[libomptarget] Remove duplicate RTLRequiresFlags per device

We have one global RTLs.RequiresFlags, I don't see a need to make a
copy per device that the runtime manages. This was problematic anyway
because the copy happened during the first __tgt_register_lib(). This
made it impossible to call __tgt_register_requires() from normal user
funtions for testing.
Hence, this change also fixes unified_shared_memory/shared_update.c for
older versions of Clang that don't call __tgt_register_requires() before
__tgt_register_lib().

Differential Revision: https://reviews.llvm.org/D66019

git-svn-id: https://llvm.org/svn/llvm-project/openmp/trunk@368465 91177308-0d34-0410-b5e6-96231b3b80d8

Author: hahnjo

libomptarget/src/api.cpp

@@ -120,7 +120,7 @@ EXTERN int omp_target_is_present(void *ptr, int device_num) {
   // getTgtPtrBegin() function which means that there is no device
   // corresponding point for ptr. This function should return false
   // in that situation.
-  if (Device.RTLRequiresFlags & OMP_REQ_UNIFIED_SHARED_MEMORY)
+  if (RTLs.RequiresFlags & OMP_REQ_UNIFIED_SHARED_MEMORY)
     rc = !IsHostPtr;
   DP("Call to omp_target_is_present returns %d\n", rc);
   return rc;

libomptarget/src/device.cpp

@@ -194,7 +194,7 @@ void *DeviceTy::getOrAllocTgtPtr(void *HstPtrBegin, void *HstPtrBase,
     // maps are respected.
     // TODO: In addition to the mapping rules above, when the close map
     // modifier is implemented, foce the mapping of the variable to the device.
-    if (RTLRequiresFlags & OMP_REQ_UNIFIED_SHARED_MEMORY) {
+    if (RTLs.RequiresFlags & OMP_REQ_UNIFIED_SHARED_MEMORY) {
       DP("Return HstPtrBegin " DPxMOD " Size=%ld RefCount=%s\n",
          DPxPTR((uintptr_t)HstPtrBegin), Size, (UpdateRefCount ? " updated" : ""));
       IsHostPtr = true;
@@ -241,7 +241,7 @@ void *DeviceTy::getTgtPtrBegin(void *HstPtrBegin, int64_t Size, bool &IsLast,
         (CONSIDERED_INF(HT.RefCount)) ? "INF" :
             std::to_string(HT.RefCount).c_str());
     rc = (void *)tp;
-  } else if (RTLRequiresFlags & OMP_REQ_UNIFIED_SHARED_MEMORY) {
+  } else if (RTLs.RequiresFlags & OMP_REQ_UNIFIED_SHARED_MEMORY) {
     // If the value isn't found in the mapping and unified shared memory
     // is on then it means we have stumbled upon a value which we need to
     // use directly from the host.
@@ -270,7 +270,7 @@ void *DeviceTy::getTgtPtrBegin(void *HstPtrBegin, int64_t Size) {
 }
 
 int DeviceTy::deallocTgtPtr(void *HstPtrBegin, int64_t Size, bool ForceDelete) {
-  if (RTLRequiresFlags & OMP_REQ_UNIFIED_SHARED_MEMORY)
+  if (RTLs.RequiresFlags & OMP_REQ_UNIFIED_SHARED_MEMORY)
     return OFFLOAD_SUCCESS;
   // Check if the pointer is contained in any sub-nodes.
   int rc;
@@ -305,7 +305,7 @@ int DeviceTy::deallocTgtPtr(void *HstPtrBegin, int64_t Size, bool ForceDelete) {
 void DeviceTy::init() {
   // Make call to init_requires if it exists for this plugin.
   if (RTL->init_requires)
-    RTL->init_requires(RTLRequiresFlags);
+    RTL->init_requires(RTLs.RequiresFlags);
   int32_t rc = RTL->init_device(RTLDeviceID);
   if (rc == OFFLOAD_SUCCESS) {
     IsInit = true;

libomptarget/src/device.h

@@ -100,13 +100,10 @@ struct DeviceTy {
   // moved into the target task in libomp.
   std::map<int32_t, uint64_t> LoopTripCnt;
 
-  int64_t RTLRequiresFlags;
-
   DeviceTy(RTLInfoTy *RTL)
       : DeviceID(-1), RTL(RTL), RTLDeviceID(-1), IsInit(false), InitFlag(),
-        HasPendingGlobals(false), HostDataToTargetMap(),
-        PendingCtorsDtors(), ShadowPtrMap(), DataMapMtx(), PendingGlobalsMtx(),
-        ShadowMtx(), RTLRequiresFlags(0) {}
+        HasPendingGlobals(false), HostDataToTargetMap(), PendingCtorsDtors(),
+        ShadowPtrMap(), DataMapMtx(), PendingGlobalsMtx(), ShadowMtx() {}
 
