smpboot.c

linux-2.6.15.6

	setup_local_APIC();

	/*
	 * Get our bogomips.
 	 *
 	 * Need to enable IRQs because it can take longer and then
	 * the NMI watchdog might kill us.
	 */
	local_irq_enable();
	calibrate_delay();
	local_irq_disable();
	Dprintk("Stack at about %p\n",&cpuid);

	disable_APIC_timer();

	/*
	 * Save our processor parameters
	 */
 	smp_store_cpu_info(cpuid);

	/*
	 * Allow the master to continue.
	 */
	cpu_set(cpuid, cpu_callin_map);
}

/* representing cpus for which sibling maps can be computed */
static cpumask_t cpu_sibling_setup_map;

static inline void set_cpu_sibling_map(int cpu)
{
	int i;
	struct cpuinfo_x86 *c = cpu_data;

	cpu_set(cpu, cpu_sibling_setup_map);

	if (smp_num_siblings > 1) {
		for_each_cpu_mask(i, cpu_sibling_setup_map) {
			if (phys_proc_id[cpu] == phys_proc_id[i] &&
			    cpu_core_id[cpu] == cpu_core_id[i]) {
				cpu_set(i, cpu_sibling_map[cpu]);
				cpu_set(cpu, cpu_sibling_map[i]);
				cpu_set(i, cpu_core_map[cpu]);
				cpu_set(cpu, cpu_core_map[i]);
			}
		}
	} else {
		cpu_set(cpu, cpu_sibling_map[cpu]);
	}

	if (current_cpu_data.x86_max_cores == 1) {
		cpu_core_map[cpu] = cpu_sibling_map[cpu];
		c[cpu].booted_cores = 1;
		return;
	}

	for_each_cpu_mask(i, cpu_sibling_setup_map) {
		if (phys_proc_id[cpu] == phys_proc_id[i]) {
			cpu_set(i, cpu_core_map[cpu]);
			cpu_set(cpu, cpu_core_map[i]);
			/*
			 *  Does this new cpu bringup a new core?
			 */
			if (cpus_weight(cpu_sibling_map[cpu]) == 1) {
				/*
				 * for each core in package, increment
				 * the booted_cores for this new cpu
				 */
				if (first_cpu(cpu_sibling_map[i]) == i)
					c[cpu].booted_cores++;
				/*
				 * increment the core count for all
				 * the other cpus in this package
				 */
				if (i != cpu)
					c[i].booted_cores++;
			} else if (i != cpu && !c[cpu].booted_cores)
				c[cpu].booted_cores = c[i].booted_cores;
		}
	}
}

/*
 * Setup code on secondary processor (after comming out of the trampoline)
 */
void __cpuinit start_secondary(void)
{
	/*
	 * Dont put anything before smp_callin(), SMP
	 * booting is too fragile that we want to limit the
	 * things done here to the most necessary things.
	 */
	cpu_init();
	preempt_disable();
	smp_callin();

	/* otherwise gcc will move up the smp_processor_id before the cpu_init */
	barrier();

	Dprintk("cpu %d: setting up apic clock\n", smp_processor_id()); 	
	setup_secondary_APIC_clock();

	Dprintk("cpu %d: enabling apic timer\n", smp_processor_id());

	if (nmi_watchdog == NMI_IO_APIC) {
		disable_8259A_irq(0);
		enable_NMI_through_LVT0(NULL);
		enable_8259A_irq(0);
	}

	enable_APIC_timer();

	/*
	 * The sibling maps must be set before turing the online map on for
	 * this cpu
	 */
	set_cpu_sibling_map(smp_processor_id());

	/* 
  	 * Wait for TSC sync to not schedule things before.
	 * We still process interrupts, which could see an inconsistent
	 * time in that window unfortunately. 
	 * Do this here because TSC sync has global unprotected state.
 	 */
	tsc_sync_wait();

	/*
	 * We need to hold call_lock, so there is no inconsistency
	 * between the time smp_call_function() determines number of
	 * IPI receipients, and the time when the determination is made
	 * for which cpus receive the IPI in genapic_flat.c. Holding this
	 * lock helps us to not include this cpu in a currently in progress
	 * smp_call_function().
	 */
	lock_ipi_call_lock();

