AMDGPU: Remove error on calls for amdgcn

Repurpose the -amdgpu-function-calls flag. Rather
than require it to emit a call, only use it to
run the always inline path or not.

llvm-svn: 310003

diff --git a/llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp b/llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp
index 9c74607..3bd334c 100644 (file)
--- a/llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp
@@ -290,11 +290,6 @@ AMDGPUTargetMachine::AMDGPUTargetMachine(const Target &T, const Triple &TT,
 
 AMDGPUTargetMachine::~AMDGPUTargetMachine() = default;
 
-bool AMDGPUTargetMachine::enableFunctionCalls() const {
-  return EnableAMDGPUFunctionCalls &&
-         getTargetTriple().getArch() == Triple::amdgcn;
-}
-
 StringRef AMDGPUTargetMachine::getGPUName(const Function &F) const {
   Attribute GPUAttr = F.getFnAttribute("target-cpu");
   return GPUAttr.hasAttribute(Attribute::None) ?
@@ -503,8 +498,10 @@ class GCNPassConfig final : public AMDGPUPassConfig {
 public:
   GCNPassConfig(LLVMTargetMachine &TM, PassManagerBase &PM)
     : AMDGPUPassConfig(TM, PM) {
-    // It is necessary to know the register usage of the entire call graph.
-    setRequiresCodeGenSCCOrder(EnableAMDGPUFunctionCalls);
+    // It is necessary to know the register usage of the entire call graph.  We
+    // allow calls without EnableAMDGPUFunctionCalls if they are marked
+    // noinline, so this is always required.
+    setRequiresCodeGenSCCOrder(true);
   }
 
   GCNTargetMachine &getGCNTargetMachine() const {
@@ -571,15 +568,18 @@ void AMDGPUPassConfig::addIRPasses() {
 
   addPass(createAMDGPULowerIntrinsicsPass());
 
-  // Function calls are not supported, so make sure we inline everything.
-  addPass(createAMDGPUAlwaysInlinePass());
-  addPass(createAlwaysInlinerLegacyPass());
-  // We need to add the barrier noop pass, otherwise adding the function
-  // inlining pass will cause all of the PassConfigs passes to be run
-  // one function at a time, which means if we have a nodule with two
-  // functions, then we will generate code for the first function
-  // without ever running any passes on the second.
-  addPass(createBarrierNoopPass());
+  if (TM.getTargetTriple().getArch() == Triple::r600 ||
+      !EnableAMDGPUFunctionCalls) {
+    // Function calls are not supported, so make sure we inline everything.
+    addPass(createAMDGPUAlwaysInlinePass());
+    addPass(createAlwaysInlinerLegacyPass());
+    // We need to add the barrier noop pass, otherwise adding the function
+    // inlining pass will cause all of the PassConfigs passes to be run
+    // one function at a time, which means if we have a nodule with two
+    // functions, then we will generate code for the first function
+    // without ever running any passes on the second.
+    addPass(createBarrierNoopPass());
+  }
 
   if (TM.getTargetTriple().getArch() == Triple::amdgcn) {
     // TODO: May want to move later or split into an early and late one.

diff --git a/llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.h b/llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.h
index 0b1fd5e..cc36e93 100644 (file)
--- a/llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.h
+++ b/llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.h
@@ -69,9 +69,6 @@ public:
       return -1;
     return 0;
   }
-
-  LLVM_READONLY
-  bool enableFunctionCalls() const;
 };
 
 //===----------------------------------------------------------------------===//

diff --git a/llvm/lib/Target/AMDGPU/SIISelLowering.cpp b/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
index da7d04b..68881ef 100644 (file)
--- a/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
+++ b/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
@@ -1962,11 +1962,6 @@ void SITargetLowering::passSpecialInputs(
 // The wave scratch offset register is used as the global base pointer.
 SDValue SITargetLowering::LowerCall(CallLoweringInfo &CLI,
                                     SmallVectorImpl<SDValue> &InVals) const {
-  const AMDGPUTargetMachine &TM =
-    static_cast<const AMDGPUTargetMachine &>(getTargetMachine());
-  if (!TM.enableFunctionCalls())
-    return AMDGPUTargetLowering::LowerCall(CLI, InVals);
-
   SelectionDAG &DAG = CLI.DAG;
   const SDLoc &DL = CLI.DL;
   SmallVector<ISD::OutputArg, 32> &Outs = CLI.Outs;

diff --git a/llvm/test/CodeGen/AMDGPU/basic-call-return.ll b/llvm/test/CodeGen/AMDGPU/basic-call-return.ll
index 6588a1c..1f2f92f 100644 (file)
--- a/llvm/test/CodeGen/AMDGPU/basic-call-return.ll
+++ b/llvm/test/CodeGen/AMDGPU/basic-call-return.ll
@@ -1,6 +1,6 @@
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -amdgpu-function-calls -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=hawaii -amdgpu-function-calls -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -amdgpu-function-calls -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=hawaii -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
 
 define void @void_func_void() #2 {
   ret void

diff --git a/llvm/test/CodeGen/AMDGPU/byval-frame-setup.ll b/llvm/test/CodeGen/AMDGPU/byval-frame-setup.ll
index 6ae5aab..723c78a 100644 (file)
--- a/llvm/test/CodeGen/AMDGPU/byval-frame-setup.ll
+++ b/llvm/test/CodeGen/AMDGPU/byval-frame-setup.ll
@@ -1,5 +1,5 @@
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -amdgpu-function-calls -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,VI %s
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=hawaii -amdgpu-function-calls -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,CI %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,VI %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=hawaii -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,CI %s
 
 %struct.ByValStruct = type { [4 x i32] }
 

diff --git a/llvm/test/CodeGen/AMDGPU/call-argument-types.ll b/llvm/test/CodeGen/AMDGPU/call-argument-types.ll
index 1f3ccba..589b333 100644 (file)
--- a/llvm/test/CodeGen/AMDGPU/call-argument-types.ll
+++ b/llvm/test/CodeGen/AMDGPU/call-argument-types.ll
@@ -1,7 +1,7 @@
-; RUN: llc -march=amdgcn -mcpu=fiji -mattr=-flat-for-global -amdgpu-function-calls -amdgpu-scalarize-global-loads=0 -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,VI,MESA %s
-; RUN: llc -march=amdgcn -mcpu=hawaii -amdgpu-function-calls -amdgpu-scalarize-global-loads=0 -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,CI,MESA %s
-; RUN: llc -march=amdgcn -mcpu=gfx900 -mattr=-flat-for-global -amdgpu-function-calls -amdgpu-scalarize-global-loads=0 -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GFX9,VI,MESA %s
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -mattr=-flat-for-global -amdgpu-function-calls -amdgpu-scalarize-global-loads=0 -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,VI,HSA %s
+; RUN: llc -march=amdgcn -mcpu=fiji -mattr=-flat-for-global -amdgpu-scalarize-global-loads=0 -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,VI,MESA %s
+; RUN: llc -march=amdgcn -mcpu=hawaii -amdgpu-scalarize-global-loads=0 -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,CI,MESA %s
+; RUN: llc -march=amdgcn -mcpu=gfx900 -mattr=-flat-for-global -amdgpu-scalarize-global-loads=0 -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GFX9,VI,MESA %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -mattr=-flat-for-global -amdgpu-scalarize-global-loads=0 -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,VI,HSA %s
 
 declare void @external_void_func_i1(i1) #0
 declare void @external_void_func_i1_signext(i1 signext) #0

diff --git a/llvm/test/CodeGen/AMDGPU/call-encoding.ll b/llvm/test/CodeGen/AMDGPU/call-encoding.ll
index 01e3536..f018e9b 100644 (file)
--- a/llvm/test/CodeGen/AMDGPU/call-encoding.ll
+++ b/llvm/test/CodeGen/AMDGPU/call-encoding.ll
@@ -1,6 +1,6 @@
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -amdgpu-function-calls -filetype=obj -verify-machineinstrs < %s | llvm-objdump -triple amdgcn--amdhsa -mcpu=fiji -d - | FileCheck -check-prefixes=GCN,VI %s
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -amdgpu-function-calls -filetype=obj -verify-machineinstrs < %s | llvm-objdump -triple amdgcn--amdhsa -mcpu=gfx900 -d - | FileCheck -check-prefixes=GCN,GFX9 %s
-; XUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=hawaii -amdgpu-function-calls -filetype=obj -verify-machineinstrs < %s | llvm-objdump -triple amdgcn--amdhsa -mcpu=hawaii -d - | FileCheck -check-prefixes=GCN,CI %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -filetype=obj -verify-machineinstrs < %s | llvm-objdump -triple amdgcn--amdhsa -mcpu=fiji -d - | FileCheck -check-prefixes=GCN,VI %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -filetype=obj -verify-machineinstrs < %s | llvm-objdump -triple amdgcn--amdhsa -mcpu=gfx900 -d - | FileCheck -check-prefixes=GCN,GFX9 %s
+; XUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=hawaii -filetype=obj -verify-machineinstrs < %s | llvm-objdump -triple amdgcn--amdhsa -mcpu=hawaii -d - | FileCheck -check-prefixes=GCN,CI %s
 
 ; GCN: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
 ; GCN-NEXT: s_setpc_b64

diff --git a/llvm/test/CodeGen/AMDGPU/call-graph-register-usage.ll b/llvm/test/CodeGen/AMDGPU/call-graph-register-usage.ll
index 7c39831..0a9468f 100644 (file)
--- a/llvm/test/CodeGen/AMDGPU/call-graph-register-usage.ll
+++ b/llvm/test/CodeGen/AMDGPU/call-graph-register-usage.ll
@@ -1,6 +1,6 @@
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -amdgpu-function-calls -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,CI %s
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -amdgpu-function-calls -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,VI,VI-NOBUG %s
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=iceland -amdgpu-function-calls -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,VI,VI-BUG %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,CI %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,VI,VI-NOBUG %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=iceland -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,VI,VI-BUG %s
 
 ; Make sure to run a GPU with the SGPR allocation bug.
 

diff --git a/llvm/test/CodeGen/AMDGPU/call-preserved-registers.ll b/llvm/test/CodeGen/AMDGPU/call-preserved-registers.ll
index 8dac066..1812261 100644 (file)
--- a/llvm/test/CodeGen/AMDGPU/call-preserved-registers.ll
+++ b/llvm/test/CodeGen/AMDGPU/call-preserved-registers.ll
@@ -1,6 +1,6 @@
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -amdgpu-function-calls -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=hawaii -amdgpu-function-calls -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -amdgpu-function-calls -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=hawaii -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
 
 declare void @external_void_func_void() #0
 

diff --git a/llvm/test/CodeGen/AMDGPU/call-return-types.ll b/llvm/test/CodeGen/AMDGPU/call-return-types.ll
index 2d6d509..3f607c1 100644 (file)
--- a/llvm/test/CodeGen/AMDGPU/call-return-types.ll
+++ b/llvm/test/CodeGen/AMDGPU/call-return-types.ll
@@ -1,6 +1,6 @@
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -mattr=-flat-for-global -amdgpu-function-calls -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=hawaii -amdgpu-function-calls -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -mattr=-flat-for-global -amdgpu-function-calls -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=hawaii -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
 
 declare void @external_void_func_void() #0
 

diff --git a/llvm/test/CodeGen/AMDGPU/call.ll b/llvm/test/CodeGen/AMDGPU/call.ll
index 769c7bb..85c3792 100644 (file)
--- a/llvm/test/CodeGen/AMDGPU/call.ll
+++ b/llvm/test/CodeGen/AMDGPU/call.ll
@@ -1,5 +1,3 @@
-; RUN: not llc -march=amdgcn -verify-machineinstrs < %s 2>&1 | FileCheck %s
-; RUN: not llc -march=amdgcn -mcpu=tonga -verify-machineinstrs < %s 2>&1 | FileCheck %s
 ; RUN: not llc -march=r600 -mcpu=cypress < %s 2>&1 | FileCheck %s
 
 ; CHECK: in function test_call_external{{.*}}: unsupported call to function external_function

diff --git a/llvm/test/CodeGen/AMDGPU/callee-frame-setup.ll b/llvm/test/CodeGen/AMDGPU/callee-frame-setup.ll
index d0edcf8..9e01267 100644 (file)
--- a/llvm/test/CodeGen/AMDGPU/callee-frame-setup.ll
+++ b/llvm/test/CodeGen/AMDGPU/callee-frame-setup.ll
@@ -1,5 +1,5 @@
-; RUN: llc -amdgpu-function-calls -march=amdgcn -mcpu=hawaii -verify-machineinstrs < %s | FileCheck  -enable-var-scope -check-prefix=GCN -check-prefix=CI %s
-; RUN: llc -amdgpu-function-calls -march=amdgcn -mcpu=gfx900 -verify-machineinstrs < %s | FileCheck  -enable-var-scope -check-prefix=GCN -check-prefix=GFX9 %s
+; RUN: llc -march=amdgcn -mcpu=hawaii -verify-machineinstrs < %s | FileCheck  -enable-var-scope -check-prefix=GCN -check-prefix=CI %s
+; RUN: llc -march=amdgcn -mcpu=gfx900 -verify-machineinstrs < %s | FileCheck  -enable-var-scope -check-prefix=GCN -check-prefix=GFX9 %s
 
 ; GCN-LABEL: {{^}}callee_no_stack:
 ; GCN: ; BB#0:

diff --git a/llvm/test/CodeGen/AMDGPU/callee-special-input-sgprs.ll b/llvm/test/CodeGen/AMDGPU/callee-special-input-sgprs.ll
index eb8c428..11e1a95 100644 (file)
--- a/llvm/test/CodeGen/AMDGPU/callee-special-input-sgprs.ll
+++ b/llvm/test/CodeGen/AMDGPU/callee-special-input-sgprs.ll
@@ -1,5 +1,5 @@
-; RUN: llc -amdgpu-function-calls -mtriple=amdgcn-amd-amdhsa -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,CIVI %s
-; RUN: llc -amdgpu-function-calls -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GFX9 %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,CIVI %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GFX9 %s
 
 ; GCN-LABEL: {{^}}use_dispatch_ptr:
 ; GCN: s_load_dword s{{[0-9]+}}, s[6:7], 0x0

diff --git a/llvm/test/CodeGen/AMDGPU/callee-special-input-vgprs.ll b/llvm/test/CodeGen/AMDGPU/callee-special-input-vgprs.ll
index fa8a2d2..10ccce6 100644 (file)
--- a/llvm/test/CodeGen/AMDGPU/callee-special-input-vgprs.ll
+++ b/llvm/test/CodeGen/AMDGPU/callee-special-input-vgprs.ll
@@ -1,4 +1,4 @@
-; RUN: llc -amdgpu-function-calls -mtriple=amdgcn-amd-amdhsa -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefix=GCN %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefix=GCN %s
 
 ; GCN-LABEL: {{^}}use_workitem_id_x:
 ; GCN: s_waitcnt

diff --git a/llvm/test/CodeGen/AMDGPU/frame-index-elimination.ll b/llvm/test/CodeGen/AMDGPU/frame-index-elimination.ll
index 727c754..61b0982 100644 (file)
--- a/llvm/test/CodeGen/AMDGPU/frame-index-elimination.ll
+++ b/llvm/test/CodeGen/AMDGPU/frame-index-elimination.ll
@@ -1,4 +1,4 @@
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -mattr=-promote-alloca -amdgpu-function-calls -amdgpu-sroa=0 -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefix=GCN %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mattr=-promote-alloca -amdgpu-sroa=0 -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefix=GCN %s
 
 ; Test that non-entry function frame indices are expanded properly to
 ; give an index relative to the scratch wave offset register
@@ -144,6 +144,7 @@ define void @func_other_fi_user_non_inline_imm_offset_i32() #0 {
   store volatile i32 %mul, i32 addrspace(3)* undef
   ret void
 }
+
 ; GCN-LABEL: {{^}}func_other_fi_user_non_inline_imm_offset_i32_vcc_live:
 ; GCN: s_sub_u32 [[DIFF:s[0-9]+]], s5, s4
 ; GCN-DAG: v_lshr_b32_e64 [[SCALED:v[0-9]+]], [[DIFF]], 6

diff --git a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.implicitarg.ptr.ll b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.implicitarg.ptr.ll
index 7108884..e071e67 100644 (file)
--- a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.implicitarg.ptr.ll
+++ b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.implicitarg.ptr.ll
@@ -1,6 +1,6 @@
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -amdgpu-function-calls -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,HSA,HSA-NOENV %s
-; RUN: llc -mtriple=amdgcn-amd-amdhsa-opencl -amdgpu-function-calls -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,HSA,HSA-OPENCL %s
-; RUN: llc -mtriple=amdgcn-mesa-mesa3d -amdgpu-function-calls -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,MESA %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,HSA,HSA-NOENV %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa-opencl -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,HSA,HSA-OPENCL %s
+; RUN: llc -mtriple=amdgcn-mesa-mesa3d -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,MESA %s
 
 ; GCN-LABEL: {{^}}kernel_implicitarg_ptr_empty:
 ; GCN: enable_sgpr_kernarg_segment_ptr = 1

diff --git a/llvm/test/CodeGen/AMDGPU/mem-builtins.ll b/llvm/test/CodeGen/AMDGPU/mem-builtins.ll
index 1cbd0c3..c7607a3 100644 (file)
--- a/llvm/test/CodeGen/AMDGPU/mem-builtins.ll
+++ b/llvm/test/CodeGen/AMDGPU/mem-builtins.ll
@@ -1,4 +1,5 @@
-; RUN: not llc -march=amdgcn -verify-machineinstrs < %s 2>&1 | FileCheck -check-prefix=ERROR %s
+; RUN: not llc -march=r600 < %s 2>&1 | FileCheck -check-prefix=ERROR %s
+; RUN: llc -march=amdgcn < %s | FileCheck -check-prefix=GCN %s
 
 declare i32 @memcmp(i8 addrspace(1)* readonly nocapture, i8 addrspace(1)* readonly nocapture, i64) #0
 declare i8 addrspace(1)* @memchr(i8 addrspace(1)* readonly nocapture, i32, i64) #1
@@ -9,6 +10,9 @@ declare i32 @strcmp(i8* nocapture, i8* nocapture) #1
 
 
 ; ERROR: error: <unknown>:0:0: in function test_memcmp void (i8 addrspace(1)*, i8 addrspace(1)*, i32*): unsupported call to function memcmp
+
+; GCN: s_add_u32 s{{[0-9]+}}, s{{[0-9]+}}, memcmp@rel32@lo+4
+; GCN: s_addc_u32 s{{[0-9]+}}, s{{[0-9]+}}, memcmp@rel32@hi+4
 define amdgpu_kernel void @test_memcmp(i8 addrspace(1)* %x, i8 addrspace(1)* %y, i32* nocapture %p) #0 {
 entry:
   %cmp = tail call i32 @memcmp(i8 addrspace(1)* %x, i8 addrspace(1)* %y, i64 2)
@@ -17,6 +21,9 @@ entry:
 }
 
 ; ERROR: error: <unknown>:0:0: in function test_memchr void (i8 addrspace(1)*, i32, i64): unsupported call to function memchr
+
+; GCN: s_add_u32 s{{[0-9]+}}, s{{[0-9]+}}, memchr@rel32@lo+4
+; GCN: s_addc_u32 s{{[0-9]+}}, s{{[0-9]+}}, memchr@rel32@hi+4
 define amdgpu_kernel void @test_memchr(i8 addrspace(1)* %src, i32 %char, i64 %len) #0 {
   %res = call i8 addrspace(1)* @memchr(i8 addrspace(1)* %src, i32 %char, i64 %len)
   store volatile i8 addrspace(1)* %res, i8 addrspace(1)* addrspace(1)* undef
@@ -24,6 +31,9 @@ define amdgpu_kernel void @test_memchr(i8 addrspace(1)* %src, i32 %char, i64 %le
 }
 
 ; ERROR: error: <unknown>:0:0: in function test_strcpy void (i8*, i8*): unsupported call to function strcpy
+
+; GCN: s_add_u32 s{{[0-9]+}}, s{{[0-9]+}}, strcpy@rel32@lo+4
+; GCN: s_addc_u32 s{{[0-9]+}}, s{{[0-9]+}}, strcpy@rel32@hi+4
 define amdgpu_kernel void @test_strcpy(i8* %dst, i8* %src) #0 {
   %res = call i8* @strcpy(i8* %dst, i8* %src)
   store volatile i8* %res, i8* addrspace(1)* undef
@@ -31,6 +41,9 @@ define amdgpu_kernel void @test_strcpy(i8* %dst, i8* %src) #0 {
 }
 
 ; ERROR: error: <unknown>:0:0: in function test_strcmp void (i8*, i8*): unsupported call to function strcmp
+
+; GCN: s_add_u32 s{{[0-9]+}}, s{{[0-9]+}}, strcmp@rel32@lo+4
+; GCN: s_addc_u32 s{{[0-9]+}}, s{{[0-9]+}}, strcmp@rel32@hi+4
 define amdgpu_kernel void @test_strcmp(i8* %src0, i8* %src1) #0 {
   %res = call i32 @strcmp(i8* %src0, i8* %src1)
   store volatile i32 %res, i32 addrspace(1)* undef
@@ -38,6 +51,9 @@ define amdgpu_kernel void @test_strcmp(i8* %src0, i8* %src1) #0 {
 }
 
 ; ERROR: error: <unknown>:0:0: in function test_strlen void (i8*): unsupported call to function strlen
+
+; GCN: s_add_u32 s{{[0-9]+}}, s{{[0-9]+}}, strlen@rel32@lo+4
+; GCN: s_addc_u32 s{{[0-9]+}}, s{{[0-9]+}}, strlen@rel32@hi+4
 define amdgpu_kernel void @test_strlen(i8* %src) #0 {
   %res = call i32 @strlen(i8* %src)
   store volatile i32 %res, i32 addrspace(1)* undef
@@ -45,6 +61,9 @@ define amdgpu_kernel void @test_strlen(i8* %src) #0 {
 }
 
 ; ERROR: error: <unknown>:0:0: in function test_strnlen void (i8*, i32): unsupported call to function strnlen
+
+; GCN: s_add_u32 s{{[0-9]+}}, s{{[0-9]+}}, strnlen@rel32@lo+4
+; GCN: s_addc_u32 s{{[0-9]+}}, s{{[0-9]+}}, strnlen@rel32@hi+4
 define amdgpu_kernel void @test_strnlen(i8* %src, i32 %size) #0 {
   %res = call i32 @strnlen(i8* %src, i32 %size)
   store volatile i32 %res, i32 addrspace(1)* undef

diff --git a/llvm/test/CodeGen/AMDGPU/nested-calls.ll b/llvm/test/CodeGen/AMDGPU/nested-calls.ll
index f8ce818..4cefa7e 100644 (file)
--- a/llvm/test/CodeGen/AMDGPU/nested-calls.ll
+++ b/llvm/test/CodeGen/AMDGPU/nested-calls.ll
@@ -1,6 +1,6 @@
-; RUN: llc -march=amdgcn -mcpu=fiji -mattr=-flat-for-global -amdgpu-sroa=0 -amdgpu-function-calls -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefix=GCN -check-prefix=VI %s
-; RUN: llc -march=amdgcn -mcpu=hawaii -amdgpu-function-calls -amdgpu-sroa=0 -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefix=GCN -check-prefix=CI %s
-; RUN: llc -march=amdgcn -mcpu=gfx900 -mattr=-flat-for-global -amdgpu-sroa=0 -amdgpu-function-calls -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefix=GCN -check-prefix=GFX9 -check-prefix=VI %s
+; RUN: llc -march=amdgcn -mcpu=fiji -mattr=-flat-for-global -amdgpu-sroa=0 -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefix=GCN -check-prefix=VI %s
+; RUN: llc -march=amdgcn -mcpu=hawaii -amdgpu-sroa=0 -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefix=GCN -check-prefix=CI %s
+; RUN: llc -march=amdgcn -mcpu=gfx900 -mattr=-flat-for-global -amdgpu-sroa=0 -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefix=GCN -check-prefix=GFX9 -check-prefix=VI %s
 
 ; Test calls when called by other callable functions rather than
 ; kernels.

diff --git a/llvm/test/CodeGen/AMDGPU/private-memory-broken.ll b/llvm/test/CodeGen/AMDGPU/private-memory-broken.ll
deleted file mode 100644 (file)
index 9b5f655..0000000
--- a/llvm/test/CodeGen/AMDGPU/private-memory-broken.ll
+++ /dev/null
@@ -1,21 +0,0 @@
-; RUN: not llc -verify-machineinstrs -march=amdgcn %s -o /dev/null 2>&1 | FileCheck %s
-; RUN: not llc -verify-machineinstrs -march=amdgcn -mcpu=tonga %s -o /dev/null 2>&1 | FileCheck %s
-
-; Make sure promote alloca pass doesn't crash
-
-; CHECK: unsupported call
-
-declare i32 @foo(i32*) nounwind
-
-define amdgpu_kernel void @call_private(i32 addrspace(1)* %out, i32 %in) nounwind {
-entry:
-  %tmp = alloca [2 x i32]
-  %tmp1 = getelementptr [2 x i32], [2 x i32]* %tmp, i32 0, i32 0
-  %tmp2 = getelementptr [2 x i32], [2 x i32]* %tmp, i32 0, i32 1
-  store i32 0, i32* %tmp1
-  store i32 1, i32* %tmp2
-  %tmp3 = getelementptr [2 x i32], [2 x i32]* %tmp, i32 0, i32 %in
-  %val = call i32 @foo(i32* %tmp3) nounwind
-  store i32 %val, i32 addrspace(1)* %out
-  ret void
-}

diff --git a/llvm/test/CodeGen/AMDGPU/promote-alloca-bitcast-function.ll b/llvm/test/CodeGen/AMDGPU/promote-alloca-bitcast-function.ll
index a5eb92d..f880d21 100644 (file)
--- a/llvm/test/CodeGen/AMDGPU/promote-alloca-bitcast-function.ll
+++ b/llvm/test/CodeGen/AMDGPU/promote-alloca-bitcast-function.ll
@@ -1,19 +1,21 @@
 ; RUN: not llc -march=amdgcn < %s 2>&1 | FileCheck %s
 
+; CHECK: LLVM ERROR: indirect calls not handled
+
 ; Make sure that AMDGPUPromoteAlloca doesn't crash if the called
 ; function is a constantexpr cast of a function.
 
 declare void @foo(float*) #0
 declare void @foo.varargs(...) #0
 
-; CHECK: in function crash_call_constexpr_cast{{.*}}: unsupported call to function foo
+; XCHECK: in function crash_call_constexpr_cast{{.*}}: unsupported call to function foo
 define amdgpu_kernel void @crash_call_constexpr_cast() #0 {
   %alloca = alloca i32
   call void bitcast (void (float*)* @foo to void (i32*)*)(i32* %alloca) #0
   ret void
 }
 
-; CHECK: in function crash_call_constexpr_cast{{.*}}: unsupported call to function foo.varargs
+; XCHECK: in function crash_call_constexpr_cast{{.*}}: unsupported call to function foo.varargs
 define amdgpu_kernel void @crash_call_constexpr_cast_varargs() #0 {
   %alloca = alloca i32
   call void bitcast (void (...)* @foo.varargs to void (i32*)*)(i32* %alloca) #0

diff --git a/llvm/test/CodeGen/AMDGPU/promote-alloca-calling-conv.ll b/llvm/test/CodeGen/AMDGPU/promote-alloca-calling-conv.ll
index f5a8b65..515cf81 100644 (file)
--- a/llvm/test/CodeGen/AMDGPU/promote-alloca-calling-conv.ll
+++ b/llvm/test/CodeGen/AMDGPU/promote-alloca-calling-conv.ll
@@ -1,5 +1,5 @@
 ; RUN: opt -S -mtriple=amdgcn-unknown-unknown -amdgpu-promote-alloca < %s | FileCheck -check-prefix=IR %s
-; RUN: llc -march=amdgcn -mcpu=tonga < %s | FileCheck -check-prefix=ASM %s
+; RUN: llc -march=amdgcn -mcpu=fiji < %s | FileCheck -check-prefix=ASM %s
 
 ; IR-LABEL: define amdgpu_vs void @promote_alloca_shaders(i32 addrspace(1)* inreg %out, i32 addrspace(1)* inreg %in) #0 {
 ; IR: alloca [5 x i32]
@@ -74,6 +74,26 @@ entry:
   ret void
 }
 
+declare i32 @foo(i32*) #0
+
+; ASM-LABEL: {{^}}call_private:
+; ASM: buffer_store_dword
+; ASM: buffer_store_dword
+; ASM: s_swappc_b64
+; ASM: ScratchSize: 16396
+define amdgpu_kernel void @call_private(i32 addrspace(1)* %out, i32 %in) #0 {
+entry:
+  %tmp = alloca [2 x i32]
+  %tmp1 = getelementptr [2 x i32], [2 x i32]* %tmp, i32 0, i32 0
+  %tmp2 = getelementptr [2 x i32], [2 x i32]* %tmp, i32 0, i32 1
+  store i32 0, i32* %tmp1
+  store i32 1, i32* %tmp2
+  %tmp3 = getelementptr [2 x i32], [2 x i32]* %tmp, i32 0, i32 %in
+  %val = call i32 @foo(i32* %tmp3)
+  store i32 %val, i32 addrspace(1)* %out
+  ret void
+}
+
 declare i32 @llvm.amdgcn.workitem.id.x() #1
 
 attributes #0 = { nounwind "amdgpu-max-work-group-size"="64" }