Merge branch 'branch-25.02' into cudf-polars-multi-groupby

Author: rjzamora

cpp/include/cudf/detail/utilities/integer_utils.hpp

@@ -1,7 +1,7 @@
 /*
  * Copyright 2019 BlazingDB, Inc.
  *     Copyright 2019 Eyal Rozenberg <[email protected]>
- * Copyright (c) 2020-2024, NVIDIA CORPORATION.
+ * Copyright (c) 2020-2025, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -23,6 +23,8 @@
  */
 
 #include <cudf/fixed_point/temporary.hpp>
+#include <cudf/types.hpp>
+#include <cudf/utilities/error.hpp>
 
 #include <cmath>
 #include <cstdlib>
@@ -44,13 +46,17 @@ namespace util {
  * `modulus` is positive. The safety is in regard to rollover.
  */
 template <typename S>
-constexpr S round_up_safe(S number_to_round, S modulus)
+CUDF_HOST_DEVICE constexpr S round_up_safe(S number_to_round, S modulus)
 {
   auto remainder = number_to_round % modulus;
   if (remainder == 0) { return number_to_round; }
   auto rounded_up = number_to_round - remainder + modulus;
   if (rounded_up < number_to_round) {
-    throw std::invalid_argument("Attempt to round up beyond the type's maximum value");
+#ifndef __CUDA_ARCH__
+    CUDF_FAIL("Attempt to round up beyond the type's maximum value", cudf::data_type_error);
+#else
+    CUDF_UNREACHABLE("Attempt to round up beyond the type's maximum value");
+#endif
   }
   return rounded_up;
 }

cpp/include/cudf/utilities/span.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2020-2024, NVIDIA CORPORATION.
+ * Copyright (c) 2020-2025, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -197,11 +197,16 @@ struct host_span : public cudf::detail::span_base<T, Extent, host_span<T, Extent
 
   constexpr host_span() noexcept : base() {}  // required to compile on centos
 
-  /// Constructor from pointer and size
-  /// @param data Pointer to the first element in the span
-  /// @param size The number of elements in the span
-  /// @param is_device_accessible Whether the data is device accessible (e.g. pinned memory)
-  constexpr host_span(T* data, std::size_t size, bool is_device_accessible)
+  /**
+   * @brief Constructor from pointer and size
+   *
+   * @note This needs to be host-device , as it's used by a host-device function in base_2dspan
+   *
+   * @param data Pointer to the first element in the span
+   * @param size The number of elements in the span
+   * @param is_device_accessible Whether the data is device accessible (e.g. pinned memory)
+   */
+  CUDF_HOST_DEVICE constexpr host_span(T* data, std::size_t size, bool is_device_accessible)
     : base(data, size), _is_device_accessible{is_device_accessible}
   {
   }
@@ -311,8 +316,8 @@ struct host_span : public cudf::detail::span_base<T, Extent, host_span<T, Extent
    * @param count The number of elements in the subspan
    * @return A subspan of the sequence, of requested count and offset
    */
-  [[nodiscard]] constexpr host_span subspan(typename base::size_type offset,
-                                            typename base::size_type count) const noexcept
+  [[nodiscard]] CUDF_HOST_DEVICE constexpr host_span subspan(
+    typename base::size_type offset, typename base::size_type count) const noexcept
   {
     return host_span{this->data() + offset, count, _is_device_accessible};
   }
@@ -434,8 +439,8 @@ struct device_span : public cudf::detail::span_base<T, Extent, device_span<T, Ex
    * @param count The number of elements in the subspan
    * @return A subspan of the sequence, of requested count and offset
    */
-  [[nodiscard]] constexpr device_span subspan(typename base::size_type offset,
-                                              typename base::size_type count) const noexcept
+  [[nodiscard]] CUDF_HOST_DEVICE constexpr device_span subspan(
+    typename base::size_type offset, typename base::size_type count) const noexcept
   {
     return device_span{this->data() + offset, count};
   }
@@ -475,28 +480,28 @@ class base_2dspan {
    *
    * @return A pointer to the first element of the span
    */
-  [[nodiscard]] constexpr auto data() const noexcept { return _flat.data(); }
+  [[nodiscard]] CUDF_HOST_DEVICE constexpr auto data() const noexcept { return _flat.data(); }
 
   /**
    * @brief Returns the size in the span as pair.
    *
    * @return pair representing rows and columns size of the span
    */
-  [[nodiscard]] constexpr auto size() const noexcept { return _size; }
+  [[nodiscard]] CUDF_HOST_DEVICE constexpr auto size() const noexcept { return _size; }
 
