Merge branch 'master' into gfs2

59 files changed, 3552 insertions, 1175 deletions

diff --git a/arch/sh/kernel/Makefile b/arch/sh/kernel/Makefile
index f05cd96f8867..5da88a43d350 100644
--- a/arch/sh/kernel/Makefile
+++ b/arch/sh/kernel/Makefile
@@ -6,9 +6,10 @@ extra-y	:= head.o init_task.o vmlinux.lds
 
 obj-y	:= process.o signal.o entry.o traps.o irq.o \
 	ptrace.o setup.o time.o sys_sh.o semaphore.o \
-	io.o io_generic.o sh_ksyms.o
+	io.o io_generic.o sh_ksyms.o syscalls.o
 
 obj-y				+= cpu/ timers/
+obj-$(CONFIG_VSYSCALL)		+= vsyscall/
 
 obj-$(CONFIG_SMP)		+= smp.o
 obj-$(CONFIG_CF_ENABLER)	+= cf-enabler.o
@@ -18,3 +19,5 @@ obj-$(CONFIG_SH_CPU_FREQ)	+= cpufreq.o
 obj-$(CONFIG_MODULES)		+= module.o
 obj-$(CONFIG_EARLY_PRINTK)	+= early_printk.o
 obj-$(CONFIG_KEXEC)		+= machine_kexec.o relocate_kernel.o
+obj-$(CONFIG_APM)		+= apm.o
+obj-$(CONFIG_PM)		+= pm.o
diff --git a/arch/sh/kernel/apm.c b/arch/sh/kernel/apm.c
new file mode 100644
index 000000000000..871e7d640002
--- /dev/null
+++ b/arch/sh/kernel/apm.c
@@ -0,0 +1,539 @@
+/*
+ * bios-less APM driver for hp680
+ *
+ * Copyright 2005 (c) Andriy Skulysh <askulysh@gmail.com>
+ *
+ * based on ARM APM driver by
+ *  Jamey Hicks <jamey@crl.dec.com>
+ *
+ * adapted from the APM BIOS driver for Linux by
+ *  Stephen Rothwell (sfr@linuxcare.com)
+ *
+ * APM 1.2 Reference:
+ *   Intel Corporation, Microsoft Corporation. Advanced Power Management
+ *   (APM) BIOS Interface Specification, Revision 1.2, February 1996.
+ *
+ * [This document is available from Microsoft at:
+ *    http://www.microsoft.com/hwdev/busbios/amp_12.htm]
+ */
+#include <linux/config.h>
+#include <linux/module.h>
+#include <linux/poll.h>
+#include <linux/timer.h>
+#include <linux/slab.h>
+#include <linux/proc_fs.h>
+#include <linux/miscdevice.h>
+#include <linux/apm_bios.h>
+#include <linux/pm.h>
+#include <linux/pm_legacy.h>
+#include <asm/apm.h>
+
+#define MODNAME "apm"
+
+/*
+ * The apm_bios device is one of the misc char devices.
+ * This is its minor number.
+ */
+#define APM_MINOR_DEV			134
+
+/*
+ * Maximum number of events stored
+ */
+#define APM_MAX_EVENTS			16
+
+struct apm_queue {
+	unsigned int		event_head;
+	unsigned int		event_tail;
+	apm_event_t		events[APM_MAX_EVENTS];
+};
+
+/*
+ * The per-file APM data
+ */
+struct apm_user {
+	struct list_head	list;
+
+	unsigned int		suser: 1;
+	unsigned int		writer: 1;
+	unsigned int		reader: 1;
+
+	int			suspend_result;
+	unsigned int		suspend_state;
+#define SUSPEND_NONE	0		/* no suspend pending */
+#define SUSPEND_PENDING	1		/* suspend pending read */
+#define SUSPEND_READ	2		/* suspend read, pending ack */
+#define SUSPEND_ACKED	3		/* suspend acked */
+#define SUSPEND_DONE	4		/* suspend completed */
+
+	struct apm_queue	queue;
+};
+
+/*
+ * Local variables
+ */
+static int suspends_pending;
+
+static DECLARE_WAIT_QUEUE_HEAD(apm_waitqueue);
+static DECLARE_WAIT_QUEUE_HEAD(apm_suspend_waitqueue);
+
+/*
+ * This is a list of everyone who has opened /dev/apm_bios
+ */
+static DECLARE_RWSEM(user_list_lock);
+static LIST_HEAD(apm_user_list);
+
+/*
+ * kapmd info.  kapmd provides us a process context to handle
+ * "APM" events within - specifically necessary if we're going
+ * to be suspending the system.
+ */
+static DECLARE_WAIT_QUEUE_HEAD(kapmd_wait);
+static DECLARE_COMPLETION(kapmd_exit);
+static DEFINE_SPINLOCK(kapmd_queue_lock);
+static struct apm_queue kapmd_queue;
+
+int apm_suspended;
+EXPORT_SYMBOL(apm_suspended);
+
+/* Platform-specific apm_read_proc(). */
+int (*apm_get_info)(char *buf, char **start, off_t fpos, int length);
+EXPORT_SYMBOL(apm_get_info);
+
+/*
+ * APM event queue management.
+ */
+static inline int queue_empty(struct apm_queue *q)
+{
+	return q->event_head == q->event_tail;
+}
+
+static inline apm_event_t queue_get_event(struct apm_queue *q)
+{
+	q->event_tail = (q->event_tail + 1) % APM_MAX_EVENTS;
+	return q->events[q->event_tail];
+}
+
+static void queue_add_event(struct apm_queue *q, apm_event_t event)
+{
+	q->event_head = (q->event_head + 1) % APM_MAX_EVENTS;
+	if (q->event_head == q->event_tail) {
+		static int notified;
+
+		if (notified++ == 0)
+			printk(KERN_ERR "apm: an event queue overflowed\n");
+
+		q->event_tail = (q->event_tail + 1) % APM_MAX_EVENTS;
+	}
+	q->events[q->event_head] = event;
+}
+
+static void queue_event_one_user(struct apm_user *as, apm_event_t event)
+{
+	if (as->suser && as->writer) {
+		switch (event) {
+		case APM_SYS_SUSPEND:
+		case APM_USER_SUSPEND:
+			/*
+			 * If this user already has a suspend pending,
+			 * don't queue another one.
+			 */
+			if (as->suspend_state != SUSPEND_NONE)
+				return;
+
+			as->suspend_state = SUSPEND_PENDING;
+			suspends_pending++;
+			break;
+		}
+	}
+	queue_add_event(&as->queue, event);
+}
+
+static void queue_event(apm_event_t event, struct apm_user *sender)
+{
+	struct apm_user *as;
+
+	down_read(&user_list_lock);
+
+	list_for_each_entry(as, &apm_user_list, list)
+		if (as != sender && as->reader)
+			queue_event_one_user(as, event);
+
+	up_read(&user_list_lock);
+	wake_up_interruptible(&apm_waitqueue);
+}
+
+/**
+ * apm_queue_event - queue an APM event for kapmd
+ * @event: APM event
+ *
+ * Queue an APM event for kapmd to process and ultimately take the
+ * appropriate action.  Only a subset of events are handled:
+ *   %APM_LOW_BATTERY
+ *   %APM_POWER_STATUS_CHANGE
+ *   %APM_USER_SUSPEND
+ *   %APM_SYS_SUSPEND
+ *   %APM_CRITICAL_SUSPEND
+ */
+void apm_queue_event(apm_event_t event)
+{
+	spin_lock_irq(&kapmd_queue_lock);
+	queue_add_event(&kapmd_queue, event);
+	spin_unlock_irq(&kapmd_queue_lock);
+
+	wake_up_interruptible(&kapmd_wait);
+}
+EXPORT_SYMBOL(apm_queue_event);
+
+static void apm_suspend(void)
+{
+	struct apm_user *as;
+	int err;
+
+	apm_suspended = 1;
+	err = pm_suspend(PM_SUSPEND_MEM);
+
+	/*
+	 * Anyone on the APM queues will think we're still suspended.
+	 * Send a message so everyone knows we're now awake again.
+	 */
+	queue_event(APM_NORMAL_RESUME, NULL);
+
+	/*
+	 * Finally, wake up anyone who is sleeping on the suspend.
+	 */
+	down_read(&user_list_lock);
+	list_for_each_entry(as, &apm_user_list, list) {
+		as->suspend_result = err;
+		as->suspend_state = SUSPEND_DONE;
+	}
+	up_read(&user_list_lock);
+
+	wake_up(&apm_suspend_waitqueue);
+	apm_suspended = 0;
+}
+
+static ssize_t apm_read(struct file *fp, char __user *buf,
+			size_t count, loff_t *ppos)
+{
+	struct apm_user *as = fp->private_data;
+	apm_event_t event;
+	int i = count, ret = 0;
+
+	if (count < sizeof(apm_event_t))
+		return -EINVAL;
+
+	if (queue_empty(&as->queue) && fp->f_flags & O_NONBLOCK)
+		return -EAGAIN;
+
+	wait_event_interruptible(apm_waitqueue, !queue_empty(&as->queue));
+
+	while ((i >= sizeof(event)) && !queue_empty(&as->queue)) {
+		event = queue_get_event(&as->queue);
+
+		ret = -EFAULT;
+		if (copy_to_user(buf, &event, sizeof(event)))
+			break;
+
+		if (event == APM_SYS_SUSPEND || event == APM_USER_SUSPEND)
+			as->suspend_state = SUSPEND_READ;
+
+		buf += sizeof(event);
+		i -= sizeof(event);
+	}
+
+	if (i < count)
+		ret = count - i;
+
+	return ret;
+}
+
+static unsigned int apm_poll(struct file *fp, poll_table * wait)
+{
+	struct apm_user *as = fp->private_data;
+
+	poll_wait(fp, &apm_waitqueue, wait);
+	return queue_empty(&as->queue) ? 0 : POLLIN | POLLRDNORM;
+}
+
+/*
+ * apm_ioctl - handle APM ioctl
+ *
+ * APM_IOC_SUSPEND
+ *   This IOCTL is overloaded, and performs two functions.  It is used to:
+ *     - initiate a suspend
+ *     - acknowledge a suspend read from /dev/apm_bios.
+ *   Only when everyone who has opened /dev/apm_bios with write permission
+ *   has acknowledge does the actual suspend happen.
+ */
+static int
+apm_ioctl(struct inode * inode, struct file *filp, u_int cmd, u_long arg)
+{
+	struct apm_user *as = filp->private_data;
+	unsigned long flags;
+	int err = -EINVAL;
+
+	if (!as->suser || !as->writer)
+		return -EPERM;
+
+	switch (cmd) {
+	case APM_IOC_SUSPEND:
+		as->suspend_result = -EINTR;
+
+		if (as->suspend_state == SUSPEND_READ) {
+			/*
+			 * If we read a suspend command from /dev/apm_bios,
+			 * then the corresponding APM_IOC_SUSPEND ioctl is
+			 * interpreted as an acknowledge.
+			 */
+			as->suspend_state = SUSPEND_ACKED;
+			suspends_pending--;
+		} else {
+			/*
+			 * Otherwise it is a request to suspend the system.
+			 * Queue an event for all readers, and expect an
+			 * acknowledge from all writers who haven't already
+			 * acknowledged.
+			 */
+			queue_event(APM_USER_SUSPEND, as);
+		}
+
+		/*
+		 * If there are no further acknowledges required, suspend
+		 * the system.
+		 */
+		if (suspends_pending == 0)
+			apm_suspend();
+
+		/*
+		 * Wait for the suspend/resume to complete.  If there are
+		 * pending acknowledges, we wait here for them.
+		 *
+		 * Note that we need to ensure that the PM subsystem does
+		 * not kick us out of the wait when it suspends the threads.
+		 */
+		flags = current->flags;
+		current->flags |= PF_NOFREEZE;
+
+		/*
+		 * Note: do not allow a thread which is acking the suspend
+		 * to escape until the resume is complete.
+		 */
+		if (as->suspend_state == SUSPEND_ACKED)
+			wait_event(apm_suspend_waitqueue,
+					 as->suspend_state == SUSPEND_DONE);
+		else
+			wait_event_interruptible(apm_suspend_waitqueue,
+					 as->suspend_state == SUSPEND_DONE);
+
+		current->flags = flags;
+		err = as->suspend_result;
+		as->suspend_state = SUSPEND_NONE;
+		break;
+	}
+
+	return err;
+}
+
+static int apm_release(struct inode * inode, struct file * filp)
+{
+	struct apm_user *as = filp->private_data;
+	filp->private_data = NULL;
+
+	down_write(&user_list_lock);
+	list_del(&as->list);
+	up_write(&user_list_lock);
+
+	/*
+	 * We are now unhooked from the chain.  As far as new
+	 * events are concerned, we no longer exist.  However, we
+	 * need to balance suspends_pending, which means the
+	 * possibility of sleeping.
+	 */
+	if (as->suspend_state != SUSPEND_NONE) {
+		suspends_pending -= 1;
+		if (suspends_pending == 0)
+			apm_suspend();
+	}
+
+	kfree(as);
+	return 0;
+}
+
+static int apm_open(struct inode * inode, struct file * filp)
+{
+	struct apm_user *as;
+
+	as = kzalloc(sizeof(*as), GFP_KERNEL);
+	if (as) {
+		/*
+		 * XXX - this is a tiny bit broken, when we consider BSD
+		 * process accounting. If the device is opened by root, we
+		 * instantly flag that we used superuser privs. Who knows,
+		 * we might close the device immediately without doing a
+		 * privileged operation -- cevans
+		 */
+		as->suser = capable(CAP_SYS_ADMIN);
+		as->writer = (filp->f_mode & FMODE_WRITE) == FMODE_WRITE;
+		as->reader = (filp->f_mode & FMODE_READ) == FMODE_READ;
+
+		down_write(&user_list_lock);
+		list_add(&as->list, &apm_user_list);
+		up_write(&user_list_lock);
+
+		filp->private_data = as;
+	}
+
+	return as ? 0 : -ENOMEM;
+}
+
+static struct file_operations apm_bios_fops = {
+	.owner		= THIS_MODULE,
+	.read		= apm_read,
+	.poll		= apm_poll,
+	.ioctl		= apm_ioctl,
+	.open		= apm_open,
+	.release	= apm_release,
+};
+
+static struct miscdevice apm_device = {
+	.minor		= APM_MINOR_DEV,
+	.name		= "apm_bios",
+	.fops		= &apm_bios_fops
+};
+
+
+#ifdef CONFIG_PROC_FS
+/*
+ * Arguments, with symbols from linux/apm_bios.h.
+ *
+ *   0) Linux driver version (this will change if format changes)
+ *   1) APM BIOS Version.  Usually 1.0, 1.1 or 1.2.
+ *   2) APM flags from APM Installation Check (0x00):
+ *	bit 0: APM_16_BIT_SUPPORT
+ *	bit 1: APM_32_BIT_SUPPORT
+ *	bit 2: APM_IDLE_SLOWS_CLOCK
+ *	bit 3: APM_BIOS_DISABLED
+ *	bit 4: APM_BIOS_DISENGAGED
+ *   3) AC line status
+ *	0x00: Off-line
+ *	0x01: On-line
+ *	0x02: On backup power (BIOS >= 1.1 only)
+ *	0xff: Unknown
+ *   4) Battery status
+ *	0x00: High
+ *	0x01: Low
+ *	0x02: Critical
+ *	0x03: Charging
+ *	0x04: Selected battery not present (BIOS >= 1.2 only)
+ *	0xff: Unknown
+ *   5) Battery flag
+ *	bit 0: High
+ *	bit 1: Low
+ *	bit 2: Critical
+ *	bit 3: Charging
+ *	bit 7: No system battery
+ *	0xff: Unknown
+ *   6) Remaining battery life (percentage of charge):
+ *	0-100: valid
+ *	-1: Unknown
+ *   7) Remaining battery life (time units):
+ *	Number of remaining minutes or seconds
+ *	-1: Unknown
+ *   8) min = minutes; sec = seconds
+ */
+static int apm_read_proc(char *buf, char **start, off_t fpos, int length)
+{
+	if (likely(apm_get_info))
+		return apm_get_info(buf, start, fpos, length);
+
+	return -EINVAL;
+}
+#endif
+
+static int kapmd(void *arg)
+{
+	daemonize("kapmd");
+	current->flags |= PF_NOFREEZE;
+
+	do {
+		apm_event_t event;
+
+		wait_event_interruptible(kapmd_wait,
+				!queue_empty(&kapmd_queue) || !pm_active);
+
+		if (!pm_active)
+			break;
+
+		spin_lock_irq(&kapmd_queue_lock);
+		event = 0;
+		if (!queue_empty(&kapmd_queue))
+			event = queue_get_event(&kapmd_queue);
+		spin_unlock_irq(&kapmd_queue_lock);
+
+		switch (event) {
+		case 0:
+			break;
+
+		case APM_LOW_BATTERY:
+		case APM_POWER_STATUS_CHANGE:
+			queue_event(event, NULL);
+			break;
+
+		case APM_USER_SUSPEND:
+		case APM_SYS_SUSPEND:
+			queue_event(event, NULL);
+			if (suspends_pending == 0)
+				apm_suspend();
+			break;
+
+		case APM_CRITICAL_SUSPEND:
+			apm_suspend();
+			break;
+		}
+	} while (1);
+
+	complete_and_exit(&kapmd_exit, 0);
+}
+
+static int __init apm_init(void)
+{
+	int ret;
+
+	pm_active = 1;
+
+	ret = kernel_thread(kapmd, NULL, CLONE_KERNEL);
+	if (unlikely(ret < 0)) {
+		pm_active = 0;
+		return ret;
+	}
+
+	create_proc_info_entry("apm", 0, NULL, apm_read_proc);
+
+	ret = misc_register(&apm_device);
+	if (unlikely(ret != 0)) {
+		remove_proc_entry("apm", NULL);
+
+		pm_active = 0;
+		wake_up(&kapmd_wait);
+		wait_for_completion(&kapmd_exit);
+	}
+
+	return ret;
+}
+
+static void __exit apm_exit(void)
+{
+	misc_deregister(&apm_device);
+	remove_proc_entry("apm", NULL);
+
+	pm_active = 0;
+	wake_up(&kapmd_wait);
+	wait_for_completion(&kapmd_exit);
+}
+
+module_init(apm_init);
+module_exit(apm_exit);
+
+MODULE_AUTHOR("Stephen Rothwell, Andriy Skulysh");
+MODULE_DESCRIPTION("Advanced Power Management");
+MODULE_LICENSE("GPL");
diff --git a/arch/sh/kernel/cf-enabler.c b/arch/sh/kernel/cf-enabler.c
index f1f9ab87f0b0..3e5fa1e24df0 100644
--- a/arch/sh/kernel/cf-enabler.c
+++ b/arch/sh/kernel/cf-enabler.c
@@ -10,7 +10,8 @@
  */
 
 #include <linux/init.h>
-
+#include <linux/mm.h>
+#include <linux/vmalloc.h>
 #include <asm/io.h>
 #include <asm/irq.h>
 
@@ -32,8 +33,6 @@
 /* SH4 can't access PCMCIA interface through P2 area.
  * we must remap it with appropreate attribute bit of the page set.
  * this part is based on Greg Banks' hd64465_ss.c implementation - Masahiro Abe */
-#include <linux/mm.h>
-#include <linux/vmalloc.h>
 
 #if defined(CONFIG_CF_AREA6)
 #define slot_no 0
@@ -41,9 +40,6 @@
 #define slot_no 1
 #endif
 
-/* defined in mm/ioremap.c */
-extern void * p3_ioremap(unsigned long phys_addr, unsigned long size, unsigned long flags);
-
 /* use this pointer to access to directly connected compact flash io area*/
 void *cf_io_base;
 
@@ -62,7 +58,7 @@ static int __init allocate_cf_area(void)
 		return -ENOMEM;
 	}
 /*	printk("p3_ioremap(paddr=0x%08lx, psize=0x%08lx, prot=0x%08lx)=0x%08lx\n",
-	    	paddrbase, psize, prot.pgprot, cf_io_base);*/
+		paddrbase, psize, prot.pgprot, cf_io_base);*/
 
 	/* XXX : do we need attribute and common-memory area also? */
 
@@ -87,7 +83,7 @@ static int __init cf_init_default(void)
 }
 
 #if defined(CONFIG_SH_SOLUTION_ENGINE)
-#include <asm/se/se.h>
+#include <asm/se.h>
 
 /*
  * SolutionEngine
diff --git a/arch/sh/kernel/cpu/Makefile b/arch/sh/kernel/cpu/Makefile
index 59d5b748752f..fb5dac069382 100644
--- a/arch/sh/kernel/cpu/Makefile
+++ b/arch/sh/kernel/cpu/Makefile
@@ -8,6 +8,5 @@ obj-$(CONFIG_CPU_SH2)		+= sh2/
 obj-$(CONFIG_CPU_SH3)		+= sh3/
 obj-$(CONFIG_CPU_SH4)		+= sh4/
 
-obj-$(CONFIG_SH_RTC)		+= rtc.o
 obj-$(CONFIG_UBC_WAKEUP)	+= ubc.o
 obj-$(CONFIG_SH_ADC)		+= adc.o
diff --git a/arch/sh/kernel/cpu/clock.c b/arch/sh/kernel/cpu/clock.c
index 97fa37f42b84..51ec64cdf348 100644
--- a/arch/sh/kernel/cpu/clock.c
+++ b/arch/sh/kernel/cpu/clock.c
@@ -1,7 +1,7 @@
 /*
  * arch/sh/kernel/cpu/clock.c - SuperH clock framework
  *
- *  Copyright (C) 2005  Paul Mundt
+ *  Copyright (C) 2005, 2006  Paul Mundt
  *
  * This clock framework is derived from the OMAP version by:
  *
@@ -15,6 +15,7 @@
 #include <linux/kernel.h>
 #include <linux/init.h>
 #include <linux/module.h>
+#include <linux/mutex.h>
 #include <linux/list.h>
 #include <linux/kref.h>
 #include <linux/seq_file.h>
@@ -24,7 +25,7 @@
 
 static LIST_HEAD(clock_list);
 static DEFINE_SPINLOCK(clock_lock);
-static DECLARE_MUTEX(clock_list_sem);
+static DEFINE_MUTEX(clock_list_sem);
 
 /*
  * Each subtype is expected to define the init routines for these clocks,
@@ -140,21 +141,21 @@ void clk_disable(struct clk *clk)
 
 int clk_register(struct clk *clk)
 {
-	down(&clock_list_sem);
+	mutex_lock(&clock_list_sem);
 
 	list_add(&clk->node, &clock_list);
 	kref_init(&clk->kref);
 
-	up(&clock_list_sem);
+	mutex_unlock(&clock_list_sem);
 
 	return 0;
 }
 
 void clk_unregister(struct clk *clk)
 {
-	down(&clock_list_sem);
+	mutex_lock(&clock_list_sem);
 	list_del(&clk->node);
-	up(&clock_list_sem);
+	mutex_unlock(&clock_list_sem);
 }
 
 inline unsigned long clk_get_rate(struct clk *clk)
@@ -198,14 +199,14 @@ struct clk *clk_get(const char *id)
 {
 	struct clk *p, *clk = ERR_PTR(-ENOENT);
 
-	down(&clock_list_sem);
+	mutex_lock(&clock_list_sem);
 	list_for_each_entry(p, &clock_list, node) {
 		if (strcmp(id, p->name) == 0 && try_module_get(p->owner)) {
 			clk = p;
 			break;
 		}
 	}
-	up(&clock_list_sem);
+	mutex_unlock(&clock_list_sem);
 
 	return clk;
 }
@@ -225,7 +226,7 @@ int __init clk_init(void)
 {
 	int i, ret = 0;
 
-	BUG_ON(unlikely(!master_clk.rate));
+	BUG_ON(!master_clk.rate);
 
 	for (i = 0; i < ARRAY_SIZE(onchip_clocks); i++) {
 		struct clk *clk = onchip_clocks[i];
diff --git a/arch/sh/kernel/cpu/init.c b/arch/sh/kernel/cpu/init.c
index 868e68b28880..bfb90eb0b7a6 100644
--- a/arch/sh/kernel/cpu/init.c
+++ b/arch/sh/kernel/cpu/init.c
@@ -4,6 +4,7 @@
  * CPU init code
  *
  * Copyright (C) 2002, 2003  Paul Mundt
+ * Copyright (C) 2003  Richard Curnow
  *
  * This file is subject to the terms and conditions of the GNU General Public
  * License.  See the file "COPYING" in the main directory of this archive
@@ -13,6 +14,7 @@
 #include <linux/kernel.h>
 #include <asm/processor.h>
 #include <asm/uaccess.h>
+#include <asm/page.h>
 #include <asm/system.h>
 #include <asm/cacheflush.h>
 #include <asm/cache.h>
@@ -51,7 +53,15 @@ static void __init cache_init(void)
 	ccr = ctrl_inl(CCR);
 
 	/*
-	 * If the cache is already enabled .. flush it.
+	 * At this point we don't know whether the cache is enabled or not - a
+	 * bootloader may have enabled it.  There are at least 2 things that
+	 * could be dirty in the cache at this point:
+	 * 1. kernel command line set up by boot loader
+	 * 2. spilled registers from the prolog of this function
+	 * => before re-initialising the cache, we must do a purge of the whole
+	 * cache out to memory for safety.  As long as nothing is spilled
+	 * during the loop to lines that have already been done, this is safe.
+	 * - RPC
 	 */
 	if (ccr & CCR_CACHE_ENABLE) {
 		unsigned long ways, waysize, addrstart;
@@ -98,6 +108,8 @@ static void __init cache_init(void)
 	/* Force EMODE if possible */
 	if (cpu_data->dcache.ways > 1)
 		flags |= CCR_CACHE_EMODE;
+	else
+		flags &= ~CCR_CACHE_EMODE;
 #endif
 
 #ifdef CONFIG_SH_WRITETHROUGH
@@ -112,6 +124,9 @@ static void __init cache_init(void)
 	/* Turn on OCRAM -- halve the OC */
 	flags |= CCR_CACHE_ORA;
 	cpu_data->dcache.sets >>= 1;
+
+	cpu_data->dcache.way_size = cpu_data->dcache.sets *
+				    cpu_data->dcache.linesz;
 #endif
 
 	ctrl_outl(flags, CCR);
@@ -184,6 +199,10 @@ asmlinkage void __init sh_cpu_init(void)
 	/* Init the cache */
 	cache_init();
 
+	shm_align_mask = max_t(unsigned long,
+			       cpu_data->dcache.way_size - 1,
+			       PAGE_SIZE - 1);
+
 	/* Disable the FPU */
 	if (fpu_disabled) {
 		printk("FPU Disabled\n");
diff --git a/arch/sh/kernel/cpu/irq/Makefile b/arch/sh/kernel/cpu/irq/Makefile
index e3cccea15e1d..1c034c283f59 100644
--- a/arch/sh/kernel/cpu/irq/Makefile
+++ b/arch/sh/kernel/cpu/irq/Makefile
@@ -3,5 +3,6 @@
 #
 obj-y	+= ipr.o imask.o
 
-obj-$(CONFIG_CPU_HAS_PINT_IRQ)	+= pint.o
-obj-$(CONFIG_CPU_HAS_INTC2_IRQ)	+= intc2.o
+obj-$(CONFIG_CPU_HAS_PINT_IRQ)		+= pint.o
+obj-$(CONFIG_CPU_HAS_MASKREG_IRQ)	+= maskreg.o
+obj-$(CONFIG_CPU_HAS_INTC2_IRQ)		+= intc2.o
diff --git a/arch/sh/kernel/cpu/irq/intc2.c b/arch/sh/kernel/cpu/irq/intc2.c
index 30064bf6e154..e30e4b7aa70e 100644
--- a/arch/sh/kernel/cpu/irq/intc2.c
+++ b/arch/sh/kernel/cpu/irq/intc2.c
@@ -241,9 +241,9 @@ static struct intc2_init {
 	/* 110-111 reserved/unused */
 #elif defined(CONFIG_CPU_SUBTYPE_SH7780)
 	{ TIMER_IRQ, 0, 24, 0, INTC_TMU0_MSK, 2},
-#ifdef CONFIG_SH_RTC
-	{ RTC_IRQ, 4, 0, 0, INTC_RTC_MSK, TIMER_PRIORITY },
-#endif
+	{ 21, 1, 0, 0, INTC_RTC_MSK, TIMER_PRIORITY },
+	{ 22, 1, 1, 0, INTC_RTC_MSK, TIMER_PRIORITY },
+	{ 23, 1, 2, 0, INTC_RTC_MSK, TIMER_PRIORITY },
 	{ SCIF0_ERI_IRQ, 8, 24, 0, INTC_SCIF0_MSK, SCIF0_PRIORITY },
 	{ SCIF0_RXI_IRQ, 8, 24, 0, INTC_SCIF0_MSK, SCIF0_PRIORITY },
 	{ SCIF0_BRI_IRQ, 8, 24, 0, INTC_SCIF0_MSK, SCIF0_PRIORITY },
diff --git a/arch/sh/kernel/cpu/irq/ipr.c b/arch/sh/kernel/cpu/irq/ipr.c
index 0f545941fb4f..f785822cd5de 100644
--- a/arch/sh/kernel/cpu/irq/ipr.c
+++ b/arch/sh/kernel/cpu/irq/ipr.c
@@ -57,31 +57,27 @@ static struct hw_interrupt_type ipr_irq_type = {
 
 static void disable_ipr_irq(unsigned int irq)
 {
-	unsigned long val, flags;
+	unsigned long val;
 	unsigned int addr = ipr_data[irq].addr;
 	unsigned short mask = 0xffff ^ (0x0f << ipr_data[irq].shift);
 
