Merge commit 'origin/HEAD' into test-merge

Manual fixup of include/asm-powerpc/pgtable-ppc64.h

37 files changed, 247 insertions, 75 deletions

diff --git a/arch/alpha/kernel/vmlinux.lds.S b/arch/alpha/kernel/vmlinux.lds.S
index f13249be17c..ef37fc1acae 100644
--- a/arch/alpha/kernel/vmlinux.lds.S
+++ b/arch/alpha/kernel/vmlinux.lds.S
@@ -25,6 +25,13 @@ SECTIONS
 	} :kernel
 	_etext = .;	/* End of text section */
 
+	NOTES :kernel :note
+	.dummy : {
+		*(.dummy)
+	} :kernel
+
+	RODATA
+
 	/* Exception table */
 	. = ALIGN(16);
 	__ex_table : {
@@ -33,13 +40,6 @@ SECTIONS
 		__stop___ex_table = .;
 	}
 
-	NOTES :kernel :note
-	.dummy : {
-		*(.dummy)
-	} :kernel
-
-	RODATA
-
 	/* Will be freed after init */
 	. = ALIGN(PAGE_SIZE);
 	/* Init code and data */
diff --git a/arch/arm/common/dmabounce.c b/arch/arm/common/dmabounce.c
index 52fc6a88328..2744673314b 100644
--- a/arch/arm/common/dmabounce.c
+++ b/arch/arm/common/dmabounce.c
@@ -650,7 +650,8 @@ EXPORT_SYMBOL(dma_map_sg);
 EXPORT_SYMBOL(dma_unmap_sg);
 EXPORT_SYMBOL(dma_sync_single_for_cpu);
 EXPORT_SYMBOL(dma_sync_single_for_device);
-EXPORT_SYMBOL(dma_sync_sg);
+EXPORT_SYMBOL(dma_sync_sg_for_cpu);
+EXPORT_SYMBOL(dma_sync_sg_for_device);
 EXPORT_SYMBOL(dmabounce_register_dev);
 EXPORT_SYMBOL(dmabounce_unregister_dev);
 
diff --git a/arch/arm/mach-omap2/gpmc.c b/arch/arm/mach-omap2/gpmc.c
index 02cede295e8..dbf68dc50ae 100644
--- a/arch/arm/mach-omap2/gpmc.c
+++ b/arch/arm/mach-omap2/gpmc.c
@@ -42,7 +42,7 @@
 #define GPMC_STATUS		0x54
 #define GPMC_PREFETCH_CONFIG1	0x1e0
 #define GPMC_PREFETCH_CONFIG2	0x1e4
-#define GPMC_PREFETCH_CONTROL	0x1e8
+#define GPMC_PREFETCH_CONTROL	0x1ec
 #define GPMC_PREFETCH_STATUS	0x1f0
 #define GPMC_ECC_CONFIG		0x1f4
 #define GPMC_ECC_CONTROL	0x1f8
diff --git a/arch/arm/mach-realview/platsmp.c b/arch/arm/mach-realview/platsmp.c
index 3e57428affe..8e813ed5751 100644
--- a/arch/arm/mach-realview/platsmp.c
+++ b/arch/arm/mach-realview/platsmp.c
@@ -74,6 +74,8 @@ static DEFINE_SPINLOCK(boot_lock);
 
 void __cpuinit platform_secondary_init(unsigned int cpu)
 {
+	trace_hardirqs_off();
+
 	/*
 	 * the primary core may have used a "cross call" soft interrupt
 	 * to get this processor out of WFI in the BootMonitor - make
diff --git a/arch/arm/plat-omap/dma.c b/arch/arm/plat-omap/dma.c
index c00eda588cd..39c637b0ffe 100644
--- a/arch/arm/plat-omap/dma.c
+++ b/arch/arm/plat-omap/dma.c
@@ -501,8 +501,6 @@ static inline void omap_enable_channel_irq(int lch)
 
 	/* Enable some nice interrupts. */
 	OMAP_DMA_CICR_REG(lch) = dma_chan[lch].enabled_irqs;
-
-	dma_chan[lch].flags |= OMAP_DMA_ACTIVE;
 }
 
 static void omap_disable_channel_irq(int lch)
diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c
index 4ae15c8c248..632cda8f2e7 100644
--- a/arch/ia64/kernel/setup.c
+++ b/arch/ia64/kernel/setup.c
@@ -547,7 +547,8 @@ setup_arch (char **cmdline_p)
 # ifdef CONFIG_ACPI_NUMA
 	acpi_numa_init();
 	per_cpu_scan_finalize((cpus_weight(early_cpu_possible_map) == 0 ?
-		32 : cpus_weight(early_cpu_possible_map)), additional_cpus);
+		32 : cpus_weight(early_cpu_possible_map)),
+		additional_cpus > 0 ? additional_cpus : 0);
 # endif
 #else
 # ifdef CONFIG_SMP
diff --git a/arch/ia64/kernel/time.c b/arch/ia64/kernel/time.c
index 8c73643f2d6..aad1b7b1fff 100644
--- a/arch/ia64/kernel/time.c
+++ b/arch/ia64/kernel/time.c
@@ -117,6 +117,7 @@ void account_system_vtime(struct task_struct *tsk)
 
 	local_irq_restore(flags);
 }
+EXPORT_SYMBOL_GPL(account_system_vtime);
 
