[AMDGPU] Do not require opencl triple environment for OpenCL

[yaxunl]

Summary

A recent change requires opencl triple environment for compiling OpenCL
program, which causes regressions in libclc.

This patch fixes that. Instead of deducing language based on triple
environment, it checks LangOptions.

[tstellar]

Thanks!

test/CodeGenOpenCL/amdgpu-alignment.cl

2 ↗

(On Diff #99921)

You might want to have two run lines for each test, one with no environment and one with the -opencl environment to make sure that doesn't break in the future. Otherwise, LGTM.

[yaxunl]

test/CodeGenOpenCL/amdgpu-alignment.cl

2 ↗

(On Diff #99921)

I will add them when committing. Thanks.

cfe/trunk/lib/Basic/Targets.cpp

AMDGPUTargetInfo(const llvm::Triple &Triple, const TargetOptions &Opts)
     }
     auto IsGenericZero = isGenericZero(Triple);
     resetDataLayout(getTriple().getArch() == llvm::Triple::amdgcn ?
                     (IsGenericZero ? DataLayoutStringSIGenericIsZero :
                         DataLayoutStringSIPrivateIsZero)
                     : DataLayoutStringR600);
     assert(DataLayout->getAllocaAddrSpace() == AS.Private);
 
-    AddrSpaceMap =
-        llvm::StringSwitch<const LangAS::Map *>(Triple.getEnvironmentName())
-            .Case("opencl", &AMDGPUOpenCLPrivateIsZeroMap)
-            .Case("amdgiz", &AMDGPUNonOpenCLGenericIsZeroMap)
-            .Case("amdgizcl", &AMDGPUOpenCLGenericIsZeroMap)
-            .Default(&AMDGPUNonOpenCLPrivateIsZeroMap);
     UseAddrSpaceMapMangling = true;
   }
 
+  void adjust(LangOptions &Opts) override {
+    TargetInfo::adjust(Opts);
+    if (isGenericZero(getTriple())) {
+      AddrSpaceMap = Opts.OpenCL ? &AMDGPUOpenCLGenericIsZeroMap
+                                 : &AMDGPUNonOpenCLGenericIsZeroMap;
+    } else {
+      AddrSpaceMap = Opts.OpenCL ? &AMDGPUOpenCLPrivateIsZeroMap
+                                 : &AMDGPUNonOpenCLPrivateIsZeroMap;
+    }
+  }
+
   uint64_t getPointerWidthV(unsigned AddrSpace) const override {
     if (GPU <= GK_CAYMAN)
       return 32;
 
     if (AddrSpace == AS.Private || AddrSpace == AS.Local) {
       return 32;
     }
     return 64;

cfe/trunk/test/CodeGenOpenCL/amdgpu-alignment.cl

 // REQUIRES: amdgpu-registered-target
+// RUN: %clang_cc1 -triple amdgcn-unknown-unknown -S -disable-llvm-passes -emit-llvm -o - %s | FileCheck %s
 // RUN: %clang_cc1 -triple amdgcn-unknown-unknown-opencl -S -disable-llvm-passes -emit-llvm -o - %s | FileCheck %s
 
 #pragma OPENCL EXTENSION cl_khr_fp64 : enable
 #pragma OPENCL EXTENSION cl_khr_fp16 : enable
 
 typedef char __attribute__((ext_vector_type(2))) char2;
 typedef char __attribute__((ext_vector_type(3))) char3;
 typedef char __attribute__((ext_vector_type(4))) char4;

cfe/trunk/test/CodeGenOpenCL/amdgpu-debug-info-pointer-address-space.cl

+// RUN: %clang -cl-std=CL2.0 -emit-llvm -g -O0 -S -target amdgcn-amd-amdhsa -mcpu=fiji -o - %s | FileCheck %s
 // RUN: %clang -cl-std=CL2.0 -emit-llvm -g -O0 -S -target amdgcn-amd-amdhsa-opencl -mcpu=fiji -o - %s | FileCheck %s
 
 // CHECK-DAG: ![[DWARF_ADDRESS_SPACE_NONE:[0-9]+]] = !DIDerivedType(tag: DW_TAG_pointer_type, baseType: !{{[0-9]+}}, size: {{[0-9]+}})
 // CHECK-DAG: ![[DWARF_ADDRESS_SPACE_LOCAL:[0-9]+]] = !DIDerivedType(tag: DW_TAG_pointer_type, baseType: !{{[0-9]+}}, size: {{[0-9]+}}, dwarfAddressSpace: 2)
 // CHECK-DAG: ![[DWARF_ADDRESS_SPACE_PRIVATE:[0-9]+]] = !DIDerivedType(tag: DW_TAG_pointer_type, baseType: !{{[0-9]+}}, size: {{[0-9]+}}, dwarfAddressSpace: 1)
 
 // CHECK-DAG: distinct !DIGlobalVariable(name: "FileVar0", scope: !{{[0-9]+}}, file: !{{[0-9]+}}, line: {{[0-9]+}}, type: ![[DWARF_ADDRESS_SPACE_NONE]], isLocal: false, isDefinition: true)
 global int *FileVar0;

cfe/trunk/test/CodeGenOpenCL/amdgpu-debug-info-variable-expression.cl

+// RUN: %clang -cl-std=CL2.0 -emit-llvm -g -O0 -S -target amdgcn-amd-amdhsa -mcpu=fiji -o - %s | FileCheck %s
 // RUN: %clang -cl-std=CL2.0 -emit-llvm -g -O0 -S -target amdgcn-amd-amdhsa-opencl -mcpu=fiji -o - %s | FileCheck %s
 
 // CHECK-DAG: ![[LOCAL:[0-9]+]] = !DIExpression(DW_OP_constu, 2, DW_OP_swap, DW_OP_xderef)
 // CHECK-DAG: ![[PRIVATE:[0-9]+]] = !DIExpression(DW_OP_constu, 1, DW_OP_swap, DW_OP_xderef)
 
 // CHECK-DAG: ![[FILEVAR0:[0-9]+]] = distinct !DIGlobalVariable(name: "FileVar0", scope: !{{[0-9]+}}, file: !{{[0-9]+}}, line: {{[0-9]+}}, type: !{{[0-9]+}}, isLocal: false, isDefinition: true)
 // CHECK-DAG: !DIGlobalVariableExpression(var: ![[FILEVAR0]])
 global int *FileVar0;

cfe/trunk/test/CodeGenOpenCL/amdgpu-nullptr.cl

+// RUN: %clang_cc1 %s -cl-std=CL2.0 -include opencl-c.h -triple amdgcn -emit-llvm -o - | FileCheck %s
+// RUN: %clang_cc1 %s -O0 -cl-std=CL2.0 -include opencl-c.h -triple amdgcn -emit-llvm -o - | FileCheck --check-prefix=NOOPT %s
 // RUN: %clang_cc1 %s -cl-std=CL2.0 -include opencl-c.h -triple amdgcn---opencl -emit-llvm -o - | FileCheck %s
-// RUN: %clang_cc1 %s -O0 -cl-std=CL2.0 -include opencl-c.h -triple amdgcn---opencl -emit-llvm -o - | FileCheck --check-prefix=NOOPT %s
 
 typedef struct {
   private char *p1;
   local char *p2;
   constant char *p3;
   global char *p4;
   generic char *p5;
 } StructTy1;

cfe/trunk/test/CodeGenOpenCL/builtins-amdgcn.cl

 // REQUIRES: amdgpu-registered-target
+// RUN: %clang_cc1 -triple amdgcn-unknown-unknown -S -emit-llvm -o - %s | FileCheck %s
 // RUN: %clang_cc1 -triple amdgcn-unknown-unknown-opencl -S -emit-llvm -o - %s | FileCheck %s
 
 #pragma OPENCL EXTENSION cl_khr_fp64 : enable
 
 typedef unsigned long ulong;
 typedef unsigned int uint;
 
 // CHECK-LABEL: @test_div_scale_f64

cfe/trunk/test/CodeGenOpenCL/byval.cl

+// RUN: %clang_cc1 -emit-llvm -o - -triple amdgcn %s | FileCheck %s
 // RUN: %clang_cc1 -emit-llvm -o - -triple amdgcn---opencl %s | FileCheck %s
 // RUN: %clang_cc1 -emit-llvm -o - -triple amdgcn---amdgizcl %s | FileCheck %s -check-prefix=AMDGIZ
 
 struct A {
   int x[100];
 };
 
 int f(struct A a);

cfe/trunk/test/CodeGenOpenCL/size_t.cl

 // RUN: %clang_cc1 %s -cl-std=CL2.0 -finclude-default-header -emit-llvm -O0 -triple spir-unknown-unknown -o - | FileCheck --check-prefix=SZ32 %s
 // RUN: %clang_cc1 %s -cl-std=CL2.0 -finclude-default-header -emit-llvm -O0 -triple spir64-unknown-unknown -o - | FileCheck --check-prefix=SZ64 --check-prefix=SZ64ONLY %s
+// RUN: %clang_cc1 %s -cl-std=CL2.0 -finclude-default-header -emit-llvm -O0 -triple amdgcn -o - | FileCheck --check-prefix=SZ64 --check-prefix=AMDONLY %s
 // RUN: %clang_cc1 %s -cl-std=CL2.0 -finclude-default-header -emit-llvm -O0 -triple amdgcn---opencl -o - | FileCheck --check-prefix=SZ64 --check-prefix=AMDONLY %s
 
 //SZ32: define{{.*}} i32 @test_ptrtoint_private(i8* %x)
 //SZ32: ptrtoint i8* %{{.*}} to i32
 //SZ64: define{{.*}} i64 @test_ptrtoint_private(i8* %x)
 //SZ64: ptrtoint i8* %{{.*}} to i64
 size_t test_ptrtoint_private(private char* x) {
   return (size_t)x;

cfe/trunk/test/Sema/sizeof-struct-non-zero-as-member.cl

+// RUN: %clang_cc1 -verify -fsyntax-only -triple amdgcn -target-cpu verde -S -emit-llvm -o - %s
 // RUN: %clang_cc1 -verify -fsyntax-only -triple amdgcn--opencl -target-cpu verde -S -emit-llvm -o - %s
 // expected-no-diagnostics
 
 // Record lowering was crashing on SI and newer targets, because it
 // was using the wrong size for test::ptr.  Since global memory
 // has 64-bit pointers, sizeof(test::ptr) should be 8.
 
 struct test_as0 {int *ptr;};