generic-library/openssl
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

_x86_64_AES_[en|de]crypt_compact: size optimization and aggressive

Browse Source 
T[ed]4 prefetch.
This commit is contained in:
Andy Polyakov 2007-07-19 14:29:41 +00:00
parent 0d7dba92c8
commit 8bae7722a2
1 changed files with 53 additions and 54 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

107
crypto/aes/asm/aes-x86_64.pl
View File

@ -354,7 +354,7 @@ ___
# it's possible to implement this by shifting tN by 8, filling least # it's possible to implement this by shifting tN by 8, filling least
# significant byte with byte load and finally bswap-ing at the end, # significant byte with byte load and finally bswap-ing at the end,
# but partial byte load kills Core 2... # but such partial register load kills Core 2...
sub enccompactvert() sub enccompactvert()
{ my ($t3,$t4,$t5)=("%r8d","%r9d","%r13d"); { my ($t3,$t4,$t5)=("%r8d","%r9d","%r13d");
@ -525,15 +525,18 @@ $code.=<<___;
rol \$24,$s3 rol \$24,$s3
xor $r20,$s2 xor $r20,$s2
xor $r21,$s3 xor $r21,$s3
mov 0($sbox),$acc0 # prefetch Te4
ror \$16,$t2 ror \$16,$t2
ror \$16,$t3 ror \$16,$t3
mov 64($sbox),$acc1
xor $t2,$s2 xor $t2,$s2
xor $t3,$s3 xor $t3,$s3
mov 128($sbox),$r20
ror \$8,$t2 ror \$8,$t2
ror \$8,$t3 ror \$8,$t3
mov 192($sbox),$r21
xor $t2,$s2 xor $t2,$s2
xor $t3,$s3 xor $t3,$s3
___ ___
} }
@ -541,38 +544,34 @@ $code.=<<___;
.type _x86_64_AES_encrypt_compact,\@abi-omnipotent .type _x86_64_AES_encrypt_compact,\@abi-omnipotent
.align 16 .align 16
_x86_64_AES_encrypt_compact: _x86_64_AES_encrypt_compact:
lea 128($sbox),$acc0 # size optimization lea 128($sbox),$inp # size optimization
mov 0-128($acc0),$acc1 # prefetch Te4 mov 0-128($inp),$acc1 # prefetch Te4
mov 32-128($acc0),$acc2 mov 32-128($inp),$acc2
mov 64-128($acc0),$t0 mov 64-128($inp),$t0
mov 96-128($acc0),$t1 mov 96-128($inp),$t1
mov 128-128($acc0),$acc1 mov 128-128($inp),$acc1
mov 160-128($acc0),$acc2 mov 160-128($inp),$acc2
mov 192-128($acc0),$t0 mov 192-128($inp),$t0
mov 224-128($acc0),$t1 mov 224-128($inp),$t1
xor 0($key),$s0 # xor with key
xor 4($key),$s1
xor 8($key),$s2
xor 12($key),$s3
lea 16($key),$key
jmp .Lenc_loop_compact jmp .Lenc_loop_compact
.align 16 .align 16
.Lenc_loop_compact: .Lenc_loop_compact:
___ xor 0($key),$s0 # xor with key
&enccompactvert();
&enctransform();
$code.=<<___;
xor 0($key),$s0
xor 4($key),$s1 xor 4($key),$s1
xor 8($key),$s2 xor 8($key),$s2
xor 12($key),$s3 xor 12($key),$s3
lea 16($key),$key lea 16($key),$key
cmp 16(%rsp),$key
jne .Lenc_loop_compact
___ ___
&enccompactvert(); &enccompactvert();
$code.=<<___; $code.=<<___;
cmp 16(%rsp),$key
je .Lenc_compact_done
___
&enctransform();
$code.=<<___;
jmp .Lenc_loop_compact
.align 16
.Lenc_compact_done:
xor 0($key),$s0 xor 0($key),$s0
xor 4($key),$s1 xor 4($key),$s1
xor 8($key),$s2 xor 8($key),$s2
@ -1003,6 +1002,7 @@ ___
sub dectransform() sub dectransform()
{ my ($tp10,$tp20,$tp40,$tp80,$acc0)=("%rax","%r8", "%r9", "%r10","%rbx"); { my ($tp10,$tp20,$tp40,$tp80,$acc0)=("%rax","%r8", "%r9", "%r10","%rbx");
