x86_64 assembler pack to comply with updated styling x86_64-xlate.pl rules.

crypto/aes/asm/aes-x86_64.pl

@@ -617,8 +617,7 @@ AES_encrypt:
 	push	$key
 
 	# pick Te4 copy which can't "overlap" with stack frame or key schedule
-	.picmeup	$sbox
+	lea	.LAES_Te+2048(%rip),$sbox
-	lea	AES_Te+2048-.($sbox),$sbox
 	lea	768(%rsp),%rbp
 	sub	$sbox,%rbp
 	and	\$0x300,%rbp
@@ -1210,8 +1209,7 @@ AES_decrypt:
 	push	$key
 
 	# pick Td4 copy which can't "overlap" with stack frame or key schedule
-	.picmeup	$sbox
+	lea	.LAES_Td+2048(%rip),$sbox
-	lea	AES_Td+2048-.($sbox),$sbox
 	lea	768(%rsp),%rbp
 	sub	$sbox,%rbp
 	and	\$0x300,%rbp
@@ -1292,8 +1290,7 @@ _x86_64_AES_set_encrypt_key:
 	test	\$-1,%rdi
 	jz	.Lbadpointer
 
-	.picmeup %rbp
+	lea	.LAES_Te(%rip),%rbp
-	lea	AES_Te-.(%rbp),%rbp
 	lea	2048+128(%rbp),%rbp
 
 	# prefetch Te4
@@ -1564,8 +1561,7 @@ AES_set_decrypt_key:
 		cmp	%rsi,%rdi
 	jne	.Linvert
 
-	.picmeup %rax
+	lea	.LAES_Te+2048+1024(%rip),%rax	# rcon
-	lea	AES_Te+2048+1024-.(%rax),%rax	# rcon
 
 	mov	40(%rax),$mask80
 	mov	48(%rax),$maskfe
@@ -1636,11 +1632,10 @@ AES_cbc_encrypt:
 	cld
 	mov	%r9d,%r9d	# clear upper half of enc
 
-	.picmeup $sbox
+	lea	.LAES_Te(%rip),$sbox
-	lea	AES_Te-.($sbox),$sbox
 	cmp	\$0,%r9
 	jne	.Lcbc_picked_te
-	lea	AES_Td-AES_Te($sbox),$sbox
+	lea	.LAES_Td(%rip),$sbox
 .Lcbc_picked_te:
 
 	mov	OPENSSL_ia32cap_P(%rip),%eax
@@ -2066,9 +2061,8 @@ ___
 }
 
 $code.=<<___;
-.globl	AES_Te
 .align	64
-AES_Te:
+.LAES_Te:
 ___
 	&_data_word(0xa56363c6, 0x847c7cf8, 0x997777ee, 0x8d7b7bf6);
 	&_data_word(0x0df2f2ff, 0xbd6b6bd6, 0xb16f6fde, 0x54c5c591);
@@ -2275,9 +2269,8 @@ $code.=<<___;
 	.long	0xfefefefe, 0xfefefefe, 0x1b1b1b1b, 0x1b1b1b1b
 ___
 $code.=<<___;
-.globl	AES_Td
 .align	64
-AES_Td:
+.LAES_Td:
 ___
 	&_data_word(0x50a7f451, 0x5365417e, 0xc3a4171a, 0x965e273a);
 	&_data_word(0xcb6bab3b, 0xf1459d1f, 0xab58faac, 0x9303e34b);

crypto/bn/asm/x86_64-gcc.c

@@ -182,7 +182,7 @@ BN_ULONG bn_add_words (BN_ULONG *rp, BN_ULONG *ap, BN_ULONG *bp,int n)
 
 	asm (
 	"	subq	%2,%2		\n"
-	".align 16			\n"
+	".p2align 4			\n"
 	"1:	movq	(%4,%2,8),%0	\n"
 	"	adcq	(%5,%2,8),%0	\n"
 	"	movq	%0,(%3,%2,8)	\n"
@@ -205,7 +205,7 @@ BN_ULONG bn_sub_words (BN_ULONG *rp, BN_ULONG *ap, BN_ULONG *bp,int n)
 
 	asm (
 	"	subq	%2,%2		\n"
-	".align 16			\n"
+	".p2align 4			\n"
 	"1:	movq	(%4,%2,8),%0	\n"
 	"	sbbq	(%5,%2,8),%0	\n"
 	"	movq	%0,(%3,%2,8)	\n"

crypto/rc4/asm/rc4-x86_64.pl

@@ -336,8 +336,7 @@ RC4_set_key:
 .type	RC4_options,\@function,0
 .align	16
 RC4_options:
-	.picmeup %rax
+	lea	.Lopts(%rip),%rax
-	lea	.Lopts-.(%rax),%rax
 	mov	OPENSSL_ia32cap_P(%rip),%edx
 	bt	\$20,%edx
 	jnc	.Ldone

crypto/sha/asm/sha512-x86_64.pl

@@ -40,14 +40,16 @@
 # sha256_block:-( This is presumably because 64-bit shifts/rotates
 # apparently are not atomic instructions, but implemented in microcode.
 
