database/doris
2
0
Fork 0
Code Issues Pull Requests Actions 3 Packages Projects Releases Wiki Activity

[Enhancement](function) optimize the upper and lower functions using the simd instruction. (#13326)

Browse Source 
optimize the `upper` and `lower` functions using the simd instruction.
This commit is contained in:
luozenglin
2022-11-03 15:12:25 +08:00
committed by GitHub
parent b1816d49e7
commit ee934483eb
1 changed files with 15 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files

25
be/src/vec/functions/function_string.cpp
View File

@ -245,21 +245,25 @@ struct NameToUpper {
static constexpr auto name = "upper";
};
using char_transter_op = int (*)(int);
template <char_transter_op op>
template <typename OpName>
struct TransferImpl {
static Status vector(const ColumnString::Chars& data, const ColumnString::Offsets& offsets,
ColumnString::Chars& res_data, ColumnString::Offsets& res_offsets) {
size_t offset_size = offsets.size();
res_offsets.resize(offsets.size());
for (size_t i = 0; i < offset_size; ++i) {
res_offsets[i] = offsets[i];
if (UNLIKELY(!offset_size)) {
return Status::OK();
}
res_offsets.resize(offset_size);
memcpy(res_offsets.data(), offsets.data(),
offset_size * sizeof(ColumnString::Offsets::value_type));
size_t data_length = data.size();
res_data.resize(data_length);
for (size_t i = 0; i < data_length; ++i) {
res_data[i] = op(data[i]);
if constexpr (std::is_same_v<OpName, NameToUpper>) {
simd::VStringFunctions::to_upper(data.data(), data_length, res_data.data());
} else if constexpr (std::is_same_v<OpName, NameToLower>) {
simd::VStringFunctions::to_lower(data.data(), data_length, res_data.data());
}
return Status::OK();
}
@ -275,7 +279,8 @@ struct InitcapImpl {
ColumnString::Chars& res_data, ColumnString::Offsets& res_offsets) {
size_t offset_size = offsets.size();
res_offsets.resize(offsets.size());
memcpy(res_offsets.data(), offsets.data(), offset_size * sizeof(offsets.data()));
memcpy(res_offsets.data(), offsets.data(),
offset_size * sizeof(ColumnString::Offsets::value_type));
size_t data_length = data.size();
res_data.resize(data_length);
@ -620,9 +625,9 @@ using FunctionStringFindInSet =
using FunctionUnHex = FunctionStringOperateToNullType<UnHexImpl>;
using FunctionToLower = FunctionStringToString<TransferImpl<::tolower>, NameToLower>;
using FunctionToLower = FunctionStringToString<TransferImpl<NameToLower>, NameToLower>;
using FunctionToUpper = FunctionStringToString<TransferImpl<::toupper>, NameToUpper>;
using FunctionToUpper = FunctionStringToString<TransferImpl<NameToUpper>, NameToUpper>;
using FunctionToInitcap = FunctionStringToString<InitcapImpl, NameToInitcap>;