diff --git a/mpn/x86_64/addaddmul_1msb0.asm b/mpn/x86_64/addaddmul_1msb0.asm
index 87c21b4..2bfa122 100644
--- a/mpn/x86_64/addaddmul_1msb0.asm
+++ b/mpn/x86_64/addaddmul_1msb0.asm
@@ -168,3 +168,4 @@ L(end):	cmp	$1, R32(n)
 	pop	%r12
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/addmul_2.asm b/mpn/x86_64/addmul_2.asm
index 18307d7..2999ce5 100644
--- a/mpn/x86_64/addmul_2.asm
+++ b/mpn/x86_64/addmul_2.asm
@@ -182,3 +182,4 @@ L(end):	xor	R32(w1), R32(w1)
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/aorrlsh1_n.asm b/mpn/x86_64/aorrlsh1_n.asm
index 6ee0872..9ebd7dc 100644
--- a/mpn/x86_64/aorrlsh1_n.asm
+++ b/mpn/x86_64/aorrlsh1_n.asm
@@ -168,3 +168,4 @@ ifdef(`OPERATION_rsblsh1_n',`
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/aorrlshC_n.asm b/mpn/x86_64/aorrlshC_n.asm
index 5a9fd4d..c3d55a6 100644
--- a/mpn/x86_64/aorrlshC_n.asm
+++ b/mpn/x86_64/aorrlshC_n.asm
@@ -158,3 +158,4 @@ ifelse(ADDSUB,add,`
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/aorrlsh_n.asm b/mpn/x86_64/aorrlsh_n.asm
index 5ca128f..7dd0bcf 100644
--- a/mpn/x86_64/aorrlsh_n.asm
+++ b/mpn/x86_64/aorrlsh_n.asm
@@ -174,3 +174,4 @@ L(end):	add	R32(%rbx), R32(%rbx)
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/aors_err1_n.asm b/mpn/x86_64/aors_err1_n.asm
index 54d0b3f..13a6af2 100644
--- a/mpn/x86_64/aors_err1_n.asm
+++ b/mpn/x86_64/aors_err1_n.asm
@@ -223,3 +223,4 @@ L(end):
 	pop	%rbx
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/aors_err2_n.asm b/mpn/x86_64/aors_err2_n.asm
index ce5c2a4..0466f06 100644
--- a/mpn/x86_64/aors_err2_n.asm
+++ b/mpn/x86_64/aors_err2_n.asm
@@ -170,3 +170,4 @@ L(end):
 	pop	%rbx
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/aors_err3_n.asm b/mpn/x86_64/aors_err3_n.asm
index bb6d0c5..cc5461f 100644
--- a/mpn/x86_64/aors_err3_n.asm
+++ b/mpn/x86_64/aors_err3_n.asm
@@ -154,3 +154,4 @@ L(end):
 	pop	%rbx
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/aors_n.asm b/mpn/x86_64/aors_n.asm
index 8941f7a..361e04d 100644
--- a/mpn/x86_64/aors_n.asm
+++ b/mpn/x86_64/aors_n.asm
@@ -167,3 +167,4 @@ L(end):	lea	32(up), up
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/aorsmul_1.asm b/mpn/x86_64/aorsmul_1.asm
index e3fc005..25d0c13 100644
--- a/mpn/x86_64/aorsmul_1.asm
+++ b/mpn/x86_64/aorsmul_1.asm
@@ -178,3 +178,4 @@ IFDOS(``pop	%rdi		'')
 IFDOS(``pop	%rsi		'')
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/atom/addmul_2.asm b/mpn/x86_64/atom/addmul_2.asm
index c1dcdc4..07ae7b8 100644
--- a/mpn/x86_64/atom/addmul_2.asm
+++ b/mpn/x86_64/atom/addmul_2.asm
@@ -184,3 +184,4 @@ L(end):	mul	v1
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/atom/aorrlsh1_n.asm b/mpn/x86_64/atom/aorrlsh1_n.asm
index f44de19..f9d7bac 100644
--- a/mpn/x86_64/atom/aorrlsh1_n.asm
+++ b/mpn/x86_64/atom/aorrlsh1_n.asm
@@ -236,3 +236,4 @@ IFDOS(`	mov	56(%rsp), %r8	')
 	sbb	R32(%rbp), R32(%rbp)	C save acy
 	jmp	L(ent)
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/atom/aorrlsh2_n.asm b/mpn/x86_64/atom/aorrlsh2_n.asm
index 02fb29d..5ea55b4 100644
--- a/mpn/x86_64/atom/aorrlsh2_n.asm
+++ b/mpn/x86_64/atom/aorrlsh2_n.asm
@@ -189,3 +189,4 @@ ifdef(`OPERATION_rsblsh2_n',`
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/atom/aorsmul_1.asm b/mpn/x86_64/atom/aorsmul_1.asm
index e953153..6a12f96 100644
--- a/mpn/x86_64/atom/aorsmul_1.asm
+++ b/mpn/x86_64/atom/aorsmul_1.asm
@@ -188,3 +188,4 @@ L(cj1):	ADDSUB	%rax, (rp,n,8)
 	ret
 EPILOGUE()
 ASM_END()
+CF_PROT
diff --git a/mpn/x86_64/atom/lshift.asm b/mpn/x86_64/atom/lshift.asm
index 1b37d5d..15786cb 100644
--- a/mpn/x86_64/atom/lshift.asm
+++ b/mpn/x86_64/atom/lshift.asm
@@ -121,3 +121,4 @@ L(end):	shl	R8(%rcx), %r10
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/atom/lshiftc.asm b/mpn/x86_64/atom/lshiftc.asm
index 7385f8f..3171d3c 100644
--- a/mpn/x86_64/atom/lshiftc.asm
+++ b/mpn/x86_64/atom/lshiftc.asm
@@ -125,3 +125,4 @@ L(end):	shl	R8(%rcx), %r10
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/atom/mul_1.asm b/mpn/x86_64/atom/mul_1.asm
index d76a3d3..304c208 100644
--- a/mpn/x86_64/atom/mul_1.asm
+++ b/mpn/x86_64/atom/mul_1.asm
@@ -141,3 +141,4 @@ IFDOS(`	mov	56(%rsp), %r8	')
 	jmp	L(com)
 EPILOGUE()
 ASM_END()
+CF_PROT
diff --git a/mpn/x86_64/atom/mul_2.asm b/mpn/x86_64/atom/mul_2.asm
index f3fc3af..c7b78a7 100644
--- a/mpn/x86_64/atom/mul_2.asm
+++ b/mpn/x86_64/atom/mul_2.asm
@@ -184,3 +184,4 @@ L(end):	mul	v1
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/atom/redc_1.asm b/mpn/x86_64/atom/redc_1.asm
index 62b9a84..eeb09d3 100644
--- a/mpn/x86_64/atom/redc_1.asm
+++ b/mpn/x86_64/atom/redc_1.asm
@@ -577,3 +577,4 @@ L(n4):	mov	-32(mp), %rax
 	jmp	L(cj)
 EPILOGUE()
 ASM_END()
+CF_PROT
diff --git a/mpn/x86_64/atom/rsh1aors_n.asm b/mpn/x86_64/atom/rsh1aors_n.asm
index 6f5f638..a589b89 100644
--- a/mpn/x86_64/atom/rsh1aors_n.asm
+++ b/mpn/x86_64/atom/rsh1aors_n.asm
@@ -285,3 +285,4 @@ L(cj1):	pop	%r15
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/atom/rshift.asm b/mpn/x86_64/atom/rshift.asm
index 29c027d..c8b78bf 100644
--- a/mpn/x86_64/atom/rshift.asm
+++ b/mpn/x86_64/atom/rshift.asm
@@ -119,3 +119,4 @@ L(end):	shr	R8(cnt), %r10
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/atom/sublsh1_n.asm b/mpn/x86_64/atom/sublsh1_n.asm
index 1306acd..574b25b 100644
--- a/mpn/x86_64/atom/sublsh1_n.asm
+++ b/mpn/x86_64/atom/sublsh1_n.asm
@@ -240,3 +240,4 @@ IFDOS(`	mov	56(%rsp), %r8	')
 	sbb	R32(%rbp), R32(%rbp)	C save acy
 	jmp	L(ent)
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/bd1/aorsmul_1.asm b/mpn/x86_64/bd1/aorsmul_1.asm
index 96fec9f..ce76154 100644
--- a/mpn/x86_64/bd1/aorsmul_1.asm
+++ b/mpn/x86_64/bd1/aorsmul_1.asm
@@ -179,3 +179,4 @@ IFDOS(``pop	%rsi		'')
 	ret
 EPILOGUE()
 ASM_END()
+CF_PROT
diff --git a/mpn/x86_64/bd1/mul_1.asm b/mpn/x86_64/bd1/mul_1.asm
index e59667c..308f336 100644
--- a/mpn/x86_64/bd1/mul_1.asm
+++ b/mpn/x86_64/bd1/mul_1.asm
@@ -182,3 +182,4 @@ IFDOS(``pop	%rsi		'')
 	ret
 EPILOGUE()
 ASM_END()
+CF_PROT
diff --git a/mpn/x86_64/bd1/mul_2.asm b/mpn/x86_64/bd1/mul_2.asm
index 4ed5f30..f40cf47 100644
--- a/mpn/x86_64/bd1/mul_2.asm
+++ b/mpn/x86_64/bd1/mul_2.asm
@@ -190,3 +190,4 @@ L(end):	mov	-8(up,n,8), %rax
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/bd1/mul_basecase.asm b/mpn/x86_64/bd1/mul_basecase.asm
index e47ba58..6d61cbc 100644
--- a/mpn/x86_64/bd1/mul_basecase.asm
+++ b/mpn/x86_64/bd1/mul_basecase.asm
@@ -414,3 +414,4 @@ L(ret2):pop	%rbp
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/bdiv_dbm1c.asm b/mpn/x86_64/bdiv_dbm1c.asm
index a53bd52..f9c4aa0 100644
--- a/mpn/x86_64/bdiv_dbm1c.asm
+++ b/mpn/x86_64/bdiv_dbm1c.asm
@@ -104,3 +104,4 @@ L(lo1):	sub	%rax, %r8
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/bdiv_q_1.asm b/mpn/x86_64/bdiv_q_1.asm
index 02eacbe..7bfa66d 100644
--- a/mpn/x86_64/bdiv_q_1.asm
+++ b/mpn/x86_64/bdiv_q_1.asm
@@ -165,3 +165,4 @@ L(one):	shr	R8(%rcx), %rax
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/bobcat/aors_n.asm b/mpn/x86_64/bobcat/aors_n.asm
index 22287b8..1df1a08 100644
--- a/mpn/x86_64/bobcat/aors_n.asm
+++ b/mpn/x86_64/bobcat/aors_n.asm
@@ -148,3 +148,4 @@ PROLOGUE(func_nc)
 IFDOS(`	mov	56(%rsp), %r8	')
 	jmp	L(ent)
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/bobcat/aorsmul_1.asm b/mpn/x86_64/bobcat/aorsmul_1.asm
index 415a17c..79d81f4 100644
--- a/mpn/x86_64/bobcat/aorsmul_1.asm
+++ b/mpn/x86_64/bobcat/aorsmul_1.asm
@@ -181,3 +181,4 @@ IFDOS(`	pop	%rdi		')
 IFDOS(`	pop	%rsi		')
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/bobcat/copyd.asm b/mpn/x86_64/bobcat/copyd.asm
index 877714e..2f781a3 100644
--- a/mpn/x86_64/bobcat/copyd.asm
+++ b/mpn/x86_64/bobcat/copyd.asm
@@ -89,3 +89,4 @@ L(end):	cmp	$-4, R32(n)
 L(ret):	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/bobcat/copyi.asm b/mpn/x86_64/bobcat/copyi.asm
index ee0f578..ff249bc 100644
--- a/mpn/x86_64/bobcat/copyi.asm
+++ b/mpn/x86_64/bobcat/copyi.asm
@@ -92,3 +92,4 @@ L(end):	cmp	$4, R32(n)
 L(ret):	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/bobcat/mul_1.asm b/mpn/x86_64/bobcat/mul_1.asm
index ab428a8..b4f401b 100644
--- a/mpn/x86_64/bobcat/mul_1.asm
+++ b/mpn/x86_64/bobcat/mul_1.asm
@@ -185,3 +185,4 @@ IFDOS(`	pop	%rdi		')
 IFDOS(`	pop	%rsi		')
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/bobcat/mul_basecase.asm b/mpn/x86_64/bobcat/mul_basecase.asm
index e7d46bf..14c7b13 100644
--- a/mpn/x86_64/bobcat/mul_basecase.asm
+++ b/mpn/x86_64/bobcat/mul_basecase.asm
@@ -484,3 +484,4 @@ L(ret):	pop	%r13
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/bobcat/redc_1.asm b/mpn/x86_64/bobcat/redc_1.asm
index d55b1e5..d686cfb 100644
--- a/mpn/x86_64/bobcat/redc_1.asm
+++ b/mpn/x86_64/bobcat/redc_1.asm
@@ -505,3 +505,4 @@ L(n3):	mov	-24(mp), %rax
 	jmp	L(ret)
 EPILOGUE()
 ASM_END()
+CF_PROT
diff --git a/mpn/x86_64/bobcat/sqr_basecase.asm b/mpn/x86_64/bobcat/sqr_basecase.asm
index 0e417a1..5693c46 100644
--- a/mpn/x86_64/bobcat/sqr_basecase.asm
+++ b/mpn/x86_64/bobcat/sqr_basecase.asm
@@ -563,3 +563,4 @@ L(esd):	add	%rbx, w0
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/cnd_aors_n.asm b/mpn/x86_64/cnd_aors_n.asm
index 13a2ab3..35f30e7 100644
--- a/mpn/x86_64/cnd_aors_n.asm
+++ b/mpn/x86_64/cnd_aors_n.asm
@@ -181,3 +181,4 @@ L(end):	neg	R32(%rax)
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/com.asm b/mpn/x86_64/com.asm
index 006acaf..56b0747 100644
--- a/mpn/x86_64/com.asm
+++ b/mpn/x86_64/com.asm
@@ -93,3 +93,4 @@ L(e10):	movq	24(up,n,8), %r9
 L(ret):	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/copyd.asm b/mpn/x86_64/copyd.asm
index a5e6e59..020e287 100644
--- a/mpn/x86_64/copyd.asm
+++ b/mpn/x86_64/copyd.asm
@@ -91,3 +91,4 @@ L(end):	shr	R32(n)
 	mov	%r9, -16(rp)
 1:	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/copyi.asm b/mpn/x86_64/copyi.asm
index bafce7a..1a4fb6d 100644
--- a/mpn/x86_64/copyi.asm
+++ b/mpn/x86_64/copyi.asm
@@ -90,3 +90,4 @@ L(end):	shr	R32(n)
 	mov	%r9, 16(rp)
 1:	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/core2/aors_err1_n.asm b/mpn/x86_64/core2/aors_err1_n.asm
index 3f875ae..5162272 100644
--- a/mpn/x86_64/core2/aors_err1_n.asm
+++ b/mpn/x86_64/core2/aors_err1_n.asm
@@ -223,3 +223,4 @@ L(end):
 	pop	%rbx
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/core2/aors_n.asm b/mpn/x86_64/core2/aors_n.asm
index 74a1bce..19078d8 100644
--- a/mpn/x86_64/core2/aors_n.asm
+++ b/mpn/x86_64/core2/aors_n.asm
@@ -139,3 +139,4 @@ IFDOS(`	mov	56(%rsp), %r8	')
 	jmp	L(start)
 EPILOGUE()
 
