Merge branch 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/epip/linux-2.6-unicore32

* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/epip/linux-2.6-unicore32: (40 commits)
  unicore32: rewrite arch-specific tlb.h to use asm-generic version
  unicore32: modify io_p2v and io_v2p macros, and adjust PKUNITY_mmio_BASEs
  unicore32: replace unicore32-specific iomap functions with generic lib implementation
  unicore32 machine related: add frame buffer driver for pkunity-v3 soc
  unicore32 machine related files: add i2c bus drivers for pkunity-v3 soc
  unicore32 io: redefine __REG(x) and re-use readl/writel funcs
  unicore32 i8042 upgrade and bugfix: adjust resource request region type
  unicore32 upgrade to v2.6.38-rc5: add one more paramter for pte_alloc_map call
  unicore32 i8042: adjust io funcs of i8042-unicore32io.h
  unicore32: rename PKUNITY_IOSPACE_BASE to PKUNITY_MMIO_BASE
  unicore32: modify function names and parameters for irq_chips
  unicore32: remove unused lines in arch/unicore32/include/asm/irq.h
  unicore32 time.c: change calculate method for clock_event_device
  unicore32: ADD MAINTAINER for unicore32 architecture
  unicore32 machine related files: ps2 driver
  unicore32 machine related files: pci bus handling
  unicore32 machine related files: hardware registers
  unicore32 machine related files: core files
  unicore32 additional architecture files: boot process
  unicore32 additional architecture files: low-level lib: misc
  ...

Acked-by: Arnd Bergmann <arnd@arndb.de>

37 files changed, 7437 insertions, 0 deletions

diff --git a/arch/unicore32/kernel/Makefile b/arch/unicore32/kernel/Makefile
new file mode 100644
index 000000000000..ec23a2fb2f50
--- /dev/null
+++ b/arch/unicore32/kernel/Makefile
@@ -0,0 +1,33 @@
+#
+# Makefile for the linux kernel.
+#
+
+# Object file lists.
+obj-y				:= dma.o elf.o entry.o process.o ptrace.o
+obj-y				+= setup.o signal.o sys.o stacktrace.o traps.o
+
+obj-$(CONFIG_MODULES)		+= ksyms.o module.o
+obj-$(CONFIG_EARLY_PRINTK)	+= early_printk.o
+
+obj-$(CONFIG_CPU_FREQ)		+= cpu-ucv2.o
+obj-$(CONFIG_UNICORE_FPU_F64)	+= fpu-ucf64.o
+
+# obj-y for architecture PKUnity v3
+obj-$(CONFIG_ARCH_PUV3)		+= clock.o irq.o time.o
+
+obj-$(CONFIG_PUV3_GPIO)		+= gpio.o
+obj-$(CONFIG_PUV3_RTC)		+= rtc.o
+obj-$(CONFIG_PUV3_PWM)		+= pwm.o
+obj-$(CONFIG_PUV3_PM)		+= pm.o sleep.o
+obj-$(CONFIG_HIBERNATION)	+= hibernate.o hibernate_asm.o
+
+obj-$(CONFIG_PCI)		+= pci.o
+
+# obj-y for specific machines
+obj-$(CONFIG_ARCH_PUV3)		+= puv3-core.o
+obj-$(CONFIG_PUV3_NB0916)	+= puv3-nb0916.o
+
+head-y				:= head.o
+obj-$(CONFIG_DEBUG_LL)		+= debug.o
+
+extra-y				:= $(head-y) init_task.o vmlinux.lds
diff --git a/arch/unicore32/kernel/asm-offsets.c b/arch/unicore32/kernel/asm-offsets.c
new file mode 100644
index 000000000000..ffcbe7536ca7
--- /dev/null
+++ b/arch/unicore32/kernel/asm-offsets.c
@@ -0,0 +1,112 @@
+/*
+ * linux/arch/unicore32/kernel/asm-offsets.c
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * Generate definitions needed by assembly language modules.
+ * This code generates raw asm output which is post-processed to extract
+ * and format the required data.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/dma-mapping.h>
+#include <linux/kbuild.h>
+#include <linux/suspend.h>
+#include <linux/thread_info.h>
+#include <asm/memory.h>
+#include <asm/suspend.h>
+
+/*
+ * GCC 3.0, 3.1: general bad code generation.
+ * GCC 3.2.0: incorrect function argument offset calculation.
+ * GCC 3.2.x: miscompiles NEW_AUX_ENT in fs/binfmt_elf.c
+ *	(http://gcc.gnu.org/PR8896) and incorrect structure
+ *		initialisation in fs/jffs2/erase.c
+ */
+#if (__GNUC__ < 4)
+#error Your compiler should upgrade to uc4
+#error	Known good compilers: 4.2.2
+#endif
+
+int main(void)
+{
+	DEFINE(TSK_ACTIVE_MM,	offsetof(struct task_struct, active_mm));
+	BLANK();
+	DEFINE(TI_FLAGS,	offsetof(struct thread_info, flags));
+	DEFINE(TI_PREEMPT,	offsetof(struct thread_info, preempt_count));
+	DEFINE(TI_ADDR_LIMIT,	offsetof(struct thread_info, addr_limit));
+	DEFINE(TI_TASK,		offsetof(struct thread_info, task));
+	DEFINE(TI_EXEC_DOMAIN,	offsetof(struct thread_info, exec_domain));
+	DEFINE(TI_CPU,		offsetof(struct thread_info, cpu));
+	DEFINE(TI_CPU_SAVE,	offsetof(struct thread_info, cpu_context));
+	DEFINE(TI_USED_CP,	offsetof(struct thread_info, used_cp));
+#ifdef CONFIG_UNICORE_FPU_F64
+	DEFINE(TI_FPSTATE,	offsetof(struct thread_info, fpstate));
+#endif
+	BLANK();
+	DEFINE(S_R0,		offsetof(struct pt_regs, UCreg_00));
+	DEFINE(S_R1,		offsetof(struct pt_regs, UCreg_01));
+	DEFINE(S_R2,		offsetof(struct pt_regs, UCreg_02));
+	DEFINE(S_R3,		offsetof(struct pt_regs, UCreg_03));
+	DEFINE(S_R4,		offsetof(struct pt_regs, UCreg_04));
+	DEFINE(S_R5,		offsetof(struct pt_regs, UCreg_05));
+	DEFINE(S_R6,		offsetof(struct pt_regs, UCreg_06));
+	DEFINE(S_R7,		offsetof(struct pt_regs, UCreg_07));
+	DEFINE(S_R8,		offsetof(struct pt_regs, UCreg_08));
+	DEFINE(S_R9,		offsetof(struct pt_regs, UCreg_09));
+	DEFINE(S_R10,		offsetof(struct pt_regs, UCreg_10));
+	DEFINE(S_R11,		offsetof(struct pt_regs, UCreg_11));
+	DEFINE(S_R12,		offsetof(struct pt_regs, UCreg_12));
+	DEFINE(S_R13,		offsetof(struct pt_regs, UCreg_13));
+	DEFINE(S_R14,		offsetof(struct pt_regs, UCreg_14));
+	DEFINE(S_R15,		offsetof(struct pt_regs, UCreg_15));
+	DEFINE(S_R16,		offsetof(struct pt_regs, UCreg_16));
+	DEFINE(S_R17,		offsetof(struct pt_regs, UCreg_17));
+	DEFINE(S_R18,		offsetof(struct pt_regs, UCreg_18));
+	DEFINE(S_R19,		offsetof(struct pt_regs, UCreg_19));
+	DEFINE(S_R20,		offsetof(struct pt_regs, UCreg_20));
+	DEFINE(S_R21,		offsetof(struct pt_regs, UCreg_21));
+	DEFINE(S_R22,		offsetof(struct pt_regs, UCreg_22));
+	DEFINE(S_R23,		offsetof(struct pt_regs, UCreg_23));
+	DEFINE(S_R24,		offsetof(struct pt_regs, UCreg_24));
+	DEFINE(S_R25,		offsetof(struct pt_regs, UCreg_25));
+	DEFINE(S_R26,		offsetof(struct pt_regs, UCreg_26));
+	DEFINE(S_FP,		offsetof(struct pt_regs, UCreg_fp));
+	DEFINE(S_IP,		offsetof(struct pt_regs, UCreg_ip));
+	DEFINE(S_SP,		offsetof(struct pt_regs, UCreg_sp));
+	DEFINE(S_LR,		offsetof(struct pt_regs, UCreg_lr));
+	DEFINE(S_PC,		offsetof(struct pt_regs, UCreg_pc));
+	DEFINE(S_PSR,		offsetof(struct pt_regs, UCreg_asr));
+	DEFINE(S_OLD_R0,	offsetof(struct pt_regs, UCreg_ORIG_00));
+	DEFINE(S_FRAME_SIZE,	sizeof(struct pt_regs));
+	BLANK();
+	DEFINE(VMA_VM_MM,	offsetof(struct vm_area_struct, vm_mm));
+	DEFINE(VMA_VM_FLAGS,	offsetof(struct vm_area_struct, vm_flags));
+	BLANK();
+	DEFINE(VM_EXEC,		VM_EXEC);
+	BLANK();
+	DEFINE(PAGE_SZ,		PAGE_SIZE);
+	BLANK();
+	DEFINE(SYS_ERROR0,	0x9f0000);
+	BLANK();
+	DEFINE(PBE_ADDRESS,		offsetof(struct pbe, address));
+	DEFINE(PBE_ORIN_ADDRESS,	offsetof(struct pbe, orig_address));
+	DEFINE(PBE_NEXT,		offsetof(struct pbe, next));
+	DEFINE(SWSUSP_CPU,		offsetof(struct swsusp_arch_regs, \
+							cpu_context));
+#ifdef	CONFIG_UNICORE_FPU_F64
+	DEFINE(SWSUSP_FPSTATE,		offsetof(struct swsusp_arch_regs, \
+							fpstate));
+#endif
+	BLANK();
+	DEFINE(DMA_BIDIRECTIONAL,	DMA_BIDIRECTIONAL);
+	DEFINE(DMA_TO_DEVICE,		DMA_TO_DEVICE);
+	DEFINE(DMA_FROM_DEVICE,		DMA_FROM_DEVICE);
+	return 0;
+}
diff --git a/arch/unicore32/kernel/clock.c b/arch/unicore32/kernel/clock.c
new file mode 100644
index 000000000000..18d4563e6fa5
--- /dev/null
+++ b/arch/unicore32/kernel/clock.c
@@ -0,0 +1,390 @@
+/*
+ * linux/arch/unicore32/kernel/clock.c
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ *	Maintained by GUAN Xue-tao <gxt@mprc.pku.edu.cn>
+ *	Copyright (C) 2001-2010 Guan Xuetao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/device.h>
+#include <linux/list.h>
+#include <linux/errno.h>
+#include <linux/err.h>
+#include <linux/string.h>
+#include <linux/clk.h>
+#include <linux/mutex.h>
+#include <linux/delay.h>
+#include <linux/io.h>
+
+#include <mach/hardware.h>
+
+/*
+ * Very simple clock implementation
+ */
+struct clk {
+	struct list_head	node;
+	unsigned long		rate;
+	const char		*name;
+};
+
+static struct clk clk_ost_clk = {
+	.name		= "OST_CLK",
+	.rate		= CLOCK_TICK_RATE,
+};
+
+static struct clk clk_mclk_clk = {
+	.name		= "MAIN_CLK",
+};
+
+static struct clk clk_bclk32_clk = {
+	.name		= "BUS32_CLK",
+};
+
+static struct clk clk_ddr_clk = {
+	.name		= "DDR_CLK",
+};
+
+static struct clk clk_vga_clk = {
+	.name		= "VGA_CLK",
+};
+
+static LIST_HEAD(clocks);
+static DEFINE_MUTEX(clocks_mutex);
+
+struct clk *clk_get(struct device *dev, const char *id)
+{
+	struct clk *p, *clk = ERR_PTR(-ENOENT);
+
+	mutex_lock(&clocks_mutex);
+	list_for_each_entry(p, &clocks, node) {
+		if (strcmp(id, p->name) == 0) {
+			clk = p;
+			break;
+		}
+	}
+	mutex_unlock(&clocks_mutex);
+
+	return clk;
+}
+EXPORT_SYMBOL(clk_get);
+
+void clk_put(struct clk *clk)
+{
+}
+EXPORT_SYMBOL(clk_put);
+
+int clk_enable(struct clk *clk)
+{
+	return 0;
+}
+EXPORT_SYMBOL(clk_enable);
+
+void clk_disable(struct clk *clk)
+{
+}
+EXPORT_SYMBOL(clk_disable);
+
+unsigned long clk_get_rate(struct clk *clk)
+{
+	return clk->rate;
+}
+EXPORT_SYMBOL(clk_get_rate);
+
+struct {
+	unsigned long rate;
+	unsigned long cfg;
+	unsigned long div;
+} vga_clk_table[] = {
+	{.rate =  25175000, .cfg = 0x00002001, .div = 0x9},
+	{.rate =  31500000, .cfg = 0x00002001, .div = 0x7},
+	{.rate =  40000000, .cfg = 0x00003801, .div = 0x9},
+	{.rate =  49500000, .cfg = 0x00003801, .div = 0x7},
+	{.rate =  65000000, .cfg = 0x00002c01, .div = 0x4},
+	{.rate =  78750000, .cfg = 0x00002400, .div = 0x7},
+	{.rate = 108000000, .cfg = 0x00002c01, .div = 0x2},
+	{.rate = 106500000, .cfg = 0x00003c01, .div = 0x3},
+	{.rate =  50650000, .cfg = 0x00106400, .div = 0x9},
+	{.rate =  61500000, .cfg = 0x00106400, .div = 0xa},
+	{.rate =  85500000, .cfg = 0x00002800, .div = 0x6},
+};
+
+struct {
+	unsigned long mrate;
+	unsigned long prate;
+} mclk_clk_table[] = {
+	{.mrate = 500000000, .prate = 0x00109801},
+	{.mrate = 525000000, .prate = 0x00104C00},
+	{.mrate = 550000000, .prate = 0x00105000},
+	{.mrate = 575000000, .prate = 0x00105400},
+	{.mrate = 600000000, .prate = 0x00105800},
+	{.mrate = 625000000, .prate = 0x00105C00},
+	{.mrate = 650000000, .prate = 0x00106000},
+	{.mrate = 675000000, .prate = 0x00106400},
+	{.mrate = 700000000, .prate = 0x00106800},
+	{.mrate = 725000000, .prate = 0x00106C00},
+	{.mrate = 750000000, .prate = 0x00107000},
+	{.mrate = 775000000, .prate = 0x00107400},
+	{.mrate = 800000000, .prate = 0x00107800},
+};
+
+int clk_set_rate(struct clk *clk, unsigned long rate)
+{
+	if (clk == &clk_vga_clk) {
+		unsigned long pll_vgacfg, pll_vgadiv;
+		int ret, i;
+
+		/* lookup vga_clk_table */
+		ret = -EINVAL;
+		for (i = 0; i < ARRAY_SIZE(vga_clk_table); i++) {
+			if (rate == vga_clk_table[i].rate) {
+				pll_vgacfg = vga_clk_table[i].cfg;
+				pll_vgadiv = vga_clk_table[i].div;
+				ret = 0;
+				break;
+			}
+		}
+
+		if (ret)
+			return ret;
+
+		if (readl(PM_PLLVGACFG) == pll_vgacfg)
+			return 0;
+
+		/* set pll vga cfg reg. */
+		writel(pll_vgacfg, PM_PLLVGACFG);
+
+		writel(PM_PMCR_CFBVGA, PM_PMCR);
+		while ((readl(PM_PLLDFCDONE) & PM_PLLDFCDONE_VGADFC)
+				!= PM_PLLDFCDONE_VGADFC)
+			udelay(100); /* about 1ms */
+
+		/* set div cfg reg. */
+		writel(readl(PM_PCGR) | PM_PCGR_VGACLK, PM_PCGR);
+
+		writel((readl(PM_DIVCFG) & ~PM_DIVCFG_VGACLK_MASK)
+				| PM_DIVCFG_VGACLK(pll_vgadiv), PM_DIVCFG);
+
+		writel(readl(PM_SWRESET) | PM_SWRESET_VGADIV, PM_SWRESET);
+		while ((readl(PM_SWRESET) & PM_SWRESET_VGADIV)
+				== PM_SWRESET_VGADIV)
+			udelay(100); /* 65536 bclk32, about 320us */
+
+		writel(readl(PM_PCGR) & ~PM_PCGR_VGACLK, PM_PCGR);
+	}
+#ifdef CONFIG_CPU_FREQ
+	if (clk == &clk_mclk_clk) {
+		u32 pll_rate, divstatus = PM_DIVSTATUS;
+		int ret, i;
+
+		/* lookup mclk_clk_table */
+		ret = -EINVAL;
+		for (i = 0; i < ARRAY_SIZE(mclk_clk_table); i++) {
+			if (rate == mclk_clk_table[i].mrate) {
+				pll_rate = mclk_clk_table[i].prate;
+				clk_mclk_clk.rate = mclk_clk_table[i].mrate;
+				ret = 0;
+				break;
+			}
+		}
+
+		if (ret)
+			return ret;
+
+		if (clk_mclk_clk.rate)
+			clk_bclk32_clk.rate = clk_mclk_clk.rate
+				/ (((divstatus & 0x0000f000) >> 12) + 1);
+
+		/* set pll sys cfg reg. */
+		PM_PLLSYSCFG = pll_rate;
+
+		PM_PMCR = PM_PMCR_CFBSYS;
+		while ((PM_PLLDFCDONE & PM_PLLDFCDONE_SYSDFC)
+				!= PM_PLLDFCDONE_SYSDFC)
+			udelay(100);
+			/* about 1ms */
+	}
+#endif
+	return 0;
+}
+EXPORT_SYMBOL(clk_set_rate);
+
+int clk_register(struct clk *clk)
+{
+	mutex_lock(&clocks_mutex);
+	list_add(&clk->node, &clocks);
+	mutex_unlock(&clocks_mutex);
+	printk(KERN_DEFAULT "PKUnity PM: %s %lu.%02luM\n", clk->name,
+		(clk->rate)/1000000, (clk->rate)/10000 % 100);
+	return 0;
+}
+EXPORT_SYMBOL(clk_register);
+
+void clk_unregister(struct clk *clk)
+{
+	mutex_lock(&clocks_mutex);
+	list_del(&clk->node);
+	mutex_unlock(&clocks_mutex);
+}
+EXPORT_SYMBOL(clk_unregister);
+
+struct {
+	unsigned long prate;
+	unsigned long rate;
+} pllrate_table[] = {
+	{.prate = 0x00002001, .rate = 250000000},
+	{.prate = 0x00104801, .rate = 250000000},
+	{.prate = 0x00104C01, .rate = 262500000},
+	{.prate = 0x00002401, .rate = 275000000},
+	{.prate = 0x00105001, .rate = 275000000},
+	{.prate = 0x00105401, .rate = 287500000},
+	{.prate = 0x00002801, .rate = 300000000},
+	{.prate = 0x00105801, .rate = 300000000},
+	{.prate = 0x00105C01, .rate = 312500000},
+	{.prate = 0x00002C01, .rate = 325000000},
+	{.prate = 0x00106001, .rate = 325000000},
+	{.prate = 0x00106401, .rate = 337500000},
+	{.prate = 0x00003001, .rate = 350000000},
+	{.prate = 0x00106801, .rate = 350000000},
+	{.prate = 0x00106C01, .rate = 362500000},
+	{.prate = 0x00003401, .rate = 375000000},
+	{.prate = 0x00107001, .rate = 375000000},
+	{.prate = 0x00107401, .rate = 387500000},
+	{.prate = 0x00003801, .rate = 400000000},
+	{.prate = 0x00107801, .rate = 400000000},
+	{.prate = 0x00107C01, .rate = 412500000},
+	{.prate = 0x00003C01, .rate = 425000000},
+	{.prate = 0x00108001, .rate = 425000000},
+	{.prate = 0x00108401, .rate = 437500000},
+	{.prate = 0x00004001, .rate = 450000000},
+	{.prate = 0x00108801, .rate = 450000000},
+	{.prate = 0x00108C01, .rate = 462500000},
+	{.prate = 0x00004401, .rate = 475000000},
+	{.prate = 0x00109001, .rate = 475000000},
+	{.prate = 0x00109401, .rate = 487500000},
+	{.prate = 0x00004801, .rate = 500000000},
+	{.prate = 0x00109801, .rate = 500000000},
+	{.prate = 0x00104C00, .rate = 525000000},
+	{.prate = 0x00002400, .rate = 550000000},
+	{.prate = 0x00105000, .rate = 550000000},
+	{.prate = 0x00105400, .rate = 575000000},
+	{.prate = 0x00002800, .rate = 600000000},
+	{.prate = 0x00105800, .rate = 600000000},
+	{.prate = 0x00105C00, .rate = 625000000},
+	{.prate = 0x00002C00, .rate = 650000000},
+	{.prate = 0x00106000, .rate = 650000000},
+	{.prate = 0x00106400, .rate = 675000000},
+	{.prate = 0x00003000, .rate = 700000000},
+	{.prate = 0x00106800, .rate = 700000000},
+	{.prate = 0x00106C00, .rate = 725000000},
+	{.prate = 0x00003400, .rate = 750000000},
+	{.prate = 0x00107000, .rate = 750000000},
+	{.prate = 0x00107400, .rate = 775000000},
+	{.prate = 0x00003800, .rate = 800000000},
+	{.prate = 0x00107800, .rate = 800000000},
+	{.prate = 0x00107C00, .rate = 825000000},
+	{.prate = 0x00003C00, .rate = 850000000},
+	{.prate = 0x00108000, .rate = 850000000},
+	{.prate = 0x00108400, .rate = 875000000},
+	{.prate = 0x00004000, .rate = 900000000},
+	{.prate = 0x00108800, .rate = 900000000},
+	{.prate = 0x00108C00, .rate = 925000000},
+	{.prate = 0x00004400, .rate = 950000000},
+	{.prate = 0x00109000, .rate = 950000000},
+	{.prate = 0x00109400, .rate = 975000000},
+	{.prate = 0x00004800, .rate = 1000000000},
+	{.prate = 0x00109800, .rate = 1000000000},
+};
+
+struct {
+	unsigned long prate;
+	unsigned long drate;
+} pddr_table[] = {
+	{.prate = 0x00100800, .drate = 44236800},
+	{.prate = 0x00100C00, .drate = 66355200},
+	{.prate = 0x00101000, .drate = 88473600},
+	{.prate = 0x00101400, .drate = 110592000},
+	{.prate = 0x00101800, .drate = 132710400},
+	{.prate = 0x00101C01, .drate = 154828800},
+	{.prate = 0x00102001, .drate = 176947200},
+	{.prate = 0x00102401, .drate = 199065600},
+	{.prate = 0x00102801, .drate = 221184000},
+	{.prate = 0x00102C01, .drate = 243302400},
+	{.prate = 0x00103001, .drate = 265420800},
+	{.prate = 0x00103401, .drate = 287539200},
+	{.prate = 0x00103801, .drate = 309657600},
+	{.prate = 0x00103C01, .drate = 331776000},
+	{.prate = 0x00104001, .drate = 353894400},
+};
+
+static int __init clk_init(void)
+{
+#ifdef CONFIG_PUV3_PM
+	u32 pllrate, divstatus = readl(PM_DIVSTATUS);
+	u32 pcgr_val = readl(PM_PCGR);
+	int i;
+
+	pcgr_val |= PM_PCGR_BCLKMME | PM_PCGR_BCLKH264E | PM_PCGR_BCLKH264D
+			| PM_PCGR_HECLK | PM_PCGR_HDCLK;
+	writel(pcgr_val, PM_PCGR);
+
+	pllrate = readl(PM_PLLSYSSTATUS);
+
+	/* lookup pmclk_table */
+	clk_mclk_clk.rate = 0;
+	for (i = 0; i < ARRAY_SIZE(pllrate_table); i++) {
+		if (pllrate == pllrate_table[i].prate) {
+			clk_mclk_clk.rate = pllrate_table[i].rate;
+			break;
+		}
+	}
+
+	if (clk_mclk_clk.rate)
+		clk_bclk32_clk.rate = clk_mclk_clk.rate /
+			(((divstatus & 0x0000f000) >> 12) + 1);
+
+	pllrate = readl(PM_PLLDDRSTATUS);
+
+	/* lookup pddr_table */
+	clk_ddr_clk.rate = 0;
+	for (i = 0; i < ARRAY_SIZE(pddr_table); i++) {
+		if (pllrate == pddr_table[i].prate) {
+			clk_ddr_clk.rate = pddr_table[i].drate;
+			break;
+		}
+	}
+
+	pllrate = readl(PM_PLLVGASTATUS);
+
+	/* lookup pvga_table */
+	clk_vga_clk.rate = 0;
+	for (i = 0; i < ARRAY_SIZE(pllrate_table); i++) {
+		if (pllrate == pllrate_table[i].prate) {
+			clk_vga_clk.rate = pllrate_table[i].rate;
+			break;
+		}
+	}
+
+	if (clk_vga_clk.rate)
+		clk_vga_clk.rate = clk_vga_clk.rate /
+			(((divstatus & 0x00f00000) >> 20) + 1);
+
+	clk_register(&clk_vga_clk);
+#endif
+#ifdef CONFIG_ARCH_FPGA
+	clk_ddr_clk.rate = 33000000;
+	clk_mclk_clk.rate = 33000000;
+	clk_bclk32_clk.rate = 33000000;
+#endif
+	clk_register(&clk_ddr_clk);
+	clk_register(&clk_mclk_clk);
+	clk_register(&clk_bclk32_clk);
+	clk_register(&clk_ost_clk);
+	return 0;
+}
+core_initcall(clk_init);
diff --git a/arch/unicore32/kernel/cpu-ucv2.c b/arch/unicore32/kernel/cpu-ucv2.c
new file mode 100644
index 000000000000..4a99f62584c7
--- /dev/null
+++ b/arch/unicore32/kernel/cpu-ucv2.c
@@ -0,0 +1,93 @@
+/*
+ * linux/arch/unicore32/kernel/cpu-ucv2.c: clock scaling for the UniCore-II
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ *	Maintained by GUAN Xue-tao <gxt@mprc.pku.edu.cn>
+ *	Copyright (C) 2001-2010 Guan Xuetao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/clk.h>
+#include <linux/cpufreq.h>
+
+#include <mach/hardware.h>
+
+static struct cpufreq_driver ucv2_driver;
+
+/* make sure that only the "userspace" governor is run
+ * -- anything else wouldn't make sense on this platform, anyway.
+ */
+int ucv2_verify_speed(struct cpufreq_policy *policy)
+{
+	if (policy->cpu)
+		return -EINVAL;
+
+	cpufreq_verify_within_limits(policy,
+			policy->cpuinfo.min_freq, policy->cpuinfo.max_freq);
+
+	return 0;
+}
+
+static unsigned int ucv2_getspeed(unsigned int cpu)
+{
+	struct clk *mclk = clk_get(NULL, "MAIN_CLK");
+
+	if (cpu)
+		return 0;
+	return clk_get_rate(mclk)/1000;
+}
+
+static int ucv2_target(struct cpufreq_policy *policy,
+			 unsigned int target_freq,
+			 unsigned int relation)
+{
+	unsigned int cur = ucv2_getspeed(0);
+	struct cpufreq_freqs freqs;
+	struct clk *mclk = clk_get(NULL, "MAIN_CLK");
+
+	cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE);
+
+	if (!clk_set_rate(mclk, target_freq * 1000)) {
+		freqs.old = cur;
+		freqs.new = target_freq;
+		freqs.cpu = 0;
+	}
+
+	cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE);
+
+	return 0;
+}
+
+static int __init ucv2_cpu_init(struct cpufreq_policy *policy)
+{
+	if (policy->cpu != 0)
+		return -EINVAL;
+	policy->cur = ucv2_getspeed(0);
+	policy->min = policy->cpuinfo.min_freq = 250000;
+	policy->max = policy->cpuinfo.max_freq = 1000000;
+	policy->cpuinfo.transition_latency = CPUFREQ_ETERNAL;
+	return 0;
+}
+
+static struct cpufreq_driver ucv2_driver = {
+	.flags		= CPUFREQ_STICKY,
+	.verify		= ucv2_verify_speed,
+	.target		= ucv2_target,
+	.get		= ucv2_getspeed,
+	.init		= ucv2_cpu_init,
+	.name		= "UniCore-II",
+};
+
+static int __init ucv2_cpufreq_init(void)
+{
+	return cpufreq_register_driver(&ucv2_driver);
+}
+
+arch_initcall(ucv2_cpufreq_init);
diff --git a/arch/unicore32/kernel/debug-macro.S b/arch/unicore32/kernel/debug-macro.S
new file mode 100644
index 000000000000..2711d6d87d8e
--- /dev/null
+++ b/arch/unicore32/kernel/debug-macro.S
@@ -0,0 +1,89 @@
+/*
+ * linux/arch/unicore32/kernel/debug-macro.S
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * Debugging macro include header
+ */
+#include <generated/asm-offsets.h>
+#include <mach/hardware.h>
+
+		.macro	put_word_ocd, rd, rx=r16
+1001:		movc		\rx, p1.c0, #0
+		cand.a	\rx, #2
+		bne	1001b
+		movc		p1.c1, \rd, #1
+		.endm
+
+#ifdef CONFIG_DEBUG_OCD
+		/* debug using UniCore On-Chip-Debugger */
+		.macro	addruart, rx
+		.endm
+
+		.macro	senduart, rd, rx
+		put_word_ocd	\rd, \rx
+		.endm
+
+		.macro	busyuart, rd, rx
+		.endm
+
+		.macro	waituart, rd, rx
+		.endm
+#else
+#define UART_CLK_DEFAULT        3686400 * 20
+	/* Uartclk = MCLK/ 2, The MCLK on my board is 3686400 * 40  */
+#define BAUD_RATE_DEFAULT	115200
+	/* The baud rate of the serial port */
+
+#define UART_DIVISOR_DEFAULT	(UART_CLK_DEFAULT \
+				/ (16 * BAUD_RATE_DEFAULT) - 1)
+
+		.macro	addruart,rx
+		mrc	p0, #0, \rx, c1, c0
+		tst	\rx, #1			@ MMU enabled?
+		moveq	\rx, #0xee000000	@ physical base address
+		movne	\rx, #0x6e000000	@ virtual address
+
+		@ We probe for the active serial port here
+		@ However, now we assume UART0 is active:	epip4d
+		@ We assume r1 and r2 can be clobbered.
+
+		movl 	r2, #UART_DIVISOR_DEFAULT
+		mov 	r1, #0x80
+		str	r1, [\rx, #UART_LCR_OFFSET]
+		and	r1, r2, #0xff00
+		mov	r1, r1, lsr #8
+		str	r1, [\rx, #UART_DLH_OFFSET]
+		and	r1, r2, #0xff
+		str	r1, [\rx, #UART_DLL_OFFSET]
+		mov 	r1, #0x7
+		str	r1, [\rx, #UART_FCR_OFFSET]
+		mov 	r1, #0x3
+		str	r1, [\rx, #UART_LCR_OFFSET]
+		mov 	r1, #0x0
+		str	r1, [\rx, #UART_IER_OFFSET]
+		.endm
+
+		.macro	senduart,rd,rx
+		str	\rd, [\rx, #UART_THR_OFFSET]
+		.endm
+
+		.macro	waituart,rd,rx
+1001:		ldr	\rd, [\rx, #UART_LSR_OFFSET]
+		tst	\rd, #UART_LSR_THRE
+		beq	1001b
+		.endm
+
+		.macro	busyuart,rd,rx
+1001:		ldr	\rd, [\rx, #UART_LSR_OFFSET]
+		tst	\rd, #UART_LSR_TEMT
+		bne	1001b
+		.endm
+#endif
+
diff --git a/arch/unicore32/kernel/debug.S b/arch/unicore32/kernel/debug.S
new file mode 100644
index 000000000000..029fd12f6ab0
--- /dev/null
+++ b/arch/unicore32/kernel/debug.S
@@ -0,0 +1,85 @@
+/*
+ * linux/arch/unicore32/kernel/debug.S
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ *  32-bit debugging code
+ */
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+		.text
+
+/*
+ * Some debugging routines (useful if you've got MM problems and
+ * printk isn't working).  For DEBUGGING ONLY!!!  Do not leave
+ * references to these in a production kernel!
