sh: Kill off the rest of arch/sh64/kernel/.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
diff --git a/arch/sh/drivers/pci/pci-sh5.c b/arch/sh/drivers/pci/pci-sh5.c
new file mode 100644
index 0000000..b4d9534
--- /dev/null
+++ b/arch/sh/drivers/pci/pci-sh5.c
@@ -0,0 +1,536 @@
+/*
+ * Copyright (C) 2001 David J. Mckay (david.mckay@st.com)
+ * Copyright (C) 2003, 2004 Paul Mundt
+ * Copyright (C) 2004 Richard Curnow
+ *
+ * May be copied or modified under the terms of the GNU General Public
+ * License.  See linux/COPYING for more information.
+ *
+ * Support functions for the SH5 PCI hardware.
+ */
+
+#include <linux/kernel.h>
+#include <linux/rwsem.h>
+#include <linux/smp.h>
+#include <linux/interrupt.h>
+#include <linux/init.h>
+#include <linux/errno.h>
+#include <linux/pci.h>
+#include <linux/delay.h>
+#include <linux/types.h>
+#include <asm/pci.h>
+#include <linux/irq.h>
+
+#include <asm/io.h>
+#include <asm/hardware.h>
+#include "pci_sh5.h"
+
+static unsigned long pcicr_virt;
+unsigned long pciio_virt;
+
+static void __init pci_fixup_ide_bases(struct pci_dev *d)
+{
+	int i;
+
+	/*
+	 * PCI IDE controllers use non-standard I/O port decoding, respect it.
+	 */
+	if ((d->class >> 8) != PCI_CLASS_STORAGE_IDE)
+		return;
+	printk("PCI: IDE base address fixup for %s\n", pci_name(d));
+	for(i=0; i<4; i++) {
+		struct resource *r = &d->resource[i];
+		if ((r->start & ~0x80) == 0x374) {
+			r->start |= 2;
+			r->end = r->start;
+		}
+	}
+}
+DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, pci_fixup_ide_bases);
+
+char * __devinit pcibios_setup(char *str)
+{
+	return str;
+}
+
+/* Rounds a number UP to the nearest power of two. Used for
+ * sizing the PCI window.
+ */
+static u32 __init r2p2(u32 num)
+{
+	int i = 31;
+	u32 tmp = num;
+
+	if (num == 0)
+		return 0;
+
+	do {
+		if (tmp & (1 << 31))
+			break;
+		i--;
+		tmp <<= 1;
+	} while (i >= 0);
+
+	tmp = 1 << i;
+	/* If the original number isn't a power of 2, round it up */
+	if (tmp != num)
+		tmp <<= 1;
+
+	return tmp;
+}
+
+extern unsigned long long memory_start, memory_end;
+
+int __init sh5pci_init(unsigned memStart, unsigned memSize)
+{
+	u32 lsr0;
+	u32 uval;
+
+	pcicr_virt = onchip_remap(SH5PCI_ICR_BASE, 1024, "PCICR");
+	if (!pcicr_virt) {
+		panic("Unable to remap PCICR\n");
+	}
+
+	pciio_virt = onchip_remap(SH5PCI_IO_BASE, 0x10000, "PCIIO");
+	if (!pciio_virt) {
+		panic("Unable to remap PCIIO\n");
+	}
+
+	pr_debug("Register base addres is 0x%08lx\n", pcicr_virt);
+
+	/* Clear snoop registers */
+        SH5PCI_WRITE(CSCR0, 0);
+        SH5PCI_WRITE(CSCR1, 0);
+
+	pr_debug("Wrote to reg\n");
+
+        /* Switch off interrupts */
+        SH5PCI_WRITE(INTM,  0);
+        SH5PCI_WRITE(AINTM, 0);
+        SH5PCI_WRITE(PINTM, 0);
+
+        /* Set bus active, take it out of reset */
+        uval = SH5PCI_READ(CR);
+
+	/* Set command Register */
+        SH5PCI_WRITE(CR, uval | CR_LOCK_MASK | CR_CFINT| CR_FTO | CR_PFE | CR_PFCS | CR_BMAM);
+
+	uval=SH5PCI_READ(CR);
+        pr_debug("CR is actually 0x%08x\n",uval);
+
+        /* Allow it to be a master */
+	/* NB - WE DISABLE I/O ACCESS to stop overlap */
+        /* set WAIT bit to enable stepping, an attempt to improve stability */
+	SH5PCI_WRITE_SHORT(CSR_CMD,
+			    PCI_COMMAND_MEMORY | PCI_COMMAND_MASTER | PCI_COMMAND_WAIT);
+
+        /*
+        ** Set translation mapping memory in order to convert the address
+        ** used for the main bus, to the PCI internal address.
+        */
+        SH5PCI_WRITE(MBR,0x40000000);
+
+        /* Always set the max size 512M */
+        SH5PCI_WRITE(MBMR, PCISH5_MEM_SIZCONV(512*1024*1024));
+
+        /*
+        ** I/O addresses are mapped at internal PCI specific address
+        ** as is described into the configuration bridge table.
+        ** These are changed to 0, to allow cards that have legacy
+        ** io such as vga to function correctly. We set the SH5 IOBAR to
+        ** 256K, which is a bit big as we can only have 64K of address space
+        */
+
+        SH5PCI_WRITE(IOBR,0x0);
+
+	pr_debug("PCI:Writing 0x%08x to IOBR\n",0);
+
+        /* Set up a 256K window. Totally pointless waste  of address space */
+        SH5PCI_WRITE(IOBMR,0);
+	pr_debug("PCI:Writing 0x%08x to IOBMR\n",0);
+
+	/* The SH5 has a HUGE 256K I/O region, which breaks the PCI spec. Ideally,
+         * we would want to map the I/O region somewhere, but it is so big this is not
+         * that easy!
