Clean up literal zero cuda_stream_view arguments

cpp/benchmarks/column/concatenate_benchmark.cpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2020, NVIDIA CORPORATION.
+ * Copyright (c) 2020-2021, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -62,7 +62,7 @@ static void BM_concatenate(benchmark::State& state)
   CHECK_CUDA(0);
 
   for (auto _ : state) {
-    cuda_event_timer raii(state, true, 0);
+    cuda_event_timer raii(state, true, rmm::cuda_stream_default);
     auto result = cudf::concatenate(column_views);
   }
 
@@ -124,7 +124,7 @@ static void BM_concatenate_tables(benchmark::State& state)
   CHECK_CUDA(0);
 
   for (auto _ : state) {
-    cuda_event_timer raii(state, true, 0);
+    cuda_event_timer raii(state, true, rmm::cuda_stream_default);
     auto result = cudf::concatenate(table_views);
   }
 
@@ -184,7 +184,7 @@ static void BM_concatenate_strings(benchmark::State& state)
   CHECK_CUDA(0);
 
   for (auto _ : state) {
-    cuda_event_timer raii(state, true, 0);
+    cuda_event_timer raii(state, true, rmm::cuda_stream_default);
     auto result = cudf::concatenate(column_views);
   }
 

cpp/benchmarks/join/join_benchmark.cu

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019, NVIDIA CORPORATION.
+ * Copyright (c) 2019-2021, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -103,7 +103,7 @@ static void BM_join(benchmark::State &state)
   // Benchmark the inner join operation
 
   for (auto _ : state) {
-    cuda_event_timer raii(state, true, 0);
+    cuda_event_timer raii(state, true, rmm::cuda_stream_default);
 
     auto result = cudf::inner_join(
       probe_table, build_table, columns_to_join, columns_to_join, cudf::null_equality::UNEQUAL);

cpp/benchmarks/sort/sort_benchmark.cpp

@@ -61,7 +61,7 @@ static void BM_sort(benchmark::State& state, bool nulls)
   auto input = cudf::table_view(column_views);
 
   for (auto _ : state) {
-    cuda_event_timer raii(state, true, 0);
+    cuda_event_timer raii(state, true, rmm::cuda_stream_default);
 
     auto result = (stable) ? cudf::stable_sorted_order(input) : cudf::sorted_order(input);
   }

cpp/benchmarks/sort/sort_strings_benchmark.cpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2020, NVIDIA CORPORATION.
+ * Copyright (c) 2020-2021, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -32,7 +32,7 @@ static void BM_sort(benchmark::State& state)
   auto const table = create_random_table({cudf::type_id::STRING}, 1, row_count{n_rows});
 
   for (auto _ : state) {
-    cuda_event_timer raii(state, true, 0);
+    cuda_event_timer raii(state, true, rmm::cuda_stream_default);
     cudf::sort(table->view());
   }
 }

cpp/benchmarks/string/case_benchmark.cpp

@@ -32,7 +32,7 @@ static void BM_case(benchmark::State& state)
   cudf::strings_column_view input(table->view().column(0));
 
   for (auto _ : state) {
-    cuda_event_timer raii(state, true, 0);
+    cuda_event_timer raii(state, true, rmm::cuda_stream_default);
     cudf::strings::to_lower(input);
   }
 

cpp/benchmarks/string/combine_benchmark.cpp

@@ -43,7 +43,7 @@ static void BM_combine(benchmark::State& state)
   cudf::string_scalar separator("+");
 
   for (auto _ : state) {
-    cuda_event_timer raii(state, true, 0);
+    cuda_event_timer raii(state, true, rmm::cuda_stream_default);
     cudf::strings::concatenate(table->view(), separator);
   }
 

cpp/benchmarks/string/contains_benchmark.cpp

@@ -35,7 +35,7 @@ static void BM_contains(benchmark::State& state, contains_type ct)
   cudf::strings_column_view input(table->view().column(0));
 
   for (auto _ : state) {
-    cuda_event_timer raii(state, true, 0);
+    cuda_event_timer raii(state, true, rmm::cuda_stream_default);
     // contains_re(), matches_re(), and count_re() all have similar functions
     // with count_re() being the most regex intensive
     switch (ct) {

cpp/benchmarks/string/copy_benchmark.cpp

@@ -54,7 +54,7 @@ static void BM_copy(benchmark::State& state, copy_type ct)
                                                                     host_map_data.end());
 
   for (auto _ : state) {
-    cuda_event_timer raii(state, true, 0);
+    cuda_event_timer raii(state, true, rmm::cuda_stream_default);
     switch (ct) {
       case gather: cudf::gather(source->view(), index_map); break;
       case scatter: cudf::scatter(source->view(), index_map, target->view()); break;

