Skip ORC and Parquet readers' benchmark cases that are not currently …

…supported (#10194)

Closes: #9961
Skipping the following:
- ORC reader: row selection through stripe selection
- ORC reader: lists column with row selection
- Parquet reader: row selection through row group selection
- Parquet reader: lists column with row selection

Authors:
  - Vukasin Milovanovic (https://github.com/vuule)

Approvers:
  - Devavret Makkar (https://github.com/devavret)
  - David Wendt (https://github.com/davidwendt)

URL: #10194

cpp/benchmarks/io/orc/orc_reader.cpp

@@ -88,13 +88,13 @@ void BM_orc_read_varying_options(benchmark::State& state)
   auto const use_np_dtypes = (flags & 2) != 0;
   auto const ts_type       = cudf::data_type{static_cast<cudf::type_id>(state.range(state_idx++))};
 
+  // skip_rows is not supported on nested types
   auto const data_types =
     dtypes_for_column_selection(get_type_or_group({int32_t(type_group_id::INTEGRAL_SIGNED),
                                                    int32_t(type_group_id::FLOATING_POINT),
                                                    int32_t(type_group_id::FIXED_POINT),
                                                    int32_t(type_group_id::TIMESTAMP),
-                                                   int32_t(cudf::type_id::STRING),
-                                                   int32_t(cudf::type_id::LIST)}),
+                                                   int32_t(cudf::type_id::STRING)}),
                                 col_sel);
   auto const tbl  = create_random_table(data_types, data_types.size(), table_size_bytes{data_size});
   auto const view = tbl->view();
@@ -181,11 +181,12 @@ BENCHMARK_REGISTER_F(OrcRead, column_selection)
   ->Unit(benchmark::kMillisecond)
   ->UseManualTime();
 
+// Need an API to get the number of stripes to enable row_selection::STRIPES here
 BENCHMARK_DEFINE_F(OrcRead, row_selection)
 (::benchmark::State& state) { BM_orc_read_varying_options(state); }
 BENCHMARK_REGISTER_F(OrcRead, row_selection)
   ->ArgsProduct({{int32_t(column_selection::ALL)},
-                 {int32_t(row_selection::STRIPES), int32_t(row_selection::NROWS)},
+                 {int32_t(row_selection::NROWS)},
                  {1, 8},
                  {0b11},  // defaults
                  {int32_t(cudf::type_id::EMPTY)}})

cpp/benchmarks/io/parquet/parquet_reader.cpp

@@ -88,13 +88,13 @@ void BM_parq_read_varying_options(benchmark::State& state)
   auto const use_pandas_metadata = (flags & 2) != 0;
   auto const ts_type = cudf::data_type{static_cast<cudf::type_id>(state.range(state_idx++))};
 
+  // No nested types here, because of https://github.com/rapidsai/cudf/issues/9970
   auto const data_types = dtypes_for_column_selection(
     get_type_or_group({static_cast<int32_t>(type_group_id::INTEGRAL),
                        static_cast<int32_t>(type_group_id::FLOATING_POINT),
                        static_cast<int32_t>(type_group_id::FIXED_POINT),
                        static_cast<int32_t>(type_group_id::TIMESTAMP),
-                       static_cast<int32_t>(cudf::type_id::STRING),
-                       static_cast<int32_t>(cudf::type_id::LIST)}),
+                       static_cast<int32_t>(cudf::type_id::STRING)}),
     col_sel);
   auto const tbl  = create_random_table(data_types, data_types.size(), table_size_bytes{data_size});
   auto const view = tbl->view();
@@ -181,20 +181,18 @@ BENCHMARK_REGISTER_F(ParquetRead, column_selection)
   ->Unit(benchmark::kMillisecond)
   ->UseManualTime();
 
-// Disabled until we add an API to read metadata from a parquet file and determine num row groups.
-// https://github.com/rapidsai/cudf/pull/9963#issuecomment-1004832863
-/*
+// row_selection::ROW_GROUPS disabled until we add an API to read metadata from a parquet file and
+// determine num row groups. https://github.com/rapidsai/cudf/pull/9963#issuecomment-1004832863
 BENCHMARK_DEFINE_F(ParquetRead, row_selection)
 (::benchmark::State& state) { BM_parq_read_varying_options(state); }
 BENCHMARK_REGISTER_F(ParquetRead, row_selection)
   ->ArgsProduct({{int32_t(column_selection::ALL)},
-                 {int32_t(row_selection::ROW_GROUPS), int32_t(row_selection::NROWS)},
+                 {int32_t(row_selection::NROWS)},
                  {1, 4},
                  {0b01},  // defaults
                  {int32_t(cudf::type_id::EMPTY)}})
   ->Unit(benchmark::kMillisecond)
   ->UseManualTime();
-*/
 
 BENCHMARK_DEFINE_F(ParquetRead, misc_options)
 (::benchmark::State& state) { BM_parq_read_varying_options(state); }