Use the RET macro.

25 files changed, 295 insertions, 296 deletions

diff --git a/sys/arm/arm/bcopy_page.S b/sys/arm/arm/bcopy_page.S
index c310143..9fb79a1 100644
--- a/sys/arm/arm/bcopy_page.S
+++ b/sys/arm/arm/bcopy_page.S
@@ -242,7 +242,7 @@ ENTRY(bcopy_page)
 	strd	r4, [r1], #0x08
 	bgt	1b
 	ldmfd	sp!, {r4, r5}
-	mov	pc, lr
+	RET
 
 /*
  * XSCALE version of bzero_page
@@ -269,5 +269,5 @@ ENTRY(bzero_page)
 	strd	r2, [r0], #8
 	subs	r1, r1, #128
 	bne	1b
-	mov	pc, lr
+	RET
 #endif	/* __XSCALE__ */
diff --git a/sys/arm/arm/bcopyinout.S b/sys/arm/arm/bcopyinout.S
index 9900bcd..af9a40e 100644
--- a/sys/arm/arm/bcopyinout.S
+++ b/sys/arm/arm/bcopyinout.S
@@ -80,7 +80,7 @@ ENTRY(copyin)
 	/* Quick exit if length is zero */	
 	teq	r2, #0
 	moveq	r0, #0
-	moveq	pc, lr
+	RETEQ
 
 	SAVE_REGS
 #ifdef MULTIPROCESSOR
@@ -280,14 +280,14 @@ ENTRY(copyin)
 	str	r5, [r4, #PCB_ONFAULT]
 	RESTORE_REGS
 
-	mov	pc, lr
+	RET
 
 .Lcopyfault:
 	mov	r0, #14 /* EFAULT */
 	str	r5, [r4, #PCB_ONFAULT]
 	RESTORE_REGS
 
-	mov	pc, lr
+	RET
 
 /*
  * r0 = kernel space address
@@ -304,7 +304,7 @@ ENTRY(copyout)
 	/* Quick exit if length is zero */	
 	teq	r2, #0
 	moveq	r0, #0
-	moveq	pc, lr
+	RETeq
 
 	SAVE_REGS
 #ifdef MULTIPROCESSOR
@@ -501,7 +501,7 @@ ENTRY(copyout)
 	str	r5, [r4, #PCB_ONFAULT]
 	RESTORE_REGS
 
-	mov	pc, lr
+	RET
 
 /*
  * r0 = kernel space source address
@@ -517,7 +517,7 @@ ENTRY(kcopy)
 	/* Quick exit if length is zero */	
 	teq	r2, #0
 	moveq	r0, #0
-	moveq	pc, lr
+	RETeq
 
 	SAVE_REGS
 #ifdef MULTIPROCESSOR
@@ -703,7 +703,7 @@ ENTRY(kcopy)
 	str	r5, [r4, #PCB_ONFAULT]
 	RESTORE_REGS
 
-	mov	pc, lr
+	RET
 #endif	/* !__XSCALE__ */
 
 /*
@@ -738,7 +738,7 @@ ENTRY(badaddr_read_1)
 	strb	r3, [r1]
 	mov	r0, #0		/* No fault */
 1:	str	ip, [r2, #PCB_ONFAULT]
-	mov	pc, lr
+	RET
 
 /*
  * int badaddr_read_2(const uint16_t *src, uint16_t *dest)
@@ -772,7 +772,7 @@ ENTRY(badaddr_read_2)
 	strh	r3, [r1]
 	mov	r0, #0		/* No fault */
 1:	str	ip, [r2, #PCB_ONFAULT]
-	mov	pc, lr
+	RET
 
 /*
  * int badaddr_read_4(const uint32_t *src, uint32_t *dest)
@@ -806,5 +806,5 @@ ENTRY(badaddr_read_4)
 	str	r3, [r1]
 	mov	r0, #0		/* No fault */
 1:	str	ip, [r2, #PCB_ONFAULT]
-	mov	pc, lr
+	RET
 
diff --git a/sys/arm/arm/bcopyinout_xscale.S b/sys/arm/arm/bcopyinout_xscale.S
index d2b95b6..744fa59 100644
--- a/sys/arm/arm/bcopyinout_xscale.S
+++ b/sys/arm/arm/bcopyinout_xscale.S
@@ -111,7 +111,7 @@ ENTRY(copyin)
 	ldrbt	ip, [r0], #0x01
 	strb	ip, [r1], #0x01
 	cmp	r2, #0x00		/* All done? */
-	moveq	pc, lr
+	RETeq
 
 	/* Destination buffer is now word aligned */
 .Lcopyin_wordaligned:
@@ -193,7 +193,7 @@ ENTRY(copyin)
 .Lcopyin_w_lessthan128:
 	adds	r2, r2, #0x80		/* Adjust for extra sub */
 	ldmeqfd	sp!, {r4-r9}
-	moveq	pc, lr			/* Return now if done */
+	RETeq
 	subs	r2, r2, #0x20
 	blt	.Lcopyin_w_lessthan32
 
@@ -218,7 +218,7 @@ ENTRY(copyin)
 .Lcopyin_w_lessthan32:
 	adds	r2, r2, #0x20		/* Adjust for extra sub */
 	ldmeqfd	sp!, {r4-r9}
-	moveq	pc, lr			/* Return now if done */
+	RETeq				/* Return now if done */
 
 	and	r4, r2, #0x18
 	rsb	r5, r4, #0x18
@@ -246,14 +246,14 @@ ENTRY(copyin)
 
 	/* Less than 8 bytes remaining */
 	ldmfd	sp!, {r4-r9}
-	moveq	pc, lr			/* Return now if done */
+	RETeq				/* Return now if done */
 	mov	r3, #0x00
 
 .Lcopyin_w_less_than8:
 	subs	r2, r2, #0x04
 	ldrget	ip, [r0], #0x04
 	strge	ip, [r1], #0x04
-	moveq	pc, lr			/* Return now if done */
+	RETeq				/* Return now if done */
 	addlt	r2, r2, #0x04
 	ldrbt	ip, [r0], #0x01
 	cmp	r2, #0x02
@@ -262,7 +262,7 @@ ENTRY(copyin)
 	ldrgtbt	ip, [r0]
 	strgeb	r2, [r1], #0x01
 	strgtb	ip, [r1]
-	mov	pc, lr
+	RET
 
 /*
  * At this point, it has not been possible to word align both buffers.
@@ -317,7 +317,7 @@ ENTRY(copyin)
 
 	adds	r2, r2, #0x10         
 	ldmeqfd	sp!, {r4-r7}
-	moveq	pc, lr			/* Return now if done */
+	RETeq				/* Return now if done */
 	subs	r2, r2, #0x04
 	sublt	r0, r0, #0x03
 	blt	.Lcopyin_l4
@@ -378,7 +378,7 @@ ENTRY(copyin)
 
 	adds	r2, r2, #0x10         
 	ldmeqfd	sp!, {r4-r7}
-	moveq	pc, lr			/* Return now if done */
+	RETeq				/* Return now if done */
 	subs	r2, r2, #0x04
 	sublt	r0, r0, #0x02
 	blt	.Lcopyin_l4
@@ -439,7 +439,7 @@ ENTRY(copyin)
 
 	adds	r2, r2, #0x10         
 	ldmeqfd	sp!, {r4-r7}
-	moveq	pc, lr			/* Return now if done */
+	RETeq				/* Return now if done */
 	subs	r2, r2, #0x04
 	sublt	r0, r0, #0x01
 	blt	.Lcopyin_l4
@@ -465,7 +465,7 @@ ENTRY(copyin)
 	ldmfd	sp!, {r4-r7}
 	mov	r3, #0x00
 	adds	r2, r2, #0x04
-	moveq	pc, lr
+	RETeq
 .Lcopyin_l4_2:
 	rsbs	r2, r2, #0x03
 	addne	pc, pc, r2, lsl #3
@@ -476,7 +476,7 @@ ENTRY(copyin)
 	strb	ip, [r1], #0x01
 	ldrbt	ip, [r0]
 	strb	ip, [r1]
-	mov	pc, lr
+	RET
 
 
 /*
@@ -541,7 +541,7 @@ ENTRY(copyout)
 	ldrb	ip, [r0], #0x01
 	strbt	ip, [r1], #0x01
 	cmp	r2, #0x00		/* All done? */
-	moveq	pc, lr
+	RETeq
 
 	/* Destination buffer is now word aligned */
 .Lcopyout_wordaligned:
@@ -639,7 +639,7 @@ ENTRY(copyout)
 .Lcopyout_w_lessthan128:
 	adds	r2, r2, #0x80		/* Adjust for extra sub */
 	ldmeqfd	sp!, {r4-r9}
-	moveq	pc, lr			/* Return now if done */
+	RETeq				/* Return now if done */
 	subs	r2, r2, #0x20
 	blt	.Lcopyout_w_lessthan32
 
@@ -668,7 +668,7 @@ ENTRY(copyout)
 .Lcopyout_w_lessthan32:
 	adds	r2, r2, #0x20		/* Adjust for extra sub */
 	ldmeqfd	sp!, {r4-r9}
-	moveq	pc, lr			/* Return now if done */
+	RETeq				/* Return now if done */
 
 	and	r4, r2, #0x18
 	rsb	r5, r4, #0x18
@@ -696,14 +696,14 @@ ENTRY(copyout)
 
 	/* Less than 8 bytes remaining */
 	ldmfd	sp!, {r4-r9}
-	moveq	pc, lr			/* Return now if done */
+	RETeq				/* Return now if done */
 	mov	r3, #0x00
 
 .Lcopyout_w_less_than8:
 	subs	r2, r2, #0x04
 	ldrge	ip, [r0], #0x04
 	strget	ip, [r1], #0x04
-	moveq	pc, lr			/* Return now if done */
+	RETeq				/* Return now if done */
 	addlt	r2, r2, #0x04
 	ldrb	ip, [r0], #0x01
 	cmp	r2, #0x02
@@ -712,7 +712,7 @@ ENTRY(copyout)
 	ldrgtb	ip, [r0]
 	strgebt	r2, [r1], #0x01
 	strgtbt	ip, [r1]
-	mov	pc, lr
+	RET
 
 /*
  * At this point, it has not been possible to word align both buffers.
@@ -767,7 +767,7 @@ ENTRY(copyout)
 
 	adds	r2, r2, #0x10         
 	ldmeqfd	sp!, {r4-r7}
-	moveq	pc, lr			/* Return now if done */
+	RETeq				/* Return now if done */
 	subs	r2, r2, #0x04
 	sublt	r0, r0, #0x03
 	blt	.Lcopyout_l4
@@ -828,7 +828,7 @@ ENTRY(copyout)
 
 	adds	r2, r2, #0x10         
 	ldmeqfd	sp!, {r4-r7}
-	moveq	pc, lr			/* Return now if done */
+	RETeq				/* Return now if done */
 	subs	r2, r2, #0x04
 	sublt	r0, r0, #0x02
 	blt	.Lcopyout_l4
@@ -889,7 +889,7 @@ ENTRY(copyout)
 
 	adds	r2, r2, #0x10         
 	ldmeqfd	sp!, {r4-r7}
-	moveq	pc, lr			/* Return now if done */
+	RETeq				/* Return now if done */
 	subs	r2, r2, #0x04
 	sublt	r0, r0, #0x01
 	blt	.Lcopyout_l4
@@ -915,7 +915,7 @@ ENTRY(copyout)
 	ldmfd	sp!, {r4-r7}
 	mov	r3, #0x00
 	adds	r2, r2, #0x04
-	moveq	pc, lr
+	RETeq
 .Lcopyout_l4_2:
 	rsbs	r2, r2, #0x03
 	addne	pc, pc, r2, lsl #3
@@ -926,7 +926,7 @@ ENTRY(copyout)
 	strbt	ip, [r1], #0x01
 	ldrb	ip, [r0]
 	strbt	ip, [r1]
-	mov	pc, lr
+	RET
 
