deps: update openssl asm and asm_obsolete files

Regenerate asm files with Makefile and CC=gcc and ASM=gcc where
gcc-4.8.4. Also asm files in asm_obsolete dir to support old compiler
and assembler are regenerated without CC and ASM envs.

Fixes: https://github.com/nodejs/node/issues/6458
PR-URL: https://github.com/nodejs/node/pull/6550
Reviewed-By: Myles Borins <myles.borins@gmail.com>
Reviewed-By: Fedor Indutny <fedor.indutny@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>

deps/openssl/asm/x64-elf-gas/aes/aes-x86_64.s

@@ -81,8 +81,8 @@ _x86_64_AES_encrypt:
 	movl	0(%r14,%rdi,8),%edi
 	movl	0(%r14,%rbp,8),%ebp
 
-	andl	$65280,%edi
-	andl	$65280,%ebp
+	andl	$0x0000ff00,%edi
+	andl	$0x0000ff00,%ebp
 
 	xorl	%edi,%r10d
 	xorl	%ebp,%r11d
@@ -94,8 +94,8 @@ _x86_64_AES_encrypt:
 	movl	0(%r14,%rsi,8),%esi
 	movl	0(%r14,%rdi,8),%edi
 
-	andl	$65280,%esi
-	andl	$65280,%edi
+	andl	$0x0000ff00,%esi
+	andl	$0x0000ff00,%edi
 	shrl	$16,%ebx
 	xorl	%esi,%r12d
 	xorl	%edi,%r8d
@@ -108,9 +108,9 @@ _x86_64_AES_encrypt:
 	movl	0(%r14,%rdi,8),%edi
 	movl	0(%r14,%rbp,8),%ebp
 
-	andl	$16711680,%esi
-	andl	$16711680,%edi
-	andl	$16711680,%ebp
+	andl	$0x00ff0000,%esi
+	andl	$0x00ff0000,%edi
+	andl	$0x00ff0000,%ebp
 
 	xorl	%esi,%r10d
 	xorl	%edi,%r11d
@@ -123,9 +123,9 @@ _x86_64_AES_encrypt:
 	movl	2(%r14,%rdi,8),%edi
 	movl	2(%r14,%rbp,8),%ebp
 
-	andl	$16711680,%esi
-	andl	$4278190080,%edi
-	andl	$4278190080,%ebp
+	andl	$0x00ff0000,%esi
+	andl	$0xff000000,%edi
+	andl	$0xff000000,%ebp
 
 	xorl	%esi,%r8d
 	xorl	%edi,%r10d
@@ -138,8 +138,8 @@ _x86_64_AES_encrypt:
 	movl	2(%r14,%rdi,8),%edi
 	movl	16+0(%r15),%eax
 
-	andl	$4278190080,%esi
-	andl	$4278190080,%edi
+	andl	$0xff000000,%esi
+	andl	$0xff000000,%edi
 
 	xorl	%esi,%r12d
 	xorl	%edi,%r8d
@@ -241,8 +241,8 @@ _x86_64_AES_encrypt_compact:
 	xorl	%r8d,%edx
 	cmpq	16(%rsp),%r15
 	je	.Lenc_compact_done
-	movl	$2155905152,%r10d
-	movl	$2155905152,%r11d
+	movl	$0x80808080,%r10d
+	movl	$0x80808080,%r11d
 	andl	%eax,%r10d
 	andl	%ebx,%r11d
 	movl	%r10d,%esi
@@ -253,10 +253,10 @@ _x86_64_AES_encrypt_compact:
 	leal	(%rbx,%rbx,1),%r9d
 	subl	%r10d,%esi
 	subl	%r11d,%edi
-	andl	$4278124286,%r8d
-	andl	$4278124286,%r9d
-	andl	$454761243,%esi
-	andl	$454761243,%edi
+	andl	$0xfefefefe,%r8d
+	andl	$0xfefefefe,%r9d
+	andl	$0x1b1b1b1b,%esi
+	andl	$0x1b1b1b1b,%edi
 	movl	%eax,%r10d
 	movl	%ebx,%r11d
 	xorl	%esi,%r8d
@@ -264,9 +264,9 @@ _x86_64_AES_encrypt_compact:
 
 	xorl	%r8d,%eax
 	xorl	%r9d,%ebx
-	movl	$2155905152,%r12d
+	movl	$0x80808080,%r12d
 	roll	$24,%eax
-	movl	$2155905152,%ebp
+	movl	$0x80808080,%ebp
 	roll	$24,%ebx
 	andl	%ecx,%r12d
 	andl	%edx,%ebp
@@ -289,10 +289,10 @@ _x86_64_AES_encrypt_compact:
 	xorl	%r10d,%eax
 	xorl	%r11d,%ebx
 
-	andl	$4278124286,%r8d
-	andl	$4278124286,%r9d
-	andl	$454761243,%esi
-	andl	$454761243,%edi
+	andl	$0xfefefefe,%r8d
+	andl	$0xfefefefe,%r9d
+	andl	$0x1b1b1b1b,%esi
+	andl	$0x1b1b1b1b,%edi
 	movl	%ecx,%r12d
 	movl	%edx,%ebp
 	xorl	%esi,%r8d
@@ -345,7 +345,7 @@ AES_encrypt:
 	andq	$-64,%rsp
 	subq	%rsp,%rcx
 	negq	%rcx
-	andq	$960,%rcx
+	andq	$0x3c0,%rcx
 	subq	%rcx,%rsp
 	subq	$32,%rsp
 
@@ -370,7 +370,7 @@ AES_encrypt:
 	leaq	.LAES_Te+2048(%rip),%r14
 	leaq	768(%rsp),%rbp
 	subq	%r14,%rbp
-	andq	$768,%rbp
+	andq	$0x300,%rbp
 	leaq	(%r14,%rbp,1),%r14
 
 	call	_x86_64_AES_encrypt_compact
@@ -792,7 +792,7 @@ AES_decrypt:
 	andq	$-64,%rsp
 	subq	%rsp,%rcx
 	negq	%rcx
-	andq	$960,%rcx
+	andq	$0x3c0,%rcx
 	subq	%rcx,%rsp
 	subq	$32,%rsp
 
@@ -817,7 +817,7 @@ AES_decrypt:
 	leaq	.LAES_Td+2048(%rip),%r14
 	leaq	768(%rsp),%rbp
 	subq	%r14,%rbp
-	andq	$768,%rbp
+	andq	$0x300,%rbp
 	leaq	(%r14,%rbp,1),%r14
 	shrq	$3,%rbp
 	addq	%rbp,%r14
@@ -1333,9 +1333,9 @@ AES_cbc_encrypt:
 	movq	%r14,%r10
 	leaq	2304(%r14),%r11
 	movq	%r15,%r12
-	andq	$4095,%r10
-	andq	$4095,%r11
-	andq	$4095,%r12
+	andq	$0xFFF,%r10
+	andq	$0xFFF,%r11
+	andq	$0xFFF,%r12
 
 	cmpq	%r11,%r12
 	jb	.Lcbc_te_break_out
@@ -1344,7 +1344,7 @@ AES_cbc_encrypt:
 	jmp	.Lcbc_te_ok
 .Lcbc_te_break_out:
 	subq	%r10,%r12
-	andq	$4095,%r12
+	andq	$0xFFF,%r12
 	addq	$320,%r12
 	subq	%r12,%r15
 .align	4
@@ -1370,7 +1370,7 @@ AES_cbc_encrypt:
 
 	movq	%r15,%r10
 	subq	%r14,%r10
-	andq	$4095,%r10
+	andq	$0xfff,%r10
 	cmpq	$2304,%r10
 	jb	.Lcbc_do_ecopy
 	cmpq	$4096-248,%r10
@@ -1557,7 +1557,7 @@ AES_cbc_encrypt:
 	leaq	-88-63(%rcx),%r10
 	subq	%rbp,%r10
 	negq	%r10
-	andq	$960,%r10
+	andq	$0x3c0,%r10
 	subq	%r10,%rbp
 
 	xchgq	%rsp,%rbp
@@ -1586,7 +1586,7 @@ AES_cbc_encrypt:
 	leaq	2048(%r14),%r14
 	leaq	768-8(%rsp),%rax
 	subq	%r14,%rax
-	andq	$768,%rax
+	andq	$0x300,%rax
 	leaq	(%r14,%rax,1),%r14
 
 	cmpq	$0,%rbx

deps/openssl/asm/x64-elf-gas/aes/aesni-mb-x86_64.s

@@ -6,6 +6,14 @@
 .type	aesni_multi_cbc_encrypt,@function
 .align	32
 aesni_multi_cbc_encrypt:
+	cmpl	$2,%edx
+	jb	.Lenc_non_avx
+	movl	OPENSSL_ia32cap_P+4(%rip),%ecx
+	testl	$268435456,%ecx
+	jnz	_avx_cbc_enc_shortcut
+	jmp	.Lenc_non_avx
+.align	16
+.Lenc_non_avx:
 	movq	%rsp,%rax
 	pushq	%rbx
 	pushq	%rbp
@@ -262,6 +270,14 @@ aesni_multi_cbc_encrypt:
 .type	aesni_multi_cbc_decrypt,@function
 .align	32
 aesni_multi_cbc_decrypt:
+	cmpl	$2,%edx
+	jb	.Ldec_non_avx
+	movl	OPENSSL_ia32cap_P+4(%rip),%ecx
+	testl	$268435456,%ecx
+	jnz	_avx_cbc_dec_shortcut
+	jmp	.Ldec_non_avx
+.align	16
+.Ldec_non_avx:
 	movq	%rsp,%rax
 	pushq	%rbx
 	pushq	%rbp
@@ -504,3 +520,916 @@ aesni_multi_cbc_decrypt:
 .Ldec4x_epilogue:
 	.byte	0xf3,0xc3
 .size	aesni_multi_cbc_decrypt,.-aesni_multi_cbc_decrypt
+.type	aesni_multi_cbc_encrypt_avx,@function
+.align	32
+aesni_multi_cbc_encrypt_avx:
+_avx_cbc_enc_shortcut:
+	movq	%rsp,%rax
+	pushq	%rbx
+	pushq	%rbp
+	pushq	%r12
+	pushq	%r13
+	pushq	%r14
+	pushq	%r15
+
+
+
+
+
+
+
+
+	subq	$192,%rsp
+	andq	$-128,%rsp
+	movq	%rax,16(%rsp)
+
+.Lenc8x_body:
+	vzeroupper
+	vmovdqu	(%rsi),%xmm15
+	leaq	120(%rsi),%rsi
+	leaq	160(%rdi),%rdi
+	shrl	$1,%edx
+
+.Lenc8x_loop_grande:
+
+	xorl	%edx,%edx
+	movl	-144(%rdi),%ecx
+	movq	-160(%rdi),%r8
+	cmpl	%edx,%ecx
+	movq	-152(%rdi),%rbx
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	-136(%rdi),%xmm2
+	movl	%ecx,32(%rsp)
+	cmovleq	%rsp,%r8
+	subq	%r8,%rbx
+	movq	%rbx,64(%rsp)
+	movl	-104(%rdi),%ecx
+	movq	-120(%rdi),%r9
+	cmpl	%edx,%ecx
+	movq	-112(%rdi),%rbp
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	-96(%rdi),%xmm3
+	movl	%ecx,36(%rsp)
+	cmovleq	%rsp,%r9
+	subq	%r9,%rbp
+	movq	%rbp,72(%rsp)
+	movl	-64(%rdi),%ecx
+	movq	-80(%rdi),%r10
+	cmpl	%edx,%ecx
+	movq	-72(%rdi),%rbp
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	-56(%rdi),%xmm4
+	movl	%ecx,40(%rsp)
+	cmovleq	%rsp,%r10
+	subq	%r10,%rbp
+	movq	%rbp,80(%rsp)
+	movl	-24(%rdi),%ecx
+	movq	-40(%rdi),%r11
+	cmpl	%edx,%ecx
+	movq	-32(%rdi),%rbp
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	-16(%rdi),%xmm5
+	movl	%ecx,44(%rsp)
+	cmovleq	%rsp,%r11
+	subq	%r11,%rbp
+	movq	%rbp,88(%rsp)
+	movl	16(%rdi),%ecx
+	movq	0(%rdi),%r12
+	cmpl	%edx,%ecx
+	movq	8(%rdi),%rbp
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	24(%rdi),%xmm6
+	movl	%ecx,48(%rsp)
+	cmovleq	%rsp,%r12
+	subq	%r12,%rbp
+	movq	%rbp,96(%rsp)
+	movl	56(%rdi),%ecx
+	movq	40(%rdi),%r13
+	cmpl	%edx,%ecx
+	movq	48(%rdi),%rbp
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	64(%rdi),%xmm7
+	movl	%ecx,52(%rsp)
+	cmovleq	%rsp,%r13
+	subq	%r13,%rbp
+	movq	%rbp,104(%rsp)
+	movl	96(%rdi),%ecx
+	movq	80(%rdi),%r14
+	cmpl	%edx,%ecx
+	movq	88(%rdi),%rbp
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	104(%rdi),%xmm8
+	movl	%ecx,56(%rsp)
+	cmovleq	%rsp,%r14
+	subq	%r14,%rbp
+	movq	%rbp,112(%rsp)
+	movl	136(%rdi),%ecx
+	movq	120(%rdi),%r15
+	cmpl	%edx,%ecx
+	movq	128(%rdi),%rbp
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	144(%rdi),%xmm9
+	movl	%ecx,60(%rsp)
+	cmovleq	%rsp,%r15
+	subq	%r15,%rbp
+	movq	%rbp,120(%rsp)
+	testl	%edx,%edx
+	jz	.Lenc8x_done
+
+	vmovups	16-120(%rsi),%xmm1
+	vmovups	32-120(%rsi),%xmm0
+	movl	240-120(%rsi),%eax
+
+	vpxor	(%r8),%xmm15,%xmm10
+	leaq	128(%rsp),%rbp
+	vpxor	(%r9),%xmm15,%xmm11
+	vpxor	(%r10),%xmm15,%xmm12
+	vpxor	(%r11),%xmm15,%xmm13
+	vpxor	%xmm10,%xmm2,%xmm2
+	vpxor	(%r12),%xmm15,%xmm10
+	vpxor	%xmm11,%xmm3,%xmm3
+	vpxor	(%r13),%xmm15,%xmm11
+	vpxor	%xmm12,%xmm4,%xmm4
+	vpxor	(%r14),%xmm15,%xmm12
+	vpxor	%xmm13,%xmm5,%xmm5
+	vpxor	(%r15),%xmm15,%xmm13
+	vpxor	%xmm10,%xmm6,%xmm6
+	movl	$1,%ecx
+	vpxor	%xmm11,%xmm7,%xmm7
+	vpxor	%xmm12,%xmm8,%xmm8
+	vpxor	%xmm13,%xmm9,%xmm9
+	jmp	.Loop_enc8x
+
+.align	32
+.Loop_enc8x:
+	vaesenc	%xmm1,%xmm2,%xmm2
+	cmpl	32+0(%rsp),%ecx
+	vaesenc	%xmm1,%xmm3,%xmm3
+	prefetcht0	31(%r8)
+	vaesenc	%xmm1,%xmm4,%xmm4
+	vaesenc	%xmm1,%xmm5,%xmm5
+	leaq	(%r8,%rbx,1),%rbx
+	cmovgeq	%rsp,%r8
+	vaesenc	%xmm1,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesenc	%xmm1,%xmm7,%xmm7
+	subq	%r8,%rbx
+	vaesenc	%xmm1,%xmm8,%xmm8
+	vpxor	16(%r8),%xmm15,%xmm10
+	movq	%rbx,64+0(%rsp)
+	vaesenc	%xmm1,%xmm9,%xmm9
+	vmovups	-72(%rsi),%xmm1
+	leaq	16(%r8,%rbx,1),%r8
+	vmovdqu	%xmm10,0(%rbp)
+	vaesenc	%xmm0,%xmm2,%xmm2
+	cmpl	32+4(%rsp),%ecx
+	movq	64+8(%rsp),%rbx
+	vaesenc	%xmm0,%xmm3,%xmm3
+	prefetcht0	31(%r9)
+	vaesenc	%xmm0,%xmm4,%xmm4
+	vaesenc	%xmm0,%xmm5,%xmm5
+	leaq	(%r9,%rbx,1),%rbx
+	cmovgeq	%rsp,%r9
+	vaesenc	%xmm0,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesenc	%xmm0,%xmm7,%xmm7
+	subq	%r9,%rbx
+	vaesenc	%xmm0,%xmm8,%xmm8
+	vpxor	16(%r9),%xmm15,%xmm11
+	movq	%rbx,64+8(%rsp)
+	vaesenc	%xmm0,%xmm9,%xmm9
+	vmovups	-56(%rsi),%xmm0
+	leaq	16(%r9,%rbx,1),%r9
+	vmovdqu	%xmm11,16(%rbp)
+	vaesenc	%xmm1,%xmm2,%xmm2
+	cmpl	32+8(%rsp),%ecx
+	movq	64+16(%rsp),%rbx
+	vaesenc	%xmm1,%xmm3,%xmm3
+	prefetcht0	31(%r10)
+	vaesenc	%xmm1,%xmm4,%xmm4
+	prefetcht0	15(%r8)
+	vaesenc	%xmm1,%xmm5,%xmm5
+	leaq	(%r10,%rbx,1),%rbx
+	cmovgeq	%rsp,%r10
+	vaesenc	%xmm1,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesenc	%xmm1,%xmm7,%xmm7
+	subq	%r10,%rbx
+	vaesenc	%xmm1,%xmm8,%xmm8
+	vpxor	16(%r10),%xmm15,%xmm12
+	movq	%rbx,64+16(%rsp)
+	vaesenc	%xmm1,%xmm9,%xmm9
+	vmovups	-40(%rsi),%xmm1
+	leaq	16(%r10,%rbx,1),%r10
+	vmovdqu	%xmm12,32(%rbp)
+	vaesenc	%xmm0,%xmm2,%xmm2
+	cmpl	32+12(%rsp),%ecx
+	movq	64+24(%rsp),%rbx
+	vaesenc	%xmm0,%xmm3,%xmm3
+	prefetcht0	31(%r11)
+	vaesenc	%xmm0,%xmm4,%xmm4
+	prefetcht0	15(%r9)
+	vaesenc	%xmm0,%xmm5,%xmm5
+	leaq	(%r11,%rbx,1),%rbx
+	cmovgeq	%rsp,%r11
+	vaesenc	%xmm0,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesenc	%xmm0,%xmm7,%xmm7
+	subq	%r11,%rbx
+	vaesenc	%xmm0,%xmm8,%xmm8
+	vpxor	16(%r11),%xmm15,%xmm13
+	movq	%rbx,64+24(%rsp)
+	vaesenc	%xmm0,%xmm9,%xmm9
+	vmovups	-24(%rsi),%xmm0
+	leaq	16(%r11,%rbx,1),%r11
+	vmovdqu	%xmm13,48(%rbp)
+	vaesenc	%xmm1,%xmm2,%xmm2
+	cmpl	32+16(%rsp),%ecx
+	movq	64+32(%rsp),%rbx
+	vaesenc	%xmm1,%xmm3,%xmm3
+	prefetcht0	31(%r12)
+	vaesenc	%xmm1,%xmm4,%xmm4
+	prefetcht0	15(%r10)
+	vaesenc	%xmm1,%xmm5,%xmm5
+	leaq	(%r12,%rbx,1),%rbx
+	cmovgeq	%rsp,%r12
+	vaesenc	%xmm1,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesenc	%xmm1,%xmm7,%xmm7
+	subq	%r12,%rbx
+	vaesenc	%xmm1,%xmm8,%xmm8
+	vpxor	16(%r12),%xmm15,%xmm10
+	movq	%rbx,64+32(%rsp)
+	vaesenc	%xmm1,%xmm9,%xmm9
+	vmovups	-8(%rsi),%xmm1
+	leaq	16(%r12,%rbx,1),%r12
+	vaesenc	%xmm0,%xmm2,%xmm2
+	cmpl	32+20(%rsp),%ecx
+	movq	64+40(%rsp),%rbx
+	vaesenc	%xmm0,%xmm3,%xmm3
+	prefetcht0	31(%r13)
+	vaesenc	%xmm0,%xmm4,%xmm4
+	prefetcht0	15(%r11)
+	vaesenc	%xmm0,%xmm5,%xmm5
+	leaq	(%rbx,%r13,1),%rbx
+	cmovgeq	%rsp,%r13
+	vaesenc	%xmm0,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesenc	%xmm0,%xmm7,%xmm7
+	subq	%r13,%rbx
+	vaesenc	%xmm0,%xmm8,%xmm8
+	vpxor	16(%r13),%xmm15,%xmm11
+	movq	%rbx,64+40(%rsp)
+	vaesenc	%xmm0,%xmm9,%xmm9
+	vmovups	8(%rsi),%xmm0
+	leaq	16(%r13,%rbx,1),%r13
+	vaesenc	%xmm1,%xmm2,%xmm2
+	cmpl	32+24(%rsp),%ecx
+	movq	64+48(%rsp),%rbx
+	vaesenc	%xmm1,%xmm3,%xmm3
+	prefetcht0	31(%r14)
+	vaesenc	%xmm1,%xmm4,%xmm4
+	prefetcht0	15(%r12)
+	vaesenc	%xmm1,%xmm5,%xmm5
+	leaq	(%r14,%rbx,1),%rbx
+	cmovgeq	%rsp,%r14
+	vaesenc	%xmm1,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesenc	%xmm1,%xmm7,%xmm7
+	subq	%r14,%rbx
+	vaesenc	%xmm1,%xmm8,%xmm8
+	vpxor	16(%r14),%xmm15,%xmm12
+	movq	%rbx,64+48(%rsp)
+	vaesenc	%xmm1,%xmm9,%xmm9
+	vmovups	24(%rsi),%xmm1
+	leaq	16(%r14,%rbx,1),%r14
+	vaesenc	%xmm0,%xmm2,%xmm2
+	cmpl	32+28(%rsp),%ecx
+	movq	64+56(%rsp),%rbx
+	vaesenc	%xmm0,%xmm3,%xmm3
+	prefetcht0	31(%r15)
+	vaesenc	%xmm0,%xmm4,%xmm4
+	prefetcht0	15(%r13)
+	vaesenc	%xmm0,%xmm5,%xmm5
+	leaq	(%r15,%rbx,1),%rbx
+	cmovgeq	%rsp,%r15
+	vaesenc	%xmm0,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesenc	%xmm0,%xmm7,%xmm7
+	subq	%r15,%rbx
+	vaesenc	%xmm0,%xmm8,%xmm8
+	vpxor	16(%r15),%xmm15,%xmm13
+	movq	%rbx,64+56(%rsp)
+	vaesenc	%xmm0,%xmm9,%xmm9
+	vmovups	40(%rsi),%xmm0
+	leaq	16(%r15,%rbx,1),%r15
+	vmovdqu	32(%rsp),%xmm14
+	prefetcht0	15(%r14)
+	prefetcht0	15(%r15)
+	cmpl	$11,%eax
+	jb	.Lenc8x_tail
+
+	vaesenc	%xmm1,%xmm2,%xmm2
+	vaesenc	%xmm1,%xmm3,%xmm3
+	vaesenc	%xmm1,%xmm4,%xmm4
+	vaesenc	%xmm1,%xmm5,%xmm5
+	vaesenc	%xmm1,%xmm6,%xmm6
+	vaesenc	%xmm1,%xmm7,%xmm7
+	vaesenc	%xmm1,%xmm8,%xmm8
+	vaesenc	%xmm1,%xmm9,%xmm9
+	vmovups	176-120(%rsi),%xmm1
+
+	vaesenc	%xmm0,%xmm2,%xmm2
+	vaesenc	%xmm0,%xmm3,%xmm3
+	vaesenc	%xmm0,%xmm4,%xmm4
+	vaesenc	%xmm0,%xmm5,%xmm5
+	vaesenc	%xmm0,%xmm6,%xmm6
+	vaesenc	%xmm0,%xmm7,%xmm7
+	vaesenc	%xmm0,%xmm8,%xmm8
+	vaesenc	%xmm0,%xmm9,%xmm9
+	vmovups	192-120(%rsi),%xmm0
+	je	.Lenc8x_tail
+
+	vaesenc	%xmm1,%xmm2,%xmm2
+	vaesenc	%xmm1,%xmm3,%xmm3
+	vaesenc	%xmm1,%xmm4,%xmm4
+	vaesenc	%xmm1,%xmm5,%xmm5
+	vaesenc	%xmm1,%xmm6,%xmm6
+	vaesenc	%xmm1,%xmm7,%xmm7
+	vaesenc	%xmm1,%xmm8,%xmm8
+	vaesenc	%xmm1,%xmm9,%xmm9
+	vmovups	208-120(%rsi),%xmm1
+
+	vaesenc	%xmm0,%xmm2,%xmm2
+	vaesenc	%xmm0,%xmm3,%xmm3
+	vaesenc	%xmm0,%xmm4,%xmm4
+	vaesenc	%xmm0,%xmm5,%xmm5
+	vaesenc	%xmm0,%xmm6,%xmm6
+	vaesenc	%xmm0,%xmm7,%xmm7
+	vaesenc	%xmm0,%xmm8,%xmm8
+	vaesenc	%xmm0,%xmm9,%xmm9
+	vmovups	224-120(%rsi),%xmm0
+
+.Lenc8x_tail:
+	vaesenc	%xmm1,%xmm2,%xmm2
+	vpxor	%xmm15,%xmm15,%xmm15
+	vaesenc	%xmm1,%xmm3,%xmm3
+	vaesenc	%xmm1,%xmm4,%xmm4
+	vpcmpgtd	%xmm15,%xmm14,%xmm15
+	vaesenc	%xmm1,%xmm5,%xmm5
+	vaesenc	%xmm1,%xmm6,%xmm6
+	vpaddd	%xmm14,%xmm15,%xmm15
+	vmovdqu	48(%rsp),%xmm14
+	vaesenc	%xmm1,%xmm7,%xmm7
+	movq	64(%rsp),%rbx
+	vaesenc	%xmm1,%xmm8,%xmm8
+	vaesenc	%xmm1,%xmm9,%xmm9
+	vmovups	16-120(%rsi),%xmm1
+
+	vaesenclast	%xmm0,%xmm2,%xmm2
+	vmovdqa	%xmm15,32(%rsp)
+	vpxor	%xmm15,%xmm15,%xmm15
+	vaesenclast	%xmm0,%xmm3,%xmm3
+	vaesenclast	%xmm0,%xmm4,%xmm4
+	vpcmpgtd	%xmm15,%xmm14,%xmm15
+	vaesenclast	%xmm0,%xmm5,%xmm5
+	vaesenclast	%xmm0,%xmm6,%xmm6
+	vpaddd	%xmm15,%xmm14,%xmm14
+	vmovdqu	-120(%rsi),%xmm15
+	vaesenclast	%xmm0,%xmm7,%xmm7
+	vaesenclast	%xmm0,%xmm8,%xmm8
+	vmovdqa	%xmm14,48(%rsp)
+	vaesenclast	%xmm0,%xmm9,%xmm9
+	vmovups	32-120(%rsi),%xmm0
+
+	vmovups	%xmm2,-16(%r8)
+	subq	%rbx,%r8
+	vpxor	0(%rbp),%xmm2,%xmm2
+	vmovups	%xmm3,-16(%r9)
+	subq	72(%rsp),%r9
+	vpxor	16(%rbp),%xmm3,%xmm3
+	vmovups	%xmm4,-16(%r10)
+	subq	80(%rsp),%r10
+	vpxor	32(%rbp),%xmm4,%xmm4
+	vmovups	%xmm5,-16(%r11)
+	subq	88(%rsp),%r11
+	vpxor	48(%rbp),%xmm5,%xmm5
+	vmovups	%xmm6,-16(%r12)
+	subq	96(%rsp),%r12
+	vpxor	%xmm10,%xmm6,%xmm6
+	vmovups	%xmm7,-16(%r13)
+	subq	104(%rsp),%r13
+	vpxor	%xmm11,%xmm7,%xmm7
+	vmovups	%xmm8,-16(%r14)
+	subq	112(%rsp),%r14
+	vpxor	%xmm12,%xmm8,%xmm8
+	vmovups	%xmm9,-16(%r15)
+	subq	120(%rsp),%r15
+	vpxor	%xmm13,%xmm9,%xmm9
+
+	decl	%edx
+	jnz	.Loop_enc8x
+
+	movq	16(%rsp),%rax
+
+
+
+
+
+.Lenc8x_done:
+	vzeroupper
+	movq	-48(%rax),%r15
+	movq	-40(%rax),%r14
+	movq	-32(%rax),%r13
+	movq	-24(%rax),%r12
+	movq	-16(%rax),%rbp
+	movq	-8(%rax),%rbx
+	leaq	(%rax),%rsp
+.Lenc8x_epilogue:
+	.byte	0xf3,0xc3
+.size	aesni_multi_cbc_encrypt_avx,.-aesni_multi_cbc_encrypt_avx
+
+.type	aesni_multi_cbc_decrypt_avx,@function
+.align	32
+aesni_multi_cbc_decrypt_avx:
+_avx_cbc_dec_shortcut:
+	movq	%rsp,%rax
+	pushq	%rbx
+	pushq	%rbp
+	pushq	%r12
+	pushq	%r13
+	pushq	%r14
+	pushq	%r15
+
+
+
+
+
+
+
+
+
+	subq	$256,%rsp
+	andq	$-256,%rsp
+	subq	$192,%rsp
+	movq	%rax,16(%rsp)
+
+.Ldec8x_body:
+	vzeroupper
+	vmovdqu	(%rsi),%xmm15
+	leaq	120(%rsi),%rsi
+	leaq	160(%rdi),%rdi
+	shrl	$1,%edx
+
+.Ldec8x_loop_grande:
+
+	xorl	%edx,%edx
+	movl	-144(%rdi),%ecx
+	movq	-160(%rdi),%r8
+	cmpl	%edx,%ecx
+	movq	-152(%rdi),%rbx
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	-136(%rdi),%xmm2
+	movl	%ecx,32(%rsp)
+	cmovleq	%rsp,%r8
+	subq	%r8,%rbx
+	movq	%rbx,64(%rsp)
+	vmovdqu	%xmm2,192(%rsp)
+	movl	-104(%rdi),%ecx
+	movq	-120(%rdi),%r9
+	cmpl	%edx,%ecx
+	movq	-112(%rdi),%rbp
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	-96(%rdi),%xmm3
+	movl	%ecx,36(%rsp)
+	cmovleq	%rsp,%r9
+	subq	%r9,%rbp
+	movq	%rbp,72(%rsp)
+	vmovdqu	%xmm3,208(%rsp)
+	movl	-64(%rdi),%ecx
+	movq	-80(%rdi),%r10
+	cmpl	%edx,%ecx
+	movq	-72(%rdi),%rbp
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	-56(%rdi),%xmm4
+	movl	%ecx,40(%rsp)
+	cmovleq	%rsp,%r10
+	subq	%r10,%rbp
+	movq	%rbp,80(%rsp)
+	vmovdqu	%xmm4,224(%rsp)
+	movl	-24(%rdi),%ecx
+	movq	-40(%rdi),%r11
+	cmpl	%edx,%ecx
+	movq	-32(%rdi),%rbp
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	-16(%rdi),%xmm5
+	movl	%ecx,44(%rsp)
+	cmovleq	%rsp,%r11
+	subq	%r11,%rbp
+	movq	%rbp,88(%rsp)
+	vmovdqu	%xmm5,240(%rsp)
+	movl	16(%rdi),%ecx
+	movq	0(%rdi),%r12
+	cmpl	%edx,%ecx
+	movq	8(%rdi),%rbp
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	24(%rdi),%xmm6
+	movl	%ecx,48(%rsp)
+	cmovleq	%rsp,%r12
+	subq	%r12,%rbp
+	movq	%rbp,96(%rsp)
+	vmovdqu	%xmm6,256(%rsp)
+	movl	56(%rdi),%ecx
+	movq	40(%rdi),%r13
+	cmpl	%edx,%ecx
+	movq	48(%rdi),%rbp
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	64(%rdi),%xmm7
+	movl	%ecx,52(%rsp)
+	cmovleq	%rsp,%r13
+	subq	%r13,%rbp
+	movq	%rbp,104(%rsp)
+	vmovdqu	%xmm7,272(%rsp)
+	movl	96(%rdi),%ecx
+	movq	80(%rdi),%r14
+	cmpl	%edx,%ecx
+	movq	88(%rdi),%rbp
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	104(%rdi),%xmm8
+	movl	%ecx,56(%rsp)
+	cmovleq	%rsp,%r14
+	subq	%r14,%rbp
+	movq	%rbp,112(%rsp)
+	vmovdqu	%xmm8,288(%rsp)
+	movl	136(%rdi),%ecx
+	movq	120(%rdi),%r15
+	cmpl	%edx,%ecx
+	movq	128(%rdi),%rbp
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	144(%rdi),%xmm9
+	movl	%ecx,60(%rsp)
+	cmovleq	%rsp,%r15
+	subq	%r15,%rbp
+	movq	%rbp,120(%rsp)
+	vmovdqu	%xmm9,304(%rsp)
+	testl	%edx,%edx
+	jz	.Ldec8x_done
+
+	vmovups	16-120(%rsi),%xmm1
+	vmovups	32-120(%rsi),%xmm0
+	movl	240-120(%rsi),%eax
+	leaq	192+128(%rsp),%rbp
+
+	vmovdqu	(%r8),%xmm2
+	vmovdqu	(%r9),%xmm3
+	vmovdqu	(%r10),%xmm4
+	vmovdqu	(%r11),%xmm5
+	vmovdqu	(%r12),%xmm6
+	vmovdqu	(%r13),%xmm7
+	vmovdqu	(%r14),%xmm8
+	vmovdqu	(%r15),%xmm9
+	vmovdqu	%xmm2,0(%rbp)
+	vpxor	%xmm15,%xmm2,%xmm2
+	vmovdqu	%xmm3,16(%rbp)
+	vpxor	%xmm15,%xmm3,%xmm3
+	vmovdqu	%xmm4,32(%rbp)
+	vpxor	%xmm15,%xmm4,%xmm4
+	vmovdqu	%xmm5,48(%rbp)
+	vpxor	%xmm15,%xmm5,%xmm5
+	vmovdqu	%xmm6,64(%rbp)
+	vpxor	%xmm15,%xmm6,%xmm6
+	vmovdqu	%xmm7,80(%rbp)
+	vpxor	%xmm15,%xmm7,%xmm7
+	vmovdqu	%xmm8,96(%rbp)
+	vpxor	%xmm15,%xmm8,%xmm8
+	vmovdqu	%xmm9,112(%rbp)
+	vpxor	%xmm15,%xmm9,%xmm9
+	xorq	$0x80,%rbp
+	movl	$1,%ecx
+	jmp	.Loop_dec8x
+
+.align	32
+.Loop_dec8x:
+	vaesdec	%xmm1,%xmm2,%xmm2
+	cmpl	32+0(%rsp),%ecx
+	vaesdec	%xmm1,%xmm3,%xmm3
+	prefetcht0	31(%r8)
+	vaesdec	%xmm1,%xmm4,%xmm4
+	vaesdec	%xmm1,%xmm5,%xmm5
+	leaq	(%r8,%rbx,1),%rbx
+	cmovgeq	%rsp,%r8
+	vaesdec	%xmm1,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesdec	%xmm1,%xmm7,%xmm7
+	subq	%r8,%rbx
+	vaesdec	%xmm1,%xmm8,%xmm8
+	vmovdqu	16(%r8),%xmm10
+	movq	%rbx,64+0(%rsp)
+	vaesdec	%xmm1,%xmm9,%xmm9
+	vmovups	-72(%rsi),%xmm1
+	leaq	16(%r8,%rbx,1),%r8
+	vmovdqu	%xmm10,128(%rsp)
+	vaesdec	%xmm0,%xmm2,%xmm2
+	cmpl	32+4(%rsp),%ecx
+	movq	64+8(%rsp),%rbx
+	vaesdec	%xmm0,%xmm3,%xmm3
+	prefetcht0	31(%r9)
+	vaesdec	%xmm0,%xmm4,%xmm4
+	vaesdec	%xmm0,%xmm5,%xmm5
+	leaq	(%r9,%rbx,1),%rbx
+	cmovgeq	%rsp,%r9
+	vaesdec	%xmm0,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesdec	%xmm0,%xmm7,%xmm7
+	subq	%r9,%rbx
+	vaesdec	%xmm0,%xmm8,%xmm8
+	vmovdqu	16(%r9),%xmm11
+	movq	%rbx,64+8(%rsp)
+	vaesdec	%xmm0,%xmm9,%xmm9
+	vmovups	-56(%rsi),%xmm0
+	leaq	16(%r9,%rbx,1),%r9
+	vmovdqu	%xmm11,144(%rsp)
+	vaesdec	%xmm1,%xmm2,%xmm2
+	cmpl	32+8(%rsp),%ecx
+	movq	64+16(%rsp),%rbx
+	vaesdec	%xmm1,%xmm3,%xmm3
+	prefetcht0	31(%r10)
+	vaesdec	%xmm1,%xmm4,%xmm4
+	prefetcht0	15(%r8)
+	vaesdec	%xmm1,%xmm5,%xmm5
+	leaq	(%r10,%rbx,1),%rbx
+	cmovgeq	%rsp,%r10
+	vaesdec	%xmm1,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesdec	%xmm1,%xmm7,%xmm7
+	subq	%r10,%rbx
+	vaesdec	%xmm1,%xmm8,%xmm8
+	vmovdqu	16(%r10),%xmm12
+	movq	%rbx,64+16(%rsp)
+	vaesdec	%xmm1,%xmm9,%xmm9
+	vmovups	-40(%rsi),%xmm1
+	leaq	16(%r10,%rbx,1),%r10
+	vmovdqu	%xmm12,160(%rsp)
+	vaesdec	%xmm0,%xmm2,%xmm2
+	cmpl	32+12(%rsp),%ecx
+	movq	64+24(%rsp),%rbx
+	vaesdec	%xmm0,%xmm3,%xmm3
+	prefetcht0	31(%r11)
+	vaesdec	%xmm0,%xmm4,%xmm4
+	prefetcht0	15(%r9)
+	vaesdec	%xmm0,%xmm5,%xmm5
+	leaq	(%r11,%rbx,1),%rbx
+	cmovgeq	%rsp,%r11
+	vaesdec	%xmm0,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesdec	%xmm0,%xmm7,%xmm7
+	subq	%r11,%rbx
+	vaesdec	%xmm0,%xmm8,%xmm8
+	vmovdqu	16(%r11),%xmm13
+	movq	%rbx,64+24(%rsp)
+	vaesdec	%xmm0,%xmm9,%xmm9
+	vmovups	-24(%rsi),%xmm0
+	leaq	16(%r11,%rbx,1),%r11
+	vmovdqu	%xmm13,176(%rsp)
+	vaesdec	%xmm1,%xmm2,%xmm2
+	cmpl	32+16(%rsp),%ecx
+	movq	64+32(%rsp),%rbx
+	vaesdec	%xmm1,%xmm3,%xmm3
+	prefetcht0	31(%r12)
+	vaesdec	%xmm1,%xmm4,%xmm4
+	prefetcht0	15(%r10)
+	vaesdec	%xmm1,%xmm5,%xmm5
+	leaq	(%r12,%rbx,1),%rbx
+	cmovgeq	%rsp,%r12
+	vaesdec	%xmm1,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesdec	%xmm1,%xmm7,%xmm7
+	subq	%r12,%rbx
+	vaesdec	%xmm1,%xmm8,%xmm8
+	vmovdqu	16(%r12),%xmm10
+	movq	%rbx,64+32(%rsp)
+	vaesdec	%xmm1,%xmm9,%xmm9
+	vmovups	-8(%rsi),%xmm1
+	leaq	16(%r12,%rbx,1),%r12
+	vaesdec	%xmm0,%xmm2,%xmm2
+	cmpl	32+20(%rsp),%ecx
+	movq	64+40(%rsp),%rbx
+	vaesdec	%xmm0,%xmm3,%xmm3
+	prefetcht0	31(%r13)
+	vaesdec	%xmm0,%xmm4,%xmm4
+	prefetcht0	15(%r11)
+	vaesdec	%xmm0,%xmm5,%xmm5
+	leaq	(%rbx,%r13,1),%rbx
+	cmovgeq	%rsp,%r13
+	vaesdec	%xmm0,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesdec	%xmm0,%xmm7,%xmm7
+	subq	%r13,%rbx
+	vaesdec	%xmm0,%xmm8,%xmm8
+	vmovdqu	16(%r13),%xmm11
+	movq	%rbx,64+40(%rsp)
+	vaesdec	%xmm0,%xmm9,%xmm9
+	vmovups	8(%rsi),%xmm0
+	leaq	16(%r13,%rbx,1),%r13
+	vaesdec	%xmm1,%xmm2,%xmm2
+	cmpl	32+24(%rsp),%ecx
+	movq	64+48(%rsp),%rbx
+	vaesdec	%xmm1,%xmm3,%xmm3
+	prefetcht0	31(%r14)
+	vaesdec	%xmm1,%xmm4,%xmm4
+	prefetcht0	15(%r12)
+	vaesdec	%xmm1,%xmm5,%xmm5
+	leaq	(%r14,%rbx,1),%rbx
+	cmovgeq	%rsp,%r14
+	vaesdec	%xmm1,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesdec	%xmm1,%xmm7,%xmm7
+	subq	%r14,%rbx
+	vaesdec	%xmm1,%xmm8,%xmm8
+	vmovdqu	16(%r14),%xmm12
+	movq	%rbx,64+48(%rsp)
+	vaesdec	%xmm1,%xmm9,%xmm9
+	vmovups	24(%rsi),%xmm1
+	leaq	16(%r14,%rbx,1),%r14
+	vaesdec	%xmm0,%xmm2,%xmm2
+	cmpl	32+28(%rsp),%ecx
+	movq	64+56(%rsp),%rbx
+	vaesdec	%xmm0,%xmm3,%xmm3
+	prefetcht0	31(%r15)
+	vaesdec	%xmm0,%xmm4,%xmm4
+	prefetcht0	15(%r13)
+	vaesdec	%xmm0,%xmm5,%xmm5
+	leaq	(%r15,%rbx,1),%rbx
+	cmovgeq	%rsp,%r15
+	vaesdec	%xmm0,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesdec	%xmm0,%xmm7,%xmm7
+	subq	%r15,%rbx
+	vaesdec	%xmm0,%xmm8,%xmm8
+	vmovdqu	16(%r15),%xmm13
+	movq	%rbx,64+56(%rsp)
+	vaesdec	%xmm0,%xmm9,%xmm9
+	vmovups	40(%rsi),%xmm0
+	leaq	16(%r15,%rbx,1),%r15
+	vmovdqu	32(%rsp),%xmm14
+	prefetcht0	15(%r14)
+	prefetcht0	15(%r15)
+	cmpl	$11,%eax
+	jb	.Ldec8x_tail
+
+	vaesdec	%xmm1,%xmm2,%xmm2
+	vaesdec	%xmm1,%xmm3,%xmm3
+	vaesdec	%xmm1,%xmm4,%xmm4
+	vaesdec	%xmm1,%xmm5,%xmm5
+	vaesdec	%xmm1,%xmm6,%xmm6
+	vaesdec	%xmm1,%xmm7,%xmm7
+	vaesdec	%xmm1,%xmm8,%xmm8
+	vaesdec	%xmm1,%xmm9,%xmm9
+	vmovups	176-120(%rsi),%xmm1
+
+	vaesdec	%xmm0,%xmm2,%xmm2
+	vaesdec	%xmm0,%xmm3,%xmm3
+	vaesdec	%xmm0,%xmm4,%xmm4
+	vaesdec	%xmm0,%xmm5,%xmm5
+	vaesdec	%xmm0,%xmm6,%xmm6
+	vaesdec	%xmm0,%xmm7,%xmm7
+	vaesdec	%xmm0,%xmm8,%xmm8
+	vaesdec	%xmm0,%xmm9,%xmm9
+	vmovups	192-120(%rsi),%xmm0
+	je	.Ldec8x_tail
+
+	vaesdec	%xmm1,%xmm2,%xmm2
+	vaesdec	%xmm1,%xmm3,%xmm3
+	vaesdec	%xmm1,%xmm4,%xmm4
+	vaesdec	%xmm1,%xmm5,%xmm5
+	vaesdec	%xmm1,%xmm6,%xmm6
+	vaesdec	%xmm1,%xmm7,%xmm7
+	vaesdec	%xmm1,%xmm8,%xmm8
+	vaesdec	%xmm1,%xmm9,%xmm9
+	vmovups	208-120(%rsi),%xmm1
+
+	vaesdec	%xmm0,%xmm2,%xmm2
+	vaesdec	%xmm0,%xmm3,%xmm3
+	vaesdec	%xmm0,%xmm4,%xmm4
+	vaesdec	%xmm0,%xmm5,%xmm5
+	vaesdec	%xmm0,%xmm6,%xmm6
+	vaesdec	%xmm0,%xmm7,%xmm7
+	vaesdec	%xmm0,%xmm8,%xmm8
+	vaesdec	%xmm0,%xmm9,%xmm9
+	vmovups	224-120(%rsi),%xmm0
+
+.Ldec8x_tail:
+	vaesdec	%xmm1,%xmm2,%xmm2
+	vpxor	%xmm15,%xmm15,%xmm15
+	vaesdec	%xmm1,%xmm3,%xmm3
+	vaesdec	%xmm1,%xmm4,%xmm4
+	vpcmpgtd	%xmm15,%xmm14,%xmm15
+	vaesdec	%xmm1,%xmm5,%xmm5
+	vaesdec	%xmm1,%xmm6,%xmm6
+	vpaddd	%xmm14,%xmm15,%xmm15
+	vmovdqu	48(%rsp),%xmm14
+	vaesdec	%xmm1,%xmm7,%xmm7
+	movq	64(%rsp),%rbx
+	vaesdec	%xmm1,%xmm8,%xmm8
+	vaesdec	%xmm1,%xmm9,%xmm9
+	vmovups	16-120(%rsi),%xmm1
+
+	vaesdeclast	%xmm0,%xmm2,%xmm2
+	vmovdqa	%xmm15,32(%rsp)
+	vpxor	%xmm15,%xmm15,%xmm15
+	vaesdeclast	%xmm0,%xmm3,%xmm3
+	vpxor	0(%rbp),%xmm2,%xmm2
+	vaesdeclast	%xmm0,%xmm4,%xmm4
+	vpxor	16(%rbp),%xmm3,%xmm3
+	vpcmpgtd	%xmm15,%xmm14,%xmm15
+	vaesdeclast	%xmm0,%xmm5,%xmm5
+	vpxor	32(%rbp),%xmm4,%xmm4
+	vaesdeclast	%xmm0,%xmm6,%xmm6
+	vpxor	48(%rbp),%xmm5,%xmm5
+	vpaddd	%xmm15,%xmm14,%xmm14
+	vmovdqu	-120(%rsi),%xmm15
+	vaesdeclast	%xmm0,%xmm7,%xmm7
+	vpxor	64(%rbp),%xmm6,%xmm6
+	vaesdeclast	%xmm0,%xmm8,%xmm8
+	vpxor	80(%rbp),%xmm7,%xmm7
+	vmovdqa	%xmm14,48(%rsp)
+	vaesdeclast	%xmm0,%xmm9,%xmm9
+	vpxor	96(%rbp),%xmm8,%xmm8
+	vmovups	32-120(%rsi),%xmm0
+
+	vmovups	%xmm2,-16(%r8)
+	subq	%rbx,%r8
+	vmovdqu	128+0(%rsp),%xmm2
+	vpxor	112(%rbp),%xmm9,%xmm9
+	vmovups	%xmm3,-16(%r9)
+	subq	72(%rsp),%r9
+	vmovdqu	%xmm2,0(%rbp)
+	vpxor	%xmm15,%xmm2,%xmm2
+	vmovdqu	128+16(%rsp),%xmm3
+	vmovups	%xmm4,-16(%r10)
+	subq	80(%rsp),%r10
+	vmovdqu	%xmm3,16(%rbp)
+	vpxor	%xmm15,%xmm3,%xmm3
+	vmovdqu	128+32(%rsp),%xmm4
+	vmovups	%xmm5,-16(%r11)
+	subq	88(%rsp),%r11
+	vmovdqu	%xmm4,32(%rbp)
+	vpxor	%xmm15,%xmm4,%xmm4
+	vmovdqu	128+48(%rsp),%xmm5
+	vmovups	%xmm6,-16(%r12)
+	subq	96(%rsp),%r12
+	vmovdqu	%xmm5,48(%rbp)
+	vpxor	%xmm15,%xmm5,%xmm5
+	vmovdqu	%xmm10,64(%rbp)
+	vpxor	%xmm10,%xmm15,%xmm6
+	vmovups	%xmm7,-16(%r13)
+	subq	104(%rsp),%r13
+	vmovdqu	%xmm11,80(%rbp)
+	vpxor	%xmm11,%xmm15,%xmm7
+	vmovups	%xmm8,-16(%r14)
+	subq	112(%rsp),%r14
+	vmovdqu	%xmm12,96(%rbp)
+	vpxor	%xmm12,%xmm15,%xmm8
+	vmovups	%xmm9,-16(%r15)
+	subq	120(%rsp),%r15
+	vmovdqu	%xmm13,112(%rbp)
+	vpxor	%xmm13,%xmm15,%xmm9
+
+	xorq	$128,%rbp
+	decl	%edx
+	jnz	.Loop_dec8x
+
+	movq	16(%rsp),%rax
+
+
+
+
+
+.Ldec8x_done:
+	vzeroupper
+	movq	-48(%rax),%r15
+	movq	-40(%rax),%r14
+	movq	-32(%rax),%r13
+	movq	-24(%rax),%r12
+	movq	-16(%rax),%rbp
+	movq	-8(%rax),%rbx
+	leaq	(%rax),%rsp
+.Ldec8x_epilogue:
+	.byte	0xf3,0xc3
+.size	aesni_multi_cbc_decrypt_avx,.-aesni_multi_cbc_decrypt_avx

deps/openssl/asm/x64-elf-gas/aes/aesni-x86_64.s

@@ -503,7 +503,7 @@ aesni_ecb_encrypt:
 	testl	%r8d,%r8d
 	jz	.Lecb_decrypt
 
-	cmpq	$128,%rdx
+	cmpq	$0x80,%rdx
 	jb	.Lecb_enc_tail
 
 	movdqu	(%rdi),%xmm2
@@ -515,7 +515,7 @@ aesni_ecb_encrypt:
 	movdqu	96(%rdi),%xmm8
 	movdqu	112(%rdi),%xmm9
 	leaq	128(%rdi),%rdi
-	subq	$128,%rdx
+	subq	$0x80,%rdx
 	jmp	.Lecb_enc_loop8_enter
 .align	16
 .Lecb_enc_loop8:
@@ -543,7 +543,7 @@ aesni_ecb_encrypt:
 
 	call	_aesni_encrypt8
 
-	subq	$128,%rdx
+	subq	$0x80,%rdx
 	jnc	.Lecb_enc_loop8
 
 	movups	%xmm2,(%rsi)
@@ -557,22 +557,22 @@ aesni_ecb_encrypt:
 	movups	%xmm8,96(%rsi)
 	movups	%xmm9,112(%rsi)
 	leaq	128(%rsi),%rsi
-	addq	$128,%rdx
+	addq	$0x80,%rdx
 	jz	.Lecb_ret
 
 .Lecb_enc_tail:
 	movups	(%rdi),%xmm2
-	cmpq	$32,%rdx
+	cmpq	$0x20,%rdx
 	jb	.Lecb_enc_one
 	movups	16(%rdi),%xmm3
 	je	.Lecb_enc_two
 	movups	32(%rdi),%xmm4
-	cmpq	$64,%rdx
+	cmpq	$0x40,%rdx
 	jb	.Lecb_enc_three
 	movups	48(%rdi),%xmm5
 	je	.Lecb_enc_four
 	movups	64(%rdi),%xmm6
-	cmpq	$96,%rdx
+	cmpq	$0x60,%rdx
 	jb	.Lecb_enc_five
 	movups	80(%rdi),%xmm7
 	je	.Lecb_enc_six
@@ -646,7 +646,7 @@ aesni_ecb_encrypt:
 
 .align	16
 .Lecb_decrypt:
-	cmpq	$128,%rdx
+	cmpq	$0x80,%rdx
 	jb	.Lecb_dec_tail
 
 	movdqu	(%rdi),%xmm2
@@ -658,7 +658,7 @@ aesni_ecb_encrypt:
 	movdqu	96(%rdi),%xmm8
 	movdqu	112(%rdi),%xmm9
 	leaq	128(%rdi),%rdi
-	subq	$128,%rdx
+	subq	$0x80,%rdx
 	jmp	.Lecb_dec_loop8_enter
 .align	16
 .Lecb_dec_loop8:
@@ -687,7 +687,7 @@ aesni_ecb_encrypt:
 	call	_aesni_decrypt8
 
 	movups	(%r11),%xmm0
-	subq	$128,%rdx
+	subq	$0x80,%rdx
 	jnc	.Lecb_dec_loop8
 
 	movups	%xmm2,(%rsi)
@@ -709,22 +709,22 @@ aesni_ecb_encrypt:
 	movups	%xmm9,112(%rsi)
 	pxor	%xmm9,%xmm9
 	leaq	128(%rsi),%rsi
-	addq	$128,%rdx
+	addq	$0x80,%rdx
 	jz	.Lecb_ret
 
 .Lecb_dec_tail:
 	movups	(%rdi),%xmm2
-	cmpq	$32,%rdx
+	cmpq	$0x20,%rdx
 	jb	.Lecb_dec_one
 	movups	16(%rdi),%xmm3
 	je	.Lecb_dec_two
 	movups	32(%rdi),%xmm4
-	cmpq	$64,%rdx
+	cmpq	$0x40,%rdx
 	jb	.Lecb_dec_three
 	movups	48(%rdi),%xmm5
 	je	.Lecb_dec_four
 	movups	64(%rdi),%xmm6
-	cmpq	$96,%rdx
+	cmpq	$0x60,%rdx
 	jb	.Lecb_dec_five
 	movups	80(%rdi),%xmm7
 	je	.Lecb_dec_six
@@ -1598,7 +1598,7 @@ aesni_xts_encrypt:
 
 	movdqa	.Lxts_magic(%rip),%xmm8
 	movdqa	%xmm2,%xmm15
-	pshufd	$95,%xmm2,%xmm9
+	pshufd	$0x5f,%xmm2,%xmm9
 	pxor	%xmm0,%xmm1
 	movdqa	%xmm9,%xmm14
 	paddd	%xmm9,%xmm9
@@ -1697,7 +1697,7 @@ aesni_xts_encrypt:
 .byte	102,15,56,220,248
 	movups	64(%r11),%xmm0
 	movdqa	%xmm8,80(%rsp)
-	pshufd	$95,%xmm15,%xmm9
+	pshufd	$0x5f,%xmm15,%xmm9
 	jmp	.Lxts_enc_loop6
 .align	32
 .Lxts_enc_loop6:
@@ -1836,13 +1836,13 @@ aesni_xts_encrypt:
 	jz	.Lxts_enc_done
 
 	pxor	%xmm0,%xmm11
-	cmpq	$32,%rdx
+	cmpq	$0x20,%rdx
 	jb	.Lxts_enc_one
 	pxor	%xmm0,%xmm12
 	je	.Lxts_enc_two
 
 	pxor	%xmm0,%xmm13
-	cmpq	$64,%rdx
+	cmpq	$0x40,%rdx
 	jb	.Lxts_enc_three
 	pxor	%xmm0,%xmm14
 	je	.Lxts_enc_four
@@ -2069,7 +2069,7 @@ aesni_xts_decrypt:
 
 	movdqa	.Lxts_magic(%rip),%xmm8
 	movdqa	%xmm2,%xmm15
-	pshufd	$95,%xmm2,%xmm9
+	pshufd	$0x5f,%xmm2,%xmm9
 	pxor	%xmm0,%xmm1
 	movdqa	%xmm9,%xmm14
 	paddd	%xmm9,%xmm9
@@ -2168,7 +2168,7 @@ aesni_xts_decrypt:
 .byte	102,15,56,222,248
 	movups	64(%r11),%xmm0
 	movdqa	%xmm8,80(%rsp)
-	pshufd	$95,%xmm15,%xmm9
+	pshufd	$0x5f,%xmm15,%xmm9
 	jmp	.Lxts_dec_loop6
 .align	32
 .Lxts_dec_loop6:
@@ -2308,13 +2308,13 @@ aesni_xts_decrypt:
 	jz	.Lxts_dec_done
 
 	pxor	%xmm0,%xmm12
-	cmpq	$32,%rdx
+	cmpq	$0x20,%rdx
 	jb	.Lxts_dec_one
 	pxor	%xmm0,%xmm13
 	je	.Lxts_dec_two
 
 	pxor	%xmm0,%xmm14
-	cmpq	$64,%rdx
+	cmpq	$0x40,%rdx
 	jb	.Lxts_dec_three
 	je	.Lxts_dec_four
 
@@ -2345,7 +2345,7 @@ aesni_xts_decrypt:
 	pcmpgtd	%xmm15,%xmm14
 	movdqu	%xmm6,64(%rsi)
 	leaq	80(%rsi),%rsi
-	pshufd	$19,%xmm14,%xmm11
+	pshufd	$0x13,%xmm14,%xmm11
 	andq	$15,%r9
 	jz	.Lxts_dec_ret
 
@@ -2634,7 +2634,7 @@ aesni_cbc_encrypt:
 	leaq	-8(%rax),%rbp
 	movups	(%r8),%xmm10
 	movl	%r10d,%eax
-	cmpq	$80,%rdx
+	cmpq	$0x50,%rdx
 	jbe	.Lcbc_dec_tail
 
 	movups	(%rcx),%xmm0
@@ -2650,14 +2650,14 @@ aesni_cbc_encrypt:
 	movdqu	80(%rdi),%xmm7
 	movdqa	%xmm6,%xmm15
 	movl	OPENSSL_ia32cap_P+4(%rip),%r9d
-	cmpq	$112,%rdx
+	cmpq	$0x70,%rdx
 	jbe	.Lcbc_dec_six_or_seven
 
 	andl	$71303168,%r9d
-	subq	$80,%rdx
+	subq	$0x50,%rdx
 	cmpl	$4194304,%r9d
 	je	.Lcbc_dec_loop6_enter
-	subq	$32,%rdx
+	subq	$0x20,%rdx
 	leaq	112(%rcx),%rcx
 	jmp	.Lcbc_dec_loop8_enter
 .align	16
@@ -2672,7 +2672,7 @@ aesni_cbc_encrypt:
 	movups	16-112(%rcx),%xmm1
 	pxor	%xmm0,%xmm4
 	xorq	%r11,%r11
-	cmpq	$112,%rdx
+	cmpq	$0x70,%rdx
 	pxor	%xmm0,%xmm5
 	pxor	%xmm0,%xmm6
 	pxor	%xmm0,%xmm7
@@ -2857,21 +2857,21 @@ aesni_cbc_encrypt:
 	movups	%xmm8,96(%rsi)
 	leaq	112(%rsi),%rsi
 
-	subq	$128,%rdx
+	subq	$0x80,%rdx
 	ja	.Lcbc_dec_loop8
 
 	movaps	%xmm9,%xmm2
 	leaq	-112(%rcx),%rcx
-	addq	$112,%rdx
+	addq	$0x70,%rdx
 	jle	.Lcbc_dec_clear_tail_collected
 	movups	%xmm9,(%rsi)
 	leaq	16(%rsi),%rsi
-	cmpq	$80,%rdx
+	cmpq	$0x50,%rdx
 	jbe	.Lcbc_dec_tail
 
 	movaps	%xmm11,%xmm2
 .Lcbc_dec_six_or_seven:
-	cmpq	$96,%rdx
+	cmpq	$0x60,%rdx
 	ja	.Lcbc_dec_seven
 
 	movaps	%xmm7,%xmm8
@@ -2964,33 +2964,33 @@ aesni_cbc_encrypt:
 	movl	%r10d,%eax
 	movdqu	%xmm6,64(%rsi)
 	leaq	80(%rsi),%rsi
-	subq	$96,%rdx
+	subq	$0x60,%rdx
 	ja	.Lcbc_dec_loop6
 
 	movdqa	%xmm7,%xmm2
-	addq	$80,%rdx
+	addq	$0x50,%rdx
 	jle	.Lcbc_dec_clear_tail_collected
 	movups	%xmm7,(%rsi)
 	leaq	16(%rsi),%rsi
 
 .Lcbc_dec_tail:
 	movups	(%rdi),%xmm2
-	subq	$16,%rdx
+	subq	$0x10,%rdx
 	jbe	.Lcbc_dec_one
 
 	movups	16(%rdi),%xmm3
 	movaps	%xmm2,%xmm11
-	subq	$16,%rdx
+	subq	$0x10,%rdx
 	jbe	.Lcbc_dec_two
 
 	movups	32(%rdi),%xmm4
 	movaps	%xmm3,%xmm12
-	subq	$16,%rdx
+	subq	$0x10,%rdx
 	jbe	.Lcbc_dec_three
 
 	movups	48(%rdi),%xmm5
 	movaps	%xmm4,%xmm13
-	subq	$16,%rdx
+	subq	$0x10,%rdx
 	jbe	.Lcbc_dec_four
 
 	movups	64(%rdi),%xmm6
@@ -3015,7 +3015,7 @@ aesni_cbc_encrypt:
 	movdqa	%xmm6,%xmm2
 	pxor	%xmm6,%xmm6
 	pxor	%xmm7,%xmm7
-	subq	$16,%rdx
+	subq	$0x10,%rdx
 	jmp	.Lcbc_dec_tail_collected
 
 .align	16
@@ -3332,7 +3332,7 @@ __aesni_set_encrypt_key:
 	pslldq	$4,%xmm0
 	pxor	%xmm3,%xmm0
 
-	pshufd	$255,%xmm0,%xmm3
+	pshufd	$0xff,%xmm0,%xmm3
 	pxor	%xmm1,%xmm3
 	pslldq	$4,%xmm1
 	pxor	%xmm1,%xmm3
@@ -3419,7 +3419,7 @@ __aesni_set_encrypt_key:
 	decl	%r10d
 	jz	.Ldone_key256
 
-	pshufd	$255,%xmm0,%xmm2
+	pshufd	$0xff,%xmm0,%xmm2
 	pxor	%xmm3,%xmm3
 .byte	102,15,56,221,211
 

deps/openssl/asm/x64-elf-gas/aes/bsaes-x86_64.s

@@ -324,45 +324,45 @@ _bsaes_encrypt8_bitslice:
 	pxor	%xmm2,%xmm5
 	decl	%r10d
 	jl	.Lenc_done
-	pshufd	$147,%xmm15,%xmm7
-	pshufd	$147,%xmm0,%xmm8
+	pshufd	$0x93,%xmm15,%xmm7
+	pshufd	$0x93,%xmm0,%xmm8
 	pxor	%xmm7,%xmm15
-	pshufd	$147,%xmm3,%xmm9
+	pshufd	$0x93,%xmm3,%xmm9
 	pxor	%xmm8,%xmm0
-	pshufd	$147,%xmm5,%xmm10
+	pshufd	$0x93,%xmm5,%xmm10
 	pxor	%xmm9,%xmm3
-	pshufd	$147,%xmm2,%xmm11
+	pshufd	$0x93,%xmm2,%xmm11
 	pxor	%xmm10,%xmm5
-	pshufd	$147,%xmm6,%xmm12
+	pshufd	$0x93,%xmm6,%xmm12
 	pxor	%xmm11,%xmm2
-	pshufd	$147,%xmm1,%xmm13
+	pshufd	$0x93,%xmm1,%xmm13
 	pxor	%xmm12,%xmm6
-	pshufd	$147,%xmm4,%xmm14
+	pshufd	$0x93,%xmm4,%xmm14
 	pxor	%xmm13,%xmm1
 	pxor	%xmm14,%xmm4
 
 	pxor	%xmm15,%xmm8
 	pxor	%xmm4,%xmm7
 	pxor	%xmm4,%xmm8
-	pshufd	$78,%xmm15,%xmm15
+	pshufd	$0x4E,%xmm15,%xmm15
 	pxor	%xmm0,%xmm9
-	pshufd	$78,%xmm0,%xmm0
+	pshufd	$0x4E,%xmm0,%xmm0
 	pxor	%xmm2,%xmm12
 	pxor	%xmm7,%xmm15
 	pxor	%xmm6,%xmm13
 	pxor	%xmm8,%xmm0
 	pxor	%xmm5,%xmm11
-	pshufd	$78,%xmm2,%xmm7
+	pshufd	$0x4E,%xmm2,%xmm7
 	pxor	%xmm1,%xmm14
-	pshufd	$78,%xmm6,%xmm8
+	pshufd	$0x4E,%xmm6,%xmm8
 	pxor	%xmm3,%xmm10
-	pshufd	$78,%xmm5,%xmm2
+	pshufd	$0x4E,%xmm5,%xmm2
 	pxor	%xmm4,%xmm10
-	pshufd	$78,%xmm4,%xmm6
+	pshufd	$0x4E,%xmm4,%xmm6
 	pxor	%xmm4,%xmm11
-	pshufd	$78,%xmm1,%xmm5
+	pshufd	$0x4E,%xmm1,%xmm5
 	pxor	%xmm11,%xmm7
-	pshufd	$78,%xmm3,%xmm1
+	pshufd	$0x4E,%xmm3,%xmm1
 	pxor	%xmm12,%xmm8
 	pxor	%xmm10,%xmm2
 	pxor	%xmm14,%xmm6
@@ -796,24 +796,24 @@ _bsaes_decrypt8:
 	decl	%r10d
 	jl	.Ldec_done
 
