Merge branch 'linus' into perf/urgent

Merge reason: we need to fix a non-trivial merge conflict.

Signed-off-by: Ingo Molnar <mingo@elte.hu>

1 files changed, 577 insertions, 570 deletions

diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index b0e28c47ab8..a8bf9994b08 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -1,23 +1,18 @@
 /*
- *  arch/s390/kernel/smp.c
+ *  SMP related functions
  *
- *    Copyright IBM Corp. 1999, 2009
- *    Author(s): Denis Joseph Barrow (djbarrow@de.ibm.com,barrow_dj@yahoo.com),
- *		 Martin Schwidefsky (schwidefsky@de.ibm.com)
- *		 Heiko Carstens (heiko.carstens@de.ibm.com)
+ *    Copyright IBM Corp. 1999,2012
+ *    Author(s): Denis Joseph Barrow,
+ *		 Martin Schwidefsky <schwidefsky@de.ibm.com>,
+ *		 Heiko Carstens <heiko.carstens@de.ibm.com>,
  *
  *  based on other smp stuff by
  *    (c) 1995 Alan Cox, CymruNET Ltd  <alan@cymru.net>
  *    (c) 1998 Ingo Molnar
  *
- * We work with logical cpu numbering everywhere we can. The only
- * functions using the real cpu address (got from STAP) are the sigp
- * functions. For all other functions we use the identity mapping.
- * That means that cpu_number_map[i] == i for every cpu. cpu_number_map is
- * used e.g. to find the idle task belonging to a logical cpu. Every array
- * in the kernel is sorted by the logical cpu number and not by the physical
- * one which is causing all the confusion with __cpu_logical_map and
- * cpu_number_map in other architectures.
+ * The code outside of smp.c uses logical cpu numbers, only smp.c does
+ * the translation of logical to physical cpu ids. All new code that
+ * operates on physical cpu numbers needs to go into smp.c.
  */
 
 #define KMSG_COMPONENT "cpu"
@@ -31,198 +26,433 @@
 #include <linux/spinlock.h>
 #include <linux/kernel_stat.h>
 #include <linux/delay.h>
-#include <linux/cache.h>
 #include <linux/interrupt.h>
 #include <linux/irqflags.h>
 #include <linux/cpu.h>
-#include <linux/timex.h>
-#include <linux/bootmem.h>
 #include <linux/slab.h>
 #include <linux/crash_dump.h>
 #include <asm/asm-offsets.h>
 #include <asm/ipl.h>
 #include <asm/setup.h>
-#include <asm/sigp.h>
-#include <asm/pgalloc.h>
 #include <asm/irq.h>
-#include <asm/cpcmd.h>
 #include <asm/tlbflush.h>
 #include <asm/timer.h>
 #include <asm/lowcore.h>
 #include <asm/sclp.h>
-#include <asm/cputime.h>
 #include <asm/vdso.h>
-#include <asm/cpu.h>
+#include <asm/debug.h>
+#include <asm/os_info.h>
 #include "entry.h"
 
-/* logical cpu to cpu address */
-unsigned short __cpu_logical_map[NR_CPUS];
+enum {
+	sigp_sense = 1,
+	sigp_external_call = 2,
+	sigp_emergency_signal = 3,
+	sigp_start = 4,
+	sigp_stop = 5,
+	sigp_restart = 6,
+	sigp_stop_and_store_status = 9,
+	sigp_initial_cpu_reset = 11,
+	sigp_cpu_reset = 12,
+	sigp_set_prefix = 13,
+	sigp_store_status_at_address = 14,
+	sigp_store_extended_status_at_address = 15,
+	sigp_set_architecture = 18,
+	sigp_conditional_emergency_signal = 19,
+	sigp_sense_running = 21,
+};
 
-static struct task_struct *current_set[NR_CPUS];
+enum {
+	sigp_order_code_accepted = 0,
+	sigp_status_stored = 1,
+	sigp_busy = 2,
+	sigp_not_operational = 3,
+};
 
-static u8 smp_cpu_type;
-static int smp_use_sigp_detection;
+enum {
+	ec_schedule = 0,
+	ec_call_function,
+	ec_call_function_single,
+	ec_stop_cpu,
+};
 
-enum s390_cpu_state {
+enum {
 	CPU_STATE_STANDBY,
 	CPU_STATE_CONFIGURED,
 };
 
+struct pcpu {
+	struct cpu cpu;
+	struct task_struct *idle;	/* idle process for the cpu */
+	struct _lowcore *lowcore;	/* lowcore page(s) for the cpu */
+	unsigned long async_stack;	/* async stack for the cpu */
+	unsigned long panic_stack;	/* panic stack for the cpu */
+	unsigned long ec_mask;		/* bit mask for ec_xxx functions */
+	int state;			/* physical cpu state */
+	u32 status;			/* last status received via sigp */
+	u16 address;			/* physical cpu address */
+};
+
+static u8 boot_cpu_type;
+static u16 boot_cpu_address;
+static struct pcpu pcpu_devices[NR_CPUS];
+
 DEFINE_MUTEX(smp_cpu_state_mutex);
-static int smp_cpu_state[NR_CPUS];
 
-static DEFINE_PER_CPU(struct cpu, cpu_devices);
+/*
+ * Signal processor helper functions.
+ */
+static inline int __pcpu_sigp(u16 addr, u8 order, u32 parm, u32 *status)
+{
+	register unsigned int reg1 asm ("1") = parm;
+	int cc;
 
-static void smp_ext_bitcall(int, int);
+	asm volatile(
+		"	sigp	%1,%2,0(%3)\n"
+		"	ipm	%0\n"
+		"	srl	%0,28\n"
+		: "=d" (cc), "+d" (reg1) : "d" (addr), "a" (order) : "cc");
+	if (status && cc == 1)
+		*status = reg1;
+	return cc;
+}
 
-static int raw_cpu_stopped(int cpu)
+static inline int __pcpu_sigp_relax(u16 addr, u8 order, u32 parm, u32 *status)
 {
-	u32 status;
+	int cc;
 
-	switch (raw_sigp_ps(&status, 0, cpu, sigp_sense)) {
-	case sigp_status_stored:
-		/* Check for stopped and check stop state */
-		if (status & 0x50)
-			return 1;
-		break;
-	default:
-		break;
+	while (1) {
+		cc = __pcpu_sigp(addr, order, parm, status);
+		if (cc != sigp_busy)
+			return cc;
+		cpu_relax();
 	}
-	return 0;
 }
 
-static inline int cpu_stopped(int cpu)
+static int pcpu_sigp_retry(struct pcpu *pcpu, u8 order, u32 parm)
+{
+	int cc, retry;
+
+	for (retry = 0; ; retry++) {
+		cc = __pcpu_sigp(pcpu->address, order, parm, &pcpu->status);
+		if (cc != sigp_busy)
+			break;
+		if (retry >= 3)
+			udelay(10);
+	}
+	return cc;
+}
+
+static inline int pcpu_stopped(struct pcpu *pcpu)
+{
+	if (__pcpu_sigp(pcpu->address, sigp_sense,
+			0, &pcpu->status) != sigp_status_stored)
+		return 0;
+	/* Check for stopped and check stop state */
+	return !!(pcpu->status & 0x50);
+}
+
+static inline int pcpu_running(struct pcpu *pcpu)
 {
-	return raw_cpu_stopped(cpu_logical_map(cpu));
+	if (__pcpu_sigp(pcpu->address, sigp_sense_running,
+			0, &pcpu->status) != sigp_status_stored)
+		return 1;
+	/* Check for running status */
+	return !(pcpu->status & 0x400);
 }
 
 /*
- * Ensure that PSW restart is done on an online CPU
+ * Find struct pcpu by cpu address.
  */
-void smp_restart_with_online_cpu(void)
+static struct pcpu *pcpu_find_address(const struct cpumask *mask, int address)
 {
 	int cpu;
 