+CF_PROT
diff --git a/mpn/x86_64/core2/aorsmul_1.asm b/mpn/x86_64/core2/aorsmul_1.asm
index 6b313dd..392f4de 100644
--- a/mpn/x86_64/core2/aorsmul_1.asm
+++ b/mpn/x86_64/core2/aorsmul_1.asm
@@ -176,3 +176,4 @@ L(n1):	mov	8(rp), %r10
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/core2/divrem_1.asm b/mpn/x86_64/core2/divrem_1.asm
index 1b3f139..0a67dc3 100644
--- a/mpn/x86_64/core2/divrem_1.asm
+++ b/mpn/x86_64/core2/divrem_1.asm
@@ -241,3 +241,4 @@ L(ret):	pop	%rbx
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/core2/gcd_1.asm b/mpn/x86_64/core2/gcd_1.asm
index bdb940c..452b763 100644
--- a/mpn/x86_64/core2/gcd_1.asm
+++ b/mpn/x86_64/core2/gcd_1.asm
@@ -144,3 +144,4 @@ L(end):	pop	%rcx
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/core2/lshift.asm b/mpn/x86_64/core2/lshift.asm
index 8ccafec..00b39b8 100644
--- a/mpn/x86_64/core2/lshift.asm
+++ b/mpn/x86_64/core2/lshift.asm
@@ -147,3 +147,4 @@ L(end):	shld	R8(cnt), %r8, %r11
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/core2/lshiftc.asm b/mpn/x86_64/core2/lshiftc.asm
index 65c7b2f..4d3acfe 100644
--- a/mpn/x86_64/core2/lshiftc.asm
+++ b/mpn/x86_64/core2/lshiftc.asm
@@ -157,3 +157,4 @@ L(end):	shld	R8(cnt), %r8, %r11
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/core2/mul_basecase.asm b/mpn/x86_64/core2/mul_basecase.asm
index d16be85..04cd4c2 100644
--- a/mpn/x86_64/core2/mul_basecase.asm
+++ b/mpn/x86_64/core2/mul_basecase.asm
@@ -973,3 +973,4 @@ L(lo3):	mul	v0
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/core2/mullo_basecase.asm b/mpn/x86_64/core2/mullo_basecase.asm
index 0f03d86..efed03d 100644
--- a/mpn/x86_64/core2/mullo_basecase.asm
+++ b/mpn/x86_64/core2/mullo_basecase.asm
@@ -425,3 +425,4 @@ L(n3):	mov	(vp_param), %r9
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/core2/redc_1.asm b/mpn/x86_64/core2/redc_1.asm
index 8c296fd..d98f56f 100644
--- a/mpn/x86_64/core2/redc_1.asm
+++ b/mpn/x86_64/core2/redc_1.asm
@@ -428,3 +428,4 @@ L(n4):	mov	-32(mp), %rax
 	jmp	L(add_n)
 EPILOGUE()
 ASM_END()
+CF_PROT
diff --git a/mpn/x86_64/core2/rsh1aors_n.asm b/mpn/x86_64/core2/rsh1aors_n.asm
index 27eed37..579fec6 100644
--- a/mpn/x86_64/core2/rsh1aors_n.asm
+++ b/mpn/x86_64/core2/rsh1aors_n.asm
@@ -167,3 +167,4 @@ L(end):	shrd	$1, %rbx, %rbp
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/core2/rshift.asm b/mpn/x86_64/core2/rshift.asm
index ab32ec8..97f4429 100644
--- a/mpn/x86_64/core2/rshift.asm
+++ b/mpn/x86_64/core2/rshift.asm
@@ -145,3 +145,4 @@ L(end):	shrd	R8(cnt), %r8, %r11
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/core2/sqr_basecase.asm b/mpn/x86_64/core2/sqr_basecase.asm
index a112c1b..0ee6ca3 100644
--- a/mpn/x86_64/core2/sqr_basecase.asm
+++ b/mpn/x86_64/core2/sqr_basecase.asm
@@ -982,3 +982,4 @@ L(n3):	mov	%rax, %r10
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/core2/sublshC_n.asm b/mpn/x86_64/core2/sublshC_n.asm
index 5acc46b..7a48dfb 100644
--- a/mpn/x86_64/core2/sublshC_n.asm
+++ b/mpn/x86_64/core2/sublshC_n.asm
@@ -156,3 +156,4 @@ L(end):	shr	$RSH, %r11
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/coreibwl/addmul_1.asm b/mpn/x86_64/coreibwl/addmul_1.asm
index aaa58e7..4fb79f9 100644
--- a/mpn/x86_64/coreibwl/addmul_1.asm
+++ b/mpn/x86_64/coreibwl/addmul_1.asm
@@ -107,33 +107,39 @@ L(tab):	JMPENT(	L(f0), L(tab))
 	JMPENT(	L(f7), L(tab))
 	TEXT
 