+         */
+	SH5PCI_WRITE(CSR_IBAR0,~0);
+	/* Set memory size value */
+        memSize = memory_end - memory_start;
+
+        /* Now we set up the mbars so the PCI bus can see the memory of the machine */
+        if (memSize < (1024 * 1024)) {
+                printk(KERN_ERR "PCISH5: Ridiculous memory size of 0x%x?\n", memSize);
+                return -EINVAL;
+        }
+
+        /* Set LSR 0 */
+        lsr0 = (memSize > (512 * 1024 * 1024)) ? 0x1ff00001 : ((r2p2(memSize) - 0x100000) | 0x1);
+        SH5PCI_WRITE(LSR0, lsr0);
+
+	pr_debug("PCI:Writing 0x%08x to LSR0\n",lsr0);
+
+        /* Set MBAR 0 */
+        SH5PCI_WRITE(CSR_MBAR0, memory_start);
+        SH5PCI_WRITE(LAR0, memory_start);
+
+        SH5PCI_WRITE(CSR_MBAR1,0);
+        SH5PCI_WRITE(LAR1,0);
+        SH5PCI_WRITE(LSR1,0);
+
+	pr_debug("PCI:Writing 0x%08llx to CSR_MBAR0\n",memory_start);
+	pr_debug("PCI:Writing 0x%08llx to LAR0\n",memory_start);
+
+        /* Enable the PCI interrupts on the device */
+        SH5PCI_WRITE(INTM,  ~0);
+        SH5PCI_WRITE(AINTM, ~0);
+        SH5PCI_WRITE(PINTM, ~0);
+
+	pr_debug("Switching on all error interrupts\n");
+
+        return(0);
+}
+
+static int sh5pci_read(struct pci_bus *bus, unsigned int devfn, int where,
+			int size, u32 *val)
+{
+	SH5PCI_WRITE(PAR, CONFIG_CMD(bus, devfn, where));
+
+	switch (size) {
+		case 1:
+			*val = (u8)SH5PCI_READ_BYTE(PDR + (where & 3));
+			break;
+		case 2:
+			*val = (u16)SH5PCI_READ_SHORT(PDR + (where & 2));
+			break;
+		case 4:
+			*val = SH5PCI_READ(PDR);
+			break;
+	}
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static int sh5pci_write(struct pci_bus *bus, unsigned int devfn, int where,
+			 int size, u32 val)
+{
+	SH5PCI_WRITE(PAR, CONFIG_CMD(bus, devfn, where));
+
+	switch (size) {
+		case 1:
+			SH5PCI_WRITE_BYTE(PDR + (where & 3), (u8)val);
+			break;
+		case 2:
+			SH5PCI_WRITE_SHORT(PDR + (where & 2), (u16)val);
+			break;
+		case 4:
+			SH5PCI_WRITE(PDR, val);
+			break;
+	}
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static struct pci_ops pci_config_ops = {
+	.read =		sh5pci_read,
+	.write =	sh5pci_write,
+};
+
+/* Everything hangs off this */
+static struct pci_bus *pci_root_bus;
+
+
+static u8 __init no_swizzle(struct pci_dev *dev, u8 * pin)
+{
+	pr_debug("swizzle for dev %d on bus %d slot %d pin is %d\n",
+	         dev->devfn,dev->bus->number, PCI_SLOT(dev->devfn),*pin);
+	return PCI_SLOT(dev->devfn);
+}
+
+static inline u8 bridge_swizzle(u8 pin, u8 slot)
+{
+	return (((pin-1) + slot) % 4) + 1;
+}
+
+u8 __init common_swizzle(struct pci_dev *dev, u8 *pinp)
+{
+	if (dev->bus->number != 0) {
+		u8 pin = *pinp;
+		do {
+			pin = bridge_swizzle(pin, PCI_SLOT(dev->devfn));
+			/* Move up the chain of bridges. */
+			dev = dev->bus->self;
+		} while (dev->bus->self);
+		*pinp = pin;
+
+		/* The slot is the slot of the last bridge. */
+	}
+
+	return PCI_SLOT(dev->devfn);
+}
+
+/* This needs to be shunted out of here into the board specific bit */
+
+static int __init map_cayman_irq(struct pci_dev *dev, u8 slot, u8 pin)
+{
+	int result = -1;
+
+	/* The complication here is that the PCI IRQ lines from the Cayman's 2
+	   5V slots get into the CPU via a different path from the IRQ lines
+	   from the 3 3.3V slots.  Thus, we have to detect whether the card's
+	   interrupts go via the 5V or 3.3V path, i.e. the 'bridge swizzling'
+	   at the point where we cross from 5V to 3.3V is not the normal case.
+
+	   The added complication is that we don't know that the 5V slots are
+	   always bus 2, because a card containing a PCI-PCI bridge may be
+	   plugged into a 3.3V slot, and this changes the bus numbering.
+
+	   Also, the Cayman has an intermediate PCI bus that goes a custom
+	   expansion board header (and to the secondary bridge).  This bus has
+	   never been used in practice.
+
+	   The 1ary onboard PCI-PCI bridge is device 3 on bus 0
+	   The 2ary onboard PCI-PCI bridge is device 0 on the 2ary bus of the 1ary bridge.
