From: bg <bg@b1d182e4-1ff8-0310-901f-bddb46175740>
Date: Mon, 27 Apr 2009 17:47:36 +0000 (+0000)
Subject: more shabal
X-Git-Url: https://git.cryptolib.org/?p=avr-crypto-lib.git;a=commitdiff_plain;h=600e82c981df8a6e446c6469e7e6be627ec027de

more shabal
---

diff --git a/host/shavs_test.rb b/host/shavs_test.rb
index 74d02c8..225c381 100644
--- a/host/shavs_test.rb
+++ b/host/shavs_test.rb
@@ -57,13 +57,14 @@ def send_md(md_string)
     $sp.print(md_string[i].chr)
 #	print("DBG s: "+ md_string[i].chr) if $debug
 	if(i%20==19)
-		sleep(0.1)
+#		sleep(0.1)
 	end		
   end
 end
 
 def run_test(filename)
   errors = 0
+  line=1
   if not File.exist?(filename)
   	puts("ERROR file "+filename+" does not exist!")
   end
@@ -91,7 +92,8 @@ def run_test(filename)
 	b = (/[\s]*MD[\s]*=[\s]*([0-9a-fA-F]*).*/.match(avr_md))[1];
 	a.upcase!
 	b.upcase!
-	puts("") if (pos%$linewidth==0 and $linewidth!=0)
+	printf("\n%4d: ", line) if (pos%$linewidth==0 and $linewidth!=0)
+	line += 1               if (pos%$linewidth==0 and $linewidth!=0)
 	#putc((a==b)?'*':'!')
 	if(a==b)
 	  putc('*')
@@ -131,9 +133,9 @@ init_system()
 for i in (5..(ARGV.size-1))
   errors = run_test(ARGV[i])
   if errors == 0
-    puts("[ok]")
+    puts("\n[ok]")
   else
-    puts("[errors: "+errors.to_s+"]")
+    puts("\n[errors: "+errors.to_s+"]")
   end
 end
  $sp.print("EXIT\r");
diff --git a/mkfiles/shabal.mk b/mkfiles/shabal.mk
index 5034854..bc12041 100644
--- a/mkfiles/shabal.mk
+++ b/mkfiles/shabal.mk
@@ -1,11 +1,12 @@
 # Makefile for Skein
-ALGO_NAME := SHABAL_C
+ALGO_NAME := SHABAL
 