 	/* Set the priority in IPR to 0 */
-	local_irq_save(flags);
 	val = ctrl_inw(addr);
 	val &= mask;
 	ctrl_outw(val, addr);
-	local_irq_restore(flags);
 }
 
 static void enable_ipr_irq(unsigned int irq)
 {
-	unsigned long val, flags;
+	unsigned long val;
 	unsigned int addr = ipr_data[irq].addr;
 	int priority = ipr_data[irq].priority;
 	unsigned short value = (priority << ipr_data[irq].shift);
 
 	/* Set priority in IPR back to original value */
-	local_irq_save(flags);
 	val = ctrl_inw(addr);
 	val |= value;
 	ctrl_outw(val, addr);
-	local_irq_restore(flags);
 }
 
 static void mask_and_ack_ipr(unsigned int irq)
@@ -89,6 +85,7 @@ static void mask_and_ack_ipr(unsigned int irq)
 	disable_ipr_irq(irq);
 
 #if defined(CONFIG_CPU_SUBTYPE_SH7707) || defined(CONFIG_CPU_SUBTYPE_SH7709) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7706) || \
     defined(CONFIG_CPU_SUBTYPE_SH7300) || defined(CONFIG_CPU_SUBTYPE_SH7705)
 	/* This is needed when we use edge triggered setting */
 	/* XXX: Is it really needed? */
@@ -123,7 +120,7 @@ void __init init_IRQ(void)
 #ifndef CONFIG_CPU_SUBTYPE_SH7780
 	make_ipr_irq(TIMER_IRQ, TIMER_IPR_ADDR, TIMER_IPR_POS, TIMER_PRIORITY);
 	make_ipr_irq(TIMER1_IRQ, TIMER1_IPR_ADDR, TIMER1_IPR_POS, TIMER1_PRIORITY);
-#if defined(CONFIG_SH_RTC)
+#ifdef RTC_IRQ
 	make_ipr_irq(RTC_IRQ, RTC_IPR_ADDR, RTC_IPR_POS, RTC_PRIORITY);
 #endif
 
@@ -162,6 +159,7 @@ void __init init_IRQ(void)
 #endif
 
 #if defined(CONFIG_CPU_SUBTYPE_SH7707) || defined(CONFIG_CPU_SUBTYPE_SH7709) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7706) || \
     defined(CONFIG_CPU_SUBTYPE_SH7300) || defined(CONFIG_CPU_SUBTYPE_SH7705)
 	/*
 	 * Initialize the Interrupt Controller (INTC)
@@ -192,6 +190,8 @@ void __init init_IRQ(void)
 	/* Perform the machine specific initialisation */
 	if (sh_mv.mv_init_irq != NULL)
 		sh_mv.mv_init_irq();
+
+	irq_ctx_init(smp_processor_id());
 }
 
 #if !defined(CONFIG_CPU_HAS_PINT_IRQ)
diff --git a/arch/sh/kernel/cpu/irq/maskreg.c b/arch/sh/kernel/cpu/irq/maskreg.c
new file mode 100644
index 000000000000..492db31b3cab
--- /dev/null
+++ b/arch/sh/kernel/cpu/irq/maskreg.c
@@ -0,0 +1,93 @@
+/*
+ * Interrupt handling for Simple external interrupt mask register
+ *
+ * Copyright (C) 2001 A&D Co., Ltd. <http://www.aandd.co.jp>
+ *
+ * This is for the machine which have single 16 bit register
+ * for masking external IRQ individually.
+ * Each bit of the register is for masking each interrupt.
+ *
+ * This file may be copied or modified under the terms of the GNU
+ * General Public License.  See linux/COPYING for more information.
+ */
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/irq.h>
+#include <asm/system.h>
+#include <asm/io.h>
+
+/* address of external interrupt mask register */
+unsigned long irq_mask_register;
+
+/* forward declaration */
+static unsigned int startup_maskreg_irq(unsigned int irq);
+static void shutdown_maskreg_irq(unsigned int irq);
+static void enable_maskreg_irq(unsigned int irq);
+static void disable_maskreg_irq(unsigned int irq);
+static void mask_and_ack_maskreg(unsigned int);
+static void end_maskreg_irq(unsigned int irq);
+
+/* hw_interrupt_type */
+static struct hw_interrupt_type maskreg_irq_type = {
+	.typename = "Mask Register",
+	.startup = startup_maskreg_irq,
+	.shutdown = shutdown_maskreg_irq,
+	.enable = enable_maskreg_irq,
+	.disable = disable_maskreg_irq,
+	.ack = mask_and_ack_maskreg,
+	.end = end_maskreg_irq
+};
+
+/* actual implementatin */
+static unsigned int startup_maskreg_irq(unsigned int irq)
+{
+	enable_maskreg_irq(irq);
+	return 0; /* never anything pending */
+}
+
+static void shutdown_maskreg_irq(unsigned int irq)
+{
+	disable_maskreg_irq(irq);
+}
+
+static void disable_maskreg_irq(unsigned int irq)
+{
+	unsigned short val, mask = 0x01 << irq;
+
+	BUG_ON(!irq_mask_register);
+
+	/* Set "irq"th bit */
+	val = ctrl_inw(irq_mask_register);
+	val |= mask;
+	ctrl_outw(val, irq_mask_register);
+}
+
+static void enable_maskreg_irq(unsigned int irq)
+{
+	unsigned short val, mask = ~(0x01 << irq);
+
+	BUG_ON(!irq_mask_register);
+
+	/* Clear "irq"th bit */
+	val = ctrl_inw(irq_mask_register);
+	val &= mask;
+	ctrl_outw(val, irq_mask_register);
+}
+
+static void mask_and_ack_maskreg(unsigned int irq)
+{
+	disable_maskreg_irq(irq);
+}
+
+static void end_maskreg_irq(unsigned int irq)
+{
+	if (!(irq_desc[irq].status & (IRQ_DISABLED|IRQ_INPROGRESS)))
+		enable_maskreg_irq(irq);
+}
+
+void make_maskreg_irq(unsigned int irq)
+{
+	disable_irq_nosync(irq);
+	irq_desc[irq].handler = &maskreg_irq_type;
+	disable_maskreg_irq(irq);
+}
diff --git a/arch/sh/kernel/cpu/irq/pint.c b/arch/sh/kernel/cpu/irq/pint.c
index 80cd8108d36a..17f47b373d6e 100644
--- a/arch/sh/kernel/cpu/irq/pint.c
+++ b/arch/sh/kernel/cpu/irq/pint.c
@@ -48,26 +48,22 @@ static struct hw_interrupt_type pint_irq_type = {
 
 static void disable_pint_irq(unsigned int irq)
 {
-	unsigned long val, flags;
+	unsigned long val;
 
-	local_irq_save(flags);
 	val = ctrl_inw(INTC_INTER);
 	val &= ~(1 << (irq - PINT_IRQ_BASE));
 	ctrl_outw(val, INTC_INTER);	/* disable PINTn */
 	portcr_mask &= ~(3 << (irq - PINT_IRQ_BASE)*2);
-	local_irq_restore(flags);
 }
 
 static void enable_pint_irq(unsigned int irq)
 {
-	unsigned long val, flags;
+	unsigned long val;
 
-	local_irq_save(flags);
 	val = ctrl_inw(INTC_INTER);
 	val |= 1 << (irq - PINT_IRQ_BASE);
 	ctrl_outw(val, INTC_INTER);	/* enable PINTn */
 	portcr_mask |= 3 << (irq - PINT_IRQ_BASE)*2;
-	local_irq_restore(flags);
 }
 
 static void mask_and_ack_pint(unsigned int irq)
diff --git a/arch/sh/kernel/cpu/rtc.c b/arch/sh/kernel/cpu/rtc.c
deleted file mode 100644
index 4304cf75cfa2..000000000000
--- a/arch/sh/kernel/cpu/rtc.c
+++ /dev/null
@@ -1,128 +0,0 @@
-/*
- * linux/arch/sh/kernel/rtc.c -- SH3 / SH4 on-chip RTC support
- *
- *  Copyright (C) 2000  Philipp Rumpf <prumpf@tux.org>
- *  Copyright (C) 1999  Tetsuya Okada & Niibe Yutaka
- */
-
-#include <linux/init.h>
-#include <linux/kernel.h>
-#include <linux/sched.h>
-#include <linux/time.h>
-#include <linux/bcd.h>
-#include <asm/io.h>
-#include <asm/rtc.h>
-
-void sh_rtc_gettimeofday(struct timespec *ts)
-{
-	unsigned int sec128, sec, sec2, min, hr, wk, day, mon, yr, yr100, cf_bit;
-	unsigned long flags;
-
- again:
-	do {
-		local_irq_save(flags);
-		ctrl_outb(0, RCR1);  /* Clear CF-bit */
-		sec128 = ctrl_inb(R64CNT);
-		sec = ctrl_inb(RSECCNT);
-		min = ctrl_inb(RMINCNT);
-		hr  = ctrl_inb(RHRCNT);
-		wk  = ctrl_inb(RWKCNT);
-		day = ctrl_inb(RDAYCNT);
-		mon = ctrl_inb(RMONCNT);
-#if defined(CONFIG_CPU_SH4)
-		yr  = ctrl_inw(RYRCNT);
-		yr100 = (yr >> 8);
-		yr &= 0xff;
-#else
-		yr  = ctrl_inb(RYRCNT);
-		yr100 = (yr == 0x99) ? 0x19 : 0x20;
-#endif
-		sec2 = ctrl_inb(R64CNT);
-		cf_bit = ctrl_inb(RCR1) & RCR1_CF;
-		local_irq_restore(flags);
-	} while (cf_bit != 0 || ((sec128 ^ sec2) & RTC_BIT_INVERTED) != 0);
-
-	BCD_TO_BIN(yr100);
-	BCD_TO_BIN(yr);
-	BCD_TO_BIN(mon);
-	BCD_TO_BIN(day);
-	BCD_TO_BIN(hr);
-	BCD_TO_BIN(min);
-	BCD_TO_BIN(sec);
-
-	if (yr > 99 || mon < 1 || mon > 12 || day > 31 || day < 1 ||
-	    hr > 23 || min > 59 || sec > 59) {
-		printk(KERN_ERR
-		       "SH RTC: invalid value, resetting to 1 Jan 2000\n");
-		local_irq_save(flags);
-		ctrl_outb(RCR2_RESET, RCR2);  /* Reset & Stop */
-		ctrl_outb(0, RSECCNT);
-		ctrl_outb(0, RMINCNT);
-		ctrl_outb(0, RHRCNT);
-		ctrl_outb(6, RWKCNT);
-		ctrl_outb(1, RDAYCNT);
-		ctrl_outb(1, RMONCNT);
-#if defined(CONFIG_CPU_SH4)
-		ctrl_outw(0x2000, RYRCNT);
-#else
-		ctrl_outb(0, RYRCNT);
-#endif
-		ctrl_outb(RCR2_RTCEN|RCR2_START, RCR2);  /* Start */
-		goto again;
-	}
-
-#if RTC_BIT_INVERTED != 0
-	if ((sec128 & RTC_BIT_INVERTED))
-		sec--;
-#endif
-
-	ts->tv_sec = mktime(yr100 * 100 + yr, mon, day, hr, min, sec);
-	ts->tv_nsec = ((sec128 * 1000000) / 128) * 1000;
-}
-
-/*
- * Changed to only care about tv_sec, and not the full timespec struct
- * (i.e. tv_nsec).  It can easily be switched to timespec for future cpus
- * that support setting usec or nsec RTC values.
- */
-int sh_rtc_settimeofday(const time_t secs)
-{
-	int retval = 0;
-	int real_seconds, real_minutes, cmos_minutes;
-	unsigned long flags;
-
-	local_irq_save(flags);
-	ctrl_outb(RCR2_RESET, RCR2);  /* Reset pre-scaler & stop RTC */
-
-	cmos_minutes = ctrl_inb(RMINCNT);
-	BCD_TO_BIN(cmos_minutes);
-
-	/*
-	 * since we're only adjusting minutes and seconds,
-	 * don't interfere with hour overflow. This avoids
-	 * messing with unknown time zones but requires your
-	 * RTC not to be off by more than 15 minutes
-	 */
-	real_seconds = secs % 60;
-	real_minutes = secs / 60;
-	if (((abs(real_minutes - cmos_minutes) + 15)/30) & 1)
-		real_minutes += 30;	/* correct for half hour time zone */
-	real_minutes %= 60;
-
-	if (abs(real_minutes - cmos_minutes) < 30) {
-		BIN_TO_BCD(real_seconds);
-		BIN_TO_BCD(real_minutes);
-		ctrl_outb(real_seconds, RSECCNT);
-		ctrl_outb(real_minutes, RMINCNT);
-	} else {
-		printk(KERN_WARNING
-		       "set_rtc_time: can't update from %d to %d\n",
-		       cmos_minutes, real_minutes);
-		retval = -1;
-	}
-
-	ctrl_outb(RCR2_RTCEN|RCR2_START, RCR2);  /* Start RTC */
-	local_irq_restore(flags);
-
-	return retval;
-}
diff --git a/arch/sh/kernel/cpu/sh3/Makefile b/arch/sh/kernel/cpu/sh3/Makefile
index b54dbb9a0c86..58d3815695ff 100644
--- a/arch/sh/kernel/cpu/sh3/Makefile
+++ b/arch/sh/kernel/cpu/sh3/Makefile
@@ -4,10 +4,21 @@
 
 obj-y	:= ex.o probe.o
 
+# CPU subtype setup
+obj-$(CONFIG_CPU_SUBTYPE_SH7705)	+= setup-sh7705.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7706)	+= setup-sh7709.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7707)	+= setup-sh7709.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7708)	+= setup-sh7708.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7709)	+= setup-sh7709.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7300)	+= setup-sh7300.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7710)	+= setup-sh7710.o
+
+# Primary on-chip clocks (common)
 clock-$(CONFIG_CPU_SH3)			:= clock-sh3.o
 clock-$(CONFIG_CPU_SUBTYPE_SH7300)	:= clock-sh7300.o
 clock-$(CONFIG_CPU_SUBTYPE_SH7705)	:= clock-sh7705.o
+clock-$(CONFIG_CPU_SUBTYPE_SH7706)	:= clock-sh7706.o
 clock-$(CONFIG_CPU_SUBTYPE_SH7709)	:= clock-sh7709.o
+clock-$(CONFIG_CPU_SUBTYPE_SH7710)	:= clock-sh7300.o
 
 obj-y	+= $(clock-y)
-
diff --git a/arch/sh/kernel/cpu/sh3/clock-sh7706.c b/arch/sh/kernel/cpu/sh3/clock-sh7706.c
new file mode 100644
index 000000000000..0cf96f9833bc
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh3/clock-sh7706.c
@@ -0,0 +1,84 @@
+/*
+ * arch/sh/kernel/cpu/sh3/clock-sh7706.c
+ *
+ * SH7706 support for the clock framework
+ *
+ *  Copyright (C) 2006  Takashi YOSHII
+ *
+ * Based on arch/sh/kernel/cpu/sh3/clock-sh7709.c
+ *  Copyright (C) 2005  Andriy Skulysh
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <asm/clock.h>
+#include <asm/freq.h>
+#include <asm/io.h>
+
+static int stc_multipliers[] = { 1, 2, 4, 1, 3, 6, 1, 1 };
+static int ifc_divisors[]    = { 1, 2, 4, 1, 3, 1, 1, 1 };
+static int pfc_divisors[]    = { 1, 2, 4, 1, 3, 6, 1, 1 };
+
+static void master_clk_init(struct clk *clk)
+{
+	int frqcr = ctrl_inw(FRQCR);
+	int idx = ((frqcr & 0x2000) >> 11) | (frqcr & 0x0003);
+
+	clk->rate *= pfc_divisors[idx];
+}
+
+static struct clk_ops sh7706_master_clk_ops = {
+	.init		= master_clk_init,
+};
+
+static void module_clk_recalc(struct clk *clk)
+{
+	int frqcr = ctrl_inw(FRQCR);
+	int idx = ((frqcr & 0x2000) >> 11) | (frqcr & 0x0003);
+
+	clk->rate = clk->parent->rate / pfc_divisors[idx];
+}
+
+static struct clk_ops sh7706_module_clk_ops = {
+	.recalc		= module_clk_recalc,
+};
+
+static void bus_clk_recalc(struct clk *clk)
+{
+	int frqcr = ctrl_inw(FRQCR);
+	int idx = ((frqcr & 0x8000) >> 13) | ((frqcr & 0x0030) >> 4);
+
+	clk->rate = clk->parent->rate / stc_multipliers[idx];
+}
+
+static struct clk_ops sh7706_bus_clk_ops = {
+	.recalc		= bus_clk_recalc,
+};
+
+static void cpu_clk_recalc(struct clk *clk)
+{
+	int frqcr = ctrl_inw(FRQCR);
+	int idx = ((frqcr & 0x4000) >> 12) | ((frqcr & 0x000c) >> 2);
+
+	clk->rate = clk->parent->rate / ifc_divisors[idx];
+}
+
+static struct clk_ops sh7706_cpu_clk_ops = {
+	.recalc		= cpu_clk_recalc,
+};
+
+static struct clk_ops *sh7706_clk_ops[] = {
+	&sh7706_master_clk_ops,
+	&sh7706_module_clk_ops,
+	&sh7706_bus_clk_ops,
+	&sh7706_cpu_clk_ops,
+};
+
+void __init arch_init_clk_ops(struct clk_ops **ops, int idx)
+{
+	if (idx < ARRAY_SIZE(sh7706_clk_ops))
+		*ops = sh7706_clk_ops[idx];
+}
diff --git a/arch/sh/kernel/cpu/sh3/ex.S b/arch/sh/kernel/cpu/sh3/ex.S
index cc04e9e239ff..44daf44833f9 100644
--- a/arch/sh/kernel/cpu/sh3/ex.S
+++ b/arch/sh/kernel/cpu/sh3/ex.S
@@ -84,8 +84,12 @@ ENTRY(interrupt_table)
 	.long	do_IRQ	!      rovi
 	.long	do_IRQ			
 	.long	do_IRQ			/* 5E0 */
-#if  defined(CONFIG_CPU_SUBTYPE_SH7707) || defined(CONFIG_CPU_SUBTYPE_SH7709) || \
-     defined(CONFIG_CPU_SUBTYPE_SH7300) || defined(CONFIG_CPU_SUBTYPE_SH7705)
+#if  defined(CONFIG_CPU_SUBTYPE_SH7707) || \
+     defined(CONFIG_CPU_SUBTYPE_SH7709) || \
+     defined(CONFIG_CPU_SUBTYPE_SH7706) || \
+     defined(CONFIG_CPU_SUBTYPE_SH7300) || \
+     defined(CONFIG_CPU_SUBTYPE_SH7705) || \
+     defined(CONFIG_CPU_SUBTYPE_SH7710)
 	.long	do_IRQ	! 32 IRQ  irq0	/* 600 */
 	.long	do_IRQ	! 33      irq1
 	.long	do_IRQ	! 34      irq2
@@ -147,6 +151,51 @@ ENTRY(interrupt_table)
 	.long   do_IRQ	! 62 PCC  pcc0i
 	.long   do_IRQ	! 63      pcc1i	/* 9E0 */
 #endif
+#if defined(CONFIG_CPU_SUBTYPE_SH7710)
+	.long   exception_none	! 61 	/* 9A0 */
+	.long   exception_none	! 62
+	.long   exception_none	! 63
+	.long   exception_none	! 64	/* A00 */
+	.long   exception_none	! 65
+	.long   exception_none	! 66
+	.long   exception_none	! 67
+	.long   exception_none	! 68
+	.long   exception_none	! 69
+	.long   exception_none	! 70
+	.long   exception_none	! 71
+	.long   exception_none	! 72	/* B00 */
+	.long   exception_none	! 73
+	.long   exception_none	! 74
+	.long   exception_none	! 75
+	.long   do_IRQ	! 76 DMAC2 dei4	/* B80 */
+	.long   do_IRQ	! 77 DMAC2 dei5
+	.long   exception_none	! 78
+	.long   do_IRQ	! 79 IPSEC ipseci /* BE0 */
+	.long   do_IRQ	! 80 EDMAC eint0 /* C00 */
+	.long   do_IRQ	! 81 EDMAC eint1
+	.long   do_IRQ	! 82 EDMAC eint2
+	.long   exception_none	! 83	/* C60 */
+	.long   exception_none	! 84
+	.long   exception_none	! 85
+	.long   exception_none	! 86
+	.long   exception_none	! 87
+	.long   exception_none	! 88	/* D00 */
+	.long   exception_none	! 89
+	.long   exception_none	! 90
+	.long   exception_none	! 91
+	.long   exception_none	! 92
+	.long   exception_none	! 93
+	.long   exception_none	! 94
+	.long   exception_none	! 95
+	.long   do_IRQ	! 96 SIOF eri0	/* E00 */
+	.long   do_IRQ	! 97      txi0
+	.long   do_IRQ	! 98      rxi0
+	.long   do_IRQ	! 99      cci0
+	.long   do_IRQ	! 100     eri1	/* E80 */
+	.long   do_IRQ	! 101     txi1
+	.long   do_IRQ	! 102     rxi2
+	.long   do_IRQ	! 103     cci3
+#endif
 #if defined(CONFIG_CPU_SUBTYPE_SH7300)
 	.long   do_IRQ	! 64
 	.long   do_IRQ	! 65
@@ -195,4 +244,3 @@ ENTRY(interrupt_table)
 	.long   do_IRQ	! 108
 #endif
 #endif
-
diff --git a/arch/sh/kernel/cpu/sh3/probe.c b/arch/sh/kernel/cpu/sh3/probe.c
index 5cdc88638601..e67098836290 100644
--- a/arch/sh/kernel/cpu/sh3/probe.c
+++ b/arch/sh/kernel/cpu/sh3/probe.c
@@ -72,6 +72,12 @@ int __init detect_cpu_and_cache_system(void)
 		cpu_data->dcache.sets		= 256;
 		cpu_data->type = CPU_SH7729;
 
+#if defined(CONFIG_CPU_SUBTYPE_SH7706)
+		cpu_data->type = CPU_SH7706;
+#endif
+#if defined(CONFIG_CPU_SUBTYPE_SH7710)
+		cpu_data->type = CPU_SH7710;
+#endif
 #if defined(CONFIG_CPU_SUBTYPE_SH7705)
 		cpu_data->type = CPU_SH7705;
 
diff --git a/arch/sh/kernel/cpu/sh3/setup-sh7300.c b/arch/sh/kernel/cpu/sh3/setup-sh7300.c
new file mode 100644
index 000000000000..ab4d204bfba5
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh3/setup-sh7300.c
@@ -0,0 +1,43 @@
+/*
+ * SH7300 Setup
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <asm/sci.h>
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xa4430000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCI,
+		.irqs		= { 80, 80, 80, 80 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct platform_device *sh7300_devices[] __initdata = {
+	&sci_device,
+};
+
+static int __init sh7300_devices_setup(void)
+{
+	return platform_add_devices(sh7300_devices,
+				    ARRAY_SIZE(sh7300_devices));
+}
+__initcall(sh7300_devices_setup);
diff --git a/arch/sh/kernel/cpu/sh3/setup-sh7705.c b/arch/sh/kernel/cpu/sh3/setup-sh7705.c
new file mode 100644
index 000000000000..a8e41c5241fa
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh3/setup-sh7705.c
@@ -0,0 +1,48 @@
+/*
+ * SH7705 Setup
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <asm/sci.h>
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xa4400000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 52, 53, 55, 54 },
+	}, {
+		.mapbase	= 0xa4410000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 56, 57, 59, 58 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct platform_device *sh7705_devices[] __initdata = {
+	&sci_device,
+};
+
+static int __init sh7705_devices_setup(void)
+{
+	return platform_add_devices(sh7705_devices,
+				    ARRAY_SIZE(sh7705_devices));
+}
+__initcall(sh7705_devices_setup);
diff --git a/arch/sh/kernel/cpu/sh3/setup-sh7708.c b/arch/sh/kernel/cpu/sh3/setup-sh7708.c
new file mode 100644
index 000000000000..f933723911ca
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh3/setup-sh7708.c
@@ -0,0 +1,43 @@
+/*
+ * SH7708 Setup
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <asm/sci.h>
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xfffffe80,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCI,
+		.irqs		= { 23, 24, 25, 0 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct platform_device *sh7708_devices[] __initdata = {
+	&sci_device,
+};
+
+static int __init sh7708_devices_setup(void)
+{
+	return platform_add_devices(sh7708_devices,
+				    ARRAY_SIZE(sh7708_devices));
+}
+__initcall(sh7708_devices_setup);
diff --git a/arch/sh/kernel/cpu/sh3/setup-sh7709.c b/arch/sh/kernel/cpu/sh3/setup-sh7709.c
new file mode 100644
index 000000000000..ff43ef2a1f0c
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh3/setup-sh7709.c
@@ -0,0 +1,53 @@
+/*
+ * SH7707/SH7709 Setup
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <asm/sci.h>
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xfffffe80,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCI,
+		.irqs		= { 23, 24, 25, 0 },
+	}, {
+		.mapbase	= 0xa4000150,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 56, 57, 59, 58 },
+	}, {
+		.mapbase	= 0xa4000140,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_IRDA,
+		.irqs		= { 52, 53, 55, 54 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct platform_device *sh7709_devices[] __initdata = {
+	&sci_device,
+};
+
+static int __init sh7709_devices_setup(void)
+{
+	return platform_add_devices(sh7709_devices,
+				    ARRAY_SIZE(sh7709_devices));
+}
+__initcall(sh7709_devices_setup);
diff --git a/arch/sh/kernel/cpu/sh3/setup-sh7710.c b/arch/sh/kernel/cpu/sh3/setup-sh7710.c
new file mode 100644
index 000000000000..895f99ee6a95
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh3/setup-sh7710.c
@@ -0,0 +1,43 @@
+/*
+ * SH7710 Setup
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <asm/sci.h>
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xa4400000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 52, 53, 55, 54 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct platform_device *sh7710_devices[] __initdata = {
+	&sci_device,
+};
+
+static int __init sh7710_devices_setup(void)
+{
+	return platform_add_devices(sh7710_devices,
+				    ARRAY_SIZE(sh7710_devices));
+}
+__initcall(sh7710_devices_setup);
diff --git a/arch/sh/kernel/cpu/sh4/Makefile b/arch/sh/kernel/cpu/sh4/Makefile
index 3d5cafc71ae3..8dbf3895ece7 100644
--- a/arch/sh/kernel/cpu/sh4/Makefile
+++ b/arch/sh/kernel/cpu/sh4/Makefile
@@ -7,6 +7,16 @@ obj-y	:= ex.o probe.o
 obj-$(CONFIG_SH_FPU)                    += fpu.o
 obj-$(CONFIG_SH_STORE_QUEUES)		+= sq.o
 
