r312024 - [OPENMP] Capture global variables in all target executable regions.

[Alexey Bataev via cfe-commits]

Author: abataev
Date: Tue Aug 29 12:30:57 2017
New Revision: 312024

URL: http://llvm.org/viewvc/llvm-project?rev=312024&view=rev
Log:
[OPENMP] Capture global variables in all target executable regions.

Capturing of the global variables occurs only in target regions. Patch
fixes it and allows capturing of globals in all target executable
directives.

Modified:
    cfe/trunk/lib/Sema/SemaOpenMP.cpp
    cfe/trunk/test/OpenMP/target_teams_codegen.cpp

Modified: cfe/trunk/lib/Sema/SemaOpenMP.cpp
URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/Sema/SemaOpenMP.cpp?rev=312024&r1=312023&r2=312024&view=diff
==============================================================================
--- cfe/trunk/lib/Sema/SemaOpenMP.cpp (original)
+++ cfe/trunk/lib/Sema/SemaOpenMP.cpp Tue Aug 29 12:30:57 2017
@@ -1278,7 +1278,7 @@ VarDecl *Sema::IsOpenMPCapturedDecl(Valu
   //
   auto *VD = dyn_cast<VarDecl>(D);
   if (VD && !VD->hasLocalStorage()) {
-    if (DSAStack->getCurrentDirective() == OMPD_target &&
+    if (isOpenMPTargetExecutionDirective(DSAStack->getCurrentDirective()) &&
         !DSAStack->isClauseParsingMode())
       return VD;
     if (DSAStack->hasDirective(

Modified: cfe/trunk/test/OpenMP/target_teams_codegen.cpp
URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/test/OpenMP/target_teams_codegen.cpp?rev=312024&r1=312023&r2=312024&view=diff
==============================================================================
--- cfe/trunk/test/OpenMP/target_teams_codegen.cpp (original)
+++ cfe/trunk/test/OpenMP/target_teams_codegen.cpp Tue Aug 29 12:30:57 2017
@@ -83,6 +83,8 @@ struct TT{
   ty Y;
 };
 
+int global;
+
 // CHECK: define {{.*}}[[FOO:@.+]](
 int foo(int n) {
   int a = 0;
@@ -178,6 +180,7 @@ int foo(int n) {
   {
     a += 1;
     aa += 1;
+    global += 1;
   }
 
   // We capture 3 VLA sizes in this target region
@@ -372,6 +375,7 @@ int foo(int n) {
 // CHECK:       [[AA_ADDR:%.+]] = alloca i[[SZ]], align
 // CHECK:       [[A_CASTED:%.+]] = alloca i[[SZ]], align
 // CHECK:       [[AA_CASTED:%.+]] = alloca i[[SZ]], align
+// CHECK:       [[GLOBAL_CASTED:%.+]] = alloca i[[SZ]], align
 // CHECK-DAG:   store i[[SZ]] %{{.+}}, i[[SZ]]* [[A_ADDR]], align
 // CHECK-DAG:   store i[[SZ]] %{{.+}}, i[[SZ]]* [[AA_ADDR]], align
 // CHECK-64-DAG:[[A_CADDR:%.+]] = bitcast i[[SZ]]* [[A_ADDR]] to i32*
@@ -384,18 +388,28 @@ int foo(int n) {
 // CHECK-DAG:   [[AA:%.+]] = load i16, i16* [[AA_CADDR]], align
 // CHECK-DAG:   [[AA_C:%.+]] = bitcast i[[SZ]]* [[AA_CASTED]] to i16*
 // CHECK-DAG:   store i16 [[AA]], i16* [[AA_C]], align
+// CHECK-DAG:   [[GLOBAL:%.+]] = load i32, i32* @global, align
+// CHECK-64-DAG:[[GLOBAL_C:%.+]] = bitcast i[[SZ]]* [[GLOBAL_CASTED]] to i32*
+// CHECK-64-DAG:store i32 [[GLOBAL]], i32* [[GLOBAL_C]], align
+// CHECK-32-DAG:store i32 [[GLOBAL]], i32* [[GLOBAL_CASTED]], align
 // CHECK-DAG:   [[PARAM1:%.+]] = load i[[SZ]], i[[SZ]]* [[A_CASTED]], align
 // CHECK-DAG:   [[PARAM2:%.+]] = load i[[SZ]], i[[SZ]]* [[AA_CASTED]], align
-// CHECK-DAG:   call {{.*}}void (%ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_teams(%ident_t* [[DEF_LOC]], i32 2, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i[[SZ]], i[[SZ]])* [[OMP_OUTLINED3:@.+]] to void (i32*, i32*, ...)*), i[[SZ]] [[PARAM1]], i[[SZ]] [[PARAM2]])
+// CHECK-DAG:   [[PARAM3:%.+]] = load i[[SZ]], i[[SZ]]* [[GLOBAL_CASTED]], align
+// CHECK-DAG:   call {{.*}}void (%ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_teams(%ident_t* [[DEF_LOC]], i32 3, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i[[SZ]], i[[SZ]], i[[SZ]])* [[OMP_OUTLINED3:@.+]] to void (i32*, i32*, ...)*), i[[SZ]] [[PARAM1]], i[[SZ]] [[PARAM2]], i[[SZ]] [[PARAM3]])
 //
 //
-// CHECK:       define internal {{.*}}void [[OMP_OUTLINED3]](i32* noalias %.global_tid., i32* noalias %.bound_tid., i[[SZ]] %{{.+}}, i[[SZ]] %{{.+}})
+// CHECK:       define internal {{.*}}void [[OMP_OUTLINED3]](i32* noalias %.global_tid., i32* noalias %.bound_tid., i[[SZ]] %{{.+}}, i[[SZ]] %{{.+}}, i[[SZ]] %{{.+}})
 // CHECK:       [[A_ADDR:%.+]] = alloca i[[SZ]], align
 // CHECK:       [[AA_ADDR:%.+]] = alloca i[[SZ]], align
+// CHECK:       [[GLOBAL_ADDR:%.+]] = alloca i[[SZ]], align
 // CHECK-DAG:   store i[[SZ]] %{{.+}}, i[[SZ]]* [[A_ADDR]], align
 // CHECK-DAG:   store i[[SZ]] %{{.+}}, i[[SZ]]* [[AA_ADDR]], align
+// CHECK-DAG:   store i[[SZ]] %{{.+}}, i[[SZ]]* [[GLOBAL_ADDR]], align
 // CHECK-64-DAG:[[A_CADDR:%.+]] = bitcast i[[SZ]]* [[A_ADDR]] to i32*
 // CHECK-DAG:   [[AA_CADDR:%.+]] = bitcast i[[SZ]]* [[AA_ADDR]] to i16*
+// CHECK-64-DAG:[[GLOBAL_CADDR:%.+]] = bitcast i[[SZ]]* [[GLOBAL_ADDR]] to i32*
+// CHECK-64:    store i32 {{.+}}, i32* [[GLOBAL_CADDR]],
+// CHECK-32:    store i32 {{.+}}, i32* [[GLOBAL_ADDR]],
 // CHECK:       ret void
 // CHECK-NEXT:  }