Merge commit 'md/for-linus' into async-tx-next

Conflicts:
	drivers/md/raid5.c

18 files changed, 195 insertions, 268 deletions

diff --git a/arch/powerpc/platforms/44x/warp.c b/arch/powerpc/platforms/44x/warp.c
index 42e09a9f77e..e5c1b096c3e 100644
--- a/arch/powerpc/platforms/44x/warp.c
+++ b/arch/powerpc/platforms/44x/warp.c
@@ -16,6 +16,7 @@
 #include <linux/interrupt.h>
 #include <linux/delay.h>
 #include <linux/of_gpio.h>
+#include <linux/of_i2c.h>
 
 #include <asm/machdep.h>
 #include <asm/prom.h>
@@ -63,9 +64,6 @@ define_machine(warp) {
 };
 
 
-static u32 post_info;
-
-/* I am not sure this is the best place for this... */
 static int __init warp_post_info(void)
 {
 	struct device_node *np;
@@ -87,10 +85,9 @@ static int __init warp_post_info(void)
 
 	iounmap(fpga);
 
-	if (post1 || post2) {
+	if (post1 || post2)
 		printk(KERN_INFO "Warp POST %08x %08x\n", post1, post2);
-		post_info = 1;
-	} else
+	else
 		printk(KERN_INFO "Warp POST OK\n");
 
 	return 0;
@@ -166,6 +163,9 @@ static irqreturn_t temp_isr(int irq, void *context)
 		value ^= 1;
 		mdelay(500);
 	}
+
+	/* Not reached */
+	return IRQ_HANDLED;
 }
 
 static int pika_setup_leds(void)
@@ -179,24 +179,19 @@ static int pika_setup_leds(void)
 	}
 
 	for_each_child_of_node(np, child)
-		if (strcmp(child->name, "green") == 0) {
+		if (strcmp(child->name, "green") == 0)
 			green_led = of_get_gpio(child, 0);
-			/* Turn back on the green LED */
-			gpio_set_value(green_led, 1);
-		} else if (strcmp(child->name, "red") == 0) {
+		else if (strcmp(child->name, "red") == 0)
 			red_led = of_get_gpio(child, 0);
-			/* Set based on post */
-			gpio_set_value(red_led, post_info);
-		}
 
 	of_node_put(np);
 
 	return 0;
 }
 
-static void pika_setup_critical_temp(struct i2c_client *client)
+static void pika_setup_critical_temp(struct device_node *np,
+				     struct i2c_client *client)
 {
-	struct device_node *np;
 	int irq, rc;
 
 	/* Do this before enabling critical temp interrupt since we
@@ -208,14 +203,7 @@ static void pika_setup_critical_temp(struct i2c_client *client)
 	i2c_smbus_write_byte_data(client, 2, 65); /* Thigh */
 	i2c_smbus_write_byte_data(client, 3,  0); /* Tlow */
 
-	np = of_find_compatible_node(NULL, NULL, "adi,ad7414");
-	if (np == NULL) {
-		printk(KERN_ERR __FILE__ ": Unable to find ad7414\n");
-		return;
-	}
-
 	irq = irq_of_parse_and_map(np, 0);
-	of_node_put(np);
 	if (irq  == NO_IRQ) {
 		printk(KERN_ERR __FILE__ ": Unable to get ad7414 irq\n");
 		return;
@@ -244,32 +232,24 @@ static inline void pika_dtm_check_fan(void __iomem *fpga)
 
 static int pika_dtm_thread(void __iomem *fpga)
 {
-	struct i2c_adapter *adap;
+	struct device_node *np;
 	struct i2c_client *client;
 
-	/* We loop in case either driver was compiled as a module and
-	 * has not been insmoded yet.
-	 */
-	while (!(adap = i2c_get_adapter(0))) {
-		set_current_state(TASK_INTERRUPTIBLE);
-		schedule_timeout(HZ);
-	}
-
-	while (1) {
-		list_for_each_entry(client, &adap->clients, list)
-			if (client->addr == 0x4a)
-				goto found_it;
+	np = of_find_compatible_node(NULL, NULL, "adi,ad7414");
+	if (np == NULL)
+		return -ENOENT;
 
-		set_current_state(TASK_INTERRUPTIBLE);
-		schedule_timeout(HZ);
+	client = of_find_i2c_device_by_node(np);
+	if (client == NULL) {
+		of_node_put(np);
+		return -ENOENT;
 	}
 
-found_it:
-	pika_setup_critical_temp(client);
+	pika_setup_critical_temp(np, client);
 
-	i2c_put_adapter(adap);
+	of_node_put(np);
 
-	printk(KERN_INFO "PIKA DTM thread running.\n");
+	printk(KERN_INFO "Warp DTM thread running.\n");
 
 	while (!kthread_should_stop()) {
 		int val;
@@ -291,7 +271,6 @@ found_it:
 	return 0;
 }
 
-
 static int __init pika_dtm_start(void)
 {
 	struct task_struct *dtm_thread;
diff --git a/arch/powerpc/platforms/82xx/pq2ads-pci-pic.c b/arch/powerpc/platforms/82xx/pq2ads-pci-pic.c
index ddf0bdc0fc8..7ee979f323d 100644
--- a/arch/powerpc/platforms/82xx/pq2ads-pci-pic.c
+++ b/arch/powerpc/platforms/82xx/pq2ads-pci-pic.c
@@ -147,7 +147,7 @@ int __init pq2ads_pci_init_irq(void)
 		goto out;
 	}
 
