Update to OpenSSL 1.0.2.o

tmp64/aesni-sha1-x86_64.asm

@@ -2795,6 +2795,7 @@ $L$prologue_shaext:
 	mov	r11d,DWORD[240+rcx]
 	sub	rsi,rdi
 	movups	xmm15,XMMWORD[rcx]
+	movups	xmm2,XMMWORD[r8]
 	movups	xmm0,XMMWORD[16+rcx]
 	lea	rcx,[112+rcx]
 

tmp64/aesni-sha256-x86_64.asm

@@ -4147,6 +4147,7 @@ $L$prologue_shaext:
 	mov	r11d,DWORD[240+rcx]
 	sub	rsi,rdi
 	movups	xmm15,XMMWORD[rcx]
+	movups	xmm6,XMMWORD[r8]
 	movups	xmm4,XMMWORD[16+rcx]
 	lea	rcx,[112+rcx]
 

tmp64/ecp_nistz256-x86_64.asm

@@ -1132,19 +1132,18 @@ DB	0x67
 	adox	r13,rbp
 DB	0x67,0x67
 	mulx	rax,rcx,rdx
-	mov	rdx,r8
+	mov	rdx,QWORD[(($L$poly+24))]
 	adox	r14,rcx
 	shlx	rcx,r8,rsi
 	adox	r15,rax
 	shrx	rax,r8,rsi
-	mov	rbp,QWORD[(($L$poly+24))]
+	mov	rbp,rdx
 
 
 	add	r9,rcx
 	adc	r10,rax
 
-	mulx	r8,rcx,rbp
-	mov	rdx,r9
+	mulx	r8,rcx,r8
 	adc	r11,rcx
 	shlx	rcx,r9,rsi
 	adc	r8,0
@@ -1154,8 +1153,7 @@ DB	0x67,0x67
 	add	r10,rcx
 	adc	r11,rax
 
-	mulx	r9,rcx,rbp
-	mov	rdx,r10
+	mulx	r9,rcx,r9
 	adc	r8,rcx
 	shlx	rcx,r10,rsi
 	adc	r9,0
@@ -1165,8 +1163,7 @@ DB	0x67,0x67
 	add	r11,rcx
 	adc	r8,rax
 
-	mulx	r10,rcx,rbp
-	mov	rdx,r11
+	mulx	r10,rcx,r10
 	adc	r9,rcx
 	shlx	rcx,r11,rsi
 	adc	r10,0
@@ -1176,12 +1173,12 @@ DB	0x67,0x67
 	add	r8,rcx
 	adc	r9,rax
 
-	mulx	r11,rcx,rbp
+	mulx	r11,rcx,r11
 	adc	r10,rcx
 	adc	r11,0
 
 	xor	rdx,rdx
-	adc	r12,r8
+	add	r12,r8
 	mov	rsi,QWORD[(($L$poly+8))]
 	adc	r13,r9
 	mov	r8,r12
@@ -1190,8 +1187,7 @@ DB	0x67,0x67
 	mov	r9,r13
 	adc	rdx,0
 
-	xor	eax,eax
-	sbb	r12,-1
+	sub	r12,-1
 	mov	r10,r14
 	sbb	r13,rsi
 	sbb	r14,0

tmp64/rsaz-avx2.asm

@@ -94,7 +94,7 @@ $L$sqr_1024_no_n_copy:
 	vmovdqu	ymm8,YMMWORD[((256-128))+rsi]
 
 	lea	rbx,[192+rsp]
-	vpbroadcastq	ymm15,QWORD[$L$and_mask]
+	vmovdqu	ymm15,YMMWORD[$L$and_mask]
 	jmp	NEAR $L$OOP_GRANDE_SQR_1024
 
