From 95becf03279b80d6328277f09ce44d2c7a71a0c8 Mon Sep 17 00:00:00 2001 From: Ken MacKay Date: Wed, 7 May 2014 22:36:06 -0700 Subject: [PATCH] Add fast multiplication for 192-bit and 256-bit curves. --- asm_avr.inc | 7715 +++++++++++++++++++++++++- scripts/mult.py | 162 - scripts/mult_avr.py | 199 + scripts/{square.py => square_avr.py} | 0 uECC.c | 5 +- 5 files changed, 7912 insertions(+), 169 deletions(-) delete mode 100755 scripts/mult.py create mode 100755 scripts/mult_avr.py rename scripts/{square.py => square_avr.py} (100%) diff --git a/asm_avr.inc b/asm_avr.inc index dcce947..f79c0dc 100644 --- a/asm_avr.inc +++ b/asm_avr.inc @@ -152,7 +152,7 @@ static uint8_t vli_sub(uint8_t *p_result, uint8_t *p_left, uint8_t *p_right) } #define asm_sub 1 -#if uECC_CURVE == uECC_secp160r1 +#if (uECC_BYTES == 20) __attribute((noinline)) static void vli_mult(uint8_t *p_result, uint8_t *p_left, uint8_t *p_right) { @@ -2067,11 +2067,7720 @@ static void vli_mult(uint8_t *p_result, uint8_t *p_left, uint8_t *p_right) ); } #define asm_mult 1 -#endif /* uECC_CURVE == uECC_secp160r1 */ +#elif (uECC_BYTES == 24) +__attribute((noinline)) +static void vli_mult(uint8_t *p_result, uint8_t *p_left, uint8_t *p_right) +{ + __asm__ volatile ( + "adiw r30, 20 \n\t" + "adiw r28, 20 \n\t" + "ld r2, x+ \n\t" + "ld r3, x+ \n\t" + "ld r4, x+ \n\t" + "ld r5, x+ \n\t" + "ld r12, y+ \n\t" + "ld r13, y+ \n\t" + "ld r14, y+ \n\t" + "ld r15, y+ \n\t" + "ldi r25, 0 \n\t" + + "ldi r23, 0 \n\t" + "mul r2, r12 \n\t" + "st z+, r0 \n\t" + "mov r22, r1 \n\t" + + "ldi r24, 0 \n\t" + "mul r2, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "mul r3, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r2, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r2, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ldi r24, 0 \n\t" + "mul r3, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r4, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "mul r5, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "st z+, r24 \n\t" + "st z+, r22 \n\t" + + "sbiw r30, 18 \n\t" + "sbiw r28, 14 \n\t" + "ld r12, y+ \n\t" + "ld r13, y+ \n\t" + "ld r14, y+ \n\t" + "ld r15, y+ \n\t" + "ld r16, y+ \n\t" + "ld r17, y+ \n\t" + "ld r18, y+ \n\t" + "ld r19, y+ \n\t" + "ld r20, y+ \n\t" + "ld r21, y+ \n\t" + "ld r6, x+ \n\t" + "ld r7, x+ \n\t" + "ld r8, x+ \n\t" + "ld r9, x+ \n\t" + "ld r10, x+ \n\t" + "ld r11, x+ \n\t" + + "ldi r23, 0 \n\t" + "mul r2, r12 \n\t" + "st z+, r0 \n\t" + "mov r22, r1 \n\t" + + "ldi r24, 0 \n\t" + "mul r2, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "mul r3, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r2, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r2, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ldi r24, 0 \n\t" + "mul r2, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r2, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r2, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ldi r24, 0 \n\t" + "mul r2, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r2, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r2, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r2, x+ \n\t" + "ldi r24, 0 \n\t" + "mul r3, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r3, x+ \n\t" + "ldi r22, 0 \n\t" + "mul r4, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r4, x+ \n\t" + "ldi r23, 0 \n\t" + "mul r5, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r5, x+ \n\t" + "ldi r24, 0 \n\t" + "mul r6, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r12, y+ \n\t" + "ldi r22, 0 \n\t" + "mul r6, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r13, y+ \n\t" + "ldi r23, 0 \n\t" + "mul r6, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r14, y+ \n\t" + "ldi r24, 0 \n\t" + "mul r6, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r15, y+ \n\t" + "ldi r22, 0 \n\t" + "mul r6, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r7, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ldi r24, 0 \n\t" + "mul r8, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r9, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r10, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ldi r24, 0 \n\t" + "mul r11, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r2, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r3, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ldi r24, 0 \n\t" + "mul r4, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "mul r5, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "st z+, r23 \n\t" + "st z+, r24 \n\t" + + "sbiw r30, 38 \n\t" + "sbiw r28, 24 \n\t" + "sbiw r26, 14 \n\t" + "ld r2, x+ \n\t" + "ld r12, y+ \n\t" + "ld r3, x+ \n\t" + "ld r13, y+ \n\t" + "ld r4, x+ \n\t" + "ld r14, y+ \n\t" + "ld r5, x+ \n\t" + "ld r15, y+ \n\t" + "ld r6, x+ \n\t" + "ld r16, y+ \n\t" + "ld r7, x+ \n\t" + "ld r17, y+ \n\t" + "ld r8, x+ \n\t" + "ld r18, y+ \n\t" + "ld r9, x+ \n\t" + "ld r19, y+ \n\t" + "ld r10, x+ \n\t" + "ld r20, y+ \n\t" + "ld r11, x+ \n\t" + "ld r21, y+ \n\t" + + "ldi r23, 0 \n\t" + "mul r2, r12 \n\t" + "st z+, r0 \n\t" + "mov r22, r1 \n\t" + + "ldi r24, 0 \n\t" + "mul r2, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "mul r3, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r2, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r2, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ldi r24, 0 \n\t" + "mul r2, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r2, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r2, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ldi r24, 0 \n\t" + "mul r2, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r2, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r2, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r2, x+ \n\t" + "ldi r24, 0 \n\t" + "mul r3, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r3, x+ \n\t" + "ldi r22, 0 \n\t" + "mul r4, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r4, x+ \n\t" + "ldi r23, 0 \n\t" + "mul r5, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r5, x+ \n\t" + "ldi r24, 0 \n\t" + "mul r6, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r6, x+ \n\t" + "ldi r22, 0 \n\t" + "mul r7, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r7, x+ \n\t" + "ldi r23, 0 \n\t" + "mul r8, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r8, x+ \n\t" + "ldi r24, 0 \n\t" + "mul r9, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r9, x+ \n\t" + "ldi r22, 0 \n\t" + "mul r10, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r10, x+ \n\t" + "ldi r23, 0 \n\t" + "mul r11, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r11, x+ \n\t" + "ldi r24, 0 \n\t" + "mul r2, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r2, x+ \n\t" + "ldi r22, 0 \n\t" + "mul r3, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r3, x+ \n\t" + "ldi r23, 0 \n\t" + "mul r4, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r4, x+ \n\t" + "ldi r24, 0 \n\t" + "mul r5, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r5, x+ \n\t" + "ldi r22, 0 \n\t" + "mul r6, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r12, y+ \n\t" + "ldi r23, 0 \n\t" + "mul r6, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r13, y+ \n\t" + "ldi r24, 0 \n\t" + "mul r6, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r14, y+ \n\t" + "ldi r22, 0 \n\t" + "mul r6, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r15, y+ \n\t" + "ldi r23, 0 \n\t" + "mul r6, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r16, y+ \n\t" + "ldi r24, 0 \n\t" + "mul r6, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r17, y+ \n\t" + "ldi r22, 0 \n\t" + "mul r6, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r18, y+ \n\t" + "ldi r23, 0 \n\t" + "mul r6, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r19, y+ \n\t" + "ldi r24, 0 \n\t" + "mul r6, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r20, y+ \n\t" + "ldi r22, 0 \n\t" + "mul r6, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r21, y+ \n\t" + "ldi r23, 0 \n\t" + "mul r6, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r12, y+ \n\t" + "ldi r24, 0 \n\t" + "mul r6, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r13, y+ \n\t" + "ldi r22, 0 \n\t" + "mul r6, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r14, y+ \n\t" + "ldi r23, 0 \n\t" + "mul r6, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r15, y+ \n\t" + "ldi r24, 0 \n\t" + "mul r6, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r7, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r8, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ldi r24, 0 \n\t" + "mul r9, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r10, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r11, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ldi r24, 0 \n\t" + "mul r2, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r3, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r4, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "mul r5, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "st z+, r22 \n\t" + "st z+, r23 \n\t" + + "eor r1, r1 \n\t" + : "+x" (p_left), "+y" (p_right), "+z" (p_result) + : + : "r0", "r2", "r3", "r4", "r5", "r6", "r7", "r8", "r9", "r10", "r11", "r12", + "r13", "r14", "r15", "r16", "r17", "r18", "r19", "r20", "r21", "r22", "r23", "r24", "r25", "cc", "memory" + ); +} +#define