-rw-r--r-- 22891 lib25519-20240321/crypto_multiscalar/ed25519/amd64-avx2-10l-maax-p3/ge25519_double.S raw
#include "crypto_asm_hidden.h"
#include "consts_namespace.h"
// ge25519_double
.p2align 5
ASM_HIDDEN _CRYPTO_NAMESPACE(ge25519_double)
.globl _CRYPTO_NAMESPACE(ge25519_double)
ASM_HIDDEN CRYPTO_NAMESPACE(ge25519_double)
.globl CRYPTO_NAMESPACE(ge25519_double)
_CRYPTO_NAMESPACE(ge25519_double):
CRYPTO_NAMESPACE(ge25519_double):
movq %rsp,%r11
andq $-32,%rsp
subq $1120,%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 %rdi,56(%rsp)
movq %rsi,%rdi
/* dbl p1p1 */
// square
xorq %r13,%r13
movq 0(%rdi),%rdx
mulx 8(%rdi),%r9,%r10
mulx 16(%rdi),%rcx,%r11
adcx %rcx,%r10
mulx 24(%rdi),%rcx,%r12
adcx %rcx,%r11
adcx %r13,%r12
movq 8(%rdi),%rdx
xorq %r14,%r14
mulx 16(%rdi),%rcx,%rdx
adcx %rcx,%r11
adox %rdx,%r12
movq 8(%rdi),%rdx
mulx 24(%rdi),%rcx,%rdx
adcx %rcx,%r12
adox %rdx,%r13
adcx %r14,%r13
xorq %r15,%r15
movq 16(%rdi),%rdx
mulx 24(%rdi),%rcx,%r14
adcx %rcx,%r13
adcx %r15,%r14
shld $1,%r14,%r15
shld $1,%r13,%r14
shld $1,%r12,%r13
shld $1,%r11,%r12
shld $1,%r10,%r11
shld $1,%r9,%r10
shlq $1,%r9
xorq %rdx,%rdx
movq 0(%rdi),%rdx
mulx %rdx,%r8,%rdx
adcx %rdx,%r9
movq 8(%rdi),%rdx
mulx %rdx,%rcx,%rdx
adcx %rcx,%r10
adcx %rdx,%r11
movq 16(%rdi),%rdx
mulx %rdx,%rcx,%rdx
adcx %rcx,%r12
adcx %rdx,%r13
movq 24(%rdi),%rdx
mulx %rdx,%rcx,%rdx
adcx %rcx,%r14
adcx %rdx,%r15
xorq %rbp,%rbp
movq $38,%rdx
mulx %r12,%rax,%r12
adcx %rax,%r8
adox %r12,%r9
mulx %r13,%rcx,%r13
adcx %rcx,%r9
adox %r13,%r10
mulx %r14,%rcx,%r14
adcx %rcx,%r10
adox %r14,%r11
mulx %r15,%rcx,%r15
adcx %rcx,%r11
adox %rbp,%r15
adcx %rbp,%r15
shld $1,%r11,%r15
andq mask63(%rip),%r11
imul $19,%r15,%r15
addq %r15,%r8
adcq $0,%r9
adcq $0,%r10
adcq $0,%r11
movq %r8,64(%rsp)
movq %r9,72(%rsp)
movq %r10,80(%rsp)
movq %r11,88(%rsp)
// square
xorq %r13,%r13
movq 32(%rdi),%rdx
mulx 40(%rdi),%r9,%r10
mulx 48(%rdi),%rcx,%r11
adcx %rcx,%r10
mulx 56(%rdi),%rcx,%r12
adcx %rcx,%r11
adcx %r13,%r12
movq 40(%rdi),%rdx
xorq %r14,%r14
mulx 48(%rdi),%rcx,%rdx
adcx %rcx,%r11
adox %rdx,%r12
movq 40(%rdi),%rdx
mulx 56(%rdi),%rcx,%rdx
adcx %rcx,%r12
adox %rdx,%r13
adcx %r14,%r13
xorq %r15,%r15
movq 48(%rdi),%rdx
mulx 56(%rdi),%rcx,%r14
adcx %rcx,%r13
adcx %r15,%r14
shld $1,%r14,%r15
shld $1,%r13,%r14
shld $1,%r12,%r13