-	pshufd	$78,%xmm15,%xmm7
-	pshufd	$78,%xmm2,%xmm13
+	pshufd	$0x4E,%xmm15,%xmm7
+	pshufd	$0x4E,%xmm2,%xmm13
 	pxor	%xmm15,%xmm7
-	pshufd	$78,%xmm4,%xmm14
+	pshufd	$0x4E,%xmm4,%xmm14
 	pxor	%xmm2,%xmm13
-	pshufd	$78,%xmm0,%xmm8
+	pshufd	$0x4E,%xmm0,%xmm8
 	pxor	%xmm4,%xmm14
-	pshufd	$78,%xmm5,%xmm9
+	pshufd	$0x4E,%xmm5,%xmm9
 	pxor	%xmm0,%xmm8
-	pshufd	$78,%xmm3,%xmm10
+	pshufd	$0x4E,%xmm3,%xmm10
 	pxor	%xmm5,%xmm9
 	pxor	%xmm13,%xmm15
 	pxor	%xmm13,%xmm0
-	pshufd	$78,%xmm1,%xmm11
+	pshufd	$0x4E,%xmm1,%xmm11
 	pxor	%xmm3,%xmm10
 	pxor	%xmm7,%xmm5
 	pxor	%xmm8,%xmm3
-	pshufd	$78,%xmm6,%xmm12
+	pshufd	$0x4E,%xmm6,%xmm12
 	pxor	%xmm1,%xmm11
 	pxor	%xmm14,%xmm0
 	pxor	%xmm9,%xmm1
@@ -827,45 +827,45 @@ _bsaes_decrypt8:
 	pxor	%xmm14,%xmm1
 	pxor	%xmm14,%xmm6
 	pxor	%xmm12,%xmm4
-	pshufd	$147,%xmm15,%xmm7
-	pshufd	$147,%xmm0,%xmm8
+	pshufd	$0x93,%xmm15,%xmm7
+	pshufd	$0x93,%xmm0,%xmm8
 	pxor	%xmm7,%xmm15
-	pshufd	$147,%xmm5,%xmm9
+	pshufd	$0x93,%xmm5,%xmm9
 	pxor	%xmm8,%xmm0
-	pshufd	$147,%xmm3,%xmm10
+	pshufd	$0x93,%xmm3,%xmm10
 	pxor	%xmm9,%xmm5
-	pshufd	$147,%xmm1,%xmm11
+	pshufd	$0x93,%xmm1,%xmm11
 	pxor	%xmm10,%xmm3
-	pshufd	$147,%xmm6,%xmm12
+	pshufd	$0x93,%xmm6,%xmm12
 	pxor	%xmm11,%xmm1
-	pshufd	$147,%xmm2,%xmm13
+	pshufd	$0x93,%xmm2,%xmm13
 	pxor	%xmm12,%xmm6
-	pshufd	$147,%xmm4,%xmm14
+	pshufd	$0x93,%xmm4,%xmm14
 	pxor	%xmm13,%xmm2
 	pxor	%xmm14,%xmm4
 
 	pxor	%xmm15,%xmm8
 	pxor	%xmm4,%xmm7
 	pxor	%xmm4,%xmm8
-	pshufd	$78,%xmm15,%xmm15
+	pshufd	$0x4E,%xmm15,%xmm15
 	pxor	%xmm0,%xmm9
-	pshufd	$78,%xmm0,%xmm0
+	pshufd	$0x4E,%xmm0,%xmm0
 	pxor	%xmm1,%xmm12
 	pxor	%xmm7,%xmm15
 	pxor	%xmm6,%xmm13
 	pxor	%xmm8,%xmm0
 	pxor	%xmm3,%xmm11
-	pshufd	$78,%xmm1,%xmm7
+	pshufd	$0x4E,%xmm1,%xmm7
 	pxor	%xmm2,%xmm14
-	pshufd	$78,%xmm6,%xmm8
+	pshufd	$0x4E,%xmm6,%xmm8
 	pxor	%xmm5,%xmm10
-	pshufd	$78,%xmm3,%xmm1
+	pshufd	$0x4E,%xmm3,%xmm1
 	pxor	%xmm4,%xmm10
-	pshufd	$78,%xmm4,%xmm6
+	pshufd	$0x4E,%xmm4,%xmm6
 	pxor	%xmm4,%xmm11
-	pshufd	$78,%xmm2,%xmm3
+	pshufd	$0x4E,%xmm2,%xmm3
 	pxor	%xmm11,%xmm7
-	pshufd	$78,%xmm5,%xmm2
+	pshufd	$0x4E,%xmm5,%xmm2
 	pxor	%xmm12,%xmm8
 	pxor	%xmm1,%xmm10
 	pxor	%xmm14,%xmm6
@@ -1552,20 +1552,20 @@ bsaes_xts_encrypt:
 	movdqa	%xmm7,(%rax)
 
 	andq	$-16,%r14
-	subq	$128,%rsp
+	subq	$0x80,%rsp
 	movdqa	32(%rbp),%xmm6
 
 	pxor	%xmm14,%xmm14
 	movdqa	.Lxts_magic(%rip),%xmm12
 	pcmpgtd	%xmm6,%xmm14
 
-	subq	$128,%r14
+	subq	$0x80,%r14
 	jc	.Lxts_enc_short
 	jmp	.Lxts_enc_loop
 
 .align	16
 .Lxts_enc_loop:
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm15
 	movdqa	%xmm6,0(%rsp)
@@ -1573,7 +1573,7 @@ bsaes_xts_encrypt:
 	pand	%xmm12,%xmm13
 	pcmpgtd	%xmm6,%xmm14
 	pxor	%xmm13,%xmm6
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm0
 	movdqa	%xmm6,16(%rsp)
@@ -1582,7 +1582,7 @@ bsaes_xts_encrypt:
 	pcmpgtd	%xmm6,%xmm14
 	pxor	%xmm13,%xmm6
 	movdqu	0(%r12),%xmm7
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm1
 	movdqa	%xmm6,32(%rsp)
@@ -1592,7 +1592,7 @@ bsaes_xts_encrypt:
 	pxor	%xmm13,%xmm6
 	movdqu	16(%r12),%xmm8
 	pxor	%xmm7,%xmm15
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm2
 	movdqa	%xmm6,48(%rsp)
@@ -1602,7 +1602,7 @@ bsaes_xts_encrypt:
 	pxor	%xmm13,%xmm6
 	movdqu	32(%r12),%xmm9
 	pxor	%xmm8,%xmm0
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm3
 	movdqa	%xmm6,64(%rsp)
@@ -1612,7 +1612,7 @@ bsaes_xts_encrypt:
 	pxor	%xmm13,%xmm6
 	movdqu	48(%r12),%xmm10
 	pxor	%xmm9,%xmm1
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm4
 	movdqa	%xmm6,80(%rsp)
@@ -1622,7 +1622,7 @@ bsaes_xts_encrypt:
 	pxor	%xmm13,%xmm6
 	movdqu	64(%r12),%xmm11
 	pxor	%xmm10,%xmm2
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm5
 	movdqa	%xmm6,96(%rsp)
@@ -1666,20 +1666,20 @@ bsaes_xts_encrypt:
 	pxor	%xmm14,%xmm14
 	movdqa	.Lxts_magic(%rip),%xmm12
 	pcmpgtd	%xmm6,%xmm14
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	paddq	%xmm6,%xmm6
 	pand	%xmm12,%xmm13
 	pcmpgtd	%xmm6,%xmm14
 	pxor	%xmm13,%xmm6
 
-	subq	$128,%r14
+	subq	$0x80,%r14
 	jnc	.Lxts_enc_loop
 
 .Lxts_enc_short:
-	addq	$128,%r14
+	addq	$0x80,%r14
 	jz	.Lxts_enc_done
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm15
 	movdqa	%xmm6,0(%rsp)
@@ -1687,7 +1687,7 @@ bsaes_xts_encrypt:
 	pand	%xmm12,%xmm13
 	pcmpgtd	%xmm6,%xmm14
 	pxor	%xmm13,%xmm6
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm0
 	movdqa	%xmm6,16(%rsp)
@@ -1698,7 +1698,7 @@ bsaes_xts_encrypt:
 	movdqu	0(%r12),%xmm7
 	cmpq	$16,%r14
 	je	.Lxts_enc_1
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm1
 	movdqa	%xmm6,32(%rsp)
@@ -1710,7 +1710,7 @@ bsaes_xts_encrypt:
 	cmpq	$32,%r14
 	je	.Lxts_enc_2
 	pxor	%xmm7,%xmm15
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm2
 	movdqa	%xmm6,48(%rsp)
@@ -1722,7 +1722,7 @@ bsaes_xts_encrypt:
 	cmpq	$48,%r14
 	je	.Lxts_enc_3
 	pxor	%xmm8,%xmm0
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm3
 	movdqa	%xmm6,64(%rsp)
@@ -1734,7 +1734,7 @@ bsaes_xts_encrypt:
 	cmpq	$64,%r14
 	je	.Lxts_enc_4
 	pxor	%xmm9,%xmm1
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm4
 	movdqa	%xmm6,80(%rsp)
@@ -1746,7 +1746,7 @@ bsaes_xts_encrypt:
 	cmpq	$80,%r14
 	je	.Lxts_enc_5
 	pxor	%xmm10,%xmm2
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm5
 	movdqa	%xmm6,96(%rsp)
@@ -2011,20 +2011,20 @@ bsaes_xts_decrypt:
 	shlq	$4,%rax
 	subq	%rax,%r14
 
-	subq	$128,%rsp
+	subq	$0x80,%rsp
 	movdqa	32(%rbp),%xmm6
 
 	pxor	%xmm14,%xmm14
 	movdqa	.Lxts_magic(%rip),%xmm12
 	pcmpgtd	%xmm6,%xmm14
 
-	subq	$128,%r14
+	subq	$0x80,%r14
 	jc	.Lxts_dec_short
 	jmp	.Lxts_dec_loop
 
 .align	16
 .Lxts_dec_loop:
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm15
 	movdqa	%xmm6,0(%rsp)
@@ -2032,7 +2032,7 @@ bsaes_xts_decrypt:
 	pand	%xmm12,%xmm13
 	pcmpgtd	%xmm6,%xmm14
 	pxor	%xmm13,%xmm6
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm0
 	movdqa	%xmm6,16(%rsp)
@@ -2041,7 +2041,7 @@ bsaes_xts_decrypt:
 	pcmpgtd	%xmm6,%xmm14
 	pxor	%xmm13,%xmm6
 	movdqu	0(%r12),%xmm7
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm1
 	movdqa	%xmm6,32(%rsp)
@@ -2051,7 +2051,7 @@ bsaes_xts_decrypt:
 	pxor	%xmm13,%xmm6
 	movdqu	16(%r12),%xmm8
 	pxor	%xmm7,%xmm15
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm2
 	movdqa	%xmm6,48(%rsp)
@@ -2061,7 +2061,7 @@ bsaes_xts_decrypt:
 	pxor	%xmm13,%xmm6
 	movdqu	32(%r12),%xmm9
 	pxor	%xmm8,%xmm0
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm3
 	movdqa	%xmm6,64(%rsp)
@@ -2071,7 +2071,7 @@ bsaes_xts_decrypt:
 	pxor	%xmm13,%xmm6
 	movdqu	48(%r12),%xmm10
 	pxor	%xmm9,%xmm1
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm4
 	movdqa	%xmm6,80(%rsp)
@@ -2081,7 +2081,7 @@ bsaes_xts_decrypt:
 	pxor	%xmm13,%xmm6
 	movdqu	64(%r12),%xmm11
 	pxor	%xmm10,%xmm2
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm5
 	movdqa	%xmm6,96(%rsp)
@@ -2125,20 +2125,20 @@ bsaes_xts_decrypt:
 	pxor	%xmm14,%xmm14
 	movdqa	.Lxts_magic(%rip),%xmm12
 	pcmpgtd	%xmm6,%xmm14
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	paddq	%xmm6,%xmm6
 	pand	%xmm12,%xmm13
 	pcmpgtd	%xmm6,%xmm14
 	pxor	%xmm13,%xmm6
 
-	subq	$128,%r14
+	subq	$0x80,%r14
 	jnc	.Lxts_dec_loop
 
 .Lxts_dec_short:
-	addq	$128,%r14
+	addq	$0x80,%r14
 	jz	.Lxts_dec_done
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm15
 	movdqa	%xmm6,0(%rsp)
@@ -2146,7 +2146,7 @@ bsaes_xts_decrypt:
 	pand	%xmm12,%xmm13
 	pcmpgtd	%xmm6,%xmm14
 	pxor	%xmm13,%xmm6
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm0
 	movdqa	%xmm6,16(%rsp)
@@ -2157,7 +2157,7 @@ bsaes_xts_decrypt:
 	movdqu	0(%r12),%xmm7
 	cmpq	$16,%r14
 	je	.Lxts_dec_1
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm1
 	movdqa	%xmm6,32(%rsp)
@@ -2169,7 +2169,7 @@ bsaes_xts_decrypt:
 	cmpq	$32,%r14
 	je	.Lxts_dec_2
 	pxor	%xmm7,%xmm15
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm2
 	movdqa	%xmm6,48(%rsp)
@@ -2181,7 +2181,7 @@ bsaes_xts_decrypt:
 	cmpq	$48,%r14
 	je	.Lxts_dec_3
 	pxor	%xmm8,%xmm0
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm3
 	movdqa	%xmm6,64(%rsp)
@@ -2193,7 +2193,7 @@ bsaes_xts_decrypt:
 	cmpq	$64,%r14
 	je	.Lxts_dec_4
 	pxor	%xmm9,%xmm1
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm4
 	movdqa	%xmm6,80(%rsp)
@@ -2205,7 +2205,7 @@ bsaes_xts_decrypt:
 	cmpq	$80,%r14
 	je	.Lxts_dec_5
 	pxor	%xmm10,%xmm2
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm5
 	movdqa	%xmm6,96(%rsp)
@@ -2382,7 +2382,7 @@ bsaes_xts_decrypt:
 	pxor	%xmm14,%xmm14
 	movdqa	.Lxts_magic(%rip),%xmm12
 	pcmpgtd	%xmm6,%xmm14
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	movdqa	%xmm6,%xmm5
 	paddq	%xmm6,%xmm6
 	pand	%xmm12,%xmm13

deps/openssl/asm/x64-elf-gas/aes/vpaes-x86_64.s

@@ -60,7 +60,7 @@ _vpaes_encrypt_core:
 	addq	$16,%r11
 	pxor	%xmm0,%xmm3
 .byte	102,15,56,0,193
-	andq	$48,%r11
+	andq	$0x30,%r11
 	subq	$1,%rax
 	pxor	%xmm3,%xmm0
 
@@ -120,10 +120,10 @@ _vpaes_decrypt_core:
 	pand	%xmm9,%xmm0
 .byte	102,15,56,0,208
 	movdqa	.Lk_dipt+16(%rip),%xmm0
-	xorq	$48,%r11
+	xorq	$0x30,%r11
 	leaq	.Lk_dsbd(%rip),%r10
 .byte	102,15,56,0,193
-	andq	$48,%r11
+	andq	$0x30,%r11
 	pxor	%xmm5,%xmm2
 	movdqa	.Lk_mc_forward+48(%rip),%xmm5
 	pxor	%xmm2,%xmm0
@@ -242,7 +242,7 @@ _vpaes_schedule_core:
 	movdqa	(%r8,%r10,1),%xmm1
 .byte	102,15,56,0,217
 	movdqu	%xmm3,(%rdx)
-	xorq	$48,%r8
+	xorq	$0x30,%r8
 
 .Lschedule_go:
 	cmpl	$192,%esi
@@ -332,7 +332,7 @@ _vpaes_schedule_core:
 	call	_vpaes_schedule_mangle
 
 
-	pshufd	$255,%xmm0,%xmm0
+	pshufd	$0xFF,%xmm0,%xmm0
 	movdqa	%xmm7,%xmm5
 	movdqa	%xmm6,%xmm7
 	call	_vpaes_schedule_low_round
@@ -399,8 +399,8 @@ _vpaes_schedule_core:
 .type	_vpaes_schedule_192_smear,@function
 .align	16
 _vpaes_schedule_192_smear:
-	pshufd	$128,%xmm6,%xmm1
-	pshufd	$254,%xmm7,%xmm0
+	pshufd	$0x80,%xmm6,%xmm1
+	pshufd	$0xFE,%xmm7,%xmm0
 	pxor	%xmm1,%xmm6
 	pxor	%xmm1,%xmm1
 	pxor	%xmm0,%xmm6
@@ -437,7 +437,7 @@ _vpaes_schedule_round:
 	pxor	%xmm1,%xmm7
 
 
-	pshufd	$255,%xmm0,%xmm0
+	pshufd	$0xFF,%xmm0,%xmm0
 .byte	102,15,58,15,192,1
 
 
@@ -596,7 +596,7 @@ _vpaes_schedule_mangle:
 	movdqa	(%r8,%r10,1),%xmm1
 .byte	102,15,56,0,217
 	addq	$-16,%r8
-	andq	$48,%r8
+	andq	$0x30,%r8
 	movdqu	%xmm3,(%rdx)
 	.byte	0xf3,0xc3
 .size	_vpaes_schedule_mangle,.-_vpaes_schedule_mangle
@@ -614,7 +614,7 @@ vpaes_set_encrypt_key:
 	movl	%eax,240(%rdx)
 
 	movl	$0,%ecx
-	movl	$48,%r8d
+	movl	$0x30,%r8d
 	call	_vpaes_schedule_core
 	xorl	%eax,%eax
 	.byte	0xf3,0xc3

deps/openssl/asm/x64-elf-gas/bn/rsaz-x86_64.s

@@ -19,6 +19,10 @@ rsaz_512_sqr:
 	movq	(%rsi),%rdx
 	movq	8(%rsi),%rax
 	movq	%rcx,128(%rsp)
+	movl	$0x80100,%r11d
+	andl	OPENSSL_ia32cap_P+8(%rip),%r11d
+	cmpl	$0x80100,%r11d
+	je	.Loop_sqrx
 	jmp	.Loop_sqr
 
 .align	32
@@ -382,6 +386,276 @@ rsaz_512_sqr:
 
 	decl	%r8d
 	jnz	.Loop_sqr
+	jmp	.Lsqr_tail
+
+.align	32
+.Loop_sqrx:
+	movl	%r8d,128+8(%rsp)
+.byte	102,72,15,110,199
+.byte	102,72,15,110,205
+
+	mulxq	%rax,%r8,%r9
+
+	mulxq	16(%rsi),%rcx,%r10
+	xorq	%rbp,%rbp
+
+	mulxq	24(%rsi),%rax,%r11
+	adcxq	%rcx,%r9
+
+	mulxq	32(%rsi),%rcx,%r12
+	adcxq	%rax,%r10
+
+	mulxq	40(%rsi),%rax,%r13
+	adcxq	%rcx,%r11
+
+.byte	0xc4,0x62,0xf3,0xf6,0xb6,0x30,0x00,0x00,0x00
+	adcxq	%rax,%r12
+	adcxq	%rcx,%r13
+
+.byte	0xc4,0x62,0xfb,0xf6,0xbe,0x38,0x00,0x00,0x00
+	adcxq	%rax,%r14
+	adcxq	%rbp,%r15
+
+	movq	%r9,%rcx
+	shldq	$1,%r8,%r9
+	shlq	$1,%r8
+
+	xorl	%ebp,%ebp
+	mulxq	%rdx,%rax,%rdx
+	adcxq	%rdx,%r8
+	movq	8(%rsi),%rdx
+	adcxq	%rbp,%r9
+
+	movq	%rax,(%rsp)
+	movq	%r8,8(%rsp)
+
+
+	mulxq	16(%rsi),%rax,%rbx
+	adoxq	%rax,%r10
+	adcxq	%rbx,%r11
+
+.byte	0xc4,0x62,0xc3,0xf6,0x86,0x18,0x00,0x00,0x00
+	adoxq	%rdi,%r11
+	adcxq	%r8,%r12
+
+	mulxq	32(%rsi),%rax,%rbx
+	adoxq	%rax,%r12
+	adcxq	%rbx,%r13
+
+	mulxq	40(%rsi),%rdi,%r8
+	adoxq	%rdi,%r13
+	adcxq	%r8,%r14
+
+.byte	0xc4,0xe2,0xfb,0xf6,0x9e,0x30,0x00,0x00,0x00
+	adoxq	%rax,%r14
+	adcxq	%rbx,%r15
+
+.byte	0xc4,0x62,0xc3,0xf6,0x86,0x38,0x00,0x00,0x00
+	adoxq	%rdi,%r15
+	adcxq	%rbp,%r8
+	adoxq	%rbp,%r8
+
+	movq	%r11,%rbx
+	shldq	$1,%r10,%r11
+	shldq	$1,%rcx,%r10
+
+	xorl	%ebp,%ebp
+	mulxq	%rdx,%rax,%rcx
+	movq	16(%rsi),%rdx
+	adcxq	%rax,%r9
+	adcxq	%rcx,%r10
+	adcxq	%rbp,%r11
+
+	movq	%r9,16(%rsp)
+.byte	0x4c,0x89,0x94,0x24,0x18,0x00,0x00,0x00
+
+
+.byte	0xc4,0x62,0xc3,0xf6,0x8e,0x18,0x00,0x00,0x00
+	adoxq	%rdi,%r12
+	adcxq	%r9,%r13
+
+	mulxq	32(%rsi),%rax,%rcx
+	adoxq	%rax,%r13
+	adcxq	%rcx,%r14
+
+	mulxq	40(%rsi),%rdi,%r9
+	adoxq	%rdi,%r14
+	adcxq	%r9,%r15
+
+.byte	0xc4,0xe2,0xfb,0xf6,0x8e,0x30,0x00,0x00,0x00
+	adoxq	%rax,%r15
+	adcxq	%rcx,%r8
+
+.byte	0xc4,0x62,0xc3,0xf6,0x8e,0x38,0x00,0x00,0x00
+	adoxq	%rdi,%r8
+	adcxq	%rbp,%r9
+	adoxq	%rbp,%r9
+
+	movq	%r13,%rcx
+	shldq	$1,%r12,%r13
+	shldq	$1,%rbx,%r12
+
+	xorl	%ebp,%ebp
+	mulxq	%rdx,%rax,%rdx
+	adcxq	%rax,%r11
+	adcxq	%rdx,%r12
+	movq	24(%rsi),%rdx
+	adcxq	%rbp,%r13
+
+	movq	%r11,32(%rsp)
+.byte	0x4c,0x89,0xa4,0x24,0x28,0x00,0x00,0x00
+
+
+.byte	0xc4,0xe2,0xfb,0xf6,0x9e,0x20,0x00,0x00,0x00
+	adoxq	%rax,%r14
+	adcxq	%rbx,%r15
+
+	mulxq	40(%rsi),%rdi,%r10
+	adoxq	%rdi,%r15
+	adcxq	%r10,%r8
+
+	mulxq	48(%rsi),%rax,%rbx
+	adoxq	%rax,%r8
+	adcxq	%rbx,%r9
+
+	mulxq	56(%rsi),%rdi,%r10
+	adoxq	%rdi,%r9
+	adcxq	%rbp,%r10
+	adoxq	%rbp,%r10
+
+.byte	0x66
+	movq	%r15,%rbx
+	shldq	$1,%r14,%r15
+	shldq	$1,%rcx,%r14
+
+	xorl	%ebp,%ebp
+	mulxq	%rdx,%rax,%rdx
+	adcxq	%rax,%r13
+	adcxq	%rdx,%r14
+	movq	32(%rsi),%rdx
+	adcxq	%rbp,%r15
+
+	movq	%r13,48(%rsp)
+	movq	%r14,56(%rsp)
+
+
+.byte	0xc4,0x62,0xc3,0xf6,0x9e,0x28,0x00,0x00,0x00
+	adoxq	%rdi,%r8
+	adcxq	%r11,%r9
+
+	mulxq	48(%rsi),%rax,%rcx
+	adoxq	%rax,%r9
+	adcxq	%rcx,%r10
+
+	mulxq	56(%rsi),%rdi,%r11
+	adoxq	%rdi,%r10
+	adcxq	%rbp,%r11
+	adoxq	%rbp,%r11
+
+	movq	%r9,%rcx
+	shldq	$1,%r8,%r9
+	shldq	$1,%rbx,%r8
+
+	xorl	%ebp,%ebp
+	mulxq	%rdx,%rax,%rdx
+	adcxq	%rax,%r15
+	adcxq	%rdx,%r8
+	movq	40(%rsi),%rdx
+	adcxq	%rbp,%r9
+
+	movq	%r15,64(%rsp)
+	movq	%r8,72(%rsp)
+
+
+.byte	0xc4,0xe2,0xfb,0xf6,0x9e,0x30,0x00,0x00,0x00
+	adoxq	%rax,%r10
+	adcxq	%rbx,%r11
+
+.byte	0xc4,0x62,0xc3,0xf6,0xa6,0x38,0x00,0x00,0x00
+	adoxq	%rdi,%r11
+	adcxq	%rbp,%r12
+	adoxq	%rbp,%r12
+
+	movq	%r11,%rbx
+	shldq	$1,%r10,%r11
+	shldq	$1,%rcx,%r10
+
+	xorl	%ebp,%ebp
+	mulxq	%rdx,%rax,%rdx
+	adcxq	%rax,%r9
+	adcxq	%rdx,%r10
+	movq	48(%rsi),%rdx
+	adcxq	%rbp,%r11
+
+	movq	%r9,80(%rsp)
+	movq	%r10,88(%rsp)
+
+
+.byte	0xc4,0x62,0xfb,0xf6,0xae,0x38,0x00,0x00,0x00
+	adoxq	%rax,%r12
+	adoxq	%rbp,%r13
+
+	xorq	%r14,%r14
+	shldq	$1,%r13,%r14
+	shldq	$1,%r12,%r13
+	shldq	$1,%rbx,%r12
+
+	xorl	%ebp,%ebp
+	mulxq	%rdx,%rax,%rdx
+	adcxq	%rax,%r11
+	adcxq	%rdx,%r12
+	movq	56(%rsi),%rdx
+	adcxq	%rbp,%r13
+
+.byte	0x4c,0x89,0x9c,0x24,0x60,0x00,0x00,0x00
+.byte	0x4c,0x89,0xa4,0x24,0x68,0x00,0x00,0x00
+
+
+	mulxq	%rdx,%rax,%rdx
+	adoxq	%rax,%r13
+	adoxq	%rbp,%rdx
+
+.byte	0x66
+	addq	%rdx,%r14
+
+	movq	%r13,112(%rsp)
+	movq	%r14,120(%rsp)
+.byte	102,72,15,126,199
+.byte	102,72,15,126,205
+
+	movq	128(%rsp),%rdx
+	movq	(%rsp),%r8
+	movq	8(%rsp),%r9
+	movq	16(%rsp),%r10
+	movq	24(%rsp),%r11
+	movq	32(%rsp),%r12
+	movq	40(%rsp),%r13
+	movq	48(%rsp),%r14
+	movq	56(%rsp),%r15
+
+	call	__rsaz_512_reducex
+
+	addq	64(%rsp),%r8
+	adcq	72(%rsp),%r9
+	adcq	80(%rsp),%r10
+	adcq	88(%rsp),%r11
+	adcq	96(%rsp),%r12
+	adcq	104(%rsp),%r13
+	adcq	112(%rsp),%r14
+	adcq	120(%rsp),%r15
+	sbbq	%rcx,%rcx
+
+	call	__rsaz_512_subtract
+
+	movq	%r8,%rdx
+	movq	%r9,%rax
+	movl	128+8(%rsp),%r8d
+	movq	%rdi,%rsi
+
+	decl	%r8d
+	jnz	.Loop_sqrx
+
+.Lsqr_tail:
 
 	leaq	128+24+48(%rsp),%rax
 	movq	-48(%rax),%r15
@@ -410,6 +684,10 @@ rsaz_512_mul:
 .byte	102,72,15,110,199
 .byte	102,72,15,110,201
 	movq	%r8,128(%rsp)
+	movl	$0x80100,%r11d
+	andl	OPENSSL_ia32cap_P+8(%rip),%r11d
+	cmpl	$0x80100,%r11d
+	je	.Lmulx
 	movq	(%rdx),%rbx
 	movq	%rdx,%rbp
 	call	__rsaz_512_mul
@@ -427,6 +705,29 @@ rsaz_512_mul:
 	movq	56(%rsp),%r15
 
 	call	__rsaz_512_reduce
+	jmp	.Lmul_tail
+
+.align	32
+.Lmulx:
+	movq	%rdx,%rbp
+	movq	(%rdx),%rdx
+	call	__rsaz_512_mulx
+
+.byte	102,72,15,126,199
+.byte	102,72,15,126,205
+
+	movq	128(%rsp),%rdx
+	movq	(%rsp),%r8
+	movq	8(%rsp),%r9
+	movq	16(%rsp),%r10
+	movq	24(%rsp),%r11
+	movq	32(%rsp),%r12
+	movq	40(%rsp),%r13
+	movq	48(%rsp),%r14
+	movq	56(%rsp),%r15
+
+	call	__rsaz_512_reducex
+.Lmul_tail:
 	addq	64(%rsp),%r8
 	adcq	72(%rsp),%r9
 	adcq	80(%rsp),%r10
@@ -517,6 +818,10 @@ rsaz_512_mul_gather4:
 	por	%xmm9,%xmm8
 	pshufd	$0x4e,%xmm8,%xmm9
 	por	%xmm9,%xmm8
+	movl	$0x80100,%r11d
+	andl	OPENSSL_ia32cap_P+8(%rip),%r11d
+	cmpl	$0x80100,%r11d
+	je	.Lmulx_gather
 .byte	102,76,15,126,195
 
 	movq	%r8,128(%rsp)
@@ -697,6 +1002,142 @@ rsaz_512_mul_gather4:
 	movq	56(%rsp),%r15
 
 	call	__rsaz_512_reduce
+	jmp	.Lmul_gather_tail
+
+.align	32
+.Lmulx_gather:
+.byte	102,76,15,126,194
+
+	movq	%r8,128(%rsp)
+	movq	%rdi,128+8(%rsp)
+	movq	%rcx,128+16(%rsp)
+
+	mulxq	(%rsi),%rbx,%r8
+	movq	%rbx,(%rsp)
+	xorl	%edi,%edi
+
+	mulxq	8(%rsi),%rax,%r9
+
+	mulxq	16(%rsi),%rbx,%r10
+	adcxq	%rax,%r8
+
+	mulxq	24(%rsi),%rax,%r11
+	adcxq	%rbx,%r9
+
+	mulxq	32(%rsi),%rbx,%r12
+	adcxq	%rax,%r10
+
+	mulxq	40(%rsi),%rax,%r13
+	adcxq	%rbx,%r11
+
+	mulxq	48(%rsi),%rbx,%r14
+	adcxq	%rax,%r12
+
+	mulxq	56(%rsi),%rax,%r15
+	adcxq	%rbx,%r13
+	adcxq	%rax,%r14
+.byte	0x67
+	movq	%r8,%rbx
+	adcxq	%rdi,%r15
+
+	movq	$-7,%rcx
+	jmp	.Loop_mulx_gather
+
+.align	32
+.Loop_mulx_gather:
+	movdqa	0(%rbp),%xmm8
+	movdqa	16(%rbp),%xmm9
+	movdqa	32(%rbp),%xmm10
+	movdqa	48(%rbp),%xmm11
+	pand	%xmm0,%xmm8
+	movdqa	64(%rbp),%xmm12
+	pand	%xmm1,%xmm9
+	movdqa	80(%rbp),%xmm13
+	pand	%xmm2,%xmm10
+	movdqa	96(%rbp),%xmm14
+	pand	%xmm3,%xmm11
+	movdqa	112(%rbp),%xmm15
+	leaq	128(%rbp),%rbp
+	pand	%xmm4,%xmm12
+	pand	%xmm5,%xmm13
+	pand	%xmm6,%xmm14
+	pand	%xmm7,%xmm15
+	por	%xmm10,%xmm8
+	por	%xmm11,%xmm9
+	por	%xmm12,%xmm8
+	por	%xmm13,%xmm9
+	por	%xmm14,%xmm8
+	por	%xmm15,%xmm9
+
+	por	%xmm9,%xmm8
+	pshufd	$0x4e,%xmm8,%xmm9
+	por	%xmm9,%xmm8
+.byte	102,76,15,126,194
+
+.byte	0xc4,0x62,0xfb,0xf6,0x86,0x00,0x00,0x00,0x00
+	adcxq	%rax,%rbx
+	adoxq	%r9,%r8
+
+	mulxq	8(%rsi),%rax,%r9
+	adcxq	%rax,%r8
+	adoxq	%r10,%r9
+
+	mulxq	16(%rsi),%rax,%r10
+	adcxq	%rax,%r9
+	adoxq	%r11,%r10
+
+.byte	0xc4,0x62,0xfb,0xf6,0x9e,0x18,0x00,0x00,0x00
+	adcxq	%rax,%r10
+	adoxq	%r12,%r11
+
+	mulxq	32(%rsi),%rax,%r12
+	adcxq	%rax,%r11
+	adoxq	%r13,%r12
+
+	mulxq	40(%rsi),%rax,%r13
+	adcxq	%rax,%r12
+	adoxq	%r14,%r13
+
+.byte	0xc4,0x62,0xfb,0xf6,0xb6,0x30,0x00,0x00,0x00
+	adcxq	%rax,%r13
+.byte	0x67
+	adoxq	%r15,%r14
+
+	mulxq	56(%rsi),%rax,%r15
+	movq	%rbx,64(%rsp,%rcx,8)
+	adcxq	%rax,%r14
+	adoxq	%rdi,%r15
+	movq	%r8,%rbx
+	adcxq	%rdi,%r15
+
+	incq	%rcx
+	jnz	.Loop_mulx_gather
+
+	movq	%r8,64(%rsp)
+	movq	%r9,64+8(%rsp)
+	movq	%r10,64+16(%rsp)
+	movq	%r11,64+24(%rsp)
+	movq	%r12,64+32(%rsp)
+	movq	%r13,64+40(%rsp)
+	movq	%r14,64+48(%rsp)
+	movq	%r15,64+56(%rsp)
+
+	movq	128(%rsp),%rdx
+	movq	128+8(%rsp),%rdi
+	movq	128+16(%rsp),%rbp
+
+	movq	(%rsp),%r8
+	movq	8(%rsp),%r9
+	movq	16(%rsp),%r10
+	movq	24(%rsp),%r11
+	movq	32(%rsp),%r12
+	movq	40(%rsp),%r13
+	movq	48(%rsp),%r14
+	movq	56(%rsp),%r15
+
+	call	__rsaz_512_reducex
+
+.Lmul_gather_tail:
 	addq	64(%rsp),%r8
 	adcq	72(%rsp),%r9
 	adcq	80(%rsp),%r10
@@ -741,6 +1182,10 @@ rsaz_512_mul_scatter4:
 	movq	%rcx,128(%rsp)
 
 	movq	%rdi,%rbp
+	movl	$0x80100,%r11d
+	andl	OPENSSL_ia32cap_P+8(%rip),%r11d
+	cmpl	$0x80100,%r11d
+	je	.Lmulx_scatter
 	movq	(%rdi),%rbx
 	call	__rsaz_512_mul
 
@@ -757,6 +1202,29 @@ rsaz_512_mul_scatter4:
 	movq	56(%rsp),%r15
 
 	call	__rsaz_512_reduce
+	jmp	.Lmul_scatter_tail
+
+.align	32
+.Lmulx_scatter:
+	movq	(%rdi),%rdx
+	call	__rsaz_512_mulx
+
+.byte	102,72,15,126,199
+.byte	102,72,15,126,205
+
+	movq	128(%rsp),%rdx
+	movq	(%rsp),%r8
+	movq	8(%rsp),%r9
+	movq	16(%rsp),%r10
+	movq	24(%rsp),%r11
+	movq	32(%rsp),%r12
+	movq	40(%rsp),%r13
+	movq	48(%rsp),%r14
+	movq	56(%rsp),%r15
+
+	call	__rsaz_512_reducex
+
+.Lmul_scatter_tail:
 	addq	64(%rsp),%r8
 	adcq	72(%rsp),%r9
 	adcq	80(%rsp),%r10
@@ -803,6 +1271,7 @@ rsaz_512_mul_by_one:
 
 	subq	$128+24,%rsp
 .Lmul_by_one_body:
+	movl	OPENSSL_ia32cap_P+8(%rip),%eax
 	movq	%rdx,%rbp
 	movq	%rcx,128(%rsp)
 
@@ -823,7 +1292,16 @@ rsaz_512_mul_by_one:
 	movdqa	%xmm0,64(%rsp)
 	movdqa	%xmm0,80(%rsp)
 	movdqa	%xmm0,96(%rsp)
+	andl	$0x80100,%eax
+	cmpl	$0x80100,%eax
+	je	.Lby_one_callx
 	call	__rsaz_512_reduce
+	jmp	.Lby_one_tail
+.align	32
+.Lby_one_callx:
+	movq	128(%rsp),%rdx
+	call	__rsaz_512_reducex
+.Lby_one_tail:
 	movq	%r8,(%rdi)
 	movq	%r9,8(%rdi)
 	movq	%r10,16(%rdi)
@@ -927,6 +1405,62 @@ __rsaz_512_reduce:
 
 	.byte	0xf3,0xc3
 .size	__rsaz_512_reduce,.-__rsaz_512_reduce
+.type	__rsaz_512_reducex,@function
+.align	32
+__rsaz_512_reducex:
+
+	imulq	%r8,%rdx
+	xorq	%rsi,%rsi
+	movl	$8,%ecx
+	jmp	.Lreduction_loopx
+
+.align	32
+.Lreduction_loopx:
+	movq	%r8,%rbx
+	mulxq	0(%rbp),%rax,%r8
+	adcxq	%rbx,%rax
+	adoxq	%r9,%r8
+
+	mulxq	8(%rbp),%rax,%r9
+	adcxq	%rax,%r8
+	adoxq	%r10,%r9
+
+	mulxq	16(%rbp),%rbx,%r10
+	adcxq	%rbx,%r9
+	adoxq	%r11,%r10
+
+	mulxq	24(%rbp),%rbx,%r11
+	adcxq	%rbx,%r10
+	adoxq	%r12,%r11
+
+.byte	0xc4,0x62,0xe3,0xf6,0xa5,0x20,0x00,0x00,0x00
+	movq	%rdx,%rax
+	movq	%r8,%rdx
+	adcxq	%rbx,%r11
+	adoxq	%r13,%r12
+
+	mulxq	128+8(%rsp),%rbx,%rdx
+	movq	%rax,%rdx
+
+	mulxq	40(%rbp),%rax,%r13
+	adcxq	%rax,%r12
+	adoxq	%r14,%r13
+
+.byte	0xc4,0x62,0xfb,0xf6,0xb5,0x30,0x00,0x00,0x00
+	adcxq	%rax,%r13
+	adoxq	%r15,%r14
+
+	mulxq	56(%rbp),%rax,%r15
+	movq	%rbx,%rdx
+	adcxq	%rax,%r14
+	adoxq	%rsi,%r15
+	adcxq	%rsi,%r15
+
+	decl	%ecx
+	jne	.Lreduction_loopx
+
+	.byte	0xf3,0xc3
+.size	__rsaz_512_reducex,.-__rsaz_512_reducex
 .type	__rsaz_512_subtract,@function
 .align	32
 __rsaz_512_subtract:
@@ -1126,6 +1660,126 @@ __rsaz_512_mul:
 
 	.byte	0xf3,0xc3
 .size	__rsaz_512_mul,.-__rsaz_512_mul
+.type	__rsaz_512_mulx,@function
+.align	32
+__rsaz_512_mulx:
+	mulxq	(%rsi),%rbx,%r8
+	movq	$-6,%rcx
+
+	mulxq	8(%rsi),%rax,%r9
+	movq	%rbx,8(%rsp)
+
+	mulxq	16(%rsi),%rbx,%r10
+	adcq	%rax,%r8
+
+	mulxq	24(%rsi),%rax,%r11
+	adcq	%rbx,%r9
+
+	mulxq	32(%rsi),%rbx,%r12
+	adcq	%rax,%r10
+
+	mulxq	40(%rsi),%rax,%r13
+	adcq	%rbx,%r11
+
+	mulxq	48(%rsi),%rbx,%r14
+	adcq	%rax,%r12
+
+	mulxq	56(%rsi),%rax,%r15
+	movq	8(%rbp),%rdx
+	adcq	%rbx,%r13
+	adcq	%rax,%r14
+	adcq	$0,%r15
+
+	xorq	%rdi,%rdi
+	jmp	.Loop_mulx
+
+.align	32
+.Loop_mulx:
+	movq	%r8,%rbx
+	mulxq	(%rsi),%rax,%r8
+	adcxq	%rax,%rbx
+	adoxq	%r9,%r8
+
+	mulxq	8(%rsi),%rax,%r9
+	adcxq	%rax,%r8
+	adoxq	%r10,%r9
+
+	mulxq	16(%rsi),%rax,%r10
+	adcxq	%rax,%r9
+	adoxq	%r11,%r10
+
+	mulxq	24(%rsi),%rax,%r11
+	adcxq	%rax,%r10
+	adoxq	%r12,%r11
+
+.byte	0x3e,0xc4,0x62,0xfb,0xf6,0xa6,0x20,0x00,0x00,0x00
+	adcxq	%rax,%r11
+	adoxq	%r13,%r12
+
+	mulxq	40(%rsi),%rax,%r13
+	adcxq	%rax,%r12
+	adoxq	%r14,%r13
+
+	mulxq	48(%rsi),%rax,%r14
+	adcxq	%rax,%r13
+	adoxq	%r15,%r14
+
+	mulxq	56(%rsi),%rax,%r15
+	movq	64(%rbp,%rcx,8),%rdx
+	movq	%rbx,8+64-8(%rsp,%rcx,8)
+	adcxq	%rax,%r14
+	adoxq	%rdi,%r15
+	adcxq	%rdi,%r15
+
+	incq	%rcx
+	jnz	.Loop_mulx
+
+	movq	%r8,%rbx
+	mulxq	(%rsi),%rax,%r8
+	adcxq	%rax,%rbx
+	adoxq	%r9,%r8
+
+.byte	0xc4,0x62,0xfb,0xf6,0x8e,0x08,0x00,0x00,0x00
+	adcxq	%rax,%r8
+	adoxq	%r10,%r9
+
+.byte	0xc4,0x62,0xfb,0xf6,0x96,0x10,0x00,0x00,0x00
+	adcxq	%rax,%r9
+	adoxq	%r11,%r10
+
+	mulxq	24(%rsi),%rax,%r11
+	adcxq	%rax,%r10
+	adoxq	%r12,%r11
+
+	mulxq	32(%rsi),%rax,%r12
+	adcxq	%rax,%r11
+	adoxq	%r13,%r12
+
+	mulxq	40(%rsi),%rax,%r13
+	adcxq	%rax,%r12
+	adoxq	%r14,%r13
+
+.byte	0xc4,0x62,0xfb,0xf6,0xb6,0x30,0x00,0x00,0x00
+	adcxq	%rax,%r13
+	adoxq	%r15,%r14
+
+.byte	0xc4,0x62,0xfb,0xf6,0xbe,0x38,0x00,0x00,0x00
+	adcxq	%rax,%r14
+	adoxq	%rdi,%r15
+	adcxq	%rdi,%r15
+
+	movq	%rbx,8+64-8(%rsp)
+	movq	%r8,8+64(%rsp)
+	movq	%r9,8+64+8(%rsp)
+	movq	%r10,8+64+16(%rsp)
+	movq	%r11,8+64+24(%rsp)
+	movq	%r12,8+64+32(%rsp)
+	movq	%r13,8+64+40(%rsp)
+	movq	%r14,8+64+48(%rsp)
+	movq	%r15,8+64+56(%rsp)
+
+	.byte	0xf3,0xc3
+.size	__rsaz_512_mulx,.-__rsaz_512_mulx
 .globl	rsaz_512_scatter4
 .type	rsaz_512_scatter4,@function
 .align	16

deps/openssl/asm/x64-elf-gas/bn/x86_64-mont.s

@@ -10,6 +10,7 @@ bn_mul_mont:
 	jnz	.Lmul_enter
 	cmpl	$8,%r9d
 	jb	.Lmul_enter
+	movl	OPENSSL_ia32cap_P+8(%rip),%r11d
 	cmpq	%rsi,%rdx
 	jne	.Lmul4x_enter
 	testl	$7,%r9d
@@ -34,6 +35,20 @@ bn_mul_mont:
 
 	movq	%r11,8(%rsp,%r9,8)
 .Lmul_body:
+
+
+
+
+
+
+	subq	%rsp,%r11
+	andq	$-4096,%r11
+.Lmul_page_walk:
+	movq	(%rsp,%r11,1),%r10
+	subq	$4096,%r11
+.byte	0x66,0x2e
+	jnc	.Lmul_page_walk
+
 	movq	%rdx,%r12
 	movq	(%r8),%r8
 	movq	(%r12),%rbx
@@ -215,6 +230,9 @@ bn_mul_mont:
 .align	16
 bn_mul4x_mont:
 .Lmul4x_enter:
+	andl	$0x80100,%r11d
+	cmpl	$0x80100,%r11d
+	je	.Lmulx4x_enter
 	pushq	%rbx
 	pushq	%rbp
 	pushq	%r12
@@ -231,6 +249,14 @@ bn_mul4x_mont:
 
 	movq	%r11,8(%rsp,%r9,8)
 .Lmul4x_body:
+	subq	%rsp,%r11
+	andq	$-4096,%r11
+.Lmul4x_page_walk:
+	movq	(%rsp,%r11,1),%r10
+	subq	$4096,%r11
+.byte	0x2e
+	jnc	.Lmul4x_page_walk
+
 	movq	%rdi,16(%rsp,%r9,8)
 	movq	%rdx,%r12
 	movq	(%r8),%r8
@@ -611,6 +637,7 @@ bn_mul4x_mont:
 .size	bn_mul4x_mont,.-bn_mul4x_mont
 
 
+
 .type	bn_sqr8x_mont,@function
 .align	32
 bn_sqr8x_mont:
@@ -653,6 +680,15 @@ bn_sqr8x_mont:
 	subq	%r11,%rsp
 .Lsqr8x_sp_done:
 	andq	$-64,%rsp
+	movq	%rax,%r11
+	subq	%rsp,%r11
+	andq	$-4096,%r11
+.Lsqr8x_page_walk:
+	movq	(%rsp,%r11,1),%r10
+	subq	$4096,%r11
+.byte	0x2e
+	jnc	.Lsqr8x_page_walk
+
 	movq	%r9,%r10
 	negq	%r9
 
@@ -664,6 +700,25 @@ bn_sqr8x_mont:
 	pxor	%xmm0,%xmm0
 .byte	102,72,15,110,207
 .byte	102,73,15,110,218
+	movl	OPENSSL_ia32cap_P+8(%rip),%eax
+	andl	$0x80100,%eax
+	cmpl	$0x80100,%eax
+	jne	.Lsqr8x_nox
+
+	call	bn_sqrx8x_internal
+
+
+
+
+	leaq	(%r8,%rcx,1),%rbx
+	movq	%rcx,%r9
+	movq	%rcx,%rdx
+.byte	102,72,15,126,207
+	sarq	$3+2,%rcx
+	jmp	.Lsqr8x_sub
+
+.align	32
+.Lsqr8x_nox:
 	call	bn_sqr8x_internal
 
 
@@ -742,5 +797,336 @@ bn_sqr8x_mont:
 .Lsqr8x_epilogue:
 	.byte	0xf3,0xc3
 .size	bn_sqr8x_mont,.-bn_sqr8x_mont
+.type	bn_mulx4x_mont,@function
+.align	32
+bn_mulx4x_mont:
+.Lmulx4x_enter:
+	movq	%rsp,%rax
+	pushq	%rbx
+	pushq	%rbp
+	pushq	%r12
+	pushq	%r13
+	pushq	%r14
+	pushq	%r15
+
+	shll	$3,%r9d
+.byte	0x67
+	xorq	%r10,%r10
+	subq	%r9,%r10
+	movq	(%r8),%r8
+	leaq	-72(%rsp,%r10,1),%rsp
+	andq	$-128,%rsp
+	movq	%rax,%r11
+	subq	%rsp,%r11
+	andq	$-4096,%r11
+.Lmulx4x_page_walk:
+	movq	(%rsp,%r11,1),%r10
+	subq	$4096,%r11
+.byte	0x66,0x2e
+	jnc	.Lmulx4x_page_walk
+
+	leaq	(%rdx,%r9,1),%r10
+
+
+
+
+
+
+
+
+
+
+
+
+	movq	%r9,0(%rsp)
+	shrq	$5,%r9
+	movq	%r10,16(%rsp)
+	subq	$1,%r9
+	movq	%r8,24(%rsp)
+	movq	%rdi,32(%rsp)
+	movq	%rax,40(%rsp)
+	movq	%r9,48(%rsp)
+	jmp	.Lmulx4x_body
+
+.align	32
+.Lmulx4x_body:
+	leaq	8(%rdx),%rdi
+	movq	(%rdx),%rdx
+	leaq	64+32(%rsp),%rbx
+	movq	%rdx,%r9
+
+	mulxq	0(%rsi),%r8,%rax
+	mulxq	8(%rsi),%r11,%r14
+	addq	%rax,%r11
+	movq	%rdi,8(%rsp)
+	mulxq	16(%rsi),%r12,%r13
+	adcq	%r14,%r12
+	adcq	$0,%r13
+
+	movq	%r8,%rdi
+	imulq	24(%rsp),%r8
+	xorq	%rbp,%rbp
+
+	mulxq	24(%rsi),%rax,%r14
+	movq	%r8,%rdx
+	leaq	32(%rsi),%rsi
+	adcxq	%rax,%r13
+	adcxq	%rbp,%r14
+
+	mulxq	0(%rcx),%rax,%r10
+	adcxq	%rax,%rdi
+	adoxq	%r11,%r10
+	mulxq	8(%rcx),%rax,%r11
+	adcxq	%rax,%r10
+	adoxq	%r12,%r11
+.byte	0xc4,0x62,0xfb,0xf6,0xa1,0x10,0x00,0x00,0x00
+	movq	48(%rsp),%rdi
+	movq	%r10,-32(%rbx)
+	adcxq	%rax,%r11
+	adoxq	%r13,%r12
+	mulxq	24(%rcx),%rax,%r15
+	movq	%r9,%rdx
+	movq	%r11,-24(%rbx)
+	adcxq	%rax,%r12
+	adoxq	%rbp,%r15
+	leaq	32(%rcx),%rcx
+	movq	%r12,-16(%rbx)
+
+	jmp	.Lmulx4x_1st
+
+.align	32
+.Lmulx4x_1st:
+	adcxq	%rbp,%r15
+	mulxq	0(%rsi),%r10,%rax
+	adcxq	%r14,%r10
+	mulxq	8(%rsi),%r11,%r14
+	adcxq	%rax,%r11
+	mulxq	16(%rsi),%r12,%rax
+	adcxq	%r14,%r12
+	mulxq	24(%rsi),%r13,%r14
+.byte	0x67,0x67
+	movq	%r8,%rdx
+	adcxq	%rax,%r13
+	adcxq	%rbp,%r14
+	leaq	32(%rsi),%rsi
+	leaq	32(%rbx),%rbx
+
+	adoxq	%r15,%r10
+	mulxq	0(%rcx),%rax,%r15
+	adcxq	%rax,%r10
+	adoxq	%r15,%r11
+	mulxq	8(%rcx),%rax,%r15
+	adcxq	%rax,%r11
+	adoxq	%r15,%r12
+	mulxq	16(%rcx),%rax,%r15
+	movq	%r10,-40(%rbx)
+	adcxq	%rax,%r12
+	movq	%r11,-32(%rbx)
+	adoxq	%r15,%r13
+	mulxq	24(%rcx),%rax,%r15
+	movq	%r9,%rdx
+	movq	%r12,-24(%rbx)
+	adcxq	%rax,%r13
+	adoxq	%rbp,%r15
+	leaq	32(%rcx),%rcx
+	movq	%r13,-16(%rbx)
+
+	decq	%rdi
+	jnz	.Lmulx4x_1st
+
+	movq	0(%rsp),%rax
+	movq	8(%rsp),%rdi
+	adcq	%rbp,%r15
+	addq	%r15,%r14
+	sbbq	%r15,%r15
+	movq	%r14,-8(%rbx)
+	jmp	.Lmulx4x_outer
+
+.align	32
+.Lmulx4x_outer:
+	movq	(%rdi),%rdx
+	leaq	8(%rdi),%rdi
+	subq	%rax,%rsi
+	movq	%r15,(%rbx)
+	leaq	64+32(%rsp),%rbx
+	subq	%rax,%rcx
+
+	mulxq	0(%rsi),%r8,%r11
+	xorl	%ebp,%ebp
+	movq	%rdx,%r9
+	mulxq	8(%rsi),%r14,%r12
+	adoxq	-32(%rbx),%r8
+	adcxq	%r14,%r11
+	mulxq	16(%rsi),%r15,%r13
+	adoxq	-24(%rbx),%r11
+	adcxq	%r15,%r12
+	adoxq	%rbp,%r12
+	adcxq	%rbp,%r13
+
+	movq	%rdi,8(%rsp)
+.byte	0x67
+	movq	%r8,%r15
+	imulq	24(%rsp),%r8
+	xorl	%ebp,%ebp
+
+	mulxq	24(%rsi),%rax,%r14
+	movq	%r8,%rdx
+	adoxq	-16(%rbx),%r12
+	adcxq	%rax,%r13
+	adoxq	-8(%rbx),%r13
+	adcxq	%rbp,%r14
+	leaq	32(%rsi),%rsi
+	adoxq	%rbp,%r14
+
+	mulxq	0(%rcx),%rax,%r10
+	adcxq	%rax,%r15
+	adoxq	%r11,%r10
+	mulxq	8(%rcx),%rax,%r11
+	adcxq	%rax,%r10
+	adoxq	%r12,%r11
+	mulxq	16(%rcx),%rax,%r12
+	movq	%r10,-32(%rbx)
+	adcxq	%rax,%r11
+	adoxq	%r13,%r12
+	mulxq	24(%rcx),%rax,%r15
+	movq	%r9,%rdx
+	movq	%r11,-24(%rbx)
+	leaq	32(%rcx),%rcx
+	adcxq	%rax,%r12
+	adoxq	%rbp,%r15
+	movq	48(%rsp),%rdi
+	movq	%r12,-16(%rbx)
+
+	jmp	.Lmulx4x_inner
+
+.align	32
+.Lmulx4x_inner:
+	mulxq	0(%rsi),%r10,%rax
+	adcxq	%rbp,%r15
+	adoxq	%r14,%r10
+	mulxq	8(%rsi),%r11,%r14
+	adcxq	0(%rbx),%r10
+	adoxq	%rax,%r11
+	mulxq	16(%rsi),%r12,%rax
+	adcxq	8(%rbx),%r11
+	adoxq	%r14,%r12
+	mulxq	24(%rsi),%r13,%r14
+	movq	%r8,%rdx
+	adcxq	16(%rbx),%r12
+	adoxq	%rax,%r13
+	adcxq	24(%rbx),%r13
+	adoxq	%rbp,%r14
+	leaq	32(%rsi),%rsi
+	leaq	32(%rbx),%rbx
+	adcxq	%rbp,%r14
+
+	adoxq	%r15,%r10
+	mulxq	0(%rcx),%rax,%r15
+	adcxq	%rax,%r10
+	adoxq	%r15,%r11
+	mulxq	8(%rcx),%rax,%r15
+	adcxq	%rax,%r11
+	adoxq	%r15,%r12
+	mulxq	16(%rcx),%rax,%r15
+	movq	%r10,-40(%rbx)
+	adcxq	%rax,%r12
+	adoxq	%r15,%r13
+	mulxq	24(%rcx),%rax,%r15
+	movq	%r9,%rdx
+	movq	%r11,-32(%rbx)
+	movq	%r12,-24(%rbx)
+	adcxq	%rax,%r13
+	adoxq	%rbp,%r15
+	leaq	32(%rcx),%rcx
+	movq	%r13,-16(%rbx)
+
+	decq	%rdi
+	jnz	.Lmulx4x_inner
+
+	movq	0(%rsp),%rax
+	movq	8(%rsp),%rdi
+	adcq	%rbp,%r15
+	subq	0(%rbx),%rbp
+	adcq	%r15,%r14
+	sbbq	%r15,%r15
+	movq	%r14,-8(%rbx)
+
+	cmpq	16(%rsp),%rdi
+	jne	.Lmulx4x_outer
+
+	leaq	64(%rsp),%rbx
+	subq	%rax,%rcx
+	negq	%r15
+	movq	%rax,%rdx
+	shrq	$3+2,%rax
+	movq	32(%rsp),%rdi
+	jmp	.Lmulx4x_sub
+
+.align	32
+.Lmulx4x_sub:
+	movq	0(%rbx),%r11
+	movq	8(%rbx),%r12
+	movq	16(%rbx),%r13
+	movq	24(%rbx),%r14
+	leaq	32(%rbx),%rbx
+	sbbq	0(%rcx),%r11
+	sbbq	8(%rcx),%r12
+	sbbq	16(%rcx),%r13
+	sbbq	24(%rcx),%r14
+	leaq	32(%rcx),%rcx
+	movq	%r11,0(%rdi)
+	movq	%r12,8(%rdi)
+	movq	%r13,16(%rdi)
+	movq	%r14,24(%rdi)
+	leaq	32(%rdi),%rdi
+	decq	%rax
+	jnz	.Lmulx4x_sub
+
+	sbbq	$0,%r15
+	leaq	64(%rsp),%rbx
+	subq	%rdx,%rdi
+
+.byte	102,73,15,110,207
+	pxor	%xmm0,%xmm0
+	pshufd	$0,%xmm1,%xmm1
+	movq	40(%rsp),%rsi
+	jmp	.Lmulx4x_cond_copy
+
+.align	32
+.Lmulx4x_cond_copy:
+	movdqa	0(%rbx),%xmm2
+	movdqa	16(%rbx),%xmm3
+	leaq	32(%rbx),%rbx
+	movdqu	0(%rdi),%xmm4
+	movdqu	16(%rdi),%xmm5
+	leaq	32(%rdi),%rdi
+	movdqa	%xmm0,-32(%rbx)
+	movdqa	%xmm0,-16(%rbx)
+	pcmpeqd	%xmm1,%xmm0
+	pand	%xmm1,%xmm2
+	pand	%xmm1,%xmm3
+	pand	%xmm0,%xmm4
+	pand	%xmm0,%xmm5
+	pxor	%xmm0,%xmm0
+	por	%xmm2,%xmm4
+	por	%xmm3,%xmm5
+	movdqu	%xmm4,-32(%rdi)
+	movdqu	%xmm5,-16(%rdi)
+	subq	$32,%rdx
+	jnz	.Lmulx4x_cond_copy
+
+	movq	%rdx,(%rbx)
+
+	movq	$1,%rax
+	movq	-48(%rsi),%r15
+	movq	-40(%rsi),%r14
+	movq	-32(%rsi),%r13
+	movq	-24(%rsi),%r12
+	movq	-16(%rsi),%rbp
+	movq	-8(%rsi),%rbx
+	leaq	(%rsi),%rsp
+.Lmulx4x_epilogue:
+	.byte	0xf3,0xc3
+.size	bn_mulx4x_mont,.-bn_mulx4x_mont
 .byte	77,111,110,116,103,111,109,101,114,121,32,77,117,108,116,105,112,108,105,99,97,116,105,111,110,32,102,111,114,32,120,56,54,95,54,52,44,32,67,82,89,80,84,79,71,65,77,83,32,98,121,32,60,97,112,112,114,111,64,111,112,101,110,115,115,108,46,111,114,103,62,0
 .align	16

deps/openssl/asm/x64-elf-gas/camellia/cmll-x86_64.s

@@ -1624,7 +1624,7 @@ Camellia_cbc_encrypt:
 	leaq	-64-63(%rcx),%r10
 	subq	%rsp,%r10
 	negq	%r10
-	andq	$960,%r10
+	andq	$0x3C0,%r10
 	subq	%r10,%rsp
 