cpp/benchmarks/string/extract_benchmark.cpp

@@ -44,7 +44,7 @@ static void BM_extract(benchmark::State& state, int re_instructions)
   std::string const pattern = "(" + raw_pattern.substr(0, re_instructions) + ")";
 
   for (auto _ : state) {
-    cuda_event_timer raii(state, true, 0);
+    cuda_event_timer raii(state, true, rmm::cuda_stream_default);
     auto results = cudf::strings::extract(input, pattern);
   }
 

cpp/benchmarks/string/factory_benchmark.cu

@@ -63,7 +63,7 @@ static void BM_factory(benchmark::State& state)
                     string_view_to_pair{});
 
   for (auto _ : state) {
-    cuda_event_timer raii(state, true, 0);
+    cuda_event_timer raii(state, true, rmm::cuda_stream_default);
     cudf::make_strings_column(pairs);
   }
 

cpp/benchmarks/string/filter_benchmark.cpp

@@ -50,7 +50,7 @@ static void BM_filter_chars(benchmark::State& state, FilterAPI api)
     {cudf::char_utf8{'a'}, cudf::char_utf8{'c'}}};
 
   for (auto _ : state) {
-    cuda_event_timer raii(state, true, 0);
+    cuda_event_timer raii(state, true, rmm::cuda_stream_default);
     switch (api) {
       case filter: cudf::strings::filter_characters_of_type(input, types); break;
       case filter_chars: cudf::strings::filter_characters(input, filter_table); break;

cpp/benchmarks/string/find_benchmark.cpp

@@ -46,7 +46,7 @@ static void BM_find_scalar(benchmark::State& state, FindAPI find_api)
   cudf::test::strings_column_wrapper targets({"+", "-"});
 
   for (auto _ : state) {
-    cuda_event_timer raii(state, true, 0);
+    cuda_event_timer raii(state, true, rmm::cuda_stream_default);
     switch (find_api) {
       case find: cudf::strings::find(input, target); break;
       case find_multi:

cpp/benchmarks/string/replace_benchmark.cpp

@@ -49,7 +49,7 @@ static void BM_replace(benchmark::State& state, replace_type rt)
   cudf::test::strings_column_wrapper repls({"", ""});
 
   for (auto _ : state) {
-    cuda_event_timer raii(state, true, 0);
+    cuda_event_timer raii(state, true, rmm::cuda_stream_default);
     switch (rt) {
       case scalar: cudf::strings::replace(input, target, repl); break;
       case slice: cudf::strings::replace_slice(input, repl, 1, 10); break;

cpp/benchmarks/string/replace_re_benchmark.cpp

@@ -43,7 +43,7 @@ static void BM_replace(benchmark::State& state, replace_type rt)
   cudf::test::strings_column_wrapper repls({"#", ""});
 
   for (auto _ : state) {
-    cuda_event_timer raii(state, true, 0);
+    cuda_event_timer raii(state, true, rmm::cuda_stream_default);
     switch (rt) {
       case replace_type::replace_re:  // contains_re and matches_re use the same main logic
         cudf::strings::replace_re(input, "\\d+");

cpp/benchmarks/string/split_benchmark.cpp

@@ -44,7 +44,7 @@ static void BM_split(benchmark::State& state, split_type rt)
   cudf::string_scalar target("+");
 
   for (auto _ : state) {
-    cuda_event_timer raii(state, true, 0);
+    cuda_event_timer raii(state, true, rmm::cuda_stream_default);
     switch (rt) {
       case split: cudf::strings::split(input, target); break;
       case split_ws: cudf::strings::split(input); break;

cpp/benchmarks/string/substring_benchmark.cpp

@@ -54,7 +54,7 @@ static void BM_substring(benchmark::State& state, substring_type rt)
   cudf::test::strings_column_wrapper delimiters(delim_itr, delim_itr + n_rows);
 
   for (auto _ : state) {
-    cuda_event_timer raii(state, true, 0);
+    cuda_event_timer raii(state, true, rmm::cuda_stream_default);
     switch (rt) {
       case position: cudf::strings::slice_strings(input, 1, max_str_length / 2); break;
       case multi_position: cudf::strings::slice_strings(input, starts, stops); break;

cpp/benchmarks/string/translate_benchmark.cpp

@@ -54,7 +54,7 @@ static void BM_translate(benchmark::State& state, int entry_count)
                  });
 
   for (auto _ : state) {
-    cuda_event_timer raii(state, true, 0);
+    cuda_event_timer raii(state, true, rmm::cuda_stream_default);
     cudf::strings::translate(input, entries);
   }
 