-	for_each_online_cpu(cpu) {
-		if (stap() == __cpu_logical_map[cpu]) {
-			/* We are online: Enable DAT again and return */
-			__load_psw_mask(psw_kernel_bits | PSW_MASK_DAT);
-			return;
-		}
+	for_each_cpu(cpu, mask)
+		if (pcpu_devices[cpu].address == address)
+			return pcpu_devices + cpu;
+	return NULL;
+}
+
+static void pcpu_ec_call(struct pcpu *pcpu, int ec_bit)
+{
+	int order;
+
+	set_bit(ec_bit, &pcpu->ec_mask);
+	order = pcpu_running(pcpu) ?
+		sigp_external_call : sigp_emergency_signal;
+	pcpu_sigp_retry(pcpu, order, 0);
+}
+
+static int __cpuinit pcpu_alloc_lowcore(struct pcpu *pcpu, int cpu)
+{
+	struct _lowcore *lc;
+
+	if (pcpu != &pcpu_devices[0]) {
+		pcpu->lowcore =	(struct _lowcore *)
+			__get_free_pages(GFP_KERNEL | GFP_DMA, LC_ORDER);
+		pcpu->async_stack = __get_free_pages(GFP_KERNEL, ASYNC_ORDER);
+		pcpu->panic_stack = __get_free_page(GFP_KERNEL);
+		if (!pcpu->lowcore || !pcpu->panic_stack || !pcpu->async_stack)
+			goto out;
 	}
-	/* We are not online: Do PSW restart on an online CPU */
-	while (sigp(cpu, sigp_restart) == sigp_busy)
-		cpu_relax();
-	/* And stop ourself */
-	while (raw_sigp(stap(), sigp_stop) == sigp_busy)
-		cpu_relax();
-	for (;;);
+	lc = pcpu->lowcore;
+	memcpy(lc, &S390_lowcore, 512);
+	memset((char *) lc + 512, 0, sizeof(*lc) - 512);
+	lc->async_stack = pcpu->async_stack + ASYNC_SIZE;
+	lc->panic_stack = pcpu->panic_stack + PAGE_SIZE;
+	lc->cpu_nr = cpu;
+#ifndef CONFIG_64BIT
+	if (MACHINE_HAS_IEEE) {
+		lc->extended_save_area_addr = get_zeroed_page(GFP_KERNEL);
+		if (!lc->extended_save_area_addr)
+			goto out;
+	}
+#else
+	if (vdso_alloc_per_cpu(lc))
+		goto out;
+#endif
+	lowcore_ptr[cpu] = lc;
+	pcpu_sigp_retry(pcpu, sigp_set_prefix, (u32)(unsigned long) lc);
+	return 0;
+out:
+	if (pcpu != &pcpu_devices[0]) {
+		free_page(pcpu->panic_stack);
+		free_pages(pcpu->async_stack, ASYNC_ORDER);
+		free_pages((unsigned long) pcpu->lowcore, LC_ORDER);
+	}
+	return -ENOMEM;
 }
 
-void smp_switch_to_ipl_cpu(void (*func)(void *), void *data)
+static void pcpu_free_lowcore(struct pcpu *pcpu)
 {
-	struct _lowcore *lc, *current_lc;
-	struct stack_frame *sf;
-	struct pt_regs *regs;
-	unsigned long sp;
-
-	if (smp_processor_id() == 0)
-		func(data);
-	__load_psw_mask(PSW_DEFAULT_KEY | PSW_MASK_BASE |
-			PSW_MASK_EA | PSW_MASK_BA);
-	/* Disable lowcore protection */
-	__ctl_clear_bit(0, 28);
-	current_lc = lowcore_ptr[smp_processor_id()];
-	lc = lowcore_ptr[0];
-	if (!lc)
-		lc = current_lc;
-	lc->restart_psw.mask =
-		PSW_DEFAULT_KEY | PSW_MASK_BASE | PSW_MASK_EA | PSW_MASK_BA;
-	lc->restart_psw.addr = PSW_ADDR_AMODE | (unsigned long) smp_restart_cpu;
-	if (!cpu_online(0))
-		smp_switch_to_cpu(func, data, 0, stap(), __cpu_logical_map[0]);
-	while (sigp(0, sigp_stop_and_store_status) == sigp_busy)
-		cpu_relax();
-	sp = lc->panic_stack;
-	sp -= sizeof(struct pt_regs);
-	regs = (struct pt_regs *) sp;
-	memcpy(&regs->gprs, &current_lc->gpregs_save_area, sizeof(regs->gprs));
-	regs->psw = current_lc->psw_save_area;
-	sp -= STACK_FRAME_OVERHEAD;
-	sf = (struct stack_frame *) sp;
-	sf->back_chain = 0;
-	smp_switch_to_cpu(func, data, sp, stap(), __cpu_logical_map[0]);
+	pcpu_sigp_retry(pcpu, sigp_set_prefix, 0);
+	lowcore_ptr[pcpu - pcpu_devices] = NULL;
+#ifndef CONFIG_64BIT
+	if (MACHINE_HAS_IEEE) {
+		struct _lowcore *lc = pcpu->lowcore;
+
+		free_page((unsigned long) lc->extended_save_area_addr);
+		lc->extended_save_area_addr = 0;
+	}
+#else
+	vdso_free_per_cpu(pcpu->lowcore);
+#endif
+	if (pcpu != &pcpu_devices[0]) {
+		free_page(pcpu->panic_stack);
+		free_pages(pcpu->async_stack, ASYNC_ORDER);
+		free_pages((unsigned long) pcpu->lowcore, LC_ORDER);
+	}
+}
+
+static void pcpu_prepare_secondary(struct pcpu *pcpu, int cpu)
+{
+	struct _lowcore *lc = pcpu->lowcore;
+
+	atomic_inc(&init_mm.context.attach_count);
+	lc->cpu_nr = cpu;
+	lc->percpu_offset = __per_cpu_offset[cpu];
+	lc->kernel_asce = S390_lowcore.kernel_asce;
+	lc->machine_flags = S390_lowcore.machine_flags;
+	lc->ftrace_func = S390_lowcore.ftrace_func;
+	lc->user_timer = lc->system_timer = lc->steal_timer = 0;
+	__ctl_store(lc->cregs_save_area, 0, 15);
+	save_access_regs((unsigned int *) lc->access_regs_save_area);
+	memcpy(lc->stfle_fac_list, S390_lowcore.stfle_fac_list,
+	       MAX_FACILITY_BIT/8);
+}
+
+static void pcpu_attach_task(struct pcpu *pcpu, struct task_struct *tsk)
+{
+	struct _lowcore *lc = pcpu->lowcore;
+	struct thread_info *ti = task_thread_info(tsk);
+
+	lc->kernel_stack = (unsigned long) task_stack_page(tsk) + THREAD_SIZE;
+	lc->thread_info = (unsigned long) task_thread_info(tsk);
+	lc->current_task = (unsigned long) tsk;
+	lc->user_timer = ti->user_timer;
+	lc->system_timer = ti->system_timer;
+	lc->steal_timer = 0;
+}
+
+static void pcpu_start_fn(struct pcpu *pcpu, void (*func)(void *), void *data)
+{
+	struct _lowcore *lc = pcpu->lowcore;
+
+	lc->restart_stack = lc->kernel_stack;
+	lc->restart_fn = (unsigned long) func;
+	lc->restart_data = (unsigned long) data;
+	lc->restart_source = -1UL;
+	pcpu_sigp_retry(pcpu, sigp_restart, 0);
+}
+
+/*
+ * Call function via PSW restart on pcpu and stop the current cpu.
+ */
+static void pcpu_delegate(struct pcpu *pcpu, void (*func)(void *),
+			  void *data, unsigned long stack)
+{
+	struct _lowcore *lc = pcpu->lowcore;
+	unsigned short this_cpu;
+
+	__load_psw_mask(psw_kernel_bits);
+	this_cpu = stap();
+	if (pcpu->address == this_cpu)
+		func(data);	/* should not return */
+	/* Stop target cpu (if func returns this stops the current cpu). */
+	pcpu_sigp_retry(pcpu, sigp_stop, 0);
+	/* Restart func on the target cpu and stop the current cpu. */
+	lc->restart_stack = stack;
+	lc->restart_fn = (unsigned long) func;
+	lc->restart_data = (unsigned long) data;
+	lc->restart_source = (unsigned long) this_cpu;
+	asm volatile(
+		"0:	sigp	0,%0,6	# sigp restart to target cpu\n"
+		"	brc	2,0b	# busy, try again\n"
+		"1:	sigp	0,%1,5	# sigp stop to current cpu\n"
+		"	brc	2,1b	# busy, try again\n"
+		: : "d" (pcpu->address), "d" (this_cpu) : "0", "1", "cc");
+	for (;;) ;
+}
+
+/*
+ * Call function on an online CPU.
+ */
+void smp_call_online_cpu(void (*func)(void *), void *data)
+{
+	struct pcpu *pcpu;
+
+	/* Use the current cpu if it is online. */
+	pcpu = pcpu_find_address(cpu_online_mask, stap());
+	if (!pcpu)
+		/* Use the first online cpu. */
+		pcpu = pcpu_devices + cpumask_first(cpu_online_mask);
+	pcpu_delegate(pcpu, func, data, (unsigned long) restart_stack);
+}
+
+/*
+ * Call function on the ipl CPU.
+ */
+void smp_call_ipl_cpu(void (*func)(void *), void *data)
+{
+	pcpu_delegate(&pcpu_devices[0], func, data,
+		      pcpu_devices->panic_stack + PAGE_SIZE);
+}
+
+int smp_find_processor_id(u16 address)
+{
+	int cpu;
+
+	for_each_present_cpu(cpu)
+		if (pcpu_devices[cpu].address == address)
+			return cpu;
+	return -1;
+}
+
+int smp_vcpu_scheduled(int cpu)
+{
+	return pcpu_running(pcpu_devices + cpu);
+}
+
+void smp_yield(void)
+{
+	if (MACHINE_HAS_DIAG44)
+		asm volatile("diag 0,0,0x44");
 }
 
-static void smp_stop_cpu(void)
+void smp_yield_cpu(int cpu)
 {
-	while (sigp(smp_processor_id(), sigp_stop) == sigp_busy)
+	if (MACHINE_HAS_DIAG9C)
+		asm volatile("diag %0,0,0x9c"
+			     : : "d" (pcpu_devices[cpu].address));
+	else if (MACHINE_HAS_DIAG44)
+		asm volatile("diag 0,0,0x44");
+}
+
+/*
+ * Send cpus emergency shutdown signal. This gives the cpus the
+ * opportunity to complete outstanding interrupts.
+ */
+void smp_emergency_stop(cpumask_t *cpumask)
+{
+	u64 end;
+	int cpu;
+
+	end = get_clock() + (1000000UL << 12);
+	for_each_cpu(cpu, cpumask) {
+		struct pcpu *pcpu = pcpu_devices + cpu;
+		set_bit(ec_stop_cpu, &pcpu->ec_mask);
+		while (__pcpu_sigp(pcpu->address, sigp_emergency_signal,
+				   0, NULL) == sigp_busy &&
+		       get_clock() < end)
+			cpu_relax();
+	}
+	while (get_clock() < end) {
+		for_each_cpu(cpu, cpumask)
+			if (pcpu_stopped(pcpu_devices + cpu))
+				cpumask_clear_cpu(cpu, cpumask);
+		if (cpumask_empty(cpumask))
+			break;
 		cpu_relax();
+	}
 }
 
+/*
+ * Stop all cpus but the current one.
+ */
 void smp_send_stop(void)
 {
 	cpumask_t cpumask;
 	int cpu;
-	u64 end;
 
 	/* Disable all interrupts/machine checks */
 	__load_psw_mask(psw_kernel_bits | PSW_MASK_DAT);
 	trace_hardirqs_off();
 
+	debug_set_critical();
 	cpumask_copy(&cpumask, cpu_online_mask);
 	cpumask_clear_cpu(smp_processor_id(), &cpumask);
 
-	if (oops_in_progress) {
-		/*
-		 * Give the other cpus the opportunity to complete
-		 * outstanding interrupts before stopping them.
-		 */
-		end = get_clock() + (1000000UL << 12);
-		for_each_cpu(cpu, &cpumask) {
-			set_bit(ec_stop_cpu, (unsigned long *)
-				&lowcore_ptr[cpu]->ext_call_fast);
-			while (sigp(cpu, sigp_emergency_signal) == sigp_busy &&
-			       get_clock() < end)
-				cpu_relax();
-		}
-		while (get_clock() < end) {
-			for_each_cpu(cpu, &cpumask)
-				if (cpu_stopped(cpu))
-					cpumask_clear_cpu(cpu, &cpumask);
-			if (cpumask_empty(&cpumask))
-				break;
-			cpu_relax();
-		}
-	}
+	if (oops_in_progress)
+		smp_emergency_stop(&cpumask);
 
 	/* stop all processors */
 	for_each_cpu(cpu, &cpumask) {
-		while (sigp(cpu, sigp_stop) == sigp_busy)
-			cpu_relax();
-		while (!cpu_stopped(cpu))
+		struct pcpu *pcpu = pcpu_devices + cpu;
+		pcpu_sigp_retry(pcpu, sigp_stop, 0);
+		while (!pcpu_stopped(pcpu))
 			cpu_relax();
 	}
 }
 
 /*
+ * Stop the current cpu.
+ */
+void smp_stop_cpu(void)
+{
+	pcpu_sigp_retry(pcpu_devices + smp_processor_id(), sigp_stop, 0);
+	for (;;) ;
+}
+
+/*
  * This is the main routine where commands issued by other
  * cpus are handled.
  */
-
-static void do_ext_call_interrupt(unsigned int ext_int_code,
+static void do_ext_call_interrupt(struct ext_code ext_code,
 				  unsigned int param32, unsigned long param64)
 {
 	unsigned long bits;
+	int cpu;
 
-	if ((ext_int_code & 0xffff) == 0x1202)
-		kstat_cpu(smp_processor_id()).irqs[EXTINT_EXC]++;
+	cpu = smp_processor_id();
+	if (ext_code.code == 0x1202)
+		kstat_cpu(cpu).irqs[EXTINT_EXC]++;
 	else
-		kstat_cpu(smp_processor_id()).irqs[EXTINT_EMS]++;
+		kstat_cpu(cpu).irqs[EXTINT_EMS]++;
 	/*
 	 * handle bit signal external calls
 	 */
-	bits = xchg(&S390_lowcore.ext_call_fast, 0);
+	bits = xchg(&pcpu_devices[cpu].ec_mask, 0);
 
 	if (test_bit(ec_stop_cpu, &bits))
 		smp_stop_cpu();
@@ -238,38 +468,17 @@ static void do_ext_call_interrupt(unsigned int ext_int_code,
 
 }
 
-/*
- * Send an external call sigp to another cpu and return without waiting
- * for its completion.
- */
-static void smp_ext_bitcall(int cpu, int sig)
-{
-	int order;
-
-	/*
-	 * Set signaling bit in lowcore of target cpu and kick it
-	 */
-	set_bit(sig, (unsigned long *) &lowcore_ptr[cpu]->ext_call_fast);
-	while (1) {
-		order = smp_vcpu_scheduled(cpu) ?
-			sigp_external_call : sigp_emergency_signal;
-		if (sigp(cpu, order) != sigp_busy)
-			break;
-		udelay(10);
-	}
-}
-
 void arch_send_call_function_ipi_mask(const struct cpumask *mask)
 {
 	int cpu;
 
 	for_each_cpu(cpu, mask)
-		smp_ext_bitcall(cpu, ec_call_function);
+		pcpu_ec_call(pcpu_devices + cpu, ec_call_function);
 }
 
 void arch_send_call_function_single_ipi(int cpu)
 {
-	smp_ext_bitcall(cpu, ec_call_function_single);
+	pcpu_ec_call(pcpu_devices + cpu, ec_call_function_single);
 }
 
