mirror of
https://github.com/ultimatepp/ultimatepp.git
synced 2026-06-02 22:03:19 -06:00
627 lines
13 KiB
C++
627 lines
13 KiB
C++
#ifdef CPU_X86
|
|
|
|
void memset8__(void *t, __m128i data, size_t len);
|
|
|
|
inline
|
|
void memset8(void *p, byte data, size_t len)
|
|
{
|
|
byte *t = (byte *)p;
|
|
if(len < 2) {
|
|
if(len)
|
|
t[0] = data;
|
|
return;
|
|
}
|
|
dword val4 = 0x1010101 * data;
|
|
if(len <= 4) {
|
|
*(word *)t = *(word *)(t + len - 2) = (word)val4;
|
|
return;
|
|
}
|
|
if(len > 16) {
|
|
memset8__(t, _mm_set1_epi32(val4), len);
|
|
return;
|
|
}
|
|
*(dword *)t = *(dword *)(t + len - 4) = val4;
|
|
if(len > 8)
|
|
*(dword *)(t + 4) = *(dword *)(t + len - 8) = val4;
|
|
}
|
|
|
|
inline
|
|
void memset16(void *p, word data, size_t len)
|
|
{
|
|
word *t = (word *)p;
|
|
if(len < 2) {
|
|
if(len)
|
|
t[0] = data;
|
|
return;
|
|
}
|
|
dword val4 = 0x10001 * data;
|
|
if(len >= 16) {
|
|
memset8__(t, _mm_set1_epi32(val4), 2 * len);
|
|
return;
|
|
}
|
|
*(dword *)(t + len - 2) = val4;
|
|
if(len & 8) {
|
|
_mm_storeu_si128((__m128i *)t, _mm_set1_epi32(val4));
|
|
t += 8;
|
|
}
|
|
if(len & 4) {
|
|
*(dword *)t = val4;
|
|
*(dword *)(t + 2) = val4;
|
|
t += 4;
|
|
}
|
|
if(len & 2)
|
|
*(dword *)t = val4;
|
|
}
|
|
|
|
inline
|
|
void memset32(void *p, dword data, size_t len)
|
|
{
|
|
dword *t = (dword *)p;
|
|
if(len < 4) {
|
|
if(len & 2) {
|
|
t[0] = t[1] = t[len - 1] = data;
|
|
return;
|
|
}
|
|
if(len & 1)
|
|
t[0] = data;
|
|
return;
|
|
}
|
|
__m128i val4 = _mm_set1_epi32(data);
|
|
if(len >= 16) {
|
|
memset8__(t, val4, 4 * len);
|
|
return;
|
|
}
|
|
auto Set4 = [&](size_t at) { _mm_storeu_si128((__m128i *)(t + at), val4); };
|
|
Set4(len - 4); // fill tail
|
|
if(len & 8) {
|
|
Set4(0); Set4(4);
|
|
t += 8;
|
|
}
|
|
if(len & 4)
|
|
Set4(0);
|
|
}
|
|
|
|
inline
|
|
void memset64(void *p, qword data, size_t len)
|
|
{
|
|
qword *t = (qword *)p;
|
|
if(len < 2) {
|
|
if(len)
|
|
t[0] = data;
|
|
return;
|
|
}
|
|
__m128i val2 = _mm_set1_epi64x(data);
|
|
if(len >= 8) {
|
|
memset8__(t, val2, 8 * len);
|
|
return;
|
|
}
|
|
auto Set128 = [&](size_t at) { _mm_storeu_si128((__m128i *)(t + at), val2); };
|
|
Set128(len - 2); // fill tail
|
|
if(len & 4) {
|
|
Set128(0); Set128(2);
|
|
t += 4;
|
|
}
|
|
if(len & 2)
|
|
Set128(0);
|
|
}
|
|
|
|
inline
|
|
void memset128(void *t, m128 data, size_t len)
|
|
{
|
|
if(len)
|
|
memset8__(t, *(__m128i *)&data, 16 * len);
|
|
}
|
|
|
|
void memcpy8__(void *p, const void *q, size_t len);
|
|
|
|
inline
|
|
void memcpy8(void *p, const void *q, size_t len)
|
|
{
|
|
byte *t = (byte *)p;
|
|
byte *s = (byte *)q;
|
|
if(len <= 4) {
|
|
if(len < 2) {
|
|
if(len)
|
|
t[0] = s[0];
|
|
return;
|
|
}
|
|
*(word *)t = *(word *)s;
|
|
*(word *)(t + len - 2) = *(word *)(s + len - 2);
|
|
return;
|
|
}
|
|
if(len <= 16) {
|
|
if(len <= 8) {
|
|
*(dword *)(t) = *(dword *)(s);
|
|
*(dword *)(t + len - 4) = *(dword *)(s + len - 4);
|
|
return;
|
|
}
|
|
*(uint64 *)t = *(uint64 *)s;
|
|
*(uint64 *)(t + len - 8) = *(uint64 *)(s + len - 8);
|
|
return;
|
|
}
|
|
memcpy8__(t, s, len);
|
|
}
|
|
|
|
inline
|
|
void memcpy16(void *p, const void *q, size_t len)
|
|
{
|
|
word *t = (word *)p;
|
|
word *s = (word *)q;
|
|
if(len <= 4) {
|
|
if(len < 2) {
|
|
if(len)
|
|
t[0] = s[0];
|
|
return;
|
|
}
|
|
*(dword *)t = *(dword *)s;
|
|
*(dword *)(t + len - 2) = *(dword *)(s + len - 2);
|
|
return;
|
|
}
|
|
if(len <= 16) {
|
|
if(len <= 8) {
|
|
*(uint64 *)(t) = *(uint64 *)(s);
|
|
*(uint64 *)(t + len - 4) = *(uint64 *)(s + len - 4);
|
|
return;
|
|
}
|
|
auto Copy128 = [&](size_t at) { _mm_storeu_si128((__m128i *)(t + at), _mm_loadu_si128((__m128i *)(s + at))); };
|
|
Copy128(0);
|
|
Copy128(len - 8);
|
|
return;
|
|
}
|
|
memcpy8__(t, s, 2 * len);
|
|
}
|
|
|
|
inline
|
|
void memcpy32(void *p, const void *q, size_t len)
|
|
{
|
|
dword *t = (dword *)p;
|
|
dword *s = (dword *)q;
|
|
|
|
#ifdef CPU_64
|
|
if(len <= 4) {
|
|
if(len) {
|
|
if(len > 1) {
|
|
*(int64 *)t = *(int64 *)s;
|
|
*(int64 *)(t + len - 2) = *(int64 *)(s + len - 2);
|
|
return;
|
|
}
|
|
*t = *s;
|
|
}
|
|
return;
|
|
}
|
|
#else
|
|
if(len < 4) {
|
|
if(len) {
|
|
if(len > 1) {
|
|
t[0] = s[0];
|
|
t[1] = s[1];
|
|
t[len - 1] = s[len - 1];
|
|
return;
|
|
}
|
|
*t = *s;
|
|
}
|
|
return;
|
|
}
|
|
#endif
|
|
|
|
auto Copy128 = [&](size_t at) { _mm_storeu_si128((__m128i *)(t + at), _mm_loadu_si128((__m128i *)(s + at))); };
|
|
|
|
if(len >= 16) {
|
|
memcpy8__(t, s, 4 * len);
|
|
return;
|
|
}
|
|
Copy128(len - 4); // copy tail
|
|
if(len & 8) {
|
|
Copy128(0); Copy128(4);
|
|
t += 8;
|
|
s += 8;
|
|
}
|
|
if(len & 4)
|
|
Copy128(0);
|
|
}
|
|
|
|
inline
|
|
void memcpy64(void *p, const void *q, size_t len)
|
|
{
|
|
qword *t = (qword *)p;
|
|
qword *s = (qword *)q;
|
|
|
|
if(len <= 2) {
|
|
if(len) {
|
|
if(len > 1) {
|
|
*(int64 *)t = *(int64 *)s;
|
|
*(int64 *)(t + len - 1) = *(int64 *)(s + len - 1);
|
|
return;
|
|
}
|
|
*t = *s;
|
|
}
|
|
return;
|
|
}
|
|
|
|
auto Copy4 = [&](size_t at) { _mm_storeu_si128((__m128i *)(t + at), _mm_loadu_si128((__m128i *)(s + at))); };
|
|
|
|
Copy4(len - 2); // copy tail
|
|
if(len >= 8) {
|
|
memcpy8__(t, s, 8 * len);
|
|
return;
|
|
}
|
|
if(len & 4) {
|
|
Copy4(0); Copy4(2);
|
|
t += 4;
|
|
s += 4;
|
|
}
|
|
if(len & 2)
|
|
Copy4(0);
|
|
}
|
|
|
|
inline
|
|
void memcpy128(void *p, const void *q, size_t len)
|
|
{
|
|
struct dqword { qword x[2]; };
|
|
|
|
static_assert(sizeof(dqword) == 16, "dqword sizeof");
|
|
|
|
dqword *t = (dqword *)p;
|
|
dqword *s = (dqword *)q;
|
|
|
|
auto Copy4 = [&](size_t at) { _mm_storeu_si128((__m128i *)(t + at), _mm_loadu_si128((__m128i *)(s + at))); };
|
|
|
|
if(len >= 8) {
|
|
memcpy8__(t, s, 16 * len);
|
|
return;
|
|
}
|
|
if(len & 4) {
|
|
Copy4(0); Copy4(1); Copy4(2); Copy4(3);
|
|
t += 4;
|
|
s += 4;
|
|
}
|
|
if(len & 2) {
|
|
Copy4(0); Copy4(1);
|
|
t += 2;
|
|
s += 2;
|
|
}
|
|
if(len & 1)
|
|
Copy4(0);
|
|
}
|
|
|
|
template <class T>
|
|
void memcpy_t(void *t, const T *s, size_t count)
|
|
{
|
|
#ifdef CPU_X86
|
|
if((sizeof(T) & 15) == 0)
|
|
memcpy128(t, s, count * (sizeof(T) >> 4));
|
|
else
|
|
if((sizeof(T) & 7) == 0)
|
|
memcpy64(t, s, count * (sizeof(T) >> 3));
|
|
else
|
|
#endif
|
|
if((sizeof(T) & 3) == 0)
|
|
memcpy32(t, s, count * (sizeof(T) >> 2));
|
|
else
|
|
if((sizeof(T) & 1) == 0)
|
|
memcpy16(t, s, count * (sizeof(T) >> 1));
|
|
else
|
|
memcpy8(t, s, count * sizeof(T));
|
|
}
|
|
|
|
force_inline
|
|
bool memeq8__(const void *p, const void *q, size_t len)
|
|
{
|
|
ASSERT(len >= 16);
|
|
const byte *t = (byte *)p;
|
|
const byte *s = (byte *)q;
|
|
|
|
auto Cmp128 = [&](size_t at) { return _mm_cmpeq_epi32(_mm_loadu_si128((__m128i *)(s + at)), _mm_loadu_si128((__m128i *)(t + at))); };
|
|
auto Neq = [](__m128i v) { return _mm_movemask_epi8(v) != 0xffff; };
|
|
auto And = [](__m128i a, __m128i b) { return _mm_and_si128(a, b); };
|
|
|
|
if(Neq(And(Cmp128(len - 16), Cmp128(0)))) // tail & alignment, also <= 32
|
|
return false;
|
|
|
|
if(len <= 32)
|
|
return true;
|
|
|
|
const byte *e = t + len; // align up
|
|
|
|
byte *t1 = (byte *)(((uintptr_t)t | 15) + 1);
|
|
s += t1 - t;
|
|
t = t1;
|
|
len = e - t;
|
|
e -= 32;
|
|
while(t <= e) {
|
|
if(Neq(And(Cmp128(0), Cmp128(1*16))))
|
|
return false;
|
|
s += 32;
|
|
t += 32;
|
|
}
|
|
if(len & 16)
|
|
if(Neq(Cmp128(0)))
|
|
return false;
|
|
return true;
|
|
}
|
|
|
|
force_inline
|
|
bool inline_memeq8_aligned(const void *p, const void *q, size_t len)
|
|
{
|
|
const byte *t = (const byte *)p;
|
|
const byte *s = (const byte *)q;
|
|
if(len >= 16) // 15..31 is the most important range for String, make it fastest
|
|
return memeq8__(t, s, len);
|
|
if(len > 4) {
|
|
if(len <= 8) // test tail first in case we are searching in sorted list
|
|
return Peek32(s + len - 4) == Peek32(t + len - 4) && Peek32(s) == Peek32(t);
|
|
return Peek64(s + len - 8) == Peek64(t + len - 8) && Peek64(s) == Peek64(t);
|
|
}
|
|
if(len < 2)
|
|
return len ? t[0] == s[0] : true;
|
|
if(Peek16(s + len - 2) != Peek16(t + len - 2))
|
|
return false;
|
|
return len > 2 ? Peek16(s) == Peek16(t) : true;
|
|
}
|
|
|
|
force_inline
|
|
bool inline_memeq16_aligned(const void *p, const void *q, size_t len)
|
|
{
|
|
const word *t = (const word *)p;
|
|
const word *s = (const word *)q;
|
|
if(len <= 2)
|
|
return len ? Peek16(s + len - 1) == Peek16(t + len - 1) && Peek16(s) == Peek16(t) : true;
|
|
if(len <= 8) {
|
|
if(len <= 4)
|
|
return Peek32(s + len - 2) == Peek32(t + len - 2) && Peek32(s) == Peek32(t);
|
|
return Peek64(s + len - 4) == Peek64(t + len - 4) && Peek64(s) == Peek64(t);
|
|
}
|
|
return memeq8__(t, s, 2 * len);
|
|
}
|
|
|
|
force_inline
|
|
bool inline_memeq32_aligned(const void *p, const void *q, size_t len)
|
|
{
|
|
const dword *t = (const dword *)p;
|
|
const dword *s = (const dword *)q;
|
|
if(len <= 4) {
|
|
if(len <= 2)
|
|
return len ? Peek32(s + len - 1) == Peek32(t + len - 1) && Peek32(s) == Peek32(t) : true;
|
|
return Peek64(s + len - 2) == Peek64(t + len - 2) && Peek64(s) == Peek64(t);
|
|
}
|
|
return memeq8__(t, s, 4 * len);
|
|
}
|
|
|
|
force_inline
|
|
bool inline_memeq64_aligned(const void *p, const void *q, size_t len)
|
|
{
|
|
const qword *t = (const qword *)p;
|
|
const qword *s = (const qword *)q;
|
|
if(len <= 2)
|
|
return len ? Peek64(s + len - 1) == Peek64(t + len - 1) && Peek64(s) == Peek64(t) : true;
|
|
return memeq8__(t, s, 8 * len);
|
|
}
|
|
|
|
force_inline
|
|
bool inline_memeq128_aligned(const void *t, const void *s, size_t len)
|
|
{
|
|
return len == 0 || memeq8__(t, s, 16 * len);
|
|
}
|
|
|
|
bool memeq8(const void *p, const void *q, size_t len);
|
|
bool memeq16(const void *p, const void *q, size_t len);
|
|
bool memeq32(const void *p, const void *q, size_t len);
|
|
bool memeq64(const void *p, const void *q, size_t len);
|
|
bool memeq128(const void *p, const void *q, size_t len);
|
|
|
|
#else
|
|
|
|
template <class T>
|
|
void memset__(void *p, T data, size_t len)
|
|
{
|
|
T *t = (T *)p;
|
|
while(len >= 8) {
|
|
t[0] = t[1] = t[2] = t[3] = t[4] = t[5] = t[6] = t[7] = data;
|
|
t += 8;
|
|
len -= 8;
|
|
}
|
|
if(len & 4) {
|
|
t[0] = t[1] = t[2] = t[3] = data;
|
|
t += 4;
|
|
}
|
|
if(len & 2) {
|
|
t[0] = t[1] = data;
|
|
t += 2;
|
|
}
|
|
if(len & 1)
|
|
t[0] = data;
|
|
}
|
|
|
|
inline
|
|
void memset8(void *p, byte data, size_t len)
|
|
{
|
|
if(len >= 64) {
|
|
memset(p, data, len);
|
|
return;
|
|
}
|
|
memset__<byte>(p, data, len);
|
|
}
|
|
|
|
inline
|
|
void memset16(void *p, word data, size_t len)
|
|
{
|
|
memset__<word>(p, data, len);
|
|
}
|
|
|
|
inline
|
|
void memset32(void *p, dword data, size_t len)
|
|
{
|
|
memset__<dword>(p, data, len);
|
|
}
|
|
|
|
inline
|
|
void memset64(void *p, dword data, size_t len)
|
|
{
|
|
memset__<qword>(p, data, len);
|
|
}
|
|
|
|
inline
|
|
void memcpy8(void *p, const void *q, size_t len)
|
|
{
|
|
memcpy(p, q, len);
|
|
}
|
|
|
|
inline
|
|
void memcpy16(void *p, const void *q, size_t len)
|
|
{
|
|
memcpy(p, q, 2 * len);
|
|
}
|
|
|
|
inline
|
|
void memcpy32(void *p, const void *q, size_t len)
|
|
{
|
|
memcpy(p, q, 4 * len);
|
|
}
|
|
|
|
inline
|
|
void memcpy64(void *p, const void *q, size_t len)
|
|
{
|
|
memcpy(p, q, 8 * len);
|
|
}
|
|
|
|
inline
|
|
void memcpy128(void *p, const void *q, size_t len)
|
|
{
|
|
memcpy(p, q, 16 * len);
|
|
}
|
|
|
|
template <class T>
|
|
void memcpy_t(void *t, const T *s, size_t count)
|
|
{
|
|
memcpy8(t, s, count * sizeof(T));
|
|
}
|
|
|
|
inline
|
|
bool inline_memeq8_aligned(const void *p, const void *q, size_t len)
|
|
{
|
|
const byte *t = (const byte *)p;
|
|
const byte *s = (const byte *)q;
|
|
while(len >= 8) {
|
|
if(*(qword *)t != *(qword *)s)
|
|
return false;
|
|
s += 8;
|
|
t += 8;
|
|
len -= 8;
|
|
}
|
|
if(len & 4) {
|
|
if(*(dword *)t != *(dword *)s)
|
|
return false;
|
|
s += 4;
|
|
t += 4;
|
|
}
|
|
if(len & 2) {
|
|
if(*(word *)t != *(word *)s)
|
|
return false;
|
|
s += 2;
|
|
t += 2;
|
|
}
|
|
if(len & 2) {
|
|
if(*t != *s)
|
|
return false;
|
|
}
|
|
return true;
|
|
}
|
|
|
|
force_inline
|
|
bool inline_memeq16_aligned(const void *p, const void *q, size_t len)
|
|
{
|
|
return memcmp(p, q, 2 * len) == 0;
|
|
}
|
|
|
|
force_inline
|
|
bool inline_memeq32_aligned(const void *p, const void *q, size_t len)
|
|
{
|
|
return memcmp(p, q, 4 * len) == 0;
|
|
}
|
|
|
|
force_inline
|
|
bool inline_memeq64_aligned(const void *p, const void *q, size_t len)
|
|
{
|
|
return memcmp(p, q, 8 * len) == 0;
|
|
}
|
|
|
|
inline
|
|
bool memeq8(const void *p, const void *q, size_t len)
|
|
{
|
|
return memcmp(p, q, len) == 0;
|
|
}
|
|
|
|
inline
|
|
bool memeq16(const void *p, const void *q, size_t len)
|
|
{
|
|
return memcmp(p, q, 2 * len) == 0;
|
|
}
|
|
|
|
inline
|
|
bool memeq32(const void *p, const void *q, size_t len)
|
|
{
|
|
return memcmp(p, q, 4 * len) == 0;
|
|
}
|
|
|
|
inline
|
|
bool memeq64(const void *p, const void *q, size_t len)
|
|
{
|
|
return memcmp(p, q, 8 * len) == 0;
|
|
}
|
|
|
|
#endif
|
|
|
|
#if defined(CPU_LE)
|
|
force_inline
|
|
int inline_memcmp_aligned(const char *a, const char *b, size_t len)
|
|
{
|
|
#ifdef CPU_64
|
|
while(len >= 8) {
|
|
uint64 a64 = *(uint64 *)a;
|
|
uint64 b64 = *(uint64 *)b;
|
|
if(a64 != b64)
|
|
return SwapEndian64(a64) < SwapEndian64(b64) ? -1 : 1;
|
|
a += 8;
|
|
b += 8;
|
|
len -= 8;
|
|
}
|
|
if(len & 4) {
|
|
uint32 a32 = *(uint32 *)a;
|
|
uint32 b32 = *(uint32 *)b;
|
|
if(a32 != b32)
|
|
return SwapEndian32(a32) < SwapEndian32(b32) ? -1 : 1;
|
|
a += 4;
|
|
b += 4;
|
|
}
|
|
#else
|
|
while(len >= 4) {
|
|
uint32 a32 = *(uint32 *)a;
|
|
uint32 b32 = *(uint32 *)b;
|
|
if(a32 != b32)
|
|
return SwapEndian32(a32) < SwapEndian32(b32) ? -1 : 1;
|
|
a += 4;
|
|
b += 4;
|
|
len -= 4;
|
|
}
|
|
#endif
|
|
if(len & 2) {
|
|
uint16 a16 = *(uint16 *)a;
|
|
uint16 b16 = *(uint16 *)b;
|
|
if(a16 != b16)
|
|
return SwapEndian16(a16) < SwapEndian16(b16) ? -1 : 1;
|
|
a += 2;
|
|
b += 2;
|
|
}
|
|
if((len & 1) != 0 && *a != *b)
|
|
return (byte)*a < (byte)*b ? -1 : 1;
|
|
return 0;
|
|
}
|
|
#else
|
|
inline
|
|
int inline_memcmp_aligned(const char *a, const char *b, size_t len)
|
|
{
|
|
return memcmp(a, b, len);
|
|
}
|
|
#endif
|
|
|
|
hash_t memhash(const void *ptr, size_t count);
|