-	priv = alloc_bootmem(sizeof(struct pq2ads_pci_pic));
+	priv = kzalloc(sizeof(*priv), GFP_KERNEL);
 	if (!priv) {
 		of_node_put(np);
 		ret = -ENOMEM;
diff --git a/arch/powerpc/platforms/85xx/mpc85xx_mds.c b/arch/powerpc/platforms/85xx/mpc85xx_mds.c
index 77f90b35635..bfb32834ab0 100644
--- a/arch/powerpc/platforms/85xx/mpc85xx_mds.c
+++ b/arch/powerpc/platforms/85xx/mpc85xx_mds.c
@@ -233,6 +233,19 @@ static void __init mpc85xx_mds_setup_arch(void)
 			/* Turn UCC1 & UCC2 on */
 			setbits8(&bcsr_regs[8], BCSR_UCC1_GETH_EN);
 			setbits8(&bcsr_regs[9], BCSR_UCC2_GETH_EN);
+		} else if (machine_is(mpc8569_mds)) {
+#define BCSR7_UCC12_GETHnRST	(0x1 << 2)
+#define BCSR8_UEM_MARVELL_RST	(0x1 << 1)
+			/*
+			 * U-Boot mangles interrupt polarity for Marvell PHYs,
+			 * so reset built-in and UEM Marvell PHYs, this puts
+			 * the PHYs into their normal state.
+			 */
+			clrbits8(&bcsr_regs[7], BCSR7_UCC12_GETHnRST);
+			setbits8(&bcsr_regs[8], BCSR8_UEM_MARVELL_RST);
+
+			setbits8(&bcsr_regs[7], BCSR7_UCC12_GETHnRST);
+			clrbits8(&bcsr_regs[8], BCSR8_UEM_MARVELL_RST);
 		}
 		iounmap(bcsr_regs);
 	}
@@ -285,6 +298,7 @@ static struct of_device_id mpc85xx_ids[] = {
 	{ .type = "qe", },
 	{ .compatible = "fsl,qe", },
 	{ .compatible = "gianfar", },
+	{ .compatible = "fsl,rapidio-delta", },
 	{},
 };
 
diff --git a/arch/powerpc/platforms/85xx/smp.c b/arch/powerpc/platforms/85xx/smp.c
index cc0b0db8a6f..62c592ede64 100644
--- a/arch/powerpc/platforms/85xx/smp.c
+++ b/arch/powerpc/platforms/85xx/smp.c
@@ -52,20 +52,19 @@ smp_85xx_kick_cpu(int nr)
 
 	pr_debug("smp_85xx_kick_cpu: kick CPU #%d\n", nr);
 
-	local_irq_save(flags);
-
 	np = of_get_cpu_node(nr, NULL);
 	cpu_rel_addr = of_get_property(np, "cpu-release-addr", NULL);
 
 	if (cpu_rel_addr == NULL) {
 		printk(KERN_ERR "No cpu-release-addr for cpu %d\n", nr);
-		local_irq_restore(flags);
 		return;
 	}
 
 	/* Map the spin table */
 	bptr_vaddr = ioremap(*cpu_rel_addr, SIZE_BOOT_ENTRY);
 
+	local_irq_save(flags);
+
 	out_be32(bptr_vaddr + BOOT_ENTRY_PIR, nr);
 	out_be32(bptr_vaddr + BOOT_ENTRY_ADDR_LOWER, __pa(__early_start));
 
@@ -73,10 +72,10 @@ smp_85xx_kick_cpu(int nr)
 	while ((__secondary_hold_acknowledge != nr) && (++n < 1000))
 		mdelay(1);
 
-	iounmap(bptr_vaddr);
-
 	local_irq_restore(flags);
 
+	iounmap(bptr_vaddr);
+
 	pr_debug("waited %d msecs for CPU #%d.\n", n, nr);
 }
 
diff --git a/arch/powerpc/platforms/85xx/socrates.c b/arch/powerpc/platforms/85xx/socrates.c
index d0e8443b12c..747d8fb3ab8 100644
--- a/arch/powerpc/platforms/85xx/socrates.c
+++ b/arch/powerpc/platforms/85xx/socrates.c
@@ -102,10 +102,11 @@ static struct of_device_id __initdata socrates_of_bus_ids[] = {
 	{},
 };
 
-static void __init socrates_init(void)
+static int __init socrates_publish_devices(void)
 {
-	of_platform_bus_probe(NULL, socrates_of_bus_ids, NULL);
+	return of_platform_bus_probe(NULL, socrates_of_bus_ids, NULL);
 }
+machine_device_initcall(socrates, socrates_publish_devices);
 
 /*
  * Called very early, device-tree isn't unflattened
@@ -124,7 +125,6 @@ define_machine(socrates) {
 	.name			= "Socrates",
 	.probe			= socrates_probe,
 	.setup_arch		= socrates_setup_arch,
-	.init			= socrates_init,
 	.init_IRQ		= socrates_pic_init,
 	.get_irq		= mpic_get_irq,
 	.restart		= fsl_rstcr_restart,
diff --git a/arch/powerpc/platforms/85xx/xes_mpc85xx.c b/arch/powerpc/platforms/85xx/xes_mpc85xx.c
index ee01532786e..1b426050a2f 100644
--- a/arch/powerpc/platforms/85xx/xes_mpc85xx.c
+++ b/arch/powerpc/platforms/85xx/xes_mpc85xx.c
@@ -32,7 +32,6 @@
 
 #include <sysdev/fsl_soc.h>
 #include <sysdev/fsl_pci.h>
-#include <linux/of_platform.h>
 
