-rw-r--r-- 28499 libmceliece-20240513/crypto_kem/348864/avx/vec_mul_asm.S raw
// 20240504 djb: add note.GNU-stack
// 20221231 djb: port hidden to macos; tnx thom wiggers
// 20221230 djb: add linker line

// linker define vec_mul_asm

#include "crypto_asm_hidden.h"

#define vec_mul_asm CRYPTO_SHARED_NAMESPACE(vec_mul_asm)
#define _vec_mul_asm _CRYPTO_SHARED_NAMESPACE(vec_mul_asm)

# qhasm: int64 input_0

# qhasm: int64 input_1

# qhasm: int64 input_2

# qhasm: int64 input_3

# qhasm: int64 input_4

# qhasm: int64 input_5

# qhasm: stack64 input_6

# qhasm: stack64 input_7

# qhasm: int64 caller_r11

# qhasm: int64 caller_r12

# qhasm: int64 caller_r13

# qhasm: int64 caller_r14

# qhasm: int64 caller_r15

# qhasm: int64 caller_rbx

# qhasm: int64 caller_rbp

# qhasm: reg256 s0

# qhasm: reg256 s1

# qhasm: reg256 s2

# qhasm: reg256 s3

# qhasm: reg256 s4

# qhasm: reg256 s5

# qhasm: reg256 t0

# qhasm: reg256 t1

# qhasm: reg256 t2

# qhasm: reg256 b0

# qhasm: reg256 b1

# qhasm: reg256 b2

# qhasm: reg256 b3

# qhasm: reg256 b4

# qhasm: reg256 b5

# qhasm: reg256 a0

# qhasm: reg256 a1

# qhasm: reg256 a2

# qhasm: reg256 a3

# qhasm: reg256 a4

# qhasm: reg256 a5

# qhasm: reg256 r0

# qhasm: reg256 r1

# qhasm: reg256 r2

# qhasm: reg256 r3

# qhasm: reg256 r4

# qhasm: reg256 r5

# qhasm: reg256 r6

# qhasm: reg256 r7

# qhasm: reg256 r8

# qhasm: reg256 r9

# qhasm: reg256 r10

# qhasm: reg256 r11

# qhasm: reg256 r12

# qhasm: reg256 r13

# qhasm: reg256 r14

# qhasm: reg256 r15

# qhasm: reg256 r16

# qhasm: reg256 r17

# qhasm: reg256 r18

# qhasm: reg256 r19

# qhasm: reg256 r20

# qhasm: reg256 r21

# qhasm: reg256 r22

# qhasm: reg256 r

# qhasm: int64 h0

# qhasm: int64 h1

# qhasm: int64 h2

# qhasm: int64 h3

# qhasm: int64 h4

# qhasm: int64 h5

# qhasm: int64 h6

# qhasm: int64 h7

# qhasm: int64 h8

# qhasm: int64 h9

# qhasm: int64 h10

# qhasm: int64 h11

# qhasm: int64 h12

# qhasm: int64 h13

# qhasm: int64 h14

# qhasm: int64 h15

# qhasm: int64 h16

# qhasm: int64 h17

# qhasm: int64 h18

# qhasm: int64 h19

# qhasm: int64 h20

# qhasm: int64 h21

# qhasm: int64 h22

# qhasm: stack4864 buf

# qhasm: int64 ptr

# qhasm: int64 tmp

# qhasm: stack64 r11_stack

# qhasm: stack64 r12_stack

# qhasm: stack64 r13_stack

# qhasm: stack64 r14_stack

# qhasm: stack64 r15_stack

# qhasm: stack64 rbx_stack

# qhasm: stack64 rbp_stack

# qhasm: enter vec_mul_asm
.p2align 5
ASM_HIDDEN _vec_mul_asm
ASM_HIDDEN vec_mul_asm
.global _vec_mul_asm
.global vec_mul_asm
_vec_mul_asm:
vec_mul_asm:
mov %rsp,%r11
and $31,%r11
add $672,%r11
sub %r11,%rsp

# qhasm: r11_stack = caller_r11
# asm 1: movq <caller_r11=int64#9,>r11_stack=stack64#1
# asm 2: movq <caller_r11=%r11,>r11_stack=608(%rsp)
movq %r11,608(%rsp)

# qhasm: r12_stack = caller_r12
# asm 1: movq <caller_r12=int64#10,>r12_stack=stack64#2
# asm 2: movq <caller_r12=%r12,>r12_stack=616(%rsp)
movq %r12,616(%rsp)

# qhasm: r13_stack = caller_r13
# asm 1: movq <caller_r13=int64#11,>r13_stack=stack64#3
# asm 2: movq <caller_r13=%r13,>r13_stack=624(%rsp)
movq %r13,624(%rsp)

# qhasm: r14_stack = caller_r14
# asm 1: movq <caller_r14=int64#12,>r14_stack=stack64#4
# asm 2: movq <caller_r14=%r14,>r14_stack=632(%rsp)
movq %r14,632(%rsp)

# qhasm: r15_stack = caller_r15
# asm 1: movq <caller_r15=int64#13,>r15_stack=stack64#5
# asm 2: movq <caller_r15=%r15,>r15_stack=640(%rsp)
movq %r15,640(%rsp)

# qhasm: rbx_stack = caller_rbx
# asm 1: movq <caller_rbx=int64#14,>rbx_stack=stack64#6
# asm 2: movq <caller_rbx=%rbx,>rbx_stack=648(%rsp)
movq %rbx,648(%rsp)

# qhasm: ptr = &buf
# asm 1: leaq <buf=stack4864#1,>ptr=int64#4
# asm 2: leaq <buf=0(%rsp),>ptr=%rcx
leaq 0(%rsp),%rcx

# qhasm: s0 = mem256[ input_1 +  0 ]
# asm 1: vmovupd   0(<input_1=int64#2),>s0=reg256#1
# asm 2: vmovupd   0(<input_1=%rsi),>s0=%ymm0
vmovupd   0(%rsi),%ymm0

# qhasm: s1 = mem256[ input_1 + 32 ]
# asm 1: vmovupd   32(<input_1=int64#2),>s1=reg256#2
# asm 2: vmovupd   32(<input_1=%rsi),>s1=%ymm1
vmovupd   32(%rsi),%ymm1

# qhasm: s2 = mem256[ input_1 + 64 ]
# asm 1: vmovupd   64(<input_1=int64#2),>s2=reg256#3
# asm 2: vmovupd   64(<input_1=%rsi),>s2=%ymm2
vmovupd   64(%rsi),%ymm2

