21 files changed, 186 insertions, 64 deletions

diff --git a/arch/arm/plat-samsung/dev-hsmmc.c b/arch/arm/plat-samsung/dev-hsmmc.c
index b0f93f11e281..9d2be0941410 100644
--- a/arch/arm/plat-samsung/dev-hsmmc.c
+++ b/arch/arm/plat-samsung/dev-hsmmc.c
@@ -70,4 +70,6 @@ void s3c_sdhci0_set_platdata(struct s3c_sdhci_platdata *pd)
 		set->cfg_gpio = pd->cfg_gpio;
 	if (pd->cfg_card)
 		set->cfg_card = pd->cfg_card;
+	if (pd->host_caps)
+		set->host_caps = pd->host_caps;
 }
diff --git a/arch/arm/plat-samsung/dev-hsmmc1.c b/arch/arm/plat-samsung/dev-hsmmc1.c
index 1504fd802865..a6c8295840af 100644
--- a/arch/arm/plat-samsung/dev-hsmmc1.c
+++ b/arch/arm/plat-samsung/dev-hsmmc1.c
@@ -70,4 +70,6 @@ void s3c_sdhci1_set_platdata(struct s3c_sdhci_platdata *pd)
 		set->cfg_gpio = pd->cfg_gpio;
 	if (pd->cfg_card)
 		set->cfg_card = pd->cfg_card;
+	if (pd->host_caps)
+		set->host_caps = pd->host_caps;
 }
diff --git a/arch/arm/plat-samsung/dev-hsmmc2.c b/arch/arm/plat-samsung/dev-hsmmc2.c
index b28ef173444d..cb0d7143381a 100644
--- a/arch/arm/plat-samsung/dev-hsmmc2.c
+++ b/arch/arm/plat-samsung/dev-hsmmc2.c
@@ -71,4 +71,6 @@ void s3c_sdhci2_set_platdata(struct s3c_sdhci_platdata *pd)
 		set->cfg_gpio = pd->cfg_gpio;
 	if (pd->cfg_card)
 		set->cfg_card = pd->cfg_card;
+	if (pd->host_caps)
+		set->host_caps = pd->host_caps;
 }
diff --git a/arch/ia64/include/asm/unistd.h b/arch/ia64/include/asm/unistd.h
index 87f1bd1efc82..954d398a54b4 100644
--- a/arch/ia64/include/asm/unistd.h
+++ b/arch/ia64/include/asm/unistd.h
@@ -356,8 +356,6 @@ asmlinkage unsigned long sys_mmap2(
 				int fd, long pgoff);
 struct pt_regs;
 struct sigaction;
-long sys_execve(const char __user *filename, char __user * __user *argv,
-			   char __user * __user *envp, struct pt_regs *regs);
 asmlinkage long sys_ia64_pipe(void);
 asmlinkage long sys_rt_sigaction(int sig,
 				 const struct sigaction __user *act,
diff --git a/arch/m68k/include/asm/ide.h b/arch/m68k/include/asm/ide.h
index 3958726664ba..492fee8a1ab2 100644
--- a/arch/m68k/include/asm/ide.h
+++ b/arch/m68k/include/asm/ide.h
@@ -1,6 +1,4 @@
 /*
- *  linux/include/asm-m68k/ide.h
- *
  *  Copyright (C) 1994-1996  Linus Torvalds & authors
  */
 
@@ -34,6 +32,8 @@
 #include <asm/io.h>
 #include <asm/irq.h>
 
+#ifdef CONFIG_MMU
+
 /*
  * Get rid of defs from io.h - ide has its private and conflicting versions
  * Since so far no single m68k platform uses ISA/PCI I/O space for IDE, we
@@ -53,5 +53,14 @@
 #define __ide_mm_outsw(port, addr, n)	raw_outsw((u16 *)port, addr, n)
 #define __ide_mm_outsl(port, addr, n)	raw_outsl((u32 *)port, addr, n)
 
