This submission is migrated from previous version of oj.uz, which used different machine for grading. This submission may have different result if resubmitted.
#include <bits/stdc++.h>
#define Loop(x,l,r) for (ll x = (l); x < (r); ++x)
#define LoopR(x,l,r) for (ll x = (r)-1; x >= (l); --x)
typedef long long ll;
typedef std::pair<int, int> pii;
typedef std::pair<ll , ll > pll;
using namespace std;
const int N = 100'010;
int a[N], b[N];
int n, q;
__attribute__((naked))
int cnt(int x, int y, int z) {
asm(
" movl %edi, %r9d\n"
" movslq n(%rip), %rdi\n"
" testq %rdi, %rdi\n"
" jle .myL9\n"
" leaq -1(%rdi), %rax\n"
" movl %esi, %r10d\n"
" movl %edx, %r11d\n"
" cmpq $6, %rax\n"
" jbe .myL10\n"
" movq %rdi, %r8\n"
" vmovd %edx, %xmm5\n"
" vmovd %esi, %xmm4\n"
" xorl %eax, %eax\n"
" shrq $3, %r8\n"
" vmovd %r9d, %xmm6\n"
" vpbroadcastd %xmm4, %ymm4\n"
" salq $5, %r8\n"
" vpbroadcastd %xmm6, %ymm6\n"
" vpbroadcastd %xmm5, %ymm5\n"
" leaq -32(%r8), %rdx\n"
" vpxor %xmm3, %xmm3, %xmm3\n"
" leaq a(%rip), %rcx\n"
" shrq $5, %rdx\n"
" leaq b(%rip), %rsi\n"
" addq $1, %rdx\n"
" andl $3, %edx\n"
" je .myL4\n"
" cmpq $1, %rdx\n"
" je .myL21\n"
" cmpq $2, %rdx\n"
" je .myL22\n"
" vmovdqa (%rcx), %ymm0\n"
" vmovdqa (%rsi), %ymm1\n"
" movl $32, %eax\n"
" vpaddd %ymm1, %ymm0, %ymm2\n"
" vpminsd %ymm1, %ymm4, %ymm1\n"
" vpcmpgtd %ymm0, %ymm6, %ymm0\n"
" vpcmpeqd %ymm1, %ymm4, %ymm1\n"
" vpandn %ymm1, %ymm0, %ymm0\n"
" vpcmpgtd %ymm2, %ymm5, %ymm1\n"
" vpandn %ymm0, %ymm1, %ymm0\n"
" vpsubd %ymm0, %ymm3, %ymm3\n"
".myL22:\n"
" vmovdqa (%rcx,%rax), %ymm0\n"
" vmovdqa (%rsi,%rax), %ymm1\n"
" addq $32, %rax\n"
" vpaddd %ymm1, %ymm0, %ymm2\n"
" vpminsd %ymm1, %ymm4, %ymm1\n"
" vpcmpgtd %ymm0, %ymm6, %ymm0\n"
" vpcmpeqd %ymm1, %ymm4, %ymm1\n"
" vpandn %ymm1, %ymm0, %ymm0\n"
" vpcmpgtd %ymm2, %ymm5, %ymm1\n"
" vpandn %ymm0, %ymm1, %ymm0\n"
" vpsubd %ymm0, %ymm3, %ymm3\n"
".myL21:\n"
" vmovdqa (%rcx,%rax), %ymm0\n"
" vmovdqa (%rsi,%rax), %ymm1\n"
" addq $32, %rax\n"
" vpaddd %ymm1, %ymm0, %ymm2\n"
" vpminsd %ymm1, %ymm4, %ymm1\n"
" vpcmpgtd %ymm0, %ymm6, %ymm0\n"
" vpcmpeqd %ymm1, %ymm4, %ymm1\n"
" vpandn %ymm1, %ymm0, %ymm0\n"
" vpcmpgtd %ymm2, %ymm5, %ymm1\n"
" vpandn %ymm0, %ymm1, %ymm0\n"
" vpsubd %ymm0, %ymm3, %ymm3\n"
" cmpq %r8, %rax\n"
" je .myL27\n"
".myL4:\n"
" vmovdqa (%rcx,%rax), %ymm0\n"
" vmovdqa (%rsi,%rax), %ymm1\n"
" leaq 32(%rax), %rdx\n"
" vpaddd %ymm1, %ymm0, %ymm2\n"
" vpminsd %ymm1, %ymm4, %ymm1\n"
" vpcmpgtd %ymm0, %ymm6, %ymm0\n"
" vpcmpeqd %ymm1, %ymm4, %ymm1\n"
" vpandn %ymm1, %ymm0, %ymm0\n"
" vpcmpgtd %ymm2, %ymm5, %ymm1\n"
" vpandn %ymm0, %ymm1, %ymm0\n"
" vmovdqa 32(%rsi,%rax), %ymm1\n"
" vpsubd %ymm0, %ymm3, %ymm3\n"
" vmovdqa 32(%rcx,%rax), %ymm0\n"
" vpaddd %ymm1, %ymm0, %ymm2\n"
" vpminsd %ymm1, %ymm4, %ymm1\n"
" vpcmpgtd %ymm0, %ymm6, %ymm0\n"
" vpcmpeqd %ymm1, %ymm4, %ymm1\n"
" vpandn %ymm1, %ymm0, %ymm0\n"
" vpcmpgtd %ymm2, %ymm5, %ymm1\n"
" vpandn %ymm0, %ymm1, %ymm0\n"
" vmovdqa 64(%rsi,%rax), %ymm1\n"
" vpsubd %ymm0, %ymm3, %ymm3\n"
" vmovdqa 64(%rcx,%rax), %ymm0\n"
" leaq 96(%rdx), %rax\n"
" vpaddd %ymm1, %ymm0, %ymm2\n"
" vpminsd %ymm1, %ymm4, %ymm1\n"
" vpcmpgtd %ymm0, %ymm6, %ymm0\n"
" vpcmpeqd %ymm1, %ymm4, %ymm1\n"
" vpandn %ymm1, %ymm0, %ymm0\n"
" vpcmpgtd %ymm2, %ymm5, %ymm1\n"
" vpandn %ymm0, %ymm1, %ymm0\n"
" vmovdqa 64(%rsi,%rdx), %ymm1\n"
" vpsubd %ymm0, %ymm3, %ymm3\n"
" vmovdqa 64(%rcx,%rdx), %ymm0\n"
" vpaddd %ymm1, %ymm0, %ymm2\n"
" vpminsd %ymm1, %ymm4, %ymm1\n"
" vpcmpgtd %ymm0, %ymm6, %ymm0\n"
" vpcmpeqd %ymm1, %ymm4, %ymm1\n"
" vpandn %ymm1, %ymm0, %ymm0\n"
" vpcmpgtd %ymm2, %ymm5, %ymm1\n"
" vpandn %ymm0, %ymm1, %ymm0\n"
" vpsubd %ymm0, %ymm3, %ymm3\n"
" cmpq %r8, %rax\n"
" jne .myL4\n"
".myL27:\n"
" vmovdqa %xmm3, %xmm0\n"
" vextracti128 $0x1, %ymm3, %xmm3\n"
" movq %rdi, %rax\n"
" vpaddd %xmm3, %xmm0, %xmm0\n"
" andq $-8, %rax\n"
" vpsrldq $8, %xmm0, %xmm1\n"
" movq %rax, %rdx\n"
" vpaddd %xmm1, %xmm0, %xmm0\n"
" vpsrldq $4, %xmm0, %xmm1\n"
" vpaddd %xmm1, %xmm0, %xmm0\n"
" vmovd %xmm0, %r8d\n"
" cmpq %rdi, %rax\n"
" je .myL33\n"
" vzeroupper\n"
".myL3:\n"
" pushq %rbp\n"
" .cfi_def_cfa_offset 16\n"
" .cfi_offset 6, -16\n"
" movq %rsp, %rbp\n"
" .cfi_def_cfa_register 6\n"
" pushq %r13\n"
" pushq %r12\n"
" pushq %rbx\n"
" .cfi_offset 13, -24\n"
" .cfi_offset 12, -32\n"
" .cfi_offset 3, -40\n"
" movq %rdi, %rbx\n"
" subq %rax, %rbx\n"
" leaq -1(%rbx), %r12\n"
" cmpq $2, %r12\n"
" jbe .myL7\n"
" vmovdqa (%rcx,%rax,4), %xmm2\n"
" vmovdqa (%rsi,%rax,4), %xmm1\n"
" vmovd %r10d, %xmm7\n"
" vpshufd $0, %xmm7, %xmm0\n"
" vmovd %r9d, %xmm7\n"
" vpaddd %xmm1, %xmm2, %xmm3\n"
" vpminsd %xmm1, %xmm0, %xmm1\n"
" vpcmpeqd %xmm1, %xmm0, %xmm1\n"
" vpshufd $0, %xmm7, %xmm0\n"
" vmovd %r11d, %xmm7\n"
" vpcmpgtd %xmm2, %xmm0, %xmm0\n"
" vpandn %xmm1, %xmm0, %xmm0\n"
" vpshufd $0, %xmm7, %xmm1\n"
" vpcmpgtd %xmm3, %xmm1, %xmm1\n"
" vpandn %xmm0, %xmm1, %xmm1\n"
" vpxor %xmm0, %xmm0, %xmm0\n"
" vpsubd %xmm1, %xmm0, %xmm0\n"
" vpsrldq $8, %xmm0, %xmm0\n"
" vpsubd %xmm1, %xmm0, %xmm0\n"
" vpsrldq $4, %xmm0, %xmm1\n"
" vpaddd %xmm1, %xmm0, %xmm0\n"
" vmovd %xmm0, %eax\n"
" addl %eax, %r8d\n"
" movq %rbx, %rax\n"
" andq $-4, %rax\n"
" addq %rax, %rdx\n"
" cmpq %rax, %rbx\n"
" je .myL1\n"
".myL7:\n"
" movl (%rcx,%rdx,4), %ebx\n"
" movl (%rsi,%rdx,4), %r13d\n"
" xorl %eax, %eax\n"
" cmpl %r9d, %ebx\n"
" setge %al\n"
" xorl %r12d, %r12d\n"
" negl %eax\n"
" cmpl %r10d, %r13d\n"
" setge %r12b\n"
" addl %r13d, %ebx\n"
" negl %r12d\n"
" andl %r12d, %eax\n"
" cmpl %r11d, %ebx\n"
" setge %bl\n"
" movzbl %bl, %ebx\n"
" negl %ebx\n"
" andl %ebx, %eax\n"
" subl %eax, %r8d\n"
" leaq 1(%rdx), %rax\n"
" cmpq %rdi, %rax\n"
" jge .myL1\n"
" movl (%rsi,%rax,4), %ebx\n"
" movl (%rcx,%rax,4), %r12d\n"
" xorl %eax, %eax\n"
" cmpl %ebx, %r10d\n"
" setle %al\n"
" xorl %r13d, %r13d\n"
" negl %eax\n"
" cmpl %r12d, %r9d\n"
" setle %r13b\n"
" addl %r12d, %ebx\n"
" negl %r13d\n"
" andl %r13d, %eax\n"
" cmpl %ebx, %r11d\n"
" setle %bl\n"
" addq $2, %rdx\n"
" movzbl %bl, %ebx\n"
" negl %ebx\n"
" andl %ebx, %eax\n"
" subl %eax, %r8d\n"
" cmpq %rdx, %rdi\n"
" jle .myL1\n"
" movl (%rcx,%rdx,4), %ecx\n"
" movl (%rsi,%rdx,4), %esi\n"
" xorl %eax, %eax\n"
" cmpl %ecx, %r9d\n"
" setle %al\n"
" xorl %edx, %edx\n"
" negl %eax\n"
" cmpl %esi, %r10d\n"
" setle %dl\n"
" addl %esi, %ecx\n"
" negl %edx\n"
" andl %edx, %eax\n"
" xorl %edx, %edx\n"
" cmpl %ecx, %r11d\n"
" setle %dl\n"
" negl %edx\n"
" andl %edx, %eax\n"
" subl %eax, %r8d\n"
".myL1:\n"
" popq %rbx\n"
" movl %r8d, %eax\n"
" popq %r12\n"
" popq %r13\n"
" popq %rbp\n"
" .cfi_def_cfa 7, 8\n"
" ret\n"
" .p2align 4,,10\n"
" .p2align 3\n"
".myL9:\n"
" .cfi_restore 3\n"
" .cfi_restore 6\n"
" .cfi_restore 12\n"
" .cfi_restore 13\n"
" xorl %r8d, %r8d\n"
".myL30:\n"
" movl %r8d, %eax\n"
" ret\n"
" .p2align 4,,10\n"
" .p2align 3\n"
".myL33:\n"
" vzeroupper\n"
" jmp .myL30\n"
".myL10:\n"
" xorl %eax, %eax\n"
" xorl %edx, %edx\n"
" leaq a(%rip), %rcx\n"
" xorl %r8d, %r8d\n"
" leaq b(%rip), %rsi\n"
" jmp .myL3\n"
);
}
int main()
{
cin.tie(0) -> sync_with_stdio(false);
cin >> n >> q;
Loop (i,0,n)
cin >> a[i] >> b[i];
Loop (i,0,q) {
int x, y, z;
cin >> x >> y >> z;
cout << cnt(x, y, z) << '\n';
}
}
| # | Verdict | Execution time | Memory | Grader output |
|---|
| Fetching results... |
| # | Verdict | Execution time | Memory | Grader output |
|---|
| Fetching results... |
| # | Verdict | Execution time | Memory | Grader output |
|---|
| Fetching results... |
| # | Verdict | Execution time | Memory | Grader output |
|---|
| Fetching results... |