 /*
  * Called from the timer interrupt handler to charge accumulated user time
diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index e5a7c5d9636..24c5dee9176 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -1006,7 +1006,7 @@ config BOOT_ELF32
 config MIPS_L1_CACHE_SHIFT
 	int
 	default "4" if MACH_DECSTATION
-	default "7" if SGI_IP27 || SGI_IP28 || SNI_RM
+	default "7" if SGI_IP22 || SGI_IP27 || SGI_IP28 || SNI_RM
 	default "4" if PMC_MSP4200_EVAL
 	default "5"
 
diff --git a/arch/mips/kernel/cevt-txx9.c b/arch/mips/kernel/cevt-txx9.c
index 795cb8fb0d7..b5fc4eb412d 100644
--- a/arch/mips/kernel/cevt-txx9.c
+++ b/arch/mips/kernel/cevt-txx9.c
@@ -161,6 +161,9 @@ void __init txx9_tmr_init(unsigned long baseaddr)
 	struct txx9_tmr_reg __iomem *tmrptr;
 
 	tmrptr = ioremap(baseaddr, sizeof(struct txx9_tmr_reg));
+	/* Start once to make CounterResetEnable effective */
+	__raw_writel(TXx9_TMTCR_CRE | TXx9_TMTCR_TCE, &tmrptr->tcr);
+	/* Stop and reset the counter */
 	__raw_writel(TXx9_TMTCR_CRE, &tmrptr->tcr);
 	__raw_writel(0, &tmrptr->tisr);
 	__raw_writel(0xffffffff, &tmrptr->cpra);
diff --git a/arch/mips/mm/c-r3k.c b/arch/mips/mm/c-r3k.c
index 76935e32021..27a5b466c85 100644
--- a/arch/mips/mm/c-r3k.c
+++ b/arch/mips/mm/c-r3k.c
@@ -26,7 +26,7 @@
 static unsigned long icache_size, dcache_size;		/* Size in bytes */
 static unsigned long icache_lsize, dcache_lsize;	/* Size in bytes */
 
-unsigned long __init r3k_cache_size(unsigned long ca_flags)
+unsigned long __cpuinit r3k_cache_size(unsigned long ca_flags)
 {
 	unsigned long flags, status, dummy, size;
 	volatile unsigned long *p;
@@ -61,7 +61,7 @@ unsigned long __init r3k_cache_size(unsigned long ca_flags)
 	return size * sizeof(*p);
 }
 
-unsigned long __init r3k_cache_lsize(unsigned long ca_flags)
+unsigned long __cpuinit r3k_cache_lsize(unsigned long ca_flags)
 {
 	unsigned long flags, status, lsize, i;
 	volatile unsigned long *p;
@@ -90,7 +90,7 @@ unsigned long __init r3k_cache_lsize(unsigned long ca_flags)
 	return lsize * sizeof(*p);
 }
 
-static void __init r3k_probe_cache(void)
+static void __cpuinit r3k_probe_cache(void)
 {
 	dcache_size = r3k_cache_size(ST0_ISC);
 	if (dcache_size)
diff --git a/arch/mips/mm/page.c b/arch/mips/mm/page.c
index 1edf0cbbeed..1417c649485 100644
--- a/arch/mips/mm/page.c
+++ b/arch/mips/mm/page.c
@@ -235,13 +235,12 @@ static void __cpuinit set_prefetch_parameters(void)
 	}
 	/*
 	 * Too much unrolling will overflow the available space in
-	 * clear_space_array / copy_page_array. 8 words sounds generous,
-	 * but a R4000 with 128 byte L2 line length can exceed even that.
+	 * clear_space_array / copy_page_array.
 	 */
-	half_clear_loop_size = min(8 * clear_word_size,
+	half_clear_loop_size = min(16 * clear_word_size,
 				   max(cache_line_size >> 1,
 				       4 * clear_word_size));
-	half_copy_loop_size = min(8 * copy_word_size,
+	half_copy_loop_size = min(16 * copy_word_size,
 				  max(cache_line_size >> 1,
 				      4 * copy_word_size));
 }
@@ -263,21 +262,23 @@ static inline void __cpuinit build_clear_pref(u32 **buf, int off)
 	if (pref_bias_clear_store) {
 		uasm_i_pref(buf, pref_dst_mode, pref_bias_clear_store + off,
 			    A0);
-	} else if (cpu_has_cache_cdex_s) {
-		uasm_i_cache(buf, Create_Dirty_Excl_SD, off, A0);
-	} else if (cpu_has_cache_cdex_p) {
-		if (R4600_V1_HIT_CACHEOP_WAR && cpu_is_r4600_v1_x()) {
-			uasm_i_nop(buf);
-			uasm_i_nop(buf);
-			uasm_i_nop(buf);
-			uasm_i_nop(buf);
-		}
+	} else if (cache_line_size == (half_clear_loop_size << 1)) {
+		if (cpu_has_cache_cdex_s) {
+			uasm_i_cache(buf, Create_Dirty_Excl_SD, off, A0);
+		} else if (cpu_has_cache_cdex_p) {
+			if (R4600_V1_HIT_CACHEOP_WAR && cpu_is_r4600_v1_x()) {
+				uasm_i_nop(buf);
+				uasm_i_nop(buf);
+				uasm_i_nop(buf);
+				uasm_i_nop(buf);
+			}
 