 /* A few bit definitions needed for fixups on some boards */
 #define MPC85xx_L2CTL_L2E		0x80000000 /* L2 enable */
diff --git a/arch/powerpc/platforms/cell/axon_msi.c b/arch/powerpc/platforms/cell/axon_msi.c
index c71498dbf21..aca5741ddc6 100644
--- a/arch/powerpc/platforms/cell/axon_msi.c
+++ b/arch/powerpc/platforms/cell/axon_msi.c
@@ -85,7 +85,7 @@ static inline void axon_msi_debug_setup(struct device_node *dn,
 
 static void msic_dcr_write(struct axon_msic *msic, unsigned int dcr_n, u32 val)
 {
-	pr_debug("axon_msi: dcr_write(0x%x, 0x%x)\n", val, dcr_n);
+	pr_devel("axon_msi: dcr_write(0x%x, 0x%x)\n", val, dcr_n);
 
 	dcr_write(msic->dcr_host, dcr_n, val);
 }
@@ -98,7 +98,7 @@ static void axon_msi_cascade(unsigned int irq, struct irq_desc *desc)
 	int retry = 0;
 
 	write_offset = dcr_read(msic->dcr_host, MSIC_WRITE_OFFSET_REG);
-	pr_debug("axon_msi: original write_offset 0x%x\n", write_offset);
+	pr_devel("axon_msi: original write_offset 0x%x\n", write_offset);
 
 	/* write_offset doesn't wrap properly, so we have to mask it */
 	write_offset &= MSIC_FIFO_SIZE_MASK;
@@ -108,7 +108,7 @@ static void axon_msi_cascade(unsigned int irq, struct irq_desc *desc)
 		msi  = le32_to_cpu(msic->fifo_virt[idx]);
 		msi &= 0xFFFF;
 
-		pr_debug("axon_msi: woff %x roff %x msi %x\n",
+		pr_devel("axon_msi: woff %x roff %x msi %x\n",
 			  write_offset, msic->read_offset, msi);
 
 		if (msi < NR_IRQS && irq_map[msi].host == msic->irq_host) {
@@ -123,12 +123,12 @@ static void axon_msi_cascade(unsigned int irq, struct irq_desc *desc)
 			 */
 			udelay(1);
 			retry++;
-			pr_debug("axon_msi: invalid irq 0x%x!\n", msi);
+			pr_devel("axon_msi: invalid irq 0x%x!\n", msi);
 			continue;
 		}
 
 		if (retry) {
-			pr_debug("axon_msi: late irq 0x%x, retry %d\n",
+			pr_devel("axon_msi: late irq 0x%x, retry %d\n",
 				 msi, retry);
 			retry = 0;
 		}
@@ -332,7 +332,7 @@ static int axon_msi_shutdown(struct of_device *device)
 	struct axon_msic *msic = dev_get_drvdata(&device->dev);
 	u32 tmp;
 
-	pr_debug("axon_msi: disabling %s\n",
+	pr_devel("axon_msi: disabling %s\n",
 		  msic->irq_host->of_node->full_name);
 	tmp  = dcr_read(msic->dcr_host, MSIC_CTRL_REG);
 	tmp &= ~MSIC_CTRL_ENABLE & ~MSIC_CTRL_IRQ_ENABLE;
@@ -349,7 +349,7 @@ static int axon_msi_probe(struct of_device *device,
 	unsigned int virq;
 	int dcr_base, dcr_len;
 
-	pr_debug("axon_msi: setting up dn %s\n", dn->full_name);
+	pr_devel("axon_msi: setting up dn %s\n", dn->full_name);
 
 	msic = kzalloc(sizeof(struct axon_msic), GFP_KERNEL);
 	if (!msic) {
@@ -403,7 +403,7 @@ static int axon_msi_probe(struct of_device *device,
 
 	set_irq_data(virq, msic);
 	set_irq_chained_handler(virq, axon_msi_cascade);
-	pr_debug("axon_msi: irq 0x%x setup for axon_msi\n", virq);
+	pr_devel("axon_msi: irq 0x%x setup for axon_msi\n", virq);
 
 	/* Enable the MSIC hardware */
 	msic_dcr_write(msic, MSIC_BASE_ADDR_HI_REG, msic->fifo_phys >> 32);
@@ -484,13 +484,13 @@ void axon_msi_debug_setup(struct device_node *dn, struct axon_msic *msic)
 
 	addr = of_translate_address(dn, of_get_property(dn, "reg", NULL));
 	if (addr == OF_BAD_ADDR) {
-		pr_debug("axon_msi: couldn't translate reg property\n");
+		pr_devel("axon_msi: couldn't translate reg property\n");
 		return;
 	}
 
 	msic->trigger = ioremap(addr, 0x4);
 	if (!msic->trigger) {
-		pr_debug("axon_msi: ioremap failed\n");
+		pr_devel("axon_msi: ioremap failed\n");
 		return;
 	}
 
@@ -498,7 +498,7 @@ void axon_msi_debug_setup(struct device_node *dn, struct axon_msic *msic)
 
 	if (!debugfs_create_file(name, 0600, powerpc_debugfs_root,
 				 msic, &fops_msic)) {
-		pr_debug("axon_msi: debugfs_create_file failed!\n");
+		pr_devel("axon_msi: debugfs_create_file failed!\n");
 		return;
 	}
 }
diff --git a/arch/powerpc/platforms/cell/smp.c b/arch/powerpc/platforms/cell/smp.c
index 9046803c827..bc97fada48c 100644
--- a/arch/powerpc/platforms/cell/smp.c
+++ b/arch/powerpc/platforms/cell/smp.c
@@ -36,7 +36,6 @@
 #include <asm/prom.h>
 #include <asm/smp.h>
 #include <asm/paca.h>
-#include <asm/time.h>
 #include <asm/machdep.h>
 #include <asm/cputable.h>
 #include <asm/firmware.h>
@@ -140,31 +139,6 @@ static void __devinit smp_cell_setup_cpu(int cpu)
 	mtspr(SPRN_DABRX, DABRX_KERNEL | DABRX_USER);
 }
 