 

deps/openssl/asm/x64-elf-gas/md5/md5-x86_64.s

@@ -493,14 +493,14 @@ md5_block_asm_data_order:
 	movl	%ecx,%r11d
 	addl	%ecx,%ebx
 	movl	0(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	xorl	%edx,%r11d
 	leal	-198630844(%rax,%r10,1),%eax
 	orl	%ebx,%r11d
 	xorl	%ecx,%r11d
 	addl	%r11d,%eax
 	movl	28(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$6,%eax
 	xorl	%ecx,%r11d
 	addl	%ebx,%eax
@@ -509,7 +509,7 @@ md5_block_asm_data_order:
 	xorl	%ebx,%r11d
 	addl	%r11d,%edx
 	movl	56(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$10,%edx
 	xorl	%ebx,%r11d
 	addl	%eax,%edx
@@ -518,7 +518,7 @@ md5_block_asm_data_order:
 	xorl	%eax,%r11d
 	addl	%r11d,%ecx
 	movl	20(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$15,%ecx
 	xorl	%eax,%r11d
 	addl	%edx,%ecx
@@ -527,7 +527,7 @@ md5_block_asm_data_order:
 	xorl	%edx,%r11d
 	addl	%r11d,%ebx
 	movl	48(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$21,%ebx
 	xorl	%edx,%r11d
 	addl	%ecx,%ebx
@@ -536,7 +536,7 @@ md5_block_asm_data_order:
 	xorl	%ecx,%r11d
 	addl	%r11d,%eax
 	movl	12(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$6,%eax
 	xorl	%ecx,%r11d
 	addl	%ebx,%eax
@@ -545,7 +545,7 @@ md5_block_asm_data_order:
 	xorl	%ebx,%r11d
 	addl	%r11d,%edx
 	movl	40(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$10,%edx
 	xorl	%ebx,%r11d
 	addl	%eax,%edx
@@ -554,7 +554,7 @@ md5_block_asm_data_order:
 	xorl	%eax,%r11d
 	addl	%r11d,%ecx
 	movl	4(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$15,%ecx
 	xorl	%eax,%r11d
 	addl	%edx,%ecx
@@ -563,7 +563,7 @@ md5_block_asm_data_order:
 	xorl	%edx,%r11d
 	addl	%r11d,%ebx
 	movl	32(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$21,%ebx
 	xorl	%edx,%r11d
 	addl	%ecx,%ebx
@@ -572,7 +572,7 @@ md5_block_asm_data_order:
 	xorl	%ecx,%r11d
 	addl	%r11d,%eax
 	movl	60(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$6,%eax
 	xorl	%ecx,%r11d
 	addl	%ebx,%eax
@@ -581,7 +581,7 @@ md5_block_asm_data_order:
 	xorl	%ebx,%r11d
 	addl	%r11d,%edx
 	movl	24(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$10,%edx
 	xorl	%ebx,%r11d
 	addl	%eax,%edx
@@ -590,7 +590,7 @@ md5_block_asm_data_order:
 	xorl	%eax,%r11d
 	addl	%r11d,%ecx
 	movl	52(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$15,%ecx
 	xorl	%eax,%r11d
 	addl	%edx,%ecx
@@ -599,7 +599,7 @@ md5_block_asm_data_order:
 	xorl	%edx,%r11d
 	addl	%r11d,%ebx
 	movl	16(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$21,%ebx
 	xorl	%edx,%r11d
 	addl	%ecx,%ebx
@@ -608,7 +608,7 @@ md5_block_asm_data_order:
 	xorl	%ecx,%r11d
 	addl	%r11d,%eax
 	movl	44(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$6,%eax
 	xorl	%ecx,%r11d
 	addl	%ebx,%eax
@@ -617,7 +617,7 @@ md5_block_asm_data_order:
 	xorl	%ebx,%r11d
 	addl	%r11d,%edx
 	movl	8(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$10,%edx
 	xorl	%ebx,%r11d
 	addl	%eax,%edx
@@ -626,7 +626,7 @@ md5_block_asm_data_order:
 	xorl	%eax,%r11d
 	addl	%r11d,%ecx
 	movl	36(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$15,%ecx
 	xorl	%eax,%r11d
 	addl	%edx,%ecx
@@ -635,7 +635,7 @@ md5_block_asm_data_order:
 	xorl	%edx,%r11d
 	addl	%r11d,%ebx
 	movl	0(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$21,%ebx
 	xorl	%edx,%r11d
 	addl	%ecx,%ebx

deps/openssl/asm/x64-elf-gas/modes/aesni-gcm-x86_64.s

@@ -1,15 +1,753 @@
 .text
 
-.globl	aesni_gcm_encrypt
-.type	aesni_gcm_encrypt,@function
-aesni_gcm_encrypt:
-	xorl	%eax,%eax
-	.byte	0xf3,0xc3
-.size	aesni_gcm_encrypt,.-aesni_gcm_encrypt
+.type	_aesni_ctr32_ghash_6x,@function
+.align	32
+_aesni_ctr32_ghash_6x:
+	vmovdqu	32(%r11),%xmm2
+	subq	$6,%rdx
+	vpxor	%xmm4,%xmm4,%xmm4
+	vmovdqu	0-128(%rcx),%xmm15
+	vpaddb	%xmm2,%xmm1,%xmm10
+	vpaddb	%xmm2,%xmm10,%xmm11
+	vpaddb	%xmm2,%xmm11,%xmm12
+	vpaddb	%xmm2,%xmm12,%xmm13
+	vpaddb	%xmm2,%xmm13,%xmm14
+	vpxor	%xmm15,%xmm1,%xmm9
+	vmovdqu	%xmm4,16+8(%rsp)
+	jmp	.Loop6x
+
+.align	32
+.Loop6x:
+	addl	$100663296,%ebx
+	jc	.Lhandle_ctr32
+	vmovdqu	0-32(%r9),%xmm3
+	vpaddb	%xmm2,%xmm14,%xmm1
+	vpxor	%xmm15,%xmm10,%xmm10
+	vpxor	%xmm15,%xmm11,%xmm11
+
+.Lresume_ctr32:
+	vmovdqu	%xmm1,(%r8)
+	vpclmulqdq	$0x10,%xmm3,%xmm7,%xmm5
+	vpxor	%xmm15,%xmm12,%xmm12
+	vmovups	16-128(%rcx),%xmm2
+	vpclmulqdq	$0x01,%xmm3,%xmm7,%xmm6
+	xorq	%r12,%r12
+	cmpq	%r14,%r15
+
+	vaesenc	%xmm2,%xmm9,%xmm9
+	vmovdqu	48+8(%rsp),%xmm0
+	vpxor	%xmm15,%xmm13,%xmm13
+	vpclmulqdq	$0x00,%xmm3,%xmm7,%xmm1
+	vaesenc	%xmm2,%xmm10,%xmm10
+	vpxor	%xmm15,%xmm14,%xmm14
+	setnc	%r12b
+	vpclmulqdq	$0x11,%xmm3,%xmm7,%xmm7
+	vaesenc	%xmm2,%xmm11,%xmm11
+	vmovdqu	16-32(%r9),%xmm3
+	negq	%r12
+	vaesenc	%xmm2,%xmm12,%xmm12
+	vpxor	%xmm5,%xmm6,%xmm6
+	vpclmulqdq	$0x00,%xmm3,%xmm0,%xmm5
+	vpxor	%xmm4,%xmm8,%xmm8
+	vaesenc	%xmm2,%xmm13,%xmm13
+	vpxor	%xmm5,%xmm1,%xmm4
+	andq	$0x60,%r12
+	vmovups	32-128(%rcx),%xmm15
+	vpclmulqdq	$0x10,%xmm3,%xmm0,%xmm1
+	vaesenc	%xmm2,%xmm14,%xmm14
+
+	vpclmulqdq	$0x01,%xmm3,%xmm0,%xmm2
+	leaq	(%r14,%r12,1),%r14
+	vaesenc	%xmm15,%xmm9,%xmm9
+	vpxor	16+8(%rsp),%xmm8,%xmm8
+	vpclmulqdq	$0x11,%xmm3,%xmm0,%xmm3
+	vmovdqu	64+8(%rsp),%xmm0
+	vaesenc	%xmm15,%xmm10,%xmm10
+	movbeq	88(%r14),%r13
+	vaesenc	%xmm15,%xmm11,%xmm11
+	movbeq	80(%r14),%r12
+	vaesenc	%xmm15,%xmm12,%xmm12
+	movq	%r13,32+8(%rsp)
+	vaesenc	%xmm15,%xmm13,%xmm13
+	movq	%r12,40+8(%rsp)
+	vmovdqu	48-32(%r9),%xmm5
+	vaesenc	%xmm15,%xmm14,%xmm14
+
+	vmovups	48-128(%rcx),%xmm15
+	vpxor	%xmm1,%xmm6,%xmm6
+	vpclmulqdq	$0x00,%xmm5,%xmm0,%xmm1
+	vaesenc	%xmm15,%xmm9,%xmm9
+	vpxor	%xmm2,%xmm6,%xmm6
+	vpclmulqdq	$0x10,%xmm5,%xmm0,%xmm2
+	vaesenc	%xmm15,%xmm10,%xmm10
+	vpxor	%xmm3,%xmm7,%xmm7
+	vpclmulqdq	$0x01,%xmm5,%xmm0,%xmm3
+	vaesenc	%xmm15,%xmm11,%xmm11
+	vpclmulqdq	$0x11,%xmm5,%xmm0,%xmm5
+	vmovdqu	80+8(%rsp),%xmm0
+	vaesenc	%xmm15,%xmm12,%xmm12
+	vaesenc	%xmm15,%xmm13,%xmm13
+	vpxor	%xmm1,%xmm4,%xmm4
+	vmovdqu	64-32(%r9),%xmm1
+	vaesenc	%xmm15,%xmm14,%xmm14
+
+	vmovups	64-128(%rcx),%xmm15
+	vpxor	%xmm2,%xmm6,%xmm6
+	vpclmulqdq	$0x00,%xmm1,%xmm0,%xmm2
+	vaesenc	%xmm15,%xmm9,%xmm9
+	vpxor	%xmm3,%xmm6,%xmm6
+	vpclmulqdq	$0x10,%xmm1,%xmm0,%xmm3
+	vaesenc	%xmm15,%xmm10,%xmm10
+	movbeq	72(%r14),%r13
+	vpxor	%xmm5,%xmm7,%xmm7
+	vpclmulqdq	$0x01,%xmm1,%xmm0,%xmm5
+	vaesenc	%xmm15,%xmm11,%xmm11
+	movbeq	64(%r14),%r12
+	vpclmulqdq	$0x11,%xmm1,%xmm0,%xmm1
+	vmovdqu	96+8(%rsp),%xmm0
+	vaesenc	%xmm15,%xmm12,%xmm12
+	movq	%r13,48+8(%rsp)
+	vaesenc	%xmm15,%xmm13,%xmm13
+	movq	%r12,56+8(%rsp)
+	vpxor	%xmm2,%xmm4,%xmm4
+	vmovdqu	96-32(%r9),%xmm2
+	vaesenc	%xmm15,%xmm14,%xmm14
+
+	vmovups	80-128(%rcx),%xmm15
+	vpxor	%xmm3,%xmm6,%xmm6
+	vpclmulqdq	$0x00,%xmm2,%xmm0,%xmm3
+	vaesenc	%xmm15,%xmm9,%xmm9
+	vpxor	%xmm5,%xmm6,%xmm6
+	vpclmulqdq	$0x10,%xmm2,%xmm0,%xmm5
+	vaesenc	%xmm15,%xmm10,%xmm10
+	movbeq	56(%r14),%r13
+	vpxor	%xmm1,%xmm7,%xmm7
+	vpclmulqdq	$0x01,%xmm2,%xmm0,%xmm1
+	vpxor	112+8(%rsp),%xmm8,%xmm8
+	vaesenc	%xmm15,%xmm11,%xmm11
+	movbeq	48(%r14),%r12
+	vpclmulqdq	$0x11,%xmm2,%xmm0,%xmm2
+	vaesenc	%xmm15,%xmm12,%xmm12
+	movq	%r13,64+8(%rsp)
+	vaesenc	%xmm15,%xmm13,%xmm13
+	movq	%r12,72+8(%rsp)
+	vpxor	%xmm3,%xmm4,%xmm4
+	vmovdqu	112-32(%r9),%xmm3
+	vaesenc	%xmm15,%xmm14,%xmm14
+
+	vmovups	96-128(%rcx),%xmm15
+	vpxor	%xmm5,%xmm6,%xmm6
+	vpclmulqdq	$0x10,%xmm3,%xmm8,%xmm5
+	vaesenc	%xmm15,%xmm9,%xmm9
+	vpxor	%xmm1,%xmm6,%xmm6
+	vpclmulqdq	$0x01,%xmm3,%xmm8,%xmm1
+	vaesenc	%xmm15,%xmm10,%xmm10
+	movbeq	40(%r14),%r13
+	vpxor	%xmm2,%xmm7,%xmm7
+	vpclmulqdq	$0x00,%xmm3,%xmm8,%xmm2
+	vaesenc	%xmm15,%xmm11,%xmm11
+	movbeq	32(%r14),%r12
+	vpclmulqdq	$0x11,%xmm3,%xmm8,%xmm8
+	vaesenc	%xmm15,%xmm12,%xmm12
+	movq	%r13,80+8(%rsp)
+	vaesenc	%xmm15,%xmm13,%xmm13
+	movq	%r12,88+8(%rsp)
+	vpxor	%xmm5,%xmm6,%xmm6
+	vaesenc	%xmm15,%xmm14,%xmm14
+	vpxor	%xmm1,%xmm6,%xmm6
+
+	vmovups	112-128(%rcx),%xmm15
+	vpslldq	$8,%xmm6,%xmm5
+	vpxor	%xmm2,%xmm4,%xmm4
+	vmovdqu	16(%r11),%xmm3
+
+	vaesenc	%xmm15,%xmm9,%xmm9
+	vpxor	%xmm8,%xmm7,%xmm7
+	vaesenc	%xmm15,%xmm10,%xmm10
+	vpxor	%xmm5,%xmm4,%xmm4
+	movbeq	24(%r14),%r13
+	vaesenc	%xmm15,%xmm11,%xmm11
+	movbeq	16(%r14),%r12
+	vpalignr	$8,%xmm4,%xmm4,%xmm0
+	vpclmulqdq	$0x10,%xmm3,%xmm4,%xmm4
+	movq	%r13,96+8(%rsp)
+	vaesenc	%xmm15,%xmm12,%xmm12
+	movq	%r12,104+8(%rsp)
+	vaesenc	%xmm15,%xmm13,%xmm13
+	vmovups	128-128(%rcx),%xmm1
+	vaesenc	%xmm15,%xmm14,%xmm14
+
+	vaesenc	%xmm1,%xmm9,%xmm9
+	vmovups	144-128(%rcx),%xmm15
+	vaesenc	%xmm1,%xmm10,%xmm10
+	vpsrldq	$8,%xmm6,%xmm6
+	vaesenc	%xmm1,%xmm11,%xmm11
+	vpxor	%xmm6,%xmm7,%xmm7
+	vaesenc	%xmm1,%xmm12,%xmm12
+	vpxor	%xmm0,%xmm4,%xmm4
+	movbeq	8(%r14),%r13
+	vaesenc	%xmm1,%xmm13,%xmm13
+	movbeq	0(%r14),%r12
+	vaesenc	%xmm1,%xmm14,%xmm14
+	vmovups	160-128(%rcx),%xmm1
+	cmpl	$11,%ebp
+	jb	.Lenc_tail
+
+	vaesenc	%xmm15,%xmm9,%xmm9
+	vaesenc	%xmm15,%xmm10,%xmm10
+	vaesenc	%xmm15,%xmm11,%xmm11
+	vaesenc	%xmm15,%xmm12,%xmm12
+	vaesenc	%xmm15,%xmm13,%xmm13
+	vaesenc	%xmm15,%xmm14,%xmm14
+
+	vaesenc	%xmm1,%xmm9,%xmm9
+	vaesenc	%xmm1,%xmm10,%xmm10
+	vaesenc	%xmm1,%xmm11,%xmm11
+	vaesenc	%xmm1,%xmm12,%xmm12
+	vaesenc	%xmm1,%xmm13,%xmm13
+	vmovups	176-128(%rcx),%xmm15
+	vaesenc	%xmm1,%xmm14,%xmm14
+	vmovups	192-128(%rcx),%xmm1
+	je	.Lenc_tail
 
+	vaesenc	%xmm15,%xmm9,%xmm9
+	vaesenc	%xmm15,%xmm10,%xmm10
+	vaesenc	%xmm15,%xmm11,%xmm11
+	vaesenc	%xmm15,%xmm12,%xmm12
+	vaesenc	%xmm15,%xmm13,%xmm13
+	vaesenc	%xmm15,%xmm14,%xmm14
+
+	vaesenc	%xmm1,%xmm9,%xmm9
+	vaesenc	%xmm1,%xmm10,%xmm10
+	vaesenc	%xmm1,%xmm11,%xmm11
+	vaesenc	%xmm1,%xmm12,%xmm12
+	vaesenc	%xmm1,%xmm13,%xmm13
+	vmovups	208-128(%rcx),%xmm15
+	vaesenc	%xmm1,%xmm14,%xmm14
+	vmovups	224-128(%rcx),%xmm1
+	jmp	.Lenc_tail
+
+.align	32
+.Lhandle_ctr32:
+	vmovdqu	(%r11),%xmm0
+	vpshufb	%xmm0,%xmm1,%xmm6
+	vmovdqu	48(%r11),%xmm5
+	vpaddd	64(%r11),%xmm6,%xmm10
+	vpaddd	%xmm5,%xmm6,%xmm11
+	vmovdqu	0-32(%r9),%xmm3
+	vpaddd	%xmm5,%xmm10,%xmm12
+	vpshufb	%xmm0,%xmm10,%xmm10
+	vpaddd	%xmm5,%xmm11,%xmm13
+	vpshufb	%xmm0,%xmm11,%xmm11
+	vpxor	%xmm15,%xmm10,%xmm10
+	vpaddd	%xmm5,%xmm12,%xmm14
+	vpshufb	%xmm0,%xmm12,%xmm12
+	vpxor	%xmm15,%xmm11,%xmm11
+	vpaddd	%xmm5,%xmm13,%xmm1
+	vpshufb	%xmm0,%xmm13,%xmm13
+	vpshufb	%xmm0,%xmm14,%xmm14
+	vpshufb	%xmm0,%xmm1,%xmm1
+	jmp	.Lresume_ctr32
+
+.align	32
+.Lenc_tail:
+	vaesenc	%xmm15,%xmm9,%xmm9
+	vmovdqu	%xmm7,16+8(%rsp)
+	vpalignr	$8,%xmm4,%xmm4,%xmm8
+	vaesenc	%xmm15,%xmm10,%xmm10
+	vpclmulqdq	$0x10,%xmm3,%xmm4,%xmm4
+	vpxor	0(%rdi),%xmm1,%xmm2
+	vaesenc	%xmm15,%xmm11,%xmm11
+	vpxor	16(%rdi),%xmm1,%xmm0
+	vaesenc	%xmm15,%xmm12,%xmm12
+	vpxor	32(%rdi),%xmm1,%xmm5
+	vaesenc	%xmm15,%xmm13,%xmm13
+	vpxor	48(%rdi),%xmm1,%xmm6
+	vaesenc	%xmm15,%xmm14,%xmm14
+	vpxor	64(%rdi),%xmm1,%xmm7
+	vpxor	80(%rdi),%xmm1,%xmm3
+	vmovdqu	(%r8),%xmm1
+
+	vaesenclast	%xmm2,%xmm9,%xmm9
+	vmovdqu	32(%r11),%xmm2
+	vaesenclast	%xmm0,%xmm10,%xmm10
+	vpaddb	%xmm2,%xmm1,%xmm0
+	movq	%r13,112+8(%rsp)
+	leaq	96(%rdi),%rdi
+	vaesenclast	%xmm5,%xmm11,%xmm11
+	vpaddb	%xmm2,%xmm0,%xmm5
+	movq	%r12,120+8(%rsp)
+	leaq	96(%rsi),%rsi
+	vmovdqu	0-128(%rcx),%xmm15
+	vaesenclast	%xmm6,%xmm12,%xmm12
+	vpaddb	%xmm2,%xmm5,%xmm6
+	vaesenclast	%xmm7,%xmm13,%xmm13
+	vpaddb	%xmm2,%xmm6,%xmm7
+	vaesenclast	%xmm3,%xmm14,%xmm14
+	vpaddb	%xmm2,%xmm7,%xmm3
+
+	addq	$0x60,%r10
+	subq	$0x6,%rdx
+	jc	.L6x_done
+
+	vmovups	%xmm9,-96(%rsi)
+	vpxor	%xmm15,%xmm1,%xmm9
+	vmovups	%xmm10,-80(%rsi)
+	vmovdqa	%xmm0,%xmm10
+	vmovups	%xmm11,-64(%rsi)
+	vmovdqa	%xmm5,%xmm11
+	vmovups	%xmm12,-48(%rsi)
+	vmovdqa	%xmm6,%xmm12
+	vmovups	%xmm13,-32(%rsi)
+	vmovdqa	%xmm7,%xmm13
+	vmovups	%xmm14,-16(%rsi)
+	vmovdqa	%xmm3,%xmm14
+	vmovdqu	32+8(%rsp),%xmm7
+	jmp	.Loop6x
+
+.L6x_done:
+	vpxor	16+8(%rsp),%xmm8,%xmm8
+	vpxor	%xmm4,%xmm8,%xmm8
+
+	.byte	0xf3,0xc3
+.size	_aesni_ctr32_ghash_6x,.-_aesni_ctr32_ghash_6x
 .globl	aesni_gcm_decrypt
 .type	aesni_gcm_decrypt,@function
+.align	32
 aesni_gcm_decrypt:
-	xorl	%eax,%eax
+	xorq	%r10,%r10
+	cmpq	$0x60,%rdx
+	jb	.Lgcm_dec_abort
+
+	leaq	(%rsp),%rax
+	pushq	%rbx
+	pushq	%rbp
+	pushq	%r12
+	pushq	%r13
+	pushq	%r14
+	pushq	%r15
+	vzeroupper
+
+	vmovdqu	(%r8),%xmm1
+	addq	$-128,%rsp
+	movl	12(%r8),%ebx
+	leaq	.Lbswap_mask(%rip),%r11
+	leaq	-128(%rcx),%r14
+	movq	$0xf80,%r15
+	vmovdqu	(%r9),%xmm8
+	andq	$-128,%rsp
+	vmovdqu	(%r11),%xmm0
+	leaq	128(%rcx),%rcx
+	leaq	32+32(%r9),%r9
+	movl	240-128(%rcx),%ebp
+	vpshufb	%xmm0,%xmm8,%xmm8
+
+	andq	%r15,%r14
+	andq	%rsp,%r15
+	subq	%r14,%r15
+	jc	.Ldec_no_key_aliasing
+	cmpq	$768,%r15
+	jnc	.Ldec_no_key_aliasing
+	subq	%r15,%rsp
+.Ldec_no_key_aliasing:
+
+	vmovdqu	80(%rdi),%xmm7
+	leaq	(%rdi),%r14
+	vmovdqu	64(%rdi),%xmm4
+	leaq	-192(%rdi,%rdx,1),%r15
+	vmovdqu	48(%rdi),%xmm5
+	shrq	$4,%rdx
+	xorq	%r10,%r10
+	vmovdqu	32(%rdi),%xmm6
+	vpshufb	%xmm0,%xmm7,%xmm7
+	vmovdqu	16(%rdi),%xmm2
+	vpshufb	%xmm0,%xmm4,%xmm4
+	vmovdqu	(%rdi),%xmm3
+	vpshufb	%xmm0,%xmm5,%xmm5
+	vmovdqu	%xmm4,48(%rsp)
+	vpshufb	%xmm0,%xmm6,%xmm6
+	vmovdqu	%xmm5,64(%rsp)
+	vpshufb	%xmm0,%xmm2,%xmm2
+	vmovdqu	%xmm6,80(%rsp)
+	vpshufb	%xmm0,%xmm3,%xmm3
+	vmovdqu	%xmm2,96(%rsp)
+	vmovdqu	%xmm3,112(%rsp)
+
+	call	_aesni_ctr32_ghash_6x
+
+	vmovups	%xmm9,-96(%rsi)
+	vmovups	%xmm10,-80(%rsi)
+	vmovups	%xmm11,-64(%rsi)
+	vmovups	%xmm12,-48(%rsi)
+	vmovups	%xmm13,-32(%rsi)
+	vmovups	%xmm14,-16(%rsi)
+
+	vpshufb	(%r11),%xmm8,%xmm8
+	vmovdqu	%xmm8,-64(%r9)
+
+	vzeroupper
+	movq	-48(%rax),%r15
+	movq	-40(%rax),%r14
+	movq	-32(%rax),%r13
+	movq	-24(%rax),%r12
+	movq	-16(%rax),%rbp
+	movq	-8(%rax),%rbx
+	leaq	(%rax),%rsp
+.Lgcm_dec_abort:
+	movq	%r10,%rax
 	.byte	0xf3,0xc3
 .size	aesni_gcm_decrypt,.-aesni_gcm_decrypt
+.type	_aesni_ctr32_6x,@function
+.align	32
+_aesni_ctr32_6x:
+	vmovdqu	0-128(%rcx),%xmm4
+	vmovdqu	32(%r11),%xmm2
+	leaq	-1(%rbp),%r13
+	vmovups	16-128(%rcx),%xmm15
+	leaq	32-128(%rcx),%r12
+	vpxor	%xmm4,%xmm1,%xmm9
+	addl	$100663296,%ebx
+	jc	.Lhandle_ctr32_2
+	vpaddb	%xmm2,%xmm1,%xmm10
+	vpaddb	%xmm2,%xmm10,%xmm11
+	vpxor	%xmm4,%xmm10,%xmm10
+	vpaddb	%xmm2,%xmm11,%xmm12
+	vpxor	%xmm4,%xmm11,%xmm11
+	vpaddb	%xmm2,%xmm12,%xmm13
+	vpxor	%xmm4,%xmm12,%xmm12
+	vpaddb	%xmm2,%xmm13,%xmm14
+	vpxor	%xmm4,%xmm13,%xmm13
+	vpaddb	%xmm2,%xmm14,%xmm1
+	vpxor	%xmm4,%xmm14,%xmm14
+	jmp	.Loop_ctr32
+
+.align	16
+.Loop_ctr32:
+	vaesenc	%xmm15,%xmm9,%xmm9
+	vaesenc	%xmm15,%xmm10,%xmm10
+	vaesenc	%xmm15,%xmm11,%xmm11
+	vaesenc	%xmm15,%xmm12,%xmm12
+	vaesenc	%xmm15,%xmm13,%xmm13
+	vaesenc	%xmm15,%xmm14,%xmm14
+	vmovups	(%r12),%xmm15
+	leaq	16(%r12),%r12
+	decl	%r13d
+	jnz	.Loop_ctr32
+
+	vmovdqu	(%r12),%xmm3
+	vaesenc	%xmm15,%xmm9,%xmm9
+	vpxor	0(%rdi),%xmm3,%xmm4
+	vaesenc	%xmm15,%xmm10,%xmm10
+	vpxor	16(%rdi),%xmm3,%xmm5
+	vaesenc	%xmm15,%xmm11,%xmm11
+	vpxor	32(%rdi),%xmm3,%xmm6
+	vaesenc	%xmm15,%xmm12,%xmm12
+	vpxor	48(%rdi),%xmm3,%xmm8
+	vaesenc	%xmm15,%xmm13,%xmm13
+	vpxor	64(%rdi),%xmm3,%xmm2
+	vaesenc	%xmm15,%xmm14,%xmm14
+	vpxor	80(%rdi),%xmm3,%xmm3
+	leaq	96(%rdi),%rdi
+
+	vaesenclast	%xmm4,%xmm9,%xmm9
+	vaesenclast	%xmm5,%xmm10,%xmm10
+	vaesenclast	%xmm6,%xmm11,%xmm11
+	vaesenclast	%xmm8,%xmm12,%xmm12
+	vaesenclast	%xmm2,%xmm13,%xmm13
+	vaesenclast	%xmm3,%xmm14,%xmm14
+	vmovups	%xmm9,0(%rsi)
+	vmovups	%xmm10,16(%rsi)
+	vmovups	%xmm11,32(%rsi)
+	vmovups	%xmm12,48(%rsi)
+	vmovups	%xmm13,64(%rsi)
+	vmovups	%xmm14,80(%rsi)
+	leaq	96(%rsi),%rsi
+
+	.byte	0xf3,0xc3
+.align	32
+.Lhandle_ctr32_2:
+	vpshufb	%xmm0,%xmm1,%xmm6
+	vmovdqu	48(%r11),%xmm5
+	vpaddd	64(%r11),%xmm6,%xmm10
+	vpaddd	%xmm5,%xmm6,%xmm11
+	vpaddd	%xmm5,%xmm10,%xmm12
+	vpshufb	%xmm0,%xmm10,%xmm10
+	vpaddd	%xmm5,%xmm11,%xmm13
+	vpshufb	%xmm0,%xmm11,%xmm11
+	vpxor	%xmm4,%xmm10,%xmm10
+	vpaddd	%xmm5,%xmm12,%xmm14
+	vpshufb	%xmm0,%xmm12,%xmm12
+	vpxor	%xmm4,%xmm11,%xmm11
+	vpaddd	%xmm5,%xmm13,%xmm1
+	vpshufb	%xmm0,%xmm13,%xmm13
+	vpxor	%xmm4,%xmm12,%xmm12
+	vpshufb	%xmm0,%xmm14,%xmm14
+	vpxor	%xmm4,%xmm13,%xmm13
+	vpshufb	%xmm0,%xmm1,%xmm1
+	vpxor	%xmm4,%xmm14,%xmm14
+	jmp	.Loop_ctr32
+.size	_aesni_ctr32_6x,.-_aesni_ctr32_6x
+
+.globl	aesni_gcm_encrypt
+.type	aesni_gcm_encrypt,@function
+.align	32
+aesni_gcm_encrypt:
+	xorq	%r10,%r10
+	cmpq	$288,%rdx
+	jb	.Lgcm_enc_abort
+
+	leaq	(%rsp),%rax
+	pushq	%rbx
+	pushq	%rbp
+	pushq	%r12
+	pushq	%r13
+	pushq	%r14
+	pushq	%r15
+	vzeroupper
+
+	vmovdqu	(%r8),%xmm1
+	addq	$-128,%rsp
+	movl	12(%r8),%ebx
+	leaq	.Lbswap_mask(%rip),%r11
+	leaq	-128(%rcx),%r14
+	movq	$0xf80,%r15
+	leaq	128(%rcx),%rcx
+	vmovdqu	(%r11),%xmm0
+	andq	$-128,%rsp
+	movl	240-128(%rcx),%ebp
+
+	andq	%r15,%r14
+	andq	%rsp,%r15
+	subq	%r14,%r15
+	jc	.Lenc_no_key_aliasing
+	cmpq	$768,%r15
+	jnc	.Lenc_no_key_aliasing
+	subq	%r15,%rsp
+.Lenc_no_key_aliasing:
+
+	leaq	(%rsi),%r14
+	leaq	-192(%rsi,%rdx,1),%r15
+	shrq	$4,%rdx
+
+	call	_aesni_ctr32_6x
+	vpshufb	%xmm0,%xmm9,%xmm8
+	vpshufb	%xmm0,%xmm10,%xmm2
+	vmovdqu	%xmm8,112(%rsp)
+	vpshufb	%xmm0,%xmm11,%xmm4
+	vmovdqu	%xmm2,96(%rsp)
+	vpshufb	%xmm0,%xmm12,%xmm5
+	vmovdqu	%xmm4,80(%rsp)
+	vpshufb	%xmm0,%xmm13,%xmm6
+	vmovdqu	%xmm5,64(%rsp)
+	vpshufb	%xmm0,%xmm14,%xmm7
+	vmovdqu	%xmm6,48(%rsp)
+
+	call	_aesni_ctr32_6x
+
+	vmovdqu	(%r9),%xmm8
+	leaq	32+32(%r9),%r9
+	subq	$12,%rdx
+	movq	$192,%r10
+	vpshufb	%xmm0,%xmm8,%xmm8
+
+	call	_aesni_ctr32_ghash_6x
+	vmovdqu	32(%rsp),%xmm7
+	vmovdqu	(%r11),%xmm0
+	vmovdqu	0-32(%r9),%xmm3
+	vpunpckhqdq	%xmm7,%xmm7,%xmm1
+	vmovdqu	32-32(%r9),%xmm15
+	vmovups	%xmm9,-96(%rsi)
+	vpshufb	%xmm0,%xmm9,%xmm9
+	vpxor	%xmm7,%xmm1,%xmm1
+	vmovups	%xmm10,-80(%rsi)
+	vpshufb	%xmm0,%xmm10,%xmm10
+	vmovups	%xmm11,-64(%rsi)
+	vpshufb	%xmm0,%xmm11,%xmm11
+	vmovups	%xmm12,-48(%rsi)
+	vpshufb	%xmm0,%xmm12,%xmm12
+	vmovups	%xmm13,-32(%rsi)
+	vpshufb	%xmm0,%xmm13,%xmm13
+	vmovups	%xmm14,-16(%rsi)
+	vpshufb	%xmm0,%xmm14,%xmm14
+	vmovdqu	%xmm9,16(%rsp)
+	vmovdqu	48(%rsp),%xmm6
+	vmovdqu	16-32(%r9),%xmm0
+	vpunpckhqdq	%xmm6,%xmm6,%xmm2
+	vpclmulqdq	$0x00,%xmm3,%xmm7,%xmm5
+	vpxor	%xmm6,%xmm2,%xmm2
+	vpclmulqdq	$0x11,%xmm3,%xmm7,%xmm7
+	vpclmulqdq	$0x00,%xmm15,%xmm1,%xmm1
+
+	vmovdqu	64(%rsp),%xmm9
+	vpclmulqdq	$0x00,%xmm0,%xmm6,%xmm4
+	vmovdqu	48-32(%r9),%xmm3
+	vpxor	%xmm5,%xmm4,%xmm4
+	vpunpckhqdq	%xmm9,%xmm9,%xmm5
+	vpclmulqdq	$0x11,%xmm0,%xmm6,%xmm6
+	vpxor	%xmm9,%xmm5,%xmm5
+	vpxor	%xmm7,%xmm6,%xmm6
+	vpclmulqdq	$0x10,%xmm15,%xmm2,%xmm2
+	vmovdqu	80-32(%r9),%xmm15
+	vpxor	%xmm1,%xmm2,%xmm2
+
+	vmovdqu	80(%rsp),%xmm1
+	vpclmulqdq	$0x00,%xmm3,%xmm9,%xmm7
+	vmovdqu	64-32(%r9),%xmm0
+	vpxor	%xmm4,%xmm7,%xmm7
+	vpunpckhqdq	%xmm1,%xmm1,%xmm4
+	vpclmulqdq	$0x11,%xmm3,%xmm9,%xmm9
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpxor	%xmm6,%xmm9,%xmm9
+	vpclmulqdq	$0x00,%xmm15,%xmm5,%xmm5
+	vpxor	%xmm2,%xmm5,%xmm5
+
+	vmovdqu	96(%rsp),%xmm2
+	vpclmulqdq	$0x00,%xmm0,%xmm1,%xmm6
+	vmovdqu	96-32(%r9),%xmm3
+	vpxor	%xmm7,%xmm6,%xmm6
+	vpunpckhqdq	%xmm2,%xmm2,%xmm7
+	vpclmulqdq	$0x11,%xmm0,%xmm1,%xmm1
+	vpxor	%xmm2,%xmm7,%xmm7
+	vpxor	%xmm9,%xmm1,%xmm1
+	vpclmulqdq	$0x10,%xmm15,%xmm4,%xmm4
+	vmovdqu	128-32(%r9),%xmm15
+	vpxor	%xmm5,%xmm4,%xmm4
+
+	vpxor	112(%rsp),%xmm8,%xmm8
+	vpclmulqdq	$0x00,%xmm3,%xmm2,%xmm5
+	vmovdqu	112-32(%r9),%xmm0
+	vpunpckhqdq	%xmm8,%xmm8,%xmm9
+	vpxor	%xmm6,%xmm5,%xmm5
+	vpclmulqdq	$0x11,%xmm3,%xmm2,%xmm2
+	vpxor	%xmm8,%xmm9,%xmm9
+	vpxor	%xmm1,%xmm2,%xmm2
+	vpclmulqdq	$0x00,%xmm15,%xmm7,%xmm7
+	vpxor	%xmm4,%xmm7,%xmm4
+
+	vpclmulqdq	$0x00,%xmm0,%xmm8,%xmm6
+	vmovdqu	0-32(%r9),%xmm3
+	vpunpckhqdq	%xmm14,%xmm14,%xmm1
+	vpclmulqdq	$0x11,%xmm0,%xmm8,%xmm8
+	vpxor	%xmm14,%xmm1,%xmm1
+	vpxor	%xmm5,%xmm6,%xmm5
+	vpclmulqdq	$0x10,%xmm15,%xmm9,%xmm9
+	vmovdqu	32-32(%r9),%xmm15
+	vpxor	%xmm2,%xmm8,%xmm7
+	vpxor	%xmm4,%xmm9,%xmm6
+
+	vmovdqu	16-32(%r9),%xmm0
+	vpxor	%xmm5,%xmm7,%xmm9
+	vpclmulqdq	$0x00,%xmm3,%xmm14,%xmm4
+	vpxor	%xmm9,%xmm6,%xmm6
+	vpunpckhqdq	%xmm13,%xmm13,%xmm2
+	vpclmulqdq	$0x11,%xmm3,%xmm14,%xmm14
+	vpxor	%xmm13,%xmm2,%xmm2
+	vpslldq	$8,%xmm6,%xmm9
+	vpclmulqdq	$0x00,%xmm15,%xmm1,%xmm1
+	vpxor	%xmm9,%xmm5,%xmm8
+	vpsrldq	$8,%xmm6,%xmm6
+	vpxor	%xmm6,%xmm7,%xmm7
+
+	vpclmulqdq	$0x00,%xmm0,%xmm13,%xmm5
+	vmovdqu	48-32(%r9),%xmm3
+	vpxor	%xmm4,%xmm5,%xmm5
+	vpunpckhqdq	%xmm12,%xmm12,%xmm9
+	vpclmulqdq	$0x11,%xmm0,%xmm13,%xmm13
+	vpxor	%xmm12,%xmm9,%xmm9
+	vpxor	%xmm14,%xmm13,%xmm13
+	vpalignr	$8,%xmm8,%xmm8,%xmm14
+	vpclmulqdq	$0x10,%xmm15,%xmm2,%xmm2
+	vmovdqu	80-32(%r9),%xmm15
+	vpxor	%xmm1,%xmm2,%xmm2
+
+	vpclmulqdq	$0x00,%xmm3,%xmm12,%xmm4
+	vmovdqu	64-32(%r9),%xmm0
+	vpxor	%xmm5,%xmm4,%xmm4
+	vpunpckhqdq	%xmm11,%xmm11,%xmm1
+	vpclmulqdq	$0x11,%xmm3,%xmm12,%xmm12
+	vpxor	%xmm11,%xmm1,%xmm1
+	vpxor	%xmm13,%xmm12,%xmm12
+	vxorps	16(%rsp),%xmm7,%xmm7
+	vpclmulqdq	$0x00,%xmm15,%xmm9,%xmm9
+	vpxor	%xmm2,%xmm9,%xmm9
+
+	vpclmulqdq	$0x10,16(%r11),%xmm8,%xmm8
+	vxorps	%xmm14,%xmm8,%xmm8
+
+	vpclmulqdq	$0x00,%xmm0,%xmm11,%xmm5
+	vmovdqu	96-32(%r9),%xmm3
+	vpxor	%xmm4,%xmm5,%xmm5
+	vpunpckhqdq	%xmm10,%xmm10,%xmm2
+	vpclmulqdq	$0x11,%xmm0,%xmm11,%xmm11
+	vpxor	%xmm10,%xmm2,%xmm2
+	vpalignr	$8,%xmm8,%xmm8,%xmm14
+	vpxor	%xmm12,%xmm11,%xmm11
+	vpclmulqdq	$0x10,%xmm15,%xmm1,%xmm1
+	vmovdqu	128-32(%r9),%xmm15
+	vpxor	%xmm9,%xmm1,%xmm1
+
+	vxorps	%xmm7,%xmm14,%xmm14
+	vpclmulqdq	$0x10,16(%r11),%xmm8,%xmm8
+	vxorps	%xmm14,%xmm8,%xmm8
+
+	vpclmulqdq	$0x00,%xmm3,%xmm10,%xmm4
+	vmovdqu	112-32(%r9),%xmm0
+	vpxor	%xmm5,%xmm4,%xmm4
+	vpunpckhqdq	%xmm8,%xmm8,%xmm9
+	vpclmulqdq	$0x11,%xmm3,%xmm10,%xmm10
+	vpxor	%xmm8,%xmm9,%xmm9
+	vpxor	%xmm11,%xmm10,%xmm10
+	vpclmulqdq	$0x00,%xmm15,%xmm2,%xmm2
+	vpxor	%xmm1,%xmm2,%xmm2
+
+	vpclmulqdq	$0x00,%xmm0,%xmm8,%xmm5
+	vpclmulqdq	$0x11,%xmm0,%xmm8,%xmm7
+	vpxor	%xmm4,%xmm5,%xmm5
+	vpclmulqdq	$0x10,%xmm15,%xmm9,%xmm6
+	vpxor	%xmm10,%xmm7,%xmm7
+	vpxor	%xmm2,%xmm6,%xmm6
+
+	vpxor	%xmm5,%xmm7,%xmm4
+	vpxor	%xmm4,%xmm6,%xmm6
+	vpslldq	$8,%xmm6,%xmm1
+	vmovdqu	16(%r11),%xmm3
+	vpsrldq	$8,%xmm6,%xmm6
+	vpxor	%xmm1,%xmm5,%xmm8
+	vpxor	%xmm6,%xmm7,%xmm7
+
+	vpalignr	$8,%xmm8,%xmm8,%xmm2
+	vpclmulqdq	$0x10,%xmm3,%xmm8,%xmm8
+	vpxor	%xmm2,%xmm8,%xmm8
+
+	vpalignr	$8,%xmm8,%xmm8,%xmm2
+	vpclmulqdq	$0x10,%xmm3,%xmm8,%xmm8
+	vpxor	%xmm7,%xmm2,%xmm2
+	vpxor	%xmm2,%xmm8,%xmm8
+	vpshufb	(%r11),%xmm8,%xmm8
+	vmovdqu	%xmm8,-64(%r9)
+
+	vzeroupper
+	movq	-48(%rax),%r15
+	movq	-40(%rax),%r14
+	movq	-32(%rax),%r13
+	movq	-24(%rax),%r12
+	movq	-16(%rax),%rbp
+	movq	-8(%rax),%rbx
+	leaq	(%rax),%rsp
+.Lgcm_enc_abort:
+	movq	%r10,%rax
+	.byte	0xf3,0xc3
+.size	aesni_gcm_encrypt,.-aesni_gcm_encrypt
+.align	64
+.Lbswap_mask:
+.byte	15,14,13,12,11,10,9,8,7,6,5,4,3,2,1,0
+.Lpoly:
+.byte	0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0xc2
+.Lone_msb:
+.byte	0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1
+.Ltwo_lsb:
+.byte	2,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0
+.Lone_lsb:
+.byte	1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0
+.byte	65,69,83,45,78,73,32,71,67,77,32,109,111,100,117,108,101,32,102,111,114,32,120,56,54,95,54,52,44,32,67,82,89,80,84,79,71,65,77,83,32,98,121,32,60,97,112,112,114,111,64,111,112,101,110,115,115,108,46,111,114,103,62,0
+.align	64

deps/openssl/asm/x64-elf-gas/modes/ghash-x86_64.s

@@ -661,10 +661,10 @@ gcm_ghash_4bit:
 gcm_init_clmul:
 .L_init_clmul:
 	movdqu	(%rsi),%xmm2
-	pshufd	$0b01001110,%xmm2,%xmm2
+	pshufd	$78,%xmm2,%xmm2
 
 
-	pshufd	$0b11111111,%xmm2,%xmm4
+	pshufd	$255,%xmm2,%xmm4
 	movdqa	%xmm2,%xmm3
 	psllq	$1,%xmm2
 	pxor	%xmm5,%xmm5
@@ -678,11 +678,11 @@ gcm_init_clmul:
 	pxor	%xmm5,%xmm2
 
 
-	pshufd	$0b01001110,%xmm2,%xmm6
+	pshufd	$78,%xmm2,%xmm6
 	movdqa	%xmm2,%xmm0
 	pxor	%xmm2,%xmm6
 	movdqa	%xmm0,%xmm1
-	pshufd	$0b01001110,%xmm0,%xmm3
+	pshufd	$78,%xmm0,%xmm3
 	pxor	%xmm0,%xmm3
 .byte	102,15,58,68,194,0
 .byte	102,15,58,68,202,17
@@ -718,8 +718,8 @@ gcm_init_clmul:
 	pxor	%xmm4,%xmm0
 	psrlq	$1,%xmm0
 	pxor	%xmm1,%xmm0
-	pshufd	$0b01001110,%xmm2,%xmm3
-	pshufd	$0b01001110,%xmm0,%xmm4
+	pshufd	$78,%xmm2,%xmm3
+	pshufd	$78,%xmm0,%xmm4
 	pxor	%xmm2,%xmm3
 	movdqu	%xmm2,0(%rdi)
 	pxor	%xmm0,%xmm4
@@ -727,7 +727,7 @@ gcm_init_clmul:
 .byte	102,15,58,15,227,8
 	movdqu	%xmm4,32(%rdi)
 	movdqa	%xmm0,%xmm1
-	pshufd	$0b01001110,%xmm0,%xmm3
+	pshufd	$78,%xmm0,%xmm3
 	pxor	%xmm0,%xmm3
 .byte	102,15,58,68,194,0
 .byte	102,15,58,68,202,17
@@ -765,7 +765,7 @@ gcm_init_clmul:
 	pxor	%xmm1,%xmm0
 	movdqa	%xmm0,%xmm5
 	movdqa	%xmm0,%xmm1
-	pshufd	$0b01001110,%xmm0,%xmm3
+	pshufd	$78,%xmm0,%xmm3
 	pxor	%xmm0,%xmm3
 .byte	102,15,58,68,194,0
 .byte	102,15,58,68,202,17
@@ -801,8 +801,8 @@ gcm_init_clmul:
 	pxor	%xmm4,%xmm0
 	psrlq	$1,%xmm0
 	pxor	%xmm1,%xmm0
-	pshufd	$0b01001110,%xmm5,%xmm3
-	pshufd	$0b01001110,%xmm0,%xmm4
+	pshufd	$78,%xmm5,%xmm3
+	pshufd	$78,%xmm0,%xmm4
 	pxor	%xmm5,%xmm3
 	movdqu	%xmm5,48(%rdi)
 	pxor	%xmm0,%xmm4
@@ -822,7 +822,7 @@ gcm_gmult_clmul:
 	movdqu	32(%rsi),%xmm4
 .byte	102,15,56,0,197
 	movdqa	%xmm0,%xmm1
-	pshufd	$0b01001110,%xmm0,%xmm3
+	pshufd	$78,%xmm0,%xmm3
 	pxor	%xmm0,%xmm3
 .byte	102,15,58,68,194,0
 .byte	102,15,58,68,202,17
@@ -899,14 +899,14 @@ gcm_ghash_clmul:
 .byte	102,65,15,56,0,218
 .byte	102,69,15,56,0,218
 	movdqa	%xmm3,%xmm5
-	pshufd	$0b01001110,%xmm3,%xmm4
+	pshufd	$78,%xmm3,%xmm4
 	pxor	%xmm3,%xmm4
 .byte	102,15,58,68,218,0
 .byte	102,15,58,68,234,17
 .byte	102,15,58,68,231,0
 
 	movdqa	%xmm11,%xmm13
-	pshufd	$0b01001110,%xmm11,%xmm12
+	pshufd	$78,%xmm11,%xmm12
 	pxor	%xmm11,%xmm12
 .byte	102,68,15,58,68,222,0
 .byte	102,68,15,58,68,238,17
@@ -921,12 +921,12 @@ gcm_ghash_clmul:
 .byte	102,69,15,56,0,218
 .byte	102,69,15,56,0,194
 	movdqa	%xmm11,%xmm13
-	pshufd	$0b01001110,%xmm11,%xmm12
+	pshufd	$78,%xmm11,%xmm12
 	pxor	%xmm8,%xmm0
 	pxor	%xmm11,%xmm12
 .byte	102,69,15,58,68,222,0
 	movdqa	%xmm0,%xmm1
-	pshufd	$0b01001110,%xmm0,%xmm8
+	pshufd	$78,%xmm0,%xmm8
 	pxor	%xmm0,%xmm8
 .byte	102,69,15,58,68,238,17
 .byte	102,68,15,58,68,231,0
@@ -949,14 +949,14 @@ gcm_ghash_clmul:
 	movdqu	32(%rdx),%xmm3
 	movdqa	%xmm11,%xmm13
 .byte	102,68,15,58,68,199,16
-	pshufd	$0b01001110,%xmm11,%xmm12
+	pshufd	$78,%xmm11,%xmm12
 	xorps	%xmm5,%xmm1
 	pxor	%xmm11,%xmm12
 .byte	102,65,15,56,0,218
 	movups	32(%rsi),%xmm7
 	xorps	%xmm4,%xmm8
 .byte	102,68,15,58,68,218,0
-	pshufd	$0b01001110,%xmm3,%xmm4
+	pshufd	$78,%xmm3,%xmm4
 
 	pxor	%xmm0,%xmm8
 	movdqa	%xmm3,%xmm5
@@ -1000,7 +1000,7 @@ gcm_ghash_clmul:
 
 	movdqa	%xmm11,%xmm13
 	pxor	%xmm12,%xmm4
-	pshufd	$0b01001110,%xmm11,%xmm12
+	pshufd	$78,%xmm11,%xmm12
 	pxor	%xmm9,%xmm0
 	pxor	%xmm8,%xmm1
 	pxor	%xmm11,%xmm12
@@ -1010,7 +1010,7 @@ gcm_ghash_clmul:
 	movdqa	%xmm0,%xmm1
 .byte	102,69,15,58,68,238,17
 	xorps	%xmm11,%xmm3
-	pshufd	$0b01001110,%xmm0,%xmm8
+	pshufd	$78,%xmm0,%xmm8
 	pxor	%xmm0,%xmm8
 
 .byte	102,68,15,58,68,231,0
@@ -1079,7 +1079,7 @@ gcm_ghash_clmul:
 	pxor	%xmm8,%xmm0
 
 	movdqa	%xmm3,%xmm5
-	pshufd	$0b01001110,%xmm3,%xmm4
+	pshufd	$78,%xmm3,%xmm4
 	pxor	%xmm3,%xmm4
 .byte	102,15,58,68,218,0
 .byte	102,15,58,68,234,17
@@ -1096,7 +1096,7 @@ gcm_ghash_clmul:
 .Lmod_loop:
 	movdqa	%xmm0,%xmm1
 	movdqa	%xmm4,%xmm8
-	pshufd	$0b01001110,%xmm0,%xmm4
+	pshufd	$78,%xmm0,%xmm4
 	pxor	%xmm0,%xmm4
 
 .byte	102,15,58,68,198,0
@@ -1134,7 +1134,7 @@ gcm_ghash_clmul:
 	pslldq	$8,%xmm0
 	psrldq	$8,%xmm8
 	pxor	%xmm9,%xmm0
-	pshufd	$0b01001110,%xmm5,%xmm4
+	pshufd	$78,%xmm5,%xmm4
 	pxor	%xmm8,%xmm1
 	pxor	%xmm5,%xmm4
 
@@ -1156,7 +1156,7 @@ gcm_ghash_clmul:
 .Leven_tail:
 	movdqa	%xmm0,%xmm1
 	movdqa	%xmm4,%xmm8
-	pshufd	$0b01001110,%xmm0,%xmm4
+	pshufd	$78,%xmm0,%xmm4
 	pxor	%xmm0,%xmm4
 
 .byte	102,15,58,68,198,0
@@ -1204,7 +1204,7 @@ gcm_ghash_clmul:
 .byte	102,69,15,56,0,194
 	pxor	%xmm8,%xmm0
 	movdqa	%xmm0,%xmm1
-	pshufd	$0b01001110,%xmm0,%xmm3
+	pshufd	$78,%xmm0,%xmm3
 	pxor	%xmm0,%xmm3
 .byte	102,15,58,68,194,0
 .byte	102,15,58,68,202,17
@@ -1249,7 +1249,108 @@ gcm_ghash_clmul:
 .type	gcm_init_avx,@function
 .align	32
 gcm_init_avx:
-	jmp	.L_init_clmul
+	vzeroupper
+
+	vmovdqu	(%rsi),%xmm2
+	vpshufd	$78,%xmm2,%xmm2
+
+
+	vpshufd	$255,%xmm2,%xmm4
+	vpsrlq	$63,%xmm2,%xmm3
+	vpsllq	$1,%xmm2,%xmm2
+	vpxor	%xmm5,%xmm5,%xmm5
+	vpcmpgtd	%xmm4,%xmm5,%xmm5
+	vpslldq	$8,%xmm3,%xmm3
+	vpor	%xmm3,%xmm2,%xmm2
+
+
+	vpand	.L0x1c2_polynomial(%rip),%xmm5,%xmm5
+	vpxor	%xmm5,%xmm2,%xmm2
+
+	vpunpckhqdq	%xmm2,%xmm2,%xmm6
+	vmovdqa	%xmm2,%xmm0
+	vpxor	%xmm2,%xmm6,%xmm6
+	movq	$4,%r10
+	jmp	.Linit_start_avx
+.align	32
+.Linit_loop_avx:
+	vpalignr	$8,%xmm3,%xmm4,%xmm5
+	vmovdqu	%xmm5,-16(%rdi)
+	vpunpckhqdq	%xmm0,%xmm0,%xmm3
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x11,%xmm2,%xmm0,%xmm1
+	vpclmulqdq	$0x00,%xmm2,%xmm0,%xmm0
+	vpclmulqdq	$0x00,%xmm6,%xmm3,%xmm3
+	vpxor	%xmm0,%xmm1,%xmm4
+	vpxor	%xmm4,%xmm3,%xmm3
+
+	vpslldq	$8,%xmm3,%xmm4
+	vpsrldq	$8,%xmm3,%xmm3
+	vpxor	%xmm4,%xmm0,%xmm0
+	vpxor	%xmm3,%xmm1,%xmm1
+	vpsllq	$57,%xmm0,%xmm3
+	vpsllq	$62,%xmm0,%xmm4
+	vpxor	%xmm3,%xmm4,%xmm4
+	vpsllq	$63,%xmm0,%xmm3
+	vpxor	%xmm3,%xmm4,%xmm4
+	vpslldq	$8,%xmm4,%xmm3
+	vpsrldq	$8,%xmm4,%xmm4
+	vpxor	%xmm3,%xmm0,%xmm0
+	vpxor	%xmm4,%xmm1,%xmm1
+
+	vpsrlq	$1,%xmm0,%xmm4
+	vpxor	%xmm0,%xmm1,%xmm1
+	vpxor	%xmm4,%xmm0,%xmm0
+	vpsrlq	$5,%xmm4,%xmm4
+	vpxor	%xmm4,%xmm0,%xmm0
+	vpsrlq	$1,%xmm0,%xmm0
+	vpxor	%xmm1,%xmm0,%xmm0
+.Linit_start_avx:
+	vmovdqa	%xmm0,%xmm5
+	vpunpckhqdq	%xmm0,%xmm0,%xmm3
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x11,%xmm2,%xmm0,%xmm1
+	vpclmulqdq	$0x00,%xmm2,%xmm0,%xmm0
+	vpclmulqdq	$0x00,%xmm6,%xmm3,%xmm3
+	vpxor	%xmm0,%xmm1,%xmm4
+	vpxor	%xmm4,%xmm3,%xmm3
+
+	vpslldq	$8,%xmm3,%xmm4
+	vpsrldq	$8,%xmm3,%xmm3
+	vpxor	%xmm4,%xmm0,%xmm0
+	vpxor	%xmm3,%xmm1,%xmm1
+	vpsllq	$57,%xmm0,%xmm3
+	vpsllq	$62,%xmm0,%xmm4
+	vpxor	%xmm3,%xmm4,%xmm4
+	vpsllq	$63,%xmm0,%xmm3
+	vpxor	%xmm3,%xmm4,%xmm4
+	vpslldq	$8,%xmm4,%xmm3
+	vpsrldq	$8,%xmm4,%xmm4
+	vpxor	%xmm3,%xmm0,%xmm0
+	vpxor	%xmm4,%xmm1,%xmm1
+
+	vpsrlq	$1,%xmm0,%xmm4
+	vpxor	%xmm0,%xmm1,%xmm1
+	vpxor	%xmm4,%xmm0,%xmm0
+	vpsrlq	$5,%xmm4,%xmm4
+	vpxor	%xmm4,%xmm0,%xmm0
+	vpsrlq	$1,%xmm0,%xmm0
+	vpxor	%xmm1,%xmm0,%xmm0
+	vpshufd	$78,%xmm5,%xmm3
+	vpshufd	$78,%xmm0,%xmm4
+	vpxor	%xmm5,%xmm3,%xmm3
+	vmovdqu	%xmm5,0(%rdi)
+	vpxor	%xmm0,%xmm4,%xmm4
+	vmovdqu	%xmm0,16(%rdi)
+	leaq	48(%rdi),%rdi
+	subq	$1,%r10
+	jnz	.Linit_loop_avx
+
+	vpalignr	$8,%xmm4,%xmm3,%xmm5
+	vmovdqu	%xmm5,-16(%rdi)
+
+	vzeroupper
+	.byte	0xf3,0xc3
 .size	gcm_init_avx,.-gcm_init_avx
 .globl	gcm_gmult_avx
 .type	gcm_gmult_avx,@function
@@ -1261,7 +1362,377 @@ gcm_gmult_avx:
 .type	gcm_ghash_avx,@function
 .align	32
 gcm_ghash_avx:
-	jmp	.L_ghash_clmul
+	vzeroupper
+
+	vmovdqu	(%rdi),%xmm10
+	leaq	.L0x1c2_polynomial(%rip),%r10
+	leaq	64(%rsi),%rsi
+	vmovdqu	.Lbswap_mask(%rip),%xmm13
+	vpshufb	%xmm13,%xmm10,%xmm10
+	cmpq	$0x80,%rcx
+	jb	.Lshort_avx
+	subq	$0x80,%rcx
+
+	vmovdqu	112(%rdx),%xmm14
+	vmovdqu	0-64(%rsi),%xmm6
+	vpshufb	%xmm13,%xmm14,%xmm14
+	vmovdqu	32-64(%rsi),%xmm7
+
+	vpunpckhqdq	%xmm14,%xmm14,%xmm9
+	vmovdqu	96(%rdx),%xmm15
+	vpclmulqdq	$0x00,%xmm6,%xmm14,%xmm0
+	vpxor	%xmm14,%xmm9,%xmm9
+	vpshufb	%xmm13,%xmm15,%xmm15
+	vpclmulqdq	$0x11,%xmm6,%xmm14,%xmm1
+	vmovdqu	16-64(%rsi),%xmm6
+	vpunpckhqdq	%xmm15,%xmm15,%xmm8
+	vmovdqu	80(%rdx),%xmm14
+	vpclmulqdq	$0x00,%xmm7,%xmm9,%xmm2
+	vpxor	%xmm15,%xmm8,%xmm8
+
+	vpshufb	%xmm13,%xmm14,%xmm14
+	vpclmulqdq	$0x00,%xmm6,%xmm15,%xmm3
+	vpunpckhqdq	%xmm14,%xmm14,%xmm9
+	vpclmulqdq	$0x11,%xmm6,%xmm15,%xmm4
+	vmovdqu	48-64(%rsi),%xmm6
+	vpxor	%xmm14,%xmm9,%xmm9
+	vmovdqu	64(%rdx),%xmm15
+	vpclmulqdq	$0x10,%xmm7,%xmm8,%xmm5
+	vmovdqu	80-64(%rsi),%xmm7
+
+	vpshufb	%xmm13,%xmm15,%xmm15
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x00,%xmm6,%xmm14,%xmm0
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpunpckhqdq	%xmm15,%xmm15,%xmm8
+	vpclmulqdq	$0x11,%xmm6,%xmm14,%xmm1
+	vmovdqu	64-64(%rsi),%xmm6
+	vpxor	%xmm2,%xmm5,%xmm5
+	vpclmulqdq	$0x00,%xmm7,%xmm9,%xmm2
+	vpxor	%xmm15,%xmm8,%xmm8
+
+	vmovdqu	48(%rdx),%xmm14
+	vpxor	%xmm3,%xmm0,%xmm0
+	vpclmulqdq	$0x00,%xmm6,%xmm15,%xmm3
+	vpxor	%xmm4,%xmm1,%xmm1
+	vpshufb	%xmm13,%xmm14,%xmm14
+	vpclmulqdq	$0x11,%xmm6,%xmm15,%xmm4
+	vmovdqu	96-64(%rsi),%xmm6
+	vpxor	%xmm5,%xmm2,%xmm2
+	vpunpckhqdq	%xmm14,%xmm14,%xmm9
+	vpclmulqdq	$0x10,%xmm7,%xmm8,%xmm5
+	vmovdqu	128-64(%rsi),%xmm7
+	vpxor	%xmm14,%xmm9,%xmm9
+
+	vmovdqu	32(%rdx),%xmm15
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x00,%xmm6,%xmm14,%xmm0
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpshufb	%xmm13,%xmm15,%xmm15
+	vpclmulqdq	$0x11,%xmm6,%xmm14,%xmm1
+	vmovdqu	112-64(%rsi),%xmm6
+	vpxor	%xmm2,%xmm5,%xmm5
+	vpunpckhqdq	%xmm15,%xmm15,%xmm8
+	vpclmulqdq	$0x00,%xmm7,%xmm9,%xmm2
+	vpxor	%xmm15,%xmm8,%xmm8
+
+	vmovdqu	16(%rdx),%xmm14
+	vpxor	%xmm3,%xmm0,%xmm0
+	vpclmulqdq	$0x00,%xmm6,%xmm15,%xmm3
+	vpxor	%xmm4,%xmm1,%xmm1
+	vpshufb	%xmm13,%xmm14,%xmm14
+	vpclmulqdq	$0x11,%xmm6,%xmm15,%xmm4
+	vmovdqu	144-64(%rsi),%xmm6
+	vpxor	%xmm5,%xmm2,%xmm2
+	vpunpckhqdq	%xmm14,%xmm14,%xmm9
+	vpclmulqdq	$0x10,%xmm7,%xmm8,%xmm5
+	vmovdqu	176-64(%rsi),%xmm7
+	vpxor	%xmm14,%xmm9,%xmm9
+
+	vmovdqu	(%rdx),%xmm15
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x00,%xmm6,%xmm14,%xmm0
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpshufb	%xmm13,%xmm15,%xmm15
+	vpclmulqdq	$0x11,%xmm6,%xmm14,%xmm1
+	vmovdqu	160-64(%rsi),%xmm6
+	vpxor	%xmm2,%xmm5,%xmm5
+	vpclmulqdq	$0x10,%xmm7,%xmm9,%xmm2
+
+	leaq	128(%rdx),%rdx
+	cmpq	$0x80,%rcx
+	jb	.Ltail_avx
+
+	vpxor	%xmm10,%xmm15,%xmm15
+	subq	$0x80,%rcx
+	jmp	.Loop8x_avx
+
+.align	32
+.Loop8x_avx:
+	vpunpckhqdq	%xmm15,%xmm15,%xmm8
+	vmovdqu	112(%rdx),%xmm14
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpxor	%xmm15,%xmm8,%xmm8
+	vpclmulqdq	$0x00,%xmm6,%xmm15,%xmm10
+	vpshufb	%xmm13,%xmm14,%xmm14
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpclmulqdq	$0x11,%xmm6,%xmm15,%xmm11
+	vmovdqu	0-64(%rsi),%xmm6
+	vpunpckhqdq	%xmm14,%xmm14,%xmm9
+	vpxor	%xmm2,%xmm5,%xmm5
+	vpclmulqdq	$0x00,%xmm7,%xmm8,%xmm12
+	vmovdqu	32-64(%rsi),%xmm7
+	vpxor	%xmm14,%xmm9,%xmm9
+
+	vmovdqu	96(%rdx),%xmm15
+	vpclmulqdq	$0x00,%xmm6,%xmm14,%xmm0
+	vpxor	%xmm3,%xmm10,%xmm10
+	vpshufb	%xmm13,%xmm15,%xmm15
+	vpclmulqdq	$0x11,%xmm6,%xmm14,%xmm1
+	vxorps	%xmm4,%xmm11,%xmm11
+	vmovdqu	16-64(%rsi),%xmm6
+	vpunpckhqdq	%xmm15,%xmm15,%xmm8
+	vpclmulqdq	$0x00,%xmm7,%xmm9,%xmm2
+	vpxor	%xmm5,%xmm12,%xmm12
+	vxorps	%xmm15,%xmm8,%xmm8
+
+	vmovdqu	80(%rdx),%xmm14
+	vpxor	%xmm10,%xmm12,%xmm12
+	vpclmulqdq	$0x00,%xmm6,%xmm15,%xmm3
+	vpxor	%xmm11,%xmm12,%xmm12
+	vpslldq	$8,%xmm12,%xmm9
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x11,%xmm6,%xmm15,%xmm4
+	vpsrldq	$8,%xmm12,%xmm12
+	vpxor	%xmm9,%xmm10,%xmm10
+	vmovdqu	48-64(%rsi),%xmm6
+	vpshufb	%xmm13,%xmm14,%xmm14
+	vxorps	%xmm12,%xmm11,%xmm11
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpunpckhqdq	%xmm14,%xmm14,%xmm9
+	vpclmulqdq	$0x10,%xmm7,%xmm8,%xmm5
+	vmovdqu	80-64(%rsi),%xmm7
+	vpxor	%xmm14,%xmm9,%xmm9
+	vpxor	%xmm2,%xmm5,%xmm5
+
+	vmovdqu	64(%rdx),%xmm15
+	vpalignr	$8,%xmm10,%xmm10,%xmm12
+	vpclmulqdq	$0x00,%xmm6,%xmm14,%xmm0
+	vpshufb	%xmm13,%xmm15,%xmm15
+	vpxor	%xmm3,%xmm0,%xmm0
+	vpclmulqdq	$0x11,%xmm6,%xmm14,%xmm1
+	vmovdqu	64-64(%rsi),%xmm6
+	vpunpckhqdq	%xmm15,%xmm15,%xmm8
+	vpxor	%xmm4,%xmm1,%xmm1
+	vpclmulqdq	$0x00,%xmm7,%xmm9,%xmm2
+	vxorps	%xmm15,%xmm8,%xmm8
+	vpxor	%xmm5,%xmm2,%xmm2
+
+	vmovdqu	48(%rdx),%xmm14
+	vpclmulqdq	$0x10,(%r10),%xmm10,%xmm10
+	vpclmulqdq	$0x00,%xmm6,%xmm15,%xmm3
+	vpshufb	%xmm13,%xmm14,%xmm14
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x11,%xmm6,%xmm15,%xmm4
+	vmovdqu	96-64(%rsi),%xmm6
+	vpunpckhqdq	%xmm14,%xmm14,%xmm9
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpclmulqdq	$0x10,%xmm7,%xmm8,%xmm5
+	vmovdqu	128-64(%rsi),%xmm7
+	vpxor	%xmm14,%xmm9,%xmm9
+	vpxor	%xmm2,%xmm5,%xmm5
+
+	vmovdqu	32(%rdx),%xmm15
+	vpclmulqdq	$0x00,%xmm6,%xmm14,%xmm0
+	vpshufb	%xmm13,%xmm15,%xmm15
+	vpxor	%xmm3,%xmm0,%xmm0
+	vpclmulqdq	$0x11,%xmm6,%xmm14,%xmm1
+	vmovdqu	112-64(%rsi),%xmm6
+	vpunpckhqdq	%xmm15,%xmm15,%xmm8
+	vpxor	%xmm4,%xmm1,%xmm1
+	vpclmulqdq	$0x00,%xmm7,%xmm9,%xmm2
+	vpxor	%xmm15,%xmm8,%xmm8
+	vpxor	%xmm5,%xmm2,%xmm2
+	vxorps	%xmm12,%xmm10,%xmm10
+
+	vmovdqu	16(%rdx),%xmm14
+	vpalignr	$8,%xmm10,%xmm10,%xmm12
+	vpclmulqdq	$0x00,%xmm6,%xmm15,%xmm3
+	vpshufb	%xmm13,%xmm14,%xmm14
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x11,%xmm6,%xmm15,%xmm4
+	vmovdqu	144-64(%rsi),%xmm6
+	vpclmulqdq	$0x10,(%r10),%xmm10,%xmm10
+	vxorps	%xmm11,%xmm12,%xmm12
+	vpunpckhqdq	%xmm14,%xmm14,%xmm9
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpclmulqdq	$0x10,%xmm7,%xmm8,%xmm5
+	vmovdqu	176-64(%rsi),%xmm7
+	vpxor	%xmm14,%xmm9,%xmm9
+	vpxor	%xmm2,%xmm5,%xmm5
+
+	vmovdqu	(%rdx),%xmm15
+	vpclmulqdq	$0x00,%xmm6,%xmm14,%xmm0
+	vpshufb	%xmm13,%xmm15,%xmm15
+	vpclmulqdq	$0x11,%xmm6,%xmm14,%xmm1
+	vmovdqu	160-64(%rsi),%xmm6
+	vpxor	%xmm12,%xmm15,%xmm15
+	vpclmulqdq	$0x10,%xmm7,%xmm9,%xmm2
+	vpxor	%xmm10,%xmm15,%xmm15
+
+	leaq	128(%rdx),%rdx
+	subq	$0x80,%rcx
+	jnc	.Loop8x_avx
+
+	addq	$0x80,%rcx
+	jmp	.Ltail_no_xor_avx
+
+.align	32
+.Lshort_avx:
+	vmovdqu	-16(%rdx,%rcx,1),%xmm14
+	leaq	(%rdx,%rcx,1),%rdx
+	vmovdqu	0-64(%rsi),%xmm6
+	vmovdqu	32-64(%rsi),%xmm7
+	vpshufb	%xmm13,%xmm14,%xmm15
+
+	vmovdqa	%xmm0,%xmm3
+	vmovdqa	%xmm1,%xmm4
+	vmovdqa	%xmm2,%xmm5
+	subq	$0x10,%rcx
+	jz	.Ltail_avx
+
+	vpunpckhqdq	%xmm15,%xmm15,%xmm8
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x00,%xmm6,%xmm15,%xmm0
+	vpxor	%xmm15,%xmm8,%xmm8
+	vmovdqu	-32(%rdx),%xmm14
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpclmulqdq	$0x11,%xmm6,%xmm15,%xmm1
+	vmovdqu	16-64(%rsi),%xmm6
+	vpshufb	%xmm13,%xmm14,%xmm15
+	vpxor	%xmm2,%xmm5,%xmm5
+	vpclmulqdq	$0x00,%xmm7,%xmm8,%xmm2
+	vpsrldq	$8,%xmm7,%xmm7
+	subq	$0x10,%rcx
+	jz	.Ltail_avx
+
+	vpunpckhqdq	%xmm15,%xmm15,%xmm8
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x00,%xmm6,%xmm15,%xmm0
+	vpxor	%xmm15,%xmm8,%xmm8
+	vmovdqu	-48(%rdx),%xmm14
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpclmulqdq	$0x11,%xmm6,%xmm15,%xmm1
+	vmovdqu	48-64(%rsi),%xmm6
+	vpshufb	%xmm13,%xmm14,%xmm15
+	vpxor	%xmm2,%xmm5,%xmm5
+	vpclmulqdq	$0x00,%xmm7,%xmm8,%xmm2
+	vmovdqu	80-64(%rsi),%xmm7
+	subq	$0x10,%rcx
+	jz	.Ltail_avx
+
+	vpunpckhqdq	%xmm15,%xmm15,%xmm8
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x00,%xmm6,%xmm15,%xmm0
+	vpxor	%xmm15,%xmm8,%xmm8
+	vmovdqu	-64(%rdx),%xmm14
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpclmulqdq	$0x11,%xmm6,%xmm15,%xmm1
+	vmovdqu	64-64(%rsi),%xmm6
+	vpshufb	%xmm13,%xmm14,%xmm15
+	vpxor	%xmm2,%xmm5,%xmm5
+	vpclmulqdq	$0x00,%xmm7,%xmm8,%xmm2
+	vpsrldq	$8,%xmm7,%xmm7
+	subq	$0x10,%rcx
+	jz	.Ltail_avx
+
+	vpunpckhqdq	%xmm15,%xmm15,%xmm8
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x00,%xmm6,%xmm15,%xmm0
+	vpxor	%xmm15,%xmm8,%xmm8
+	vmovdqu	-80(%rdx),%xmm14
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpclmulqdq	$0x11,%xmm6,%xmm15,%xmm1
+	vmovdqu	96-64(%rsi),%xmm6
+	vpshufb	%xmm13,%xmm14,%xmm15
+	vpxor	%xmm2,%xmm5,%xmm5
+	vpclmulqdq	$0x00,%xmm7,%xmm8,%xmm2
+	vmovdqu	128-64(%rsi),%xmm7
+	subq	$0x10,%rcx
+	jz	.Ltail_avx
+
+	vpunpckhqdq	%xmm15,%xmm15,%xmm8
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x00,%xmm6,%xmm15,%xmm0
+	vpxor	%xmm15,%xmm8,%xmm8
+	vmovdqu	-96(%rdx),%xmm14
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpclmulqdq	$0x11,%xmm6,%xmm15,%xmm1
+	vmovdqu	112-64(%rsi),%xmm6
+	vpshufb	%xmm13,%xmm14,%xmm15
+	vpxor	%xmm2,%xmm5,%xmm5
+	vpclmulqdq	$0x00,%xmm7,%xmm8,%xmm2
+	vpsrldq	$8,%xmm7,%xmm7
+	subq	$0x10,%rcx
+	jz	.Ltail_avx
+
+	vpunpckhqdq	%xmm15,%xmm15,%xmm8
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x00,%xmm6,%xmm15,%xmm0
+	vpxor	%xmm15,%xmm8,%xmm8
+	vmovdqu	-112(%rdx),%xmm14
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpclmulqdq	$0x11,%xmm6,%xmm15,%xmm1
+	vmovdqu	144-64(%rsi),%xmm6
+	vpshufb	%xmm13,%xmm14,%xmm15
+	vpxor	%xmm2,%xmm5,%xmm5
+	vpclmulqdq	$0x00,%xmm7,%xmm8,%xmm2
+	vmovq	184-64(%rsi),%xmm7
+	subq	$0x10,%rcx
+	jmp	.Ltail_avx
+
+.align	32
+.Ltail_avx:
+	vpxor	%xmm10,%xmm15,%xmm15
+.Ltail_no_xor_avx:
+	vpunpckhqdq	%xmm15,%xmm15,%xmm8
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x00,%xmm6,%xmm15,%xmm0
+	vpxor	%xmm15,%xmm8,%xmm8
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpclmulqdq	$0x11,%xmm6,%xmm15,%xmm1
+	vpxor	%xmm2,%xmm5,%xmm5
+	vpclmulqdq	$0x00,%xmm7,%xmm8,%xmm2
+
+	vmovdqu	(%r10),%xmm12
+
+	vpxor	%xmm0,%xmm3,%xmm10
+	vpxor	%xmm1,%xmm4,%xmm11
+	vpxor	%xmm2,%xmm5,%xmm5
+
+	vpxor	%xmm10,%xmm5,%xmm5
+	vpxor	%xmm11,%xmm5,%xmm5
+	vpslldq	$8,%xmm5,%xmm9
+	vpsrldq	$8,%xmm5,%xmm5
+	vpxor	%xmm9,%xmm10,%xmm10
+	vpxor	%xmm5,%xmm11,%xmm11
+
+	vpclmulqdq	$0x10,%xmm12,%xmm10,%xmm9
+	vpalignr	$8,%xmm10,%xmm10,%xmm10
+	vpxor	%xmm9,%xmm10,%xmm10
+
+	vpclmulqdq	$0x10,%xmm12,%xmm10,%xmm9
+	vpalignr	$8,%xmm10,%xmm10,%xmm10
+	vpxor	%xmm11,%xmm10,%xmm10
+	vpxor	%xmm9,%xmm10,%xmm10
+
+	cmpq	$0,%rcx
+	jne	.Lshort_avx
+
+	vpshufb	%xmm13,%xmm10,%xmm10
+	vmovdqu	%xmm10,(%rdi)
+	vzeroupper
+	.byte	0xf3,0xc3
 .size	gcm_ghash_avx,.-gcm_ghash_avx
 .align	64
 .Lbswap_mask:

deps/openssl/asm/x64-elf-gas/x86_64cpuid.s

@@ -44,43 +44,43 @@ OPENSSL_ia32_cpuid:
 	movl	%eax,%r11d
 
 	xorl	%eax,%eax
-	cmpl	$1970169159,%ebx
+	cmpl	$0x756e6547,%ebx
 	setne	%al
 	movl	%eax,%r9d
-	cmpl	$1231384169,%edx
+	cmpl	$0x49656e69,%edx
 	setne	%al
 	orl	%eax,%r9d
-	cmpl	$1818588270,%ecx
+	cmpl	$0x6c65746e,%ecx
 	setne	%al
 	orl	%eax,%r9d
 	jz	.Lintel
 
