crypto/sha/asm/sha1-ppc.pl

   1 #! /usr/bin/env perl
   2 # Copyright 2006-2016 The OpenSSL Project Authors. All Rights Reserved.
   3 #
   4 # Licensed under the OpenSSL license (the "License").  You may not use
   5 # this file except in compliance with the License.  You can obtain a copy
   6 # in the file LICENSE in the source distribution or at
   7 # https://www.openssl.org/source/license.html
   8
   9
  10 # ====================================================================
  11 # Written by Andy Polyakov <appro@fy.chalmers.se> for the OpenSSL
  12 # project. The module is, however, dual licensed under OpenSSL and
  13 # CRYPTOGAMS licenses depending on where you obtain it. For further
  14 # details see http://www.openssl.org/~appro/cryptogams/.
  15 # ====================================================================
  16
  17 # I let hardware handle unaligned input(*), except on page boundaries
  18 # (see below for details). Otherwise straightforward implementation
  19 # with X vector in register bank.
  20 #
  21 # (*) this means that this module is inappropriate for PPC403? Does
  22 #     anybody know if pre-POWER3 can sustain unaligned load?
  23
  24 #                       -m64    -m32
  25 # ----------------------------------
  26 # PPC970,gcc-4.0.0      +76%    +59%
  27 # Power6,xlc-7          +68%    +33%
  28
  29 $flavour = shift;
  30
  31 if ($flavour =~ /64/) {
  32         $SIZE_T =8;
  33         $LRSAVE =2*$SIZE_T;
  34         $UCMP   ="cmpld";
  35         $STU    ="stdu";
  36         $POP    ="ld";
  37         $PUSH   ="std";
  38 } elsif ($flavour =~ /32/) {
  39         $SIZE_T =4;
  40         $LRSAVE =$SIZE_T;
  41         $UCMP   ="cmplw";
  42         $STU    ="stwu";
  43         $POP    ="lwz";
  44         $PUSH   ="stw";
  45 } else { die "nonsense $flavour"; }
  46
  47 # Define endianess based on flavour
  48 # i.e.: linux64le
  49 $LITTLE_ENDIAN = ($flavour=~/le$/) ? $SIZE_T : 0;
  50
  51 $0 =~ m/(.*[\/\\])[^\/\\]+$/; $dir=$1;
  52 ( $xlate="${dir}ppc-xlate.pl" and -f $xlate ) or
  53 ( $xlate="${dir}../../perlasm/ppc-xlate.pl" and -f $xlate) or
  54 die "can't locate ppc-xlate.pl";
  55
  56 open STDOUT,"| $^X $xlate $flavour ".shift || die "can't call $xlate: $!";
  57
  58 $FRAME=24*$SIZE_T+64;
  59 $LOCALS=6*$SIZE_T;
  60
  61 $K  ="r0";
  62 $sp ="r1";
  63 $toc="r2";
  64 $ctx="r3";
  65 $inp="r4";
  66 $num="r5";
  67 $t0 ="r15";
  68 $t1 ="r6";
  69
  70 $A  ="r7";
  71 $B  ="r8";
  72 $C  ="r9";
  73 $D  ="r10";
  74 $E  ="r11";
  75 $T  ="r12";
  76
  77 @V=($A,$B,$C,$D,$E,$T);
  78 @X=("r16","r17","r18","r19","r20","r21","r22","r23",
  79     "r24","r25","r26","r27","r28","r29","r30","r31");
  80
  81 sub loadbe {
  82 my ($dst, $src, $temp_reg) = @_;
  83 $code.=<<___ if (!$LITTLE_ENDIAN);
  84         lwz     $dst,$src
  85 ___
  86 $code.=<<___ if ($LITTLE_ENDIAN);
  87         lwz     $temp_reg,$src
  88         rotlwi  $dst,$temp_reg,8
  89         rlwimi  $dst,$temp_reg,24,0,7
  90         rlwimi  $dst,$temp_reg,24,16,23
  91 ___
  92 }
  93
  94 sub BODY_00_19 {
  95 my ($i,$a,$b,$c,$d,$e,$f)=@_;
  96 my $j=$i+1;
  97
  98         # Since the last value of $f is discarded, we can use
  99         # it as a temp reg to swap byte-order when needed.
 100         loadbe("@X[$i]","`$i*4`($inp)",$f) if ($i==0);
 101         loadbe("@X[$j]","`$j*4`($inp)",$f) if ($i<15);
 102 $code.=<<___ if ($i<15);
 103         add     $f,$K,$e
 104         rotlwi  $e,$a,5
 105         add     $f,$f,@X[$i]
 106         and     $t0,$c,$b
 107         add     $f,$f,$e
 108         andc    $t1,$d,$b
 109         rotlwi  $b,$b,30
 110         or      $t0,$t0,$t1
 111         add     $f,$f,$t0
 112 ___
 113 $code.=<<___ if ($i>=15);
 114         add     $f,$K,$e
 115         rotlwi  $e,$a,5
 116         xor     @X[$j%16],@X[$j%16],@X[($j+2)%16]
 117         add     $f,$f,@X[$i%16]
 118         and     $t0,$c,$b
 119         xor     @X[$j%16],@X[$j%16],@X[($j+8)%16]
 120         add     $f,$f,$e
 121         andc    $t1,$d,$b
 122         rotlwi  $b,$b,30
 123         or      $t0,$t0,$t1
 124         xor     @X[$j%16],@X[$j%16],@X[($j+13)%16]
 125         add     $f,$f,$t0
 126         rotlwi  @X[$j%16],@X[$j%16],1
 127 ___
 128 }
 129
 130 sub BODY_20_39 {
 131 my ($i,$a,$b,$c,$d,$e,$f)=@_;
 132 my $j=$i+1;
 133 $code.=<<___ if ($i<79);
 134         add     $f,$K,$e
 135         xor     $t0,$b,$d
 136         rotlwi  $e,$a,5
 137         xor     @X[$j%16],@X[$j%16],@X[($j+2)%16]
 138         add     $f,$f,@X[$i%16]
 139         xor     $t0,$t0,$c
 140         xor     @X[$j%16],@X[$j%16],@X[($j+8)%16]
 141         add     $f,$f,$t0
 142         rotlwi  $b,$b,30
 143         xor     @X[$j%16],@X[$j%16],@X[($j+13)%16]
 144         add     $f,$f,$e
 145         rotlwi  @X[$j%16],@X[$j%16],1
 146 ___
 147 $code.=<<___ if ($i==79);
 148         add     $f,$K,$e
 149         xor     $t0,$b,$d
 150         rotlwi  $e,$a,5
 151         lwz     r16,0($ctx)
 152         add     $f,$f,@X[$i%16]
 153         xor     $t0,$t0,$c
 154         lwz     r17,4($ctx)
 155         add     $f,$f,$t0
 156         rotlwi  $b,$b,30
 157         lwz     r18,8($ctx)
 158         lwz     r19,12($ctx)
 159         add     $f,$f,$e
 160         lwz     r20,16($ctx)
 161 ___
 162 }
 163
 164 sub BODY_40_59 {
 165 my ($i,$a,$b,$c,$d,$e,$f)=@_;
 166 my $j=$i+1;
 167 $code.=<<___;
 168         add     $f,$K,$e
 169         rotlwi  $e,$a,5
 170         xor     @X[$j%16],@X[$j%16],@X[($j+2)%16]
 171         add     $f,$f,@X[$i%16]
 172         and     $t0,$b,$c
 173         xor     @X[$j%16],@X[$j%16],@X[($j+8)%16]
 174         add     $f,$f,$e
 175         or      $t1,$b,$c
 176         rotlwi  $b,$b,30
 177         xor     @X[$j%16],@X[$j%16],@X[($j+13)%16]
 178         and     $t1,$t1,$d
 179         or      $t0,$t0,$t1
 180         rotlwi  @X[$j%16],@X[$j%16],1
 181         add     $f,$f,$t0
 182 ___
 183 }
 184
 185 $code=<<___;
 186 .machine        "any"
 187 .text
 188
 189 .globl  .sha1_block_data_order
 190 .align  4
 191 .sha1_block_data_order:
 192         $STU    $sp,-$FRAME($sp)
 193         mflr    r0
 194         $PUSH   r15,`$FRAME-$SIZE_T*17`($sp)
 195         $PUSH   r16,`$FRAME-$SIZE_T*16`($sp)
 196         $PUSH   r17,`$FRAME-$SIZE_T*15`($sp)
 197         $PUSH   r18,`$FRAME-$SIZE_T*14`($sp)
 198         $PUSH   r19,`$FRAME-$SIZE_T*13`($sp)
 199         $PUSH   r20,`$FRAME-$SIZE_T*12`($sp)
 200         $PUSH   r21,`$FRAME-$SIZE_T*11`($sp)
 201         $PUSH   r22,`$FRAME-$SIZE_T*10`($sp)
 202         $PUSH   r23,`$FRAME-$SIZE_T*9`($sp)
 203         $PUSH   r24,`$FRAME-$SIZE_T*8`($sp)
 204         $PUSH   r25,`$FRAME-$SIZE_T*7`($sp)
 205         $PUSH   r26,`$FRAME-$SIZE_T*6`($sp)
 206         $PUSH   r27,`$FRAME-$SIZE_T*5`($sp)
 207         $PUSH   r28,`$FRAME-$SIZE_T*4`($sp)
 208         $PUSH   r29,`$FRAME-$SIZE_T*3`($sp)
 209         $PUSH   r30,`$FRAME-$SIZE_T*2`($sp)
 210         $PUSH   r31,`$FRAME-$SIZE_T*1`($sp)
 211         $PUSH   r0,`$FRAME+$LRSAVE`($sp)
 212         lwz     $A,0($ctx)
 213         lwz     $B,4($ctx)
 214         lwz     $C,8($ctx)
 215         lwz     $D,12($ctx)
 216         lwz     $E,16($ctx)
 217         andi.   r0,$inp,3
 218         bne     Lunaligned
 219 Laligned:
 220         mtctr   $num
 221         bl      Lsha1_block_private
 222         b       Ldone
 223
 224 ; PowerPC specification allows an implementation to be ill-behaved
 225 ; upon unaligned access which crosses page boundary. "Better safe
 226 ; than sorry" principle makes me treat it specially. But I don't
 227 ; look for particular offending word, but rather for 64-byte input
 228 ; block which crosses the boundary. Once found that block is aligned
 229 ; and hashed separately...
 230 .align  4
 231 Lunaligned:
 232         subfic  $t1,$inp,4096
 233         andi.   $t1,$t1,4095    ; distance to closest page boundary
 234         srwi.   $t1,$t1,6       ; t1/=64
 235         beq     Lcross_page
 236         $UCMP   $num,$t1
 237         ble     Laligned        ; didn't cross the page boundary
 238         mtctr   $t1
 239         subfc   $num,$t1,$num
 240         bl      Lsha1_block_private
 241 Lcross_page:
 242         li      $t1,16
 243         mtctr   $t1
 244         addi    r20,$sp,$LOCALS ; spot within the frame
 245 Lmemcpy:
 246         lbz     r16,0($inp)
 247         lbz     r17,1($inp)
 248         lbz     r18,2($inp)
 249         lbz     r19,3($inp)
 250         addi    $inp,$inp,4
 251         stb     r16,0(r20)
 252         stb     r17,1(r20)
 253         stb     r18,2(r20)
 254         stb     r19,3(r20)
 255         addi    r20,r20,4
 256         bdnz    Lmemcpy
 257
 258         $PUSH   $inp,`$FRAME-$SIZE_T*18`($sp)
 259         li      $t1,1
 260         addi    $inp,$sp,$LOCALS
 261         mtctr   $t1
 262         bl      Lsha1_block_private
 263         $POP    $inp,`$FRAME-$SIZE_T*18`($sp)
 264         addic.  $num,$num,-1
 265         bne     Lunaligned
 266
 267 Ldone:
 268         $POP    r0,`$FRAME+$LRSAVE`($sp)
 269         $POP    r15,`$FRAME-$SIZE_T*17`($sp)
