int __init check_nmi_watchdog (void)
{
int *counts;
int cpu;
if ((nmi_watchdog == NMI_NONE) || (nmi_watchdog == NMI_DISABLED))
return 0;
if (!atomic_read(&nmi_active))
return 0;
counts = kmalloc(NR_CPUS * sizeof(int), GFP_KERNEL);
if (!counts)
return -1;
printk(KERN_INFO "testing NMI watchdog ... ");
#ifdef CONFIG_SMP
if (nmi_watchdog == NMI_LOCAL_APIC)
smp_call_function(nmi_cpu_busy, (void *)&endflag, 0, 0);
#endif
for (cpu = 0; cpu < NR_CPUS; cpu++)
counts[cpu] = cpu_pda(cpu)->__nmi_count;
local_irq_enable();
mdelay((20*1000)/nmi_hz); // wait 20 ticks
for_each_online_cpu(cpu) {
if (!per_cpu(wd_enabled, cpu))
continue;
if (cpu_pda(cpu)->__nmi_count - counts[cpu] <= 5) {
printk(KERN_WARNING "WARNING: CPU#%d: NMI "
"appears to be stuck (%d->%d)!\n",
cpu,
counts[cpu],
cpu_pda(cpu)->__nmi_count);
per_cpu(wd_enabled, cpu) = 0;
atomic_dec(&nmi_active);
}
}
if (!atomic_read(&nmi_active)) {
kfree(counts);
atomic_set(&nmi_active, -1);
endflag = 1;
return -1;
}
endflag = 1;
printk("OK.\n");
/* now that we know it works we can reduce NMI frequency to
something more reasonable; makes a difference in some configs */
if (nmi_watchdog == NMI_LOCAL_APIC)
nmi_hz = lapic_adjust_nmi_hz(1);
kfree(counts);
return 0;
}
/*
* Great future plan:
* Declare PDA itself and support (irqstack,tss,pgd) as per cpu data.
* Always point %gs to its beginning
*/
void __init setup_per_cpu_areas(void)
{
int i;
unsigned long size;
#ifdef CONFIG_HOTPLUG_CPU
prefill_possible_map();
#endif
/* Copy section for each CPU (we discard the original) */
size = PERCPU_ENOUGH_ROOM;
printk(KERN_INFO "PERCPU: Allocating %lu bytes of per cpu data\n", size);
for_each_cpu_mask (i, cpu_possible_map) {
char *ptr;
if (!NODE_DATA(cpu_to_node(i))) {
printk("cpu with no node %d, num_online_nodes %d\n",
i, num_online_nodes());
ptr = alloc_bootmem(size);
} else {
ptr = alloc_bootmem_node(NODE_DATA(cpu_to_node(i)), size);
}
if (!ptr)
panic("Cannot allocate cpu data for CPU %d\n", i);
cpu_pda(i)->data_offset = ptr - __per_cpu_start;
memcpy(ptr, __per_cpu_start, __per_cpu_end - __per_cpu_start);
}
/**
* This initializes a per-CPU area for each CPU.
*
* TODO: The PDA and per-CPU areas are pretty tightly wound. It should be
* possible to make the per-CPU area *be* the PDA, or put another way,
* point %GS at the per-CPU area rather than the PDA. All of the PDA's
* current contents would become normal per-CPU variables.
*/
static void __init
setup_per_cpu_areas(void)
{
int i;
size_t size;
/*
* There is an ELF section containing all per-CPU variables
* surrounded by __per_cpu_start and __per_cpu_end symbols.
* We create a copy of this ELF section for each CPU.
