Update copyright year

[openssl.git] / crypto / aes / asm / vpaes-x86_64.pl

diff --git a/crypto/aes/asm/vpaes-x86_64.pl b/crypto/aes/asm/vpaes-x86_64.pl
index 40ef342d97b5f3fed725854ec9d45c2ff553d77f..33d293e623d5ba79b5d0ef6c6a3fea65c7a6a817 100644 (file)
--- a/crypto/aes/asm/vpaes-x86_64.pl
+++ b/crypto/aes/asm/vpaes-x86_64.pl
@@ -1,4 +1,11 @@
-#!/usr/bin/env perl
+#! /usr/bin/env perl
+# Copyright 2011-2019 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
+# in the file LICENSE in the source distribution or at
+# https://www.openssl.org/source/license.html
+
 
 ######################################################################
 ## Constant-time SSSE3 AES core implementation.
@@ -30,6 +37,8 @@
 # Core 2(**)   29.6/41.1/14.3          21.9/25.2(***)
 # Nehalem      29.6/40.3/14.6          10.0/11.8
 # Atom         57.3/74.2/32.1          60.9/77.2(***)
+# Silvermont   52.7/64.0/19.5          48.8/60.8(***)
+# Goldmont     38.9/49.0/17.8          10.6/12.6
 #
 # (*)  "Hyper-threading" in the context refers rather to cache shared
 #      among multiple cores, than to specifically Intel HTT. As vast
@@ -56,7 +65,7 @@ $0 =~ m/(.*[\/\\])[^\/\\]+$/; $dir=$1;
 ( $xlate="${dir}../../perlasm/x86_64-xlate.pl" and -f $xlate) or
 die "can't locate x86_64-xlate.pl";
 
-open OUT,"| \"$^X\" $xlate $flavour $output";
+open OUT,"| \"$^X\" \"$xlate\" $flavour \"$output\"";
 *STDOUT=*OUT;
 
 $PREFIX="vpaes";
@@ -82,6 +91,7 @@ $code.=<<___;
 .type  _vpaes_encrypt_core,\@abi-omnipotent
 .align 16
 _vpaes_encrypt_core:
+.cfi_startproc
        mov     %rdx,   %r9
        mov     \$16,   %r11
        mov     240(%rdx),%eax
@@ -162,8 +172,9 @@ _vpaes_encrypt_core:
        pxor    %xmm4,  %xmm0   # 0 = A
        pshufb  %xmm1,  %xmm0
        ret
+.cfi_endproc
 .size  _vpaes_encrypt_core,.-_vpaes_encrypt_core
-       
+
 ##
 ##  Decryption core
 ##
@@ -172,6 +183,7 @@ _vpaes_encrypt_core:
 .type  _vpaes_decrypt_core,\@abi-omnipotent
 .align 16
 _vpaes_decrypt_core:
+.cfi_startproc
        mov     %rdx,   %r9             # load key
        mov     240(%rdx),%eax
        movdqa  %xmm9,  %xmm1
@@ -268,6 +280,7 @@ _vpaes_decrypt_core:
        pxor    %xmm4,  %xmm0   # 0 = A
        pshufb  %xmm2,  %xmm0
        ret
+.cfi_endproc
 .size  _vpaes_decrypt_core,.-_vpaes_decrypt_core
 
 ########################################################
@@ -278,6 +291,7 @@ _vpaes_decrypt_core:
 .type  _vpaes_schedule_core,\@abi-omnipotent
 .align 16
 _vpaes_schedule_core:
+.cfi_startproc
        # rdi = key
        # rsi = size in bits
        # rdx = buffer
@@ -324,7 +338,7 @@ _vpaes_schedule_core:
 ##
 .Lschedule_128:
        mov     \$10, %esi
-       
+
 .Loop_schedule_128:
        call    _vpaes_schedule_round
        dec     %rsi
@@ -358,7 +372,7 @@ _vpaes_schedule_core:
 