+#else
+
+#define __ide_mm_insw(port, addr, n)	io_insw((unsigned int)port, addr, n)
+#define __ide_mm_insl(port, addr, n)	io_insl((unsigned int)port, addr, n)
+#define __ide_mm_outsw(port, addr, n)	io_outsw((unsigned int)port, addr, n)
+#define __ide_mm_outsl(port, addr, n)	io_outsl((unsigned int)port, addr, n)
+
+#endif /* CONFIG_MMU */
+
 #endif /* __KERNEL__ */
 #endif /* _M68K_IDE_H */
diff --git a/arch/m68knommu/kernel/process.c b/arch/m68knommu/kernel/process.c
index 4d090d3c0897..6d3390590e5b 100644
--- a/arch/m68knommu/kernel/process.c
+++ b/arch/m68knommu/kernel/process.c
@@ -316,14 +316,14 @@ void dump(struct pt_regs *fp)
 		fp->d0, fp->d1, fp->d2, fp->d3);
 	printk(KERN_EMERG "d4: %08lx    d5: %08lx    a0: %08lx    a1: %08lx\n",
 		fp->d4, fp->d5, fp->a0, fp->a1);
-	printk(KERN_EMERG "\nUSP: %08x   TRAPFRAME: %08x\n",
-		(unsigned int) rdusp(), (unsigned int) fp);
+	printk(KERN_EMERG "\nUSP: %08x   TRAPFRAME: %p\n",
+		(unsigned int) rdusp(), fp);
 
 	printk(KERN_EMERG "\nCODE:");
 	tp = ((unsigned char *) fp->pc) - 0x20;
 	for (sp = (unsigned long *) tp, i = 0; (i < 0x40);  i += 4) {
 		if ((i % 0x10) == 0)
-			printk(KERN_EMERG "%08x: ", (int) (tp + i));
+			printk(KERN_EMERG "%p: ", tp + i);
 		printk("%08x ", (int) *sp++);
 	}
 	printk(KERN_EMERG "\n");
@@ -332,7 +332,7 @@ void dump(struct pt_regs *fp)
 	tp = ((unsigned char *) fp) - 0x40;
 	for (sp = (unsigned long *) tp, i = 0; (i < 0xc0); i += 4) {
 		if ((i % 0x10) == 0)
-			printk(KERN_EMERG "%08x: ", (int) (tp + i));
+			printk(KERN_EMERG "%p: ", tp + i);
 		printk("%08x ", (int) *sp++);
 	}
 	printk(KERN_EMERG "\n");
@@ -341,7 +341,7 @@ void dump(struct pt_regs *fp)
 	tp = (unsigned char *) (rdusp() - 0x10);
 	for (sp = (unsigned long *) tp, i = 0; (i < 0x80); i += 4) {
 		if ((i % 0x10) == 0)
-			printk(KERN_EMERG "%08x: ", (int) (tp + i));
+			printk(KERN_EMERG "%p: ", tp + i);
 		printk("%08x ", (int) *sp++);
 	}
 	printk(KERN_EMERG "\n");
diff --git a/arch/microblaze/kernel/prom_parse.c b/arch/microblaze/kernel/prom_parse.c
index d33ba17601fa..99d9b61cccb5 100644
--- a/arch/microblaze/kernel/prom_parse.c
+++ b/arch/microblaze/kernel/prom_parse.c
@@ -73,7 +73,7 @@ int of_irq_map_pci(struct pci_dev *pdev, struct of_irq *out_irq)
 		/* We can only get here if we hit a P2P bridge with no node,
 		 * let's do standard swizzling and try again
 		 */
-		lspec = of_irq_pci_swizzle(PCI_SLOT(pdev->devfn), lspec);
+		lspec = pci_swizzle_interrupt_pin(pdev, lspec);
 		pdev = ppdev;
 	}
 
diff --git a/arch/microblaze/pci/pci-common.c b/arch/microblaze/pci/pci-common.c
index 23be25fec4d6..55ef532f32be 100644
--- a/arch/microblaze/pci/pci-common.c
+++ b/arch/microblaze/pci/pci-common.c
@@ -27,10 +27,11 @@
 #include <linux/irq.h>
 #include <linux/vmalloc.h>
 #include <linux/slab.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
 