-	cmpl	$1752462657,%ebx
+	cmpl	$0x68747541,%ebx
 	setne	%al
 	movl	%eax,%r10d
-	cmpl	$1769238117,%edx
+	cmpl	$0x69746E65,%edx
 	setne	%al
 	orl	%eax,%r10d
-	cmpl	$1145913699,%ecx
+	cmpl	$0x444D4163,%ecx
 	setne	%al
 	orl	%eax,%r10d
 	jnz	.Lintel
 
 
-	movl	$2147483648,%eax
+	movl	$0x80000000,%eax
 	cpuid
-	cmpl	$2147483649,%eax
+	cmpl	$0x80000001,%eax
 	jb	.Lintel
 	movl	%eax,%r10d
-	movl	$2147483649,%eax
+	movl	$0x80000001,%eax
 	cpuid
 	orl	%ecx,%r9d
-	andl	$2049,%r9d
+	andl	$0x00000801,%r9d
 
-	cmpl	$2147483656,%r10d
+	cmpl	$0x80000008,%r10d
 	jb	.Lintel
 
-	movl	$2147483656,%eax
+	movl	$0x80000008,%eax
 	cpuid
 	movzbq	%cl,%r10
 	incq	%r10
@@ -92,7 +92,7 @@ OPENSSL_ia32_cpuid:
 	shrl	$16,%ebx
 	cmpb	%r10b,%bl
 	ja	.Lgeneric
-	andl	$4026531839,%edx
+	andl	$0xefffffff,%edx
 	jmp	.Lgeneric
 
 .Lintel:
@@ -105,7 +105,7 @@ OPENSSL_ia32_cpuid:
 	cpuid
 	movl	%eax,%r10d
 	shrl	$14,%r10d
-	andl	$4095,%r10d
+	andl	$0xfff,%r10d
 
 	cmpl	$7,%r11d
 	jb	.Lnocacheinfo
@@ -118,29 +118,29 @@ OPENSSL_ia32_cpuid:
 .Lnocacheinfo:
 	movl	$1,%eax
 	cpuid
-	andl	$3220176895,%edx
+	andl	$0xbfefffff,%edx
 	cmpl	$0,%r9d
 	jne	.Lnotintel
-	orl	$1073741824,%edx
+	orl	$0x40000000,%edx
 	andb	$15,%ah
 	cmpb	$15,%ah
 	jne	.Lnotintel
-	orl	$1048576,%edx
+	orl	$0x00100000,%edx
 .Lnotintel:
 	btl	$28,%edx
 	jnc	.Lgeneric
-	andl	$4026531839,%edx
+	andl	$0xefffffff,%edx
 	cmpl	$0,%r10d
 	je	.Lgeneric
 
-	orl	$268435456,%edx
+	orl	$0x10000000,%edx
 	shrl	$16,%ebx
 	cmpb	$1,%bl
 	ja	.Lgeneric
-	andl	$4026531839,%edx
+	andl	$0xefffffff,%edx
 .Lgeneric:
-	andl	$2048,%r9d
-	andl	$4294965247,%ecx
+	andl	$0x00000800,%r9d
+	andl	$0xfffff7ff,%ecx
 	orl	%ecx,%r9d
 
 	movl	%edx,%r10d
@@ -152,9 +152,9 @@ OPENSSL_ia32_cpuid:
 	cmpl	$6,%eax
 	je	.Ldone
 .Lclear_avx:
-	movl	$4026525695,%eax
+	movl	$0xefffe7ff,%eax
 	andl	%eax,%r9d
-	andl	$4294967263,8(%rdi)
+	andl	$0xffffffdf,8(%rdi)
 .Ldone:
 	shlq	$32,%r9
 	movl	%r10d,%eax

deps/openssl/asm/x64-macosx-gas/aes/aes-x86_64.s

@@ -81,8 +81,8 @@ L$enc_loop:
 	movl	0(%r14,%rdi,8),%edi
 	movl	0(%r14,%rbp,8),%ebp
 
-	andl	$65280,%edi
-	andl	$65280,%ebp
+	andl	$0x0000ff00,%edi
+	andl	$0x0000ff00,%ebp
 
 	xorl	%edi,%r10d
 	xorl	%ebp,%r11d
@@ -94,8 +94,8 @@ L$enc_loop:
 	movl	0(%r14,%rsi,8),%esi
 	movl	0(%r14,%rdi,8),%edi
 
-	andl	$65280,%esi
-	andl	$65280,%edi
+	andl	$0x0000ff00,%esi
+	andl	$0x0000ff00,%edi
 	shrl	$16,%ebx
 	xorl	%esi,%r12d
 	xorl	%edi,%r8d
@@ -108,9 +108,9 @@ L$enc_loop:
 	movl	0(%r14,%rdi,8),%edi
 	movl	0(%r14,%rbp,8),%ebp
 
-	andl	$16711680,%esi
-	andl	$16711680,%edi
-	andl	$16711680,%ebp
+	andl	$0x00ff0000,%esi
+	andl	$0x00ff0000,%edi
+	andl	$0x00ff0000,%ebp
 
 	xorl	%esi,%r10d
 	xorl	%edi,%r11d
@@ -123,9 +123,9 @@ L$enc_loop:
 	movl	2(%r14,%rdi,8),%edi
 	movl	2(%r14,%rbp,8),%ebp
 
-	andl	$16711680,%esi
-	andl	$4278190080,%edi
-	andl	$4278190080,%ebp
+	andl	$0x00ff0000,%esi
+	andl	$0xff000000,%edi
+	andl	$0xff000000,%ebp
 
 	xorl	%esi,%r8d
 	xorl	%edi,%r10d
@@ -138,8 +138,8 @@ L$enc_loop:
 	movl	2(%r14,%rdi,8),%edi
 	movl	16+0(%r15),%eax
 
-	andl	$4278190080,%esi
-	andl	$4278190080,%edi
+	andl	$0xff000000,%esi
+	andl	$0xff000000,%edi
 
 	xorl	%esi,%r12d
 	xorl	%edi,%r8d
@@ -241,8 +241,8 @@ L$enc_loop_compact:
 	xorl	%r8d,%edx
 	cmpq	16(%rsp),%r15
 	je	L$enc_compact_done
-	movl	$2155905152,%r10d
-	movl	$2155905152,%r11d
+	movl	$0x80808080,%r10d
+	movl	$0x80808080,%r11d
 	andl	%eax,%r10d
 	andl	%ebx,%r11d
 	movl	%r10d,%esi
@@ -253,10 +253,10 @@ L$enc_loop_compact:
 	leal	(%rbx,%rbx,1),%r9d
 	subl	%r10d,%esi
 	subl	%r11d,%edi
-	andl	$4278124286,%r8d
-	andl	$4278124286,%r9d
-	andl	$454761243,%esi
-	andl	$454761243,%edi
+	andl	$0xfefefefe,%r8d
+	andl	$0xfefefefe,%r9d
+	andl	$0x1b1b1b1b,%esi
+	andl	$0x1b1b1b1b,%edi
 	movl	%eax,%r10d
 	movl	%ebx,%r11d
 	xorl	%esi,%r8d
@@ -264,9 +264,9 @@ L$enc_loop_compact:
 
 	xorl	%r8d,%eax
 	xorl	%r9d,%ebx
-	movl	$2155905152,%r12d
+	movl	$0x80808080,%r12d
 	roll	$24,%eax
-	movl	$2155905152,%ebp
+	movl	$0x80808080,%ebp
 	roll	$24,%ebx
 	andl	%ecx,%r12d
 	andl	%edx,%ebp
@@ -289,10 +289,10 @@ L$enc_loop_compact:
 	xorl	%r10d,%eax
 	xorl	%r11d,%ebx
 
-	andl	$4278124286,%r8d
-	andl	$4278124286,%r9d
-	andl	$454761243,%esi
-	andl	$454761243,%edi
+	andl	$0xfefefefe,%r8d
+	andl	$0xfefefefe,%r9d
+	andl	$0x1b1b1b1b,%esi
+	andl	$0x1b1b1b1b,%edi
 	movl	%ecx,%r12d
 	movl	%edx,%ebp
 	xorl	%esi,%r8d
@@ -345,7 +345,7 @@ _AES_encrypt:
 	andq	$-64,%rsp
 	subq	%rsp,%rcx
 	negq	%rcx
-	andq	$960,%rcx
+	andq	$0x3c0,%rcx
 	subq	%rcx,%rsp
 	subq	$32,%rsp
 
@@ -370,7 +370,7 @@ L$enc_prologue:
 	leaq	L$AES_Te+2048(%rip),%r14
 	leaq	768(%rsp),%rbp
 	subq	%r14,%rbp
-	andq	$768,%rbp
+	andq	$0x300,%rbp
 	leaq	(%r14,%rbp,1),%r14
 
 	call	_x86_64_AES_encrypt_compact
@@ -792,7 +792,7 @@ _AES_decrypt:
 	andq	$-64,%rsp
 	subq	%rsp,%rcx
 	negq	%rcx
-	andq	$960,%rcx
+	andq	$0x3c0,%rcx
 	subq	%rcx,%rsp
 	subq	$32,%rsp
 
@@ -817,7 +817,7 @@ L$dec_prologue:
 	leaq	L$AES_Td+2048(%rip),%r14
 	leaq	768(%rsp),%rbp
 	subq	%r14,%rbp
-	andq	$768,%rbp
+	andq	$0x300,%rbp
 	leaq	(%r14,%rbp,1),%r14
 	shrq	$3,%rbp
 	addq	%rbp,%r14
@@ -1333,9 +1333,9 @@ L$cbc_picked_te:
 	movq	%r14,%r10
 	leaq	2304(%r14),%r11
 	movq	%r15,%r12
-	andq	$4095,%r10
-	andq	$4095,%r11
-	andq	$4095,%r12
+	andq	$0xFFF,%r10
+	andq	$0xFFF,%r11
+	andq	$0xFFF,%r12
 
 	cmpq	%r11,%r12
 	jb	L$cbc_te_break_out
@@ -1344,7 +1344,7 @@ L$cbc_picked_te:
 	jmp	L$cbc_te_ok
 L$cbc_te_break_out:
 	subq	%r10,%r12
-	andq	$4095,%r12
+	andq	$0xFFF,%r12
 	addq	$320,%r12
 	subq	%r12,%r15
 .p2align	2
@@ -1370,7 +1370,7 @@ L$cbc_fast_body:
 
 	movq	%r15,%r10
 	subq	%r14,%r10
-	andq	$4095,%r10
+	andq	$0xfff,%r10
 	cmpq	$2304,%r10
 	jb	L$cbc_do_ecopy
 	cmpq	$4096-248,%r10
@@ -1557,7 +1557,7 @@ L$cbc_slow_prologue:
 	leaq	-88-63(%rcx),%r10
 	subq	%rbp,%r10
 	negq	%r10
-	andq	$960,%r10
+	andq	$0x3c0,%r10
 	subq	%r10,%rbp
 
 	xchgq	%rsp,%rbp
@@ -1586,7 +1586,7 @@ L$cbc_slow_body:
 	leaq	2048(%r14),%r14
 	leaq	768-8(%rsp),%rax
 	subq	%r14,%rax
-	andq	$768,%rax
+	andq	$0x300,%rax
 	leaq	(%r14,%rax,1),%r14
 
 	cmpq	$0,%rbx

deps/openssl/asm/x64-macosx-gas/aes/aesni-mb-x86_64.s

@@ -6,6 +6,14 @@
 
 .p2align	5
 _aesni_multi_cbc_encrypt:
+	cmpl	$2,%edx
+	jb	L$enc_non_avx
+	movl	_OPENSSL_ia32cap_P+4(%rip),%ecx
+	testl	$268435456,%ecx
+	jnz	_avx_cbc_enc_shortcut
+	jmp	L$enc_non_avx
+.p2align	4
+L$enc_non_avx:
 	movq	%rsp,%rax
 	pushq	%rbx
 	pushq	%rbp
@@ -262,6 +270,14 @@ L$enc4x_epilogue:
 
 .p2align	5
 _aesni_multi_cbc_decrypt:
+	cmpl	$2,%edx
+	jb	L$dec_non_avx
+	movl	_OPENSSL_ia32cap_P+4(%rip),%ecx
+	testl	$268435456,%ecx
+	jnz	_avx_cbc_dec_shortcut
+	jmp	L$dec_non_avx
+.p2align	4
+L$dec_non_avx:
 	movq	%rsp,%rax
 	pushq	%rbx
 	pushq	%rbp
@@ -503,3 +519,916 @@ L$dec4x_done:
 	leaq	(%rax),%rsp
 L$dec4x_epilogue:
 	.byte	0xf3,0xc3
+
+
+.p2align	5
+aesni_multi_cbc_encrypt_avx:
+_avx_cbc_enc_shortcut:
+	movq	%rsp,%rax
+	pushq	%rbx
+	pushq	%rbp
+	pushq	%r12
+	pushq	%r13
+	pushq	%r14
+	pushq	%r15
+
+
+
+
+
+
+
+
+	subq	$192,%rsp
+	andq	$-128,%rsp
+	movq	%rax,16(%rsp)
+
+L$enc8x_body:
+	vzeroupper
+	vmovdqu	(%rsi),%xmm15
+	leaq	120(%rsi),%rsi
+	leaq	160(%rdi),%rdi
+	shrl	$1,%edx
+
+L$enc8x_loop_grande:
+
+	xorl	%edx,%edx
+	movl	-144(%rdi),%ecx
+	movq	-160(%rdi),%r8
+	cmpl	%edx,%ecx
+	movq	-152(%rdi),%rbx
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	-136(%rdi),%xmm2
+	movl	%ecx,32(%rsp)
+	cmovleq	%rsp,%r8
+	subq	%r8,%rbx
+	movq	%rbx,64(%rsp)
+	movl	-104(%rdi),%ecx
+	movq	-120(%rdi),%r9
+	cmpl	%edx,%ecx
+	movq	-112(%rdi),%rbp
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	-96(%rdi),%xmm3
+	movl	%ecx,36(%rsp)
+	cmovleq	%rsp,%r9
+	subq	%r9,%rbp
+	movq	%rbp,72(%rsp)
+	movl	-64(%rdi),%ecx
+	movq	-80(%rdi),%r10
+	cmpl	%edx,%ecx
+	movq	-72(%rdi),%rbp
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	-56(%rdi),%xmm4
+	movl	%ecx,40(%rsp)
+	cmovleq	%rsp,%r10
+	subq	%r10,%rbp
+	movq	%rbp,80(%rsp)
+	movl	-24(%rdi),%ecx
+	movq	-40(%rdi),%r11
+	cmpl	%edx,%ecx
+	movq	-32(%rdi),%rbp
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	-16(%rdi),%xmm5
+	movl	%ecx,44(%rsp)
+	cmovleq	%rsp,%r11
+	subq	%r11,%rbp
+	movq	%rbp,88(%rsp)
+	movl	16(%rdi),%ecx
+	movq	0(%rdi),%r12
+	cmpl	%edx,%ecx
+	movq	8(%rdi),%rbp
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	24(%rdi),%xmm6
+	movl	%ecx,48(%rsp)
+	cmovleq	%rsp,%r12
+	subq	%r12,%rbp
+	movq	%rbp,96(%rsp)
+	movl	56(%rdi),%ecx
+	movq	40(%rdi),%r13
+	cmpl	%edx,%ecx
+	movq	48(%rdi),%rbp
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	64(%rdi),%xmm7
+	movl	%ecx,52(%rsp)
+	cmovleq	%rsp,%r13
+	subq	%r13,%rbp
+	movq	%rbp,104(%rsp)
+	movl	96(%rdi),%ecx
+	movq	80(%rdi),%r14
+	cmpl	%edx,%ecx
+	movq	88(%rdi),%rbp
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	104(%rdi),%xmm8
+	movl	%ecx,56(%rsp)
+	cmovleq	%rsp,%r14
+	subq	%r14,%rbp
+	movq	%rbp,112(%rsp)
+	movl	136(%rdi),%ecx
+	movq	120(%rdi),%r15
+	cmpl	%edx,%ecx
+	movq	128(%rdi),%rbp
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	144(%rdi),%xmm9
+	movl	%ecx,60(%rsp)
+	cmovleq	%rsp,%r15
+	subq	%r15,%rbp
+	movq	%rbp,120(%rsp)
+	testl	%edx,%edx
+	jz	L$enc8x_done
+
+	vmovups	16-120(%rsi),%xmm1
+	vmovups	32-120(%rsi),%xmm0
+	movl	240-120(%rsi),%eax
+
+	vpxor	(%r8),%xmm15,%xmm10
+	leaq	128(%rsp),%rbp
+	vpxor	(%r9),%xmm15,%xmm11
+	vpxor	(%r10),%xmm15,%xmm12
+	vpxor	(%r11),%xmm15,%xmm13
+	vpxor	%xmm10,%xmm2,%xmm2
+	vpxor	(%r12),%xmm15,%xmm10
+	vpxor	%xmm11,%xmm3,%xmm3
+	vpxor	(%r13),%xmm15,%xmm11
+	vpxor	%xmm12,%xmm4,%xmm4
+	vpxor	(%r14),%xmm15,%xmm12
+	vpxor	%xmm13,%xmm5,%xmm5
+	vpxor	(%r15),%xmm15,%xmm13
+	vpxor	%xmm10,%xmm6,%xmm6
+	movl	$1,%ecx
+	vpxor	%xmm11,%xmm7,%xmm7
+	vpxor	%xmm12,%xmm8,%xmm8
+	vpxor	%xmm13,%xmm9,%xmm9
+	jmp	L$oop_enc8x
+
+.p2align	5
+L$oop_enc8x:
+	vaesenc	%xmm1,%xmm2,%xmm2
+	cmpl	32+0(%rsp),%ecx
+	vaesenc	%xmm1,%xmm3,%xmm3
+	prefetcht0	31(%r8)
+	vaesenc	%xmm1,%xmm4,%xmm4
+	vaesenc	%xmm1,%xmm5,%xmm5
+	leaq	(%r8,%rbx,1),%rbx
+	cmovgeq	%rsp,%r8
+	vaesenc	%xmm1,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesenc	%xmm1,%xmm7,%xmm7
+	subq	%r8,%rbx
+	vaesenc	%xmm1,%xmm8,%xmm8
+	vpxor	16(%r8),%xmm15,%xmm10
+	movq	%rbx,64+0(%rsp)
+	vaesenc	%xmm1,%xmm9,%xmm9
+	vmovups	-72(%rsi),%xmm1
+	leaq	16(%r8,%rbx,1),%r8
+	vmovdqu	%xmm10,0(%rbp)
+	vaesenc	%xmm0,%xmm2,%xmm2
+	cmpl	32+4(%rsp),%ecx
+	movq	64+8(%rsp),%rbx
+	vaesenc	%xmm0,%xmm3,%xmm3
+	prefetcht0	31(%r9)
+	vaesenc	%xmm0,%xmm4,%xmm4
+	vaesenc	%xmm0,%xmm5,%xmm5
+	leaq	(%r9,%rbx,1),%rbx
+	cmovgeq	%rsp,%r9
+	vaesenc	%xmm0,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesenc	%xmm0,%xmm7,%xmm7
+	subq	%r9,%rbx
+	vaesenc	%xmm0,%xmm8,%xmm8
+	vpxor	16(%r9),%xmm15,%xmm11
+	movq	%rbx,64+8(%rsp)
+	vaesenc	%xmm0,%xmm9,%xmm9
+	vmovups	-56(%rsi),%xmm0
+	leaq	16(%r9,%rbx,1),%r9
+	vmovdqu	%xmm11,16(%rbp)
+	vaesenc	%xmm1,%xmm2,%xmm2
+	cmpl	32+8(%rsp),%ecx
+	movq	64+16(%rsp),%rbx
+	vaesenc	%xmm1,%xmm3,%xmm3
+	prefetcht0	31(%r10)
+	vaesenc	%xmm1,%xmm4,%xmm4
+	prefetcht0	15(%r8)
+	vaesenc	%xmm1,%xmm5,%xmm5
+	leaq	(%r10,%rbx,1),%rbx
+	cmovgeq	%rsp,%r10
+	vaesenc	%xmm1,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesenc	%xmm1,%xmm7,%xmm7
+	subq	%r10,%rbx
+	vaesenc	%xmm1,%xmm8,%xmm8
+	vpxor	16(%r10),%xmm15,%xmm12
+	movq	%rbx,64+16(%rsp)
+	vaesenc	%xmm1,%xmm9,%xmm9
+	vmovups	-40(%rsi),%xmm1
+	leaq	16(%r10,%rbx,1),%r10
+	vmovdqu	%xmm12,32(%rbp)
+	vaesenc	%xmm0,%xmm2,%xmm2
+	cmpl	32+12(%rsp),%ecx
+	movq	64+24(%rsp),%rbx
+	vaesenc	%xmm0,%xmm3,%xmm3
+	prefetcht0	31(%r11)
+	vaesenc	%xmm0,%xmm4,%xmm4
+	prefetcht0	15(%r9)
+	vaesenc	%xmm0,%xmm5,%xmm5
+	leaq	(%r11,%rbx,1),%rbx
+	cmovgeq	%rsp,%r11
+	vaesenc	%xmm0,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesenc	%xmm0,%xmm7,%xmm7
+	subq	%r11,%rbx
+	vaesenc	%xmm0,%xmm8,%xmm8
+	vpxor	16(%r11),%xmm15,%xmm13
+	movq	%rbx,64+24(%rsp)
+	vaesenc	%xmm0,%xmm9,%xmm9
+	vmovups	-24(%rsi),%xmm0
+	leaq	16(%r11,%rbx,1),%r11
+	vmovdqu	%xmm13,48(%rbp)
+	vaesenc	%xmm1,%xmm2,%xmm2
+	cmpl	32+16(%rsp),%ecx
+	movq	64+32(%rsp),%rbx
+	vaesenc	%xmm1,%xmm3,%xmm3
+	prefetcht0	31(%r12)
+	vaesenc	%xmm1,%xmm4,%xmm4
+	prefetcht0	15(%r10)
+	vaesenc	%xmm1,%xmm5,%xmm5
+	leaq	(%r12,%rbx,1),%rbx
+	cmovgeq	%rsp,%r12
+	vaesenc	%xmm1,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesenc	%xmm1,%xmm7,%xmm7
+	subq	%r12,%rbx
+	vaesenc	%xmm1,%xmm8,%xmm8
+	vpxor	16(%r12),%xmm15,%xmm10
+	movq	%rbx,64+32(%rsp)
+	vaesenc	%xmm1,%xmm9,%xmm9
+	vmovups	-8(%rsi),%xmm1
+	leaq	16(%r12,%rbx,1),%r12
+	vaesenc	%xmm0,%xmm2,%xmm2
+	cmpl	32+20(%rsp),%ecx
+	movq	64+40(%rsp),%rbx
+	vaesenc	%xmm0,%xmm3,%xmm3
+	prefetcht0	31(%r13)
+	vaesenc	%xmm0,%xmm4,%xmm4
+	prefetcht0	15(%r11)
+	vaesenc	%xmm0,%xmm5,%xmm5
+	leaq	(%rbx,%r13,1),%rbx
+	cmovgeq	%rsp,%r13
+	vaesenc	%xmm0,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesenc	%xmm0,%xmm7,%xmm7
+	subq	%r13,%rbx
+	vaesenc	%xmm0,%xmm8,%xmm8
+	vpxor	16(%r13),%xmm15,%xmm11
+	movq	%rbx,64+40(%rsp)
+	vaesenc	%xmm0,%xmm9,%xmm9
+	vmovups	8(%rsi),%xmm0
+	leaq	16(%r13,%rbx,1),%r13
+	vaesenc	%xmm1,%xmm2,%xmm2
+	cmpl	32+24(%rsp),%ecx
+	movq	64+48(%rsp),%rbx
+	vaesenc	%xmm1,%xmm3,%xmm3
+	prefetcht0	31(%r14)
+	vaesenc	%xmm1,%xmm4,%xmm4
+	prefetcht0	15(%r12)
+	vaesenc	%xmm1,%xmm5,%xmm5
+	leaq	(%r14,%rbx,1),%rbx
+	cmovgeq	%rsp,%r14
+	vaesenc	%xmm1,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesenc	%xmm1,%xmm7,%xmm7
+	subq	%r14,%rbx
+	vaesenc	%xmm1,%xmm8,%xmm8
+	vpxor	16(%r14),%xmm15,%xmm12
+	movq	%rbx,64+48(%rsp)
+	vaesenc	%xmm1,%xmm9,%xmm9
+	vmovups	24(%rsi),%xmm1
+	leaq	16(%r14,%rbx,1),%r14
+	vaesenc	%xmm0,%xmm2,%xmm2
+	cmpl	32+28(%rsp),%ecx
+	movq	64+56(%rsp),%rbx
+	vaesenc	%xmm0,%xmm3,%xmm3
+	prefetcht0	31(%r15)
+	vaesenc	%xmm0,%xmm4,%xmm4
+	prefetcht0	15(%r13)
+	vaesenc	%xmm0,%xmm5,%xmm5
+	leaq	(%r15,%rbx,1),%rbx
+	cmovgeq	%rsp,%r15
+	vaesenc	%xmm0,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesenc	%xmm0,%xmm7,%xmm7
+	subq	%r15,%rbx
+	vaesenc	%xmm0,%xmm8,%xmm8
+	vpxor	16(%r15),%xmm15,%xmm13
+	movq	%rbx,64+56(%rsp)
+	vaesenc	%xmm0,%xmm9,%xmm9
+	vmovups	40(%rsi),%xmm0
+	leaq	16(%r15,%rbx,1),%r15
+	vmovdqu	32(%rsp),%xmm14
+	prefetcht0	15(%r14)
+	prefetcht0	15(%r15)
+	cmpl	$11,%eax
+	jb	L$enc8x_tail
+
+	vaesenc	%xmm1,%xmm2,%xmm2
+	vaesenc	%xmm1,%xmm3,%xmm3
+	vaesenc	%xmm1,%xmm4,%xmm4
+	vaesenc	%xmm1,%xmm5,%xmm5
+	vaesenc	%xmm1,%xmm6,%xmm6
+	vaesenc	%xmm1,%xmm7,%xmm7
+	vaesenc	%xmm1,%xmm8,%xmm8
+	vaesenc	%xmm1,%xmm9,%xmm9
+	vmovups	176-120(%rsi),%xmm1
+
+	vaesenc	%xmm0,%xmm2,%xmm2
+	vaesenc	%xmm0,%xmm3,%xmm3
+	vaesenc	%xmm0,%xmm4,%xmm4
+	vaesenc	%xmm0,%xmm5,%xmm5
+	vaesenc	%xmm0,%xmm6,%xmm6
+	vaesenc	%xmm0,%xmm7,%xmm7
+	vaesenc	%xmm0,%xmm8,%xmm8
+	vaesenc	%xmm0,%xmm9,%xmm9
+	vmovups	192-120(%rsi),%xmm0
+	je	L$enc8x_tail
+
+	vaesenc	%xmm1,%xmm2,%xmm2
+	vaesenc	%xmm1,%xmm3,%xmm3
+	vaesenc	%xmm1,%xmm4,%xmm4
+	vaesenc	%xmm1,%xmm5,%xmm5
+	vaesenc	%xmm1,%xmm6,%xmm6
+	vaesenc	%xmm1,%xmm7,%xmm7
+	vaesenc	%xmm1,%xmm8,%xmm8
+	vaesenc	%xmm1,%xmm9,%xmm9
+	vmovups	208-120(%rsi),%xmm1
+
+	vaesenc	%xmm0,%xmm2,%xmm2
+	vaesenc	%xmm0,%xmm3,%xmm3
+	vaesenc	%xmm0,%xmm4,%xmm4
+	vaesenc	%xmm0,%xmm5,%xmm5
+	vaesenc	%xmm0,%xmm6,%xmm6
+	vaesenc	%xmm0,%xmm7,%xmm7
+	vaesenc	%xmm0,%xmm8,%xmm8
+	vaesenc	%xmm0,%xmm9,%xmm9
+	vmovups	224-120(%rsi),%xmm0
+
+L$enc8x_tail:
+	vaesenc	%xmm1,%xmm2,%xmm2
+	vpxor	%xmm15,%xmm15,%xmm15
+	vaesenc	%xmm1,%xmm3,%xmm3
+	vaesenc	%xmm1,%xmm4,%xmm4
+	vpcmpgtd	%xmm15,%xmm14,%xmm15
+	vaesenc	%xmm1,%xmm5,%xmm5
+	vaesenc	%xmm1,%xmm6,%xmm6
+	vpaddd	%xmm14,%xmm15,%xmm15
+	vmovdqu	48(%rsp),%xmm14
+	vaesenc	%xmm1,%xmm7,%xmm7
+	movq	64(%rsp),%rbx
+	vaesenc	%xmm1,%xmm8,%xmm8
+	vaesenc	%xmm1,%xmm9,%xmm9
+	vmovups	16-120(%rsi),%xmm1
+
+	vaesenclast	%xmm0,%xmm2,%xmm2
+	vmovdqa	%xmm15,32(%rsp)
+	vpxor	%xmm15,%xmm15,%xmm15
+	vaesenclast	%xmm0,%xmm3,%xmm3
+	vaesenclast	%xmm0,%xmm4,%xmm4
+	vpcmpgtd	%xmm15,%xmm14,%xmm15
+	vaesenclast	%xmm0,%xmm5,%xmm5
+	vaesenclast	%xmm0,%xmm6,%xmm6
+	vpaddd	%xmm15,%xmm14,%xmm14
+	vmovdqu	-120(%rsi),%xmm15
+	vaesenclast	%xmm0,%xmm7,%xmm7
+	vaesenclast	%xmm0,%xmm8,%xmm8
+	vmovdqa	%xmm14,48(%rsp)
+	vaesenclast	%xmm0,%xmm9,%xmm9
+	vmovups	32-120(%rsi),%xmm0
+
+	vmovups	%xmm2,-16(%r8)
+	subq	%rbx,%r8
+	vpxor	0(%rbp),%xmm2,%xmm2
+	vmovups	%xmm3,-16(%r9)
+	subq	72(%rsp),%r9
+	vpxor	16(%rbp),%xmm3,%xmm3
+	vmovups	%xmm4,-16(%r10)
+	subq	80(%rsp),%r10
+	vpxor	32(%rbp),%xmm4,%xmm4
+	vmovups	%xmm5,-16(%r11)
+	subq	88(%rsp),%r11
+	vpxor	48(%rbp),%xmm5,%xmm5
+	vmovups	%xmm6,-16(%r12)
+	subq	96(%rsp),%r12
+	vpxor	%xmm10,%xmm6,%xmm6
+	vmovups	%xmm7,-16(%r13)
+	subq	104(%rsp),%r13
+	vpxor	%xmm11,%xmm7,%xmm7
+	vmovups	%xmm8,-16(%r14)
+	subq	112(%rsp),%r14
+	vpxor	%xmm12,%xmm8,%xmm8
+	vmovups	%xmm9,-16(%r15)
+	subq	120(%rsp),%r15
+	vpxor	%xmm13,%xmm9,%xmm9
+
+	decl	%edx
+	jnz	L$oop_enc8x
+
+	movq	16(%rsp),%rax
+
+
+
+
+
+L$enc8x_done:
+	vzeroupper
+	movq	-48(%rax),%r15
+	movq	-40(%rax),%r14
+	movq	-32(%rax),%r13
+	movq	-24(%rax),%r12
+	movq	-16(%rax),%rbp
+	movq	-8(%rax),%rbx
+	leaq	(%rax),%rsp
+L$enc8x_epilogue:
+	.byte	0xf3,0xc3
+
+
+
+.p2align	5
+aesni_multi_cbc_decrypt_avx:
+_avx_cbc_dec_shortcut:
+	movq	%rsp,%rax
+	pushq	%rbx
+	pushq	%rbp
+	pushq	%r12
+	pushq	%r13
+	pushq	%r14
+	pushq	%r15
+
+
+
+
+
+
+
+
+
+	subq	$256,%rsp
+	andq	$-256,%rsp
+	subq	$192,%rsp
+	movq	%rax,16(%rsp)
+
+L$dec8x_body:
+	vzeroupper
+	vmovdqu	(%rsi),%xmm15
+	leaq	120(%rsi),%rsi
+	leaq	160(%rdi),%rdi
+	shrl	$1,%edx
+
+L$dec8x_loop_grande:
+
+	xorl	%edx,%edx
+	movl	-144(%rdi),%ecx
+	movq	-160(%rdi),%r8
+	cmpl	%edx,%ecx
+	movq	-152(%rdi),%rbx
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	-136(%rdi),%xmm2
+	movl	%ecx,32(%rsp)
+	cmovleq	%rsp,%r8
+	subq	%r8,%rbx
+	movq	%rbx,64(%rsp)
+	vmovdqu	%xmm2,192(%rsp)
+	movl	-104(%rdi),%ecx
+	movq	-120(%rdi),%r9
+	cmpl	%edx,%ecx
+	movq	-112(%rdi),%rbp
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	-96(%rdi),%xmm3
+	movl	%ecx,36(%rsp)
+	cmovleq	%rsp,%r9
+	subq	%r9,%rbp
+	movq	%rbp,72(%rsp)
+	vmovdqu	%xmm3,208(%rsp)
+	movl	-64(%rdi),%ecx
+	movq	-80(%rdi),%r10
+	cmpl	%edx,%ecx
+	movq	-72(%rdi),%rbp
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	-56(%rdi),%xmm4
+	movl	%ecx,40(%rsp)
+	cmovleq	%rsp,%r10
+	subq	%r10,%rbp
+	movq	%rbp,80(%rsp)
+	vmovdqu	%xmm4,224(%rsp)
+	movl	-24(%rdi),%ecx
+	movq	-40(%rdi),%r11
+	cmpl	%edx,%ecx
+	movq	-32(%rdi),%rbp
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	-16(%rdi),%xmm5
+	movl	%ecx,44(%rsp)
+	cmovleq	%rsp,%r11
+	subq	%r11,%rbp
+	movq	%rbp,88(%rsp)
+	vmovdqu	%xmm5,240(%rsp)
+	movl	16(%rdi),%ecx
+	movq	0(%rdi),%r12
+	cmpl	%edx,%ecx
+	movq	8(%rdi),%rbp
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	24(%rdi),%xmm6
+	movl	%ecx,48(%rsp)
+	cmovleq	%rsp,%r12
+	subq	%r12,%rbp
+	movq	%rbp,96(%rsp)
+	vmovdqu	%xmm6,256(%rsp)
+	movl	56(%rdi),%ecx
+	movq	40(%rdi),%r13
+	cmpl	%edx,%ecx
+	movq	48(%rdi),%rbp
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	64(%rdi),%xmm7
+	movl	%ecx,52(%rsp)
+	cmovleq	%rsp,%r13
+	subq	%r13,%rbp
+	movq	%rbp,104(%rsp)
+	vmovdqu	%xmm7,272(%rsp)
+	movl	96(%rdi),%ecx
+	movq	80(%rdi),%r14
+	cmpl	%edx,%ecx
+	movq	88(%rdi),%rbp
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	104(%rdi),%xmm8
+	movl	%ecx,56(%rsp)
+	cmovleq	%rsp,%r14
+	subq	%r14,%rbp
+	movq	%rbp,112(%rsp)
+	vmovdqu	%xmm8,288(%rsp)
+	movl	136(%rdi),%ecx
+	movq	120(%rdi),%r15
+	cmpl	%edx,%ecx
+	movq	128(%rdi),%rbp
+	cmovgl	%ecx,%edx
+	testl	%ecx,%ecx
+	vmovdqu	144(%rdi),%xmm9
+	movl	%ecx,60(%rsp)
+	cmovleq	%rsp,%r15
+	subq	%r15,%rbp
+	movq	%rbp,120(%rsp)
+	vmovdqu	%xmm9,304(%rsp)
+	testl	%edx,%edx
+	jz	L$dec8x_done
+
+	vmovups	16-120(%rsi),%xmm1
+	vmovups	32-120(%rsi),%xmm0
+	movl	240-120(%rsi),%eax
+	leaq	192+128(%rsp),%rbp
+
+	vmovdqu	(%r8),%xmm2
+	vmovdqu	(%r9),%xmm3
+	vmovdqu	(%r10),%xmm4
+	vmovdqu	(%r11),%xmm5
+	vmovdqu	(%r12),%xmm6
+	vmovdqu	(%r13),%xmm7
+	vmovdqu	(%r14),%xmm8
+	vmovdqu	(%r15),%xmm9
+	vmovdqu	%xmm2,0(%rbp)
+	vpxor	%xmm15,%xmm2,%xmm2
+	vmovdqu	%xmm3,16(%rbp)
+	vpxor	%xmm15,%xmm3,%xmm3
+	vmovdqu	%xmm4,32(%rbp)
+	vpxor	%xmm15,%xmm4,%xmm4
+	vmovdqu	%xmm5,48(%rbp)
+	vpxor	%xmm15,%xmm5,%xmm5
+	vmovdqu	%xmm6,64(%rbp)
+	vpxor	%xmm15,%xmm6,%xmm6
+	vmovdqu	%xmm7,80(%rbp)
+	vpxor	%xmm15,%xmm7,%xmm7
+	vmovdqu	%xmm8,96(%rbp)
+	vpxor	%xmm15,%xmm8,%xmm8
+	vmovdqu	%xmm9,112(%rbp)
+	vpxor	%xmm15,%xmm9,%xmm9
+	xorq	$0x80,%rbp
+	movl	$1,%ecx
+	jmp	L$oop_dec8x
+
+.p2align	5
+L$oop_dec8x:
+	vaesdec	%xmm1,%xmm2,%xmm2
+	cmpl	32+0(%rsp),%ecx
+	vaesdec	%xmm1,%xmm3,%xmm3
+	prefetcht0	31(%r8)
+	vaesdec	%xmm1,%xmm4,%xmm4
+	vaesdec	%xmm1,%xmm5,%xmm5
+	leaq	(%r8,%rbx,1),%rbx
+	cmovgeq	%rsp,%r8
+	vaesdec	%xmm1,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesdec	%xmm1,%xmm7,%xmm7
+	subq	%r8,%rbx
+	vaesdec	%xmm1,%xmm8,%xmm8
+	vmovdqu	16(%r8),%xmm10
+	movq	%rbx,64+0(%rsp)
+	vaesdec	%xmm1,%xmm9,%xmm9
+	vmovups	-72(%rsi),%xmm1
+	leaq	16(%r8,%rbx,1),%r8
+	vmovdqu	%xmm10,128(%rsp)
+	vaesdec	%xmm0,%xmm2,%xmm2
+	cmpl	32+4(%rsp),%ecx
+	movq	64+8(%rsp),%rbx
+	vaesdec	%xmm0,%xmm3,%xmm3
+	prefetcht0	31(%r9)
+	vaesdec	%xmm0,%xmm4,%xmm4
+	vaesdec	%xmm0,%xmm5,%xmm5
+	leaq	(%r9,%rbx,1),%rbx
+	cmovgeq	%rsp,%r9
+	vaesdec	%xmm0,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesdec	%xmm0,%xmm7,%xmm7
+	subq	%r9,%rbx
+	vaesdec	%xmm0,%xmm8,%xmm8
+	vmovdqu	16(%r9),%xmm11
+	movq	%rbx,64+8(%rsp)
+	vaesdec	%xmm0,%xmm9,%xmm9
+	vmovups	-56(%rsi),%xmm0
+	leaq	16(%r9,%rbx,1),%r9
+	vmovdqu	%xmm11,144(%rsp)
+	vaesdec	%xmm1,%xmm2,%xmm2
+	cmpl	32+8(%rsp),%ecx
+	movq	64+16(%rsp),%rbx
+	vaesdec	%xmm1,%xmm3,%xmm3
+	prefetcht0	31(%r10)
+	vaesdec	%xmm1,%xmm4,%xmm4
+	prefetcht0	15(%r8)
+	vaesdec	%xmm1,%xmm5,%xmm5
+	leaq	(%r10,%rbx,1),%rbx
+	cmovgeq	%rsp,%r10
+	vaesdec	%xmm1,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesdec	%xmm1,%xmm7,%xmm7
+	subq	%r10,%rbx
+	vaesdec	%xmm1,%xmm8,%xmm8
+	vmovdqu	16(%r10),%xmm12
+	movq	%rbx,64+16(%rsp)
+	vaesdec	%xmm1,%xmm9,%xmm9
+	vmovups	-40(%rsi),%xmm1
+	leaq	16(%r10,%rbx,1),%r10
+	vmovdqu	%xmm12,160(%rsp)
+	vaesdec	%xmm0,%xmm2,%xmm2
+	cmpl	32+12(%rsp),%ecx
+	movq	64+24(%rsp),%rbx
+	vaesdec	%xmm0,%xmm3,%xmm3
+	prefetcht0	31(%r11)
+	vaesdec	%xmm0,%xmm4,%xmm4
+	prefetcht0	15(%r9)
+	vaesdec	%xmm0,%xmm5,%xmm5
+	leaq	(%r11,%rbx,1),%rbx
+	cmovgeq	%rsp,%r11
+	vaesdec	%xmm0,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesdec	%xmm0,%xmm7,%xmm7
+	subq	%r11,%rbx
+	vaesdec	%xmm0,%xmm8,%xmm8
+	vmovdqu	16(%r11),%xmm13
+	movq	%rbx,64+24(%rsp)
+	vaesdec	%xmm0,%xmm9,%xmm9
+	vmovups	-24(%rsi),%xmm0
+	leaq	16(%r11,%rbx,1),%r11
+	vmovdqu	%xmm13,176(%rsp)
+	vaesdec	%xmm1,%xmm2,%xmm2
+	cmpl	32+16(%rsp),%ecx
+	movq	64+32(%rsp),%rbx
+	vaesdec	%xmm1,%xmm3,%xmm3
+	prefetcht0	31(%r12)
+	vaesdec	%xmm1,%xmm4,%xmm4
+	prefetcht0	15(%r10)
+	vaesdec	%xmm1,%xmm5,%xmm5
+	leaq	(%r12,%rbx,1),%rbx
+	cmovgeq	%rsp,%r12
+	vaesdec	%xmm1,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesdec	%xmm1,%xmm7,%xmm7
+	subq	%r12,%rbx
+	vaesdec	%xmm1,%xmm8,%xmm8
+	vmovdqu	16(%r12),%xmm10
+	movq	%rbx,64+32(%rsp)
+	vaesdec	%xmm1,%xmm9,%xmm9
+	vmovups	-8(%rsi),%xmm1
+	leaq	16(%r12,%rbx,1),%r12
+	vaesdec	%xmm0,%xmm2,%xmm2
+	cmpl	32+20(%rsp),%ecx
+	movq	64+40(%rsp),%rbx
+	vaesdec	%xmm0,%xmm3,%xmm3
+	prefetcht0	31(%r13)
+	vaesdec	%xmm0,%xmm4,%xmm4
+	prefetcht0	15(%r11)
+	vaesdec	%xmm0,%xmm5,%xmm5
+	leaq	(%rbx,%r13,1),%rbx
+	cmovgeq	%rsp,%r13
+	vaesdec	%xmm0,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesdec	%xmm0,%xmm7,%xmm7
+	subq	%r13,%rbx
+	vaesdec	%xmm0,%xmm8,%xmm8
+	vmovdqu	16(%r13),%xmm11
+	movq	%rbx,64+40(%rsp)
+	vaesdec	%xmm0,%xmm9,%xmm9
+	vmovups	8(%rsi),%xmm0
+	leaq	16(%r13,%rbx,1),%r13
+	vaesdec	%xmm1,%xmm2,%xmm2
+	cmpl	32+24(%rsp),%ecx
+	movq	64+48(%rsp),%rbx
+	vaesdec	%xmm1,%xmm3,%xmm3
+	prefetcht0	31(%r14)
+	vaesdec	%xmm1,%xmm4,%xmm4
+	prefetcht0	15(%r12)
+	vaesdec	%xmm1,%xmm5,%xmm5
+	leaq	(%r14,%rbx,1),%rbx
+	cmovgeq	%rsp,%r14
+	vaesdec	%xmm1,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesdec	%xmm1,%xmm7,%xmm7
+	subq	%r14,%rbx
+	vaesdec	%xmm1,%xmm8,%xmm8
+	vmovdqu	16(%r14),%xmm12
+	movq	%rbx,64+48(%rsp)
+	vaesdec	%xmm1,%xmm9,%xmm9
+	vmovups	24(%rsi),%xmm1
+	leaq	16(%r14,%rbx,1),%r14
+	vaesdec	%xmm0,%xmm2,%xmm2
+	cmpl	32+28(%rsp),%ecx
+	movq	64+56(%rsp),%rbx
+	vaesdec	%xmm0,%xmm3,%xmm3
+	prefetcht0	31(%r15)
+	vaesdec	%xmm0,%xmm4,%xmm4
+	prefetcht0	15(%r13)
+	vaesdec	%xmm0,%xmm5,%xmm5
+	leaq	(%r15,%rbx,1),%rbx
+	cmovgeq	%rsp,%r15
+	vaesdec	%xmm0,%xmm6,%xmm6
+	cmovgq	%rsp,%rbx
+	vaesdec	%xmm0,%xmm7,%xmm7
+	subq	%r15,%rbx
+	vaesdec	%xmm0,%xmm8,%xmm8
+	vmovdqu	16(%r15),%xmm13
+	movq	%rbx,64+56(%rsp)
+	vaesdec	%xmm0,%xmm9,%xmm9
+	vmovups	40(%rsi),%xmm0
+	leaq	16(%r15,%rbx,1),%r15
+	vmovdqu	32(%rsp),%xmm14
+	prefetcht0	15(%r14)
+	prefetcht0	15(%r15)
+	cmpl	$11,%eax
+	jb	L$dec8x_tail
+
+	vaesdec	%xmm1,%xmm2,%xmm2
+	vaesdec	%xmm1,%xmm3,%xmm3
+	vaesdec	%xmm1,%xmm4,%xmm4
+	vaesdec	%xmm1,%xmm5,%xmm5
+	vaesdec	%xmm1,%xmm6,%xmm6
+	vaesdec	%xmm1,%xmm7,%xmm7
+	vaesdec	%xmm1,%xmm8,%xmm8
+	vaesdec	%xmm1,%xmm9,%xmm9
+	vmovups	176-120(%rsi),%xmm1
+
+	vaesdec	%xmm0,%xmm2,%xmm2
+	vaesdec	%xmm0,%xmm3,%xmm3
+	vaesdec	%xmm0,%xmm4,%xmm4
+	vaesdec	%xmm0,%xmm5,%xmm5
+	vaesdec	%xmm0,%xmm6,%xmm6
+	vaesdec	%xmm0,%xmm7,%xmm7
+	vaesdec	%xmm0,%xmm8,%xmm8
+	vaesdec	%xmm0,%xmm9,%xmm9
+	vmovups	192-120(%rsi),%xmm0
+	je	L$dec8x_tail
+
+	vaesdec	%xmm1,%xmm2,%xmm2
+	vaesdec	%xmm1,%xmm3,%xmm3
+	vaesdec	%xmm1,%xmm4,%xmm4
+	vaesdec	%xmm1,%xmm5,%xmm5
+	vaesdec	%xmm1,%xmm6,%xmm6
+	vaesdec	%xmm1,%xmm7,%xmm7
+	vaesdec	%xmm1,%xmm8,%xmm8
+	vaesdec	%xmm1,%xmm9,%xmm9
+	vmovups	208-120(%rsi),%xmm1
+
+	vaesdec	%xmm0,%xmm2,%xmm2
+	vaesdec	%xmm0,%xmm3,%xmm3
+	vaesdec	%xmm0,%xmm4,%xmm4
+	vaesdec	%xmm0,%xmm5,%xmm5
+	vaesdec	%xmm0,%xmm6,%xmm6
+	vaesdec	%xmm0,%xmm7,%xmm7
+	vaesdec	%xmm0,%xmm8,%xmm8
+	vaesdec	%xmm0,%xmm9,%xmm9
+	vmovups	224-120(%rsi),%xmm0
+
+L$dec8x_tail:
+	vaesdec	%xmm1,%xmm2,%xmm2
+	vpxor	%xmm15,%xmm15,%xmm15
+	vaesdec	%xmm1,%xmm3,%xmm3
+	vaesdec	%xmm1,%xmm4,%xmm4
+	vpcmpgtd	%xmm15,%xmm14,%xmm15
+	vaesdec	%xmm1,%xmm5,%xmm5
+	vaesdec	%xmm1,%xmm6,%xmm6
+	vpaddd	%xmm14,%xmm15,%xmm15
+	vmovdqu	48(%rsp),%xmm14
+	vaesdec	%xmm1,%xmm7,%xmm7
+	movq	64(%rsp),%rbx
+	vaesdec	%xmm1,%xmm8,%xmm8
+	vaesdec	%xmm1,%xmm9,%xmm9
+	vmovups	16-120(%rsi),%xmm1
+
+	vaesdeclast	%xmm0,%xmm2,%xmm2
+	vmovdqa	%xmm15,32(%rsp)
+	vpxor	%xmm15,%xmm15,%xmm15
+	vaesdeclast	%xmm0,%xmm3,%xmm3
+	vpxor	0(%rbp),%xmm2,%xmm2
+	vaesdeclast	%xmm0,%xmm4,%xmm4
+	vpxor	16(%rbp),%xmm3,%xmm3
+	vpcmpgtd	%xmm15,%xmm14,%xmm15
+	vaesdeclast	%xmm0,%xmm5,%xmm5
+	vpxor	32(%rbp),%xmm4,%xmm4
+	vaesdeclast	%xmm0,%xmm6,%xmm6
+	vpxor	48(%rbp),%xmm5,%xmm5
+	vpaddd	%xmm15,%xmm14,%xmm14
+	vmovdqu	-120(%rsi),%xmm15
+	vaesdeclast	%xmm0,%xmm7,%xmm7
+	vpxor	64(%rbp),%xmm6,%xmm6
+	vaesdeclast	%xmm0,%xmm8,%xmm8
+	vpxor	80(%rbp),%xmm7,%xmm7
+	vmovdqa	%xmm14,48(%rsp)
+	vaesdeclast	%xmm0,%xmm9,%xmm9
+	vpxor	96(%rbp),%xmm8,%xmm8
+	vmovups	32-120(%rsi),%xmm0
+
+	vmovups	%xmm2,-16(%r8)
+	subq	%rbx,%r8
+	vmovdqu	128+0(%rsp),%xmm2
+	vpxor	112(%rbp),%xmm9,%xmm9
+	vmovups	%xmm3,-16(%r9)
+	subq	72(%rsp),%r9
+	vmovdqu	%xmm2,0(%rbp)
+	vpxor	%xmm15,%xmm2,%xmm2
+	vmovdqu	128+16(%rsp),%xmm3
+	vmovups	%xmm4,-16(%r10)
+	subq	80(%rsp),%r10
+	vmovdqu	%xmm3,16(%rbp)
+	vpxor	%xmm15,%xmm3,%xmm3
+	vmovdqu	128+32(%rsp),%xmm4
+	vmovups	%xmm5,-16(%r11)
+	subq	88(%rsp),%r11
+	vmovdqu	%xmm4,32(%rbp)
+	vpxor	%xmm15,%xmm4,%xmm4
+	vmovdqu	128+48(%rsp),%xmm5
+	vmovups	%xmm6,-16(%r12)
+	subq	96(%rsp),%r12
+	vmovdqu	%xmm5,48(%rbp)
+	vpxor	%xmm15,%xmm5,%xmm5
+	vmovdqu	%xmm10,64(%rbp)
+	vpxor	%xmm10,%xmm15,%xmm6
+	vmovups	%xmm7,-16(%r13)
+	subq	104(%rsp),%r13
+	vmovdqu	%xmm11,80(%rbp)
+	vpxor	%xmm11,%xmm15,%xmm7
+	vmovups	%xmm8,-16(%r14)
+	subq	112(%rsp),%r14
+	vmovdqu	%xmm12,96(%rbp)
+	vpxor	%xmm12,%xmm15,%xmm8
+	vmovups	%xmm9,-16(%r15)
+	subq	120(%rsp),%r15
+	vmovdqu	%xmm13,112(%rbp)
+	vpxor	%xmm13,%xmm15,%xmm9
+
+	xorq	$128,%rbp
+	decl	%edx
+	jnz	L$oop_dec8x
+
+	movq	16(%rsp),%rax
+
+
+
+
+
+L$dec8x_done:
+	vzeroupper
+	movq	-48(%rax),%r15
+	movq	-40(%rax),%r14
+	movq	-32(%rax),%r13
+	movq	-24(%rax),%r12
+	movq	-16(%rax),%rbp
+	movq	-8(%rax),%rbx
+	leaq	(%rax),%rsp
+L$dec8x_epilogue:
+	.byte	0xf3,0xc3

deps/openssl/asm/x64-macosx-gas/aes/aesni-x86_64.s

@@ -503,7 +503,7 @@ _aesni_ecb_encrypt:
 	testl	%r8d,%r8d
 	jz	L$ecb_decrypt
 
-	cmpq	$128,%rdx
+	cmpq	$0x80,%rdx
 	jb	L$ecb_enc_tail
 
 	movdqu	(%rdi),%xmm2
@@ -515,7 +515,7 @@ _aesni_ecb_encrypt:
 	movdqu	96(%rdi),%xmm8
 	movdqu	112(%rdi),%xmm9
 	leaq	128(%rdi),%rdi
-	subq	$128,%rdx
+	subq	$0x80,%rdx
 	jmp	L$ecb_enc_loop8_enter
 .p2align	4
 L$ecb_enc_loop8:
@@ -543,7 +543,7 @@ L$ecb_enc_loop8_enter:
 
 	call	_aesni_encrypt8
 
-	subq	$128,%rdx
+	subq	$0x80,%rdx
 	jnc	L$ecb_enc_loop8
 
 	movups	%xmm2,(%rsi)
@@ -557,22 +557,22 @@ L$ecb_enc_loop8_enter:
 	movups	%xmm8,96(%rsi)
 	movups	%xmm9,112(%rsi)
 	leaq	128(%rsi),%rsi
-	addq	$128,%rdx
+	addq	$0x80,%rdx
 	jz	L$ecb_ret
 
 L$ecb_enc_tail:
 	movups	(%rdi),%xmm2
-	cmpq	$32,%rdx
+	cmpq	$0x20,%rdx
 	jb	L$ecb_enc_one
 	movups	16(%rdi),%xmm3
 	je	L$ecb_enc_two
 	movups	32(%rdi),%xmm4
-	cmpq	$64,%rdx
+	cmpq	$0x40,%rdx
 	jb	L$ecb_enc_three
 	movups	48(%rdi),%xmm5
 	je	L$ecb_enc_four
 	movups	64(%rdi),%xmm6
-	cmpq	$96,%rdx
+	cmpq	$0x60,%rdx
 	jb	L$ecb_enc_five
 	movups	80(%rdi),%xmm7
 	je	L$ecb_enc_six
@@ -646,7 +646,7 @@ L$ecb_enc_six:
 
 .p2align	4
 L$ecb_decrypt:
-	cmpq	$128,%rdx
+	cmpq	$0x80,%rdx
 	jb	L$ecb_dec_tail
 
 	movdqu	(%rdi),%xmm2
@@ -658,7 +658,7 @@ L$ecb_decrypt:
 	movdqu	96(%rdi),%xmm8
 	movdqu	112(%rdi),%xmm9
 	leaq	128(%rdi),%rdi
-	subq	$128,%rdx
+	subq	$0x80,%rdx
 	jmp	L$ecb_dec_loop8_enter
 .p2align	4
 L$ecb_dec_loop8:
@@ -687,7 +687,7 @@ L$ecb_dec_loop8_enter:
 	call	_aesni_decrypt8
 
 	movups	(%r11),%xmm0
-	subq	$128,%rdx
+	subq	$0x80,%rdx
 	jnc	L$ecb_dec_loop8
 
 	movups	%xmm2,(%rsi)
@@ -709,22 +709,22 @@ L$ecb_dec_loop8_enter:
 	movups	%xmm9,112(%rsi)
 	pxor	%xmm9,%xmm9
 	leaq	128(%rsi),%rsi
-	addq	$128,%rdx
+	addq	$0x80,%rdx
 	jz	L$ecb_ret
 
 L$ecb_dec_tail:
 	movups	(%rdi),%xmm2
-	cmpq	$32,%rdx
+	cmpq	$0x20,%rdx
 	jb	L$ecb_dec_one
 	movups	16(%rdi),%xmm3
 	je	L$ecb_dec_two
 	movups	32(%rdi),%xmm4
-	cmpq	$64,%rdx
+	cmpq	$0x40,%rdx
 	jb	L$ecb_dec_three
 	movups	48(%rdi),%xmm5
 	je	L$ecb_dec_four
 	movups	64(%rdi),%xmm6
-	cmpq	$96,%rdx
+	cmpq	$0x60,%rdx
 	jb	L$ecb_dec_five
 	movups	80(%rdi),%xmm7
 	je	L$ecb_dec_six
@@ -1598,7 +1598,7 @@ L$oop_enc1_8:
 
 	movdqa	L$xts_magic(%rip),%xmm8
 	movdqa	%xmm2,%xmm15
-	pshufd	$95,%xmm2,%xmm9
+	pshufd	$0x5f,%xmm2,%xmm9
 	pxor	%xmm0,%xmm1
 	movdqa	%xmm9,%xmm14
 	paddd	%xmm9,%xmm9
@@ -1697,7 +1697,7 @@ L$xts_enc_grandloop:
 .byte	102,15,56,220,248
 	movups	64(%r11),%xmm0
 	movdqa	%xmm8,80(%rsp)
-	pshufd	$95,%xmm15,%xmm9
+	pshufd	$0x5f,%xmm15,%xmm9
 	jmp	L$xts_enc_loop6
 .p2align	5
 L$xts_enc_loop6:
@@ -1836,13 +1836,13 @@ L$xts_enc_short:
 	jz	L$xts_enc_done
 
 	pxor	%xmm0,%xmm11
-	cmpq	$32,%rdx
+	cmpq	$0x20,%rdx
 	jb	L$xts_enc_one
 	pxor	%xmm0,%xmm12
 	je	L$xts_enc_two
 
 	pxor	%xmm0,%xmm13
-	cmpq	$64,%rdx
+	cmpq	$0x40,%rdx
 	jb	L$xts_enc_three
 	pxor	%xmm0,%xmm14
 	je	L$xts_enc_four
@@ -2069,7 +2069,7 @@ L$oop_enc1_11:
 
 	movdqa	L$xts_magic(%rip),%xmm8
 	movdqa	%xmm2,%xmm15
-	pshufd	$95,%xmm2,%xmm9
+	pshufd	$0x5f,%xmm2,%xmm9
 	pxor	%xmm0,%xmm1
 	movdqa	%xmm9,%xmm14
 	paddd	%xmm9,%xmm9
@@ -2168,7 +2168,7 @@ L$xts_dec_grandloop:
 .byte	102,15,56,222,248
 	movups	64(%r11),%xmm0
 	movdqa	%xmm8,80(%rsp)
-	pshufd	$95,%xmm15,%xmm9
+	pshufd	$0x5f,%xmm15,%xmm9
 	jmp	L$xts_dec_loop6
 .p2align	5
 L$xts_dec_loop6:
@@ -2308,13 +2308,13 @@ L$xts_dec_short:
 	jz	L$xts_dec_done
 
 	pxor	%xmm0,%xmm12
-	cmpq	$32,%rdx
+	cmpq	$0x20,%rdx
 	jb	L$xts_dec_one
 	pxor	%xmm0,%xmm13
 	je	L$xts_dec_two
 
 	pxor	%xmm0,%xmm14
-	cmpq	$64,%rdx
+	cmpq	$0x40,%rdx
 	jb	L$xts_dec_three
 	je	L$xts_dec_four
 
@@ -2345,7 +2345,7 @@ L$xts_dec_short:
 	pcmpgtd	%xmm15,%xmm14
 	movdqu	%xmm6,64(%rsi)
 	leaq	80(%rsi),%rsi
-	pshufd	$19,%xmm14,%xmm11
+	pshufd	$0x13,%xmm14,%xmm11
 	andq	$15,%r9
 	jz	L$xts_dec_ret
 
@@ -2634,7 +2634,7 @@ L$cbc_decrypt_bulk:
 	leaq	-8(%rax),%rbp
 	movups	(%r8),%xmm10
 	movl	%r10d,%eax
-	cmpq	$80,%rdx
+	cmpq	$0x50,%rdx
 	jbe	L$cbc_dec_tail
 
 	movups	(%rcx),%xmm0
@@ -2650,14 +2650,14 @@ L$cbc_decrypt_bulk:
 	movdqu	80(%rdi),%xmm7
 	movdqa	%xmm6,%xmm15
 	movl	_OPENSSL_ia32cap_P+4(%rip),%r9d
-	cmpq	$112,%rdx
+	cmpq	$0x70,%rdx
 	jbe	L$cbc_dec_six_or_seven
 
 	andl	$71303168,%r9d
-	subq	$80,%rdx
+	subq	$0x50,%rdx
 	cmpl	$4194304,%r9d
 	je	L$cbc_dec_loop6_enter
-	subq	$32,%rdx
+	subq	$0x20,%rdx
 	leaq	112(%rcx),%rcx
 	jmp	L$cbc_dec_loop8_enter
 .p2align	4
@@ -2672,7 +2672,7 @@ L$cbc_dec_loop8_enter:
 	movups	16-112(%rcx),%xmm1
 	pxor	%xmm0,%xmm4
 	xorq	%r11,%r11
-	cmpq	$112,%rdx
+	cmpq	$0x70,%rdx
 	pxor	%xmm0,%xmm5
 	pxor	%xmm0,%xmm6
 	pxor	%xmm0,%xmm7
@@ -2857,21 +2857,21 @@ L$cbc_dec_done:
 	movups	%xmm8,96(%rsi)
 	leaq	112(%rsi),%rsi
 
-	subq	$128,%rdx
+	subq	$0x80,%rdx
 	ja	L$cbc_dec_loop8
 
 	movaps	%xmm9,%xmm2
 	leaq	-112(%rcx),%rcx
-	addq	$112,%rdx
+	addq	$0x70,%rdx
 	jle	L$cbc_dec_clear_tail_collected
 	movups	%xmm9,(%rsi)
 	leaq	16(%rsi),%rsi
-	cmpq	$80,%rdx
+	cmpq	$0x50,%rdx
 	jbe	L$cbc_dec_tail
 
 	movaps	%xmm11,%xmm2
 L$cbc_dec_six_or_seven:
-	cmpq	$96,%rdx
+	cmpq	$0x60,%rdx
 	ja	L$cbc_dec_seven
 
 	movaps	%xmm7,%xmm8
@@ -2964,33 +2964,33 @@ L$cbc_dec_loop6_enter:
 	movl	%r10d,%eax
 	movdqu	%xmm6,64(%rsi)
 	leaq	80(%rsi),%rsi
-	subq	$96,%rdx
+	subq	$0x60,%rdx
 	ja	L$cbc_dec_loop6
 
 	movdqa	%xmm7,%xmm2
-	addq	$80,%rdx
+	addq	$0x50,%rdx
 	jle	L$cbc_dec_clear_tail_collected
 	movups	%xmm7,(%rsi)
 	leaq	16(%rsi),%rsi
 
