-rw-r--r-- 25525 lib25519-20240321/crypto_multiscalar/ed25519/amd64-avx2-9l-mxaa-p3/ge25519_add.S raw
#include "crypto_asm_hidden.h"
#include "consts_namespace.h"
// ge25519_add
.p2align 5
ASM_HIDDEN _CRYPTO_NAMESPACE(ge25519_add)
.globl _CRYPTO_NAMESPACE(ge25519_add)
ASM_HIDDEN CRYPTO_NAMESPACE(ge25519_add)
.globl CRYPTO_NAMESPACE(ge25519_add)
_CRYPTO_NAMESPACE(ge25519_add):
CRYPTO_NAMESPACE(ge25519_add):
movq %rsp,%r11
andq $-32,%rsp
subq $1536,%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)
movq 64(%rsi),%r8
movq 72(%rsi),%r9
movq 80(%rsi),%r10
movq 88(%rsi),%r11
movq %r8,216(%rsp)
movq %r9,224(%rsp)
movq %r10,232(%rsp)
movq %r11,240(%rsp)
movq 96(%rsi),%r8
movq 104(%rsi),%r9
movq 112(%rsi),%r10
movq 120(%rsi),%r11
movq %r8,256(%rsp)
movq %r9,264(%rsp)
movq %r10,272(%rsp)
movq %r11,280(%rsp)
movq 64(%rdx),%r8
movq 72(%rdx),%r9
movq 80(%rdx),%r10
movq 88(%rdx),%r11
movq %r8,288(%rsp)
movq %r9,296(%rsp)
movq %r10,304(%rsp)
movq %r11,312(%rsp)
movq 96(%rdx),%r8
movq 104(%rdx),%r9
movq 112(%rdx),%r10
movq 120(%rdx),%r11
movq %r8,320(%rsp)
movq %r9,328(%rsp)
movq %r10,336(%rsp)
movq %r11,344(%rsp)
movq %rdx,%rcx
/* add p1p1 */
// load
movq 32(%rsi),%rdx
movq 40(%rsi),%r8
movq 48(%rsi),%r9
movq 56(%rsi),%rax
// copy
movq %rdx,%r10
movq %r8,%r11
movq %r9,%r12
movq %rax,%r13
// sub
subq 0(%rsi),%rdx
sbbq 8(%rsi),%r8
sbbq 16(%rsi),%r9
sbbq 24(%rsi),%rax
movq $0,%r14
movq $38,%r15
cmovae %r14,%r15
subq %r15,%rdx
sbbq %r14,%r8
sbbq %r14,%r9
sbbq %r14,%rax
cmovc %r15,%r14
subq %r14,%rdx
// add
addq 0(%rsi),%r10
adcq 8(%rsi),%r11
adcq 16(%rsi),%r12
adcq 24(%rsi),%r13
movq $0,%r14
movq $38,%r15
cmovae %r14,%r15
addq %r15,%r10
adcq %r14,%r11
adcq %r14,%r12
adcq %r14,%r13
cmovc %r15,%r14
addq %r14,%r10
// store
movq %rdx,56(%rsp)
movq %r8,64(%rsp)
movq %r9,72(%rsp)
movq %rax,80(%rsp)
// store
movq %r10,88(%rsp)
movq %r11,96(%rsp)
movq %r12,104(%rsp)
movq %r13,112(%rsp)
// load
movq 32(%rcx),%rdx
movq 40(%rcx),%r8
movq 48(%rcx),%r9
movq 56(%rcx),%rax
// copy
movq %rdx,%r10
movq %r8,%r11
movq %r9,%r12
movq %rax,%r13
// sub
subq 0(%rcx),%rdx
sbbq 8(%rcx),%r8
sbbq 16(%rcx),%r9
sbbq 24(%rcx),%rax
movq $0,%r14
movq $38,%r15
cmovae %r14,%r15
subq %r15,%rdx
sbbq %r14,%r8
sbbq %r14,%r9
sbbq %r14,%rax
cmovc %r15,%r14
subq %r14,%rdx
// add
addq 0(%rcx),%r10
adcq 8(%rcx),%r11
adcq 16(%rcx),%r12
adcq 24(%rcx),%r13
movq $0,%r14
movq $38,%r15
cmovae %r14,%r15
addq %r15,%r10
adcq %r14,%r11
adcq %r14,%r12
adcq %r14,%r13
cmovc %r15,%r14
addq %r14,%r10
// store
movq %rdx,120(%rsp)
movq %r8,128(%rsp)
movq %r9,136(%rsp)
movq %rax,144(%rsp)
// store
movq %r10,152(%rsp)
movq %r11,160(%rsp)
movq %r12,168(%rsp)
movq %r13,176(%rsp)
// mul
movq 56(%rsp),%rdx
mulx 120(%rsp),%r8,%r9
mulx 128(%rsp),%rcx,%r10
addq %rcx,%r9
mulx 136(%rsp),%rcx,%r11
adcq %rcx,%r10
mulx 144(%rsp),%rcx,%r12
