Handle find-doc-nits script rename

[openssl.git] / crypto / x86_64cpuid.pl

diff --git a/crypto/x86_64cpuid.pl b/crypto/x86_64cpuid.pl
index 0cf68f1360ef2e9b49960c64f038ec78435e7a4a..c2a7d72b0ee9b241993211fb0ebd51b6b09179a1 100644 (file)
--- a/crypto/x86_64cpuid.pl
+++ b/crypto/x86_64cpuid.pl
@@ -1,4 +1,11 @@
-#!/usr/bin/env perl
+#! /usr/bin/env perl
+# Copyright 2005-2016 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
+
 
 $flavour = shift;
 $output  = shift;
@@ -11,7 +18,7 @@ $0 =~ m/(.*[\/\\])[^\/\\]+$/; $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;
 
 ($arg1,$arg2,$arg3,$arg4)=$win64?("%rcx","%rdx","%r8", "%r9") :        # Win64 order
@@ -56,13 +63,25 @@ OPENSSL_rdtsc:
 .type  OPENSSL_ia32_cpuid,\@function,1
 .align 16
 OPENSSL_ia32_cpuid:
+.cfi_startproc
        mov     %rbx,%r8                # save %rbx
+.cfi_register  %rbx,%r8
 
        xor     %eax,%eax
        mov     %eax,8(%rdi)            # clear 3rd word
        cpuid
        mov     %eax,%r11d              # max value for standard query level
 
+       cmp     \$7,%eax
+       jb      .Lno_extended_info
+
+       mov     \$7,%eax
+       xor     %ecx,%ecx
+       cpuid
+       mov     %ebx,8(%rdi)
+
+.Lno_extended_info:
+
        xor     %eax,%eax
        cmp     \$0x756e6547,%ebx       # "Genu"
        setne   %al
@@ -127,14 +146,6 @@ OPENSSL_ia32_cpuid:
        shr     \$14,%r10d
        and     \$0xfff,%r10d           # number of cores -1 per L1D
 
-       cmp     \$7,%r11d
-       jb      .Lnocacheinfo
-
-       mov     \$7,%eax
-       xor     %ecx,%ecx
-       cpuid
-       mov     %ebx,8(%rdi)
-
 .Lnocacheinfo:
        mov     \$1,%eax
        cpuid
@@ -168,19 +179,29 @@ OPENSSL_ia32_cpuid:
        jnc     .Lclear_avx
        xor     %ecx,%ecx               # XCR0
        .byte   0x0f,0x01,0xd0          # xgetbv
+       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)
        and     \$6,%eax                # isolate XMM and YMM state support
        cmp     \$6,%eax
        je      .Ldone
 .Lclear_avx:
        mov     \$0xefffe7ff,%eax       # ~(1<<28|1<<12|1<<11)
        and     %eax,%r9d               # clear AVX, FMA and AMD XOP bits
-       andl    \$0xffffffdf,8(%rdi)    # cleax AVX2, ~(1<<5)
+       mov     \$0x3fdeffdf,%eax       # ~(1<<31|1<<30|1<<21|1<<16|1<<5)
+       and     %eax,8(%rdi)            # cleax AVX2 and AVX512* bits
 .Ldone:
        shl     \$32,%r9
        mov     %r10d,%eax
        mov     %r8,%rbx                # restore %rbx
+.cfi_restore   %rbx
        or      %r9,%rax
        ret
+.cfi_endproc
 .size  OPENSSL_ia32_cpuid,.-OPENSSL_ia32_cpuid
 
 .globl  OPENSSL_cleanse
@@ -217,6 +238,28 @@ OPENSSL_cleanse:
        jne     .Little
        ret
 .size  OPENSSL_cleanse,.-OPENSSL_cleanse
+
+.globl  CRYPTO_memcmp
+.type   CRYPTO_memcmp,\@abi-omnipotent
+.align  16
+CRYPTO_memcmp:
+       xor     %rax,%rax
+       xor     %r10,%r10
+       cmp     \$0,$arg3
+       je      .Lno_data
+.Loop_cmp:
+       mov     ($arg1),%r10b
+       lea     1($arg1),$arg1
+       xor     ($arg2),%r10b
+       lea     1($arg2),$arg2
+       or      %r10b,%al
+       dec     $arg3
+       jnz     .Loop_cmp
+       neg     %rax
+       shr     \$63,%rax
+.Lno_data:
+       ret
+.size  CRYPTO_memcmp,.-CRYPTO_memcmp
 ___
 
 print<<___ if (!$win64);
@@ -364,21 +407,67 @@ OPENSSL_instrument_bus2:
 ___
 }
 
+sub gen_random {
+my $rdop = shift;
 print<<___;
-.globl OPENSSL_ia32_rdrand
-.type  OPENSSL_ia32_rdrand,\@abi-omnipotent
+.globl OPENSSL_ia32_${rdop}
+.type  OPENSSL_ia32_${rdop},\@abi-omnipotent
 .align 16
-OPENSSL_ia32_rdrand:
+OPENSSL_ia32_${rdop}:
        mov     \$8,%ecx
-.Loop_rdrand:
-       rdrand  %rax
-       jc      .Lbreak_rdrand
-       loop    .Loop_rdrand
-.Lbreak_rdrand:
+.Loop_${rdop}:
+       ${rdop} %rax
+       jc      .Lbreak_${rdop}
+       loop    .Loop_${rdop}
+.Lbreak_${rdop}:
        cmp     \$0,%rax
        cmove   %rcx,%rax
        ret
-.size  OPENSSL_ia32_rdrand,.-OPENSSL_ia32_rdrand
+.size  OPENSSL_ia32_${rdop},.-OPENSSL_ia32_${rdop}
+
+.globl OPENSSL_ia32_${rdop}_bytes
+.type  OPENSSL_ia32_${rdop}_bytes,\@abi-omnipotent
+.align 16
+OPENSSL_ia32_${rdop}_bytes:
+       xor     %rax, %rax      # return value
+       cmp     \$0,$arg2
+       je      .Ldone_${rdop}_bytes
+
+       mov     \$8,%r11
+.Loop_${rdop}_bytes:
+       ${rdop} %r10
+       jc      .Lbreak_${rdop}_bytes
+       dec     %r11
+       jnz     .Loop_${rdop}_bytes
+       jmp     .Ldone_${rdop}_bytes
+
+.align 16
+.Lbreak_${rdop}_bytes:
+       cmp     \$8,$arg2
+       jb      .Ltail_${rdop}_bytes
+       mov     %r10,($arg1)
+       lea     8($arg1),$arg1
+       add     \$8,%rax
+       sub     \$8,$arg2
+       jz      .Ldone_${rdop}_bytes
+       mov     \$8,%r11
+       jmp     .Loop_${rdop}_bytes
+
+.align 16
+.Ltail_${rdop}_bytes:
+       mov     %r10b,($arg1)
+       lea     1($arg1),$arg1
+       inc     %rax
+       shr     \$8,%r8
+       dec     $arg2
+       jnz     .Ltail_${rdop}_bytes
+
+.Ldone_${rdop}_bytes:
+       ret
+.size  OPENSSL_ia32_${rdop}_bytes,.-OPENSSL_ia32_${rdop}_bytes
 ___
+}
+gen_random("rdrand");
+gen_random("rdseed");
 
 close STDOUT;  # flush