-rw-r--r-- 30880 lib25519-20240928/crypto_nP/montgomery25519/amd64-avx512-8x1-ns10l-maax/mladder.S raw
#include "crypto_asm_hidden.h"
// linker define mladder
// linker use hh1_p1
// linker use hh1_p2
// linker use hh1_p3
// linker use h2h_p1
// linker use h2h_p2
// linker use h2h_p3
// linker use hh1_xor
// linker use h2h_xor
// linker use dup_mask1
// linker use dup_mask2
// linker use dup_mask3
// linker use swap_c
// linker use swap_mask
// linker use h2h_mask
// linker use vecmask1
// linker use vecmask2
// linker use vec19
// linker use vec608
// linker use vecmask21
// linker use vecmask26
// linker use vecmask32
// linker use a24
/* Assembly for Montgomery ladder. */
#include "consts_namespace.h"
.p2align 5
ASM_HIDDEN _CRYPTO_SHARED_NAMESPACE(mladder)
.globl _CRYPTO_SHARED_NAMESPACE(mladder)
ASM_HIDDEN CRYPTO_SHARED_NAMESPACE(mladder)
.globl CRYPTO_SHARED_NAMESPACE(mladder)
_CRYPTO_SHARED_NAMESPACE(mladder):
CRYPTO_SHARED_NAMESPACE(mladder):
movq %rsp,%r11
andq $-64,%rsp
subq $704,%rsp
movq %r11,0(%rsp)
movq %r12,8(%rsp)
movq %r13,16(%rsp)
movq %r14,24(%rsp)
movq %r15,32(%rsp)
movq %rbx,40(%rsp)
movq %rbp,48(%rsp)
// blending masks
movl $52428,%eax
kmovw %eax,%k1
movl $65280,%eax
kmovw %eax,%k2
movl $255,%eax
kmovw %eax,%k3
// load <0,0,1,X1>
vmovdqa64 0(%rsi),%zmm0
vmovdqa64 64(%rsi),%zmm1
vmovdqa64 128(%rsi),%zmm2
vmovdqa64 192(%rsi),%zmm3
vmovdqa64 256(%rsi),%zmm4
vmovdqa64 320(%rsi),%zmm5
vmovdqa64 384(%rsi),%zmm6
vmovdqa64 448(%rsi),%zmm7
vmovdqa64 512(%rsi),%zmm8
vmovdqa64 576(%rsi),%zmm9
// <0',0',1',X1'> ← Pack-D2N(<0,0,1,X1>)
vpshufd $68,%zmm5,%zmm5
vpshufd $68,%zmm6,%zmm6
vpshufd $68,%zmm7,%zmm7
vpshufd $68,%zmm8,%zmm8
vpshufd $68,%zmm9,%zmm9
vpblendmd %zmm5,%zmm0,%zmm0{%k1}
vpblendmd %zmm6,%zmm1,%zmm1{%k1}
vpblendmd %zmm7,%zmm2,%zmm2{%k1}
vpblendmd %zmm8,%zmm3,%zmm3{%k1}
vpblendmd %zmm9,%zmm4,%zmm4{%k1}
vmovdqa64 %zmm0,0(%rsi)
vmovdqa64 %zmm1,64(%rsi)
vmovdqa64 %zmm2,128(%rsi)
vmovdqa64 %zmm3,192(%rsi)
vmovdqa64 %zmm4,256(%rsi)
// load <X2,Z2,X3,Z3>
vmovdqa64 0(%rdi),%zmm0
vmovdqa64 64(%rdi),%zmm1
vmovdqa64 128(%rdi),%zmm2
vmovdqa64 192(%rdi),%zmm3
vmovdqa64 256(%rdi),%zmm4
vmovdqa64 320(%rdi),%zmm5
vmovdqa64 384(%rdi),%zmm6
vmovdqa64 448(%rdi),%zmm7
vmovdqa64 512(%rdi),%zmm8
vmovdqa64 576(%rdi),%zmm9
// <X2',Z2',X3',Z3'> ← Pack-D2N(<X2,Z2,X3,Z3>)
vpshufd $68,%zmm5,%zmm5
vpshufd $68,%zmm6,%zmm6
vpshufd $68,%zmm7,%zmm7
vpshufd $68,%zmm8,%zmm8
vpshufd $68,%zmm9,%zmm9
vpblendmd %zmm5,%zmm0,%zmm0{%k1}
vpblendmd %zmm6,%zmm1,%zmm1{%k1}
vpblendmd %zmm7,%zmm2,%zmm2{%k1}
vpblendmd %zmm8,%zmm3,%zmm3{%k1}
vpblendmd %zmm9,%zmm4,%zmm4{%k1}
vpsllq $32,%zmm1,%zmm1
vporq %zmm1,%zmm0,%zmm0
vpsllq $32,%zmm3,%zmm3
vporq %zmm3,%zmm2,%zmm2
movq $31,%r15
movq $6,%rcx
movb $0,%r8b
movq %rdx,%rax
.L1:
addq %r15,%rax
movb 0(%rax),%r14b
movq %rdx,%rax
.L2:
movb %r14b,%bl
shrb %cl,%bl
andb $1,%bl
movb %bl,%r9b
xorb %r8b,%bl
movb %r9b,%r8b
// <X2',Z2',X3',Z3'> ← Dense-Swap(<X2',Z2',X3',Z3'>,b)
movzbl %bl,%ebx
imul $4,%ebx,%ebx
movl %ebx,56(%rsp)
vpbroadcastq 56(%rsp),%zmm5
vpaddq swap_c(%rip),%zmm5,%zmm5
vpandq swap_mask(%rip),%zmm5,%zmm5
vpermq %zmm0,%zmm5,%zmm0
vpermq %zmm2,%zmm5,%zmm2
vpermq %zmm4,%zmm5,%zmm4
// permutation indices to generate duplicates
vmovdqa64 dup_mask2(%rip),%zmm10
vmovdqa64 dup_mask3(%rip),%zmm11
vpermq %zmm0,%zmm10,%zmm5
vpermq %zmm0,%zmm11,%zmm6
vpaddd hh1_p1(%rip),%zmm5,%zmm5
vpxorq hh1_xor(%rip),%zmm6,%zmm6
vpaddd %zmm5,%zmm6,%zmm0
vpermq %zmm2,%zmm10,%zmm5
