X-Git-Url: https://git.librecmc.org/?a=blobdiff_plain;f=crypto%2Fx86_64cpuid.pl;h=c54b9e3681c6cebba14a9509a43c13865079dccc;hb=cec2af75102df52c2f270b3751e3487f47fccb6c;hp=777d557783d8b6a85862fdb030a329d95feb104a;hpb=5f1841cdcae459924c3d1d92fcaf3110068c7cda;p=oweals%2Fopenssl.git

diff --git a/crypto/x86_64cpuid.pl b/crypto/x86_64cpuid.pl
index 777d557783..c54b9e3681 100644
--- a/crypto/x86_64cpuid.pl
+++ b/crypto/x86_64cpuid.pl
@@ -1,72 +1,39 @@
 #!/usr/bin/env perl
 
-$output=shift;
-$win64a=1 if ($output =~ /win64a\.[s|asm]/);
-open STDOUT,">$output" || die "can't open $output: $!";
+$flavour = shift;
+$output  = shift;
+if ($flavour =~ /\./) { $output = $flavour; undef $flavour; }
 
-print<<___ if(defined($win64a));
-_TEXT	SEGMENT
-PUBLIC	OPENSSL_rdtsc
-ALIGN	16
-OPENSSL_rdtsc	PROC
-	rdtsc
-	shl	rdx,32
-	or	rax,rdx
-	ret
-OPENSSL_rdtsc	ENDP
+$win64=0; $win64=1 if ($flavour =~ /[nm]asm|mingw64/ || $output =~ /\.asm$/);
 
-PUBLIC	OPENSSL_atomic_add
-ALIGN	16
-OPENSSL_atomic_add	PROC
-	mov	eax,DWORD PTR[rcx]
-\$Lspin:	lea	r8,DWORD PTR[rdx+rax]
-lock	cmpxchg	DWORD PTR[rcx],r8d
-	jne	\$Lspin
-	mov	eax,r8d
-	cdqe    
-	ret
-OPENSSL_atomic_add	ENDP
+$0 =~ m/(.*[\/\\])[^\/\\]+$/; $dir=$1;
+open STDOUT,"| $^X ${dir}perlasm/x86_64-xlate.pl $flavour $output";
 
-PUBLIC	OPENSSL_wipe_cpu
-ALIGN	16
-OPENSSL_wipe_cpu	PROC
-	pxor	xmm0,xmm0
-	pxor	xmm1,xmm1
-	pxor	xmm2,xmm2
-	pxor	xmm3,xmm3
-	pxor	xmm4,xmm4
-	pxor	xmm5,xmm5
-	xor	rcx,rcx
-	xor	rdx,rdx
-	xor	r8,r8
-	xor	r9,r9
-	xor	r10,r10
-	xor	r11,r11
-	lea	rax,QWORD PTR[rsp+8]
-	ret
-OPENSSL_wipe_cpu	ENDP
+if ($win64)	{ $arg1="%rcx"; $arg2="%rdx"; }
+else		{ $arg1="%rdi"; $arg2="%rsi"; }
+print<<___;
+.extern		OPENSSL_cpuid_setup
+.section	.init
+	call	OPENSSL_cpuid_setup
 
-OPENSSL_ia32_cpuid	PROC
-	mov	r8,rbx
-	mov	eax,1
-	cpuid
-	shl	rcx,32
-	mov	eax,edx
-	mov	rbx,r8
-	or	rax,rcx
+.text
+
+.globl	OPENSSL_atomic_add
+.type	OPENSSL_atomic_add,\@abi-omnipotent
+.align	16
+OPENSSL_atomic_add:
+	movl	($arg1),%eax
+.Lspin:	leaq	($arg2,%rax),%r8
+	.byte	0xf0		# lock
+	cmpxchgl	%r8d,($arg1)
+	jne	.Lspin
+	movl	%r8d,%eax
+	.byte	0x48,0x98	# cltq/cdqe
 	ret
-OPENSSL_ia32_cpuid	ENDP
-_TEXT	ENDS
+.size	OPENSSL_atomic_add,.-OPENSSL_atomic_add
 
-CRT\$XIU	SEGMENT
-EXTRN	OPENSSL_cpuid_setup:PROC
-DQ	OPENSSL_cpuid_setup
-CRT\$XIU	ENDS
-END
-___
-print<<___ if(!defined($win64a));
-.text
 .globl	OPENSSL_rdtsc
+.type	OPENSSL_rdtsc,\@abi-omnipotent
 .align	16
 OPENSSL_rdtsc:
 	rdtsc
@@ -75,21 +42,85 @@ OPENSSL_rdtsc:
 	ret
 .size	OPENSSL_rdtsc,.-OPENSSL_rdtsc
 