asm_mult 1 +#elif (uECC_BYTES == 32) +__attribute((noinline)) +static void vli_mult(uint8_t *p_result, uint8_t *p_left, uint8_t *p_right) +{ + __asm__ volatile ( + "adiw r30, 30 \n\t" + "adiw r28, 30 \n\t" + "ld r2, x+ \n\t" + "ld r3, x+ \n\t" + "ld r12, y+ \n\t" + "ld r13, y+ \n\t" + "ldi r25, 0 \n\t" + + "ldi r23, 0 \n\t" + "mul r2, r12 \n\t" + "st z+, r0 \n\t" + "mov r22, r1 \n\t" + + "ldi r24, 0 \n\t" + "mul r2, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "mul r3, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "mul r3, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "st z+, r23 \n\t" + "st z+, r24 \n\t" + + "sbiw r30, 14 \n\t" + "sbiw r28, 12 \n\t" + "ld r12, y+ \n\t" + "ld r13, y+ \n\t" + "ld r14, y+ \n\t" + "ld r15, y+ \n\t" + "ld r16, y+ \n\t" + "ld r17, y+ \n\t" + "ld r18, y+ \n\t" + "ld r19, y+ \n\t" + "ld r20, y+ \n\t" + "ld r21, y+ \n\t" + "ld r4, x+ \n\t" + "ld r5, x+ \n\t" + "ld r6, x+ \n\t" + "ld r7, x+ \n\t" + "ld r8, x+ \n\t" + "ld r9, x+ \n\t" + "ld r10, x+ \n\t" + "ld r11, x+ \n\t" + + "ldi r23, 0 \n\t" + "mul r2, r12 \n\t" + "st z+, r0 \n\t" + "mov r22, r1 \n\t" + + "ldi r24, 0 \n\t" + "mul r2, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "mul r3, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r2, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r2, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ldi r24, 0 \n\t" + "mul r2, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r2, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r2, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ldi r24, 0 \n\t" + "mul r2, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r2, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r2, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r2, x+ \n\t" + "ldi r24, 0 \n\t" + "mul r3, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r3, x+ \n\t" + "ldi r22, 0 \n\t" + "mul r4, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r12, y+ \n\t" + "ldi r23, 0 \n\t" + "mul r4, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r13, y+ \n\t" + "ldi r24, 0 \n\t" + "mul r4, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r5, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r6, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ldi r24, 0 \n\t" + "mul r7, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r8, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r9, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ldi r24, 0 \n\t" + "mul r10, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r11, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r2, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "mul r3, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "st z+, r22 \n\t" + "st z+, r23 \n\t" + + "sbiw r30, 34 \n\t" + "sbiw r28, 22 \n\t" + "sbiw r26, 12 \n\t" + "ld r2, x+ \n\t" + "ld r12, y+ \n\t" + "ld r3, x+ \n\t" + "ld r13, y+ \n\t" + "ld r4, x+ \n\t" + "ld r14, y+ \n\t" + "ld r5, x+ \n\t" + "ld r15, y+ \n\t" + "ld r6, x+ \n\t" + "ld r16, y+ \n\t" + "ld r7, x+ \n\t" + "ld r17, y+ \n\t" + "ld r8, x+ \n\t" + "ld r18, y+ \n\t" + "ld r9, x+ \n\t" + "ld r19, y+ \n\t" + "ld r10, x+ \n\t" + "ld r20, y+ \n\t" + "ld r11, x+ \n\t" + "ld r21, y+ \n\t" + + "ldi r23, 0 \n\t" + "mul r2, r12 \n\t" + "st z+, r0 \n\t" + "mov r22, r1 \n\t" + + "ldi r24, 0 \n\t" + "mul r2, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "mul r3, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r2, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r2, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ldi r24, 0 \n\t" + "mul r2, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r2, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r2, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ldi r24, 0 \n\t" + "mul r2, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r2, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r2, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r2, x+ \n\t" + "ldi r24, 0 \n\t" + "mul r3, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r3, x+ \n\t" + "ldi r22, 0 \n\t" + "mul r4, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r4, x+ \n\t" + "ldi r23, 0 \n\t" + "mul r5, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r5, x+ \n\t" + "ldi r24, 0 \n\t" + "mul r6, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r6, x+ \n\t" + "ldi r22, 0 \n\t" + "mul r7, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r7, x+ \n\t" + "ldi r23, 0 \n\t" + "mul r8, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r8, x+ \n\t" + "ldi r24, 0 \n\t" + "mul