1 files changed, 168 insertions, 92 deletions

diff --git a/arch/powerpc/platforms/powermac/smp.c b/arch/powerpc/platforms/powermac/smp.c
index c95215f4f8b..5cbd4d67d5c 100644
--- a/arch/powerpc/platforms/powermac/smp.c
+++ b/arch/powerpc/platforms/powermac/smp.c
@@ -35,7 +35,7 @@
 #include <linux/compiler.h>
 
 #include <asm/ptrace.h>
-#include <asm/atomic.h>
+#include <linux/atomic.h>
 #include <asm/code-patching.h>
 #include <asm/irq.h>
 #include <asm/page.h>
@@ -70,7 +70,7 @@ static void (*pmac_tb_freeze)(int freeze);
 static u64 timebase;
 static int tb_req;
 
-#ifdef CONFIG_PPC32
+#ifdef CONFIG_PPC_PMAC32_PSURGE
 
 /*
  * Powersurge (old powermac SMP) support.
@@ -124,6 +124,10 @@ static volatile u32 __iomem *psurge_start;
 /* what sort of powersurge board we have */
 static int psurge_type = PSURGE_NONE;
 
+/* irq for secondary cpus to report */
+static struct irq_domain *psurge_host;
+int psurge_secondary_virq;
+
 /*
  * Set and clear IPIs for powersurge.
  */
@@ -156,51 +160,51 @@ static inline void psurge_clr_ipi(int cpu)
 /*
  * On powersurge (old SMP powermac architecture) we don't have
  * separate IPIs for separate messages like openpic does.  Instead
- * we have a bitmap for each processor, where a 1 bit means that
- * the corresponding message is pending for that processor.
- * Ideally each cpu's entry would be in a different cache line.
+ * use the generic demux helpers
  *  -- paulus.
  */
-static unsigned long psurge_smp_message[NR_CPUS];
-
-void psurge_smp_message_recv(void)
+static irqreturn_t psurge_ipi_intr(int irq, void *d)
 {
-	int cpu = smp_processor_id();
-	int msg;
-
-	/* clear interrupt */
-	psurge_clr_ipi(cpu);
+	psurge_clr_ipi(smp_processor_id());
+	smp_ipi_demux();
 
-	if (num_online_cpus() < 2)
-		return;
+	return IRQ_HANDLED;
+}
 
-	/* make sure there is a message there */
-	for (msg = 0; msg < 4; msg++)
-		if (test_and_clear_bit(msg, &psurge_smp_message[cpu]))
-			smp_message_recv(msg);
+static void smp_psurge_cause_ipi(int cpu, unsigned long data)
+{
+	psurge_set_ipi(cpu);
 }
 
-irqreturn_t psurge_primary_intr(int irq, void *d)
+static int psurge_host_map(struct irq_domain *h, unsigned int virq,
+			 irq_hw_number_t hw)
 {
-	psurge_smp_message_recv();
-	return IRQ_HANDLED;
+	irq_set_chip_and_handler(virq, &dummy_irq_chip, handle_percpu_irq);
+
+	return 0;
 }
 
-static void smp_psurge_message_pass(int target, int msg)
+static const struct irq_domain_ops psurge_host_ops = {
+	.map	= psurge_host_map,
+};
+
+static int psurge_secondary_ipi_init(void)
 {
-	int i;
+	int rc = -ENOMEM;
 
-	if (num_online_cpus() < 2)
-		return;
+	psurge_host = irq_domain_add_nomap(NULL, ~0, &psurge_host_ops, NULL);
 
-	for_each_online_cpu(i) {
-		if (target == MSG_ALL
-		    || (target == MSG_ALL_BUT_SELF && i != smp_processor_id())
-		    || target == i) {
-			set_bit(msg, &psurge_smp_message[i]);
-			psurge_set_ipi(i);
-		}
-	}
+	if (psurge_host)
+		psurge_secondary_virq = irq_create_direct_mapping(psurge_host);
+
+	if (psurge_secondary_virq)
+		rc = request_irq(psurge_secondary_virq, psurge_ipi_intr,
+			IRQF_PERCPU | IRQF_NO_THREAD, "IPI", NULL);
+
+	if (rc)
+		pr_err("Failed to setup secondary cpu IPI\n");
+
+	return rc;
 }
 