+ */
+#include "debug-macro.S"
+
+/*
+ * Useful debugging routines
+ */
+ENTRY(printhex8)
+		mov	r1, #8
+		b	printhex
+ENDPROC(printhex8)
+
+ENTRY(printhex4)
+		mov	r1, #4
+		b	printhex
+ENDPROC(printhex4)
+
+ENTRY(printhex2)
+		mov	r1, #2
+printhex:	adr	r2, hexbuf
+		add	r3, r2, r1
+		mov	r1, #0
+		stb	r1, [r3]
+1:		and	r1, r0, #15
+		mov	r0, r0 >> #4
+		csub.a	r1, #10
+		beg	2f
+		add	r1, r1, #'0' - 'a' + 10
+2:		add	r1, r1, #'a' - 10
+		stb.w	r1, [r3+], #-1
+		cxor.a	r3, r2
+		bne	1b
+		mov	r0, r2
+		b	printascii
+ENDPROC(printhex2)
+
+		.ltorg
+
+ENTRY(printascii)
+		addruart r3
+		b	2f
+1:		waituart r2, r3
+		senduart r1, r3
+		busyuart r2, r3
+		cxor.a	r1, #'\n'
+		cmoveq	r1, #'\r'
+		beq	1b
+2:		cxor.a	r0, #0
+		beq	3f
+		ldb.w	r1, [r0]+, #1
+		cxor.a	r1, #0
+		bne	1b
+3:		mov	pc, lr
+ENDPROC(printascii)
+
+ENTRY(printch)
+		addruart r3
+		mov	r1, r0
+		mov	r0, #0
+		b	1b
+ENDPROC(printch)
+
+hexbuf:		.space 16
+
diff --git a/arch/unicore32/kernel/dma.c b/arch/unicore32/kernel/dma.c
new file mode 100644
index 000000000000..ae441bc3122c
--- /dev/null
+++ b/arch/unicore32/kernel/dma.c
@@ -0,0 +1,183 @@
+/*
+ * linux/arch/unicore32/kernel/dma.c
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ *	Maintained by GUAN Xue-tao <gxt@mprc.pku.edu.cn>
+ *	Copyright (C) 2001-2010 Guan Xuetao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/interrupt.h>
+#include <linux/errno.h>
+#include <linux/io.h>
+
+#include <asm/system.h>
+#include <asm/irq.h>
+#include <mach/hardware.h>
+#include <mach/dma.h>
+
+struct dma_channel {
+	char *name;
+	puv3_dma_prio prio;
+	void (*irq_handler)(int, void *);
+	void (*err_handler)(int, void *);
+	void *data;
+};
+
+static struct dma_channel dma_channels[MAX_DMA_CHANNELS];
+
+int puv3_request_dma(char *name, puv3_dma_prio prio,
+			 void (*irq_handler)(int, void *),
+			 void (*err_handler)(int, void *),
+			 void *data)
+{
+	unsigned long flags;
+	int i, found = 0;
+
+	/* basic sanity checks */
+	if (!name)
+		return -EINVAL;
+
+	local_irq_save(flags);
+
+	do {
+		/* try grabbing a DMA channel with the requested priority */
+		for (i = 0; i < MAX_DMA_CHANNELS; i++) {
+			if ((dma_channels[i].prio == prio) &&
+			    !dma_channels[i].name) {
+				found = 1;
+				break;
+			}
+		}
+		/* if requested prio group is full, try a hier priority */
+	} while (!found && prio--);
+
+	if (found) {
+		dma_channels[i].name = name;
+		dma_channels[i].irq_handler = irq_handler;
+		dma_channels[i].err_handler = err_handler;
+		dma_channels[i].data = data;
+	} else {
+		printk(KERN_WARNING "No more available DMA channels for %s\n",
+				name);
+		i = -ENODEV;
+	}
+
+	local_irq_restore(flags);
+	return i;
+}
+EXPORT_SYMBOL(puv3_request_dma);
+
+void puv3_free_dma(int dma_ch)
+{
+	unsigned long flags;
+
+	if (!dma_channels[dma_ch].name) {
+		printk(KERN_CRIT
+			"%s: trying to free channel %d which is already freed\n",
+			__func__, dma_ch);
+		return;
+	}
+
+	local_irq_save(flags);
+	dma_channels[dma_ch].name = NULL;
+	dma_channels[dma_ch].err_handler = NULL;
+	local_irq_restore(flags);
+}
+EXPORT_SYMBOL(puv3_free_dma);
+
+static irqreturn_t dma_irq_handler(int irq, void *dev_id)
+{
+	int i, dint;
+
+	dint = readl(DMAC_ITCSR);
+	for (i = 0; i < MAX_DMA_CHANNELS; i++) {
+		if (dint & DMAC_CHANNEL(i)) {
+			struct dma_channel *channel = &dma_channels[i];
+
+			/* Clear TC interrupt of channel i */
+			writel(DMAC_CHANNEL(i), DMAC_ITCCR);
+			writel(0, DMAC_ITCCR);
+
+			if (channel->name && channel->irq_handler) {
+				channel->irq_handler(i, channel->data);
+			} else {
+				/*
+				 * IRQ for an unregistered DMA channel:
+				 * let's clear the interrupts and disable it.
+				 */
+				printk(KERN_WARNING "spurious IRQ for"
+						" DMA channel %d\n", i);
+			}
+		}
+	}
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t dma_err_handler(int irq, void *dev_id)
+{
+	int i, dint;
+
+	dint = readl(DMAC_IESR);
+	for (i = 0; i < MAX_DMA_CHANNELS; i++) {
+		if (dint & DMAC_CHANNEL(i)) {
+			struct dma_channel *channel = &dma_channels[i];
+
+			/* Clear Err interrupt of channel i */
+			writel(DMAC_CHANNEL(i), DMAC_IECR);
+			writel(0, DMAC_IECR);
+
+			if (channel->name && channel->err_handler) {
+				channel->err_handler(i, channel->data);
+			} else {
+				/*
+				 * IRQ for an unregistered DMA channel:
+				 * let's clear the interrupts and disable it.
+				 */
+				printk(KERN_WARNING "spurious IRQ for"
+						" DMA channel %d\n", i);
+			}
+		}
+	}
+	return IRQ_HANDLED;
+}
+
+int __init puv3_init_dma(void)
+{
+	int i, ret;
+
+	/* dma channel priorities on v8 processors:
+	 * ch 0 - 1  <--> (0) DMA_PRIO_HIGH
+	 * ch 2 - 3  <--> (1) DMA_PRIO_MEDIUM
+	 * ch 4 - 5  <--> (2) DMA_PRIO_LOW
+	 */
+	for (i = 0; i < MAX_DMA_CHANNELS; i++) {
+		puv3_stop_dma(i);
+		dma_channels[i].name = NULL;
+		dma_channels[i].prio = min((i & 0x7) >> 1, DMA_PRIO_LOW);
+	}
+
+	ret = request_irq(IRQ_DMA, dma_irq_handler, 0, "DMA", NULL);
+	if (ret) {
+		printk(KERN_CRIT "Can't register IRQ for DMA\n");
+		return ret;
+	}
+
+	ret = request_irq(IRQ_DMAERR, dma_err_handler, 0, "DMAERR", NULL);
+	if (ret) {
+		printk(KERN_CRIT "Can't register IRQ for DMAERR\n");
+		free_irq(IRQ_DMA, "DMA");
+		return ret;
+	}
+
+	return 0;
+}
+
+postcore_initcall(puv3_init_dma);
diff --git a/arch/unicore32/kernel/early_printk.c b/arch/unicore32/kernel/early_printk.c
new file mode 100644
index 000000000000..3922255f1fa8
--- /dev/null
+++ b/arch/unicore32/kernel/early_printk.c
@@ -0,0 +1,59 @@
+/*
+ * linux/arch/unicore32/kernel/early_printk.c
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/console.h>
+#include <linux/init.h>
+#include <linux/string.h>
+#include <mach/ocd.h>
+
+/* On-Chip-Debugger functions */
+
+static void early_ocd_write(struct console *con, const char *s, unsigned n)
+{
+	while (*s && n-- > 0) {
+		if (*s == '\n')
+			ocd_putc((int)'\r');
+		ocd_putc((int)*s);
+		s++;
+	}
+}
+
+static struct console early_ocd_console = {
+	.name =		"earlyocd",
+	.write =	early_ocd_write,
+	.flags =	CON_PRINTBUFFER,
+	.index =	-1,
+};
+
+/* Direct interface for emergencies */
+static struct console *early_console = &early_ocd_console;
+
+static int __initdata keep_early;
+
+static int __init setup_early_printk(char *buf)
+{
+	if (!buf)
+		return 0;
+
+	if (strstr(buf, "keep"))
+		keep_early = 1;
+
+	if (!strncmp(buf, "ocd", 3))
+		early_console = &early_ocd_console;
+
+	if (keep_early)
+		early_console->flags &= ~CON_BOOT;
+	else
+		early_console->flags |= CON_BOOT;
+	register_console(early_console);
+	return 0;
+}
+early_param("earlyprintk", setup_early_printk);
diff --git a/arch/unicore32/kernel/elf.c b/arch/unicore32/kernel/elf.c
new file mode 100644
index 000000000000..0a176734fefa
--- /dev/null
+++ b/arch/unicore32/kernel/elf.c
@@ -0,0 +1,38 @@
+/*
+ * linux/arch/unicore32/kernel/elf.c
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/personality.h>
+#include <linux/binfmts.h>
+#include <linux/elf.h>
+
+int elf_check_arch(const struct elf32_hdr *x)
+{
+	/* Make sure it's an UniCore executable */
+	if (x->e_machine != EM_UNICORE)
+		return 0;
+
+	/* Make sure the entry address is reasonable */
+	if (x->e_entry & 3)
+		return 0;
+
+	return 1;
+}
+EXPORT_SYMBOL(elf_check_arch);
+
+void elf_set_personality(const struct elf32_hdr *x)
+{
+	unsigned int personality = PER_LINUX;
+
+	set_personality(personality);
+}
+EXPORT_SYMBOL(elf_set_personality);
diff --git a/arch/unicore32/kernel/entry.S b/arch/unicore32/kernel/entry.S
new file mode 100644
index 000000000000..00a259f9819e
--- /dev/null
+++ b/arch/unicore32/kernel/entry.S
@@ -0,0 +1,824 @@
+/*
+ * linux/arch/unicore32/kernel/entry.S
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ *  Low-level vector interface routines
+ */
+#include <linux/init.h>
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+#include <asm/errno.h>
+#include <asm/thread_info.h>
+#include <asm/memory.h>
+#include <asm/unistd.h>
+#include <generated/asm-offsets.h>
+#include "debug-macro.S"
+
+@
+@ Most of the stack format comes from struct pt_regs, but with
+@ the addition of 8 bytes for storing syscall args 5 and 6.
+@
+#define S_OFF		8
+
+/*
+ * The SWI code relies on the fact that R0 is at the bottom of the stack
+ * (due to slow/fast restore user regs).
+ */
+#if S_R0 != 0
+#error "Please fix"
+#endif
+
+	.macro	zero_fp
+#ifdef CONFIG_FRAME_POINTER
+	mov	fp, #0
+#endif
+	.endm
+
+	.macro	alignment_trap, rtemp
+#ifdef CONFIG_ALIGNMENT_TRAP
+	ldw	\rtemp, .LCcralign
+	ldw	\rtemp, [\rtemp]
+	movc	p0.c1, \rtemp, #0
+#endif
+	.endm
+
+	.macro	load_user_sp_lr, rd, rtemp, offset = 0
+	mov	\rtemp, asr
+	xor	\rtemp, \rtemp, #(PRIV_MODE ^ SUSR_MODE)
+	mov.a	asr, \rtemp			@ switch to the SUSR mode
+
+	ldw	sp, [\rd+], #\offset		@ load sp_user
+	ldw	lr, [\rd+], #\offset + 4	@ load lr_user
+
+	xor	\rtemp, \rtemp, #(PRIV_MODE ^ SUSR_MODE)
+	mov.a	asr, \rtemp			@ switch back to the PRIV mode
+	.endm
+
+	.macro	priv_exit, rpsr
+	mov.a	bsr, \rpsr
+	ldm.w	(r0 - r15), [sp]+
+	ldm.b	(r16 - pc), [sp]+		@ load r0 - pc, asr
+	.endm
+
+	.macro	restore_user_regs, fast = 0, offset = 0
+	ldw	r1, [sp+], #\offset + S_PSR	@ get calling asr
+	ldw	lr, [sp+], #\offset + S_PC	@ get pc
+	mov.a	bsr, r1				@ save in bsr_priv
+	.if	\fast
+	add	sp, sp, #\offset + S_R1		@ r0 is syscall return value
+	ldm.w	(r1 - r15), [sp]+		@ get calling r1 - r15
+	ldur	(r16 - lr), [sp]+		@ get calling r16 - lr
+	.else
+	ldm.w	(r0 - r15), [sp]+		@ get calling r0 - r15
+	ldur	(r16 - lr), [sp]+		@ get calling r16 - lr
+	.endif
+	nop
+	add	sp, sp, #S_FRAME_SIZE - S_R16
+	mov.a	pc, lr				@ return
+						@ and move bsr_priv into asr
+	.endm
+
+	.macro	get_thread_info, rd
+	mov	\rd, sp >> #13
+	mov	\rd, \rd << #13
+	.endm
+
+	.macro	get_irqnr_and_base, irqnr, irqstat, base, tmp
+	ldw	\base, =(PKUNITY_INTC_BASE)
+	ldw	\irqstat, [\base+], #0xC	@ INTC_ICIP
+	ldw	\tmp,	  [\base+], #0x4	@ INTC_ICMR
+	and.a	\irqstat, \irqstat, \tmp
+	beq	1001f
+	cntlz	\irqnr, \irqstat
+	rsub	\irqnr, \irqnr, #31
+1001:	/* EQ will be set if no irqs pending */
+	.endm
+
+#ifdef CONFIG_DEBUG_LL
+	.macro	printreg, reg, temp
+		adr	\temp, 901f
+		stm	(r0-r3), [\temp]+
+		stw	lr, [\temp+], #0x10
+		mov	r0, \reg
+		b.l	printhex8
+		mov	r0, #':'
+		b.l	printch
+		mov	r0, pc
+		b.l	printhex8
+		adr	r0, 902f
+		b.l	printascii
+		adr	\temp, 901f
+		ldm	(r0-r3), [\temp]+
+		ldw	lr, [\temp+], #0x10
+		b	903f
+901:	.word	0, 0, 0, 0, 0	@ r0-r3, lr
+902:	.asciz	": epip4d\n"
+	.align
+903:
+	.endm
+#endif
+
+/*
+ * These are the registers used in the syscall handler, and allow us to
+ * have in theory up to 7 arguments to a function - r0 to r6.
+ *
+ * Note that tbl == why is intentional.
+ *
+ * We must set at least "tsk" and "why" when calling ret_with_reschedule.
+ */
+scno	.req	r21		@ syscall number
+tbl	.req	r22		@ syscall table pointer
+why	.req	r22		@ Linux syscall (!= 0)
+tsk	.req	r23		@ current thread_info
+
+/*
+ * Interrupt handling.  Preserves r17, r18, r19
+ */
+	.macro	intr_handler
+1:	get_irqnr_and_base r0, r6, r5, lr
+	beq	2f
+	mov	r1, sp
+	@
+	@ routine called with r0 = irq number, r1 = struct pt_regs *
+	@
+	adr	lr, 1b
+	b	asm_do_IRQ
+2:
+	.endm
+
+/*
+ * PRIV mode handlers
+ */
+	.macro	priv_entry
+	sub	sp, sp, #(S_FRAME_SIZE - 4)
+	stm	(r1 - r15), [sp]+
+	add	r5, sp, #S_R15
+	stm	(r16 - r28), [r5]+
+
+	ldm	(r1 - r3), [r0]+
+	add	r5, sp, #S_SP - 4	@ here for interlock avoidance
+	mov	r4, #-1			@  ""  ""      ""       ""
+	add	r0, sp, #(S_FRAME_SIZE - 4)
+	stw.w	r1, [sp+], #-4		@ save the "real" r0 copied
+					@ from the exception stack
+
+	mov	r1, lr
+
+	@
+	@ We are now ready to fill in the remaining blanks on the stack:
+	@
+	@  r0 - sp_priv
+	@  r1 - lr_priv
+	@  r2 - lr_<exception>, already fixed up for correct return/restart
+	@  r3 - bsr_<exception>
+	@  r4 - orig_r0 (see pt_regs definition in ptrace.h)
+	@
+	stm	(r0 - r4), [r5]+
+	.endm
+
+/*
+ * User mode handlers
+ *
+ */
+	.macro	user_entry
+	sub	sp, sp, #S_FRAME_SIZE
+	stm	(r1 - r15), [sp+]
+	add	r4, sp, #S_R16
+	stm	(r16 - r28), [r4]+
+
+	ldm	(r1 - r3), [r0]+
+	add	r0, sp, #S_PC		@ here for interlock avoidance
+	mov	r4, #-1			@  ""  ""     ""        ""
+
+	stw	r1, [sp]		@ save the "real" r0 copied
+					@ from the exception stack
+
+	@
+	@ We are now ready to fill in the remaining blanks on the stack:
+	@
+	@  r2 - lr_<exception>, already fixed up for correct return/restart
+	@  r3 - bsr_<exception>
+	@  r4 - orig_r0 (see pt_regs definition in ptrace.h)
+	@
+	@ Also, separately save sp_user and lr_user
+	@
+	stm	(r2 - r4), [r0]+
+	stur	(sp, lr), [r0-]
+
+	@
+	@ Enable the alignment trap while in kernel mode
+	@
+	alignment_trap r0
+
+	@
+	@ Clear FP to mark the first stack frame
+	@
+	zero_fp
+	.endm
+
+	.text
+
+@
+@ __invalid - generic code for failed exception
+@			(re-entrant version of handlers)
+@
+__invalid:
+	sub	sp, sp, #S_FRAME_SIZE
+	stm	(r1 - r15), [sp+]
+	add	r1, sp, #S_R16
+	stm	(r16 - r28, sp, lr), [r1]+
+
+	zero_fp
+
+	ldm	(r4 - r6), [r0]+
+	add	r0, sp, #S_PC		@ here for interlock avoidance
+	mov	r7, #-1			@  ""   ""    ""        ""
+	stw	r4, [sp]		@ save preserved r0
+	stm	(r5 - r7), [r0]+	@ lr_<exception>,
+					@ asr_<exception>, "old_r0"
+
+	mov	r0, sp
+	mov	r1, asr
+	b	bad_mode
+ENDPROC(__invalid)
+
+	.align	5
+__dabt_priv:
+	priv_entry
+
+	@
+	@ get ready to re-enable interrupts if appropriate
+	@
+	mov	r17, asr
+	cand.a	r3, #PSR_I_BIT
+	bne	1f
+	andn	r17, r17, #PSR_I_BIT
+1:
+
+	@
+	@ Call the processor-specific abort handler:
+	@
+	@  r2 - aborted context pc
+	@  r3 - aborted context asr
+	@
+	@ The abort handler must return the aborted address in r0, and
+	@ the fault status register in r1.
+	@
+	movc	r1, p0.c3, #0		@ get FSR
+	movc	r0, p0.c4, #0		@ get FAR
+
+	@
+	@ set desired INTR state, then call main handler
+	@
+	mov.a	asr, r17
+	mov	r2, sp
+	b.l	do_DataAbort
+
+	@
+	@ INTRs off again before pulling preserved data off the stack
+	@
+	disable_irq r0
+
+	@
+	@ restore BSR and restart the instruction
+	@
+	ldw	r2, [sp+], #S_PSR
+	priv_exit r2				@ return from exception
+ENDPROC(__dabt_priv)
+
+	.align	5
+__intr_priv:
+	priv_entry
+
+	intr_handler
+
+	mov	r0, #0				@ epip4d
+	movc	p0.c5, r0, #14
+	nop; nop; nop; nop; nop; nop; nop; nop
+
+	ldw	r4, [sp+], #S_PSR		@ irqs are already disabled
+
+	priv_exit r4				@ return from exception
+ENDPROC(__intr_priv)
+
+	.ltorg
+
+	.align	5
+__extn_priv:
+	priv_entry
+
+	mov	r0, sp				@ struct pt_regs *regs
+	mov	r1, asr
+	b	bad_mode			@ not supported
+ENDPROC(__extn_priv)
+
+	.align	5
+__pabt_priv:
+	priv_entry
+
+	@
+	@ re-enable interrupts if appropriate
+	@
+	mov	r17, asr
+	cand.a	r3, #PSR_I_BIT
+	bne	1f
+	andn	r17, r17, #PSR_I_BIT
+1:
+
+	@
+	@ set args, then call main handler
+	@
+	@  r0 - address of faulting instruction
+	@  r1 - pointer to registers on stack
+	@
+	mov	r0, r2			@ pass address of aborted instruction
+	mov	r1, #5
+	mov.a	asr, r17
+	mov	r2, sp			@ regs
+	b.l	do_PrefetchAbort	@ call abort handler
+
+	@
+	@ INTRs off again before pulling preserved data off the stack
+	@
+	disable_irq r0
+
+	@
+	@ restore BSR and restart the instruction
+	@
+	ldw	r2, [sp+], #S_PSR
+	priv_exit r2			@ return from exception
+ENDPROC(__pabt_priv)
+
+	.align	5
+.LCcralign:
+	.word	cr_alignment
+
+	.align	5
+__dabt_user:
+	user_entry
+
+#ifdef CONFIG_UNICORE_FPU_F64
+	cff	ip, s31
+	cand.a	ip, #0x08000000		@ FPU execption traps?
+	beq	209f
+
+	ldw	ip, [sp+], #S_PC
+	add	ip, ip, #4
+	stw	ip, [sp+], #S_PC
+	@
+	@ fall through to the emulation code, which returns using r19 if
+	@ it has emulated the instruction, or the more conventional lr
+	@ if we are to treat this as a real extended instruction
+	@
+	@  r0 - instruction
+	@
+1:	ldw.u	r0, [r2]
+	adr	r19, ret_from_exception
+	adr	lr, 209f
+	@
+	@ fallthrough to call do_uc_f64
+	@
+/*
+ * Check whether the instruction is a co-processor instruction.
+ * If yes, we need to call the relevant co-processor handler.
+ *
+ * Note that we don't do a full check here for the co-processor
+ * instructions; all instructions with bit 27 set are well
+ * defined.  The only instructions that should fault are the
+ * co-processor instructions.
+ *
+ * Emulators may wish to make use of the following registers:
+ *  r0  = instruction opcode.
+ *  r2  = PC
+ *  r19 = normal "successful" return address
+ *  r20 = this threads thread_info structure.
+ *  lr  = unrecognised instruction return address
+ */
+	get_thread_info r20			@ get current thread
+	and	r8, r0, #0x00003c00		@ mask out CP number
+	mov	r7, #1
+	stb	r7, [r20+], #TI_USED_CP + 2	@ set appropriate used_cp[]
+
+	@ F64 hardware support entry point.
+	@  r0  = faulted instruction
+	@  r19 = return address
+	@  r20 = fp_state
+	enable_irq r4
+	add	r20, r20, #TI_FPSTATE	@ r20 = workspace
+	cff	r1, s31			@ get fpu FPSCR
+	andn    r2, r1, #0x08000000
+	ctf     r2, s31			@ clear 27 bit
+	mov	r2, sp			@ nothing stacked - regdump is at TOS
+	mov	lr, r19			@ setup for a return to the user code
+
+	@ Now call the C code to package up the bounce to the support code
+	@   r0 holds the trigger instruction
+	@   r1 holds the FPSCR value
+	@   r2 pointer to register dump
+	b	ucf64_exchandler
+209:
+#endif
+	@
+	@ Call the processor-specific abort handler:
+	@
+	@  r2 - aborted context pc
+	@  r3 - aborted context asr
+	@
+	@ The abort handler must return the aborted address in r0, and
+	@ the fault status register in r1.
+	@
+	movc	r1, p0.c3, #0		@ get FSR
+	movc	r0, p0.c4, #0		@ get FAR
+
+	@
+	@ INTRs on, then call the main handler
+	@
+	enable_irq r2
+	mov	r2, sp
+	adr	lr, ret_from_exception
+	b	do_DataAbort
+ENDPROC(__dabt_user)
+
+	.align	5
+__intr_user:
+	user_entry
+
+	get_thread_info tsk
+
+	intr_handler
+
+	mov	why, #0
+	b	ret_to_user
+ENDPROC(__intr_user)
+
+	.ltorg
+
+	.align	5
+__extn_user:
+	user_entry
+
+	mov	r0, sp
+	mov	r1, asr
+	b	bad_mode
+ENDPROC(__extn_user)
+
+	.align	5
+__pabt_user:
+	user_entry
+
+	mov	r0, r2			@ pass address of aborted instruction.
+	mov	r1, #5
+	enable_irq r1			@ Enable interrupts
+	mov	r2, sp			@ regs
+	b.l	do_PrefetchAbort	@ call abort handler
+	/* fall through */
+/*
+ * This is the return code to user mode for abort handlers
+ */
+ENTRY(ret_from_exception)
+	get_thread_info tsk
+	mov	why, #0
+	b	ret_to_user
+ENDPROC(__pabt_user)
+ENDPROC(ret_from_exception)
+
+/*
+ * Register switch for UniCore V2 processors
+ * r0 = previous task_struct, r1 = previous thread_info, r2 = next thread_info
+ * previous and next are guaranteed not to be the same.
+ */
+ENTRY(__switch_to)
+	add	ip, r1, #TI_CPU_SAVE
+	stm.w	(r4 - r15), [ip]+
+	stm.w	(r16 - r27, sp, lr), [ip]+
+
+#ifdef	CONFIG_UNICORE_FPU_F64
+	add	ip, r1, #TI_FPSTATE
+	sfm.w	(f0  - f7 ), [ip]+
+	sfm.w	(f8  - f15), [ip]+
+	sfm.w	(f16 - f23), [ip]+
+	sfm.w	(f24 - f31), [ip]+
+	cff	r4, s31
+	stw	r4, [ip]
+
+	add	ip, r2, #TI_FPSTATE
+	lfm.w	(f0  - f7 ), [ip]+
+	lfm.w	(f8  - f15), [ip]+
+	lfm.w	(f16 - f23), [ip]+
+	lfm.w	(f24 - f31), [ip]+
+	ldw	r4, [ip]
+	ctf	r4, s31
+#endif
+	add	ip, r2, #TI_CPU_SAVE
+	ldm.w	(r4 - r15), [ip]+
+	ldm	(r16 - r27, sp, pc), [ip]+	@ Load all regs saved previously
+ENDPROC(__switch_to)
+
+	.align	5
+/*
+ * This is the fast syscall return path.  We do as little as
+ * possible here, and this includes saving r0 back into the PRIV
+ * stack.
+ */
+ret_fast_syscall:
+	disable_irq r1				@ disable interrupts
+	ldw	r1, [tsk+], #TI_FLAGS
+	cand.a	r1, #_TIF_WORK_MASK
+	bne	fast_work_pending
+
+	@ fast_restore_user_regs
+	restore_user_regs fast = 1, offset = S_OFF
+
+/*
+ * Ok, we need to do extra processing, enter the slow path.
+ */
+fast_work_pending:
+	stw.w	r0, [sp+], #S_R0+S_OFF		@ returned r0
+work_pending:
+	cand.a	r1, #_TIF_NEED_RESCHED
+	bne	work_resched
+	cand.a	r1, #_TIF_SIGPENDING|_TIF_NOTIFY_RESUME
+	beq	no_work_pending
+	mov	r0, sp				@ 'regs'
+	mov	r2, why				@ 'syscall'
+	cand.a	r1, #_TIF_SIGPENDING		@ delivering a signal?
+	cmovne	why, #0				@ prevent further restarts
+	b.l	do_notify_resume
+	b	ret_slow_syscall		@ Check work again
+
+work_resched:
+	b.l	schedule
+/*
+ * "slow" syscall return path.  "why" tells us if this was a real syscall.
+ */
+ENTRY(ret_to_user)
+ret_slow_syscall:
+	disable_irq r1				@ disable interrupts
+	get_thread_info tsk			@ epip4d, one path error?!
+	ldw	r1, [tsk+], #TI_FLAGS
+	cand.a	r1, #_TIF_WORK_MASK
+	bne	work_pending
+no_work_pending:
+	@ slow_restore_user_regs
+	restore_user_regs fast = 0, offset = 0
+ENDPROC(ret_to_user)
+
+/*
+ * This is how we return from a fork.
+ */
+ENTRY(ret_from_fork)
+	b.l	schedule_tail
+	get_thread_info tsk
+	ldw	r1, [tsk+], #TI_FLAGS		@ check for syscall tracing
+	mov	why, #1
+	cand.a	r1, #_TIF_SYSCALL_TRACE		@ are we tracing syscalls?
+	beq	ret_slow_syscall
+	mov	r1, sp
+	mov	r0, #1				@ trace exit [IP = 1]
+	b.l	syscall_trace
+	b	ret_slow_syscall
+ENDPROC(ret_from_fork)
+
+/*=============================================================================
+ * SWI handler
+ *-----------------------------------------------------------------------------
+ */
+	.align	5
+ENTRY(vector_swi)
+	sub	sp, sp, #S_FRAME_SIZE
+	stm	(r0 - r15), [sp]+		@ Calling r0 - r15
+	add	r8, sp, #S_R16
+	stm	(r16 - r28), [r8]+		@ Calling r16 - r28
+	add	r8, sp, #S_PC
+	stur	(sp, lr), [r8-]			@ Calling sp, lr
+	mov	r8, bsr				@ called from non-REAL mode
+	stw	lr, [sp+], #S_PC		@ Save calling PC
+	stw	r8, [sp+], #S_PSR		@ Save ASR
+	stw	r0, [sp+], #S_OLD_R0		@ Save OLD_R0
+	zero_fp
+
+	/*
+	 * Get the system call number.
+	 */
+	sub	ip, lr, #4
+	ldw.u	scno, [ip]			@ get SWI instruction
+
+#ifdef CONFIG_ALIGNMENT_TRAP
+	ldw	ip, __cr_alignment
+	ldw	ip, [ip]
+	movc	p0.c1, ip, #0                   @ update control register
+#endif
+	enable_irq ip
+
+	get_thread_info tsk
+	ldw	tbl, =sys_call_table		@ load syscall table pointer
+
+	andn	scno, scno, #0xff000000		@ mask off SWI op-code
+	andn	scno, scno, #0x00ff0000		@ mask off SWI op-code
+
+	stm.w	(r4, r5), [sp-]			@ push fifth and sixth args
+	ldw	ip, [tsk+], #TI_FLAGS		@ check for syscall tracing
+	cand.a	ip, #_TIF_SYSCALL_TRACE		@ are we tracing syscalls?
+	bne	__sys_trace
+
+	csub.a	scno, #__NR_syscalls		@ check upper syscall limit
+	adr	lr, ret_fast_syscall		@ return address
+	bea	1f
+	ldw	pc, [tbl+], scno << #2		@ call sys_* routine
+1:
+	add	r1, sp, #S_OFF
+2:	mov	why, #0				@ no longer a real syscall
+	b	sys_ni_syscall			@ not private func
+
+	/*
+	 * This is the really slow path.  We're going to be doing
+	 * context switches, and waiting for our parent to respond.
+	 */
+__sys_trace:
+	mov	r2, scno
+	add	r1, sp, #S_OFF
+	mov	r0, #0				@ trace entry [IP = 0]
+	b.l	syscall_trace
+
+	adr	lr, __sys_trace_return		@ return address
+	mov	scno, r0			@ syscall number (possibly new)
+	add	r1, sp, #S_R0 + S_OFF		@ pointer to regs
+	csub.a	scno, #__NR_syscalls		@ check upper syscall limit
+	bea	2b
+	ldm	(r0 - r3), [r1]+		@ have to reload r0 - r3
+	ldw	pc, [tbl+], scno << #2		@ call sys_* routine
+
+__sys_trace_return:
+	stw.w	r0, [sp+], #S_R0 + S_OFF	@ save returned r0
+	mov	r2, scno
+	mov	r1, sp
+	mov	r0, #1				@ trace exit [IP = 1]
+	b.l	syscall_trace
+	b	ret_slow_syscall
+
+	.align	5
+#ifdef CONFIG_ALIGNMENT_TRAP
+	.type	__cr_alignment, #object
+__cr_alignment:
+	.word	cr_alignment
+#endif
+	.ltorg
+
+ENTRY(sys_execve)
+		add	r3, sp, #S_OFF
+		b	__sys_execve
+ENDPROC(sys_execve)
+
+ENTRY(sys_clone)
+		add	ip, sp, #S_OFF
+		stw	ip, [sp+], #4
+		b	__sys_clone
+ENDPROC(sys_clone)
+
+ENTRY(sys_rt_sigreturn)
+		add	r0, sp, #S_OFF
+		mov	why, #0		@ prevent syscall restart handling
+		b	__sys_rt_sigreturn
+ENDPROC(sys_rt_sigreturn)
+
+ENTRY(sys_sigaltstack)
+		ldw	r2, [sp+], #S_OFF + S_SP
+		b	do_sigaltstack
+ENDPROC(sys_sigaltstack)
+
+	__INIT
+
+/*
+ * Vector stubs.
