-rw-r--r-- 33373 lib25519-20220726/crypto_sign/ed25519/amd64-avx2-9l-mxaa/ge25519_base.S raw
/* Assembly for fixed base scalar multiplication. * * This assembly has been developed after studying the * amd64-64-24k implementation of the work "High speed * high security signatures" by Bernstein et al. */ #include "consts_namespace.h" .p2align 5 .globl _CRYPTO_NAMESPACE(base) .globl CRYPTO_NAMESPACE(base) _CRYPTO_NAMESPACE(base): CRYPTO_NAMESPACE(base): 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 %rdi,56(%rsp) movq %rsi,64(%rsp) movq %rdx,72(%rsp) /* choose t and initialize r */ movq %rdx,%rcx movzbl 0(%rsi),%eax movsbq %al,%rdx movq $0,%rsi imulq $768,%rsi,%rdi movq %rdx,%rsi sarq $7,%rsi movq %rdx,%r8 addq %rsi,%r8 xorq %rsi,%r8 movq $1,%rsi movq $0,%r9 movq $0,%rax movq $0,%r10 movq $1,%r11 movq $0,%r12 movq $0,%r13 movq $0,%r14 cmpq $1,%r8 movq 0(%rcx,%rdi),%r15 cmove %r15,%rsi movq 8(%rcx,%rdi),%r15 cmove %r15,%r9 movq 16(%rcx,%rdi),%r15 cmove %r15,%rax movq 24(%rcx,%rdi),%r15 cmove %r15,%r10 movq 32(%rcx,%rdi),%r15 cmove %r15,%r11 movq 40(%rcx,%rdi),%r15 cmove %r15,%r12 movq 48(%rcx,%rdi),%r15 cmove %r15,%r13 movq 56(%rcx,%rdi),%r15 cmove %r15,%r14 cmpq $2,%r8 movq 96(%rcx,%rdi),%r15 cmove %r15,%rsi movq 104(%rcx,%rdi),%r15 cmove %r15,%r9 movq 112(%rcx,%rdi),%r15 cmove %r15,%rax movq 120(%rcx,%rdi),%r15 cmove %r15,%r10 movq 128(%rcx,%rdi),%r15 cmove %r15,%r11 movq 136(%rcx,%rdi),%r15 cmove %r15,%r12 movq 144(%rcx,%rdi),%r15 cmove %r15,%r13 movq 152(%rcx,%rdi),%r15 cmove %r15,%r14 cmpq $3,%r8 movq 192(%rcx,%rdi),%r15 cmove %r15,%rsi movq 200(%rcx,%rdi),%r15 cmove %r15,%r9 movq 208(%rcx,%rdi),%r15 cmove %r15,%rax movq 216(%rcx,%rdi),%r15 cmove %r15,%r10 movq 224(%rcx,%rdi),%r15 cmove %r15,%r11 movq 232(%rcx,%rdi),%r15 cmove %r15,%r12 movq 240(%rcx,%rdi),%r15 cmove %r15,%r13 movq 248(%rcx,%rdi),%r15 cmove %r15,%r14 cmpq $4,%r8 movq 288(%rcx,%rdi),%r15 cmove %r15,%rsi movq 296(%rcx,%rdi),%r15 cmove %r15,%r9 movq 304(%rcx,%rdi),%r15 cmove %r15,%rax movq 312(%rcx,%rdi),%r15 cmove %r15,%r10 movq 320(%rcx,%rdi),%r15 cmove %r15,%r11 movq 328(%rcx,%rdi),%r15 cmove %r15,%r12 movq 336(%rcx,%rdi),%r15 cmove %r15,%r13 movq 344(%rcx,%rdi),%r15 cmove %r15,%r14 cmpq $5,%r8 movq 384(%rcx,%rdi),%r15 cmove %r15,%rsi movq 392(%rcx,%rdi),%r15 cmove %r15,%r9 movq 400(%rcx,%rdi),%r15 cmove %r15,%rax movq 408(%rcx,%rdi),%r15 cmove %r15,%r10 movq 416(%rcx,%rdi),%r15 cmove %r15,%r11 movq 424(%rcx,%rdi),%r15 cmove %r15,%r12 movq 432(%rcx,%rdi),%r15 cmove %r15,%r13 movq 440(%rcx,%rdi),%r15 cmove %r15,%r14 cmpq $6,%r8 movq 480(%rcx,%rdi),%r15 cmove %r15,%rsi movq 488(%rcx,%rdi),%r15 cmove %r15,%r9 movq 496(%rcx,%rdi),%r15 cmove %r15,%rax movq 504(%rcx,%rdi),%r15 cmove %r15,%r10 movq 512(%rcx,%rdi),%r15 cmove %r15,%r11 movq 520(%rcx,%rdi),%r15 cmove %r15,%r12 movq 528(%rcx,%rdi),%r15 cmove %r15,%r13 movq 536(%rcx,%rdi),%r15 cmove %r15,%r14 cmpq $7,%r8 movq 576(%rcx,%rdi),%r15 cmove %r15,%rsi movq 584(%rcx,%rdi),%r15 cmove %r15,%r9 movq 592(%rcx,%rdi),%r15 cmove %r15,%rax movq 600(%rcx,%rdi),%r15 cmove %r15,%r10 movq 608(%rcx,%rdi),%r15 cmove %r15,%r11 movq 616(%rcx,%rdi),%r15 cmove %r15,%r12 movq 624(%rcx,%rdi),%r15 cmove %r15,%r13 movq 632(%rcx,%rdi),%r15 cmove %r15,%r14 cmpq $8,%r8 movq 672(%rcx,%rdi),%r15 cmove %r15,%rsi movq 680(%rcx,%rdi),%r15 cmove %r15,%r9 movq 688(%rcx,%rdi),%r15 cmove %r15,%rax movq 696(%rcx,%rdi),%r15 cmove %r15,%r10 movq 704(%rcx,%rdi),%r15 cmove %r15,%r11 movq 712(%rcx,%rdi),%r15 cmove %r15,%r12 movq 720(%rcx,%rdi),%r15 cmove %r15,%r13 movq 728(%rcx,%rdi),%r15 cmove %r15,%r14 cmpq $0,%rdx movq %rsi,%r15 cmovl %r11,%rsi cmovl %r15,%r11 movq %r9,%r15 cmovl %r12,%r9 cmovl %r15,%r12 movq %rax,%r15 cmovl %r13,%rax cmovl %r15,%r13 movq %r10,%r15 cmovl %r14,%r10 cmovl %r15,%r14 movq %r11,%r15 movq %r12,80(%rsp) movq %r13,88(%rsp) movq %r14,96(%rsp) // sub subq %rsi,%r11 sbbq %r9,%r12 sbbq %rax,%r13 sbbq %r10,%r14 movq $0,%rbx movq $38,%rbp cmovae %rbx,%rbp subq %rbp,%r11 sbbq %rbx,%r12 sbbq %rbx,%r13 sbbq %rbx,%r14 cmovc %rbp,%rbx subq %rbx,%r11 movq %r11,128(%rsp) movq %r12,136(%rsp) movq %r13,144(%rsp) movq %r14,152(%rsp) // add addq %r15,%rsi adcq 80(%rsp),%r9 adcq 88(%rsp),%rax adcq 96(%rsp),%r10 movq $0,%rbx movq $38,%rbp cmovae %rbx,%rbp addq %rbp,%rsi adcq %rbx,%r9 adcq %rbx,%rax adcq %rbx,%r10 cmovc %rbp,%rbx addq %rbx,%rsi movq %rsi,160(%rsp) movq %r9,168(%rsp) movq %rax,176(%rsp) movq %r10,184(%rsp) movq $0,%rsi movq $0,%r9 movq $0,%rax movq $0,%r10 cmpq $1,%r8 movq 64(%rcx,%rdi),%r11 cmove %r11,%rsi movq 72(%rcx,%rdi),%r11 cmove %r11,%r9 movq 80(%rcx,%rdi),%r11 cmove %r11,%rax movq 88(%rcx,%rdi),%r11 cmove %r11,%r10 cmpq $2,%r8 movq 160(%rcx,%rdi),%r11 cmove %r11,%rsi movq 168(%rcx,%rdi),%r11 cmove %r11,%r9 movq 176(%rcx,%rdi),%r11 cmove %r11,%rax movq 184(%rcx,%rdi),%r11 cmove %r11,%r10 cmpq $3,%r8 movq 256(%rcx,%rdi),%r11 cmove %r11,%rsi movq 264(%rcx,%rdi),%r11 cmove %r11,%r9 movq 272(%rcx,%rdi),%r11 cmove %r11,%rax movq 280(%rcx,%rdi),%r11 cmove %r11,%r10 cmpq $4,%r8 movq 352(%rcx,%rdi),%r11 cmove %r11,%rsi movq 360(%rcx,%rdi),%r11 cmove %r11,%r9 movq 368(%rcx,%rdi),%r11 cmove %r11,%rax movq 376(%rcx,%rdi),%r11 cmove %r11,%r10 cmpq $5,%r8 movq 448(%rcx,%rdi),%r11 cmove %r11,%rsi movq 456(%rcx,%rdi),%r11 cmove %r11,%r9 movq 464(%rcx,%rdi),%r11 cmove %r11,%rax movq 472(%rcx,%rdi),%r11 cmove %r11,%r10 cmpq $6,%r8 movq 544(%rcx,%rdi),%r11 cmove %r11,%rsi movq 552(%rcx,%rdi),%r11 cmove %r11,%r9 movq 560(%rcx,%rdi),%r11 cmove %r11,%rax movq 568(%rcx,%rdi),%r11 cmove %r11,%r10 cmpq $7,%r8 movq 640(%rcx,%rdi),%r11 cmove %r11,%rsi movq 648(%rcx,%rdi),%r11 cmove %r11,%r9 movq 656(%rcx,%rdi),%r11 cmove %r11,%rax movq 664(%rcx,%rdi),%r11 cmove %r11,%r10 cmpq $8,%r8 movq 736(%rcx,%rdi),%r8 cmove %r8,%rsi movq 744(%rcx,%rdi),%r8 cmove %r8,%r9 movq 752(%rcx,%rdi),%r8 cmove %r8,%rax movq 760(%rcx,%rdi),%rdi cmove %rdi,%r10 movq $0,%rdi movq $0,%rcx movq $0,%r8 movq $0,%r11 subq %rsi,%rdi sbbq %r9,%rcx sbbq %rax,%r8 sbbq %r10,%r11 movq $0,%r12 movq $38,%r13 cmovae %r12,%r13 subq %r13,%rdi sbbq %r12,%rcx sbbq %r12,%r8 sbbq %r12,%r11 cmovc %r13,%r12 subq %r12,%rdi cmpq $0,%rdx cmovl %rdi,%rsi cmovl %rcx,%r9 cmovl %r8,%rax cmovl %r11,%r10 movq %rsi,224(%rsp) movq %r9,232(%rsp) movq %rax,240(%rsp) movq %r10,248(%rsp) movq $2,192(%rsp) movq $0,200(%rsp) movq $0,208(%rsp) movq $0,216(%rsp) /* loop: i=1,i<64,i=i+1 */ movq $1,80(%rsp) .L: /* choose t */ movq 72(%rsp),%rcx movq 80(%rsp),%rsi movq 64(%rsp),%rdx movzbl 0(%rdx,%rsi),%eax movsbq %al,%rdx imulq $768,%rsi,%rdi movq %rdx,%rsi sarq $7,%rsi movq %rdx,%r8 addq %rsi,%r8 xorq %rsi,%r8 movq $1,%rsi movq $0,%r9 movq $0,%rax movq $0,%r10 movq $1,%r11 movq $0,%r12 movq $0,%r13 movq $0,%r14 cmpq $1,%r8 movq 0(%rcx,%rdi),%r15 cmove %r15,%rsi movq 8(%rcx,%rdi),%r15 cmove %r15,%r9 movq 16(%rcx,%rdi),%r15 cmove %r15,%rax movq 24(%rcx,%rdi),%r15 cmove %r15,%r10 movq 32(%rcx,%rdi),%r15 cmove %r15,%r11 movq 40(%rcx,%rdi),%r15 cmove %r15,%r12 movq 48(%rcx,%rdi),%r15 cmove %r15,%r13 movq 56(%rcx,%rdi),%r15 cmove %r15,%r14 cmpq $2,%r8 movq 96(%rcx,%rdi),%r15 cmove %r15,%rsi movq 104(%rcx,%rdi),%r15 cmove %r15,%r9 movq 112(%rcx,%rdi),%r15 cmove %r15,%rax movq 120(%rcx,%rdi),%r15 cmove %r15,%r10 movq 128(%rcx,%rdi),%r15 cmove %r15,%r11 movq 136(%rcx,%rdi),%r15 cmove %r15,%r12 movq 144(%rcx,%rdi),%r15 cmove %r15,%r13 movq 152(%rcx,%rdi),%r15 cmove %r15,%r14 cmpq $3,%r8 movq 192(%rcx,%rdi),%r15 cmove %r15,%rsi movq 200(%rcx,%rdi),%r15 cmove %r15,%r9 movq 208(%rcx,%rdi),%r15 cmove %r15,%rax movq 216(%rcx,%rdi),%r15 cmove %r15,%r10 movq 224(%rcx,%rdi),%r15 cmove %r15,%r11 movq 232(%rcx,%rdi),%r15 cmove %r15,%r12 movq 240(%rcx,%rdi),%r15 cmove %r15,%r13 movq 248(%rcx,%rdi),%r15 cmove %r15,%r14 cmpq $4,%r8 movq 288(%rcx,%rdi),%r15 cmove %r15,%rsi movq 296(%rcx,%rdi),%r15 cmove %r15,%r9 movq 304(%rcx,%rdi),%r15 cmove %r15,%rax movq 312(%rcx,%rdi),%r15 cmove %r15,%r10 movq 320(%rcx,%rdi),%r15 cmove %r15,%r11 movq 328(%rcx,%rdi),%r15 cmove %r15,%r12 movq 336(%rcx,%rdi),%r15 cmove %r15,%r13 movq 344(%rcx,%rdi),%r15 cmove %r15,%r14 cmpq $5,%r8 movq 384(%rcx,%rdi),%r15 cmove %r15,%rsi movq 392(%rcx,%rdi),%r15 cmove %r15,%r9 movq 400(%rcx,%rdi),%r15 cmove %r15,%rax movq 408(%rcx,%rdi),%r15 cmove %r15,%r10 movq 416(%rcx,%rdi),%r15 cmove %r15,%r11 movq 424(%rcx,%rdi),%r15 cmove %r15,%r12 movq 432(%rcx,%rdi),%r15 cmove %r15,%r13 movq 440(%rcx,%rdi),%r15 cmove %r15,%r14 cmpq $6,%r8 movq 480(%rcx,%rdi),%r15 cmove %r15,%rsi movq 488(%rcx,%rdi),%r15 cmove %r15,%r9 movq 496(%rcx,%rdi),%r15 cmove %r15,%rax movq 504(%rcx,%rdi),%r15 cmove %r15,%r10 movq 512(%rcx,%rdi),%r15 cmove %r15,%r11 movq 520(%rcx,%rdi),%r15 cmove %r15,%r12 movq 528(%rcx,%rdi),%r15 cmove %r15,%r13 movq 536(%rcx,%rdi),%r15 cmove %r15,%r14 cmpq $7,%r8 movq 576(%rcx,%rdi),%r15 cmove %r15,%rsi movq 584(%rcx,%rdi),%r15 cmove %r15,%r9 movq 592(%rcx,%rdi),%r15 cmove %r15,%rax movq 600(%rcx,%rdi),%r15 cmove %r15,%r10 movq 608(%rcx,%rdi),%r15 cmove %r15,%r11 movq 616(%rcx,%rdi),%r15 cmove %r15,%r12 movq 624(%rcx,%rdi),%r15 cmove %r15,%r13 movq 632(%rcx,%rdi),%r15 cmove %r15,%r14 cmpq $8,%r8 movq 672(%rcx,%rdi),%r15 cmove %r15,%rsi movq 680(%rcx,%rdi),%r15 cmove %r15,%r9 movq 688(%rcx,%rdi),%r15 cmove %r15,%rax movq 696(%rcx,%rdi),%r15 cmove %r15,%r10 movq 704(%rcx,%rdi),%r15 cmove %r15,%r11 movq 712(%rcx,%rdi),%r15 cmove %r15,%r12 movq 720(%rcx,%rdi),%r15 cmove %r15,%r13 movq 728(%rcx,%rdi),%r15 cmove %r15,%r14 cmpq $0,%rdx movq %rsi,%r15 cmovl %r11,%rsi cmovl %r15,%r11 movq %r9,%r15 cmovl %r12,%r9 cmovl %r15,%r12 movq %rax,%r15 cmovl %r13,%rax cmovl %r15,%r13 movq %r10,%r15 cmovl %r14,%r10 cmovl %r15,%r14 movq %rsi,256(%rsp) movq %r9,264(%rsp) movq %rax,272(%rsp) movq %r10,280(%rsp) movq %r11,288(%rsp) movq %r12,296(%rsp) movq %r13,304(%rsp) movq %r14,312(%rsp) movq $0,%rsi movq $0,%r9 movq $0,%rax movq $0,%r10 cmpq $1,%r8 movq 64(%rcx,%rdi),%r11 cmove %r11,%rsi movq 72(%rcx,%rdi),%r11 cmove %r11,%r9 movq 80(%rcx,%rdi),%r11 cmove %r11,%rax movq 88(%rcx,%rdi),%r11 cmove %r11,%r10 cmpq $2,%r8 movq 160(%rcx,%rdi),%r11 cmove %r11,%rsi movq 168(%rcx,%rdi),%r11 cmove %r11,%r9 movq 176(%rcx,%rdi),%r11 cmove %r11,%rax movq 184(%rcx,%rdi),%r11 cmove %r11,%r10 cmpq $3,%r8 movq 256(%rcx,%rdi),%r11 cmove %r11,%rsi movq 264(%rcx,%rdi),%r11 cmove %r11,%r9 movq 272(%rcx,%rdi),%r11 cmove %r11,%rax movq 280(%rcx,%rdi),%r11 cmove %r11,%r10 cmpq $4,%r8 movq 352(%rcx,%rdi),%r11 cmove %r11,%rsi movq 360(%rcx,%rdi),%r11 cmove %r11,%r9 movq 368(%rcx,%rdi),%r11 cmove %r11,%rax movq 376(%rcx,%rdi),%r11 cmove %r11,%r10 cmpq $5,%r8 movq 448(%rcx,%rdi),%r11 cmove %r11,%rsi movq 456(%rcx,%rdi),%r11 cmove %r11,%r9 movq 464(%rcx,%rdi),%r11 cmove %r11,%rax movq 472(%rcx,%rdi),%r11 cmove %r11,%r10 cmpq $6,%r8 movq 544(%rcx,%rdi),%r11 cmove %r11,%rsi movq 552(%rcx,%rdi),%r11 cmove %r11,%r9 movq 560(%rcx,%rdi),%r11 cmove %r11,%rax movq 568(%rcx,%rdi),%r11 cmove %r11,%r10 cmpq $7,%r8 movq 640(%rcx,%rdi),%r11 cmove %r11,%rsi movq 648(%rcx,%rdi),%r11 cmove %r11,%r9 movq 656(%rcx,%rdi),%r11 cmove %r11,%rax movq 664(%rcx,%rdi),%r11 cmove %r11,%r10 cmpq $8,%r8 movq 736(%rcx,%rdi),%r8 cmove %r8,%rsi movq 744(%rcx,%rdi),%r8 cmove %r8,%r9 movq 752(%rcx,%rdi),%r8 cmove %r8,%rax movq 760(%rcx,%rdi),%rdi cmove %rdi,%r10 movq $0,%rdi movq $0,%rcx movq $0,%r8 movq $0,%r11 subq %rsi,%rdi sbbq %r9,%rcx sbbq %rax,%r8 sbbq %r10,%r11 movq $0,%r12 movq $38,%r13 cmovae %r12,%r13 subq %r13,%rdi sbbq %r12,%rcx sbbq %r12,%r8 sbbq %r12,%r11 cmovc %r13,%r12 subq %r12,%rdi cmpq $0,%rdx cmovl %rdi,%rsi cmovl %rcx,%r9 cmovl %r8,%rax cmovl %r11,%r10 movq %rsi,320(%rsp) movq %r9,328(%rsp) movq %rax,336(%rsp) movq %r10,344(%rsp) /* nielsadd2 */ // load movq 160(%rsp),%rdx movq 168(%rsp),%rcx movq 176(%rsp),%r8 movq 184(%rsp),%r9 // copy movq %rdx,%rax movq %rcx,%r10 movq %r8,%r11 movq %r9,%r12 // sub subq 128(%rsp),%rdx sbbq 136(%rsp),%rcx sbbq 144(%rsp),%r8 sbbq 152(%rsp),%r9 movq $0,%r13 movq $38,%r14 cmovae %r13,%r14 subq %r14,%rdx sbbq %r13,%rcx sbbq %r13,%r8 sbbq %r13,%r9 cmovc %r14,%r13 subq %r13,%rdx movq %rdx,352(%rsp) movq %rcx,360(%rsp) movq %r8,368(%rsp) movq %r9,376(%rsp) // add addq 128(%rsp),%rax adcq 136(%rsp),%r10 adcq 144(%rsp),%r11 adcq 152(%rsp),%r12 movq $0,%r13 movq $38,%r14 cmovae %r13,%r14 addq %r14,%rax adcq %r13,%r10 adcq %r13,%r11 adcq %r13,%r12 cmovc %r14,%r13 addq %r13,%rax movq %rax,384(%rsp) movq %r10,392(%rsp) movq %r11,400(%rsp) movq %r12,408(%rsp) // mul movq 256(%rsp),%rdx mulx 352(%rsp),%r8,%r9 mulx 360(%rsp),%rcx,%r10 addq %rcx,%r9 mulx 368(%rsp),%rcx,%r11 adcq %rcx,%r10 mulx 376(%rsp),%rcx,%r12 adcq %rcx,%r11 adcq $0,%r12 movq 264(%rsp),%rdx mulx 352(%rsp),%rax,%rbx mulx 360(%rsp),%rcx,%rbp addq %rcx,%rbx mulx 368(%rsp),%rcx,%rsi adcq %rcx,%rbp mulx 376(%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 352(%rsp),%rax,%rbx mulx 360(%rsp),%rcx,%rbp addq %rcx,%rbx mulx 368(%rsp),%rcx,%rsi adcq %rcx,%rbp mulx 376(%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 352(%rsp),%rax,%rbx mulx 360(%rsp),%rcx,%rbp addq %rcx,%rbx mulx 368(%rsp),%rcx,%rsi adcq %rcx,%rbp mulx 376(%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 movq %r8,352(%rsp) movq %r9,360(%rsp) movq %r10,368(%rsp) movq %r11,376(%rsp) // mul movq 288(%rsp),%rdx mulx 384(%rsp),%r8,%r9 mulx 392(%rsp),%rcx,%r10 addq %rcx,%r9 mulx 400(%rsp),%rcx,%r11 adcq %rcx,%r10 mulx 408(%rsp),%rcx,%r12 adcq %rcx,%r11 adcq $0,%r12 movq 296(%rsp),%rdx mulx 384(%rsp),%rax,%rbx mulx 392(%rsp),%rcx,%rbp addq %rcx,%rbx mulx 400(%rsp),%rcx,%rsi adcq %rcx,%rbp mulx 408(%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 304(%rsp),%rdx mulx 384(%rsp),%rax,%rbx mulx 392(%rsp),%rcx,%rbp addq %rcx,%rbx mulx 400(%rsp),%rcx,%rsi adcq %rcx,%rbp mulx 408(%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 312(%rsp),%rdx mulx 384(%rsp),%rax,%rbx mulx 392(%rsp),%rcx,%rbp addq %rcx,%rbx mulx 400(%rsp),%rcx,%rsi adcq %rcx,%rbp mulx 408(%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 // sub subq 352(%rsp),%r8 sbbq 360(%rsp),%r9 sbbq 368(%rsp),%r10 sbbq 376(%rsp),%r11 movq $0,%rax movq $38,%rbx cmovae %rax,%rbx subq %rbx,%r8 sbbq %rax,%r9 sbbq %rax,%r10 sbbq %rax,%r11 cmovc %rbx,%rax subq %rax,%r8 movq %r8,384(%rsp) movq %r9,392(%rsp) movq %r10,400(%rsp) movq %r11,408(%rsp) // add addq 352(%rsp),%r12 adcq 360(%rsp),%r13 adcq 368(%rsp),%r14 adcq 376(%rsp),%r15 movq $0,%rax movq $38,%rbx cmovae %rax,%rbx addq %rbx,%r12 adcq %rax,%r13 adcq %rax,%r14 adcq %rax,%r15 cmovc %rbx,%rax addq %rax,%r12 movq %r12,352(%rsp) movq %r13,360(%rsp) movq %r14,368(%rsp) movq %r15,376(%rsp) // mul movq 320(%rsp),%rdx mulx 224(%rsp),%r8,%r9 mulx 232(%rsp),%rcx,%r10 addq %rcx,%r9 mulx 240(%rsp),%rcx,%r11 adcq %rcx,%r10 mulx 248(%rsp),%rcx,%r12 adcq %rcx,%r11 adcq $0,%r12 movq 328(%rsp),%rdx mulx 224(%rsp),%rax,%rbx mulx 232(%rsp),%rcx,%rbp addq %rcx,%rbx mulx 240(%rsp),%rcx,%rsi adcq %rcx,%rbp mulx 248(%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 336(%rsp),%rdx mulx 224(%rsp),%rax,%rbx mulx 232(%rsp),%rcx,%rbp addq %rcx,%rbx mulx 240(%rsp),%rcx,%rsi adcq %rcx,%rbp mulx 248(%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 344(%rsp),%rdx mulx 224(%rsp),%rax,%rbx mulx 232(%rsp),%rcx,%rbp addq %rcx,%rbx mulx 240(%rsp),%rcx,%rsi adcq %rcx,%rbp mulx 248(%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 movq 192(%rsp),%r12 movq 200(%rsp),%r13 movq 208(%rsp),%r14 movq 216(%rsp),%r15 // double addq %r12,%r12 adcq %r13,%r13 adcq %r14,%r14 adcq %r15,%r15 movq $0,%rbx movq $38,%rax cmovae %rbx,%rax addq %rax,%r12 adcq %rbx,%r13 adcq %rbx,%r14 adcq %rbx,%r15 cmovc %rax,%rbx addq %rbx,%r12 // copy movq %r12,%rsi movq %r13,%rax movq %r14,%rcx movq %r15,%rdx // sub subq %r8,%r12 sbbq %r9,%r13 sbbq %r10,%r14 sbbq %r11,%r15 movq $0,%rbx movq $38,%rbp cmovae %rbx,%rbp subq %rbp,%r12 sbbq %rbx,%r13 sbbq %rbx,%r14 sbbq %rbx,%r15 cmovc %rbp,%rbx subq %rbx,%r12 movq %r12,448(%rsp) movq %r13,456(%rsp) movq %r14,464(%rsp) movq %r15,472(%rsp) // add addq %r8,%rsi adcq %r9,%rax adcq %r10,%rcx adcq %r11,%rdx movq $0,%rbx movq $38,%rbp cmovae %rbx,%rbp addq %rbp,%rsi adcq %rbx,%rax adcq %rbx,%rcx adcq %rbx,%rdx cmovc %rbp,%rbx addq %rbx,%rsi movq %rsi,416(%rsp) movq %rax,424(%rsp) movq %rcx,432(%rsp) movq %rdx,440(%rsp) /* p1p1 to p3 */ // convert to 9x4 form vmovdqa 384(%rsp),%ymm8 vmovdqa 352(%rsp),%ymm9 vmovdqa 416(%rsp),%ymm10 vmovdqa 352(%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 448(%rsp),%ymm8 vmovdqa 416(%rsp),%ymm9 vmovdqa 448(%rsp),%ymm10 vmovdqa 384(%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 %ymm10,128(%rsp) vmovdqa %ymm11,160(%rsp) vmovdqa %ymm12,192(%rsp) vmovdqa %ymm13,224(%rsp) movq 80(%rsp),%r15 addq $1,%r15 movq %r15,80(%rsp) cmpq $63,%r15 jle .L /* store final value of r */ movq 56(%rsp),%rdi movq 128(%rsp),%r8 movq 136(%rsp),%r9 movq 144(%rsp),%r10 movq 152(%rsp),%r11 movq %r8,0(%rdi) movq %r9,8(%rdi) movq %r10,16(%rdi) movq %r11,24(%rdi) movq 160(%rsp),%r8 movq 168(%rsp),%r9 movq 176(%rsp),%r10 movq 184(%rsp),%r11 movq %r8,32(%rdi) movq %r9,40(%rdi) movq %r10,48(%rdi) movq %r11,56(%rdi) movq 192(%rsp),%r8 movq 200(%rsp),%r9 movq 208(%rsp),%r10 movq 216(%rsp),%r11 movq %r8,64(%rdi) movq %r9,72(%rdi) movq %r10,80(%rdi) movq %r11,88(%rdi) movq 224(%rsp),%r8 movq 232(%rsp),%r9 movq 240(%rsp),%r10 movq 248(%rsp),%r11 movq %r8,96(%rdi) movq %r9,104(%rdi) movq %r10,112(%rdi) movq %r11,120(%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