diff -urpN aa/john-1.7.6/src/x86-64.S john-1.7.6/src/x86-64.S
--- aa/john-1.7.6/src/x86-64.S	2010-06-15 01:27:35.000000000 +0000
+++ john-1.7.6/src/x86-64.S	2010-09-02 00:20:06.595802029 +0000
@@ -1071,6 +1071,14 @@ DES_bs_crypt_LM_loop:
  * work on Solaris (unless the use of GNU assembler is forced).
  */
 
+/* Sun's assembler can't multiply, but at least it can add... */
+#define MUL_16(x)  (x+x+x+x+x+x+x+x+x+x+x+x+x+x+x+x)
+#define MUL_32(x)  (MUL_16(x)  + MUL_16(x))
+#define MUL_64(x)  (MUL_32(x)  + MUL_32(x))
+#define MUL_128(x) (MUL_64(x)  + MUL_64(x))
+#define MUL_256(x) (MUL_128(x) + MUL_128(x))
+#define MUL_512(x) (MUL_256(x) + MUL_256(x))
+
 #ifdef UNDERSCORES
 #define nt_crypt_all_x86_64 _nt_crypt_all_x86_64
 #define nt_buffer8x _nt_buffer8x
@@ -1135,8 +1143,8 @@ const_stage3:
 #define t23 %xmm13
 
 #define STEP1(aa, bb, cc, dd, aa3, bb3, cc3, dd3, x, s, base)	\
-	paddd (512*base)+(x*32)+nt_buffer8x(%rip), aa;		\
-	paddd (512*base)+(x*32)+16+nt_buffer8x(%rip), aa3;	\
+	paddd MUL_512(base)+MUL_32(x)+nt_buffer8x(%rip), aa;		\
+	paddd MUL_512(base)+MUL_32(x)+16+nt_buffer8x(%rip), aa3;	\
 	movdqa cc, t1;						\
 	movdqa cc3, t13;					\
 	pxor dd, t1;						\
@@ -1157,8 +1165,8 @@ const_stage3:
 	por t23, aa3;
 
 #define STEP2(aa, bb, cc, dd, aa3, bb3, cc3, dd3, x, s, base)	\
-	paddd (512*base)+(x*32)+nt_buffer8x(%rip), aa;		\
-	paddd (512*base)+(x*32)+16+nt_buffer8x(%rip), aa3;	\
+	paddd MUL_512(base)+MUL_32(x)+nt_buffer8x(%rip), aa;		\
+	paddd MUL_512(base)+MUL_32(x)+16+nt_buffer8x(%rip), aa3;	\
 	movdqa cc, t1;						\
 	movdqa cc3, t13;					\
 	movdqa cc, t2;						\
@@ -1185,8 +1193,8 @@ const_stage3:
 	por t13, aa3;
 
 #define STEP3(aa, bb, cc, dd, aa3, bb3, cc3, dd3, x, s, base)	\
-	paddd (512*base)+(x*32)+nt_buffer8x(%rip), aa;		\
-	paddd (512*base)+(x*32)+16+nt_buffer8x(%rip), aa3;	\
+	paddd MUL_512(base)+MUL_32(x)+nt_buffer8x(%rip), aa;		\
+	paddd MUL_512(base)+MUL_32(x)+16+nt_buffer8x(%rip), aa3;	\
 	movdqa dd, t1;						\
 	movdqa dd3, t13;					\
 	pxor cc, t1;						\
@@ -1216,8 +1224,8 @@ const_stage3:
 	movdqa const_init_d(%rip), d;				\
 	movdqa const_init_d(%rip), d3;				\
 								\
-	paddd (512*base)+nt_buffer8x(%rip), a;			\
-	paddd (512*base)+16+nt_buffer8x(%rip), a3;		\
+	paddd MUL_512(base)+nt_buffer8x(%rip), a;			\
+	paddd MUL_512(base)+16+nt_buffer8x(%rip), a3;		\
 	pslld $3, a;						\
 	pslld $3, a3;						\
 								\
@@ -1267,8 +1275,8 @@ const_stage3:
 	STEP3(c, d, a, b, c3, d3, a3, b3, 5 , 11, base)		\
 	movdqa a, t1;						\
 	movdqa a3, t13;						\
-	paddd (512*base)+416+nt_buffer8x(%rip), b;		\
-	paddd (512*base)+416+16+nt_buffer8x(%rip), b3;		\
+	paddd MUL_512(base)+416+nt_buffer8x(%rip), b;		\
+	paddd MUL_512(base)+416+16+nt_buffer8x(%rip), b3;		\
 	pxor d, t1;						\
 	pxor d3,t13;						\
 	pxor c, t1;						\
@@ -1276,14 +1284,14 @@ const_stage3:
 	paddd t1, b;						\
 	paddd t13,b3;						\
 								\
-	movdqa a,  (128*base)+output8x(%rip);			\
-	movdqa a3,  (128*base)+16+output8x(%rip);		\
-	movdqa b, (128*base)+32+output8x(%rip);			\
-	movdqa b3, (128*base)+32+16+output8x(%rip);		\
-	movdqa c, (128*base)+64+output8x(%rip);			\
-	movdqa c3, (128*base)+64+16+output8x(%rip);		\
-	movdqa d, (128*base)+96+output8x(%rip);			\
-	movdqa d3, (128*base)+96+16+output8x(%rip);
+	movdqa a,  MUL_128(base)+output8x(%rip);			\
+	movdqa a3,  MUL_128(base)+16+output8x(%rip);		\
+	movdqa b, MUL_128(base)+32+output8x(%rip);			\
+	movdqa b3, MUL_128(base)+32+16+output8x(%rip);		\
+	movdqa c, MUL_128(base)+64+output8x(%rip);			\
+	movdqa c3, MUL_128(base)+64+16+output8x(%rip);		\
+	movdqa d, MUL_128(base)+96+output8x(%rip);			\
+	movdqa d3, MUL_128(base)+96+16+output8x(%rip);
 	