+ *
+ * This code is copied to 0xffff0200 so we can use branches in the
+ * vectors, rather than ldr's.  Note that this code must not
+ * exceed 0x300 bytes.
+ *
+ * Common stub entry macro:
+ *   Enter in INTR mode, bsr = PRIV/USER ASR, lr = PRIV/USER PC
+ *
+ * SP points to a minimal amount of processor-private memory, the address
+ * of which is copied into r0 for the mode specific abort handler.
+ */
+	.macro	vector_stub, name, mode
+	.align	5
+
+vector_\name:
+	@
+	@ Save r0, lr_<exception> (parent PC) and bsr_<exception>
+	@ (parent ASR)
+	@
+	stw	r0, [sp]
+	stw	lr, [sp+], #4		@ save r0, lr
+	mov	lr, bsr
+	stw	lr, [sp+], #8		@ save bsr
+
+	@
+	@ Prepare for PRIV mode.  INTRs remain disabled.
+	@
+	mov	r0, asr
+	xor	r0, r0, #(\mode ^ PRIV_MODE)
+	mov.a	bsr, r0
+
+	@
+	@ the branch table must immediately follow this code
+	@
+	and	lr, lr, #0x03
+	add	lr, lr, #1
+	mov	r0, sp
+	ldw	lr, [pc+], lr << #2
+	mov.a	pc, lr			@ branch to handler in PRIV mode
+ENDPROC(vector_\name)
+	.align	2
+	@ handler addresses follow this label
+	.endm
+
+	.globl	__stubs_start
+__stubs_start:
+/*
+ * Interrupt dispatcher
+ */
+	vector_stub	intr, INTR_MODE
+
+	.long	__intr_user			@  0  (USER)
+	.long	__invalid			@  1
+	.long	__invalid			@  2
+	.long	__intr_priv			@  3  (PRIV)
+
+/*
+ * Data abort dispatcher
+ * Enter in ABT mode, bsr = USER ASR, lr = USER PC
+ */
+	vector_stub	dabt, ABRT_MODE
+
+	.long	__dabt_user			@  0  (USER)
+	.long	__invalid			@  1
+	.long	__invalid			@  2  (INTR)
+	.long	__dabt_priv			@  3  (PRIV)
+
+/*
+ * Prefetch abort dispatcher
+ * Enter in ABT mode, bsr = USER ASR, lr = USER PC
+ */
+	vector_stub	pabt, ABRT_MODE
+
+	.long	__pabt_user			@  0 (USER)
+	.long	__invalid			@  1
+	.long	__invalid			@  2 (INTR)
+	.long	__pabt_priv			@  3 (PRIV)
+
+/*
+ * Undef instr entry dispatcher
+ * Enter in EXTN mode, bsr = PRIV/USER ASR, lr = PRIV/USER PC
+ */
+	vector_stub	extn, EXTN_MODE
+
+	.long	__extn_user			@  0 (USER)
+	.long	__invalid			@  1
+	.long	__invalid			@  2 (INTR)
+	.long	__extn_priv			@  3 (PRIV)
+
+/*
+ * We group all the following data together to optimise
+ * for CPUs with separate I & D caches.
+ */
+	.align	5
+
+.LCvswi:
+	.word	vector_swi
+
+	.globl	__stubs_end
+__stubs_end:
+
+	.equ	stubs_offset, __vectors_start + 0x200 - __stubs_start
+
+	.globl	__vectors_start
+__vectors_start:
+	jepriv	SYS_ERROR0
+	b	vector_extn + stubs_offset
+	ldw	pc, .LCvswi + stubs_offset
+	b	vector_pabt + stubs_offset
+	b	vector_dabt + stubs_offset
+	jepriv	SYS_ERROR0
+	b	vector_intr + stubs_offset
+	jepriv	SYS_ERROR0
+
+	.globl	__vectors_end
+__vectors_end:
+
+	.data
+
+	.globl	cr_alignment
+	.globl	cr_no_alignment
+cr_alignment:
+	.space	4
+cr_no_alignment:
+	.space	4
diff --git a/arch/unicore32/kernel/fpu-ucf64.c b/arch/unicore32/kernel/fpu-ucf64.c
new file mode 100644
index 000000000000..282a60ac82ba
--- /dev/null
+++ b/arch/unicore32/kernel/fpu-ucf64.c
@@ -0,0 +1,126 @@
+/*
+ * linux/arch/unicore32/kernel/fpu-ucf64.c
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/signal.h>
+#include <linux/sched.h>
+#include <linux/init.h>
+
+#include <asm/fpu-ucf64.h>
+
+/*
+ * A special flag to tell the normalisation code not to normalise.
+ */
+#define F64_NAN_FLAG	0x100
+
+/*
+ * A bit pattern used to indicate the initial (unset) value of the
+ * exception mask, in case nothing handles an instruction.  This
+ * doesn't include the NAN flag, which get masked out before
+ * we check for an error.
+ */
+#define F64_EXCEPTION_ERROR	((u32)-1 & ~F64_NAN_FLAG)
+
+/*
+ * Since we aren't building with -mfpu=f64, we need to code
+ * these instructions using their MRC/MCR equivalents.
+ */
+#define f64reg(_f64_) #_f64_
+
+#define cff(_f64_) ({			\
+	u32 __v;			\
+	asm("cff %0, " f64reg(_f64_) "@ fmrx	%0, " #_f64_	\
+	    : "=r" (__v) : : "cc");	\
+	__v;				\
+	})
+
+#define ctf(_f64_, _var_)		\
+	asm("ctf %0, " f64reg(_f64_) "@ fmxr	" #_f64_ ", %0"	\
+	   : : "r" (_var_) : "cc")
+
+/*
+ * Raise a SIGFPE for the current process.
+ * sicode describes the signal being raised.
+ */
+void ucf64_raise_sigfpe(unsigned int sicode, struct pt_regs *regs)
+{
+	siginfo_t info;
+
+	memset(&info, 0, sizeof(info));
+
+	info.si_signo = SIGFPE;
+	info.si_code = sicode;
+	info.si_addr = (void __user *)(instruction_pointer(regs) - 4);
+
+	/*
+	 * This is the same as NWFPE, because it's not clear what
+	 * this is used for
+	 */
+	current->thread.error_code = 0;
+	current->thread.trap_no = 6;
+
+	send_sig_info(SIGFPE, &info, current);
+}
+
+/*
+ * Handle exceptions of UniCore-F64.
+ */
+void ucf64_exchandler(u32 inst, u32 fpexc, struct pt_regs *regs)
+{
+	u32 tmp = fpexc;
+	u32 exc = F64_EXCEPTION_ERROR & fpexc;
+
+	pr_debug("UniCore-F64: instruction %08x fpscr %08x\n",
+			inst, fpexc);
+
+	if (exc & FPSCR_CMPINSTR_BIT) {
+		if (exc & FPSCR_CON)
+			tmp |= FPSCR_CON;
+		else
+			tmp &= ~(FPSCR_CON);
+		exc &= ~(FPSCR_CMPINSTR_BIT | FPSCR_CON);
+	} else {
+		pr_debug(KERN_ERR "UniCore-F64 Error: unhandled exceptions\n");
+		pr_debug(KERN_ERR "UniCore-F64 FPSCR 0x%08x INST 0x%08x\n",
+				cff(FPSCR), inst);
+
+		ucf64_raise_sigfpe(0, regs);
+		return;
+	}
+
+	/*
+	 * Update the FPSCR with the additional exception flags.
+	 * Comparison instructions always return at least one of
+	 * these flags set.
+	 */
+	tmp &= ~(FPSCR_TRAP | FPSCR_IOS | FPSCR_OFS | FPSCR_UFS |
+			FPSCR_IXS | FPSCR_HIS | FPSCR_IOC | FPSCR_OFC |
+			FPSCR_UFC | FPSCR_IXC | FPSCR_HIC);
+
+	tmp |= exc;
+	ctf(FPSCR, tmp);
+}
+
+/*
+ * F64 support code initialisation.
+ */
+static int __init ucf64_init(void)
+{
+	ctf(FPSCR, 0x0);     /* FPSCR_UFE | FPSCR_NDE perhaps better */
+
+	printk(KERN_INFO "Enable UniCore-F64 support.\n");
+
+	return 0;
+}
+
+late_initcall(ucf64_init);
diff --git a/arch/unicore32/kernel/gpio.c b/arch/unicore32/kernel/gpio.c
new file mode 100644
index 000000000000..cb12ec39552c
--- /dev/null
+++ b/arch/unicore32/kernel/gpio.c
@@ -0,0 +1,122 @@
+/*
+ * linux/arch/unicore32/kernel/gpio.c
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ *	Maintained by GUAN Xue-tao <gxt@mprc.pku.edu.cn>
+ *	Copyright (C) 2001-2010 Guan Xuetao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+/* in FPGA, no GPIO support */
+
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/gpio.h>
+#include <mach/hardware.h>
+
+#ifdef CONFIG_LEDS
+#include <linux/leds.h>
+#include <linux/platform_device.h>
+
+static const struct gpio_led puv3_gpio_leds[] = {
+	{ .name = "cpuhealth", .gpio = GPO_CPU_HEALTH, .active_low = 0,
+		.default_trigger = "heartbeat",	},
+	{ .name = "hdd_led", .gpio = GPO_HDD_LED, .active_low = 1,
+		.default_trigger = "ide-disk", },
+};
+
+static const struct gpio_led_platform_data puv3_gpio_led_data = {
+	.num_leds =	ARRAY_SIZE(puv3_gpio_leds),
+	.leds =		(void *) puv3_gpio_leds,
+};
+
+static struct platform_device puv3_gpio_gpio_leds = {
+	.name =		"leds-gpio",
+	.id =		-1,
+	.dev = {
+		.platform_data = (void *) &puv3_gpio_led_data,
+	}
+};
+
+static int __init puv3_gpio_leds_init(void)
+{
+	platform_device_register(&puv3_gpio_gpio_leds);
+	return 0;
+}
+
+device_initcall(puv3_gpio_leds_init);
+#endif
+
+static int puv3_gpio_get(struct gpio_chip *chip, unsigned offset)
+{
+	return readl(GPIO_GPLR) & GPIO_GPIO(offset);
+}
+
+static void puv3_gpio_set(struct gpio_chip *chip, unsigned offset, int value)
+{
+	if (value)
+		writel(GPIO_GPIO(offset), GPIO_GPSR);
+	else
+		writel(GPIO_GPIO(offset), GPIO_GPCR);
+}
+
+static int puv3_direction_input(struct gpio_chip *chip, unsigned offset)
+{
+	unsigned long flags;
+
+	local_irq_save(flags);
+	writel(readl(GPIO_GPDR) & ~GPIO_GPIO(offset), GPIO_GPDR);
+	local_irq_restore(flags);
+	return 0;
+}
+
+static int puv3_direction_output(struct gpio_chip *chip, unsigned offset,
+		int value)
+{
+	unsigned long flags;
+
+	local_irq_save(flags);
+	puv3_gpio_set(chip, offset, value);
+	writel(readl(GPIO_GPDR) | GPIO_GPIO(offset), GPIO_GPDR);
+	local_irq_restore(flags);
+	return 0;
+}
+
+static struct gpio_chip puv3_gpio_chip = {
+	.label			= "gpio",
+	.direction_input	= puv3_direction_input,
+	.direction_output	= puv3_direction_output,
+	.set			= puv3_gpio_set,
+	.get			= puv3_gpio_get,
+	.base			= 0,
+	.ngpio			= GPIO_MAX + 1,
+};
+
+void __init puv3_init_gpio(void)
+{
+	writel(GPIO_DIR, GPIO_GPDR);
+#if	defined(CONFIG_PUV3_NB0916) || defined(CONFIG_PUV3_SMW0919)	\
+	|| defined(CONFIG_PUV3_DB0913)
+	gpio_set_value(GPO_WIFI_EN, 1);
+	gpio_set_value(GPO_HDD_LED, 1);
+	gpio_set_value(GPO_VGA_EN, 1);
+	gpio_set_value(GPO_LCD_EN, 1);
+	gpio_set_value(GPO_CAM_PWR_EN, 0);
+	gpio_set_value(GPO_LCD_VCC_EN, 1);
+	gpio_set_value(GPO_SOFT_OFF, 1);
+	gpio_set_value(GPO_BT_EN, 1);
+	gpio_set_value(GPO_FAN_ON, 0);
+	gpio_set_value(GPO_SPKR, 0);
+	gpio_set_value(GPO_CPU_HEALTH, 1);
+	gpio_set_value(GPO_LAN_SEL, 1);
+/*
+ * DO NOT modify the GPO_SET_V1 and GPO_SET_V2 in kernel
+ *	gpio_set_value(GPO_SET_V1, 1);
+ *	gpio_set_value(GPO_SET_V2, 1);
+ */
+#endif
+	gpiochip_add(&puv3_gpio_chip);
+}
diff --git a/arch/unicore32/kernel/head.S b/arch/unicore32/kernel/head.S
new file mode 100644
index 000000000000..92255f3ab6a7
--- /dev/null
+++ b/arch/unicore32/kernel/head.S
@@ -0,0 +1,252 @@
+/*
+ * linux/arch/unicore32/kernel/head.S
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+#include <linux/init.h>
+
+#include <asm/assembler.h>
+#include <asm/ptrace.h>
+#include <generated/asm-offsets.h>
+#include <asm/memory.h>
+#include <asm/thread_info.h>
+#include <asm/system.h>
+#include <asm/pgtable-hwdef.h>
+
+#if (PHYS_OFFSET & 0x003fffff)
+#error "PHYS_OFFSET must be at an even 4MiB boundary!"
+#endif
+
+#define KERNEL_RAM_VADDR	(PAGE_OFFSET + KERNEL_IMAGE_START)
+#define KERNEL_RAM_PADDR	(PHYS_OFFSET + KERNEL_IMAGE_START)
+
+#define KERNEL_PGD_PADDR	(KERNEL_RAM_PADDR - 0x1000)
+#define KERNEL_PGD_VADDR	(KERNEL_RAM_VADDR - 0x1000)
+
+#define KERNEL_START		KERNEL_RAM_VADDR
+#define KERNEL_END		_end
+
+/*
+ * swapper_pg_dir is the virtual address of the initial page table.
+ * We place the page tables 4K below KERNEL_RAM_VADDR.  Therefore, we must
+ * make sure that KERNEL_RAM_VADDR is correctly set.  Currently, we expect
+ * the least significant 16 bits to be 0x8000, but we could probably
+ * relax this restriction to KERNEL_RAM_VADDR >= PAGE_OFFSET + 0x1000.
+ */
+#if (KERNEL_RAM_VADDR & 0xffff) != 0x8000
+#error KERNEL_RAM_VADDR must start at 0xXXXX8000
+#endif
+
+	.globl	swapper_pg_dir
+	.equ	swapper_pg_dir, KERNEL_RAM_VADDR - 0x1000
+
+/*
+ * Kernel startup entry point.
+ * ---------------------------
+ *
+ * This is normally called from the decompressor code.  The requirements
+ * are: MMU = off, D-cache = off, I-cache = dont care
+ *
+ * This code is mostly position independent, so if you link the kernel at
+ * 0xc0008000, you call this at __pa(0xc0008000).
+ */
+	__HEAD
+ENTRY(stext)
+	@ set asr
+	mov	r0, #PRIV_MODE			@ ensure priv mode
+	or	r0, #PSR_R_BIT | PSR_I_BIT	@ disable irqs
+	mov.a	asr, r0
+
+	@ process identify
+	movc	r0, p0.c0, #0			@ cpuid
+	movl	r1, 0xff00ffff			@ mask
+	movl	r2, 0x4d000863			@ value
+	and	r0, r1, r0
+	cxor.a	r0, r2
+	bne	__error_p			@ invalid processor id
+
+	/*
+	 * Clear the 4K level 1 swapper page table
+	 */
+	movl	r0, #KERNEL_PGD_PADDR		@ page table address
+	mov	r1, #0
+	add	r2, r0, #0x1000
+101:	stw.w	r1, [r0]+, #4
+	stw.w	r1, [r0]+, #4
+	stw.w	r1, [r0]+, #4
+	stw.w	r1, [r0]+, #4
+	cxor.a	r0, r2
+	bne	101b
+
+	movl	r4, #KERNEL_PGD_PADDR		@ page table address
+	mov	r7, #PMD_TYPE_SECT | PMD_PRESENT	@ page size: section
+	or	r7, r7, #PMD_SECT_CACHEABLE		@ cacheable
+	or	r7, r7, #PMD_SECT_READ | PMD_SECT_WRITE | PMD_SECT_EXEC
+
+	/*
+	 * Create identity mapping for first 4MB of kernel to
+	 * cater for the MMU enable.  This identity mapping
+	 * will be removed by paging_init().  We use our current program
+	 * counter to determine corresponding section base address.
+	 */
+	mov	r6, pc
+	mov	r6, r6 >> #22			@ start of kernel section
+	or	r1, r7, r6 << #22		@ flags + kernel base
+	stw	r1, [r4+], r6 << #2		@ identity mapping
+
+	/*
+	 * Now setup the pagetables for our kernel direct
+	 * mapped region.
+	 */
+	add	r0, r4,  #(KERNEL_START & 0xff000000) >> 20
+	stw.w	r1, [r0+], #(KERNEL_START & 0x00c00000) >> 20
+	movl	r6, #(KERNEL_END - 1)
+	add	r0, r0, #4
+	add	r6, r4, r6 >> #20
+102:	csub.a	r0, r6
+	add	r1, r1, #1 << 22
+	bua	103f
+	stw.w	r1, [r0]+, #4
+	b	102b
+103:
+	/*
+	 * Then map first 4MB of ram in case it contains our boot params.
+	 */
+	add	r0, r4, #PAGE_OFFSET >> 20
+	or	r6, r7, #(PHYS_OFFSET & 0xffc00000)
+	stw	r6, [r0]
+
+	ldw	r15, __switch_data		@ address to jump to after
+
+	/*
+	 * Initialise TLB, Caches, and MMU state ready to switch the MMU
+	 * on.
+	 */
+	mov	r0, #0
+	movc	p0.c5, r0, #28			@ cache invalidate all
+	nop8
+	movc	p0.c6, r0, #6			@ TLB invalidate all
+	nop8
+
+	/*
+	 * ..V. .... ..TB IDAM
+	 * ..1. .... ..01 1111
+	 */
+	movl	r0, #0x201f			@ control register setting
+
+	/*
+	 * Setup common bits before finally enabling the MMU.  Essentially
+	 * this is just loading the page table pointer and domain access
+	 * registers.
+	 */
+	#ifndef CONFIG_ALIGNMENT_TRAP
+		andn	r0, r0, #CR_A
+	#endif
+	#ifdef CONFIG_CPU_DCACHE_DISABLE
+		andn	r0, r0, #CR_D
+	#endif
+	#ifdef CONFIG_CPU_DCACHE_WRITETHROUGH
+		andn	r0, r0, #CR_B
+	#endif
+	#ifdef CONFIG_CPU_ICACHE_DISABLE
+		andn	r0, r0, #CR_I
+	#endif
+
+	movc	p0.c2, r4, #0			@ set pgd
+	b	__turn_mmu_on
+ENDPROC(stext)
+
+/*
+ * Enable the MMU.  This completely changes the stucture of the visible
+ * memory space.  You will not be able to trace execution through this.
+ *
+ *  r0  = cp#0 control register
+ *  r15 = *virtual* address to jump to upon completion
+ */
+	.align	5
+__turn_mmu_on:
+	mov	r0, r0
+	movc	p0.c1, r0, #0			@ write control reg
+	nop					@ fetch inst by phys addr
+	mov	pc, r15
+	nop8					@ fetch inst by phys addr
+ENDPROC(__turn_mmu_on)
+
+/*
+ * Setup the initial page tables.  We only setup the barest
+ * amount which are required to get the kernel running, which
+ * generally means mapping in the kernel code.
+ *
+ * r9  = cpuid
+ * r10 = procinfo
+ *
+ * Returns:
+ *  r0, r3, r6, r7 corrupted
+ *  r4 = physical page table address
+ */
+	.ltorg
+
+	.align	2
+	.type	__switch_data, %object
+__switch_data:
+	.long	__mmap_switched
+	.long	__bss_start			@ r6
+	.long	_end				@ r7
+	.long	cr_alignment			@ r8
+	.long	init_thread_union + THREAD_START_SP @ sp
+
+/*
+ * The following fragment of code is executed with the MMU on in MMU mode,
+ * and uses absolute addresses; this is not position independent.
+ *
+ *  r0  = cp#0 control register
+ */
+__mmap_switched:
+	adr	r3, __switch_data + 4
+
+	ldm.w	(r6, r7, r8), [r3]+
+	ldw	sp, [r3]
+
+	mov	fp, #0				@ Clear BSS (and zero fp)
+203:	csub.a	r6, r7
+	bea	204f
+	stw.w	fp, [r6]+,#4
+	b	203b
+204:
+	andn	r1, r0, #CR_A			@ Clear 'A' bit
+	stm	(r0, r1), [r8]+			@ Save control register values
+	b	start_kernel
+ENDPROC(__mmap_switched)
+
+/*
+ * Exception handling.  Something went wrong and we can't proceed.  We
+ * ought to tell the user, but since we don't have any guarantee that
+ * we're even running on the right architecture, we do virtually nothing.
+ *
+ * If CONFIG_DEBUG_LL is set we try to print out something about the error
+ * and hope for the best (useful if bootloader fails to pass a proper
+ * machine ID for example).
+ */
+__error_p:
+#ifdef CONFIG_DEBUG_LL
+	adr	r0, str_p1
+	b.l	printascii
+	mov	r0, r9
+	b.l	printhex8
+	adr	r0, str_p2
+	b.l	printascii
+901:	nop8
+	b	901b
+str_p1:	.asciz	"\nError: unrecognized processor variant (0x"
+str_p2:	.asciz	").\n"
+	.align
+#endif
+ENDPROC(__error_p)
+
diff --git a/arch/unicore32/kernel/hibernate.c b/arch/unicore32/kernel/hibernate.c
new file mode 100644
index 000000000000..7d0f0b7983a0
--- /dev/null
+++ b/arch/unicore32/kernel/hibernate.c
@@ -0,0 +1,160 @@
+/*
+ *  linux/arch/unicore32/kernel/hibernate.c
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ *	Maintained by GUAN Xue-tao <gxt@mprc.pku.edu.cn>
+ *	Copyright (C) 2001-2010 Guan Xuetao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/gfp.h>
+#include <linux/suspend.h>
+#include <linux/bootmem.h>
+
+#include <asm/system.h>
+#include <asm/page.h>
+#include <asm/pgtable.h>
+#include <asm/pgalloc.h>
+#include <asm/suspend.h>
+
+#include "mach/pm.h"
+
+/* Pointer to the temporary resume page tables */
+pgd_t *resume_pg_dir;
+
+struct swsusp_arch_regs swsusp_arch_regs_cpu0;
+
+/*
+ * Create a middle page table on a resume-safe page and put a pointer to it in
+ * the given global directory entry.  This only returns the gd entry
+ * in non-PAE compilation mode, since the middle layer is folded.
+ */
+static pmd_t *resume_one_md_table_init(pgd_t *pgd)
+{
+	pud_t *pud;
+	pmd_t *pmd_table;
+
+	pud = pud_offset(pgd, 0);
+	pmd_table = pmd_offset(pud, 0);
+
+	return pmd_table;
+}
+
+/*
+ * Create a page table on a resume-safe page and place a pointer to it in
+ * a middle page directory entry.
+ */
+static pte_t *resume_one_page_table_init(pmd_t *pmd)
+{
+	if (pmd_none(*pmd)) {
+		pte_t *page_table = (pte_t *)get_safe_page(GFP_ATOMIC);
+		if (!page_table)
+			return NULL;
+
+		set_pmd(pmd, __pmd(__pa(page_table) | _PAGE_KERNEL_TABLE));
+
+		BUG_ON(page_table != pte_offset_kernel(pmd, 0));
+
+		return page_table;
+	}
+
+	return pte_offset_kernel(pmd, 0);
+}
+
+/*
+ * This maps the physical memory to kernel virtual address space, a total
+ * of max_low_pfn pages, by creating page tables starting from address
+ * PAGE_OFFSET.  The page tables are allocated out of resume-safe pages.
+ */
+static int resume_physical_mapping_init(pgd_t *pgd_base)
+{
+	unsigned long pfn;
+	pgd_t *pgd;
+	pmd_t *pmd;
+	pte_t *pte;
+	int pgd_idx, pmd_idx;
+
+	pgd_idx = pgd_index(PAGE_OFFSET);
+	pgd = pgd_base + pgd_idx;
+	pfn = 0;
+
+	for (; pgd_idx < PTRS_PER_PGD; pgd++, pgd_idx++) {
+		pmd = resume_one_md_table_init(pgd);
+		if (!pmd)
+			return -ENOMEM;
+
+		if (pfn >= max_low_pfn)
+			continue;
+
+		for (pmd_idx = 0; pmd_idx < PTRS_PER_PMD; pmd++, pmd_idx++) {
+			pte_t *max_pte;
+
+			if (pfn >= max_low_pfn)
+				break;
+
+			/* Map with normal page tables.
+			 * NOTE: We can mark everything as executable here
+			 */
+			pte = resume_one_page_table_init(pmd);
+			if (!pte)
+				return -ENOMEM;
+
+			max_pte = pte + PTRS_PER_PTE;
+			for (; pte < max_pte; pte++, pfn++) {
+				if (pfn >= max_low_pfn)
+					break;
+
+				set_pte(pte, pfn_pte(pfn, PAGE_KERNEL_EXEC));
+			}
+		}
+	}
+
+	return 0;
+}
+
+static inline void resume_init_first_level_page_table(pgd_t *pg_dir)
+{
+}
+
+int swsusp_arch_resume(void)
+{
+	int error;
+
+	resume_pg_dir = (pgd_t *)get_safe_page(GFP_ATOMIC);
+	if (!resume_pg_dir)
+		return -ENOMEM;
+
+	resume_init_first_level_page_table(resume_pg_dir);
+	error = resume_physical_mapping_init(resume_pg_dir);
+	if (error)
+		return error;
+
+	/* We have got enough memory and from now on we cannot recover */
+	restore_image(resume_pg_dir, restore_pblist);
+	return 0;
+}
+
+/*
+ *	pfn_is_nosave - check if given pfn is in the 'nosave' section
+ */
+
+int pfn_is_nosave(unsigned long pfn)
+{
+	unsigned long begin_pfn = __pa(&__nosave_begin) >> PAGE_SHIFT;
+	unsigned long end_pfn = PAGE_ALIGN(__pa(&__nosave_end)) >> PAGE_SHIFT;
+
+	return (pfn >= begin_pfn) && (pfn < end_pfn);
+}
+
+void save_processor_state(void)
+{
+}
+
+void restore_processor_state(void)
+{
+	local_flush_tlb_all();
+}
diff --git a/arch/unicore32/kernel/hibernate_asm.S b/arch/unicore32/kernel/hibernate_asm.S
new file mode 100644
index 000000000000..cc3c65253c8c
--- /dev/null
+++ b/arch/unicore32/kernel/hibernate_asm.S
@@ -0,0 +1,117 @@
+/*
+ * linux/arch/unicore32/kernel/hibernate_asm.S
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ *	Maintained by GUAN Xue-tao <gxt@mprc.pku.edu.cn>
+ *	Copyright (C) 2001-2010 Guan Xuetao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/sys.h>
+#include <linux/errno.h>
+#include <linux/linkage.h>
+#include <generated/asm-offsets.h>
+#include <asm/page.h>
+#include <asm/pgtable.h>
+#include <asm/assembler.h>
+
+@ restore_image(pgd_t *resume_pg_dir, struct pbe *restore_pblist)
+@ r0: resume_pg_dir
+@ r1: restore_pblist
+@ copy restore_pblist pages
+@ restore registers from swsusp_arch_regs_cpu0
+@
+ENTRY(restore_image)
+	sub	r0, r0, #PAGE_OFFSET
+	mov	r5, #0
+	movc	p0.c6, r5, #6	@invalidate ITLB & DTLB
+	movc	p0.c2, r0, #0
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+
+	.p2align 4,,7
+101:
+	csub.a	r1, #0
+	beq	109f
+
+	ldw	r6, [r1+], #PBE_ADDRESS
+	ldw	r7, [r1+], #PBE_ORIN_ADDRESS
+
+	movl	ip, #128
+102:	ldm.w	(r8 - r15), [r6]+
+	stm.w	(r8 - r15), [r7]+
+	sub.a	ip, ip, #1
+	bne	102b
+
+	ldw	r1, [r1+], #PBE_NEXT
+	b	101b
+
+	.p2align 4,,7
+109:
+	/* go back to the original page tables */
+	ldw	r0, =swapper_pg_dir
+	sub	r0, r0, #PAGE_OFFSET
+	mov	r5, #0
+	movc	p0.c6, r5, #6
+	movc	p0.c2, r0, #0
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+
+#ifdef	CONFIG_UNICORE_FPU_F64
+	ldw	ip, 1f
+	add	ip, ip, #SWSUSP_FPSTATE
+	lfm.w	(f0  - f7 ), [ip]+
+	lfm.w	(f8  - f15), [ip]+
+	lfm.w	(f16 - f23), [ip]+
+	lfm.w	(f24 - f31), [ip]+
+	ldw	r4, [ip]
+	ctf	r4, s31
+#endif
+	mov	r0, #0x0
+	ldw	ip, 1f
+	add	ip, ip, #SWSUSP_CPU
+	ldm.w	(r4 - r15), [ip]+
+	ldm	(r16 - r27, sp, pc), [ip]+	@ Load all regs saved previously
+
+	.align	2
+1:	.long	swsusp_arch_regs_cpu0
+
+
+@ swsusp_arch_suspend()
+@ - prepare pc for resume, return from function without swsusp_save on resume
+@ - save registers in swsusp_arch_regs_cpu0
+@ - call swsusp_save write suspend image
+
+ENTRY(swsusp_arch_suspend)
+	ldw	ip, 1f
+	add	ip, ip, #SWSUSP_CPU
+	stm.w	(r4 - r15), [ip]+
+	stm.w	(r16 - r27, sp, lr), [ip]+
+
+#ifdef	CONFIG_UNICORE_FPU_F64
+	ldw	ip, 1f
+	add	ip, ip, #SWSUSP_FPSTATE
+	sfm.w	(f0  - f7 ), [ip]+
+	sfm.w	(f8  - f15), [ip]+
+	sfm.w	(f16 - f23), [ip]+
+	sfm.w	(f24 - f31), [ip]+
+	cff	r4, s31
+	stw	r4, [ip]
+#endif
+	b	swsusp_save			@ no return
+
+1:	.long	swsusp_arch_regs_cpu0
diff --git a/arch/unicore32/kernel/init_task.c b/arch/unicore32/kernel/init_task.c
new file mode 100644
index 000000000000..a35a1e50e4f4
--- /dev/null
+++ b/arch/unicore32/kernel/init_task.c
@@ -0,0 +1,44 @@
+/*
+ * linux/arch/unicore32/kernel/init_task.c
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/mm.h>
+#include <linux/module.h>
+#include <linux/fs.h>
+#include <linux/sched.h>
+#include <linux/init.h>
+#include <linux/init_task.h>
+#include <linux/mqueue.h>
+#include <linux/uaccess.h>
+
+#include <asm/pgtable.h>
+
+static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
+static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
+/*
+ * Initial thread structure.
+ *
+ * We need to make sure that this is 8192-byte aligned due to the
+ * way process stacks are handled. This is done by making sure
+ * the linker maps this in the .text segment right after head.S,
+ * and making head.S ensure the proper alignment.
