yaxunl created this revision.
Herald added subscribers: t-tye, Anastasia, tpr, dstuttard, nhaehnle, wdng,
kzhuravl.
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.
https://reviews.llvm.org/D33445
Files:
lib/Basic/Targets.cpp
test/CodeGenOpenCL/amdgpu-alignment.cl
test/CodeGenOpenCL/amdgpu-debug-info-pointer-address-space.cl
test/CodeGenOpenCL/amdgpu-debug-info-variable-expression.cl
test/CodeGenOpenCL/amdgpu-nullptr.cl
test/CodeGenOpenCL/builtins-amdgcn.cl
test/CodeGenOpenCL/byval.cl
test/CodeGenOpenCL/size_t.cl
test/Sema/sizeof-struct-non-zero-as-member.cl
Index: test/Sema/sizeof-struct-non-zero-as-member.cl
===================================================================
--- test/Sema/sizeof-struct-non-zero-as-member.cl
+++ test/Sema/sizeof-struct-non-zero-as-member.cl
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -verify -fsyntax-only -triple amdgcn--opencl -target-cpu verde -S -emit-llvm -o - %s
+// RUN: %clang_cc1 -verify -fsyntax-only -triple amdgcn -target-cpu verde -S -emit-llvm -o - %s
// expected-no-diagnostics
// Record lowering was crashing on SI and newer targets, because it
Index: test/CodeGenOpenCL/size_t.cl
===================================================================
--- test/CodeGenOpenCL/size_t.cl
+++ test/CodeGenOpenCL/size_t.cl
@@ -1,6 +1,6 @@
// 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---opencl -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 -o - | FileCheck --check-prefix=SZ64 --check-prefix=AMDONLY %s
//SZ32: define{{.*}} i32 @test_ptrtoint_private(i8* %x)
//SZ32: ptrtoint i8* %{{.*}} to i32
Index: test/CodeGenOpenCL/byval.cl
===================================================================
--- test/CodeGenOpenCL/byval.cl
+++ test/CodeGenOpenCL/byval.cl
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -emit-llvm -o - -triple amdgcn---opencl %s | FileCheck %s
+// RUN: %clang_cc1 -emit-llvm -o - -triple amdgcn %s | FileCheck %s
// RUN: %clang_cc1 -emit-llvm -o - -triple amdgcn---amdgizcl %s | FileCheck %s -check-prefix=AMDGIZ
struct A {
Index: test/CodeGenOpenCL/builtins-amdgcn.cl
===================================================================
--- test/CodeGenOpenCL/builtins-amdgcn.cl
+++ test/CodeGenOpenCL/builtins-amdgcn.cl
@@ -1,5 +1,5 @@
// REQUIRES: amdgpu-registered-target
-// RUN: %clang_cc1 -triple amdgcn-unknown-unknown-opencl -S -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple amdgcn-unknown-unknown -S -emit-llvm -o - %s | FileCheck %s
#pragma OPENCL EXTENSION cl_khr_fp64 : enable
Index: test/CodeGenOpenCL/amdgpu-nullptr.cl
===================================================================
--- test/CodeGenOpenCL/amdgpu-nullptr.cl
+++ test/CodeGenOpenCL/amdgpu-nullptr.cl
@@ -1,5 +1,5 @@
-// 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
+// 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
typedef struct {
private char *p1;
Index: test/CodeGenOpenCL/amdgpu-debug-info-variable-expression.cl
===================================================================
--- test/CodeGenOpenCL/amdgpu-debug-info-variable-expression.cl
+++ test/CodeGenOpenCL/amdgpu-debug-info-variable-expression.cl
@@ -1,4 +1,4 @@
-// RUN: %clang -cl-std=CL2.0 -emit-llvm -g -O0 -S -target amdgcn-amd-amdhsa-opencl -mcpu=fiji -o - %s | FileCheck %s
+// RUN: %clang -cl-std=CL2.0 -emit-llvm -g -O0 -S -target amdgcn-amd-amdhsa -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)
Index: test/CodeGenOpenCL/amdgpu-debug-info-pointer-address-space.cl
===================================================================
--- test/CodeGenOpenCL/amdgpu-debug-info-pointer-address-space.cl
+++ test/CodeGenOpenCL/amdgpu-debug-info-pointer-address-space.cl
@@ -1,4 +1,4 @@
-// RUN: %clang -cl-std=CL2.0 -emit-llvm -g -O0 -S -target amdgcn-amd-amdhsa-opencl -mcpu=fiji -o - %s | FileCheck %s
+// RUN: %clang -cl-std=CL2.0 -emit-llvm -g -O0 -S -target amdgcn-amd-amdhsa -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)
Index: test/CodeGenOpenCL/amdgpu-alignment.cl
===================================================================
--- test/CodeGenOpenCL/amdgpu-alignment.cl
+++ test/CodeGenOpenCL/amdgpu-alignment.cl
@@ -1,5 +1,5 @@
// REQUIRES: amdgpu-registered-target
-// RUN: %clang_cc1 -triple amdgcn-unknown-unknown-opencl -S -disable-llvm-passes -emit-llvm -o - %s | FileCheck %s
+// RUN: %clang_cc1 -triple amdgcn-unknown-unknown -S -disable-llvm-passes -emit-llvm -o - %s | FileCheck %s
#pragma OPENCL EXTENSION cl_khr_fp64 : enable
#pragma OPENCL EXTENSION cl_khr_fp16 : enable
Index: lib/Basic/Targets.cpp
===================================================================
--- lib/Basic/Targets.cpp
+++ lib/Basic/Targets.cpp
@@ -2169,15 +2169,20 @@
: 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;
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
