25 files changed, 132 insertions, 135 deletions

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 123b7924904f..995b55311c55 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -20,7 +20,6 @@ config ARM
 	select GENERIC_STRNCPY_FROM_USER
 	select GENERIC_STRNLEN_USER
 	select HARDIRQS_SW_RESEND
-	select HAVE_AOUT
 	select HAVE_ARCH_JUMP_LABEL if !XIP_KERNEL
 	select HAVE_ARCH_KGDB
 	select HAVE_ARCH_SECCOMP_FILTER
diff --git a/arch/arm/Kconfig.debug b/arch/arm/Kconfig.debug
index e401a766c0bd..583f4a00ec32 100644
--- a/arch/arm/Kconfig.debug
+++ b/arch/arm/Kconfig.debug
@@ -804,9 +804,19 @@ config DEBUG_LL_INCLUDE
 
 config DEBUG_UNCOMPRESS
 	bool
-	default y if ARCH_MULTIPLATFORM && DEBUG_LL && \
-		     !DEBUG_OMAP2PLUS_UART && \
+	depends on ARCH_MULTIPLATFORM
+	default y if DEBUG_LL && !DEBUG_OMAP2PLUS_UART && \
 		     !DEBUG_TEGRA_UART
+	help
+	  This option influences the normal decompressor output for
+	  multiplatform kernels.  Normally, multiplatform kernels disable
+	  decompressor output because it is not possible to know where to
+	  send the decompressor output.
+
+	  When this option is set, the selected DEBUG_LL output method
+	  will be re-used for normal decompressor output on multiplatform
+	  kernels.
+	  
 
 config UNCOMPRESS_INCLUDE
 	string
diff --git a/arch/arm/Makefile b/arch/arm/Makefile
index c0ac0f5e5e5c..6fd2ceae305a 100644
--- a/arch/arm/Makefile
+++ b/arch/arm/Makefile
@@ -153,6 +153,7 @@ machine-$(CONFIG_ARCH_DAVINCI)		+= davinci
 machine-$(CONFIG_ARCH_DOVE)		+= dove
 machine-$(CONFIG_ARCH_EBSA110)		+= ebsa110
 machine-$(CONFIG_ARCH_EP93XX)		+= ep93xx
+machine-$(CONFIG_ARCH_EXYNOS)		+= exynos
 machine-$(CONFIG_ARCH_GEMINI)		+= gemini
 machine-$(CONFIG_ARCH_HIGHBANK)		+= highbank
 machine-$(CONFIG_ARCH_INTEGRATOR)	+= integrator
@@ -160,15 +161,16 @@ machine-$(CONFIG_ARCH_IOP13XX)		+= iop13xx
 machine-$(CONFIG_ARCH_IOP32X)		+= iop32x
 machine-$(CONFIG_ARCH_IOP33X)		+= iop33x
 machine-$(CONFIG_ARCH_IXP4XX)		+= ixp4xx
+machine-$(CONFIG_ARCH_KEYSTONE)		+= keystone
 machine-$(CONFIG_ARCH_KIRKWOOD)		+= kirkwood
 machine-$(CONFIG_ARCH_KS8695)		+= ks8695
 machine-$(CONFIG_ARCH_LPC32XX)		+= lpc32xx
 machine-$(CONFIG_ARCH_MMP)		+= mmp
 machine-$(CONFIG_ARCH_MSM)		+= msm
 machine-$(CONFIG_ARCH_MV78XX0)		+= mv78xx0
+machine-$(CONFIG_ARCH_MVEBU)		+= mvebu
 machine-$(CONFIG_ARCH_MXC)		+= imx
 machine-$(CONFIG_ARCH_MXS)		+= mxs
-machine-$(CONFIG_ARCH_MVEBU)		+= mvebu
 machine-$(CONFIG_ARCH_NETX)		+= netx
 machine-$(CONFIG_ARCH_NOMADIK)		+= nomadik
 machine-$(CONFIG_ARCH_NSPIRE)		+= nspire
@@ -176,7 +178,6 @@ machine-$(CONFIG_ARCH_OMAP1)		+= omap1
 machine-$(CONFIG_ARCH_OMAP2PLUS)	+= omap2
 machine-$(CONFIG_ARCH_ORION5X)		+= orion5x
 machine-$(CONFIG_ARCH_PICOXCELL)	+= picoxcell
-machine-$(CONFIG_ARCH_SIRF)		+= prima2
 machine-$(CONFIG_ARCH_PXA)		+= pxa
 machine-$(CONFIG_ARCH_REALVIEW)		+= realview
 machine-$(CONFIG_ARCH_ROCKCHIP)		+= rockchip
@@ -186,25 +187,24 @@ machine-$(CONFIG_ARCH_S3C64XX)		+= s3c64xx
 machine-$(CONFIG_ARCH_S5P64X0)		+= s5p64x0
 machine-$(CONFIG_ARCH_S5PC100)		+= s5pc100
 machine-$(CONFIG_ARCH_S5PV210)		+= s5pv210
-machine-$(CONFIG_ARCH_EXYNOS)		+= exynos
 machine-$(CONFIG_ARCH_SA1100)		+= sa1100
 machine-$(CONFIG_ARCH_SHARK)		+= shark
 machine-$(CONFIG_ARCH_SHMOBILE) 	+= shmobile
+machine-$(CONFIG_ARCH_SIRF)		+= prima2
+machine-$(CONFIG_ARCH_SOCFPGA)		+= socfpga
+machine-$(CONFIG_ARCH_STI)		+= sti
+machine-$(CONFIG_ARCH_SUNXI)		+= sunxi
 machine-$(CONFIG_ARCH_TEGRA)		+= tegra
 machine-$(CONFIG_ARCH_U300)		+= u300
 machine-$(CONFIG_ARCH_U8500)		+= ux500
 machine-$(CONFIG_ARCH_VERSATILE)	+= versatile
 machine-$(CONFIG_ARCH_VEXPRESS)		+= vexpress
+machine-$(CONFIG_ARCH_VIRT)		+= virt
 machine-$(CONFIG_ARCH_VT8500)		+= vt8500
 machine-$(CONFIG_ARCH_W90X900)		+= w90x900
+machine-$(CONFIG_ARCH_ZYNQ)		+= zynq
 machine-$(CONFIG_FOOTBRIDGE)		+= footbridge
-machine-$(CONFIG_ARCH_SOCFPGA)		+= socfpga
 machine-$(CONFIG_PLAT_SPEAR)		+= spear
-machine-$(CONFIG_ARCH_STI)		+= sti
-machine-$(CONFIG_ARCH_VIRT)		+= virt
-machine-$(CONFIG_ARCH_ZYNQ)		+= zynq
-machine-$(CONFIG_ARCH_SUNXI)		+= sunxi
-machine-$(CONFIG_ARCH_KEYSTONE)		+= keystone
 