+ *
+ * The things we do for performance..
+ */
+union thread_union init_thread_union __init_task_data = {
+	INIT_THREAD_INFO(init_task) };
+
+/*
+ * Initial task structure.
+ *
+ * All other task structs will be allocated on slabs in fork.c
+ */
+struct task_struct init_task = INIT_TASK(init_task);
+EXPORT_SYMBOL(init_task);
diff --git a/arch/unicore32/kernel/irq.c b/arch/unicore32/kernel/irq.c
new file mode 100644
index 000000000000..b23624cf3062
--- /dev/null
+++ b/arch/unicore32/kernel/irq.c
@@ -0,0 +1,426 @@
+/*
+ * linux/arch/unicore32/kernel/irq.c
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/kernel_stat.h>
+#include <linux/module.h>
+#include <linux/signal.h>
+#include <linux/ioport.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/random.h>
+#include <linux/smp.h>
+#include <linux/init.h>
+#include <linux/seq_file.h>
+#include <linux/errno.h>
+#include <linux/list.h>
+#include <linux/kallsyms.h>
+#include <linux/proc_fs.h>
+#include <linux/sysdev.h>
+#include <linux/gpio.h>
+
+#include <asm/system.h>
+#include <mach/hardware.h>
+
+#include "setup.h"
+
+/*
+ * PKUnity GPIO edge detection for IRQs:
+ * IRQs are generated on Falling-Edge, Rising-Edge, or both.
+ * Use this instead of directly setting GRER/GFER.
+ */
+static int GPIO_IRQ_rising_edge;
+static int GPIO_IRQ_falling_edge;
+static int GPIO_IRQ_mask = 0;
+
+#define GPIO_MASK(irq)		(1 << (irq - IRQ_GPIO0))
+
+static int puv3_gpio_type(struct irq_data *d, unsigned int type)
+{
+	unsigned int mask;
+
+	if (d->irq < IRQ_GPIOHIGH)
+		mask = 1 << d->irq;
+	else
+		mask = GPIO_MASK(d->irq);
+
+	if (type == IRQ_TYPE_PROBE) {
+		if ((GPIO_IRQ_rising_edge | GPIO_IRQ_falling_edge) & mask)
+			return 0;
+		type = IRQ_TYPE_EDGE_RISING | IRQ_TYPE_EDGE_FALLING;
+	}
+
+	if (type & IRQ_TYPE_EDGE_RISING)
+		GPIO_IRQ_rising_edge |= mask;
+	else
+		GPIO_IRQ_rising_edge &= ~mask;
+	if (type & IRQ_TYPE_EDGE_FALLING)
+		GPIO_IRQ_falling_edge |= mask;
+	else
+		GPIO_IRQ_falling_edge &= ~mask;
+
+	writel(GPIO_IRQ_rising_edge & GPIO_IRQ_mask, GPIO_GRER);
+	writel(GPIO_IRQ_falling_edge & GPIO_IRQ_mask, GPIO_GFER);
+
+	return 0;
+}
+
+/*
+ * GPIO IRQs must be acknowledged.  This is for IRQs from 0 to 7.
+ */
+static void puv3_low_gpio_ack(struct irq_data *d)
+{
+	writel((1 << d->irq), GPIO_GEDR);
+}
+
+static void puv3_low_gpio_mask(struct irq_data *d)
+{
+	writel(readl(INTC_ICMR) & ~(1 << d->irq), INTC_ICMR);
+}
+
+static void puv3_low_gpio_unmask(struct irq_data *d)
+{
+	writel(readl(INTC_ICMR) | (1 << d->irq), INTC_ICMR);
+}
+
+static int puv3_low_gpio_wake(struct irq_data *d, unsigned int on)
+{
+	if (on)
+		writel(readl(PM_PWER) | (1 << d->irq), PM_PWER);
+	else
+		writel(readl(PM_PWER) & ~(1 << d->irq), PM_PWER);
+	return 0;
+}
+
+static struct irq_chip puv3_low_gpio_chip = {
+	.name		= "GPIO-low",
+	.irq_ack	= puv3_low_gpio_ack,
+	.irq_mask	= puv3_low_gpio_mask,
+	.irq_unmask	= puv3_low_gpio_unmask,
+	.irq_set_type	= puv3_gpio_type,
+	.irq_set_wake	= puv3_low_gpio_wake,
+};
+
+/*
+ * IRQ8 (GPIO0 through 27) handler.  We enter here with the
+ * irq_controller_lock held, and IRQs disabled.  Decode the IRQ
+ * and call the handler.
+ */
+static void
+puv3_gpio_handler(unsigned int irq, struct irq_desc *desc)
+{
+	unsigned int mask;
+
+	mask = readl(GPIO_GEDR);
+	do {
+		/*
+		 * clear down all currently active IRQ sources.
+		 * We will be processing them all.
+		 */
+		writel(mask, GPIO_GEDR);
+
+		irq = IRQ_GPIO0;
+		do {
+			if (mask & 1)
+				generic_handle_irq(irq);
+			mask >>= 1;
+			irq++;
+		} while (mask);
+		mask = readl(GPIO_GEDR);
+	} while (mask);
+}
+
+/*
+ * GPIO0-27 edge IRQs need to be handled specially.
+ * In addition, the IRQs are all collected up into one bit in the
+ * interrupt controller registers.
+ */
+static void puv3_high_gpio_ack(struct irq_data *d)
+{
+	unsigned int mask = GPIO_MASK(d->irq);
+
+	writel(mask, GPIO_GEDR);
+}
+
+static void puv3_high_gpio_mask(struct irq_data *d)
+{
+	unsigned int mask = GPIO_MASK(d->irq);
+
+	GPIO_IRQ_mask &= ~mask;
+
+	writel(readl(GPIO_GRER) & ~mask, GPIO_GRER);
+	writel(readl(GPIO_GFER) & ~mask, GPIO_GFER);
+}
+
+static void puv3_high_gpio_unmask(struct irq_data *d)
+{
+	unsigned int mask = GPIO_MASK(d->irq);
+
+	GPIO_IRQ_mask |= mask;
+
+	writel(GPIO_IRQ_rising_edge & GPIO_IRQ_mask, GPIO_GRER);
+	writel(GPIO_IRQ_falling_edge & GPIO_IRQ_mask, GPIO_GFER);
+}
+
+static int puv3_high_gpio_wake(struct irq_data *d, unsigned int on)
+{
+	if (on)
+		writel(readl(PM_PWER) | PM_PWER_GPIOHIGH, PM_PWER);
+	else
+		writel(readl(PM_PWER) & ~PM_PWER_GPIOHIGH, PM_PWER);
+	return 0;
+}
+
+static struct irq_chip puv3_high_gpio_chip = {
+	.name		= "GPIO-high",
+	.irq_ack	= puv3_high_gpio_ack,
+	.irq_mask	= puv3_high_gpio_mask,
+	.irq_unmask	= puv3_high_gpio_unmask,
+	.irq_set_type	= puv3_gpio_type,
+	.irq_set_wake	= puv3_high_gpio_wake,
+};
+
+/*
+ * We don't need to ACK IRQs on the PKUnity unless they're GPIOs
+ * this is for internal IRQs i.e. from 8 to 31.
+ */
+static void puv3_mask_irq(struct irq_data *d)
+{
+	writel(readl(INTC_ICMR) & ~(1 << d->irq), INTC_ICMR);
+}
+
+static void puv3_unmask_irq(struct irq_data *d)
+{
+	writel(readl(INTC_ICMR) | (1 << d->irq), INTC_ICMR);
+}
+
+/*
+ * Apart form GPIOs, only the RTC alarm can be a wakeup event.
+ */
+static int puv3_set_wake(struct irq_data *d, unsigned int on)
+{
+	if (d->irq == IRQ_RTCAlarm) {
+		if (on)
+			writel(readl(PM_PWER) | PM_PWER_RTC, PM_PWER);
+		else
+			writel(readl(PM_PWER) & ~PM_PWER_RTC, PM_PWER);
+		return 0;
+	}
+	return -EINVAL;
+}
+
+static struct irq_chip puv3_normal_chip = {
+	.name		= "PKUnity-v3",
+	.irq_ack	= puv3_mask_irq,
+	.irq_mask	= puv3_mask_irq,
+	.irq_unmask	= puv3_unmask_irq,
+	.irq_set_wake	= puv3_set_wake,
+};
+
+static struct resource irq_resource = {
+	.name	= "irqs",
+	.start	= io_v2p(PKUNITY_INTC_BASE),
+	.end	= io_v2p(PKUNITY_INTC_BASE) + 0xFFFFF,
+};
+
+static struct puv3_irq_state {
+	unsigned int	saved;
+	unsigned int	icmr;
+	unsigned int	iclr;
+	unsigned int	iccr;
+} puv3_irq_state;
+
+static int puv3_irq_suspend(struct sys_device *dev, pm_message_t state)
+{
+	struct puv3_irq_state *st = &puv3_irq_state;
+
+	st->saved = 1;
+	st->icmr = readl(INTC_ICMR);
+	st->iclr = readl(INTC_ICLR);
+	st->iccr = readl(INTC_ICCR);
+
+	/*
+	 * Disable all GPIO-based interrupts.
+	 */
+	writel(readl(INTC_ICMR) & ~(0x1ff), INTC_ICMR);
+
+	/*
+	 * Set the appropriate edges for wakeup.
+	 */
+	writel(readl(PM_PWER) & GPIO_IRQ_rising_edge, GPIO_GRER);
+	writel(readl(PM_PWER) & GPIO_IRQ_falling_edge, GPIO_GFER);
+
+	/*
+	 * Clear any pending GPIO interrupts.
+	 */
+	writel(readl(GPIO_GEDR), GPIO_GEDR);
+
+	return 0;
+}
+
+static int puv3_irq_resume(struct sys_device *dev)
+{
+	struct puv3_irq_state *st = &puv3_irq_state;
+
+	if (st->saved) {
+		writel(st->iccr, INTC_ICCR);
+		writel(st->iclr, INTC_ICLR);
+
+		writel(GPIO_IRQ_rising_edge & GPIO_IRQ_mask, GPIO_GRER);
+		writel(GPIO_IRQ_falling_edge & GPIO_IRQ_mask, GPIO_GFER);
+
+		writel(st->icmr, INTC_ICMR);
+	}
+	return 0;
+}
+
+static struct sysdev_class puv3_irq_sysclass = {
+	.name		= "pkunity-irq",
+	.suspend	= puv3_irq_suspend,
+	.resume		= puv3_irq_resume,
+};
+
+static struct sys_device puv3_irq_device = {
+	.id		= 0,
+	.cls		= &puv3_irq_sysclass,
+};
+
+static int __init puv3_irq_init_devicefs(void)
+{
+	sysdev_class_register(&puv3_irq_sysclass);
+	return sysdev_register(&puv3_irq_device);
+}
+
+device_initcall(puv3_irq_init_devicefs);
+
+void __init init_IRQ(void)
+{
+	unsigned int irq;
+
+	request_resource(&iomem_resource, &irq_resource);
+
+	/* disable all IRQs */
+	writel(0, INTC_ICMR);
+
+	/* all IRQs are IRQ, not REAL */
+	writel(0, INTC_ICLR);
+
+	/* clear all GPIO edge detects */
+	writel(FMASK(8, 0) & ~FIELD(1, 1, GPI_SOFF_REQ), GPIO_GPIR);
+	writel(0, GPIO_GFER);
+	writel(0, GPIO_GRER);
+	writel(0x0FFFFFFF, GPIO_GEDR);
+
+	writel(1, INTC_ICCR);
+
+	for (irq = 0; irq < IRQ_GPIOHIGH; irq++) {
+		set_irq_chip(irq, &puv3_low_gpio_chip);
+		set_irq_handler(irq, handle_edge_irq);
+		irq_modify_status(irq,
+			IRQ_NOREQUEST | IRQ_NOPROBE | IRQ_NOAUTOEN,
+			0);
+	}
+
+	for (irq = IRQ_GPIOHIGH + 1; irq < IRQ_GPIO0; irq++) {
+		set_irq_chip(irq, &puv3_normal_chip);
+		set_irq_handler(irq, handle_level_irq);
+		irq_modify_status(irq,
+			IRQ_NOREQUEST | IRQ_NOAUTOEN,
+			IRQ_NOPROBE);
+	}
+
+	for (irq = IRQ_GPIO0; irq <= IRQ_GPIO27; irq++) {
+		set_irq_chip(irq, &puv3_high_gpio_chip);
+		set_irq_handler(irq, handle_edge_irq);
+		irq_modify_status(irq,
+			IRQ_NOREQUEST | IRQ_NOPROBE | IRQ_NOAUTOEN,
+			0);
+	}
+
+	/*
+	 * Install handler for GPIO 0-27 edge detect interrupts
+	 */
+	set_irq_chip(IRQ_GPIOHIGH, &puv3_normal_chip);
+	set_irq_chained_handler(IRQ_GPIOHIGH, puv3_gpio_handler);
+
+#ifdef CONFIG_PUV3_GPIO
+	puv3_init_gpio();
+#endif
+}
+
+int show_interrupts(struct seq_file *p, void *v)
+{
+	int i = *(loff_t *) v, cpu;
+	struct irq_desc *desc;
+	struct irqaction *action;
+	unsigned long flags;
+
+	if (i == 0) {
+		char cpuname[12];
+
+		seq_printf(p, "    ");
+		for_each_present_cpu(cpu) {
+			sprintf(cpuname, "CPU%d", cpu);
+			seq_printf(p, " %10s", cpuname);
+		}
+		seq_putc(p, '\n');
+	}
+
+	if (i < nr_irqs) {
+		desc = irq_to_desc(i);
+		raw_spin_lock_irqsave(&desc->lock, flags);
+		action = desc->action;
+		if (!action)
+			goto unlock;
+
+		seq_printf(p, "%3d: ", i);
+		for_each_present_cpu(cpu)
+			seq_printf(p, "%10u ", kstat_irqs_cpu(i, cpu));
+		seq_printf(p, " %10s", desc->irq_data.chip->name ? : "-");
+		seq_printf(p, "  %s", action->name);
+		for (action = action->next; action; action = action->next)
+			seq_printf(p, ", %s", action->name);
+
+		seq_putc(p, '\n');
+unlock:
+		raw_spin_unlock_irqrestore(&desc->lock, flags);
+	} else if (i == nr_irqs) {
+		seq_printf(p, "Error in interrupt!\n");
+	}
+	return 0;
+}
+
+/*
+ * do_IRQ handles all hardware IRQ's.  Decoded IRQs should not
+ * come via this function.  Instead, they should provide their
+ * own 'handler'
+ */
+asmlinkage void asm_do_IRQ(unsigned int irq, struct pt_regs *regs)
+{
+	struct pt_regs *old_regs = set_irq_regs(regs);
+
+	irq_enter();
+
+	/*
+	 * Some hardware gives randomly wrong interrupts.  Rather
+	 * than crashing, do something sensible.
+	 */
+	if (unlikely(irq >= nr_irqs)) {
+		if (printk_ratelimit())
+			printk(KERN_WARNING "Bad IRQ%u\n", irq);
+		ack_bad_irq(irq);
+	} else {
+		generic_handle_irq(irq);
+	}
+
+	irq_exit();
+	set_irq_regs(old_regs);
+}
+
diff --git a/arch/unicore32/kernel/ksyms.c b/arch/unicore32/kernel/ksyms.c
new file mode 100644
index 000000000000..a8970809428a
--- /dev/null
+++ b/arch/unicore32/kernel/ksyms.c
@@ -0,0 +1,99 @@
+/*
+ * linux/arch/unicore32/kernel/ksyms.c
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/string.h>
+#include <linux/cryptohash.h>
+#include <linux/delay.h>
+#include <linux/in6.h>
+#include <linux/syscalls.h>
+#include <linux/uaccess.h>
+#include <linux/io.h>
+
+#include <asm/checksum.h>
+#include <asm/system.h>
+
+#include "ksyms.h"
+
+EXPORT_SYMBOL(__uc32_find_next_zero_bit);
+EXPORT_SYMBOL(__uc32_find_next_bit);
+
+EXPORT_SYMBOL(__backtrace);
+
+	/* platform dependent support */
+EXPORT_SYMBOL(__udelay);
+EXPORT_SYMBOL(__const_udelay);
+
+	/* networking */
+EXPORT_SYMBOL(csum_partial);
+EXPORT_SYMBOL(csum_partial_copy_from_user);
+EXPORT_SYMBOL(csum_partial_copy_nocheck);
+EXPORT_SYMBOL(__csum_ipv6_magic);
+
+	/* io */
+#ifndef __raw_readsb
+EXPORT_SYMBOL(__raw_readsb);
+#endif
+#ifndef __raw_readsw
+EXPORT_SYMBOL(__raw_readsw);
+#endif
+#ifndef __raw_readsl
+EXPORT_SYMBOL(__raw_readsl);
+#endif
+#ifndef __raw_writesb
+EXPORT_SYMBOL(__raw_writesb);
+#endif
+#ifndef __raw_writesw
+EXPORT_SYMBOL(__raw_writesw);
+#endif
+#ifndef __raw_writesl
+EXPORT_SYMBOL(__raw_writesl);
+#endif
+
+	/* string / mem functions */
+EXPORT_SYMBOL(strchr);
+EXPORT_SYMBOL(strrchr);
+EXPORT_SYMBOL(memset);
+EXPORT_SYMBOL(memcpy);
+EXPORT_SYMBOL(memmove);
+EXPORT_SYMBOL(memchr);
+
+	/* user mem (segment) */
+EXPORT_SYMBOL(__strnlen_user);
+EXPORT_SYMBOL(__strncpy_from_user);
+
+EXPORT_SYMBOL(copy_page);
+
+EXPORT_SYMBOL(__copy_from_user);
+EXPORT_SYMBOL(__copy_to_user);
+EXPORT_SYMBOL(__clear_user);
+
+EXPORT_SYMBOL(__get_user_1);
+EXPORT_SYMBOL(__get_user_2);
+EXPORT_SYMBOL(__get_user_4);
+
+EXPORT_SYMBOL(__put_user_1);
+EXPORT_SYMBOL(__put_user_2);
+EXPORT_SYMBOL(__put_user_4);
+EXPORT_SYMBOL(__put_user_8);
+
+EXPORT_SYMBOL(__ashldi3);
+EXPORT_SYMBOL(__ashrdi3);
+EXPORT_SYMBOL(__divsi3);
+EXPORT_SYMBOL(__lshrdi3);
+EXPORT_SYMBOL(__modsi3);
+EXPORT_SYMBOL(__muldi3);
+EXPORT_SYMBOL(__ucmpdi2);
+EXPORT_SYMBOL(__udivsi3);
+EXPORT_SYMBOL(__umodsi3);
+EXPORT_SYMBOL(__bswapsi2);
+
diff --git a/arch/unicore32/kernel/ksyms.h b/arch/unicore32/kernel/ksyms.h
new file mode 100644
index 000000000000..185cdc712d03
--- /dev/null
+++ b/arch/unicore32/kernel/ksyms.h
@@ -0,0 +1,15 @@
+/*
+ * libgcc functions - functions that are used internally by the
+ * compiler...  (prototypes are not correct though, but that
+ * doesn't really matter since they're not versioned).
+ */
+extern void __ashldi3(void);
+extern void __ashrdi3(void);
+extern void __divsi3(void);
+extern void __lshrdi3(void);
+extern void __modsi3(void);
+extern void __muldi3(void);
+extern void __ucmpdi2(void);
+extern void __udivsi3(void);
+extern void __umodsi3(void);
+extern void __bswapsi2(void);
diff --git a/arch/unicore32/kernel/module.c b/arch/unicore32/kernel/module.c
new file mode 100644
index 000000000000..3e5a38d71a1e
--- /dev/null
+++ b/arch/unicore32/kernel/module.c
@@ -0,0 +1,152 @@
+/*
+ * linux/arch/unicore32/kernel/module.c
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/module.h>
+#include <linux/moduleloader.h>
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/elf.h>
+#include <linux/vmalloc.h>
+#include <linux/fs.h>
+#include <linux/string.h>
+#include <linux/gfp.h>
+
+#include <asm/pgtable.h>
+#include <asm/sections.h>
+
+void *module_alloc(unsigned long size)
+{
+	struct vm_struct *area;
+
+	size = PAGE_ALIGN(size);
+	if (!size)
+		return NULL;
+
+	area = __get_vm_area(size, VM_ALLOC, MODULES_VADDR, MODULES_END);
+	if (!area)
+		return NULL;
+
+	return __vmalloc_area(area, GFP_KERNEL, PAGE_KERNEL_EXEC);
+}
+
+void module_free(struct module *module, void *region)
+{
+	vfree(region);
+}
+
+int module_frob_arch_sections(Elf_Ehdr *hdr,
+			      Elf_Shdr *sechdrs,
+			      char *secstrings,
+			      struct module *mod)
+{
+	return 0;
+}
+
+int
+apply_relocate(Elf32_Shdr *sechdrs, const char *strtab, unsigned int symindex,
+	       unsigned int relindex, struct module *module)
+{
+	Elf32_Shdr *symsec = sechdrs + symindex;
+	Elf32_Shdr *relsec = sechdrs + relindex;
+	Elf32_Shdr *dstsec = sechdrs + relsec->sh_info;
+	Elf32_Rel *rel = (void *)relsec->sh_addr;
+	unsigned int i;
+
+	for (i = 0; i < relsec->sh_size / sizeof(Elf32_Rel); i++, rel++) {
+		unsigned long loc;
+		Elf32_Sym *sym;
+		s32 offset;
+
+		offset = ELF32_R_SYM(rel->r_info);
+		if (offset < 0 || offset >
+				(symsec->sh_size / sizeof(Elf32_Sym))) {
+			printk(KERN_ERR "%s: bad relocation, "
+					"section %d reloc %d\n",
+					module->name, relindex, i);
+			return -ENOEXEC;
+		}
+
+		sym = ((Elf32_Sym *)symsec->sh_addr) + offset;
+
+		if (rel->r_offset < 0 || rel->r_offset >
+				dstsec->sh_size - sizeof(u32)) {
+			printk(KERN_ERR "%s: out of bounds relocation, "
+				"section %d reloc %d offset %d size %d\n",
+				module->name, relindex, i, rel->r_offset,
+				dstsec->sh_size);
+			return -ENOEXEC;
+		}
+
+		loc = dstsec->sh_addr + rel->r_offset;
+
+		switch (ELF32_R_TYPE(rel->r_info)) {
+		case R_UNICORE_NONE:
+			/* ignore */
+			break;
+
+		case R_UNICORE_ABS32:
+			*(u32 *)loc += sym->st_value;
+			break;
+
+		case R_UNICORE_PC24:
+		case R_UNICORE_CALL:
+		case R_UNICORE_JUMP24:
+			offset = (*(u32 *)loc & 0x00ffffff) << 2;
+			if (offset & 0x02000000)
+				offset -= 0x04000000;
+
+			offset += sym->st_value - loc;
+			if (offset & 3 ||
+			    offset <= (s32)0xfe000000 ||
+			    offset >= (s32)0x02000000) {
+				printk(KERN_ERR
+				       "%s: relocation out of range, section "
+				       "%d reloc %d sym '%s'\n", module->name,
+				       relindex, i, strtab + sym->st_name);
+				return -ENOEXEC;
+			}
+
+			offset >>= 2;
+
+			*(u32 *)loc &= 0xff000000;
+			*(u32 *)loc |= offset & 0x00ffffff;
+			break;
+
+		default:
+			printk(KERN_ERR "%s: unknown relocation: %u\n",
+			       module->name, ELF32_R_TYPE(rel->r_info));
+			return -ENOEXEC;
+		}
+	}
+	return 0;
+}
+
+int
+apply_relocate_add(Elf32_Shdr *sechdrs, const char *strtab,
+		   unsigned int symindex, unsigned int relsec,
+		   struct module *module)
+{
+	printk(KERN_ERR "module %s: ADD RELOCATION unsupported\n",
+	       module->name);
+	return -ENOEXEC;
+}
+
+int
+module_finalize(const Elf32_Ehdr *hdr, const Elf_Shdr *sechdrs,
+		struct module *module)
+{
+	return 0;
+}
+
+void
+module_arch_cleanup(struct module *mod)
+{
+}
diff --git a/arch/unicore32/kernel/pci.c b/arch/unicore32/kernel/pci.c
new file mode 100644
index 000000000000..100eab842e66
--- /dev/null
+++ b/arch/unicore32/kernel/pci.c
@@ -0,0 +1,404 @@
+/*
+ * linux/arch/unicore32/kernel/pci.c
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ *  PCI bios-type initialisation for PCI machines
+ *
+ */
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/interrupt.h>
+#include <linux/pci.h>
+#include <linux/slab.h>
+#include <linux/init.h>
+#include <linux/io.h>
+
+static int debug_pci;
+static int use_firmware;
+
+#define CONFIG_CMD(bus, devfn, where)	\
+	(0x80000000 | (bus->number << 16) | (devfn << 8) | (where & ~3))
+
+static int
+puv3_read_config(struct pci_bus *bus, unsigned int devfn, int where,
+			int size, u32 *value)
+{
+	writel(CONFIG_CMD(bus, devfn, where), PCICFG_ADDR);
+	switch (size) {
+	case 1:
+		*value = (readl(PCICFG_DATA) >> ((where & 3) * 8)) & 0xFF;
+		break;
+	case 2:
+		*value = (readl(PCICFG_DATA) >> ((where & 2) * 8)) & 0xFFFF;
+		break;
+	case 4:
+		*value = readl(PCICFG_DATA);
+		break;
+	}
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static int
+puv3_write_config(struct pci_bus *bus, unsigned int devfn, int where,
+			int size, u32 value)
+{
+	writel(CONFIG_CMD(bus, devfn, where), PCICFG_ADDR);
+	switch (size) {
+	case 1:
+		writel((readl(PCICFG_DATA) & ~FMASK(8, (where&3)*8))
+			| FIELD(value, 8, (where&3)*8), PCICFG_DATA);
+		break;
+	case 2:
+		writel((readl(PCICFG_DATA) & ~FMASK(16, (where&2)*8))
+			| FIELD(value, 16, (where&2)*8), PCICFG_DATA);
+		break;
+	case 4:
+		writel(value, PCICFG_DATA);
+		break;
+	}
+	return PCIBIOS_SUCCESSFUL;
+}
+
+struct pci_ops pci_puv3_ops = {
+	.read  = puv3_read_config,
+	.write = puv3_write_config,
+};
+
+void pci_puv3_preinit(void)
+{
+	printk(KERN_DEBUG "PCI: PKUnity PCI Controller Initializing ...\n");
+	/* config PCI bridge base */
+	writel(io_v2p(PKUNITY_PCIBRI_BASE), PCICFG_BRIBASE);
+
+	writel(0, PCIBRI_AHBCTL0);
+	writel(io_v2p(PKUNITY_PCIBRI_BASE) | PCIBRI_BARx_MEM, PCIBRI_AHBBAR0);
+	writel(0xFFFF0000, PCIBRI_AHBAMR0);
+	writel(0, PCIBRI_AHBTAR0);
+
+	writel(PCIBRI_CTLx_AT, PCIBRI_AHBCTL1);
+	writel(io_v2p(PKUNITY_PCILIO_BASE) | PCIBRI_BARx_IO, PCIBRI_AHBBAR1);
+	writel(0xFFFF0000, PCIBRI_AHBAMR1);
+	writel(0x00000000, PCIBRI_AHBTAR1);
+
+	writel(PCIBRI_CTLx_PREF, PCIBRI_AHBCTL2);
+	writel(io_v2p(PKUNITY_PCIMEM_BASE) | PCIBRI_BARx_MEM, PCIBRI_AHBBAR2);
+	writel(0xF8000000, PCIBRI_AHBAMR2);
+	writel(0, PCIBRI_AHBTAR2);
+
+	writel(io_v2p(PKUNITY_PCIAHB_BASE) | PCIBRI_BARx_MEM, PCIBRI_BAR1);
+
+	writel(PCIBRI_CTLx_AT | PCIBRI_CTLx_PREF, PCIBRI_PCICTL0);
+	writel(io_v2p(PKUNITY_PCIAHB_BASE) | PCIBRI_BARx_MEM, PCIBRI_PCIBAR0);
+	writel(0xF8000000, PCIBRI_PCIAMR0);
+	writel(PKUNITY_SDRAM_BASE, PCIBRI_PCITAR0);
+
+	writel(readl(PCIBRI_CMD) | PCIBRI_CMD_IO | PCIBRI_CMD_MEM, PCIBRI_CMD);
+}
+
+static int __init pci_puv3_map_irq(struct pci_dev *dev, u8 slot, u8 pin)
+{
+	if (dev->bus->number == 0) {
+#ifdef CONFIG_ARCH_FPGA /* 4 pci slots */
+		if      (dev->devfn == 0x00)
+			return IRQ_PCIINTA;
+		else if (dev->devfn == 0x08)
+			return IRQ_PCIINTB;
+		else if (dev->devfn == 0x10)
+			return IRQ_PCIINTC;
+		else if (dev->devfn == 0x18)
+			return IRQ_PCIINTD;
+#endif
+#ifdef CONFIG_PUV3_DB0913 /* 3 pci slots */
+		if      (dev->devfn == 0x30)
+			return IRQ_PCIINTB;
+		else if (dev->devfn == 0x60)
+			return IRQ_PCIINTC;
+		else if (dev->devfn == 0x58)
+			return IRQ_PCIINTD;
+#endif
+#if	defined(CONFIG_PUV3_NB0916) || defined(CONFIG_PUV3_SMW0919)
+		/* only support 2 pci devices */
+		if      (dev->devfn == 0x00)
+			return IRQ_PCIINTC; /* sata */
+#endif
+	}
+	return -1;
+}
+
+/*
+ * Only first 128MB of memory can be accessed via PCI.
+ * We use GFP_DMA to allocate safe buffers to do map/unmap.
+ * This is really ugly and we need a better way of specifying
+ * DMA-capable regions of memory.
+ */
+void __init puv3_pci_adjust_zones(unsigned long *zone_size,
+	unsigned long *zhole_size)
+{
+	unsigned int sz = SZ_128M >> PAGE_SHIFT;
+
+	/*
+	 * Only adjust if > 128M on current system
+	 */
+	if (zone_size[0] <= sz)
+		return;
+
+	zone_size[1] = zone_size[0] - sz;
+	zone_size[0] = sz;
+	zhole_size[1] = zhole_size[0];
+	zhole_size[0] = 0;
+}
+
+void __devinit pcibios_update_irq(struct pci_dev *dev, int irq)
+{
+	if (debug_pci)
+		printk(KERN_DEBUG "PCI: Assigning IRQ %02d to %s\n",
+				irq, pci_name(dev));
+	pci_write_config_byte(dev, PCI_INTERRUPT_LINE, irq);
+}
+
+/*
+ * If the bus contains any of these devices, then we must not turn on
+ * parity checking of any kind.
+ */
+static inline int pdev_bad_for_parity(struct pci_dev *dev)
+{
+	return 0;
+}
+
+/*
+ * pcibios_fixup_bus - Called after each bus is probed,
+ * but before its children are examined.
+ */
+void __devinit pcibios_fixup_bus(struct pci_bus *bus)
+{
+	struct pci_dev *dev;
+	u16 features = PCI_COMMAND_SERR
+		| PCI_COMMAND_PARITY
+		| PCI_COMMAND_FAST_BACK;
+
+	bus->resource[0] = &ioport_resource;
+	bus->resource[1] = &iomem_resource;
+
+	/*
+	 * Walk the devices on this bus, working out what we can
+	 * and can't support.
+	 */
+	list_for_each_entry(dev, &bus->devices, bus_list) {
+		u16 status;
+
+		pci_read_config_word(dev, PCI_STATUS, &status);
+
+		/*
+		 * If any device on this bus does not support fast back
+		 * to back transfers, then the bus as a whole is not able
+		 * to support them.  Having fast back to back transfers
+		 * on saves us one PCI cycle per transaction.
