[Libomptarget] Reduce shared memory stack size to 512 and a message when it is exceeded

[jhuber6]

Summary

Reduces the shared memory size used for globalization to 512 bytes from
2048 to reduce the pressure on shared memory. This patch ado adds a
debug mesage to indicate when the shared memory was insufficient.

[jdoerfert]

typo in the commit message.

openmp/libomptarget/DeviceRTL/src/State.cpp

140

Be more explicit here, or, probably better, use a "runtime remark" [OMPR1000] Shared memory stack full, fallback to dynamic allocation of global memory (=slow)

[jhuber6]

Being more specific.

[jhuber6]

fix missing word

[jdoerfert]

LG

openmp/libomptarget/DeviceRTL/CMakeLists.txt

set(src_files
   ${source_directory}/Reduction.cpp
   ${source_directory}/State.cpp
   ${source_directory}/Synchronization.cpp
   ${source_directory}/Tasking.cpp
   ${source_directory}/Utils.cpp
   ${source_directory}/Workshare.cpp
 )
 
-set(clang_opt_flags -O1 -mllvm -openmp-opt-disable -DSHARED_SCRATCHPAD_SIZE=2048)
+set(clang_opt_flags -O1 -mllvm -openmp-opt-disable -DSHARED_SCRATCHPAD_SIZE=512)
 set(link_opt_flags  -O1        -openmp-opt-disable)
 
 # Prepend -I to each list element
 set (LIBOMPTARGET_LLVM_INCLUDE_DIRS_DEVICERTL "${LIBOMPTARGET_LLVM_INCLUDE_DIRS}")
 list(TRANSFORM LIBOMPTARGET_LLVM_INCLUDE_DIRS_DEVICERTL PREPEND "-I")
 
 # Set flags for LLVM Bitcode compilation.
 set(bc_flags -S -x c++ -std=c++17 -fvisibility=hidden

openmp/libomptarget/DeviceRTL/src/State.cpp

void *SharedMemorySmartStackTy::push(uint64_t Bytes) {
 
   int TId = mapping::getThreadIdInBlock();
   if (Usage[TId] + AlignedBytes <= StorageTotal) {
     void *Ptr = getThreadDataTop(TId);
     Usage[TId] += AlignedBytes;
     return Ptr;
   }
 
+  if (config::isDebugMode(config::DebugKind::CommonIssues))
+    PRINT("Shared memory stack full, fallback to dynamic allocation of global "
+          "memory will negatively impact performance.");
   void *GlobalMemory = memory::allocGlobal(
       AlignedBytes, "Slow path shared memory allocation, insufficient "
                     "shared memory stack memory!");
   ASSERT(GlobalMemory != nullptr && "nullptr returned by malloc!");
 
   return GlobalMemory;
 }