[PM] Port LoadStoreVectorizer to the new pass manager.

[markus]

test/Transforms/LoadStoreVectorizer/X86/correct-order.ll

2 ↗

(On Diff #175099)

Unclear to my why -aa-pipeline=basic-aa is needed with the new PM (otherwise if not used AAResult contains no AAs and the test fails) while with the legacy PM -basicaa was not needed. Actually not clear to me if the two options are to be considered equivalent.

[fedor.sergeev]

LGTM.

test/Transforms/LoadStoreVectorizer/X86/correct-order.ll

2 ↗

(On Diff #175099)

BasicAA is always available for function passes in legacy PM (see AAResultsWrapperPass::runOnFunction).
There is no unconditional AA analyses inserted into AA pipeline in new PM.

[markus]

Thanks for reviewing :)

test/Transforms/LoadStoreVectorizer/X86/correct-order.ll

2 ↗

(On Diff #175099)

In that case maybe we should have a -aa-pipeline=basic-aa option in all the .ll tests (I see that there are a few missing)? Not that it seems to matter for the tests passing but the pass has proven do depend on it for a few of them so might make sense.

[fedor.sergeev]

test/Transforms/LoadStoreVectorizer/X86/correct-order.ll

2 ↗

(On Diff #175099)

I'm not sure about this.
If we need it all the time then perhaps we should implement a pipeline-by-default, similar to legacy pm?
I'll try to ping @chandlerc on that.

[fedor.sergeev]

Btw, @markus, you can go push this thing.

llvm/trunk/include/llvm/InitializePasses.h

 void initializeLintPass(PassRegistry&);
 void initializeLiveDebugValuesPass(PassRegistry&);
 void initializeLiveDebugVariablesPass(PassRegistry&);
 void initializeLiveIntervalsPass(PassRegistry&);
 void initializeLiveRangeShrinkPass(PassRegistry&);
 void initializeLiveRegMatrixPass(PassRegistry&);
 void initializeLiveStacksPass(PassRegistry&);
 void initializeLiveVariablesPass(PassRegistry&);
-void initializeLoadStoreVectorizerPass(PassRegistry&);
+void initializeLoadStoreVectorizerLegacyPassPass(PassRegistry&);
 void initializeLoaderPassPass(PassRegistry&);
 void initializeLocalStackSlotPassPass(PassRegistry&);
 void initializeLocalizerPass(PassRegistry&);
 void initializeLoopAccessLegacyAnalysisPass(PassRegistry&);
 void initializeLoopDataPrefetchLegacyPassPass(PassRegistry&);
 void initializeLoopDeletionLegacyPassPass(PassRegistry&);
 void initializeLoopDistributeLegacyPass(PassRegistry&);
 void initializeLoopExtractorPass(PassRegistry&);

llvm/trunk/include/llvm/Transforms/Vectorize/LoadStoreVectorizer.h

+//===- LoadStoreVectorizer.cpp - GPU Load & Store Vectorizer --------------===//
+//
+//                     The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_TRANSFORMS_VECTORIZE_LOADSTOREVECTORIZER_H
+#define LLVM_TRANSFORMS_VECTORIZE_LOADSTOREVECTORIZER_H
+
+#include "llvm/IR/PassManager.h"
+
+namespace llvm {
+
+class LoadStoreVectorizerPass : public PassInfoMixin<LoadStoreVectorizerPass> {
+public:
+  PreservedAnalyses run(Function &F, FunctionAnalysisManager &AM);
+};
+
+/// Create a legacy pass manager instance of the LoadStoreVectorizer pass
+Pass *createLoadStoreVectorizerPass();
+
+}
+
+#endif /* LLVM_TRANSFORMS_VECTORIZE_LOADSTOREVECTORIZER_H */

llvm/trunk/lib/Passes/PassBuilder.cpp

 #include "llvm/Transforms/Utils/EntryExitInstrumenter.h"
 #include "llvm/Transforms/Utils/LCSSA.h"
 #include "llvm/Transforms/Utils/LibCallsShrinkWrap.h"
 #include "llvm/Transforms/Utils/LoopSimplify.h"
 #include "llvm/Transforms/Utils/LowerInvoke.h"
 #include "llvm/Transforms/Utils/Mem2Reg.h"
 #include "llvm/Transforms/Utils/NameAnonGlobals.h"
 #include "llvm/Transforms/Utils/SymbolRewriter.h"
+#include "llvm/Transforms/Vectorize/LoadStoreVectorizer.h"
 #include "llvm/Transforms/Vectorize/LoopVectorize.h"
 #include "llvm/Transforms/Vectorize/SLPVectorizer.h"
 