# qhasm: t0 = mem256[ input_2 +  0 ]
# asm 1: vmovupd   0(<input_2=int64#3),>t0=reg256#4
# asm 2: vmovupd   0(<input_2=%rdx),>t0=%ymm3
vmovupd   0(%rdx),%ymm3

# qhasm: t1 = mem256[ input_2 + 32 ]
# asm 1: vmovupd   32(<input_2=int64#3),>t1=reg256#5
# asm 2: vmovupd   32(<input_2=%rdx),>t1=%ymm4
vmovupd   32(%rdx),%ymm4

# qhasm: t2 = mem256[ input_2 + 64 ]
# asm 1: vmovupd   64(<input_2=int64#3),>t2=reg256#6
# asm 2: vmovupd   64(<input_2=%rdx),>t2=%ymm5
vmovupd   64(%rdx),%ymm5

# qhasm: a5[0,1,2,3] = s2[2,2,3,3]
# asm 1: vpermq $0xfa,<s2=reg256#3,>a5=reg256#7
# asm 2: vpermq $0xfa,<s2=%ymm2,>a5=%ymm6
vpermq $0xfa,%ymm2,%ymm6

# qhasm: b5[0,1,2,3] = t2[2,3,2,3]
# asm 1: vpermq $0xee,<t2=reg256#6,>b5=reg256#8
# asm 2: vpermq $0xee,<t2=%ymm5,>b5=%ymm7
vpermq $0xee,%ymm5,%ymm7

# qhasm: r10 = a5 & b5
# asm 1: vpand <a5=reg256#7,<b5=reg256#8,>r10=reg256#9
# asm 2: vpand <a5=%ymm6,<b5=%ymm7,>r10=%ymm8
vpand %ymm6,%ymm7,%ymm8

# qhasm: mem256[ ptr + 320 ] = r10
# asm 1: vmovupd   <r10=reg256#9,320(<ptr=int64#4)
# asm 2: vmovupd   <r10=%ymm8,320(<ptr=%rcx)
vmovupd   %ymm8,320(%rcx)

# qhasm: b4[0,1,2,3] = t2[0,1,0,1]
# asm 1: vpermq $0x44,<t2=reg256#6,>b4=reg256#6
# asm 2: vpermq $0x44,<t2=%ymm5,>b4=%ymm5
vpermq $0x44,%ymm5,%ymm5

# qhasm: r9 = a5 & b4
# asm 1: vpand <a5=reg256#7,<b4=reg256#6,>r9=reg256#9
# asm 2: vpand <a5=%ymm6,<b4=%ymm5,>r9=%ymm8
vpand %ymm6,%ymm5,%ymm8

# qhasm: b3[0,1,2,3] = t1[2,3,2,3]
# asm 1: vpermq $0xee,<t1=reg256#5,>b3=reg256#10
# asm 2: vpermq $0xee,<t1=%ymm4,>b3=%ymm9
vpermq $0xee,%ymm4,%ymm9

# qhasm: r8 = a5 & b3
# asm 1: vpand <a5=reg256#7,<b3=reg256#10,>r8=reg256#11
# asm 2: vpand <a5=%ymm6,<b3=%ymm9,>r8=%ymm10
vpand %ymm6,%ymm9,%ymm10

# qhasm: b2[0,1,2,3] = t1[0,1,0,1]
# asm 1: vpermq $0x44,<t1=reg256#5,>b2=reg256#5
# asm 2: vpermq $0x44,<t1=%ymm4,>b2=%ymm4
vpermq $0x44,%ymm4,%ymm4

# qhasm: r7 = a5 & b2
# asm 1: vpand <a5=reg256#7,<b2=reg256#5,>r7=reg256#12
# asm 2: vpand <a5=%ymm6,<b2=%ymm4,>r7=%ymm11
vpand %ymm6,%ymm4,%ymm11

# qhasm: b1[0,1,2,3] = t0[2,3,2,3]
# asm 1: vpermq $0xee,<t0=reg256#4,>b1=reg256#13
# asm 2: vpermq $0xee,<t0=%ymm3,>b1=%ymm12
vpermq $0xee,%ymm3,%ymm12

# qhasm: r6 = a5 & b1
# asm 1: vpand <a5=reg256#7,<b1=reg256#13,>r6=reg256#14
# asm 2: vpand <a5=%ymm6,<b1=%ymm12,>r6=%ymm13
vpand %ymm6,%ymm12,%ymm13

# qhasm: b0[0,1,2,3] = t0[0,1,0,1]
# asm 1: vpermq $0x44,<t0=reg256#4,>b0=reg256#4
# asm 2: vpermq $0x44,<t0=%ymm3,>b0=%ymm3
vpermq $0x44,%ymm3,%ymm3

# qhasm: r5 = a5 & b0
# asm 1: vpand <a5=reg256#7,<b0=reg256#4,>r5=reg256#7
# asm 2: vpand <a5=%ymm6,<b0=%ymm3,>r5=%ymm6
vpand %ymm6,%ymm3,%ymm6

# qhasm: a4[0,1,2,3] = s2[0,0,1,1]
# asm 1: vpermq $0x50,<s2=reg256#3,>a4=reg256#3
# asm 2: vpermq $0x50,<s2=%ymm2,>a4=%ymm2
vpermq $0x50,%ymm2,%ymm2

# qhasm: r = a4 & b5
# asm 1: vpand <a4=reg256#3,<b5=reg256#8,>r=reg256#15
# asm 2: vpand <a4=%ymm2,<b5=%ymm7,>r=%ymm14
vpand %ymm2,%ymm7,%ymm14

# qhasm: r9 ^= r
# asm 1: vpxor <r=reg256#15,<r9=reg256#9,<r9=reg256#9
# asm 2: vpxor <r=%ymm14,<r9=%ymm8,<r9=%ymm8
vpxor %ymm14,%ymm8,%ymm8

# qhasm: mem256[ ptr + 288 ] = r9
# asm 1: vmovupd   <r9=reg256#9,288(<ptr=int64#4)
# asm 2: vmovupd   <r9=%ymm8,288(<ptr=%rcx)
vmovupd   %ymm8,288(%rcx)

# qhasm: r = a4 & b4
# asm 1: vpand <a4=reg256#3,<b4=reg256#6,>r=reg256#9
# asm 2: vpand <a4=%ymm2,<b4=%ymm5,>r=%ymm8
vpand %ymm2,%ymm5,%ymm8