-L(f0):	mulx(	(up), %r10, %r8)
+L(f0):	CFPROT_ENDBR
+	mulx(	(up), %r10, %r8)
 	lea	-8(up), up
 	lea	-8(rp), rp
 	lea	-1(n), n
 	jmp	L(b0)
 
-L(f3):	mulx(	(up), %r9, %rax)
+L(f3):	CFPROT_ENDBR
+	mulx(	(up), %r9, %rax)
 	lea	16(up), up
 	lea	-48(rp), rp
 	jmp	L(b3)
 
-L(f4):	mulx(	(up), %r10, %r8)
+L(f4):	CFPROT_ENDBR
+	mulx(	(up), %r10, %r8)
 	lea	24(up), up
 	lea	-40(rp), rp
 	jmp	L(b4)
 
-L(f5):	mulx(	(up), %r9, %rax)
+L(f5):	CFPROT_ENDBR
+	mulx(	(up), %r9, %rax)
 	lea	32(up), up
 	lea	-32(rp), rp
 	jmp	L(b5)
 
-L(f6):	mulx(	(up), %r10, %r8)
+L(f6):	CFPROT_ENDBR
+	mulx(	(up), %r10, %r8)
 	lea	40(up), up
 	lea	-24(rp), rp
 	jmp	L(b6)
 
-L(f1):	mulx(	(up), %r9, %rax)
+L(f1):	CFPROT_ENDBR
+	mulx(	(up), %r9, %rax)
 	jrcxz	L(1)
 	jmp	L(b1)
 L(1):	add	(rp), %r9
@@ -151,7 +157,8 @@ ifdef(`PIC',
 `	nop;nop;nop;nop',
 `	nop;nop;nop;nop;nop;nop;nop;nop;nop;nop;nop')
 
-L(f2):	mulx(	(up), %r10, %r8)
+L(f2):	CFPROT_ENDBR
+	mulx(	(up), %r10, %r8)
 	lea	8(up), up
 	lea	8(rp), rp
 	mulx(	(up), %r9, %rax)
@@ -195,9 +202,11 @@ L(b3):	adox(	48,(rp), %r9)
 	mulx(	(up), %r9, %rax)
 	jmp	L(top)
 
-L(f7):	mulx(	(up), %r9, %rax)
+L(f7):	CFPROT_ENDBR
+	mulx(	(up), %r9, %rax)
 	lea	-16(up), up
 	lea	-16(rp), rp
 	jmp	L(b7)
 EPILOGUE()
 ASM_END()
+CF_PROT
diff --git a/mpn/x86_64/coreibwl/mul_1.asm b/mpn/x86_64/coreibwl/mul_1.asm
index a271e6c..4fe4822 100644
--- a/mpn/x86_64/coreibwl/mul_1.asm
+++ b/mpn/x86_64/coreibwl/mul_1.asm
@@ -106,48 +106,56 @@ L(tab):	JMPENT(	L(f0), L(tab))
 	JMPENT(	L(f7), L(tab))
 	TEXT
 
-L(f0):	mulx(	(up), %r10, %r8)
+L(f0):	CFPROT_ENDBR
+	mulx(	(up), %r10, %r8)
 	lea	56(up), up
 	lea	-8(rp), rp
 	jmp	L(b0)
 
-L(f3):	mulx(	(up), %r9, %rax)
+L(f3):	CFPROT_ENDBR
+	mulx(	(up), %r9, %rax)
 	lea	16(up), up
 	lea	16(rp), rp
 	inc	n
 	jmp	L(b3)
 
-L(f4):	mulx(	(up), %r10, %r8)
+L(f4):	CFPROT_ENDBR
+	mulx(	(up), %r10, %r8)
 	lea	24(up), up
 	lea	24(rp), rp
 	inc	n
 	jmp	L(b4)
 
-L(f5):	mulx(	(up), %r9, %rax)
+L(f5):	CFPROT_ENDBR
+	mulx(	(up), %r9, %rax)
 	lea	32(up), up
 	lea	32(rp), rp
 	inc	n
 	jmp	L(b5)
 
-L(f6):	mulx(	(up), %r10, %r8)
+L(f6):	CFPROT_ENDBR
+	mulx(	(up), %r10, %r8)
 	lea	40(up), up
 	lea	40(rp), rp
 	inc	n
 	jmp	L(b6)
 
-L(f7):	mulx(	(up), %r9, %rax)
+L(f7):	CFPROT_ENDBR
+	mulx(	(up), %r9, %rax)
 	lea	48(up), up
 	lea	48(rp), rp
 	inc	n
 	jmp	L(b7)
 
-L(f1):	mulx(	(up), %r9, %rax)
+L(f1):	CFPROT_ENDBR
+	mulx(	(up), %r9, %rax)
 	test	n, n
 	jnz	L(b1)
 L(1):	mov	%r9, (rp)
 	ret
 
-L(f2):	mulx(	(up), %r10, %r8)
+L(f2):	CFPROT_ENDBR
+	mulx(	(up), %r10, %r8)
 	lea	8(up), up
 	lea	8(rp), rp
 	mulx(	(up), %r9, %rax)
@@ -191,3 +199,4 @@ L(end):	mov	%r10, -8(rp)
 	ret
 EPILOGUE()
 ASM_END()
+CF_PROT
diff --git a/mpn/x86_64/coreibwl/mul_basecase.asm b/mpn/x86_64/coreibwl/mul_basecase.asm
index 50f3ce5..74cd67c 100644
--- a/mpn/x86_64/coreibwl/mul_basecase.asm
+++ b/mpn/x86_64/coreibwl/mul_basecase.asm
@@ -155,45 +155,53 @@ ifdef(`PIC',
 	jmp	*(%r10,%rax,8)
 ')
 
-L(mf0):	mulx(	(up), w2, w3)
+L(mf0):	CFPROT_ENDBR
+	mulx(	(up), w2, w3)
 	lea	56(up), up
 	lea	-8(rp), rp
 	jmp	L(mb0)
 
-L(mf3):	mulx(	(up), w0, w1)
+L(mf3):	CFPROT_ENDBR
+	mulx(	(up), w0, w1)
 	lea	16(up), up
 	lea	16(rp), rp
 	inc	n
 	jmp	L(mb3)
 
-L(mf4):	mulx(	(up), w2, w3)
+L(mf4):	CFPROT_ENDBR
+	mulx(	(up), w2, w3)
 	lea	24(up), up
 	lea	24(rp), rp
 	inc	n
 	jmp	L(mb4)
 
-L(mf5):	mulx(	(up), w0, w1)
+L(mf5):	CFPROT_ENDBR
+	mulx(	(up), w0, w1)
 	lea	32(up), up
 	lea	32(rp), rp
 	inc	n
 	jmp	L(mb5)
 
-L(mf6):	mulx(	(up), w2, w3)
+L(mf6):	CFPROT_ENDBR
+	mulx(	(up), w2, w3)
 	lea	40(up), up
 	lea	40(rp), rp
 	inc	n
 	jmp	L(mb6)
 
-L(mf7):	mulx(	(up), w0, w1)
+L(mf7):	CFPROT_ENDBR
+	mulx(	(up), w0, w1)
 	lea	48(up), up
 	lea	48(rp), rp
 	inc	n
 	jmp	L(mb7)
 
-L(mf1):	mulx(	(up), w0, w1)
+L(mf1):	CFPROT_ENDBR
+	mulx(	(up), w0, w1)
 	jmp	L(mb1)
 
-L(mf2):	mulx(	(up), w2, w3)
+L(mf2):	CFPROT_ENDBR
+	mulx(	(up), w2, w3)
 	lea	8(up), up
 	lea	8(rp), rp
 	mulx(	(up), w0, w1)
@@ -254,32 +262,39 @@ L(outer):
 	lea	8(vp), vp
 	jmp	*jaddr
 
-L(f0):	mulx(	8,(up), w2, w3)
+L(f0):	CFPROT_ENDBR
+	mulx(	8,(up), w2, w3)
 	lea	8(rp,unneg,8), rp
 	lea	-1(n), n
 	jmp	L(b0)
 
-L(f3):	mulx(	-16,(up), w0, w1)
+L(f3):	CFPROT_ENDBR
+	mulx(	-16,(up), w0, w1)
 	lea	-56(rp,unneg,8), rp
 	jmp	L(b3)
 
-L(f4):	mulx(	-24,(up), w2, w3)
+L(f4):	CFPROT_ENDBR
+	mulx(	-24,(up), w2, w3)
 	lea	-56(rp,unneg,8), rp
 	jmp	L(b4)
 
-L(f5):	mulx(	-32,(up), w0, w1)
+L(f5):	CFPROT_ENDBR
+	mulx(	-32,(up), w0, w1)
 	lea	-56(rp,unneg,8), rp
 	jmp	L(b5)
 
-L(f6):	mulx(	-40,(up), w2, w3)
+L(f6):	CFPROT_ENDBR
+	mulx(	-40,(up), w2, w3)
 	lea	-56(rp,unneg,8), rp
 	jmp	L(b6)
 
-L(f7):	mulx(	16,(up), w0, w1)
+L(f7):	CFPROT_ENDBR
+	mulx(	16,(up), w0, w1)
 	lea	8(rp,unneg,8), rp
 	jmp	L(b7)
 
