summaryrefslogtreecommitdiffstats
path: root/arch/x86/kernel/apic
diff options
context:
space:
mode:
authorDaniel Baumann <daniel.baumann@progress-linux.org>2024-04-11 08:27:49 +0000
committerDaniel Baumann <daniel.baumann@progress-linux.org>2024-04-11 08:27:49 +0000
commitace9429bb58fd418f0c81d4c2835699bddf6bde6 (patch)
treeb2d64bc10158fdd5497876388cd68142ca374ed3 /arch/x86/kernel/apic
parentInitial commit. (diff)
downloadlinux-ace9429bb58fd418f0c81d4c2835699bddf6bde6.tar.xz
linux-ace9429bb58fd418f0c81d4c2835699bddf6bde6.zip
Adding upstream version 6.6.15.upstream/6.6.15
Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>
Diffstat (limited to 'arch/x86/kernel/apic')
-rw-r--r--arch/x86/kernel/apic/Makefile30
-rw-r--r--arch/x86/kernel/apic/apic.c2877
-rw-r--r--arch/x86/kernel/apic/apic_common.c57
-rw-r--r--arch/x86/kernel/apic/apic_flat_64.c191
-rw-r--r--arch/x86/kernel/apic/apic_noop.c78
-rw-r--r--arch/x86/kernel/apic/apic_numachip.c292
-rw-r--r--arch/x86/kernel/apic/bigsmp_32.c126
-rw-r--r--arch/x86/kernel/apic/hw_nmi.c61
-rw-r--r--arch/x86/kernel/apic/init.c110
-rw-r--r--arch/x86/kernel/apic/io_apic.c3116
-rw-r--r--arch/x86/kernel/apic/ipi.c311
-rw-r--r--arch/x86/kernel/apic/local.h85
-rw-r--r--arch/x86/kernel/apic/msi.c395
-rw-r--r--arch/x86/kernel/apic/probe_32.c140
-rw-r--r--arch/x86/kernel/apic/probe_64.c42
-rw-r--r--arch/x86/kernel/apic/vector.c1394
-rw-r--r--arch/x86/kernel/apic/x2apic_cluster.c262
-rw-r--r--arch/x86/kernel/apic/x2apic_phys.c177
-rw-r--r--arch/x86/kernel/apic/x2apic_uv_x.c1872
19 files changed, 11616 insertions, 0 deletions
diff --git a/arch/x86/kernel/apic/Makefile b/arch/x86/kernel/apic/Makefile
new file mode 100644
index 0000000000..2ee867d796
--- /dev/null
+++ b/arch/x86/kernel/apic/Makefile
@@ -0,0 +1,30 @@
+# SPDX-License-Identifier: GPL-2.0
+#
+# Makefile for local APIC drivers and for the IO-APIC code
+#
+
+# Leads to non-deterministic coverage that is not a function of syscall inputs.
+# In particualr, smp_apic_timer_interrupt() is called in random places.
+KCOV_INSTRUMENT := n
+
+obj-$(CONFIG_X86_LOCAL_APIC) += apic.o apic_common.o apic_noop.o ipi.o vector.o init.o
+obj-y += hw_nmi.o
+
+obj-$(CONFIG_X86_IO_APIC) += io_apic.o
+obj-$(CONFIG_PCI_MSI) += msi.o
+obj-$(CONFIG_SMP) += ipi.o
+
+ifeq ($(CONFIG_X86_64),y)
+# APIC probe will depend on the listing order here
+obj-$(CONFIG_X86_NUMACHIP) += apic_numachip.o
+obj-$(CONFIG_X86_UV) += x2apic_uv_x.o
+obj-$(CONFIG_X86_X2APIC) += x2apic_phys.o
+obj-$(CONFIG_X86_X2APIC) += x2apic_cluster.o
+obj-y += apic_flat_64.o
+endif
+
+# APIC probe will depend on the listing order here
+obj-$(CONFIG_X86_BIGSMP) += bigsmp_32.o
+
+# For 32bit, probe_32 need to be listed last
+obj-$(CONFIG_X86_LOCAL_APIC) += probe_$(BITS).o
diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c
new file mode 100644
index 0000000000..3cdf484935
--- /dev/null
+++ b/arch/x86/kernel/apic/apic.c
@@ -0,0 +1,2877 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * 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/perf_event.h>
+#include <linux/kernel_stat.h>
+#include <linux/mc146818rtc.h>
+#include <linux/acpi_pmtmr.h>
+#include <linux/clockchips.h>
+#include <linux/interrupt.h>
+#include <linux/memblock.h>
+#include <linux/ftrace.h>
+#include <linux/ioport.h>
+#include <linux/export.h>
+#include <linux/syscore_ops.h>
+#include <linux/delay.h>
+#include <linux/timex.h>
+#include <linux/i8253.h>
+#include <linux/dmar.h>
+#include <linux/init.h>
+#include <linux/cpu.h>
+#include <linux/dmi.h>
+#include <linux/smp.h>
+#include <linux/mm.h>
+
+#include <xen/xen.h>
+
+#include <asm/trace/irq_vectors.h>
+#include <asm/irq_remapping.h>
+#include <asm/pc-conf-reg.h>
+#include <asm/perf_event.h>
+#include <asm/x86_init.h>
+#include <linux/atomic.h>
+#include <asm/barrier.h>
+#include <asm/mpspec.h>
+#include <asm/i8259.h>
+#include <asm/proto.h>
+#include <asm/traps.h>
+#include <asm/apic.h>
+#include <asm/acpi.h>
+#include <asm/io_apic.h>
+#include <asm/desc.h>
+#include <asm/hpet.h>
+#include <asm/mtrr.h>
+#include <asm/time.h>
+#include <asm/smp.h>
+#include <asm/mce.h>
+#include <asm/tsc.h>
+#include <asm/hypervisor.h>
+#include <asm/cpu_device_id.h>
+#include <asm/intel-family.h>
+#include <asm/irq_regs.h>
+#include <asm/cpu.h>
+
+#include "local.h"
+
+unsigned int num_processors;
+
+unsigned disabled_cpus;
+
+/* Processor that is doing the boot up */
+unsigned int boot_cpu_physical_apicid __ro_after_init = -1U;
+EXPORT_SYMBOL_GPL(boot_cpu_physical_apicid);
+
+u8 boot_cpu_apic_version __ro_after_init;
+
+/*
+ * Bitmask of physically existing CPUs:
+ */
+physid_mask_t phys_cpu_present_map;
+
+/*
+ * Processor to be disabled specified by kernel parameter
+ * disable_cpu_apicid=<int>, mostly used for the kdump 2nd kernel to
+ * avoid undefined behaviour caused by sending INIT from AP to BSP.
+ */
+static unsigned int disabled_cpu_apicid __ro_after_init = BAD_APICID;
+
+/*
+ * This variable controls which CPUs receive external NMIs. By default,
+ * external NMIs are delivered only to the BSP.
+ */
+static int apic_extnmi __ro_after_init = APIC_EXTNMI_BSP;
+
+/*
+ * Hypervisor supports 15 bits of APIC ID in MSI Extended Destination ID
+ */
+static bool virt_ext_dest_id __ro_after_init;
+
+/* For parallel bootup. */
+unsigned long apic_mmio_base __ro_after_init;
+
+static inline bool apic_accessible(void)
+{
+ return x2apic_mode || apic_mmio_base;
+}
+
+/*
+ * Map cpu index to physical APIC ID
+ */
+DEFINE_EARLY_PER_CPU_READ_MOSTLY(u16, x86_cpu_to_apicid, BAD_APICID);
+DEFINE_EARLY_PER_CPU_READ_MOSTLY(u32, x86_cpu_to_acpiid, U32_MAX);
+EXPORT_EARLY_PER_CPU_SYMBOL(x86_cpu_to_apicid);
+EXPORT_EARLY_PER_CPU_SYMBOL(x86_cpu_to_acpiid);
+
+#ifdef CONFIG_X86_32
+/* Local APIC was disabled by the BIOS and enabled by the kernel */
+static int enabled_via_apicbase __ro_after_init;
+
+/*
+ * Handle interrupt mode configuration register (IMCR).
+ * This register controls whether the interrupt signals
+ * that reach the BSP come from the master PIC or from the
+ * local APIC. Before entering Symmetric I/O Mode, either
+ * the BIOS or the operating system must switch out of
+ * PIC Mode by changing the IMCR.
+ */
+static inline void imcr_pic_to_apic(void)
+{
+ /* NMI and 8259 INTR go through APIC */
+ pc_conf_set(PC_CONF_MPS_IMCR, 0x01);
+}
+
+static inline void imcr_apic_to_pic(void)
+{
+ /* NMI and 8259 INTR go directly to BSP */
+ pc_conf_set(PC_CONF_MPS_IMCR, 0x00);
+}
+#endif
+
+/*
+ * Knob to control our willingness to enable the local APIC.
+ *
+ * +1=force-enable
+ */
+static int force_enable_local_apic __initdata;
+
+/*
+ * APIC command line parameters
+ */
+static int __init parse_lapic(char *arg)
+{
+ if (IS_ENABLED(CONFIG_X86_32) && !arg)
+ force_enable_local_apic = 1;
+ else if (arg && !strncmp(arg, "notscdeadline", 13))
+ setup_clear_cpu_cap(X86_FEATURE_TSC_DEADLINE_TIMER);
+ return 0;
+}
+early_param("lapic", parse_lapic);
+
+#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 1;
+}
+__setup("apicpmtimer", setup_apicpmtimer);
+#endif
+
+static unsigned long mp_lapic_addr __ro_after_init;
+bool apic_is_disabled __ro_after_init;
+/* Disable local APIC timer from the kernel commandline or via dmi quirk */
+static int disable_apic_timer __initdata;
+/* Local APIC timer works in C2 */
+int local_apic_timer_c2_ok __ro_after_init;
+EXPORT_SYMBOL_GPL(local_apic_timer_c2_ok);
+
+/*
+ * Debug level, exported for io_apic.c
+ */
+int apic_verbosity __ro_after_init;
+
+int pic_mode __ro_after_init;
+
+/* Have we found an MP table */
+int smp_found_config __ro_after_init;
+
+static struct resource lapic_resource = {
+ .name = "Local APIC",
+ .flags = IORESOURCE_MEM | IORESOURCE_BUSY,
+};
+
+unsigned int lapic_timer_period = 0;
+
+static void apic_pm_activate(void);
+
+/*
+ * 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)
+{
+ return APIC_INTEGRATED(lapic_get_version());
+}
+
+/*
+ * 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;
+
+ /* Hygon systems use modern APIC */
+ if (boot_cpu_data.x86_vendor == X86_VENDOR_HYGON)
+ return 1;
+
+ return lapic_get_version() >= 0x14;
+}
+
+/*
+ * right after this call apic become NOOP driven
+ * so apic->write/read doesn't do anything
+ */
+static void __init apic_disable(void)
+{
+ apic_install_driver(&apic_noop);
+}
+
+void native_apic_icr_write(u32 low, u32 id)
+{
+ unsigned long flags;
+
+ local_irq_save(flags);
+ apic_write(APIC_ICR2, SET_XAPIC_DEST_FIELD(id));
+ apic_write(APIC_ICR, low);
+ local_irq_restore(flags);
+}
+
+u64 native_apic_icr_read(void)
+{
+ u32 icr1, icr2;
+
+ icr2 = apic_read(APIC_ICR2);
+ icr1 = apic_read(APIC_ICR);
+
+ return icr1 | ((u64)icr2 << 32);
+}
+
+#ifdef CONFIG_X86_32
+/**
+ * get_physical_broadcast - Get number of physical broadcast IDs
+ */
+int get_physical_broadcast(void)
+{
+ return modern_apic() ? 0xff : 0xf;
+}
+#endif
+
+/**
+ * lapic_get_maxlvt - get the maximum number of local vector table entries
+ */
+int lapic_get_maxlvt(void)
+{
+ /*
+ * - we always have APIC integrated on 64bit mode
+ * - 82489DXs do not report # of LVT entries
+ */
+ return lapic_is_integrated() ? GET_APIC_MAXLVT(apic_read(APIC_LVR)) : 2;
+}
+
+/*
+ * Local APIC timer
+ */
+
+/* Clock divisor */
+#define APIC_DIVISOR 16
+#define TSC_DIVISOR 8
+
+/* i82489DX specific */
+#define I82489DX_BASE_DIVIDER (((0x2) << 18))
+
+/*
+ * This function sets up the local APIC timer, with a timeout of
+ * 'clocks' APIC bus clock. During calibration we actually call
+ * 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;
+ else if (boot_cpu_has(X86_FEATURE_TSC_DEADLINE_TIMER))
+ lvtt_value |= APIC_LVT_TIMER_TSCDEADLINE;
+
+ /*
+ * The i82489DX APIC uses bit 18 and 19 for the base divider. This
+ * overlaps with bit 18 on integrated APICs, but is not documented
+ * in the SDM. No problem though. i82489DX equipped systems do not
+ * have TSC deadline timer.
+ */
+ if (!lapic_is_integrated())
+ lvtt_value |= I82489DX_BASE_DIVIDER;
+
+ if (!irqen)
+ lvtt_value |= APIC_LVT_MASKED;
+
+ apic_write(APIC_LVTT, lvtt_value);
+
+ if (lvtt_value & APIC_LVT_TIMER_TSCDEADLINE) {
+ /*
+ * See Intel SDM: TSC-Deadline Mode chapter. In xAPIC mode,
+ * writing to the APIC LVTT and TSC_DEADLINE MSR isn't serialized.
+ * According to Intel, MFENCE can do the serialization here.
+ */
+ asm volatile("mfence" : : : "memory");
+ return;
+ }
+
+ /*
+ * 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
+ *
+ * Software should use the LVT offsets the BIOS provides. The offsets
+ * are determined by the subsystems using it like those for MCE
+ * threshold or IBS. On K8 only offset 0 (APIC500) and MCE interrupts
+ * are supported. Beginning with family 10h at least 4 offsets are
+ * available.
+ *
+ * Since the offsets must be consistent for all cores, we keep track
+ * of the LVT offsets in software and reserve the offset for the same
+ * vector also to be used on other cores. An offset is freed by
+ * setting the entry to APIC_EILVT_MASKED.
+ *
+ * If the BIOS is right, there should be no conflicts. Otherwise a
+ * "[Firmware Bug]: ..." error message is generated. However, if
+ * software does not properly determines the offsets, it is not
+ * necessarily a BIOS bug.
+ */
+
+static atomic_t eilvt_offsets[APIC_EILVT_NR_MAX];
+
+static inline int eilvt_entry_is_changeable(unsigned int old, unsigned int new)
+{
+ return (old & APIC_EILVT_MASKED)
+ || (new == APIC_EILVT_MASKED)
+ || ((new & ~APIC_EILVT_MASKED) == old);
+}
+
+static unsigned int reserve_eilvt_offset(int offset, unsigned int new)
+{
+ unsigned int rsvd, vector;
+
+ if (offset >= APIC_EILVT_NR_MAX)
+ return ~0;
+
+ rsvd = atomic_read(&eilvt_offsets[offset]);
+ do {
+ vector = rsvd & ~APIC_EILVT_MASKED; /* 0: unassigned */
+ if (vector && !eilvt_entry_is_changeable(vector, new))
+ /* may not change if vectors are different */
+ return rsvd;
+ } while (!atomic_try_cmpxchg(&eilvt_offsets[offset], &rsvd, new));
+
+ rsvd = new & ~APIC_EILVT_MASKED;
+ if (rsvd && rsvd != vector)
+ pr_info("LVT offset %d assigned for vector 0x%02x\n",
+ offset, rsvd);
+
+ return new;
+}
+
+/*
+ * If mask=1, the LVT entry does not generate interrupts while mask=0
+ * enables the vector. See also the BKDGs. Must be called with
+ * preemption disabled.
+ */
+
+int setup_APIC_eilvt(u8 offset, u8 vector, u8 msg_type, u8 mask)
+{
+ unsigned long reg = APIC_EILVTn(offset);
+ unsigned int new, old, reserved;
+
+ new = (mask << 16) | (msg_type << 8) | vector;
+ old = apic_read(reg);
+ reserved = reserve_eilvt_offset(offset, new);
+
+ if (reserved != new) {
+ pr_err(FW_BUG "cpu %d, try to use APIC%lX (LVT offset %d) for "
+ "vector 0x%x, but the register is already in use for "
+ "vector 0x%x on another cpu\n",
+ smp_processor_id(), reg, offset, new, reserved);
+ return -EINVAL;
+ }
+
+ if (!eilvt_entry_is_changeable(old, new)) {
+ pr_err(FW_BUG "cpu %d, try to use APIC%lX (LVT offset %d) for "
+ "vector 0x%x, but the register is already in use for "
+ "vector 0x%x on this cpu\n",
+ smp_processor_id(), reg, offset, new, old);
+ return -EBUSY;
+ }
+
+ apic_write(reg, new);
+
+ return 0;
+}
+EXPORT_SYMBOL_GPL(setup_APIC_eilvt);
+
+/*
+ * 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;
+}
+
+static int lapic_next_deadline(unsigned long delta,
+ struct clock_event_device *evt)
+{
+ u64 tsc;
+
+ /* This MSR is special and need a special fence: */
+ weak_wrmsr_fence();
+
+ tsc = rdtsc();
+ wrmsrl(MSR_IA32_TSC_DEADLINE, tsc + (((u64) delta) * TSC_DIVISOR));
+ return 0;
+}
+
+static int lapic_timer_shutdown(struct clock_event_device *evt)
+{
+ unsigned int v;
+
+ /* Lapic used as dummy for broadcast ? */
+ if (evt->features & CLOCK_EVT_FEAT_DUMMY)
+ return 0;
+
+ v = apic_read(APIC_LVTT);
+ v |= (APIC_LVT_MASKED | LOCAL_TIMER_VECTOR);
+ apic_write(APIC_LVTT, v);
+ apic_write(APIC_TMICT, 0);
+ return 0;
+}
+
+static inline int
+lapic_timer_set_periodic_oneshot(struct clock_event_device *evt, bool oneshot)
+{
+ /* Lapic used as dummy for broadcast ? */
+ if (evt->features & CLOCK_EVT_FEAT_DUMMY)
+ return 0;
+
+ __setup_APIC_LVTT(lapic_timer_period, oneshot, 1);
+ return 0;
+}
+
+static int lapic_timer_set_periodic(struct clock_event_device *evt)
+{
+ return lapic_timer_set_periodic_oneshot(evt, false);
+}
+
+static int lapic_timer_set_oneshot(struct clock_event_device *evt)
+{
+ return lapic_timer_set_periodic_oneshot(evt, true);
+}
+
+/*
+ * 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
+}
+
+
+/*
+ * 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_state_shutdown = lapic_timer_shutdown,
+ .set_state_periodic = lapic_timer_set_periodic,
+ .set_state_oneshot = lapic_timer_set_oneshot,
+ .set_state_oneshot_stopped = lapic_timer_shutdown,
+ .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 const struct x86_cpu_id deadline_match[] __initconst = {
+ X86_MATCH_INTEL_FAM6_MODEL_STEPPINGS(HASWELL_X, X86_STEPPINGS(0x2, 0x2), 0x3a), /* EP */
+ X86_MATCH_INTEL_FAM6_MODEL_STEPPINGS(HASWELL_X, X86_STEPPINGS(0x4, 0x4), 0x0f), /* EX */
+
+ X86_MATCH_INTEL_FAM6_MODEL( BROADWELL_X, 0x0b000020),
+
+ X86_MATCH_INTEL_FAM6_MODEL_STEPPINGS(BROADWELL_D, X86_STEPPINGS(0x2, 0x2), 0x00000011),
+ X86_MATCH_INTEL_FAM6_MODEL_STEPPINGS(BROADWELL_D, X86_STEPPINGS(0x3, 0x3), 0x0700000e),
+ X86_MATCH_INTEL_FAM6_MODEL_STEPPINGS(BROADWELL_D, X86_STEPPINGS(0x4, 0x4), 0x0f00000c),
+ X86_MATCH_INTEL_FAM6_MODEL_STEPPINGS(BROADWELL_D, X86_STEPPINGS(0x5, 0x5), 0x0e000003),
+
+ X86_MATCH_INTEL_FAM6_MODEL_STEPPINGS(SKYLAKE_X, X86_STEPPINGS(0x3, 0x3), 0x01000136),
+ X86_MATCH_INTEL_FAM6_MODEL_STEPPINGS(SKYLAKE_X, X86_STEPPINGS(0x4, 0x4), 0x02000014),
+ X86_MATCH_INTEL_FAM6_MODEL_STEPPINGS(SKYLAKE_X, X86_STEPPINGS(0x5, 0xf), 0),
+
+ X86_MATCH_INTEL_FAM6_MODEL( HASWELL, 0x22),
+ X86_MATCH_INTEL_FAM6_MODEL( HASWELL_L, 0x20),
+ X86_MATCH_INTEL_FAM6_MODEL( HASWELL_G, 0x17),
+
+ X86_MATCH_INTEL_FAM6_MODEL( BROADWELL, 0x25),
+ X86_MATCH_INTEL_FAM6_MODEL( BROADWELL_G, 0x17),
+
+ X86_MATCH_INTEL_FAM6_MODEL( SKYLAKE_L, 0xb2),
+ X86_MATCH_INTEL_FAM6_MODEL( SKYLAKE, 0xb2),
+
+ X86_MATCH_INTEL_FAM6_MODEL( KABYLAKE_L, 0x52),
+ X86_MATCH_INTEL_FAM6_MODEL( KABYLAKE, 0x52),
+
+ {},
+};
+
+static __init bool apic_validate_deadline_timer(void)
+{
+ const struct x86_cpu_id *m;
+ u32 rev;
+
+ if (!boot_cpu_has(X86_FEATURE_TSC_DEADLINE_TIMER))
+ return false;
+ if (boot_cpu_has(X86_FEATURE_HYPERVISOR))
+ return true;
+
+ m = x86_match_cpu(deadline_match);
+ if (!m)
+ return true;
+
+ rev = (u32)m->driver_data;
+
+ if (boot_cpu_data.microcode >= rev)
+ return true;
+
+ setup_clear_cpu_cap(X86_FEATURE_TSC_DEADLINE_TIMER);
+ pr_err(FW_BUG "TSC_DEADLINE disabled due to Errata; "
+ "please update microcode to version: 0x%x (or later)\n", rev);
+ return false;
+}
+
+/*
+ * Setup the local APIC timer for this CPU. Copy the initialized values
+ * of the boot CPU and register the clock event in the framework.
+ */
+static void setup_APIC_timer(void)
+{
+ struct clock_event_device *levt = this_cpu_ptr(&lapic_events);
+
+ if (this_cpu_has(X86_FEATURE_ARAT)) {
+ lapic_clockevent.features &= ~CLOCK_EVT_FEAT_C3STOP;
+ /* Make LAPIC timer preferable over percpu HPET */
+ lapic_clockevent.rating = 150;
+ }
+
+ memcpy(levt, &lapic_clockevent, sizeof(*levt));
+ levt->cpumask = cpumask_of(smp_processor_id());
+
+ if (this_cpu_has(X86_FEATURE_TSC_DEADLINE_TIMER)) {
+ levt->name = "lapic-deadline";
+ levt->features &= ~(CLOCK_EVT_FEAT_PERIODIC |
+ CLOCK_EVT_FEAT_DUMMY);
+ levt->set_next_event = lapic_next_deadline;
+ clockevents_config_and_register(levt,
+ tsc_khz * (1000 / TSC_DIVISOR),
+ 0xF, ~0UL);
+ } else
+ clockevents_register_device(levt);
+}
+
+/*
+ * Install the updated TSC frequency from recalibration at the TSC
+ * deadline clockevent devices.
+ */
+static void __lapic_update_tsc_freq(void *info)
+{
+ struct clock_event_device *levt = this_cpu_ptr(&lapic_events);
+
+ if (!this_cpu_has(X86_FEATURE_TSC_DEADLINE_TIMER))
+ return;
+
+ clockevents_update_freq(levt, tsc_khz * (1000 / TSC_DIVISOR));
+}
+
+void lapic_update_tsc_freq(void)
+{
+ /*
+ * The clockevent device's ->mult and ->shift can both be
+ * changed. In order to avoid races, schedule the frequency
+ * update code on each CPU.
+ */
+ on_each_cpu(__lapic_update_tsc_freq, NULL, 0);
+}
+
+/*
+ * 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 synchronous. CPUs connected by the same APIC bus have the very same bus
+ * frequency.
+ *
+ * This was previously done by reading the PIT/HPET and waiting for a wrap
+ * 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 and polled calibration function.
+ */
+static void __init lapic_cal_handler(struct clock_event_device *dev)
+{
+ unsigned long long tsc = 0;
+ long tapic = apic_read(APIC_TMCCT);
+ unsigned long pm = acpi_pm_read_early();
+
+ if (boot_cpu_has(X86_FEATURE_TSC))
+ tsc = rdtsc();
+
+ 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_warn("APIC calibration not consistent "
+ "with PM-Timer: %ldms instead of 100ms\n", (long)res);
+
+ /* Correct the lapic counter value */
+ res = (((u64)(*delta)) * pm_100ms);
+ 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 (boot_cpu_has(X86_FEATURE_TSC)) {
+ res = (((u64)(*deltatsc)) * pm_100ms);
+ do_div(res, deltapm);
+ apic_printk(APIC_VERBOSE, "TSC delta adjusted to "
+ "PM-Timer: %lu (%ld)\n",
+ (unsigned long)res, *deltatsc);
+ *deltatsc = (long)res;
+ }
+
+ return 0;
+}
+
+static int __init lapic_init_clockevent(void)
+{
+ if (!lapic_timer_period)
+ return -1;
+
+ /* Calculate the scaled math multiplication factor */
+ lapic_clockevent.mult = div_sc(lapic_timer_period/APIC_DIVISOR,
+ TICK_NSEC, lapic_clockevent.shift);
+ lapic_clockevent.max_delta_ns =
+ clockevent_delta2ns(0x7FFFFFFF, &lapic_clockevent);
+ lapic_clockevent.max_delta_ticks = 0x7FFFFFFF;
+ lapic_clockevent.min_delta_ns =
+ clockevent_delta2ns(0xF, &lapic_clockevent);
+ lapic_clockevent.min_delta_ticks = 0xF;
+
+ return 0;
+}
+
+bool __init apic_needs_pit(void)
+{
+ /*
+ * If the frequencies are not known, PIT is required for both TSC
+ * and apic timer calibration.
+ */
+ if (!tsc_khz || !cpu_khz)
+ return true;
+
+ /* Is there an APIC at all or is it disabled? */
+ if (!boot_cpu_has(X86_FEATURE_APIC) || apic_is_disabled)
+ return true;
+
+ /*
+ * If interrupt delivery mode is legacy PIC or virtual wire without
+ * configuration, the local APIC timer wont be set up. Make sure
+ * that the PIT is initialized.
+ */
+ if (apic_intr_mode == APIC_PIC ||
+ apic_intr_mode == APIC_VIRTUAL_WIRE_NO_CONFIG)
+ return true;
+
+ /* Virt guests may lack ARAT, but still have DEADLINE */
+ if (!boot_cpu_has(X86_FEATURE_ARAT))
+ return true;
+
+ /* Deadline timer is based on TSC so no further PIT action required */
+ if (boot_cpu_has(X86_FEATURE_TSC_DEADLINE_TIMER))
+ return false;
+
+ /* APIC timer disabled? */
+ if (disable_apic_timer)
+ return true;
+ /*
+ * The APIC timer frequency is known already, no PIT calibration
+ * required. If unknown, let the PIT be initialized.
+ */
+ return lapic_timer_period == 0;
+}
+
+static int __init calibrate_APIC_clock(void)
+{
+ struct clock_event_device *levt = this_cpu_ptr(&lapic_events);
+ u64 tsc_perj = 0, tsc_start = 0;
+ unsigned long jif_start;
+ unsigned long deltaj;
+ long delta, deltatsc;
+ int pm_referenced = 0;
+
+ if (boot_cpu_has(X86_FEATURE_TSC_DEADLINE_TIMER))
+ return 0;
+
+ /*
+ * Check if lapic timer has already been calibrated by platform
+ * specific routine, such as tsc calibration code. If so just fill
+ * in the clockevent structure and return.
+ */
+ if (!lapic_init_clockevent()) {
+ apic_printk(APIC_VERBOSE, "lapic timer already calibrated %d\n",
+ lapic_timer_period);
+ /*
+ * Direct calibration methods must have an always running
+ * local APIC timer, no need for broadcast timer.
+ */
+ lapic_clockevent.features &= ~CLOCK_EVT_FEAT_DUMMY;
+ return 0;
+ }
+
+ apic_printk(APIC_VERBOSE, "Using local APIC timer interrupts.\n"
+ "calibrating APIC timer ...\n");
+
+ /*
+ * There are platforms w/o global clockevent devices. Instead of
+ * making the calibration conditional on that, use a polling based
+ * approach everywhere.
+ */
+ local_irq_disable();
+
+ /*
+ * 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);
+
+ /*
+ * Methods to terminate the calibration loop:
+ * 1) Global clockevent if available (jiffies)
+ * 2) TSC if available and frequency is known
+ */
+ jif_start = READ_ONCE(jiffies);
+
+ if (tsc_khz) {
+ tsc_start = rdtsc();
+ tsc_perj = div_u64((u64)tsc_khz * 1000, HZ);
+ }
+
+ /*
+ * Enable interrupts so the tick can fire, if a global
+ * clockevent device is available
+ */
+ local_irq_enable();
+
+ while (lapic_cal_loops <= LAPIC_CAL_LOOPS) {
+ /* Wait for a tick to elapse */
+ while (1) {
+ if (tsc_khz) {
+ u64 tsc_now = rdtsc();
+ if ((tsc_now - tsc_start) >= tsc_perj) {
+ tsc_start += tsc_perj;
+ break;
+ }
+ } else {
+ unsigned long jif_now = READ_ONCE(jiffies);
+
+ if (time_after(jif_now, jif_start)) {
+ jif_start = jif_now;
+ break;
+ }
+ }
+ cpu_relax();
+ }
+
+ /* Invoke the calibration routine */
+ local_irq_disable();
+ lapic_cal_handler(NULL);
+ local_irq_enable();
+ }
+
+ local_irq_disable();
+
+ /* Build delta t1-t2 as apic timer counts down */
+ delta = lapic_cal_t1 - lapic_cal_t2;
+ apic_printk(APIC_VERBOSE, "... lapic delta = %ld\n", delta);
+
+ 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);
+
+ lapic_timer_period = (delta * APIC_DIVISOR) / LAPIC_CAL_LOOPS;
+ lapic_init_clockevent();
+
+ apic_printk(APIC_VERBOSE, "..... delta %ld\n", delta);
+ apic_printk(APIC_VERBOSE, "..... mult: %u\n", lapic_clockevent.mult);
+ apic_printk(APIC_VERBOSE, "..... calibration result: %u\n",
+ lapic_timer_period);
+
+ if (boot_cpu_has(X86_FEATURE_TSC)) {
+ apic_printk(APIC_VERBOSE, "..... CPU clock speed is "
+ "%ld.%04ld MHz.\n",
+ (deltatsc / LAPIC_CAL_LOOPS) / (1000000 / HZ),
+ (deltatsc / LAPIC_CAL_LOOPS) % (1000000 / HZ));
+ }
+
+ apic_printk(APIC_VERBOSE, "..... host bus clock speed is "
+ "%u.%04u MHz.\n",
+ lapic_timer_period / (1000000 / HZ),
+ lapic_timer_period % (1000000 / HZ));
+
+ /*
+ * Do a sanity check on the APIC calibration result
+ */
+ if (lapic_timer_period < (1000000 / HZ)) {
+ local_irq_enable();
+ pr_warn("APIC frequency too slow, disabling apic timer\n");
+ return -1;
+ }
+
+ levt->features &= ~CLOCK_EVT_FEAT_DUMMY;
+
+ /*
+ * PM timer calibration failed or not turned on so lets try APIC
+ * timer based calibration, if a global clockevent device is
+ * available.
+ */
+ if (!pm_referenced && global_clock_event) {
+ apic_printk(APIC_VERBOSE, "... verify APIC timer\n");
+
+ /*
+ * Setup the apic timer manually
+ */
+ levt->event_handler = lapic_cal_handler;
+ lapic_timer_set_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 */
+ local_irq_disable();
+ lapic_timer_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;
+ }
+ local_irq_enable();
+
+ if (levt->features & CLOCK_EVT_FEAT_DUMMY) {
+ pr_warn("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;
+ }
+
+ 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.
+ */
+ lapic_clockevent.features &= ~CLOCK_EVT_FEAT_DUMMY;
+
+ /* Setup the lapic or request the broadcast */
+ setup_APIC_timer();
+ amd_e400_c1e_apic_setup();
+}
+
+void setup_secondary_APIC_clock(void)
+{
+ setup_APIC_timer();
+ amd_e400_c1e_apic_setup();
+}
+
+/*
+ * The guts of the apic timer interrupt
+ */
+static void local_apic_timer_interrupt(void)
+{
+ struct clock_event_device *evt = this_cpu_ptr(&lapic_events);
+
+ /*
+ * 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_warn("Spurious LAPIC timer interrupt on cpu %d\n",
+ smp_processor_id());
+ /* Switch it off */
+ lapic_timer_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 ... ]
+ */
+DEFINE_IDTENTRY_SYSVEC(sysvec_apic_timer_interrupt)
+{
+ struct pt_regs *old_regs = set_irq_regs(regs);
+
+ apic_eoi();
+ trace_local_timer_entry(LOCAL_TIMER_VECTOR);
+ local_apic_timer_interrupt();
+ trace_local_timer_exit(LOCAL_TIMER_VECTOR);
+
+ set_irq_regs(old_regs);
+}
+
+/*
+ * 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;
+
+ if (!apic_accessible())
+ 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 */
+#ifdef CONFIG_X86_THERMAL_VECTOR
+ if (maxlvt >= 5) {
+ v = apic_read(APIC_LVTTHMR);
+ apic_write(APIC_LVTTHMR, v | APIC_LVT_MASKED);
+ }
+#endif
+#ifdef CONFIG_X86_MCE_INTEL
+ if (maxlvt >= 6) {
+ v = apic_read(APIC_LVTCMCI);
+ if (!(v & APIC_LVT_MASKED))
+ apic_write(APIC_LVTCMCI, 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);
+ }
+}
+
+/**
+ * apic_soft_disable - Clears and software disables the local APIC on hotplug
+ *
+ * Contrary to disable_local_APIC() this does not touch the enable bit in
+ * MSR_IA32_APICBASE. Clearing that bit on systems based on the 3 wire APIC
+ * bus would require a hardware reset as the APIC would lose track of bus
+ * arbitration. On systems with FSB delivery APICBASE could be disabled,
+ * but it has to be guaranteed that no interrupt is sent to the APIC while
+ * in that state and it's not clear from the SDM whether it still responds
+ * to INIT/SIPI messages. Stay on the safe side and use software disable.
+ */
+void apic_soft_disable(void)
+{
+ u32 value;
+
+ clear_local_APIC();
+
+ /* Soft disable APIC (implies clearing of registers for 82489DX!). */
+ value = apic_read(APIC_SPIV);
+ value &= ~APIC_SPIV_APIC_ENABLED;
+ apic_write(APIC_SPIV, value);
+}
+
+/**
+ * disable_local_APIC - clear and disable the local APIC
+ */
+void disable_local_APIC(void)
+{
+ if (!apic_accessible())
+ return;
+
+ apic_soft_disable();
+
+#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 (!boot_cpu_has(X86_FEATURE_APIC) && !apic_from_smp_config())
+ 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);
+}
+
+/**
+ * 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);
+}
+
+enum apic_intr_mode_id apic_intr_mode __ro_after_init;
+
+static int __init __apic_intr_mode_select(void)
+{
+ /* Check kernel option */
+ if (apic_is_disabled) {
+ pr_info("APIC disabled via kernel command line\n");
+ return APIC_PIC;
+ }
+
+ /* Check BIOS */
+#ifdef CONFIG_X86_64
+ /* On 64-bit, the APIC must be integrated, Check local APIC only */
+ if (!boot_cpu_has(X86_FEATURE_APIC)) {
+ apic_is_disabled = true;
+ pr_info("APIC disabled by BIOS\n");
+ return APIC_PIC;
+ }
+#else
+ /* On 32-bit, the APIC may be integrated APIC or 82489DX */
+
+ /* Neither 82489DX nor integrated APIC ? */
+ if (!boot_cpu_has(X86_FEATURE_APIC) && !smp_found_config) {
+ apic_is_disabled = true;
+ return APIC_PIC;
+ }
+
+ /* If the BIOS pretends there is an integrated APIC ? */
+ if (!boot_cpu_has(X86_FEATURE_APIC) &&
+ APIC_INTEGRATED(boot_cpu_apic_version)) {
+ apic_is_disabled = true;
+ pr_err(FW_BUG "Local APIC not detected, force emulation\n");
+ return APIC_PIC;
+ }
+#endif
+
+ /* Check MP table or ACPI MADT configuration */
+ if (!smp_found_config) {
+ disable_ioapic_support();
+ if (!acpi_lapic) {
+ pr_info("APIC: ACPI MADT or MP tables are not detected\n");
+ return APIC_VIRTUAL_WIRE_NO_CONFIG;
+ }
+ return APIC_VIRTUAL_WIRE;
+ }
+
+#ifdef CONFIG_SMP
+ /* If SMP should be disabled, then really disable it! */
+ if (!setup_max_cpus) {
+ pr_info("APIC: SMP mode deactivated\n");
+ return APIC_SYMMETRIC_IO_NO_ROUTING;
+ }
+#endif
+
+ return APIC_SYMMETRIC_IO;
+}
+
+/* Select the interrupt delivery mode for the BSP */
+void __init apic_intr_mode_select(void)
+{
+ apic_intr_mode = __apic_intr_mode_select();
+}
+
+/*
+ * An initial setup of the virtual wire mode.
+ */
+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 || !boot_cpu_has(X86_FEATURE_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;
+ if (apic_extnmi == APIC_EXTNMI_NONE)
+ value |= APIC_LVT_MASKED;
+ apic_write(APIC_LVT1, value);
+}
+
+static void __init apic_bsp_setup(bool upmode);
+
+/* Init the interrupt delivery mode for the BSP */
+void __init apic_intr_mode_init(void)
+{
+ bool upmode = IS_ENABLED(CONFIG_UP_LATE_INIT);
+
+ switch (apic_intr_mode) {
+ case APIC_PIC:
+ pr_info("APIC: Keep in PIC mode(8259)\n");
+ return;
+ case APIC_VIRTUAL_WIRE:
+ pr_info("APIC: Switch to virtual wire mode setup\n");
+ break;
+ case APIC_VIRTUAL_WIRE_NO_CONFIG:
+ pr_info("APIC: Switch to virtual wire mode setup with no configuration\n");
+ upmode = true;
+ break;
+ case APIC_SYMMETRIC_IO:
+ pr_info("APIC: Switch to symmetric I/O mode setup\n");
+ break;
+ case APIC_SYMMETRIC_IO_NO_ROUTING:
+ pr_info("APIC: Switch to symmetric I/O mode setup in no SMP routine\n");
+ break;
+ }
+
+ x86_64_probe_apic();
+
+ x86_32_install_bigsmp();
+
+ if (x86_platform.apic_post_init)
+ x86_platform.apic_post_init();
+
+ apic_bsp_setup(upmode);
+}
+
+static void 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);
+}
+
+#define APIC_IR_REGS APIC_ISR_NR
+#define APIC_IR_BITS (APIC_IR_REGS * 32)
+#define APIC_IR_MAPSIZE (APIC_IR_BITS / BITS_PER_LONG)
+
+union apic_ir {
+ unsigned long map[APIC_IR_MAPSIZE];
+ u32 regs[APIC_IR_REGS];
+};
+
+static bool apic_check_and_ack(union apic_ir *irr, union apic_ir *isr)
+{
+ int i, bit;
+
+ /* Read the IRRs */
+ for (i = 0; i < APIC_IR_REGS; i++)
+ irr->regs[i] = apic_read(APIC_IRR + i * 0x10);
+
+ /* Read the ISRs */
+ for (i = 0; i < APIC_IR_REGS; i++)
+ isr->regs[i] = apic_read(APIC_ISR + i * 0x10);
+
+ /*
+ * If the ISR map is not empty. ACK the APIC and run another round
+ * to verify whether a pending IRR has been unblocked and turned
+ * into a ISR.
+ */
+ if (!bitmap_empty(isr->map, APIC_IR_BITS)) {
+ /*
+ * There can be multiple ISR bits set when a high priority
+ * interrupt preempted a lower priority one. Issue an ACK
+ * per set bit.
+ */
+ for_each_set_bit(bit, isr->map, APIC_IR_BITS)
+ apic_eoi();
+ return true;
+ }
+
+ return !bitmap_empty(irr->map, APIC_IR_BITS);
+}
+
+/*
+ * 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 the CPU has serviced that pending interrupt and it
+ * might not have done the apic_eoi() 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 serviced the interrupt. Hence
+ * a vector might get locked. It was noticed for timer irq (vector
+ * 0x31). Issue an extra EOI to clear ISR.
+ *
+ * If there are pending IRR bits they turn into ISR bits after a higher
+ * priority ISR bit has been acked.
+ */
+static void apic_pending_intr_clear(void)
+{
+ union apic_ir irr, isr;
+ unsigned int i;
+
+ /* 512 loops are way oversized and give the APIC a chance to obey. */
+ for (i = 0; i < 512; i++) {
+ if (!apic_check_and_ack(&irr, &isr))
+ return;
+ }
+ /* Dump the IRR/ISR content if that failed */
+ pr_warn("APIC: Stale IRR: %256pb ISR: %256pb\n", irr.map, isr.map);
+}
+
+/**
+ * setup_local_APIC - setup the local APIC
+ *
+ * Used to setup local APIC while initializing BSP or bringing up APs.
+ * Always called with preemption disabled.
+ */
+static void setup_local_APIC(void)
+{
+ int cpu = smp_processor_id();
+ unsigned int value;
+
+ if (apic_is_disabled) {
+ disable_ioapic_support();
+ return;
+ }
+
+ /*
+ * If this comes from kexec/kcrash the APIC might be enabled in
+ * SPIV. Soft disable it before doing further initialization.
+ */
+ value = apic_read(APIC_SPIV);
+ value &= ~APIC_SPIV_APIC_ENABLED;
+ apic_write(APIC_SPIV, value);
+
+#ifdef CONFIG_X86_32
+ /* Pound the ESR really hard over the head with a big hammer - mbligh */
+ if (lapic_is_integrated() && apic->disable_esr) {
+ apic_write(APIC_ESR, 0);
+ apic_write(APIC_ESR, 0);
+ apic_write(APIC_ESR, 0);
+ apic_write(APIC_ESR, 0);
+ }
+#endif
+ /* Validate that the APIC is registered if required */
+ BUG_ON(apic->apic_id_registered && !apic->apic_id_registered());
+
+ /*
+ * Intel recommends to set DFR, LDR and TPR before enabling
+ * an APIC. See e.g. "AP-388 82489DX User's Manual" (Intel
+ * document number 292116).
+ *
+ * Except for APICs which operate in physical destination mode.
+ */
+ if (apic->init_apic_ldr)
+ apic->init_apic_ldr();
+
+ /*
+ * Set Task Priority to 'accept all except vectors 0-31'. An APIC
+ * vector in the 16-31 range could be delivered if TPR == 0, but we
+ * would think it's an exception and terrible things will happen. We
+ * never change this later on.
+ */
+ value = apic_read(APIC_TASKPRI);
+ value &= ~APIC_TPRI_MASK;
+ value |= 0x10;
+ apic_write(APIC_TASKPRI, value);
+
+ /* Clear eventually stale ISR/IRR bits */
+ apic_pending_intr_clear();
+
+ /*
+ * 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 distribution model be more
+ * like LRU than MRU (the short-term load is more even across CPUs).
+ */
+
+ /*
+ * - 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);
+
+ perf_events_lapic_init();
+
+ /*
+ * Set up LVT0, LVT1:
+ *
+ * set up through-local-APIC on the boot CPU'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 (!cpu && (pic_mode || !value || ioapic_is_disabled)) {
+ value = APIC_DM_EXTINT;
+ apic_printk(APIC_VERBOSE, "enabled ExtINT on CPU#%d\n", cpu);
+ } else {
+ value = APIC_DM_EXTINT | APIC_LVT_MASKED;
+ apic_printk(APIC_VERBOSE, "masked ExtINT on CPU#%d\n", cpu);
+ }
+ apic_write(APIC_LVT0, value);
+
+ /*
+ * Only the BSP sees the LINT1 NMI signal by default. This can be
+ * modified by apic_extnmi= boot option.
+ */
+ if ((!cpu && apic_extnmi != APIC_EXTNMI_NONE) ||
+ apic_extnmi == APIC_EXTNMI_ALL)
+ value = APIC_DM_NMI;
+ else
+ value = APIC_DM_NMI | APIC_LVT_MASKED;
+
+ /* Is 82489DX ? */
+ if (!lapic_is_integrated())
+ value |= APIC_LVT_LEVEL_TRIGGER;
+ apic_write(APIC_LVT1, value);
+
+#ifdef CONFIG_X86_MCE_INTEL
+ /* Recheck CMCI information after local APIC is up on CPU #0 */
+ if (!cpu)
+ cmci_recheck();
+#endif
+}
+
+static void 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
+
+ apic_pm_activate();
+}
+
+/*
+ * APIC setup function for application processors. Called from smpboot.c
+ */
+void apic_ap_setup(void)
+{
+ setup_local_APIC();
+ end_local_APIC_setup();
+}
+
+static __init void cpu_set_boot_apic(void);
+
+static __init void apic_read_boot_cpu_id(bool x2apic)
+{
+ /*
+ * This can be invoked from check_x2apic() before the APIC has been
+ * selected. But that code knows for sure that the BIOS enabled
+ * X2APIC.
+ */
+ if (x2apic) {
+ boot_cpu_physical_apicid = native_apic_msr_read(APIC_ID);
+ boot_cpu_apic_version = GET_APIC_VERSION(native_apic_msr_read(APIC_LVR));
+ } else {
+ boot_cpu_physical_apicid = read_apic_id();
+ boot_cpu_apic_version = GET_APIC_VERSION(apic_read(APIC_LVR));
+ }
+ cpu_set_boot_apic();
+}
+
+#ifdef CONFIG_X86_X2APIC
+int x2apic_mode;
+EXPORT_SYMBOL_GPL(x2apic_mode);
+
+enum {
+ X2APIC_OFF,
+ X2APIC_DISABLED,
+ /* All states below here have X2APIC enabled */
+ X2APIC_ON,
+ X2APIC_ON_LOCKED
+};
+static int x2apic_state;
+
+static bool x2apic_hw_locked(void)
+{
+ u64 ia32_cap;
+ u64 msr;
+
+ ia32_cap = x86_read_arch_cap_msr();
+ if (ia32_cap & ARCH_CAP_XAPIC_DISABLE) {
+ rdmsrl(MSR_IA32_XAPIC_DISABLE_STATUS, msr);
+ return (msr & LEGACY_XAPIC_DISABLED);
+ }
+ return false;
+}
+
+static void __x2apic_disable(void)
+{
+ u64 msr;
+
+ if (!boot_cpu_has(X86_FEATURE_APIC))
+ return;
+
+ rdmsrl(MSR_IA32_APICBASE, msr);
+ if (!(msr & X2APIC_ENABLE))
+ return;
+ /* Disable xapic and x2apic first and then reenable xapic mode */
+ wrmsrl(MSR_IA32_APICBASE, msr & ~(X2APIC_ENABLE | XAPIC_ENABLE));
+ wrmsrl(MSR_IA32_APICBASE, msr & ~X2APIC_ENABLE);
+ printk_once(KERN_INFO "x2apic disabled\n");
+}
+
+static void __x2apic_enable(void)
+{
+ u64 msr;
+
+ rdmsrl(MSR_IA32_APICBASE, msr);
+ if (msr & X2APIC_ENABLE)
+ return;
+ wrmsrl(MSR_IA32_APICBASE, msr | X2APIC_ENABLE);
+ printk_once(KERN_INFO "x2apic enabled\n");
+}
+
+static int __init setup_nox2apic(char *str)
+{
+ if (x2apic_enabled()) {
+ int apicid = native_apic_msr_read(APIC_ID);
+
+ if (apicid >= 255) {
+ pr_warn("Apicid: %08x, cannot enforce nox2apic\n",
+ apicid);
+ return 0;
+ }
+ if (x2apic_hw_locked()) {
+ pr_warn("APIC locked in x2apic mode, can't disable\n");
+ return 0;
+ }
+ pr_warn("x2apic already enabled.\n");
+ __x2apic_disable();
+ }
+ setup_clear_cpu_cap(X86_FEATURE_X2APIC);
+ x2apic_state = X2APIC_DISABLED;
+ x2apic_mode = 0;
+ return 0;
+}
+early_param("nox2apic", setup_nox2apic);
+
+/* Called from cpu_init() to enable x2apic on (secondary) cpus */
+void x2apic_setup(void)
+{
+ /*
+ * Try to make the AP's APIC state match that of the BSP, but if the
+ * BSP is unlocked and the AP is locked then there is a state mismatch.
+ * Warn about the mismatch in case a GP fault occurs due to a locked AP
+ * trying to be turned off.
+ */
+ if (x2apic_state != X2APIC_ON_LOCKED && x2apic_hw_locked())
+ pr_warn("x2apic lock mismatch between BSP and AP.\n");
+ /*
+ * If x2apic is not in ON or LOCKED state, disable it if already enabled
+ * from BIOS.
+ */
+ if (x2apic_state < X2APIC_ON) {
+ __x2apic_disable();
+ return;
+ }
+ __x2apic_enable();
+}
+
+static __init void apic_set_fixmap(void);
+
+static __init void x2apic_disable(void)
+{
+ u32 x2apic_id, state = x2apic_state;
+
+ x2apic_mode = 0;
+ x2apic_state = X2APIC_DISABLED;
+
+ if (state != X2APIC_ON)
+ return;
+
+ x2apic_id = read_apic_id();
+ if (x2apic_id >= 255)
+ panic("Cannot disable x2apic, id: %08x\n", x2apic_id);
+
+ if (x2apic_hw_locked()) {
+ pr_warn("Cannot disable locked x2apic, id: %08x\n", x2apic_id);
+ return;
+ }
+
+ __x2apic_disable();
+ apic_set_fixmap();
+}
+
+static __init void x2apic_enable(void)
+{
+ if (x2apic_state != X2APIC_OFF)
+ return;
+
+ x2apic_mode = 1;
+ x2apic_state = X2APIC_ON;
+ __x2apic_enable();
+}
+
+static __init void try_to_enable_x2apic(int remap_mode)
+{
+ if (x2apic_state == X2APIC_DISABLED)
+ return;
+
+ if (remap_mode != IRQ_REMAP_X2APIC_MODE) {
+ u32 apic_limit = 255;
+
+ /*
+ * Using X2APIC without IR is not architecturally supported
+ * on bare metal but may be supported in guests.
+ */
+ if (!x86_init.hyper.x2apic_available()) {
+ pr_info("x2apic: IRQ remapping doesn't support X2APIC mode\n");
+ x2apic_disable();
+ return;
+ }
+
+ /*
+ * If the hypervisor supports extended destination ID in
+ * MSI, that increases the maximum APIC ID that can be
+ * used for non-remapped IRQ domains.
+ */
+ if (x86_init.hyper.msi_ext_dest_id()) {
+ virt_ext_dest_id = 1;
+ apic_limit = 32767;
+ }
+
+ /*
+ * Without IR, all CPUs can be addressed by IOAPIC/MSI only
+ * in physical mode, and CPUs with an APIC ID that cannot
+ * be addressed must not be brought online.
+ */
+ x2apic_set_max_apicid(apic_limit);
+ x2apic_phys = 1;
+ }
+ x2apic_enable();
+}
+
+void __init check_x2apic(void)
+{
+ if (x2apic_enabled()) {
+ pr_info("x2apic: enabled by BIOS, switching to x2apic ops\n");
+ x2apic_mode = 1;
+ if (x2apic_hw_locked())
+ x2apic_state = X2APIC_ON_LOCKED;
+ else
+ x2apic_state = X2APIC_ON;
+ apic_read_boot_cpu_id(true);
+ } else if (!boot_cpu_has(X86_FEATURE_X2APIC)) {
+ x2apic_state = X2APIC_DISABLED;
+ }
+}
+#else /* CONFIG_X86_X2APIC */
+void __init check_x2apic(void)
+{
+ if (!apic_is_x2apic_enabled())
+ return;
+ /*
+ * Checkme: Can we simply turn off x2APIC here instead of disabling the APIC?
+ */
+ pr_err("Kernel does not support x2APIC, please recompile with CONFIG_X86_X2APIC.\n");
+ pr_err("Disabling APIC, expect reduced performance and functionality.\n");
+
+ apic_is_disabled = true;
+ setup_clear_cpu_cap(X86_FEATURE_APIC);
+}
+
+static inline void try_to_enable_x2apic(int remap_mode) { }
+static inline void __x2apic_enable(void) { }
+#endif /* !CONFIG_X86_X2APIC */
+
+void __init enable_IR_x2apic(void)
+{
+ unsigned long flags;
+ int ret, ir_stat;
+
+ if (ioapic_is_disabled) {
+ pr_info("Not enabling interrupt remapping due to skipped IO-APIC setup\n");
+ return;
+ }
+
+ ir_stat = irq_remapping_prepare();
+ if (ir_stat < 0 && !x2apic_supported())
+ return;
+
+ ret = save_ioapic_entries();
+ if (ret) {
+ pr_info("Saving IO-APIC state failed: %d\n", ret);
+ return;
+ }
+
+ local_irq_save(flags);
+ legacy_pic->mask_all();
+ mask_ioapic_entries();
+
+ /* If irq_remapping_prepare() succeeded, try to enable it */
+ if (ir_stat >= 0)
+ ir_stat = irq_remapping_enable();
+ /* ir_stat contains the remap mode or an error code */
+ try_to_enable_x2apic(ir_stat);
+
+ if (ir_stat < 0)
+ restore_ioapic_entries();
+ legacy_pic->restore_mask();
+ local_irq_restore(flags);
+}
+
+#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 bool __init detect_init_APIC(void)
+{
+ if (!boot_cpu_has(X86_FEATURE_APIC)) {
+ pr_info("No local APIC present\n");
+ return false;
+ }
+
+ register_lapic_address(APIC_DEFAULT_PHYS_BASE);
+ return true;
+}
+#else
+
+static bool __init apic_verify(unsigned long addr)
+{
+ u32 features, h, l;
+
+ /*
+ * The APIC feature bit should now be enabled
+ * in `cpuid'
+ */
+ features = cpuid_edx(1);
+ if (!(features & (1 << X86_FEATURE_APIC))) {
+ pr_warn("Could not enable APIC!\n");
+ return false;
+ }
+ set_cpu_cap(&boot_cpu_data, X86_FEATURE_APIC);
+
+ /* The BIOS may have set up the APIC at some other address */
+ if (boot_cpu_data.x86 >= 6) {
+ rdmsr(MSR_IA32_APICBASE, l, h);
+ if (l & MSR_IA32_APICBASE_ENABLE)
+ addr = l & MSR_IA32_APICBASE_BASE;
+ }
+
+ register_lapic_address(addr);
+ pr_info("Found and enabled local APIC!\n");
+ return true;
+}
+
+bool __init apic_force_enable(unsigned long addr)
+{
+ u32 h, l;
+
+ if (apic_is_disabled)
+ return false;
+
+ /*
+ * 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.
+ */
+ if (boot_cpu_data.x86 >= 6) {
+ 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 | addr;
+ wrmsr(MSR_IA32_APICBASE, l, h);
+ enabled_via_apicbase = 1;
+ }
+ }
+ return apic_verify(addr);
+}
+
+/*
+ * Detect and initialize APIC
+ */
+static bool __init detect_init_APIC(void)
+{
+ /* Disabled by kernel option? */
+ if (apic_is_disabled)
+ return false;
+
+ 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_HYGON:
+ break;
+ case X86_VENDOR_INTEL:
+ if (boot_cpu_data.x86 == 6 || boot_cpu_data.x86 == 15 ||
+ (boot_cpu_data.x86 == 5 && boot_cpu_has(X86_FEATURE_APIC)))
+ break;
+ goto no_apic;
+ default:
+ goto no_apic;
+ }
+
+ if (!boot_cpu_has(X86_FEATURE_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 false;
+ }
+ if (!apic_force_enable(APIC_DEFAULT_PHYS_BASE))
+ return false;
+ } else {
+ if (!apic_verify(APIC_DEFAULT_PHYS_BASE))
+ return false;
+ }
+
+ apic_pm_activate();
+
+ return true;
+
+no_apic:
+ pr_info("No local APIC present or hardware disabled\n");
+ return false;
+}
+#endif
+
+/**
+ * init_apic_mappings - initialize APIC mappings
+ */
+void __init init_apic_mappings(void)
+{
+ if (apic_validate_deadline_timer())
+ pr_info("TSC deadline timer available\n");
+
+ if (x2apic_mode)
+ return;
+
+ if (!smp_found_config) {
+ if (!detect_init_APIC()) {
+ pr_info("APIC: disable apic facility\n");
+ apic_disable();
+ }
+ num_processors = 1;
+ }
+}
+
+static __init void apic_set_fixmap(void)
+{
+ set_fixmap_nocache(FIX_APIC_BASE, mp_lapic_addr);
+ apic_mmio_base = APIC_BASE;
+ apic_printk(APIC_VERBOSE, "mapped APIC to %16lx (%16lx)\n",
+ apic_mmio_base, mp_lapic_addr);
+ apic_read_boot_cpu_id(false);
+}
+
+void __init register_lapic_address(unsigned long address)
+{
+ /* This should only happen once */
+ WARN_ON_ONCE(mp_lapic_addr);
+ mp_lapic_addr = address;
+
+ if (!x2apic_mode)
+ apic_set_fixmap();
+}
+
+/*
+ * Local APIC interrupts
+ */
+
+/*
+ * Common handling code for spurious_interrupt and spurious_vector entry
+ * points below. No point in allowing the compiler to inline it twice.
+ */
+static noinline void handle_spurious_interrupt(u8 vector)
+{
+ u32 v;
+
+ trace_spurious_apic_entry(vector);
+
+ inc_irq_stat(irq_spurious_count);
+
+ /*
+ * If this is a spurious interrupt then do not acknowledge
+ */
+ if (vector == SPURIOUS_APIC_VECTOR) {
+ /* See SDM vol 3 */
+ pr_info("Spurious APIC interrupt (vector 0xFF) on CPU#%d, should never happen.\n",
+ smp_processor_id());
+ goto out;
+ }
+
+ /*
+ * If it is a vectored one, verify it's set in the ISR. If set,
+ * acknowledge it.
+ */
+ v = apic_read(APIC_ISR + ((vector & ~0x1f) >> 1));
+ if (v & (1 << (vector & 0x1f))) {
+ pr_info("Spurious interrupt (vector 0x%02x) on CPU#%d. Acked\n",
+ vector, smp_processor_id());
+ apic_eoi();
+ } else {
+ pr_info("Spurious interrupt (vector 0x%02x) on CPU#%d. Not pending!\n",
+ vector, smp_processor_id());
+ }
+out:
+ trace_spurious_apic_exit(vector);
+}
+
+/**
+ * spurious_interrupt - Catch all for interrupts raised on unused vectors
+ * @regs: Pointer to pt_regs on stack
+ * @vector: The vector number
+ *
+ * This is invoked from ASM entry code to catch all interrupts which
+ * trigger on an entry which is routed to the common_spurious idtentry
+ * point.
+ */
+DEFINE_IDTENTRY_IRQ(spurious_interrupt)
+{
+ handle_spurious_interrupt(vector);
+}
+
+DEFINE_IDTENTRY_SYSVEC(sysvec_spurious_apic_interrupt)
+{
+ handle_spurious_interrupt(SPURIOUS_APIC_VECTOR);
+}
+
+/*
+ * This interrupt should never happen with our APIC/SMP architecture
+ */
+DEFINE_IDTENTRY_SYSVEC(sysvec_error_interrupt)
+{
+ static const char * const error_interrupt_reason[] = {
+ "Send CS error", /* APIC Error Bit 0 */
+ "Receive CS error", /* APIC Error Bit 1 */
+ "Send accept error", /* APIC Error Bit 2 */
+ "Receive accept error", /* APIC Error Bit 3 */
+ "Redirectable IPI", /* APIC Error Bit 4 */
+ "Send illegal vector", /* APIC Error Bit 5 */
+ "Received illegal vector", /* APIC Error Bit 6 */
+ "Illegal register address", /* APIC Error Bit 7 */
+ };
+ u32 v, i = 0;
+
+ trace_error_apic_entry(ERROR_APIC_VECTOR);
+
+ /* First tickle the hardware, only then report what went on. -- REW */
+ if (lapic_get_maxlvt() > 3) /* Due to the Pentium erratum 3AP. */
+ apic_write(APIC_ESR, 0);
+ v = apic_read(APIC_ESR);
+ apic_eoi();
+ atomic_inc(&irq_err_count);
+
+ apic_printk(APIC_DEBUG, KERN_DEBUG "APIC error on CPU%d: %02x",
+ smp_processor_id(), v);
+
+ v &= 0xff;
+ while (v) {
+ if (v & 0x1)
+ apic_printk(APIC_DEBUG, KERN_CONT " : %s", error_interrupt_reason[i]);
+ i++;
+ v >>= 1;
+ }
+
+ apic_printk(APIC_DEBUG, KERN_CONT "\n");
+
+ trace_error_apic_exit(ERROR_APIC_VECTOR);
+}
+
+/**
+ * connect_bsp_APIC - attach the APIC to the interrupt system
+ */
+static 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");
+ imcr_pic_to_apic();
+ }
+#endif
+}
+
+/**
+ * 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");
+ imcr_apic_to_pic();
+ 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);
+}
+
+/*
+ * The number of allocated logical CPU IDs. Since logical CPU IDs are allocated
+ * contiguously, it equals to current allocated max logical CPU ID plus 1.
+ * All allocated CPU IDs should be in the [0, nr_logical_cpuids) range,
+ * so the maximum of nr_logical_cpuids is nr_cpu_ids.
+ *
+ * NOTE: Reserve 0 for BSP.
+ */
+static int nr_logical_cpuids = 1;
+
+/*
+ * Used to store mapping between logical CPU IDs and APIC IDs.
+ */
+int cpuid_to_apicid[] = {
+ [0 ... NR_CPUS - 1] = -1,
+};
+
+bool arch_match_cpu_phys_id(int cpu, u64 phys_id)
+{
+ return phys_id == cpuid_to_apicid[cpu];
+}
+
+#ifdef CONFIG_SMP
+static void cpu_mark_primary_thread(unsigned int cpu, unsigned int apicid)
+{
+ /* Isolate the SMT bit(s) in the APICID and check for 0 */
+ u32 mask = (1U << (fls(smp_num_siblings) - 1)) - 1;
+
+ if (smp_num_siblings == 1 || !(apicid & mask))
+ cpumask_set_cpu(cpu, &__cpu_primary_thread_mask);
+}
+
+/*
+ * Due to the utter mess of CPUID evaluation smp_num_siblings is not valid
+ * during early boot. Initialize the primary thread mask before SMP
+ * bringup.
+ */
+static int __init smp_init_primary_thread_mask(void)
+{
+ unsigned int cpu;
+
+ /*
+ * XEN/PV provides either none or useless topology information.
+ * Pretend that all vCPUs are primary threads.
+ */
+ if (xen_pv_domain()) {
+ cpumask_copy(&__cpu_primary_thread_mask, cpu_possible_mask);
+ return 0;
+ }
+
+ for (cpu = 0; cpu < nr_logical_cpuids; cpu++)
+ cpu_mark_primary_thread(cpu, cpuid_to_apicid[cpu]);
+ return 0;
+}
+early_initcall(smp_init_primary_thread_mask);
+#else
+static inline void cpu_mark_primary_thread(unsigned int cpu, unsigned int apicid) { }
+#endif
+
+/*
+ * Should use this API to allocate logical CPU IDs to keep nr_logical_cpuids
+ * and cpuid_to_apicid[] synchronized.
+ */
+static int allocate_logical_cpuid(int apicid)
+{
+ int i;
+
+ /*
+ * cpuid <-> apicid mapping is persistent, so when a cpu is up,
+ * check if the kernel has allocated a cpuid for it.
+ */
+ for (i = 0; i < nr_logical_cpuids; i++) {
+ if (cpuid_to_apicid[i] == apicid)
+ return i;
+ }
+
+ /* Allocate a new cpuid. */
+ if (nr_logical_cpuids >= nr_cpu_ids) {
+ WARN_ONCE(1, "APIC: NR_CPUS/possible_cpus limit of %u reached. "
+ "Processor %d/0x%x and the rest are ignored.\n",
+ nr_cpu_ids, nr_logical_cpuids, apicid);
+ return -EINVAL;
+ }
+
+ cpuid_to_apicid[nr_logical_cpuids] = apicid;
+ return nr_logical_cpuids++;
+}
+
+static void cpu_update_apic(int cpu, int apicid)
+{
+#if defined(CONFIG_SMP) || defined(CONFIG_X86_64)
+ early_per_cpu(x86_cpu_to_apicid, cpu) = apicid;
+#endif
+ set_cpu_possible(cpu, true);
+ physid_set(apicid, phys_cpu_present_map);
+ set_cpu_present(cpu, true);
+ num_processors++;
+
+ if (system_state != SYSTEM_BOOTING)
+ cpu_mark_primary_thread(cpu, apicid);
+}
+
+static __init void cpu_set_boot_apic(void)
+{
+ cpuid_to_apicid[0] = boot_cpu_physical_apicid;
+ cpu_update_apic(0, boot_cpu_physical_apicid);
+ x86_32_probe_bigsmp_early();
+}
+
+int generic_processor_info(int apicid)
+{
+ int cpu, max = nr_cpu_ids;
+
+ /* The boot CPU must be set before MADT/MPTABLE parsing happens */
+ if (cpuid_to_apicid[0] == BAD_APICID)
+ panic("Boot CPU APIC not registered yet\n");
+
+ if (apicid == boot_cpu_physical_apicid)
+ return 0;
+
+ if (disabled_cpu_apicid == apicid) {
+ int thiscpu = num_processors + disabled_cpus;
+
+ pr_warn("APIC: Disabling requested cpu. Processor %d/0x%x ignored.\n",
+ thiscpu, apicid);
+
+ disabled_cpus++;
+ return -ENODEV;
+ }
+
+ if (num_processors >= nr_cpu_ids) {
+ int thiscpu = max + disabled_cpus;
+
+ pr_warn("APIC: NR_CPUS/possible_cpus limit of %i reached. "
+ "Processor %d/0x%x ignored.\n", max, thiscpu, apicid);
+
+ disabled_cpus++;
+ return -EINVAL;
+ }
+
+ cpu = allocate_logical_cpuid(apicid);
+ if (cpu < 0) {
+ disabled_cpus++;
+ return -EINVAL;
+ }
+
+ cpu_update_apic(cpu, apicid);
+ return cpu;
+}
+
+
+void __irq_msi_compose_msg(struct irq_cfg *cfg, struct msi_msg *msg,
+ bool dmar)
+{
+ memset(msg, 0, sizeof(*msg));
+
+ msg->arch_addr_lo.base_address = X86_MSI_BASE_ADDRESS_LOW;
+ msg->arch_addr_lo.dest_mode_logical = apic->dest_mode_logical;
+ msg->arch_addr_lo.destid_0_7 = cfg->dest_apicid & 0xFF;
+
+ msg->arch_data.delivery_mode = APIC_DELIVERY_MODE_FIXED;
+ msg->arch_data.vector = cfg->vector;
+
+ msg->address_hi = X86_MSI_BASE_ADDRESS_HIGH;
+ /*
+ * Only the IOMMU itself can use the trick of putting destination
+ * APIC ID into the high bits of the address. Anything else would
+ * just be writing to memory if it tried that, and needs IR to
+ * address APICs which can't be addressed in the normal 32-bit
+ * address range at 0xFFExxxxx. That is typically just 8 bits, but
+ * some hypervisors allow the extended destination ID field in bits
+ * 5-11 to be used, giving support for 15 bits of APIC IDs in total.
+ */
+ if (dmar)
+ msg->arch_addr_hi.destid_8_31 = cfg->dest_apicid >> 8;
+ else if (virt_ext_dest_id && cfg->dest_apicid < 0x8000)
+ msg->arch_addr_lo.virt_destid_8_14 = cfg->dest_apicid >> 8;
+ else
+ WARN_ON_ONCE(cfg->dest_apicid > 0xFF);
+}
+
+u32 x86_msi_msg_get_destid(struct msi_msg *msg, bool extid)
+{
+ u32 dest = msg->arch_addr_lo.destid_0_7;
+
+ if (extid)
+ dest |= msg->arch_addr_hi.destid_8_31 << 8;
+ return dest;
+}
+EXPORT_SYMBOL_GPL(x86_msi_msg_get_destid);
+
+static void __init apic_bsp_up_setup(void)
+{
+#ifdef CONFIG_X86_64
+ apic_write(APIC_ID, apic->set_apic_id(boot_cpu_physical_apicid));
+#endif
+ physid_set_mask_of_physid(boot_cpu_physical_apicid, &phys_cpu_present_map);
+}
+
+/**
+ * apic_bsp_setup - Setup function for local apic and io-apic
+ * @upmode: Force UP mode (for APIC_init_uniprocessor)
+ */
+static void __init apic_bsp_setup(bool upmode)
+{
+ connect_bsp_APIC();
+ if (upmode)
+ apic_bsp_up_setup();
+ setup_local_APIC();
+
+ enable_IO_APIC();
+ end_local_APIC_setup();
+ irq_remap_enable_fault_handling();
+ setup_IO_APIC();
+ lapic_update_legacy_vectors();
+}
+
+#ifdef CONFIG_UP_LATE_INIT
+void __init up_late_init(void)
+{
+ if (apic_intr_mode == APIC_PIC)
+ return;
+
+ /* Setup local timer */
+ x86_init.timers.setup_percpu_clockev();
+}
+#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;
+ unsigned int apic_cmci;
+} apic_pm_state;
+
+static int lapic_suspend(void)
+{
+ 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);
+#ifdef CONFIG_X86_THERMAL_VECTOR
+ if (maxlvt >= 5)
+ apic_pm_state.apic_thmr = apic_read(APIC_LVTTHMR);
+#endif
+#ifdef CONFIG_X86_MCE_INTEL
+ if (maxlvt >= 6)
+ apic_pm_state.apic_cmci = apic_read(APIC_LVTCMCI);
+#endif
+
+ local_irq_save(flags);
+
+ /*
+ * Mask IOAPIC before disabling the local APIC to prevent stale IRR
+ * entries on some implementations.
+ */
+ mask_ioapic_entries();
+
+ disable_local_APIC();
+
+ irq_remapping_disable();
+
+ local_irq_restore(flags);
+ return 0;
+}
+
+static void lapic_resume(void)
+{
+ unsigned int l, h;
+ unsigned long flags;
+ int maxlvt;
+
+ if (!apic_pm_state.active)
+ return;
+
+ local_irq_save(flags);
+
+ /*
+ * IO-APIC and PIC have their own resume routines.
+ * We just mask them here to make sure the interrupt
+ * subsystem is completely quiet while we enable x2apic
+ * and interrupt-remapping.
+ */
+ mask_ioapic_entries();
+ legacy_pic->mask_all();
+
+ if (x2apic_mode) {
+ __x2apic_enable();
+ } else {
+ /*
+ * 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!
+ */
+ if (boot_cpu_data.x86 >= 6) {
+ 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);
+ }
+ }
+
+ maxlvt = lapic_get_maxlvt();
+ 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);
+#ifdef CONFIG_X86_THERMAL_VECTOR
+ if (maxlvt >= 5)
+ apic_write(APIC_LVTTHMR, apic_pm_state.apic_thmr);
+#endif
+#ifdef CONFIG_X86_MCE_INTEL
+ if (maxlvt >= 6)
+ apic_write(APIC_LVTCMCI, apic_pm_state.apic_cmci);
+#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);
+
+ irq_remapping_reenable(x2apic_mode);
+
+ local_irq_restore(flags);
+}
+
+/*
+ * This device has no shutdown method - fully functioning local APICs
+ * are needed on every CPU up until machine_halt/restart/poweroff.
+ */
+
+static struct syscore_ops lapic_syscore_ops = {
+ .resume = lapic_resume,
+ .suspend = lapic_suspend,
+};
+
+static void apic_pm_activate(void)
+{
+ apic_pm_state.active = 1;
+}
+
+static int __init init_lapic_sysfs(void)
+{
+ /* XXX: remove suspend/resume procs if !apic_pm_state.active? */
+ if (boot_cpu_has(X86_FEATURE_APIC))
+ register_syscore_ops(&lapic_syscore_ops);
+
+ return 0;
+}
+
+/* local apic needs to resume before other devices access its registers. */
+core_initcall(init_lapic_sysfs);
+
+#else /* CONFIG_PM */
+
+static void apic_pm_activate(void) { }
+
+#endif /* CONFIG_PM */
+
+#ifdef CONFIG_X86_64
+
+static int multi_checked;
+static int multi;
+
+static int set_multi(const struct dmi_system_id *d)
+{
+ if (multi)
+ return 0;
+ pr_info("APIC: %s detected, Multi Chassis\n", d->ident);
+ multi = 1;
+ return 0;
+}
+
+static const struct dmi_system_id multi_dmi_table[] = {
+ {
+ .callback = set_multi,
+ .ident = "IBM System Summit2",
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "IBM"),
+ DMI_MATCH(DMI_PRODUCT_NAME, "Summit2"),
+ },
+ },
+ {}
+};
+
+static void dmi_check_multi(void)
+{
+ if (multi_checked)
+ return;
+
+ dmi_check_system(multi_dmi_table);
+ multi_checked = 1;
+}
+
+/*
+ * 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 DMI to check them
+ */
+int apic_is_clustered_box(void)
+{
+ dmi_check_multi();
+ return multi;
+}
+#endif
+
+/*
+ * APIC command line parameters
+ */
+static int __init setup_disableapic(char *arg)
+{
+ apic_is_disabled = true;
+ setup_clear_cpu_cap(X86_FEATURE_APIC);
+ return 0;
+}
+early_param("disableapic", setup_disableapic);
+
+/* same as disableapic, for compatibility */
+static int __init setup_nolapic(char *arg)
+{
+ return setup_disableapic(arg);
+}
+early_param("nolapic", setup_nolapic);
+
+static int __init parse_lapic_timer_c2_ok(char *arg)
+{
+ 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) {
+ if (IS_ENABLED(CONFIG_X86_32))
+ return -EINVAL;
+
+ ioapic_is_disabled = false;
+ return 0;
+ }
+
+ if (strcmp("debug", arg) == 0)
+ apic_verbosity = APIC_DEBUG;
+ else if (strcmp("verbose", arg) == 0)
+ apic_verbosity = APIC_VERBOSE;
+#ifdef CONFIG_X86_64
+ else {
+ pr_warn("APIC Verbosity level %s not recognised"
+ " use apic=verbose or apic=debug\n", arg);
+ return -EINVAL;
+ }
+#endif
+
+ return 0;
+}
+early_param("apic", apic_set_verbosity);
+
+static int __init lapic_insert_resource(void)
+{
+ if (!apic_mmio_base)
+ return -1;
+
+ /* Put local APIC into the resource map. */
+ lapic_resource.start = apic_mmio_base;
+ 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);
+
+static int __init apic_set_disabled_cpu_apicid(char *arg)
+{
+ if (!arg || !get_option(&arg, &disabled_cpu_apicid))
+ return -EINVAL;
+
+ return 0;
+}
+early_param("disable_cpu_apicid", apic_set_disabled_cpu_apicid);
+
+static int __init apic_set_extnmi(char *arg)
+{
+ if (!arg)
+ return -EINVAL;
+
+ if (!strncmp("all", arg, 3))
+ apic_extnmi = APIC_EXTNMI_ALL;
+ else if (!strncmp("none", arg, 4))
+ apic_extnmi = APIC_EXTNMI_NONE;
+ else if (!strncmp("bsp", arg, 3))
+ apic_extnmi = APIC_EXTNMI_BSP;
+ else {
+ pr_warn("Unknown external NMI delivery mode `%s' ignored\n", arg);
+ return -EINVAL;
+ }
+
+ return 0;
+}
+early_param("apic_extnmi", apic_set_extnmi);
diff --git a/arch/x86/kernel/apic/apic_common.c b/arch/x86/kernel/apic/apic_common.c
new file mode 100644
index 0000000000..7bc5d9bf59
--- /dev/null
+++ b/arch/x86/kernel/apic/apic_common.c
@@ -0,0 +1,57 @@
+/*
+ * Common functions shared between the various APIC flavours
+ *
+ * SPDX-License-Identifier: GPL-2.0
+ */
+#include <linux/irq.h>
+#include <asm/apic.h>
+
+#include "local.h"
+
+u32 apic_default_calc_apicid(unsigned int cpu)
+{
+ return per_cpu(x86_cpu_to_apicid, cpu);
+}
+
+u32 apic_flat_calc_apicid(unsigned int cpu)
+{
+ return 1U << cpu;
+}
+
+bool default_check_apicid_used(physid_mask_t *map, int apicid)
+{
+ return physid_isset(apicid, *map);
+}
+
+void default_ioapic_phys_id_map(physid_mask_t *phys_map, physid_mask_t *retmap)
+{
+ *retmap = *phys_map;
+}
+
+int default_cpu_present_to_apicid(int mps_cpu)
+{
+ if (mps_cpu < nr_cpu_ids && cpu_present(mps_cpu))
+ return (int)per_cpu(x86_cpu_to_apicid, mps_cpu);
+ else
+ return BAD_APICID;
+}
+EXPORT_SYMBOL_GPL(default_cpu_present_to_apicid);
+
+bool default_apic_id_registered(void)
+{
+ return physid_isset(read_apic_id(), phys_cpu_present_map);
+}
+
+/*
+ * Set up the logical destination ID when the APIC operates in logical
+ * destination mode.
+ */
+void default_init_apic_ldr(void)
+{
+ unsigned long val;
+
+ apic_write(APIC_DFR, APIC_DFR_FLAT);
+ val = apic_read(APIC_LDR) & ~APIC_LDR_MASK;
+ val |= SET_APIC_LOGICAL_ID(1UL << smp_processor_id());
+ apic_write(APIC_LDR, val);
+}
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 0000000000..032a84e2c3
--- /dev/null
+++ b/arch/x86/kernel/apic/apic_flat_64.c
@@ -0,0 +1,191 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright 2004 James Cleverdon, IBM.
+ *
+ * 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/cpumask.h>
+#include <linux/export.h>
+#include <linux/acpi.h>
+
+#include <asm/jailhouse_para.h>
+#include <asm/apic.h>
+
+#include "local.h"
+
+static struct apic apic_physflat;
+static struct apic apic_flat;
+
+struct apic *apic __ro_after_init = &apic_flat;
+EXPORT_SYMBOL_GPL(apic);
+
+static int flat_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
+{
+ return 1;
+}
+
+static 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 unsigned int flat_get_apic_id(unsigned long x)
+{
+ return (x >> 24) & 0xFF;
+}
+
+static u32 set_apic_id(unsigned int id)
+{
+ return (id & 0xFF) << 24;
+}
+
+static int flat_phys_pkg_id(int initial_apic_id, int index_msb)
+{
+ return initial_apic_id >> index_msb;
+}
+
+static int flat_probe(void)
+{
+ return 1;
+}
+
+static struct apic apic_flat __ro_after_init = {
+ .name = "flat",
+ .probe = flat_probe,
+ .acpi_madt_oem_check = flat_acpi_madt_oem_check,
+ .apic_id_registered = default_apic_id_registered,
+
+ .delivery_mode = APIC_DELIVERY_MODE_FIXED,
+ .dest_mode_logical = true,
+
+ .disable_esr = 0,
+
+ .init_apic_ldr = default_init_apic_ldr,
+ .cpu_present_to_apicid = default_cpu_present_to_apicid,
+ .phys_pkg_id = flat_phys_pkg_id,
+
+ .max_apic_id = 0xFE,
+ .get_apic_id = flat_get_apic_id,
+ .set_apic_id = set_apic_id,
+
+ .calc_dest_apicid = apic_flat_calc_apicid,
+
+ .send_IPI = default_send_IPI_single,
+ .send_IPI_mask = flat_send_IPI_mask,
+ .send_IPI_mask_allbutself = flat_send_IPI_mask_allbutself,
+ .send_IPI_allbutself = default_send_IPI_allbutself,
+ .send_IPI_all = default_send_IPI_all,
+ .send_IPI_self = default_send_IPI_self,
+
+ .read = native_apic_mem_read,
+ .write = native_apic_mem_write,
+ .eoi = native_apic_mem_eoi,
+ .icr_read = native_apic_icr_read,
+ .icr_write = native_apic_icr_write,
+ .wait_icr_idle = apic_mem_wait_icr_idle,
+ .safe_wait_icr_idle = apic_mem_wait_icr_idle_timeout,
+};
+
+/*
+ * Physflat mode is used when there are more than 8 CPUs on a 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;
+ }
+
+ if (!strncmp(oem_id, "IBM", 3) && !strncmp(oem_table_id, "EXA", 3)) {
+ printk(KERN_DEBUG "IBM Summit detected, will use apic physical");
+ return 1;
+ }
+#endif
+
+ return 0;
+}
+
+static int physflat_probe(void)
+{
+ return apic == &apic_physflat || num_possible_cpus() > 8 || jailhouse_paravirt();
+}
+
+static struct apic apic_physflat __ro_after_init = {
+
+ .name = "physical flat",
+ .probe = physflat_probe,
+ .acpi_madt_oem_check = physflat_acpi_madt_oem_check,
+ .apic_id_registered = default_apic_id_registered,
+
+ .delivery_mode = APIC_DELIVERY_MODE_FIXED,
+ .dest_mode_logical = false,
+
+ .disable_esr = 0,
+
+ .check_apicid_used = NULL,
+ .ioapic_phys_id_map = NULL,
+ .cpu_present_to_apicid = default_cpu_present_to_apicid,
+ .phys_pkg_id = flat_phys_pkg_id,
+
+ .max_apic_id = 0xFE,
+ .get_apic_id = flat_get_apic_id,
+ .set_apic_id = set_apic_id,
+
+ .calc_dest_apicid = apic_default_calc_apicid,
+
+ .send_IPI = default_send_IPI_single_phys,
+ .send_IPI_mask = default_send_IPI_mask_sequence_phys,
+ .send_IPI_mask_allbutself = default_send_IPI_mask_allbutself_phys,
+ .send_IPI_allbutself = default_send_IPI_allbutself,
+ .send_IPI_all = default_send_IPI_all,
+ .send_IPI_self = default_send_IPI_self,
+
+ .read = native_apic_mem_read,
+ .write = native_apic_mem_write,
+ .eoi = native_apic_mem_eoi,
+ .icr_read = native_apic_icr_read,
+ .icr_write = native_apic_icr_write,
+ .wait_icr_idle = apic_mem_wait_icr_idle,
+ .safe_wait_icr_idle = apic_mem_wait_icr_idle_timeout,
+};
+
+/*
+ * We need to check for physflat first, so this order is important.
+ */
+apic_drivers(apic_physflat, apic_flat);
diff --git a/arch/x86/kernel/apic/apic_noop.c b/arch/x86/kernel/apic/apic_noop.c
new file mode 100644
index 0000000000..966d7cf10b
--- /dev/null
+++ b/arch/x86/kernel/apic/apic_noop.c
@@ -0,0 +1,78 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * NOOP APIC driver.
+ *
+ * Does almost nothing and should be substituted by a real apic driver via
+ * probe routine.
+ *
+ * Though in case if apic is disabled (for some reason) we try
+ * to not uglify the caller's code and allow to call (some) apic routines
+ * like self-ipi, etc...
+ *
+ * FIXME: Remove this gunk. The above argument which was intentionally left
+ * in place is silly to begin with because none of the callbacks except for
+ * APIC::read/write() have a WARN_ON_ONCE() in them. Sigh...
+ */
+#include <linux/cpumask.h>
+#include <linux/thread_info.h>
+
+#include <asm/apic.h>
+
+static void noop_send_IPI(int cpu, int vector) { }
+static void noop_send_IPI_mask(const struct cpumask *cpumask, int vector) { }
+static void noop_send_IPI_mask_allbutself(const struct cpumask *cpumask, int vector) { }
+static void noop_send_IPI_allbutself(int vector) { }
+static void noop_send_IPI_all(int vector) { }
+static void noop_send_IPI_self(int vector) { }
+static void noop_apic_icr_write(u32 low, u32 id) { }
+static int noop_wakeup_secondary_cpu(int apicid, unsigned long start_eip) { return -1; }
+static u64 noop_apic_icr_read(void) { return 0; }
+static int noop_phys_pkg_id(int cpuid_apic, int index_msb) { return 0; }
+static unsigned int noop_get_apic_id(unsigned long x) { return 0; }
+static void noop_apic_eoi(void) { }
+
+static u32 noop_apic_read(u32 reg)
+{
+ WARN_ON_ONCE(boot_cpu_has(X86_FEATURE_APIC) && !apic_is_disabled);
+ return 0;
+}
+
+static void noop_apic_write(u32 reg, u32 val)
+{
+ WARN_ON_ONCE(boot_cpu_has(X86_FEATURE_APIC) && !apic_is_disabled);
+}
+
+struct apic apic_noop __ro_after_init = {
+ .name = "noop",
+
+ .delivery_mode = APIC_DELIVERY_MODE_FIXED,
+ .dest_mode_logical = true,
+
+ .disable_esr = 0,
+
+ .check_apicid_used = default_check_apicid_used,
+ .ioapic_phys_id_map = default_ioapic_phys_id_map,
+ .cpu_present_to_apicid = default_cpu_present_to_apicid,
+
+ .phys_pkg_id = noop_phys_pkg_id,
+
+ .max_apic_id = 0xFE,
+ .get_apic_id = noop_get_apic_id,
+
+ .calc_dest_apicid = apic_flat_calc_apicid,
+
+ .send_IPI = noop_send_IPI,
+ .send_IPI_mask = noop_send_IPI_mask,
+ .send_IPI_mask_allbutself = noop_send_IPI_mask_allbutself,
+ .send_IPI_allbutself = noop_send_IPI_allbutself,
+ .send_IPI_all = noop_send_IPI_all,
+ .send_IPI_self = noop_send_IPI_self,
+
+ .wakeup_secondary_cpu = noop_wakeup_secondary_cpu,
+
+ .read = noop_apic_read,
+ .write = noop_apic_write,
+ .eoi = noop_apic_eoi,
+ .icr_read = noop_apic_icr_read,
+ .icr_write = noop_apic_icr_write,
+};
diff --git a/arch/x86/kernel/apic/apic_numachip.c b/arch/x86/kernel/apic/apic_numachip.c
new file mode 100644
index 0000000000..63f3d7be9d
--- /dev/null
+++ b/arch/x86/kernel/apic/apic_numachip.c
@@ -0,0 +1,292 @@
+/*
+ * 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.
+ *
+ * Numascale NumaConnect-Specific APIC Code
+ *
+ * Copyright (C) 2011 Numascale AS. All rights reserved.
+ *
+ * Send feedback to <support@numascale.com>
+ *
+ */
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/pgtable.h>
+
+#include <asm/numachip/numachip.h>
+#include <asm/numachip/numachip_csr.h>
+
+
+#include "local.h"
+
+u8 numachip_system __read_mostly;
+static const struct apic apic_numachip1;
+static const struct apic apic_numachip2;
+static void (*numachip_apic_icr_write)(int apicid, unsigned int val) __read_mostly;
+
+static unsigned int numachip1_get_apic_id(unsigned long x)
+{
+ unsigned long value;
+ unsigned int id = (x >> 24) & 0xff;
+
+ if (static_cpu_has(X86_FEATURE_NODEID_MSR)) {
+ rdmsrl(MSR_FAM10H_NODE_ID, value);
+ id |= (value << 2) & 0xff00;
+ }
+
+ return id;
+}
+
+static u32 numachip1_set_apic_id(unsigned int id)
+{
+ return (id & 0xff) << 24;
+}
+
+static unsigned int numachip2_get_apic_id(unsigned long x)
+{
+ u64 mcfg;
+
+ rdmsrl(MSR_FAM10H_MMIO_CONF_BASE, mcfg);
+ return ((mcfg >> (28 - 8)) & 0xfff00) | (x >> 24);
+}
+
+static u32 numachip2_set_apic_id(unsigned int id)
+{
+ return id << 24;
+}
+
+static int numachip_phys_pkg_id(int initial_apic_id, int index_msb)
+{
+ return initial_apic_id >> index_msb;
+}
+
+static void numachip1_apic_icr_write(int apicid, unsigned int val)
+{
+ write_lcsr(CSR_G3_EXT_IRQ_GEN, (apicid << 16) | val);
+}
+
+static void numachip2_apic_icr_write(int apicid, unsigned int val)
+{
+ numachip2_write32_lcsr(NUMACHIP2_APIC_ICR, (apicid << 12) | val);
+}
+
+static int numachip_wakeup_secondary(int phys_apicid, unsigned long start_rip)
+{
+ numachip_apic_icr_write(phys_apicid, APIC_DM_INIT);
+ numachip_apic_icr_write(phys_apicid, APIC_DM_STARTUP |
+ (start_rip >> 12));
+
+ return 0;
+}
+
+static void numachip_send_IPI_one(int cpu, int vector)
+{
+ int local_apicid, apicid = per_cpu(x86_cpu_to_apicid, cpu);
+ unsigned int dmode;
+
+ preempt_disable();
+ local_apicid = __this_cpu_read(x86_cpu_to_apicid);
+
+ /* Send via local APIC where non-local part matches */
+ if (!((apicid ^ local_apicid) >> NUMACHIP_LAPIC_BITS)) {
+ unsigned long flags;
+
+ local_irq_save(flags);
+ __default_send_IPI_dest_field(apicid, vector,
+ APIC_DEST_PHYSICAL);
+ local_irq_restore(flags);
+ preempt_enable();
+ return;
+ }
+ preempt_enable();
+
+ dmode = (vector == NMI_VECTOR) ? APIC_DM_NMI : APIC_DM_FIXED;
+ numachip_apic_icr_write(apicid, dmode | vector);
+}
+
+static void numachip_send_IPI_mask(const struct cpumask *mask, int vector)
+{
+ unsigned int cpu;
+
+ for_each_cpu(cpu, mask)
+ numachip_send_IPI_one(cpu, vector);
+}
+
+static void numachip_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)
+ numachip_send_IPI_one(cpu, vector);
+ }
+}
+
+static void numachip_send_IPI_allbutself(int vector)
+{
+ unsigned int this_cpu = smp_processor_id();
+ unsigned int cpu;
+
+ for_each_online_cpu(cpu) {
+ if (cpu != this_cpu)
+ numachip_send_IPI_one(cpu, vector);
+ }
+}
+
+static void numachip_send_IPI_all(int vector)
+{
+ numachip_send_IPI_mask(cpu_online_mask, vector);
+}
+
+static void numachip_send_IPI_self(int vector)
+{
+ apic_write(APIC_SELF_IPI, vector);
+}
+
+static int __init numachip1_probe(void)
+{
+ return apic == &apic_numachip1;
+}
+
+static int __init numachip2_probe(void)
+{
+ return apic == &apic_numachip2;
+}
+
+static void fixup_cpu_id(struct cpuinfo_x86 *c, int node)
+{
+ u64 val;
+ u32 nodes = 1;
+
+ this_cpu_write(cpu_llc_id, node);
+
+ /* Account for nodes per socket in multi-core-module processors */
+ if (boot_cpu_has(X86_FEATURE_NODEID_MSR)) {
+ rdmsrl(MSR_FAM10H_NODE_ID, val);
+ nodes = ((val >> 3) & 7) + 1;
+ }
+
+ c->phys_proc_id = node / nodes;
+}
+
+static int __init numachip_system_init(void)
+{
+ /* Map the LCSR area and set up the apic_icr_write function */
+ switch (numachip_system) {
+ case 1:
+ init_extra_mapping_uc(NUMACHIP_LCSR_BASE, NUMACHIP_LCSR_SIZE);
+ numachip_apic_icr_write = numachip1_apic_icr_write;
+ break;
+ case 2:
+ init_extra_mapping_uc(NUMACHIP2_LCSR_BASE, NUMACHIP2_LCSR_SIZE);
+ numachip_apic_icr_write = numachip2_apic_icr_write;
+ break;
+ default:
+ return 0;
+ }
+
+ x86_cpuinit.fixup_cpu_id = fixup_cpu_id;
+ x86_init.pci.arch_init = pci_numachip_init;
+
+ return 0;
+}
+early_initcall(numachip_system_init);
+
+static int numachip1_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
+{
+ if ((strncmp(oem_id, "NUMASC", 6) != 0) ||
+ (strncmp(oem_table_id, "NCONNECT", 8) != 0))
+ return 0;
+
+ numachip_system = 1;
+
+ return 1;
+}
+
+static int numachip2_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
+{
+ if ((strncmp(oem_id, "NUMASC", 6) != 0) ||
+ (strncmp(oem_table_id, "NCONECT2", 8) != 0))
+ return 0;
+
+ numachip_system = 2;
+
+ return 1;
+}
+
+static const struct apic apic_numachip1 __refconst = {
+ .name = "NumaConnect system",
+ .probe = numachip1_probe,
+ .acpi_madt_oem_check = numachip1_acpi_madt_oem_check,
+
+ .delivery_mode = APIC_DELIVERY_MODE_FIXED,
+ .dest_mode_logical = false,
+
+ .disable_esr = 0,
+
+ .cpu_present_to_apicid = default_cpu_present_to_apicid,
+ .phys_pkg_id = numachip_phys_pkg_id,
+
+ .max_apic_id = UINT_MAX,
+ .get_apic_id = numachip1_get_apic_id,
+ .set_apic_id = numachip1_set_apic_id,
+
+ .calc_dest_apicid = apic_default_calc_apicid,
+
+ .send_IPI = numachip_send_IPI_one,
+ .send_IPI_mask = numachip_send_IPI_mask,
+ .send_IPI_mask_allbutself = numachip_send_IPI_mask_allbutself,
+ .send_IPI_allbutself = numachip_send_IPI_allbutself,
+ .send_IPI_all = numachip_send_IPI_all,
+ .send_IPI_self = numachip_send_IPI_self,
+
+ .wakeup_secondary_cpu = numachip_wakeup_secondary,
+
+ .read = native_apic_mem_read,
+ .write = native_apic_mem_write,
+ .eoi = native_apic_mem_eoi,
+ .icr_read = native_apic_icr_read,
+ .icr_write = native_apic_icr_write,
+};
+
+apic_driver(apic_numachip1);
+
+static const struct apic apic_numachip2 __refconst = {
+ .name = "NumaConnect2 system",
+ .probe = numachip2_probe,
+ .acpi_madt_oem_check = numachip2_acpi_madt_oem_check,
+
+ .delivery_mode = APIC_DELIVERY_MODE_FIXED,
+ .dest_mode_logical = false,
+
+ .disable_esr = 0,
+
+ .cpu_present_to_apicid = default_cpu_present_to_apicid,
+ .phys_pkg_id = numachip_phys_pkg_id,
+
+ .max_apic_id = UINT_MAX,
+ .get_apic_id = numachip2_get_apic_id,
+ .set_apic_id = numachip2_set_apic_id,
+
+ .calc_dest_apicid = apic_default_calc_apicid,
+
+ .send_IPI = numachip_send_IPI_one,
+ .send_IPI_mask = numachip_send_IPI_mask,
+ .send_IPI_mask_allbutself = numachip_send_IPI_mask_allbutself,
+ .send_IPI_allbutself = numachip_send_IPI_allbutself,
+ .send_IPI_all = numachip_send_IPI_all,
+ .send_IPI_self = numachip_send_IPI_self,
+
+ .wakeup_secondary_cpu = numachip_wakeup_secondary,
+
+ .read = native_apic_mem_read,
+ .write = native_apic_mem_write,
+ .eoi = native_apic_mem_eoi,
+ .icr_read = native_apic_icr_read,
+ .icr_write = native_apic_icr_write,
+};
+
+apic_driver(apic_numachip2);
diff --git a/arch/x86/kernel/apic/bigsmp_32.c b/arch/x86/kernel/apic/bigsmp_32.c
new file mode 100644
index 0000000000..0e5535add4
--- /dev/null
+++ b/arch/x86/kernel/apic/bigsmp_32.c
@@ -0,0 +1,126 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * APIC driver for "bigsmp" xAPIC machines with more than 8 virtual CPUs.
+ *
+ * Drives the local APIC in "clustered mode".
+ */
+#include <linux/cpumask.h>
+#include <linux/dmi.h>
+#include <linux/smp.h>
+
+#include <asm/apic.h>
+#include <asm/io_apic.h>
+
+#include "local.h"
+
+static unsigned bigsmp_get_apic_id(unsigned long x)
+{
+ return (x >> 24) & 0xFF;
+}
+
+static bool bigsmp_check_apicid_used(physid_mask_t *map, int apicid)
+{
+ return false;
+}
+
+static void bigsmp_ioapic_phys_id_map(physid_mask_t *phys_map, physid_mask_t *retmap)
+{
+ /* For clustered we don't have a good way to do this yet - hack */
+ physids_promote(0xFFL, retmap);
+}
+
+static int bigsmp_phys_pkg_id(int cpuid_apic, int index_msb)
+{
+ return cpuid_apic >> index_msb;
+}
+
+static void bigsmp_send_IPI_allbutself(int vector)
+{
+ default_send_IPI_mask_allbutself_phys(cpu_online_mask, vector);
+}
+
+static void bigsmp_send_IPI_all(int vector)
+{
+ default_send_IPI_mask_sequence_phys(cpu_online_mask, vector);
+}
+
+static int dmi_bigsmp; /* can be set by dmi scanners */
+
+static int hp_ht_bigsmp(const struct dmi_system_id *d)
+{
+ printk(KERN_NOTICE "%s detected: force use of apic=bigsmp\n", d->ident);
+ dmi_bigsmp = 1;
+
+ return 0;
+}
+
+
+static const struct dmi_system_id bigsmp_dmi_table[] = {
+ { hp_ht_bigsmp, "HP ProLiant DL760 G2",
+ { DMI_MATCH(DMI_BIOS_VENDOR, "HP"),
+ DMI_MATCH(DMI_BIOS_VERSION, "P44-"),
+ }
+ },
+
+ { hp_ht_bigsmp, "HP ProLiant DL740",
+ { DMI_MATCH(DMI_BIOS_VENDOR, "HP"),
+ DMI_MATCH(DMI_BIOS_VERSION, "P47-"),
+ }
+ },
+ { } /* NULL entry stops DMI scanning */
+};
+
+static int probe_bigsmp(void)
+{
+ return dmi_check_system(bigsmp_dmi_table);
+}
+
+static struct apic apic_bigsmp __ro_after_init = {
+
+ .name = "bigsmp",
+ .probe = probe_bigsmp,
+
+ .delivery_mode = APIC_DELIVERY_MODE_FIXED,
+ .dest_mode_logical = false,
+
+ .disable_esr = 1,
+
+ .check_apicid_used = bigsmp_check_apicid_used,
+ .ioapic_phys_id_map = bigsmp_ioapic_phys_id_map,
+ .cpu_present_to_apicid = default_cpu_present_to_apicid,
+ .phys_pkg_id = bigsmp_phys_pkg_id,
+
+ .max_apic_id = 0xFE,
+ .get_apic_id = bigsmp_get_apic_id,
+ .set_apic_id = NULL,
+
+ .calc_dest_apicid = apic_default_calc_apicid,
+
+ .send_IPI = default_send_IPI_single_phys,
+ .send_IPI_mask = default_send_IPI_mask_sequence_phys,
+ .send_IPI_mask_allbutself = NULL,
+ .send_IPI_allbutself = bigsmp_send_IPI_allbutself,
+ .send_IPI_all = bigsmp_send_IPI_all,
+ .send_IPI_self = default_send_IPI_self,
+
+ .read = native_apic_mem_read,
+ .write = native_apic_mem_write,
+ .eoi = native_apic_mem_eoi,
+ .icr_read = native_apic_icr_read,
+ .icr_write = native_apic_icr_write,
+ .wait_icr_idle = apic_mem_wait_icr_idle,
+ .safe_wait_icr_idle = apic_mem_wait_icr_idle_timeout,
+};
+
+bool __init apic_bigsmp_possible(bool cmdline_override)
+{
+ return apic == &apic_bigsmp || !cmdline_override;
+}
+
+void __init apic_bigsmp_force(void)
+{
+ if (apic != &apic_bigsmp)
+ apic_install_driver(&apic_bigsmp);
+}
+
+apic_driver(apic_bigsmp);
diff --git a/arch/x86/kernel/apic/hw_nmi.c b/arch/x86/kernel/apic/hw_nmi.c
new file mode 100644
index 0000000000..45af535c44
--- /dev/null
+++ b/arch/x86/kernel/apic/hw_nmi.c
@@ -0,0 +1,61 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * HW NMI watchdog support
+ *
+ * started by Don Zickus, Copyright (C) 2010 Red Hat, Inc.
+ *
+ * Arch specific calls to support NMI watchdog
+ *
+ * Bits copied from original nmi.c file
+ *
+ */
+#include <linux/thread_info.h>
+#include <asm/apic.h>
+#include <asm/nmi.h>
+
+#include <linux/cpumask.h>
+#include <linux/kdebug.h>
+#include <linux/notifier.h>
+#include <linux/kprobes.h>
+#include <linux/nmi.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+
+#include "local.h"
+
+#ifdef CONFIG_HARDLOCKUP_DETECTOR_PERF
+u64 hw_nmi_get_sample_period(int watchdog_thresh)
+{
+ return (u64)(cpu_khz) * 1000 * watchdog_thresh;
+}
+#endif
+
+#ifdef arch_trigger_cpumask_backtrace
+static void nmi_raise_cpu_backtrace(cpumask_t *mask)
+{
+ __apic_send_IPI_mask(mask, NMI_VECTOR);
+}
+
+void arch_trigger_cpumask_backtrace(const cpumask_t *mask, int exclude_cpu)
+{
+ nmi_trigger_cpumask_backtrace(mask, exclude_cpu,
+ nmi_raise_cpu_backtrace);
+}
+
+static int nmi_cpu_backtrace_handler(unsigned int cmd, struct pt_regs *regs)
+{
+ if (nmi_cpu_backtrace(regs))
+ return NMI_HANDLED;
+
+ return NMI_DONE;
+}
+NOKPROBE_SYMBOL(nmi_cpu_backtrace_handler);
+
+static int __init register_nmi_cpu_backtrace_handler(void)
+{
+ register_nmi_handler(NMI_LOCAL, nmi_cpu_backtrace_handler,
+ 0, "arch_bt");
+ return 0;
+}
+early_initcall(register_nmi_cpu_backtrace_handler);
+#endif
diff --git a/arch/x86/kernel/apic/init.c b/arch/x86/kernel/apic/init.c
new file mode 100644
index 0000000000..821e2e536f
--- /dev/null
+++ b/arch/x86/kernel/apic/init.c
@@ -0,0 +1,110 @@
+// SPDX-License-Identifier: GPL-2.0-only
+#define pr_fmt(fmt) "APIC: " fmt
+
+#include <asm/apic.h>
+
+#include "local.h"
+
+/*
+ * Use DEFINE_STATIC_CALL_NULL() to avoid having to provide stub functions
+ * for each callback. The callbacks are setup during boot and all except
+ * wait_icr_idle() must be initialized before usage. The IPI wrappers
+ * use static_call() and not static_call_cond() to catch any fails.
+ */
+#define DEFINE_APIC_CALL(__cb) \
+ DEFINE_STATIC_CALL_NULL(apic_call_##__cb, *apic->__cb)
+
+DEFINE_APIC_CALL(eoi);
+DEFINE_APIC_CALL(native_eoi);
+DEFINE_APIC_CALL(icr_read);
+DEFINE_APIC_CALL(icr_write);
+DEFINE_APIC_CALL(read);
+DEFINE_APIC_CALL(send_IPI);
+DEFINE_APIC_CALL(send_IPI_mask);
+DEFINE_APIC_CALL(send_IPI_mask_allbutself);
+DEFINE_APIC_CALL(send_IPI_allbutself);
+DEFINE_APIC_CALL(send_IPI_all);
+DEFINE_APIC_CALL(send_IPI_self);
+DEFINE_APIC_CALL(wait_icr_idle);
+DEFINE_APIC_CALL(wakeup_secondary_cpu);
+DEFINE_APIC_CALL(wakeup_secondary_cpu_64);
+DEFINE_APIC_CALL(write);
+
+EXPORT_STATIC_CALL_TRAMP_GPL(apic_call_send_IPI_mask);
+EXPORT_STATIC_CALL_TRAMP_GPL(apic_call_send_IPI_self);
+
+/* The container for function call overrides */
+struct apic_override __x86_apic_override __initdata;
+
+#define apply_override(__cb) \
+ if (__x86_apic_override.__cb) \
+ apic->__cb = __x86_apic_override.__cb
+
+static __init void restore_override_callbacks(void)
+{
+ apply_override(eoi);
+ apply_override(native_eoi);
+ apply_override(write);
+ apply_override(read);
+ apply_override(send_IPI);
+ apply_override(send_IPI_mask);
+ apply_override(send_IPI_mask_allbutself);
+ apply_override(send_IPI_allbutself);
+ apply_override(send_IPI_all);
+ apply_override(send_IPI_self);
+ apply_override(icr_read);
+ apply_override(icr_write);
+ apply_override(wakeup_secondary_cpu);
+ apply_override(wakeup_secondary_cpu_64);
+}
+
+#define update_call(__cb) \
+ static_call_update(apic_call_##__cb, *apic->__cb)
+
+static __init void update_static_calls(void)
+{
+ update_call(eoi);
+ update_call(native_eoi);
+ update_call(write);
+ update_call(read);
+ update_call(send_IPI);
+ update_call(send_IPI_mask);
+ update_call(send_IPI_mask_allbutself);
+ update_call(send_IPI_allbutself);
+ update_call(send_IPI_all);
+ update_call(send_IPI_self);
+ update_call(icr_read);
+ update_call(icr_write);
+ update_call(wait_icr_idle);
+ update_call(wakeup_secondary_cpu);
+ update_call(wakeup_secondary_cpu_64);
+}
+
+void __init apic_setup_apic_calls(void)
+{
+ /* Ensure that the default APIC has native_eoi populated */
+ apic->native_eoi = apic->eoi;
+ update_static_calls();
+ pr_info("Static calls initialized\n");
+}
+
+void __init apic_install_driver(struct apic *driver)
+{
+ if (apic == driver)
+ return;
+
+ apic = driver;
+
+ if (IS_ENABLED(CONFIG_X86_X2APIC) && apic->x2apic_set_max_apicid)
+ apic->max_apic_id = x2apic_max_apicid;
+
+ /* Copy the original eoi() callback as KVM/HyperV might overwrite it */
+ if (!apic->native_eoi)
+ apic->native_eoi = apic->eoi;
+
+ /* Apply any already installed callback overrides */
+ restore_override_callbacks();
+ update_static_calls();
+
+ pr_info("Switched APIC routing to: %s\n", driver->name);
+}
diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
new file mode 100644
index 0000000000..00da6cf6b0
--- /dev/null
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -0,0 +1,3116 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * 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
+ *
+ * Historical information which is worth to be preserved:
+ *
+ * - SiS APIC rmw bug:
+ *
+ * We used to have a workaround for a bug in SiS chips which
+ * required to rewrite the index register for a read-modify-write
+ * operation as the chip lost the index information which was
+ * setup for the read already. We cache the data now, so that
+ * workaround has been removed.
+ */
+
+#include <linux/mm.h>
+#include <linux/interrupt.h>
+#include <linux/irq.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/export.h>
+#include <linux/syscore_ops.h>
+#include <linux/freezer.h>
+#include <linux/kthread.h>
+#include <linux/jiffies.h> /* time_after() */
+#include <linux/slab.h>
+#include <linux/memblock.h>
+#include <linux/msi.h>
+
+#include <asm/irqdomain.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/time.h>
+#include <asm/i8259.h>
+#include <asm/setup.h>
+#include <asm/irq_remapping.h>
+#include <asm/hw_irq.h>
+#include <asm/apic.h>
+#include <asm/pgtable.h>
+#include <asm/x86_init.h>
+
+#define for_each_ioapic(idx) \
+ for ((idx) = 0; (idx) < nr_ioapics; (idx)++)
+#define for_each_ioapic_reverse(idx) \
+ for ((idx) = nr_ioapics - 1; (idx) >= 0; (idx)--)
+#define for_each_pin(idx, pin) \
+ for ((pin) = 0; (pin) < ioapics[(idx)].nr_registers; (pin)++)
+#define for_each_ioapic_pin(idx, pin) \
+ for_each_ioapic((idx)) \
+ for_each_pin((idx), (pin))
+#define for_each_irq_pin(entry, head) \
+ list_for_each_entry(entry, &head, list)
+
+static DEFINE_RAW_SPINLOCK(ioapic_lock);
+static DEFINE_MUTEX(ioapic_mutex);
+static unsigned int ioapic_dynirq_base;
+static int ioapic_initialized;
+
+struct irq_pin_list {
+ struct list_head list;
+ int apic, pin;
+};
+
+struct mp_chip_data {
+ struct list_head irq_2_pin;
+ struct IO_APIC_route_entry entry;
+ bool is_level;
+ bool active_low;
+ bool isa_irq;
+ u32 count;
+};
+
+struct mp_ioapic_gsi {
+ u32 gsi_base;
+ u32 gsi_end;
+};
+
+static struct ioapic {
+ /*
+ * # of IRQ routing registers
+ */
+ int nr_registers;
+ /*
+ * Saved state during suspend/resume, or while enabling intr-remap.
+ */
+ struct IO_APIC_route_entry *saved_registers;
+ /* I/O APIC config */
+ struct mpc_ioapic mp_config;
+ /* IO APIC gsi routing info */
+ struct mp_ioapic_gsi gsi_config;
+ struct ioapic_domain_cfg irqdomain_cfg;
+ struct irq_domain *irqdomain;
+ struct resource *iomem_res;
+} ioapics[MAX_IO_APICS];
+
+#define mpc_ioapic_ver(ioapic_idx) ioapics[ioapic_idx].mp_config.apicver
+
+int mpc_ioapic_id(int ioapic_idx)
+{
+ return ioapics[ioapic_idx].mp_config.apicid;
+}
+
+unsigned int mpc_ioapic_addr(int ioapic_idx)
+{
+ return ioapics[ioapic_idx].mp_config.apicaddr;
+}
+
+static inline struct mp_ioapic_gsi *mp_ioapic_gsi_routing(int ioapic_idx)
+{
+ return &ioapics[ioapic_idx].gsi_config;
+}
+
+static inline int mp_ioapic_pin_count(int ioapic)
+{
+ struct mp_ioapic_gsi *gsi_cfg = mp_ioapic_gsi_routing(ioapic);
+
+ return gsi_cfg->gsi_end - gsi_cfg->gsi_base + 1;
+}
+
+static inline u32 mp_pin_to_gsi(int ioapic, int pin)
+{
+ return mp_ioapic_gsi_routing(ioapic)->gsi_base + pin;
+}
+
+static inline bool mp_is_legacy_irq(int irq)
+{
+ return irq >= 0 && irq < nr_legacy_irqs();
+}
+
+static inline struct irq_domain *mp_ioapic_irqdomain(int ioapic)
+{
+ return ioapics[ioapic].irqdomain;
+}
+
+int nr_ioapics;
+
+/* The one past the highest gsi number used */
+u32 gsi_top;
+
+/* MP IRQ source entries */
+struct mpc_intsrc mp_irqs[MAX_IRQ_SOURCES];
+
+/* # of MP IRQ source entries */
+int mp_irq_entries;
+
+#ifdef CONFIG_EISA
+int mp_bus_id_to_type[MAX_MP_BUSSES];
+#endif
+
+DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES);
+
+bool ioapic_is_disabled __ro_after_init;
+
+/**
+ * disable_ioapic_support() - disables ioapic support at runtime
+ */
+void disable_ioapic_support(void)
+{
+#ifdef CONFIG_PCI
+ noioapicquirk = 1;
+ noioapicreroute = -1;
+#endif
+ ioapic_is_disabled = true;
+}
+
+static int __init parse_noapic(char *str)
+{
+ /* disable IO-APIC */
+ disable_ioapic_support();
+ return 0;
+}
+early_param("noapic", parse_noapic);
+
+/* Will be called in mpparse/ACPI codes for saving IRQ info */
+void mp_save_irq(struct mpc_intsrc *m)
+{
+ int i;
+
+ apic_printk(APIC_VERBOSE, "Int: type %d, pol %d, trig %d, bus %02x,"
+ " IRQ %02x, APIC ID %x, APIC INT %02x\n",
+ m->irqtype, m->irqflag & 3, (m->irqflag >> 2) & 3, m->srcbus,
+ m->srcbusirq, m->dstapic, m->dstirq);
+
+ for (i = 0; i < mp_irq_entries; i++) {
+ if (!memcmp(&mp_irqs[i], m, sizeof(*m)))
+ return;
+ }
+
+ memcpy(&mp_irqs[mp_irq_entries], m, sizeof(*m));
+ if (++mp_irq_entries == MAX_IRQ_SOURCES)
+ panic("Max # of irq sources exceeded!!\n");
+}
+
+static void alloc_ioapic_saved_registers(int idx)
+{
+ size_t size;
+
+ if (ioapics[idx].saved_registers)
+ return;
+
+ size = sizeof(struct IO_APIC_route_entry) * ioapics[idx].nr_registers;
+ ioapics[idx].saved_registers = kzalloc(size, GFP_KERNEL);
+ if (!ioapics[idx].saved_registers)
+ pr_err("IOAPIC %d: suspend/resume impossible!\n", idx);
+}
+
+static void free_ioapic_saved_registers(int idx)
+{
+ kfree(ioapics[idx].saved_registers);
+ ioapics[idx].saved_registers = NULL;
+}
+
+int __init arch_early_ioapic_init(void)
+{
+ int i;
+
+ if (!nr_legacy_irqs())
+ io_apic_irqs = ~0UL;
+
+ for_each_ioapic(i)
+ alloc_ioapic_saved_registers(i);
+
+ return 0;
+}
+
+struct io_apic {
+ unsigned int index;
+ unsigned int unused[3];
+ unsigned int data;
+ unsigned int unused2[11];
+ unsigned int eoi;
+};
+
+static __attribute_const__ struct io_apic __iomem *io_apic_base(int idx)
+{
+ return (void __iomem *) __fix_to_virt(FIX_IO_APIC_BASE_0 + idx)
+ + (mpc_ioapic_addr(idx) & ~PAGE_MASK);
+}
+
+static inline void io_apic_eoi(unsigned int apic, unsigned int vector)
+{
+ struct io_apic __iomem *io_apic = io_apic_base(apic);
+ writel(vector, &io_apic->eoi);
+}
+
+unsigned int native_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 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);
+}
+
+static struct IO_APIC_route_entry __ioapic_read_entry(int apic, int pin)
+{
+ struct IO_APIC_route_entry entry;
+
+ entry.w1 = io_apic_read(apic, 0x10 + 2 * pin);
+ entry.w2 = io_apic_read(apic, 0x11 + 2 * pin);
+
+ return entry;
+}
+
+static struct IO_APIC_route_entry ioapic_read_entry(int apic, int pin)
+{
+ struct IO_APIC_route_entry entry;
+ unsigned long flags;
+
+ raw_spin_lock_irqsave(&ioapic_lock, flags);
+ entry = __ioapic_read_entry(apic, pin);
+ raw_spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ return 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)
+{
+ io_apic_write(apic, 0x11 + 2*pin, e.w2);
+ io_apic_write(apic, 0x10 + 2*pin, e.w1);
+}
+
+static void ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e)
+{
+ unsigned long flags;
+
+ raw_spin_lock_irqsave(&ioapic_lock, flags);
+ __ioapic_write_entry(apic, pin, e);
+ raw_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)
+{
+ struct IO_APIC_route_entry e = { .masked = true };
+ unsigned long flags;
+
+ raw_spin_lock_irqsave(&ioapic_lock, flags);
+ io_apic_write(apic, 0x10 + 2*pin, e.w1);
+ io_apic_write(apic, 0x11 + 2*pin, e.w2);
+ raw_spin_unlock_irqrestore(&ioapic_lock, flags);
+}
+
+/*
+ * 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 int __add_pin_to_irq_node(struct mp_chip_data *data,
+ int node, int apic, int pin)
+{
+ struct irq_pin_list *entry;
+
+ /* don't allow duplicates */
+ for_each_irq_pin(entry, data->irq_2_pin)
+ if (entry->apic == apic && entry->pin == pin)
+ return 0;
+
+ entry = kzalloc_node(sizeof(struct irq_pin_list), GFP_ATOMIC, node);
+ if (!entry) {
+ pr_err("can not alloc irq_pin_list (%d,%d,%d)\n",
+ node, apic, pin);
+ return -ENOMEM;
+ }
+ entry->apic = apic;
+ entry->pin = pin;
+ list_add_tail(&entry->list, &data->irq_2_pin);
+
+ return 0;
+}
+
+static void __remove_pin_from_irq(struct mp_chip_data *data, int apic, int pin)
+{
+ struct irq_pin_list *tmp, *entry;
+
+ list_for_each_entry_safe(entry, tmp, &data->irq_2_pin, list)
+ if (entry->apic == apic && entry->pin == pin) {
+ list_del(&entry->list);
+ kfree(entry);
+ return;
+ }
+}
+
+static void add_pin_to_irq_node(struct mp_chip_data *data,
+ int node, int apic, int pin)
+{
+ if (__add_pin_to_irq_node(data, node, apic, pin))
+ panic("IO-APIC: failed to add irq-pin. Can not proceed\n");
+}
+
+/*
+ * Reroute an IRQ to a different pin.
+ */
+static void __init replace_pin_at_irq_node(struct mp_chip_data *data, int node,
+ int oldapic, int oldpin,
+ int newapic, int newpin)
+{
+ struct irq_pin_list *entry;
+
+ for_each_irq_pin(entry, data->irq_2_pin) {
+ if (entry->apic == oldapic && entry->pin == oldpin) {
+ entry->apic = newapic;
+ entry->pin = newpin;
+ /* every one is different, right? */
+ return;
+ }
+ }
+
+ /* old apic/pin didn't exist, so just add new ones */
+ add_pin_to_irq_node(data, node, newapic, newpin);
+}
+
+static void io_apic_modify_irq(struct mp_chip_data *data, bool masked,
+ void (*final)(struct irq_pin_list *entry))
+{
+ struct irq_pin_list *entry;
+
+ data->entry.masked = masked;
+
+ for_each_irq_pin(entry, data->irq_2_pin) {
+ io_apic_write(entry->apic, 0x10 + 2 * entry->pin, data->entry.w1);
+ if (final)
+ final(entry);
+ }
+}
+
+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_ioapic_irq(struct irq_data *irq_data)
+{
+ struct mp_chip_data *data = irq_data->chip_data;
+ unsigned long flags;
+
+ raw_spin_lock_irqsave(&ioapic_lock, flags);
+ io_apic_modify_irq(data, true, &io_apic_sync);
+ raw_spin_unlock_irqrestore(&ioapic_lock, flags);
+}
+
+static void __unmask_ioapic(struct mp_chip_data *data)
+{
+ io_apic_modify_irq(data, false, NULL);
+}
+
+static void unmask_ioapic_irq(struct irq_data *irq_data)
+{
+ struct mp_chip_data *data = irq_data->chip_data;
+ unsigned long flags;
+
+ raw_spin_lock_irqsave(&ioapic_lock, flags);
+ __unmask_ioapic(data);
+ raw_spin_unlock_irqrestore(&ioapic_lock, flags);
+}
+
+/*
+ * IO-APIC versions below 0x20 don't support EOI register.
+ * For the record, here is the information about various versions:
+ * 0Xh 82489DX
+ * 1Xh I/OAPIC or I/O(x)APIC which are not PCI 2.2 Compliant
+ * 2Xh I/O(x)APIC which is PCI 2.2 Compliant
+ * 30h-FFh Reserved
+ *
+ * Some of the Intel ICH Specs (ICH2 to ICH5) documents the io-apic
+ * version as 0x2. This is an error with documentation and these ICH chips
+ * use io-apic's of version 0x20.
+ *
+ * For IO-APIC's with EOI register, we use that to do an explicit EOI.
+ * Otherwise, we simulate the EOI message manually by changing the trigger
+ * mode to edge and then back to level, with RTE being masked during this.
+ */
+static void __eoi_ioapic_pin(int apic, int pin, int vector)
+{
+ if (mpc_ioapic_ver(apic) >= 0x20) {
+ io_apic_eoi(apic, vector);
+ } else {
+ struct IO_APIC_route_entry entry, entry1;
+
+ entry = entry1 = __ioapic_read_entry(apic, pin);
+
+ /*
+ * Mask the entry and change the trigger mode to edge.
+ */
+ entry1.masked = true;
+ entry1.is_level = false;
+
+ __ioapic_write_entry(apic, pin, entry1);
+
+ /*
+ * Restore the previous level triggered entry.
+ */
+ __ioapic_write_entry(apic, pin, entry);
+ }
+}
+
+static void eoi_ioapic_pin(int vector, struct mp_chip_data *data)
+{
+ unsigned long flags;
+ struct irq_pin_list *entry;
+
+ raw_spin_lock_irqsave(&ioapic_lock, flags);
+ for_each_irq_pin(entry, data->irq_2_pin)
+ __eoi_ioapic_pin(entry->apic, entry->pin, vector);
+ raw_spin_unlock_irqrestore(&ioapic_lock, flags);
+}
+
+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 == APIC_DELIVERY_MODE_SMI)
+ return;
+
+ /*
+ * Make sure the entry is masked and re-read the contents to check
+ * if it is a level triggered pin and if the remote-IRR is set.
+ */
+ if (!entry.masked) {
+ entry.masked = true;
+ ioapic_write_entry(apic, pin, entry);
+ entry = ioapic_read_entry(apic, pin);
+ }
+
+ if (entry.irr) {
+ unsigned long flags;
+
+ /*
+ * Make sure the trigger mode is set to level. Explicit EOI
+ * doesn't clear the remote-IRR if the trigger mode is not
+ * set to level.
+ */
+ if (!entry.is_level) {
+ entry.is_level = true;
+ ioapic_write_entry(apic, pin, entry);
+ }
+ raw_spin_lock_irqsave(&ioapic_lock, flags);
+ __eoi_ioapic_pin(apic, pin, entry.vector);
+ raw_spin_unlock_irqrestore(&ioapic_lock, flags);
+ }
+
+ /*
+ * Clear the rest of the bits in the IO-APIC RTE except for the mask
+ * bit.
+ */
+ ioapic_mask_entry(apic, pin);
+ entry = ioapic_read_entry(apic, pin);
+ if (entry.irr)
+ pr_err("Unable to reset IRR for apic: %d, pin :%d\n",
+ mpc_ioapic_id(apic), pin);
+}
+
+void clear_IO_APIC (void)
+{
+ int apic, pin;
+
+ for_each_ioapic_pin(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 */
+
+/*
+ * Saves all the IO-APIC RTE's
+ */
+int save_ioapic_entries(void)
+{
+ int apic, pin;
+ int err = 0;
+
+ for_each_ioapic(apic) {
+ if (!ioapics[apic].saved_registers) {
+ err = -ENOMEM;
+ continue;
+ }
+
+ for_each_pin(apic, pin)
+ ioapics[apic].saved_registers[pin] =
+ ioapic_read_entry(apic, pin);
+ }
+
+ return err;
+}
+
+/*
+ * Mask all IO APIC entries.
+ */
+void mask_ioapic_entries(void)
+{
+ int apic, pin;
+
+ for_each_ioapic(apic) {
+ if (!ioapics[apic].saved_registers)
+ continue;
+
+ for_each_pin(apic, pin) {
+ struct IO_APIC_route_entry entry;
+
+ entry = ioapics[apic].saved_registers[pin];
+ if (!entry.masked) {
+ entry.masked = true;
+ ioapic_write_entry(apic, pin, entry);
+ }
+ }
+ }
+}
+
+/*
+ * Restore IO APIC entries which was saved in the ioapic structure.
+ */
+int restore_ioapic_entries(void)
+{
+ int apic, pin;
+
+ for_each_ioapic(apic) {
+ if (!ioapics[apic].saved_registers)
+ continue;
+
+ for_each_pin(apic, pin)
+ ioapic_write_entry(apic, pin,
+ ioapics[apic].saved_registers[pin]);
+ }
+ return 0;
+}
+
+/*
+ * Find the IRQ entry number of a certain pin.
+ */
+static int find_irq_entry(int ioapic_idx, int pin, int type)
+{
+ int i;
+
+ for (i = 0; i < mp_irq_entries; i++)
+ if (mp_irqs[i].irqtype == type &&
+ (mp_irqs[i].dstapic == mpc_ioapic_id(ioapic_idx) ||
+ 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 ioapic_idx;
+
+ for_each_ioapic(ioapic_idx)
+ if (mpc_ioapic_id(ioapic_idx) == mp_irqs[i].dstapic)
+ return ioapic_idx;
+ }
+
+ return -1;
+}
+
+static bool irq_active_low(int idx)
+{
+ int bus = mp_irqs[idx].srcbus;
+
+ /*
+ * Determine IRQ line polarity (high active or low active):
+ */
+ switch (mp_irqs[idx].irqflag & MP_IRQPOL_MASK) {
+ case MP_IRQPOL_DEFAULT:
+ /*
+ * Conforms to spec, ie. bus-type dependent polarity. PCI
+ * defaults to low active. [E]ISA defaults to high active.
+ */
+ return !test_bit(bus, mp_bus_not_pci);
+ case MP_IRQPOL_ACTIVE_HIGH:
+ return false;
+ case MP_IRQPOL_RESERVED:
+ pr_warn("IOAPIC: Invalid polarity: 2, defaulting to low\n");
+ fallthrough;
+ case MP_IRQPOL_ACTIVE_LOW:
+ default: /* Pointless default required due to do gcc stupidity */
+ return true;
+ }
+}
+
+#ifdef CONFIG_EISA
+/*
+ * EISA Edge/Level control register, ELCR
+ */
+static bool EISA_ELCR(unsigned int irq)
+{
+ if (irq < nr_legacy_irqs()) {
+ unsigned int port = PIC_ELCR1 + (irq >> 3);
+ return (inb(port) >> (irq & 7)) & 1;
+ }
+ apic_printk(APIC_VERBOSE, KERN_INFO
+ "Broken MPtable reports ISA irq %d\n", irq);
+ return false;
+}
+
+/*
+ * EISA interrupts are always active high and can be edge or level
+ * triggered 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.
+ */
+static bool eisa_irq_is_level(int idx, int bus, bool level)
+{
+ switch (mp_bus_id_to_type[bus]) {
+ case MP_BUS_PCI:
+ case MP_BUS_ISA:
+ return level;
+ case MP_BUS_EISA:
+ return EISA_ELCR(mp_irqs[idx].srcbusirq);
+ }
+ pr_warn("IOAPIC: Invalid srcbus: %d defaulting to level\n", bus);
+ return true;
+}
+#else
+static inline int eisa_irq_is_level(int idx, int bus, bool level)
+{
+ return level;
+}
+#endif
+
+static bool irq_is_level(int idx)
+{
+ int bus = mp_irqs[idx].srcbus;
+ bool level;
+
+ /*
+ * Determine IRQ trigger mode (edge or level sensitive):
+ */
+ switch (mp_irqs[idx].irqflag & MP_IRQTRIG_MASK) {
+ case MP_IRQTRIG_DEFAULT:
+ /*
+ * Conforms to spec, ie. bus-type dependent trigger
+ * mode. PCI defaults to level, ISA to edge.
+ */
+ level = !test_bit(bus, mp_bus_not_pci);
+ /* Take EISA into account */
+ return eisa_irq_is_level(idx, bus, level);
+ case MP_IRQTRIG_EDGE:
+ return false;
+ case MP_IRQTRIG_RESERVED:
+ pr_warn("IOAPIC: Invalid trigger mode 2 defaulting to level\n");
+ fallthrough;
+ case MP_IRQTRIG_LEVEL:
+ default: /* Pointless default required due to do gcc stupidity */
+ return true;
+ }
+}
+
+static int __acpi_get_override_irq(u32 gsi, bool *trigger, bool *polarity)
+{
+ int ioapic, pin, idx;
+
+ if (ioapic_is_disabled)
+ return -1;
+
+ ioapic = mp_find_ioapic(gsi);
+ if (ioapic < 0)
+ return -1;
+
+ pin = mp_find_ioapic_pin(ioapic, gsi);
+ if (pin < 0)
+ return -1;
+
+ idx = find_irq_entry(ioapic, pin, mp_INT);
+ if (idx < 0)
+ return -1;
+
+ *trigger = irq_is_level(idx);
+ *polarity = irq_active_low(idx);
+ return 0;
+}
+
+#ifdef CONFIG_ACPI
+int acpi_get_override_irq(u32 gsi, int *is_level, int *active_low)
+{
+ *is_level = *active_low = 0;
+ return __acpi_get_override_irq(gsi, (bool *)is_level,
+ (bool *)active_low);
+}
+#endif
+
+void ioapic_set_alloc_attr(struct irq_alloc_info *info, int node,
+ int trigger, int polarity)
+{
+ init_irq_alloc_info(info, NULL);
+ info->type = X86_IRQ_ALLOC_TYPE_IOAPIC;
+ info->ioapic.node = node;
+ info->ioapic.is_level = trigger;
+ info->ioapic.active_low = polarity;
+ info->ioapic.valid = 1;
+}
+
+static void ioapic_copy_alloc_attr(struct irq_alloc_info *dst,
+ struct irq_alloc_info *src,
+ u32 gsi, int ioapic_idx, int pin)
+{
+ bool level, pol_low;
+
+ copy_irq_alloc_info(dst, src);
+ dst->type = X86_IRQ_ALLOC_TYPE_IOAPIC;
+ dst->devid = mpc_ioapic_id(ioapic_idx);
+ dst->ioapic.pin = pin;
+ dst->ioapic.valid = 1;
+ if (src && src->ioapic.valid) {
+ dst->ioapic.node = src->ioapic.node;
+ dst->ioapic.is_level = src->ioapic.is_level;
+ dst->ioapic.active_low = src->ioapic.active_low;
+ } else {
+ dst->ioapic.node = NUMA_NO_NODE;
+ if (__acpi_get_override_irq(gsi, &level, &pol_low) >= 0) {
+ dst->ioapic.is_level = level;
+ dst->ioapic.active_low = pol_low;
+ } else {
+ /*
+ * PCI interrupts are always active low level
+ * triggered.
+ */
+ dst->ioapic.is_level = true;
+ dst->ioapic.active_low = true;
+ }
+ }
+}
+
+static int ioapic_alloc_attr_node(struct irq_alloc_info *info)
+{
+ return (info && info->ioapic.valid) ? info->ioapic.node : NUMA_NO_NODE;
+}
+
+static void mp_register_handler(unsigned int irq, bool level)
+{
+ irq_flow_handler_t hdl;
+ bool fasteoi;
+
+ if (level) {
+ irq_set_status_flags(irq, IRQ_LEVEL);
+ fasteoi = true;
+ } else {
+ irq_clear_status_flags(irq, IRQ_LEVEL);
+ fasteoi = false;
+ }
+
+ hdl = fasteoi ? handle_fasteoi_irq : handle_edge_irq;
+ __irq_set_handler(irq, hdl, 0, fasteoi ? "fasteoi" : "edge");
+}
+
+static bool mp_check_pin_attr(int irq, struct irq_alloc_info *info)
+{
+ struct mp_chip_data *data = irq_get_chip_data(irq);
+
+ /*
+ * setup_IO_APIC_irqs() programs all legacy IRQs with default trigger
+ * and polarity attributes. So allow the first user to reprogram the
+ * pin with real trigger and polarity attributes.
+ */
+ if (irq < nr_legacy_irqs() && data->count == 1) {
+ if (info->ioapic.is_level != data->is_level)
+ mp_register_handler(irq, info->ioapic.is_level);
+ data->entry.is_level = data->is_level = info->ioapic.is_level;
+ data->entry.active_low = data->active_low = info->ioapic.active_low;
+ }
+
+ return data->is_level == info->ioapic.is_level &&
+ data->active_low == info->ioapic.active_low;
+}
+
+static int alloc_irq_from_domain(struct irq_domain *domain, int ioapic, u32 gsi,
+ struct irq_alloc_info *info)
+{
+ bool legacy = false;
+ int irq = -1;
+ int type = ioapics[ioapic].irqdomain_cfg.type;
+
+ switch (type) {
+ case IOAPIC_DOMAIN_LEGACY:
+ /*
+ * Dynamically allocate IRQ number for non-ISA IRQs in the first
+ * 16 GSIs on some weird platforms.
+ */
+ if (!ioapic_initialized || gsi >= nr_legacy_irqs())
+ irq = gsi;
+ legacy = mp_is_legacy_irq(irq);
+ break;
+ case IOAPIC_DOMAIN_STRICT:
+ irq = gsi;
+ break;
+ case IOAPIC_DOMAIN_DYNAMIC:
+ break;
+ default:
+ WARN(1, "ioapic: unknown irqdomain type %d\n", type);
+ return -1;
+ }
+
+ return __irq_domain_alloc_irqs(domain, irq, 1,
+ ioapic_alloc_attr_node(info),
+ info, legacy, NULL);
+}
+
+/*
+ * Need special handling for ISA IRQs because there may be multiple IOAPIC pins
+ * sharing the same ISA IRQ number and irqdomain only supports 1:1 mapping
+ * between IOAPIC pin and IRQ number. A typical IOAPIC has 24 pins, pin 0-15 are
+ * used for legacy IRQs and pin 16-23 are used for PCI IRQs (PIRQ A-H).
+ * When ACPI is disabled, only legacy IRQ numbers (IRQ0-15) are available, and
+ * some BIOSes may use MP Interrupt Source records to override IRQ numbers for
+ * PIRQs instead of reprogramming the interrupt routing logic. Thus there may be
+ * multiple pins sharing the same legacy IRQ number when ACPI is disabled.
+ */
+static int alloc_isa_irq_from_domain(struct irq_domain *domain,
+ int irq, int ioapic, int pin,
+ struct irq_alloc_info *info)
+{
+ struct mp_chip_data *data;
+ struct irq_data *irq_data = irq_get_irq_data(irq);
+ int node = ioapic_alloc_attr_node(info);
+
+ /*
+ * Legacy ISA IRQ has already been allocated, just add pin to
+ * the pin list associated with this IRQ and program the IOAPIC
+ * entry. The IOAPIC entry
+ */
+ if (irq_data && irq_data->parent_data) {
+ if (!mp_check_pin_attr(irq, info))
+ return -EBUSY;
+ if (__add_pin_to_irq_node(irq_data->chip_data, node, ioapic,
+ info->ioapic.pin))
+ return -ENOMEM;
+ } else {
+ info->flags |= X86_IRQ_ALLOC_LEGACY;
+ irq = __irq_domain_alloc_irqs(domain, irq, 1, node, info, true,
+ NULL);
+ if (irq >= 0) {
+ irq_data = irq_domain_get_irq_data(domain, irq);
+ data = irq_data->chip_data;
+ data->isa_irq = true;
+ }
+ }
+
+ return irq;
+}
+
+static int mp_map_pin_to_irq(u32 gsi, int idx, int ioapic, int pin,
+ unsigned int flags, struct irq_alloc_info *info)
+{
+ int irq;
+ bool legacy = false;
+ struct irq_alloc_info tmp;
+ struct mp_chip_data *data;
+ struct irq_domain *domain = mp_ioapic_irqdomain(ioapic);
+
+ if (!domain)
+ return -ENOSYS;
+
+ if (idx >= 0 && test_bit(mp_irqs[idx].srcbus, mp_bus_not_pci)) {
+ irq = mp_irqs[idx].srcbusirq;
+ legacy = mp_is_legacy_irq(irq);
+ /*
+ * IRQ2 is unusable for historical reasons on systems which
+ * have a legacy PIC. See the comment vs. IRQ2 further down.