cpp/benchmarks/string/url_decode_benchmark.cpp

@@ -80,7 +80,7 @@ void BM_url_decode(benchmark::State& state)
   auto strings_view = cudf::strings_column_view(column);
 
   for (auto _ : state) {
-    cuda_event_timer raii(state, true, 0);
+    cuda_event_timer raii(state, true, rmm::cuda_stream_default);
     auto result = cudf::strings::url_decode(strings_view);
   }
 

cpp/benchmarks/text/normalize_benchmark.cpp

@@ -41,7 +41,7 @@ static void BM_normalize(benchmark::State& state, bool to_lower)
   cudf::strings_column_view input(table->view().column(0));
 
   for (auto _ : state) {
-    cuda_event_timer raii(state, true, 0);
+    cuda_event_timer raii(state, true, rmm::cuda_stream_default);
     nvtext::normalize_characters(input, to_lower);
   }
 

cpp/benchmarks/text/normalize_spaces_benchmark.cpp

@@ -42,7 +42,7 @@ static void BM_normalize(benchmark::State& state)
   cudf::strings_column_view input(table->view().column(0));
 
   for (auto _ : state) {
-    cuda_event_timer raii(state, true, 0);
+    cuda_event_timer raii(state, true, rmm::cuda_stream_default);
     nvtext::normalize_spaces(input);
   }
 

cpp/benchmarks/text/tokenize_benchmark.cpp

@@ -46,7 +46,7 @@ static void BM_tokenize(benchmark::State& state, tokenize_type tt)
   cudf::test::strings_column_wrapper delimiters({" ", "+", "-"});
 
   for (auto _ : state) {
-    cuda_event_timer raii(state, true, 0);
+    cuda_event_timer raii(state, true, rmm::cuda_stream_default);
     switch (tt) {
       case tokenize_type::single: nvtext::tokenize(input); break;
       case tokenize_type::multi:

cpp/include/cudf/scalar/scalar.hpp

@@ -151,7 +151,7 @@ class fixed_width_scalar : public scalar {
   /**
    * @brief Implicit conversion operator to get the value of the scalar on the host
    */
-  explicit operator value_type() const { return this->value(0); }
+  explicit operator value_type() const { return this->value(rmm::cuda_stream_default); }
 
   /**
    * @brief Get the value of the scalar
@@ -449,7 +449,7 @@ class string_scalar : public scalar {
   /**
    * @brief Implicit conversion operator to get the value of the scalar in a host std::string
    */
-  explicit operator std::string() const { return this->to_string(0); }
+  explicit operator std::string() const { return this->to_string(rmm::cuda_stream_default); }
 
   /**
    * @brief Get the value of the scalar in a host std::string

cpp/include/cudf/types.hpp

@@ -64,8 +64,8 @@ class list_scalar;
 class string_scalar;
 template <typename T> class numeric_scalar;
 template <typename T> class fixed_point_scalar;
-template <typename T> class timestamp_scalar;
-template <typename T> class duration_scalar;
+template <typename T> struct timestamp_scalar;
+template <typename T> struct duration_scalar;
 
 class string_scalar_device_view;
 template <typename T> class numeric_scalar_device_view;

cpp/src/copying/concatenate.cu

@@ -455,7 +455,8 @@ rmm::device_buffer concatenate_masks(host_span<column_view const> views,
     rmm::device_buffer null_mask =
       create_null_mask(total_element_count, mask_state::UNINITIALIZED, mr);
 
-    detail::concatenate_masks(views, static_cast<bitmask_type*>(null_mask.data()), 0);
+    detail::concatenate_masks(
+      views, static_cast<bitmask_type*>(null_mask.data()), rmm::cuda_stream_default);
 
     return null_mask;
   }

cpp/src/copying/pack.cpp

@@ -17,6 +17,8 @@
 #include <cudf/detail/copy.hpp>
 #include <cudf/detail/nvtx/ranges.hpp>
 
+#include <rmm/cuda_stream_view.hpp>
+
 namespace cudf {
 namespace detail {
 
@@ -216,7 +218,7 @@ table_view unpack(uint8_t const* metadata, uint8_t const* gpu_data)
 packed_columns pack(cudf::table_view const& input, rmm::mr::device_memory_resource* mr)
 {
   CUDF_FUNC_RANGE();
-  return detail::pack(input, 0, mr);
+  return detail::pack(input, rmm::cuda_stream_default, mr);
 }
 
 /**

cpp/src/io/csv/writer_impl.cu

@@ -37,6 +37,7 @@
 
 #include <rmm/cuda_stream_view.hpp>
 #include <rmm/exec_policy.hpp>
+#include <rmm/mr/device/per_device_resource.hpp>
 