-L(f1):	mulx(	(up), w0, w1)
+L(f1):	CFPROT_ENDBR
+	mulx(	(up), w0, w1)
 	lea	8(rp,unneg,8), rp
 	jmp	L(b1)
 
@@ -300,7 +315,7 @@ L(done):
 	FUNC_EXIT()
 	ret
 
-L(f2):
+L(f2):	CFPROT_ENDBR
 	mulx(	-8,(up), w2, w3)
 	lea	8(rp,unneg,8), rp
 	mulx(	(up), w0, w1)
@@ -365,3 +380,4 @@ L(atab):JMPENT(	L(f0), L(atab))
 	JMPENT(	L(f7), L(atab))
 	TEXT
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/coreibwl/sqr_basecase.asm b/mpn/x86_64/coreibwl/sqr_basecase.asm
index 447ba00..64e8298 100644
--- a/mpn/x86_64/coreibwl/sqr_basecase.asm
+++ b/mpn/x86_64/coreibwl/sqr_basecase.asm
@@ -184,42 +184,50 @@ ifdef(`PIC',
 	jmp	*(%r10,%rax,8)
 ')
 
-L(mf0):	mulx(	8,(up), w2, w3)
+L(mf0):	CFPROT_ENDBR
+	mulx(	8,(up), w2, w3)
 	lea	64(up), up
 C	lea	(rp), rp
 	jmp	L(mb0)
 
-L(mf3):	mulx(	8,(up), w0, w1)
+L(mf3):	CFPROT_ENDBR
+	mulx(	8,(up), w0, w1)
 	lea	24(up), up
 	lea	24(rp), rp
 	jmp	L(mb3)
 
-L(mf4):	mulx(	8,(up), w2, w3)
+L(mf4):	CFPROT_ENDBR
+	mulx(	8,(up), w2, w3)
 	lea	32(up), up
 	lea	32(rp), rp
 	jmp	L(mb4)
 
-L(mf5):	mulx(	8,(up), w0, w1)
+L(mf5):	CFPROT_ENDBR
+	mulx(	8,(up), w0, w1)
 	lea	40(up), up
 	lea	40(rp), rp
 	jmp	L(mb5)
 
-L(mf6):	mulx(	8,(up), w2, w3)
+L(mf6):	CFPROT_ENDBR
+	mulx(	8,(up), w2, w3)
 	lea	48(up), up
 	lea	48(rp), rp
 	jmp	L(mb6)
 
-L(mf7):	mulx(	8,(up), w0, w1)
+L(mf7):	CFPROT_ENDBR
+	mulx(	8,(up), w0, w1)
 	lea	56(up), up
 	lea	56(rp), rp
 	jmp	L(mb7)
 
-L(mf1):	mulx(	8,(up), w0, w1)
+L(mf1):	CFPROT_ENDBR
+	mulx(	8,(up), w0, w1)
 	lea	8(up), up
 	lea	8(rp), rp
 	jmp	L(mb1)
 