 #ifndef CONFIG_64BIT
@@ -295,15 +504,16 @@ EXPORT_SYMBOL(smp_ptlb_all);
  */
 void smp_send_reschedule(int cpu)
 {
-	smp_ext_bitcall(cpu, ec_schedule);
+	pcpu_ec_call(pcpu_devices + cpu, ec_schedule);
 }
 
 /*
  * parameter area for the set/clear control bit callbacks
  */
 struct ec_creg_mask_parms {
-	unsigned long orvals[16];
-	unsigned long andvals[16];
+	unsigned long orval;
+	unsigned long andval;
+	int cr;
 };
 
 /*
@@ -313,11 +523,9 @@ static void smp_ctl_bit_callback(void *info)
 {
 	struct ec_creg_mask_parms *pp = info;
 	unsigned long cregs[16];
-	int i;
 
 	__ctl_store(cregs, 0, 15);
-	for (i = 0; i <= 15; i++)
-		cregs[i] = (cregs[i] & pp->andvals[i]) | pp->orvals[i];
+	cregs[pp->cr] = (cregs[pp->cr] & pp->andval) | pp->orval;
 	__ctl_load(cregs, 0, 15);
 }
 
@@ -326,11 +534,8 @@ static void smp_ctl_bit_callback(void *info)
  */
 void smp_ctl_set_bit(int cr, int bit)
 {
-	struct ec_creg_mask_parms parms;
+	struct ec_creg_mask_parms parms = { 1UL << bit, -1UL, cr };
 
-	memset(&parms.orvals, 0, sizeof(parms.orvals));
-	memset(&parms.andvals, 0xff, sizeof(parms.andvals));
-	parms.orvals[cr] = 1UL << bit;
 	on_each_cpu(smp_ctl_bit_callback, &parms, 1);
 }
 EXPORT_SYMBOL(smp_ctl_set_bit);
@@ -340,220 +545,178 @@ EXPORT_SYMBOL(smp_ctl_set_bit);
  */
 void smp_ctl_clear_bit(int cr, int bit)
 {
-	struct ec_creg_mask_parms parms;
+	struct ec_creg_mask_parms parms = { 0, ~(1UL << bit), cr };
 
-	memset(&parms.orvals, 0, sizeof(parms.orvals));
-	memset(&parms.andvals, 0xff, sizeof(parms.andvals));
-	parms.andvals[cr] = ~(1UL << bit);
 	on_each_cpu(smp_ctl_bit_callback, &parms, 1);
 }
 EXPORT_SYMBOL(smp_ctl_clear_bit);
 
 #if defined(CONFIG_ZFCPDUMP) || defined(CONFIG_CRASH_DUMP)
 
-static void __init smp_get_save_area(unsigned int cpu, unsigned int phy_cpu)
+struct save_area *zfcpdump_save_areas[NR_CPUS + 1];
+EXPORT_SYMBOL_GPL(zfcpdump_save_areas);
+
+static void __init smp_get_save_area(int cpu, u16 address)
 {
-	if (ipl_info.type != IPL_TYPE_FCP_DUMP && !OLDMEM_BASE)
-		return;
+	void *lc = pcpu_devices[0].lowcore;
+	struct save_area *save_area;
+
 	if (is_kdump_kernel())
 		return;
+	if (!OLDMEM_BASE && (address == boot_cpu_address ||
+			     ipl_info.type != IPL_TYPE_FCP_DUMP))
+		return;
 	if (cpu >= NR_CPUS) {
-		pr_warning("CPU %i exceeds the maximum %i and is excluded from "
-			   "the dump\n", cpu, NR_CPUS - 1);
+		pr_warning("CPU %i exceeds the maximum %i and is excluded "
+			   "from the dump\n", cpu, NR_CPUS - 1);
 		return;
 	}
-	zfcpdump_save_areas[cpu] = kmalloc(sizeof(struct save_area), GFP_KERNEL);
-	while (raw_sigp(phy_cpu, sigp_stop_and_store_status) == sigp_busy)
-		cpu_relax();
-	memcpy_real(zfcpdump_save_areas[cpu],
-		    (void *)(unsigned long) store_prefix() + SAVE_AREA_BASE,
-		    sizeof(struct save_area));
+	save_area = kmalloc(sizeof(struct save_area), GFP_KERNEL);
+	if (!save_area)
+		panic("could not allocate memory for save area\n");
+	zfcpdump_save_areas[cpu] = save_area;
+#ifdef CONFIG_CRASH_DUMP
+	if (address == boot_cpu_address) {
+		/* Copy the registers of the boot cpu. */
+		copy_oldmem_page(1, (void *) save_area, sizeof(*save_area),
+				 SAVE_AREA_BASE - PAGE_SIZE, 0);
+		return;
+	}
+#endif
+	/* Get the registers of a non-boot cpu. */
+	__pcpu_sigp_relax(address, sigp_stop_and_store_status, 0, NULL);
+	memcpy_real(save_area, lc + SAVE_AREA_BASE, sizeof(*save_area));
 }
 
-struct save_area *zfcpdump_save_areas[NR_CPUS + 1];
-EXPORT_SYMBOL_GPL(zfcpdump_save_areas);
-
-#else
-
-static inline void smp_get_save_area(unsigned int cpu, unsigned int phy_cpu) { }
-
-#endif /* CONFIG_ZFCPDUMP */
-
-static int cpu_known(int cpu_id)
+int smp_store_status(int cpu)
 {
-	int cpu;
+	struct pcpu *pcpu;
 
-	for_each_present_cpu(cpu) {
-		if (__cpu_logical_map[cpu] == cpu_id)
-			return 1;
-	}
+	pcpu = pcpu_devices + cpu;
+	if (__pcpu_sigp_relax(pcpu->address, sigp_stop_and_store_status,
+			      0, NULL) != sigp_order_code_accepted)
+		return -EIO;
 	return 0;
 }
 
-static int smp_rescan_cpus_sigp(cpumask_t avail)
-{
-	int cpu_id, logical_cpu;
+#else /* CONFIG_ZFCPDUMP || CONFIG_CRASH_DUMP */
 
-	logical_cpu = cpumask_first(&avail);
-	if (logical_cpu >= nr_cpu_ids)
-		return 0;
-	for (cpu_id = 0; cpu_id <= MAX_CPU_ADDRESS; cpu_id++) {
-		if (cpu_known(cpu_id))
-			continue;
-		__cpu_logical_map[logical_cpu] = cpu_id;
-		cpu_set_polarization(logical_cpu, POLARIZATION_UNKNOWN);
-		if (!cpu_stopped(logical_cpu))
-			continue;
-		set_cpu_present(logical_cpu, true);
-		smp_cpu_state[logical_cpu] = CPU_STATE_CONFIGURED;
-		logical_cpu = cpumask_next(logical_cpu, &avail);
-		if (logical_cpu >= nr_cpu_ids)
-			break;
-	}
-	return 0;
-}
+static inline void smp_get_save_area(int cpu, u16 address) { }
 
