP for enabling intr-remapping " + " and x2apic\n"); +#endif + + return; +} +#endif /* CONFIG_X86_X2APIC */ + +#ifdef CONFIG_X86_64 +/* + * Detect and enable local APICs on non-SMP boards. + * Original code written by Keir Fraser. + * 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) +{ + if (!cpu_has_apic) { + pr_info("No local APIC present\n"); + return -1; + } + + mp_lapic_addr = APIC_DEFAULT_PHYS_BASE; + boot_cpu_physical_apicid = 0; + return 0; +} +#else +/* + * Detect and initialize APIC + */ +static int __init detect_init_APIC(void) +{ + u32 h, l, features; + + /* Disabled by kernel option? */ + if (disable_apic) + return -1; + + switch (boot_cpu_data.x86_vendor) { + case X86_VENDOR_AMD: + if ((boot_cpu_data.x86 == 6 && boot_cpu_data.x86_model > 1) || + (boot_cpu_data.x86 >= 15)) + break; + goto no_apic; + case X86_VENDOR_INTEL: + if (boot_cpu_data.x86 == 6 || boot_cpu_data.x86 == 15 || + (boot_cpu_data.x86 == 5 && cpu_has_apic)) + break; + goto no_apic; + default: + goto no_apic; + } + + if (!cpu_has_apic) { + /* + * Over-ride BIOS and try to enable the local APIC only if + * "lapic" specified. + */ + if (!force_enable_local_apic) { + pr_info("Local APIC disabled by BIOS -- " + "you can enable it with \"lapic\"\n"); + return -1; + } + /* + * Some BIOSes disable the local APIC in the APIC_BASE + * MSR. This can only be done in software for Intel P6 or later + * and AMD K7 (Model > 1) or later. + */ + rdmsr(MSR_IA32_APICBASE, l, h); + if (!(l & MSR_IA32_APICBASE_ENABLE)) { + pr_info("Local APIC disabled by BIOS -- reenabling.\n"); + l &= ~MSR_IA32_APICBASE_BASE; + l |= MSR_IA32_APICBASE_ENABLE | APIC_DEFAULT_PHYS_BASE; + wrmsr(MSR_IA32_APICBASE, l, h); + enabled_via_apicbase = 1; + } + } + /* + * The APIC feature bit should now be enabled + * in `cpuid' + */ + features = cpuid_edx(1); + if (!(features & (1 << X86_FEATURE_APIC))) { + pr_warning("Could not enable APIC!\n"); + return -1; + } + 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 */ + rdmsr(MSR_IA32_APICBASE, l, h); + if (l & MSR_IA32_APICBASE_ENABLE) + mp_lapic_addr = l & MSR_IA32_APICBASE_BASE; + + pr_info("Found and enabled local APIC!\n"); + + apic_pm_activate(); + + return 0; + +no_apic: + pr_info("No local APIC present or hardware disabled\n"); + return -1; +} +#endif + +#ifdef CONFIG_X86_64 +void __init early_init_lapic_mapping(void) +{ + unsigned long phys_addr; + + /* + * If no local APIC can be found then go out + * : it means there is no mpatable and MADT + */ + if (!smp_found_config) + return; + + phys_addr = mp_lapic_addr; + + set_fixmap_nocache(FIX_APIC_BASE, phys_addr); + apic_printk(APIC_VERBOSE, "mapped APIC to %16lx (%16lx)\n", + APIC_BASE, phys_addr); + + /* + * Fetch the APIC ID of the BSP in case we have a + * default configuration (or the MP table is broken). + */ + boot_cpu_physical_apicid = read_apic_id(); +} +#endif + +/** + * init_apic_mappings - initialize APIC mappings + */ +void __init init_apic_mappings(void) +{ +#ifdef CONFIG_X86_X2APIC + if (x2apic) { + boot_cpu_physical_apicid = read_apic_id(); + return; + } +#endif + + /* + * If no local APIC can be found then set up a fake all + * zeroes page to simulate the local APIC and another + * one for the IO-APIC. + */ + if (!smp_found_config && detect_init_APIC()) { + apic_phys = (unsigned long) alloc_bootmem_pages(PAGE_SIZE); + apic_phys = __pa(apic_phys); + } else + apic_phys = mp_lapic_addr; + + set_fixmap_nocache(FIX_APIC_BASE, apic_phys); + apic_printk(APIC_VERBOSE, "mapped APIC to %08lx (%08lx)\n", + APIC_BASE, apic_phys); + + /* + * Fetch the APIC ID of the BSP in case we have a + * default configuration (or the MP table is broken). + */ + if (boot_cpu_physical_apicid == -1U) + boot_cpu_physical_apicid = read_apic_id(); +} + +/* + * This initializes the IO-APIC and APIC hardware if this is + * a UP kernel. + */ +int apic_version[MAX_APICS]; + +int __init APIC_init_uniprocessor(void) +{ + if (disable_apic) { + pr_info("Apic disabled\n"); + return -1; + } +#ifdef CONFIG_X86_64 + if (!cpu_has_apic) { + disable_apic = 1; + pr_info("Apic disabled by BIOS\n"); + return -1; + } +#else + if (!smp_found_config && !cpu_has_apic) + return -1; + + /* + * Complain if the BIOS pretends there is one. + */ + if (!cpu_has_apic && + APIC_INTEGRATED(apic_version[boot_cpu_physical_apicid])) { + pr_err("BIOS bug, local APIC 0x%x not detected!...\n", + boot_cpu_physical_apicid); + clear_cpu_cap(&boot_cpu_data, X86_FEATURE_APIC); + return -1; + } +#endif + + enable_IR_x2apic(); +#ifdef CONFIG_X86_64 + default_setup_apic_routing(); +#endif + + verify_local_APIC(); + connect_bsp_APIC(); + +#ifdef CONFIG_X86_64 + apic_write(APIC_ID, 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); + setup_local_APIC(); + +#ifdef CONFIG_X86_IO_APIC + /* + * Now enable IO-APICs, actually call clear_IO_APIC + * We need clear_IO_APIC before enabling error vector + */ + if (!skip_ioapic_setup && nr_ioapics) + enable_IO_APIC(); +#endif + + end_local_APIC_setup(); + +#ifdef CONFIG_X86_IO_APIC + if (smp_found_config && !skip_ioapic_setup && nr_ioapics) + setup_IO_APIC(); + else { + nr_ioapics = 0; + localise_nmi_watchdog(); + } +#else + localise_nmi_watchdog(); +#endif + + setup_boot_clock(); +#ifdef CONFIG_X86_64 + check_nmi_watchdog(); +#endif + + return 0; +} + +/* + * Local APIC interrupts + */ + +/* + * This interrupt should _never_ happen with our APIC/SMP architecture + */ +void smp_spurious_interrupt(struct pt_regs *regs) +{ + u32 v; + + exit_idle(); + irq_enter(); + /* + * 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. + */ + v = apic_read(APIC_ISR + ((SPURIOUS_APIC_VECTOR & ~0x1f) >> 1)); + if (v & (1 << (SPURIOUS_APIC_VECTOR & 0x1f))) + ack_APIC_irq(); + + inc_irq_stat(irq_spurious_count); + + /* see sw-dev-man vol 3, chapter 7.4.13.5 */ + pr_info("spurious APIC interrupt on CPU#%d, " + "should never happen.\n", smp_processor_id()); + irq_exit(); +} + +/* + * This interrupt should never happen with our APIC/SMP architecture + */ +void smp_error_interrupt(struct pt_regs *regs) +{ + u32 v, v1; + + exit_idle(); + irq_enter(); + /* First tickle the hardware, only then report what went on. -- REW */ + v = apic_read(APIC_ESR); + apic_write(APIC_ESR, 0); + v1 = apic_read(APIC_ESR); + ack_APIC_irq(); + atomic_inc(&irq_err_count); + + /* + * Here is what the APIC error bits mean: + * 0: Send CS error + * 1: Receive CS error + * 2: Send accept error + * 3: Receive accept error + * 4: Reserved + * 5: Send illegal vector + * 6: Received illegal vector + * 7: Illegal register address + */ + pr_debug("APIC error on CPU%d: %02x(%02x)\n", + smp_processor_id(), v , v1); + irq_exit(); +} + +/** + * connect_bsp_APIC - attach the APIC to the interrupt system + */ +void __init connect_bsp_APIC(void) +{ +#ifdef CONFIG_X86_32 + if (pic_mode) { + /* + * Do not trust the local APIC being empty at bootup. + */ + clear_local_APIC(); + /* + * 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"); + outb(0x70, 0x22); + outb(0x01, 0x23); + } +#endif + if (apic->enable_apic_mode) + apic->enable_apic_mode(); +} + +/** + * disconnect_bsp_APIC - detach the APIC from the interrupt system + * @virt_wire_setup: indicates, whether virtual wire mode is selected + * + * Virtual wire mode is necessary to deliver legacy interrupts even when the + * APIC is disabled. + */ +void disconnect_bsp_APIC(int virt_wire_setup) +{ + unsigned int value; + +#ifdef CONFIG_X86_32 + if (pic_mode) { + /* + * Put the board back into PIC mode (has an effect only on + * certain older boards). Note that APIC interrupts, including + * IPIs, won't work beyond this point! The only exception are + * INIT IPIs. + */ + apic_printk(APIC_VERBOSE, "disabling APIC mode, " + "entering PIC mode.\n"); + outb(0x70, 0x22); + outb(0x00, 0x23); + return; + } +#endif + + /* Go back to Virtual Wire compatibility mode */ + + /* For the spurious interrupt use vector F, and enable it */ + value = apic_read(APIC_SPIV); + value &= ~APIC_VECTOR_MASK; + value |= APIC_SPIV_APIC_ENABLED; + value |= 0xf; + apic_write(APIC_SPIV, value); + + if (!virt_wire_setup) { + /* + * For LVT0 make it edge triggered, active high, + * external and enabled + */ + value = apic_read(APIC_LVT0); + value &= ~(APIC_MODE_MASK | APIC_SEND_PENDING | + APIC_INPUT_POLARITY | APIC_LVT_REMOTE_IRR | + APIC_LVT_LEVEL_TRIGGER | APIC_LVT_MASKED); + value |= APIC_LVT_REMOTE_IRR | APIC_SEND_PENDING; + value = SET_APIC_DELIVERY_MODE(value, APIC_MODE_EXTINT); + apic_write(APIC_LVT0, value); + } else { + /* Disable LVT0 */ + apic_write(APIC_LVT0, APIC_LVT_MASKED); + } + + /* + * For LVT1 make it edge triggered, active high, + * nmi and enabled + */ + value = apic_read(APIC_LVT1); + value &= ~(APIC_MODE_MASK | APIC_SEND_PENDING | + APIC_INPUT_POLARITY | APIC_LVT_REMOTE_IRR | + APIC_LVT_LEVEL_TRIGGER | APIC_LVT_MASKED); + value |= APIC_LVT_REMOTE_IRR | APIC_SEND_PENDING; + value = SET_APIC_DELIVERY_MODE(value, APIC_MODE_NMI); + apic_write(APIC_LVT1, value); +} + +void __cpuinit generic_processor_info(int apicid, int version) +{ + int cpu; + + /* + * Validate version + */ + if (version == 0x0) { + pr_warning("BIOS bug, APIC version is 0 for CPU#%d! " + "fixing up to 0x10. (tell your hw vendor)\n", + version); + version = 0x10; + } + apic_version[apicid] = version; + + if (num_processors >= nr_cpu_ids) { + int max = nr_cpu_ids; + int thiscpu = max + disabled_cpus; + + pr_warning( + "ACPI: NR_CPUS/possible_cpus limit of %i reached." + " Processor %d/0x%x ignored.\n", max, thiscpu, apicid); + + disabled_cpus++; + return; + } + + num_processors++; + cpu = cpumask_next_zero(-1, cpu_present_mask); + + if (version != apic_version[boot_cpu_physical_apicid]) + WARN_ONCE(1, + "ACPI: apic version mismatch, bootcpu: %x cpu %d: %x\n", + apic_version[boot_cpu_physical_apicid], cpu, version); + + physid_set(apicid, phys_cpu_present_map); + 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. + */ + cpu = 0; + } + if (apicid > max_physical_apicid) + max_physical_apicid = apicid; + +#ifdef CONFIG_X86_32 + /* + * Would be preferable to switch to bigsmp when CONFIG_HOTPLUG_CPU=y + * but we need to work other dependencies like SMP_SUSPEND etc + * before this can be done without some confusion. + * if (CPU_HOTPLUG_ENABLED || num_processors > 8) + * - Ashok Raj <ashok.raj@intel.com> + */ + if (max_physical_apicid >= 8) { + switch (boot_cpu_data.x86_vendor) { + case X86_VENDOR_INTEL: + if (!APIC_XAPIC(version)) { + def_to_bigsmp = 0; + break; + } + /* If P4 and above fall through */ + case X86_VENDOR_AMD: + def_to_bigsmp = 1; + } + } +#endif + +#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 + + set_cpu_possible(cpu, true); + set_cpu_present(cpu, true); +} + +int hard_smp_processor_id(void) +{ + return read_apic_id(); +} + +void default_init_apic_ldr(void) +{ + unsigned long val; + + apic_write(APIC_DFR, APIC_DFR_VALUE); + val = apic_read(APIC_LDR) & ~APIC_LDR_MASK; + val |= SET_APIC_LOGICAL_ID(1UL << smp_processor_id()); + apic_write(APIC_LDR, val); +} + +#ifdef CONFIG_X86_32 +int default_apicid_to_node(int logical_apicid) +{ +#ifdef CONFIG_SMP + return apicid_2_node[hard_smp_processor_id()]; +#else + return 0; +#endif +} +#endif + +/* + * Power management + */ +#ifdef CONFIG_PM + +static struct { + /* + * 'active' is true if the local APIC was enabled by us and + * not the BIOS; this signifies that we are also responsible + * for disabling it before entering apm/acpi suspend + */ + int active; + /* r/w apic fields */ + unsigned int apic_id; + unsigned int apic_taskpri; + unsigned int apic_ldr; + unsigned int apic_dfr; + unsigned int apic_spiv; + unsigned int apic_lvtt; + unsigned int apic_lvtpc; + unsigned int apic_lvt0; + unsigned int apic_lvt1; + unsigned int apic_lvterr; + unsigned int apic_tmict; + unsigned int apic_tdcr; + unsigned int apic_thmr; +} apic_pm_state; + +static int lapic_suspend(struct sys_device *dev, pm_message_t state) +{ + unsigned long flags; + int maxlvt; + + if (!apic_pm_state.active) + return 0; + + maxlvt = lapic_get_maxlvt(); + + apic_pm_state.apic_id = apic_read(APIC_ID); + apic_pm_state.apic_taskpri = apic_read(APIC_TASKPRI); + apic_pm_state.apic_ldr = apic_read(APIC_LDR); + apic_pm_state.apic_dfr = apic_read(APIC_DFR); + apic_pm_state.apic_spiv = apic_read(APIC_SPIV); + apic_pm_state.apic_lvtt = apic_read(APIC_LVTT); + if (maxlvt >= 4) + apic_pm_state.apic_lvtpc = apic_read(APIC_LVTPC); + apic_pm_state.apic_lvt0 = apic_read(APIC_LVT0); + apic_pm_state.apic_lvt1 = apic_read(APIC_LVT1); + apic_pm_state.apic_lvterr = apic_read(APIC_LVTERR); + apic_pm_state.apic_tmict = apic_read(APIC_TMICT); + apic_pm_state.apic_tdcr = apic_read(APIC_TDCR); +#if defined(CONFIG_X86_MCE_P4THERMAL) || defined(CONFIG_X86_MCE_INTEL) + if (maxlvt >= 5) + apic_pm_state.apic_thmr = apic_read(APIC_LVTTHMR); +#endif + + local_irq_save(flags); + disable_local_APIC(); + local_irq_restore(flags); + return 0; +} + +static int lapic_resume(struct sys_device *dev) +{ + unsigned int l, h; + unsigned long flags; + int maxlvt; + + if (!apic_pm_state.active) + return 0; + + maxlvt = lapic_get_maxlvt(); + + local_irq_save(flags); + +#ifdef CONFIG_X86_X2APIC + if (x2apic) + enable_x2apic(); + else +#endif + { + /* + * Make sure the APICBASE points to the right address + * + * FIXME! This will be wrong if we ever support suspend on + * SMP! We'll need to do this as part of the CPU restore! + */ + rdmsr(MSR_IA32_APICBASE, l, h); + l &= ~MSR_IA32_APICBASE_BASE; + l |= MSR_IA32_APICBASE_ENABLE | mp_lapic_addr; + wrmsr(MSR_IA32_APICBASE, l, h); + } + + apic_write(APIC_LVTERR, ERROR_APIC_VECTOR | APIC_LVT_MASKED); + apic_write(APIC_ID, apic_pm_state.apic_id); + apic_write(APIC_DFR, apic_pm_state.apic_dfr); + apic_write(APIC_LDR, apic_pm_state.apic_ldr); + apic_write(APIC_TASKPRI, apic_pm_state.apic_taskpri); + apic_write(APIC_SPIV, apic_pm_state.apic_spiv); + apic_write(APIC_LVT0, apic_pm_state.apic_lvt0); + apic_write(APIC_LVT1, apic_pm_state.apic_lvt1); +#if defined(CONFIG_X86_MCE_P4THERMAL) || defined(CONFIG_X86_MCE_INTEL) + if (maxlvt >= 5) + apic_write(APIC_LVTTHMR, apic_pm_state.apic_thmr); +#endif + if (maxlvt >= 4) + apic_write(APIC_LVTPC, apic_pm_state.apic_lvtpc); + apic_write(APIC_LVTT, apic_pm_state.apic_lvtt); + apic_write(APIC_TDCR, apic_pm_state.apic_tdcr); + apic_write(APIC_TMICT, apic_pm_state.apic_tmict); + apic_write(APIC_ESR, 0); + apic_read(APIC_ESR); + apic_write(APIC_LVTERR, apic_pm_state.apic_lvterr); + apic_write(APIC_ESR, 0); + apic_read(APIC_ESR); + + local_irq_restore(flags); + + return 0; +} + +/* + * This device has no shutdown method - fully functioning local APICs + * are needed on every CPU up until machine_halt/restart/poweroff. + */ + +static struct sysdev_class lapic_sysclass = { + .name = "lapic", + .resume = lapic_resume, + .suspend = lapic_suspend, +}; + +static struct sys_device device_lapic = { + .id = 0, + .cls = &lapic_sysclass, +}; + +static void __cpuinit apic_pm_activate(void) +{ + apic_pm_state.active = 1; +} + +static int __init init_lapic_sysfs(void) +{ + int error; + + if (!cpu_has_apic) + return 0; + /* XXX: remove suspend/resume procs if !apic_pm_state.active? */ + + error = sysdev_class_register(&lapic_sysclass); + if (!error) + error = sysdev_register(&device_lapic); + return error; +} +device_initcall(init_lapic_sysfs); + +#else /* CONFIG_PM */ + +static void apic_pm_activate(void) { } + +#endif /* CONFIG_PM */ + +#ifdef CONFIG_X86_64 +/* + * apic_is_clustered_box() -- Check if we can expect good TSC + * + * Thus far, the major user of this is IBM's Summit2 series: + * + * Clustered boxes may have unsynced TSC problems if they are + * multi-chassis. Use available data to take a good guess. + * If in doubt, go HPET. + */ +__cpuinit int apic_is_clustered_box(void) +{ + int i, clusters, zeros; + unsigned id; + u16 *bios_cpu_apicid; + DECLARE_BITMAP(clustermap, NUM_APIC_CLUSTERS); + + /* + * there is not this kind of box with AMD CPU yet. + * Some AMD box with quadcore cpu and 8 sockets apicid + * will be [4, 0x23] or [8, 0x27] could be thought to + * vsmp box still need checking... + */ + if ((boot_cpu_data.x86_vendor == X86_VENDOR_AMD) && !is_vsmp_box()) + return 0; + + bios_cpu_apicid = early_per_cpu_ptr(x86_bios_cpu_apicid); + bitmap_zero(clustermap, NUM_APIC_CLUSTERS); + + for (i = 0; i < nr_cpu_ids; i++) { + /* are we being called early in kernel startup? */ + if (bios_cpu_apicid) { + id = bios_cpu_apicid[i]; + } else if (i < nr_cpu_ids) { + if (cpu_present(i)) + id = per_cpu(x86_bios_cpu_apicid, i); + else + continue; + } else + break; + + if (id != BAD_APICID) + __set_bit(APIC_CLUSTERID(id), clustermap); + } + + /* Problem: Partially populated chassis may not have CPUs in some of + * the APIC clusters they have been allocated. Only present CPUs have + * x86_bios_cpu_apicid entries, thus causing zeroes in the bitmap. + * Since clusters are allocated sequentially, count zeros only if + * they are bounded by ones. + */ + clusters = 0; + zeros = 0; + for (i = 0; i < NUM_APIC_CLUSTERS; i++) { + if (test_bit(i, clustermap)) { + clusters += 1 + zeros; + zeros = 0; + } else + ++zeros; + } + + /* ScaleMP vSMPowered boxes have one cluster per board and TSCs are + * not guaranteed to be synced between boards + */ + if (is_vsmp_box() && clusters > 1) + return 1; + + /* + * If clusters > 2, then should be multi-chassis. + * May have to revisit this when multi-core + hyperthreaded CPUs come + * out, but AFAIK this will work even for them. + */ + return (clusters > 2); +} +#endif + +/* + * APIC command line parameters + */ +static int __init setup_disableapic(char *arg) +{ + disable_apic = 1; + 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) +{ + local_apic_timer_c2_ok = 1; + return 0; +} +early_param("lapic_timer_c2_ok", parse_lapic_timer_c2_ok); + +static int __init parse_disable_apic_timer(char *arg) +{ + disable_apic_timer = 1; + return 0; +} +early_param("noapictimer", parse_disable_apic_timer); + +static int __init parse_nolapic_timer(char *arg) +{ + disable_apic_timer = 1; + return 0; +} +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; + return 0; +#endif + return -EINVAL; + } + + if (strcmp("debug", arg) == 0) + apic_verbosity = APIC_DEBUG; + else if (strcmp("verbose", arg) == 0) + apic_verbosity = APIC_VERBOSE; + else { + pr_warning("APIC Verbosity level %s not recognised" + " use apic=verbose or apic=debug\n", arg); + return -EINVAL; + } + + return 0; +} +early_param("apic", apic_set_verbosity); + +static int __init lapic_insert_resource(void) +{ + if (!apic_phys) + return -1; + + /* Put local APIC into the resource map. */ + lapic_resource.start = apic_phys; + lapic_resource.end = lapic_resource.start + PAGE_SIZE - 1; + insert_resource(&iomem_resource, &lapic_resource); + + return 0; +} + +/* + * need call insert after e820_reserve_resources() + * that is using request_resource + */ +late_initcall(lapic_insert_resource); diff --git a/arch/x86/kernel/apic/apic_64.c b/arch/x86/kernel/apic/apic_64.c new file mode 100644 index 000000000000..70935dd904db --- /dev/null +++ b/arch/x86/kernel/apic/apic_64.c @@ -0,0 +1,89 @@ +/* + * Copyright 2004 James Cleverdon, IBM. + * Subject to the GNU Public License, v.2 + * + * Generic APIC sub-arch probe layer. + * + * Hacked for x86-64 by James Cleverdon from i386 architecture code by + * Martin Bligh, Andi Kleen, James Bottomley, John Stultz, and + * James Cleverdon. + */ +#include <linux/threads.h> +#include <linux/cpumask.h> +#include <linux/string.h> +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/ctype.h> +#include <linux/init.h> +#include <linux/hardirq.h> +#include <linux/dmar.h> + +#include <asm/smp.h> +#include <asm/apic.h> +#include <asm/ipi.h> +#include <asm/setup.h> + +extern struct apic apic_flat; +extern struct apic apic_physflat; +extern struct apic apic_x2xpic_uv_x; +extern struct apic apic_x2apic_phys; +extern struct apic apic_x2apic_cluster; + +struct apic __read_mostly *apic = &apic_flat; +EXPORT_SYMBOL_GPL(apic); + +static struct apic *apic_probe[] __initdata = { +#ifdef CONFIG_X86_UV + &apic_x2apic_uv_x, +#endif +#ifdef CONFIG_X86_X2APIC + &apic_x2apic_phys, + &apic_x2apic_cluster, +#endif + &apic_physflat, + NULL, +}; + +/* + * Check the APIC IDs in bios_cpu_apicid and choose the APIC mode. + */ +void __init default_setup_apic_routing(void) +{ +#ifdef CONFIG_X86_X2APIC + if (apic == &apic_x2apic_phys || apic == &apic_x2apic_cluster) { + if (!intr_remapping_enabled) + apic = &apic_flat; + } +#endif + + if (apic == &apic_flat) { + if (max_physical_apicid >= 8) + apic = &apic_physflat; + printk(KERN_INFO "Setting APIC routing to %s\n", apic->name); + } + + if (x86_quirks->update_apic) + x86_quirks->update_apic(); +} + +/* Same for both flat and physical. */ + +void apic_send_IPI_self(int vector) +{ + __default_send_IPI_shortcut(APIC_DEST_SELF, vector, APIC_DEST_PHYSICAL); +} + +int __init default_acpi_madt_oem_check(char *oem_id, char *oem_table_id) +{ + int i; + + for (i = 0; apic_probe[i]; ++i) { + if (apic_probe[i]->acpi_madt_oem_check(oem_id, oem_table_id)) { + apic = apic_probe[i]; + printk(KERN_INFO "Setting APIC routing to %s.