+# CPU subtype setup
+obj-$(CONFIG_CPU_SUBTYPE_SH7750)	+= setup-sh7750.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7751)	+= setup-sh7750.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7760)	+= setup-sh7760.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7770)	+= setup-sh7770.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7780)	+= setup-sh7780.o
+obj-$(CONFIG_CPU_SUBTYPE_SH73180)	+= setup-sh73180.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7343)	+= setup-sh7343.o
+obj-$(CONFIG_CPU_SUBTYPE_SH4_202)	+= setup-sh4-202.o
+
 # Primary on-chip clocks (common)
 clock-$(CONFIG_CPU_SH4)			:= clock-sh4.o
 clock-$(CONFIG_CPU_SUBTYPE_SH73180)	:= clock-sh73180.o
diff --git a/arch/sh/kernel/cpu/sh4/ex.S b/arch/sh/kernel/cpu/sh4/ex.S
index 26a27df06505..7146893a6cca 100644
--- a/arch/sh/kernel/cpu/sh4/ex.S
+++ b/arch/sh/kernel/cpu/sh4/ex.S
@@ -72,6 +72,7 @@ ENTRY(interrupt_table)
 	.long	do_IRQ	! 1110
 	.long	exception_error		
 	! Internal hardware
+#ifndef CONFIG_CPU_SUBTYPE_SH7780
 	.long	do_IRQ	! TMU0 tuni0	/* 400 */
 	.long	do_IRQ	! TMU1 tuni1
 	.long	do_IRQ	! TMU2 tuni2
@@ -122,6 +123,13 @@ ENTRY(interrupt_table)
 	.long	do_IRQ	! 45      dmte5
 	.long	do_IRQ	! 46      dmte6
 	.long	do_IRQ	! 47      dmte7		/* 7E0 */
+#elif defined(CONFIG_CPU_SUBTYPE_SH7343)
+	.long	do_IRQ	! 44 IIC1 ali		/* 780 */
+	.long	do_IRQ	! 45      tacki
+	.long	do_IRQ	! 46      waiti
+	.long	do_IRQ	! 47      dtei		/* 7E0 */
+	.long	do_IRQ	! 48 DMAC dei0		/* 800 */
+	.long	do_IRQ	! 49      dei1		/* 820 */
 #else
 	.long	exception_error			! 44	/* 780 */
 	.long	exception_error			! 45
@@ -131,7 +139,8 @@ ENTRY(interrupt_table)
 #if defined(CONFIG_SH_FPU)
 	.long	do_fpu_state_restore	! 48	/* 800 */
 	.long	do_fpu_state_restore	! 49	/* 820 */
-#else
+#elif !defined(CONFIG_CPU_SUBTYPE_SH7343) && \
+      !defined(CONFIG_CPU_SUBTYPE_SH73180)
 	.long	exception_error
 	.long	exception_error
 #endif
@@ -224,7 +233,7 @@ ENTRY(interrupt_table)
 	.long	exception_error
 	.long	do_IRQ	! ADC	adi
 	.long	do_IRQ	! CMT	cmti	/* FA0 */
-#elif defined(CONFIG_CPU_SUBTYPE_SH73180)
+#elif defined(CONFIG_CPU_SUBTYPE_SH73180) || defined(CONFIG_CPU_SUBTYPE_SH7343)
 	.long	do_IRQ	!  50 0x840
 	.long	do_IRQ	!  51 0x860
 	.long	do_IRQ	!  52 0x880
@@ -379,5 +388,168 @@ ENTRY(interrupt_table)
 	.long	exception_error			! 141 0x13a0
 	.long	exception_error			! 142 0x13c0
 	.long	exception_error			! 143 0x13e0
+#elif defined(CONFIG_CPU_SUBTYPE_SH7770)
+	.long	do_IRQ	!  50 0x840
+	.long	do_IRQ	!  51 0x860
+	.long	do_IRQ	!  52 0x880
+	.long	do_IRQ	!  53 0x8a0
+	.long	do_IRQ	!  54 0x8c0
+	.long	do_IRQ	!  55 0x8e0
+	.long	do_IRQ	!  56 0x900
+	.long	do_IRQ	!  57 0x920
+	.long	do_IRQ	!  58 0x940
+	.long	do_IRQ	!  59 0x960
+	.long	do_IRQ	!  60 0x980
+	.long	do_IRQ	!  61 0x9a0
+	.long	do_IRQ	!  62 0x9c0
+	.long	do_IRQ	!  63 0x9e0
+	.long	do_IRQ	!  64 0xa00
+	.long	do_IRQ	!  65 0xa20
+	.long	do_IRQ	!  66 0xa4d
+	.long	do_IRQ	!  67 0xa60
+	.long	do_IRQ	!  68 0xa80
+	.long	do_IRQ	!  69 0xaa0
+	.long	do_IRQ	!  70 0xac0
+	.long	do_IRQ	!  71 0xae0
+	.long	do_IRQ	!  72 0xb00
+	.long	do_IRQ	!  73 0xb20
+	.long	do_IRQ	!  74 0xb40
+	.long	do_IRQ	!  75 0xb60
+	.long	do_IRQ	!  76 0xb80
+	.long	do_IRQ	!  77 0xba0
+	.long	do_IRQ	!  78 0xbc0
+	.long	do_IRQ	!  79 0xbe0
+	.long	do_IRQ	!  80 0xc00
+	.long	do_IRQ	!  81 0xc20
+	.long	do_IRQ	!  82 0xc40
+	.long	do_IRQ	!  83 0xc60
+	.long	do_IRQ	!  84 0xc80
+	.long	do_IRQ	!  85 0xca0
+	.long	do_IRQ	!  86 0xcc0
+	.long	do_IRQ	!  87 0xce0
+	.long	do_IRQ	!  88 0xd00
+	.long	do_IRQ	!  89 0xd20
+	.long	do_IRQ	!  90 0xd40
+	.long	do_IRQ	!  91 0xd60
+	.long	do_IRQ	!  92 0xd80
+	.long	do_IRQ	!  93 0xda0
+	.long	do_IRQ	!  94 0xdc0
+	.long	do_IRQ	!  95 0xde0
+	.long	do_IRQ	!  96 0xe00
+	.long	do_IRQ	!  97 0xe20
+	.long	do_IRQ	!  98 0xe40
+	.long	do_IRQ	!  99 0xe60
+	.long	do_IRQ	! 100 0xe80
+	.long	do_IRQ	! 101 0xea0
+	.long	do_IRQ	! 102 0xec0
+	.long	do_IRQ	! 103 0xee0
+	.long	do_IRQ	! 104 0xf00
+	.long	do_IRQ	! 105 0xf20
+	.long	do_IRQ	! 106 0xf40
+	.long	do_IRQ	! 107 0xf60
+	.long	do_IRQ	! 108 0xf80
+#endif
+#else
+	.long	exception_error		/* 400 */
+	.long	exception_error
+	.long	exception_error
+	.long	exception_error
+	.long	do_IRQ	! RTC	ati
+	.long	do_IRQ	!	pri
+	.long	do_IRQ	!	cui
+	.long	exception_error
+	.long	exception_error		/* 500 */
+	.long	exception_error
+	.long	exception_error
+	.long	do_IRQ	! WDT	iti	/* 560 */
+	.long	do_IRQ	! TMU-ch0
+	.long	do_IRQ	! TMU-ch1
+	.long	do_IRQ	! TMU-ch2
+	.long	do_IRQ	! ticpi2	/* 5E0 */
+	.long	do_IRQ	! 32 Hitachi UDI	/* 600 */
+	.long	exception_error
+	.long	do_IRQ	! 34 DMAC dmte0
+	.long	do_IRQ	! 35	  dmte1
+	.long	do_IRQ	! 36	  dmte2
+	.long	do_IRQ	! 37	  dmte3
+	.long	do_IRQ	! 38	  dmae
+	.long	exception_error			! 39	/* 6E0 */
+	.long	do_IRQ	! 40 SCIF-ch0 eri		/* 700 */
+	.long	do_IRQ	! 41	      rxi
+	.long	do_IRQ	! 42	      bri
+	.long	do_IRQ	! 43	      txi
+	.long	do_IRQ	! 44 DMAC dmte4		/* 780 */
+	.long	do_IRQ	! 45	  dmte5
+	.long	do_IRQ	! 46	  dmte6
+	.long	do_IRQ	! 47	  dmte7		/* 7E0 */
+#if defined(CONFIG_SH_FPU)
+	.long	do_fpu_state_restore	! 48	/* 800 */
+	.long	do_fpu_state_restore	! 49	/* 820 */
+#else
+	.long	exception_error
+	.long	exception_error
+#endif
+	.long	exception_error			/* 840 */
+	.long	exception_error
+	.long	exception_error
+	.long	exception_error
+	.long	exception_error
+	.long	exception_error
+	.long	do_IRQ	! 56 CMT	/* 900 */
+	.long	exception_error
+	.long	exception_error
+	.long	exception_error
+	.long	do_IRQ	! 60 HAC
+	.long	exception_error
+	.long	exception_error
+	.long	exception_error
+	.long	do_IRQ	! PCI serr	/* A00 */
+	.long	do_IRQ	!     INTA
+	.long	do_IRQ	!     INTB
+	.long	do_IRQ	!     INTC
+	.long	do_IRQ	!     INTD
+	.long	do_IRQ	!     err
+	.long	do_IRQ	!     pwd3
+	.long	do_IRQ	!     pwd2
+	.long	do_IRQ	!     pwd1	/* B00 */
+	.long	do_IRQ	!     pwd0
+	.long	exception_error
+	.long	exception_error
+	.long	do_IRQ	! SCIF-ch1 eri	/* B80 */
+	.long	do_IRQ	!	   rxi
+	.long	do_IRQ	!	   bri
+	.long	do_IRQ	!	   txi
+	.long	do_IRQ	! SIOF		/* C00 */
+	.long	exception_error
+	.long	exception_error
+	.long	exception_error
+	.long	do_IRQ	! HSPI		/* C80 */
+	.long	exception_error
+	.long	exception_error
+	.long	exception_error
+	.long	do_IRQ	! MMCIF	fatat	/* D00 */
+	.long	do_IRQ	!	tran
+	.long	do_IRQ	!	err
+	.long	do_IRQ	!	frdy
+	.long	do_IRQ	! DMAC dmint8	/* D80 */
+	.long	do_IRQ	!      dmint9
+	.long	do_IRQ	!      dmint10
+	.long	do_IRQ	!      dmint11
+	.long	do_IRQ	! TMU-ch3	/* E00 */
+	.long	do_IRQ	! TMU-ch4
+	.long	do_IRQ	! TMU-ch5
+	.long	exception_error
+	.long	do_IRQ	! SSI
+	.long	exception_error
+	.long	exception_error
+	.long	exception_error
+	.long	do_IRQ	! FLCTL	flste	/* F00 */
+	.long	do_IRQ	!	fltend
+	.long	do_IRQ	!	fltrq0
+	.long	do_IRQ	!	fltrq1
+	.long	do_IRQ	! GPIO gpioi0	/* F80 */
+	.long	do_IRQ	!      gpioi1
+	.long	do_IRQ	!      gpioi2
+	.long	do_IRQ	!      gpioi3
 #endif
 
diff --git a/arch/sh/kernel/cpu/sh4/probe.c b/arch/sh/kernel/cpu/sh4/probe.c
index 42427b79697b..c294de1e14a3 100644
--- a/arch/sh/kernel/cpu/sh4/probe.c
+++ b/arch/sh/kernel/cpu/sh4/probe.c
@@ -3,7 +3,7 @@
  *
  * CPU Subtype Probing for SH-4.
  *
- * Copyright (C) 2001, 2002, 2003, 2004  Paul Mundt
+ * Copyright (C) 2001 - 2006  Paul Mundt
  * Copyright (C) 2003  Richard Curnow
  *
  * This file is subject to the terms and conditions of the GNU General Public
@@ -29,7 +29,7 @@ int __init detect_cpu_and_cache_system(void)
 		[9] = (1 << 16)
 	};
 
-	pvr = (ctrl_inl(CCN_PVR) >> 8) & 0xffff;
+	pvr = (ctrl_inl(CCN_PVR) >> 8) & 0xffffff;
 	prr = (ctrl_inl(CCN_PRR) >> 4) & 0xff;
 	cvr = (ctrl_inl(CCN_CVR));
 
@@ -38,7 +38,6 @@ int __init detect_cpu_and_cache_system(void)
 	 */
 	cpu_data->icache.way_incr	= (1 << 13);
 	cpu_data->icache.entry_shift	= 5;
-	cpu_data->icache.entry_mask	= 0x1fe0;
 	cpu_data->icache.sets		= 256;
 	cpu_data->icache.ways		= 1;
 	cpu_data->icache.linesz		= L1_CACHE_BYTES;
@@ -48,13 +47,29 @@ int __init detect_cpu_and_cache_system(void)
 	 */
 	cpu_data->dcache.way_incr	= (1 << 14);
 	cpu_data->dcache.entry_shift	= 5;
-	cpu_data->dcache.entry_mask	= 0x3fe0;
 	cpu_data->dcache.sets		= 512;
 	cpu_data->dcache.ways		= 1;
 	cpu_data->dcache.linesz		= L1_CACHE_BYTES;
 
-	/* Set the FPU flag, virtually all SH-4's have one */
-	cpu_data->flags |= CPU_HAS_FPU;
+	/*
+	 * Setup some generic flags we can probe
+	 * (L2 and DSP detection only work on SH-4A)
+	 */
+	if (((pvr >> 16) & 0xff) == 0x10) {
+		if ((cvr & 0x02000000) == 0)
+			cpu_data->flags |= CPU_HAS_L2_CACHE;
+		if ((cvr & 0x10000000) == 0)
+			cpu_data->flags |= CPU_HAS_DSP;
+
+		cpu_data->flags |= CPU_HAS_LLSC;
+	}
+
+	/* FPU detection works for everyone */
+	if ((cvr & 0x20000000) == 1)
+		cpu_data->flags |= CPU_HAS_FPU;
+
+	/* Mask off the upper chip ID */
+	pvr &= 0xffff;
 
 	/*
 	 * Probe the underlying processor version/revision and
@@ -63,56 +78,101 @@ int __init detect_cpu_and_cache_system(void)
 	switch (pvr) {
 	case 0x205:
 		cpu_data->type = CPU_SH7750;
-		cpu_data->flags |= CPU_HAS_P2_FLUSH_BUG | CPU_HAS_PERF_COUNTER;
+		cpu_data->flags |= CPU_HAS_P2_FLUSH_BUG | CPU_HAS_FPU |
+				   CPU_HAS_PERF_COUNTER | CPU_HAS_PTEA;
 		break;
 	case 0x206:
 		cpu_data->type = CPU_SH7750S;
-		cpu_data->flags |= CPU_HAS_P2_FLUSH_BUG | CPU_HAS_PERF_COUNTER;
+		cpu_data->flags |= CPU_HAS_P2_FLUSH_BUG | CPU_HAS_FPU |
+				   CPU_HAS_PERF_COUNTER | CPU_HAS_PTEA;
 		break;
 	case 0x1100:
 		cpu_data->type = CPU_SH7751;
+		cpu_data->flags |= CPU_HAS_FPU | CPU_HAS_PTEA;
 		break;
 	case 0x2000:
 		cpu_data->type = CPU_SH73180;
 		cpu_data->icache.ways = 4;
 		cpu_data->dcache.ways = 4;
-		cpu_data->flags &= ~CPU_HAS_FPU;
+		cpu_data->flags |= CPU_HAS_LLSC;
+		break;
+	case 0x2001:
+	case 0x2004:
+		cpu_data->type = CPU_SH7770;
+		cpu_data->icache.ways = 4;
+		cpu_data->dcache.ways = 4;
+
+		cpu_data->flags |= CPU_HAS_FPU | CPU_HAS_LLSC;
+		break;
+	case 0x2006:
+	case 0x200A:
+		if (prr == 0x61)
+			cpu_data->type = CPU_SH7781;
+		else
+			cpu_data->type = CPU_SH7780;
+
+		cpu_data->icache.ways = 4;
+		cpu_data->dcache.ways = 4;
+
+		cpu_data->flags |= CPU_HAS_FPU | CPU_HAS_PERF_COUNTER |
+				   CPU_HAS_LLSC;
+		break;
+	case 0x3000:
+	case 0x3003:
+		cpu_data->type = CPU_SH7343;
+		cpu_data->icache.ways = 4;
+		cpu_data->dcache.ways = 4;
+		cpu_data->flags |= CPU_HAS_LLSC;
 		break;
 	case 0x8000:
 		cpu_data->type = CPU_ST40RA;
+		cpu_data->flags |= CPU_HAS_FPU | CPU_HAS_PTEA;
 		break;
 	case 0x8100:
 		cpu_data->type = CPU_ST40GX1;
+		cpu_data->flags |= CPU_HAS_FPU | CPU_HAS_PTEA;
 		break;
 	case 0x700:
 		cpu_data->type = CPU_SH4_501;
 		cpu_data->icache.ways = 2;
 		cpu_data->dcache.ways = 2;
-
-		/* No FPU on the SH4-500 series.. */
-		cpu_data->flags &= ~CPU_HAS_FPU;
+		cpu_data->flags |= CPU_HAS_PTEA;
 		break;
 	case 0x600:
 		cpu_data->type = CPU_SH4_202;
 		cpu_data->icache.ways = 2;
 		cpu_data->dcache.ways = 2;
+		cpu_data->flags |= CPU_HAS_FPU | CPU_HAS_PTEA;
 		break;
 	case 0x500 ... 0x501:
 		switch (prr) {
-		    case 0x10: cpu_data->type = CPU_SH7750R; break;
-		    case 0x11: cpu_data->type = CPU_SH7751R; break;
-		    case 0x50: cpu_data->type = CPU_SH7760;  break;
+		case 0x10:
+			cpu_data->type = CPU_SH7750R;
+			break;
+		case 0x11:
+			cpu_data->type = CPU_SH7751R;
+			break;
+		case 0x50 ... 0x5f:
+			cpu_data->type = CPU_SH7760;
+			break;
 		}
 
 		cpu_data->icache.ways = 2;
 		cpu_data->dcache.ways = 2;
 
+		cpu_data->flags |= CPU_HAS_FPU | CPU_HAS_PTEA;
+
 		break;
 	default:
 		cpu_data->type = CPU_SH_NONE;
 		break;
 	}
 
+#ifdef CONFIG_SH_DIRECT_MAPPED
+	cpu_data->icache.ways = 1;
+	cpu_data->dcache.ways = 1;
+#endif
+
 	/*
 	 * On anything that's not a direct-mapped cache, look to the CVR
 	 * for I/D-cache specifics.
@@ -121,18 +181,56 @@ int __init detect_cpu_and_cache_system(void)
 		size = sizes[(cvr >> 20) & 0xf];
 		cpu_data->icache.way_incr	= (size >> 1);
 		cpu_data->icache.sets		= (size >> 6);
-		cpu_data->icache.entry_mask	=
-			(cpu_data->icache.way_incr - (1 << 5));
+
 	}
 
+	/* Setup the rest of the I-cache info */
+	cpu_data->icache.entry_mask = cpu_data->icache.way_incr -
+				      cpu_data->icache.linesz;
+
+	cpu_data->icache.way_size = cpu_data->icache.sets *
+				    cpu_data->icache.linesz;
+
+	/* And the rest of the D-cache */
 	if (cpu_data->dcache.ways > 1) {
 		size = sizes[(cvr >> 16) & 0xf];
 		cpu_data->dcache.way_incr	= (size >> 1);
 		cpu_data->dcache.sets		= (size >> 6);
-		cpu_data->dcache.entry_mask	=
-			(cpu_data->dcache.way_incr - (1 << 5));
+	}
+
+	cpu_data->dcache.entry_mask = cpu_data->dcache.way_incr -
+				      cpu_data->dcache.linesz;
+
+	cpu_data->dcache.way_size = cpu_data->dcache.sets *
+				    cpu_data->dcache.linesz;
+
+	/*
+	 * Setup the L2 cache desc
+	 *
+	 * SH-4A's have an optional PIPT L2.
+	 */
+	if (cpu_data->flags & CPU_HAS_L2_CACHE) {
+		/*
+		 * Size calculation is much more sensible
+		 * than it is for the L1.
+		 *
+		 * Sizes are 128KB, 258KB, 512KB, and 1MB.
+		 */
+		size = (cvr & 0xf) << 17;
+
+		BUG_ON(!size);
+
+		cpu_data->scache.way_incr	= (1 << 16);
+		cpu_data->scache.entry_shift	= 5;
+		cpu_data->scache.ways		= 4;
+		cpu_data->scache.linesz		= L1_CACHE_BYTES;
+		cpu_data->scache.entry_mask	=
+			(cpu_data->scache.way_incr - cpu_data->scache.linesz);
+		cpu_data->scache.sets		= size /
+			(cpu_data->scache.linesz * cpu_data->scache.ways);
+		cpu_data->scache.way_size	=
+			(cpu_data->scache.sets * cpu_data->scache.linesz);
 	}
 
 	return 0;
 }
-
diff --git a/arch/sh/kernel/cpu/sh4/setup-sh4-202.c b/arch/sh/kernel/cpu/sh4/setup-sh4-202.c
new file mode 100644
index 000000000000..6e4e96541358
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh4/setup-sh4-202.c
@@ -0,0 +1,43 @@
+/*
+ * SH4-202 Setup
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <asm/sci.h>
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xffe80000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 40, 41, 43, 42 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct platform_device *sh4202_devices[] __initdata = {
+	&sci_device,
+};
+
+static int __init sh4202_devices_setup(void)
+{
+	return platform_add_devices(sh4202_devices,
+				    ARRAY_SIZE(sh4202_devices));
+}
+__initcall(sh4202_devices_setup);
diff --git a/arch/sh/kernel/cpu/sh4/setup-sh73180.c b/arch/sh/kernel/cpu/sh4/setup-sh73180.c
new file mode 100644
index 000000000000..cc9ea1e2e5df
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh4/setup-sh73180.c
@@ -0,0 +1,43 @@
+/*
+ * SH73180 Setup
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <asm/sci.h>
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xffe80000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 80, 81, 83, 82 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct platform_device *sh73180_devices[] __initdata = {
+	&sci_device,
+};
+
+static int __init sh73180_devices_setup(void)
+{
+	return platform_add_devices(sh73180_devices,
+				    ARRAY_SIZE(sh73180_devices));
+}
+__initcall(sh73180_devices_setup);
diff --git a/arch/sh/kernel/cpu/sh4/setup-sh7343.c b/arch/sh/kernel/cpu/sh4/setup-sh7343.c
new file mode 100644
index 000000000000..91d61cf91ba1
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh4/setup-sh7343.c
@@ -0,0 +1,43 @@
+/*
+ * SH7343 Setup
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <asm/sci.h>
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xffe00000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 80, 81, 83, 82 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct platform_device *sh7343_devices[] __initdata = {
+	&sci_device,
+};
+
+static int __init sh7343_devices_setup(void)
+{
+	return platform_add_devices(sh7343_devices,
+				    ARRAY_SIZE(sh7343_devices));
+}
+__initcall(sh7343_devices_setup);
diff --git a/arch/sh/kernel/cpu/sh4/setup-sh7750.c b/arch/sh/kernel/cpu/sh4/setup-sh7750.c
new file mode 100644
index 000000000000..50812d57c1c1
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh4/setup-sh7750.c
@@ -0,0 +1,48 @@
+/*
+ * SH7750/SH7751 Setup
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <asm/sci.h>
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xffe00000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCI,
+		.irqs		= { 23, 24, 25, 0 },
+	}, {
+		.mapbase	= 0xffe80000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 40, 41, 43, 42 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct platform_device *sh7750_devices[] __initdata = {
+	&sci_device,
+};
+
+static int __init sh7750_devices_setup(void)
+{
+	return platform_add_devices(sh7750_devices,
+				    ARRAY_SIZE(sh7750_devices));
+}
+__initcall(sh7750_devices_setup);
diff --git a/arch/sh/kernel/cpu/sh4/setup-sh7760.c b/arch/sh/kernel/cpu/sh4/setup-sh7760.c
new file mode 100644
index 000000000000..97f1c9af35d6
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh4/setup-sh7760.c
@@ -0,0 +1,53 @@
+/*
+ * SH7760 Setup
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <asm/sci.h>
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xfe600000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 52, 53, 55, 54 },
+	}, {
+		.mapbase	= 0xfe610000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 72, 73, 75, 74 },
+	}, {
+		.mapbase	= 0xfe620000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 76, 77, 79, 78 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct platform_device *sh7760_devices[] __initdata = {
+	&sci_device,
+};
+
+static int __init sh7760_devices_setup(void)
+{
+	return platform_add_devices(sh7760_devices,
+				    ARRAY_SIZE(sh7760_devices));
+}
+__initcall(sh7760_devices_setup);
diff --git a/arch/sh/kernel/cpu/sh4/setup-sh7770.c b/arch/sh/kernel/cpu/sh4/setup-sh7770.c
new file mode 100644
index 000000000000..6a04cc5f5aca
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh4/setup-sh7770.c
@@ -0,0 +1,53 @@
+/*
+ * SH7770 Setup
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <asm/sci.h>
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xff923000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 61, 61, 61, 61 },
+	}, {
+		.mapbase	= 0xff924000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 62, 62, 62, 62 },
+	}, {
+		.mapbase	= 0xff925000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 63, 63, 63, 63 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct platform_device *sh7770_devices[] __initdata = {
+	&sci_device,
+};
+
+static int __init sh7770_devices_setup(void)
+{
+	return platform_add_devices(sh7770_devices,
+				    ARRAY_SIZE(sh7770_devices));
+}
+__initcall(sh7770_devices_setup);
diff --git a/arch/sh/kernel/cpu/sh4/setup-sh7780.c b/arch/sh/kernel/cpu/sh4/setup-sh7780.c
new file mode 100644
index 000000000000..72493f259edc
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh4/setup-sh7780.c
@@ -0,0 +1,79 @@
+/*
+ * SH7780 Setup
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <asm/sci.h>
+
+static struct resource rtc_resources[] = {
+	[0] = {
+		.start	= 0xffe80000,
+		.end	= 0xffe80000 + 0x58 - 1,
+		.flags	= IORESOURCE_IO,
+	},
+	[1] = {
+		/* Period IRQ */
+		.start	= 21,
+		.flags	= IORESOURCE_IRQ,
+	},
+	[2] = {
+		/* Carry IRQ */
+		.start	= 22,
+		.flags	= IORESOURCE_IRQ,
+	},
+	[3] = {
+		/* Alarm IRQ */
+		.start	= 23,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device rtc_device = {
+	.name		= "sh-rtc",
+	.id		= -1,
+	.num_resources	= ARRAY_SIZE(rtc_resources),
+	.resource	= rtc_resources,
+};
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xffe00000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 40, 41, 43, 42 },
+	}, {
+		.mapbase	= 0xffe10000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 76, 77, 79, 78 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct platform_device *sh7780_devices[] __initdata = {
+	&rtc_device,
+	&sci_device,
+};
+
+static int __init sh7780_devices_setup(void)
+{
+	return platform_add_devices(sh7780_devices,
+				    ARRAY_SIZE(sh7780_devices));
+}
+__initcall(sh7780_devices_setup);
diff --git a/arch/sh/kernel/cpu/sh4/sq.c b/arch/sh/kernel/cpu/sh4/sq.c
index b09805f3ee23..7bcc73f9b8df 100644
--- a/arch/sh/kernel/cpu/sh4/sq.c
+++ b/arch/sh/kernel/cpu/sh4/sq.c
@@ -1,49 +1,52 @@
 /*
- * arch/sh/kernel/cpu/sq.c
+ * arch/sh/kernel/cpu/sh4/sq.c
  *
  * General management API for SH-4 integrated Store Queues
  *
- * Copyright (C) 2001, 2002, 2003, 2004  Paul Mundt
+ * Copyright (C) 2001 - 2006  Paul Mundt
  * Copyright (C) 2001, 2002  M. R. Brown
  *
- * Some of this code has been adopted directly from the old arch/sh/mm/sq.c
- * hack that was part of the LinuxDC project. For all intents and purposes,
- * this is a completely new interface that really doesn't have much in common
- * with the old zone-based approach at all. In fact, it's only listed here for
- * general completeness.
- *
  * This file is subject to the terms and conditions of the GNU General Public
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  */
 #include <linux/init.h>
+#include <linux/cpu.h>
+#include <linux/bitmap.h>
+#include <linux/sysdev.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/slab.h>
-#include <linux/list.h>
-#include <linux/proc_fs.h>
-#include <linux/miscdevice.h>
 #include <linux/vmalloc.h>
-
+#include <linux/mm.h>
 #include <asm/io.h>
 #include <asm/page.h>
-#include <asm/mmu_context.h>
+#include <asm/cacheflush.h>
 #include <asm/cpu/sq.h>
 
-static LIST_HEAD(sq_mapping_list);
+struct sq_mapping;
+
+struct sq_mapping {
+	const char *name;
+
+	unsigned long sq_addr;
+	unsigned long addr;
+	unsigned int size;
+
+	struct sq_mapping *next;
+};
+
+static struct sq_mapping *sq_mapping_list;
 static DEFINE_SPINLOCK(sq_mapping_lock);
+static kmem_cache_t *sq_cache;
+static unsigned long *sq_bitmap;
 
-/**
- * sq_flush - Flush (prefetch) the store queue cache
- * @addr: the store queue address to flush
- *
- * Executes a prefetch instruction on the specified store queue cache,
- * so that the cached data is written to physical memory.
- */
-inline void sq_flush(void *addr)
-{
-	__asm__ __volatile__ ("pref @%0" : : "r" (addr) : "memory");
-}
+#define store_queue_barrier()			\
+do {						\
+	(void)ctrl_inl(P4SEG_STORE_QUE);	\
+	ctrl_outl(0, P4SEG_STORE_QUE + 0);	\
+	ctrl_outl(0, P4SEG_STORE_QUE + 8);	\
+} while (0);
 
