[PR41010][OpenCL] Allow OpenCL C style vector initialization in C++

[Anastasia]

Summary

Not sure it's the right approach but the idea is that we would like to accept vector contraction syntax from OpenCL C.

This commit fixes Sema checks when creating a vector using compound vector syntax. We should pass the type of initializing expression consistently, to avoid mismatch with the initializer.

This change also improves testing for the vector creation.

[rjmccall]

lib/Sema/SemaInit.cpp

1295

There should at least be a comment here explaining this, but mostly it's been so long since I've really understood this code that I'd like Richard to look at it.

[Anastasia]

@rsmith, do you have any comments here? It would be quite useful to have this patch in Clang9.

[Anastasia]

ping!

[Anastasia]

Created a follow up review in https://reviews.llvm.org/D65286

lib/Sema/SemaInit.cpp

void InitListChecker::CheckSubElementType(const InitializedEntity &Entity,
   if (SemaRef.getLangOpts().CPlusPlus) {
     // C++ [dcl.init.aggr]p2:
     //   Each member is copy-initialized from the corresponding
     //   initializer-clause.
 
     // FIXME: Better EqualLoc?
     InitializationKind Kind =
         InitializationKind::CreateCopy(expr->getBeginLoc(), SourceLocation());
-    InitializationSequence Seq(SemaRef, Entity, Kind, expr,
+    auto TmpEntity =
+        (ElemType->isExtVectorType() && !Entity.getType()->isExtVectorType())
+            ? InitializedEntity::InitializeTemporary(ElemType)
+            : Entity;
+
+    InitializationSequence Seq(SemaRef, TmpEntity, Kind, expr,
                                /*TopLevelOfInitList*/ true);
 
     // C++14 [dcl.init.aggr]p13:
     //   If the assignment-expression can initialize a member, the member is
     //   initialized. Otherwise [...] brace elision is assumed
     //
     // Brace elision is never performed if the element is not an
     // assignment-expression.
     if (Seq || isa<InitListExpr>(expr)) {
       if (!VerifyOnly) {
-        ExprResult Result =
-          Seq.Perform(SemaRef, Entity, Kind, expr);
+        ExprResult Result = Seq.Perform(SemaRef, TmpEntity, Kind, expr);
         if (Result.isInvalid())
           hadError = true;
 
         UpdateStructuredListElement(StructuredList, StructuredIndex,
                                     Result.getAs<Expr>());
       } else if (!Seq)
         hadError = true;
       ++Index;

test/CodeGenOpenCL/vector_literals_valid.cl

-// RUN: %clang_cc1 -emit-llvm %s -o %t
+// RUN: %clang_cc1 -emit-llvm %s -o - -O0 | FileCheck %s
+// RUN: %clang_cc1 -emit-llvm %s -o - -cl-std=c++ -O0 | FileCheck %s
 
-typedef __attribute__(( ext_vector_type(2) ))  int int2;
-typedef __attribute__(( ext_vector_type(3) ))  int int3;
+typedef __attribute__((ext_vector_type(2))) int int2;
+typedef __attribute__((ext_vector_type(3))) int int3;
 typedef __attribute__(( ext_vector_type(4) ))  int int4;
 typedef __attribute__(( ext_vector_type(8) ))  int int8;
-typedef __attribute__(( ext_vector_type(4) ))  float float4;
+typedef __attribute__((ext_vector_type(4))) float float4;
 