 # comment out the following line for removement of Skein from the build process
 HASHES += $(ALGO_NAME)
 
 
-$(ALGO_NAME)_OBJ      := shabal.o
+$(ALGO_NAME)_OBJ      := shabal-asm.o shabal192-asm.o shabal224-asm.o \
+                         shabal256-asm.o shabal384-asm.o shabal512-asm.o
 $(ALGO_NAME)_TEST_BIN := main-shabal-test.o debug.o uart.o hexdigit_tab.o  \
                          dbz_strings.o nessie_common.o cli.o string-extras.o performance_test.o \
 			 nessie_hash_test.o hfal-basic.o hfal_shabal.o shavs.o
diff --git a/mkfiles/shabal_c.mk b/mkfiles/shabal_c.mk
new file mode 100644
index 0000000..9521743
--- /dev/null
+++ b/mkfiles/shabal_c.mk
@@ -0,0 +1,14 @@
+# Makefile for Skein
+ALGO_NAME := SHABAL_C
+
+# comment out the following line for removement of Skein from the build process
+HASHES += $(ALGO_NAME)
+
+
+$(ALGO_NAME)_OBJ      := shabal.o shabal192.o shabal224.o shabal256.o shabal384.o shabal512.o
+$(ALGO_NAME)_TEST_BIN := main-shabal-test.o debug.o uart.o hexdigit_tab.o  \
+                         dbz_strings.o nessie_common.o cli.o string-extras.o performance_test.o \
+			 nessie_hash_test.o hfal-basic.o hfal_shabal.o shavs.o
+$(ALGO_NAME)_NESSIE_TEST      := test nessie
+$(ALGO_NAME)_PERFORMANCE_TEST := performance
+
diff --git a/shabal-asm.S b/shabal-asm.S
new file mode 100644
index 0000000..31879b3
--- /dev/null
+++ b/shabal-asm.S
@@ -0,0 +1,739 @@
+/* shabal-asm.S */
+/*
+    This file is part of the AVR-Crypto-Lib.
+    Copyright (C) 2009  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/>.
+*/
+/*
+ * \file    shabal-asm.S
+ * \author  Daniel Otte
+ * \email   daniel.otte@rub.de
+ * \date    2009-04-27
+ * \license GPLv3 or later
+ */ 
+
+#include "avr-asm-macros.S"
+
+/******************************************************************************/
+/*
+void shabal_p(shabal_ctx_t* ctx, const void* m){
+	uint8_t i,j;
+	for(i=0;i<16;++i){
+		ctx->b[i] = ROTL32(ctx->b[i],17);
+	}
+	for(j=0;j<SHABAL_P;++j){
+		for(i=0;i<16;++i){
+			ctx->a[(i+16*j)%SHABAL_R] =
+				shabal_u(ctx->a[(i+16*j)%SHABAL_R] 
+					^ shabal_v(ROTL32(ctx->a[(i+16*j+SHABAL_R-1)%SHABAL_R],15))
+				    ^ ctx->c[(8-i+16)%16])
+				^ ctx->b[(i+SHABAL_O1)%16]
+				^ ((ctx->b[(i+SHABAL_O2)%16]) & ~(ctx->b[(i+SHABAL_O3)%16]))
+				^ ((uint32_t*)m)[i];
+			ctx->b[i] = ROTL32(ctx->b[i], 1) ^ ~(ctx->a[(i+16*j)%SHABAL_R]);
+		}
+	}
+	
+	for(j=0;j<36;++j){
+		ctx->a[j%SHABAL_R] += ctx->c[(j+3)%16];
+	}
+}
+*/
+MB0 =  2
+MB1 =  3
+AB0 =  4
+AB1 =  5
+BB0 =  6
+BB1 =  7
+CB0 =  8
+CB1 =  9
+AL0 = 10
+AL1 = 11
+AL2 = 12
+AL3 = 13
+A0  = 14
+A1  = 15
+A2  = 16
+A3  = 17
+B0  = 18
+B1  = 19
+B2  = 20
+B3  = 21
+I   = 22
+J   = 23
+T0  = 26
+T1  = 27
+T2  = 28
+T3  = 29
+/*
+ * param ctx: r24:r25
+ * param m:   r22:r23 
+ */
+; .global shabal_p
+shabal_p:
+	push_range 2, 17
+	push r28
+	push r29
+	movw MB0, r22
+	movw r30, r24
+	adiw r30, 8
+	ld BB0, Z+
+	ld BB1, Z+
+	ld CB0, Z+
+	ld CB1, Z+
+	movw AB0, r30
+	movw r30, BB0
+	adiw r30, 16*4-1
+	adiw r30, 1
+	ldi r24, 16
+1:
+	ld A3, -Z
+	ld A2, -Z
+	ld A1, -Z
+	ld A0, -Z	
+	mov r0, A3
+	rol r0
+	rol A0
+	rol A1
+	rol A2
+	rol A3
+	std Z+0, A2
+	std Z+1, A3
+	std Z+2, A0
+	std Z+3, A1
+	dec r24
+	brne 1b
+	movw B0, A2
+	movw B2, A0
+	/* load ctx->a[(i+16*j-1)%12]*/
+	movw r26, AB0
+	adiw r26, 4*11
+	ld AL0, X+
+	ld AL1, X+
+	ld AL2, X+
+	ld AL3, X+
+	clr I
+	clr J
+1:
+	/* ROTL32(AL, 15)*/
+	movw T0, AL2
+	movw T2, AL0
+	mov r0, T0
+	ror r0
+	ror T3
+	ror T2
+	ror T1
+	ror T0
+	movw AL0, T0
+	movw AL2, T2
+	/* apply V to AL */
+	movw A0, AL0
+	movw A2, AL2
+	lsl A0
+	rol A1
+	rol A2
+	rol A3
+	lsl A0
+	rol A1
+	rol A2
+	rol A3
+	add A0, AL0
+	adc A1, AL1	
+	adc A2, AL2
+	adc A3, AL3
+	/* xor in ctx->c[(8-i+16)%16] */
+	ldi T0, 24
+	sub T0, I
+	andi T0, 0x0f
+	lsl T0
+	lsl T0
+	movw r30, CB0
+	add r30, T0
+	adc r31, r1
+	ld r0, Z+
+	eor A0, r0
+	ld r0, Z+
+	eor A1, r0
+	ld r0, Z+
+	eor A2, r0
+	ld r0, Z+
+	eor A3, r0
+	/* xor in ctx->a[(i+16*j)%12] */
+	mov T0, J
+	swap T0 /* *=16 */
+	add T0, I
+	ldi r30, lo8(mod12table)
+	ldi r31, hi8(mod12table)
+	add r30, T0
+	adc r31, r1
+	lpm T0, Z
+	movw r30, AB0
+	add r30, T0
+	adc r31, r1
+	movw T2, r30
+	ld r0, Z+
+	eor A0, r0
+	ld r0, Z+
+	eor A1, r0
+	ld r0, Z+
+	eor A2, r0
+	ld r0, Z+
+	eor A3, r0
+	/* AL = 3*A */
+	movw AL0, A0
+	movw AL2, A2
+	lsl AL0
+	rol AL1
+	rol AL2
+	rol AL3
+	add AL0, A0
+	adc AL1, A1
+	adc AL2, A2
+	adc AL3, A3
+	/* xor in ctx->b[(i+13)%16] */
+	ldi T0, 13
+	add T0, I
+	andi T0, 0x0f
+	lsl T0
+	lsl T0
+	movw r30, BB0
+	add r30, T0
+	adc r31, r1
+	ld r0, Z+
+	eor AL0, r0
+	ld r0, Z+
+	eor AL1, r0
+	ld r0, Z+
+	eor AL2, r0
+	ld r0, Z+
+	eor AL3, r0
+	/* load ctx->b[(i+9)%16] into A */
+	ldi T0, 9
+	add T0, I
+	andi T0, 0x0f
+	lsl T0
+	lsl T0
+	movw r30, BB0
+	add r30, T0
+	adc r31, r1
+	ld A0, Z+
+	ld A1, Z+
+	ld A2, Z+
+	ld A3, Z+
+	/* and in ~(ctx->b[(i+6)%16]) */
+	ldi T0, 6
+	add T0, I
+	andi T0, 0x0f
+	lsl T0
+	lsl T0
+	movw r30, BB0
+	add r30, T0
+	adc r31, r1
+	ld r0, Z+
+	com r0
+	and A0, r0
+	ld r0, Z+
+	com r0
+	and A1, r0
+	ld r0, Z+
+	com r0
+	and A2, r0
+	ld r0, Z+
+	com r0
+	and A3, r0
+	/* xor A into AL */
+	eor AL0, A0
+	eor AL1, A1
+	eor AL2, A2
+	eor AL3, A3
+	/* xor m[i] into AL */
+	mov T0, I
+	lsl T0
+	lsl T0
+	movw r30, MB0
+	add r30, T0
+	adc r31, r1
+	ld r0, Z+
+	eor AL0, r0
+	ld r0, Z+
+	eor AL1, r0
+	ld r0, Z+
+	eor AL2, r0
+	ld r0, Z+
+	eor AL3, r0
+	/* A (AL) is done, now store it */
+	movw r30, T2
+	st Z+, AL0
+	st Z+, AL1
+	st Z+, AL2
+	st Z+, AL3
+	/* process ctx->b[i] */
+	/* ROTL32(b, 1)*/
+	mov r0, B3
+	rol r0
+	rol B0
+	rol B1
+	rol B2
+	rol B3
+	/* xor in ~(ctx->a[(i+16*j)%SHABAL_R]) */
+	movw A0, AL0
+	movw A2, AL2
+	com A0
+	com A1
+	com A2
+	com A3
+	eor B0, A0
+	eor B1, A1
+	eor B2, A2
+	eor B3, A3
+	/* store B */
+	movw r30, BB0
+	mov T0, I
+	lsl T0
+	lsl T0
+	add r30, T0
+	adc r31, r1
+	st Z+, B0
+	st Z+, B1
+	st Z+, B2
+	st Z+, B3
+	inc I
+	cpi I, 16
+	brne local_reload
+	inc J
+	cpi J, 3
+	brne global_reload
+	rjmp addition
+global_reload:
+	clr I
+local_reload:	
+	mov T0, I
+	lsl T0
+	lsl T0
+	movw r30, BB0 
+	add r30, T0
+	adc r31, r1
+	ld B0, Z+
+	ld B1, Z+
+	ld B2, Z+
+	ld B3, Z+
+	
+	rjmp 1b
+addition:
+	clr J
+	movw r30, AB0
+	movw r26, CB0
+	adiw r26, 3*4
+1:
+	/* J = 0..11 */
+	ld AL0, X+
+	ld A0, Z
+	add A0, AL0
+	st Z+, A0
+	ld AL0, X+
+	ld A0, Z
+	adc A0, AL0
+	st Z+, A0
+	ld AL0, X+
+	ld A0, Z
+	adc A0, AL0
+	st Z+, A0
+	ld AL0, X+
+	ld A0, Z
+	adc A0, AL0
+	st Z+, A0
+	inc J
+	cpi J, 12
+	brne 1b
+	/* J = 12 */
+	movw r30, AB0
+	ld AL0, X+
+	ld A0, Z
+	add A0, AL0
+	st Z+, A0
+	ld AL0, X+
+	ld A0, Z
+	adc A0, AL0
+	st Z+, A0
+	ld AL0, X+
+	ld A0, Z
+	adc A0, AL0
+	st Z+, A0
+	ld AL0, X+
+	ld A0, Z
+	adc A0, AL0
+	st Z+, A0
+	inc J
+	/* J= 13..23*/
+	movw r26, CB0
+1:	
+	ld AL0, X+
+	ld A0, Z
+	add A0, AL0
+	st Z+, A0
+	ld AL0, X+
+	ld A0, Z
+	adc A0, AL0
+	st Z+, A0
+	ld AL0, X+
+	ld A0, Z
+	adc A0, AL0
+	st Z+, A0
+	ld AL0, X+
+	ld A0, Z
+	adc A0, AL0
+	st Z+, A0
+	inc J
+	cpi J, 24
+	brne 1b
+	/* J= 24..28*/
+	movw r30, AB0
+1:	
+	ld AL0, X+
+	ld A0, Z
+	add A0, AL0
+	st Z+, A0
+	ld AL0, X+
+	ld A0, Z
+	adc A0, AL0
+	st Z+, A0
+	ld AL0, X+
+	ld A0, Z
+	adc A0, AL0
+	st Z+, A0
+	ld AL0, X+
+	ld A0, Z
+	adc A0, AL0
+	st Z+, A0
+	inc J
+	cpi J, 29
+	brne 1b
+
+	/* J= 29..35*/
+	movw r26, CB0
+1:	
+	ld AL0, X+
+	ld A0, Z
+	add A0, AL0
+	st Z+, A0
+	ld AL0, X+
+	ld A0, Z
+	adc A0, AL0
+	st Z+, A0
+	ld AL0, X+
+	ld A0, Z
+	adc A0, AL0
+	st Z+, A0
+	ld AL0, X+
+	ld A0, Z
+	adc A0, AL0
+	st Z+, A0
+	inc J
+	cpi J, 36
+	brne 1b	
+exit:	
+	pop r29
+	pop r28
+	pop_range 2, 17
+	ret
+
+mod12table:
+	.byte  0,  4,  8, 12, 16, 20, 24, 28
+	.byte 32, 36, 40, 44,  0,  4,  8, 12
+	.byte 16, 20, 24, 28, 32, 36, 40, 44
+	.byte  0,  4,  8, 12, 16, 20, 24, 28
+	.byte 32, 36, 40, 44,  0,  4,  8, 12
+	.byte 16, 20, 24, 28, 32, 36, 40, 44
+	
+/******************************************************************************/
+/*
+void shabal_nextBlock(shabal_ctx_t* ctx, const void* block){
+	uint8_t i;
+	uint32_t* t;
+	for(i=0;i<16;++i){
+		ctx->b[i] += ((uint32_t*)block)[i];
+	}
+	ctx->a[0] ^= ctx->w.w32[0];
+	ctx->a[1] ^= ctx->w.w32[1];
+	shabal_p(ctx, block);
+	for(i=0;i<16;++i){
+		ctx->c[i] -= ((uint32_t*)block)[i];
+	}
+	ctx->w.w64++;
+	t = ctx->c;
+	ctx->c = ctx->b;
+	ctx->b = t;
+}
+*/
+/*
+ * param ctx:    r24:r25
+ * param block:  r22:r23
+ */
+MB0  = 14
+MB1  = 15
+CTX0 = 16
+CTX1 = 17
+.global shabal_nextBlock
+shabal_nextBlock:
+	push_range 14, 17
+	movw CTX0, r24
+	movw MB0, r22
+	/* xor W into A and increment W */
+	movw r30, CTX0
+	ldi r19, 8
+	sec
+1:	
+	ld r20, Z
+	ldd r21, Z+(8+4)
+	eor r21, r20
+	std Z+(8+4), r21
+	adc r20, r1
+	st Z+, r20 
+	dec r19
+	brne 1b
+	/* add block to ctx->b */
+	ld r26, Z+
+	ld r27, Z
+	movw r30, MB0
+	ldi r19, 16
+1:	
+	ld r0, X
+	ld r18, Z+
+	add r0, r18
+	st X+, r0
+	ld r0, X
+	ld r18, Z+
+	adc r0, r18
+	st X+, r0
+	ld r0, X
+	ld r18, Z+
+	adc r0, r18
+	st X+, r0
+	ld r0, X
+	ld r18, Z+
+	adc r0, r18
+	st X+, r0
+	dec r19
+	brne 1b
+	/* call shbal_p */
+	rcall shabal_p
+	/* sub block from ctx->c */
+	movw r30, CTX0
+	adiw r30, 8+2
+	ld r26, Z+
+	ld r27, Z
+	movw r30, MB0
+	ldi r19, 16
+1:	
+	ld r0, X
+	ld r18, Z+
+	sub r0, r18
+	st X+, r0
+	ld r0, X
+	ld r18, Z+
+	sbc r0, r18
+	st X+, r0
+	ld r0, X
+	ld r18, Z+
+	sbc r0, r18
+	st X+, r0
+	ld r0, X
+	ld r18, Z+
+	sbc r0, r18
+	st X+, r0
+	dec r19
+	brne 1b
+	/* xchange ctx->b with ctx->c*/
+	movw r30, CTX0
+	ldd r22, Z+8
+	ldd r23, Z+9
+	ldd r24, Z+10
+	ldd r25, Z+11
+	std Z+10, r22
+	std Z+11, r23
+	std Z+8,  r24
+	std Z+9,  r25
+	pop_range 14, 17
+	ret
+
+/******************************************************************************/
+/*
+void shabal_lastBlock(shabal_ctx_t* ctx, const void* block, uint16_t length_b){
+	uint8_t i,j;
+	uint32_t* t;
+	uint8_t buffer[64];
+	while(length_b>=SHABAL_BLOCKSIZE){
+		shabal_nextBlock(ctx, block);
+		block = (uint8_t*)block + SHABAL_BLOCKSIZE_B;
+		length_b -= SHABAL_BLOCKSIZE;
+	}
+	memset(buffer, 0, 64);
+	memcpy(buffer, block, (length_b+7)/8);
+	buffer[length_b/8] |= 0x80>>(length_b%8);
+	for(i=0;i<16;++i){
+		ctx->b[i] += ((uint32_t*)buffer)[i];
+	}
+	for(j=0; j<4;++j){
+		ctx->a[0] ^= ctx->w.w32[0];
+		ctx->a[1] ^= ctx->w.w32[1];
+		shabal_p(ctx, buffer);
+		t = ctx->c;
+		ctx->c = ctx->b;
+		ctx->b = t;
+	}	 
+}
+*/
+I    = 16
+LEN0 = 16
+LEN1 = 17
+CTX0 = 14
+CTX1 = 15
+MB0  = 12
+MB1  = 13
+/*
+ * param ctx:       r24:r25
+ * param block:     r22:r23
+ * param length_b:  r20:r21
+ */
+.global shabal_lastBlock
+shabal_lastBlock:
+	push_range 12, 17
+	movw CTX0, r24
+	movw MB0,  r22
+	movw LEN0, r20
+1:
+	cpi LEN1, 0x02
+	brlo 2f
+	movw r24, CTX0
+	movw r22, MB0
+	rcall shabal_nextBlock
+	subi LEN1, 0x02
+	ldi r18, 64
+	add	MB0, r18
+	adc MB1, r1
+	rjmp 1b
+2:	
+	stack_alloc_large 64
+	adiw r30, 1 /* Z points at buffer */
+	movw r26, MB0
+	/* r24 = LEN/8*/
+	movw r24, LEN0
+	lsr r25
+	ror r24
+	lsr r24
+	lsr r24
+	ldi r25, 64-1
+	sub r25, r24
+	tst r24
+	breq 32f
+31:
+	ld r0, X+
+	st Z+, r0
+	dec r24
+	brne 31b
+32:
+	ldi r18, 0x80
+	andi LEN0, 0x07
+	breq append_0x80
+	ld r0, X+
+33:
+	lsr r18
+	dec LEN0
+	brne 33b
+	or r0, r18
+	st Z+, r0
+	rjmp append_zeros
+append_0x80:
+	st Z+, r18
+append_zeros:
+	tst r25
+	breq 4f
+34: st Z+, r1
+	dec r25
+	brne 34b
+4:		
+	sbiw r30, 63
+	sbiw r30,  1
+	movw MB0, r30
+	movw r26, CTX0
+	adiw r26, 8
+	ld r24, X+
+	ld r25, X
+	movw r26, r24
+	ldi r18, 16
+41: 
+	ld r24, X
+	ld r25, Z+
+	add r24, r25
+	st X+, r24
+	ld r24, X
+	ld r25, Z+
+	adc r24, r25
+	st X+, r24	
+	ld r24, X
+	ld r25, Z+
+	adc r24, r25
+	st X+, r24	
+	ld r24, X
+	ld r25, Z+
+	adc r24, r25
+	st X+, r24	
+	dec r18
+	brne 41b
+	/* final loop */
+	ldi I, 4
+5:
+	/* xor W into A */
+	movw r30, CTX0
+	ldi r19, 8
+51:	
+	ld  r24, Z+
+	ldd r25, Z+(8+4-1)
+	eor r24, r25
+	std Z+(8+4-1), r24	
+	dec r19
+	brne 51b
+	movw r24, CTX0
+	movw r22, MB0
+	rcall shabal_p
+	movw r30, CTX0
+	ldd r22, Z+8
+	ldd r23, Z+9
+	ldd r24, Z+10
+	ldd r25, Z+11
+	std Z+10, r22
+	std Z+11, r23
+	std Z+8,  r24
+	std Z+9,  r25
+	dec I
+	brne 5b
+	
+	stack_free_large 64
+	pop_range 12, 17
+	ret
+
+	
+
+	
+	
+	
+	
+	
+	
+	
+	
+	
+	
+	
diff --git a/shabal.c b/shabal.c
index 1fa0f6d..f18908a 100644
--- a/shabal.c
+++ b/shabal.c
@@ -27,7 +27,6 @@
 