 /**
  * sq_flush_range - Flush (prefetch) a specific SQ range
@@ -56,152 +59,73 @@ inline void sq_flush(void *addr)
 void sq_flush_range(unsigned long start, unsigned int len)
 {
 	volatile unsigned long *sq = (unsigned long *)start;
-	unsigned long dummy;
 
 	/* Flush the queues */
 	for (len >>= 5; len--; sq += 8)
-		sq_flush((void *)sq);
+		prefetchw((void *)sq);
 
 	/* Wait for completion */
-	dummy = ctrl_inl(P4SEG_STORE_QUE);
-
-	ctrl_outl(0, P4SEG_STORE_QUE + 0);
-	ctrl_outl(0, P4SEG_STORE_QUE + 8);
+	store_queue_barrier();
 }
 
-static struct sq_mapping *__sq_alloc_mapping(unsigned long virt, unsigned long phys, unsigned long size, const char *name)
+static inline void sq_mapping_list_add(struct sq_mapping *map)
 {
-	struct sq_mapping *map;
-
-	if (virt + size > SQ_ADDRMAX)
-		return ERR_PTR(-ENOSPC);
-
-	map = kmalloc(sizeof(struct sq_mapping), GFP_KERNEL);
-	if (!map)
-		return ERR_PTR(-ENOMEM);
+	struct sq_mapping **p, *tmp;
 
-	INIT_LIST_HEAD(&map->list);
+	spin_lock_irq(&sq_mapping_lock);
 
-	map->sq_addr	= virt;
-	map->addr	= phys;
-	map->size	= size + 1;
-	map->name	= name;
+	p = &sq_mapping_list;
+	while ((tmp = *p) != NULL)
+		p = &tmp->next;
 
-	list_add(&map->list, &sq_mapping_list);
+	map->next = tmp;
+	*p = map;
 
-	return map;
+	spin_unlock_irq(&sq_mapping_lock);
 }
 
-static unsigned long __sq_get_next_addr(void)
+static inline void sq_mapping_list_del(struct sq_mapping *map)
 {
-	if (!list_empty(&sq_mapping_list)) {
-		struct list_head *pos, *tmp;
-
-		/*
-		 * Read one off the list head, as it will have the highest
-		 * mapped allocation. Set the next one up right above it.
-		 *
-		 * This is somewhat sub-optimal, as we don't look at
-		 * gaps between allocations or anything lower then the
-		 * highest-level allocation.
-		 *
-		 * However, in the interest of performance and the general
-		 * lack of desire to do constant list rebalancing, we don't
-		 * worry about it.
-		 */
-		list_for_each_safe(pos, tmp, &sq_mapping_list) {
-			struct sq_mapping *entry;
-
-			entry = list_entry(pos, typeof(*entry), list);
-
-			return entry->sq_addr + entry->size;
+	struct sq_mapping **p, *tmp;
+
+	spin_lock_irq(&sq_mapping_lock);
+
+	for (p = &sq_mapping_list; (tmp = *p); p = &tmp->next)
+		if (tmp == map) {
+			*p = tmp->next;
+			break;
 		}
-	}
 
-	return P4SEG_STORE_QUE;
+	spin_unlock_irq(&sq_mapping_lock);
 }
 
-/**
- * __sq_remap - Perform a translation from the SQ to a phys addr
- * @map: sq mapping containing phys and store queue addresses.
- *
- * Maps the store queue address specified in the mapping to the physical
- * address specified in the mapping.
- */
-static struct sq_mapping *__sq_remap(struct sq_mapping *map)
+static int __sq_remap(struct sq_mapping *map, unsigned long flags)
 {
-	unsigned long flags, pteh, ptel;
+#if defined(CONFIG_MMU)
 	struct vm_struct *vma;
-	pgprot_t pgprot;
-
-	/*
-	 * Without an MMU (or with it turned off), this is much more
-	 * straightforward, as we can just load up each queue's QACR with
-	 * the physical address appropriately masked.
-	 */
-
-	ctrl_outl(((map->addr >> 26) << 2) & 0x1c, SQ_QACR0);
-	ctrl_outl(((map->addr >> 26) << 2) & 0x1c, SQ_QACR1);
 
-#ifdef CONFIG_MMU
-	/*
-	 * With an MMU on the other hand, things are slightly more involved.
-	 * Namely, we have to have a direct mapping between the SQ addr and
-	 * the associated physical address in the UTLB by way of setting up
-	 * a virt<->phys translation by hand. We do this by simply specifying
-	 * the SQ addr in UTLB.VPN and the associated physical address in
-	 * UTLB.PPN.
-	 *
-	 * Notably, even though this is a special case translation, and some
-	 * of the configuration bits are meaningless, we're still required
-	 * to have a valid ASID context in PTEH.
-	 *
-	 * We could also probably get by without explicitly setting PTEA, but
-	 * we do it here just for good measure.
-	 */
-	spin_lock_irqsave(&sq_mapping_lock, flags);
-
-	pteh = map->sq_addr;
-	ctrl_outl((pteh & MMU_VPN_MASK) | get_asid(), MMU_PTEH);
-
-	ptel = map->addr & PAGE_MASK;
-	ctrl_outl(((ptel >> 28) & 0xe) | (ptel & 0x1), MMU_PTEA);
-
-	pgprot = pgprot_noncached(PAGE_KERNEL);
-
-	ptel &= _PAGE_FLAGS_HARDWARE_MASK;
-	ptel |= pgprot_val(pgprot);
-	ctrl_outl(ptel, MMU_PTEL);
-
-	__asm__ __volatile__ ("ldtlb" : : : "memory");
-
-	spin_unlock_irqrestore(&sq_mapping_lock, flags);
-
-	/*
-	 * Next, we need to map ourselves in the kernel page table, so that
-	 * future accesses after a TLB flush will be handled when we take a
-	 * page fault.
-	 *
-	 * Theoretically we could just do this directly and not worry about
-	 * setting up the translation by hand ahead of time, but for the
-	 * cases where we want a one-shot SQ mapping followed by a quick
-	 * writeout before we hit the TLB flush, we do it anyways. This way
-	 * we at least save ourselves the initial page fault overhead.
-	 */
 	vma = __get_vm_area(map->size, VM_ALLOC, map->sq_addr, SQ_ADDRMAX);
 	if (!vma)
-		return ERR_PTR(-ENOMEM);
+		return -ENOMEM;
 
 	vma->phys_addr = map->addr;
 
 	if (remap_area_pages((unsigned long)vma->addr, vma->phys_addr,
-			     map->size, pgprot_val(pgprot))) {
+			     map->size, flags)) {
 		vunmap(vma->addr);
-		return NULL;
+		return -EAGAIN;
 	}
-#endif /* CONFIG_MMU */
+#else
+	/*
+	 * Without an MMU (or with it turned off), this is much more
+	 * straightforward, as we can just load up each queue's QACR with
+	 * the physical address appropriately masked.
+	 */
+	ctrl_outl(((map->addr >> 26) << 2) & 0x1c, SQ_QACR0);
+	ctrl_outl(((map->addr >> 26) << 2) & 0x1c, SQ_QACR1);
+#endif
 
-	return map;
+	return 0;
 }
 
 /**
@@ -209,42 +133,65 @@ static struct sq_mapping *__sq_remap(struct sq_mapping *map)
  * @phys: Physical address of mapping.
  * @size: Length of mapping.
  * @name: User invoking mapping.
+ * @flags: Protection flags.
  *
  * Remaps the physical address @phys through the next available store queue
  * address of @size length. @name is logged at boot time as well as through
- * the procfs interface.
- *
- * A pre-allocated and filled sq_mapping pointer is returned, and must be
- * cleaned up with a call to sq_unmap() when the user is done with the
- * mapping.
+ * the sysfs interface.
  */
-struct sq_mapping *sq_remap(unsigned long phys, unsigned int size, const char *name)
+unsigned long sq_remap(unsigned long phys, unsigned int size,
+		       const char *name, unsigned long flags)
 {
 	struct sq_mapping *map;
-	unsigned long virt, end;
+	unsigned long end;
 	unsigned int psz;
+	int ret, page;
 
 	/* Don't allow wraparound or zero size */
 	end = phys + size - 1;
-	if (!size || end < phys)
-		return NULL;
+	if (unlikely(!size || end < phys))
+		return -EINVAL;
 	/* Don't allow anyone to remap normal memory.. */
-	if (phys < virt_to_phys(high_memory))
-		return NULL;
+	if (unlikely(phys < virt_to_phys(high_memory)))
+		return -EINVAL;
 
 	phys &= PAGE_MASK;
+	size = PAGE_ALIGN(end + 1) - phys;
+
+	map = kmem_cache_alloc(sq_cache, GFP_KERNEL);
+	if (unlikely(!map))
+		return -ENOMEM;
+
+	map->addr = phys;
+	map->size = size;
+	map->name = name;
+
+	page = bitmap_find_free_region(sq_bitmap, 0x04000000,
+				       get_order(map->size));
+	if (unlikely(page < 0)) {
+		ret = -ENOSPC;
+		goto out;
+	}
+
+	map->sq_addr = P4SEG_STORE_QUE + (page << PAGE_SHIFT);
+
+	ret = __sq_remap(map, flags);
+	if (unlikely(ret != 0))
+		goto out;
+
+	psz = (size + (PAGE_SIZE - 1)) >> PAGE_SHIFT;
+	pr_info("sqremap: %15s  [%4d page%s]  va 0x%08lx   pa 0x%08lx\n",
+		likely(map->name) ? map->name : "???",
+		psz, psz == 1 ? " " : "s",
+		map->sq_addr, map->addr);
 
-	size  = PAGE_ALIGN(end + 1) - phys;
-	virt  = __sq_get_next_addr();
-	psz   = (size + (PAGE_SIZE - 1)) / PAGE_SIZE;
-	map   = __sq_alloc_mapping(virt, phys, size, name);
+	sq_mapping_list_add(map);
 
-	printk("sqremap: %15s  [%4d page%s]  va 0x%08lx   pa 0x%08lx\n",
-	       map->name ? map->name : "???",
-	       psz, psz == 1 ? " " : "s",
-	       map->sq_addr, map->addr);
+	return map->sq_addr;
 
-	return __sq_remap(map);
+out:
+	kmem_cache_free(sq_cache, map);
+	return ret;
 }
 
 /**
@@ -255,188 +202,198 @@ struct sq_mapping *sq_remap(unsigned long phys, unsigned int size, const char *n
  * sq_remap(). Also frees up the pte that was previously inserted into
  * the kernel page table and discards the UTLB translation.
  */
-void sq_unmap(struct sq_mapping *map)
+void sq_unmap(unsigned long vaddr)
 {
-	if (map->sq_addr > (unsigned long)high_memory)
-		vfree((void *)(map->sq_addr & PAGE_MASK));
+	struct sq_mapping **p, *map;
+	struct vm_struct *vma;
+	int page;
 
-	list_del(&map->list);
-	kfree(map);
-}
+	for (p = &sq_mapping_list; (map = *p); p = &map->next)
+		if (map->sq_addr == vaddr)
+			break;
 
-/**
- * sq_clear - Clear a store queue range
- * @addr: Address to start clearing from.
- * @len: Length to clear.
- *
- * A quick zero-fill implementation for clearing out memory that has been
- * remapped through the store queues.
- */
-void sq_clear(unsigned long addr, unsigned int len)
-{
-	int i;
+	if (unlikely(!map)) {
+		printk("%s: bad store queue address 0x%08lx\n",
+		       __FUNCTION__, vaddr);
+		return;
+	}
 
-	/* Clear out both queues linearly */
-	for (i = 0; i < 8; i++) {
-		ctrl_outl(0, addr + i + 0);
-		ctrl_outl(0, addr + i + 8);
+	page = (map->sq_addr - P4SEG_STORE_QUE) >> PAGE_SHIFT;
+	bitmap_release_region(sq_bitmap, page, get_order(map->size));
+
+#ifdef CONFIG_MMU
+	vma = remove_vm_area((void *)(map->sq_addr & PAGE_MASK));
+	if (!vma) {
+		printk(KERN_ERR "%s: bad address 0x%08lx\n",
+		       __FUNCTION__, map->sq_addr);
+		return;
 	}
+#endif
+
+	sq_mapping_list_del(map);
 
-	sq_flush_range(addr, len);
+	kmem_cache_free(sq_cache, map);
 }
 
-/**
- * sq_vma_unmap - Unmap a VMA range
- * @area: VMA containing range.
- * @addr: Start of range.
- * @len: Length of range.
+/*
+ * Needlessly complex sysfs interface. Unfortunately it doesn't seem like
+ * there is any other easy way to add things on a per-cpu basis without
+ * putting the directory entries somewhere stupid and having to create
+ * links in sysfs by hand back in to the per-cpu directories.
  *
- * Searches the sq_mapping_list for a mapping matching the sq addr @addr,
- * and subsequently frees up the entry. Further cleanup is done by generic
- * code.
+ * Some day we may want to have an additional abstraction per store
+ * queue, but considering the kobject hell we already have to deal with,
+ * it's simply not worth the trouble.
  */
-static void sq_vma_unmap(struct vm_area_struct *area,
-			 unsigned long addr, size_t len)
-{
-	struct list_head *pos, *tmp;
+static struct kobject *sq_kobject[NR_CPUS];
 
-	list_for_each_safe(pos, tmp, &sq_mapping_list) {
-		struct sq_mapping *entry;
+struct sq_sysfs_attr {
+	struct attribute attr;
+	ssize_t (*show)(char *buf);
+	ssize_t (*store)(const char *buf, size_t count);
+};
 
-		entry = list_entry(pos, typeof(*entry), list);
+#define to_sq_sysfs_attr(attr)	container_of(attr, struct sq_sysfs_attr, attr)
 
-		if (entry->sq_addr == addr) {
-			/*
-			 * We could probably get away without doing the tlb flush
-			 * here, as generic code should take care of most of this
-			 * when unmapping the rest of the VMA range for us. Leave
-			 * it in for added sanity for the time being..
-			 */
-			__flush_tlb_page(get_asid(), entry->sq_addr & PAGE_MASK);
+static ssize_t sq_sysfs_show(struct kobject *kobj, struct attribute *attr,
+			     char *buf)
+{
+	struct sq_sysfs_attr *sattr = to_sq_sysfs_attr(attr);
 
-			list_del(&entry->list);
-			kfree(entry);
+	if (likely(sattr->show))
+		return sattr->show(buf);
 
-			return;
-		}
-	}
+	return -EIO;
 }
 
-/**
- * sq_vma_sync - Sync a VMA range
- * @area: VMA containing range.
- * @start: Start of range.
- * @len: Length of range.
- * @flags: Additional flags.
- *
- * Synchronizes an sq mapped range by flushing the store queue cache for
- * the duration of the mapping.
- *
- * Used internally for user mappings, which must use msync() to prefetch
- * the store queue cache.
- */
-static int sq_vma_sync(struct vm_area_struct *area,
-		       unsigned long start, size_t len, unsigned int flags)
+static ssize_t sq_sysfs_store(struct kobject *kobj, struct attribute *attr,
+			      const char *buf, size_t count)
 {
-	sq_flush_range(start, len);
+	struct sq_sysfs_attr *sattr = to_sq_sysfs_attr(attr);
 
-	return 0;
+	if (likely(sattr->store))
+		return sattr->store(buf, count);
+
+	return -EIO;
 }
 
-static struct vm_operations_struct sq_vma_ops = {
-	.unmap	= sq_vma_unmap,
-	.sync	= sq_vma_sync,
-};
+static ssize_t mapping_show(char *buf)
+{
+	struct sq_mapping **list, *entry;
+	char *p = buf;
 
-/**
- * sq_mmap - mmap() for /dev/cpu/sq
- * @file: unused.
- * @vma: VMA to remap.
- *
- * Remap the specified vma @vma through the store queues, and setup associated
- * information for the new mapping. Also build up the page tables for the new
- * area.
- */
-static int sq_mmap(struct file *file, struct vm_area_struct *vma)
+	for (list = &sq_mapping_list; (entry = *list); list = &entry->next)
+		p += sprintf(p, "%08lx-%08lx [%08lx]: %s\n",
+			     entry->sq_addr, entry->sq_addr + entry->size,
+			     entry->addr, entry->name);
+
+	return p - buf;
+}
+
+static ssize_t mapping_store(const char *buf, size_t count)
 {
-	unsigned long offset = vma->vm_pgoff << PAGE_SHIFT;
-	unsigned long size = vma->vm_end - vma->vm_start;
-	struct sq_mapping *map;
+	unsigned long base = 0, len = 0;
 
-	/*
-	 * We're not interested in any arbitrary virtual address that has
-	 * been stuck in the VMA, as we already know what addresses we
-	 * want. Save off the size, and reposition the VMA to begin at
-	 * the next available sq address.
-	 */
-	vma->vm_start = __sq_get_next_addr();
-	vma->vm_end   = vma->vm_start + size;
+	sscanf(buf, "%lx %lx", &base, &len);
+	if (!base)
+		return -EIO;
 
-	vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
+	if (likely(len)) {
+		int ret = sq_remap(base, len, "Userspace",
+				   pgprot_val(PAGE_SHARED));
+		if (ret < 0)
+			return ret;
+	} else
+		sq_unmap(base);
 
-	vma->vm_flags |= VM_IO | VM_RESERVED;
+	return count;
+}
 
-	map = __sq_alloc_mapping(vma->vm_start, offset, size, "Userspace");
+static struct sq_sysfs_attr mapping_attr =
+	__ATTR(mapping, 0644, mapping_show, mapping_store);
 
-	if (io_remap_pfn_range(vma, map->sq_addr, map->addr >> PAGE_SHIFT,
-				size, vma->vm_page_prot))
-		return -EAGAIN;
+static struct attribute *sq_sysfs_attrs[] = {
+	&mapping_attr.attr,
+	NULL,
+};
 
-	vma->vm_ops = &sq_vma_ops;
+static struct sysfs_ops sq_sysfs_ops = {
+	.show	= sq_sysfs_show,
+	.store	= sq_sysfs_store,
+};
 
-	return 0;
-}
+static struct kobj_type ktype_percpu_entry = {
+	.sysfs_ops	= &sq_sysfs_ops,
+	.default_attrs	= sq_sysfs_attrs,
+};
 
-#ifdef CONFIG_PROC_FS
-static int sq_mapping_read_proc(char *buf, char **start, off_t off,
-				int len, int *eof, void *data)
+static int __devinit sq_sysdev_add(struct sys_device *sysdev)
 {
-	struct list_head *pos;
-	char *p = buf;
+	unsigned int cpu = sysdev->id;
+	struct kobject *kobj;
 
-	list_for_each_prev(pos, &sq_mapping_list) {
-		struct sq_mapping *entry;
+	sq_kobject[cpu] = kzalloc(sizeof(struct kobject), GFP_KERNEL);
+	if (unlikely(!sq_kobject[cpu]))
+		return -ENOMEM;
 
-		entry = list_entry(pos, typeof(*entry), list);
+	kobj = sq_kobject[cpu];
+	kobj->parent = &sysdev->kobj;
+	kobject_set_name(kobj, "%s", "sq");
+	kobj->ktype = &ktype_percpu_entry;
 
-		p += sprintf(p, "%08lx-%08lx [%08lx]: %s\n", entry->sq_addr,
-			     entry->sq_addr + entry->size - 1, entry->addr,
-			     entry->name);
-	}
-
-	return p - buf;
+	return kobject_register(kobj);
 }
-#endif
 
-static struct file_operations sq_fops = {
-	.owner		= THIS_MODULE,
-	.mmap		= sq_mmap,
-};
+static int __devexit sq_sysdev_remove(struct sys_device *sysdev)
+{
+	unsigned int cpu = sysdev->id;
+	struct kobject *kobj = sq_kobject[cpu];
 
-static struct miscdevice sq_dev = {
-	.minor		= STORE_QUEUE_MINOR,
-	.name		= "sq",
-	.fops		= &sq_fops,
+	kobject_unregister(kobj);
+	return 0;
+}
+
+static struct sysdev_driver sq_sysdev_driver = {
+	.add		= sq_sysdev_add,
+	.remove		= __devexit_p(sq_sysdev_remove),
 };
 
 static int __init sq_api_init(void)
 {
-	int ret;
+	unsigned int nr_pages = 0x04000000 >> PAGE_SHIFT;
+	unsigned int size = (nr_pages + (BITS_PER_LONG - 1)) / BITS_PER_LONG;
+	int ret = -ENOMEM;
+
 	printk(KERN_NOTICE "sq: Registering store queue API.\n");
 
-	create_proc_read_entry("sq_mapping", 0, 0, sq_mapping_read_proc, 0);
+	sq_cache = kmem_cache_create("store_queue_cache",
+				sizeof(struct sq_mapping), 0, 0,
+				NULL, NULL);
+	if (unlikely(!sq_cache))
+		return ret;
 
-	ret = misc_register(&sq_dev);
-	if (ret)
-		remove_proc_entry("sq_mapping", NULL);
+	sq_bitmap = kzalloc(size, GFP_KERNEL);
+	if (unlikely(!sq_bitmap))
+		goto out;
+
+	ret = sysdev_driver_register(&cpu_sysdev_class, &sq_sysdev_driver);
+	if (unlikely(ret != 0))
+		goto out;
+
+	return 0;
+
+out:
+	kfree(sq_bitmap);
+	kmem_cache_destroy(sq_cache);
 
 	return ret;
 }
 
 static void __exit sq_api_exit(void)
 {
-	misc_deregister(&sq_dev);
-	remove_proc_entry("sq_mapping", NULL);
+	sysdev_driver_unregister(&cpu_sysdev_class, &sq_sysdev_driver);
+	kfree(sq_bitmap);
+	kmem_cache_destroy(sq_cache);
 }
 
 module_init(sq_api_init);
@@ -445,11 +402,7 @@ module_exit(sq_api_exit);
 MODULE_AUTHOR("Paul Mundt <lethal@linux-sh.org>, M. R. Brown <mrbrown@0xd6.org>");
 MODULE_DESCRIPTION("Simple API for SH-4 integrated Store Queues");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(STORE_QUEUE_MINOR);
 
 EXPORT_SYMBOL(sq_remap);
 EXPORT_SYMBOL(sq_unmap);
-EXPORT_SYMBOL(sq_clear);
-EXPORT_SYMBOL(sq_flush);
 EXPORT_SYMBOL(sq_flush_range);
-
diff --git a/arch/sh/kernel/early_printk.c b/arch/sh/kernel/early_printk.c
index 1378db375e17..a00022722e9e 100644
--- a/arch/sh/kernel/early_printk.c
+++ b/arch/sh/kernel/early_printk.c
@@ -3,7 +3,7 @@
  *
  *  Copyright (C) 1999, 2000  Niibe Yutaka
  *  Copyright (C) 2002  M. R. Brown
- *  Copyright (C) 2004  Paul Mundt
+ *  Copyright (C) 2004 - 2006  Paul Mundt
  *
  * This file is subject to the terms and conditions of the GNU General Public
  * License.  See the file "COPYING" in the main directory of this archive
@@ -49,7 +49,7 @@ static int __init sh_console_setup(struct console *co, char *options)
 	return 0;
 }
 
-static struct console early_console = {
+static struct console bios_console = {
 	.name		= "bios",
 	.write		= sh_console_write,
 	.setup		= sh_console_setup,
@@ -59,34 +59,43 @@ static struct console early_console = {
 #endif
 
 #ifdef CONFIG_EARLY_SCIF_CONSOLE
+#include <linux/serial_core.h>
+#include "../../../drivers/serial/sh-sci.h"
+
+#ifdef CONFIG_CPU_SH4
 #define SCIF_REG	0xffe80000
+#elif defined(CONFIG_CPU_SUBTYPE_SH72060)
+#define SCIF_REG	0xfffe9800
+#else
+#error "Undefined SCIF for this subtype"
+#endif
+
+static struct uart_port scif_port = {
+	.mapbase	= SCIF_REG,
+	.membase	= (char __iomem *)SCIF_REG,
+};
 
 static void scif_sercon_putc(int c)
 {
-	while (!(ctrl_inw(SCIF_REG + 0x10) & 0x20)) ;
+	while (((sci_in(&scif_port, SCFDR) & 0x1f00 >> 8) == 16))
+		;
 
-	ctrl_outb(c, SCIF_REG + 12);
-	ctrl_outw((ctrl_inw(SCIF_REG + 0x10) & 0x9f), SCIF_REG + 0x10);
+	sci_out(&scif_port, SCxTDR, c);
+	sci_in(&scif_port, SCxSR);
+	sci_out(&scif_port, SCxSR, 0xf3 & ~(0x20 | 0x40));
+
+	while ((sci_in(&scif_port, SCxSR) & 0x40) == 0);
+		;
 
 	if (c == '\n')
 		scif_sercon_putc('\r');
 }
 
-static void scif_sercon_flush(void)
-{
-	ctrl_outw((ctrl_inw(SCIF_REG + 0x10) & 0xbf), SCIF_REG + 0x10);
-
-	while (!(ctrl_inw(SCIF_REG + 0x10) & 0x40)) ;
-
-	ctrl_outw((ctrl_inw(SCIF_REG + 0x10) & 0xbf), SCIF_REG + 0x10);
-}
-
-static void scif_sercon_write(struct console *con, const char *s, unsigned count)
+static void scif_sercon_write(struct console *con, const char *s,
+			      unsigned count)
 {
 	while (count-- > 0)
 		scif_sercon_putc(*s++);
-
-	scif_sercon_flush();
 }
 
 static int __init scif_sercon_setup(struct console *con, char *options)
@@ -96,7 +105,7 @@ static int __init scif_sercon_setup(struct console *con, char *options)
 	return 0;
 }
 
-static struct console early_console = {
+static struct console scif_console = {
 	.name		= "sercon",
 	.write		= scif_sercon_write,
 	.setup		= scif_sercon_setup,
@@ -104,7 +113,7 @@ static struct console early_console = {
 	.index		= -1,
 };
 
-void scif_sercon_init(int baud)
+static void scif_sercon_init(int baud)
 {
 	ctrl_outw(0, SCIF_REG + 8);
 	ctrl_outw(0, SCIF_REG);
@@ -122,16 +131,61 @@ void scif_sercon_init(int baud)
 }
 #endif
 
-void __init enable_early_printk(void)
+/*
+ * Setup a default console, if more than one is compiled in, rely on the
+ * earlyprintk= parsing to give priority.
+ */
+static struct console *early_console =
+#ifdef CONFIG_SH_STANDARD_BIOS
+	&bios_console
+#elif defined(CONFIG_EARLY_SCIF_CONSOLE)
+	&scif_console
+#else
+	NULL
+#endif
+	;
+
+static int __initdata keep_early;
+
+int __init setup_early_printk(char *opt)
 {
-#ifdef CONFIG_EARLY_SCIF_CONSOLE
-	scif_sercon_init(115200);
+	char *space;
+	char buf[256];
+
+	strlcpy(buf, opt, sizeof(buf));
+	space = strchr(buf, ' ');
+	if (space)
+		*space = 0;
+
+	if (strstr(buf, "keep"))
+		keep_early = 1;
+
+#ifdef CONFIG_SH_STANDARD_BIOS
+	if (!strncmp(buf, "bios", 4))
+		early_console = &bios_console;
+#endif
+#if defined(CONFIG_EARLY_SCIF_CONSOLE)
+	if (!strncmp(buf, "serial", 6)) {
+		early_console = &scif_console;
+
+#ifdef CONFIG_CPU_SH4
+		scif_sercon_init(115200);
+#endif
+	}
 #endif
-	register_console(&early_console);
+
+	if (likely(early_console))
+		register_console(early_console);
+
+	return 1;
 }
+__setup("earlyprintk=", setup_early_printk);
 
-void disable_early_printk(void)
+void __init disable_early_printk(void)
 {
-	unregister_console(&early_console);
+	if (!keep_early) {
+		printk("disabling early console\n");
+		unregister_console(early_console);
+	} else
+		printk("keeping early console\n");
 }
-
diff --git a/arch/sh/kernel/entry.S b/arch/sh/kernel/entry.S
index 7dfd2ba75f7f..fe8221855b28 100644
--- a/arch/sh/kernel/entry.S
+++ b/arch/sh/kernel/entry.S
@@ -18,24 +18,6 @@
 #include <asm/cpu/mmu_context.h>
 #include <asm/unistd.h>
 