 
 /*
@@ -990,7 +990,7 @@ ENTRY(kcopy)
 	ldrb	ip, [r0], #0x01
 	strb	ip, [r1], #0x01
 	cmp	r2, #0x00		/* All done? */
-	moveq	pc, lr
+	RETeq
 
 	/* Destination buffer is now word aligned */
 .Lkcopy_wordaligned:
@@ -1071,7 +1071,7 @@ ENTRY(kcopy)
 .Lkcopy_w_lessthan128:
 	adds	r2, r2, #0x80		/* Adjust for extra sub */
 	ldmeqfd	sp!, {r4-r9}
-	moveq	pc, lr			/* Return now if done */
+	RETeq				/* Return now if done */
 	subs	r2, r2, #0x20
 	blt	.Lkcopy_w_lessthan32
 
@@ -1096,7 +1096,7 @@ ENTRY(kcopy)
 .Lkcopy_w_lessthan32:
 	adds	r2, r2, #0x20		/* Adjust for extra sub */
 	ldmeqfd	sp!, {r4-r9}
-	moveq	pc, lr			/* Return now if done */
+	RETeq				/* Return now if done */
 
 	and	r4, r2, #0x18
 	rsb	r5, r4, #0x18
@@ -1124,14 +1124,14 @@ ENTRY(kcopy)
 
 	/* Less than 8 bytes remaining */
 	ldmfd	sp!, {r4-r9}
-	moveq	pc, lr			/* Return now if done */
+	RETeq				/* Return now if done */
 	mov	r3, #0x00
 
 .Lkcopy_w_less_than8:
 	subs	r2, r2, #0x04
 	ldrge	ip, [r0], #0x04
 	strge	ip, [r1], #0x04
-	moveq	pc, lr			/* Return now if done */
+	RETeq				/* Return now if done */
 	addlt	r2, r2, #0x04
 	ldrb	ip, [r0], #0x01
 	cmp	r2, #0x02
@@ -1140,7 +1140,7 @@ ENTRY(kcopy)
 	ldrgtb	ip, [r0]
 	strgeb	r2, [r1], #0x01
 	strgtb	ip, [r1]
-	mov	pc, lr
+	RET
 
 /*
  * At this point, it has not been possible to word align both buffers.
@@ -1195,7 +1195,7 @@ ENTRY(kcopy)
 
 	adds	r2, r2, #0x10         
 	ldmeqfd	sp!, {r4-r7}
-	moveq	pc, lr			/* Return now if done */
+	RETeq				/* Return now if done */
 	subs	r2, r2, #0x04
 	sublt	r0, r0, #0x03
 	blt	.Lkcopy_bad_endgame
@@ -1256,7 +1256,7 @@ ENTRY(kcopy)
 
 	adds	r2, r2, #0x10         
 	ldmeqfd	sp!, {r4-r7}
-	moveq	pc, lr			/* Return now if done */
+	RETeq				/* Return now if done */
 	subs	r2, r2, #0x04
 	sublt	r0, r0, #0x02
 	blt	.Lkcopy_bad_endgame
@@ -1317,7 +1317,7 @@ ENTRY(kcopy)
 
 	adds	r2, r2, #0x10         
 	ldmeqfd	sp!, {r4-r7}
-	moveq	pc, lr			/* Return now if done */
+	RETeq				/* Return now if done */
 	subs	r2, r2, #0x04
 	sublt	r0, r0, #0x01
 	blt	.Lkcopy_bad_endgame
@@ -1343,7 +1343,7 @@ ENTRY(kcopy)
 	ldmfd	sp!, {r4-r7}
 	mov	r3, #0x00
 	adds	r2, r2, #0x04
-	moveq	pc, lr
+	RETeq
 .Lkcopy_bad_endgame2:
 	rsbs	r2, r2, #0x03
 	addne	pc, pc, r2, lsl #3
@@ -1354,4 +1354,4 @@ ENTRY(kcopy)
 	strb	ip, [r1], #0x01
 	ldrb	ip, [r0]
 	strb	ip, [r1]
-	mov	pc, lr
+	RET
diff --git a/sys/arm/arm/blockio.S b/sys/arm/arm/blockio.S
index 95a8efa..a300aa9 100644
--- a/sys/arm/arm/blockio.S
+++ b/sys/arm/arm/blockio.S
@@ -92,7 +92,7 @@ ENTRY(read_multi_1)
 .Lrm1_l4:
 	adds	r2, r2, #4			/* r2 = length again */
 	ldmeqdb	fp, {fp, sp, pc}
-	moveq	pc, r14
+	RETeq
 	cmp	r2, #2
 	ldrb	r3, [r0]
 	strb	r3, [r1], #1
@@ -182,7 +182,7 @@ ENTRY(insw)
 	strb	r3, [r1], #0x0001
 	bgt	.Linswloop
 
-	mov	pc, lr
+	RET
 
 /* Word aligned insw */
 
@@ -198,7 +198,7 @@ ENTRY(insw)
 	subs	r2, r2, #0x00000002	/* Next */
 	bgt	.Lfastinswloop
 
-	mov	pc, lr
+	RET
 
 
 /*
@@ -231,7 +231,7 @@ ENTRY(outsw)
 	str	r3, [r0]
 	bgt	.Loutswloop
 
-	mov	pc, lr
+	RET
 
 /* Word aligned outsw */
 
@@ -259,7 +259,7 @@ ENTRY(outsw)
 
 	bgt	.Lfastoutswloop
 
-	mov	pc, lr
+	RET
 
 /*
  * reads short ints (16 bits) from an I/O address into a block of memory
diff --git a/sys/arm/arm/bus_space_asm_generic.S b/sys/arm/arm/bus_space_asm_generic.S
index f8635a6..4c34537 100644
--- a/sys/arm/arm/bus_space_asm_generic.S
+++ b/sys/arm/arm/bus_space_asm_generic.S
@@ -49,17 +49,17 @@ __FBSDID("$FreeBSD$");
 
 ENTRY(generic_bs_r_1)
 	ldrb	r0, [r1, r2]
-	mov	pc, lr
+	RET
 
 #if (ARM_ARCH_4 + ARM_ARCH_5) > 0
 ENTRY(generic_armv4_bs_r_2)
 	ldrh	r0, [r1, r2]
-	mov	pc, lr
+	RET
 #endif
 
 ENTRY(generic_bs_r_4)
 	ldr	r0, [r1, r2]
-	mov	pc, lr
+	RET
 
 /*
  * write single
@@ -67,17 +67,17 @@ ENTRY(generic_bs_r_4)
 
 ENTRY(generic_bs_w_1)
 	strb	r3, [r1, r2]
-	mov	pc, lr
+	RET
 
 #if (ARM_ARCH_4 + ARM_ARCH_5) > 0
 ENTRY(generic_armv4_bs_w_2)
 	strh	r3, [r1, r2]
-	mov	pc, lr
+	RET
 #endif
 
 ENTRY(generic_bs_w_4)
 	str	r3, [r1, r2]
-	mov	pc, lr
+	RET
 
 /*
  * read multiple
@@ -88,14 +88,14 @@ ENTRY(generic_bs_rm_1)
 	mov	r1, r3
 	ldr	r2, [sp, #0]
 	teq	r2, #0
-	moveq	pc, lr
+	RETeq
 
 1:	ldrb	r3, [r0]
 	strb	r3, [r1], #1
 	subs	r2, r2, #1
 	bne	1b
 
-	mov	pc, lr
+	RET
 
 #if (ARM_ARCH_4 + ARM_ARCH_5) > 0
 ENTRY(generic_armv4_bs_rm_2)
@@ -103,14 +103,14 @@ ENTRY(generic_armv4_bs_rm_2)
 	mov	r1, r3
 	ldr	r2, [sp, #0]
 	teq	r2, #0
-	moveq	pc, lr
+	RETeq
 
 1:	ldrh	r3, [r0]
 	strh	r3, [r1], #2
 	subs	r2, r2, #1
 	bne	1b
 
-	mov	pc, lr
+	RET
 #endif
 
 ENTRY(generic_bs_rm_4)
@@ -118,14 +118,14 @@ ENTRY(generic_bs_rm_4)
 	mov	r1, r3
 	ldr	r2, [sp, #0]
 	teq	r2, #0
-	moveq	pc, lr
+	RETeq
 
 1:	ldr	r3, [r0]
 	str	r3, [r1], #4
 	subs	r2, r2, #1
 	bne	1b
 
-	mov	pc, lr
+	RET
 
 /*
  * write multiple
@@ -136,14 +136,14 @@ ENTRY(generic_bs_wm_1)
 	mov	r1, r3
 	ldr	r2, [sp, #0]
 	teq	r2, #0
-	moveq	pc, lr
+	RETeq
 
 1:	ldrb	r3, [r1], #1
 	strb	r3, [r0]
 	subs	r2, r2, #1
 	bne	1b
 
-	mov	pc, lr
+	RET
 
 #if (ARM_ARCH_4 + ARM_ARCH_5) > 0
 ENTRY(generic_armv4_bs_wm_2)
@@ -151,14 +151,14 @@ ENTRY(generic_armv4_bs_wm_2)
 	mov	r1, r3
 	ldr	r2, [sp, #0]
 	teq	r2, #0
-	moveq	pc, lr
+	RETeq
 
 1:	ldrh	r3, [r1], #2
 	strh	r3, [r0]
 	subs	r2, r2, #1
 	bne	1b
 
-	mov	pc, lr
+	RET
 #endif
 
 ENTRY(generic_bs_wm_4)
@@ -166,14 +166,14 @@ ENTRY(generic_bs_wm_4)
 	mov	r1, r3
 	ldr	r2, [sp, #0]
 	teq	r2, #0
-	moveq	pc, lr
+	RETeq
 
 1:	ldr	r3, [r1], #4
 	str	r3, [r0]
 	subs	r2, r2, #1
 	bne	1b
 
-	mov	pc, lr
+	RET
 
 /*
  * read region
@@ -184,14 +184,14 @@ ENTRY(generic_bs_rr_1)
 	mov	r1, r3
 	ldr	r2, [sp, #0]
 	teq	r2, #0
-	moveq	pc, lr
+	RETeq
 
