weaselab/conflict-set
2
0
Fork 0
Code Issues 4 Packages Releases 13 Activity

Remove memmove call for inserting into Node16

Browse Source
This commit is contained in:
Andrew Noyes
2024-08-02 13:44:53 -07:00
parent 72168ef6a3
commit c8d9dc034d
1 changed files with 10 additions and 60 deletions
Download Patch File Download Diff File Expand all files Collapse all files

70
ConflictSet.cpp
View File

@@ -1148,71 +1148,21 @@ Node *&getOrCreateChild(Node *&self, uint8_t index, WriteContext *tls) {
}
insert16:
auto *self16 = static_cast<Node16 *>(self);
assert(self->getType() == Type_Node16);
++self->numChildren;
#ifdef HAS_AVX
__m128i key_vec = _mm_set1_epi8(index);
__m128i indices;
memcpy(&indices, self16->index, sizeof(self16->index));
__m128i results = _mm_cmpeq_epi8(key_vec, _mm_min_epu8(key_vec, indices));
int mask = (1 << (self->numChildren - 1)) - 1;
uint32_t bitfield = _mm_movemask_epi8(results) & mask;
bitfield |= uint32_t(1) << (self->numChildren - 1);
int i = std::countr_zero(bitfield);
if (i < self->numChildren - 1) {
memmove(self16->index + i + 1, self16->index + i,
self->numChildren - (i + 1));
memmove(self16->children + i + 1, self16->children + i,
(self->numChildren - (i + 1)) *
sizeof(self16->children[0])); // NOLINT
memmove(self16->childMaxVersion + i + 1, self16->childMaxVersion + i,
(self->numChildren - (i + 1)) *
sizeof(self16->childMaxVersion[0]));
}
#elif defined(HAS_ARM_NEON)
uint8x16_t indices;
memcpy(&indices, self16->index, sizeof(self16->index));
// 0xff for each leq
auto results = vcleq_u8(vdupq_n_u8(index), indices);
uint64_t mask = (uint64_t(1) << ((self->numChildren - 1) * 4)) - 1;
// 0xf for each 0xff (within mask)
uint64_t bitfield =
vget_lane_u64(
vreinterpret_u64_u8(vshrn_n_u16(vreinterpretq_u16_u8(results), 4)),
0) &
mask;
bitfield |= uint64_t(0xf) << ((self->numChildren - 1) * 4);
int i = std::countr_zero(bitfield) / 4;
if (i < self->numChildren - 1) {
memmove(self16->index + i + 1, self16->index + i,
self->numChildren - (i + 1));
memmove(self16->children + i + 1, self16->children + i,
(self->numChildren - (i + 1)) *
sizeof(self16->children[0])); // NOLINT
memmove(self16->childMaxVersion + i + 1, self16->childMaxVersion + i,
(self->numChildren - (i + 1)) *
sizeof(self16->childMaxVersion[0]));
}
#else
int i = 0;
for (; i < int(self->numChildren) - 1; ++i) {
if (int(self16->index[i]) > int(index)) {
memmove(self16->index + i + 1, self16->index + i,
self->numChildren - (i + 1));
memmove(self16->children + i + 1, self16->children + i,
(self->numChildren - (i + 1)) * sizeof(self16->children[0]));
memmove(self16->childMaxVersion + i + 1, self16->childMaxVersion + i,
(self->numChildren - (i + 1)) *
sizeof(self16->childMaxVersion[0]));
auto *self16 = static_cast<Node16 *>(self);
int i = self->numChildren - 1;
for (; i >= 0; --i) {
if (int(self16->index[i]) < int(index)) {
break;
}
self16->index[i + 1] = self16->index[i];
self16->children[i + 1] = self16->children[i];
self16->childMaxVersion[i + 1] = self16->childMaxVersion[i];
}
#endif
self16->index[i] = index;
auto &result = self16->children[i];
self16->index[i + 1] = index;
auto &result = self16->children[i + 1];
result = nullptr;
++self->numChildren;
return result;
}
case Type_Node48: {