 using namespace llvm;
 
 static cl::opt<unsigned> MaxDevirtIterations("pm-max-devirt-iterations",
                                              cl::ReallyHidden, cl::init(4));
 static cl::opt<bool>

llvm/trunk/lib/Passes/PassRegistry.def

 FUNCTION_PASS("float2int", Float2IntPass())
 FUNCTION_PASS("no-op-function", NoOpFunctionPass())
 FUNCTION_PASS("libcalls-shrinkwrap", LibCallsShrinkWrapPass())
 FUNCTION_PASS("loweratomic", LowerAtomicPass())
 FUNCTION_PASS("lower-expect", LowerExpectIntrinsicPass())
 FUNCTION_PASS("lower-guard-intrinsic", LowerGuardIntrinsicPass())
 FUNCTION_PASS("guard-widening", GuardWideningPass())
 FUNCTION_PASS("gvn", GVN())
+FUNCTION_PASS("load-store-vectorizer", LoadStoreVectorizerPass())
 FUNCTION_PASS("loop-simplify", LoopSimplifyPass())
 FUNCTION_PASS("loop-sink", LoopSinkPass())
 FUNCTION_PASS("lowerinvoke", LowerInvokePass())
 FUNCTION_PASS("mem2reg", PromotePass())
 FUNCTION_PASS("memcpyopt", MemCpyOptPass())
 FUNCTION_PASS("mldst-motion", MergedLoadStoreMotionPass())
 FUNCTION_PASS("nary-reassociate", NaryReassociatePass())
 FUNCTION_PASS("newgvn", NewGVNPass())

llvm/trunk/lib/Transforms/Vectorize/LoadStoreVectorizer.cpp

 #include "llvm/IR/Value.h"
 #include "llvm/Pass.h"
 #include "llvm/Support/Casting.h"
 #include "llvm/Support/Debug.h"
 #include "llvm/Support/KnownBits.h"
 #include "llvm/Support/MathExtras.h"
 #include "llvm/Support/raw_ostream.h"
 #include "llvm/Transforms/Vectorize.h"
+#include "llvm/Transforms/Vectorize/LoadStoreVectorizer.h"
 #include <algorithm>
 #include <cassert>
 #include <cstdlib>
 #include <tuple>
 #include <utility>
 
 using namespace llvm;
 
private:
   vectorizeStoreChain(ArrayRef<Instruction *> Chain,
                       SmallPtrSet<Instruction *, 16> *InstructionsProcessed);
 
   /// Check if this load/store access is misaligned accesses.
   bool accessIsMisaligned(unsigned SzInBytes, unsigned AddressSpace,
                           unsigned Alignment);
 };
 
-class LoadStoreVectorizer : public FunctionPass {
+class LoadStoreVectorizerLegacyPass : public FunctionPass {
 public:
   static char ID;
 
-  LoadStoreVectorizer() : FunctionPass(ID) {
-    initializeLoadStoreVectorizerPass(*PassRegistry::getPassRegistry());
+  LoadStoreVectorizerLegacyPass() : FunctionPass(ID) {
+    initializeLoadStoreVectorizerLegacyPassPass(*PassRegistry::getPassRegistry());
   }
 
   bool runOnFunction(Function &F) override;
 
   StringRef getPassName() const override {
     return "GPU Load and Store Vectorizer";
   }
 
   void getAnalysisUsage(AnalysisUsage &AU) const override {
     AU.addRequired<AAResultsWrapperPass>();
     AU.addRequired<ScalarEvolutionWrapperPass>();
     AU.addRequired<DominatorTreeWrapperPass>();
     AU.addRequired<TargetTransformInfoWrapperPass>();
     AU.setPreservesCFG();
   }
 };
 
 } // end anonymous namespace
 
-char LoadStoreVectorizer::ID = 0;
+char LoadStoreVectorizerLegacyPass::ID = 0;
 