r9, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r9, x+ \n\t" + "ldi r22, 0 \n\t" + "mul r10, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r10, x+ \n\t" + "ldi r23, 0 \n\t" + "mul r11, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r11, x+ \n\t" + "ldi r24, 0 \n\t" + "mul r2, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r2, x+ \n\t" + "ldi r22, 0 \n\t" + "mul r3, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r3, x+ \n\t" + "ldi r23, 0 \n\t" + "mul r4, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r12, y+ \n\t" + "ldi r24, 0 \n\t" + "mul r4, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r13, y+ \n\t" + "ldi r22, 0 \n\t" + "mul r4, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r14, y+ \n\t" + "ldi r23, 0 \n\t" + "mul r4, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r15, y+ \n\t" + "ldi r24, 0 \n\t" + "mul r4, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r16, y+ \n\t" + "ldi r22, 0 \n\t" + "mul r4, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r17, y+ \n\t" + "ldi r23, 0 \n\t" + "mul r4, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r18, y+ \n\t" + "ldi r24, 0 \n\t" + "mul r4, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r19, y+ \n\t" + "ldi r22, 0 \n\t" + "mul r4, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r20, y+ \n\t" + "ldi r23, 0 \n\t" + "mul r4, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r21, y+ \n\t" + "ldi r24, 0 \n\t" + "mul r4, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r12, y+ \n\t" + "ldi r22, 0 \n\t" + "mul r4, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r13, y+ \n\t" + "ldi r23, 0 \n\t" + "mul r4, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ldi r24, 0 \n\t" + "mul r5, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r6, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r7, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ldi r24, 0 \n\t" + "mul r8, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r9, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r10, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ldi r24, 0 \n\t" + "mul r11, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r2, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "mul r3, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "st z+, r24 \n\t" + "st z+, r22 \n\t" + + "sbiw r30, 54 \n\t" + "sbiw r28, 32 \n\t" + "sbiw r26, 22 \n\t" + "ld r2, x+ \n\t" + "ld r12, y+ \n\t" + "ld r3, x+ \n\t" + "ld r13, y+ \n\t" + "ld r4, x+ \n\t" + "ld r14, y+ \n\t" + "ld r5, x+ \n\t" + "ld r15, y+ \n\t" + "ld r6, x+ \n\t" + "ld r16, y+ \n\t" + "ld r7, x+ \n\t" + "ld r17, y+ \n\t" + "ld r8, x+ \n\t" + "ld r18, y+ \n\t" + "ld r9, x+ \n\t" + "ld r19, y+ \n\t" + "ld r10, x+ \n\t" + "ld r20, y+ \n\t" + "ld r11, x+ \n\t" + "ld r21, y+ \n\t" + + "ldi r23, 0 \n\t" + "mul r2, r12 \n\t" + "st z+, r0 \n\t" + "mov r22, r1 \n\t" + + "ldi r24, 0 \n\t" + "mul r2, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "mul r3, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r2, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r2, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ldi r24, 0 \n\t" + "mul r2, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r2, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r2, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ldi r24, 0 \n\t" + "mul r2, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r2, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r2, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r2, x+ \n\t" + "ldi r24, 0 \n\t" + "mul r3, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r3, x+ \n\t" + "ldi r22, 0 \n\t" + "mul r4, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r4, x+ \n\t" + "ldi r23, 0 \n\t" + "mul r5, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r5, x+ \n\t" + "ldi r24, 0 \n\t" + "mul r6, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r6, x+ \n\t" + "ldi r22, 0 \n\t" + "mul r7, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r7, x+ \n\t" + "ldi r23, 0 \n\t" + "mul r8, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r8, x+ \n\t" + "ldi r24, 0 \n\t" + "mul r9, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r9, x+ \n\t" + "ldi r22, 0 \n\t" + "mul r10, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r10, x+ \n\t" + "ldi r23, 0 \n\t" + "mul r11, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r11, x+ \n\t" + "ldi r24, 0 \n\t" + "mul r2, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r2, x+ \n\t" + "ldi r22, 0 \n\t" + "mul r3, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r3, x+ \n\t" + "ldi r23, 0 \n\t" + "mul