-L(mf2):	mulx(	8,(up), w2, w3)
+L(mf2):	CFPROT_ENDBR
+	mulx(	8,(up), w2, w3)
 	lea	16(up), up
 	lea	16(rp), rp
 	dec	R32(n)
@@ -275,7 +283,8 @@ L(ed0):	adox(	(rp), w0)
 	mov	w0, (rp)
 	adc	%rcx, w1		C relies on rcx = 0
 	mov	w1, 8(rp)
-L(f7):	lea	-64(up,un_save,8), up
+L(f7):	CFPROT_ENDBR
+	lea	-64(up,un_save,8), up
 	or	R32(un_save), R32(n)
 	mov	8(up), u0
 	mulx(	16,(up), w0, w1)
@@ -326,7 +335,8 @@ L(ed1):	adox(	(rp), w0)
 	mov	w0, (rp)
 	adc	%rcx, w1		C relies on rcx = 0
 	mov	w1, 8(rp)
-L(f0):	lea	-64(up,un_save,8), up
+L(f0):	CFPROT_ENDBR
+	lea	-64(up,un_save,8), up
 	or	R32(un_save), R32(n)
 	mov	(up), u0
 	mulx(	8,(up), w2, w3)
@@ -377,7 +387,8 @@ L(ed2):	adox(	(rp), w0)
 	mov	w0, (rp)
 	adc	%rcx, w1		C relies on rcx = 0
 	mov	w1, 8(rp)
-L(f1):	lea	(up,un_save,8), up
+L(f1):	CFPROT_ENDBR
+	lea	(up,un_save,8), up
 	or	R32(un_save), R32(n)
 	lea	8(un_save), un_save
 	mov	-8(up), u0
@@ -429,7 +440,8 @@ L(ed3):	adox(	(rp), w0)
 	mov	w0, (rp)
 	adc	%rcx, w1		C relies on rcx = 0
 	mov	w1, 8(rp)
-L(f2):	lea	(up,un_save,8), up
+L(f2):	CFPROT_ENDBR
+	lea	(up,un_save,8), up
 	or	R32(un_save), R32(n)
 	jz	L(corner2)
 	mov	-16(up), u0
@@ -482,7 +494,8 @@ L(ed4):	adox(	(rp), w0)
 	mov	w0, (rp)
 	adc	%rcx, w1		C relies on rcx = 0
 	mov	w1, 8(rp)
-L(f3):	lea	(up,un_save,8), up
+L(f3):	CFPROT_ENDBR
+	lea	(up,un_save,8), up
 	or	R32(un_save), R32(n)
 	jz	L(corner3)
 	mov	-24(up), u0
@@ -534,7 +547,8 @@ L(ed5):	adox(	(rp), w0)
 	mov	w0, (rp)
 	adc	%rcx, w1		C relies on rcx = 0
 	mov	w1, 8(rp)
-L(f4):	lea	(up,un_save,8), up
+L(f4):	CFPROT_ENDBR
+	lea	(up,un_save,8), up
 	or	R32(un_save), R32(n)
 	mov	-32(up), u0
 	mulx(	-24,(up), w2, w3)
@@ -585,7 +599,8 @@ L(ed6):	adox(	(rp), w0)
 	mov	w0, (rp)
 	adc	%rcx, w1		C relies on rcx = 0
 	mov	w1, 8(rp)
-L(f5):	lea	(up,un_save,8), up
+L(f5):	CFPROT_ENDBR
+	lea	(up,un_save,8), up
 	or	R32(un_save), R32(n)
 	mov	-40(up), u0
 	mulx(	-32,(up), w0, w1)
@@ -636,7 +651,8 @@ L(ed7):	adox(	(rp), w0)
 	mov	w0, (rp)
 	adc	%rcx, w1		C relies on rcx = 0
 	mov	w1, 8(rp)
-L(f6):	lea	(up,un_save,8), up
+L(f6):	CFPROT_ENDBR
+	lea	(up,un_save,8), up
 	or	R32(un_save), R32(n)
 	mov	-48(up), u0
 	mulx(	-40,(up), w2, w3)
@@ -838,3 +854,4 @@ L(atab):JMPENT(	L(f6), L(atab))
 	JMPENT(	L(f5), L(atab))
 	TEXT
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/coreihwl/addmul_2.asm b/mpn/x86_64/coreihwl/addmul_2.asm
index 54aebc8..2a5f996 100644
--- a/mpn/x86_64/coreihwl/addmul_2.asm
+++ b/mpn/x86_64/coreihwl/addmul_2.asm
@@ -236,3 +236,4 @@ L(end):	mulx(	v0, %rax, w3)
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/coreihwl/aorsmul_1.asm b/mpn/x86_64/coreihwl/aorsmul_1.asm
index fd5a26d..8c03b17 100644
--- a/mpn/x86_64/coreihwl/aorsmul_1.asm
+++ b/mpn/x86_64/coreihwl/aorsmul_1.asm
@@ -196,3 +196,4 @@ L(ret):	pop	%r13
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/coreihwl/mul_1.asm b/mpn/x86_64/coreihwl/mul_1.asm
index 1e3c338..b6463f9 100644
--- a/mpn/x86_64/coreihwl/mul_1.asm
+++ b/mpn/x86_64/coreihwl/mul_1.asm
@@ -153,3 +153,4 @@ L(cj1):	mov	%rbx, 24(rp)
 	ret
 EPILOGUE()
 ASM_END()
+CF_PROT
diff --git a/mpn/x86_64/coreihwl/mul_2.asm b/mpn/x86_64/coreihwl/mul_2.asm
index 5bdb1aa..21defe9 100644
--- a/mpn/x86_64/coreihwl/mul_2.asm
+++ b/mpn/x86_64/coreihwl/mul_2.asm
@@ -171,3 +171,4 @@ L(end):	mulx(	v1, %rdx, %rax)
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/coreihwl/mul_basecase.asm b/mpn/x86_64/coreihwl/mul_basecase.asm
index b2656c8..e4a8381 100644
--- a/mpn/x86_64/coreihwl/mul_basecase.asm
+++ b/mpn/x86_64/coreihwl/mul_basecase.asm
@@ -439,3 +439,4 @@ L(ret2):pop	%rbp
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/coreihwl/mullo_basecase.asm b/mpn/x86_64/coreihwl/mullo_basecase.asm
index 9986e8b..6756802 100644
--- a/mpn/x86_64/coreihwl/mullo_basecase.asm
+++ b/mpn/x86_64/coreihwl/mullo_basecase.asm
@@ -424,3 +424,4 @@ L(n3):	mov	(vp), %r9
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/coreihwl/redc_1.asm b/mpn/x86_64/coreihwl/redc_1.asm
index b1d6c0a..b8b4a9e 100644
--- a/mpn/x86_64/coreihwl/redc_1.asm
+++ b/mpn/x86_64/coreihwl/redc_1.asm
@@ -435,3 +435,4 @@ L(ret):	pop	%r15
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/coreihwl/sqr_basecase.asm b/mpn/x86_64/coreihwl/sqr_basecase.asm
index 641cdf3..8e83470 100644
--- a/mpn/x86_64/coreihwl/sqr_basecase.asm
+++ b/mpn/x86_64/coreihwl/sqr_basecase.asm
@@ -504,3 +504,4 @@ L(dend):adc	%rbx, %rdx
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/coreinhm/aorrlsh_n.asm b/mpn/x86_64/coreinhm/aorrlsh_n.asm
index eed64e7..b1a4610 100644
--- a/mpn/x86_64/coreinhm/aorrlsh_n.asm
+++ b/mpn/x86_64/coreinhm/aorrlsh_n.asm
@@ -198,3 +198,4 @@ IFDOS(`	mov	64(%rsp), %r9	')	C cy
 	sbb	R32(%rbx), R32(%rbx)	C initialise CF save register
 	jmp	L(ent)
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/coreinhm/aorsmul_1.asm b/mpn/x86_64/coreinhm/aorsmul_1.asm
index b768905..e2d96a8 100644
--- a/mpn/x86_64/coreinhm/aorsmul_1.asm
+++ b/mpn/x86_64/coreinhm/aorsmul_1.asm
@@ -185,3 +185,4 @@ L(end):	mul	v0
 	ret
 EPILOGUE()
 ASM_END()
+CF_PROT
diff --git a/mpn/x86_64/coreinhm/redc_1.asm b/mpn/x86_64/coreinhm/redc_1.asm
index fc71c1b..782da6b 100644
--- a/mpn/x86_64/coreinhm/redc_1.asm
+++ b/mpn/x86_64/coreinhm/redc_1.asm
@@ -547,3 +547,4 @@ L(n3):	mov	-24(mp), %rax
 	jmp	L(ret)
 EPILOGUE()
 ASM_END()
+CF_PROT
diff --git a/mpn/x86_64/coreisbr/addmul_2.asm b/mpn/x86_64/coreisbr/addmul_2.asm
index 21f0bf4..e6ffe3e 100644
--- a/mpn/x86_64/coreisbr/addmul_2.asm
+++ b/mpn/x86_64/coreisbr/addmul_2.asm
@@ -222,3 +222,4 @@ L(end):	mul	v1
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/coreisbr/aorrlshC_n.asm b/mpn/x86_64/coreisbr/aorrlshC_n.asm
index 23ace41..75a9b8c 100644
--- a/mpn/x86_64/coreisbr/aorrlshC_n.asm
+++ b/mpn/x86_64/coreisbr/aorrlshC_n.asm
@@ -171,3 +171,4 @@ L(end):	shr	$RSH, %rbp
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/coreisbr/aorrlsh_n.asm b/mpn/x86_64/coreisbr/aorrlsh_n.asm
index db8ee68..611dcb2 100644
--- a/mpn/x86_64/coreisbr/aorrlsh_n.asm
+++ b/mpn/x86_64/coreisbr/aorrlsh_n.asm
@@ -213,3 +213,4 @@ IFDOS(`	mov	64(%rsp), %r9	')	C cy
 	sbb	R32(%rbx), R32(%rbx)	C initialise CF save register
 	jmp	L(ent)
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/coreisbr/aors_n.asm b/mpn/x86_64/coreisbr/aors_n.asm
index 01abf78..07fef16 100644
--- a/mpn/x86_64/coreisbr/aors_n.asm
+++ b/mpn/x86_64/coreisbr/aors_n.asm
@@ -196,3 +196,4 @@ PROLOGUE(func_nc)
 IFDOS(`	mov	56(%rsp), %r8	')
 	jmp	L(ent)
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/coreisbr/aorsmul_1.asm b/mpn/x86_64/coreisbr/aorsmul_1.asm
index 9f01d9c..41b8016 100644
--- a/mpn/x86_64/coreisbr/aorsmul_1.asm
+++ b/mpn/x86_64/coreisbr/aorsmul_1.asm
@@ -207,3 +207,4 @@ IFDOS(``pop	%rsi		'')
 	ret
 EPILOGUE()
 ASM_END()
+CF_PROT
diff --git a/mpn/x86_64/coreisbr/mul_1.asm b/mpn/x86_64/coreisbr/mul_1.asm
index ded7d89..a30f00b 100644
--- a/mpn/x86_64/coreisbr/mul_1.asm
+++ b/mpn/x86_64/coreisbr/mul_1.asm
@@ -159,3 +159,4 @@ IFDOS(``pop	%rdi		'')
 IFDOS(``pop	%rsi		'')
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/coreisbr/mul_2.asm b/mpn/x86_64/coreisbr/mul_2.asm
index ffee78a..991820b 100644
--- a/mpn/x86_64/coreisbr/mul_2.asm
+++ b/mpn/x86_64/coreisbr/mul_2.asm
@@ -161,3 +161,4 @@ L(end):	mul	v0
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/coreisbr/mul_basecase.asm b/mpn/x86_64/coreisbr/mul_basecase.asm
index 35fd1cc..063664b 100644
--- a/mpn/x86_64/coreisbr/mul_basecase.asm
+++ b/mpn/x86_64/coreisbr/mul_basecase.asm
@@ -405,3 +405,4 @@ L(ret2):pop	%rbp
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/coreisbr/mullo_basecase.asm b/mpn/x86_64/coreisbr/mullo_basecase.asm
index a41a8ac..1b75c78 100644
--- a/mpn/x86_64/coreisbr/mullo_basecase.asm
+++ b/mpn/x86_64/coreisbr/mullo_basecase.asm
@@ -382,3 +382,4 @@ L(n3):	mov	(vp_param), %r9
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/coreisbr/popcount.asm b/mpn/x86_64/coreisbr/popcount.asm
index a5be33e..426d3a6 100644
--- a/mpn/x86_64/coreisbr/popcount.asm
+++ b/mpn/x86_64/coreisbr/popcount.asm
@@ -116,3 +116,4 @@ L(cj1):	add	%r11, %rax
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/coreisbr/redc_1.asm b/mpn/x86_64/coreisbr/redc_1.asm
index f0dbe07..710e60e 100644
--- a/mpn/x86_64/coreisbr/redc_1.asm
+++ b/mpn/x86_64/coreisbr/redc_1.asm
@@ -544,3 +544,4 @@ L(n3):	mov	-32(mp), %rax
 	jmp	L(cj)
 EPILOGUE()
 ASM_END()
+CF_PROT
diff --git a/mpn/x86_64/coreisbr/rsh1aors_n.asm b/mpn/x86_64/coreisbr/rsh1aors_n.asm
index fd2eaea..d390ff3 100644
--- a/mpn/x86_64/coreisbr/rsh1aors_n.asm
+++ b/mpn/x86_64/coreisbr/rsh1aors_n.asm
@@ -191,3 +191,4 @@ L(end):	shrd	$1, %rbx, %rbp
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/coreisbr/sqr_basecase.asm b/mpn/x86_64/coreisbr/sqr_basecase.asm
index 46a3612..4d4e545 100644
--- a/mpn/x86_64/coreisbr/sqr_basecase.asm
+++ b/mpn/x86_64/coreisbr/sqr_basecase.asm
@@ -482,3 +482,4 @@ L(dend):add	%r8, %r10
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/div_qr_1n_pi1.asm b/mpn/x86_64/div_qr_1n_pi1.asm
index cb072e9..5a4f195 100644
--- a/mpn/x86_64/div_qr_1n_pi1.asm
+++ b/mpn/x86_64/div_qr_1n_pi1.asm
@@ -245,3 +245,4 @@ L(q_incr_loop):
 	lea	8(U1), U1
 	jmp	L(q_incr_loop)
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/div_qr_2n_pi1.asm b/mpn/x86_64/div_qr_2n_pi1.asm
index 5e59a0a..252781c 100644
--- a/mpn/x86_64/div_qr_2n_pi1.asm
+++ b/mpn/x86_64/div_qr_2n_pi1.asm
@@ -156,3 +156,4 @@ L(fix):	C Unlikely update. u2 >= d1
 	sbb	d1, u2
 	jmp	L(bck)
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/div_qr_2u_pi1.asm b/mpn/x86_64/div_qr_2u_pi1.asm
index 85af96f..b47209e 100644
--- a/mpn/x86_64/div_qr_2u_pi1.asm
+++ b/mpn/x86_64/div_qr_2u_pi1.asm
@@ -198,3 +198,4 @@ L(fix_qh):	C Unlikely update. u2 >= d1
 	sbb	d1, u2
 	jmp	L(bck_qh)
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/dive_1.asm b/mpn/x86_64/dive_1.asm
index 988bdab..b401112 100644
--- a/mpn/x86_64/dive_1.asm
+++ b/mpn/x86_64/dive_1.asm
@@ -156,3 +156,4 @@ L(one):	shr	R8(%rcx), %rax
 	ret
 