-INITIALIZE_PASS_BEGIN(LoadStoreVectorizer, DEBUG_TYPE,
+INITIALIZE_PASS_BEGIN(LoadStoreVectorizerLegacyPass, DEBUG_TYPE,
                       "Vectorize load and Store instructions", false, false)
 INITIALIZE_PASS_DEPENDENCY(SCEVAAWrapperPass)
 INITIALIZE_PASS_DEPENDENCY(DominatorTreeWrapperPass)
 INITIALIZE_PASS_DEPENDENCY(AAResultsWrapperPass)
 INITIALIZE_PASS_DEPENDENCY(GlobalsAAWrapperPass)
 INITIALIZE_PASS_DEPENDENCY(TargetTransformInfoWrapperPass)
-INITIALIZE_PASS_END(LoadStoreVectorizer, DEBUG_TYPE,
+INITIALIZE_PASS_END(LoadStoreVectorizerLegacyPass, DEBUG_TYPE,
                     "Vectorize load and store instructions", false, false)
 
 Pass *llvm::createLoadStoreVectorizerPass() {
-  return new LoadStoreVectorizer();
+  return new LoadStoreVectorizerLegacyPass();
 }
 
-// The real propagateMetadata expects a SmallVector<Value*>, but we deal in
-// vectors of Instructions.
-static void propagateMetadata(Instruction *I, ArrayRef<Instruction *> IL) {
-  SmallVector<Value *, 8> VL(IL.begin(), IL.end());
-  propagateMetadata(I, VL);
-}
-
-bool LoadStoreVectorizer::runOnFunction(Function &F) {
+bool LoadStoreVectorizerLegacyPass::runOnFunction(Function &F) {
   // Don't vectorize when the attribute NoImplicitFloat is used.
   if (skipFunction(F) || F.hasFnAttribute(Attribute::NoImplicitFloat))
     return false;
 
   AliasAnalysis &AA = getAnalysis<AAResultsWrapperPass>().getAAResults();
   DominatorTree &DT = getAnalysis<DominatorTreeWrapperPass>().getDomTree();
   ScalarEvolution &SE = getAnalysis<ScalarEvolutionWrapperPass>().getSE();
   TargetTransformInfo &TTI =
       getAnalysis<TargetTransformInfoWrapperPass>().getTTI(F);
 
   Vectorizer V(F, AA, DT, SE, TTI);
   return V.run();
 }
 
+PreservedAnalyses LoadStoreVectorizerPass::run(Function &F, FunctionAnalysisManager &AM) {
+  // Don't vectorize when the attribute NoImplicitFloat is used.
+  if (F.hasFnAttribute(Attribute::NoImplicitFloat))
+    return PreservedAnalyses::all();
+
+  AliasAnalysis &AA = AM.getResult<AAManager>(F);
+  DominatorTree &DT = AM.getResult<DominatorTreeAnalysis>(F);
+  ScalarEvolution &SE = AM.getResult<ScalarEvolutionAnalysis>(F);
+  TargetTransformInfo &TTI = AM.getResult<TargetIRAnalysis>(F);
+
+  Vectorizer V(F, AA, DT, SE, TTI);
+  bool Changed = V.run();
+  PreservedAnalyses PA;
+  PA.preserveSet<CFGAnalyses>();
+  return Changed ? PA : PreservedAnalyses::all();
+}
+
+// The real propagateMetadata expects a SmallVector<Value*>, but we deal in
+// vectors of Instructions.
+static void propagateMetadata(Instruction *I, ArrayRef<Instruction *> IL) {
+  SmallVector<Value *, 8> VL(IL.begin(), IL.end());
+  propagateMetadata(I, VL);
+}
+
 // Vectorizer Implementation
 bool Vectorizer::run() {
   bool Changed = false;
 
   // Scan the blocks in the function in post order.
   for (BasicBlock *BB : post_order(&F)) {
     InstrListMap LoadRefs, StoreRefs;
     std::tie(LoadRefs, StoreRefs) = collectInstructions(BB);

llvm/trunk/lib/Transforms/Vectorize/Vectorize.cpp

 
 using namespace llvm;
 
