CMake: adding flags for profiling and debugging (shouldn't affect performance)

2026-04-30 14:29:51 +02:00
parent 2fc5cf7573
commit bd92c1ff8f
2 changed files with 4 additions and 3 deletions
@@ -8,8 +8,8 @@ SET(CMAKE_POLICY_DEFAULT_CMP0077 NEW)
 SET(CMAKE_CXX_STANDARD 20)
 SET(CMAKE_CXX_STANDARD_REQUIRED True)

-SET(CMAKE_CXX_FLAGS_RELEASE "-O3 -Wno-deprecated-enum-enum-conversion")
-SET(CMAKE_C_FLAGS_RELEASE "-O3")
+SET(CMAKE_CXX_FLAGS_RELEASE "-O3 -Wno-deprecated-enum-enum-conversion -g -fno-omit-frame-pointer")
+SET(CMAKE_C_FLAGS_RELEASE "-O3 -g -fno-omit-frame-pointer")

 SET(JFJOCH_WRITER_ONLY           OFF CACHE BOOL "Compile HDF5 writer only")
 SET(JFJOCH_INSTALL_DRIVER_SOURCE OFF CACHE BOOL "Install kernel driver source (ignored if building writer only; necessary for RPM building)")
@@ -41,7 +41,7 @@ CHECK_LANGUAGE(CUDA)

 SET(CMAKE_CUDA_ARCHITECTURES 75 80 86 89 90 100 120) # T4, A100, RTX A4000, L4
 SET(CMAKE_CUDA_STANDARD 17)
-SET(CMAKE_CUDA_FLAGS_RELEASE "-O3")
+SET(CMAKE_CUDA_FLAGS_RELEASE "-O3 -lineinfo")
 SET(CMAKE_CUDA_RUNTIME_LIBRARY Static)

 SET(JFJOCH_CUDA_AVAILABLE OFF)
@@ -3,6 +3,7 @@
 ### 1.0.0-rc.142
 This is an UNSTABLE release. The release has significant modifications and bug fixes, if things go wrong, it is better to revert to 1.0.0-rc.132.

+* Add flags for debugging and profiling, while keeping optimizations to allow for profiling with VTune
 * Support for newer CUDA architectures (notably Blackwell); minimum CUDA version 12.8
 * Minor changes to jfjoch_process, jfjoch_fpga_test and jfjoch_lite_perf_test to make them more consistent