Merge remote-tracking branch 'upstream/master' into fix25

.gitmodules

@@ -93,10 +93,13 @@
 	url = https://github.com/ClickHouse-Extras/libunwind.git
 [submodule "contrib/simdjson"]
 	path = contrib/simdjson
-	url = https://github.com/lemire/simdjson.git
+	url = https://github.com/ClickHouse-Extras/simdjson.git
 [submodule "contrib/rapidjson"]
 	path = contrib/rapidjson
 	url = https://github.com/Tencent/rapidjson
 [submodule "contrib/mimalloc"]
 	path = contrib/mimalloc
 	url = https://github.com/ClickHouse-Extras/mimalloc
+[submodule "contrib/fastops"]
+	path = contrib/fastops
+	url = https://github.com/ClickHouse-Extras/fastops

CMakeLists.txt

@@ -476,6 +476,7 @@ include (cmake/find_hyperscan.cmake)
 include (cmake/find_mimalloc.cmake)
 include (cmake/find_simdjson.cmake)
 include (cmake/find_rapidjson.cmake)
+include (cmake/find_fastops.cmake)
 
 find_contrib_lib(cityhash)
 find_contrib_lib(farmhash)
@@ -568,4 +569,5 @@ if (GLIBC_COMPATIBILITY OR USE_INTERNAL_UNWIND_LIBRARY_FOR_EXCEPTION_HANDLING)
     add_default_dependencies(base64)
     add_default_dependencies(readpassphrase)
     add_default_dependencies(unwind_static)
+    add_default_dependencies(fastops)
 endif ()

cmake/find_fastops.cmake

@@ -0,0 +1,15 @@
+option (ENABLE_FASTOPS "Enable fast vectorized mathematical functions library by Michael Parakhin" ${NOT_UNBUNDLED})
+
+if (ENABLE_FASTOPS)
+    if(NOT EXISTS "${ClickHouse_SOURCE_DIR}/contrib/fastops/fastops/fastops.h")
+        message(FATAL_ERROR "submodule contrib/fastops is missing. to fix try run: \n git submodule update --init --recursive")
+        set(USE_FASTOPS 0)
+    endif()
+    set (USE_FASTOPS 1)
+    set (FASTOPS_INCLUDE_DIR ${ClickHouse_SOURCE_DIR}/contrib/fastops/)
+    set (FASTOPS_LIBRARY fastops)
+else ()
+    set(USE_FASTOPS 0)
+endif ()
+
+message (STATUS "Using fastops")

contrib/CMakeLists.txt

@@ -330,3 +330,7 @@ endif()
 if (USE_MIMALLOC)
     add_subdirectory (mimalloc)
 endif()
+
+if (USE_FASTOPS)
+    add_subdirectory (fastops-cmake)
+endif()

contrib/fastops

@@ -0,0 +1 @@
+Subproject commit d2c85c5d6549cfd648a7f31ef7b14341881ff8ae

contrib/fastops-cmake/CMakeLists.txt

@@ -0,0 +1,20 @@
+set(LIBRARY_DIR ${ClickHouse_SOURCE_DIR}/contrib/fastops)
+
+set(SRCS "")
+
+if(HAVE_AVX)
+    set (SRCS ${SRCS} ${LIBRARY_DIR}/fastops/avx/ops_avx.cpp ${LIBRARY_DIR}/fastops/core/FastIntrinsics.cpp)
+    set_source_files_properties(${LIBRARY_DIR}/fastops/avx/ops_avx.cpp PROPERTIES COMPILE_FLAGS "-mavx -DNO_AVX2")
+    set_source_files_properties(${LIBRARY_DIR}/fastops/core/FastIntrinsics.cpp PROPERTIES COMPILE_FLAGS "-mavx -DNO_AVX2")
+endif()
+
+if(HAVE_AVX2)
+    set (SRCS ${SRCS} ${LIBRARY_DIR}/fastops/avx2/ops_avx2.cpp)
+    set_source_files_properties(${LIBRARY_DIR}/fastops/avx2/ops_avx2.cpp PROPERTIES COMPILE_FLAGS "-mavx2 -mfma")
+endif()
+
+set (SRCS ${SRCS} ${LIBRARY_DIR}/fastops/plain/ops_plain.cpp ${LIBRARY_DIR}/fastops/core/avx_id.cpp ${LIBRARY_DIR}/fastops/fastops.cpp)
+
+add_library(fastops ${SRCS})
+
+target_include_directories(fastops SYSTEM PUBLIC "${LIBRARY_DIR}")

contrib/simdjson

@@ -1 +1 @@
-Subproject commit 3bd3116cf8faf6d482dc31423b16533bfa2696f7
+Subproject commit e3f6322af762213ff2087ce3366bf9541c7fd355

dbms/src/DataStreams/PushingToViewsBlockOutputStream.cpp

@@ -10,7 +10,7 @@
 #include <Common/getNumberOfPhysicalCPUCores.h>
 #include <Common/ThreadPool.h>
 #include <Storages/MergeTree/ReplicatedMergeTreeBlockOutputStream.h>
-#include <Storages/StorageBlockInserted.h>
+#include <Storages/StorageValues.h>
 
 namespace DB
 {
@@ -183,7 +183,7 @@ void PushingToViewsBlockOutputStream::process(const Block & block, size_t view_n
         ///  but it will contain single block (that is INSERT-ed into main table).
         /// InterpreterSelectQuery will do processing of alias columns.
         Context local_context = *views_context;
-        local_context.addViewSource(StorageBlockInserted::create(block, storage));
+        local_context.addViewSource(StorageValues::create(storage->getDatabaseName(), storage->getTableName(), storage->getColumns(), block));
         InterpreterSelectQuery select(view.query, local_context, SelectQueryOptions());
 
         BlockInputStreamPtr in = std::make_shared<MaterializingBlockInputStream>(select.execute().in);

dbms/src/Functions/CMakeLists.txt

@@ -20,6 +20,7 @@ target_link_libraries(clickhouse_functions
         ${METROHASH_LIBRARIES}
         murmurhash
         ${BASE64_LIBRARY}
+        ${FASTOPS_LIBRARY}
 
     PRIVATE
         ${ZLIB_LIBRARIES}
@@ -31,7 +32,7 @@ if (OPENSSL_CRYPTO_LIBRARY)
 endif()
 
 target_include_directories(clickhouse_functions PRIVATE ${CMAKE_CURRENT_BINARY_DIR}/include)
-target_include_directories(clickhouse_functions SYSTEM BEFORE PUBLIC ${DIVIDE_INCLUDE_DIR} ${METROHASH_INCLUDE_DIR})
+target_include_directories(clickhouse_functions SYSTEM PRIVATE ${DIVIDE_INCLUDE_DIR} ${METROHASH_INCLUDE_DIR})
 
 if (CONSISTENT_HASHING_INCLUDE_DIR)
     target_include_directories (clickhouse_functions PRIVATE ${CONSISTENT_HASHING_INCLUDE_DIR})
@@ -48,7 +49,11 @@ if (USE_ICU)
 endif ()
 
 if (USE_VECTORCLASS)
-    target_include_directories (clickhouse_functions SYSTEM BEFORE PUBLIC ${VECTORCLASS_INCLUDE_DIR})
+    target_include_directories (clickhouse_functions SYSTEM PRIVATE ${VECTORCLASS_INCLUDE_DIR})
+endif ()
+
+if (USE_FASTOPS)
+    target_include_directories (clickhouse_functions SYSTEM PRIVATE ${FASTOPS_INCLUDE_DIR})
 endif ()
 
 if (ENABLE_TESTS)

dbms/src/Functions/FunctionMathUnary.h

@@ -31,6 +31,14 @@
 #endif
 
 
+/** FastOps is a fast vector math library from Michael Parakhin (former Yandex CTO),
+  * Enabled by default.
+  */
+#if USE_FASTOPS
+#include <fastops/fastops.h>
+#endif
+
+
 namespace DB
 {
 
@@ -41,16 +49,14 @@ namespace ErrorCodes
 
 
 template <typename Impl>
-class FunctionMathUnaryFloat64 : public IFunction
+class FunctionMathUnary : public IFunction
 {
 public:
     static constexpr auto name = Impl::name;
-    static FunctionPtr create(const Context &) { return std::make_shared<FunctionMathUnaryFloat64>(); }
+    static FunctionPtr create(const Context &) { return std::make_shared<FunctionMathUnary>(); }
-    static_assert(Impl::rows_per_iteration > 0, "Impl must process at least one row per iteration");
 
 private:
     String getName() const override { return name; }
-
     size_t getNumberOfArguments() const override { return 1; }
 
     DataTypePtr getReturnTypeImpl(const DataTypes & arguments) const override
@@ -60,38 +66,63 @@ private:
             throw Exception{"Illegal type " + arg->getName() + " of argument of function " + getName(),
                 ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT};
 
-        return std::make_shared<DataTypeFloat64>();
+        /// Integers are converted to Float64.
+        if (Impl::always_returns_float64 || !isFloat(arg))
+            return std::make_shared<DataTypeFloat64>();
+        else
+            return arg;
     }
 
-    template <typename T>
+    template <typename T, typename ReturnType>
-    static void executeInIterations(const T * src_data, Float64 * dst_data, size_t size)
+    static void executeInIterations(const T * src_data, ReturnType * dst_data, size_t size)
     {
-        const size_t rows_remaining = size % Impl::rows_per_iteration;
+        if constexpr (Impl::rows_per_iteration == 0)
-        const size_t rows_size = size - rows_remaining;
-
-        for (size_t i = 0; i < rows_size; i += Impl::rows_per_iteration)
-            Impl::execute(&src_data[i], &dst_data[i]);
-
-        if (rows_remaining != 0)
         {
-            T src_remaining[Impl::rows_per_iteration];
+            /// Process all data as a whole and use FastOps implementation
-            memcpy(src_remaining, &src_data[rows_size], rows_remaining * sizeof(T));
-            memset(src_remaining + rows_remaining, 0, (Impl::rows_per_iteration - rows_remaining) * sizeof(T));
-            Float64 dst_remaining[Impl::rows_per_iteration];
 
-            Impl::execute(src_remaining, dst_remaining);
+            /// If the argument is integer, convert to Float64 beforehand
+            if constexpr (!std::is_floating_point_v<T>)
+            {
+                PODArray<Float64> tmp_vec(size);
+                for (size_t i = 0; i < size; ++i)
+                    tmp_vec[i] = src_data[i];
 
-            memcpy(&dst_data[rows_size], dst_remaining, rows_remaining * sizeof(Float64));
+                Impl::execute(tmp_vec.data(), size, dst_data);
+            }
+            else
+            {
+                Impl::execute(src_data, size, dst_data);
+            }
+        }
+        else
+        {
+            const size_t rows_remaining = size % Impl::rows_per_iteration;
+            const size_t rows_size = size - rows_remaining;
+
+            for (size_t i = 0; i < rows_size; i += Impl::rows_per_iteration)
+                Impl::execute(&src_data[i], &dst_data[i]);
+
+            if (rows_remaining != 0)
+            {
+                T src_remaining[Impl::rows_per_iteration];
+                memcpy(src_remaining, &src_data[rows_size], rows_remaining * sizeof(T));
+                memset(src_remaining + rows_remaining, 0, (Impl::rows_per_iteration - rows_remaining) * sizeof(T));
+                ReturnType dst_remaining[Impl::rows_per_iteration];
+
+                Impl::execute(src_remaining, dst_remaining);
+
+                memcpy(&dst_data[rows_size], dst_remaining, rows_remaining * sizeof(ReturnType));
+            }
         }
     }
 
-    template <typename T>
+    template <typename T, typename ReturnType>
     static bool execute(Block & block, const ColumnVector<T> * col, const size_t result)
     {
         const auto & src_data = col->getData();
         const size_t size = src_data.size();
 
-        auto dst = ColumnVector<Float64>::create();
+        auto dst = ColumnVector<ReturnType>::create();
         auto & dst_data = dst->getData();
         dst_data.resize(size);
 
@@ -101,19 +132,19 @@ private:
         return true;
     }
 
-    template <typename T>
+    template <typename T, typename ReturnType>
     static bool execute(Block & block, const ColumnDecimal<T> * col, const size_t result)
     {
         const auto & src_data = col->getData();
         const size_t size = src_data.size();
         UInt32 scale = src_data.getScale();
 
-        auto dst = ColumnVector<Float64>::create();
+        auto dst = ColumnVector<ReturnType>::create();
         auto & dst_data = dst->getData();
         dst_data.resize(size);
 
         for (size_t i = 0; i < size; ++i)
-            dst_data[i] = convertFromDecimal<DataTypeDecimal<T>, DataTypeNumber<Float64>>(src_data[i], scale);
+            dst_data[i] = convertFromDecimal<DataTypeDecimal<T>, DataTypeNumber<ReturnType>>(src_data[i], scale);
 
         executeInIterations(dst_data.data(), dst_data.data(), size);
 
@@ -131,10 +162,11 @@ private:
         {
             using Types = std::decay_t<decltype(types)>;
             using Type = typename Types::RightType;
+            using ReturnType = std::conditional_t<Impl::always_returns_float64 || !std::is_floating_point_v<Type>, Float64, Type>;
             using ColVecType = std::conditional_t<IsDecimalNumber<Type>, ColumnDecimal<Type>, ColumnVector<Type>>;
 
             const auto col_vec = checkAndGetColumn<ColVecType>(col.column.get());
-            return execute<Type>(block, col_vec, result);
+            return execute<Type, ReturnType>(block, col_vec, result);
         };
 
         if (!callOnBasicType<void, true, true, true, false>(col.type->getTypeId(), call))
@@ -149,6 +181,7 @@ struct UnaryFunctionPlain
 {
     static constexpr auto name = Name::name;
     static constexpr auto rows_per_iteration = 1;
+    static constexpr bool always_returns_float64 = true;
 
     template <typename T>
     static void execute(const T * src, Float64 * dst)
@@ -164,6 +197,7 @@ struct UnaryFunctionVectorized
 {
     static constexpr auto name = Name::name;
     static constexpr auto rows_per_iteration = 2;
+    static constexpr bool always_returns_float64 = true;
 
     template <typename T>
     static void execute(const T * src, Float64 * dst)

dbms/src/Functions/SimdJSONParser.h

@@ -25,14 +25,14 @@ struct SimdJSONParser
 
     void preallocate(size_t max_size)
     {
-        if (!pj.allocateCapacity(max_size))
+        if (!pj.allocate_capacity(max_size))
             throw Exception{"Can not allocate memory for " + std::to_string(max_size) + " units when parsing JSON",
                             ErrorCodes::CANNOT_ALLOCATE_MEMORY};
     }
 
     bool parse(const StringRef & json) { return !json_parse(json.data, json.size, pj); }
 
-    using Iterator = simdjson::ParsedJson::iterator;
+    using Iterator = simdjson::ParsedJson::Iterator;
     Iterator getRoot() { return Iterator{pj}; }
 
     static bool isInt64(const Iterator & it) { return it.is_integer(); }

dbms/src/Functions/acos.cpp

@@ -1,11 +1,11 @@
-#include <Functions/FunctionMathUnaryFloat64.h>
+#include <Functions/FunctionMathUnary.h>
 #include <Functions/FunctionFactory.h>
 
 namespace DB
 {
 
 struct AcosName { static constexpr auto name = "acos"; };
-using FunctionAcos = FunctionMathUnaryFloat64<UnaryFunctionVectorized<AcosName, acos>>;
+using FunctionAcos = FunctionMathUnary<UnaryFunctionVectorized<AcosName, acos>>;
 
 void registerFunctionAcos(FunctionFactory & factory)
 {

dbms/src/Functions/asin.cpp

@@ -1,11 +1,11 @@
-#include <Functions/FunctionMathUnaryFloat64.h>
+#include <Functions/FunctionMathUnary.h>
 #include <Functions/FunctionFactory.h>
 
 namespace DB
 {
 
 struct AsinName { static constexpr auto name = "asin"; };
-using FunctionAsin = FunctionMathUnaryFloat64<UnaryFunctionVectorized<AsinName, asin>>;
+using FunctionAsin = FunctionMathUnary<UnaryFunctionVectorized<AsinName, asin>>;
 
 void registerFunctionAsin(FunctionFactory & factory)
 {

dbms/src/Functions/atan.cpp

@@ -1,11 +1,11 @@
-#include <Functions/FunctionMathUnaryFloat64.h>
+#include <Functions/FunctionMathUnary.h>
 #include <Functions/FunctionFactory.h>
 
 namespace DB
 {
 
 struct AtanName { static constexpr auto name = "atan"; };
-using FunctionAtan = FunctionMathUnaryFloat64<UnaryFunctionVectorized<AtanName, atan>>;
+using FunctionAtan = FunctionMathUnary<UnaryFunctionVectorized<AtanName, atan>>;
 
 void registerFunctionAtan(FunctionFactory & factory)
 {

dbms/src/Functions/cbrt.cpp

@@ -1,11 +1,11 @@
-#include <Functions/FunctionMathUnaryFloat64.h>
+#include <Functions/FunctionMathUnary.h>
 #include <Functions/FunctionFactory.h>
 
 namespace DB
 {
 
 struct CbrtName { static constexpr auto name = "cbrt"; };
-using FunctionCbrt = FunctionMathUnaryFloat64<UnaryFunctionVectorized<CbrtName, cbrt>>;
+using FunctionCbrt = FunctionMathUnary<UnaryFunctionVectorized<CbrtName, cbrt>>;
 
 void registerFunctionCbrt(FunctionFactory & factory)
 {

dbms/src/Functions/config_functions.h.in

@@ -9,3 +9,4 @@
 #cmakedefine01 USE_SIMDJSON
 #cmakedefine01 USE_RAPIDJSON
 #cmakedefine01 USE_H3
+#cmakedefine01 USE_FASTOPS

dbms/src/Functions/cos.cpp

@@ -1,11 +1,11 @@
-#include <Functions/FunctionMathUnaryFloat64.h>
+#include <Functions/FunctionMathUnary.h>
 #include <Functions/FunctionFactory.h>
 
 namespace DB
 {
 
 struct CosName { static constexpr auto name = "cos"; };
-using FunctionCos = FunctionMathUnaryFloat64<UnaryFunctionVectorized<CosName, cos>>;
+using FunctionCos = FunctionMathUnary<UnaryFunctionVectorized<CosName, cos>>;
 
 void registerFunctionCos(FunctionFactory & factory)
 {

dbms/src/Functions/erf.cpp

@@ -1,11 +1,11 @@
-#include <Functions/FunctionMathUnaryFloat64.h>
+#include <Functions/FunctionMathUnary.h>
 #include <Functions/FunctionFactory.h>
 
 namespace DB
 {
 
 struct ErfName { static constexpr auto name = "erf"; };
-using FunctionErf = FunctionMathUnaryFloat64<UnaryFunctionPlain<ErfName, std::erf>>;
+using FunctionErf = FunctionMathUnary<UnaryFunctionPlain<ErfName, std::erf>>;
 
 void registerFunctionErf(FunctionFactory & factory)
 {

dbms/src/Functions/erfc.cpp

@@ -1,11 +1,11 @@
-#include <Functions/FunctionMathUnaryFloat64.h>
+#include <Functions/FunctionMathUnary.h>
 #include <Functions/FunctionFactory.h>
 
 namespace DB
 {
 
 struct ErfcName { static constexpr auto name = "erfc"; };
-using FunctionErfc = FunctionMathUnaryFloat64<UnaryFunctionPlain<ErfcName, std::erfc>>;
+using FunctionErfc = FunctionMathUnary<UnaryFunctionPlain<ErfcName, std::erfc>>;
 
 void registerFunctionErfc(FunctionFactory & factory)
 {

dbms/src/Functions/exp.cpp

@@ -1,11 +1,34 @@
-#include <Functions/FunctionMathUnaryFloat64.h>
+#include <Functions/FunctionMathUnary.h>
 #include <Functions/FunctionFactory.h>
 
 namespace DB
 {
 
 struct ExpName { static constexpr auto name = "exp"; };
-using FunctionExp = FunctionMathUnaryFloat64<UnaryFunctionVectorized<ExpName, exp>>;
+
+#if USE_FASTOPS
+
+namespace
+{
+    struct Impl
+    {
+        static constexpr auto name = ExpName::name;
+        static constexpr auto rows_per_iteration = 0;
+        static constexpr bool always_returns_float64 = false;
+
+        template <typename T>
+        static void execute(const T * src, size_t size, T * dst)
+        {
+            NFastOps::Exp<true>(src, size, dst);
+        }
+    };
+}
+
+using FunctionExp = FunctionMathUnary<Impl>;
+
+#else
+using FunctionExp = FunctionMathUnary<UnaryFunctionVectorized<ExpName, exp>>;
+#endif
 
 void registerFunctionExp(FunctionFactory & factory)
 {

dbms/src/Functions/exp10.cpp

@@ -1,4 +1,4 @@
-#include <Functions/FunctionMathUnaryFloat64.h>
+#include <Functions/FunctionMathUnary.h>
 #include <Functions/FunctionFactory.h>
 #if !USE_VECTORCLASS
 #   include <common/preciseExp10.h>
@@ -9,7 +9,7 @@ namespace DB
 
 struct Exp10Name { static constexpr auto name = "exp10"; };
 
-using FunctionExp10 = FunctionMathUnaryFloat64<UnaryFunctionVectorized<Exp10Name,
+using FunctionExp10 = FunctionMathUnary<UnaryFunctionVectorized<Exp10Name,
 #if USE_VECTORCLASS
     exp10
 #else

dbms/src/Functions/exp2.cpp

@@ -1,11 +1,11 @@
-#include <Functions/FunctionMathUnaryFloat64.h>
+#include <Functions/FunctionMathUnary.h>
 #include <Functions/FunctionFactory.h>
 
 namespace DB
 {
 
 struct Exp2Name { static constexpr auto name = "exp2"; };
-using FunctionExp2 = FunctionMathUnaryFloat64<UnaryFunctionVectorized<Exp2Name, exp2>>;
+using FunctionExp2 = FunctionMathUnary<UnaryFunctionVectorized<Exp2Name, exp2>>;
 
 void registerFunctionExp2(FunctionFactory & factory)
 {

dbms/src/Functions/lgamma.cpp

@@ -1,11 +1,11 @@
-#include <Functions/FunctionMathUnaryFloat64.h>
+#include <Functions/FunctionMathUnary.h>
 #include <Functions/FunctionFactory.h>
 
 namespace DB
 {
 
 struct LGammaName { static constexpr auto name = "lgamma"; };
-using FunctionLGamma = FunctionMathUnaryFloat64<UnaryFunctionPlain<LGammaName, std::lgamma>>;
+using FunctionLGamma = FunctionMathUnary<UnaryFunctionPlain<LGammaName, std::lgamma>>;
 
 void registerFunctionLGamma(FunctionFactory & factory)
 {

dbms/src/Functions/log.cpp

@@ -1,11 +1,34 @@
-#include <Functions/FunctionMathUnaryFloat64.h>
+#include <Functions/FunctionMathUnary.h>
 #include <Functions/FunctionFactory.h>
 
 namespace DB
 {
 
 struct LogName { static constexpr auto name = "log"; };
-using FunctionLog = FunctionMathUnaryFloat64<UnaryFunctionVectorized<LogName, log>>;
+
+#if USE_FASTOPS
+
+namespace
+{
+    struct Impl
+    {
+        static constexpr auto name = LogName::name;
+        static constexpr auto rows_per_iteration = 0;
+        static constexpr bool always_returns_float64 = false;
+
+        template <typename T>
+        static void execute(const T * src, size_t size, T * dst)
+        {
+            NFastOps::Log<true>(src, size, dst);
+        }
+    };
+}
+
+using FunctionLog = FunctionMathUnary<Impl>;
+
+#else
+using FunctionLog = FunctionMathUnary<UnaryFunctionVectorized<LogName, log>>;
+#endif
 
 void registerFunctionLog(FunctionFactory & factory)
 {

dbms/src/Functions/log10.cpp

@@ -1,11 +1,11 @@
-#include <Functions/FunctionMathUnaryFloat64.h>
+#include <Functions/FunctionMathUnary.h>
 #include <Functions/FunctionFactory.h>
 
 namespace DB
 {
 
 struct Log10Name { static constexpr auto name = "log10"; };
-using FunctionLog10 = FunctionMathUnaryFloat64<UnaryFunctionVectorized<Log10Name, log10>>;
+using FunctionLog10 = FunctionMathUnary<UnaryFunctionVectorized<Log10Name, log10>>;
 
 void registerFunctionLog10(FunctionFactory & factory)
 {

dbms/src/Functions/log2.cpp

@@ -1,11 +1,11 @@
-#include <Functions/FunctionMathUnaryFloat64.h>
+#include <Functions/FunctionMathUnary.h>
 #include <Functions/FunctionFactory.h>
 
 namespace DB
 {
 
 struct Log2Name { static constexpr auto name = "log2"; };
-using FunctionLog2 = FunctionMathUnaryFloat64<UnaryFunctionVectorized<Log2Name, log2>>;
+using FunctionLog2 = FunctionMathUnary<UnaryFunctionVectorized<Log2Name, log2>>;
 
 void registerFunctionLog2(FunctionFactory & factory)
 {

dbms/src/Functions/registerFunctionsMath.cpp

@@ -23,6 +23,8 @@ void registerFunctionTan(FunctionFactory & factory);
 void registerFunctionAsin(FunctionFactory & factory);
 void registerFunctionAcos(FunctionFactory & factory);
 void registerFunctionAtan(FunctionFactory & factory);
+void registerFunctionSigmoid(FunctionFactory & factory);
+void registerFunctionTanh(FunctionFactory & factory);
 void registerFunctionPow(FunctionFactory & factory);
 
 void registerFunctionsMath(FunctionFactory & factory)
@@ -47,6 +49,8 @@ void registerFunctionsMath(FunctionFactory & factory)
     registerFunctionAsin(factory);
     registerFunctionAcos(factory);
     registerFunctionAtan(factory);
+    registerFunctionSigmoid(factory);
+    registerFunctionTanh(factory);
     registerFunctionPow(factory);
 }
 

dbms/src/Functions/sigmoid.cpp

@@ -0,0 +1,46 @@
+#include <Functions/FunctionMathUnary.h>
+#include <Functions/FunctionFactory.h>
+
+namespace DB
+{
+
+struct SigmoidName { static constexpr auto name = "sigmoid"; };
+
+#if USE_FASTOPS
+
+namespace
+{
+    struct Impl
+    {
+        static constexpr auto name = SigmoidName::name;
+        static constexpr auto rows_per_iteration = 0;
+        static constexpr bool always_returns_float64 = false;
+
+        template <typename T>
+        static void execute(const T * src, size_t size, T * dst)
+        {
+            NFastOps::Sigmoid<>(src, size, dst);
+        }
+    };
+}
+
+using FunctionSigmoid = FunctionMathUnary<Impl>;
+
+#else
+
+static double sigmoid(double x)
+{
+    return 1.0 / (1.0 + exp(-x));
+}
+
+using FunctionSigmoid = FunctionMathUnary<UnaryFunctionVectorized<SigmoidName, sigmoid>>;
+
+#endif
+
+void registerFunctionSigmoid(FunctionFactory & factory)
+{
+    factory.registerFunction<FunctionSigmoid>();
+}
+
+}
+

dbms/src/Functions/sin.cpp

@@ -1,11 +1,11 @@
-#include <Functions/FunctionMathUnaryFloat64.h>
+#include <Functions/FunctionMathUnary.h>
 #include <Functions/FunctionFactory.h>
 
 namespace DB
 {
 
 struct SinName { static constexpr auto name = "sin"; };
-using FunctionSin = FunctionMathUnaryFloat64<UnaryFunctionVectorized<SinName, sin>>;
+using FunctionSin = FunctionMathUnary<UnaryFunctionVectorized<SinName, sin>>;
 
 void registerFunctionSin(FunctionFactory & factory)
 {

dbms/src/Functions/sqrt.cpp

@@ -1,11 +1,11 @@
-#include <Functions/FunctionMathUnaryFloat64.h>
+#include <Functions/FunctionMathUnary.h>
 #include <Functions/FunctionFactory.h>
 
 namespace DB
 {
 
 struct SqrtName { static constexpr auto name = "sqrt"; };
-using FunctionSqrt = FunctionMathUnaryFloat64<UnaryFunctionVectorized<SqrtName, sqrt>>;
+using FunctionSqrt = FunctionMathUnary<UnaryFunctionVectorized<SqrtName, sqrt>>;
 
 void registerFunctionSqrt(FunctionFactory & factory)
 {

dbms/src/Functions/tan.cpp

@@ -1,11 +1,11 @@
-#include <Functions/FunctionMathUnaryFloat64.h>
+#include <Functions/FunctionMathUnary.h>
 #include <Functions/FunctionFactory.h>
 
 namespace DB
 {
 
 struct TanName { static constexpr auto name = "tan"; };
-using FunctionTan = FunctionMathUnaryFloat64<UnaryFunctionVectorized<TanName, tan>>;
+using FunctionTan = FunctionMathUnary<UnaryFunctionVectorized<TanName, tan>>;
 
 void registerFunctionTan(FunctionFactory & factory)
 {

dbms/src/Functions/tanh.cpp

@@ -0,0 +1,46 @@
+#include <Functions/FunctionMathUnary.h>
+#include <Functions/FunctionFactory.h>
+
+namespace DB
+{
+
+struct TanhName { static constexpr auto name = "tanh"; };
+
+#if USE_FASTOPS
+
+namespace
+{
+    struct Impl
+    {
+        static constexpr auto name = TanhName::name;
+        static constexpr auto rows_per_iteration = 0;
+        static constexpr bool always_returns_float64 = false;
+
+        template <typename T>
+        static void execute(const T * src, size_t size, T * dst)
+        {
+            NFastOps::Tanh<>(src, size, dst);
+        }
+    };
+}
+
+using FunctionTanh = FunctionMathUnary<Impl>;
+
+#else
+
+static double tanh(double x)
+{
+    return 2 / (1.0 + exp(-2 * x)) - 1;
+}
+
+using FunctionTanh = FunctionMathUnary<UnaryFunctionVectorized<TanhName, tanh>>;
+#endif
+
+void registerFunctionTanh(FunctionFactory & factory)
+{
+    factory.registerFunction<FunctionTanh>(FunctionFactory::CaseInsensitive);
+}
+
+}
+
+

dbms/src/Functions/tgamma.cpp

@@ -1,11 +1,11 @@
-#include <Functions/FunctionMathUnaryFloat64.h>
+#include <Functions/FunctionMathUnary.h>
 #include <Functions/FunctionFactory.h>
 
 namespace DB
 {
 
 struct TGammaName { static constexpr auto name = "tgamma"; };
-using FunctionTGamma = FunctionMathUnaryFloat64<UnaryFunctionPlain<TGammaName, std::tgamma>>;
+using FunctionTGamma = FunctionMathUnary<UnaryFunctionPlain<TGammaName, std::tgamma>>;
 
 void registerFunctionTGamma(FunctionFactory & factory)
 {

dbms/src/Interpreters/Context.h

@@ -133,7 +133,7 @@ private:
     String default_format;  /// Format, used when server formats data by itself and if query does not have FORMAT specification.
                             /// Thus, used in HTTP interface. If not specified - then some globally default format is used.
     TableAndCreateASTs external_tables;     /// Temporary tables.
-    StoragePtr view_source;                 /// Temporary StorageBlockInserted used to generate alias columns for materialized views
+    StoragePtr view_source;                 /// Temporary StorageValues used to generate alias columns for materialized views
     Tables table_function_results;          /// Temporary tables obtained by execution of table functions. Keyed by AST tree id.
     Context * query_context = nullptr;
     Context * session_context = nullptr;    /// Session context or nullptr. Could be equal to this.

dbms/src/Interpreters/InterpreterSelectQuery.cpp

@@ -48,9 +48,7 @@
 #include <Storages/MergeTree/MergeTreeData.h>
 #include <Storages/MergeTree/MergeTreeWhereOptimizer.h>
 #include <Storages/IStorage.h>
-#include <Storages/StorageBlockInserted.h>
+#include <Storages/StorageValues.h>
-#include <Storages/StorageMergeTree.h>
-#include <Storages/StorageReplicatedMergeTree.h>
 
 #include <TableFunctions/ITableFunction.h>
 #include <TableFunctions/TableFunctionFactory.h>
@@ -274,8 +272,8 @@ InterpreterSelectQuery::InterpreterSelectQuery(
 
             if (auto view_source = context.getViewSource())
             {
-                auto & storage_block = static_cast<const StorageBlockInserted &>(*view_source);
+                auto & storage_values = static_cast<const StorageValues &>(*view_source);
-                if (storage_block.getDatabaseName() == database_name && storage_block.getTableName() == table_name)
+                if (storage_values.getDatabaseName() == database_name && storage_values.getTableName() == table_name)
                 {
                     /// Read from view source.
                     storage = context.getViewSource();

dbms/src/Storages/StorageBlock.h

@@ -1,59 +0,0 @@
-#pragma once
-
-#include <ext/shared_ptr_helper.h>
-
-#include <Core/NamesAndTypes.h>
-#include <DataStreams/OneBlockInputStream.h>
-#include <Storages/IStorage.h>
-
-
-namespace DB
-{
-/// The table has all the properties of another storage,
-///  but will read single prepared block instead.
-/// Used in PushingToViewsBlockOutputStream for generating alias columns
-/// NOTE: Some of the properties seems redundant.
-class StorageBlock : public ext::shared_ptr_helper<StorageBlock>, public IStorage
-{
-public:
-    std::string getName() const override { return storage->getName(); }
-    std::string getTableName() const override { return storage->getTableName(); }
-    std::string getDatabaseName() const override { return storage->getDatabaseName(); }
-    bool isRemote() const override { return storage->isRemote(); }
-    bool supportsSampling() const override { return storage->supportsSampling(); }
-    bool supportsFinal() const override { return storage->supportsFinal(); }
-    bool supportsPrewhere() const override { return storage->supportsPrewhere(); }
-    bool supportsReplication() const override { return storage->supportsReplication(); }
-    bool supportsDeduplication() const override { return storage->supportsDeduplication(); }
-    bool supportsIndexForIn() const override { return storage->supportsIndexForIn(); }
-    bool mayBenefitFromIndexForIn(const ASTPtr & left_in_operand, const Context & query_context) const override
-    {
-        return storage->mayBenefitFromIndexForIn(left_in_operand, query_context);
-    }
-    ASTPtr getPartitionKeyAST() const override { return storage->getPartitionKeyAST(); }
-    ASTPtr getSortingKeyAST() const override { return storage->getSortingKeyAST(); }
-    ASTPtr getPrimaryKeyAST() const override { return storage->getPrimaryKeyAST(); }
-    ASTPtr getSamplingKeyAST() const override { return storage->getSamplingKeyAST(); }
-    Names getColumnsRequiredForPartitionKey() const override { return storage->getColumnsRequiredForPartitionKey(); }
-    Names getColumnsRequiredForSortingKey() const override { return storage->getColumnsRequiredForSortingKey(); }
-    Names getColumnsRequiredForPrimaryKey() const override { return storage->getColumnsRequiredForPrimaryKey(); }
-    Names getColumnsRequiredForSampling() const override { return storage->getColumnsRequiredForSampling(); }
-    Names getColumnsRequiredForFinal() const override { return storage->getColumnsRequiredForFinal(); }
-
-    BlockInputStreams read(const Names &, const SelectQueryInfo &, const Context &, QueryProcessingStage::Enum, size_t, unsigned) override
-    {
-        return {std::make_shared<OneBlockInputStream>(std::move(block))};
-    }
-
-private:
-    Block block;
-    StoragePtr storage;
-
-protected:
-    StorageBlock(Block block_, StoragePtr storage_)
-        : IStorage{storage_->getColumns()}, block(std::move(block_)), storage(storage_)
-    {
-    }
-};
-
-}

dbms/src/Storages/StorageBlockInserted.h

@@ -1,63 +0,0 @@
-#pragma once
-
-#include <ext/shared_ptr_helper.h>
-
-#include <Core/NamesAndTypes.h>
-#include <DataStreams/OneBlockInputStream.h>
-#include <Storages/IStorage.h>
-
-
-namespace DB
-{
-
-/// Allows to look at the Block that was inserted or is going to be inserted
-///  to a table as if we read this block from that table.
-
-/// The table has all the properties of another storage,
-///  but will read single prepared block instead.
-/// Used in PushingToViewsBlockOutputStream for generating alias columns
-/// NOTE: Some of the properties seems redundant.
-class StorageBlockInserted : public ext::shared_ptr_helper<StorageBlockInserted>, public IStorage
-{
-public:
-    std::string getName() const override { return storage->getName(); }
-    std::string getTableName() const override { return storage->getTableName(); }
-    std::string getDatabaseName() const override { return storage->getDatabaseName(); }
-    bool isRemote() const override { return storage->isRemote(); }
-    bool supportsSampling() const override { return storage->supportsSampling(); }
-    bool supportsFinal() const override { return storage->supportsFinal(); }
-    bool supportsPrewhere() const override { return storage->supportsPrewhere(); }
-    bool supportsReplication() const override { return storage->supportsReplication(); }
-    bool supportsDeduplication() const override { return storage->supportsDeduplication(); }
-    bool supportsIndexForIn() const override { return storage->supportsIndexForIn(); }
-    bool mayBenefitFromIndexForIn(const ASTPtr & left_in_operand, const Context & query_context) const override
-    {
-        return storage->mayBenefitFromIndexForIn(left_in_operand, query_context);
-    }
-    ASTPtr getPartitionKeyAST() const override { return storage->getPartitionKeyAST(); }
-    ASTPtr getSortingKeyAST() const override { return storage->getSortingKeyAST(); }
-    ASTPtr getPrimaryKeyAST() const override { return storage->getPrimaryKeyAST(); }
-    ASTPtr getSamplingKeyAST() const override { return storage->getSamplingKeyAST(); }
-    Names getColumnsRequiredForPartitionKey() const override { return storage->getColumnsRequiredForPartitionKey(); }
-    Names getColumnsRequiredForSortingKey() const override { return storage->getColumnsRequiredForSortingKey(); }
-    Names getColumnsRequiredForPrimaryKey() const override { return storage->getColumnsRequiredForPrimaryKey(); }
-    Names getColumnsRequiredForSampling() const override { return storage->getColumnsRequiredForSampling(); }
-    Names getColumnsRequiredForFinal() const override { return storage->getColumnsRequiredForFinal(); }
-
-    BlockInputStreams read(const Names &, const SelectQueryInfo &, const Context &, QueryProcessingStage::Enum, size_t, unsigned) override
-    {
-        return {std::make_shared<OneBlockInputStream>(std::move(block))};
-    }
-
-private:
-    Block block;
-    StoragePtr storage;
-
-protected:
-    StorageBlockInserted(Block block_, StoragePtr storage_)
-        : IStorage{storage_->getColumns()}, block(std::move(block_)), storage(storage_)
-    {
-    }
-};
-
-}

dbms/tests/performance/math.xml

@@ -0,0 +1,46 @@
+<test>
+    <type>once</type>
+
+    <stop_conditions>
+        <any_of>
+            <average_speed_not_changing_for_ms>1000</average_speed_not_changing_for_ms>
+            <total_time_ms>10000</total_time_ms>
+        </any_of>
+    </stop_conditions>
+
+    <main_metric>
+        <max_rows_per_second />
+    </main_metric>
+
+    <substitutions>
+        <substitution>
+           <name>func</name>
+           <values>
+               <value>exp</value>
+               <value>log</value>
+               <value>exp2</value>
+               <value>log2</value>
+               <value>exp10</value>
+               <value>log10</value>
+               <value>sqrt</value>
+               <value>cbrt</value>
+               <value>erf</value>
+               <value>erfc</value>
+               <value>lgamma</value>
+               <value>tgamma</value>
+               <value>sin</value>
+               <value>cos</value>
+               <value>tan</value>
+               <value>asin</value>
+               <value>acos</value>
+               <value>atan</value>
+               <value>sigmoid</value>
+               <value>tanh</value>
+           </values>
+        </substitution>
+    </substitutions>
+
+    <query>SELECT count() FROM system.numbers WHERE NOT ignore({func}(toFloat64(number)))</query>
+    <query>SELECT count() FROM system.numbers WHERE NOT ignore({func}(toFloat32(number)))</query>
+    <query>SELECT count() FROM system.numbers WHERE NOT ignore({func}(number))</query>
+</test>

dbms/tests/queries/0_stateless/00087_math_functions.sql

@@ -32,16 +32,16 @@ select tgamma(2) = 1;
 select tgamma(3) = 2;
 select tgamma(4) = 6;
 
-select sum(abs(lgamma(x + 1) - log(tgamma(x + 1))) < 1.0e-9) / count() from system.one array join range(10) as x;
+select sum(abs(lgamma(x + 1) - log(tgamma(x + 1))) < 1.0e-8) / count() from system.one array join range(10) as x;
 
 select abs(e() - arraySum(arrayMap(x -> 1 / tgamma(x + 1), range(13)))) < 1.0e-9;
 
 select log(0) = -inf;
 select log(1) = 0;
-select log(e()) = 1;
+select abs(log(e()) - 1) < 1e-8;
-select log(exp(1)) = 1;
+select abs(log(exp(1)) - 1) < 1e-8;
-select log(exp(2)) = 2;
+select abs(log(exp(2)) - 2) < 1e-8;
-select sum(abs(log(exp(x)) - x) < 1.0e-9) / count() from system.one array join range(100) as x;
+select sum(abs(log(exp(x)) - x) < 1e-8) / count() from system.one array join range(100) as x;
 
 select exp2(-1) = 1/2;
 select exp2(0) = 1;

dbms/tests/queries/0_stateless/00232_format_readable_size.reference

@@ -1,70 +1,70 @@
-1	1.00 B	1	1.00 B	1	1.00 B
+1.00 B	1.00 B	1.00 B
-2.718282	2.72 B	2	2.00 B	2	2.00 B
+2.72 B	2.00 B	2.00 B
-7.389056	7.39 B	7	7.00 B	7	7.00 B
+7.39 B	7.00 B	7.00 B
-20.085537	20.09 B	20	20.00 B	20	20.00 B
+20.09 B	20.00 B	20.00 B
-54.59815	54.60 B	54	54.00 B	54	54.00 B
+54.60 B	54.00 B	54.00 B
-148.413159	148.41 B	148	148.00 B	148	148.00 B
+148.41 B	148.00 B	148.00 B
-403.428793	403.43 B	403	403.00 B	403	403.00 B
+403.43 B	403.00 B	403.00 B
-1096.633158	1.07 KiB	1096	1.07 KiB	1096	1.07 KiB
+1.07 KiB	1.07 KiB	1.07 KiB
-2980.957987	2.91 KiB	2980	2.91 KiB	2980	2.91 KiB
+2.91 KiB	2.91 KiB	2.91 KiB
-8103.083928	7.91 KiB	8103	7.91 KiB	8103	7.91 KiB
+7.91 KiB	7.91 KiB	7.91 KiB
-22026.465795	21.51 KiB	22026	21.51 KiB	22026	21.51 KiB
+21.51 KiB	21.51 KiB	21.51 KiB
-59874.141715	58.47 KiB	59874	58.47 KiB	59874	58.47 KiB
+58.47 KiB	58.47 KiB	58.47 KiB
-162754.791419	158.94 KiB	162754	158.94 KiB	162754	158.94 KiB
+158.94 KiB	158.94 KiB	158.94 KiB
-442413.392009	432.04 KiB	442413	432.04 KiB	442413	432.04 KiB
+432.04 KiB	432.04 KiB	432.04 KiB
-1202604.284165	1.15 MiB	1202604	1.15 MiB	1202604	1.15 MiB
+1.15 MiB	1.15 MiB	1.15 MiB
-3269017.372472	3.12 MiB	3269017	3.12 MiB	3269017	3.12 MiB
+3.12 MiB	3.12 MiB	3.12 MiB
-8886110.520508	8.47 MiB	8886110	8.47 MiB	8886110	8.47 MiB
+8.47 MiB	8.47 MiB	8.47 MiB
-24154952.753575	23.04 MiB	24154952	23.04 MiB	24154952	23.04 MiB
+23.04 MiB	23.04 MiB	23.04 MiB
-65659969.137331	62.62 MiB	65659969	62.62 MiB	65659969	62.62 MiB
+62.62 MiB	62.62 MiB	62.62 MiB
-178482300.963187	170.21 MiB	178482300	170.21 MiB	178482300	170.21 MiB
+170.21 MiB	170.21 MiB	170.21 MiB
-485165195.40979	462.69 MiB	485165195	462.69 MiB	485165195	462.69 MiB
+462.69 MiB	462.69 MiB	462.69 MiB
-1318815734.483214	1.23 GiB	1318815734	1.23 GiB	1318815734	1.23 GiB
+1.23 GiB	1.23 GiB	1.23 GiB
-3584912846.131592	3.34 GiB	3584912846	3.34 GiB	-710054450	-677.16 MiB
+3.34 GiB	3.34 GiB	-2.00 GiB
-9744803446.248903	9.08 GiB	9744803446	9.08 GiB	1154868854	1.08 GiB
+9.08 GiB	9.08 GiB	-2.00 GiB
-26489122129.84347	24.67 GiB	26489122129	24.67 GiB	719318353	686.00 MiB
+24.67 GiB	24.67 GiB	-2.00 GiB
-72004899337.38588	67.06 GiB	72004899337	67.06 GiB	-1009544695	-962.78 MiB
+67.06 GiB	67.06 GiB	-2.00 GiB
-195729609428.83878	182.29 GiB	195729609428	182.29 GiB	-1838886188	-1.71 GiB
+182.29 GiB	182.29 GiB	-2.00 GiB
-532048240601.79865	495.51 GiB	532048240601	495.51 GiB	-527704103	-503.26 MiB
+495.51 GiB	495.51 GiB	-2.00 GiB
-1446257064291.475	1.32 TiB	1446257064291	1.32 TiB	-1146914461	-1.07 GiB
+1.32 TiB	1.32 TiB	-2.00 GiB
-3931334297144.042	3.58 TiB	3931334297144	3.58 TiB	1439221304	1.34 GiB
+3.58 TiB	3.58 TiB	-2.00 GiB
-10686474581524.463	9.72 TiB	10686474581524	9.72 TiB	595949076	568.34 MiB
+9.72 TiB	9.72 TiB	-2.00 GiB
-29048849665247.426	26.42 TiB	29048849665247	26.42 TiB	1985842399	1.85 GiB
+26.42 TiB	26.42 TiB	-2.00 GiB
-78962960182680.69	71.82 TiB	78962960182680	71.82 TiB	-13554280	-12.93 MiB
+71.82 TiB	71.82 TiB	-2.00 GiB
-214643579785916.06	195.22 TiB	214643579785916	195.22 TiB	-1705798980	-1.59 GiB
+195.22 TiB	195.22 TiB	-2.00 GiB
-583461742527454.9	530.66 TiB	583461742527454	530.66 TiB	-974699554	-929.55 MiB
+530.66 TiB	530.66 TiB	-2.00 GiB
-1586013452313430.8	1.41 PiB	1586013452313430	1.41 PiB	-2005982378	-1.87 GiB
+1.41 PiB	1.41 PiB	-2.00 GiB
-4311231547115195	3.83 PiB	4311231547115195	3.83 PiB	-790034757	-753.44 MiB
+3.83 PiB	3.83 PiB	-2.00 GiB
-11719142372802612	10.41 PiB	11719142372802612	10.41 PiB	1983119412	1.85 GiB
+10.41 PiB	10.41 PiB	-2.00 GiB
-31855931757113756	28.29 PiB	31855931757113756	28.29 PiB	408891804	389.95 MiB
+28.29 PiB	28.29 PiB	-2.00 GiB
-86593400423993740	76.91 PiB	86593400423993744	76.91 PiB	673862032	642.64 MiB
+76.91 PiB	76.91 PiB	-2.00 GiB
-235385266837020000	209.06 PiB	235385266837020000	209.06 PiB	791567712	754.90 MiB
+209.06 PiB	209.06 PiB	-2.00 GiB
-639843493530054900	568.30 PiB	639843493530054912	568.30 PiB	1874080000	1.75 GiB
+568.30 PiB	568.30 PiB	-2.00 GiB
-1739274941520501000	1.51 EiB	1739274941520500992	1.51 EiB	538007808	513.08 MiB
+1.51 EiB	1.51 EiB	-2.00 GiB
-4727839468229346000	4.10 EiB	4727839468229346304	4.10 EiB	2061616128	1.92 GiB
+4.10 EiB	4.10 EiB	-2.00 GiB
-12851600114359308000	11.15 EiB	12851600114359308288	11.15 EiB	-1681813504	-1.57 GiB
+11.15 EiB	11.15 EiB	-2.00 GiB
-34934271057485095000	30.30 EiB	0	0.00 B	0	0.00 B
+30.30 EiB	0.00 B	-2.00 GiB
-94961194206024480000	82.37 EiB	0	0.00 B	0	0.00 B
+82.37 EiB	0.00 B	-2.00 GiB
-258131288619006750000	223.89 EiB	0	0.00 B	0	0.00 B
+223.89 EiB	0.00 B	-2.00 GiB
-701673591209763100000	608.60 EiB	0	0.00 B	0	0.00 B
+608.60 EiB	0.00 B	-2.00 GiB
-1.9073465724950998e21	1.62 ZiB	0	0.00 B	0	0.00 B
+1.62 ZiB	0.00 B	-2.00 GiB
-5.184705528587072e21	4.39 ZiB	0	0.00 B	0	0.00 B
+4.39 ZiB	0.00 B	-2.00 GiB
-1.4093490824269389e22	11.94 ZiB	0	0.00 B	0	0.00 B
+11.94 ZiB	0.00 B	-2.00 GiB
-3.831008000716577e22	32.45 ZiB	0	0.00 B	0	0.00 B
+32.45 ZiB	0.00 B	-2.00 GiB
-1.0413759433029089e23	88.21 ZiB	0	0.00 B	0	0.00 B
+88.21 ZiB	0.00 B	-2.00 GiB
-2.830753303274694e23	239.77 ZiB	0	0.00 B	0	0.00 B
+239.77 ZiB	0.00 B	-2.00 GiB
-7.694785265142018e23	651.77 ZiB	0	0.00 B	0	0.00 B
+651.77 ZiB	0.00 B	-2.00 GiB
-2.091659496012996e24	1.73 YiB	0	0.00 B	0	0.00 B
+1.73 YiB	0.00 B	-2.00 GiB
-5.685719999335932e24	4.70 YiB	0	0.00 B	0	0.00 B
+4.70 YiB	0.00 B	-2.00 GiB
-1.545538935590104e25	12.78 YiB	0	0.00 B	0	0.00 B
+12.78 YiB	0.00 B	-2.00 GiB
-4.2012104037905144e25	34.75 YiB	0	0.00 B	0	0.00 B
+34.75 YiB	0.00 B	-2.00 GiB
-1.1420073898156842e26	94.46 YiB	0	0.00 B	0	0.00 B
+94.46 YiB	0.00 B	-2.00 GiB
-3.10429793570192e26	256.78 YiB	0	0.00 B	0	0.00 B
+256.78 YiB	0.00 B	-2.00 GiB
-8.438356668741454e26	698.00 YiB	0	0.00 B	0	0.00 B
+698.00 YiB	0.00 B	-2.00 GiB
-2.29378315946961e27	1897.37 YiB	0	0.00 B	0	0.00 B
+1897.37 YiB	0.00 B	-2.00 GiB
-6.235149080811617e27	5157.59 YiB	0	0.00 B	0	0.00 B
+5157.59 YiB	0.00 B	-2.00 GiB
-1.6948892444103338e28	14019.80 YiB	0	0.00 B	0	0.00 B
+14019.80 YiB	0.00 B	-2.00 GiB
-4.607186634331292e28	38109.75 YiB	0	0.00 B	0	0.00 B
+38109.75 YiB	0.00 B	-2.00 GiB
-1.2523631708422137e29	103593.05 YiB	0	0.00 B	0	0.00 B
+103593.05 YiB	0.00 B	-2.00 GiB
-3.404276049931741e29	281595.11 YiB	0	0.00 B	0	0.00 B
+281595.11 YiB	0.00 B	-2.00 GiB
-9.253781725587787e29	765454.88 YiB	0	0.00 B	0	0.00 B
+765454.88 YiB	0.00 B	-2.00 GiB

dbms/tests/queries/0_stateless/00232_format_readable_size.sql

@@ -1 +1,4 @@
-SELECT round(exp(number), 6) AS x, formatReadableSize(x), toUInt64(x) AS y, formatReadableSize(y), toInt32(y) AS z, formatReadableSize(z) FROM system.numbers LIMIT 70;
+WITH round(exp(number), 6) AS x, toUInt64(x) AS y, toInt32(x) AS z
+SELECT formatReadableSize(x), formatReadableSize(y), formatReadableSize(z)
+FROM system.numbers
+LIMIT 70;

dbms/tests/queries/0_stateless/00332_quantile_timing_memory_leak.sql

@@ -1,2 +1,2 @@
 SELECT quantileTiming(number) FROM (SELECT * FROM system.numbers LIMIT 10000);
-SELECT floor(log(1 + number) / log(1.5)) AS k, count() AS c, quantileTiming(number % 10000) AS q FROM (SELECT * FROM system.numbers LIMIT 1000000) GROUP BY k ORDER BY k;
+SELECT floor(log2(1 + number) / log2(1.5)) AS k, count() AS c, quantileTiming(number % 10000) AS q FROM (SELECT * FROM system.numbers LIMIT 1000000) GROUP BY k ORDER BY k;

dbms/tests/queries/0_stateless/00700_decimal_math.reference

@@ -1,6 +1,6 @@
-42.4200	3.7476	42.419153766068966
+42.4200	3.7476	42.419154
-42.4200	5.4066	42.41786197045111
+42.4200	5.4066	42.417862
-42.4200	1.6275	42.413098391048806
+42.4200	1.6275	42.413098
 42.4200	6.513	42.419169
 42.4200	3.4875	42.417263671875
 1.00000	0.8427007929497149	0.15729920705028513
@@ -8,9 +8,9 @@
 0.00	0	1	0
 3.14159265	0	-1	-0
 1.00	1.5707963267948966	0	0.7853981633974483
-42.4200	3.7476	42.419153766068966
+42.4200	3.7476	42.419154
-42.4200	5.4066	42.41786197045111
+42.4200	5.4066	42.417862
-42.4200	1.6275	42.413098391048806
+42.4200	1.6275	42.413098
 42.4200	6.513	42.419169
 42.4200	3.4875	42.417263671875
 1.00000	0.8427007929497149	0.15729920705028513
@@ -18,9 +18,9 @@
 0.00	0	1	0
 3.14159265358979328	0	-1	-0
 1.00	1.5707963267948966	0	0.7853981633974483
-42.4200	3.7476	42.419153766068966
+42.4200	3.7476	42.419154
-42.4200	5.4066	42.41786197045111
+42.4200	5.4066	42.417862
-42.4200	1.6275	42.413098391048806
+42.4200	1.6275	42.413098
 42.4200	6.513	42.419169
 42.4200	3.4875	42.417263671875
 1.00000	0.8427007929497149	0.15729920705028513

dbms/tests/queries/0_stateless/00700_decimal_math.sql

@@ -1,6 +1,6 @@
-SELECT toDecimal32('42.42', 4) AS x, toDecimal32(log(x), 4) AS y, exp(y);
+SELECT toDecimal32('42.42', 4) AS x, toDecimal32(log(x), 4) AS y, round(exp(y), 6);
-SELECT toDecimal32('42.42', 4) AS x, toDecimal32(log2(x), 4) AS y, exp2(y);
+SELECT toDecimal32('42.42', 4) AS x, toDecimal32(log2(x), 4) AS y, round(exp2(y), 6);
-SELECT toDecimal32('42.42', 4) AS x, toDecimal32(log10(x), 4) AS y, exp10(y);
+SELECT toDecimal32('42.42', 4) AS x, toDecimal32(log10(x), 4) AS y, round(exp10(y), 6);
 
 SELECT toDecimal32('42.42', 4) AS x, toDecimal32(sqrt(x), 3) AS y, y * y;
 SELECT toDecimal32('42.42', 4) AS x, toDecimal32(cbrt(x), 4) AS y, toDecimal64(y, 4) * y * y;
@@ -12,9 +12,9 @@ SELECT toDecimal32(pi(), 8) AS x, round(sin(x), 8), round(cos(x), 8), round(tan(
 SELECT toDecimal32('1.0', 2) AS x, asin(x), acos(x), atan(x);
 
 
-SELECT toDecimal64('42.42', 4) AS x, toDecimal32(log(x), 4) AS y, exp(y);
+SELECT toDecimal64('42.42', 4) AS x, toDecimal32(log(x), 4) AS y, round(exp(y), 6);
-SELECT toDecimal64('42.42', 4) AS x, toDecimal32(log2(x), 4) AS y, exp2(y);
+SELECT toDecimal64('42.42', 4) AS x, toDecimal32(log2(x), 4) AS y, round(exp2(y), 6);
-SELECT toDecimal64('42.42', 4) AS x, toDecimal32(log10(x), 4) AS y, exp10(y);
+SELECT toDecimal64('42.42', 4) AS x, toDecimal32(log10(x), 4) AS y, round(exp10(y), 6);
 
 SELECT toDecimal64('42.42', 4) AS x, toDecimal32(sqrt(x), 3) AS y, y * y;
 SELECT toDecimal64('42.42', 4) AS x, toDecimal32(cbrt(x), 4) AS y, toDecimal64(y, 4) * y * y;
@@ -26,9 +26,9 @@ SELECT toDecimal64(pi(), 17) AS x, round(sin(x), 8), round(cos(x), 8), round(tan
 SELECT toDecimal64('1.0', 2) AS x, asin(x), acos(x), atan(x);
 
 
-SELECT toDecimal128('42.42', 4) AS x, toDecimal32(log(x), 4) AS y, exp(y);
+SELECT toDecimal128('42.42', 4) AS x, toDecimal32(log(x), 4) AS y, round(exp(y), 6);
-SELECT toDecimal128('42.42', 4) AS x, toDecimal32(log2(x), 4) AS y, exp2(y);
+SELECT toDecimal128('42.42', 4) AS x, toDecimal32(log2(x), 4) AS y, round(exp2(y), 6);
-SELECT toDecimal128('42.42', 4) AS x, toDecimal32(log10(x), 4) AS y, exp10(y);
+SELECT toDecimal128('42.42', 4) AS x, toDecimal32(log10(x), 4) AS y, round(exp10(y), 6);
 
 SELECT toDecimal128('42.42', 4) AS x, toDecimal32(sqrt(x), 3) AS y, y * y;
 SELECT toDecimal128('42.42', 4) AS x, toDecimal32(cbrt(x), 4) AS y, toDecimal64(y, 4) * y * y;

dbms/tests/queries/0_stateless/00712_nan_comparison.reference

@@ -19,7 +19,7 @@
 0	1	0	0	0	0
 0	1	0	0	0	0
 nan	nan	nan	nan	nan	nan
-nan	nan	nan	nan	nan	nan	nan	nan	nan
+nan	0	nan	nan	nan	nan	nan	nan	nan
 nan	nan	nan	nan	nan
 nan	nan	nan	nan	nan	nan	nan
 -1	1

dbms/tests/queries/0_stateless/00712_nan_comparison.sql

@@ -22,7 +22,7 @@ SELECT -nan = toFloat64(0.0), -nan != toFloat64(0.0), -nan < toFloat64(0.0), -na
 
 --SELECT 1 % nan, nan % 1, pow(x, 1), pow(1, x); -- TODO
 SELECT 1 + nan, 1 - nan, nan - 1, 1 * nan, 1 / nan, nan / 1;
-SELECT nan AS x, exp(x), exp2(x), exp10(x), log(x), log2(x), log10(x), sqrt(x), cbrt(x);
+SELECT nan AS x, isFinite(exp(x)) /* exp(nan) is allowed to return inf */, exp2(x), exp10(x), log(x), log2(x), log10(x), sqrt(x), cbrt(x);
 SELECT nan AS x, erf(x), erfc(x), lgamma(x), tgamma(x);
 SELECT nan AS x, sin(x), cos(x), tan(x), asin(x), acos(x), atan(x);
 

dbms/tests/queries/0_stateless/00715_bounding_ratio.sql

@@ -23,4 +23,4 @@ SELECT number % 10 AS k, boundingRatio(1000 + number, number * 1.5 - 111) FROM n
 SELECT boundingRatio(1000 + number, number * 1.5 - 111) FROM numbers(2);
 SELECT boundingRatio(1000 + number, number * 1.5 - 111) FROM numbers(1);
 SELECT boundingRatio(1000 + number, number * 1.5 - 111) FROM numbers(1) WHERE 0;
-SELECT boundingRatio(number, exp(number)) = e() - 1 FROM numbers(2);
+SELECT boundingRatio(number, exp(number)) = exp(1) - 1 FROM numbers(2);

dbms/tests/queries/0_stateless/00760_insert_json_with_defaults.reference

@@ -1,10 +1,10 @@
 0	0	6	6	6
-0	5	5	1.7917595	5
+0	5	5	1.79176	5
-1	1	2	1.0986123	42
+1	1	2	1.09861	42
-1	1	2	1.0986123	42
+1	1	2	1.09861	42
-2	2	4	1.609438	2
+2	2	4	1.60944	2
 3	3	3	3	3
-4	0	4	1.609438	42
+4	0	4	1.60944	42
-{"x":7,"y":7,"a":"14","b":2.7080503,"c":42,"n.a":["1","2"],"n.b":["a","b"]}
+{"x":7,"y":7,"a":"14","b":2.70805,"c":42,"n.a":["1","2"],"n.b":["a","b"]}
-{"x":8,"y":8,"a":"16","b":2.8332133,"c":42,"n.a":["3","4"],"n.c":[0,0],"n.b":["c","d"]}
+{"x":8,"y":8,"a":"16","b":2.83321,"c":42,"n.a":["3","4"],"n.c":[0,0],"n.b":["c","d"]}
-{"x":9,"y":9,"a":"18","b":2.944439,"c":42,"n.a":[],"n.c":[],"n.b":[]}
+{"x":9,"y":9,"a":"18","b":2.94444,"c":42,"n.a":[],"n.c":[],"n.b":[]}

dbms/tests/queries/0_stateless/00760_insert_json_with_defaults.sql

@@ -6,7 +6,7 @@ CREATE TABLE defaults
     x UInt32,
     y UInt32,
     a DEFAULT x + y,
-    b Float32 DEFAULT log(1 + x + y),
+    b Float32 DEFAULT round(log(1 + x + y), 5),
     c UInt32 DEFAULT 42,
     e MATERIALIZED x + y,
     f ALIAS x + y

dbms/tests/queries/0_stateless/00794_materialized_view_with_column_defaults.sql

@@ -1,5 +1,3 @@
-USE test;
-
 DROP TABLE IF EXISTS test1;
 
 CREATE TABLE test1 (
@@ -15,7 +13,7 @@ AS SELECT
   date,
   zoneId,
   test
-FROM test.test1;
+FROM test1;
 
 INSERT INTO test1 VALUES ('2018-12-10', '2018-12-10 23:59:59', 1);
 

dbms/tests/queries/0_stateless/00851_http_insert_json_defaults.reference

@@ -1,6 +1,6 @@
 {"x":0,"y":0,"a":"6","b":6,"c":6}
-{"x":0,"y":5,"a":"5","b":1.7917595,"c":5}
+{"x":0,"y":5,"a":"5","b":1.79176,"c":5}
-{"x":1,"y":1,"a":"2","b":1.0986123,"c":42}
+{"x":1,"y":1,"a":"2","b":1.09861,"c":42}
-{"x":2,"y":2,"a":"4","b":1.609438,"c":2}
+{"x":2,"y":2,"a":"4","b":1.60944,"c":2}
 {"x":3,"y":3,"a":"3","b":3,"c":3}
-{"x":4,"y":0,"a":"4","b":1.609438,"c":42}
+{"x":4,"y":0,"a":"4","b":1.60944,"c":42}

dbms/tests/queries/0_stateless/00851_http_insert_json_defaults.sh

@@ -4,7 +4,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
 . $CURDIR/../shell_config.sh
 
 $CLICKHOUSE_CLIENT --query="DROP TABLE IF EXISTS defaults"
-$CLICKHOUSE_CLIENT --query="CREATE TABLE defaults (x UInt32, y UInt32, a DEFAULT x + y, b Float32 DEFAULT log(1 + x + y), c UInt32 DEFAULT 42, e MATERIALIZED x + y, f ALIAS x + y) ENGINE = Memory"
+$CLICKHOUSE_CLIENT --query="CREATE TABLE defaults (x UInt32, y UInt32, a DEFAULT x + y, b Float32 DEFAULT round(log(1 + x + y), 5), c UInt32 DEFAULT 42, e MATERIALIZED x + y, f ALIAS x + y) ENGINE = Memory"
 
 echo -ne '{"x":1, "y":1}\n' | ${CLICKHOUSE_CURL} -sS "${CLICKHOUSE_URL_PARAMS}&query=INSERT%20INTO%20defaults%20FORMAT%20JSONEachRow%20SETTINGS%20input_format_defaults_for_omitted_fields=1" --data-binary @-
 echo -ne '{"x":2, "y":2, "c":2}\n' | ${CLICKHOUSE_CURL} -sS "${CLICKHOUSE_URL_PARAMS}&query=INSERT+INTO+defaults+FORMAT+JSONEachRow+SETTINGS+input_format_defaults_for_omitted_fields=1" --data-binary @-

dbms/tests/queries/0_stateless/00978_ml_math.reference

@@ -0,0 +1,3 @@
+0.26894	0.26894	0.26894	-0.76159	-0.76159	-0.76159
+0.5	0.5	0.5	0	0	0
+0.73106	0.73106	0.73106	0.76159	0.76159	0.76159

dbms/tests/queries/0_stateless/00978_ml_math.sql

@@ -0,0 +1,4 @@
+SELECT
+    round(sigmoid(x), 5), round(sigmoid(toFloat32(x)), 5), round(sigmoid(toFloat64(x)), 5),
+    round(tanh(x), 5), round(TANH(toFloat32(x)), 5), round(TANh(toFloat64(x)), 5)
+FROM (SELECT arrayJoin([-1, 0, 1]) AS x);

libs/libglibc-compatibility/musl/__math_divzerof.c

@@ -0,0 +1,17 @@
+#include <stdint.h>
+
+/* fp_barrier returns its input, but limits code transformations
+   as if it had a side-effect (e.g. observable io) and returned
+   an arbitrary value.  */
+
+static inline float fp_barrierf(float x)
+{
+	volatile float y = x;
+	return y;
+}
+
+
+float __math_divzerof(uint32_t sign)
+{
+	return fp_barrierf(sign ? -1.0f : 1.0f) / 0.0f;
+}

libs/libglibc-compatibility/musl/__math_invalidf.c

@@ -0,0 +1,4 @@
+float __math_invalidf(float x)
+{
+	return (x - x) / (x - x);
+}

libs/libglibc-compatibility/musl/logf.c

@@ -0,0 +1,88 @@
+/*
+ * Single-precision log function.
+ *
+ * Copyright (c) 2017-2018, Arm Limited.
+ * SPDX-License-Identifier: MIT
+ */
+
+#include <math.h>
+#include <stdint.h>
+#include "logf_data.h"
+
+float __math_invalidf(float);
+float __math_divzerof(uint32_t);
+
+/*
+LOGF_TABLE_BITS = 4
+LOGF_POLY_ORDER = 4
+
+ULP error: 0.818 (nearest rounding.)
+Relative error: 1.957 * 2^-26 (before rounding.)
+*/
+
+#define T __logf_data.tab
+#define A __logf_data.poly
+#define Ln2 __logf_data.ln2
+#define N (1 << LOGF_TABLE_BITS)
+#define OFF 0x3f330000
+#define WANT_ROUNDING 1
+
+#define asuint(f) ((union{float _f; uint32_t _i;}){f})._i
+#define asfloat(i) ((union{uint32_t _i; float _f;}){i})._f
+
+/* Evaluate an expression as the specified type. With standard excess
+   precision handling a type cast or assignment is enough (with
+   -ffloat-store an assignment is required, in old compilers argument
+   passing and return statement may not drop excess precision).  */
+
+static inline float eval_as_float(float x)
+{
+    float y = x;
+    return y;
+}
+
+float logf(float x)
+{
+	double_t z, r, r2, y, y0, invc, logc;
+	uint32_t ix, iz, tmp;
+	int k, i;
+
+	ix = asuint(x);
+	/* Fix sign of zero with downward rounding when x==1.  */
+	if (WANT_ROUNDING && __builtin_expect(ix == 0x3f800000, 0))
+		return 0;
+	if (__builtin_expect(ix - 0x00800000 >= 0x7f800000 - 0x00800000, 0)) {
+		/* x < 0x1p-126 or inf or nan.  */
+		if (ix * 2 == 0)
+			return __math_divzerof(1);
+		if (ix == 0x7f800000) /* log(inf) == inf.  */
+			return x;
+		if ((ix & 0x80000000) || ix * 2 >= 0xff000000)
+			return __math_invalidf(x);
+		/* x is subnormal, normalize it.  */
+		ix = asuint(x * 0x1p23f);
+		ix -= 23 << 23;
+	}
+
+	/* x = 2^k z; where z is in range [OFF,2*OFF] and exact.
+	   The range is split into N subintervals.
+	   The ith subinterval contains z and c is near its center.  */
+	tmp = ix - OFF;
+	i = (tmp >> (23 - LOGF_TABLE_BITS)) % N;
+	k = (int32_t)tmp >> 23; /* arithmetic shift */
+	iz = ix - (tmp & 0x1ff << 23);
+	invc = T[i].invc;
+	logc = T[i].logc;
+	z = (double_t)asfloat(iz);
+
+	/* log(x) = log1p(z/c-1) + log(c) + k*Ln2 */
+	r = z * invc - 1;
+	y0 = logc + (double_t)k * Ln2;
+
+	/* Pipelined polynomial evaluation to approximate log1p(r).  */
+	r2 = r * r;
+	y = A[1] * r + A[2];
+	y = A[0] * r2 + y;
+	y = y * r2 + (y0 + r);
+	return eval_as_float(y);
+}

libs/libglibc-compatibility/musl/logf_data.c

@@ -0,0 +1,33 @@
+/*
+ * Data definition for logf.
+ *
+ * Copyright (c) 2017-2018, Arm Limited.
+ * SPDX-License-Identifier: MIT
+ */
+
+#include "logf_data.h"
+
+const struct logf_data __logf_data = {
+  .tab = {
+  { 0x1.661ec79f8f3bep+0, -0x1.57bf7808caadep-2 },
+  { 0x1.571ed4aaf883dp+0, -0x1.2bef0a7c06ddbp-2 },
+  { 0x1.49539f0f010bp+0, -0x1.01eae7f513a67p-2 },
+  { 0x1.3c995b0b80385p+0, -0x1.b31d8a68224e9p-3 },
+  { 0x1.30d190c8864a5p+0, -0x1.6574f0ac07758p-3 },
+  { 0x1.25e227b0b8eap+0, -0x1.1aa2bc79c81p-3 },
+  { 0x1.1bb4a4a1a343fp+0, -0x1.a4e76ce8c0e5ep-4 },
+  { 0x1.12358f08ae5bap+0, -0x1.1973c5a611cccp-4 },
+  { 0x1.0953f419900a7p+0, -0x1.252f438e10c1ep-5 },
+  { 0x1p+0, 0x0p+0 },
+  { 0x1.e608cfd9a47acp-1, 0x1.aa5aa5df25984p-5 },
+  { 0x1.ca4b31f026aap-1, 0x1.c5e53aa362eb4p-4 },
+  { 0x1.b2036576afce6p-1, 0x1.526e57720db08p-3 },
+  { 0x1.9c2d163a1aa2dp-1, 0x1.bc2860d22477p-3 },
+  { 0x1.886e6037841edp-1, 0x1.1058bc8a07ee1p-2 },
+  { 0x1.767dcf5534862p-1, 0x1.4043057b6ee09p-2 },
+  },
+  .ln2 = 0x1.62e42fefa39efp-1,
+  .poly = {
+  -0x1.00ea348b88334p-2, 0x1.5575b0be00b6ap-2, -0x1.ffffef20a4123p-2,
+  }
+};

libs/libglibc-compatibility/musl/logf_data.h

@@ -0,0 +1,18 @@
+/*
+ * Copyright (c) 2017-2018, Arm Limited.
+ * SPDX-License-Identifier: MIT
+ */
+#ifndef _LOGF_DATA_H
+#define _LOGF_DATA_H
+
+#define LOGF_TABLE_BITS 4
+#define LOGF_POLY_ORDER 4
+extern __attribute__((__visibility__("hidden"))) const struct logf_data {
+	struct {
+		double invc, logc;
+	} tab[1 << LOGF_TABLE_BITS];
+	double ln2;
+	double poly[LOGF_POLY_ORDER - 1]; /* First order coefficient is 1.  */
+} __logf_data;
+
+#endif