+ *
+ * If this gets removed at some point then the related code
+ * in lapic_assign_system_vectors() needs to be adjusted as
+ * well.
+ */
+ if (legacy && irq == PIC_CASCADE_IR)
+ return -EINVAL;
+ }
+
+ mutex_lock(&ioapic_mutex);
+ if (!(flags & IOAPIC_MAP_ALLOC)) {
+ if (!legacy) {
+ irq = irq_find_mapping(domain, pin);
+ if (irq == 0)
+ irq = -ENOENT;
+ }
+ } else {
+ ioapic_copy_alloc_attr(&tmp, info, gsi, ioapic, pin);
+ if (legacy)
+ irq = alloc_isa_irq_from_domain(domain, irq,
+ ioapic, pin, &tmp);
+ else if ((irq = irq_find_mapping(domain, pin)) == 0)
+ irq = alloc_irq_from_domain(domain, ioapic, gsi, &tmp);
+ else if (!mp_check_pin_attr(irq, &tmp))
+ irq = -EBUSY;
+ if (irq >= 0) {
+ data = irq_get_chip_data(irq);
+ data->count++;
+ }
+ }
+ mutex_unlock(&ioapic_mutex);
+
+ return irq;
+}
+
+static int pin_2_irq(int idx, int ioapic, int pin, unsigned int flags)
+{
+ u32 gsi = mp_pin_to_gsi(ioapic, pin);
+
+ /*
+ * Debugging check, we are in big trouble if this message pops up!
+ */
+ if (mp_irqs[idx].dstirq != pin)
+ pr_err("broken BIOS or MPTABLE parser, ayiee!!\n");
+
+#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 {
+ int irq = pirq_entries[pin-16];
+ apic_printk(APIC_VERBOSE, KERN_DEBUG
+ "using PIRQ%d -> IRQ %d\n",
+ pin-16, irq);
+ return irq;
+ }
+ }
+ }
+#endif
+
+ return mp_map_pin_to_irq(gsi, idx, ioapic, pin, flags, NULL);
+}
+
+int mp_map_gsi_to_irq(u32 gsi, unsigned int flags, struct irq_alloc_info *info)
+{
+ int ioapic, pin, idx;
+
+ ioapic = mp_find_ioapic(gsi);
+ if (ioapic < 0)
+ return -ENODEV;
+
+ pin = mp_find_ioapic_pin(ioapic, gsi);
+ idx = find_irq_entry(ioapic, pin, mp_INT);
+ if ((flags & IOAPIC_MAP_CHECK) && idx < 0)
+ return -ENODEV;
+
+ return mp_map_pin_to_irq(gsi, idx, ioapic, pin, flags, info);
+}
+
+void mp_unmap_irq(int irq)
+{
+ struct irq_data *irq_data = irq_get_irq_data(irq);
+ struct mp_chip_data *data;
+
+ if (!irq_data || !irq_data->domain)
+ return;
+
+ data = irq_data->chip_data;
+ if (!data || data->isa_irq)
+ return;
+
+ mutex_lock(&ioapic_mutex);
+ if (--data->count == 0)
+ irq_domain_free_irqs(irq, 1);
+ mutex_unlock(&ioapic_mutex);
+}
+
+/*
+ * Find a specific PCI IRQ entry.
+ * Not an __init, possibly needed by modules
+ */
+int IO_APIC_get_PCI_irq_vector(int bus, int slot, int pin)
+{
+ int irq, i, best_ioapic = -1, best_idx = -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;
+ int ioapic_idx, found = 0;
+
+ if (bus != lbus || mp_irqs[i].irqtype != mp_INT ||
+ slot != ((mp_irqs[i].srcbusirq >> 2) & 0x1f))
+ continue;
+
+ for_each_ioapic(ioapic_idx)
+ if (mpc_ioapic_id(ioapic_idx) == mp_irqs[i].dstapic ||
+ mp_irqs[i].dstapic == MP_APIC_ALL) {
+ found = 1;
+ break;
+ }
+ if (!found)
+ continue;
+
+ /* Skip ISA IRQs */
+ irq = pin_2_irq(i, ioapic_idx, mp_irqs[i].dstirq, 0);
+ if (irq > 0 && !IO_APIC_IRQ(irq))
+ continue;
+
+ if (pin == (mp_irqs[i].srcbusirq & 3)) {
+ best_idx = i;
+ best_ioapic = ioapic_idx;
+ goto out;
+ }
+
+ /*
+ * Use the first all-but-pin matching entry as a
+ * best-guess fuzzy result for broken mptables.
+ */
+ if (best_idx < 0) {
+ best_idx = i;
+ best_ioapic = ioapic_idx;
+ }
+ }
+ if (best_idx < 0)
+ return -1;
+
+out:
+ return pin_2_irq(best_idx, best_ioapic, mp_irqs[best_idx].dstirq,
+ IOAPIC_MAP_ALLOC);
+}
+EXPORT_SYMBOL(IO_APIC_get_PCI_irq_vector);
+
+static struct irq_chip ioapic_chip, ioapic_ir_chip;
+
+static void __init setup_IO_APIC_irqs(void)
+{
+ unsigned int ioapic, pin;
+ int idx;
+
+ apic_printk(APIC_VERBOSE, KERN_DEBUG "init IO_APIC IRQs\n");
+
+ for_each_ioapic_pin(ioapic, pin) {
+ idx = find_irq_entry(ioapic, pin, mp_INT);
+ if (idx < 0)
+ apic_printk(APIC_VERBOSE,
+ KERN_DEBUG " apic %d pin %d not connected\n",
+ mpc_ioapic_id(ioapic), pin);
+ else
+ pin_2_irq(idx, ioapic, pin,
+ ioapic ? 0 : IOAPIC_MAP_ALLOC);
+ }
+}
+
+void ioapic_zap_locks(void)
+{
+ raw_spin_lock_init(&ioapic_lock);
+}
+
+static void io_apic_print_entries(unsigned int apic, unsigned int nr_entries)
+{
+ struct IO_APIC_route_entry entry;
+ char buf[256];
+ int i;
+
+ printk(KERN_DEBUG "IOAPIC %d:\n", apic);
+ for (i = 0; i <= nr_entries; i++) {
+ entry = ioapic_read_entry(apic, i);
+ snprintf(buf, sizeof(buf),
+ " pin%02x, %s, %s, %s, V(%02X), IRR(%1d), S(%1d)",
+ i,
+ entry.masked ? "disabled" : "enabled ",
+ entry.is_level ? "level" : "edge ",
+ entry.active_low ? "low " : "high",
+ entry.vector, entry.irr, entry.delivery_status);
+ if (entry.ir_format) {
+ printk(KERN_DEBUG "%s, remapped, I(%04X), Z(%X)\n",
+ buf,
+ (entry.ir_index_15 << 15) | entry.ir_index_0_14,
+ entry.ir_zero);
+ } else {
+ printk(KERN_DEBUG "%s, %s, D(%02X%02X), M(%1d)\n", buf,
+ entry.dest_mode_logical ? "logical " : "physical",
+ entry.virt_destid_8_14, entry.destid_0_7,
+ entry.delivery_mode);
+ }
+ }
+}
+
+static void __init print_IO_APIC(int ioapic_idx)
+{
+ 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;
+
+ raw_spin_lock_irqsave(&ioapic_lock, flags);
+ reg_00.raw = io_apic_read(ioapic_idx, 0);
+ reg_01.raw = io_apic_read(ioapic_idx, 1);
+ if (reg_01.bits.version >= 0x10)
+ reg_02.raw = io_apic_read(ioapic_idx, 2);
+ if (reg_01.bits.version >= 0x20)
+ reg_03.raw = io_apic_read(ioapic_idx, 3);
+ raw_spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ printk(KERN_DEBUG "IO APIC #%d......\n", mpc_ioapic_id(ioapic_idx));
+ 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 *)&reg_01);
+ printk(KERN_DEBUG "....... : max redirection entries: %02X\n",
+ reg_01.bits.entries);
+
+ printk(KERN_DEBUG "....... : PRQ implemented: %X\n", reg_01.bits.PRQ);
+ printk(KERN_DEBUG "....... : IO APIC version: %02X\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");
+ io_apic_print_entries(ioapic_idx, reg_01.bits.entries);
+}
+
+void __init print_IO_APICs(void)
+{
+ int ioapic_idx;
+ unsigned int irq;
+
+ printk(KERN_DEBUG "number of MP IRQ sources: %d.\n", mp_irq_entries);
+ for_each_ioapic(ioapic_idx)
+ printk(KERN_DEBUG "number of IO-APIC #%d registers: %d.\n",
+ mpc_ioapic_id(ioapic_idx),
+ ioapics[ioapic_idx].nr_registers);
+
+ /*
+ * 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_each_ioapic(ioapic_idx)
+ print_IO_APIC(ioapic_idx);
+
+ printk(KERN_DEBUG "IRQ to pin mappings:\n");
+ for_each_active_irq(irq) {
+ struct irq_pin_list *entry;
+ struct irq_chip *chip;
+ struct mp_chip_data *data;
+
+ chip = irq_get_chip(irq);
+ if (chip != &ioapic_chip && chip != &ioapic_ir_chip)
+ continue;
+ data = irq_get_chip_data(irq);
+ if (!data)
+ continue;
+ if (list_empty(&data->irq_2_pin))
+ continue;
+
+ printk(KERN_DEBUG "IRQ%d ", irq);
+ for_each_irq_pin(entry, data->irq_2_pin)
+ pr_cont("-> %d:%d", entry->apic, entry->pin);
+ pr_cont("\n");
+ }
+
+ printk(KERN_INFO ".................................... done.\n");
+}
+
+/* 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)
+{
+ int i8259_apic, i8259_pin;
+ int apic, pin;
+
+ if (ioapic_is_disabled)
+ nr_ioapics = 0;
+
+ if (!nr_legacy_irqs() || !nr_ioapics)
+ return;
+
+ for_each_ioapic_pin(apic, pin) {
+ /* See if any of the pins is in ExtINT mode */
+ struct IO_APIC_route_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.masked &&
+ entry.delivery_mode == APIC_DELIVERY_MODE_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();
+}
+
+void native_restore_boot_irq_mode(void)
+{
+ /*
+ * 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;
+ u32 apic_id = read_apic_id();
+
+ memset(&entry, 0, sizeof(entry));
+ entry.masked = false;
+ entry.is_level = false;
+ entry.active_low = false;
+ entry.dest_mode_logical = false;
+ entry.delivery_mode = APIC_DELIVERY_MODE_EXTINT;
+ entry.destid_0_7 = apic_id & 0xFF;
+ entry.virt_destid_8_14 = apic_id >> 8;
+
+ /*
+ * Add it to the IO-APIC irq-routing table:
+ */
+ ioapic_write_entry(ioapic_i8259.apic, ioapic_i8259.pin, entry);
+ }
+
+ if (boot_cpu_has(X86_FEATURE_APIC) || apic_from_smp_config())
+ disconnect_bsp_APIC(ioapic_i8259.pin != -1);
+}
+
+void restore_boot_irq_mode(void)
+{
+ if (!nr_legacy_irqs())
+ return;
+
+ x86_apic_ops.restore();
+}
+
+#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
+ */
+void __init setup_ioapic_ids_from_mpc_nocheck(void)
+{
+ union IO_APIC_reg_00 reg_00;
+ physid_mask_t phys_id_present_map;
+ int ioapic_idx;
+ int i;
+ unsigned char old_id;
+ unsigned long flags;
+
+ /*
+ * This is broken; anything with a real cpu count has to
+ * circumvent this idiocy regardless.
+ */
+ apic->ioapic_phys_id_map(&phys_cpu_present_map, &phys_id_present_map);
+
+ /*
+ * Set the IOAPIC ID to the value stored in the MPC table.
+ */
+ for_each_ioapic(ioapic_idx) {
+ /* Read the register 0 value */
+ raw_spin_lock_irqsave(&ioapic_lock, flags);
+ reg_00.raw = io_apic_read(ioapic_idx, 0);
+ raw_spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ old_id = mpc_ioapic_id(ioapic_idx);
+
+ if (mpc_ioapic_id(ioapic_idx) >= get_physical_broadcast()) {
+ printk(KERN_ERR "BIOS bug, IO-APIC#%d ID is %d in the MPC table!...\n",
+ ioapic_idx, mpc_ioapic_id(ioapic_idx));
+ printk(KERN_ERR "... fixing up to %d. (tell your hw vendor)\n",
+ reg_00.bits.ID);
+ ioapics[ioapic_idx].mp_config.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,
+ mpc_ioapic_id(ioapic_idx))) {
+ printk(KERN_ERR "BIOS bug, IO-APIC#%d ID %d is already used!...\n",
+ ioapic_idx, mpc_ioapic_id(ioapic_idx));
+ 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);
+ ioapics[ioapic_idx].mp_config.apicid = i;
+ } else {
+ apic_printk(APIC_VERBOSE, "Setting %d in the phys_id_present_map\n",
+ mpc_ioapic_id(ioapic_idx));
+ physid_set(mpc_ioapic_id(ioapic_idx), phys_id_present_map);
+ }
+
+ /*
+ * We need to adjust the IRQ routing table
+ * if the ID changed.
+ */
+ if (old_id != mpc_ioapic_id(ioapic_idx))
+ for (i = 0; i < mp_irq_entries; i++)
+ if (mp_irqs[i].dstapic == old_id)
+ mp_irqs[i].dstapic
+ = mpc_ioapic_id(ioapic_idx);
+
+ /*
+ * Update the ID register according to the right value
+ * from the MPC table if they are different.
+ */
+ if (mpc_ioapic_id(ioapic_idx) == reg_00.bits.ID)
+ continue;
+
+ apic_printk(APIC_VERBOSE, KERN_INFO
+ "...changing IO-APIC physical APIC ID to %d ...",
+ mpc_ioapic_id(ioapic_idx));
+
+ reg_00.bits.ID = mpc_ioapic_id(ioapic_idx);
+ raw_spin_lock_irqsave(&ioapic_lock, flags);
+ io_apic_write(ioapic_idx, 0, reg_00.raw);
+ raw_spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ /*
+ * Sanity check
+ */
+ raw_spin_lock_irqsave(&ioapic_lock, flags);
+ reg_00.raw = io_apic_read(ioapic_idx, 0);
+ raw_spin_unlock_irqrestore(&ioapic_lock, flags);
+ if (reg_00.bits.ID != mpc_ioapic_id(ioapic_idx))
+ pr_cont("could not set ID!\n");
+ else
+ apic_printk(APIC_VERBOSE, " ok.\n");
+ }
+}
+
+void __init setup_ioapic_ids_from_mpc(void)
+{
+
+ if (acpi_ioapic)
+ 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(boot_cpu_apic_version))
+ return;
+ setup_ioapic_ids_from_mpc_nocheck();
+}
+#endif
+
+int no_timer_check __initdata;
+
+static int __init notimercheck(char *s)
+{
+ no_timer_check = 1;
+ return 1;
+}
+__setup("no_timer_check", notimercheck);
+
+static void __init delay_with_tsc(void)
+{
+ unsigned long long start, now;
+ unsigned long end = jiffies + 4;
+
+ start = rdtsc();
+
+ /*
+ * We don't know the TSC frequency yet, but waiting for
+ * 40000000000/HZ TSC cycles is safe:
+ * 4 GHz == 10 jiffies
+ * 1 GHz == 40 jiffies
+ */
+ do {
+ rep_nop();
+ now = rdtsc();
+ } while ((now - start) < 40000000000ULL / HZ &&
+ time_before_eq(jiffies, end));
+}
+
+static void __init delay_without_tsc(void)
+{
+ unsigned long end = jiffies + 4;
+ int band = 1;
+
+ /*
+ * We don't know any frequency yet, but waiting for
+ * 40940000000/HZ cycles is safe:
+ * 4 GHz == 10 jiffies
+ * 1 GHz == 40 jiffies
+ * 1 << 1 + 1 << 2 +...+ 1 << 11 = 4094
+ */
+ do {
+ __delay(((1U << band++) * 10000000UL) / HZ);
+ } while (band < 12 && time_before_eq(jiffies, end));
+}
+
+/*
+ * 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;
+
+ if (no_timer_check)
+ return 1;
+
+ local_irq_enable();
+ if (boot_cpu_has(X86_FEATURE_TSC))
+ delay_with_tsc();
+ else
+ delay_without_tsc();
+
+ /*
+ * 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.
+ */
+
+ local_irq_disable();
+
+ /* Did jiffies advance? */
+ return time_after(jiffies, t1 + 4);
+}
+
+/*
+ * 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(struct irq_data *data)
+{
+ int was_pending = 0, irq = data->irq;
+ unsigned long flags;
+
+ raw_spin_lock_irqsave(&ioapic_lock, flags);
+ if (irq < nr_legacy_irqs()) {
+ legacy_pic->mask(irq);
+ if (legacy_pic->irq_pending(irq))
+ was_pending = 1;
+ }
+ __unmask_ioapic(data->chip_data);
+ raw_spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ return was_pending;
+}
+
+atomic_t irq_mis_count;
+
+#ifdef CONFIG_GENERIC_PENDING_IRQ
+static bool io_apic_level_ack_pending(struct mp_chip_data *data)
+{
+ struct irq_pin_list *entry;
+ unsigned long flags;
+
+ raw_spin_lock_irqsave(&ioapic_lock, flags);
+ for_each_irq_pin(entry, data->irq_2_pin) {
+ struct IO_APIC_route_entry e;
+ int pin;
+
+ pin = entry->pin;
+ e.w1 = io_apic_read(entry->apic, 0x10 + pin*2);
+ /* Is the remote IRR bit set? */
+ if (e.irr) {
+ raw_spin_unlock_irqrestore(&ioapic_lock, flags);
+ return true;
+ }
+ }
+ raw_spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ return false;
+}
+
+static inline bool ioapic_prepare_move(struct irq_data *data)
+{
+ /* If we are moving the IRQ we need to mask it */
+ if (unlikely(irqd_is_setaffinity_pending(data))) {
+ if (!irqd_irq_masked(data))
+ mask_ioapic_irq(data);
+ return true;
+ }
+ return false;
+}
+
+static inline void ioapic_finish_move(struct irq_data *data, bool moveit)
+{
+ if (unlikely(moveit)) {
+ /* 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
+ * completely 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.
+ */
+ if (!io_apic_level_ack_pending(data->chip_data))
+ irq_move_masked_irq(data);
+ /* If the IRQ is masked in the core, leave it: */
+ if (!irqd_irq_masked(data))
+ unmask_ioapic_irq(data);
+ }
+}
+#else
+static inline bool ioapic_prepare_move(struct irq_data *data)
+{
+ return false;
+}
+static inline void ioapic_finish_move(struct irq_data *data, bool moveit)
+{
+}
+#endif
+
+static void ioapic_ack_level(struct irq_data *irq_data)
+{
+ struct irq_cfg *cfg = irqd_cfg(irq_data);
+ unsigned long v;
+ bool moveit;
+ int i;
+
+ irq_complete_move(cfg);
+ moveit = ioapic_prepare_move(irq_data);
+
+ /*
+ * 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
+ *
+ * Also in the case when cpu goes offline, fixup_irqs() will forward
+ * any unhandled interrupt on the offlined cpu to the new cpu
+ * destination that is handling the corresponding interrupt. This
+ * interrupt forwarding is done via IPI's. Hence, in this case also
+ * level-triggered io-apic interrupt will be seen as an edge
+ * interrupt in the IRR. And we can't rely on the cpu's EOI
+ * to be broadcasted to the IO-APIC's which will clear the remoteIRR
+ * corresponding to the level-triggered interrupt. Hence on IO-APIC's
+ * supporting EOI register, we do an explicit EOI to clear the
+ * remote IRR and on IO-APIC's which don't have an EOI register,
+ * we use the above logic (mask+edge followed by unmask+level) from
+ * Manfred Spraul to clear the remote IRR.
+ */
+ i = cfg->vector;
+ v = apic_read(APIC_TMR + ((i & ~0x1f) >> 1));
+
+ /*
+ * We must acknowledge the irq before we move it or the acknowledge will
+ * not propagate properly.
+ */
+ apic_eoi();
+
+ /*
+ * Tail end of clearing remote IRR bit (either by delivering the EOI
+ * message via io-apic EOI register write or simulating it using
+ * mask+edge followed by unmask+level logic) manually when the
+ * level triggered interrupt is seen as the edge triggered interrupt
+ * at the cpu.
+ */
+ if (!(v & (1 << (i & 0x1f)))) {
+ atomic_inc(&irq_mis_count);
+ eoi_ioapic_pin(cfg->vector, irq_data->chip_data);
+ }
+
+ ioapic_finish_move(irq_data, moveit);
+}
+
+static void ioapic_ir_ack_level(struct irq_data *irq_data)
+{
+ struct mp_chip_data *data = irq_data->chip_data;
+
+ /*
+ * Intr-remapping uses pin number as the virtual vector
+ * in the RTE. Actual vector is programmed in
+ * intr-remapping table entry. Hence for the io-apic
+ * EOI we use the pin number.
+ */
+ apic_ack_irq(irq_data);
+ eoi_ioapic_pin(data->entry.vector, data);
+}
+
+/*
+ * The I/OAPIC is just a device for generating MSI messages from legacy
+ * interrupt pins. Various fields of the RTE translate into bits of the
+ * resulting MSI which had a historical meaning.
+ *
+ * With interrupt remapping, many of those bits have different meanings
+ * in the underlying MSI, but the way that the I/OAPIC transforms them
+ * from its RTE to the MSI message is the same. This function allows
+ * the parent IRQ domain to compose the MSI message, then takes the
+ * relevant bits to put them in the appropriate places in the RTE in
+ * order to generate that message when the IRQ happens.
+ *
+ * The setup here relies on a preconfigured route entry (is_level,
+ * active_low, masked) because the parent domain is merely composing the
+ * generic message routing information which is used for the MSI.
+ */
+static void ioapic_setup_msg_from_msi(struct irq_data *irq_data,
+ struct IO_APIC_route_entry *entry)
+{
+ struct msi_msg msg;
+
+ /* Let the parent domain compose the MSI message */
+ irq_chip_compose_msi_msg(irq_data, &msg);
+
+ /*
+ * - Real vector
+ * - DMAR/IR: 8bit subhandle (ioapic.pin)
+ * - AMD/IR: 8bit IRTE index
+ */
+ entry->vector = msg.arch_data.vector;
+ /* Delivery mode (for DMAR/IR all 0) */
+ entry->delivery_mode = msg.arch_data.delivery_mode;
+ /* Destination mode or DMAR/IR index bit 15 */
+ entry->dest_mode_logical = msg.arch_addr_lo.dest_mode_logical;
+ /* DMAR/IR: 1, 0 for all other modes */
+ entry->ir_format = msg.arch_addr_lo.dmar_format;
+ /*
+ * - DMAR/IR: index bit 0-14.
+ *
+ * - Virt: If the host supports x2apic without a virtualized IR
+ * unit then bit 0-6 of dmar_index_0_14 are providing bit
+ * 8-14 of the destination id.
+ *
+ * All other modes have bit 0-6 of dmar_index_0_14 cleared and the
+ * topmost 8 bits are destination id bit 0-7 (entry::destid_0_7).
+ */
+ entry->ir_index_0_14 = msg.arch_addr_lo.dmar_index_0_14;
+}
+
+static void ioapic_configure_entry(struct irq_data *irqd)
+{
+ struct mp_chip_data *mpd = irqd->chip_data;
+ struct irq_pin_list *entry;
+
+ ioapic_setup_msg_from_msi(irqd, &mpd->entry);
+
+ for_each_irq_pin(entry, mpd->irq_2_pin)
+ __ioapic_write_entry(entry->apic, entry->pin, mpd->entry);
+}
+
+static int ioapic_set_affinity(struct irq_data *irq_data,
+ const struct cpumask *mask, bool force)
+{
+ struct irq_data *parent = irq_data->parent_data;
+ unsigned long flags;
+ int ret;
+
+ ret = parent->chip->irq_set_affinity(parent, mask, force);
+ raw_spin_lock_irqsave(&ioapic_lock, flags);
+ if (ret >= 0 && ret != IRQ_SET_MASK_OK_DONE)
+ ioapic_configure_entry(irq_data);
+ raw_spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ return ret;
+}
+
+/*
+ * Interrupt shutdown masks the ioapic pin, but the interrupt might already
+ * be in flight, but not yet serviced by the target CPU. That means
+ * __synchronize_hardirq() would return and claim that everything is calmed
+ * down. So free_irq() would proceed and deactivate the interrupt and free
+ * resources.
+ *
+ * Once the target CPU comes around to service it it will find a cleared
+ * vector and complain. While the spurious interrupt is harmless, the full
+ * release of resources might prevent the interrupt from being acknowledged
+ * which keeps the hardware in a weird state.
+ *
+ * Verify that the corresponding Remote-IRR bits are clear.
+ */
+static int ioapic_irq_get_chip_state(struct irq_data *irqd,
+ enum irqchip_irq_state which,
+ bool *state)
+{
+ struct mp_chip_data *mcd = irqd->chip_data;
+ struct IO_APIC_route_entry rentry;
+ struct irq_pin_list *p;
+
+ if (which != IRQCHIP_STATE_ACTIVE)
+ return -EINVAL;
+
+ *state = false;
+ raw_spin_lock(&ioapic_lock);
+ for_each_irq_pin(p, mcd->irq_2_pin) {
+ rentry = __ioapic_read_entry(p->apic, p->pin);
+ /*
+ * The remote IRR is only valid in level trigger mode. It's
+ * meaning is undefined for edge triggered interrupts and
+ * irrelevant because the IO-APIC treats them as fire and
+ * forget.
+ */
+ if (rentry.irr && rentry.is_level) {
+ *state = true;
+ break;
+ }
+ }
+ raw_spin_unlock(&ioapic_lock);
+ return 0;
+}
+
+static struct irq_chip ioapic_chip __read_mostly = {
+ .name = "IO-APIC",
+ .irq_startup = startup_ioapic_irq,
+ .irq_mask = mask_ioapic_irq,
+ .irq_unmask = unmask_ioapic_irq,
+ .irq_ack = irq_chip_ack_parent,
+ .irq_eoi = ioapic_ack_level,
+ .irq_set_affinity = ioapic_set_affinity,
+ .irq_retrigger = irq_chip_retrigger_hierarchy,
+ .irq_get_irqchip_state = ioapic_irq_get_chip_state,
+ .flags = IRQCHIP_SKIP_SET_WAKE |
+ IRQCHIP_AFFINITY_PRE_STARTUP,
+};
+
+static struct irq_chip ioapic_ir_chip __read_mostly = {
+ .name = "IR-IO-APIC",
+ .irq_startup = startup_ioapic_irq,
+ .irq_mask = mask_ioapic_irq,
+ .irq_unmask = unmask_ioapic_irq,
+ .irq_ack = irq_chip_ack_parent,
+ .irq_eoi = ioapic_ir_ack_level,
+ .irq_set_affinity = ioapic_set_affinity,
+ .irq_retrigger = irq_chip_retrigger_hierarchy,
+ .irq_get_irqchip_state = ioapic_irq_get_chip_state,
+ .flags = IRQCHIP_SKIP_SET_WAKE |
+ IRQCHIP_AFFINITY_PRE_STARTUP,
+};
+
+static inline void init_IO_APIC_traps(void)
+{
+ struct irq_cfg *cfg;
+ unsigned int irq;
+
+ for_each_active_irq(irq) {
+ cfg = irq_cfg(irq);
+ 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_legacy_irqs())
+ legacy_pic->make_irq(irq);
+ else
+ /* Strange. Oh, well.. */
+ irq_set_chip(irq, &no_irq_chip);
+ }
+ }
+}
+
+/*
+ * The local APIC irq-chip implementation:
+ */
+
+static void mask_lapic_irq(struct irq_data *data)
+{
+ unsigned long v;
+
+ v = apic_read(APIC_LVT0);
+ apic_write(APIC_LVT0, v | APIC_LVT_MASKED);
+}
+
+static void unmask_lapic_irq(struct irq_data *data)
+{
+ unsigned long v;
+
+ v = apic_read(APIC_LVT0);
+ apic_write(APIC_LVT0, v & ~APIC_LVT_MASKED);
+}
+
+static void ack_lapic_irq(struct irq_data *data)
+{
+ apic_eoi();
+}
+
+static struct irq_chip lapic_chip __read_mostly = {
+ .name = "local-APIC",
+ .irq_mask = mask_lapic_irq,
+ .irq_unmask = unmask_lapic_irq,
+ .irq_ack = ack_lapic_irq,
+};
+
+static void lapic_register_intr(int irq)
+{
+ irq_clear_status_flags(irq, IRQ_LEVEL);
+ irq_set_chip_and_handler_name(irq, &lapic_chip, handle_edge_irq,
+ "edge");
+}
+
+/*
+ * 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;
+ u32 apic_id;
+
+ 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);
+
+ apic_id = read_apic_id();
+ memset(&entry1, 0, sizeof(entry1));
+
+ entry1.dest_mode_logical = true;
+ entry1.masked = false;
+ entry1.destid_0_7 = apic_id & 0xFF;
+ entry1.virt_destid_8_14 = apic_id >> 8;
+ entry1.delivery_mode = APIC_DELIVERY_MODE_EXTINT;
+ entry1.active_low = entry0.active_low;
+ entry1.is_level = false;
+ 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);
+
+static int mp_alloc_timer_irq(int ioapic, int pin)
+{
+ int irq = -1;
+ struct irq_domain *domain = mp_ioapic_irqdomain(ioapic);
+
+ if (domain) {
+ struct irq_alloc_info info;
+
+ ioapic_set_alloc_attr(&info, NUMA_NO_NODE, 0, 0);
+ info.devid = mpc_ioapic_id(ioapic);
+ info.ioapic.pin = pin;
+ mutex_lock(&ioapic_mutex);
+ irq = alloc_isa_irq_from_domain(domain, 0, ioapic, pin, &info);
+ mutex_unlock(&ioapic_mutex);
+ }
+
+ return irq;
+}
+
+/*
+ * 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_data *irq_data = irq_get_irq_data(0);
+ struct mp_chip_data *data = irq_data->chip_data;
+ struct irq_cfg *cfg = irqd_cfg(irq_data);
+ int node = cpu_to_node(0);
+ int apic1, pin1, apic2, pin2;
+ int no_pin1 = 0;
+
+ if (!global_clock_event)
+ return;
+
+ local_irq_disable();
+
+ /*
+ * get/set the timer IRQ vector:
+ */
+ legacy_pic->mask(0);
+
+ /*
+ * 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);
+ legacy_pic->init(1);
+
+ 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) {
+ panic_if_irq_remap("BIOS bug: timer not connected to IO-APIC");
+ 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) {
+ mp_alloc_timer_irq(apic1, pin1);
+ } else {
+ /*
+ * for edge trigger, it's already unmasked,
+ * so only need to unmask if it is level-trigger
+ * do we really have level trigger timer?
+ */
+ int idx = find_irq_entry(apic1, pin1, mp_INT);
+
+ if (idx != -1 && irq_is_level(idx))
+ unmask_ioapic_irq(irq_get_irq_data(0));
+ }
+ irq_domain_deactivate_irq(irq_data);
+ irq_domain_activate_irq(irq_data, false);
+ if (timer_irq_works()) {
+ if (disable_timer_pin_1 > 0)
+ clear_IO_APIC_pin(0, pin1);
+ goto out;
+ }
+ panic_if_irq_remap("timer doesn't work through Interrupt-remapped IO-APIC");
+ 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_node(data, node, apic1, pin1, apic2, pin2);
+ irq_domain_deactivate_irq(irq_data);
+ irq_domain_activate_irq(irq_data, false);
+ legacy_pic->unmask(0);
+ if (timer_irq_works()) {
+ apic_printk(APIC_QUIET, KERN_INFO "....... works.\n");
+ goto out;
+ }
+ /*
+ * Cleanup, just in case ...
+ */
+ legacy_pic->mask(0);
+ clear_IO_APIC_pin(apic2, pin2);
+ apic_printk(APIC_QUIET, KERN_INFO "....... failed.\n");
+ }
+
+ apic_printk(APIC_QUIET, KERN_INFO
+ "...trying to set up timer as Virtual Wire IRQ...\n");
+
+ lapic_register_intr(0);
+ apic_write(APIC_LVT0, APIC_DM_FIXED | cfg->vector); /* Fixed mode */
+ legacy_pic->unmask(0);
+
+ if (timer_irq_works()) {
+ apic_printk(APIC_QUIET, KERN_INFO "..... works.\n");
+ goto out;
+ }
+ legacy_pic->mask(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");
+
+ legacy_pic->init(0);
+ legacy_pic->make_irq(0);
+ apic_write(APIC_LVT0, APIC_DM_EXTINT);
+ legacy_pic->unmask(0);
+
+ unlock_ExtINT_logic();
+
+ if (timer_irq_works()) {
+ apic_printk(APIC_QUIET, KERN_INFO "..... works.\n");
+ goto out;
+ }
+ apic_printk(APIC_QUIET, KERN_INFO "..... failed :(.\n");
+ if (apic_is_x2apic_enabled())
+ apic_printk(APIC_QUIET, KERN_INFO
+ "Perhaps problem with the pre-enabled x2apic mode\n"
+ "Try booting with x2apic and interrupt-remapping disabled in the bios.\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_enable();
+}
+
+/*
+ * 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 (1UL << PIC_CASCADE_IR)
+
+static int mp_irqdomain_create(int ioapic)
+{
+ struct irq_domain *parent;
+ int hwirqs = mp_ioapic_pin_count(ioapic);
+ struct ioapic *ip = &ioapics[ioapic];
+ struct ioapic_domain_cfg *cfg = &ip->irqdomain_cfg;
+ struct mp_ioapic_gsi *gsi_cfg = mp_ioapic_gsi_routing(ioapic);
+ struct fwnode_handle *fn;
+ struct irq_fwspec fwspec;
+
+ if (cfg->type == IOAPIC_DOMAIN_INVALID)
+ return 0;
+
+ /* Handle device tree enumerated APICs proper */
+ if (cfg->dev) {
+ fn = of_node_to_fwnode(cfg->dev);
+ } else {
+ fn = irq_domain_alloc_named_id_fwnode("IO-APIC", mpc_ioapic_id(ioapic));
+ if (!fn)
+ return -ENOMEM;
+ }
+
+ fwspec.fwnode = fn;
+ fwspec.param_count = 1;
+ fwspec.param[0] = mpc_ioapic_id(ioapic);
+
+ parent = irq_find_matching_fwspec(&fwspec, DOMAIN_BUS_ANY);
+ if (!parent) {
+ if (!cfg->dev)
+ irq_domain_free_fwnode(fn);
+ return -ENODEV;
+ }
+
+ ip->irqdomain = irq_domain_create_hierarchy(parent, 0, hwirqs, fn, cfg->ops,
+ (void *)(long)ioapic);
+ if (!ip->irqdomain) {
+ /* Release fw handle if it was allocated above */
+ if (!cfg->dev)
+ irq_domain_free_fwnode(fn);
+ return -ENOMEM;
+ }
+
+ if (cfg->type == IOAPIC_DOMAIN_LEGACY ||
+ cfg->type == IOAPIC_DOMAIN_STRICT)
+ ioapic_dynirq_base = max(ioapic_dynirq_base,
+ gsi_cfg->gsi_end + 1);
+
+ return 0;
+}
+
+static void ioapic_destroy_irqdomain(int idx)
+{
+ struct ioapic_domain_cfg *cfg = &ioapics[idx].irqdomain_cfg;
+ struct fwnode_handle *fn = ioapics[idx].irqdomain->fwnode;
+
+ if (ioapics[idx].irqdomain) {
+ irq_domain_remove(ioapics[idx].irqdomain);
+ if (!cfg->dev)
+ irq_domain_free_fwnode(fn);
+ ioapics[idx].irqdomain = NULL;
+ }
+}
+
+void __init setup_IO_APIC(void)
+{
+ int ioapic;
+
+ if (ioapic_is_disabled || !nr_ioapics)
+ return;
+
+ io_apic_irqs = nr_legacy_irqs() ? ~PIC_IRQS : ~0UL;
+
+ apic_printk(APIC_VERBOSE, "ENABLING IO-APIC IRQs\n");
+ for_each_ioapic(ioapic)
+ BUG_ON(mp_irqdomain_create(ioapic));
+
+ /*
+ * Set up IO-APIC IRQ routing.
+ */
+ x86_init.mpparse.setup_ioapic_ids();
+
+ sync_Arb_IDs();
+ setup_IO_APIC_irqs();
+ init_IO_APIC_traps();
+ if (nr_legacy_irqs())
+ check_timer();
+
+ ioapic_initialized = 1;
+}
+
+static void resume_ioapic_id(int ioapic_idx)
+{
+ unsigned long flags;
+ union IO_APIC_reg_00 reg_00;
+
+ raw_spin_lock_irqsave(&ioapic_lock, flags);
+ reg_00.raw = io_apic_read(ioapic_idx, 0);
+ if (reg_00.bits.ID != mpc_ioapic_id(ioapic_idx)) {
+ reg_00.bits.ID = mpc_ioapic_id(ioapic_idx);
+ io_apic_write(ioapic_idx, 0, reg_00.raw);
+ }
+ raw_spin_unlock_irqrestore(&ioapic_lock, flags);
+}
+
+static void ioapic_resume(void)
+{
+ int ioapic_idx;
+
+ for_each_ioapic_reverse(ioapic_idx)
+ resume_ioapic_id(ioapic_idx);
+
+ restore_ioapic_entries();
+}
+
+static struct syscore_ops ioapic_syscore_ops = {
+ .suspend = save_ioapic_entries,
+ .resume = ioapic_resume,
+};
+
+static int __init ioapic_init_ops(void)
+{
+ register_syscore_ops(&ioapic_syscore_ops);
+
+ return 0;
+}
+
+device_initcall(ioapic_init_ops);
+
+static int io_apic_get_redir_entries(int ioapic)
+{
+ union IO_APIC_reg_01 reg_01;
+ unsigned long flags;
+
+ raw_spin_lock_irqsave(&ioapic_lock, flags);
+ reg_01.raw = io_apic_read(ioapic, 1);
+ raw_spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ /* The register returns the maximum index redir index
+ * supported, which is one less than the total number of redir
+ * entries.
+ */
+ return reg_01.bits.entries + 1;
+}
+
+unsigned int arch_dynirq_lower_bound(unsigned int from)
+{
+ unsigned int ret;
+
+ /*
+ * dmar_alloc_hwirq() may be called before setup_IO_APIC(), so use
+ * gsi_top if ioapic_dynirq_base hasn't been initialized yet.
+ */
+ ret = ioapic_dynirq_base ? : gsi_top;
+
+ /*
+ * For DT enabled machines ioapic_dynirq_base is irrelevant and
+ * always 0. gsi_top can be 0 if there is no IO/APIC registered.
+ * 0 is an invalid interrupt number for dynamic allocations. Return
+ * @from instead.
+ */
+ return ret ? : from;
+}
+
+#ifdef CONFIG_X86_32
+static int 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->ioapic_phys_id_map(&phys_cpu_present_map, &apic_id_map);
+
+ raw_spin_lock_irqsave(&ioapic_lock, flags);
+ reg_00.raw = io_apic_read(ioapic, 0);
+ raw_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;
+ }
+
+ physid_set_mask_of_physid(apic_id, &tmp);
+ physids_or(apic_id_map, apic_id_map, tmp);
+
+ if (reg_00.bits.ID != apic_id) {
+ reg_00.bits.ID = apic_id;
+
+ raw_spin_lock_irqsave(&ioapic_lock, flags);
+ io_apic_write(ioapic, 0, reg_00.raw);
+ reg_00.raw = io_apic_read(ioapic, 0);
+ raw_spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ /* Sanity check */
+ if (reg_00.bits.ID != apic_id) {
+ pr_err("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;
+}
+
+static u8 io_apic_unique_id(int idx, u8 id)
+{
+ if ((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) &&
+ !APIC_XAPIC(boot_cpu_apic_version))
+ return io_apic_get_unique_id(idx, id);
+ else
+ return id;
+}
+#else
+static u8 io_apic_unique_id(int idx, u8 id)
+{
+ union IO_APIC_reg_00 reg_00;
+ DECLARE_BITMAP(used, 256);
+ unsigned long flags;
+ u8 new_id;
+ int i;
+
+ bitmap_zero(used, 256);
+ for_each_ioapic(i)
+ __set_bit(mpc_ioapic_id(i), used);
+
+ /* Hand out the requested id if available */
+ if (!test_bit(id, used))
+ return id;
+
+ /*
+ * Read the current id from the ioapic and keep it if
+ * available.
+ */
+ raw_spin_lock_irqsave(&ioapic_lock, flags);
+ reg_00.raw = io_apic_read(idx, 0);
+ raw_spin_unlock_irqrestore(&ioapic_lock, flags);
+ new_id = reg_00.bits.ID;
+ if (!test_bit(new_id, used)) {
+ apic_printk(APIC_VERBOSE, KERN_INFO
+ "IOAPIC[%d]: Using reg apic_id %d instead of %d\n",
+ idx, new_id, id);
+ return new_id;
+ }
+
+ /*
+ * Get the next free id and write it to the ioapic.
+ */
+ new_id = find_first_zero_bit(used, 256);
+ reg_00.bits.ID = new_id;
+ raw_spin_lock_irqsave(&ioapic_lock, flags);
+ io_apic_write(idx, 0, reg_00.raw);
+ reg_00.raw = io_apic_read(idx, 0);
+ raw_spin_unlock_irqrestore(&ioapic_lock, flags);
+ /* Sanity check */
+ BUG_ON(reg_00.bits.ID != new_id);
+
+ return new_id;
+}
+#endif
+
+static int io_apic_get_version(int ioapic)
+{
+ union IO_APIC_reg_01 reg_01;
+ unsigned long flags;
+
+ raw_spin_lock_irqsave(&ioapic_lock, flags);
+ reg_01.raw = io_apic_read(ioapic, 1);
+ raw_spin_unlock_irqrestore(&ioapic_lock, flags);
+
+ return reg_01.bits.version;
+}
+
+/*
+ * This function updates target affinity of IOAPIC interrupts to include
+ * the CPUs which came online during SMP bringup.