 1:	ldrb	r3, [r0], #1
 	strb	r3, [r1], #1
 	subs	r2, r2, #1
 	bne	1b
 
-	mov	pc, lr
+	RET
 
 #if (ARM_ARCH_4 + ARM_ARCH_5) > 0
 ENTRY(generic_armv4_bs_rr_2)
@@ -199,14 +199,14 @@ ENTRY(generic_armv4_bs_rr_2)
 	mov	r1, r3
 	ldr	r2, [sp, #0]
 	teq	r2, #0
-	moveq	pc, lr
+	RETeq
 
 1:	ldrh	r3, [r0], #2
 	strh	r3, [r1], #2
 	subs	r2, r2, #1
 	bne	1b
 
-	mov	pc, lr
+	RET
 #endif
 
 ENTRY(generic_bs_rr_4)
@@ -214,14 +214,14 @@ ENTRY(generic_bs_rr_4)
 	mov	r1, r3
 	ldr	r2, [sp, #0]
 	teq	r2, #0
-	moveq	pc, lr
+	RETeq
 
 1:	ldr	r3, [r0], #4
 	str	r3, [r1], #4
 	subs	r2, r2, #1
 	bne	1b
 
-	mov	pc, lr
+	RET
 
 /*
  * write region.
@@ -232,14 +232,14 @@ ENTRY(generic_bs_wr_1)
 	mov	r1, r3
 	ldr	r2, [sp, #0]
 	teq	r2, #0
-	moveq	pc, lr
+	RETeq
 
 1:	ldrb	r3, [r1], #1
 	strb	r3, [r0], #1
 	subs	r2, r2, #1
 	bne	1b
 
-	mov	pc, lr
+	RET
 
 #if (ARM_ARCH_4 + ARM_ARCH_5) > 0
 ENTRY(generic_armv4_bs_wr_2)
@@ -247,14 +247,14 @@ ENTRY(generic_armv4_bs_wr_2)
 	mov	r1, r3
 	ldr	r2, [sp, #0]
 	teq	r2, #0
-	moveq	pc, lr
+	RETeq
 
 1:	ldrh	r3, [r1], #2
 	strh	r3, [r0], #2
 	subs	r2, r2, #1
 	bne	1b
 
-	mov	pc, lr
+	RET
 #endif
 
 ENTRY(generic_bs_wr_4)
@@ -262,14 +262,14 @@ ENTRY(generic_bs_wr_4)
 	mov	r1, r3
 	ldr	r2, [sp, #0]
 	teq	r2, #0
-	moveq	pc, lr
+	RETeq
 
 1:	ldr	r3, [r1], #4
 	str	r3, [r0], #4
 	subs	r2, r2, #1
 	bne	1b
 
-	mov	pc, lr
+	RET
 
 /*
  * set region
@@ -280,13 +280,13 @@ ENTRY(generic_bs_sr_1)
 	mov	r1, r3
 	ldr	r2, [sp, #0]
 	teq	r2, #0
-	moveq	pc, lr
+	RETeq
 
 1:	strb	r1, [r0], #1
 	subs	r2, r2, #1
 	bne	1b
 
-	mov	pc, lr
+	RET
 
 #if (ARM_ARCH_4 + ARM_ARCH_5) > 0
 ENTRY(generic_armv4_bs_sr_2)
@@ -294,13 +294,13 @@ ENTRY(generic_armv4_bs_sr_2)
 	mov	r1, r3
 	ldr	r2, [sp, #0]
 	teq	r2, #0
-	moveq	pc, lr
+	RETeq
 
 1:	strh	r1, [r0], #2
 	subs	r2, r2, #1
 	bne	1b
 
-	mov	pc, lr
+	RET
 #endif
 
 ENTRY(generic_bs_sr_4)
@@ -308,13 +308,13 @@ ENTRY(generic_bs_sr_4)
 	mov	r1, r3
 	ldr	r2, [sp, #0]
 	teq	r2, #0
-	moveq	pc, lr
+	RETeq
 
 1:	str	r1, [r0], #4
 	subs	r2, r2, #1
 	bne	1b
 
-	mov	pc, lr
+	RET
 
 /*
  * copy region
@@ -327,7 +327,7 @@ ENTRY(generic_armv4_bs_c_2)
 	add	r1, r2, r3
 	ldr	r2, [sp, #4]
 	teq	r2, #0
-	moveq	pc, lr
+	RETeq
 
 	cmp	r0, r1
 	blt	2f
@@ -337,7 +337,7 @@ ENTRY(generic_armv4_bs_c_2)
 	subs	r2, r2, #1
 	bne	1b
 
-	mov	pc, lr
+	RET
 
 2:	add	r0, r0, r2, lsl #1
 	add	r1, r1, r2, lsl #1
@@ -349,5 +349,5 @@ ENTRY(generic_armv4_bs_c_2)
 	subs	r2, r2, #1
 	bne	3b
 
-	mov	pc, lr
+	RET
 #endif
diff --git a/sys/arm/arm/copystr.S b/sys/arm/arm/copystr.S
index 9058b69..7c91cfa 100644
--- a/sys/arm/arm/copystr.S
+++ b/sys/arm/arm/copystr.S
@@ -87,7 +87,7 @@ ENTRY(copystr)
 	strne	r5, [r3]
 
 	ldmfd	sp!, {r4-r5}			/* stack is 8 byte aligned */
-	mov	pc, lr
+	RET
 
 #define SAVE_REGS	stmfd	sp!, {r4-r6}
 #define RESTORE_REGS	ldmfd	sp!, {r4-r6}
@@ -147,7 +147,7 @@ ENTRY(copyinstr)
 	strne	r6, [r3]
 
 	RESTORE_REGS
-	mov	pc, lr
+	RET
 
 /*
  * r0 - kernel space address
@@ -204,14 +204,14 @@ ENTRY(copyoutstr)
 	strne	r6, [r3]
 
 	RESTORE_REGS
-	mov	pc, lr
+	RET
 
 /* A fault occurred during the copy */
 .Lcopystrfault:
 	mov	r1, #0x00000000
 	str	r1, [r4, #PCB_ONFAULT]
 	RESTORE_REGS
-	mov	pc, lr
+	RET
 
 #ifdef DIAGNOSTIC
 .Lcopystrpcbfault:
diff --git a/sys/arm/arm/cpufunc_asm.S b/sys/arm/arm/cpufunc_asm.S
index 6a1456d..fa634af 100644
--- a/sys/arm/arm/cpufunc_asm.S
+++ b/sys/arm/arm/cpufunc_asm.S
@@ -49,7 +49,7 @@ __FBSDID("$FreeBSD$");
 	.align	0
 
 ENTRY(cpufunc_nullop)
-	mov	pc, lr
+	RET
 
 /*
  * Generic functions to read the internal coprocessor registers
@@ -63,23 +63,23 @@ ENTRY(cpufunc_nullop)
 
 ENTRY(cpufunc_id)
 	mrc	p15, 0, r0, c0, c0, 0
-	mov	pc, lr
+	RET
 
 ENTRY(cpu_get_control)
 	mrc	p15, 0, r0, c1, c0, 0
-	mov	pc, lr
+	RET
 
 ENTRY(cpu_read_cache_config)
 	mrc	p15, 0, r0, c0, c0, 1
-	mov	pc, lr
+	RET
 
 ENTRY(cpufunc_faultstatus)
 	mrc	p15, 0, r0, c5, c0, 0
-	mov	pc, lr
+	RET
 
 ENTRY(cpufunc_faultaddress)
 	mrc	p15, 0, r0, c6, c0, 0
-	mov	pc, lr
+	RET
 
 
 /*
@@ -96,12 +96,12 @@ ENTRY(cpufunc_faultaddress)
 #if 0 /* See below. */
 ENTRY(cpufunc_control)
 	mcr	p15, 0, r0, c1, c0, 0
-	mov	pc, lr
+	RET
 #endif
 
 ENTRY(cpufunc_domains)
 	mcr	p15, 0, r0, c3, c0, 0
-	mov	pc, lr
+	RET
 
 /*
  * Generic functions to read/modify/write the internal coprocessor registers
@@ -125,7 +125,7 @@ ENTRY(cpufunc_control)
 	mov	r0, r3			/* Return old value */
 	#endif
 
-	mov	pc, lr
+	RET
 .Lglou:
 	.asciz "plop %p\n"
 	.align 0
diff --git a/sys/arm/arm/cpufunc_asm_arm3.S b/sys/arm/arm/cpufunc_asm_arm3.S
index 83fc054..dd1558c 100644
--- a/sys/arm/arm/cpufunc_asm_arm3.S
+++ b/sys/arm/arm/cpufunc_asm_arm3.S
@@ -50,7 +50,7 @@ ENTRY(arm3_control)
 	teq	r2, r3			/* Only write if there is a change */
 	mcrne	p15, 0, r2, c2, c0, 0	/* Write new control register */
 	mov	r0, r3			/* Return old value */
-	mov	pc, lr
+	RET
 
 /*
  * Cache functions.
@@ -58,4 +58,4 @@ ENTRY(arm3_control)
 
 ENTRY(arm3_cache_flush)
 	mcr	p15, 0, r0, c1, c0, 0
-	mov	pc, lr
+	RET
diff --git a/sys/arm/arm/cpufunc_asm_arm67.S b/sys/arm/arm/cpufunc_asm_arm67.S
index 2d7dbb8..08107b1 100644
--- a/sys/arm/arm/cpufunc_asm_arm67.S
+++ b/sys/arm/arm/cpufunc_asm_arm67.S
@@ -61,25 +61,25 @@ ENTRY(arm67_setttb)
 	mov	r0, r0
 	mov	r0, r0
 
-	mov	pc, lr
+	RET
 
 /*
  * TLB functions
  */
 ENTRY(arm67_tlb_flush)
 	mcr	p15, 0, r0, c5, c0, 0 
-	mov	pc, lr
+	RET
 
 ENTRY(arm67_tlb_purge)
 	mcr	p15, 0, r0, c6, c0, 0
-	mov	pc, lr
+	RET
 
 /*
  * Cache functions
  */
 ENTRY(arm67_cache_flush)
 	mcr	p15, 0, r0, c7, c0, 0
-	mov	pc, lr
+	RET
 
 /*
  * Context switch.
@@ -108,4 +108,4 @@ ENTRY(arm67_context_switch)
 	/* Make sure that pipeline is emptied */
 	mov	r0, r0
 	mov	r0, r0
-	mov	pc, lr
+	RET
diff --git a/sys/arm/arm/cpufunc_asm_arm7tdmi.S b/sys/arm/arm/cpufunc_asm_arm7tdmi.S
index 0eee07a..a55d323 100644
--- a/sys/arm/arm/cpufunc_asm_arm7tdmi.S
+++ b/sys/arm/arm/cpufunc_asm_arm7tdmi.S
@@ -67,11 +67,11 @@ ENTRY(arm7tdmi_setttb)
 ENTRY(arm7tdmi_tlb_flushID)
 	mov	r0, #0
 	mcr	p15, 0, r0, c8, c7, 0
-	mov	pc, lr
+	RET
 