   // The existence of mutexes makes DeviceTy non-copyable. We need to
   // provide a copy constructor and an assignment operator explicitly.
@@ -115,9 +112,8 @@ struct DeviceTy {
         IsInit(d.IsInit), InitFlag(), HasPendingGlobals(d.HasPendingGlobals),
         HostDataToTargetMap(d.HostDataToTargetMap),
         PendingCtorsDtors(d.PendingCtorsDtors), ShadowPtrMap(d.ShadowPtrMap),
-        DataMapMtx(), PendingGlobalsMtx(),
-        ShadowMtx(), LoopTripCnt(d.LoopTripCnt),
-        RTLRequiresFlags(d.RTLRequiresFlags) {}
+        DataMapMtx(), PendingGlobalsMtx(), ShadowMtx(),
+        LoopTripCnt(d.LoopTripCnt) {}
 
   DeviceTy& operator=(const DeviceTy &d) {
     DeviceID = d.DeviceID;
@@ -129,7 +125,6 @@ struct DeviceTy {
     PendingCtorsDtors = d.PendingCtorsDtors;
     ShadowPtrMap = d.ShadowPtrMap;
     LoopTripCnt = d.LoopTripCnt;
-    RTLRequiresFlags = d.RTLRequiresFlags;
 
     return *this;
   }

libomptarget/src/omptarget.cpp

@@ -290,7 +290,7 @@ int target_data_begin(DeviceTy &Device, int32_t arg_num,
 
     if (arg_types[i] & OMP_TGT_MAPTYPE_TO) {
       bool copy = false;
-      if (!(Device.RTLRequiresFlags & OMP_REQ_UNIFIED_SHARED_MEMORY)) {
+      if (!(RTLs.RequiresFlags & OMP_REQ_UNIFIED_SHARED_MEMORY)) {
         if (IsNew || (arg_types[i] & OMP_TGT_MAPTYPE_ALWAYS)) {
           copy = true;
         } else if (arg_types[i] & OMP_TGT_MAPTYPE_MEMBER_OF) {
@@ -390,7 +390,7 @@ int target_data_end(DeviceTy &Device, int32_t arg_num, void **args_base,
       if (arg_types[i] & OMP_TGT_MAPTYPE_FROM) {
         bool Always = arg_types[i] & OMP_TGT_MAPTYPE_ALWAYS;
         bool CopyMember = false;
-        if (!(Device.RTLRequiresFlags & OMP_REQ_UNIFIED_SHARED_MEMORY)) {
+        if (!(RTLs.RequiresFlags & OMP_REQ_UNIFIED_SHARED_MEMORY)) {
           if ((arg_types[i] & OMP_TGT_MAPTYPE_MEMBER_OF) &&
               !(arg_types[i] & OMP_TGT_MAPTYPE_PTR_AND_OBJ)) {
             // Copy data only if the "parent" struct has RefCount==1.
@@ -404,7 +404,7 @@ int target_data_end(DeviceTy &Device, int32_t arg_num, void **args_base,
         }
 
         if ((DelEntry || Always || CopyMember) &&
-            !(Device.RTLRequiresFlags & OMP_REQ_UNIFIED_SHARED_MEMORY &&
+            !(RTLs.RequiresFlags & OMP_REQ_UNIFIED_SHARED_MEMORY &&
               TgtPtrBegin == HstPtrBegin)) {
           DP("Moving %" PRId64 " bytes (tgt:" DPxMOD ") -> (hst:" DPxMOD ")\n",
               data_size, DPxPTR(TgtPtrBegin), DPxPTR(HstPtrBegin));
@@ -486,7 +486,7 @@ int target_data_update(DeviceTy &Device, int32_t arg_num,
       continue;
     }
 
-    if (Device.RTLRequiresFlags & OMP_REQ_UNIFIED_SHARED_MEMORY &&
+    if (RTLs.RequiresFlags & OMP_REQ_UNIFIED_SHARED_MEMORY &&
         TgtPtrBegin == HstPtrBegin) {
       DP("hst data:" DPxMOD " unified and shared, becomes a noop\n",
          DPxPTR(HstPtrBegin));
@@ -669,7 +669,7 @@ int target(int64_t device_id, void *host_ptr, int32_t arg_num,
              DPxPTR(HstPtrVal));
           continue;
         }
-        if (Device.RTLRequiresFlags & OMP_REQ_UNIFIED_SHARED_MEMORY &&
+        if (RTLs.RequiresFlags & OMP_REQ_UNIFIED_SHARED_MEMORY &&
             TgtPtrBegin == HstPtrBegin) {
           DP("Unified memory is active, no need to map lambda captured"
              "variable (" DPxMOD ")\n", DPxPTR(HstPtrVal));

libomptarget/src/rtl.cpp

@@ -266,8 +266,6 @@ void RTLsTy::RegisterLib(__tgt_bin_desc *desc) {
           Devices[start + device_id].DeviceID = start + device_id;
           // RTL local device ID
           Devices[start + device_id].RTLDeviceID = device_id;
-          // RTL requires flags
-          Devices[start + device_id].RTLRequiresFlags = RequiresFlags;
         }
 
         // Initialize the index of this RTL and save it in the used RTLs.