# qhasm: r8 ^= r
# asm 1: vpxor <r=reg256#9,<r8=reg256#11,<r8=reg256#11
# asm 2: vpxor <r=%ymm8,<r8=%ymm10,<r8=%ymm10
vpxor %ymm8,%ymm10,%ymm10

# qhasm: r = a4 & b3
# asm 1: vpand <a4=reg256#3,<b3=reg256#10,>r=reg256#9
# asm 2: vpand <a4=%ymm2,<b3=%ymm9,>r=%ymm8
vpand %ymm2,%ymm9,%ymm8

# qhasm: r7 ^= r
# asm 1: vpxor <r=reg256#9,<r7=reg256#12,<r7=reg256#12
# asm 2: vpxor <r=%ymm8,<r7=%ymm11,<r7=%ymm11
vpxor %ymm8,%ymm11,%ymm11

# qhasm: r = a4 & b2
# asm 1: vpand <a4=reg256#3,<b2=reg256#5,>r=reg256#9
# asm 2: vpand <a4=%ymm2,<b2=%ymm4,>r=%ymm8
vpand %ymm2,%ymm4,%ymm8

# qhasm: r6 ^= r
# asm 1: vpxor <r=reg256#9,<r6=reg256#14,<r6=reg256#14
# asm 2: vpxor <r=%ymm8,<r6=%ymm13,<r6=%ymm13
vpxor %ymm8,%ymm13,%ymm13

# qhasm: r = a4 & b1
# asm 1: vpand <a4=reg256#3,<b1=reg256#13,>r=reg256#9
# asm 2: vpand <a4=%ymm2,<b1=%ymm12,>r=%ymm8
vpand %ymm2,%ymm12,%ymm8

# qhasm: r5 ^= r
# asm 1: vpxor <r=reg256#9,<r5=reg256#7,<r5=reg256#7
# asm 2: vpxor <r=%ymm8,<r5=%ymm6,<r5=%ymm6
vpxor %ymm8,%ymm6,%ymm6

# qhasm: r4 = a4 & b0
# asm 1: vpand <a4=reg256#3,<b0=reg256#4,>r4=reg256#3
# asm 2: vpand <a4=%ymm2,<b0=%ymm3,>r4=%ymm2
vpand %ymm2,%ymm3,%ymm2

# qhasm: a3[0,1,2,3] = s1[2,2,3,3]
# asm 1: vpermq $0xfa,<s1=reg256#2,>a3=reg256#9
# asm 2: vpermq $0xfa,<s1=%ymm1,>a3=%ymm8
vpermq $0xfa,%ymm1,%ymm8

# qhasm: r = a3 & b5
# asm 1: vpand <a3=reg256#9,<b5=reg256#8,>r=reg256#15
# asm 2: vpand <a3=%ymm8,<b5=%ymm7,>r=%ymm14
vpand %ymm8,%ymm7,%ymm14

# qhasm: r8 ^= r
# asm 1: vpxor <r=reg256#15,<r8=reg256#11,<r8=reg256#11
# asm 2: vpxor <r=%ymm14,<r8=%ymm10,<r8=%ymm10
vpxor %ymm14,%ymm10,%ymm10

# qhasm: mem256[ ptr + 256 ] = r8
# asm 1: vmovupd   <r8=reg256#11,256(<ptr=int64#4)
# asm 2: vmovupd   <r8=%ymm10,256(<ptr=%rcx)
vmovupd   %ymm10,256(%rcx)

# qhasm: r = a3 & b4
# asm 1: vpand <a3=reg256#9,<b4=reg256#6,>r=reg256#11
# asm 2: vpand <a3=%ymm8,<b4=%ymm5,>r=%ymm10
vpand %ymm8,%ymm5,%ymm10

# qhasm: r7 ^= r
# asm 1: vpxor <r=reg256#11,<r7=reg256#12,<r7=reg256#12
# asm 2: vpxor <r=%ymm10,<r7=%ymm11,<r7=%ymm11
vpxor %ymm10,%ymm11,%ymm11

# qhasm: r = a3 & b3
# asm 1: vpand <a3=reg256#9,<b3=reg256#10,>r=reg256#11
# asm 2: vpand <a3=%ymm8,<b3=%ymm9,>r=%ymm10
vpand %ymm8,%ymm9,%ymm10

# qhasm: r6 ^= r
# asm 1: vpxor <r=reg256#11,<r6=reg256#14,<r6=reg256#14
# asm 2: vpxor <r=%ymm10,<r6=%ymm13,<r6=%ymm13
vpxor %ymm10,%ymm13,%ymm13

# qhasm: r = a3 & b2
# asm 1: vpand <a3=reg256#9,<b2=reg256#5,>r=reg256#11
# asm 2: vpand <a3=%ymm8,<b2=%ymm4,>r=%ymm10
vpand %ymm8,%ymm4,%ymm10

# qhasm: r5 ^= r
# asm 1: vpxor <r=reg256#11,<r5=reg256#7,<r5=reg256#7
# asm 2: vpxor <r=%ymm10,<r5=%ymm6,<r5=%ymm6
vpxor %ymm10,%ymm6,%ymm6

# qhasm: r = a3 & b1
# asm 1: vpand <a3=reg256#9,<b1=reg256#13,>r=reg256#11
# asm 2: vpand <a3=%ymm8,<b1=%ymm12,>r=%ymm10
vpand %ymm8,%ymm12,%ymm10

# qhasm: r4 ^= r
# asm 1: vpxor <r=reg256#11,<r4=reg256#3,<r4=reg256#3
# asm 2: vpxor <r=%ymm10,<r4=%ymm2,<r4=%ymm2
vpxor %ymm10,%ymm2,%ymm2

# qhasm: r3 = a3 & b0
# asm 1: vpand <a3=reg256#9,<b0=reg256#4,>r3=reg256#9
# asm 2: vpand <a3=%ymm8,<b0=%ymm3,>r3=%ymm8
vpand %ymm8,%ymm3,%ymm8

# qhasm: a2[0,1,2,3] = s1[0,0,1,1]
# asm 1: vpermq $0x50,<s1=reg256#2,>a2=reg256#2
# asm 2: vpermq $0x50,<s1=%ymm1,>a2=%ymm1
vpermq $0x50,%ymm1,%ymm1

# qhasm: r = a2 & b5
# asm 1: vpand <a2=reg256#2,<b5=reg256#8,>r=reg256#11
# asm 2: vpand <a2=%ymm1,<b5=%ymm7,>r=%ymm10
vpand %ymm1,%ymm7,%ymm10

