X-Git-Url: https://git.openssl.org/gitweb/?p=openssl.git;a=blobdiff_plain;f=crypto%2Fx86_64cpuid.pl;h=6423e803b759542dcc23999a8d1c239feccec4a7;hp=7c8952e7214c8108090a2cf5c62e47317c38f04d;hb=0e2ad9593a0ff494fab3234e49499a2e68542228;hpb=d6ee8f3dc4414cd97bd63b801f8644f0ff8a1f17

diff --git a/crypto/x86_64cpuid.pl b/crypto/x86_64cpuid.pl
index 7c8952e721..6423e803b7 100644
--- a/crypto/x86_64cpuid.pl
+++ b/crypto/x86_64cpuid.pl
@@ -1,5 +1,5 @@
 #! /usr/bin/env perl
-# Copyright 2005-2016 The OpenSSL Project Authors. All Rights Reserved.
+# Copyright 2005-2018 The OpenSSL Project Authors. All Rights Reserved.
 #
 # Licensed under the OpenSSL license (the "License").  You may not use
 # this file except in compliance with the License.  You can obtain a copy
@@ -139,6 +139,7 @@ OPENSSL_ia32_cpuid:
 .Lnocacheinfo:
 	mov	\$1,%eax
 	cpuid
+	movd	%eax,%xmm0		# put aside processor id
 	and	\$0xbfefffff,%edx	# force reserved bits to 0
 	cmp	\$0,%r9d
 	jne	.Lnotintel
@@ -186,6 +187,13 @@ OPENSSL_ia32_cpuid:
 	jc	.Lnotknights
 	and	\$0xfff7ffff,%ebx	# clear ADCX/ADOX flag
 .Lnotknights:
+	movd	%xmm0,%eax		# restore processor id
+	and	\$0x0fff0ff0,%eax
+	cmp	\$0x00050650,%eax	# Skylake-X
+	jne	.Lnotskylakex
+	and	\$0xfffeffff,%ebx	# ~(1<<16)
+					# suppress AVX512F flag on Skylake-X
+.Lnotskylakex:
 	mov	%ebx,8(%rdi)		# save extended feature flags
 	mov	%ecx,12(%rdi)
 .Lno_extended_info:
@@ -197,10 +205,11 @@ OPENSSL_ia32_cpuid:
 	and	\$0xe6,%eax		# isolate XMM, YMM and ZMM state support
 	cmp	\$0xe6,%eax
 	je	.Ldone
-	andl	\$0xfffeffff,8(%rdi)	# clear AVX512F, ~(1<<16)
-					# note that we don't touch other AVX512
-					# extensions, because they can be used
-					# with YMM (without opmasking though)
+	andl	\$0x3fdeffff,8(%rdi)	# ~(1<<31|1<<30|1<<21|1<<16)
+					# clear AVX512F+BW+VL+FIMA, all of
+					# them are EVEX-encoded, which requires
+					# ZMM state support even if one uses
+					# only XMM and YMM :-(
 	and	\$6,%eax		# isolate XMM and YMM state support
 	cmp	\$6,%eax
 	je	.Ldone
@@ -262,6 +271,18 @@ CRYPTO_memcmp:
 	xor	%r10,%r10
 	cmp	\$0,$arg3
 	je	.Lno_data
+	cmp	\$16,$arg3
+	jne	.Loop_cmp
+	mov	($arg1),%r10
+	mov	8($arg1),%r11
+	mov	\$1,$arg3
+	xor	($arg2),%r10
+	xor	8($arg2),%r11
+	or	%r11,%r10
+	cmovnz	$arg3,%rax
+	ret
+
+.align	16
 .Loop_cmp:
 	mov	($arg1),%r10b
 	lea	1($arg1),$arg1
@@ -425,21 +446,6 @@ ___
 sub gen_random {
 my $rdop = shift;
 print<<___;
-.globl	OPENSSL_ia32_${rdop}
-.type	OPENSSL_ia32_${rdop},\@abi-omnipotent
-.align	16
-OPENSSL_ia32_${rdop}:
-	mov	\$8,%ecx
-.Loop_${rdop}:
-	${rdop}	%rax
-	jc	.Lbreak_${rdop}
-	loop	.Loop_${rdop}
-.Lbreak_${rdop}:
-	cmp	\$0,%rax
-	cmove	%rcx,%rax
-	ret
-.size	OPENSSL_ia32_${rdop},.-OPENSSL_ia32_${rdop}
-
 .globl	OPENSSL_ia32_${rdop}_bytes
 .type	OPENSSL_ia32_${rdop}_bytes,\@abi-omnipotent
 .align	16
@@ -473,11 +479,12 @@ OPENSSL_ia32_${rdop}_bytes:
 	mov	%r10b,($arg1)
 	lea	1($arg1),$arg1
 	inc	%rax
-	shr	\$8,%r8
+	shr	\$8,%r10
 	dec	$arg2
 	jnz	.Ltail_${rdop}_bytes
 
 .Ldone_${rdop}_bytes:
+	xor	%r10,%r10	# Clear sensitive data from register
 	ret
 .size	OPENSSL_ia32_${rdop}_bytes,.-OPENSSL_ia32_${rdop}_bytes
 ___