*/
size = ALIGN(__per_cpu_end - __per_cpu_start, SMP_CACHE_BYTES);
for_each_cpu_mask (i, cpu_present_map) {
char *ptr;
ptr = alloc_bootmem_aligned(size, PAGE_SIZE);
if (!ptr)
panic("Cannot allocate cpu data for CPU %d\n", i);
/*
* Pre-bias data_offset by subtracting its offset from
* __per_cpu_start. Later, per_cpu() will calculate a
* per_cpu variable's address with:
*
* addr = offset_in_percpu_ELF_section + data_offset
* = (__per_cpu_start + offset) + (ptr - __per_cpu_start)
* = offset + ptr
*/
cpu_pda(i)->data_offset = ptr - __per_cpu_start;
memcpy(ptr, __per_cpu_start, __per_cpu_end - __per_cpu_start);
}
void __init x86_64_start_kernel(char * real_mode_data)
{
int i;
/* clear bss before set_intr_gate with early_idt_handler */
clear_bss();
/* Make NULL pointers segfault */
zap_identity_mappings();
for (i = 0; i < IDT_ENTRIES; i++)
set_intr_gate(i, early_idt_handler);
load_idt((const struct desc_ptr *)&idt_descr);
early_printk("Kernel alive\n");
for (i = 0; i < NR_CPUS; i++)
cpu_pda(i) = &boot_cpu_pda[i];
pda_init(0);
copy_bootdata(__va(real_mode_data));
#ifdef CONFIG_SMP
cpu_set(0, cpu_online_map);
#endif
start_kernel();
}
void __cpuinit numa_set_node(int cpu, int node)
{
int *cpu_to_node_map = early_per_cpu_ptr(x86_cpu_to_node_map);
if (cpu_pda(cpu) && node != NUMA_NO_NODE)
cpu_pda(cpu)->nodenumber = node;
if (cpu_to_node_map)
cpu_to_node_map[cpu] = node;
else if (per_cpu_offset(cpu))
per_cpu(x86_cpu_to_node_map, cpu) = node;
else
Dprintk(KERN_INFO "Setting node for non-present cpu %d\n", cpu);
}
/*
* Allocate cpu_pda pointer table and array via alloc_bootmem.
*/
static void __init setup_cpu_pda_map(void)
{
char *pda;
struct x8664_pda **new_cpu_pda;
unsigned long size;
int cpu;
size = roundup(sizeof(struct x8664_pda), cache_line_size());
/* allocate cpu_pda array and pointer table */
{
unsigned long tsize = nr_cpu_ids * sizeof(void *);
unsigned long asize = size * (nr_cpu_ids - 1);
tsize = roundup(tsize, cache_line_size());
new_cpu_pda = alloc_bootmem(tsize + asize);
pda = (char *)new_cpu_pda + tsize;
}
/* initialize pointer table to static pda's */
for_each_possible_cpu(cpu) {
if (cpu == 0) {
/* leave boot cpu pda in place */
new_cpu_pda[0] = cpu_pda(0);
continue;
}
new_cpu_pda[cpu] = (struct x8664_pda *)pda;
new_cpu_pda[cpu]->in_bootmem = 1;
pda += size;
}
/* point to new pointer table */
_cpu_pda = new_cpu_pda;
}
int show_interrupts(struct seq_file *p, void *v)
{
int i = *(loff_t *) v, j;
struct irqaction * action;
unsigned long flags;
if (i == 0) {
seq_printf(p, " ");
for_each_online_cpu(j)
seq_printf(p, "CPU%-8d",j);
seq_putc(p, '\n');
}
if (i < NR_IRQS) {
spin_lock_irqsave(&irq_desc[i].lock, flags);
action = irq_desc[i].action;
if (!action)
goto skip;
seq_printf(p, "%3d: ",i);
#ifndef CONFIG_SMP
seq_printf(p, "%10u ", kstat_irqs(i));