-#if !defined(CONFIG_NFSD) && !defined(CONFIG_NFSD_MODULE)
-#define sys_nfsservctl		sys_ni_syscall
-#endif
-
-#if !defined(CONFIG_MMU)
-#define sys_madvise		sys_ni_syscall
-#define sys_readahead		sys_ni_syscall
-#define sys_mprotect		sys_ni_syscall
-#define sys_msync		sys_ni_syscall
-#define sys_mlock		sys_ni_syscall
-#define sys_munlock		sys_ni_syscall
-#define sys_mlockall		sys_ni_syscall
-#define sys_munlockall		sys_ni_syscall
-#define sys_mremap		sys_ni_syscall
-#define sys_mincore		sys_ni_syscall
-#define sys_remap_file_pages	sys_ni_syscall
-#endif
-
 ! NOTE:
 ! GNU as (as of 2.9.1) changes bf/s into bt/s and bra, when the address
 ! to be jumped is too far, but it causes illegal slot exception.
@@ -326,7 +308,7 @@ ENTRY(exception_error)
 	.align	2
 ret_from_exception:
 	preempt_stop()
-ret_from_irq:
+ENTRY(ret_from_irq)
 	!
 	mov	#OFF_SR, r0
 	mov.l	@(r0,r15), r0	! get status register
@@ -389,11 +371,12 @@ work_pending:
 	! r8: current_thread_info
 	! t:  result of "tst	#_TIF_NEED_RESCHED, r0"
 	bf/s	work_resched
-	 tst	#_TIF_SIGPENDING, r0
+	 tst	#(_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK), r0
 work_notifysig:
 	bt/s	restore_all
 	 mov	r15, r4
-	mov	#0, r5
+	mov	r12, r5		! set arg1(save_r0)
+	mov	r0, r6
 	mov.l	2f, r1
 	mova	restore_all, r0
 	jmp	@r1
@@ -431,7 +414,7 @@ work_resched:
 
 	.align	2
 1:	.long	schedule
-2:	.long	do_signal
+2:	.long	do_notify_resume
 
 	.align	2
 syscall_exit_work:
@@ -552,6 +535,7 @@ syscall_call:
 	mov.l	@r9, r8
 	jsr	@r8	    	! jump to specific syscall handler
 	 nop
+	mov.l	@(OFF_R0,r15), r12		! save r0
 	mov.l	r0, @(OFF_R0,r15)		! save the return value
 	!
 syscall_exit:
@@ -644,7 +628,7 @@ skip_restore:
 	!
 #if defined(CONFIG_KGDB_NMI)
 	! Clear in_nmi
-	mov.l	4f, k0
+	mov.l	6f, k0
 	mov	#0, k1
 	mov.b	k1, @k0
 #endif
@@ -722,7 +706,7 @@ interrupt:
 !
 !
 	.align	2
-handle_exception:
+ENTRY(handle_exception)
 	! Using k0, k1 for scratch registers (r0_bank1, r1_bank),
 	! save all registers onto stack.
 	!
@@ -732,8 +716,8 @@ handle_exception:
 	bt/s	1f		! It's a kernel to kernel transition.
 	 mov	r15, k0		! save original stack to k0
 	/* User space to kernel */
-	mov	#0x20, k1
-	shll8	k1		! k1 := 8192 (== THREAD_SIZE)
+	mov	#(THREAD_SIZE >> 8), k1
+	shll8	k1		! k1 := THREAD_SIZE
 	add	current, k1
 	mov	k1, r15		! change to kernel stack
 	!
@@ -838,300 +822,3 @@ ENTRY(exception_none)
 	rts
 	 nop
 
-	.data
-ENTRY(sys_call_table)
-	.long sys_ni_syscall	/* 0  -  old "setup()" system call*/
-	.long sys_exit
-	.long sys_fork
-	.long sys_read
-	.long sys_write
-	.long sys_open		/* 5 */
-	.long sys_close
-	.long sys_waitpid
-	.long sys_creat
-	.long sys_link
-	.long sys_unlink		/* 10 */
-	.long sys_execve
-	.long sys_chdir
-	.long sys_time
-	.long sys_mknod
-	.long sys_chmod		/* 15 */
-	.long sys_lchown16
-	.long sys_ni_syscall	/* old break syscall holder */
-	.long sys_stat
-	.long sys_lseek
-	.long sys_getpid		/* 20 */
-	.long sys_mount
-	.long sys_oldumount
-	.long sys_setuid16
-	.long sys_getuid16
-	.long sys_stime		/* 25 */
-	.long sys_ptrace
-	.long sys_alarm
-	.long sys_fstat
-	.long sys_pause
-	.long sys_utime		/* 30 */
-	.long sys_ni_syscall	/* old stty syscall holder */
-	.long sys_ni_syscall	/* old gtty syscall holder */
-	.long sys_access
-	.long sys_nice
-	.long sys_ni_syscall	/* 35 */		/* old ftime syscall holder */
-	.long sys_sync
-	.long sys_kill
-	.long sys_rename
-	.long sys_mkdir
-	.long sys_rmdir		/* 40 */
-	.long sys_dup
-	.long sys_pipe
-	.long sys_times
-	.long sys_ni_syscall	/* old prof syscall holder */
-	.long sys_brk		/* 45 */
-	.long sys_setgid16
-	.long sys_getgid16
-	.long sys_signal
-	.long sys_geteuid16
-	.long sys_getegid16	/* 50 */
-	.long sys_acct
-	.long sys_umount		/* recycled never used phys() */
-	.long sys_ni_syscall	/* old lock syscall holder */
-	.long sys_ioctl
-	.long sys_fcntl		/* 55 */
-	.long sys_ni_syscall	/* old mpx syscall holder */
-	.long sys_setpgid
-	.long sys_ni_syscall	/* old ulimit syscall holder */
-	.long sys_ni_syscall	/* sys_olduname */
-	.long sys_umask		/* 60 */
-	.long sys_chroot
-	.long sys_ustat
-	.long sys_dup2
-	.long sys_getppid
-	.long sys_getpgrp		/* 65 */
-	.long sys_setsid
-	.long sys_sigaction
-	.long sys_sgetmask
-	.long sys_ssetmask
-	.long sys_setreuid16	/* 70 */
-	.long sys_setregid16
-	.long sys_sigsuspend
-	.long sys_sigpending
-	.long sys_sethostname
-	.long sys_setrlimit	/* 75 */
-	.long sys_old_getrlimit
-	.long sys_getrusage
-	.long sys_gettimeofday
-	.long sys_settimeofday
-	.long sys_getgroups16	/* 80 */
-	.long sys_setgroups16
-	.long sys_ni_syscall	/* sys_oldselect */
-	.long sys_symlink
-	.long sys_lstat
-	.long sys_readlink		/* 85 */
-	.long sys_uselib
-	.long sys_swapon
-	.long sys_reboot
-	.long old_readdir
-	.long old_mmap		/* 90 */
-	.long sys_munmap
-	.long sys_truncate
-	.long sys_ftruncate
-	.long sys_fchmod
-	.long sys_fchown16		/* 95 */
-	.long sys_getpriority
-	.long sys_setpriority
-	.long sys_ni_syscall	/* old profil syscall holder */
-	.long sys_statfs
-	.long sys_fstatfs		/* 100 */
-	.long sys_ni_syscall	/* ioperm */
-	.long sys_socketcall
-	.long sys_syslog
-	.long sys_setitimer
-	.long sys_getitimer	/* 105 */
-	.long sys_newstat
-	.long sys_newlstat
-	.long sys_newfstat
-	.long sys_uname
-	.long sys_ni_syscall	/* 110 */ /* iopl */
-	.long sys_vhangup
-	.long sys_ni_syscall	/* idle */
-	.long sys_ni_syscall	/* vm86old */
-	.long sys_wait4
-	.long sys_swapoff		/* 115 */
-	.long sys_sysinfo
-	.long sys_ipc
-	.long sys_fsync
-	.long sys_sigreturn
-	.long sys_clone		/* 120 */
-	.long sys_setdomainname
-	.long sys_newuname
-	.long sys_ni_syscall	/* sys_modify_ldt */
-	.long sys_adjtimex
-	.long sys_mprotect		/* 125 */
-	.long sys_sigprocmask
-	.long sys_ni_syscall	/* old "create_module" */
-	.long sys_init_module
-	.long sys_delete_module
-	.long sys_ni_syscall	/* 130: old "get_kernel_syms" */
-	.long sys_quotactl
-	.long sys_getpgid
-	.long sys_fchdir
-	.long sys_bdflush
-	.long sys_sysfs		/* 135 */
-	.long sys_personality
-	.long sys_ni_syscall	/* for afs_syscall */
-	.long sys_setfsuid16
-	.long sys_setfsgid16
-	.long sys_llseek		/* 140 */
-	.long sys_getdents
-	.long sys_select
-	.long sys_flock
-	.long sys_msync
-	.long sys_readv		/* 145 */
-	.long sys_writev
-	.long sys_getsid
-	.long sys_fdatasync
-	.long sys_sysctl
-	.long sys_mlock		/* 150 */
-	.long sys_munlock
-	.long sys_mlockall
-	.long sys_munlockall
-	.long sys_sched_setparam
-	.long sys_sched_getparam   /* 155 */
-	.long sys_sched_setscheduler
-	.long sys_sched_getscheduler
-	.long sys_sched_yield
-	.long sys_sched_get_priority_max
-	.long sys_sched_get_priority_min  /* 160 */
-	.long sys_sched_rr_get_interval
-	.long sys_nanosleep
-	.long sys_mremap
-	.long sys_setresuid16
-	.long sys_getresuid16	/* 165 */
-	.long sys_ni_syscall	/* vm86 */
-	.long sys_ni_syscall	/* old "query_module" */
-	.long sys_poll
-	.long sys_nfsservctl
-	.long sys_setresgid16	/* 170 */
-	.long sys_getresgid16
-	.long sys_prctl
-	.long sys_rt_sigreturn
-	.long sys_rt_sigaction
-	.long sys_rt_sigprocmask	/* 175 */
-	.long sys_rt_sigpending
-	.long sys_rt_sigtimedwait
-	.long sys_rt_sigqueueinfo
-	.long sys_rt_sigsuspend
-	.long sys_pread_wrapper	   /* 180 */
-	.long sys_pwrite_wrapper
-	.long sys_chown16
-	.long sys_getcwd
-	.long sys_capget
-	.long sys_capset           /* 185 */
-	.long sys_sigaltstack
-	.long sys_sendfile
-	.long sys_ni_syscall	/* streams1 */
-	.long sys_ni_syscall	/* streams2 */
-	.long sys_vfork            /* 190 */
-	.long sys_getrlimit
-	.long sys_mmap2
-	.long sys_truncate64
-	.long sys_ftruncate64
-	.long sys_stat64		/* 195 */
-	.long sys_lstat64
-	.long sys_fstat64
-	.long sys_lchown
-	.long sys_getuid
-	.long sys_getgid		/* 200 */
-	.long sys_geteuid
-	.long sys_getegid
-	.long sys_setreuid
-	.long sys_setregid
-	.long sys_getgroups	/* 205 */
-	.long sys_setgroups
-	.long sys_fchown
-	.long sys_setresuid
-	.long sys_getresuid
-	.long sys_setresgid	/* 210 */
-	.long sys_getresgid
-	.long sys_chown
-	.long sys_setuid
-	.long sys_setgid
-	.long sys_setfsuid		/* 215 */
-	.long sys_setfsgid
-	.long sys_pivot_root
-	.long sys_mincore
-	.long sys_madvise
-	.long sys_getdents64	/* 220 */
-	.long sys_fcntl64
-	.long sys_ni_syscall	/* reserved for TUX */
-	.long sys_ni_syscall	/* Reserved for Security */
-	.long sys_gettid
-	.long sys_readahead	/* 225 */
-	.long sys_setxattr
-	.long sys_lsetxattr
-	.long sys_fsetxattr
-	.long sys_getxattr
-	.long sys_lgetxattr	/* 230 */
-	.long sys_fgetxattr
-	.long sys_listxattr
-	.long sys_llistxattr
-	.long sys_flistxattr
-	.long sys_removexattr	/* 235 */
-	.long sys_lremovexattr
-	.long sys_fremovexattr
-	.long sys_tkill
-	.long sys_sendfile64
-	.long sys_futex		/* 240 */
-	.long sys_sched_setaffinity
-	.long sys_sched_getaffinity
-	.long sys_ni_syscall
-	.long sys_ni_syscall
-	.long sys_io_setup	/* 245 */
-	.long sys_io_destroy
-	.long sys_io_getevents
-	.long sys_io_submit
-	.long sys_io_cancel
-	.long sys_fadvise64	/* 250 */
-	.long sys_ni_syscall
-	.long sys_exit_group
-	.long sys_lookup_dcookie
-	.long sys_epoll_create
-	.long sys_epoll_ctl	/* 255 */
-	.long sys_epoll_wait
- 	.long sys_remap_file_pages
- 	.long sys_set_tid_address
- 	.long sys_timer_create
- 	.long sys_timer_settime		/* 260 */
- 	.long sys_timer_gettime
- 	.long sys_timer_getoverrun
- 	.long sys_timer_delete
- 	.long sys_clock_settime
- 	.long sys_clock_gettime		/* 265 */
- 	.long sys_clock_getres
- 	.long sys_clock_nanosleep
-	.long sys_statfs64
-	.long sys_fstatfs64     
-	.long sys_tgkill		/* 270 */
-	.long sys_utimes
- 	.long sys_fadvise64_64_wrapper
-	.long sys_ni_syscall	/* Reserved for vserver */
-	.long sys_ni_syscall	/* Reserved for mbind */
-	.long sys_ni_syscall	/* 275 - get_mempolicy */
-	.long sys_ni_syscall	/* set_mempolicy */
-	.long sys_mq_open
-	.long sys_mq_unlink
-	.long sys_mq_timedsend
-	.long sys_mq_timedreceive       /* 280 */
-	.long sys_mq_notify
-	.long sys_mq_getsetattr
-	.long sys_ni_syscall	/* Reserved for kexec */
-	.long sys_waitid
-	.long sys_add_key		/* 285 */
-	.long sys_request_key
-	.long sys_keyctl
-	.long sys_ioprio_set
-	.long sys_ioprio_get
-	.long sys_inotify_init		/* 290 */
-	.long sys_inotify_add_watch
-	.long sys_inotify_rm_watch
-
-/* End of entry.S */
diff --git a/arch/sh/kernel/head.S b/arch/sh/kernel/head.S
index 9b9e6ef626ce..f5f53d14f245 100644
--- a/arch/sh/kernel/head.S
+++ b/arch/sh/kernel/head.S
@@ -11,6 +11,18 @@
  * Head.S contains the SH exception handlers and startup code.
  */
 #include <linux/linkage.h>
+#include <asm/thread_info.h>
+
+#ifdef CONFIG_CPU_SH4A
+#define SYNCO()		synco
+
+#define PREFI(label, reg)	\
+	mov.l	label, reg;	\
+	prefi	@reg
+#else
+#define SYNCO()
+#define PREFI(label, reg)
+#endif
 
 	.section	.empty_zero_page, "aw"
 ENTRY(empty_zero_page)
@@ -42,18 +54,25 @@ ENTRY(_stext)
 	!			Initialize global interrupt mask
 	mov	#0, r0
 	ldc	r0, r6_bank
+
+	/*
+	 * Prefetch if possible to reduce cache miss penalty.
+	 *
+	 * We do this early on for SH-4A as a micro-optimization,
+	 * as later on we will have speculative execution enabled
+	 * and this will become less of an issue.
+	 */
+	PREFI(5f, r0)
+	PREFI(6f, r0)
+
 	!
 	mov.l	2f, r0
 	mov	r0, r15		! Set initial r15 (stack pointer)
-	mov	#0x20, r1	!
-	shll8	r1		! r1 = 8192
+	mov	#(THREAD_SIZE >> 8), r1
+	shll8	r1		! r1 = THREAD_SIZE
 	sub	r1, r0		!
 	ldc	r0, r7_bank	! ... and initial thread_info
-	!
-	!			Additional CPU initialization
-	mov.l	6f, r0
-	jsr	@r0
-	 nop
+
 	!			Clear BSS area
 	mov.l	3f, r1
 	add	#4, r1
@@ -62,6 +81,14 @@ ENTRY(_stext)
 9:	cmp/hs	r2, r1
 	bf/s	9b		! while (r1 < r2)
 	 mov.l	r0,@-r2
+
+	!			Additional CPU initialization
+	mov.l	6f, r0
+	jsr	@r0
+	 nop
+
+	SYNCO()			! Wait for pending instructions..
+
 	!			Start kernel
 	mov.l	5f, r0
 	jmp	@r0
@@ -69,7 +96,7 @@ ENTRY(_stext)
 
 	.balign 4
 1:	.long	0x400080F0		! MD=1, RB=0, BL=0, FD=1, IMASK=0xF
-2:	.long	stack
+2:	.long	init_thread_union+THREAD_SIZE
 3:	.long	__bss_start
 4:	.long	_end
 5:	.long	start_kernel
diff --git a/arch/sh/kernel/io.c b/arch/sh/kernel/io.c
index 71c9fde2fd90..501fe03e3715 100644
--- a/arch/sh/kernel/io.c
+++ b/arch/sh/kernel/io.c
@@ -61,6 +61,73 @@ void memset_io(volatile void __iomem *dst, int c, unsigned long count)
 }
 EXPORT_SYMBOL(memset_io);
 
+void __raw_readsl(unsigned long addr, void *datap, int len)
+{
+	u32 *data;
+
+	for (data = datap; (len != 0) && (((u32)data & 0x1f) != 0); len--)
+		*data++ = ctrl_inl(addr);
+
+	if (likely(len >= (0x20 >> 2))) {
+		int tmp2, tmp3, tmp4, tmp5, tmp6;
+
+		__asm__ __volatile__(
+			"1:			\n\t"
+			"mov.l	@%7, r0		\n\t"
+			"mov.l	@%7, %2		\n\t"
+#ifdef CONFIG_CPU_SH4
+			"movca.l r0, @%0	\n\t"
+#else
+			"mov.l	r0, @%0		\n\t"
+#endif
+			"mov.l	@%7, %3		\n\t"
+			"mov.l	@%7, %4		\n\t"
+			"mov.l	@%7, %5		\n\t"
+			"mov.l	@%7, %6		\n\t"
+			"mov.l	@%7, r7		\n\t"
+			"mov.l	@%7, r0		\n\t"
+			"mov.l	%2, @(0x04,%0)	\n\t"
+			"mov	#0x20>>2, %2	\n\t"
+			"mov.l	%3, @(0x08,%0)	\n\t"
+			"sub	%2, %1		\n\t"
+			"mov.l	%4, @(0x0c,%0)	\n\t"
+			"cmp/hi	%1, %2		! T if 32 > len	\n\t"
+			"mov.l	%5, @(0x10,%0)	\n\t"
+			"mov.l	%6, @(0x14,%0)	\n\t"
+			"mov.l	r7, @(0x18,%0)	\n\t"
+			"mov.l	r0, @(0x1c,%0)	\n\t"
+			"bf.s	1b		\n\t"
+			" add	#0x20, %0	\n\t"
+			: "=&r" (data), "=&r" (len),
+			  "=&r" (tmp2), "=&r" (tmp3), "=&r" (tmp4),
+			  "=&r" (tmp5), "=&r" (tmp6)
+			: "r"(addr), "0" (data), "1" (len)
+			: "r0", "r7", "t", "memory");
+	}
+
+	for (; len != 0; len--)
+		*data++ = ctrl_inl(addr);
+}
+EXPORT_SYMBOL(__raw_readsl);
+
+void __raw_writesl(unsigned long addr, const void *data, int len)
+{
+	if (likely(len != 0)) {
+		int tmp1;
+
+		__asm__ __volatile__ (
+			"1:				\n\t"
+			"mov.l	@%0+, %1	\n\t"
+			"dt		%3		\n\t"
+			"bf.s		1b		\n\t"
+			" mov.l	%1, @%4		\n\t"
+			: "=&r" (data), "=&r" (tmp1)
+			: "0" (data), "r" (len), "r"(addr)
+			: "t", "memory");
+	}
+}
+EXPORT_SYMBOL(__raw_writesl);
+
 void __iomem *ioport_map(unsigned long port, unsigned int nr)
 {
 	return sh_mv.mv_ioport_map(port, nr);
diff --git a/arch/sh/kernel/irq.c b/arch/sh/kernel/irq.c
index c2e07f7f3496..c7ebd6aec951 100644
--- a/arch/sh/kernel/irq.c
+++ b/arch/sh/kernel/irq.c
@@ -1,5 +1,4 @@
-/* $Id: irq.c,v 1.20 2004/01/13 05:52:11 kkojima Exp $
- *
+/*
  * linux/arch/sh/kernel/irq.c
  *
  *	Copyright (C) 1992, 1998 Linus Torvalds, Ingo Molnar
@@ -7,13 +6,15 @@
  *
  * SuperH version:  Copyright (C) 1999  Niibe Yutaka
  */
-
 #include <linux/irq.h>
 #include <linux/interrupt.h>
+#include <linux/module.h>
 #include <linux/kernel_stat.h>
 #include <linux/seq_file.h>
 #include <asm/irq.h>
 #include <asm/processor.h>
+#include <asm/uaccess.h>
+#include <asm/thread_info.h>
 #include <asm/cpu/mmu_context.h>
 
 /*
@@ -60,15 +61,46 @@ unlock:
 }
 #endif
 
+#ifdef CONFIG_4KSTACKS
+/*
+ * per-CPU IRQ handling contexts (thread information and stack)
+ */
+union irq_ctx {
+	struct thread_info	tinfo;
+	u32			stack[THREAD_SIZE/sizeof(u32)];
+};
+
+static union irq_ctx *hardirq_ctx[NR_CPUS];
+static union irq_ctx *softirq_ctx[NR_CPUS];
+#endif
 
 asmlinkage int do_IRQ(unsigned long r4, unsigned long r5,
 		      unsigned long r6, unsigned long r7,
 		      struct pt_regs regs)
 {
 	int irq = r4;
+#ifdef CONFIG_4KSTACKS
+	union irq_ctx *curctx, *irqctx;
+#endif
 
 	irq_enter();
 
+#ifdef CONFIG_DEBUG_STACKOVERFLOW
+	/* Debugging check for stack overflow: is there less than 1KB free? */
+	{
+		long sp;
+
+		__asm__ __volatile__ ("and r15, %0" :
+					"=r" (sp) : "0" (THREAD_SIZE - 1));
+
+		if (unlikely(sp < (sizeof(struct thread_info) + STACK_WARN))) {
+			printk("do_IRQ: stack overflow: %ld\n",
+			       sp - sizeof(struct thread_info));
+			dump_stack();
+		}
+	}
+#endif
+
 #ifdef CONFIG_CPU_HAS_INTEVT
 	__asm__ __volatile__ (
 #ifdef CONFIG_CPU_HAS_SR_RB
@@ -87,7 +119,135 @@ asmlinkage int do_IRQ(unsigned long r4, unsigned long r5,
 #endif
 
 	irq = irq_demux(irq);
-	__do_IRQ(irq, &regs);
+
+#ifdef CONFIG_4KSTACKS
+	curctx = (union irq_ctx *)current_thread_info();
+	irqctx = hardirq_ctx[smp_processor_id()];
+
+	/*
+	 * this is where we switch to the IRQ stack. However, if we are
+	 * already using the IRQ stack (because we interrupted a hardirq
+	 * handler) we can't do that and just have to keep using the
+	 * current stack (which is the irq stack already after all)
+	 */
+	if (curctx != irqctx) {
+		u32 *isp;
+
+		isp = (u32 *)((char *)irqctx + sizeof(*irqctx));
+		irqctx->tinfo.task = curctx->tinfo.task;
+		irqctx->tinfo.previous_sp = current_stack_pointer;
+
+		__asm__ __volatile__ (
+			"mov	%0, r4		\n"
+			"mov	%1, r5		\n"
+			"mov	r15, r9		\n"
+			"jsr	@%2		\n"
+			/* swith to the irq stack */
+			" mov	%3, r15		\n"
+			/* restore the stack (ring zero) */
+			"mov	r9, r15		\n"
+			: /* no outputs */
+			: "r" (irq), "r" (&regs), "r" (__do_IRQ), "r" (isp)
+			/* XXX: A somewhat excessive clobber list? -PFM */
+			: "memory", "r0", "r1", "r2", "r3", "r4",
+			  "r5", "r6", "r7", "r8", "t", "pr"
+		);
+	} else
+#endif
+		__do_IRQ(irq, &regs);
+
 	irq_exit();
+
 	return 1;
 }
+
+#ifdef CONFIG_4KSTACKS
+/*
+ * These should really be __section__(".bss.page_aligned") as well, but
+ * gcc's 3.0 and earlier don't handle that correctly.
+ */
+static char softirq_stack[NR_CPUS * THREAD_SIZE]
+		__attribute__((__aligned__(THREAD_SIZE)));
+
+static char hardirq_stack[NR_CPUS * THREAD_SIZE]
+		__attribute__((__aligned__(THREAD_SIZE)));
+
+/*
+ * allocate per-cpu stacks for hardirq and for softirq processing
+ */
+void irq_ctx_init(int cpu)
+{
+	union irq_ctx *irqctx;
+
+	if (hardirq_ctx[cpu])
+		return;
+
+	irqctx = (union irq_ctx *)&hardirq_stack[cpu * THREAD_SIZE];
+	irqctx->tinfo.task		= NULL;
+	irqctx->tinfo.exec_domain	= NULL;
+	irqctx->tinfo.cpu		= cpu;
+	irqctx->tinfo.preempt_count	= HARDIRQ_OFFSET;
+	irqctx->tinfo.addr_limit	= MAKE_MM_SEG(0);
+
+	hardirq_ctx[cpu] = irqctx;
+
+	irqctx = (union irq_ctx *)&softirq_stack[cpu * THREAD_SIZE];
+	irqctx->tinfo.task		= NULL;
+	irqctx->tinfo.exec_domain	= NULL;
+	irqctx->tinfo.cpu		= cpu;
+	irqctx->tinfo.preempt_count	= SOFTIRQ_OFFSET;
+	irqctx->tinfo.addr_limit	= MAKE_MM_SEG(0);
+
+	softirq_ctx[cpu] = irqctx;
+
+	printk("CPU %u irqstacks, hard=%p soft=%p\n",
+		cpu, hardirq_ctx[cpu], softirq_ctx[cpu]);
+}
+
+void irq_ctx_exit(int cpu)
+{
+	hardirq_ctx[cpu] = NULL;
+}
+
+extern asmlinkage void __do_softirq(void);
+
+asmlinkage void do_softirq(void)
+{
+	unsigned long flags;
+	struct thread_info *curctx;
+	union irq_ctx *irqctx;
+	u32 *isp;
+
+	if (in_interrupt())
+		return;
+
+	local_irq_save(flags);
+
+	if (local_softirq_pending()) {
+		curctx = current_thread_info();
+		irqctx = softirq_ctx[smp_processor_id()];
+		irqctx->tinfo.task = curctx->task;
+		irqctx->tinfo.previous_sp = current_stack_pointer;
+
+		/* build the stack frame on the softirq stack */
+		isp = (u32 *)((char *)irqctx + sizeof(*irqctx));
+
+		__asm__ __volatile__ (
+			"mov	r15, r9		\n"
+			"jsr	@%0		\n"
+			/* switch to the softirq stack */
+			" mov	%1, r15		\n"
+			/* restore the thread stack */
+			"mov	r9, r15		\n"
+			: /* no outputs */
+			: "r" (__do_softirq), "r" (isp)
+			/* XXX: A somewhat excessive clobber list? -PFM */
+			: "memory", "r0", "r1", "r2", "r3", "r4",
+			  "r5", "r6", "r7", "r8", "r9", "r15", "t", "pr"
+		);
+	}
+
+	local_irq_restore(flags);
+}
+EXPORT_SYMBOL(do_softirq);
+#endif
diff --git a/arch/sh/kernel/kgdb_stub.c b/arch/sh/kernel/kgdb_stub.c
index 42638b92b51c..9c6315f0335d 100644
--- a/arch/sh/kernel/kgdb_stub.c
+++ b/arch/sh/kernel/kgdb_stub.c
@@ -101,16 +101,17 @@
 #include <linux/linkage.h>
 #include <linux/init.h>
 
+#ifdef CONFIG_SH_KGDB_CONSOLE
+#include <linux/console.h>
+#endif
+
 #include <asm/system.h>
 #include <asm/current.h>
 #include <asm/signal.h>
 #include <asm/pgtable.h>
 #include <asm/ptrace.h>
 #include <asm/kgdb.h>
-
-#ifdef CONFIG_SH_KGDB_CONSOLE
-#include <linux/console.h>
-#endif
+#include <asm/io.h>
 
 /* Function pointers for linkage */
 kgdb_debug_hook_t *kgdb_debug_hook;
@@ -240,7 +241,6 @@ static jmp_buf rem_com_env;
 /* Misc static */
 static int stepped_address;
 static short stepped_opcode;
-static const char hexchars[] = "0123456789abcdef";
 static char in_buffer[BUFMAX];
 static char out_buffer[OUTBUFMAX];
 
@@ -253,29 +253,6 @@ typedef unsigned char threadref[8];
 #define BUF_THREAD_ID_SIZE 16
 #endif
 
