[CodeGenPrepare] Make -addr-sink-using-gep work with address spaces.

[efriedma]

Summary

When we construct addressing modes, we use isNoopAddrSpaceCast to ignore addrspacecast instructions, so pointers can have a mismatched address space. Make sure we insert the correct addrspacecast when we sink the address computation.

[arsenm]

LGTM

llvm/trunk/lib/CodeGen/CodeGenPrepare.cpp

bool CodeGenPrepare::optimizeMemoryInst(Instruction *MemoryInst, Value *Addr,
   // that we have to sink it into this block.  Check to see if we have already
   // done this for some other load/store instr in this block.  If so, reuse the
   // computation.
   Value *&SunkAddr = SunkAddrs[Addr];
   if (SunkAddr) {
     DEBUG(dbgs() << "CGP: Reusing nonlocal addrmode: " << AddrMode << " for "
                  << *MemoryInst << "\n");
     if (SunkAddr->getType() != Addr->getType())
-      SunkAddr = Builder.CreateBitCast(SunkAddr, Addr->getType());
+      SunkAddr = Builder.CreatePointerCast(SunkAddr, Addr->getType());
   } else if (AddrSinkUsingGEPs ||
              (!AddrSinkUsingGEPs.getNumOccurrences() && TM &&
               SubtargetInfo->useAA())) {
     // By default, we use the GEP-based method when AA is used later. This
     // prevents new inttoptr/ptrtoint pairs from degrading AA capabilities.
     DEBUG(dbgs() << "CGP: SINKING nonlocal addrmode: " << AddrMode << " for "
                  << *MemoryInst << "\n");
     Type *IntPtrTy = DL->getIntPtrType(Addr->getType());
if (!ResultPtr &&
 
       // Add in the Base Offset if present.
       if (AddrMode.BaseOffs) {
         Value *V = ConstantInt::get(IntPtrTy, AddrMode.BaseOffs);
         if (ResultIndex) {
           // We need to add this separately from the scale above to help with
           // SDAG consecutive load/store merging.
           if (ResultPtr->getType() != I8PtrTy)
-            ResultPtr = Builder.CreateBitCast(ResultPtr, I8PtrTy);
+            ResultPtr = Builder.CreatePointerCast(ResultPtr, I8PtrTy);
           ResultPtr = Builder.CreateGEP(I8Ty, ResultPtr, ResultIndex, "sunkaddr");
         }
 
         ResultIndex = V;
       }
 
       if (!ResultIndex) {
         SunkAddr = ResultPtr;
       } else {
         if (ResultPtr->getType() != I8PtrTy)
-          ResultPtr = Builder.CreateBitCast(ResultPtr, I8PtrTy);
+          ResultPtr = Builder.CreatePointerCast(ResultPtr, I8PtrTy);
         SunkAddr = Builder.CreateGEP(I8Ty, ResultPtr, ResultIndex, "sunkaddr");
       }
 
       if (SunkAddr->getType() != Addr->getType())
-        SunkAddr = Builder.CreateBitCast(SunkAddr, Addr->getType());
+        SunkAddr = Builder.CreatePointerCast(SunkAddr, Addr->getType());
     }
   } else {
     DEBUG(dbgs() << "CGP: SINKING nonlocal addrmode: " << AddrMode << " for "
                  << *MemoryInst << "\n");
     Type *IntPtrTy = DL->getIntPtrType(Addr->getType());
     Value *Result = nullptr;
 
     // Start with the base register. Do this first so that subsequent address

llvm/trunk/test/Transforms/CodeGenPrepare/X86/sink-addrspacecast.ll

-; RUN: opt -S -codegenprepare < %s | FileCheck %s
+; RUN: opt -S -codegenprepare < %s | FileCheck %s -check-prefix=CHECK -check-prefix=INT
+; RUN: opt -S -codegenprepare -addr-sink-using-gep=true < %s | FileCheck %s -check-prefix=CHECK -check-prefix=GEP
 
 target datalayout =
 "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128"
 target triple = "x86_64-unknown-linux-gnu"
 
 ; CHECK-LABEL: @load_cast_gep
-; CHECK: add i64 %sunkaddr, 40
+; INT: add i64 %sunkaddr, 40
+; GEP: [[CAST:%[0-9]+]] = addrspacecast i64* %base to i8 addrspace(1)*
+; GEP: getelementptr i8, i8 addrspace(1)* [[CAST]], i64 40
 define void @load_cast_gep(i1 %cond, i64* %base) {
 entry:
   %addr = getelementptr inbounds i64, i64* %base, i64 5
   %casted = addrspacecast i64* %addr to i32 addrspace(1)*
   br i1 %cond, label %if.then, label %fallthrough
 
 if.then:
   %v = load i32, i32 addrspace(1)* %casted, align 4
   br label %fallthrough
 
 fallthrough:
   ret void
 }
 
 ; CHECK-LABEL: @store_gep_cast
-; CHECK: add i64 %sunkaddr, 20
+; INT: add i64 %sunkaddr, 20
+; GEP: [[CAST:%[0-9]+]] = addrspacecast i64* %base to i8 addrspace(1)*
+; GEP: getelementptr i8, i8 addrspace(1)* [[CAST]], i64 20
 define void @store_gep_cast(i1 %cond, i64* %base) {
 entry:
   %casted = addrspacecast i64* %base to i32 addrspace(1)*
   %addr = getelementptr inbounds i32, i32 addrspace(1)* %casted, i64 5
   br i1 %cond, label %if.then, label %fallthrough
 
 if.then:
   store i32 0, i32 addrspace(1)* %addr, align 4
   br label %fallthrough
 
 fallthrough:
   ret void
 }