 L$cbc_dec_tail:
 	movups	(%rdi),%xmm2
-	subq	$16,%rdx
+	subq	$0x10,%rdx
 	jbe	L$cbc_dec_one
 
 	movups	16(%rdi),%xmm3
 	movaps	%xmm2,%xmm11
-	subq	$16,%rdx
+	subq	$0x10,%rdx
 	jbe	L$cbc_dec_two
 
 	movups	32(%rdi),%xmm4
 	movaps	%xmm3,%xmm12
-	subq	$16,%rdx
+	subq	$0x10,%rdx
 	jbe	L$cbc_dec_three
 
 	movups	48(%rdi),%xmm5
 	movaps	%xmm4,%xmm13
-	subq	$16,%rdx
+	subq	$0x10,%rdx
 	jbe	L$cbc_dec_four
 
 	movups	64(%rdi),%xmm6
@@ -3015,7 +3015,7 @@ L$cbc_dec_tail:
 	movdqa	%xmm6,%xmm2
 	pxor	%xmm6,%xmm6
 	pxor	%xmm7,%xmm7
-	subq	$16,%rdx
+	subq	$0x10,%rdx
 	jmp	L$cbc_dec_tail_collected
 
 .p2align	4
@@ -3332,7 +3332,7 @@ L$oop_key192:
 	pslldq	$4,%xmm0
 	pxor	%xmm3,%xmm0
 
-	pshufd	$255,%xmm0,%xmm3
+	pshufd	$0xff,%xmm0,%xmm3
 	pxor	%xmm1,%xmm3
 	pslldq	$4,%xmm1
 	pxor	%xmm1,%xmm3
@@ -3419,7 +3419,7 @@ L$oop_key256:
 	decl	%r10d
 	jz	L$done_key256
 
-	pshufd	$255,%xmm0,%xmm2
+	pshufd	$0xff,%xmm0,%xmm2
 	pxor	%xmm3,%xmm3
 .byte	102,15,56,221,211
 

deps/openssl/asm/x64-macosx-gas/aes/bsaes-x86_64.s

@@ -324,45 +324,45 @@ L$enc_sbox:
 	pxor	%xmm2,%xmm5
 	decl	%r10d
 	jl	L$enc_done
-	pshufd	$147,%xmm15,%xmm7
-	pshufd	$147,%xmm0,%xmm8
+	pshufd	$0x93,%xmm15,%xmm7
+	pshufd	$0x93,%xmm0,%xmm8
 	pxor	%xmm7,%xmm15
-	pshufd	$147,%xmm3,%xmm9
+	pshufd	$0x93,%xmm3,%xmm9
 	pxor	%xmm8,%xmm0
-	pshufd	$147,%xmm5,%xmm10
+	pshufd	$0x93,%xmm5,%xmm10
 	pxor	%xmm9,%xmm3
-	pshufd	$147,%xmm2,%xmm11
+	pshufd	$0x93,%xmm2,%xmm11
 	pxor	%xmm10,%xmm5
-	pshufd	$147,%xmm6,%xmm12
+	pshufd	$0x93,%xmm6,%xmm12
 	pxor	%xmm11,%xmm2
-	pshufd	$147,%xmm1,%xmm13
+	pshufd	$0x93,%xmm1,%xmm13
 	pxor	%xmm12,%xmm6
-	pshufd	$147,%xmm4,%xmm14
+	pshufd	$0x93,%xmm4,%xmm14
 	pxor	%xmm13,%xmm1
 	pxor	%xmm14,%xmm4
 
 	pxor	%xmm15,%xmm8
 	pxor	%xmm4,%xmm7
 	pxor	%xmm4,%xmm8
-	pshufd	$78,%xmm15,%xmm15
+	pshufd	$0x4E,%xmm15,%xmm15
 	pxor	%xmm0,%xmm9
-	pshufd	$78,%xmm0,%xmm0
+	pshufd	$0x4E,%xmm0,%xmm0
 	pxor	%xmm2,%xmm12
 	pxor	%xmm7,%xmm15
 	pxor	%xmm6,%xmm13
 	pxor	%xmm8,%xmm0
 	pxor	%xmm5,%xmm11
-	pshufd	$78,%xmm2,%xmm7
+	pshufd	$0x4E,%xmm2,%xmm7
 	pxor	%xmm1,%xmm14
-	pshufd	$78,%xmm6,%xmm8
+	pshufd	$0x4E,%xmm6,%xmm8
 	pxor	%xmm3,%xmm10
-	pshufd	$78,%xmm5,%xmm2
+	pshufd	$0x4E,%xmm5,%xmm2
 	pxor	%xmm4,%xmm10
-	pshufd	$78,%xmm4,%xmm6
+	pshufd	$0x4E,%xmm4,%xmm6
 	pxor	%xmm4,%xmm11
-	pshufd	$78,%xmm1,%xmm5
+	pshufd	$0x4E,%xmm1,%xmm5
 	pxor	%xmm11,%xmm7
-	pshufd	$78,%xmm3,%xmm1
+	pshufd	$0x4E,%xmm3,%xmm1
 	pxor	%xmm12,%xmm8
 	pxor	%xmm10,%xmm2
 	pxor	%xmm14,%xmm6
@@ -796,24 +796,24 @@ L$dec_sbox:
 	decl	%r10d
 	jl	L$dec_done
 
-	pshufd	$78,%xmm15,%xmm7
-	pshufd	$78,%xmm2,%xmm13
+	pshufd	$0x4E,%xmm15,%xmm7
+	pshufd	$0x4E,%xmm2,%xmm13
 	pxor	%xmm15,%xmm7
-	pshufd	$78,%xmm4,%xmm14
+	pshufd	$0x4E,%xmm4,%xmm14
 	pxor	%xmm2,%xmm13
-	pshufd	$78,%xmm0,%xmm8
+	pshufd	$0x4E,%xmm0,%xmm8
 	pxor	%xmm4,%xmm14
-	pshufd	$78,%xmm5,%xmm9
+	pshufd	$0x4E,%xmm5,%xmm9
 	pxor	%xmm0,%xmm8
-	pshufd	$78,%xmm3,%xmm10
+	pshufd	$0x4E,%xmm3,%xmm10
 	pxor	%xmm5,%xmm9
 	pxor	%xmm13,%xmm15
 	pxor	%xmm13,%xmm0
-	pshufd	$78,%xmm1,%xmm11
+	pshufd	$0x4E,%xmm1,%xmm11
 	pxor	%xmm3,%xmm10
 	pxor	%xmm7,%xmm5
 	pxor	%xmm8,%xmm3
-	pshufd	$78,%xmm6,%xmm12
+	pshufd	$0x4E,%xmm6,%xmm12
 	pxor	%xmm1,%xmm11
 	pxor	%xmm14,%xmm0
 	pxor	%xmm9,%xmm1
@@ -827,45 +827,45 @@ L$dec_sbox:
 	pxor	%xmm14,%xmm1
 	pxor	%xmm14,%xmm6
 	pxor	%xmm12,%xmm4
-	pshufd	$147,%xmm15,%xmm7
-	pshufd	$147,%xmm0,%xmm8
+	pshufd	$0x93,%xmm15,%xmm7
+	pshufd	$0x93,%xmm0,%xmm8
 	pxor	%xmm7,%xmm15
-	pshufd	$147,%xmm5,%xmm9
+	pshufd	$0x93,%xmm5,%xmm9
 	pxor	%xmm8,%xmm0
-	pshufd	$147,%xmm3,%xmm10
+	pshufd	$0x93,%xmm3,%xmm10
 	pxor	%xmm9,%xmm5
-	pshufd	$147,%xmm1,%xmm11
+	pshufd	$0x93,%xmm1,%xmm11
 	pxor	%xmm10,%xmm3
-	pshufd	$147,%xmm6,%xmm12
+	pshufd	$0x93,%xmm6,%xmm12
 	pxor	%xmm11,%xmm1
-	pshufd	$147,%xmm2,%xmm13
+	pshufd	$0x93,%xmm2,%xmm13
 	pxor	%xmm12,%xmm6
-	pshufd	$147,%xmm4,%xmm14
+	pshufd	$0x93,%xmm4,%xmm14
 	pxor	%xmm13,%xmm2
 	pxor	%xmm14,%xmm4
 
 	pxor	%xmm15,%xmm8
 	pxor	%xmm4,%xmm7
 	pxor	%xmm4,%xmm8
-	pshufd	$78,%xmm15,%xmm15
+	pshufd	$0x4E,%xmm15,%xmm15
 	pxor	%xmm0,%xmm9
-	pshufd	$78,%xmm0,%xmm0
+	pshufd	$0x4E,%xmm0,%xmm0
 	pxor	%xmm1,%xmm12
 	pxor	%xmm7,%xmm15
 	pxor	%xmm6,%xmm13
 	pxor	%xmm8,%xmm0
 	pxor	%xmm3,%xmm11
-	pshufd	$78,%xmm1,%xmm7
+	pshufd	$0x4E,%xmm1,%xmm7
 	pxor	%xmm2,%xmm14
-	pshufd	$78,%xmm6,%xmm8
+	pshufd	$0x4E,%xmm6,%xmm8
 	pxor	%xmm5,%xmm10
-	pshufd	$78,%xmm3,%xmm1
+	pshufd	$0x4E,%xmm3,%xmm1
 	pxor	%xmm4,%xmm10
-	pshufd	$78,%xmm4,%xmm6
+	pshufd	$0x4E,%xmm4,%xmm6
 	pxor	%xmm4,%xmm11
-	pshufd	$78,%xmm2,%xmm3
+	pshufd	$0x4E,%xmm2,%xmm3
 	pxor	%xmm11,%xmm7
-	pshufd	$78,%xmm5,%xmm2
+	pshufd	$0x4E,%xmm5,%xmm2
 	pxor	%xmm12,%xmm8
 	pxor	%xmm1,%xmm10
 	pxor	%xmm14,%xmm6
@@ -1552,20 +1552,20 @@ L$xts_enc_prologue:
 	movdqa	%xmm7,(%rax)
 
 	andq	$-16,%r14
-	subq	$128,%rsp
+	subq	$0x80,%rsp
 	movdqa	32(%rbp),%xmm6
 
 	pxor	%xmm14,%xmm14
 	movdqa	L$xts_magic(%rip),%xmm12
 	pcmpgtd	%xmm6,%xmm14
 
-	subq	$128,%r14
+	subq	$0x80,%r14
 	jc	L$xts_enc_short
 	jmp	L$xts_enc_loop
 
 .p2align	4
 L$xts_enc_loop:
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm15
 	movdqa	%xmm6,0(%rsp)
@@ -1573,7 +1573,7 @@ L$xts_enc_loop:
 	pand	%xmm12,%xmm13
 	pcmpgtd	%xmm6,%xmm14
 	pxor	%xmm13,%xmm6
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm0
 	movdqa	%xmm6,16(%rsp)
@@ -1582,7 +1582,7 @@ L$xts_enc_loop:
 	pcmpgtd	%xmm6,%xmm14
 	pxor	%xmm13,%xmm6
 	movdqu	0(%r12),%xmm7
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm1
 	movdqa	%xmm6,32(%rsp)
@@ -1592,7 +1592,7 @@ L$xts_enc_loop:
 	pxor	%xmm13,%xmm6
 	movdqu	16(%r12),%xmm8
 	pxor	%xmm7,%xmm15
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm2
 	movdqa	%xmm6,48(%rsp)
@@ -1602,7 +1602,7 @@ L$xts_enc_loop:
 	pxor	%xmm13,%xmm6
 	movdqu	32(%r12),%xmm9
 	pxor	%xmm8,%xmm0
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm3
 	movdqa	%xmm6,64(%rsp)
@@ -1612,7 +1612,7 @@ L$xts_enc_loop:
 	pxor	%xmm13,%xmm6
 	movdqu	48(%r12),%xmm10
 	pxor	%xmm9,%xmm1
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm4
 	movdqa	%xmm6,80(%rsp)
@@ -1622,7 +1622,7 @@ L$xts_enc_loop:
 	pxor	%xmm13,%xmm6
 	movdqu	64(%r12),%xmm11
 	pxor	%xmm10,%xmm2
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm5
 	movdqa	%xmm6,96(%rsp)
@@ -1666,20 +1666,20 @@ L$xts_enc_loop:
 	pxor	%xmm14,%xmm14
 	movdqa	L$xts_magic(%rip),%xmm12
 	pcmpgtd	%xmm6,%xmm14
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	paddq	%xmm6,%xmm6
 	pand	%xmm12,%xmm13
 	pcmpgtd	%xmm6,%xmm14
 	pxor	%xmm13,%xmm6
 
-	subq	$128,%r14
+	subq	$0x80,%r14
 	jnc	L$xts_enc_loop
 
 L$xts_enc_short:
-	addq	$128,%r14
+	addq	$0x80,%r14
 	jz	L$xts_enc_done
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm15
 	movdqa	%xmm6,0(%rsp)
@@ -1687,7 +1687,7 @@ L$xts_enc_short:
 	pand	%xmm12,%xmm13
 	pcmpgtd	%xmm6,%xmm14
 	pxor	%xmm13,%xmm6
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm0
 	movdqa	%xmm6,16(%rsp)
@@ -1698,7 +1698,7 @@ L$xts_enc_short:
 	movdqu	0(%r12),%xmm7
 	cmpq	$16,%r14
 	je	L$xts_enc_1
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm1
 	movdqa	%xmm6,32(%rsp)
@@ -1710,7 +1710,7 @@ L$xts_enc_short:
 	cmpq	$32,%r14
 	je	L$xts_enc_2
 	pxor	%xmm7,%xmm15
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm2
 	movdqa	%xmm6,48(%rsp)
@@ -1722,7 +1722,7 @@ L$xts_enc_short:
 	cmpq	$48,%r14
 	je	L$xts_enc_3
 	pxor	%xmm8,%xmm0
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm3
 	movdqa	%xmm6,64(%rsp)
@@ -1734,7 +1734,7 @@ L$xts_enc_short:
 	cmpq	$64,%r14
 	je	L$xts_enc_4
 	pxor	%xmm9,%xmm1
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm4
 	movdqa	%xmm6,80(%rsp)
@@ -1746,7 +1746,7 @@ L$xts_enc_short:
 	cmpq	$80,%r14
 	je	L$xts_enc_5
 	pxor	%xmm10,%xmm2
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm5
 	movdqa	%xmm6,96(%rsp)
@@ -2011,20 +2011,20 @@ L$xts_dec_prologue:
 	shlq	$4,%rax
 	subq	%rax,%r14
 
-	subq	$128,%rsp
+	subq	$0x80,%rsp
 	movdqa	32(%rbp),%xmm6
 
 	pxor	%xmm14,%xmm14
 	movdqa	L$xts_magic(%rip),%xmm12
 	pcmpgtd	%xmm6,%xmm14
 
-	subq	$128,%r14
+	subq	$0x80,%r14
 	jc	L$xts_dec_short
 	jmp	L$xts_dec_loop
 
 .p2align	4
 L$xts_dec_loop:
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm15
 	movdqa	%xmm6,0(%rsp)
@@ -2032,7 +2032,7 @@ L$xts_dec_loop:
 	pand	%xmm12,%xmm13
 	pcmpgtd	%xmm6,%xmm14
 	pxor	%xmm13,%xmm6
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm0
 	movdqa	%xmm6,16(%rsp)
@@ -2041,7 +2041,7 @@ L$xts_dec_loop:
 	pcmpgtd	%xmm6,%xmm14
 	pxor	%xmm13,%xmm6
 	movdqu	0(%r12),%xmm7
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm1
 	movdqa	%xmm6,32(%rsp)
@@ -2051,7 +2051,7 @@ L$xts_dec_loop:
 	pxor	%xmm13,%xmm6
 	movdqu	16(%r12),%xmm8
 	pxor	%xmm7,%xmm15
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm2
 	movdqa	%xmm6,48(%rsp)
@@ -2061,7 +2061,7 @@ L$xts_dec_loop:
 	pxor	%xmm13,%xmm6
 	movdqu	32(%r12),%xmm9
 	pxor	%xmm8,%xmm0
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm3
 	movdqa	%xmm6,64(%rsp)
@@ -2071,7 +2071,7 @@ L$xts_dec_loop:
 	pxor	%xmm13,%xmm6
 	movdqu	48(%r12),%xmm10
 	pxor	%xmm9,%xmm1
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm4
 	movdqa	%xmm6,80(%rsp)
@@ -2081,7 +2081,7 @@ L$xts_dec_loop:
 	pxor	%xmm13,%xmm6
 	movdqu	64(%r12),%xmm11
 	pxor	%xmm10,%xmm2
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm5
 	movdqa	%xmm6,96(%rsp)
@@ -2125,20 +2125,20 @@ L$xts_dec_loop:
 	pxor	%xmm14,%xmm14
 	movdqa	L$xts_magic(%rip),%xmm12
 	pcmpgtd	%xmm6,%xmm14
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	paddq	%xmm6,%xmm6
 	pand	%xmm12,%xmm13
 	pcmpgtd	%xmm6,%xmm14
 	pxor	%xmm13,%xmm6
 
-	subq	$128,%r14
+	subq	$0x80,%r14
 	jnc	L$xts_dec_loop
 
 L$xts_dec_short:
-	addq	$128,%r14
+	addq	$0x80,%r14
 	jz	L$xts_dec_done
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm15
 	movdqa	%xmm6,0(%rsp)
@@ -2146,7 +2146,7 @@ L$xts_dec_short:
 	pand	%xmm12,%xmm13
 	pcmpgtd	%xmm6,%xmm14
 	pxor	%xmm13,%xmm6
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm0
 	movdqa	%xmm6,16(%rsp)
@@ -2157,7 +2157,7 @@ L$xts_dec_short:
 	movdqu	0(%r12),%xmm7
 	cmpq	$16,%r14
 	je	L$xts_dec_1
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm1
 	movdqa	%xmm6,32(%rsp)
@@ -2169,7 +2169,7 @@ L$xts_dec_short:
 	cmpq	$32,%r14
 	je	L$xts_dec_2
 	pxor	%xmm7,%xmm15
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm2
 	movdqa	%xmm6,48(%rsp)
@@ -2181,7 +2181,7 @@ L$xts_dec_short:
 	cmpq	$48,%r14
 	je	L$xts_dec_3
 	pxor	%xmm8,%xmm0
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm3
 	movdqa	%xmm6,64(%rsp)
@@ -2193,7 +2193,7 @@ L$xts_dec_short:
 	cmpq	$64,%r14
 	je	L$xts_dec_4
 	pxor	%xmm9,%xmm1
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm4
 	movdqa	%xmm6,80(%rsp)
@@ -2205,7 +2205,7 @@ L$xts_dec_short:
 	cmpq	$80,%r14
 	je	L$xts_dec_5
 	pxor	%xmm10,%xmm2
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	pxor	%xmm14,%xmm14
 	movdqa	%xmm6,%xmm5
 	movdqa	%xmm6,96(%rsp)
@@ -2382,7 +2382,7 @@ L$xts_dec_done:
 	pxor	%xmm14,%xmm14
 	movdqa	L$xts_magic(%rip),%xmm12
 	pcmpgtd	%xmm6,%xmm14
-	pshufd	$19,%xmm14,%xmm13
+	pshufd	$0x13,%xmm14,%xmm13
 	movdqa	%xmm6,%xmm5
 	paddq	%xmm6,%xmm6
 	pand	%xmm12,%xmm13

deps/openssl/asm/x64-macosx-gas/aes/vpaes-x86_64.s

@@ -60,7 +60,7 @@ L$enc_loop:
 	addq	$16,%r11
 	pxor	%xmm0,%xmm3
 .byte	102,15,56,0,193
-	andq	$48,%r11
+	andq	$0x30,%r11
 	subq	$1,%rax
 	pxor	%xmm3,%xmm0
 
@@ -120,10 +120,10 @@ _vpaes_decrypt_core:
 	pand	%xmm9,%xmm0
 .byte	102,15,56,0,208
 	movdqa	L$k_dipt+16(%rip),%xmm0
-	xorq	$48,%r11
+	xorq	$0x30,%r11
 	leaq	L$k_dsbd(%rip),%r10
 .byte	102,15,56,0,193
-	andq	$48,%r11
+	andq	$0x30,%r11
 	pxor	%xmm5,%xmm2
 	movdqa	L$k_mc_forward+48(%rip),%xmm5
 	pxor	%xmm2,%xmm0
@@ -242,7 +242,7 @@ L$schedule_am_decrypting:
 	movdqa	(%r8,%r10,1),%xmm1
 .byte	102,15,56,0,217
 	movdqu	%xmm3,(%rdx)
-	xorq	$48,%r8
+	xorq	$0x30,%r8
 
 L$schedule_go:
 	cmpl	$192,%esi
@@ -332,7 +332,7 @@ L$oop_schedule_256:
 	call	_vpaes_schedule_mangle
 
 
-	pshufd	$255,%xmm0,%xmm0
+	pshufd	$0xFF,%xmm0,%xmm0
 	movdqa	%xmm7,%xmm5
 	movdqa	%xmm6,%xmm7
 	call	_vpaes_schedule_low_round
@@ -399,8 +399,8 @@ L$schedule_mangle_last_dec:
 
 .p2align	4
 _vpaes_schedule_192_smear:
-	pshufd	$128,%xmm6,%xmm1
-	pshufd	$254,%xmm7,%xmm0
+	pshufd	$0x80,%xmm6,%xmm1
+	pshufd	$0xFE,%xmm7,%xmm0
 	pxor	%xmm1,%xmm6
 	pxor	%xmm1,%xmm1
 	pxor	%xmm0,%xmm6
@@ -437,7 +437,7 @@ _vpaes_schedule_round:
 	pxor	%xmm1,%xmm7
 
 
-	pshufd	$255,%xmm0,%xmm0
+	pshufd	$0xFF,%xmm0,%xmm0
 .byte	102,15,58,15,192,1
 
 
@@ -596,7 +596,7 @@ L$schedule_mangle_both:
 	movdqa	(%r8,%r10,1),%xmm1
 .byte	102,15,56,0,217
 	addq	$-16,%r8
-	andq	$48,%r8
+	andq	$0x30,%r8
 	movdqu	%xmm3,(%rdx)
 	.byte	0xf3,0xc3
 
@@ -614,7 +614,7 @@ _vpaes_set_encrypt_key:
 	movl	%eax,240(%rdx)
 
 	movl	$0,%ecx
-	movl	$48,%r8d
+	movl	$0x30,%r8d
 	call	_vpaes_schedule_core
 	xorl	%eax,%eax
 	.byte	0xf3,0xc3

deps/openssl/asm/x64-macosx-gas/bn/rsaz-x86_64.s

@@ -19,6 +19,10 @@ L$sqr_body:
 	movq	(%rsi),%rdx
 	movq	8(%rsi),%rax
 	movq	%rcx,128(%rsp)
+	movl	$0x80100,%r11d
+	andl	_OPENSSL_ia32cap_P+8(%rip),%r11d
+	cmpl	$0x80100,%r11d
+	je	L$oop_sqrx
 	jmp	L$oop_sqr
 
 .p2align	5
@@ -382,6 +386,276 @@ L$oop_sqr:
 
 	decl	%r8d
 	jnz	L$oop_sqr
+	jmp	L$sqr_tail
+
+.p2align	5
+L$oop_sqrx:
+	movl	%r8d,128+8(%rsp)
+.byte	102,72,15,110,199
+.byte	102,72,15,110,205
+
+	mulxq	%rax,%r8,%r9
+
+	mulxq	16(%rsi),%rcx,%r10
+	xorq	%rbp,%rbp
+
+	mulxq	24(%rsi),%rax,%r11
+	adcxq	%rcx,%r9
+
+	mulxq	32(%rsi),%rcx,%r12
+	adcxq	%rax,%r10
+
+	mulxq	40(%rsi),%rax,%r13
+	adcxq	%rcx,%r11
+
+.byte	0xc4,0x62,0xf3,0xf6,0xb6,0x30,0x00,0x00,0x00
+	adcxq	%rax,%r12
+	adcxq	%rcx,%r13
+
+.byte	0xc4,0x62,0xfb,0xf6,0xbe,0x38,0x00,0x00,0x00
+	adcxq	%rax,%r14
+	adcxq	%rbp,%r15
+
+	movq	%r9,%rcx
+	shldq	$1,%r8,%r9
+	shlq	$1,%r8
+
+	xorl	%ebp,%ebp
+	mulxq	%rdx,%rax,%rdx
+	adcxq	%rdx,%r8
+	movq	8(%rsi),%rdx
+	adcxq	%rbp,%r9
+
+	movq	%rax,(%rsp)
+	movq	%r8,8(%rsp)
+
+
+	mulxq	16(%rsi),%rax,%rbx
+	adoxq	%rax,%r10
+	adcxq	%rbx,%r11
+
+.byte	0xc4,0x62,0xc3,0xf6,0x86,0x18,0x00,0x00,0x00
+	adoxq	%rdi,%r11
+	adcxq	%r8,%r12
+
+	mulxq	32(%rsi),%rax,%rbx
+	adoxq	%rax,%r12
+	adcxq	%rbx,%r13
+
+	mulxq	40(%rsi),%rdi,%r8
+	adoxq	%rdi,%r13
+	adcxq	%r8,%r14
+
+.byte	0xc4,0xe2,0xfb,0xf6,0x9e,0x30,0x00,0x00,0x00
+	adoxq	%rax,%r14
+	adcxq	%rbx,%r15
+
+.byte	0xc4,0x62,0xc3,0xf6,0x86,0x38,0x00,0x00,0x00
+	adoxq	%rdi,%r15
+	adcxq	%rbp,%r8
+	adoxq	%rbp,%r8
+
+	movq	%r11,%rbx
+	shldq	$1,%r10,%r11
+	shldq	$1,%rcx,%r10
+
+	xorl	%ebp,%ebp
+	mulxq	%rdx,%rax,%rcx
+	movq	16(%rsi),%rdx
+	adcxq	%rax,%r9
+	adcxq	%rcx,%r10
+	adcxq	%rbp,%r11
+
+	movq	%r9,16(%rsp)
+.byte	0x4c,0x89,0x94,0x24,0x18,0x00,0x00,0x00
+
+
+.byte	0xc4,0x62,0xc3,0xf6,0x8e,0x18,0x00,0x00,0x00
+	adoxq	%rdi,%r12
+	adcxq	%r9,%r13
+
+	mulxq	32(%rsi),%rax,%rcx
+	adoxq	%rax,%r13
+	adcxq	%rcx,%r14
+
+	mulxq	40(%rsi),%rdi,%r9
+	adoxq	%rdi,%r14
+	adcxq	%r9,%r15
+
+.byte	0xc4,0xe2,0xfb,0xf6,0x8e,0x30,0x00,0x00,0x00
+	adoxq	%rax,%r15
+	adcxq	%rcx,%r8
+
+.byte	0xc4,0x62,0xc3,0xf6,0x8e,0x38,0x00,0x00,0x00
+	adoxq	%rdi,%r8
+	adcxq	%rbp,%r9
+	adoxq	%rbp,%r9
+
+	movq	%r13,%rcx
+	shldq	$1,%r12,%r13
+	shldq	$1,%rbx,%r12
+
+	xorl	%ebp,%ebp
+	mulxq	%rdx,%rax,%rdx
+	adcxq	%rax,%r11
+	adcxq	%rdx,%r12
+	movq	24(%rsi),%rdx
+	adcxq	%rbp,%r13
+
+	movq	%r11,32(%rsp)
+.byte	0x4c,0x89,0xa4,0x24,0x28,0x00,0x00,0x00
+
+
+.byte	0xc4,0xe2,0xfb,0xf6,0x9e,0x20,0x00,0x00,0x00
+	adoxq	%rax,%r14
+	adcxq	%rbx,%r15
+
+	mulxq	40(%rsi),%rdi,%r10
+	adoxq	%rdi,%r15
+	adcxq	%r10,%r8
+
+	mulxq	48(%rsi),%rax,%rbx
+	adoxq	%rax,%r8
+	adcxq	%rbx,%r9
+
+	mulxq	56(%rsi),%rdi,%r10
+	adoxq	%rdi,%r9
+	adcxq	%rbp,%r10
+	adoxq	%rbp,%r10
+
+.byte	0x66
+	movq	%r15,%rbx
+	shldq	$1,%r14,%r15
+	shldq	$1,%rcx,%r14
+
+	xorl	%ebp,%ebp
+	mulxq	%rdx,%rax,%rdx
+	adcxq	%rax,%r13
+	adcxq	%rdx,%r14
+	movq	32(%rsi),%rdx
+	adcxq	%rbp,%r15
+
+	movq	%r13,48(%rsp)
+	movq	%r14,56(%rsp)
+
+
+.byte	0xc4,0x62,0xc3,0xf6,0x9e,0x28,0x00,0x00,0x00
+	adoxq	%rdi,%r8
+	adcxq	%r11,%r9
+
+	mulxq	48(%rsi),%rax,%rcx
+	adoxq	%rax,%r9
+	adcxq	%rcx,%r10
+
+	mulxq	56(%rsi),%rdi,%r11
+	adoxq	%rdi,%r10
+	adcxq	%rbp,%r11
+	adoxq	%rbp,%r11
+
+	movq	%r9,%rcx
+	shldq	$1,%r8,%r9
+	shldq	$1,%rbx,%r8
+
+	xorl	%ebp,%ebp
+	mulxq	%rdx,%rax,%rdx
+	adcxq	%rax,%r15
+	adcxq	%rdx,%r8
+	movq	40(%rsi),%rdx
+	adcxq	%rbp,%r9
+
+	movq	%r15,64(%rsp)
+	movq	%r8,72(%rsp)
+
+
+.byte	0xc4,0xe2,0xfb,0xf6,0x9e,0x30,0x00,0x00,0x00
+	adoxq	%rax,%r10
+	adcxq	%rbx,%r11
+
+.byte	0xc4,0x62,0xc3,0xf6,0xa6,0x38,0x00,0x00,0x00
+	adoxq	%rdi,%r11
+	adcxq	%rbp,%r12
+	adoxq	%rbp,%r12
+
+	movq	%r11,%rbx
+	shldq	$1,%r10,%r11
+	shldq	$1,%rcx,%r10
+
+	xorl	%ebp,%ebp
+	mulxq	%rdx,%rax,%rdx
+	adcxq	%rax,%r9
+	adcxq	%rdx,%r10
+	movq	48(%rsi),%rdx
+	adcxq	%rbp,%r11
+
+	movq	%r9,80(%rsp)
+	movq	%r10,88(%rsp)
+
+
+.byte	0xc4,0x62,0xfb,0xf6,0xae,0x38,0x00,0x00,0x00
+	adoxq	%rax,%r12
+	adoxq	%rbp,%r13
+
+	xorq	%r14,%r14
+	shldq	$1,%r13,%r14
+	shldq	$1,%r12,%r13
+	shldq	$1,%rbx,%r12
+
+	xorl	%ebp,%ebp
+	mulxq	%rdx,%rax,%rdx
+	adcxq	%rax,%r11
+	adcxq	%rdx,%r12
+	movq	56(%rsi),%rdx
+	adcxq	%rbp,%r13
+
+.byte	0x4c,0x89,0x9c,0x24,0x60,0x00,0x00,0x00
+.byte	0x4c,0x89,0xa4,0x24,0x68,0x00,0x00,0x00
+
+
+	mulxq	%rdx,%rax,%rdx
+	adoxq	%rax,%r13
+	adoxq	%rbp,%rdx
+
+.byte	0x66
+	addq	%rdx,%r14
+
+	movq	%r13,112(%rsp)
+	movq	%r14,120(%rsp)
+.byte	102,72,15,126,199
+.byte	102,72,15,126,205
+
+	movq	128(%rsp),%rdx
+	movq	(%rsp),%r8
+	movq	8(%rsp),%r9
+	movq	16(%rsp),%r10
+	movq	24(%rsp),%r11
+	movq	32(%rsp),%r12
+	movq	40(%rsp),%r13
+	movq	48(%rsp),%r14
+	movq	56(%rsp),%r15
+
+	call	__rsaz_512_reducex
+
+	addq	64(%rsp),%r8
+	adcq	72(%rsp),%r9
+	adcq	80(%rsp),%r10
+	adcq	88(%rsp),%r11
+	adcq	96(%rsp),%r12
+	adcq	104(%rsp),%r13
+	adcq	112(%rsp),%r14
+	adcq	120(%rsp),%r15
+	sbbq	%rcx,%rcx
+
+	call	__rsaz_512_subtract
+
+	movq	%r8,%rdx
+	movq	%r9,%rax
+	movl	128+8(%rsp),%r8d
+	movq	%rdi,%rsi
+
+	decl	%r8d
+	jnz	L$oop_sqrx
+
+L$sqr_tail:
 
 	leaq	128+24+48(%rsp),%rax
 	movq	-48(%rax),%r15
@@ -410,6 +684,10 @@ L$mul_body:
 .byte	102,72,15,110,199
 .byte	102,72,15,110,201
 	movq	%r8,128(%rsp)
+	movl	$0x80100,%r11d
+	andl	_OPENSSL_ia32cap_P+8(%rip),%r11d
+	cmpl	$0x80100,%r11d
+	je	L$mulx
 	movq	(%rdx),%rbx
 	movq	%rdx,%rbp
 	call	__rsaz_512_mul
@@ -427,6 +705,29 @@ L$mul_body:
 	movq	56(%rsp),%r15
 
 	call	__rsaz_512_reduce
+	jmp	L$mul_tail
+
+.p2align	5
+L$mulx:
+	movq	%rdx,%rbp
+	movq	(%rdx),%rdx
+	call	__rsaz_512_mulx
+
+.byte	102,72,15,126,199
+.byte	102,72,15,126,205
+
+	movq	128(%rsp),%rdx
+	movq	(%rsp),%r8
+	movq	8(%rsp),%r9
+	movq	16(%rsp),%r10
+	movq	24(%rsp),%r11
+	movq	32(%rsp),%r12
+	movq	40(%rsp),%r13
+	movq	48(%rsp),%r14
+	movq	56(%rsp),%r15
+
+	call	__rsaz_512_reducex
+L$mul_tail:
 	addq	64(%rsp),%r8
 	adcq	72(%rsp),%r9
 	adcq	80(%rsp),%r10
@@ -517,6 +818,10 @@ L$mul_gather4_body:
 	por	%xmm9,%xmm8
 	pshufd	$0x4e,%xmm8,%xmm9
 	por	%xmm9,%xmm8
+	movl	$0x80100,%r11d
+	andl	_OPENSSL_ia32cap_P+8(%rip),%r11d
+	cmpl	$0x80100,%r11d
+	je	L$mulx_gather
 .byte	102,76,15,126,195
 
 	movq	%r8,128(%rsp)
@@ -697,6 +1002,142 @@ L$oop_mul_gather:
 	movq	56(%rsp),%r15
 
 	call	__rsaz_512_reduce
+	jmp	L$mul_gather_tail
+
+.p2align	5
+L$mulx_gather:
+.byte	102,76,15,126,194
+
+	movq	%r8,128(%rsp)
+	movq	%rdi,128+8(%rsp)
+	movq	%rcx,128+16(%rsp)
+
+	mulxq	(%rsi),%rbx,%r8
+	movq	%rbx,(%rsp)
+	xorl	%edi,%edi
+
+	mulxq	8(%rsi),%rax,%r9
+
+	mulxq	16(%rsi),%rbx,%r10
+	adcxq	%rax,%r8
+
+	mulxq	24(%rsi),%rax,%r11
+	adcxq	%rbx,%r9
+
+	mulxq	32(%rsi),%rbx,%r12
+	adcxq	%rax,%r10
+
+	mulxq	40(%rsi),%rax,%r13
+	adcxq	%rbx,%r11
+
+	mulxq	48(%rsi),%rbx,%r14
+	adcxq	%rax,%r12
+
+	mulxq	56(%rsi),%rax,%r15
+	adcxq	%rbx,%r13
+	adcxq	%rax,%r14
+.byte	0x67
+	movq	%r8,%rbx
+	adcxq	%rdi,%r15
+
+	movq	$-7,%rcx
+	jmp	L$oop_mulx_gather
+
+.p2align	5
+L$oop_mulx_gather:
+	movdqa	0(%rbp),%xmm8
+	movdqa	16(%rbp),%xmm9
+	movdqa	32(%rbp),%xmm10
+	movdqa	48(%rbp),%xmm11
+	pand	%xmm0,%xmm8
+	movdqa	64(%rbp),%xmm12
+	pand	%xmm1,%xmm9
+	movdqa	80(%rbp),%xmm13
+	pand	%xmm2,%xmm10
+	movdqa	96(%rbp),%xmm14
+	pand	%xmm3,%xmm11
+	movdqa	112(%rbp),%xmm15
+	leaq	128(%rbp),%rbp
+	pand	%xmm4,%xmm12
+	pand	%xmm5,%xmm13
+	pand	%xmm6,%xmm14
+	pand	%xmm7,%xmm15
+	por	%xmm10,%xmm8
+	por	%xmm11,%xmm9
+	por	%xmm12,%xmm8
+	por	%xmm13,%xmm9
+	por	%xmm14,%xmm8
+	por	%xmm15,%xmm9
+
+	por	%xmm9,%xmm8
+	pshufd	$0x4e,%xmm8,%xmm9
+	por	%xmm9,%xmm8
+.byte	102,76,15,126,194
+
+.byte	0xc4,0x62,0xfb,0xf6,0x86,0x00,0x00,0x00,0x00
+	adcxq	%rax,%rbx
+	adoxq	%r9,%r8
+
+	mulxq	8(%rsi),%rax,%r9
+	adcxq	%rax,%r8
+	adoxq	%r10,%r9
+
+	mulxq	16(%rsi),%rax,%r10
+	adcxq	%rax,%r9
+	adoxq	%r11,%r10
+
+.byte	0xc4,0x62,0xfb,0xf6,0x9e,0x18,0x00,0x00,0x00
+	adcxq	%rax,%r10
+	adoxq	%r12,%r11
+
+	mulxq	32(%rsi),%rax,%r12
+	adcxq	%rax,%r11
+	adoxq	%r13,%r12
+
+	mulxq	40(%rsi),%rax,%r13
+	adcxq	%rax,%r12
+	adoxq	%r14,%r13
+
+.byte	0xc4,0x62,0xfb,0xf6,0xb6,0x30,0x00,0x00,0x00
+	adcxq	%rax,%r13
+.byte	0x67
+	adoxq	%r15,%r14
+
+	mulxq	56(%rsi),%rax,%r15
+	movq	%rbx,64(%rsp,%rcx,8)
+	adcxq	%rax,%r14
+	adoxq	%rdi,%r15
+	movq	%r8,%rbx
+	adcxq	%rdi,%r15
+
+	incq	%rcx
+	jnz	L$oop_mulx_gather
+
+	movq	%r8,64(%rsp)
+	movq	%r9,64+8(%rsp)
+	movq	%r10,64+16(%rsp)
+	movq	%r11,64+24(%rsp)
+	movq	%r12,64+32(%rsp)
+	movq	%r13,64+40(%rsp)
+	movq	%r14,64+48(%rsp)
+	movq	%r15,64+56(%rsp)
+
+	movq	128(%rsp),%rdx
+	movq	128+8(%rsp),%rdi
+	movq	128+16(%rsp),%rbp
+
+	movq	(%rsp),%r8
+	movq	8(%rsp),%r9
+	movq	16(%rsp),%r10
+	movq	24(%rsp),%r11
+	movq	32(%rsp),%r12
+	movq	40(%rsp),%r13
+	movq	48(%rsp),%r14
+	movq	56(%rsp),%r15
+
+	call	__rsaz_512_reducex
+
+L$mul_gather_tail:
 	addq	64(%rsp),%r8
 	adcq	72(%rsp),%r9
 	adcq	80(%rsp),%r10
@@ -741,6 +1182,10 @@ L$mul_scatter4_body:
 	movq	%rcx,128(%rsp)
 
 	movq	%rdi,%rbp
+	movl	$0x80100,%r11d
+	andl	_OPENSSL_ia32cap_P+8(%rip),%r11d
+	cmpl	$0x80100,%r11d
+	je	L$mulx_scatter
 	movq	(%rdi),%rbx
 	call	__rsaz_512_mul
 
@@ -757,6 +1202,29 @@ L$mul_scatter4_body:
 	movq	56(%rsp),%r15
 
 	call	__rsaz_512_reduce
+	jmp	L$mul_scatter_tail
+
+.p2align	5
+L$mulx_scatter:
+	movq	(%rdi),%rdx
+	call	__rsaz_512_mulx
+
+.byte	102,72,15,126,199
+.byte	102,72,15,126,205
+
+	movq	128(%rsp),%rdx
+	movq	(%rsp),%r8
+	movq	8(%rsp),%r9
+	movq	16(%rsp),%r10
+	movq	24(%rsp),%r11
+	movq	32(%rsp),%r12
+	movq	40(%rsp),%r13
+	movq	48(%rsp),%r14
+	movq	56(%rsp),%r15
+
+	call	__rsaz_512_reducex
+
+L$mul_scatter_tail:
 	addq	64(%rsp),%r8
 	adcq	72(%rsp),%r9
 	adcq	80(%rsp),%r10
@@ -803,6 +1271,7 @@ _rsaz_512_mul_by_one:
 
 	subq	$128+24,%rsp
 L$mul_by_one_body:
+	movl	_OPENSSL_ia32cap_P+8(%rip),%eax
 	movq	%rdx,%rbp
 	movq	%rcx,128(%rsp)
 
@@ -823,7 +1292,16 @@ L$mul_by_one_body:
 	movdqa	%xmm0,64(%rsp)
 	movdqa	%xmm0,80(%rsp)
 	movdqa	%xmm0,96(%rsp)
+	andl	$0x80100,%eax
+	cmpl	$0x80100,%eax
+	je	L$by_one_callx
 	call	__rsaz_512_reduce
+	jmp	L$by_one_tail
+.p2align	5
+L$by_one_callx:
+	movq	128(%rsp),%rdx
+	call	__rsaz_512_reducex
+L$by_one_tail:
 	movq	%r8,(%rdi)
 	movq	%r9,8(%rdi)
 	movq	%r10,16(%rdi)
@@ -928,6 +1406,62 @@ L$reduction_loop:
 	.byte	0xf3,0xc3
 
 
+.p2align	5
+__rsaz_512_reducex:
+
+	imulq	%r8,%rdx
+	xorq	%rsi,%rsi
+	movl	$8,%ecx
+	jmp	L$reduction_loopx
+
+.p2align	5
+L$reduction_loopx:
+	movq	%r8,%rbx
+	mulxq	0(%rbp),%rax,%r8
+	adcxq	%rbx,%rax
+	adoxq	%r9,%r8
+
+	mulxq	8(%rbp),%rax,%r9
+	adcxq	%rax,%r8
+	adoxq	%r10,%r9
+
+	mulxq	16(%rbp),%rbx,%r10
+	adcxq	%rbx,%r9
+	adoxq	%r11,%r10
+
+	mulxq	24(%rbp),%rbx,%r11
+	adcxq	%rbx,%r10
+	adoxq	%r12,%r11
+
+.byte	0xc4,0x62,0xe3,0xf6,0xa5,0x20,0x00,0x00,0x00
+	movq	%rdx,%rax
+	movq	%r8,%rdx
+	adcxq	%rbx,%r11
+	adoxq	%r13,%r12
+
+	mulxq	128+8(%rsp),%rbx,%rdx
+	movq	%rax,%rdx
+
+	mulxq	40(%rbp),%rax,%r13
+	adcxq	%rax,%r12
+	adoxq	%r14,%r13
+
+.byte	0xc4,0x62,0xfb,0xf6,0xb5,0x30,0x00,0x00,0x00
+	adcxq	%rax,%r13
+	adoxq	%r15,%r14
+
+	mulxq	56(%rbp),%rax,%r15
+	movq	%rbx,%rdx
+	adcxq	%rax,%r14
+	adoxq	%rsi,%r15
+	adcxq	%rsi,%r15
+
+	decl	%ecx
+	jne	L$reduction_loopx
+
+	.byte	0xf3,0xc3
+
+
 .p2align	5
 __rsaz_512_subtract:
 	movq	%r8,(%rdi)
@@ -1126,6 +1660,126 @@ L$oop_mul:
 
 	.byte	0xf3,0xc3
 
+
+.p2align	5
+__rsaz_512_mulx:
+	mulxq	(%rsi),%rbx,%r8
+	movq	$-6,%rcx
+
+	mulxq	8(%rsi),%rax,%r9
+	movq	%rbx,8(%rsp)
+
+	mulxq	16(%rsi),%rbx,%r10
+	adcq	%rax,%r8
+
+	mulxq	24(%rsi),%rax,%r11
+	adcq	%rbx,%r9
+
+	mulxq	32(%rsi),%rbx,%r12
+	adcq	%rax,%r10
+
+	mulxq	40(%rsi),%rax,%r13
+	adcq	%rbx,%r11
+
+	mulxq	48(%rsi),%rbx,%r14
+	adcq	%rax,%r12
+
+	mulxq	56(%rsi),%rax,%r15
+	movq	8(%rbp),%rdx
+	adcq	%rbx,%r13
+	adcq	%rax,%r14
+	adcq	$0,%r15
+
+	xorq	%rdi,%rdi
+	jmp	L$oop_mulx
+
+.p2align	5
+L$oop_mulx:
+	movq	%r8,%rbx
+	mulxq	(%rsi),%rax,%r8
+	adcxq	%rax,%rbx
+	adoxq	%r9,%r8
+
+	mulxq	8(%rsi),%rax,%r9
+	adcxq	%rax,%r8
+	adoxq	%r10,%r9
+
+	mulxq	16(%rsi),%rax,%r10
+	adcxq	%rax,%r9
+	adoxq	%r11,%r10
+
+	mulxq	24(%rsi),%rax,%r11
+	adcxq	%rax,%r10
+	adoxq	%r12,%r11
+
+.byte	0x3e,0xc4,0x62,0xfb,0xf6,0xa6,0x20,0x00,0x00,0x00
+	adcxq	%rax,%r11
+	adoxq	%r13,%r12
+
+	mulxq	40(%rsi),%rax,%r13
+	adcxq	%rax,%r12
+	adoxq	%r14,%r13
+
+	mulxq	48(%rsi),%rax,%r14
+	adcxq	%rax,%r13
+	adoxq	%r15,%r14
+
+	mulxq	56(%rsi),%rax,%r15
+	movq	64(%rbp,%rcx,8),%rdx
+	movq	%rbx,8+64-8(%rsp,%rcx,8)
+	adcxq	%rax,%r14
+	adoxq	%rdi,%r15
+	adcxq	%rdi,%r15
+
+	incq	%rcx
+	jnz	L$oop_mulx
+
+	movq	%r8,%rbx
+	mulxq	(%rsi),%rax,%r8
+	adcxq	%rax,%rbx
+	adoxq	%r9,%r8
+
+.byte	0xc4,0x62,0xfb,0xf6,0x8e,0x08,0x00,0x00,0x00
+	adcxq	%rax,%r8
+	adoxq	%r10,%r9
+
+.byte	0xc4,0x62,0xfb,0xf6,0x96,0x10,0x00,0x00,0x00
+	adcxq	%rax,%r9
+	adoxq	%r11,%r10
+
+	mulxq	24(%rsi),%rax,%r11
+	adcxq	%rax,%r10
+	adoxq	%r12,%r11
+
+	mulxq	32(%rsi),%rax,%r12
+	adcxq	%rax,%r11
+	adoxq	%r13,%r12
+
+	mulxq	40(%rsi),%rax,%r13
+	adcxq	%rax,%r12
+	adoxq	%r14,%r13
+
+.byte	0xc4,0x62,0xfb,0xf6,0xb6,0x30,0x00,0x00,0x00
+	adcxq	%rax,%r13
+	adoxq	%r15,%r14
+
+.byte	0xc4,0x62,0xfb,0xf6,0xbe,0x38,0x00,0x00,0x00
+	adcxq	%rax,%r14
+	adoxq	%rdi,%r15
+	adcxq	%rdi,%r15
+
+	movq	%rbx,8+64-8(%rsp)
+	movq	%r8,8+64(%rsp)
+	movq	%r9,8+64+8(%rsp)
+	movq	%r10,8+64+16(%rsp)
+	movq	%r11,8+64+24(%rsp)
+	movq	%r12,8+64+32(%rsp)
+	movq	%r13,8+64+40(%rsp)
+	movq	%r14,8+64+48(%rsp)
+	movq	%r15,8+64+56(%rsp)
+
+	.byte	0xf3,0xc3
+
 .globl	_rsaz_512_scatter4
 
 .p2align	4

deps/openssl/asm/x64-macosx-gas/bn/x86_64-mont.s

@@ -10,6 +10,7 @@ _bn_mul_mont:
 	jnz	L$mul_enter
 	cmpl	$8,%r9d
 	jb	L$mul_enter
+	movl	_OPENSSL_ia32cap_P+8(%rip),%r11d
 	cmpq	%rsi,%rdx
 	jne	L$mul4x_enter
 	testl	$7,%r9d
@@ -34,6 +35,20 @@ L$mul_enter:
 
 	movq	%r11,8(%rsp,%r9,8)
 L$mul_body:
+
+
+
+
+
+
+	subq	%rsp,%r11
+	andq	$-4096,%r11
+L$mul_page_walk:
+	movq	(%rsp,%r11,1),%r10
+	subq	$4096,%r11
+.byte	0x66,0x2e
+	jnc	L$mul_page_walk
+
 	movq	%rdx,%r12
 	movq	(%r8),%r8
 	movq	(%r12),%rbx
@@ -215,6 +230,9 @@ L$mul_epilogue:
 .p2align	4
 bn_mul4x_mont:
 L$mul4x_enter:
+	andl	$0x80100,%r11d
+	cmpl	$0x80100,%r11d
+	je	L$mulx4x_enter
 	pushq	%rbx
 	pushq	%rbp
 	pushq	%r12
@@ -231,6 +249,14 @@ L$mul4x_enter:
 
 	movq	%r11,8(%rsp,%r9,8)
 L$mul4x_body:
+	subq	%rsp,%r11
+	andq	$-4096,%r11
+L$mul4x_page_walk:
+	movq	(%rsp,%r11,1),%r10
+	subq	$4096,%r11
+.byte	0x2e
+	jnc	L$mul4x_page_walk
+
 	movq	%rdi,16(%rsp,%r9,8)
 	movq	%rdx,%r12
 	movq	(%r8),%r8
@@ -612,6 +638,7 @@ L$mul4x_epilogue:
 
 
 
+
 .p2align	5
 bn_sqr8x_mont:
 L$sqr8x_enter:
@@ -653,6 +680,15 @@ L$sqr8x_sp_alt:
 	subq	%r11,%rsp
 L$sqr8x_sp_done:
 	andq	$-64,%rsp
+	movq	%rax,%r11
+	subq	%rsp,%r11
+	andq	$-4096,%r11
+L$sqr8x_page_walk:
+	movq	(%rsp,%r11,1),%r10
+	subq	$4096,%r11
+.byte	0x2e
+	jnc	L$sqr8x_page_walk
+
 	movq	%r9,%r10
 	negq	%r9
 
@@ -664,6 +700,25 @@ L$sqr8x_body:
 	pxor	%xmm0,%xmm0
 .byte	102,72,15,110,207
 .byte	102,73,15,110,218
+	movl	_OPENSSL_ia32cap_P+8(%rip),%eax
+	andl	$0x80100,%eax
+	cmpl	$0x80100,%eax
+	jne	L$sqr8x_nox
+
+	call	_bn_sqrx8x_internal
+
+
+
+
+	leaq	(%r8,%rcx,1),%rbx
+	movq	%rcx,%r9
+	movq	%rcx,%rdx
+.byte	102,72,15,126,207
+	sarq	$3+2,%rcx
+	jmp	L$sqr8x_sub
+
+.p2align	5
+L$sqr8x_nox:
 	call	_bn_sqr8x_internal
 
 
@@ -742,5 +797,336 @@ L$sqr8x_cond_copy:
 L$sqr8x_epilogue:
 	.byte	0xf3,0xc3
 
+
+.p2align	5
+bn_mulx4x_mont:
+L$mulx4x_enter:
+	movq	%rsp,%rax
+	pushq	%rbx
+	pushq	%rbp
+	pushq	%r12
+	pushq	%r13
+	pushq	%r14
+	pushq	%r15
+
+	shll	$3,%r9d
+.byte	0x67
+	xorq	%r10,%r10
+	subq	%r9,%r10
+	movq	(%r8),%r8
+	leaq	-72(%rsp,%r10,1),%rsp
+	andq	$-128,%rsp
+	movq	%rax,%r11
+	subq	%rsp,%r11
+	andq	$-4096,%r11
+L$mulx4x_page_walk:
+	movq	(%rsp,%r11,1),%r10
+	subq	$4096,%r11
+.byte	0x66,0x2e
+	jnc	L$mulx4x_page_walk
+
+	leaq	(%rdx,%r9,1),%r10
+
+
+
+
+
+
+
+
+
+
+
+
+	movq	%r9,0(%rsp)
+	shrq	$5,%r9
+	movq	%r10,16(%rsp)
+	subq	$1,%r9
+	movq	%r8,24(%rsp)
+	movq	%rdi,32(%rsp)
+	movq	%rax,40(%rsp)
+	movq	%r9,48(%rsp)
+	jmp	L$mulx4x_body
+
+.p2align	5
+L$mulx4x_body:
+	leaq	8(%rdx),%rdi
+	movq	(%rdx),%rdx
+	leaq	64+32(%rsp),%rbx
+	movq	%rdx,%r9
+
+	mulxq	0(%rsi),%r8,%rax
+	mulxq	8(%rsi),%r11,%r14
+	addq	%rax,%r11
+	movq	%rdi,8(%rsp)
+	mulxq	16(%rsi),%r12,%r13
+	adcq	%r14,%r12
+	adcq	$0,%r13
+
+	movq	%r8,%rdi
+	imulq	24(%rsp),%r8
+	xorq	%rbp,%rbp
+
+	mulxq	24(%rsi),%rax,%r14
+	movq	%r8,%rdx
+	leaq	32(%rsi),%rsi
+	adcxq	%rax,%r13
+	adcxq	%rbp,%r14
+
+	mulxq	0(%rcx),%rax,%r10
+	adcxq	%rax,%rdi
+	adoxq	%r11,%r10
+	mulxq	8(%rcx),%rax,%r11
+	adcxq	%rax,%r10
+	adoxq	%r12,%r11
+.byte	0xc4,0x62,0xfb,0xf6,0xa1,0x10,0x00,0x00,0x00
+	movq	48(%rsp),%rdi
+	movq	%r10,-32(%rbx)
+	adcxq	%rax,%r11
+	adoxq	%r13,%r12
+	mulxq	24(%rcx),%rax,%r15
+	movq	%r9,%rdx
+	movq	%r11,-24(%rbx)
+	adcxq	%rax,%r12
+	adoxq	%rbp,%r15
+	leaq	32(%rcx),%rcx
+	movq	%r12,-16(%rbx)
+
+	jmp	L$mulx4x_1st
+
+.p2align	5
+L$mulx4x_1st:
+	adcxq	%rbp,%r15
+	mulxq	0(%rsi),%r10,%rax
+	adcxq	%r14,%r10
+	mulxq	8(%rsi),%r11,%r14
+	adcxq	%rax,%r11
+	mulxq	16(%rsi),%r12,%rax
+	adcxq	%r14,%r12
+	mulxq	24(%rsi),%r13,%r14
+.byte	0x67,0x67
+	movq	%r8,%rdx
+	adcxq	%rax,%r13
+	adcxq	%rbp,%r14
+	leaq	32(%rsi),%rsi
+	leaq	32(%rbx),%rbx
+
+	adoxq	%r15,%r10
+	mulxq	0(%rcx),%rax,%r15
+	adcxq	%rax,%r10
+	adoxq	%r15,%r11
+	mulxq	8(%rcx),%rax,%r15
+	adcxq	%rax,%r11
+	adoxq	%r15,%r12
+	mulxq	16(%rcx),%rax,%r15
+	movq	%r10,-40(%rbx)
+	adcxq	%rax,%r12
+	movq	%r11,-32(%rbx)
+	adoxq	%r15,%r13
+	mulxq	24(%rcx),%rax,%r15
+	movq	%r9,%rdx
+	movq	%r12,-24(%rbx)
+	adcxq	%rax,%r13
+	adoxq	%rbp,%r15
+	leaq	32(%rcx),%rcx
+	movq	%r13,-16(%rbx)
+
+	decq	%rdi
+	jnz	L$mulx4x_1st
+
+	movq	0(%rsp),%rax
+	movq	8(%rsp),%rdi
+	adcq	%rbp,%r15
+	addq	%r15,%r14
+	sbbq	%r15,%r15
+	movq	%r14,-8(%rbx)
+	jmp	L$mulx4x_outer
+
+.p2align	5
+L$mulx4x_outer:
+	movq	(%rdi),%rdx
+	leaq	8(%rdi),%rdi
+	subq	%rax,%rsi
+	movq	%r15,(%rbx)
+	leaq	64+32(%rsp),%rbx
+	subq	%rax,%rcx
+
+	mulxq	0(%rsi),%r8,%r11
+	xorl	%ebp,%ebp
+	movq	%rdx,%r9
+	mulxq	8(%rsi),%r14,%r12
+	adoxq	-32(%rbx),%r8
+	adcxq	%r14,%r11
+	mulxq	16(%rsi),%r15,%r13
+	adoxq	-24(%rbx),%r11
+	adcxq	%r15,%r12
+	adoxq	%rbp,%r12
+	adcxq	%rbp,%r13
+
+	movq	%rdi,8(%rsp)
+.byte	0x67
+	movq	%r8,%r15
+	imulq	24(%rsp),%r8
+	xorl	%ebp,%ebp
+
+	mulxq	24(%rsi),%rax,%r14
+	movq	%r8,%rdx
+	adoxq	-16(%rbx),%r12
+	adcxq	%rax,%r13
+	adoxq	-8(%rbx),%r13
+	adcxq	%rbp,%r14
+	leaq	32(%rsi),%rsi
+	adoxq	%rbp,%r14
+
+	mulxq	0(%rcx),%rax,%r10
+	adcxq	%rax,%r15
+	adoxq	%r11,%r10
+	mulxq	8(%rcx),%rax,%r11
+	adcxq	%rax,%r10
+	adoxq	%r12,%r11
+	mulxq	16(%rcx),%rax,%r12
+	movq	%r10,-32(%rbx)
+	adcxq	%rax,%r11
+	adoxq	%r13,%r12
+	mulxq	24(%rcx),%rax,%r15
+	movq	%r9,%rdx
+	movq	%r11,-24(%rbx)
+	leaq	32(%rcx),%rcx
+	adcxq	%rax,%r12
+	adoxq	%rbp,%r15
+	movq	48(%rsp),%rdi
+	movq	%r12,-16(%rbx)
+
+	jmp	L$mulx4x_inner
+
+.p2align	5
+L$mulx4x_inner:
+	mulxq	0(%rsi),%r10,%rax
+	adcxq	%rbp,%r15
+	adoxq	%r14,%r10
+	mulxq	8(%rsi),%r11,%r14
+	adcxq	0(%rbx),%r10
+	adoxq	%rax,%r11
+	mulxq	16(%rsi),%r12,%rax
+	adcxq	8(%rbx),%r11
+	adoxq	%r14,%r12
+	mulxq	24(%rsi),%r13,%r14
+	movq	%r8,%rdx
+	adcxq	16(%rbx),%r12
+	adoxq	%rax,%r13
+	adcxq	24(%rbx),%r13
+	adoxq	%rbp,%r14
+	leaq	32(%rsi),%rsi
+	leaq	32(%rbx),%rbx
+	adcxq	%rbp,%r14
+
+	adoxq	%r15,%r10
+	mulxq	0(%rcx),%rax,%r15
+	adcxq	%rax,%r10
+	adoxq	%r15,%r11
+	mulxq	8(%rcx),%rax,%r15
+	adcxq	%rax,%r11
+	adoxq	%r15,%r12
+	mulxq	16(%rcx),%rax,%r15
+	movq	%r10,-40(%rbx)
+	adcxq	%rax,%r12
+	adoxq	%r15,%r13
+	mulxq	24(%rcx),%rax,%r15
+	movq	%r9,%rdx
+	movq	%r11,-32(%rbx)
+	movq	%r12,-24(%rbx)
+	adcxq	%rax,%r13
+	adoxq	%rbp,%r15
+	leaq	32(%rcx),%rcx
+	movq	%r13,-16(%rbx)
+
+	decq	%rdi
+	jnz	L$mulx4x_inner
+
+	movq	0(%rsp),%rax
+	movq	8(%rsp),%rdi
+	adcq	%rbp,%r15
+	subq	0(%rbx),%rbp
+	adcq	%r15,%r14
+	sbbq	%r15,%r15
+	movq	%r14,-8(%rbx)
+
+	cmpq	16(%rsp),%rdi
+	jne	L$mulx4x_outer
+
+	leaq	64(%rsp),%rbx
+	subq	%rax,%rcx
+	negq	%r15
+	movq	%rax,%rdx
+	shrq	$3+2,%rax
+	movq	32(%rsp),%rdi
+	jmp	L$mulx4x_sub
+
+.p2align	5
+L$mulx4x_sub:
+	movq	0(%rbx),%r11
+	movq	8(%rbx),%r12
+	movq	16(%rbx),%r13
+	movq	24(%rbx),%r14
+	leaq	32(%rbx),%rbx
+	sbbq	0(%rcx),%r11
+	sbbq	8(%rcx),%r12
+	sbbq	16(%rcx),%r13
+	sbbq	24(%rcx),%r14
+	leaq	32(%rcx),%rcx
+	movq	%r11,0(%rdi)
+	movq	%r12,8(%rdi)
+	movq	%r13,16(%rdi)
+	movq	%r14,24(%rdi)
+	leaq	32(%rdi),%rdi
+	decq	%rax
+	jnz	L$mulx4x_sub
+
+	sbbq	$0,%r15
+	leaq	64(%rsp),%rbx
+	subq	%rdx,%rdi
+
+.byte	102,73,15,110,207
+	pxor	%xmm0,%xmm0
+	pshufd	$0,%xmm1,%xmm1
+	movq	40(%rsp),%rsi
+	jmp	L$mulx4x_cond_copy
+
+.p2align	5
+L$mulx4x_cond_copy:
+	movdqa	0(%rbx),%xmm2
+	movdqa	16(%rbx),%xmm3
+	leaq	32(%rbx),%rbx
+	movdqu	0(%rdi),%xmm4
+	movdqu	16(%rdi),%xmm5
+	leaq	32(%rdi),%rdi
+	movdqa	%xmm0,-32(%rbx)
+	movdqa	%xmm0,-16(%rbx)
+	pcmpeqd	%xmm1,%xmm0
+	pand	%xmm1,%xmm2
+	pand	%xmm1,%xmm3
+	pand	%xmm0,%xmm4
+	pand	%xmm0,%xmm5
+	pxor	%xmm0,%xmm0
+	por	%xmm2,%xmm4
+	por	%xmm3,%xmm5
+	movdqu	%xmm4,-32(%rdi)
+	movdqu	%xmm5,-16(%rdi)
+	subq	$32,%rdx
+	jnz	L$mulx4x_cond_copy
+
+	movq	%rdx,(%rbx)
+
+	movq	$1,%rax
+	movq	-48(%rsi),%r15
+	movq	-40(%rsi),%r14
+	movq	-32(%rsi),%r13
+	movq	-24(%rsi),%r12
+	movq	-16(%rsi),%rbp
+	movq	-8(%rsi),%rbx
+	leaq	(%rsi),%rsp
+L$mulx4x_epilogue:
+	.byte	0xf3,0xc3
+
 .byte	77,111,110,116,103,111,109,101,114,121,32,77,117,108,116,105,112,108,105,99,97,116,105,111,110,32,102,111,114,32,120,56,54,95,54,52,44,32,67,82,89,80,84,79,71,65,77,83,32,98,121,32,60,97,112,112,114,111,64,111,112,101,110,115,115,108,46,111,114,103,62,0
 .p2align	4

deps/openssl/asm/x64-macosx-gas/camellia/cmll-x86_64.s

@@ -1624,7 +1624,7 @@ L$cbc_prologue:
 	leaq	-64-63(%rcx),%r10
 	subq	%rsp,%r10
 	negq	%r10
-	andq	$960,%r10
+	andq	$0x3C0,%r10
 	subq	%r10,%rsp
 