+	   */
+
+	struct slot_pin {
+		int slot;
+		int pin;
+	} path[4];
+	int i=0;
+
+	while (dev->bus->number > 0) {
+
+		slot = path[i].slot = PCI_SLOT(dev->devfn);
+		pin = path[i].pin = bridge_swizzle(pin, slot);
+		dev = dev->bus->self;
+		i++;
+		if (i > 3) panic("PCI path to root bus too long!\n");
+	}
+
+	slot = PCI_SLOT(dev->devfn);
+	/* This is the slot on bus 0 through which the device is eventually
+	   reachable. */
+
+	/* Now work back up. */
+	if ((slot < 3) || (i == 0)) {
+		/* Bus 0 (incl. PCI-PCI bridge itself) : perform the final
+		   swizzle now. */
+		result = IRQ_INTA + bridge_swizzle(pin, slot) - 1;
+	} else {
+		i--;
+		slot = path[i].slot;
+		pin  = path[i].pin;
+		if (slot > 0) {
+			panic("PCI expansion bus device found - not handled!\n");
+		} else {
+			if (i > 0) {
+				/* 5V slots */
+				i--;
+				slot = path[i].slot;
+				pin  = path[i].pin;
+				/* 'pin' was swizzled earlier wrt slot, don't do it again. */
+				result = IRQ_P2INTA + (pin - 1);
+			} else {
+				/* IRQ for 2ary PCI-PCI bridge : unused */
+				result = -1;
+			}
+		}
+	}
+
+	return result;
+}
+
+static irqreturn_t pcish5_err_irq(int irq, void *dev_id)
+{
+	struct pt_regs *regs = get_irq_regs();
+	unsigned pci_int, pci_air, pci_cir, pci_aint;
+
+	pci_int = SH5PCI_READ(INT);
+	pci_cir = SH5PCI_READ(CIR);
+	pci_air = SH5PCI_READ(AIR);
+
+	if (pci_int) {
+		printk("PCI INTERRUPT (at %08llx)!\n", regs->pc);
+		printk("PCI INT -> 0x%x\n", pci_int & 0xffff);
+		printk("PCI AIR -> 0x%x\n", pci_air);
+		printk("PCI CIR -> 0x%x\n", pci_cir);
+		SH5PCI_WRITE(INT, ~0);
+	}
+
+	pci_aint = SH5PCI_READ(AINT);
+	if (pci_aint) {
+		printk("PCI ARB INTERRUPT!\n");
+		printk("PCI AINT -> 0x%x\n", pci_aint);
+		printk("PCI AIR -> 0x%x\n", pci_air);
+		printk("PCI CIR -> 0x%x\n", pci_cir);
+		SH5PCI_WRITE(AINT, ~0);
+	}
+
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t pcish5_serr_irq(int irq, void *dev_id)
+{
+	printk("SERR IRQ\n");
+
+	return IRQ_NONE;
+}
+
+static void __init
+pcibios_size_bridge(struct pci_bus *bus, struct resource *ior,
+		    struct resource *memr)
+{
+	struct resource io_res, mem_res;
+	struct pci_dev *dev;
+	struct pci_dev *bridge = bus->self;
+	struct list_head *ln;
+
+	if (!bridge)
+		return;	/* host bridge, nothing to do */
+
+	/* set reasonable default locations for pcibios_align_resource */
+	io_res.start = PCIBIOS_MIN_IO;
+	mem_res.start = PCIBIOS_MIN_MEM;
+
+	io_res.end = io_res.start;
+	mem_res.end = mem_res.start;
+
+	/* Collect information about how our direct children are layed out. */
+	for (ln=bus->devices.next; ln != &bus->devices; ln=ln->next) {
+		int i;
+		dev = pci_dev_b(ln);
+
+		/* Skip bridges for now */
+		if (dev->class >> 8 == PCI_CLASS_BRIDGE_PCI)
+			continue;
+
+		for (i = 0; i < PCI_NUM_RESOURCES; i++) {
+			struct resource res;
+			unsigned long size;
+
+			memcpy(&res, &dev->resource[i], sizeof(res));
+			size = res.end - res.start + 1;
+
+			if (res.flags & IORESOURCE_IO) {
+				res.start = io_res.end;
+				pcibios_align_resource(dev, &res, size, 0);
+				io_res.end = res.start + size;
+			} else if (res.flags & IORESOURCE_MEM) {
+				res.start = mem_res.end;
+				pcibios_align_resource(dev, &res, size, 0);
+				mem_res.end = res.start + size;
+			}
+		}
+	}
+
+	/* And for all of the subordinate busses. */
+	for (ln=bus->children.next; ln != &bus->children; ln=ln->next)
+		pcibios_size_bridge(pci_bus_b(ln), &io_res, &mem_res);
+
+	/* turn the ending locations into sizes (subtract start) */
+	io_res.end -= io_res.start;
+	mem_res.end -= mem_res.start;
+
+	/* Align the sizes up by bridge rules */
+	io_res.end = ALIGN(io_res.end, 4*1024) - 1;
+	mem_res.end = ALIGN(mem_res.end, 1*1024*1024) - 1;
+
+	/* Adjust the bridge's allocation requirements */
+	bridge->resource[0].end = bridge->resource[0].start + io_res.end;
+	bridge->resource[1].end = bridge->resource[1].start + mem_res.end;
+
+	bridge->resource[PCI_BRIDGE_RESOURCES].end =
+	    bridge->resource[PCI_BRIDGE_RESOURCES].start + io_res.end;
+	bridge->resource[PCI_BRIDGE_RESOURCES+1].end =
+	    bridge->resource[PCI_BRIDGE_RESOURCES+1].start + mem_res.end;
+
+	/* adjust parent's resource requirements */
+	if (ior) {
+		ior->end = ALIGN(ior->end, 4*1024);
+		ior->end += io_res.end;
+	}
+
+	if (memr) {
+		memr->end = ALIGN(memr->end, 1*1024*1024);
+		memr->end += mem_res.end;
+	}
+}
+
+static void __init pcibios_size_bridges(void)
+{
+	struct resource io_res, mem_res;
+
+	memset(&io_res, 0, sizeof(io_res));
+	memset(&mem_res, 0, sizeof(mem_res));
+
+	pcibios_size_bridge(pci_root_bus, &io_res, &mem_res);
+}
+
+static int __init pcibios_init(void)
+{
+        if (request_irq(IRQ_ERR, pcish5_err_irq,
+                        IRQF_DISABLED, "PCI Error",NULL) < 0) {
+                printk(KERN_ERR "PCISH5: Cannot hook PCI_PERR interrupt\n");
+                return -EINVAL;
+        }
+
+        if (request_irq(IRQ_SERR, pcish5_serr_irq,
+                        IRQF_DISABLED, "PCI SERR interrupt", NULL) < 0) {
+                printk(KERN_ERR "PCISH5: Cannot hook PCI_SERR interrupt\n");
+                return -EINVAL;
+        }
+
+	/* The pci subsystem needs to know where memory is and how much
+	 * of it there is. I've simply made these globals. A better mechanism
+	 * is probably needed.
