diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index db524e75c4a2..290a4b57617f 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -1229,7 +1229,7 @@ config SMP
 config SMP_ON_UP
 	bool "Allow booting SMP kernel on uniprocessor systems (EXPERIMENTAL)"
 	depends on EXPERIMENTAL
-	depends on SMP && !XIP && !THUMB2_KERNEL
+	depends on SMP && !XIP
 	default y
 	help
 	  SMP kernels contain instructions which fail on non-SMP processors.
diff --git a/arch/arm/include/asm/assembler.h b/arch/arm/include/asm/assembler.h
index 749bb6622404..72d3389e9c12 100644
--- a/arch/arm/include/asm/assembler.h
+++ b/arch/arm/include/asm/assembler.h
@@ -157,16 +157,24 @@
 #ifdef CONFIG_SMP
 #define ALT_SMP(instr...)					\
 9998:	instr
+/*
+ * Note: if you get assembler errors from ALT_UP() when building with
+ * CONFIG_THUMB2_KERNEL, you almost certainly need to use
+ * ALT_SMP( W(instr) ... )
+ */
 #define ALT_UP(instr...)					\
 	.pushsection ".alt.smp.init", "a"			;\
 	.long	9998b						;\
-	instr							;\
+9997:	instr							;\
+	.if . - 9997b != 4					;\
+		.error "ALT_UP() content must assemble to exactly 4 bytes";\
+	.endif							;\
 	.popsection
 #define ALT_UP_B(label)					\
 	.equ	up_b_offset, label - 9998b			;\
 	.pushsection ".alt.smp.init", "a"			;\
 	.long	9998b						;\
-	b	. + up_b_offset					;\
+	W(b)	. + up_b_offset					;\
 	.popsection
 #else
 #define ALT_SMP(instr...)
@@ -177,16 +185,24 @@
 /*
  * SMP data memory barrier
  */
-	.macro	smp_dmb
+	.macro	smp_dmb mode
 #ifdef CONFIG_SMP
 #if __LINUX_ARM_ARCH__ >= 7
+	.ifeqs "\mode","arm"
 	ALT_SMP(dmb)
+	.else
+	ALT_SMP(W(dmb))
+	.endif
 #elif __LINUX_ARM_ARCH__ == 6
 	ALT_SMP(mcr	p15, 0, r0, c7, c10, 5)	@ dmb
 #else
 #error Incompatible SMP platform
 #endif
+	.ifeqs "\mode","arm"
 	ALT_UP(nop)
+	.else
+	ALT_UP(W(nop))
+	.endif
 #endif
 	.endm
 
diff --git a/arch/arm/kernel/entry-armv.S b/arch/arm/kernel/entry-armv.S
index 955cf5f539ed..7f22a11a5105 100644
--- a/arch/arm/kernel/entry-armv.S
+++ b/arch/arm/kernel/entry-armv.S
@@ -842,7 +842,7 @@ __kuser_helper_start:
  */
 
 __kuser_memory_barrier:				@ 0xffff0fa0
-	smp_dmb
+	smp_dmb	arm
 	usr_ret	lr
 
 	.align	5
@@ -959,7 +959,7 @@ kuser_cmpxchg_fixup:
 
 #else
 
-	smp_dmb
+	smp_dmb	arm
 1:	ldrex	r3, [r2]
 	subs	r3, r3, r0
 	strexeq	r3, r1, [r2]
diff --git a/arch/arm/kernel/head.S b/arch/arm/kernel/head.S
index fd94e4e82fc9..359e54e83bd5 100644
--- a/arch/arm/kernel/head.S
+++ b/arch/arm/kernel/head.S
@@ -412,10 +412,17 @@ __fixup_smp_on_up:
 	add	r4, r4, r3
 	add	r5, r5, r3
 2:	cmp	r4, r5
+	movhs	pc, lr
 	ldmia	r4!, {r0, r6}
-	strlo	r6, [r0, r3]
-	blo	2b
-	mov	pc, lr
+ ARM(	str	r6, [r0, r3]	)
+ THUMB(	add	r0, r0, r3	)
+#ifdef __ARMEB__
+ THUMB(	mov	r6, r6, ror #16	)	@ Convert word order for big-endian.
+#endif
+ THUMB(	strh	r6, [r0], #2	)	@ For Thumb-2, store as two halfwords
+ THUMB(	mov	r6, r6, lsr #16	)	@ to be robust against misaligned r3.
+ THUMB(	strh	r6, [r0]	)
+	b	2b
 ENDPROC(__fixup_smp)
 
 1:	.word	.