# qhasm: r7 ^= r
# asm 1: vpxor <r=reg256#11,<r7=reg256#12,<r7=reg256#12
# asm 2: vpxor <r=%ymm10,<r7=%ymm11,<r7=%ymm11
vpxor %ymm10,%ymm11,%ymm11

# qhasm: mem256[ ptr + 224 ] = r7
# asm 1: vmovupd   <r7=reg256#12,224(<ptr=int64#4)
# asm 2: vmovupd   <r7=%ymm11,224(<ptr=%rcx)
vmovupd   %ymm11,224(%rcx)

# qhasm: r = a2 & b4
# asm 1: vpand <a2=reg256#2,<b4=reg256#6,>r=reg256#11
# asm 2: vpand <a2=%ymm1,<b4=%ymm5,>r=%ymm10
vpand %ymm1,%ymm5,%ymm10

# qhasm: r6 ^= r
# asm 1: vpxor <r=reg256#11,<r6=reg256#14,<r6=reg256#14
# asm 2: vpxor <r=%ymm10,<r6=%ymm13,<r6=%ymm13
vpxor %ymm10,%ymm13,%ymm13

# qhasm: r = a2 & b3
# asm 1: vpand <a2=reg256#2,<b3=reg256#10,>r=reg256#11
# asm 2: vpand <a2=%ymm1,<b3=%ymm9,>r=%ymm10
vpand %ymm1,%ymm9,%ymm10

# qhasm: r5 ^= r
# asm 1: vpxor <r=reg256#11,<r5=reg256#7,<r5=reg256#7
# asm 2: vpxor <r=%ymm10,<r5=%ymm6,<r5=%ymm6
vpxor %ymm10,%ymm6,%ymm6

# qhasm: r = a2 & b2
# asm 1: vpand <a2=reg256#2,<b2=reg256#5,>r=reg256#11
# asm 2: vpand <a2=%ymm1,<b2=%ymm4,>r=%ymm10
vpand %ymm1,%ymm4,%ymm10

# qhasm: r4 ^= r
# asm 1: vpxor <r=reg256#11,<r4=reg256#3,<r4=reg256#3
# asm 2: vpxor <r=%ymm10,<r4=%ymm2,<r4=%ymm2
vpxor %ymm10,%ymm2,%ymm2

# qhasm: r = a2 & b1
# asm 1: vpand <a2=reg256#2,<b1=reg256#13,>r=reg256#11
# asm 2: vpand <a2=%ymm1,<b1=%ymm12,>r=%ymm10
vpand %ymm1,%ymm12,%ymm10

# qhasm: r3 ^= r
# asm 1: vpxor <r=reg256#11,<r3=reg256#9,<r3=reg256#9
# asm 2: vpxor <r=%ymm10,<r3=%ymm8,<r3=%ymm8
vpxor %ymm10,%ymm8,%ymm8

# qhasm: r2 = a2 & b0
# asm 1: vpand <a2=reg256#2,<b0=reg256#4,>r2=reg256#2
# asm 2: vpand <a2=%ymm1,<b0=%ymm3,>r2=%ymm1
vpand %ymm1,%ymm3,%ymm1

# qhasm: a1[0,1,2,3] = s0[2,2,3,3]
# asm 1: vpermq $0xfa,<s0=reg256#1,>a1=reg256#11
# asm 2: vpermq $0xfa,<s0=%ymm0,>a1=%ymm10
vpermq $0xfa,%ymm0,%ymm10

# qhasm: r = a1 & b5
# asm 1: vpand <a1=reg256#11,<b5=reg256#8,>r=reg256#12
# asm 2: vpand <a1=%ymm10,<b5=%ymm7,>r=%ymm11
vpand %ymm10,%ymm7,%ymm11

# qhasm: r6 ^= r
# asm 1: vpxor <r=reg256#12,<r6=reg256#14,<r6=reg256#14
# asm 2: vpxor <r=%ymm11,<r6=%ymm13,<r6=%ymm13
vpxor %ymm11,%ymm13,%ymm13

# qhasm: mem256[ ptr + 192 ] = r6
# asm 1: vmovupd   <r6=reg256#14,192(<ptr=int64#4)
# asm 2: vmovupd   <r6=%ymm13,192(<ptr=%rcx)
vmovupd   %ymm13,192(%rcx)

# qhasm: r = a1 & b4
# asm 1: vpand <a1=reg256#11,<b4=reg256#6,>r=reg256#12
# asm 2: vpand <a1=%ymm10,<b4=%ymm5,>r=%ymm11
vpand %ymm10,%ymm5,%ymm11

# qhasm: r5 ^= r
# asm 1: vpxor <r=reg256#12,<r5=reg256#7,<r5=reg256#7
# asm 2: vpxor <r=%ymm11,<r5=%ymm6,<r5=%ymm6
vpxor %ymm11,%ymm6,%ymm6

# qhasm: r = a1 & b3
# asm 1: vpand <a1=reg256#11,<b3=reg256#10,>r=reg256#12
# asm 2: vpand <a1=%ymm10,<b3=%ymm9,>r=%ymm11
vpand %ymm10,%ymm9,%ymm11

# qhasm: r4 ^= r
# asm 1: vpxor <r=reg256#12,<r4=reg256#3,<r4=reg256#3
# asm 2: vpxor <r=%ymm11,<r4=%ymm2,<r4=%ymm2
vpxor %ymm11,%ymm2,%ymm2

# qhasm: r = a1 & b2
# asm 1: vpand <a1=reg256#11,<b2=reg256#5,>r=reg256#12
# asm 2: vpand <a1=%ymm10,<b2=%ymm4,>r=%ymm11
vpand %ymm10,%ymm4,%ymm11

# qhasm: r3 ^= r
# asm 1: vpxor <r=reg256#12,<r3=reg256#9,<r3=reg256#9
# asm 2: vpxor <r=%ymm11,<r3=%ymm8,<r3=%ymm8
vpxor %ymm11,%ymm8,%ymm8

# qhasm: r = a1 & b1
# asm 1: vpand <a1=reg256#11,<b1=reg256#13,>r=reg256#12
# asm 2: vpand <a1=%ymm10,<b1=%ymm12,>r=%ymm11
vpand %ymm10,%ymm12,%ymm11

# qhasm: r2 ^= r
# asm 1: vpxor <r=reg256#12,<r2=reg256#2,<r2=reg256#2
# asm 2: vpxor <r=%ymm11,<r2=%ymm1,<r2=%ymm1
vpxor %ymm11,%ymm1,%ymm1