+	 */
+	sh5pci_init(__pa(memory_start),
+		     __pa(memory_end) - __pa(memory_start));
+
+	pci_root_bus = pci_scan_bus(0, &pci_config_ops, NULL);
+	pcibios_size_bridges();
+	pci_assign_unassigned_resources();
+	pci_fixup_irqs(no_swizzle, map_cayman_irq);
+
+	return 0;
+}
+
+subsys_initcall(pcibios_init);
+
+void __devinit pcibios_fixup_bus(struct pci_bus *bus)
+{
+	struct pci_dev *dev = bus->self;
+	int i;
+
+#if 1
+	if(dev) {
+		for(i=0; i<3; i++) {
+			bus->resource[i] =
+				&dev->resource[PCI_BRIDGE_RESOURCES+i];
+			bus->resource[i]->name = bus->name;
+		}
+		bus->resource[0]->flags |= IORESOURCE_IO;
+		bus->resource[1]->flags |= IORESOURCE_MEM;
+
+		/* For now, propagate host limits to the bus;
+		 * we'll adjust them later. */
+
+#if 1
+		bus->resource[0]->end = 64*1024 - 1 ;
+		bus->resource[1]->end = PCIBIOS_MIN_MEM+(256*1024*1024)-1;
+		bus->resource[0]->start = PCIBIOS_MIN_IO;
+		bus->resource[1]->start = PCIBIOS_MIN_MEM;
+#else
+		bus->resource[0]->end = 0;
+		bus->resource[1]->end = 0;
+		bus->resource[0]->start =0;
+		bus->resource[1]->start = 0;
+#endif
+		/* Turn off downstream PF memory address range by default */
+		bus->resource[2]->start = 1024*1024;
+		bus->resource[2]->end = bus->resource[2]->start - 1;
+	}
+#endif
+
+}
+
diff --git a/arch/sh/drivers/pci/pci-sh5.h b/arch/sh/drivers/pci/pci-sh5.h
new file mode 100644
index 0000000..c71159d
--- /dev/null
+++ b/arch/sh/drivers/pci/pci-sh5.h
@@ -0,0 +1,107 @@
+/*
+ * Copyright (C) 2001 David J. Mckay (david.mckay@st.com)
+ *
+ * May be copied or modified under the terms of the GNU General Public
+ * License.  See linux/COPYING for more information.
+ *
+ * Definitions for the SH5 PCI hardware.
+ */
+
+/* Product ID */
+#define PCISH5_PID		0x350d
+
+/* vendor ID */
+#define PCISH5_VID		0x1054
+
+/* Configuration types */
+#define ST_TYPE0                0x00    /* Configuration cycle type 0 */
+#define ST_TYPE1                0x01    /* Configuration cycle type 1 */
+
+/* VCR data */
+#define PCISH5_VCR_STATUS      0x00
+#define PCISH5_VCR_VERSION     0x08
+
+/*
+** ICR register offsets and bits
+*/
+#define PCISH5_ICR_CR          0x100   /* PCI control register values */
+#define CR_PBAM                 (1<<12)
+#define CR_PFCS                 (1<<11)
+#define CR_FTO                  (1<<10)
+#define CR_PFE                  (1<<9)
+#define CR_TBS                  (1<<8)
+#define CR_SPUE                 (1<<7)
+#define CR_BMAM                 (1<<6)
+#define CR_HOST                 (1<<5)
+#define CR_CLKEN                (1<<4)
+#define CR_SOCS                 (1<<3)
+#define CR_IOCS                 (1<<2)
+#define CR_RSTCTL               (1<<1)
+#define CR_CFINT                (1<<0)
+#define CR_LOCK_MASK            0xa5000000
+
+#define PCISH5_ICR_INT         0x114   /* Interrupt registert values     */
+#define INT_MADIM               (1<<2)
+
+#define PCISH5_ICR_LSR0        0X104   /* Local space register values    */
+#define PCISH5_ICR_LSR1        0X108   /* Local space register values    */
+#define PCISH5_ICR_LAR0        0x10c   /* Local address register values  */
+#define PCISH5_ICR_LAR1        0x110   /* Local address register values  */
+#define PCISH5_ICR_INTM        0x118   /* Interrupt mask register values                         */
+#define PCISH5_ICR_AIR         0x11c   /* Interrupt error address information register values    */
+#define PCISH5_ICR_CIR         0x120   /* Interrupt error command information register values    */
+#define PCISH5_ICR_AINT        0x130   /* Interrupt error arbiter interrupt register values      */
+#define PCISH5_ICR_AINTM       0x134   /* Interrupt error arbiter interrupt mask register values */
+#define PCISH5_ICR_BMIR        0x138   /* Interrupt error info register of bus master values     */
+#define PCISH5_ICR_PAR         0x1c0   /* Pio address register values                            */
+#define PCISH5_ICR_MBR         0x1c4   /* Memory space bank register values                      */
+#define PCISH5_ICR_IOBR        0x1c8   /* I/O space bank register values                         */
+#define PCISH5_ICR_PINT        0x1cc   /* power management interrupt register values             */