#else
for_each_online_cpu(j)
seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]);
#endif
seq_printf(p, " %8s", irq_desc[i].chip->name);
seq_printf(p, "-%-8s", irq_desc[i].name);
seq_printf(p, " %s", action->name);
for (action=action->next; action; action = action->next)
seq_printf(p, ", %s", action->name);
seq_putc(p, '\n');
skip:
spin_unlock_irqrestore(&irq_desc[i].lock, flags);
} else if (i == NR_IRQS) {
seq_printf(p, "NMI: ");
for_each_online_cpu(j)
seq_printf(p, "%10u ", cpu_pda(j)->__nmi_count);
seq_putc(p, '\n');
seq_printf(p, "LOC: ");
for_each_online_cpu(j)
seq_printf(p, "%10u ", cpu_pda(j)->apic_timer_irqs);
seq_putc(p, '\n');
seq_printf(p, "ERR: %10u\n", atomic_read(&irq_err_count));
}
return 0;
}
int __init check_nmi_watchdog (void)
{
volatile int endflag = 0;
int *counts;
int cpu;
counts = kmalloc(NR_CPUS * sizeof(int), GFP_KERNEL);
if (!counts)
return -1;
printk(KERN_INFO "testing NMI watchdog ... ");
#ifdef CONFIG_SMP
if (nmi_watchdog == NMI_LOCAL_APIC)
smp_call_function(nmi_cpu_busy, (void *)&endflag, 0, 0);
#endif
for (cpu = 0; cpu < NR_CPUS; cpu++)
counts[cpu] = cpu_pda(cpu)->__nmi_count;
local_irq_enable();
mdelay((10*1000)/nmi_hz); // wait 10 ticks
for_each_online_cpu(cpu) {
if (cpu_pda(cpu)->__nmi_count - counts[cpu] <= 5) {
endflag = 1;
printk("CPU#%d: NMI appears to be stuck (%d->%d)!\n",
cpu,
counts[cpu],
cpu_pda(cpu)->__nmi_count);
nmi_active = 0;
lapic_nmi_owner &= ~LAPIC_NMI_WATCHDOG;
nmi_perfctr_msr = 0;
kfree(counts);
return -1;
}
}
endflag = 1;
printk("OK.\n");
/* now that we know it works we can reduce NMI frequency to
something more reasonable; makes a difference in some configs */
if (nmi_watchdog == NMI_LOCAL_APIC)
nmi_hz = 1;
kfree(counts);
return 0;
}
static int __cpuinit xen_cpu_up(unsigned int cpu)
{
struct task_struct *idle = idle_task(cpu);
int rc;
#ifdef CONFIG_X86_64
/* Allocate node local memory for AP pdas */
WARN_ON(cpu == 0);
if (cpu > 0) {
rc = get_local_pda(cpu);
if (rc)
return rc;
}
#endif
#ifdef CONFIG_X86_32
init_gdt(cpu);
per_cpu(current_task, cpu) = idle;
irq_ctx_init(cpu);
#else
cpu_pda(cpu)->pcurrent = idle;
clear_tsk_thread_flag(idle, TIF_FORK);
#endif
xen_setup_timer(cpu);
xen_init_lock_cpu(cpu);
per_cpu(cpu_state, cpu) = CPU_UP_PREPARE;
/* make sure interrupts start blocked */
per_cpu(xen_vcpu, cpu)->evtchn_upcall_mask = 1;
rc = cpu_initialize_context(cpu, idle);
if (rc)
return rc;
if (num_online_cpus() == 1)
alternatives_smp_switch(1);
rc = xen_smp_intr_init(cpu);
if (rc)
return rc;
rc = HYPERVISOR_vcpu_op(VCPUOP_up, cpu, NULL);
BUG_ON(rc);
while(per_cpu(cpu_state, cpu) != CPU_ONLINE) {
HYPERVISOR_sched_op(SCHEDOP_yield, 0);
barrier();
}
return 0;
}
static void
cpu_initialize_context(unsigned int cpu)
{
/* vcpu_guest_context_t is too large to allocate on the stack.
