Page 206 - ARM 64 Bit Assembly Language

194 Chapter 7

3 y: .8byte 0x75
4 msg: .asciz "%lx * %lx = %lx\n"
5 .text
6 .type main %function
7 .global main
8 main: stp x29, x30, [sp, #-16]!
9 mov x0, #0 // x0 is result
10 ldr x1, x // x1 is multiplicand
11 ldr x2, y // x2 is multiplier
12 loop: tst x2, 0x1 // check LSB
13 lsr x2, x2, #1 // shift multiplier right
14 add x3, x0, x1
15 csel x0, x3, x0, ne // Add if LSB was 1
16 lsl x1, x1, #1 // shift multiplicand left
17 cmp x2, xzr
18 bne loop // if (multiplier == 0), we are done
19 // printf("%lx * %lx = %lx\n")
20 mov x3, x0
21 ldr x2, x
22 ldr x1, y
23 adr x0, msg
24 bl printf
25 mov w0, #0 // return 0
26 ldp x29, x30, [sp], #16
27 ret
28 .size main,(. - main)

7.2.3 Signed multiplication

Consider the two multiplication problems shown in Fig. 7.1 and Fig. 7.2. Note that the result
of a multiply depends on whether the numbers are interpreted as unsigned numbers or signed
numbers. For this reason, most computer CPU’s have two different multiply operations for
signed and unsigned numbers.

11011001
−39
× 01001001
× 73
1111111111011001
657 =
1111111011001
219
1111011001
−2847
1111010011100001

Figure 7.1: In signed 8-bit math, 11011001 2 is −39 10 .