+#define PCISH5_ICR_PINTM       0x1d0   /* power management interrupt mask register values        */
+#define PCISH5_ICR_MBMR        0x1d8   /* memory space bank mask register values                 */
+#define PCISH5_ICR_IOBMR       0x1dc   /* I/O space bank mask register values                    */
+#define PCISH5_ICR_CSCR0       0x210   /* PCI cache snoop control register 0                     */
+#define PCISH5_ICR_CSCR1       0x214   /* PCI cache snoop control register 1                     */
+#define PCISH5_ICR_PDR         0x220   /* Pio data register values                               */
+
+/* These are configs space registers */
+#define PCISH5_ICR_CSR_VID     0x000	/* Vendor id                           */
+#define PCISH5_ICR_CSR_DID     0x002   /* Device id                           */
+#define PCISH5_ICR_CSR_CMD     0x004   /* Command register                    */
+#define PCISH5_ICR_CSR_STATUS  0x006   /* Stautus                             */
+#define PCISH5_ICR_CSR_IBAR0   0x010   /* I/O base address register           */
+#define PCISH5_ICR_CSR_MBAR0   0x014   /* First  Memory base address register */
+#define PCISH5_ICR_CSR_MBAR1   0x018   /* Second Memory base address register */
+
+
+
+/* Base address of registers */
+#define SH5PCI_ICR_BASE (PHYS_PCI_BLOCK + 0x00040000)
+#define SH5PCI_IO_BASE  (PHYS_PCI_BLOCK + 0x00800000)
+/* #define SH5PCI_VCR_BASE (P2SEG_PCICB_BLOCK + P2SEG)    */
+
+/* Register selection macro */
+#define PCISH5_ICR_REG(x)                ( pcicr_virt + (PCISH5_ICR_##x))
+/* #define PCISH5_VCR_REG(x)                ( SH5PCI_VCR_BASE (PCISH5_VCR_##x)) */
+
+/* Write I/O functions */
+#define SH5PCI_WRITE(reg,val)        ctrl_outl((u32)(val),PCISH5_ICR_REG(reg))
+#define SH5PCI_WRITE_SHORT(reg,val)  ctrl_outw((u16)(val),PCISH5_ICR_REG(reg))
+#define SH5PCI_WRITE_BYTE(reg,val)   ctrl_outb((u8)(val),PCISH5_ICR_REG(reg))
+
+/* Read I/O functions */
+#define SH5PCI_READ(reg)             ctrl_inl(PCISH5_ICR_REG(reg))
+#define SH5PCI_READ_SHORT(reg)       ctrl_inw(PCISH5_ICR_REG(reg))
+#define SH5PCI_READ_BYTE(reg)        ctrl_inb(PCISH5_ICR_REG(reg))
+
+/* Set PCI config bits */
+#define SET_CONFIG_BITS(bus,devfn,where)  ((((bus) << 16) | ((devfn) << 8) | ((where) & ~3)) | 0x80000000)
+
+/* Set PCI command register */
+#define CONFIG_CMD(bus, devfn, where)            SET_CONFIG_BITS(bus->number,devfn,where)
+
+/* Size converters */
+#define PCISH5_MEM_SIZCONV(x)		  (((x / 0x40000) - 1) << 18)
+#define PCISH5_IO_SIZCONV(x)		  (((x / 0x40000) - 1) << 18)
+
+
diff --git a/arch/sh/kernel/Makefile_32 b/arch/sh/kernel/Makefile_32
index 990ba74..c892898 100644
--- a/arch/sh/kernel/Makefile_32
+++ b/arch/sh/kernel/Makefile_32
@@ -6,7 +6,7 @@
 
 obj-y	:= debugtraps.o io.o io_generic.o irq.o machvec.o process_32.o \
 	   ptrace_32.o semaphore.o setup.o signal_32.o sys_sh.o sys_sh32.o \
-	   syscalls_32.o time.o topology.o traps.o traps_32.o
+	   syscalls_32.o time_32.o topology.o traps.o traps_32.o
 
 obj-y				+= cpu/ timers/
 obj-$(CONFIG_VSYSCALL)		+= vsyscall/
diff --git a/arch/sh/kernel/Makefile_64 b/arch/sh/kernel/Makefile_64
index 10e3ae1..1ef21cc 100644
--- a/arch/sh/kernel/Makefile_64
+++ b/arch/sh/kernel/Makefile_64
@@ -2,7 +2,7 @@
 
 obj-y	:= debugtraps.o io.o io_generic.o irq.o machvec.o process_64.o \
 	   ptrace_64.o semaphore.o setup.o signal_64.o sys_sh.o sys_sh64.o \
-	   syscalls_64.o time.o topology.o traps.o traps_64.o
+	   syscalls_64.o time_64.o topology.o traps.o traps_64.o
 
 obj-y				+= cpu/ timers/
 obj-$(CONFIG_VSYSCALL)		+= vsyscall/
diff --git a/arch/sh/kernel/time.c b/arch/sh/kernel/time_32.c
similarity index 100%
rename from arch/sh/kernel/time.c
rename to arch/sh/kernel/time_32.c
diff --git a/arch/sh/kernel/time_64.c b/arch/sh/kernel/time_64.c
new file mode 100644
index 0000000..4c52fee
--- /dev/null
+++ b/arch/sh/kernel/time_64.c
@@ -0,0 +1,528 @@
+/*
+ * 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.
+ *
+ * arch/sh64/kernel/time.c
+ *
+ * Copyright (C) 2000, 2001  Paolo Alberelli
+ * Copyright (C) 2003 - 2007  Paul Mundt
+ * Copyright (C) 2003  Richard Curnow
+ *
+ *    Original TMU/RTC code taken from sh version.
+ *    Copyright (C) 1999  Tetsuya Okada & Niibe Yutaka
+ *      Some code taken from i386 version.