# qhasm: r1 = a1 & b0
# asm 1: vpand <a1=reg256#11,<b0=reg256#4,>r1=reg256#11
# asm 2: vpand <a1=%ymm10,<b0=%ymm3,>r1=%ymm10
vpand %ymm10,%ymm3,%ymm10

# qhasm: a0[0,1,2,3] = s0[0,0,1,1]
# asm 1: vpermq $0x50,<s0=reg256#1,>a0=reg256#1
# asm 2: vpermq $0x50,<s0=%ymm0,>a0=%ymm0
vpermq $0x50,%ymm0,%ymm0

# qhasm: r = a0 & b5
# asm 1: vpand <a0=reg256#1,<b5=reg256#8,>r=reg256#8
# asm 2: vpand <a0=%ymm0,<b5=%ymm7,>r=%ymm7
vpand %ymm0,%ymm7,%ymm7

# qhasm: r5 ^= r
# asm 1: vpxor <r=reg256#8,<r5=reg256#7,<r5=reg256#7
# asm 2: vpxor <r=%ymm7,<r5=%ymm6,<r5=%ymm6
vpxor %ymm7,%ymm6,%ymm6

# qhasm: mem256[ ptr + 160 ] = r5
# asm 1: vmovupd   <r5=reg256#7,160(<ptr=int64#4)
# asm 2: vmovupd   <r5=%ymm6,160(<ptr=%rcx)
vmovupd   %ymm6,160(%rcx)

# qhasm: r = a0 & b4
# asm 1: vpand <a0=reg256#1,<b4=reg256#6,>r=reg256#6
# asm 2: vpand <a0=%ymm0,<b4=%ymm5,>r=%ymm5
vpand %ymm0,%ymm5,%ymm5

# qhasm: r4 ^= r
# asm 1: vpxor <r=reg256#6,<r4=reg256#3,<r4=reg256#3
# asm 2: vpxor <r=%ymm5,<r4=%ymm2,<r4=%ymm2
vpxor %ymm5,%ymm2,%ymm2

# qhasm: r = a0 & b3
# asm 1: vpand <a0=reg256#1,<b3=reg256#10,>r=reg256#6
# asm 2: vpand <a0=%ymm0,<b3=%ymm9,>r=%ymm5
vpand %ymm0,%ymm9,%ymm5

# qhasm: r3 ^= r
# asm 1: vpxor <r=reg256#6,<r3=reg256#9,<r3=reg256#9
# asm 2: vpxor <r=%ymm5,<r3=%ymm8,<r3=%ymm8
vpxor %ymm5,%ymm8,%ymm8

# qhasm: r = a0 & b2
# asm 1: vpand <a0=reg256#1,<b2=reg256#5,>r=reg256#5
# asm 2: vpand <a0=%ymm0,<b2=%ymm4,>r=%ymm4
vpand %ymm0,%ymm4,%ymm4

# qhasm: r2 ^= r
# asm 1: vpxor <r=reg256#5,<r2=reg256#2,<r2=reg256#2
# asm 2: vpxor <r=%ymm4,<r2=%ymm1,<r2=%ymm1
vpxor %ymm4,%ymm1,%ymm1

# qhasm: r = a0 & b1
# asm 1: vpand <a0=reg256#1,<b1=reg256#13,>r=reg256#5
# asm 2: vpand <a0=%ymm0,<b1=%ymm12,>r=%ymm4
vpand %ymm0,%ymm12,%ymm4

# qhasm: r1 ^= r
# asm 1: vpxor <r=reg256#5,<r1=reg256#11,<r1=reg256#11
# asm 2: vpxor <r=%ymm4,<r1=%ymm10,<r1=%ymm10
vpxor %ymm4,%ymm10,%ymm10

# qhasm: r0 = a0 & b0
# asm 1: vpand <a0=reg256#1,<b0=reg256#4,>r0=reg256#1
# asm 2: vpand <a0=%ymm0,<b0=%ymm3,>r0=%ymm0
vpand %ymm0,%ymm3,%ymm0

# qhasm: mem256[ ptr + 128 ] = r4
# asm 1: vmovupd   <r4=reg256#3,128(<ptr=int64#4)
# asm 2: vmovupd   <r4=%ymm2,128(<ptr=%rcx)
vmovupd   %ymm2,128(%rcx)

# qhasm: mem256[ ptr + 96 ] = r3
# asm 1: vmovupd   <r3=reg256#9,96(<ptr=int64#4)
# asm 2: vmovupd   <r3=%ymm8,96(<ptr=%rcx)
vmovupd   %ymm8,96(%rcx)

# qhasm: mem256[ ptr + 64 ] = r2
# asm 1: vmovupd   <r2=reg256#2,64(<ptr=int64#4)
# asm 2: vmovupd   <r2=%ymm1,64(<ptr=%rcx)
vmovupd   %ymm1,64(%rcx)

# qhasm: mem256[ ptr + 32 ] = r1
# asm 1: vmovupd   <r1=reg256#11,32(<ptr=int64#4)
# asm 2: vmovupd   <r1=%ymm10,32(<ptr=%rcx)
vmovupd   %ymm10,32(%rcx)

# qhasm: mem256[ ptr + 0 ] = r0
# asm 1: vmovupd   <r0=reg256#1,0(<ptr=int64#4)
# asm 2: vmovupd   <r0=%ymm0,0(<ptr=%rcx)
vmovupd   %ymm0,0(%rcx)

# qhasm: h22 = mem64[ ptr + 344 ]
# asm 1: movq   344(<ptr=int64#4),>h22=int64#2
# asm 2: movq   344(<ptr=%rcx),>h22=%rsi
movq   344(%rcx),%rsi

# qhasm: h13 = h22
# asm 1: mov  <h22=int64#2,>h13=int64#3
# asm 2: mov  <h22=%rsi,>h13=%rdx
mov  %rsi,%rdx

# qhasm: h10 = h22
# asm 1: mov  <h22=int64#2,>h10=int64#2
# asm 2: mov  <h22=%rsi,>h10=%rsi
mov  %rsi,%rsi

# qhasm: h21 = mem64[ ptr + 336 ]
# asm 1: movq   336(<ptr=int64#4),>h21=int64#5
# asm 2: movq   336(<ptr=%rcx),>h21=%r8
movq   336(%rcx),%r8