* Hence we allocate statically and protect it with a lock */
vm_page_t m[NPGPTD + 2];
static vcpu_guest_context_t ctxt;
vm_offset_t boot_stack;
vm_offset_t newPTD;
vm_paddr_t ma[NPGPTD];
int i;
/*
* Page 0,[0-3] PTD
* Page 1, [4] boot stack
* Page [5] PDPT
*
*/
for (i = 0; i < NPGPTD + 2; i++) {
m[i] = vm_page_alloc(NULL, 0,
VM_ALLOC_NORMAL | VM_ALLOC_NOOBJ | VM_ALLOC_WIRED |
VM_ALLOC_ZERO);
pmap_zero_page(m[i]);
}
boot_stack = kmem_alloc_nofault(kernel_map, PAGE_SIZE);
newPTD = kmem_alloc_nofault(kernel_map, NPGPTD * PAGE_SIZE);
ma[0] = VM_PAGE_TO_MACH(m[0])|PG_V;
#ifdef PAE
pmap_kenter(boot_stack, VM_PAGE_TO_PHYS(m[NPGPTD + 1]));
for (i = 0; i < NPGPTD; i++) {
((vm_paddr_t *)boot_stack)[i] =
ma[i] = VM_PAGE_TO_MACH(m[i])|PG_V;
}
#endif
/*
* Copy cpu0 IdlePTD to new IdlePTD - copying only
* kernel mappings
*/
pmap_qenter(newPTD, m, 4);
memcpy((uint8_t *)newPTD + KPTDI*sizeof(vm_paddr_t),
(uint8_t *)PTOV(IdlePTD) + KPTDI*sizeof(vm_paddr_t),
nkpt*sizeof(vm_paddr_t));
pmap_qremove(newPTD, 4);
kmem_free(kernel_map, newPTD, 4 * PAGE_SIZE);
/*
* map actual idle stack to boot_stack
*/
pmap_kenter(boot_stack, VM_PAGE_TO_PHYS(m[NPGPTD]));
xen_pgdpt_pin(VM_PAGE_TO_MACH(m[NPGPTD + 1]));
rw_wlock(&pvh_global_lock);
for (i = 0; i < 4; i++) {
int pdir = (PTDPTDI + i) / NPDEPG;
int curoffset = (PTDPTDI + i) % NPDEPG;
xen_queue_pt_update((vm_paddr_t)
((ma[pdir] & ~PG_V) + (curoffset*sizeof(vm_paddr_t))),
ma[i]);
}
PT_UPDATES_FLUSH();
rw_wunlock(&pvh_global_lock);
memset(&ctxt, 0, sizeof(ctxt));
ctxt.flags = VGCF_IN_KERNEL;
ctxt.user_regs.ds = GSEL(GDATA_SEL, SEL_KPL);
ctxt.user_regs.es = GSEL(GDATA_SEL, SEL_KPL);
ctxt.user_regs.fs = GSEL(GPRIV_SEL, SEL_KPL);
ctxt.user_regs.gs = GSEL(GDATA_SEL, SEL_KPL);
ctxt.user_regs.cs = GSEL(GCODE_SEL, SEL_KPL);
ctxt.user_regs.ss = GSEL(GDATA_SEL, SEL_KPL);
ctxt.user_regs.eip = (unsigned long)init_secondary;
ctxt.user_regs.eflags = PSL_KERNEL | 0x1000; /* IOPL_RING1 */
memset(&ctxt.fpu_ctxt, 0, sizeof(ctxt.fpu_ctxt));
smp_trap_init(ctxt.trap_ctxt);
ctxt.ldt_ents = 0;
ctxt.gdt_frames[0] = (uint32_t)((uint64_t)vtomach(bootAPgdt) >> PAGE_SHIFT);
ctxt.gdt_ents = 512;
#ifdef __i386__
ctxt.user_regs.esp = boot_stack + PAGE_SIZE;
ctxt.kernel_ss = GSEL(GDATA_SEL, SEL_KPL);
ctxt.kernel_sp = boot_stack + PAGE_SIZE;
ctxt.event_callback_cs = GSEL(GCODE_SEL, SEL_KPL);
ctxt.event_callback_eip = (unsigned long)Xhypervisor_callback;
ctxt.failsafe_callback_cs = GSEL(GCODE_SEL, SEL_KPL);
ctxt.failsafe_callback_eip = (unsigned long)failsafe_callback;
ctxt.ctrlreg[3] = VM_PAGE_TO_MACH(m[NPGPTD + 1]);
#else /* __x86_64__ */