-/* Return addr as a real volatile address */
-static inline unsigned int ctrl_inl(const unsigned long addr)
-{
-	return *(volatile unsigned long *) addr;
-}
-
-/* Correctly set *addr using volatile */
-static inline void ctrl_outl(const unsigned int b, unsigned long addr)
-{
-	*(volatile unsigned long *) addr = b;
-}
-
-/* Get high hex bits */
-static char highhex(const int x)
-{
-	return hexchars[(x >> 4) & 0xf];
-}
-
-/* Get low hex bits */
-static char lowhex(const int x)
-{
-	return hexchars[x & 0xf];
-}
 
 /* Convert ch to hex */
 static int hex(const char ch)
diff --git a/arch/sh/kernel/machine_kexec.c b/arch/sh/kernel/machine_kexec.c
index 6bcd8d92399f..08587cdb64d6 100644
--- a/arch/sh/kernel/machine_kexec.c
+++ b/arch/sh/kernel/machine_kexec.c
@@ -29,12 +29,6 @@ extern const unsigned char relocate_new_kernel[];
 extern const unsigned int relocate_new_kernel_size;
 extern void *gdb_vbr_vector;
 
-/*
- * Provide a dummy crash_notes definition while crash dump arrives to ppc.
- * This prevents breakage of crash_notes attribute in kernel/ksysfs.c.
- */
-void *crash_notes = NULL;
-
 void machine_shutdown(void)
 {
 }
diff --git a/arch/sh/kernel/pm.c b/arch/sh/kernel/pm.c
new file mode 100644
index 000000000000..10ab62c9aede
--- /dev/null
+++ b/arch/sh/kernel/pm.c
@@ -0,0 +1,88 @@
+/*
+ * Generic Power Management Routine
+ *
+ * Copyright (c) 2006 Andriy Skulysh <askulsyh@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License.
+ */
+#include <linux/suspend.h>
+#include <linux/delay.h>
+#include <linux/gfp.h>
+#include <asm/freq.h>
+#include <asm/io.h>
+#include <asm/watchdog.h>
+#include <asm/pm.h>
+
+#define INTR_OFFSET	0x600
+
+#define STBCR		0xffffff82
+#define STBCR2		0xffffff88
+
+#define STBCR_STBY	0x80
+#define STBCR_MSTP2	0x04
+
+#define MCR		0xffffff68
+#define RTCNT		0xffffff70
+
+#define MCR_RMODE	2
+#define MCR_RFSH	4
+
+void pm_enter(void)
+{
+	u8 stbcr, csr;
+	u16 frqcr, mcr;
+	u32 vbr_new, vbr_old;
+
+	set_bl_bit();
+
+	/* set wdt */
+	csr = sh_wdt_read_csr();
+	csr &= ~WTCSR_TME;
+	csr |= WTCSR_CKS_4096;
+	sh_wdt_write_csr(csr);
+	csr = sh_wdt_read_csr();
+	sh_wdt_write_cnt(0);
+
+	/* disable PLL1 */
+	frqcr = ctrl_inw(FRQCR);
+	frqcr &= ~(FRQCR_PLLEN | FRQCR_PSTBY);
+	ctrl_outw(frqcr, FRQCR);
+
+	/* enable standby */
+	stbcr = ctrl_inb(STBCR);
+	ctrl_outb(stbcr | STBCR_STBY | STBCR_MSTP2, STBCR);
+
+	/* set self-refresh */
+	mcr = ctrl_inw(MCR);
+	ctrl_outw(mcr & ~MCR_RFSH, MCR);
+
+	/* set interrupt handler */
+	asm volatile("stc vbr, %0" : "=r" (vbr_old));
+	vbr_new = get_zeroed_page(GFP_ATOMIC);
+	udelay(50);
+	memcpy((void*)(vbr_new + INTR_OFFSET),
+	       &wakeup_start, &wakeup_end - &wakeup_start);
+	asm volatile("ldc %0, vbr" : : "r" (vbr_new));
+
+	ctrl_outw(0, RTCNT);
+	ctrl_outw(mcr | MCR_RFSH | MCR_RMODE, MCR);
+
+	cpu_sleep();
+
+	asm volatile("ldc %0, vbr" : : "r" (vbr_old));
+
+	free_page(vbr_new);
+
+	/* enable PLL1 */
+	frqcr = ctrl_inw(FRQCR);
+	frqcr |= FRQCR_PSTBY;
+	ctrl_outw(frqcr, FRQCR);
+	udelay(50);
+	frqcr |= FRQCR_PLLEN;
+	ctrl_outw(frqcr, FRQCR);
+
+	ctrl_outb(stbcr, STBCR);
+
+	clear_bl_bit();
+}
diff --git a/arch/sh/kernel/process.c b/arch/sh/kernel/process.c
index f2031314cb2b..0b1d5dd7a93b 100644
--- a/arch/sh/kernel/process.c
+++ b/arch/sh/kernel/process.c
@@ -81,16 +81,6 @@ void cpu_idle(void)
 
 void machine_restart(char * __unused)
 {
-
-#ifdef CONFIG_KEXEC
-	struct kimage *image;
-	image = xchg(&kexec_image, 0);
-	if (image) {
-		machine_shutdown();
-		machine_kexec(image);
-	}
-#endif
-
 	/* SR.BL=1 and invoke address error to let CPU reset (manual reset) */
 	asm volatile("ldc %0, sr\n\t"
 		     "mov.l @%1, %0" : : "r" (0x10000000), "r" (0x80000001));
@@ -263,6 +253,7 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long usp,
 		unsigned long unused,
 		struct task_struct *p, struct pt_regs *regs)
 {
+	struct thread_info *ti = task_thread_info(p);
 	struct pt_regs *childregs;
 #if defined(CONFIG_SH_FPU)
 	struct task_struct *tsk = current;
@@ -277,8 +268,10 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long usp,
 
 	if (user_mode(regs)) {
 		childregs->regs[15] = usp;
+		ti->addr_limit = USER_DS;
 	} else {
 		childregs->regs[15] = (unsigned long)task_stack_page(p) + THREAD_SIZE;
+		ti->addr_limit = KERNEL_DS;
 	}
         if (clone_flags & CLONE_SETTLS) {
 		childregs->gbr = childregs->regs[0];
@@ -299,13 +292,15 @@ ubc_set_tracing(int asid, unsigned long pc)
 {
 	ctrl_outl(pc, UBC_BARA);
 
+#ifdef CONFIG_MMU
 	/* We don't have any ASID settings for the SH-2! */
 	if (cpu_data->type != CPU_SH7604)
 		ctrl_outb(asid, UBC_BASRA);
+#endif
 
 	ctrl_outl(0, UBC_BAMRA);
 
-	if (cpu_data->type == CPU_SH7729) {
+	if (cpu_data->type == CPU_SH7729 || cpu_data->type == CPU_SH7710) {
 		ctrl_outw(BBR_INST | BBR_READ | BBR_CPU, UBC_BBRA);
 		ctrl_outl(BRCR_PCBA | BRCR_PCTE, UBC_BRCR);
 	} else {
@@ -344,6 +339,7 @@ struct task_struct *__switch_to(struct task_struct *prev, struct task_struct *ne
 	}
 #endif
 
+#ifdef CONFIG_MMU
 	/*
 	 * Restore the kernel mode register
 	 *   	k7 (r7_bank1)
@@ -351,19 +347,21 @@ struct task_struct *__switch_to(struct task_struct *prev, struct task_struct *ne
 	asm volatile("ldc	%0, r7_bank"
 		     : /* no output */
 		     : "r" (task_thread_info(next)));
+#endif
 
-#ifdef CONFIG_MMU
 	/* If no tasks are using the UBC, we're done */
 	if (ubc_usercnt == 0)
 		/* If no tasks are using the UBC, we're done */;
 	else if (next->thread.ubc_pc && next->mm) {
-		ubc_set_tracing(next->mm->context & MMU_CONTEXT_ASID_MASK,
-				next->thread.ubc_pc);
+		int asid = 0;
+#ifdef CONFIG_MMU
+		asid |= next->mm->context.id & MMU_CONTEXT_ASID_MASK;
+#endif
+		ubc_set_tracing(asid, next->thread.ubc_pc);
 	} else {
 		ctrl_outw(0, UBC_BBRA);
 		ctrl_outw(0, UBC_BBRB);
 	}
-#endif
 
 	return prev;
 }
diff --git a/arch/sh/kernel/ptrace.c b/arch/sh/kernel/ptrace.c
index f7eebbde3291..04ca13a041c1 100644
--- a/arch/sh/kernel/ptrace.c
+++ b/arch/sh/kernel/ptrace.c
@@ -224,7 +224,6 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 
 	case PTRACE_SETDSPREGS: {
 		unsigned long dp;
-		int i;
 
 		ret = -EIO;
 		dp = ((unsigned long) child) + THREAD_SIZE -
diff --git a/arch/sh/kernel/semaphore.c b/arch/sh/kernel/semaphore.c
index a3c24dcbf01d..184119eeae56 100644
--- a/arch/sh/kernel/semaphore.c
+++ b/arch/sh/kernel/semaphore.c
@@ -14,7 +14,7 @@
 #include <asm/semaphore.h>
 #include <asm/semaphore-helper.h>
 
-spinlock_t semaphore_wake_lock;
+DEFINE_SPINLOCK(semaphore_wake_lock);
 
 /*
  * Semaphores are implemented using a two-way counter:
diff --git a/arch/sh/kernel/setup.c b/arch/sh/kernel/setup.c
index e75189cb1db7..5f587332234a 100644
--- a/arch/sh/kernel/setup.c
+++ b/arch/sh/kernel/setup.c
@@ -43,27 +43,14 @@ extern void * __rd_start, * __rd_end;
  * The bigger value means no problem.
  */
 struct sh_cpuinfo boot_cpu_data = { CPU_SH_NONE, 10000000, };
+#ifdef CONFIG_VT
 struct screen_info screen_info;
+#endif
 
 #if defined(CONFIG_SH_UNKNOWN)
 struct sh_machine_vector sh_mv;
 #endif
 
-/* We need this to satisfy some external references. */
-struct screen_info screen_info = {
-        0, 25,                  /* orig-x, orig-y */
-        0,                      /* unused */
-        0,                      /* orig-video-page */
-        0,                      /* orig-video-mode */
-        80,                     /* orig-video-cols */
-        0,0,0,                  /* ega_ax, ega_bx, ega_cx */
-        25,                     /* orig-video-lines */
-        0,                      /* orig-video-isVGA */
-        16                      /* orig-video-points */
-};
-
-extern void platform_setup(void);
-extern char *get_system_type(void);
 extern int root_mountflags;
 
 #define MV_NAME_SIZE 32
@@ -90,29 +77,8 @@ static struct sh_machine_vector* __init get_mv_byname(const char* name);
 
 static char command_line[COMMAND_LINE_SIZE] = { 0, };
 
-struct resource standard_io_resources[] = {
-	{ "dma1", 0x00, 0x1f },
-	{ "pic1", 0x20, 0x3f },
-	{ "timer", 0x40, 0x5f },
-	{ "keyboard", 0x60, 0x6f },
-	{ "dma page reg", 0x80, 0x8f },
-	{ "pic2", 0xa0, 0xbf },
-	{ "dma2", 0xc0, 0xdf },
-	{ "fpu", 0xf0, 0xff }
-};
-
-#define STANDARD_IO_RESOURCES (sizeof(standard_io_resources)/sizeof(struct resource))
-
-/* System RAM - interrupted by the 640kB-1M hole */
-#define code_resource (ram_resources[3])
-#define data_resource (ram_resources[4])
-static struct resource ram_resources[] = {
-	{ "System RAM", 0x000000, 0x09ffff, IORESOURCE_BUSY },
-	{ "System RAM", 0x100000, 0x100000, IORESOURCE_BUSY },
-	{ "Video RAM area", 0x0a0000, 0x0bffff },
-	{ "Kernel code", 0x100000, 0 },
-	{ "Kernel data", 0, 0 }
-};
+static struct resource code_resource = { .name = "Kernel code", };
+static struct resource data_resource = { .name = "Kernel data", };
 
 unsigned long memory_start, memory_end;
 
@@ -145,6 +111,24 @@ static inline void parse_cmdline (char ** cmdline_p, char mv_name[MV_NAME_SIZE],
 				memory_end = memory_start + mem_size;
 			}
 		}
+
+#ifdef CONFIG_EARLY_PRINTK
+		if (c == ' ' && !memcmp(from, "earlyprintk=", 12)) {
+			char *ep_end;
+
+			if (to != command_line)
+				to--;
+
+			from += 12;
+			ep_end = strchr(from, ' ');
+
+			setup_early_printk(from);
+			printk("early console enabled\n");
+
+			from = ep_end;
+		}
+#endif
+
 		if (c == ' ' && !memcmp(from, "sh_mv=", 6)) {
 			char* mv_end;
 			char* mv_comma;
@@ -237,6 +221,9 @@ static int __init sh_mv_setup(char **cmdline_p)
 	__set_io_port_base(mv_io_base);
 #endif
 
+	if (!sh_mv.mv_nr_irqs)
+		sh_mv.mv_nr_irqs = NR_IRQS;
+
 	return 0;
 }
 
@@ -245,11 +232,6 @@ void __init setup_arch(char **cmdline_p)
 	unsigned long bootmap_size;
 	unsigned long start_pfn, max_pfn, max_low_pfn;
 
-#ifdef CONFIG_EARLY_PRINTK
-	extern void enable_early_printk(void);
-
-	enable_early_printk();
-#endif
 #ifdef CONFIG_CMDLINE_BOOL
         strcpy(COMMAND_LINE, CONFIG_CMDLINE);
 #endif
@@ -368,14 +350,14 @@ void __init setup_arch(char **cmdline_p)
 #endif
 
 	/* Perform the machine specific initialisation */
-	platform_setup();
+	if (likely(sh_mv.mv_setup))
+		sh_mv.mv_setup(cmdline_p);
 
 	paging_init();
 }
 
 struct sh_machine_vector* __init get_mv_byname(const char* name)
 {
-	extern int strcasecmp(const char *, const char *);
 	extern long __machvec_start, __machvec_end;
 	struct sh_machine_vector *all_vecs =
 		(struct sh_machine_vector *)&__machvec_start;
@@ -410,25 +392,18 @@ static int __init topology_init(void)
 subsys_initcall(topology_init);
 
 static const char *cpu_name[] = {
-	[CPU_SH7604]	= "SH7604",
-	[CPU_SH7705]	= "SH7705",
-	[CPU_SH7708]	= "SH7708",
-	[CPU_SH7729]	= "SH7729",
-	[CPU_SH7300]	= "SH7300",
-	[CPU_SH7750]	= "SH7750",
-	[CPU_SH7750S]	= "SH7750S",
-	[CPU_SH7750R]	= "SH7750R",
-	[CPU_SH7751]	= "SH7751",
-	[CPU_SH7751R]	= "SH7751R",
-	[CPU_SH7760]	= "SH7760",
-	[CPU_SH73180]	= "SH73180",
-	[CPU_ST40RA]	= "ST40RA",
-	[CPU_ST40GX1]	= "ST40GX1",
-	[CPU_SH4_202]	= "SH4-202",
-	[CPU_SH4_501]	= "SH4-501",
-	[CPU_SH7770]	= "SH7770",
-	[CPU_SH7780]	= "SH7780",
-	[CPU_SH7781]	= "SH7781",
+	[CPU_SH7604]	= "SH7604",	[CPU_SH7300]	= "SH7300",
+	[CPU_SH7705]	= "SH7705",	[CPU_SH7706]	= "SH7706",
+	[CPU_SH7707]	= "SH7707",	[CPU_SH7708]	= "SH7708",
+	[CPU_SH7709]	= "SH7709",	[CPU_SH7710]	= "SH7710",
+	[CPU_SH7729]	= "SH7729",	[CPU_SH7750]	= "SH7750",
+	[CPU_SH7750S]	= "SH7750S",	[CPU_SH7750R]	= "SH7750R",
+	[CPU_SH7751]	= "SH7751",	[CPU_SH7751R]	= "SH7751R",
+	[CPU_SH7760]	= "SH7760",	[CPU_SH73180]	= "SH73180",
+	[CPU_ST40RA]	= "ST40RA",	[CPU_ST40GX1]	= "ST40GX1",
+	[CPU_SH4_202]	= "SH4-202",	[CPU_SH4_501]	= "SH4-501",
+	[CPU_SH7770]	= "SH7770",	[CPU_SH7780]	= "SH7780",
+	[CPU_SH7781]	= "SH7781",	[CPU_SH7343]	= "SH7343",
 	[CPU_SH_NONE]	= "Unknown"
 };
 
@@ -438,8 +413,10 @@ const char *get_cpu_subtype(void)
 }
 
 #ifdef CONFIG_PROC_FS
+/* Symbolic CPU flags, keep in sync with asm/cpu-features.h */
 static const char *cpu_flags[] = {
-	"none", "fpu", "p2flush", "mmuassoc", "dsp", "perfctr", "ptea", NULL
+	"none", "fpu", "p2flush", "mmuassoc", "dsp", "perfctr",
+	"ptea", "llsc", "l2", NULL
 };
 
 static void show_cpuflags(struct seq_file *m)
@@ -460,7 +437,8 @@ static void show_cpuflags(struct seq_file *m)
 	seq_printf(m, "\n");
 }
 
-static void show_cacheinfo(struct seq_file *m, const char *type, struct cache_info info)
+static void show_cacheinfo(struct seq_file *m, const char *type,
+			   struct cache_info info)
 {
 	unsigned int cache_size;
 
@@ -493,7 +471,7 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 	 * unified cache on the SH-2 and SH-3, as well as the harvard
 	 * style cache on the SH-4.
 	 */
-	if (test_bit(SH_CACHE_COMBINED, &(boot_cpu_data.icache.flags))) {
+	if (boot_cpu_data.icache.flags & SH_CACHE_COMBINED) {
 		seq_printf(m, "unified\n");
 		show_cacheinfo(m, "cache", boot_cpu_data.icache);
 	} else {
@@ -502,6 +480,10 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 		show_cacheinfo(m, "dcache", boot_cpu_data.dcache);
 	}
 
+	/* Optional secondary cache */
+	if (boot_cpu_data.flags & CPU_HAS_L2_CACHE)
+		show_cacheinfo(m, "scache", boot_cpu_data.scache);
+
 	seq_printf(m, "bogomips\t: %lu.%02lu\n",
 		     boot_cpu_data.loops_per_jiffy/(500000/HZ),
 		     (boot_cpu_data.loops_per_jiffy/(5000/HZ)) % 100);
@@ -617,4 +599,3 @@ static int __init kgdb_parse_options(char *options)
 }
 __setup("kgdb=", kgdb_parse_options);
 #endif /* CONFIG_SH_KGDB */
-
diff --git a/arch/sh/kernel/sh_ksyms.c b/arch/sh/kernel/sh_ksyms.c
index 245ed8f945e8..d3cbfa2ad4a7 100644
--- a/arch/sh/kernel/sh_ksyms.c
+++ b/arch/sh/kernel/sh_ksyms.c
@@ -27,21 +27,11 @@ EXPORT_SYMBOL(sh_mv);
 
 /* platform dependent support */
 EXPORT_SYMBOL(dump_fpu);
-EXPORT_SYMBOL(iounmap);
-EXPORT_SYMBOL(enable_irq);
-EXPORT_SYMBOL(disable_irq);
-EXPORT_SYMBOL(probe_irq_mask);
 EXPORT_SYMBOL(kernel_thread);
-EXPORT_SYMBOL(disable_irq_nosync);
 EXPORT_SYMBOL(irq_desc);
 EXPORT_SYMBOL(no_irq_type);
 
-EXPORT_SYMBOL(strstr);
 EXPORT_SYMBOL(strlen);
-EXPORT_SYMBOL(strnlen);
-EXPORT_SYMBOL(strchr);
-EXPORT_SYMBOL(strcat);
-EXPORT_SYMBOL(strncat);
 
 /* PCI exports */
 #ifdef CONFIG_PCI
@@ -52,13 +42,8 @@ EXPORT_SYMBOL(pci_free_consistent);
 /* mem exports */
 EXPORT_SYMBOL(memchr);
 EXPORT_SYMBOL(memcpy);
-EXPORT_SYMBOL(memcpy_fromio);
-EXPORT_SYMBOL(memcpy_toio);
 EXPORT_SYMBOL(memset);
-EXPORT_SYMBOL(memset_io);
 EXPORT_SYMBOL(memmove);
-EXPORT_SYMBOL(memcmp);
-EXPORT_SYMBOL(memscan);
 EXPORT_SYMBOL(__copy_user);
 EXPORT_SYMBOL(boot_cpu_data);
 
@@ -94,7 +79,9 @@ EXPORT_SYMBOL(strcpy);
 DECLARE_EXPORT(__movstr_i4_even);
 DECLARE_EXPORT(__movstr_i4_odd);
 DECLARE_EXPORT(__movstrSI12_i4);
+#endif
 
+#if defined(CONFIG_CPU_SH4) || defined(CONFIG_SH7705_CACHE_32KB)
 /* needed by some modules */
 EXPORT_SYMBOL(flush_cache_all);
 EXPORT_SYMBOL(flush_cache_range);
@@ -102,11 +89,9 @@ EXPORT_SYMBOL(flush_dcache_page);
 EXPORT_SYMBOL(__flush_purge_region);
 #endif
 
-#if defined(CONFIG_SH7705_CACHE_32KB)
-EXPORT_SYMBOL(flush_cache_all);
-EXPORT_SYMBOL(flush_cache_range);
-EXPORT_SYMBOL(flush_dcache_page);
-EXPORT_SYMBOL(__flush_purge_region);
+#if defined(CONFIG_MMU) && (defined(CONFIG_CPU_SH4) || \
+	defined(CONFIG_SH7705_CACHE_32KB))
+EXPORT_SYMBOL(clear_user_page);
 #endif
 
 EXPORT_SYMBOL(flush_tlb_page);
@@ -116,7 +101,12 @@ EXPORT_SYMBOL(__down_trylock);
 EXPORT_SYMBOL(synchronize_irq);
 #endif
 
+#ifdef CONFIG_PM
+EXPORT_SYMBOL(pm_suspend);
+#endif
+
 EXPORT_SYMBOL(csum_partial);
+#ifdef CONFIG_IPV6
 EXPORT_SYMBOL(csum_ipv6_magic);
-EXPORT_SYMBOL(consistent_sync);
+#endif
 EXPORT_SYMBOL(clear_page);
diff --git a/arch/sh/kernel/signal.c b/arch/sh/kernel/signal.c
index b475c4d2405f..5213f5bc6ce0 100644
--- a/arch/sh/kernel/signal.c
+++ b/arch/sh/kernel/signal.c
@@ -8,7 +8,6 @@
  *  SuperH version:  Copyright (C) 1999, 2000  Niibe Yutaka & Kaz Kojima
  *
  */
-
 #include <linux/sched.h>
 #include <linux/mm.h>
 #include <linux/smp.h>
@@ -21,6 +20,7 @@
 #include <linux/unistd.h>
 #include <linux/stddef.h>
 #include <linux/tty.h>
+#include <linux/elf.h>
 #include <linux/personality.h>
 #include <linux/binfmts.h>
 
@@ -29,12 +29,8 @@
 #include <asm/pgtable.h>
 #include <asm/cacheflush.h>
 
-#define DEBUG_SIG 0
-
 #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
 
-asmlinkage int do_signal(struct pt_regs *regs, sigset_t *oldset);
-
 /*
  * Atomically swap in the new signal mask, and wait for a signal.
  */
@@ -43,51 +39,17 @@ sys_sigsuspend(old_sigset_t mask,
 	       unsigned long r5, unsigned long r6, unsigned long r7,
 	       struct pt_regs regs)
 {
-	sigset_t saveset;
-
 	mask &= _BLOCKABLE;
 	spin_lock_irq(&current->sighand->siglock);
-	saveset = current->blocked;
+	current->saved_sigmask = current->blocked;
 	siginitset(&current->blocked, mask);
 	recalc_sigpending();
 	spin_unlock_irq(&current->sighand->siglock);
 
-	regs.regs[0] = -EINTR;
-	while (1) {
-		current->state = TASK_INTERRUPTIBLE;
-		schedule();
-		if (do_signal(&regs, &saveset))
-			return -EINTR;
-	}
-}
-
-asmlinkage int
-sys_rt_sigsuspend(sigset_t *unewset, size_t sigsetsize,
-		  unsigned long r6, unsigned long r7,
-		  struct pt_regs regs)
-{
-	sigset_t saveset, newset;
-
-	/* XXX: Don't preclude handling different sized sigset_t's.  */
-	if (sigsetsize != sizeof(sigset_t))
-		return -EINVAL;
-
-	if (copy_from_user(&newset, unewset, sizeof(newset)))
-		return -EFAULT;
-	sigdelsetmask(&newset, ~_BLOCKABLE);
-	spin_lock_irq(&current->sighand->siglock);
-	saveset = current->blocked;
-	current->blocked = newset;
-	recalc_sigpending();
-	spin_unlock_irq(&current->sighand->siglock);
-
-	regs.regs[0] = -EINTR;
-	while (1) {
-		current->state = TASK_INTERRUPTIBLE;
-		schedule();
-		if (do_signal(&regs, &saveset))
-			return -EINTR;
-	}
+	current->state = TASK_INTERRUPTIBLE;
+	schedule();
+	set_thread_flag(TIF_RESTORE_SIGMASK);
+	return -ERESTARTNOHAND;
 }
 
 asmlinkage int 
@@ -348,7 +310,12 @@ get_sigframe(struct k_sigaction *ka, unsigned long sp, size_t frame_size)
 	return (void __user *)((sp - frame_size) & -8ul);
 }
 
-static void setup_frame(int sig, struct k_sigaction *ka,
+/* These symbols are defined with the addresses in the vsyscall page.
+   See vsyscall-trapa.S.  */
+extern void __user __kernel_sigreturn;
+extern void __user __kernel_rt_sigreturn;
+
+static int setup_frame(int sig, struct k_sigaction *ka,
 			sigset_t *set, struct pt_regs *regs)
 {
 	struct sigframe __user *frame;
@@ -368,15 +335,18 @@ static void setup_frame(int sig, struct k_sigaction *ka,
 
 	err |= setup_sigcontext(&frame->sc, regs, set->sig[0]);
 
-	if (_NSIG_WORDS > 1) {
+	if (_NSIG_WORDS > 1)
 		err |= __copy_to_user(frame->extramask, &set->sig[1],
 				      sizeof(frame->extramask));
-	}
 
 	/* Set up to return from userspace.  If provided, use a stub
 	   already in userspace.  */
 	if (ka->sa.sa_flags & SA_RESTORER) {
 		regs->pr = (unsigned long) ka->sa.sa_restorer;
+#ifdef CONFIG_VSYSCALL
+	} else if (likely(current->mm->context.vdso)) {
+		regs->pr = VDSO_SYM(&__kernel_sigreturn);
+#endif
 	} else {
 		/* Generate return code (system call to sigreturn) */
 		err |= __put_user(MOVW(7), &frame->retcode[0]);
@@ -402,21 +372,22 @@ static void setup_frame(int sig, struct k_sigaction *ka,
 
 	set_fs(USER_DS);
 
-#if DEBUG_SIG
-	printk("SIG deliver (%s:%d): sp=%p pc=%08lx pr=%08lx\n",
-		current->comm, current->pid, frame, regs->pc, regs->pr);
-#endif
+	pr_debug("SIG deliver (%s:%d): sp=%p pc=%08lx pr=%08lx\n",
+		 current->comm, current->pid, frame, regs->pc, regs->pr);
 
 	flush_cache_sigtramp(regs->pr);
+
 	if ((-regs->pr & (L1_CACHE_BYTES-1)) < sizeof(frame->retcode))
 		flush_cache_sigtramp(regs->pr + L1_CACHE_BYTES);
-	return;
+
+	return 0;
 
 give_sigsegv:
 	force_sigsegv(sig, current);
+	return -EFAULT;
 }
 
-static void setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
+static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 			   sigset_t *set, struct pt_regs *regs)
 {
 	struct rt_sigframe __user *frame;
@@ -452,6 +423,10 @@ static void setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 	   already in userspace.  */
 	if (ka->sa.sa_flags & SA_RESTORER) {
 		regs->pr = (unsigned long) ka->sa.sa_restorer;
+#ifdef CONFIG_VSYSCALL
+	} else if (likely(current->mm->context.vdso)) {
+		regs->pr = VDSO_SYM(&__kernel_rt_sigreturn);
+#endif
 	} else {
 		/* Generate return code (system call to rt_sigreturn) */
 		err |= __put_user(MOVW(7), &frame->retcode[0]);
@@ -477,28 +452,31 @@ static void setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 
 	set_fs(USER_DS);
 
