From eaa0ad97872ac1becf437c14909afe7a01b77364 Mon Sep 17 00:00:00 2001
From: Shigeki Ohtsu <ohtsu@ohtsu.org>
Date: Tue, 14 Aug 2018 23:14:19 +0900
Subject: deps: update archs files for OpenSSL-1.1.0i

`cd deps/openssl/config; make` updates all archs dependant files.

PR-URL: https://github.com/nodejs/node/pull/22318
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Rod Vagg <rod@vagg.org>
---
 .../asm/crypto/bn/x86_64-mont.s                    | 81 +++++++++++-----------
 1 file changed, 40 insertions(+), 41 deletions(-)

(limited to 'deps/openssl/config/archs/solaris64-x86_64-gcc/asm/crypto/bn/x86_64-mont.s')

diff --git a/deps/openssl/config/archs/solaris64-x86_64-gcc/asm/crypto/bn/x86_64-mont.s b/deps/openssl/config/archs/solaris64-x86_64-gcc/asm/crypto/bn/x86_64-mont.s
index d19d4662b4..3a78cd8440 100644
--- a/deps/openssl/config/archs/solaris64-x86_64-gcc/asm/crypto/bn/x86_64-mont.s
+++ b/deps/openssl/config/archs/solaris64-x86_64-gcc/asm/crypto/bn/x86_64-mont.s
@@ -1,4 +1,4 @@
-.text
+.text	
 
 
 
@@ -197,30 +197,30 @@ bn_mul_mont:
 
 	xorq	%r14,%r14
 	movq	(%rsp),%rax
-	leaq	(%rsp),%rsi
 	movq	%r9,%r15
-	jmp	.Lsub
+
 .align	16
 .Lsub:	sbbq	(%rcx,%r14,8),%rax
 	movq	%rax,(%rdi,%r14,8)
-	movq	8(%rsi,%r14,8),%rax
+	movq	8(%rsp,%r14,8),%rax
 	leaq	1(%r14),%r14
 	decq	%r15
 	jnz	.Lsub
 
 	sbbq	$0,%rax
+	movq	$-1,%rbx
+	xorq	%rax,%rbx
 	xorq	%r14,%r14
-	andq	%rax,%rsi
-	notq	%rax
-	movq	%rdi,%rcx
-	andq	%rax,%rcx
 	movq	%r9,%r15
-	orq	%rcx,%rsi
-.align	16
+
 .Lcopy:
-	movq	(%rsi,%r14,8),%rax
-	movq	%r14,(%rsp,%r14,8)
-	movq	%rax,(%rdi,%r14,8)
+	movq	(%rdi,%r14,8),%rcx
+	movq	(%rsp,%r14,8),%rdx
+	andq	%rbx,%rcx
+	andq	%rax,%rdx
+	movq	%r9,(%rsp,%r14,8)
+	orq	%rcx,%rdx
+	movq	%rdx,(%rdi,%r14,8)
 	leaq	1(%r14),%r14
 	subq	$1,%r15
 	jnz	.Lcopy
@@ -574,10 +574,10 @@ bn_mul4x_mont:
 	cmpq	%r9,%r14
 	jb	.Louter4x
 	movq	16(%rsp,%r9,8),%rdi
+	leaq	-4(%r9),%r15
 	movq	0(%rsp),%rax
-	pxor	%xmm0,%xmm0
 	movq	8(%rsp),%rdx
-	shrq	$2,%r9
+	shrq	$2,%r15
 	leaq	(%rsp),%rsi
 	xorq	%r14,%r14
 
@@ -585,9 +585,7 @@ bn_mul4x_mont:
 	movq	16(%rsi),%rbx
 	movq	24(%rsi),%rbp
 	sbbq	8(%rcx),%rdx
-	leaq	-1(%r9),%r15
-	jmp	.Lsub4x
-.align	16
+
 .Lsub4x:
 	movq	%rax,0(%rdi,%r14,8)
 	movq	%rdx,8(%rdi,%r14,8)
@@ -614,34 +612,35 @@ bn_mul4x_mont:
 
 	sbbq	$0,%rax
 	movq	%rbp,24(%rdi,%r14,8)
-	xorq	%r14,%r14
-	andq	%rax,%rsi
-	notq	%rax
-	movq	%rdi,%rcx
-	andq	%rax,%rcx
-	leaq	-1(%r9),%r15
-	orq	%rcx,%rsi
-
-	movdqu	(%rsi),%xmm1
-	movdqa	%xmm0,(%rsp)
-	movdqu	%xmm1,(%rdi)
+	pxor	%xmm0,%xmm0
+.byte	102,72,15,110,224
+	pcmpeqd	%xmm5,%xmm5
+	pshufd	$0,%xmm4,%xmm4
+	movq	%r9,%r15
+	pxor	%xmm4,%xmm5
+	shrq	$2,%r15
+	xorl	%eax,%eax
+
 	jmp	.Lcopy4x
 .align	16
 .Lcopy4x:
-	movdqu	16(%rsi,%r14,1),%xmm2
-	movdqu	32(%rsi,%r14,1),%xmm1
-	movdqa	%xmm0,16(%rsp,%r14,1)
-	movdqu	%xmm2,16(%rdi,%r14,1)
-	movdqa	%xmm0,32(%rsp,%r14,1)
-	movdqu	%xmm1,32(%rdi,%r14,1)
-	leaq	32(%r14),%r14
+	movdqa	(%rsp,%rax,1),%xmm1
+	movdqu	(%rdi,%rax,1),%xmm2
+	pand	%xmm4,%xmm1
+	pand	%xmm5,%xmm2
+	movdqa	16(%rsp,%rax,1),%xmm3
+	movdqa	%xmm0,(%rsp,%rax,1)
+	por	%xmm2,%xmm1
+	movdqu	16(%rdi,%rax,1),%xmm2
+	movdqu	%xmm1,(%rdi,%rax,1)
+	pand	%xmm4,%xmm3
+	pand	%xmm5,%xmm2
+	movdqa	%xmm0,16(%rsp,%rax,1)
+	por	%xmm2,%xmm3
+	movdqu	%xmm3,16(%rdi,%rax,1)
+	leaq	32(%rax),%rax
 	decq	%r15
 	jnz	.Lcopy4x
-
-	shlq	$2,%r9
-	movdqu	16(%rsi,%r14,1),%xmm2
-	movdqa	%xmm0,16(%rsp,%r14,1)
-	movdqu	%xmm2,16(%rdi,%r14,1)
 	movq	8(%rsp,%r9,8),%rsi
 	movq	$1,%rax
 	movq	-48(%rsi),%r15
-- 
cgit v1.2.3