deps: update openssl asm and asm_obsolete files

Regenerate asm files with Makefile and CC=gcc and ASM=nasm where gcc
version was 5.4.0 and nasm version was 2.11.08.

Also asm files in asm_obsolete dir to support old compiler and
assembler are regenerated without CC and ASM envs.

PR-URL: #16691
Reviewed-By: Ben Noordhuis <[email protected]>
Reviewed-By: Tobias Nießen <[email protected]>
Reviewed-By: Michael Dawson <[email protected]>
Reviewed-By: Myles Borins <[email protected]>
Reviewed-By: Colin Ihrig <[email protected]>
Reviewed-By: Rod Vagg <[email protected]>
Reviewed-By: Gireesh Punathil <[email protected]>
Reviewed-By: James M Snell <[email protected]>

Author: shigeki

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

@@ -3075,11 +3075,19 @@ __bn_sqrx8x_internal:
 
 .align	32
 .Lsqrx8x_break:
-	subq	16+8(%rsp),%r8
+	xorq	%rbp,%rbp
+	subq	16+8(%rsp),%rbx
+	adcxq	%rbp,%r8
 	movq	24+8(%rsp),%rcx
+	adcxq	%rbp,%r9
 	movq	0(%rsi),%rdx
-	xorl	%ebp,%ebp
+	adcq	$0,%r10
 	movq	%r8,0(%rdi)
+	adcq	$0,%r11
+	adcq	$0,%r12
+	adcq	$0,%r13
+	adcq	$0,%r14
+	adcq	$0,%r15
 	cmpq	%rcx,%rdi
 	je	.Lsqrx8x_outer_loop
 

deps/openssl/asm/x64-elf-gas/ec/ecp_nistz256-x86_64.s

@@ -1036,19 +1036,18 @@ __ecp_nistz256_sqr_montx:
 	adoxq	%rbp,%r13
 .byte	0x67,0x67
 	mulxq	%rdx,%rcx,%rax
-	movq	%r8,%rdx
+	movq	.Lpoly+24(%rip),%rdx
 	adoxq	%rcx,%r14
 	shlxq	%rsi,%r8,%rcx
 	adoxq	%rax,%r15
 	shrxq	%rsi,%r8,%rax
-	movq	.Lpoly+24(%rip),%rbp
+	movq	%rdx,%rbp
 
 
 	addq	%rcx,%r9
 	adcq	%rax,%r10
 
-	mulxq	%rbp,%rcx,%r8
-	movq	%r9,%rdx
+	mulxq	%r8,%rcx,%r8
 	adcq	%rcx,%r11
 	shlxq	%rsi,%r9,%rcx
 	adcq	$0,%r8
@@ -1058,8 +1057,7 @@ __ecp_nistz256_sqr_montx:
 	addq	%rcx,%r10
 	adcq	%rax,%r11
 
-	mulxq	%rbp,%rcx,%r9
-	movq	%r10,%rdx
+	mulxq	%r9,%rcx,%r9
 	adcq	%rcx,%r8
 	shlxq	%rsi,%r10,%rcx
 	adcq	$0,%r9
@@ -1069,8 +1067,7 @@ __ecp_nistz256_sqr_montx:
 	addq	%rcx,%r11
 	adcq	%rax,%r8
 
-	mulxq	%rbp,%rcx,%r10
-	movq	%r11,%rdx
+	mulxq	%r10,%rcx,%r10
 	adcq	%rcx,%r9
 	shlxq	%rsi,%r11,%rcx
 	adcq	$0,%r10
@@ -1080,12 +1077,12 @@ __ecp_nistz256_sqr_montx:
 	addq	%rcx,%r8
 	adcq	%rax,%r9
 
-	mulxq	%rbp,%rcx,%r11
+	mulxq	%r11,%rcx,%r11
 	adcq	%rcx,%r10
 	adcq	$0,%r11
 
 	xorq	%rdx,%rdx
-	adcq	%r8,%r12
+	addq	%r8,%r12
 	movq	.Lpoly+8(%rip),%rsi
 	adcq	%r9,%r13
 	movq	%r12,%r8
@@ -1094,8 +1091,7 @@ __ecp_nistz256_sqr_montx:
 	movq	%r13,%r9
 	adcq	$0,%rdx
 
-	xorl	%eax,%eax
-	sbbq	$-1,%r12
+	subq	$-1,%r12
 	movq	%r14,%r10
 	sbbq	%rsi,%r13
 	sbbq	$0,%r14