-#if DEBUG_SIG
-	printk("SIG deliver (%s:%d): sp=%p pc=%08lx pr=%08lx\n",
-		current->comm, current->pid, frame, regs->pc, regs->pr);
-#endif
+	pr_debug("SIG deliver (%s:%d): sp=%p pc=%08lx pr=%08lx\n",
+		 current->comm, current->pid, frame, regs->pc, regs->pr);
 
 	flush_cache_sigtramp(regs->pr);
+
 	if ((-regs->pr & (L1_CACHE_BYTES-1)) < sizeof(frame->retcode))
 		flush_cache_sigtramp(regs->pr + L1_CACHE_BYTES);
-	return;
+
+	return 0;
 
 give_sigsegv:
 	force_sigsegv(sig, current);
+	return -EFAULT;
 }
 
 /*
  * OK, we're invoking a handler
  */
 
-static void
+static int
 handle_signal(unsigned long sig, struct k_sigaction *ka, siginfo_t *info,
 	      sigset_t *oldset, struct pt_regs *regs)
 {
+	int ret;
+
 	/* Are we from a system call? */
 	if (regs->tra >= 0) {
 		/* If so, check system call restarting.. */
@@ -539,19 +517,23 @@ handle_signal(unsigned long sig, struct k_sigaction *ka, siginfo_t *info,
 
 	/* Set up the stack frame */
 	if (ka->sa.sa_flags & SA_SIGINFO)
-		setup_rt_frame(sig, ka, info, oldset, regs);
+		ret = setup_rt_frame(sig, ka, info, oldset, regs);
 	else
-		setup_frame(sig, ka, oldset, regs);
+		ret = setup_frame(sig, ka, oldset, regs);
 
 	if (ka->sa.sa_flags & SA_ONESHOT)
 		ka->sa.sa_handler = SIG_DFL;
 
-	spin_lock_irq(&current->sighand->siglock);
-	sigorsets(&current->blocked,&current->blocked,&ka->sa.sa_mask);
-	if (!(ka->sa.sa_flags & SA_NODEFER))
-		sigaddset(&current->blocked,sig);
-	recalc_sigpending();
-	spin_unlock_irq(&current->sighand->siglock);
+	if (ret == 0) {
+		spin_lock_irq(&current->sighand->siglock);
+		sigorsets(&current->blocked,&current->blocked,&ka->sa.sa_mask);
+		if (!(ka->sa.sa_flags & SA_NODEFER))
+			sigaddset(&current->blocked,sig);
+		recalc_sigpending();
+		spin_unlock_irq(&current->sighand->siglock);
+	}
+
+	return ret;
 }
 
 /*
@@ -563,11 +545,12 @@ handle_signal(unsigned long sig, struct k_sigaction *ka, siginfo_t *info,
  * the kernel can handle, and then we build all the user-level signal handling
  * stack-frames in one go after that.
  */
-int do_signal(struct pt_regs *regs, sigset_t *oldset)
+static void do_signal(struct pt_regs *regs, unsigned int save_r0)
 {
 	siginfo_t info;
 	int signr;
 	struct k_sigaction ka;
+	sigset_t *oldset;
 
 	/*
 	 * We want the common case to go fast, which
@@ -576,19 +559,27 @@ int do_signal(struct pt_regs *regs, sigset_t *oldset)
 	 * if so.
 	 */
 	if (!user_mode(regs))
-		return 1;
+		return;
 
 	if (try_to_freeze())
 		goto no_signal;
 
-	if (!oldset)
+	if (test_thread_flag(TIF_RESTORE_SIGMASK))
+		oldset = &current->saved_sigmask;
+	else
 		oldset = &current->blocked;
 
 	signr = get_signal_to_deliver(&info, &ka, regs, NULL);
 	if (signr > 0) {
 		/* Whee!  Actually deliver the signal.  */
-		handle_signal(signr, &ka, &info, oldset, regs);
-		return 1;
+		if (handle_signal(signr, &ka, &info, oldset, regs) == 0) {
+			/* a signal was successfully delivered; the saved
+			 * sigmask will have been stored in the signal frame,
+			 * and will be restored by sigreturn, so we can simply
+			 * clear the TIF_RESTORE_SIGMASK flag */
+			if (test_thread_flag(TIF_RESTORE_SIGMASK))
+				clear_thread_flag(TIF_RESTORE_SIGMASK);
+		}
 	}
 
  no_signal:
@@ -597,10 +588,27 @@ int do_signal(struct pt_regs *regs, sigset_t *oldset)
 		/* Restart the system call - no handlers present */
 		if (regs->regs[0] == -ERESTARTNOHAND ||
 		    regs->regs[0] == -ERESTARTSYS ||
-		    regs->regs[0] == -ERESTARTNOINTR ||
-		    regs->regs[0] == -ERESTART_RESTARTBLOCK) {
+		    regs->regs[0] == -ERESTARTNOINTR) {
+		    	regs->regs[0] = save_r0;
+			regs->pc -= 2;
+		} else if (regs->regs[0] == -ERESTART_RESTARTBLOCK) {
 			regs->pc -= 2;
+			regs->regs[3] = __NR_restart_syscall;
 		}
 	}
-	return 0;
+
+	/* if there's no signal to deliver, we just put the saved sigmask
+	 * back */
+	if (test_thread_flag(TIF_RESTORE_SIGMASK)) {
+		clear_thread_flag(TIF_RESTORE_SIGMASK);
+		sigprocmask(SIG_SETMASK, &current->saved_sigmask, NULL);
+	}
+}
+
+asmlinkage void do_notify_resume(struct pt_regs *regs, unsigned int save_r0,
+				 __u32 thread_info_flags)
+{
+	/* deal with pending signal delivery */
+	if (thread_info_flags & (_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK))
+		do_signal(regs, save_r0);
 }
diff --git a/arch/sh/kernel/sys_sh.c b/arch/sh/kernel/sys_sh.c
index 917b2f32f260..b68ff705f067 100644
--- a/arch/sh/kernel/sys_sh.c
+++ b/arch/sh/kernel/sys_sh.c
@@ -21,7 +21,8 @@
 #include <linux/mman.h>
 #include <linux/file.h>
 #include <linux/utsname.h>
-
+#include <linux/module.h>
+#include <asm/cacheflush.h>
 #include <asm/uaccess.h>
 #include <asm/ipc.h>
 
@@ -44,11 +45,16 @@ asmlinkage int sys_pipe(unsigned long r4, unsigned long r5,
 	return error;
 }
 
-#if defined(HAVE_ARCH_UNMAPPED_AREA)
+unsigned long shm_align_mask = PAGE_SIZE - 1;	/* Sane caches */
+
+EXPORT_SYMBOL(shm_align_mask);
+
 /*
- * To avoid cache alias, we map the shard page with same color.
+ * To avoid cache aliases, we map the shared page with same color.
  */
-#define COLOUR_ALIGN(addr)	(((addr)+SHMLBA-1)&~(SHMLBA-1))
+#define COLOUR_ALIGN(addr, pgoff)				\
+	((((addr) + shm_align_mask) & ~shm_align_mask) +	\
+	 (((pgoff) << PAGE_SHIFT) & shm_align_mask))
 
 unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr,
 	unsigned long len, unsigned long pgoff, unsigned long flags)
@@ -56,43 +62,52 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr,
 	struct mm_struct *mm = current->mm;
 	struct vm_area_struct *vma;
 	unsigned long start_addr;
+	int do_colour_align;
 
 	if (flags & MAP_FIXED) {
 		/* We do not accept a shared mapping if it would violate
 		 * cache aliasing constraints.
 		 */
-		if ((flags & MAP_SHARED) && (addr & (SHMLBA - 1)))
+		if ((flags & MAP_SHARED) && (addr & shm_align_mask))
 			return -EINVAL;
 		return addr;
 	}
 
-	if (len > TASK_SIZE)
+	if (unlikely(len > TASK_SIZE))
 		return -ENOMEM;
 
+	do_colour_align = 0;
+	if (filp || (flags & MAP_SHARED))
+		do_colour_align = 1;
+
 	if (addr) {
-		if (flags & MAP_PRIVATE)
-			addr = PAGE_ALIGN(addr);
+		if (do_colour_align)
+			addr = COLOUR_ALIGN(addr, pgoff);
 		else
-			addr = COLOUR_ALIGN(addr);
+			addr = PAGE_ALIGN(addr);
+
 		vma = find_vma(mm, addr);
 		if (TASK_SIZE - len >= addr &&
 		    (!vma || addr + len <= vma->vm_start))
 			return addr;
 	}
-	if (len <= mm->cached_hole_size) {
+
+	if (len > mm->cached_hole_size) {
+		start_addr = addr = mm->free_area_cache;
+	} else {
 	        mm->cached_hole_size = 0;
-		mm->free_area_cache = TASK_UNMAPPED_BASE;
+		start_addr = addr = TASK_UNMAPPED_BASE;
 	}
-	if (flags & MAP_PRIVATE)
-		addr = PAGE_ALIGN(mm->free_area_cache);
-	else
-		addr = COLOUR_ALIGN(mm->free_area_cache);
-	start_addr = addr;
 
 full_search:
+	if (do_colour_align)
+		addr = COLOUR_ALIGN(addr, pgoff);
+	else
+		addr = PAGE_ALIGN(mm->free_area_cache);
+
 	for (vma = find_vma(mm, addr); ; vma = vma->vm_next) {
 		/* At this point:  (!vma || addr < vma->vm_end). */
-		if (TASK_SIZE - len < addr) {
+		if (unlikely(TASK_SIZE - len < addr)) {
 			/*
 			 * Start a new search - just in case we missed
 			 * some holes.
@@ -104,7 +119,7 @@ full_search:
 			}
 			return -ENOMEM;
 		}
-		if (!vma || addr + len <= vma->vm_start) {
+		if (likely(!vma || addr + len <= vma->vm_start)) {
 			/*
 			 * Remember the place where we stopped the search:
 			 */
@@ -115,11 +130,10 @@ full_search:
 		        mm->cached_hole_size = vma->vm_start - addr;
 
 		addr = vma->vm_end;
-		if (!(flags & MAP_PRIVATE))
-			addr = COLOUR_ALIGN(addr);
+		if (do_colour_align)
+			addr = COLOUR_ALIGN(addr, pgoff);
 	}
 }
-#endif
 
 static inline long
 do_mmap2(unsigned long addr, unsigned long len, unsigned long prot, 
diff --git a/arch/sh/kernel/syscalls.S b/arch/sh/kernel/syscalls.S
new file mode 100644
index 000000000000..768334e95075
--- /dev/null
+++ b/arch/sh/kernel/syscalls.S
@@ -0,0 +1,353 @@
+/*
+ * arch/sh/kernel/syscalls.S
+ *
+ * System call table for SuperH
+ *
+ *  Copyright (C) 1999, 2000, 2002  Niibe Yutaka
+ *  Copyright (C) 2003  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ */
+#include <linux/sys.h>
+#include <linux/linkage.h>
+
+#if !defined(CONFIG_NFSD) && !defined(CONFIG_NFSD_MODULE)
+#define sys_nfsservctl		sys_ni_syscall
+#endif
+
+#if !defined(CONFIG_MMU)
+#define sys_madvise		sys_ni_syscall
+#define sys_readahead		sys_ni_syscall
+#define sys_mprotect		sys_ni_syscall
+#define sys_msync		sys_ni_syscall
+#define sys_mlock		sys_ni_syscall
+#define sys_munlock		sys_ni_syscall
+#define sys_mlockall		sys_ni_syscall
+#define sys_munlockall		sys_ni_syscall
+#define sys_mremap		sys_ni_syscall
+#define sys_mincore		sys_ni_syscall
+#define sys_remap_file_pages	sys_ni_syscall
+#endif
+
+	.data
+ENTRY(sys_call_table)
+	.long sys_restart_syscall	/* 0  -  old "setup()" system call*/
+	.long sys_exit
+	.long sys_fork
+	.long sys_read
+	.long sys_write
+	.long sys_open		/* 5 */
+	.long sys_close
+	.long sys_waitpid
+	.long sys_creat
+	.long sys_link
+	.long sys_unlink		/* 10 */
+	.long sys_execve
+	.long sys_chdir
+	.long sys_time
+	.long sys_mknod
+	.long sys_chmod		/* 15 */
+	.long sys_lchown16
+	.long sys_ni_syscall	/* old break syscall holder */
+	.long sys_stat
+	.long sys_lseek
+	.long sys_getpid		/* 20 */
+	.long sys_mount
+	.long sys_oldumount
+	.long sys_setuid16
+	.long sys_getuid16
+	.long sys_stime		/* 25 */
+	.long sys_ptrace
+	.long sys_alarm
+	.long sys_fstat
+	.long sys_pause
+	.long sys_utime		/* 30 */
+	.long sys_ni_syscall	/* old stty syscall holder */
+	.long sys_ni_syscall	/* old gtty syscall holder */
+	.long sys_access
+	.long sys_nice
+	.long sys_ni_syscall	/* 35 */		/* old ftime syscall holder */
+	.long sys_sync
+	.long sys_kill
+	.long sys_rename
+	.long sys_mkdir
+	.long sys_rmdir		/* 40 */
+	.long sys_dup
+	.long sys_pipe
+	.long sys_times
+	.long sys_ni_syscall	/* old prof syscall holder */
+	.long sys_brk		/* 45 */
+	.long sys_setgid16
+	.long sys_getgid16
+	.long sys_signal
+	.long sys_geteuid16
+	.long sys_getegid16	/* 50 */
+	.long sys_acct
+	.long sys_umount		/* recycled never used phys() */
+	.long sys_ni_syscall	/* old lock syscall holder */
+	.long sys_ioctl
+	.long sys_fcntl		/* 55 */
+	.long sys_ni_syscall	/* old mpx syscall holder */
+	.long sys_setpgid
+	.long sys_ni_syscall	/* old ulimit syscall holder */
+	.long sys_ni_syscall	/* sys_olduname */
+	.long sys_umask		/* 60 */
+	.long sys_chroot
+	.long sys_ustat
+	.long sys_dup2
+	.long sys_getppid
+	.long sys_getpgrp		/* 65 */
+	.long sys_setsid
+	.long sys_sigaction
+	.long sys_sgetmask
+	.long sys_ssetmask
+	.long sys_setreuid16	/* 70 */
+	.long sys_setregid16
+	.long sys_sigsuspend
+	.long sys_sigpending
+	.long sys_sethostname
+	.long sys_setrlimit	/* 75 */
+	.long sys_old_getrlimit
+	.long sys_getrusage
+	.long sys_gettimeofday
+	.long sys_settimeofday
+	.long sys_getgroups16	/* 80 */
+	.long sys_setgroups16
+	.long sys_ni_syscall	/* sys_oldselect */
+	.long sys_symlink
+	.long sys_lstat
+	.long sys_readlink		/* 85 */
+	.long sys_uselib
+	.long sys_swapon
+	.long sys_reboot
+	.long old_readdir
+	.long old_mmap		/* 90 */
+	.long sys_munmap
+	.long sys_truncate
+	.long sys_ftruncate
+	.long sys_fchmod
+	.long sys_fchown16		/* 95 */
+	.long sys_getpriority
+	.long sys_setpriority
+	.long sys_ni_syscall	/* old profil syscall holder */
+	.long sys_statfs
+	.long sys_fstatfs		/* 100 */
+	.long sys_ni_syscall	/* ioperm */
+	.long sys_socketcall
+	.long sys_syslog
+	.long sys_setitimer
+	.long sys_getitimer	/* 105 */
+	.long sys_newstat
+	.long sys_newlstat
+	.long sys_newfstat
+	.long sys_uname
+	.long sys_ni_syscall	/* 110 */ /* iopl */
+	.long sys_vhangup
+	.long sys_ni_syscall	/* idle */
+	.long sys_ni_syscall	/* vm86old */
+	.long sys_wait4
+	.long sys_swapoff		/* 115 */
+	.long sys_sysinfo
+	.long sys_ipc
+	.long sys_fsync
+	.long sys_sigreturn
+	.long sys_clone		/* 120 */
+	.long sys_setdomainname
+	.long sys_newuname
+	.long sys_ni_syscall	/* sys_modify_ldt */
+	.long sys_adjtimex
+	.long sys_mprotect		/* 125 */
+	.long sys_sigprocmask
+	.long sys_ni_syscall	/* old "create_module" */
+	.long sys_init_module
+	.long sys_delete_module
+	.long sys_ni_syscall	/* 130: old "get_kernel_syms" */
+	.long sys_quotactl
+	.long sys_getpgid
+	.long sys_fchdir
+	.long sys_bdflush
+	.long sys_sysfs		/* 135 */
+	.long sys_personality
+	.long sys_ni_syscall	/* for afs_syscall */
+	.long sys_setfsuid16
+	.long sys_setfsgid16
+	.long sys_llseek		/* 140 */
+	.long sys_getdents
+	.long sys_select
+	.long sys_flock
+	.long sys_msync
+	.long sys_readv		/* 145 */
+	.long sys_writev
+	.long sys_getsid
+	.long sys_fdatasync
+	.long sys_sysctl
+	.long sys_mlock		/* 150 */
+	.long sys_munlock
+	.long sys_mlockall
+	.long sys_munlockall
+	.long sys_sched_setparam
+	.long sys_sched_getparam   /* 155 */
+	.long sys_sched_setscheduler
+	.long sys_sched_getscheduler
+	.long sys_sched_yield
+	.long sys_sched_get_priority_max
+	.long sys_sched_get_priority_min  /* 160 */
+	.long sys_sched_rr_get_interval
+	.long sys_nanosleep
+	.long sys_mremap
+	.long sys_setresuid16
+	.long sys_getresuid16	/* 165 */
+	.long sys_ni_syscall	/* vm86 */
+	.long sys_ni_syscall	/* old "query_module" */
+	.long sys_poll
+	.long sys_nfsservctl
+	.long sys_setresgid16	/* 170 */
+	.long sys_getresgid16
+	.long sys_prctl
+	.long sys_rt_sigreturn
+	.long sys_rt_sigaction
+	.long sys_rt_sigprocmask	/* 175 */
+	.long sys_rt_sigpending
+	.long sys_rt_sigtimedwait
+	.long sys_rt_sigqueueinfo
+	.long sys_rt_sigsuspend
+	.long sys_pread_wrapper	   /* 180 */
+	.long sys_pwrite_wrapper
+	.long sys_chown16
+	.long sys_getcwd
+	.long sys_capget
+	.long sys_capset           /* 185 */
+	.long sys_sigaltstack
+	.long sys_sendfile
+	.long sys_ni_syscall	/* streams1 */
+	.long sys_ni_syscall	/* streams2 */
+	.long sys_vfork            /* 190 */
+	.long sys_getrlimit
+	.long sys_mmap2
+	.long sys_truncate64
+	.long sys_ftruncate64
+	.long sys_stat64		/* 195 */
+	.long sys_lstat64
+	.long sys_fstat64
+	.long sys_lchown
+	.long sys_getuid
+	.long sys_getgid		/* 200 */
+	.long sys_geteuid
+	.long sys_getegid
+	.long sys_setreuid
+	.long sys_setregid
+	.long sys_getgroups	/* 205 */
+	.long sys_setgroups
+	.long sys_fchown
+	.long sys_setresuid
+	.long sys_getresuid
+	.long sys_setresgid	/* 210 */
+	.long sys_getresgid
+	.long sys_chown
+	.long sys_setuid
+	.long sys_setgid
+	.long sys_setfsuid		/* 215 */
+	.long sys_setfsgid
+	.long sys_pivot_root
+	.long sys_mincore
+	.long sys_madvise
+	.long sys_getdents64	/* 220 */
+	.long sys_fcntl64
+	.long sys_ni_syscall	/* reserved for TUX */
+	.long sys_ni_syscall	/* Reserved for Security */
+	.long sys_gettid
+	.long sys_readahead	/* 225 */
+	.long sys_setxattr
+	.long sys_lsetxattr
+	.long sys_fsetxattr
+	.long sys_getxattr
+	.long sys_lgetxattr	/* 230 */
+	.long sys_fgetxattr
+	.long sys_listxattr
+	.long sys_llistxattr
+	.long sys_flistxattr
+	.long sys_removexattr	/* 235 */
+	.long sys_lremovexattr
+	.long sys_fremovexattr
+	.long sys_tkill
+	.long sys_sendfile64
+	.long sys_futex		/* 240 */
+	.long sys_sched_setaffinity
+	.long sys_sched_getaffinity
+	.long sys_ni_syscall
+	.long sys_ni_syscall
+	.long sys_io_setup	/* 245 */
+	.long sys_io_destroy
+	.long sys_io_getevents
+	.long sys_io_submit
+	.long sys_io_cancel
+	.long sys_fadvise64	/* 250 */
+	.long sys_ni_syscall
+	.long sys_exit_group
+	.long sys_lookup_dcookie
+	.long sys_epoll_create
+	.long sys_epoll_ctl	/* 255 */
+	.long sys_epoll_wait
+ 	.long sys_remap_file_pages
+ 	.long sys_set_tid_address
+ 	.long sys_timer_create
+ 	.long sys_timer_settime		/* 260 */
+ 	.long sys_timer_gettime
+ 	.long sys_timer_getoverrun
+ 	.long sys_timer_delete
+ 	.long sys_clock_settime
+ 	.long sys_clock_gettime		/* 265 */
+ 	.long sys_clock_getres
+ 	.long sys_clock_nanosleep
+	.long sys_statfs64
+	.long sys_fstatfs64
+	.long sys_tgkill		/* 270 */
+	.long sys_utimes
+ 	.long sys_fadvise64_64_wrapper
+	.long sys_ni_syscall	/* Reserved for vserver */
+	.long sys_ni_syscall	/* Reserved for mbind */
+	.long sys_ni_syscall	/* 275 - get_mempolicy */
+	.long sys_ni_syscall	/* set_mempolicy */
+	.long sys_mq_open
+	.long sys_mq_unlink
+	.long sys_mq_timedsend
+	.long sys_mq_timedreceive       /* 280 */
+	.long sys_mq_notify
+	.long sys_mq_getsetattr
+	.long sys_kexec_load
+	.long sys_waitid
+	.long sys_ni_syscall		/* 285 */
+	.long sys_add_key
+	.long sys_request_key
+	.long sys_keyctl
+	.long sys_ioprio_set
+	.long sys_ioprio_get		/* 290 */
+	.long sys_inotify_init
+	.long sys_inotify_add_watch
+	.long sys_inotify_rm_watch
+	.long sys_migrate_pages
+	.long sys_openat		/* 295 */
+	.long sys_mkdirat
+	.long sys_mknodat
+	.long sys_fchownat
+	.long sys_futimesat
+	.long sys_fstatat64		/* 300 */
+	.long sys_unlinkat
+	.long sys_renameat
+	.long sys_linkat
+	.long sys_symlinkat
+	.long sys_readlinkat		/* 305 */
+	.long sys_fchmodat
+	.long sys_faccessat
+	.long sys_pselect6
+	.long sys_ppoll
+	.long sys_unshare		/* 310 */
+	.long sys_set_robust_list
+	.long sys_get_robust_list
+	.long sys_splice
+	.long sys_sync_file_range
+	.long sys_tee			/* 315 */
+	.long sys_vmsplice
diff --git a/arch/sh/kernel/time.c b/arch/sh/kernel/time.c
index a1589f85499d..149d9713eddf 100644
--- a/arch/sh/kernel/time.c
+++ b/arch/sh/kernel/time.c
@@ -3,13 +3,12 @@
  *
  *  Copyright (C) 1999  Tetsuya Okada & Niibe Yutaka
  *  Copyright (C) 2000  Philipp Rumpf <prumpf@tux.org>
- *  Copyright (C) 2002, 2003, 2004, 2005  Paul Mundt
+ *  Copyright (C) 2002 - 2006  Paul Mundt
  *  Copyright (C) 2002  M. R. Brown  <mrbrown@linux-sh.org>
  *
  *  Some code taken from i386 version.
  *    Copyright (C) 1991, 1992, 1995  Linus Torvalds
  */
-
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/init.h>
@@ -26,15 +25,20 @@ struct sys_timer *sys_timer;
 DEFINE_SPINLOCK(rtc_lock);
 EXPORT_SYMBOL(rtc_lock);
 
-/* XXX: Can we initialize this in a routine somewhere?  Dreamcast doesn't want
- * these routines anywhere... */
-#ifdef CONFIG_SH_RTC
-void (*rtc_get_time)(struct timespec *) = sh_rtc_gettimeofday;
-int (*rtc_set_time)(const time_t) = sh_rtc_settimeofday;
-#else
-void (*rtc_get_time)(struct timespec *);
-int (*rtc_set_time)(const time_t);
-#endif
+/* Dummy RTC ops */
+static void null_rtc_get_time(struct timespec *tv)
+{
+	tv->tv_sec = mktime(2000, 1, 1, 0, 0, 0);
+	tv->tv_nsec = 0;
+}
+
+static int null_rtc_set_time(const time_t secs)
+{
+	return 0;
+}
+
+void (*rtc_sh_get_time)(struct timespec *) = null_rtc_get_time;
+int (*rtc_sh_set_time)(const time_t) = null_rtc_set_time;
 
 /*
  * Scheduler clock - returns current time in nanosec units.
@@ -70,7 +74,6 @@ void do_gettimeofday(struct timeval *tv)
 	tv->tv_sec = sec;
 	tv->tv_usec = usec;
 }
-
 EXPORT_SYMBOL(do_gettimeofday);
 
 int do_settimeofday(struct timespec *tv)
@@ -103,7 +106,6 @@ int do_settimeofday(struct timespec *tv)
 
 	return 0;
 }
-
 EXPORT_SYMBOL(do_settimeofday);
 
 /* last time the RTC clock got updated */
@@ -135,7 +137,7 @@ void handle_timer_tick(struct pt_regs *regs)
 	    xtime.tv_sec > last_rtc_update + 660 &&
 	    (xtime.tv_nsec / 1000) >= 500000 - ((unsigned) TICK_SIZE) / 2 &&
 	    (xtime.tv_nsec / 1000) <= 500000 + ((unsigned) TICK_SIZE) / 2) {
-		if (rtc_set_time(xtime.tv_sec) == 0)
+		if (rtc_sh_set_time(xtime.tv_sec) == 0)
 			last_rtc_update = xtime.tv_sec;
 		else
 			/* do it again in 60s */
@@ -143,8 +145,33 @@ void handle_timer_tick(struct pt_regs *regs)
 	}
 }
 
+#ifdef CONFIG_PM
+int timer_suspend(struct sys_device *dev, pm_message_t state)
+{
+	struct sys_timer *sys_timer = container_of(dev, struct sys_timer, dev);
+
+	sys_timer->ops->stop();
+
+	return 0;
+}
+
+int timer_resume(struct sys_device *dev)
+{
+	struct sys_timer *sys_timer = container_of(dev, struct sys_timer, dev);
+
+	sys_timer->ops->start();
+
+	return 0;
+}
+#else
+#define timer_suspend NULL
+#define timer_resume NULL
+#endif
+
 static struct sysdev_class timer_sysclass = {
 	set_kset_name("timer"),
+	.suspend = timer_suspend,
+	.resume	 = timer_resume,
 };
 
 static int __init timer_init_sysfs(void)
@@ -156,7 +183,6 @@ static int __init timer_init_sysfs(void)
 	sys_timer->dev.cls = &timer_sysclass;
 	return sysdev_register(&sys_timer->dev);
 }
-
 device_initcall(timer_init_sysfs);
 
 void (*board_time_init)(void);
@@ -168,15 +194,9 @@ void __init time_init(void)
 
 	clk_init();
 
-	if (rtc_get_time) {
-		rtc_get_time(&xtime);
-	} else {
-		xtime.tv_sec = mktime(2000, 1, 1, 0, 0, 0);
-		xtime.tv_nsec = 0;
-	}
-
-        set_normalized_timespec(&wall_to_monotonic,
-                                -xtime.tv_sec, -xtime.tv_nsec);
+	rtc_sh_get_time(&xtime);
+	set_normalized_timespec(&wall_to_monotonic,
+				-xtime.tv_sec, -xtime.tv_nsec);
 
 	/*
 	 * Find the timer to use as the system timer, it will be
diff --git a/arch/sh/kernel/timers/timer-tmu.c b/arch/sh/kernel/timers/timer-tmu.c
index d4212add53b2..205816fcf0da 100644
--- a/arch/sh/kernel/timers/timer-tmu.c
+++ b/arch/sh/kernel/timers/timer-tmu.c
@@ -132,17 +132,17 @@ static unsigned long tmu_timer_get_frequency(void)
 	ctrl_outl(0xffffffff, TMU0_TCOR);
 	ctrl_outl(0xffffffff, TMU0_TCNT);
 
-	rtc_get_time(&ts2);
+	rtc_sh_get_time(&ts2);
 
 	do {
-		rtc_get_time(&ts1);
+		rtc_sh_get_time(&ts1);
 	} while (ts1.tv_nsec == ts2.tv_nsec && ts1.tv_sec == ts2.tv_sec);
 