r4, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r4, x+ \n\t" + "ldi r24, 0 \n\t" + "mul r5, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r5, x+ \n\t" + "ldi r22, 0 \n\t" + "mul r6, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r6, x+ \n\t" + "ldi r23, 0 \n\t" + "mul r7, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r7, x+ \n\t" + "ldi r24, 0 \n\t" + "mul r8, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r8, x+ \n\t" + "ldi r22, 0 \n\t" + "mul r9, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r9, x+ \n\t" + "ldi r23, 0 \n\t" + "mul r10, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r10, x+ \n\t" + "ldi r24, 0 \n\t" + "mul r11, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r4, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r11, x+ \n\t" + "ldi r22, 0 \n\t" + "mul r2, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r4, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r2, x+ \n\t" + "ldi r23, 0 \n\t" + "mul r3, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r4, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r3, x+ \n\t" + "ldi r24, 0 \n\t" + "mul r4, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r12, y+ \n\t" + "ldi r22, 0 \n\t" + "mul r4, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r13, y+ \n\t" + "ldi r23, 0 \n\t" + "mul r4, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r14, y+ \n\t" + "ldi r24, 0 \n\t" + "mul r4, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r15, y+ \n\t" + "ldi r22, 0 \n\t" + "mul r4, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r16, y+ \n\t" + "ldi r23, 0 \n\t" + "mul r4, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r17, y+ \n\t" + "ldi r24, 0 \n\t" + "mul r4, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r18, y+ \n\t" + "ldi r22, 0 \n\t" + "mul r4, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r19, y+ \n\t" + "ldi r23, 0 \n\t" + "mul r4, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r20, y+ \n\t" + "ldi r24, 0 \n\t" + "mul r4, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r21, y+ \n\t" + "ldi r22, 0 \n\t" + "mul r4, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r12, y+ \n\t" + "ldi r23, 0 \n\t" + "mul r4, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r13, y+ \n\t" + "ldi r24, 0 \n\t" + "mul r4, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r14, y+ \n\t" + "ldi r22, 0 \n\t" + "mul r4, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r15, y+ \n\t" + "ldi r23, 0 \n\t" + "mul r4, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r16, y+ \n\t" + "ldi r24, 0 \n\t" + "mul r4, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r17, y+ \n\t" + "ldi r22, 0 \n\t" + "mul r4, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r18, y+ \n\t" + "ldi r23, 0 \n\t" + "mul r4, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r19, y+ \n\t" + "ldi r24, 0 \n\t" + "mul r4, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r20, y+ \n\t" + "ldi r22, 0 \n\t" + "mul r4, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ld r21, y+ \n\t" + "ldi r23, 0 \n\t" + "mul r4, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r5, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r14 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "ld r0, z \n\t" + "add r24, r0 \n\t" + "adc r22, r25 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ld r12, y+ \n\t" + "ldi r24, 0 \n\t" + "mul r4, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r5, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r6, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r15 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r14 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "ld r0, z \n\t" + "add r22, r0 \n\t" + "adc r23, r25 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ld r13, y+ \n\t" + "ldi r22, 0 \n\t" + "mul r4, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r5, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r6, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r7, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r16 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r15 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r14 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "ld r0, z \n\t" + "add r23, r0 \n\t" + "adc r24, r25 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r5, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r6, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r7, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r8, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r17 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r16 