 .Loop_schedule_192:
        call    _vpaes_schedule_round
-       palignr \$8,%xmm6,%xmm0 
+       palignr \$8,%xmm6,%xmm0
        call    _vpaes_schedule_mangle  # save key n
        call    _vpaes_schedule_192_smear
        call    _vpaes_schedule_mangle  # save key n+1
@@ -384,7 +398,7 @@ _vpaes_schedule_core:
        movdqu  16(%rdi),%xmm0          # load key part 2 (unaligned)
        call    _vpaes_schedule_transform       # input transform
        mov     \$7, %esi
-       
+
 .Loop_schedule_256:
        call    _vpaes_schedule_mangle  # output low result
        movdqa  %xmm0,  %xmm6           # save cur_lo in xmm6
@@ -393,7 +407,7 @@ _vpaes_schedule_core:
        call    _vpaes_schedule_round
        dec     %rsi
        jz      .Lschedule_mangle_last
-       call    _vpaes_schedule_mangle  
+       call    _vpaes_schedule_mangle
 
        # low round. swap xmm7 and xmm6
        pshufd  \$0xFF, %xmm0,  %xmm0
@@ -401,10 +415,10 @@ _vpaes_schedule_core:
        movdqa  %xmm6,  %xmm7
        call    _vpaes_schedule_low_round
        movdqa  %xmm5,  %xmm7
-       
+
        jmp     .Loop_schedule_256
 
-       
+
 ##
 ##  .aes_schedule_mangle_last
 ##
@@ -444,6 +458,7 @@ _vpaes_schedule_core:
        pxor    %xmm6,  %xmm6
        pxor    %xmm7,  %xmm7
        ret
+.cfi_endproc
 .size  _vpaes_schedule_core,.-_vpaes_schedule_core
 
 ##
@@ -463,6 +478,7 @@ _vpaes_schedule_core:
 .type  _vpaes_schedule_192_smear,\@abi-omnipotent
 .align 16
 _vpaes_schedule_192_smear:
+.cfi_startproc
        pshufd  \$0x80, %xmm6,  %xmm1   # d c 0 0 -> c 0 0 0
        pshufd  \$0xFE, %xmm7,  %xmm0   # b a _ _ -> b b b a
        pxor    %xmm1,  %xmm6           # -> c+d c 0 0
@@ -471,6 +487,7 @@ _vpaes_schedule_192_smear:
        movdqa  %xmm6,  %xmm0
        movhlps %xmm1,  %xmm6           # clobber low side with zeros
        ret
+.cfi_endproc
 .size  _vpaes_schedule_192_smear,.-_vpaes_schedule_192_smear
 
 ##
@@ -494,6 +511,7 @@ _vpaes_schedule_192_smear:
 .type  _vpaes_schedule_round,\@abi-omnipotent
 .align 16
 _vpaes_schedule_round:
+.cfi_startproc
        # extract rcon from xmm8
        pxor    %xmm1,  %xmm1
        palignr \$15,   %xmm8,  %xmm1
@@ -503,9 +521,9 @@ _vpaes_schedule_round:
        # rotate
        pshufd  \$0xFF, %xmm0,  %xmm0
        palignr \$1,    %xmm0,  %xmm0
-       
+
        # fall through...
-       
+
        # low round: same as high round, but no rotation and no rcon.
 _vpaes_schedule_low_round:
        # smear xmm7
@@ -544,9 +562,10 @@ _vpaes_schedule_low_round:
        pxor    %xmm4,  %xmm0           # 0 = sbox output
 
        # add in smeared stuff
-       pxor    %xmm7,  %xmm0   
+       pxor    %xmm7,  %xmm0
        movdqa  %xmm0,  %xmm7
        ret
+.cfi_endproc
 .size  _vpaes_schedule_round,.-_vpaes_schedule_round
 
 ##
@@ -561,6 +580,7 @@ _vpaes_schedule_low_round:
 .type  _vpaes_schedule_transform,\@abi-omnipotent
 .align 16
 _vpaes_schedule_transform:
+.cfi_startproc
        movdqa  %xmm9,  %xmm1
        pandn   %xmm0,  %xmm1
        psrld   \$4,    %xmm1
@@ -571,6 +591,7 @@ _vpaes_schedule_transform:
        pshufb  %xmm1,  %xmm0
        pxor    %xmm2,  %xmm0
        ret
+.cfi_endproc
 .size  _vpaes_schedule_transform,.-_vpaes_schedule_transform
 