   /**
    * @brief Returns the number of elements in the span.
    *
    * @return Number of elements in the span
    */
-  [[nodiscard]] constexpr auto count() const noexcept { return _flat.size(); }
+  [[nodiscard]] CUDF_HOST_DEVICE constexpr auto count() const noexcept { return _flat.size(); }
 
   /**
    * @brief Checks if the span is empty.
    *
    * @return True if the span is empty, false otherwise
    */
-  [[nodiscard]] constexpr bool is_empty() const noexcept { return count() == 0; }
+  [[nodiscard]] CUDF_HOST_DEVICE constexpr bool is_empty() const noexcept { return count() == 0; }
 
   /**
    * @brief Returns a reference to the row-th element of the sequence.
@@ -507,7 +512,7 @@ class base_2dspan {
    * @param row the index of the element to access
    * @return A reference to the row-th element of the sequence, i.e., `data()[row]`
    */
-  constexpr RowType<T, dynamic_extent> operator[](size_t row) const
+  CUDF_HOST_DEVICE constexpr RowType<T, dynamic_extent> operator[](size_t row) const
   {
     return _flat.subspan(row * _size.second, _size.second);
   }
@@ -517,7 +522,10 @@ class base_2dspan {
    *
    * @return A flattened span of the 2D span
    */
-  [[nodiscard]] constexpr RowType<T, dynamic_extent> flat_view() const { return _flat; }
+  [[nodiscard]] CUDF_HOST_DEVICE constexpr RowType<T, dynamic_extent> flat_view() const
+  {
+    return _flat;
+  }
 
   /**
    * @brief Construct a 2D span from another 2D span of convertible type

cpp/src/io/json/read_json.cu

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2022-2024, NVIDIA CORPORATION.
+ * Copyright (c) 2022-2025, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -30,19 +30,33 @@
 #include <cudf/utilities/memory_resource.hpp>
 #include <cudf/utilities/span.hpp>
 
+#include <rmm/cuda_stream_pool.hpp>
 #include <rmm/device_uvector.hpp>
 #include <rmm/exec_policy.hpp>
 
 #include <thrust/distance.h>
 #include <thrust/iterator/constant_iterator.h>
 #include <thrust/scatter.h>
 
+#include <BS_thread_pool.hpp>
+#include <BS_thread_pool_utils.hpp>
+
 #include <numeric>
 
 namespace cudf::io::json::detail {
 
 namespace {
 
+namespace pools {
+
+BS::thread_pool& tpool()
+{
+  static BS::thread_pool _tpool(std::thread::hardware_concurrency());
+  return _tpool;
+}
+
+}  // namespace pools
+
 class compressed_host_buffer_source final : public datasource {
  public:
   explicit compressed_host_buffer_source(std::unique_ptr<datasource> const& src,
@@ -51,8 +65,8 @@ class compressed_host_buffer_source final : public datasource {
   {
     auto ch_buffer = host_span<uint8_t const>(reinterpret_cast<uint8_t const*>(_dbuf_ptr->data()),
                                               _dbuf_ptr->size());
-    if (comptype == compression_type::GZIP || comptype == compression_type::ZIP ||
-        comptype == compression_type::SNAPPY) {
+    if (_comptype == compression_type::GZIP || _comptype == compression_type::ZIP ||
+        _comptype == compression_type::SNAPPY) {
       _decompressed_ch_buffer_size = cudf::io::detail::get_uncompressed_size(_comptype, ch_buffer);
     } else {
       _decompressed_buffer         = cudf::io::detail::decompress(_comptype, ch_buffer);
@@ -96,7 +110,22 @@ class compressed_host_buffer_source final : public datasource {
     return std::make_unique<non_owning_buffer>(_decompressed_buffer.data() + offset, count);
   }
 
-  [[nodiscard]] bool supports_device_read() const override { return false; }
+  std::future<size_t> device_read_async(size_t offset,
+                                        size_t size,
+                                        uint8_t* dst,
+                                        rmm::cuda_stream_view stream) override
+  {
+    auto& thread_pool = pools::tpool();
+    return thread_pool.submit_task([this, offset, size, dst, stream] {
+      auto hbuf = host_read(offset, size);
+      CUDF_CUDA_TRY(
+        cudaMemcpyAsync(dst, hbuf->data(), hbuf->size(), cudaMemcpyHostToDevice, stream.value()));
+      stream.synchronize();
+      return hbuf->size();
+    });
+  }
+
+  [[nodiscard]] bool supports_device_read() const override { return true; }
 