-$output=shift;
+$flavour = shift;
+$output  = shift;
+if ($flavour =~ /\./) { $output = $flavour; undef $flavour; }
 
 $0 =~ m/(.*[\/\\])[^\/\\]+$/; $dir=$1;
 ( $xlate="${dir}x86_64-xlate.pl" and -f $xlate ) or
 ( $xlate="${dir}../../perlasm/x86_64-xlate.pl" and -f $xlate) or
 die "can't locate x86_64-xlate.pl";
 
-open STDOUT,"| $^X $xlate $output";
+open STDOUT,"| $^X $xlate $flavour $output";
 
 if ($output =~ /512/) {
 	$func="sha512_block_data_order";
@@ -196,8 +198,7 @@ $func:
 	mov	%rdx,$_end		# save end pointer, "3rd" arg
 	mov	%rbp,$_rsp		# save copy of %rsp
 
-	.picmeup $Tbl
+	lea	$TABLE(%rip),$Tbl
-	lea	$TABLE-.($Tbl),$Tbl
 
 	mov	$SZ*0($ctx),$A
 	mov	$SZ*1($ctx),$B

crypto/whrlpool/asm/wp-x86_64.pl

@@ -71,8 +71,7 @@ $func:
 	mov	%rdx,16(%rbx)
 	mov	%rax,32(%rbx)		# saved stack pointer
 
-	.picmeup %rbp
+	lea	$table(%rip),%rbp
-	lea	$table-.(%rbp),%rbp
 
 	xor	%rcx,%rcx
 	xor	%rdx,%rdx

crypto/x86_64cpuid.pl

@@ -1,110 +1,37 @@
 #!/usr/bin/env perl
 
-$output=shift;
+$flavour = shift;
-$masm=1 if ($output =~ /\.asm/);
+$output  = shift;
-open STDOUT,">$output" || die "can't open $output: $!";
+if ($flavour =~ /\./) { $output = $flavour; undef $flavour; }
 
-print<<___ if(defined($masm));
+$win64=0; $win64=1 if ($flavour =~ /[nm]asm|mingw64/ || $output =~ /\.asm$/);
-_TEXT	SEGMENT
-PUBLIC	OPENSSL_rdtsc
 
-PUBLIC	OPENSSL_atomic_add
+$0 =~ m/(.*[\/\\])[^\/\\]+$/; $dir=$1;
-ALIGN	16
+open STDOUT,"| $^X ${dir}perlasm/x86_64-xlate.pl $flavour $output";
-OPENSSL_atomic_add	PROC
-	mov	eax,DWORD PTR[rcx]
-\$Lspin:	lea	r8,DWORD PTR[rdx+rax]
-lock	cmpxchg	DWORD PTR[rcx],r8d
-	jne	\$Lspin
-	mov	eax,r8d
-	cdqe    
-	ret
-OPENSSL_atomic_add	ENDP
-
-PUBLIC	OPENSSL_wipe_cpu
-ALIGN	16
-OPENSSL_wipe_cpu	PROC
-	pxor	xmm0,xmm0
-	pxor	xmm1,xmm1
-	pxor	xmm2,xmm2
-	pxor	xmm3,xmm3
-	pxor	xmm4,xmm4
-	pxor	xmm5,xmm5
-	xor	rcx,rcx
-	xor	rdx,rdx
-	xor	r8,r8
-	xor	r9,r9
-	xor	r10,r10
-	xor	r11,r11
-	lea	rax,QWORD PTR[rsp+8]
-	ret
-OPENSSL_wipe_cpu	ENDP
-_TEXT	ENDS
-
-CRT\$XIU	SEGMENT
-EXTRN	OPENSSL_cpuid_setup:PROC
-DQ	OPENSSL_cpuid_setup
-CRT\$XIU	ENDS
-
-___
-print<<___ if(!defined($masm));
-.text
-
-.globl	OPENSSL_atomic_add
-.type	OPENSSL_atomic_add,\@function
-.align	16
-OPENSSL_atomic_add:
-	movl	(%rdi),%eax
-.Lspin:	leaq	(%rsi,%rax),%r8
-lock;	cmpxchgl	%r8d,(%rdi)
-	jne	.Lspin
-	movl	%r8d,%eax
-	.byte	0x48,0x98
-	ret
-.size	OPENSSL_atomic_add,.-OPENSSL_atomic_add
-
-.globl	OPENSSL_wipe_cpu
-.type	OPENSSL_wipe_cpu,\@function
-.align	16
-OPENSSL_wipe_cpu:
-	pxor	%xmm0,%xmm0
-	pxor	%xmm1,%xmm1
-	pxor	%xmm2,%xmm2
-	pxor	%xmm3,%xmm3
-	pxor	%xmm4,%xmm4
-	pxor	%xmm5,%xmm5
-	pxor	%xmm6,%xmm6
-	pxor	%xmm7,%xmm7
-	pxor	%xmm8,%xmm8
-	pxor	%xmm9,%xmm9
-	pxor	%xmm10,%xmm10
-	pxor	%xmm11,%xmm11
-	pxor	%xmm12,%xmm12
-	pxor	%xmm13,%xmm13
-	pxor	%xmm14,%xmm14
-	pxor	%xmm15,%xmm15
-	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
 