# qhasm: h21 ^= *(uint64 *) ( ptr + 328 )
# asm 1: xorq 328(<ptr=int64#4),<h21=int64#5
# asm 2: xorq 328(<ptr=%rcx),<h21=%r8
xorq 328(%rcx),%r8

# qhasm: h12 = h21
# asm 1: mov  <h21=int64#5,>h12=int64#6
# asm 2: mov  <h21=%r8,>h12=%r9
mov  %r8,%r9

# qhasm: h9 = h21
# asm 1: mov  <h21=int64#5,>h9=int64#5
# asm 2: mov  <h21=%r8,>h9=%r8
mov  %r8,%r8

# qhasm: h20 = mem64[ ptr + 312 ]
# asm 1: movq   312(<ptr=int64#4),>h20=int64#7
# asm 2: movq   312(<ptr=%rcx),>h20=%rax
movq   312(%rcx),%rax

# qhasm: h20 ^= *(uint64 *) ( ptr + 320 )
# asm 1: xorq 320(<ptr=int64#4),<h20=int64#7
# asm 2: xorq 320(<ptr=%rcx),<h20=%rax
xorq 320(%rcx),%rax

# qhasm: h11 = h20
# asm 1: mov  <h20=int64#7,>h11=int64#8
# asm 2: mov  <h20=%rax,>h11=%r10
mov  %rax,%r10

# qhasm: h8 = h20
# asm 1: mov  <h20=int64#7,>h8=int64#7
# asm 2: mov  <h20=%rax,>h8=%rax
mov  %rax,%rax

# qhasm: h19 = mem64[ ptr + 304 ]
# asm 1: movq   304(<ptr=int64#4),>h19=int64#9
# asm 2: movq   304(<ptr=%rcx),>h19=%r11
movq   304(%rcx),%r11

# qhasm: h19 ^= *(uint64 *) ( ptr + 296 )
# asm 1: xorq 296(<ptr=int64#4),<h19=int64#9
# asm 2: xorq 296(<ptr=%rcx),<h19=%r11
xorq 296(%rcx),%r11

# qhasm: h10 ^= h19
# asm 1: xor  <h19=int64#9,<h10=int64#2
# asm 2: xor  <h19=%r11,<h10=%rsi
xor  %r11,%rsi

# qhasm: h7 = h19
# asm 1: mov  <h19=int64#9,>h7=int64#9
# asm 2: mov  <h19=%r11,>h7=%r11
mov  %r11,%r11

# qhasm: h18 = mem64[ ptr + 280 ]
# asm 1: movq   280(<ptr=int64#4),>h18=int64#10
# asm 2: movq   280(<ptr=%rcx),>h18=%r12
movq   280(%rcx),%r12

# qhasm: h18 ^= *(uint64 *) ( ptr + 288 )
# asm 1: xorq 288(<ptr=int64#4),<h18=int64#10
# asm 2: xorq 288(<ptr=%rcx),<h18=%r12
xorq 288(%rcx),%r12

# qhasm: h9 ^= h18
# asm 1: xor  <h18=int64#10,<h9=int64#5
# asm 2: xor  <h18=%r12,<h9=%r8
xor  %r12,%r8

# qhasm: h6 = h18
# asm 1: mov  <h18=int64#10,>h6=int64#10
# asm 2: mov  <h18=%r12,>h6=%r12
mov  %r12,%r12

# qhasm: h17 = mem64[ ptr + 272 ]
# asm 1: movq   272(<ptr=int64#4),>h17=int64#11
# asm 2: movq   272(<ptr=%rcx),>h17=%r13
movq   272(%rcx),%r13

# qhasm: h17 ^= *(uint64 *) ( ptr + 264 )
# asm 1: xorq 264(<ptr=int64#4),<h17=int64#11
# asm 2: xorq 264(<ptr=%rcx),<h17=%r13
xorq 264(%rcx),%r13

# qhasm: h8 ^= h17
# asm 1: xor  <h17=int64#11,<h8=int64#7
# asm 2: xor  <h17=%r13,<h8=%rax
xor  %r13,%rax

# qhasm: h5 = h17
# asm 1: mov  <h17=int64#11,>h5=int64#11
# asm 2: mov  <h17=%r13,>h5=%r13
mov  %r13,%r13

# qhasm: h16 = mem64[ ptr + 248 ]
# asm 1: movq   248(<ptr=int64#4),>h16=int64#12
# asm 2: movq   248(<ptr=%rcx),>h16=%r14
movq   248(%rcx),%r14

# qhasm: h16 ^= *(uint64 *) ( ptr + 256 )
# asm 1: xorq 256(<ptr=int64#4),<h16=int64#12
# asm 2: xorq 256(<ptr=%rcx),<h16=%r14
xorq 256(%rcx),%r14

# qhasm: h7 ^= h16
# asm 1: xor  <h16=int64#12,<h7=int64#9
# asm 2: xor  <h16=%r14,<h7=%r11
xor  %r14,%r11

# qhasm: h4 = h16
# asm 1: mov  <h16=int64#12,>h4=int64#12
# asm 2: mov  <h16=%r14,>h4=%r14
mov  %r14,%r14

# qhasm: h15 = mem64[ ptr + 240 ]
# asm 1: movq   240(<ptr=int64#4),>h15=int64#13
# asm 2: movq   240(<ptr=%rcx),>h15=%r15
movq   240(%rcx),%r15

# qhasm: h15 ^= *(uint64 *) ( ptr + 232 )
# asm 1: xorq 232(<ptr=int64#4),<h15=int64#13
# asm 2: xorq 232(<ptr=%rcx),<h15=%r15
xorq 232(%rcx),%r15

# qhasm: h6 ^= h15
# asm 1: xor  <h15=int64#13,<h6=int64#10
# asm 2: xor  <h15=%r15,<h6=%r12
xor  %r15,%r12

# qhasm: h3 = h15
# asm 1: mov  <h15=int64#13,>h3=int64#13
# asm 2: mov  <h15=%r15,>h3=%r15
mov  %r15,%r15

# qhasm: h14 = mem64[ ptr + 216 ]
# asm 1: movq   216(<ptr=int64#4),>h14=int64#14
# asm 2: movq   216(<ptr=%rcx),>h14=%rbx
movq   216(%rcx),%rbx

# qhasm: h14 ^= *(uint64 *) ( ptr + 224 )
# asm 1: xorq 224(<ptr=int64#4),<h14=int64#14
# asm 2: xorq 224(<ptr=%rcx),<h14=%rbx
xorq 224(%rcx),%rbx