vpermq %zmm2,%zmm11,%zmm6
vpaddd hh1_p2(%rip),%zmm5,%zmm5
vpxorq hh1_xor(%rip),%zmm6,%zmm6
vpaddd %zmm5,%zmm6,%zmm2
vpermq %zmm4,%zmm10,%zmm5
vpermq %zmm4,%zmm11,%zmm6
vpaddd hh1_p3(%rip),%zmm5,%zmm5
vpxorq hh1_xor(%rip),%zmm6,%zmm6
vpaddd %zmm5,%zmm6,%zmm4
vpsrlq $32,%zmm0,%zmm1
vpsrlq $32,%zmm2,%zmm3
// <T1',T2',T1',T2'> ← Dense-Dup(<T1',T2',T4',T3'>)
vmovdqa64 dup_mask1(%rip),%zmm10
vpermq %zmm0,%zmm10,%zmm5
vpermq %zmm2,%zmm10,%zmm7
vpermq %zmm4,%zmm10,%zmm9
vpsrlq $32,%zmm5,%zmm6
vpsrlq $32,%zmm7,%zmm8
// <T5',T6',T7',T8'> ← Mul(<T1',T2',T4',T3'>,<T1',T2',T1',T2'>)
vpmuludq %zmm5,%zmm0,%zmm10
vpmuludq %zmm6,%zmm0,%zmm11
vpmuludq %zmm5,%zmm1,%zmm20
vpaddq %zmm20,%zmm11,%zmm11
vpmuludq %zmm7,%zmm0,%zmm12
vpmuludq %zmm6,%zmm1,%zmm21
vpaddq %zmm21,%zmm12,%zmm12
vpmuludq %zmm5,%zmm2,%zmm22
vpaddq %zmm22,%zmm12,%zmm12
vpmuludq %zmm8,%zmm0,%zmm13
vpmuludq %zmm7,%zmm1,%zmm23
vpaddq %zmm23,%zmm13,%zmm13
vpmuludq %zmm6,%zmm2,%zmm24
vpmuludq %zmm5,%zmm3,%zmm25
vpaddq %zmm24,%zmm25,%zmm25
vpaddq %zmm25,%zmm13,%zmm13
vpmuludq %zmm9,%zmm0,%zmm14
vpmuludq %zmm8,%zmm1,%zmm26
vpaddq %zmm26,%zmm14,%zmm14
vpmuludq %zmm7,%zmm2,%zmm27
vpmuludq %zmm6,%zmm3,%zmm28
vpaddq %zmm27,%zmm28,%zmm15
vpmuludq %zmm5,%zmm4,%zmm29
vpaddq %zmm14,%zmm15,%zmm14
vpaddq %zmm29,%zmm14,%zmm14
vpmuludq %zmm9,%zmm1,%zmm15
vpmuludq %zmm8,%zmm2,%zmm30
vpaddq %zmm30,%zmm15,%zmm15
vpmuludq %zmm7,%zmm3,%zmm31
vpmuludq %zmm6,%zmm4,%zmm16
vpaddq %zmm31,%zmm16,%zmm19
vpaddq %zmm19,%zmm15,%zmm15
vpmuludq %zmm9,%zmm2,%zmm16
vpmuludq %zmm8,%zmm3,%zmm20
vpaddq %zmm20,%zmm16,%zmm16
vpmuludq %zmm7,%zmm4,%zmm21
vpaddq %zmm21,%zmm16,%zmm16
vpmuludq %zmm9,%zmm3,%zmm17
vpmuludq %zmm8,%zmm4,%zmm31
vpaddq %zmm31,%zmm17,%zmm17
vpmuludq %zmm9,%zmm4,%zmm18
vmovdqa64 %zmm15,64(%rsp)
vmovdqa64 %zmm16,128(%rsp)
vmovdqa64 %zmm17,192(%rsp)
vmovdqa64 %zmm18,256(%rsp)
vpshufd $218,%zmm0,%zmm20
vpshufd $218,%zmm1,%zmm21
vpshufd $218,%zmm2,%zmm22
vpshufd $218,%zmm3,%zmm23
vpshufd $218,%zmm4,%zmm24
vpshufd $218,%zmm5,%zmm25
vpshufd $218,%zmm6,%zmm26
vpshufd $218,%zmm7,%zmm27
vpshufd $218,%zmm8,%zmm28
vpshufd $218,%zmm9,%zmm29
vpaddq %zmm20,%zmm0,%zmm0
vpaddq %zmm21,%zmm1,%zmm1
vpaddq %zmm22,%zmm2,%zmm2
vpaddq %zmm23,%zmm3,%zmm3
vpaddq %zmm24,%zmm4,%zmm4
vpaddq %zmm25,%zmm5,%zmm5
vpaddq %zmm26,%zmm6,%zmm6
vpaddq %zmm27,%zmm7,%zmm7
vpaddq %zmm28,%zmm8,%zmm8
vpaddq %zmm29,%zmm9,%zmm9
vpmuludq %zmm5,%zmm0,%zmm15
vpmuludq %zmm6,%zmm0,%zmm16
vpmuludq %zmm5,%zmm1,%zmm31
vpaddq %zmm31,%zmm16,%zmm16
vpmuludq %zmm7,%zmm0,%zmm17
vpmuludq %zmm6,%zmm1,%zmm31
vpaddq %zmm31,%zmm17,%zmm17
vpmuludq %zmm5,%zmm2,%zmm31
vpaddq %zmm31,%zmm17,%zmm17
vpmuludq %zmm8,%zmm0,%zmm18
vpmuludq %zmm7,%zmm1,%zmm31
vpaddq %zmm31,%zmm18,%zmm18
vpmuludq %zmm6,%zmm2,%zmm31
vpaddq %zmm31,%zmm18,%zmm18
vpmuludq %zmm5,%zmm3,%zmm31
vpaddq %zmm31,%zmm18,%zmm18
vpmuludq %zmm9,%zmm0,%zmm19
vpmuludq %zmm8,%zmm1,%zmm31
vpaddq %zmm31,%zmm19,%zmm19
vpmuludq %zmm7,%zmm2,%zmm31
vpaddq %zmm31,%zmm19,%zmm19
vpmuludq %zmm6,%zmm3,%zmm31
vpaddq %zmm31,%zmm19,%zmm19
vpmuludq %zmm5,%zmm4,%zmm31
vpaddq %zmm31,%zmm19,%zmm19
vpmuludq %zmm9,%zmm1,%zmm20
vpmuludq %zmm8,%zmm2,%zmm31
vpaddq %zmm31,%zmm20,%zmm20
vpmuludq %zmm7,%zmm3,%zmm31
vpaddq %zmm31,%zmm20,%zmm20
vpmuludq %zmm6,%zmm4,%zmm31
vpaddq %zmm31,%zmm20,%zmm20
vpmuludq %zmm9,%zmm2,%zmm21
vpmuludq %zmm8,%zmm3,%zmm31
vpaddq %zmm31,%zmm21,%zmm21
vpmuludq %zmm7,%zmm4,%zmm31
vpaddq %zmm31,%zmm21,%zmm21
vpmuludq %zmm9,%zmm3,%zmm22
vpmuludq %zmm8,%zmm4,%zmm31
vpaddq %zmm31,%zmm22,%zmm22