+		 */
+		if (!(status & PCI_STATUS_FAST_BACK))
+			features &= ~PCI_COMMAND_FAST_BACK;
+
+		if (pdev_bad_for_parity(dev))
+			features &= ~(PCI_COMMAND_SERR
+					| PCI_COMMAND_PARITY);
+
+		switch (dev->class >> 8) {
+		case PCI_CLASS_BRIDGE_PCI:
+			pci_read_config_word(dev, PCI_BRIDGE_CONTROL, &status);
+			status |= PCI_BRIDGE_CTL_PARITY
+				| PCI_BRIDGE_CTL_MASTER_ABORT;
+			status &= ~(PCI_BRIDGE_CTL_BUS_RESET
+				| PCI_BRIDGE_CTL_FAST_BACK);
+			pci_write_config_word(dev, PCI_BRIDGE_CONTROL, status);
+			break;
+
+		case PCI_CLASS_BRIDGE_CARDBUS:
+			pci_read_config_word(dev, PCI_CB_BRIDGE_CONTROL,
+					&status);
+			status |= PCI_CB_BRIDGE_CTL_PARITY
+				| PCI_CB_BRIDGE_CTL_MASTER_ABORT;
+			pci_write_config_word(dev, PCI_CB_BRIDGE_CONTROL,
+					status);
+			break;
+		}
+	}
+
+	/*
+	 * Now walk the devices again, this time setting them up.
+	 */
+	list_for_each_entry(dev, &bus->devices, bus_list) {
+		u16 cmd;
+
+		pci_read_config_word(dev, PCI_COMMAND, &cmd);
+		cmd |= features;
+		pci_write_config_word(dev, PCI_COMMAND, cmd);
+
+		pci_write_config_byte(dev, PCI_CACHE_LINE_SIZE,
+				      L1_CACHE_BYTES >> 2);
+	}
+
+	/*
+	 * Propagate the flags to the PCI bridge.
+	 */
+	if (bus->self && bus->self->hdr_type == PCI_HEADER_TYPE_BRIDGE) {
+		if (features & PCI_COMMAND_FAST_BACK)
+			bus->bridge_ctl |= PCI_BRIDGE_CTL_FAST_BACK;
+		if (features & PCI_COMMAND_PARITY)
+			bus->bridge_ctl |= PCI_BRIDGE_CTL_PARITY;
+	}
+
+	/*
+	 * Report what we did for this bus
+	 */
+	printk(KERN_INFO "PCI: bus%d: Fast back to back transfers %sabled\n",
+		bus->number, (features & PCI_COMMAND_FAST_BACK) ? "en" : "dis");
+}
+#ifdef CONFIG_HOTPLUG
+EXPORT_SYMBOL(pcibios_fixup_bus);
+#endif
+
+static int __init pci_common_init(void)
+{
+	struct pci_bus *puv3_bus;
+
+	pci_puv3_preinit();
+
+	puv3_bus = pci_scan_bus(0, &pci_puv3_ops, NULL);
+
+	if (!puv3_bus)
+		panic("PCI: unable to scan bus!");
+
+	pci_fixup_irqs(pci_common_swizzle, pci_puv3_map_irq);
+
+	if (!use_firmware) {
+		/*
+		 * Size the bridge windows.
+		 */
+		pci_bus_size_bridges(puv3_bus);
+
+		/*
+		 * Assign resources.
+		 */
+		pci_bus_assign_resources(puv3_bus);
+	}
+
+	/*
+	 * Tell drivers about devices found.
+	 */
+	pci_bus_add_devices(puv3_bus);
+
+	return 0;
+}
+subsys_initcall(pci_common_init);
+
+char * __devinit pcibios_setup(char *str)
+{
+	if (!strcmp(str, "debug")) {
+		debug_pci = 1;
+		return NULL;
+	} else if (!strcmp(str, "firmware")) {
+		use_firmware = 1;
+		return NULL;
+	}
+	return str;
+}
+
+/*
+ * From arch/i386/kernel/pci-i386.c:
+ *
+ * We need to avoid collisions with `mirrored' VGA ports
+ * and other strange ISA hardware, so we always want the
+ * addresses to be allocated in the 0x000-0x0ff region
+ * modulo 0x400.
+ *
+ * Why? Because some silly external IO cards only decode
+ * the low 10 bits of the IO address. The 0x00-0xff region
+ * is reserved for motherboard devices that decode all 16
+ * bits, so it's ok to allocate at, say, 0x2800-0x28ff,
+ * but we want to try to avoid allocating at 0x2900-0x2bff
+ * which might be mirrored at 0x0100-0x03ff..
+ */
+resource_size_t pcibios_align_resource(void *data, const struct resource *res,
+				resource_size_t size, resource_size_t align)
+{
+	resource_size_t start = res->start;
+
+	if (res->flags & IORESOURCE_IO && start & 0x300)
+		start = (start + 0x3ff) & ~0x3ff;
+
+	start = (start + align - 1) & ~(align - 1);
+
+	return start;
+}
+
+/**
+ * pcibios_enable_device - Enable I/O and memory.
+ * @dev: PCI device to be enabled
+ */
+int pcibios_enable_device(struct pci_dev *dev, int mask)
+{
+	u16 cmd, old_cmd;
+	int idx;
+	struct resource *r;
+
+	pci_read_config_word(dev, PCI_COMMAND, &cmd);
+	old_cmd = cmd;
+	for (idx = 0; idx < 6; idx++) {
+		/* Only set up the requested stuff */
+		if (!(mask & (1 << idx)))
+			continue;
+
+		r = dev->resource + idx;
+		if (!r->start && r->end) {
+			printk(KERN_ERR "PCI: Device %s not available because"
+			       " of resource collisions\n", pci_name(dev));
+			return -EINVAL;
+		}
+		if (r->flags & IORESOURCE_IO)
+			cmd |= PCI_COMMAND_IO;
+		if (r->flags & IORESOURCE_MEM)
+			cmd |= PCI_COMMAND_MEMORY;
+	}
+
+	/*
+	 * Bridges (eg, cardbus bridges) need to be fully enabled
+	 */
+	if ((dev->class >> 16) == PCI_BASE_CLASS_BRIDGE)
+		cmd |= PCI_COMMAND_IO | PCI_COMMAND_MEMORY;
+
+	if (cmd != old_cmd) {
+		printk("PCI: enabling device %s (%04x -> %04x)\n",
+		       pci_name(dev), old_cmd, cmd);
+		pci_write_config_word(dev, PCI_COMMAND, cmd);
+	}
+	return 0;
+}
+
+int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma,
+			enum pci_mmap_state mmap_state, int write_combine)
+{
+	unsigned long phys;
+
+	if (mmap_state == pci_mmap_io)
+		return -EINVAL;
+
+	phys = vma->vm_pgoff;
+
+	/*
+	 * Mark this as IO
+	 */
+	vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
+
+	if (remap_pfn_range(vma, vma->vm_start, phys,
+			     vma->vm_end - vma->vm_start,
+			     vma->vm_page_prot))
+		return -EAGAIN;
+
+	return 0;
+}
diff --git a/arch/unicore32/kernel/pm.c b/arch/unicore32/kernel/pm.c
new file mode 100644
index 000000000000..784bc2db3b28
--- /dev/null
+++ b/arch/unicore32/kernel/pm.c
@@ -0,0 +1,123 @@
+/*
+ * linux/arch/unicore32/kernel/pm.c
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ *	Maintained by GUAN Xue-tao <gxt@mprc.pku.edu.cn>
+ *	Copyright (C) 2001-2010 Guan Xuetao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/suspend.h>
+#include <linux/errno.h>
+#include <linux/slab.h>
+#include <linux/io.h>
+
+#include <mach/hardware.h>
+#include <mach/pm.h>
+
+#include "setup.h"
+
+struct puv3_cpu_pm_fns *puv3_cpu_pm_fns;
+static unsigned long *sleep_save;
+
+int puv3_pm_enter(suspend_state_t state)
+{
+	unsigned long sleep_save_checksum = 0, checksum = 0;
+	int i;
+
+	/* skip registers saving for standby */
+	if (state != PM_SUSPEND_STANDBY) {
+		puv3_cpu_pm_fns->save(sleep_save);
+		/* before sleeping, calculate and save a checksum */
+		for (i = 0; i < puv3_cpu_pm_fns->save_count - 1; i++)
+			sleep_save_checksum += sleep_save[i];
+	}
+
+	/* *** go zzz *** */
+	puv3_cpu_pm_fns->enter(state);
+	cpu_init();
+#ifdef CONFIG_INPUT_KEYBOARD
+	puv3_ps2_init();
+#endif
+#ifdef CONFIG_PCI
+	pci_puv3_preinit();
+#endif
+	if (state != PM_SUSPEND_STANDBY) {
+		/* after sleeping, validate the checksum */
+		for (i = 0; i < puv3_cpu_pm_fns->save_count - 1; i++)
+			checksum += sleep_save[i];
+
+		/* if invalid, display message and wait for a hardware reset */
+		if (checksum != sleep_save_checksum) {
+			while (1)
+				puv3_cpu_pm_fns->enter(state);
+		}
+		puv3_cpu_pm_fns->restore(sleep_save);
+	}
+
+	pr_debug("*** made it back from resume\n");
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(puv3_pm_enter);
+
+unsigned long sleep_phys_sp(void *sp)
+{
+	return virt_to_phys(sp);
+}
+
+static int puv3_pm_valid(suspend_state_t state)
+{
+	if (puv3_cpu_pm_fns)
+		return puv3_cpu_pm_fns->valid(state);
+
+	return -EINVAL;
+}
+
+static int puv3_pm_prepare(void)
+{
+	int ret = 0;
+
+	if (puv3_cpu_pm_fns && puv3_cpu_pm_fns->prepare)
+		ret = puv3_cpu_pm_fns->prepare();
+
+	return ret;
+}
+
+static void puv3_pm_finish(void)
+{
+	if (puv3_cpu_pm_fns && puv3_cpu_pm_fns->finish)
+		puv3_cpu_pm_fns->finish();
+}
+
+static struct platform_suspend_ops puv3_pm_ops = {
+	.valid		= puv3_pm_valid,
+	.enter		= puv3_pm_enter,
+	.prepare	= puv3_pm_prepare,
+	.finish		= puv3_pm_finish,
+};
+
+static int __init puv3_pm_init(void)
+{
+	if (!puv3_cpu_pm_fns) {
+		printk(KERN_ERR "no valid puv3_cpu_pm_fns defined\n");
+		return -EINVAL;
+	}
+
+	sleep_save = kmalloc(puv3_cpu_pm_fns->save_count
+				* sizeof(unsigned long), GFP_KERNEL);
+	if (!sleep_save) {
+		printk(KERN_ERR "failed to alloc memory for pm save\n");
+		return -ENOMEM;
+	}
+
+	suspend_set_ops(&puv3_pm_ops);
+	return 0;
+}
+
+device_initcall(puv3_pm_init);
diff --git a/arch/unicore32/kernel/process.c b/arch/unicore32/kernel/process.c
new file mode 100644
index 000000000000..ba401df971ed
--- /dev/null
+++ b/arch/unicore32/kernel/process.c
@@ -0,0 +1,389 @@
+/*
+ * linux/arch/unicore32/kernel/process.c
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <stdarg.h>
+
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/stddef.h>
+#include <linux/unistd.h>
+#include <linux/delay.h>
+#include <linux/reboot.h>
+#include <linux/interrupt.h>
+#include <linux/kallsyms.h>
+#include <linux/init.h>
+#include <linux/cpu.h>
+#include <linux/elfcore.h>
+#include <linux/pm.h>
+#include <linux/tick.h>
+#include <linux/utsname.h>
+#include <linux/uaccess.h>
+#include <linux/random.h>
+#include <linux/gpio.h>
+#include <linux/stacktrace.h>
+
+#include <asm/cacheflush.h>
+#include <asm/processor.h>
+#include <asm/system.h>
+#include <asm/stacktrace.h>
+
+#include "setup.h"
+
+static const char * const processor_modes[] = {
+	"UK00", "UK01", "UK02", "UK03", "UK04", "UK05", "UK06", "UK07",
+	"UK08", "UK09", "UK0A", "UK0B", "UK0C", "UK0D", "UK0E", "UK0F",
+	"USER", "REAL", "INTR", "PRIV", "UK14", "UK15", "UK16", "ABRT",
+	"UK18", "UK19", "UK1A", "EXTN", "UK1C", "UK1D", "UK1E", "SUSR"
+};
+
+/*
+ * The idle thread, has rather strange semantics for calling pm_idle,
+ * but this is what x86 does and we need to do the same, so that
+ * things like cpuidle get called in the same way.
+ */
+void cpu_idle(void)
+{
+	/* endless idle loop with no priority at all */
+	while (1) {
+		tick_nohz_stop_sched_tick(1);
+		while (!need_resched()) {
+			local_irq_disable();
+			stop_critical_timings();
+			cpu_do_idle();
+			local_irq_enable();
+			start_critical_timings();
+		}
+		tick_nohz_restart_sched_tick();
+		preempt_enable_no_resched();
+		schedule();
+		preempt_disable();
+	}
+}
+
+static char reboot_mode = 'h';
+
+int __init reboot_setup(char *str)
+{
+	reboot_mode = str[0];
+	return 1;
+}
+
+__setup("reboot=", reboot_setup);
+
+void machine_halt(void)
+{
+	gpio_set_value(GPO_SOFT_OFF, 0);
+}
+
+/*
+ * Function pointers to optional machine specific functions
+ */
+void (*pm_power_off)(void) = NULL;
+
+void machine_power_off(void)
+{
+	if (pm_power_off)
+		pm_power_off();
+	machine_halt();
+}
+
+void machine_restart(char *cmd)
+{
+	/* Disable interrupts first */
+	local_irq_disable();
+
+	/*
+	 * Tell the mm system that we are going to reboot -
+	 * we may need it to insert some 1:1 mappings so that
+	 * soft boot works.
+	 */
+	setup_mm_for_reboot(reboot_mode);
+
+	/* Clean and invalidate caches */
+	flush_cache_all();
+
+	/* Turn off caching */
+	cpu_proc_fin();
+
+	/* Push out any further dirty data, and ensure cache is empty */
+	flush_cache_all();
+
+	/*
+	 * Now handle reboot code.
+	 */
+	if (reboot_mode == 's') {
+		/* Jump into ROM at address 0xffff0000 */
+		cpu_reset(VECTORS_BASE);
+	} else {
+		writel(0x00002001, PM_PLLSYSCFG); /* cpu clk = 250M */
+		writel(0x00100800, PM_PLLDDRCFG); /* ddr clk =  44M */
+		writel(0x00002001, PM_PLLVGACFG); /* vga clk = 250M */
+
+		/* Use on-chip reset capability */
+		/* following instructions must be in one icache line */
+		__asm__ __volatile__(
+			"	.align 5\n\t"
+			"	stw	%1, [%0]\n\t"
+			"201:	ldw	r0, [%0]\n\t"
+			"	cmpsub.a	r0, #0\n\t"
+			"	bne	201b\n\t"
+			"	stw	%3, [%2]\n\t"
+			"	nop; nop; nop\n\t"
+			/* prefetch 3 instructions at most */
+			:
+			: "r" (PM_PMCR),
+			  "r" (PM_PMCR_CFBSYS | PM_PMCR_CFBDDR
+				| PM_PMCR_CFBVGA),
+			  "r" (RESETC_SWRR),
+			  "r" (RESETC_SWRR_SRB)
+			: "r0", "memory");
+	}
+
+	/*
+	 * Whoops - the architecture was unable to reboot.
+	 * Tell the user!
+	 */
+	mdelay(1000);
+	printk(KERN_EMERG "Reboot failed -- System halted\n");
+	do { } while (1);
+}
+
+void __show_regs(struct pt_regs *regs)
+{
+	unsigned long flags;
+	char buf[64];
+
+	printk(KERN_DEFAULT "CPU: %d    %s  (%s %.*s)\n",
+		raw_smp_processor_id(), print_tainted(),
+		init_utsname()->release,
+		(int)strcspn(init_utsname()->version, " "),
+		init_utsname()->version);
+	print_symbol("PC is at %s\n", instruction_pointer(regs));
+	print_symbol("LR is at %s\n", regs->UCreg_lr);
+	printk(KERN_DEFAULT "pc : [<%08lx>]    lr : [<%08lx>]    psr: %08lx\n"
+	       "sp : %08lx  ip : %08lx  fp : %08lx\n",
+		regs->UCreg_pc, regs->UCreg_lr, regs->UCreg_asr,
+		regs->UCreg_sp, regs->UCreg_ip, regs->UCreg_fp);
+	printk(KERN_DEFAULT "r26: %08lx  r25: %08lx  r24: %08lx\n",
+		regs->UCreg_26, regs->UCreg_25,
+		regs->UCreg_24);
+	printk(KERN_DEFAULT "r23: %08lx  r22: %08lx  r21: %08lx  r20: %08lx\n",
+		regs->UCreg_23, regs->UCreg_22,
+		regs->UCreg_21, regs->UCreg_20);
+	printk(KERN_DEFAULT "r19: %08lx  r18: %08lx  r17: %08lx  r16: %08lx\n",
+		regs->UCreg_19, regs->UCreg_18,
+		regs->UCreg_17, regs->UCreg_16);
+	printk(KERN_DEFAULT "r15: %08lx  r14: %08lx  r13: %08lx  r12: %08lx\n",
+		regs->UCreg_15, regs->UCreg_14,
+		regs->UCreg_13, regs->UCreg_12);
+	printk(KERN_DEFAULT "r11: %08lx  r10: %08lx  r9 : %08lx  r8 : %08lx\n",
+		regs->UCreg_11, regs->UCreg_10,
+		regs->UCreg_09, regs->UCreg_08);
+	printk(KERN_DEFAULT "r7 : %08lx  r6 : %08lx  r5 : %08lx  r4 : %08lx\n",
+		regs->UCreg_07, regs->UCreg_06,
+		regs->UCreg_05, regs->UCreg_04);
+	printk(KERN_DEFAULT "r3 : %08lx  r2 : %08lx  r1 : %08lx  r0 : %08lx\n",
+		regs->UCreg_03, regs->UCreg_02,
+		regs->UCreg_01, regs->UCreg_00);
+
+	flags = regs->UCreg_asr;
+	buf[0] = flags & PSR_S_BIT ? 'S' : 's';
+	buf[1] = flags & PSR_Z_BIT ? 'Z' : 'z';
+	buf[2] = flags & PSR_C_BIT ? 'C' : 'c';
+	buf[3] = flags & PSR_V_BIT ? 'V' : 'v';
+	buf[4] = '\0';
+
+	printk(KERN_DEFAULT "Flags: %s  INTR o%s  REAL o%s  Mode %s  Segment %s\n",
+		buf, interrupts_enabled(regs) ? "n" : "ff",
+		fast_interrupts_enabled(regs) ? "n" : "ff",
+		processor_modes[processor_mode(regs)],
+		segment_eq(get_fs(), get_ds()) ? "kernel" : "user");
+	{
+		unsigned int ctrl;
+
+		buf[0] = '\0';
+		{
+			unsigned int transbase;
+			asm("movc %0, p0.c2, #0\n"
+			    : "=r" (transbase));
+			snprintf(buf, sizeof(buf), "  Table: %08x", transbase);
+		}
+		asm("movc %0, p0.c1, #0\n" : "=r" (ctrl));
+
+		printk(KERN_DEFAULT "Control: %08x%s\n", ctrl, buf);
+	}
+}
+
+void show_regs(struct pt_regs *regs)
+{
+	printk(KERN_DEFAULT "\n");
+	printk(KERN_DEFAULT "Pid: %d, comm: %20s\n",
+			task_pid_nr(current), current->comm);
+	__show_regs(regs);
+	__backtrace();
+}
+
+/*
+ * Free current thread data structures etc..
+ */
+void exit_thread(void)
+{
+}
+
+void flush_thread(void)
+{
+	struct thread_info *thread = current_thread_info();
+	struct task_struct *tsk = current;
+
+	memset(thread->used_cp, 0, sizeof(thread->used_cp));
+	memset(&tsk->thread.debug, 0, sizeof(struct debug_info));
+#ifdef CONFIG_UNICORE_FPU_F64
+	memset(&thread->fpstate, 0, sizeof(struct fp_state));
+#endif
+}
+
+void release_thread(struct task_struct *dead_task)
+{
+}
+
+asmlinkage void ret_from_fork(void) __asm__("ret_from_fork");
+
+int
+copy_thread(unsigned long clone_flags, unsigned long stack_start,
+	    unsigned long stk_sz, struct task_struct *p, struct pt_regs *regs)
+{
+	struct thread_info *thread = task_thread_info(p);
+	struct pt_regs *childregs = task_pt_regs(p);
+
+	*childregs = *regs;
+	childregs->UCreg_00 = 0;
+	childregs->UCreg_sp = stack_start;
+
+	memset(&thread->cpu_context, 0, sizeof(struct cpu_context_save));
+	thread->cpu_context.sp = (unsigned long)childregs;
+	thread->cpu_context.pc = (unsigned long)ret_from_fork;
+
+	if (clone_flags & CLONE_SETTLS)
+		childregs->UCreg_16 = regs->UCreg_03;
+
+	return 0;
+}
+
+/*
+ * Fill in the task's elfregs structure for a core dump.
+ */
+int dump_task_regs(struct task_struct *t, elf_gregset_t *elfregs)
+{
+	elf_core_copy_regs(elfregs, task_pt_regs(t));
+	return 1;
+}
+
+/*
+ * fill in the fpe structure for a core dump...
+ */
+int dump_fpu(struct pt_regs *regs, elf_fpregset_t *fp)
+{
+	struct thread_info *thread = current_thread_info();
+	int used_math = thread->used_cp[1] | thread->used_cp[2];
+
+#ifdef CONFIG_UNICORE_FPU_F64
+	if (used_math)
+		memcpy(fp, &thread->fpstate, sizeof(*fp));
+#endif
+	return used_math != 0;
+}
+EXPORT_SYMBOL(dump_fpu);
+
+/*
+ * Shuffle the argument into the correct register before calling the
+ * thread function.  r1 is the thread argument, r2 is the pointer to
+ * the thread function, and r3 points to the exit function.
+ */
+asm(".pushsection .text\n"
+"	.align\n"
+"	.type	kernel_thread_helper, #function\n"
+"kernel_thread_helper:\n"
+"	mov.a	asr, r7\n"
+"	mov	r0, r4\n"
+"	mov	lr, r6\n"
+"	mov	pc, r5\n"
+"	.size	kernel_thread_helper, . - kernel_thread_helper\n"
+"	.popsection");
+
+/*
+ * Create a kernel thread.
+ */
+pid_t kernel_thread(int (*fn)(void *), void *arg, unsigned long flags)
+{
+	struct pt_regs regs;
+
+	memset(&regs, 0, sizeof(regs));
+
+	regs.UCreg_04 = (unsigned long)arg;
+	regs.UCreg_05 = (unsigned long)fn;
+	regs.UCreg_06 = (unsigned long)do_exit;
+	regs.UCreg_07 = PRIV_MODE;
+	regs.UCreg_pc = (unsigned long)kernel_thread_helper;
+	regs.UCreg_asr = regs.UCreg_07 | PSR_I_BIT;
+
+	return do_fork(flags|CLONE_VM|CLONE_UNTRACED, 0, &regs, 0, NULL, NULL);
+}
+EXPORT_SYMBOL(kernel_thread);
+
+unsigned long get_wchan(struct task_struct *p)
+{
+	struct stackframe frame;
+	int count = 0;
+	if (!p || p == current || p->state == TASK_RUNNING)
+		return 0;
+
+	frame.fp = thread_saved_fp(p);
+	frame.sp = thread_saved_sp(p);
+	frame.lr = 0;			/* recovered from the stack */
+	frame.pc = thread_saved_pc(p);
+	do {
+		int ret = unwind_frame(&frame);
+		if (ret < 0)
+			return 0;
+		if (!in_sched_functions(frame.pc))
+			return frame.pc;
+	} while ((count++) < 16);
+	return 0;
+}
+
+unsigned long arch_randomize_brk(struct mm_struct *mm)
+{
+	unsigned long range_end = mm->brk + 0x02000000;
+	return randomize_range(mm->brk, range_end, 0) ? : mm->brk;
+}
+
+/*
+ * The vectors page is always readable from user space for the
+ * atomic helpers and the signal restart code.  Let's declare a mapping
+ * for it so it is visible through ptrace and /proc/<pid>/mem.
+ */
+
+int vectors_user_mapping(void)
+{
+	struct mm_struct *mm = current->mm;
+	return install_special_mapping(mm, 0xffff0000, PAGE_SIZE,
+				       VM_READ | VM_EXEC |
+				       VM_MAYREAD | VM_MAYEXEC |
+				       VM_ALWAYSDUMP | VM_RESERVED,
+				       NULL);
+}
+
+const char *arch_vma_name(struct vm_area_struct *vma)
+{
+	return (vma->vm_start == 0xffff0000) ? "[vectors]" : NULL;
+}
diff --git a/arch/unicore32/kernel/ptrace.c b/arch/unicore32/kernel/ptrace.c
new file mode 100644
index 000000000000..9f07c08da050
--- /dev/null
+++ b/arch/unicore32/kernel/ptrace.c
@@ -0,0 +1,149 @@
+/*
+ * linux/arch/unicore32/kernel/ptrace.c
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * By Ross Biro 1/23/92
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/kernel.h>
+#include <linux/ptrace.h>
+#include <linux/signal.h>
+#include <linux/uaccess.h>
+
+/*
+ * this routine will get a word off of the processes privileged stack.
+ * the offset is how far from the base addr as stored in the THREAD.
+ * this routine assumes that all the privileged stacks are in our
+ * data space.
+ */
+static inline long get_user_reg(struct task_struct *task, int offset)
+{
+	return task_pt_regs(task)->uregs[offset];
+}
+
+/*
+ * this routine will put a word on the processes privileged stack.
+ * the offset is how far from the base addr as stored in the THREAD.
+ * this routine assumes that all the privileged stacks are in our
+ * data space.
+ */
+static inline int
+put_user_reg(struct task_struct *task, int offset, long data)
+{
+	struct pt_regs newregs, *regs = task_pt_regs(task);
+	int ret = -EINVAL;
+
+	newregs = *regs;
+	newregs.uregs[offset] = data;
+
+	if (valid_user_regs(&newregs)) {
+		regs->uregs[offset] = data;
+		ret = 0;
+	}
+
+	return ret;
+}
+
+/*
+ * Called by kernel/ptrace.c when detaching..
+ */
+void ptrace_disable(struct task_struct *child)
+{
+}
+
+/*
+ * We actually access the pt_regs stored on the kernel stack.
+ */
+static int ptrace_read_user(struct task_struct *tsk, unsigned long off,
+			    unsigned long __user *ret)
+{
+	unsigned long tmp;
+
+	tmp = 0;
+	if (off < sizeof(struct pt_regs))
+		tmp = get_user_reg(tsk, off >> 2);
+
+	return put_user(tmp, ret);
+}
+
+/*
+ * We actually access the pt_regs stored on the kernel stack.
+ */
+static int ptrace_write_user(struct task_struct *tsk, unsigned long off,
+			     unsigned long val)
+{
+	if (off >= sizeof(struct pt_regs))
+		return 0;
+
+	return put_user_reg(tsk, off >> 2, val);
+}
+
+long arch_ptrace(struct task_struct *child, long request,
+		 unsigned long addr, unsigned long data)
+{
+	int ret;
+	unsigned long __user *datap = (unsigned long __user *) data;
+
+	switch (request) {
+	case PTRACE_PEEKUSR:
+		ret = ptrace_read_user(child, addr, datap);
+		break;
+
+	case PTRACE_POKEUSR:
+		ret = ptrace_write_user(child, addr, data);
+		break;
+
+	case PTRACE_GET_THREAD_AREA:
+		ret = put_user(task_pt_regs(child)->UCreg_16,
+			       datap);
+		break;
+
+	default:
+		ret = ptrace_request(child, request, addr, data);
+		break;
+	}
+
+	return ret;
+}
+
+asmlinkage int syscall_trace(int why, struct pt_regs *regs, int scno)
+{
+	unsigned long ip;
+
+	if (!test_thread_flag(TIF_SYSCALL_TRACE))
+		return scno;
+	if (!(current->ptrace & PT_PTRACED))
+		return scno;
+
+	/*
+	 * Save IP.  IP is used to denote syscall entry/exit:
+	 *  IP = 0 -> entry, = 1 -> exit
+	 */
+	ip = regs->UCreg_ip;
+	regs->UCreg_ip = why;
+
+	current_thread_info()->syscall = scno;
+
+	/* the 0x80 provides a way for the tracing parent to distinguish
+	   between a syscall stop and SIGTRAP delivery */
+	ptrace_notify(SIGTRAP | ((current->ptrace & PT_TRACESYSGOOD)
+				 ? 0x80 : 0));
+	/*
+	 * this isn't the same as continuing with a signal, but it will do
+	 * for normal use.  strace only continues with a signal if the
+	 * stopping signal is not SIGTRAP.  -brl
+	 */
+	if (current->exit_code) {
+		send_sig(current->exit_code, current, 1);
+		current->exit_code = 0;
+	}
+	regs->UCreg_ip = ip;
+
+	return current_thread_info()->syscall;
+}
diff --git a/arch/unicore32/kernel/puv3-core.c b/arch/unicore32/kernel/puv3-core.c
new file mode 100644
index 000000000000..8b1b6beb858e
--- /dev/null
+++ b/arch/unicore32/kernel/puv3-core.c
@@ -0,0 +1,285 @@
+/*
+ *  linux/arch/unicore32/kernel/puv3-core.c
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ *	Maintained by GUAN Xue-tao <gxt@mprc.pku.edu.cn>
+ *	Copyright (C) 2001-2010 Guan Xuetao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/init.h>
+#include <linux/device.h>
+#include <linux/sysdev.h>
+#include <linux/amba/bus.h>
+#include <linux/platform_device.h>
+#include <linux/io.h>
+#include <linux/cnt32_to_63.h>
+#include <linux/usb/musb.h>
+
+#include <asm/irq.h>
+#include <mach/hardware.h>
+#include <mach/pm.h>
+
+/*
+ * This is the PKUnity sched_clock implementation.  This has
+ * a resolution of 271ns, and a maximum value of 32025597s (370 days).
+ *
+ * The return value is guaranteed to be monotonic in that range as
+ * long as there is always less than 582 seconds between successive
+ * calls to this function.