-		if (R4600_V2_HIT_CACHEOP_WAR && cpu_is_r4600_v2_x())
-			uasm_i_lw(buf, ZERO, ZERO, AT);
+			if (R4600_V2_HIT_CACHEOP_WAR && cpu_is_r4600_v2_x())
+				uasm_i_lw(buf, ZERO, ZERO, AT);
 
-		uasm_i_cache(buf, Create_Dirty_Excl_D, off, A0);
-	}
+			uasm_i_cache(buf, Create_Dirty_Excl_D, off, A0);
+		}
+		}
 }
 
 void __cpuinit build_clear_page(void)
@@ -403,20 +404,22 @@ static inline void build_copy_store_pref(u32 **buf, int off)
 	if (pref_bias_copy_store) {
 		uasm_i_pref(buf, pref_dst_mode, pref_bias_copy_store + off,
 			    A0);
-	} else if (cpu_has_cache_cdex_s) {
-		uasm_i_cache(buf, Create_Dirty_Excl_SD, off, A0);
-	} else if (cpu_has_cache_cdex_p) {
-		if (R4600_V1_HIT_CACHEOP_WAR && cpu_is_r4600_v1_x()) {
-			uasm_i_nop(buf);
-			uasm_i_nop(buf);
-			uasm_i_nop(buf);
-			uasm_i_nop(buf);
-		}
+	} else if (cache_line_size == (half_copy_loop_size << 1)) {
+		if (cpu_has_cache_cdex_s) {
+			uasm_i_cache(buf, Create_Dirty_Excl_SD, off, A0);
+		} else if (cpu_has_cache_cdex_p) {
+			if (R4600_V1_HIT_CACHEOP_WAR && cpu_is_r4600_v1_x()) {
+				uasm_i_nop(buf);
+				uasm_i_nop(buf);
+				uasm_i_nop(buf);
+				uasm_i_nop(buf);
+			}
 
-		if (R4600_V2_HIT_CACHEOP_WAR && cpu_is_r4600_v2_x())
-			uasm_i_lw(buf, ZERO, ZERO, AT);
+			if (R4600_V2_HIT_CACHEOP_WAR && cpu_is_r4600_v2_x())
+				uasm_i_lw(buf, ZERO, ZERO, AT);
 
-		uasm_i_cache(buf, Create_Dirty_Excl_D, off, A0);
+			uasm_i_cache(buf, Create_Dirty_Excl_D, off, A0);
+		}
 	}
 }
 
diff --git a/arch/mips/mm/sc-rm7k.c b/arch/mips/mm/sc-rm7k.c
index fc227f3b119..e3abfb2d7e8 100644
--- a/arch/mips/mm/sc-rm7k.c
+++ b/arch/mips/mm/sc-rm7k.c
@@ -86,7 +86,7 @@ static void rm7k_sc_inv(unsigned long addr, unsigned long size)
 /*
  * This function is executed in uncached address space.
  */
-static __init void __rm7k_sc_enable(void)
+static __cpuinit void __rm7k_sc_enable(void)
 {
 	int i;
 
@@ -107,7 +107,7 @@ static __init void __rm7k_sc_enable(void)
 	}
 }
 
-static __init void rm7k_sc_enable(void)
+static __cpuinit void rm7k_sc_enable(void)
 {
 	if (read_c0_config() & RM7K_CONF_SE)
 		return;
diff --git a/arch/mips/sgi-ip32/ip32-irq.c b/arch/mips/sgi-ip32/ip32-irq.c
index b0ea0e43ba4..0d6b6663d5f 100644
--- a/arch/mips/sgi-ip32/ip32-irq.c
+++ b/arch/mips/sgi-ip32/ip32-irq.c
@@ -425,6 +425,11 @@ static void ip32_irq0(void)
 	BUILD_BUG_ON(MACEISA_SERIAL2_RDMAOR_IRQ - MACEISA_AUDIO_SW_IRQ != 31);
 
 	crime_int = crime->istat & crime_mask;
+
+	/* crime sometime delivers spurious interrupts, ignore them */
+	if (unlikely(crime_int == 0))
+		return;
+
 	irq = MACE_VID_IN1_IRQ + __ffs(crime_int);
 
 	if (crime_int & CRIME_MACEISA_INT_MASK) {
diff --git a/arch/mn10300/kernel/mn10300_ksyms.c b/arch/mn10300/kernel/mn10300_ksyms.c
index 6d19628634e..f9eb9753a40 100644
--- a/arch/mn10300/kernel/mn10300_ksyms.c
+++ b/arch/mn10300/kernel/mn10300_ksyms.c
@@ -10,8 +10,11 @@
  */
 #include <linux/module.h>
 #include <asm/uaccess.h>
+#include <asm/pgtable.h>
 
 
+EXPORT_SYMBOL(empty_zero_page);
+
 EXPORT_SYMBOL(change_bit);
 EXPORT_SYMBOL(test_and_change_bit);
 
@@ -31,7 +34,9 @@ extern u64 __ashrdi3(u64, unsigned);
 extern u64 __ashldi3(u64, unsigned);
 extern u64 __lshrdi3(u64, unsigned);
 extern s64 __negdi2(s64);
+extern int __ucmpdi2(u64, u64);
 EXPORT_SYMBOL(__ashrdi3);
 EXPORT_SYMBOL(__ashldi3);
 EXPORT_SYMBOL(__lshrdi3);
 EXPORT_SYMBOL(__negdi2);
+EXPORT_SYMBOL(__ucmpdi2);
diff --git a/arch/mn10300/kernel/process.c b/arch/mn10300/kernel/process.c
index 9c623c88387..b28c9a60445 100644
--- a/arch/mn10300/kernel/process.c
+++ b/arch/mn10300/kernel/process.c
@@ -153,6 +153,7 @@ int kernel_thread(int (*fn)(void *), void *arg, unsigned long flags)
 	return do_fork(flags | CLONE_VM | CLONE_UNTRACED, 0, &regs, 0,
 		       NULL, NULL);
 }
+EXPORT_SYMBOL(kernel_thread);
 
 /*
  * free current thread data structures etc..
diff --git a/arch/mn10300/lib/Makefile b/arch/mn10300/lib/Makefile
index fdfa9ec5b5b..0cd2346f4c1 100644
--- a/arch/mn10300/lib/Makefile
+++ b/arch/mn10300/lib/Makefile
@@ -4,4 +4,4 @@
 
 lib-y = delay.o usercopy.o checksum.o bitops.o memcpy.o memmove.o memset.o
 lib-y += do_csum.o
-lib-y += __ashldi3.o __ashrdi3.o __lshrdi3.o negdi2.o
+lib-y += __ashldi3.o __ashrdi3.o __lshrdi3.o negdi2.o __ucmpdi2.o
diff --git a/arch/mn10300/lib/__ucmpdi2.S b/arch/mn10300/lib/__ucmpdi2.S
new file mode 100644
index 00000000000..60dcbdfe386
--- /dev/null
+++ b/arch/mn10300/lib/__ucmpdi2.S
@@ -0,0 +1,43 @@
+/* __ucmpdi2.S: 64-bit unsigned compare
+ *
+ * Copyright (C) 2008 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells (dhowells@redhat.com)
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+
+        .text
+        .p2align	4
+
+###############################################################################
+#
+# int __ucmpdi2(unsigned long long a [D0:D1],
+#		unsigned long long b [(SP,12),(SP,16)])
+#
+# - returns 0, 1, or 2 as a <, =, > b respectively.
+#
+###############################################################################
+        .globl		__ucmpdi2
+        .type		__ucmpdi2,@function
+__ucmpdi2:
+	mov		(12,sp),a0		# b.lsw
+	mov		(16,sp),a1		# b.msw
+
+	sub		a0,d0
+	subc		a1,d1			# may clear Z, never sets it
+	bne		__ucmpdi2_differ	# a.msw != b.msw
+	mov		+1,d0
+	rets
+
+__ucmpdi2_differ:
+	# C flag is set if LE, clear if GE
+	subc		d0,d0			# -1 if LE, 0 if GE
+	add		+1,d0			#  0 if LE, 1 if GE
+	add		d0,d0			#  0 if LE, 2 if GE
+	rets
+
+	.size		__ucmpdi2, .-__ucmpdi2
diff --git a/arch/powerpc/boot/Makefile b/arch/powerpc/boot/Makefile
index 751a6e6070d..0055edbc774 100644
--- a/arch/powerpc/boot/Makefile
+++ b/arch/powerpc/boot/Makefile
@@ -279,7 +279,8 @@ endif
 initrd-  := $(patsubst zImage%, zImage.initrd%, $(image-n) $(image-))
 initrd-y := $(patsubst zImage%, zImage.initrd%, \
 		$(patsubst dtbImage%, dtbImage.initrd%, \
-		$(patsubst treeImage%, treeImage.initrd%, $(image-y))))
+		$(patsubst simpleImage%, simpleImage.initrd%, \
+		$(patsubst treeImage%, treeImage.initrd%, $(image-y)))))
 initrd-y := $(filter-out $(image-y), $(initrd-y))
 targets	+= $(image-y) $(initrd-y)
 
diff --git a/arch/powerpc/kernel/legacy_serial.c b/arch/powerpc/kernel/legacy_serial.c
index 61dd17449dd..4d96e1db55e 100644
--- a/arch/powerpc/kernel/legacy_serial.c
+++ b/arch/powerpc/kernel/legacy_serial.c
@@ -33,13 +33,14 @@ static struct legacy_serial_info {
 	phys_addr_t			taddr;
 } legacy_serial_infos[MAX_LEGACY_SERIAL_PORTS];
 
-static struct __initdata of_device_id parents[] = {
+static struct __initdata of_device_id legacy_serial_parents[] = {
 	{.type = "soc",},
 	{.type = "tsi-bridge",},
 	{.type = "opb", },
 	{.compatible = "ibm,opb",},
 	{.compatible = "simple-bus",},
 	{.compatible = "wrs,epld-localbus",},
+	{},
 };
 
 static unsigned int legacy_serial_count;
@@ -136,6 +137,11 @@ static int __init add_legacy_soc_port(struct device_node *np,
 	if (of_get_property(np, "clock-frequency", NULL) == NULL)
 		return -1;
 
+	/* if reg-shift or offset, don't try to use it */
+	if ((of_get_property(np, "reg-shift", NULL) != NULL) ||
+		(of_get_property(np, "reg-offset", NULL) != NULL))
+		return -1;
+
 	/* if rtas uses this device, don't try to use it as well */
 	if (of_get_property(np, "used-by-rtas", NULL) != NULL)
 		return -1;
@@ -322,7 +328,7 @@ void __init find_legacy_serial_ports(void)
 		struct device_node *parent = of_get_parent(np);
 		if (!parent)
 			continue;
-		if (of_match_node(parents, parent) != NULL) {
+		if (of_match_node(legacy_serial_parents, parent) != NULL) {
 			index = add_legacy_soc_port(np, np);
 			if (index >= 0 && np == stdout)
 				legacy_serial_console = index;
diff --git a/arch/powerpc/kernel/of_platform.c b/arch/powerpc/kernel/of_platform.c
index e79ad8afda0..3f37a6e6277 100644
--- a/arch/powerpc/kernel/of_platform.c
+++ b/arch/powerpc/kernel/of_platform.c
@@ -76,6 +76,8 @@ struct of_device* of_platform_device_create(struct device_node *np,
 		return NULL;
 
 	dev->dma_mask = 0xffffffffUL;
+	dev->dev.coherent_dma_mask = DMA_32BIT_MASK;
+
 	dev->dev.bus = &of_platform_bus_type;
 
 	/* We do not fill the DMA ops for platform devices by default.
diff --git a/arch/powerpc/platforms/52xx/lite5200_pm.c b/arch/powerpc/platforms/52xx/lite5200_pm.c
index 41c7fd91e99..fe92e65103e 100644
--- a/arch/powerpc/platforms/52xx/lite5200_pm.c
+++ b/arch/powerpc/platforms/52xx/lite5200_pm.c
@@ -14,6 +14,7 @@ static struct mpc52xx_sdma __iomem *bes;
 static struct mpc52xx_xlb __iomem *xlb;
 static struct mpc52xx_gpio __iomem *gps;
 static struct mpc52xx_gpio_wkup __iomem *gpw;
+static void __iomem *pci;
 static void __iomem *sram;
 static const int sram_size = 0x4000;	/* 16 kBytes */
 static void __iomem *mbar;
@@ -50,6 +51,8 @@ static int lite5200_pm_prepare(void)
 		{ .type = "builtin", .compatible = "mpc5200", }, /* efika */
 		{}
 	};
+	u64 regaddr64 = 0;
+	const u32 *regaddr_p;
 
 	/* deep sleep? let mpc52xx code handle that */
 	if (lite5200_pm_target_state == PM_SUSPEND_STANDBY)
@@ -60,8 +63,12 @@ static int lite5200_pm_prepare(void)
 
 	/* map registers */
 	np = of_find_matching_node(NULL, immr_ids);
-	mbar = of_iomap(np, 0);
+	regaddr_p = of_get_address(np, 0, NULL, NULL);
+	if (regaddr_p)
+		regaddr64 = of_translate_address(np, regaddr_p);
 	of_node_put(np);
+
+	mbar = ioremap((u32) regaddr64, 0xC000);
 	if (!mbar) {
 		printk(KERN_ERR "%s:%i Error mapping registers\n", __func__, __LINE__);
 		return -ENOSYS;
@@ -71,6 +78,7 @@ static int lite5200_pm_prepare(void)
 	pic = mbar + 0x500;
 	gps = mbar + 0xb00;
 	gpw = mbar + 0xc00;
+	pci = mbar + 0xd00;
 	bes = mbar + 0x1200;
 	xlb = mbar + 0x1f00;
 	sram = mbar + 0x8000;
@@ -85,6 +93,7 @@ static struct mpc52xx_sdma sbes;
 static struct mpc52xx_xlb sxlb;
 static struct mpc52xx_gpio sgps;
 static struct mpc52xx_gpio_wkup sgpw;
+static char spci[0x200];
 
 static void lite5200_save_regs(void)
 {
@@ -94,6 +103,7 @@ static void lite5200_save_regs(void)
 	_memcpy_fromio(&sxlb, xlb, sizeof(*xlb));
 	_memcpy_fromio(&sgps, gps, sizeof(*gps));
 	_memcpy_fromio(&sgpw, gpw, sizeof(*gpw));
+	_memcpy_fromio(spci, pci, 0x200);
 
 	_memcpy_fromio(saved_sram, sram, sram_size);
 }
@@ -103,6 +113,8 @@ static void lite5200_restore_regs(void)
 	int i;
 	_memcpy_toio(sram, saved_sram, sram_size);
 
+	/* PCI Configuration */
+	_memcpy_toio(pci, spci, 0x200);
 
 	/*
 	 * GPIOs. Interrupt Master Enable has higher address then other
diff --git a/arch/um/Makefile b/arch/um/Makefile
index dbeab15e7bb..ca40397017b 100644
--- a/arch/um/Makefile
+++ b/arch/um/Makefile
@@ -77,7 +77,6 @@ include $(srctree)/$(ARCH_DIR)/Makefile-os-$(OS)
 KERNEL_DEFINES = $(strip -Derrno=kernel_errno -Dsigprocmask=kernel_sigprocmask \
 			 -Dmktime=kernel_mktime $(ARCH_KERNEL_DEFINES))
 KBUILD_CFLAGS += $(KERNEL_DEFINES)
-KBUILD_CFLAGS += $(call cc-option,-fno-unit-at-a-time,)
 
 PHONY += linux
 
diff --git a/arch/um/Makefile-i386 b/arch/um/Makefile-i386
index 561e373bd85..302cbe50454 100644
--- a/arch/um/Makefile-i386
+++ b/arch/um/Makefile-i386
@@ -32,4 +32,11 @@ cflags-y += $(call cc-option,-mpreferred-stack-boundary=2)
 # an unresolved reference.
 cflags-y += -ffreestanding
 
+# Disable unit-at-a-time mode on pre-gcc-4.0 compilers, it makes gcc use
+# a lot more stack due to the lack of sharing of stacklots.  Also, gcc
+# 4.3.0 needs -funit-at-a-time for extern inline functions.
+KBUILD_CFLAGS += $(shell if [ $(call cc-version) -lt 0400 ] ; then \
+			echo $(call cc-option,-fno-unit-at-a-time); \
+			else echo $(call cc-option,-funit-at-a-time); fi ;)
+
 KBUILD_CFLAGS += $(cflags-y)
diff --git a/arch/um/Makefile-x86_64 b/arch/um/Makefile-x86_64
index 8ed362f9358..a9cd7e77a7a 100644
--- a/arch/um/Makefile-x86_64
+++ b/arch/um/Makefile-x86_64
@@ -21,3 +21,6 @@ HEADER_ARCH := x86
 
 LINK-$(CONFIG_LD_SCRIPT_DYN) += -Wl,-rpath,/lib64
 LINK-y += -m64
+
+# Do unit-at-a-time unconditionally on x86_64, following the host
+KBUILD_CFLAGS += $(call cc-option,-funit-at-a-time)
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index e0edaaa6920..bf07b6f50fa 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -966,8 +966,8 @@ config NUMA_EMU
 	  number of nodes. This is only useful for debugging.
 
 config NODES_SHIFT
-	int "Max num nodes shift(1-15)"
-	range 1 15  if X86_64
+	int "Max num nodes shift(1-9)"
+	range 1 9  if X86_64
 	default "6" if X86_64
 	default "4" if X86_NUMAQ
 	default "3"
diff --git a/arch/x86/kernel/.gitignore b/arch/x86/kernel/.gitignore
index 4ea38a39aed..08f4fd73146 100644
--- a/arch/x86/kernel/.gitignore
+++ b/arch/x86/kernel/.gitignore
@@ -1,2 +1,3 @@
 vsyscall.lds
 vsyscall_32.lds
+vmlinux.lds
diff --git a/arch/x86/kernel/acpi/realmode/wakeup.S b/arch/x86/kernel/acpi/realmode/wakeup.S
index f9b77fb37e5..3355973b12a 100644
--- a/arch/x86/kernel/acpi/realmode/wakeup.S
+++ b/arch/x86/kernel/acpi/realmode/wakeup.S
@@ -5,6 +5,7 @@
 #include <asm/msr-index.h>
 #include <asm/page.h>
 #include <asm/pgtable.h>
+#include <asm/processor-flags.h>
 
 	.code16
 	.section ".header", "a"
@@ -24,6 +25,11 @@ pmode_gdt:	.quad	0
 realmode_flags:	.long	0
 real_magic:	.long	0
 trampoline_segment:	.word 0
+_pad1:		.byte	0
+wakeup_jmp:	.byte	0xea	/* ljmpw */
+wakeup_jmp_off:	.word	3f
+wakeup_jmp_seg:	.word	0
+wakeup_gdt:	.quad	0, 0, 0
 signature:	.long	0x51ee1111
 
 	.text
@@ -34,11 +40,34 @@ _start:
 	cli
 	cld
 
+	/* Apparently some dimwit BIOS programmers don't know how to
+	   program a PM to RM transition, and we might end up here with
+	   junk in the data segment descriptor registers.  The only way
+	   to repair that is to go into PM and fix it ourselves... */
+	movw	$16, %cx
+	lgdtl	%cs:wakeup_gdt
+	movl	%cr0, %eax
+	orb	$X86_CR0_PE, %al
+	movl	%eax, %cr0
+	jmp	1f
+1:	ljmpw	$8, $2f
+2:
+	movw	%cx, %ds
+	movw	%cx, %es
+	movw	%cx, %ss
+	movw	%cx, %fs
+	movw	%cx, %gs
+
+	andb	$~X86_CR0_PE, %al
+	movl	%eax, %cr0
+	jmp	wakeup_jmp
+3:
 	/* Set up segments */
 	movw	%cs, %ax
 	movw	%ax, %ds
 	movw	%ax, %es
 	movw	%ax, %ss
+	lidtl	wakeup_idt
 