+ *      Copyright (C) 1991, 1992, 1995  Linus Torvalds
+ */
+#include <linux/errno.h>
+#include <linux/rwsem.h>
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/param.h>
+#include <linux/string.h>
+#include <linux/mm.h>
+#include <linux/interrupt.h>
+#include <linux/time.h>
+#include <linux/delay.h>
+#include <linux/init.h>
+#include <linux/profile.h>
+#include <linux/smp.h>
+#include <linux/module.h>
+#include <linux/bcd.h>
+#include <linux/timex.h>
+#include <linux/irq.h>
+#include <linux/io.h>
+#include <linux/platform_device.h>
+#include <asm/cpu/registers.h>	 /* required by inline __asm__ stmt. */
+#include <asm/cpu/irq.h>
+#include <asm/addrspace.h>
+#include <asm/processor.h>
+#include <asm/uaccess.h>
+#include <asm/delay.h>
+
+#define TMU_TOCR_INIT	0x00
+#define TMU0_TCR_INIT	0x0020
+#define TMU_TSTR_INIT	1
+#define TMU_TSTR_OFF	0
+
+/* Real Time Clock */
+#define	RTC_BLOCK_OFF	0x01040000
+#define RTC_BASE	PHYS_PERIPHERAL_BLOCK + RTC_BLOCK_OFF
+#define RTC_RCR1_CIE	0x10	/* Carry Interrupt Enable */
+#define RTC_RCR1	(rtc_base + 0x38)
+
+/* Clock, Power and Reset Controller */
+#define	CPRC_BLOCK_OFF	0x01010000
+#define CPRC_BASE	PHYS_PERIPHERAL_BLOCK + CPRC_BLOCK_OFF
+
+#define FRQCR		(cprc_base+0x0)
+#define WTCSR		(cprc_base+0x0018)
+#define STBCR		(cprc_base+0x0030)
+
+/* Time Management Unit */
+#define	TMU_BLOCK_OFF	0x01020000
+#define TMU_BASE	PHYS_PERIPHERAL_BLOCK + TMU_BLOCK_OFF
+#define TMU0_BASE	tmu_base + 0x8 + (0xc * 0x0)
+#define TMU1_BASE	tmu_base + 0x8 + (0xc * 0x1)
+#define TMU2_BASE	tmu_base + 0x8 + (0xc * 0x2)
+
+#define TMU_TOCR	tmu_base+0x0	/* Byte access */
+#define TMU_TSTR	tmu_base+0x4	/* Byte access */
+
+#define TMU0_TCOR	TMU0_BASE+0x0	/* Long access */
+#define TMU0_TCNT	TMU0_BASE+0x4	/* Long access */
+#define TMU0_TCR	TMU0_BASE+0x8	/* Word access */
+
+#define TICK_SIZE (tick_nsec / 1000)
+
+static unsigned long tmu_base, rtc_base;
+unsigned long cprc_base;
+
+/* Variables to allow interpolation of time of day to resolution better than a
+ * jiffy. */
+
+/* This is effectively protected by xtime_lock */
+static unsigned long ctc_last_interrupt;
+static unsigned long long usecs_per_jiffy = 1000000/HZ; /* Approximation */
+
+#define CTC_JIFFY_SCALE_SHIFT 40
+
+/* 2**CTC_JIFFY_SCALE_SHIFT / ctc_ticks_per_jiffy */
+static unsigned long long scaled_recip_ctc_ticks_per_jiffy;
+
+/* Estimate number of microseconds that have elapsed since the last timer tick,
+   by scaling the delta that has occurred in the CTC register.
+
+   WARNING WARNING WARNING : This algorithm relies on the CTC decrementing at
+   the CPU clock rate.  If the CPU sleeps, the CTC stops counting.  Bear this
+   in mind if enabling SLEEP_WORKS in process.c.  In that case, this algorithm
+   probably needs to use TMU.TCNT0 instead.  This will work even if the CPU is
+   sleeping, though will be coarser.
+
+   FIXME : What if usecs_per_tick is moving around too much, e.g. if an adjtime
+   is running or if the freq or tick arguments of adjtimex are modified after
+   we have calibrated the scaling factor?  This will result in either a jump at
+   the end of a tick period, or a wrap backwards at the start of the next one,
+   if the application is reading the time of day often enough.  I think we
+   ought to do better than this.  For this reason, usecs_per_jiffy is left
+   separated out in the calculation below.  This allows some future hook into
+   the adjtime-related stuff in kernel/timer.c to remove this hazard.
+
+*/
+
+static unsigned long usecs_since_tick(void)
+{
+	unsigned long long current_ctc;
+	long ctc_ticks_since_interrupt;
+	unsigned long long ull_ctc_ticks_since_interrupt;
+	unsigned long result;
+
+	unsigned long long mul1_out;
+	unsigned long long mul1_out_high;
+	unsigned long long mul2_out_low, mul2_out_high;
+
+	/* Read CTC register */
+	asm ("getcon cr62, %0" : "=r" (current_ctc));
+	/* Note, the CTC counts down on each CPU clock, not up.
+	   Note(2), use long type to get correct wraparound arithmetic when
+	   the counter crosses zero. */
+	ctc_ticks_since_interrupt = (long) ctc_last_interrupt - (long) current_ctc;
+	ull_ctc_ticks_since_interrupt = (unsigned long long) ctc_ticks_since_interrupt;
+
+	/* Inline assembly to do 32x32x32->64 multiplier */
+	asm volatile ("mulu.l %1, %2, %0" :
+	     "=r" (mul1_out) :
+	     "r" (ull_ctc_ticks_since_interrupt), "r" (usecs_per_jiffy));
+
+	mul1_out_high = mul1_out >> 32;
+
+	asm volatile ("mulu.l %1, %2, %0" :
+	     "=r" (mul2_out_low) :
+	     "r" (mul1_out), "r" (scaled_recip_ctc_ticks_per_jiffy));
+
+#if 1
+	asm volatile ("mulu.l %1, %2, %0" :
+	     "=r" (mul2_out_high) :
+	     "r" (mul1_out_high), "r" (scaled_recip_ctc_ticks_per_jiffy));
+#endif
+
+	result = (unsigned long) (((mul2_out_high << 32) + mul2_out_low) >> CTC_JIFFY_SCALE_SHIFT);
+
+	return result;
+}
+
+void do_gettimeofday(struct timeval *tv)
+{
+	unsigned long flags;
+	unsigned long seq;
+	unsigned long usec, sec;
+
+	do {
+		seq = read_seqbegin_irqsave(&xtime_lock, flags);
+		usec = usecs_since_tick();
+		sec = xtime.tv_sec;
+		usec += xtime.tv_nsec / 1000;
+	} while (read_seqretry_irqrestore(&xtime_lock, seq, flags));
+
+	while (usec >= 1000000) {
+		usec -= 1000000;
+		sec++;
+	}
+
+	tv->tv_sec = sec;
+	tv->tv_usec = usec;
+}
+
+int do_settimeofday(struct timespec *tv)
+{
+	time_t wtm_sec, sec = tv->tv_sec;
+	long wtm_nsec, nsec = tv->tv_nsec;
+
+	if ((unsigned long)tv->tv_nsec >= NSEC_PER_SEC)
+		return -EINVAL;
+
+	write_seqlock_irq(&xtime_lock);
+	/*
+	 * This is revolting. We need to set "xtime" correctly. However, the
+	 * value in this location is the value at the most recent update of
+	 * wall time.  Discover what correction gettimeofday() would have
+	 * made, and then undo it!