+ *
+ *  ( * 1E9 / CLOCK_TICK_RATE ) -> about 2235/32
+ */
+unsigned long long sched_clock(void)
+{
+	unsigned long long v = cnt32_to_63(readl(OST_OSCR));
+
+	/* original conservative method, but overflow frequently
+	 * v *= NSEC_PER_SEC >> 12;
+	 * do_div(v, CLOCK_TICK_RATE >> 12);
+	 */
+	v = ((v & 0x7fffffffffffffffULL) * 2235) >> 5;
+
+	return v;
+}
+
+static struct resource puv3_usb_resources[] = {
+	/* order is significant! */
+	{
+		.start		= io_v2p(PKUNITY_USB_BASE),
+		.end		= io_v2p(PKUNITY_USB_BASE) + 0x3ff,
+		.flags		= IORESOURCE_MEM,
+	}, {
+		.start		= IRQ_USB,
+		.flags		= IORESOURCE_IRQ,
+	}, {
+		.start		= IRQ_USB,
+		.flags		= IORESOURCE_IRQ,
+	},
+};
+
+static struct musb_hdrc_config	puv3_usb_config[] = {
+	{
+		.num_eps = 16,
+		.multipoint = 1,
+#ifdef CONFIG_USB_INVENTRA_DMA
+		.dma = 1,
+		.dma_channels = 8,
+#endif
+	},
+};
+
+static struct musb_hdrc_platform_data puv3_usb_plat = {
+	.mode		= MUSB_HOST,
+	.min_power	= 100,
+	.clock		= 0,
+	.config		= puv3_usb_config,
+};
+
+static struct resource puv3_mmc_resources[] = {
+	[0] = {
+		.start	= io_v2p(PKUNITY_SDC_BASE),
+		.end	= io_v2p(PKUNITY_SDC_BASE) + 0xfff,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= IRQ_SDC,
+		.end	= IRQ_SDC,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct resource puv3_unigfx_resources[] = {
+	[0] = {
+		.start	= io_v2p(PKUNITY_UNIGFX_BASE),
+		.end	= io_v2p(PKUNITY_UNIGFX_BASE) + 0xfff,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= PKUNITY_UNIGFX_MMAP_BASE,
+		.end	= PKUNITY_UNIGFX_MMAP_BASE + PKUNITY_UNIGFX_MMAP_SIZE,
+		.flags	= IORESOURCE_MEM,
+	},
+};
+
+static struct resource puv3_rtc_resources[] = {
+	[0] = {
+		.start = io_v2p(PKUNITY_RTC_BASE),
+		.end   = io_v2p(PKUNITY_RTC_BASE) + 0xff,
+		.flags = IORESOURCE_MEM,
+	},
+	[1] = {
+		.start = IRQ_RTCAlarm,
+		.end   = IRQ_RTCAlarm,
+		.flags = IORESOURCE_IRQ,
+	},
+	[2] = {
+		.start = IRQ_RTC,
+		.end   = IRQ_RTC,
+		.flags = IORESOURCE_IRQ
+	}
+};
+
+static struct resource puv3_pwm_resources[] = {
+	[0] = {
+		.start	= io_v2p(PKUNITY_OST_BASE) + 0x80,
+		.end	= io_v2p(PKUNITY_OST_BASE) + 0xff,
+		.flags	= IORESOURCE_MEM,
+	},
+};
+
+static struct resource puv3_uart0_resources[] = {
+	[0] = {
+		.start = io_v2p(PKUNITY_UART0_BASE),
+		.end   = io_v2p(PKUNITY_UART0_BASE) + 0xff,
+		.flags = IORESOURCE_MEM,
+	},
+	[1] = {
+		.start = IRQ_UART0,
+		.end   = IRQ_UART0,
+		.flags = IORESOURCE_IRQ
+	}
+};
+
+static struct resource puv3_uart1_resources[] = {
+	[0] = {
+		.start = io_v2p(PKUNITY_UART1_BASE),
+		.end   = io_v2p(PKUNITY_UART1_BASE) + 0xff,
+		.flags = IORESOURCE_MEM,
+	},
+	[1] = {
+		.start = IRQ_UART1,
+		.end   = IRQ_UART1,
+		.flags = IORESOURCE_IRQ
+	}
+};
+
+static struct resource puv3_umal_resources[] = {
+	[0] = {
+		.start = io_v2p(PKUNITY_UMAL_BASE),
+		.end   = io_v2p(PKUNITY_UMAL_BASE) + 0x1fff,
+		.flags = IORESOURCE_MEM,
+	},
+	[1] = {
+		.start = IRQ_UMAL,
+		.end   = IRQ_UMAL,
+		.flags = IORESOURCE_IRQ
+	}
+};
+
+#ifdef CONFIG_PUV3_PM
+
+#define SAVE(x)		sleep_save[SLEEP_SAVE_##x] = x
+#define RESTORE(x)	x = sleep_save[SLEEP_SAVE_##x]
+
+/*
+ * List of global PXA peripheral registers to preserve.
+ * More ones like CP and general purpose register values are preserved
+ * with the stack pointer in sleep.S.
+ */
+enum {
+	SLEEP_SAVE_PM_PLLDDRCFG,
+	SLEEP_SAVE_COUNT
+};
+
+
+static void puv3_cpu_pm_save(unsigned long *sleep_save)
+{
+/*	SAVE(PM_PLLDDRCFG); */
+}
+
+static void puv3_cpu_pm_restore(unsigned long *sleep_save)
+{
+/*	RESTORE(PM_PLLDDRCFG); */
+}
+
+static int puv3_cpu_pm_prepare(void)
+{
+	/* set resume return address */
+	writel(virt_to_phys(puv3_cpu_resume), PM_DIVCFG);
+	return 0;
+}
+
+static void puv3_cpu_pm_enter(suspend_state_t state)
+{
+	/* Clear reset status */
+	writel(RESETC_RSSR_HWR | RESETC_RSSR_WDR
+			| RESETC_RSSR_SMR | RESETC_RSSR_SWR, RESETC_RSSR);
+
+	switch (state) {
+/*	case PM_SUSPEND_ON:
+		puv3_cpu_idle();
+		break; */
+	case PM_SUSPEND_MEM:
+		puv3_cpu_pm_prepare();
+		puv3_cpu_suspend(PM_PMCR_SFB);
+		break;
+	}
+}
+
+static int puv3_cpu_pm_valid(suspend_state_t state)
+{
+	return state == PM_SUSPEND_MEM;
+}
+
+static void puv3_cpu_pm_finish(void)
+{
+	/* ensure not to come back here if it wasn't intended */
+	/* PSPR = 0; */
+}
+
+static struct puv3_cpu_pm_fns puv3_cpu_pm_fnss = {
+	.save_count	= SLEEP_SAVE_COUNT,
+	.valid		= puv3_cpu_pm_valid,
+	.save		= puv3_cpu_pm_save,
+	.restore	= puv3_cpu_pm_restore,
+	.enter		= puv3_cpu_pm_enter,
+	.prepare	= puv3_cpu_pm_prepare,
+	.finish		= puv3_cpu_pm_finish,
+};
+
+static void __init puv3_init_pm(void)
+{
+	puv3_cpu_pm_fns = &puv3_cpu_pm_fnss;
+}
+#else
+static inline void puv3_init_pm(void) {}
+#endif
+
+void puv3_ps2_init(void)
+{
+	struct clk *bclk32;
+
+	bclk32 = clk_get(NULL, "BUS32_CLK");
+	writel(clk_get_rate(bclk32) / 200000, PS2_CNT); /* should > 5us */
+}
+
+void __init puv3_core_init(void)
+{
+	puv3_init_pm();
+	puv3_ps2_init();
+
+	platform_device_register_simple("PKUnity-v3-RTC", -1,
+			puv3_rtc_resources, ARRAY_SIZE(puv3_rtc_resources));
+	platform_device_register_simple("PKUnity-v3-UMAL", -1,
+			puv3_umal_resources, ARRAY_SIZE(puv3_umal_resources));
+	platform_device_register_simple("PKUnity-v3-MMC", -1,
+			puv3_mmc_resources, ARRAY_SIZE(puv3_mmc_resources));
+	platform_device_register_simple("PKUnity-v3-UNIGFX", -1,
+			puv3_unigfx_resources, ARRAY_SIZE(puv3_unigfx_resources));
+	platform_device_register_simple("PKUnity-v3-PWM", -1,
+			puv3_pwm_resources, ARRAY_SIZE(puv3_pwm_resources));
+	platform_device_register_simple("PKUnity-v3-UART", 0,
+			puv3_uart0_resources, ARRAY_SIZE(puv3_uart0_resources));
+	platform_device_register_simple("PKUnity-v3-UART", 1,
+			puv3_uart1_resources, ARRAY_SIZE(puv3_uart1_resources));
+	platform_device_register_simple("PKUnity-v3-AC97", -1, NULL, 0);
+	platform_device_register_resndata(&platform_bus, "musb_hdrc", -1,
+			puv3_usb_resources, ARRAY_SIZE(puv3_usb_resources),
+			&puv3_usb_plat, sizeof(puv3_usb_plat));
+}
+
diff --git a/arch/unicore32/kernel/puv3-nb0916.c b/arch/unicore32/kernel/puv3-nb0916.c
new file mode 100644
index 000000000000..e731c561ed4e
--- /dev/null
+++ b/arch/unicore32/kernel/puv3-nb0916.c
@@ -0,0 +1,145 @@
+/*
+ * linux/arch/unicore32/kernel/puv3-nb0916.c
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ *	Maintained by GUAN Xue-tao <gxt@mprc.pku.edu.cn>
+ *	Copyright (C) 2001-2010 Guan Xuetao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/init.h>
+#include <linux/device.h>
+#include <linux/sysdev.h>
+#include <linux/platform_device.h>
+#include <linux/mtd/physmap.h>
+#include <linux/io.h>
+#include <linux/reboot.h>
+#include <linux/interrupt.h>
+#include <linux/i2c.h>
+#include <linux/pwm_backlight.h>
+#include <linux/gpio.h>
+#include <linux/gpio_keys.h>
+#include <linux/input.h>
+
+#include <mach/hardware.h>
+
+static struct physmap_flash_data physmap_flash_data = {
+	.width		= 1,
+};
+
+static struct resource physmap_flash_resource = {
+	.start		= 0xFFF80000,
+	.end		= 0xFFFFFFFF,
+	.flags		= IORESOURCE_MEM,
+};
+
+static struct resource puv3_i2c_resources[] = {
+	[0] = {
+		.start = io_v2p(PKUNITY_I2C_BASE),
+		.end   = io_v2p(PKUNITY_I2C_BASE) + 0xff,
+		.flags = IORESOURCE_MEM,
+	},
+	[1] = {
+		.start = IRQ_I2C,
+		.end   = IRQ_I2C,
+		.flags = IORESOURCE_IRQ,
+	}
+};
+
+static struct platform_pwm_backlight_data nb0916_backlight_data = {
+	.pwm_id		= 0,
+	.max_brightness	= 100,
+	.dft_brightness	= 100,
+	.pwm_period_ns	= 70 * 1024,
+};
+
+static struct gpio_keys_button nb0916_gpio_keys[] = {
+	{
+		.type	= EV_KEY,
+		.code	= KEY_POWER,
+		.gpio	= GPI_SOFF_REQ,
+		.desc	= "Power Button",
+		.wakeup = 1,
+		.active_low = 1,
+	},
+	{
+		.type	= EV_KEY,
+		.code	= BTN_TOUCH,
+		.gpio	= GPI_BTN_TOUCH,
+		.desc	= "Touchpad Button",
+		.wakeup = 1,
+		.active_low = 1,
+	},
+};
+
+static struct gpio_keys_platform_data nb0916_gpio_button_data = {
+	.buttons	= nb0916_gpio_keys,
+	.nbuttons	= ARRAY_SIZE(nb0916_gpio_keys),
+};
+
+static irqreturn_t nb0916_lcdcaseoff_handler(int irq, void *dev_id)
+{
+	if (gpio_get_value(GPI_LCD_CASE_OFF))
+		gpio_set_value(GPO_LCD_EN, 1);
+	else
+		gpio_set_value(GPO_LCD_EN, 0);
+
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t nb0916_overheat_handler(int irq, void *dev_id)
+{
+	machine_halt();
+	/* SYSTEM HALT, NO RETURN */
+	return IRQ_HANDLED;
+}
+
+static struct i2c_board_info __initdata puv3_i2c_devices[] = {
+	{	I2C_BOARD_INFO("lm75",		I2C_TAR_THERMAL),	},
+	{	I2C_BOARD_INFO("bq27200",	I2C_TAR_PWIC),		},
+	{	I2C_BOARD_INFO("24c02",		I2C_TAR_EEPROM),	},
+};
+
+int __init mach_nb0916_init(void)
+{
+	i2c_register_board_info(0, puv3_i2c_devices,
+			ARRAY_SIZE(puv3_i2c_devices));
+
+	platform_device_register_simple("PKUnity-v3-I2C", -1,
+			puv3_i2c_resources, ARRAY_SIZE(puv3_i2c_resources));
+
+	platform_device_register_data(&platform_bus, "pwm-backlight", -1,
+			&nb0916_backlight_data, sizeof(nb0916_backlight_data));
+
+	platform_device_register_data(&platform_bus, "gpio-keys", -1,
+			&nb0916_gpio_button_data, sizeof(nb0916_gpio_button_data));
+
+	platform_device_register_resndata(&platform_bus, "physmap-flash", -1,
+			&physmap_flash_resource, 1,
+			&physmap_flash_data, sizeof(physmap_flash_data));
+
+	if (request_irq(gpio_to_irq(GPI_LCD_CASE_OFF),
+		&nb0916_lcdcaseoff_handler,
+		IRQF_DISABLED | IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING,
+		"NB0916 lcd case off", NULL) < 0) {
+
+		printk(KERN_DEBUG "LCD-Case-OFF IRQ %d not available\n",
+			gpio_to_irq(GPI_LCD_CASE_OFF));
+	}
+
+	if (request_irq(gpio_to_irq(GPI_OTP_INT), &nb0916_overheat_handler,
+		IRQF_DISABLED | IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING,
+		"NB0916 overheating protection", NULL) < 0) {
+
+		printk(KERN_DEBUG "Overheating Protection IRQ %d not available\n",
+			gpio_to_irq(GPI_OTP_INT));
+	}
+
+	return 0;
+}
+
+subsys_initcall_sync(mach_nb0916_init);
diff --git a/arch/unicore32/kernel/pwm.c b/arch/unicore32/kernel/pwm.c
new file mode 100644
index 000000000000..4615d51e3ba6
--- /dev/null
+++ b/arch/unicore32/kernel/pwm.c
@@ -0,0 +1,263 @@
+/*
+ * linux/arch/unicore32/kernel/pwm.c
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ *	Maintained by GUAN Xue-tao <gxt@mprc.pku.edu.cn>
+ *	Copyright (C) 2001-2010 Guan Xuetao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/platform_device.h>
+#include <linux/slab.h>
+#include <linux/err.h>
+#include <linux/clk.h>
+#include <linux/io.h>
+#include <linux/pwm.h>
+
+#include <asm/div64.h>
+#include <mach/hardware.h>
+
+struct pwm_device {
+	struct list_head	node;
+	struct platform_device *pdev;
+
+	const char	*label;
+	struct clk	*clk;
+	int		clk_enabled;
+
+	unsigned int	use_count;
+	unsigned int	pwm_id;
+};
+
+/*
+ * period_ns = 10^9 * (PRESCALE + 1) * (PV + 1) / PWM_CLK_RATE
+ * duty_ns   = 10^9 * (PRESCALE + 1) * DC / PWM_CLK_RATE
+ */
+int pwm_config(struct pwm_device *pwm, int duty_ns, int period_ns)
+{
+	unsigned long long c;
+	unsigned long period_cycles, prescale, pv, dc;
+
+	if (pwm == NULL || period_ns == 0 || duty_ns > period_ns)
+		return -EINVAL;
+
+	c = clk_get_rate(pwm->clk);
+	c = c * period_ns;
+	do_div(c, 1000000000);
+	period_cycles = c;
+
+	if (period_cycles < 1)
+		period_cycles = 1;
+	prescale = (period_cycles - 1) / 1024;
+	pv = period_cycles / (prescale + 1) - 1;
+
+	if (prescale > 63)
+		return -EINVAL;
+
+	if (duty_ns == period_ns)
+		dc = OST_PWMDCCR_FDCYCLE;
+	else
+		dc = (pv + 1) * duty_ns / period_ns;
+
+	/* NOTE: the clock to PWM has to be enabled first
+	 * before writing to the registers
+	 */
+	clk_enable(pwm->clk);
+	OST_PWMPWCR = prescale;
+	OST_PWMDCCR = pv - dc;
+	OST_PWMPCR  = pv;
+	clk_disable(pwm->clk);
+
+	return 0;
+}
+EXPORT_SYMBOL(pwm_config);
+
+int pwm_enable(struct pwm_device *pwm)
+{
+	int rc = 0;
+
+	if (!pwm->clk_enabled) {
+		rc = clk_enable(pwm->clk);
+		if (!rc)
+			pwm->clk_enabled = 1;
+	}
+	return rc;
+}
+EXPORT_SYMBOL(pwm_enable);
+
+void pwm_disable(struct pwm_device *pwm)
+{
+	if (pwm->clk_enabled) {
+		clk_disable(pwm->clk);
+		pwm->clk_enabled = 0;
+	}
+}
+EXPORT_SYMBOL(pwm_disable);
+
+static DEFINE_MUTEX(pwm_lock);
+static LIST_HEAD(pwm_list);
+
+struct pwm_device *pwm_request(int pwm_id, const char *label)
+{
+	struct pwm_device *pwm;
+	int found = 0;
+
+	mutex_lock(&pwm_lock);
+
+	list_for_each_entry(pwm, &pwm_list, node) {
+		if (pwm->pwm_id == pwm_id) {
+			found = 1;
+			break;
+		}
+	}
+
+	if (found) {
+		if (pwm->use_count == 0) {
+			pwm->use_count++;
+			pwm->label = label;
+		} else
+			pwm = ERR_PTR(-EBUSY);
+	} else
+		pwm = ERR_PTR(-ENOENT);
+
+	mutex_unlock(&pwm_lock);
+	return pwm;
+}
+EXPORT_SYMBOL(pwm_request);
+
+void pwm_free(struct pwm_device *pwm)
+{
+	mutex_lock(&pwm_lock);
+
+	if (pwm->use_count) {
+		pwm->use_count--;
+		pwm->label = NULL;
+	} else
+		pr_warning("PWM device already freed\n");
+
+	mutex_unlock(&pwm_lock);
+}
+EXPORT_SYMBOL(pwm_free);
+
+static inline void __add_pwm(struct pwm_device *pwm)
+{
+	mutex_lock(&pwm_lock);
+	list_add_tail(&pwm->node, &pwm_list);
+	mutex_unlock(&pwm_lock);
+}
+
+static struct pwm_device *pwm_probe(struct platform_device *pdev,
+		unsigned int pwm_id, struct pwm_device *parent_pwm)
+{
+	struct pwm_device *pwm;
+	struct resource *r;
+	int ret = 0;
+
+	pwm = kzalloc(sizeof(struct pwm_device), GFP_KERNEL);
+	if (pwm == NULL) {
+		dev_err(&pdev->dev, "failed to allocate memory\n");
+		return ERR_PTR(-ENOMEM);
+	}
+
+	pwm->clk = clk_get(NULL, "OST_CLK");
+	if (IS_ERR(pwm->clk)) {
+		ret = PTR_ERR(pwm->clk);
+		goto err_free;
+	}
+	pwm->clk_enabled = 0;
+
+	pwm->use_count = 0;
+	pwm->pwm_id = pwm_id;
+	pwm->pdev = pdev;
+
+	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (r == NULL) {
+		dev_err(&pdev->dev, "no memory resource defined\n");
+		ret = -ENODEV;
+		goto err_free_clk;
+	}
+
+	r = request_mem_region(r->start, resource_size(r), pdev->name);
+	if (r == NULL) {
+		dev_err(&pdev->dev, "failed to request memory resource\n");
+		ret = -EBUSY;
+		goto err_free_clk;
+	}
+
+	__add_pwm(pwm);
+	platform_set_drvdata(pdev, pwm);
+	return pwm;
+
+err_free_clk:
+	clk_put(pwm->clk);
+err_free:
+	kfree(pwm);
+	return ERR_PTR(ret);
+}
+
+static int __devinit puv3_pwm_probe(struct platform_device *pdev)
+{
+	struct pwm_device *pwm = pwm_probe(pdev, pdev->id, NULL);
+
+	if (IS_ERR(pwm))
+		return PTR_ERR(pwm);
+
+	return 0;
+}
+
+static int __devexit pwm_remove(struct platform_device *pdev)
+{
+	struct pwm_device *pwm;
+	struct resource *r;
+
+	pwm = platform_get_drvdata(pdev);
+	if (pwm == NULL)
+		return -ENODEV;
+
+	mutex_lock(&pwm_lock);
+	list_del(&pwm->node);
+	mutex_unlock(&pwm_lock);
+
+	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	release_mem_region(r->start, resource_size(r));
+
+	clk_put(pwm->clk);
+	kfree(pwm);
+	return 0;
+}
+
+static struct platform_driver puv3_pwm_driver = {
+	.driver		= {
+		.name	= "PKUnity-v3-PWM",
+	},
+	.probe		= puv3_pwm_probe,
+	.remove		= __devexit_p(pwm_remove),
+};
+
+static int __init pwm_init(void)
+{
+	int ret = 0;
+
+	ret = platform_driver_register(&puv3_pwm_driver);
+	if (ret) {
+		printk(KERN_ERR "failed to register puv3_pwm_driver\n");
+		return ret;
+	}
+
+	return ret;
+}
+arch_initcall(pwm_init);
+
+static void __exit pwm_exit(void)
+{
+	platform_driver_unregister(&puv3_pwm_driver);
+}
+module_exit(pwm_exit);
+
+MODULE_LICENSE("GPL v2");
diff --git a/arch/unicore32/kernel/rtc.c b/arch/unicore32/kernel/rtc.c
new file mode 100644
index 000000000000..c5f068295b51
--- /dev/null
+++ b/arch/unicore32/kernel/rtc.c
@@ -0,0 +1,380 @@
+/*
+ * linux/arch/unicore32/kernel/rtc.c
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ *	Maintained by GUAN Xue-tao <gxt@mprc.pku.edu.cn>
+ *	Copyright (C) 2001-2010 Guan Xuetao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/module.h>
+#include <linux/fs.h>
+#include <linux/string.h>
+#include <linux/init.h>
+#include <linux/platform_device.h>
+#include <linux/interrupt.h>
+#include <linux/rtc.h>
+#include <linux/bcd.h>
+#include <linux/clk.h>
+#include <linux/log2.h>
+#include <linux/slab.h>
+#include <linux/uaccess.h>
+#include <linux/io.h>
+
+#include <asm/irq.h>
+#include <mach/hardware.h>
+
+static struct resource *puv3_rtc_mem;
+
+static int puv3_rtc_alarmno = IRQ_RTCAlarm;
+static int puv3_rtc_tickno  = IRQ_RTC;
+
+static DEFINE_SPINLOCK(puv3_rtc_pie_lock);
+
+/* IRQ Handlers */
+
+static irqreturn_t puv3_rtc_alarmirq(int irq, void *id)
+{
+	struct rtc_device *rdev = id;
+
+	writel(readl(RTC_RTSR) | RTC_RTSR_AL, RTC_RTSR);
+	rtc_update_irq(rdev, 1, RTC_AF | RTC_IRQF);
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t puv3_rtc_tickirq(int irq, void *id)
+{
+	struct rtc_device *rdev = id;
+
+	writel(readl(RTC_RTSR) | RTC_RTSR_HZ, RTC_RTSR);
+	rtc_update_irq(rdev, 1, RTC_PF | RTC_IRQF);
+	return IRQ_HANDLED;
+}
+
+/* Update control registers */
+static void puv3_rtc_setaie(int to)
+{
+	unsigned int tmp;
+
+	pr_debug("%s: aie=%d\n", __func__, to);
+
+	tmp = readl(RTC_RTSR) & ~RTC_RTSR_ALE;
+
+	if (to)
+		tmp |= RTC_RTSR_ALE;
+
+	writel(tmp, RTC_RTSR);
+}
+
+static int puv3_rtc_setpie(struct device *dev, int enabled)
+{
+	unsigned int tmp;
+
+	pr_debug("%s: pie=%d\n", __func__, enabled);
+
+	spin_lock_irq(&puv3_rtc_pie_lock);
+	tmp = readl(RTC_RTSR) & ~RTC_RTSR_HZE;
+
+	if (enabled)
+		tmp |= RTC_RTSR_HZE;
+
+	writel(tmp, RTC_RTSR);
+	spin_unlock_irq(&puv3_rtc_pie_lock);
+
+	return 0;
+}
+
+static int puv3_rtc_setfreq(struct device *dev, int freq)
+{
+	return 0;
+}
+
+/* Time read/write */
+
+static int puv3_rtc_gettime(struct device *dev, struct rtc_time *rtc_tm)
+{
+	rtc_time_to_tm(readl(RTC_RCNR), rtc_tm);
+
+	pr_debug("read time %02x.%02x.%02x %02x/%02x/%02x\n",
+		 rtc_tm->tm_year, rtc_tm->tm_mon, rtc_tm->tm_mday,
+		 rtc_tm->tm_hour, rtc_tm->tm_min, rtc_tm->tm_sec);
+
+	return 0;
+}
+
+static int puv3_rtc_settime(struct device *dev, struct rtc_time *tm)
+{
+	unsigned long rtc_count = 0;
+
+	pr_debug("set time %02d.%02d.%02d %02d/%02d/%02d\n",
+		 tm->tm_year, tm->tm_mon, tm->tm_mday,
+		 tm->tm_hour, tm->tm_min, tm->tm_sec);
+
+	rtc_tm_to_time(tm, &rtc_count);
+	writel(rtc_count, RTC_RCNR);
+
+	return 0;
+}
+
+static int puv3_rtc_getalarm(struct device *dev, struct rtc_wkalrm *alrm)
+{
+	struct rtc_time *alm_tm = &alrm->time;
+
+	rtc_time_to_tm(readl(RTC_RTAR), alm_tm);
+
+	alrm->enabled = readl(RTC_RTSR) & RTC_RTSR_ALE;
+
+	pr_debug("read alarm %02x %02x.%02x.%02x %02x/%02x/%02x\n",
+		 alrm->enabled,
+		 alm_tm->tm_year, alm_tm->tm_mon, alm_tm->tm_mday,
+		 alm_tm->tm_hour, alm_tm->tm_min, alm_tm->tm_sec);
+
+	return 0;
+}
+
+static int puv3_rtc_setalarm(struct device *dev, struct rtc_wkalrm *alrm)
+{
+	struct rtc_time *tm = &alrm->time;
+	unsigned long rtcalarm_count = 0;
+
+	pr_debug("puv3_rtc_setalarm: %d, %02x/%02x/%02x %02x.%02x.%02x\n",
+		 alrm->enabled,
+		 tm->tm_mday & 0xff, tm->tm_mon & 0xff, tm->tm_year & 0xff,
+		 tm->tm_hour & 0xff, tm->tm_min & 0xff, tm->tm_sec);
+
+	rtc_tm_to_time(tm, &rtcalarm_count);
+	writel(rtcalarm_count, RTC_RTAR);
+
+	puv3_rtc_setaie(alrm->enabled);
+
+	if (alrm->enabled)
+		enable_irq_wake(puv3_rtc_alarmno);
+	else
+		disable_irq_wake(puv3_rtc_alarmno);
+
+	return 0;
+}
+
+static int puv3_rtc_proc(struct device *dev, struct seq_file *seq)
+{
+	seq_printf(seq, "periodic_IRQ\t: %s\n",
+		     (readl(RTC_RTSR) & RTC_RTSR_HZE) ? "yes" : "no");
+	return 0;
+}
+
+static int puv3_rtc_open(struct device *dev)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct rtc_device *rtc_dev = platform_get_drvdata(pdev);
+	int ret;
+
+	ret = request_irq(puv3_rtc_alarmno, puv3_rtc_alarmirq,
+			  IRQF_DISABLED,  "pkunity-rtc alarm", rtc_dev);
+
+	if (ret) {
+		dev_err(dev, "IRQ%d error %d\n", puv3_rtc_alarmno, ret);
+		return ret;
+	}
+
+	ret = request_irq(puv3_rtc_tickno, puv3_rtc_tickirq,
+			  IRQF_DISABLED,  "pkunity-rtc tick", rtc_dev);
+
+	if (ret) {
+		dev_err(dev, "IRQ%d error %d\n", puv3_rtc_tickno, ret);
+		goto tick_err;
+	}
+
+	return ret;
+
+ tick_err:
+	free_irq(puv3_rtc_alarmno, rtc_dev);
+	return ret;
+}
+
+static void puv3_rtc_release(struct device *dev)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct rtc_device *rtc_dev = platform_get_drvdata(pdev);
+
+	/* do not clear AIE here, it may be needed for wake */
+
+	puv3_rtc_setpie(dev, 0);
+	free_irq(puv3_rtc_alarmno, rtc_dev);
+	free_irq(puv3_rtc_tickno, rtc_dev);
+}
+
+static const struct rtc_class_ops puv3_rtcops = {
+	.open		= puv3_rtc_open,
+	.release	= puv3_rtc_release,
+	.read_time	= puv3_rtc_gettime,
+	.set_time	= puv3_rtc_settime,
+	.read_alarm	= puv3_rtc_getalarm,
+	.set_alarm	= puv3_rtc_setalarm,
+	.irq_set_freq	= puv3_rtc_setfreq,
+	.irq_set_state	= puv3_rtc_setpie,
+	.proc	        = puv3_rtc_proc,
+};
+
+static void puv3_rtc_enable(struct platform_device *pdev, int en)
+{
+	if (!en) {
+		writel(readl(RTC_RTSR) & ~RTC_RTSR_HZE, RTC_RTSR);
+	} else {
+		/* re-enable the device, and check it is ok */
+
+		if ((readl(RTC_RTSR) & RTC_RTSR_HZE) == 0) {
+			dev_info(&pdev->dev, "rtc disabled, re-enabling\n");
+			writel(readl(RTC_RTSR) | RTC_RTSR_HZE, RTC_RTSR);
+		}
+	}
+}
+
+static int puv3_rtc_remove(struct platform_device *dev)
+{
+	struct rtc_device *rtc = platform_get_drvdata(dev);
+
+	platform_set_drvdata(dev, NULL);
+	rtc_device_unregister(rtc);
+
+	puv3_rtc_setpie(&dev->dev, 0);
+	puv3_rtc_setaie(0);
+
+	release_resource(puv3_rtc_mem);
+	kfree(puv3_rtc_mem);
+
+	return 0;
+}
+
+static int puv3_rtc_probe(struct platform_device *pdev)
+{
+	struct rtc_device *rtc;
+	struct resource *res;
+	int ret;
+
+	pr_debug("%s: probe=%p\n", __func__, pdev);
+
+	/* find the IRQs */
+
+	puv3_rtc_tickno = platform_get_irq(pdev, 1);
+	if (puv3_rtc_tickno < 0) {
+		dev_err(&pdev->dev, "no irq for rtc tick\n");
+		return -ENOENT;
+	}
+
+	puv3_rtc_alarmno = platform_get_irq(pdev, 0);
+	if (puv3_rtc_alarmno < 0) {
+		dev_err(&pdev->dev, "no irq for alarm\n");
+		return -ENOENT;
+	}
+
+	pr_debug("PKUnity_rtc: tick irq %d, alarm irq %d\n",
+		 puv3_rtc_tickno, puv3_rtc_alarmno);
+
+	/* get the memory region */
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (res == NULL) {
+		dev_err(&pdev->dev, "failed to get memory region resource\n");
+		return -ENOENT;
+	}
+
+	puv3_rtc_mem = request_mem_region(res->start,
+					 res->end-res->start+1,
+					 pdev->name);
+
+	if (puv3_rtc_mem == NULL) {
+		dev_err(&pdev->dev, "failed to reserve memory region\n");
+		ret = -ENOENT;
+		goto err_nores;
+	}
+
+	puv3_rtc_enable(pdev, 1);
+
+	puv3_rtc_setfreq(&pdev->dev, 1);
+
+	/* register RTC and exit */
+
+	rtc = rtc_device_register("pkunity", &pdev->dev, &puv3_rtcops,
+				  THIS_MODULE);
+
+	if (IS_ERR(rtc)) {
+		dev_err(&pdev->dev, "cannot attach rtc\n");
+		ret = PTR_ERR(rtc);
+		goto err_nortc;
+	}
+
+	/* platform setup code should have handled this; sigh */
+	if (!device_can_wakeup(&pdev->dev))
+		device_init_wakeup(&pdev->dev, 1);
+
+	platform_set_drvdata(pdev, rtc);
+	return 0;
+
+ err_nortc:
+	puv3_rtc_enable(pdev, 0);
+	release_resource(puv3_rtc_mem);
+
+ err_nores:
+	return ret;
+}
+
+#ifdef CONFIG_PM
+
+/* RTC Power management control */
+
+static int ticnt_save;
+
+static int puv3_rtc_suspend(struct platform_device *pdev, pm_message_t state)
+{
+	/* save RTAR for anyone using periodic interrupts */
+	ticnt_save = readl(RTC_RTAR);
+	puv3_rtc_enable(pdev, 0);
+	return 0;
+}
+
+static int puv3_rtc_resume(struct platform_device *pdev)
+{
+	puv3_rtc_enable(pdev, 1);
+	writel(ticnt_save, RTC_RTAR);
+	return 0;
+}
+#else
+#define puv3_rtc_suspend NULL
+#define puv3_rtc_resume  NULL
+#endif
+
+static struct platform_driver puv3_rtcdrv = {
+	.probe		= puv3_rtc_probe,
+	.remove		= __devexit_p(puv3_rtc_remove),
+	.suspend	= puv3_rtc_suspend,
+	.resume		= puv3_rtc_resume,
+	.driver		= {
+		.name	= "PKUnity-v3-RTC",
+		.owner	= THIS_MODULE,
+	}
+};
+
+static char __initdata banner[] = "PKUnity-v3 RTC, (c) 2009 PKUnity Co.\n";
+
+static int __init puv3_rtc_init(void)
+{
+	printk(banner);
+	return platform_driver_register(&puv3_rtcdrv);
+}
+
+static void __exit puv3_rtc_exit(void)
+{
+	platform_driver_unregister(&puv3_rtcdrv);
+}
+
+module_init(puv3_rtc_init);
+module_exit(puv3_rtc_exit);
+
+MODULE_DESCRIPTION("RTC Driver for the PKUnity v3 chip");
+MODULE_AUTHOR("Hu Dongliang");
+MODULE_LICENSE("GPL v2");
+
diff --git a/arch/unicore32/kernel/setup.c b/arch/unicore32/kernel/setup.c
new file mode 100644
index 000000000000..1e175a82844d
--- /dev/null
+++ b/arch/unicore32/kernel/setup.c
@@ -0,0 +1,360 @@
+/*
+ * linux/arch/unicore32/kernel/setup.c
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/stddef.h>
+#include <linux/ioport.h>
+#include <linux/delay.h>
+#include <linux/utsname.h>
+#include <linux/initrd.h>
+#include <linux/console.h>
+#include <linux/bootmem.h>
+#include <linux/seq_file.h>
+#include <linux/screen_info.h>
+#include <linux/init.h>
+#include <linux/root_dev.h>
+#include <linux/cpu.h>
+#include <linux/interrupt.h>
+#include <linux/smp.h>
+#include <linux/fs.h>
+#include <linux/proc_fs.h>
+#include <linux/memblock.h>
+#include <linux/elf.h>
+#include <linux/io.h>
+
+#include <asm/cputype.h>
+#include <asm/sections.h>
+#include <asm/setup.h>
+#include <asm/cacheflush.h>
+#include <asm/tlbflush.h>
+#include <asm/traps.h>
+
+#include "setup.h"
+
+#ifndef MEM_SIZE
+#define MEM_SIZE	(16*1024*1024)
+#endif
+
+struct stack {
+	u32 irq[3];
+	u32 abt[3];
+	u32 und[3];
+} ____cacheline_aligned;
+
+static struct stack stacks[NR_CPUS];
+
+char elf_platform[ELF_PLATFORM_SIZE];
+EXPORT_SYMBOL(elf_platform);
+
+static char __initdata cmd_line[COMMAND_LINE_SIZE];
+
+static char default_command_line[COMMAND_LINE_SIZE] __initdata = CONFIG_CMDLINE;
+
+/*
+ * Standard memory resources
+ */
+static struct resource mem_res[] = {
+	{
+		.name = "Video RAM",
+		.start = 0,
+		.end = 0,
+		.flags = IORESOURCE_MEM
+	},
+	{
+		.name = "Kernel text",
+		.start = 0,
+		.end = 0,
+		.flags = IORESOURCE_MEM
+	},
+	{
+		.name = "Kernel data",
+		.start = 0,
+		.end = 0,
+		.flags = IORESOURCE_MEM
+	}
+};
+
+#define video_ram   mem_res[0]
+#define kernel_code mem_res[1]
+#define kernel_data mem_res[2]
+
+/*
+ * These functions re-use the assembly code in head.S, which
+ * already provide the required functionality.