vpmuludq %zmm9,%zmm4,%zmm23
vpshufd $238,%zmm10,%zmm0
vpshufd $238,%zmm11,%zmm1
vpshufd $238,%zmm12,%zmm2
vpshufd $238,%zmm13,%zmm3
vpshufd $238,%zmm14,%zmm24
vpshufd $238,64(%rsp),%zmm25
vpshufd $238,128(%rsp),%zmm26
vpshufd $238,192(%rsp),%zmm27
vpshufd $238,256(%rsp),%zmm28
vpsubq %zmm10,%zmm15,%zmm15
vpsubq %zmm11,%zmm16,%zmm16
vpsubq %zmm12,%zmm17,%zmm17
vpsubq %zmm13,%zmm18,%zmm18
vpsubq %zmm14,%zmm19,%zmm19
vpsubq 64(%rsp),%zmm20,%zmm20
vpsubq 128(%rsp),%zmm21,%zmm21
vpsubq 192(%rsp),%zmm22,%zmm22
vpsubq 256(%rsp),%zmm23,%zmm23
vpsubq %zmm0,%zmm15,%zmm15
vpsubq %zmm1,%zmm16,%zmm16
vpsubq %zmm2,%zmm17,%zmm17
vpsubq %zmm3,%zmm18,%zmm18
vpsubq %zmm24,%zmm19,%zmm19
vpsubq %zmm25,%zmm20,%zmm20
vpsubq %zmm26,%zmm21,%zmm21
vpsubq %zmm27,%zmm22,%zmm22
vpsubq %zmm28,%zmm23,%zmm23
vpaddq 64(%rsp),%zmm15,%zmm15
vpaddq 128(%rsp),%zmm16,%zmm16
vpaddq 192(%rsp),%zmm17,%zmm17
vpaddq 256(%rsp),%zmm18,%zmm18
vpaddq %zmm0,%zmm20,%zmm20
vpaddq %zmm1,%zmm21,%zmm21
vpaddq %zmm2,%zmm22,%zmm22
vpaddq %zmm3,%zmm23,%zmm23
vpsrlq $26,%zmm20,%zmm31
vpaddq %zmm31,%zmm21,%zmm21
vpandq vecmask26(%rip),%zmm20,%zmm20
vpmuludq vec608(%rip),%zmm20,%zmm20
vpaddq %zmm20,%zmm10,%zmm10
vpsrlq $26,%zmm21,%zmm31
vpaddq %zmm31,%zmm22,%zmm22
vpandq vecmask26(%rip),%zmm21,%zmm21
vpmuludq vec608(%rip),%zmm21,%zmm21
vpaddq %zmm21,%zmm11,%zmm11
vpsrlq $26,%zmm22,%zmm31
vpaddq %zmm31,%zmm23,%zmm23
vpandq vecmask26(%rip),%zmm22,%zmm22
vpmuludq vec608(%rip),%zmm22,%zmm22
vpaddq %zmm22,%zmm12,%zmm12
vpsrlq $26,%zmm23,%zmm31
vpaddq %zmm31,%zmm24,%zmm24
vpandq vecmask26(%rip),%zmm23,%zmm23
vpmuludq vec608(%rip),%zmm23,%zmm23
vpaddq %zmm23,%zmm13,%zmm13
vpsrlq $26,%zmm24,%zmm31
vpaddq %zmm31,%zmm25,%zmm25
vpandq vecmask26(%rip),%zmm24,%zmm24
vpmuludq vec608(%rip),%zmm24,%zmm24
vpaddq %zmm24,%zmm14,%zmm14
vpsrlq $26,%zmm25,%zmm31
vpaddq %zmm31,%zmm26,%zmm26
vpandq vecmask26(%rip),%zmm25,%zmm25
vpmuludq vec608(%rip),%zmm25,%zmm25
vpaddq %zmm25,%zmm15,%zmm15
vpsrlq $26,%zmm26,%zmm31
vpaddq %zmm31,%zmm27,%zmm27
vpandq vecmask26(%rip),%zmm26,%zmm26
vpmuludq vec608(%rip),%zmm26,%zmm26
vpaddq %zmm26,%zmm16,%zmm16
vpsrlq $26,%zmm27,%zmm31
vpaddq %zmm31,%zmm28,%zmm28
vpandq vecmask26(%rip),%zmm27,%zmm27
vpmuludq vec608(%rip),%zmm27,%zmm27
vpaddq %zmm27,%zmm17,%zmm17
vpsrlq $26,%zmm28,%zmm31
vpandq vecmask26(%rip),%zmm28,%zmm28
vpmuludq vec608(%rip),%zmm28,%zmm28
vpaddq %zmm28,%zmm18,%zmm18
vpmuludq vec608(%rip),%zmm31,%zmm31
vpaddq %zmm31,%zmm19,%zmm19
vpshufd $68,%zmm15,%zmm0
vpshufd $68,%zmm16,%zmm1
vpshufd $68,%zmm17,%zmm2
vpshufd $68,%zmm18,%zmm3
vpshufd $68,%zmm19,%zmm4
vpblendmd %zmm0,%zmm10,%zmm10{%k1}
vpblendmd %zmm1,%zmm11,%zmm11{%k1}
vpblendmd %zmm2,%zmm12,%zmm12{%k1}
vpblendmd %zmm3,%zmm13,%zmm13{%k1}
vpblendmd %zmm4,%zmm14,%zmm14{%k1}
vpsrlq $26,%zmm10,%zmm31
vpaddq %zmm31,%zmm11,%zmm11
vpandq vecmask26(%rip),%zmm10,%zmm10
vpsrlq $26,%zmm11,%zmm31
vpaddq %zmm31,%zmm12,%zmm12
vpandq vecmask26(%rip),%zmm11,%zmm11
vpsrlq $26,%zmm12,%zmm31
vpaddq %zmm31,%zmm13,%zmm13
vpandq vecmask26(%rip),%zmm12,%zmm12
vpsrlq $26,%zmm13,%zmm31
vpaddq %zmm31,%zmm14,%zmm14
vpandq vecmask26(%rip),%zmm13,%zmm13
vmovdqa64 %zmm14,%zmm18
vpsrlq $26,%zmm14,%zmm31
vpshufd $78,%zmm31,%zmm31
vpandq vecmask1(%rip),%zmm31,%zmm31
vpaddq %zmm31,%zmm10,%zmm10
vpandq vecmask26(%rip),%zmm14,%zmm14
vpsrlq $21,%zmm18,%zmm31
vpshufd $78,%zmm31,%zmm31
vpandq vecmask2(%rip),%zmm31,%zmm31
vpaddq %zmm31,%zmm10,%zmm10
vpaddq %zmm31,%zmm31,%zmm31
vpaddq %zmm31,%zmm10,%zmm10
