Benchmarks: Add Benchmark - Add the source code of rocm kernel launch overhead benchmark. (#136)

**Description**
Add the source code of rocm kernel launch overhead benchmark. 

**Major Revision**
- Revise cmake build logic to support both cuda and rocm

superbench/benchmarks/micro_benchmarks/kernel_launch_overhead/CMakeLists.txt

@@ -2,10 +2,39 @@
 # Licensed under the MIT License.
 
 cmake_minimum_required(VERSION 3.18)
-project(kernel_launch_overhead LANGUAGES CUDA CXX)
+
+project(kernel_launch_overhead LANGUAGES CXX)
 
 include(../cuda_common.cmake)
+find_package(CUDAToolkit QUIET)
+include(../rocm_common.cmake)
+find_package(HIP QUIET)
+
+# Cuda environment
+if(CUDAToolkit_FOUND)
+    message(STATUS "Found CUDA: " ${CUDAToolkit_VERSION})
+    enable_language(CUDA)
+
+    add_executable(kernel_launch_overhead kernel_launch.cu)
+    set_property(TARGET kernel_launch_overhead PROPERTY CUDA_ARCHITECTURES ${NVCC_ARCHS_SUPPORTED})
+    install(TARGETS kernel_launch_overhead RUNTIME DESTINATION bin)
+
+# ROCm environment
+elseif(HIP_FOUND)
+    message(STATUS "Found HIP: " ${HIP_VERSION})
+    
+    # Convert cuda code to hip code inplace
+    execute_process(COMMAND hipify-perl -inplace -print-stats kernel_launch.cu
+                     WORKING_DIRECTORY ${CMAKE_CURRENT_SOURCE_DIR}/)
+
+    # Add HIP targets
+    set_source_files_properties(kernel_launch.cu PROPERTIES HIP_SOURCE_PROPERTY_FORMAT 1)
+    # Link with HIP
+    hip_add_executable(kernel_launch_overhead kernel_launch.cu)
+    # Install tergets
+    install(TARGETS kernel_launch_overhead RUNTIME DESTINATION bin)
+
+else()
+    message(FATAL_ERROR "No CUDA or ROCm environment found.")
+endif()
 
-add_executable(kernel_launch_overhead cuda_kernel_launch.cu)
-set_property(TARGET kernel_launch_overhead PROPERTY CUDA_ARCHITECTURES ${NVCC_ARCHS_SUPPORTED})
-install(TARGETS kernel_launch_overhead RUNTIME DESTINATION bin)

superbench/benchmarks/micro_benchmarks/kernel_launch_overhead/kernel_launch.cu

@@ -2,7 +2,7 @@
 // Licensed under the MIT License.
 
 // Kernel launch benchmark which will launch one empty kernel and record the cost in event mode and wall mode.
-//   event mode: using cuda event to record the elapsed time of kernel launch on device.
+//   event mode: using cuda/hip event to record the elapsed time of kernel launch on device.
 //   wall mode: using host timer to record the elapsed time kernel launch on both host and device.
 
 #include <algorithm>