+ */
+static void __init setup_processor(void)
+{
+	printk(KERN_DEFAULT "CPU: UniCore-II [%08x] revision %d, cr=%08lx\n",
+	       uc32_cpuid, (int)(uc32_cpuid >> 16) & 15, cr_alignment);
+
+	sprintf(init_utsname()->machine, "puv3");
+	sprintf(elf_platform, "ucv2");
+}
+
+/*
+ * cpu_init - initialise one CPU.
+ *
+ * cpu_init sets up the per-CPU stacks.
+ */
+void cpu_init(void)
+{
+	unsigned int cpu = smp_processor_id();
+	struct stack *stk = &stacks[cpu];
+
+	/*
+	 * setup stacks for re-entrant exception handlers
+	 */
+	__asm__ (
+	"mov.a	asr, %1\n\t"
+	"add	sp, %0, %2\n\t"
+	"mov.a	asr, %3\n\t"
+	"add	sp, %0, %4\n\t"
+	"mov.a	asr, %5\n\t"
+	"add	sp, %0, %6\n\t"
+	"mov.a	asr, %7"
+	    :
+	    : "r" (stk),
+	      "r" (PSR_R_BIT | PSR_I_BIT | INTR_MODE),
+	      "I" (offsetof(struct stack, irq[0])),
+	      "r" (PSR_R_BIT | PSR_I_BIT | ABRT_MODE),
+	      "I" (offsetof(struct stack, abt[0])),
+	      "r" (PSR_R_BIT | PSR_I_BIT | EXTN_MODE),
+	      "I" (offsetof(struct stack, und[0])),
+	      "r" (PSR_R_BIT | PSR_I_BIT | PRIV_MODE)
+	: "r30", "cc");
+}
+
+static int __init uc32_add_memory(unsigned long start, unsigned long size)
+{
+	struct membank *bank = &meminfo.bank[meminfo.nr_banks];
+
+	if (meminfo.nr_banks >= NR_BANKS) {
+		printk(KERN_CRIT "NR_BANKS too low, "
+			"ignoring memory at %#lx\n", start);
+		return -EINVAL;
+	}
+
+	/*
+	 * Ensure that start/size are aligned to a page boundary.
+	 * Size is appropriately rounded down, start is rounded up.
+	 */
+	size -= start & ~PAGE_MASK;
+
+	bank->start = PAGE_ALIGN(start);
+	bank->size  = size & PAGE_MASK;
+
+	/*
+	 * Check whether this memory region has non-zero size or
+	 * invalid node number.
+	 */
+	if (bank->size == 0)
+		return -EINVAL;
+
+	meminfo.nr_banks++;
+	return 0;
+}
+
+/*
+ * Pick out the memory size.  We look for mem=size@start,
+ * where start and size are "size[KkMm]"
+ */
+static int __init early_mem(char *p)
+{
+	static int usermem __initdata = 1;
+	unsigned long size, start;
+	char *endp;
+
+	/*
+	 * If the user specifies memory size, we
+	 * blow away any automatically generated
+	 * size.
+	 */
+	if (usermem) {
+		usermem = 0;
+		meminfo.nr_banks = 0;
+	}
+
+	start = PHYS_OFFSET;
+	size  = memparse(p, &endp);
+	if (*endp == '@')
+		start = memparse(endp + 1, NULL);
+
+	uc32_add_memory(start, size);
+
+	return 0;
+}
+early_param("mem", early_mem);
+
+static void __init
+request_standard_resources(struct meminfo *mi)
+{
+	struct resource *res;
+	int i;
+
+	kernel_code.start   = virt_to_phys(_stext);
+	kernel_code.end     = virt_to_phys(_etext - 1);
+	kernel_data.start   = virt_to_phys(_sdata);
+	kernel_data.end     = virt_to_phys(_end - 1);
+
+	for (i = 0; i < mi->nr_banks; i++) {
+		if (mi->bank[i].size == 0)
+			continue;
+
+		res = alloc_bootmem_low(sizeof(*res));
+		res->name  = "System RAM";
+		res->start = mi->bank[i].start;
+		res->end   = mi->bank[i].start + mi->bank[i].size - 1;
+		res->flags = IORESOURCE_MEM | IORESOURCE_BUSY;
+
+		request_resource(&iomem_resource, res);
+
+		if (kernel_code.start >= res->start &&
+		    kernel_code.end <= res->end)
+			request_resource(res, &kernel_code);
+		if (kernel_data.start >= res->start &&
+		    kernel_data.end <= res->end)
+			request_resource(res, &kernel_data);
+	}
+
+	video_ram.start = PKUNITY_UNIGFX_MMAP_BASE;
+	video_ram.end   = PKUNITY_UNIGFX_MMAP_BASE + PKUNITY_UNIGFX_MMAP_SIZE;
+	request_resource(&iomem_resource, &video_ram);
+}
+
+static void (*init_machine)(void) __initdata;
+
+static int __init customize_machine(void)
+{
+	/* customizes platform devices, or adds new ones */
+	if (init_machine)
+		init_machine();
+	return 0;
+}
+arch_initcall(customize_machine);
+
+void __init setup_arch(char **cmdline_p)
+{
+	char *from = default_command_line;
+
+	setup_processor();
+
+	init_mm.start_code = (unsigned long) _stext;
+	init_mm.end_code   = (unsigned long) _etext;
+	init_mm.end_data   = (unsigned long) _edata;
+	init_mm.brk	   = (unsigned long) _end;
+
+	/* parse_early_param needs a boot_command_line */
+	strlcpy(boot_command_line, from, COMMAND_LINE_SIZE);
+
+	/* populate cmd_line too for later use, preserving boot_command_line */
+	strlcpy(cmd_line, boot_command_line, COMMAND_LINE_SIZE);
+	*cmdline_p = cmd_line;
+
+	parse_early_param();
+
+	uc32_memblock_init(&meminfo);
+
+	paging_init();
+	request_standard_resources(&meminfo);
+
+	cpu_init();
+
+	/*
+	 * Set up various architecture-specific pointers
+	 */
+	init_machine = puv3_core_init;
+
+#ifdef CONFIG_VT
+#if defined(CONFIG_VGA_CONSOLE)
+	conswitchp = &vga_con;
+#elif defined(CONFIG_DUMMY_CONSOLE)
+	conswitchp = &dummy_con;
+#endif
+#endif
+	early_trap_init();
+}
+
+static struct cpu cpuinfo_unicore;
+
+static int __init topology_init(void)
+{
+	int i;
+
+	for_each_possible_cpu(i)
+		register_cpu(&cpuinfo_unicore, i);
+
+	return 0;
+}
+subsys_initcall(topology_init);
+
+#ifdef CONFIG_HAVE_PROC_CPU
+static int __init proc_cpu_init(void)
+{
+	struct proc_dir_entry *res;
+
+	res = proc_mkdir("cpu", NULL);
+	if (!res)
+		return -ENOMEM;
+	return 0;
+}
+fs_initcall(proc_cpu_init);
+#endif
+
+static int c_show(struct seq_file *m, void *v)
+{
+	seq_printf(m, "Processor\t: UniCore-II rev %d (%s)\n",
+		   (int)(uc32_cpuid >> 16) & 15, elf_platform);
+
+	seq_printf(m, "BogoMIPS\t: %lu.%02lu\n",
+		   loops_per_jiffy / (500000/HZ),
+		   (loops_per_jiffy / (5000/HZ)) % 100);
+
+	/* dump out the processor features */
+	seq_puts(m, "Features\t: CMOV UC-F64");
+
+	seq_printf(m, "\nCPU implementer\t: 0x%02x\n", uc32_cpuid >> 24);
+	seq_printf(m, "CPU architecture: 2\n");
+	seq_printf(m, "CPU revision\t: %d\n", (uc32_cpuid >> 16) & 15);
+
+	seq_printf(m, "Cache type\t: write-back\n"
+			"Cache clean\t: cp0 c5 ops\n"
+			"Cache lockdown\t: not support\n"
+			"Cache format\t: Harvard\n");
+
+	seq_puts(m, "\n");
+
+	seq_printf(m, "Hardware\t: PKUnity v3\n");
+
+	return 0;
+}
+
+static void *c_start(struct seq_file *m, loff_t *pos)
+{
+	return *pos < 1 ? (void *)1 : NULL;
+}
+
+static void *c_next(struct seq_file *m, void *v, loff_t *pos)
+{
+	++*pos;
+	return NULL;
+}
+
+static void c_stop(struct seq_file *m, void *v)
+{
+}
+
+const struct seq_operations cpuinfo_op = {
+	.start	= c_start,
+	.next	= c_next,
+	.stop	= c_stop,
+	.show	= c_show
+};
diff --git a/arch/unicore32/kernel/setup.h b/arch/unicore32/kernel/setup.h
new file mode 100644
index 000000000000..dcd1306eb5c6
--- /dev/null
+++ b/arch/unicore32/kernel/setup.h
@@ -0,0 +1,30 @@
+/*
+ * linux/arch/unicore32/kernel/setup.h
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __UNICORE_KERNEL_SETUP_H__
+#define __UNICORE_KERNEL_SETUP_H__
+
+extern void paging_init(void);
+extern void puv3_core_init(void);
+
+extern void puv3_ps2_init(void);
+extern void pci_puv3_preinit(void);
+extern void __init puv3_init_gpio(void);
+
+extern void setup_mm_for_reboot(char mode);
+
+extern char __stubs_start[], __stubs_end[];
+extern char __vectors_start[], __vectors_end[];
+
+extern void kernel_thread_helper(void);
+
+extern void __init early_signal_init(void);
+#endif
diff --git a/arch/unicore32/kernel/signal.c b/arch/unicore32/kernel/signal.c
new file mode 100644
index 000000000000..b163fca56789
--- /dev/null
+++ b/arch/unicore32/kernel/signal.c
@@ -0,0 +1,494 @@
+/*
+ * linux/arch/unicore32/kernel/signal.c
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/errno.h>
+#include <linux/signal.h>
+#include <linux/personality.h>
+#include <linux/freezer.h>
+#include <linux/uaccess.h>
+#include <linux/tracehook.h>
+#include <linux/elf.h>
+#include <linux/unistd.h>
+
+#include <asm/cacheflush.h>
+#include <asm/ucontext.h>
+
+#define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
+
+/*
+ * For UniCore syscalls, we encode the syscall number into the instruction.
+ */
+#define SWI_SYS_SIGRETURN	(0xff000000) /* error number for new abi */
+#define SWI_SYS_RT_SIGRETURN	(0xff000000 | (__NR_rt_sigreturn))
+#define SWI_SYS_RESTART		(0xff000000 | (__NR_restart_syscall))
+
+#define KERN_SIGRETURN_CODE	(KUSER_VECPAGE_BASE + 0x00000500)
+#define KERN_RESTART_CODE	(KERN_SIGRETURN_CODE + sizeof(sigreturn_codes))
+
+const unsigned long sigreturn_codes[3] = {
+	SWI_SYS_SIGRETURN, SWI_SYS_RT_SIGRETURN,
+};
+
+const unsigned long syscall_restart_code[2] = {
+	SWI_SYS_RESTART,	/* swi	__NR_restart_syscall */
+	0x69efc004,		/* ldr	pc, [sp], #4 */
+};
+
+/*
+ * Do a signal return; undo the signal stack.  These are aligned to 64-bit.
+ */
+struct sigframe {
+	struct ucontext uc;
+	unsigned long retcode[2];
+};
+
+struct rt_sigframe {
+	struct siginfo info;
+	struct sigframe sig;
+};
+
+static int restore_sigframe(struct pt_regs *regs, struct sigframe __user *sf)
+{
+	sigset_t set;
+	int err;
+
+	err = __copy_from_user(&set, &sf->uc.uc_sigmask, sizeof(set));
+	if (err == 0) {
+		sigdelsetmask(&set, ~_BLOCKABLE);
+		spin_lock_irq(&current->sighand->siglock);
+		current->blocked = set;
+		recalc_sigpending();
+		spin_unlock_irq(&current->sighand->siglock);
+	}
+
+	err |= __get_user(regs->UCreg_00, &sf->uc.uc_mcontext.regs.UCreg_00);
+	err |= __get_user(regs->UCreg_01, &sf->uc.uc_mcontext.regs.UCreg_01);
+	err |= __get_user(regs->UCreg_02, &sf->uc.uc_mcontext.regs.UCreg_02);
+	err |= __get_user(regs->UCreg_03, &sf->uc.uc_mcontext.regs.UCreg_03);
+	err |= __get_user(regs->UCreg_04, &sf->uc.uc_mcontext.regs.UCreg_04);
+	err |= __get_user(regs->UCreg_05, &sf->uc.uc_mcontext.regs.UCreg_05);
+	err |= __get_user(regs->UCreg_06, &sf->uc.uc_mcontext.regs.UCreg_06);
+	err |= __get_user(regs->UCreg_07, &sf->uc.uc_mcontext.regs.UCreg_07);
+	err |= __get_user(regs->UCreg_08, &sf->uc.uc_mcontext.regs.UCreg_08);
+	err |= __get_user(regs->UCreg_09, &sf->uc.uc_mcontext.regs.UCreg_09);
+	err |= __get_user(regs->UCreg_10, &sf->uc.uc_mcontext.regs.UCreg_10);
+	err |= __get_user(regs->UCreg_11, &sf->uc.uc_mcontext.regs.UCreg_11);
+	err |= __get_user(regs->UCreg_12, &sf->uc.uc_mcontext.regs.UCreg_12);
+	err |= __get_user(regs->UCreg_13, &sf->uc.uc_mcontext.regs.UCreg_13);
+	err |= __get_user(regs->UCreg_14, &sf->uc.uc_mcontext.regs.UCreg_14);
+	err |= __get_user(regs->UCreg_15, &sf->uc.uc_mcontext.regs.UCreg_15);
+	err |= __get_user(regs->UCreg_16, &sf->uc.uc_mcontext.regs.UCreg_16);
+	err |= __get_user(regs->UCreg_17, &sf->uc.uc_mcontext.regs.UCreg_17);
+	err |= __get_user(regs->UCreg_18, &sf->uc.uc_mcontext.regs.UCreg_18);
+	err |= __get_user(regs->UCreg_19, &sf->uc.uc_mcontext.regs.UCreg_19);
+	err |= __get_user(regs->UCreg_20, &sf->uc.uc_mcontext.regs.UCreg_20);
+	err |= __get_user(regs->UCreg_21, &sf->uc.uc_mcontext.regs.UCreg_21);
+	err |= __get_user(regs->UCreg_22, &sf->uc.uc_mcontext.regs.UCreg_22);
+	err |= __get_user(regs->UCreg_23, &sf->uc.uc_mcontext.regs.UCreg_23);
+	err |= __get_user(regs->UCreg_24, &sf->uc.uc_mcontext.regs.UCreg_24);
+	err |= __get_user(regs->UCreg_25, &sf->uc.uc_mcontext.regs.UCreg_25);
+	err |= __get_user(regs->UCreg_26, &sf->uc.uc_mcontext.regs.UCreg_26);
+	err |= __get_user(regs->UCreg_fp, &sf->uc.uc_mcontext.regs.UCreg_fp);
+	err |= __get_user(regs->UCreg_ip, &sf->uc.uc_mcontext.regs.UCreg_ip);
+	err |= __get_user(regs->UCreg_sp, &sf->uc.uc_mcontext.regs.UCreg_sp);
+	err |= __get_user(regs->UCreg_lr, &sf->uc.uc_mcontext.regs.UCreg_lr);
+	err |= __get_user(regs->UCreg_pc, &sf->uc.uc_mcontext.regs.UCreg_pc);
+	err |= __get_user(regs->UCreg_asr, &sf->uc.uc_mcontext.regs.UCreg_asr);
+
+	err |= !valid_user_regs(regs);
+
+	return err;
+}
+
+asmlinkage int __sys_rt_sigreturn(struct pt_regs *regs)
+{
+	struct rt_sigframe __user *frame;
+
+	/* Always make any pending restarted system calls return -EINTR */
+	current_thread_info()->restart_block.fn = do_no_restart_syscall;
+
+	/*
+	 * Since we stacked the signal on a 64-bit boundary,
+	 * then 'sp' should be word aligned here.  If it's
+	 * not, then the user is trying to mess with us.
+	 */
+	if (regs->UCreg_sp & 7)
+		goto badframe;
+
+	frame = (struct rt_sigframe __user *)regs->UCreg_sp;
+
+	if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
+		goto badframe;
+
+	if (restore_sigframe(regs, &frame->sig))
+		goto badframe;
+
+	if (do_sigaltstack(&frame->sig.uc.uc_stack, NULL, regs->UCreg_sp)
+			== -EFAULT)
+		goto badframe;
+
+	return regs->UCreg_00;
+
+badframe:
+	force_sig(SIGSEGV, current);
+	return 0;
+}
+
+static int setup_sigframe(struct sigframe __user *sf, struct pt_regs *regs,
+		sigset_t *set)
+{
+	int err = 0;
+
+	err |= __put_user(regs->UCreg_00, &sf->uc.uc_mcontext.regs.UCreg_00);
+	err |= __put_user(regs->UCreg_01, &sf->uc.uc_mcontext.regs.UCreg_01);
+	err |= __put_user(regs->UCreg_02, &sf->uc.uc_mcontext.regs.UCreg_02);
+	err |= __put_user(regs->UCreg_03, &sf->uc.uc_mcontext.regs.UCreg_03);
+	err |= __put_user(regs->UCreg_04, &sf->uc.uc_mcontext.regs.UCreg_04);
+	err |= __put_user(regs->UCreg_05, &sf->uc.uc_mcontext.regs.UCreg_05);
+	err |= __put_user(regs->UCreg_06, &sf->uc.uc_mcontext.regs.UCreg_06);
+	err |= __put_user(regs->UCreg_07, &sf->uc.uc_mcontext.regs.UCreg_07);
+	err |= __put_user(regs->UCreg_08, &sf->uc.uc_mcontext.regs.UCreg_08);
+	err |= __put_user(regs->UCreg_09, &sf->uc.uc_mcontext.regs.UCreg_09);
+	err |= __put_user(regs->UCreg_10, &sf->uc.uc_mcontext.regs.UCreg_10);
+	err |= __put_user(regs->UCreg_11, &sf->uc.uc_mcontext.regs.UCreg_11);
+	err |= __put_user(regs->UCreg_12, &sf->uc.uc_mcontext.regs.UCreg_12);
+	err |= __put_user(regs->UCreg_13, &sf->uc.uc_mcontext.regs.UCreg_13);
+	err |= __put_user(regs->UCreg_14, &sf->uc.uc_mcontext.regs.UCreg_14);
+	err |= __put_user(regs->UCreg_15, &sf->uc.uc_mcontext.regs.UCreg_15);
+	err |= __put_user(regs->UCreg_16, &sf->uc.uc_mcontext.regs.UCreg_16);
+	err |= __put_user(regs->UCreg_17, &sf->uc.uc_mcontext.regs.UCreg_17);
+	err |= __put_user(regs->UCreg_18, &sf->uc.uc_mcontext.regs.UCreg_18);
+	err |= __put_user(regs->UCreg_19, &sf->uc.uc_mcontext.regs.UCreg_19);
+	err |= __put_user(regs->UCreg_20, &sf->uc.uc_mcontext.regs.UCreg_20);
+	err |= __put_user(regs->UCreg_21, &sf->uc.uc_mcontext.regs.UCreg_21);
+	err |= __put_user(regs->UCreg_22, &sf->uc.uc_mcontext.regs.UCreg_22);
+	err |= __put_user(regs->UCreg_23, &sf->uc.uc_mcontext.regs.UCreg_23);
+	err |= __put_user(regs->UCreg_24, &sf->uc.uc_mcontext.regs.UCreg_24);
+	err |= __put_user(regs->UCreg_25, &sf->uc.uc_mcontext.regs.UCreg_25);
+	err |= __put_user(regs->UCreg_26, &sf->uc.uc_mcontext.regs.UCreg_26);
+	err |= __put_user(regs->UCreg_fp, &sf->uc.uc_mcontext.regs.UCreg_fp);
+	err |= __put_user(regs->UCreg_ip, &sf->uc.uc_mcontext.regs.UCreg_ip);
+	err |= __put_user(regs->UCreg_sp, &sf->uc.uc_mcontext.regs.UCreg_sp);
+	err |= __put_user(regs->UCreg_lr, &sf->uc.uc_mcontext.regs.UCreg_lr);
+	err |= __put_user(regs->UCreg_pc, &sf->uc.uc_mcontext.regs.UCreg_pc);
+	err |= __put_user(regs->UCreg_asr, &sf->uc.uc_mcontext.regs.UCreg_asr);
+
+	err |= __put_user(current->thread.trap_no,
+			&sf->uc.uc_mcontext.trap_no);
+	err |= __put_user(current->thread.error_code,
+			&sf->uc.uc_mcontext.error_code);
+	err |= __put_user(current->thread.address,
+			&sf->uc.uc_mcontext.fault_address);
+	err |= __put_user(set->sig[0], &sf->uc.uc_mcontext.oldmask);
+
+	err |= __copy_to_user(&sf->uc.uc_sigmask, set, sizeof(*set));
+
+	return err;
+}
+
+static inline void __user *get_sigframe(struct k_sigaction *ka,
+		struct pt_regs *regs, int framesize)
+{
+	unsigned long sp = regs->UCreg_sp;
+	void __user *frame;
+
+	/*
+	 * This is the X/Open sanctioned signal stack switching.
+	 */
+	if ((ka->sa.sa_flags & SA_ONSTACK) && !sas_ss_flags(sp))
+		sp = current->sas_ss_sp + current->sas_ss_size;
+
+	/*
+	 * ATPCS B01 mandates 8-byte alignment
+	 */
+	frame = (void __user *)((sp - framesize) & ~7);
+
+	/*
+	 * Check that we can actually write to the signal frame.
+	 */
+	if (!access_ok(VERIFY_WRITE, frame, framesize))
+		frame = NULL;
+
+	return frame;
+}
+
+static int setup_return(struct pt_regs *regs, struct k_sigaction *ka,
+	     unsigned long __user *rc, void __user *frame, int usig)
+{
+	unsigned long handler = (unsigned long)ka->sa.sa_handler;
+	unsigned long retcode;
+	unsigned long asr = regs->UCreg_asr & ~PSR_f;
+
+	unsigned int idx = 0;
+
+	if (ka->sa.sa_flags & SA_SIGINFO)
+		idx += 1;
+
+	if (__put_user(sigreturn_codes[idx],   rc) ||
+	    __put_user(sigreturn_codes[idx+1], rc+1))
+		return 1;
+
+	retcode = KERN_SIGRETURN_CODE + (idx << 2);
+
+	regs->UCreg_00 = usig;
+	regs->UCreg_sp = (unsigned long)frame;
+	regs->UCreg_lr = retcode;
+	regs->UCreg_pc = handler;
+	regs->UCreg_asr = asr;
+
+	return 0;
+}
+
+static int setup_frame(int usig, struct k_sigaction *ka,
+		sigset_t *set, struct pt_regs *regs)
+{
+	struct sigframe __user *frame = get_sigframe(ka, regs, sizeof(*frame));
+	int err = 0;
+
+	if (!frame)
+		return 1;
+
+	/*
+	 * Set uc.uc_flags to a value which sc.trap_no would never have.
+	 */
+	err |= __put_user(0x5ac3c35a, &frame->uc.uc_flags);
+
+	err |= setup_sigframe(frame, regs, set);
+	if (err == 0)
+		err |= setup_return(regs, ka, frame->retcode, frame, usig);
+
+	return err;
+}
+
+static int setup_rt_frame(int usig, struct k_sigaction *ka, siginfo_t *info,
+	       sigset_t *set, struct pt_regs *regs)
+{
+	struct rt_sigframe __user *frame =
+			get_sigframe(ka, regs, sizeof(*frame));
+	stack_t stack;
+	int err = 0;
+
+	if (!frame)
+		return 1;
+
+	err |= copy_siginfo_to_user(&frame->info, info);
+
+	err |= __put_user(0, &frame->sig.uc.uc_flags);
+	err |= __put_user(NULL, &frame->sig.uc.uc_link);
+
+	memset(&stack, 0, sizeof(stack));
+	stack.ss_sp = (void __user *)current->sas_ss_sp;
+	stack.ss_flags = sas_ss_flags(regs->UCreg_sp);
+	stack.ss_size = current->sas_ss_size;
+	err |= __copy_to_user(&frame->sig.uc.uc_stack, &stack, sizeof(stack));
+
+	err |= setup_sigframe(&frame->sig, regs, set);
+	if (err == 0)
+		err |= setup_return(regs, ka, frame->sig.retcode, frame, usig);
+
+	if (err == 0) {
+		/*
+		 * For realtime signals we must also set the second and third
+		 * arguments for the signal handler.
+		 */
+		regs->UCreg_01 = (unsigned long)&frame->info;
+		regs->UCreg_02 = (unsigned long)&frame->sig.uc;
+	}
+
+	return err;
+}
+
+static inline void setup_syscall_restart(struct pt_regs *regs)
+{
+	regs->UCreg_00 = regs->UCreg_ORIG_00;
+	regs->UCreg_pc -= 4;
+}
+
+/*
+ * OK, we're invoking a handler
+ */
+static int handle_signal(unsigned long sig, struct k_sigaction *ka,
+	      siginfo_t *info, sigset_t *oldset,
+	      struct pt_regs *regs, int syscall)
+{
+	struct thread_info *thread = current_thread_info();
+	struct task_struct *tsk = current;
+	int usig = sig;
+	int ret;
+
+	/*
+	 * If we were from a system call, check for system call restarting...
+	 */
+	if (syscall) {
+		switch (regs->UCreg_00) {
+		case -ERESTART_RESTARTBLOCK:
+		case -ERESTARTNOHAND:
+			regs->UCreg_00 = -EINTR;
+			break;
+		case -ERESTARTSYS:
+			if (!(ka->sa.sa_flags & SA_RESTART)) {
+				regs->UCreg_00 = -EINTR;
+				break;
+			}
+			/* fallthrough */
+		case -ERESTARTNOINTR:
+			setup_syscall_restart(regs);
+		}
+	}
+
+	/*
+	 * translate the signal
+	 */
+	if (usig < 32 && thread->exec_domain
+			&& thread->exec_domain->signal_invmap)
+		usig = thread->exec_domain->signal_invmap[usig];
+
+	/*
+	 * Set up the stack frame
+	 */
+	if (ka->sa.sa_flags & SA_SIGINFO)
+		ret = setup_rt_frame(usig, ka, info, oldset, regs);
+	else
+		ret = setup_frame(usig, ka, oldset, regs);
+
+	/*
+	 * Check that the resulting registers are actually sane.
+	 */
+	ret |= !valid_user_regs(regs);
+
+	if (ret != 0) {
+		force_sigsegv(sig, tsk);
+		return ret;
+	}
+
+	/*
+	 * Block the signal if we were successful.
+	 */
+	spin_lock_irq(&tsk->sighand->siglock);
+	sigorsets(&tsk->blocked, &tsk->blocked,
+		  &ka->sa.sa_mask);
+	if (!(ka->sa.sa_flags & SA_NODEFER))
+		sigaddset(&tsk->blocked, sig);
+	recalc_sigpending();
+	spin_unlock_irq(&tsk->sighand->siglock);
+
+	return 0;
+}
+
+/*
+ * Note that 'init' is a special process: it doesn't get signals it doesn't
+ * want to handle. Thus you cannot kill init even with a SIGKILL even by
+ * mistake.
+ *
+ * Note that we go through the signals twice: once to check the signals that
+ * the kernel can handle, and then we build all the user-level signal handling
+ * stack-frames in one go after that.
+ */
+static void do_signal(struct pt_regs *regs, int syscall)
+{
+	struct k_sigaction ka;
+	siginfo_t info;
+	int signr;
+
+	/*
+	 * We want the common case to go fast, which
+	 * is why we may in certain cases get here from
+	 * kernel mode. Just return without doing anything
+	 * if so.
+	 */
+	if (!user_mode(regs))
+		return;
+
+	if (try_to_freeze())
+		goto no_signal;
+
+	signr = get_signal_to_deliver(&info, &ka, regs, NULL);
+	if (signr > 0) {
+		sigset_t *oldset;
+
+		if (test_thread_flag(TIF_RESTORE_SIGMASK))
+			oldset = &current->saved_sigmask;
+		else
+			oldset = &current->blocked;
+		if (handle_signal(signr, &ka, &info, oldset, regs, syscall)
+				== 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);
+		}
+		return;
+	}
+
+ no_signal:
+	/*
+	 * No signal to deliver to the process - restart the syscall.
+	 */
+	if (syscall) {
+		if (regs->UCreg_00 == -ERESTART_RESTARTBLOCK) {
+				u32 __user *usp;
+
+				regs->UCreg_sp -= 4;
+				usp = (u32 __user *)regs->UCreg_sp;
+
+				if (put_user(regs->UCreg_pc, usp) == 0) {
+					regs->UCreg_pc = KERN_RESTART_CODE;
+				} else {
+					regs->UCreg_sp += 4;
+					force_sigsegv(0, current);
+				}
+		}
+		if (regs->UCreg_00 == -ERESTARTNOHAND ||
+		    regs->UCreg_00 == -ERESTARTSYS ||
+		    regs->UCreg_00 == -ERESTARTNOINTR) {
+			setup_syscall_restart(regs);
+		}
+
+		/* 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 thread_flags, int syscall)
+{
+	if (thread_flags & _TIF_SIGPENDING)
+		do_signal(regs, syscall);
+
+	if (thread_flags & _TIF_NOTIFY_RESUME) {
+		clear_thread_flag(TIF_NOTIFY_RESUME);
+		tracehook_notify_resume(regs);
+		if (current->replacement_session_keyring)
+			key_replace_session_keyring();
+	}
+}
+
+/*
+ * Copy signal return handlers into the vector page, and
+ * set sigreturn to be a pointer to these.