 /*
@@ -311,6 +315,9 @@ static int __init smp_psurge_probe(void)
 		ncpus = 2;
 	}
 
+	if (psurge_secondary_ipi_init())
+		return 1;
+
 	psurge_start = ioremap(PSURGE_START, 4);
 	psurge_pri_intr = ioremap(PSURGE_PRI_INTR, 4);
 
@@ -329,7 +336,7 @@ static int __init smp_psurge_probe(void)
 	return ncpus;
 }
 
-static void __init smp_psurge_kick_cpu(int nr)
+static int __init smp_psurge_kick_cpu(int nr)
 {
 	unsigned long start = __pa(__secondary_start_pmac_0) + nr * 8;
 	unsigned long a, flags;
@@ -394,17 +401,19 @@ static void __init smp_psurge_kick_cpu(int nr)
 		psurge_set_ipi(1);
 
 	if (ppc_md.progress) ppc_md.progress("smp_psurge_kick_cpu - done", 0x354);
+
+	return 0;
 }
 
 static struct irqaction psurge_irqaction = {
-	.handler = psurge_primary_intr,
-	.flags = IRQF_DISABLED,
+	.handler = psurge_ipi_intr,
+	.flags = IRQF_PERCPU | IRQF_NO_THREAD,
 	.name = "primary IPI",
 };
 
 static void __init smp_psurge_setup_cpu(int cpu_nr)
 {
-	if (cpu_nr != 0)
+	if (cpu_nr != 0 || !psurge_start)
 		return;
 
 	/* reset the entry point so if we get another intr we won't
@@ -437,14 +446,15 @@ void __init smp_psurge_give_timebase(void)
 
 /* PowerSurge-style Macs */
 struct smp_ops_t psurge_smp_ops = {
-	.message_pass	= smp_psurge_message_pass,
+	.message_pass	= NULL,	/* Use smp_muxed_ipi_message_pass */
+	.cause_ipi	= smp_psurge_cause_ipi,
 	.probe		= smp_psurge_probe,
 	.kick_cpu	= smp_psurge_kick_cpu,
 	.setup_cpu	= smp_psurge_setup_cpu,
 	.give_timebase	= smp_psurge_give_timebase,
 	.take_timebase	= smp_psurge_take_timebase,
 };
-#endif /* CONFIG_PPC32 - actually powersurge support */
+#endif /* CONFIG_PPC_PMAC32_PSURGE */
 
 /*
  * Core 99 and later support
@@ -474,7 +484,7 @@ static void smp_core99_give_timebase(void)
 }
 
 
-static void __devinit smp_core99_take_timebase(void)
+static void smp_core99_take_timebase(void)
 {
 	unsigned long flags;
 
@@ -659,7 +669,7 @@ static void smp_core99_gpio_tb_freeze(int freeze)
 volatile static long int core99_l2_cache;
 volatile static long int core99_l3_cache;
 
-static void __devinit core99_init_caches(int cpu)
+static void core99_init_caches(int cpu)
 {
 #ifndef CONFIG_PPC64
 	if (!cpu_has_feature(CPU_FTR_L2CR))
@@ -791,14 +801,14 @@ static int __init smp_core99_probe(void)
 	return ncpus;
 }
 
-static void __devinit smp_core99_kick_cpu(int nr)
+static int smp_core99_kick_cpu(int nr)
 {
 	unsigned int save_vector;
 	unsigned long target, flags;
 	unsigned int *vector = (unsigned int *)(PAGE_OFFSET+0x100);
 
 	if (nr < 0 || nr > 3)
-		return;
+		return -ENOENT;
 
 	if (ppc_md.progress)
 		ppc_md.progress("smp_core99_kick_cpu", 0x346);
@@ -830,9 +840,11 @@ static void __devinit smp_core99_kick_cpu(int nr)
 
 	local_irq_restore(flags);
 	if (ppc_md.progress) ppc_md.progress("smp_core99_kick_cpu done", 0x347);
+
+	return 0;
 }
 