-static DEFINE_SPINLOCK(timebase_lock);
-static unsigned long timebase = 0;
-
-static void __devinit cell_give_timebase(void)
-{
-	spin_lock(&timebase_lock);
-	rtas_call(rtas_token("freeze-time-base"), 0, 1, NULL);
-	timebase = get_tb();
-	spin_unlock(&timebase_lock);
-
-	while (timebase)
-		barrier();
-	rtas_call(rtas_token("thaw-time-base"), 0, 1, NULL);
-}
-
-static void __devinit cell_take_timebase(void)
-{
-	while (!timebase)
-		barrier();
-	spin_lock(&timebase_lock);
-	set_tb(timebase >> 32, timebase & 0xffffffff);
-	timebase = 0;
-	spin_unlock(&timebase_lock);
-}
-
 static void __devinit smp_cell_kick_cpu(int nr)
 {
 	BUG_ON(nr < 0 || nr >= NR_CPUS);
@@ -224,8 +198,8 @@ void __init smp_init_cell(void)
 
 	/* Non-lpar has additional take/give timebase */
 	if (rtas_token("freeze-time-base") != RTAS_UNKNOWN_SERVICE) {
-		smp_ops->give_timebase = cell_give_timebase;
-		smp_ops->take_timebase = cell_take_timebase;
+		smp_ops->give_timebase = rtas_give_timebase;
+		smp_ops->take_timebase = rtas_take_timebase;
 	}
 
 	DBG(" <- smp_init_cell()\n");
diff --git a/arch/powerpc/platforms/chrp/smp.c b/arch/powerpc/platforms/chrp/smp.c
index 10a4a4d063b..02cafecc90e 100644
--- a/arch/powerpc/platforms/chrp/smp.c
+++ b/arch/powerpc/platforms/chrp/smp.c
@@ -26,7 +26,6 @@
 #include <asm/io.h>
 #include <asm/prom.h>
 #include <asm/smp.h>
-#include <asm/time.h>
 #include <asm/machdep.h>
 #include <asm/mpic.h>
 #include <asm/rtas.h>
@@ -42,40 +41,12 @@ static void __devinit smp_chrp_setup_cpu(int cpu_nr)
 	mpic_setup_this_cpu();
 }
 
-static DEFINE_SPINLOCK(timebase_lock);
-static unsigned int timebase_upper = 0, timebase_lower = 0;
-
-void __devinit smp_chrp_give_timebase(void)
-{
-	spin_lock(&timebase_lock);
-	rtas_call(rtas_token("freeze-time-base"), 0, 1, NULL);
-	timebase_upper = get_tbu();
-	timebase_lower = get_tbl();
-	spin_unlock(&timebase_lock);
-
-	while (timebase_upper || timebase_lower)
-		barrier();
-	rtas_call(rtas_token("thaw-time-base"), 0, 1, NULL);
-}
-
-void __devinit smp_chrp_take_timebase(void)
-{
-	while (!(timebase_upper || timebase_lower))
-		barrier();
-	spin_lock(&timebase_lock);
-	set_tb(timebase_upper, timebase_lower);
-	timebase_upper = 0;
-	timebase_lower = 0;
-	spin_unlock(&timebase_lock);
-	printk("CPU %i taken timebase\n", smp_processor_id());
-}
-
 /* CHRP with openpic */
 struct smp_ops_t chrp_smp_ops = {
 	.message_pass = smp_mpic_message_pass,
 	.probe = smp_mpic_probe,
 	.kick_cpu = smp_chrp_kick_cpu,
 	.setup_cpu = smp_chrp_setup_cpu,
-	.give_timebase = smp_chrp_give_timebase,
-	.take_timebase = smp_chrp_take_timebase,
+	.give_timebase = rtas_give_timebase,
+	.take_timebase = rtas_take_timebase,
 };
diff --git a/arch/powerpc/platforms/pasemi/setup.c b/arch/powerpc/platforms/pasemi/setup.c
index 153051eb6d9..a4619347aa7 100644
--- a/arch/powerpc/platforms/pasemi/setup.c
+++ b/arch/powerpc/platforms/pasemi/setup.c
@@ -71,20 +71,25 @@ static void pas_restart(char *cmd)
 }
 
 #ifdef CONFIG_SMP
-static DEFINE_SPINLOCK(timebase_lock);
+static raw_spinlock_t timebase_lock;
 static unsigned long timebase;
 
 static void __devinit pas_give_timebase(void)
 {
-	spin_lock(&timebase_lock);
+	unsigned long flags;
+
+	local_irq_save(flags);
+	hard_irq_disable();
+	__raw_spin_lock(&timebase_lock);
 	mtspr(SPRN_TBCTL, TBCTL_FREEZE);
 	isync();
 	timebase = get_tb();
-	spin_unlock(&timebase_lock);
+	__raw_spin_unlock(&timebase_lock);
 
 	while (timebase)
 		barrier();
 	mtspr(SPRN_TBCTL, TBCTL_RESTART);
+	local_irq_restore(flags);
 }
 
 static void __devinit pas_take_timebase(void)
@@ -92,10 +97,10 @@ static void __devinit pas_take_timebase(void)
 	while (!timebase)
 		smp_rmb();
 