+ */
+void __init early_signal_init(void)
+{
+	memcpy((void *)kuser_vecpage_to_vectors(KERN_SIGRETURN_CODE),
+			sigreturn_codes, sizeof(sigreturn_codes));
+	memcpy((void *)kuser_vecpage_to_vectors(KERN_RESTART_CODE),
+			syscall_restart_code, sizeof(syscall_restart_code));
+	/* Need not to flush icache, since early_trap_init will do it last. */
+}
diff --git a/arch/unicore32/kernel/sleep.S b/arch/unicore32/kernel/sleep.S
new file mode 100644
index 000000000000..607a104aec59
--- /dev/null
+++ b/arch/unicore32/kernel/sleep.S
@@ -0,0 +1,202 @@
+/*
+ * linux/arch/unicore32/kernel/sleep.S
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ *	Maintained by GUAN Xue-tao <gxt@mprc.pku.edu.cn>
+ *	Copyright (C) 2001-2010 Guan Xuetao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+#include <mach/hardware.h>
+
+		.text
+
+pkunity_cpu_save_cp:
+
+	@ get coprocessor registers
+
+	movc	r3, p0.c7, #0			@ PID
+	movc	r4, p0.c2, #0			@ translation table base addr
+	movc	r5, p0.c1, #0			@ control reg
+
+
+	@ store them plus current virtual stack ptr on stack
+	mov	r6, sp
+	stm.w	(r3 - r6), [sp-]
+
+	mov	pc, lr
+
+pkunity_cpu_save_sp:
+	@ preserve phys address of stack
+	mov	r0, sp
+	stw.w	lr, [sp+], #-4
+	b.l	sleep_phys_sp
+	ldw	r1, =sleep_save_sp
+	stw	r0, [r1]
+	ldw.w	pc, [sp]+, #4
+
+/*
+ * puv3_cpu_suspend()
+ *
+ * Forces CPU into sleep state.
+ *
+ * r0 = value for PWRMODE M field for desired sleep state
+ */
+
+ENTRY(puv3_cpu_suspend)
+	stm.w	(r16 - r27, lr), [sp-]		@ save registers on stack
+	stm.w	(r4 - r15), [sp-]		@ save registers on stack
+
+#ifdef	CONFIG_UNICORE_FPU_F64
+	sfm.w	(f0  - f7 ), [sp-]
+	sfm.w	(f8  - f15), [sp-]
+	sfm.w	(f16 - f23), [sp-]
+	sfm.w	(f24 - f31), [sp-]
+	cff	r4, s31
+	stm.w	(r4), [sp-]
+#endif
+	b.l	pkunity_cpu_save_cp
+
+	b.l	pkunity_cpu_save_sp
+
+	@ clean data cache
+	mov	r1, #0
+	movc	p0.c5, r1, #14
+	nop
+	nop
+	nop
+	nop
+
+
+
+	@ DDR2 BaseAddr
+	ldw	r0, =(PKUNITY_DDR2CTRL_BASE)
+
+	@ PM BaseAddr
+	ldw	r1, =(PKUNITY_PM_BASE)
+
+	@ set PLL_SYS_CFG reg, 275
+	movl	r6, #0x00002401
+	stw	r6, [r1+], #0x18
+	@ set PLL_DDR_CFG reg, 66MHz
+	movl	r6, #0x00100c00
+	stw	r6, [r1+], #0x1c
+
+	@ set wake up source
+	movl	r8, #0x800001ff		@ epip4d
+	stw	r8, [r1+], #0xc
+
+	@ set PGSR
+	movl	r5, #0x40000
+	stw	r5, [r1+], #0x10
+
+	@ prepare DDR2 refresh settings
+	ldw	r5, [r0+], #0x24
+	or	r5, r5, #0x00000001
+
+	@ prepare PMCR for PLL changing
+	movl	r6, #0xc
+
+	@ prepare for closing PLL
+	movl	r7, #0x1
+
+	@ prepare sleep mode
+	mov	r8, #0x1
+
+@	movl	r0, 0x11111111
+@	put_word_ocd r0
+	b	pkunity_cpu_do_suspend
+
+	.ltorg
+	.align	5
+pkunity_cpu_do_suspend:
+	b	101f
+	@ put DDR2 into self-refresh
+100:	stw	r5, [r0+], #0x24
+	@ change PLL
+	stw	r6, [r1]
+	b	1f
+
+	.ltorg
+	.align	5
+101:	b	102f
+	@ wait for PLL changing complete
+1:	ldw	r6, [r1+], #0x44
+	csub.a	r6, #0x1
+	bne	1b
+	b	2f
+
+	.ltorg
+	.align	5
+102:	b	100b
+	@ close PLL
+2:	stw	r7, [r1+], #0x4
+	@ enter sleep mode
+	stw	r8, [r1]
+3:	b	3b
+
+
+
+
+/*
+ * puv3_cpu_resume()
+ *
+ * entry point from bootloader into kernel during resume
+ *
+ * Note: Yes, part of the following code is located into the .data section.
+ *       This is to allow sleep_save_sp to be accessed with a relative load
+ *       while we can't rely on any MMU translation.  We could have put
+ *       sleep_save_sp in the .text section as well, but some setups might
+ *       insist on it to be truly read-only.
+ */
+
+	.data
+	.align 5
+ENTRY(puv3_cpu_resume)
+@	movl	r0, 0x20202020
+@	put_word_ocd r0
+
+	ldw	r0, sleep_save_sp		@ stack phys addr
+	ldw	r2, =resume_after_mmu		@ its absolute virtual address
+	ldm	(r3 - r6), [r0]+		@ CP regs + virt stack ptr
+	mov	sp, r6				@ CP regs + virt stack ptr
+
+	mov	r1, #0
+	movc	p0.c6, r1, #6			@ invalidate I & D TLBs
+	movc	p0.c5, r1, #28			@ invalidate I & D caches, BTB
+
+	movc	p0.c7, r3, #0			@ PID
+	movc	p0.c2, r4, #0			@ translation table base addr
+	movc	p0.c1, r5, #0			@ control reg, turn on mmu
+	nop
+	jump	r2
+	nop
+	nop
+	nop
+	nop
+	nop
+
+sleep_save_sp:
+	.word	0				@ preserve stack phys ptr here
+
+	.text
+resume_after_mmu:
+@	movl	r0, 0x30303030
+@	put_word_ocd r0
+
+#ifdef	CONFIG_UNICORE_FPU_F64
+	lfm.w	(f0  - f7 ), [sp]+
+	lfm.w	(f8  - f15), [sp]+
+	lfm.w	(f16 - f23), [sp]+
+	lfm.w	(f24 - f31), [sp]+
+	ldm.w	(r4), [sp]+
+	ctf	r4, s31
+#endif
+	ldm.w	(r4 - r15), [sp]+		@ restore registers from stack
+	ldm.w	(r16 - r27, pc), [sp]+		@ return to caller
diff --git a/arch/unicore32/kernel/stacktrace.c b/arch/unicore32/kernel/stacktrace.c
new file mode 100644
index 000000000000..b34030bdabe3
--- /dev/null
+++ b/arch/unicore32/kernel/stacktrace.c
@@ -0,0 +1,131 @@
+/*
+ * linux/arch/unicore32/kernel/stacktrace.c
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/stacktrace.h>
+
+#include <asm/stacktrace.h>
+
+#if defined(CONFIG_FRAME_POINTER)
+/*
+ * Unwind the current stack frame and store the new register values in the
+ * structure passed as argument. Unwinding is equivalent to a function return,
+ * hence the new PC value rather than LR should be used for backtrace.
+ *
+ * With framepointer enabled, a simple function prologue looks like this:
+ *	mov	ip, sp
+ *	stmdb	sp!, {fp, ip, lr, pc}
+ *	sub	fp, ip, #4
+ *
+ * A simple function epilogue looks like this:
+ *	ldm	sp, {fp, sp, pc}
+ *
+ * Note that with framepointer enabled, even the leaf functions have the same
+ * prologue and epilogue, therefore we can ignore the LR value in this case.
+ */
+int notrace unwind_frame(struct stackframe *frame)
+{
+	unsigned long high, low;
+	unsigned long fp = frame->fp;
+
+	/* only go to a higher address on the stack */
+	low = frame->sp;
+	high = ALIGN(low, THREAD_SIZE);
+
+	/* check current frame pointer is within bounds */
+	if (fp < (low + 12) || fp + 4 >= high)
+		return -EINVAL;
+
+	/* restore the registers from the stack frame */
+	frame->fp = *(unsigned long *)(fp - 12);
+	frame->sp = *(unsigned long *)(fp - 8);
+	frame->pc = *(unsigned long *)(fp - 4);
+
+	return 0;
+}
+#endif
+
+void notrace walk_stackframe(struct stackframe *frame,
+		     int (*fn)(struct stackframe *, void *), void *data)
+{
+	while (1) {
+		int ret;
+
+		if (fn(frame, data))
+			break;
+		ret = unwind_frame(frame);
+		if (ret < 0)
+			break;
+	}
+}
+EXPORT_SYMBOL(walk_stackframe);
+
+#ifdef CONFIG_STACKTRACE
+struct stack_trace_data {
+	struct stack_trace *trace;
+	unsigned int no_sched_functions;
+	unsigned int skip;
+};
+
+static int save_trace(struct stackframe *frame, void *d)
+{
+	struct stack_trace_data *data = d;
+	struct stack_trace *trace = data->trace;
+	unsigned long addr = frame->pc;
+
+	if (data->no_sched_functions && in_sched_functions(addr))
+		return 0;
+	if (data->skip) {
+		data->skip--;
+		return 0;
+	}
+
+	trace->entries[trace->nr_entries++] = addr;
+
+	return trace->nr_entries >= trace->max_entries;
+}
+
+void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace)
+{
+	struct stack_trace_data data;
+	struct stackframe frame;
+
+	data.trace = trace;
+	data.skip = trace->skip;
+
+	if (tsk != current) {
+		data.no_sched_functions = 1;
+		frame.fp = thread_saved_fp(tsk);
+		frame.sp = thread_saved_sp(tsk);
+		frame.lr = 0;		/* recovered from the stack */
+		frame.pc = thread_saved_pc(tsk);
+	} else {
+		register unsigned long current_sp asm("sp");
+
+		data.no_sched_functions = 0;
+		frame.fp = (unsigned long)__builtin_frame_address(0);
+		frame.sp = current_sp;
+		frame.lr = (unsigned long)__builtin_return_address(0);
+		frame.pc = (unsigned long)save_stack_trace_tsk;
+	}
+
+	walk_stackframe(&frame, save_trace, &data);
+	if (trace->nr_entries < trace->max_entries)
+		trace->entries[trace->nr_entries++] = ULONG_MAX;
+}
+
+void save_stack_trace(struct stack_trace *trace)
+{
+	save_stack_trace_tsk(current, trace);
+}
+EXPORT_SYMBOL_GPL(save_stack_trace);
+#endif
diff --git a/arch/unicore32/kernel/sys.c b/arch/unicore32/kernel/sys.c
new file mode 100644
index 000000000000..3afe60a39ac9
--- /dev/null
+++ b/arch/unicore32/kernel/sys.c
@@ -0,0 +1,126 @@
+/*
+ * linux/arch/unicore32/kernel/sys.c
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/module.h>
+#include <linux/errno.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <linux/mm.h>
+#include <linux/sem.h>
+#include <linux/msg.h>
+#include <linux/shm.h>
+#include <linux/stat.h>
+#include <linux/syscalls.h>
+#include <linux/mman.h>
+#include <linux/fs.h>
+#include <linux/file.h>
+#include <linux/ipc.h>
+#include <linux/uaccess.h>
+
+#include <asm/syscalls.h>
+#include <asm/cacheflush.h>
+
+/* Clone a task - this clones the calling program thread.
+ * This is called indirectly via a small wrapper
+ */
+asmlinkage long __sys_clone(unsigned long clone_flags, unsigned long newsp,
+			 void __user *parent_tid, void __user *child_tid,
+			 struct pt_regs *regs)
+{
+	if (!newsp)
+		newsp = regs->UCreg_sp;
+
+	return do_fork(clone_flags, newsp, regs, 0,
+			parent_tid, child_tid);
+}
+
+/* sys_execve() executes a new program.
+ * This is called indirectly via a small wrapper
+ */
+asmlinkage long __sys_execve(const char __user *filename,
+			  const char __user *const __user *argv,
+			  const char __user *const __user *envp,
+			  struct pt_regs *regs)
+{
+	int error;
+	char *fn;
+
+	fn = getname(filename);
+	error = PTR_ERR(fn);
+	if (IS_ERR(fn))
+		goto out;
+	error = do_execve(fn, argv, envp, regs);
+	putname(fn);
+out:
+	return error;
+}
+
+int kernel_execve(const char *filename,
+		  const char *const argv[],
+		  const char *const envp[])
+{
+	struct pt_regs regs;
+	int ret;
+
+	memset(&regs, 0, sizeof(struct pt_regs));
+	ret = do_execve(filename,
+			(const char __user *const __user *)argv,
+			(const char __user *const __user *)envp, &regs);
+	if (ret < 0)
+		goto out;
+
+	/*
+	 * Save argc to the register structure for userspace.
+	 */
+	regs.UCreg_00 = ret;
+
+	/*
+	 * We were successful.  We won't be returning to our caller, but
+	 * instead to user space by manipulating the kernel stack.
+	 */
+	asm("add	r0, %0, %1\n\t"
+		"mov	r1, %2\n\t"
+		"mov	r2, %3\n\t"
+		"mov	r22, #0\n\t"	/* not a syscall */
+		"mov	r23, %0\n\t"	/* thread structure */
+		"b.l	memmove\n\t"	/* copy regs to top of stack */
+		"mov	sp, r0\n\t"	/* reposition stack pointer */
+		"b	ret_to_user"
+		:
+		: "r" (current_thread_info()),
+		  "Ir" (THREAD_START_SP - sizeof(regs)),
+		  "r" (&regs),
+		  "Ir" (sizeof(regs))
+		: "r0", "r1", "r2", "r3", "ip", "lr", "memory");
+
+ out:
+	return ret;
+}
+EXPORT_SYMBOL(kernel_execve);
+
+/* Note: used by the compat code even in 64-bit Linux. */
+SYSCALL_DEFINE6(mmap2, unsigned long, addr, unsigned long, len,
+		unsigned long, prot, unsigned long, flags,
+		unsigned long, fd, unsigned long, off_4k)
+{
+	return sys_mmap_pgoff(addr, len, prot, flags, fd,
+			      off_4k);
+}
+
+/* Provide the actual syscall number to call mapping. */
+#undef __SYSCALL
+#define __SYSCALL(nr, call)	[nr] = (call),
+
+/* Note that we don't include <linux/unistd.h> but <asm/unistd.h> */
+void *sys_call_table[__NR_syscalls] = {
+	[0 ... __NR_syscalls-1] = sys_ni_syscall,
+#include <asm/unistd.h>
+};
diff --git a/arch/unicore32/kernel/time.c b/arch/unicore32/kernel/time.c
new file mode 100644
index 000000000000..080710c09241
--- /dev/null
+++ b/arch/unicore32/kernel/time.c
@@ -0,0 +1,143 @@
+/*
+ * linux/arch/unicore32/kernel/time.c
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ *	Maintained by GUAN Xue-tao <gxt@mprc.pku.edu.cn>
+ *	Copyright (C) 2001-2010 Guan Xuetao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/init.h>
+#include <linux/errno.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/timex.h>
+#include <linux/clockchips.h>
+
+#include <mach/hardware.h>
+
+#define MIN_OSCR_DELTA 2
+
+static irqreturn_t puv3_ost0_interrupt(int irq, void *dev_id)
+{
+	struct clock_event_device *c = dev_id;
+
+	/* Disarm the compare/match, signal the event. */
+	writel(readl(OST_OIER) & ~OST_OIER_E0, OST_OIER);
+	writel(readl(OST_OSSR) & ~OST_OSSR_M0, OST_OSSR);
+	c->event_handler(c);
+
+	return IRQ_HANDLED;
+}
+
+static int
+puv3_osmr0_set_next_event(unsigned long delta, struct clock_event_device *c)
+{
+	unsigned long next, oscr;
+
+	writel(readl(OST_OIER) | OST_OIER_E0, OST_OIER);
+	next = readl(OST_OSCR) + delta;
+	writel(next, OST_OSMR0);
+	oscr = readl(OST_OSCR);
+
+	return (signed)(next - oscr) <= MIN_OSCR_DELTA ? -ETIME : 0;
+}
+
+static void
+puv3_osmr0_set_mode(enum clock_event_mode mode, struct clock_event_device *c)
+{
+	switch (mode) {
+	case CLOCK_EVT_MODE_ONESHOT:
+	case CLOCK_EVT_MODE_UNUSED:
+	case CLOCK_EVT_MODE_SHUTDOWN:
+		writel(readl(OST_OIER) & ~OST_OIER_E0, OST_OIER);
+		writel(readl(OST_OSSR) & ~OST_OSSR_M0, OST_OSSR);
+		break;
+
+	case CLOCK_EVT_MODE_RESUME:
+	case CLOCK_EVT_MODE_PERIODIC:
+		break;
+	}
+}
+
+static struct clock_event_device ckevt_puv3_osmr0 = {
+	.name		= "osmr0",
+	.features	= CLOCK_EVT_FEAT_ONESHOT,
+	.rating		= 200,
+	.set_next_event	= puv3_osmr0_set_next_event,
+	.set_mode	= puv3_osmr0_set_mode,
+};
+
+static cycle_t puv3_read_oscr(struct clocksource *cs)
+{
+	return readl(OST_OSCR);
+}
+
+static struct clocksource cksrc_puv3_oscr = {
+	.name		= "oscr",
+	.rating		= 200,
+	.read		= puv3_read_oscr,
+	.mask		= CLOCKSOURCE_MASK(32),
+	.flags		= CLOCK_SOURCE_IS_CONTINUOUS,
+};
+
+static struct irqaction puv3_timer_irq = {
+	.name		= "ost0",
+	.flags		= IRQF_DISABLED | IRQF_TIMER | IRQF_IRQPOLL,
+	.handler	= puv3_ost0_interrupt,
+	.dev_id		= &ckevt_puv3_osmr0,
+};
+
+void __init time_init(void)
+{
+	writel(0, OST_OIER);		/* disable any timer interrupts */
+	writel(0, OST_OSSR);		/* clear status on all timers */
+
+	clockevents_calc_mult_shift(&ckevt_puv3_osmr0, CLOCK_TICK_RATE, 5);
+
+	ckevt_puv3_osmr0.max_delta_ns =
+		clockevent_delta2ns(0x7fffffff, &ckevt_puv3_osmr0);
+	ckevt_puv3_osmr0.min_delta_ns =
+		clockevent_delta2ns(MIN_OSCR_DELTA * 2, &ckevt_puv3_osmr0) + 1;
+	ckevt_puv3_osmr0.cpumask = cpumask_of(0);
+
+	setup_irq(IRQ_TIMER0, &puv3_timer_irq);
+
+	clocksource_register_hz(&cksrc_puv3_oscr, CLOCK_TICK_RATE);
+	clockevents_register_device(&ckevt_puv3_osmr0);
+}
+
+#ifdef CONFIG_PM
+unsigned long osmr[4], oier;
+
+void puv3_timer_suspend(void)
+{
+	osmr[0] = readl(OST_OSMR0);
+	osmr[1] = readl(OST_OSMR1);
+	osmr[2] = readl(OST_OSMR2);
+	osmr[3] = readl(OST_OSMR3);
+	oier = readl(OST_OIER);
+}
+
+void puv3_timer_resume(void)
+{
+	writel(0, OST_OSSR);
+	writel(osmr[0], OST_OSMR0);
+	writel(osmr[1], OST_OSMR1);
+	writel(osmr[2], OST_OSMR2);
+	writel(osmr[3], OST_OSMR3);
+	writel(oier, OST_OIER);
+
+	/*
+	 * OSMR0 is the system timer: make sure OSCR is sufficiently behind
+	 */
+	writel(readl(OST_OSMR0) - LATCH, OST_OSCR);
+}
+#else
+void puv3_timer_suspend(void) { };
+void puv3_timer_resume(void) { };
+#endif
+
diff --git a/arch/unicore32/kernel/traps.c b/arch/unicore32/kernel/traps.c
new file mode 100644
index 000000000000..25abbb101729
--- /dev/null
+++ b/arch/unicore32/kernel/traps.c
@@ -0,0 +1,333 @@
+/*
+ * linux/arch/unicore32/kernel/traps.c
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ *  'traps.c' handles hardware exceptions after we have saved some state.
+ *  Mostly a debugging aid, but will probably kill the offending process.
+ */
+#include <linux/module.h>
+#include <linux/signal.h>
+#include <linux/spinlock.h>
+#include <linux/personality.h>
+#include <linux/kallsyms.h>
+#include <linux/kdebug.h>
+#include <linux/uaccess.h>
+#include <linux/delay.h>
+#include <linux/hardirq.h>
+#include <linux/init.h>
+#include <linux/uaccess.h>
+#include <linux/atomic.h>
+#include <linux/unistd.h>
+
+#include <asm/cacheflush.h>
+#include <asm/system.h>
+#include <asm/traps.h>
+
+#include "setup.h"
+
+static void dump_mem(const char *, const char *, unsigned long, unsigned long);
+
+void dump_backtrace_entry(unsigned long where,
+		unsigned long from, unsigned long frame)
+{
+#ifdef CONFIG_KALLSYMS
+	printk(KERN_DEFAULT "[<%08lx>] (%pS) from [<%08lx>] (%pS)\n",
+			where, (void *)where, from, (void *)from);
+#else
+	printk(KERN_DEFAULT "Function entered at [<%08lx>] from [<%08lx>]\n",
+			where, from);
+#endif
+}
+
+/*
+ * Stack pointers should always be within the kernels view of
+ * physical memory.  If it is not there, then we can't dump
+ * out any information relating to the stack.
+ */
+static int verify_stack(unsigned long sp)
+{
+	if (sp < PAGE_OFFSET ||
+	    (sp > (unsigned long)high_memory && high_memory != NULL))
+		return -EFAULT;
+
+	return 0;
+}
+
+/*
+ * Dump out the contents of some memory nicely...
+ */
+static void dump_mem(const char *lvl, const char *str, unsigned long bottom,
+		     unsigned long top)
+{
+	unsigned long first;
+	mm_segment_t fs;
+	int i;
+
+	/*
+	 * We need to switch to kernel mode so that we can use __get_user
+	 * to safely read from kernel space.  Note that we now dump the
+	 * code first, just in case the backtrace kills us.
+	 */
+	fs = get_fs();
+	set_fs(KERNEL_DS);
+
+	printk(KERN_DEFAULT "%s%s(0x%08lx to 0x%08lx)\n",
+			lvl, str, bottom, top);
+
+	for (first = bottom & ~31; first < top; first += 32) {
+		unsigned long p;
+		char str[sizeof(" 12345678") * 8 + 1];
+
+		memset(str, ' ', sizeof(str));
+		str[sizeof(str) - 1] = '\0';
+
+		for (p = first, i = 0; i < 8 && p < top; i++, p += 4) {
+			if (p >= bottom && p < top) {
+				unsigned long val;
+				if (__get_user(val, (unsigned long *)p) == 0)
+					sprintf(str + i * 9, " %08lx", val);
+				else
+					sprintf(str + i * 9, " ????????");
+			}
+		}
+		printk(KERN_DEFAULT "%s%04lx:%s\n", lvl, first & 0xffff, str);
+	}
+
+	set_fs(fs);
+}
+
+static void dump_instr(const char *lvl, struct pt_regs *regs)
+{
+	unsigned long addr = instruction_pointer(regs);
+	const int width = 8;
+	mm_segment_t fs;
+	char str[sizeof("00000000 ") * 5 + 2 + 1], *p = str;
+	int i;
+
+	/*
+	 * We need to switch to kernel mode so that we can use __get_user
+	 * to safely read from kernel space.  Note that we now dump the
+	 * code first, just in case the backtrace kills us.
+	 */
+	fs = get_fs();
+	set_fs(KERNEL_DS);
+
+	for (i = -4; i < 1; i++) {
+		unsigned int val, bad;
+
+		bad = __get_user(val, &((u32 *)addr)[i]);
+
+		if (!bad)
+			p += sprintf(p, i == 0 ? "(%0*x) " : "%0*x ",
+					width, val);
+		else {
+			p += sprintf(p, "bad PC value");
+			break;
+		}
+	}
+	printk(KERN_DEFAULT "%sCode: %s\n", lvl, str);
+
+	set_fs(fs);
+}
+
+static void dump_backtrace(struct pt_regs *regs, struct task_struct *tsk)
+{
+	unsigned int fp, mode;
+	int ok = 1;
+
+	printk(KERN_DEFAULT "Backtrace: ");
+
+	if (!tsk)
+		tsk = current;
+
+	if (regs) {
+		fp = regs->UCreg_fp;
+		mode = processor_mode(regs);
+	} else if (tsk != current) {
+		fp = thread_saved_fp(tsk);
+		mode = 0x10;
+	} else {
+		asm("mov %0, fp" : "=r" (fp) : : "cc");
+		mode = 0x10;
+	}
+
+	if (!fp) {
+		printk("no frame pointer");
+		ok = 0;
+	} else if (verify_stack(fp)) {
+		printk("invalid frame pointer 0x%08x", fp);
+		ok = 0;
+	} else if (fp < (unsigned long)end_of_stack(tsk))
+		printk("frame pointer underflow");
+	printk("\n");
+
+	if (ok)
+		c_backtrace(fp, mode);
+}
+
+void dump_stack(void)
+{
+	dump_backtrace(NULL, NULL);
+}
+EXPORT_SYMBOL(dump_stack);
+
+void show_stack(struct task_struct *tsk, unsigned long *sp)
+{
+	dump_backtrace(NULL, tsk);
+	barrier();
+}
+
+static int __die(const char *str, int err, struct thread_info *thread,
+		struct pt_regs *regs)
+{
+	struct task_struct *tsk = thread->task;
+	static int die_counter;
+	int ret;
+
+	printk(KERN_EMERG "Internal error: %s: %x [#%d]\n",
+	       str, err, ++die_counter);
+	sysfs_printk_last_file();
+
+	/* trap and error numbers are mostly meaningless on UniCore */
+	ret = notify_die(DIE_OOPS, str, regs, err, tsk->thread.trap_no, \
+			SIGSEGV);
+	if (ret == NOTIFY_STOP)
+		return ret;
+
+	print_modules();
+	__show_regs(regs);
+	printk(KERN_EMERG "Process %.*s (pid: %d, stack limit = 0x%p)\n",
+		TASK_COMM_LEN, tsk->comm, task_pid_nr(tsk), thread + 1);
+
+	if (!user_mode(regs) || in_interrupt()) {
+		dump_mem(KERN_EMERG, "Stack: ", regs->UCreg_sp,
+			 THREAD_SIZE + (unsigned long)task_stack_page(tsk));
+		dump_backtrace(regs, tsk);
+		dump_instr(KERN_EMERG, regs);
+	}
+
+	return ret;
+}
+
+DEFINE_SPINLOCK(die_lock);
+
+/*
+ * This function is protected against re-entrancy.
+ */
+void die(const char *str, struct pt_regs *regs, int err)
+{
+	struct thread_info *thread = current_thread_info();
+	int ret;
+
+	oops_enter();
+
+	spin_lock_irq(&die_lock);
+	console_verbose();
+	bust_spinlocks(1);
+	ret = __die(str, err, thread, regs);
+
+	bust_spinlocks(0);
+	add_taint(TAINT_DIE);
+	spin_unlock_irq(&die_lock);
+	oops_exit();
+
+	if (in_interrupt())
+		panic("Fatal exception in interrupt");
+	if (panic_on_oops)
+		panic("Fatal exception");
+	if (ret != NOTIFY_STOP)
+		do_exit(SIGSEGV);
+}
+
+void uc32_notify_die(const char *str, struct pt_regs *regs,
+		struct siginfo *info, unsigned long err, unsigned long trap)
+{
+	if (user_mode(regs)) {
+		current->thread.error_code = err;
+		current->thread.trap_no = trap;
+
+		force_sig_info(info->si_signo, info, current);
+	} else
+		die(str, regs, err);
+}
+
+/*
+ * bad_mode handles the impossible case in the vectors.  If you see one of
+ * these, then it's extremely serious, and could mean you have buggy hardware.
+ * It never returns, and never tries to sync.  We hope that we can at least
+ * dump out some state information...
+ */
+asmlinkage void bad_mode(struct pt_regs *regs, unsigned int reason)
+{
+	console_verbose();
+
+	printk(KERN_CRIT "Bad mode detected with reason 0x%x\n", reason);
+
+	die("Oops - bad mode", regs, 0);
+	local_irq_disable();
+	panic("bad mode");
+}
+
+void __pte_error(const char *file, int line, unsigned long val)
+{
+	printk(KERN_DEFAULT "%s:%d: bad pte %08lx.\n", file, line, val);
+}
+
+void __pmd_error(const char *file, int line, unsigned long val)
+{
+	printk(KERN_DEFAULT "%s:%d: bad pmd %08lx.\n", file, line, val);
+}
+
+void __pgd_error(const char *file, int line, unsigned long val)
+{
+	printk(KERN_DEFAULT "%s:%d: bad pgd %08lx.\n", file, line, val);
+}
+
+asmlinkage void __div0(void)
+{
+	printk(KERN_DEFAULT "Division by zero in kernel.\n");
+	dump_stack();
+}
+EXPORT_SYMBOL(__div0);
+
+void abort(void)
+{
+	BUG();
+
+	/* if that doesn't kill us, halt */
+	panic("Oops failed to kill thread");
+}
+EXPORT_SYMBOL(abort);
+
+void __init trap_init(void)
+{
+	return;
+}
+
+void __init early_trap_init(void)
+{
+	unsigned long vectors = VECTORS_BASE;
+
+	/*
+	 * Copy the vectors, stubs (in entry-unicore.S)
+	 * into the vector page, mapped at 0xffff0000, and ensure these
+	 * are visible to the instruction stream.
+	 */
+	memcpy((void *)vectors,
+			__vectors_start,
+			__vectors_end - __vectors_start);
+	memcpy((void *)vectors + 0x200,
+			__stubs_start,
+			__stubs_end - __stubs_start);
+
+	early_signal_init();
+
+	flush_icache_range(vectors, vectors + PAGE_SIZE);
+}
diff --git a/arch/unicore32/kernel/vmlinux.lds.S b/arch/unicore32/kernel/vmlinux.lds.S
new file mode 100644
index 000000000000..0b4eb89729e7
--- /dev/null
+++ b/arch/unicore32/kernel/vmlinux.lds.S
@@ -0,0 +1,61 @@
+/*
+ * linux/arch/unicore32/kernel/vmlinux.lds.S
+ *
+ * Code specific to PKUnity SoC and UniCore ISA
+ *
+ * Copyright (C) 2001-2010 GUAN Xue-tao
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <asm-generic/vmlinux.lds.h>
+#include <asm/thread_info.h>
+#include <asm/memory.h>
+#include <asm/page.h>
+
+OUTPUT_ARCH(unicore32)
+ENTRY(stext)
+
+jiffies = jiffies_64;
+
+SECTIONS
+{
+	. = PAGE_OFFSET + KERNEL_IMAGE_START;
+
+	_text = .;
+	__init_begin = .;
+	HEAD_TEXT_SECTION
+	INIT_TEXT_SECTION(PAGE_SIZE)
+	INIT_DATA_SECTION(16)
+	PERCPU(PAGE_SIZE)
+	__init_end = .;
+
+	_stext = .;
+	.text : {		/* Real text segment */
+		TEXT_TEXT
+		SCHED_TEXT
+		LOCK_TEXT
+
+		*(.fixup)
+		*(.gnu.warning)
+	}
+	_etext = .;
+
+	_sdata = .;
+	RO_DATA_SECTION(PAGE_SIZE)
+	RW_DATA_SECTION(32, PAGE_SIZE, THREAD_SIZE)
+	_edata = .;
+
+	EXCEPTION_TABLE(32)
+	NOTES
+
+	BSS_SECTION(0, 0, 0)
+	_end = .;
+
+	STABS_DEBUG
+	DWARF_DEBUG
+
+	DISCARDS		/* Exit code and data */
+}