renaming to AVR-Crypto-Lib

[avr-crypto-lib.git] / noekeon_asm.S

diff --git a/noekeon_asm.S b/noekeon_asm.S
index 43ffa93a33f9a39714bfe8a5a420ecd144ecb311..7fc05069005d2acb79a2c847609017bffadaea0a 100644 (file)
--- a/noekeon_asm.S
+++ b/noekeon_asm.S
@@ -1,3 +1,21 @@
+/* noekeon_asm.S */
+/*
+    This file is part of the This file is part of the AVR-Crypto-Lib.
+    Copyright (C) 2008  Daniel Otte (daniel.otte@rub.de)
+
+    This program is free software: you can redistribute it and/or modify
+    it under the terms of the GNU General Public License as published by
+    the Free Software Foundation, either version 3 of the License, or
+    (at your option) any later version.
+
+    This program is distributed in the hope that it will be useful,
+    but WITHOUT ANY WARRANTY; without even the implied warranty of
+    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+    GNU General Public License for more details.
+
+    You should have received a copy of the GNU General Public License
+    along with this program.  If not, see <http://www.gnu.org/licenses/>.
+*/
 /* 
  * noekeon assembler implementation for avr
  * author: Daniel Otte
@@ -26,13 +44,9 @@
        push r17
        push r28
        push r29
-       in r28, _SFR_IO_ADDR(SREG)
-       push r28
 .endm
 
 .macro pop_all
-       pop r28
-       out _SFR_IO_ADDR(SREG), r28
        pop r29
        pop r28
        pop r17
@@ -54,6 +68,18 @@
        clr r1
 .endm
 
+push_all_func:
+       pop r31
+       pop r30
+       push_all
+       ijmp
+
+pop_all_func:
+    pop r31
+    pop r30
+    pop_all
+    ijmp
+
 .macro xchg a b
        eor \a, \b
        eor \b, \a
@@ -83,7 +109,6 @@
        \op \p\()\d, \q\()\z
 .endm
 
-.global bigendian_rotl32
 ; === bigendian_rotl32 ===
 ; this function rotates a 32bit bigendian word n bits to the left
 ;  param1: the 32-bit value
@@ -94,7 +119,6 @@
 ;   given in r25,r24,r23,r22   
 
 bigendian_rotl32:
-       in r0, _SFR_IO_ADDR(SREG)
        /* copy high bit of r22 to carry */
        mov r1, r22
 2:
@@ -109,13 +133,11 @@ bigendian_rotl32:
        brne 2b
 bigendian_rotl32_exit:
        clr r1
-       out _SFR_IO_ADDR(SREG), r0
        ret
        
        
 /******************************************************************************/
 
-.global bigendian_rotr32
 ; === bigendian_rotl32 ===
 ; this function rotates a 32bit bigendian word n bits to the right
 ;  param1: the 32-bit value
@@ -126,7 +148,6 @@ bigendian_rotl32_exit:
 ;   given in r25,r24,r23,r22   
 
 bigendian_rotr32:
-       in r0, _SFR_IO_ADDR(SREG)
        /* copy high bit of r25 to carry */
 
        mov r1, r25
@@ -141,7 +162,6 @@ bigendian_rotr32:
        brne 2b
 bigendian_rotr32_exit:
        clr r1
-       out _SFR_IO_ADDR(SREG), r0
        ret
 
 /******************************************************************************/
@@ -267,14 +287,15 @@ theta:
        ret
 
 /******************************************************************************/
+#ifndef NOEKEON_NO_ENC
 ; === noekeon_enc ===
 ;
-;  param1: pointer to buffer/state (r24,r25)
+;  param1: pointer to buffer (r24,r25)
 ;  param2: pointer to k (r22,r23) 
 ;
 .global noekeon_enc
 noekeon_enc:
-       push_all
+       rcall push_all_func
        /* load state */
        movw r26, r22
        ldi r28, 2
@@ -304,7 +325,7 @@ noekeon_enc:
        lpm r0, Z
        push r0
 3:
-       call round /* pops rc2 & rc1 */
+       rcall round /* pops rc2 & rc1 */
        pop r22
        dec r22
        push r22
@@ -314,7 +335,7 @@ noekeon_enc:
 
        ldi r22, 0xD4
        eor state0_3, r22
-       call theta
+       rcall theta
 
        pop r31
        pop r30
@@ -327,10 +348,13 @@ noekeon_enc:
        dec r22
        brne 1b
        