adcq %rcx,%r11
adcq $0,%r12
movq 64(%rsp),%rdx
mulx 120(%rsp),%rax,%rbx
mulx 128(%rsp),%rcx,%rbp
addq %rcx,%rbx
mulx 136(%rsp),%rcx,%rsi
adcq %rcx,%rbp
mulx 144(%rsp),%rcx,%r13
adcq %rcx,%rsi
adcq $0,%r13
addq %rax,%r9
adcq %rbx,%r10
adcq %rbp,%r11
adcq %rsi,%r12
adcq $0,%r13
movq 72(%rsp),%rdx
mulx 120(%rsp),%rax,%rbx
mulx 128(%rsp),%rcx,%rbp
addq %rcx,%rbx
mulx 136(%rsp),%rcx,%rsi
adcq %rcx,%rbp
mulx 144(%rsp),%rcx,%r14
adcq %rcx,%rsi
adcq $0,%r14
addq %rax,%r10
adcq %rbx,%r11
adcq %rbp,%r12
adcq %rsi,%r13
adcq $0,%r14
movq 80(%rsp),%rdx
mulx 120(%rsp),%rax,%rbx
mulx 128(%rsp),%rcx,%rbp
addq %rcx,%rbx
mulx 136(%rsp),%rcx,%rsi
adcq %rcx,%rbp
mulx 144(%rsp),%rcx,%r15
adcq %rcx,%rsi
adcq $0,%r15
addq %rax,%r11
adcq %rbx,%r12
adcq %rbp,%r13
adcq %rsi,%r14
adcq $0,%r15
movq $38,%rdx
mulx %r12,%r12,%rbx
mulx %r13,%r13,%rcx
addq %rbx,%r13
mulx %r14,%r14,%rbx
adcq %rcx,%r14
mulx %r15,%r15,%rcx
adcq %rbx,%r15
adcq $0,%rcx
addq %r12,%r8
adcq %r13,%r9
adcq %r14,%r10
adcq %r15,%r11
adcq $0,%rcx
shld $1,%r11,%rcx
andq mask63(%rip),%r11
imul $19,%rcx,%rcx
addq %rcx,%r8
adcq $0,%r9
adcq $0,%r10
adcq $0,%r11
// store
movq %r8,56(%rsp)
movq %r9,64(%rsp)
movq %r10,72(%rsp)
movq %r11,80(%rsp)
// mul
movq 88(%rsp),%rdx
mulx 152(%rsp),%r8,%r9
mulx 160(%rsp),%rcx,%r10
addq %rcx,%r9
mulx 168(%rsp),%rcx,%r11
adcq %rcx,%r10
mulx 176(%rsp),%rcx,%r12
adcq %rcx,%r11
adcq $0,%r12
movq 96(%rsp),%rdx
mulx 152(%rsp),%rax,%rbx
mulx 160(%rsp),%rcx,%rbp
addq %rcx,%rbx
mulx 168(%rsp),%rcx,%rsi
adcq %rcx,%rbp
mulx 176(%rsp),%rcx,%r13
adcq %rcx,%rsi
adcq $0,%r13
addq %rax,%r9
adcq %rbx,%r10
adcq %rbp,%r11
adcq %rsi,%r12
adcq $0,%r13
movq 104(%rsp),%rdx
mulx 152(%rsp),%rax,%rbx
mulx 160(%rsp),%rcx,%rbp
addq %rcx,%rbx
mulx 168(%rsp),%rcx,%rsi
adcq %rcx,%rbp
mulx 176(%rsp),%rcx,%r14
adcq %rcx,%rsi
adcq $0,%r14
addq %rax,%r10
adcq %rbx,%r11
adcq %rbp,%r12
adcq %rsi,%r13
adcq $0,%r14
movq 112(%rsp),%rdx
mulx 152(%rsp),%rax,%rbx
mulx 160(%rsp),%rcx,%rbp
addq %rcx,%rbx
mulx 168(%rsp),%rcx,%rsi
adcq %rcx,%rbp
mulx 176(%rsp),%rcx,%r15
adcq %rcx,%rsi
adcq $0,%r15
addq %rax,%r11
adcq %rbx,%r12
adcq %rbp,%r13
adcq %rsi,%r14
adcq $0,%r15
movq $38,%rdx
mulx %r12,%r12,%rbx
mulx %r13,%r13,%rcx
addq %rbx,%r13
mulx %r14,%r14,%rbx
adcq %rcx,%r14
mulx %r15,%r15,%rcx
adcq %rbx,%r15
adcq $0,%rcx
addq %r12,%r8
adcq %r13,%r9
adcq %r14,%r10
adcq %r15,%r11
adcq $0,%rcx
shld $1,%r11,%rcx
andq mask63(%rip),%r11
imul $19,%rcx,%rcx
addq %rcx,%r8
adcq $0,%r9
adcq $0,%r10
adcq $0,%r11
// copy
movq %r8,%r12
movq %r9,%r13
movq %r10,%r14
movq %r11,%r15
// add
addq 56(%rsp),%r12
adcq 64(%rsp),%r13
adcq 72(%rsp),%r14
adcq 80(%rsp),%r15
movq $0,%rdx
movq $38,%rax
cmovae %rdx,%rax
addq %rax,%r12
adcq %rdx,%r13
adcq %rdx,%r14
adcq %rdx,%r15
cmovc %rax,%rdx
addq %rdx,%r12
// sub
subq 56(%rsp),%r8
sbbq 64(%rsp),%r9
sbbq 72(%rsp),%r10
sbbq 80(%rsp),%r11
movq $0,%rdx
mov $38,%rax
cmovae %rdx,%rax
subq %rax,%r8
sbbq %rdx,%r9
sbbq %rdx,%r10
sbbq %rdx,%r11
cmovc %rax,%rdx
subq %rdx,%r8
// store
movq %r8,64(%rsp)
movq %r9,72(%rsp)
movq %r10,80(%rsp)
movq %r11,88(%rsp)
// store
movq %r12,128(%rsp)
movq %r13,136(%rsp)
movq %r14,144(%rsp)
movq %r15,152(%rsp)
// mul
movq 256(%rsp),%rdx
mulx 320(%rsp),%r8,%r9
mulx 328(%rsp),%rcx,%r10
addq %rcx,%r9
mulx 336(%rsp),%rcx,%r11
adcq %rcx,%r10
mulx 344(%rsp),%rcx,%r12
adcq %rcx,%r11
adcq $0,%r12
movq 264(%rsp),%rdx
mulx 320(%rsp),%rax,%rbx
mulx 328(%rsp),%rcx,%rbp
addq %rcx,%rbx
mulx 336(%rsp),%rcx,%rsi
adcq %rcx,%rbp
mulx 344(%rsp),%rcx,%r13
adcq %rcx,%rsi
adcq $0,%r13
addq %rax,%r9
adcq %rbx,%r10
adcq %rbp,%r11
adcq %rsi,%r12
adcq $0,%r13
movq 272(%rsp),%rdx
mulx 320(%rsp),%rax,%rbx
mulx 328(%rsp),%rcx,%rbp
addq %rcx,%rbx
mulx 336(%rsp),%rcx,%rsi
adcq %rcx,%rbp
mulx 344(%rsp),%rcx,%r14
adcq %rcx,%rsi
adcq $0,%r14
addq %rax,%r10
adcq %rbx,%r11
adcq %rbp,%r12
adcq %rsi,%r13
adcq $0,%r14
movq 280(%rsp),%rdx
mulx 320(%rsp),%rax,%rbx
mulx 328(%rsp),%rcx,%rbp
addq %rcx,%rbx
mulx 336(%rsp),%rcx,%rsi
adcq %rcx,%rbp
mulx 344(%rsp),%rcx,%r15
adcq %rcx,%rsi
adcq $0,%r15
addq %rax,%r11
adcq %rbx,%r12
adcq %rbp,%r13
adcq %rsi,%r14
adcq $0,%r15
movq $38,%rdx
mulx %r12,%r12,%rbx
mulx %r13,%r13,%rcx
addq %rbx,%r13
mulx %r14,%r14,%rbx
adcq %rcx,%r14
mulx %r15,%r15,%rcx
adcq %rbx,%r15
adcq $0,%rcx
addq %r12,%r8
adcq %r13,%r9
adcq %r14,%r10
adcq %r15,%r11
adcq $0,%rcx
shld $1,%r11,%rcx
andq mask63(%rip),%r11
imul $19,%rcx,%rcx
addq %rcx,%r8
adcq $0,%r9
adcq $0,%r10
adcq $0,%r11
// store
movq %r8,184(%rsp)
movq %r9,192(%rsp)
movq %r10,200(%rsp)
movq %r11,208(%rsp)
// mul
movq EC2D0(%rip),%rdx
mulx 184(%rsp),%r8,%r9
mulx 192(%rsp),%rcx,%r10
addq %rcx,%r9
mulx 200(%rsp),%rcx,%r11
adcq %rcx,%r10
mulx 208(%rsp),%rcx,%r12
adcq %rcx,%r11
adcq $0,%r12
movq EC2D1(%rip),%rdx
mulx 184(%rsp),%rax,%rbx
mulx 192(%rsp),%rcx,%rbp
addq %rcx,%rbx
mulx 200(%rsp),%rcx,%rsi
adcq %rcx,%rbp
mulx 208(%rsp),%rcx,%r13
adcq %rcx,%rsi
adcq $0,%r13
addq %rax,%r9
adcq %rbx,%r10
adcq %rbp,%r11
adcq %rsi,%r12
adcq $0,%r13
movq EC2D2(%rip),%rdx
mulx 184(%rsp),%rax,%rbx
mulx 192(%rsp),%rcx,%rbp
addq %rcx,%rbx
mulx 200(%rsp),%rcx,%rsi
adcq %rcx,%rbp
mulx 208(%rsp),%rcx,%r14
adcq %rcx,%rsi
adcq $0,%r14
addq %rax,%r10
adcq %rbx,%r11
adcq %rbp,%r12
adcq %rsi,%r13
adcq $0,%r14
movq EC2D3(%rip),%rdx
mulx 184(%rsp),%rax,%rbx
mulx 192(%rsp),%rcx,%rbp
addq %rcx,%rbx
mulx 200(%rsp),%rcx,%rsi
adcq %rcx,%rbp
mulx 208(%rsp),%rcx,%r15
adcq %rcx,%rsi
adcq $0,%r15
addq %rax,%r11
adcq %rbx,%r12
adcq %rbp,%r13
adcq %rsi,%r14
adcq $0,%r15
movq $38,%rdx
mulx %r12,%r12,%rbx
mulx %r13,%r13,%rcx
addq %rbx,%r13
mulx %r14,%r14,%rbx
adcq %rcx,%r14
mulx %r15,%r15,%rcx
adcq %rbx,%r15
adcq $0,%rcx
addq %r12,%r8
adcq %r13,%r9
adcq %r14,%r10
adcq %r15,%r11
adcq $0,%rcx
shld $1,%r11,%rcx
andq mask63(%rip),%r11
imul $19,%rcx,%rcx
addq %rcx,%r8
adcq $0,%r9
adcq $0,%r10
adcq $0,%r11
// store
movq %r8,184(%rsp)
movq %r9,192(%rsp)
movq %r10,200(%rsp)
movq %r11,208(%rsp)
// mul
movq 216(%rsp),%rdx
mulx 288(%rsp),%r8,%r9
mulx 296(%rsp),%rcx,%r10
addq %rcx,%r9
mulx 304(%rsp),%rcx,%r11
adcq %rcx,%r10
mulx 312(%rsp),%rcx,%r12
adcq %rcx,%r11
adcq $0,%r12
movq 224(%rsp),%rdx
mulx 288(%rsp),%rax,%rbx
mulx 296(%rsp),%rcx,%rbp
addq %rcx,%rbx
mulx 304(%rsp),%rcx,%rsi
adcq %rcx,%rbp
mulx 312(%rsp),%rcx,%r13
adcq %rcx,%rsi
adcq $0,%r13
addq %rax,%r9
adcq %rbx,%r10
adcq %rbp,%r11
adcq %rsi,%r12
adcq $0,%r13
movq 232(%rsp),%rdx
mulx 288(%rsp),%rax,%rbx
mulx 296(%rsp),%rcx,%rbp
addq %rcx,%rbx
mulx 304(%rsp),%rcx,%rsi
adcq %rcx,%rbp
mulx 312(%rsp),%rcx,%r14
adcq %rcx,%rsi
adcq $0,%r14
addq %rax,%r10
adcq %rbx,%r11
adcq %rbp,%r12
adcq %rsi,%r13
adcq $0,%r14
movq 240(%rsp),%rdx
mulx 288(%rsp),%rax,%rbx
mulx 296(%rsp),%rcx,%rbp
addq %rcx,%rbx
mulx 304(%rsp),%rcx,%rsi
adcq %rcx,%rbp
mulx 312(%rsp),%rcx,%r15
adcq %rcx,%rsi
adcq $0,%r15
addq %rax,%r11
adcq %rbx,%r12
adcq %rbp,%r13
adcq %rsi,%r14
adcq $0,%r15
movq $38,%rdx
mulx %r12,%r12,%rbx
mulx %r13,%r13,%rcx
addq %rbx,%r13
mulx %r14,%r14,%rbx
adcq %rcx,%r14
mulx %r15,%r15,%rcx
adcq %rbx,%r15
adcq $0,%rcx
addq %r12,%r8
adcq %r13,%r9
adcq %r14,%r10
adcq %r15,%r11
adcq $0,%rcx
shld $1,%r11,%rcx
andq mask63(%rip),%r11
imul $19,%rcx,%rcx
addq %rcx,%r8
adcq $0,%r9
adcq $0,%r10
adcq $0,%r11
// double
addq %r8,%r8
adcq %r9,%r9
adcq %r10,%r10
adcq %r11,%r11
movq $0,%rdx
movq $38,%rcx
cmovae %rdx,%rcx
addq %rcx,%r8
adcq %rdx,%r9
adcq %rdx,%r10
adcq %rdx,%r11
cmovc %rcx,%rdx
addq %rdx,%r8
// copy
movq %r8,%r12
movq %r9,%r13
movq %r10,%r14
movq %r11,%r15
// add
addq 184(%rsp),%r12
adcq 192(%rsp),%r13
adcq 200(%rsp),%r14
adcq 208(%rsp),%r15
movq $0,%rdx
movq $38,%rax
cmovae %rdx,%rax
addq %rax,%r12
adcq %rdx,%r13
adcq %rdx,%r14
adcq %rdx,%r15
cmovc %rax,%rdx
addq %rdx,%r12
// sub
subq 184(%rsp),%r8
sbbq 192(%rsp),%r9
sbbq 200(%rsp),%r10
sbbq 208(%rsp),%r11
movq $0,%rdx
mov $38,%rax
cmovae %rdx,%rax
subq %rax,%r8
sbbq %rdx,%r9
sbbq %rdx,%r10
sbbq %rdx,%r11
cmovc %rax,%rdx
subq %rdx,%r8
// store
movq %r12,96(%rsp)
movq %r13,104(%rsp)
movq %r14,112(%rsp)
movq %r15,120(%rsp)
// store
movq %r8,160(%rsp)
movq %r9,168(%rsp)
movq %r10,176(%rsp)
movq %r11,184(%rsp)
/* p1p1 to p3 */
// convert to 9x4 form
vmovdqa 64(%rsp),%ymm8
vmovdqa 96(%rsp),%ymm9
vmovdqa 96(%rsp),%ymm10
vmovdqa 64(%rsp),%ymm11
vpunpcklqdq %ymm9,%ymm8,%ymm12
vpunpckhqdq %ymm9,%ymm8,%ymm13
vpunpcklqdq %ymm11,%ymm10,%ymm14
vpunpckhqdq %ymm11,%ymm10,%ymm15
vpermq $68,%ymm14,%ymm7
vpblendd $240,%ymm7,%ymm12,%ymm1
vpermq $68,%ymm15,%ymm7
vpblendd $240,%ymm7,%ymm13,%ymm2
vpermq $238,%ymm12,%ymm7
vpblendd $240,%ymm14,%ymm7,%ymm3
vpermq $238,%ymm13,%ymm7
vpblendd $240,%ymm15,%ymm7,%ymm4
vpand pmask1(%rip),%ymm1,%ymm10
vpand pmask2(%rip),%ymm1,%ymm11
vpsrlq $29,%ymm11,%ymm11
vpand pmask3(%rip),%ymm1,%ymm7
vpsrlq $58,%ymm7,%ymm7
vpand pmask4(%rip),%ymm2,%ymm9
vpsllq $6,%ymm9,%ymm9
vpor %ymm9,%ymm7,%ymm12
vpand pmask5(%rip),%ymm2,%ymm13
vpsrlq $23,%ymm13,%ymm13
vpand pmask6(%rip),%ymm2,%ymm7
vpsrlq $52,%ymm7,%ymm7
vpand pmask7(%rip),%ymm3,%ymm9
vpsllq $12,%ymm9,%ymm9
vpor %ymm9,%ymm7,%ymm5
vpand pmask8(%rip),%ymm3,%ymm6
vpsrlq $17,%ymm6,%ymm6
vpand pmask9(%rip),%ymm3,%ymm7
vpsrlq $46,%ymm7,%ymm7
vpand pmask10(%rip),%ymm4,%ymm9
vpsllq $18,%ymm9,%ymm9
vpor %ymm9,%ymm7,%ymm7
vpand pmask11(%rip),%ymm4,%ymm8
vpsrlq $11,%ymm8,%ymm8
vpand pmask12(%rip),%ymm4,%ymm9
vpsrlq $40,%ymm9,%ymm9
vmovdqa %ymm10,1248(%rsp)
vmovdqa %ymm11,1280(%rsp)
vmovdqa %ymm12,1312(%rsp)
vmovdqa %ymm13,1344(%rsp)
vmovdqa %ymm5,1376(%rsp)
vmovdqa %ymm6,1408(%rsp)
vmovdqa %ymm7,1440(%rsp)
vmovdqa %ymm8,1472(%rsp)
vmovdqa %ymm9,1504(%rsp)
// convert to 9x4 form
vmovdqa 160(%rsp),%ymm8
vmovdqa 128(%rsp),%ymm9
vmovdqa 160(%rsp),%ymm10
vmovdqa 128(%rsp),%ymm11
vpunpcklqdq %ymm9,%ymm8,%ymm5
vpunpckhqdq %ymm9,%ymm8,%ymm6
vpunpcklqdq %ymm11,%ymm10,%ymm7
vpunpckhqdq %ymm11,%ymm10,%ymm8
vpermq $68,%ymm7,%ymm9
vpblendd $240,%ymm9,%ymm5,%ymm3
vpermq $68,%ymm8,%ymm9
vpblendd $240,%ymm9,%ymm6,%ymm4
vpermq $238,%ymm5,%ymm9
vpblendd $240,%ymm7,%ymm9,%ymm5
vpermq $238,%ymm6,%ymm9
vpblendd $240,%ymm8,%ymm9,%ymm6
vpand pmask1(%rip),%ymm3,%ymm10
vpand pmask2(%rip),%ymm3,%ymm11
vpsrlq $29,%ymm11,%ymm11
vpand pmask3(%rip),%ymm3,%ymm7
vpsrlq $58,%ymm7,%ymm7
vpand pmask4(%rip),%ymm4,%ymm9
vpsllq $6,%ymm9,%ymm9
vpor %ymm9,%ymm7,%ymm12
vpand pmask5(%rip),%ymm4,%ymm13
vpsrlq $23,%ymm13,%ymm13
vpand pmask6(%rip),%ymm4,%ymm7
vpsrlq $52,%ymm7,%ymm7
vpand pmask7(%rip),%ymm5,%ymm9
vpsllq $12,%ymm9,%ymm9
vpor %ymm9,%ymm7,%ymm0
vpand pmask8(%rip),%ymm5,%ymm1
vpsrlq $17,%ymm1,%ymm1
vpand pmask9(%rip),%ymm5,%ymm7
vpsrlq $46,%ymm7,%ymm7
vpand pmask10(%rip),%ymm6,%ymm9
vpsllq $18,%ymm9,%ymm9
vpor %ymm9,%ymm7,%ymm2
vpand pmask11(%rip),%ymm6,%ymm3
vpsrlq $11,%ymm3,%ymm3
vpand pmask12(%rip),%ymm6,%ymm4
vpsrlq $40,%ymm4,%ymm4
vmovdqa 1376(%rsp),%ymm5
vmovdqa 1408(%rsp),%ymm6
vmovdqa 1440(%rsp),%ymm7
vmovdqa 1472(%rsp),%ymm8
vmovdqa 1504(%rsp),%ymm9
// mul4x1
vpmuludq %ymm5,%ymm0,%ymm15
vmovdqa %ymm15,480(%rsp)
vpmuludq %ymm6,%ymm0,%ymm15
vpmuludq %ymm5,%ymm1,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vmovdqa %ymm15,512(%rsp)
vpmuludq %ymm7,%ymm0,%ymm15
vpmuludq %ymm6,%ymm1,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vpmuludq %ymm5,%ymm2,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vmovdqa %ymm15,544(%rsp)
vpmuludq %ymm8,%ymm0,%ymm15
vpmuludq %ymm7,%ymm1,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vpmuludq %ymm6,%ymm2,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vpmuludq %ymm5,%ymm3,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vmovdqa %ymm15,576(%rsp)
vpmuludq %ymm9,%ymm0,%ymm15
vpmuludq %ymm8,%ymm1,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vpmuludq %ymm7,%ymm2,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vpmuludq %ymm6,%ymm3,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vpmuludq %ymm5,%ymm4,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vmovdqa %ymm15,608(%rsp)
vpmuludq %ymm9,%ymm1,%ymm15
vpmuludq %ymm8,%ymm2,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vpmuludq %ymm7,%ymm3,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vpmuludq %ymm6,%ymm4,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vmovdqa %ymm15,640(%rsp)
vpmuludq %ymm9,%ymm2,%ymm15
vpmuludq %ymm8,%ymm3,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vpmuludq %ymm7,%ymm4,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vmovdqa %ymm15,672(%rsp)
vpmuludq %ymm9,%ymm3,%ymm15
vpmuludq %ymm8,%ymm4,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vmovdqa %ymm15,704(%rsp)
vpmuludq %ymm9,%ymm4,%ymm15
vmovdqa %ymm15,736(%rsp)
vpaddq %ymm10,%ymm0,%ymm0
vpaddq %ymm11,%ymm1,%ymm1
vpaddq %ymm12,%ymm2,%ymm2
vpaddq %ymm13,%ymm3,%ymm3
vpaddq 1248(%rsp),%ymm5,%ymm5
vpaddq 1280(%rsp),%ymm6,%ymm6
vpaddq 1312(%rsp),%ymm7,%ymm7
vpaddq 1344(%rsp),%ymm8,%ymm8
vpmuludq 1248(%rsp),%ymm10,%ymm15
vmovdqa %ymm15,768(%rsp)
vpaddq 480(%rsp),%ymm15,%ymm15
vmovdqa %ymm15,992(%rsp)
vpmuludq 1280(%rsp),%ymm10,%ymm15
vpmuludq 1248(%rsp),%ymm11,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vmovdqa %ymm15,800(%rsp)
vpaddq 512(%rsp),%ymm15,%ymm15
vmovdqa %ymm15,1024(%rsp)
vpmuludq 1312(%rsp),%ymm10,%ymm15
vpmuludq 1280(%rsp),%ymm11,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vpmuludq 1248(%rsp),%ymm12,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vmovdqa %ymm15,832(%rsp)
vpaddq 544(%rsp),%ymm15,%ymm15
vmovdqa %ymm15,1056(%rsp)
vpmuludq 1344(%rsp),%ymm10,%ymm15
vpmuludq 1312(%rsp),%ymm11,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vpmuludq 1280(%rsp),%ymm12,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vpmuludq 1248(%rsp),%ymm13,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vmovdqa %ymm15,864(%rsp)
vpaddq 576(%rsp),%ymm15,%ymm15
vmovdqa %ymm15,1088(%rsp)
vpmuludq 1344(%rsp),%ymm11,%ymm15
vpmuludq 1312(%rsp),%ymm12,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vpmuludq 1280(%rsp),%ymm13,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vmovdqa %ymm15,896(%rsp)
vpaddq 608(%rsp),%ymm15,%ymm15
vmovdqa %ymm15,1120(%rsp)
vpmuludq 1344(%rsp),%ymm12,%ymm15
vpmuludq 1312(%rsp),%ymm13,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vmovdqa %ymm15,928(%rsp)
vpaddq 640(%rsp),%ymm15,%ymm15
vmovdqa %ymm15,1152(%rsp)
vpmuludq 1344(%rsp),%ymm13,%ymm15
vmovdqa %ymm15,960(%rsp)
vpaddq 672(%rsp),%ymm15,%ymm15
vmovdqa %ymm15,1184(%rsp)
vpmuludq %ymm5,%ymm0,%ymm15
vmovdqa %ymm15,1216(%rsp)
vpmuludq %ymm6,%ymm0,%ymm15
vpmuludq %ymm5,%ymm1,%ymm14
vpaddq %ymm14,%ymm15,%ymm10
vpmuludq %ymm7,%ymm0,%ymm15
vpmuludq %ymm6,%ymm1,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vpmuludq %ymm5,%ymm2,%ymm14
vpaddq %ymm14,%ymm15,%ymm11
vpmuludq %ymm8,%ymm0,%ymm15
vpmuludq %ymm7,%ymm1,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vpmuludq %ymm6,%ymm2,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vpmuludq %ymm5,%ymm3,%ymm14
vpaddq %ymm14,%ymm15,%ymm12
vpmuludq %ymm9,%ymm0,%ymm15
vpmuludq %ymm8,%ymm1,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vpmuludq %ymm7,%ymm2,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vpmuludq %ymm6,%ymm3,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vpmuludq %ymm5,%ymm4,%ymm14
vpaddq %ymm14,%ymm15,%ymm13
vpmuludq %ymm9,%ymm1,%ymm15
vpmuludq %ymm8,%ymm2,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vpmuludq %ymm7,%ymm3,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vpmuludq %ymm6,%ymm4,%ymm14
vpaddq %ymm14,%ymm15,%ymm0
vpmuludq %ymm9,%ymm2,%ymm15
vpmuludq %ymm8,%ymm3,%ymm14
vpaddq %ymm14,%ymm15,%ymm15
vpmuludq %ymm7,%ymm4,%ymm14
vpaddq %ymm14,%ymm15,%ymm1
vpmuludq %ymm9,%ymm3,%ymm15
vpmuludq %ymm8,%ymm4,%ymm14
vpaddq %ymm14,%ymm15,%ymm2
vpmuludq %ymm9,%ymm4,%ymm3
vmovdqa 1216(%rsp),%ymm9
vpsubq 992(%rsp),%ymm9,%ymm9
vpaddq 896(%rsp),%ymm9,%ymm9
vpsubq 1024(%rsp),%ymm10,%ymm10
vpaddq 928(%rsp),%ymm10,%ymm10
vpsubq 1056(%rsp),%ymm11,%ymm11
vpaddq 960(%rsp),%ymm11,%ymm11
vpsubq 1088(%rsp),%ymm12,%ymm12
vpsubq 1120(%rsp),%ymm13,%ymm13
vpaddq 480(%rsp),%ymm13,%ymm13
vpsubq 1152(%rsp),%ymm0,%ymm0
vpaddq 512(%rsp),%ymm0,%ymm0
vpsubq 1184(%rsp),%ymm1,%ymm1
vpaddq 544(%rsp),%ymm1,%ymm1
vpsubq 704(%rsp),%ymm2,%ymm2
vpaddq 576(%rsp),%ymm2,%ymm2
vpsubq 736(%rsp),%ymm3,%ymm3
vpaddq 608(%rsp),%ymm3,%ymm3
vpsrlq $29,%ymm0,%ymm14
vpaddq %ymm14,%ymm1,%ymm1
vpand vecmask29(%rip),%ymm0,%ymm0
vpmuludq vec1216(%rip),%ymm0,%ymm0
vpaddq 768(%rsp),%ymm0,%ymm0
vpsrlq $29,%ymm1,%ymm14
vpaddq %ymm14,%ymm2,%ymm2
vpand vecmask29(%rip),%ymm1,%ymm1
vpmuludq vec1216(%rip),%ymm1,%ymm1
vpaddq 800(%rsp),%ymm1,%ymm1
vpsrlq $29,%ymm2,%ymm14
vpaddq %ymm14,%ymm3,%ymm3
vpand vecmask29(%rip),%ymm2,%ymm2
vpmuludq vec1216(%rip),%ymm2,%ymm2
vpaddq 832(%rsp),%ymm2,%ymm2
vpsrlq $29,%ymm3,%ymm14
vpaddq 640(%rsp),%ymm14,%ymm14
vpand vecmask29(%rip),%ymm3,%ymm3
vpmuludq vec1216(%rip),%ymm3,%ymm3
vpaddq 864(%rsp),%ymm3,%ymm3
vpsrlq $29,%ymm14,%ymm15
vpaddq 672(%rsp),%ymm15,%ymm15
vpand vecmask29(%rip),%ymm14,%ymm4
vpmuludq vec1216(%rip),%ymm4,%ymm4
vpaddq %ymm9,%ymm4,%ymm4
vpsrlq $29,%ymm15,%ymm14
vpaddq 704(%rsp),%ymm14,%ymm14
vpand vecmask29(%rip),%ymm15,%ymm5
vpmuludq vec1216(%rip),%ymm5,%ymm5
vpaddq %ymm10,%ymm5,%ymm5
vpsrlq $29,%ymm14,%ymm15
vpaddq 736(%rsp),%ymm15,%ymm15
vpand vecmask29(%rip),%ymm14,%ymm6
vpmuludq vec1216(%rip),%ymm6,%ymm6
vpaddq %ymm11,%ymm6,%ymm6
vpsrlq $29,%ymm15,%ymm8
vpand vecmask29(%rip),%ymm15,%ymm7
vpmuludq vec1216(%rip),%ymm7,%ymm7
vpaddq %ymm12,%ymm7,%ymm7
vpmuludq vec1216(%rip),%ymm8,%ymm8
vpaddq %ymm13,%ymm8,%ymm8
vpsrlq $29,%ymm7,%ymm15
vpaddq %ymm15,%ymm8,%ymm8
vpand vecmask29(%rip),%ymm7,%ymm7
vpsrlq $23,%ymm8,%ymm15
vpaddq %ymm15,%ymm0,%ymm0
vpaddq %ymm15,%ymm15,%ymm15
vpaddq %ymm15,%ymm0,%ymm0
vpsllq $3,%ymm15,%ymm15
vpaddq %ymm15,%ymm0,%ymm0
vpand vecmask23(%rip),%ymm8,%ymm8
vpsrlq $29,%ymm0,%ymm15
vpaddq %ymm15,%ymm1,%ymm1
vpand vecmask29(%rip),%ymm0,%ymm0
vpsrlq $29,%ymm1,%ymm15
vpaddq %ymm15,%ymm2,%ymm2
vpand vecmask29(%rip),%ymm1,%ymm1
vpsrlq $29,%ymm2,%ymm15
vpaddq %ymm15,%ymm3,%ymm3
vpand vecmask29(%rip),%ymm2,%ymm2
vpsrlq $29,%ymm3,%ymm15
vpaddq %ymm15,%ymm4,%ymm4
vpand vecmask29(%rip),%ymm3,%ymm3
vpsrlq $29,%ymm4,%ymm15
vpaddq %ymm15,%ymm5,%ymm5
vpand vecmask29(%rip),%ymm4,%ymm4
vpsrlq $29,%ymm5,%ymm15
vpaddq %ymm15,%ymm6,%ymm6
vpand vecmask29(%rip),%ymm5,%ymm5
vpsrlq $29,%ymm6,%ymm15
vpaddq %ymm15,%ymm7,%ymm7
vpand vecmask29(%rip),%ymm6,%ymm6
vpsrlq $29,%ymm7,%ymm15
vpaddq %ymm15,%ymm8,%ymm8
vpand vecmask29(%rip),%ymm7,%ymm7
// get back to 4x4 form
vpand upmask1(%rip),%ymm0,%ymm10
vpand upmask1(%rip),%ymm1,%ymm11
vpsllq $29,%ymm11,%ymm11
vpor %ymm10,%ymm11,%ymm10
vpand upmask2(%rip),%ymm2,%ymm11
vpsllq $58,%ymm11,%ymm11
vpor %ymm10,%ymm11,%ymm10
vpand upmask6(%rip),%ymm2,%ymm11
vpsrlq $6,%ymm11,%ymm11
vpand upmask1(%rip),%ymm3,%ymm12
vpsllq $23,%ymm12,%ymm12
vpor %ymm11,%ymm12,%ymm11
vpand upmask3(%rip),%ymm4,%ymm12
vpsllq $52,%ymm12,%ymm12
vpor %ymm11,%ymm12,%ymm11
vpand upmask7(%rip),%ymm4,%ymm12
vpsrlq $12,%ymm12,%ymm12
vpand upmask1(%rip),%ymm5,%ymm13
vpsllq $17,%ymm13,%ymm13
vpor %ymm12,%ymm13,%ymm12
vpand upmask4(%rip),%ymm6,%ymm13
vpsllq $46,%ymm13,%ymm13
vpor %ymm12,%ymm13,%ymm12
vpand upmask8(%rip),%ymm6,%ymm13
vpsrlq $18,%ymm13,%ymm13
vpand upmask1(%rip),%ymm7,%ymm14
vpsllq $11,%ymm14,%ymm14
vpor %ymm13,%ymm14,%ymm13
vpand upmask5(%rip),%ymm8,%ymm14
vpsllq $40,%ymm14,%ymm14
vpor %ymm13,%ymm14,%ymm13
vpunpcklqdq %ymm11,%ymm10,%ymm2
vpunpckhqdq %ymm11,%ymm10,%ymm3
vpunpcklqdq %ymm13,%ymm12,%ymm4
vpunpckhqdq %ymm13,%ymm12,%ymm5
vpermq $68,%ymm4,%ymm7
vpblendd $240,%ymm7,%ymm2,%ymm10
vpermq $68,%ymm5,%ymm7
vpblendd $240,%ymm7,%ymm3,%ymm11
vpermq $238,%ymm2,%ymm7
vpblendd $240,%ymm4,%ymm7,%ymm12
vpermq $238,%ymm3,%ymm7
vpblendd $240,%ymm5,%ymm7,%ymm13
vmovdqa %ymm10,0(%rdi)
vmovdqa %ymm11,32(%rdi)
vmovdqa %ymm12,64(%rdi)
vmovdqa %ymm13,96(%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