 /// initializeVectorizationPasses - Initialize all passes linked into the
 /// Vectorization library.
 void llvm::initializeVectorization(PassRegistry &Registry) {
   initializeLoopVectorizePass(Registry);
   initializeSLPVectorizerPass(Registry);
-  initializeLoadStoreVectorizerPass(Registry);
+  initializeLoadStoreVectorizerLegacyPassPass(Registry);
 }
 
 void LLVMInitializeVectorization(LLVMPassRegistryRef R) {
   initializeVectorization(*unwrap(R));
 }
 
 void LLVMAddLoopVectorizePass(LLVMPassManagerRef PM) {
   unwrap(PM)->add(createLoopVectorizePass());
 }
 
 void LLVMAddSLPVectorizePass(LLVMPassManagerRef PM) {
   unwrap(PM)->add(createSLPVectorizerPass());
 }

llvm/trunk/test/Transforms/LoadStoreVectorizer/AMDGPU/adjust-alloca-alignment.ll

 ; RUN: opt -S -load-store-vectorizer -mattr=-unaligned-buffer-access,+max-private-element-size-16 < %s | FileCheck -check-prefix=ALIGNED -check-prefix=ALL %s
 ; RUN: opt -S -load-store-vectorizer -mattr=+unaligned-buffer-access,+unaligned-scratch-access,+max-private-element-size-16 < %s | FileCheck -check-prefix=UNALIGNED -check-prefix=ALL %s
+; RUN: opt -S -passes='function(load-store-vectorizer)' -mattr=-unaligned-buffer-access,+max-private-element-size-16 < %s | FileCheck -check-prefix=ALIGNED -check-prefix=ALL %s
+; RUN: opt -S -passes='function(load-store-vectorizer)' -mattr=+unaligned-buffer-access,+unaligned-scratch-access,+max-private-element-size-16 < %s | FileCheck -check-prefix=UNALIGNED -check-prefix=ALL %s
 
 target triple = "amdgcn--"
 target datalayout = "e-p:64:64-p1:64:64-p2:32:32-p3:32:32-p4:64:64-p5:32:32-p6:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64-S32-A5"
 
 ; ALL-LABEL: @load_unknown_offset_align1_i8(
 ; ALL: alloca [128 x i8], align 1
 ; UNALIGNED: load <2 x i8>, <2 x i8> addrspace(5)* %{{[0-9]+}}, align 1{{$}}
 

llvm/trunk/test/Transforms/LoadStoreVectorizer/AMDGPU/complex-index.ll

 ; RUN: opt -mtriple=amdgcn-amd-amdhsa -basicaa -load-store-vectorizer -S -o - %s | FileCheck %s
+; RUN: opt -mtriple=amdgcn-amd-amdhsa -aa-pipeline=basic-aa -passes='function(load-store-vectorizer)' -S -o - %s | FileCheck %s
 
 target datalayout = "e-p:64:64-p1:64:64-p2:32:32-p3:32:32-p4:64:64-p5:32:32-p6:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64-S32-A5"
 
 declare i64 @_Z12get_local_idj(i32)
 
 declare i64 @_Z12get_group_idj(i32)
 
 declare double @llvm.fmuladd.f64(double, double, double)
entry:
   %conv25 = fpext float %load2 to double
   %mul26 = fmul double %conv25, 0x3FEAB481D8F35506
   %conv27 = fptrunc double %mul26 to float
   %conv34 = fpext float %conv27 to double
   %storeval2 = tail call double @llvm.fmuladd.f64(double 0x3FF4FFAFBBEC946A, double 0.000000e+00, double %conv34)
   %cstoreval2 = fptrunc double %storeval2 to float
   store float %cstoreval2, float addrspace(1)* %arrayidx24, align 4
   ret void
-}
+}
 

llvm/trunk/test/Transforms/LoadStoreVectorizer/AMDGPU/extended-index.ll

 ; RUN: opt -mtriple=amdgcn-amd-amdhsa -basicaa -load-store-vectorizer -S -o - %s | FileCheck %s
+; RUN: opt -mtriple=amdgcn-amd-amdhsa -aa-pipeline=basic-aa -passes='function(load-store-vectorizer)' -S -o - %s | FileCheck %s
 
 target datalayout = "e-p:64:64-p1:64:64-p2:32:32-p3:32:32-p4:64:64-p5:32:32-p6:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64-S32-A5"
 
 declare i32 @llvm.amdgcn.workitem.id.x() #1
 
 ; CHECK-LABEL: @basic_merge_sext_index(
 ; CHECK: sext i32 %id.x to i64
 ; CHECK: load <2 x float>

llvm/trunk/test/Transforms/LoadStoreVectorizer/AMDGPU/gep-bitcast.ll

 ; RUN: opt -S -mtriple=amdgcn--amdhsa -load-store-vectorizer < %s | FileCheck %s
+; RUN: opt -S -mtriple=amdgcn--amdhsa -passes='function(load-store-vectorizer)' < %s | FileCheck %s
 
 target datalayout = "e-p:64:64-p1:64:64-p2:32:32-p3:32:32-p4:64:64-p5:32:32-p6:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64-S32-A5"
 