-       pop_all
+       rcall pop_all_func
        ret
+#endif
 /******************************************************************************/
 /******************************************************************************/
+#ifndef NOEKEON_NO_DEC
+
 ; === noekeon_dec ===
 ;
 ;  param1: pointer to buffer/state (r24,r25)
@@ -338,7 +362,7 @@ noekeon_enc:
 ;
 .global noekeon_dec
 noekeon_dec:
-       push_all
+       rcall push_all_func
        /* allocate 16 bytes on the stack */
        in r30, _SFR_IO_ADDR(SPL)
        in r31, _SFR_IO_ADDR(SPH)
@@ -369,7 +393,7 @@ noekeon_dec:
 
        movw r26, r30
        sbiw r26, 16 /* set X back to begining of stack key */
-       call theta
+       rcall theta
        
        /* mov state to stackkey */
        clr r29
@@ -411,7 +435,7 @@ noekeon_dec:
        push r0
        push r1
 3:
-       call round /* pops rc2 & rc1 */
+       rcall round /* pops rc2 & rc1 */
        pop r22
        dec r22
        push r22
@@ -419,7 +443,7 @@ noekeon_dec:
 ;----
        pop r22
 
-       call theta
+       rcall theta
        ldi r22, 0x80
        eor state0_3, r22
        
@@ -442,25 +466,28 @@ write_state_back:
        adiw r30, 16 
        out  _SFR_IO_ADDR(SPH), r31
        out  _SFR_IO_ADDR(SPL), r30
-       pop_all
+       rcall pop_all_func
        ret
+#endif
 /******************************************************************************/
        
+       
 round: 
        pop r24
        pop r25
        pop r1
        eor state0_3, r1
-       call theta
+       rcall theta
        pop r1
        eor state0_3, r1
        push r25
        push r24
 pi_gamma_pi:
-       clc
-       call pi
+       ldi r30, pm_lo8(bigendian_rotl32)
+       ldi r31, pm_hi8(bigendian_rotl32)
+       rcall pi
        /* pi1 done; now gamma */
-       call gamma_1
+       rcall gamma_1
        /* a[0] <-> a[3] */
        xchg state0_0, state3_0
        xchg state0_1, state3_1
@@ -470,23 +497,11 @@ pi_gamma_pi:
        op32 eor, state2, state0
        op32 eor, state2, state1
        op32 eor, state2, state3
-/*     
-       eor state2_0, state0_0
-       eor state2_1, state0_1
-       eor state2_2, state0_2
-       eor state2_3, state0_3
-       eor state2_0, state1_0
-       eor state2_1, state1_1
-       eor state2_2, state1_2
-       eor state2_3, state1_3
-       eor state2_0, state3_0
-       eor state2_1, state3_1
-       eor state2_2, state3_2
-       eor state2_3, state3_3
-*/
-       call gamma_1
-       sec
-       call pi
+
+       rcall gamma_1
+       ldi r30, pm_lo8(bigendian_rotr32)
+       ldi r31, pm_hi8(bigendian_rotr32)
+       rcall pi
        ret
        
 gamma_1:
@@ -529,17 +544,7 @@ gamma_1:
        eor state0_3, r1
        ret
        
-pi:
-       brcs 1f
-       ldi r30, lo8(bigendian_rotl32)
-       ldi r31, hi8(bigendian_rotl32)
-       rjmp 2f
-1:
-       ldi r30, lo8(bigendian_rotr32)
-       ldi r31, hi8(bigendian_rotr32)
-2:     
-       lsr r31
-       ror r30
+pi:    
        /* a[1] <<<= 1*/
        mov r22, state1_0
        mov r23, state1_1
@@ -586,6 +591,9 @@ void noekeon_init(void* key, noekeon_ctx_t* ctx){
        noekeon_enc(ctx, nullv);
 }
 */
+
+#ifndef NOEKEON_NO_INIT
+
 .global noekeon_init
 noekeon_init:
 ; === noekeon_init ===
@@ -619,7 +627,7 @@ noekeon_init:
        
        sbiw r26, 16
        movw r24, r26
-       call noekeon_enc
+       rcall noekeon_enc
        
        in r30, _SFR_IO_ADDR(SPL)
        in r31, _SFR_IO_ADDR(SPH)
@@ -628,6 +636,6 @@ noekeon_init:
        out  _SFR_IO_ADDR(SPL), r30     
        ret
        
-
+#endif