 # Platform directory name.  This list is sorted alphanumerically
 # by CONFIG_* macro name.
diff --git a/arch/arm/include/asm/a.out-core.h b/arch/arm/include/asm/a.out-core.h
deleted file mode 100644
index 92f10cb5c70c..000000000000
--- a/arch/arm/include/asm/a.out-core.h
+++ /dev/null
@@ -1,45 +0,0 @@
-/* a.out coredump register dumper
- *
- * Copyright (C) 2007 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 Licence
- * as published by the Free Software Foundation; either version
- * 2 of the Licence, or (at your option) any later version.
- */
-
-#ifndef _ASM_A_OUT_CORE_H
-#define _ASM_A_OUT_CORE_H
-
-#ifdef __KERNEL__
-
-#include <linux/user.h>
-#include <linux/elfcore.h>
-
-/*
- * fill in the user structure for an a.out core dump
- */
-static inline void aout_dump_thread(struct pt_regs *regs, struct user *dump)
-{
-	struct task_struct *tsk = current;
-
-	dump->magic = CMAGIC;
-	dump->start_code = tsk->mm->start_code;
-	dump->start_stack = regs->ARM_sp & ~(PAGE_SIZE - 1);
-
-	dump->u_tsize = (tsk->mm->end_code - tsk->mm->start_code) >> PAGE_SHIFT;
-	dump->u_dsize = (tsk->mm->brk - tsk->mm->start_data + PAGE_SIZE - 1) >> PAGE_SHIFT;
-	dump->u_ssize = 0;
-
-	memset(dump->u_debugreg, 0, sizeof(dump->u_debugreg));
-
-	if (dump->start_stack < 0x04000000)
-		dump->u_ssize = (0x04000000 - dump->start_stack) >> PAGE_SHIFT;
-
-	dump->regs = *regs;
-	dump->u_fpvalid = dump_fpu (regs, &dump->u_fp);
-}
-
-#endif /* __KERNEL__ */
-#endif /* _ASM_A_OUT_CORE_H */
diff --git a/arch/arm/include/asm/cputype.h b/arch/arm/include/asm/cputype.h
index 8c25dc4e9851..9672e978d50d 100644
--- a/arch/arm/include/asm/cputype.h
+++ b/arch/arm/include/asm/cputype.h
@@ -89,13 +89,18 @@ extern unsigned int processor_id;
 		__val;							\
 	})
 