\n", + apic->name); + return 1; + } + } + return 0; +} diff --git a/arch/x86/kernel/apic/apic_flat_64.c b/arch/x86/kernel/apic/apic_flat_64.c new file mode 100644 index 000000000000..3b002995e145 --- /dev/null +++ b/arch/x86/kernel/apic/apic_flat_64.c @@ -0,0 +1,389 @@ +/* + * Copyright 2004 James Cleverdon, IBM. + * Subject to the GNU Public License, v.2 + * + * Flat APIC subarch code. + * + * Hacked for x86-64 by James Cleverdon from i386 architecture code by + * Martin Bligh, Andi Kleen, James Bottomley, John Stultz, and + * James Cleverdon. + */ +#include <linux/errno.h> +#include <linux/threads.h> +#include <linux/cpumask.h> +#include <linux/string.h> +#include <linux/kernel.h> +#include <linux/ctype.h> +#include <linux/init.h> +#include <linux/hardirq.h> +#include <asm/smp.h> +#include <asm/apic.h> +#include <asm/ipi.h> + +#ifdef CONFIG_ACPI +#include <acpi/acpi_bus.h> +#endif + +static int flat_acpi_madt_oem_check(char *oem_id, char *oem_table_id) +{ + return 1; +} + +static const struct cpumask *flat_target_cpus(void) +{ + return cpu_online_mask; +} + +static void flat_vector_allocation_domain(int cpu, struct cpumask *retmask) +{ + /* Careful. Some cpus do not strictly honor the set of cpus + * specified in the interrupt destination when using lowest + * priority interrupt delivery mode. + * + * In particular there was a hyperthreading cpu observed to + * deliver interrupts to the wrong hyperthread when only one + * hyperthread was specified in the interrupt desitination. + */ + cpumask_clear(retmask); + cpumask_bits(retmask)[0] = APIC_ALL_CPUS; +} + +/* + * Set up the logical destination ID. + * + * 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... + */ +static void flat_init_apic_ldr(void) +{ + unsigned long val; + unsigned long num, id; + + num = smp_processor_id(); + id = 1UL << num; + apic_write(APIC_DFR, APIC_DFR_FLAT); + val = apic_read(APIC_LDR) & ~APIC_LDR_MASK; + val |= SET_APIC_LOGICAL_ID(id); + apic_write(APIC_LDR, val); +} + +static inline void _flat_send_IPI_mask(unsigned long mask, int vector) +{ + unsigned long flags; + + local_irq_save(flags); + __default_send_IPI_dest_field(mask, vector, apic->dest_logical); + local_irq_restore(flags); +} + +static void flat_send_IPI_mask(const struct cpumask *cpumask, int vector) +{ + unsigned long mask = cpumask_bits(cpumask)[0]; + + _flat_send_IPI_mask(mask, vector); +} + +static void + flat_send_IPI_mask_allbutself(const struct cpumask *cpumask, int vector) +{ + unsigned long mask = cpumask_bits(cpumask)[0]; + int cpu = smp_processor_id(); + + if (cpu < BITS_PER_LONG) + clear_bit(cpu, &mask); + + _flat_send_IPI_mask(mask, vector); +} + +static void flat_send_IPI_allbutself(int vector) +{ + int cpu = smp_processor_id(); +#ifdef CONFIG_HOTPLUG_CPU + int hotplug = 1; +#else + int hotplug = 0; +#endif + if (hotplug || vector == NMI_VECTOR) { + if (!cpumask_equal(cpu_online_mask, cpumask_of(cpu))) { + unsigned long mask = cpumask_bits(cpu_online_mask)[0]; + + if (cpu < BITS_PER_LONG) + clear_bit(cpu, &mask); + + _flat_send_IPI_mask(mask, vector); + } + } else if (num_online_cpus() > 1) { + __default_send_IPI_shortcut(APIC_DEST_ALLBUT, + vector, apic->dest_logical); + } +} + +static void flat_send_IPI_all(int vector) +{ + if (vector == NMI_VECTOR) { + flat_send_IPI_mask(cpu_online_mask, vector); + } else { + __default_send_IPI_shortcut(APIC_DEST_ALLINC, + vector, apic->dest_logical); + } +} + +static unsigned int flat_get_apic_id(unsigned long x) +{ + unsigned int id; + + id = (((x)>>24) & 0xFFu); + + return id; +} + +static unsigned long set_apic_id(unsigned int id) +{ + unsigned long x; + + x = ((id & 0xFFu)<<24); + return x; +} + +static unsigned int read_xapic_id(void) +{ + unsigned int id; + + id = flat_get_apic_id(apic_read(APIC_ID)); + return id; +} + +static int flat_apic_id_registered(void) +{ + return physid_isset(read_xapic_id(), phys_cpu_present_map); +} + +static unsigned int flat_cpu_mask_to_apicid(const struct cpumask *cpumask) +{ + return cpumask_bits(cpumask)[0] & APIC_ALL_CPUS; +} + +static unsigned int flat_cpu_mask_to_apicid_and(const struct cpumask *cpumask, + const struct cpumask *andmask) +{ + unsigned long mask1 = cpumask_bits(cpumask)[0] & APIC_ALL_CPUS; + unsigned long mask2 = cpumask_bits(andmask)[0] & APIC_ALL_CPUS; + + return mask1 & mask2; +} + +static int flat_phys_pkg_id(int initial_apic_id, int index_msb) +{ + return hard_smp_processor_id() >> index_msb; +} + +struct apic apic_flat = { + .name = "flat", + .probe = NULL, + .acpi_madt_oem_check = flat_acpi_madt_oem_check, + .apic_id_registered = flat_apic_id_registered, + + .irq_delivery_mode = dest_LowestPrio, + .irq_dest_mode = 1, /* logical */ + + .target_cpus = flat_target_cpus, + .disable_esr = 0, + .dest_logical = APIC_DEST_LOGICAL, + .check_apicid_used = NULL, + .check_apicid_present = NULL, + + .vector_allocation_domain = flat_vector_allocation_domain, + .init_apic_ldr = flat_init_apic_ldr, + + .ioapic_phys_id_map = NULL, + .setup_apic_routing = NULL, + .multi_timer_check = NULL, + .apicid_to_node = NULL, + .cpu_to_logical_apicid = NULL, + .cpu_present_to_apicid = default_cpu_present_to_apicid, + .apicid_to_cpu_present = NULL, + .setup_portio_remap = NULL, + .check_phys_apicid_present = default_check_phys_apicid_present, + .enable_apic_mode = NULL, + .phys_pkg_id = flat_phys_pkg_id, + .mps_oem_check = NULL, + + .get_apic_id = flat_get_apic_id, + .set_apic_id = set_apic_id, + .apic_id_mask = 0xFFu << 24, + + .cpu_mask_to_apicid = flat_cpu_mask_to_apicid, + .cpu_mask_to_apicid_and = flat_cpu_mask_to_apicid_and, + + .send_IPI_mask = flat_send_IPI_mask, + .send_IPI_mask_allbutself = flat_send_IPI_mask_allbutself, + .send_IPI_allbutself = flat_send_IPI_allbutself, + .send_IPI_all = flat_send_IPI_all, + .send_IPI_self = apic_send_IPI_self, + + .wakeup_cpu = NULL, + .trampoline_phys_low = DEFAULT_TRAMPOLINE_PHYS_LOW, + .trampoline_phys_high = DEFAULT_TRAMPOLINE_PHYS_HIGH, + .wait_for_init_deassert = NULL, + .smp_callin_clear_local_apic = NULL, + .inquire_remote_apic = NULL, + + .read = native_apic_mem_read, + .write = native_apic_mem_write, + .icr_read = native_apic_icr_read, + .icr_write = native_apic_icr_write, + .wait_icr_idle = native_apic_wait_icr_idle, + .safe_wait_icr_idle = native_safe_apic_wait_icr_idle, +}; + +/* + * Physflat mode is used when there are more than 8 CPUs on a AMD system. + * We cannot use logical delivery in this case because the mask + * overflows, so use physical mode. + */ +static int physflat_acpi_madt_oem_check(char *oem_id, char *oem_table_id) +{ +#ifdef CONFIG_ACPI + /* + * Quirk: some x86_64 machines can only use physical APIC mode + * regardless of how many processors are present (x86_64 ES7000 + * is an example). + */ + if (acpi_gbl_FADT.header.revision > FADT2_REVISION_ID && + (acpi_gbl_FADT.flags & ACPI_FADT_APIC_PHYSICAL)) { + printk(KERN_DEBUG "system APIC only can use physical flat"); + return 1; + } +#endif + + return 0; +} + +static const struct cpumask *physflat_target_cpus(void) +{ + return cpu_online_mask; +} + +static void physflat_vector_allocation_domain(int cpu, struct cpumask *retmask) +{ + cpumask_clear(retmask); + cpumask_set_cpu(cpu, retmask); +} + +static void physflat_send_IPI_mask(const struct cpumask *cpumask, int vector) +{ + default_send_IPI_mask_sequence_phys(cpumask, vector); +} + +static void physflat_send_IPI_mask_allbutself(const struct cpumask *cpumask, + int vector) +{ + default_send_IPI_mask_allbutself_phys(cpumask, vector); +} + +static void physflat_send_IPI_allbutself(int vector) +{ + default_send_IPI_mask_allbutself_phys(cpu_online_mask, vector); +} + +static void physflat_send_IPI_all(int vector) +{ + physflat_send_IPI_mask(cpu_online_mask, vector); +} + +static unsigned int physflat_cpu_mask_to_apicid(const struct cpumask *cpumask) +{ + int cpu; + + /* + * We're using fixed IRQ delivery, can only return one phys APIC ID. + * May as well be the first. + */ + cpu = cpumask_first(cpumask); + if ((unsigned)cpu < nr_cpu_ids) + return per_cpu(x86_cpu_to_apicid, cpu); + else + return BAD_APICID; +} + +static unsigned int +physflat_cpu_mask_to_apicid_and(const struct cpumask *cpumask, + const struct cpumask *andmask) +{ + int cpu; + + /* + * We're using fixed IRQ delivery, can only return one phys APIC ID. + * May as well be the first. + */ + for_each_cpu_and(cpu, cpumask, andmask) { + if (cpumask_test_cpu(cpu, cpu_online_mask)) + break; + } + if (cpu < nr_cpu_ids) + return per_cpu(x86_cpu_to_apicid, cpu); + + return BAD_APICID; +} + +struct apic apic_physflat = { + + .name = "physical flat", + .probe = NULL, + .acpi_madt_oem_check = physflat_acpi_madt_oem_check, + .apic_id_registered = flat_apic_id_registered, + + .irq_delivery_mode = dest_Fixed, + .irq_dest_mode = 0, /* physical */ + + .target_cpus = physflat_target_cpus, + .disable_esr = 0, + .dest_logical = 0, + .check_apicid_used = NULL, + .check_apicid_present = NULL, + + .vector_allocation_domain = physflat_vector_allocation_domain, + /* not needed, but shouldn't hurt: */ + .init_apic_ldr = flat_init_apic_ldr, + + .ioapic_phys_id_map = NULL, + .setup_apic_routing = NULL, + .multi_timer_check = NULL, + .apicid_to_node = NULL, + .cpu_to_logical_apicid = NULL, + .cpu_present_to_apicid = default_cpu_present_to_apicid, + .apicid_to_cpu_present = NULL, + .setup_portio_remap = NULL, + .check_phys_apicid_present = default_check_phys_apicid_present, + .enable_apic_mode = NULL, + .phys_pkg_id = flat_phys_pkg_id, + .mps_oem_check = NULL, + + .get_apic_id = flat_get_apic_id, + .set_apic_id = set_apic_id, + .apic_id_mask = 0xFFu << 24, + + .cpu_mask_to_apicid = physflat_cpu_mask_to_apicid, + .cpu_mask_to_apicid_and = physflat_cpu_mask_to_apicid_and, + + .send_IPI_mask = physflat_send_IPI_mask, + .send_IPI_mask_allbutself = physflat_send_IPI_mask_allbutself, + .send_IPI_allbutself = physflat_send_IPI_allbutself, + .send_IPI_all = physflat_send_IPI_all, + .send_IPI_self = apic_send_IPI_self, + + .wakeup_cpu = NULL, + .trampoline_phys_low = DEFAULT_TRAMPOLINE_PHYS_LOW, + .trampoline_phys_high = DEFAULT_TRAMPOLINE_PHYS_HIGH, + .wait_for_init_deassert = NULL, + .smp_callin_clear_local_apic = NULL, + .inquire_remote_apic = NULL, + + .read = native_apic_mem_read, + .write = native_apic_mem_write, + .icr_read = native_apic_icr_read, + .icr_write = native_apic_icr_write, + .wait_icr_idle = native_apic_wait_icr_idle, + .safe_wait_icr_idle = native_safe_apic_wait_icr_idle, +}; diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c new file mode 100644 index 000000000000..00e6071cefc4 --- /dev/null +++ b/arch/x86/kernel/apic/io_apic.c @@ -0,0 +1,4160 @@ +/* + * Intel IO-APIC support for multi-Pentium hosts. + * + * Copyright (C) 1997, 1998, 1999, 2000, 2009 Ingo Molnar, Hajnalka Szabo + * + * Many thanks to Stig Venaas for trying out countless experimental + * patches and reporting/debugging problems patiently! + * + * (c) 1999, Multiple IO-APIC support, developed by + * Ken-ichi Yaku <yaku@css1.kbnes.nec.co.jp> and + * Hidemi Kishimoto <kisimoto@css1.kbnes.nec.co.jp>, + * further tested and cleaned up by Zach Brown <zab@redhat.com> + * and Ingo Molnar <mingo@redhat.com> + * + * Fixes + * Maciej W. Rozycki : Bits for genuine 82489DX APICs; + * thanks to Eric Gilmore + * and Rolf G. Tews + * for testing these extensively + * Paul Diefenbaugh : Added full ACPI support + */ + +#include <linux/mm.h> +#include <linux/interrupt.h> +#include <linux/init.h> +#include <linux/delay.h> +#include <linux/sched.h> +#include <linux/pci.h> +#include <linux/mc146818rtc.h> +#include <linux/compiler.h> +#include <linux/acpi.h> +#include <linux/module.h> +#include <linux/sysdev.h> +#include <linux/msi.h> +#include <linux/htirq.h> +#include <linux/freezer.h> +#include <linux/kthread.h> +#include <linux/jiffies.h> /* time_after() */ +#ifdef CONFIG_ACPI +#include <acpi/acpi_bus.h> +#endif +#include <linux/bootmem.h> +#include <linux/dmar.h> +#include <linux/hpet.h> + +#include <asm/idle.h> +#include <asm/io.h> +#include <asm/smp.h> +#include <asm/cpu.h> +#include <asm/desc.h> +#include <asm/proto.h> +#include <asm/acpi.h> +#include <asm/dma.h> +#include <asm/timer.h> +#include <asm/i8259.h> +#include <asm/nmi.h> +#include <asm/msidef.h> +#include <asm/hypertransport.h> +#include <asm/setup.h> +#include <asm/irq_remapping.h> +#include <asm/hpet.h> +#include <asm/uv/uv_hub.h> +#include <asm/uv/uv_irq.h> + +#include <asm/apic.h> + +#define __apicdebuginit(type) static type __init + +/* + * Is the SiS APIC rmw bug present ? + * -1 = don't know, 0 = no, 1 = yes + */ +int sis_apic_bug = -1; + +static DEFINE_SPINLOCK(ioapic_lock); +static DEFINE_SPINLOCK(vector_lock); + +/* + * # of IRQ routing registers + */ +int nr_ioapic_registers[MAX_IO_APICS]; + +/* I/O APIC entries */ +struct mpc_ioapic mp_ioapics[MAX_IO_APICS]; +int nr_ioapics; + +/* MP IRQ source entries */ +struct mpc_intsrc mp_irqs[MAX_IRQ_SOURCES]; + +/* # of MP IRQ source entries */ +int mp_irq_entries; + +#if defined (CONFIG_MCA) || defined (CONFIG_EISA) +int mp_bus_id_to_type[MAX_MP_BUSSES]; +#endif + +DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES); + +int skip_ioapic_setup; + +void arch_disable_smp_support(void) +{ +#ifdef CONFIG_PCI + noioapicquirk = 1; + noioapicreroute = -1; +#endif + skip_ioapic_setup = 1; +} + +static int __init parse_noapic(char *str) +{ + /* disable IO-APIC */ + arch_disable_smp_support(); + return 0; +} +early_param("noapic", parse_noapic); + +struct irq_pin_list; + +/* + * This is performance-critical, we want to do it O(1) + * + * the indexing order of this array favors 1:1 mappings + * between pins and IRQs. + */ + +struct irq_pin_list { + int apic, pin; + struct irq_pin_list *next; +}; + +static struct irq_pin_list *get_one_free_irq_2_pin(int cpu) +{ + struct irq_pin_list *pin; + int node; + + node = cpu_to_node(cpu); + + pin = kzalloc_node(sizeof(*pin), GFP_ATOMIC, node); + + return pin; +} + +struct irq_cfg { + struct irq_pin_list *irq_2_pin; + cpumask_var_t domain; + cpumask_var_t old_domain; + unsigned move_cleanup_count; + u8 vector; + u8 move_in_progress : 1; +#ifdef CONFIG_NUMA_MIGRATE_IRQ_DESC + u8 move_desc_pending : 1; +#endif +}; + +/* irq_cfg is indexed by the sum of all RTEs in all I/O APICs. */ +#ifdef CONFIG_SPARSE_IRQ +static struct irq_cfg irq_cfgx[] = { +#else +static struct irq_cfg irq_cfgx[NR_IRQS] = { +#endif + [0] = { .vector = IRQ0_VECTOR, }, + [1] = { .vector = IRQ1_VECTOR, }, + [2] = { .vector = IRQ2_VECTOR, }, + [3] = { .vector = IRQ3_VECTOR, }, + [4] = { .vector = IRQ4_VECTOR, }, + [5] = { .vector = IRQ5_VECTOR, }, + [6] = { .vector = IRQ6_VECTOR, }, + [7] = { .vector = IRQ7_VECTOR, }, + [8] = { .vector = IRQ8_VECTOR, }, + [9] = { .vector = IRQ9_VECTOR, }, + [10] = { .vector = IRQ10_VECTOR, }, + [11] = { .vector = IRQ11_VECTOR, }, + [12] = { .vector = IRQ12_VECTOR, }, + [13] = { .vector = IRQ13_VECTOR, }, + [14] = { .vector = IRQ14_VECTOR, }, + [15] = { .vector = IRQ15_VECTOR, }, +}; + +int __init arch_early_irq_init(void) +{ + struct irq_cfg *cfg; + struct irq_desc *desc; + int count; + int i; + + cfg = irq_cfgx; + count = ARRAY_SIZE(irq_cfgx); + + for (i = 0; i < count; i++) { + desc = irq_to_desc(i); + desc->chip_data = &cfg[i]; + alloc_bootmem_cpumask_var(&cfg[i].domain); + alloc_bootmem_cpumask_var(&cfg[i].old_domain); + if (i < NR_IRQS_LEGACY) + cpumask_setall(cfg[i].domain); + } + + return 0; +} + +#ifdef CONFIG_SPARSE_IRQ +static struct irq_cfg *irq_cfg(unsigned int irq) +{ + struct irq_cfg *cfg = NULL; + struct irq_desc *desc; + + desc = irq_to_desc(irq); + if (desc) + cfg = desc->chip_data; + + return cfg; +} + +static struct irq_cfg *get_one_free_irq_cfg(int cpu) +{ + struct irq_cfg *cfg; + int node; + + node = cpu_to_node(cpu); + + cfg = kzalloc_node(sizeof(*cfg), GFP_ATOMIC, node); + if (cfg) { + if (!alloc_cpumask_var_node(&cfg->domain, GFP_ATOMIC, node)) { + kfree(cfg); + cfg = NULL; + } else if (!alloc_cpumask_var_node(&cfg->old_domain, + GFP_ATOMIC, node)) { + free_cpumask_var(cfg->domain); + kfree(cfg); + cfg = NULL; + } else { + cpumask_clear(cfg->domain); + cpumask_clear(cfg->old_domain); + } + } + + return cfg; +} + +int arch_init_chip_data(struct irq_desc *desc, int cpu) +{ + struct irq_cfg *cfg; + + cfg = desc->chip_data; + if (!cfg) { + desc->chip_data = get_one_free_irq_cfg(cpu); + if (!desc->chip_data) { + printk(KERN_ERR "can not alloc irq_cfg\n"); + BUG_ON(1); + } + } + + return 0; +} + +#ifdef CONFIG_NUMA_MIGRATE_IRQ_DESC + +static void +init_copy_irq_2_pin(struct irq_cfg *old_cfg, struct irq_cfg *cfg, int cpu) +{ + struct irq_pin_list *old_entry, *head, *tail, *entry; + + cfg->irq_2_pin = NULL; + old_entry = old_cfg->irq_2_pin; + if (!old_entry) + return; + + entry = get_one_free_irq_2_pin(cpu); + if (!entry) + return; + + entry->apic = old_entry->apic; + entry->pin = old_entry->pin; + head = entry; + tail = entry; + old_entry = old_entry->next; + while (old_entry) { + entry = get_one_free_irq_2_pin(cpu); + if (!entry) { + entry = head; + while (entry) { + head = entry->next; + kfree(entry); + entry = head; + } + /* still use the old one */ + return; + } + entry->apic = old_entry->apic; + entry->pin = old_entry->pin; + tail->next = entry; + tail = entry; + old_entry = old_entry->next; + } + + tail->next = NULL; + cfg->irq_2_pin = head; +} + +static void free_irq_2_pin(struct irq_cfg *old_cfg, struct irq_cfg *cfg) +{ + struct irq_pin_list *entry, *next; + + if (old_cfg->irq_2_pin == cfg->irq_2_pin) + return; + + entry = old_cfg->irq_2_pin; + + while (entry) { + next = entry->next; + kfree(entry); + entry = next; + } + old_cfg->irq_2_pin = NULL; +} + +void arch_init_copy_chip_data(struct irq_desc *old_desc, + struct irq_desc *desc, int cpu) +{ + struct irq_cfg *cfg; + struct irq_cfg *old_cfg; + + cfg = get_one_free_irq_cfg(cpu); + + if (!cfg) + return; + + desc->chip_data = cfg; + + old_cfg = old_desc->chip_data; + + memcpy(cfg, old_cfg, sizeof(struct irq_cfg)); + + init_copy_irq_2_pin(old_cfg, cfg, cpu); +} + +static void free_irq_cfg(struct irq_cfg *old_cfg) +{ + kfree(old_cfg); +} + +void arch_free_chip_data(struct irq_desc *old_desc, struct irq_desc *desc) +{ + struct irq_cfg *old_cfg, *cfg; + + old_cfg = old_desc->chip_data; + cfg = desc->chip_data; + + if (old_cfg == cfg) + return; + + if (old_cfg) { + free_irq_2_pin(old_cfg, cfg); + free_irq_cfg(old_cfg); + old_desc->chip_data = NULL; + } +} + +static void +set_extra_move_desc(struct irq_desc *desc, const struct cpumask *mask) +{ + struct irq_cfg *cfg = desc->chip_data; + + if (!cfg->move_in_progress) { + /* it means that domain is not changed */ + if (!cpumask_intersects(desc->affinity, mask)) + cfg->move_desc_pending = 1; + } +} +#endif + +#else +static struct irq_cfg *irq_cfg(unsigned int irq) +{ + return irq < nr_irqs ? irq_cfgx + irq : NULL; +} + +#endif + +#ifndef CONFIG_NUMA_MIGRATE_IRQ_DESC +static inline void +set_extra_move_desc(struct irq_desc *desc, const struct cpumask *mask) +{ +} +#endif + +struct io_apic { + unsigned int index; + unsigned int unused[3]; + unsigned int data; +}; + +static __attribute_const__ struct io_apic __iomem *io_apic_base(int idx) +{ + return (void __iomem *) __fix_to_virt(FIX_IO_APIC_BASE_0 + idx) + + (mp_ioapics[idx].apicaddr & ~PAGE_MASK); +} + +static inline unsigned int io_apic_read(unsigned int apic, unsigned int reg) +{ + struct io_apic __iomem *io_apic = io_apic_base(apic); + writel(reg, &io_apic->index); + return readl(&io_apic->data); +} + +static inline void io_apic_write(unsigned int apic, unsigned int reg, unsigned int value) +{ + struct io_apic __iomem *io_apic = io_apic_base(apic); + writel(reg, &io_apic->index); + writel(value, &io_apic->data); +} + +/* + * Re-write a value: to be used for read-modify-write + * cycles where the read already set up the index register. + * + * Older SiS APIC requires we rewrite the index register + */ +static inline void io_apic_modify(unsigned int apic, unsigned int reg, unsigned int value) +{ + struct io_apic __iomem *io_apic = io_apic_base(apic); + + if (sis_apic_bug) + writel(reg, &io_apic->index); + writel(value, &io_apic->data); +} + +static bool io_apic_level_ack_pending(struct irq_cfg *cfg) +{ + struct irq_pin_list *entry; + unsigned long flags; + + spin_lock_irqsave(&ioapic_lock, flags); + entry = cfg->irq_2_pin; + for (;;) { + unsigned int reg; + int pin; + + if (!entry) + break; + pin = entry->pin; + reg = io_apic_read(entry->apic, 0x10 + pin*2); + /* Is the remote IRR bit set? */ + if (reg & IO_APIC_REDIR_REMOTE_IRR) { + spin_unlock_irqrestore(&ioapic_lock, flags); + return true; + } + if (!entry->next) + break; + entry = entry->next; + } + spin_unlock_irqrestore(&ioapic_lock, flags); + + return false; +} + +union entry_union { + struct { u32 w1, w2; }; + struct IO_APIC_route_entry entry; +}; + +static struct IO_APIC_route_entry ioapic_read_entry(int apic, int pin) +{ + union entry_union eu; + unsigned long flags; + spin_lock_irqsave(&ioapic_lock, flags); + eu.w1 = io_apic_read(apic, 0x10 + 2 * pin); + eu.w2 = io_apic_read(apic, 0x11 + 2 * pin); + spin_unlock_irqrestore(&ioapic_lock, flags); + return eu.entry; +} + +/* + * When we write a new IO APIC routing entry, we need to write the high + * word first! If the mask bit in the low word is clear, we will enable + * the interrupt, and we need to make sure the entry is fully populated + * before that happens. + */ +static void +__ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e) +{ + union entry_union eu; + eu.entry = e; + io_apic_write(apic, 0x11 + 2*pin, eu.w2); + io_apic_write(apic, 0x10 + 2*pin, eu.w1); +} + +void ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e) +{ + unsigned long flags; + spin_lock_irqsave(&ioapic_lock, flags); + __ioapic_write_entry(apic, pin, e); + spin_unlock_irqrestore(&ioapic_lock, flags); +} + +/* + * When we mask an IO APIC routing entry, we need to write the low + * word first, in order to set the mask bit before we change the + * high bits! + */ +static void ioapic_mask_entry(int apic, int pin) +{ + unsigned long flags; + union entry_union eu = { .entry.mask = 1 }; + + spin_lock_irqsave(&ioapic_lock, flags); + io_apic_write(apic, 0x10 + 2*pin, eu.w1); + io_apic_write(apic, 0x11 + 2*pin, eu.w2); + spin_unlock_irqrestore(&ioapic_lock, flags); +} + +#ifdef CONFIG_SMP +static void send_cleanup_vector(struct irq_cfg *cfg) +{ + cpumask_var_t cleanup_mask; + + if (unlikely(!alloc_cpumask_var(&cleanup_mask, GFP_ATOMIC))) { + unsigned int i; + cfg->move_cleanup_count = 0; + for_each_cpu_and(i, cfg->old_domain, cpu_online_mask) + cfg->move_cleanup_count++; + for_each_cpu_and(i, cfg->old_domain, cpu_online_mask) + apic->send_IPI_mask(cpumask_of(i), IRQ_MOVE_CLEANUP_VECTOR); + } else { + cpumask_and(cleanup_mask, cfg->old_domain, cpu_online_mask); + cfg->move_cleanup_count = cpumask_weight(cleanup_mask); + apic->send_IPI_mask(cleanup_mask, IRQ_MOVE_CLEANUP_VECTOR); + free_cpumask_var(cleanup_mask); + } + cfg->move_in_progress = 0; +} + +static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, struct irq_cfg *cfg) +{ + int apic, pin; + struct irq_pin_list *entry; + u8 vector = cfg->vector; + + entry = cfg->irq_2_pin; + for (;;) { + unsigned int reg; + + if (!entry) + break; + + apic = entry->apic; + pin = entry->pin; +#ifdef CONFIG_INTR_REMAP + /* + * With interrupt-remapping, destination information comes + * from interrupt-remapping table entry. + */ + if (!irq_remapped(irq)) + io_apic_write(apic, 0x11 + pin*2, dest); +#else + io_apic_write(apic, 0x11 + pin*2, dest); +#endif + reg = io_apic_read(apic, 0x10 + pin*2); + reg &= ~IO_APIC_REDIR_VECTOR_MASK; + reg |= vector; + io_apic_modify(apic, 0x10 + pin*2, reg); + if (!entry->next) + break; + entry = entry->next; + } +} + +static int +assign_irq_vector(int irq, struct irq_cfg *cfg, const struct cpumask *mask); + +/* + * Either sets desc->affinity to a valid value, and returns + * ->cpu_mask_to_apicid of that, or returns BAD_APICID and + * leaves desc->affinity untouched. + */ +static unsigned int +set_desc_affinity(struct irq_desc *desc, const