 ; Check that vectorizer can find a GEP through bitcast
 ; CHECK-LABEL: @vect_zext_bitcast_f32_to_i32_idx
 ; CHECK: load <4 x i32>
 define void @vect_zext_bitcast_f32_to_i32_idx(float addrspace(1)* %arg1, i32 %base) {
   %add1 = add nuw i32 %base, 0

llvm/trunk/test/Transforms/LoadStoreVectorizer/AMDGPU/insertion-point.ll

 ; RUN: opt -mtriple=amdgcn-amd-amdhsa -basicaa -load-store-vectorizer -S -o - %s | FileCheck %s
+; RUN: opt -mtriple=amdgcn-amd-amdhsa -aa-pipeline=basic-aa -passes='function(load-store-vectorizer)' -S -o - %s | FileCheck %s
 
 target datalayout = "e-p:64:64-p1:64:64-p2:32:32-p3:32:32-p4:64:64-p5:32:32-p6:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64-S32-A5"
 
 ; Check position of the inserted vector load/store.  Vectorized loads should be
 ; inserted at the position of the first load in the chain, and stores should be
 ; inserted at the position of the last store.
 
 ; CHECK-LABEL: @insert_load_point(

llvm/trunk/test/Transforms/LoadStoreVectorizer/AMDGPU/interleaved-mayalias-store.ll

 ; RUN: opt -mtriple=amdgcn-amd-amdhsa -basicaa -load-store-vectorizer -S -o - %s | FileCheck %s
+; RUN: opt -mtriple=amdgcn-amd-amdhsa -aa-pipeline=basic-aa -passes='function(load-store-vectorizer)' -S -o - %s | FileCheck %s
 
 target datalayout = "e-p:64:64-p1:64:64-p2:32:32-p3:32:32-p4:64:64-p5:32:32-p6:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64-S32-A5"
 
 ; This is NOT OK to vectorize, as either load may alias either store.
 
 ; CHECK: load double
 ; CHECK: store double 0.000000e+00, double addrspace(1)* %a,
 ; CHECK: load double

llvm/trunk/test/Transforms/LoadStoreVectorizer/AMDGPU/invariant-load.ll

 ; RUN: opt -mtriple=amdgcn-amd-amdhsa -basicaa -load-store-vectorizer -S -o - %s | FileCheck %s
+; RUN: opt -mtriple=amdgcn-amd-amdhsa -aa-pipeline=basic-aa -passes='function(load-store-vectorizer)' -S -o - %s | FileCheck %s
 
 target datalayout = "e-p:64:64-p1:64:64-p2:32:32-p3:32:32-p4:64:64-p5:32:32-p6:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64-S32-A5"
 
 ; CHECK-LABEL: @interleave
 ; CHECK: load <2 x double>, <2 x double> addrspace(1)* %{{.}}, align 8{{$}}
 ; CHECK: store <2 x double> zeroinitializer
 ; CHECK: store double %add
 define amdgpu_kernel void @interleave(double addrspace(1)* nocapture %a, double addrspace(1)* nocapture %b, double addrspace(1)* nocapture readonly %c) #0 {

llvm/trunk/test/Transforms/LoadStoreVectorizer/X86/codegenprepare-produced-address-math.ll

 ; RUN: opt -codegenprepare -load-store-vectorizer %s -S -o - | FileCheck %s
 ; RUN: opt                 -load-store-vectorizer %s -S -o - | FileCheck %s
+; RUN: opt -codegenprepare -aa-pipeline=basic-aa -passes='function(load-store-vectorizer)' %s -S -o - | FileCheck %s
+; RUN: opt                 -aa-pipeline=basic-aa -passes='function(load-store-vectorizer)' %s -S -o - | FileCheck %s
 
 target triple = "x86_64--"
 
