external/boringssl: bump revision.

This change bumps the BoringSSL revision to the current tip-of-tree.

Change-Id: I91d5bf467e16e8d86cb19a4de873985f524e5faa

2 files changed, 445 insertions, 172 deletions

diff --git a/mac-x86/crypto/bn/bn-586.S b/mac-x86/crypto/bn/bn-586.S
index 34cf56f..0f0a94e 100644
--- a/mac-x86/crypto/bn/bn-586.S
+++ b/mac-x86/crypto/bn/bn-586.S
@@ -6,6 +6,102 @@
 .align	4
 _bn_mul_add_words:
 L_bn_mul_add_words_begin:
+	call	L000PIC_me_up
+L000PIC_me_up:
+	popl	%eax
+	movl	L_OPENSSL_ia32cap_P$non_lazy_ptr-L000PIC_me_up(%eax),%eax
+	btl	$26,(%eax)
+	jnc	L001maw_non_sse2
+	movl	4(%esp),%eax
+	movl	8(%esp),%edx
+	movl	12(%esp),%ecx
+	movd	16(%esp),%mm0
+	pxor	%mm1,%mm1
+	jmp	L002maw_sse2_entry
+.align	4,0x90
+L003maw_sse2_unrolled:
+	movd	(%eax),%mm3
+	paddq	%mm3,%mm1
+	movd	(%edx),%mm2
+	pmuludq	%mm0,%mm2
+	movd	4(%edx),%mm4
+	pmuludq	%mm0,%mm4
+	movd	8(%edx),%mm6
+	pmuludq	%mm0,%mm6
+	movd	12(%edx),%mm7
+	pmuludq	%mm0,%mm7
+	paddq	%mm2,%mm1
+	movd	4(%eax),%mm3
+	paddq	%mm4,%mm3
+	movd	8(%eax),%mm5
+	paddq	%mm6,%mm5
+	movd	12(%eax),%mm4
+	paddq	%mm4,%mm7
+	movd	%mm1,(%eax)
+	movd	16(%edx),%mm2
+	pmuludq	%mm0,%mm2
+	psrlq	$32,%mm1
+	movd	20(%edx),%mm4
+	pmuludq	%mm0,%mm4
+	paddq	%mm3,%mm1
+	movd	24(%edx),%mm6
+	pmuludq	%mm0,%mm6
+	movd	%mm1,4(%eax)
+	psrlq	$32,%mm1
+	movd	28(%edx),%mm3
+	addl	$32,%edx
+	pmuludq	%mm0,%mm3
+	paddq	%mm5,%mm1
+	movd	16(%eax),%mm5
+	paddq	%mm5,%mm2
+	movd	%mm1,8(%eax)
+	psrlq	$32,%mm1
+	paddq	%mm7,%mm1
+	movd	20(%eax),%mm5
+	paddq	%mm5,%mm4
+	movd	%mm1,12(%eax)
+	psrlq	$32,%mm1
+	paddq	%mm2,%mm1
+	movd	24(%eax),%mm5
+	paddq	%mm5,%mm6
+	movd	%mm1,16(%eax)
+	psrlq	$32,%mm1
+	paddq	%mm4,%mm1
+	movd	28(%eax),%mm5
+	paddq	%mm5,%mm3
+	movd	%mm1,20(%eax)
+	psrlq	$32,%mm1
+	paddq	%mm6,%mm1
+	movd	%mm1,24(%eax)
+	psrlq	$32,%mm1
+	paddq	%mm3,%mm1
+	movd	%mm1,28(%eax)
+	leal	32(%eax),%eax
+	psrlq	$32,%mm1
+	subl	$8,%ecx
+	jz	L004maw_sse2_exit
+L002maw_sse2_entry:
+	testl	$4294967288,%ecx
+	jnz	L003maw_sse2_unrolled
+.align	2,0x90
+L005maw_sse2_loop:
+	movd	(%edx),%mm2
+	movd	(%eax),%mm3
+	pmuludq	%mm0,%mm2
+	leal	4(%edx),%edx
+	paddq	%mm3,%mm1
+	paddq	%mm2,%mm1
+	movd	%mm1,(%eax)
+	subl	$1,%ecx
+	psrlq	$32,%mm1
+	leal	4(%eax),%eax
+	jnz	L005maw_sse2_loop
+L004maw_sse2_exit:
+	movd	%mm1,%eax
+	emms
+	ret
+.align	4,0x90
+L001maw_non_sse2:
 	pushl	%ebp
 	pushl	%ebx
 	pushl	%esi
@@ -18,9 +114,9 @@ L_bn_mul_add_words_begin:
 	andl	$4294967288,%ecx
 	movl	32(%esp),%ebp
 	pushl	%ecx
-	jz	L000maw_finish
+	jz	L006maw_finish
 .align	4,0x90
-L001maw_loop:
+L007maw_loop:
 	# Round 0 
 	movl	(%ebx),%eax
 	mull	%ebp
@@ -97,13 +193,13 @@ L001maw_loop:
 	subl	$8,%ecx
 	leal	32(%ebx),%ebx
 	leal	32(%edi),%edi
-	jnz	L001maw_loop
-L000maw_finish:
+	jnz	L007maw_loop
+L006maw_finish:
 	movl	32(%esp),%ecx
 	andl	$7,%ecx
-	jnz	L002maw_finish2
-	jmp	L003maw_end
-L002maw_finish2:
+	jnz	L008maw_finish2
+	jmp	L009maw_end
+L008maw_finish2:
 	# Tail Round 0 
 	movl	(%ebx),%eax
 	mull	%ebp
@@ -114,7 +210,7 @@ L002maw_finish2:
 	decl	%ecx
 	movl	%eax,(%edi)
 	movl	%edx,%esi
-	jz	L003maw_end
+	jz	L009maw_end
 	# Tail Round 1 
 	movl	4(%ebx),%eax
 	mull	%ebp
@@ -125,7 +221,7 @@ L002maw_finish2:
 	decl	%ecx
 	movl	%eax,4(%edi)
 	movl	%edx,%esi
-	jz	L003maw_end
+	jz	L009maw_end
 	# Tail Round 2 
 	movl	8(%ebx),%eax
 	mull	%ebp