ctxt.user_regs.esp = idle->thread.rsp0 - sizeof(struct pt_regs);
ctxt.kernel_ss = GSEL(GDATA_SEL, SEL_KPL);
ctxt.kernel_sp = idle->thread.rsp0;
ctxt.event_callback_eip = (unsigned long)hypervisor_callback;
ctxt.failsafe_callback_eip = (unsigned long)failsafe_callback;
ctxt.syscall_callback_eip = (unsigned long)system_call;
ctxt.ctrlreg[3] = xen_pfn_to_cr3(virt_to_mfn(init_level4_pgt));
ctxt.gs_base_kernel = (unsigned long)(cpu_pda(cpu));
#endif
printf("gdtpfn=%lx pdptpfn=%lx\n",
ctxt.gdt_frames[0],
ctxt.ctrlreg[3] >> PAGE_SHIFT);
PANIC_IF(HYPERVISOR_vcpu_op(VCPUOP_initialise, cpu, &ctxt));
DELAY(3000);
PANIC_IF(HYPERVISOR_vcpu_op(VCPUOP_up, cpu, NULL));
}
void __init x86_64_init_pda(void)
{
_cpu_pda = __cpu_pda;
cpu_pda(0) = &_boot_cpu_pda;
pda_init(0);
}
void __init x86_64_start_kernel(char * real_mode_data)
{
int i;
/*
* Build-time sanity checks on the kernel image and module
* area mappings. (these are purely build-time and produce no code)
*/
BUILD_BUG_ON(MODULES_VADDR < KERNEL_IMAGE_START);
BUILD_BUG_ON(MODULES_VADDR-KERNEL_IMAGE_START < KERNEL_IMAGE_SIZE);
BUILD_BUG_ON(MODULES_LEN + KERNEL_IMAGE_SIZE > 2*PUD_SIZE);
BUILD_BUG_ON((KERNEL_IMAGE_START & ~PMD_MASK) != 0);
BUILD_BUG_ON((MODULES_VADDR & ~PMD_MASK) != 0);
BUILD_BUG_ON(!(MODULES_VADDR > __START_KERNEL));
BUILD_BUG_ON(!(((MODULES_END - 1) & PGDIR_MASK) ==
(__START_KERNEL & PGDIR_MASK)));
/* clear bss before set_intr_gate with early_idt_handler */
clear_bss();
/* Make NULL pointers segfault */
zap_identity_mappings();
/* Cleanup the over mapped high alias */
cleanup_highmap();
for (i = 0; i < NUM_EXCEPTION_VECTORS; i++) {
#ifdef CONFIG_EARLY_PRINTK
set_intr_gate(i, &early_idt_handlers[i]);
#else
set_intr_gate(i, early_idt_handler);
#endif
}
load_idt((const struct desc_ptr *)&idt_descr);
early_printk("Kernel alive\n");
for (i = 0; i < NR_CPUS; i++)
cpu_pda(i) = &boot_cpu_pda[i];
pda_init(0);
copy_bootdata(__va(real_mode_data));
reserve_early(__pa_symbol(&_text), __pa_symbol(&_end), "TEXT DATA BSS");
#ifdef CONFIG_BLK_DEV_INITRD
/* Reserve INITRD */
if (boot_params.hdr.type_of_loader && boot_params.hdr.ramdisk_image) {
unsigned long ramdisk_image = boot_params.hdr.ramdisk_image;
unsigned long ramdisk_size = boot_params.hdr.ramdisk_size;
unsigned long ramdisk_end = ramdisk_image + ramdisk_size;
reserve_early(ramdisk_image, ramdisk_end, "RAMDISK");
}
#endif
reserve_ebda_region();
reserve_setup_data();
/*
* At this point everything still needed from the boot loader
* or BIOS or kernel text should be early reserved or marked not
* RAM in e820. All other memory is free game.
*/
start_kernel();
}