shld $1,%r11,%r12
shld $1,%r10,%r11
shld $1,%r9,%r10
shlq $1,%r9
xorq %rdx,%rdx
movq 32(%rdi),%rdx
mulx %rdx,%r8,%rdx
adcx %rdx,%r9
movq 40(%rdi),%rdx
mulx %rdx,%rcx,%rdx
adcx %rcx,%r10
adcx %rdx,%r11
movq 48(%rdi),%rdx
mulx %rdx,%rcx,%rdx
adcx %rcx,%r12
adcx %rdx,%r13
movq 56(%rdi),%rdx
mulx %rdx,%rcx,%rdx
adcx %rcx,%r14
adcx %rdx,%r15
xorq %rbp,%rbp
movq $38,%rdx
mulx %r12,%rax,%r12
adcx %rax,%r8
adox %r12,%r9
mulx %r13,%rcx,%r13
adcx %rcx,%r9
adox %r13,%r10
mulx %r14,%rcx,%r14
adcx %rcx,%r10
adox %r14,%r11
mulx %r15,%rcx,%r15
adcx %rcx,%r11
adox %rbp,%r15
adcx %rbp,%r15
shld $1,%r11,%r15
andq mask63(%rip),%r11
imul $19,%r15,%r15
addq %r15,%r8
adcq $0,%r9
adcq $0,%r10
adcq $0,%r11
movq %r8,96(%rsp)
movq %r9,104(%rsp)
movq %r10,112(%rsp)
movq %r11,120(%rsp)
// square
xorq %r13,%r13
movq 64(%rdi),%rdx
mulx 72(%rdi),%r9,%r10
mulx 80(%rdi),%rcx,%r11
adcx %rcx,%r10
mulx 88(%rdi),%rcx,%r12
adcx %rcx,%r11
adcx %r13,%r12
movq 72(%rdi),%rdx
xorq %r14,%r14
mulx 80(%rdi),%rcx,%rdx
adcx %rcx,%r11
adox %rdx,%r12
movq 72(%rdi),%rdx
mulx 88(%rdi),%rcx,%rdx
adcx %rcx,%r12
adox %rdx,%r13
adcx %r14,%r13
xorq %r15,%r15
movq 80(%rdi),%rdx
mulx 88(%rdi),%rcx,%r14
adcx %rcx,%r13
adcx %r15,%r14
shld $1,%r14,%r15
shld $1,%r13,%r14
shld $1,%r12,%r13
shld $1,%r11,%r12
shld $1,%r10,%r11
shld $1,%r9,%r10
shlq $1,%r9
xorq %rdx,%rdx
movq 64(%rdi),%rdx
mulx %rdx,%r8,%rdx
adcx %rdx,%r9
movq 72(%rdi),%rdx
mulx %rdx,%rcx,%rdx
adcx %rcx,%r10
adcx %rdx,%r11
movq 80(%rdi),%rdx
mulx %rdx,%rcx,%rdx
adcx %rcx,%r12
adcx %rdx,%r13
movq 88(%rdi),%rdx
mulx %rdx,%rcx,%rdx
adcx %rcx,%r14
adcx %rdx,%r15
xorq %rbp,%rbp
movq $38,%rdx
mulx %r12,%rax,%r12
adcx %rax,%r8
adox %r12,%r9
mulx %r13,%rcx,%r13
adcx %rcx,%r9
adox %r13,%r10
mulx %r14,%rcx,%r14
adcx %rcx,%r10
adox %r14,%r11
mulx %r15,%rcx,%r15
adcx %rcx,%r11
adox %rbp,%r15
adcx %rbp,%r15
shld $1,%r11,%r15
andq mask63(%rip),%r11
imul $19,%r15,%r15
addq %r15,%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
movq %r8,128(%rsp)
movq %r9,136(%rsp)
movq %r10,144(%rsp)
movq %r11,152(%rsp)
// neg
movq $0,%r8
movq $0,%r9
movq $0,%r10
movq $0,%r11
subq 64(%rsp),%r8
sbbq 72(%rsp),%r9
sbbq 80(%rsp),%r10
sbbq 88(%rsp),%r11
movq $0,%rdx
movq $38,%rax
cmovae %rdx,%rax
subq %rax,%r8
sbbq %rdx,%r9
sbbq %rdx,%r10
sbbq %rdx,%r11
cmovc %rax,%rdx
subq %rdx,%r8
movq %r8,64(%rsp)
movq %r9,72(%rsp)
movq %r10,80(%rsp)
movq %r11,88(%rsp)
// copy
movq %r8,%r12
movq %r9,%r13
movq %r10,%r14
movq %r11,%r15
// sub
subq 96(%rsp),%r8
sbbq 104(%rsp),%r9
sbbq 112(%rsp),%r10
sbbq 120(%rsp),%r11
movq $0,%rdx
movq $38,%rax
cmovae %rdx,%rax
subq %rax,%r8
sbbq %rdx,%r9
sbbq %rdx,%r10
sbbq %rdx,%r11
cmovc %rax,%rdx
subq %rdx,%r8
movq %r8,224(%rsp)
movq %r9,232(%rsp)
movq %r10,240(%rsp)
movq %r11,248(%rsp)
// add
addq 96(%rsp),%r12
adcq 104(%rsp),%r13
adcq 112(%rsp),%r14
adcq 120(%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
subq %rdx,%r12
movq %r12,192(%rsp)
movq %r13,200(%rsp)
movq %r14,208(%rsp)
movq %r15,216(%rsp)
// sub
subq 128(%rsp),%r12
sbbq 136(%rsp),%r13
sbbq 144(%rsp),%r14
sbbq 152(%rsp),%r15
movq $0,%rdx
movq $38,%rax
cmovae %rdx,%rax
subq %rax,%r12
sbbq %rdx,%r13
sbbq %rdx,%r14
sbbq %rdx,%r15
cmovc %rax,%rdx
subq %rdx,%r12
movq %r12,256(%rsp)
movq %r13,264(%rsp)
movq %r14,272(%rsp)
movq %r15,280(%rsp)
// add
movq 0(%rdi),%r8
movq 8(%rdi),%r9
movq 16(%rdi),%r10
movq 24(%rdi),%r11
addq 32(%rdi),%r8
adcq 40(%rdi),%r9
adcq 48(%rdi),%r10
adcq 56(%rdi),%r11
movq $0,%rdx
movq $38,%rax
cmovae %rdx,%rax
addq %rax,%r8
adcq %rdx,%r9
adcq %rdx,%r10
adcq %rdx,%r11
cmovc %rax,%rdx
addq %rdx,%r8
movq %r8,160(%rsp)
movq %r9,168(%rsp)
movq %r10,176(%rsp)
movq %r11,184(%rsp)
// square
xorq %r13,%r13
movq 160(%rsp),%rdx
mulx 168(%rsp),%r9,%r10
mulx 176(%rsp),%rcx,%r11
adcx %rcx,%r10
mulx 184(%rsp),%rcx,%r12
adcx %rcx,%r11
adcx %r13,%r12
movq 168(%rsp),%rdx
xorq %r14,%r14
mulx 176(%rsp),%rcx,%rdx
adcx %rcx,%r11
adox %rdx,%r12
movq 168(%rsp),%rdx
mulx 184(%rsp),%rcx,%rdx
adcx %rcx,%r12
adox %rdx,%r13
adcx %r14,%r13
xorq %r15,%r15
movq 176(%rsp),%rdx
mulx 184(%rsp),%rcx,%r14
adcx %rcx,%r13
adcx %r15,%r14
shld $1,%r14,%r15
shld $1,%r13,%r14
shld $1,%r12,%r13
shld $1,%r11,%r12
shld $1,%r10,%r11
shld $1,%r9,%r10
shlq $1,%r9
xorq %rdx,%rdx
movq 160(%rsp),%rdx
mulx %rdx,%r8,%rdx
adcx %rdx,%r9
movq 168(%rsp),%rdx
mulx %rdx,%rcx,%rdx
adcx %rcx,%r10
adcx %rdx,%r11
movq 176(%rsp),%rdx
mulx %rdx,%rcx,%rdx
adcx %rcx,%r12
adcx %rdx,%r13
movq 184(%rsp),%rdx
mulx %rdx,%rcx,%rdx
adcx %rcx,%r14
adcx %rdx,%r15
xorq %rbp,%rbp
movq $38,%rdx
mulx %r12,%rax,%r12
adcx %rax,%r8
adox %r12,%r9
mulx %r13,%rcx,%r13
adcx %rcx,%r9
adox %r13,%r10
mulx %r14,%rcx,%r14
adcx %rcx,%r10
adox %r14,%r11
mulx %r15,%rcx,%r15
adcx %rcx,%r11
adox %rbp,%r15
adcx %rbp,%r15
shld $1,%r11,%r15
andq mask63(%rip),%r11
imul $19,%r15,%r15
addq %r15,%r8
adcq $0,%r9
adcq $0,%r10
adcq $0,%r11
// add
addq 64(%rsp),%r8
adcq 72(%rsp),%r9
adcq 80(%rsp),%r10
adcq 88(%rsp),%r11
movq $0,%rdx
movq $38,%rax
cmovae %rdx,%rax
addq %rax,%r8
adcq %rdx,%r9
adcq %rdx,%r10
adcq %rdx,%r11
cmovc %rax,%rdx
addq %rdx,%r8
// sub
subq 96(%rsp),%r8
sbbq 104(%rsp),%r9
sbbq 112(%rsp),%r10
sbbq 120(%rsp),%r11
movq $0,%rdx
movq $38,%rax
cmovae %rdx,%rax
subq %rax,%r8
sbbq %rdx,%r9
sbbq %rdx,%r10
sbbq %rdx,%r11
cmovc %rax,%rdx
subq %rdx,%r8
movq %r8,160(%rsp)
movq %r9,168(%rsp)
movq %r10,176(%rsp)
movq %r11,184(%rsp)
/* p1p1 to p3 */
movq 56(%rsp),%rdi
// convert to 10x4 form
vmovdqa 160(%rsp),%ymm8
vmovdqa 192(%rsp),%ymm9
vmovdqa 192(%rsp),%ymm10
vmovdqa 160(%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,%ymm3
vpermq $238,%ymm12,%ymm7
vpblendd $240,%ymm14,%ymm7,%ymm5
vpermq $238,%ymm13,%ymm7
vpblendd $240,%ymm15,%ymm7,%ymm7
vpand pmask1(%rip),%ymm1,%ymm0
vpand pmask2(%rip),%ymm1,%ymm11
vpsrlq $26,%ymm11,%ymm11
vpand pmask3(%rip),%ymm1,%ymm2
vpsrlq $51,%ymm2,%ymm2
vpand pmask4(%rip),%ymm3,%ymm9
vpsllq $13,%ymm9,%ymm9
vpor %ymm9,%ymm2,%ymm2
vpand pmask5(%rip),%ymm3,%ymm12
vpsrlq $13,%ymm12,%ymm12
vpand pmask6(%rip),%ymm3,%ymm4
vpsrlq $38,%ymm4,%ymm4
vpand pmask7(%rip),%ymm5,%ymm13
vpand pmask8(%rip),%ymm5,%ymm6
vpsrlq $25,%ymm6,%ymm6
vpand pmask9(%rip),%ymm5,%ymm14
vpsrlq $51,%ymm14,%ymm14
vpand pmask10(%rip),%ymm7,%ymm9
vpsllq $13,%ymm9,%ymm9
vpor %ymm9,%ymm14,%ymm14
vpand pmask11(%rip),%ymm7,%ymm8
vpsrlq $12,%ymm8,%ymm8
vpand pmask12(%rip),%ymm7,%ymm15
vpsrlq $38,%ymm15,%ymm15
vmovdqa %ymm0,480(%rsp)
vmovdqa %ymm11,512(%rsp)
vmovdqa %ymm2,544(%rsp)
vmovdqa %ymm12,576(%rsp)
vmovdqa %ymm4,608(%rsp)
vmovdqa %ymm13,640(%rsp)
vmovdqa %ymm6,672(%rsp)
vmovdqa %ymm14,704(%rsp)
vmovdqa %ymm8,736(%rsp)
vmovdqa %ymm15,768(%rsp)
// convert to 10x4 form
vmovdqa 256(%rsp),%ymm8
vmovdqa 224(%rsp),%ymm9
vmovdqa 256(%rsp),%ymm10
vmovdqa 224(%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,%ymm10
vpermq $68,%ymm15,%ymm7
vpblendd $240,%ymm7,%ymm13,%ymm11
vpermq $238,%ymm12,%ymm7
vpblendd $240,%ymm14,%ymm7,%ymm12
vpermq $238,%ymm13,%ymm7
vpblendd $240,%ymm15,%ymm7,%ymm13
vpand pmask1(%rip),%ymm10,%ymm0
vpand pmask2(%rip),%ymm10,%ymm1
vpsrlq $26,%ymm1,%ymm1
vpand pmask3(%rip),%ymm10,%ymm2
vpsrlq $51,%ymm2,%ymm2
vpand pmask4(%rip),%ymm11,%ymm3
vpsllq $13,%ymm3,%ymm3
vpor %ymm3,%ymm2,%ymm2
vpand pmask5(%rip),%ymm11,%ymm3
vpsrlq $13,%ymm3,%ymm3
vpand pmask6(%rip),%ymm11,%ymm4
vpsrlq $38,%ymm4,%ymm4
vpand pmask7(%rip),%ymm12,%ymm5
vpand pmask8(%rip),%ymm12,%ymm6
vpsrlq $25,%ymm6,%ymm6
vpand pmask9(%rip),%ymm12,%ymm7
vpsrlq $51,%ymm7,%ymm7
vpand pmask10(%rip),%ymm13,%ymm8
vpsllq $13,%ymm8,%ymm8
vpor %ymm8,%ymm7,%ymm7
vpand pmask11(%rip),%ymm13,%ymm8
vpsrlq $12,%ymm8,%ymm8
vpand pmask12(%rip),%ymm13,%ymm9
vpsrlq $38,%ymm9,%ymm9
vmovdqa %ymm0,800(%rsp)
vmovdqa %ymm1,832(%rsp)
vmovdqa %ymm2,864(%rsp)
vmovdqa %ymm3,896(%rsp)
vmovdqa %ymm4,928(%rsp)
vmovdqa %ymm5,960(%rsp)
vmovdqa %ymm6,992(%rsp)
vmovdqa %ymm7,1024(%rsp)
vmovdqa %ymm8,1056(%rsp)
vmovdqa %ymm9,1088(%rsp)
// mul4x1
vmovdqa 512(%rsp),%ymm11
vmovdqa 576(%rsp),%ymm12
vmovdqa 640(%rsp),%ymm13
vmovdqa 704(%rsp),%ymm14
vmovdqa 768(%rsp),%ymm15
vpaddq %ymm11,%ymm11,%ymm11
vpaddq %ymm12,%ymm12,%ymm12
vpaddq %ymm13,%ymm13,%ymm13
vpaddq %ymm14,%ymm14,%ymm14
vpaddq %ymm15,%ymm15,%ymm15
vpmuludq vec19(%rip),%ymm1,%ymm1
vpmuludq vec19(%rip),%ymm2,%ymm2
vpmuludq vec19(%rip),%ymm3,%ymm3
vpmuludq vec19(%rip),%ymm4,%ymm4
vpmuludq vec19(%rip),%ymm5,%ymm5
vpmuludq vec19(%rip),%ymm6,%ymm6
vpmuludq vec19(%rip),%ymm7,%ymm7
vpmuludq vec19(%rip),%ymm8,%ymm8
vpmuludq vec19(%rip),%ymm9,%ymm9
vpmuludq %ymm15,%ymm1,%ymm0
vpmuludq %ymm14,%ymm3,%ymm10
vpaddq %ymm10,%ymm0,%ymm0
vpmuludq %ymm13,%ymm5,%ymm10
vpaddq %ymm10,%ymm0,%ymm0
vpmuludq %ymm12,%ymm7,%ymm10
vpaddq %ymm10,%ymm0,%ymm0
vpmuludq %ymm11,%ymm9,%ymm10
vpaddq %ymm10,%ymm0,%ymm0
vpmuludq 736(%rsp),%ymm2,%ymm10
vpaddq %ymm10,%ymm0,%ymm0
vpmuludq 672(%rsp),%ymm4,%ymm10
vpaddq %ymm10,%ymm0,%ymm0
vpmuludq 608(%rsp),%ymm6,%ymm10
vpaddq %ymm10,%ymm0,%ymm0
vpmuludq 544(%rsp),%ymm8,%ymm10
vpaddq %ymm10,%ymm0,%ymm0
vpmuludq 768(%rsp),%ymm2,%ymm1
vpmuludq 736(%rsp),%ymm3,%ymm10
vpaddq %ymm10,%ymm1,%ymm1
vpmuludq 704(%rsp),%ymm4,%ymm10
vpaddq %ymm10,%ymm1,%ymm1
vpmuludq 672(%rsp),%ymm5,%ymm10
vpaddq %ymm10,%ymm1,%ymm1
vpmuludq 640(%rsp),%ymm6,%ymm10
vpaddq %ymm10,%ymm1,%ymm1
vpmuludq 608(%rsp),%ymm7,%ymm10
vpaddq %ymm10,%ymm1,%ymm1
vpmuludq 576(%rsp),%ymm8,%ymm10
vpaddq %ymm10,%ymm1,%ymm1
vpmuludq 544(%rsp),%ymm9,%ymm10
vpaddq %ymm10,%ymm1,%ymm1
vpmuludq %ymm15,%ymm3,%ymm2
vpmuludq %ymm14,%ymm5,%ymm10
vpaddq %ymm10,%ymm2,%ymm2
vpmuludq %ymm13,%ymm7,%ymm10
vpaddq %ymm10,%ymm2,%ymm2
vpmuludq %ymm12,%ymm9,%ymm10
vpaddq %ymm10,%ymm2,%ymm2
vpmuludq 736(%rsp),%ymm4,%ymm10
vpaddq %ymm10,%ymm2,%ymm2
vpmuludq 672(%rsp),%ymm6,%ymm10
vpaddq %ymm10,%ymm2,%ymm2
vpmuludq 608(%rsp),%ymm8,%ymm10
vpaddq %ymm10,%ymm2,%ymm2
vpmuludq 832(%rsp),%ymm11,%ymm10
vpaddq %ymm10,%ymm2,%ymm2
vpmuludq 768(%rsp),%ymm4,%ymm3
vpmuludq 736(%rsp),%ymm5,%ymm10
vpaddq %ymm10,%ymm3,%ymm3
vpmuludq 704(%rsp),%ymm6,%ymm10
vpaddq %ymm10,%ymm3,%ymm3
vpmuludq 672(%rsp),%ymm7,%ymm10
vpaddq %ymm10,%ymm3,%ymm3
vpmuludq 640(%rsp),%ymm8,%ymm10
vpaddq %ymm10,%ymm3,%ymm3
vpmuludq 608(%rsp),%ymm9,%ymm10
vpaddq %ymm10,%ymm3,%ymm3
vpmuludq %ymm15,%ymm5,%ymm4
vpmuludq %ymm14,%ymm7,%ymm10
vpaddq %ymm10,%ymm4,%ymm4
vpmuludq %ymm13,%ymm9,%ymm10
vpaddq %ymm10,%ymm4,%ymm4
vpmuludq 736(%rsp),%ymm6,%ymm10
vpaddq %ymm10,%ymm4,%ymm4
vpmuludq 672(%rsp),%ymm8,%ymm10
vpaddq %ymm10,%ymm4,%ymm4
vpmuludq 832(%rsp),%ymm12,%ymm10
vpaddq %ymm10,%ymm4,%ymm4
vpmuludq 896(%rsp),%ymm11,%ymm10
vpaddq %ymm10,%ymm4,%ymm4
vpmuludq 768(%rsp),%ymm6,%ymm5
vpmuludq 736(%rsp),%ymm7,%ymm10
vpaddq %ymm10,%ymm5,%ymm5
vpmuludq 704(%rsp),%ymm8,%ymm10
vpaddq %ymm10,%ymm5,%ymm5
vpmuludq 672(%rsp),%ymm9,%ymm10
vpaddq %ymm10,%ymm5,%ymm5
vpmuludq %ymm15,%ymm7,%ymm6
vpmuludq %ymm14,%ymm9,%ymm10
vpaddq %ymm10,%ymm6,%ymm6
vpmuludq 736(%rsp),%ymm8,%ymm10
vpaddq %ymm10,%ymm6,%ymm6
vpmuludq 832(%rsp),%ymm13,%ymm10
vpaddq %ymm10,%ymm6,%ymm6
vpmuludq 896(%rsp),%ymm12,%ymm10
vpaddq %ymm10,%ymm6,%ymm6
vpmuludq 960(%rsp),%ymm11,%ymm10
vpaddq %ymm10,%ymm6,%ymm6
vpmuludq 768(%rsp),%ymm8,%ymm7
vpmuludq 736(%rsp),%ymm9,%ymm10
vpaddq %ymm10,%ymm7,%ymm7
vpmuludq %ymm15,%ymm9,%ymm8
vpmuludq 832(%rsp),%ymm14,%ymm10
vpaddq %ymm10,%ymm8,%ymm8
vpmuludq 896(%rsp),%ymm13,%ymm10
vpaddq %ymm10,%ymm8,%ymm8
vpmuludq 960(%rsp),%ymm12,%ymm10
vpaddq %ymm10,%ymm8,%ymm8
vpmuludq 1024(%rsp),%ymm11,%ymm10
vpaddq %ymm10,%ymm8,%ymm8
vmovdqa 480(%rsp),%ymm11
vmovdqa 512(%rsp),%ymm12
vmovdqa 544(%rsp),%ymm13
vmovdqa 576(%rsp),%ymm14
vmovdqa 608(%rsp),%ymm15
vmovdqa 640(%rsp),%ymm9
vpmuludq 800(%rsp),%ymm11,%ymm10
vpaddq %ymm10,%ymm0,%ymm0
vpmuludq 864(%rsp),%ymm11,%ymm10
vpaddq %ymm10,%ymm2,%ymm2
vpmuludq 800(%rsp),%ymm13,%ymm10
vpaddq %ymm10,%ymm2,%ymm2
vpmuludq 896(%rsp),%ymm11,%ymm10
vpaddq %ymm10,%ymm3,%ymm3
vpmuludq 864(%rsp),%ymm12,%ymm10
vpaddq %ymm10,%ymm3,%ymm3
vpmuludq 832(%rsp),%ymm13,%ymm10
vpaddq %ymm10,%ymm3,%ymm3
vpmuludq 800(%rsp),%ymm14,%ymm10
vpaddq %ymm10,%ymm3,%ymm3
vpmuludq 832(%rsp),%ymm11,%ymm10
vpaddq %ymm10,%ymm1,%ymm1
vpmuludq 800(%rsp),%ymm12,%ymm10
vpaddq %ymm10,%ymm1,%ymm1
vpmuludq 928(%rsp),%ymm11,%ymm10
vpaddq %ymm10,%ymm4,%ymm4
vpmuludq 864(%rsp),%ymm13,%ymm10
vpaddq %ymm10,%ymm4,%ymm4
vpmuludq 800(%rsp),%ymm15,%ymm10
vpaddq %ymm10,%ymm4,%ymm4
vpmuludq 960(%rsp),%ymm11,%ymm10
vpaddq %ymm10,%ymm5,%ymm5
vpmuludq 928(%rsp),%ymm12,%ymm10
vpaddq %ymm10,%ymm5,%ymm5
vpmuludq 896(%rsp),%ymm13,%ymm10
vpaddq %ymm10,%ymm5,%ymm5
vpmuludq 864(%rsp),%ymm14,%ymm10
vpaddq %ymm10,%ymm5,%ymm5
vpmuludq 832(%rsp),%ymm15,%ymm10
vpaddq %ymm10,%ymm5,%ymm5
vpmuludq 800(%rsp),%ymm9,%ymm10
vpaddq %ymm10,%ymm5,%ymm5
vpmuludq 992(%rsp),%ymm11,%ymm10
vpaddq %ymm10,%ymm6,%ymm6
vpmuludq 928(%rsp),%ymm13,%ymm10
vpaddq %ymm10,%ymm6,%ymm6
vpmuludq 864(%rsp),%ymm15,%ymm10
vpaddq %ymm10,%ymm6,%ymm6
vpmuludq 1024(%rsp),%ymm11,%ymm10
vpaddq %ymm10,%ymm7,%ymm7
vpmuludq 992(%rsp),%ymm12,%ymm10
vpaddq %ymm10,%ymm7,%ymm7
vpmuludq 960(%rsp),%ymm13,%ymm10
vpaddq %ymm10,%ymm7,%ymm7
vpmuludq 928(%rsp),%ymm14,%ymm10
vpaddq %ymm10,%ymm7,%ymm7
vpmuludq 896(%rsp),%ymm15,%ymm10
vpaddq %ymm10,%ymm7,%ymm7
vpmuludq 864(%rsp),%ymm9,%ymm10
vpaddq %ymm10,%ymm7,%ymm7
vpmuludq 1056(%rsp),%ymm11,%ymm10
vpaddq %ymm10,%ymm8,%ymm8
vpmuludq 992(%rsp),%ymm13,%ymm10
vpaddq %ymm10,%ymm8,%ymm8
vpmuludq 928(%rsp),%ymm15,%ymm10
vpaddq %ymm10,%ymm8,%ymm8
vpmuludq 928(%rsp),%ymm9,%ymm9
vpmuludq 1088(%rsp),%ymm11,%ymm10
vpaddq %ymm10,%ymm9,%ymm9
vpmuludq 1056(%rsp),%ymm12,%ymm10
vpaddq %ymm10,%ymm9,%ymm9
vpmuludq 1024(%rsp),%ymm13,%ymm10
vpaddq %ymm10,%ymm9,%ymm9
vpmuludq 992(%rsp),%ymm14,%ymm10
vpaddq %ymm10,%ymm9,%ymm9
vpmuludq 960(%rsp),%ymm15,%ymm10
vpaddq %ymm10,%ymm9,%ymm9
vmovdqa 672(%rsp),%ymm11
vmovdqa 704(%rsp),%ymm12
vmovdqa 736(%rsp),%ymm13
vmovdqa 768(%rsp),%ymm14
vpmuludq 800(%rsp),%ymm11,%ymm10
vpaddq %ymm10,%ymm6,%ymm6
vpmuludq 832(%rsp),%ymm11,%ymm10
vpaddq %ymm10,%ymm7,%ymm7
vpmuludq 800(%rsp),%ymm12,%ymm10
vpaddq %ymm10,%ymm7,%ymm7
vpmuludq 864(%rsp),%ymm11,%ymm10
vpaddq %ymm10,%ymm8,%ymm8
vpmuludq 800(%rsp),%ymm13,%ymm10
vpaddq %ymm10,%ymm8,%ymm8
vpmuludq 896(%rsp),%ymm11,%ymm10
vpaddq %ymm10,%ymm9,%ymm9
vpmuludq 864(%rsp),%ymm12,%ymm10
vpaddq %ymm10,%ymm9,%ymm9
vpmuludq 832(%rsp),%ymm13,%ymm10
vpaddq %ymm10,%ymm9,%ymm9
vpmuludq 800(%rsp),%ymm14,%ymm10
vpaddq %ymm10,%ymm9,%ymm9
vpsrlq $26,%ymm8,%ymm10
vpaddq %ymm10,%ymm9,%ymm9
vpand vecmask26(%rip),%ymm8,%ymm8
vpsrlq $25,%ymm9,%ymm10
vpmuludq vec19(%rip),%ymm10,%ymm10
vpaddq %ymm10,%ymm0,%ymm0
vpand vecmask25(%rip),%ymm9,%ymm9
vpsrlq $26,%ymm0,%ymm10
vpaddq %ymm10,%ymm1,%ymm1
vpand vecmask26(%rip),%ymm0,%ymm0
vpsrlq $25,%ymm1,%ymm10
vpaddq %ymm10,%ymm2,%ymm2
vpand vecmask25(%rip),%ymm1,%ymm1
vpsrlq $26,%ymm2,%ymm10
vpaddq %ymm10,%ymm3,%ymm3
vpand vecmask26(%rip),%ymm2,%ymm2
vpsrlq $25,%ymm3,%ymm10
vpaddq %ymm10,%ymm4,%ymm4
vpand vecmask25(%rip),%ymm3,%ymm3
vpsrlq $26,%ymm4,%ymm10
vpaddq %ymm10,%ymm5,%ymm5
vpand vecmask26(%rip),%ymm4,%ymm4
vpsrlq $25,%ymm5,%ymm10
vpaddq %ymm10,%ymm6,%ymm6
vpand vecmask25(%rip),%ymm5,%ymm5
vpsrlq $26,%ymm6,%ymm10
vpaddq %ymm10,%ymm7,%ymm7
vpand vecmask26(%rip),%ymm6,%ymm6
vpsrlq $25,%ymm7,%ymm10
vpaddq %ymm10,%ymm8,%ymm8
vpand vecmask25(%rip),%ymm7,%ymm7
vpsrlq $26,%ymm8,%ymm10
vpaddq %ymm10,%ymm9,%ymm9
vpand vecmask26(%rip),%ymm8,%ymm8
// get back to 4x4 form
vpand upmask1(%rip),%ymm0,%ymm10
vpand upmask2(%rip),%ymm1,%ymm11
vpsllq $26,%ymm11,%ymm11
vpor %ymm10,%ymm11,%ymm10
vpand upmask3(%rip),%ymm2,%ymm11
vpsllq $51,%ymm11,%ymm11
vpor %ymm10,%ymm11,%ymm10
vpand upmask4(%rip),%ymm2,%ymm11
vpsrlq $13,%ymm11,%ymm11
vpand upmask2(%rip),%ymm3,%ymm12
vpsllq $13,%ymm12,%ymm12
vpor %ymm11,%ymm12,%ymm11
vpand upmask1(%rip),%ymm4,%ymm12
vpsllq $38,%ymm12,%ymm12
vpor %ymm11,%ymm12,%ymm11
vpand upmask2(%rip),%ymm5,%ymm12
vpand upmask1(%rip),%ymm6,%ymm13
vpsllq $25,%ymm13,%ymm13
vpor %ymm12,%ymm13,%ymm12
vpand upmask3(%rip),%ymm7,%ymm13
vpsllq $51,%ymm13,%ymm13
vpor %ymm12,%ymm13,%ymm12
vpand upmask5(%rip),%ymm7,%ymm13
vpsrlq $13,%ymm13,%ymm13
vpand upmask1(%rip),%ymm8,%ymm14
vpsllq $12,%ymm14,%ymm14
vpor %ymm13,%ymm14,%ymm13
vpand upmask1(%rip),%ymm9,%ymm14
vpsllq $38,%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