 .text
 
diff -urpN aa/john-1.7.6/src/x86-sse.S john-1.7.6/src/x86-sse.S
--- aa/john-1.7.6/src/x86-sse.S	2010-06-15 01:29:35.000000000 +0000
+++ john-1.7.6/src/x86-sse.S	2010-09-02 00:24:47.063922107 +0000
@@ -1319,6 +1319,15 @@ DES_bs_crypt_LM_loop:
  * work on Solaris (unless the use of GNU assembler is forced).
  */
 
+/* Sun's assembler can't multiply, but at least it can add... */
+#define MUL_4(x)   (x+x+x+x)
+#define MUL_16(x)  (x+x+x+x+x+x+x+x+x+x+x+x+x+x+x+x)
+#define MUL_32(x)  (MUL_16(x)  + MUL_16(x))
+#define MUL_64(x)  (MUL_32(x)  + MUL_32(x))
+#define MUL_128(x) (MUL_64(x)  + MUL_64(x))
+#define MUL_256(x) (MUL_128(x) + MUL_128(x))
+#define MUL_512(x) (MUL_256(x) + MUL_256(x))
+
 #ifdef UNDERSCORES
 #define nt_crypt_all_sse2 _nt_crypt_all_sse2
 #define nt_buffer1x _nt_buffer1x
@@ -1387,8 +1396,8 @@ const_stage3:
 #define Q3 $0x6ed9eba1
 
 #define STEP1(aa, bb, cc, dd, aa3, bb3, cc3, dd3, x, s, base)	\
-	paddd (256*base)+(x*16)+nt_buffer4x, aa;		\
-	addl (64*base)+(x*4)+nt_buffer1x, aa3;			\
+	paddd MUL_256(base)+MUL_16(x)+nt_buffer4x, aa;		\
+	addl MUL_64(base)+MUL_4(x)+nt_buffer1x, aa3;			\
 	movdqa cc, t1;						\
 	movl cc3, t13;						\
 	pxor dd, t1;						\
@@ -1406,8 +1415,8 @@ const_stage3:
 	por t2, aa;
 
 #define STEP2(aa, bb, cc, dd, aa3, bb3, cc3, dd3, x, s, base)	\
-	paddd (256*base)+(x*16)+nt_buffer4x, aa;		\
-	addl (64*base)+(x*4)+nt_buffer1x, aa3;			\
+	paddd MUL_256(base)+MUL_16(x)+nt_buffer4x, aa;		\
+	addl MUL_64(base)+MUL_4(x)+nt_buffer1x, aa3;			\
 	movdqa cc, t1;						\
 	movl cc3, t13;						\
 	movdqa cc, t2;						\
@@ -1431,8 +1440,8 @@ const_stage3:
 	por t1, aa;
 
 #define STEP3(aa, bb, cc, dd, aa3, bb3, cc3, dd3, x, s, base)	\
-	paddd (256*base)+(x*16)+nt_buffer4x, aa;		\
-	addl (64*base)+(x*4)+nt_buffer1x, aa3;			\
+	paddd MUL_256(base)+MUL_16(x)+nt_buffer4x, aa;		\
+	addl MUL_64(base)+MUL_4(x)+nt_buffer1x, aa3;			\
 	movdqa dd, t1;						\
 	movl dd3, t13;						\
 	pxor cc, t1;						\
@@ -1459,8 +1468,8 @@ const_stage3:
 	movdqa const_init_d, d;					\
 	movl const_init_d, d3;					\
 								\
-	paddd (256*base)+nt_buffer4x, a;			\
-	addl (64*base)+nt_buffer1x, a3;				\
+	paddd MUL_256(base)+nt_buffer4x, a;			\
+	addl MUL_64(base)+nt_buffer1x, a3;				\
 	pslld $3, a;						\
 	roll $3, a3;						\
 								\
@@ -1510,8 +1519,8 @@ const_stage3:
 	STEP3(c, d, a, b, c3, d3, a3, b3, 5 , 11, base)		\
 	movdqa a, t1;						\
 	movl a3, t13;						\
-	paddd (256*base)+208+nt_buffer4x, b;			\
-	addl (64*base)+52+nt_buffer1x, b3;			\
+	paddd MUL_256(base)+208+nt_buffer4x, b;			\
+	addl MUL_64(base)+52+nt_buffer1x, b3;			\
 	pxor d, t1;						\
 	xorl d3,t13;						\
 	pxor c, t1;						\
@@ -1519,14 +1528,14 @@ const_stage3:
 	paddd t1, b;						\
 	addl t13,b3;						\
 								\
-	movdqa a,  (64*base)+output4x;				\
-	movl a3,  (16*base)+output1x;				\
-	movdqa b, (64*base)+16+output4x;			\
-	movl b3,  (16*base)+4+output1x;				\
-	movdqa c, (64*base)+32+output4x;			\
-	movl c3,  (16*base)+8+output1x;				\
-	movdqa d, (64*base)+48+output4x;			\
-	movl d3,  (16*base)+12+output1x;
+	movdqa a,  MUL_64(base)+output4x;				\
+	movl a3,  MUL_16(base)+output1x;				\
+	movdqa b, MUL_64(base)+16+output4x;			\
+	movl b3,  MUL_16(base)+4+output1x;				\
+	movdqa c, MUL_64(base)+32+output4x;			\
+	movl c3,  MUL_16(base)+8+output1x;				\
+	movdqa d, MUL_64(base)+48+output4x;			\
+	movl d3,  MUL_16(base)+12+output1x;
 	
 .text