   [[nodiscard]] size_t size() const override { return _decompressed_ch_buffer_size; }
 
@@ -431,6 +460,8 @@ device_span<char> ingest_raw_input(device_span<char> buffer,
   // line of file i+1 don't end up on the same JSON line, if file i does not already end with a line
   // delimiter.
   auto constexpr num_delimiter_chars = 1;
+  std::vector<std::future<size_t>> thread_tasks;
+  auto stream_pool = cudf::detail::fork_streams(stream, pools::tpool().get_thread_count());
 
   auto delimiter_map = cudf::detail::make_empty_host_vector<std::size_t>(sources.size(), stream);
   std::vector<std::size_t> prefsum_source_sizes(sources.size());
@@ -447,13 +478,17 @@ device_span<char> ingest_raw_input(device_span<char> buffer,
 
   auto const total_bytes_to_read = std::min(range_size, prefsum_source_sizes.back() - range_offset);
   range_offset -= start_source ? prefsum_source_sizes[start_source - 1] : 0;
-  for (std::size_t i = start_source; i < sources.size() && bytes_read < total_bytes_to_read; i++) {
+  for (std::size_t i = start_source, cur_stream = 0;
+       i < sources.size() && bytes_read < total_bytes_to_read;
+       i++) {
     if (sources[i]->is_empty()) continue;
     auto data_size = std::min(sources[i]->size() - range_offset, total_bytes_to_read - bytes_read);
     auto destination = reinterpret_cast<uint8_t*>(buffer.data()) + bytes_read +
                        (num_delimiter_chars * delimiter_map.size());
-    if (sources[i]->is_device_read_preferred(data_size)) {
-      bytes_read += sources[i]->device_read(range_offset, data_size, destination, stream);
+    if (sources[i]->supports_device_read()) {
+      thread_tasks.emplace_back(sources[i]->device_read_async(
+        range_offset, data_size, destination, stream_pool[cur_stream++ % stream_pool.size()]));
+      bytes_read += data_size;
     } else {
       h_buffers.emplace_back(sources[i]->host_read(range_offset, data_size));
       auto const& h_buffer = h_buffers.back();
@@ -481,6 +516,15 @@ device_span<char> ingest_raw_input(device_span<char> buffer,
                     buffer.data());
   }
   stream.synchronize();
+
+  if (thread_tasks.size()) {
+    auto const bytes_read = std::accumulate(
+      thread_tasks.begin(), thread_tasks.end(), std::size_t{0}, [](std::size_t sum, auto& task) {
+        return sum + task.get();
+      });
+    CUDF_EXPECTS(bytes_read == total_bytes_to_read, "something's fishy");
+  }
+
   return buffer.first(bytes_read + (delimiter_map.size() * num_delimiter_chars));
 }
 
@@ -505,10 +549,17 @@ table_with_metadata read_json(host_span<std::unique_ptr<datasource>> sources,
     return read_json_impl(sources, reader_opts, stream, mr);
 
   std::vector<std::unique_ptr<datasource>> compressed_sources;
-  for (size_t i = 0; i < sources.size(); i++) {
-    compressed_sources.emplace_back(
-      std::make_unique<compressed_host_buffer_source>(sources[i], reader_opts.get_compression()));
+  std::vector<std::future<std::unique_ptr<compressed_host_buffer_source>>> thread_tasks;
+  auto& thread_pool = pools::tpool();
+  for (auto& src : sources) {
+    thread_tasks.emplace_back(thread_pool.submit_task([&reader_opts, &src] {
+      return std::make_unique<compressed_host_buffer_source>(src, reader_opts.get_compression());
+    }));
   }
+  std::transform(thread_tasks.begin(),
+                 thread_tasks.end(),
+                 std::back_inserter(compressed_sources),
+                 [](auto& task) { return task.get(); });
   // in read_json_impl, we need the compressed source size to actually be the
   // uncompressed source size for correct batching
   return read_json_impl(compressed_sources, reader_opts, stream, mr);

cpp/src/io/parquet/reader_impl.cpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019-2024, NVIDIA CORPORATION.
+ * Copyright (c) 2019-2025, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -97,38 +97,24 @@ void reader::impl::decode_page_data(read_mode mode, size_t skip_rows, size_t num
                              _stream);
     }
 
-    // Compute column string sizes (using page string offsets) for this subpass
+    // Compute column string sizes (using page string offsets) for this output table chunk
     col_string_sizes = calculate_page_string_offsets();
 
-    // ensure cumulative column string sizes have been initialized
-    if (pass.cumulative_col_string_sizes.empty()) {
-      pass.cumulative_col_string_sizes.resize(_input_columns.size(), 0);
-    }
-
-    // Add to the cumulative column string sizes of this pass
-    std::transform(pass.cumulative_col_string_sizes.begin(),
-                   pass.cumulative_col_string_sizes.end(),
-                   col_string_sizes.begin(),
-                   pass.cumulative_col_string_sizes.begin(),
-                   std::plus<>{});
-
     // Check for overflow in cumulative column string sizes of this pass so that the page string
     // offsets of overflowing (large) string columns are treated as 64-bit.
     auto const threshold         = static_cast<size_t>(strings::detail::get_offset64_threshold());
-    auto const has_large_strings = std::any_of(pass.cumulative_col_string_sizes.cbegin(),
-                                               pass.cumulative_col_string_sizes.cend(),
+    auto const has_large_strings = std::any_of(col_string_sizes.cbegin(),
+                                               col_string_sizes.cend(),
                                                [=](std::size_t sz) { return sz > threshold; });
     if (has_large_strings and not strings::detail::is_large_strings_enabled()) {
       CUDF_FAIL("String column exceeds the column size limit", std::overflow_error);
     }
 
-    // Mark any chunks for which the cumulative column string size has exceeded the
-    // large strings threshold
-    if (has_large_strings) {
-      for (auto& chunk : pass.chunks) {
-        auto const idx = chunk.src_col_index;
-        if (pass.cumulative_col_string_sizes[idx] > threshold) { chunk.is_large_string_col = true; }
-      }
+    // Mark/unmark column-chunk descriptors depending on the string sizes of corresponding output
+    // column chunks and the large strings threshold.
+    for (auto& chunk : pass.chunks) {
+      auto const idx            = chunk.src_col_index;
+      chunk.is_large_string_col = (col_string_sizes[idx] > threshold);
     }
   }
 
@@ -210,11 +196,9 @@ void reader::impl::decode_page_data(read_mode mode, size_t skip_rows, size_t num
         // only do string buffer for leaf
         if (idx == max_depth - 1 and out_buf.string_size() == 0 and
             col_string_sizes[pass.chunks[c].src_col_index] > 0) {
-          out_buf.create_string_data(
-            col_string_sizes[pass.chunks[c].src_col_index],
-            pass.cumulative_col_string_sizes[pass.chunks[c].src_col_index] >
-              static_cast<size_t>(strings::detail::get_offset64_threshold()),
-            _stream);
+          out_buf.create_string_data(col_string_sizes[pass.chunks[c].src_col_index],
+                                     pass.chunks[c].is_large_string_col,
+                                     _stream);
         }
         if (has_strings) { str_data[idx] = out_buf.string_data(); }
         out_buf.user_data |=
@@ -416,11 +400,11 @@ void reader::impl::decode_page_data(read_mode mode, size_t skip_rows, size_t num
         final_offsets.emplace_back(offset);
         out_buf.user_data |= PARQUET_COLUMN_BUFFER_FLAG_LIST_TERMINATED;
       } else if (out_buf.type.id() == type_id::STRING) {
-        // need to cap off the string offsets column
-        auto const sz = static_cast<size_type>(col_string_sizes[idx]);
-        if (sz <= strings::detail::get_offset64_threshold()) {
+        // only if it is not a large strings column
+        if (col_string_sizes[idx] <=
+            static_cast<size_t>(strings::detail::get_offset64_threshold())) {
           out_buffers.emplace_back(static_cast<size_type*>(out_buf.data()) + out_buf.size);
-          final_offsets.emplace_back(sz);
+          final_offsets.emplace_back(static_cast<size_type>(col_string_sizes[idx]));
         }
       }
     }

cpp/src/io/parquet/reader_impl_chunking.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2023-2024, NVIDIA CORPORATION.
+ * Copyright (c) 2023-2025, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -130,9 +130,6 @@ struct pass_intermediate_data {
   rmm::device_buffer decomp_dict_data{0, cudf::get_default_stream()};
   rmm::device_uvector<string_index_pair> str_dict_index{0, cudf::get_default_stream()};
 
-  // cumulative strings column sizes.
-  std::vector<size_t> cumulative_col_string_sizes{};
-
   int level_type_size{0};
 
   // skip_rows / num_rows for this pass.