vpsllq $3,%zmm31,%zmm31
vpaddq %zmm31,%zmm10,%zmm10
vpandq vecmask21(%rip),%zmm18,%zmm18
vpblendmd %zmm18,%zmm14,%zmm14{%k1}
vpsrlq $26,%zmm10,%zmm31
vpaddq %zmm31,%zmm11,%zmm11
vpandq vecmask26(%rip),%zmm10,%zmm10
// <T9',T10',T11',T12'> ← Dense-H2-H(<T5',T6',T7',T8'>)
vpsllq $32,%zmm11,%zmm1
vporq %zmm1,%zmm10,%zmm0
vpsllq $32,%zmm13,%zmm3
vporq %zmm3,%zmm12,%zmm2
vmovdqa64 dup_mask2(%rip),%zmm15
vmovdqa64 dup_mask3(%rip),%zmm16
vpermq %zmm0,%zmm15,%zmm5
vpandq h2h_mask(%rip),%zmm5,%zmm5
vpermq %zmm0,%zmm16,%zmm6
vpaddd h2h_p1(%rip),%zmm5,%zmm5
vpxorq h2h_xor(%rip),%zmm6,%zmm6
vpaddd %zmm5,%zmm6,%zmm0
vpermq %zmm2,%zmm15,%zmm5
vpandq h2h_mask(%rip),%zmm5,%zmm5
vpermq %zmm2,%zmm16,%zmm6
vpaddd h2h_p2(%rip),%zmm5,%zmm5
vpxorq h2h_xor(%rip),%zmm6,%zmm6
vpaddd %zmm5,%zmm6,%zmm2
vpermq %zmm14,%zmm15,%zmm5
vpandq h2h_mask(%rip),%zmm5,%zmm5
vpermq %zmm14,%zmm16,%zmm6
vpaddd h2h_p3(%rip),%zmm5,%zmm5
vpxorq h2h_xor(%rip),%zmm6,%zmm6
vpaddd %zmm5,%zmm6,%zmm4
vpsrlq $32,%zmm0,%zmm1
vpsrlq $32,%zmm2,%zmm3
// <T9',T10',1',X1'> ← Blend(<0',0',1',X1'>,<T9',T10',T11',T12'>,1100)
vpblendmd 0(%rsi),%zmm0,%zmm5{%k2}
vmovdqa64 %zmm5,64(%rsp)
vpblendmd 64(%rsi),%zmm1,%zmm6{%k2}
vmovdqa64 %zmm6,128(%rsp)
vpblendmd 128(%rsi),%zmm2,%zmm7{%k2}
vmovdqa64 %zmm7,192(%rsp)
vpblendmd 192(%rsi),%zmm3,%zmm8{%k2}
vmovdqa64 %zmm8,256(%rsp)
vpblendmd 256(%rsi),%zmm4,%zmm9{%k2}
vmovdqa64 %zmm9,320(%rsp)
// <0,T13',0,0> ← Unreduced-Mulc(<T9',T10',T11',T12'>,<0,a24',0,0>)
// <T5',T14',T7',T8'> ← Add(<0,T13',0,0>,<T5',T6',T7',T8'>)
vpmuludq a24(%rip),%zmm0,%zmm15
vpmuludq a24(%rip),%zmm1,%zmm16
vpmuludq a24(%rip),%zmm2,%zmm17
vpmuludq a24(%rip),%zmm3,%zmm18
vpmuludq a24(%rip),%zmm4,%zmm19
vpaddq %zmm10,%zmm15,%zmm15
vpaddq %zmm11,%zmm16,%zmm16
vpaddq %zmm12,%zmm17,%zmm17
vpaddq %zmm13,%zmm18,%zmm18
vpaddq %zmm14,%zmm19,%zmm19
vpsrlq $26,%zmm15,%zmm31
vpaddq %zmm31,%zmm16,%zmm16
vpandq vecmask26(%rip),%zmm15,%zmm15
vpsrlq $26,%zmm16,%zmm31
vpaddq %zmm31,%zmm17,%zmm17
vpandq vecmask26(%rip),%zmm16,%zmm16
vpsrlq $26,%zmm17,%zmm31
vpaddq %zmm31,%zmm18,%zmm18
vpandq vecmask26(%rip),%zmm17,%zmm17
vpsrlq $26,%zmm18,%zmm31
vpaddq %zmm31,%zmm19,%zmm19
vpandq vecmask26(%rip),%zmm18,%zmm18
vmovdqa64 %zmm19,%zmm30
vpsrlq $26,%zmm19,%zmm31
vpshufd $78,%zmm31,%zmm31
vpandq vecmask1(%rip),%zmm31,%zmm31
vpaddq %zmm31,%zmm15,%zmm15
vpandq vecmask26(%rip),%zmm19,%zmm19
vpsrlq $21,%zmm30,%zmm31
vpshufd $78,%zmm31,%zmm31
vpandq vecmask2(%rip),%zmm31,%zmm31
vpaddq %zmm31,%zmm15,%zmm15
vpaddq %zmm31,%zmm31,%zmm31
vpaddq %zmm31,%zmm15,%zmm15
vpsllq $3,%zmm31,%zmm31
vpaddq %zmm31,%zmm15,%zmm15
vpandq vecmask21(%rip),%zmm30,%zmm30
vpblendmd %zmm30,%zmm19,%zmm19{%k1}
vpsrlq $26,%zmm15,%zmm31
vpaddq %zmm31,%zmm16,%zmm16
vpandq vecmask26(%rip),%zmm15,%zmm15
vmovdqa64 %zmm15,384(%rsp)
vmovdqa64 %zmm16,448(%rsp)
vmovdqa64 %zmm17,512(%rsp)
vmovdqa64 %zmm18,576(%rsp)
vmovdqa64 %zmm19,640(%rsp)
// <*,*,T15',T16'> ← Sqr(<T9',T10',T11',T12'>)
vpmuludq %zmm0,%zmm0,%zmm10
vpmuludq %zmm1,%zmm0,%zmm11
vpaddq %zmm11,%zmm11,%zmm11
vpmuludq %zmm2,%zmm0,%zmm12
vpaddq %zmm12,%zmm12,%zmm12
vpmuludq %zmm1,%zmm1,%zmm22
vpaddq %zmm22,%zmm12,%zmm12
vpmuludq %zmm3,%zmm0,%zmm13
vpmuludq %zmm2,%zmm1,%zmm23
vpaddq %zmm23,%zmm13,%zmm13
vpaddq %zmm13,%zmm13,%zmm13
vpmuludq %zmm4,%zmm0,%zmm14
vpmuludq %zmm3,%zmm1,%zmm26
vpaddq %zmm26,%zmm14,%zmm14
vpaddq %zmm14,%zmm14,%zmm14
vpmuludq %zmm2,%zmm2,%zmm27
vpaddq %zmm27,%zmm14,%zmm14
vpmuludq %zmm4,%zmm1,%zmm15