+	 */
+	nsec -= 1000 * usecs_since_tick();
+
+	wtm_sec  = wall_to_monotonic.tv_sec + (xtime.tv_sec - sec);
+	wtm_nsec = wall_to_monotonic.tv_nsec + (xtime.tv_nsec - nsec);
+
+	set_normalized_timespec(&xtime, sec, nsec);
+	set_normalized_timespec(&wall_to_monotonic, wtm_sec, wtm_nsec);
+
+	ntp_clear();
+	write_sequnlock_irq(&xtime_lock);
+	clock_was_set();
+
+	return 0;
+}
+EXPORT_SYMBOL(do_settimeofday);
+
+/* 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;
+
+/* last time the RTC clock got updated */
+static long last_rtc_update;
+
+/*
+ * timer_interrupt() needs to keep up the real-time clock,
+ * as well as call the "do_timer()" routine every clocktick
+ */
+static inline void do_timer_interrupt(void)
+{
+	unsigned long long current_ctc;
+	asm ("getcon cr62, %0" : "=r" (current_ctc));
+	ctc_last_interrupt = (unsigned long) current_ctc;
+
+	do_timer(1);
+#ifndef CONFIG_SMP
+	update_process_times(user_mode(get_irq_regs()));
+#endif
+	if (current->pid)
+		profile_tick(CPU_PROFILING);
+
+#ifdef CONFIG_HEARTBEAT
+	if (sh_mv.mv_heartbeat != NULL)
+		sh_mv.mv_heartbeat();
+#endif
+
+	/*
+	 * If we have an externally synchronized Linux clock, then update
+	 * RTC clock accordingly every ~11 minutes. Set_rtc_mmss() has to be
+	 * called as close as possible to 500 ms before the new second starts.
+	 */
+	if (ntp_synced() &&
+	    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_sh_set_time(xtime.tv_sec) == 0)
+			last_rtc_update = xtime.tv_sec;
+		else
+			/* do it again in 60 s */
+			last_rtc_update = xtime.tv_sec - 600;
+	}
+}
+
+/*
+ * This is the same as the above, except we _also_ save the current
+ * Time Stamp Counter value at the time of the timer interrupt, so that
+ * we later on can estimate the time of day more exactly.
+ */
+static irqreturn_t timer_interrupt(int irq, void *dev_id)
+{
+	unsigned long timer_status;
+
+	/* Clear UNF bit */
+	timer_status = ctrl_inw(TMU0_TCR);
+	timer_status &= ~0x100;
+	ctrl_outw(timer_status, TMU0_TCR);
+
+	/*
+	 * Here we are in the timer irq handler. We just have irqs locally
+	 * disabled but we don't know if the timer_bh is running on the other
+	 * CPU. We need to avoid to SMP race with it. NOTE: we don' t need
+	 * the irq version of write_lock because as just said we have irq
+	 * locally disabled. -arca
+	 */
+	write_lock(&xtime_lock);
+	do_timer_interrupt();
+	write_unlock(&xtime_lock);
+
+	return IRQ_HANDLED;
+}
+
+
+static __init unsigned int get_cpu_hz(void)
+{
+	unsigned int count;
+	unsigned long __dummy;
+	unsigned long ctc_val_init, ctc_val;
+
+	/*
+	** Regardless the toolchain, force the compiler to use the
+	** arbitrary register r3 as a clock tick counter.
+	** NOTE: r3 must be in accordance with sh64_rtc_interrupt()
+	*/
+	register unsigned long long  __rtc_irq_flag __asm__ ("r3");
+
+	local_irq_enable();
+	do {} while (ctrl_inb(rtc_base) != 0);
+	ctrl_outb(RTC_RCR1_CIE, RTC_RCR1); /* Enable carry interrupt */
+
+	/*
+	 * r3 is arbitrary. CDC does not support "=z".
+	 */
+	ctc_val_init = 0xffffffff;
+	ctc_val = ctc_val_init;
+
+	asm volatile("gettr	tr0, %1\n\t"
+		     "putcon	%0, " __CTC "\n\t"
+		     "and	%2, r63, %2\n\t"
+		     "pta	$+4, tr0\n\t"
+		     "beq/l	%2, r63, tr0\n\t"
+		     "ptabs	%1, tr0\n\t"
+		     "getcon	" __CTC ", %0\n\t"
+		: "=r"(ctc_val), "=r" (__dummy), "=r" (__rtc_irq_flag)
+		: "0" (0));
+	local_irq_disable();
+	/*
+	 * SH-3:
+	 * CPU clock = 4 stages * loop
+	 * tst    rm,rm      if id ex
+	 * bt/s   1b            if id ex
+	 * add    #1,rd            if id ex
+         *                            (if) pipe line stole
+	 * tst    rm,rm                  if id ex
+         * ....
+	 *
+	 *
+	 * SH-4:
+	 * CPU clock = 6 stages * loop
+	 * I don't know why.
+         * ....
+	 *
+	 * SH-5:
+	 * Use CTC register to count.  This approach returns the right value
+	 * even if the I-cache is disabled (e.g. whilst debugging.)
+	 *
+	 */
+
+	count = ctc_val_init - ctc_val; /* CTC counts down */
+
+#if defined (CONFIG_SH_SIMULATOR)
+	/*
+	 * Let's pretend we are a 5MHz SH-5 to avoid a too
+	 * little timer interval. Also to keep delay
+	 * calibration within a reasonable time.
+	 */
+	return 5000000;
+#else
+	/*
+	 * This really is count by the number of clock cycles
+         * by the ratio between a complete R64CNT
+         * wrap-around (128) and CUI interrupt being raised (64).