-static void __devinit smp_core99_setup_cpu(int cpu_nr)
+static void smp_core99_setup_cpu(int cpu_nr)
 {
 	/* Setup L2/L3 */
 	if (cpu_nr != 0)
@@ -840,92 +852,151 @@ static void __devinit smp_core99_setup_cpu(int cpu_nr)
 
 	/* Setup openpic */
 	mpic_setup_this_cpu();
+}
 
-	if (cpu_nr == 0) {
 #ifdef CONFIG_PPC64
-		extern void g5_phy_disable_cpu1(void);
+#ifdef CONFIG_HOTPLUG_CPU
+static int smp_core99_cpu_notify(struct notifier_block *self,
+				 unsigned long action, void *hcpu)
+{
+	int rc;
 
-		/* Close i2c bus if it was used for tb sync */
+	switch(action) {
+	case CPU_UP_PREPARE:
+	case CPU_UP_PREPARE_FROZEN:
+		/* Open i2c bus if it was used for tb sync */
 		if (pmac_tb_clock_chip_host) {
-			pmac_i2c_close(pmac_tb_clock_chip_host);
-			pmac_tb_clock_chip_host	= NULL;
+			rc = pmac_i2c_open(pmac_tb_clock_chip_host, 1);
+			if (rc) {
+				pr_err("Failed to open i2c bus for time sync\n");
+				return notifier_from_errno(rc);
+			}
 		}
+		break;
+	case CPU_ONLINE:
+	case CPU_UP_CANCELED:
+		/* Close i2c bus if it was used for tb sync */
+		if (pmac_tb_clock_chip_host)
+			pmac_i2c_close(pmac_tb_clock_chip_host);
+		break;
+	default:
+		break;
+	}
+	return NOTIFY_OK;
+}
 
-		/* If we didn't start the second CPU, we must take
-		 * it off the bus
-		 */
-		if (of_machine_is_compatible("MacRISC4") &&
-		    num_online_cpus() < 2)		
-			g5_phy_disable_cpu1();
-#endif /* CONFIG_PPC64 */
+static struct notifier_block smp_core99_cpu_nb = {
+	.notifier_call	= smp_core99_cpu_notify,
+};
+#endif /* CONFIG_HOTPLUG_CPU */
+
+static void __init smp_core99_bringup_done(void)
+{
+	extern void g5_phy_disable_cpu1(void);
 
-		if (ppc_md.progress)
-			ppc_md.progress("core99_setup_cpu 0 done", 0x349);
+	/* Close i2c bus if it was used for tb sync */
+	if (pmac_tb_clock_chip_host)
+		pmac_i2c_close(pmac_tb_clock_chip_host);
+
+	/* If we didn't start the second CPU, we must take
+	 * it off the bus.
+	 */
+	if (of_machine_is_compatible("MacRISC4") &&
+	    num_online_cpus() < 2) {
+		set_cpu_present(1, false);
+		g5_phy_disable_cpu1();
 	}
-}
+#ifdef CONFIG_HOTPLUG_CPU
+	register_cpu_notifier(&smp_core99_cpu_nb);
+#endif
 
+	if (ppc_md.progress)
+		ppc_md.progress("smp_core99_bringup_done", 0x349);
+}
+#endif /* CONFIG_PPC64 */
 
-#if defined(CONFIG_HOTPLUG_CPU) && defined(CONFIG_PPC32)
+#ifdef CONFIG_HOTPLUG_CPU
 
-int smp_core99_cpu_disable(void)
+static int smp_core99_cpu_disable(void)
 {
-	set_cpu_online(smp_processor_id(), false);
+	int rc = generic_cpu_disable();
+	if (rc)
+		return rc;
 
-	/* XXX reset cpu affinity here */
 	mpic_cpu_set_priority(0xf);
-	asm volatile("mtdec %0" : : "r" (0x7fffffff));
-	mb();
-	udelay(20);
-	asm volatile("mtdec %0" : : "r" (0x7fffffff));
+
 	return 0;
 }
 
