diff --git a/apple-x86/crypto/fipsmodule/sha256-586-apple.S b/apple-x86/crypto/fipsmodule/sha256-586-apple.S
index d43510a49..53421104c 100644
--- a/apple-x86/crypto/fipsmodule/sha256-586-apple.S
+++ b/apple-x86/crypto/fipsmodule/sha256-586-apple.S
@@ -33,26 +33,24 @@ L000pic_point:
 	movl	L_OPENSSL_ia32cap_P$non_lazy_ptr-L001K256(%ebp),%edx
 	movl	(%edx),%ecx
 	movl	4(%edx),%ebx
-	testl	$1048576,%ecx
-	jnz	L002loop
 	movl	8(%edx),%edx
 	testl	$16777216,%ecx
-	jz	L003no_xmm
+	jz	L002no_xmm
 	andl	$1073741824,%ecx
 	andl	$268435968,%ebx
 	orl	%ebx,%ecx
 	andl	$1342177280,%ecx
 	cmpl	$1342177280,%ecx
-	je	L004AVX
+	je	L003AVX
 	testl	$512,%ebx
-	jnz	L005SSSE3
-L003no_xmm:
+	jnz	L004SSSE3
+L002no_xmm:
 	subl	%edi,%eax
 	cmpl	$256,%eax
-	jae	L006unrolled
-	jmp	L002loop
+	jae	L005unrolled
+	jmp	L006loop
 .align	4,0x90
-L002loop:
+L006loop:
 	movl	(%edi),%eax
 	movl	4(%edi),%ebx
 	movl	8(%edi),%ecx
@@ -245,7 +243,7 @@ L00816_63:
 	leal	356(%esp),%esp
 	subl	$256,%ebp
 	cmpl	8(%esp),%edi
-	jb	L002loop
+	jb	L006loop
 	movl	12(%esp),%esp
 	popl	%edi
 	popl	%esi
@@ -262,7 +260,7 @@ L001K256:
 .byte	112,112,114,111,64,111,112,101,110,115,115,108,46,111,114,103
 .byte	62,0
 .align	4,0x90
-L006unrolled:
+L005unrolled:
 	leal	-96(%esp),%esp
 	movl	(%esi),%eax
 	movl	4(%esi),%ebp
@@ -3169,7 +3167,7 @@ L009grand_loop:
 	popl	%ebp
 	ret
 .align	5,0x90
-L005SSSE3:
+L004SSSE3:
 	leal	-96(%esp),%esp
 	movl	(%esi),%eax
 	movl	4(%esi),%ebx
@@ -4380,7 +4378,7 @@ L011ssse3_00_47:
 	popl	%ebp
 	ret
 .align	5,0x90
-L004AVX:
+L003AVX:
 	leal	-96(%esp),%esp
 	vzeroall
 	movl	(%esi),%eax
diff --git a/linux-x86/crypto/fipsmodule/sha256-586-linux.S b/linux-x86/crypto/fipsmodule/sha256-586-linux.S
index ee41b78cb..7da9d1eb4 100644
--- a/linux-x86/crypto/fipsmodule/sha256-586-linux.S
+++ b/linux-x86/crypto/fipsmodule/sha256-586-linux.S
@@ -34,26 +34,24 @@ sha256_block_data_order:
 	leal	OPENSSL_ia32cap_P-.L001K256(%ebp),%edx
 	movl	(%edx),%ecx
 	movl	4(%edx),%ebx
-	testl	$1048576,%ecx
-	jnz	.L002loop
 	movl	8(%edx),%edx
 	testl	$16777216,%ecx
-	jz	.L003no_xmm
+	jz	.L002no_xmm
 	andl	$1073741824,%ecx
 	andl	$268435968,%ebx
 	orl	%ebx,%ecx
 	andl	$1342177280,%ecx
 	cmpl	$1342177280,%ecx
-	je	.L004AVX
+	je	.L003AVX
 	testl	$512,%ebx
-	jnz	.L005SSSE3
-.L003no_xmm:
+	jnz	.L004SSSE3
+.L002no_xmm:
 	subl	%edi,%eax
 	cmpl	$256,%eax
-	jae	.L006unrolled
-	jmp	.L002loop
+	jae	.L005unrolled
+	jmp	.L006loop
 .align	16
-.L002loop:
+.L006loop:
 	movl	(%edi),%eax
 	movl	4(%edi),%ebx
 	movl	8(%edi),%ecx
@@ -246,7 +244,7 @@ sha256_block_data_order:
 	leal	356(%esp),%esp
 	subl	$256,%ebp
 	cmpl	8(%esp),%edi
-	jb	.L002loop
+	jb	.L006loop
 	movl	12(%esp),%esp
 	popl	%edi
 	popl	%esi
@@ -263,7 +261,7 @@ sha256_block_data_order:
 .byte	112,112,114,111,64,111,112,101,110,115,115,108,46,111,114,103
 .byte	62,0
 .align	16
-.L006unrolled:
+.L005unrolled:
 	leal	-96(%esp),%esp
 	movl	(%esi),%eax
 	movl	4(%esi),%ebp
@@ -3170,7 +3168,7 @@ sha256_block_data_order:
 	popl	%ebp
 	ret
 .align	32
-.L005SSSE3:
+.L004SSSE3:
 	leal	-96(%esp),%esp
 	movl	(%esi),%eax
 	movl	4(%esi),%ebx
@@ -4381,7 +4379,7 @@ sha256_block_data_order:
 	popl	%ebp
 	ret
 .align	32
-.L004AVX:
+.L003AVX:
 	leal	-96(%esp),%esp
 	vzeroall
 	movl	(%esi),%eax
diff --git a/src/crypto/fipsmodule/sha/asm/sha256-586.pl b/src/crypto/fipsmodule/sha/asm/sha256-586.pl
index d23d81fbf..ab821e745 100644
--- a/src/crypto/fipsmodule/sha/asm/sha256-586.pl
+++ b/src/crypto/fipsmodule/sha/asm/sha256-586.pl
@@ -211,8 +211,6 @@ sub BODY_00_15() {
 	&picmeup("edx","OPENSSL_ia32cap_P",$K256,&label("K256"));
 	&mov	("ecx",&DWP(0,"edx"));
 	&mov	("ebx",&DWP(4,"edx"));
-	&test	("ecx",1<<20);		# check for P4
-	&jnz	(&label("loop"));
 	&mov	("edx",&DWP(8,"edx"))	if ($xmm);
 	&test	("ecx",1<<24);		# check for FXSR
 	&jz	($unroll_after?&label("no_xmm"):&label("loop"));
diff --git a/win-x86/crypto/fipsmodule/sha256-586-win.asm b/win-x86/crypto/fipsmodule/sha256-586-win.asm
index 4e0278b0a..65f3b549a 100644
--- a/win-x86/crypto/fipsmodule/sha256-586-win.asm
+++ b/win-x86/crypto/fipsmodule/sha256-586-win.asm
@@ -41,26 +41,24 @@ L$000pic_point:
 	lea	edx,[_OPENSSL_ia32cap_P]
 	mov	ecx,DWORD [edx]
 	mov	ebx,DWORD [4+edx]
-	test	ecx,1048576
-	jnz	NEAR L$002loop
 	mov	edx,DWORD [8+edx]
 	test	ecx,16777216
-	jz	NEAR L$003no_xmm
+	jz	NEAR L$002no_xmm
 	and	ecx,1073741824
 	and	ebx,268435968
 	or	ecx,ebx
 	and	ecx,1342177280
 	cmp	ecx,1342177280
-	je	NEAR L$004AVX
+	je	NEAR L$003AVX
 	test	ebx,512
-	jnz	NEAR L$005SSSE3
-L$003no_xmm:
+	jnz	NEAR L$004SSSE3
+L$002no_xmm:
 	sub	eax,edi
 	cmp	eax,256
-	jae	NEAR L$006unrolled
-	jmp	NEAR L$002loop
+	jae	NEAR L$005unrolled
+	jmp	NEAR L$006loop
 align	16
-L$002loop:
+L$006loop:
 	mov	eax,DWORD [edi]
 	mov	ebx,DWORD [4+edi]
 	mov	ecx,DWORD [8+edi]
@@ -253,7 +251,7 @@ L$00816_63:
 	lea	esp,[356+esp]
 	sub	ebp,256
 	cmp	edi,DWORD [8+esp]
-	jb	NEAR L$002loop
+	jb	NEAR L$006loop
 	mov	esp,DWORD [12+esp]
 	pop	edi
 	pop	esi
@@ -270,7 +268,7 @@ db	67,82,89,80,84,79,71,65,77,83,32,98,121,32,60,97
 db	112,112,114,111,64,111,112,101,110,115,115,108,46,111,114,103
 db	62,0
 align	16
-L$006unrolled:
+L$005unrolled:
 	lea	esp,[esp-96]
 	mov	eax,DWORD [esi]
 	mov	ebp,DWORD [4+esi]
@@ -3177,7 +3175,7 @@ L$009grand_loop:
 	pop	ebp
 	ret
 align	32
-L$005SSSE3:
+L$004SSSE3:
 	lea	esp,[esp-96]
 	mov	eax,DWORD [esi]
 	mov	ebx,DWORD [4+esi]
@@ -4388,7 +4386,7 @@ db	102,15,58,15,249,4
 	pop	ebp
 	ret
 align	32
-L$004AVX:
+L$003AVX:
 	lea	esp,[esp-96]
 	vzeroall
 	mov	eax,DWORD [esi]