+ */
+#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 = memblock_alloc(n, SMP_CACHE_BYTES);
+ if (!mem)
+ panic("%s: Failed to allocate %lu bytes\n", __func__, n);
+ res = (void *)mem;
+
+ mem += sizeof(struct resource) * nr_ioapics;
+
+ for_each_ioapic(i) {
+ res[i].name = mem;
+ res[i].flags = IORESOURCE_MEM | IORESOURCE_BUSY;
+ snprintf(mem, IOAPIC_RESOURCE_NAME_SIZE, "IOAPIC %u", i);
+ mem += IOAPIC_RESOURCE_NAME_SIZE;
+ ioapics[i].iomem_res = &res[i];
+ }
+
+ ioapic_resources = res;
+
+ return res;
+}
+
+static void io_apic_set_fixmap(enum fixed_addresses idx, phys_addr_t phys)
+{
+ pgprot_t flags = FIXMAP_PAGE_NOCACHE;
+
+ /*
+ * Ensure fixmaps for IO-APIC MMIO respect memory encryption pgprot
+ * bits, just like normal ioremap():
+ */
+ if (cc_platform_has(CC_ATTR_GUEST_MEM_ENCRYPT)) {
+ if (x86_platform.hyper.is_private_mmio(phys))
+ flags = pgprot_encrypted(flags);
+ else
+ flags = pgprot_decrypted(flags);
+ }
+
+ __set_fixmap(idx, phys, flags);
+}
+
+void __init io_apic_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_each_ioapic(i) {
+ if (smp_found_config) {
+ ioapic_phys = mpc_ioapic_addr(i);
+#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;
+ ioapic_is_disabled = true;
+ goto fake_ioapic_page;
+ }
+#endif
+ } else {
+#ifdef CONFIG_X86_32
+fake_ioapic_page:
+#endif
+ ioapic_phys = (unsigned long)memblock_alloc(PAGE_SIZE,
+ PAGE_SIZE);
+ if (!ioapic_phys)
+ panic("%s: Failed to allocate %lu bytes align=0x%lx\n",
+ __func__, PAGE_SIZE, PAGE_SIZE);
+ ioapic_phys = __pa(ioapic_phys);
+ }
+ io_apic_set_fixmap(idx, ioapic_phys);
+ apic_printk(APIC_VERBOSE, "mapped IOAPIC to %08lx (%08lx)\n",
+ __fix_to_virt(idx) + (ioapic_phys & ~PAGE_MASK),
+ ioapic_phys);
+ idx++;
+
+ ioapic_res->start = ioapic_phys;
+ ioapic_res->end = ioapic_phys + IO_APIC_SLOT_SIZE - 1;
+ ioapic_res++;
+ }
+}
+
+void __init ioapic_insert_resources(void)
+{
+ int i;
+ struct resource *r = ioapic_resources;
+
+ if (!r) {
+ if (nr_ioapics > 0)
+ printk(KERN_ERR
+ "IO APIC resources couldn't be allocated.\n");
+ return;
+ }
+
+ for_each_ioapic(i) {
+ insert_resource(&iomem_resource, r);
+ r++;
+ }
+}
+
+int mp_find_ioapic(u32 gsi)
+{
+ int i;
+
+ if (nr_ioapics == 0)
+ return -1;
+
+ /* Find the IOAPIC that manages this GSI. */
+ for_each_ioapic(i) {
+ struct mp_ioapic_gsi *gsi_cfg = mp_ioapic_gsi_routing(i);
+ if (gsi >= gsi_cfg->gsi_base && gsi <= gsi_cfg->gsi_end)
+ return i;
+ }
+
+ printk(KERN_ERR "ERROR: Unable to locate IOAPIC for GSI %d\n", gsi);
+ return -1;
+}
+
+int mp_find_ioapic_pin(int ioapic, u32 gsi)
+{
+ struct mp_ioapic_gsi *gsi_cfg;
+
+ if (WARN_ON(ioapic < 0))
+ return -1;
+
+ gsi_cfg = mp_ioapic_gsi_routing(ioapic);
+ if (WARN_ON(gsi > gsi_cfg->gsi_end))
+ return -1;
+
+ return gsi - gsi_cfg->gsi_base;
+}
+
+static int bad_ioapic_register(int idx)
+{
+ union IO_APIC_reg_00 reg_00;
+ union IO_APIC_reg_01 reg_01;
+ union IO_APIC_reg_02 reg_02;
+
+ reg_00.raw = io_apic_read(idx, 0);
+ reg_01.raw = io_apic_read(idx, 1);
+ reg_02.raw = io_apic_read(idx, 2);
+
+ if (reg_00.raw == -1 && reg_01.raw == -1 && reg_02.raw == -1) {
+ pr_warn("I/O APIC 0x%x registers return all ones, skipping!\n",
+ mpc_ioapic_addr(idx));
+ return 1;
+ }
+
+ return 0;
+}
+
+static int find_free_ioapic_entry(void)
+{
+ int idx;
+
+ for (idx = 0; idx < MAX_IO_APICS; idx++)
+ if (ioapics[idx].nr_registers == 0)
+ return idx;
+
+ return MAX_IO_APICS;
+}
+
+/**
+ * mp_register_ioapic - Register an IOAPIC device
+ * @id: hardware IOAPIC ID
+ * @address: physical address of IOAPIC register area
+ * @gsi_base: base of GSI associated with the IOAPIC
+ * @cfg: configuration information for the IOAPIC
+ */
+int mp_register_ioapic(int id, u32 address, u32 gsi_base,
+ struct ioapic_domain_cfg *cfg)
+{
+ bool hotplug = !!ioapic_initialized;
+ struct mp_ioapic_gsi *gsi_cfg;
+ int idx, ioapic, entries;
+ u32 gsi_end;
+
+ if (!address) {
+ pr_warn("Bogus (zero) I/O APIC address found, skipping!\n");
+ return -EINVAL;
+ }
+ for_each_ioapic(ioapic)
+ if (ioapics[ioapic].mp_config.apicaddr == address) {
+ pr_warn("address 0x%x conflicts with IOAPIC%d\n",
+ address, ioapic);
+ return -EEXIST;
+ }
+
+ idx = find_free_ioapic_entry();
+ if (idx >= MAX_IO_APICS) {
+ pr_warn("Max # of I/O APICs (%d) exceeded (found %d), skipping\n",
+ MAX_IO_APICS, idx);
+ return -ENOSPC;
+ }
+
+ ioapics[idx].mp_config.type = MP_IOAPIC;
+ ioapics[idx].mp_config.flags = MPC_APIC_USABLE;
+ ioapics[idx].mp_config.apicaddr = address;
+
+ io_apic_set_fixmap(FIX_IO_APIC_BASE_0 + idx, address);
+ if (bad_ioapic_register(idx)) {
+ clear_fixmap(FIX_IO_APIC_BASE_0 + idx);
+ return -ENODEV;
+ }
+
+ ioapics[idx].mp_config.apicid = io_apic_unique_id(idx, id);
+ ioapics[idx].mp_config.apicver = io_apic_get_version(idx);
+
+ /*
+ * Build basic GSI lookup table to facilitate gsi->io_apic lookups
+ * and to prevent reprogramming of IOAPIC pins (PCI GSIs).
+ */
+ entries = io_apic_get_redir_entries(idx);
+ gsi_end = gsi_base + entries - 1;
+ for_each_ioapic(ioapic) {
+ gsi_cfg = mp_ioapic_gsi_routing(ioapic);
+ if ((gsi_base >= gsi_cfg->gsi_base &&
+ gsi_base <= gsi_cfg->gsi_end) ||
+ (gsi_end >= gsi_cfg->gsi_base &&
+ gsi_end <= gsi_cfg->gsi_end)) {
+ pr_warn("GSI range [%u-%u] for new IOAPIC conflicts with GSI[%u-%u]\n",
+ gsi_base, gsi_end,
+ gsi_cfg->gsi_base, gsi_cfg->gsi_end);
+ clear_fixmap(FIX_IO_APIC_BASE_0 + idx);
+ return -ENOSPC;
+ }
+ }
+ gsi_cfg = mp_ioapic_gsi_routing(idx);
+ gsi_cfg->gsi_base = gsi_base;
+ gsi_cfg->gsi_end = gsi_end;
+
+ ioapics[idx].irqdomain = NULL;
+ ioapics[idx].irqdomain_cfg = *cfg;
+
+ /*
+ * If mp_register_ioapic() is called during early boot stage when
+ * walking ACPI/DT tables, it's too early to create irqdomain,
+ * we are still using bootmem allocator. So delay it to setup_IO_APIC().
+ */
+ if (hotplug) {
+ if (mp_irqdomain_create(idx)) {
+ clear_fixmap(FIX_IO_APIC_BASE_0 + idx);
+ return -ENOMEM;
+ }
+ alloc_ioapic_saved_registers(idx);
+ }
+
+ if (gsi_cfg->gsi_end >= gsi_top)
+ gsi_top = gsi_cfg->gsi_end + 1;
+ if (nr_ioapics <= idx)
+ nr_ioapics = idx + 1;
+
+ /* Set nr_registers to mark entry present */
+ ioapics[idx].nr_registers = entries;
+
+ pr_info("IOAPIC[%d]: apic_id %d, version %d, address 0x%x, GSI %d-%d\n",
+ idx, mpc_ioapic_id(idx),
+ mpc_ioapic_ver(idx), mpc_ioapic_addr(idx),
+ gsi_cfg->gsi_base, gsi_cfg->gsi_end);
+
+ return 0;
+}
+
+int mp_unregister_ioapic(u32 gsi_base)
+{
+ int ioapic, pin;
+ int found = 0;
+
+ for_each_ioapic(ioapic)
+ if (ioapics[ioapic].gsi_config.gsi_base == gsi_base) {
+ found = 1;
+ break;
+ }
+ if (!found) {
+ pr_warn("can't find IOAPIC for GSI %d\n", gsi_base);
+ return -ENODEV;
+ }
+
+ for_each_pin(ioapic, pin) {
+ u32 gsi = mp_pin_to_gsi(ioapic, pin);
+ int irq = mp_map_gsi_to_irq(gsi, 0, NULL);
+ struct mp_chip_data *data;
+
+ if (irq >= 0) {
+ data = irq_get_chip_data(irq);
+ if (data && data->count) {
+ pr_warn("pin%d on IOAPIC%d is still in use.\n",
+ pin, ioapic);
+ return -EBUSY;
+ }
+ }
+ }
+
+ /* Mark entry not present */
+ ioapics[ioapic].nr_registers = 0;
+ ioapic_destroy_irqdomain(ioapic);
+ free_ioapic_saved_registers(ioapic);
+ if (ioapics[ioapic].iomem_res)
+ release_resource(ioapics[ioapic].iomem_res);
+ clear_fixmap(FIX_IO_APIC_BASE_0 + ioapic);
+ memset(&ioapics[ioapic], 0, sizeof(ioapics[ioapic]));
+
+ return 0;
+}
+
+int mp_ioapic_registered(u32 gsi_base)
+{
+ int ioapic;
+
+ for_each_ioapic(ioapic)
+ if (ioapics[ioapic].gsi_config.gsi_base == gsi_base)
+ return 1;
+
+ return 0;
+}
+
+static void mp_irqdomain_get_attr(u32 gsi, struct mp_chip_data *data,
+ struct irq_alloc_info *info)
+{
+ if (info && info->ioapic.valid) {
+ data->is_level = info->ioapic.is_level;
+ data->active_low = info->ioapic.active_low;
+ } else if (__acpi_get_override_irq(gsi, &data->is_level,
+ &data->active_low) < 0) {
+ /* PCI interrupts are always active low level triggered. */
+ data->is_level = true;
+ data->active_low = true;
+ }
+}
+
+/*
+ * Configure the I/O-APIC specific fields in the routing entry.
+ *
+ * This is important to setup the I/O-APIC specific bits (is_level,
+ * active_low, masked) because the underlying parent domain will only
+ * provide the routing information and is oblivious of the I/O-APIC
+ * specific bits.
+ *
+ * The entry is just preconfigured at this point and not written into the
+ * RTE. This happens later during activation which will fill in the actual
+ * routing information.
+ */
+static void mp_preconfigure_entry(struct mp_chip_data *data)
+{
+ struct IO_APIC_route_entry *entry = &data->entry;
+
+ memset(entry, 0, sizeof(*entry));
+ entry->is_level = data->is_level;
+ entry->active_low = data->active_low;
+ /*
+ * Mask level triggered irqs. Edge triggered irqs are masked
+ * by the irq core code in case they fire.
+ */
+ entry->masked = data->is_level;
+}
+
+int mp_irqdomain_alloc(struct irq_domain *domain, unsigned int virq,
+ unsigned int nr_irqs, void *arg)
+{
+ struct irq_alloc_info *info = arg;
+ struct mp_chip_data *data;
+ struct irq_data *irq_data;
+ int ret, ioapic, pin;
+ unsigned long flags;
+
+ if (!info || nr_irqs > 1)
+ return -EINVAL;
+ irq_data = irq_domain_get_irq_data(domain, virq);
+ if (!irq_data)
+ return -EINVAL;
+
+ ioapic = mp_irqdomain_ioapic_idx(domain);
+ pin = info->ioapic.pin;
+ if (irq_find_mapping(domain, (irq_hw_number_t)pin) > 0)
+ return -EEXIST;
+
+ data = kzalloc(sizeof(*data), GFP_KERNEL);
+ if (!data)
+ return -ENOMEM;
+
+ ret = irq_domain_alloc_irqs_parent(domain, virq, nr_irqs, info);
+ if (ret < 0) {
+ kfree(data);
+ return ret;
+ }
+
+ INIT_LIST_HEAD(&data->irq_2_pin);
+ irq_data->hwirq = info->ioapic.pin;
+ irq_data->chip = (domain->parent == x86_vector_domain) ?
+ &ioapic_chip : &ioapic_ir_chip;
+ irq_data->chip_data = data;
+ mp_irqdomain_get_attr(mp_pin_to_gsi(ioapic, pin), data, info);
+
+ add_pin_to_irq_node(data, ioapic_alloc_attr_node(info), ioapic, pin);
+
+ mp_preconfigure_entry(data);
+ mp_register_handler(virq, data->is_level);
+
+ local_irq_save(flags);
+ if (virq < nr_legacy_irqs())
+ legacy_pic->mask(virq);
+ local_irq_restore(flags);
+
+ apic_printk(APIC_VERBOSE, KERN_DEBUG
+ "IOAPIC[%d]: Preconfigured routing entry (%d-%d -> IRQ %d Level:%i ActiveLow:%i)\n",
+ ioapic, mpc_ioapic_id(ioapic), pin, virq,
+ data->is_level, data->active_low);
+ return 0;
+}
+
+void mp_irqdomain_free(struct irq_domain *domain, unsigned int virq,
+ unsigned int nr_irqs)
+{
+ struct irq_data *irq_data;
+ struct mp_chip_data *data;
+
+ BUG_ON(nr_irqs != 1);
+ irq_data = irq_domain_get_irq_data(domain, virq);
+ if (irq_data && irq_data->chip_data) {
+ data = irq_data->chip_data;
+ __remove_pin_from_irq(data, mp_irqdomain_ioapic_idx(domain),
+ (int)irq_data->hwirq);
+ WARN_ON(!list_empty(&data->irq_2_pin));
+ kfree(irq_data->chip_data);
+ }
+ irq_domain_free_irqs_top(domain, virq, nr_irqs);
+}
+
+int mp_irqdomain_activate(struct irq_domain *domain,
+ struct irq_data *irq_data, bool reserve)
+{
+ unsigned long flags;
+
+ raw_spin_lock_irqsave(&ioapic_lock, flags);
+ ioapic_configure_entry(irq_data);
+ raw_spin_unlock_irqrestore(&ioapic_lock, flags);
+ return 0;
+}
+
+void mp_irqdomain_deactivate(struct irq_domain *domain,
+ struct irq_data *irq_data)
+{
+ /* It won't be called for IRQ with multiple IOAPIC pins associated */
+ ioapic_mask_entry(mp_irqdomain_ioapic_idx(domain),
+ (int)irq_data->hwirq);
+}
+
+int mp_irqdomain_ioapic_idx(struct irq_domain *domain)
+{
+ return (int)(long)domain->host_data;
+}
+
+const struct irq_domain_ops mp_ioapic_irqdomain_ops = {
+ .alloc = mp_irqdomain_alloc,
+ .free = mp_irqdomain_free,
+ .activate = mp_irqdomain_activate,
+ .deactivate = mp_irqdomain_deactivate,
+};
diff --git a/arch/x86/kernel/apic/ipi.c b/arch/x86/kernel/apic/ipi.c
new file mode 100644
index 0000000000..a44ba7209e
--- /dev/null
+++ b/arch/x86/kernel/apic/ipi.c
@@ -0,0 +1,311 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <linux/cpumask.h>
+#include <linux/delay.h>
+#include <linux/smp.h>
+
+#include <asm/io_apic.h>
+
+#include "local.h"
+
+DEFINE_STATIC_KEY_FALSE(apic_use_ipi_shorthand);
+
+#ifdef CONFIG_SMP
+static int apic_ipi_shorthand_off __ro_after_init;
+
+static __init int apic_ipi_shorthand(char *str)
+{
+ get_option(&str, &apic_ipi_shorthand_off);
+ return 1;
+}
+__setup("no_ipi_broadcast=", apic_ipi_shorthand);
+
+static int __init print_ipi_mode(void)
+{
+ pr_info("IPI shorthand broadcast: %s\n",
+ apic_ipi_shorthand_off ? "disabled" : "enabled");
+ return 0;
+}
+late_initcall(print_ipi_mode);
+
+void apic_smt_update(void)
+{
+ /*
+ * Do not switch to broadcast mode if:
+ * - Disabled on the command line
+ * - Only a single CPU is online
+ * - Not all present CPUs have been at least booted once
+ *
+ * The latter is important as the local APIC might be in some
+ * random state and a broadcast might cause havoc. That's
+ * especially true for NMI broadcasting.
+ */
+ if (apic_ipi_shorthand_off || num_online_cpus() == 1 ||
+ !cpumask_equal(cpu_present_mask, &cpus_booted_once_mask)) {
+ static_branch_disable(&apic_use_ipi_shorthand);
+ } else {
+ static_branch_enable(&apic_use_ipi_shorthand);
+ }
+}
+
+void apic_send_IPI_allbutself(unsigned int vector)
+{
+ if (num_online_cpus() < 2)
+ return;
+
+ if (static_branch_likely(&apic_use_ipi_shorthand))
+ __apic_send_IPI_allbutself(vector);
+ else
+ __apic_send_IPI_mask_allbutself(cpu_online_mask, vector);
+}
+
+/*
+ * Send a 'reschedule' IPI to another CPU. It goes straight through and
+ * wastes no time serializing anything. Worst case is that we lose a
+ * reschedule ...
+ */
+void native_smp_send_reschedule(int cpu)
+{
+ if (unlikely(cpu_is_offline(cpu))) {
+ WARN(1, "sched: Unexpected reschedule of offline CPU#%d!\n", cpu);
+ return;
+ }
+ __apic_send_IPI(cpu, RESCHEDULE_VECTOR);
+}
+
+void native_send_call_func_single_ipi(int cpu)
+{
+ __apic_send_IPI(cpu, CALL_FUNCTION_SINGLE_VECTOR);
+}
+
+void native_send_call_func_ipi(const struct cpumask *mask)
+{
+ if (static_branch_likely(&apic_use_ipi_shorthand)) {
+ unsigned int cpu = smp_processor_id();
+
+ if (!cpumask_or_equal(mask, cpumask_of(cpu), cpu_online_mask))
+ goto sendmask;
+
+ if (cpumask_test_cpu(cpu, mask))
+ __apic_send_IPI_all(CALL_FUNCTION_VECTOR);
+ else if (num_online_cpus() > 1)
+ __apic_send_IPI_allbutself(CALL_FUNCTION_VECTOR);
+ return;
+ }
+
+sendmask:
+ __apic_send_IPI_mask(mask, CALL_FUNCTION_VECTOR);
+}
+
+#endif /* CONFIG_SMP */
+
+static inline int __prepare_ICR2(unsigned int mask)
+{
+ return SET_XAPIC_DEST_FIELD(mask);
+}
+
+u32 apic_mem_wait_icr_idle_timeout(void)
+{
+ int cnt;
+
+ for (cnt = 0; cnt < 1000; cnt++) {
+ if (!(apic_read(APIC_ICR) & APIC_ICR_BUSY))
+ return 0;
+ inc_irq_stat(icr_read_retry_count);
+ udelay(100);
+ }
+ return APIC_ICR_BUSY;
+}
+
+void apic_mem_wait_icr_idle(void)
+{
+ while (native_apic_mem_read(APIC_ICR) & APIC_ICR_BUSY)
+ cpu_relax();
+}
+
+/*
+ * This is safe against interruption because it only writes the lower 32
+ * bits of the APIC_ICR register. The destination field is ignored for
+ * short hand IPIs.
+ *
+ * wait_icr_idle()
+ * write(ICR2, dest)
+ * NMI
+ * wait_icr_idle()
+ * write(ICR)
+ * wait_icr_idle()
+ * write(ICR)
+ *
+ * This function does not need to disable interrupts as there is no ICR2
+ * interaction. The memory write is direct except when the machine is
+ * affected by the 11AP Pentium erratum, which turns the plain write into
+ * an XCHG operation.
+ */
+static void __default_send_IPI_shortcut(unsigned int shortcut, int vector)
+{
+ /*
+ * Wait for the previous ICR command to complete. Use
+ * safe_apic_wait_icr_idle() for the NMI vector as there have been
+ * issues where otherwise the system hangs when the panic CPU tries
+ * to stop the others before launching the kdump kernel.
+ */
+ if (unlikely(vector == NMI_VECTOR))
+ apic_mem_wait_icr_idle_timeout();
+ else
+ apic_mem_wait_icr_idle();
+
+ /* Destination field (ICR2) and the destination mode are ignored */
+ native_apic_mem_write(APIC_ICR, __prepare_ICR(shortcut, vector, 0));
+}
+
+/*
+ * This is used to send an IPI with no shorthand notation (the destination is
+ * specified in bits 56 to 63 of the ICR).
+ */
+void __default_send_IPI_dest_field(unsigned int dest_mask, int vector,
+ unsigned int dest_mode)
+{
+ /* See comment in __default_send_IPI_shortcut() */
+ if (unlikely(vector == NMI_VECTOR))
+ apic_mem_wait_icr_idle_timeout();
+ else
+ apic_mem_wait_icr_idle();
+
+ /* Set the IPI destination field in the ICR */
+ native_apic_mem_write(APIC_ICR2, __prepare_ICR2(dest_mask));
+ /* Send it with the proper destination mode */
+ native_apic_mem_write(APIC_ICR, __prepare_ICR(0, vector, dest_mode));
+}
+
+void default_send_IPI_single_phys(int cpu, int vector)
+{
+ unsigned long flags;
+
+ local_irq_save(flags);
+ __default_send_IPI_dest_field(per_cpu(x86_cpu_to_apicid, cpu),
+ vector, APIC_DEST_PHYSICAL);
+ local_irq_restore(flags);
+}
+
+void default_send_IPI_mask_sequence_phys(const struct cpumask *mask, int vector)
+{
+ unsigned long flags;
+ unsigned long cpu;
+
+ local_irq_save(flags);
+ for_each_cpu(cpu, mask) {
+ __default_send_IPI_dest_field(per_cpu(x86_cpu_to_apicid,
+ cpu), vector, APIC_DEST_PHYSICAL);
+ }
+ local_irq_restore(flags);
+}
+
+void default_send_IPI_mask_allbutself_phys(const struct cpumask *mask,
+ int vector)
+{
+ unsigned int cpu, this_cpu = smp_processor_id();
+ unsigned long flags;
+
+ local_irq_save(flags);
+ for_each_cpu(cpu, mask) {
+ if (cpu == this_cpu)
+ continue;
+ __default_send_IPI_dest_field(per_cpu(x86_cpu_to_apicid,
+ cpu), vector, APIC_DEST_PHYSICAL);
+ }
+ local_irq_restore(flags);
+}
+
+/*
+ * Helper function for APICs which insist on cpumasks
+ */
+void default_send_IPI_single(int cpu, int vector)
+{
+ __apic_send_IPI_mask(cpumask_of(cpu), vector);
+}
+
+void default_send_IPI_allbutself(int vector)
+{
+ __default_send_IPI_shortcut(APIC_DEST_ALLBUT, vector);
+}
+
+void default_send_IPI_all(int vector)
+{
+ __default_send_IPI_shortcut(APIC_DEST_ALLINC, vector);
+}
+
+void default_send_IPI_self(int vector)
+{
+ __default_send_IPI_shortcut(APIC_DEST_SELF, vector);
+}
+
+#ifdef CONFIG_X86_32
+void default_send_IPI_mask_sequence_logical(const struct cpumask *mask, int vector)
+{
+ unsigned long flags;
+ unsigned int cpu;
+
+ local_irq_save(flags);
+ for_each_cpu(cpu, mask)
+ __default_send_IPI_dest_field(1U << cpu, vector, APIC_DEST_LOGICAL);
+ local_irq_restore(flags);
+}
+
+void default_send_IPI_mask_allbutself_logical(const struct cpumask *mask,
+ int vector)
+{
+ unsigned int cpu, this_cpu = smp_processor_id();
+ unsigned long flags;
+
+ local_irq_save(flags);
+ for_each_cpu(cpu, mask) {
+ if (cpu == this_cpu)
+ continue;
+ __default_send_IPI_dest_field(1U << cpu, vector, APIC_DEST_LOGICAL);
+ }
+ local_irq_restore(flags);
+}
+
+void default_send_IPI_mask_logical(const struct cpumask *cpumask, int vector)
+{
+ unsigned long mask = cpumask_bits(cpumask)[0];
+ unsigned long flags;
+
+ if (!mask)
+ return;
+
+ 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);
+}
+
+#ifdef CONFIG_SMP
+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 = read_apic_id();
+ if (apicid == BAD_APICID)
+ return 0;
+
+ cpuid = convert_apicid_to_cpu(apicid);
+
+ return cpuid >= 0 ? cpuid : 0;
+}
+#endif
+#endif
diff --git a/arch/x86/kernel/apic/local.h b/arch/x86/kernel/apic/local.h
new file mode 100644
index 0000000000..ec219c659c
--- /dev/null
+++ b/arch/x86/kernel/apic/local.h
@@ -0,0 +1,85 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Historical copyright notices:
+ *
+ * Copyright 2004 James Cleverdon, IBM.
+ * (c) 1995 Alan Cox, Building #3 <alan@redhat.com>
+ * (c) 1998-99, 2000 Ingo Molnar <mingo@redhat.com>
+ * (c) 2002,2003 Andi Kleen, SuSE Labs.
+ */
+
+#include <linux/jump_label.h>
+
+#include <asm/irq_vectors.h>
+#include <asm/apic.h>
+
+/* X2APIC */
+void __x2apic_send_IPI_dest(unsigned int apicid, int vector, unsigned int dest);
+unsigned int x2apic_get_apic_id(unsigned long id);
+u32 x2apic_set_apic_id(unsigned int id);
+int x2apic_phys_pkg_id(int initial_apicid, int index_msb);
+
+void x2apic_send_IPI_all(int vector);
+void x2apic_send_IPI_allbutself(int vector);
+void x2apic_send_IPI_self(int vector);
+extern u32 x2apic_max_apicid;
+
+/* IPI */
+
+DECLARE_STATIC_KEY_FALSE(apic_use_ipi_shorthand);
+
+static inline unsigned int __prepare_ICR(unsigned int shortcut, int vector,
+ unsigned int dest)
+{
+ unsigned int icr = shortcut | dest;
+
+ switch (vector) {
+ default:
+ icr |= APIC_DM_FIXED | vector;
+ break;
+ case NMI_VECTOR:
+ icr |= APIC_DM_NMI;
+ break;
+ }
+ return icr;
+}
+
+void default_init_apic_ldr(void);
+
+void apic_mem_wait_icr_idle(void);
+u32 apic_mem_wait_icr_idle_timeout(void);
+
+/*
+ * This is used to send an IPI with no shorthand notation (the destination is
+ * specified in bits 56 to 63 of the ICR).
+ */
+void __default_send_IPI_dest_field(unsigned int mask, int vector, unsigned int dest);
+
+void default_send_IPI_single(int cpu, int vector);
+void default_send_IPI_single_phys(int cpu, int vector);
+void default_send_IPI_mask_sequence_phys(const struct cpumask *mask, int vector);
+void default_send_IPI_mask_allbutself_phys(const struct cpumask *mask, int vector);
+void default_send_IPI_allbutself(int vector);
+void default_send_IPI_all(int vector);
+void default_send_IPI_self(int vector);
+
+bool default_apic_id_registered(void);
+
+#ifdef CONFIG_X86_32
+void default_send_IPI_mask_sequence_logical(const struct cpumask *mask, int vector);
+void default_send_IPI_mask_allbutself_logical(const struct cpumask *mask, int vector);
+void default_send_IPI_mask_logical(const struct cpumask *mask, int vector);
+void x86_32_probe_bigsmp_early(void);
+void x86_32_install_bigsmp(void);
+#else
+static inline void x86_32_probe_bigsmp_early(void) { }
+static inline void x86_32_install_bigsmp(void) { }
+#endif
+
+#ifdef CONFIG_X86_BIGSMP
+bool apic_bigsmp_possible(bool cmdline_selected);
+void apic_bigsmp_force(void);
+#else
+static inline bool apic_bigsmp_possible(bool cmdline_selected) { return false; };
+static inline void apic_bigsmp_force(void) { }
+#endif
diff --git a/arch/x86/kernel/apic/msi.c b/arch/x86/kernel/apic/msi.c
new file mode 100644
index 0000000000..d9651f15ae
--- /dev/null
+++ b/arch/x86/kernel/apic/msi.c
@@ -0,0 +1,395 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Support of MSI, HPET and DMAR interrupts.
+ *
+ * Copyright (C) 1997, 1998, 1999, 2000, 2009 Ingo Molnar, Hajnalka Szabo
+ * Moved from arch/x86/kernel/apic/io_apic.c.
+ * Jiang Liu <jiang.liu@linux.intel.com>
+ * Convert to hierarchical irqdomain
+ */
+#include <linux/mm.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/pci.h>
+#include <linux/dmar.h>
+#include <linux/hpet.h>
+#include <linux/msi.h>
+#include <asm/irqdomain.h>
+#include <asm/hpet.h>
+#include <asm/hw_irq.h>
+#include <asm/apic.h>
+#include <asm/irq_remapping.h>
+#include <asm/xen/hypervisor.h>
+
+struct irq_domain *x86_pci_msi_default_domain __ro_after_init;
+
+static void irq_msi_update_msg(struct irq_data *irqd, struct irq_cfg *cfg)
+{
+ struct msi_msg msg[2] = { [1] = { }, };
+
+ __irq_msi_compose_msg(cfg, msg, false);
+ irq_data_get_irq_chip(irqd)->irq_write_msi_msg(irqd, msg);
+}
+
+static int
+msi_set_affinity(struct irq_data *irqd, const struct cpumask *mask, bool force)
+{
+ struct irq_cfg old_cfg, *cfg = irqd_cfg(irqd);
+ struct irq_data *parent = irqd->parent_data;
+ unsigned int cpu;
+ int ret;
+
+ /* Save the current configuration */
+ cpu = cpumask_first(irq_data_get_effective_affinity_mask(irqd));
+ old_cfg = *cfg;
+
+ /* Allocate a new target vector */
+ ret = parent->chip->irq_set_affinity(parent, mask, force);
+ if (ret < 0 || ret == IRQ_SET_MASK_OK_DONE)
+ return ret;
+
+ /*
+ * For non-maskable and non-remapped MSI interrupts the migration
+ * to a different destination CPU and a different vector has to be
+ * done careful to handle the possible stray interrupt which can be
+ * caused by the non-atomic update of the address/data pair.
+ *
+ * Direct update is possible when:
+ * - The MSI is maskable (remapped MSI does not use this code path).
+ * The reservation mode bit is set in this case.
+ * - The new vector is the same as the old vector
+ * - The old vector is MANAGED_IRQ_SHUTDOWN_VECTOR (interrupt starts up)
+ * - The interrupt is not yet started up
+ * - The new destination CPU is the same as the old destination CPU
+ */
+ if (!irqd_can_reserve(irqd) ||
+ cfg->vector == old_cfg.vector ||
+ old_cfg.vector == MANAGED_IRQ_SHUTDOWN_VECTOR ||
+ !irqd_is_started(irqd) ||
+ cfg->dest_apicid == old_cfg.dest_apicid) {
+ irq_msi_update_msg(irqd, cfg);
+ return ret;
+ }
+
+ /*
+ * Paranoia: Validate that the interrupt target is the local
+ * CPU.
+ */
+ if (WARN_ON_ONCE(cpu != smp_processor_id())) {
+ irq_msi_update_msg(irqd, cfg);
+ return ret;
+ }
+
+ /*
+ * Redirect the interrupt to the new vector on the current CPU
+ * first. This might cause a spurious interrupt on this vector if
+ * the device raises an interrupt right between this update and the
+ * update to the final destination CPU.
+ *
+ * If the vector is in use then the installed device handler will
+ * denote it as spurious which is no harm as this is a rare event
+ * and interrupt handlers have to cope with spurious interrupts
+ * anyway. If the vector is unused, then it is marked so it won't
+ * trigger the 'No irq handler for vector' warning in
+ * common_interrupt().
+ *
+ * This requires to hold vector lock to prevent concurrent updates to
+ * the affected vector.
+ */
+ lock_vector_lock();
+
+ /*
+ * Mark the new target vector on the local CPU if it is currently
+ * unused. Reuse the VECTOR_RETRIGGERED state which is also used in
+ * the CPU hotplug path for a similar purpose. This cannot be
+ * undone here as the current CPU has interrupts disabled and
+ * cannot handle the interrupt before the whole set_affinity()
+ * section is done. In the CPU unplug case, the current CPU is
+ * about to vanish and will not handle any interrupts anymore. The
+ * vector is cleaned up when the CPU comes online again.
+ */
+ if (IS_ERR_OR_NULL(this_cpu_read(vector_irq[cfg->vector])))
+ this_cpu_write(vector_irq[cfg->vector], VECTOR_RETRIGGERED);
+
+ /* Redirect it to the new vector on the local CPU temporarily */
+ old_cfg.vector = cfg->vector;
+ irq_msi_update_msg(irqd, &old_cfg);
+
+ /* Now transition it to the target CPU */
+ irq_msi_update_msg(irqd, cfg);
+
+ /*
+ * All interrupts after this point are now targeted at the new
+ * vector/CPU.
+ *
+ * Drop vector lock before testing whether the temporary assignment
+ * to the local CPU was hit by an interrupt raised in the device,
+ * because the retrigger function acquires vector lock again.
+ */
+ unlock_vector_lock();
+
+ /*
+ * Check whether the transition raced with a device interrupt and
+ * is pending in the local APICs IRR. It is safe to do this outside
+ * of vector lock as the irq_desc::lock of this interrupt is still
+ * held and interrupts are disabled: The check is not accessing the
+ * underlying vector store. It's just checking the local APIC's
+ * IRR.
+ */
+ if (lapic_vector_set_in_irr(cfg->vector))
+ irq_data_get_irq_chip(irqd)->irq_retrigger(irqd);
+
+ return ret;
+}
+
+/**
+ * pci_dev_has_default_msi_parent_domain - Check whether the device has the default
+ * MSI parent domain associated
+ * @dev: Pointer to the PCI device
+ */
+bool pci_dev_has_default_msi_parent_domain(struct pci_dev *dev)
+{
+ struct irq_domain *domain = dev_get_msi_domain(&dev->dev);
+
+ if (!domain)
+ domain = dev_get_msi_domain(&dev->bus->dev);
+ if (!domain)
+ return false;
+
+ return domain == x86_vector_domain;
+}
+
+/**
+ * x86_msi_prepare - Setup of msi_alloc_info_t for allocations
+ * @domain: The domain for which this setup happens
+ * @dev: The device for which interrupts are allocated
+ * @nvec: The number of vectors to allocate
+ * @alloc: The allocation info structure to initialize
+ *
+ * This function is to be used for all types of MSI domains above the x86
+ * vector domain and any intermediates. It is always invoked from the
+ * top level interrupt domain. The domain specific allocation
+ * functionality is determined via the @domain's bus token which allows to
+ * map the X86 specific allocation type.
+ */
+static int x86_msi_prepare(struct irq_domain *domain, struct device *dev,
+ int nvec, msi_alloc_info_t *alloc)
+{
+ struct msi_domain_info *info = domain->host_data;
+
+ init_irq_alloc_info(alloc, NULL);
+
+ switch (info->bus_token) {
+ case DOMAIN_BUS_PCI_DEVICE_MSI:
+ alloc->type = X86_IRQ_ALLOC_TYPE_PCI_MSI;
+ return 0;
+ case DOMAIN_BUS_PCI_DEVICE_MSIX:
+ case DOMAIN_BUS_PCI_DEVICE_IMS:
+ alloc->type = X86_IRQ_ALLOC_TYPE_PCI_MSIX;
+ return 0;
+ default:
+ return -EINVAL;
+ }
+}
+
+/**
+ * x86_init_dev_msi_info - Domain info setup for MSI domains
+ * @dev: The device for which the domain should be created
+ * @domain: The (root) domain providing this callback
+ * @real_parent: The real parent domain of the to initialize domain
+ * @info: The domain info for the to initialize domain
+ *
+ * This function is to be used for all types of MSI domains above the x86
+ * vector domain and any intermediates. The domain specific functionality
+ * is determined via the @real_parent.
+ */
+static bool x86_init_dev_msi_info(struct device *dev, struct irq_domain *domain,
+ struct irq_domain *real_parent, struct msi_domain_info *info)
+{
+ const struct msi_parent_ops *pops = real_parent->msi_parent_ops;
+
+ /* MSI parent domain specific settings */
+ switch (real_parent->bus_token) {
+ case DOMAIN_BUS_ANY:
+ /* Only the vector domain can have the ANY token */
+ if (WARN_ON_ONCE(domain != real_parent))
+ return false;
+ info->chip->irq_set_affinity = msi_set_affinity;
+ break;
+ case DOMAIN_BUS_DMAR:
+ case DOMAIN_BUS_AMDVI:
+ break;
+ default:
+ WARN_ON_ONCE(1);
+ return false;
+ }
+
+ /* Is the target supported? */
+ switch(info->bus_token) {
+ case DOMAIN_BUS_PCI_DEVICE_MSI:
+ case DOMAIN_BUS_PCI_DEVICE_MSIX:
+ break;
+ case DOMAIN_BUS_PCI_DEVICE_IMS:
+ if (!(pops->supported_flags & MSI_FLAG_PCI_IMS))
+ return false;
+ break;
+ default:
+ WARN_ON_ONCE(1);
+ return false;
+ }
+
+ /*
+ * Mask out the domain specific MSI feature flags which are not
+ * supported by the real parent.
+ */
+ info->flags &= pops->supported_flags;
+ /* Enforce the required flags */
+ info->flags |= X86_VECTOR_MSI_FLAGS_REQUIRED;
+
+ /* This is always invoked from the top level MSI domain! */
+ info->ops->msi_prepare = x86_msi_prepare;
+
+ info->chip->irq_ack = irq_chip_ack_parent;
+ info->chip->irq_retrigger = irq_chip_retrigger_hierarchy;
+ info->chip->flags |= IRQCHIP_SKIP_SET_WAKE |
+ IRQCHIP_AFFINITY_PRE_STARTUP;
+
+ info->handler = handle_edge_irq;
+ info->handler_name = "edge";
+
+ return true;
+}
+
+static const struct msi_parent_ops x86_vector_msi_parent_ops = {
+ .supported_flags = X86_VECTOR_MSI_FLAGS_SUPPORTED,
+ .init_dev_msi_info = x86_init_dev_msi_info,
+};
+
+struct irq_domain * __init native_create_pci_msi_domain(void)
+{
+ if (apic_is_disabled)
+ return NULL;
+
+ x86_vector_domain->flags |= IRQ_DOMAIN_FLAG_MSI_PARENT;
+ x86_vector_domain->msi_parent_ops = &x86_vector_msi_parent_ops;
+ return x86_vector_domain;
+}
+
+void __init x86_create_pci_msi_domain(void)
+{
+ x86_pci_msi_default_domain = x86_init.irqs.create_pci_msi_domain();
+}
+
+/* Keep around for hyperV */
+int pci_msi_prepare(struct irq_domain *domain, struct device *dev, int nvec,
+ msi_alloc_info_t *arg)
+{
+ init_irq_alloc_info(arg, NULL);
+
+ if (to_pci_dev(dev)->msix_enabled)
+ arg->type = X86_IRQ_ALLOC_TYPE_PCI_MSIX;
+ else
+ arg->type = X86_IRQ_ALLOC_TYPE_PCI_MSI;
+ return 0;
+}
+EXPORT_SYMBOL_GPL(pci_msi_prepare);
+
+#ifdef CONFIG_DMAR_TABLE
+/*
+ * The Intel IOMMU (ab)uses the high bits of the MSI address to contain the
+ * high bits of the destination APIC ID. This can't be done in the general
+ * case for MSIs as it would be targeting real memory above 4GiB not the
+ * APIC.
+ */
+static void dmar_msi_compose_msg(struct irq_data *data, struct msi_msg *msg)
+{
+ __irq_msi_compose_msg(irqd_cfg(data), msg, true);
+}
+
+static void dmar_msi_write_msg(struct irq_data *data, struct msi_msg *msg)
+{
+ dmar_msi_write(data->irq, msg);
+}
+
+static struct irq_chip dmar_msi_controller = {
+ .name = "DMAR-MSI",
+ .irq_unmask = dmar_msi_unmask,
+ .irq_mask = dmar_msi_mask,
+ .irq_ack = irq_chip_ack_parent,
+ .irq_set_affinity = msi_domain_set_affinity,
+ .irq_retrigger = irq_chip_retrigger_hierarchy,
+ .irq_compose_msi_msg = dmar_msi_compose_msg,
+ .irq_write_msi_msg = dmar_msi_write_msg,
+ .flags = IRQCHIP_SKIP_SET_WAKE |
+ IRQCHIP_AFFINITY_PRE_STARTUP,
+};
+
+static int dmar_msi_init(struct irq_domain *domain,
+ struct msi_domain_info *info, unsigned int virq,
+ irq_hw_number_t hwirq, msi_alloc_info_t *arg)
+{
+ irq_domain_set_info(domain, virq, arg->devid, info->chip, NULL,
+ handle_edge_irq, arg->data, "edge");
+
+ return 0;
+}
+
+static struct msi_domain_ops dmar_msi_domain_ops = {
+ .msi_init = dmar_msi_init,
+};
+
+static struct msi_domain_info dmar_msi_domain_info = {
+ .ops = &dmar_msi_domain_ops,
+ .chip = &dmar_msi_controller,
+ .flags = MSI_FLAG_USE_DEF_DOM_OPS,
+};
+
+static struct irq_domain *dmar_get_irq_domain(void)
+{
+ static struct irq_domain *dmar_domain;
+ static DEFINE_MUTEX(dmar_lock);
+ struct fwnode_handle *fn;
+
+ mutex_lock(&dmar_lock);
+ if (dmar_domain)
+ goto out;
+
+ fn = irq_domain_alloc_named_fwnode("DMAR-MSI");
+ if (fn) {
+ dmar_domain = msi_create_irq_domain(fn, &dmar_msi_domain_info,
+ x86_vector_domain);
+ if (!dmar_domain)
+ irq_domain_free_fwnode(fn);
+ }
+out:
+ mutex_unlock(&dmar_lock);
+ return dmar_domain;
+}
+
+int dmar_alloc_hwirq(int id, int node, void *arg)
+{
+ struct irq_domain *domain = dmar_get_irq_domain();
+ struct irq_alloc_info info;
+
+ if (!domain)
+ return -1;
+
+ init_irq_alloc_info(&info, NULL);
+ info.type = X86_IRQ_ALLOC_TYPE_DMAR;
+ info.devid = id;
+ info.hwirq = id;
+ info.data = arg;
+
+ return irq_domain_alloc_irqs(domain, 1, node, &info);
+}
+
+void dmar_free_hwirq(int irq)
+{
+ irq_domain_free_irqs(irq, 1);
+}
+#endif
+
+bool arch_restore_msi_irqs(struct pci_dev *dev)
+{
+ return xen_initdom_restore_msi(dev);
+}
diff --git a/arch/x86/kernel/apic/probe_32.c b/arch/x86/kernel/apic/probe_32.c
new file mode 100644
index 0000000000..9a06df6cdd
--- /dev/null
+++ b/arch/x86/kernel/apic/probe_32.c
@@ -0,0 +1,140 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Default generic APIC driver. This handles up to 8 CPUs.
+ *
+ * Copyright 2003 Andi Kleen, SuSE Labs.
+ *
+ * Generic x86 APIC driver probe layer.
+ */
+#include <linux/export.h>
+#include <linux/errno.h>
+#include <linux/smp.h>
+
+#include <xen/xen.h>
+
+#include <asm/io_apic.h>
+#include <asm/apic.h>
+#include <asm/acpi.h>
+
+#include "local.h"
+
+static int default_phys_pkg_id(int cpuid_apic, int index_msb)
+{
+ return cpuid_apic >> index_msb;
+}
+
+/* should be called last. */
+static int probe_default(void)
+{
+ return 1;
+}
+
+static struct apic apic_default __ro_after_init = {
+
+ .name = "default",
+ .probe = probe_default,
+ .apic_id_registered = default_apic_id_registered,
+
+ .delivery_mode = APIC_DELIVERY_MODE_FIXED,
+ .dest_mode_logical = true,
+
+ .disable_esr = 0,
+
+ .check_apicid_used = default_check_apicid_used,
+ .init_apic_ldr = default_init_apic_ldr,
+ .ioapic_phys_id_map = default_ioapic_phys_id_map,
+ .cpu_present_to_apicid = default_cpu_present_to_apicid,
+ .phys_pkg_id = default_phys_pkg_id,
+
+ .max_apic_id = 0xFE,
+ .get_apic_id = default_get_apic_id,
+
+ .calc_dest_apicid = apic_flat_calc_apicid,
+
+ .send_IPI = default_send_IPI_single,
+ .send_IPI_mask = default_send_IPI_mask_logical,
+ .send_IPI_mask_allbutself = default_send_IPI_mask_allbutself_logical,
+ .send_IPI_allbutself = default_send_IPI_allbutself,
+ .send_IPI_all = default_send_IPI_all,
+ .send_IPI_self = default_send_IPI_self,
+
+ .read = native_apic_mem_read,
+ .write = native_apic_mem_write,
+ .eoi = native_apic_mem_eoi,
+ .icr_read = native_apic_icr_read,
+ .icr_write = native_apic_icr_write,
+ .wait_icr_idle = apic_mem_wait_icr_idle,
+ .safe_wait_icr_idle = apic_mem_wait_icr_idle_timeout,
+};
+
+apic_driver(apic_default);
+
+struct apic *apic __ro_after_init = &apic_default;
+EXPORT_SYMBOL_GPL(apic);
+
+static int cmdline_apic __initdata;
+static int __init parse_apic(char *arg)
+{
+ struct apic **drv;
+
+ if (!arg)
+ return -EINVAL;
+
+ for (drv = __apicdrivers; drv < __apicdrivers_end; drv++) {
+ if (!strcmp((*drv)->name, arg)) {
+ apic_install_driver(*drv);
+ cmdline_apic = 1;
+ return 0;
+ }
+ }
+
+ /* Parsed again by __setup for debug/verbose */
+ return 0;
+}
+early_param("apic", parse_apic);
+
+void __init x86_32_probe_bigsmp_early(void)
+{
+ if (nr_cpu_ids <= 8 || xen_pv_domain())
+ return;
+
+ if (IS_ENABLED(CONFIG_X86_BIGSMP)) {
+ switch (boot_cpu_data.x86_vendor) {
+ case X86_VENDOR_INTEL:
+ if (!APIC_XAPIC(boot_cpu_apic_version))
+ break;
+ /* P4 and above */
+ fallthrough;
+ case X86_VENDOR_HYGON:
+ case X86_VENDOR_AMD:
+ if (apic_bigsmp_possible(cmdline_apic))
+ return;
+ break;
+ }
+ }
+ pr_info("Limiting to 8 possible CPUs\n");
+ set_nr_cpu_ids(8);
+}
+
+void __init x86_32_install_bigsmp(void)
+{
+ if (nr_cpu_ids > 8 && !xen_pv_domain())
+ apic_bigsmp_force();
+}
+
+void __init x86_32_probe_apic(void)
+{
+ if (!cmdline_apic) {
+ struct apic **drv;
+
+ for (drv = __apicdrivers; drv < __apicdrivers_end; drv++) {
+ if ((*drv)->probe()) {
+ apic_install_driver(*drv);
+ break;
+ }
+ }
+ /* Not visible without early console */
+ if (drv == __apicdrivers_end)
+ panic("Didn't find an APIC driver");
+ }
+}
diff --git a/arch/x86/kernel/apic/probe_64.c b/arch/x86/kernel/apic/probe_64.c
new file mode 100644
index 0000000000..ecdf0c4121
--- /dev/null
+++ b/arch/x86/kernel/apic/probe_64.c
@@ -0,0 +1,42 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright 2004 James Cleverdon, IBM.