 #include <asm/processor.h>
 #include <asm/io.h>
-#include <asm/prom.h>
 #include <asm/pci-bridge.h>
 #include <asm/byteorder.h>
 
@@ -1077,7 +1078,7 @@ void __devinit pcibios_setup_bus_devices(struct pci_bus *bus)
 		struct dev_archdata *sd = &dev->dev.archdata;
 
 		/* Setup OF node pointer in archdata */
-		sd->of_node = pci_device_to_OF_node(dev);
+		dev->dev.of_node = pci_device_to_OF_node(dev);
 
 		/* Fixup NUMA node as it may not be setup yet by the generic
 		 * code and is needed by the DMA init
diff --git a/arch/microblaze/pci/xilinx_pci.c b/arch/microblaze/pci/xilinx_pci.c
index 7869a41b0f94..0687a42a5bd4 100644
--- a/arch/microblaze/pci/xilinx_pci.c
+++ b/arch/microblaze/pci/xilinx_pci.c
@@ -16,6 +16,7 @@
 
 #include <linux/ioport.h>
 #include <linux/of.h>
+#include <linux/of_address.h>
 #include <linux/pci.h>
 #include <asm/io.h>
 
diff --git a/arch/um/include/asm/dma-mapping.h b/arch/um/include/asm/dma-mapping.h
index 17a2cb5a4178..1f469e80fdd3 100644
--- a/arch/um/include/asm/dma-mapping.h
+++ b/arch/um/include/asm/dma-mapping.h
@@ -95,13 +95,6 @@ dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg, int nelems,
 #define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f)
 #define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h)
 
-static inline int
-dma_get_cache_alignment(void)
-{
-	BUG();
-	return(0);
-}
-
 static inline void
 dma_cache_sync(struct device *dev, void *vaddr, size_t size,
 	       enum dma_data_direction direction)
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index a84fc34c8f77..cea0cd9a316f 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -245,6 +245,11 @@ config ARCH_HWEIGHT_CFLAGS
 
 config KTIME_SCALAR
 	def_bool X86_32
+
+config ARCH_CPU_PROBE_RELEASE
+	def_bool y
+	depends on HOTPLUG_CPU
+
 source "init/Kconfig"
 source "kernel/Kconfig.freezer"
 
@@ -749,11 +754,11 @@ config IOMMU_API
 	def_bool (AMD_IOMMU || DMAR)
 
 config MAXSMP
-	bool "Configure Maximum number of SMP Processors and NUMA Nodes"
+	bool "Enable Maximum number of SMP Processors and NUMA Nodes"
 	depends on X86_64 && SMP && DEBUG_KERNEL && EXPERIMENTAL
 	select CPUMASK_OFFSTACK
 	---help---
-	  Configure maximum number of CPUS and NUMA Nodes for this architecture.
+	  Enable maximum number of CPUS and NUMA Nodes for this architecture.
 	  If unsure, say N.
 
 config NR_CPUS
diff --git a/arch/x86/include/asm/pgtable_32.h b/arch/x86/include/asm/pgtable_32.h
index 2984a25ff383..f686f49e8b7b 100644
--- a/arch/x86/include/asm/pgtable_32.h
+++ b/arch/x86/include/asm/pgtable_32.h
@@ -26,6 +26,7 @@ struct mm_struct;
 struct vm_area_struct;
 
 extern pgd_t swapper_pg_dir[1024];
+extern pgd_t trampoline_pg_dir[1024];
 
 static inline void pgtable_cache_init(void) { }
 static inline void check_pgt_cache(void) { }
diff --git a/arch/x86/include/asm/trampoline.h b/arch/x86/include/asm/trampoline.h
index cb507bb05d79..4dde797c0578 100644
--- a/arch/x86/include/asm/trampoline.h
+++ b/arch/x86/include/asm/trampoline.h
@@ -13,14 +13,17 @@ extern unsigned char *trampoline_base;
 
 extern unsigned long init_rsp;
 extern unsigned long initial_code;
+extern unsigned long initial_page_table;
 extern unsigned long initial_gs;
 