+/*
+ * The memory clobber prevents gcc 4.5 from reordering the mrc before
+ * any is_smp() tests, which can cause undefined instruction aborts on
+ * ARM1136 r0 due to the missing extended CP15 registers.
+ */
 #define read_cpuid_ext(ext_reg)						\
 	({								\
 		unsigned int __val;					\
 		asm("mrc	p15, 0, %0, c0, " ext_reg		\
 		    : "=r" (__val)					\
 		    :							\
-		    : "cc");						\
+		    : "memory");					\
 		__val;							\
 	})
 
diff --git a/arch/arm/include/asm/mmu.h b/arch/arm/include/asm/mmu.h
index 7345e37155d5..6f18da09668b 100644
--- a/arch/arm/include/asm/mmu.h
+++ b/arch/arm/include/asm/mmu.h
@@ -6,6 +6,8 @@
 typedef struct {
 #ifdef CONFIG_CPU_HAS_ASID
 	atomic64_t	id;
+#else
+	int		switch_pending;
 #endif
 	unsigned int	vmalloc_seq;
 	unsigned long	sigpage;
diff --git a/arch/arm/include/asm/mmu_context.h b/arch/arm/include/asm/mmu_context.h
index b5792b7fd8d3..9b32f76bb0dd 100644
--- a/arch/arm/include/asm/mmu_context.h
+++ b/arch/arm/include/asm/mmu_context.h
@@ -56,7 +56,7 @@ static inline void check_and_switch_context(struct mm_struct *mm,
 		 * on non-ASID CPUs, the old mm will remain valid until the
 		 * finish_arch_post_lock_switch() call.
 		 */
-		set_ti_thread_flag(task_thread_info(tsk), TIF_SWITCH_MM);
+		mm->context.switch_pending = 1;
 	else
 		cpu_switch_mm(mm->pgd, mm);
 }
@@ -65,9 +65,21 @@ static inline void check_and_switch_context(struct mm_struct *mm,
 	finish_arch_post_lock_switch
 static inline void finish_arch_post_lock_switch(void)
 {
-	if (test_and_clear_thread_flag(TIF_SWITCH_MM)) {
-		struct mm_struct *mm = current->mm;
-		cpu_switch_mm(mm->pgd, mm);
+	struct mm_struct *mm = current->mm;
+
+	if (mm && mm->context.switch_pending) {
+		/*
+		 * Preemption must be disabled during cpu_switch_mm() as we
+		 * have some stateful cache flush implementations. Check
+		 * switch_pending again in case we were preempted and the
+		 * switch to this mm was already done.
+		 */
+		preempt_disable();
+		if (mm->context.switch_pending) {
+			mm->context.switch_pending = 0;
+			cpu_switch_mm(mm->pgd, mm);
+		}
+		preempt_enable_no_resched();
 	}
 }
 
diff --git a/arch/arm/include/asm/processor.h b/arch/arm/include/asm/processor.h
index 06e7d509eaac..413f3876341c 100644
--- a/arch/arm/include/asm/processor.h
+++ b/arch/arm/include/asm/processor.h
@@ -54,7 +54,6 @@ struct thread_struct {
 