struct cpumask *mask) +{ + struct irq_cfg *cfg; + unsigned int irq; + + if (!cpumask_intersects(mask, cpu_online_mask)) + return BAD_APICID; + + irq = desc->irq; + cfg = desc->chip_data; + if (assign_irq_vector(irq, cfg, mask)) + return BAD_APICID; + + cpumask_and(desc->affinity, cfg->domain, mask); + set_extra_move_desc(desc, mask); + + return apic->cpu_mask_to_apicid_and(desc->affinity, cpu_online_mask); +} + +static void +set_ioapic_affinity_irq_desc(struct irq_desc *desc, const struct cpumask *mask) +{ + struct irq_cfg *cfg; + unsigned long flags; + unsigned int dest; + unsigned int irq; + + irq = desc->irq; + cfg = desc->chip_data; + + spin_lock_irqsave(&ioapic_lock, flags); + dest = set_desc_affinity(desc, mask); + if (dest != BAD_APICID) { + /* Only the high 8 bits are valid. */ + dest = SET_APIC_LOGICAL_ID(dest); + __target_IO_APIC_irq(irq, dest, cfg); + } + spin_unlock_irqrestore(&ioapic_lock, flags); +} + +static void +set_ioapic_affinity_irq(unsigned int irq, const struct cpumask *mask) +{ + struct irq_desc *desc; + + desc = irq_to_desc(irq); + + set_ioapic_affinity_irq_desc(desc, mask); +} +#endif /* CONFIG_SMP */ + +/* + * The common case is 1:1 IRQ<->pin mappings. Sometimes there are + * shared ISA-space IRQs, so we have to support them. We are super + * fast in the common case, and fast for shared ISA-space IRQs. + */ +static void add_pin_to_irq_cpu(struct irq_cfg *cfg, int cpu, int apic, int pin) +{ + struct irq_pin_list *entry; + + entry = cfg->irq_2_pin; + if (!entry) { + entry = get_one_free_irq_2_pin(cpu); + if (!entry) { + printk(KERN_ERR "can not alloc irq_2_pin to add %d - %d\n", + apic, pin); + return; + } + cfg->irq_2_pin = entry; + entry->apic = apic; + entry->pin = pin; + return; + } + + while (entry->next) { + /* not again, please */ + if (entry->apic == apic && entry->pin == pin) + return; + + entry = entry->next; + } + + entry->next = get_one_free_irq_2_pin(cpu); + entry = entry->next; + entry->apic = apic; + entry->pin = pin; +} + +/* + * Reroute an IRQ to a different pin. + */ +static void __init replace_pin_at_irq_cpu(struct irq_cfg *cfg, int cpu, + int oldapic, int oldpin, + int newapic, int newpin) +{ + struct irq_pin_list *entry = cfg->irq_2_pin; + int replaced = 0; + + while (entry) { + if (entry->apic == oldapic && entry->pin == oldpin) { + entry->apic = newapic; + entry->pin = newpin; + replaced = 1; + /* every one is different, right? */ + break; + } + entry = entry->next; + } + + /* why? call replace before add? */ + if (!replaced) + add_pin_to_irq_cpu(cfg, cpu, newapic, newpin); +} + +static inline void io_apic_modify_irq(struct irq_cfg *cfg, + int mask_and, int mask_or, + void (*final)(struct irq_pin_list *entry)) +{ + int pin; + struct irq_pin_list *entry; + + for (entry = cfg->irq_2_pin; entry != NULL; entry = entry->next) { + unsigned int reg; + pin = entry->pin; + reg = io_apic_read(entry->apic, 0x10 + pin * 2); + reg &= mask_and; + reg |= mask_or; + io_apic_modify(entry->apic, 0x10 + pin * 2, reg); + if (final) + final(entry); + } +} + +static void __unmask_IO_APIC_irq(struct irq_cfg *cfg) +{ + io_apic_modify_irq(cfg, ~IO_APIC_REDIR_MASKED, 0, NULL); +} + +#ifdef CONFIG_X86_64 +static void io_apic_sync(struct irq_pin_list *entry) +{ + /* + * Synchronize the IO-APIC and the CPU by doing + * a dummy read from the IO-APIC + */ + struct io_apic __iomem *io_apic; + io_apic = io_apic_base(entry->apic); + readl(&io_apic->data); +} + +static void __mask_IO_APIC_irq(struct irq_cfg *cfg) +{ + io_apic_modify_irq(cfg, ~0, IO_APIC_REDIR_MASKED, &io_apic_sync); +} +#else /* CONFIG_X86_32 */ +static void __mask_IO_APIC_irq(struct irq_cfg *cfg) +{ + io_apic_modify_irq(cfg, ~0, IO_APIC_REDIR_MASKED, NULL); +} + +static void __mask_and_edge_IO_APIC_irq(struct irq_cfg *cfg) +{ + io_apic_modify_irq(cfg, ~IO_APIC_REDIR_LEVEL_TRIGGER, + IO_APIC_REDIR_MASKED, NULL); +} + +static void __unmask_and_level_IO_APIC_irq(struct irq_cfg *cfg) +{ + io_apic_modify_irq(cfg, ~IO_APIC_REDIR_MASKED, + IO_APIC_REDIR_LEVEL_TRIGGER, NULL); +} +#endif /* CONFIG_X86_32 */ + +static void mask_IO_APIC_irq_desc(struct irq_desc *desc) +{ + struct irq_cfg *cfg = desc->chip_data; + unsigned long flags; + + BUG_ON(!cfg); + + spin_lock_irqsave(&ioapic_lock, flags); + __mask_IO_APIC_irq(cfg); + spin_unlock_irqrestore(&ioapic_lock, flags); +} + +static void unmask_IO_APIC_irq_desc(struct irq_desc *desc) +{ + struct irq_cfg *cfg = desc->chip_data; + unsigned long flags; + + spin_lock_irqsave(&ioapic_lock, flags); + __unmask_IO_APIC_irq(cfg); + spin_unlock_irqrestore(&ioapic_lock, flags); +} + +static void mask_IO_APIC_irq(unsigned int irq) +{ + struct irq_desc *desc = irq_to_desc(irq); + + mask_IO_APIC_irq_desc(desc); +} +static void unmask_IO_APIC_irq(unsigned int irq) +{ + struct irq_desc *desc = irq_to_desc(irq); + + unmask_IO_APIC_irq_desc(desc); +} + +static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin) +{ + struct IO_APIC_route_entry entry; + + /* Check delivery_mode to be sure we're not clearing an SMI pin */ + entry = ioapic_read_entry(apic, pin); + if (entry.delivery_mode == dest_SMI) + return; + /* + * Disable it in the IO-APIC irq-routing table: + */ + ioapic_mask_entry(apic, pin); +} + +static void clear_IO_APIC (void) +{ + int apic, pin; + + for (apic = 0; apic < nr_ioapics; apic++) + for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) + clear_IO_APIC_pin(apic, pin); +} + +#ifdef CONFIG_X86_32 +/* + * support for broken MP BIOSs, enables hand-redirection of PIRQ0-7 to + * specific CPU-side IRQs. + */ + +#define MAX_PIRQS 8 +static int pirq_entries[MAX_PIRQS] = { + [0 ... MAX_PIRQS - 1] = -1 +}; + +static int __init ioapic_pirq_setup(char *str) +{ + int i, max; + int ints[MAX_PIRQS+1]; + + get_options(str, ARRAY_SIZE(ints), ints); + + apic_printk(APIC_VERBOSE, KERN_INFO + "PIRQ redirection, working around broken MP-BIOS.\n"); + max = MAX_PIRQS; + if (ints[0] < MAX_PIRQS) + max = ints[0]; + + for (i = 0; i < max; i++) { + apic_printk(APIC_VERBOSE, KERN_DEBUG + "... PIRQ%d -> IRQ %d\n", i, ints[i+1]); + /* + * PIRQs are mapped upside down, usually. + */ + pirq_entries[MAX_PIRQS-i-1] = ints[i+1]; + } + return 1; +} + +__setup("pirq=", ioapic_pirq_setup); +#endif /* CONFIG_X86_32 */ + +#ifdef CONFIG_INTR_REMAP +/* I/O APIC RTE contents at the OS boot up */ +static struct IO_APIC_route_entry *early_ioapic_entries[MAX_IO_APICS]; + +/* + * Saves and masks all the unmasked IO-APIC RTE's + */ +int save_mask_IO_APIC_setup(void) +{ + union IO_APIC_reg_01 reg_01; + unsigned long flags; + int apic, pin; + + /* + * The number of IO-APIC IRQ registers (== #pins): + */ + for (apic = 0; apic < nr_ioapics; apic++) { + spin_lock_irqsave(&ioapic_lock, flags); + reg_01.raw = io_apic_read(apic, 1); + spin_unlock_irqrestore(&ioapic_lock, flags); + nr_ioapic_registers[apic] = reg_01.bits.entries+1; + } + + for (apic = 0; apic < nr_ioapics; apic++) { + early_ioapic_entries[apic] = + kzalloc(sizeof(struct IO_APIC_route_entry) * + nr_ioapic_registers[apic], GFP_KERNEL); + if (!early_ioapic_entries[apic]) + goto nomem; + } + + for (apic = 0; apic < nr_ioapics; apic++) + for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) { + struct IO_APIC_route_entry entry; + + entry = early_ioapic_entries[apic][pin] = + ioapic_read_entry(apic, pin); + if (!entry.mask) { + entry.mask = 1; + ioapic_write_entry(apic, pin, entry); + } + } + + return 0; + +nomem: + while (apic >= 0) + kfree(early_ioapic_entries[apic--]); + memset(early_ioapic_entries, 0, + ARRAY_SIZE(early_ioapic_entries)); + + return -ENOMEM; +} + +void restore_IO_APIC_setup(void) +{ + int apic, pin; + + for (apic = 0; apic < nr_ioapics; apic++) { + if (!early_ioapic_entries[apic]) + break; + for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) + ioapic_write_entry(apic, pin, + early_ioapic_entries[apic][pin]); + kfree(early_ioapic_entries[apic]); + early_ioapic_entries[apic] = NULL; + } +} + +void reinit_intr_remapped_IO_APIC(int intr_remapping) +{ + /* + * for now plain restore of previous settings. + * TBD: In the case of OS enabling interrupt-remapping, + * IO-APIC RTE's need to be setup to point to interrupt-remapping + * table entries. for now, do a plain restore, and wait for + * the setup_IO_APIC_irqs() to do proper initialization. + */ + restore_IO_APIC_setup(); +} +#endif + +/* + * Find the IRQ entry number of a certain pin. + */ +static int find_irq_entry(int apic, int pin, int type) +{ + int i; + + for (i = 0; i < mp_irq_entries; i++) + if (mp_irqs[i].irqtype == type && + (mp_irqs[i].dstapic == mp_ioapics[apic].apicid || + mp_irqs[i].dstapic == MP_APIC_ALL) && + mp_irqs[i].dstirq == pin) + return i; + + return -1; +} + +/* + * Find the pin to which IRQ[irq] (ISA) is connected + */ +static int __init find_isa_irq_pin(int irq, int type) +{ + int i; + + for (i = 0; i < mp_irq_entries; i++) { + int lbus = mp_irqs[i].srcbus; + + if (test_bit(lbus, mp_bus_not_pci) && + (mp_irqs[i].irqtype == type) && + (mp_irqs[i].srcbusirq == irq)) + + return mp_irqs[i].dstirq; + } + return -1; +} + +static int __init find_isa_irq_apic(int irq, int type) +{ + int i; + + for (i = 0; i < mp_irq_entries; i++) { + int lbus = mp_irqs[i].srcbus; + + if (test_bit(lbus, mp_bus_not_pci) && + (mp_irqs[i].irqtype == type) && + (mp_irqs[i].srcbusirq == irq)) + break; + } + if (i < mp_irq_entries) { + int apic; + for(apic = 0; apic < nr_ioapics; apic++) { + if (mp_ioapics[apic].apicid == mp_irqs[i].dstapic) + return apic; + } + } + + return -1; +} + +/* + * Find a specific PCI IRQ entry. + * Not an __init, possibly needed by modules + */ +static int pin_2_irq(int idx, int apic, int pin); + +int IO_APIC_get_PCI_irq_vector(int bus, int slot, int pin) +{ + int apic, i, best_guess = -1; + + apic_printk(APIC_DEBUG, "querying PCI -> IRQ mapping bus:%d, slot:%d, pin:%d.\n", + bus, slot, pin); + if (test_bit(bus, mp_bus_not_pci)) { + apic_printk(APIC_VERBOSE, "PCI BIOS passed nonexistent PCI bus %d!\n", bus); + return -1; + } + for (i = 0; i < mp_irq_entries; i++) { + int lbus = mp_irqs[i].srcbus; + + for (apic = 0; apic < nr_ioapics; apic++) + if (mp_ioapics[apic].apicid == mp_irqs[i].dstapic || + mp_irqs[i].dstapic == MP_APIC_ALL) + break; + + if (!test_bit(lbus, mp_bus_not_pci) && + !mp_irqs[i].irqtype && + (bus == lbus) && + (slot == ((mp_irqs[i].srcbusirq >> 2) & 0x1f))) { + int irq = pin_2_irq(i, apic, mp_irqs[i].dstirq); + + if (!(apic || IO_APIC_IRQ(irq))) + continue; + + if (pin == (mp_irqs[i].srcbusirq & 3)) + return irq; + /* + * Use the first all-but-pin matching entry as a + * best-guess fuzzy result for broken mptables. + */ + if (best_guess < 0) + best_guess = irq; + } + } + return best_guess; +} + +EXPORT_SYMBOL(IO_APIC_get_PCI_irq_vector); + +#if defined(CONFIG_EISA) || defined(CONFIG_MCA) +/* + * EISA Edge/Level control register, ELCR + */ +static int EISA_ELCR(unsigned int irq) +{ + if (irq < NR_IRQS_LEGACY) { + unsigned int port = 0x4d0 + (irq >> 3); + return (inb(port) >> (irq & 7)) & 1; + } + apic_printk(APIC_VERBOSE, KERN_INFO + "Broken MPtable reports ISA irq %d\n", irq); + return 0; +} + +#endif + +/* ISA interrupts are always polarity zero edge triggered, + * when listed as conforming in the MP table. */ + +#define default_ISA_trigger(idx) (0) +#define default_ISA_polarity(idx) (0) + +/* EISA interrupts are always polarity zero and can be edge or level + * trigger depending on the ELCR value. If an interrupt is listed as + * EISA conforming in the MP table, that means its trigger type must + * be read in from the ELCR */ + +#define default_EISA_trigger(idx) (EISA_ELCR(mp_irqs[idx].srcbusirq)) +#define default_EISA_polarity(idx) default_ISA_polarity(idx) + +/* PCI interrupts are always polarity one level triggered, + * when listed as conforming in the MP table. */ + +#define default_PCI_trigger(idx) (1) +#define default_PCI_polarity(idx) (1) + +/* MCA interrupts are always polarity zero level triggered, + * when listed as conforming in the MP table. */ + +#define default_MCA_trigger(idx) (1) +#define default_MCA_polarity(idx) default_ISA_polarity(idx) + +static int MPBIOS_polarity(int idx) +{ + int bus = mp_irqs[idx].srcbus; + int polarity; + + /* + * Determine IRQ line polarity (high active or low active): + */ + switch (mp_irqs[idx].irqflag & 3) + { + case 0: /* conforms, ie. bus-type dependent polarity */ + if (test_bit(bus, mp_bus_not_pci)) + polarity = default_ISA_polarity(idx); + else + polarity = default_PCI_polarity(idx); + break; + case 1: /* high active */ + { + polarity = 0; + break; + } + case 2: /* reserved */ + { + printk(KERN_WARNING "broken BIOS!!\n"); + polarity = 1; + break; + } + case 3: /* low active */ + { + polarity = 1; + break; + } + default: /* invalid */ + { + printk(KERN_WARNING "broken BIOS!!\n"); + polarity = 1; + break; + } + } + return polarity; +} + +static int MPBIOS_trigger(int idx) +{ + int bus = mp_irqs[idx].srcbus; + int trigger; + + /* + * Determine IRQ trigger mode (edge or level sensitive): + */ + switch ((mp_irqs[idx].irqflag>>2) & 3) + { + case 0: /* conforms, ie. bus-type dependent */ + if (test_bit(bus, mp_bus_not_pci)) + trigger = default_ISA_trigger(idx); + else + trigger = default_PCI_trigger(idx); +#if defined(CONFIG_EISA) || defined(CONFIG_MCA) + switch (mp_bus_id_to_type[bus]) { + case MP_BUS_ISA: /* ISA pin */ + { + /* set before the switch */ + break; + } + case MP_BUS_EISA: /* EISA pin */ + { + trigger = default_EISA_trigger(idx); + break; + } + case MP_BUS_PCI: /* PCI pin */ + { + /* set before the switch */ + break; + } + case MP_BUS_MCA: /* MCA pin */ + { + trigger = default_MCA_trigger(idx); + break; + } + default: + { + printk(KERN_WARNING "broken BIOS!!\n"); + trigger = 1; + break; + } + } +#endif + break; + case 1: /* edge */ + { + trigger = 0; + break; + } + case 2: /* reserved */ + { + printk(KERN_WARNING "broken BIOS!!\n"); + trigger = 1; + break; + } + case 3: /* level */ + { + trigger = 1; + break; + } + default: /* invalid */ + { + printk(KERN_WARNING "broken BIOS!!\n"); + trigger = 0; + break; + } + } + return trigger; +} + +static inline int irq_polarity(int idx) +{ + return MPBIOS_polarity(idx); +} + +static inline int irq_trigger(int idx) +{ + return MPBIOS_trigger(idx); +} + +int (*ioapic_renumber_irq)(int ioapic, int irq); +static int pin_2_irq(int idx, int apic, int pin) +{ + int irq, i; + int bus = mp_irqs[idx].srcbus; + + /* + * Debugging check, we are in big trouble if this message pops up! + */ + if (mp_irqs[idx].dstirq != pin) + printk(KERN_ERR "broken BIOS or MPTABLE parser, ayiee!!\n"); + + if (test_bit(bus, mp_bus_not_pci)) { + irq = mp_irqs[idx].srcbusirq; + } else { + /* + * PCI IRQs are mapped in order + */ + i = irq = 0; + while (i < apic) + irq += nr_ioapic_registers[i++]; + irq += pin; + /* + * For MPS mode, so far only needed by ES7000 platform + */ + if (ioapic_renumber_irq) + irq = ioapic_renumber_irq(apic, irq); + } + +#ifdef CONFIG_X86_32 + /* + * PCI IRQ command line redirection. Yes, limits are hardcoded. + */ + if ((pin >= 16) && (pin <= 23)) { + if (pirq_entries[pin-16] != -1) { + if (!pirq_entries[pin-16]) { + apic_printk(APIC_VERBOSE, KERN_DEBUG + "disabling PIRQ%d\n", pin-16); + } else { + irq = pirq_entries[pin-16]; + apic_printk(APIC_VERBOSE, KERN_DEBUG + "using PIRQ%d -> IRQ %d\n", + pin-16, irq); + } + } + } +#endif + + return irq; +} + +void lock_vector_lock(void) +{ + /* Used to the online set of cpus does not change + * during assign_irq_vector. + */ + spin_lock(&vector_lock); +} + +void unlock_vector_lock(void) +{ + spin_unlock(&vector_lock); +} + +static int +__assign_irq_vector(int irq, struct irq_cfg *cfg, const struct cpumask *mask) +{ + /* + * NOTE! The local APIC isn't very good at handling + * multiple interrupts at the same interrupt level. + * As the interrupt level is determined by taking the + * vector number and shifting that right by 4, we + * want to spread these out a bit so that they don't + * all fall in the same interrupt level. + * + * Also, we've got to be careful not to trash gate + * 0x80, because int 0x80 is hm, kind of importantish. ;) + */ + static int current_vector = FIRST_DEVICE_VECTOR, current_offset = 0; + unsigned int old_vector; + int cpu, err; + cpumask_var_t tmp_mask; + + if ((cfg->move_in_progress) || cfg->move_cleanup_count) + return -EBUSY; + + if (!alloc_cpumask_var(&tmp_mask, GFP_ATOMIC)) + return -ENOMEM; + + old_vector = cfg->vector; + if (old_vector) { + cpumask_and(tmp_mask, mask, cpu_online_mask); + cpumask_and(tmp_mask, cfg->domain, tmp_mask); + if (!cpumask_empty(tmp_mask)) { + free_cpumask_var(tmp_mask); + return 0; + } + } + + /* Only try and allocate irqs on cpus that are present */ + err = -ENOSPC; + for_each_cpu_and(cpu, mask, cpu_online_mask) { + int new_cpu; + int vector, offset; + + apic->vector_allocation_domain(cpu, tmp_mask); + + vector = current_vector; + offset = current_offset; +next: + vector += 8; + if (vector >= first_system_vector) { + /* If out of vectors on large boxen, must share them. */ + offset = (offset + 1) % 8; + vector = FIRST_DEVICE_VECTOR + offset; + } + if (unlikely(current_vector == vector)) + continue; + + if (test_bit(vector, used_vectors)) + goto next; + + for_each_cpu_and(new_cpu, tmp_mask, cpu_online_mask) + if (per_cpu(vector_irq, new_cpu)[vector] != -1) + goto next; + /* Found one! */ + current_vector = vector; + current_offset = offset; + if (old_vector) { + cfg->move_in_progress = 1; + cpumask_copy(cfg->old_domain, cfg->domain); + } + for_each_cpu_and(new_cpu, tmp_mask, cpu_online_mask) + per_cpu(vector_irq, new_cpu)[vector] = irq; + cfg->vector = vector; + cpumask_copy(cfg->domain, tmp_mask); + err = 0; + break; + } + free_cpumask_var(tmp_mask); + return err; +} + +static int +assign_irq_vector(int irq, struct irq_cfg *cfg, const struct cpumask *mask) +{ + int err; + unsigned long flags; + + spin_lock_irqsave(&vector_lock, flags); + err = __assign_irq_vector(irq, cfg, mask); + spin_unlock_irqrestore(&vector_lock, flags); + return err; +} + +static void __clear_irq_vector(int irq, struct irq_cfg *cfg) +{ + int cpu, vector; + + BUG_ON(!cfg->vector); + + vector = cfg->vector; + for_each_cpu_and(cpu, cfg->domain, cpu_online_mask) + per_cpu(vector_irq, cpu)[vector] = -1; + + cfg->vector = 0; + cpumask_clear(cfg->domain); + + if (likely(!cfg->move_in_progress)) + return; + for_each_cpu_and(cpu, cfg->old_domain, cpu_online_mask) { + for (vector = FIRST_EXTERNAL_VECTOR; vector < NR_VECTORS; + vector++) { + if (per_cpu(vector_irq, cpu)[vector] != irq) + continue; + per_cpu(vector_irq, cpu)[vector] = -1; + break; + } + } + cfg->move_in_progress = 0; +} + +void __setup_vector_irq(int cpu) +{ + /* Initialize vector_irq on a new cpu */ + /* This function must be called with vector_lock held */ + int irq, vector; + struct irq_cfg *cfg; + struct irq_desc *desc; + + /* Mark the inuse vectors */ + for_each_irq_desc(irq, desc) { + cfg = desc->chip_data; + if (!cpumask_test_cpu(cpu, cfg->domain)) + continue; + vector = cfg->vector; + per_cpu(vector_irq, cpu)[vector] = irq; + } + /* Mark the free vectors */ + for (vector = 0; vector < NR_VECTORS; ++vector) { + irq = per_cpu(vector_irq, cpu)[vector]; + if (irq < 0) + continue; + + cfg = irq_cfg(irq); + if (!cpumask_test_cpu(cpu, cfg->domain)) + per_cpu(vector_irq, cpu)[vector] = -1; + } +} + +static struct irq_chip ioapic_chip; +#ifdef CONFIG_INTR_REMAP +static struct irq_chip ir_ioapic_chip; +#endif + +#define IOAPIC_AUTO -1 +#define IOAPIC_EDGE 0 +#define IOAPIC_LEVEL 1 + +#ifdef CONFIG_X86_32 +static inline int IO_APIC_irq_trigger(int irq) +{ + int apic, idx, pin; + + for (apic = 0; apic < nr_ioapics; apic++) { + for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) { + idx = find_irq_entry(apic, pin, mp_INT); + if ((idx != -1) && (irq == pin_2_irq(idx, apic, pin))) + return irq_trigger(idx); + } + } + /* + * nonexistent IRQs are edge default + */ + return 0; +} +#else +static inline int IO_APIC_irq_trigger(int irq) +{ + return 1; +} +#endif + +static void ioapic_register_intr(int irq, struct irq_desc *desc, unsigned long trigger) +{ + + if ((trigger == IOAPIC_AUTO && IO_APIC_irq_trigger(irq)) || + trigger == IOAPIC_LEVEL) + desc->status |= IRQ_LEVEL; + else + desc->status &= ~IRQ_LEVEL; + +#ifdef CONFIG_INTR_REMAP + if (irq_remapped(irq)) { + desc->status |= IRQ_MOVE_PCNTXT; + if (trigger) + set_irq_chip_and_handler_name(irq, &ir_ioapic_chip, + handle_fasteoi_irq, + "fasteoi"); + else + set_irq_chip_and_handler_name(irq, &ir_ioapic_chip, + handle_edge_irq, "edge"); + return; + } +#endif + if ((trigger == IOAPIC_AUTO && IO_APIC_irq_trigger(irq)) || + trigger == IOAPIC_LEVEL) + set_irq_chip_and_handler_name(irq, &ioapic_chip, + handle_fasteoi_irq, + "fasteoi"); + else + set_irq_chip_and_handler_name(irq, &ioapic_chip, + handle_edge_irq, "edge"); +} + +int setup_ioapic_entry(int apic_id, int irq, + struct IO_APIC_route_entry *entry, + unsigned int destination, int trigger, + int polarity, int vector) +{ + /* + * add it to the IO-APIC irq-routing table: + */ + memset(entry,0,sizeof(*entry)); + +#ifdef CONFIG_INTR_REMAP + if (intr_remapping_enabled) { + struct intel_iommu *iommu = map_ioapic_to_ir(apic_id); + struct irte irte; + struct IR_IO_APIC_route_entry *ir_entry = + (struct IR_IO_APIC_route_entry *) entry; + int index; + + if (!iommu) + panic("No mapping iommu for ioapic %d\n", apic_id); + + index = alloc_irte(iommu, irq, 1); + if (index < 0) + panic("Failed to allocate IRTE for ioapic %d\n", apic_id); + + memset(&irte, 0, sizeof(irte)); + + irte.present = 1; + irte.dst_mode = apic->irq_dest_mode; + irte.trigger_mode = trigger; + irte.dlvry_mode = apic->irq_delivery_mode; + irte.vector = vector; + irte.dest_id = IRTE_DEST(destination); + + modify_irte(irq, &irte); + + ir_entry->index2 = (index >> 15) & 0x1; + ir_entry->zero = 0; + ir_entry->format = 1; + ir_entry->index = (index & 0x7fff); + } else +#endif + { + entry->delivery_mode = apic->irq_delivery_mode; + entry->dest_mode = apic->irq_dest_mode; + entry->dest = destination; + } + + entry->mask = 0; /* enable IRQ */ + entry->trigger = trigger; + entry->polarity = polarity; + entry->vector = vector; + + /* Mask level triggered irqs. + * Use IRQ_DELAYED_DISABLE for edge triggered irqs. + */ + if (trigger) + entry->mask = 1; + return 0; +} + +static void setup_IO_APIC_irq(int apic_id, int pin, unsigned int irq, struct irq_desc *desc, + int trigger, int polarity) +{ + struct irq_cfg *cfg; + struct IO_APIC_route_entry entry; + unsigned int dest; + + if (!IO_APIC_IRQ(irq)) + return; + + cfg = desc->chip_data; + + if (assign_irq_vector(irq, cfg, apic->target_cpus())) + return; + + dest = apic->cpu_mask_to_apicid_and(cfg->domain, apic->target_cpus()); + + apic_printk(APIC_VERBOSE,KERN_DEBUG + "IOAPIC[%d]: Set routing entry (%d-%d -> 0x%x -> " + "IRQ %d Mode:%i Active:%i)\n", + apic_id, mp_ioapics[apic_id].apicid, pin, cfg->vector, + irq, trigger, polarity); + + + if (setup_ioapic_entry(mp_ioapics[apic_id].apicid, irq, &entry, + dest, trigger, polarity, cfg->vector)) { + printk("Failed to setup ioapic entry for ioapic %d, pin %d\n", + mp_ioapics[apic_id].apicid, pin); + __clear_irq_vector(irq, cfg); + return; + } + + ioapic_register_intr(irq, desc, trigger); + if (irq < NR_IRQS_LEGACY) + disable_8259A_irq(irq); + + ioapic_write_entry(apic_id, pin, entry); +} + +static void __init setup_IO_APIC_irqs(void) +{ + int apic_id, pin, idx, irq; + int notcon = 0; + struct irq_desc *desc; + struct irq_cfg *cfg; + int cpu = boot_cpu_id; + + apic_printk(APIC_VERBOSE, KERN_DEBUG "init IO_APIC IRQs\n"); + + for (apic_id = 0; apic_id < nr_ioapics; apic_id++) { + for (pin = 0; pin < nr_ioapic_registers[apic_id]; pin++) { + + idx = find_irq_entry(apic_id, pin, mp_INT); + if (idx == -1) { + if (!notcon) { + notcon = 1; + apic_printk(APIC_VERBOSE, + KERN_DEBUG " %d-%d", + mp_ioapics[apic_id].apicid, pin); + } else + apic_printk(APIC_VERBOSE, " %d-%d", + mp_ioapics[apic_id].apicid, pin); + continue; + } + if (notcon) { + apic_printk(APIC_VERBOSE, + " (apicid-pin) not connected\n"); + notcon = 0; + } + + irq = pin_2_irq(idx, apic_id, pin); + + /* + * Skip the timer IRQ if there's a quirk handler + * installed and if it returns 1: + */ + if (apic->multi_timer_check && + apic->multi_timer_check(apic_id, irq)) + continue; + + desc = irq_to_desc_alloc_cpu(irq, cpu); + if (!desc) { + printk(KERN_INFO "can not get irq_desc for %d\n", irq); + continue; + } + cfg = desc->chip_data; + add_pin_to_irq_cpu(cfg, cpu, apic_id, pin); + + setup_IO_APIC_irq(apic_id, pin, irq, desc, + irq_trigger(idx), irq_polarity(idx)); + } + } + + if (notcon) + apic_printk(APIC_VERBOSE, + " (apicid-pin) not connected\n"); +} + +/* + * Set up the timer pin, possibly with the 8259A-master behind. + */ +static void __init setup_timer_IRQ0_pin(unsigned int apic_id, unsigned int pin, + int vector) +{ + struct IO_APIC_route_entry entry; + +#ifdef CONFIG_INTR_REMAP + if (intr_remapping_enabled) + return; +#endif + + memset(&entry, 0, sizeof(entry)); + + /* + * We use logical delivery to get the timer IRQ + * to the first CPU. + */ + entry.dest_mode = apic->irq_dest_mode; + entry.mask = 0; /* don't mask IRQ for edge */ + entry.dest = apic->cpu_mask_to_apicid(apic->target_cpus()); + entry.delivery_mode = apic->irq_delivery_mode; + entry.polarity = 0; + entry.trigger = 0; + entry.vector = vector; + + /* + * The timer IRQ doesn't have to know that behind the + * scene we may have a 8259A-master in AEOI mode ... + */ + set_irq_chip_and_handler_name(0, &ioapic_chip, handle_edge_irq, "edge"); + + /* + * Add it to the IO-APIC irq-routing table: + */ + ioapic_write_entry(apic_id, pin, entry); +} + + +__apicdebuginit(void) print_IO_APIC(void) +{ + int apic, i; + union IO_APIC_reg_00 reg_00; + union IO_APIC_reg_01 reg_01; + union IO_APIC_reg_02 reg_02; + union IO_APIC_reg_03 reg_03; + unsigned long flags; + struct irq_cfg *cfg; + struct irq_desc *desc; + unsigned int irq; + + if (apic_verbosity == APIC_QUIET) + return; + + printk(KERN_DEBUG "number of MP IRQ sources: %d.\n", mp_irq_entries); + for (i = 0; i < nr_ioapics; i++) + printk(KERN_DEBUG "number of IO-APIC #%d registers: %d.\n", + mp_ioapics[i].apicid, nr_ioapic_registers[i]); + + /* + * We are a bit conservative about what we expect. We have to + * know about every hardware change ASAP. + */ + printk(KERN_INFO "testing the IO APIC.......................\n"); + + for (apic = 0; apic < nr_ioapics; apic++) { + + spin_lock_irqsave(&ioapic_lock, flags); + reg_00.raw = io_apic_read(apic, 0); + reg_01.raw = io_apic_read(apic, 1); + if (reg_01.bits.version >= 0x10) + reg_02.raw = io_apic_read(apic, 2); + if (reg_01.bits.version >= 0x20) + reg_03.raw = io_apic_read(apic, 3); + spin_unlock_irqrestore(&ioapic_lock, flags); + + printk("\n"); + printk(KERN_DEBUG "IO APIC #%d......\n", mp_ioapics[apic].apicid); + printk(KERN_DEBUG ".... register #00: %08X\n", reg_00.raw); + printk(KERN_DEBUG "....... : physical APIC id: %02X\n", reg_00.bits.ID); + printk(KERN_DEBUG "....... : Delivery Type: %X\n", reg_00.bits.delivery_type); + printk(KERN_DEBUG "....... : LTS : %X\n", reg_00.bits.LTS); + + printk(KERN_DEBUG ".... register #01: %08X\n", *(int *)®_01); + printk(KERN_DEBUG "....... : max redirection entries: %04X\n", reg_01.bits.entries); + + printk(KERN_DEBUG "....... : PRQ implemented: %X\n", reg_01.bits.PRQ); + printk(KERN_DEBUG "....... : IO APIC version: %04X\n", reg_01.bits.version); + + /* + * Some Intel chipsets with IO APIC VERSION of 0x1? don't have reg_02, + * but the value of reg_02 is read as the previous read register + * value, so ignore it if reg_02 == reg_01. + */ + if (reg_01.bits.version >= 0x10 && reg_02.raw != reg_01.raw) { + printk(KERN_DEBUG ".... register #02: %08X\n", reg_02.raw); + printk(KERN_DEBUG "....... : arbitration: %02X\n", reg_02.bits.arbitration); + } + + /* + * Some Intel chipsets with IO APIC VERSION of 0x2? don't have reg_02 + * or reg_03, but the value of reg_0[23] is read as the previous read + * register value, so ignore it if reg_03 == reg_0[12]. + */ + if (reg_01.bits.version >= 0x20 && reg_03.raw != reg_02.raw && + reg_03.raw != reg_01.raw) { + printk(KERN_DEBUG ".... register #03: %08X\n", reg_03.raw); + printk(KERN_DEBUG "....... : Boot DT : %X\n", reg_03.bits.boot_DT); + } + + printk(KERN_DEBUG ".... IRQ redirection table:\n"); + + printk(KERN_DEBUG " NR Dst Mask Trig IRR Pol" + " Stat Dmod Deli Vect: \n"); + + for (i = 0; i <= reg_01.bits.entries; i++) { + struct IO_APIC_route_entry entry; + + entry = ioapic_read_entry(apic, i); + + printk(KERN_DEBUG " %02x %03X ", + i, + entry.dest + ); + + printk("%1d %1d %1d %1d %1d %1d %1d %02X\n", + entry.mask, + entry.trigger, + entry.irr, + entry.polarity, + entry.delivery_status, + entry.dest_mode, + entry.delivery_mode, + entry.vector + ); + } + } + printk(KERN_DEBUG "IRQ to pin mappings:\n"); + for_each_irq_desc(irq, desc) { + struct irq_pin_list *entry; + + cfg = desc->chip_data; + entry = cfg->irq_2_pin; + if (!entry) + continue; + printk(KERN_DEBUG "IRQ%d ", irq); + for (;;) { + printk("-> %d:%d", entry->apic, entry->pin); + if (!entry->next) + break; + entry = entry->next; + } + printk("\n"); + } + + printk(KERN_INFO ".................................... done.\n"); + + return; +} + +__apicdebuginit(void) print_APIC_bitfield(int base) +{ + unsigned int v; + int i, j; + + if (apic_verbosity == APIC_QUIET) + return; + + printk(KERN_DEBUG "0123456789abcdef0123456789abcdef\n" KERN_DEBUG); + for (i = 0; i < 8; i++) { + v = apic_read(base + i*0x10); + for (j = 0; j < 32; j++) { + if (v & (1<<j)) + printk("1"); + else + printk("0"); + } + printk("\n"); + } +} + +__apicdebuginit(void) print_local_APIC(void *dummy) +{ + unsigned int v, ver, maxlvt; + u64 icr; + + if (apic_verbosity == APIC_QUIET) + return; + + printk("\n" KERN_DEBUG "printing local APIC contents on CPU#%d/%d:\n", + smp_processor_id(), hard_smp_processor_id()); + v = apic_read(APIC_ID); + printk(KERN_INFO "... APIC ID: %08x (%01x)\n", v, read_apic_id()); + v = apic_read(APIC_LVR); + printk(KERN_INFO "... APIC VERSION: %08x\n", v); + ver = GET_APIC_VERSION(v); + maxlvt = lapic_get_maxlvt(); + + v = apic_read(APIC_TASKPRI); + printk(KERN_DEBUG "... APIC TASKPRI: %08x (%02x)\n", v, v & APIC_TPRI_MASK); + + if (APIC_INTEGRATED(ver)) { /* !82489DX */ + if (!APIC_XAPIC(ver)) { + v = apic_read(APIC_ARBPRI); + printk(KERN_DEBUG "... APIC ARBPRI: %08x (%02x)\n", v, + v & APIC_ARBPRI_MASK); + } + v = apic_read(APIC_PROCPRI); + printk(KERN_DEBUG "... APIC PROCPRI: %08x\n", v); + } + + /* + * Remote read supported only in the 82489DX and local APIC for + * Pentium processors. + */ + if (!APIC_INTEGRATED(ver) || maxlvt == 3) { + v = apic_read(APIC_RRR); + printk(KERN_DEBUG "... APIC RRR: %08x\n", v); + } + + v = apic_read(APIC_LDR); + printk(KERN_DEBUG "... APIC LDR: %08x\n", v); + if (!x2apic_enabled()) { + v = apic_read(APIC_DFR); + printk(KERN_DEBUG "... APIC DFR: %08x\n", v); + } + v = apic_read(APIC_SPIV); + printk(KERN_DEBUG "... APIC SPIV: %08x\n", v); + + printk(KERN_DEBUG "... APIC ISR field:\n"); + print_APIC_bitfield(APIC_ISR); + printk(KERN_DEBUG "... APIC TMR field:\n"); + print_APIC_bitfield(APIC_TMR); + printk(KERN_DEBUG "... APIC IRR field:\n"); + print_APIC_bitfield(APIC_IRR); + + if (APIC_INTEGRATED(ver)) { /* !82489DX */ + if (maxlvt > 3) /* Due to the Pentium erratum 3AP. */ + apic_write(APIC_ESR, 0); + + v = apic_read(APIC_ESR); + printk(KERN_DEBUG "... APIC ESR: %08x\n", v); + } + + icr = apic_icr_read(); + printk(KERN_DEBUG "... APIC ICR: %08x\n", (u32)icr); + printk(KERN_DEBUG "... APIC ICR2: %08x\n", (u32)(icr >> 32)); + + v = apic_read(APIC_LVTT); + printk(KERN_DEBUG "... APIC LVTT: %08x\n", v); + + if (maxlvt > 3) { /* PC is LVT#4. */ + v = apic_read(APIC_LVTPC); + printk(KERN_DEBUG "... APIC LVTPC: %08x\n", v); + } + v = apic_read(APIC_LVT0); + printk(KERN_DEBUG "... APIC LVT0: %08x\n", v); + v = apic_read(APIC_LVT1); + printk(KERN_DEBUG "... APIC LVT1: %08x\n", v); + + if (maxlvt > 2) { /* ERR is LVT#3. */ + v = apic_read(APIC_LVTERR); + printk(KERN_DEBUG "... APIC LVTERR: %08x\n", v); + } + + v = apic_read(APIC_TMICT); + printk(KERN_DEBUG "... APIC TMICT: %08x\n", v); + v = apic_read(APIC_TMCCT); + printk(KERN_DEBUG "... APIC TMCCT: %08x\n", v); + v = apic_read(APIC_TDCR); + printk(KERN_DEBUG "... APIC TDCR: %08x\n", v); + printk("\n"); +} + +__apicdebuginit(void) print_all_local_APICs(void) +{ + int cpu; + + preempt_disable(); + for_each_online_cpu(cpu) + smp_call_function_single(cpu, print_local_APIC, NULL, 1); + preempt_enable(); +} + +__apicdebuginit(void) print_PIC(void) +{ + unsigned int v; + unsigned long flags; + + if (apic_verbosity == APIC_QUIET) + return; + + printk(KERN_DEBUG "\nprinting PIC contents\n"); + + spin_lock_irqsave(&i8259A_lock, flags); + + v = inb(0xa1) << 8 | inb(0x21); + printk(KERN_DEBUG "... PIC IMR: %04x\n", v); + + v = inb(0xa0) << 8 | inb(0x20); + printk(KERN_DEBUG "... PIC IRR: %04x\n", v); + + outb(0x0b,0xa0); + outb(0x0b,0x20); + v = inb(0xa0) << 8 | inb(0x20); + outb(0x0a,0xa0); + outb(0x0a,0x20); + + spin_unlock_irqrestore(&i8259A_lock, flags); + + printk(KERN_DEBUG "... PIC ISR: %04x\n", v); + + v = inb(0x4d1) << 8 | inb(0x4d0); + printk(KERN_DEBUG "... PIC ELCR: %04x\n", v); +} + +__apicdebuginit(int) print_all_ICs(void) +{ + print_PIC(); + print_all_local_APICs(); + print_IO_APIC(); + + return 0; +} + +fs_initcall(print_all_ICs); + + +/* Where if anywhere is the i8259 connect in external int mode */ +static struct { int pin, apic; } ioapic_i8259 = { -1, -1 }; + +void __init enable_IO_APIC(void) +{ + union IO_APIC_reg_01 reg_01; + int i8259_apic, i8259_pin; + int apic; + unsigned long flags; + + /* + * The number of IO-APIC IRQ registers (== #pins): + */ + for (apic = 0; apic < nr_ioapics; apic++) { + spin_lock_irqsave(&ioapic_lock, flags); + reg_01.raw = io_apic_read(apic, 1); + spin_unlock_irqrestore(&ioapic_lock, flags); + nr_ioapic_registers[apic] = reg_01.bits.entries+1; + } + for(apic = 0; apic < nr_ioapics; apic++) { + int pin; + /* See if any of the pins is in ExtINT mode */ + for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) { + struct IO_APIC_route_entry entry; + entry = ioapic_read_entry(apic, pin); + + /* If the interrupt line is enabled and in ExtInt mode + * I have found the pin where the i8259 is connected. + */ + if ((entry.mask == 0) && (entry.delivery_mode == dest_ExtINT)) { + ioapic_i8259.apic = apic; + ioapic_i8259.pin = pin; + goto found_i8259; + } + } + } + found_i8259: + /* Look to see what if the MP table has reported the ExtINT */ + /* If we could not find the appropriate pin by looking at the ioapic + * the i8259 probably is not connected the ioapic but give the + * mptable a chance anyway. + */ + i8259_pin = find_isa_irq_pin(0, mp_ExtINT); + i8259_apic = find_isa_irq_apic(0, mp_ExtINT); + /* Trust the MP table if nothing is setup in the hardware */ + if ((ioapic_i8259.pin == -1) && (i8259_pin >= 0)) { + printk(KERN_WARNING "ExtINT not setup in hardware but reported by MP table\n"); + ioapic_i8259.pin = i8259_pin; + ioapic_i8259.apic = i8259_apic; + } + /* Complain if the MP table and the hardware disagree */ + if (((ioapic_i8259.apic != i8259_apic) || (ioapic_i8259.pin != i8259_pin)) && + (i8259_pin >= 0) && (ioapic_i8259.pin >= 0)) + { + printk(KERN_WARNING "ExtINT in hardware and MP table differ\n"); + } + + /* + * Do not trust the IO-APIC being empty at bootup + */ + clear_IO_APIC(); +} + +/* + * Not an __init, needed by the reboot code + */ +void disable_IO_APIC(void) +{ + /* + * Clear the IO-APIC before rebooting: + */ + clear_IO_APIC(); + + /* + * If the i8259 is routed through an IOAPIC + * Put that IOAPIC in virtual wire mode + * so legacy interrupts can be delivered. + */ + if (ioapic_i8259.pin != -1) { + struct IO_APIC_route_entry entry; + + memset(&entry, 0, sizeof(entry)); + entry.mask = 0; /* Enabled */ + entry.trigger = 0; /* Edge */ + entry.irr = 0; + entry.polarity = 0; /* High */ + entry.delivery_status = 0; + entry.dest_mode = 0; /* Physical */ + entry.delivery_mode = dest_ExtINT; /* ExtInt */ + entry.vector = 0; + entry.dest = read_apic_id(); + + /* + * Add it to the IO-APIC irq-routing table: + */ + ioapic_write_entry(ioapic_i8259.apic, ioapic_i8259.pin, entry); + } + + disconnect_bsp_APIC(ioapic_i8259.pin != -1); +} + +#ifdef CONFIG_X86_32 +/* + * function to set the IO-APIC physical IDs based on the + * values stored in the MPC table. + * + * by Matt Domsch <Matt_Domsch@dell.com> Tue Dec 21 12:25:05 CST 1999 + */ + +static void __init setup_ioapic_ids_from_mpc(void) +{ + union IO_APIC_reg_00 reg_00; + physid_mask_t phys_id_present_map; + int apic_id; + int i; + unsigned char old_id; + unsigned long flags; + + if (x86_quirks->setup_ioapic_ids && x86_quirks->setup_ioapic_ids()) + return; + + /* + * Don't check I/O APIC IDs for xAPIC systems. They have + * no meaning without the serial APIC bus. + */ + if (!(boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) + || APIC_XAPIC(apic_version[boot_cpu_physical_apicid])) + return; + /* + * This is broken; anything with a real cpu count has to + * circumvent this idiocy regardless. + */ + phys_id_present_map = apic->ioapic_phys_id_map(phys_cpu_present_map); + + /* + * Set the IOAPIC ID to the value stored in the MPC table. + */ + for (apic_id = 0; apic_id < nr_ioapics; apic_id++) { + + /* Read the register 0 value */ + spin_lock_irqsave(&ioapic_lock, flags); + reg_00.raw = io_apic_read(apic_id, 0); + spin_unlock_irqrestore(&ioapic_lock, flags); + + old_id = mp_ioapics[apic_id].apicid; + + if (mp_ioapics[apic_id].apicid >= get_physical_broadcast()) { + printk(KERN_ERR "BIOS bug, IO-APIC#%d ID is %d in the MPC table!...\n", + apic_id, mp_ioapics[apic_id].apicid); + printk(KERN_ERR "... fixing up to %d. (tell your hw vendor)\n", + reg_00.bits.ID); + mp_ioapics[apic_id].apicid = reg_00.bits.ID; + } + + /* + * Sanity check, is the ID really free? Every APIC in a + * system must have a unique ID or we get lots of nice + * 'stuck on smp_invalidate_needed IPI wait' messages. + */ + if (apic->check_apicid_used(phys_id_present_map, + mp_ioapics[apic_id].apicid)) { + printk(KERN_ERR "BIOS bug, IO-APIC#%d ID %d is already used!...\n", + apic_id, mp_ioapics[apic_id].apicid); + for (i = 0; i < get_physical_broadcast(); i++) + if (!physid_isset(i, phys_id_present_map)) + break; + if (i >= get_physical_broadcast()) + panic("Max APIC ID exceeded!\n"); + printk(KERN_ERR "... fixing up to %d. (tell your hw vendor)\n", + i); + physid_set(i, phys_id_present_map); + mp_ioapics[apic_id].apicid = i; + } else { + physid_mask_t tmp; + tmp = apic->apicid_to_cpu_present(mp_ioapics[apic_id].apicid); + apic_printk(APIC_VERBOSE, "Setting %d in the " + "phys_id_present_map\n", + mp_ioapics[apic_id].apicid); + physids_or(phys_id_present_map, phys_id_present_map, tmp); + } + + + /* + * We need to adjust the IRQ routing table + * if the ID changed. + */ + if (old_id != mp_ioapics[apic_id].apicid) + for (i = 0; i < mp_irq_entries; i++) + if (mp_irqs[i].dstapic == old_id) + mp_irqs[i].dstapic + = mp_ioapics[apic_id].apicid; + + /* + * Read the right value from the MPC table and + * write it into the ID register. + */ + apic_printk(APIC_VERBOSE, KERN_INFO + "...changing IO-APIC physical APIC ID to %d ...", + mp_ioapics[apic_id].apicid); + + reg_00.bits.ID = mp_ioapics[apic_id].apicid; + spin_lock_irqsave(&ioapic_lock, flags); + io_apic_write(apic_id, 0, reg_00.raw); + spin_unlock_irqrestore(&ioapic_lock, flags); + + /* + * Sanity check + */ + spin_lock_irqsave(&ioapic_lock, flags); + reg_00.raw = io_apic_read(apic_id, 0); + spin_unlock_irqrestore(&ioapic_lock, flags); + if (reg_00.bits.ID != mp_ioapics[apic_id].apicid) + printk("could not set ID!\n"); + else + apic_printk(APIC_VERBOSE, " ok.\n"); + } +} +#endif + +int no_timer_check __initdata; + +static int __init notimercheck(char *s) +{ + no_timer_check = 1; + return 1; +} +__setup("no_timer_check", notimercheck); + +/* + * There is a nasty bug in some older SMP boards, their mptable lies + * about the timer IRQ. We do the following to work around the situation: + * + * - timer IRQ defaults to IO-APIC IRQ + * - if this function detects that timer IRQs are defunct, then we fall + * back to ISA timer IRQs + */ +static int __init timer_irq_works(void) +{ + unsigned long t1 = jiffies; + unsigned long flags; + + if (no_timer_check) + return 1; + + local_save_flags(flags); + local_irq_enable(); + /* Let ten ticks pass... */ + mdelay((10 * 1000) / HZ); + local_irq_restore(flags); + + /* + * Expect a few ticks at least, to be sure some possible + * glue logic does not lock up after one or two first + * ticks in a non-ExtINT mode. Also the local APIC + * might have cached one ExtINT interrupt. Finally, at + * least one tick may be lost due to delays. + */ + + /* jiffies wrap? */ + if (time_after(jiffies, t1 + 4)) + return 1; + return 0; +} + +/* + * In the SMP+IOAPIC case it might happen that there are an unspecified + * number of pending IRQ events unhandled. These cases are very rare, + * so we 'resend' these IRQs via IPIs, to the same CPU. It's much + * better to do it this way as thus we do not have to be aware of + * 'pending' interrupts in the IRQ path, except at this point. + */ +/* + * Edge triggered needs to resend any interrupt + * that was delayed but this is now handled in the device + * independent code. + */ + +/* + * Starting up a edge-triggered IO-APIC interrupt is + * nasty - we need to make sure that we get the edge. + * If it is already asserted for some reason, we need + * return 1 to indicate that is was pending. + * + * This is not complete - we should be able to fake + * an edge even if it isn't on the 8259A... + */ + +static unsigned int startup_ioapic_irq(unsigned int irq) +{ + int was_pending = 0; + unsigned long flags; + struct irq_cfg *cfg; + + spin_lock_irqsave(&ioapic_lock, flags); + if (irq < NR_IRQS_LEGACY) { + disable_8259A_irq(irq); + if (i8259A_irq_pending(irq)) + was_pending = 1; + } + cfg = irq_cfg(irq); + __unmask_IO_APIC_irq(cfg); + spin_unlock_irqrestore(&ioapic_lock, flags); + + return was_pending; +} + +#ifdef CONFIG_X86_64 +static int ioapic_retrigger_irq(unsigned int irq) +{ + + struct irq_cfg *cfg = irq_cfg(irq); + unsigned long flags; + + spin_lock_irqsave(&vector_lock, flags); + apic->send_IPI_mask(cpumask_of(cpumask_first(cfg->domain)), cfg->vector); + spin_unlock_irqrestore(&vector_lock, flags); + + return 1; +} +#else +static int ioapic_retrigger_irq(unsigned int irq) +{ + apic->send_IPI_self(irq_cfg(irq)->vector); + + return 1; +} +#endif + +/* + * Level and edge triggered IO-APIC interrupts need different handling, + * so we use two separate IRQ descriptors. Edge triggered IRQs can be + * handled with the level-triggered descriptor, but that one has slightly + * more overhead. Level-triggered interrupts cannot be handled with the + * edge-triggered handler, without risking IRQ storms and other ugly + * races. + */ + +#ifdef CONFIG_SMP + +#ifdef CONFIG_INTR_REMAP +static void ir_irq_migration(struct work_struct *work); + +static DECLARE_DELAYED_WORK(ir_migration_work, ir_irq_migration); + +/* + * Migrate the IO-APIC irq in the presence of intr-remapping. + * + * For edge triggered, irq migration is a simple atomic update(of vector + * and cpu destination) of IRTE and flush the hardware cache. + * + * For level triggered, we need to modify the io-apic RTE aswell with the update + * vector information, along with modifying IRTE with vector and destination. + * So irq migration for level triggered is little bit more complex compared to + * edge triggered migration. But the good news is, we use the same algorithm + * for level triggered migration as we have today, only difference being, + * we now initiate the irq migration from process context instead of the + * interrupt context. + * + * In future, when we do a directed EOI (combined with cpu EOI broadcast + * suppression) to the IO-APIC, level triggered irq migration will also be + * as simple as edge triggered migration and we can do the irq migration + * with a simple atomic update to IO-APIC RTE. + */ +static void +migrate_ioapic_irq_desc(struct irq_desc *desc, const struct cpumask *mask) +{ + struct irq_cfg *cfg; + struct irte irte; + int modify_ioapic_rte; + unsigned int dest; + unsigned long flags; + unsigned int irq; + + if (!cpumask_intersects(mask, cpu_online_mask)) + return; + + irq = desc->irq; + if (get_irte(irq, &irte)) + return; + + cfg = desc->chip_data; + if (assign_irq_vector(irq, cfg, mask)) + return; + + set_extra_move_desc(desc, mask); + + dest = apic->cpu_mask_to_apicid_and(cfg->domain, mask); + + modify_ioapic_rte = desc->status & IRQ_LEVEL; + if (modify_ioapic_rte) { + spin_lock_irqsave(&ioapic_lock, flags); + __target_IO_APIC_irq(irq, dest, cfg); + spin_unlock_irqrestore(&ioapic_lock, flags); + } + + irte.vector = cfg->vector; + irte.dest_id = IRTE_DEST(dest); + + /* + * Modified the IRTE and flushes the Interrupt entry cache. + */ + modify_irte(irq, &irte); + + if (cfg->move_in_progress) + send_cleanup_vector(cfg); + + cpumask_copy(desc->affinity, mask); +} + +static int migrate_irq_remapped_level_desc(struct irq_desc *desc) +{ + int ret = -1; + struct irq_cfg *cfg = desc->chip_data; + + mask_IO_APIC_irq_desc(desc); + + if (io_apic_level_ack_pending(cfg)) { + /* + * Interrupt in progress. Migrating irq now will change the + * vector information in the IO-APIC RTE and that will confuse + * the EOI broadcast performed by cpu. + * So, delay the irq migration to the next instance. + */ + schedule_delayed_work(&ir_migration_work, 1); + goto unmask; + } + + /* everthing is clear. we have right of way */ + migrate_ioapic_irq_desc(desc, desc->pending_mask); + + ret = 0; + desc->status &= ~IRQ_MOVE_PENDING; + cpumask_clear(desc->pending_mask); + +unmask: + unmask_IO_APIC_irq_desc(desc); + + return ret; +} + +static void ir_irq_migration(struct work_struct *work) +{ + unsigned int irq; + struct irq_desc *desc; + + for_each_irq_desc(irq, desc) { + if (desc->status & IRQ_MOVE_PENDING) { + unsigned long flags; + + spin_lock_irqsave(&desc->lock, flags); + if (!desc->chip->set_affinity || + !(desc->status & IRQ_MOVE_PENDING)) { + desc->status &= ~IRQ_MOVE_PENDING; + spin_unlock_irqrestore(&desc->lock, flags); + continue; + } + + desc->chip->set_affinity(irq, desc->pending_mask); + spin_unlock_irqrestore(&desc->lock, flags); + } + } +} + +/* + * Migrates the IRQ destination in the process context. + */ +static void set_ir_ioapic_affinity_irq_desc(struct irq_desc *desc, + const struct cpumask *mask) +{ + if (desc->status & IRQ_LEVEL) { + desc->status |= IRQ_MOVE_PENDING; + cpumask_copy(desc->pending_mask, mask); + migrate_irq_remapped_level_desc(desc); + return; + } + + migrate_ioapic_irq_desc(desc, mask); +} +static void set_ir_ioapic_affinity_irq(unsigned int irq, + const struct cpumask *mask) +{ + struct irq_desc *desc = irq_to_desc(irq); + + set_ir_ioapic_affinity_irq_desc(desc, mask); +} +#endif + +asmlinkage void smp_irq_move_cleanup_interrupt(void) +{ + unsigned vector, me; + + ack_APIC_irq(); + exit_idle(); + irq_enter(); + + me = smp_processor_id(); + for (vector = FIRST_EXTERNAL_VECTOR; vector < NR_VECTORS; vector++) { + unsigned int irq; + struct irq_desc *desc; + struct irq_cfg *cfg; + irq = __get_cpu_var(vector_irq)[vector]; + + if (irq == -1) + continue; + + desc = irq_to_desc(irq); + if (!desc) + continue; + + cfg = irq_cfg(irq); + spin_lock(&desc->lock); + if (!cfg->move_cleanup_count) + goto unlock; + + if (vector == cfg->vector && cpumask_test_cpu(me, cfg->domain)) + goto unlock; + + __get_cpu_var(vector_irq)[vector] = -1; + cfg->move_cleanup_count--; +unlock: + spin_unlock(&desc->lock); + } + + irq_exit(); +} + +static void irq_complete_move(struct irq_desc **descp) +{ + struct irq_desc *desc = *descp; + struct irq_cfg *cfg = desc->chip_data; + unsigned vector, me; + + if (likely(!cfg->move_in_progress)) { +#ifdef CONFIG_NUMA_MIGRATE_IRQ_DESC + if (likely(!cfg->move_desc_pending)) + return; + + /* domain has not changed, but affinity did */ + me = smp_processor_id(); + if (cpumask_test_cpu(me, desc->affinity)) { + *descp = desc = move_irq_desc(desc, me); + /* get the new one */ + cfg = desc->chip_data; + cfg->move_desc_pending = 0; + } +#endif + return; + } + + vector = ~get_irq_regs()->orig_ax; + me = smp_processor_id(); + + if (vector == cfg->vector && cpumask_test_cpu(me, cfg->domain)) { +#ifdef CONFIG_NUMA_MIGRATE_IRQ_DESC + *descp = desc = move_irq_desc(desc, me); + /* get the new one */ + cfg = desc->chip_data; +#endif + send_cleanup_vector(cfg); + } +} +#else +static inline void irq_complete_move(struct irq_desc **descp) {} +#endif + +#ifdef CONFIG_INTR_REMAP +static void ack_x2apic_level(unsigned int irq) +{ + ack_x2APIC_irq(); +} + +static void ack_x2apic_edge(unsigned int irq) +{ + ack_x2APIC_irq(); +} + +#endif + +static void ack_apic_edge(unsigned int irq) +{ + struct irq_desc *desc = irq_to_desc(irq); + + irq_complete_move(&desc); + move_native_irq(irq); + ack_APIC_irq(); +} + +atomic_t irq_mis_count; + +static void ack_apic_level(unsigned int irq) +{ + struct irq_desc *desc = irq_to_desc(irq); + +#ifdef CONFIG_X86_32 + unsigned long v; + int i; +#endif + struct irq_cfg *cfg; + int do_unmask_irq = 0; + + irq_complete_move(&desc); +#ifdef CONFIG_GENERIC_PENDING_IRQ + /* If we are moving the irq we need to mask it */ + if (unlikely(desc->status & IRQ_MOVE_PENDING)) { + do_unmask_irq = 1; + mask_IO_APIC_irq_desc(desc); + } +#endif + +#ifdef CONFIG_X86_32 + /* + * It appears there is an erratum which affects at least version 0x11 + * of I/O APIC (that's the 82093AA and cores integrated into various + * chipsets). Under certain conditions a level-triggered interrupt is + * erroneously delivered as edge-triggered one but the respective IRR + * bit gets set nevertheless. As a result the I/O unit expects an EOI + * message but it will never arrive and further interrupts are blocked + * from the source. The exact reason is so far unknown, but the + * phenomenon was observed when two consecutive interrupt requests + * from a given source get delivered to the same CPU and the source is + * temporarily disabled in between. + * + * A workaround is to simulate an EOI message manually. We achieve it + * by setting the trigger mode to edge and then to level when the edge + * trigger mode gets detected in the TMR of a local APIC for a + * level-triggered interrupt. We mask the source for the time of the + * operation to prevent an edge-triggered interrupt escaping meanwhile. + * The idea is from Manfred Spraul. --macro + */ + cfg = desc->chip_data; + i = cfg->vector; + + v = apic_read(APIC_TMR + ((i & ~0x1f) >> 1)); +#endif + + /* + * We must acknowledge the irq before we move it or the acknowledge will + * not propagate properly. + */ + ack_APIC_irq(); + + /* Now we can move and renable the irq */ + if (unlikely(do_unmask_irq)) { + /* Only migrate the irq if the ack has been received. + * + * On rare occasions the broadcast level triggered ack gets + * delayed going to ioapics, and if we reprogram the + * vector while Remote IRR is still set the irq will never + * fire again. + * + * To prevent this scenario we read the Remote IRR bit + * of the ioapic. This has two effects. + * - On any sane system the read of the ioapic will + * flush writes (and acks) going to the ioapic from + * this cpu. + * - We get to see if the ACK has actually been delivered. + * + * Based on failed experiments of reprogramming the + * ioapic entry from outside of irq context starting + * with masking the ioapic entry and then polling until + * Remote IRR was clear before reprogramming the + * ioapic I don't trust the Remote IRR bit to be + * completey accurate. + * + * However there appears to be no other way to plug + * this race, so if the Remote IRR bit is not + * accurate and is causing problems then it is a hardware bug + * and you can go talk to the chipset vendor about it. + */ + cfg = desc->chip_data; + if (!io_apic_level_ack_pending(cfg)) + move_masked_irq(irq); + unmask_IO_APIC_irq_desc(desc); + } + +#ifdef CONFIG_X86_32 + if (!(v & (1 << (i & 0x1f)))) { + atomic_inc(&irq_mis_count); + spin_lock(&ioapic_lock); + __mask_and_edge_IO_APIC_irq(cfg); + __unmask_and_level_IO_APIC_irq(cfg); + spin_unlock(&ioapic_lock); + } +#endif +} + +static struct irq_chip ioapic_chip __read_mostly = { + .name = "IO-APIC", + .startup = startup_ioapic_irq, + .mask = mask_IO_APIC_irq, + .unmask = unmask_IO_APIC_irq, + .ack = ack_apic_edge, + .eoi = ack_apic_level, +#ifdef CONFIG_SMP + .set_affinity = set_ioapic_affinity_irq, +#endif + .retrigger = ioapic_retrigger_irq, +}; + +#ifdef CONFIG_INTR_REMAP +static struct irq_chip ir_ioapic_chip __read_mostly = { + .name = "IR-IO-APIC", + .startup = startup_ioapic_irq, + .mask = mask_IO_APIC_irq, + .unmask = unmask_IO_APIC_irq, + .ack = ack_x2apic_edge, + .eoi = ack_x2apic_level, +#ifdef CONFIG_SMP + .set_affinity = set_ir_ioapic_affinity_irq, +#endif + .retrigger = ioapic_retrigger_irq, +}; +#endif + +static inline void init_IO_APIC_traps(void) +{ + int irq; + struct irq_desc *desc; + struct irq_cfg *cfg; + + /* + * NOTE! The local APIC isn't very good at handling + * multiple interrupts at the same interrupt level. + * As the interrupt level is determined by taking the + * vector number and shifting that right by 4, we + * want to spread these out a bit so that they don't + * all fall in the same interrupt level. + * + * Also, we've got to be careful not to trash gate + * 0x80, because int 0x80 is hm, kind of importantish. ;) + */ + for_each_irq_desc(irq, desc) { + cfg = desc->chip_data; + if (IO_APIC_IRQ(irq) && cfg && !cfg->vector) { + /* + * Hmm.. We don't have an entry for this, + * so default to an old-fashioned 8259 + * interrupt if we can.. + */ + if (irq < NR_IRQS_LEGACY) + make_8259A_irq(irq); + else + /* Strange. Oh, well.. */ + desc->chip = &no_irq_chip; + } + } +} + +/* + * The local APIC irq-chip implementation: + */ + +static void mask_lapic_irq(unsigned int irq) +{ + unsigned long v; + + v = apic_read(APIC_LVT0); + apic_write(APIC_LVT0, v | APIC_LVT_MASKED); +} + +static void unmask_lapic_irq(unsigned int irq) +{ + unsigned long v; + + v = apic_read(APIC_LVT0); + apic_write(APIC_LVT0, v & ~APIC_LVT_MASKED); +} + +static void ack_lapic_irq(unsigned int irq) +{ + ack_APIC_irq(); +} + +static struct irq_chip lapic_chip __read_mostly = { + .name = "local-APIC", + .mask = mask_lapic_irq, + .unmask = unmask_lapic_irq, + .ack = ack_lapic_irq, +}; + +static void lapic_register_intr(int irq, struct irq_desc *desc) +{ + desc->status &= ~IRQ_LEVEL; + set_irq_chip_and_handler_name(irq, &lapic_chip, handle_edge_irq, + "edge"); +} + +static void __init setup_nmi(void) +{ + /* + * Dirty trick to enable the NMI watchdog ... + * We put the 8259A master into AEOI mode and + * unmask on all local APICs LVT0 as NMI. + * + * The idea to use the 8259A in AEOI mode ('8259A Virtual Wire') + * is from Maciej W. Rozycki - so we do not have to EOI from + * the NMI handler or the timer interrupt. + */ + apic_printk(APIC_VERBOSE, KERN_INFO "activating NMI Watchdog ..."); + + enable_NMI_through_LVT0(); + + apic_printk(APIC_VERBOSE, " done.\n"); +} + +/* + * This looks a bit hackish but it's about the only one way of sending + * a few INTA cycles to 8259As and any associated glue logic. ICR does + * not support the ExtINT mode, unfortunately. We need to send these + * cycles as some i82489DX-based boards have glue logic that keeps the + * 8259A interrupt line asserted until INTA. --macro + */ +static inline void __init unlock_ExtINT_logic(void) +{ + int apic, pin, i; + struct IO_APIC_route_entry entry0, entry1; + unsigned char save_control, save_freq_select; + + pin = find_isa_irq_pin(8, mp_INT); + if (pin == -1) { + WARN_ON_ONCE(1); + return; + } + apic = find_isa_irq_apic(8, mp_INT); + if (apic == -1) { + WARN_ON_ONCE(1); + return; + } + + entry0 = ioapic_read_entry(apic, pin); + clear_IO_APIC_pin(apic, pin); + + memset(&entry1, 0, sizeof(entry1)); + + entry1.dest_mode = 0; /* physical delivery */ + entry1.mask = 0; /* unmask IRQ now */ + entry1.dest = hard_smp_processor_id(); + entry1.delivery_mode = dest_ExtINT; + entry1.polarity = entry0.polarity; + entry1.trigger = 0; + entry1.vector = 0; + + ioapic_write_entry(apic, pin, entry1); + + save_control = CMOS_READ(RTC_CONTROL); + save_freq_select = CMOS_READ(RTC_FREQ_SELECT); + CMOS_WRITE((save_freq_select & ~RTC_RATE_SELECT) | 0x6, + RTC_FREQ_SELECT); + CMOS_WRITE(save_control | RTC_PIE, RTC_CONTROL); + + i = 100; + while (i-- > 0) { + mdelay(10); + if ((CMOS_READ(RTC_INTR_FLAGS) & RTC_PF) == RTC_PF) + i -= 10; + } + + CMOS_WRITE(save_control, RTC_CONTROL); + CMOS_WRITE(save_freq_select, RTC_FREQ_SELECT); + clear_IO_APIC_pin(apic, pin); + + ioapic_write_entry(apic, pin, entry0); +} + +static int disable_timer_pin_1 __initdata; +/* Actually the next is obsolete, but keep it for paranoid reasons -AK */ +static int __init disable_timer_pin_setup(char *arg) +{ + disable_timer_pin_1 = 1; + return 0; +} +early_param("disable_timer_pin_1", disable_timer_pin_setup); + +int timer_through_8259 __initdata; + +/* + * This code may look a bit paranoid, but it's supposed to cooperate with + * a wide range of boards and BIOS bugs. Fortunately only the timer IRQ + * is so screwy. Thanks to Brian Perkins for testing/hacking this beast + * fanatically on his truly buggy board. + * + * FIXME: really need to revamp this for all platforms. + */ +static inline void __init check_timer(void) +{ + struct irq_desc *desc = irq_to_desc(0); + struct irq_cfg *cfg = desc->chip_data; + int cpu = boot_cpu_id; + int apic1, pin1, apic2, pin2; + unsigned long flags; + int no_pin1 = 0; + + local_irq_save(flags); + + /* + * get/set the timer IRQ vector: + */ + disable_8259A_irq(0); + assign_irq_vector(0, cfg, apic->target_cpus()); + + /* + * As IRQ0 is to be enabled in the 8259A, the virtual + * wire has to be disabled in the local APIC. Also + * timer interrupts need to be acknowledged manually in + * the 8259A for the i82489DX when using the NMI + * watchdog as that APIC treats NMIs as level-triggered. + * The AEOI mode will finish them in the 8259A + * automatically. + */ + apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT); + init_8259A(1); +#ifdef CONFIG_X86_32 + { + unsigned int ver; + + ver = apic_read(APIC_LVR); + ver = GET_APIC_VERSION(ver); + timer_ack = (nmi_watchdog == NMI_IO_APIC && !APIC_INTEGRATED(ver)); + } +#endif + + pin1 = find_isa_irq_pin(0, mp_INT); + apic1 = find_isa_irq_apic(0, mp_INT); + pin2 = ioapic_i8259.pin; + apic2 = ioapic_i8259.apic; + + apic_printk(APIC_QUIET, KERN_INFO "..TIMER: vector=0x%02X " + "apic1=%d pin1=%d apic2=%d pin2=%d\n", + cfg->vector, apic1, pin1, apic2, pin2); + + /* + * Some BIOS writers are clueless and report the ExtINTA + * I/O APIC input from the cascaded 8259A as the timer + * interrupt input. So just in case, if only one pin + * was found above, try it both directly and through the + * 8259A. + */ + if (pin1 == -1) { +#ifdef CONFIG_INTR_REMAP + if (intr_remapping_enabled) + panic("BIOS bug: timer not connected to IO-APIC"); +#endif + pin1 = pin2; + apic1 = apic2; + no_pin1 = 1; + } else if (pin2 == -1) { + pin2 = pin1; + apic2 = apic1; + } + + if (pin1 != -1) { + /* + * Ok, does IRQ0 through the IOAPIC work? + */ + if (no_pin1) { + add_pin_to_irq_cpu(cfg, cpu, apic1, pin1); + setup_timer_IRQ0_pin(apic1, pin1, cfg->vector); + } else { + /* for edge trigger, setup_IO_APIC_irq already + * leave it unmasked. + * so only need to unmask if it is level-trigger + * do we really have level trigger timer? + */ + int idx; + idx = find_irq_entry(apic1, pin1, mp_INT); + if (idx != -1 && irq_trigger(idx)) + unmask_IO_APIC_irq_desc(desc); + } + if (timer_irq_works()) { + if (nmi_watchdog == NMI_IO_APIC) { + setup_nmi(); + enable_8259A_irq(0); + } + if (disable_timer_pin_1 > 0) + clear_IO_APIC_pin(0, pin1); + goto out; + } +#ifdef CONFIG_INTR_REMAP + if (intr_remapping_enabled) + panic("timer doesn't work through Interrupt-remapped IO-APIC"); +#endif + local_irq_disable(); + clear_IO_APIC_pin(apic1, pin1); + if (!no_pin1) + apic_printk(APIC_QUIET, KERN_ERR "..MP-BIOS bug: " + "8254 timer not connected to IO-APIC\n"); + + apic_printk(APIC_QUIET, KERN_INFO "...trying to set up timer " + "(IRQ0) through the 8259A ...\n"); + apic_printk(APIC_QUIET, KERN_INFO + "..... (found apic %d pin %d) ...\n", apic2, pin2); + /* + * legacy devices should be connected to IO APIC #0 + */ + replace_pin_at_irq_cpu(cfg, cpu, apic1, pin1, apic2, pin2); + setup_timer_IRQ0_pin(apic2, pin2, cfg->vector); + enable_8259A_irq(0); + if (timer_irq_works()) { + apic_printk(APIC_QUIET, KERN_INFO "....... works.\n"); + timer_through_8259 = 1; + if (nmi_watchdog == NMI_IO_APIC) { + disable_8259A_irq(0); + setup_nmi(); + enable_8259A_irq(0); + } + goto out; + } + /* + * Cleanup, just in case ... + */ + local_irq_disable(); + disable_8259A_irq(0); + clear_IO_APIC_pin(apic2, pin2); + apic_printk(APIC_QUIET, KERN_INFO "....... failed.\n"); + } + + if (nmi_watchdog == NMI_IO_APIC) { + apic_printk(APIC_QUIET, KERN_WARNING "timer doesn't work " + "through the IO-APIC - disabling NMI Watchdog!\n"); + nmi_watchdog = NMI_NONE; + } +#ifdef CONFIG_X86_32 + timer_ack = 0; +#endif + + apic_printk(APIC_QUIET, KERN_INFO + "...trying to set up timer as Virtual Wire IRQ...\n"); + + lapic_register_intr(0, desc); + apic_write(APIC_LVT0, APIC_DM_FIXED | cfg->vector); /* Fixed mode */ + enable_8259A_irq(0); + + if (timer_irq_works()) { + apic_printk(APIC_QUIET, KERN_INFO "..... works.\n"); + goto out; + } + local_irq_disable(); + disable_8259A_irq(0); + apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_FIXED | cfg->vector); + apic_printk(APIC_QUIET, KERN_INFO "..... failed.\n"); + + apic_printk(APIC_QUIET, KERN_INFO + "...trying to set up timer as ExtINT IRQ...\n"); + + init_8259A(0); + make_8259A_irq(0); + apic_write(APIC_LVT0, APIC_DM_EXTINT); + + unlock_ExtINT_logic(); + + if (timer_irq_works()) { + apic_printk(APIC_QUIET, KERN_INFO "..... works.\n"); + goto out; + } + local_irq_disable(); + apic_printk(APIC_QUIET, KERN_INFO "..... failed :(.\n"); + panic("IO-APIC + timer doesn't work! Boot with apic=debug and send a " + "report. Then try booting with the 'noapic' option.\n"); +out: + local_irq_restore(flags); +} + +/* + * Traditionally ISA IRQ2 is the cascade IRQ, and is not available + * to devices. However there may be an I/O APIC pin available for + * this interrupt regardless. The pin may be left unconnected, but + * typically it will be reused as an ExtINT cascade interrupt for + * the master 8259A. In the MPS case such a pin will normally be + * reported as an ExtINT interrupt in the MP table. With ACPI + * there is no provision for ExtINT interrupts, and in the absence + * of an override it would be treated as an ordinary ISA I/O APIC + * interrupt, that is edge-triggered and unmasked by default. We + * used to do this, but it caused problems on some systems because + * of the NMI watchdog and sometimes IRQ0 of the 8254 timer using + * the same ExtINT cascade interrupt to drive the local APIC of the + * bootstrap processor. Therefore we refrain from routing IRQ2 to + * the I/O APIC in all cases now. No actual device should request + * it anyway. --macro + */ +#define PIC_IRQS (1 << PIC_CASCADE_IR) + +void __init setup_IO_APIC(void) +{ + + /* + * calling enable_IO_APIC() is moved to setup_local_APIC for BP + */ + + io_apic_irqs = ~PIC_IRQS; + + apic_printk(APIC_VERBOSE, "ENABLING IO-APIC IRQs\n"); + /* + * Set up IO-APIC IRQ routing. + */ +#ifdef CONFIG_X86_32 + if (!acpi_ioapic) + setup_ioapic_ids_from_mpc(); +#endif + sync_Arb_IDs(); + setup_IO_APIC_irqs(); + init_IO_APIC_traps(); + check_timer(); +} + +/* + * Called after all the initialization is done. If we didnt find any + * APIC bugs then we can allow the modify fast path + */ + +static int __init io_apic_bug_finalize(void) +{ + if (sis_apic_bug == -1) + sis_apic_bug = 0; + return 0; +} + +late_initcall(io_apic_bug_finalize); + +struct sysfs_ioapic_data { + struct sys_device dev; + struct IO_APIC_route_entry entry[0]; +}; +static struct sysfs_ioapic_data * mp_ioapic_data[MAX_IO_APICS]; + +static int ioapic_suspend(struct sys_device *dev, pm_message_t state) +{ + struct IO_APIC_route_entry *entry; + struct sysfs_ioapic_data *data; + int i; + + data = container_of(dev, struct sysfs_ioapic_data, dev); + entry = data->entry; + for (i = 0; i < nr_ioapic_registers[dev->id]; i ++, entry ++ ) + *entry = ioapic_read_entry(dev->id, i); + + return 0; +} + +static int ioapic_resume(struct sys_device *dev) +{ + struct IO_APIC_route_entry *entry; + struct sysfs_ioapic_data *data; + unsigned long flags; + union IO_APIC_reg_00 reg_00; + int i; + + data = container_of(dev, struct sysfs_ioapic_data, dev); + entry = data->entry; + + spin_lock_irqsave(&ioapic_lock, flags); + reg_00.raw = io_apic_read(dev->id, 0); + if (reg_00.bits.ID != mp_ioapics[dev->id].apicid) { + reg_00.bits.ID = mp_ioapics[dev->id].apicid; + io_apic_write(dev->id, 0, reg_00.raw); + } + spin_unlock_irqrestore(&ioapic_lock, flags); + for (i = 0; i < nr_ioapic_registers[dev->id]; i++) + ioapic_write_entry(dev->id, i, entry[i]); + + return 0; +} + +static struct sysdev_class ioapic_sysdev_class = { + .name = "ioapic", + .suspend = ioapic_suspend, + .resume = ioapic_resume, +}; + +static int __init ioapic_init_sysfs(void) +{ + struct sys_device * dev; + int i, size, error; + + error = sysdev_class_register(&ioapic_sysdev_class); + if (error) + return error; + + for (i = 0; i < nr_ioapics; i++ ) { + size = sizeof(struct sys_device) + nr_ioapic_registers[i] + * sizeof(struct IO_APIC_route_entry); + mp_ioapic_data[i] = kzalloc(size, GFP_KERNEL); + if (!mp_ioapic_data[i]) { + printk(KERN_ERR "Can't suspend/resume IOAPIC %d\n", i); + continue; + } + dev = &mp_ioapic_data[i]->dev; + dev->id = i; + dev->cls = &ioapic_sysdev_class; + error = sysdev_register(dev); + if (error) { + kfree(mp_ioapic_data[i]); + mp_ioapic_data[i] = NULL; + printk(KERN_ERR "Can't suspend/resume IOAPIC %d\n", i); + continue; + } + } + + return 0; +} + +device_initcall(ioapic_init_sysfs); + +static int nr_irqs_gsi = NR_IRQS_LEGACY; +/* + * Dynamic irq allocate and deallocation + */ +unsigned int create_irq_nr(unsigned int irq_want) +{ + /* Allocate an unused irq */ + unsigned int irq; + unsigned int new; + unsigned long flags; + struct irq_cfg *cfg_new = NULL; + int cpu = boot_cpu_id; + struct irq_desc *desc_new = NULL; + + irq = 0; + if (irq_want < nr_irqs_gsi) + irq_want = nr_irqs_gsi; + + spin_lock_irqsave(&vector_lock, flags); + for (new = irq_want; new < nr_irqs; new++) { + desc_new = irq_to_desc_alloc_cpu(new, cpu); + if (!desc_new) { + printk(KERN_INFO "can not get irq_desc for %d\n", new); + continue; + } + cfg_new = desc_new->chip_data; + + if (cfg_new->vector != 0) + continue; + if (__assign_irq_vector(new, cfg_new, apic->target_cpus()) == 0) + irq = new; + break; + } + spin_unlock_irqrestore(&vector_lock, flags); + + if (irq > 0) { + dynamic_irq_init(irq); + /* restore it, in case dynamic_irq_init clear it */ + if (desc_new) + desc_new->chip_data = cfg_new; + } + return irq; +} + +int create_irq(void) +{ + unsigned int irq_want; + int irq; + + irq_want = nr_irqs_gsi; + irq = create_irq_nr(irq_want); + + if (irq == 0) + irq = -1; + + return irq; +} + +void destroy_irq(unsigned int irq) +{ + unsigned long flags; + struct irq_cfg *cfg; + struct irq_desc *desc; + + /* store it, in case dynamic_irq_cleanup clear it */ + desc = irq_to_desc(irq); + cfg = desc->chip_data; + dynamic_irq_cleanup(irq); + /* connect back irq_cfg */ + if (desc) + desc->chip_data = cfg; + +#ifdef CONFIG_INTR_REMAP + free_irte(irq); +#endif + spin_lock_irqsave(&vector_lock, flags); + __clear_irq_vector(irq, cfg); + spin_unlock_irqrestore(&vector_lock, flags); +} + +/* + * MSI message composition + */ +#ifdef CONFIG_PCI_MSI +static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_msg *msg) +{ + struct irq_cfg *cfg; + int err; + unsigned dest; + + if (disable_apic) + return -ENXIO; + + cfg = irq_cfg(irq); + err = assign_irq_vector(irq, cfg, apic->target_cpus()); + if (err) + return err; + + dest = apic->cpu_mask_to_apicid_and(cfg->domain, apic->target_cpus()); + +#ifdef CONFIG_INTR_REMAP + if (irq_remapped(irq)) { + struct irte irte; + int ir_index; + u16 sub_handle; + + ir_index = map_irq_to_irte_handle(irq, &sub_handle); + BUG_ON(ir_index == -1); + + memset (&irte, 0, sizeof(irte)); + + irte.present = 1; + irte.dst_mode = apic->irq_dest_mode; + irte.trigger_mode = 0; /* edge */ + irte.dlvry_mode = apic->irq_delivery_mode; + irte.vector = cfg->vector; + irte.dest_id = IRTE_DEST(dest); + + modify_irte(irq, &irte); + + msg->address_hi = MSI_ADDR_BASE_HI; + msg->data = sub_handle; + msg->address_lo = MSI_ADDR_BASE_LO | MSI_ADDR_IR_EXT_INT | + MSI_ADDR_IR_SHV | + MSI_ADDR_IR_INDEX1(ir_index) | + MSI_ADDR_IR_INDEX2(ir_index); + } else +#endif + { + msg->address_hi = MSI_ADDR_BASE_HI; + msg->address_lo = + MSI_ADDR_BASE_LO | + ((apic->irq_dest_mode == 0) ? + MSI_ADDR_DEST_MODE_PHYSICAL: + MSI_ADDR_DEST_MODE_LOGICAL) | + ((apic->irq_delivery_mode != dest_LowestPrio) ? + MSI_ADDR_REDIRECTION_CPU: + MSI_ADDR_REDIRECTION_LOWPRI) | + MSI_ADDR_DEST_ID(dest); + + msg->data = + MSI_DATA_TRIGGER_EDGE | + MSI_DATA_LEVEL_ASSERT | + ((apic->irq_delivery_mode != dest_LowestPrio) ? + MSI_DATA_DELIVERY_FIXED: + MSI_DATA_DELIVERY_LOWPRI) | + MSI_DATA_VECTOR(cfg->vector); + } + return err; +} + +#ifdef CONFIG_SMP +static void set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask) +{ + struct irq_desc *desc = irq_to_desc(irq); + struct irq_cfg *cfg; + struct msi_msg msg; + unsigned int dest; + + dest = set_desc_affinity(desc, mask); + if (dest == BAD_APICID) + return; + + cfg = desc->chip_data; + + read_msi_msg_desc(desc, &msg); + + msg.data &= ~MSI_DATA_VECTOR_MASK; + msg.data |= MSI_DATA_VECTOR(cfg->vector); + msg.address_lo &= ~MSI_ADDR_DEST_ID_MASK; + msg.address_lo |= MSI_ADDR_DEST_ID(dest); + + write_msi_msg_desc(desc, &msg); +} +#ifdef CONFIG_INTR_REMAP +/* + * Migrate the MSI irq to another cpumask. This migration is + * done in the process context using interrupt-remapping hardware. + */ +static void +ir_set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask) +{ + struct irq_desc *desc = irq_to_desc(irq); + struct irq_cfg *cfg = desc->chip_data; + unsigned int dest; + struct irte irte; + + if (get_irte(irq, &irte)) + return; + + dest = set_desc_affinity(desc, mask); + if (dest == BAD_APICID) + return; + + irte.vector = cfg->vector; + irte.dest_id = IRTE_DEST(dest); + + /* + * atomically update the IRTE with the new destination and vector. + */ + modify_irte(irq, &irte); + + /* + * After this point, all the interrupts will start arriving + * at the new destination. So, time to cleanup the previous + * vector allocation. + */ + if (cfg->move_in_progress) + send_cleanup_vector(cfg); +} + +#endif +#endif /* CONFIG_SMP */ + +/* + * IRQ Chip for MSI PCI/PCI-X/PCI-Express Devices, + * which implement the MSI or MSI-X Capability Structure. + */ +static struct irq_chip msi_chip = { + .name = "PCI-MSI", + .unmask = unmask_msi_irq, + .mask = mask_msi_irq, + .ack = ack_apic_edge, +#ifdef CONFIG_SMP + .set_affinity = set_msi_irq_affinity, +#endif + .retrigger = ioapic_retrigger_irq, +}; + +#ifdef CONFIG_INTR_REMAP +static struct irq_chip msi_ir_chip = { + .name = "IR-PCI-MSI", + .unmask = unmask_msi_irq, + .mask = mask_msi_irq, + .ack = ack_x2apic_edge, +#ifdef CONFIG_SMP + .set_affinity = ir_set_msi_irq_affinity, +#endif + .retrigger = ioapic_retrigger_irq, +}; + +/* + * Map the PCI dev to the corresponding remapping hardware unit + * and allocate 'nvec' consecutive interrupt-remapping table entries + * in it. + */ +static int msi_alloc_irte(struct pci_dev *dev, int irq, int nvec) +{ + struct intel_iommu *iommu; + int index; + + iommu = map_dev_to_ir(dev); + if (!iommu) { + printk(KERN_ERR + "Unable to map PCI %s to iommu\n", pci_name(dev)); + return -ENOENT; + } + + index = alloc_irte(iommu, irq, nvec); + if (index < 0) { + printk(KERN_ERR + "Unable to allocate %d IRTE for PCI %s\n", nvec, + pci_name(dev)); + return -ENOSPC; + } + return index; +} +#endif + +static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq) +{ + int ret; + struct msi_msg msg; + + ret = msi_compose_msg(dev, irq, &msg); + if (ret < 0) + return ret; + + set_irq_msi(irq, msidesc); + write_msi_msg(irq, &msg); + +#ifdef CONFIG_INTR_REMAP + if (irq_remapped(irq)) { + struct irq_desc *desc = irq_to_desc(irq); + /* + * irq migration in process context + */ + desc->status |= IRQ_MOVE_PCNTXT; + set_irq_chip_and_handler_name(irq, &msi_ir_chip, handle_edge_irq, "edge"); + } else +#endif + set_irq_chip_and_handler_name(irq, &msi_chip, handle_edge_irq, "edge"); + + dev_printk(KERN_DEBUG, &dev->dev, "irq %d for MSI/MSI-X\n", irq); + + return 0; +} + +int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) +{ + unsigned int irq; + int ret, sub_handle; + struct msi_desc *msidesc; + unsigned int irq_want; + +#ifdef CONFIG_INTR_REMAP + struct intel_iommu *iommu = 0; + int index = 0; +#endif + + irq_want = nr_irqs_gsi; + sub_handle = 0; + list_for_each_entry(msidesc, &dev->msi_list, list) { + irq = create_irq_nr(irq_want); + if (irq == 0) + return -1; + irq_want = irq + 1; +#ifdef CONFIG_INTR_REMAP + if (!intr_remapping_enabled) + goto no_ir; + + if (!sub_handle) { + /* + * allocate the consecutive block of IRTE's + * for 'nvec' + */ + index = msi_alloc_irte(dev, irq, nvec); + if (index < 0) { + ret = index; + goto error; + } + } else { + iommu = map_dev_to_ir(dev); + if (!iommu) { + ret = -ENOENT; + goto error; + } + /* + * setup the mapping between the irq and the IRTE + * base index, the sub_handle pointing to the + * appropriate interrupt remap table entry. + */ + set_irte_irq(irq, iommu, index, sub_handle); + } +no_ir: +#endif + ret = setup_msi_irq(dev, msidesc, irq); + if (ret < 0) + goto error; + sub_handle++; + } + return 0; + +error: + destroy_irq(irq); + return ret; +} + +void arch_teardown_msi_irq(unsigned int irq) +{ + destroy_irq(irq); +} + +#ifdef CONFIG_DMAR +#ifdef CONFIG_SMP +static void dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask) +{ + struct irq_desc *desc = irq_to_desc(irq); + struct irq_cfg *cfg; + struct msi_msg msg; + unsigned int dest; + + dest = set_desc_affinity(desc, mask); + if (dest == BAD_APICID) + return; + + cfg = desc->chip_data; + + dmar_msi_read(irq, &msg); + + msg.data &= ~MSI_DATA_VECTOR_MASK; + msg.data |= MSI_DATA_VECTOR(cfg->vector); + msg.address_lo &= ~MSI_ADDR_DEST_ID_MASK; + msg.address_lo |= MSI_ADDR_DEST_ID(dest); + + dmar_msi_write(irq, &msg); +} + +#endif /* CONFIG_SMP */ + +struct irq_chip dmar_msi_type = { + .name = "DMAR_MSI", + .unmask = dmar_msi_unmask, + .mask = dmar_msi_mask, + .ack = ack_apic_edge, +#ifdef CONFIG_SMP + .set_affinity = dmar_msi_set_affinity, +#endif + .retrigger = ioapic_retrigger_irq, +}; + +int arch_setup_dmar_msi(unsigned int irq) +{ + int ret; + struct msi_msg msg; + + ret = msi_compose_msg(NULL, irq, &msg); + if (ret < 0) + return ret; + dmar_msi_write(irq, &msg); + set_irq_chip_and_handler_name(irq, &dmar_msi_type, handle_edge_irq, + "edge"); + return 0; +} +#endif + +#ifdef CONFIG_HPET_TIMER + +#ifdef CONFIG_SMP +static void hpet_msi_set_affinity(unsigned int irq, const struct cpumask *mask) +{ + struct irq_desc *desc = irq_to_desc(irq); + struct irq_cfg *cfg; + struct msi_msg msg; + unsigned int dest; + + dest = set_desc_affinity(desc, mask); + if (dest == BAD_APICID) + return; + + cfg = desc->chip_data; + + hpet_msi_read(irq, &msg); + + msg.data &= ~MSI_DATA_VECTOR_MASK; + msg.data |= MSI_DATA_VECTOR(cfg->vector); + msg.address_lo &= ~MSI_ADDR_DEST_ID_MASK; + msg.address_lo |= MSI_ADDR_DEST_ID(dest); + + hpet_msi_write(irq, &msg); +} + +#endif /* CONFIG_SMP */ + +struct irq_chip hpet_msi_type = { + .name = "HPET_MSI", + .unmask = hpet_msi_unmask, + .mask = hpet_msi_mask, + .ack = ack_apic_edge, +#ifdef CONFIG_SMP + .set_affinity = hpet_msi_set_affinity, +#endif + .retrigger = ioapic_retrigger_irq, +}; + +int arch_setup_hpet_msi(unsigned int irq) +{ + int ret; + struct msi_msg msg; + + ret = msi_compose_msg(NULL, irq, &msg); + if (ret < 0) + return ret; + + hpet_msi_write(irq, &msg); + set_irq_chip_and_handler_name(irq, &hpet_msi_type, handle_edge_irq, + "edge"); + + return 0; +} +#endif + +#endif /* CONFIG_PCI_MSI */ +/* + * Hypertransport interrupt support + */ +#ifdef CONFIG_HT_IRQ + +#ifdef CONFIG_SMP + +static void target_ht_irq(unsigned int irq, unsigned int dest, u8 vector) +{ + struct ht_irq_msg msg; + fetch_ht_irq_msg(irq, &msg); + + msg.address_lo &= ~(HT_IRQ_LOW_VECTOR_MASK | HT_IRQ_LOW_DEST_ID_MASK); + msg.address_hi &= ~(HT_IRQ_HIGH_DEST_ID_MASK); + + msg.address_lo |= HT_IRQ_LOW_VECTOR(vector) | HT_IRQ_LOW_DEST_ID(dest); + msg.address_hi |= HT_IRQ_HIGH_DEST_ID(dest); + + write_ht_irq_msg(irq, &msg); +} + +static void set_ht_irq_affinity(unsigned int irq, const struct cpumask *mask) +{ + struct irq_desc *desc = irq_to_desc(irq); + struct irq_cfg *cfg; + unsigned int dest; + + dest = set_desc_affinity(desc, mask); + if (dest == BAD_APICID) + return; + + cfg = desc->chip_data; + + target_ht_irq(irq, dest, cfg->vector); +} + +#endif + +static struct irq_chip ht_irq_chip = { + .name = "PCI-HT", + .mask = mask_ht_irq, + .unmask = unmask_ht_irq, + .ack = ack_apic_edge, +#ifdef CONFIG_SMP + .set_affinity = set_ht_irq_affinity, +#endif + .retrigger = ioapic_retrigger_irq, +}; + +int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) +{ + struct irq_cfg *cfg; + int err; + + if (disable_apic) + return -ENXIO; + + cfg = irq_cfg(irq); + err = assign_irq_vector(irq, cfg, apic->target_cpus()); + if (!err) { + struct ht_irq_msg msg; + unsigned dest; + + dest = apic->cpu_mask_to_apicid_and(cfg->domain, + apic->target_cpus()); + + msg.address_hi = HT_IRQ_HIGH_DEST_ID(dest); + + msg.address_lo = + HT_IRQ_LOW_BASE | + HT_IRQ_LOW_DEST_ID(dest) | + HT_IRQ_LOW_VECTOR(cfg->vector) | + ((apic->irq_dest_mode == 0) ? + HT_IRQ_LOW_DM_PHYSICAL : + HT_IRQ_LOW_DM_LOGICAL) | + HT_IRQ_LOW_RQEOI_EDGE | + ((apic->irq_delivery_mode != dest_LowestPrio) ? + HT_IRQ_LOW_MT_FIXED : + HT_IRQ_LOW_MT_ARBITRATED) | + HT_IRQ_LOW_IRQ_MASKED; + + write_ht_irq_msg(irq, &msg); + + set_irq_chip_and_handler_name(irq, &ht_irq_chip, + handle_edge_irq, "edge"); + + dev_printk(KERN_DEBUG, &dev->dev, "irq %d for HT\n", irq); + } + return err; +} +#endif /* CONFIG_HT_IRQ */ + +#ifdef CONFIG_X86_UV +/* + * Re-target the irq to the specified CPU and enable the specified MMR located + * on the specified blade to allow the sending of MSIs to the specified CPU. + */ +int arch_enable_uv_irq(char *irq_name, unsigned int irq, int cpu, int mmr_blade, + unsigned long mmr_offset) +{ + const struct cpumask *eligible_cpu = cpumask_of(cpu); + struct irq_cfg *cfg; + int mmr_pnode; + unsigned long mmr_value; + struct uv_IO_APIC_route_entry *entry; + unsigned long flags; + int err; + + cfg = irq_cfg(irq); + + err = assign_irq_vector(irq, cfg, eligible_cpu); + if (err != 0) + return err; + + spin_lock_irqsave(&vector_lock, flags); + set_irq_chip_and_handler_name(irq, &uv_irq_chip, handle_percpu_irq, + irq_name); + spin_unlock_irqrestore(&vector_lock, flags); + + mmr_value = 0; + entry = (struct uv_IO_APIC_route_entry *)&mmr_value; + BUG_ON(sizeof(struct uv_IO_APIC_route_entry) != sizeof(unsigned long)); + + entry->vector = cfg->vector; + entry->delivery_mode = apic->irq_delivery_mode; + entry->dest_mode = apic->irq_dest_mode; + entry->polarity = 0; + entry->trigger = 0; + entry->mask = 0; + entry->dest = apic->cpu_mask_to_apicid(eligible_cpu); + + mmr_pnode = uv_blade_to_pnode(mmr_blade); + uv_write_global_mmr64(mmr_pnode, mmr_offset, mmr_value); + + return irq; +} + +/* + * Disable the specified MMR located on the specified blade so that MSIs are + * longer allowed to be sent. + */ +void arch_disable_uv_irq(int mmr_blade, unsigned long mmr_offset) +{ + unsigned long mmr_value; + struct uv_IO_APIC_route_entry *entry; + int mmr_pnode; + + mmr_value = 0; + entry = (struct uv_IO_APIC_route_entry *)&mmr_value; + BUG_ON(sizeof(struct uv_IO_APIC_route_entry) != sizeof(unsigned long)); + + entry->mask = 1; + + mmr_pnode = uv_blade_to_pnode(mmr_blade); + uv_write_global_mmr64(mmr_pnode, mmr_offset, mmr_value); +} +#endif /* CONFIG_X86_64 */ + +int __init io_apic_get_redir_entries (int ioapic) +{ + union IO_APIC_reg_01 reg_01; + unsigned long flags; + + spin_lock_irqsave(&ioapic_lock, flags); + reg_01.raw = io_apic_read(ioapic, 1); + spin_unlock_irqrestore(&ioapic_lock, flags); + + return reg_01.bits.entries; +} + +void __init probe_nr_irqs_gsi(void) +{ + int nr = 0; + + nr = acpi_probe_gsi(); + if (nr > nr_irqs_gsi) { + nr_irqs_gsi = nr; + } else { + /* for acpi=off or acpi is not compiled in */ + int idx; + + nr = 0; + for (idx = 0; idx < nr_ioapics; idx++) + nr += io_apic_get_redir_entries(idx) + 1; + + if (nr > nr_irqs_gsi) + nr_irqs_gsi = nr; + } + + printk(KERN_DEBUG "nr_irqs_gsi: %d\n", nr_irqs_gsi); +} + +#ifdef CONFIG_SPARSE_IRQ +int __init arch_probe_nr_irqs(void) +{ + int nr; + + if (nr_irqs > (NR_VECTORS * nr_cpu_ids)) + nr_irqs = NR_VECTORS * nr_cpu_ids; + + nr = nr_irqs_gsi + 8 * nr_cpu_ids; +#if defined(CONFIG_PCI_MSI) || defined(CONFIG_HT_IRQ) + /* + * for MSI and HT dyn irq + */ + nr += nr_irqs_gsi * 16; +#endif + if (nr < nr_irqs) + nr_irqs = nr; + + return 0; +} +#endif + +/* -------------------------------------------------------------------------- + ACPI-based IOAPIC Configuration + -------------------------------------------------------------------------- */ + +#ifdef CONFIG_ACPI + +#ifdef CONFIG_X86_32 +int __init io_apic_get_unique_id(int ioapic, int apic_id) +{ + union IO_APIC_reg_00 reg_00; + static physid_mask_t apic_id_map = PHYSID_MASK_NONE; + physid_mask_t tmp; + unsigned long flags; + int i = 0; + + /* + * The P4 platform supports up to 256 APIC IDs on two separate APIC + * buses (one for LAPICs, one for IOAPICs), where predecessors only + * supports up to 16 on one shared APIC bus. + * + * TBD: Expand LAPIC/IOAPIC support on P4-class systems to take full + * advantage of new APIC bus architecture. + */ + + if (physids_empty(apic_id_map)) + apic_id_map = apic->ioapic_phys_id_map(phys_cpu_present_map); + + spin_lock_irqsave(&ioapic_lock, flags); + reg_00.raw = io_apic_read(ioapic, 0); + spin_unlock_irqrestore(&ioapic_lock, flags); + + if (apic_id >= get_physical_broadcast()) { + printk(KERN_WARNING "IOAPIC[%d]: Invalid apic_id %d, trying " + "%d\n", ioapic, apic_id, reg_00.bits.ID); + apic_id = reg_00.bits.ID; + } + + /* + * Every APIC in a system must have a unique ID or we get lots of nice + * 'stuck on smp_invalidate_needed IPI wait' messages. + */ + if (apic->check_apicid_used(apic_id_map, apic_id)) { + + for (i = 0; i < get_physical_broadcast(); i++) { + if (!apic->check_apicid_used(apic_id_map, i)) + break; + } + + if (i == get_physical_broadcast()) + panic("Max apic_id exceeded!\n"); + + printk(KERN_WARNING "IOAPIC[%d]: apic_id %d already used, " + "trying %d\n", ioapic, apic_id, i); + + apic_id = i; + } + + tmp = apic->apicid_to_cpu_present(apic_id); + physids_or(apic_id_map, apic_id_map, tmp); + + if (reg_00.bits.ID != apic_id) { + reg_00.bits.ID = apic_id; + + spin_lock_irqsave(&ioapic_lock, flags); + io_apic_write(ioapic, 0, reg_00.raw); + reg_00.raw = io_apic_read(ioapic, 0); + spin_unlock_irqrestore(&ioapic_lock, flags); + + /* Sanity check */ + if (reg_00.bits.ID != apic_id) { + printk("IOAPIC[%d]: Unable to change apic_id!\n", ioapic); + return -1; + } + } + + apic_printk(APIC_VERBOSE, KERN_INFO + "IOAPIC[%d]: Assigned apic_id %d\n", ioapic, apic_id); + + return apic_id; +} + +int __init io_apic_get_version(int ioapic) +{ + union IO_APIC_reg_01 reg_01; + unsigned long flags; + + spin_lock_irqsave(&ioapic_lock, flags); + reg_01.raw = io_apic_read(ioapic, 1); + spin_unlock_irqrestore(&ioapic_lock, flags); + + return reg_01.bits.version; +} +#endif + +int io_apic_set_pci_routing (int ioapic, int pin, int irq, int triggering, int polarity) +{ + struct irq_desc *desc; + struct irq_cfg *cfg; + int cpu = boot_cpu_id; + + if (!IO_APIC_IRQ(irq)) { + apic_printk(APIC_QUIET,KERN_ERR "IOAPIC[%d]: Invalid reference to IRQ 0\n", + ioapic); + return -EINVAL; + } + + desc = irq_to_desc_alloc_cpu(irq, cpu); + if (!desc) { + printk(KERN_INFO "can not get irq_desc %d\n", irq); + return 0; + } + + /* + * IRQs < 16 are already in the irq_2_pin[] map + */ + if (irq >= NR_IRQS_LEGACY) { + cfg = desc->chip_data; + add_pin_to_irq_cpu(cfg, cpu, ioapic, pin); + } + + setup_IO_APIC_irq(ioapic, pin, irq, desc, triggering, polarity); + + return 0; +} + + +int acpi_get_override_irq(int bus_irq, int *trigger, int *polarity) +{ + int i; + + if (skip_ioapic_setup) + return -1; + + for (i = 0; i < mp_irq_entries; i++) + if (mp_irqs[i].irqtype == mp_INT && + mp_irqs[i].srcbusirq == bus_irq) + break; + if (i >= mp_irq_entries) + return -1; + + *trigger = irq_trigger(i); + *polarity = irq_polarity(i); + return 0; +} + +#endif /* CONFIG_ACPI */ + +/* + * This function currently is only a helper for the i386 smp boot process where + * we need to reprogram the ioredtbls to cater for the cpus which have come online + * so mask in all cases should simply be apic->target_cpus() + */ +#ifdef CONFIG_SMP +void __init setup_ioapic_dest(void) +{ + int pin, ioapic, irq, irq_entry; + struct irq_desc *desc; + struct irq_cfg *cfg; + const struct cpumask *mask; + + if (skip_ioapic_setup == 1) + return; + + for (ioapic = 0; ioapic < nr_ioapics; ioapic++) { + for (pin = 0; pin < nr_ioapic_registers[ioapic]; pin++) { + irq_entry = find_irq_entry(ioapic, pin, mp_INT); + if (irq_entry == -1) + continue; + irq = pin_2_irq(irq_entry, ioapic, pin); + + /* setup_IO_APIC_irqs could fail to get vector for some device + * when you have too many devices, because at that time only boot + * cpu is online. + */ + desc = irq_to_desc(irq); + cfg = desc->chip_data; + if (!cfg->vector) { + setup_IO_APIC_irq(ioapic, pin, irq, desc, + irq_trigger(irq_entry), + irq_polarity(irq_entry)); + continue; + + } + + /* + * Honour affinities which have been set in early boot + */ + if (desc->status & + (IRQ_NO_BALANCING | IRQ_AFFINITY_SET)) + mask = desc->affinity; + else + mask = apic->target_cpus(); + +#ifdef CONFIG_INTR_REMAP + if (intr_remapping_enabled) + set_ir_ioapic_affinity_irq_desc(desc, mask); + else +#endif + set_ioapic_affinity_irq_desc(desc, mask); + } + + } +} +#endif + +#define IOAPIC_RESOURCE_NAME_SIZE 11 + +static struct resource *ioapic_resources; + +static struct resource * __init ioapic_setup_resources(void) +{ + unsigned long n; + struct resource *res; + char *mem; + int i; + + if (nr_ioapics <= 0) + return NULL; + + n = IOAPIC_RESOURCE_NAME_SIZE + sizeof(struct resource); + n *= nr_ioapics; + + mem = alloc_bootmem(n); + res = (void *)mem; + + if (mem != NULL) { + mem += sizeof(struct resource) * nr_ioapics; + + for (i = 0; i < nr_ioapics; i++) { + res[i].name = mem; + res[i].flags = IORESOURCE_MEM | IORESOURCE_BUSY; + sprintf(mem, "IOAPIC %u", i); + mem += IOAPIC_RESOURCE_NAME_SIZE; + } + } + + ioapic_resources = res; + + return res; +} + +void __init ioapic_init_mappings(void) +{ + unsigned long ioapic_phys, idx = FIX_IO_APIC_BASE_0; + struct resource *ioapic_res; + int i; + + ioapic_res = ioapic_setup_resources(); + for (i = 0; i < nr_ioapics; i++) { + if (smp_found_config) { + ioapic_phys = mp_ioapics[i].apicaddr; +#ifdef CONFIG_X86_32 + if (!ioapic_phys) { + printk(KERN_ERR + "WARNING: bogus zero IO-APIC " + "address found in MPTABLE, " + "disabling IO/APIC support!\n"); + smp_found_config = 0; + skip_ioapic_setup = 1; + goto fake_ioapic_page; + } +#endif + } else { +#ifdef CONFIG_X86_32 +fake_ioapic_page: +#endif + ioapic_phys = (unsigned long) + alloc_bootmem_pages(PAGE_SIZE); + ioapic_phys = __pa(ioapic_phys); + } + set_fixmap_nocache(idx, ioapic_phys); + apic_printk(APIC_VERBOSE, + "mapped IOAPIC to %08lx (%08lx)\n", + __fix_to_virt(idx), ioapic_phys); + idx++; + + if (ioapic_res != NULL) { + ioapic_res->start = ioapic_phys; + ioapic_res->end = ioapic_phys + (4 * 1024) - 1; + ioapic_res++; + } + } +} + +static int __init ioapic_insert_resources(void) +{ + int i; + struct resource *r = ioapic_resources; + + if (!r) { + printk(KERN_ERR + "IO APIC resources could be not be allocated.\n"); + return -1; + } + + for (i = 0; i < nr_ioapics; i++) { + insert_resource(&iomem_resource, r); + r++; + } + + return 0; +} + +/* Insert the IO APIC resources after PCI initialization has occured to handle + * IO APICS that are mapped in on a BAR in PCI space. */ +late_initcall(ioapic_insert_resources); diff --git a/arch/x86/kernel/apic/ipi.c b/arch/x86/kernel/apic/ipi.c new file mode 100644 index 000000000000..dbf5445727a9 --- /dev/null +++ b/arch/x86/kernel/apic/ipi.c @@ -0,0 +1,164 @@ +#include <linux/cpumask.h> +#include <linux/interrupt.h> +#include <linux/init.h> + +#include <linux/mm.h> +#include <linux/delay.h> +#include <linux/spinlock.h> +#include <linux/kernel_stat.h> +#include <linux/mc146818rtc.h> +#include <linux/cache.h> +#include <linux/cpu.h> +#include <linux/module.h> + +#include <asm/smp.h> +#include <asm/mtrr.h> +#include <asm/tlbflush.h> +#include <asm/mmu_context.h> +#include <asm/apic.h> +#include <asm/proto.h> +#include <asm/ipi.h> + +void default_send_IPI_mask_sequence_phys(const struct cpumask *mask, int vector) +{ + unsigned long query_cpu; + unsigned long flags; + + /* + * Hack. The clustered APIC addressing mode doesn't allow us to send + * to an arbitrary mask, so I do a unicast to each CPU instead. + * - mbligh + */ + local_irq_save(flags); + for_each_cpu(query_cpu, mask) { + __default_send_IPI_dest_field(per_cpu(x86_cpu_to_apicid, + query_cpu), vector, APIC_DEST_PHYSICAL); + } + local_irq_restore(flags); +} + +void default_send_IPI_mask_allbutself_phys(const struct cpumask *mask, + int vector) +{ + unsigned int this_cpu = smp_processor_id(); + unsigned int query_cpu; + unsigned long flags; + + /* See Hack comment above */ + + local_irq_save(flags); + for_each_cpu(query_cpu, mask) { + if (query_cpu == this_cpu) + continue; + __default_send_IPI_dest_field(per_cpu(x86_cpu_to_apicid, + query_cpu), vector, APIC_DEST_PHYSICAL); + } + local_irq_restore(flags); +} + +void default_send_IPI_mask_sequence_logical(const struct cpumask *mask, + int vector) +{ + unsigned long flags; + unsigned int query_cpu; + + /* + * Hack. The clustered APIC addressing mode doesn't allow us to send + * to an arbitrary mask, so I do a unicasts to each CPU instead. This + * should be modified to do 1 message per cluster ID - mbligh + */ + + local_irq_save(flags); + for_each_cpu(query_cpu, mask) + __default_send_IPI_dest_field( + apic->cpu_to_logical_apicid(query_cpu), vector, + apic->dest_logical); + local_irq_restore(flags); +} + +void default_send_IPI_mask_allbutself_logical(const struct cpumask *mask, + int vector) +{ + unsigned long flags; + unsigned int query_cpu; + unsigned int this_cpu = smp_processor_id(); + + /* See Hack comment above */ + + local_irq_save(flags); + for_each_cpu(query_cpu, mask) { + if (query_cpu == this_cpu) + continue; + __default_send_IPI_dest_field( + apic->cpu_to_logical_apicid(query_cpu), vector, + apic->dest_logical); + } + local_irq_restore(flags); +} + +#ifdef CONFIG_X86_32 + +/* + * This is only used on smaller machines. + */ +void default_send_IPI_mask_logical(const struct cpumask *cpumask, int vector) +{ + unsigned long mask = cpumask_bits(cpumask)[0]; + unsigned long flags; + + local_irq_save(flags); + WARN_ON(mask & ~cpumask_bits(cpu_online_mask)[0]); + __default_send_IPI_dest_field(mask, vector, apic->dest_logical); + local_irq_restore(flags); +} + +void default_send_IPI_allbutself(int vector) +{ + /* + * if there are no other CPUs in the system then we get an APIC send + * error if we try to broadcast, thus avoid sending IPIs in this case. + */ + if (!(num_online_cpus() > 1)) + return; + + __default_local_send_IPI_allbutself(vector); +} + +void default_send_IPI_all(int vector) +{ + __default_local_send_IPI_all(vector); +} + +void default_send_IPI_self(int vector) +{ + __default_send_IPI_shortcut(APIC_DEST_SELF, vector, apic->dest_logical); +} + +/* must come after the send_IPI functions above for inlining */ +static int convert_apicid_to_cpu(int apic_id) +{ + int i; + + for_each_possible_cpu(i) { + if (per_cpu(x86_cpu_to_apicid, i) == apic_id) + return i; + } + return -1; +} + +int safe_smp_processor_id(void) +{ + int apicid, cpuid; + + if (!boot_cpu_has(X86_FEATURE_APIC)) + return 0; + + apicid = hard_smp_processor_id(); + if (apicid == BAD_APICID) + return 0; + + cpuid = convert_apicid_to_cpu(apicid); + + return cpuid >= 0 ? cpuid : 0; +} +#endif diff --git a/arch/x86/kernel/apic/nmi.c b/arch/x86/kernel/apic/nmi.c new file mode 100644 index 000000000000..bdfad80c3cf1 --- /dev/null +++ b/arch/x86/kernel/apic/nmi.c @@ -0,0 +1,564 @@ +/* + * NMI watchdog support on APIC systems + * + * Started by Ingo Molnar <mingo@redhat.com> + * + * Fixes: + * Mikael Pettersson : AMD K7 support for local APIC NMI watchdog. + * Mikael Pettersson : Power Management for local APIC NMI watchdog. + * Mikael Pettersson : Pentium 4 support for local APIC NMI watchdog. + * Pavel Machek and + * Mikael Pettersson : PM converted to driver model. Disable/enable API. + */ + +#include <asm/apic.h> + +#include <linux/nmi.h> +#include <linux/mm.h> +#include <linux/delay.h> +#include <linux/interrupt.h> +#include <linux/module.h> +#include <linux/sysdev.h> +#include <linux/sysctl.h> +#include <linux/percpu.h> +#include <linux/kprobes.h> +#include <linux/cpumask.h> +#include <linux/kernel_stat.h> +#include <linux/kdebug.h> +#include <linux/smp.h> + +#include <asm/i8259.h> +#include <asm/io_apic.h> +#include <asm/proto.h> +#include <asm/timer.h> + +#include <asm/mce.h> + +#include <asm/mach_traps.h> + +int unknown_nmi_panic; +int nmi_watchdog_enabled; + +static cpumask_t backtrace_mask = CPU_MASK_NONE; + +/* nmi_active: + * >0: the lapic NMI watchdog is active, but can be disabled + * <0: the lapic NMI watchdog has not been set up, and cannot + * be enabled + * 0: the lapic NMI watchdog is disabled, but can be enabled + */ +atomic_t nmi_active = ATOMIC_INIT(0); /* oprofile uses this */ +EXPORT_SYMBOL(nmi_active); + +unsigned int nmi_watchdog = NMI_NONE; +EXPORT_SYMBOL(nmi_watchdog); + +static int panic_on_timeout; + +static unsigned int nmi_hz = HZ; +static DEFINE_PER_CPU(short, wd_enabled); +static int endflag __initdata; + +static inline unsigned int get_nmi_count(int cpu) +{ + return per_cpu(irq_stat, cpu).__nmi_count; +} + +static inline int mce_in_progress(void) +{ +#if defined(CONFIG_X86_64) && defined(CONFIG_X86_MCE) + return atomic_read(&mce_entry) > 0; +#endif + return 0; +} + +/* + * Take the local apic timer and PIT/HPET into account. We don't + * know which one is active, when we have highres/dyntick on + */ +static inline unsigned int get_timer_irqs(int cpu) +{ + return per_cpu(irq_stat, cpu).apic_timer_irqs + + per_cpu(irq_stat, cpu).irq0_irqs; +} + +#ifdef CONFIG_SMP +/* + * The performance counters used by NMI_LOCAL_APIC don't trigger when + * the CPU is idle. To make sure the NMI watchdog really ticks on all + * CPUs during the test make them busy. + */ +static __init void nmi_cpu_busy(void *data) +{ + local_irq_enable_in_hardirq(); + /* + * Intentionally don't use cpu_relax here. This is + * to make sure that the performance counter really ticks, + * even if there is a simulator or similar that catches the + * pause instruction. On a real HT machine this is fine because + * all other CPUs are busy with "useless" delay loops and don't + * care if they get somewhat less cycles. + */ + while (endflag == 0) + mb(); +} +#endif + +static void report_broken_nmi(int cpu, int *prev_nmi_count) +{ + printk(KERN_CONT "\n"); + + printk(KERN_WARNING + "WARNING: CPU#%d: NMI appears to be stuck (%d->%d)!\n", + cpu, prev_nmi_count[cpu], get_nmi_count(cpu)); + + printk(KERN_WARNING + "Please report this to bugzilla.kernel.org,\n"); + printk(KERN_WARNING + "and attach the output of the 'dmesg' command.\n"); + + per_cpu(wd_enabled, cpu) = 0; + atomic_dec(&nmi_active); +} + +static void __acpi_nmi_disable(void *__unused) +{ + apic_write(APIC_LVT0, APIC_DM_NMI | APIC_LVT_MASKED); +} + +int __init check_nmi_watchdog(void) +{ + unsigned int *prev_nmi_count; + int cpu; + + if (!nmi_watchdog_active() || !atomic_read(&nmi_active)) + return 0; + + prev_nmi_count = kmalloc(nr_cpu_ids * sizeof(int), GFP_KERNEL); + if (!prev_nmi_count) + goto error; + + printk(KERN_INFO "Testing NMI watchdog ... "); + +#ifdef CONFIG_SMP + if (nmi_watchdog == NMI_LOCAL_APIC) + smp_call_function(nmi_cpu_busy, (void *)&endflag, 0); +#endif + + for_each_possible_cpu(cpu) + prev_nmi_count[cpu] = get_nmi_count(cpu); + local_irq_enable(); + mdelay((20 * 1000) / nmi_hz); /* wait 20 ticks */ + + for_each_online_cpu(cpu) { + if (!per_cpu(wd_enabled, cpu)) + continue; + if (get_nmi_count(cpu) - prev_nmi_count[cpu] <= 5) + report_broken_nmi(cpu, prev_nmi_count); + } + endflag = 1; + if (!atomic_read(&nmi_active)) { + kfree(prev_nmi_count); + atomic_set(&nmi_active, -1); + goto error; + } + printk("OK.\n"); + + /* + * now that we know it works we can reduce NMI frequency to + * something more reasonable; makes a difference in some configs + */ + if (nmi_watchdog == NMI_LOCAL_APIC) + nmi_hz = lapic_adjust_nmi_hz(1); + + kfree(prev_nmi_count); + return 0; +error: + if (nmi_watchdog == NMI_IO_APIC) { + if (!timer_through_8259) + disable_8259A_irq(0); + on_each_cpu(__acpi_nmi_disable, NULL, 1); + } + +#ifdef CONFIG_X86_32 + timer_ack = 0; +#endif + return -1; +} + +static int __init setup_nmi_watchdog(char *str) +{ + unsigned int nmi; + + if (!strncmp(str, "panic", 5)) { + panic_on_timeout = 1; + str = strchr(str, ','); + if (!str) + return 1; + ++str; + } + + if (!strncmp(str, "lapic", 5)) + nmi_watchdog = NMI_LOCAL_APIC; + else if (!strncmp(str, "ioapic", 6)) + nmi_watchdog = NMI_IO_APIC; + else { + get_option(&str, &nmi); + if (nmi >= NMI_INVALID) + return 0; + nmi_watchdog = nmi; + } + + return 1; +} +__setup("nmi_watchdog=", setup_nmi_watchdog); + +/* + * Suspend/resume support + */ +#ifdef CONFIG_PM + +static int nmi_pm_active; /* nmi_active before suspend */ + +static int lapic_nmi_suspend(struct sys_device *dev, pm_message_t state) +{ + /* only CPU0 goes here, other CPUs should be offline */ + nmi_pm_active = atomic_read(&nmi_active); + stop_apic_nmi_watchdog(NULL); + BUG_ON(atomic_read(&nmi_active) != 0); + return 0; +} + +static int lapic_nmi_resume(struct sys_device *dev) +{ + /* only CPU0 goes here, other CPUs should be offline */ + if (nmi_pm_active > 0) { + setup_apic_nmi_watchdog(NULL); + touch_nmi_watchdog(); + } + return 0; +} + +static struct sysdev_class nmi_sysclass = { + .name = "lapic_nmi", + .resume = lapic_nmi_resume, + .suspend = lapic_nmi_suspend, +}; + +static struct sys_device device_lapic_nmi = { + .id = 0, + .cls = &nmi_sysclass, +}; + +static int __init init_lapic_nmi_sysfs(void) +{ + int error; + + /* + * should really be a BUG_ON but b/c this is an + * init call, it just doesn't work. -dcz + */ + if (nmi_watchdog != NMI_LOCAL_APIC) + return 0; + + if (atomic_read(&nmi_active) < 0) + return 0; + + error = sysdev_class_register(&nmi_sysclass); + if (!error) + error = sysdev_register(&device_lapic_nmi); + return error; +} + +/* must come after the local APIC's device_initcall() */ +late_initcall(init_lapic_nmi_sysfs); + +#endif /* CONFIG_PM */ + +static void __acpi_nmi_enable(void *__unused) +{ + apic_write(APIC_LVT0, APIC_DM_NMI); +} + +/* + * Enable timer based NMIs on all CPUs: + */ +void acpi_nmi_enable(void) +{ + if (atomic_read(&nmi_active) && nmi_watchdog == NMI_IO_APIC) + on_each_cpu(__acpi_nmi_enable, NULL, 1); +} + +/* + * Disable timer based NMIs on all CPUs: + */ +void acpi_nmi_disable(void) +{ + if (atomic_read(&nmi_active) && nmi_watchdog == NMI_IO_APIC) + on_each_cpu(__acpi_nmi_disable, NULL, 1); +} + +/* + * This function is called as soon the LAPIC NMI watchdog driver has everything + * in place and it's ready to check if the NMIs belong to the NMI watchdog + */ +void cpu_nmi_set_wd_enabled(void) +{ + __get_cpu_var(wd_enabled) = 1; +} + +void setup_apic_nmi_watchdog(void *unused) +{ + if (__get_cpu_var(wd_enabled)) + return; + + /* cheap hack to support suspend/resume */ + /* if cpu0 is not active neither should the other cpus */ + if (smp_processor_id() != 0 && atomic_read(&nmi_active) <= 0) + return; + + switch (nmi_watchdog) { + case NMI_LOCAL_APIC: + if (lapic_watchdog_init(nmi_hz) < 0) { + __get_cpu_var(wd_enabled) = 0; + return; + } + /* FALL THROUGH */ + case NMI_IO_APIC: + __get_cpu_var(wd_enabled) = 1; + atomic_inc(&nmi_active); + } +} + +void stop_apic_nmi_watchdog(void *unused) +{ + /* only support LOCAL and IO APICs for now */ + if (!nmi_watchdog_active()) + return; + if (__get_cpu_var(wd_enabled) == 0) + return; + if (nmi_watchdog == NMI_LOCAL_APIC) + lapic_watchdog_stop(); + else + __acpi_nmi_disable(NULL); + __get_cpu_var(wd_enabled) = 0; + atomic_dec(&nmi_active); +} + +/* + * the best way to detect whether a CPU has a 'hard lockup' problem + * is to check it's local APIC timer IRQ counts. If they are not + * changing then that CPU has some problem. + * + * as these watchdog NMI IRQs are generated on every CPU, we only + * have to check the current processor. + * + * since NMIs don't listen to _any_ locks, we have to be extremely + * careful not to rely on unsafe variables. The printk might lock + * up though, so we have to break up any console locks first ... + * [when there will be more tty-related locks, break them up here too!] + */ + +static DEFINE_PER_CPU(unsigned, last_irq_sum); +static DEFINE_PER_CPU(local_t, alert_counter); +static DEFINE_PER_CPU(int, nmi_touch); + +void touch_nmi_watchdog(void) +{ + if (nmi_watchdog_active()) { + unsigned cpu; + + /* + * Tell other CPUs to reset their alert counters. We cannot + * do it ourselves because the alert count increase is not + * atomic. + */ + for_each_present_cpu(cpu) { + if (per_cpu(nmi_touch, cpu) != 1) + per_cpu(nmi_touch, cpu) = 1; + } + } + + /* + * Tickle the softlockup detector too: + */ + touch_softlockup_watchdog(); +} +EXPORT_SYMBOL(touch_nmi_watchdog); + +notrace __kprobes int +nmi_watchdog_tick(struct pt_regs *regs, unsigned reason) +{ + /* + * Since current_thread_info()-> is always on the stack, and we + * always switch the stack NMI-atomically, it's safe to use + * smp_processor_id(). + */ + unsigned int sum; + int touched = 0; + int cpu = smp_processor_id(); + int rc = 0; + + /* check for other users first */ + if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT) + == NOTIFY_STOP) { + rc = 1; + touched = 1; + } + + sum = get_timer_irqs(cpu); + + if (__get_cpu_var(nmi_touch)) { + __get_cpu_var(nmi_touch) = 0; + touched = 1; + } + + if (cpu_isset(cpu, backtrace_mask)) { + static DEFINE_SPINLOCK(lock); /* Serialise the printks */ + + spin_lock(&lock); + printk(KERN_WARNING "NMI backtrace for cpu %d\n", cpu); + dump_stack(); + spin_unlock(&lock); + cpu_clear(cpu, backtrace_mask); + } + + /* Could check oops_in_progress here too, but it's safer not to */ + if (mce_in_progress()) + touched = 1; + + /* if the none of the timers isn't firing, this cpu isn't doing much */ + if (!touched && __get_cpu_var(last_irq_sum) == sum) { + /* + * Ayiee, looks like this CPU is stuck ... + * wait a few IRQs (5 seconds) before doing the oops ... + */ + local_inc(&__get_cpu_var(alert_counter)); + if (local_read(&__get_cpu_var(alert_counter)) == 5 * nmi_hz) + /* + * die_nmi will return ONLY if NOTIFY_STOP happens.. + */ + die_nmi("BUG: NMI Watchdog detected LOCKUP", + regs, panic_on_timeout); + } else { + __get_cpu_var(last_irq_sum) = sum; + local_set(&__get_cpu_var(alert_counter), 0); + } + + /* see if the nmi watchdog went off */ + if (!__get_cpu_var(wd_enabled)) + return rc; + switch (nmi_watchdog) { + case NMI_LOCAL_APIC: + rc |= lapic_wd_event(nmi_hz); + break; + case NMI_IO_APIC: + /* + * don't know how to accurately check for this. + * just assume it was a watchdog timer interrupt + * This matches the old behaviour. + */ + rc = 1; + break; + } + return rc; +} + +#ifdef CONFIG_SYSCTL + +static void enable_ioapic_nmi_watchdog_single(void *unused) +{ + __get_cpu_var(wd_enabled) = 1; + atomic_inc(&nmi_active); + __acpi_nmi_enable(NULL); +} + +static void enable_ioapic_nmi_watchdog(void) +{ + on_each_cpu(enable_ioapic_nmi_watchdog_single, NULL, 1); + touch_nmi_watchdog(); +} + +static void disable_ioapic_nmi_watchdog(void) +{ + on_each_cpu(stop_apic_nmi_watchdog, NULL, 1); +} + +static int __init setup_unknown_nmi_panic(char *str) +{ + unknown_nmi_panic = 1; + return 1; +} +__setup("unknown_nmi_panic", setup_unknown_nmi_panic); + +static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu) +{ + unsigned char reason = get_nmi_reason(); + char buf[64]; + + sprintf(buf, "NMI received for unknown reason %02x\n", reason); + die_nmi(buf, regs, 1); /* Always panic here */ + return 0; +} + +/* + * proc handler for /proc/sys/kernel/nmi + */ +int proc_nmi_enabled(struct ctl_table *table, int write, struct file *file, + void __user *buffer, size_t *length, loff_t *ppos) +{ + int old_state; + + nmi_watchdog_enabled = (atomic_read(&nmi_active) > 0) ? 1 : 0; + old_state = nmi_watchdog_enabled; + proc_dointvec(table, write, file, buffer, length, ppos); + if (!!old_state == !!nmi_watchdog_enabled) + return 0; + + if (atomic_read(&nmi_active) < 0 || !nmi_watchdog_active()) { + printk(KERN_WARNING + "NMI watchdog is permanently disabled\n"); + return -EIO; + } + + if (nmi_watchdog == NMI_LOCAL_APIC) { + if (nmi_watchdog_enabled) + enable_lapic_nmi_watchdog(); + else + disable_lapic_nmi_watchdog(); + } else if (nmi_watchdog == NMI_IO_APIC) { + if (nmi_watchdog_enabled) + enable_ioapic_nmi_watchdog(); + else + disable_ioapic_nmi_watchdog(); + } else { + printk(KERN_WARNING + "NMI watchdog doesn't know what hardware to touch\n"); + return -EIO; + } + return 0; +} + +#endif /* CONFIG_SYSCTL */ + +int do_nmi_callback(struct pt_regs *regs, int cpu) +{ +#ifdef CONFIG_SYSCTL + if (unknown_nmi_panic) + return unknown_nmi_panic_callback(regs, cpu); +#endif + return 0; +} + +void __trigger_all_cpu_backtrace(void) +{ + int i; + + backtrace_mask = cpu_online_map; + /* Wait for up to 10 seconds for all CPUs to do the backtrace */ + for (i = 0; i < 10 * 1000; i++) { + if (cpus_empty(backtrace_mask)) + break; + mdelay(1); + } +} diff --git a/arch/x86/kernel/apic/x2apic_cluster.c b/arch/x86/kernel/apic/x2apic_cluster.c new file mode 100644 index 000000000000..4e39d9ad4d52 --- /dev/null +++ b/arch/x86/kernel/apic/x2apic_cluster.c @@ -0,0 +1,243 @@ +#include <linux/threads.h> +#include <linux/cpumask.h> +#include <linux/string.h> +#include <linux/kernel.h> +#include <linux/ctype.h> +#include <linux/init.h> +#include <linux/dmar.h> + +#include <asm/smp.h> +#include <asm/apic.h> +#include <asm/ipi.h> + +DEFINE_PER_CPU(u32, x86_cpu_to_logical_apicid); + +static int x2apic_acpi_madt_oem_check(char *oem_id, char *oem_table_id) +{ + if (cpu_has_x2apic) + return 1; + + return 0; +} + +/* Start with all IRQs pointing to boot CPU. IRQ balancing will shift them. */ + +static const struct cpumask *x2apic_target_cpus(void) +{ + return cpumask_of(0); +} + +/* + * for now each logical cpu is in its own vector allocation domain. + */ +static void x2apic_vector_allocation_domain(int cpu, struct cpumask *retmask) +{ + cpumask_clear(retmask); + cpumask_set_cpu(cpu, retmask); +} + +static void + __x2apic_send_IPI_dest(unsigned int apicid, int vector, unsigned int dest) +{ + unsigned long cfg; + + cfg = __prepare_ICR(0, vector, dest); + + /* + * send the IPI. + */ + native_x2apic_icr_write(cfg, apicid); +} + +/* + * for now, we send the IPI's one by one in the cpumask. + * TBD: Based on the cpu mask, we can send the IPI's to the cluster group + * at once. We have 16 cpu's in a cluster. This will minimize IPI register + * writes. + */ +static void x2apic_send_IPI_mask(const struct cpumask *mask, int vector) +{ + unsigned long query_cpu; + unsigned long flags; + + local_irq_save(flags); + for_each_cpu(query_cpu, mask) { + __x2apic_send_IPI_dest( + per_cpu(x86_cpu_to_logical_apicid, query_cpu), + vector, apic->dest_logical); + } + local_irq_restore(flags); +} + +static void + x2apic_send_IPI_mask_allbutself(const struct cpumask *mask, int vector) +{ + unsigned long this_cpu = smp_processor_id(); + unsigned long query_cpu; + unsigned long flags; + + local_irq_save(flags); + for_each_cpu(query_cpu, mask) { + if (query_cpu == this_cpu) + continue; + __x2apic_send_IPI_dest( + per_cpu(x86_cpu_to_logical_apicid, query_cpu), + vector, apic->dest_logical); + } + local_irq_restore(flags); +} + +static void x2apic_send_IPI_allbutself(int vector) +{ + unsigned long this_cpu = smp_processor_id(); + unsigned long query_cpu; + unsigned long flags; + + local_irq_save(flags); + for_each_online_cpu(query_cpu) { + if (query_cpu == this_cpu) + continue; + __x2apic_send_IPI_dest( + per_cpu(x86_cpu_to_logical_apicid, query_cpu), + vector, apic->dest_logical); + } + local_irq_restore(flags); +} + +static void x2apic_send_IPI_all(int vector) +{ + x2apic_send_IPI_mask(cpu_online_mask, vector); +} + +static int x2apic_apic_id_registered(void) +{ + return 1; +} + +static unsigned int x2apic_cpu_mask_to_apicid(const struct cpumask *cpumask) +{ + /* + * We're using fixed IRQ delivery, can only return one logical APIC ID. + * May as well be the first. + */ + int cpu = cpumask_first(cpumask); + + if ((unsigned)cpu < nr_cpu_ids) + return per_cpu(x86_cpu_to_logical_apicid, cpu); + else + return BAD_APICID; +} + +static unsigned int +x2apic_cpu_mask_to_apicid_and(const struct cpumask *cpumask, + const struct cpumask *andmask) +{ + int cpu; + + /* + * We're using fixed IRQ delivery, can only return one logical APIC ID. + * May as well be the first. + */ + for_each_cpu_and(cpu, cpumask, andmask) { + if (cpumask_test_cpu(cpu, cpu_online_mask)) + break; + } + + if (cpu < nr_cpu_ids) + return per_cpu(x86_cpu_to_logical_apicid, cpu); + + return BAD_APICID; +} + +static unsigned int x2apic_cluster_phys_get_apic_id(unsigned long x) +{ + unsigned int id; + + id = x; + return id; +} + +static unsigned long set_apic_id(unsigned int id) +{ + unsigned long x; + + x = id; + return x; +} + +static int x2apic_cluster_phys_pkg_id(int initial_apicid, int index_msb) +{ + return current_cpu_data.initial_apicid >> index_msb; +} + +static void x2apic_send_IPI_self(int vector) +{ + apic_write(APIC_SELF_IPI, vector); +} + +static void init_x2apic_ldr(void) +{ + int cpu = smp_processor_id(); + + per_cpu(x86_cpu_to_logical_apicid, cpu) = apic_read(APIC_LDR); +} + +struct apic apic_x2apic_cluster = { + + .name = "cluster x2apic", + .probe = NULL, + .acpi_madt_oem_check = x2apic_acpi_madt_oem_check, + .apic_id_registered = x2apic_apic_id_registered, + + .irq_delivery_mode = dest_LowestPrio, + .irq_dest_mode = 1, /* logical */ + + .target_cpus = x2apic_target_cpus, + .disable_esr = 0, + .dest_logical = APIC_DEST_LOGICAL, + .check_apicid_used = NULL, + .check_apicid_present = NULL, + + .vector_allocation_domain = x2apic_vector_allocation_domain, + .init_apic_ldr = init_x2apic_ldr, + + .ioapic_phys_id_map = NULL, + .setup_apic_routing = NULL, + .multi_timer_check = NULL, + .apicid_to_node = NULL, + .cpu_to_logical_apicid = NULL, + .cpu_present_to_apicid = default_cpu_present_to_apicid, + .apicid_to_cpu_present = NULL, + .setup_portio_remap = NULL, + .check_phys_apicid_present = default_check_phys_apicid_present, + .enable_apic_mode = NULL, + .phys_pkg_id = x2apic_cluster_phys_pkg_id, + .mps_oem_check = NULL, + + .get_apic_id = x2apic_cluster_phys_get_apic_id, + .set_apic_id = set_apic_id, + .apic_id_mask = 0xFFFFFFFFu, + + .cpu_mask_to_apicid = x2apic_cpu_mask_to_apicid, + .cpu_mask_to_apicid_and = x2apic_cpu_mask_to_apicid_and, + + .send_IPI_mask = x2apic_send_IPI_mask, + .send_IPI_mask_allbutself = x2apic_send_IPI_mask_allbutself, + .send_IPI_allbutself = x2apic_send_IPI_allbutself, + .send_IPI_all = x2apic_send_IPI_all, + .send_IPI_self = x2apic_send_IPI_self, + + .wakeup_cpu = NULL, + .trampoline_phys_low = DEFAULT_TRAMPOLINE_PHYS_LOW, + .trampoline_phys_high = DEFAULT_TRAMPOLINE_PHYS_HIGH, + .wait_for_init_deassert = NULL, + .smp_callin_clear_local_apic = NULL, + .inquire_remote_apic = NULL, + + .read = native_apic_msr_read, + .write = native_apic_msr_write, + .icr_read = native_x2apic_icr_read, + .icr_write = native_x2apic_icr_write, + .wait_icr_idle = native_x2apic_wait_icr_idle, + .safe_wait_icr_idle = native_safe_x2apic_wait_icr_idle, +}; diff --git a/arch/x86/kernel/apic/x2apic_phys.c b/arch/x86/kernel/apic/x2apic_phys.c new file mode 100644 