[skip e2e]: update cardinal version (#1206)

* update cardinal version

Signed-off-by: xianliang.li <xianliang.li@zilliz.com>

* Revert "enhance: knowhere support metric mhjaccard and index minhash lsh for minhash vector (#1203)"

This reverts commit fd0871f.

Signed-off-by: xianliang.li <xianliang.li@zilliz.com>

---------

Signed-off-by: xianliang.li <xianliang.li@zilliz.com>

Author: foxspy

cmake/libs/libcardinal.cmake

@@ -1,5 +1,5 @@
 # Use short SHA1 as version
-set(CARDINAL_VERSION v2.5.8)
+set(CARDINAL_VERSION v2.5.9)
 set(CARDINAL_REPO_URL "https://github.com/zilliztech/cardinal.git")
 
 set(CARDINAL_REPO_DIR "${CMAKE_CURRENT_SOURCE_DIR}/thirdparty/cardinal")

cmake/libs/libfaiss.cmake

@@ -50,7 +50,6 @@ if(__X86_64)
     ${UTILS_SRC} $<TARGET_OBJECTS:utils_sse> $<TARGET_OBJECTS:utils_avx>
     $<TARGET_OBJECTS:utils_avx512> $<TARGET_OBJECTS:utils_avx512icx>)
   target_link_libraries(knowhere_utils PUBLIC glog::glog)
-  target_link_libraries(knowhere_utils PUBLIC xxHash::xxhash)
 endif()
 
 if(__AARCH64)
@@ -100,22 +99,19 @@ if(__AARCH64)
   endif()
 
   target_link_libraries(knowhere_utils PUBLIC glog::glog)
-  target_link_libraries(knowhere_utils PUBLIC xxHash::xxhash)
 endif()
 
 if(__RISCV64)
   set(UTILS_SRC src/simd/hook.cc src/simd/distances_ref.cc)
   add_library(knowhere_utils STATIC ${UTILS_SRC})
   target_link_libraries(knowhere_utils PUBLIC glog::glog)
-  target_link_libraries(knowhere_utils PUBLIC xxHash::xxhash)
 endif()
 
 # ToDo: Add distances_vsx.cc for powerpc64 SIMD acceleration
 if(__PPC64)
   set(UTILS_SRC src/simd/hook.cc src/simd/distances_ref.cc src/simd/distances_powerpc.cc)
   add_library(knowhere_utils STATIC ${UTILS_SRC})
   target_link_libraries(knowhere_utils PUBLIC glog::glog)
-  target_link_libraries(knowhere_utils PUBLIC xxHash::xxhash)
 endif()
 
 
@@ -135,9 +131,6 @@ else()
   find_package(BLAS REQUIRED)
 endif()
 
-find_package(xxHash REQUIRED)
-include_directories(${xxHash_INCLUDE_DIRS})
-
 if(__X86_64)
   list(REMOVE_ITEM FAISS_SRCS ${FAISS_AVX2_SRCS})
 

conanfile.py

@@ -106,7 +106,6 @@ def requirements(self):
         self.requires("folly/2023.10.30.09@milvus/dev")
         self.requires("libcurl/8.2.1")
         self.requires("simde/0.8.2")
-        self.requires("xxhash/0.8.2")
         if self.settings.os == "Android":
             self.requires("openblas/0.3.27")
         if not self.options.with_light:

include/knowhere/comp/bloomfilter.h

@@ -56,7 +56,6 @@ constexpr const char* INDEX_HNSW_PQ = "HNSW_PQ";
 constexpr const char* INDEX_HNSW_PRQ = "HNSW_PRQ";
 
 constexpr const char* INDEX_DISKANN = "DISKANN";
-constexpr const char* INDEX_MINHASH_INDEX = "MINHASH_LSH";
 
 constexpr const char* INDEX_SPARSE_INVERTED_INDEX = "SPARSE_INVERTED_INDEX";
 constexpr const char* INDEX_SPARSE_WAND = "SPARSE_WAND";
@@ -188,16 +187,6 @@ constexpr const char* DROP_RATIO_SEARCH = "drop_ratio_search";
 // RaBitQ Params
 constexpr const char* RABITQ_QUERY_BITS = "rbq_bits_query";
 
-// minhash Params
-constexpr const char* ALIGNED_BLOCK_SIZE = "aligned_block_size";
-constexpr const char* BAND = "band";
-constexpr const char* SHARED_BLOOM_FILTER = "shared_bloom_filter";
-constexpr const char* BLOOM_FALSE_POSITIVE_RPOB = "bloom_false_positive_prob";
-constexpr const char* HASH_CODE_IN_MEM = "hash_code_in_mem";
-constexpr const char* SEARCH_WITH_JACCARD = "search_with_jaccard";
-constexpr const char* MH_ELEMENT_BIT_WIDTH = "mh_element_bit_width";
-constexpr const char* MH_LSH_REFINE_K = "refine_k";
-constexpr const char* BATCH_SEARCH = "batch_search";
 }  // namespace indexparam
 
 using MetricType = std::string;
@@ -208,7 +197,6 @@ constexpr const char* L2 = "L2";
 constexpr const char* COSINE = "COSINE";
 constexpr const char* HAMMING = "HAMMING";
 constexpr const char* JACCARD = "JACCARD";
-constexpr const char* MHJACCARD = "MHJACCARD";
 constexpr const char* SUBSTRUCTURE = "SUBSTRUCTURE";
 constexpr const char* SUPERSTRUCTURE = "SUPERSTRUCTURE";
 constexpr const char* BM25 = "BM25";

include/knowhere/comp/index_param.h

@@ -633,12 +633,6 @@ class BaseConfig : public Config {
      */
     CFG_INT refine_type;
     CFG_BOOL refine_with_quant;
-    /*
-     * band is a special parameters of BF search and MinHash index node train.
-     */
-    CFG_INT band;
-    CFG_BOOL search_with_jaccard;
-    CFG_INT mh_element_bit_width;
     KNOHWERE_DECLARE_CONFIG(BaseConfig) {
         KNOWHERE_CONFIG_DECLARE_FIELD(dim).allow_empty_without_default().description("vector dim").for_train();
         KNOWHERE_CONFIG_DECLARE_FIELD(metric_type)
@@ -787,17 +781,6 @@ class BaseConfig : public Config {
             .for_search()
             .for_range_search()
             .for_iterator();
-        KNOWHERE_CONFIG_DECLARE_FIELD(band).description("param of MinHashLSH").set_default(1).for_train().for_search();
-        KNOWHERE_CONFIG_DECLARE_FIELD(mh_element_bit_width)
-            .description("sizeof(hash code), the hash element should be aligned on 8 bits")
-            .set_default(8)
-            .set_range(8, 256)
-            .for_train()
-            .for_search();
-        KNOWHERE_CONFIG_DECLARE_FIELD(search_with_jaccard)
-            .description("return the jaccard distance of minhash vector search or minhashlsh hit flag.")
-            .set_default(false)
-            .for_search();
     }
 };
 }  // namespace knowhere

include/knowhere/config.h

@@ -143,10 +143,6 @@ class IndexFactory {
 #define KNOWHERE_MOCK_REGISTER_DENSE_INT_GLOBAL(name, index_node, features, ...) \
     KNOWHERE_MOCK_REGISTER_GLOBAL(name, index_node, int8, (features | knowhere::feature::INT8), ##__VA_ARGS__);
 
-// register vector index supporting binary data types
-#define KNOWHERE_MOCK_REGISTER_DENSE_BINARY_ALL_GLOBAL(name, index_node, features, ...) \
-    KNOWHERE_SIMPLE_REGISTER_GLOBAL(name, index_node, bin1, (features | knowhere::feature::BINARY), ##__VA_ARGS__);
-
 // register vector index supporting ALL_DENSE_FLOAT_TYPE(float32, bf16, fp16) data types, but mocked bf16 and fp16
 #define KNOWHERE_MOCK_REGISTER_DENSE_FLOAT_ALL_GLOBAL(name, index_node, features, ...)                          \
     KNOWHERE_MOCK_REGISTER_GLOBAL(name, index_node, bf16, (features | knowhere::feature::BF16), ##__VA_ARGS__); \

include/knowhere/index/index_factory.h

@@ -98,8 +98,6 @@ static std::set<std::pair<std::string, VecType>> legal_knowhere_index = {
     // sparse index
     {IndexEnum::INDEX_SPARSE_INVERTED_INDEX, VecType::VECTOR_SPARSE_FLOAT},
     {IndexEnum::INDEX_SPARSE_WAND, VecType::VECTOR_SPARSE_FLOAT},
-    //  minhash index
-    {IndexEnum::INDEX_MINHASH_INDEX, VecType::VECTOR_BINARY},
 };
 
 static std::set<std::string> legal_support_mmap_knowhere_index = {

include/knowhere/index/index_table.h

@@ -14,7 +14,6 @@
 #include <strings.h>
 
 #include <algorithm>
-#include <fstream>
 #include <vector>
 
 #include "knowhere/binaryset.h"
@@ -227,25 +226,6 @@ readBinaryPOD(R& in, T& podRef) {
     in.read((char*)&podRef, sizeof(T));
 }
 
-inline void
-load_binary_vec(const std::string& bin_file, std::unique_ptr<char[]>& data, size_t& npts, size_t& dim) {
-    std::ifstream file(bin_file, std::ios::binary);
-    if (!file.is_open()) {
-        throw std::runtime_error("fail to open file: " + bin_file);
-    }
-    uint32_t n, d;
-    file.read(reinterpret_cast<char*>(&n), sizeof(uint32_t));
-    file.read(reinterpret_cast<char*>(&d), sizeof(uint32_t));
-    npts = n;
-    dim = d;
-    if (dim % 8 != 0) {
-        throw std::runtime_error("fail to load binary vector base file, dim % 8 != 0 ");
-    }
-    uint64_t total_size = dim * npts / 8;
-    data = std::make_unique<char[]>(total_size);
-    file.read(reinterpret_cast<char*>(data.get()), total_size);
-}
-
 // taken from
 // https://github.com/Microsoft/BLAS-on-flash/blob/master/include/utils.h
 // round up X to the nearest multiple of Y