-static int smp_rescan_cpus_sclp(cpumask_t avail)
+#endif /* CONFIG_ZFCPDUMP || CONFIG_CRASH_DUMP */
+
+static struct sclp_cpu_info *smp_get_cpu_info(void)
 {
+	static int use_sigp_detection;
 	struct sclp_cpu_info *info;
-	int cpu_id, logical_cpu, cpu;
-	int rc;
-
-	logical_cpu = cpumask_first(&avail);
-	if (logical_cpu >= nr_cpu_ids)
-		return 0;
-	info = kmalloc(sizeof(*info), GFP_KERNEL);
-	if (!info)
-		return -ENOMEM;
-	rc = sclp_get_cpu_info(info);
-	if (rc)
-		goto out;
-	for (cpu = 0; cpu < info->combined; cpu++) {
-		if (info->has_cpu_type && info->cpu[cpu].type != smp_cpu_type)
-			continue;
-		cpu_id = info->cpu[cpu].address;
-		if (cpu_known(cpu_id))
-			continue;
-		__cpu_logical_map[logical_cpu] = cpu_id;
-		cpu_set_polarization(logical_cpu, POLARIZATION_UNKNOWN);
-		set_cpu_present(logical_cpu, true);
-		if (cpu >= info->configured)
-			smp_cpu_state[logical_cpu] = CPU_STATE_STANDBY;
-		else
-			smp_cpu_state[logical_cpu] = CPU_STATE_CONFIGURED;
-		logical_cpu = cpumask_next(logical_cpu, &avail);
-		if (logical_cpu >= nr_cpu_ids)
-			break;
+	int address;
+
+	info = kzalloc(sizeof(*info), GFP_KERNEL);
+	if (info && (use_sigp_detection || sclp_get_cpu_info(info))) {
+		use_sigp_detection = 1;
+		for (address = 0; address <= MAX_CPU_ADDRESS; address++) {
+			if (__pcpu_sigp_relax(address, sigp_sense, 0, NULL) ==
+			    sigp_not_operational)
+				continue;
+			info->cpu[info->configured].address = address;
+			info->configured++;
+		}
+		info->combined = info->configured;
 	}
-out:
-	kfree(info);
-	return rc;
+	return info;
 }
 
-static int __smp_rescan_cpus(void)
+static int __devinit smp_add_present_cpu(int cpu);
+
+static int __devinit __smp_rescan_cpus(struct sclp_cpu_info *info,
+				       int sysfs_add)
 {
+	struct pcpu *pcpu;
 	cpumask_t avail;
+	int cpu, nr, i;
 
+	nr = 0;
 	cpumask_xor(&avail, cpu_possible_mask, cpu_present_mask);
-	if (smp_use_sigp_detection)
-		return smp_rescan_cpus_sigp(avail);
-	else
-		return smp_rescan_cpus_sclp(avail);
+	cpu = cpumask_first(&avail);
+	for (i = 0; (i < info->combined) && (cpu < nr_cpu_ids); i++) {
+		if (info->has_cpu_type && info->cpu[i].type != boot_cpu_type)
+			continue;
+		if (pcpu_find_address(cpu_present_mask, info->cpu[i].address))
+			continue;
+		pcpu = pcpu_devices + cpu;
+		pcpu->address = info->cpu[i].address;
+		pcpu->state = (cpu >= info->configured) ?
+			CPU_STATE_STANDBY : CPU_STATE_CONFIGURED;
+		cpu_set_polarization(cpu, POLARIZATION_UNKNOWN);
+		set_cpu_present(cpu, true);
+		if (sysfs_add && smp_add_present_cpu(cpu) != 0)
+			set_cpu_present(cpu, false);
+		else
+			nr++;
+		cpu = cpumask_next(cpu, &avail);
+	}
+	return nr;
 }
 
 static void __init smp_detect_cpus(void)
 {
 	unsigned int cpu, c_cpus, s_cpus;
 	struct sclp_cpu_info *info;
-	u16 boot_cpu_addr, cpu_addr;
 
-	c_cpus = 1;
-	s_cpus = 0;
-	boot_cpu_addr = __cpu_logical_map[0];
-	info = kmalloc(sizeof(*info), GFP_KERNEL);
+	info = smp_get_cpu_info();
 	if (!info)
 		panic("smp_detect_cpus failed to allocate memory\n");
-#ifdef CONFIG_CRASH_DUMP
-	if (OLDMEM_BASE && !is_kdump_kernel()) {
-		struct save_area *save_area;
-
-		save_area = kmalloc(sizeof(*save_area), GFP_KERNEL);
-		if (!save_area)
-			panic("could not allocate memory for save area\n");
-		copy_oldmem_page(1, (void *) save_area, sizeof(*save_area),
-				 0x200, 0);
-		zfcpdump_save_areas[0] = save_area;
-	}
-#endif
-	/* Use sigp detection algorithm if sclp doesn't work. */
-	if (sclp_get_cpu_info(info)) {
-		smp_use_sigp_detection = 1;
-		for (cpu = 0; cpu <= MAX_CPU_ADDRESS; cpu++) {
-			if (cpu == boot_cpu_addr)
-				continue;
-			if (!raw_cpu_stopped(cpu))
-				continue;
-			smp_get_save_area(c_cpus, cpu);
-			c_cpus++;
-		}
-		goto out;
-	}
-
 	if (info->has_cpu_type) {
 		for (cpu = 0; cpu < info->combined; cpu++) {
-			if (info->cpu[cpu].address == boot_cpu_addr) {
-				smp_cpu_type = info->cpu[cpu].type;
-				break;
-			}
+			if (info->cpu[cpu].address != boot_cpu_address)
+				continue;
+			/* The boot cpu dictates the cpu type. */
+			boot_cpu_type = info->cpu[cpu].type;
+			break;
 		}
 	}
-
+	c_cpus = s_cpus = 0;
 	for (cpu = 0; cpu < info->combined; cpu++) {
-		if (info->has_cpu_type && info->cpu[cpu].type != smp_cpu_type)
+		if (info->has_cpu_type && info->cpu[cpu].type != boot_cpu_type)
 			continue;
-		cpu_addr = info->cpu[cpu].address;
-		if (cpu_addr == boot_cpu_addr)
-			continue;
-		if (!raw_cpu_stopped(cpu_addr)) {
+		if (cpu < info->configured) {
+			smp_get_save_area(c_cpus, info->cpu[cpu].address);
+			c_cpus++;
+		} else
 			s_cpus++;
-			continue;
-		}
-		smp_get_save_area(c_cpus, cpu_addr);
-		c_cpus++;
 	}
-out:
-	kfree(info);
 	pr_info("%d configured CPUs, %d standby CPUs\n", c_cpus, s_cpus);
 	get_online_cpus();
-	__smp_rescan_cpus();
+	__smp_rescan_cpus(info, 0);
 	put_online_cpus();
+	kfree(info);
 }
 
 /*
  *	Activate a secondary processor.
  */
-int __cpuinit start_secondary(void *cpuvoid)
+static void __cpuinit smp_start_secondary(void *cpuvoid)
 {
+	S390_lowcore.last_update_clock = get_clock();
+	S390_lowcore.restart_stack = (unsigned long) restart_stack;
+	S390_lowcore.restart_fn = (unsigned long) do_restart;
+	S390_lowcore.restart_data = 0;
+	S390_lowcore.restart_source = -1UL;
+	restore_access_regs(S390_lowcore.access_regs_save_area);
+	__ctl_load(S390_lowcore.cregs_save_area, 0, 15);
+	__load_psw_mask(psw_kernel_bits | PSW_MASK_DAT);
 	cpu_init();
 	preempt_disable();
 	init_cpu_timer();
 	init_cpu_vtimer();
 	pfault_init();
-
 	notify_cpu_starting(smp_processor_id());
 	ipi_call_lock();
 	set_cpu_online(smp_processor_id(), true);
 	ipi_call_unlock();
-	__ctl_clear_bit(0, 28); /* Disable lowcore protection */
-	S390_lowcore.restart_psw.mask =
-		PSW_DEFAULT_KEY | PSW_MASK_BASE | PSW_MASK_EA | PSW_MASK_BA;
-	S390_lowcore.restart_psw.addr =
-		PSW_ADDR_AMODE | (unsigned long) psw_restart_int_handler;
-	__ctl_set_bit(0, 28); /* Enable lowcore protection */
 	local_irq_enable();
 	/* cpu_idle will call schedule for us */
 	cpu_idle();
-	return 0;
 }
 
 struct create_idle {
@@ -572,82 +735,20 @@ static void __cpuinit smp_fork_idle(struct work_struct *work)
 	complete(&c_idle->done);
 }
 