 #define start_thread(regs,pc,sp)					\
 ({									\
-	unsigned long *stack = (unsigned long *)sp;			\
 	memset(regs->uregs, 0, sizeof(regs->uregs));			\
 	if (current->personality & ADDR_LIMIT_32BIT)			\
 		regs->ARM_cpsr = USR_MODE;				\
@@ -65,9 +64,6 @@ struct thread_struct {
 	regs->ARM_cpsr |= PSR_ENDSTATE;					\
 	regs->ARM_pc = pc & ~1;		/* pc */			\
 	regs->ARM_sp = sp;		/* sp */			\
-	regs->ARM_r2 = stack[2];	/* r2 (envp) */			\
-	regs->ARM_r1 = stack[1];	/* r1 (argv) */			\
-	regs->ARM_r0 = stack[0];	/* r0 (argc) */			\
 	nommu_start_thread(regs);					\
 })
 
diff --git a/arch/arm/include/asm/thread_info.h b/arch/arm/include/asm/thread_info.h
index 214d4158089a..2b8114fcba09 100644
--- a/arch/arm/include/asm/thread_info.h
+++ b/arch/arm/include/asm/thread_info.h
@@ -156,7 +156,6 @@ extern int vfp_restore_user_hwstate(struct user_vfp __user *,
 #define TIF_USING_IWMMXT	17
 #define TIF_MEMDIE		18	/* is terminating due to OOM killer */
 #define TIF_RESTORE_SIGMASK	20
-#define TIF_SWITCH_MM		22	/* deferred switch_mm */
 
 #define _TIF_SIGPENDING		(1 << TIF_SIGPENDING)
 #define _TIF_NEED_RESCHED	(1 << TIF_NEED_RESCHED)
diff --git a/arch/arm/include/asm/tlbflush.h b/arch/arm/include/asm/tlbflush.h
index fdbb9e369745..f467e9b3f8d5 100644
--- a/arch/arm/include/asm/tlbflush.h
+++ b/arch/arm/include/asm/tlbflush.h
@@ -443,7 +443,18 @@ static inline void local_flush_bp_all(void)
 		isb();
 }
 
+#include <asm/cputype.h>
 #ifdef CONFIG_ARM_ERRATA_798181
+static inline int erratum_a15_798181(void)
+{
+	unsigned int midr = read_cpuid_id();
+
+	/* Cortex-A15 r0p0..r3p2 affected */
+	if ((midr & 0xff0ffff0) != 0x410fc0f0 || midr > 0x413fc0f2)
+		return 0;
+	return 1;
+}
+
 static inline void dummy_flush_tlb_a15_erratum(void)
 {
 	/*
@@ -453,6 +464,11 @@ static inline void dummy_flush_tlb_a15_erratum(void)
 	dsb();
 }
 #else
+static inline int erratum_a15_798181(void)
+{
+	return 0;
+}
+
 static inline void dummy_flush_tlb_a15_erratum(void)
 {
 }
diff --git a/arch/arm/include/asm/virt.h b/arch/arm/include/asm/virt.h
index 50af92bac737..4371f45c5784 100644
--- a/arch/arm/include/asm/virt.h
+++ b/arch/arm/include/asm/virt.h
@@ -29,6 +29,7 @@
 #define BOOT_CPU_MODE_MISMATCH	PSR_N_BIT
 
 #ifndef __ASSEMBLY__
+#include <asm/cacheflush.h>
 
 #ifdef CONFIG_ARM_VIRT_EXT
 /*
@@ -41,10 +42,21 @@
  */
 extern int __boot_cpu_mode;
 
+static inline void sync_boot_mode(void)
+{
+	/*
+	 * As secondaries write to __boot_cpu_mode with caches disabled, we
+	 * must flush the corresponding cache entries to ensure the visibility
+	 * of their writes.
+	 */
+	sync_cache_r(&__boot_cpu_mode);
+}
+
 void __hyp_set_vectors(unsigned long phys_vector_base);
 unsigned long __hyp_get_vectors(void);
 #else
 #define __boot_cpu_mode	(SVC_MODE)
+#define sync_boot_mode()
 #endif
 
 #ifndef ZIMAGE
diff --git a/arch/arm/include/uapi/asm/Kbuild b/arch/arm/include/uapi/asm/Kbuild
index 47bcb2d254af..18d76fd5a2af 100644
--- a/arch/arm/include/uapi/asm/Kbuild
+++ b/arch/arm/include/uapi/asm/Kbuild
@@ -1,7 +1,6 @@
 # UAPI Header export list
 include include/uapi/asm-generic/Kbuild.asm
 
