* useful topology information for the kernel to make use of. As a
* result, all CPUs are treated as if they're single-core and
* single-threaded.
- *
- * This does not handle HOTPLUG_CPU yet.
*/
#include <linux/sched.h>
#include <linux/err.h>
+#include <linux/slab.h>
#include <linux/smp.h>
#include <asm/paravirt.h>
#include "xen-ops.h"
#include "mmu.h"
-static cpumask_t xen_cpu_initialized_map;
-static DEFINE_PER_CPU(int, resched_irq) = -1;
-static DEFINE_PER_CPU(int, callfunc_irq) = -1;
-static DEFINE_PER_CPU(int, debug_irq) = -1;
+cpumask_var_t xen_cpu_initialized_map;
-/*
- * Structure and data for smp_call_function(). This is designed to minimise
- * static memory requirements. It also looks cleaner.
- */
-static DEFINE_SPINLOCK(call_lock);
-
-struct call_data_struct {
- void (*func) (void *info);
- void *info;
- atomic_t started;
- atomic_t finished;
- int wait;
-};
+static DEFINE_PER_CPU(int, xen_resched_irq);
+static DEFINE_PER_CPU(int, xen_callfunc_irq);
+static DEFINE_PER_CPU(int, xen_callfuncsingle_irq);
+static DEFINE_PER_CPU(int, xen_debug_irq) = -1;
static irqreturn_t xen_call_function_interrupt(int irq, void *dev_id);
-
-static struct call_data_struct *call_data;
+static irqreturn_t xen_call_function_single_interrupt(int irq, void *dev_id);
/*
* Reschedule call back. Nothing to do,
*/
static irqreturn_t xen_reschedule_interrupt(int irq, void *dev_id)
{
+ inc_irq_stat(irq_resched_count);
+
return IRQ_HANDLED;
}
-static __cpuinit void cpu_bringup_and_idle(void)
+static __cpuinit void cpu_bringup(void)
{
int cpu = smp_processor_id();
cpu_init();
+ touch_softlockup_watchdog();
+ preempt_disable();
+
xen_enable_sysenter();
+ xen_enable_syscall();
- preempt_disable();
- per_cpu(cpu_state, cpu) = CPU_ONLINE;
+ cpu = smp_processor_id();
+ smp_store_cpu_info(cpu);
+ cpu_data(cpu).x86_max_cores = 1;
+ set_cpu_sibling_map(cpu);
xen_setup_cpu_clockevents();
+ set_cpu_online(cpu, true);
+ percpu_write(cpu_state, CPU_ONLINE);
+ wmb();
+
/* We can take interrupts now: we're officially "up". */
local_irq_enable();
wmb(); /* make sure everything is out */
+}
+
+static __cpuinit void cpu_bringup_and_idle(void)
+{
+ cpu_bringup();
cpu_idle();
}
NULL);
if (rc < 0)
goto fail;
- per_cpu(resched_irq, cpu) = rc;
+ per_cpu(xen_resched_irq, cpu) = rc;
callfunc_name = kasprintf(GFP_KERNEL, "callfunc%d", cpu);
rc = bind_ipi_to_irqhandler(XEN_CALL_FUNCTION_VECTOR,
NULL);
if (rc < 0)
goto fail;
- per_cpu(callfunc_irq, cpu) = rc;
+ per_cpu(xen_callfunc_irq, cpu) = rc;
debug_name = kasprintf(GFP_KERNEL, "debug%d", cpu);
