nudt-compiler-cpp/2026test_results/performance/transpose2.s

    .data
    .globl matrix
    .p2align 2
matrix:
    .zero 80000000
    .globl a
    .p2align 2
a:
    .zero 400000

    .text
    .globl transpose
    .p2align 2
transpose:
.L.transpose.0:
    stp x29, x30, [sp, #-16]!
    mov x29, sp
    sub sp, sp, #48
    str x22, [sp, #0]
    str x21, [sp, #8]
    str x20, [sp, #16]
    str x19, [sp, #24]
    mov w8, w0
    mov x11, x1
    mov w10, w2
    sdiv w12, w8, w10
    mov w8, #0
    mov w9, w8
    b .L.transpose.1
.L.transpose.1:
    cmp w9, w12
    b.lt .L.transpose.2
    b .L.transpose.3
.L.transpose.2:
    mov w8, #0
    mov w19, w8
    mov w15, w9
    b .L.transpose.4
.L.transpose.3:
    mov w8, #-1
    mov w0, w8
    ldr x22, [sp, #0]
    ldr x21, [sp, #8]
    ldr x20, [sp, #16]
    ldr x19, [sp, #24]
    add sp, sp, #48
    ldp x29, x30, [sp], #16
    ret
.L.transpose.4:
    cmp w19, w10
    b.lt .L.transpose.5
    b .L.transpose.6
.L.transpose.5:
    cmp w15, w19
    b.lt .L.transpose.7
    b .L.transpose.8
.L.transpose.6:
    mov w8, #1
    add w8, w15, w8
    mov w9, w8
    b .L.transpose.1
.L.transpose.7:
    mov w8, #1
    add w8, w19, w8
    mov w19, w8
    b .L.transpose.4
.L.transpose.8:
    mul w8, w15, w10
    add w20, w8, w19
    sxtw x8, w20
    lsl x8, x8, #2
    add x8, x11, x8
    ldr w21, [x8]
    mul w8, w19, w12
    add w22, w8, w15
    sxtw x8, w20
    lsl x8, x8, #2
    add x8, x11, x8
    ldr w9, [x8]
    sxtw x8, w22
    lsl x8, x8, #2
    add x8, x11, x8
    str w9, [x8]
    sxtw x8, w20
    lsl x8, x8, #2
    add x8, x11, x8
    str w21, [x8]
    mov w8, #1
    add w8, w19, w8
    mov w19, w8
    b .L.transpose.4

    .text
    .globl main
    .p2align 2
main:
.L.main.0:
    stp x29, x30, [sp, #-16]!
    mov x29, sp
    sub sp, sp, #48
    str x19, [sp, #0]
    str x21, [sp, #8]
    str x20, [sp, #16]
    bl getint
    mov w20, w0
    adrp x8, a
    add x8, x8, :lo12:a
    mov x0, x8
    bl getarray
    mov w19, w0
    mov w8, #28
    mov w0, w8
    bl _sysy_starttime
    mov w8, #0
    mov w10, w8
    b .L.main.1
.L.main.1:
    cmp w10, w20
    b.lt .L.main.2
    b .L.main.3
.L.main.2:
    adrp x8, matrix
    add x8, x8, :lo12:matrix
    sxtw x9, w10
    lsl x9, x9, #2
    add x8, x8, x9
    str w10, [x8]
    mov w8, #4
    add w8, w10, #3
    cmp w10, #0
    csel w8, w8, w10, lt
    asr w8, w8, #2
    mov w9, #4
    msub w9, w8, w9, w10
    mov w8, #0
    cmp w9, w8
    b.eq .L.main.4
    b .L.main.5
.L.main.3:
    mov w8, #0
    mov w21, w8
    b .L.main.6
.L.main.4:
    adrp x8, matrix
    add x8, x8, :lo12:matrix
    sxtw x9, w10
    lsl x9, x9, #2
    add x8, x8, x9
    mov w9, #4
    str w9, [x8]
    b .L.main.5
.L.main.5:
    mov w8, #1
    add w8, w10, w8
    mov w10, w8
    b .L.main.1
.L.main.6:
    cmp w21, w19
    b.lt .L.main.7
    b .L.main.8
.L.main.7:
    adrp x8, a
    add x8, x8, :lo12:a
    sxtw x9, w21
    lsl x9, x9, #2
    add x8, x8, x9
    ldr w9, [x8]
    mov w0, w20
    adrp x8, matrix
    add x8, x8, :lo12:matrix
    mov x1, x8
    mov w2, w9
    bl transpose
    mov w8, w0
    mov w8, #1
    add w8, w21, w8
    mov w21, w8
    b .L.main.6
.L.main.8:
    mov w8, #0
    mov w11, w8
    mov w12, w8
    b .L.main.9
.L.main.9:
    cmp w12, w19
    b.lt .L.main.10
    b .L.main.11
.L.main.10:
    mul w10, w12, w12
    adrp x9, matrix
    add x9, x9, :lo12:matrix
    sxtw x8, w12
    lsl x8, x8, #2
    add x8, x9, x8
    ldr w8, [x8]
    mul w8, w10, w8
    add w9, w11, w8
    mov w8, #1
    add w8, w12, w8
    mov w11, w9
    mov w12, w8
    b .L.main.9
.L.main.11:
    mov w8, #0
    cmp w11, w8
    b.lt .L.main.12
    mov w19, w11
    b .L.main.13
.L.main.12:
    mov w8, #0
    sub w8, w8, w11
    mov w19, w8
    b .L.main.13
.L.main.13:
    mov w8, #49
    mov w0, w8
    bl _sysy_stoptime
    mov w0, w19
    bl putint
    mov w8, #10
    mov w0, w8
    bl putch
    mov w8, #0
    mov w0, w8
    ldr x19, [sp, #0]
    ldr x21, [sp, #8]
    ldr x20, [sp, #16]
    add sp, sp, #48
    ldp x29, x30, [sp], #16
    ret