-.globl	OPENSSL_atomic_add
-.type	OPENSSL_atomic_add,\@function
+.globl	OPENSSL_ia32_cpuid
+.type	OPENSSL_ia32_cpuid,\@abi-omnipotent
 .align	16
-OPENSSL_atomic_add:
-	movl	(%rdi),%eax
-.Lspin:	lea	(%rsi,%rax),%r8
-lock;	cmpxchg	%r8d,(%rdi)
-	jne	.Lspin
-	mov	%r8d,%eax
-	cdqe
+OPENSSL_ia32_cpuid:
+	mov	%rbx,%r8
+
+	xor	%eax,%eax
+	cpuid
+	xor	%eax,%eax
+	cmp	\$0x756e6547,%ebx	# "Genu"
+	setne	%al
+	mov	%eax,%r9d
+	cmp	\$0x49656e69,%edx	# "ineI"
+	setne	%al
+	or	%eax,%r9d
+	cmp	\$0x6c65746e,%ecx	# "ntel"
+	setne	%al
+	or	%eax,%r9d
+
+	mov	\$1,%eax
+	cpuid
+	cmp	\$0,%r9d
+	jne	.Lnotintel
+	or	\$0x00100000,%edx	# use reserved 20th bit to engage RC4_CHAR
+	and	\$15,%ah
+	cmp	\$15,%ah		# examine Family ID
+	je	.Lnotintel
+	or	\$0x40000000,%edx	# use reserved bit to skip unrolled loop
+.Lnotintel:
+	bt	\$28,%edx		# test hyper-threading bit
+	jnc	.Ldone
+	shr	\$16,%ebx
+	cmp	\$1,%bl			# see if cache is shared
+	ja	.Ldone
+	and	\$0xefffffff,%edx	# ~(1<<28)
+.Ldone:
+	shl	\$32,%rcx
+	mov	%edx,%eax
+	mov	%r8,%rbx
+	or	%rcx,%rax
 	ret
-.size	OPENSSL_atomic_add,.-OPENSSL_atomic_add
+.size	OPENSSL_ia32_cpuid,.-OPENSSL_ia32_cpuid
+
+.globl  OPENSSL_cleanse
+.type   OPENSSL_cleanse,\@abi-omnipotent
+.align  16
+OPENSSL_cleanse:
+	xor	%rax,%rax
+	cmp	\$15,$arg2
+	jae	.Lot
+.Little:
+	mov	%al,($arg1)
+	sub	\$1,$arg2
+	lea	1($arg1),$arg1
+	jnz	.Little
+	ret
+.align	16
+.Lot:
+	test	\$7,$arg1
+	jz	.Laligned
+	mov	%al,($arg1)
+	lea	-1($arg2),$arg2
+	lea	1($arg1),$arg1
+	jmp	.Lot
+.Laligned:
+	mov	%rax,($arg1)
+	lea	-8($arg2),$arg2
+	test	\$-8,$arg2
+	lea	8($arg1),$arg1
+	jnz	.Laligned
+	cmp	\$0,$arg2
+	jne	.Little
+	ret
+.size	OPENSSL_cleanse,.-OPENSSL_cleanse
+___
 
+print<<___ if (!$win64);
 .globl	OPENSSL_wipe_cpu
-.type	OPENSSL_wipe_cpu,\@function
+.type	OPENSSL_wipe_cpu,\@abi-omnipotent
 .align	16
 OPENSSL_wipe_cpu:
 	pxor	%xmm0,%xmm0
@@ -108,31 +139,38 @@ OPENSSL_wipe_cpu:
 	pxor	%xmm13,%xmm13
 	pxor	%xmm14,%xmm14
 	pxor	%xmm15,%xmm15
-	xor	%rcx,%rcx
-	xor	%rdx,%rdx
-	xor	%rsi,%rsi
-	xor	%rdi,%rdi
-	xor	%r8,%r8
-	xor	%r9,%r9
-	xor	%r10,%r10
-	xor	%r11,%r11
-	lea	8(%rsp),%rax
+	xorq	%rcx,%rcx
+	xorq	%rdx,%rdx
+	xorq	%rsi,%rsi
+	xorq	%rdi,%rdi
+	xorq	%r8,%r8
+	xorq	%r9,%r9
+	xorq	%r10,%r10
+	xorq	%r11,%r11
+	leaq	8(%rsp),%rax
 	ret
 .size	OPENSSL_wipe_cpu,.-OPENSSL_wipe_cpu
-
-.globl	OPENSSL_ia32_cpuid
+___
+print<<___ if ($win64);
+.globl	OPENSSL_wipe_cpu
+.type	OPENSSL_wipe_cpu,\@abi-omnipotent
 .align	16
-OPENSSL_ia32_cpuid:
-	mov	%rbx,%r8
-	mov	\$1,%eax
-	cpuid
-	shl	\$32,%rcx
-	mov	%edx,%eax
-	mov	%r8,%rbx
-	or	%rcx,%rax
+OPENSSL_wipe_cpu:
+	pxor	%xmm0,%xmm0
+	pxor	%xmm1,%xmm1
+	pxor	%xmm2,%xmm2
+	pxor	%xmm3,%xmm3
+	pxor	%xmm4,%xmm4
+	pxor	%xmm5,%xmm5
+	xorq	%rcx,%rcx
+	xorq	%rdx,%rdx
+	xorq	%r8,%r8
+	xorq	%r9,%r9
+	xorq	%r10,%r10
+	xorq	%r11,%r11
+	leaq	8(%rsp),%rax
 	ret
-.size	OPENSSL_ia32_cpuid,.-OPENSSL_ia32_cpuid
-
-.section	.init
-	call	OPENSSL_cpuid_setup
+.size	OPENSSL_wipe_cpu,.-OPENSSL_wipe_cpu
 ___
+
+close STDOUT;	# flush