 #define TRAMPOLINE_SIZE roundup(trampoline_end - trampoline_data, PAGE_SIZE)
 
 extern unsigned long setup_trampoline(void);
+extern void __init setup_trampoline_page_table(void);
 extern void __init reserve_trampoline_memory(void);
 #else
-static inline void reserve_trampoline_memory(void) {};
+static inline void setup_trampoline_page_table(void) {}
+static inline void reserve_trampoline_memory(void) {}
 #endif /* CONFIG_X86_TRAMPOLINE */
 
 #endif /* __ASSEMBLY__ */
diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index 4dc0084ec1b1..f1efebaf5510 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -1728,6 +1728,8 @@ __apicdebuginit(void) print_IO_APIC(void)
 		struct irq_pin_list *entry;
 
 		cfg = desc->chip_data;
+		if (!cfg)
+			continue;
 		entry = cfg->irq_2_pin;
 		if (!entry)
 			continue;
diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index 60a57b13082d..ba5f62f45f01 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -669,7 +669,7 @@ bool cpu_has_amd_erratum(const int *erratum)
 	}
 
 	/* OSVW unavailable or ID unknown, match family-model-stepping range */
-	ms = (cpu->x86_model << 8) | cpu->x86_mask;
+	ms = (cpu->x86_model << 4) | cpu->x86_mask;
 	while ((range = *erratum++))
 		if ((cpu->x86 == AMD_MODEL_RANGE_FAMILY(range)) &&
 		    (ms >= AMD_MODEL_RANGE_START(range)) &&
diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c
index 214ac860ebe0..d8d86d014008 100644
--- a/arch/x86/kernel/cpu/perf_event_intel.c
+++ b/arch/x86/kernel/cpu/perf_event_intel.c
@@ -491,33 +491,78 @@ static void intel_pmu_enable_all(int added)
  *   Intel Errata AAP53  (model 30)
  *   Intel Errata BD53   (model 44)
  *
- * These chips need to be 'reset' when adding counters by programming
- * the magic three (non counting) events 0x4300D2, 0x4300B1 and 0x4300B5
- * either in sequence on the same PMC or on different PMCs.
+ * The official story:
+ *   These chips need to be 'reset' when adding counters by programming the
+ *   magic three (non-counting) events 0x4300B5, 0x4300D2, and 0x4300B1 either
+ *   in sequence on the same PMC or on different PMCs.
+ *
+ * In practise it appears some of these events do in fact count, and
+ * we need to programm all 4 events.
  */
-static void intel_pmu_nhm_enable_all(int added)
+static void intel_pmu_nhm_workaround(void)
 {
-	if (added) {
-		struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
-		int i;
+	struct cpu_hw_events *cpuc = &__get_cpu_var(cpu_hw_events);
+	static const unsigned long nhm_magic[4] = {
+		0x4300B5,
+		0x4300D2,
+		0x4300B1,
+		0x4300B1
+	};
+	struct perf_event *event;
+	int i;
+
+	/*
+	 * The Errata requires below steps:
+	 * 1) Clear MSR_IA32_PEBS_ENABLE and MSR_CORE_PERF_GLOBAL_CTRL;
+	 * 2) Configure 4 PERFEVTSELx with the magic events and clear
+	 *    the corresponding PMCx;
+	 * 3) set bit0~bit3 of MSR_CORE_PERF_GLOBAL_CTRL;
+	 * 4) Clear MSR_CORE_PERF_GLOBAL_CTRL;
+	 * 5) Clear 4 pairs of ERFEVTSELx and PMCx;
+	 */
+
+	/*
+	 * The real steps we choose are a little different from above.
+	 * A) To reduce MSR operations, we don't run step 1) as they
+	 *    are already cleared before this function is called;
+	 * B) Call x86_perf_event_update to save PMCx before configuring
+	 *    PERFEVTSELx with magic number;
+	 * C) With step 5), we do clear only when the PERFEVTSELx is
+	 *    not used currently.
+	 * D) Call x86_perf_event_set_period to restore PMCx;
+	 */
 