 	movl	$wakeup_stack_end, %esp
 
@@ -98,7 +127,14 @@ bogus_real_magic:
 	jmp	1b
 
 	.data
-	.balign	4
+	.balign	8
+
+	/* This is the standard real-mode IDT */
+wakeup_idt:
+	.word	0xffff		/* limit */
+	.long	0		/* address */
+	.word	0
+
 	.globl	HEAP, heap_end
 HEAP:
 	.long	wakeup_heap
diff --git a/arch/x86/kernel/acpi/realmode/wakeup.h b/arch/x86/kernel/acpi/realmode/wakeup.h
index ef8166fe802..69d38d0b2b6 100644
--- a/arch/x86/kernel/acpi/realmode/wakeup.h
+++ b/arch/x86/kernel/acpi/realmode/wakeup.h
@@ -24,6 +24,11 @@ struct wakeup_header {
 	u32 realmode_flags;
 	u32 real_magic;
 	u16 trampoline_segment;	/* segment with trampoline code, 64-bit only */
+	u8  _pad1;
+	u8  wakeup_jmp;
+	u16 wakeup_jmp_off;
+	u16 wakeup_jmp_seg;
+	u64 wakeup_gdt[3];
 	u32 signature;		/* To check we have correct structure */
 } __attribute__((__packed__));
 
diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c
index afc25ee9964..36af01f029e 100644
--- a/arch/x86/kernel/acpi/sleep.c
+++ b/arch/x86/kernel/acpi/sleep.c
@@ -50,6 +50,20 @@ int acpi_save_state_mem(void)
 
 	header->video_mode = saved_video_mode;
 
+	header->wakeup_jmp_seg = acpi_wakeup_address >> 4;
+	/* GDT[0]: GDT self-pointer */
+	header->wakeup_gdt[0] =
+		(u64)(sizeof(header->wakeup_gdt) - 1) +
+		((u64)(acpi_wakeup_address +
+			((char *)&header->wakeup_gdt - (char *)acpi_realmode))
+				<< 16);
+	/* GDT[1]: real-mode-like code segment */
+	header->wakeup_gdt[1] = (0x009bULL << 40) +
+		((u64)acpi_wakeup_address << 16) + 0xffff;
+	/* GDT[2]: real-mode-like data segment */
+	header->wakeup_gdt[2] = (0x0093ULL << 40) +
+		((u64)acpi_wakeup_address << 16) + 0xffff;
+
 #ifndef CONFIG_64BIT
 	store_gdt((struct desc_ptr *)&header->pmode_gdt);
 
@@ -111,7 +125,7 @@ void __init acpi_reserve_bootmem(void)
 		return;
 	}
 
-	acpi_wakeup_address = acpi_realmode;
+	acpi_wakeup_address = virt_to_phys((void *)acpi_realmode);
 }
 
 
diff --git a/arch/x86/kernel/efi_32.c b/arch/x86/kernel/efi_32.c
index 5d23d85624d..4b63c8e1f13 100644
--- a/arch/x86/kernel/efi_32.c
+++ b/arch/x86/kernel/efi_32.c
@@ -49,13 +49,13 @@ void efi_call_phys_prelog(void)
 	local_irq_save(efi_rt_eflags);
 
 	/*
-	 * If I don't have PSE, I should just duplicate two entries in page
-	 * directory. If I have PSE, I just need to duplicate one entry in
+	 * If I don't have PAE, I should just duplicate two entries in page
+	 * directory. If I have PAE, I just need to duplicate one entry in
 	 * page directory.
 	 */
 	cr4 = read_cr4();
 
-	if (cr4 & X86_CR4_PSE) {
+	if (cr4 & X86_CR4_PAE) {
 		efi_bak_pg_dir_pointer[0].pgd =
 		    swapper_pg_dir[pgd_index(0)].pgd;
 		swapper_pg_dir[0].pgd =
@@ -93,7 +93,7 @@ void efi_call_phys_epilog(void)
 
 	cr4 = read_cr4();
 
-	if (cr4 & X86_CR4_PSE) {
+	if (cr4 & X86_CR4_PAE) {
 		swapper_pg_dir[pgd_index(0)].pgd =
 		    efi_bak_pg_dir_pointer[0].pgd;
 	} else {
diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S
index 10a1955bb1d..b817974ef94 100644
--- a/arch/x86/kernel/head_64.S
+++ b/arch/x86/kernel/head_64.S
@@ -128,7 +128,7 @@ ident_complete:
 	/* Fixup phys_base */
 	addq	%rbp, phys_base(%rip)
 
-#ifdef CONFIG_SMP
+#ifdef CONFIG_X86_TRAMPOLINE
 	addq	%rbp, trampoline_level4_pgt + 0(%rip)
 	addq	%rbp, trampoline_level4_pgt + (511*8)(%rip)
 #endif
diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c
index eb9ddd8efb8..95e80e5033c 100644
--- a/arch/x86/kernel/i387.c
+++ b/arch/x86/kernel/i387.c
@@ -162,7 +162,7 @@ int xfpregs_get(struct task_struct *target, const struct user_regset *regset,
 	int ret;
 
 	if (!cpu_has_fxsr)
-		return -ENODEV;
+		return -EIO;
 
 	ret = init_fpu(target);
 	if (ret)
@@ -179,7 +179,7 @@ int xfpregs_set(struct task_struct *target, const struct user_regset *regset,
 	int ret;
 
 	if (!cpu_has_fxsr)
-		return -ENODEV;
+		return -EIO;
 
 	ret = init_fpu(target);
 	if (ret)
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index 56078d61c79..3e1cecedde4 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -996,7 +996,6 @@ do_rest:
 #endif
 		cpu_clear(cpu, cpu_callout_map); /* was set by do_boot_cpu() */
 		cpu_clear(cpu, cpu_initialized); /* was set by cpu_init() */
-		cpu_clear(cpu, cpu_possible_map);
 		cpu_clear(cpu, cpu_present_map);
 		per_cpu(x86_cpu_to_apicid, cpu) = BAD_APICID;
 	}
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index 156e6d7b0e3..819dad973b1 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -135,7 +135,7 @@ static __init void *spp_getpage(void)
 	return ptr;
 }
 