 

deps/openssl/asm/x64-macosx-gas/md5/md5-x86_64.s

@@ -493,14 +493,14 @@ L$loop:
 	movl	%ecx,%r11d
 	addl	%ecx,%ebx
 	movl	0(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	xorl	%edx,%r11d
 	leal	-198630844(%rax,%r10,1),%eax
 	orl	%ebx,%r11d
 	xorl	%ecx,%r11d
 	addl	%r11d,%eax
 	movl	28(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$6,%eax
 	xorl	%ecx,%r11d
 	addl	%ebx,%eax
@@ -509,7 +509,7 @@ L$loop:
 	xorl	%ebx,%r11d
 	addl	%r11d,%edx
 	movl	56(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$10,%edx
 	xorl	%ebx,%r11d
 	addl	%eax,%edx
@@ -518,7 +518,7 @@ L$loop:
 	xorl	%eax,%r11d
 	addl	%r11d,%ecx
 	movl	20(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$15,%ecx
 	xorl	%eax,%r11d
 	addl	%edx,%ecx
@@ -527,7 +527,7 @@ L$loop:
 	xorl	%edx,%r11d
 	addl	%r11d,%ebx
 	movl	48(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$21,%ebx
 	xorl	%edx,%r11d
 	addl	%ecx,%ebx
@@ -536,7 +536,7 @@ L$loop:
 	xorl	%ecx,%r11d
 	addl	%r11d,%eax
 	movl	12(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$6,%eax
 	xorl	%ecx,%r11d
 	addl	%ebx,%eax
@@ -545,7 +545,7 @@ L$loop:
 	xorl	%ebx,%r11d
 	addl	%r11d,%edx
 	movl	40(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$10,%edx
 	xorl	%ebx,%r11d
 	addl	%eax,%edx
@@ -554,7 +554,7 @@ L$loop:
 	xorl	%eax,%r11d
 	addl	%r11d,%ecx
 	movl	4(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$15,%ecx
 	xorl	%eax,%r11d
 	addl	%edx,%ecx
@@ -563,7 +563,7 @@ L$loop:
 	xorl	%edx,%r11d
 	addl	%r11d,%ebx
 	movl	32(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$21,%ebx
 	xorl	%edx,%r11d
 	addl	%ecx,%ebx
@@ -572,7 +572,7 @@ L$loop:
 	xorl	%ecx,%r11d
 	addl	%r11d,%eax
 	movl	60(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$6,%eax
 	xorl	%ecx,%r11d
 	addl	%ebx,%eax
@@ -581,7 +581,7 @@ L$loop:
 	xorl	%ebx,%r11d
 	addl	%r11d,%edx
 	movl	24(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$10,%edx
 	xorl	%ebx,%r11d
 	addl	%eax,%edx
@@ -590,7 +590,7 @@ L$loop:
 	xorl	%eax,%r11d
 	addl	%r11d,%ecx
 	movl	52(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$15,%ecx
 	xorl	%eax,%r11d
 	addl	%edx,%ecx
@@ -599,7 +599,7 @@ L$loop:
 	xorl	%edx,%r11d
 	addl	%r11d,%ebx
 	movl	16(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$21,%ebx
 	xorl	%edx,%r11d
 	addl	%ecx,%ebx
@@ -608,7 +608,7 @@ L$loop:
 	xorl	%ecx,%r11d
 	addl	%r11d,%eax
 	movl	44(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$6,%eax
 	xorl	%ecx,%r11d
 	addl	%ebx,%eax
@@ -617,7 +617,7 @@ L$loop:
 	xorl	%ebx,%r11d
 	addl	%r11d,%edx
 	movl	8(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$10,%edx
 	xorl	%ebx,%r11d
 	addl	%eax,%edx
@@ -626,7 +626,7 @@ L$loop:
 	xorl	%eax,%r11d
 	addl	%r11d,%ecx
 	movl	36(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$15,%ecx
 	xorl	%eax,%r11d
 	addl	%edx,%ecx
@@ -635,7 +635,7 @@ L$loop:
 	xorl	%edx,%r11d
 	addl	%r11d,%ebx
 	movl	0(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$21,%ebx
 	xorl	%edx,%r11d
 	addl	%ecx,%ebx

deps/openssl/asm/x64-macosx-gas/modes/aesni-gcm-x86_64.s

@@ -1,14 +1,753 @@
 .text
 
-.globl	_aesni_gcm_encrypt
 
-_aesni_gcm_encrypt:
-	xorl	%eax,%eax
-	.byte	0xf3,0xc3
+.p2align	5
+_aesni_ctr32_ghash_6x:
+	vmovdqu	32(%r11),%xmm2
+	subq	$6,%rdx
+	vpxor	%xmm4,%xmm4,%xmm4
+	vmovdqu	0-128(%rcx),%xmm15
+	vpaddb	%xmm2,%xmm1,%xmm10
+	vpaddb	%xmm2,%xmm10,%xmm11
+	vpaddb	%xmm2,%xmm11,%xmm12
+	vpaddb	%xmm2,%xmm12,%xmm13
+	vpaddb	%xmm2,%xmm13,%xmm14
+	vpxor	%xmm15,%xmm1,%xmm9
+	vmovdqu	%xmm4,16+8(%rsp)
+	jmp	L$oop6x
+
+.p2align	5
+L$oop6x:
+	addl	$100663296,%ebx
+	jc	L$handle_ctr32
+	vmovdqu	0-32(%r9),%xmm3
+	vpaddb	%xmm2,%xmm14,%xmm1
+	vpxor	%xmm15,%xmm10,%xmm10
+	vpxor	%xmm15,%xmm11,%xmm11
+
+L$resume_ctr32:
+	vmovdqu	%xmm1,(%r8)
+	vpclmulqdq	$0x10,%xmm3,%xmm7,%xmm5
+	vpxor	%xmm15,%xmm12,%xmm12
+	vmovups	16-128(%rcx),%xmm2
+	vpclmulqdq	$0x01,%xmm3,%xmm7,%xmm6
+	xorq	%r12,%r12
+	cmpq	%r14,%r15
+
+	vaesenc	%xmm2,%xmm9,%xmm9
+	vmovdqu	48+8(%rsp),%xmm0
+	vpxor	%xmm15,%xmm13,%xmm13
+	vpclmulqdq	$0x00,%xmm3,%xmm7,%xmm1
+	vaesenc	%xmm2,%xmm10,%xmm10
+	vpxor	%xmm15,%xmm14,%xmm14
+	setnc	%r12b
+	vpclmulqdq	$0x11,%xmm3,%xmm7,%xmm7
+	vaesenc	%xmm2,%xmm11,%xmm11
+	vmovdqu	16-32(%r9),%xmm3
+	negq	%r12
+	vaesenc	%xmm2,%xmm12,%xmm12
+	vpxor	%xmm5,%xmm6,%xmm6
+	vpclmulqdq	$0x00,%xmm3,%xmm0,%xmm5
+	vpxor	%xmm4,%xmm8,%xmm8
+	vaesenc	%xmm2,%xmm13,%xmm13
+	vpxor	%xmm5,%xmm1,%xmm4
+	andq	$0x60,%r12
+	vmovups	32-128(%rcx),%xmm15
+	vpclmulqdq	$0x10,%xmm3,%xmm0,%xmm1
+	vaesenc	%xmm2,%xmm14,%xmm14
+
+	vpclmulqdq	$0x01,%xmm3,%xmm0,%xmm2
+	leaq	(%r14,%r12,1),%r14
+	vaesenc	%xmm15,%xmm9,%xmm9
+	vpxor	16+8(%rsp),%xmm8,%xmm8
+	vpclmulqdq	$0x11,%xmm3,%xmm0,%xmm3
+	vmovdqu	64+8(%rsp),%xmm0
+	vaesenc	%xmm15,%xmm10,%xmm10
+	movbeq	88(%r14),%r13
+	vaesenc	%xmm15,%xmm11,%xmm11
+	movbeq	80(%r14),%r12
+	vaesenc	%xmm15,%xmm12,%xmm12
+	movq	%r13,32+8(%rsp)
+	vaesenc	%xmm15,%xmm13,%xmm13
+	movq	%r12,40+8(%rsp)
+	vmovdqu	48-32(%r9),%xmm5
+	vaesenc	%xmm15,%xmm14,%xmm14
+
+	vmovups	48-128(%rcx),%xmm15
+	vpxor	%xmm1,%xmm6,%xmm6
+	vpclmulqdq	$0x00,%xmm5,%xmm0,%xmm1
+	vaesenc	%xmm15,%xmm9,%xmm9
+	vpxor	%xmm2,%xmm6,%xmm6
+	vpclmulqdq	$0x10,%xmm5,%xmm0,%xmm2
+	vaesenc	%xmm15,%xmm10,%xmm10
+	vpxor	%xmm3,%xmm7,%xmm7
+	vpclmulqdq	$0x01,%xmm5,%xmm0,%xmm3
+	vaesenc	%xmm15,%xmm11,%xmm11
+	vpclmulqdq	$0x11,%xmm5,%xmm0,%xmm5
+	vmovdqu	80+8(%rsp),%xmm0
+	vaesenc	%xmm15,%xmm12,%xmm12
+	vaesenc	%xmm15,%xmm13,%xmm13
+	vpxor	%xmm1,%xmm4,%xmm4
+	vmovdqu	64-32(%r9),%xmm1
+	vaesenc	%xmm15,%xmm14,%xmm14
+
+	vmovups	64-128(%rcx),%xmm15
+	vpxor	%xmm2,%xmm6,%xmm6
+	vpclmulqdq	$0x00,%xmm1,%xmm0,%xmm2
+	vaesenc	%xmm15,%xmm9,%xmm9
+	vpxor	%xmm3,%xmm6,%xmm6
+	vpclmulqdq	$0x10,%xmm1,%xmm0,%xmm3
+	vaesenc	%xmm15,%xmm10,%xmm10
+	movbeq	72(%r14),%r13
+	vpxor	%xmm5,%xmm7,%xmm7
+	vpclmulqdq	$0x01,%xmm1,%xmm0,%xmm5
+	vaesenc	%xmm15,%xmm11,%xmm11
+	movbeq	64(%r14),%r12
+	vpclmulqdq	$0x11,%xmm1,%xmm0,%xmm1
+	vmovdqu	96+8(%rsp),%xmm0
+	vaesenc	%xmm15,%xmm12,%xmm12
+	movq	%r13,48+8(%rsp)
+	vaesenc	%xmm15,%xmm13,%xmm13
+	movq	%r12,56+8(%rsp)
+	vpxor	%xmm2,%xmm4,%xmm4
+	vmovdqu	96-32(%r9),%xmm2
+	vaesenc	%xmm15,%xmm14,%xmm14
+
+	vmovups	80-128(%rcx),%xmm15
+	vpxor	%xmm3,%xmm6,%xmm6
+	vpclmulqdq	$0x00,%xmm2,%xmm0,%xmm3
+	vaesenc	%xmm15,%xmm9,%xmm9
+	vpxor	%xmm5,%xmm6,%xmm6
+	vpclmulqdq	$0x10,%xmm2,%xmm0,%xmm5
+	vaesenc	%xmm15,%xmm10,%xmm10
+	movbeq	56(%r14),%r13
+	vpxor	%xmm1,%xmm7,%xmm7
+	vpclmulqdq	$0x01,%xmm2,%xmm0,%xmm1
+	vpxor	112+8(%rsp),%xmm8,%xmm8
+	vaesenc	%xmm15,%xmm11,%xmm11
+	movbeq	48(%r14),%r12
+	vpclmulqdq	$0x11,%xmm2,%xmm0,%xmm2
+	vaesenc	%xmm15,%xmm12,%xmm12
+	movq	%r13,64+8(%rsp)
+	vaesenc	%xmm15,%xmm13,%xmm13
+	movq	%r12,72+8(%rsp)
+	vpxor	%xmm3,%xmm4,%xmm4
+	vmovdqu	112-32(%r9),%xmm3
+	vaesenc	%xmm15,%xmm14,%xmm14
+
+	vmovups	96-128(%rcx),%xmm15
+	vpxor	%xmm5,%xmm6,%xmm6
+	vpclmulqdq	$0x10,%xmm3,%xmm8,%xmm5
+	vaesenc	%xmm15,%xmm9,%xmm9
+	vpxor	%xmm1,%xmm6,%xmm6
+	vpclmulqdq	$0x01,%xmm3,%xmm8,%xmm1
+	vaesenc	%xmm15,%xmm10,%xmm10
+	movbeq	40(%r14),%r13
+	vpxor	%xmm2,%xmm7,%xmm7
+	vpclmulqdq	$0x00,%xmm3,%xmm8,%xmm2
+	vaesenc	%xmm15,%xmm11,%xmm11
+	movbeq	32(%r14),%r12
+	vpclmulqdq	$0x11,%xmm3,%xmm8,%xmm8
+	vaesenc	%xmm15,%xmm12,%xmm12
+	movq	%r13,80+8(%rsp)
+	vaesenc	%xmm15,%xmm13,%xmm13
+	movq	%r12,88+8(%rsp)
+	vpxor	%xmm5,%xmm6,%xmm6
+	vaesenc	%xmm15,%xmm14,%xmm14
+	vpxor	%xmm1,%xmm6,%xmm6
+
+	vmovups	112-128(%rcx),%xmm15
+	vpslldq	$8,%xmm6,%xmm5
+	vpxor	%xmm2,%xmm4,%xmm4
+	vmovdqu	16(%r11),%xmm3
+
+	vaesenc	%xmm15,%xmm9,%xmm9
+	vpxor	%xmm8,%xmm7,%xmm7
+	vaesenc	%xmm15,%xmm10,%xmm10
+	vpxor	%xmm5,%xmm4,%xmm4
+	movbeq	24(%r14),%r13
+	vaesenc	%xmm15,%xmm11,%xmm11
+	movbeq	16(%r14),%r12
+	vpalignr	$8,%xmm4,%xmm4,%xmm0
+	vpclmulqdq	$0x10,%xmm3,%xmm4,%xmm4
+	movq	%r13,96+8(%rsp)
+	vaesenc	%xmm15,%xmm12,%xmm12
+	movq	%r12,104+8(%rsp)
+	vaesenc	%xmm15,%xmm13,%xmm13
+	vmovups	128-128(%rcx),%xmm1
+	vaesenc	%xmm15,%xmm14,%xmm14
+
+	vaesenc	%xmm1,%xmm9,%xmm9
+	vmovups	144-128(%rcx),%xmm15
+	vaesenc	%xmm1,%xmm10,%xmm10
+	vpsrldq	$8,%xmm6,%xmm6
+	vaesenc	%xmm1,%xmm11,%xmm11
+	vpxor	%xmm6,%xmm7,%xmm7
+	vaesenc	%xmm1,%xmm12,%xmm12
+	vpxor	%xmm0,%xmm4,%xmm4
+	movbeq	8(%r14),%r13
+	vaesenc	%xmm1,%xmm13,%xmm13
+	movbeq	0(%r14),%r12
+	vaesenc	%xmm1,%xmm14,%xmm14
+	vmovups	160-128(%rcx),%xmm1
+	cmpl	$11,%ebp
+	jb	L$enc_tail
+
+	vaesenc	%xmm15,%xmm9,%xmm9
+	vaesenc	%xmm15,%xmm10,%xmm10
+	vaesenc	%xmm15,%xmm11,%xmm11
+	vaesenc	%xmm15,%xmm12,%xmm12
+	vaesenc	%xmm15,%xmm13,%xmm13
+	vaesenc	%xmm15,%xmm14,%xmm14
+
+	vaesenc	%xmm1,%xmm9,%xmm9
+	vaesenc	%xmm1,%xmm10,%xmm10
+	vaesenc	%xmm1,%xmm11,%xmm11
+	vaesenc	%xmm1,%xmm12,%xmm12
+	vaesenc	%xmm1,%xmm13,%xmm13
+	vmovups	176-128(%rcx),%xmm15
+	vaesenc	%xmm1,%xmm14,%xmm14
+	vmovups	192-128(%rcx),%xmm1
+	je	L$enc_tail
+
+	vaesenc	%xmm15,%xmm9,%xmm9
+	vaesenc	%xmm15,%xmm10,%xmm10
+	vaesenc	%xmm15,%xmm11,%xmm11
+	vaesenc	%xmm15,%xmm12,%xmm12
+	vaesenc	%xmm15,%xmm13,%xmm13
+	vaesenc	%xmm15,%xmm14,%xmm14
+
+	vaesenc	%xmm1,%xmm9,%xmm9
+	vaesenc	%xmm1,%xmm10,%xmm10
+	vaesenc	%xmm1,%xmm11,%xmm11
+	vaesenc	%xmm1,%xmm12,%xmm12
+	vaesenc	%xmm1,%xmm13,%xmm13
+	vmovups	208-128(%rcx),%xmm15
+	vaesenc	%xmm1,%xmm14,%xmm14
+	vmovups	224-128(%rcx),%xmm1
+	jmp	L$enc_tail
 
+.p2align	5
+L$handle_ctr32:
+	vmovdqu	(%r11),%xmm0
+	vpshufb	%xmm0,%xmm1,%xmm6
+	vmovdqu	48(%r11),%xmm5
+	vpaddd	64(%r11),%xmm6,%xmm10
+	vpaddd	%xmm5,%xmm6,%xmm11
+	vmovdqu	0-32(%r9),%xmm3
+	vpaddd	%xmm5,%xmm10,%xmm12
+	vpshufb	%xmm0,%xmm10,%xmm10
+	vpaddd	%xmm5,%xmm11,%xmm13
+	vpshufb	%xmm0,%xmm11,%xmm11
+	vpxor	%xmm15,%xmm10,%xmm10
+	vpaddd	%xmm5,%xmm12,%xmm14
+	vpshufb	%xmm0,%xmm12,%xmm12
+	vpxor	%xmm15,%xmm11,%xmm11
+	vpaddd	%xmm5,%xmm13,%xmm1
+	vpshufb	%xmm0,%xmm13,%xmm13
+	vpshufb	%xmm0,%xmm14,%xmm14
+	vpshufb	%xmm0,%xmm1,%xmm1
+	jmp	L$resume_ctr32
+
+.p2align	5
+L$enc_tail:
+	vaesenc	%xmm15,%xmm9,%xmm9
+	vmovdqu	%xmm7,16+8(%rsp)
+	vpalignr	$8,%xmm4,%xmm4,%xmm8
+	vaesenc	%xmm15,%xmm10,%xmm10
+	vpclmulqdq	$0x10,%xmm3,%xmm4,%xmm4
+	vpxor	0(%rdi),%xmm1,%xmm2
+	vaesenc	%xmm15,%xmm11,%xmm11
+	vpxor	16(%rdi),%xmm1,%xmm0
+	vaesenc	%xmm15,%xmm12,%xmm12
+	vpxor	32(%rdi),%xmm1,%xmm5
+	vaesenc	%xmm15,%xmm13,%xmm13
+	vpxor	48(%rdi),%xmm1,%xmm6
+	vaesenc	%xmm15,%xmm14,%xmm14
+	vpxor	64(%rdi),%xmm1,%xmm7
+	vpxor	80(%rdi),%xmm1,%xmm3
+	vmovdqu	(%r8),%xmm1
+
+	vaesenclast	%xmm2,%xmm9,%xmm9
+	vmovdqu	32(%r11),%xmm2
+	vaesenclast	%xmm0,%xmm10,%xmm10
+	vpaddb	%xmm2,%xmm1,%xmm0
+	movq	%r13,112+8(%rsp)
+	leaq	96(%rdi),%rdi
+	vaesenclast	%xmm5,%xmm11,%xmm11
+	vpaddb	%xmm2,%xmm0,%xmm5
+	movq	%r12,120+8(%rsp)
+	leaq	96(%rsi),%rsi
+	vmovdqu	0-128(%rcx),%xmm15
+	vaesenclast	%xmm6,%xmm12,%xmm12
+	vpaddb	%xmm2,%xmm5,%xmm6
+	vaesenclast	%xmm7,%xmm13,%xmm13
+	vpaddb	%xmm2,%xmm6,%xmm7
+	vaesenclast	%xmm3,%xmm14,%xmm14
+	vpaddb	%xmm2,%xmm7,%xmm3
+
+	addq	$0x60,%r10
+	subq	$0x6,%rdx
+	jc	L$6x_done
+
+	vmovups	%xmm9,-96(%rsi)
+	vpxor	%xmm15,%xmm1,%xmm9
+	vmovups	%xmm10,-80(%rsi)
+	vmovdqa	%xmm0,%xmm10
+	vmovups	%xmm11,-64(%rsi)
+	vmovdqa	%xmm5,%xmm11
+	vmovups	%xmm12,-48(%rsi)
+	vmovdqa	%xmm6,%xmm12
+	vmovups	%xmm13,-32(%rsi)
+	vmovdqa	%xmm7,%xmm13
+	vmovups	%xmm14,-16(%rsi)
+	vmovdqa	%xmm3,%xmm14
+	vmovdqu	32+8(%rsp),%xmm7
+	jmp	L$oop6x
+
+L$6x_done:
+	vpxor	16+8(%rsp),%xmm8,%xmm8
+	vpxor	%xmm4,%xmm8,%xmm8
+
+	.byte	0xf3,0xc3
 
 .globl	_aesni_gcm_decrypt
 
+.p2align	5
 _aesni_gcm_decrypt:
-	xorl	%eax,%eax
+	xorq	%r10,%r10
+	cmpq	$0x60,%rdx
+	jb	L$gcm_dec_abort
+
+	leaq	(%rsp),%rax
+	pushq	%rbx
+	pushq	%rbp
+	pushq	%r12
+	pushq	%r13
+	pushq	%r14
+	pushq	%r15
+	vzeroupper
+
+	vmovdqu	(%r8),%xmm1
+	addq	$-128,%rsp
+	movl	12(%r8),%ebx
+	leaq	L$bswap_mask(%rip),%r11
+	leaq	-128(%rcx),%r14
+	movq	$0xf80,%r15
+	vmovdqu	(%r9),%xmm8
+	andq	$-128,%rsp
+	vmovdqu	(%r11),%xmm0
+	leaq	128(%rcx),%rcx
+	leaq	32+32(%r9),%r9
+	movl	240-128(%rcx),%ebp
+	vpshufb	%xmm0,%xmm8,%xmm8
+
+	andq	%r15,%r14
+	andq	%rsp,%r15
+	subq	%r14,%r15
+	jc	L$dec_no_key_aliasing
+	cmpq	$768,%r15
+	jnc	L$dec_no_key_aliasing
+	subq	%r15,%rsp
+L$dec_no_key_aliasing:
+
+	vmovdqu	80(%rdi),%xmm7
+	leaq	(%rdi),%r14
+	vmovdqu	64(%rdi),%xmm4
+	leaq	-192(%rdi,%rdx,1),%r15
+	vmovdqu	48(%rdi),%xmm5
+	shrq	$4,%rdx
+	xorq	%r10,%r10
+	vmovdqu	32(%rdi),%xmm6
+	vpshufb	%xmm0,%xmm7,%xmm7
+	vmovdqu	16(%rdi),%xmm2
+	vpshufb	%xmm0,%xmm4,%xmm4
+	vmovdqu	(%rdi),%xmm3
+	vpshufb	%xmm0,%xmm5,%xmm5
+	vmovdqu	%xmm4,48(%rsp)
+	vpshufb	%xmm0,%xmm6,%xmm6
+	vmovdqu	%xmm5,64(%rsp)
+	vpshufb	%xmm0,%xmm2,%xmm2
+	vmovdqu	%xmm6,80(%rsp)
+	vpshufb	%xmm0,%xmm3,%xmm3
+	vmovdqu	%xmm2,96(%rsp)
+	vmovdqu	%xmm3,112(%rsp)
+
+	call	_aesni_ctr32_ghash_6x
+
+	vmovups	%xmm9,-96(%rsi)
+	vmovups	%xmm10,-80(%rsi)
+	vmovups	%xmm11,-64(%rsi)
+	vmovups	%xmm12,-48(%rsi)
+	vmovups	%xmm13,-32(%rsi)
+	vmovups	%xmm14,-16(%rsi)
+
+	vpshufb	(%r11),%xmm8,%xmm8
+	vmovdqu	%xmm8,-64(%r9)
+
+	vzeroupper
+	movq	-48(%rax),%r15
+	movq	-40(%rax),%r14
+	movq	-32(%rax),%r13
+	movq	-24(%rax),%r12
+	movq	-16(%rax),%rbp
+	movq	-8(%rax),%rbx
+	leaq	(%rax),%rsp
+L$gcm_dec_abort:
+	movq	%r10,%rax
+	.byte	0xf3,0xc3
+
+
+.p2align	5
+_aesni_ctr32_6x:
+	vmovdqu	0-128(%rcx),%xmm4
+	vmovdqu	32(%r11),%xmm2
+	leaq	-1(%rbp),%r13
+	vmovups	16-128(%rcx),%xmm15
+	leaq	32-128(%rcx),%r12
+	vpxor	%xmm4,%xmm1,%xmm9
+	addl	$100663296,%ebx
+	jc	L$handle_ctr32_2
+	vpaddb	%xmm2,%xmm1,%xmm10
+	vpaddb	%xmm2,%xmm10,%xmm11
+	vpxor	%xmm4,%xmm10,%xmm10
+	vpaddb	%xmm2,%xmm11,%xmm12
+	vpxor	%xmm4,%xmm11,%xmm11
+	vpaddb	%xmm2,%xmm12,%xmm13
+	vpxor	%xmm4,%xmm12,%xmm12
+	vpaddb	%xmm2,%xmm13,%xmm14
+	vpxor	%xmm4,%xmm13,%xmm13
+	vpaddb	%xmm2,%xmm14,%xmm1
+	vpxor	%xmm4,%xmm14,%xmm14
+	jmp	L$oop_ctr32
+
+.p2align	4
+L$oop_ctr32:
+	vaesenc	%xmm15,%xmm9,%xmm9
+	vaesenc	%xmm15,%xmm10,%xmm10
+	vaesenc	%xmm15,%xmm11,%xmm11
+	vaesenc	%xmm15,%xmm12,%xmm12
+	vaesenc	%xmm15,%xmm13,%xmm13
+	vaesenc	%xmm15,%xmm14,%xmm14
+	vmovups	(%r12),%xmm15
+	leaq	16(%r12),%r12
+	decl	%r13d
+	jnz	L$oop_ctr32
+
+	vmovdqu	(%r12),%xmm3
+	vaesenc	%xmm15,%xmm9,%xmm9
+	vpxor	0(%rdi),%xmm3,%xmm4
+	vaesenc	%xmm15,%xmm10,%xmm10
+	vpxor	16(%rdi),%xmm3,%xmm5
+	vaesenc	%xmm15,%xmm11,%xmm11
+	vpxor	32(%rdi),%xmm3,%xmm6
+	vaesenc	%xmm15,%xmm12,%xmm12
+	vpxor	48(%rdi),%xmm3,%xmm8
+	vaesenc	%xmm15,%xmm13,%xmm13
+	vpxor	64(%rdi),%xmm3,%xmm2
+	vaesenc	%xmm15,%xmm14,%xmm14
+	vpxor	80(%rdi),%xmm3,%xmm3
+	leaq	96(%rdi),%rdi
+
+	vaesenclast	%xmm4,%xmm9,%xmm9
+	vaesenclast	%xmm5,%xmm10,%xmm10
+	vaesenclast	%xmm6,%xmm11,%xmm11
+	vaesenclast	%xmm8,%xmm12,%xmm12
+	vaesenclast	%xmm2,%xmm13,%xmm13
+	vaesenclast	%xmm3,%xmm14,%xmm14
+	vmovups	%xmm9,0(%rsi)
+	vmovups	%xmm10,16(%rsi)
+	vmovups	%xmm11,32(%rsi)
+	vmovups	%xmm12,48(%rsi)
+	vmovups	%xmm13,64(%rsi)
+	vmovups	%xmm14,80(%rsi)
+	leaq	96(%rsi),%rsi
+
+	.byte	0xf3,0xc3
+.p2align	5
+L$handle_ctr32_2:
+	vpshufb	%xmm0,%xmm1,%xmm6
+	vmovdqu	48(%r11),%xmm5
+	vpaddd	64(%r11),%xmm6,%xmm10
+	vpaddd	%xmm5,%xmm6,%xmm11
+	vpaddd	%xmm5,%xmm10,%xmm12
+	vpshufb	%xmm0,%xmm10,%xmm10
+	vpaddd	%xmm5,%xmm11,%xmm13
+	vpshufb	%xmm0,%xmm11,%xmm11
+	vpxor	%xmm4,%xmm10,%xmm10
+	vpaddd	%xmm5,%xmm12,%xmm14
+	vpshufb	%xmm0,%xmm12,%xmm12
+	vpxor	%xmm4,%xmm11,%xmm11
+	vpaddd	%xmm5,%xmm13,%xmm1
+	vpshufb	%xmm0,%xmm13,%xmm13
+	vpxor	%xmm4,%xmm12,%xmm12
+	vpshufb	%xmm0,%xmm14,%xmm14
+	vpxor	%xmm4,%xmm13,%xmm13
+	vpshufb	%xmm0,%xmm1,%xmm1
+	vpxor	%xmm4,%xmm14,%xmm14
+	jmp	L$oop_ctr32
+
+
+.globl	_aesni_gcm_encrypt
+
+.p2align	5
+_aesni_gcm_encrypt:
+	xorq	%r10,%r10
+	cmpq	$288,%rdx
+	jb	L$gcm_enc_abort
+
+	leaq	(%rsp),%rax
+	pushq	%rbx
+	pushq	%rbp
+	pushq	%r12
+	pushq	%r13
+	pushq	%r14
+	pushq	%r15
+	vzeroupper
+
+	vmovdqu	(%r8),%xmm1
+	addq	$-128,%rsp
+	movl	12(%r8),%ebx
+	leaq	L$bswap_mask(%rip),%r11
+	leaq	-128(%rcx),%r14
+	movq	$0xf80,%r15
+	leaq	128(%rcx),%rcx
+	vmovdqu	(%r11),%xmm0
+	andq	$-128,%rsp
+	movl	240-128(%rcx),%ebp
+
+	andq	%r15,%r14
+	andq	%rsp,%r15
+	subq	%r14,%r15
+	jc	L$enc_no_key_aliasing
+	cmpq	$768,%r15
+	jnc	L$enc_no_key_aliasing
+	subq	%r15,%rsp
+L$enc_no_key_aliasing:
+
+	leaq	(%rsi),%r14
+	leaq	-192(%rsi,%rdx,1),%r15
+	shrq	$4,%rdx
+
+	call	_aesni_ctr32_6x
+	vpshufb	%xmm0,%xmm9,%xmm8
+	vpshufb	%xmm0,%xmm10,%xmm2
+	vmovdqu	%xmm8,112(%rsp)
+	vpshufb	%xmm0,%xmm11,%xmm4
+	vmovdqu	%xmm2,96(%rsp)
+	vpshufb	%xmm0,%xmm12,%xmm5
+	vmovdqu	%xmm4,80(%rsp)
+	vpshufb	%xmm0,%xmm13,%xmm6
+	vmovdqu	%xmm5,64(%rsp)
+	vpshufb	%xmm0,%xmm14,%xmm7
+	vmovdqu	%xmm6,48(%rsp)
+
+	call	_aesni_ctr32_6x
+
+	vmovdqu	(%r9),%xmm8
+	leaq	32+32(%r9),%r9
+	subq	$12,%rdx
+	movq	$192,%r10
+	vpshufb	%xmm0,%xmm8,%xmm8
+
+	call	_aesni_ctr32_ghash_6x
+	vmovdqu	32(%rsp),%xmm7
+	vmovdqu	(%r11),%xmm0
+	vmovdqu	0-32(%r9),%xmm3
+	vpunpckhqdq	%xmm7,%xmm7,%xmm1
+	vmovdqu	32-32(%r9),%xmm15
+	vmovups	%xmm9,-96(%rsi)
+	vpshufb	%xmm0,%xmm9,%xmm9
+	vpxor	%xmm7,%xmm1,%xmm1
+	vmovups	%xmm10,-80(%rsi)
+	vpshufb	%xmm0,%xmm10,%xmm10
+	vmovups	%xmm11,-64(%rsi)
+	vpshufb	%xmm0,%xmm11,%xmm11
+	vmovups	%xmm12,-48(%rsi)
+	vpshufb	%xmm0,%xmm12,%xmm12
+	vmovups	%xmm13,-32(%rsi)
+	vpshufb	%xmm0,%xmm13,%xmm13
+	vmovups	%xmm14,-16(%rsi)
+	vpshufb	%xmm0,%xmm14,%xmm14
+	vmovdqu	%xmm9,16(%rsp)
+	vmovdqu	48(%rsp),%xmm6
+	vmovdqu	16-32(%r9),%xmm0
+	vpunpckhqdq	%xmm6,%xmm6,%xmm2
+	vpclmulqdq	$0x00,%xmm3,%xmm7,%xmm5
+	vpxor	%xmm6,%xmm2,%xmm2
+	vpclmulqdq	$0x11,%xmm3,%xmm7,%xmm7
+	vpclmulqdq	$0x00,%xmm15,%xmm1,%xmm1
+
+	vmovdqu	64(%rsp),%xmm9
+	vpclmulqdq	$0x00,%xmm0,%xmm6,%xmm4
+	vmovdqu	48-32(%r9),%xmm3
+	vpxor	%xmm5,%xmm4,%xmm4
+	vpunpckhqdq	%xmm9,%xmm9,%xmm5
+	vpclmulqdq	$0x11,%xmm0,%xmm6,%xmm6
+	vpxor	%xmm9,%xmm5,%xmm5
+	vpxor	%xmm7,%xmm6,%xmm6
+	vpclmulqdq	$0x10,%xmm15,%xmm2,%xmm2
+	vmovdqu	80-32(%r9),%xmm15
+	vpxor	%xmm1,%xmm2,%xmm2
+
+	vmovdqu	80(%rsp),%xmm1
+	vpclmulqdq	$0x00,%xmm3,%xmm9,%xmm7
+	vmovdqu	64-32(%r9),%xmm0
+	vpxor	%xmm4,%xmm7,%xmm7
+	vpunpckhqdq	%xmm1,%xmm1,%xmm4
+	vpclmulqdq	$0x11,%xmm3,%xmm9,%xmm9
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpxor	%xmm6,%xmm9,%xmm9
+	vpclmulqdq	$0x00,%xmm15,%xmm5,%xmm5
+	vpxor	%xmm2,%xmm5,%xmm5
+
+	vmovdqu	96(%rsp),%xmm2
+	vpclmulqdq	$0x00,%xmm0,%xmm1,%xmm6
+	vmovdqu	96-32(%r9),%xmm3
+	vpxor	%xmm7,%xmm6,%xmm6
+	vpunpckhqdq	%xmm2,%xmm2,%xmm7
+	vpclmulqdq	$0x11,%xmm0,%xmm1,%xmm1
+	vpxor	%xmm2,%xmm7,%xmm7
+	vpxor	%xmm9,%xmm1,%xmm1
+	vpclmulqdq	$0x10,%xmm15,%xmm4,%xmm4
+	vmovdqu	128-32(%r9),%xmm15
+	vpxor	%xmm5,%xmm4,%xmm4
+
+	vpxor	112(%rsp),%xmm8,%xmm8
+	vpclmulqdq	$0x00,%xmm3,%xmm2,%xmm5
+	vmovdqu	112-32(%r9),%xmm0
+	vpunpckhqdq	%xmm8,%xmm8,%xmm9
+	vpxor	%xmm6,%xmm5,%xmm5
+	vpclmulqdq	$0x11,%xmm3,%xmm2,%xmm2
+	vpxor	%xmm8,%xmm9,%xmm9
+	vpxor	%xmm1,%xmm2,%xmm2
+	vpclmulqdq	$0x00,%xmm15,%xmm7,%xmm7
+	vpxor	%xmm4,%xmm7,%xmm4
+
+	vpclmulqdq	$0x00,%xmm0,%xmm8,%xmm6
+	vmovdqu	0-32(%r9),%xmm3
+	vpunpckhqdq	%xmm14,%xmm14,%xmm1
+	vpclmulqdq	$0x11,%xmm0,%xmm8,%xmm8
+	vpxor	%xmm14,%xmm1,%xmm1
+	vpxor	%xmm5,%xmm6,%xmm5
+	vpclmulqdq	$0x10,%xmm15,%xmm9,%xmm9
+	vmovdqu	32-32(%r9),%xmm15
+	vpxor	%xmm2,%xmm8,%xmm7
+	vpxor	%xmm4,%xmm9,%xmm6
+
+	vmovdqu	16-32(%r9),%xmm0
+	vpxor	%xmm5,%xmm7,%xmm9
+	vpclmulqdq	$0x00,%xmm3,%xmm14,%xmm4
+	vpxor	%xmm9,%xmm6,%xmm6
+	vpunpckhqdq	%xmm13,%xmm13,%xmm2
+	vpclmulqdq	$0x11,%xmm3,%xmm14,%xmm14
+	vpxor	%xmm13,%xmm2,%xmm2
+	vpslldq	$8,%xmm6,%xmm9
+	vpclmulqdq	$0x00,%xmm15,%xmm1,%xmm1
+	vpxor	%xmm9,%xmm5,%xmm8
+	vpsrldq	$8,%xmm6,%xmm6
+	vpxor	%xmm6,%xmm7,%xmm7
+
+	vpclmulqdq	$0x00,%xmm0,%xmm13,%xmm5
+	vmovdqu	48-32(%r9),%xmm3
+	vpxor	%xmm4,%xmm5,%xmm5
+	vpunpckhqdq	%xmm12,%xmm12,%xmm9
+	vpclmulqdq	$0x11,%xmm0,%xmm13,%xmm13
+	vpxor	%xmm12,%xmm9,%xmm9
+	vpxor	%xmm14,%xmm13,%xmm13
+	vpalignr	$8,%xmm8,%xmm8,%xmm14
+	vpclmulqdq	$0x10,%xmm15,%xmm2,%xmm2
+	vmovdqu	80-32(%r9),%xmm15
+	vpxor	%xmm1,%xmm2,%xmm2
+
+	vpclmulqdq	$0x00,%xmm3,%xmm12,%xmm4
+	vmovdqu	64-32(%r9),%xmm0
+	vpxor	%xmm5,%xmm4,%xmm4
+	vpunpckhqdq	%xmm11,%xmm11,%xmm1
+	vpclmulqdq	$0x11,%xmm3,%xmm12,%xmm12
+	vpxor	%xmm11,%xmm1,%xmm1
+	vpxor	%xmm13,%xmm12,%xmm12
+	vxorps	16(%rsp),%xmm7,%xmm7
+	vpclmulqdq	$0x00,%xmm15,%xmm9,%xmm9
+	vpxor	%xmm2,%xmm9,%xmm9
+
+	vpclmulqdq	$0x10,16(%r11),%xmm8,%xmm8
+	vxorps	%xmm14,%xmm8,%xmm8
+
+	vpclmulqdq	$0x00,%xmm0,%xmm11,%xmm5
+	vmovdqu	96-32(%r9),%xmm3
+	vpxor	%xmm4,%xmm5,%xmm5
+	vpunpckhqdq	%xmm10,%xmm10,%xmm2
+	vpclmulqdq	$0x11,%xmm0,%xmm11,%xmm11
+	vpxor	%xmm10,%xmm2,%xmm2
+	vpalignr	$8,%xmm8,%xmm8,%xmm14
+	vpxor	%xmm12,%xmm11,%xmm11
+	vpclmulqdq	$0x10,%xmm15,%xmm1,%xmm1
+	vmovdqu	128-32(%r9),%xmm15
+	vpxor	%xmm9,%xmm1,%xmm1
+
+	vxorps	%xmm7,%xmm14,%xmm14
+	vpclmulqdq	$0x10,16(%r11),%xmm8,%xmm8
+	vxorps	%xmm14,%xmm8,%xmm8
+
+	vpclmulqdq	$0x00,%xmm3,%xmm10,%xmm4
+	vmovdqu	112-32(%r9),%xmm0
+	vpxor	%xmm5,%xmm4,%xmm4
+	vpunpckhqdq	%xmm8,%xmm8,%xmm9
+	vpclmulqdq	$0x11,%xmm3,%xmm10,%xmm10
+	vpxor	%xmm8,%xmm9,%xmm9
+	vpxor	%xmm11,%xmm10,%xmm10
+	vpclmulqdq	$0x00,%xmm15,%xmm2,%xmm2
+	vpxor	%xmm1,%xmm2,%xmm2
+
+	vpclmulqdq	$0x00,%xmm0,%xmm8,%xmm5
+	vpclmulqdq	$0x11,%xmm0,%xmm8,%xmm7
+	vpxor	%xmm4,%xmm5,%xmm5
+	vpclmulqdq	$0x10,%xmm15,%xmm9,%xmm6
+	vpxor	%xmm10,%xmm7,%xmm7
+	vpxor	%xmm2,%xmm6,%xmm6
+
+	vpxor	%xmm5,%xmm7,%xmm4
+	vpxor	%xmm4,%xmm6,%xmm6
+	vpslldq	$8,%xmm6,%xmm1
+	vmovdqu	16(%r11),%xmm3
+	vpsrldq	$8,%xmm6,%xmm6
+	vpxor	%xmm1,%xmm5,%xmm8
+	vpxor	%xmm6,%xmm7,%xmm7
+
+	vpalignr	$8,%xmm8,%xmm8,%xmm2
+	vpclmulqdq	$0x10,%xmm3,%xmm8,%xmm8
+	vpxor	%xmm2,%xmm8,%xmm8
+
+	vpalignr	$8,%xmm8,%xmm8,%xmm2
+	vpclmulqdq	$0x10,%xmm3,%xmm8,%xmm8
+	vpxor	%xmm7,%xmm2,%xmm2
+	vpxor	%xmm2,%xmm8,%xmm8
+	vpshufb	(%r11),%xmm8,%xmm8
+	vmovdqu	%xmm8,-64(%r9)
+
+	vzeroupper
+	movq	-48(%rax),%r15
+	movq	-40(%rax),%r14
+	movq	-32(%rax),%r13
+	movq	-24(%rax),%r12
+	movq	-16(%rax),%rbp
+	movq	-8(%rax),%rbx
+	leaq	(%rax),%rsp
+L$gcm_enc_abort:
+	movq	%r10,%rax
 	.byte	0xf3,0xc3
+
+.p2align	6
+L$bswap_mask:
+.byte	15,14,13,12,11,10,9,8,7,6,5,4,3,2,1,0
+L$poly:
+.byte	0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0xc2
+L$one_msb:
+.byte	0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1
+L$two_lsb:
+.byte	2,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0
+L$one_lsb:
+.byte	1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0
+.byte	65,69,83,45,78,73,32,71,67,77,32,109,111,100,117,108,101,32,102,111,114,32,120,56,54,95,54,52,44,32,67,82,89,80,84,79,71,65,77,83,32,98,121,32,60,97,112,112,114,111,64,111,112,101,110,115,115,108,46,111,114,103,62,0
+.p2align	6

deps/openssl/asm/x64-macosx-gas/modes/ghash-x86_64.s

@@ -661,10 +661,10 @@ L$ghash_epilogue:
 _gcm_init_clmul:
 L$_init_clmul:
 	movdqu	(%rsi),%xmm2
-	pshufd	$0b01001110,%xmm2,%xmm2
+	pshufd	$78,%xmm2,%xmm2
 
 
-	pshufd	$0b11111111,%xmm2,%xmm4
+	pshufd	$255,%xmm2,%xmm4
 	movdqa	%xmm2,%xmm3
 	psllq	$1,%xmm2
 	pxor	%xmm5,%xmm5
@@ -678,11 +678,11 @@ L$_init_clmul:
 	pxor	%xmm5,%xmm2
 
 
-	pshufd	$0b01001110,%xmm2,%xmm6
+	pshufd	$78,%xmm2,%xmm6
 	movdqa	%xmm2,%xmm0
 	pxor	%xmm2,%xmm6
 	movdqa	%xmm0,%xmm1
-	pshufd	$0b01001110,%xmm0,%xmm3
+	pshufd	$78,%xmm0,%xmm3
 	pxor	%xmm0,%xmm3
 .byte	102,15,58,68,194,0
 .byte	102,15,58,68,202,17
@@ -718,8 +718,8 @@ L$_init_clmul:
 	pxor	%xmm4,%xmm0
 	psrlq	$1,%xmm0
 	pxor	%xmm1,%xmm0
-	pshufd	$0b01001110,%xmm2,%xmm3
-	pshufd	$0b01001110,%xmm0,%xmm4
+	pshufd	$78,%xmm2,%xmm3
+	pshufd	$78,%xmm0,%xmm4
 	pxor	%xmm2,%xmm3
 	movdqu	%xmm2,0(%rdi)
 	pxor	%xmm0,%xmm4
@@ -727,7 +727,7 @@ L$_init_clmul:
 .byte	102,15,58,15,227,8
 	movdqu	%xmm4,32(%rdi)
 	movdqa	%xmm0,%xmm1
-	pshufd	$0b01001110,%xmm0,%xmm3
+	pshufd	$78,%xmm0,%xmm3
 	pxor	%xmm0,%xmm3
 .byte	102,15,58,68,194,0
 .byte	102,15,58,68,202,17
@@ -765,7 +765,7 @@ L$_init_clmul:
 	pxor	%xmm1,%xmm0
 	movdqa	%xmm0,%xmm5
 	movdqa	%xmm0,%xmm1
-	pshufd	$0b01001110,%xmm0,%xmm3
+	pshufd	$78,%xmm0,%xmm3
 	pxor	%xmm0,%xmm3
 .byte	102,15,58,68,194,0
 .byte	102,15,58,68,202,17
@@ -801,8 +801,8 @@ L$_init_clmul:
 	pxor	%xmm4,%xmm0
 	psrlq	$1,%xmm0
 	pxor	%xmm1,%xmm0
-	pshufd	$0b01001110,%xmm5,%xmm3
-	pshufd	$0b01001110,%xmm0,%xmm4
+	pshufd	$78,%xmm5,%xmm3
+	pshufd	$78,%xmm0,%xmm4
 	pxor	%xmm5,%xmm3
 	movdqu	%xmm5,48(%rdi)
 	pxor	%xmm0,%xmm4
@@ -822,7 +822,7 @@ L$_gmult_clmul:
 	movdqu	32(%rsi),%xmm4
 .byte	102,15,56,0,197
 	movdqa	%xmm0,%xmm1
-	pshufd	$0b01001110,%xmm0,%xmm3
+	pshufd	$78,%xmm0,%xmm3
 	pxor	%xmm0,%xmm3
 .byte	102,15,58,68,194,0
 .byte	102,15,58,68,202,17
@@ -899,14 +899,14 @@ L$_ghash_clmul:
 .byte	102,65,15,56,0,218
 .byte	102,69,15,56,0,218
 	movdqa	%xmm3,%xmm5
-	pshufd	$0b01001110,%xmm3,%xmm4
+	pshufd	$78,%xmm3,%xmm4
 	pxor	%xmm3,%xmm4
 .byte	102,15,58,68,218,0
 .byte	102,15,58,68,234,17
 .byte	102,15,58,68,231,0
 
 	movdqa	%xmm11,%xmm13
-	pshufd	$0b01001110,%xmm11,%xmm12
+	pshufd	$78,%xmm11,%xmm12
 	pxor	%xmm11,%xmm12
 .byte	102,68,15,58,68,222,0
 .byte	102,68,15,58,68,238,17
@@ -921,12 +921,12 @@ L$_ghash_clmul:
 .byte	102,69,15,56,0,218
 .byte	102,69,15,56,0,194
 	movdqa	%xmm11,%xmm13
-	pshufd	$0b01001110,%xmm11,%xmm12
+	pshufd	$78,%xmm11,%xmm12
 	pxor	%xmm8,%xmm0
 	pxor	%xmm11,%xmm12
 .byte	102,69,15,58,68,222,0
 	movdqa	%xmm0,%xmm1
-	pshufd	$0b01001110,%xmm0,%xmm8
+	pshufd	$78,%xmm0,%xmm8
 	pxor	%xmm0,%xmm8
 .byte	102,69,15,58,68,238,17
 .byte	102,68,15,58,68,231,0
@@ -949,14 +949,14 @@ L$mod4_loop:
 	movdqu	32(%rdx),%xmm3
 	movdqa	%xmm11,%xmm13
 .byte	102,68,15,58,68,199,16
-	pshufd	$0b01001110,%xmm11,%xmm12
+	pshufd	$78,%xmm11,%xmm12
 	xorps	%xmm5,%xmm1
 	pxor	%xmm11,%xmm12
 .byte	102,65,15,56,0,218
 	movups	32(%rsi),%xmm7
 	xorps	%xmm4,%xmm8
 .byte	102,68,15,58,68,218,0
-	pshufd	$0b01001110,%xmm3,%xmm4
+	pshufd	$78,%xmm3,%xmm4
 
 	pxor	%xmm0,%xmm8
 	movdqa	%xmm3,%xmm5
@@ -1000,7 +1000,7 @@ L$mod4_loop:
 
 	movdqa	%xmm11,%xmm13
 	pxor	%xmm12,%xmm4
-	pshufd	$0b01001110,%xmm11,%xmm12
+	pshufd	$78,%xmm11,%xmm12
 	pxor	%xmm9,%xmm0
 	pxor	%xmm8,%xmm1
 	pxor	%xmm11,%xmm12
@@ -1010,7 +1010,7 @@ L$mod4_loop:
 	movdqa	%xmm0,%xmm1
 .byte	102,69,15,58,68,238,17
 	xorps	%xmm11,%xmm3
-	pshufd	$0b01001110,%xmm0,%xmm8
+	pshufd	$78,%xmm0,%xmm8
 	pxor	%xmm0,%xmm8
 
 .byte	102,68,15,58,68,231,0
@@ -1079,7 +1079,7 @@ L$skip4x:
 	pxor	%xmm8,%xmm0
 
 	movdqa	%xmm3,%xmm5
-	pshufd	$0b01001110,%xmm3,%xmm4
+	pshufd	$78,%xmm3,%xmm4
 	pxor	%xmm3,%xmm4
 .byte	102,15,58,68,218,0
 .byte	102,15,58,68,234,17
@@ -1096,7 +1096,7 @@ L$skip4x:
 L$mod_loop:
 	movdqa	%xmm0,%xmm1
 	movdqa	%xmm4,%xmm8
-	pshufd	$0b01001110,%xmm0,%xmm4
+	pshufd	$78,%xmm0,%xmm4
 	pxor	%xmm0,%xmm4
 
 .byte	102,15,58,68,198,0
@@ -1134,7 +1134,7 @@ L$mod_loop:
 	pslldq	$8,%xmm0
 	psrldq	$8,%xmm8
 	pxor	%xmm9,%xmm0
-	pshufd	$0b01001110,%xmm5,%xmm4
+	pshufd	$78,%xmm5,%xmm4
 	pxor	%xmm8,%xmm1
 	pxor	%xmm5,%xmm4
 
@@ -1156,7 +1156,7 @@ L$mod_loop:
 L$even_tail:
 	movdqa	%xmm0,%xmm1
 	movdqa	%xmm4,%xmm8
-	pshufd	$0b01001110,%xmm0,%xmm4
+	pshufd	$78,%xmm0,%xmm4
 	pxor	%xmm0,%xmm4
 
 .byte	102,15,58,68,198,0
@@ -1204,7 +1204,7 @@ L$odd_tail:
 .byte	102,69,15,56,0,194
 	pxor	%xmm8,%xmm0
 	movdqa	%xmm0,%xmm1
-	pshufd	$0b01001110,%xmm0,%xmm3
+	pshufd	$78,%xmm0,%xmm3
 	pxor	%xmm0,%xmm3
 .byte	102,15,58,68,194,0
 .byte	102,15,58,68,202,17
@@ -1249,7 +1249,108 @@ L$done:
 
 .p2align	5
 _gcm_init_avx:
-	jmp	L$_init_clmul
+	vzeroupper
+
+	vmovdqu	(%rsi),%xmm2
+	vpshufd	$78,%xmm2,%xmm2
+
+
+	vpshufd	$255,%xmm2,%xmm4
+	vpsrlq	$63,%xmm2,%xmm3
+	vpsllq	$1,%xmm2,%xmm2
+	vpxor	%xmm5,%xmm5,%xmm5
+	vpcmpgtd	%xmm4,%xmm5,%xmm5
+	vpslldq	$8,%xmm3,%xmm3
+	vpor	%xmm3,%xmm2,%xmm2
+
+
+	vpand	L$0x1c2_polynomial(%rip),%xmm5,%xmm5
+	vpxor	%xmm5,%xmm2,%xmm2
+
+	vpunpckhqdq	%xmm2,%xmm2,%xmm6
+	vmovdqa	%xmm2,%xmm0
+	vpxor	%xmm2,%xmm6,%xmm6
+	movq	$4,%r10
+	jmp	L$init_start_avx
+.p2align	5
+L$init_loop_avx:
+	vpalignr	$8,%xmm3,%xmm4,%xmm5
+	vmovdqu	%xmm5,-16(%rdi)
+	vpunpckhqdq	%xmm0,%xmm0,%xmm3
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x11,%xmm2,%xmm0,%xmm1
+	vpclmulqdq	$0x00,%xmm2,%xmm0,%xmm0
+	vpclmulqdq	$0x00,%xmm6,%xmm3,%xmm3
+	vpxor	%xmm0,%xmm1,%xmm4
+	vpxor	%xmm4,%xmm3,%xmm3
+
+	vpslldq	$8,%xmm3,%xmm4
+	vpsrldq	$8,%xmm3,%xmm3
+	vpxor	%xmm4,%xmm0,%xmm0
+	vpxor	%xmm3,%xmm1,%xmm1
+	vpsllq	$57,%xmm0,%xmm3
+	vpsllq	$62,%xmm0,%xmm4
+	vpxor	%xmm3,%xmm4,%xmm4
+	vpsllq	$63,%xmm0,%xmm3
+	vpxor	%xmm3,%xmm4,%xmm4
+	vpslldq	$8,%xmm4,%xmm3
+	vpsrldq	$8,%xmm4,%xmm4
+	vpxor	%xmm3,%xmm0,%xmm0
+	vpxor	%xmm4,%xmm1,%xmm1
+
+	vpsrlq	$1,%xmm0,%xmm4
+	vpxor	%xmm0,%xmm1,%xmm1
+	vpxor	%xmm4,%xmm0,%xmm0
+	vpsrlq	$5,%xmm4,%xmm4
+	vpxor	%xmm4,%xmm0,%xmm0
+	vpsrlq	$1,%xmm0,%xmm0
+	vpxor	%xmm1,%xmm0,%xmm0
+L$init_start_avx:
+	vmovdqa	%xmm0,%xmm5
+	vpunpckhqdq	%xmm0,%xmm0,%xmm3
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x11,%xmm2,%xmm0,%xmm1
+	vpclmulqdq	$0x00,%xmm2,%xmm0,%xmm0
+	vpclmulqdq	$0x00,%xmm6,%xmm3,%xmm3
+	vpxor	%xmm0,%xmm1,%xmm4
+	vpxor	%xmm4,%xmm3,%xmm3
+
+	vpslldq	$8,%xmm3,%xmm4
+	vpsrldq	$8,%xmm3,%xmm3
+	vpxor	%xmm4,%xmm0,%xmm0
+	vpxor	%xmm3,%xmm1,%xmm1
+	vpsllq	$57,%xmm0,%xmm3
+	vpsllq	$62,%xmm0,%xmm4
+	vpxor	%xmm3,%xmm4,%xmm4
+	vpsllq	$63,%xmm0,%xmm3
+	vpxor	%xmm3,%xmm4,%xmm4
+	vpslldq	$8,%xmm4,%xmm3
+	vpsrldq	$8,%xmm4,%xmm4
+	vpxor	%xmm3,%xmm0,%xmm0
+	vpxor	%xmm4,%xmm1,%xmm1
+
+	vpsrlq	$1,%xmm0,%xmm4
+	vpxor	%xmm0,%xmm1,%xmm1
+	vpxor	%xmm4,%xmm0,%xmm0
+	vpsrlq	$5,%xmm4,%xmm4
+	vpxor	%xmm4,%xmm0,%xmm0
+	vpsrlq	$1,%xmm0,%xmm0
+	vpxor	%xmm1,%xmm0,%xmm0
+	vpshufd	$78,%xmm5,%xmm3
+	vpshufd	$78,%xmm0,%xmm4
+	vpxor	%xmm5,%xmm3,%xmm3
+	vmovdqu	%xmm5,0(%rdi)
+	vpxor	%xmm0,%xmm4,%xmm4
+	vmovdqu	%xmm0,16(%rdi)
+	leaq	48(%rdi),%rdi
+	subq	$1,%r10
+	jnz	L$init_loop_avx
+
+	vpalignr	$8,%xmm4,%xmm3,%xmm5
+	vmovdqu	%xmm5,-16(%rdi)
+
+	vzeroupper
+	.byte	0xf3,0xc3
 
 .globl	_gcm_gmult_avx
 
@@ -1261,7 +1362,377 @@ _gcm_gmult_avx:
 
 .p2align	5
 _gcm_ghash_avx:
-	jmp	L$_ghash_clmul
+	vzeroupper
+
+	vmovdqu	(%rdi),%xmm10
+	leaq	L$0x1c2_polynomial(%rip),%r10
+	leaq	64(%rsi),%rsi
+	vmovdqu	L$bswap_mask(%rip),%xmm13
+	vpshufb	%xmm13,%xmm10,%xmm10
+	cmpq	$0x80,%rcx
+	jb	L$short_avx
+	subq	$0x80,%rcx
+
+	vmovdqu	112(%rdx),%xmm14
+	vmovdqu	0-64(%rsi),%xmm6
+	vpshufb	%xmm13,%xmm14,%xmm14
+	vmovdqu	32-64(%rsi),%xmm7
+
+	vpunpckhqdq	%xmm14,%xmm14,%xmm9
+	vmovdqu	96(%rdx),%xmm15
+	vpclmulqdq	$0x00,%xmm6,%xmm14,%xmm0
+	vpxor	%xmm14,%xmm9,%xmm9
+	vpshufb	%xmm13,%xmm15,%xmm15
+	vpclmulqdq	$0x11,%xmm6,%xmm14,%xmm1
+	vmovdqu	16-64(%rsi),%xmm6
+	vpunpckhqdq	%xmm15,%xmm15,%xmm8
+	vmovdqu	80(%rdx),%xmm14
+	vpclmulqdq	$0x00,%xmm7,%xmm9,%xmm2
+	vpxor	%xmm15,%xmm8,%xmm8
+
+	vpshufb	%xmm13,%xmm14,%xmm14
+	vpclmulqdq	$0x00,%xmm6,%xmm15,%xmm3
+	vpunpckhqdq	%xmm14,%xmm14,%xmm9
+	vpclmulqdq	$0x11,%xmm6,%xmm15,%xmm4
+	vmovdqu	48-64(%rsi),%xmm6
+	vpxor	%xmm14,%xmm9,%xmm9
+	vmovdqu	64(%rdx),%xmm15
+	vpclmulqdq	$0x10,%xmm7,%xmm8,%xmm5
+	vmovdqu	80-64(%rsi),%xmm7
+
+	vpshufb	%xmm13,%xmm15,%xmm15
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x00,%xmm6,%xmm14,%xmm0
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpunpckhqdq	%xmm15,%xmm15,%xmm8
+	vpclmulqdq	$0x11,%xmm6,%xmm14,%xmm1
+	vmovdqu	64-64(%rsi),%xmm6
+	vpxor	%xmm2,%xmm5,%xmm5
+	vpclmulqdq	$0x00,%xmm7,%xmm9,%xmm2
+	vpxor	%xmm15,%xmm8,%xmm8
+
+	vmovdqu	48(%rdx),%xmm14
+	vpxor	%xmm3,%xmm0,%xmm0
+	vpclmulqdq	$0x00,%xmm6,%xmm15,%xmm3
+	vpxor	%xmm4,%xmm1,%xmm1
+	vpshufb	%xmm13,%xmm14,%xmm14
+	vpclmulqdq	$0x11,%xmm6,%xmm15,%xmm4
+	vmovdqu	96-64(%rsi),%xmm6
+	vpxor	%xmm5,%xmm2,%xmm2
+	vpunpckhqdq	%xmm14,%xmm14,%xmm9
+	vpclmulqdq	$0x10,%xmm7,%xmm8,%xmm5
+	vmovdqu	128-64(%rsi),%xmm7
+	vpxor	%xmm14,%xmm9,%xmm9
+
+	vmovdqu	32(%rdx),%xmm15
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x00,%xmm6,%xmm14,%xmm0
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpshufb	%xmm13,%xmm15,%xmm15
+	vpclmulqdq	$0x11,%xmm6,%xmm14,%xmm1
+	vmovdqu	112-64(%rsi),%xmm6
+	vpxor	%xmm2,%xmm5,%xmm5
+	vpunpckhqdq	%xmm15,%xmm15,%xmm8
+	vpclmulqdq	$0x00,%xmm7,%xmm9,%xmm2
+	vpxor	%xmm15,%xmm8,%xmm8
+
+	vmovdqu	16(%rdx),%xmm14
+	vpxor	%xmm3,%xmm0,%xmm0
+	vpclmulqdq	$0x00,%xmm6,%xmm15,%xmm3
+	vpxor	%xmm4,%xmm1,%xmm1
+	vpshufb	%xmm13,%xmm14,%xmm14
+	vpclmulqdq	$0x11,%xmm6,%xmm15,%xmm4
+	vmovdqu	144-64(%rsi),%xmm6
+	vpxor	%xmm5,%xmm2,%xmm2
+	vpunpckhqdq	%xmm14,%xmm14,%xmm9
+	vpclmulqdq	$0x10,%xmm7,%xmm8,%xmm5
+	vmovdqu	176-64(%rsi),%xmm7
+	vpxor	%xmm14,%xmm9,%xmm9
+
+	vmovdqu	(%rdx),%xmm15
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x00,%xmm6,%xmm14,%xmm0
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpshufb	%xmm13,%xmm15,%xmm15
+	vpclmulqdq	$0x11,%xmm6,%xmm14,%xmm1
+	vmovdqu	160-64(%rsi),%xmm6
+	vpxor	%xmm2,%xmm5,%xmm5
+	vpclmulqdq	$0x10,%xmm7,%xmm9,%xmm2
+
+	leaq	128(%rdx),%rdx
+	cmpq	$0x80,%rcx
+	jb	L$tail_avx
+
+	vpxor	%xmm10,%xmm15,%xmm15
+	subq	$0x80,%rcx
+	jmp	L$oop8x_avx
+
+.p2align	5
+L$oop8x_avx:
+	vpunpckhqdq	%xmm15,%xmm15,%xmm8
+	vmovdqu	112(%rdx),%xmm14
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpxor	%xmm15,%xmm8,%xmm8
+	vpclmulqdq	$0x00,%xmm6,%xmm15,%xmm10
+	vpshufb	%xmm13,%xmm14,%xmm14
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpclmulqdq	$0x11,%xmm6,%xmm15,%xmm11
+	vmovdqu	0-64(%rsi),%xmm6
+	vpunpckhqdq	%xmm14,%xmm14,%xmm9
+	vpxor	%xmm2,%xmm5,%xmm5
+	vpclmulqdq	$0x00,%xmm7,%xmm8,%xmm12
+	vmovdqu	32-64(%rsi),%xmm7
+	vpxor	%xmm14,%xmm9,%xmm9
+
+	vmovdqu	96(%rdx),%xmm15
+	vpclmulqdq	$0x00,%xmm6,%xmm14,%xmm0
+	vpxor	%xmm3,%xmm10,%xmm10
+	vpshufb	%xmm13,%xmm15,%xmm15
+	vpclmulqdq	$0x11,%xmm6,%xmm14,%xmm1
+	vxorps	%xmm4,%xmm11,%xmm11
+	vmovdqu	16-64(%rsi),%xmm6
+	vpunpckhqdq	%xmm15,%xmm15,%xmm8
+	vpclmulqdq	$0x00,%xmm7,%xmm9,%xmm2
+	vpxor	%xmm5,%xmm12,%xmm12
+	vxorps	%xmm15,%xmm8,%xmm8
+
+	vmovdqu	80(%rdx),%xmm14
+	vpxor	%xmm10,%xmm12,%xmm12
+	vpclmulqdq	$0x00,%xmm6,%xmm15,%xmm3
+	vpxor	%xmm11,%xmm12,%xmm12
+	vpslldq	$8,%xmm12,%xmm9
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x11,%xmm6,%xmm15,%xmm4
+	vpsrldq	$8,%xmm12,%xmm12
+	vpxor	%xmm9,%xmm10,%xmm10
+	vmovdqu	48-64(%rsi),%xmm6
+	vpshufb	%xmm13,%xmm14,%xmm14
+	vxorps	%xmm12,%xmm11,%xmm11
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpunpckhqdq	%xmm14,%xmm14,%xmm9
+	vpclmulqdq	$0x10,%xmm7,%xmm8,%xmm5
+	vmovdqu	80-64(%rsi),%xmm7
+	vpxor	%xmm14,%xmm9,%xmm9
+	vpxor	%xmm2,%xmm5,%xmm5
+
+	vmovdqu	64(%rdx),%xmm15
+	vpalignr	$8,%xmm10,%xmm10,%xmm12
+	vpclmulqdq	$0x00,%xmm6,%xmm14,%xmm0
+	vpshufb	%xmm13,%xmm15,%xmm15
+	vpxor	%xmm3,%xmm0,%xmm0
+	vpclmulqdq	$0x11,%xmm6,%xmm14,%xmm1
+	vmovdqu	64-64(%rsi),%xmm6
+	vpunpckhqdq	%xmm15,%xmm15,%xmm8
+	vpxor	%xmm4,%xmm1,%xmm1
+	vpclmulqdq	$0x00,%xmm7,%xmm9,%xmm2
+	vxorps	%xmm15,%xmm8,%xmm8
+	vpxor	%xmm5,%xmm2,%xmm2
+
+	vmovdqu	48(%rdx),%xmm14
+	vpclmulqdq	$0x10,(%r10),%xmm10,%xmm10
+	vpclmulqdq	$0x00,%xmm6,%xmm15,%xmm3
+	vpshufb	%xmm13,%xmm14,%xmm14
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x11,%xmm6,%xmm15,%xmm4
+	vmovdqu	96-64(%rsi),%xmm6
+	vpunpckhqdq	%xmm14,%xmm14,%xmm9
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpclmulqdq	$0x10,%xmm7,%xmm8,%xmm5
+	vmovdqu	128-64(%rsi),%xmm7
+	vpxor	%xmm14,%xmm9,%xmm9
+	vpxor	%xmm2,%xmm5,%xmm5
+
+	vmovdqu	32(%rdx),%xmm15
+	vpclmulqdq	$0x00,%xmm6,%xmm14,%xmm0
+	vpshufb	%xmm13,%xmm15,%xmm15
+	vpxor	%xmm3,%xmm0,%xmm0
+	vpclmulqdq	$0x11,%xmm6,%xmm14,%xmm1
+	vmovdqu	112-64(%rsi),%xmm6
+	vpunpckhqdq	%xmm15,%xmm15,%xmm8
+	vpxor	%xmm4,%xmm1,%xmm1
+	vpclmulqdq	$0x00,%xmm7,%xmm9,%xmm2
+	vpxor	%xmm15,%xmm8,%xmm8
+	vpxor	%xmm5,%xmm2,%xmm2
+	vxorps	%xmm12,%xmm10,%xmm10
+
+	vmovdqu	16(%rdx),%xmm14
+	vpalignr	$8,%xmm10,%xmm10,%xmm12
+	vpclmulqdq	$0x00,%xmm6,%xmm15,%xmm3
+	vpshufb	%xmm13,%xmm14,%xmm14
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x11,%xmm6,%xmm15,%xmm4
+	vmovdqu	144-64(%rsi),%xmm6
+	vpclmulqdq	$0x10,(%r10),%xmm10,%xmm10
+	vxorps	%xmm11,%xmm12,%xmm12
+	vpunpckhqdq	%xmm14,%xmm14,%xmm9
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpclmulqdq	$0x10,%xmm7,%xmm8,%xmm5
+	vmovdqu	176-64(%rsi),%xmm7
+	vpxor	%xmm14,%xmm9,%xmm9
+	vpxor	%xmm2,%xmm5,%xmm5
+
+	vmovdqu	(%rdx),%xmm15
+	vpclmulqdq	$0x00,%xmm6,%xmm14,%xmm0
+	vpshufb	%xmm13,%xmm15,%xmm15
+	vpclmulqdq	$0x11,%xmm6,%xmm14,%xmm1
+	vmovdqu	160-64(%rsi),%xmm6
+	vpxor	%xmm12,%xmm15,%xmm15
+	vpclmulqdq	$0x10,%xmm7,%xmm9,%xmm2
+	vpxor	%xmm10,%xmm15,%xmm15
+
+	leaq	128(%rdx),%rdx
+	subq	$0x80,%rcx
+	jnc	L$oop8x_avx
+
+	addq	$0x80,%rcx
+	jmp	L$tail_no_xor_avx
+
+.p2align	5
+L$short_avx:
+	vmovdqu	-16(%rdx,%rcx,1),%xmm14
+	leaq	(%rdx,%rcx,1),%rdx
+	vmovdqu	0-64(%rsi),%xmm6
+	vmovdqu	32-64(%rsi),%xmm7
+	vpshufb	%xmm13,%xmm14,%xmm15
+
+	vmovdqa	%xmm0,%xmm3
+	vmovdqa	%xmm1,%xmm4
+	vmovdqa	%xmm2,%xmm5
+	subq	$0x10,%rcx
+	jz	L$tail_avx
+
+	vpunpckhqdq	%xmm15,%xmm15,%xmm8
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x00,%xmm6,%xmm15,%xmm0
+	vpxor	%xmm15,%xmm8,%xmm8
+	vmovdqu	-32(%rdx),%xmm14
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpclmulqdq	$0x11,%xmm6,%xmm15,%xmm1
+	vmovdqu	16-64(%rsi),%xmm6
+	vpshufb	%xmm13,%xmm14,%xmm15
+	vpxor	%xmm2,%xmm5,%xmm5
+	vpclmulqdq	$0x00,%xmm7,%xmm8,%xmm2
+	vpsrldq	$8,%xmm7,%xmm7
+	subq	$0x10,%rcx
+	jz	L$tail_avx
+
+	vpunpckhqdq	%xmm15,%xmm15,%xmm8
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x00,%xmm6,%xmm15,%xmm0
+	vpxor	%xmm15,%xmm8,%xmm8
+	vmovdqu	-48(%rdx),%xmm14
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpclmulqdq	$0x11,%xmm6,%xmm15,%xmm1
+	vmovdqu	48-64(%rsi),%xmm6
+	vpshufb	%xmm13,%xmm14,%xmm15
+	vpxor	%xmm2,%xmm5,%xmm5
+	vpclmulqdq	$0x00,%xmm7,%xmm8,%xmm2
+	vmovdqu	80-64(%rsi),%xmm7
+	subq	$0x10,%rcx
+	jz	L$tail_avx
+
+	vpunpckhqdq	%xmm15,%xmm15,%xmm8
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x00,%xmm6,%xmm15,%xmm0
+	vpxor	%xmm15,%xmm8,%xmm8
+	vmovdqu	-64(%rdx),%xmm14
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpclmulqdq	$0x11,%xmm6,%xmm15,%xmm1
+	vmovdqu	64-64(%rsi),%xmm6
+	vpshufb	%xmm13,%xmm14,%xmm15
+	vpxor	%xmm2,%xmm5,%xmm5
+	vpclmulqdq	$0x00,%xmm7,%xmm8,%xmm2
+	vpsrldq	$8,%xmm7,%xmm7
+	subq	$0x10,%rcx
+	jz	L$tail_avx
+
+	vpunpckhqdq	%xmm15,%xmm15,%xmm8
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x00,%xmm6,%xmm15,%xmm0
+	vpxor	%xmm15,%xmm8,%xmm8
+	vmovdqu	-80(%rdx),%xmm14
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpclmulqdq	$0x11,%xmm6,%xmm15,%xmm1
+	vmovdqu	96-64(%rsi),%xmm6
+	vpshufb	%xmm13,%xmm14,%xmm15
+	vpxor	%xmm2,%xmm5,%xmm5
+	vpclmulqdq	$0x00,%xmm7,%xmm8,%xmm2
+	vmovdqu	128-64(%rsi),%xmm7
+	subq	$0x10,%rcx
+	jz	L$tail_avx
+
+	vpunpckhqdq	%xmm15,%xmm15,%xmm8
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x00,%xmm6,%xmm15,%xmm0
+	vpxor	%xmm15,%xmm8,%xmm8
+	vmovdqu	-96(%rdx),%xmm14
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpclmulqdq	$0x11,%xmm6,%xmm15,%xmm1
+	vmovdqu	112-64(%rsi),%xmm6
+	vpshufb	%xmm13,%xmm14,%xmm15
+	vpxor	%xmm2,%xmm5,%xmm5
+	vpclmulqdq	$0x00,%xmm7,%xmm8,%xmm2
+	vpsrldq	$8,%xmm7,%xmm7
+	subq	$0x10,%rcx
+	jz	L$tail_avx
+
+	vpunpckhqdq	%xmm15,%xmm15,%xmm8
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x00,%xmm6,%xmm15,%xmm0
+	vpxor	%xmm15,%xmm8,%xmm8
+	vmovdqu	-112(%rdx),%xmm14
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpclmulqdq	$0x11,%xmm6,%xmm15,%xmm1
+	vmovdqu	144-64(%rsi),%xmm6
+	vpshufb	%xmm13,%xmm14,%xmm15
+	vpxor	%xmm2,%xmm5,%xmm5
+	vpclmulqdq	$0x00,%xmm7,%xmm8,%xmm2
+	vmovq	184-64(%rsi),%xmm7
+	subq	$0x10,%rcx
+	jmp	L$tail_avx
+
+.p2align	5
+L$tail_avx:
+	vpxor	%xmm10,%xmm15,%xmm15
+L$tail_no_xor_avx:
+	vpunpckhqdq	%xmm15,%xmm15,%xmm8
+	vpxor	%xmm0,%xmm3,%xmm3
+	vpclmulqdq	$0x00,%xmm6,%xmm15,%xmm0
+	vpxor	%xmm15,%xmm8,%xmm8
+	vpxor	%xmm1,%xmm4,%xmm4
+	vpclmulqdq	$0x11,%xmm6,%xmm15,%xmm1
+	vpxor	%xmm2,%xmm5,%xmm5
+	vpclmulqdq	$0x00,%xmm7,%xmm8,%xmm2
+
+	vmovdqu	(%r10),%xmm12
+
+	vpxor	%xmm0,%xmm3,%xmm10
+	vpxor	%xmm1,%xmm4,%xmm11
+	vpxor	%xmm2,%xmm5,%xmm5
+
+	vpxor	%xmm10,%xmm5,%xmm5
+	vpxor	%xmm11,%xmm5,%xmm5
+	vpslldq	$8,%xmm5,%xmm9
+	vpsrldq	$8,%xmm5,%xmm5
+	vpxor	%xmm9,%xmm10,%xmm10
+	vpxor	%xmm5,%xmm11,%xmm11
+
+	vpclmulqdq	$0x10,%xmm12,%xmm10,%xmm9
+	vpalignr	$8,%xmm10,%xmm10,%xmm10
+	vpxor	%xmm9,%xmm10,%xmm10
+
+	vpclmulqdq	$0x10,%xmm12,%xmm10,%xmm9
+	vpalignr	$8,%xmm10,%xmm10,%xmm10
+	vpxor	%xmm11,%xmm10,%xmm10
+	vpxor	%xmm9,%xmm10,%xmm10
+
+	cmpq	$0,%rcx
+	jne	L$short_avx
+
+	vpshufb	%xmm13,%xmm10,%xmm10
+	vmovdqu	%xmm10,(%rdi)
+	vzeroupper
+	.byte	0xf3,0xc3
 