-header-y += a.out.h
 header-y += byteorder.h
 header-y += fcntl.h
 header-y += hwcap.h
diff --git a/arch/arm/include/uapi/asm/a.out.h b/arch/arm/include/uapi/asm/a.out.h
deleted file mode 100644
index 083894b2e3bc..000000000000
--- a/arch/arm/include/uapi/asm/a.out.h
+++ /dev/null
@@ -1,34 +0,0 @@
-#ifndef __ARM_A_OUT_H__
-#define __ARM_A_OUT_H__
-
-#include <linux/personality.h>
-#include <linux/types.h>
-
-struct exec
-{
-  __u32 a_info;		/* Use macros N_MAGIC, etc for access */
-  __u32 a_text;		/* length of text, in bytes */
-  __u32 a_data;		/* length of data, in bytes */
-  __u32 a_bss;		/* length of uninitialized data area for file, in bytes */
-  __u32 a_syms;		/* length of symbol table data in file, in bytes */
-  __u32 a_entry;	/* start address */
-  __u32 a_trsize;	/* length of relocation info for text, in bytes */
-  __u32 a_drsize;	/* length of relocation info for data, in bytes */
-};
-
-/*
- * This is always the same
- */
-#define N_TXTADDR(a)	(0x00008000)
-
-#define N_TRSIZE(a)	((a).a_trsize)
-#define N_DRSIZE(a)	((a).a_drsize)
-#define N_SYMSIZE(a)	((a).a_syms)
-
-#define M_ARM 103
-
-#ifndef LIBRARY_START_TEXT
-#define LIBRARY_START_TEXT	(0x00c00000)
-#endif
-
-#endif /* __A_OUT_GNU_H__ */
diff --git a/arch/arm/kernel/entry-v7m.S b/arch/arm/kernel/entry-v7m.S
index e00621f1403f..52b26432c9a9 100644
--- a/arch/arm/kernel/entry-v7m.S
+++ b/arch/arm/kernel/entry-v7m.S
@@ -49,7 +49,7 @@ __irq_entry:
 	mov	r1, sp
 	stmdb	sp!, {lr}
 	@ routine called with r0 = irq number, r1 = struct pt_regs *
-	bl	nvic_do_IRQ
+	bl	nvic_handle_irq
 
 	pop	{lr}
 	@
diff --git a/arch/arm/kernel/head-nommu.S b/arch/arm/kernel/head-nommu.S
index b361de143756..14235ba64a90 100644
--- a/arch/arm/kernel/head-nommu.S
+++ b/arch/arm/kernel/head-nommu.S
@@ -87,6 +87,7 @@ ENTRY(stext)
 ENDPROC(stext)
 
 #ifdef CONFIG_SMP
+	.text
 ENTRY(secondary_startup)
 	/*
 	 * Common entry point for secondary CPUs.
diff --git a/arch/arm/kernel/head.S b/arch/arm/kernel/head.S
index 9cf6063020ae..2c7cc1e03473 100644
--- a/arch/arm/kernel/head.S
+++ b/arch/arm/kernel/head.S
@@ -343,6 +343,7 @@ __turn_mmu_on_loc:
 	.long	__turn_mmu_on_end
 
 #if defined(CONFIG_SMP)
+	.text
 ENTRY(secondary_startup)
 	/*
 	 * Common entry point for secondary CPUs.
diff --git a/arch/arm/kernel/hyp-stub.S b/arch/arm/kernel/hyp-stub.S
index 4910232c4833..797b1a6a4906 100644
--- a/arch/arm/kernel/hyp-stub.S
+++ b/arch/arm/kernel/hyp-stub.S
@@ -56,8 +56,8 @@ ENTRY(__boot_cpu_mode)
 	ldr	\reg3, [\reg2]
 	ldr	\reg1, [\reg2, \reg3]
 	cmp	\mode, \reg1		@ matches primary CPU boot mode?
-	orrne	r7, r7, #BOOT_CPU_MODE_MISMATCH
-	strne	r7, [r5, r6]		@ record what happened and give up
+	orrne	\reg1, \reg1, #BOOT_CPU_MODE_MISMATCH
+	strne	\reg1, [\reg2, \reg3]	@ record what happened and give up
 	.endm
 
 #else	/* ZIMAGE */
diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c
index d03b5bd889c5..536c85fe72a8 100644
--- a/arch/arm/kernel/process.c
+++ b/arch/arm/kernel/process.c
@@ -197,6 +197,7 @@ void machine_shutdown(void)
  */
 void machine_halt(void)
 {
+	local_irq_disable();
 	smp_send_stop();
 
 	local_irq_disable();
@@ -211,6 +212,7 @@ void machine_halt(void)
  */
 void machine_power_off(void)
 {
+	local_irq_disable();
 	smp_send_stop();
 
 	if (pm_power_off)
@@ -230,6 +232,7 @@ void machine_power_off(void)
  */
 void machine_restart(char *cmd)
 {
+	local_irq_disable();
 	smp_send_stop();
 
 	arm_pm_restart(reboot_mode, cmd);
diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c
index 63af9a7ae512..afc2489ee13b 100644
--- a/arch/arm/kernel/setup.c
+++ b/arch/arm/kernel/setup.c
@@ -836,6 +836,8 @@ static int __init meminfo_cmp(const void *_a, const void *_b)
 void __init hyp_mode_check(void)
 {
 #ifdef CONFIG_ARM_VIRT_EXT
+	sync_boot_mode();
+
 	if (is_hyp_mode_available()) {
 		pr_info("CPU: All CPU(s) started in HYP mode.\n");
 		pr_info("CPU: Virtualization extensions available.\n");
@@ -971,6 +973,7 @@ static const char *hwcap_str[] = {
 	"vfpv4",
 	"idiva",
 	"idivt",
+	"vfpd32",
 	"lpae",
 	NULL
 };
diff --git a/arch/arm/kernel/smp_tlb.c b/arch/arm/kernel/smp_tlb.c
index a98b62dca2fa..c2edfff573c2 100644
--- a/arch/arm/kernel/smp_tlb.c
+++ b/arch/arm/kernel/smp_tlb.c
@@ -70,23 +70,6 @@ static inline void ipi_flush_bp_all(void *ignored)
 	local_flush_bp_all();
 }
 
-#ifdef CONFIG_ARM_ERRATA_798181
-static int erratum_a15_798181(void)
-{
-	unsigned int midr = read_cpuid_id();
-
-	/* Cortex-A15 r0p0..r3p2 affected */
-	if ((midr & 0xff0ffff0) != 0x410fc0f0 || midr > 0x413fc0f2)
-		return 0;
-	return 1;
-}
-#else
-static int erratum_a15_798181(void)
-{
-	return 0;
-}
-#endif
-
 static void ipi_flush_tlb_a15_erratum(void *arg)
 {
 	dmb();
diff --git a/arch/arm/mm/context.c b/arch/arm/mm/context.c
index b55b1015724b..4a0544492f10 100644
--- a/arch/arm/mm/context.c
+++ b/arch/arm/mm/context.c
@@ -245,7 +245,8 @@ void check_and_switch_context(struct mm_struct *mm, struct task_struct *tsk)
 	if (cpumask_test_and_clear_cpu(cpu, &tlb_flush_pending)) {
 		local_flush_bp_all();
 		local_flush_tlb_all();
-		dummy_flush_tlb_a15_erratum();
+		if (erratum_a15_798181())
+			dummy_flush_tlb_a15_erratum();
 	}
 
 	atomic64_set(&per_cpu(active_asids, cpu), asid);
diff --git a/arch/arm/mm/mmu.c b/arch/arm/mm/mmu.c
index ca46f413d867..53cdbd39ec8e 100644
--- a/arch/arm/mm/mmu.c
+++ b/arch/arm/mm/mmu.c
@@ -989,6 +989,7 @@ phys_addr_t arm_lowmem_limit __initdata = 0;
 
 void __init sanity_check_meminfo(void)
 {
+	phys_addr_t memblock_limit = 0;
 	int i, j, highmem = 0;
 	phys_addr_t vmalloc_limit = __pa(vmalloc_min - 1) + 1;
 
@@ -1052,9 +1053,32 @@ void __init sanity_check_meminfo(void)
 			bank->size = size_limit;
 		}
 #endif
-		if (!bank->highmem && bank->start + bank->size > arm_lowmem_limit)
-			arm_lowmem_limit = bank->start + bank->size;
+		if (!bank->highmem) {
+			phys_addr_t bank_end = bank->start + bank->size;
 
+			if (bank_end > arm_lowmem_limit)
+				arm_lowmem_limit = bank_end;
+
+			/*
+			 * Find the first non-section-aligned page, and point
+			 * memblock_limit at it. This relies on rounding the
+			 * limit down to be section-aligned, which happens at
+			 * the end of this function.
+			 *
+			 * With this algorithm, the start or end of almost any
+			 * bank can be non-section-aligned. The only exception
+			 * is that the start of the bank 0 must be section-
+			 * aligned, since otherwise memory would need to be
+			 * allocated when mapping the start of bank 0, which
+			 * occurs before any free memory is mapped.
+			 */
+			if (!memblock_limit) {
+				if (!IS_ALIGNED(bank->start, SECTION_SIZE))
+					memblock_limit = bank->start;
+				else if (!IS_ALIGNED(bank_end, SECTION_SIZE))
+					memblock_limit = bank_end;
+			}
+		}
 		j++;
 	}
 #ifdef CONFIG_HIGHMEM
@@ -1079,7 +1103,18 @@ void __init sanity_check_meminfo(void)
 #endif
 	meminfo.nr_banks = j;
 	high_memory = __va(arm_lowmem_limit - 1) + 1;
-	memblock_set_current_limit(arm_lowmem_limit);
+
+	/*
+	 * Round the memblock limit down to a section size.  This
+	 * helps to ensure that we will allocate memory from the
+	 * last full section, which should be mapped.
+	 */
+	if (memblock_limit)
+		memblock_limit = round_down(memblock_limit, SECTION_SIZE);
+	if (!memblock_limit)
+		memblock_limit = arm_lowmem_limit;
+
+	memblock_set_current_limit(memblock_limit);
 }
 
 static inline void prepare_page_table(void)
@@ -1288,8 +1323,6 @@ void __init paging_init(struct machine_desc *mdesc)
 {
 	void *zero_page;
 
-	memblock_set_current_limit(arm_lowmem_limit);
-
 	build_mem_type_table();
 	prepare_page_table();
 	map_lowmem();
diff --git a/arch/arm/mm/proc-v7-2level.S b/arch/arm/mm/proc-v7-2level.S
index f64afb9f1bd5..bdd3be4be77a 100644
--- a/arch/arm/mm/proc-v7-2level.S
+++ b/arch/arm/mm/proc-v7-2level.S
@@ -110,7 +110,7 @@ ENTRY(cpu_v7_set_pte_ext)
  ARM(	str	r3, [r0, #2048]! )
  THUMB(	add	r0, r0, #2048 )
  THUMB(	str	r3, [r0] )
-	ALT_SMP(mov	pc,lr)
+	ALT_SMP(W(nop))
 	ALT_UP (mcr	p15, 0, r0, c7, c10, 1)		@ flush_pte
 #endif
 	mov	pc, lr
diff --git a/arch/arm/mm/proc-v7-3level.S b/arch/arm/mm/proc-v7-3level.S
index c36ac69488c8..01a719e18bb0 100644
--- a/arch/arm/mm/proc-v7-3level.S
+++ b/arch/arm/mm/proc-v7-3level.S
@@ -81,7 +81,7 @@ ENTRY(cpu_v7_set_pte_ext)
 	tst	r3, #1 << (55 - 32)		@ L_PTE_DIRTY
 	orreq	r2, #L_PTE_RDONLY
 1:	strd	r2, r3, [r0]
-	ALT_SMP(mov	pc, lr)
+	ALT_SMP(W(nop))
 	ALT_UP (mcr	p15, 0, r0, c7, c10, 1)		@ flush_pte
 #endif
 	mov	pc, lr
diff --git a/arch/arm/mm/proc-v7.S b/arch/arm/mm/proc-v7.S
index 5c6d5a3050ea..73398bcf9bd8 100644
--- a/arch/arm/mm/proc-v7.S
+++ b/arch/arm/mm/proc-v7.S
@@ -75,13 +75,14 @@ ENTRY(cpu_v7_do_idle)
 ENDPROC(cpu_v7_do_idle)
 
 ENTRY(cpu_v7_dcache_clean_area)
-	ALT_SMP(mov	pc, lr)			@ MP extensions imply L1 PTW
-	ALT_UP(W(nop))
-	dcache_line_size r2, r3
-1:	mcr	p15, 0, r0, c7, c10, 1		@ clean D entry
+	ALT_SMP(W(nop))			@ MP extensions imply L1 PTW
+	ALT_UP_B(1f)
+	mov	pc, lr
+1:	dcache_line_size r2, r3
+2:	mcr	p15, 0, r0, c7, c10, 1		@ clean D entry
 	add	r0, r0, r2
 	subs	r1, r1, r2
-	bhi	1b
+	bhi	2b
 	dsb
 	mov	pc, lr
 ENDPROC(cpu_v7_dcache_clean_area)