	/*
	 * Allow the master to continue.
	 */
	cpu_set(smp_processor_id(), cpu_online_map);
	per_cpu(cpu_state, smp_processor_id()) = CPU_ONLINE;
	unlock_ipi_call_lock();

	cpu_idle();
}

extern volatile unsigned long init_rsp;
extern void (*initial_code)(void);

#ifdef APIC_DEBUG
static void inquire_remote_apic(int apicid)
{
	unsigned i, regs[] = { APIC_ID >> 4, APIC_LVR >> 4, APIC_SPIV >> 4 };
	char *names[] = { "ID", "VERSION", "SPIV" };
	int timeout, status;

	printk(KERN_INFO "Inquiring remote APIC #%d...\n", apicid);

	for (i = 0; i < sizeof(regs) / sizeof(*regs); i++) {
		printk("... APIC #%d %s: ", apicid, names[i]);

		/*
		 * Wait for idle.
		 */
		apic_wait_icr_idle();

		apic_write(APIC_ICR2, SET_APIC_DEST_FIELD(apicid));
		apic_write(APIC_ICR, APIC_DM_REMRD | regs[i]);

		timeout = 0;
		do {
			udelay(100);
			status = apic_read(APIC_ICR) & APIC_ICR_RR_MASK;
		} while (status == APIC_ICR_RR_INPROG && timeout++ < 1000);

		switch (status) {
		case APIC_ICR_RR_VALID:
			status = apic_read(APIC_RRR);
			printk("%08x\n", status);
			break;
		default:
			printk("failed\n");
		}
	}
}
#endif

/*
 * Kick the secondary to wake up.
 */
static int __cpuinit wakeup_secondary_via_INIT(int phys_apicid, unsigned int start_rip)
{
	unsigned long send_status = 0, accept_status = 0;
	int maxlvt, timeout, num_starts, j;

	Dprintk("Asserting INIT.\n");

	/*
	 * Turn INIT on target chip
	 */
	apic_write(APIC_ICR2, SET_APIC_DEST_FIELD(phys_apicid));

	/*
	 * Send IPI
	 */
	apic_write(APIC_ICR, APIC_INT_LEVELTRIG | APIC_INT_ASSERT
				| APIC_DM_INIT);

	Dprintk("Waiting for send to finish...\n");
	timeout = 0;
	do {
		Dprintk("+");
		udelay(100);
		send_status = apic_read(APIC_ICR) & APIC_ICR_BUSY;
	} while (send_status && (timeout++ < 1000));

	mdelay(10);

	Dprintk("Deasserting INIT.\n");

	/* Target chip */
	apic_write(APIC_ICR2, SET_APIC_DEST_FIELD(phys_apicid));

	/* Send IPI */
	apic_write(APIC_ICR, APIC_INT_LEVELTRIG | APIC_DM_INIT);

	Dprintk("Waiting for send to finish...\n");
	timeout = 0;
	do {
		Dprintk("+");
		udelay(100);
		send_status = apic_read(APIC_ICR) & APIC_ICR_BUSY;
	} while (send_status && (timeout++ < 1000));

	atomic_set(&init_deasserted, 1);

	num_starts = 2;

	/*
	 * Run STARTUP IPI loop.
	 */
	Dprintk("#startup loops: %d.\n", num_starts);

	maxlvt = get_maxlvt();

	for (j = 1; j <= num_starts; j++) {
		Dprintk("Sending STARTUP #%d.\n",j);
		apic_read_around(APIC_SPIV);
		apic_write(APIC_ESR, 0);
		apic_read(APIC_ESR);
		Dprintk("After apic_write.\n");

		/*
		 * STARTUP IPI
		 */

		/* Target chip */
		apic_write(APIC_ICR2, SET_APIC_DEST_FIELD(phys_apicid));

		/* Boot on the stack */
		/* Kick the second */
		apic_write(APIC_ICR, APIC_DM_STARTUP | (start_rip >> 12));

		/*
		 * Give the other CPU some time to accept the IPI.
		 */
		udelay(300);

		Dprintk("Startup point 1.\n");

		Dprintk("Waiting for send to finish...\n");
		timeout = 0;
		do {
			Dprintk("+");
			udelay(100);
			send_status = apic_read(APIC_ICR) & APIC_ICR_BUSY;
		} while (send_status && (timeout++ < 1000));