 .p2align	6
 L$bswap_mask:

deps/openssl/asm/x64-macosx-gas/x86_64cpuid.s

@@ -45,43 +45,43 @@ _OPENSSL_ia32_cpuid:
 	movl	%eax,%r11d
 
 	xorl	%eax,%eax
-	cmpl	$1970169159,%ebx
+	cmpl	$0x756e6547,%ebx
 	setne	%al
 	movl	%eax,%r9d
-	cmpl	$1231384169,%edx
+	cmpl	$0x49656e69,%edx
 	setne	%al
 	orl	%eax,%r9d
-	cmpl	$1818588270,%ecx
+	cmpl	$0x6c65746e,%ecx
 	setne	%al
 	orl	%eax,%r9d
 	jz	L$intel
 
-	cmpl	$1752462657,%ebx
+	cmpl	$0x68747541,%ebx
 	setne	%al
 	movl	%eax,%r10d
-	cmpl	$1769238117,%edx
+	cmpl	$0x69746E65,%edx
 	setne	%al
 	orl	%eax,%r10d
-	cmpl	$1145913699,%ecx
+	cmpl	$0x444D4163,%ecx
 	setne	%al
 	orl	%eax,%r10d
 	jnz	L$intel
 
 
-	movl	$2147483648,%eax
+	movl	$0x80000000,%eax
 	cpuid
-	cmpl	$2147483649,%eax
+	cmpl	$0x80000001,%eax
 	jb	L$intel
 	movl	%eax,%r10d
-	movl	$2147483649,%eax
+	movl	$0x80000001,%eax
 	cpuid
 	orl	%ecx,%r9d
-	andl	$2049,%r9d
+	andl	$0x00000801,%r9d
 
-	cmpl	$2147483656,%r10d
+	cmpl	$0x80000008,%r10d
 	jb	L$intel
 
-	movl	$2147483656,%eax
+	movl	$0x80000008,%eax
 	cpuid
 	movzbq	%cl,%r10
 	incq	%r10
@@ -93,7 +93,7 @@ _OPENSSL_ia32_cpuid:
 	shrl	$16,%ebx
 	cmpb	%r10b,%bl
 	ja	L$generic
-	andl	$4026531839,%edx
+	andl	$0xefffffff,%edx
 	jmp	L$generic
 
 L$intel:
@@ -106,7 +106,7 @@ L$intel:
 	cpuid
 	movl	%eax,%r10d
 	shrl	$14,%r10d
-	andl	$4095,%r10d
+	andl	$0xfff,%r10d
 
 	cmpl	$7,%r11d
 	jb	L$nocacheinfo
@@ -119,29 +119,29 @@ L$intel:
 L$nocacheinfo:
 	movl	$1,%eax
 	cpuid
-	andl	$3220176895,%edx
+	andl	$0xbfefffff,%edx
 	cmpl	$0,%r9d
 	jne	L$notintel
-	orl	$1073741824,%edx
+	orl	$0x40000000,%edx
 	andb	$15,%ah
 	cmpb	$15,%ah
 	jne	L$notintel
-	orl	$1048576,%edx
+	orl	$0x00100000,%edx
 L$notintel:
 	btl	$28,%edx
 	jnc	L$generic
-	andl	$4026531839,%edx
+	andl	$0xefffffff,%edx
 	cmpl	$0,%r10d
 	je	L$generic
 
-	orl	$268435456,%edx
+	orl	$0x10000000,%edx
 	shrl	$16,%ebx
 	cmpb	$1,%bl
 	ja	L$generic
-	andl	$4026531839,%edx
+	andl	$0xefffffff,%edx
 L$generic:
-	andl	$2048,%r9d
-	andl	$4294965247,%ecx
+	andl	$0x00000800,%r9d
+	andl	$0xfffff7ff,%ecx
 	orl	%ecx,%r9d
 
 	movl	%edx,%r10d
@@ -153,9 +153,9 @@ L$generic:
 	cmpl	$6,%eax
 	je	L$done
 L$clear_avx:
-	movl	$4026525695,%eax
+	movl	$0xefffe7ff,%eax
 	andl	%eax,%r9d
-	andl	$4294967263,8(%rdi)
+	andl	$0xffffffdf,8(%rdi)
 L$done:
 	shlq	$32,%r9
 	movl	%r10d,%eax

deps/openssl/asm/x64-win32-masm/bn/x86_64-mont.asm

@@ -48,6 +48,20 @@ $L$mul_enter::
 
 	mov	QWORD PTR[8+r9*8+rsp],r11
 $L$mul_body::
+
+
+
+
+
+
+	sub	r11,rsp
+	and	r11,-4096
+$L$mul_page_walk::
+	mov	r10,QWORD PTR[r11*1+rsp]
+	sub	r11,4096
+DB	066h,02eh
+	jnc	$L$mul_page_walk
+
 	mov	r12,rdx
 	mov	r8,QWORD PTR[r8]
 	mov	rbx,QWORD PTR[r12]
@@ -263,6 +277,14 @@ $L$mul4x_enter::
 
 	mov	QWORD PTR[8+r9*8+rsp],r11
 $L$mul4x_body::
+	sub	r11,rsp
+	and	r11,-4096
+$L$mul4x_page_walk::
+	mov	r10,QWORD PTR[r11*1+rsp]
+	sub	r11,4096
+DB	02eh
+	jnc	$L$mul4x_page_walk
+
 	mov	QWORD PTR[16+r9*8+rsp],rdi
 	mov	r12,rdx
 	mov	r8,QWORD PTR[r8]
@@ -701,6 +723,15 @@ $L$sqr8x_sp_alt::
 	sub	rsp,r11
 $L$sqr8x_sp_done::
 	and	rsp,-64
+	mov	r11,rax
+	sub	r11,rsp
+	and	r11,-4096
+$L$sqr8x_page_walk::
+	mov	r10,QWORD PTR[r11*1+rsp]
+	sub	r11,4096
+DB	02eh
+	jnc	$L$sqr8x_page_walk
+
 	mov	r10,r9
 	neg	r9
 
@@ -842,8 +873,17 @@ DB	067h
 	sub	r10,r9
 	mov	r8,QWORD PTR[r8]
 	lea	rsp,QWORD PTR[((-72))+r10*1+rsp]
-	lea	r10,QWORD PTR[r9*1+rdx]
 	and	rsp,-128
+	mov	r11,rax
+	sub	r11,rsp
+	and	r11,-4096
+$L$mulx4x_page_walk::
+	mov	r10,QWORD PTR[r11*1+rsp]
+	sub	r11,4096
+DB	066h,02eh
+	jnc	$L$mulx4x_page_walk
+
+	lea	r10,QWORD PTR[r9*1+rdx]
 
 
 

deps/openssl/asm/x64-win32-masm/bn/x86_64-mont5.asm

@@ -44,6 +44,20 @@ $L$mul_enter::
 
 	mov	QWORD PTR[8+r9*8+rsp],rax
 $L$mul_body::
+
+
+
+
+
+
+	sub	rax,rsp
+	and	rax,-4096
+$L$mul_page_walk::
+	mov	r11,QWORD PTR[rax*1+rsp]
+	sub	rax,4096
+DB	02eh
+	jnc	$L$mul_page_walk
+
 	lea	r12,QWORD PTR[128+rdx]
 	movdqa	xmm0,XMMWORD PTR[r10]
 	movdqa	xmm1,XMMWORD PTR[16+r10]
@@ -474,6 +488,15 @@ $L$mul4xsp_alt::
 	sub	rsp,r11
 $L$mul4xsp_done::
 	and	rsp,-64
+	mov	r11,rax
+	sub	r11,rsp
+	and	r11,-4096
+$L$mul4x_page_walk::
+	mov	r10,QWORD PTR[r11*1+rsp]
+	sub	r11,4096
+DB	02eh
+	jnc	$L$mul4x_page_walk
+
 	neg	r9
 
 	mov	QWORD PTR[40+rsp],rax
@@ -1082,6 +1105,15 @@ $L$pwr_sp_alt::
 	sub	rsp,r11
 $L$pwr_sp_done::
 	and	rsp,-64
+	mov	r11,rax
+	sub	r11,rsp
+	and	r11,-4096
+$L$pwr_page_walk::
+	mov	r10,QWORD PTR[r11*1+rsp]
+	sub	r11,4096
+DB	02eh
+	jnc	$L$pwr_page_walk
+
 	mov	r10,r9
 	neg	r9
 
@@ -2038,6 +2070,15 @@ $L$from_sp_alt::
 	sub	rsp,r11
 $L$from_sp_done::
 	and	rsp,-64
+	mov	r11,rax
+	sub	r11,rsp
+	and	r11,-4096
+$L$from_page_walk::
+	mov	r10,QWORD PTR[r11*1+rsp]
+	sub	r11,4096
+DB	02eh
+	jnc	$L$from_page_walk
+
 	mov	r10,r9
 	neg	r9
 
@@ -2186,6 +2227,15 @@ $L$mulx4xsp_alt::
 	sub	rsp,r11
 $L$mulx4xsp_done::
 	and	rsp,-64
+	mov	r11,rax
+	sub	r11,rsp
+	and	r11,-4096
+$L$mulx4x_page_walk::
+	mov	r10,QWORD PTR[r11*1+rsp]
+	sub	r11,4096
+DB	02eh
+	jnc	$L$mulx4x_page_walk
+
 
 
 
@@ -2697,6 +2747,15 @@ $L$pwrx_sp_alt::
 	sub	rsp,r11
 $L$pwrx_sp_done::
 	and	rsp,-64
+	mov	r11,rax
+	sub	r11,rsp
+	and	r11,-4096
+$L$pwrx_page_walk::
+	mov	r10,QWORD PTR[r11*1+rsp]
+	sub	r11,4096
+DB	02eh
+	jnc	$L$pwrx_page_walk
+
 	mov	r10,r9
 	neg	r9
 

deps/openssl/asm/x86-elf-gas/bn/x86-mont.s

@@ -29,6 +29,14 @@ bn_mul_mont:
 	xorl	$2048,%edx
 	subl	%edx,%esp
 	andl	$-64,%esp
+	movl	%ebp,%eax
+	subl	%esp,%eax
+	andl	$-4096,%eax
+.L001page_walk:
+	movl	(%esp,%eax,1),%edx
+	subl	$4096,%eax
+.byte	46
+	jnc	.L001page_walk
 	movl	(%esi),%eax
 	movl	4(%esi),%ebx
 	movl	8(%esi),%ecx
@@ -44,7 +52,7 @@ bn_mul_mont:
 	movl	%ebp,24(%esp)
 	leal	OPENSSL_ia32cap_P,%eax
 	btl	$26,(%eax)
-	jnc	.L001non_sse2
+	jnc	.L002non_sse2
 	movl	$-1,%eax
 	movd	%eax,%mm7
 	movl	8(%esp),%esi
@@ -68,7 +76,7 @@ bn_mul_mont:
 	psrlq	$32,%mm3
 	incl	%ecx
 .align	16
-.L0021st:
+.L0031st:
 	pmuludq	%mm4,%mm0
 	pmuludq	%mm5,%mm1
 	paddq	%mm0,%mm2
@@ -83,7 +91,7 @@ bn_mul_mont:
 	psrlq	$32,%mm3
 	leal	1(%ecx),%ecx
 	cmpl	%ebx,%ecx
-	jl	.L0021st
+	jl	.L0031st
 	pmuludq	%mm4,%mm0
 	pmuludq	%mm5,%mm1
 	paddq	%mm0,%mm2
@@ -97,7 +105,7 @@ bn_mul_mont:
 	paddq	%mm2,%mm3
 	movq	%mm3,32(%esp,%ebx,4)
 	incl	%edx
-.L003outer:
+.L004outer:
 	xorl	%ecx,%ecx
 	movd	(%edi,%edx,4),%mm4
 	movd	(%esi),%mm5
@@ -119,7 +127,7 @@ bn_mul_mont:
 	paddq	%mm6,%mm2
 	incl	%ecx
 	decl	%ebx
-.L004inner:
+.L005inner:
 	pmuludq	%mm4,%mm0
 	pmuludq	%mm5,%mm1
 	paddq	%mm0,%mm2
@@ -136,7 +144,7 @@ bn_mul_mont:
 	paddq	%mm6,%mm2
 	decl	%ebx
 	leal	1(%ecx),%ecx
-	jnz	.L004inner
+	jnz	.L005inner
 	movl	%ecx,%ebx
 	pmuludq	%mm4,%mm0
 	pmuludq	%mm5,%mm1
@@ -154,11 +162,11 @@ bn_mul_mont:
 	movq	%mm3,32(%esp,%ebx,4)
 	leal	1(%edx),%edx
 	cmpl	%ebx,%edx
-	jle	.L003outer
+	jle	.L004outer
 	emms
-	jmp	.L005common_tail
+	jmp	.L006common_tail
 .align	16
-.L001non_sse2:
+.L002non_sse2:
 	movl	8(%esp),%esi
 	leal	1(%ebx),%ebp
 	movl	12(%esp),%edi
@@ -169,12 +177,12 @@ bn_mul_mont:
 	leal	4(%edi,%ebx,4),%eax
 	orl	%edx,%ebp
 	movl	(%edi),%edi
-	jz	.L006bn_sqr_mont
+	jz	.L007bn_sqr_mont
 	movl	%eax,28(%esp)
 	movl	(%esi),%eax
 	xorl	%edx,%edx
 .align	16
-.L007mull:
+.L008mull:
 	movl	%edx,%ebp
 	mull	%edi
 	addl	%eax,%ebp
@@ -183,7 +191,7 @@ bn_mul_mont:
 	movl	(%esi,%ecx,4),%eax
 	cmpl	%ebx,%ecx
 	movl	%ebp,28(%esp,%ecx,4)
-	jl	.L007mull
+	jl	.L008mull
 	movl	%edx,%ebp
 	mull	%edi
 	movl	20(%esp),%edi
@@ -201,9 +209,9 @@ bn_mul_mont:
 	movl	4(%esi),%eax
 	adcl	$0,%edx
 	incl	%ecx
-	jmp	.L0082ndmadd
+	jmp	.L0092ndmadd
 .align	16
-.L0091stmadd:
+.L0101stmadd:
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ecx,4),%ebp
@@ -214,7 +222,7 @@ bn_mul_mont:
 	adcl	$0,%edx
 	cmpl	%ebx,%ecx
 	movl	%ebp,28(%esp,%ecx,4)
-	jl	.L0091stmadd
+	jl	.L0101stmadd
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ebx,4),%eax
@@ -237,7 +245,7 @@ bn_mul_mont:
 	adcl	$0,%edx
 	movl	$1,%ecx
 .align	16
-.L0082ndmadd:
+.L0092ndmadd:
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ecx,4),%ebp
@@ -248,7 +256,7 @@ bn_mul_mont:
 	adcl	$0,%edx
 	cmpl	%ebx,%ecx
 	movl	%ebp,24(%esp,%ecx,4)
-	jl	.L0082ndmadd
+	jl	.L0092ndmadd
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ebx,4),%ebp
@@ -264,16 +272,16 @@ bn_mul_mont:
 	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	.L0091stmadd
+	jmp	.L0101stmadd
 .align	16
-.L006bn_sqr_mont:
+.L007bn_sqr_mont:
 	movl	%ebx,(%esp)
 	movl	%ecx,12(%esp)
 	movl	%edi,%eax
@@ -284,7 +292,7 @@ bn_mul_mont:
 	andl	$1,%ebx
 	incl	%ecx
 .align	16
-.L010sqr:
+.L011sqr:
 	movl	(%esi,%ecx,4),%eax
 	movl	%edx,%ebp
 	mull	%edi
@@ -296,7 +304,7 @@ bn_mul_mont:
 	cmpl	(%esp),%ecx
 	movl	%eax,%ebx
 	movl	%ebp,28(%esp,%ecx,4)
-	jl	.L010sqr
+	jl	.L011sqr
 	movl	(%esi,%ecx,4),%eax
 	movl	%edx,%ebp
 	mull	%edi
@@ -320,7 +328,7 @@ bn_mul_mont:
 	movl	4(%esi),%eax
 	movl	$1,%ecx
 .align	16
-.L0113rdmadd:
+.L0123rdmadd:
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ecx,4),%ebp
@@ -339,7 +347,7 @@ bn_mul_mont:
 	adcl	$0,%edx
 	cmpl	%ebx,%ecx
 	movl	%ebp,24(%esp,%ecx,4)
-	jl	.L0113rdmadd
+	jl	.L0123rdmadd
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ebx,4),%ebp
@@ -355,7 +363,7 @@ bn_mul_mont:
 	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
@@ -367,12 +375,12 @@ bn_mul_mont:
 	xorl	%ebp,%ebp
 	cmpl	%ebx,%ecx
 	leal	1(%ecx),%ecx
-	je	.L012sqrlast
+	je	.L013sqrlast
 	movl	%edx,%ebx
 	shrl	$1,%edx
 	andl	$1,%ebx
 .align	16
-.L013sqradd:
+.L014sqradd:
 	movl	(%esi,%ecx,4),%eax
 	movl	%edx,%ebp
 	mull	%edi
@@ -388,13 +396,13 @@ bn_mul_mont:
 	cmpl	(%esp),%ecx
 	movl	%ebp,28(%esp,%ecx,4)
 	movl	%eax,%ebx
-	jle	.L013sqradd
+	jle	.L014sqradd
 	movl	%edx,%ebp
 	addl	%edx,%edx
 	shrl	$31,%ebp
 	addl	%ebx,%edx
 	adcl	$0,%ebp
-.L012sqrlast:
+.L013sqrlast:
 	movl	20(%esp),%edi
 	movl	16(%esp),%esi
 	imull	32(%esp),%edi
@@ -409,9 +417,9 @@ bn_mul_mont:
 	adcl	$0,%edx
 	movl	$1,%ecx
 	movl	4(%esi),%eax
-	jmp	.L0113rdmadd
+	jmp	.L0123rdmadd
 .align	16
-.L005common_tail:
+.L006common_tail:
 	movl	16(%esp),%ebp
 	movl	4(%esp),%edi
 	leal	32(%esp),%esi
@@ -419,13 +427,13 @@ bn_mul_mont:
 	movl	%ebx,%ecx
 	xorl	%edx,%edx
 .align	16
-.L014sub:
+.L015sub:
 	sbbl	(%ebp,%edx,4),%eax
 	movl	%eax,(%edi,%edx,4)
 	decl	%ecx
 	movl	4(%esi,%edx,4),%eax
 	leal	1(%edx),%edx
-	jge	.L014sub
+	jge	.L015sub
 	sbbl	$0,%eax
 	andl	%eax,%esi
 	notl	%eax
@@ -433,12 +441,12 @@ bn_mul_mont:
 	andl	%eax,%ebp
 	orl	%ebp,%esi
 .align	16
-.L015copy:
+.L016copy:
 	movl	(%esi,%ebx,4),%eax
 	movl	%eax,(%edi,%ebx,4)
 	movl	%ecx,32(%esp,%ebx,4)
 	decl	%ebx
-	jge	.L015copy
+	jge	.L016copy
 	movl	24(%esp),%esp
 	movl	$1,%eax
 .L000just_leave:

deps/openssl/asm/x86-macosx-gas/bn/x86-mont.s

@@ -28,6 +28,14 @@ L_bn_mul_mont_begin:
 	xorl	$2048,%edx
 	subl	%edx,%esp
 	andl	$-64,%esp
+	movl	%ebp,%eax
+	subl	%esp,%eax
+	andl	$-4096,%eax
+L001page_walk:
+	movl	(%esp,%eax,1),%edx
+	subl	$4096,%eax
+.byte	46
+	jnc	L001page_walk
 	movl	(%esi),%eax
 	movl	4(%esi),%ebx
 	movl	8(%esi),%ecx
@@ -41,12 +49,12 @@ L_bn_mul_mont_begin:
 	movl	%esi,20(%esp)
 	leal	-3(%edi),%ebx
 	movl	%ebp,24(%esp)
-	call	L001PIC_me_up
-L001PIC_me_up:
+	call	L002PIC_me_up
+L002PIC_me_up:
 	popl	%eax
-	movl	L_OPENSSL_ia32cap_P$non_lazy_ptr-L001PIC_me_up(%eax),%eax
+	movl	L_OPENSSL_ia32cap_P$non_lazy_ptr-L002PIC_me_up(%eax),%eax
 	btl	$26,(%eax)
-	jnc	L002non_sse2
+	jnc	L003non_sse2
 	movl	$-1,%eax
 	movd	%eax,%mm7
 	movl	8(%esp),%esi
@@ -70,7 +78,7 @@ L001PIC_me_up:
 	psrlq	$32,%mm3
 	incl	%ecx
 .align	4,0x90
-L0031st:
+L0041st:
 	pmuludq	%mm4,%mm0
 	pmuludq	%mm5,%mm1
 	paddq	%mm0,%mm2
@@ -85,7 +93,7 @@ L0031st:
 	psrlq	$32,%mm3
 	leal	1(%ecx),%ecx
 	cmpl	%ebx,%ecx
-	jl	L0031st
+	jl	L0041st
 	pmuludq	%mm4,%mm0
 	pmuludq	%mm5,%mm1
 	paddq	%mm0,%mm2
@@ -99,7 +107,7 @@ L0031st:
 	paddq	%mm2,%mm3
 	movq	%mm3,32(%esp,%ebx,4)
 	incl	%edx
-L004outer:
+L005outer:
 	xorl	%ecx,%ecx
 	movd	(%edi,%edx,4),%mm4
 	movd	(%esi),%mm5
@@ -121,7 +129,7 @@ L004outer:
 	paddq	%mm6,%mm2
 	incl	%ecx
 	decl	%ebx
-L005inner:
+L006inner:
 	pmuludq	%mm4,%mm0
 	pmuludq	%mm5,%mm1
 	paddq	%mm0,%mm2
@@ -138,7 +146,7 @@ L005inner:
 	paddq	%mm6,%mm2
 	decl	%ebx
 	leal	1(%ecx),%ecx
-	jnz	L005inner
+	jnz	L006inner
 	movl	%ecx,%ebx
 	pmuludq	%mm4,%mm0
 	pmuludq	%mm5,%mm1
@@ -156,11 +164,11 @@ L005inner:
 	movq	%mm3,32(%esp,%ebx,4)
 	leal	1(%edx),%edx
 	cmpl	%ebx,%edx
-	jle	L004outer
+	jle	L005outer
 	emms
-	jmp	L006common_tail
+	jmp	L007common_tail
 .align	4,0x90
-L002non_sse2:
+L003non_sse2:
 	movl	8(%esp),%esi
 	leal	1(%ebx),%ebp
 	movl	12(%esp),%edi
@@ -171,12 +179,12 @@ L002non_sse2:
 	leal	4(%edi,%ebx,4),%eax
 	orl	%edx,%ebp
 	movl	(%edi),%edi
-	jz	L007bn_sqr_mont
+	jz	L008bn_sqr_mont
 	movl	%eax,28(%esp)
 	movl	(%esi),%eax
 	xorl	%edx,%edx
 .align	4,0x90
-L008mull:
+L009mull:
 	movl	%edx,%ebp
 	mull	%edi
 	addl	%eax,%ebp
@@ -185,7 +193,7 @@ L008mull:
 	movl	(%esi,%ecx,4),%eax
 	cmpl	%ebx,%ecx
 	movl	%ebp,28(%esp,%ecx,4)
-	jl	L008mull
+	jl	L009mull
 	movl	%edx,%ebp
 	mull	%edi
 	movl	20(%esp),%edi
@@ -203,9 +211,9 @@ L008mull:
 	movl	4(%esi),%eax
 	adcl	$0,%edx
 	incl	%ecx
-	jmp	L0092ndmadd
+	jmp	L0102ndmadd
 .align	4,0x90
-L0101stmadd:
+L0111stmadd:
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ecx,4),%ebp
@@ -216,7 +224,7 @@ L0101stmadd:
 	adcl	$0,%edx
 	cmpl	%ebx,%ecx
 	movl	%ebp,28(%esp,%ecx,4)
-	jl	L0101stmadd
+	jl	L0111stmadd
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ebx,4),%eax
@@ -239,7 +247,7 @@ L0101stmadd:
 	adcl	$0,%edx
 	movl	$1,%ecx
 .align	4,0x90
-L0092ndmadd:
+L0102ndmadd:
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ecx,4),%ebp
@@ -250,7 +258,7 @@ L0092ndmadd:
 	adcl	$0,%edx
 	cmpl	%ebx,%ecx
 	movl	%ebp,24(%esp,%ecx,4)
-	jl	L0092ndmadd
+	jl	L0102ndmadd
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ebx,4),%ebp
@@ -266,16 +274,16 @@ L0092ndmadd:
 	movl	%edx,32(%esp,%ebx,4)
 	cmpl	28(%esp),%ecx
 	movl	%eax,36(%esp,%ebx,4)
-	je	L006common_tail
+	je	L007common_tail
 	movl	(%ecx),%edi
 	movl	8(%esp),%esi
 	movl	%ecx,12(%esp)
 	xorl	%ecx,%ecx
 	xorl	%edx,%edx
 	movl	(%esi),%eax
-	jmp	L0101stmadd
+	jmp	L0111stmadd
 .align	4,0x90
-L007bn_sqr_mont:
+L008bn_sqr_mont:
 	movl	%ebx,(%esp)
 	movl	%ecx,12(%esp)
 	movl	%edi,%eax
@@ -286,7 +294,7 @@ L007bn_sqr_mont:
 	andl	$1,%ebx
 	incl	%ecx
 .align	4,0x90
-L011sqr:
+L012sqr:
 	movl	(%esi,%ecx,4),%eax
 	movl	%edx,%ebp
 	mull	%edi
@@ -298,7 +306,7 @@ L011sqr:
 	cmpl	(%esp),%ecx
 	movl	%eax,%ebx
 	movl	%ebp,28(%esp,%ecx,4)
-	jl	L011sqr
+	jl	L012sqr
 	movl	(%esi,%ecx,4),%eax
 	movl	%edx,%ebp
 	mull	%edi
@@ -322,7 +330,7 @@ L011sqr:
 	movl	4(%esi),%eax
 	movl	$1,%ecx
 .align	4,0x90
-L0123rdmadd:
+L0133rdmadd:
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ecx,4),%ebp
@@ -341,7 +349,7 @@ L0123rdmadd:
 	adcl	$0,%edx
 	cmpl	%ebx,%ecx
 	movl	%ebp,24(%esp,%ecx,4)
-	jl	L0123rdmadd
+	jl	L0133rdmadd
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ebx,4),%ebp
@@ -357,7 +365,7 @@ L0123rdmadd:
 	movl	%edx,32(%esp,%ebx,4)
 	cmpl	%ebx,%ecx
 	movl	%eax,36(%esp,%ebx,4)
-	je	L006common_tail
+	je	L007common_tail
 	movl	4(%esi,%ecx,4),%edi
 	leal	1(%ecx),%ecx
 	movl	%edi,%eax
@@ -369,12 +377,12 @@ L0123rdmadd:
 	xorl	%ebp,%ebp
 	cmpl	%ebx,%ecx
 	leal	1(%ecx),%ecx
-	je	L013sqrlast
+	je	L014sqrlast
 	movl	%edx,%ebx
 	shrl	$1,%edx
 	andl	$1,%ebx
 .align	4,0x90
-L014sqradd:
+L015sqradd:
 	movl	(%esi,%ecx,4),%eax
 	movl	%edx,%ebp
 	mull	%edi
@@ -390,13 +398,13 @@ L014sqradd:
 	cmpl	(%esp),%ecx
 	movl	%ebp,28(%esp,%ecx,4)
 	movl	%eax,%ebx
-	jle	L014sqradd
+	jle	L015sqradd
 	movl	%edx,%ebp
 	addl	%edx,%edx
 	shrl	$31,%ebp
 	addl	%ebx,%edx
 	adcl	$0,%ebp
-L013sqrlast:
+L014sqrlast:
 	movl	20(%esp),%edi
 	movl	16(%esp),%esi
 	imull	32(%esp),%edi
@@ -411,9 +419,9 @@ L013sqrlast:
 	adcl	$0,%edx
 	movl	$1,%ecx
 	movl	4(%esi),%eax
-	jmp	L0123rdmadd
+	jmp	L0133rdmadd
 .align	4,0x90
-L006common_tail:
+L007common_tail:
 	movl	16(%esp),%ebp
 	movl	4(%esp),%edi
 	leal	32(%esp),%esi
@@ -421,13 +429,13 @@ L006common_tail:
 	movl	%ebx,%ecx
 	xorl	%edx,%edx
 .align	4,0x90
-L015sub:
+L016sub:
 	sbbl	(%ebp,%edx,4),%eax
 	movl	%eax,(%edi,%edx,4)
 	decl	%ecx
 	movl	4(%esi,%edx,4),%eax
 	leal	1(%edx),%edx
-	jge	L015sub
+	jge	L016sub
 	sbbl	$0,%eax
 	andl	%eax,%esi
 	notl	%eax
@@ -435,12 +443,12 @@ L015sub:
 	andl	%eax,%ebp
 	orl	%ebp,%esi
 .align	4,0x90
-L016copy:
+L017copy:
 	movl	(%esi,%ebx,4),%eax
 	movl	%eax,(%edi,%ebx,4)
 	movl	%ecx,32(%esp,%ebx,4)
 	decl	%ebx
-	jge	L016copy
+	jge	L017copy
 	movl	24(%esp),%esp
 	movl	$1,%eax
 L000just_leave:

deps/openssl/asm/x86-win32-masm/bn/x86-mont.asm

@@ -45,6 +45,14 @@ $L_bn_mul_mont_begin::
 	xor	edx,2048
 	sub	esp,edx
 	and	esp,-64
+	mov	eax,ebp
+	sub	eax,esp
+	and	eax,-4096
+$L001page_walk:
+	mov	edx,DWORD PTR [eax*1+esp]
+	sub	eax,4096
+DB	46
+	jnc	$L001page_walk
 	mov	eax,DWORD PTR [esi]
 	mov	ebx,DWORD PTR 4[esi]
 	mov	ecx,DWORD PTR 8[esi]
@@ -60,7 +68,7 @@ $L_bn_mul_mont_begin::
 	mov	DWORD PTR 24[esp],ebp
 	lea	eax,DWORD PTR _OPENSSL_ia32cap_P
 	bt	DWORD PTR [eax],26
-	jnc	$L001non_sse2
+	jnc	$L002non_sse2
 	mov	eax,-1
 	movd	mm7,eax
 	mov	esi,DWORD PTR 8[esp]
@@ -84,7 +92,7 @@ $L_bn_mul_mont_begin::
 	psrlq	mm3,32
 	inc	ecx
 ALIGN	16
-$L0021st:
+$L0031st:
 	pmuludq	mm0,mm4
 	pmuludq	mm1,mm5
 	paddq	mm2,mm0
@@ -99,7 +107,7 @@ $L0021st:
 	psrlq	mm3,32
 	lea	ecx,DWORD PTR 1[ecx]
 	cmp	ecx,ebx
-	jl	$L0021st
+	jl	$L0031st
 	pmuludq	mm0,mm4
 	pmuludq	mm1,mm5
 	paddq	mm2,mm0
@@ -113,7 +121,7 @@ $L0021st:
 	paddq	mm3,mm2
 	movq	QWORD PTR 32[ebx*4+esp],mm3
 	inc	edx
-$L003outer:
+$L004outer:
 	xor	ecx,ecx
 	movd	mm4,DWORD PTR [edx*4+edi]
 	movd	mm5,DWORD PTR [esi]
@@ -135,7 +143,7 @@ $L003outer:
 	paddq	mm2,mm6
 	inc	ecx
 	dec	ebx
-$L004inner:
+$L005inner:
 	pmuludq	mm0,mm4
 	pmuludq	mm1,mm5
 	paddq	mm2,mm0
@@ -152,7 +160,7 @@ $L004inner:
 	paddq	mm2,mm6
 	dec	ebx
 	lea	ecx,DWORD PTR 1[ecx]
-	jnz	$L004inner
+	jnz	$L005inner
 	mov	ebx,ecx
 	pmuludq	mm0,mm4
 	pmuludq	mm1,mm5
@@ -170,11 +178,11 @@ $L004inner:
 	movq	QWORD PTR 32[ebx*4+esp],mm3
 	lea	edx,DWORD PTR 1[edx]
 	cmp	edx,ebx
-	jle	$L003outer
+	jle	$L004outer
 	emms
-	jmp	$L005common_tail
+	jmp	$L006common_tail
 ALIGN	16
-$L001non_sse2:
+$L002non_sse2:
 	mov	esi,DWORD PTR 8[esp]
 	lea	ebp,DWORD PTR 1[ebx]
 	mov	edi,DWORD PTR 12[esp]
@@ -185,12 +193,12 @@ $L001non_sse2:
 	lea	eax,DWORD PTR 4[ebx*4+edi]
 	or	ebp,edx
 	mov	edi,DWORD PTR [edi]
-	jz	$L006bn_sqr_mont
+	jz	$L007bn_sqr_mont
 	mov	DWORD PTR 28[esp],eax
 	mov	eax,DWORD PTR [esi]
 	xor	edx,edx
 ALIGN	16
-$L007mull:
+$L008mull:
 	mov	ebp,edx
 	mul	edi
 	add	ebp,eax
@@ -199,7 +207,7 @@ $L007mull:
 	mov	eax,DWORD PTR [ecx*4+esi]
 	cmp	ecx,ebx
 	mov	DWORD PTR 28[ecx*4+esp],ebp
-	jl	$L007mull
+	jl	$L008mull
 	mov	ebp,edx
 	mul	edi
 	mov	edi,DWORD PTR 20[esp]
@@ -217,9 +225,9 @@ $L007mull:
 	mov	eax,DWORD PTR 4[esi]
 	adc	edx,0
 	inc	ecx
-	jmp	$L0082ndmadd
+	jmp	$L0092ndmadd
 ALIGN	16
-$L0091stmadd:
+$L0101stmadd:
 	mov	ebp,edx
 	mul	edi
 	add	ebp,DWORD PTR 32[ecx*4+esp]
@@ -230,7 +238,7 @@ $L0091stmadd:
 	adc	edx,0
 	cmp	ecx,ebx
 	mov	DWORD PTR 28[ecx*4+esp],ebp
-	jl	$L0091stmadd
+	jl	$L0101stmadd
 	mov	ebp,edx
 	mul	edi
 	add	eax,DWORD PTR 32[ebx*4+esp]
@@ -253,7 +261,7 @@ $L0091stmadd:
 	adc	edx,0
 	mov	ecx,1
 ALIGN	16
-$L0082ndmadd:
+$L0092ndmadd:
 	mov	ebp,edx
 	mul	edi
 	add	ebp,DWORD PTR 32[ecx*4+esp]
@@ -264,7 +272,7 @@ $L0082ndmadd:
 	adc	edx,0
 	cmp	ecx,ebx
 	mov	DWORD PTR 24[ecx*4+esp],ebp
-	jl	$L0082ndmadd
+	jl	$L0092ndmadd
 	mov	ebp,edx
 	mul	edi
 	add	ebp,DWORD PTR 32[ebx*4+esp]
@@ -280,16 +288,16 @@ $L0082ndmadd:
 	mov	DWORD PTR 32[ebx*4+esp],edx
 	cmp	ecx,DWORD PTR 28[esp]
 	mov	DWORD PTR 36[ebx*4+esp],eax
-	je	$L005common_tail
+	je	$L006common_tail
 	mov	edi,DWORD PTR [ecx]
 	mov	esi,DWORD PTR 8[esp]
 	mov	DWORD PTR 12[esp],ecx
 	xor	ecx,ecx
 	xor	edx,edx
 	mov	eax,DWORD PTR [esi]
-	jmp	$L0091stmadd
+	jmp	$L0101stmadd
 ALIGN	16
-$L006bn_sqr_mont:
+$L007bn_sqr_mont:
 	mov	DWORD PTR [esp],ebx
 	mov	DWORD PTR 12[esp],ecx
 	mov	eax,edi
@@ -300,7 +308,7 @@ $L006bn_sqr_mont:
 	and	ebx,1
 	inc	ecx
 ALIGN	16
-$L010sqr:
+$L011sqr:
 	mov	eax,DWORD PTR [ecx*4+esi]
 	mov	ebp,edx
 	mul	edi
@@ -312,7 +320,7 @@ $L010sqr:
 	cmp	ecx,DWORD PTR [esp]
 	mov	ebx,eax
 	mov	DWORD PTR 28[ecx*4+esp],ebp
-	jl	$L010sqr
+	jl	$L011sqr
 	mov	eax,DWORD PTR [ecx*4+esi]
 	mov	ebp,edx
 	mul	edi
@@ -336,7 +344,7 @@ $L010sqr:
 	mov	eax,DWORD PTR 4[esi]
 	mov	ecx,1
 ALIGN	16
-$L0113rdmadd:
+$L0123rdmadd:
 	mov	ebp,edx
 	mul	edi
 	add	ebp,DWORD PTR 32[ecx*4+esp]
@@ -355,7 +363,7 @@ $L0113rdmadd:
 	adc	edx,0
 	cmp	ecx,ebx
 	mov	DWORD PTR 24[ecx*4+esp],ebp
-	jl	$L0113rdmadd
+	jl	$L0123rdmadd
 	mov	ebp,edx
 	mul	edi
 	add	ebp,DWORD PTR 32[ebx*4+esp]
@@ -371,7 +379,7 @@ $L0113rdmadd:
 	mov	DWORD PTR 32[ebx*4+esp],edx
 	cmp	ecx,ebx
 	mov	DWORD PTR 36[ebx*4+esp],eax
-	je	$L005common_tail
+	je	$L006common_tail
 	mov	edi,DWORD PTR 4[ecx*4+esi]
 	lea	ecx,DWORD PTR 1[ecx]
 	mov	eax,edi
@@ -383,12 +391,12 @@ $L0113rdmadd:
 	xor	ebp,ebp
 	cmp	ecx,ebx
 	lea	ecx,DWORD PTR 1[ecx]
-	je	$L012sqrlast
+	je	$L013sqrlast
 	mov	ebx,edx
 	shr	edx,1
 	and	ebx,1
 ALIGN	16
-$L013sqradd:
+$L014sqradd:
 	mov	eax,DWORD PTR [ecx*4+esi]
 	mov	ebp,edx
 	mul	edi
@@ -404,13 +412,13 @@ $L013sqradd:
 	cmp	ecx,DWORD PTR [esp]
 	mov	DWORD PTR 28[ecx*4+esp],ebp
 	mov	ebx,eax
-	jle	$L013sqradd
+	jle	$L014sqradd
 	mov	ebp,edx
 	add	edx,edx
 	shr	ebp,31
 	add	edx,ebx
 	adc	ebp,0
-$L012sqrlast:
+$L013sqrlast:
 	mov	edi,DWORD PTR 20[esp]
 	mov	esi,DWORD PTR 16[esp]
 	imul	edi,DWORD PTR 32[esp]
@@ -425,9 +433,9 @@ $L012sqrlast:
 	adc	edx,0
 	mov	ecx,1
 	mov	eax,DWORD PTR 4[esi]
-	jmp	$L0113rdmadd
+	jmp	$L0123rdmadd
 ALIGN	16
-$L005common_tail:
+$L006common_tail:
 	mov	ebp,DWORD PTR 16[esp]
 	mov	edi,DWORD PTR 4[esp]
 	lea	esi,DWORD PTR 32[esp]
@@ -435,13 +443,13 @@ $L005common_tail:
 	mov	ecx,ebx
 	xor	edx,edx
 ALIGN	16
-$L014sub:
+$L015sub:
 	sbb	eax,DWORD PTR [edx*4+ebp]
 	mov	DWORD PTR [edx*4+edi],eax
 	dec	ecx
 	mov	eax,DWORD PTR 4[edx*4+esi]
 	lea	edx,DWORD PTR 1[edx]
-	jge	$L014sub
+	jge	$L015sub
 	sbb	eax,0
 	and	esi,eax
 	not	eax
@@ -449,12 +457,12 @@ $L014sub:
 	and	ebp,eax
 	or	esi,ebp
 ALIGN	16
-$L015copy:
+$L016copy:
 	mov	eax,DWORD PTR [ebx*4+esi]
 	mov	DWORD PTR [ebx*4+edi],eax
 	mov	DWORD PTR 32[ebx*4+esp],ecx
 	dec	ebx
-	jge	$L015copy
+	jge	$L016copy
 	mov	esp,DWORD PTR 24[esp]
 	mov	eax,1
 $L000just_leave:

deps/openssl/asm_obsolete/x64-elf-gas/aes/aesni-sha1-x86_64.s

@@ -1392,8 +1392,8 @@ aesni_cbc_sha1_enc_shaext:
 	movups	16(%rcx),%xmm0
 	leaq	112(%rcx),%rcx
 
-	pshufd	$0b00011011,%xmm8,%xmm8
-	pshufd	$0b00011011,%xmm9,%xmm9
+	pshufd	$27,%xmm8,%xmm8
+	pshufd	$27,%xmm9,%xmm9
 	jmp	.Loop_shaext
 
 .align	16
@@ -1672,8 +1672,8 @@ aesni_cbc_sha1_enc_shaext:
 	leaq	64(%rdi),%rdi
 	jnz	.Loop_shaext
 
-	pshufd	$0b00011011,%xmm8,%xmm8
-	pshufd	$0b00011011,%xmm9,%xmm9
+	pshufd	$27,%xmm8,%xmm8
+	pshufd	$27,%xmm9,%xmm9
 	movups	%xmm2,(%r8)
 	movdqu	%xmm8,(%r9)
 	movd	%xmm9,16(%r9)

deps/openssl/asm_obsolete/x64-elf-gas/aes/aesni-x86_64.s

@@ -3462,11 +3462,11 @@ __aesni_set_encrypt_key:
 	movups	%xmm0,(%rax)
 	leaq	16(%rax),%rax
 .Lkey_expansion_128_cold:
-	shufps	$0b00010000,%xmm0,%xmm4
+	shufps	$16,%xmm0,%xmm4
 	xorps	%xmm4,%xmm0
-	shufps	$0b10001100,%xmm0,%xmm4
+	shufps	$140,%xmm0,%xmm4
 	xorps	%xmm4,%xmm0
-	shufps	$0b11111111,%xmm1,%xmm1
+	shufps	$255,%xmm1,%xmm1
 	xorps	%xmm1,%xmm0
 	.byte	0xf3,0xc3
 
@@ -3477,25 +3477,25 @@ __aesni_set_encrypt_key:
 .Lkey_expansion_192a_cold:
 	movaps	%xmm2,%xmm5
 .Lkey_expansion_192b_warm:
-	shufps	$0b00010000,%xmm0,%xmm4
+	shufps	$16,%xmm0,%xmm4
 	movdqa	%xmm2,%xmm3
 	xorps	%xmm4,%xmm0
-	shufps	$0b10001100,%xmm0,%xmm4
+	shufps	$140,%xmm0,%xmm4
 	pslldq	$4,%xmm3
 	xorps	%xmm4,%xmm0
-	pshufd	$0b01010101,%xmm1,%xmm1
+	pshufd	$85,%xmm1,%xmm1
 	pxor	%xmm3,%xmm2
 	pxor	%xmm1,%xmm0
-	pshufd	$0b11111111,%xmm0,%xmm3
+	pshufd	$255,%xmm0,%xmm3
 	pxor	%xmm3,%xmm2
 	.byte	0xf3,0xc3
 
 .align	16
 .Lkey_expansion_192b:
 	movaps	%xmm0,%xmm3
-	shufps	$0b01000100,%xmm0,%xmm5
+	shufps	$68,%xmm0,%xmm5
 	movups	%xmm5,(%rax)
-	shufps	$0b01001110,%xmm2,%xmm3
+	shufps	$78,%xmm2,%xmm3
 	movups	%xmm3,16(%rax)
 	leaq	32(%rax),%rax
 	jmp	.Lkey_expansion_192b_warm
@@ -3505,11 +3505,11 @@ __aesni_set_encrypt_key:
 	movups	%xmm2,(%rax)
 	leaq	16(%rax),%rax
 .Lkey_expansion_256a_cold:
-	shufps	$0b00010000,%xmm0,%xmm4
+	shufps	$16,%xmm0,%xmm4
 	xorps	%xmm4,%xmm0
-	shufps	$0b10001100,%xmm0,%xmm4
+	shufps	$140,%xmm0,%xmm4
 	xorps	%xmm4,%xmm0
-	shufps	$0b11111111,%xmm1,%xmm1
+	shufps	$255,%xmm1,%xmm1
 	xorps	%xmm1,%xmm0
 	.byte	0xf3,0xc3
 
@@ -3518,11 +3518,11 @@ __aesni_set_encrypt_key:
 	movups	%xmm0,(%rax)
 	leaq	16(%rax),%rax
 
-	shufps	$0b00010000,%xmm2,%xmm4
+	shufps	$16,%xmm2,%xmm4
 	xorps	%xmm4,%xmm2
-	shufps	$0b10001100,%xmm2,%xmm4
+	shufps	$140,%xmm2,%xmm4
 	xorps	%xmm4,%xmm2
-	shufps	$0b10101010,%xmm1,%xmm1
+	shufps	$170,%xmm1,%xmm1
 	xorps	%xmm1,%xmm2
 	.byte	0xf3,0xc3
 .size	aesni_set_encrypt_key,.-aesni_set_encrypt_key

deps/openssl/asm_obsolete/x64-elf-gas/bn/x86_64-mont.s

@@ -34,6 +34,20 @@ bn_mul_mont:
 
 	movq	%r11,8(%rsp,%r9,8)
 .Lmul_body:
+
+
+
+
+
+
+	subq	%rsp,%r11
+	andq	$-4096,%r11
+.Lmul_page_walk:
+	movq	(%rsp,%r11,1),%r10
+	subq	$4096,%r11
+.byte	0x66,0x2e
+	jnc	.Lmul_page_walk
+
 	movq	%rdx,%r12
 	movq	(%r8),%r8
 	movq	(%r12),%rbx
@@ -231,6 +245,14 @@ bn_mul4x_mont:
 
 	movq	%r11,8(%rsp,%r9,8)
 .Lmul4x_body:
+	subq	%rsp,%r11
+	andq	$-4096,%r11
+.Lmul4x_page_walk:
+	movq	(%rsp,%r11,1),%r10
+	subq	$4096,%r11
+.byte	0x2e
+	jnc	.Lmul4x_page_walk
+
 	movq	%rdi,16(%rsp,%r9,8)
 	movq	%rdx,%r12
 	movq	(%r8),%r8
@@ -653,6 +675,15 @@ bn_sqr8x_mont:
 	subq	%r11,%rsp
 .Lsqr8x_sp_done:
 	andq	$-64,%rsp
+	movq	%rax,%r11
+	subq	%rsp,%r11
+	andq	$-4096,%r11
+.Lsqr8x_page_walk:
+	movq	(%rsp,%r11,1),%r10
+	subq	$4096,%r11
+.byte	0x2e
+	jnc	.Lsqr8x_page_walk
+
 	movq	%r9,%r10
 	negq	%r9
 

deps/openssl/asm_obsolete/x64-elf-gas/bn/x86_64-mont5.s

@@ -30,6 +30,20 @@ bn_mul_mont_gather5:
 
 	movq	%rax,8(%rsp,%r9,8)
 .Lmul_body:
+
+
+
+
+
+
+	subq	%rsp,%rax
+	andq	$-4096,%rax
+.Lmul_page_walk:
+	movq	(%rsp,%rax,1),%r11
+	subq	$4096,%rax
+.byte	0x2e
+	jnc	.Lmul_page_walk
+
 	leaq	128(%rdx),%r12
 	movdqa	0(%r10),%xmm0
 	movdqa	16(%r10),%xmm1
@@ -442,6 +456,15 @@ bn_mul4x_mont_gather5:
 	subq	%r11,%rsp
 .Lmul4xsp_done:
 	andq	$-64,%rsp
+	movq	%rax,%r11
+	subq	%rsp,%r11
+	andq	$-4096,%r11
+.Lmul4x_page_walk:
+	movq	(%rsp,%r11,1),%r10
+	subq	$4096,%r11
+.byte	0x2e
+	jnc	.Lmul4x_page_walk
+
 	negq	%r9
 
 	movq	%rax,40(%rsp)
@@ -1031,6 +1054,15 @@ bn_power5:
 	subq	%r11,%rsp
 .Lpwr_sp_done:
 	andq	$-64,%rsp
+	movq	%rax,%r11
+	subq	%rsp,%r11
+	andq	$-4096,%r11
+.Lpwr_page_walk:
+	movq	(%rsp,%r11,1),%r10
+	subq	$4096,%r11
+.byte	0x2e
+	jnc	.Lpwr_page_walk
+
 	movq	%r9,%r10
 	negq	%r9
 
@@ -1972,6 +2004,15 @@ bn_from_mont8x:
 	subq	%r11,%rsp
 .Lfrom_sp_done:
 	andq	$-64,%rsp
+	movq	%rax,%r11
+	subq	%rsp,%r11
+	andq	$-4096,%r11
+.Lfrom_page_walk:
+	movq	(%rsp,%r11,1),%r10
+	subq	$4096,%r11
+.byte	0x2e
+	jnc	.Lfrom_page_walk
+
 	movq	%r9,%r10
 	negq	%r9
 

deps/openssl/asm_obsolete/x64-elf-gas/md5/md5-x86_64.s

@@ -493,14 +493,14 @@ md5_block_asm_data_order:
 	movl	%ecx,%r11d
 	addl	%ecx,%ebx
 	movl	0(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	xorl	%edx,%r11d
 	leal	-198630844(%rax,%r10,1),%eax
 	orl	%ebx,%r11d
 	xorl	%ecx,%r11d
 	addl	%r11d,%eax
 	movl	28(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$6,%eax
 	xorl	%ecx,%r11d
 	addl	%ebx,%eax
@@ -509,7 +509,7 @@ md5_block_asm_data_order:
 	xorl	%ebx,%r11d
 	addl	%r11d,%edx
 	movl	56(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$10,%edx
 	xorl	%ebx,%r11d
 	addl	%eax,%edx
@@ -518,7 +518,7 @@ md5_block_asm_data_order:
 	xorl	%eax,%r11d
 	addl	%r11d,%ecx
 	movl	20(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$15,%ecx
 	xorl	%eax,%r11d
 	addl	%edx,%ecx
@@ -527,7 +527,7 @@ md5_block_asm_data_order:
 	xorl	%edx,%r11d
 	addl	%r11d,%ebx
 	movl	48(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$21,%ebx
 	xorl	%edx,%r11d
 	addl	%ecx,%ebx
@@ -536,7 +536,7 @@ md5_block_asm_data_order:
 	xorl	%ecx,%r11d
 	addl	%r11d,%eax
 	movl	12(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$6,%eax
 	xorl	%ecx,%r11d
 	addl	%ebx,%eax
@@ -545,7 +545,7 @@ md5_block_asm_data_order:
 	xorl	%ebx,%r11d
 	addl	%r11d,%edx
 	movl	40(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$10,%edx
 	xorl	%ebx,%r11d
 	addl	%eax,%edx
@@ -554,7 +554,7 @@ md5_block_asm_data_order:
 	xorl	%eax,%r11d
 	addl	%r11d,%ecx
 	movl	4(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$15,%ecx
 	xorl	%eax,%r11d
 	addl	%edx,%ecx
@@ -563,7 +563,7 @@ md5_block_asm_data_order:
 	xorl	%edx,%r11d
 	addl	%r11d,%ebx
 	movl	32(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$21,%ebx
 	xorl	%edx,%r11d
 	addl	%ecx,%ebx
@@ -572,7 +572,7 @@ md5_block_asm_data_order:
 	xorl	%ecx,%r11d
 	addl	%r11d,%eax
 	movl	60(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$6,%eax
 	xorl	%ecx,%r11d
 	addl	%ebx,%eax
@@ -581,7 +581,7 @@ md5_block_asm_data_order:
 	xorl	%ebx,%r11d
 	addl	%r11d,%edx
 	movl	24(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$10,%edx
 	xorl	%ebx,%r11d
 	addl	%eax,%edx
@@ -590,7 +590,7 @@ md5_block_asm_data_order:
 	xorl	%eax,%r11d
 	addl	%r11d,%ecx
 	movl	52(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$15,%ecx
 	xorl	%eax,%r11d
 	addl	%edx,%ecx
@@ -599,7 +599,7 @@ md5_block_asm_data_order:
 	xorl	%edx,%r11d
 	addl	%r11d,%ebx
 	movl	16(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$21,%ebx
 	xorl	%edx,%r11d
 	addl	%ecx,%ebx
@@ -608,7 +608,7 @@ md5_block_asm_data_order:
 	xorl	%ecx,%r11d
 	addl	%r11d,%eax
 	movl	44(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$6,%eax
 	xorl	%ecx,%r11d
 	addl	%ebx,%eax
@@ -617,7 +617,7 @@ md5_block_asm_data_order:
 	xorl	%ebx,%r11d
 	addl	%r11d,%edx
 	movl	8(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$10,%edx
 	xorl	%ebx,%r11d
 	addl	%eax,%edx
@@ -626,7 +626,7 @@ md5_block_asm_data_order:
 	xorl	%eax,%r11d
 	addl	%r11d,%ecx
 	movl	36(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$15,%ecx
 	xorl	%eax,%r11d
 	addl	%edx,%ecx
@@ -635,7 +635,7 @@ md5_block_asm_data_order:
 	xorl	%edx,%r11d
 	addl	%r11d,%ebx
 	movl	0(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$21,%ebx
 	xorl	%edx,%r11d
 	addl	%ecx,%ebx

deps/openssl/asm_obsolete/x64-elf-gas/modes/ghash-x86_64.s

@@ -661,10 +661,10 @@ gcm_ghash_4bit:
 gcm_init_clmul:
 .L_init_clmul:
 	movdqu	(%rsi),%xmm2
-	pshufd	$0b01001110,%xmm2,%xmm2
+	pshufd	$78,%xmm2,%xmm2
 
 
-	pshufd	$0b11111111,%xmm2,%xmm4
+	pshufd	$255,%xmm2,%xmm4
 	movdqa	%xmm2,%xmm3
 	psllq	$1,%xmm2
 	pxor	%xmm5,%xmm5
@@ -678,11 +678,11 @@ gcm_init_clmul:
 	pxor	%xmm5,%xmm2
 
 
-	pshufd	$0b01001110,%xmm2,%xmm6
+	pshufd	$78,%xmm2,%xmm6
 	movdqa	%xmm2,%xmm0
 	pxor	%xmm2,%xmm6
 	movdqa	%xmm0,%xmm1
-	pshufd	$0b01001110,%xmm0,%xmm3
+	pshufd	$78,%xmm0,%xmm3
 	pxor	%xmm0,%xmm3
 .byte	102,15,58,68,194,0
 .byte	102,15,58,68,202,17
@@ -718,8 +718,8 @@ gcm_init_clmul:
 	pxor	%xmm4,%xmm0
 	psrlq	$1,%xmm0
 	pxor	%xmm1,%xmm0
-	pshufd	$0b01001110,%xmm2,%xmm3
-	pshufd	$0b01001110,%xmm0,%xmm4
+	pshufd	$78,%xmm2,%xmm3
+	pshufd	$78,%xmm0,%xmm4
 	pxor	%xmm2,%xmm3
 	movdqu	%xmm2,0(%rdi)
 	pxor	%xmm0,%xmm4
@@ -727,7 +727,7 @@ gcm_init_clmul:
 .byte	102,15,58,15,227,8
 	movdqu	%xmm4,32(%rdi)
 	movdqa	%xmm0,%xmm1
-	pshufd	$0b01001110,%xmm0,%xmm3
+	pshufd	$78,%xmm0,%xmm3
 	pxor	%xmm0,%xmm3
 .byte	102,15,58,68,194,0
 .byte	102,15,58,68,202,17
@@ -765,7 +765,7 @@ gcm_init_clmul:
 	pxor	%xmm1,%xmm0
 	movdqa	%xmm0,%xmm5
 	movdqa	%xmm0,%xmm1
-	pshufd	$0b01001110,%xmm0,%xmm3
+	pshufd	$78,%xmm0,%xmm3
 	pxor	%xmm0,%xmm3
 .byte	102,15,58,68,194,0
 .byte	102,15,58,68,202,17
@@ -801,8 +801,8 @@ gcm_init_clmul:
 	pxor	%xmm4,%xmm0
 	psrlq	$1,%xmm0
 	pxor	%xmm1,%xmm0
-	pshufd	$0b01001110,%xmm5,%xmm3
-	pshufd	$0b01001110,%xmm0,%xmm4
+	pshufd	$78,%xmm5,%xmm3
+	pshufd	$78,%xmm0,%xmm4
 	pxor	%xmm5,%xmm3
 	movdqu	%xmm5,48(%rdi)
 	pxor	%xmm0,%xmm4
@@ -822,7 +822,7 @@ gcm_gmult_clmul:
 	movdqu	32(%rsi),%xmm4
 .byte	102,15,56,0,197
 	movdqa	%xmm0,%xmm1
-	pshufd	$0b01001110,%xmm0,%xmm3
+	pshufd	$78,%xmm0,%xmm3
 	pxor	%xmm0,%xmm3
 .byte	102,15,58,68,194,0
 .byte	102,15,58,68,202,17
@@ -899,14 +899,14 @@ gcm_ghash_clmul:
 .byte	102,65,15,56,0,218
 .byte	102,69,15,56,0,218
 	movdqa	%xmm3,%xmm5
-	pshufd	$0b01001110,%xmm3,%xmm4
+	pshufd	$78,%xmm3,%xmm4
 	pxor	%xmm3,%xmm4
 .byte	102,15,58,68,218,0
 .byte	102,15,58,68,234,17
 .byte	102,15,58,68,231,0
 
 	movdqa	%xmm11,%xmm13
-	pshufd	$0b01001110,%xmm11,%xmm12
+	pshufd	$78,%xmm11,%xmm12
 	pxor	%xmm11,%xmm12
 .byte	102,68,15,58,68,222,0
 .byte	102,68,15,58,68,238,17
@@ -921,12 +921,12 @@ gcm_ghash_clmul:
 .byte	102,69,15,56,0,218
 .byte	102,69,15,56,0,194
 	movdqa	%xmm11,%xmm13
-	pshufd	$0b01001110,%xmm11,%xmm12
+	pshufd	$78,%xmm11,%xmm12
 	pxor	%xmm8,%xmm0
 	pxor	%xmm11,%xmm12
 .byte	102,69,15,58,68,222,0
 	movdqa	%xmm0,%xmm1
-	pshufd	$0b01001110,%xmm0,%xmm8
+	pshufd	$78,%xmm0,%xmm8
 	pxor	%xmm0,%xmm8
 .byte	102,69,15,58,68,238,17
 .byte	102,68,15,58,68,231,0
@@ -949,14 +949,14 @@ gcm_ghash_clmul:
 	movdqu	32(%rdx),%xmm3
 	movdqa	%xmm11,%xmm13
 .byte	102,68,15,58,68,199,16
-	pshufd	$0b01001110,%xmm11,%xmm12
+	pshufd	$78,%xmm11,%xmm12
 	xorps	%xmm5,%xmm1
 	pxor	%xmm11,%xmm12
 .byte	102,65,15,56,0,218
 	movups	32(%rsi),%xmm7
 	xorps	%xmm4,%xmm8
 .byte	102,68,15,58,68,218,0
-	pshufd	$0b01001110,%xmm3,%xmm4
+	pshufd	$78,%xmm3,%xmm4
 
 	pxor	%xmm0,%xmm8
 	movdqa	%xmm3,%xmm5
@@ -1000,7 +1000,7 @@ gcm_ghash_clmul:
 
 	movdqa	%xmm11,%xmm13
 	pxor	%xmm12,%xmm4
-	pshufd	$0b01001110,%xmm11,%xmm12
+	pshufd	$78,%xmm11,%xmm12
 	pxor	%xmm9,%xmm0
 	pxor	%xmm8,%xmm1
 	pxor	%xmm11,%xmm12
@@ -1010,7 +1010,7 @@ gcm_ghash_clmul:
 	movdqa	%xmm0,%xmm1
 .byte	102,69,15,58,68,238,17
 	xorps	%xmm11,%xmm3
-	pshufd	$0b01001110,%xmm0,%xmm8
+	pshufd	$78,%xmm0,%xmm8
 	pxor	%xmm0,%xmm8
 