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/divrem_1.asm b/mpn/x86_64/divrem_1.asm
index d4d61ad..0417756 100644
--- a/mpn/x86_64/divrem_1.asm
+++ b/mpn/x86_64/divrem_1.asm
@@ -312,3 +312,4 @@ L(ret):	pop	%rbx
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/divrem_2.asm b/mpn/x86_64/divrem_2.asm
index 296c9b6..73aa740 100644
--- a/mpn/x86_64/divrem_2.asm
+++ b/mpn/x86_64/divrem_2.asm
@@ -188,3 +188,4 @@ L(fix):	seta	%dl
 	sbb	%r11, %rbx
 	jmp	L(bck)
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/fastavx/copyd.asm b/mpn/x86_64/fastavx/copyd.asm
index 56d472f..8d4f651 100644
--- a/mpn/x86_64/fastavx/copyd.asm
+++ b/mpn/x86_64/fastavx/copyd.asm
@@ -170,3 +170,4 @@ L(bc):	test	$4, R8(n)
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/fastavx/copyi.asm b/mpn/x86_64/fastavx/copyi.asm
index 7607747..3364aa9 100644
--- a/mpn/x86_64/fastavx/copyi.asm
+++ b/mpn/x86_64/fastavx/copyi.asm
@@ -167,3 +167,4 @@ L(bc):	test	$4, R8(n)
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/fastsse/com-palignr.asm b/mpn/x86_64/fastsse/com-palignr.asm
index c7155d1..191e5d9 100644
--- a/mpn/x86_64/fastsse/com-palignr.asm
+++ b/mpn/x86_64/fastsse/com-palignr.asm
@@ -308,3 +308,4 @@ L(end):	test	$1, R8(n)
 1:	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/fastsse/com.asm b/mpn/x86_64/fastsse/com.asm
index 307fb75..5dfc8e4 100644
--- a/mpn/x86_64/fastsse/com.asm
+++ b/mpn/x86_64/fastsse/com.asm
@@ -165,3 +165,4 @@ L(sma):	add	$14, n
 L(don):	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/fastsse/copyd-palignr.asm b/mpn/x86_64/fastsse/copyd-palignr.asm
index fac6f8a..a69812c 100644
--- a/mpn/x86_64/fastsse/copyd-palignr.asm
+++ b/mpn/x86_64/fastsse/copyd-palignr.asm
@@ -252,3 +252,4 @@ L(end):	test	$1, R8(n)
 1:	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/fastsse/copyd.asm b/mpn/x86_64/fastsse/copyd.asm
index 5b8b8bf..f03affa 100644
--- a/mpn/x86_64/fastsse/copyd.asm
+++ b/mpn/x86_64/fastsse/copyd.asm
@@ -156,3 +156,4 @@ L(sma):	test	$8, R8(n)
 L(don):	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/fastsse/copyi-palignr.asm b/mpn/x86_64/fastsse/copyi-palignr.asm
index 22f13f1..e50f604 100644
--- a/mpn/x86_64/fastsse/copyi-palignr.asm
+++ b/mpn/x86_64/fastsse/copyi-palignr.asm
@@ -296,3 +296,4 @@ L(end):	test	$1, R8(n)
 1:	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/fastsse/copyi.asm b/mpn/x86_64/fastsse/copyi.asm
index b2f3b9d..a506942 100644
--- a/mpn/x86_64/fastsse/copyi.asm
+++ b/mpn/x86_64/fastsse/copyi.asm
@@ -175,3 +175,4 @@ dnl	jnc	1b
 L(ret):	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/fastsse/lshift-movdqu2.asm b/mpn/x86_64/fastsse/lshift-movdqu2.asm
index a05e850..df8ee6d 100644
--- a/mpn/x86_64/fastsse/lshift-movdqu2.asm
+++ b/mpn/x86_64/fastsse/lshift-movdqu2.asm
@@ -180,3 +180,4 @@ L(end8):movq	(ap), %xmm0
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/fastsse/lshift.asm b/mpn/x86_64/fastsse/lshift.asm
index f76972a..7d0f0fc 100644
--- a/mpn/x86_64/fastsse/lshift.asm
+++ b/mpn/x86_64/fastsse/lshift.asm
@@ -167,3 +167,4 @@ L(end8):movq	(ap), %xmm0
 	movq	%xmm0, (rp)
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/fastsse/lshiftc-movdqu2.asm b/mpn/x86_64/fastsse/lshiftc-movdqu2.asm
index 8250910..4878dad 100644
--- a/mpn/x86_64/fastsse/lshiftc-movdqu2.asm
+++ b/mpn/x86_64/fastsse/lshiftc-movdqu2.asm
@@ -191,3 +191,4 @@ L(end8):movq	(ap), %xmm0
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/fastsse/lshiftc.asm b/mpn/x86_64/fastsse/lshiftc.asm
index d252069..f042ec0 100644
--- a/mpn/x86_64/fastsse/lshiftc.asm
+++ b/mpn/x86_64/fastsse/lshiftc.asm
@@ -177,3 +177,4 @@ L(end8):movq	(ap), %xmm0
 	movq	%xmm0, (rp)
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/fastsse/rshift-movdqu2.asm b/mpn/x86_64/fastsse/rshift-movdqu2.asm
index 1e270b1..8149717 100644
--- a/mpn/x86_64/fastsse/rshift-movdqu2.asm
+++ b/mpn/x86_64/fastsse/rshift-movdqu2.asm
@@ -199,3 +199,4 @@ L(bc):	dec	R32(n)
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/fastsse/sec_tabselect.asm b/mpn/x86_64/fastsse/sec_tabselect.asm
index e3df110..9975eca 100644
--- a/mpn/x86_64/fastsse/sec_tabselect.asm
+++ b/mpn/x86_64/fastsse/sec_tabselect.asm
@@ -190,3 +190,4 @@ L(tp1):	movdqa	%xmm8, %xmm0
 L(b000):FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/fat/fat_entry.asm b/mpn/x86_64/fat/fat_entry.asm
index 8f7599d..5f78553 100644
--- a/mpn/x86_64/fat/fat_entry.asm
+++ b/mpn/x86_64/fat/fat_entry.asm
@@ -205,3 +205,4 @@ PROLOGUE(__gmpn_cpuid)
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/gcd_1.asm b/mpn/x86_64/gcd_1.asm
index ac4aced..bf32cc0 100644
--- a/mpn/x86_64/gcd_1.asm
+++ b/mpn/x86_64/gcd_1.asm
@@ -163,3 +163,4 @@ L(shift_alot):
 	mov	%rax, %rcx
 	jmp	L(mid)
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/invert_limb.asm b/mpn/x86_64/invert_limb.asm
index cc79b89..829861f 100644
--- a/mpn/x86_64/invert_limb.asm
+++ b/mpn/x86_64/invert_limb.asm
@@ -113,3 +113,4 @@ ifdef(`DARWIN',`
 	ret
 EPILOGUE()
 ASM_END()
+CF_PROT
diff --git a/mpn/x86_64/invert_limb_table.asm b/mpn/x86_64/invert_limb_table.asm
index 739d59e..16fe314 100644
--- a/mpn/x86_64/invert_limb_table.asm
+++ b/mpn/x86_64/invert_limb_table.asm
@@ -48,3 +48,4 @@ forloop(i,256,512-1,dnl
 `	.value	eval(0x7fd00/i)
 ')dnl
 ASM_END()
+CF_PROT
diff --git a/mpn/x86_64/k10/hamdist.asm b/mpn/x86_64/k10/hamdist.asm
index 44b67b5..83e4e86 100644
--- a/mpn/x86_64/k10/hamdist.asm
+++ b/mpn/x86_64/k10/hamdist.asm
@@ -101,3 +101,4 @@ L(top):	mov	(ap,n,8), %r8
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/k10/popcount.asm b/mpn/x86_64/k10/popcount.asm
index 3814aea..17e7a73 100644
--- a/mpn/x86_64/k10/popcount.asm
+++ b/mpn/x86_64/k10/popcount.asm
@@ -136,3 +136,4 @@ C 1 = n mod 8
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/k8/aorrlsh_n.asm b/mpn/x86_64/k8/aorrlsh_n.asm
index ff3a184..8eff29e 100644
--- a/mpn/x86_64/k8/aorrlsh_n.asm
+++ b/mpn/x86_64/k8/aorrlsh_n.asm
@@ -215,3 +215,4 @@ L(cj1):	mov	%r9, 8(rp,n,8)
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/k8/div_qr_1n_pi1.asm b/mpn/x86_64/k8/div_qr_1n_pi1.asm
index 861402b..fef3a09 100644
--- a/mpn/x86_64/k8/div_qr_1n_pi1.asm
+++ b/mpn/x86_64/k8/div_qr_1n_pi1.asm
@@ -247,3 +247,4 @@ L(q_incr_loop):
 	lea	8(U1), U1
 	jmp	L(q_incr_loop)
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/k8/mul_basecase.asm b/mpn/x86_64/k8/mul_basecase.asm
index ca2efb9..61b6e0e 100644
--- a/mpn/x86_64/k8/mul_basecase.asm
+++ b/mpn/x86_64/k8/mul_basecase.asm
@@ -467,3 +467,4 @@ L(ret):	pop	%r15
 	ret
 
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/k8/mullo_basecase.asm b/mpn/x86_64/k8/mullo_basecase.asm
index fa00f42..b1f5b20 100644
--- a/mpn/x86_64/k8/mullo_basecase.asm
+++ b/mpn/x86_64/k8/mullo_basecase.asm
@@ -99,12 +99,14 @@ dnl	JMPENT(	L(2m4), L(tab))			C 10
 dnl	JMPENT(	L(3m4), L(tab))			C 11
 	TEXT
 
-L(1):	imul	%r8, %rax
+L(1):	CFPROT_ENDBR
+	imul	%r8, %rax
 	mov	%rax, (rp)
 	FUNC_EXIT()
 	ret
 
-L(2):	mov	8(vp_param), %r11
+L(2):	CFPROT_ENDBR
+	mov	8(vp_param), %r11
 	imul	%rax, %r11		C u0 x v1
 	mul	%r8			C u0 x v0
 	mov	%rax, (rp)
@@ -115,7 +117,8 @@ L(2):	mov	8(vp_param), %r11
 	FUNC_EXIT()
 	ret
 
-L(3):	mov	8(vp_param), %r9	C v1
+L(3):	CFPROT_ENDBR
+	mov	8(vp_param), %r9	C v1
 	mov	16(vp_param), %r11
 	mul	%r8			C u0 x v0 -> <r1,r0>
 	mov	%rax, (rp)		C r0
@@ -144,7 +147,8 @@ L(0m4):
 L(1m4):
 L(2m4):
 L(3m4):
-L(gen):	push	%rbx
+L(gen):	CFPROT_ENDBR
+	push	%rbx
 	push	%rbp
 	push	%r13
 	push	%r14
@@ -434,3 +438,4 @@ L(ret):	pop	%r15
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/k8/mulmid_basecase.asm b/mpn/x86_64/k8/mulmid_basecase.asm
index 86f1414..0ace1ba 100644
--- a/mpn/x86_64/k8/mulmid_basecase.asm
+++ b/mpn/x86_64/k8/mulmid_basecase.asm
@@ -557,3 +557,4 @@ L(ret):	pop	%r15
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/k8/redc_1.asm b/mpn/x86_64/k8/redc_1.asm
index 9327b21..b00103f 100644
--- a/mpn/x86_64/k8/redc_1.asm
+++ b/mpn/x86_64/k8/redc_1.asm
@@ -124,8 +124,9 @@ L(tab):	JMPENT(	L(0), L(tab))
 	JMPENT(	L(3m4), L(tab))
 	TEXT
 
