Rebase fix features

Author: fel-cab

openmp/libomptarget/include/Shared/Profile.h

@@ -97,4 +97,16 @@ class Profiler {
   std::string RTM = RegionTypeMsg;                                             \
   llvm::TimeTraceScope TimeScope(__FUNCTION__, ProfileLocation + RTM)
 
+/// Time spend in the current scope, assigned to the regionType
+/// with details from runtime
+#define TIMESCOPE_WITH_DETAILS_AND_IDENT(RegionTypeMsg, Details, IDENT)        \
+  SourceInfo SI(IDENT);                                                        \
+  std::string ProfileLocation = SI.getProfileLocation();                       \
+  llvm::TimeTraceScope TimeScope(RegionTypeMsg, ProfileLocation + Details)
+
+/// Time spend in the current scope, assigned to the function name and source
+/// with details
+#define TIMESCOPE_WITH_DETAILS(Details)                                        \
+  llvm::TimeTraceScope TimeScope(__FUNCTION__, Details)
+
 #endif // OMPTARGET_SHARED_PROFILE_H

openmp/libomptarget/src/interface.cpp

@@ -65,7 +65,7 @@ targetData(ident_t *Loc, int64_t DeviceId, int32_t ArgNum, void **ArgsBase,
   static_assert(std::is_convertible_v<TargetAsyncInfoTy, AsyncInfoTy>,
                 "TargetAsyncInfoTy must be convertible to AsyncInfoTy.");
 
-  TIMESCOPE_WITH_DETAILS_AND_IDENT("Runtime Data Copy:",
+  TIMESCOPE_WITH_DETAILS_AND_IDENT("Runtime: Data Copy",
                                    "NumArgs=" + std::to_string(ArgNum), Loc);
 
   DP("Entering data %s region for device %" PRId64 " with %d mappings\n",
@@ -263,7 +263,7 @@ static inline int targetKernel(ident_t *Loc, int64_t DeviceId, int32_t NumTeams,
          !KernelArgs->ThreadLimit[1] && !KernelArgs->ThreadLimit[2] &&
          "OpenMP interface should not use multiple dimensions");
   TIMESCOPE_WITH_DETAILS_AND_IDENT(
-      "Runtime target exe:",
+      "Runtime: target exe",
       "NumTeams=" + std::to_string(NumTeams) +
           ";NumArgs=" + std::to_string(KernelArgs->NumArgs),
       Loc);
@@ -297,9 +297,10 @@ static inline int targetKernel(ident_t *Loc, int64_t DeviceId, int32_t NumTeams,
 
   int Rc = OFFLOAD_SUCCESS;
   Rc = target(Loc, *DeviceOrErr, HostPtr, *KernelArgs, AsyncInfo);
-
-  if (Rc == OFFLOAD_SUCCESS)
-    Rc = AsyncInfo.synchronize();
+  { // required to show syncronization
+    TIMESCOPE_WITH_DETAILS_AND_IDENT("Runtime: syncronize", "", Loc);
+    if (Rc == OFFLOAD_SUCCESS)
+      Rc = AsyncInfo.synchronize();
 
     handleTargetOutcome(Rc == OFFLOAD_SUCCESS, Loc);
     assert(Rc == OFFLOAD_SUCCESS && "__tgt_target_kernel unexpected failure!");

openmp/libomptarget/src/omptarget.cpp

@@ -534,7 +534,7 @@ int targetDataBegin(ident_t *Loc, DeviceTy &Device, int32_t ArgNum,
         (ArgTypes[I] & OMP_TGT_MAPTYPE_PRIVATE))
       continue;
     TIMESCOPE_WITH_DETAILS_AND_IDENT(
-        "HostToDev:", "Size=" + std::to_string(ArgSizes[I]) + "B", Loc);
+        "HostToDev", "Size=" + std::to_string(ArgSizes[I]) + "B", Loc);
     if (ArgMappers && ArgMappers[I]) {
       // Instead of executing the regular path of targetDataBegin, call the
       // targetDataMapper variant which will call targetDataBegin again
@@ -910,7 +910,7 @@ int targetDataEnd(ident_t *Loc, DeviceTy &Device, int32_t ArgNum,
       DP("Moving %" PRId64 " bytes (tgt:" DPxMOD ") -> (hst:" DPxMOD ")\n",
          DataSize, DPxPTR(TgtPtrBegin), DPxPTR(HstPtrBegin));
       TIMESCOPE_WITH_DETAILS_AND_IDENT(
-          "DevToHost:", "Size=" + std::to_string(DataSize) + "B", Loc);
+          "DevToHost", "Size=" + std::to_string(DataSize) + "B", Loc);
       // Wait for any previous transfer if an event is present.
       if (void *Event = TPR.getEntry()->getEvent()) {
         if (Device.waitEvent(Event, AsyncInfo) != OFFLOAD_SUCCESS) {
@@ -1636,7 +1636,7 @@ int target(ident_t *Loc, DeviceTy &Device, void *HostPtr,
   {
     assert(KernelArgs.NumArgs == TgtArgs.size() && "Argument count mismatch!");
     TIMESCOPE_WITH_DETAILS_AND_IDENT(
-        "Kernel Target:",
+        "Kernel Target",
         "NumArguments=" + std::to_string(KernelArgs.NumArgs) +
             ";NumTeams=" + std::to_string(KernelArgs.NumTeams[0]) +
             ";TripCount=" + std::to_string(KernelArgs.Tripcount),