@@ -136,7 +232,7 @@ L002maw_finish2:
 	decl	%ecx
 	movl	%eax,8(%edi)
 	movl	%edx,%esi
-	jz	L003maw_end
+	jz	L009maw_end
 	# Tail Round 3 
 	movl	12(%ebx),%eax
 	mull	%ebp
@@ -147,7 +243,7 @@ L002maw_finish2:
 	decl	%ecx
 	movl	%eax,12(%edi)
 	movl	%edx,%esi
-	jz	L003maw_end
+	jz	L009maw_end
 	# Tail Round 4 
 	movl	16(%ebx),%eax
 	mull	%ebp
@@ -158,7 +254,7 @@ L002maw_finish2:
 	decl	%ecx
 	movl	%eax,16(%edi)
 	movl	%edx,%esi
-	jz	L003maw_end
+	jz	L009maw_end
 	# Tail Round 5 
 	movl	20(%ebx),%eax
 	mull	%ebp
@@ -169,7 +265,7 @@ L002maw_finish2:
 	decl	%ecx
 	movl	%eax,20(%edi)
 	movl	%edx,%esi
-	jz	L003maw_end
+	jz	L009maw_end
 	# Tail Round 6 
 	movl	24(%ebx),%eax
 	mull	%ebp
@@ -179,7 +275,7 @@ L002maw_finish2:
 	adcl	$0,%edx
 	movl	%eax,24(%edi)
 	movl	%edx,%esi
-L003maw_end:
+L009maw_end:
 	movl	%esi,%eax
 	popl	%ecx
 	popl	%edi
@@ -192,6 +288,33 @@ L003maw_end:
 .align	4
 _bn_mul_words:
 L_bn_mul_words_begin:
+	call	L010PIC_me_up
+L010PIC_me_up:
+	popl	%eax
+	movl	L_OPENSSL_ia32cap_P$non_lazy_ptr-L010PIC_me_up(%eax),%eax
+	btl	$26,(%eax)
+	jnc	L011mw_non_sse2
+	movl	4(%esp),%eax
+	movl	8(%esp),%edx
+	movl	12(%esp),%ecx
+	movd	16(%esp),%mm0
+	pxor	%mm1,%mm1
+.align	4,0x90
+L012mw_sse2_loop:
+	movd	(%edx),%mm2
+	pmuludq	%mm0,%mm2
+	leal	4(%edx),%edx
+	paddq	%mm2,%mm1
+	movd	%mm1,(%eax)
+	subl	$1,%ecx
+	psrlq	$32,%mm1
+	leal	4(%eax),%eax
+	jnz	L012mw_sse2_loop
+	movd	%mm1,%eax
+	emms
+	ret
+.align	4,0x90
+L011mw_non_sse2:
 	pushl	%ebp
 	pushl	%ebx
 	pushl	%esi
@@ -203,8 +326,8 @@ L_bn_mul_words_begin:
 	movl	28(%esp),%ebp
 	movl	32(%esp),%ecx
 	andl	$4294967288,%ebp
-	jz	L004mw_finish
-L005mw_loop:
+	jz	L013mw_finish
+L014mw_loop:
 	# Round 0 
 	movl	(%ebx),%eax
 	mull	%ecx
@@ -265,14 +388,14 @@ L005mw_loop:
 	addl	$32,%ebx
 	addl	$32,%edi
 	subl	$8,%ebp
-	jz	L004mw_finish
-	jmp	L005mw_loop
-L004mw_finish:
+	jz	L013mw_finish
+	jmp	L014mw_loop
+L013mw_finish:
 	movl	28(%esp),%ebp
 	andl	$7,%ebp
-	jnz	L006mw_finish2
-	jmp	L007mw_end
-L006mw_finish2:
+	jnz	L015mw_finish2
+	jmp	L016mw_end
+L015mw_finish2:
 	# Tail Round 0 
 	movl	(%ebx),%eax
 	mull	%ecx
@@ -281,7 +404,7 @@ L006mw_finish2:
 	movl	%eax,(%edi)
 	movl	%edx,%esi
 	decl	%ebp
-	jz	L007mw_end
+	jz	L016mw_end
 	# Tail Round 1 
 	movl	4(%ebx),%eax
 	mull	%ecx
@@ -290,7 +413,7 @@ L006mw_finish2:
 	movl	%eax,4(%edi)
 	movl	%edx,%esi
 	decl	%ebp
-	jz	L007mw_end
+	jz	L016mw_end
 	# Tail Round 2 
 	movl	8(%ebx),%eax
 	mull	%ecx
@@ -299,7 +422,7 @@ L006mw_finish2:
 	movl	%eax,8(%edi)
 	movl	%edx,%esi
 	decl	%ebp
-	jz	L007mw_end
+	jz	L016mw_end
 	# Tail Round 3 
 	movl	12(%ebx),%eax
 	mull	%ecx
@@ -308,7 +431,7 @@ L006mw_finish2:
 	movl	%eax,12(%edi)
 	movl	%edx,%esi
 	decl	%ebp
-	jz	L007mw_end
+	jz	L016mw_end
 	# Tail Round 4 
 	movl	16(%ebx),%eax
 	mull	%ecx
@@ -317,7 +440,7 @@ L006mw_finish2:
 	movl	%eax,16(%edi)
 	movl	%edx,%esi
 	decl	%ebp
-	jz	L007mw_end
+	jz	L016mw_end
 	# Tail Round 5 
 	movl	20(%ebx),%eax
 	mull	%ecx
@@ -326,7 +449,7 @@ L006mw_finish2:
 	movl	%eax,20(%edi)
 	movl	%edx,%esi
 	decl	%ebp
-	jz	L007mw_end
+	jz	L016mw_end
 	# Tail Round 6 
 	movl	24(%ebx),%eax
 	mull	%ecx
@@ -334,7 +457,7 @@ L006mw_finish2:
 	adcl	$0,%edx
 	movl	%eax,24(%edi)
 	movl	%edx,%esi
-L007mw_end:
+L016mw_end:
 	movl	%esi,%eax
 	popl	%edi
 	popl	%esi
@@ -346,6 +469,28 @@ L007mw_end:
 .align	4
 _bn_sqr_words:
 L_bn_sqr_words_begin:
+	call	L017PIC_me_up
+L017PIC_me_up:
+	popl	%eax
+	movl	L_OPENSSL_ia32cap_P$non_lazy_ptr-L017PIC_me_up(%eax),%eax
+	btl	$26,(%eax)
+	jnc	L018sqr_non_sse2
+	movl	4(%esp),%eax
+	movl	8(%esp),%edx
+	movl	12(%esp),%ecx
+.align	4,0x90
+L019sqr_sse2_loop:
+	movd	(%edx),%mm0
+	pmuludq	%mm0,%mm0
+	leal	4(%edx),%edx
+	movq	%mm0,(%eax)
+	subl	$1,%ecx
+	leal	8(%eax),%eax
+	jnz	L019sqr_sse2_loop
+	emms
+	ret
+.align	4,0x90
+L018sqr_non_sse2:
 	pushl	%ebp
 	pushl	%ebx
 	pushl	%esi
@@ -355,8 +500,8 @@ L_bn_sqr_words_begin:
 	movl	24(%esp),%edi
 	movl	28(%esp),%ebx
 	andl	$4294967288,%ebx
-	jz	L008sw_finish
-L009sw_loop:
+	jz	L020sw_finish
+L021sw_loop:
 	# Round 0 
 	movl	(%edi),%eax
 	mull	%eax
@@ -401,59 +546,59 @@ L009sw_loop:
 	addl	$32,%edi
 	addl	$64,%esi
 	subl	$8,%ebx
-	jnz	L009sw_loop
-L008sw_finish:
+	jnz	L021sw_loop
+L020sw_finish:
 	movl	28(%esp),%ebx
 	andl	$7,%ebx
-	jz	L010sw_end
+	jz	L022sw_end
 	# Tail Round 0 
 	movl	(%edi),%eax
 	mull	%eax
 	movl	%eax,(%esi)
 	decl	%ebx
 	movl	%edx,4(%esi)
-	jz	L010sw_end
+	jz	L022sw_end
 	# Tail Round 1 
 	movl	4(%edi),%eax
 	mull	%eax
 	movl	%eax,8(%esi)
 	decl	%ebx
 	movl	%edx,12(%esi)
-	jz	L010sw_end
+	jz	L022sw_end
 	# Tail Round 2 
 	movl	8(%edi),%eax
 	mull	%eax
 	movl	%eax,16(%esi)
 	decl	%ebx
 	movl	%edx,20(%esi)
-	jz	L010sw_end
+	jz	L022sw_end
 	# Tail Round 3 
 	movl	12(%edi),%eax
 	mull	%eax
 	movl	%eax,24(%esi)
 	decl	%ebx
 	movl	%edx,28(%esi)
-	jz	L010sw_end
+	jz	L022sw_end
 	# Tail Round 4 
 	movl	16(%edi),%eax
 	mull	%eax
 	movl	%eax,32(%esi)
 	decl	%ebx
 	movl	%edx,36(%esi)
-	jz	L010sw_end
+	jz	L022sw_end
 	# Tail Round 5 
 	movl	20(%edi),%eax
 	mull	%eax
 	movl	%eax,40(%esi)
 	decl	%ebx
 	movl	%edx,44(%esi)
-	jz	L010sw_end
+	jz	L022sw_end
 	# Tail Round 6 
 	movl	24(%edi),%eax
 	mull	%eax
 	movl	%eax,48(%esi)
 	movl	%edx,52(%esi)
-L010sw_end:
+L022sw_end:
 	popl	%edi
 	popl	%esi
 	popl	%ebx
@@ -485,8 +630,8 @@ L_bn_add_words_begin:
 	movl	32(%esp),%ebp
 	xorl	%eax,%eax
 	andl	$4294967288,%ebp
-	jz	L011aw_finish
-L012aw_loop:
+	jz	L023aw_finish
+L024aw_loop:
 	# Round 0 
 	movl	(%esi),%ecx
 	movl	(%edi),%edx
@@ -564,11 +709,11 @@ L012aw_loop:
 	addl	$32,%edi
 	addl	$32,%ebx
 	subl	$8,%ebp
-	jnz	L012aw_loop
-L011aw_finish:
+	jnz	L024aw_loop
+L023aw_finish:
 	movl	32(%esp),%ebp
 	andl	$7,%ebp
-	jz	L013aw_end
+	jz	L025aw_end
 	# Tail Round 0 
 	movl	(%esi),%ecx
 	movl	(%edi),%edx
@@ -579,7 +724,7 @@ L011aw_finish:
 	adcl	$0,%eax
 	decl	%ebp
 	movl	%ecx,(%ebx)
-	jz	L013aw_end
+	jz	L025aw_end
 	# Tail Round 1 
 	movl	4(%esi),%ecx
 	movl	4(%edi),%edx
@@ -590,7 +735,7 @@ L011aw_finish:
 	adcl	$0,%eax
 	decl	%ebp
 	movl	%ecx,4(%ebx)
-	jz	L013aw_end
+	jz	L025aw_end
 	# Tail Round 2 
 	movl	8(%esi),%ecx
 	movl	8(%edi),%edx
@@ -601,7 +746,7 @@ L011aw_finish:
 	adcl	$0,%eax
 	decl	%ebp
 	movl	%ecx,8(%ebx)
-	jz	L013aw_end
+	jz	L025aw_end
 	# Tail Round 3 
 	movl	12(%esi),%ecx
 	movl	12(%edi),%edx
@@ -612,7 +757,7 @@ L011aw_finish:
 	adcl	$0,%eax
 	decl	%ebp
 	movl	%ecx,12(%ebx)
-	jz	L013aw_end
+	jz	L025aw_end
 	# Tail Round 4 
 	movl	16(%esi),%ecx
 	movl	16(%edi),%edx
@@ -623,7 +768,7 @@ L011aw_finish:
 	adcl	$0,%eax
 	decl	%ebp
 	movl	%ecx,16(%ebx)
-	jz	L013aw_end
+	jz	L025aw_end
 	# Tail Round 5 
 	movl	20(%esi),%ecx
 	movl	20(%edi),%edx
@@ -634,7 +779,7 @@ L011aw_finish:
 	adcl	$0,%eax
 	decl	%ebp