 #include <stdint.h>
 #include "shabal.h"
-#include <avr/pgmspace.h>
 #include <string.h>
 
 #define SHABAL_O1 13
@@ -53,193 +52,22 @@ void shabal_p(shabal_ctx_t* ctx, const void* m){
 	for(i=0;i<16;++i){
 		ctx->b[i] = ROTL32(ctx->b[i],17);
 	}
-	for(j=0;j<SHABAL_P;++j){
+	for(j=0;j<SHABAL_P;j++){
 		for(i=0;i<16;++i){
-			ctx->a[(i+16*j)%SHABAL_R] =
-				shabal_u(ctx->a[(i+16*j)%SHABAL_R] 
-					^ shabal_v(ROTL32(ctx->a[(i+16*j+SHABAL_R-1)%SHABAL_R],15))
+			ctx->a[(i+j*16)%SHABAL_R] = 
+				shabal_u(ctx->a[(i+j*16)%SHABAL_R] 
+					^ shabal_v(ROTL32(ctx->a[(i+j*16+SHABAL_R-1)%SHABAL_R],15))
 				    ^ ctx->c[(8-i+16)%16])
 				^ ctx->b[(i+SHABAL_O1)%16]
 				^ ((ctx->b[(i+SHABAL_O2)%16]) & ~(ctx->b[(i+SHABAL_O3)%16]))
 				^ ((uint32_t*)m)[i];
-			ctx->b[i] = ROTL32(ctx->b[i], 1) ^ ~(ctx->a[(i+16*j)%SHABAL_R]);
+			ctx->b[i] = ROTL32(ctx->b[i], 1) ^ ~(ctx->a[(i+j*16)%SHABAL_R]);
 		}
 	}
 	
