thevar1able/ClickHouse
1
0
Fork 0
mirror of https://github.com/ClickHouse/ClickHouse.git synced 2024-11-23 16:12:01 +00:00
Code Issues Packages Projects Releases Wiki Activity

functions: vectorize bitHammingDistance

Browse Source
This commit is contained in:
ricky 2024-10-18 18:01:42 +08:00
parent 7fd2207626
commit 03534fa386
2 changed files with 34 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
contrib/CMakeLists.txt vendored
View File

@ -211,10 +211,10 @@ if (ARCH_S390X)
add_contrib(crc32-s390x-cmake crc32-s390x) add_contrib(crc32-s390x-cmake crc32-s390x)
endif() endif()
add_contrib (SimSIMD-cmake SimSIMD)
option(ENABLE_USEARCH "Enable USearch" ${ENABLE_LIBRARIES}) option(ENABLE_USEARCH "Enable USearch" ${ENABLE_LIBRARIES})
if (ENABLE_USEARCH) if (ENABLE_USEARCH)
add_contrib (FP16-cmake FP16) add_contrib (FP16-cmake FP16)
add_contrib (SimSIMD-cmake SimSIMD)
add_contrib (usearch-cmake usearch) # requires: FP16, SimdSIMD add_contrib (usearch-cmake usearch) # requires: FP16, SimdSIMD
else () else ()
message(STATUS "Not using USearch") message(STATUS "Not using USearch")

41
src/Functions/bitHammingDistance.cpp
View File

@ -1,6 +1,8 @@
#include <bit>
#include <Functions/FunctionBinaryArithmetic.h> #include <Functions/FunctionBinaryArithmetic.h>
#include <Functions/FunctionFactory.h> #include <Functions/FunctionFactory.h>
#include <bit> #pragma clang diagnostic ignored "-Wused-but-marked-unused"
#include <simsimd/simsimd.h>
namespace DB namespace DB
@ -26,17 +28,40 @@ struct BitHammingDistanceImpl
if constexpr (sizeof(A) <= sizeof(UInt64) && sizeof(B) <= sizeof(UInt64)) if constexpr (sizeof(A) <= sizeof(UInt64) && sizeof(B) <= sizeof(UInt64))
{ {
UInt64 res = static_cast<UInt64>(a) ^ static_cast<UInt64>(b); if constexpr (std::is_same_v<A, B>)
return std::popcount(res); {
simsimd_distance_t distance;
simsimd_hamming_b8(
reinterpret_cast<const simsimd_b8_t *>(&a), reinterpret_cast<const simsimd_b8_t *>(&b), sizeof(A), &distance);
return static_cast<UInt64>(distance);
}
else
{
UInt64 res = static_cast<UInt64>(a) ^ static_cast<UInt64>(b);
return std::popcount(res);
}
} }
else if constexpr (is_big_int_v<A> && is_big_int_v<B>) else if constexpr (is_big_int_v<A> && is_big_int_v<B>)
{ {
auto xored = a ^ b; if constexpr (std::is_same_v<A, B>)
{
simsimd_distance_t distance;
simsimd_hamming_b8(
reinterpret_cast<const simsimd_b8_t *>(&a.items),
reinterpret_cast<const simsimd_b8_t *>(&b.items),
sizeof(a.items),
&distance);
return static_cast<ResultType>(distance);
}
else
{
auto xored = a ^ b;
ResultType res = 0; ResultType res = 0;
for (auto item : xored.items) for (auto item : xored.items)
res += std::popcount(item); res += std::popcount(item);
return res; return res;
}
} }
else else
throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Unsupported data type combination in function 'bitHammingDistance'"); throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Unsupported data type combination in function 'bitHammingDistance'");