rc = bind_virq_to_irqhandler(VIRQ_DEBUG, cpu, xen_debug_interrupt,
debug_name, NULL);
if (rc < 0)
goto fail;
- per_cpu(debug_irq, cpu) = rc;
+ per_cpu(xen_debug_irq, cpu) = rc;
+
+ callfunc_name = kasprintf(GFP_KERNEL, "callfuncsingle%d", cpu);
+ rc = bind_ipi_to_irqhandler(XEN_CALL_FUNCTION_SINGLE_VECTOR,
+ cpu,
+ xen_call_function_single_interrupt,
+ IRQF_DISABLED|IRQF_PERCPU|IRQF_NOBALANCING,
+ callfunc_name,
+ NULL);
+ if (rc < 0)
+ goto fail;
+ per_cpu(xen_callfuncsingle_irq, cpu) = rc;
return 0;
fail:
- if (per_cpu(resched_irq, cpu) >= 0)
- unbind_from_irqhandler(per_cpu(resched_irq, cpu), NULL);
- if (per_cpu(callfunc_irq, cpu) >= 0)
- unbind_from_irqhandler(per_cpu(callfunc_irq, cpu), NULL);
- if (per_cpu(debug_irq, cpu) >= 0)
- unbind_from_irqhandler(per_cpu(debug_irq, cpu), NULL);
+ if (per_cpu(xen_resched_irq, cpu) >= 0)
+ unbind_from_irqhandler(per_cpu(xen_resched_irq, cpu), NULL);
+ if (per_cpu(xen_callfunc_irq, cpu) >= 0)
+ unbind_from_irqhandler(per_cpu(xen_callfunc_irq, cpu), NULL);
+ if (per_cpu(xen_debug_irq, cpu) >= 0)
+ unbind_from_irqhandler(per_cpu(xen_debug_irq, cpu), NULL);
+ if (per_cpu(xen_callfuncsingle_irq, cpu) >= 0)
+ unbind_from_irqhandler(per_cpu(xen_callfuncsingle_irq, cpu),
+ NULL);
+
return rc;
}
-void __init xen_fill_possible_map(void)
+static void __init xen_fill_possible_map(void)
{
int i, rc;
- for (i = 0; i < NR_CPUS; i++) {
+ for (i = 0; i < nr_cpu_ids; i++) {
rc = HYPERVISOR_vcpu_op(VCPUOP_is_up, i, NULL);
- if (rc >= 0)
- cpu_set(i, cpu_possible_map);
+ if (rc >= 0) {
+ num_processors++;
+ set_cpu_possible(i, true);
+ }
}
}
-void __init xen_smp_prepare_boot_cpu(void)
+static void __init xen_smp_prepare_boot_cpu(void)
{
- int cpu;
-
BUG_ON(smp_processor_id() != 0);
native_smp_prepare_boot_cpu();
/* We've switched to the "real" per-cpu gdt, so make sure the
old memory can be recycled */
- make_lowmem_page_readwrite(&per_cpu__gdt_page);
-
- for_each_possible_cpu(cpu) {
- cpus_clear(per_cpu(cpu_sibling_map, cpu));
- /*
- * cpu_core_map lives in a per cpu area that is cleared
- * when the per cpu array is allocated.
- *
- * cpus_clear(per_cpu(cpu_core_map, cpu));
- */
- }
+ make_lowmem_page_readwrite(xen_initial_gdt);
xen_setup_vcpu_info_placement();
}
-void __init xen_smp_prepare_cpus(unsigned int max_cpus)
+static void __init xen_smp_prepare_cpus(unsigned int max_cpus)
{
unsigned cpu;
- for_each_possible_cpu(cpu) {
- cpus_clear(per_cpu(cpu_sibling_map, cpu));
- /*
- * cpu_core_ map will be zeroed when the per
- * cpu area is allocated.
