diff --git a/arch/arm/include/asm/assembler.h b/arch/arm/include/asm/assembler.h
index b146918..8d42ef4 100644
--- a/arch/arm/include/asm/assembler.h
+++ b/arch/arm/include/asm/assembler.h
@@ -58,16 +58,22 @@
 #endif
 
 /*
- * We only support cores that support at least Thumb-1 and thus we use
- * 'bx lr'
+ * Use 'bx lr' everywhere except ARMv4 (without 'T') where only 'mov pc, lr'
+ * works
  */
 	.irp	c,,eq,ne,cs,cc,mi,pl,vs,vc,hi,ls,ge,lt,gt,le,hs,lo
 	.macro	ret\c, reg
+
+	/* ARMv4- don't know bx lr but the assembler fails to see that */
+#ifdef __ARM_ARCH_4__
+	mov\c	pc, \reg
+#else
 	.ifeqs	"\reg", "lr"
 	bx\c	\reg
 	.else
 	mov\c	pc, \reg
 	.endif
+#endif
 	.endm
 	.endr
 
diff --git a/arch/arm/lib/lib1funcs.S b/arch/arm/lib/lib1funcs.S
index 700eee5..7ff4446 100644
--- a/arch/arm/lib/lib1funcs.S
+++ b/arch/arm/lib/lib1funcs.S
@@ -377,7 +377,7 @@
 	lsls	r1, r1, #1
 	add	lr, lr, r1
 	pop	{r1}
-	bx	lr
+	ret	lr
 ENDPROC(__gnu_thumb1_case_sqi)
 .popsection
 
@@ -391,7 +391,7 @@
 	lsls	r1, r1, #1
 	add	lr, lr, r1
 	pop	{r1}
-	bx	lr
+	ret	lr
 ENDPROC(__gnu_thumb1_case_uqi)
 .popsection
 
@@ -406,7 +406,7 @@
 	lsls	r1, r1, #1
 	add	lr, lr, r1
 	pop	{r0, r1}
-	bx	lr
+	ret	lr
 ENDPROC(__gnu_thumb1_case_shi)
 .popsection
 
@@ -421,7 +421,7 @@
 	lsls	r1, r1, #1
 	add	lr, lr, r1
 	pop	{r0, r1}
-	bx	lr
+	ret	lr
 ENDPROC(__gnu_thumb1_case_uhi)
 .popsection
 #endif
diff --git a/arch/arm/lib/memcpy.S b/arch/arm/lib/memcpy.S
index eee7a21..a1c996f 100644
--- a/arch/arm/lib/memcpy.S
+++ b/arch/arm/lib/memcpy.S
@@ -59,7 +59,7 @@
 #endif
 ENTRY(memcpy)
 		cmp	r0, r1
-		bxeq	lr
+		reteq	lr
 
 		enter	r4, lr
 
@@ -148,7 +148,7 @@
 		str1b	r0, ip, cs, abort=21f
 
 		exit	r4, lr
-		bx	lr
+		ret	lr
 
 9:		rsb	ip, ip, #4
 		cmp	ip, #2
@@ -258,7 +258,7 @@
 
 	.macro	copy_abort_end
 	ldmfd	sp!, {r4, lr}
-	bx	lr
+	ret	lr
 	.endm
 
 ENDPROC(memcpy)
diff --git a/arch/arm/lib/relocate.S b/arch/arm/lib/relocate.S
index 5102bfa..dd6f2e3 100644
--- a/arch/arm/lib/relocate.S
+++ b/arch/arm/lib/relocate.S
@@ -61,7 +61,7 @@
 	stmia	r1!, {r2-r8,r10}
 #endif
 #endif
-	bx	lr
+	ret	lr
 
 ENDPROC(relocate_vectors)
 
@@ -127,13 +127,7 @@
 	mcr	p15, 0, r0, c7, c10, 4	/* drain write buffer */
 #endif
 
-	/* ARMv4- don't know bx lr but the assembler fails to see that */
-
-#ifdef __ARM_ARCH_4__
-	mov	pc, lr
-#else
-	bx	lr
-#endif
+	ret	lr
 
 ENDPROC(relocate_code)
 
diff --git a/arch/arm/lib/setjmp.S b/arch/arm/lib/setjmp.S
index 176a1d5..2f041ae 100644
--- a/arch/arm/lib/setjmp.S
+++ b/arch/arm/lib/setjmp.S
@@ -17,7 +17,7 @@
 	mov  ip, sp
 	stm  a1, {v1-v8, ip, lr}
 	mov  a1, #0
-	bx   lr
+	ret  lr
 ENDPROC(setjmp)
 .popsection
 
@@ -31,6 +31,6 @@
 	bne  1f
 	mov  a1, #1
 1:
-	bx   lr
+	ret  lr
 ENDPROC(longjmp)
 .popsection