 ENTRY(arm7tdmi_tlb_flushID_SE)
 	mcr	p15, 0, r0, c8, c7, 1
-	mov	pc, lr
+	RET
 
 /*
  * Cache functions
@@ -85,7 +85,7 @@ ENTRY(arm7tdmi_cache_flushID)
 	mov	r0, r0
 	mov	r0, r0
 
-	mov	pc, lr
+	RET
 
 /*
  * Context switch.
diff --git a/sys/arm/arm/cpufunc_asm_arm8.S b/sys/arm/arm/cpufunc_asm_arm8.S
index 1afcb16..aeff06b 100644
--- a/sys/arm/arm/cpufunc_asm_arm8.S
+++ b/sys/arm/arm/cpufunc_asm_arm8.S
@@ -57,7 +57,7 @@ ENTRY(arm8_clock_config)
 	mov	r0, r0			/* NOP */
 	mcr	p15, 0, r2, c15, c0, 0 	/* Write clock register */
 	mov	r0, r3			/* Return old value */
-	mov	pc, lr
+	RET
 
 /*
  * Functions to set the MMU Translation Table Base register
@@ -89,29 +89,29 @@ ENTRY(arm8_setttb)
 	mov	r0, r0
 	msr	cpsr_all, r3
 
-	mov	pc, lr
+	RET
 
 /*
  * TLB functions
  */
 ENTRY(arm8_tlb_flushID)
 	mcr	p15, 0, r0, c8, c7, 0	/* flush I+D tlb */
-	mov	pc, lr
+	RET
 
 ENTRY(arm8_tlb_flushID_SE)
 	mcr	p15, 0, r0, c8, c7, 1	/* flush I+D tlb single entry */
-	mov	pc, lr
+	RET
 
 /*
  * Cache functions
  */
 ENTRY(arm8_cache_flushID)
 	mcr	p15, 0, r0, c7, c7, 0	/* flush I+D cache */
-	mov	pc, lr
+	RET
 
 ENTRY(arm8_cache_flushID_E)
 	mcr	p15, 0, r0, c7, c7, 1	/* flush I+D single entry */
-	mov	pc, lr
+	RET
 
 ENTRY(arm8_cache_cleanID)
 	mov	r0, #0x00000000
@@ -152,11 +152,11 @@ ENTRY(arm8_cache_cleanID)
 	adds	r0, r0, #0x04000000
 	bne	1b
 
-	mov	pc, lr
+	RET
 
 ENTRY(arm8_cache_cleanID_E)
 	mcr	p15, 0, r0, c7, c11, 1	/* clean I+D single entry */
-	mov	pc, lr
+	RET
 
 ENTRY(arm8_cache_purgeID)
 	/*
@@ -231,7 +231,7 @@ ENTRY(arm8_cache_purgeID)
 	bne	1b
 
 	msr	cpsr_all, r3
-	mov	pc, lr
+	RET
 
 ENTRY(arm8_cache_purgeID_E)
 	/*
@@ -252,7 +252,7 @@ ENTRY(arm8_cache_purgeID_E)
 	mcr	p15, 0, r0, c7, c11, 1	/* clean I+D single entry */
 	mcr	p15, 0, r0, c7, c7, 1	/* flush I+D single entry */
 	msr	cpsr_all, r3
-	mov	pc, lr
+	RET
 
 /*
  * Context switch.
@@ -281,4 +281,4 @@ ENTRY(arm8_context_switch)
 	/* Make sure that pipeline is emptied */
 	mov	r0, r0
 	mov	r0, r0
-	mov	pc, lr
+	RET
diff --git a/sys/arm/arm/cpufunc_asm_arm9.S b/sys/arm/arm/cpufunc_asm_arm9.S
index 3237bf0..958a32d 100644
--- a/sys/arm/arm/cpufunc_asm_arm9.S
+++ b/sys/arm/arm/cpufunc_asm_arm9.S
@@ -52,7 +52,7 @@ ENTRY(arm9_setttb)
 	mcr	p15, 0, r0, c2, c0, 0	/* load new TTB */
 
 	mcr	p15, 0, r0, c8, c7, 0	/* invalidate I+D TLBs */
-	mov	pc, lr
+	RET
 
 /*
  * TLB functions
@@ -60,46 +60,46 @@ ENTRY(arm9_setttb)
 ENTRY(arm9_tlb_flushID_SE)
 	mcr	p15, 0, r0, c8, c6, 1	/* flush D tlb single entry */
 	mcr	p15, 0, r0, c8, c5, 1	/* flush I tlb single entry */
-	mov	pc, lr
+	RET
 
 /*
  * Cache functions
  */
 ENTRY(arm9_cache_flushID)
 	mcr	p15, 0, r0, c7, c7, 0	/* flush I+D cache */
-	mov	pc, lr
+	RET
 
 ENTRY(arm9_cache_flushID_SE)
 	mcr	p15, 0, r0, c7, c5, 1	/* flush one entry from I cache */
 	mcr	p15, 0, r0, c7, c6, 1	/* flush one entry from D cache */
-	mov	pc, lr
+	RET
 
 ENTRY(arm9_cache_flushI)
 	mcr	p15, 0, r0, c7, c5, 0	/* flush I cache */
-	mov	pc, lr
+	RET
 
 ENTRY(arm9_cache_flushI_SE)
 	mcr	p15, 0, r0, c7, c5, 1	/* flush one entry from I cache */
-	mov	pc, lr
+	RET
 
 ENTRY(arm9_cache_flushD)
 	mcr	p15, 0, r0, c7, c6, 0	/* flush D cache */
-	mov	pc, lr
+	RET
 
 ENTRY(arm9_cache_flushD_SE)
 	mcr	p15, 0, r0, c7, c6, 1	/* flush one entry from D cache */
-	mov	pc, lr
+	RET
 
 ENTRY(arm9_cache_cleanID)
 	mcr	p15, 0, r0, c7, c10, 4
-	mov	pc, lr
+	RET
 
 /*
  * Soft functions
  */
 ENTRY(arm9_cache_syncI) 
 	mcr	p15, 0, r0, c7, c7, 0	/* flush I+D caches */
-	mov	pc, lr
+	RET
 
 ENTRY_NP(arm9_cache_flushID_rng)
 	b	_C_LABEL(arm9_cache_flushID)
@@ -134,4 +134,4 @@ ENTRY(arm9_context_switch)
 	nop
 	nop
 	nop
-	mov	pc, lr
+	RET
diff --git a/sys/arm/arm/cpufunc_asm_armv4.S b/sys/arm/arm/cpufunc_asm_armv4.S
index f1e7d3b..6914818 100644
--- a/sys/arm/arm/cpufunc_asm_armv4.S
+++ b/sys/arm/arm/cpufunc_asm_armv4.S
@@ -45,23 +45,23 @@ __FBSDID("$FreeBSD$");
  */
 ENTRY(armv4_tlb_flushID)
 	mcr	p15, 0, r0, c8, c7, 0	/* flush I+D tlb */
-	mov	pc, lr
+	RET
 
 ENTRY(armv4_tlb_flushI)
 	mcr	p15, 0, r0, c8, c5, 0	/* flush I tlb */
-	mov	pc, lr
+	RET
 
 ENTRY(armv4_tlb_flushD)
 	mcr	p15, 0, r0, c8, c6, 0	/* flush D tlb */
-	mov	pc, lr
+	RET
 
 ENTRY(armv4_tlb_flushD_SE)
 	mcr	p15, 0, r0, c8, c6, 1	/* flush D tlb single entry */
-	mov	pc, lr
+	RET
 
 /*
  * Other functions
  */
 ENTRY(armv4_drain_writebuf)
 	mcr	p15, 0, r0, c7, c10, 4	/* drain write buffer */
-	mov	pc, lr
+	RET
diff --git a/sys/arm/arm/cpufunc_asm_ixp12x0.S b/sys/arm/arm/cpufunc_asm_ixp12x0.S
index 142682c..dc20496 100644
--- a/sys/arm/arm/cpufunc_asm_ixp12x0.S
+++ b/sys/arm/arm/cpufunc_asm_ixp12x0.S
@@ -60,11 +60,11 @@ ENTRY(ixp12x0_context_switch)
 	/* Make sure that pipeline is emptied */
 	mov	r0, r0
 	mov	r0, r0
-	mov	pc, lr
+	RET
 
 ENTRY(ixp12x0_drain_readbuf)
 	mcr	p15, 0, r0, c9, c0, 0		/* drain read buffer */
-	mov	pc, lr
+	RET
 
 /*
  * Information for the IXP12X0 cache clean/purge functions:
diff --git a/sys/arm/arm/cpufunc_asm_sa1.S b/sys/arm/arm/cpufunc_asm_sa1.S
index 77768ff..6b10ba1 100644
--- a/sys/arm/arm/cpufunc_asm_sa1.S
+++ b/sys/arm/arm/cpufunc_asm_sa1.S
@@ -84,7 +84,7 @@ ENTRY(sa1_setttb)
 #else
 	str	r2, [r3]
 #endif
-	mov	pc, lr
+	RET
 
 /*
  * TLB functions
@@ -92,30 +92,30 @@ ENTRY(sa1_setttb)
 ENTRY(sa1_tlb_flushID_SE)
 	mcr	p15, 0, r0, c8, c6, 1	/* flush D tlb single entry */
 	mcr	p15, 0, r0, c8, c5, 0	/* flush I tlb */
-	mov	pc, lr
+	RET
 
 /*
  * Cache functions
  */
 ENTRY(sa1_cache_flushID)
 	mcr	p15, 0, r0, c7, c7, 0	/* flush I+D cache */
-	mov	pc, lr
+	RET
 
 ENTRY(sa1_cache_flushI)
 	mcr	p15, 0, r0, c7, c5, 0	/* flush I cache */
-	mov	pc, lr
+	RET
 
 ENTRY(sa1_cache_flushD)
 	mcr	p15, 0, r0, c7, c6, 0	/* flush D cache */
-	mov	pc, lr
+	RET
 
 ENTRY(sa1_cache_flushD_SE)
 	mcr	p15, 0, r0, c7, c6, 1	/* flush D cache single entry */
-	mov	pc, lr
+	RET
 
 ENTRY(sa1_cache_cleanD_E)
 	mcr	p15, 0, r0, c7, c10, 1	/* clean D cache entry */
-	mov	pc, lr
+	RET
 
 /*
  * Information for the SA-1 cache clean/purge functions:
@@ -195,20 +195,20 @@ ENTRY(sa1_cache_cleanD)
 	mcr	p15, 0, r0, c7, c10, 4	/* drain write buffer */
 
 	SA1_CACHE_CLEAN_EPILOGUE
-	mov	pc, lr
+	RET
 
 ENTRY(sa1_cache_purgeID_E)
 	mcr	p15, 0, r0, c7, c10, 1	/* clean dcache entry */
 	mcr	p15, 0, r0, c7, c10, 4	/* drain write buffer */
 	mcr	p15, 0, r0, c7, c5, 0	/* flush I cache */
 	mcr	p15, 0, r0, c7, c6, 1	/* flush D cache single entry */
-	mov	pc, lr
+	RET
 