 ALIGN	32
@@ -863,10 +863,10 @@ $L$oop_mul_1024:
 	vpmuludq	ymm12,ymm11,YMMWORD[((192-128))+rcx]
 	vpaddq	ymm6,ymm6,ymm12
 	vpmuludq	ymm13,ymm11,YMMWORD[((224-128))+rcx]
-	vpblendd	ymm9,ymm9,ymm14,3
+	vpblendd	ymm12,ymm9,ymm14,3
 	vpaddq	ymm7,ymm7,ymm13
 	vpmuludq	ymm0,ymm11,YMMWORD[((256-128))+rcx]
-	vpaddq	ymm3,ymm3,ymm9
+	vpaddq	ymm3,ymm3,ymm12
 	vpaddq	ymm8,ymm8,ymm0
 
 	mov	rax,rbx
@@ -879,7 +879,9 @@ $L$oop_mul_1024:
 	vmovdqu	ymm13,YMMWORD[((-8+64-128))+rsi]
 
 	mov	rax,r10
+	vpblendd	ymm9,ymm9,ymm14,0xfc
 	imul	eax,r8d
+	vpaddq	ymm4,ymm4,ymm9
 	and	eax,0x1fffffff
 
 	imul	rbx,QWORD[((16-128))+rsi]
@@ -1108,7 +1110,6 @@ $L$oop_mul_1024:
 
 	dec	r14d
 	jnz	NEAR $L$oop_mul_1024
-	vpermq	ymm15,ymm15,0
 	vpaddq	ymm0,ymm12,YMMWORD[rsp]
 
 	vpsrlq	ymm12,ymm0,29
@@ -1785,7 +1786,7 @@ rsaz_avx2_eligible:
 
 ALIGN	64
 $L$and_mask:
-	DQ	0x1fffffff,0x1fffffff,0x1fffffff,-1
+	DQ	0x1fffffff,0x1fffffff,0x1fffffff,0x1fffffff
 $L$scatter_permd:
 	DD	0,2,4,6,7,7,7,7
 $L$gather_permd:

tmp64/x86_64-mont5.asm

@@ -3164,11 +3164,19 @@ DB	0x67
 
 ALIGN	32
 $L$sqrx8x_break:
-	sub	r8,QWORD[((16+8))+rsp]
+	xor	rbp,rbp
+	sub	rbx,QWORD[((16+8))+rsp]
+	adcx	r8,rbp
 	mov	rcx,QWORD[((24+8))+rsp]
+	adcx	r9,rbp
 	mov	rdx,QWORD[rsi]
-	xor	ebp,ebp
+	adc	r10,0
 	mov	QWORD[rdi],r8
+	adc	r11,0
+	adc	r12,0
+	adc	r13,0
+	adc	r14,0
+	adc	r15,0
 	cmp	rdi,rcx
 	je	NEAR $L$sqrx8x_outer_loop
 

tmp64/x86_64cpuid.asm

@@ -119,14 +119,6 @@ $L$intel:
 	shr	r10d,14
 	and	r10d,0xfff
 
-	cmp	r11d,7
-	jb	NEAR $L$nocacheinfo
-
-	mov	eax,7
-	xor	ecx,ecx
-	cpuid
-	mov	DWORD[8+rdi],ebx
-
 $L$nocacheinfo:
 	mov	eax,1
 	cpuid
@@ -136,8 +128,19 @@ $L$nocacheinfo:
 	or	edx,0x40000000
 	and	ah,15
 	cmp	ah,15
-	jne	NEAR $L$notintel
+	jne	NEAR $L$notP4
 	or	edx,0x00100000
+$L$notP4:
+	cmp	ah,6
+	jne	NEAR $L$notintel
+	and	eax,0x0fff0ff0
+	cmp	eax,0x00050670
+	je	NEAR $L$knights
+	cmp	eax,0x00080650
+	jne	NEAR $L$notintel
+$L$knights:
+	and	ecx,0xfbffffff
+
 $L$notintel:
 	bt	edx,28
 	jnc	NEAR $L$generic
@@ -156,6 +159,19 @@ $L$generic:
 	or	r9d,ecx
 
 	mov	r10d,edx
+
+	cmp	r11d,7
+	jb	NEAR $L$no_extended_info
+	mov	eax,7
+	xor	ecx,ecx
+	cpuid
+	bt	r9d,26
+	jc	NEAR $L$notknights
+	and	ebx,0xfff7ffff
+$L$notknights:
+	mov	DWORD[8+rdi],ebx
+$L$no_extended_info:
+
 	bt	r9d,27
 	jnc	NEAR $L$clear_avx
 	xor	ecx,ecx