 %union = type { { [4 x [4 x [4 x [16 x float]]]], [4 x [4 x [4 x [16 x float]]]], [10 x [10 x [4 x float]]] } }
 
 @global_pointer = external unnamed_addr global { %union, [2000 x i8] }, align 4
 
 ; Function Attrs: convergent nounwind

llvm/trunk/test/Transforms/LoadStoreVectorizer/X86/compare-scev-by-complexity.ll

 ; RUN: opt -load-store-vectorizer %s -S | FileCheck %s
+; RUN: opt -aa-pipeline=basic-aa -passes='function(load-store-vectorizer)' %s -S | FileCheck %s
 
 ; Check that setting wrapping flags after a SCEV node is created
 ; does not invalidate "sorted by complexity" invariant for
 ; operands of commutative and associative SCEV operators.
 
 target triple = "x86_64--"
 
 @global_value0 = external constant i32

llvm/trunk/test/Transforms/LoadStoreVectorizer/X86/correct-order.ll

 ; RUN: opt -mtriple=x86_64-unknown-linux-gnu -load-store-vectorizer -S -o - %s | FileCheck %s
+; RUN: opt -mtriple=x86_64-unknown-linux-gnu -aa-pipeline=basic-aa -passes='function(load-store-vectorizer)' -S -o - %s | FileCheck %s
 
 target datalayout = "e-m:e-i64:64-i128:128-n32:64-S128"
 
 ; CHECK-LABEL: @correct_order(
 ; CHECK: [[LOAD_PTR:%[0-9]+]] = bitcast i32* %next.gep1
 ; CHECK: load <2 x i32>, <2 x i32>* [[LOAD_PTR]]
 ; CHECK: load i32, i32* %next.gep
 ; CHECK: [[STORE_PTR:%[0-9]+]] = bitcast i32* %next.gep

llvm/trunk/test/Transforms/LoadStoreVectorizer/X86/load-width.ll

 ; RUN: opt -mtriple=x86_64-unknown-linux-gnu -load-store-vectorizer -mcpu haswell -S -o - %s | FileCheck --check-prefix=CHECK-HSW %s
 ; RUN: opt -mtriple=x86_64-unknown-linux-gnu -load-store-vectorizer -mcpu knl -S -o - %s | FileCheck --check-prefix=CHECK-KNL %s
+; RUN: opt -mtriple=x86_64-unknown-linux-gnu -aa-pipeline=basic-aa -passes='function(load-store-vectorizer)' -mcpu haswell -S -o - %s | FileCheck --check-prefix=CHECK-HSW %s
+; RUN: opt -mtriple=x86_64-unknown-linux-gnu -aa-pipeline=basic-aa -passes='function(load-store-vectorizer)' -mcpu knl -S -o - %s | FileCheck --check-prefix=CHECK-KNL %s
 
 define <8 x double> @loadwidth_insert_extract(double* %ptr) {
     %a = bitcast double* %ptr to <2 x double> *
     %b = getelementptr <2 x double>, <2 x double>* %a, i32 1
     %c = getelementptr <2 x double>, <2 x double>* %a, i32 2
     %d = getelementptr <2 x double>, <2 x double>* %a, i32 3
 ; CHECK-HSW: load <4 x double>
 ; CHECK-HSW: load <4 x double>

llvm/trunk/test/Transforms/LoadStoreVectorizer/X86/merge-tbaa.ll

 ; RUN: opt -mtriple=x86_64-unknown-linux-gnu -load-store-vectorizer -S < %s | \
 ; RUN:     FileCheck %s
+; RUN: opt -mtriple=x86_64-unknown-linux-gnu -aa-pipeline=basic-aa -passes='function(load-store-vectorizer)' -S < %s | \
+; RUN:     FileCheck %s
 ;
 ; The GPU Load & Store Vectorizer may merge differently-typed accesses into a
 ; single instruction. This test checks that we merge TBAA tags for such
 ; accesses correctly.
 
 target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
 
 ; struct S {

llvm/trunk/test/Transforms/LoadStoreVectorizer/X86/non-byte-size.ll

 ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
 ; RUN: opt < %s -load-store-vectorizer -S -mtriple=x86_64-unknown-linux-gnu | FileCheck %s
+; RUN: opt < %s -aa-pipeline=basic-aa -passes='function(load-store-vectorizer)' -S -mtriple=x86_64-unknown-linux-gnu | FileCheck %s
 