 	movl	%ecx,20(%ebx)
-	jz	L013aw_end
+	jz	L025aw_end
 	# Tail Round 6 
 	movl	24(%esi),%ecx
 	movl	24(%edi),%edx
@@ -644,7 +789,7 @@ L011aw_finish:
 	addl	%edx,%ecx
 	adcl	$0,%eax
 	movl	%ecx,24(%ebx)
-L013aw_end:
+L025aw_end:
 	popl	%edi
 	popl	%esi
 	popl	%ebx
@@ -666,8 +811,8 @@ L_bn_sub_words_begin:
 	movl	32(%esp),%ebp
 	xorl	%eax,%eax
 	andl	$4294967288,%ebp
-	jz	L014aw_finish
-L015aw_loop:
+	jz	L026aw_finish
+L027aw_loop:
 	# Round 0 
 	movl	(%esi),%ecx
 	movl	(%edi),%edx
@@ -745,11 +890,11 @@ L015aw_loop:
 	addl	$32,%edi
 	addl	$32,%ebx
 	subl	$8,%ebp
-	jnz	L015aw_loop
-L014aw_finish:
+	jnz	L027aw_loop
+L026aw_finish:
 	movl	32(%esp),%ebp
 	andl	$7,%ebp
-	jz	L016aw_end
+	jz	L028aw_end
 	# Tail Round 0 
 	movl	(%esi),%ecx
 	movl	(%edi),%edx
@@ -760,7 +905,7 @@ L014aw_finish:
 	adcl	$0,%eax
 	decl	%ebp
 	movl	%ecx,(%ebx)
-	jz	L016aw_end
+	jz	L028aw_end
 	# Tail Round 1 
 	movl	4(%esi),%ecx
 	movl	4(%edi),%edx
@@ -771,7 +916,7 @@ L014aw_finish:
 	adcl	$0,%eax
 	decl	%ebp
 	movl	%ecx,4(%ebx)
-	jz	L016aw_end
+	jz	L028aw_end
 	# Tail Round 2 
 	movl	8(%esi),%ecx
 	movl	8(%edi),%edx
@@ -782,7 +927,7 @@ L014aw_finish:
 	adcl	$0,%eax
 	decl	%ebp
 	movl	%ecx,8(%ebx)
-	jz	L016aw_end
+	jz	L028aw_end
 	# Tail Round 3 
 	movl	12(%esi),%ecx
 	movl	12(%edi),%edx
@@ -793,7 +938,7 @@ L014aw_finish:
 	adcl	$0,%eax
 	decl	%ebp
 	movl	%ecx,12(%ebx)
-	jz	L016aw_end
+	jz	L028aw_end
 	# Tail Round 4 
 	movl	16(%esi),%ecx
 	movl	16(%edi),%edx
@@ -804,7 +949,7 @@ L014aw_finish:
 	adcl	$0,%eax
 	decl	%ebp
 	movl	%ecx,16(%ebx)
-	jz	L016aw_end
+	jz	L028aw_end
 	# Tail Round 5 
 	movl	20(%esi),%ecx
 	movl	20(%edi),%edx
@@ -815,7 +960,7 @@ L014aw_finish:
 	adcl	$0,%eax
 	decl	%ebp
 	movl	%ecx,20(%ebx)
-	jz	L016aw_end
+	jz	L028aw_end
 	# Tail Round 6 
 	movl	24(%esi),%ecx
 	movl	24(%edi),%edx
@@ -825,7 +970,7 @@ L014aw_finish:
 	subl	%edx,%ecx
 	adcl	$0,%eax
 	movl	%ecx,24(%ebx)
-L016aw_end:
+L028aw_end:
 	popl	%edi
 	popl	%esi
 	popl	%ebx
@@ -847,8 +992,8 @@ L_bn_sub_part_words_begin:
 	movl	32(%esp),%ebp
 	xorl	%eax,%eax
 	andl	$4294967288,%ebp
-	jz	L017aw_finish
-L018aw_loop:
+	jz	L029aw_finish
+L030aw_loop:
 	# Round 0 
 	movl	(%esi),%ecx
 	movl	(%edi),%edx
@@ -926,11 +1071,11 @@ L018aw_loop:
 	addl	$32,%edi
 	addl	$32,%ebx
 	subl	$8,%ebp
-	jnz	L018aw_loop
-L017aw_finish:
+	jnz	L030aw_loop
+L029aw_finish:
 	movl	32(%esp),%ebp
 	andl	$7,%ebp
-	jz	L019aw_end
+	jz	L031aw_end
 	# Tail Round 0 
 	movl	(%esi),%ecx
 	movl	(%edi),%edx
@@ -944,7 +1089,7 @@ L017aw_finish:
 	addl	$4,%edi
 	addl	$4,%ebx
 	decl	%ebp
-	jz	L019aw_end
+	jz	L031aw_end
 	# Tail Round 1 
 	movl	(%esi),%ecx
 	movl	(%edi),%edx
@@ -958,7 +1103,7 @@ L017aw_finish:
 	addl	$4,%edi
 	addl	$4,%ebx
 	decl	%ebp
-	jz	L019aw_end
+	jz	L031aw_end
 	# Tail Round 2 
 	movl	(%esi),%ecx
 	movl	(%edi),%edx
@@ -972,7 +1117,7 @@ L017aw_finish:
 	addl	$4,%edi
 	addl	$4,%ebx
 	decl	%ebp
-	jz	L019aw_end
+	jz	L031aw_end
 	# Tail Round 3 
 	movl	(%esi),%ecx
 	movl	(%edi),%edx
@@ -986,7 +1131,7 @@ L017aw_finish:
 	addl	$4,%edi
 	addl	$4,%ebx
 	decl	%ebp
-	jz	L019aw_end
+	jz	L031aw_end
 	# Tail Round 4 
 	movl	(%esi),%ecx
 	movl	(%edi),%edx
@@ -1000,7 +1145,7 @@ L017aw_finish:
 	addl	$4,%edi
 	addl	$4,%ebx
 	decl	%ebp
-	jz	L019aw_end
+	jz	L031aw_end
 	# Tail Round 5 
 	movl	(%esi),%ecx
 	movl	(%edi),%edx
@@ -1014,7 +1159,7 @@ L017aw_finish:
 	addl	$4,%edi
 	addl	$4,%ebx
 	decl	%ebp
-	jz	L019aw_end
+	jz	L031aw_end
 	# Tail Round 6 
 	movl	(%esi),%ecx
 	movl	(%edi),%edx
@@ -1027,20 +1172,20 @@ L017aw_finish:
 	addl	$4,%esi
 	addl	$4,%edi
 	addl	$4,%ebx
-L019aw_end:
+L031aw_end:
 	cmpl	$0,36(%esp)
-	je	L020pw_end
+	je	L032pw_end
 	movl	36(%esp),%ebp
 	cmpl	$0,%ebp
-	je	L020pw_end
-	jge	L021pw_pos
+	je	L032pw_end
+	jge	L033pw_pos
 	# pw_neg 
 	movl	$0,%edx
 	subl	%ebp,%edx
 	movl	%edx,%ebp
 	andl	$4294967288,%ebp
-	jz	L022pw_neg_finish
-L023pw_neg_loop:
+	jz	L034pw_neg_finish
+L035pw_neg_loop:
 	# dl<0 Round 0 
 	movl	$0,%ecx
 	movl	(%edi),%edx
@@ -1117,13 +1262,13 @@ L023pw_neg_loop:
 	addl	$32,%edi
 	addl	$32,%ebx
 	subl	$8,%ebp
-	jnz	L023pw_neg_loop
-L022pw_neg_finish:
+	jnz	L035pw_neg_loop
+L034pw_neg_finish:
 	movl	36(%esp),%edx
 	movl	$0,%ebp
 	subl	%edx,%ebp
 	andl	$7,%ebp
-	jz	L020pw_end
+	jz	L032pw_end
 	# dl<0 Tail Round 0 
 	movl	$0,%ecx
 	movl	(%edi),%edx
@@ -1134,7 +1279,7 @@ L022pw_neg_finish:
 	adcl	$0,%eax
 	decl	%ebp
 	movl	%ecx,(%ebx)
-	jz	L020pw_end
+	jz	L032pw_end
 	# dl<0 Tail Round 1 
 	movl	$0,%ecx
 	movl	4(%edi),%edx
@@ -1145,7 +1290,7 @@ L022pw_neg_finish:
 	adcl	$0,%eax
 	decl	%ebp
 	movl	%ecx,4(%ebx)
-	jz	L020pw_end
+	jz	L032pw_end
 	# dl<0 Tail Round 2 
 	movl	$0,%ecx
 	movl	8(%edi),%edx
@@ -1156,7 +1301,7 @@ L022pw_neg_finish:
 	adcl	$0,%eax
 	decl	%ebp
 	movl	%ecx,8(%ebx)
-	jz	L020pw_end
+	jz	L032pw_end
 	# dl<0 Tail Round 3 
 	movl	$0,%ecx
 	movl	12(%edi),%edx
@@ -1167,7 +1312,7 @@ L022pw_neg_finish:
 	adcl	$0,%eax
 	decl	%ebp
 	movl	%ecx,12(%ebx)
-	jz	L020pw_end
+	jz	L032pw_end
 	# dl<0 Tail Round 4 
 	movl	$0,%ecx
 	movl	16(%edi),%edx
@@ -1178,7 +1323,7 @@ L022pw_neg_finish:
 	adcl	$0,%eax
 	decl	%ebp
 	movl	%ecx,16(%ebx)
-	jz	L020pw_end
+	jz	L032pw_end
 	# dl<0 Tail Round 5 
 	movl	$0,%ecx
 	movl	20(%edi),%edx
@@ -1189,7 +1334,7 @@ L022pw_neg_finish:
 	adcl	$0,%eax
 	decl	%ebp
 	movl	%ecx,20(%ebx)
-	jz	L020pw_end
+	jz	L032pw_end
 	# dl<0 Tail Round 6 
 	movl	$0,%ecx
 	movl	24(%edi),%edx
@@ -1199,181 +1344,185 @@ L022pw_neg_finish:
 	subl	%edx,%ecx
 	adcl	$0,%eax
 	movl	%ecx,24(%ebx)
-	jmp	L020pw_end
-L021pw_pos:
+	jmp	L032pw_end
+L033pw_pos:
 	andl	$4294967288,%ebp
-	jz	L024pw_pos_finish
-L025pw_pos_loop:
+	jz	L036pw_pos_finish
+L037pw_pos_loop:
 	# dl>0 Round 0 
 	movl	(%esi),%ecx
 	subl	%eax,%ecx
 	movl	%ecx,(%ebx)
-	jnc	L026pw_nc0
+	jnc	L038pw_nc0
 	# dl>0 Round 1 
 	movl	4(%esi),%ecx
 	subl	%eax,%ecx
 	movl	%ecx,4(%ebx)
-	jnc	L027pw_nc1
+	jnc	L039pw_nc1
 	# dl>0 Round 2 
 	movl	8(%esi),%ecx
 	subl	%eax,%ecx
 	movl	%ecx,8(%ebx)
-	jnc	L028pw_nc2
+	jnc	L040pw_nc2
 	# dl>0 Round 3 
 	movl	12(%esi),%ecx
 	subl	%eax,%ecx
 	movl	%ecx,12(%ebx)
-	jnc	L029pw_nc3
+	jnc	L041pw_nc3
 	# dl>0 Round 4 
 	movl	16(%esi),%ecx
 	subl	%eax,%ecx
 	movl	%ecx,16(%ebx)
-	jnc	L030pw_nc4
+	jnc	L042pw_nc4
 	# dl>0 Round 5 
 	movl	20(%esi),%ecx
 	subl	%eax,%ecx
 	movl	%ecx,20(%ebx)
-	jnc	L031pw_nc5
+	jnc	L043pw_nc5
 	# dl>0 Round 6 
 	movl	24(%esi),%ecx
 	subl	%eax,%ecx
 	movl	%ecx,24(%ebx)
-	jnc	L032pw_nc6
+	jnc	L044pw_nc6
 	# dl>0 Round 7 
 	movl	28(%esi),%ecx
 	subl	%eax,%ecx
 	movl	%ecx,28(%ebx)
-	jnc	L033pw_nc7
+	jnc	L045pw_nc7
 
 	addl	$32,%esi
 	addl	$32,%ebx
 	subl	$8,%ebp
-	jnz	L025pw_pos_loop
-L024pw_pos_finish:
+	jnz	L037pw_pos_loop
+L036pw_pos_finish:
 	movl	36(%esp),%ebp