-static int __cpuinit smp_alloc_lowcore(int cpu)
-{
-	unsigned long async_stack, panic_stack;
-	struct _lowcore *lowcore;
-
-	lowcore = (void *) __get_free_pages(GFP_KERNEL | GFP_DMA, LC_ORDER);
-	if (!lowcore)
-		return -ENOMEM;
-	async_stack = __get_free_pages(GFP_KERNEL, ASYNC_ORDER);
-	panic_stack = __get_free_page(GFP_KERNEL);
-	if (!panic_stack || !async_stack)
-		goto out;
-	memcpy(lowcore, &S390_lowcore, 512);
-	memset((char *)lowcore + 512, 0, sizeof(*lowcore) - 512);
-	lowcore->async_stack = async_stack + ASYNC_SIZE;
-	lowcore->panic_stack = panic_stack + PAGE_SIZE;
-	lowcore->restart_psw.mask =
-		PSW_DEFAULT_KEY | PSW_MASK_BASE | PSW_MASK_EA | PSW_MASK_BA;
-	lowcore->restart_psw.addr =
-		PSW_ADDR_AMODE | (unsigned long) restart_int_handler;
-	if (user_mode != HOME_SPACE_MODE)
-		lowcore->restart_psw.mask |= PSW_ASC_HOME;
-#ifndef CONFIG_64BIT
-	if (MACHINE_HAS_IEEE) {
-		unsigned long save_area;
-
-		save_area = get_zeroed_page(GFP_KERNEL);
-		if (!save_area)
-			goto out;
-		lowcore->extended_save_area_addr = (u32) save_area;
-	}
-#else
-	if (vdso_alloc_per_cpu(cpu, lowcore))
-		goto out;
-#endif
-	lowcore_ptr[cpu] = lowcore;
-	return 0;
-
-out:
-	free_page(panic_stack);
-	free_pages(async_stack, ASYNC_ORDER);
-	free_pages((unsigned long) lowcore, LC_ORDER);
-	return -ENOMEM;
-}
-
-static void smp_free_lowcore(int cpu)
-{
-	struct _lowcore *lowcore;
-
-	lowcore = lowcore_ptr[cpu];
-#ifndef CONFIG_64BIT
-	if (MACHINE_HAS_IEEE)
-		free_page((unsigned long) lowcore->extended_save_area_addr);
-#else
-	vdso_free_per_cpu(cpu, lowcore);
-#endif
-	free_page(lowcore->panic_stack - PAGE_SIZE);
-	free_pages(lowcore->async_stack - ASYNC_SIZE, ASYNC_ORDER);
-	free_pages((unsigned long) lowcore, LC_ORDER);
-	lowcore_ptr[cpu] = NULL;
-}
-
 /* Upping and downing of CPUs */
 int __cpuinit __cpu_up(unsigned int cpu)
 {
-	struct _lowcore *cpu_lowcore;
 	struct create_idle c_idle;
-	struct task_struct *idle;
-	struct stack_frame *sf;
-	u32 lowcore;
-	int ccode;
+	struct pcpu *pcpu;
+	int rc;
 
-	if (smp_cpu_state[cpu] != CPU_STATE_CONFIGURED)
+	pcpu = pcpu_devices + cpu;
+	if (pcpu->state != CPU_STATE_CONFIGURED)
+		return -EIO;
+	if (pcpu_sigp_retry(pcpu, sigp_initial_cpu_reset, 0) !=
+	    sigp_order_code_accepted)
 		return -EIO;
-	idle = current_set[cpu];
-	if (!idle) {
+	if (!pcpu->idle) {
 		c_idle.done = COMPLETION_INITIALIZER_ONSTACK(c_idle.done);
 		INIT_WORK_ONSTACK(&c_idle.work, smp_fork_idle);
 		c_idle.cpu = cpu;
@@ -655,68 +756,28 @@ int __cpuinit __cpu_up(unsigned int cpu)
 		wait_for_completion(&c_idle.done);
 		if (IS_ERR(c_idle.idle))
 			return PTR_ERR(c_idle.idle);
-		idle = c_idle.idle;
-		current_set[cpu] = c_idle.idle;
+		pcpu->idle = c_idle.idle;
 	}
-	init_idle(idle, cpu);
-	if (smp_alloc_lowcore(cpu))
-		return -ENOMEM;
-	do {
-		ccode = sigp(cpu, sigp_initial_cpu_reset);
-		if (ccode == sigp_busy)
-			udelay(10);
-		if (ccode == sigp_not_operational)
-			goto err_out;
-	} while (ccode == sigp_busy);
-
-	lowcore = (u32)(unsigned long)lowcore_ptr[cpu];
-	while (sigp_p(lowcore, cpu, sigp_set_prefix) == sigp_busy)
-		udelay(10);
-
-	cpu_lowcore = lowcore_ptr[cpu];
-	cpu_lowcore->kernel_stack = (unsigned long)
-		task_stack_page(idle) + THREAD_SIZE;
-	cpu_lowcore->thread_info = (unsigned long) task_thread_info(idle);
-	sf = (struct stack_frame *) (cpu_lowcore->kernel_stack
-				     - sizeof(struct pt_regs)
-				     - sizeof(struct stack_frame));
-	memset(sf, 0, sizeof(struct stack_frame));
-	sf->gprs[9] = (unsigned long) sf;
-	cpu_lowcore->gpregs_save_area[15] = (unsigned long) sf;
-	__ctl_store(cpu_lowcore->cregs_save_area, 0, 15);
-	atomic_inc(&init_mm.context.attach_count);
-	asm volatile(
-		"	stam	0,15,0(%0)"
-		: : "a" (&cpu_lowcore->access_regs_save_area) : "memory");
-	cpu_lowcore->percpu_offset = __per_cpu_offset[cpu];
-	cpu_lowcore->current_task = (unsigned long) idle;
-	cpu_lowcore->cpu_nr = cpu;
-	cpu_lowcore->kernel_asce = S390_lowcore.kernel_asce;
-	cpu_lowcore->machine_flags = S390_lowcore.machine_flags;
-	cpu_lowcore->ftrace_func = S390_lowcore.ftrace_func;
-	memcpy(cpu_lowcore->stfle_fac_list, S390_lowcore.stfle_fac_list,
-	       MAX_FACILITY_BIT/8);
-	eieio();
-
-	while (sigp(cpu, sigp_restart) == sigp_busy)
-		udelay(10);
-
+	init_idle(pcpu->idle, cpu);
+	rc = pcpu_alloc_lowcore(pcpu, cpu);
+	if (rc)
+		return rc;
+	pcpu_prepare_secondary(pcpu, cpu);
+	pcpu_attach_task(pcpu, pcpu->idle);
+	pcpu_start_fn(pcpu, smp_start_secondary, NULL);
 	while (!cpu_online(cpu))
 		cpu_relax();
 	return 0;
-
-err_out:
-	smp_free_lowcore(cpu);
-	return -EIO;
 }
 
 static int __init setup_possible_cpus(char *s)
 {
-	int pcpus, cpu;
+	int max, cpu;
 
