[clang][HLSL] Add GroupMemoryBarrierWithGroupSync intrinsic (#111883)

partially fixes #70103 

### Changes
* Implemented `GroupMemoryBarrierWithGroupSync` clang builtin
* Linked `GroupMemoryBarrierWithGroupSync` clang builtin with
`hlsl_intrinsics.h`
* Added sema checks for `GroupMemoryBarrierWithGroupSync` to
`CheckHLSLBuiltinFunctionCall` in
`SemaChecking.cpp`
* Add codegen for `GroupMemoryBarrierWithGroupSync` to
`EmitHLSLBuiltinExpr` in `CGBuiltin.cpp`
* Add codegen tests to
`clang/test/CodeGenHLSL/builtins/GroupMemoryBarrierWithGroupSync.hlsl`
* Add sema tests to
`clang/test/SemaHLSL/BuiltIns/GroupMemoryBarrierWithGroupSync-errors.hlsl`

### Related PRs
* [[DXIL] Add GroupMemoryBarrierWithGroupSync intrinsic
#111884](https://github.com/llvm/llvm-project/pull/111884)
* [[SPIRV] Add GroupMemoryBarrierWithGroupSync intrinsic
#111888](https://github.com/llvm/llvm-project/pull/111888)

clang/include/clang/Basic/Builtins.td

@@ -4930,6 +4930,12 @@ def HLSLClip: LangBuiltin<"HLSL_LANG"> {
   let Prototype = "void(...)";
 }
 
+def HLSLGroupMemoryBarrierWithGroupSync: LangBuiltin<"HLSL_LANG"> {
+  let Spellings = ["__builtin_hlsl_group_memory_barrier_with_group_sync"];
+  let Attributes = [NoThrow, Const];
+  let Prototype = "void()";
+}
+
 // Builtins for XRay.
 def XRayCustomEvent : Builtin {
   let Spellings = ["__xray_customevent"];

clang/lib/CodeGen/CGBuiltin.cpp

@@ -19456,6 +19456,12 @@ case Builtin::BI__builtin_hlsl_elementwise_isinf: {
     assert(E->getArg(0)->getType()->hasFloatingRepresentation() &&
            "clip operands types mismatch");
     return handleHlslClip(E, this);
+  case Builtin::BI__builtin_hlsl_group_memory_barrier_with_group_sync: {
+    Intrinsic::ID ID =
+        CGM.getHLSLRuntime().getGroupMemoryBarrierWithGroupSyncIntrinsic();
+    return EmitRuntimeCall(
+        Intrinsic::getOrInsertDeclaration(&CGM.getModule(), ID));
+  }
   }
   return nullptr;
 }

clang/lib/CodeGen/CGHLSLRuntime.h

@@ -103,6 +103,8 @@ public:
 
   GENERATE_HLSL_INTRINSIC_FUNCTION(CreateHandleFromBinding, handle_fromBinding)
   GENERATE_HLSL_INTRINSIC_FUNCTION(BufferUpdateCounter, bufferUpdateCounter)
+  GENERATE_HLSL_INTRINSIC_FUNCTION(GroupMemoryBarrierWithGroupSync,
+                                   group_memory_barrier_with_group_sync)
 
   //===----------------------------------------------------------------------===//
   // End of reserved area for HLSL intrinsic getters.

clang/lib/Headers/hlsl/hlsl_intrinsics.h

@@ -2481,5 +2481,17 @@ float3 radians(float3);
 _HLSL_BUILTIN_ALIAS(__builtin_hlsl_elementwise_radians)
 float4 radians(float4);
 
+//===----------------------------------------------------------------------===//
+// GroupMemoryBarrierWithGroupSync builtins
+//===----------------------------------------------------------------------===//
+
+/// \fn void GroupMemoryBarrierWithGroupSync(void)
+/// \brief Blocks execution of all threads in a group until all group shared
+/// accesses have been completed and all threads in the group have reached this
+/// call.
+
+_HLSL_BUILTIN_ALIAS(__builtin_hlsl_group_memory_barrier_with_group_sync)
+void GroupMemoryBarrierWithGroupSync(void);
+
 } // namespace hlsl
 #endif //_HLSL_HLSL_INTRINSICS_H_

clang/test/CodeGenHLSL/builtins/GroupMemoryBarrierWithGroupSync.hlsl

@@ -0,0 +1,20 @@
+// RUN: %clang_cc1 -finclude-default-header -x hlsl -triple \
+// RUN:   dxil-pc-shadermodel6.3-library %s \
+// RUN:   -emit-llvm -disable-llvm-passes -o - | FileCheck %s \
+// RUN:   -DTARGET=dx -DFNATTRS=noundef -check-prefixes=CHECK,CHECK-DXIL
+// RUN: %clang_cc1 -finclude-default-header -x hlsl -triple \
+// RUN:   spirv-unknown-vulkan-compute %s \
+// RUN:   -emit-llvm -disable-llvm-passes -o - | FileCheck %s \
+// RUN:   -DTARGET=spv -DFNATTRS="spir_func noundef" -check-prefixes=CHECK,CHECK-SPIRV
+
+// CHECK-DXIL: define void @
+// CHECK-SPIRV: define spir_func void @
+void test_GroupMemoryBarrierWithGroupSync() {
+// CHECK-DXIL: call void @llvm.[[TARGET]].group.memory.barrier.with.group.sync()
+// CHECK-SPIRV: call spir_func void @llvm.[[TARGET]].group.memory.barrier.with.group.sync()
+  GroupMemoryBarrierWithGroupSync();
+}
+
+// CHECK: declare void @llvm.[[TARGET]].group.memory.barrier.with.group.sync() #[[ATTRS:[0-9]+]]
+// CHECK-NOT: attributes #[[ATTRS]] = {{.+}}memory(none){{.+}}
+// CHECK: attributes #[[ATTRS]] = {

clang/test/SemaHLSL/BuiltIns/GroupMemoryBarrierWithGroupSync-errors.hlsl

@@ -0,0 +1,6 @@
+// RUN: %clang_cc1 -finclude-default-header -triple dxil-pc-shadermodel6.6-library %s -emit-llvm-only -disable-llvm-passes -verify
+
+void test_too_many_arg() {
+  __builtin_hlsl_group_memory_barrier_with_group_sync(0);
+  // expected-error@-1 {{too many arguments to function call, expected 0, have 1}}
+}