vpmuludq %zmm3,%zmm2,%zmm30
vpaddq %zmm30,%zmm15,%zmm15
vpaddq %zmm15,%zmm15,%zmm5
vpmuludq %zmm4,%zmm2,%zmm16
vpaddq %zmm16,%zmm16,%zmm16
vpmuludq %zmm3,%zmm3,%zmm21
vpaddq %zmm21,%zmm16,%zmm6
vpmuludq %zmm4,%zmm3,%zmm17
vpaddq %zmm17,%zmm17,%zmm7
vpmuludq %zmm4,%zmm4,%zmm8
vpshufd $238,%zmm0,%zmm20
vpshufd $238,%zmm1,%zmm21
vpshufd $238,%zmm2,%zmm22
vpshufd $238,%zmm3,%zmm23
vpshufd $238,%zmm4,%zmm24
vpaddq %zmm20,%zmm0,%zmm0
vpaddq %zmm21,%zmm1,%zmm1
vpaddq %zmm22,%zmm2,%zmm2
vpaddq %zmm23,%zmm3,%zmm3
vpaddq %zmm24,%zmm4,%zmm4
vpmuludq %zmm0,%zmm0,%zmm15
vpmuludq %zmm1,%zmm0,%zmm16
vpaddq %zmm16,%zmm16,%zmm16
vpmuludq %zmm2,%zmm0,%zmm17
vpaddq %zmm17,%zmm17,%zmm17
vpmuludq %zmm1,%zmm1,%zmm22
vpaddq %zmm22,%zmm17,%zmm17
vpmuludq %zmm3,%zmm0,%zmm18
vpmuludq %zmm2,%zmm1,%zmm23
vpaddq %zmm23,%zmm18,%zmm18
vpaddq %zmm18,%zmm18,%zmm18
vpmuludq %zmm4,%zmm0,%zmm19
vpmuludq %zmm3,%zmm1,%zmm26
vpaddq %zmm26,%zmm19,%zmm19
vpaddq %zmm19,%zmm19,%zmm19
vpmuludq %zmm2,%zmm2,%zmm27
vpaddq %zmm27,%zmm19,%zmm19
vpmuludq %zmm4,%zmm1,%zmm20
vpmuludq %zmm3,%zmm2,%zmm30
vpaddq %zmm30,%zmm20,%zmm20
vpaddq %zmm20,%zmm20,%zmm20
vpmuludq %zmm4,%zmm2,%zmm21
vpaddq %zmm21,%zmm21,%zmm21
vpmuludq %zmm3,%zmm3,%zmm31
vpaddq %zmm31,%zmm21,%zmm21
vpmuludq %zmm4,%zmm3,%zmm22
vpaddq %zmm22,%zmm22,%zmm22
vpmuludq %zmm4,%zmm4,%zmm23
vpshufd $238,%zmm10,%zmm0
vpshufd $238,%zmm11,%zmm1
vpshufd $238,%zmm12,%zmm2
vpshufd $238,%zmm13,%zmm3
vpshufd $238,%zmm14,%zmm24
vpshufd $238,%zmm5,%zmm25
vpshufd $238,%zmm6,%zmm26
vpshufd $238,%zmm7,%zmm27
vpshufd $238,%zmm8,%zmm28
vpsubq %zmm10,%zmm15,%zmm15
vpsubq %zmm11,%zmm16,%zmm16
vpsubq %zmm12,%zmm17,%zmm17
vpsubq %zmm13,%zmm18,%zmm18
vpsubq %zmm14,%zmm19,%zmm19
vpsubq %zmm5,%zmm20,%zmm20
vpsubq %zmm6,%zmm21,%zmm21
vpsubq %zmm7,%zmm22,%zmm22
vpsubq %zmm8,%zmm23,%zmm23
vpsubq %zmm0,%zmm15,%zmm15
vpsubq %zmm1,%zmm16,%zmm16
vpsubq %zmm2,%zmm17,%zmm17
vpsubq %zmm3,%zmm18,%zmm18
vpsubq %zmm24,%zmm19,%zmm19
vpsubq %zmm25,%zmm20,%zmm20
vpsubq %zmm26,%zmm21,%zmm21
vpsubq %zmm27,%zmm22,%zmm22
vpsubq %zmm28,%zmm23,%zmm23
vpaddq %zmm5,%zmm15,%zmm15
vpaddq %zmm6,%zmm16,%zmm16
vpaddq %zmm7,%zmm17,%zmm17
vpaddq %zmm8,%zmm18,%zmm18
vpaddq %zmm0,%zmm20,%zmm20
vpaddq %zmm1,%zmm21,%zmm21
vpaddq %zmm2,%zmm22,%zmm22
vpaddq %zmm3,%zmm23,%zmm23
vpsrlq $26,%zmm20,%zmm31
vpaddq %zmm31,%zmm21,%zmm21
vpandq vecmask26(%rip),%zmm20,%zmm20
vpmuludq vec608(%rip),%zmm20,%zmm20
vpaddq %zmm20,%zmm10,%zmm10
vpsrlq $26,%zmm21,%zmm31
vpaddq %zmm31,%zmm22,%zmm22
vpandq vecmask26(%rip),%zmm21,%zmm21
vpmuludq vec608(%rip),%zmm21,%zmm21
vpaddq %zmm21,%zmm11,%zmm11
vpsrlq $26,%zmm22,%zmm31
vpaddq %zmm31,%zmm23,%zmm23
vpandq vecmask26(%rip),%zmm22,%zmm22
vpmuludq vec608(%rip),%zmm22,%zmm22
vpaddq %zmm22,%zmm12,%zmm12
vpsrlq $26,%zmm23,%zmm31
vpaddq %zmm31,%zmm24,%zmm24
vpandq vecmask26(%rip),%zmm23,%zmm23
vpmuludq vec608(%rip),%zmm23,%zmm23
vpaddq %zmm23,%zmm13,%zmm13
vpsrlq $26,%zmm24,%zmm31
vpaddq %zmm31,%zmm25,%zmm25
vpandq vecmask26(%rip),%zmm24,%zmm24
vpmuludq vec608(%rip),%zmm24,%zmm24
vpaddq %zmm24,%zmm14,%zmm14
vpsrlq $26,%zmm25,%zmm31
vpaddq %zmm31,%zmm26,%zmm26
vpandq vecmask26(%rip),%zmm25,%zmm25
vpmuludq vec608(%rip),%zmm25,%zmm25
vpaddq %zmm25,%zmm15,%zmm15
vpsrlq $26,%zmm26,%zmm31
vpaddq %zmm31,%zmm27,%zmm27
vpandq vecmask26(%rip),%zmm26,%zmm26
vpmuludq vec608(%rip),%zmm26,%zmm26
vpaddq %zmm26,%zmm16,%zmm16
vpsrlq $26,%zmm27,%zmm31
vpaddq %zmm31,%zmm28,%zmm28
vpandq vecmask26(%rip),%zmm27,%zmm27
vpmuludq vec608(%rip),%zmm27,%zmm27