 	for(j=0;j<36;++j){
 		ctx->a[j%SHABAL_R] += ctx->c[(j+3)%16];
 	}
-/*
-	for (j=0; j<36; j++){
-        ctx->a[(36 - 1 - j)%12] = 
-                ctx->a[(36 - 1 - j) % 12]
-                + ctx->c[(36 * 16 + 6 - j) % 16];
-
-	}
-*/
-}
-
-uint32_t shabal192_iv[] PROGMEM = {
-	/* A */
-	0xFD749ED4, 0xB798E530, 0x33904B6F, 0x46BDA85E,
-	0x076934B4, 0x454B4058, 0x77F74527, 0xFB4CF465,
-    0x62931DA9, 0xE778C8DB, 0x22B3998E, 0xAC15CFB9,
-	/* B */
-	0x58BCBAC4, 0xEC47A08E, 0xAEE933B2, 0xDFCBC824,
-	0xA7944804, 0xBF65BDB0, 0x5A9D4502, 0x59979AF7,
-    0xC5CEA54E, 0x4B6B8150, 0x16E71909, 0x7D632319,
-	0x930573A0, 0xF34C63D1, 0xCAF914B4, 0xFDD6612C,
-	/* C */
-	0x61550878, 0x89EF2B75, 0xA1660C46, 0x7EF3855B,
-	0x7297B58C, 0x1BC67793, 0x7FB1C723, 0xB66FC640,
-    0x1A48B71C, 0xF0976D17, 0x088CE80A, 0xA454EDF3,
-	0x1C096BF4, 0xAC76224B, 0x5215781C, 0xCD5D2669
-};
-
-
-uint32_t shabal224_iv[] PROGMEM = {
-	/* A */
-	0xA5201467, 0xA9B8D94A, 0xD4CED997, 0x68379D7B, 
-	0xA7FC73BA, 0xF1A2546B, 0x606782BF, 0xE0BCFD0F,
-	0x2F25374E, 0x069A149F, 0x5E2DFF25, 0xFAECF061,
-	/* B */ 
-	0xEC9905D8, 0xF21850CF, 0xC0A746C8, 0x21DAD498,
-	0x35156EEB, 0x088C97F2, 0x26303E40, 0x8A2D4FB5,
-    0xFEEE44B6, 0x8A1E9573, 0x7B81111A, 0xCBC139F0, 
-	0xA3513861, 0x1D2C362E, 0x918C580E, 0xB58E1B9C,
-	/* C */
-	0xE4B573A1, 0x4C1A0880, 0x1E907C51, 0x04807EFD, 
-	0x3AD8CDE5, 0x16B21302, 0x02512C53, 0x2204CB18,
-    0x99405F2D, 0xE5B648A1, 0x70AB1D43, 0xA10C25C2, 
-	0x16F1AC05, 0x38BBEB56, 0x9B01DC60, 0xB1096D83
-};
-
-uint32_t shabal256_iv[] PROGMEM = {
-	/* A */ 
-	0x52F84552, 0xE54B7999, 0x2D8EE3EC, 0xB9645191, 
-	0xE0078B86, 0xBB7C44C9, 0xD2B5C1CA, 0xB0D2EB8C,
-	0x14CE5A45, 0x22AF50DC, 0xEFFDBC6B, 0xEB21B74A,
-	/* B */ 
-	0xB555C6EE, 0x3E710596, 0xA72A652F, 0x9301515F, 
-	0xDA28C1FA, 0x696FD868, 0x9CB6BF72, 0x0AFE4002,
-    0xA6E03615, 0x5138C1D4, 0xBE216306, 0xB38B8890, 
-	0x3EA8B96B, 0x3299ACE4, 0x30924DD4, 0x55CB34A5,
-	/* C */ 
-	0xB405F031, 0xC4233EBA, 0xB3733979, 0xC0DD9D55,
-	0xC51C28AE, 0xA327B8E1, 0x56C56167, 0xED614433,
-	0x88B59D60, 0x60E2CEBA, 0x758B4B8B, 0x83E82A7F, 
-	0xBC968828, 0xE6E00BF7, 0xBA839E55, 0x9B491C60
-};
-
-uint32_t shabal384_iv[] PROGMEM = {
-	/* A */ 
-	0xC8FCA331, 0xE55C504E, 0x003EBF26, 0xBB6B8D83, 
-	0x7B0448C1, 0x41B82789, 0x0A7C9601, 0x8D659CFF,
-	0xB6E2673E, 0xCA54C77B, 0x1460FD7E, 0x3FCB8F2D,
-	/* B */ 
-	0x527291FC, 0x2A16455F, 0x78E627E5, 0x944F169F, 
-	0x1CA6F016, 0xA854EA25, 0x8DB98ABE, 0xF2C62641,
-    0x30117DCB, 0xCF5C4309, 0x93711A25, 0xF9F671B8, 
-	0xB01D2116, 0x333F4B89, 0xB285D165, 0x86829B36,
-	/* C */ 
-	0xF764B11A, 0x76172146, 0xCEF6934D, 0xC6D28399, 
-	0xFE095F61, 0x5E6018B4, 0x5048ECF5, 0x51353261,
-    0x6E6E36DC, 0x63130DAD, 0xA9C69BD6, 0x1E90EA0C, 
-	0x7C35073B, 0x28D95E6D, 0xAA340E0D, 0xCB3DEE70
-};
-
-uint32_t shabal512_iv[] PROGMEM = {
-	/* A */ 
-	0x20728DFD, 0x46C0BD53, 0xE782B699, 0x55304632, 
-	0x71B4EF90, 0x0EA9E82C, 0xDBB930F1, 0xFAD06B8B,
-	0xBE0CAE40, 0x8BD14410, 0x76D2ADAC, 0x28ACAB7F,
-	/* B */ 
-	0xC1099CB7, 0x07B385F3, 0xE7442C26, 0xCC8AD640, 
-	0xEB6F56C7, 0x1EA81AA9, 0x73B9D314, 0x1DE85D08,
-	0x48910A5A, 0x893B22DB, 0xC5A0DF44, 0xBBC4324E, 
-	0x72D2F240, 0x75941D99, 0x6D8BDE82, 0xA1A7502B,
-	/* C */
-	0xD9BF68D1, 0x58BAD750, 0x56028CB2, 0x8134F359, 
-	0xB5D469D8, 0x941A8CC2, 0x418B2A6E, 0x04052780,
-	0x7F07D787, 0x5194358F, 0x3C60D665, 0xBE97D79A, 
-	0x950C3434, 0xAED9A06D, 0x2537DC8D, 0x7CDB5969,
-};
-
-void shabal192_init(shabal_ctx_t* ctx){
-	uint8_t i;
-	ctx->b = ctx->b_buffer;
-	ctx->c = ctx->c_buffer;
-	ctx->w.w64 = 1LL;
-	for(i=0;i<SHABAL_R;++i){
-		ctx->a[i] = pgm_read_dword(&(shabal192_iv[i]));
-	}
-	for(i=0;i<16;++i){
-		ctx->b[i] = pgm_read_dword(&(shabal192_iv[SHABAL_R+i]));
-	}
-	for(i=0;i<16;++i){
-		ctx->c[i] = pgm_read_dword(&(shabal192_iv[SHABAL_R+16+i]));
-	}
-}
-
-void shabal224_init(shabal_ctx_t* ctx){
-	uint8_t i;
-	ctx->b = ctx->b_buffer;
-	ctx->c = ctx->c_buffer;
-	ctx->w.w64 = 1LL;
-	for(i=0;i<SHABAL_R;++i){
-		ctx->a[i] = pgm_read_dword(&(shabal224_iv[i]));
-	}
-	for(i=0;i<16;++i){
-		ctx->b[i] = pgm_read_dword(&(shabal224_iv[SHABAL_R+i]));
-	}
-	for(i=0;i<16;++i){
-		ctx->c[i] = pgm_read_dword(&(shabal224_iv[SHABAL_R+16+i]));
-	}
-}
-void shabal256_init(shabal_ctx_t* ctx){
-	uint8_t i;
-	ctx->b = ctx->b_buffer;
-	ctx->c = ctx->c_buffer;
-	ctx->w.w64 = 1LL;
-	for(i=0;i<SHABAL_R;++i){
-		ctx->a[i] = pgm_read_dword(&(shabal256_iv[i]));
-	}
-	for(i=0;i<16;++i){
-		ctx->b[i] = pgm_read_dword(&(shabal256_iv[SHABAL_R+i]));
-	}
-	for(i=0;i<16;++i){
-		ctx->c[i] = pgm_read_dword(&(shabal256_iv[SHABAL_R+16+i]));
-	}
-}
-void shabal384_init(shabal_ctx_t* ctx){
-	uint8_t i;
-	ctx->b = ctx->b_buffer;
-	ctx->c = ctx->c_buffer;
-	ctx->w.w64 = 1LL;
-	for(i=0;i<SHABAL_R;++i){
-		ctx->a[i] = pgm_read_dword(&(shabal384_iv[i]));
-	}
-	for(i=0;i<16;++i){
-		ctx->b[i] = pgm_read_dword(&(shabal384_iv[SHABAL_R+i]));
-	}
-	for(i=0;i<16;++i){
-		ctx->c[i] = pgm_read_dword(&(shabal384_iv[SHABAL_R+16+i]));
-	}
-}
-void shabal512_init(shabal_ctx_t* ctx){
-	uint8_t i;
-	ctx->b = ctx->b_buffer;
-	ctx->c = ctx->c_buffer;
-	ctx->w.w64 = 1LL;
-	for(i=0;i<SHABAL_R;++i){
-		ctx->a[i] = pgm_read_dword(&(shabal512_iv[i]));
-	}
-	for(i=0;i<16;++i){
-		ctx->b[i] = pgm_read_dword(&(shabal512_iv[SHABAL_R+i]));
-	}
-	for(i=0;i<16;++i){
-		ctx->c[i] = pgm_read_dword(&(shabal512_iv[SHABAL_R+16+i]));
-	}
 }
 
 void shabal_nextBlock(shabal_ctx_t* ctx, const void* block){
@@ -276,14 +104,8 @@ void shabal_lastBlock(shabal_ctx_t* ctx, const void* block, uint16_t length_b){
 	for(i=0;i<16;++i){
 		ctx->b[i] += ((uint32_t*)buffer)[i];
 	}
-	ctx->a[0] ^= ctx->w.w32[0];
-	ctx->a[1] ^= ctx->w.w32[1];
-	shabal_p(ctx, buffer);
-	t = ctx->c;
-	ctx->c = ctx->b;
-	ctx->b = t;
 
-	for(j=0; j<3;++j){
+	for(j=0; j<4;++j){
 		ctx->a[0] ^= ctx->w.w32[0];
 		ctx->a[1] ^= ctx->w.w32[1];
 		shabal_p(ctx, buffer);
@@ -294,87 +116,6 @@ void shabal_lastBlock(shabal_ctx_t* ctx, const void* block, uint16_t length_b){
 	 