-	spin_lock(&timebase_lock);
+	__raw_spin_lock(&timebase_lock);
 	set_tb(timebase >> 32, timebase & 0xffffffff);
 	timebase = 0;
-	spin_unlock(&timebase_lock);
+	__raw_spin_unlock(&timebase_lock);
 }
 
 struct smp_ops_t pas_smp_ops = {
diff --git a/arch/powerpc/platforms/powermac/cpufreq_64.c b/arch/powerpc/platforms/powermac/cpufreq_64.c
index 22ecfbe7183..708c7513337 100644
--- a/arch/powerpc/platforms/powermac/cpufreq_64.c
+++ b/arch/powerpc/platforms/powermac/cpufreq_64.c
@@ -251,7 +251,7 @@ static void g5_pfunc_switch_volt(int speed_mode)
 static struct pmf_function *pfunc_cpu_setfreq_high;
 static struct pmf_function *pfunc_cpu_setfreq_low;
 static struct pmf_function *pfunc_cpu_getfreq;
-static struct pmf_function *pfunc_slewing_done;;
+static struct pmf_function *pfunc_slewing_done;
 
 static int g5_pfunc_switch_freq(int speed_mode)
 {
diff --git a/arch/powerpc/platforms/powermac/pic.c b/arch/powerpc/platforms/powermac/pic.c
index dce73634910..d212006a5b3 100644
--- a/arch/powerpc/platforms/powermac/pic.c
+++ b/arch/powerpc/platforms/powermac/pic.c
@@ -609,7 +609,7 @@ static int pmacpic_find_viaint(void)
 	np = of_find_node_by_name(NULL, "via-pmu");
 	if (np == NULL)
 		goto not_found;
-	viaint = irq_of_parse_and_map(np, 0);;
+	viaint = irq_of_parse_and_map(np, 0);
 
 not_found:
 #endif /* CONFIG_ADB_PMU */
diff --git a/arch/powerpc/platforms/powermac/setup.c b/arch/powerpc/platforms/powermac/setup.c
index 86f69a4eb49..c2052265636 100644
--- a/arch/powerpc/platforms/powermac/setup.c
+++ b/arch/powerpc/platforms/powermac/setup.c
@@ -103,11 +103,6 @@ unsigned long smu_cmdbuf_abs;
 EXPORT_SYMBOL(smu_cmdbuf_abs);
 #endif
 
-#ifdef CONFIG_SMP
-extern struct smp_ops_t psurge_smp_ops;
-extern struct smp_ops_t core99_smp_ops;
-#endif /* CONFIG_SMP */
-
 static void pmac_show_cpuinfo(struct seq_file *m)
 {
 	struct device_node *np;
@@ -341,34 +336,6 @@ static void __init pmac_setup_arch(void)
 		ROOT_DEV = DEFAULT_ROOT_DEVICE;
 #endif
 
-#ifdef CONFIG_SMP
-	/* Check for Core99 */
-	ic = of_find_node_by_name(NULL, "uni-n");
-	if (!ic)
-		ic = of_find_node_by_name(NULL, "u3");
-	if (!ic)
-		ic = of_find_node_by_name(NULL, "u4");
-	if (ic) {
-		of_node_put(ic);
-		smp_ops = &core99_smp_ops;
-	}
-#ifdef CONFIG_PPC32
-	else {
-		/*
-		 * We have to set bits in cpu_possible_map here since the
-		 * secondary CPU(s) aren't in the device tree, and
-		 * setup_per_cpu_areas only allocates per-cpu data for
-		 * CPUs in the cpu_possible_map.
-		 */
-		int cpu;
-
-		for (cpu = 1; cpu < 4 && cpu < NR_CPUS; ++cpu)
-			cpu_set(cpu, cpu_possible_map);
-		smp_ops = &psurge_smp_ops;
-	}
-#endif
-#endif /* CONFIG_SMP */
-
 #ifdef CONFIG_ADB
 	if (strstr(cmd_line, "adb_sync")) {
 		extern int __adb_probe_sync;
@@ -512,6 +479,14 @@ static void __init pmac_init_early(void)
 #ifdef CONFIG_PPC64
 	iommu_init_early_dart();
 #endif
+
+	/* SMP Init has to be done early as we need to patch up
+	 * cpu_possible_map before interrupt stacks are allocated
+	 * or kaboom...
+	 */
+#ifdef CONFIG_SMP
+	pmac_setup_smp();
+#endif
 }
 
 static int __init pmac_declare_of_platform_devices(void)
diff --git a/arch/powerpc/platforms/powermac/smp.c b/arch/powerpc/platforms/powermac/smp.c
index cf1dbe75889..6d4da7b46b4 100644
--- a/arch/powerpc/platforms/powermac/smp.c
+++ b/arch/powerpc/platforms/powermac/smp.c
@@ -64,10 +64,11 @@
 extern void __secondary_start_pmac_0(void);
 extern int pmac_pfunc_base_install(void);
 
-#ifdef CONFIG_PPC32
+static void (*pmac_tb_freeze)(int freeze);
+static u64 timebase;
+static int tb_req;
 
-/* Sync flag for HW tb sync */
-static volatile int sec_tb_reset = 0;
+#ifdef CONFIG_PPC32
 
 /*
  * Powersurge (old powermac SMP) support.
@@ -294,6 +295,9 @@ static int __init smp_psurge_probe(void)
 		psurge_quad_init();
 		/* All released cards using this HW design have 4 CPUs */
 		ncpus = 4;
+		/* No sure how timebase sync works on those, let's use SW */
+		smp_ops->give_timebase = smp_generic_give_timebase;
+		smp_ops->take_timebase = smp_generic_take_timebase;
 	} else {
 		iounmap(quad_base);
 		if ((in_8(hhead_base + HHEAD_CONFIG) & 0x02) == 0) {
@@ -308,18 +312,15 @@ static int __init smp_psurge_probe(void)
 	psurge_start = ioremap(PSURGE_START, 4);
 	psurge_pri_intr = ioremap(PSURGE_PRI_INTR, 4);
 