-		wrmsrl(MSR_ARCH_PERFMON_EVENTSEL0 + 0, 0x4300D2);
-		wrmsrl(MSR_ARCH_PERFMON_EVENTSEL0 + 1, 0x4300B1);
-		wrmsrl(MSR_ARCH_PERFMON_EVENTSEL0 + 2, 0x4300B5);
+	/* We always operate 4 pairs of PERF Counters */
+	for (i = 0; i < 4; i++) {
+		event = cpuc->events[i];
+		if (event)
+			x86_perf_event_update(event);
+	}
 
-		wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0x3);
-		wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0x0);
+	for (i = 0; i < 4; i++) {
+		wrmsrl(MSR_ARCH_PERFMON_EVENTSEL0 + i, nhm_magic[i]);
+		wrmsrl(MSR_ARCH_PERFMON_PERFCTR0 + i, 0x0);
+	}
 
-		for (i = 0; i < 3; i++) {
-			struct perf_event *event = cpuc->events[i];
+	wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0xf);
+	wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0x0);
 
-			if (!event)
-				continue;
+	for (i = 0; i < 4; i++) {
+		event = cpuc->events[i];
 
+		if (event) {
+			x86_perf_event_set_period(event);
 			__x86_pmu_enable_event(&event->hw,
-					       ARCH_PERFMON_EVENTSEL_ENABLE);
-		}
+					ARCH_PERFMON_EVENTSEL_ENABLE);
+		} else
+			wrmsrl(MSR_ARCH_PERFMON_EVENTSEL0 + i, 0x0);
 	}
+}
+
+static void intel_pmu_nhm_enable_all(int added)
+{
+	if (added)
+		intel_pmu_nhm_workaround();
 	intel_pmu_enable_all(added);
 }
 
diff --git a/arch/x86/kernel/head_32.S b/arch/x86/kernel/head_32.S
index ff4c453e13f3..fa8c1b8e09fb 100644
--- a/arch/x86/kernel/head_32.S
+++ b/arch/x86/kernel/head_32.S
@@ -334,7 +334,7 @@ ENTRY(startup_32_smp)
 /*
  * Enable paging
  */
-	movl $pa(swapper_pg_dir),%eax
+	movl pa(initial_page_table), %eax
 	movl %eax,%cr3		/* set the page table pointer.. */
 	movl %cr0,%eax
 	orl  $X86_CR0_PG,%eax
@@ -614,6 +614,8 @@ ignore_int:
 .align 4
 ENTRY(initial_code)
 	.long i386_start_kernel
+ENTRY(initial_page_table)
+	.long pa(swapper_pg_dir)
 
 /*
  * BSS section
@@ -629,6 +631,10 @@ ENTRY(swapper_pg_dir)
 #endif
 swapper_pg_fixmap:
 	.fill 1024,4,0
+#ifdef CONFIG_X86_TRAMPOLINE
+ENTRY(trampoline_pg_dir)
+	.fill 1024,4,0
+#endif
 ENTRY(empty_zero_page)
 	.fill 4096,1,0
 
diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c
index 1bfb6cf4dd55..770ebfb349e9 100644
--- a/arch/x86/kernel/kprobes.c
+++ b/arch/x86/kernel/kprobes.c
@@ -709,6 +709,7 @@ static __used __kprobes void *trampoline_handler(struct pt_regs *regs)
 	struct hlist_node *node, *tmp;
 	unsigned long flags, orig_ret_address = 0;
 	unsigned long trampoline_address = (unsigned long)&kretprobe_trampoline;
+	kprobe_opcode_t *correct_ret_addr = NULL;
 
 	INIT_HLIST_HEAD(&empty_rp);
 	kretprobe_hash_lock(current, &head, &flags);
@@ -740,14 +741,34 @@ static __used __kprobes void *trampoline_handler(struct pt_regs *regs)
 			/* another task is sharing our hash bucket */
 			continue;
 
+		orig_ret_address = (unsigned long)ri->ret_addr;
+
+		if (orig_ret_address != trampoline_address)
+			/*
+			 * This is the real return address. Any other
+			 * instances associated with this task are for
+			 * other calls deeper on the call stack
+			 */
+			break;
+	}
+
+	kretprobe_assert(ri, orig_ret_address, trampoline_address);
+
+	correct_ret_addr = ri->ret_addr;
+	hlist_for_each_entry_safe(ri, node, tmp, head, hlist) {
+		if (ri->task != current)
+			/* another task is sharing our hash bucket */
+			continue;
+
+		orig_ret_address = (unsigned long)ri->ret_addr;
 		if (ri->rp && ri->rp->handler) {
 			__get_cpu_var(current_kprobe) = &ri->rp->kp;
 			get_kprobe_ctlblk()->kprobe_status = KPROBE_HIT_ACTIVE;
+			ri->ret_addr = correct_ret_addr;
 			ri->rp->handler(ri, regs);
 			__get_cpu_var(current_kprobe) = NULL;
 		}
 
-		orig_ret_address = (unsigned long)ri->ret_addr;
 		recycle_rp_inst(ri, &empty_rp);
 
 		if (orig_ret_address != trampoline_address)
@@ -759,8 +780,6 @@ static __used __kprobes void *trampoline_handler(struct pt_regs *regs)
 			break;
 	}
 
-	kretprobe_assert(ri, orig_ret_address, trampoline_address);
-
 	kretprobe_hash_unlock(current, &flags);
 
 	hlist_for_each_entry_safe(ri, node, tmp, &empty_rp, hlist) {
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index b008e7883207..c3a4fbb2b996 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -1014,6 +1014,8 @@ void __init setup_arch(char **cmdline_p)
 	paging_init();
 	x86_init.paging.pagetable_setup_done(swapper_pg_dir);
 
+	setup_trampoline_page_table();
+
 	tboot_probe();
 
 #ifdef CONFIG_X86_64
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index a5e928b0cb5f..8b3bfc4dd708 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -73,7 +73,6 @@
 
 #ifdef CONFIG_X86_32
 u8 apicid_2_node[MAX_APICID];
-static int low_mappings;
 #endif
 
 /* State of each CPU */
@@ -91,6 +90,25 @@ DEFINE_PER_CPU(int, cpu_state) = { 0 };
 static DEFINE_PER_CPU(struct task_struct *, idle_thread_array);
 #define get_idle_for_cpu(x)      (per_cpu(idle_thread_array, x))
 #define set_idle_for_cpu(x, p)   (per_cpu(idle_thread_array, x) = (p))
+
+/*
+ * We need this for trampoline_base protection from concurrent accesses when
+ * off- and onlining cores wildly.
+ */
+static DEFINE_MUTEX(x86_cpu_hotplug_driver_mutex);
+
+void cpu_hotplug_driver_lock()
+{
+        mutex_lock(&x86_cpu_hotplug_driver_mutex);
+}
+
+void cpu_hotplug_driver_unlock()
+{
+        mutex_unlock(&x86_cpu_hotplug_driver_mutex);
+}
+
+ssize_t arch_cpu_probe(const char *buf, size_t count) { return -1; }
+ssize_t arch_cpu_release(const char *buf, size_t count) { return -1; }
 #else
 static struct task_struct *idle_thread_array[NR_CPUS] __cpuinitdata ;
 #define get_idle_for_cpu(x)      (idle_thread_array[(x)])
@@ -281,6 +299,18 @@ notrace static void __cpuinit start_secondary(void *unused)
 	 * fragile that we want to limit the things done here to the
 	 * most necessary things.
 	 */
+
+#ifdef CONFIG_X86_32
+	/*
+	 * Switch away from the trampoline page-table
+	 *
+	 * Do this before cpu_init() because it needs to access per-cpu
+	 * data which may not be mapped in the trampoline page-table.
+	 */
+	load_cr3(swapper_pg_dir);
+	__flush_tlb_all();
+#endif
+
 	vmi_bringup();
 	cpu_init();
 	preempt_disable();
@@ -299,12 +329,6 @@ notrace static void __cpuinit start_secondary(void *unused)
 		legacy_pic->chip->unmask(0);
 	}
 