 }
 
-static inline
 void shabal_ctx2hash(void* dest, const shabal_ctx_t* ctx, uint16_t outlength_b){
 	memcpy(dest, &(ctx->c[16-outlength_b/32]), outlength_b/8);
 }
-
-void shabal192_ctx2hash(void* dest, const shabal_ctx_t* ctx){
-	shabal_ctx2hash(dest, ctx, 192);
-}
-
-void shabal224_ctx2hash(void* dest, const shabal_ctx_t* ctx){
-	shabal_ctx2hash(dest, ctx, 224);
-}
-
-void shabal256_ctx2hash(void* dest, const shabal_ctx_t* ctx){
-	shabal_ctx2hash(dest, ctx, 256);
-}
-
-void shabal384_ctx2hash(void* dest, const shabal_ctx_t* ctx){
-	shabal_ctx2hash(dest, ctx, 384);
-}
-
-void shabal512_ctx2hash(void* dest, const shabal_ctx_t* ctx){
-	shabal_ctx2hash(dest, ctx, 512);
-}
-
-void shabal192(void* dest, void* msg, uint32_t length_b){
-	shabal_ctx_t ctx;
-	shabal192_init(&ctx);
-	while(length_b>=SHABAL_BLOCKSIZE){
-		shabal_nextBlock(&ctx, msg);
-		msg = (uint8_t*)msg+SHABAL_BLOCKSIZE_B;
-		length_b -= SHABAL_BLOCKSIZE;
-	}
-	shabal_lastBlock(&ctx, msg, length_b);
-	shabal192_ctx2hash(dest, &ctx);
-}
-
-void shabal224(void* dest, void* msg, uint32_t length_b){
-	shabal_ctx_t ctx;
-	shabal224_init(&ctx);
-	while(length_b>=SHABAL_BLOCKSIZE){
-		shabal_nextBlock(&ctx, msg);
-		msg = (uint8_t*)msg+SHABAL_BLOCKSIZE_B;
-		length_b -= SHABAL_BLOCKSIZE;
-	}
-	shabal_lastBlock(&ctx, msg, length_b);
-	shabal224_ctx2hash(dest, &ctx);
-}
-
-void shabal256(void* dest, void* msg, uint32_t length_b){
-	shabal_ctx_t ctx;
-	shabal256_init(&ctx);
-	while(length_b>=SHABAL_BLOCKSIZE){
-		shabal_nextBlock(&ctx, msg);
-		msg = (uint8_t*)msg+SHABAL_BLOCKSIZE_B;
-		length_b -= SHABAL_BLOCKSIZE;
-	}
-	shabal_lastBlock(&ctx, msg, length_b);
-	shabal256_ctx2hash(dest, &ctx);
-}
-
-void shabal384(void* dest, void* msg, uint32_t length_b){
-	shabal_ctx_t ctx;
-	shabal384_init(&ctx);
-	while(length_b>=SHABAL_BLOCKSIZE){
-		shabal_nextBlock(&ctx, msg);
-		msg = (uint8_t*)msg+SHABAL_BLOCKSIZE_B;
-		length_b -= SHABAL_BLOCKSIZE;
-	}
-	shabal_lastBlock(&ctx, msg, length_b);
-	shabal384_ctx2hash(dest, &ctx);
-}
-
-void shabal512(void* dest, void* msg, uint32_t length_b){
-	shabal_ctx_t ctx;
-	shabal512_init(&ctx);
-	while(length_b>=SHABAL_BLOCKSIZE){
-		shabal_nextBlock(&ctx, msg);
-		msg = (uint8_t*)msg+SHABAL_BLOCKSIZE_B;
-		length_b -= SHABAL_BLOCKSIZE;
-	}
-	shabal_lastBlock(&ctx, msg, length_b);
-	shabal512_ctx2hash(dest, &ctx);
-}
diff --git a/shabal.h b/shabal.h
index 5a9f921..fab23a1 100644
--- a/shabal.h
+++ b/shabal.h
@@ -38,13 +38,13 @@
 
 typedef struct{
 	union{
-	uint64_t w64;
-	uint32_t w32[2];
+		uint64_t w64;
+		uint32_t w32[2];
 	} w; /* the counter */
-	uint32_t a[SHABAL_R];
 	uint32_t *b;
-	uint32_t b_buffer[16];
 	uint32_t *c;
+	uint32_t a[SHABAL_R];
+	uint32_t b_buffer[16];
 	uint32_t c_buffer[16];
 }shabal_ctx_t;
 
@@ -70,4 +70,6 @@ void shabal256(void* dest, void* msg, uint32_t length_b);
 void shabal384(void* dest, void* msg, uint32_t length_b);
 void shabal512(void* dest, void* msg, uint32_t length_b);
 