 	andl	$7,%ebp
-	jz	L020pw_end
+	jz	L032pw_end
 	# dl>0 Tail Round 0 
 	movl	(%esi),%ecx
 	subl	%eax,%ecx
 	movl	%ecx,(%ebx)
-	jnc	L034pw_tail_nc0
+	jnc	L046pw_tail_nc0
 	decl	%ebp
-	jz	L020pw_end
+	jz	L032pw_end
 	# dl>0 Tail Round 1 
 	movl	4(%esi),%ecx
 	subl	%eax,%ecx
 	movl	%ecx,4(%ebx)
-	jnc	L035pw_tail_nc1
+	jnc	L047pw_tail_nc1
 	decl	%ebp
-	jz	L020pw_end
+	jz	L032pw_end
 	# dl>0 Tail Round 2 
 	movl	8(%esi),%ecx
 	subl	%eax,%ecx
 	movl	%ecx,8(%ebx)
-	jnc	L036pw_tail_nc2
+	jnc	L048pw_tail_nc2
 	decl	%ebp
-	jz	L020pw_end
+	jz	L032pw_end
 	# dl>0 Tail Round 3 
 	movl	12(%esi),%ecx
 	subl	%eax,%ecx
 	movl	%ecx,12(%ebx)
-	jnc	L037pw_tail_nc3
+	jnc	L049pw_tail_nc3
 	decl	%ebp
-	jz	L020pw_end
+	jz	L032pw_end
 	# dl>0 Tail Round 4 
 	movl	16(%esi),%ecx
 	subl	%eax,%ecx
 	movl	%ecx,16(%ebx)
-	jnc	L038pw_tail_nc4
+	jnc	L050pw_tail_nc4
 	decl	%ebp
-	jz	L020pw_end
+	jz	L032pw_end
 	# dl>0 Tail Round 5 
 	movl	20(%esi),%ecx
 	subl	%eax,%ecx
 	movl	%ecx,20(%ebx)
-	jnc	L039pw_tail_nc5
+	jnc	L051pw_tail_nc5
 	decl	%ebp
-	jz	L020pw_end
+	jz	L032pw_end
 	# dl>0 Tail Round 6 
 	movl	24(%esi),%ecx
 	subl	%eax,%ecx
 	movl	%ecx,24(%ebx)
-	jnc	L040pw_tail_nc6
+	jnc	L052pw_tail_nc6
 	movl	$1,%eax
-	jmp	L020pw_end
-L041pw_nc_loop:
+	jmp	L032pw_end
+L053pw_nc_loop:
 	movl	(%esi),%ecx
 	movl	%ecx,(%ebx)
-L026pw_nc0:
+L038pw_nc0:
 	movl	4(%esi),%ecx
 	movl	%ecx,4(%ebx)
-L027pw_nc1:
+L039pw_nc1:
 	movl	8(%esi),%ecx
 	movl	%ecx,8(%ebx)
-L028pw_nc2:
+L040pw_nc2:
 	movl	12(%esi),%ecx
 	movl	%ecx,12(%ebx)
-L029pw_nc3:
+L041pw_nc3:
 	movl	16(%esi),%ecx
 	movl	%ecx,16(%ebx)
-L030pw_nc4:
+L042pw_nc4:
 	movl	20(%esi),%ecx
 	movl	%ecx,20(%ebx)
-L031pw_nc5:
+L043pw_nc5:
 	movl	24(%esi),%ecx
 	movl	%ecx,24(%ebx)
-L032pw_nc6:
+L044pw_nc6:
 	movl	28(%esi),%ecx
 	movl	%ecx,28(%ebx)
-L033pw_nc7:
+L045pw_nc7:
 
 	addl	$32,%esi
 	addl	$32,%ebx
 	subl	$8,%ebp
-	jnz	L041pw_nc_loop
+	jnz	L053pw_nc_loop
 	movl	36(%esp),%ebp
 	andl	$7,%ebp
-	jz	L042pw_nc_end
+	jz	L054pw_nc_end
 	movl	(%esi),%ecx
 	movl	%ecx,(%ebx)
-L034pw_tail_nc0:
+L046pw_tail_nc0:
 	decl	%ebp
-	jz	L042pw_nc_end
+	jz	L054pw_nc_end
 	movl	4(%esi),%ecx
 	movl	%ecx,4(%ebx)
-L035pw_tail_nc1:
+L047pw_tail_nc1:
 	decl	%ebp
-	jz	L042pw_nc_end
+	jz	L054pw_nc_end
 	movl	8(%esi),%ecx
 	movl	%ecx,8(%ebx)
-L036pw_tail_nc2:
+L048pw_tail_nc2:
 	decl	%ebp
-	jz	L042pw_nc_end
+	jz	L054pw_nc_end
 	movl	12(%esi),%ecx
 	movl	%ecx,12(%ebx)
-L037pw_tail_nc3:
+L049pw_tail_nc3:
 	decl	%ebp
-	jz	L042pw_nc_end
+	jz	L054pw_nc_end
 	movl	16(%esi),%ecx
 	movl	%ecx,16(%ebx)
-L038pw_tail_nc4:
+L050pw_tail_nc4:
 	decl	%ebp
-	jz	L042pw_nc_end
+	jz	L054pw_nc_end
 	movl	20(%esi),%ecx
 	movl	%ecx,20(%ebx)
-L039pw_tail_nc5:
+L051pw_tail_nc5:
 	decl	%ebp
-	jz	L042pw_nc_end
+	jz	L054pw_nc_end
 	movl	24(%esi),%ecx
 	movl	%ecx,24(%ebx)
-L040pw_tail_nc6:
-L042pw_nc_end:
+L052pw_tail_nc6:
+L054pw_nc_end:
 	movl	$0,%eax
-L020pw_end:
+L032pw_end:
 	popl	%edi
 	popl	%esi
 	popl	%ebx
 	popl	%ebp
 	ret
+.section __IMPORT,__pointers,non_lazy_symbol_pointers
+L_OPENSSL_ia32cap_P$non_lazy_ptr:
+.indirect_symbol	_OPENSSL_ia32cap_P
+.long	0
 #endif
diff --git a/mac-x86/crypto/bn/x86-mont.S b/mac-x86/crypto/bn/x86-mont.S
index 1b79c5f..234034b 100644
--- a/mac-x86/crypto/bn/x86-mont.S
+++ b/mac-x86/crypto/bn/x86-mont.S
@@ -43,6 +43,126 @@ L_bn_mul_mont_begin:
 	movl	%esi,20(%esp)