 %rec = type { i32, i28 }
 
 ; We currently do not optimize this scenario.
 ; But we verify that we no longer crash when compiling this.
 define void @test1(%rec* %out, %rec* %in) {
 ; CHECK-LABEL: @test1(
 ; CHECK-NEXT:    [[IN1:%.*]] = getelementptr [[REC:%.*]], %rec* [[IN:%.*]], i16 0, i32 0

llvm/trunk/test/Transforms/LoadStoreVectorizer/X86/preserve-order32.ll

 ; RUN: opt -mtriple=x86-linux -load-store-vectorizer -S -o - %s | FileCheck %s
+; RUN: opt -mtriple=x86-linux -aa-pipeline=basic-aa -passes='function(load-store-vectorizer)' -S -o - %s | FileCheck %s
 
 target datalayout = "e-p:32:32-p1:64:64-p2:64:64-p3:32:32-p4:64:64-p5:32:32-p24:64:64-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64"
 
 %struct.buffer_t = type { i32, i8* }
 
 ; Check an i32 and i8* get vectorized, and that the two accesses
 ; (load into buff.val and store to buff.p) preserve their order.
 ; Vectorized loads should be inserted at the position of the first load,

llvm/trunk/test/Transforms/LoadStoreVectorizer/X86/preserve-order64.ll

 ; RUN: opt -mtriple=x86_64-unknown-linux-gnu -load-store-vectorizer -S -o - %s | FileCheck %s
+; RUN: opt -mtriple=x86_64-unknown-linux-gnu -aa-pipeline=basic-aa -passes='function(load-store-vectorizer)' -S -o - %s | FileCheck %s
 
 target datalayout = "e-m:e-i64:64-i128:128-n32:64-S128"
 
 %struct.buffer_t = type { i64, i8* }
 %struct.nested.buffer = type { %struct.buffer_t, %struct.buffer_t }
 
 ; Check an i64 and i8* get vectorized, and that the two accesses
 ; (load into buff.val and store to buff.p) preserve their order.

llvm/trunk/test/Transforms/LoadStoreVectorizer/X86/subchain-interleaved.ll

 ; RUN: opt -mtriple=x86_64-unknown-linux-gnu -load-store-vectorizer -S -o - %s | FileCheck %s
+; RUN: opt -mtriple=x86_64-unknown-linux-gnu -aa-pipeline=basic-aa -passes='function(load-store-vectorizer)' -S -o - %s | FileCheck %s
 
 target datalayout = "e-m:e-i64:64-i128:128-n32:64-S128"
 
 ; Vectorized subsets of the load/store chains in the presence of
 ; interleaved loads/stores
 
 ; CHECK-LABEL: @interleave_2L_2S(
 ; CHECK: load <2 x i32>

llvm/trunk/test/Transforms/LoadStoreVectorizer/X86/vector-scalar.ll

 ; RUN: opt -mtriple=x86_64-unknown-linux-gnu -load-store-vectorizer -mcpu haswell -S -o - %s | FileCheck %s
+; RUN: opt -mtriple=x86_64-unknown-linux-gnu -aa-pipeline=basic-aa -passes='function(load-store-vectorizer)' -mcpu haswell -S -o - %s | FileCheck %s
 
 ; Check that the LoadStoreVectorizer does not crash due to not differentiating <1 x T> and T.
 
 ; CHECK-LABEL: @vector_scalar(
 ; CHECK: store double
 ; CHECK: store <1 x double>
 define void @vector_scalar(double* %ptr, double %a, <1 x double> %b) {
   %1 = bitcast double* %ptr to <1 x double>*
   %2 = getelementptr <1 x double>, <1 x double>* %1, i32 1
   store double %a, double* %ptr, align 8
   store <1 x double> %b, <1 x double>* %2, align 8
   ret void
 }

llvm/trunk/test/Transforms/LoadStoreVectorizer/int_sideeffect.ll

 ; RUN: opt -S < %s -load-store-vectorizer | FileCheck %s
+; RUN: opt -S < %s -passes='function(load-store-vectorizer)' | FileCheck %s
 
 declare void @llvm.sideeffect()
 
 ; load-store vectorization across a @llvm.sideeffect.
 
 ; CHECK-LABEL: test
 ; CHECK: load <4 x float>
 ; CHECK: store <4 x float>