+void shabal_ctx2hash(void* dest, const shabal_ctx_t* ctx, uint16_t outlength_b);
+
 #endif /* SHABAL_H_ */
diff --git a/shabal192-asm.S b/shabal192-asm.S
new file mode 100644
index 0000000..09b41d5
--- /dev/null
+++ b/shabal192-asm.S
@@ -0,0 +1,191 @@
+/* shabal192-asm.S */
+/*
+    This file is part of the AVR-Crypto-Lib.
+    Copyright (C) 2009  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/>.
+*/
+/*
+ * \file    shabal192-asm.S
+ * \author  Daniel Otte
+ * \email   daniel.otte@rub.de
+ * \date    2009-04-27
+ * \license GPLv3 or later
+ */ 
+
+#include "avr-asm-macros.S"
+
+shabal192_iv:
+	/* A */
+.long	0xFD749ED4, 0xB798E530, 0x33904B6F, 0x46BDA85E
+.long	0x076934B4, 0x454B4058, 0x77F74527, 0xFB4CF465
+.long	0x62931DA9, 0xE778C8DB, 0x22B3998E, 0xAC15CFB9
+	/* B */
+.long	0x58BCBAC4, 0xEC47A08E, 0xAEE933B2, 0xDFCBC824
+.long	0xA7944804, 0xBF65BDB0, 0x5A9D4502, 0x59979AF7
+.long	0xC5CEA54E, 0x4B6B8150, 0x16E71909, 0x7D632319
+.long	0x930573A0, 0xF34C63D1, 0xCAF914B4, 0xFDD6612C
+	/* C */
+.long	0x61550878, 0x89EF2B75, 0xA1660C46, 0x7EF3855B
+.long	0x7297B58C, 0x1BC67793, 0x7FB1C723, 0xB66FC640
+.long	0x1A48B71C, 0xF0976D17, 0x088CE80A, 0xA454EDF3
+.long	0x1C096BF4, 0xAC76224B, 0x5215781C, 0xCD5D2669
+
+
+/******************************************************************************/
+/*
+void shabal192_init(shabal_ctx_t* ctx){
+	uint8_t i;
+	ctx->b = ctx->b_buffer;
+	ctx->c = ctx->c_buffer;
+	ctx->w.w64 = 1LL;
+	for(i=0;i<SHABAL_R;++i){
+		ctx->a[i] = pgm_read_dword(&(shabal192_iv[i]));
+	}
+	for(i=0;i<16;++i){
+		ctx->b[i] = pgm_read_dword(&(shabal192_iv[SHABAL_R+i]));
+	}
+	for(i=0;i<16;++i){
+		ctx->c[i] = pgm_read_dword(&(shabal192_iv[SHABAL_R+16+i]));
+	}
+}
+*/
+/*
+ * param ctx: r24,r25
+ */
+.global shabal192_init
+shabal192_init:
+	movw r26, r24
+	ldi r24, 1
+	st X+, r24
+	st X+, r1
+	st X+, r1
+	st X+, r1
+	st X+, r1
+	st X+, r1
+	st X+, r1
+	st X+, r1
+	movw r24, r26
+	adiw r24, 12*4+2+2
+	st X+, r24
+	st X+, r25
+	adiw r24, 4*16-1
+	adiw r24, 1
+	st X+, r24
+	st X+, r25 
+	ldi r24, (12+16+16)
+	ldi r30, lo8(shabal192_iv)
+	ldi r31, hi8(shabal192_iv)
+1:	lpm r0, Z+
+	st X+, r0
+	lpm r0, Z+
+	st X+, r0
+	lpm r0, Z+
+	st X+, r0
+	lpm r0, Z+
+	st X+, r0
+	dec r24
+	brne 1b
+	ret	
+
+/******************************************************************************/
+/*
+ * param dest: r24:r25
+ * param ctx:  r22:r23
+ */
+.global shabal192_ctx2hash
+shabal192_ctx2hash:
+	movw r26, r24
+	movw r30, r22
+	ldd r24, Z+(8+2)
+	ldd r25, Z+(8+2+1)
+	movw r30, r24
+	adiw r30, (16-192/32)*4
+	ldi r24, 192/8
+1:  ld r0, Z+
+	st X+, r0
+	dec r24
+	brne 1b
+	ret	
+	
+/******************************************************************************/
+/*
+void shabal192(void* dest, void* msg, uint32_t length_b){
+	shabal_ctx_t ctx;
+	shabal192_init(&ctx);
+	while(length_b>=SHABAL_BLOCKSIZE){
+		shabal_nextBlock(&ctx, msg);
+		msg = (uint8_t*)msg+SHABAL_BLOCKSIZE_B;
+		length_b -= SHABAL_BLOCKSIZE;
+	}
+	shabal_lastBlock(&ctx, msg, length_b);
+	shabal192_ctx2hash(dest, &ctx);
+}
+*/
+/*
+ * param dest:      r24:r25
+ * param msg:       r22:r23
+ * param length_b:  r18:r21
+ */
+DST0 =  8
+DST1 =  9
+CTX0 = 10
+CTX1 = 11
+MSG0 = 12
+MSG1 = 13
+LEN2 = 14
+LEN3 = 15
+LEN0 = 16
+LEN1 = 17
+.global shabal192
+shabal192: 
+	push_range 8, 17
+	stack_alloc_large 188
+	adiw r30, 1 /* Z points to stack space (ctx) */
+	movw CTX0, r30
+	movw DST0, r24
+	movw MSG0, r22
+	movw LEN0, r18
+	movw LEN2, r20
+	movw r24, r30
+	rcall shabal192_init
+2:
+	tst LEN2
+	brne 3f
+	tst LEN3
+	breq 4f
+3:
+	movw r24, CTX0
+	movw r22, MSG0
+	rcall shabal_nextBlock
+	subi LEN1, 0x02
+	sbc LEN2, r1
+	sbc LEN3, r1
+	ldi r18, 64
+	add MSG0, r18
+	adc MSG1, r1	
+	rjmp 2b
+4:	
+	movw r24, CTX0
+	movw r22, MSG0
+	movw r20, LEN0
+	rcall shabal_lastBlock
+	movw r24, DST0
+	movw r22, CTX0
+	rcall shabal192_ctx2hash
+	stack_free_large2 188
+	pop_range 8, 17
+	ret
+	
+	
diff --git a/shabal224-asm.S b/shabal224-asm.S
new file mode 100644
index 0000000..1168d1b
--- /dev/null
+++ b/shabal224-asm.S
@@ -0,0 +1,160 @@
+/* shabal224-asm.S */
+/*
+    This file is part of the AVR-Crypto-Lib.
+    Copyright (C) 2009  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/>.
+*/
+/*
+ * \file    shabal224-asm.S
+ * \author  Daniel Otte
+ * \email   daniel.otte@rub.de
+ * \date    2009-04-27
+ * \license GPLv3 or later
+ */ 
+
+#include "avr-asm-macros.S"
+
+/******************************************************************************/	
+
+shabal224_iv:
+	/* A */
+.long	0xA5201467, 0xA9B8D94A, 0xD4CED997, 0x68379D7B 
+.long	0xA7FC73BA, 0xF1A2546B, 0x606782BF, 0xE0BCFD0F
+.long	0x2F25374E, 0x069A149F, 0x5E2DFF25, 0xFAECF061
+	/* B */ 
+.long	0xEC9905D8, 0xF21850CF, 0xC0A746C8, 0x21DAD498
+.long	0x35156EEB, 0x088C97F2, 0x26303E40, 0x8A2D4FB5
+.long	0xFEEE44B6, 0x8A1E9573, 0x7B81111A, 0xCBC139F0 
+.long	0xA3513861, 0x1D2C362E, 0x918C580E, 0xB58E1B9C
+	/* C */
+.long	0xE4B573A1, 0x4C1A0880, 0x1E907C51, 0x04807EFD 
+.long	0x3AD8CDE5, 0x16B21302, 0x02512C53, 0x2204CB18
+.long	0x99405F2D, 0xE5B648A1, 0x70AB1D43, 0xA10C25C2 
+.long	0x16F1AC05, 0x38BBEB56, 0x9B01DC60, 0xB1096D83
+
+
+/******************************************************************************/	
+/*
+ * param ctx: r24,r25
+ */
+.global shabal224_init
+shabal224_init:
+	movw r26, r24
+	ldi r24, 1
+	st X+, r24
+	st X+, r1
+	st X+, r1
+	st X+, r1
+	st X+, r1
+	st X+, r1
+	st X+, r1
+	st X+, r1
+	movw r24, r26
+	adiw r24, 12*4+4
+	st X+, r24
+	st X+, r25
+	adiw r24, 4*16-1
+	adiw r24, 1
+	st X+, r24
+	st X+, r25 
+	ldi r24, (12+16+16)
+	ldi r30, lo8(shabal224_iv)
+	ldi r31, hi8(shabal224_iv)
+1:	lpm r0, Z+
+	st X+, r0
+	lpm r0, Z+
+	st X+, r0
+	lpm r0, Z+
+	st X+, r0
+	lpm r0, Z+
+	st X+, r0
+	dec r24
+	brne 1b
+	ret	
+
+/******************************************************************************/	
+	
+.global shabal224_ctx2hash
+shabal224_ctx2hash:
+	movw r26, r24
+	movw r30, r22
+	ldd r24, Z+(8+2)
+	ldd r25, Z+(8+2+1)
+	movw r30, r24
+	adiw r30, (16-224/32)*4
+	ldi r24, 224/8
+1:  ld r0, Z+
+	st X+, r0
+	dec r24
+	brne 1b
+	ret		
+	
+/******************************************************************************/	
+/*
+ * param dest:      r24:r25
+ * param msg:       r22:r23
+ * param length_b:  r18:r21
+ */
+DST0 =  8
+DST1 =  9
+CTX0 = 10
+CTX1 = 11
+MSG0 = 12
+MSG1 = 13
+LEN2 = 14
+LEN3 = 15
+LEN0 = 16
+LEN1 = 17
+.global shabal224
+shabal224: 
+	push_range 8, 17
+	stack_alloc_large 188
+	adiw r30, 1 /* Z points to stack space (ctx) */
+	movw CTX0, r30
+	movw DST0, r24
+	movw MSG0, r22
+	movw LEN0, r18
+	movw LEN2, r20
+	movw r24, r30
+	rcall shabal224_init
+2:
+	tst LEN2
+	brne 3f
+	tst LEN3
+	breq 4f
+3:
+	movw r24, CTX0
+	movw r22, MSG0
+	rcall shabal_nextBlock
+	subi LEN1, 0x02
+	sbc LEN2, r1
+	sbc LEN3, r1
+	ldi r18, 64
+	add MSG0, r18
+	adc MSG1, r1	
+	rjmp 2b
+4:	
+	movw r24, CTX0
+	movw r22, MSG0
+	movw r20, LEN0
+	rcall shabal_lastBlock
+	movw r24, DST0
+	movw r22, CTX0
+	rcall shabal224_ctx2hash
+	stack_free_large2 188
+	pop_range 8, 17
+	ret
+	
+
diff --git a/shabal256-asm.S b/shabal256-asm.S
new file mode 100644
index 0000000..f95e6c7
--- /dev/null
+++ b/shabal256-asm.S
@@ -0,0 +1,159 @@
+/* shabal256-asm.S */
+/*
+    This file is part of the AVR-Crypto-Lib.
+    Copyright (C) 2009  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/>.
+*/
+/*
+ * \file    shabal256-asm.S
+ * \author  Daniel Otte
+ * \email   daniel.otte@rub.de
+ * \date    2009-04-27
+ * \license GPLv3 or later
+ */ 
+
+#include "avr-asm-macros.S"
+
+/******************************************************************************/	
+
+shabal256_iv:
+	/* A */ 
+.long	0x52F84552, 0xE54B7999, 0x2D8EE3EC, 0xB9645191 
+.long	0xE0078B86, 0xBB7C44C9, 0xD2B5C1CA, 0xB0D2EB8C
+.long	0x14CE5A45, 0x22AF50DC, 0xEFFDBC6B, 0xEB21B74A
+	/* B */ 
+.long	0xB555C6EE, 0x3E710596, 0xA72A652F, 0x9301515F 
+.long	0xDA28C1FA, 0x696FD868, 0x9CB6BF72, 0x0AFE4002
+.long	0xA6E03615, 0x5138C1D4, 0xBE216306, 0xB38B8890 
+.long	0x3EA8B96B, 0x3299ACE4, 0x30924DD4, 0x55CB34A5
+	/* C */ 
+.long	0xB405F031, 0xC4233EBA, 0xB3733979, 0xC0DD9D55
+.long	0xC51C28AE, 0xA327B8E1, 0x56C56167, 0xED614433
+.long	0x88B59D60, 0x60E2CEBA, 0x758B4B8B, 0x83E82A7F 
+.long	0xBC968828, 0xE6E00BF7, 0xBA839E55, 0x9B491C60
+
+/******************************************************************************/	
+/*
+ * param ctx: r24,r25
+ */
+.global shabal256_init
+shabal256_init:
+	movw r26, r24
+	ldi r24, 1
+	st X+, r24
+	st X+, r1
+	st X+, r1
+	st X+, r1
+	st X+, r1
+	st X+, r1
+	st X+, r1
+	st X+, r1
+	movw r24, r26
+	adiw r24, 12*4+4
+	st X+, r24
+	st X+, r25
+	adiw r24, 4*16-1
+	adiw r24, 1
+	st X+, r24
+	st X+, r25 
+	ldi r24, (12+16+16)
+	ldi r30, lo8(shabal256_iv)
+	ldi r31, hi8(shabal256_iv)
+1:	lpm r0, Z+
+	st X+, r0
+	lpm r0, Z+
+	st X+, r0
+	lpm r0, Z+
+	st X+, r0
+	lpm r0, Z+
+	st X+, r0
+	dec r24
+	brne 1b
+	ret	
+
+/******************************************************************************/	
+
+.global shabal256_ctx2hash
+shabal256_ctx2hash:
+	movw r26, r24
+	movw r30, r22
+	ldd r24, Z+(8+2)
+	ldd r25, Z+(8+2+1)
+	movw r30, r24
+	adiw r30, (16-256/32)*4
+	ldi r24, 256/8
+1:  ld r0, Z+
+	st X+, r0
+	dec r24
+	brne 1b
+	ret		
+	
+/******************************************************************************/	
+/*
+ * param dest:      r24:r25
+ * param msg:       r22:r23
+ * param length_b:  r18:r21
+ */
+DST0 =  8
+DST1 =  9
+CTX0 = 10
+CTX1 = 11
+MSG0 = 12
+MSG1 = 13
+LEN2 = 14
+LEN3 = 15
+LEN0 = 16
+LEN1 = 17
+.global shabal256
+shabal256: 
+	push_range 8, 17
+	stack_alloc_large 188
+	adiw r30, 1 /* Z points to stack space (ctx) */
+	movw CTX0, r30
+	movw DST0, r24
+	movw MSG0, r22
+	movw LEN0, r18
+	movw LEN2, r20
+	movw r24, r30
+	rcall shabal256_init
+2:
+	tst LEN2
+	brne 3f
+	tst LEN3
+	breq 4f
+3:
+	movw r24, CTX0
+	movw r22, MSG0
+	rcall shabal_nextBlock
+	subi LEN1, 0x02
+	sbc LEN2, r1
+	sbc LEN3, r1
+	ldi r18, 64
+	add MSG0, r18
+	adc MSG1, r1	
+	rjmp 2b
+4:	
+	movw r24, CTX0
+	movw r22, MSG0
+	movw r20, LEN0
+	rcall shabal_lastBlock
+	movw r24, DST0
+	movw r22, CTX0
+	rcall shabal256_ctx2hash
+	stack_free_large2 188
+	pop_range 8, 17
+	ret
+	
+
diff --git a/shabal384-asm.S b/shabal384-asm.S
new file mode 100644
index 0000000..e9a053d
--- /dev/null
+++ b/shabal384-asm.S
@@ -0,0 +1,158 @@
+/* shabal384-asm.S */
+/*
+    This file is part of the AVR-Crypto-Lib.
+    Copyright (C) 2009  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/>.
+*/
+/*
+ * \file    shabal384-asm.S
+ * \author  Daniel Otte
+ * \email   daniel.otte@rub.de
+ * \date    2009-04-27
+ * \license GPLv3 or later
+ */ 
+
+#include "avr-asm-macros.S"
+
+/******************************************************************************/	
+
+shabal384_iv:
+	/* A */ 
+.long	0xC8FCA331, 0xE55C504E, 0x003EBF26, 0xBB6B8D83 
+.long	0x7B0448C1, 0x41B82789, 0x0A7C9601, 0x8D659CFF
+.long	0xB6E2673E, 0xCA54C77B, 0x1460FD7E, 0x3FCB8F2D
+	/* B */ 
+.long	0x527291FC, 0x2A16455F, 0x78E627E5, 0x944F169F 
+.long	0x1CA6F016, 0xA854EA25, 0x8DB98ABE, 0xF2C62641
+.long	0x30117DCB, 0xCF5C4309, 0x93711A25, 0xF9F671B8 
+.long	0xB01D2116, 0x333F4B89, 0xB285D165, 0x86829B36
+	/* C */ 
+.long	0xF764B11A, 0x76172146, 0xCEF6934D, 0xC6D28399 
+.long	0xFE095F61, 0x5E6018B4, 0x5048ECF5, 0x51353261
+.long	0x6E6E36DC, 0x63130DAD, 0xA9C69BD6, 0x1E90EA0C 
+.long	0x7C35073B, 0x28D95E6D, 0xAA340E0D, 0xCB3DEE70
+
+/******************************************************************************/	
+/*
+ * param ctx: r24,r25
+ */
+.global shabal384_init
+shabal384_init:
+	movw r26, r24
+	ldi r24, 1
+	st X+, r24
+	st X+, r1
+	st X+, r1
+	st X+, r1
+	st X+, r1
+	st X+, r1
+	st X+, r1
+	st X+, r1
+	movw r24, r26
+	adiw r24, 12*4+4
+	st X+, r24
+	st X+, r25
+	adiw r24, 4*16-1
+	adiw r24, 1
+	st X+, r24
+	st X+, r25 
+	ldi r24, (12+16+16)
+	ldi r30, lo8(shabal384_iv)
+	ldi r31, hi8(shabal384_iv)
+1:	lpm r0, Z+
+	st X+, r0
+	lpm r0, Z+
+	st X+, r0
+	lpm r0, Z+
+	st X+, r0
+	lpm r0, Z+
+	st X+, r0
+	dec r24
+	brne 1b
+	ret	
+
+/******************************************************************************/	
+
+.global shabal384_ctx2hash
+shabal384_ctx2hash:
+	movw r26, r24
+	movw r30, r22
+	ldd r24, Z+(8+2)
+	ldd r25, Z+(8+2+1)
+	movw r30, r24
+	adiw r30, (16-384/32)*4
+	ldi r24, 384/8
+1:  ld r0, Z+
+	st X+, r0
+	dec r24
+	brne 1b
+	ret		
+
+/******************************************************************************/	
+/*
+ * param dest:      r24:r25
+ * param msg:       r22:r23
+ * param length_b:  r18:r21
+ */
+DST0 =  8
+DST1 =  9
+CTX0 = 10
+CTX1 = 11
+MSG0 = 12
+MSG1 = 13
+LEN2 = 14
+LEN3 = 15
+LEN0 = 16
+LEN1 = 17
+.global shabal384
+shabal384: 
+	push_range 8, 17
+	stack_alloc_large 188
+	adiw r30, 1 /* Z points to stack space (ctx) */
+	movw CTX0, r30
+	movw DST0, r24
+	movw MSG0, r22
+	movw LEN0, r18
+	movw LEN2, r20
+	movw r24, r30
+	rcall shabal384_init
+2:
+	tst LEN2
+	brne 3f
+	tst LEN3
+	breq 4f
+3:
+	movw r24, CTX0
+	movw r22, MSG0
+	rcall shabal_nextBlock
+	subi LEN1, 0x02
+	sbc LEN2, r1
+	sbc LEN3, r1
+	ldi r18, 64
+	add MSG0, r18
+	adc MSG1, r1	
+	rjmp 2b
+4:	
+	movw r24, CTX0
+	movw r22, MSG0
+	movw r20, LEN0
+	rcall shabal_lastBlock
+	movw r24, DST0
+	movw r22, CTX0
+	rcall shabal384_ctx2hash
+	stack_free_large2 188
+	pop_range 8, 17
+	ret
+	
diff --git a/shabal512-asm.S b/shabal512-asm.S
new file mode 100644
index 0000000..0d412ff
--- /dev/null
+++ b/shabal512-asm.S
@@ -0,0 +1,158 @@
+/* shabal512-asm.S */
+/*
+    This file is part of the AVR-Crypto-Lib.
+    Copyright (C) 2009  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/>.
+*/
+/*
+ * \file    shabal512-asm.S
+ * \author  Daniel Otte
+ * \email   daniel.otte@rub.de
+ * \date    2009-04-27
+ * \license GPLv3 or later
+ */ 
+
+#include "avr-asm-macros.S"
+
+/******************************************************************************/	
+
+shabal512_iv:
+	/* A */ 
+.long	0x20728DFD, 0x46C0BD53, 0xE782B699, 0x55304632 
+.long	0x71B4EF90, 0x0EA9E82C, 0xDBB930F1, 0xFAD06B8B
+.long	0xBE0CAE40, 0x8BD14410, 0x76D2ADAC, 0x28ACAB7F
+	/* B */ 
+.long	0xC1099CB7, 0x07B385F3, 0xE7442C26, 0xCC8AD640 
+.long	0xEB6F56C7, 0x1EA81AA9, 0x73B9D314, 0x1DE85D08
+.long	0x48910A5A, 0x893B22DB, 0xC5A0DF44, 0xBBC4324E 
+.long	0x72D2F240, 0x75941D99, 0x6D8BDE82, 0xA1A7502B
+	/* C */
+.long	0xD9BF68D1, 0x58BAD750, 0x56028CB2, 0x8134F359 
+.long	0xB5D469D8, 0x941A8CC2, 0x418B2A6E, 0x04052780
+.long	0x7F07D787, 0x5194358F, 0x3C60D665, 0xBE97D79A 
+.long	0x950C3434, 0xAED9A06D, 0x2537DC8D, 0x7CDB5969
+
+/******************************************************************************/	
+/*
+ * param ctx: r24,r25
+ */
+.global shabal512_init
+shabal512_init:
+	movw r26, r24
+	ldi r24, 1
+	st X+, r24
+	st X+, r1
+	st X+, r1
+	st X+, r1
+	st X+, r1
+	st X+, r1
+	st X+, r1
+	st X+, r1
+	movw r24, r26
+	adiw r24, 12*4+4
+	st X+, r24
+	st X+, r25
+	adiw r24, 4*16-1
+	adiw r24, 1
+	st X+, r24
+	st X+, r25 
+	ldi r24, (12+16+16)
+	ldi r30, lo8(shabal512_iv)
+	ldi r31, hi8(shabal512_iv)
+1:	lpm r0, Z+
+	st X+, r0
+	lpm r0, Z+
+	st X+, r0
+	lpm r0, Z+
+	st X+, r0
+	lpm r0, Z+
+	st X+, r0
+	dec r24
+	brne 1b
+	ret	
+
+/******************************************************************************/	
+
+.global shabal512_ctx2hash
+shabal512_ctx2hash:
+	movw r26, r24
+	movw r30, r22
+	ldd r24, Z+(8+2)
+	ldd r25, Z+(8+2+1)
+	movw r30, r24
+	; adiw r30, (16-512/32)*4
+	ldi r24, 512/8
+1:  ld r0, Z+
+	st X+, r0
+	dec r24
+	brne 1b
+	ret	
+	
+/******************************************************************************/	
+/*
+ * param dest:      r24:r25
+ * param msg:       r22:r23
+ * param length_b:  r18:r21
+ */
+DST0 =  8
+DST1 =  9
+CTX0 = 10
+CTX1 = 11
+MSG0 = 12
+MSG1 = 13
+LEN2 = 14
+LEN3 = 15
+LEN0 = 16
+LEN1 = 17
+.global shabal512
+shabal512: 
+	push_range 8, 17
+	stack_alloc_large 188
+	adiw r30, 1 /* Z points to stack space (ctx) */
+	movw CTX0, r30
+	movw DST0, r24
+	movw MSG0, r22
+	movw LEN0, r18
+	movw LEN2, r20
+	movw r24, r30
+	rcall shabal512_init
+2:
+	tst LEN2
+	brne 3f
+	tst LEN3
+	breq 4f
+3:
+	movw r24, CTX0
+	movw r22, MSG0
+	rcall shabal_nextBlock
+	subi LEN1, 0x02
+	sbc LEN2, r1
+	sbc LEN3, r1
+	ldi r18, 64
+	add MSG0, r18
+	adc MSG1, r1	
+	rjmp 2b
+4:	
+	movw r24, CTX0
+	movw r22, MSG0
+	movw r20, LEN0
+	rcall shabal_lastBlock
+	movw r24, DST0
+	movw r22, CTX0
+	rcall shabal512_ctx2hash
+	stack_free_large2 188
+	pop_range 8, 17
+	ret
+		
diff --git a/test_src/main-shabal-test.c b/test_src/main-shabal-test.c
index 3e8a9d6..c103529 100644
--- a/test_src/main-shabal-test.c
+++ b/test_src/main-shabal-test.c
@@ -123,6 +123,86 @@ void testrun_stdtest_shabal(void){
 	testrun_stdtest_shabal512(mb, strlen(mb)*8);
 }
 