vpaddq %zmm27,%zmm17,%zmm17
vpsrlq $26,%zmm28,%zmm31
vpandq vecmask26(%rip),%zmm28,%zmm28
vpmuludq vec608(%rip),%zmm28,%zmm28
vpaddq %zmm28,%zmm18,%zmm18
vpmuludq vec608(%rip),%zmm31,%zmm31
vpaddq %zmm31,%zmm19,%zmm19
vpshufd $68,%zmm15,%zmm0
vpshufd $68,%zmm16,%zmm1
vpshufd $68,%zmm17,%zmm2
vpshufd $68,%zmm18,%zmm3
vpshufd $68,%zmm19,%zmm4
vpblendmd %zmm0,%zmm10,%zmm10{%k1}
vpblendmd %zmm1,%zmm11,%zmm11{%k1}
vpblendmd %zmm2,%zmm12,%zmm12{%k1}
vpblendmd %zmm3,%zmm13,%zmm13{%k1}
vpblendmd %zmm4,%zmm14,%zmm14{%k1}
vpsrlq $26,%zmm10,%zmm31
vpaddq %zmm31,%zmm11,%zmm11
vpandq vecmask26(%rip),%zmm10,%zmm10
vpsrlq $26,%zmm11,%zmm31
vpaddq %zmm31,%zmm12,%zmm12
vpandq vecmask26(%rip),%zmm11,%zmm11
vpsrlq $26,%zmm12,%zmm31
vpaddq %zmm31,%zmm13,%zmm13
vpandq vecmask26(%rip),%zmm12,%zmm12
vpsrlq $26,%zmm13,%zmm31
vpaddq %zmm31,%zmm14,%zmm14
vpandq vecmask26(%rip),%zmm13,%zmm13
vmovdqa64 %zmm14,%zmm18
vpsrlq $26,%zmm14,%zmm31
vpshufd $78,%zmm31,%zmm31
vpandq vecmask1(%rip),%zmm31,%zmm31
vpaddq %zmm31,%zmm10,%zmm10
vpandq vecmask26(%rip),%zmm14,%zmm14
vpsrlq $21,%zmm18,%zmm31
vpshufd $78,%zmm31,%zmm31
vpandq vecmask2(%rip),%zmm31,%zmm31
vpaddq %zmm31,%zmm10,%zmm10
vpaddq %zmm31,%zmm31,%zmm31
vpaddq %zmm31,%zmm10,%zmm10
vpsllq $3,%zmm31,%zmm31
vpaddq %zmm31,%zmm10,%zmm10
vpandq vecmask21(%rip),%zmm18,%zmm18
vpblendmd %zmm18,%zmm14,%zmm14{%k1}
vpsrlq $26,%zmm10,%zmm31
vpaddq %zmm31,%zmm11,%zmm11
vpandq vecmask26(%rip),%zmm10,%zmm10
// <T5',T14',T15',T16'> ← Blend(<T5',T14',T7',T8'>,<*,*,T15',T16'>,9)
vpblendmd 384(%rsp),%zmm10,%zmm0{%k3}
vpblendmd 448(%rsp),%zmm11,%zmm1{%k3}
vpblendmd 512(%rsp),%zmm12,%zmm2{%k3}
vpblendmd 576(%rsp),%zmm13,%zmm3{%k3}
vpblendmd 640(%rsp),%zmm14,%zmm4{%k3}
// <X2',Z2',X3',Z3'> ← Mul(<T5',T14',T15',T16'>,<T9',T10',1',X1'>)
vmovdqa64 64(%rsp),%zmm5
vmovdqa64 128(%rsp),%zmm6
vmovdqa64 192(%rsp),%zmm7
vmovdqa64 256(%rsp),%zmm8
vmovdqa64 320(%rsp),%zmm9
vpmuludq %zmm5,%zmm0,%zmm10
vpmuludq %zmm6,%zmm0,%zmm11
vpmuludq %zmm5,%zmm1,%zmm20
vpaddq %zmm20,%zmm11,%zmm11
vpmuludq %zmm7,%zmm0,%zmm12
vpmuludq %zmm6,%zmm1,%zmm21
vpaddq %zmm21,%zmm12,%zmm12
vpmuludq %zmm5,%zmm2,%zmm22
vpaddq %zmm22,%zmm12,%zmm12
vpmuludq %zmm8,%zmm0,%zmm13
vpmuludq %zmm7,%zmm1,%zmm23
vpaddq %zmm23,%zmm13,%zmm13
vpmuludq %zmm6,%zmm2,%zmm24
vpmuludq %zmm5,%zmm3,%zmm25
vpaddq %zmm24,%zmm25,%zmm25
vpaddq %zmm25,%zmm13,%zmm13
vpmuludq %zmm9,%zmm0,%zmm14
vpmuludq %zmm8,%zmm1,%zmm26
vpaddq %zmm26,%zmm14,%zmm14
vpmuludq %zmm7,%zmm2,%zmm27
vpmuludq %zmm6,%zmm3,%zmm28
vpaddq %zmm27,%zmm28,%zmm15
vpmuludq %zmm5,%zmm4,%zmm29
vpaddq %zmm14,%zmm15,%zmm14
vpaddq %zmm29,%zmm14,%zmm14
vpmuludq %zmm9,%zmm1,%zmm15
vpmuludq %zmm8,%zmm2,%zmm30
vpaddq %zmm30,%zmm15,%zmm15
vpmuludq %zmm7,%zmm3,%zmm31
vpmuludq %zmm6,%zmm4,%zmm16
vpaddq %zmm31,%zmm16,%zmm19
vpaddq %zmm19,%zmm15,%zmm15
vpmuludq %zmm9,%zmm2,%zmm16
vpmuludq %zmm8,%zmm3,%zmm20
vpaddq %zmm20,%zmm16,%zmm16
vpmuludq %zmm7,%zmm4,%zmm21
vpaddq %zmm21,%zmm16,%zmm16
vpmuludq %zmm9,%zmm3,%zmm17
vpmuludq %zmm8,%zmm4,%zmm31
vpaddq %zmm31,%zmm17,%zmm17
vpmuludq %zmm9,%zmm4,%zmm18
vmovdqa64 %zmm15,64(%rsp)
vmovdqa64 %zmm16,128(%rsp)
vmovdqa64 %zmm17,192(%rsp)
vmovdqa64 %zmm18,256(%rsp)
vpshufd $218,%zmm0,%zmm20
vpshufd $218,%zmm1,%zmm21
vpshufd $218,%zmm2,%zmm22
vpshufd $218,%zmm3,%zmm23
vpshufd $218,%zmm4,%zmm24
vpshufd $218,%zmm5,%zmm25
vpshufd $218,%zmm6,%zmm26
vpshufd $218,%zmm7,%zmm27
vpshufd $218,%zmm8,%zmm28
vpshufd $218,%zmm9,%zmm29
vpaddq %zmm20,%zmm0,%zmm0