 270         $POP    r16,`$FRAME-$SIZE_T*16`($sp)
 271         $POP    r17,`$FRAME-$SIZE_T*15`($sp)
 272         $POP    r18,`$FRAME-$SIZE_T*14`($sp)
 273         $POP    r19,`$FRAME-$SIZE_T*13`($sp)
 274         $POP    r20,`$FRAME-$SIZE_T*12`($sp)
 275         $POP    r21,`$FRAME-$SIZE_T*11`($sp)
 276         $POP    r22,`$FRAME-$SIZE_T*10`($sp)
 277         $POP    r23,`$FRAME-$SIZE_T*9`($sp)
 278         $POP    r24,`$FRAME-$SIZE_T*8`($sp)
 279         $POP    r25,`$FRAME-$SIZE_T*7`($sp)
 280         $POP    r26,`$FRAME-$SIZE_T*6`($sp)
 281         $POP    r27,`$FRAME-$SIZE_T*5`($sp)
 282         $POP    r28,`$FRAME-$SIZE_T*4`($sp)
 283         $POP    r29,`$FRAME-$SIZE_T*3`($sp)
 284         $POP    r30,`$FRAME-$SIZE_T*2`($sp)
 285         $POP    r31,`$FRAME-$SIZE_T*1`($sp)
 286         mtlr    r0
 287         addi    $sp,$sp,$FRAME
 288         blr
 289         .long   0
 290         .byte   0,12,4,1,0x80,18,3,0
 291         .long   0
 292 ___
 293
 294 # This is private block function, which uses tailored calling
 295 # interface, namely upon entry SHA_CTX is pre-loaded to given
 296 # registers and counter register contains amount of chunks to
 297 # digest...
 298 $code.=<<___;
 299 .align  4
 300 Lsha1_block_private:
 301 ___
 302 $code.=<<___;   # load K_00_19
 303         lis     $K,0x5a82
 304         ori     $K,$K,0x7999
 305 ___
 306 for($i=0;$i<20;$i++)    { &BODY_00_19($i,@V); unshift(@V,pop(@V)); }
 307 $code.=<<___;   # load K_20_39
 308         lis     $K,0x6ed9
 309         ori     $K,$K,0xeba1
 310 ___
 311 for(;$i<40;$i++)        { &BODY_20_39($i,@V); unshift(@V,pop(@V)); }
 312 $code.=<<___;   # load K_40_59
 313         lis     $K,0x8f1b
 314         ori     $K,$K,0xbcdc
 315 ___
 316 for(;$i<60;$i++)        { &BODY_40_59($i,@V); unshift(@V,pop(@V)); }
 317 $code.=<<___;   # load K_60_79
 318         lis     $K,0xca62
 319         ori     $K,$K,0xc1d6
 320 ___
 321 for(;$i<80;$i++)        { &BODY_20_39($i,@V); unshift(@V,pop(@V)); }
 322 $code.=<<___;
 323         add     r16,r16,$E
 324         add     r17,r17,$T
 325         add     r18,r18,$A
 326         add     r19,r19,$B
 327         add     r20,r20,$C
 328         stw     r16,0($ctx)
 329         mr      $A,r16
 330         stw     r17,4($ctx)
 331         mr      $B,r17
 332         stw     r18,8($ctx)
 333         mr      $C,r18
 334         stw     r19,12($ctx)
 335         mr      $D,r19
 336         stw     r20,16($ctx)
 337         mr      $E,r20
 338         addi    $inp,$inp,`16*4`
 339         bdnz    Lsha1_block_private
 340         blr
 341         .long   0
 342         .byte   0,12,0x14,0,0,0,0,0
 343 .size   .sha1_block_data_order,.-.sha1_block_data_order
 344 ___
 345 $code.=<<___;
 346 .asciz  "SHA1 block transform for PPC, CRYPTOGAMS by <appro\@fy.chalmers.se>"
 347 ___
 348
 349 $code =~ s/\`([^\`]*)\`/eval $1/gem;
 350 print $code;
 351 close STDOUT;