 ENTRY(sa1_cache_purgeD_E)
 	mcr	p15, 0, r0, c7, c10, 1	/* clean dcache entry */
 	mcr	p15, 0, r0, c7, c10, 4	/* drain write buffer */
 	mcr	p15, 0, r0, c7, c6, 1	/* flush D cache single entry */
-	mov	pc, lr
+	RET
 
 /*
  * Soft functions
@@ -230,7 +230,7 @@ ENTRY(sa1_cache_cleanD_rng)
 	bhi	1b
 
 	mcr	p15, 0, r0, c7, c10, 4	/* drain write buffer */
-	mov	pc, lr
+	RET
 
 ENTRY(sa1_cache_purgeID_rng)
 	cmp	r1, #0x4000
@@ -248,7 +248,7 @@ ENTRY(sa1_cache_purgeID_rng)
 
 	mcr	p15, 0, r0, c7, c10, 4	/* drain write buffer */
 	mcr	p15, 0, r0, c7, c5, 0	/* flush I cache */
-	mov	pc, lr
+	RET
 
 ENTRY(sa1_cache_purgeD_rng)
 	cmp	r1, #0x4000
@@ -265,7 +265,7 @@ ENTRY(sa1_cache_purgeD_rng)
 	bhi	1b
 
 	mcr	p15, 0, r0, c7, c10, 4	/* drain write buffer */
-	mov	pc, lr
+	RET
 
 ENTRY(sa1_cache_syncI_rng)
 	cmp	r1, #0x4000
@@ -283,7 +283,7 @@ ENTRY(sa1_cache_syncI_rng)
 	mcr	p15, 0, r0, c7, c10, 4	/* drain write buffer */
 	mcr	p15, 0, r0, c7, c5, 0	/* flush I cache */
 
-	mov	pc, lr
+	RET
 
 /*
  * Context switch.
@@ -312,5 +312,5 @@ ENTRY(sa110_context_switch)
 	/* Make sure that pipeline is emptied */
 	mov	r0, r0
 	mov	r0, r0
-	mov	pc, lr
+	RET
 #endif
diff --git a/sys/arm/arm/cpufunc_asm_sa11x0.S b/sys/arm/arm/cpufunc_asm_sa11x0.S
index c4c04bc..a208f6f 100644
--- a/sys/arm/arm/cpufunc_asm_sa11x0.S
+++ b/sys/arm/arm/cpufunc_asm_sa11x0.S
@@ -94,7 +94,7 @@ ENTRY(sa11x0_cpu_sleep)
 
 	/* Restore interrupts (which will cause them to be serviced). */
 	msr	cpsr_all, r3
-	mov	pc, lr
+	RET
 
 
 /*
@@ -118,8 +118,8 @@ ENTRY(sa11x0_context_switch)
 	/* Make sure that pipeline is emptied */
 	mov	r0, r0
 	mov	r0, r0
-	mov	pc, lr
+	RET
 
 ENTRY(sa11x0_drain_readbuf)
 	mcr	p15, 0, r0, c9, c0, 0		/* drain read buffer */
-	mov	pc, lr
+	RET
diff --git a/sys/arm/arm/cpufunc_asm_xscale.S b/sys/arm/arm/cpufunc_asm_xscale.S
index 5b690cc..8e9e606 100644
--- a/sys/arm/arm/cpufunc_asm_xscale.S
+++ b/sys/arm/arm/cpufunc_asm_xscale.S
@@ -166,7 +166,7 @@ ENTRY(xscale_setttb)
 #else
 	str	r2, [r3]
 #endif
-	mov	pc, lr
+	RET
 
 /*
  * TLB functions
@@ -177,7 +177,7 @@ ENTRY(xscale_setttb)
 ENTRY(xscale_tlb_flushID_SE)
 	mcr	p15, 0, r0, c8, c6, 1	/* flush D tlb single entry */
 	mcr	p15, 0, r0, c8, c5, 1	/* flush I tlb single entry */
-	mov	pc, lr
+	RET
 
 /*
  * Cache functions
@@ -317,7 +317,7 @@ ENTRY(xscale_cache_cleanD)
 	CPWAIT(r0)
 
 	XSCALE_CACHE_CLEAN_EPILOGUE
-	mov	pc, lr
+	RET
 
 /*
  * Clean the mini-data cache.
@@ -494,4 +494,4 @@ ENTRY(xscale_cpu_sleep)
 	mcr	p14, 0, r0, c7, c0, 0
 
 1:
-	mov	pc, lr
+	RET
diff --git a/sys/arm/arm/fiq_subr.S b/sys/arm/arm/fiq_subr.S
index 054037f..d86e97f 100644
--- a/sys/arm/arm/fiq_subr.S
+++ b/sys/arm/arm/fiq_subr.S
@@ -73,7 +73,7 @@ ENTRY(fiq_getregs)
 	stmia	r0, {r8-r13}
 
 	BACK_TO_SVC_MODE
-	mov	pc, lr
+	RET
 
 /*
  * fiq_setregs:
@@ -87,7 +87,7 @@ ENTRY(fiq_setregs)
 	ldmia	r0, {r8-r13}
 
 	BACK_TO_SVC_MODE
-	mov	pc, lr
+	RET
 
 /*
  * fiq_nullhandler:
diff --git a/sys/arm/arm/fusu.S b/sys/arm/arm/fusu.S
index b7ad506..79be1d5 100644
--- a/sys/arm/arm/fusu.S
+++ b/sys/arm/arm/fusu.S
@@ -77,7 +77,7 @@ ENTRY(casuptr)
 	ldrt	r3, [r0]
 	cmp	r3, r1
 	movne	r0, r3
-	movne	pc, lr
+	RETne
 	strt	r2, [r0]
 	mov	r0, r1
 #ifdef MULTIPROCESSOR
@@ -94,7 +94,7 @@ ENTRY(casuptr)
 #endif
 	mov	r1, #0x00000000
 	str	r1, [r3, #PCB_ONFAULT]
-	mov	pc, lr
+	RET
 
 	
 /*
@@ -130,7 +130,7 @@ ENTRY(fuword)
 	mov	r1, #0x00000000
 	str	r1, [r2, #PCB_ONFAULT]
 	mov	r0, r3
-	mov	pc, lr
+	RET
 
 /*
  * fusword(caddr_t uaddr);
@@ -168,7 +168,7 @@ ENTRY(fusword)
 #endif
 	mov	r1, #0x00000000
 	str	r1, [r2, #PCB_ONFAULT]
-	mov	pc, lr
+	RET
 
 /*
  * fuswintr(caddr_t uaddr);
@@ -181,7 +181,7 @@ ENTRY(fuswintr)
 	ldr	r2, [r2]
 	teq	r2, #0
 	mvnne	r0, #0x00000000
-	movne	pc, lr
+	RETne
 
 #ifdef MULTIPROCESSOR
 	/* XXX Probably not appropriate for non-Hydra SMPs */
@@ -214,7 +214,7 @@ ENTRY(fuswintr)
 
 	mov	r1, #0x00000000
 	str	r1, [r2, #PCB_ONFAULT]
-	mov	pc, lr
+	RET
 
 Lblock_userspace_access:
 	.word	_C_LABEL(block_userspace_access)
@@ -258,7 +258,7 @@ ENTRY(fubyte)
 	mov	r1, #0x00000000
 	str	r1, [r2, #PCB_ONFAULT]
 	mov	r0, r3
-	mov	pc, lr
+	RET
 
 /*
  * Handle faults from [fs]u*().  Clean up and return -1.
@@ -268,7 +268,7 @@ ENTRY(fubyte)
 	mov	r0, #0x00000000
 	str	r0, [r2, #PCB_ONFAULT]
 	mvn	r0, #0x00000000
-	mov	pc, lr
+	RET
 
 /*
  * Handle faults from [fs]u*().  Clean up and return -1.  This differs from
@@ -282,7 +282,7 @@ _C_LABEL(fusubailout):
 	mov	r0, #0x00000000
 	str	r0, [r2, #PCB_ONFAULT]
 	mvn	r0, #0x00000000
-	mov	pc, lr
+	RET
 
 #ifdef DIAGNOSTIC
 /*
@@ -331,7 +331,7 @@ ENTRY(suword)
 
 	mov	r0, #0x00000000
 	str	r0, [r2, #PCB_ONFAULT]
-	mov	pc, lr
+	RET
 
 /*
  * suswintr(caddr_t uaddr, short x);
@@ -344,7 +344,7 @@ ENTRY(suswintr)
 	ldr	r2, [r2]
 	teq	r2, #0
 	mvnne	r0, #0x00000000
-	movne	pc, lr
+	RETne
 
 #ifdef MULTIPROCESSOR
 	stmfd	sp!, {r0, r1, r14}
@@ -377,7 +377,7 @@ ENTRY(suswintr)
 
 	mov	r0, #0x00000000
 	str	r0, [r2, #PCB_ONFAULT]
-	mov	pc, lr
+	RET
 
 /*
  * susword(caddr_t uaddr, short x);
@@ -416,7 +416,7 @@ ENTRY(susword)
 
 	mov	r0, #0x00000000
 	str	r0, [r2, #PCB_ONFAULT]
-	mov	pc, lr
+	RET
 
 /*
  * subyte(caddr_t uaddr, char x);
@@ -448,4 +448,4 @@ ENTRY(subyte)
 	strbt	r1, [r0]
 	mov	r0, #0x00000000
 	str	r0, [r2, #PCB_ONFAULT]
-	mov	pc, lr
+	RET
diff --git a/sys/arm/arm/in_cksum_arm.S b/sys/arm/arm/in_cksum_arm.S
index 57f43e3..09233be 100644
--- a/sys/arm/arm/in_cksum_arm.S
+++ b/sys/arm/arm/in_cksum_arm.S
@@ -274,7 +274,7 @@ ASENTRY_NP(L_cksumdata)
 	orrne	r2, r2, r6, lsl #24
 #endif
 	subs	r1, r1, r7		/* Update length */
-	moveq	pc, lr			/* All done? */
+	RETeq			/* All done? */
 
 	/* Buffer is now word aligned */
 .Lcksumdata_wordaligned:
@@ -366,7 +366,7 @@ ASENTRY_NP(L_cksumdata)
 #endif
 
 	adds	r1, r1, #0x40
-	moveq	pc, lr
+	RETeq
 	cmp	r1, #0x20
 
 #ifdef __XSCALE__
@@ -399,7 +399,7 @@ ASENTRY_NP(L_cksumdata)
 #endif
 	adc	r2, r2, #0x00
 	subs	r1, r1, #0x20
-	moveq	pc, lr
+	RETeq
 
 .Lcksumdata_less_than_32:
 	/* There are less than 32 bytes left */
@@ -442,7 +442,7 @@ ASENTRY_NP(L_cksumdata)
 	/* Deal with < 4 bytes remaining */
 .Lcksumdata_lessthan4:
 	adds	r1, r1, #0x04
-	moveq	pc, lr
+	RETeq
 
 	/* Deal with 1 to 3 remaining bytes, possibly misaligned */
 .Lcksumdata_endgame:
@@ -467,4 +467,4 @@ ASENTRY_NP(L_cksumdata)
 #endif
 	adds	r2, r2, r3
 	adc	r2, r2, #0x00
-	mov	pc, lr
+	RET
diff --git a/sys/arm/arm/locore.S b/sys/arm/arm/locore.S
index 38aca85..390e940 100644
--- a/sys/arm/arm/locore.S
+++ b/sys/arm/arm/locore.S
@@ -317,12 +317,12 @@ ENTRY_NP(Execute)
 ENTRY(setjmp)
 	stmia	r0, {r4-r14}
 	mov	r0, #0x00000000
-	mov	pc, lr
+	RET
 
 ENTRY(longjmp)
 	ldmia	r0, {r4-r14}
 	mov	r0, #0x00000001
-	mov	pc, lr
+	RET
 
 	.data
 	.global _C_LABEL(esym)
diff --git a/sys/arm/arm/nexus_io_asm.S b/sys/arm/arm/nexus_io_asm.S
index 0129f7f..1b5a166 100644
--- a/sys/arm/arm/nexus_io_asm.S
+++ b/sys/arm/arm/nexus_io_asm.S
@@ -47,17 +47,17 @@ __FBSDID("$FreeBSD$");
 
 ENTRY(nexus_bs_r_1)
 	ldrb	r0, [r1, r2, lsl #2]
-	mov	pc, lr
+	RET
 
 ENTRY(nexus_bs_r_2)
 	ldr	r0, [r1, r2, lsl #2]
 	bic	r0, r0, #0xff000000
 	bic	r0, r0, #0x00ff0000
-	mov	pc, lr
+	RET
 
 ENTRY(nexus_bs_r_4)
 	ldr	r0, [r1, r2, lsl #2]
-	mov	pc, lr
+	RET
 
 /*
  * write single
@@ -65,17 +65,17 @@ ENTRY(nexus_bs_r_4)
 
 ENTRY(nexus_bs_w_1)
 	strb	r3, [r1, r2, lsl #2]
-	mov	pc, lr
+	RET
 
 ENTRY(nexus_bs_w_2)
 	mov	r3, r3, lsl #16
 	orr	r3, r3, r3, lsr #16
 	str	r3, [r1, r2, lsl #2]
-	mov	pc, lr
+	RET
 
 ENTRY(nexus_bs_w_4)
 	str	r3, [r1, r2, lsl #2]
-	mov	pc, lr
+	RET
 
 /*
  * read multiple
@@ -105,7 +105,7 @@ nexus_wm_1_loop:
 	subs	r2, r2, #0x00000001
 	bgt	nexus_wm_1_loop
 
-	mov	pc, lr
+	RET
 
 ENTRY(nexus_bs_wm_2)
 	add	r0, r1, r2, lsl #2
diff --git a/sys/arm/arm/setcpsr.S b/sys/arm/arm/setcpsr.S
index fa70e1b..f2385b9 100644
--- a/sys/arm/arm/setcpsr.S
+++ b/sys/arm/arm/setcpsr.S
@@ -65,7 +65,7 @@ ENTRY_NP(SetCPSR)
 
         mov	r0, r3			/* Return the old CPSR */
 
-	mov	pc, lr
+	RET
 
 
 /* Gets the CPSR register
@@ -76,5 +76,5 @@ ENTRY_NP(SetCPSR)
 ENTRY_NP(GetCPSR)
         mrs     r0, cpsr		/* Get the CPSR */
 
-	mov	pc, lr
+	RET
 
diff --git a/sys/arm/arm/support.S b/sys/arm/arm/support.S
index 5a8c738..8db9f87 100644
--- a/sys/arm/arm/support.S
+++ b/sys/arm/arm/support.S
@@ -117,7 +117,7 @@ do_memset:
 	stmgeia	ip!, {r2-r3}
 #endif
 	bgt	.Lmemset_loop128
-	moveq	pc, lr			/* Zero length so just exit */
+	RETeq			/* Zero length so just exit */
 
 	add	r1, r1, #0x80		/* Adjust for extra sub */
 
@@ -136,7 +136,7 @@ do_memset:
 	stmgeia	ip!, {r2-r3}
 #endif
 	bgt	.Lmemset_loop32
-	moveq	pc, lr			/* Zero length so just exit */
+	RETeq			/* Zero length so just exit */
 
 	adds	r1, r1, #0x10		/* Partially adjust for extra sub */
 
@@ -148,7 +148,7 @@ do_memset:
 	stmgeia	ip!, {r2-r3}
 	stmgeia	ip!, {r2-r3}
 #endif
-	moveq	pc, lr			/* Zero length so just exit */
+	RETeq			/* Zero length so just exit */
 
 	addlt	r1, r1, #0x10		/* Possibly adjust for extra sub */
 
@@ -157,12 +157,12 @@ do_memset:
 	subs	r1, r1, #0x04
 	strge	r3, [ip], #0x04
 	bgt	.Lmemset_loop4
-	moveq	pc, lr			/* Zero length so just exit */
+	RETeq			/* Zero length so just exit */
 
 #ifdef __XSCALE__
 	/* Compensate for 64-bit alignment check */
 	adds	r1, r1, #0x04
-	moveq	pc, lr
+	RETeq
 	cmp	r1, #2
 #else
 	cmp	r1, #-2
@@ -171,7 +171,7 @@ do_memset:
 	strb	r3, [ip], #0x01		/* Set 1 byte */
 	strgeb	r3, [ip], #0x01		/* Set another byte */
 	strgtb	r3, [ip]		/* and a third */
-	mov	pc, lr			/* Exit */
+	RET			/* Exit */
 
 .Lmemset_wordunaligned:
 	rsb	r2, r2, #0x004
@@ -185,12 +185,12 @@ do_memset:
 
 .Lmemset_lessthanfour:
 	cmp	r1, #0x00
-	moveq	pc, lr			/* Zero length so exit */
+	RETeq			/* Zero length so exit */
 	strb	r3, [ip], #0x01		/* Set 1 byte */
 	cmp	r1, #0x02
 	strgeb	r3, [ip], #0x01		/* Set another byte */
 	strgtb	r3, [ip]		/* and a third */
-	mov	pc, lr			/* Exit */
+	RET			/* Exit */
 
 ENTRY(memcmp)
 	mov	ip, r0
@@ -201,7 +201,7 @@ ENTRY(memcmp)
 	/* Are both addresses aligned the same way? */
 	cmp	r2, #0x00
 	eornes	r3, ip, r1
-	moveq	pc, lr			/* len == 0, or same addresses! */
+	RETeq			/* len == 0, or same addresses! */
 	tst	r3, #0x03
 	subne	r2, r2, #0x01
 	bne	.Lmemcmp_bytewise2	/* Badly aligned. Do it the slow way */
@@ -217,25 +217,25 @@ ENTRY(memcmp)
 	ldrb	r0, [ip], #0x01
 	ldrb	r3, [r1], #0x01
 	subs	r0, r0, r3
-	movne	pc, lr
+	RETne
 	subs	r2, r2, #0x01
-	moveq	pc, lr
+	RETeq
 
 	/* Compare up to 2 bytes */
 	ldrb	r0, [ip], #0x01
 	ldrb	r3, [r1], #0x01
 	subs	r0, r0, r3
-	movne	pc, lr
+	RETne
 	subs	r2, r2, #0x01
-	moveq	pc, lr
+	RETeq
 
 	/* Compare 1 byte */
 	ldrb	r0, [ip], #0x01
 	ldrb	r3, [r1], #0x01
 	subs	r0, r0, r3
-	movne	pc, lr
+	RETne
 	subs	r2, r2, #0x01
-	moveq	pc, lr
+	RETeq
 
 	/* Compare 4 bytes at a time, if possible */
 	subs	r2, r2, #0x04
@@ -251,7 +251,7 @@ ENTRY(memcmp)
 	/* Correct for extra subtraction, and check if done */
 	adds	r2, r2, #0x04
 	cmpeq	r0, #0x00		/* If done, did all bytes match? */
-	moveq	pc, lr			/* Yup. Just return */
+	RETeq			/* Yup. Just return */
 
 	/* Re-do the final word byte-wise */
 	sub	ip, ip, #0x04
@@ -266,7 +266,7 @@ ENTRY(memcmp)
 	cmpcs	r0, r3
 	beq	.Lmemcmp_bytewise2
 	sub	r0, r0, r3
-	mov	pc, lr
+	RET
 
 	/*
 	 * 6 byte compares are very common, thanks to the network stack.
@@ -281,25 +281,25 @@ ENTRY(memcmp)
 	ldrb	r2, [r1, #0x01]		/* r2 = b2#1 */
 	subs	r0, r0, r3		/* r0 = b1#0 - b2#0 */
 	ldreqb	r3, [ip, #0x01]		/* r3 = b1#1 */
-	movne	pc, lr			/* Return if mismatch on #0 */
+	RETne			/* Return if mismatch on #0 */
 	subs	r0, r3, r2		/* r0 = b1#1 - b2#1 */
 	ldreqb	r3, [r1, #0x02]		/* r3 = b2#2 */
 	ldreqb	r0, [ip, #0x02]		/* r0 = b1#2 */
-	movne	pc, lr			/* Return if mismatch on #1 */
+	RETne			/* Return if mismatch on #1 */
 	ldrb	r2, [r1, #0x03]		/* r2 = b2#3 */
 	subs	r0, r0, r3		/* r0 = b1#2 - b2#2 */
 	ldreqb	r3, [ip, #0x03]		/* r3 = b1#3 */
-	movne	pc, lr			/* Return if mismatch on #2 */
+	RETne			/* Return if mismatch on #2 */
 	subs	r0, r3, r2		/* r0 = b1#3 - b2#3 */
 	ldreqb	r3, [r1, #0x04]		/* r3 = b2#4 */
 	ldreqb	r0, [ip, #0x04]		/* r0 = b1#4 */
-	movne	pc, lr			/* Return if mismatch on #3 */
+	RETne			/* Return if mismatch on #3 */
 	ldrb	r2, [r1, #0x05]		/* r2 = b2#5 */
 	subs	r0, r0, r3		/* r0 = b1#4 - b2#4 */
 	ldreqb	r3, [ip, #0x05]		/* r3 = b1#5 */
-	movne	pc, lr			/* Return if mismatch on #4 */
+	RETne			/* Return if mismatch on #4 */
 	sub	r0, r3, r2		/* r0 = b1#5 - b2#5 */
-	mov	pc, lr
+	RET
 
 ENTRY(bcopy)
 	mov r3, r0
@@ -607,7 +607,7 @@ ENTRY(memcpy)
 .Lmemcpy_w_lessthan128:
 	adds	r2, r2, #0x80		/* Adjust for extra sub */
 	ldmeqfd	sp!, {r4-r9}
-	moveq	pc, lr			/* Return now if done */
+	RETeq			/* Return now if done */
 	subs	r2, r2, #0x20
 	blt	.Lmemcpy_w_lessthan32
 
@@ -632,7 +632,7 @@ ENTRY(memcpy)
 .Lmemcpy_w_lessthan32:
 	adds	r2, r2, #0x20		/* Adjust for extra sub */
 	ldmeqfd	sp!, {r4-r9}
-	moveq	pc, lr			/* Return now if done */
+	RETeq			/* Return now if done */
 
 	and	r4, r2, #0x18
 	rsbs	r4, r4, #0x18
@@ -659,11 +659,11 @@ ENTRY(memcpy)
 
 	/* Less than 8 bytes remaining */
 	ldmfd	sp!, {r4-r9}
-	moveq	pc, lr			/* Return now if done */
+	RETeq			/* Return now if done */
 	subs	r2, r2, #0x04
 	ldrge	ip, [r1], #0x04
 	strge	ip, [r3], #0x04
-	moveq	pc, lr			/* Return now if done */
+	RETeq			/* Return now if done */
 	addlt	r2, r2, #0x04
 	ldrb	ip, [r1], #0x01
 	cmp	r2, #0x02
@@ -672,7 +672,7 @@ ENTRY(memcpy)
 	ldrgtb	ip, [r1]
 	strgeb	r2, [r3], #0x01
 	strgtb	ip, [r3]
-	mov	pc, lr
+	RET
 
 
 /*
@@ -726,7 +726,7 @@ ENTRY(memcpy)
 
 	adds	r2, r2, #0x10         
 	ldmeqfd	sp!, {r4-r7}
-	moveq	pc, lr			/* Return now if done */
+	RETeq			/* Return now if done */
 	subs	r2, r2, #0x04
 	sublt	r1, r1, #0x03
 	blt	.Lmemcpy_bad_done
@@ -787,7 +787,7 @@ ENTRY(memcpy)
 
 	adds	r2, r2, #0x10         
 	ldmeqfd	sp!, {r4-r7}
-	moveq	pc, lr			/* Return now if done */
+	RETeq			/* Return now if done */
 	subs	r2, r2, #0x04
 	sublt	r1, r1, #0x02
 	blt	.Lmemcpy_bad_done
@@ -848,7 +848,7 @@ ENTRY(memcpy)
 
 	adds	r2, r2, #0x10         
 	ldmeqfd	sp!, {r4-r7}
-	moveq	pc, lr			/* Return now if done */
+	RETeq			/* Return now if done */
 	subs	r2, r2, #0x04
 	sublt	r1, r1, #0x01
 	blt	.Lmemcpy_bad_done
@@ -873,7 +873,7 @@ ENTRY(memcpy)
 .Lmemcpy_bad_done:
 	ldmfd	sp!, {r4-r7}
 	adds	r2, r2, #0x04
-	moveq	pc, lr
+	RETeq
 	ldrb	ip, [r1], #0x01
 	cmp	r2, #0x02
 	ldrgeb	r2, [r1], #0x01
@@ -881,7 +881,7 @@ ENTRY(memcpy)
 	ldrgtb	ip, [r1]
 	strgeb	r2, [r3], #0x01
 	strgtb	ip, [r3]
-	mov	pc, lr
+	RET
 
 
 /*
@@ -892,7 +892,7 @@ ENTRY(memcpy)
 .Lmemcpy_short:
 	add	pc, pc, r2, lsl #2
 	nop
-	mov	pc, lr			/* 0x00 */
+	RET			/* 0x00 */
 	b	.Lmemcpy_bytewise	/* 0x01 */
 	b	.Lmemcpy_bytewise	/* 0x02 */
 	b	.Lmemcpy_bytewise	/* 0x03 */
@@ -912,7 +912,7 @@ ENTRY(memcpy)
 	strb	ip, [r3], #0x01
 	ldrneb	ip, [r1], #0x01
 	bne	1b
-	mov	pc, lr
+	RET
 
 /******************************************************************************
  * Special case for 4 byte copies
@@ -932,7 +932,7 @@ ENTRY(memcpy)
  */
 	ldr	r2, [r1]
 	str	r2, [r0]
-	mov	pc, lr
+	RET
 	LMEMCPY_4_PAD
 
 /*
@@ -948,7 +948,7 @@ ENTRY(memcpy)
 	orr	r3, r3, r2, lsl #24	/* r3 = 3210 */
 #endif
 	str	r3, [r0]
-	mov	pc, lr
+	RET
 	LMEMCPY_4_PAD
 
 /*
@@ -963,7 +963,7 @@ ENTRY(memcpy)
 #endif
 	orr	r3, r2, r3, lsl #16
 	str	r3, [r0]
-	mov	pc, lr
+	RET
 	LMEMCPY_4_PAD
 
 /*
@@ -979,7 +979,7 @@ ENTRY(memcpy)
 	orr	r3, r3, r2, lsl #8	/* r3 = 3210 */
 #endif
 	str	r3, [r0]
-	mov	pc, lr
+	RET
 	LMEMCPY_4_PAD
 
 /*
@@ -998,7 +998,7 @@ ENTRY(memcpy)
 	strb	r1, [r0, #0x03]
 #endif
 	strh	r3, [r0, #0x01]
-	mov	pc, lr
+	RET
 	LMEMCPY_4_PAD
 
 /*
@@ -1010,7 +1010,7 @@ ENTRY(memcpy)
 	strb	r2, [r0]
 	strh	r3, [r0, #0x01]
 	strb	r1, [r0, #0x03]
-	mov	pc, lr
+	RET
 	LMEMCPY_4_PAD
 
 /*
@@ -1031,7 +1031,7 @@ ENTRY(memcpy)
 #endif
 	strh	r2, [r0, #0x01]
 	strb	r3, [r0, #0x03]
-	mov	pc, lr
+	RET
 	LMEMCPY_4_PAD
 
 /*
@@ -1043,7 +1043,7 @@ ENTRY(memcpy)
 	strb	r2, [r0]
 	strh	r3, [r0, #0x01]
 	strb	r1, [r0, #0x03]
-	mov	pc, lr
+	RET
 	LMEMCPY_4_PAD
 
 /*
@@ -1059,7 +1059,7 @@ ENTRY(memcpy)
 	mov	r3, r2, lsr #16
 	strh	r3, [r0, #0x02]
 #endif
-	mov	pc, lr
+	RET
 	LMEMCPY_4_PAD
 
 /*
@@ -1077,7 +1077,7 @@ ENTRY(memcpy)
 	orr	r2, r2, r3, lsl #8	/* r2 = xx32 */
 #endif
 	strh	r2, [r0, #0x02]
-	mov	pc, lr
+	RET
 	LMEMCPY_4_PAD
 
 /*
@@ -1087,7 +1087,7 @@ ENTRY(memcpy)
 	ldrh	r3, [r1, #0x02]
 	strh	r2, [r0]
 	strh	r3, [r0, #0x02]
-	mov	pc, lr
+	RET
 	LMEMCPY_4_PAD
 
 /*
@@ -1105,7 +1105,7 @@ ENTRY(memcpy)
 	orr	r3, r3, r2, lsr #24	/* r3 = 3210 */
 #endif
 	strh	r3, [r0]
-	mov	pc, lr
+	RET
 	LMEMCPY_4_PAD
 
 /*
@@ -1125,7 +1125,7 @@ ENTRY(memcpy)
 	strh	r3, [r0, #0x01]
 	strb	r1, [r0, #0x03]
 #endif
-	mov	pc, lr
+	RET
 	LMEMCPY_4_PAD
 
 /*
@@ -1137,7 +1137,7 @@ ENTRY(memcpy)
 	strb	r2, [r0]
 	strh	r3, [r0, #0x01]
 	strb	r1, [r0, #0x03]
-	mov	pc, lr
+	RET
 	LMEMCPY_4_PAD
 
 /*
@@ -1162,7 +1162,7 @@ ENTRY(memcpy)
 	mov	r3, r3, lsr #8		/* r3 = ...3 */
 	strb	r3, [r0, #0x03]
 #endif
-	mov	pc, lr
+	RET
 	LMEMCPY_4_PAD
 
 /*
@@ -1174,7 +1174,7 @@ ENTRY(memcpy)
 	strb	r2, [r0]
 	strh	r3, [r0, #0x01]
 	strb	r1, [r0, #0x03]
-	mov	pc, lr
+	RET
 	LMEMCPY_4_PAD
 
 
@@ -1198,7 +1198,7 @@ ENTRY(memcpy)
 	ldrh	r3, [r1, #0x04]
 	str	r2, [r0]
 	strh	r3, [r0, #0x04]
-	mov	pc, lr
+	RET
 	LMEMCPY_6_PAD
 
 /*
@@ -1216,7 +1216,7 @@ ENTRY(memcpy)
 	mov	r3, r3, lsr #8		/* BE:r3 = .345  LE:r3 = .x54 */
 	str	r2, [r0]
 	strh	r3, [r0, #0x04]
-	mov	pc, lr
+	RET
 	LMEMCPY_6_PAD
 
 /*
@@ -1235,7 +1235,7 @@ ENTRY(memcpy)
 	str	r2, [r0]
 	strh	r1, [r0, #0x04]
 #endif
-	mov	pc, lr
+	RET
 	LMEMCPY_6_PAD
 
 /*
@@ -1257,7 +1257,7 @@ ENTRY(memcpy)
 #endif
 	str	r2, [r0]
 	strh	r1, [r0, #0x04]
-	mov	pc, lr
+	RET
 	LMEMCPY_6_PAD
 
 /*
@@ -1280,7 +1280,7 @@ ENTRY(memcpy)
 #endif
 	strh	r3, [r0, #0x03]
 	strb	r2, [r0, #0x05]
-	mov	pc, lr
+	RET
 	LMEMCPY_6_PAD
 
 /*
@@ -1294,7 +1294,7 @@ ENTRY(memcpy)
 	strh	r3, [r0, #0x01]
 	strh	ip, [r0, #0x03]
 	strb	r1, [r0, #0x05]
-	mov	pc, lr
+	RET
 	LMEMCPY_6_PAD
 
 /*
@@ -1321,7 +1321,7 @@ ENTRY(memcpy)
 	orr	r3, r3, r1, lsl #8	/* r3 = 4321 */
 	strh	r3, [r0, #0x01]
 #endif
-	mov	pc, lr
+	RET
 	LMEMCPY_6_PAD
 
 /*
@@ -1335,7 +1335,7 @@ ENTRY(memcpy)
 	strh	r3, [r0, #0x01]
 	strh	ip, [r0, #0x03]
 	strb	r1, [r0, #0x05]
-	mov	pc, lr
+	RET
 	LMEMCPY_6_PAD
 
 /*
@@ -1356,7 +1356,7 @@ ENTRY(memcpy)
 	strh	r3, [r0]
 	str	r2, [r0, #0x02]
 #endif
-	mov	pc, lr
+	RET
 	LMEMCPY_6_PAD
 
 /*
@@ -1374,7 +1374,7 @@ ENTRY(memcpy)
 #endif
 	strh	r1, [r0]
 	str	r2, [r0, #0x02]
-	mov	pc, lr
+	RET
 	LMEMCPY_6_PAD
 
 /*
@@ -1384,7 +1384,7 @@ ENTRY(memcpy)
 	ldr	r3, [r1, #0x02]
 	strh	r2, [r0]
 	str	r3, [r0, #0x02]
-	mov	pc, lr
+	RET
 	LMEMCPY_6_PAD
 
 /*
@@ -1404,7 +1404,7 @@ ENTRY(memcpy)
 #endif
 	strh	r3, [r0]
 	str	r1, [r0, #0x02]
-	mov	pc, lr
+	RET
 	LMEMCPY_6_PAD
 
 /*
@@ -1425,7 +1425,7 @@ ENTRY(memcpy)
 #endif
 	str	r2, [r0, #0x01]
 	strb	r1, [r0, #0x05]
-	mov	pc, lr
+	RET
 	LMEMCPY_6_PAD
 
 /*
@@ -1439,7 +1439,7 @@ ENTRY(memcpy)
 	strh	r3, [r0, #0x01]
 	strh	ip, [r0, #0x03]
 	strb	r1, [r0, #0x05]
-	mov	pc, lr
+	RET
 	LMEMCPY_6_PAD
 
 /*
@@ -1460,7 +1460,7 @@ ENTRY(memcpy)
 #endif
 	str	r2, [r0, #0x01]
 	strb	r1, [r0, #0x05]
-	mov	pc, lr
+	RET
 	LMEMCPY_6_PAD
 
 /*
@@ -1472,7 +1472,7 @@ ENTRY(memcpy)
 	strb	r2, [r0]
 	str	r3, [r0, #0x01]
 	strb	r1, [r0, #0x05]
-	mov	pc, lr
+	RET
 	LMEMCPY_6_PAD
 
 
@@ -1496,7 +1496,7 @@ ENTRY(memcpy)
 	ldr	r3, [r1, #0x04]
 	str	r2, [r0]
 	str	r3, [r0, #0x04]
-	mov	pc, lr
+	RET
 	LMEMCPY_8_PAD
 
 /*
@@ -1517,7 +1517,7 @@ ENTRY(memcpy)
 #endif
 	str	r3, [r0]
 	str	r2, [r0, #0x04]
-	mov	pc, lr
+	RET
 	LMEMCPY_8_PAD
 
 /*
@@ -1537,7 +1537,7 @@ ENTRY(memcpy)
 #endif
 	str	r2, [r0]
 	str	r3, [r0, #0x04]
-	mov	pc, lr
+	RET
 	LMEMCPY_8_PAD
 
 /*
@@ -1558,7 +1558,7 @@ ENTRY(memcpy)
 #endif
 	str	r3, [r0]
 	str	r2, [r0, #0x04]
-	mov	pc, lr
+	RET
 	LMEMCPY_8_PAD
 
 /*
@@ -1583,7 +1583,7 @@ ENTRY(memcpy)
 #endif
 	strh	r1, [r0, #0x01]
 	str	r3, [r0, #0x03]
-	mov	pc, lr
+	RET
 	LMEMCPY_8_PAD
 
 /*
@@ -1597,7 +1597,7 @@ ENTRY(memcpy)
 	strh	r3, [r0, #0x01]
 	str	ip, [r0, #0x03]
 	strb	r1, [r0, #0x07]
-	mov	pc, lr
+	RET
 	LMEMCPY_8_PAD
 
 /*
@@ -1625,7 +1625,7 @@ ENTRY(memcpy)
 #endif
 	strh	ip, [r0, #0x01]
 	str	r3, [r0, #0x03]
-	mov	pc, lr
+	RET
 	LMEMCPY_8_PAD
 
 /*
@@ -1646,7 +1646,7 @@ ENTRY(memcpy)
 #endif
 	str	r2, [r0, #0x03]
 	strb	r1, [r0, #0x07]
-	mov	pc, lr
+	RET
 	LMEMCPY_8_PAD
 
 /*
@@ -1666,7 +1666,7 @@ ENTRY(memcpy)
 #endif
 	str	r2, [r0, #0x02]
 	strh	r3, [r0, #0x06]
-	mov	pc, lr
+	RET
 	LMEMCPY_8_PAD
 
 /*
@@ -1689,7 +1689,7 @@ ENTRY(memcpy)
 #endif
 	str	r1, [r0, #0x02]
 	strh	r3, [r0, #0x06]
-	mov	pc, lr
+	RET
 	LMEMCPY_8_PAD
 
 /*
@@ -1701,7 +1701,7 @@ ENTRY(memcpy)
 	strh	r2, [r0]
 	str	ip, [r0, #0x02]
 	strh	r3, [r0, #0x06]
-	mov	pc, lr
+	RET
 	LMEMCPY_8_PAD
 
 /*
@@ -1724,7 +1724,7 @@ ENTRY(memcpy)
 #endif
 	str	r3, [r0, #0x02]
 	strh	r2, [r0]
-	mov	pc, lr
+	RET
 	LMEMCPY_8_PAD
 
 /*
@@ -1749,7 +1749,7 @@ ENTRY(memcpy)
 	orr	r2, r2, r3, lsl #24	/* r2 = 4321 */
 	str	r2, [r0, #0x01]
 #endif
-	mov	pc, lr
+	RET
 	LMEMCPY_8_PAD
 
 /*
@@ -1770,7 +1770,7 @@ ENTRY(memcpy)
 #endif
 	str	r2, [r0, #0x01]
 	strb	r1, [r0, #0x07]
-	mov	pc, lr
+	RET
 	LMEMCPY_8_PAD
 
 /*
@@ -1798,7 +1798,7 @@ ENTRY(memcpy)
 #endif
 	str	ip, [r0, #0x01]
 	strh	r1, [r0, #0x05]
-	mov	pc, lr
+	RET
 	LMEMCPY_8_PAD
 
 /*
@@ -1812,7 +1812,7 @@ ENTRY(memcpy)
 	str	ip, [r0, #0x01]
 	strh	r3, [r0, #0x05]
 	strb	r1, [r0, #0x07]
-	mov	pc, lr
+	RET
 	LMEMCPY_8_PAD
 
 /******************************************************************************
@@ -1837,7 +1837,7 @@ ENTRY(memcpy)
 	str	r2, [r0]
 	str	r3, [r0, #0x04]
 	str	r1, [r0, #0x08]
-	mov	pc, lr
+	RET
 	LMEMCPY_C_PAD
 
 /*
@@ -1865,7 +1865,7 @@ ENTRY(memcpy)
 #endif
 	str	r2, [r0, #0x04]
 	str	r1, [r0]
-	mov	pc, lr
+	RET
 	LMEMCPY_C_PAD
 
 /*
@@ -1892,7 +1892,7 @@ ENTRY(memcpy)
 #endif
 	str	r3, [r0, #0x04]
 	str	r1, [r0, #0x08]
-	mov	pc, lr
+	RET
 	LMEMCPY_C_PAD
 
 /*
@@ -1920,7 +1920,7 @@ ENTRY(memcpy)
 #endif
 	str	r3, [r0, #0x04]
 	str	r1, [r0, #0x08]
-	mov	pc, lr
+	RET
 	LMEMCPY_C_PAD
 
 /*
@@ -1949,7 +1949,7 @@ ENTRY(memcpy)
 	str	r2, [r0, #0x03]
 	str	r1, [r0, #0x07]
 	strb	ip, [r0, #0x0b]
-	mov	pc, lr
+	RET
 	LMEMCPY_C_PAD
 
 /*
@@ -1965,7 +1965,7 @@ ENTRY(memcpy)
 	str	ip, [r0, #0x03]
 	str	r2, [r0, #0x07]
 	strb	r1, [r0, #0x0b]
-	mov	pc, lr
+	RET
 	LMEMCPY_C_PAD
 
 /*
@@ -1999,7 +1999,7 @@ ENTRY(memcpy)
 	str	r3, [r0, #0x03]
 	str	r2, [r0, #0x07]
 	strb	r1, [r0, #0x0b]
-	mov	pc, lr
+	RET
 	LMEMCPY_C_PAD
 
 /*
@@ -2029,7 +2029,7 @@ ENTRY(memcpy)
 	str	r3, [r0, #0x03]
 	str	ip, [r0, #0x07]
 	strb	r1, [r0, #0x0b]
-	mov	pc, lr
+	RET
 	LMEMCPY_C_PAD
 
 /*
@@ -2055,7 +2055,7 @@ ENTRY(memcpy)
 	str	r1, [r0, #0x02]
 	str	r3, [r0, #0x06]
 	strh	r2, [r0, #0x0a]
-	mov	pc, lr
+	RET
 	LMEMCPY_C_PAD
 
 /*
@@ -2084,7 +2084,7 @@ ENTRY(memcpy)
 	str	r2, [r0, #0x02]
 	str	r3, [r0, #0x06]
 	strh	r1, [r0, #0x0a]
-	mov	pc, lr
+	RET
 	LMEMCPY_C_PAD
 
 /*
@@ -2098,7 +2098,7 @@ ENTRY(memcpy)
 	str	r3, [r0, #0x02]
 	str	ip, [r0, #0x06]
 	strh	r1, [r0, #0x0a]
-	mov	pc, lr
+	RET
 	LMEMCPY_C_PAD
 
 /*
@@ -2127,7 +2127,7 @@ ENTRY(memcpy)
 	str	r2, [r0, #0x06]
 	str	r3, [r0, #0x02]
 	strh	r1, [r0]
-	mov	pc, lr
+	RET
 	LMEMCPY_C_PAD
 
 /*
@@ -2161,7 +2161,7 @@ ENTRY(memcpy)
 	mov	r1, r1, lsr #16		/* r1 = ...B */
 	strb	r1, [r0, #0x0b]
 #endif
-	mov	pc, lr
+	RET
 	LMEMCPY_C_PAD
 
 /*
@@ -2191,7 +2191,7 @@ ENTRY(memcpy)
 	str	r3, [r0, #0x05]
 	str	ip, [r0, #0x01]
 	strb	r1, [r0]
-	mov	pc, lr
+	RET
 	LMEMCPY_C_PAD
 
 /*
@@ -2232,7 +2232,7 @@ ENTRY(memcpy)
 	strh	ip, [r0, #0x09]
 	strb	r1, [r0, #0x0b]
 #endif
-	mov	pc, lr
+	RET
 	LMEMCPY_C_PAD
 
 /*
@@ -2248,7 +2248,7 @@ ENTRY(memcpy)
 	str	ip, [r0, #0x05]
 	strh	r2, [r0, #0x09]
 	strb	r1, [r0, #0x0b]
-	mov	pc, lr
+	RET
 #endif /* __XSCALE__ */
 
 #ifdef GPROF
diff --git a/sys/arm/arm/swtch.S b/sys/arm/arm/swtch.S
index 70c854f..1cfd1f6 100644
--- a/sys/arm/arm/swtch.S
+++ b/sys/arm/arm/swtch.S
@@ -317,7 +317,6 @@ ENTRY(cpu_switch)
 
 	mov     r2, #DOMAIN_CLIENT
 	cmp     r5, r2, lsl #(PMAP_DOMAIN_KERNEL * 2) /* Sw to kernel thread? */
-	
 	beq     .Lcs_cache_purge_skipped        /* Yup. Don't flush cache */
 		
 	/*
@@ -464,7 +463,7 @@ ENTRY(cpu_switch)
 	.asciz	"cpu_switch: sched_qs empty with non-zero sched_whichqs!\n"
 #endif
 ENTRY(savectx)
-	mov	pc, lr
+	RET
 ENTRY(fork_trampoline)
 	mov	r1, r5
 	mov	r2, sp