Remove any references to determinism

Author: gonidelis

cub/cub/device/device_run_length_encode.cuh

@@ -31,7 +31,6 @@
 
 #include <cuda/std/__execution/env.h>
 #include <cuda/std/__type_traits/enable_if.h>
-#include <cuda/std/__type_traits/is_integral.h>
 #include <cuda/std/__type_traits/is_same.h>
 
 CUB_NAMESPACE_BEGIN
@@ -231,9 +230,6 @@ struct DeviceRunLengthEncode
   //!   - ``[d_num_runs_out, d_num_runs_out + 1)``
   //!   - ``[d_in, d_in + num_items)``
   //!
-  //! - The algorithm is deterministic: it produces the same output for the same input across runs
-  //!   on the same or different GPUs. No determinism requirements need to be specified via the environment.
-  //!
   //! Snippet
   //! +++++++++++++++++++++++++++++++++++++++++++++
   //!
@@ -480,9 +476,6 @@ struct DeviceRunLengthEncode
   //!   - ``[d_num_runs_out, d_num_runs_out + 1)``
   //!   - ``[d_in, d_in + num_items)``
   //!
-  //! - The algorithm is deterministic: it produces the same output for the same input across runs
-  //!   on the same or different GPUs. No determinism requirements need to be specified via the environment.
-  //!
   //! Snippet
   //! +++++++++++++++++++++++++++++++++++++++++++++
   //!

cub/cub/device/device_select.cuh

@@ -23,7 +23,6 @@
 #include <cub/device/dispatch/dispatch_select_if.cuh>
 #include <cub/device/dispatch/dispatch_unique_by_key.cuh>
 
-#include <cuda/__execution/determinism.h>
 #include <cuda/__execution/require.h>
 #include <cuda/std/__execution/env.h>
 #include <cuda/std/__functional/operations.h>

cub/test/catch2_test_device_run_length_encode_env.cu

@@ -33,10 +33,9 @@ TEST_CASE("DeviceRunLengthEncode::Encode works with default environment", "[run_
   auto d_counts_out   = c2h::device_vector<int>(8);
   auto d_num_runs_out = c2h::device_vector<int>(1);
 
-  REQUIRE(
-    cudaSuccess
-    == cub::DeviceRunLengthEncode::Encode(
-      d_in.begin(), d_unique_out.begin(), d_counts_out.begin(), d_num_runs_out.begin(), (int) d_in.size()));
+  REQUIRE(cudaSuccess
+          == cub::DeviceRunLengthEncode::Encode(
+            d_in.begin(), d_unique_out.begin(), d_counts_out.begin(), d_num_runs_out.begin(), (int) d_in.size()));
 
   c2h::device_vector<int> expected_unique{0, 2, 9, 5, 8};
   c2h::device_vector<int> expected_counts{1, 2, 1, 3, 1};
@@ -56,10 +55,9 @@ TEST_CASE("DeviceRunLengthEncode::NonTrivialRuns works with default environment"
   auto d_lengths_out  = c2h::device_vector<int>(8);
   auto d_num_runs_out = c2h::device_vector<int>(1);
 
-  REQUIRE(
-    cudaSuccess
-    == cub::DeviceRunLengthEncode::NonTrivialRuns(
-      d_in.begin(), d_offsets_out.begin(), d_lengths_out.begin(), d_num_runs_out.begin(), (int) d_in.size()));
+  REQUIRE(cudaSuccess
+          == cub::DeviceRunLengthEncode::NonTrivialRuns(
+            d_in.begin(), d_offsets_out.begin(), d_lengths_out.begin(), d_num_runs_out.begin(), (int) d_in.size()));
 
   c2h::device_vector<int> expected_offsets{1, 4};
   c2h::device_vector<int> expected_lengths{2, 3};

cub/test/catch2_test_device_run_length_encode_env_api.cu

@@ -2,7 +2,6 @@
 // SPDX-License-Identifier: BSD-3-Clause
 
 #include "insert_nested_NVTX_range_guard.h"
-// above header needs to be included first
 
 #include <cub/device/device_run_length_encode.cuh>