- *
- * cpus_clear(per_cpu(cpu_core_map, cpu));
- */
- }
+ xen_init_lock_cpu(0);
smp_store_cpu_info(0);
+ cpu_data(0).x86_max_cores = 1;
set_cpu_sibling_map(0);
if (xen_smp_intr_init(0))
BUG();
- xen_cpu_initialized_map = cpumask_of_cpu(0);
+ if (!alloc_cpumask_var(&xen_cpu_initialized_map, GFP_KERNEL))
+ panic("could not allocate xen_cpu_initialized_map\n");
+
+ cpumask_copy(xen_cpu_initialized_map, cpumask_of(0));
/* Restrict the possible_map according to max_cpus. */
while ((num_possible_cpus() > 1) && (num_possible_cpus() > max_cpus)) {
- for (cpu = NR_CPUS - 1; !cpu_possible(cpu); cpu--)
+ for (cpu = nr_cpu_ids - 1; !cpu_possible(cpu); cpu--)
continue;
- cpu_clear(cpu, cpu_possible_map);
+ set_cpu_possible(cpu, false);
}
for_each_possible_cpu (cpu) {
if (IS_ERR(idle))
panic("failed fork for CPU %d", cpu);
- cpu_set(cpu, cpu_present_map);
+ set_cpu_present(cpu, true);
}
-
- //init_xenbus_allowed_cpumask();
}
static __cpuinit int
cpu_initialize_context(unsigned int cpu, struct task_struct *idle)
{
struct vcpu_guest_context *ctxt;
- struct gdt_page *gdt = &per_cpu(gdt_page, cpu);
+ struct desc_struct *gdt;
+ unsigned long gdt_mfn;
- if (cpu_test_and_set(cpu, xen_cpu_initialized_map))
+ if (cpumask_test_and_set_cpu(cpu, xen_cpu_initialized_map))
return 0;
ctxt = kzalloc(sizeof(*ctxt), GFP_KERNEL);
if (ctxt == NULL)
return -ENOMEM;
+ gdt = get_cpu_gdt_table(cpu);
+
ctxt->flags = VGCF_IN_KERNEL;
ctxt->user_regs.ds = __USER_DS;
ctxt->user_regs.es = __USER_DS;
- ctxt->user_regs.fs = __KERNEL_PERCPU;
- ctxt->user_regs.gs = 0;
ctxt->user_regs.ss = __KERNEL_DS;
+#ifdef CONFIG_X86_32
+ ctxt->user_regs.fs = __KERNEL_PERCPU;
+ ctxt->user_regs.gs = __KERNEL_STACK_CANARY;
+#else
+ ctxt->gs_base_kernel = per_cpu_offset(cpu);
+#endif
ctxt->user_regs.eip = (unsigned long)cpu_bringup_and_idle;
ctxt->user_regs.eflags = 0x1000; /* IOPL_RING1 */
ctxt->ldt_ents = 0;
- BUG_ON((unsigned long)gdt->gdt & ~PAGE_MASK);
- make_lowmem_page_readonly(gdt->gdt);
+ BUG_ON((unsigned long)gdt & ~PAGE_MASK);
- ctxt->gdt_frames[0] = virt_to_mfn(gdt->gdt);
- ctxt->gdt_ents = ARRAY_SIZE(gdt->gdt);
+ gdt_mfn = arbitrary_virt_to_mfn(gdt);
+ make_lowmem_page_readonly(gdt);
+ make_lowmem_page_readonly(mfn_to_virt(gdt_mfn));
+
+ ctxt->gdt_frames[0] = gdt_mfn;
+ ctxt->gdt_ents = GDT_ENTRIES;
ctxt->user_regs.cs = __KERNEL_CS;
ctxt->user_regs.esp = idle->thread.sp0 - sizeof(struct pt_regs);
ctxt->kernel_ss = __KERNEL_DS;
ctxt->kernel_sp = idle->thread.sp0;
+#ifdef CONFIG_X86_32
ctxt->event_callback_cs = __KERNEL_CS;
- ctxt->event_callback_eip = (unsigned long)xen_hypervisor_callback;
ctxt->failsafe_callback_cs = __KERNEL_CS;
+#endif
+ ctxt->event_callback_eip = (unsigned long)xen_hypervisor_callback;
ctxt->failsafe_callback_eip = (unsigned long)xen_failsafe_callback;
per_cpu(xen_cr3, cpu) = __pa(swapper_pg_dir);
return 0;
}
-int __cpuinit xen_cpu_up(unsigned int cpu)
+static int __cpuinit xen_cpu_up(unsigned int cpu)
{
struct task_struct *idle = idle_task(cpu);
int rc;
-#if 0
- rc = cpu_up_check(cpu);
- if (rc)