+if ($win64)	{ $arg1="%rcx"; $arg2="%rdx"; }
+else		{ $arg1="%rdi"; $arg2="%rsi"; }
+print<<___;
+.extern		OPENSSL_cpuid_setup
 .section	.init
 	call	OPENSSL_cpuid_setup
 
-___
-
-$0 =~ m/(.*[\/\\])[^\/\\]+$/; $dir=$1;
-open STDOUT,"| $^X ${dir}perlasm/x86_64-xlate.pl $output";
-
-print<<___;
 .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
+.size	OPENSSL_atomic_add,.-OPENSSL_atomic_add
+
 .globl	OPENSSL_rdtsc
 .type	OPENSSL_rdtsc,\@abi-omnipotent
 .align	16
@@ -159,35 +86,91 @@ OPENSSL_ia32_cpuid:
 .size	OPENSSL_ia32_cpuid,.-OPENSSL_ia32_cpuid
 
 .globl  OPENSSL_cleanse
-.type   OPENSSL_cleanse,\@function,2
+.type   OPENSSL_cleanse,\@abi-omnipotent
 .align  16
 OPENSSL_cleanse:
 	xor	%rax,%rax
-	cmp	\$15,%rsi
+	cmp	\$15,$arg2
 	jae	.Lot
 .Little:
-	mov	%al,(%rdi)
+	mov	%al,($arg1)
-	sub	\$1,%rsi
+	sub	\$1,$arg2
-	lea	1(%rdi),%rdi
+	lea	1($arg1),$arg1
 	jnz	.Little
 	ret
 .align	16
 .Lot:
-	test	\$7,%rdi
+	test	\$7,$arg1
 	jz	.Laligned
-	mov	%al,(%rdi)
+	mov	%al,($arg1)
-	lea	-1(%rsi),%rsi
+	lea	-1($arg2),$arg2
-	lea	1(%rdi),%rdi
+	lea	1($arg1),$arg1
 	jmp	.Lot
 .Laligned:
-	mov	%rax,(%rdi)
+	mov	%rax,($arg1)
-	lea	-8(%rsi),%rsi
+	lea	-8($arg2),$arg2
-	test	\$-8,%rsi
+	test	\$-8,$arg2
-	lea	8(%rdi),%rdi
+	lea	8($arg1),$arg1
 	jnz	.Laligned
-	cmp	\$0,%rsi
+	cmp	\$0,$arg2
 	jne	.Little
 	ret
 .size	OPENSSL_cleanse,.-OPENSSL_cleanse
 ___
+
+print<<___ if (!$win64);
+.globl	OPENSSL_wipe_cpu
+.type	OPENSSL_wipe_cpu,\@abi-omnipotent
+.align	16
+OPENSSL_wipe_cpu:
+	pxor	%xmm0,%xmm0
+	pxor	%xmm1,%xmm1
+	pxor	%xmm2,%xmm2
+	pxor	%xmm3,%xmm3
+	pxor	%xmm4,%xmm4
+	pxor	%xmm5,%xmm5
+	pxor	%xmm6,%xmm6
+	pxor	%xmm7,%xmm7
+	pxor	%xmm8,%xmm8
+	pxor	%xmm9,%xmm9
+	pxor	%xmm10,%xmm10
+	pxor	%xmm11,%xmm11
+	pxor	%xmm12,%xmm12
+	pxor	%xmm13,%xmm13
+	pxor	%xmm14,%xmm14
+	pxor	%xmm15,%xmm15
+	xorq	%rcx,%rcx
+	xorq	%rdx,%rdx
+	xorq	%rsi,%rsi
+	xorq	%rdi,%rdi
+	xorq	%r8,%r8
+	xorq	%r9,%r9
+	xorq	%r10,%r10
+	xorq	%r11,%r11
+	leaq	8(%rsp),%rax
+	ret
+.size	OPENSSL_wipe_cpu,.-OPENSSL_wipe_cpu
+___
+print<<___ if ($win64);
+.globl	OPENSSL_wipe_cpu
+.type	OPENSSL_wipe_cpu,\@abi-omnipotent
+.align	16
+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_wipe_cpu,.-OPENSSL_wipe_cpu
+___
+
 close STDOUT;	# flush