+ *
+ * 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/thread_info.h>
+#include <asm/apic.h>
+
+#include "local.h"
+
+/* Select the appropriate APIC driver */
+void __init x86_64_probe_apic(void)
+{
+ struct apic **drv;
+
+ enable_IR_x2apic();
+
+ for (drv = __apicdrivers; drv < __apicdrivers_end; drv++) {
+ if ((*drv)->probe && (*drv)->probe()) {
+ apic_install_driver(*drv);
+ break;
+ }
+ }
+}
+
+int __init default_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
+{
+ struct apic **drv;
+
+ for (drv = __apicdrivers; drv < __apicdrivers_end; drv++) {
+ if ((*drv)->acpi_madt_oem_check(oem_id, oem_table_id)) {
+ apic_install_driver(*drv);
+ return 1;
+ }
+ }
+ return 0;
+}
diff --git a/arch/x86/kernel/apic/vector.c b/arch/x86/kernel/apic/vector.c
new file mode 100644
index 0000000000..319448d87b
--- /dev/null
+++ b/arch/x86/kernel/apic/vector.c
@@ -0,0 +1,1394 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Local APIC related interfaces to support IOAPIC, MSI, etc.
+ *
+ * Copyright (C) 1997, 1998, 1999, 2000, 2009 Ingo Molnar, Hajnalka Szabo
+ * Moved from arch/x86/kernel/apic/io_apic.c.
+ * Jiang Liu <jiang.liu@linux.intel.com>
+ * Enable support of hierarchical irqdomains
+ */
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/seq_file.h>
+#include <linux/init.h>
+#include <linux/compiler.h>
+#include <linux/slab.h>
+#include <asm/irqdomain.h>
+#include <asm/hw_irq.h>
+#include <asm/traps.h>
+#include <asm/apic.h>
+#include <asm/i8259.h>
+#include <asm/desc.h>
+#include <asm/irq_remapping.h>
+
+#include <asm/trace/irq_vectors.h>
+
+struct apic_chip_data {
+ struct irq_cfg hw_irq_cfg;
+ unsigned int vector;
+ unsigned int prev_vector;
+ unsigned int cpu;
+ unsigned int prev_cpu;
+ unsigned int irq;
+ struct hlist_node clist;
+ unsigned int move_in_progress : 1,
+ is_managed : 1,
+ can_reserve : 1,
+ has_reserved : 1;
+};
+
+struct irq_domain *x86_vector_domain;
+EXPORT_SYMBOL_GPL(x86_vector_domain);
+static DEFINE_RAW_SPINLOCK(vector_lock);
+static cpumask_var_t vector_searchmask;
+static struct irq_chip lapic_controller;
+static struct irq_matrix *vector_matrix;
+#ifdef CONFIG_SMP
+
+static void vector_cleanup_callback(struct timer_list *tmr);
+
+struct vector_cleanup {
+ struct hlist_head head;
+ struct timer_list timer;
+};
+
+static DEFINE_PER_CPU(struct vector_cleanup, vector_cleanup) = {
+ .head = HLIST_HEAD_INIT,
+ .timer = __TIMER_INITIALIZER(vector_cleanup_callback, TIMER_PINNED),
+};
+#endif
+
+void lock_vector_lock(void)
+{
+ /* Used to the online set of cpus does not change
+ * during assign_irq_vector.
+ */
+ raw_spin_lock(&vector_lock);
+}
+
+void unlock_vector_lock(void)
+{
+ raw_spin_unlock(&vector_lock);
+}
+
+void init_irq_alloc_info(struct irq_alloc_info *info,
+ const struct cpumask *mask)
+{
+ memset(info, 0, sizeof(*info));
+ info->mask = mask;
+}
+
+void copy_irq_alloc_info(struct irq_alloc_info *dst, struct irq_alloc_info *src)
+{
+ if (src)
+ *dst = *src;
+ else
+ memset(dst, 0, sizeof(*dst));
+}
+
+static struct apic_chip_data *apic_chip_data(struct irq_data *irqd)
+{
+ if (!irqd)
+ return NULL;
+
+ while (irqd->parent_data)
+ irqd = irqd->parent_data;
+
+ return irqd->chip_data;
+}
+
+struct irq_cfg *irqd_cfg(struct irq_data *irqd)
+{
+ struct apic_chip_data *apicd = apic_chip_data(irqd);
+
+ return apicd ? &apicd->hw_irq_cfg : NULL;
+}
+EXPORT_SYMBOL_GPL(irqd_cfg);
+
+struct irq_cfg *irq_cfg(unsigned int irq)
+{
+ return irqd_cfg(irq_get_irq_data(irq));
+}
+
+static struct apic_chip_data *alloc_apic_chip_data(int node)
+{
+ struct apic_chip_data *apicd;
+
+ apicd = kzalloc_node(sizeof(*apicd), GFP_KERNEL, node);
+ if (apicd)
+ INIT_HLIST_NODE(&apicd->clist);
+ return apicd;
+}
+
+static void free_apic_chip_data(struct apic_chip_data *apicd)
+{
+ kfree(apicd);
+}
+
+static void apic_update_irq_cfg(struct irq_data *irqd, unsigned int vector,
+ unsigned int cpu)
+{
+ struct apic_chip_data *apicd = apic_chip_data(irqd);
+
+ lockdep_assert_held(&vector_lock);
+
+ apicd->hw_irq_cfg.vector = vector;
+ apicd->hw_irq_cfg.dest_apicid = apic->calc_dest_apicid(cpu);
+ irq_data_update_effective_affinity(irqd, cpumask_of(cpu));
+ trace_vector_config(irqd->irq, vector, cpu,
+ apicd->hw_irq_cfg.dest_apicid);
+}
+
+static void apic_update_vector(struct irq_data *irqd, unsigned int newvec,
+ unsigned int newcpu)
+{
+ struct apic_chip_data *apicd = apic_chip_data(irqd);
+ struct irq_desc *desc = irq_data_to_desc(irqd);
+ bool managed = irqd_affinity_is_managed(irqd);
+
+ lockdep_assert_held(&vector_lock);
+
+ trace_vector_update(irqd->irq, newvec, newcpu, apicd->vector,
+ apicd->cpu);
+
+ /*
+ * If there is no vector associated or if the associated vector is
+ * the shutdown vector, which is associated to make PCI/MSI
+ * shutdown mode work, then there is nothing to release. Clear out
+ * prev_vector for this and the offlined target case.
+ */
+ apicd->prev_vector = 0;
+ if (!apicd->vector || apicd->vector == MANAGED_IRQ_SHUTDOWN_VECTOR)
+ goto setnew;
+ /*
+ * If the target CPU of the previous vector is online, then mark
+ * the vector as move in progress and store it for cleanup when the
+ * first interrupt on the new vector arrives. If the target CPU is
+ * offline then the regular release mechanism via the cleanup
+ * vector is not possible and the vector can be immediately freed
+ * in the underlying matrix allocator.
+ */
+ if (cpu_online(apicd->cpu)) {
+ apicd->move_in_progress = true;
+ apicd->prev_vector = apicd->vector;
+ apicd->prev_cpu = apicd->cpu;
+ WARN_ON_ONCE(apicd->cpu == newcpu);
+ } else {
+ irq_matrix_free(vector_matrix, apicd->cpu, apicd->vector,
+ managed);
+ }
+
+setnew:
+ apicd->vector = newvec;
+ apicd->cpu = newcpu;
+ BUG_ON(!IS_ERR_OR_NULL(per_cpu(vector_irq, newcpu)[newvec]));
+ per_cpu(vector_irq, newcpu)[newvec] = desc;
+}
+
+static void vector_assign_managed_shutdown(struct irq_data *irqd)
+{
+ unsigned int cpu = cpumask_first(cpu_online_mask);
+
+ apic_update_irq_cfg(irqd, MANAGED_IRQ_SHUTDOWN_VECTOR, cpu);
+}
+
+static int reserve_managed_vector(struct irq_data *irqd)
+{
+ const struct cpumask *affmsk = irq_data_get_affinity_mask(irqd);
+ struct apic_chip_data *apicd = apic_chip_data(irqd);
+ unsigned long flags;
+ int ret;
+
+ raw_spin_lock_irqsave(&vector_lock, flags);
+ apicd->is_managed = true;
+ ret = irq_matrix_reserve_managed(vector_matrix, affmsk);
+ raw_spin_unlock_irqrestore(&vector_lock, flags);
+ trace_vector_reserve_managed(irqd->irq, ret);
+ return ret;
+}
+
+static void reserve_irq_vector_locked(struct irq_data *irqd)
+{
+ struct apic_chip_data *apicd = apic_chip_data(irqd);
+
+ irq_matrix_reserve(vector_matrix);
+ apicd->can_reserve = true;
+ apicd->has_reserved = true;
+ irqd_set_can_reserve(irqd);
+ trace_vector_reserve(irqd->irq, 0);
+ vector_assign_managed_shutdown(irqd);
+}
+
+static int reserve_irq_vector(struct irq_data *irqd)
+{
+ unsigned long flags;
+
+ raw_spin_lock_irqsave(&vector_lock, flags);
+ reserve_irq_vector_locked(irqd);
+ raw_spin_unlock_irqrestore(&vector_lock, flags);
+ return 0;
+}
+
+static int
+assign_vector_locked(struct irq_data *irqd, const struct cpumask *dest)
+{
+ struct apic_chip_data *apicd = apic_chip_data(irqd);
+ bool resvd = apicd->has_reserved;
+ unsigned int cpu = apicd->cpu;
+ int vector = apicd->vector;
+
+ lockdep_assert_held(&vector_lock);
+
+ /*
+ * If the current target CPU is online and in the new requested
+ * affinity mask, there is no point in moving the interrupt from
+ * one CPU to another.
+ */
+ if (vector && cpu_online(cpu) && cpumask_test_cpu(cpu, dest))
+ return 0;
+
+ /*
+ * Careful here. @apicd might either have move_in_progress set or
+ * be enqueued for cleanup. Assigning a new vector would either
+ * leave a stale vector on some CPU around or in case of a pending
+ * cleanup corrupt the hlist.
+ */
+ if (apicd->move_in_progress || !hlist_unhashed(&apicd->clist))
+ return -EBUSY;
+
+ vector = irq_matrix_alloc(vector_matrix, dest, resvd, &cpu);
+ trace_vector_alloc(irqd->irq, vector, resvd, vector);
+ if (vector < 0)
+ return vector;
+ apic_update_vector(irqd, vector, cpu);
+ apic_update_irq_cfg(irqd, vector, cpu);
+
+ return 0;
+}
+
+static int assign_irq_vector(struct irq_data *irqd, const struct cpumask *dest)
+{
+ unsigned long flags;
+ int ret;
+
+ raw_spin_lock_irqsave(&vector_lock, flags);
+ cpumask_and(vector_searchmask, dest, cpu_online_mask);
+ ret = assign_vector_locked(irqd, vector_searchmask);
+ raw_spin_unlock_irqrestore(&vector_lock, flags);
+ return ret;
+}
+
+static int assign_irq_vector_any_locked(struct irq_data *irqd)
+{
+ /* Get the affinity mask - either irq_default_affinity or (user) set */
+ const struct cpumask *affmsk = irq_data_get_affinity_mask(irqd);
+ int node = irq_data_get_node(irqd);
+
+ if (node != NUMA_NO_NODE) {
+ /* Try the intersection of @affmsk and node mask */
+ cpumask_and(vector_searchmask, cpumask_of_node(node), affmsk);
+ if (!assign_vector_locked(irqd, vector_searchmask))
+ return 0;
+ }
+
+ /* Try the full affinity mask */
+ cpumask_and(vector_searchmask, affmsk, cpu_online_mask);
+ if (!assign_vector_locked(irqd, vector_searchmask))
+ return 0;
+
+ if (node != NUMA_NO_NODE) {
+ /* Try the node mask */
+ if (!assign_vector_locked(irqd, cpumask_of_node(node)))
+ return 0;
+ }
+
+ /* Try the full online mask */
+ return assign_vector_locked(irqd, cpu_online_mask);
+}
+
+static int
+assign_irq_vector_policy(struct irq_data *irqd, struct irq_alloc_info *info)
+{
+ if (irqd_affinity_is_managed(irqd))
+ return reserve_managed_vector(irqd);
+ if (info->mask)
+ return assign_irq_vector(irqd, info->mask);
+ /*
+ * Make only a global reservation with no guarantee. A real vector
+ * is associated at activation time.
+ */
+ return reserve_irq_vector(irqd);
+}
+
+static int
+assign_managed_vector(struct irq_data *irqd, const struct cpumask *dest)
+{
+ const struct cpumask *affmsk = irq_data_get_affinity_mask(irqd);
+ struct apic_chip_data *apicd = apic_chip_data(irqd);
+ int vector, cpu;
+
+ cpumask_and(vector_searchmask, dest, affmsk);
+
+ /* set_affinity might call here for nothing */
+ if (apicd->vector && cpumask_test_cpu(apicd->cpu, vector_searchmask))
+ return 0;
+ vector = irq_matrix_alloc_managed(vector_matrix, vector_searchmask,
+ &cpu);
+ trace_vector_alloc_managed(irqd->irq, vector, vector);
+ if (vector < 0)
+ return vector;
+ apic_update_vector(irqd, vector, cpu);
+ apic_update_irq_cfg(irqd, vector, cpu);
+ return 0;
+}
+
+static void clear_irq_vector(struct irq_data *irqd)
+{
+ struct apic_chip_data *apicd = apic_chip_data(irqd);
+ bool managed = irqd_affinity_is_managed(irqd);
+ unsigned int vector = apicd->vector;
+
+ lockdep_assert_held(&vector_lock);
+
+ if (!vector)
+ return;
+
+ trace_vector_clear(irqd->irq, vector, apicd->cpu, apicd->prev_vector,
+ apicd->prev_cpu);
+
+ per_cpu(vector_irq, apicd->cpu)[vector] = VECTOR_SHUTDOWN;
+ irq_matrix_free(vector_matrix, apicd->cpu, vector, managed);
+ apicd->vector = 0;
+
+ /* Clean up move in progress */
+ vector = apicd->prev_vector;
+ if (!vector)
+ return;
+
+ per_cpu(vector_irq, apicd->prev_cpu)[vector] = VECTOR_SHUTDOWN;
+ irq_matrix_free(vector_matrix, apicd->prev_cpu, vector, managed);
+ apicd->prev_vector = 0;
+ apicd->move_in_progress = 0;
+ hlist_del_init(&apicd->clist);
+}
+
+static void x86_vector_deactivate(struct irq_domain *dom, struct irq_data *irqd)
+{
+ struct apic_chip_data *apicd = apic_chip_data(irqd);
+ unsigned long flags;
+
+ trace_vector_deactivate(irqd->irq, apicd->is_managed,
+ apicd->can_reserve, false);
+
+ /* Regular fixed assigned interrupt */
+ if (!apicd->is_managed && !apicd->can_reserve)
+ return;
+ /* If the interrupt has a global reservation, nothing to do */
+ if (apicd->has_reserved)
+ return;
+
+ raw_spin_lock_irqsave(&vector_lock, flags);
+ clear_irq_vector(irqd);
+ if (apicd->can_reserve)
+ reserve_irq_vector_locked(irqd);
+ else
+ vector_assign_managed_shutdown(irqd);
+ raw_spin_unlock_irqrestore(&vector_lock, flags);
+}
+
+static int activate_reserved(struct irq_data *irqd)
+{
+ struct apic_chip_data *apicd = apic_chip_data(irqd);
+ int ret;
+
+ ret = assign_irq_vector_any_locked(irqd);
+ if (!ret) {
+ apicd->has_reserved = false;
+ /*
+ * Core might have disabled reservation mode after
+ * allocating the irq descriptor. Ideally this should
+ * happen before allocation time, but that would require
+ * completely convoluted ways of transporting that
+ * information.
+ */
+ if (!irqd_can_reserve(irqd))
+ apicd->can_reserve = false;
+ }
+
+ /*
+ * Check to ensure that the effective affinity mask is a subset
+ * the user supplied affinity mask, and warn the user if it is not
+ */
+ if (!cpumask_subset(irq_data_get_effective_affinity_mask(irqd),
+ irq_data_get_affinity_mask(irqd))) {
+ pr_warn("irq %u: Affinity broken due to vector space exhaustion.\n",
+ irqd->irq);
+ }
+
+ return ret;
+}
+
+static int activate_managed(struct irq_data *irqd)
+{
+ const struct cpumask *dest = irq_data_get_affinity_mask(irqd);
+ int ret;
+
+ cpumask_and(vector_searchmask, dest, cpu_online_mask);
+ if (WARN_ON_ONCE(cpumask_empty(vector_searchmask))) {
+ /* Something in the core code broke! Survive gracefully */
+ pr_err("Managed startup for irq %u, but no CPU\n", irqd->irq);
+ return -EINVAL;
+ }
+
+ ret = assign_managed_vector(irqd, vector_searchmask);
+ /*
+ * This should not happen. The vector reservation got buggered. Handle
+ * it gracefully.
+ */
+ if (WARN_ON_ONCE(ret < 0)) {
+ pr_err("Managed startup irq %u, no vector available\n",
+ irqd->irq);
+ }
+ return ret;
+}
+
+static int x86_vector_activate(struct irq_domain *dom, struct irq_data *irqd,
+ bool reserve)
+{
+ struct apic_chip_data *apicd = apic_chip_data(irqd);
+ unsigned long flags;
+ int ret = 0;
+
+ trace_vector_activate(irqd->irq, apicd->is_managed,
+ apicd->can_reserve, reserve);
+
+ raw_spin_lock_irqsave(&vector_lock, flags);
+ if (!apicd->can_reserve && !apicd->is_managed)
+ assign_irq_vector_any_locked(irqd);
+ else if (reserve || irqd_is_managed_and_shutdown(irqd))
+ vector_assign_managed_shutdown(irqd);
+ else if (apicd->is_managed)
+ ret = activate_managed(irqd);
+ else if (apicd->has_reserved)
+ ret = activate_reserved(irqd);
+ raw_spin_unlock_irqrestore(&vector_lock, flags);
+ return ret;
+}
+
+static void vector_free_reserved_and_managed(struct irq_data *irqd)
+{
+ const struct cpumask *dest = irq_data_get_affinity_mask(irqd);
+ struct apic_chip_data *apicd = apic_chip_data(irqd);
+
+ trace_vector_teardown(irqd->irq, apicd->is_managed,
+ apicd->has_reserved);
+
+ if (apicd->has_reserved)
+ irq_matrix_remove_reserved(vector_matrix);
+ if (apicd->is_managed)
+ irq_matrix_remove_managed(vector_matrix, dest);
+}
+
+static void x86_vector_free_irqs(struct irq_domain *domain,
+ unsigned int virq, unsigned int nr_irqs)
+{
+ struct apic_chip_data *apicd;
+ struct irq_data *irqd;
+ unsigned long flags;
+ int i;
+
+ for (i = 0; i < nr_irqs; i++) {
+ irqd = irq_domain_get_irq_data(x86_vector_domain, virq + i);
+ if (irqd && irqd->chip_data) {
+ raw_spin_lock_irqsave(&vector_lock, flags);
+ clear_irq_vector(irqd);
+ vector_free_reserved_and_managed(irqd);
+ apicd = irqd->chip_data;
+ irq_domain_reset_irq_data(irqd);
+ raw_spin_unlock_irqrestore(&vector_lock, flags);
+ free_apic_chip_data(apicd);
+ }
+ }
+}
+
+static bool vector_configure_legacy(unsigned int virq, struct irq_data *irqd,
+ struct apic_chip_data *apicd)
+{
+ unsigned long flags;
+ bool realloc = false;
+
+ apicd->vector = ISA_IRQ_VECTOR(virq);
+ apicd->cpu = 0;
+
+ raw_spin_lock_irqsave(&vector_lock, flags);
+ /*
+ * If the interrupt is activated, then it must stay at this vector
+ * position. That's usually the timer interrupt (0).
+ */
+ if (irqd_is_activated(irqd)) {
+ trace_vector_setup(virq, true, 0);
+ apic_update_irq_cfg(irqd, apicd->vector, apicd->cpu);
+ } else {
+ /* Release the vector */
+ apicd->can_reserve = true;
+ irqd_set_can_reserve(irqd);
+ clear_irq_vector(irqd);
+ realloc = true;
+ }
+ raw_spin_unlock_irqrestore(&vector_lock, flags);
+ return realloc;
+}
+
+static int x86_vector_alloc_irqs(struct irq_domain *domain, unsigned int virq,
+ unsigned int nr_irqs, void *arg)
+{
+ struct irq_alloc_info *info = arg;
+ struct apic_chip_data *apicd;
+ struct irq_data *irqd;
+ int i, err, node;
+
+ if (apic_is_disabled)
+ return -ENXIO;
+
+ /*
+ * Catch any attempt to touch the cascade interrupt on a PIC
+ * equipped system.
+ */
+ if (WARN_ON_ONCE(info->flags & X86_IRQ_ALLOC_LEGACY &&
+ virq == PIC_CASCADE_IR))
+ return -EINVAL;
+
+ for (i = 0; i < nr_irqs; i++) {
+ irqd = irq_domain_get_irq_data(domain, virq + i);
+ BUG_ON(!irqd);
+ node = irq_data_get_node(irqd);
+ WARN_ON_ONCE(irqd->chip_data);
+ apicd = alloc_apic_chip_data(node);
+ if (!apicd) {
+ err = -ENOMEM;
+ goto error;
+ }
+
+ apicd->irq = virq + i;
+ irqd->chip = &lapic_controller;
+ irqd->chip_data = apicd;
+ irqd->hwirq = virq + i;
+ irqd_set_single_target(irqd);
+ /*
+ * Prevent that any of these interrupts is invoked in
+ * non interrupt context via e.g. generic_handle_irq()
+ * as that can corrupt the affinity move state.
+ */
+ irqd_set_handle_enforce_irqctx(irqd);
+
+ /* Don't invoke affinity setter on deactivated interrupts */
+ irqd_set_affinity_on_activate(irqd);
+
+ /*
+ * Legacy vectors are already assigned when the IOAPIC
+ * takes them over. They stay on the same vector. This is
+ * required for check_timer() to work correctly as it might
+ * switch back to legacy mode. Only update the hardware
+ * config.
+ */
+ if (info->flags & X86_IRQ_ALLOC_LEGACY) {
+ if (!vector_configure_legacy(virq + i, irqd, apicd))
+ continue;
+ }
+
+ err = assign_irq_vector_policy(irqd, info);
+ trace_vector_setup(virq + i, false, err);
+ if (err) {
+ irqd->chip_data = NULL;
+ free_apic_chip_data(apicd);
+ goto error;
+ }
+ }
+
+ return 0;
+
+error:
+ x86_vector_free_irqs(domain, virq, i);
+ return err;
+}
+
+#ifdef CONFIG_GENERIC_IRQ_DEBUGFS
+static void x86_vector_debug_show(struct seq_file *m, struct irq_domain *d,
+ struct irq_data *irqd, int ind)
+{
+ struct apic_chip_data apicd;
+ unsigned long flags;
+ int irq;
+
+ if (!irqd) {
+ irq_matrix_debug_show(m, vector_matrix, ind);
+ return;
+ }
+
+ irq = irqd->irq;
+ if (irq < nr_legacy_irqs() && !test_bit(irq, &io_apic_irqs)) {
+ seq_printf(m, "%*sVector: %5d\n", ind, "", ISA_IRQ_VECTOR(irq));
+ seq_printf(m, "%*sTarget: Legacy PIC all CPUs\n", ind, "");
+ return;
+ }
+
+ if (!irqd->chip_data) {
+ seq_printf(m, "%*sVector: Not assigned\n", ind, "");
+ return;
+ }
+
+ raw_spin_lock_irqsave(&vector_lock, flags);
+ memcpy(&apicd, irqd->chip_data, sizeof(apicd));
+ raw_spin_unlock_irqrestore(&vector_lock, flags);
+
+ seq_printf(m, "%*sVector: %5u\n", ind, "", apicd.vector);
+ seq_printf(m, "%*sTarget: %5u\n", ind, "", apicd.cpu);
+ if (apicd.prev_vector) {
+ seq_printf(m, "%*sPrevious vector: %5u\n", ind, "", apicd.prev_vector);
+ seq_printf(m, "%*sPrevious target: %5u\n", ind, "", apicd.prev_cpu);
+ }
+ seq_printf(m, "%*smove_in_progress: %u\n", ind, "", apicd.move_in_progress ? 1 : 0);
+ seq_printf(m, "%*sis_managed: %u\n", ind, "", apicd.is_managed ? 1 : 0);
+ seq_printf(m, "%*scan_reserve: %u\n", ind, "", apicd.can_reserve ? 1 : 0);
+ seq_printf(m, "%*shas_reserved: %u\n", ind, "", apicd.has_reserved ? 1 : 0);
+ seq_printf(m, "%*scleanup_pending: %u\n", ind, "", !hlist_unhashed(&apicd.clist));
+}
+#endif
+
+int x86_fwspec_is_ioapic(struct irq_fwspec *fwspec)
+{
+ if (fwspec->param_count != 1)
+ return 0;
+
+ if (is_fwnode_irqchip(fwspec->fwnode)) {
+ const char *fwname = fwnode_get_name(fwspec->fwnode);
+ return fwname && !strncmp(fwname, "IO-APIC-", 8) &&
+ simple_strtol(fwname+8, NULL, 10) == fwspec->param[0];
+ }
+ return to_of_node(fwspec->fwnode) &&
+ of_device_is_compatible(to_of_node(fwspec->fwnode),
+ "intel,ce4100-ioapic");
+}
+
+int x86_fwspec_is_hpet(struct irq_fwspec *fwspec)
+{
+ if (fwspec->param_count != 1)
+ return 0;
+
+ if (is_fwnode_irqchip(fwspec->fwnode)) {
+ const char *fwname = fwnode_get_name(fwspec->fwnode);
+ return fwname && !strncmp(fwname, "HPET-MSI-", 9) &&
+ simple_strtol(fwname+9, NULL, 10) == fwspec->param[0];
+ }
+ return 0;
+}
+
+static int x86_vector_select(struct irq_domain *d, struct irq_fwspec *fwspec,
+ enum irq_domain_bus_token bus_token)
+{
+ /*
+ * HPET and I/OAPIC cannot be parented in the vector domain
+ * if IRQ remapping is enabled. APIC IDs above 15 bits are
+ * only permitted if IRQ remapping is enabled, so check that.
+ */
+ if (apic_id_valid(32768))
+ return 0;
+
+ return x86_fwspec_is_ioapic(fwspec) || x86_fwspec_is_hpet(fwspec);
+}
+
+static const struct irq_domain_ops x86_vector_domain_ops = {
+ .select = x86_vector_select,
+ .alloc = x86_vector_alloc_irqs,
+ .free = x86_vector_free_irqs,
+ .activate = x86_vector_activate,
+ .deactivate = x86_vector_deactivate,
+#ifdef CONFIG_GENERIC_IRQ_DEBUGFS
+ .debug_show = x86_vector_debug_show,
+#endif
+};
+
+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 = (gsi_top + nr_legacy_irqs()) + 8 * nr_cpu_ids;
+#if defined(CONFIG_PCI_MSI)
+ /*
+ * for MSI and HT dyn irq
+ */
+ if (gsi_top <= NR_IRQS_LEGACY)
+ nr += 8 * nr_cpu_ids;
+ else
+ nr += gsi_top * 16;
+#endif
+ if (nr < nr_irqs)
+ nr_irqs = nr;
+
+ /*
+ * We don't know if PIC is present at this point so we need to do
+ * probe() to get the right number of legacy IRQs.
+ */
+ return legacy_pic->probe();
+}
+
+void lapic_assign_legacy_vector(unsigned int irq, bool replace)
+{
+ /*
+ * Use assign system here so it wont get accounted as allocated
+ * and moveable in the cpu hotplug check and it prevents managed
+ * irq reservation from touching it.
+ */
+ irq_matrix_assign_system(vector_matrix, ISA_IRQ_VECTOR(irq), replace);
+}
+
+void __init lapic_update_legacy_vectors(void)
+{
+ unsigned int i;
+
+ if (IS_ENABLED(CONFIG_X86_IO_APIC) && nr_ioapics > 0)
+ return;
+
+ /*
+ * If the IO/APIC is disabled via config, kernel command line or
+ * lack of enumeration then all legacy interrupts are routed
+ * through the PIC. Make sure that they are marked as legacy
+ * vectors. PIC_CASCADE_IRQ has already been marked in
+ * lapic_assign_system_vectors().
+ */
+ for (i = 0; i < nr_legacy_irqs(); i++) {
+ if (i != PIC_CASCADE_IR)
+ lapic_assign_legacy_vector(i, true);
+ }
+}
+
+void __init lapic_assign_system_vectors(void)
+{
+ unsigned int i, vector;
+
+ for_each_set_bit(vector, system_vectors, NR_VECTORS)
+ irq_matrix_assign_system(vector_matrix, vector, false);
+
+ if (nr_legacy_irqs() > 1)
+ lapic_assign_legacy_vector(PIC_CASCADE_IR, false);
+
+ /* System vectors are reserved, online it */
+ irq_matrix_online(vector_matrix);
+
+ /* Mark the preallocated legacy interrupts */
+ for (i = 0; i < nr_legacy_irqs(); i++) {
+ /*
+ * Don't touch the cascade interrupt. It's unusable
+ * on PIC equipped machines. See the large comment
+ * in the IO/APIC code.
+ */
+ if (i != PIC_CASCADE_IR)
+ irq_matrix_assign(vector_matrix, ISA_IRQ_VECTOR(i));
+ }
+}
+
+int __init arch_early_irq_init(void)
+{
+ struct fwnode_handle *fn;
+
+ fn = irq_domain_alloc_named_fwnode("VECTOR");
+ BUG_ON(!fn);
+ x86_vector_domain = irq_domain_create_tree(fn, &x86_vector_domain_ops,
+ NULL);
+ BUG_ON(x86_vector_domain == NULL);
+ irq_set_default_host(x86_vector_domain);
+
+ BUG_ON(!alloc_cpumask_var(&vector_searchmask, GFP_KERNEL));
+
+ /*
+ * Allocate the vector matrix allocator data structure and limit the
+ * search area.
+ */
+ vector_matrix = irq_alloc_matrix(NR_VECTORS, FIRST_EXTERNAL_VECTOR,
+ FIRST_SYSTEM_VECTOR);
+ BUG_ON(!vector_matrix);
+
+ return arch_early_ioapic_init();
+}
+
+#ifdef CONFIG_SMP
+
+static struct irq_desc *__setup_vector_irq(int vector)
+{
+ int isairq = vector - ISA_IRQ_VECTOR(0);
+
+ /* Check whether the irq is in the legacy space */
+ if (isairq < 0 || isairq >= nr_legacy_irqs())
+ return VECTOR_UNUSED;
+ /* Check whether the irq is handled by the IOAPIC */
+ if (test_bit(isairq, &io_apic_irqs))
+ return VECTOR_UNUSED;
+ return irq_to_desc(isairq);
+}
+
+/* Online the local APIC infrastructure and initialize the vectors */
+void lapic_online(void)
+{
+ unsigned int vector;
+
+ lockdep_assert_held(&vector_lock);
+
+ /* Online the vector matrix array for this CPU */
+ irq_matrix_online(vector_matrix);
+
+ /*
+ * The interrupt affinity logic never targets interrupts to offline
+ * CPUs. The exception are the legacy PIC interrupts. In general
+ * they are only targeted to CPU0, but depending on the platform
+ * they can be distributed to any online CPU in hardware. The
+ * kernel has no influence on that. So all active legacy vectors
+ * must be installed on all CPUs. All non legacy interrupts can be
+ * cleared.
+ */
+ for (vector = 0; vector < NR_VECTORS; vector++)
+ this_cpu_write(vector_irq[vector], __setup_vector_irq(vector));
+}
+
+static void __vector_cleanup(struct vector_cleanup *cl, bool check_irr);
+
+void lapic_offline(void)
+{
+ struct vector_cleanup *cl = this_cpu_ptr(&vector_cleanup);
+
+ lock_vector_lock();
+
+ /* In case the vector cleanup timer has not expired */
+ __vector_cleanup(cl, false);
+
+ irq_matrix_offline(vector_matrix);
+ WARN_ON_ONCE(try_to_del_timer_sync(&cl->timer) < 0);
+ WARN_ON_ONCE(!hlist_empty(&cl->head));
+
+ unlock_vector_lock();
+}
+
+static int apic_set_affinity(struct irq_data *irqd,
+ const struct cpumask *dest, bool force)
+{
+ int err;
+
+ if (WARN_ON_ONCE(!irqd_is_activated(irqd)))
+ return -EIO;
+
+ raw_spin_lock(&vector_lock);
+ cpumask_and(vector_searchmask, dest, cpu_online_mask);
+ if (irqd_affinity_is_managed(irqd))
+ err = assign_managed_vector(irqd, vector_searchmask);
+ else
+ err = assign_vector_locked(irqd, vector_searchmask);
+ raw_spin_unlock(&vector_lock);
+ return err ? err : IRQ_SET_MASK_OK;
+}
+
+#else
+# define apic_set_affinity NULL
+#endif
+
+static int apic_retrigger_irq(struct irq_data *irqd)
+{
+ struct apic_chip_data *apicd = apic_chip_data(irqd);
+ unsigned long flags;
+
+ raw_spin_lock_irqsave(&vector_lock, flags);
+ __apic_send_IPI(apicd->cpu, apicd->vector);
+ raw_spin_unlock_irqrestore(&vector_lock, flags);
+
+ return 1;
+}
+
+void apic_ack_irq(struct irq_data *irqd)
+{
+ irq_move_irq(irqd);
+ apic_eoi();
+}
+
+void apic_ack_edge(struct irq_data *irqd)
+{
+ irq_complete_move(irqd_cfg(irqd));
+ apic_ack_irq(irqd);
+}
+
+static void x86_vector_msi_compose_msg(struct irq_data *data,
+ struct msi_msg *msg)
+{
+ __irq_msi_compose_msg(irqd_cfg(data), msg, false);
+}
+
+static struct irq_chip lapic_controller = {
+ .name = "APIC",
+ .irq_ack = apic_ack_edge,
+ .irq_set_affinity = apic_set_affinity,
+ .irq_compose_msi_msg = x86_vector_msi_compose_msg,
+ .irq_retrigger = apic_retrigger_irq,
+};
+
+#ifdef CONFIG_SMP
+
+static void free_moved_vector(struct apic_chip_data *apicd)
+{
+ unsigned int vector = apicd->prev_vector;
+ unsigned int cpu = apicd->prev_cpu;
+ bool managed = apicd->is_managed;
+
+ /*
+ * Managed interrupts are usually not migrated away
+ * from an online CPU, but CPU isolation 'managed_irq'
+ * can make that happen.
+ * 1) Activation does not take the isolation into account
+ * to keep the code simple
+ * 2) Migration away from an isolated CPU can happen when
+ * a non-isolated CPU which is in the calculated
+ * affinity mask comes online.
+ */
+ trace_vector_free_moved(apicd->irq, cpu, vector, managed);
+ irq_matrix_free(vector_matrix, cpu, vector, managed);
+ per_cpu(vector_irq, cpu)[vector] = VECTOR_UNUSED;
+ hlist_del_init(&apicd->clist);
+ apicd->prev_vector = 0;
+ apicd->move_in_progress = 0;
+}
+
+static void __vector_cleanup(struct vector_cleanup *cl, bool check_irr)
+{
+ struct apic_chip_data *apicd;
+ struct hlist_node *tmp;
+ bool rearm = false;
+
+ lockdep_assert_held(&vector_lock);
+
+ hlist_for_each_entry_safe(apicd, tmp, &cl->head, clist) {
+ unsigned int irr, vector = apicd->prev_vector;
+
+ /*
+ * Paranoia: Check if the vector that needs to be cleaned
+ * up is registered at the APICs IRR. That's clearly a
+ * hardware issue if the vector arrived on the old target
+ * _after_ interrupts were disabled above. Keep @apicd
+ * on the list and schedule the timer again to give the CPU
+ * a chance to handle the pending interrupt.
+ *
+ * Do not check IRR when called from lapic_offline(), because
+ * fixup_irqs() was just called to scan IRR for set bits and
+ * forward them to new destination CPUs via IPIs.
+ */
+ irr = check_irr ? apic_read(APIC_IRR + (vector / 32 * 0x10)) : 0;
+ if (irr & (1U << (vector % 32))) {
+ pr_warn_once("Moved interrupt pending in old target APIC %u\n", apicd->irq);
+ rearm = true;
+ continue;
+ }
+ free_moved_vector(apicd);
+ }
+
+ /*
+ * Must happen under vector_lock to make the timer_pending() check
+ * in __vector_schedule_cleanup() race free against the rearm here.
+ */
+ if (rearm)
+ mod_timer(&cl->timer, jiffies + 1);
+}
+
+static void vector_cleanup_callback(struct timer_list *tmr)
+{
+ struct vector_cleanup *cl = container_of(tmr, typeof(*cl), timer);
+
+ /* Prevent vectors vanishing under us */
+ raw_spin_lock_irq(&vector_lock);
+ __vector_cleanup(cl, true);
+ raw_spin_unlock_irq(&vector_lock);
+}
+
+static void __vector_schedule_cleanup(struct apic_chip_data *apicd)
+{
+ unsigned int cpu = apicd->prev_cpu;
+
+ raw_spin_lock(&vector_lock);
+ apicd->move_in_progress = 0;
+ if (cpu_online(cpu)) {
+ struct vector_cleanup *cl = per_cpu_ptr(&vector_cleanup, cpu);
+
+ hlist_add_head(&apicd->clist, &cl->head);
+
+ /*
+ * The lockless timer_pending() check is safe here. If it
+ * returns true, then the callback will observe this new
+ * apic data in the hlist as everything is serialized by
+ * vector lock.
+ *
+ * If it returns false then the timer is either not armed
+ * or the other CPU executes the callback, which again
+ * would be blocked on vector lock. Rearming it in the
+ * latter case makes it fire for nothing.
+ *
+ * This is also safe against the callback rearming the timer
+ * because that's serialized via vector lock too.
+ */
+ if (!timer_pending(&cl->timer)) {
+ cl->timer.expires = jiffies + 1;
+ add_timer_on(&cl->timer, cpu);
+ }
+ } else {
+ apicd->prev_vector = 0;
+ }
+ raw_spin_unlock(&vector_lock);
+}
+
+void vector_schedule_cleanup(struct irq_cfg *cfg)
+{
+ struct apic_chip_data *apicd;
+
+ apicd = container_of(cfg, struct apic_chip_data, hw_irq_cfg);
+ if (apicd->move_in_progress)
+ __vector_schedule_cleanup(apicd);
+}
+
+void irq_complete_move(struct irq_cfg *cfg)
+{
+ struct apic_chip_data *apicd;
+
+ apicd = container_of(cfg, struct apic_chip_data, hw_irq_cfg);
+ if (likely(!apicd->move_in_progress))
+ return;
+
+ /*
+ * If the interrupt arrived on the new target CPU, cleanup the
+ * vector on the old target CPU. A vector check is not required
+ * because an interrupt can never move from one vector to another
+ * on the same CPU.
+ */
+ if (apicd->cpu == smp_processor_id())
+ __vector_schedule_cleanup(apicd);
+}
+
+/*
+ * Called from fixup_irqs() with @desc->lock held and interrupts disabled.
+ */
+void irq_force_complete_move(struct irq_desc *desc)
+{
+ struct apic_chip_data *apicd;
+ struct irq_data *irqd;
+ unsigned int vector;
+
+ /*
+ * The function is called for all descriptors regardless of which
+ * irqdomain they belong to. For example if an IRQ is provided by
+ * an irq_chip as part of a GPIO driver, the chip data for that
+ * descriptor is specific to the irq_chip in question.
+ *
+ * Check first that the chip_data is what we expect
+ * (apic_chip_data) before touching it any further.
+ */
+ irqd = irq_domain_get_irq_data(x86_vector_domain,
+ irq_desc_get_irq(desc));
+ if (!irqd)
+ return;
+
+ raw_spin_lock(&vector_lock);
+ apicd = apic_chip_data(irqd);
+ if (!apicd)
+ goto unlock;
+
+ /*
+ * If prev_vector is empty, no action required.
+ */
+ vector = apicd->prev_vector;
+ if (!vector)
+ goto unlock;
+
+ /*
+ * This is tricky. If the cleanup of the old vector has not been
+ * done yet, then the following setaffinity call will fail with
+ * -EBUSY. This can leave the interrupt in a stale state.
+ *
+ * All CPUs are stuck in stop machine with interrupts disabled so
+ * calling __irq_complete_move() would be completely pointless.
+ *
+ * 1) The interrupt is in move_in_progress state. That means that we
+ * have not seen an interrupt since the io_apic was reprogrammed to
+ * the new vector.
+ *
+ * 2) The interrupt has fired on the new vector, but the cleanup IPIs
+ * have not been processed yet.
+ */
+ if (apicd->move_in_progress) {
+ /*
+ * In theory there is a race:
+ *
+ * set_ioapic(new_vector) <-- Interrupt is raised before update
+ * is effective, i.e. it's raised on
+ * the old vector.
+ *
+ * So if the target cpu cannot handle that interrupt before
+ * the old vector is cleaned up, we get a spurious interrupt
+ * and in the worst case the ioapic irq line becomes stale.
+ *
+ * But in case of cpu hotplug this should be a non issue
+ * because if the affinity update happens right before all
+ * cpus rendezvous in stop machine, there is no way that the
+ * interrupt can be blocked on the target cpu because all cpus
+ * loops first with interrupts enabled in stop machine, so the
+ * old vector is not yet cleaned up when the interrupt fires.
+ *
+ * So the only way to run into this issue is if the delivery
+ * of the interrupt on the apic/system bus would be delayed
+ * beyond the point where the target cpu disables interrupts
+ * in stop machine. I doubt that it can happen, but at least
+ * there is a theoretical chance. Virtualization might be
+ * able to expose this, but AFAICT the IOAPIC emulation is not
+ * as stupid as the real hardware.
+ *
+ * Anyway, there is nothing we can do about that at this point
+ * w/o refactoring the whole fixup_irq() business completely.
+ * We print at least the irq number and the old vector number,
+ * so we have the necessary information when a problem in that
+ * area arises.
+ */
+ pr_warn("IRQ fixup: irq %d move in progress, old vector %d\n",
+ irqd->irq, vector);
+ }
+ free_moved_vector(apicd);
+unlock:
+ raw_spin_unlock(&vector_lock);
+}
+
+#ifdef CONFIG_HOTPLUG_CPU
+/*
+ * Note, this is not accurate accounting, but at least good enough to
+ * prevent that the actual interrupt move will run out of vectors.
+ */
+int lapic_can_unplug_cpu(void)
+{
+ unsigned int rsvd, avl, tomove, cpu = smp_processor_id();
+ int ret = 0;
+
+ raw_spin_lock(&vector_lock);
+ tomove = irq_matrix_allocated(vector_matrix);
+ avl = irq_matrix_available(vector_matrix, true);
+ if (avl < tomove) {
+ pr_warn("CPU %u has %u vectors, %u available. Cannot disable CPU\n",
+ cpu, tomove, avl);
+ ret = -ENOSPC;
+ goto out;
+ }
+ rsvd = irq_matrix_reserved(vector_matrix);
+ if (avl < rsvd) {
+ pr_warn("Reserved vectors %u > available %u. IRQ request may fail\n",
+ rsvd, avl);
+ }
+out:
+ raw_spin_unlock(&vector_lock);
+ return ret;
+}
+#endif /* HOTPLUG_CPU */
+#endif /* SMP */
+
+static void __init print_APIC_field(int base)
+{
+ int i;
+
+ printk(KERN_DEBUG);
+
+ for (i = 0; i < 8; i++)
+ pr_cont("%08x", apic_read(base + i*0x10));
+
+ pr_cont("\n");
+}
+
+static void __init print_local_APIC(void *dummy)
+{
+ unsigned int i, v, ver, maxlvt;
+ u64 icr;
+
+ pr_debug("printing local APIC contents on CPU#%d/%d:\n",
+ smp_processor_id(), read_apic_id());
+ v = apic_read(APIC_ID);
+ pr_info("... APIC ID: %08x (%01x)\n", v, read_apic_id());
+ v = apic_read(APIC_LVR);
+ pr_info("... APIC VERSION: %08x\n", v);
+ ver = GET_APIC_VERSION(v);
+ maxlvt = lapic_get_maxlvt();
+
+ v = apic_read(APIC_TASKPRI);
+ pr_debug("... APIC TASKPRI: %08x (%02x)\n", v, v & APIC_TPRI_MASK);
+
+ /* !82489DX */
+ if (APIC_INTEGRATED(ver)) {
+ if (!APIC_XAPIC(ver)) {
+ v = apic_read(APIC_ARBPRI);
+ pr_debug("... APIC ARBPRI: %08x (%02x)\n",
+ v, v & APIC_ARBPRI_MASK);
+ }
+ v = apic_read(APIC_PROCPRI);
+ pr_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);
+ pr_debug("... APIC RRR: %08x\n", v);
+ }
+
+ v = apic_read(APIC_LDR);
+ pr_debug("... APIC LDR: %08x\n", v);
+ if (!x2apic_enabled()) {
+ v = apic_read(APIC_DFR);
+ pr_debug("... APIC DFR: %08x\n", v);
+ }
+ v = apic_read(APIC_SPIV);
+ pr_debug("... APIC SPIV: %08x\n", v);
+
+ pr_debug("... APIC ISR field:\n");
+ print_APIC_field(APIC_ISR);
+ pr_debug("... APIC TMR field:\n");
+ print_APIC_field(APIC_TMR);
+ pr_debug("... APIC IRR field:\n");
+ print_APIC_field(APIC_IRR);
+
+ /* !82489DX */
+ if (APIC_INTEGRATED(ver)) {
+ /* Due to the Pentium erratum 3AP. */
+ if (maxlvt > 3)
+ apic_write(APIC_ESR, 0);
+
+ v = apic_read(APIC_ESR);
+ pr_debug("... APIC ESR: %08x\n", v);
+ }
+
+ icr = apic_icr_read();
+ pr_debug("... APIC ICR: %08x\n", (u32)icr);
+ pr_debug("... APIC ICR2: %08x\n", (u32)(icr >> 32));
+
+ v = apic_read(APIC_LVTT);
+ pr_debug("... APIC LVTT: %08x\n", v);
+
+ if (maxlvt > 3) {
+ /* PC is LVT#4. */
+ v = apic_read(APIC_LVTPC);
+ pr_debug("... APIC LVTPC: %08x\n", v);
+ }
+ v = apic_read(APIC_LVT0);
+ pr_debug("... APIC LVT0: %08x\n", v);
+ v = apic_read(APIC_LVT1);
+ pr_debug("... APIC LVT1: %08x\n", v);
+
+ if (maxlvt > 2) {
+ /* ERR is LVT#3. */
+ v = apic_read(APIC_LVTERR);
+ pr_debug("... APIC LVTERR: %08x\n", v);
+ }
+
+ v = apic_read(APIC_TMICT);
+ pr_debug("... APIC TMICT: %08x\n", v);
+ v = apic_read(APIC_TMCCT);
+ pr_debug("... APIC TMCCT: %08x\n", v);
+ v = apic_read(APIC_TDCR);
+ pr_debug("... APIC TDCR: %08x\n", v);
+
+ if (boot_cpu_has(X86_FEATURE_EXTAPIC)) {
+ v = apic_read(APIC_EFEAT);
+ maxlvt = (v >> 16) & 0xff;
+ pr_debug("... APIC EFEAT: %08x\n", v);
+ v = apic_read(APIC_ECTRL);
+ pr_debug("... APIC ECTRL: %08x\n", v);
+ for (i = 0; i < maxlvt; i++) {
+ v = apic_read(APIC_EILVTn(i));
+ pr_debug("... APIC EILVT%d: %08x\n", i, v);
+ }
+ }
+ pr_cont("\n");
+}
+
+static void __init print_local_APICs(int maxcpu)
+{
+ int cpu;
+
+ if (!maxcpu)
+ return;
+
+ preempt_disable();
+ for_each_online_cpu(cpu) {
+ if (cpu >= maxcpu)
+ break;
+ smp_call_function_single(cpu, print_local_APIC, NULL, 1);
+ }
+ preempt_enable();
+}
+
+static void __init print_PIC(void)
+{
+ unsigned int v;
+ unsigned long flags;
+
+ if (!nr_legacy_irqs())
+ return;
+
+ pr_debug("\nprinting PIC contents\n");
+
+ raw_spin_lock_irqsave(&i8259A_lock, flags);
+
+ v = inb(0xa1) << 8 | inb(0x21);
+ pr_debug("... PIC IMR: %04x\n", v);
+
+ v = inb(0xa0) << 8 | inb(0x20);
+ pr_debug("... PIC IRR: %04x\n", v);
+
+ outb(0x0b, 0xa0);
+ outb(0x0b, 0x20);
+ v = inb(0xa0) << 8 | inb(0x20);
+ outb(0x0a, 0xa0);
+ outb(0x0a, 0x20);
+
+ raw_spin_unlock_irqrestore(&i8259A_lock, flags);
+
+ pr_debug("... PIC ISR: %04x\n", v);
+
+ v = inb(PIC_ELCR2) << 8 | inb(PIC_ELCR1);
+ pr_debug("... PIC ELCR: %04x\n", v);
+}
+
+static int show_lapic __initdata = 1;
+static __init int setup_show_lapic(char *arg)
+{
+ int num = -1;
+
+ if (strcmp(arg, "all") == 0) {
+ show_lapic = CONFIG_NR_CPUS;
+ } else {
+ get_option(&arg, &num);
+ if (num >= 0)
+ show_lapic = num;
+ }
+
+ return 1;
+}
+__setup("show_lapic=", setup_show_lapic);
+
+static int __init print_ICs(void)
+{
+ if (apic_verbosity == APIC_QUIET)
+ return 0;
+
+ print_PIC();
+
+ /* don't print out if apic is not there */
+ if (!boot_cpu_has(X86_FEATURE_APIC) && !apic_from_smp_config())
+ return 0;
+
+ print_local_APICs(show_lapic);
+ print_IO_APICs();
+
+ return 0;
+}
+
+late_initcall(print_ICs);
diff --git a/arch/x86/kernel/apic/x2apic_cluster.c b/arch/x86/kernel/apic/x2apic_cluster.c
new file mode 100644
index 0000000000..affbff65e4
--- /dev/null
+++ b/arch/x86/kernel/apic/x2apic_cluster.c
@@ -0,0 +1,262 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <linux/cpuhotplug.h>
+#include <linux/cpumask.h>
+#include <linux/slab.h>
+#include <linux/mm.h>
+
+#include <asm/apic.h>
+
+#include "local.h"
+
+#define apic_cluster(apicid) ((apicid) >> 4)
+
+/*
+ * __x2apic_send_IPI_mask() possibly needs to read
+ * x86_cpu_to_logical_apicid for all online cpus in a sequential way.
+ * Using per cpu variable would cost one cache line per cpu.
+ */
+static u32 *x86_cpu_to_logical_apicid __read_mostly;
+
+static DEFINE_PER_CPU(cpumask_var_t, ipi_mask);
+static DEFINE_PER_CPU_READ_MOSTLY(struct cpumask *, cluster_masks);
+
+static int x2apic_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
+{
+ return x2apic_enabled();
+}
+
+static void x2apic_send_IPI(int cpu, int vector)
+{
+ u32 dest = x86_cpu_to_logical_apicid[cpu];
+
+ /* x2apic MSRs are special and need a special fence: */
+ weak_wrmsr_fence();
+ __x2apic_send_IPI_dest(dest, vector, APIC_DEST_LOGICAL);
+}
+
+static void
+__x2apic_send_IPI_mask(const struct cpumask *mask, int vector, int apic_dest)
+{
+ unsigned int cpu, clustercpu;
+ struct cpumask *tmpmsk;
+ unsigned long flags;
+ u32 dest;
+
+ /* x2apic MSRs are special and need a special fence: */
+ weak_wrmsr_fence();
+ local_irq_save(flags);
+
+ tmpmsk = this_cpu_cpumask_var_ptr(ipi_mask);
+ cpumask_copy(tmpmsk, mask);
+ /* If IPI should not be sent to self, clear current CPU */
+ if (apic_dest != APIC_DEST_ALLINC)
+ __cpumask_clear_cpu(smp_processor_id(), tmpmsk);
+
+ /* Collapse cpus in a cluster so a single IPI per cluster is sent */
+ for_each_cpu(cpu, tmpmsk) {
+ struct cpumask *cmsk = per_cpu(cluster_masks, cpu);
+
+ dest = 0;
+ for_each_cpu_and(clustercpu, tmpmsk, cmsk)
+ dest |= x86_cpu_to_logical_apicid[clustercpu];
+
+ if (!dest)
+ continue;
+
+ __x2apic_send_IPI_dest(dest, vector, APIC_DEST_LOGICAL);
+ /* Remove cluster CPUs from tmpmask */
+ cpumask_andnot(tmpmsk, tmpmsk, cmsk);
+ }
+
+ local_irq_restore(flags);
+}
+
+static void x2apic_send_IPI_mask(const struct cpumask *mask, int vector)
+{
+ __x2apic_send_IPI_mask(mask, vector, APIC_DEST_ALLINC);
+}
+
+static void
+x2apic_send_IPI_mask_allbutself(const struct cpumask *mask, int vector)
+{
+ __x2apic_send_IPI_mask(mask, vector, APIC_DEST_ALLBUT);
+}
+
+static u32 x2apic_calc_apicid(unsigned int cpu)
+{
+ return x86_cpu_to_logical_apicid[cpu];
+}
+
+static void init_x2apic_ldr(void)
+{
+ struct cpumask *cmsk = this_cpu_read(cluster_masks);
+
+ BUG_ON(!cmsk);
+
+ cpumask_set_cpu(smp_processor_id(), cmsk);
+}
+
+/*
+ * As an optimisation during boot, set the cluster_mask for all present
+ * CPUs at once, to prevent each of them having to iterate over the others
+ * to find the existing cluster_mask.
+ */
+static void prefill_clustermask(struct cpumask *cmsk, unsigned int cpu, u32 cluster)
+{
+ int cpu_i;
+
+ for_each_present_cpu(cpu_i) {
+ struct cpumask **cpu_cmsk = &per_cpu(cluster_masks, cpu_i);
+ u32 apicid = apic->cpu_present_to_apicid(cpu_i);
+
+ if (apicid == BAD_APICID || cpu_i == cpu || apic_cluster(apicid) != cluster)
+ continue;
+
+ if (WARN_ON_ONCE(*cpu_cmsk == cmsk))
+ continue;
+
+ BUG_ON(*cpu_cmsk);
+ *cpu_cmsk = cmsk;
+ }
+}
+
+static int alloc_clustermask(unsigned int cpu, u32 cluster, int node)
+{
+ struct cpumask *cmsk = NULL;
+ unsigned int cpu_i;
+
+ /*
+ * At boot time, the CPU present mask is stable. The cluster mask is
+ * allocated for the first CPU in the cluster and propagated to all
+ * present siblings in the cluster. If the cluster mask is already set
+ * on entry to this function for a given CPU, there is nothing to do.
+ */
+ if (per_cpu(cluster_masks, cpu))
+ return 0;
+
+ if (system_state < SYSTEM_RUNNING)
+ goto alloc;
+
+ /*
+ * On post boot hotplug for a CPU which was not present at boot time,
+ * iterate over all possible CPUs (even those which are not present
+ * any more) to find any existing cluster mask.
+ */
+ for_each_possible_cpu(cpu_i) {
+ u32 apicid = apic->cpu_present_to_apicid(cpu_i);
+
+ if (apicid != BAD_APICID && apic_cluster(apicid) == cluster) {
+ cmsk = per_cpu(cluster_masks, cpu_i);
+ /*
+ * If the cluster is already initialized, just store
+ * the mask and return. There's no need to propagate.
+ */
+ if (cmsk) {
+ per_cpu(cluster_masks, cpu) = cmsk;
+ return 0;
+ }
+ }
+ }
+ /*
+ * No CPU in the cluster has ever been initialized, so fall through to
+ * the boot time code which will also populate the cluster mask for any
+ * other CPU in the cluster which is (now) present.
+ */
+alloc:
+ cmsk = kzalloc_node(sizeof(*cmsk), GFP_KERNEL, node);
+ if (!cmsk)
+ return -ENOMEM;
+ per_cpu(cluster_masks, cpu) = cmsk;
+ prefill_clustermask(cmsk, cpu, cluster);
+
+ return 0;
+}
+
+static int x2apic_prepare_cpu(unsigned int cpu)
+{
+ u32 phys_apicid = apic->cpu_present_to_apicid(cpu);
+ u32 cluster = apic_cluster(phys_apicid);
+ u32 logical_apicid = (cluster << 16) | (1 << (phys_apicid & 0xf));
+
+ x86_cpu_to_logical_apicid[cpu] = logical_apicid;
+
+ if (alloc_clustermask(cpu, cluster, cpu_to_node(cpu)) < 0)
+ return -ENOMEM;
+ if (!zalloc_cpumask_var(&per_cpu(ipi_mask, cpu), GFP_KERNEL))
+ return -ENOMEM;
+ return 0;
+}
+
+static int x2apic_dead_cpu(unsigned int dead_cpu)
+{
+ struct cpumask *cmsk = per_cpu(cluster_masks, dead_cpu);
+
+ if (cmsk)
+ cpumask_clear_cpu(dead_cpu, cmsk);
+ free_cpumask_var(per_cpu(ipi_mask, dead_cpu));
+ return 0;
+}
+
+static int x2apic_cluster_probe(void)
+{
+ u32 slots;
+
+ if (!x2apic_mode)
+ return 0;
+
+ slots = max_t(u32, L1_CACHE_BYTES/sizeof(u32), nr_cpu_ids);
+ x86_cpu_to_logical_apicid = kcalloc(slots, sizeof(u32), GFP_KERNEL);
+ if (!x86_cpu_to_logical_apicid)
+ return 0;
+
+ if (cpuhp_setup_state(CPUHP_X2APIC_PREPARE, "x86/x2apic:prepare",
+ x2apic_prepare_cpu, x2apic_dead_cpu) < 0) {
+ pr_err("Failed to register X2APIC_PREPARE\n");
+ kfree(x86_cpu_to_logical_apicid);
+ x86_cpu_to_logical_apicid = NULL;
+ return 0;
+ }
+ init_x2apic_ldr();
+ return 1;
+}
+
+static struct apic apic_x2apic_cluster __ro_after_init = {
+
+ .name = "cluster x2apic",
+ .probe = x2apic_cluster_probe,
+ .acpi_madt_oem_check = x2apic_acpi_madt_oem_check,
+
+ .delivery_mode = APIC_DELIVERY_MODE_FIXED,
+ .dest_mode_logical = true,
+
+ .disable_esr = 0,
+
+ .check_apicid_used = NULL,
+ .init_apic_ldr = init_x2apic_ldr,
+ .ioapic_phys_id_map = NULL,
+ .cpu_present_to_apicid = default_cpu_present_to_apicid,
+ .phys_pkg_id = x2apic_phys_pkg_id,
+
+ .max_apic_id = UINT_MAX,
+ .x2apic_set_max_apicid = true,
+ .get_apic_id = x2apic_get_apic_id,
+ .set_apic_id = x2apic_set_apic_id,
+
+ .calc_dest_apicid = x2apic_calc_apicid,
+
+ .send_IPI = x2apic_send_IPI,
+ .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,
+
+ .read = native_apic_msr_read,
+ .write = native_apic_msr_write,
+ .eoi = native_apic_msr_eoi,
+ .icr_read = native_x2apic_icr_read,
+ .icr_write = native_x2apic_icr_write,
+};
+
+apic_driver(apic_x2apic_cluster);
diff --git a/arch/x86/kernel/apic/x2apic_phys.c b/arch/x86/kernel/apic/x2apic_phys.c
new file mode 100644
index 0000000000..788cdb4ee3
--- /dev/null
+++ b/arch/x86/kernel/apic/x2apic_phys.c
@@ -0,0 +1,177 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <linux/cpumask.h>
+#include <linux/acpi.h>
+
+#include "local.h"
+
+int x2apic_phys;
+
+static struct apic apic_x2apic_phys;
+u32 x2apic_max_apicid __ro_after_init = UINT_MAX;
+
+void __init x2apic_set_max_apicid(u32 apicid)
+{
+ x2apic_max_apicid = apicid;
+ if (apic->x2apic_set_max_apicid)
+ apic->max_apic_id = apicid;
+}
+
+static int __init set_x2apic_phys_mode(char *arg)
+{
+ x2apic_phys = 1;
+ return 0;
+}
+early_param("x2apic_phys", set_x2apic_phys_mode);
+
+static bool x2apic_fadt_phys(void)
+{
+#ifdef CONFIG_ACPI
+ if ((acpi_gbl_FADT.header.revision >= FADT2_REVISION_ID) &&
+ (acpi_gbl_FADT.flags & ACPI_FADT_APIC_PHYSICAL)) {
+ printk(KERN_DEBUG "System requires x2apic physical mode\n");
+ return true;
+ }
+#endif
+ return false;
+}
+
+static int x2apic_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
+{
+ return x2apic_enabled() && (x2apic_phys || x2apic_fadt_phys());
+}
+
+static void x2apic_send_IPI(int cpu, int vector)
+{
+ u32 dest = per_cpu(x86_cpu_to_apicid, cpu);
+
+ /* x2apic MSRs are special and need a special fence: */
+ weak_wrmsr_fence();
+ __x2apic_send_IPI_dest(dest, vector, APIC_DEST_PHYSICAL);
+}
+
+static void
+__x2apic_send_IPI_mask(const struct cpumask *mask, int vector, int apic_dest)
+{
+ unsigned long query_cpu;
+ unsigned long this_cpu;
+ unsigned long flags;
+
+ /* x2apic MSRs are special and need a special fence: */
+ weak_wrmsr_fence();
+
+ local_irq_save(flags);
+
+ this_cpu = smp_processor_id();
+ for_each_cpu(query_cpu, mask) {
+ if (apic_dest == APIC_DEST_ALLBUT && this_cpu == query_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_mask(const struct cpumask *mask, int vector)
+{
+ __x2apic_send_IPI_mask(mask, vector, APIC_DEST_ALLINC);
+}
+
+static void
+ x2apic_send_IPI_mask_allbutself(const struct cpumask *mask, int vector)
+{
+ __x2apic_send_IPI_mask(mask, vector, APIC_DEST_ALLBUT);
+}
+
+static void __x2apic_send_IPI_shorthand(int vector, u32 which)
+{
+ unsigned long cfg = __prepare_ICR(which, vector, 0);
+
+ /* x2apic MSRs are special and need a special fence: */
+ weak_wrmsr_fence();
+ native_x2apic_icr_write(cfg, 0);
+}
+
+void x2apic_send_IPI_allbutself(int vector)
+{
+ __x2apic_send_IPI_shorthand(vector, APIC_DEST_ALLBUT);
+}
+
+void x2apic_send_IPI_all(int vector)
+{
+ __x2apic_send_IPI_shorthand(vector, APIC_DEST_ALLINC);
+}
+
+void x2apic_send_IPI_self(int vector)
+{
+ apic_write(APIC_SELF_IPI, vector);
+}
+
+void __x2apic_send_IPI_dest(unsigned int apicid, int vector, unsigned int dest)
+{
+ unsigned long cfg = __prepare_ICR(0, vector, dest);
+ native_x2apic_icr_write(cfg, apicid);
+}
+
+static int x2apic_phys_probe(void)
+{
+ if (!x2apic_mode)
+ return 0;
+
+ if (x2apic_phys || x2apic_fadt_phys())
+ return 1;
+
+ return apic == &apic_x2apic_phys;
+}
+
+unsigned int x2apic_get_apic_id(unsigned long id)
+{
+ return id;
+}
+
+u32 x2apic_set_apic_id(unsigned int id)
+{
+ return id;
+}
+
+int x2apic_phys_pkg_id(int initial_apicid, int index_msb)
+{
+ return initial_apicid >> index_msb;
+}
+
+static struct apic apic_x2apic_phys __ro_after_init = {
+
+ .name = "physical x2apic",
+ .probe = x2apic_phys_probe,
+ .acpi_madt_oem_check = x2apic_acpi_madt_oem_check,
+
+ .delivery_mode = APIC_DELIVERY_MODE_FIXED,
+ .dest_mode_logical = false,
+
+ .disable_esr = 0,
+
+ .cpu_present_to_apicid = default_cpu_present_to_apicid,
+ .phys_pkg_id = x2apic_phys_pkg_id,
+
+ .max_apic_id = UINT_MAX,
+ .x2apic_set_max_apicid = true,
+ .get_apic_id = x2apic_get_apic_id,
+ .set_apic_id = x2apic_set_apic_id,
+
+ .calc_dest_apicid = apic_default_calc_apicid,
+
+ .send_IPI = x2apic_send_IPI,
+ .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,
+
+ .read = native_apic_msr_read,
+ .write = native_apic_msr_write,
+ .eoi = native_apic_msr_eoi,
+ .icr_read = native_x2apic_icr_read,
+ .icr_write = native_x2apic_icr_write,
+};
+
+apic_driver(apic_x2apic_phys);
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 0000000000..205cee5676
--- /dev/null
+++ b/arch/x86/kernel/apic/x2apic_uv_x.c
@@ -0,0 +1,1872 @@
+/*
+ * 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)
+ *
+ * (C) Copyright 2020 Hewlett Packard Enterprise Development LP
+ * Copyright (C) 2007-2014 Silicon Graphics, Inc. All rights reserved.
+ */
+#include <linux/crash_dump.h>
+#include <linux/cpuhotplug.h>
+#include <linux/cpumask.h>
+#include <linux/proc_fs.h>
+#include <linux/memory.h>
+#include <linux/export.h>
+#include <linux/pci.h>
+#include <linux/acpi.h>
+#include <linux/efi.h>
+
+#include <asm/e820/api.h>
+#include <asm/uv/uv_mmrs.h>
+#include <asm/uv/uv_hub.h>
+#include <asm/uv/bios.h>
+#include <asm/uv/uv.h>
+#include <asm/apic.h>
+
+#include "local.h"
+
+static enum uv_system_type uv_system_type;
+static int uv_hubbed_system;
+static int uv_hubless_system;
+static u64 gru_start_paddr, gru_end_paddr;
+static union uvh_apicid uvh_apicid;
+static int uv_node_id;
+
+/* Unpack AT/OEM/TABLE ID's to be NULL terminated strings */
+static u8 uv_archtype[UV_AT_SIZE + 1];
+static u8 oem_id[ACPI_OEM_ID_SIZE + 1];
+static u8 oem_table_id[ACPI_OEM_TABLE_ID_SIZE + 1];
+
+/* Information derived from CPUID and some UV MMRs */
+static struct {
+ unsigned int apicid_shift;
+ unsigned int apicid_mask;
+ unsigned int socketid_shift; /* aka pnode_shift for UV2/3 */
+ unsigned int pnode_mask;
+ unsigned int nasid_shift;
+ unsigned int gpa_shift;
+ unsigned int gnode_shift;
+ unsigned int m_skt;
+ unsigned int n_skt;
+} uv_cpuid;
+
+static int uv_min_hub_revision_id;
+
+static struct apic apic_x2apic_uv_x;
+static struct uv_hub_info_s uv_hub_info_node0;
+
+/* Set this to use hardware error handler instead of kernel panic: */
+static int disable_uv_undefined_panic = 1;
+
+unsigned long uv_undefined(char *str)
+{
+ if (likely(!disable_uv_undefined_panic))
+ panic("UV: error: undefined MMR: %s\n", str);
+ else
+ pr_crit("UV: error: undefined MMR: %s\n", str);
+
+ /* Cause a machine fault: */
+ return ~0ul;
+}
+EXPORT_SYMBOL(uv_undefined);
+
+static unsigned long __init uv_early_read_mmr(unsigned long addr)
+{
+ unsigned long val, *mmr;
+
+ mmr = early_ioremap(UV_LOCAL_MMR_BASE | addr, sizeof(*mmr));
+ val = *mmr;
+ early_iounmap(mmr, sizeof(*mmr));
+
+ return val;
+}
+
+static inline bool is_GRU_range(u64 start, u64 end)
+{
+ if (!gru_start_paddr)
+ return false;
+
+ return start >= gru_start_paddr && end <= gru_end_paddr;
+}
+
+static bool uv_is_untracked_pat_range(u64 start, u64 end)
+{
+ return is_ISA_range(start, end) || is_GRU_range(start, end);
+}
+
+static void __init early_get_pnodeid(void)
+{
+ int pnode;
+
+ uv_cpuid.m_skt = 0;
+ if (UVH_RH10_GAM_ADDR_MAP_CONFIG) {
+ union uvh_rh10_gam_addr_map_config_u m_n_config;
+
+ m_n_config.v = uv_early_read_mmr(UVH_RH10_GAM_ADDR_MAP_CONFIG);
+ uv_cpuid.n_skt = m_n_config.s.n_skt;
+ uv_cpuid.nasid_shift = 0;
+ } else if (UVH_RH_GAM_ADDR_MAP_CONFIG) {
+ union uvh_rh_gam_addr_map_config_u m_n_config;
+
+ m_n_config.v = uv_early_read_mmr(UVH_RH_GAM_ADDR_MAP_CONFIG);
+ uv_cpuid.n_skt = m_n_config.s.n_skt;
+ if (is_uv(UV3))
+ uv_cpuid.m_skt = m_n_config.s3.m_skt;
+ if (is_uv(UV2))
+ uv_cpuid.m_skt = m_n_config.s2.m_skt;
+ uv_cpuid.nasid_shift = 1;
+ } else {
+ unsigned long GAM_ADDR_MAP_CONFIG = 0;
+
+ WARN(GAM_ADDR_MAP_CONFIG == 0,
+ "UV: WARN: GAM_ADDR_MAP_CONFIG is not available\n");
+ uv_cpuid.n_skt = 0;
+ uv_cpuid.nasid_shift = 0;
+ }
+
+ if (is_uv(UV4|UVY))
+ uv_cpuid.gnode_shift = 2; /* min partition is 4 sockets */
+
+ uv_cpuid.pnode_mask = (1 << uv_cpuid.n_skt) - 1;
+ pnode = (uv_node_id >> uv_cpuid.nasid_shift) & uv_cpuid.pnode_mask;
+ uv_cpuid.gpa_shift = 46; /* Default unless changed */
+
+ pr_info("UV: n_skt:%d pnmsk:%x pn:%x\n",
+ uv_cpuid.n_skt, uv_cpuid.pnode_mask, pnode);
+}
+
+/* Running on a UV Hubbed system, determine which UV Hub Type it is */
+static int __init early_set_hub_type(void)
+{
+ union uvh_node_id_u node_id;
+
+ /*
+ * The NODE_ID MMR is always at offset 0.
+ * Contains the chip part # + revision.
+ * Node_id field started with 15 bits,
+ * ... now 7 but upper 8 are masked to 0.
+ * All blades/nodes have the same part # and hub revision.
+ */
+ node_id.v = uv_early_read_mmr(UVH_NODE_ID);
+ uv_node_id = node_id.sx.node_id;
+
+ switch (node_id.s.part_number) {
+
+ case UV5_HUB_PART_NUMBER:
+ uv_min_hub_revision_id = node_id.s.revision
+ + UV5_HUB_REVISION_BASE;
+ uv_hub_type_set(UV5);
+ break;
+
+ /* UV4/4A only have a revision difference */
+ case UV4_HUB_PART_NUMBER:
+ uv_min_hub_revision_id = node_id.s.revision
+ + UV4_HUB_REVISION_BASE - 1;
+ uv_hub_type_set(UV4);
+ if (uv_min_hub_revision_id == UV4A_HUB_REVISION_BASE)
+ uv_hub_type_set(UV4|UV4A);
+ break;
+
+ case UV3_HUB_PART_NUMBER:
+ case UV3_HUB_PART_NUMBER_X:
+ uv_min_hub_revision_id = node_id.s.revision
+ + UV3_HUB_REVISION_BASE;
+ uv_hub_type_set(UV3);
+ break;
+
+ case UV2_HUB_PART_NUMBER:
+ case UV2_HUB_PART_NUMBER_X:
+ uv_min_hub_revision_id = node_id.s.revision
+ + UV2_HUB_REVISION_BASE - 1;
+ uv_hub_type_set(UV2);
+ break;
+
+ default:
+ return 0;
+ }
+
+ pr_info("UV: part#:%x rev:%d rev_id:%d UVtype:0x%x\n",
+ node_id.s.part_number, node_id.s.revision,
+ uv_min_hub_revision_id, is_uv(~0));
+
+ return 1;
+}
+
+static void __init uv_tsc_check_sync(void)
+{
+ u64 mmr;
+ int sync_state;
+ int mmr_shift;
+ char *state;
+
+ /* UV5 guarantees synced TSCs; do not zero TSC_ADJUST */
+ if (!is_uv(UV2|UV3|UV4)) {
+ mark_tsc_async_resets("UV5+");
+ return;
+ }
+
+ /* UV2,3,4, UV BIOS TSC sync state available */
+ mmr = uv_early_read_mmr(UVH_TSC_SYNC_MMR);
+ mmr_shift =
+ is_uv2_hub() ? UVH_TSC_SYNC_SHIFT_UV2K : UVH_TSC_SYNC_SHIFT;
+ sync_state = (mmr >> mmr_shift) & UVH_TSC_SYNC_MASK;
+
+ /* Check if TSC is valid for all sockets */
+ switch (sync_state) {
+ case UVH_TSC_SYNC_VALID:
+ state = "in sync";
+ mark_tsc_async_resets("UV BIOS");
+ break;
+
+ /* If BIOS state unknown, don't do anything */
+ case UVH_TSC_SYNC_UNKNOWN:
+ state = "unknown";
+ break;
+
+ /* Otherwise, BIOS indicates problem with TSC */
+ default:
+ state = "unstable";
+ mark_tsc_unstable("UV BIOS");
+ break;
+ }
+ pr_info("UV: TSC sync state from BIOS:0%d(%s)\n", sync_state, state);
+}
+
+/* Selector for (4|4A|5) structs */
+#define uvxy_field(sname, field, undef) ( \
+ is_uv(UV4A) ? sname.s4a.field : \
+ is_uv(UV4) ? sname.s4.field : \
+ is_uv(UV3) ? sname.s3.field : \
+ undef)
+
+/* [Copied from arch/x86/kernel/cpu/topology.c:detect_extended_topology()] */
+
+#define SMT_LEVEL 0 /* Leaf 0xb SMT level */
+#define INVALID_TYPE 0 /* Leaf 0xb sub-leaf types */
+#define SMT_TYPE 1
+#define CORE_TYPE 2
+#define LEAFB_SUBTYPE(ecx) (((ecx) >> 8) & 0xff)
+#define BITS_SHIFT_NEXT_LEVEL(eax) ((eax) & 0x1f)
+
+static void set_x2apic_bits(void)
+{
+ unsigned int eax, ebx, ecx, edx, sub_index;
+ unsigned int sid_shift;
+
+ cpuid(0, &eax, &ebx, &ecx, &edx);
+ if (eax < 0xb) {
+ pr_info("UV: CPU does not have CPUID.11\n");
+ return;
+ }
+
+ cpuid_count(0xb, SMT_LEVEL, &eax, &ebx, &ecx, &edx);
+ if (ebx == 0 || (LEAFB_SUBTYPE(ecx) != SMT_TYPE)) {
+ pr_info("UV: CPUID.11 not implemented\n");
+ return;
+ }
+
+ sid_shift = BITS_SHIFT_NEXT_LEVEL(eax);
+ sub_index = 1;
+ do {
+ cpuid_count(0xb, sub_index, &eax, &ebx, &ecx, &edx);
+ if (LEAFB_SUBTYPE(ecx) == CORE_TYPE) {
+ sid_shift = BITS_SHIFT_NEXT_LEVEL(eax);
+ break;
+ }
+ sub_index++;
+ } while (LEAFB_SUBTYPE(ecx) != INVALID_TYPE);
+
+ uv_cpuid.apicid_shift = 0;
+ uv_cpuid.apicid_mask = (~(-1 << sid_shift));
+ uv_cpuid.socketid_shift = sid_shift;
+}
+
+static void __init early_get_apic_socketid_shift(void)
+{
+ if (is_uv2_hub() || is_uv3_hub())
+ uvh_apicid.v = uv_early_read_mmr(UVH_APICID);
+
+ set_x2apic_bits();
+
+ pr_info("UV: apicid_shift:%d apicid_mask:0x%x\n", uv_cpuid.apicid_shift, uv_cpuid.apicid_mask);
+ pr_info("UV: socketid_shift:%d pnode_mask:0x%x\n", uv_cpuid.socketid_shift, uv_cpuid.pnode_mask);
+}
+
+static void __init uv_stringify(int len, char *to, char *from)
+{
+ strscpy(to, from, len);
+
+ /* Trim trailing spaces */
+ (void)strim(to);
+}
+
+/* Find UV arch type entry in UVsystab */
+static unsigned long __init early_find_archtype(struct uv_systab *st)
+{
+ int i;
+
+ for (i = 0; st->entry[i].type != UV_SYSTAB_TYPE_UNUSED; i++) {
+ unsigned long ptr = st->entry[i].offset;
+
+ if (!ptr)
+ continue;
+ ptr += (unsigned long)st;
+ if (st->entry[i].type == UV_SYSTAB_TYPE_ARCH_TYPE)
+ return ptr;
+ }
+ return 0;
+}
+
+/* Validate UV arch type field in UVsystab */
+static int __init decode_arch_type(unsigned long ptr)
+{
+ struct uv_arch_type_entry *uv_ate = (struct uv_arch_type_entry *)ptr;
+ int n = strlen(uv_ate->archtype);
+
+ if (n > 0 && n < sizeof(uv_ate->archtype)) {
+ pr_info("UV: UVarchtype received from BIOS\n");
+ uv_stringify(sizeof(uv_archtype), uv_archtype, uv_ate->archtype);
+ return 1;
+ }
+ return 0;
+}
+
+/* Determine if UV arch type entry might exist in UVsystab */
+static int __init early_get_arch_type(void)
+{
+ unsigned long uvst_physaddr, uvst_size, ptr;
+ struct uv_systab *st;
+ u32 rev;
+ int ret;
+
+ uvst_physaddr = get_uv_systab_phys(0);
+ if (!uvst_physaddr)
+ return 0;
+
+ st = early_memremap_ro(uvst_physaddr, sizeof(struct uv_systab));
+ if (!st) {
+ pr_err("UV: Cannot access UVsystab, remap failed\n");
+ return 0;
+ }
+
+ rev = st->revision;
+ if (rev < UV_SYSTAB_VERSION_UV5) {
+ early_memunmap(st, sizeof(struct uv_systab));
+ return 0;
+ }
+
+ uvst_size = st->size;
+ early_memunmap(st, sizeof(struct uv_systab));
+ st = early_memremap_ro(uvst_physaddr, uvst_size);
+ if (!st) {
+ pr_err("UV: Cannot access UVarchtype, remap failed\n");
+ return 0;
+ }
+
+ ptr = early_find_archtype(st);
+ if (!ptr) {
+ early_memunmap(st, uvst_size);
+ return 0;
+ }
+
+ ret = decode_arch_type(ptr);
+ early_memunmap(st, uvst_size);
+ return ret;
+}
+
+/* UV system found, check which APIC MODE BIOS already selected */
+static void __init early_set_apic_mode(void)
+{
+ if (x2apic_enabled())
+ uv_system_type = UV_X2APIC;
+ else
+ uv_system_type = UV_LEGACY_APIC;
+}
+
+static int __init uv_set_system_type(char *_oem_id, char *_oem_table_id)
+{
+ /* Save OEM_ID passed from ACPI MADT */
+ uv_stringify(sizeof(oem_id), oem_id, _oem_id);
+
+ /* Check if BIOS sent us a UVarchtype */
+ if (!early_get_arch_type())
+
+ /* If not use OEM ID for UVarchtype */
+ uv_stringify(sizeof(uv_archtype), uv_archtype, oem_id);
+
+ /* Check if not hubbed */
+ if (strncmp(uv_archtype, "SGI", 3) != 0) {
+
+ /* (Not hubbed), check if not hubless */
+ if (strncmp(uv_archtype, "NSGI", 4) != 0)
+
+ /* (Not hubless), not a UV */
+ return 0;
+
+ /* Is UV hubless system */
+ uv_hubless_system = 0x01;
+
+ /* UV5 Hubless */
+ if (strncmp(uv_archtype, "NSGI5", 5) == 0)
+ uv_hubless_system |= 0x20;
+
+ /* UV4 Hubless: CH */
+ else if (strncmp(uv_archtype, "NSGI4", 5) == 0)
+ uv_hubless_system |= 0x10;
+
+ /* UV3 Hubless: UV300/MC990X w/o hub */
+ else
+ uv_hubless_system |= 0x8;
+
+ /* Copy OEM Table ID */
+ uv_stringify(sizeof(oem_table_id), oem_table_id, _oem_table_id);
+
+ pr_info("UV: OEM IDs %s/%s, SystemType %d, HUBLESS ID %x\n",
+ oem_id, oem_table_id, uv_system_type, uv_hubless_system);
+
+ return 0;
+ }
+
+ if (numa_off) {
+ pr_err("UV: NUMA is off, disabling UV support\n");
+ return 0;
+ }
+
+ /* Set hubbed type if true */
+ uv_hub_info->hub_revision =
+ !strncmp(uv_archtype, "SGI5", 4) ? UV5_HUB_REVISION_BASE :
+ !strncmp(uv_archtype, "SGI4", 4) ? UV4_HUB_REVISION_BASE :
+ !strncmp(uv_archtype, "SGI3", 4) ? UV3_HUB_REVISION_BASE :
+ !strcmp(uv_archtype, "SGI2") ? UV2_HUB_REVISION_BASE : 0;
+
+ switch (uv_hub_info->hub_revision) {
+ case UV5_HUB_REVISION_BASE:
+ uv_hubbed_system = 0x21;
+ uv_hub_type_set(UV5);
+ break;
+
+ case UV4_HUB_REVISION_BASE:
+ uv_hubbed_system = 0x11;
+ uv_hub_type_set(UV4);
+ break;
+
+ case UV3_HUB_REVISION_BASE:
+ uv_hubbed_system = 0x9;
+ uv_hub_type_set(UV3);
+ break;
+
+ case UV2_HUB_REVISION_BASE:
+ uv_hubbed_system = 0x5;
+ uv_hub_type_set(UV2);
+ break;
+
+ default:
+ return 0;
+ }
+
+ /* Get UV hub chip part number & revision */
+ early_set_hub_type();
+
+ /* Other UV setup functions */
+ early_set_apic_mode();
+ early_get_pnodeid();
+ early_get_apic_socketid_shift();
+ x86_platform.is_untracked_pat_range = uv_is_untracked_pat_range;
+ x86_platform.nmi_init = uv_nmi_init;
+ uv_tsc_check_sync();
+
+ return 1;
+}
+
+/* Called early to probe for the correct APIC driver */
+static int __init uv_acpi_madt_oem_check(char *_oem_id, char *_oem_table_id)
+{
+ /* Set up early hub info fields for Node 0 */
+ uv_cpu_info->p_uv_hub_info = &uv_hub_info_node0;
+
+ /* If not UV, return. */
+ if (uv_set_system_type(_oem_id, _oem_table_id) == 0)
+ return 0;
+
+ /* Save for display of the OEM Table ID */
+ uv_stringify(sizeof(oem_table_id), oem_table_id, _oem_table_id);
+
+ pr_info("UV: OEM IDs %s/%s, System/UVType %d/0x%x, HUB RevID %d\n",
+ oem_id, oem_table_id, uv_system_type, is_uv(UV_ANY),
+ uv_min_hub_revision_id);
+
+ return 0;
+}
+
+enum uv_system_type get_uv_system_type(void)
+{
+ return uv_system_type;
+}
+
+int uv_get_hubless_system(void)
+{
+ return uv_hubless_system;
+}
+EXPORT_SYMBOL_GPL(uv_get_hubless_system);
+
+ssize_t uv_get_archtype(char *buf, int len)
+{
+ return scnprintf(buf, len, "%s/%s", uv_archtype, oem_table_id);
+}
+EXPORT_SYMBOL_GPL(uv_get_archtype);
+
+int is_uv_system(void)
+{
+ return uv_system_type != UV_NONE;
+}
+EXPORT_SYMBOL_GPL(is_uv_system);
+
+int is_uv_hubbed(int uvtype)
+{
+ return (uv_hubbed_system & uvtype);
+}
+EXPORT_SYMBOL_GPL(is_uv_hubbed);
+
+static int is_uv_hubless(int uvtype)
+{
+ return (uv_hubless_system & uvtype);
+}
+
+void **__uv_hub_info_list;
+EXPORT_SYMBOL_GPL(__uv_hub_info_list);
+
+DEFINE_PER_CPU(struct uv_cpu_info_s, __uv_cpu_info);
+EXPORT_PER_CPU_SYMBOL_GPL(__uv_cpu_info);
+
+short uv_possible_blades;
+EXPORT_SYMBOL_GPL(uv_possible_blades);
+
+unsigned long sn_rtc_cycles_per_second;
+EXPORT_SYMBOL(sn_rtc_cycles_per_second);
+
+/* The following values are used for the per node hub info struct */
+static __initdata unsigned short _min_socket, _max_socket;
+static __initdata unsigned short _min_pnode, _max_pnode, _gr_table_len;
+static __initdata struct uv_gam_range_entry *uv_gre_table;
+static __initdata struct uv_gam_parameters *uv_gp_table;
+static __initdata unsigned short *_socket_to_node;
+static __initdata unsigned short *_socket_to_pnode;
+static __initdata unsigned short *_pnode_to_socket;
+static __initdata unsigned short *_node_to_socket;
+
+static __initdata struct uv_gam_range_s *_gr_table;
+
+#define SOCK_EMPTY ((unsigned short)~0)
+
+/* Default UV memory block size is 2GB */
+static unsigned long mem_block_size __initdata = (2UL << 30);
+
+/* Kernel parameter to specify UV mem block size */
+static int __init parse_mem_block_size(char *ptr)
+{
+ unsigned long size = memparse(ptr, NULL);
+
+ /* Size will be rounded down by set_block_size() below */
+ mem_block_size = size;
+ return 0;
+}
+early_param("uv_memblksize", parse_mem_block_size);
+
+static __init int adj_blksize(u32 lgre)
+{
+ unsigned long base = (unsigned long)lgre << UV_GAM_RANGE_SHFT;
+ unsigned long size;
+
+ for (size = mem_block_size; size > MIN_MEMORY_BLOCK_SIZE; size >>= 1)
+ if (IS_ALIGNED(base, size))
+ break;
+
+ if (size >= mem_block_size)
+ return 0;
+
+ mem_block_size = size;
+ return 1;
+}
+
+static __init void set_block_size(void)
+{
+ unsigned int order = ffs(mem_block_size);
+
+ if (order) {
+ /* adjust for ffs return of 1..64 */
+ set_memory_block_size_order(order - 1);
+ pr_info("UV: mem_block_size set to 0x%lx\n", mem_block_size);
+ } else {
+ /* bad or zero value, default to 1UL << 31 (2GB) */
+ pr_err("UV: mem_block_size error with 0x%lx\n", mem_block_size);
+ set_memory_block_size_order(31);
+ }
+}
+
+/* Build GAM range lookup table: */
+static __init void build_uv_gr_table(void)
+{
+ struct uv_gam_range_entry *gre = uv_gre_table;
+ struct uv_gam_range_s *grt;
+ unsigned long last_limit = 0, ram_limit = 0;
+ int bytes, i, sid, lsid = -1, indx = 0, lindx = -1;
+
+ if (!gre)
+ return;
+
+ bytes = _gr_table_len * sizeof(struct uv_gam_range_s);
+ grt = kzalloc(bytes, GFP_KERNEL);
+ if (WARN_ON_ONCE(!grt))
+ return;
+ _gr_table = grt;
+
+ for (; gre->type != UV_GAM_RANGE_TYPE_UNUSED; gre++) {
+ if (gre->type == UV_GAM_RANGE_TYPE_HOLE) {
+ if (!ram_limit) {
+ /* Mark hole between RAM/non-RAM: */
+ ram_limit = last_limit;
+ last_limit = gre->limit;
+ lsid++;
+ continue;
+ }
+ last_limit = gre->limit;