# qhasm: h5 ^= h14
# asm 1: xor  <h14=int64#14,<h5=int64#11
# asm 2: xor  <h14=%rbx,<h5=%r13
xor  %rbx,%r13

# qhasm: h2 = h14
# asm 1: mov  <h14=int64#14,>h2=int64#14
# asm 2: mov  <h14=%rbx,>h2=%rbx
mov  %rbx,%rbx

# qhasm: h13 ^= *(uint64 *) ( ptr + 208 )
# asm 1: xorq 208(<ptr=int64#4),<h13=int64#3
# asm 2: xorq 208(<ptr=%rcx),<h13=%rdx
xorq 208(%rcx),%rdx

# qhasm: h13 ^= *(uint64 *) ( ptr + 200 )
# asm 1: xorq 200(<ptr=int64#4),<h13=int64#3
# asm 2: xorq 200(<ptr=%rcx),<h13=%rdx
xorq 200(%rcx),%rdx

# qhasm: h4 ^= h13
# asm 1: xor  <h13=int64#3,<h4=int64#12
# asm 2: xor  <h13=%rdx,<h4=%r14
xor  %rdx,%r14

# qhasm: h1 = h13
# asm 1: mov  <h13=int64#3,>h1=int64#3
# asm 2: mov  <h13=%rdx,>h1=%rdx
mov  %rdx,%rdx

# qhasm: h12 ^= *(uint64 *) ( ptr + 184 )
# asm 1: xorq 184(<ptr=int64#4),<h12=int64#6
# asm 2: xorq 184(<ptr=%rcx),<h12=%r9
xorq 184(%rcx),%r9

# qhasm: h12 ^= *(uint64 *) ( ptr + 192 )
# asm 1: xorq 192(<ptr=int64#4),<h12=int64#6
# asm 2: xorq 192(<ptr=%rcx),<h12=%r9
xorq 192(%rcx),%r9

# qhasm: h3 ^= h12
# asm 1: xor  <h12=int64#6,<h3=int64#13
# asm 2: xor  <h12=%r9,<h3=%r15
xor  %r9,%r15

# qhasm: h0 = h12
# asm 1: mov  <h12=int64#6,>h0=int64#6
# asm 2: mov  <h12=%r9,>h0=%r9
mov  %r9,%r9

# qhasm: h11 ^= *(uint64 *) ( ptr + 176 )
# asm 1: xorq 176(<ptr=int64#4),<h11=int64#8
# asm 2: xorq 176(<ptr=%rcx),<h11=%r10
xorq 176(%rcx),%r10

# qhasm: h11 ^= *(uint64 *) ( ptr + 168 )
# asm 1: xorq 168(<ptr=int64#4),<h11=int64#8
# asm 2: xorq 168(<ptr=%rcx),<h11=%r10
xorq 168(%rcx),%r10

# qhasm: mem64[ input_0 + 88 ] = h11
# asm 1: movq   <h11=int64#8,88(<input_0=int64#1)
# asm 2: movq   <h11=%r10,88(<input_0=%rdi)
movq   %r10,88(%rdi)

# qhasm: h10 ^= *(uint64 *) ( ptr + 152 )
# asm 1: xorq 152(<ptr=int64#4),<h10=int64#2
# asm 2: xorq 152(<ptr=%rcx),<h10=%rsi
xorq 152(%rcx),%rsi

# qhasm: h10 ^= *(uint64 *) ( ptr + 160 )
# asm 1: xorq 160(<ptr=int64#4),<h10=int64#2
# asm 2: xorq 160(<ptr=%rcx),<h10=%rsi
xorq 160(%rcx),%rsi

# qhasm: mem64[ input_0 + 80 ] = h10
# asm 1: movq   <h10=int64#2,80(<input_0=int64#1)
# asm 2: movq   <h10=%rsi,80(<input_0=%rdi)
movq   %rsi,80(%rdi)

# qhasm: h9 ^= *(uint64 *) ( ptr + 144 )
# asm 1: xorq 144(<ptr=int64#4),<h9=int64#5
# asm 2: xorq 144(<ptr=%rcx),<h9=%r8
xorq 144(%rcx),%r8

# qhasm: h9 ^= *(uint64 *) ( ptr + 136 )
# asm 1: xorq 136(<ptr=int64#4),<h9=int64#5
# asm 2: xorq 136(<ptr=%rcx),<h9=%r8
xorq 136(%rcx),%r8

# qhasm: mem64[ input_0 + 72 ] = h9
# asm 1: movq   <h9=int64#5,72(<input_0=int64#1)
# asm 2: movq   <h9=%r8,72(<input_0=%rdi)
movq   %r8,72(%rdi)

# qhasm: h8 ^= *(uint64 *) ( ptr + 120 )
# asm 1: xorq 120(<ptr=int64#4),<h8=int64#7
# asm 2: xorq 120(<ptr=%rcx),<h8=%rax
xorq 120(%rcx),%rax

# qhasm: h8 ^= *(uint64 *) ( ptr + 128 )
# asm 1: xorq 128(<ptr=int64#4),<h8=int64#7
# asm 2: xorq 128(<ptr=%rcx),<h8=%rax
xorq 128(%rcx),%rax

# qhasm: mem64[ input_0 + 64 ] = h8
# asm 1: movq   <h8=int64#7,64(<input_0=int64#1)
# asm 2: movq   <h8=%rax,64(<input_0=%rdi)
movq   %rax,64(%rdi)

# qhasm: h7 ^= *(uint64 *) ( ptr + 112 )
# asm 1: xorq 112(<ptr=int64#4),<h7=int64#9
# asm 2: xorq 112(<ptr=%rcx),<h7=%r11
xorq 112(%rcx),%r11

# qhasm: h7 ^= *(uint64 *) ( ptr + 104 )
# asm 1: xorq 104(<ptr=int64#4),<h7=int64#9
# asm 2: xorq 104(<ptr=%rcx),<h7=%r11
xorq 104(%rcx),%r11

# qhasm: mem64[ input_0 + 56 ] = h7
# asm 1: movq   <h7=int64#9,56(<input_0=int64#1)
# asm 2: movq   <h7=%r11,56(<input_0=%rdi)
movq   %r11,56(%rdi)

# qhasm: h6 ^= *(uint64 *) ( ptr + 88 )
# asm 1: xorq 88(<ptr=int64#4),<h6=int64#10
# asm 2: xorq 88(<ptr=%rcx),<h6=%r12
xorq 88(%rcx),%r12