-#ifdef CONFIG_X86_32
-	while (low_mappings)
-		cpu_relax();
-	__flush_tlb_all();
-#endif
-
 	/* This must be done before setting cpu_online_mask */
 	set_cpu_sibling_map(raw_smp_processor_id());
 	wmb();
@@ -750,6 +774,7 @@ do_rest:
 #ifdef CONFIG_X86_32
 	/* Stack for startup_32 can be just as for start_secondary onwards */
 	irq_ctx_init(cpu);
+	initial_page_table = __pa(&trampoline_pg_dir);
 #else
 	clear_tsk_thread_flag(c_idle.idle, TIF_FORK);
 	initial_gs = per_cpu_offset(cpu);
@@ -897,20 +922,8 @@ int __cpuinit native_cpu_up(unsigned int cpu)
 
 	per_cpu(cpu_state, cpu) = CPU_UP_PREPARE;
 
-#ifdef CONFIG_X86_32
-	/* init low mem mapping */
-	clone_pgd_range(swapper_pg_dir, swapper_pg_dir + KERNEL_PGD_BOUNDARY,
-		min_t(unsigned long, KERNEL_PGD_PTRS, KERNEL_PGD_BOUNDARY));
-	flush_tlb_all();
-	low_mappings = 1;
-
 	err = do_boot_cpu(apicid, cpu);
 
-	zap_low_mappings(false);
-	low_mappings = 0;
-#else
-	err = do_boot_cpu(apicid, cpu);
-#endif
 	if (err) {
 		pr_debug("do_boot_cpu failed %d\n", err);
 		return -EIO;
diff --git a/arch/x86/kernel/trampoline.c b/arch/x86/kernel/trampoline.c
index c652ef62742d..a874495b3673 100644
--- a/arch/x86/kernel/trampoline.c
+++ b/arch/x86/kernel/trampoline.c
@@ -1,6 +1,7 @@
 #include <linux/io.h>
 
 #include <asm/trampoline.h>
+#include <asm/pgtable.h>
 #include <asm/e820.h>
 
 #if defined(CONFIG_X86_64) && defined(CONFIG_ACPI_SLEEP)
@@ -37,3 +38,20 @@ unsigned long __trampinit setup_trampoline(void)
 	memcpy(trampoline_base, trampoline_data, TRAMPOLINE_SIZE);
 	return virt_to_phys(trampoline_base);
 }
+
+void __init setup_trampoline_page_table(void)
+{
+#ifdef CONFIG_X86_32
+	/* Copy kernel address range */
+	clone_pgd_range(trampoline_pg_dir + KERNEL_PGD_BOUNDARY,
+			swapper_pg_dir + KERNEL_PGD_BOUNDARY,
+			min_t(unsigned long, KERNEL_PGD_PTRS,
+			      KERNEL_PGD_BOUNDARY));
+
+	/* Initialize low mappings */
+	clone_pgd_range(trampoline_pg_dir,
+			swapper_pg_dir + KERNEL_PGD_BOUNDARY,
+			min_t(unsigned long, KERNEL_PGD_PTRS,
+			      KERNEL_PGD_BOUNDARY));
+#endif
+}