- return rc;
-#endif
-
- init_gdt(cpu);
per_cpu(current_task, cpu) = idle;
+#ifdef CONFIG_X86_32
irq_ctx_init(cpu);
+#else
+ clear_tsk_thread_flag(idle, TIF_FORK);
+ per_cpu(kernel_stack, cpu) =
+ (unsigned long)task_stack_page(idle) -
+ KERNEL_STACK_OFFSET + THREAD_SIZE;
+#endif
+ xen_setup_runstate_info(cpu);
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;
if (rc)
return rc;
- smp_store_cpu_info(cpu);
- set_cpu_sibling_map(cpu);
- /* This must be done before setting cpu_online_map */
- wmb();
-
- cpu_set(cpu, cpu_online_map);
-
rc = HYPERVISOR_vcpu_op(VCPUOP_up, cpu, NULL);
BUG_ON(rc);
+ while(per_cpu(cpu_state, cpu) != CPU_ONLINE) {
+ HYPERVISOR_sched_op(SCHEDOP_yield, NULL);
+ barrier();
+ }
+
+ return 0;
+}
+
+static void xen_smp_cpus_done(unsigned int max_cpus)
+{
+}
+
+#ifdef CONFIG_HOTPLUG_CPU
+static int xen_cpu_disable(void)
+{
+ unsigned int cpu = smp_processor_id();
+ if (cpu == 0)
+ return -EBUSY;
+
+ cpu_disable_common();
+
+ load_cr3(swapper_pg_dir);
return 0;
}
-void xen_smp_cpus_done(unsigned int max_cpus)
+static void xen_cpu_die(unsigned int cpu)
+{
+ while (HYPERVISOR_vcpu_op(VCPUOP_is_up, cpu, NULL)) {
+ current->state = TASK_UNINTERRUPTIBLE;
+ schedule_timeout(HZ/10);
+ }
+ unbind_from_irqhandler(per_cpu(xen_resched_irq, cpu), NULL);
+ unbind_from_irqhandler(per_cpu(xen_callfunc_irq, cpu), NULL);
+ unbind_from_irqhandler(per_cpu(xen_debug_irq, cpu), NULL);
+ unbind_from_irqhandler(per_cpu(xen_callfuncsingle_irq, cpu), NULL);
+ xen_uninit_lock_cpu(cpu);
+ xen_teardown_timer(cpu);
+
+ if (num_online_cpus() == 1)
+ alternatives_smp_switch(0);
+}
+
+static void __cpuinit xen_play_dead(void) /* used only with HOTPLUG_CPU */
+{
+ play_dead_common();
+ HYPERVISOR_vcpu_op(VCPUOP_down, smp_processor_id(), NULL);
+ cpu_bringup();
+}
+
+#else /* !CONFIG_HOTPLUG_CPU */
+static int xen_cpu_disable(void)
{
+ return -ENOSYS;
}
+static void xen_cpu_die(unsigned int cpu)
+{
+ BUG();
+}
+
+static void xen_play_dead(void)
+{
+ BUG();
+}
+
+#endif
static void stop_self(void *v)
{
int cpu = smp_processor_id();
BUG();
}
-void xen_smp_send_stop(void)
+static void xen_smp_send_stop(void)
{
- smp_call_function(stop_self, NULL, 0, 0);
+ smp_call_function(stop_self, NULL, 0);
}
-void xen_smp_send_reschedule(int cpu)
+static void xen_smp_send_reschedule(int cpu)
{
xen_send_IPI_one(cpu, XEN_RESCHEDULE_VECTOR);
}
-
-static void xen_send_IPI_mask(cpumask_t mask, enum ipi_vector vector)
+static void xen_send_IPI_mask(const struct cpumask *mask,
+ enum ipi_vector vector)
{
unsigned cpu;
- cpus_and(mask, mask, cpu_online_map);
-
- for_each_cpu_mask_nr(cpu, mask)
+ for_each_cpu_and(cpu, mask, cpu_online_mask)
xen_send_IPI_one(cpu, vector);
}
-static irqreturn_t xen_call_function_interrupt(int irq, void *dev_id)
+static void xen_smp_send_call_function_ipi(const struct cpumask *mask)
{
- void (*func) (void *info) = call_data->func;
- void *info = call_data->info;
- int wait = call_data->wait;
-
- /*
- * Notify initiating CPU that I've grabbed the data and am
- * about to execute the function
