[AMDGPU] Change enqueue kernel handle type

[yaxunl]

Summary

Currently the handle type is a global pointer which holds 8 bytes.
We need a larger type which hold 16 bytes, therefore change it
to [i64 x 2].

[b-sumner]

Looks good to me.

lib/Target/AMDGPU/AMDGPUOpenCLEnqueuedBlockLowering.cpp

 // enqueue related hidden kernel arguments.
 //
 //===----------------------------------------------------------------------===//
 
 #include "AMDGPU.h"
 #include "llvm/ADT/DenseSet.h"
 #include "llvm/ADT/StringRef.h"
 #include "llvm/IR/Constants.h"
+#include "llvm/IR/DerivedTypes.h"
 #include "llvm/IR/Instructions.h"
 #include "llvm/IR/Mangler.h"
 #include "llvm/IR/Module.h"
 #include "llvm/IR/User.h"
 #include "llvm/Pass.h"
 #include "llvm/Support/Debug.h"
 #include "llvm/Support/raw_ostream.h"
 
if (F.hasFnAttribute("enqueued-block")) {
       if (!F.hasName()) {
         SmallString<64> Name;
         Mangler::getNameWithPrefix(Name, "__amdgpu_enqueued_kernel",
                                    M.getDataLayout());
         F.setName(Name);
       }
       LLVM_DEBUG(dbgs() << "found enqueued kernel: " << F.getName() << '\n');
       auto RuntimeHandle = (F.getName() + ".runtime_handle").str();
-      auto T = Type::getInt8Ty(C)->getPointerTo(AMDGPUAS::GLOBAL_ADDRESS);
+      auto T = ArrayType::get(Type::getInt64Ty(C), 2);
       auto *GV = new GlobalVariable(
           M, T,
           /*IsConstant=*/false, GlobalValue::ExternalLinkage,
           /*Initializer=*/Constant::getNullValue(T), RuntimeHandle,
           /*InsertBefore=*/nullptr, GlobalValue::NotThreadLocal,
           AMDGPUAS::GLOBAL_ADDRESS,
           /*IsExternallyInitialized=*/false);
       LLVM_DEBUG(dbgs() << "runtime handle created: " << *GV << '\n');

test/CodeGen/AMDGPU/enqueue-kernel.ll

 ; RUN: opt -data-layout=A5 -amdgpu-lower-enqueued-block -S < %s | FileCheck %s
 
-; CHECK: @__test_block_invoke_kernel.runtime_handle = addrspace(1) global i8 addrspace(1)* null
-; CHECK: @__test_block_invoke_2_kernel.runtime_handle = addrspace(1) global i8 addrspace(1)* null
-; CHECK: @__amdgpu_enqueued_kernel.runtime_handle = addrspace(1) global i8 addrspace(1)* null
-; CHECK: @__amdgpu_enqueued_kernel.1.runtime_handle = addrspace(1) global i8 addrspace(1)* null
+; CHECK: @__test_block_invoke_kernel.runtime_handle = addrspace(1) global [2 x i64] zeroinitializer
+; CHECK: @__test_block_invoke_2_kernel.runtime_handle = addrspace(1) global [2 x i64] zeroinitializer
+; CHECK: @__amdgpu_enqueued_kernel.runtime_handle = addrspace(1) global [2 x i64] zeroinitializer
+; CHECK: @__amdgpu_enqueued_kernel.1.runtime_handle = addrspace(1) global [2 x i64] zeroinitializer
 
 %struct.ndrange_t = type { i32 }
 %opencl.queue_t = type opaque
 
 ; CHECK-LABEL: define amdgpu_kernel void @non_caller
 ; CHECK-NOT: #{{[0-9]+}}
 define amdgpu_kernel void @non_caller(i8 addrspace(1)* %a, i8 %b, i64 addrspace(1)* %c, i64 %d) local_unnamed_addr
   !kernel_arg_addr_space !3 !kernel_arg_access_qual !4 !kernel_arg_type !5 !kernel_arg_base_type !5 !kernel_arg_type_qual !6 {
%tmp9 = call i32 @__enqueue_kernel_basic(%opencl.queue_t addrspace(1)* undef, i32 0, %struct.ndrange_t addrspace(5)* byval nonnull %tmp3,
     i8* bitcast (void (<{ i32, i32, i8 addrspace(1)*, i64 addrspace(1)*, i64, i8 }>)* @__test_block_invoke_2_kernel to i8*), i8* nonnull %tmp8) #2
   ret void
 }
 
 ; __enqueue_kernel* functions may get inlined
 ; CHECK-LABEL: define amdgpu_kernel void @inlined_caller
 ; CHECK-SAME: #[[AT_CALLER]]
 ; CHECK-NOT: @__test_block_invoke_kernel
-; CHECK: load i64, i64 addrspace(1)* bitcast (i8 addrspace(1)* addrspace(1)* @__test_block_invoke_kernel.runtime_handle to i64 addrspace(1)*)
+; CHECK: load i64, i64 addrspace(1)* getelementptr inbounds ([2 x i64], [2 x i64] addrspace(1)* @__test_block_invoke_kernel.runtime_handle, i32 0, i32 0)
 define amdgpu_kernel void @inlined_caller(i8 addrspace(1)* %a, i8 %b, i64 addrspace(1)* %c, i64 %d) local_unnamed_addr
   !kernel_arg_addr_space !3 !kernel_arg_access_qual !4 !kernel_arg_type !5 !kernel_arg_base_type !5 !kernel_arg_type_qual !6 {
 entry:
   %tmp = load i64, i64 addrspace(1)* addrspacecast (i64* bitcast (void (<{ i32, i32, i8 addrspace(1)*, i8 }>)* @__test_block_invoke_kernel to i64*) to i64 addrspace(1)*)
   store i64 %tmp, i64 addrspace(1)* %c
   ret void
 }