-static int cpu_dead[NR_CPUS];
+#ifdef CONFIG_PPC32
 
-void pmac32_cpu_die(void)
+static void pmac_cpu_die(void)
 {
+	int cpu = smp_processor_id();
+
 	local_irq_disable();
-	cpu_dead[smp_processor_id()] = 1;
+	idle_task_exit();
+	pr_debug("CPU%d offline\n", cpu);
+	generic_set_cpu_dead(cpu);
+	smp_wmb();
 	mb();
 	low_cpu_die();
 }
 
-void smp_core99_cpu_die(unsigned int cpu)
+#else /* CONFIG_PPC32 */
+
+static void pmac_cpu_die(void)
 {
-	int timeout;
+	int cpu = smp_processor_id();
 
-	timeout = 1000;
-	while (!cpu_dead[cpu]) {
-		if (--timeout == 0) {
-			printk("CPU %u refused to die!\n", cpu);
-			break;
-		}
-		msleep(1);
+	local_irq_disable();
+	idle_task_exit();
+
+	/*
+	 * turn off as much as possible, we'll be
+	 * kicked out as this will only be invoked
+	 * on core99 platforms for now ...
+	 */
+
+	printk(KERN_INFO "CPU#%d offline\n", cpu);
+	generic_set_cpu_dead(cpu);
+	smp_wmb();
+
+	/*
+	 * Re-enable interrupts. The NAP code needs to enable them
+	 * anyways, do it now so we deal with the case where one already
+	 * happened while soft-disabled.
+	 * We shouldn't get any external interrupts, only decrementer, and the
+	 * decrementer handler is safe for use on offline CPUs
+	 */
+	local_irq_enable();
+
+	while (1) {
+		/* let's not take timer interrupts too often ... */
+		set_dec(0x7fffffff);
+
+		/* Enter NAP mode */
+		power4_idle();
 	}
-	cpu_dead[cpu] = 0;
 }
 
-#endif /* CONFIG_HOTPLUG_CPU && CONFIG_PP32 */
+#endif /* else CONFIG_PPC32 */
+#endif /* CONFIG_HOTPLUG_CPU */
 
 /* Core99 Macs (dual G4s and G5s) */
 struct smp_ops_t core99_smp_ops = {
 	.message_pass	= smp_mpic_message_pass,
 	.probe		= smp_core99_probe,
+#ifdef CONFIG_PPC64
+	.bringup_done	= smp_core99_bringup_done,
+#endif
 	.kick_cpu	= smp_core99_kick_cpu,
 	.setup_cpu	= smp_core99_setup_cpu,
 	.give_timebase	= smp_core99_give_timebase,
 	.take_timebase	= smp_core99_take_timebase,
 #if defined(CONFIG_HOTPLUG_CPU)
-# if defined(CONFIG_PPC32)
 	.cpu_disable	= smp_core99_cpu_disable,
-	.cpu_die	= smp_core99_cpu_die,
-# endif
-# if defined(CONFIG_PPC64)
-	.cpu_disable	= generic_cpu_disable,
 	.cpu_die	= generic_cpu_die,
-	/* intentionally do *NOT* assign cpu_enable,
-	 * the generic code will use kick_cpu then! */
-# endif
 #endif
 };
 
@@ -943,7 +1014,7 @@ void __init pmac_setup_smp(void)
 		of_node_put(np);
 		smp_ops = &core99_smp_ops;
 	}
-#ifdef CONFIG_PPC32
+#ifdef CONFIG_PPC_PMAC32_PSURGE
 	else {
 		/* We have to set bits in cpu_possible_mask here since the
 		 * secondary CPU(s) aren't in the device tree. Various
@@ -956,6 +1027,11 @@ void __init pmac_setup_smp(void)
 			set_cpu_possible(cpu, true);
 		smp_ops = &psurge_smp_ops;
 	}
-#endif /* CONFIG_PPC32 */
+#endif /* CONFIG_PPC_PMAC32_PSURGE */
+
+#ifdef CONFIG_HOTPLUG_CPU
+	ppc_md.cpu_die = pmac_cpu_die;
+#endif
 }
 
+