 ##
@@ -599,6 +620,7 @@ _vpaes_schedule_transform:
 .type  _vpaes_schedule_mangle,\@abi-omnipotent
 .align 16
 _vpaes_schedule_mangle:
+.cfi_startproc
        movdqa  %xmm0,  %xmm4   # save xmm0 for later
        movdqa  .Lk_mc_forward(%rip),%xmm5
        test    %rcx,   %rcx
@@ -663,6 +685,7 @@ _vpaes_schedule_mangle:
        and     \$0x30, %r8
        movdqu  %xmm3,  (%rdx)
        ret
+.cfi_endproc
 .size  _vpaes_schedule_mangle,.-_vpaes_schedule_mangle
 
 #
@@ -672,6 +695,7 @@ _vpaes_schedule_mangle:
 .type  ${PREFIX}_set_encrypt_key,\@function,3
 .align 16
 ${PREFIX}_set_encrypt_key:
+.cfi_startproc
 ___
 $code.=<<___ if ($win64);
        lea     -0xb8(%rsp),%rsp
@@ -714,12 +738,14 @@ ___
 $code.=<<___;
        xor     %eax,%eax
        ret
+.cfi_endproc
 .size  ${PREFIX}_set_encrypt_key,.-${PREFIX}_set_encrypt_key
 
 .globl ${PREFIX}_set_decrypt_key
 .type  ${PREFIX}_set_decrypt_key,\@function,3
 .align 16
 ${PREFIX}_set_decrypt_key:
+.cfi_startproc
 ___
 $code.=<<___ if ($win64);
        lea     -0xb8(%rsp),%rsp
@@ -767,12 +793,14 @@ ___
 $code.=<<___;
        xor     %eax,%eax
        ret
+.cfi_endproc
 .size  ${PREFIX}_set_decrypt_key,.-${PREFIX}_set_decrypt_key
 
 .globl ${PREFIX}_encrypt
 .type  ${PREFIX}_encrypt,\@function,3
 .align 16
 ${PREFIX}_encrypt:
+.cfi_startproc
 ___
 $code.=<<___ if ($win64);
        lea     -0xb8(%rsp),%rsp
@@ -810,12 +838,14 @@ $code.=<<___ if ($win64);
 ___
 $code.=<<___;
        ret
+.cfi_endproc
 .size  ${PREFIX}_encrypt,.-${PREFIX}_encrypt
 
 .globl ${PREFIX}_decrypt
 .type  ${PREFIX}_decrypt,\@function,3
 .align 16
 ${PREFIX}_decrypt:
+.cfi_startproc
 ___
 $code.=<<___ if ($win64);
        lea     -0xb8(%rsp),%rsp
@@ -853,6 +883,7 @@ $code.=<<___ if ($win64);
 ___
 $code.=<<___;
        ret
+.cfi_endproc
 .size  ${PREFIX}_decrypt,.-${PREFIX}_decrypt
 ___
 {
@@ -865,6 +896,7 @@ $code.=<<___;
 .type  ${PREFIX}_cbc_encrypt,\@function,6
 .align 16
 ${PREFIX}_cbc_encrypt:
+.cfi_startproc
        xchg    $key,$len
 ___
 ($len,$key)=($key,$len);
@@ -935,6 +967,7 @@ ___
 $code.=<<___;
 .Lcbc_abort:
        ret
+.cfi_endproc
 .size  ${PREFIX}_cbc_encrypt,.-${PREFIX}_cbc_encrypt
 ___
 }
@@ -948,6 +981,7 @@ $code.=<<___;
 .type  _vpaes_preheat,\@abi-omnipotent
 .align 16
 _vpaes_preheat:
+.cfi_startproc
        lea     .Lk_s0F(%rip), %r10
        movdqa  -0x20(%r10), %xmm10     # .Lk_inv
        movdqa  -0x10(%r10), %xmm11     # .Lk_inv+16
@@ -957,6 +991,7 @@ _vpaes_preheat:
        movdqa  0x50(%r10), %xmm15      # .Lk_sb2
        movdqa  0x60(%r10), %xmm14      # .Lk_sb2+16
        ret
+.cfi_endproc
 .size  _vpaes_preheat,.-_vpaes_preheat
 ########################################################
 ##                                                    ##