+L(1):	CFPROT_ENDBR
 	ALIGN(16)
-L(1):	mov	(mp_param), %rax
+	mov	(mp_param), %rax
 	mul	q0
 	add	8(up), %rax
 	adc	16(up), %rdx
@@ -135,8 +136,9 @@ L(1):	mov	(mp_param), %rax
 	jmp	L(ret)
 
 
+L(2):	CFPROT_ENDBR
 	ALIGN(16)
-L(2):	mov	(mp_param), %rax
+  mov	(mp_param), %rax
 	mul	q0
 	xor	R32(%r14), R32(%r14)
 	mov	%rax, %r10
@@ -171,7 +173,8 @@ L(2):	mov	(mp_param), %rax
 	jmp	L(ret)
 
 
-L(3):	mov	(mp_param), %rax
+L(3):	CFPROT_ENDBR
+	mov	(mp_param), %rax
 	mul	q0
 	mov	%rax, %rbx
 	mov	%rdx, %r10
@@ -247,8 +250,8 @@ L(3):	mov	(mp_param), %rax
 	jmp	L(ret)
 
 
+L(2m4):	CFPROT_ENDBR
 	ALIGN(16)
-L(2m4):
 L(lo2):	mov	(mp,nneg,8), %rax
 	mul	q0
 	xor	R32(%r14), R32(%r14)
@@ -323,8 +326,8 @@ L(le2):	add	%r10, (up)
 	jmp	L(addx)
 
 
+L(1m4):	CFPROT_ENDBR
 	ALIGN(16)
-L(1m4):
 L(lo1):	mov	(mp,nneg,8), %rax
 	xor	%r9, %r9
 	xor	R32(%rbx), R32(%rbx)
@@ -396,9 +399,9 @@ L(le1):	add	%r10, (up)
 	jmp	L(addx)
 
 
-	ALIGN(16)
 L(0):
-L(0m4):
+L(0m4):	CFPROT_ENDBR
+	ALIGN(16)
 L(lo0):	mov	(mp,nneg,8), %rax
 	mov	nneg, i
 	mul	q0
@@ -462,8 +465,8 @@ L(le0):	add	%r10, (up)
 	jmp	L(addy)
 
 
+L(3m4):	CFPROT_ENDBR
 	ALIGN(16)
-L(3m4):
 L(lo3):	mov	(mp,nneg,8), %rax
 	mul	q0
 	mov	%rax, %rbx
@@ -589,3 +592,4 @@ L(ret):	pop	%r15
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/k8/sqr_basecase.asm b/mpn/x86_64/k8/sqr_basecase.asm
index 60cf945..e6a545d 100644
--- a/mpn/x86_64/k8/sqr_basecase.asm
+++ b/mpn/x86_64/k8/sqr_basecase.asm
@@ -131,7 +131,8 @@ L(tab):	JMPENT(	L(4), L(tab))
 	JMPENT(	L(3m4), L(tab))
 	TEXT
 
-L(1):	mov	(up), %rax
+L(1):	CFPROT_ENDBR
+	mov	(up), %rax
 	mul	%rax
 	add	$40, %rsp
 	mov	%rax, (rp)
@@ -139,7 +140,8 @@ L(1):	mov	(up), %rax
 	FUNC_EXIT()
 	ret
 
-L(2):	mov	(up), %rax
+L(2):	CFPROT_ENDBR
+	mov	(up), %rax
 	mov	%rax, %r8
 	mul	%rax
 	mov	8(up), %r11
@@ -165,7 +167,8 @@ L(2):	mov	(up), %rax
 	FUNC_EXIT()
 	ret
 
-L(3):	mov	(up), %rax
+L(3):	CFPROT_ENDBR
+	mov	(up), %rax
 	mov	%rax, %r10
 	mul	%rax
 	mov	8(up), %r11
@@ -210,7 +213,8 @@ L(3):	mov	(up), %rax
 	FUNC_EXIT()
 	ret
 
-L(4):	mov	(up), %rax
+L(4):	CFPROT_ENDBR
+	mov	(up), %rax
 	mov	%rax, %r11
 	mul	%rax
 	mov	8(up), %rbx
@@ -281,7 +285,7 @@ L(4):	mov	(up), %rax
 	ret
 
 
-L(0m4):
+L(0m4):	CFPROT_ENDBR
 	lea	-16(rp,n,8), tp		C point tp in middle of result operand
 	mov	(up), v0
 	mov	8(up), %rax
@@ -339,7 +343,7 @@ L(L3):	xor	R32(w1), R32(w1)
 	jmp	L(dowhile)
 
 
-L(1m4):
+L(1m4):	CFPROT_ENDBR
 	lea	8(rp,n,8), tp		C point tp in middle of result operand
 	mov	(up), v0		C u0
 	mov	8(up), %rax		C u1
@@ -417,7 +421,7 @@ L(m2x):	mov	(up,j,8), %rax
 	jmp	L(dowhile_end)
 
 
-L(2m4):
+L(2m4):	CFPROT_ENDBR
 	lea	-16(rp,n,8), tp		C point tp in middle of result operand
 	mov	(up), v0
 	mov	8(up), %rax
@@ -474,7 +478,7 @@ L(L1):	xor	R32(w0), R32(w0)
 	jmp	L(dowhile_mid)
 
 
-L(3m4):
+L(3m4):	CFPROT_ENDBR
 	lea	8(rp,n,8), tp		C point tp in middle of result operand
 	mov	(up), v0		C u0
 	mov	8(up), %rax		C u1
@@ -805,3 +809,4 @@ L(d1):	mov	%r11, 24(rp,j,8)
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/logops_n.asm b/mpn/x86_64/logops_n.asm
index b277f58..b2c640c 100644
--- a/mpn/x86_64/logops_n.asm
+++ b/mpn/x86_64/logops_n.asm
@@ -134,6 +134,7 @@ L(e10):	movq	24(vp,n,8), %r9
 L(ret):	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
 ')
 
 ifdef(`VARIANT_2',`
@@ -187,6 +188,7 @@ L(e10):	movq	24(vp,n,8), %r9
 L(ret):	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
 ')
 
 ifdef(`VARIANT_3',`
@@ -241,4 +243,5 @@ L(e10):	movq	24(vp,n,8), %r9
 L(ret):	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
 ')
diff --git a/mpn/x86_64/lshift.asm b/mpn/x86_64/lshift.asm
index f368944..990b3b8 100644
--- a/mpn/x86_64/lshift.asm
+++ b/mpn/x86_64/lshift.asm
@@ -245,3 +245,4 @@ L(ast):	mov	(up), %r10
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/lshiftc.asm b/mpn/x86_64/lshiftc.asm
index c4ba04a..4fd4430 100644
--- a/mpn/x86_64/lshiftc.asm
+++ b/mpn/x86_64/lshiftc.asm
@@ -180,3 +180,4 @@ L(ast):	mov	(up), %r10
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/lshsub_n.asm b/mpn/x86_64/lshsub_n.asm
index 4d428c0..d263565 100644
--- a/mpn/x86_64/lshsub_n.asm
+++ b/mpn/x86_64/lshsub_n.asm
@@ -170,3 +170,4 @@ L(end):
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/missing.asm b/mpn/x86_64/missing.asm
index 9b65c89..7914b82 100644
--- a/mpn/x86_64/missing.asm
+++ b/mpn/x86_64/missing.asm
@@ -128,3 +128,4 @@ PROLOGUE(__gmp_adcx)
 	ret
 EPILOGUE()
 PROTECT(__gmp_adcx)
+CF_PROT
diff --git a/mpn/x86_64/mod_1_1.asm b/mpn/x86_64/mod_1_1.asm
index 09b5dd1..287f61d 100644
--- a/mpn/x86_64/mod_1_1.asm
+++ b/mpn/x86_64/mod_1_1.asm
@@ -234,3 +234,4 @@ L(z):
 	ret
 EPILOGUE()
 ASM_END()
+CF_PROT
diff --git a/mpn/x86_64/mod_1_2.asm b/mpn/x86_64/mod_1_2.asm
index 09d856e..1cd6dd1 100644
--- a/mpn/x86_64/mod_1_2.asm
+++ b/mpn/x86_64/mod_1_2.asm
@@ -237,3 +237,4 @@ ifdef(`SHLD_SLOW',`
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/mod_1_4.asm b/mpn/x86_64/mod_1_4.asm
index ae34617..fb685ef 100644
--- a/mpn/x86_64/mod_1_4.asm
+++ b/mpn/x86_64/mod_1_4.asm
@@ -268,3 +268,4 @@ ifdef(`SHLD_SLOW',`
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/mod_34lsub1.asm b/mpn/x86_64/mod_34lsub1.asm
index 62bdcfa..2cf5751 100644
--- a/mpn/x86_64/mod_34lsub1.asm
+++ b/mpn/x86_64/mod_34lsub1.asm
@@ -135,46 +135,55 @@ L(tab):	JMPENT(	L(0), L(tab))
 	JMPENT(	L(8), L(tab))
 	TEXT
 
-L(6):	add	(ap), %rax
+L(6):	CFPROT_ENDBR
+	add	(ap), %rax
 	adc	8(ap), %rcx
 	adc	16(ap), %rdx
 	adc	$0, %r9
 	add	$24, ap
-L(3):	add	(ap), %rax
+L(3):	CFPROT_ENDBR
+	add	(ap), %rax
 	adc	8(ap), %rcx
 	adc	16(ap), %rdx
 	jmp	L(cj1)
 