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

@@ -116,8 +116,19 @@ OPENSSL_ia32_cpuid:
 	orl	$0x40000000,%edx
 	andb	$15,%ah
 	cmpb	$15,%ah
-	jne	.Lnotintel
+	jne	.LnotP4
 	orl	$0x00100000,%edx
+.LnotP4:
+	cmpb	$6,%ah
+	jne	.Lnotintel
+	andl	$0x0fff0ff0,%eax
+	cmpl	$0x00050670,%eax
+	je	.Lknights
+	cmpl	$0x00080650,%eax
+	jne	.Lnotintel
+.Lknights:
+	andl	$0xfbffffff,%ecx
+
 .Lnotintel:
 	btl	$28,%edx
 	jnc	.Lgeneric
@@ -142,6 +153,10 @@ OPENSSL_ia32_cpuid:
 	movl	$7,%eax
 	xorl	%ecx,%ecx
 	cpuid
+	btl	$26,%r9d
+	jc	.Lnotknights
+	andl	$0xfff7ffff,%ebx
+.Lnotknights:
 	movl	%ebx,8(%rdi)
 .Lno_extended_info:
 

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

@@ -3075,11 +3075,19 @@ L$sqrx8x_loop:
 
 .p2align	5
 L$sqrx8x_break:
-	subq	16+8(%rsp),%r8
+	xorq	%rbp,%rbp
+	subq	16+8(%rsp),%rbx
+	adcxq	%rbp,%r8
 	movq	24+8(%rsp),%rcx
+	adcxq	%rbp,%r9
 	movq	0(%rsi),%rdx
-	xorl	%ebp,%ebp
+	adcq	$0,%r10
 	movq	%r8,0(%rdi)
+	adcq	$0,%r11
+	adcq	$0,%r12
+	adcq	$0,%r13
+	adcq	$0,%r14
+	adcq	$0,%r15
 	cmpq	%rcx,%rdi
 	je	L$sqrx8x_outer_loop
 

deps/openssl/asm/x64-macosx-gas/ec/ecp_nistz256-x86_64.s

@@ -1036,19 +1036,18 @@ __ecp_nistz256_sqr_montx:
 	adoxq	%rbp,%r13
 .byte	0x67,0x67
 	mulxq	%rdx,%rcx,%rax
-	movq	%r8,%rdx
+	movq	L$poly+24(%rip),%rdx
 	adoxq	%rcx,%r14
 	shlxq	%rsi,%r8,%rcx
 	adoxq	%rax,%r15
 	shrxq	%rsi,%r8,%rax
-	movq	L$poly+24(%rip),%rbp
+	movq	%rdx,%rbp
 
 
 	addq	%rcx,%r9
 	adcq	%rax,%r10
 
-	mulxq	%rbp,%rcx,%r8
-	movq	%r9,%rdx
+	mulxq	%r8,%rcx,%r8
 	adcq	%rcx,%r11
 	shlxq	%rsi,%r9,%rcx
 	adcq	$0,%r8
@@ -1058,8 +1057,7 @@ __ecp_nistz256_sqr_montx:
 	addq	%rcx,%r10
 	adcq	%rax,%r11
 
-	mulxq	%rbp,%rcx,%r9
-	movq	%r10,%rdx
+	mulxq	%r9,%rcx,%r9
 	adcq	%rcx,%r8
 	shlxq	%rsi,%r10,%rcx
 	adcq	$0,%r9
@@ -1069,8 +1067,7 @@ __ecp_nistz256_sqr_montx:
 	addq	%rcx,%r11
 	adcq	%rax,%r8
 
-	mulxq	%rbp,%rcx,%r10
-	movq	%r11,%rdx
+	mulxq	%r10,%rcx,%r10
 	adcq	%rcx,%r9
 	shlxq	%rsi,%r11,%rcx
 	adcq	$0,%r10
@@ -1080,12 +1077,12 @@ __ecp_nistz256_sqr_montx:
 	addq	%rcx,%r8
 	adcq	%rax,%r9
 
-	mulxq	%rbp,%rcx,%r11
+	mulxq	%r11,%rcx,%r11
 	adcq	%rcx,%r10
 	adcq	$0,%r11
 
 	xorq	%rdx,%rdx
-	adcq	%r8,%r12
+	addq	%r8,%r12
 	movq	L$poly+8(%rip),%rsi
 	adcq	%r9,%r13
 	movq	%r12,%r8
@@ -1094,8 +1091,7 @@ __ecp_nistz256_sqr_montx:
 	movq	%r13,%r9
 	adcq	$0,%rdx
 
-	xorl	%eax,%eax
-	sbbq	$-1,%r12
+	subq	$-1,%r12
 	movq	%r14,%r10
 	sbbq	%rsi,%r13
 	sbbq	$0,%r14