+void testshort(void){
+	uint8_t ma[64];
+	memset(ma, 0, 64);
+	testrun_stdtest_shabal192(ma, 64*8);
+}
+
+void shabal_ctx_dump(shabal_ctx_t* ctx){
+	uint8_t i;
+	void* p;
+	cli_putstr_P(PSTR("\r\n=== shabal ctx dump ===\r\n  size = "));
+	i=sizeof(shabal_ctx_t);
+	if(i>=100)
+		cli_putc('0'+i/100);
+	if(i>=10)
+		cli_putc('0'+(i/10)%10);
+	cli_putc('0'+i%10);	
+	cli_putstr_P(PSTR("\r\n  a = "));
+	cli_hexdump_block(ctx->a, 12*4, 5, 4*8);
+	cli_putstr_P(PSTR("\r\n  b_buffer = "));
+	cli_hexdump_block(ctx->b_buffer, 12*4, 5, 4*8);
+	cli_putstr_P(PSTR("\r\n  c_buffer = "));
+	cli_hexdump_block(ctx->c_buffer, 12*4, 5, 4*8);
+	if(ctx->b == &(ctx->b_buffer[0]))
+		cli_putstr_P(PSTR("\r\nb --> b_buffer"));
+	if(ctx->b == &(ctx->c_buffer[0]))
+		cli_putstr_P(PSTR("\r\nb --> c_buffer"));	
+	if(ctx->c == &(ctx->b_buffer[0]))
+		cli_putstr_P(PSTR("\r\nc --> b_buffer"));
+	if(ctx->c == &(ctx->c_buffer[0]))
+		cli_putstr_P(PSTR("\r\nc --> c_buffer"));
+	cli_putstr_P(PSTR("\r\n b = "));
+	cli_hexdump(&(ctx->b), 2);
+	p = ctx->b_buffer;
+	cli_putstr_P(PSTR("\r\n b (should) = "));
+	cli_hexdump(&p, 2);
+	cli_putstr_P(PSTR("\r\n c = "));
+	cli_hexdump(&(ctx->c), 2);	
+	p = ctx->c_buffer;
+	cli_putstr_P(PSTR("\r\n c (should) = "));
+	cli_hexdump(&p, 2);
+}
+
+
+void testinit_192(void){
+	shabal_ctx_t ctx;
+	shabal192_init(&ctx);
+	shabal_ctx_dump(&ctx);
+}
+
+void testinit_224(void){
+	shabal_ctx_t ctx;
+	shabal224_init(&ctx);
+	shabal_ctx_dump(&ctx);
+}
+
+void testinit_256(void){
+	shabal_ctx_t ctx;
+	shabal256_init(&ctx);
+	shabal_ctx_dump(&ctx);
+}
+
+void testinit_384(void){
+	shabal_ctx_t ctx;
+	shabal384_init(&ctx);
+	shabal_ctx_dump(&ctx);
+}
+
+void testinit_512(void){
+	shabal_ctx_t ctx;
+	shabal512_init(&ctx);
+	shabal_ctx_dump(&ctx);
+}
+void testinit(void){
+	testinit_192();
+	testinit_224();
+	testinit_256();
+	testinit_384();
+	testinit_512();
+}
+
 void performance_shabal(void){
 	uint64_t t;
 	char str[16];
@@ -143,28 +223,28 @@ void performance_shabal(void){
 	cli_putstr(str);
 	