index 000000000000..d2d52eb9f7ea --- /dev/null +++ b/arch/x86/kernel/apic/x2apic_phys.c @@ -0,0 +1,229 @@ +#include <linux/threads.h> +#include <linux/cpumask.h> +#include <linux/string.h> +#include <linux/kernel.h> +#include <linux/ctype.h> +#include <linux/init.h> +#include <linux/dmar.h> + +#include <asm/smp.h> +#include <asm/apic.h> +#include <asm/ipi.h> + +static int x2apic_phys; + +static int set_x2apic_phys_mode(char *arg) +{ + x2apic_phys = 1; + return 0; +} +early_param("x2apic_phys", set_x2apic_phys_mode); + +static int x2apic_acpi_madt_oem_check(char *oem_id, char *oem_table_id) +{ + if (cpu_has_x2apic && x2apic_phys) + return 1; + + return 0; +} + +/* Start with all IRQs pointing to boot CPU. IRQ balancing will shift them. */ + +static const struct cpumask *x2apic_target_cpus(void) +{ + return cpumask_of(0); +} + +static void x2apic_vector_allocation_domain(int cpu, struct cpumask *retmask) +{ + cpumask_clear(retmask); + cpumask_set_cpu(cpu, retmask); +} + +static void __x2apic_send_IPI_dest(unsigned int apicid, int vector, + unsigned int dest) +{ + unsigned long cfg; + + cfg = __prepare_ICR(0, vector, dest); + + /* + * send the IPI. + */ + native_x2apic_icr_write(cfg, apicid); +} + +static void x2apic_send_IPI_mask(const struct cpumask *mask, int vector) +{ + unsigned long query_cpu; + unsigned long flags; + + local_irq_save(flags); + for_each_cpu(query_cpu, mask) { + __x2apic_send_IPI_dest(per_cpu(x86_cpu_to_apicid, query_cpu), + vector, APIC_DEST_PHYSICAL); + } + local_irq_restore(flags); +} + +static void + x2apic_send_IPI_mask_allbutself(const struct cpumask *mask, int vector) +{ + unsigned long this_cpu = smp_processor_id(); + unsigned long query_cpu; + unsigned long flags; + + local_irq_save(flags); + for_each_cpu(query_cpu, mask) { + if (query_cpu != this_cpu) + __x2apic_send_IPI_dest( + per_cpu(x86_cpu_to_apicid, query_cpu), + vector, APIC_DEST_PHYSICAL); + } + local_irq_restore(flags); +} + +static void x2apic_send_IPI_allbutself(int vector) +{ + unsigned long this_cpu = smp_processor_id(); + unsigned long query_cpu; + unsigned long flags; + + local_irq_save(flags); + for_each_online_cpu(query_cpu) { + if (query_cpu == this_cpu) + continue; + __x2apic_send_IPI_dest(per_cpu(x86_cpu_to_apicid, query_cpu), + vector, APIC_DEST_PHYSICAL); + } + local_irq_restore(flags); +} + +static void x2apic_send_IPI_all(int vector) +{ + x2apic_send_IPI_mask(cpu_online_mask, vector); +} + +static int x2apic_apic_id_registered(void) +{ + return 1; +} + +static unsigned int x2apic_cpu_mask_to_apicid(const struct cpumask *cpumask) +{ + /* + * We're using fixed IRQ delivery, can only return one phys APIC ID. + * May as well be the first. + */ + int cpu = cpumask_first(cpumask); + + if ((unsigned)cpu < nr_cpu_ids) + return per_cpu(x86_cpu_to_apicid, cpu); + else + return BAD_APICID; +} + +static unsigned int +x2apic_cpu_mask_to_apicid_and(const struct cpumask *cpumask, + const struct cpumask *andmask) +{ + int cpu; + + /* + * We're using fixed IRQ delivery, can only return one phys APIC ID. + * May as well be the first. + */ + for_each_cpu_and(cpu, cpumask, andmask) { + if (cpumask_test_cpu(cpu, cpu_online_mask)) + break; + } + + if (cpu < nr_cpu_ids) + return per_cpu(x86_cpu_to_apicid, cpu); + + return BAD_APICID; +} + +static unsigned int x2apic_phys_get_apic_id(unsigned long x) +{ + return x; +} + +static unsigned long set_apic_id(unsigned int id) +{ + return id; +} + +static int x2apic_phys_pkg_id(int initial_apicid, int index_msb) +{ + return current_cpu_data.initial_apicid >> index_msb; +} + +static void x2apic_send_IPI_self(int vector) +{ + apic_write(APIC_SELF_IPI, vector); +} + +static void init_x2apic_ldr(void) +{ +} + +struct apic apic_x2apic_phys = { + + .name = "physical x2apic", + .probe = NULL, + .acpi_madt_oem_check = x2apic_acpi_madt_oem_check, + .apic_id_registered = x2apic_apic_id_registered, + + .irq_delivery_mode = dest_Fixed, + .irq_dest_mode = 0, /* physical */ + + .target_cpus = x2apic_target_cpus, + .disable_esr = 0, + .dest_logical = 0, + .check_apicid_used = NULL, + .check_apicid_present = NULL, + + .vector_allocation_domain = x2apic_vector_allocation_domain, + .init_apic_ldr = init_x2apic_ldr, + + .ioapic_phys_id_map = NULL, + .setup_apic_routing = NULL, + .multi_timer_check = NULL, + .apicid_to_node = NULL, + .cpu_to_logical_apicid = NULL, + .cpu_present_to_apicid = default_cpu_present_to_apicid, + .apicid_to_cpu_present = NULL, + .setup_portio_remap = NULL, + .check_phys_apicid_present = default_check_phys_apicid_present, + .enable_apic_mode = NULL, + .phys_pkg_id = x2apic_phys_pkg_id, + .mps_oem_check = NULL, + + .get_apic_id = x2apic_phys_get_apic_id, + .set_apic_id = set_apic_id, + .apic_id_mask = 0xFFFFFFFFu, + + .cpu_mask_to_apicid = x2apic_cpu_mask_to_apicid, + .cpu_mask_to_apicid_and = x2apic_cpu_mask_to_apicid_and, + + .send_IPI_mask = x2apic_send_IPI_mask, + .send_IPI_mask_allbutself = x2apic_send_IPI_mask_allbutself, + .send_IPI_allbutself = x2apic_send_IPI_allbutself, + .send_IPI_all = x2apic_send_IPI_all, + .send_IPI_self = x2apic_send_IPI_self, + + .wakeup_cpu = NULL, + .trampoline_phys_low = DEFAULT_TRAMPOLINE_PHYS_LOW, + .trampoline_phys_high = DEFAULT_TRAMPOLINE_PHYS_HIGH, + .wait_for_init_deassert = NULL, + .smp_callin_clear_local_apic = NULL, + .inquire_remote_apic = NULL, + + .read = native_apic_msr_read, + .write = native_apic_msr_write, + .icr_read = native_x2apic_icr_read, + .icr_write = native_x2apic_icr_write, + .wait_icr_idle = native_x2apic_wait_icr_idle, + .safe_wait_icr_idle = native_safe_x2apic_wait_icr_idle, +}; diff --git a/arch/x86/kernel/apic/x2apic_uv_x.c b/arch/x86/kernel/apic/x2apic_uv_x.c new file mode 100644 index 000000000000..20b4ad07c3a1 --- /dev/null +++ b/arch/x86/kernel/apic/x2apic_uv_x.c @@ -0,0 +1,643 @@ +/* + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * SGI UV APIC functions (note: not an Intel compatible APIC) + * + * Copyright (C) 2007-2008 Silicon Graphics, Inc. All rights reserved. + */ + +#include <linux/kernel.h> +#include <linux/threads.h> +#include <linux/cpu.h> +#include <linux/cpumask.h> +#include <linux/string.h> +#include <linux/ctype.h> +#include <linux/init.h> +#include <linux/sched.h> +#include <linux/module.h> +#include <linux/hardirq.h> +#include <linux/timer.h> +#include <linux/proc_fs.h> +#include <asm/current.h> +#include <asm/smp.h> +#include <asm/apic.h> +#include <asm/ipi.h> +#include <asm/pgtable.h> +#include <asm/uv/uv.h> +#include <asm/uv/uv_mmrs.h> +#include <asm/uv/uv_hub.h> +#include <asm/uv/bios.h> + +DEFINE_PER_CPU(int, x2apic_extra_bits); + +static enum uv_system_type uv_system_type; + +static int uv_acpi_madt_oem_check(char *oem_id, char *oem_table_id) +{ + if (!strcmp(oem_id, "SGI")) { + if (!strcmp(oem_table_id, "UVL")) + uv_system_type = UV_LEGACY_APIC; + else if (!strcmp(oem_table_id, "UVX")) + uv_system_type = UV_X2APIC; + else if (!strcmp(oem_table_id, "UVH")) { + uv_system_type = UV_NON_UNIQUE_APIC; + return 1; + } + } + return 0; +} + +enum uv_system_type get_uv_system_type(void) +{ + return uv_system_type; +} + +int is_uv_system(void) +{ + return uv_system_type != UV_NONE; +} +EXPORT_SYMBOL_GPL(is_uv_system); + +DEFINE_PER_CPU(struct uv_hub_info_s, __uv_hub_info); +EXPORT_PER_CPU_SYMBOL_GPL(__uv_hub_info); + +struct uv_blade_info *uv_blade_info; +EXPORT_SYMBOL_GPL(uv_blade_info); + +short *uv_node_to_blade; +EXPORT_SYMBOL_GPL(uv_node_to_blade); + +short *uv_cpu_to_blade; +EXPORT_SYMBOL_GPL(uv_cpu_to_blade); + +short uv_possible_blades; +EXPORT_SYMBOL_GPL(uv_possible_blades); + +unsigned long sn_rtc_cycles_per_second; +EXPORT_SYMBOL(sn_rtc_cycles_per_second); + +/* Start with all IRQs pointing to boot CPU. IRQ balancing will shift them. */ + +static const struct cpumask *uv_target_cpus(void) +{ + return cpumask_of(0); +} + +static void uv_vector_allocation_domain(int cpu, struct cpumask *retmask) +{ + cpumask_clear(retmask); + cpumask_set_cpu(cpu, retmask); +} + +int uv_wakeup_secondary(int phys_apicid, unsigned int start_rip) +{ + unsigned long val; + int pnode; + + pnode = uv_apicid_to_pnode(phys_apicid); + val = (1UL << UVH_IPI_INT_SEND_SHFT) | + (phys_apicid << UVH_IPI_INT_APIC_ID_SHFT) | + (((long)start_rip << UVH_IPI_INT_VECTOR_SHFT) >> 12) | + APIC_DM_INIT; + uv_write_global_mmr64(pnode, UVH_IPI_INT, val); + mdelay(10); + + val = (1UL << UVH_IPI_INT_SEND_SHFT) | + (phys_apicid << UVH_IPI_INT_APIC_ID_SHFT) | + (((long)start_rip << UVH_IPI_INT_VECTOR_SHFT) >> 12) | + APIC_DM_STARTUP; + uv_write_global_mmr64(pnode, UVH_IPI_INT, val); + return 0; +} + +static void uv_send_IPI_one(int cpu, int vector) +{ + unsigned long val, apicid; + int pnode; + + apicid = per_cpu(x86_cpu_to_apicid, cpu); + pnode = uv_apicid_to_pnode(apicid); + + val = (1UL << UVH_IPI_INT_SEND_SHFT) | + (apicid << UVH_IPI_INT_APIC_ID_SHFT) | + (vector << UVH_IPI_INT_VECTOR_SHFT); + + uv_write_global_mmr64(pnode, UVH_IPI_INT, val); +} + +static void uv_send_IPI_mask(const struct cpumask *mask, int vector) +{ + unsigned int cpu; + + for_each_cpu(cpu, mask) + uv_send_IPI_one(cpu, vector); +} + +static void uv_send_IPI_mask_allbutself(const struct cpumask *mask, int vector) +{ + unsigned int this_cpu = smp_processor_id(); + unsigned int cpu; + + for_each_cpu(cpu, mask) { + if (cpu != this_cpu) + uv_send_IPI_one(cpu, vector); + } +} + +static void uv_send_IPI_allbutself(int vector) +{ + unsigned int this_cpu = smp_processor_id(); + unsigned int cpu; + + for_each_online_cpu(cpu) { + if (cpu != this_cpu) + uv_send_IPI_one(cpu, vector); + } +} + +static void uv_send_IPI_all(int vector) +{ + uv_send_IPI_mask(cpu_online_mask, vector); +} + +static int uv_apic_id_registered(void) +{ + return 1; +} + +static void uv_init_apic_ldr(void) +{ +} + +static unsigned int uv_cpu_mask_to_apicid(const struct cpumask *cpumask) +{ + /* + * We're using fixed IRQ delivery, can only return one phys APIC ID. + * May as well be the first. + */ + int cpu = cpumask_first(cpumask); + + if ((unsigned)cpu < nr_cpu_ids) + return per_cpu(x86_cpu_to_apicid, cpu); + else + return BAD_APICID; +} + +static unsigned int +uv_cpu_mask_to_apicid_and(const struct cpumask *cpumask, + const struct cpumask *andmask) +{ + int cpu; + + /* + * We're using fixed IRQ delivery, can only return one phys APIC ID. + * May as well be the first. + */ + for_each_cpu_and(cpu, cpumask, andmask) { + if (cpumask_test_cpu(cpu, cpu_online_mask)) + break; + } + if (cpu < nr_cpu_ids) + return per_cpu(x86_cpu_to_apicid, cpu); + + return BAD_APICID; +} + +static unsigned int x2apic_get_apic_id(unsigned long x) +{ + unsigned int id; + + WARN_ON(preemptible() && num_online_cpus() > 1); + id = x | __get_cpu_var(x2apic_extra_bits); + + return id; +} + +static unsigned long set_apic_id(unsigned int id) +{ + unsigned long x; + + /* maskout x2apic_extra_bits ? */ + x = id; + return x; +} + +static unsigned int uv_read_apic_id(void) +{ + + return x2apic_get_apic_id(apic_read(APIC_ID)); +} + +static int uv_phys_pkg_id(int initial_apicid, int index_msb) +{ + return uv_read_apic_id() >> index_msb; +} + +static void uv_send_IPI_self(int vector) +{ + apic_write(APIC_SELF_IPI, vector); +} + +struct apic apic_x2apic_uv_x = { + + .name = "UV large system", + .probe = NULL, + .acpi_madt_oem_check = uv_acpi_madt_oem_check, + .apic_id_registered = uv_apic_id_registered, + + .irq_delivery_mode = dest_Fixed, + .irq_dest_mode = 1, /* logical */ + + .target_cpus = uv_target_cpus, + .disable_esr = 0, + .dest_logical = APIC_DEST_LOGICAL, + .check_apicid_used = NULL, + .check_apicid_present = NULL, + + .vector_allocation_domain = uv_vector_allocation_domain, + .init_apic_ldr = uv_init_apic_ldr, + + .ioapic_phys_id_map = NULL, + .setup_apic_routing = NULL, + .multi_timer_check = NULL, + .apicid_to_node = NULL, + .cpu_to_logical_apicid = NULL, + .cpu_present_to_apicid = default_cpu_present_to_apicid, + .apicid_to_cpu_present = NULL, + .setup_portio_remap = NULL, + .check_phys_apicid_present = default_check_phys_apicid_present, + .enable_apic_mode = NULL, + .phys_pkg_id = uv_phys_pkg_id, + .mps_oem_check = NULL, + + .get_apic_id = x2apic_get_apic_id, + .set_apic_id = set_apic_id, + .apic_id_mask = 0xFFFFFFFFu, + + .cpu_mask_to_apicid = uv_cpu_mask_to_apicid, + .cpu_mask_to_apicid_and = uv_cpu_mask_to_apicid_and, + + .send_IPI_mask = uv_send_IPI_mask, + .send_IPI_mask_allbutself = uv_send_IPI_mask_allbutself, + .send_IPI_allbutself = uv_send_IPI_allbutself, + .send_IPI_all = uv_send_IPI_all, + .send_IPI_self = uv_send_IPI_self, + + .wakeup_cpu = NULL, + .trampoline_phys_low = DEFAULT_TRAMPOLINE_PHYS_LOW, + .trampoline_phys_high = DEFAULT_TRAMPOLINE_PHYS_HIGH, + .wait_for_init_deassert = NULL, + .smp_callin_clear_local_apic = NULL, + .inquire_remote_apic = NULL, + + .read = native_apic_msr_read, + .write = native_apic_msr_write, + .icr_read = native_x2apic_icr_read, + .icr_write = native_x2apic_icr_write, + .wait_icr_idle = native_x2apic_wait_icr_idle, + .safe_wait_icr_idle = native_safe_x2apic_wait_icr_idle, +}; + +static __cpuinit void set_x2apic_extra_bits(int pnode) +{ + __get_cpu_var(x2apic_extra_bits) = (pnode << 6); +} + +/* + * Called on boot cpu. + */ +static __init int boot_pnode_to_blade(int pnode) +{ + int blade; + + for (blade = 0; blade < uv_num_possible_blades(); blade++) + if (pnode == uv_blade_info[blade].pnode) + return blade; + BUG(); +} + +struct redir_addr { + unsigned long redirect; + unsigned long alias; +}; + +#define DEST_SHIFT UVH_RH_GAM_ALIAS210_REDIRECT_CONFIG_0_MMR_DEST_BASE_SHFT + +static __initdata struct redir_addr redir_addrs[] = { + {UVH_RH_GAM_ALIAS210_REDIRECT_CONFIG_0_MMR, UVH_SI_ALIAS0_OVERLAY_CONFIG}, + {UVH_RH_GAM_ALIAS210_REDIRECT_CONFIG_1_MMR, UVH_SI_ALIAS1_OVERLAY_CONFIG}, + {UVH_RH_GAM_ALIAS210_REDIRECT_CONFIG_2_MMR, UVH_SI_ALIAS2_OVERLAY_CONFIG}, +}; + +static __init void get_lowmem_redirect(unsigned long *base, unsigned long *size) +{ + union uvh_si_alias0_overlay_config_u alias; + union uvh_rh_gam_alias210_redirect_config_2_mmr_u redirect; + int i; + + for (i = 0; i < ARRAY_SIZE(redir_addrs); i++) { + alias.v = uv_read_local_mmr(redir_addrs[i].alias); + if (alias.s.base == 0) { + *size = (1UL << alias.s.m_alias); + redirect.v = uv_read_local_mmr(redir_addrs[i].redirect); + *base = (unsigned long)redirect.s.dest_base << DEST_SHIFT; + return; + } + } + BUG(); +} + +static __init void map_low_mmrs(void) +{ + init_extra_mapping_uc(UV_GLOBAL_MMR32_BASE, UV_GLOBAL_MMR32_SIZE); + init_extra_mapping_uc(UV_LOCAL_MMR_BASE, UV_LOCAL_MMR_SIZE); +} + +enum map_type {map_wb, map_uc}; + +static __init void map_high(char *id, unsigned long base, int shift, + int max_pnode, enum map_type map_type) +{ + unsigned long bytes, paddr; + + paddr = base << shift; + bytes = (1UL << shift) * (max_pnode + 1); + printk(KERN_INFO "UV: Map %s_HI 0x%lx - 0x%lx\n", id, paddr, + paddr + bytes); + if (map_type == map_uc) + init_extra_mapping_uc(paddr, bytes); + else + init_extra_mapping_wb(paddr, bytes); + +} +static __init void map_gru_high(int max_pnode) +{ + union uvh_rh_gam_gru_overlay_config_mmr_u gru; + int shift = UVH_RH_GAM_GRU_OVERLAY_CONFIG_MMR_BASE_SHFT; + + gru.v = uv_read_local_mmr(UVH_RH_GAM_GRU_OVERLAY_CONFIG_MMR); + if (gru.s.enable) + map_high("GRU", gru.s.base, shift, max_pnode, map_wb); +} + +static __init void map_config_high(int max_pnode) +{ + union uvh_rh_gam_cfg_overlay_config_mmr_u cfg; + int shift = UVH_RH_GAM_CFG_OVERLAY_CONFIG_MMR_BASE_SHFT; + + cfg.v = uv_read_local_mmr(UVH_RH_GAM_CFG_OVERLAY_CONFIG_MMR); + if (cfg.s.enable) + map_high("CONFIG", cfg.s.base, shift, max_pnode, map_uc); +} + +static __init void map_mmr_high(int max_pnode) +{ + union uvh_rh_gam_mmr_overlay_config_mmr_u mmr; + int shift = UVH_RH_GAM_MMR_OVERLAY_CONFIG_MMR_BASE_SHFT; + + mmr.v = uv_read_local_mmr(UVH_RH_GAM_MMR_OVERLAY_CONFIG_MMR); + if (mmr.s.enable) + map_high("MMR", mmr.s.base, shift, max_pnode, map_uc); +} + +static __init void map_mmioh_high(int max_pnode) +{ + union uvh_rh_gam_mmioh_overlay_config_mmr_u mmioh; + int shift = UVH_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR_BASE_SHFT; + + mmioh.v = uv_read_local_mmr(UVH_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR); + if (mmioh.s.enable) + map_high("MMIOH", mmioh.s.base, shift, max_pnode, map_uc); +} + +static __init void uv_rtc_init(void) +{ + long status; + u64 ticks_per_sec; + + status = uv_bios_freq_base(BIOS_FREQ_BASE_REALTIME_CLOCK, + &ticks_per_sec); + if (status != BIOS_STATUS_SUCCESS || ticks_per_sec < 100000) { + printk(KERN_WARNING + "unable to determine platform RTC clock frequency, " + "guessing.\n"); + /* BIOS gives wrong value for clock freq. so guess */ + sn_rtc_cycles_per_second = 1000000000000UL / 30000UL; + } else + sn_rtc_cycles_per_second = ticks_per_sec; +} + +/* + * percpu heartbeat timer + */ +static void uv_heartbeat(unsigned long ignored) +{ + struct timer_list *timer = &uv_hub_info->scir.timer; + unsigned char bits = uv_hub_info->scir.state; + + /* flip heartbeat bit */ + bits ^= SCIR_CPU_HEARTBEAT; + + /* is this cpu idle? */ + if (idle_cpu(raw_smp_processor_id())) + bits &= ~SCIR_CPU_ACTIVITY; + else + bits |= SCIR_CPU_ACTIVITY; + + /* update system controller interface reg */ + uv_set_scir_bits(bits); + + /* enable next timer period */ + mod_timer(timer, jiffies + SCIR_CPU_HB_INTERVAL); +} + +static void __cpuinit uv_heartbeat_enable(int cpu) +{ + if (!uv_cpu_hub_info(cpu)->scir.enabled) { + struct timer_list *timer = &uv_cpu_hub_info(cpu)->scir.timer; + + uv_set_cpu_scir_bits(cpu, SCIR_CPU_HEARTBEAT|SCIR_CPU_ACTIVITY); + setup_timer(timer, uv_heartbeat, cpu); + timer->expires = jiffies + SCIR_CPU_HB_INTERVAL; + add_timer_on(timer, cpu); + uv_cpu_hub_info(cpu)->scir.enabled = 1; + } + + /* check boot cpu */ + if (!uv_cpu_hub_info(0)->scir.enabled) + uv_heartbeat_enable(0); +} + +#ifdef CONFIG_HOTPLUG_CPU +static void __cpuinit uv_heartbeat_disable(int cpu) +{ + if (uv_cpu_hub_info(cpu)->scir.enabled) { + uv_cpu_hub_info(cpu)->scir.enabled = 0; + del_timer(&uv_cpu_hub_info(cpu)->scir.timer); + } + uv_set_cpu_scir_bits(cpu, 0xff); +} + +/* + * cpu hotplug notifier + */ +static __cpuinit int uv_scir_cpu_notify(struct notifier_block *self, + unsigned long action, void *hcpu) +{ + long cpu = (long)hcpu; + + switch (action) { + case CPU_ONLINE: + uv_heartbeat_enable(cpu); + break; + case CPU_DOWN_PREPARE: + uv_heartbeat_disable(cpu); + break; + default: + break; + } + return NOTIFY_OK; +} + +static __init void uv_scir_register_cpu_notifier(void) +{ + hotcpu_notifier(uv_scir_cpu_notify, 0); +} + +#else /* !CONFIG_HOTPLUG_CPU */ + +static __init void uv_scir_register_cpu_notifier(void) +{ +} + +static __init int uv_init_heartbeat(void) +{ + int cpu; + + if (is_uv_system()) + for_each_online_cpu(cpu) + uv_heartbeat_enable(cpu); + return 0; +} + +late_initcall(uv_init_heartbeat); + +#endif /* !CONFIG_HOTPLUG_CPU */ + +/* + * Called on each cpu to initialize the per_cpu UV data area. + * ZZZ hotplug not supported yet + */ +void __cpuinit uv_cpu_init(void) +{ + /* CPU 0 initilization will be done via uv_system_init. */ + if (!uv_blade_info) + return; + + uv_blade_info[uv_numa_blade_id()].nr_online_cpus++; + + if (get_uv_system_type() == UV_NON_UNIQUE_APIC) + set_x2apic_extra_bits(uv_hub_info->pnode); +} + + +void __init uv_system_init(void) +{ + union uvh_si_addr_map_config_u m_n_config; + union uvh_node_id_u node_id; + unsigned long gnode_upper, lowmem_redir_base, lowmem_redir_size; + int bytes, nid, cpu, lcpu, pnode, blade, i, j, m_val, n_val; + int max_pnode = 0; + unsigned long mmr_base, present; + + map_low_mmrs(); + + m_n_config.v = uv_read_local_mmr(UVH_SI_ADDR_MAP_CONFIG); + m_val = m_n_config.s.m_skt; + n_val = m_n_config.s.n_skt; + mmr_base = + uv_read_local_mmr(UVH_RH_GAM_MMR_OVERLAY_CONFIG_MMR) & + ~UV_MMR_ENABLE; + printk(KERN_DEBUG "UV: global MMR base 0x%lx\n", mmr_base); + + for(i = 0; i < UVH_NODE_PRESENT_TABLE_DEPTH; i++) + uv_possible_blades += + hweight64(uv_read_local_mmr( UVH_NODE_PRESENT_TABLE + i * 8)); + printk(KERN_DEBUG "UV: Found %d blades\n", uv_num_possible_blades()); + + bytes = sizeof(struct uv_blade_info) * uv_num_possible_blades(); + uv_blade_info = kmalloc(bytes, GFP_KERNEL); + + get_lowmem_redirect(&lowmem_redir_base, &lowmem_redir_size); + + bytes = sizeof(uv_node_to_blade[0]) * num_possible_nodes(); + uv_node_to_blade = kmalloc(bytes, GFP_KERNEL); + memset(uv_node_to_blade, 255, bytes); + + bytes = sizeof(uv_cpu_to_blade[0]) * num_possible_cpus(); + uv_cpu_to_blade = kmalloc(bytes, GFP_KERNEL); + memset(uv_cpu_to_blade, 255, bytes); + + blade = 0; + for (i = 0; i < UVH_NODE_PRESENT_TABLE_DEPTH; i++) { + present = uv_read_local_mmr(UVH_NODE_PRESENT_TABLE + i * 8); + for (j = 0; j < 64; j++) { + if (!test_bit(j, &present)) + continue; + uv_blade_info[blade].pnode = (i * 64 + j); + uv_blade_info[blade].nr_possible_cpus = 0; + uv_blade_info[blade].nr_online_cpus = 0; + blade++; + } + } + + node_id.v = uv_read_local_mmr(UVH_NODE_ID); + gnode_upper = (((unsigned long)node_id.s.node_id) & + ~((1 << n_val) - 1)) << m_val; + + uv_bios_init(); + uv_bios_get_sn_info(0, &uv_type, &sn_partition_id, + &sn_coherency_id, &sn_region_size); + uv_rtc_init(); + + for_each_present_cpu(cpu) { + nid = cpu_to_node(cpu); + pnode = uv_apicid_to_pnode(per_cpu(x86_cpu_to_apicid, cpu)); + blade = boot_pnode_to_blade(pnode); + lcpu = uv_blade_info[blade].nr_possible_cpus; + uv_blade_info[blade].nr_possible_cpus++; + + uv_cpu_hub_info(cpu)->lowmem_remap_base = lowmem_redir_base; + uv_cpu_hub_info(cpu)->lowmem_remap_top = lowmem_redir_size; + uv_cpu_hub_info(cpu)->m_val = m_val; + uv_cpu_hub_info(cpu)->n_val = m_val; + uv_cpu_hub_info(cpu)->numa_blade_id = blade; + uv_cpu_hub_info(cpu)->blade_processor_id = lcpu; + uv_cpu_hub_info(cpu)->pnode = pnode; + uv_cpu_hub_info(cpu)->pnode_mask = (1 << n_val) - 1; + uv_cpu_hub_info(cpu)->gpa_mask = (1 << (m_val + n_val)) - 1; + uv_cpu_hub_info(cpu)->gnode_upper = gnode_upper; + uv_cpu_hub_info(cpu)->global_mmr_base = mmr_base; + uv_cpu_hub_info(cpu)->coherency_domain_number = sn_coherency_id; + uv_cpu_hub_info(cpu)->scir.offset = SCIR_LOCAL_MMR_BASE + lcpu; + uv_node_to_blade[nid] = blade; + uv_cpu_to_blade[cpu] = blade; + max_pnode = max(pnode, max_pnode); + + printk(KERN_DEBUG "UV: cpu %d, apicid 0x%x, pnode %d, nid %d, " + "lcpu %d, blade %d\n", + cpu, per_cpu(x86_cpu_to_apicid, cpu), pnode, nid, + lcpu, blade); + } + + map_gru_high(max_pnode); + map_mmr_high(max_pnode); + map_config_high(max_pnode); + map_mmioh_high(max_pnode); + + uv_cpu_init(); + uv_scir_register_cpu_notifier(); + proc_mkdir("sgi_uv", NULL); +} |