-	pcpus = simple_strtoul(s, NULL, 0);
+	if (kstrtoint(s, 0, &max) < 0)
+		return 0;
 	init_cpu_possible(cpumask_of(0));
-	for (cpu = 1; cpu < pcpus && cpu < nr_cpu_ids; cpu++)
+	for (cpu = 1; cpu < max && cpu < nr_cpu_ids; cpu++)
 		set_cpu_possible(cpu, true);
 	return 0;
 }
@@ -726,113 +787,79 @@ early_param("possible_cpus", setup_possible_cpus);
 
 int __cpu_disable(void)
 {
-	struct ec_creg_mask_parms cr_parms;
-	int cpu = smp_processor_id();
-
-	set_cpu_online(cpu, false);
+	unsigned long cregs[16];
 
-	/* Disable pfault pseudo page faults on this cpu. */
+	set_cpu_online(smp_processor_id(), false);
+	/* Disable pseudo page faults on this cpu. */
 	pfault_fini();
-
-	memset(&cr_parms.orvals, 0, sizeof(cr_parms.orvals));
-	memset(&cr_parms.andvals, 0xff, sizeof(cr_parms.andvals));
-
-	/* disable all external interrupts */
-	cr_parms.orvals[0] = 0;
-	cr_parms.andvals[0] = ~(1 << 15 | 1 << 14 | 1 << 13 | 1 << 11 |
-				1 << 10 | 1 <<	9 | 1 <<  6 | 1 <<  5 |
-				1 <<  4);
-	/* disable all I/O interrupts */
-	cr_parms.orvals[6] = 0;
-	cr_parms.andvals[6] = ~(1 << 31 | 1 << 30 | 1 << 29 | 1 << 28 |
-				1 << 27 | 1 << 26 | 1 << 25 | 1 << 24);
-	/* disable most machine checks */
-	cr_parms.orvals[14] = 0;
-	cr_parms.andvals[14] = ~(1 << 28 | 1 << 27 | 1 << 26 |
-				 1 << 25 | 1 << 24);
-
-	smp_ctl_bit_callback(&cr_parms);
-
+	/* Disable interrupt sources via control register. */
+	__ctl_store(cregs, 0, 15);
+	cregs[0]  &= ~0x0000ee70UL;	/* disable all external interrupts */
+	cregs[6]  &= ~0xff000000UL;	/* disable all I/O interrupts */
+	cregs[14] &= ~0x1f000000UL;	/* disable most machine checks */
+	__ctl_load(cregs, 0, 15);
 	return 0;
 }
 
 void __cpu_die(unsigned int cpu)
 {
+	struct pcpu *pcpu;
+
 	/* Wait until target cpu is down */
-	while (!cpu_stopped(cpu))
+	pcpu = pcpu_devices + cpu;
+	while (!pcpu_stopped(pcpu))
 		cpu_relax();
-	while (sigp_p(0, cpu, sigp_set_prefix) == sigp_busy)
-		udelay(10);
-	smp_free_lowcore(cpu);
+	pcpu_free_lowcore(pcpu);
 	atomic_dec(&init_mm.context.attach_count);
 }
 
 void __noreturn cpu_die(void)
 {
 	idle_task_exit();
-	while (sigp(smp_processor_id(), sigp_stop) == sigp_busy)
-		cpu_relax();
-	for (;;);
+	pcpu_sigp_retry(pcpu_devices + smp_processor_id(), sigp_stop, 0);
+	for (;;) ;
 }
 
 #endif /* CONFIG_HOTPLUG_CPU */
 
-void __init smp_prepare_cpus(unsigned int max_cpus)
+static void smp_call_os_info_init_fn(void)
 {
-#ifndef CONFIG_64BIT
-	unsigned long save_area = 0;
-#endif
-	unsigned long async_stack, panic_stack;
-	struct _lowcore *lowcore;
+	int (*init_fn)(void);
+	unsigned long size;
 
-	smp_detect_cpus();
+	init_fn = os_info_old_entry(OS_INFO_INIT_FN, &size);
+	if (!init_fn)
+		return;
+	init_fn();
+}
 
+void __init smp_prepare_cpus(unsigned int max_cpus)
+{
 	/* request the 0x1201 emergency signal external interrupt */
 	if (register_external_interrupt(0x1201, do_ext_call_interrupt) != 0)
 		panic("Couldn't request external interrupt 0x1201");
 	/* request the 0x1202 external call external interrupt */
 	if (register_external_interrupt(0x1202, do_ext_call_interrupt) != 0)
 		panic("Couldn't request external interrupt 0x1202");
-
-	/* Reallocate current lowcore, but keep its contents. */
-	lowcore = (void *) __get_free_pages(GFP_KERNEL | GFP_DMA, LC_ORDER);
-	panic_stack = __get_free_page(GFP_KERNEL);
-	async_stack = __get_free_pages(GFP_KERNEL, ASYNC_ORDER);
-	BUG_ON(!lowcore || !panic_stack || !async_stack);
-#ifndef CONFIG_64BIT
-	if (MACHINE_HAS_IEEE)
-		save_area = get_zeroed_page(GFP_KERNEL);
-#endif
-	local_irq_disable();
-	local_mcck_disable();
-	lowcore_ptr[smp_processor_id()] = lowcore;
-	*lowcore = S390_lowcore;
-	lowcore->panic_stack = panic_stack + PAGE_SIZE;
-	lowcore->async_stack = async_stack + ASYNC_SIZE;
-#ifndef CONFIG_64BIT
-	if (MACHINE_HAS_IEEE)
-		lowcore->extended_save_area_addr = (u32) save_area;
-#endif
-	set_prefix((u32)(unsigned long) lowcore);
-	local_mcck_enable();
-	local_irq_enable();
-#ifdef CONFIG_64BIT
-	if (vdso_alloc_per_cpu(smp_processor_id(), &S390_lowcore))
-		BUG();
-#endif
+	smp_call_os_info_init_fn();
+	smp_detect_cpus();
 }
 
 void __init smp_prepare_boot_cpu(void)
 {
-	BUG_ON(smp_processor_id() != 0);
-
-	current_thread_info()->cpu = 0;
-	set_cpu_present(0, true);
-	set_cpu_online(0, true);
+	struct pcpu *pcpu = pcpu_devices;
+
+	boot_cpu_address = stap();
+	pcpu->idle = current;
+	pcpu->state = CPU_STATE_CONFIGURED;
+	pcpu->address = boot_cpu_address;
+	pcpu->lowcore = (struct _lowcore *)(unsigned long) store_prefix();
+	pcpu->async_stack = S390_lowcore.async_stack - ASYNC_SIZE;
+	pcpu->panic_stack = S390_lowcore.panic_stack - PAGE_SIZE;
 	S390_lowcore.percpu_offset = __per_cpu_offset[0];
-	current_set[0] = current;
-	smp_cpu_state[0] = CPU_STATE_CONFIGURED;
 	cpu_set_polarization(0, POLARIZATION_UNKNOWN);
+	set_cpu_present(0, true);
+	set_cpu_online(0, true);
 }
 
 void __init smp_cpus_done(unsigned int max_cpus)
@@ -842,7 +869,6 @@ void __init smp_cpus_done(unsigned int max_cpus)
 void __init smp_setup_processor_id(void)
 {
 	S390_lowcore.cpu_nr = 0;
-	__cpu_logical_map[0] = stap();
 }
 
 /*
@@ -858,56 +884,57 @@ int setup_profiling_timer(unsigned int multiplier)
 