-	/*
-	 * This is necessary because OF doesn't know about the
+	/* This is necessary because OF doesn't know about the
 	 * secondary cpu(s), and thus there aren't nodes in the
 	 * device tree for them, and smp_setup_cpu_maps hasn't
-	 * set their bits in cpu_possible_map and cpu_present_map.
+	 * set their bits in cpu_present_map.
 	 */
 	if (ncpus > NR_CPUS)
 		ncpus = NR_CPUS;
-	for (i = 1; i < ncpus ; ++i) {
+	for (i = 1; i < ncpus ; ++i)
 		cpu_set(i, cpu_present_map);
-		set_hard_smp_processor_id(i, i);
-	}
 
 	if (ppc_md.progress) ppc_md.progress("smp_psurge_probe - done", 0x352);
 
@@ -329,8 +330,14 @@ static int __init smp_psurge_probe(void)
 static void __init smp_psurge_kick_cpu(int nr)
 {
 	unsigned long start = __pa(__secondary_start_pmac_0) + nr * 8;
-	unsigned long a;
-	int i;
+	unsigned long a, flags;
+	int i, j;
+
+	/* Defining this here is evil ... but I prefer hiding that
+	 * crap to avoid giving people ideas that they can do the
+	 * same.
+	 */
+	extern volatile unsigned int cpu_callin_map[NR_CPUS];
 
 	/* may need to flush here if secondary bats aren't setup */
 	for (a = KERNELBASE; a < KERNELBASE + 0x800000; a += 32)
@@ -339,47 +346,52 @@ static void __init smp_psurge_kick_cpu(int nr)
 
 	if (ppc_md.progress) ppc_md.progress("smp_psurge_kick_cpu", 0x353);
 
+	/* This is going to freeze the timeebase, we disable interrupts */
+	local_irq_save(flags);
+
 	out_be32(psurge_start, start);
 	mb();
 
 	psurge_set_ipi(nr);
+
 	/*
 	 * We can't use udelay here because the timebase is now frozen.
 	 */
 	for (i = 0; i < 2000; ++i)
-		barrier();
+		asm volatile("nop" : : : "memory");
 	psurge_clr_ipi(nr);
 
-	if (ppc_md.progress) ppc_md.progress("smp_psurge_kick_cpu - done", 0x354);
-}
-
-/*
- * With the dual-cpu powersurge board, the decrementers and timebases
- * of both cpus are frozen after the secondary cpu is started up,
- * until we give the secondary cpu another interrupt.  This routine
- * uses this to get the timebases synchronized.
- *  -- paulus.
- */
-static void __init psurge_dual_sync_tb(int cpu_nr)
-{
-	int t;
-
-	set_dec(tb_ticks_per_jiffy);
-	/* XXX fixme */
-	set_tb(0, 0);
-
-	if (cpu_nr > 0) {
+	/*
+	 * Also, because the timebase is frozen, we must not return to the
+	 * caller which will try to do udelay's etc... Instead, we wait -here-
+	 * for the CPU to callin.
+	 */
+	for (i = 0; i < 100000 && !cpu_callin_map[nr]; ++i) {
+		for (j = 1; j < 10000; j++)
+			asm volatile("nop" : : : "memory");
+		asm volatile("sync" : : : "memory");
+	}
+	if (!cpu_callin_map[nr])
+		goto stuck;
+
+	/* And we do the TB sync here too for standard dual CPU cards */
+	if (psurge_type == PSURGE_DUAL) {
+		while(!tb_req)
+			barrier();
+		tb_req = 0;
+		mb();
+		timebase = get_tb();
+		mb();
+		while (timebase)
+			barrier();
 		mb();
-		sec_tb_reset = 1;
-		return;
 	}
+ stuck:
+	/* now interrupt the secondary, restarting both TBs */
+	if (psurge_type == PSURGE_DUAL)
+		psurge_set_ipi(1);
 
-	/* wait for the secondary to have reset its TB before proceeding */
-	for (t = 10000000; t > 0 && !sec_tb_reset; --t)
-		;
-
-	/* now interrupt the secondary, starting both TBs */
-	psurge_set_ipi(1);
+	if (ppc_md.progress) ppc_md.progress("smp_psurge_kick_cpu - done", 0x354);
 }
 
 static struct irqaction psurge_irqaction = {
@@ -390,36 +402,35 @@ static struct irqaction psurge_irqaction = {
 
 static void __init smp_psurge_setup_cpu(int cpu_nr)
 {
+	if (cpu_nr != 0)
+		return;
 
-	if (cpu_nr == 0) {
-		/* If we failed to start the second CPU, we should still
-		 * send it an IPI to start the timebase & DEC or we might
-		 * have them stuck.
-		 */
-		if (num_online_cpus() < 2) {
-			if (psurge_type == PSURGE_DUAL)
-				psurge_set_ipi(1);
-			return;
-		}
-		/* reset the entry point so if we get another intr we won't
-		 * try to startup again */
-		out_be32(psurge_start, 0x100);
-		if (setup_irq(30, &psurge_irqaction))
-			printk(KERN_ERR "Couldn't get primary IPI interrupt");
-	}
-
-	if (psurge_type == PSURGE_DUAL)
-		psurge_dual_sync_tb(cpu_nr);
+	/* reset the entry point so if we get another intr we won't
+	 * try to startup again */
+	out_be32(psurge_start, 0x100);
+	if (setup_irq(30, &psurge_irqaction))
+		printk(KERN_ERR "Couldn't get primary IPI interrupt");
 }
 
 void __init smp_psurge_take_timebase(void)
 {
-	/* Dummy implementation */
+	if (psurge_type != PSURGE_DUAL)
+		return;
+
+	tb_req = 1;
+	mb();
+	while (!timebase)
+		barrier();
+	mb();
+	set_tb(timebase >> 32, timebase & 0xffffffff);
+	timebase = 0;
+	mb();
+	set_dec(tb_ticks_per_jiffy/2);
 }
 
 void __init smp_psurge_give_timebase(void)
 {
-	/* Dummy implementation */
+	/* Nothing to do here */
 }
 