 #include <thrust/count.h>
 #include <thrust/execution_policy.h>
@@ -211,10 +212,11 @@ struct column_to_strings_fn {
                (cudf::is_timestamp<column_type>()) || (cudf::is_duration<column_type>()));
   }
 
-  explicit column_to_strings_fn(csv_writer_options const& options,
-                                rmm::mr::device_memory_resource* mr = nullptr,
-                                rmm::cuda_stream_view stream        = nullptr)
-    : options_(options), mr_(mr), stream_(stream)
+  explicit column_to_strings_fn(
+    csv_writer_options const& options,
+    rmm::cuda_stream_view stream        = rmm::cuda_stream_default,
+    rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource())
+    : options_(options), stream_(stream), mr_(mr)
   {
   }
 
@@ -345,8 +347,8 @@ struct column_to_strings_fn {
 
  private:
   csv_writer_options const& options_;
-  rmm::mr::device_memory_resource* mr_;
   rmm::cuda_stream_view stream_;
+  rmm::mr::device_memory_resource* mr_;
 };
 }  // unnamed namespace
 
@@ -495,7 +497,7 @@ void writer::impl::write(table_view const& table,
 
     // convert each chunk to CSV:
     //
-    column_to_strings_fn converter{options_, mr_};
+    column_to_strings_fn converter{options_, stream, mr_};
     for (auto&& sub_view : vector_views) {
       // Skip if the table has no rows
       if (sub_view.num_rows() == 0) continue;

cpp/src/replace/nans.cu

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2020, NVIDIA CORPORATION.
+ * Copyright (c) 2020-2021, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -146,7 +146,7 @@ std::unique_ptr<column> replace_nans(column_view const& input,
                                      rmm::mr::device_memory_resource* mr)
 {
   CUDF_FUNC_RANGE();
-  return detail::replace_nans(input, replacement, 0, mr);
+  return detail::replace_nans(input, replacement, rmm::cuda_stream_default, mr);
 }
 
 std::unique_ptr<column> replace_nans(column_view const& input,

cpp/src/strings/substring.cu

@@ -402,8 +402,9 @@ std::unique_ptr<column> slice_strings(strings_column_view const& strings,
                "Strings and delimiters column sizes do not match");
 
   CUDF_FUNC_RANGE();
-  auto delimiters_dev_view_ptr = cudf::column_device_view::create(delimiters.parent(), 0);
-  auto delimiters_dev_view     = *delimiters_dev_view_ptr;
+  auto delimiters_dev_view_ptr =
+    cudf::column_device_view::create(delimiters.parent(), rmm::cuda_stream_default);
+  auto delimiters_dev_view = *delimiters_dev_view_ptr;
   return (delimiters_dev_view.nullable())
            ? detail::slice_strings(
                strings,

cpp/src/text/subword/subword_tokenize.cu

@@ -265,7 +265,7 @@ tokenizer_result subword_tokenize(cudf::strings_column_view const& strings,
                                   do_lower_case,
                                   do_truncate,
                                   max_rows_tensor,
-                                  0,
+                                  rmm::cuda_stream_default,
                                   mr);
 }
 
@@ -286,7 +286,7 @@ tokenizer_result subword_tokenize(cudf::strings_column_view const& strings,
                                   do_lower_case,
                                   do_truncate,
                                   max_rows_tensor,
-                                  0,
+                                  rmm::cuda_stream_default,
                                   mr);
 }
 

cpp/tests/copying/gather_struct_tests.cu

@@ -189,7 +189,8 @@ TYPED_TEST(TypedStructGatherTest, TestGatherStructOfLists)
       cudf::detail::make_counting_transform_iterator(0, [](auto i) { return !(i % 3); })};
   };
 
-  auto lists_column = std::make_unique<cudf::column>(cudf::column(lists_column_exemplar(), 0));
+  auto lists_column =
+    std::make_unique<cudf::column>(cudf::column(lists_column_exemplar(), rmm::cuda_stream_default));
 
   // Assemble struct column.
   std::vector<std::unique_ptr<cudf::column>> vector_of_columns;
@@ -242,7 +243,8 @@ TYPED_TEST(TypedStructGatherTest, TestGatherStructOfListsOfLists)
       cudf::detail::make_counting_transform_iterator(0, [](auto i) { return !(i % 3); })};
   };
 
-  auto lists_column = std::make_unique<cudf::column>(cudf::column(lists_column_exemplar(), 0));
+  auto lists_column =
+    std::make_unique<cudf::column>(cudf::column(lists_column_exemplar(), rmm::cuda_stream_default));
 
   // Assemble struct column.
   std::vector<std::unique_ptr<cudf::column>> vector_of_columns;