vpaddq %zmm21,%zmm1,%zmm1
vpaddq %zmm22,%zmm2,%zmm2
vpaddq %zmm23,%zmm3,%zmm3
vpaddq %zmm24,%zmm4,%zmm4
vpaddq %zmm25,%zmm5,%zmm5
vpaddq %zmm26,%zmm6,%zmm6
vpaddq %zmm27,%zmm7,%zmm7
vpaddq %zmm28,%zmm8,%zmm8
vpaddq %zmm29,%zmm9,%zmm9
vpmuludq %zmm5,%zmm0,%zmm15
vpmuludq %zmm6,%zmm0,%zmm16
vpmuludq %zmm5,%zmm1,%zmm31
vpaddq %zmm31,%zmm16,%zmm16
vpmuludq %zmm7,%zmm0,%zmm17
vpmuludq %zmm6,%zmm1,%zmm31
vpaddq %zmm31,%zmm17,%zmm17
vpmuludq %zmm5,%zmm2,%zmm31
vpaddq %zmm31,%zmm17,%zmm17
vpmuludq %zmm8,%zmm0,%zmm18
vpmuludq %zmm7,%zmm1,%zmm31
vpaddq %zmm31,%zmm18,%zmm18
vpmuludq %zmm6,%zmm2,%zmm31
vpaddq %zmm31,%zmm18,%zmm18
vpmuludq %zmm5,%zmm3,%zmm31
vpaddq %zmm31,%zmm18,%zmm18
vpmuludq %zmm9,%zmm0,%zmm19
vpmuludq %zmm8,%zmm1,%zmm31
vpaddq %zmm31,%zmm19,%zmm19
vpmuludq %zmm7,%zmm2,%zmm31
vpaddq %zmm31,%zmm19,%zmm19
vpmuludq %zmm6,%zmm3,%zmm31
vpaddq %zmm31,%zmm19,%zmm19
vpmuludq %zmm5,%zmm4,%zmm31
vpaddq %zmm31,%zmm19,%zmm19
vpmuludq %zmm9,%zmm1,%zmm20
vpmuludq %zmm8,%zmm2,%zmm31
vpaddq %zmm31,%zmm20,%zmm20
vpmuludq %zmm7,%zmm3,%zmm31
vpaddq %zmm31,%zmm20,%zmm20
vpmuludq %zmm6,%zmm4,%zmm31
vpaddq %zmm31,%zmm20,%zmm20
vpmuludq %zmm9,%zmm2,%zmm21
vpmuludq %zmm8,%zmm3,%zmm31
vpaddq %zmm31,%zmm21,%zmm21
vpmuludq %zmm7,%zmm4,%zmm31
vpaddq %zmm31,%zmm21,%zmm21
vpmuludq %zmm9,%zmm3,%zmm22
vpmuludq %zmm8,%zmm4,%zmm31
vpaddq %zmm31,%zmm22,%zmm22
vpmuludq %zmm9,%zmm4,%zmm23
vpshufd $238,%zmm10,%zmm0
vpshufd $238,%zmm11,%zmm1
vpshufd $238,%zmm12,%zmm2
vpshufd $238,%zmm13,%zmm3
vpshufd $238,%zmm14,%zmm24
vpshufd $238,64(%rsp),%zmm25
vpshufd $238,128(%rsp),%zmm26
vpshufd $238,192(%rsp),%zmm27
vpshufd $238,256(%rsp),%zmm28
vpsubq %zmm10,%zmm15,%zmm15
vpsubq %zmm11,%zmm16,%zmm16
vpsubq %zmm12,%zmm17,%zmm17
vpsubq %zmm13,%zmm18,%zmm18
vpsubq %zmm14,%zmm19,%zmm19
vpsubq 64(%rsp),%zmm20,%zmm20
vpsubq 128(%rsp),%zmm21,%zmm21
vpsubq 192(%rsp),%zmm22,%zmm22
vpsubq 256(%rsp),%zmm23,%zmm23
vpsubq %zmm0,%zmm15,%zmm15
vpsubq %zmm1,%zmm16,%zmm16
vpsubq %zmm2,%zmm17,%zmm17
vpsubq %zmm3,%zmm18,%zmm18
vpsubq %zmm24,%zmm19,%zmm19
vpsubq %zmm25,%zmm20,%zmm20
vpsubq %zmm26,%zmm21,%zmm21
vpsubq %zmm27,%zmm22,%zmm22
vpsubq %zmm28,%zmm23,%zmm23
vpaddq 64(%rsp),%zmm15,%zmm15
vpaddq 128(%rsp),%zmm16,%zmm16
vpaddq 192(%rsp),%zmm17,%zmm17
vpaddq 256(%rsp),%zmm18,%zmm18
vpaddq %zmm0,%zmm20,%zmm20
vpaddq %zmm1,%zmm21,%zmm21
vpaddq %zmm2,%zmm22,%zmm22
vpaddq %zmm3,%zmm23,%zmm23
vpsrlq $26,%zmm20,%zmm31
vpaddq %zmm31,%zmm21,%zmm21
vpandq vecmask26(%rip),%zmm20,%zmm20
vpmuludq vec608(%rip),%zmm20,%zmm20
vpaddq %zmm20,%zmm10,%zmm10
vpsrlq $26,%zmm21,%zmm31
vpaddq %zmm31,%zmm22,%zmm22
vpandq vecmask26(%rip),%zmm21,%zmm21
vpmuludq vec608(%rip),%zmm21,%zmm21
vpaddq %zmm21,%zmm11,%zmm11
vpsrlq $26,%zmm22,%zmm31
vpaddq %zmm31,%zmm23,%zmm23
vpandq vecmask26(%rip),%zmm22,%zmm22
vpmuludq vec608(%rip),%zmm22,%zmm22
vpaddq %zmm22,%zmm12,%zmm12
vpsrlq $26,%zmm23,%zmm31
vpaddq %zmm31,%zmm24,%zmm24
vpandq vecmask26(%rip),%zmm23,%zmm23
vpmuludq vec608(%rip),%zmm23,%zmm23
vpaddq %zmm23,%zmm13,%zmm13
vpsrlq $26,%zmm24,%zmm31
vpaddq %zmm31,%zmm25,%zmm25
vpandq vecmask26(%rip),%zmm24,%zmm24
vpmuludq vec608(%rip),%zmm24,%zmm24
vpaddq %zmm24,%zmm14,%zmm14
vpsrlq $26,%zmm25,%zmm31
vpaddq %zmm31,%zmm26,%zmm26
vpandq vecmask26(%rip),%zmm25,%zmm25
vpmuludq vec608(%rip),%zmm25,%zmm25
vpaddq %zmm25,%zmm15,%zmm15
vpsrlq $26,%zmm26,%zmm31
vpaddq %zmm31,%zmm27,%zmm27
vpandq vecmask26(%rip),%zmm26,%zmm26
vpmuludq vec608(%rip),%zmm26,%zmm26
vpaddq %zmm26,%zmm16,%zmm16
vpsrlq $26,%zmm27,%zmm31
vpaddq %zmm31,%zmm28,%zmm28
vpandq vecmask26(%rip),%zmm27,%zmm27
vpmuludq vec608(%rip),%zmm27,%zmm27
vpaddq %zmm27,%zmm17,%zmm17
vpsrlq $26,%zmm28,%zmm31
vpandq vecmask26(%rip),%zmm28,%zmm28
vpmuludq vec608(%rip),%zmm28,%zmm28
vpaddq %zmm28,%zmm18,%zmm18
vpmuludq vec608(%rip),%zmm31,%zmm31
vpaddq %zmm31,%zmm19,%zmm19
vpshufd $68,%zmm15,%zmm0
vpshufd $68,%zmm16,%zmm1
vpshufd $68,%zmm17,%zmm2
vpshufd $68,%zmm18,%zmm3
vpshufd $68,%zmm19,%zmm4
vpblendmd %zmm0,%zmm10,%zmm0{%k1}
vpblendmd %zmm1,%zmm11,%zmm1{%k1}
vpblendmd %zmm2,%zmm12,%zmm2{%k1}
vpblendmd %zmm3,%zmm13,%zmm3{%k1}
vpblendmd %zmm4,%zmm14,%zmm4{%k1}
vpsrlq $26,%zmm0,%zmm31
vpaddq %zmm31,%zmm1,%zmm1
vpandq vecmask26(%rip),%zmm0,%zmm0
vpsrlq $26,%zmm1,%zmm31
vpaddq %zmm31,%zmm2,%zmm2
vpandq vecmask26(%rip),%zmm1,%zmm1
vpsrlq $26,%zmm2,%zmm31
vpaddq %zmm31,%zmm3,%zmm3
vpandq vecmask26(%rip),%zmm2,%zmm2
vpsrlq $26,%zmm3,%zmm31
vpaddq %zmm31,%zmm4,%zmm4
vpandq vecmask26(%rip),%zmm3,%zmm3
vmovdqa64 %zmm4,%zmm5
vpsrlq $26,%zmm4,%zmm31
vpshufd $78,%zmm31,%zmm31
vpandq vecmask1(%rip),%zmm31,%zmm31
vpaddq %zmm31,%zmm0,%zmm0
vpandq vecmask26(%rip),%zmm4,%zmm4
vpsrlq $21,%zmm5,%zmm31
vpshufd $78,%zmm31,%zmm31
vpandq vecmask2(%rip),%zmm31,%zmm31
vpaddq %zmm31,%zmm0,%zmm0
vpaddq %zmm31,%zmm31,%zmm31
vpaddq %zmm31,%zmm0,%zmm0
vpsllq $3,%zmm31,%zmm31
vpaddq %zmm31,%zmm0,%zmm0
vpandq vecmask21(%rip),%zmm5,%zmm5
vpblendmd %zmm5,%zmm4,%zmm4{%k1}
vpsrlq $26,%zmm0,%zmm31
vpaddq %zmm31,%zmm1,%zmm1
vpandq vecmask26(%rip),%zmm0,%zmm0
vpsllq $32,%zmm1,%zmm1
vporq %zmm1,%zmm0,%zmm0
vpsllq $32,%zmm3,%zmm3
vporq %zmm3,%zmm2,%zmm2
subb $1,%cl
cmpb $0,%cl
jge .L2
movb $7,%cl
subq $1,%r15
cmpq $0,%r15
jge .L1
// <X2,Z2,X3,Z3> ← Pack-D2N(<X2',Z2',X3',Z3'>)
vpsrlq $32,%zmm0,%zmm1
vpandq vecmask32(%rip),%zmm0,%zmm0
vpsrlq $32,%zmm2,%zmm3
vpandq vecmask32(%rip),%zmm2,%zmm2
vpshufd $78,%zmm0,%zmm5
vpshufd $78,%zmm1,%zmm6
vpshufd $78,%zmm2,%zmm7
vpshufd $78,%zmm3,%zmm8
vpshufd $78,%zmm4,%zmm9
// <X2,Z2,X3,Z3> ← Reduce(<X2,Z2,X3,Z3>)
vpsrlq $26,%zmm0,%zmm20
vpaddq %zmm20,%zmm1,%zmm1
vpandq vecmask26(%rip),%zmm0,%zmm0
vpsrlq $26,%zmm1,%zmm20
vpaddq %zmm20,%zmm2,%zmm2
vpandq vecmask26(%rip),%zmm1,%zmm1
vpsrlq $26,%zmm2,%zmm20
vpaddq %zmm20,%zmm3,%zmm3
vpandq vecmask26(%rip),%zmm2,%zmm2
vpsrlq $26,%zmm3,%zmm20
vpaddq %zmm20,%zmm4,%zmm4
vpandq vecmask26(%rip),%zmm3,%zmm3
vpsrlq $26,%zmm4,%zmm20
vpaddq %zmm20,%zmm5,%zmm5
vpandq vecmask26(%rip),%zmm4,%zmm4
vpsrlq $26,%zmm5,%zmm20
vpaddq %zmm20,%zmm6,%zmm6
vpandq vecmask26(%rip),%zmm5,%zmm5
vpsrlq $26,%zmm6,%zmm20
vpaddq %zmm20,%zmm7,%zmm7
vpandq vecmask26(%rip),%zmm6,%zmm6
vpsrlq $26,%zmm7,%zmm20
vpaddq %zmm20,%zmm8,%zmm8
vpandq vecmask26(%rip),%zmm7,%zmm7
vpsrlq $26,%zmm8,%zmm20
vpaddq %zmm20,%zmm9,%zmm9
vpandq vecmask26(%rip),%zmm8,%zmm8
vpsrlq $21,%zmm9,%zmm20
vpmuludq vec19(%rip),%zmm20,%zmm20
vpaddq %zmm20,%zmm0,%zmm0
vpandq vecmask21(%rip),%zmm9,%zmm9
vpsrlq $26,%zmm0,%zmm20
vpaddq %zmm20,%zmm1,%zmm1
vpandq vecmask26(%rip),%zmm0,%zmm0
// store <X2,Z2,X3,Z3>
vmovdqa64 %zmm0,0(%rdi)
vmovdqa64 %zmm1,64(%rdi)
vmovdqa64 %zmm2,128(%rdi)
vmovdqa64 %zmm3,192(%rdi)
vmovdqa64 %zmm4,256(%rdi)
vmovdqa64 %zmm5,320(%rdi)
vmovdqa64 %zmm6,384(%rdi)
vmovdqa64 %zmm7,448(%rdi)
vmovdqa64 %zmm8,512(%rdi)
vmovdqa64 %zmm9,576(%rdi)
movq 0(%rsp),%r11
movq 8(%rsp),%r12
movq 16(%rsp),%r13
movq 24(%rsp),%r14
movq 32(%rsp),%r15
movq 40(%rsp),%rbx
movq 48(%rsp),%rbp
movq %r11,%rsp
ret
.section .note.GNU-stack,"",@progbits