 /* PowerSurge-style Macs */
@@ -437,9 +448,6 @@ struct smp_ops_t psurge_smp_ops = {
  * Core 99 and later support
  */
 
-static void (*pmac_tb_freeze)(int freeze);
-static u64 timebase;
-static int tb_req;
 
 static void smp_core99_give_timebase(void)
 {
@@ -478,7 +486,6 @@ static void __devinit smp_core99_take_timebase(void)
 	set_tb(timebase >> 32, timebase & 0xffffffff);
 	timebase = 0;
 	mb();
-	set_dec(tb_ticks_per_jiffy/2);
 
 	local_irq_restore(flags);
 }
@@ -920,3 +927,34 @@ struct smp_ops_t core99_smp_ops = {
 # endif
 #endif
 };
+
+void __init pmac_setup_smp(void)
+{
+	struct device_node *np;
+
+	/* Check for Core99 */
+	np = of_find_node_by_name(NULL, "uni-n");
+	if (!np)
+		np = of_find_node_by_name(NULL, "u3");
+	if (!np)
+		np = of_find_node_by_name(NULL, "u4");
+	if (np) {
+		of_node_put(np);
+		smp_ops = &core99_smp_ops;
+	}
+#ifdef CONFIG_PPC32
+	else {
+		/* We have to set bits in cpu_possible_map here since the
+		 * secondary CPU(s) aren't in the device tree. Various
+		 * things won't be initialized for CPUs not in the possible
+		 * map, so we really need to fix it up here.
+		 */
+		int cpu;
+
+		for (cpu = 1; cpu < 4 && cpu < NR_CPUS; ++cpu)
+			cpu_set(cpu, cpu_possible_map);
+		smp_ops = &psurge_smp_ops;
+	}
+#endif /* CONFIG_PPC32 */
+}
+
diff --git a/arch/powerpc/platforms/ps3/system-bus.c b/arch/powerpc/platforms/ps3/system-bus.c
index 9fead0faf38..3f763c5284a 100644
--- a/arch/powerpc/platforms/ps3/system-bus.c
+++ b/arch/powerpc/platforms/ps3/system-bus.c
@@ -284,7 +284,6 @@ static int ps3_sb_free_mmio_region(struct ps3_mmio_region *r)
 	int result;
 
 	dump_mmio_region(r);
-;
 	result = lv1_unmap_device_mmio_region(r->dev->bus_id, r->dev->dev_id,
 		r->lpar_addr);
 
diff --git a/arch/powerpc/platforms/pseries/lpar.c b/arch/powerpc/platforms/pseries/lpar.c
index e3139fa5e55..903eb9eec68 100644
--- a/arch/powerpc/platforms/pseries/lpar.c
+++ b/arch/powerpc/platforms/pseries/lpar.c
@@ -286,7 +286,7 @@ static long pSeries_lpar_hpte_insert(unsigned long hpte_group,
 	unsigned long hpte_v, hpte_r;
 
 	if (!(vflags & HPTE_V_BOLTED))
-		pr_debug("hpte_insert(group=%lx, va=%016lx, pa=%016lx, "
+		pr_devel("hpte_insert(group=%lx, va=%016lx, pa=%016lx, "
 			 "rflags=%lx, vflags=%lx, psize=%d)\n",
 			 hpte_group, va, pa, rflags, vflags, psize);
 
@@ -294,7 +294,7 @@ static long pSeries_lpar_hpte_insert(unsigned long hpte_group,
 	hpte_r = hpte_encode_r(pa, psize) | rflags;
 
 	if (!(vflags & HPTE_V_BOLTED))
-		pr_debug(" hpte_v=%016lx, hpte_r=%016lx\n", hpte_v, hpte_r);
+		pr_devel(" hpte_v=%016lx, hpte_r=%016lx\n", hpte_v, hpte_r);
 
 	/* Now fill in the actual HPTE */
 	/* Set CEC cookie to 0         */
@@ -311,7 +311,7 @@ static long pSeries_lpar_hpte_insert(unsigned long hpte_group,
 	lpar_rc = plpar_pte_enter(flags, hpte_group, hpte_v, hpte_r, &slot);
 	if (unlikely(lpar_rc == H_PTEG_FULL)) {
 		if (!(vflags & HPTE_V_BOLTED))
-			pr_debug(" full\n");
+			pr_devel(" full\n");
 		return -1;
 	}
 
@@ -322,11 +322,11 @@ static long pSeries_lpar_hpte_insert(unsigned long hpte_group,
 	 */
 	if (unlikely(lpar_rc != H_SUCCESS)) {
 		if (!(vflags & HPTE_V_BOLTED))
-			pr_debug(" lpar err %lu\n", lpar_rc);
+			pr_devel(" lpar err %lu\n", lpar_rc);
 		return -2;
 	}
 	if (!(vflags & HPTE_V_BOLTED))
-		pr_debug(" -> slot: %lu\n", slot & 7);
+		pr_devel(" -> slot: %lu\n", slot & 7);
 