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

@@ -117,8 +117,19 @@ L$nocacheinfo:
 	orl	$0x40000000,%edx
 	andb	$15,%ah
 	cmpb	$15,%ah
-	jne	L$notintel
+	jne	L$notP4
 	orl	$0x00100000,%edx
+L$notP4:
+	cmpb	$6,%ah
+	jne	L$notintel
+	andl	$0x0fff0ff0,%eax
+	cmpl	$0x00050670,%eax
+	je	L$knights
+	cmpl	$0x00080650,%eax
+	jne	L$notintel
+L$knights:
+	andl	$0xfbffffff,%ecx
+
 L$notintel:
 	btl	$28,%edx
 	jnc	L$generic
@@ -143,6 +154,10 @@ L$generic:
 	movl	$7,%eax
 	xorl	%ecx,%ecx
 	cpuid
+	btl	$26,%r9d
+	jc	L$notknights
+	andl	$0xfff7ffff,%ebx
+L$notknights:
 	movl	%ebx,8(%rdi)
 L$no_extended_info:
 

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

@@ -3166,11 +3166,19 @@ DB	067h
 
 ALIGN	32
 $L$sqrx8x_break::
-	sub	r8,QWORD PTR[((16+8))+rsp]
+	xor	rbp,rbp
+	sub	rbx,QWORD PTR[((16+8))+rsp]
+	adcx	r8,rbp
 	mov	rcx,QWORD PTR[((24+8))+rsp]
+	adcx	r9,rbp
 	mov	rdx,QWORD PTR[rsi]
-	xor	ebp,ebp
+	adc	r10,0
 	mov	QWORD PTR[rdi],r8
+	adc	r11,0
+	adc	r12,0
+	adc	r13,0
+	adc	r14,0
+	adc	r15,0
 	cmp	rdi,rcx
 	je	$L$sqrx8x_outer_loop
 

deps/openssl/asm/x64-win32-masm/ec/ecp_nistz256-x86_64.asm

@@ -1137,19 +1137,18 @@ DB	067h
 	adox	r13,rbp
 DB	067h,067h
 	mulx	rax,rcx,rdx
-	mov	rdx,r8
+	mov	rdx,QWORD PTR[(($L$poly+24))]
 	adox	r14,rcx
 	shlx	rcx,r8,rsi
 	adox	r15,rax
 	shrx	rax,r8,rsi
-	mov	rbp,QWORD PTR[(($L$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
@@ -1159,8 +1158,7 @@ DB	067h,067h
 	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
@@ -1170,8 +1168,7 @@ DB	067h,067h
 	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
@@ -1181,12 +1178,12 @@ DB	067h,067h
 	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 PTR[(($L$poly+8))]
 	adc	r13,r9
 	mov	r8,r12
@@ -1195,8 +1192,7 @@ DB	067h,067h
 	mov	r9,r13
 	adc	rdx,0
 
-	xor	eax,eax
-	sbb	r12,-1
+	sub	r12,-1
 	mov	r10,r14
 	sbb	r13,rsi
 	sbb	r14,0

deps/openssl/asm/x64-win32-masm/x86_64cpuid.asm

@@ -127,8 +127,19 @@ $L$nocacheinfo::
 	or	edx,040000000h
 	and	ah,15
 	cmp	ah,15
-	jne	$L$notintel
+	jne	$L$notP4
 	or	edx,000100000h
+$L$notP4::
+	cmp	ah,6
+	jne	$L$notintel
+	and	eax,00fff0ff0h
+	cmp	eax,000050670h
+	je	$L$knights
+	cmp	eax,000080650h
+	jne	$L$notintel
+$L$knights::
+	and	ecx,0fbffffffh
+
 $L$notintel::
 	bt	edx,28
 	jnc	$L$generic
@@ -153,6 +164,10 @@ $L$generic::
 	mov	eax,7
 	xor	ecx,ecx
 	cpuid
+	bt	r9d,26
+	jc	$L$notknights
+	and	ebx,0fff7ffffh
+$L$notknights::
 	mov	DWORD PTR[8+rdi],ebx
 $L$no_extended_info::
 

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

@@ -116,8 +116,19 @@ OPENSSL_ia32_cpuid:
 	orl	$0x40000000,%edx
 	andb	$15,%ah
 	cmpb	$15,%ah
-	jne	.Lnotintel
+	jne	.LnotP4
 	orl	$0x00100000,%edx
+.LnotP4:
+	cmpb	$6,%ah
+	jne	.Lnotintel
+	andl	$0x0fff0ff0,%eax
+	cmpl	$0x00050670,%eax
+	je	.Lknights
+	cmpl	$0x00080650,%eax
+	jne	.Lnotintel
+.Lknights:
+	andl	$0xfbffffff,%ecx
+
 .Lnotintel:
 	btl	$28,%edx
 	jnc	.Lgeneric
@@ -142,6 +153,10 @@ OPENSSL_ia32_cpuid:
 	movl	$7,%eax
 	xorl	%ecx,%ecx
 	cpuid
+	btl	$26,%r9d
+	jc	.Lnotknights
+	andl	$0xfff7ffff,%ebx
+.Lnotknights:
 	movl	%ebx,8(%rdi)
 .Lno_extended_info: