Fix hash join when the input tables have nulls on only one side

cpp/benchmarks/join/join.cu

@@ -27,7 +27,10 @@ void nvbench_inner_join(nvbench::state& state,
                  cudf::table_view const& right_input,
                  cudf::null_equality compare_nulls,
                  rmm::cuda_stream_view stream) {
-    cudf::hash_join hj_obj(left_input, compare_nulls, stream);
+    auto const has_nulls = cudf::has_nested_nulls(left_input) || cudf::has_nested_nulls(right_input)
+                             ? cudf::nullable_join::YES
+                             : cudf::nullable_join::NO;
+    cudf::hash_join hj_obj(left_input, has_nulls, compare_nulls, stream);
     return hj_obj.inner_join(right_input, std::nullopt, stream);
   };
 
@@ -44,7 +47,10 @@ void nvbench_left_join(nvbench::state& state,
                  cudf::table_view const& right_input,
                  cudf::null_equality compare_nulls,
                  rmm::cuda_stream_view stream) {
-    cudf::hash_join hj_obj(left_input, compare_nulls, stream);
+    auto const has_nulls = cudf::has_nested_nulls(left_input) || cudf::has_nested_nulls(right_input)
+                             ? cudf::nullable_join::YES
+                             : cudf::nullable_join::NO;
+    cudf::hash_join hj_obj(left_input, has_nulls, compare_nulls, stream);
     return hj_obj.left_join(right_input, std::nullopt, stream);
   };
 
@@ -61,7 +67,10 @@ void nvbench_full_join(nvbench::state& state,
                  cudf::table_view const& right_input,
                  cudf::null_equality compare_nulls,
                  rmm::cuda_stream_view stream) {
-    cudf::hash_join hj_obj(left_input, compare_nulls, stream);
+    auto const has_nulls = cudf::has_nested_nulls(left_input) || cudf::has_nested_nulls(right_input)
+                             ? cudf::nullable_join::YES
+                             : cudf::nullable_join::NO;
+    cudf::hash_join hj_obj(left_input, has_nulls, compare_nulls, stream);
     return hj_obj.full_join(right_input, std::nullopt, stream);
   };
 

cpp/include/cudf/detail/join.hpp

@@ -73,10 +73,10 @@ struct hash_join {
   hash_join& operator=(hash_join&&) = delete;
 
  private:
-  bool const _is_empty;                         ///< true if `_hash_table` is empty
-  rmm::device_buffer const _composite_bitmask;  ///< Bitmask to denote whether a row is valid
-  cudf::null_equality const _nulls_equal;       ///< whether to consider nulls as equal
-  cudf::table_view _build;                      ///< input table to build the hash map
+  bool const _is_empty;   ///< true if `_hash_table` is empty
+  bool const _has_nulls;  ///< true if nulls are present in either build table or any probe table
+  cudf::null_equality const _nulls_equal;  ///< whether to consider nulls as equal
+  cudf::table_view _build;                 ///< input table to build the hash map
   std::shared_ptr<cudf::experimental::row::equality::preprocessed_table>
     _preprocessed_build;  ///< input table preprocssed for row operators
   map_type _hash_table;   ///< hash table built on `_build`
@@ -89,10 +89,13 @@ struct hash_join {
    * @throw cudf::logic_error if the number of rows in `build` table exceeds MAX_JOIN_SIZE.
    *
    * @param build The build table, from which the hash table is built.
+   * @param has_nulls Flag to indicate if the there exists any nulls in the `build` table or
+   *        any `probe` table that will be used later for join.
    * @param compare_nulls Controls whether null join-key values should match or not.
    * @param stream CUDA stream used for device memory operations and kernel launches.
    */
   hash_join(cudf::table_view const& build,
+            bool has_nulls,
             cudf::null_equality compare_nulls,
             rmm::cuda_stream_view stream);
 

cpp/include/cudf/join.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019-2022, NVIDIA CORPORATION.
+ * Copyright (c) 2019-2023, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -256,6 +256,16 @@ std::unique_ptr<cudf::table> cross_join(
   cudf::table_view const& right,
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 
+/**
+ * @brief The enum class to specify if any of the input join tables (`build` table and any later
+ * `probe` table) has nulls.
+ *
+ * This is used upon hash_join object construction to specify the existence of nulls in all the
+ * possible input tables. If such null existence is unknown, `YES` should be used as the default
+ * option.
+ */
+enum class nullable_join : bool { YES, NO };
+
 /**
  * @brief Hash join that builds hash table in creation and probes results in subsequent `*_join`
  * member functions.
@@ -289,6 +299,17 @@ class hash_join {
             null_equality compare_nulls,
             rmm::cuda_stream_view stream = cudf::get_default_stream());
 
+  /**
+   * @copydoc hash_join(cudf::table_view const&, null_equality, rmm::cuda_stream_view)
+   *
+   * @param has_nulls Flag to indicate if the there exists any nulls in the `build` table or
+   *        any `probe` table that will be used later for join
+   */
+  hash_join(cudf::table_view const& build,
+            nullable_join has_nulls,
+            null_equality compare_nulls,
+            rmm::cuda_stream_view stream = cudf::get_default_stream());
+
   /**
    * Returns the row indices that can be used to construct the result of performing
    * an inner join between two tables. @see cudf::inner_join(). Behavior is undefined if the
@@ -300,6 +321,9 @@ class hash_join {
    * @param mr Device memory resource used to allocate the returned table and columns' device
    * memory.
    *
+   * @throw cudf::logic_error If the input probe table has nulls while this hash_join object was not
+   * constructed with null check.
+   *
    * @return A pair of columns [`left_indices`, `right_indices`] that can be used to construct
    * the result of performing an inner join between two tables with `build` and `probe`
    * as the the join keys .
@@ -322,6 +346,9 @@ class hash_join {
    * @param mr Device memory resource used to allocate the returned table and columns' device
    * memory.
    *
+   * @throw cudf::logic_error If the input probe table has nulls while this hash_join object was not
+   * constructed with null check.
+   *
    * @return A pair of columns [`left_indices`, `right_indices`] that can be used to construct
    * the result of performing a left join between two tables with `build` and `probe`
    * as the the join keys .
@@ -344,6 +371,9 @@ class hash_join {
    * @param mr Device memory resource used to allocate the returned table and columns' device
    * memory.
    *
+   * @throw cudf::logic_error If the input probe table has nulls while this hash_join object was not
+   * constructed with null check.
+   *
    * @return A pair of columns [`left_indices`, `right_indices`] that can be used to construct
    * the result of performing a full join between two tables with `build` and `probe`
    * as the the join keys .
@@ -362,6 +392,9 @@ class hash_join {
    * @param probe The probe table, from which the tuples are probed
    * @param stream CUDA stream used for device memory operations and kernel launches
    *
+   * @throw cudf::logic_error If the input probe table has nulls while this hash_join object was not
+   * constructed with null check.
+   *
    * @return The exact number of output when performing an inner join between two tables with
    * `build` and `probe` as the the join keys .
    */
@@ -375,6 +408,9 @@ class hash_join {
    * @param probe The probe table, from which the tuples are probed
    * @param stream CUDA stream used for device memory operations and kernel launches
    *
+   * @throw cudf::logic_error If the input probe table has nulls while this hash_join object was not
+   * constructed with null check.
+   *
    * @return The exact number of output when performing a left join between two tables with `build`
    * and `probe` as the the join keys .
    */
@@ -390,6 +426,9 @@ class hash_join {
    * @param mr Device memory resource used to allocate the intermediate table and columns' device
    * memory.
    *
+   * @throw cudf::logic_error If the input probe table has nulls while this hash_join object was not
+   * constructed with null check.
+   *
    * @return The exact number of output when performing a full join between two tables with `build`
    * and `probe` as the the join keys .
    */

cpp/src/join/hash_join.cu

@@ -359,11 +359,11 @@ std::size_t get_full_join_size(
 
 template <typename Hasher>
 hash_join<Hasher>::hash_join(cudf::table_view const& build,
+                             bool has_nulls,
                              cudf::null_equality compare_nulls,
                              rmm::cuda_stream_view stream)
-  : _is_empty{build.num_rows() == 0},
-    _composite_bitmask{
-      cudf::detail::bitmask_and(build, stream, rmm::mr::get_current_device_resource()).first},
+  : _has_nulls(has_nulls),
+    _is_empty{build.num_rows() == 0},
     _nulls_equal{compare_nulls},
     _hash_table{compute_hash_table_size(build.num_rows()),
                 cuco::empty_key{std::numeric_limits<hash_value_type>::max()},
@@ -381,11 +381,14 @@ hash_join<Hasher>::hash_join(cudf::table_view const& build,
 
   if (_is_empty) { return; }
 
+  auto const row_bitmask =
+    cudf::detail::bitmask_and(build, stream, rmm::mr::get_current_device_resource()).first;
   cudf::detail::build_join_hash_table(_build,
                                       _preprocessed_build,
                                       _hash_table,
+                                      _has_nulls,
                                       _nulls_equal,
-                                      static_cast<bitmask_type const*>(_composite_bitmask.data()),
+                                      reinterpret_cast<bitmask_type const*>(row_bitmask.data()),
                                       stream);
 }
 
@@ -434,19 +437,21 @@ std::size_t hash_join<Hasher>::inner_join_size(cudf::table_view const& probe,
   // Return directly if build table is empty
   if (_is_empty) { return 0; }
 
+  CUDF_EXPECTS(_has_nulls || !cudf::has_nested_nulls(probe),
+               "Probe table has nulls while build table was not hashed with null check.");
+
   auto const preprocessed_probe =
     cudf::experimental::row::equality::preprocessed_table::create(probe, stream);
 
-  return cudf::detail::compute_join_output_size(
-    _build,
-    probe,
-    _preprocessed_build,
-    preprocessed_probe,
-    _hash_table,
-    cudf::detail::join_kind::INNER_JOIN,
-    cudf::has_nested_nulls(probe) || cudf::has_nested_nulls(_build),
-    _nulls_equal,
-    stream);
+  return cudf::detail::compute_join_output_size(_build,
+                                                probe,
+                                                _preprocessed_build,
+                                                preprocessed_probe,
+                                                _hash_table,
+                                                cudf::detail::join_kind::INNER_JOIN,
+                                                _has_nulls,
+                                                _nulls_equal,
+                                                stream);
 }
 
 template <typename Hasher>
@@ -458,19 +463,21 @@ std::size_t hash_join<Hasher>::left_join_size(cudf::table_view const& probe,
   // Trivial left join case - exit early
   if (_is_empty) { return probe.num_rows(); }
 
+  CUDF_EXPECTS(_has_nulls || !cudf::has_nested_nulls(probe),
+               "Probe table has nulls while build table was not hashed with null check.");
+
   auto const preprocessed_probe =
     cudf::experimental::row::equality::preprocessed_table::create(probe, stream);
 
-  return cudf::detail::compute_join_output_size(
-    _build,
-    probe,
-    _preprocessed_build,
-    preprocessed_probe,
-    _hash_table,
-    cudf::detail::join_kind::LEFT_JOIN,
-    cudf::has_nested_nulls(probe) || cudf::has_nested_nulls(_build),
-    _nulls_equal,
-    stream);
+  return cudf::detail::compute_join_output_size(_build,
+                                                probe,
+                                                _preprocessed_build,
+                                                preprocessed_probe,
+                                                _hash_table,
+                                                cudf::detail::join_kind::LEFT_JOIN,
+                                                _has_nulls,
+                                                _nulls_equal,
+                                                stream);
 }
 
 template <typename Hasher>
@@ -483,19 +490,21 @@ std::size_t hash_join<Hasher>::full_join_size(cudf::table_view const& probe,
   // Trivial left join case - exit early
   if (_is_empty) { return probe.num_rows(); }
 
+  CUDF_EXPECTS(_has_nulls || !cudf::has_nested_nulls(probe),
+               "Probe table has nulls while build table was not hashed with null check.");
+
   auto const preprocessed_probe =
     cudf::experimental::row::equality::preprocessed_table::create(probe, stream);
 
-  return cudf::detail::get_full_join_size(
-    _build,
-    probe,
-    _preprocessed_build,
-    preprocessed_probe,
-    _hash_table,
-    cudf::has_nested_nulls(probe) || cudf::has_nested_nulls(_build),
-    _nulls_equal,
-    stream,
-    mr);
+  return cudf::detail::get_full_join_size(_build,
+                                          probe,
+                                          _preprocessed_build,
+                                          preprocessed_probe,
+                                          _hash_table,
+                                          _has_nulls,
+                                          _nulls_equal,
+                                          stream,
+                                          mr);
 }
 
 template <typename Hasher>
@@ -514,20 +523,22 @@ hash_join<Hasher>::probe_join_indices(cudf::table_view const& probe_table,
 
   CUDF_EXPECTS(!_is_empty, "Hash table of hash join is null.");
 
+  CUDF_EXPECTS(_has_nulls || !cudf::has_nested_nulls(probe_table),
+               "Probe table has nulls while build table was not hashed with null check.");
+
   auto const preprocessed_probe =
     cudf::experimental::row::equality::preprocessed_table::create(probe_table, stream);
-  auto join_indices = cudf::detail::probe_join_hash_table(
-    _build,
-    probe_table,
-    _preprocessed_build,
-    preprocessed_probe,
-    _hash_table,
-    join,
-    cudf::has_nested_nulls(probe_table) || cudf::has_nested_nulls(_build),
-    _nulls_equal,
-    output_size,
-    stream,
-    mr);
+  auto join_indices = cudf::detail::probe_join_hash_table(_build,
+                                                          probe_table,
+                                                          _preprocessed_build,
+                                                          preprocessed_probe,
+                                                          _hash_table,
+                                                          join,
+                                                          _has_nulls,
+                                                          _nulls_equal,
+                                                          output_size,
+                                                          stream,
+                                                          mr);
 
   if (join == cudf::detail::join_kind::FULL_JOIN) {
     auto complement_indices = detail::get_left_join_indices_complement(
@@ -553,6 +564,9 @@ hash_join<Hasher>::compute_hash_join(cudf::table_view const& probe,
   CUDF_EXPECTS(_build.num_columns() == probe.num_columns(),
                "Mismatch in number of columns to be joined on");
 
+  CUDF_EXPECTS(_has_nulls || !cudf::has_nested_nulls(probe),
+               "Probe table has nulls while build table was not hashed with null check.");
+
   if (is_trivial_join(probe, _build, join)) {
     return std::pair(std::make_unique<rmm::device_uvector<size_type>>(0, stream, mr),
                      std::make_unique<rmm::device_uvector<size_type>>(0, stream, mr));
@@ -574,7 +588,17 @@ hash_join::~hash_join() = default;
 hash_join::hash_join(cudf::table_view const& build,
                      null_equality compare_nulls,
                      rmm::cuda_stream_view stream)
-  : _impl{std::make_unique<const impl_type>(build, compare_nulls, stream)}
+  // If we cannot know beforehand about null existence then let's assume that there are nulls.
+  : hash_join(build, nullable_join::YES, compare_nulls, stream)
+{
+}
+
+hash_join::hash_join(cudf::table_view const& build,
+                     nullable_join has_nulls,
+                     null_equality compare_nulls,
+                     rmm::cuda_stream_view stream)
+  : _impl{std::make_unique<const impl_type>(
+      build, has_nulls == nullable_join::YES, compare_nulls, stream)}
 {
 }