- */
- mb();
- atomic_inc(&call_data->started);
- /*
- * At this point the info structure may be out of scope unless wait==1
- */
- irq_enter();
- (*func)(info);
- __get_cpu_var(irq_stat).irq_call_count++;
- irq_exit();
+ int cpu;
- if (wait) {
- mb(); /* commit everything before setting finished */
- atomic_inc(&call_data->finished);
- }
+ xen_send_IPI_mask(mask, XEN_CALL_FUNCTION_VECTOR);
- return IRQ_HANDLED;
+ /* Make sure other vcpus get a chance to run if they need to. */
+ for_each_cpu(cpu, mask) {
+ if (xen_vcpu_stolen(cpu)) {
+ HYPERVISOR_sched_op(SCHEDOP_yield, NULL);
+ break;
+ }
+ }
}
-int xen_smp_call_function_mask(cpumask_t mask, void (*func)(void *),
- void *info, int wait)
+static void xen_smp_send_call_function_single_ipi(int cpu)
{
- struct call_data_struct data;
- int cpus, cpu;
- bool yield;
-
- /* Holding any lock stops cpus from going down. */
- spin_lock(&call_lock);
-
- cpu_clear(smp_processor_id(), mask);
-
- cpus = cpus_weight(mask);
- if (!cpus) {
- spin_unlock(&call_lock);
- return 0;
- }
+ xen_send_IPI_mask(cpumask_of(cpu),
+ XEN_CALL_FUNCTION_SINGLE_VECTOR);
+}
- /* Can deadlock when called with interrupts disabled */
- WARN_ON(irqs_disabled());
+static irqreturn_t xen_call_function_interrupt(int irq, void *dev_id)
+{
+ irq_enter();
+ generic_smp_call_function_interrupt();
+ inc_irq_stat(irq_call_count);
+ irq_exit();
- data.func = func;
- data.info = info;
- atomic_set(&data.started, 0);
- data.wait = wait;
- if (wait)
- atomic_set(&data.finished, 0);
+ return IRQ_HANDLED;
+}
- call_data = &data;
- mb(); /* write everything before IPI */
+static irqreturn_t xen_call_function_single_interrupt(int irq, void *dev_id)
+{
+ irq_enter();
+ generic_smp_call_function_single_interrupt();
+ inc_irq_stat(irq_call_count);
+ irq_exit();
- /* Send a message to other CPUs and wait for them to respond */
- xen_send_IPI_mask(mask, XEN_CALL_FUNCTION_VECTOR);
+ return IRQ_HANDLED;
+}
- /* Make sure other vcpus get a chance to run if they need to. */
- yield = false;
- for_each_cpu_mask_nr(cpu, mask)
- if (xen_vcpu_stolen(cpu))
- yield = true;
+static const struct smp_ops xen_smp_ops __initdata = {
+ .smp_prepare_boot_cpu = xen_smp_prepare_boot_cpu,
+ .smp_prepare_cpus = xen_smp_prepare_cpus,
+ .smp_cpus_done = xen_smp_cpus_done,
- if (yield)
- HYPERVISOR_sched_op(SCHEDOP_yield, 0);
+ .cpu_up = xen_cpu_up,
+ .cpu_die = xen_cpu_die,
+ .cpu_disable = xen_cpu_disable,
+ .play_dead = xen_play_dead,
- /* Wait for response */
- while (atomic_read(&data.started) != cpus ||
- (wait && atomic_read(&data.finished) != cpus))
- cpu_relax();
+ .smp_send_stop = xen_smp_send_stop,
+ .smp_send_reschedule = xen_smp_send_reschedule,
- spin_unlock(&call_lock);
+ .send_call_func_ipi = xen_smp_send_call_function_ipi,
+ .send_call_func_single_ipi = xen_smp_send_call_function_single_ipi,
+};
- return 0;
+void __init xen_smp_init(void)
+{
+ smp_ops = xen_smp_ops;
+ xen_fill_possible_map();
+ xen_init_spinlocks();
}
git://ftp.safe.ca / safe / jmp / linux-2.6 / blobdiff