 #ifdef CONFIG_HOTPLUG_CPU
 static ssize_t cpu_configure_show(struct device *dev,
-				struct device_attribute *attr, char *buf)
+				  struct device_attribute *attr, char *buf)
 {
 	ssize_t count;
 
 	mutex_lock(&smp_cpu_state_mutex);
-	count = sprintf(buf, "%d\n", smp_cpu_state[dev->id]);
+	count = sprintf(buf, "%d\n", pcpu_devices[dev->id].state);
 	mutex_unlock(&smp_cpu_state_mutex);
 	return count;
 }
 
 static ssize_t cpu_configure_store(struct device *dev,
-				  struct device_attribute *attr,
-				  const char *buf, size_t count)
+				   struct device_attribute *attr,
+				   const char *buf, size_t count)
 {
-	int cpu = dev->id;
-	int val, rc;
+	struct pcpu *pcpu;
+	int cpu, val, rc;
 	char delim;
 
 	if (sscanf(buf, "%d %c", &val, &delim) != 1)
 		return -EINVAL;
 	if (val != 0 && val != 1)
 		return -EINVAL;
-
 	get_online_cpus();
 	mutex_lock(&smp_cpu_state_mutex);
 	rc = -EBUSY;
 	/* disallow configuration changes of online cpus and cpu 0 */
+	cpu = dev->id;
 	if (cpu_online(cpu) || cpu == 0)
 		goto out;
+	pcpu = pcpu_devices + cpu;
 	rc = 0;
 	switch (val) {
 	case 0:
-		if (smp_cpu_state[cpu] == CPU_STATE_CONFIGURED) {
-			rc = sclp_cpu_deconfigure(__cpu_logical_map[cpu]);
-			if (!rc) {
-				smp_cpu_state[cpu] = CPU_STATE_STANDBY;
-				cpu_set_polarization(cpu, POLARIZATION_UNKNOWN);
-				topology_expect_change();
-			}
-		}
+		if (pcpu->state != CPU_STATE_CONFIGURED)
+			break;
+		rc = sclp_cpu_deconfigure(pcpu->address);
+		if (rc)
+			break;
+		pcpu->state = CPU_STATE_STANDBY;
+		cpu_set_polarization(cpu, POLARIZATION_UNKNOWN);
+		topology_expect_change();
 		break;
 	case 1:
-		if (smp_cpu_state[cpu] == CPU_STATE_STANDBY) {
-			rc = sclp_cpu_configure(__cpu_logical_map[cpu]);
-			if (!rc) {
-				smp_cpu_state[cpu] = CPU_STATE_CONFIGURED;
-				cpu_set_polarization(cpu, POLARIZATION_UNKNOWN);
-				topology_expect_change();
-			}
-		}
+		if (pcpu->state != CPU_STATE_STANDBY)
+			break;
+		rc = sclp_cpu_configure(pcpu->address);
+		if (rc)
+			break;
+		pcpu->state = CPU_STATE_CONFIGURED;
+		cpu_set_polarization(cpu, POLARIZATION_UNKNOWN);
+		topology_expect_change();
 		break;
 	default:
 		break;
@@ -923,7 +950,7 @@ static DEVICE_ATTR(configure, 0644, cpu_configure_show, cpu_configure_store);
 static ssize_t show_cpu_address(struct device *dev,
 				struct device_attribute *attr, char *buf)
 {
-	return sprintf(buf, "%d\n", __cpu_logical_map[dev->id]);
+	return sprintf(buf, "%d\n", pcpu_devices[dev->id].address);
 }
 static DEVICE_ATTR(address, 0444, show_cpu_address, NULL);
 
@@ -955,22 +982,16 @@ static DEVICE_ATTR(capability, 0444, show_capability, NULL);
 static ssize_t show_idle_count(struct device *dev,
 				struct device_attribute *attr, char *buf)
 {
-	struct s390_idle_data *idle;
+	struct s390_idle_data *idle = &per_cpu(s390_idle, dev->id);
 	unsigned long long idle_count;
 	unsigned int sequence;
 
-	idle = &per_cpu(s390_idle, dev->id);
-repeat:
-	sequence = idle->sequence;
-	smp_rmb();
-	if (sequence & 1)
-		goto repeat;
-	idle_count = idle->idle_count;
-	if (idle->idle_enter)
-		idle_count++;
-	smp_rmb();
-	if (idle->sequence != sequence)
-		goto repeat;
+	do {
+		sequence = ACCESS_ONCE(idle->sequence);
+		idle_count = ACCESS_ONCE(idle->idle_count);
+		if (ACCESS_ONCE(idle->idle_enter))
+			idle_count++;
+	} while ((sequence & 1) || (idle->sequence != sequence));
 	return sprintf(buf, "%llu\n", idle_count);
 }
 static DEVICE_ATTR(idle_count, 0444, show_idle_count, NULL);
@@ -978,24 +999,18 @@ static DEVICE_ATTR(idle_count, 0444, show_idle_count, NULL);
 static ssize_t show_idle_time(struct device *dev,
 				struct device_attribute *attr, char *buf)
 {
-	struct s390_idle_data *idle;
-	unsigned long long now, idle_time, idle_enter;
+	struct s390_idle_data *idle = &per_cpu(s390_idle, dev->id);
+	unsigned long long now, idle_time, idle_enter, idle_exit;
 	unsigned int sequence;
 
-	idle = &per_cpu(s390_idle, dev->id);
-	now = get_clock();
-repeat:
-	sequence = idle->sequence;
-	smp_rmb();
-	if (sequence & 1)
-		goto repeat;
-	idle_time = idle->idle_time;
-	idle_enter = idle->idle_enter;
-	if (idle_enter != 0ULL && idle_enter < now)
-		idle_time += now - idle_enter;
-	smp_rmb();
-	if (idle->sequence != sequence)
-		goto repeat;
+	do {
+		now = get_clock();
+		sequence = ACCESS_ONCE(idle->sequence);
+		idle_time = ACCESS_ONCE(idle->idle_time);
+		idle_enter = ACCESS_ONCE(idle->idle_enter);
+		idle_exit = ACCESS_ONCE(idle->idle_exit);
+	} while ((sequence & 1) || (idle->sequence != sequence));
+	idle_time += idle_enter ? ((idle_exit ? : now) - idle_enter) : 0;
 	return sprintf(buf, "%llu\n", idle_time >> 12);
 }
 static DEVICE_ATTR(idle_time_us, 0444, show_idle_time, NULL);
@@ -1015,7 +1030,7 @@ static int __cpuinit smp_cpu_notify(struct notifier_block *self,
 				    unsigned long action, void *hcpu)
 {
 	unsigned int cpu = (unsigned int)(long)hcpu;
-	struct cpu *c = &per_cpu(cpu_devices, cpu);
+	struct cpu *c = &pcpu_devices[cpu].cpu;
 	struct device *s = &c->dev;
 	struct s390_idle_data *idle;
 	int err = 0;
@@ -1041,7 +1056,7 @@ static struct notifier_block __cpuinitdata smp_cpu_nb = {
 
 static int __devinit smp_add_present_cpu(int cpu)
 {
-	struct cpu *c = &per_cpu(cpu_devices, cpu);
+	struct cpu *c = &pcpu_devices[cpu].cpu;
 	struct device *s = &c->dev;
 	int rc;
 
@@ -1079,29 +1094,21 @@ out:
 
 int __ref smp_rescan_cpus(void)
 {
-	cpumask_t newcpus;
-	int cpu;
-	int rc;
+	struct sclp_cpu_info *info;
+	int nr;
 
+	info = smp_get_cpu_info();
+	if (!info)
+		return -ENOMEM;
 	get_online_cpus();
 	mutex_lock(&smp_cpu_state_mutex);
-	cpumask_copy(&newcpus, cpu_present_mask);
-	rc = __smp_rescan_cpus();
-	if (rc)
-		goto out;
-	cpumask_andnot(&newcpus, cpu_present_mask, &newcpus);
-	for_each_cpu(cpu, &newcpus) {
-		rc = smp_add_present_cpu(cpu);
-		if (rc)
-			set_cpu_present(cpu, false);
-	}
-	rc = 0;
-out:
+	nr = __smp_rescan_cpus(info, 1);
 	mutex_unlock(&smp_cpu_state_mutex);
 	put_online_cpus();
-	if (!cpumask_empty(&newcpus))
+	kfree(info);
+	if (nr)
 		topology_schedule_update();
-	return rc;
+	return 0;
 }
 
 static ssize_t __ref rescan_store(struct device *dev,