# qhasm: h6 ^= *(uint64 *) ( ptr + 96 )
# asm 1: xorq 96(<ptr=int64#4),<h6=int64#10
# asm 2: xorq 96(<ptr=%rcx),<h6=%r12
xorq 96(%rcx),%r12

# qhasm: mem64[ input_0 + 48 ] = h6
# asm 1: movq   <h6=int64#10,48(<input_0=int64#1)
# asm 2: movq   <h6=%r12,48(<input_0=%rdi)
movq   %r12,48(%rdi)

# qhasm: h5 ^= *(uint64 *) ( ptr + 80 )
# asm 1: xorq 80(<ptr=int64#4),<h5=int64#11
# asm 2: xorq 80(<ptr=%rcx),<h5=%r13
xorq 80(%rcx),%r13

# qhasm: h5 ^= *(uint64 *) ( ptr + 72 )
# asm 1: xorq 72(<ptr=int64#4),<h5=int64#11
# asm 2: xorq 72(<ptr=%rcx),<h5=%r13
xorq 72(%rcx),%r13

# qhasm: mem64[ input_0 + 40 ] = h5
# asm 1: movq   <h5=int64#11,40(<input_0=int64#1)
# asm 2: movq   <h5=%r13,40(<input_0=%rdi)
movq   %r13,40(%rdi)

# qhasm: h4 ^= *(uint64 *) ( ptr + 56 )
# asm 1: xorq 56(<ptr=int64#4),<h4=int64#12
# asm 2: xorq 56(<ptr=%rcx),<h4=%r14
xorq 56(%rcx),%r14

# qhasm: h4 ^= *(uint64 *) ( ptr + 64 )
# asm 1: xorq 64(<ptr=int64#4),<h4=int64#12
# asm 2: xorq 64(<ptr=%rcx),<h4=%r14
xorq 64(%rcx),%r14

# qhasm: mem64[ input_0 + 32 ] = h4
# asm 1: movq   <h4=int64#12,32(<input_0=int64#1)
# asm 2: movq   <h4=%r14,32(<input_0=%rdi)
movq   %r14,32(%rdi)

# qhasm: h3 ^= *(uint64 *) ( ptr + 48 )
# asm 1: xorq 48(<ptr=int64#4),<h3=int64#13
# asm 2: xorq 48(<ptr=%rcx),<h3=%r15
xorq 48(%rcx),%r15

# qhasm: h3 ^= *(uint64 *) ( ptr + 40 )
# asm 1: xorq 40(<ptr=int64#4),<h3=int64#13
# asm 2: xorq 40(<ptr=%rcx),<h3=%r15
xorq 40(%rcx),%r15

# qhasm: mem64[ input_0 + 24 ] = h3
# asm 1: movq   <h3=int64#13,24(<input_0=int64#1)
# asm 2: movq   <h3=%r15,24(<input_0=%rdi)
movq   %r15,24(%rdi)

# qhasm: h2 ^= *(uint64 *) ( ptr + 24 )
# asm 1: xorq 24(<ptr=int64#4),<h2=int64#14
# asm 2: xorq 24(<ptr=%rcx),<h2=%rbx
xorq 24(%rcx),%rbx

# qhasm: h2 ^= *(uint64 *) ( ptr + 32 )
# asm 1: xorq 32(<ptr=int64#4),<h2=int64#14
# asm 2: xorq 32(<ptr=%rcx),<h2=%rbx
xorq 32(%rcx),%rbx

# qhasm: mem64[ input_0 + 16 ] = h2
# asm 1: movq   <h2=int64#14,16(<input_0=int64#1)
# asm 2: movq   <h2=%rbx,16(<input_0=%rdi)
movq   %rbx,16(%rdi)

# qhasm: h1 ^= *(uint64 *) ( ptr + 16 )
# asm 1: xorq 16(<ptr=int64#4),<h1=int64#3
# asm 2: xorq 16(<ptr=%rcx),<h1=%rdx
xorq 16(%rcx),%rdx

# qhasm: h1 ^= *(uint64 *) ( ptr + 8 )
# asm 1: xorq 8(<ptr=int64#4),<h1=int64#3
# asm 2: xorq 8(<ptr=%rcx),<h1=%rdx
xorq 8(%rcx),%rdx

# qhasm: mem64[ input_0 + 8 ] = h1
# asm 1: movq   <h1=int64#3,8(<input_0=int64#1)
# asm 2: movq   <h1=%rdx,8(<input_0=%rdi)
movq   %rdx,8(%rdi)

# qhasm: h0 ^= *(uint64 *) ( ptr + 0 )
# asm 1: xorq 0(<ptr=int64#4),<h0=int64#6
# asm 2: xorq 0(<ptr=%rcx),<h0=%r9
xorq 0(%rcx),%r9

# qhasm: mem64[ input_0 + 0 ] = h0
# asm 1: movq   <h0=int64#6,0(<input_0=int64#1)
# asm 2: movq   <h0=%r9,0(<input_0=%rdi)
movq   %r9,0(%rdi)

# qhasm: caller_r11 = r11_stack
# asm 1: movq <r11_stack=stack64#1,>caller_r11=int64#9
# asm 2: movq <r11_stack=608(%rsp),>caller_r11=%r11
movq 608(%rsp),%r11

# qhasm: caller_r12 = r12_stack
# asm 1: movq <r12_stack=stack64#2,>caller_r12=int64#10
# asm 2: movq <r12_stack=616(%rsp),>caller_r12=%r12
movq 616(%rsp),%r12

# qhasm: caller_r13 = r13_stack
# asm 1: movq <r13_stack=stack64#3,>caller_r13=int64#11
# asm 2: movq <r13_stack=624(%rsp),>caller_r13=%r13
movq 624(%rsp),%r13

# qhasm: caller_r14 = r14_stack
# asm 1: movq <r14_stack=stack64#4,>caller_r14=int64#12
# asm 2: movq <r14_stack=632(%rsp),>caller_r14=%r14
movq 632(%rsp),%r14

# qhasm: caller_r15 = r15_stack
# asm 1: movq <r15_stack=stack64#5,>caller_r15=int64#13
# asm 2: movq <r15_stack=640(%rsp),>caller_r15=%r15
movq 640(%rsp),%r15

# qhasm: caller_rbx = rbx_stack
# asm 1: movq <rbx_stack=stack64#6,>caller_rbx=int64#14
# asm 2: movq <rbx_stack=648(%rsp),>caller_rbx=%rbx
movq 648(%rsp),%rbx

# qhasm: return
add %r11,%rsp
ret
.section	.note.GNU-stack,"",@progbits