 	/* actually start the timer */
 	ctrl_outb(TMU_TSTR_INIT, TMU_TSTR);
 
 	do {
-		rtc_get_time(&ts2);
+		rtc_sh_get_time(&ts2);
 	} while (ts1.tv_nsec == ts2.tv_nsec && ts1.tv_sec == ts2.tv_sec);
 
 	freq = 0xffffffff - ctrl_inl(TMU0_TCNT);
@@ -188,6 +188,18 @@ static struct clk tmu0_clk = {
 	.ops		= &tmu_clk_ops,
 };
 
+static int tmu_timer_start(void)
+{
+	ctrl_outb(TMU_TSTR_INIT, TMU_TSTR);
+	return 0;
+}
+
+static int tmu_timer_stop(void)
+{
+	ctrl_outb(0, TMU_TSTR);
+	return 0;
+}
+
 static int tmu_timer_init(void)
 {
 	unsigned long interval;
@@ -197,7 +209,7 @@ static int tmu_timer_init(void)
 	tmu0_clk.parent = clk_get("module_clk");
 
 	/* Start TMU0 */
-	ctrl_outb(0, TMU_TSTR);
+	tmu_timer_stop();
 #if !defined(CONFIG_CPU_SUBTYPE_SH7300) && !defined(CONFIG_CPU_SUBTYPE_SH7760)
 	ctrl_outb(TMU_TOCR_INIT, TMU_TOCR);
 #endif
@@ -211,13 +223,15 @@ static int tmu_timer_init(void)
 	ctrl_outl(interval, TMU0_TCOR);
 	ctrl_outl(interval, TMU0_TCNT);
 
-	ctrl_outb(TMU_TSTR_INIT, TMU_TSTR);
+	tmu_timer_start();
 
 	return 0;
 }
 
 struct sys_timer_ops tmu_timer_ops = {
 	.init		= tmu_timer_init,
+	.start		= tmu_timer_start,
+	.stop		= tmu_timer_stop,
 	.get_frequency	= tmu_timer_get_frequency,
 	.get_offset	= tmu_timer_get_offset,
 };
diff --git a/arch/sh/kernel/traps.c b/arch/sh/kernel/traps.c
index d9db1180f770..c2c597e09482 100644
--- a/arch/sh/kernel/traps.c
+++ b/arch/sh/kernel/traps.c
@@ -36,40 +36,15 @@
 
 #ifdef CONFIG_SH_KGDB
 #include <asm/kgdb.h>
-#define CHK_REMOTE_DEBUG(regs)                                               \
-{                                                                            \
-  if ((kgdb_debug_hook != (kgdb_debug_hook_t *) NULL) && (!user_mode(regs))) \
-  {                                                                          \
-    (*kgdb_debug_hook)(regs);                                                \
-  }                                                                          \
+#define CHK_REMOTE_DEBUG(regs)                 	\
+{       					\
+	if (kgdb_debug_hook && !user_mode(regs))\
+		(*kgdb_debug_hook)(regs);       \
 }
 #else
 #define CHK_REMOTE_DEBUG(regs)
 #endif
 
-#define DO_ERROR(trapnr, signr, str, name, tsk)				\
-asmlinkage void do_##name(unsigned long r4, unsigned long r5,		\
-			  unsigned long r6, unsigned long r7,		\
-			  struct pt_regs regs)				\
-{									\
-	unsigned long error_code;					\
- 									\
-	/* Check if it's a DSP instruction */				\
- 	if (is_dsp_inst(&regs)) {					\
-		/* Enable DSP mode, and restart instruction. */		\
-		regs.sr |= SR_DSP;					\
-		return;							\
-	}								\
-									\
-	asm volatile("stc	r2_bank, %0": "=r" (error_code));	\
-	local_irq_enable();						\
-	tsk->thread.error_code = error_code;				\
-	tsk->thread.trap_no = trapnr;					\
-        CHK_REMOTE_DEBUG(&regs);					\
-	force_sig(signr, tsk);						\
-	die_if_no_fixup(str,&regs,error_code);				\
-}
-
 #ifdef CONFIG_CPU_SH2
 #define TRAP_RESERVED_INST	4
 #define TRAP_ILLEGAL_SLOT_INST	6
@@ -86,7 +61,7 @@ asmlinkage void do_##name(unsigned long r4, unsigned long r5,		\
 #define VMALLOC_OFFSET (8*1024*1024)
 #define MODULE_RANGE (8*1024*1024)
 
-spinlock_t die_lock;
+DEFINE_SPINLOCK(die_lock);
 
 void die(const char * str, struct pt_regs * regs, long err)
 {
@@ -575,8 +550,117 @@ int is_dsp_inst(struct pt_regs *regs)
 #define is_dsp_inst(regs)	(0)
 #endif /* CONFIG_SH_DSP */
 
-DO_ERROR(TRAP_RESERVED_INST, SIGILL, "reserved instruction", reserved_inst, current)
-DO_ERROR(TRAP_ILLEGAL_SLOT_INST, SIGILL, "illegal slot instruction", illegal_slot_inst, current)
+extern int do_fpu_inst(unsigned short, struct pt_regs*);
+
+asmlinkage void do_reserved_inst(unsigned long r4, unsigned long r5,
+				unsigned long r6, unsigned long r7,
+				struct pt_regs regs)
+{
+	unsigned long error_code;
+	struct task_struct *tsk = current;
+
+#ifdef CONFIG_SH_FPU_EMU
+	unsigned short inst;
+	int err;
+
+	get_user(inst, (unsigned short*)regs.pc);
+
+	err = do_fpu_inst(inst, &regs);
+	if (!err) {
+		regs.pc += 2;
+		return;
+	}
+	/* not a FPU inst. */
+#endif
+
+#ifdef CONFIG_SH_DSP
+	/* Check if it's a DSP instruction */
+ 	if (is_dsp_inst(&regs)) {
+		/* Enable DSP mode, and restart instruction. */
+		regs.sr |= SR_DSP;
+		return;
+	}
+#endif
+
+	asm volatile("stc	r2_bank, %0": "=r" (error_code));
+	local_irq_enable();
+	tsk->thread.error_code = error_code;
+	tsk->thread.trap_no = TRAP_RESERVED_INST;
+	CHK_REMOTE_DEBUG(&regs);
+	force_sig(SIGILL, tsk);
+	die_if_no_fixup("reserved instruction", &regs, error_code);
+}
+
+#ifdef CONFIG_SH_FPU_EMU
+static int emulate_branch(unsigned short inst, struct pt_regs* regs)
+{
+	/*
+	 * bfs: 8fxx: PC+=d*2+4;
+	 * bts: 8dxx: PC+=d*2+4;
+	 * bra: axxx: PC+=D*2+4;
+	 * bsr: bxxx: PC+=D*2+4  after PR=PC+4;
+	 * braf:0x23: PC+=Rn*2+4;
+	 * bsrf:0x03: PC+=Rn*2+4 after PR=PC+4;
+	 * jmp: 4x2b: PC=Rn;
+	 * jsr: 4x0b: PC=Rn      after PR=PC+4;
+	 * rts: 000b: PC=PR;
+	 */
+	if ((inst & 0xfd00) == 0x8d00) {
+		regs->pc += SH_PC_8BIT_OFFSET(inst);
+		return 0;
+	}
+
+	if ((inst & 0xe000) == 0xa000) {
+		regs->pc += SH_PC_12BIT_OFFSET(inst);
+		return 0;
+	}
+
+	if ((inst & 0xf0df) == 0x0003) {
+		regs->pc += regs->regs[(inst & 0x0f00) >> 8] + 4;
+		return 0;
+	}
+
+	if ((inst & 0xf0df) == 0x400b) {
+		regs->pc = regs->regs[(inst & 0x0f00) >> 8];
+		return 0;
+	}
+
+	if ((inst & 0xffff) == 0x000b) {
+		regs->pc = regs->pr;
+		return 0;
+	}
+
+	return 1;
+}
+#endif
+
+asmlinkage void do_illegal_slot_inst(unsigned long r4, unsigned long r5,
+				unsigned long r6, unsigned long r7,
+				struct pt_regs regs)
+{
+	unsigned long error_code;
+	struct task_struct *tsk = current;
+#ifdef CONFIG_SH_FPU_EMU
+	unsigned short inst;
+
+	get_user(inst, (unsigned short *)regs.pc + 1);
+	if (!do_fpu_inst(inst, &regs)) {
+		get_user(inst, (unsigned short *)regs.pc);
+		if (!emulate_branch(inst, &regs))
+			return;
+		/* fault in branch.*/
+	}
+	/* not a FPU inst. */
+#endif
+
+	asm volatile("stc	r2_bank, %0": "=r" (error_code));
+	local_irq_enable();
+	tsk->thread.error_code = error_code;
+	tsk->thread.trap_no = TRAP_RESERVED_INST;
+	CHK_REMOTE_DEBUG(&regs);
+	force_sig(SIGILL, tsk);
+	die_if_no_fixup("illegal slot instruction", &regs, error_code);
+}
 
 asmlinkage void do_exception_error(unsigned long r4, unsigned long r5,
 				   unsigned long r6, unsigned long r7,
@@ -634,14 +718,16 @@ void __init trap_init(void)
 	exception_handling_table[TRAP_ILLEGAL_SLOT_INST]
 		= (void *)do_illegal_slot_inst;
 
-#ifdef CONFIG_CPU_SH4
-	if (!(cpu_data->flags & CPU_HAS_FPU)) {
-		/* For SH-4 lacking an FPU, treat floating point instructions
-		   as reserved. */
-		/* entry 64 corresponds to EXPEVT=0x800 */
-		exception_handling_table[64] = (void *)do_reserved_inst;
-		exception_handling_table[65] = (void *)do_illegal_slot_inst;
-	}
+#if defined(CONFIG_CPU_SH4) && !defined(CONFIG_SH_FPU) || \
+    defined(CONFIG_SH_FPU_EMU)
+	/*
+	 * For SH-4 lacking an FPU, treat floating point instructions as
+	 * reserved. They'll be handled in the math-emu case, or faulted on
+	 * otherwise.
+	 */
+	/* entry 64 corresponds to EXPEVT=0x800 */
+	exception_handling_table[64] = (void *)do_reserved_inst;
+	exception_handling_table[65] = (void *)do_illegal_slot_inst;
 #endif
 		
 	/* Setup VBR for boot cpu */
@@ -655,20 +741,12 @@ void show_stack(struct task_struct *tsk, unsigned long *sp)
 	unsigned long module_end = VMALLOC_END;
 	int i = 1;
 
-	if (tsk && !sp) {
+	if (!tsk)
+		tsk = current;
+	if (tsk == current)
+		sp = (unsigned long *)current_stack_pointer;
+	else
 		sp = (unsigned long *)tsk->thread.sp;
-	}
-
-	if (!sp) {
-		__asm__ __volatile__ (
-			"mov r15, %0\n\t"
-			"stc r7_bank, %1\n\t"
-			: "=r" (module_start),
-			  "=r" (module_end)
-		);
-		
-		sp = (unsigned long *)module_start;
-	}
 
 	stack = sp;
 
diff --git a/arch/sh/kernel/vmlinux.lds.S b/arch/sh/kernel/vmlinux.lds.S
index 95fdd9135fcf..5eb930918186 100644
--- a/arch/sh/kernel/vmlinux.lds.S
+++ b/arch/sh/kernel/vmlinux.lds.S
@@ -2,6 +2,7 @@
  * ld script to make SuperH Linux kernel
  * Written by Niibe Yutaka
  */
+#include <asm/thread_info.h>
 #include <asm-generic/vmlinux.lds.h>
 
 #ifdef CONFIG_CPU_LITTLE_ENDIAN
@@ -13,7 +14,7 @@ OUTPUT_ARCH(sh)
 ENTRY(_start)
 SECTIONS
 {
-  . = 0x80000000 + CONFIG_MEMORY_START + CONFIG_ZERO_PAGE_OFFSET;
+  . = CONFIG_PAGE_OFFSET + CONFIG_MEMORY_START + CONFIG_ZERO_PAGE_OFFSET;
   _text = .;			/* Text and read-only data */
   text = .;			/* Text and read-only data */
   .empty_zero_page : {
@@ -40,16 +41,16 @@ SECTIONS
 	*(.data)
 
  	 /* Align the initial ramdisk image (INITRD) on page boundaries. */
- 	 . = ALIGN(4096);
+ 	 . = ALIGN(PAGE_SIZE);
  	 __rd_start = .;
  	 *(.initrd)
- 	 . = ALIGN(4096);
+ 	 . = ALIGN(PAGE_SIZE);
  	 __rd_end = .;
 
 	CONSTRUCTORS
 	}
 
-  . = ALIGN(4096);
+  . = ALIGN(PAGE_SIZE);
   .data.page_aligned : { *(.data.idt) }
 
   . = ALIGN(32);
@@ -60,12 +61,10 @@ SECTIONS
 
   _edata = .;			/* End of data section */
 
-  . = ALIGN(8192);		/* init_task */
+  . = ALIGN(THREAD_SIZE);		/* init_task */
   .data.init_task : { *(.data.init_task) }
-  /* stack */
-  .stack : { stack = .;  _stack = .; }
 
-  . = ALIGN(4096);		/* Init code and data */
+  . = ALIGN(PAGE_SIZE);		/* Init code and data */
   __init_begin = .;
   _sinittext = .;
   .init.text : { *(.init.text) }
@@ -96,7 +95,7 @@ SECTIONS
   __machvec_start = .;
   .init.machvec : { *(.init.machvec) }
   __machvec_end = .;
-  . = ALIGN(4096);
+  . = ALIGN(PAGE_SIZE);
   __init_end = .;
 
   . = ALIGN(4);
diff --git a/arch/sh/kernel/vsyscall/Makefile b/arch/sh/kernel/vsyscall/Makefile
new file mode 100644
index 000000000000..4bbce1cfa359
--- /dev/null
+++ b/arch/sh/kernel/vsyscall/Makefile
@@ -0,0 +1,36 @@
+obj-y += vsyscall.o vsyscall-syscall.o
+
+$(obj)/vsyscall-syscall.o: \
+	$(foreach F,trapa,$(obj)/vsyscall-$F.so)
+
+# Teach kbuild about targets
+targets += $(foreach F,trapa,vsyscall-$F.o vsyscall-$F.so)
+targets += vsyscall-note.o vsyscall.lds
+
+# The DSO images are built using a special linker script
+quiet_cmd_syscall = SYSCALL $@
+      cmd_syscall = $(CC) -nostdlib $(SYSCFLAGS_$(@F)) \
+			   -Wl,-T,$(filter-out FORCE,$^) -o $@
+
+export CPPFLAGS_vsyscall.lds += -P -C -Ush
+
+vsyscall-flags = -shared -s -Wl,-soname=linux-gate.so.1 \
+		$(call ld-option, -Wl$(comma)--hash-style=sysv)
+
+SYSCFLAGS_vsyscall-trapa.so	= $(vsyscall-flags)
+
+$(obj)/vsyscall-trapa.so: \
+$(obj)/vsyscall-%.so: $(src)/vsyscall.lds $(obj)/vsyscall-%.o FORCE
+	$(call if_changed,syscall)
+
+# We also create a special relocatable object that should mirror the symbol
+# table and layout of the linked DSO.  With ld -R we can then refer to
+# these symbols in the kernel code rather than hand-coded addresses.
+extra-y += vsyscall-syms.o
+$(obj)/built-in.o: $(obj)/vsyscall-syms.o
+$(obj)/built-in.o: ld_flags += -R $(obj)/vsyscall-syms.o
+
+SYSCFLAGS_vsyscall-syms.o = -r
+$(obj)/vsyscall-syms.o: $(src)/vsyscall.lds \
+			$(obj)/vsyscall-trapa.o $(obj)/vsyscall-note.o FORCE
+	$(call if_changed,syscall)
diff --git a/arch/sh/kernel/vsyscall/vsyscall-note.S b/arch/sh/kernel/vsyscall/vsyscall-note.S
new file mode 100644
index 000000000000..d4b5be4f3d5f
--- /dev/null
+++ b/arch/sh/kernel/vsyscall/vsyscall-note.S
@@ -0,0 +1,25 @@
+/*
+ * This supplies .note.* sections to go into the PT_NOTE inside the vDSO text.
+ * Here we can supply some information useful to userland.
+ */
+
+#include <linux/uts.h>
+#include <linux/version.h>
+
+#define ASM_ELF_NOTE_BEGIN(name, flags, vendor, type)			      \
+	.section name, flags;						      \
+	.balign 4;							      \
+	.long 1f - 0f;		/* name length */			      \
+	.long 3f - 2f;		/* data length */			      \
+	.long type;		/* note type */				      \
+0:	.asciz vendor;		/* vendor name */			      \
+1:	.balign 4;							      \
+2:
+
+#define ASM_ELF_NOTE_END						      \
+3:	.balign 4;		/* pad out section */			      \
+	.previous
+
+	ASM_ELF_NOTE_BEGIN(".note.kernel-version", "a", UTS_SYSNAME, 0)
+	.long LINUX_VERSION_CODE
+	ASM_ELF_NOTE_END
diff --git a/arch/sh/kernel/vsyscall/vsyscall-sigreturn.S b/arch/sh/kernel/vsyscall/vsyscall-sigreturn.S
new file mode 100644
index 000000000000..555a64f124ca
--- /dev/null
+++ b/arch/sh/kernel/vsyscall/vsyscall-sigreturn.S
@@ -0,0 +1,39 @@
+#include <asm/unistd.h>
+
+	.text
+	.balign 32
+	.globl __kernel_sigreturn
+	.type __kernel_sigreturn,@function
+__kernel_sigreturn:
+.LSTART_sigreturn:
+	mov.w	1f, r3
+	trapa	#0x10
+	or	r0, r0
+	or	r0, r0
+	or	r0, r0
+	or	r0, r0
+	or	r0, r0
+
+1:	.short	__NR_sigreturn
+.LEND_sigreturn:
+	.size __kernel_sigreturn,.-.LSTART_sigreturn
+
+	.balign 32
+	.globl __kernel_rt_sigreturn
+	.type __kernel_rt_sigreturn,@function
+__kernel_rt_sigreturn:
+.LSTART_rt_sigreturn:
+	mov.w	1f, r3
+	trapa	#0x10
+	or	r0, r0
+	or	r0, r0
+	or	r0, r0
+	or	r0, r0
+	or	r0, r0
+
+1:	.short	__NR_rt_sigreturn
+.LEND_rt_sigreturn:
+	.size __kernel_rt_sigreturn,.-.LSTART_rt_sigreturn
+
+	.section .eh_frame,"a",@progbits
+	.previous
diff --git a/arch/sh/kernel/vsyscall/vsyscall-syscall.S b/arch/sh/kernel/vsyscall/vsyscall-syscall.S
new file mode 100644
index 000000000000..c2ac7f0282b3
--- /dev/null
+++ b/arch/sh/kernel/vsyscall/vsyscall-syscall.S
@@ -0,0 +1,10 @@
+#include <linux/init.h>
+
+__INITDATA
+
+	.globl vsyscall_trapa_start, vsyscall_trapa_end
+vsyscall_trapa_start:
+	.incbin "arch/sh/kernel/vsyscall/vsyscall-trapa.so"
+vsyscall_trapa_end:
+
+__FINIT
diff --git a/arch/sh/kernel/vsyscall/vsyscall-trapa.S b/arch/sh/kernel/vsyscall/vsyscall-trapa.S
new file mode 100644
index 000000000000..3b6eb34c43fa
--- /dev/null
+++ b/arch/sh/kernel/vsyscall/vsyscall-trapa.S
@@ -0,0 +1,42 @@
+	.text
+	.globl __kernel_vsyscall
+	.type __kernel_vsyscall,@function
+__kernel_vsyscall:
+.LSTART_vsyscall:
+	/* XXX: We'll have to do something here once we opt to use the vDSO
+	 * page for something other than the signal trampoline.. as well as
+	 * fill out .eh_frame -- PFM. */
+.LEND_vsyscall:
+	.size __kernel_vsyscall,.-.LSTART_vsyscall
+	.previous
+
+	.section .eh_frame,"a",@progbits
+.LCIE:
+	.ualong	.LCIE_end - .LCIE_start
+.LCIE_start:
+	.ualong	0		/* CIE ID */
+	.byte	0x1		/* Version number */
+	.string	"zRS"		/* NUL-terminated augmentation string */
+	.uleb128 0x1		/* Code alignment factor */
+	.sleb128 -4		/* Data alignment factor */
+	.byte	0x11		/* Return address register column */
+				/* Augmentation length and data (none) */
+	.byte	0xc		/* DW_CFA_def_cfa */
+	.uleb128 0xf		/* r15 */
+	.uleb128 0x0		/* offset 0 */
+
+	.align 2
+.LCIE_end:
+
+	.ualong	.LFDE_end-.LFDE_start	/* Length FDE */
+.LFDE_start:
+	.ualong	.LCIE			/* CIE pointer */
+	.ualong	.LSTART_vsyscall-.	/* start address */
+	.ualong	.LEND_vsyscall-.LSTART_vsyscall
+	.uleb128 0
+	.align 2
+.LFDE_end:
+	.previous
+
+/* Get the common code for the sigreturn entry points */
+#include "vsyscall-sigreturn.S"
diff --git a/arch/sh/kernel/vsyscall/vsyscall.c b/arch/sh/kernel/vsyscall/vsyscall.c
new file mode 100644
index 000000000000..075d6cc1a2d7
--- /dev/null
+++ b/arch/sh/kernel/vsyscall/vsyscall.c
@@ -0,0 +1,150 @@
+/*
+ * arch/sh/kernel/vsyscall.c
+ *
+ *  Copyright (C) 2006 Paul Mundt
+ *
+ * vDSO randomization
+ * Copyright(C) 2005-2006, Red Hat, Inc., Ingo Molnar
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/mm.h>
+#include <linux/slab.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/gfp.h>
+#include <linux/module.h>
+#include <linux/elf.h>
+
+/*
+ * Should the kernel map a VDSO page into processes and pass its
+ * address down to glibc upon exec()?
+ */
+unsigned int __read_mostly vdso_enabled = 1;
+EXPORT_SYMBOL_GPL(vdso_enabled);
+
+static int __init vdso_setup(char *s)
+{
+	vdso_enabled = simple_strtoul(s, NULL, 0);
+	return 1;
+}
+__setup("vdso=", vdso_setup);
+
+/*
+ * These symbols are defined by vsyscall.o to mark the bounds
+ * of the ELF DSO images included therein.
+ */
+extern const char vsyscall_trapa_start, vsyscall_trapa_end;
+static void *syscall_page;
+
+int __init vsyscall_init(void)
+{
+	syscall_page = (void *)get_zeroed_page(GFP_ATOMIC);
+
+	/*
+	 * XXX: Map this page to a fixmap entry if we get around
+	 * to adding the page to ELF core dumps
+	 */
+
+	memcpy(syscall_page,
+	       &vsyscall_trapa_start,
+	       &vsyscall_trapa_end - &vsyscall_trapa_start);
+
+	return 0;
+}
+
+static struct page *syscall_vma_nopage(struct vm_area_struct *vma,
+				       unsigned long address, int *type)
+{
+	unsigned long offset = address - vma->vm_start;
+	struct page *page;
+
+	if (address < vma->vm_start || address > vma->vm_end)
+		return NOPAGE_SIGBUS;
+
+	page = virt_to_page(syscall_page + offset);
+
+	get_page(page);
+
+	return page;
+}
+
+/* Prevent VMA merging */
+static void syscall_vma_close(struct vm_area_struct *vma)
+{
+}
+
+static struct vm_operations_struct syscall_vm_ops = {
+	.nopage	= syscall_vma_nopage,
+	.close	= syscall_vma_close,
+};
+
+/* Setup a VMA at program startup for the vsyscall page */
+int arch_setup_additional_pages(struct linux_binprm *bprm,
+				int executable_stack)
+{
+	struct vm_area_struct *vma;
+	struct mm_struct *mm = current->mm;
+	unsigned long addr;
+	int ret;
+
+	down_write(&mm->mmap_sem);
+	addr = get_unmapped_area(NULL, 0, PAGE_SIZE, 0, 0);
+	if (IS_ERR_VALUE(addr)) {
+		ret = addr;
+		goto up_fail;
+	}
+
+	vma = kmem_cache_zalloc(vm_area_cachep, SLAB_KERNEL);
+	if (!vma) {
+		ret = -ENOMEM;
+		goto up_fail;
+	}
+
+	vma->vm_start = addr;
+	vma->vm_end = addr + PAGE_SIZE;
+	/* MAYWRITE to allow gdb to COW and set breakpoints */
+	vma->vm_flags = VM_READ|VM_EXEC|VM_MAYREAD|VM_MAYEXEC|VM_MAYWRITE;
+	vma->vm_flags |= mm->def_flags;
+	vma->vm_page_prot = protection_map[vma->vm_flags & 7];
+	vma->vm_ops = &syscall_vm_ops;
+	vma->vm_mm = mm;
+
+	ret = insert_vm_struct(mm, vma);
+	if (unlikely(ret)) {
+		kmem_cache_free(vm_area_cachep, vma);
+		goto up_fail;
+	}
+
+	current->mm->context.vdso = (void *)addr;
+
+	mm->total_vm++;
+up_fail:
+	up_write(&mm->mmap_sem);
+	return ret;
+}
+
+const char *arch_vma_name(struct vm_area_struct *vma)
+{
+	if (vma->vm_mm && vma->vm_start == (long)vma->vm_mm->context.vdso)
+		return "[vdso]";
+
+	return NULL;
+}
+
+struct vm_area_struct *get_gate_vma(struct task_struct *task)
+{
+	return NULL;
+}
+
+int in_gate_area(struct task_struct *task, unsigned long address)
+{
+	return 0;
+}
+
+int in_gate_area_no_task(unsigned long address)
+{
+	return 0;
+}
diff --git a/arch/sh/kernel/vsyscall/vsyscall.lds.S b/arch/sh/kernel/vsyscall/vsyscall.lds.S
new file mode 100644
index 000000000000..b13c3d439fee
--- /dev/null
+++ b/arch/sh/kernel/vsyscall/vsyscall.lds.S
@@ -0,0 +1,74 @@
+/*
+ * Linker script for vsyscall DSO.  The vsyscall page is an ELF shared
+ * object prelinked to its virtual address, and with only one read-only
+ * segment (that fits in one page).  This script controls its layout.
+ */
+#include <asm/asm-offsets.h>
+
+#ifdef CONFIG_CPU_LITTLE_ENDIAN
+OUTPUT_FORMAT("elf32-sh-linux", "elf32-sh-linux", "elf32-sh-linux")
+#else
+OUTPUT_FORMAT("elf32-shbig-linux", "elf32-shbig-linux", "elf32-shbig-linux")
+#endif
+OUTPUT_ARCH(sh)
+
+/* The ELF entry point can be used to set the AT_SYSINFO value.  */
+ENTRY(__kernel_vsyscall);
+
+SECTIONS
+{
+  . = SIZEOF_HEADERS;
+
+  .hash           : { *(.hash) }		:text
+  .gnu.hash       : { *(.gnu.hash) }
+  .dynsym         : { *(.dynsym) }
+  .dynstr         : { *(.dynstr) }
+  .gnu.version    : { *(.gnu.version) }
+  .gnu.version_d  : { *(.gnu.version_d) }
+  .gnu.version_r  : { *(.gnu.version_r) }
+
+  /* This linker script is used both with -r and with -shared.
+     For the layouts to match, we need to skip more than enough
+     space for the dynamic symbol table et al.  If this amount
+     is insufficient, ld -shared will barf.  Just increase it here.  */
+  . = 0x400;
+
+  .text           : { *(.text) }		:text =0x90909090
+  .note		  : { *(.note.*) }		:text :note
+  .eh_frame_hdr   : { *(.eh_frame_hdr) }	:text :eh_frame_hdr
+  .eh_frame       : { KEEP (*(.eh_frame)) }	:text
+  .dynamic        : { *(.dynamic) }		:text :dynamic
+  .useless        : {
+  	*(.got.plt) *(.got)
+	*(.data .data.* .gnu.linkonce.d.*)
+	*(.dynbss)
+	*(.bss .bss.* .gnu.linkonce.b.*)
+  }						:text
+}
+
+/*
+ * We must supply the ELF program headers explicitly to get just one
+ * PT_LOAD segment, and set the flags explicitly to make segments read-only.
+ */
+PHDRS
+{
+  text PT_LOAD FILEHDR PHDRS FLAGS(5); /* PF_R|PF_X */
+  dynamic PT_DYNAMIC FLAGS(4); /* PF_R */
+  note PT_NOTE FLAGS(4); /* PF_R */
+  eh_frame_hdr 0x6474e550; /* PT_GNU_EH_FRAME, but ld doesn't match the name */
+}
+
+/*
+ * This controls what symbols we export from the DSO.
+ */
+VERSION
+{
+  LINUX_2.6 {
+    global:
+    	__kernel_vsyscall;
+    	__kernel_sigreturn;
+    	__kernel_rt_sigreturn;
+
+    local: *;
+  };
+}