my ($tp18,$tp28,$tp48,$tp88,$acc8)=("%rcx","%r11","%r12","%r13","%rdx"); my ($tp18,$tp28,$tp48,$tp88,$acc8)=("%rcx","%r11","%r12","%r13","%rdx");
my $prefetch = shift;
$code.=<<___; $code.=<<___;
mov $tp10,$acc0 mov $tp10,$acc0
@ -1107,17 +1107,21 @@ $code.=<<___;
xor `&LO("$tp80")`,`&LO("$acc0")` xor `&LO("$tp80")`,`&LO("$acc0")`
xor `&LO("$tp88")`,`&LO("$acc8")` xor `&LO("$tp88")`,`&LO("$acc8")`
`"mov 0($sbox),$mask80" if ($prefetch)`
shr \$32,$tp20 shr \$32,$tp20
shr \$32,$tp28 shr \$32,$tp28
`"mov 64($sbox),$maskfe" if ($prefetch)`
rol \$16,`&LO("$tp40")` # ROTATE(tp4^tp1^tp8,16) rol \$16,`&LO("$tp40")` # ROTATE(tp4^tp1^tp8,16)
rol \$16,`&LO("$tp48")` # ROTATE(tp4^tp1^tp8,16) rol \$16,`&LO("$tp48")` # ROTATE(tp4^tp1^tp8,16)
`"mov 128($sbox),$mask1b" if ($prefetch)`
rol \$16,`&LO("$tp20")` # ROTATE(tp4^tp1^tp8,16) rol \$16,`&LO("$tp20")` # ROTATE(tp4^tp1^tp8,16)
rol \$16,`&LO("$tp28")` # ROTATE(tp4^tp1^tp8,16) rol \$16,`&LO("$tp28")` # ROTATE(tp4^tp1^tp8,16)
`"mov 192($sbox),$tp80" if ($prefetch)`
xor `&LO("$tp40")`,`&LO("$tp10")` xor `&LO("$tp40")`,`&LO("$tp10")`
xor `&LO("$tp48")`,`&LO("$tp18")` xor `&LO("$tp48")`,`&LO("$tp18")`
`"mov 256($sbox),$tp88" if ($prefetch)`
xor `&LO("$tp20")`,`&LO("$acc0")` xor `&LO("$tp20")`,`&LO("$acc0")`
xor `&LO("$tp28")`,`&LO("$acc8")` xor `&LO("$tp28")`,`&LO("$acc8")`
___ ___
} }
@ -1125,28 +1129,30 @@ $code.=<<___;
.type _x86_64_AES_decrypt_compact,\@abi-omnipotent .type _x86_64_AES_decrypt_compact,\@abi-omnipotent
.align 16 .align 16
_x86_64_AES_decrypt_compact: _x86_64_AES_decrypt_compact:
lea 128($sbox),$acc0 # size optimization lea 128($sbox),$inp # size optimization
mov 0-128($acc0),$acc1 # prefetch Td4 mov 0-128($inp),$acc1 # prefetch Td4
mov 32-128($acc0),$acc2 mov 32-128($inp),$acc2
mov 64-128($acc0),$t0 mov 64-128($inp),$t0
mov 96-128($acc0),$t1 mov 96-128($inp),$t1
mov 128-128($acc0),$acc1 mov 128-128($inp),$acc1
mov 160-128($acc0),$acc2 mov 160-128($inp),$acc2
mov 192-128($acc0),$t0 mov 192-128($inp),$t0
mov 224-128($acc0),$t1 mov 224-128($inp),$t1
jmp .Ldec_loop_compact
xor 0($key),$s0 # xor with key
xor 4($key),$s1
xor 8($key),$s2
xor 12($key),$s3
lea 16($key),$key
jmp .Ldec_compact_loop
.align 16 .align 16
.Ldec_compact_loop: .Ldec_loop_compact:
xor 0($key),$s0 # xor with key
xor 4($key),$s1
xor 8($key),$s2
xor 12($key),$s3
lea 16($key),$key
___ ___
&deccompactvert(); &deccompactvert();
$code.=<<___; $code.=<<___;
cmp 16(%rsp),$key
je .Ldec_compact_done
mov 256+0($sbox),$mask80 mov 256+0($sbox),$mask80
shl \$32,%rbx shl \$32,%rbx
shl \$32,%rdx shl \$32,%rdx
@ -1155,18 +1161,11 @@ $code.=<<___;
or %rdx,%rcx or %rdx,%rcx
mov 256+16($sbox),$mask1b mov 256+16($sbox),$mask1b
___ ___
&dectransform(); &dectransform(1);
$code.=<<___;
xor 0($key),$s0
xor 4($key),$s1
xor 8($key),$s2
xor 12($key),$s3
lea 16($key),$key
cmp 16(%rsp),$key
jne .Ldec_compact_loop
___
&deccompactvert();
$code.=<<___; $code.=<<___;
jmp .Ldec_loop_compact
.align 16
.Ldec_compact_done:
xor 0($key),$s0 xor 0($key),$s0
xor 4($key),$s1 xor 4($key),$s1
xor 8($key),$s2 xor 8($key),$s2