 void vector_literals_valid() {
-  int4 a_1_1_1_1 = (int4)(1,2,3,4);
-  int4 a_2_1_1 = (int4)((int2)(1,2),3,4);
-  int4 a_1_2_1 = (int4)(1,(int2)(2,3),4);
-  int4 a_1_1_2 = (int4)(1,2,(int2)(3,4));
-  int4 a_2_2 = (int4)((int2)(1,2),(int2)(3,4));
-  int4 a_3_1 = (int4)((int3)(1,2,3),4);
-  int4 a_1_3 = (int4)(1,(int3)(2,3,4));
+  //CHECK: store <4 x i32> <i32 1, i32 2, i32 3, i32 4>, <4 x i32>*
+
+  int4 a_1_1_1_1 = (int4)(1, 2, 3, 4);
+
+  //CHECK: store <2 x i32> <i32 1, i32 2>, <2 x i32>*
+  //CHECK: shufflevector <2 x i32> %{{[0-9]+}}, <2 x i32> undef, <4 x i32> <i32 0, i32 1, i32 undef, i32 undef>
+  //CHECK: shufflevector <4 x i32> %{{.+}}, <4 x i32> undef, <4 x i32> <i32 0, i32 1, i32 undef, i32 undef>
+  //CHECK: insertelement <4 x i32> %{{.+}}, i32 3, i32 2
+  //CHECK: insertelement <4 x i32> %{{.+}}, i32 4, i32 3
+  int4 a_2_1_1 = (int4)((int2)(1, 2), 3, 4);
+
+  //CHECK: store <2 x i32> <i32 2, i32 3>, <2 x i32>*
+  //CHECK: shufflevector <2 x i32> %{{[0-9]+}}, <2 x i32> undef, <4 x i32> <i32 0, i32 1, i32 undef, i32 undef>
+  //CHECK: shufflevector <4 x i32> <i32 1, i32 undef, i32 undef, i32 undef>, <4 x i32> %{{.+}}, <4 x i32> <i32 0, i32 4, i32 5, i32 undef>
+  //CHECK: insertelement <4 x i32> %{{.+}}, i32 4, i32 3
+  int4 a_1_2_1 = (int4)(1, (int2)(2, 3), 4);
+
+  //CHECK: store <2 x i32> <i32 3, i32 4>, <2 x i32>*
+  //CHECK: shufflevector <2 x i32> %{{[0-9]+}}, <2 x i32> undef, <4 x i32> <i32 0, i32 1, i32 undef, i32 undef>
+  //CHECK: shufflevector <4 x i32> <i32 1, i32 2, i32 undef, i32 undef>, <4 x i32> %{{.+}}, <4 x i32> <i32 0, i32 1, i32 4, i32 5>
+  int4 a_1_1_2 = (int4)(1, 2, (int2)(3, 4));
+
+  //CHECK: store <2 x i32> <i32 1, i32 2>, <2 x i32>*
+  //CHECK: shufflevector <2 x i32> %{{[0-9]+}}, <2 x i32> undef, <4 x i32> <i32 0, i32 1, i32 undef, i32 undef>
+  //CHECK: shufflevector <4 x i32> %{{.+}}, <4 x i32> undef, <4 x i32> <i32 0, i32 1, i32 undef, i32 undef>
+  //CHECK: store <2 x i32> <i32 3, i32 4>, <2 x i32>*
+  //CHECK: shufflevector <2 x i32> %{{[0-9]+}}, <2 x i32> undef, <4 x i32> <i32 0, i32 1, i32 undef, i32 undef>
+  //CHECK: shufflevector <4 x i32> %{{.+}}, <4 x i32> %{{.+}}, <4 x i32> <i32 0, i32 1, i32 4, i32 5>
+  int4 a_2_2 = (int4)((int2)(1, 2), (int2)(3, 4));
+
+  //CHECK: store <4 x i32> <i32 2, i32 3, i32 4, i32 undef>, <4 x i32>*
+  //CHECK: shufflevector <4 x i32> %{{.+}}, <4 x i32> undef, <3 x i32> <i32 0, i32 1, i32 2>
+  //CHECK: shufflevector <3 x i32> %{{.+}}, <3 x i32> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 undef>
+  //CHECK: shufflevector <4 x i32> <i32 1, i32 undef, i32 undef, i32 undef>, <4 x i32> %{{.+}}, <4 x i32> <i32 0, i32 4, i32 5, i32 6>
+  int4 a_1_3 = (int4)(1, (int3)(2, 3, 4));
+
+  //CHECK: store <4 x i32> <i32 1, i32 1, i32 1, i32 1>, <4 x i32>* %a
   int4 a = (int4)(1);
-  int8 b = (int8)(1,2,a.xy,a);
-  float4 V2 = (float4) (1);
-}
 
+  //CHECK: load <4 x i32>, <4 x i32>* %a, align 16
+  //CHECK: shufflevector <4 x i32> %{{[0-9]+}}, <4 x i32> undef, <2 x i32> <i32 0, i32 1>
+  //CHECK: shufflevector <2 x i32> %{{[0-9]+}}, <2 x i32> undef, <8 x i32> <i32 0, i32 1, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
+  //CHECK: shufflevector <8 x i32> <i32 1, i32 2, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>, <8 x i32> %{{.+}}, <8 x i32> <i32 0, i32 1, i32 8, i32 9, i32 undef, i32 undef, i32 undef, i32 undef>
+  //CHECK: load <4 x i32>, <4 x i32>* %a, align 16
+  //CHECK: shufflevector <4 x i32> %{{[0-9]+}}, <4 x i32> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef>
+  //CHECK: shufflevector <8 x i32> %{{.+}}, <8 x i32> %{{.+}}, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 8, i32 9, i32 10, i32 11>
+  int8 b = (int8)(1, 2, a.xy, a);
 
+  //CHECK: store <4 x float> <float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00>, <4 x float>* %V2
+  float4 V2 = (float4)(1);
+}