-L(7):	add	(ap), %rax
+L(7):	CFPROT_ENDBR
+	add	(ap), %rax
 	adc	8(ap), %rcx
 	adc	16(ap), %rdx
 	adc	$0, %r9
 	add	$24, ap
-L(4):	add	(ap), %rax
+L(4):	CFPROT_ENDBR
+	add	(ap), %rax
 	adc	8(ap), %rcx
 	adc	16(ap), %rdx
 	adc	$0, %r9
 	add	$24, ap
-L(1):	add	(ap), %rax
+L(1):	CFPROT_ENDBR
+	add	(ap), %rax
 	adc	$0, %rcx
 	jmp	L(cj2)
 
-L(8):	add	(ap), %rax
+L(8):	CFPROT_ENDBR
+	add	(ap), %rax
 	adc	8(ap), %rcx
 	adc	16(ap), %rdx
 	adc	$0, %r9
 	add	$24, ap
-L(5):	add	(ap), %rax
+L(5):	CFPROT_ENDBR
+	add	(ap), %rax
 	adc	8(ap), %rcx
 	adc	16(ap), %rdx
 	adc	$0, %r9
 	add	$24, ap
-L(2):	add	(ap), %rax
+L(2):	CFPROT_ENDBR
+	add	(ap), %rax
 	adc	8(ap), %rcx
 
 L(cj2):	adc	$0, %rdx
 L(cj1):	adc	$0, %r9
-L(0):	add	%r9, %rax
+L(0):	CFPROT_ENDBR
+	add	%r9, %rax
 	adc	$0, %rcx
 	adc	$0, %rdx
 	adc	$0, %rax
@@ -203,3 +212,4 @@ L(0):	add	%r9, %rax
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/mode1o.asm b/mpn/x86_64/mode1o.asm
index 2cd2b08..c10a5a6 100644
--- a/mpn/x86_64/mode1o.asm
+++ b/mpn/x86_64/mode1o.asm
@@ -169,3 +169,4 @@ L(one):
 
 EPILOGUE(mpn_modexact_1c_odd)
 EPILOGUE(mpn_modexact_1_odd)
+CF_PROT
diff --git a/mpn/x86_64/mul_1.asm b/mpn/x86_64/mul_1.asm
index b032afc..6ea9a4a 100644
--- a/mpn/x86_64/mul_1.asm
+++ b/mpn/x86_64/mul_1.asm
@@ -181,3 +181,4 @@ IFDOS(``pop	%rdi		'')
 IFDOS(``pop	%rsi		'')
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/mul_2.asm b/mpn/x86_64/mul_2.asm
index f408c52..6b73737 100644
--- a/mpn/x86_64/mul_2.asm
+++ b/mpn/x86_64/mul_2.asm
@@ -190,3 +190,4 @@ L(m22):	mul	v1
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/mulx/aorsmul_1.asm b/mpn/x86_64/mulx/aorsmul_1.asm
index 285c073..942cf6a 100644
--- a/mpn/x86_64/mulx/aorsmul_1.asm
+++ b/mpn/x86_64/mulx/aorsmul_1.asm
@@ -159,3 +159,4 @@ L(wd1):	ADCSBB	%rbx, 24(rp)
 	ret
 EPILOGUE()
 ASM_END()
+CF_PROT
diff --git a/mpn/x86_64/mulx/mul_1.asm b/mpn/x86_64/mulx/mul_1.asm
index 34a044d..4a0e6ef 100644
--- a/mpn/x86_64/mulx/mul_1.asm
+++ b/mpn/x86_64/mulx/mul_1.asm
@@ -152,3 +152,4 @@ L(wd1):	adc	%r12, %rbx
 	ret
 EPILOGUE()
 ASM_END()
+CF_PROT
diff --git a/mpn/x86_64/nano/dive_1.asm b/mpn/x86_64/nano/dive_1.asm
index e9a0763..d57c444 100644
--- a/mpn/x86_64/nano/dive_1.asm
+++ b/mpn/x86_64/nano/dive_1.asm
@@ -164,3 +164,4 @@ L(one):	shr	R8(%rcx), %rax
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/pentium4/aors_n.asm b/mpn/x86_64/pentium4/aors_n.asm
index 8e6ee1b..d3daf6f 100644
--- a/mpn/x86_64/pentium4/aors_n.asm
+++ b/mpn/x86_64/pentium4/aors_n.asm
@@ -194,3 +194,4 @@ L(ret):	mov	R32(%rbx), R32(%rax)
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/pentium4/aorslshC_n.asm b/mpn/x86_64/pentium4/aorslshC_n.asm
index d03c6a3..a4cd689 100644
--- a/mpn/x86_64/pentium4/aorslshC_n.asm
+++ b/mpn/x86_64/pentium4/aorslshC_n.asm
@@ -201,3 +201,4 @@ L(c3):	mov	$1, R8(%rax)
 	jmp	L(rc3)
 EPILOGUE()
 ASM_END()
+CF_PROT
diff --git a/mpn/x86_64/pentium4/lshift.asm b/mpn/x86_64/pentium4/lshift.asm
index d3b5213..baa4820 100644
--- a/mpn/x86_64/pentium4/lshift.asm
+++ b/mpn/x86_64/pentium4/lshift.asm
@@ -164,3 +164,4 @@ L(ast):	movq	(up), %mm2
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/pentium4/lshiftc.asm b/mpn/x86_64/pentium4/lshiftc.asm
index fc64676..e7ed07f 100644
--- a/mpn/x86_64/pentium4/lshiftc.asm
+++ b/mpn/x86_64/pentium4/lshiftc.asm
@@ -177,3 +177,4 @@ L(ast):	movq	(up), %mm2
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/pentium4/mod_34lsub1.asm b/mpn/x86_64/pentium4/mod_34lsub1.asm
index f34b3f0..adb4ae6 100644
--- a/mpn/x86_64/pentium4/mod_34lsub1.asm
+++ b/mpn/x86_64/pentium4/mod_34lsub1.asm
@@ -165,3 +165,4 @@ L(combine):
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/pentium4/rsh1aors_n.asm b/mpn/x86_64/pentium4/rsh1aors_n.asm
index 5528ce4..64a6322 100644
--- a/mpn/x86_64/pentium4/rsh1aors_n.asm
+++ b/mpn/x86_64/pentium4/rsh1aors_n.asm
@@ -332,3 +332,4 @@ L(cj1):	or	%r14, %rbx
 L(c3):	mov	$1, R8(%rax)
 	jmp	L(rc3)
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/pentium4/rshift.asm b/mpn/x86_64/pentium4/rshift.asm
index b7c1ee2..758ca64 100644
--- a/mpn/x86_64/pentium4/rshift.asm
+++ b/mpn/x86_64/pentium4/rshift.asm
@@ -167,3 +167,4 @@ L(ast):	movq	(up), %mm2
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/popham.asm b/mpn/x86_64/popham.asm
index 9005f81..a52ea0f 100644
--- a/mpn/x86_64/popham.asm
+++ b/mpn/x86_64/popham.asm
@@ -175,3 +175,4 @@ L(end):
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/rsh1aors_n.asm b/mpn/x86_64/rsh1aors_n.asm
index a3e9cc5..d28cc32 100644
--- a/mpn/x86_64/rsh1aors_n.asm
+++ b/mpn/x86_64/rsh1aors_n.asm
@@ -187,3 +187,4 @@ L(end):	mov	%rbx, (rp)
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/rshift.asm b/mpn/x86_64/rshift.asm
index 3f344f1..2c45172 100644
--- a/mpn/x86_64/rshift.asm
+++ b/mpn/x86_64/rshift.asm
@@ -174,3 +174,4 @@ L(ast):	mov	(up), %r10
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/sec_tabselect.asm b/mpn/x86_64/sec_tabselect.asm
index e8aed26..2198b4b 100644
--- a/mpn/x86_64/sec_tabselect.asm
+++ b/mpn/x86_64/sec_tabselect.asm
@@ -174,3 +174,4 @@ L(b00):	pop	%r15
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/sqr_diag_addlsh1.asm b/mpn/x86_64/sqr_diag_addlsh1.asm
index 4ad034c..6db16f6 100644
--- a/mpn/x86_64/sqr_diag_addlsh1.asm
+++ b/mpn/x86_64/sqr_diag_addlsh1.asm
@@ -114,3 +114,4 @@ L(end):	add	%r10, %r8
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/sublsh1_n.asm b/mpn/x86_64/sublsh1_n.asm
index c6d829f..2f0fe01 100644
--- a/mpn/x86_64/sublsh1_n.asm
+++ b/mpn/x86_64/sublsh1_n.asm
@@ -158,3 +158,4 @@ L(end):	add	R32(%rbp), R32(%rax)
 	FUNC_EXIT()
 	ret
 EPILOGUE()
+CF_PROT
diff --git a/mpn/x86_64/x86_64-defs.m4 b/mpn/x86_64/x86_64-defs.m4
index a626419..80f549e 100644
--- a/mpn/x86_64/x86_64-defs.m4
+++ b/mpn/x86_64/x86_64-defs.m4
@@ -93,8 +93,38 @@ m4_assert_numargs(1)
 `	GLOBL	$1
 	TYPE($1,`function')
 $1:
+	CFPROT_ENDBR
 ')
 
+dnl Generates the endbr64 instructions
+dnl Using macro, so it can be easily extended to use some arch specific conditional defines
+define(`CFPROT_ENDBR',
+``
+	endbr64''
+)
+
+dnl Append the .gnu-property to the end of files
+dnl This is needed for a -fcf-protection 
+dnl Again, using macro for easy arch specific defines
+dnl
+define(`CF_PROT',``
+	.section	.note.gnu.property,"a"
+	.align 8
+	.long	 1f - 0f
+	.long	 4f - 1f
+	.long	 5
+0:
+	.string	 "GNU"
+1:
+	.align 8
+	.long	 0xc0000002
+	.long	 3f - 2f
+2:
+	.long	 0x3
+3:
+ 	.align 8
+4:
+'')
 
 dnl  Usage: ASSERT([cond][,instructions])
 dnl