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r15 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ldi r24, 0 \n\t" + "mul r6, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r7, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r8, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r9, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r18 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r17 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r16 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r7, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r8, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r9, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r10, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r19 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r18 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r17 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r8, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r9, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r10, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r11, r20 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r19 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r18 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ldi r24, 0 \n\t" + "mul r9, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r10, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r11, r21 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r2, r20 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r19 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "ldi r22, 0 \n\t" + "mul r10, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r11, r12 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r2, r21 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "mul r3, r20 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "adc r22, r25 \n\t" + "st z+, r23 \n\t" + + "ldi r23, 0 \n\t" + "mul r11, r13 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r2, r12 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "mul r3, r21 \n\t" + "add r24, r0 \n\t" + "adc r22, r1 \n\t" + "adc r23, r25 \n\t" + "st z+, r24 \n\t" + + "ldi r24, 0 \n\t" + "mul r2, r13 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "mul r3, r12 \n\t" + "add r22, r0 \n\t" + "adc r23, r1 \n\t" + "adc r24, r25 \n\t" + "st z+, r22 \n\t" + + "mul r3, r13 \n\t" + "add r23, r0 \n\t" + "adc r24, r1 \n\t" + "st z+, r23 \n\t" + "st z+, r24 \n\t" + + "eor r1, r1 \n\t" + : "+x" (p_left), "+y" (p_right), "+z" (p_result) + : + : "r0", "r2", "r3", "r4", "r5", "r6", "r7", "r8", "r9", "r10", "r11", "r12", + "r13", "r14", "r15", "r16", "r17", "r18", "r19", "r20", "r21", "r22", "r23", "r24", "r25", "cc", "memory" + ); +} +#define asm_mult 1 +#endif /* uECC_BYTES == 32 */ #if uECC_SQUARE_FUNC -#if uECC_CURVE == uECC_secp160r1 +#if (uECC_BYTES == 20) static void vli_square(uint8_t *p_result, uint8_t *p_left) { __asm__ volatile ( diff --git a/scripts/mult.py b/scripts/mult.py deleted file mode 100755 index a1765af..0000000 --- a/scripts/mult.py +++ /dev/null @@ -1,162 +0,0 @@ -#!/usr/bin/env python - -def rx(i): - return i + 2 - -def ry(i): - return i + 12 - -def emit(line, *args): - s = '"' + line + r' \n\t"' - print s % args - -#### set up registers -emit("adiw r30, 10") -emit("adiw r28, 10") - -for i in xrange(10): - emit("ld r%s, x+", rx(i)) -for i in xrange(10): - emit("ld r%s, y+", ry(i)) - -#### first two multiplications of initial block (x = 0-9, y = 10-19) -emit("ldi r25, 0") -print "" -emit("ldi r23, 0") -emit("mul r2, r12") -emit("st z+, r0") -emit("mov r22, r1") -print "" -emit("ldi r24, 0") -emit("mul r2, r13") -emit("add r22, r0") -emit("adc r23, r1") -emit("mul r3, r12") -emit("add r22, r0") -emit("adc r23, r1") -emit("adc r24, r25") -emit("st z+, r22") -print "" - -#### rest of initial block, with moving accumulator registers -acc = [23, 24, 22] -for r in xrange(2, 10): - emit("ldi r%s, 0", acc[2]) - for i in xrange(0, r+1): - emit("mul r%s, r%s", rx(i), ry(r - i)) - emit("add r%s, r0", acc[0]) - emit("adc r%s, r1", acc[1]) - emit("adc r%s, r25", acc[2]) - emit("st z+, r%s", acc[0]) - print "" - acc = acc[1:] + acc[:1] -for r in xrange(1, 9): - emit("ldi r%s, 0", acc[2]) - for i in xrange(0, 10-r): - emit("mul r%s, r%s", rx(r+i), ry(9 - i)) - emit("add r%s, r0", acc[0]) - emit("adc r%s, r1", acc[1]) - emit("adc r%s, r25", acc[2]) - emit("st z+, r%s", acc[0]) - print "" - acc = acc[1:] + acc[:1] -emit("mul r%s, r%s", rx(9), ry(9)) -emit("add r%s, r0", acc[0]) -emit("adc r%s, r1", acc[1]) -emit("st z+, r%s", acc[0]) -emit("st z+, r%s", acc[1]) -print "" - -#### reset y and z pointers (x still points to left + 10) -emit("sbiw r30, 30") -emit("sbiw r28, 20") - -#### load y registers -for i in xrange(10): - emit("ld r%s, y+", ry(i)) -print "" - -#### do x = 0-9, y = 0-9 multiplications -emit("ldi r23, 0") -emit("mul r2, r12") -emit("st z+, r0") -emit("mov r22, r1") -print "" -emit("ldi r24, 0") -emit("mul r2, r13") -emit("add r22, r0") -emit("adc r23, r1") -emit("mul r3, r12") -emit("add r22, r0") -emit("adc r23, r1") -emit("adc r24, r25") -emit("st z+, r22") -print "" - -acc = [23, 24, 22] -for r in xrange(2, 10): - emit("ldi r%s, 0", acc[2]) - for i in xrange(0, r+1): - emit("mul r%s, r%s", rx(i), ry(r - i)) - emit("add r%s, r0", acc[0]) - emit("adc r%s, r1", acc[1]) - emit("adc r%s, r25", acc[2]) - emit("st z+, r%s", acc[0]) - print "" - acc = acc[1:] + acc[:1] - -#### now we need to start shifting x and loading from z -x_regs = [2, 3, 4, 5, 6, 7, 8, 9, 10, 11] -for r in xrange(0, 10): - x_regs = x_regs[1:] + x_regs[:1] - emit("ld r%s, x+", x_regs[9]) # load next byte of left - emit("ldi r%s, 0", acc[2]) - for i in xrange(0, 10): - emit("mul r%s, r%s", x_regs[i], ry(9 - i)) - emit("add r%s, r0", acc[0]) - emit("adc r%s, r1", acc[1]) - emit("adc r%s, r25", acc[2]) - emit("ld r0, z") # load stored value from initial block, and add to accumulator (note z does not increment) - emit("add r%s, r0", acc[0]) - emit("adc r%s, r25", acc[1]) - emit("adc r%s, r25", acc[2]) - emit("st z+, r%s", acc[0]) # store next byte (z increments) - print "" - acc = acc[1:] + acc[:1] - -# done shifting x, start shifting y -y_regs = [12, 13, 14, 15, 16, 17, 18, 19, 20, 21] -for r in xrange(0, 10): - y_regs = y_regs[1:] + y_regs[:1] - emit("ld r%s, y+", y_regs[9]) # load next byte of right - emit("ldi r%s, 0", acc[2]) - for i in xrange(0, 10): - emit("mul r%s, r%s", x_regs[i], y_regs[9 -i]) - emit("add r%s, r0", acc[0]) - emit("adc r%s, r1", acc[1]) - emit("adc r%s, r25", acc[2]) - emit("ld r0, z") # load stored value from initial block, and add to accumulator (note z does not increment) - emit("add r%s, r0", acc[0]) - emit("adc r%s, r25", acc[1]) - emit("adc r%s, r25", acc[2]) - emit("st z+, r%s", acc[0]) # store next byte (z increments) - print "" - acc = acc[1:] + acc[:1] - -# done both shifts, do remaining corner -for r in xrange(1, 9): - emit("ldi r%s, 0", acc[2]) - for i in xrange(0, 10-r): - emit("mul r%s, r%s", x_regs[r+i], y_regs[9 - i]) - emit("add r%s, r0", acc[0]) - emit("adc r%s, r1", acc[1]) - emit("adc r%s, r25", acc[2]) - emit("st z+, r%s", acc[0]) - print "" - acc = acc[1:] + acc[:1] -emit("mul r%s, r%s", x_regs[9], y_regs[9]) -emit("add r%s, r0", acc[0]) -emit("adc r%s, r1", acc[1]) -emit("st z+, r%s", acc[0]) -emit("st z+, r%s", acc[1]) -emit("eor r1, r1") diff --git a/scripts/mult_avr.py b/scripts/mult_avr.py new file mode 100755 index 0000000..02f0510 --- /dev/null +++ b/scripts/mult_avr.py @@ -0,0 +1,199 @@ +#!/usr/bin/env python + +import sys + +if len(sys.argv) < 2: + print "Provide the integer size in bytes" + sys.exit(1) + +size = int(sys.argv[1]) + +full_rows = size // 10 +init_size = size % 10 + +if init_size == 0: + full_rows = full_rows - 1 + init_size = 10 + +def rx(i): + return i + 2 + +def ry(i): + return i + 12 + +def emit(line, *args): + s = '"' + line + r' \n\t"' + print s % args + +#### set up registers +emit("adiw r30, %s" % (size - init_size)) # move z +emit("adiw r28, %s" % (size - init_size)) # move y + +for i in xrange(init_size): + emit("ld r%s, x+", rx(i)) +for i in xrange(init_size): + emit("ld r%s, y+", ry(i)) + +# note that this code assume that init_size > 1 +#### first two multiplications of initial block +emit("ldi r25, 0") +print "" +emit("ldi r23, 0") +emit("mul r2, r12") +emit("st z+, r0") +emit("mov r22, r1") +print "" +emit("ldi r24, 0") +emit("mul r2, r13") +emit("add r22, r0") +emit("adc r23, r1") +emit("mul r3, r12") +emit("add r22, r0") +emit("adc r23, r1") +emit("adc r24, r25") +emit("st z+, r22") +print "" + +#### rest of initial block, with moving accumulator registers +acc = [23, 24, 22] +for r in xrange(2, init_size): + emit("ldi r%s, 0", acc[2]) + for i in xrange(0, r+1): + emit("mul r%s, r%s", rx(i), ry(r - i)) + emit("add r%s, r0", acc[0]) + emit("adc r%s, r1", acc[1]) + emit("adc r%s, r25", acc[2]) + emit("st z+, r%s", acc[0]) + print "" + acc = acc[1:] + acc[:1] +for r in xrange(1, init_size-1): + emit("ldi r%s, 0", acc[2]) + for i in xrange(0, init_size-r): + emit("mul r%s, r%s", rx(r+i), ry((init_size-1) - i)) + emit("add r%s, r0", acc[0]) + emit("adc r%s, r1", acc[1]) + emit("adc r%s, r25", acc[2]) + emit("st z+, r%s", acc[0]) + print "" + acc = acc[1:] + acc[:1] +emit("mul r%s, r%s", rx(init_size-1), ry(init_size-1)) +emit("add r%s, r0", acc[0]) +emit("adc r%s, r1", acc[1]) +emit("st z+, r%s", acc[0]) +emit("st z+, r%s", acc[1]) +print "" + +#### reset y and z pointers +emit("sbiw r30, %s" % (2 * init_size + 10)) +emit("sbiw r28, %s" % (init_size + 10)) + +#### load y registers +for i in xrange(10): + emit("ld r%s, y+", ry(i)) + +#### load additional x registers +for i in xrange(init_size, 10): + emit("ld r%s, x+", rx(i)) +print "" + +prev_size = init_size +for row in xrange(full_rows): + #### do x = 0-9, y = 0-9 multiplications + emit("ldi r23, 0") + emit("mul r2, r12") + emit("st z+, r0") + emit("mov r22, r1") + print "" + emit("ldi r24, 0") + emit("mul r2, r13") + emit("add r22, r0") + emit("adc r23, r1") + emit("mul r3, r12") + emit("add r22, r0") + emit("adc r23, r1") + emit("adc r24, r25") + emit("st z+, r22") + print "" + + acc = [23, 24, 22] + for r in xrange(2, 10): + emit("ldi r%s, 0", acc[2]) + for i in xrange(0, r+1): + emit("mul r%s, r%s", rx(i), ry(r - i)) + emit("add r%s, r0", acc[0]) + emit("adc r%s, r1", acc[1]) + emit("adc r%s, r25", acc[2]) + emit("st z+, r%s", acc[0]) + print "" + acc = acc[1:] + acc[:1] + + #### now we need to start shifting x and loading from z + x_regs = [2, 3, 4, 5, 6, 7, 8, 9, 10, 11] + for r in xrange(0, prev_size): + x_regs = x_regs[1:] + x_regs[:1] + emit("ld r%s, x+", x_regs[9]) # load next byte of left + emit("ldi r%s, 0", acc[2]) + for i in xrange(0, 10): + emit("mul r%s, r%s", x_regs[i], ry(9 - i)) + emit("add r%s, r0", acc[0]) + emit("adc r%s, r1", acc[1]) + emit("adc r%s, r25", acc[2]) + emit("ld r0, z") # load stored value from initial block, and add to accumulator (note z does not increment) + emit("add r%s, r0", acc[0]) + emit("adc r%s, r25", acc[1]) + emit("adc r%s, r25", acc[2]) + emit("st z+, r%s", acc[0]) # store next byte (z increments) + print "" + acc = acc[1:] + acc[:1] + + # done shifting x, start shifting y + y_regs = [12, 13, 14, 15, 16, 17, 18, 19, 20, 21] + for r in xrange(0, prev_size): + y_regs = y_regs[1:] + y_regs[:1] + emit("ld r%s, y+", y_regs[9]) # load next byte of right + emit("ldi r%s, 0", acc[2]) + for i in xrange(0, 10): + emit("mul r%s, r%s", x_regs[i], y_regs[9 -i]) + emit("add r%s, r0", acc[0]) + emit("adc r%s, r1", acc[1]) + emit("adc r%s, r25", acc[2]) + emit("ld r0, z") # load stored value from initial block, and add to accumulator (note z does not increment) + emit("add r%s, r0", acc[0]) + emit("adc r%s, r25", acc[1]) + emit("adc r%s, r25", acc[2]) + emit("st z+, r%s", acc[0]) # store next byte (z increments) + print "" + acc = acc[1:] + acc[:1] + + # done both shifts, do remaining corner + for r in xrange(1, 9): + emit("ldi r%s, 0", acc[2]) + for i in xrange(0, 10-r): + emit("mul r%s, r%s", x_regs[r+i], y_regs[9 - i]) + emit("add r%s, r0", acc[0]) + emit("adc r%s, r1", acc[1]) + emit("adc r%s, r25", acc[2]) + emit("st z+, r%s", acc[0]) + print "" + acc = acc[1:] + acc[:1] + emit("mul r%s, r%s", x_regs[9], y_regs[9]) + emit("add r%s, r0", acc[0]) + emit("adc r%s, r1", acc[1]) + emit("st z+, r%s", acc[0]) + emit("st z+, r%s", acc[1]) + print "" + + prev_size = prev_size + 10 + if row < full_rows - 1: + #### reset x, y and z pointers + emit("sbiw r30, %s" % (2 * prev_size + 10)) + emit("sbiw r28, %s" % (prev_size + 10)) + emit("sbiw r26, %s" % (prev_size)) + + #### load x and y registers + for i in xrange(10): + emit("ld r%s, x+", rx(i)) + emit("ld r%s, y+", ry(i)) + print "" + +emit("eor r1, r1") diff --git a/scripts/square.py b/scripts/square_avr.py similarity index 100% rename from scripts/square.py rename to scripts/square_avr.py diff --git a/uECC.c b/uECC.c index ba4bb2c..abeca4a 100644 --- a/uECC.c +++ b/uECC.c @@ -561,7 +561,7 @@ static uECC_word_t vli_sub(uECC_word_t *p_result, uECC_word_t *p_left, uECC_word } #endif -#if (!asm_mult || !asm_square) +#if (!asm_mult || !asm_square || uECC_CURVE == uECC_secp256k1) static void muladd(uECC_word_t a, uECC_word_t b, uECC_word_t *r0, uECC_word_t *r1, uECC_word_t *r2) { #if uECC_WORD_SIZE == 8 && !SUPPORTS_INT128 @@ -1312,7 +1312,6 @@ static void omega_mult(uint64_t * RESTRICT p_result, uint64_t * RESTRICT p_right #endif /* uECC_CURVE */ #endif /* !asm_mmod_fast */ - /* Computes p_result = (p_left * p_right) % curve_p. */ static void vli_modMult_fast(uECC_word_t *p_result, uECC_word_t *p_left, uECC_word_t *p_right) { @@ -1698,8 +1697,6 @@ static void mod_sqrt(uECC_word_t *a) vli_set(a, l_result); } - - #if uECC_WORD_SIZE == 1 static void vli_nativeToBytes(uint8_t * RESTRICT p_dest, const uint8_t * RESTRICT p_src)