-static void
+static __init void
 set_pte_phys(unsigned long vaddr, unsigned long phys, pgprot_t prot)
 {
 	pgd_t *pgd;
@@ -214,7 +214,7 @@ void __init cleanup_highmap(void)
 }
 
 /* NOTE: this is meant to be run only at boot */
-void __set_fixmap(enum fixed_addresses idx, unsigned long phys, pgprot_t prot)
+void __init __set_fixmap(enum fixed_addresses idx, unsigned long phys, pgprot_t prot)
 {
 	unsigned long address = __fix_to_virt(idx);
 
@@ -526,7 +526,8 @@ static void __init early_memtest(unsigned long start, unsigned long end)
 				t_size = end - t_start;
 
 			printk(KERN_CONT "\n  %016llx - %016llx pattern %d",
-				t_start, t_start + t_size, pattern);
+				(unsigned long long)t_start,
+				(unsigned long long)t_start + t_size, pattern);
 
 			memtest(t_start, t_size, pattern);
 
diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c
index 2b2bb3f9b68..d1b867101e5 100644
--- a/arch/x86/mm/ioremap.c
+++ b/arch/x86/mm/ioremap.c
@@ -300,6 +300,29 @@ void __iomem *ioremap_cache(resource_size_t phys_addr, unsigned long size)
 }
 EXPORT_SYMBOL(ioremap_cache);
 
+static void __iomem *ioremap_default(resource_size_t phys_addr,
+					unsigned long size)
+{
+	unsigned long flags;
+	void *ret;
+	int err;
+
+	/*
+	 * - WB for WB-able memory and no other conflicting mappings
+	 * - UC_MINUS for non-WB-able memory with no other conflicting mappings
+	 * - Inherit from confliting mappings otherwise
+	 */
+	err = reserve_memtype(phys_addr, phys_addr + size, -1, &flags);
+	if (err < 0)
+		return NULL;
+
+	ret = (void *) __ioremap_caller(phys_addr, size, flags,
+					__builtin_return_address(0));
+
+	free_memtype(phys_addr, phys_addr + size);
+	return (void __iomem *)ret;
+}
+
 /**
  * iounmap - Free a IO remapping
  * @addr: virtual address from ioremap_*
@@ -365,7 +388,7 @@ void *xlate_dev_mem_ptr(unsigned long phys)
 	if (page_is_ram(start >> PAGE_SHIFT))
 		return __va(phys);
 
-	addr = (void *)ioremap(start, PAGE_SIZE);
+	addr = (void *)ioremap_default(start, PAGE_SIZE);
 	if (addr)
 		addr = (void *)((unsigned long)addr | (phys & ~PAGE_MASK));
 
diff --git a/arch/x86/pci/common.c b/arch/x86/pci/common.c
index 940185ecaed..6e64aaf00d1 100644
--- a/arch/x86/pci/common.c
+++ b/arch/x86/pci/common.c
@@ -328,18 +328,18 @@ static struct dmi_system_id __devinitdata pciprobe_dmi_table[] = {
 #endif
 	{
 		.callback = set_bf_sort,
-		.ident = "HP ProLiant DL360",
+		.ident = "HP ProLiant DL385 G2",
 		.matches = {
 			DMI_MATCH(DMI_SYS_VENDOR, "HP"),
-			DMI_MATCH(DMI_PRODUCT_NAME, "ProLiant DL360"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "ProLiant DL385 G2"),
 		},
 	},
 	{
 		.callback = set_bf_sort,
-		.ident = "HP ProLiant DL380",
+		.ident = "HP ProLiant DL585 G2",
 		.matches = {
 			DMI_MATCH(DMI_SYS_VENDOR, "HP"),
-			DMI_MATCH(DMI_PRODUCT_NAME, "ProLiant DL380"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "ProLiant DL585 G2"),
 		},
 	},
 	{}
diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c
index df40bf74ea7..4e527e7893a 100644
--- a/arch/x86/xen/mmu.c
+++ b/arch/x86/xen/mmu.c
@@ -185,7 +185,7 @@ static pteval_t pte_mfn_to_pfn(pteval_t val)
 	if (val & _PAGE_PRESENT) {
 		unsigned long mfn = (val & PTE_MASK) >> PAGE_SHIFT;
 		pteval_t flags = val & ~PTE_MASK;
-		val = (mfn_to_pfn(mfn) << PAGE_SHIFT) | flags;
+		val = ((pteval_t)mfn_to_pfn(mfn) << PAGE_SHIFT) | flags;
 	}
 
 	return val;
@@ -196,7 +196,7 @@ static pteval_t pte_pfn_to_mfn(pteval_t val)
 	if (val & _PAGE_PRESENT) {
 		unsigned long pfn = (val & PTE_MASK) >> PAGE_SHIFT;
 		pteval_t flags = val & ~PTE_MASK;
-		val = (pfn_to_mfn(pfn) << PAGE_SHIFT) | flags;
+		val = ((pteval_t)pfn_to_mfn(pfn) << PAGE_SHIFT) | flags;
 	}
 
 	return val;