 .byte	102,68,15,58,68,231,0
@@ -1079,7 +1079,7 @@ gcm_ghash_clmul:
 	pxor	%xmm8,%xmm0
 
 	movdqa	%xmm3,%xmm5
-	pshufd	$0b01001110,%xmm3,%xmm4
+	pshufd	$78,%xmm3,%xmm4
 	pxor	%xmm3,%xmm4
 .byte	102,15,58,68,218,0
 .byte	102,15,58,68,234,17
@@ -1096,7 +1096,7 @@ gcm_ghash_clmul:
 .Lmod_loop:
 	movdqa	%xmm0,%xmm1
 	movdqa	%xmm4,%xmm8
-	pshufd	$0b01001110,%xmm0,%xmm4
+	pshufd	$78,%xmm0,%xmm4
 	pxor	%xmm0,%xmm4
 
 .byte	102,15,58,68,198,0
@@ -1134,7 +1134,7 @@ gcm_ghash_clmul:
 	pslldq	$8,%xmm0
 	psrldq	$8,%xmm8
 	pxor	%xmm9,%xmm0
-	pshufd	$0b01001110,%xmm5,%xmm4
+	pshufd	$78,%xmm5,%xmm4
 	pxor	%xmm8,%xmm1
 	pxor	%xmm5,%xmm4
 
@@ -1156,7 +1156,7 @@ gcm_ghash_clmul:
 .Leven_tail:
 	movdqa	%xmm0,%xmm1
 	movdqa	%xmm4,%xmm8
-	pshufd	$0b01001110,%xmm0,%xmm4
+	pshufd	$78,%xmm0,%xmm4
 	pxor	%xmm0,%xmm4
 
 .byte	102,15,58,68,198,0
@@ -1204,7 +1204,7 @@ gcm_ghash_clmul:
 .byte	102,69,15,56,0,194
 	pxor	%xmm8,%xmm0
 	movdqa	%xmm0,%xmm1
-	pshufd	$0b01001110,%xmm0,%xmm3
+	pshufd	$78,%xmm0,%xmm3
 	pxor	%xmm0,%xmm3
 .byte	102,15,58,68,194,0
 .byte	102,15,58,68,202,17

deps/openssl/asm_obsolete/x64-elf-gas/sha/sha1-mb-x86_64.s

@@ -2599,10 +2599,10 @@ _shaext_shortcut:
 	punpcklqdq	%xmm5,%xmm0
 	punpckhqdq	%xmm5,%xmm8
 
-	pshufd	$0b00111111,%xmm7,%xmm1
-	pshufd	$0b01111111,%xmm7,%xmm9
-	pshufd	$0b00011011,%xmm0,%xmm0
-	pshufd	$0b00011011,%xmm8,%xmm8
+	pshufd	$63,%xmm7,%xmm1
+	pshufd	$127,%xmm7,%xmm9
+	pshufd	$27,%xmm0,%xmm0
+	pshufd	$27,%xmm8,%xmm8
 	jmp	.Loop_shaext
 
 .align	32
@@ -2888,8 +2888,8 @@ _shaext_shortcut:
 
 	movl	280(%rsp),%edx
 
-	pshufd	$0b00011011,%xmm0,%xmm0
-	pshufd	$0b00011011,%xmm8,%xmm8
+	pshufd	$27,%xmm0,%xmm0
+	pshufd	$27,%xmm8,%xmm8
 
 	movdqa	%xmm0,%xmm6
 	punpckldq	%xmm8,%xmm0

deps/openssl/asm_obsolete/x64-elf-gas/sha/sha1-x86_64.s

@@ -1240,9 +1240,9 @@ _shaext_shortcut:
 	movdqa	K_XX_XX+160(%rip),%xmm3
 
 	movdqu	(%rsi),%xmm4
-	pshufd	$0b00011011,%xmm0,%xmm0
+	pshufd	$27,%xmm0,%xmm0
 	movdqu	16(%rsi),%xmm5
-	pshufd	$0b00011011,%xmm1,%xmm1
+	pshufd	$27,%xmm1,%xmm1
 	movdqu	32(%rsi),%xmm6
 .byte	102,15,56,0,227
 	movdqu	48(%rsi),%xmm7
@@ -1392,8 +1392,8 @@ _shaext_shortcut:
 
 	jnz	.Loop_shaext
 
-	pshufd	$0b00011011,%xmm0,%xmm0
-	pshufd	$0b00011011,%xmm1,%xmm1
+	pshufd	$27,%xmm0,%xmm0
+	pshufd	$27,%xmm1,%xmm1
 	movdqu	%xmm0,(%rdi)
 	movd	%xmm1,16(%rdi)
 	.byte	0xf3,0xc3

deps/openssl/asm_obsolete/x64-elf-gas/sha/sha256-mb-x86_64.s

@@ -2677,10 +2677,10 @@ _shaext_shortcut:
 	punpckhqdq	%xmm8,%xmm14
 	punpckhqdq	%xmm10,%xmm15
 
-	pshufd	$0b00011011,%xmm12,%xmm12
-	pshufd	$0b00011011,%xmm13,%xmm13
-	pshufd	$0b00011011,%xmm14,%xmm14
-	pshufd	$0b00011011,%xmm15,%xmm15
+	pshufd	$27,%xmm12,%xmm12
+	pshufd	$27,%xmm13,%xmm13
+	pshufd	$27,%xmm14,%xmm14
+	pshufd	$27,%xmm15,%xmm15
 	jmp	.Loop_shaext
 
 .align	32
@@ -3066,10 +3066,10 @@ _shaext_shortcut:
 
 	movl	280(%rsp),%edx
 
-	pshufd	$0b00011011,%xmm12,%xmm12
-	pshufd	$0b00011011,%xmm13,%xmm13
-	pshufd	$0b00011011,%xmm14,%xmm14
-	pshufd	$0b00011011,%xmm15,%xmm15
+	pshufd	$27,%xmm12,%xmm12
+	pshufd	$27,%xmm13,%xmm13
+	pshufd	$27,%xmm14,%xmm14
+	pshufd	$27,%xmm15,%xmm15
 
 	movdqa	%xmm12,%xmm5
 	movdqa	%xmm13,%xmm6

deps/openssl/asm_obsolete/x64-macosx-gas/aes/aesni-sha1-x86_64.s

@@ -1392,8 +1392,8 @@ aesni_cbc_sha1_enc_shaext:
 	movups	16(%rcx),%xmm0
 	leaq	112(%rcx),%rcx
 
-	pshufd	$0b00011011,%xmm8,%xmm8
-	pshufd	$0b00011011,%xmm9,%xmm9
+	pshufd	$27,%xmm8,%xmm8
+	pshufd	$27,%xmm9,%xmm9
 	jmp	L$oop_shaext
 
 .p2align	4
@@ -1672,8 +1672,8 @@ L$aesenclast9:
 	leaq	64(%rdi),%rdi
 	jnz	L$oop_shaext
 
-	pshufd	$0b00011011,%xmm8,%xmm8
-	pshufd	$0b00011011,%xmm9,%xmm9
+	pshufd	$27,%xmm8,%xmm8
+	pshufd	$27,%xmm9,%xmm9
 	movups	%xmm2,(%r8)
 	movdqu	%xmm8,(%r9)
 	movd	%xmm9,16(%r9)

deps/openssl/asm_obsolete/x64-macosx-gas/aes/aesni-x86_64.s

@@ -3462,11 +3462,11 @@ L$key_expansion_128:
 	movups	%xmm0,(%rax)
 	leaq	16(%rax),%rax
 L$key_expansion_128_cold:
-	shufps	$0b00010000,%xmm0,%xmm4
+	shufps	$16,%xmm0,%xmm4
 	xorps	%xmm4,%xmm0
-	shufps	$0b10001100,%xmm0,%xmm4
+	shufps	$140,%xmm0,%xmm4
 	xorps	%xmm4,%xmm0
-	shufps	$0b11111111,%xmm1,%xmm1
+	shufps	$255,%xmm1,%xmm1
 	xorps	%xmm1,%xmm0
 	.byte	0xf3,0xc3
 
@@ -3477,25 +3477,25 @@ L$key_expansion_192a:
 L$key_expansion_192a_cold:
 	movaps	%xmm2,%xmm5
 L$key_expansion_192b_warm:
-	shufps	$0b00010000,%xmm0,%xmm4
+	shufps	$16,%xmm0,%xmm4
 	movdqa	%xmm2,%xmm3
 	xorps	%xmm4,%xmm0
-	shufps	$0b10001100,%xmm0,%xmm4
+	shufps	$140,%xmm0,%xmm4
 	pslldq	$4,%xmm3
 	xorps	%xmm4,%xmm0
-	pshufd	$0b01010101,%xmm1,%xmm1
+	pshufd	$85,%xmm1,%xmm1
 	pxor	%xmm3,%xmm2
 	pxor	%xmm1,%xmm0
-	pshufd	$0b11111111,%xmm0,%xmm3
+	pshufd	$255,%xmm0,%xmm3
 	pxor	%xmm3,%xmm2
 	.byte	0xf3,0xc3
 
 .p2align	4
 L$key_expansion_192b:
 	movaps	%xmm0,%xmm3
-	shufps	$0b01000100,%xmm0,%xmm5
+	shufps	$68,%xmm0,%xmm5
 	movups	%xmm5,(%rax)
-	shufps	$0b01001110,%xmm2,%xmm3
+	shufps	$78,%xmm2,%xmm3
 	movups	%xmm3,16(%rax)
 	leaq	32(%rax),%rax
 	jmp	L$key_expansion_192b_warm
@@ -3505,11 +3505,11 @@ L$key_expansion_256a:
 	movups	%xmm2,(%rax)
 	leaq	16(%rax),%rax
 L$key_expansion_256a_cold:
-	shufps	$0b00010000,%xmm0,%xmm4
+	shufps	$16,%xmm0,%xmm4
 	xorps	%xmm4,%xmm0
-	shufps	$0b10001100,%xmm0,%xmm4
+	shufps	$140,%xmm0,%xmm4
 	xorps	%xmm4,%xmm0
-	shufps	$0b11111111,%xmm1,%xmm1
+	shufps	$255,%xmm1,%xmm1
 	xorps	%xmm1,%xmm0
 	.byte	0xf3,0xc3
 
@@ -3518,11 +3518,11 @@ L$key_expansion_256b:
 	movups	%xmm0,(%rax)
 	leaq	16(%rax),%rax
 
-	shufps	$0b00010000,%xmm2,%xmm4
+	shufps	$16,%xmm2,%xmm4
 	xorps	%xmm4,%xmm2
-	shufps	$0b10001100,%xmm2,%xmm4
+	shufps	$140,%xmm2,%xmm4
 	xorps	%xmm4,%xmm2
-	shufps	$0b10101010,%xmm1,%xmm1
+	shufps	$170,%xmm1,%xmm1
 	xorps	%xmm1,%xmm2
 	.byte	0xf3,0xc3
 

deps/openssl/asm_obsolete/x64-macosx-gas/bn/x86_64-mont.s

@@ -34,6 +34,20 @@ L$mul_enter:
 
 	movq	%r11,8(%rsp,%r9,8)
 L$mul_body:
+
+
+
+
+
+
+	subq	%rsp,%r11
+	andq	$-4096,%r11
+L$mul_page_walk:
+	movq	(%rsp,%r11,1),%r10
+	subq	$4096,%r11
+.byte	0x66,0x2e
+	jnc	L$mul_page_walk
+
 	movq	%rdx,%r12
 	movq	(%r8),%r8
 	movq	(%r12),%rbx
@@ -231,6 +245,14 @@ L$mul4x_enter:
 
 	movq	%r11,8(%rsp,%r9,8)
 L$mul4x_body:
+	subq	%rsp,%r11
+	andq	$-4096,%r11
+L$mul4x_page_walk:
+	movq	(%rsp,%r11,1),%r10
+	subq	$4096,%r11
+.byte	0x2e
+	jnc	L$mul4x_page_walk
+
 	movq	%rdi,16(%rsp,%r9,8)
 	movq	%rdx,%r12
 	movq	(%r8),%r8
@@ -653,6 +675,15 @@ L$sqr8x_sp_alt:
 	subq	%r11,%rsp
 L$sqr8x_sp_done:
 	andq	$-64,%rsp
+	movq	%rax,%r11
+	subq	%rsp,%r11
+	andq	$-4096,%r11
+L$sqr8x_page_walk:
+	movq	(%rsp,%r11,1),%r10
+	subq	$4096,%r11
+.byte	0x2e
+	jnc	L$sqr8x_page_walk
+
 	movq	%r9,%r10
 	negq	%r9
 

deps/openssl/asm_obsolete/x64-macosx-gas/bn/x86_64-mont5.s

@@ -30,6 +30,20 @@ L$mul_enter:
 
 	movq	%rax,8(%rsp,%r9,8)
 L$mul_body:
+
+
+
+
+
+
+	subq	%rsp,%rax
+	andq	$-4096,%rax
+L$mul_page_walk:
+	movq	(%rsp,%rax,1),%r11
+	subq	$4096,%rax
+.byte	0x2e
+	jnc	L$mul_page_walk
+
 	leaq	128(%rdx),%r12
 	movdqa	0(%r10),%xmm0
 	movdqa	16(%r10),%xmm1
@@ -442,6 +456,15 @@ L$mul4xsp_alt:
 	subq	%r11,%rsp
 L$mul4xsp_done:
 	andq	$-64,%rsp
+	movq	%rax,%r11
+	subq	%rsp,%r11
+	andq	$-4096,%r11
+L$mul4x_page_walk:
+	movq	(%rsp,%r11,1),%r10
+	subq	$4096,%r11
+.byte	0x2e
+	jnc	L$mul4x_page_walk
+
 	negq	%r9
 
 	movq	%rax,40(%rsp)
@@ -1031,6 +1054,15 @@ L$pwr_sp_alt:
 	subq	%r11,%rsp
 L$pwr_sp_done:
 	andq	$-64,%rsp
+	movq	%rax,%r11
+	subq	%rsp,%r11
+	andq	$-4096,%r11
+L$pwr_page_walk:
+	movq	(%rsp,%r11,1),%r10
+	subq	$4096,%r11
+.byte	0x2e
+	jnc	L$pwr_page_walk
+
 	movq	%r9,%r10
 	negq	%r9
 
@@ -1972,6 +2004,15 @@ L$from_sp_alt:
 	subq	%r11,%rsp
 L$from_sp_done:
 	andq	$-64,%rsp
+	movq	%rax,%r11
+	subq	%rsp,%r11
+	andq	$-4096,%r11
+L$from_page_walk:
+	movq	(%rsp,%r11,1),%r10
+	subq	$4096,%r11
+.byte	0x2e
+	jnc	L$from_page_walk
+
 	movq	%r9,%r10
 	negq	%r9
 

deps/openssl/asm_obsolete/x64-macosx-gas/md5/md5-x86_64.s

@@ -493,14 +493,14 @@ L$loop:
 	movl	%ecx,%r11d
 	addl	%ecx,%ebx
 	movl	0(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	xorl	%edx,%r11d
 	leal	-198630844(%rax,%r10,1),%eax
 	orl	%ebx,%r11d
 	xorl	%ecx,%r11d
 	addl	%r11d,%eax
 	movl	28(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$6,%eax
 	xorl	%ecx,%r11d
 	addl	%ebx,%eax
@@ -509,7 +509,7 @@ L$loop:
 	xorl	%ebx,%r11d
 	addl	%r11d,%edx
 	movl	56(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$10,%edx
 	xorl	%ebx,%r11d
 	addl	%eax,%edx
@@ -518,7 +518,7 @@ L$loop:
 	xorl	%eax,%r11d
 	addl	%r11d,%ecx
 	movl	20(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$15,%ecx
 	xorl	%eax,%r11d
 	addl	%edx,%ecx
@@ -527,7 +527,7 @@ L$loop:
 	xorl	%edx,%r11d
 	addl	%r11d,%ebx
 	movl	48(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$21,%ebx
 	xorl	%edx,%r11d
 	addl	%ecx,%ebx
@@ -536,7 +536,7 @@ L$loop:
 	xorl	%ecx,%r11d
 	addl	%r11d,%eax
 	movl	12(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$6,%eax
 	xorl	%ecx,%r11d
 	addl	%ebx,%eax
@@ -545,7 +545,7 @@ L$loop:
 	xorl	%ebx,%r11d
 	addl	%r11d,%edx
 	movl	40(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$10,%edx
 	xorl	%ebx,%r11d
 	addl	%eax,%edx
@@ -554,7 +554,7 @@ L$loop:
 	xorl	%eax,%r11d
 	addl	%r11d,%ecx
 	movl	4(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$15,%ecx
 	xorl	%eax,%r11d
 	addl	%edx,%ecx
@@ -563,7 +563,7 @@ L$loop:
 	xorl	%edx,%r11d
 	addl	%r11d,%ebx
 	movl	32(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$21,%ebx
 	xorl	%edx,%r11d
 	addl	%ecx,%ebx
@@ -572,7 +572,7 @@ L$loop:
 	xorl	%ecx,%r11d
 	addl	%r11d,%eax
 	movl	60(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$6,%eax
 	xorl	%ecx,%r11d
 	addl	%ebx,%eax
@@ -581,7 +581,7 @@ L$loop:
 	xorl	%ebx,%r11d
 	addl	%r11d,%edx
 	movl	24(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$10,%edx
 	xorl	%ebx,%r11d
 	addl	%eax,%edx
@@ -590,7 +590,7 @@ L$loop:
 	xorl	%eax,%r11d
 	addl	%r11d,%ecx
 	movl	52(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$15,%ecx
 	xorl	%eax,%r11d
 	addl	%edx,%ecx
@@ -599,7 +599,7 @@ L$loop:
 	xorl	%edx,%r11d
 	addl	%r11d,%ebx
 	movl	16(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$21,%ebx
 	xorl	%edx,%r11d
 	addl	%ecx,%ebx
@@ -608,7 +608,7 @@ L$loop:
 	xorl	%ecx,%r11d
 	addl	%r11d,%eax
 	movl	44(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$6,%eax
 	xorl	%ecx,%r11d
 	addl	%ebx,%eax
@@ -617,7 +617,7 @@ L$loop:
 	xorl	%ebx,%r11d
 	addl	%r11d,%edx
 	movl	8(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$10,%edx
 	xorl	%ebx,%r11d
 	addl	%eax,%edx
@@ -626,7 +626,7 @@ L$loop:
 	xorl	%eax,%r11d
 	addl	%r11d,%ecx
 	movl	36(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$15,%ecx
 	xorl	%eax,%r11d
 	addl	%edx,%ecx
@@ -635,7 +635,7 @@ L$loop:
 	xorl	%edx,%r11d
 	addl	%r11d,%ebx
 	movl	0(%rsi),%r10d
-	movl	$4294967295,%r11d
+	movl	$0xffffffff,%r11d
 	roll	$21,%ebx
 	xorl	%edx,%r11d
 	addl	%ecx,%ebx

deps/openssl/asm_obsolete/x64-macosx-gas/modes/ghash-x86_64.s

@@ -661,10 +661,10 @@ L$ghash_epilogue:
 _gcm_init_clmul:
 L$_init_clmul:
 	movdqu	(%rsi),%xmm2
-	pshufd	$0b01001110,%xmm2,%xmm2
+	pshufd	$78,%xmm2,%xmm2
 
 
-	pshufd	$0b11111111,%xmm2,%xmm4
+	pshufd	$255,%xmm2,%xmm4
 	movdqa	%xmm2,%xmm3
 	psllq	$1,%xmm2
 	pxor	%xmm5,%xmm5
@@ -678,11 +678,11 @@ L$_init_clmul:
 	pxor	%xmm5,%xmm2
 
 
-	pshufd	$0b01001110,%xmm2,%xmm6
+	pshufd	$78,%xmm2,%xmm6
 	movdqa	%xmm2,%xmm0
 	pxor	%xmm2,%xmm6
 	movdqa	%xmm0,%xmm1
-	pshufd	$0b01001110,%xmm0,%xmm3
+	pshufd	$78,%xmm0,%xmm3
 	pxor	%xmm0,%xmm3
 .byte	102,15,58,68,194,0
 .byte	102,15,58,68,202,17
@@ -718,8 +718,8 @@ L$_init_clmul:
 	pxor	%xmm4,%xmm0
 	psrlq	$1,%xmm0
 	pxor	%xmm1,%xmm0
-	pshufd	$0b01001110,%xmm2,%xmm3
-	pshufd	$0b01001110,%xmm0,%xmm4
+	pshufd	$78,%xmm2,%xmm3
+	pshufd	$78,%xmm0,%xmm4
 	pxor	%xmm2,%xmm3
 	movdqu	%xmm2,0(%rdi)
 	pxor	%xmm0,%xmm4
@@ -727,7 +727,7 @@ L$_init_clmul:
 .byte	102,15,58,15,227,8
 	movdqu	%xmm4,32(%rdi)
 	movdqa	%xmm0,%xmm1
-	pshufd	$0b01001110,%xmm0,%xmm3
+	pshufd	$78,%xmm0,%xmm3
 	pxor	%xmm0,%xmm3
 .byte	102,15,58,68,194,0
 .byte	102,15,58,68,202,17
@@ -765,7 +765,7 @@ L$_init_clmul:
 	pxor	%xmm1,%xmm0
 	movdqa	%xmm0,%xmm5
 	movdqa	%xmm0,%xmm1
-	pshufd	$0b01001110,%xmm0,%xmm3
+	pshufd	$78,%xmm0,%xmm3
 	pxor	%xmm0,%xmm3
 .byte	102,15,58,68,194,0
 .byte	102,15,58,68,202,17
@@ -801,8 +801,8 @@ L$_init_clmul:
 	pxor	%xmm4,%xmm0
 	psrlq	$1,%xmm0
 	pxor	%xmm1,%xmm0
-	pshufd	$0b01001110,%xmm5,%xmm3
-	pshufd	$0b01001110,%xmm0,%xmm4
+	pshufd	$78,%xmm5,%xmm3
+	pshufd	$78,%xmm0,%xmm4
 	pxor	%xmm5,%xmm3
 	movdqu	%xmm5,48(%rdi)
 	pxor	%xmm0,%xmm4
@@ -822,7 +822,7 @@ L$_gmult_clmul:
 	movdqu	32(%rsi),%xmm4
 .byte	102,15,56,0,197
 	movdqa	%xmm0,%xmm1
-	pshufd	$0b01001110,%xmm0,%xmm3
+	pshufd	$78,%xmm0,%xmm3
 	pxor	%xmm0,%xmm3
 .byte	102,15,58,68,194,0
 .byte	102,15,58,68,202,17
@@ -899,14 +899,14 @@ L$_ghash_clmul:
 .byte	102,65,15,56,0,218
 .byte	102,69,15,56,0,218
 	movdqa	%xmm3,%xmm5
-	pshufd	$0b01001110,%xmm3,%xmm4
+	pshufd	$78,%xmm3,%xmm4
 	pxor	%xmm3,%xmm4
 .byte	102,15,58,68,218,0
 .byte	102,15,58,68,234,17
 .byte	102,15,58,68,231,0
 
 	movdqa	%xmm11,%xmm13
-	pshufd	$0b01001110,%xmm11,%xmm12
+	pshufd	$78,%xmm11,%xmm12
 	pxor	%xmm11,%xmm12
 .byte	102,68,15,58,68,222,0
 .byte	102,68,15,58,68,238,17
@@ -921,12 +921,12 @@ L$_ghash_clmul:
 .byte	102,69,15,56,0,218
 .byte	102,69,15,56,0,194
 	movdqa	%xmm11,%xmm13
-	pshufd	$0b01001110,%xmm11,%xmm12
+	pshufd	$78,%xmm11,%xmm12
 	pxor	%xmm8,%xmm0
 	pxor	%xmm11,%xmm12
 .byte	102,69,15,58,68,222,0
 	movdqa	%xmm0,%xmm1
-	pshufd	$0b01001110,%xmm0,%xmm8
+	pshufd	$78,%xmm0,%xmm8
 	pxor	%xmm0,%xmm8
 .byte	102,69,15,58,68,238,17
 .byte	102,68,15,58,68,231,0
@@ -949,14 +949,14 @@ L$mod4_loop:
 	movdqu	32(%rdx),%xmm3
 	movdqa	%xmm11,%xmm13
 .byte	102,68,15,58,68,199,16
-	pshufd	$0b01001110,%xmm11,%xmm12
+	pshufd	$78,%xmm11,%xmm12
 	xorps	%xmm5,%xmm1
 	pxor	%xmm11,%xmm12
 .byte	102,65,15,56,0,218
 	movups	32(%rsi),%xmm7
 	xorps	%xmm4,%xmm8
 .byte	102,68,15,58,68,218,0
-	pshufd	$0b01001110,%xmm3,%xmm4
+	pshufd	$78,%xmm3,%xmm4
 
 	pxor	%xmm0,%xmm8
 	movdqa	%xmm3,%xmm5
@@ -1000,7 +1000,7 @@ L$mod4_loop:
 
 	movdqa	%xmm11,%xmm13
 	pxor	%xmm12,%xmm4
-	pshufd	$0b01001110,%xmm11,%xmm12
+	pshufd	$78,%xmm11,%xmm12
 	pxor	%xmm9,%xmm0
 	pxor	%xmm8,%xmm1
 	pxor	%xmm11,%xmm12
@@ -1010,7 +1010,7 @@ L$mod4_loop:
 	movdqa	%xmm0,%xmm1
 .byte	102,69,15,58,68,238,17
 	xorps	%xmm11,%xmm3
-	pshufd	$0b01001110,%xmm0,%xmm8
+	pshufd	$78,%xmm0,%xmm8
 	pxor	%xmm0,%xmm8
 
 .byte	102,68,15,58,68,231,0
@@ -1079,7 +1079,7 @@ L$skip4x:
 	pxor	%xmm8,%xmm0
 
 	movdqa	%xmm3,%xmm5
-	pshufd	$0b01001110,%xmm3,%xmm4
+	pshufd	$78,%xmm3,%xmm4
 	pxor	%xmm3,%xmm4
 .byte	102,15,58,68,218,0
 .byte	102,15,58,68,234,17
@@ -1096,7 +1096,7 @@ L$skip4x:
 L$mod_loop:
 	movdqa	%xmm0,%xmm1
 	movdqa	%xmm4,%xmm8
-	pshufd	$0b01001110,%xmm0,%xmm4
+	pshufd	$78,%xmm0,%xmm4
 	pxor	%xmm0,%xmm4
 
 .byte	102,15,58,68,198,0
@@ -1134,7 +1134,7 @@ L$mod_loop:
 	pslldq	$8,%xmm0
 	psrldq	$8,%xmm8
 	pxor	%xmm9,%xmm0
-	pshufd	$0b01001110,%xmm5,%xmm4
+	pshufd	$78,%xmm5,%xmm4
 	pxor	%xmm8,%xmm1
 	pxor	%xmm5,%xmm4
 
@@ -1156,7 +1156,7 @@ L$mod_loop:
 L$even_tail:
 	movdqa	%xmm0,%xmm1
 	movdqa	%xmm4,%xmm8
-	pshufd	$0b01001110,%xmm0,%xmm4
+	pshufd	$78,%xmm0,%xmm4
 	pxor	%xmm0,%xmm4
 
 .byte	102,15,58,68,198,0
@@ -1204,7 +1204,7 @@ L$odd_tail:
 .byte	102,69,15,56,0,194
 	pxor	%xmm8,%xmm0
 	movdqa	%xmm0,%xmm1
-	pshufd	$0b01001110,%xmm0,%xmm3
+	pshufd	$78,%xmm0,%xmm3
 	pxor	%xmm0,%xmm3
 .byte	102,15,58,68,194,0
 .byte	102,15,58,68,202,17

deps/openssl/asm_obsolete/x64-macosx-gas/sha/sha1-mb-x86_64.s

@@ -2599,10 +2599,10 @@ L$oop_grande_shaext:
 	punpcklqdq	%xmm5,%xmm0
 	punpckhqdq	%xmm5,%xmm8
 
-	pshufd	$0b00111111,%xmm7,%xmm1
-	pshufd	$0b01111111,%xmm7,%xmm9
-	pshufd	$0b00011011,%xmm0,%xmm0
-	pshufd	$0b00011011,%xmm8,%xmm8
+	pshufd	$63,%xmm7,%xmm1
+	pshufd	$127,%xmm7,%xmm9
+	pshufd	$27,%xmm0,%xmm0
+	pshufd	$27,%xmm8,%xmm8
 	jmp	L$oop_shaext
 
 .p2align	5
@@ -2888,8 +2888,8 @@ L$oop_shaext:
 
 	movl	280(%rsp),%edx
 
-	pshufd	$0b00011011,%xmm0,%xmm0
-	pshufd	$0b00011011,%xmm8,%xmm8
+	pshufd	$27,%xmm0,%xmm0
+	pshufd	$27,%xmm8,%xmm8
 
 	movdqa	%xmm0,%xmm6
 	punpckldq	%xmm8,%xmm0

deps/openssl/asm_obsolete/x64-macosx-gas/sha/sha1-x86_64.s

@@ -1240,9 +1240,9 @@ _shaext_shortcut:
 	movdqa	K_XX_XX+160(%rip),%xmm3
 
 	movdqu	(%rsi),%xmm4
-	pshufd	$0b00011011,%xmm0,%xmm0
+	pshufd	$27,%xmm0,%xmm0
 	movdqu	16(%rsi),%xmm5
-	pshufd	$0b00011011,%xmm1,%xmm1
+	pshufd	$27,%xmm1,%xmm1
 	movdqu	32(%rsi),%xmm6
 .byte	102,15,56,0,227
 	movdqu	48(%rsi),%xmm7
@@ -1392,8 +1392,8 @@ L$oop_shaext:
 
 	jnz	L$oop_shaext
 
-	pshufd	$0b00011011,%xmm0,%xmm0
-	pshufd	$0b00011011,%xmm1,%xmm1
+	pshufd	$27,%xmm0,%xmm0
+	pshufd	$27,%xmm1,%xmm1
 	movdqu	%xmm0,(%rdi)
 	movd	%xmm1,16(%rdi)
 	.byte	0xf3,0xc3

deps/openssl/asm_obsolete/x64-macosx-gas/sha/sha256-mb-x86_64.s

@@ -2677,10 +2677,10 @@ L$oop_grande_shaext:
 	punpckhqdq	%xmm8,%xmm14
 	punpckhqdq	%xmm10,%xmm15
 
-	pshufd	$0b00011011,%xmm12,%xmm12
-	pshufd	$0b00011011,%xmm13,%xmm13
-	pshufd	$0b00011011,%xmm14,%xmm14
-	pshufd	$0b00011011,%xmm15,%xmm15
+	pshufd	$27,%xmm12,%xmm12
+	pshufd	$27,%xmm13,%xmm13
+	pshufd	$27,%xmm14,%xmm14
+	pshufd	$27,%xmm15,%xmm15
 	jmp	L$oop_shaext
 
 .p2align	5
@@ -3066,10 +3066,10 @@ L$oop_shaext:
 
 	movl	280(%rsp),%edx
 
-	pshufd	$0b00011011,%xmm12,%xmm12
-	pshufd	$0b00011011,%xmm13,%xmm13
-	pshufd	$0b00011011,%xmm14,%xmm14
-	pshufd	$0b00011011,%xmm15,%xmm15
+	pshufd	$27,%xmm12,%xmm12
+	pshufd	$27,%xmm13,%xmm13
+	pshufd	$27,%xmm14,%xmm14
+	pshufd	$27,%xmm15,%xmm15
 
 	movdqa	%xmm12,%xmm5
 	movdqa	%xmm13,%xmm6

deps/openssl/asm_obsolete/x64-win32-masm/bn/x86_64-mont.asm

@@ -47,6 +47,20 @@ $L$mul_enter::
 
 	mov	QWORD PTR[8+r9*8+rsp],r11
 $L$mul_body::
+
+
+
+
+
+
+	sub	r11,rsp
+	and	r11,-4096
+$L$mul_page_walk::
+	mov	r10,QWORD PTR[r11*1+rsp]
+	sub	r11,4096
+DB	066h,02eh
+	jnc	$L$mul_page_walk
+
 	mov	r12,rdx
 	mov	r8,QWORD PTR[r8]
 	mov	rbx,QWORD PTR[r12]
@@ -259,6 +273,14 @@ $L$mul4x_enter::
 
 	mov	QWORD PTR[8+r9*8+rsp],r11
 $L$mul4x_body::
+	sub	r11,rsp
+	and	r11,-4096
+$L$mul4x_page_walk::
+	mov	r10,QWORD PTR[r11*1+rsp]
+	sub	r11,4096
+DB	02eh
+	jnc	$L$mul4x_page_walk
+
 	mov	QWORD PTR[16+r9*8+rsp],rdi
 	mov	r12,rdx
 	mov	r8,QWORD PTR[r8]
@@ -696,6 +718,15 @@ $L$sqr8x_sp_alt::
 	sub	rsp,r11
 $L$sqr8x_sp_done::
 	and	rsp,-64
+	mov	r11,rax
+	sub	r11,rsp
+	and	r11,-4096
+$L$sqr8x_page_walk::
+	mov	r10,QWORD PTR[r11*1+rsp]
+	sub	r11,4096
+DB	02eh
+	jnc	$L$sqr8x_page_walk
+
 	mov	r10,r9
 	neg	r9
 

deps/openssl/asm_obsolete/x64-win32-masm/bn/x86_64-mont5.asm

@@ -43,6 +43,20 @@ $L$mul_enter::
 
 	mov	QWORD PTR[8+r9*8+rsp],rax
 $L$mul_body::
+
+
+
+
+
+
+	sub	rax,rsp
+	and	rax,-4096
+$L$mul_page_walk::
+	mov	r11,QWORD PTR[rax*1+rsp]
+	sub	rax,4096
+DB	02eh
+	jnc	$L$mul_page_walk
+
 	lea	r12,QWORD PTR[128+rdx]
 	movdqa	xmm0,XMMWORD PTR[r10]
 	movdqa	xmm1,XMMWORD PTR[16+r10]
@@ -470,6 +484,15 @@ $L$mul4xsp_alt::
 	sub	rsp,r11
 $L$mul4xsp_done::
 	and	rsp,-64
+	mov	r11,rax
+	sub	r11,rsp
+	and	r11,-4096
+$L$mul4x_page_walk::
+	mov	r10,QWORD PTR[r11*1+rsp]
+	sub	r11,4096
+DB	02eh
+	jnc	$L$mul4x_page_walk
+
 	neg	r9
 
 	mov	QWORD PTR[40+rsp],rax
@@ -1074,6 +1097,15 @@ $L$pwr_sp_alt::
 	sub	rsp,r11
 $L$pwr_sp_done::
 	and	rsp,-64
+	mov	r11,rax
+	sub	r11,rsp
+	and	r11,-4096
+$L$pwr_page_walk::
+	mov	r10,QWORD PTR[r11*1+rsp]
+	sub	r11,4096
+DB	02eh
+	jnc	$L$pwr_page_walk
+
 	mov	r10,r9
 	neg	r9
 
@@ -2030,6 +2062,15 @@ $L$from_sp_alt::
 	sub	rsp,r11
 $L$from_sp_done::
 	and	rsp,-64
+	mov	r11,rax
+	sub	r11,rsp
+	and	r11,-4096
+$L$from_page_walk::
+	mov	r10,QWORD PTR[r11*1+rsp]
+	sub	r11,4096
+DB	02eh
+	jnc	$L$from_page_walk
+
 	mov	r10,r9
 	neg	r9
 

deps/openssl/asm_obsolete/x86-elf-gas/bn/x86-mont.s

@@ -29,6 +29,14 @@ bn_mul_mont:
 	xorl	$2048,%edx
 	subl	%edx,%esp
 	andl	$-64,%esp
+	movl	%ebp,%eax
+	subl	%esp,%eax
+	andl	$-4096,%eax
+.L001page_walk:
+	movl	(%esp,%eax,1),%edx
+	subl	$4096,%eax
+.byte	46
+	jnc	.L001page_walk
 	movl	(%esi),%eax
 	movl	4(%esi),%ebx
 	movl	8(%esi),%ecx
@@ -44,7 +52,7 @@ bn_mul_mont:
 	movl	%ebp,24(%esp)
 	leal	OPENSSL_ia32cap_P,%eax
 	btl	$26,(%eax)
-	jnc	.L001non_sse2
+	jnc	.L002non_sse2
 	movl	$-1,%eax
 	movd	%eax,%mm7
 	movl	8(%esp),%esi
@@ -68,7 +76,7 @@ bn_mul_mont:
 	psrlq	$32,%mm3
 	incl	%ecx
 .align	16
-.L0021st:
+.L0031st:
 	pmuludq	%mm4,%mm0
 	pmuludq	%mm5,%mm1
 	paddq	%mm0,%mm2
@@ -83,7 +91,7 @@ bn_mul_mont:
 	psrlq	$32,%mm3
 	leal	1(%ecx),%ecx
 	cmpl	%ebx,%ecx
-	jl	.L0021st
+	jl	.L0031st
 	pmuludq	%mm4,%mm0
 	pmuludq	%mm5,%mm1
 	paddq	%mm0,%mm2
@@ -97,7 +105,7 @@ bn_mul_mont:
 	paddq	%mm2,%mm3
 	movq	%mm3,32(%esp,%ebx,4)
 	incl	%edx
-.L003outer:
+.L004outer:
 	xorl	%ecx,%ecx
 	movd	(%edi,%edx,4),%mm4
 	movd	(%esi),%mm5
@@ -119,7 +127,7 @@ bn_mul_mont:
 	paddq	%mm6,%mm2
 	incl	%ecx
 	decl	%ebx
-.L004inner:
+.L005inner:
 	pmuludq	%mm4,%mm0
 	pmuludq	%mm5,%mm1
 	paddq	%mm0,%mm2
@@ -136,7 +144,7 @@ bn_mul_mont:
 	paddq	%mm6,%mm2
 	decl	%ebx
 	leal	1(%ecx),%ecx
-	jnz	.L004inner
+	jnz	.L005inner
 	movl	%ecx,%ebx
 	pmuludq	%mm4,%mm0
 	pmuludq	%mm5,%mm1
@@ -154,11 +162,11 @@ bn_mul_mont:
 	movq	%mm3,32(%esp,%ebx,4)
 	leal	1(%edx),%edx
 	cmpl	%ebx,%edx
-	jle	.L003outer
+	jle	.L004outer
 	emms
-	jmp	.L005common_tail
+	jmp	.L006common_tail
 .align	16
-.L001non_sse2:
+.L002non_sse2:
 	movl	8(%esp),%esi
 	leal	1(%ebx),%ebp
 	movl	12(%esp),%edi
@@ -169,12 +177,12 @@ bn_mul_mont:
 	leal	4(%edi,%ebx,4),%eax
 	orl	%edx,%ebp
 	movl	(%edi),%edi
-	jz	.L006bn_sqr_mont
+	jz	.L007bn_sqr_mont
 	movl	%eax,28(%esp)
 	movl	(%esi),%eax
 	xorl	%edx,%edx
 .align	16
-.L007mull:
+.L008mull:
 	movl	%edx,%ebp
 	mull	%edi
 	addl	%eax,%ebp
@@ -183,7 +191,7 @@ bn_mul_mont:
 	movl	(%esi,%ecx,4),%eax
 	cmpl	%ebx,%ecx
 	movl	%ebp,28(%esp,%ecx,4)
-	jl	.L007mull
+	jl	.L008mull
 	movl	%edx,%ebp
 	mull	%edi
 	movl	20(%esp),%edi
@@ -201,9 +209,9 @@ bn_mul_mont:
 	movl	4(%esi),%eax
 	adcl	$0,%edx
 	incl	%ecx
-	jmp	.L0082ndmadd
+	jmp	.L0092ndmadd
 .align	16
-.L0091stmadd:
+.L0101stmadd:
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ecx,4),%ebp
@@ -214,7 +222,7 @@ bn_mul_mont:
 	adcl	$0,%edx
 	cmpl	%ebx,%ecx
 	movl	%ebp,28(%esp,%ecx,4)
-	jl	.L0091stmadd
+	jl	.L0101stmadd
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ebx,4),%eax
@@ -237,7 +245,7 @@ bn_mul_mont:
 	adcl	$0,%edx
 	movl	$1,%ecx
 .align	16
-.L0082ndmadd:
+.L0092ndmadd:
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ecx,4),%ebp
@@ -248,7 +256,7 @@ bn_mul_mont:
 	adcl	$0,%edx
 	cmpl	%ebx,%ecx
 	movl	%ebp,24(%esp,%ecx,4)
-	jl	.L0082ndmadd
+	jl	.L0092ndmadd
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ebx,4),%ebp
@@ -264,16 +272,16 @@ bn_mul_mont:
 	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	.L0091stmadd
+	jmp	.L0101stmadd
 .align	16
-.L006bn_sqr_mont:
+.L007bn_sqr_mont:
 	movl	%ebx,(%esp)
 	movl	%ecx,12(%esp)
 	movl	%edi,%eax
@@ -284,7 +292,7 @@ bn_mul_mont:
 	andl	$1,%ebx
 	incl	%ecx
 .align	16
-.L010sqr:
+.L011sqr:
 	movl	(%esi,%ecx,4),%eax
 	movl	%edx,%ebp
 	mull	%edi
@@ -296,7 +304,7 @@ bn_mul_mont:
 	cmpl	(%esp),%ecx
 	movl	%eax,%ebx
 	movl	%ebp,28(%esp,%ecx,4)
-	jl	.L010sqr
+	jl	.L011sqr
 	movl	(%esi,%ecx,4),%eax
 	movl	%edx,%ebp
 	mull	%edi
@@ -320,7 +328,7 @@ bn_mul_mont:
 	movl	4(%esi),%eax
 	movl	$1,%ecx
 .align	16
-.L0113rdmadd:
+.L0123rdmadd:
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ecx,4),%ebp
@@ -339,7 +347,7 @@ bn_mul_mont:
 	adcl	$0,%edx
 	cmpl	%ebx,%ecx
 	movl	%ebp,24(%esp,%ecx,4)
-	jl	.L0113rdmadd
+	jl	.L0123rdmadd
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ebx,4),%ebp
@@ -355,7 +363,7 @@ bn_mul_mont:
 	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
@@ -367,12 +375,12 @@ bn_mul_mont:
 	xorl	%ebp,%ebp
 	cmpl	%ebx,%ecx
 	leal	1(%ecx),%ecx
-	je	.L012sqrlast
+	je	.L013sqrlast
 	movl	%edx,%ebx
 	shrl	$1,%edx
 	andl	$1,%ebx
 .align	16
-.L013sqradd:
+.L014sqradd:
 	movl	(%esi,%ecx,4),%eax
 	movl	%edx,%ebp
 	mull	%edi
@@ -388,13 +396,13 @@ bn_mul_mont:
 	cmpl	(%esp),%ecx
 	movl	%ebp,28(%esp,%ecx,4)
 	movl	%eax,%ebx
-	jle	.L013sqradd
+	jle	.L014sqradd
 	movl	%edx,%ebp
 	addl	%edx,%edx
 	shrl	$31,%ebp
 	addl	%ebx,%edx
 	adcl	$0,%ebp
-.L012sqrlast:
+.L013sqrlast:
 	movl	20(%esp),%edi
 	movl	16(%esp),%esi
 	imull	32(%esp),%edi
@@ -409,9 +417,9 @@ bn_mul_mont:
 	adcl	$0,%edx
 	movl	$1,%ecx
 	movl	4(%esi),%eax
-	jmp	.L0113rdmadd
+	jmp	.L0123rdmadd
 .align	16
-.L005common_tail:
+.L006common_tail:
 	movl	16(%esp),%ebp
 	movl	4(%esp),%edi
 	leal	32(%esp),%esi
@@ -419,13 +427,13 @@ bn_mul_mont:
 	movl	%ebx,%ecx
 	xorl	%edx,%edx
 .align	16
-.L014sub:
+.L015sub:
 	sbbl	(%ebp,%edx,4),%eax
 	movl	%eax,(%edi,%edx,4)
 	decl	%ecx
 	movl	4(%esi,%edx,4),%eax
 	leal	1(%edx),%edx
-	jge	.L014sub
+	jge	.L015sub
 	sbbl	$0,%eax
 	andl	%eax,%esi
 	notl	%eax
@@ -433,12 +441,12 @@ bn_mul_mont:
 	andl	%eax,%ebp
 	orl	%ebp,%esi
 .align	16
-.L015copy:
+.L016copy:
 	movl	(%esi,%ebx,4),%eax
 	movl	%eax,(%edi,%ebx,4)
 	movl	%ecx,32(%esp,%ebx,4)
 	decl	%ebx
-	jge	.L015copy
+	jge	.L016copy
 	movl	24(%esp),%esp
 	movl	$1,%eax
 .L000just_leave:

deps/openssl/asm_obsolete/x86-macosx-gas/bn/x86-mont.s

@@ -28,6 +28,14 @@ L_bn_mul_mont_begin:
 	xorl	$2048,%edx
 	subl	%edx,%esp
 	andl	$-64,%esp
+	movl	%ebp,%eax
+	subl	%esp,%eax
+	andl	$-4096,%eax
+L001page_walk:
+	movl	(%esp,%eax,1),%edx
+	subl	$4096,%eax
+.byte	46
+	jnc	L001page_walk
 	movl	(%esi),%eax
 	movl	4(%esi),%ebx
 	movl	8(%esi),%ecx
@@ -41,12 +49,12 @@ L_bn_mul_mont_begin:
 	movl	%esi,20(%esp)
 	leal	-3(%edi),%ebx
 	movl	%ebp,24(%esp)
-	call	L001PIC_me_up
-L001PIC_me_up:
+	call	L002PIC_me_up
+L002PIC_me_up:
 	popl	%eax
-	movl	L_OPENSSL_ia32cap_P$non_lazy_ptr-L001PIC_me_up(%eax),%eax
+	movl	L_OPENSSL_ia32cap_P$non_lazy_ptr-L002PIC_me_up(%eax),%eax
 	btl	$26,(%eax)
-	jnc	L002non_sse2
+	jnc	L003non_sse2
 	movl	$-1,%eax
 	movd	%eax,%mm7
 	movl	8(%esp),%esi
@@ -70,7 +78,7 @@ L001PIC_me_up:
 	psrlq	$32,%mm3
 	incl	%ecx
 .align	4,0x90
-L0031st:
+L0041st:
 	pmuludq	%mm4,%mm0
 	pmuludq	%mm5,%mm1
 	paddq	%mm0,%mm2
@@ -85,7 +93,7 @@ L0031st:
 	psrlq	$32,%mm3
 	leal	1(%ecx),%ecx
 	cmpl	%ebx,%ecx
-	jl	L0031st
+	jl	L0041st
 	pmuludq	%mm4,%mm0
 	pmuludq	%mm5,%mm1
 	paddq	%mm0,%mm2
@@ -99,7 +107,7 @@ L0031st:
 	paddq	%mm2,%mm3
 	movq	%mm3,32(%esp,%ebx,4)
 	incl	%edx
-L004outer:
+L005outer:
 	xorl	%ecx,%ecx
 	movd	(%edi,%edx,4),%mm4
 	movd	(%esi),%mm5
@@ -121,7 +129,7 @@ L004outer:
 	paddq	%mm6,%mm2
 	incl	%ecx
 	decl	%ebx
-L005inner:
+L006inner:
 	pmuludq	%mm4,%mm0
 	pmuludq	%mm5,%mm1
 	paddq	%mm0,%mm2
@@ -138,7 +146,7 @@ L005inner:
 	paddq	%mm6,%mm2
 	decl	%ebx
 	leal	1(%ecx),%ecx
-	jnz	L005inner
+	jnz	L006inner
 	movl	%ecx,%ebx
 	pmuludq	%mm4,%mm0
 	pmuludq	%mm5,%mm1
@@ -156,11 +164,11 @@ L005inner:
 	movq	%mm3,32(%esp,%ebx,4)
 	leal	1(%edx),%edx
 	cmpl	%ebx,%edx
-	jle	L004outer
+	jle	L005outer
 	emms
-	jmp	L006common_tail
+	jmp	L007common_tail
 .align	4,0x90
-L002non_sse2:
+L003non_sse2:
 	movl	8(%esp),%esi
 	leal	1(%ebx),%ebp
 	movl	12(%esp),%edi
@@ -171,12 +179,12 @@ L002non_sse2:
 	leal	4(%edi,%ebx,4),%eax
 	orl	%edx,%ebp
 	movl	(%edi),%edi
-	jz	L007bn_sqr_mont
+	jz	L008bn_sqr_mont
 	movl	%eax,28(%esp)
 	movl	(%esi),%eax
 	xorl	%edx,%edx
 .align	4,0x90
-L008mull:
+L009mull:
 	movl	%edx,%ebp
 	mull	%edi
 	addl	%eax,%ebp
@@ -185,7 +193,7 @@ L008mull:
 	movl	(%esi,%ecx,4),%eax
 	cmpl	%ebx,%ecx
 	movl	%ebp,28(%esp,%ecx,4)
-	jl	L008mull
+	jl	L009mull
 	movl	%edx,%ebp
 	mull	%edi
 	movl	20(%esp),%edi
@@ -203,9 +211,9 @@ L008mull:
 	movl	4(%esi),%eax
 	adcl	$0,%edx
 	incl	%ecx
-	jmp	L0092ndmadd
+	jmp	L0102ndmadd
 .align	4,0x90
-L0101stmadd:
+L0111stmadd:
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ecx,4),%ebp
@@ -216,7 +224,7 @@ L0101stmadd:
 	adcl	$0,%edx
 	cmpl	%ebx,%ecx
 	movl	%ebp,28(%esp,%ecx,4)
-	jl	L0101stmadd
+	jl	L0111stmadd
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ebx,4),%eax
@@ -239,7 +247,7 @@ L0101stmadd:
 	adcl	$0,%edx
 	movl	$1,%ecx
 .align	4,0x90
-L0092ndmadd:
+L0102ndmadd:
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ecx,4),%ebp
@@ -250,7 +258,7 @@ L0092ndmadd:
 	adcl	$0,%edx
 	cmpl	%ebx,%ecx
 	movl	%ebp,24(%esp,%ecx,4)
-	jl	L0092ndmadd
+	jl	L0102ndmadd
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ebx,4),%ebp
@@ -266,16 +274,16 @@ L0092ndmadd:
 	movl	%edx,32(%esp,%ebx,4)
 	cmpl	28(%esp),%ecx
 	movl	%eax,36(%esp,%ebx,4)
-	je	L006common_tail
+	je	L007common_tail
 	movl	(%ecx),%edi
 	movl	8(%esp),%esi
 	movl	%ecx,12(%esp)
 	xorl	%ecx,%ecx
 	xorl	%edx,%edx
 	movl	(%esi),%eax
-	jmp	L0101stmadd
+	jmp	L0111stmadd
 .align	4,0x90
-L007bn_sqr_mont:
+L008bn_sqr_mont:
 	movl	%ebx,(%esp)
 	movl	%ecx,12(%esp)
 	movl	%edi,%eax
@@ -286,7 +294,7 @@ L007bn_sqr_mont:
 	andl	$1,%ebx
 	incl	%ecx
 .align	4,0x90
-L011sqr:
+L012sqr:
 	movl	(%esi,%ecx,4),%eax
 	movl	%edx,%ebp
 	mull	%edi
@@ -298,7 +306,7 @@ L011sqr:
 	cmpl	(%esp),%ecx
 	movl	%eax,%ebx
 	movl	%ebp,28(%esp,%ecx,4)
-	jl	L011sqr
+	jl	L012sqr
 	movl	(%esi,%ecx,4),%eax
 	movl	%edx,%ebp
 	mull	%edi
@@ -322,7 +330,7 @@ L011sqr:
 	movl	4(%esi),%eax
 	movl	$1,%ecx
 .align	4,0x90
-L0123rdmadd:
+L0133rdmadd:
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ecx,4),%ebp
@@ -341,7 +349,7 @@ L0123rdmadd:
 	adcl	$0,%edx
 	cmpl	%ebx,%ecx
 	movl	%ebp,24(%esp,%ecx,4)
-	jl	L0123rdmadd
+	jl	L0133rdmadd
 	movl	%edx,%ebp
 	mull	%edi
 	addl	32(%esp,%ebx,4),%ebp
@@ -357,7 +365,7 @@ L0123rdmadd:
 	movl	%edx,32(%esp,%ebx,4)
 	cmpl	%ebx,%ecx
 	movl	%eax,36(%esp,%ebx,4)
-	je	L006common_tail
+	je	L007common_tail
 	movl	4(%esi,%ecx,4),%edi
 	leal	1(%ecx),%ecx
 	movl	%edi,%eax
@@ -369,12 +377,12 @@ L0123rdmadd:
 	xorl	%ebp,%ebp
 	cmpl	%ebx,%ecx
 	leal	1(%ecx),%ecx
-	je	L013sqrlast
+	je	L014sqrlast
 	movl	%edx,%ebx
 	shrl	$1,%edx
 	andl	$1,%ebx
 .align	4,0x90
-L014sqradd:
+L015sqradd:
 	movl	(%esi,%ecx,4),%eax
 	movl	%edx,%ebp
 	mull	%edi
@@ -390,13 +398,13 @@ L014sqradd:
 	cmpl	(%esp),%ecx
 	movl	%ebp,28(%esp,%ecx,4)
 	movl	%eax,%ebx
-	jle	L014sqradd
+	jle	L015sqradd
 	movl	%edx,%ebp
 	addl	%edx,%edx
 	shrl	$31,%ebp
 	addl	%ebx,%edx
 	adcl	$0,%ebp
-L013sqrlast:
+L014sqrlast:
 	movl	20(%esp),%edi
 	movl	16(%esp),%esi
 	imull	32(%esp),%edi
@@ -411,9 +419,9 @@ L013sqrlast:
 	adcl	$0,%edx
 	movl	$1,%ecx
 	movl	4(%esi),%eax
-	jmp	L0123rdmadd
+	jmp	L0133rdmadd
 .align	4,0x90
-L006common_tail:
+L007common_tail:
 	movl	16(%esp),%ebp
 	movl	4(%esp),%edi
 	leal	32(%esp),%esi
@@ -421,13 +429,13 @@ L006common_tail:
 	movl	%ebx,%ecx
 	xorl	%edx,%edx
 .align	4,0x90
-L015sub:
+L016sub:
 	sbbl	(%ebp,%edx,4),%eax
 	movl	%eax,(%edi,%edx,4)
 	decl	%ecx
 	movl	4(%esi,%edx,4),%eax
 	leal	1(%edx),%edx
-	jge	L015sub
+	jge	L016sub
 	sbbl	$0,%eax
 	andl	%eax,%esi
 	notl	%eax
@@ -435,12 +443,12 @@ L015sub:
 	andl	%eax,%ebp
 	orl	%ebp,%esi
 .align	4,0x90
-L016copy:
+L017copy:
 	movl	(%esi,%ebx,4),%eax
 	movl	%eax,(%edi,%ebx,4)
 	movl	%ecx,32(%esp,%ebx,4)
 	decl	%ebx
-	jge	L016copy
+	jge	L017copy
 	movl	24(%esp),%esp
 	movl	$1,%eax
 L000just_leave:

deps/openssl/asm_obsolete/x86-win32-masm/bn/x86-mont.asm

@@ -45,6 +45,14 @@ $L_bn_mul_mont_begin::
 	xor	edx,2048
 	sub	esp,edx
 	and	esp,-64
+	mov	eax,ebp
+	sub	eax,esp
+	and	eax,-4096
+$L001page_walk:
+	mov	edx,DWORD PTR [eax*1+esp]
+	sub	eax,4096
+DB	46
+	jnc	$L001page_walk
 	mov	eax,DWORD PTR [esi]
 	mov	ebx,DWORD PTR 4[esi]
 	mov	ecx,DWORD PTR 8[esi]
@@ -60,7 +68,7 @@ $L_bn_mul_mont_begin::
 	mov	DWORD PTR 24[esp],ebp
 	lea	eax,DWORD PTR _OPENSSL_ia32cap_P
 	bt	DWORD PTR [eax],26
-	jnc	$L001non_sse2
+	jnc	$L002non_sse2
 	mov	eax,-1
 	movd	mm7,eax
 	mov	esi,DWORD PTR 8[esp]
@@ -84,7 +92,7 @@ $L_bn_mul_mont_begin::
 	psrlq	mm3,32
 	inc	ecx
 ALIGN	16
-$L0021st:
+$L0031st:
 	pmuludq	mm0,mm4
 	pmuludq	mm1,mm5
 	paddq	mm2,mm0
@@ -99,7 +107,7 @@ $L0021st:
 	psrlq	mm3,32
 	lea	ecx,DWORD PTR 1[ecx]
 	cmp	ecx,ebx
-	jl	$L0021st
+	jl	$L0031st
 	pmuludq	mm0,mm4
 	pmuludq	mm1,mm5
 	paddq	mm2,mm0
@@ -113,7 +121,7 @@ $L0021st:
 	paddq	mm3,mm2
 	movq	QWORD PTR 32[ebx*4+esp],mm3
 	inc	edx
-$L003outer:
+$L004outer:
 	xor	ecx,ecx
 	movd	mm4,DWORD PTR [edx*4+edi]
 	movd	mm5,DWORD PTR [esi]
@@ -135,7 +143,7 @@ $L003outer:
 	paddq	mm2,mm6
 	inc	ecx
 	dec	ebx
-$L004inner:
+$L005inner:
 	pmuludq	mm0,mm4
 	pmuludq	mm1,mm5
 	paddq	mm2,mm0
@@ -152,7 +160,7 @@ $L004inner:
 	paddq	mm2,mm6
 	dec	ebx
 	lea	ecx,DWORD PTR 1[ecx]
-	jnz	$L004inner
+	jnz	$L005inner
 	mov	ebx,ecx
 	pmuludq	mm0,mm4
 	pmuludq	mm1,mm5
@@ -170,11 +178,11 @@ $L004inner:
 	movq	QWORD PTR 32[ebx*4+esp],mm3
 	lea	edx,DWORD PTR 1[edx]
 	cmp	edx,ebx
-	jle	$L003outer
+	jle	$L004outer
 	emms
-	jmp	$L005common_tail
+	jmp	$L006common_tail
 ALIGN	16
-$L001non_sse2:
+$L002non_sse2:
 	mov	esi,DWORD PTR 8[esp]
 	lea	ebp,DWORD PTR 1[ebx]
 	mov	edi,DWORD PTR 12[esp]
@@ -185,12 +193,12 @@ $L001non_sse2:
 	lea	eax,DWORD PTR 4[ebx*4+edi]
 	or	ebp,edx
 	mov	edi,DWORD PTR [edi]
-	jz	$L006bn_sqr_mont
+	jz	$L007bn_sqr_mont
 	mov	DWORD PTR 28[esp],eax
 	mov	eax,DWORD PTR [esi]
 	xor	edx,edx
 ALIGN	16
-$L007mull:
+$L008mull:
 	mov	ebp,edx
 	mul	edi
 	add	ebp,eax
@@ -199,7 +207,7 @@ $L007mull:
 	mov	eax,DWORD PTR [ecx*4+esi]
 	cmp	ecx,ebx
 	mov	DWORD PTR 28[ecx*4+esp],ebp
-	jl	$L007mull
+	jl	$L008mull
 	mov	ebp,edx
 	mul	edi
 	mov	edi,DWORD PTR 20[esp]
@@ -217,9 +225,9 @@ $L007mull:
 	mov	eax,DWORD PTR 4[esi]
 	adc	edx,0
 	inc	ecx
-	jmp	$L0082ndmadd
+	jmp	$L0092ndmadd
 ALIGN	16
-$L0091stmadd:
+$L0101stmadd:
 	mov	ebp,edx
 	mul	edi
 	add	ebp,DWORD PTR 32[ecx*4+esp]
@@ -230,7 +238,7 @@ $L0091stmadd:
 	adc	edx,0
 	cmp	ecx,ebx
 	mov	DWORD PTR 28[ecx*4+esp],ebp
-	jl	$L0091stmadd
+	jl	$L0101stmadd
 	mov	ebp,edx
 	mul	edi
 	add	eax,DWORD PTR 32[ebx*4+esp]
@@ -253,7 +261,7 @@ $L0091stmadd:
 	adc	edx,0
 	mov	ecx,1
 ALIGN	16
-$L0082ndmadd:
+$L0092ndmadd:
 	mov	ebp,edx
 	mul	edi
 	add	ebp,DWORD PTR 32[ecx*4+esp]
@@ -264,7 +272,7 @@ $L0082ndmadd:
 	adc	edx,0
 	cmp	ecx,ebx
 	mov	DWORD PTR 24[ecx*4+esp],ebp
-	jl	$L0082ndmadd
+	jl	$L0092ndmadd
 	mov	ebp,edx
 	mul	edi
 	add	ebp,DWORD PTR 32[ebx*4+esp]
@@ -280,16 +288,16 @@ $L0082ndmadd:
 	mov	DWORD PTR 32[ebx*4+esp],edx
 	cmp	ecx,DWORD PTR 28[esp]
 	mov	DWORD PTR 36[ebx*4+esp],eax
-	je	$L005common_tail
+	je	$L006common_tail
 	mov	edi,DWORD PTR [ecx]
 	mov	esi,DWORD PTR 8[esp]
 	mov	DWORD PTR 12[esp],ecx
 	xor	ecx,ecx
 	xor	edx,edx
 	mov	eax,DWORD PTR [esi]
-	jmp	$L0091stmadd
+	jmp	$L0101stmadd
 ALIGN	16
-$L006bn_sqr_mont:
+$L007bn_sqr_mont:
 	mov	DWORD PTR [esp],ebx
 	mov	DWORD PTR 12[esp],ecx
 	mov	eax,edi
@@ -300,7 +308,7 @@ $L006bn_sqr_mont:
 	and	ebx,1
 	inc	ecx
 ALIGN	16
-$L010sqr:
+$L011sqr:
 	mov	eax,DWORD PTR [ecx*4+esi]
 	mov	ebp,edx
 	mul	edi
@@ -312,7 +320,7 @@ $L010sqr:
 	cmp	ecx,DWORD PTR [esp]
 	mov	ebx,eax
 	mov	DWORD PTR 28[ecx*4+esp],ebp
-	jl	$L010sqr
+	jl	$L011sqr
 	mov	eax,DWORD PTR [ecx*4+esi]
 	mov	ebp,edx
 	mul	edi
@@ -336,7 +344,7 @@ $L010sqr:
 	mov	eax,DWORD PTR 4[esi]
 	mov	ecx,1
 ALIGN	16
-$L0113rdmadd:
+$L0123rdmadd:
 	mov	ebp,edx
 	mul	edi
 	add	ebp,DWORD PTR 32[ecx*4+esp]
@@ -355,7 +363,7 @@ $L0113rdmadd:
 	adc	edx,0
 	cmp	ecx,ebx
 	mov	DWORD PTR 24[ecx*4+esp],ebp
-	jl	$L0113rdmadd
+	jl	$L0123rdmadd
 	mov	ebp,edx
 	mul	edi
 	add	ebp,DWORD PTR 32[ebx*4+esp]
@@ -371,7 +379,7 @@ $L0113rdmadd:
 	mov	DWORD PTR 32[ebx*4+esp],edx
 	cmp	ecx,ebx
 	mov	DWORD PTR 36[ebx*4+esp],eax
-	je	$L005common_tail
+	je	$L006common_tail
 	mov	edi,DWORD PTR 4[ecx*4+esi]
 	lea	ecx,DWORD PTR 1[ecx]
 	mov	eax,edi
@@ -383,12 +391,12 @@ $L0113rdmadd:
 	xor	ebp,ebp
 	cmp	ecx,ebx
 	lea	ecx,DWORD PTR 1[ecx]
-	je	$L012sqrlast
+	je	$L013sqrlast
 	mov	ebx,edx
 	shr	edx,1
 	and	ebx,1
 ALIGN	16
-$L013sqradd:
+$L014sqradd:
 	mov	eax,DWORD PTR [ecx*4+esi]
 	mov	ebp,edx
 	mul	edi
@@ -404,13 +412,13 @@ $L013sqradd:
 	cmp	ecx,DWORD PTR [esp]
 	mov	DWORD PTR 28[ecx*4+esp],ebp
 	mov	ebx,eax
-	jle	$L013sqradd
+	jle	$L014sqradd
 	mov	ebp,edx
 	add	edx,edx
 	shr	ebp,31
 	add	edx,ebx
 	adc	ebp,0
-$L012sqrlast:
+$L013sqrlast:
 	mov	edi,DWORD PTR 20[esp]
 	mov	esi,DWORD PTR 16[esp]
 	imul	edi,DWORD PTR 32[esp]
@@ -425,9 +433,9 @@ $L012sqrlast:
 	adc	edx,0
 	mov	ecx,1
 	mov	eax,DWORD PTR 4[esi]
-	jmp	$L0113rdmadd
+	jmp	$L0123rdmadd
 ALIGN	16
-$L005common_tail:
+$L006common_tail:
 	mov	ebp,DWORD PTR 16[esp]
 	mov	edi,DWORD PTR 4[esp]
 	lea	esi,DWORD PTR 32[esp]
@@ -435,13 +443,13 @@ $L005common_tail:
 	mov	ecx,ebx
 	xor	edx,edx
 ALIGN	16
-$L014sub:
+$L015sub:
 	sbb	eax,DWORD PTR [edx*4+ebp]
 	mov	DWORD PTR [edx*4+edi],eax
 	dec	ecx
 	mov	eax,DWORD PTR 4[edx*4+esi]
 	lea	edx,DWORD PTR 1[edx]
-	jge	$L014sub
+	jge	$L015sub
 	sbb	eax,0
 	and	esi,eax
 	not	eax
@@ -449,12 +457,12 @@ $L014sub:
 	and	ebp,eax
 	or	esi,ebp
 ALIGN	16
-$L015copy:
+$L016copy:
 	mov	eax,DWORD PTR [ebx*4+esi]
 	mov	DWORD PTR [ebx*4+edi],eax
 	mov	DWORD PTR 32[ebx*4+esp],ecx
 	dec	ebx
-	jge	$L015copy
+	jge	$L016copy
 	mov	esp,DWORD PTR 24[esp]
 	mov	eax,1
 $L000just_leave: