Fix pointer EVT in SelectionDAGBuilder::visitAlloca

[yaxunl]

Summary

SelectionDAGBuilder::visitAlloca assumes alloca address space is 0, which is
incorrect for triple amdgcn---amdgiz and causes isel failure.

This patch fixes that.

[rampitec]

lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp

3526 ↗

(On Diff #123072)

Should not DAG.getIntPtrConstant() take an address space and pass it to the TLI.getPointerTy()?

[yaxunl]

lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp

3526 ↗

(On Diff #123072)

getIntPtrConstant gets a constant with the size of a default pointer. It seems not so useful to let it accept an extra address space parameter except here. However, even here we don't need it since we can just use getConstant with EVT IntPtr.

[rampitec]

LGTM

llvm/trunk/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp

void SelectionDAGBuilder::visitAlloca(const AllocaInst &I) {
   const TargetLowering &TLI = DAG.getTargetLoweringInfo();
   auto &DL = DAG.getDataLayout();
   uint64_t TySize = DL.getTypeAllocSize(Ty);
   unsigned Align =
       std::max((unsigned)DL.getPrefTypeAlignment(Ty), I.getAlignment());
 
   SDValue AllocSize = getValue(I.getArraySize());
 
-  EVT IntPtr = TLI.getPointerTy(DAG.getDataLayout());
+  EVT IntPtr = TLI.getPointerTy(DAG.getDataLayout(), DL.getAllocaAddrSpace());
   if (AllocSize.getValueType() != IntPtr)
     AllocSize = DAG.getZExtOrTrunc(AllocSize, dl, IntPtr);
 
   AllocSize = DAG.getNode(ISD::MUL, dl, IntPtr,
                           AllocSize,
                           DAG.getConstant(TySize, dl, IntPtr));
 
   // Handle alignment.  If the requested alignment is less than or equal to
   // the stack alignment, ignore it.  If the size is greater than or equal to
   // the stack alignment, we note this in the DYNAMIC_STACKALLOC node.
   unsigned StackAlign =
       DAG.getSubtarget().getFrameLowering()->getStackAlignment();
   if (Align <= StackAlign)
     Align = 0;
 
   // Round the size of the allocation up to the stack alignment size
   // by add SA-1 to the size. This doesn't overflow because we're computing
   // an address inside an alloca.
   SDNodeFlags Flags;
   Flags.setNoUnsignedWrap(true);
-  AllocSize = DAG.getNode(ISD::ADD, dl,
-                          AllocSize.getValueType(), AllocSize,
-                          DAG.getIntPtrConstant(StackAlign - 1, dl), Flags);
+  AllocSize = DAG.getNode(ISD::ADD, dl, AllocSize.getValueType(), AllocSize,
+                          DAG.getConstant(StackAlign - 1, dl, IntPtr), Flags);
 
   // Mask out the low bits for alignment purposes.
-  AllocSize = DAG.getNode(ISD::AND, dl,
-                          AllocSize.getValueType(), AllocSize,
-                          DAG.getIntPtrConstant(~(uint64_t)(StackAlign - 1),
-                                                dl));
+  AllocSize =
+      DAG.getNode(ISD::AND, dl, AllocSize.getValueType(), AllocSize,
+                  DAG.getConstant(~(uint64_t)(StackAlign - 1), dl, IntPtr));
 
-  SDValue Ops[] = { getRoot(), AllocSize, DAG.getIntPtrConstant(Align, dl) };
+  SDValue Ops[] = {getRoot(), AllocSize, DAG.getConstant(Align, dl, IntPtr)};
   SDVTList VTs = DAG.getVTList(AllocSize.getValueType(), MVT::Other);
   SDValue DSA = DAG.getNode(ISD::DYNAMIC_STACKALLOC, dl, VTs, Ops);
   setValue(&I, DSA);
   DAG.setRoot(DSA.getValue(1));
 
   assert(FuncInfo.MF->getFrameInfo().hasVarSizedObjects());
 }
 

llvm/trunk/test/CodeGen/AMDGPU/dynamic_stackalloc.ll

-; RUN: not llc -march=amdgcn -mcpu=tahiti -mattr=+promote-alloca -verify-machineinstrs < %s 2>&1 | FileCheck %s
-; RUN: not llc -march=amdgcn -mcpu=tahiti -mattr=-promote-alloca -verify-machineinstrs < %s 2>&1 | FileCheck %s
-; RUN: not llc -march=r600 -mcpu=cypress < %s 2>&1 | FileCheck %s
+; RUN: not llc -march=amdgcn -mtriple=amdgcn---amdgiz -mcpu=tahiti -mattr=+promote-alloca -verify-machineinstrs < %s 2>&1 | FileCheck %s
+; RUN: not llc -march=amdgcn -mtriple=amdgcn---amdgiz -mcpu=tahiti -mattr=-promote-alloca -verify-machineinstrs < %s 2>&1 | FileCheck %s
+; RUN: not llc -march=r600 -mtriple=r600---amdgiz -mcpu=cypress < %s 2>&1 | FileCheck %s
+target datalayout = "A5"
 
 ; CHECK: in function test_dynamic_stackalloc{{.*}}: unsupported dynamic alloca
 
 define amdgpu_kernel void @test_dynamic_stackalloc(i32 addrspace(1)* %out, i32 %n) {
-  %alloca = alloca i32, i32 %n
-  store volatile i32 0, i32* %alloca
+  %alloca = alloca i32, i32 %n, addrspace(5)
+  store volatile i32 0, i32 addrspace(5)* %alloca
   ret void
 }