 	leal	-3(%edi),%ebx
 	movl	%ebp,24(%esp)
+	call	L001PIC_me_up
+L001PIC_me_up:
+	popl	%eax
+	movl	L_OPENSSL_ia32cap_P$non_lazy_ptr-L001PIC_me_up(%eax),%eax
+	btl	$26,(%eax)
+	jnc	L002non_sse2
+	movl	$-1,%eax
+	movd	%eax,%mm7
+	movl	8(%esp),%esi
+	movl	12(%esp),%edi
+	movl	16(%esp),%ebp
+	xorl	%edx,%edx
+	xorl	%ecx,%ecx
+	movd	(%edi),%mm4
+	movd	(%esi),%mm5
+	movd	(%ebp),%mm3
+	pmuludq	%mm4,%mm5
+	movq	%mm5,%mm2
+	movq	%mm5,%mm0
+	pand	%mm7,%mm0
+	pmuludq	20(%esp),%mm5
+	pmuludq	%mm5,%mm3
+	paddq	%mm0,%mm3
+	movd	4(%ebp),%mm1
+	movd	4(%esi),%mm0
+	psrlq	$32,%mm2
+	psrlq	$32,%mm3
+	incl	%ecx
+.align	4,0x90
+L0031st:
+	pmuludq	%mm4,%mm0
+	pmuludq	%mm5,%mm1
+	paddq	%mm0,%mm2
+	paddq	%mm1,%mm3
+	movq	%mm2,%mm0
+	pand	%mm7,%mm0
+	movd	4(%ebp,%ecx,4),%mm1
+	paddq	%mm0,%mm3
+	movd	4(%esi,%ecx,4),%mm0
+	psrlq	$32,%mm2
+	movd	%mm3,28(%esp,%ecx,4)
+	psrlq	$32,%mm3
+	leal	1(%ecx),%ecx
+	cmpl	%ebx,%ecx
+	jl	L0031st
+	pmuludq	%mm4,%mm0
+	pmuludq	%mm5,%mm1
+	paddq	%mm0,%mm2
+	paddq	%mm1,%mm3
+	movq	%mm2,%mm0
+	pand	%mm7,%mm0
+	paddq	%mm0,%mm3
+	movd	%mm3,28(%esp,%ecx,4)
+	psrlq	$32,%mm2
+	psrlq	$32,%mm3
+	paddq	%mm2,%mm3
+	movq	%mm3,32(%esp,%ebx,4)
+	incl	%edx
+L004outer:
+	xorl	%ecx,%ecx
+	movd	(%edi,%edx,4),%mm4
+	movd	(%esi),%mm5
+	movd	32(%esp),%mm6
+	movd	(%ebp),%mm3
+	pmuludq	%mm4,%mm5
+	paddq	%mm6,%mm5
+	movq	%mm5,%mm0
+	movq	%mm5,%mm2
+	pand	%mm7,%mm0
+	pmuludq	20(%esp),%mm5
+	pmuludq	%mm5,%mm3
+	paddq	%mm0,%mm3
+	movd	36(%esp),%mm6
+	movd	4(%ebp),%mm1
+	movd	4(%esi),%mm0
+	psrlq	$32,%mm2
+	psrlq	$32,%mm3
+	paddq	%mm6,%mm2
+	incl	%ecx
+	decl	%ebx
+L005inner:
+	pmuludq	%mm4,%mm0
+	pmuludq	%mm5,%mm1
+	paddq	%mm0,%mm2
+	paddq	%mm1,%mm3
+	movq	%mm2,%mm0
+	movd	36(%esp,%ecx,4),%mm6
+	pand	%mm7,%mm0
+	movd	4(%ebp,%ecx,4),%mm1
+	paddq	%mm0,%mm3
+	movd	4(%esi,%ecx,4),%mm0
+	psrlq	$32,%mm2
+	movd	%mm3,28(%esp,%ecx,4)
+	psrlq	$32,%mm3
+	paddq	%mm6,%mm2
+	decl	%ebx
+	leal	1(%ecx),%ecx
+	jnz	L005inner
+	movl	%ecx,%ebx
+	pmuludq	%mm4,%mm0
+	pmuludq	%mm5,%mm1
+	paddq	%mm0,%mm2
+	paddq	%mm1,%mm3
+	movq	%mm2,%mm0
+	pand	%mm7,%mm0
+	paddq	%mm0,%mm3
+	movd	%mm3,28(%esp,%ecx,4)
+	psrlq	$32,%mm2
+	psrlq	$32,%mm3
+	movd	36(%esp,%ebx,4),%mm6
+	paddq	%mm2,%mm3
+	paddq	%mm6,%mm3
+	movq	%mm3,32(%esp,%ebx,4)
+	leal	1(%edx),%edx
+	cmpl	%ebx,%edx
+	jle	L004outer
+	emms
+	jmp	L006common_tail
+.align	4,0x90
+L002non_sse2:
 	movl	8(%esp),%esi
 	leal	1(%ebx),%ebp
 	movl	12(%esp),%edi
@@ -53,12 +173,12 @@ L_bn_mul_mont_begin:
 	leal	4(%edi,%ebx,4),%eax
 	orl	%edx,%ebp
 	movl	(%edi),%edi
-	jz	L001bn_sqr_mont
+	jz	L007bn_sqr_mont
 	movl	%eax,28(%esp)
 	movl	(%esi),%eax
 	xorl	%edx,%edx
 .align	4,0x90
-L002mull:
+L008mull:
 	movl	%edx,%ebp
 	mull	%edi
 	addl	%eax,%ebp
@@ -67,7 +187,7 @@ L002mull:
 	movl	(%esi,%ecx,4),%eax
 	cmpl	%ebx,%ecx
 	movl	%ebp,28(%esp,%ecx,4)
-	jl	L002mull
+	jl	L008mull
 	movl	%edx,%ebp
 	mull	%edi
 	movl	20(%esp),%edi
@@ -85,9 +205,9 @@ L002mull:
 	movl	4(%esi),%eax
 	adcl	$0,%edx
 	incl	%ecx
-	jmp	L0032ndmadd
+	jmp	L0092ndmadd
 .align	4,0x90
-L0041stmadd:
+L0101stmadd:
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ecx,4),%ebp
@@ -98,7 +218,7 @@ L0041stmadd:
 	adcl	$0,%edx
 	cmpl	%ebx,%ecx
 	movl	%ebp,28(%esp,%ecx,4)
-	jl	L0041stmadd
+	jl	L0101stmadd
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ebx,4),%eax
@@ -121,7 +241,7 @@ L0041stmadd:
 	adcl	$0,%edx