		/*
		 * Give the other CPU some time to accept the IPI.
		 */
		udelay(200);
		/*
		 * Due to the Pentium erratum 3AP.
		 */
		if (maxlvt > 3) {
			apic_read_around(APIC_SPIV);
			apic_write(APIC_ESR, 0);
		}
		accept_status = (apic_read(APIC_ESR) & 0xEF);
		if (send_status || accept_status)
			break;
	}
	Dprintk("After Startup.\n");

	if (send_status)
		printk(KERN_ERR "APIC never delivered???\n");
	if (accept_status)
		printk(KERN_ERR "APIC delivery error (%lx).\n", accept_status);

	return (send_status | accept_status);
}

struct create_idle {
	struct task_struct *idle;
	struct completion done;
	int cpu;
};

void do_fork_idle(void *_c_idle)
{
	struct create_idle *c_idle = _c_idle;

	c_idle->idle = fork_idle(c_idle->cpu);
	complete(&c_idle->done);
}

/*
 * Boot one CPU.
 */
static int __cpuinit do_boot_cpu(int cpu, int apicid)
{
	unsigned long boot_error;
	int timeout;
	unsigned long start_rip;
	struct create_idle c_idle = {
		.cpu = cpu,
		.done = COMPLETION_INITIALIZER(c_idle.done),
	};
	DECLARE_WORK(work, do_fork_idle, &c_idle);

	c_idle.idle = get_idle_for_cpu(cpu);

	if (c_idle.idle) {
		c_idle.idle->thread.rsp = (unsigned long) (((struct pt_regs *)
			(THREAD_SIZE + (unsigned long) c_idle.idle->thread_info)) - 1);
		init_idle(c_idle.idle, cpu);
		goto do_rest;
	}

	/*
	 * During cold boot process, keventd thread is not spun up yet.
	 * When we do cpu hot-add, we create idle threads on the fly, we should
	 * not acquire any attributes from the calling context. Hence the clean
	 * way to create kernel_threads() is to do that from keventd().
	 * We do the current_is_keventd() due to the fact that ACPI notifier
	 * was also queuing to keventd() and when the caller is already running
	 * in context of keventd(), we would end up with locking up the keventd
	 * thread.
	 */
	if (!keventd_up() || current_is_keventd())
		work.func(work.data);
	else {
		schedule_work(&work);
		wait_for_completion(&c_idle.done);
	}

	if (IS_ERR(c_idle.idle)) {
		printk("failed fork for CPU %d\n", cpu);
		return PTR_ERR(c_idle.idle);
	}

	set_idle_for_cpu(cpu, c_idle.idle);

do_rest:

	cpu_pda[cpu].pcurrent = c_idle.idle;

	start_rip = setup_trampoline();

	init_rsp = c_idle.idle->thread.rsp;
	per_cpu(init_tss,cpu).rsp0 = init_rsp;
	initial_code = start_secondary;
	clear_ti_thread_flag(c_idle.idle->thread_info, TIF_FORK);

	printk(KERN_INFO "Booting processor %d/%d APIC 0x%x\n", cpu,
		cpus_weight(cpu_present_map),
		apicid);

	/*
	 * This grunge runs the startup process for
	 * the targeted processor.
	 */

	atomic_set(&init_deasserted, 0);

	Dprintk("Setting warm reset code and vector.\n");

	CMOS_WRITE(0xa, 0xf);
	local_flush_tlb();
	Dprintk("1.\n");
	*((volatile unsigned short *) phys_to_virt(0x469)) = start_rip >> 4;
	Dprintk("2.\n");
	*((volatile unsigned short *) phys_to_virt(0x467)) = start_rip & 0xf;
	Dprintk("3.\n");

	/*
	 * Be paranoid about clearing APIC errors.
	 */
	if (APIC_INTEGRATED(apic_version[apicid])) {
		apic_read_around(APIC_SPIV);
		apic_write(APIC_ESR, 0);
		apic_read(APIC_ESR);
	}

	/*
	 * Status is now clean
	 */
	boot_error = 0;

	/*
	 * Starting actual IPI sequence...
	 */
	boot_error = wakeup_secondary_via_INIT(apicid, start_rip);