 	startTimer(1);
-	shabal192_init(&ctx);
+	shabal224_init(&ctx);
 	t = stopTimer();
 	cli_putstr_P(PSTR("\r\n\tctx-gen time (224): "));
 	ultoa((unsigned long)t, str, 10);
 	cli_putstr(str);
 	
 	startTimer(1);
-	shabal192_init(&ctx);
+	shabal256_init(&ctx);
 	t = stopTimer();
 	cli_putstr_P(PSTR("\r\n\tctx-gen time (256): "));
 	ultoa((unsigned long)t, str, 10);
 	cli_putstr(str);
 	
 	startTimer(1);
-	shabal192_init(&ctx);
+	shabal384_init(&ctx);
 	t = stopTimer();
 	cli_putstr_P(PSTR("\r\n\tctx-gen time (384): "));
 	ultoa((unsigned long)t, str, 10);
 	cli_putstr(str);
 	
 	startTimer(1);
-	shabal192_init(&ctx);
+	shabal512_init(&ctx);
 	t = stopTimer();
 	cli_putstr_P(PSTR("\r\n\tctx-gen time (512): "));
 	ultoa((unsigned long)t, str, 10);
@@ -281,6 +361,9 @@ const hfdesc_t* algolist[] PROGMEM = {
 
 const char nessie_str[]      PROGMEM = "nessie";
 const char test_str[]        PROGMEM = "test";
+const char testinit192_str[] PROGMEM = "testinit192";
+const char testinit_str[]    PROGMEM = "testinit";
+const char testshort_str[]   PROGMEM = "short";
 const char ztest_str[]       PROGMEM = "zerotest";
 const char performance_str[] PROGMEM = "performance";
 const char echo_str[]        PROGMEM = "echo";
@@ -291,6 +374,9 @@ const char shavs_test1_str[] PROGMEM = "shavs_test1";
 cmdlist_entry_t cmdlist[] PROGMEM = {
 	{ nessie_str,          NULL, testrun_nessie_shabal},
 	{ test_str,            NULL, testrun_stdtest_shabal},
+	{ testinit192_str,     NULL, testinit_192},
+	{ testinit_str,        NULL, testinit},
+	{ testshort_str,       NULL, testshort},
 	{ performance_str,     NULL, performance_shabal},
 	{ shavs_list_str,      NULL, shavs_listalgos},
 	{ shavs_set_str,   (void*)1, (void_fpt)shavs_setalgo},
diff --git a/test_src/main-skein-test.c b/test_src/main-skein-test.c
index 98102d8..4392314 100644
--- a/test_src/main-skein-test.c
+++ b/test_src/main-skein-test.c
@@ -174,6 +174,10 @@ void zeromsg_test_common(char* p){
 	}
 }
 
+void performance_skein(void){
+}
+
+
 /*****************************************************************************
  *  main																	 *
  *****************************************************************************/
@@ -215,6 +219,7 @@ const char shavs_test1_str[] PROGMEM = "shavs_test1";
 
 cmdlist_entry_t cmdlist[] PROGMEM = {
 //	{ nessie_str,          NULL, testrun_nessie_skein},
+	{ performance_str,     NULL, performance_skein},
 	{ test_str,            NULL, testrun_stdtest_skein},
 	{ ztest_str,       (void*)1, (void_fpt)zeromsg_test_common},
 	{ shavs_list_str,      NULL, shavs_listalgos},