 	movl	$1,%ecx
 .align	4,0x90
-L0032ndmadd:
+L0092ndmadd:
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ecx,4),%ebp
@@ -132,7 +252,7 @@ L0032ndmadd:
 	adcl	$0,%edx
 	cmpl	%ebx,%ecx
 	movl	%ebp,24(%esp,%ecx,4)
-	jl	L0032ndmadd
+	jl	L0092ndmadd
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ebx,4),%ebp
@@ -148,16 +268,16 @@ L0032ndmadd:
 	movl	%edx,32(%esp,%ebx,4)
 	cmpl	28(%esp),%ecx
 	movl	%eax,36(%esp,%ebx,4)
-	je	L005common_tail
+	je	L006common_tail
 	movl	(%ecx),%edi
 	movl	8(%esp),%esi
 	movl	%ecx,12(%esp)
 	xorl	%ecx,%ecx
 	xorl	%edx,%edx
 	movl	(%esi),%eax
-	jmp	L0041stmadd
+	jmp	L0101stmadd
 .align	4,0x90
-L001bn_sqr_mont:
+L007bn_sqr_mont:
 	movl	%ebx,(%esp)
 	movl	%ecx,12(%esp)
 	movl	%edi,%eax
@@ -168,7 +288,7 @@ L001bn_sqr_mont:
 	andl	$1,%ebx
 	incl	%ecx
 .align	4,0x90
-L006sqr:
+L011sqr:
 	movl	(%esi,%ecx,4),%eax
 	movl	%edx,%ebp
 	mull	%edi
@@ -180,7 +300,7 @@ L006sqr:
 	cmpl	(%esp),%ecx
 	movl	%eax,%ebx
 	movl	%ebp,28(%esp,%ecx,4)
-	jl	L006sqr
+	jl	L011sqr
 	movl	(%esi,%ecx,4),%eax
 	movl	%edx,%ebp
 	mull	%edi
@@ -204,7 +324,7 @@ L006sqr:
 	movl	4(%esi),%eax
 	movl	$1,%ecx
 .align	4,0x90
-L0073rdmadd:
+L0123rdmadd:
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ecx,4),%ebp
@@ -223,7 +343,7 @@ L0073rdmadd:
 	adcl	$0,%edx
 	cmpl	%ebx,%ecx
 	movl	%ebp,24(%esp,%ecx,4)
-	jl	L0073rdmadd
+	jl	L0123rdmadd
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ebx,4),%ebp
@@ -239,7 +359,7 @@ L0073rdmadd:
 	movl	%edx,32(%esp,%ebx,4)
 	cmpl	%ebx,%ecx
 	movl	%eax,36(%esp,%ebx,4)
-	je	L005common_tail
+	je	L006common_tail
 	movl	4(%esi,%ecx,4),%edi
 	leal	1(%ecx),%ecx
 	movl	%edi,%eax
@@ -251,12 +371,12 @@ L0073rdmadd:
 	xorl	%ebp,%ebp
 	cmpl	%ebx,%ecx
 	leal	1(%ecx),%ecx
-	je	L008sqrlast
+	je	L013sqrlast
 	movl	%edx,%ebx
 	shrl	$1,%edx
 	andl	$1,%ebx
 .align	4,0x90
-L009sqradd:
+L014sqradd:
 	movl	(%esi,%ecx,4),%eax
 	movl	%edx,%ebp
 	mull	%edi
@@ -272,13 +392,13 @@ L009sqradd:
 	cmpl	(%esp),%ecx
 	movl	%ebp,28(%esp,%ecx,4)
 	movl	%eax,%ebx
-	jle	L009sqradd
+	jle	L014sqradd
 	movl	%edx,%ebp
 	addl	%edx,%edx
 	shrl	$31,%ebp
 	addl	%ebx,%edx
 	adcl	$0,%ebp
-L008sqrlast:
+L013sqrlast:
 	movl	20(%esp),%edi
 	movl	16(%esp),%esi
 	imull	32(%esp),%edi
@@ -293,9 +413,9 @@ L008sqrlast:
 	adcl	$0,%edx
 	movl	$1,%ecx
 	movl	4(%esi),%eax
-	jmp	L0073rdmadd
+	jmp	L0123rdmadd
 .align	4,0x90
-L005common_tail:
+L006common_tail:
 	movl	16(%esp),%ebp
 	movl	4(%esp),%edi
 	leal	32(%esp),%esi
@@ -303,16 +423,16 @@ L005common_tail:
 	movl	%ebx,%ecx
 	xorl	%edx,%edx
 .align	4,0x90
-L010sub:
+L015sub:
 	sbbl	(%ebp,%edx,4),%eax
 	movl	%eax,(%edi,%edx,4)
 	decl	%ecx
 	movl	4(%esi,%edx,4),%eax
 	leal	1(%edx),%edx
-	jge	L010sub
+	jge	L015sub
 	sbbl	$0,%eax
 .align	4,0x90
-L011copy:
+L016copy:
 	movl	(%esi,%ebx,4),%edx
 	movl	(%edi,%ebx,4),%ebp
 	xorl	%ebp,%edx
@@ -321,7 +441,7 @@ L011copy:
 	movl	%ecx,(%esi,%ebx,4)
 	movl	%edx,(%edi,%ebx,4)
 	decl	%ebx
-	jge	L011copy
+	jge	L016copy
 	movl	24(%esp),%esp
 	movl	$1,%eax
 L000just_leave:
@@ -335,4 +455,8 @@ L000just_leave:
 .byte	54,44,32,67,82,89,80,84,79,71,65,77,83,32,98,121
 .byte	32,60,97,112,112,114,111,64,111,112,101,110,115,115,108,46
 .byte	111,114,103,62,0
+.section __IMPORT,__pointers,non_lazy_symbol_pointers
+L_OPENSSL_ia32cap_P$non_lazy_ptr:
+.indirect_symbol	_OPENSSL_ia32cap_P
+.long	0
 #endif