+	 */
+	return count*2;
+#endif
+}
+
+static irqreturn_t sh64_rtc_interrupt(int irq, void *dev_id)
+{
+	struct pt_regs *regs = get_irq_regs();
+
+	ctrl_outb(0, RTC_RCR1);	/* Disable Carry Interrupts */
+	regs->regs[3] = 1;	/* Using r3 */
+
+	return IRQ_HANDLED;
+}
+
+static struct irqaction irq0  = {
+	.handler = timer_interrupt,
+	.flags = IRQF_DISABLED,
+	.mask = CPU_MASK_NONE,
+	.name = "timer",
+};
+static struct irqaction irq1  = {
+	.handler = sh64_rtc_interrupt,
+	.flags = IRQF_DISABLED,
+	.mask = CPU_MASK_NONE,
+	.name = "rtc",
+};
+
+void __init time_init(void)
+{
+	unsigned int cpu_clock, master_clock, bus_clock, module_clock;
+	unsigned long interval;
+	unsigned long frqcr, ifc, pfc;
+	static int ifc_table[] = { 2, 4, 6, 8, 10, 12, 16, 24 };
+#define bfc_table ifc_table	/* Same */
+#define pfc_table ifc_table	/* Same */
+
+	tmu_base = onchip_remap(TMU_BASE, 1024, "TMU");
+	if (!tmu_base) {
+		panic("Unable to remap TMU\n");
+	}
+
+	rtc_base = onchip_remap(RTC_BASE, 1024, "RTC");
+	if (!rtc_base) {
+		panic("Unable to remap RTC\n");
+	}
+
+	cprc_base = onchip_remap(CPRC_BASE, 1024, "CPRC");
+	if (!cprc_base) {
+		panic("Unable to remap CPRC\n");
+	}
+
+	rtc_sh_get_time(&xtime);
+
+	setup_irq(TIMER_IRQ, &irq0);
+	setup_irq(RTC_IRQ, &irq1);
+
+	/* Check how fast it is.. */
+	cpu_clock = get_cpu_hz();
+
+	/* Note careful order of operations to maintain reasonable precision and avoid overflow. */
+	scaled_recip_ctc_ticks_per_jiffy = ((1ULL << CTC_JIFFY_SCALE_SHIFT) / (unsigned long long)(cpu_clock / HZ));
+
+	free_irq(RTC_IRQ, NULL);
+
+	printk("CPU clock: %d.%02dMHz\n",
+	       (cpu_clock / 1000000), (cpu_clock % 1000000)/10000);
+	{
+		unsigned short bfc;
+		frqcr = ctrl_inl(FRQCR);
+		ifc  = ifc_table[(frqcr>> 6) & 0x0007];
+		bfc  = bfc_table[(frqcr>> 3) & 0x0007];
+		pfc  = pfc_table[(frqcr>> 12) & 0x0007];
+		master_clock = cpu_clock * ifc;
+		bus_clock = master_clock/bfc;
+	}
+
+	printk("Bus clock: %d.%02dMHz\n",
+	       (bus_clock/1000000), (bus_clock % 1000000)/10000);
+	module_clock = master_clock/pfc;
+	printk("Module clock: %d.%02dMHz\n",
+	       (module_clock/1000000), (module_clock % 1000000)/10000);
+	interval = (module_clock/(HZ*4));
+
+	printk("Interval = %ld\n", interval);
+
+	current_cpu_data.cpu_clock    = cpu_clock;
+	current_cpu_data.master_clock = master_clock;
+	current_cpu_data.bus_clock    = bus_clock;
+	current_cpu_data.module_clock = module_clock;
+
+	/* Start TMU0 */
+	ctrl_outb(TMU_TSTR_OFF, TMU_TSTR);
+	ctrl_outb(TMU_TOCR_INIT, TMU_TOCR);
+	ctrl_outw(TMU0_TCR_INIT, TMU0_TCR);
+	ctrl_outl(interval, TMU0_TCOR);
+	ctrl_outl(interval, TMU0_TCNT);
+	ctrl_outb(TMU_TSTR_INIT, TMU_TSTR);
+}
+
+void enter_deep_standby(void)
+{
+	/* Disable watchdog timer */
+	ctrl_outl(0xa5000000, WTCSR);
+	/* Configure deep standby on sleep */
+	ctrl_outl(0x03, STBCR);
+
+#ifdef CONFIG_SH_ALPHANUMERIC
+	{
+		extern void mach_alphanum(int position, unsigned char value);
+		extern void mach_alphanum_brightness(int setting);
+		char halted[] = "Halted. ";
+		int i;
+		mach_alphanum_brightness(6); /* dimmest setting above off */
+		for (i=0; i<8; i++) {
+			mach_alphanum(i, halted[i]);
+		}
+		asm __volatile__ ("synco");
+	}
+#endif
+
+	asm __volatile__ ("sleep");
+	asm __volatile__ ("synci");
+	asm __volatile__ ("nop");
+	asm __volatile__ ("nop");
+	asm __volatile__ ("nop");
+	asm __volatile__ ("nop");
+	panic("Unexpected wakeup!\n");
+}
+
+static struct resource rtc_resources[] = {
+	[0] = {
+		/* RTC base, filled in by rtc_init */
+		.flags	= IORESOURCE_IO,
+	},
+	[1] = {
+		/* Period IRQ */
+		.start	= IRQ_PRI,
+		.flags	= IORESOURCE_IRQ,
+	},
+	[2] = {
+		/* Carry IRQ */
+		.start	= IRQ_CUI,
+		.flags	= IORESOURCE_IRQ,
+	},
+	[3] = {
+		/* Alarm IRQ */
+		.start	= IRQ_ATI,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device rtc_device = {
+	.name		= "sh-rtc",
+	.id		= -1,
+	.num_resources	= ARRAY_SIZE(rtc_resources),
+	.resource	= rtc_resources,
+};
+
+static int __init rtc_init(void)
+{
+	rtc_resources[0].start	= rtc_base;
+	rtc_resources[0].end	= rtc_resources[0].start + 0x58 - 1;
+
+	return platform_device_register(&rtc_device);
+}
+device_initcall(rtc_init);