+ pr_info("UV: extra hole in GAM RE table @%d\n", (int)(gre - uv_gre_table));
+ continue;
+ }
+ if (_max_socket < gre->sockid) {
+ pr_err("UV: GAM table sockid(%d) too large(>%d) @%d\n", gre->sockid, _max_socket, (int)(gre - uv_gre_table));
+ continue;
+ }
+ sid = gre->sockid - _min_socket;
+ if (lsid < sid) {
+ /* New range: */
+ grt = &_gr_table[indx];
+ grt->base = lindx;
+ grt->nasid = gre->nasid;
+ grt->limit = last_limit = gre->limit;
+ lsid = sid;
+ lindx = indx++;
+ continue;
+ }
+ /* Update range: */
+ if (lsid == sid && !ram_limit) {
+ /* .. if contiguous: */
+ if (grt->limit == last_limit) {
+ grt->limit = last_limit = gre->limit;
+ continue;
+ }
+ }
+ /* Non-contiguous RAM range: */
+ if (!ram_limit) {
+ grt++;
+ grt->base = lindx;
+ grt->nasid = gre->nasid;
+ grt->limit = last_limit = gre->limit;
+ continue;
+ }
+ /* Non-contiguous/non-RAM: */
+ grt++;
+ /* base is this entry */
+ grt->base = grt - _gr_table;
+ grt->nasid = gre->nasid;
+ grt->limit = last_limit = gre->limit;
+ lsid++;
+ }
+
+ /* Shorten table if possible */
+ grt++;
+ i = grt - _gr_table;
+ if (i < _gr_table_len) {
+ void *ret;
+
+ bytes = i * sizeof(struct uv_gam_range_s);
+ ret = krealloc(_gr_table, bytes, GFP_KERNEL);
+ if (ret) {
+ _gr_table = ret;
+ _gr_table_len = i;
+ }
+ }
+
+ /* Display resultant GAM range table: */
+ for (i = 0, grt = _gr_table; i < _gr_table_len; i++, grt++) {
+ unsigned long start, end;
+ int gb = grt->base;
+
+ start = gb < 0 ? 0 : (unsigned long)_gr_table[gb].limit << UV_GAM_RANGE_SHFT;
+ end = (unsigned long)grt->limit << UV_GAM_RANGE_SHFT;
+
+ pr_info("UV: GAM Range %2d %04x 0x%013lx-0x%013lx (%d)\n", i, grt->nasid, start, end, gb);
+ }
+}
+
+static int uv_wakeup_secondary(int phys_apicid, unsigned long 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) |
+ ((start_rip << UVH_IPI_INT_VECTOR_SHFT) >> 12) |
+ APIC_DM_INIT;
+
+ uv_write_global_mmr64(pnode, UVH_IPI_INT, val);
+
+ val = (1UL << UVH_IPI_INT_SEND_SHFT) |
+ (phys_apicid << UVH_IPI_INT_APIC_ID_SHFT) |
+ ((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 apicid = per_cpu(x86_cpu_to_apicid, cpu);
+ int pnode = uv_apicid_to_pnode(apicid);
+ unsigned long dmode, val;
+
+ if (vector == NMI_VECTOR)
+ dmode = APIC_DELIVERY_MODE_NMI;
+ else
+ dmode = APIC_DELIVERY_MODE_FIXED;
+
+ val = (1UL << UVH_IPI_INT_SEND_SHFT) |
+ (apicid << UVH_IPI_INT_APIC_ID_SHFT) |
+ (dmode << UVH_IPI_INT_DELIVERY_MODE_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 u32 set_apic_id(unsigned int id)
+{
+ return id;
+}
+
+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 int uv_probe(void)
+{
+ return apic == &apic_x2apic_uv_x;
+}
+
+static struct apic apic_x2apic_uv_x __ro_after_init = {
+
+ .name = "UV large system",
+ .probe = uv_probe,
+ .acpi_madt_oem_check = uv_acpi_madt_oem_check,
+
+ .delivery_mode = APIC_DELIVERY_MODE_FIXED,
+ .dest_mode_logical = false,
+
+ .disable_esr = 0,
+
+ .cpu_present_to_apicid = default_cpu_present_to_apicid,
+ .phys_pkg_id = uv_phys_pkg_id,
+
+ .max_apic_id = UINT_MAX,
+ .get_apic_id = x2apic_get_apic_id,
+ .set_apic_id = set_apic_id,
+
+ .calc_dest_apicid = apic_default_calc_apicid,
+
+ .send_IPI = uv_send_IPI_one,
+ .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 = x2apic_send_IPI_self,
+
+ .wakeup_secondary_cpu = uv_wakeup_secondary,
+
+ .read = native_apic_msr_read,
+ .write = native_apic_msr_write,
+ .eoi = native_apic_msr_eoi,
+ .icr_read = native_x2apic_icr_read,
+ .icr_write = native_x2apic_icr_write,
+};
+
+#define UVH_RH_GAM_ALIAS210_REDIRECT_CONFIG_LENGTH 3
+#define DEST_SHIFT UVXH_RH_GAM_ALIAS_0_REDIRECT_CONFIG_DEST_BASE_SHFT
+
+static __init void get_lowmem_redirect(unsigned long *base, unsigned long *size)
+{
+ union uvh_rh_gam_alias_2_overlay_config_u alias;
+ union uvh_rh_gam_alias_2_redirect_config_u redirect;
+ unsigned long m_redirect;
+ unsigned long m_overlay;
+ int i;
+
+ for (i = 0; i < UVH_RH_GAM_ALIAS210_REDIRECT_CONFIG_LENGTH; i++) {
+ switch (i) {
+ case 0:
+ m_redirect = UVH_RH_GAM_ALIAS_0_REDIRECT_CONFIG;
+ m_overlay = UVH_RH_GAM_ALIAS_0_OVERLAY_CONFIG;
+ break;
+ case 1:
+ m_redirect = UVH_RH_GAM_ALIAS_1_REDIRECT_CONFIG;
+ m_overlay = UVH_RH_GAM_ALIAS_1_OVERLAY_CONFIG;
+ break;
+ case 2:
+ m_redirect = UVH_RH_GAM_ALIAS_2_REDIRECT_CONFIG;
+ m_overlay = UVH_RH_GAM_ALIAS_2_OVERLAY_CONFIG;
+ break;
+ }
+ alias.v = uv_read_local_mmr(m_overlay);
+ if (alias.s.enable && alias.s.base == 0) {
+ *size = (1UL << alias.s.m_alias);
+ redirect.v = uv_read_local_mmr(m_redirect);
+ *base = (unsigned long)redirect.s.dest_base << DEST_SHIFT;
+ return;
+ }
+ }
+ *base = *size = 0;
+}
+
+enum map_type {map_wb, map_uc};
+static const char * const mt[] = { "WB", "UC" };
+
+static __init void map_high(char *id, unsigned long base, int pshift, int bshift, int max_pnode, enum map_type map_type)
+{
+ unsigned long bytes, paddr;
+
+ paddr = base << pshift;
+ bytes = (1UL << bshift) * (max_pnode + 1);
+ if (!paddr) {
+ pr_info("UV: Map %s_HI base address NULL\n", id);
+ return;
+ }
+ if (map_type == map_uc)
+ init_extra_mapping_uc(paddr, bytes);
+ else
+ init_extra_mapping_wb(paddr, bytes);
+
+ pr_info("UV: Map %s_HI 0x%lx - 0x%lx %s (%d segments)\n",
+ id, paddr, paddr + bytes, mt[map_type], max_pnode + 1);
+}
+
+static __init void map_gru_high(int max_pnode)
+{
+ union uvh_rh_gam_gru_overlay_config_u gru;
+ unsigned long mask, base;
+ int shift;
+
+ if (UVH_RH_GAM_GRU_OVERLAY_CONFIG) {
+ gru.v = uv_read_local_mmr(UVH_RH_GAM_GRU_OVERLAY_CONFIG);
+ shift = UVH_RH_GAM_GRU_OVERLAY_CONFIG_BASE_SHFT;
+ mask = UVH_RH_GAM_GRU_OVERLAY_CONFIG_BASE_MASK;
+ } else if (UVH_RH10_GAM_GRU_OVERLAY_CONFIG) {
+ gru.v = uv_read_local_mmr(UVH_RH10_GAM_GRU_OVERLAY_CONFIG);
+ shift = UVH_RH10_GAM_GRU_OVERLAY_CONFIG_BASE_SHFT;
+ mask = UVH_RH10_GAM_GRU_OVERLAY_CONFIG_BASE_MASK;
+ } else {
+ pr_err("UV: GRU unavailable (no MMR)\n");
+ return;
+ }
+
+ if (!gru.s.enable) {
+ pr_info("UV: GRU disabled (by BIOS)\n");
+ return;
+ }
+
+ base = (gru.v & mask) >> shift;
+ map_high("GRU", base, shift, shift, max_pnode, map_wb);
+ gru_start_paddr = ((u64)base << shift);
+ gru_end_paddr = gru_start_paddr + (1UL << shift) * (max_pnode + 1);
+}
+
+static __init void map_mmr_high(int max_pnode)
+{
+ unsigned long base;
+ int shift;
+ bool enable;
+
+ if (UVH_RH10_GAM_MMR_OVERLAY_CONFIG) {
+ union uvh_rh10_gam_mmr_overlay_config_u mmr;
+
+ mmr.v = uv_read_local_mmr(UVH_RH10_GAM_MMR_OVERLAY_CONFIG);
+ enable = mmr.s.enable;
+ base = mmr.s.base;
+ shift = UVH_RH10_GAM_MMR_OVERLAY_CONFIG_BASE_SHFT;
+ } else if (UVH_RH_GAM_MMR_OVERLAY_CONFIG) {
+ union uvh_rh_gam_mmr_overlay_config_u mmr;
+
+ mmr.v = uv_read_local_mmr(UVH_RH_GAM_MMR_OVERLAY_CONFIG);
+ enable = mmr.s.enable;
+ base = mmr.s.base;
+ shift = UVH_RH_GAM_MMR_OVERLAY_CONFIG_BASE_SHFT;
+ } else {
+ pr_err("UV:%s:RH_GAM_MMR_OVERLAY_CONFIG MMR undefined?\n",
+ __func__);
+ return;
+ }
+
+ if (enable)
+ map_high("MMR", base, shift, shift, max_pnode, map_uc);
+ else
+ pr_info("UV: MMR disabled\n");
+}
+
+/* Arch specific ENUM cases */
+enum mmioh_arch {
+ UV2_MMIOH = -1,
+ UVY_MMIOH0, UVY_MMIOH1,
+ UVX_MMIOH0, UVX_MMIOH1,
+};
+
+/* Calculate and Map MMIOH Regions */
+static void __init calc_mmioh_map(enum mmioh_arch index,
+ int min_pnode, int max_pnode,
+ int shift, unsigned long base, int m_io, int n_io)
+{
+ unsigned long mmr, nasid_mask;
+ int nasid, min_nasid, max_nasid, lnasid, mapped;
+ int i, fi, li, n, max_io;
+ char id[8];
+
+ /* One (UV2) mapping */
+ if (index == UV2_MMIOH) {
+ strscpy(id, "MMIOH", sizeof(id));
+ max_io = max_pnode;
+ mapped = 0;
+ goto map_exit;
+ }
+
+ /* small and large MMIOH mappings */
+ switch (index) {
+ case UVY_MMIOH0:
+ mmr = UVH_RH10_GAM_MMIOH_REDIRECT_CONFIG0;
+ nasid_mask = UVYH_RH10_GAM_MMIOH_REDIRECT_CONFIG0_NASID_MASK;
+ n = UVH_RH10_GAM_MMIOH_REDIRECT_CONFIG0_DEPTH;
+ min_nasid = min_pnode;
+ max_nasid = max_pnode;
+ mapped = 1;
+ break;
+ case UVY_MMIOH1:
+ mmr = UVH_RH10_GAM_MMIOH_REDIRECT_CONFIG1;
+ nasid_mask = UVYH_RH10_GAM_MMIOH_REDIRECT_CONFIG1_NASID_MASK;
+ n = UVH_RH10_GAM_MMIOH_REDIRECT_CONFIG1_DEPTH;
+ min_nasid = min_pnode;
+ max_nasid = max_pnode;
+ mapped = 1;
+ break;
+ case UVX_MMIOH0:
+ mmr = UVH_RH_GAM_MMIOH_REDIRECT_CONFIG0;
+ nasid_mask = UVH_RH_GAM_MMIOH_REDIRECT_CONFIG0_NASID_MASK;
+ n = UVH_RH_GAM_MMIOH_REDIRECT_CONFIG0_DEPTH;
+ min_nasid = min_pnode * 2;
+ max_nasid = max_pnode * 2;
+ mapped = 1;
+ break;
+ case UVX_MMIOH1:
+ mmr = UVH_RH_GAM_MMIOH_REDIRECT_CONFIG1;
+ nasid_mask = UVH_RH_GAM_MMIOH_REDIRECT_CONFIG1_NASID_MASK;
+ n = UVH_RH_GAM_MMIOH_REDIRECT_CONFIG1_DEPTH;
+ min_nasid = min_pnode * 2;
+ max_nasid = max_pnode * 2;
+ mapped = 1;
+ break;
+ default:
+ pr_err("UV:%s:Invalid mapping type:%d\n", __func__, index);
+ return;
+ }
+
+ /* enum values chosen so (index mod 2) is MMIOH 0/1 (low/high) */
+ snprintf(id, sizeof(id), "MMIOH%d", index%2);
+
+ max_io = lnasid = fi = li = -1;
+ for (i = 0; i < n; i++) {
+ unsigned long m_redirect = mmr + i * 8;
+ unsigned long redirect = uv_read_local_mmr(m_redirect);
+
+ nasid = redirect & nasid_mask;
+ if (i == 0)
+ pr_info("UV: %s redirect base 0x%lx(@0x%lx) 0x%04x\n",
+ id, redirect, m_redirect, nasid);
+
+ /* Invalid NASID check */
+ if (nasid < min_nasid || max_nasid < nasid) {
+ /* Not an error: unused table entries get "poison" values */
+ pr_debug("UV:%s:Invalid NASID(%x):%x (range:%x..%x)\n",
+ __func__, index, nasid, min_nasid, max_nasid);
+ nasid = -1;
+ }
+
+ if (nasid == lnasid) {
+ li = i;
+ /* Last entry check: */
+ if (i != n-1)
+ continue;
+ }
+
+ /* Check if we have a cached (or last) redirect to print: */
+ if (lnasid != -1 || (i == n-1 && nasid != -1)) {
+ unsigned long addr1, addr2;
+ int f, l;
+
+ if (lnasid == -1) {
+ f = l = i;
+ lnasid = nasid;
+ } else {
+ f = fi;
+ l = li;
+ }
+ addr1 = (base << shift) + f * (1ULL << m_io);
+ addr2 = (base << shift) + (l + 1) * (1ULL << m_io);
+ pr_info("UV: %s[%03d..%03d] NASID 0x%04x ADDR 0x%016lx - 0x%016lx\n",
+ id, fi, li, lnasid, addr1, addr2);
+ if (max_io < l)
+ max_io = l;
+ }
+ fi = li = i;
+ lnasid = nasid;
+ }
+
+map_exit:
+ pr_info("UV: %s base:0x%lx shift:%d m_io:%d max_io:%d max_pnode:0x%x\n",
+ id, base, shift, m_io, max_io, max_pnode);
+
+ if (max_io >= 0 && !mapped)
+ map_high(id, base, shift, m_io, max_io, map_uc);
+}
+
+static __init void map_mmioh_high(int min_pnode, int max_pnode)
+{
+ /* UVY flavor */
+ if (UVH_RH10_GAM_MMIOH_OVERLAY_CONFIG0) {
+ union uvh_rh10_gam_mmioh_overlay_config0_u mmioh0;
+ union uvh_rh10_gam_mmioh_overlay_config1_u mmioh1;
+
+ mmioh0.v = uv_read_local_mmr(UVH_RH10_GAM_MMIOH_OVERLAY_CONFIG0);
+ if (unlikely(mmioh0.s.enable == 0))
+ pr_info("UV: MMIOH0 disabled\n");
+ else
+ calc_mmioh_map(UVY_MMIOH0, min_pnode, max_pnode,
+ UVH_RH10_GAM_MMIOH_OVERLAY_CONFIG0_BASE_SHFT,
+ mmioh0.s.base, mmioh0.s.m_io, mmioh0.s.n_io);
+
+ mmioh1.v = uv_read_local_mmr(UVH_RH10_GAM_MMIOH_OVERLAY_CONFIG1);
+ if (unlikely(mmioh1.s.enable == 0))
+ pr_info("UV: MMIOH1 disabled\n");
+ else
+ calc_mmioh_map(UVY_MMIOH1, min_pnode, max_pnode,
+ UVH_RH10_GAM_MMIOH_OVERLAY_CONFIG1_BASE_SHFT,
+ mmioh1.s.base, mmioh1.s.m_io, mmioh1.s.n_io);
+ return;
+ }
+ /* UVX flavor */
+ if (UVH_RH_GAM_MMIOH_OVERLAY_CONFIG0) {
+ union uvh_rh_gam_mmioh_overlay_config0_u mmioh0;
+ union uvh_rh_gam_mmioh_overlay_config1_u mmioh1;
+
+ mmioh0.v = uv_read_local_mmr(UVH_RH_GAM_MMIOH_OVERLAY_CONFIG0);
+ if (unlikely(mmioh0.s.enable == 0))
+ pr_info("UV: MMIOH0 disabled\n");
+ else {
+ unsigned long base = uvxy_field(mmioh0, base, 0);
+ int m_io = uvxy_field(mmioh0, m_io, 0);
+ int n_io = uvxy_field(mmioh0, n_io, 0);
+
+ calc_mmioh_map(UVX_MMIOH0, min_pnode, max_pnode,
+ UVH_RH_GAM_MMIOH_OVERLAY_CONFIG0_BASE_SHFT,
+ base, m_io, n_io);
+ }
+
+ mmioh1.v = uv_read_local_mmr(UVH_RH_GAM_MMIOH_OVERLAY_CONFIG1);
+ if (unlikely(mmioh1.s.enable == 0))
+ pr_info("UV: MMIOH1 disabled\n");
+ else {
+ unsigned long base = uvxy_field(mmioh1, base, 0);
+ int m_io = uvxy_field(mmioh1, m_io, 0);
+ int n_io = uvxy_field(mmioh1, n_io, 0);
+
+ calc_mmioh_map(UVX_MMIOH1, min_pnode, max_pnode,
+ UVH_RH_GAM_MMIOH_OVERLAY_CONFIG1_BASE_SHFT,
+ base, m_io, n_io);
+ }
+ return;
+ }
+
+ /* UV2 flavor */
+ if (UVH_RH_GAM_MMIOH_OVERLAY_CONFIG) {
+ union uvh_rh_gam_mmioh_overlay_config_u mmioh;
+
+ mmioh.v = uv_read_local_mmr(UVH_RH_GAM_MMIOH_OVERLAY_CONFIG);
+ if (unlikely(mmioh.s2.enable == 0))
+ pr_info("UV: MMIOH disabled\n");
+ else
+ calc_mmioh_map(UV2_MMIOH, min_pnode, max_pnode,
+ UV2H_RH_GAM_MMIOH_OVERLAY_CONFIG_BASE_SHFT,
+ mmioh.s2.base, mmioh.s2.m_io, mmioh.s2.n_io);
+ return;
+ }
+}
+
+static __init void map_low_mmrs(void)
+{
+ if (UV_GLOBAL_MMR32_BASE)
+ init_extra_mapping_uc(UV_GLOBAL_MMR32_BASE, UV_GLOBAL_MMR32_SIZE);
+
+ if (UV_LOCAL_MMR_BASE)
+ init_extra_mapping_uc(UV_LOCAL_MMR_BASE, UV_LOCAL_MMR_SIZE);
+}
+
+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) {
+ pr_warn("UV: unable to determine platform RTC clock frequency, guessing.\n");
+
+ /* BIOS gives wrong value for clock frequency, so guess: */
+ sn_rtc_cycles_per_second = 1000000000000UL / 30000UL;
+ } else {
+ sn_rtc_cycles_per_second = ticks_per_sec;
+ }
+}
+
+/* Direct Legacy VGA I/O traffic to designated IOH */
+static int uv_set_vga_state(struct pci_dev *pdev, bool decode, unsigned int command_bits, u32 flags)
+{
+ int domain, bus, rc;
+
+ if (!(flags & PCI_VGA_STATE_CHANGE_BRIDGE))
+ return 0;
+
+ if ((command_bits & PCI_COMMAND_IO) == 0)
+ return 0;
+
+ domain = pci_domain_nr(pdev->bus);
+ bus = pdev->bus->number;
+
+ rc = uv_bios_set_legacy_vga_target(decode, domain, bus);
+
+ return rc;
+}
+
+/*
+ * Called on each CPU to initialize the per_cpu UV data area.
+ * FIXME: hotplug not supported yet
+ */
+void uv_cpu_init(void)
+{
+ /* CPU 0 initialization will be done via uv_system_init. */
+ if (smp_processor_id() == 0)
+ return;
+
+ uv_hub_info->nr_online_cpus++;
+}
+
+struct mn {
+ unsigned char m_val;
+ unsigned char n_val;
+ unsigned char m_shift;
+ unsigned char n_lshift;
+};
+
+/* Initialize caller's MN struct and fill in values */
+static void get_mn(struct mn *mnp)
+{
+ memset(mnp, 0, sizeof(*mnp));
+ mnp->n_val = uv_cpuid.n_skt;
+ if (is_uv(UV4|UVY)) {
+ mnp->m_val = 0;
+ mnp->n_lshift = 0;
+ } else if (is_uv3_hub()) {
+ union uvyh_gr0_gam_gr_config_u m_gr_config;
+
+ mnp->m_val = uv_cpuid.m_skt;
+ m_gr_config.v = uv_read_local_mmr(UVH_GR0_GAM_GR_CONFIG);
+ mnp->n_lshift = m_gr_config.s3.m_skt;
+ } else if (is_uv2_hub()) {
+ mnp->m_val = uv_cpuid.m_skt;
+ mnp->n_lshift = mnp->m_val == 40 ? 40 : 39;
+ }
+ mnp->m_shift = mnp->m_val ? 64 - mnp->m_val : 0;
+}
+
+static void __init uv_init_hub_info(struct uv_hub_info_s *hi)
+{
+ struct mn mn;
+
+ get_mn(&mn);
+ hi->gpa_mask = mn.m_val ?
+ (1UL << (mn.m_val + mn.n_val)) - 1 :
+ (1UL << uv_cpuid.gpa_shift) - 1;
+
+ hi->m_val = mn.m_val;
+ hi->n_val = mn.n_val;
+ hi->m_shift = mn.m_shift;
+ hi->n_lshift = mn.n_lshift ? mn.n_lshift : 0;
+ hi->hub_revision = uv_hub_info->hub_revision;
+ hi->hub_type = uv_hub_info->hub_type;
+ hi->pnode_mask = uv_cpuid.pnode_mask;
+ hi->nasid_shift = uv_cpuid.nasid_shift;
+ hi->min_pnode = _min_pnode;
+ hi->min_socket = _min_socket;
+ hi->node_to_socket = _node_to_socket;
+ hi->pnode_to_socket = _pnode_to_socket;
+ hi->socket_to_node = _socket_to_node;
+ hi->socket_to_pnode = _socket_to_pnode;
+ hi->gr_table_len = _gr_table_len;
+ hi->gr_table = _gr_table;
+
+ uv_cpuid.gnode_shift = max_t(unsigned int, uv_cpuid.gnode_shift, mn.n_val);
+ hi->gnode_extra = (uv_node_id & ~((1 << uv_cpuid.gnode_shift) - 1)) >> 1;
+ if (mn.m_val)
+ hi->gnode_upper = (u64)hi->gnode_extra << mn.m_val;
+
+ if (uv_gp_table) {
+ hi->global_mmr_base = uv_gp_table->mmr_base;
+ hi->global_mmr_shift = uv_gp_table->mmr_shift;
+ hi->global_gru_base = uv_gp_table->gru_base;
+ hi->global_gru_shift = uv_gp_table->gru_shift;
+ hi->gpa_shift = uv_gp_table->gpa_shift;
+ hi->gpa_mask = (1UL << hi->gpa_shift) - 1;
+ } else {
+ hi->global_mmr_base =
+ uv_read_local_mmr(UVH_RH_GAM_MMR_OVERLAY_CONFIG) &
+ ~UV_MMR_ENABLE;
+ hi->global_mmr_shift = _UV_GLOBAL_MMR64_PNODE_SHIFT;
+ }
+
+ get_lowmem_redirect(&hi->lowmem_remap_base, &hi->lowmem_remap_top);
+
+ hi->apic_pnode_shift = uv_cpuid.socketid_shift;
+
+ /* Show system specific info: */
+ pr_info("UV: N:%d M:%d m_shift:%d n_lshift:%d\n", hi->n_val, hi->m_val, hi->m_shift, hi->n_lshift);
+ pr_info("UV: gpa_mask/shift:0x%lx/%d pnode_mask:0x%x apic_pns:%d\n", hi->gpa_mask, hi->gpa_shift, hi->pnode_mask, hi->apic_pnode_shift);
+ pr_info("UV: mmr_base/shift:0x%lx/%ld\n", hi->global_mmr_base, hi->global_mmr_shift);
+ if (hi->global_gru_base)
+ pr_info("UV: gru_base/shift:0x%lx/%ld\n",
+ hi->global_gru_base, hi->global_gru_shift);
+
+ pr_info("UV: gnode_upper:0x%lx gnode_extra:0x%x\n", hi->gnode_upper, hi->gnode_extra);
+}
+
+static void __init decode_gam_params(unsigned long ptr)
+{
+ uv_gp_table = (struct uv_gam_parameters *)ptr;
+
+ pr_info("UV: GAM Params...\n");
+ pr_info("UV: mmr_base/shift:0x%llx/%d gru_base/shift:0x%llx/%d gpa_shift:%d\n",
+ uv_gp_table->mmr_base, uv_gp_table->mmr_shift,
+ uv_gp_table->gru_base, uv_gp_table->gru_shift,
+ uv_gp_table->gpa_shift);
+}
+
+static void __init decode_gam_rng_tbl(unsigned long ptr)
+{
+ struct uv_gam_range_entry *gre = (struct uv_gam_range_entry *)ptr;
+ unsigned long lgre = 0, gend = 0;
+ int index = 0;
+ int sock_min = INT_MAX, pnode_min = INT_MAX;
+ int sock_max = -1, pnode_max = -1;
+
+ uv_gre_table = gre;
+ for (; gre->type != UV_GAM_RANGE_TYPE_UNUSED; gre++) {
+ unsigned long size = ((unsigned long)(gre->limit - lgre)
+ << UV_GAM_RANGE_SHFT);
+ int order = 0;
+ char suffix[] = " KMGTPE";
+ int flag = ' ';
+
+ while (size > 9999 && order < sizeof(suffix)) {
+ size /= 1024;
+ order++;
+ }
+
+ /* adjust max block size to current range start */
+ if (gre->type == 1 || gre->type == 2)
+ if (adj_blksize(lgre))
+ flag = '*';
+
+ if (!index) {
+ pr_info("UV: GAM Range Table...\n");
+ pr_info("UV: # %20s %14s %6s %4s %5s %3s %2s\n", "Range", "", "Size", "Type", "NASID", "SID", "PN");
+ }
+ pr_info("UV: %2d: 0x%014lx-0x%014lx%c %5lu%c %3d %04x %02x %02x\n",
+ index++,
+ (unsigned long)lgre << UV_GAM_RANGE_SHFT,
+ (unsigned long)gre->limit << UV_GAM_RANGE_SHFT,
+ flag, size, suffix[order],
+ gre->type, gre->nasid, gre->sockid, gre->pnode);
+
+ if (gre->type == UV_GAM_RANGE_TYPE_HOLE)
+ gend = (unsigned long)gre->limit << UV_GAM_RANGE_SHFT;
+
+ /* update to next range start */
+ lgre = gre->limit;
+ if (sock_min > gre->sockid)
+ sock_min = gre->sockid;
+ if (sock_max < gre->sockid)
+ sock_max = gre->sockid;
+ if (pnode_min > gre->pnode)
+ pnode_min = gre->pnode;
+ if (pnode_max < gre->pnode)
+ pnode_max = gre->pnode;
+ }
+ _min_socket = sock_min;
+ _max_socket = sock_max;
+ _min_pnode = pnode_min;
+ _max_pnode = pnode_max;
+ _gr_table_len = index;
+
+ pr_info("UV: GRT: %d entries, sockets(min:%x,max:%x), pnodes(min:%x,max:%x), gap_end(%d)\n",
+ index, _min_socket, _max_socket, _min_pnode, _max_pnode, fls64(gend));
+}
+
+/* Walk through UVsystab decoding the fields */
+static int __init decode_uv_systab(void)
+{
+ struct uv_systab *st;
+ int i;
+
+ /* Get mapped UVsystab pointer */
+ st = uv_systab;
+
+ /* If UVsystab is version 1, there is no extended UVsystab */
+ if (st && st->revision == UV_SYSTAB_VERSION_1)
+ return 0;
+
+ if ((!st) || (st->revision < UV_SYSTAB_VERSION_UV4_LATEST)) {
+ int rev = st ? st->revision : 0;
+
+ pr_err("UV: BIOS UVsystab mismatch, (%x < %x)\n",
+ rev, UV_SYSTAB_VERSION_UV4_LATEST);
+ pr_err("UV: Does not support UV, switch to non-UV x86_64\n");
+ uv_system_type = UV_NONE;
+
+ return -EINVAL;
+ }
+
+ for (i = 0; st->entry[i].type != UV_SYSTAB_TYPE_UNUSED; i++) {
+ unsigned long ptr = st->entry[i].offset;
+
+ if (!ptr)
+ continue;
+
+ /* point to payload */
+ ptr += (unsigned long)st;
+
+ switch (st->entry[i].type) {
+ case UV_SYSTAB_TYPE_GAM_PARAMS:
+ decode_gam_params(ptr);
+ break;
+
+ case UV_SYSTAB_TYPE_GAM_RNG_TBL:
+ decode_gam_rng_tbl(ptr);
+ break;
+
+ case UV_SYSTAB_TYPE_ARCH_TYPE:
+ /* already processed in early startup */
+ break;
+
+ default:
+ pr_err("UV:%s:Unrecognized UV_SYSTAB_TYPE:%d, skipped\n",
+ __func__, st->entry[i].type);
+ break;
+ }
+ }
+ return 0;
+}
+
+/*
+ * Given a bitmask 'bits' representing presnt blades, numbered
+ * starting at 'base', masking off unused high bits of blade number
+ * with 'mask', update the minimum and maximum blade numbers that we
+ * have found. (Masking with 'mask' necessary because of BIOS
+ * treatment of system partitioning when creating this table we are
+ * interpreting.)
+ */
+static inline void blade_update_min_max(unsigned long bits, int base, int mask, int *min, int *max)
+{
+ int first, last;
+
+ if (!bits)
+ return;
+ first = (base + __ffs(bits)) & mask;
+ last = (base + __fls(bits)) & mask;
+
+ if (*min > first)
+ *min = first;
+ if (*max < last)
+ *max = last;
+}
+
+/* Set up physical blade translations from UVH_NODE_PRESENT_TABLE */
+static __init void boot_init_possible_blades(struct uv_hub_info_s *hub_info)
+{
+ unsigned long np;
+ int i, uv_pb = 0;
+ int sock_min = INT_MAX, sock_max = -1, s_mask;
+
+ s_mask = (1 << uv_cpuid.n_skt) - 1;
+
+ if (UVH_NODE_PRESENT_TABLE) {
+ pr_info("UV: NODE_PRESENT_DEPTH = %d\n",
+ UVH_NODE_PRESENT_TABLE_DEPTH);
+ for (i = 0; i < UVH_NODE_PRESENT_TABLE_DEPTH; i++) {
+ np = uv_read_local_mmr(UVH_NODE_PRESENT_TABLE + i * 8);
+ pr_info("UV: NODE_PRESENT(%d) = 0x%016lx\n", i, np);
+ blade_update_min_max(np, i * 64, s_mask, &sock_min, &sock_max);
+ }
+ }
+ if (UVH_NODE_PRESENT_0) {
+ np = uv_read_local_mmr(UVH_NODE_PRESENT_0);
+ pr_info("UV: NODE_PRESENT_0 = 0x%016lx\n", np);
+ blade_update_min_max(np, 0, s_mask, &sock_min, &sock_max);
+ }
+ if (UVH_NODE_PRESENT_1) {
+ np = uv_read_local_mmr(UVH_NODE_PRESENT_1);
+ pr_info("UV: NODE_PRESENT_1 = 0x%016lx\n", np);
+ blade_update_min_max(np, 64, s_mask, &sock_min, &sock_max);
+ }
+
+ /* Only update if we actually found some bits indicating blades present */
+ if (sock_max >= sock_min) {
+ _min_socket = sock_min;
+ _max_socket = sock_max;
+ uv_pb = sock_max - sock_min + 1;
+ }
+ if (uv_possible_blades != uv_pb)
+ uv_possible_blades = uv_pb;
+
+ pr_info("UV: number nodes/possible blades %d (%d - %d)\n",
+ uv_pb, sock_min, sock_max);
+}
+
+static int __init alloc_conv_table(int num_elem, unsigned short **table)
+{
+ int i;
+ size_t bytes;
+
+ bytes = num_elem * sizeof(*table[0]);
+ *table = kmalloc(bytes, GFP_KERNEL);
+ if (WARN_ON_ONCE(!*table))
+ return -ENOMEM;
+ for (i = 0; i < num_elem; i++)
+ ((unsigned short *)*table)[i] = SOCK_EMPTY;
+ return 0;
+}
+
+/* Remove conversion table if it's 1:1 */
+#define FREE_1_TO_1_TABLE(tbl, min, max, max2) free_1_to_1_table(&tbl, #tbl, min, max, max2)
+
+static void __init free_1_to_1_table(unsigned short **tp, char *tname, int min, int max, int max2)
+{
+ int i;
+ unsigned short *table = *tp;
+
+ if (table == NULL)
+ return;
+ if (max != max2)
+ return;
+ for (i = 0; i < max; i++) {
+ if (i != table[i])
+ return;
+ }
+ kfree(table);
+ *tp = NULL;
+ pr_info("UV: %s is 1:1, conversion table removed\n", tname);
+}
+
+/*
+ * Build Socket Tables
+ * If the number of nodes is >1 per socket, socket to node table will
+ * contain lowest node number on that socket.
+ */
+static void __init build_socket_tables(void)
+{
+ struct uv_gam_range_entry *gre = uv_gre_table;
+ int nums, numn, nump;
+ int i, lnid, apicid;
+ int minsock = _min_socket;
+ int maxsock = _max_socket;
+ int minpnode = _min_pnode;
+ int maxpnode = _max_pnode;
+
+ if (!gre) {
+ if (is_uv2_hub() || is_uv3_hub()) {
+ pr_info("UV: No UVsystab socket table, ignoring\n");
+ return;
+ }
+ pr_err("UV: Error: UVsystab address translations not available!\n");
+ WARN_ON_ONCE(!gre);
+ return;
+ }
+
+ numn = num_possible_nodes();
+ nump = maxpnode - minpnode + 1;
+ nums = maxsock - minsock + 1;
+
+ /* Allocate and clear tables */
+ if ((alloc_conv_table(nump, &_pnode_to_socket) < 0)
+ || (alloc_conv_table(nums, &_socket_to_pnode) < 0)
+ || (alloc_conv_table(numn, &_node_to_socket) < 0)
+ || (alloc_conv_table(nums, &_socket_to_node) < 0)) {
+ kfree(_pnode_to_socket);
+ kfree(_socket_to_pnode);
+ kfree(_node_to_socket);
+ return;
+ }
+
+ /* Fill in pnode/node/addr conversion list values: */
+ for (; gre->type != UV_GAM_RANGE_TYPE_UNUSED; gre++) {
+ if (gre->type == UV_GAM_RANGE_TYPE_HOLE)
+ continue;
+ i = gre->sockid - minsock;
+ if (_socket_to_pnode[i] == SOCK_EMPTY)
+ _socket_to_pnode[i] = gre->pnode;
+
+ i = gre->pnode - minpnode;
+ if (_pnode_to_socket[i] == SOCK_EMPTY)
+ _pnode_to_socket[i] = gre->sockid;
+
+ pr_info("UV: sid:%02x type:%d nasid:%04x pn:%02x pn2s:%2x\n",
+ gre->sockid, gre->type, gre->nasid,
+ _socket_to_pnode[gre->sockid - minsock],
+ _pnode_to_socket[gre->pnode - minpnode]);
+ }
+
+ /* Set socket -> node values: */
+ lnid = NUMA_NO_NODE;
+ for (apicid = 0; apicid < ARRAY_SIZE(__apicid_to_node); apicid++) {
+ int nid = __apicid_to_node[apicid];
+ int sockid;
+
+ if ((nid == NUMA_NO_NODE) || (lnid == nid))
+ continue;
+ lnid = nid;
+
+ sockid = apicid >> uv_cpuid.socketid_shift;
+
+ if (_socket_to_node[sockid - minsock] == SOCK_EMPTY)
+ _socket_to_node[sockid - minsock] = nid;
+
+ if (_node_to_socket[nid] == SOCK_EMPTY)
+ _node_to_socket[nid] = sockid;
+
+ pr_info("UV: sid:%02x: apicid:%04x socket:%02d node:%03x s2n:%03x\n",
+ sockid,
+ apicid,
+ _node_to_socket[nid],
+ nid,
+ _socket_to_node[sockid - minsock]);
+ }
+
+ /*
+ * If e.g. socket id == pnode for all pnodes,
+ * system runs faster by removing corresponding conversion table.
+ */
+ FREE_1_TO_1_TABLE(_socket_to_node, _min_socket, nums, numn);
+ FREE_1_TO_1_TABLE(_node_to_socket, _min_socket, nums, numn);
+ FREE_1_TO_1_TABLE(_socket_to_pnode, _min_pnode, nums, nump);
+ FREE_1_TO_1_TABLE(_pnode_to_socket, _min_pnode, nums, nump);
+}
+
+/* Check which reboot to use */
+static void check_efi_reboot(void)
+{
+ /* If EFI reboot not available, use ACPI reboot */
+ if (!efi_enabled(EFI_BOOT))
+ reboot_type = BOOT_ACPI;
+}
+
+/*
+ * User proc fs file handling now deprecated.
+ * Recommend using /sys/firmware/sgi_uv/... instead.
+ */
+static int __maybe_unused proc_hubbed_show(struct seq_file *file, void *data)
+{
+ pr_notice_once("%s: using deprecated /proc/sgi_uv/hubbed, use /sys/firmware/sgi_uv/hub_type\n",
+ current->comm);
+ seq_printf(file, "0x%x\n", uv_hubbed_system);
+ return 0;
+}
+
+static int __maybe_unused proc_hubless_show(struct seq_file *file, void *data)
+{
+ pr_notice_once("%s: using deprecated /proc/sgi_uv/hubless, use /sys/firmware/sgi_uv/hubless\n",
+ current->comm);
+ seq_printf(file, "0x%x\n", uv_hubless_system);
+ return 0;
+}
+
+static int __maybe_unused proc_archtype_show(struct seq_file *file, void *data)
+{
+ pr_notice_once("%s: using deprecated /proc/sgi_uv/archtype, use /sys/firmware/sgi_uv/archtype\n",
+ current->comm);
+ seq_printf(file, "%s/%s\n", uv_archtype, oem_table_id);
+ return 0;
+}
+
+static __init void uv_setup_proc_files(int hubless)
+{
+ struct proc_dir_entry *pde;
+
+ pde = proc_mkdir(UV_PROC_NODE, NULL);
+ proc_create_single("archtype", 0, pde, proc_archtype_show);
+ if (hubless)
+ proc_create_single("hubless", 0, pde, proc_hubless_show);
+ else
+ proc_create_single("hubbed", 0, pde, proc_hubbed_show);
+}
+
+/* Initialize UV hubless systems */
+static __init int uv_system_init_hubless(void)
+{
+ int rc;
+
+ /* Setup PCH NMI handler */
+ uv_nmi_setup_hubless();
+
+ /* Init kernel/BIOS interface */
+ rc = uv_bios_init();
+ if (rc < 0)
+ return rc;
+
+ /* Process UVsystab */
+ rc = decode_uv_systab();
+ if (rc < 0)
+ return rc;
+
+ /* Set section block size for current node memory */
+ set_block_size();
+
+ /* Create user access node */
+ if (rc >= 0)
+ uv_setup_proc_files(1);
+
+ check_efi_reboot();
+
+ return rc;
+}
+
+static void __init uv_system_init_hub(void)
+{
+ struct uv_hub_info_s hub_info = {0};
+ int bytes, cpu, nodeid, bid;
+ unsigned short min_pnode = USHRT_MAX, max_pnode = 0;
+ char *hub = is_uv5_hub() ? "UV500" :
+ is_uv4_hub() ? "UV400" :
+ is_uv3_hub() ? "UV300" :
+ is_uv2_hub() ? "UV2000/3000" : NULL;
+ struct uv_hub_info_s **uv_hub_info_list_blade;
+
+ if (!hub) {
+ pr_err("UV: Unknown/unsupported UV hub\n");
+ return;
+ }
+ pr_info("UV: Found %s hub\n", hub);
+
+ map_low_mmrs();
+
+ /* Get uv_systab for decoding, setup UV BIOS calls */
+ uv_bios_init();
+
+ /* If there's an UVsystab problem then abort UV init: */
+ if (decode_uv_systab() < 0) {
+ pr_err("UV: Mangled UVsystab format\n");
+ return;
+ }
+
+ build_socket_tables();
+ build_uv_gr_table();
+ set_block_size();
+ uv_init_hub_info(&hub_info);
+ /* If UV2 or UV3 may need to get # blades from HW */
+ if (is_uv(UV2|UV3) && !uv_gre_table)
+ boot_init_possible_blades(&hub_info);
+ else
+ /* min/max sockets set in decode_gam_rng_tbl */
+ uv_possible_blades = (_max_socket - _min_socket) + 1;
+
+ /* uv_num_possible_blades() is really the hub count: */
+ pr_info("UV: Found %d hubs, %d nodes, %d CPUs\n", uv_num_possible_blades(), num_possible_nodes(), num_possible_cpus());
+
+ uv_bios_get_sn_info(0, &uv_type, &sn_partition_id, &sn_coherency_id, &sn_region_size, &system_serial_number);
+ hub_info.coherency_domain_number = sn_coherency_id;
+ uv_rtc_init();
+
+ /*
+ * __uv_hub_info_list[] is indexed by node, but there is only
+ * one hub_info structure per blade. First, allocate one
+ * structure per blade. Further down we create a per-node
+ * table (__uv_hub_info_list[]) pointing to hub_info
+ * structures for the correct blade.
+ */
+
+ bytes = sizeof(void *) * uv_num_possible_blades();
+ uv_hub_info_list_blade = kzalloc(bytes, GFP_KERNEL);
+ if (WARN_ON_ONCE(!uv_hub_info_list_blade))
+ return;
+
+ bytes = sizeof(struct uv_hub_info_s);
+ for_each_possible_blade(bid) {
+ struct uv_hub_info_s *new_hub;
+
+ /* Allocate & fill new per hub info list */
+ new_hub = (bid == 0) ? &uv_hub_info_node0
+ : kzalloc_node(bytes, GFP_KERNEL, uv_blade_to_node(bid));
+ if (WARN_ON_ONCE(!new_hub)) {
+ /* do not kfree() bid 0, which is statically allocated */
+ while (--bid > 0)
+ kfree(uv_hub_info_list_blade[bid]);
+ kfree(uv_hub_info_list_blade);
+ return;
+ }
+
+ uv_hub_info_list_blade[bid] = new_hub;
+ *new_hub = hub_info;
+
+ /* Use information from GAM table if available: */
+ if (uv_gre_table)
+ new_hub->pnode = uv_blade_to_pnode(bid);
+ else /* Or fill in during CPU loop: */
+ new_hub->pnode = 0xffff;
+
+ new_hub->numa_blade_id = bid;
+ new_hub->memory_nid = NUMA_NO_NODE;
+ new_hub->nr_possible_cpus = 0;
+ new_hub->nr_online_cpus = 0;
+ }
+
+ /*
+ * Now populate __uv_hub_info_list[] for each node with the
+ * pointer to the struct for the blade it resides on.
+ */
+
+ bytes = sizeof(void *) * num_possible_nodes();
+ __uv_hub_info_list = kzalloc(bytes, GFP_KERNEL);
+ if (WARN_ON_ONCE(!__uv_hub_info_list)) {
+ for_each_possible_blade(bid)
+ /* bid 0 is statically allocated */
+ if (bid != 0)
+ kfree(uv_hub_info_list_blade[bid]);
+ kfree(uv_hub_info_list_blade);
+ return;
+ }
+
+ for_each_node(nodeid)
+ __uv_hub_info_list[nodeid] = uv_hub_info_list_blade[uv_node_to_blade_id(nodeid)];
+
+ /* Initialize per CPU info: */
+ for_each_possible_cpu(cpu) {
+ int apicid = per_cpu(x86_cpu_to_apicid, cpu);
+ unsigned short bid;
+ unsigned short pnode;
+
+ pnode = uv_apicid_to_pnode(apicid);
+ bid = uv_pnode_to_socket(pnode) - _min_socket;
+
+ uv_cpu_info_per(cpu)->p_uv_hub_info = uv_hub_info_list_blade[bid];
+ uv_cpu_info_per(cpu)->blade_cpu_id = uv_cpu_hub_info(cpu)->nr_possible_cpus++;
+ if (uv_cpu_hub_info(cpu)->memory_nid == NUMA_NO_NODE)
+ uv_cpu_hub_info(cpu)->memory_nid = cpu_to_node(cpu);
+
+ if (uv_cpu_hub_info(cpu)->pnode == 0xffff)
+ uv_cpu_hub_info(cpu)->pnode = pnode;
+ }
+
+ for_each_possible_blade(bid) {
+ unsigned short pnode = uv_hub_info_list_blade[bid]->pnode;
+
+ if (pnode == 0xffff)
+ continue;
+
+ min_pnode = min(pnode, min_pnode);
+ max_pnode = max(pnode, max_pnode);
+ pr_info("UV: HUB:%2d pn:%02x nrcpus:%d\n",
+ bid,
+ uv_hub_info_list_blade[bid]->pnode,
+ uv_hub_info_list_blade[bid]->nr_possible_cpus);
+ }
+
+ pr_info("UV: min_pnode:%02x max_pnode:%02x\n", min_pnode, max_pnode);
+ map_gru_high(max_pnode);
+ map_mmr_high(max_pnode);
+ map_mmioh_high(min_pnode, max_pnode);
+
+ kfree(uv_hub_info_list_blade);
+ uv_hub_info_list_blade = NULL;
+
+ uv_nmi_setup();
+ uv_cpu_init();
+ uv_setup_proc_files(0);
+
+ /* Register Legacy VGA I/O redirection handler: */
+ pci_register_set_vga_state(uv_set_vga_state);
+
+ check_efi_reboot();
+}
+
+/*
+ * There is a different code path needed to initialize a UV system that does
+ * not have a "UV HUB" (referred to as "hubless").
+ */
+void __init uv_system_init(void)
+{
+ if (likely(!is_uv_system() && !is_uv_hubless(1)))
+ return;
+
+ if (is_uv_system())
+ uv_system_init_hub();
+ else
+ uv_system_init_hubless();
+}
+
+apic_driver(apic_x2apic_uv_x);