 	/* Because of iSeries, we have to pass down the secondary
 	 * bucket bit here as well
@@ -418,17 +418,17 @@ static long pSeries_lpar_hpte_updatepp(unsigned long slot,
 
 	want_v = hpte_encode_avpn(va, psize, ssize);
 
-	pr_debug("    update: avpnv=%016lx, hash=%016lx, f=%lx, psize: %d ...",
+	pr_devel("    update: avpnv=%016lx, hash=%016lx, f=%lx, psize: %d ...",
 		 want_v, slot, flags, psize);
 
 	lpar_rc = plpar_pte_protect(flags, slot, want_v);
 
 	if (lpar_rc == H_NOT_FOUND) {
-		pr_debug("not found !\n");
+		pr_devel("not found !\n");
 		return -1;
 	}
 
-	pr_debug("ok\n");
+	pr_devel("ok\n");
 
 	BUG_ON(lpar_rc != H_SUCCESS);
 
@@ -503,7 +503,7 @@ static void pSeries_lpar_hpte_invalidate(unsigned long slot, unsigned long va,
 	unsigned long lpar_rc;
 	unsigned long dummy1, dummy2;
 
-	pr_debug("    inval : slot=%lx, va=%016lx, psize: %d, local: %d\n",
+	pr_devel("    inval : slot=%lx, va=%016lx, psize: %d, local: %d\n",
 		 slot, va, psize, local);
 
 	want_v = hpte_encode_avpn(va, psize, ssize);
diff --git a/arch/powerpc/platforms/pseries/smp.c b/arch/powerpc/platforms/pseries/smp.c
index 1a231c389ba..1f8f6cfb94f 100644
--- a/arch/powerpc/platforms/pseries/smp.c
+++ b/arch/powerpc/platforms/pseries/smp.c
@@ -35,7 +35,6 @@
 #include <asm/prom.h>
 #include <asm/smp.h>
 #include <asm/paca.h>
-#include <asm/time.h>
 #include <asm/machdep.h>
 #include <asm/cputable.h>
 #include <asm/firmware.h>
@@ -118,31 +117,6 @@ static void __devinit smp_xics_setup_cpu(int cpu)
 }
 #endif /* CONFIG_XICS */
 
-static DEFINE_SPINLOCK(timebase_lock);
-static unsigned long timebase = 0;
-
-static void __devinit pSeries_give_timebase(void)
-{
-	spin_lock(&timebase_lock);
-	rtas_call(rtas_token("freeze-time-base"), 0, 1, NULL);
-	timebase = get_tb();
-	spin_unlock(&timebase_lock);
-
-	while (timebase)
-		barrier();
-	rtas_call(rtas_token("thaw-time-base"), 0, 1, NULL);
-}
-
-static void __devinit pSeries_take_timebase(void)
-{
-	while (!timebase)
-		barrier();
-	spin_lock(&timebase_lock);
-	set_tb(timebase >> 32, timebase & 0xffffffff);
-	timebase = 0;
-	spin_unlock(&timebase_lock);
-}
-
 static void __devinit smp_pSeries_kick_cpu(int nr)
 {
 	BUG_ON(nr < 0 || nr >= NR_CPUS);
@@ -209,8 +183,8 @@ static void __init smp_init_pseries(void)
 
 	/* Non-lpar has additional take/give timebase */
 	if (rtas_token("freeze-time-base") != RTAS_UNKNOWN_SERVICE) {
-		smp_ops->give_timebase = pSeries_give_timebase;
-		smp_ops->take_timebase = pSeries_take_timebase;
+		smp_ops->give_timebase = rtas_give_timebase;
+		smp_ops->take_timebase = rtas_take_timebase;
 	}
 
 	pr_debug(" <- smp_init_pSeries()\n");
diff --git a/arch/powerpc/platforms/pseries/xics.c b/arch/powerpc/platforms/pseries/xics.c
index be3581a8c29..419f8a637ff 100644
--- a/arch/powerpc/platforms/pseries/xics.c
+++ b/arch/powerpc/platforms/pseries/xics.c
@@ -190,10 +190,10 @@ static void xics_unmask_irq(unsigned int virq)
 	int call_status;
 	int server;
 
-	pr_debug("xics: unmask virq %d\n", virq);
+	pr_devel("xics: unmask virq %d\n", virq);
 
 	irq = (unsigned int)irq_map[virq].hwirq;
-	pr_debug(" -> map to hwirq 0x%x\n", irq);
+	pr_devel(" -> map to hwirq 0x%x\n", irq);
 	if (irq == XICS_IPI || irq == XICS_IRQ_SPURIOUS)
 		return;
 
@@ -252,7 +252,7 @@ static void xics_mask_irq(unsigned int virq)
 {
 	unsigned int irq;
 
-	pr_debug("xics: mask virq %d\n", virq);
+	pr_devel("xics: mask virq %d\n", virq);
 
 	irq = (unsigned int)irq_map[virq].hwirq;
 	if (irq == XICS_IPI || irq == XICS_IRQ_SPURIOUS)
@@ -414,7 +414,7 @@ static int xics_host_match(struct irq_host *h, struct device_node *node)
 static int xics_host_map(struct irq_host *h, unsigned int virq,
 			 irq_hw_number_t hw)
 {
-	pr_debug("xics: map virq %d, hwirq 0x%lx\n", virq, hw);
+	pr_devel("xics: map virq %d, hwirq 0x%lx\n", virq, hw);
 
 	/* Insert the interrupt mapping into the radix tree for fast lookup */
 	irq_radix_revmap_insert(xics_host, virq, hw);