Fix: Compiling w/out Intel TBB

Author: ashvardanian

.vscode/settings.json

@@ -17,7 +17,7 @@
     "bootcamps",
     "Byrne",
     "Cawley",
-    "cblas",
+    "cBLAS",
     "CCCL",
     "chriskohlhoffasio",
     "clflush",
@@ -29,7 +29,8 @@
     "cppcoro",
     "cstddef",
     "CTRE",
-    "cublas",
+    "cuBLAS",
+    "cuBLASLt",
     "CUDA",
     "denormal",
     "DOTPROD",
@@ -97,6 +98,7 @@
     "prefetcher",
     "pthread",
     "PTXAS",
+    "quadpair",
     "RDMA",
     "reorderable",
     "Shankhdhar",
@@ -127,6 +129,7 @@
     "VHDL",
     "VNNI",
     "VPCLMULQDQ",
+    "warpgroup",
     "Weis",
     "WGMMA",
     "wmma",

README.md

@@ -83,6 +83,13 @@ The build will pull and compile several third-party dependencies from the source
 - Nvidia's [CCCL](https://github.com/nvidia/cccl) for GPU-accelerated algorithms.
 - Nvidia's [CUTLASS](https://github.com/nvidia/cutlass) for GPU-accelerated Linear Algebra.
 
+To build without Parallel STL, Intel TBB, and CUDA:
+
+```sh
+cmake -B build_release -D CMAKE_BUILD_TYPE=Release -D USE_INTEL_TBB=OFF -D USE_NVIDIA_CCCL=OFF
+cmake --build build_release --config Release
+```
+
 To control the output or run specific benchmarks, use the following flags:
 
 ```sh

less_slow.cpp

@@ -465,8 +465,11 @@ BENCHMARK(sorting)->Args({8196, false})->Args({8196, true});
  *
  *  @see Feature testing macros: https://en.cppreference.com/w/cpp/utility/feature_test
  */
+#if !defined(USE_INTEL_TBB)
+#define USE_INTEL_TBB 1
+#endif // !defined(USE_INTEL_TBB)
 
-#if defined(__cpp_lib_parallel_algorithm)
+#if defined(__cpp_lib_parallel_algorithm) && USE_INTEL_TBB
 #include <execution> // `std::execution::par_unseq`
 
 template <typename execution_policy_>
@@ -548,7 +551,7 @@ BENCHMARK_CAPTURE(sorting_with_executors, par_unseq, std::execution::par_unseq)
  *       by Bryce Adelstein Lelbach at CppCon 2016: https://youtu.be/Vck6kzWjY88
  */
 
-#endif // defined(__cpp_lib_parallel_algorithm)
+#endif // defined(__cpp_lib_parallel_algorithm) && USE_INTEL_TBB
 
 #if defined(_OPENMP)
 /**