[AArch64][LoopIdiom] Generalize AArch64LoopIdiomTransform into LoopIdiomVectorize (#94081)

mshockwave · web-flow · commit 37e309f16354 · 2024-06-07T14:06:11.000-07:00
To facilitate sharing LoopIdiomTransform between AArch64 and RISC-V, this first patch moves AArch64LoopIdiomTransform from lib/Target/AArch64 to lib/Transforms/Vectorize and renames it to LoopIdiomVectorize. The following patch (#94082) will teach LoopIdiomVectorize how to generate VP intrinsics (in addition to the current masked vector style) in favor of RVV.
diff --git a/llvm/include/llvm/Transforms/Vectorize/LoopIdiomVectorize.h b/llvm/include/llvm/Transforms/Vectorize/LoopIdiomVectorize.h
@@ -1,25 +1,21 @@
-//===- AArch64LoopIdiomTransform.h --------------------------------------===//
+//===----------LoopIdiomVectorize.h -----------------------------*- C++ -*-===//
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 //
 //===----------------------------------------------------------------------===//
 
-#ifndef LLVM_LIB_TARGET_AARCH64_AARCH64LOOPIDIOMTRANSFORM_H
-#define LLVM_LIB_TARGET_AARCH64_AARCH64LOOPIDIOMTRANSFORM_H
+#ifndef LLVM_LIB_TRANSFORMS_VECTORIZE_LOOPIDIOMVECTORIZE_H
+#define LLVM_LIB_TRANSFORMS_VECTORIZE_LOOPIDIOMVECTORIZE_H
 
 #include "llvm/IR/PassManager.h"
 #include "llvm/Transforms/Scalar/LoopPassManager.h"
 
 namespace llvm {
-
-struct AArch64LoopIdiomTransformPass
-    : PassInfoMixin<AArch64LoopIdiomTransformPass> {
+struct LoopIdiomVectorizePass : PassInfoMixin<LoopIdiomVectorizePass> {
   PreservedAnalyses run(Loop &L, LoopAnalysisManager &AM,
                         LoopStandardAnalysisResults &AR, LPMUpdater &U);
 };
-
 } // namespace llvm
-
-#endif // LLVM_LIB_TARGET_AARCH64_AARCH64LOOPIDIOMTRANSFORM_H
+#endif // LLVM_LIB_TRANSFORMS_VECTORIZE_LOOPIDIOMVECTORIZE_H
diff --git a/llvm/lib/Passes/PassBuilder.cpp b/llvm/lib/Passes/PassBuilder.cpp
@@ -299,6 +299,7 @@
 #include "llvm/Transforms/Utils/UnifyFunctionExitNodes.h"
 #include "llvm/Transforms/Utils/UnifyLoopExits.h"
 #include "llvm/Transforms/Vectorize/LoadStoreVectorizer.h"
+#include "llvm/Transforms/Vectorize/LoopIdiomVectorize.h"
 #include "llvm/Transforms/Vectorize/LoopVectorize.h"
 #include "llvm/Transforms/Vectorize/SLPVectorizer.h"
 #include "llvm/Transforms/Vectorize/VectorCombine.h"
diff --git a/llvm/lib/Passes/PassRegistry.def b/llvm/lib/Passes/PassRegistry.def
@@ -622,6 +622,7 @@ LOOP_PASS("invalidate<all>", InvalidateAllAnalysesPass())
 LOOP_PASS("loop-bound-split", LoopBoundSplitPass())
 LOOP_PASS("loop-deletion", LoopDeletionPass())
 LOOP_PASS("loop-idiom", LoopIdiomRecognizePass())
+LOOP_PASS("loop-idiom-vectorize", LoopIdiomVectorizePass())
 LOOP_PASS("loop-instsimplify", LoopInstSimplifyPass())
 LOOP_PASS("loop-predication", LoopPredicationPass())
 LOOP_PASS("loop-reduce", LoopStrengthReducePass())
diff --git a/llvm/lib/Target/AArch64/AArch64.h b/llvm/lib/Target/AArch64/AArch64.h
@@ -90,7 +90,6 @@ void initializeAArch64DeadRegisterDefinitionsPass(PassRegistry&);
 void initializeAArch64ExpandPseudoPass(PassRegistry &);
 void initializeAArch64GlobalsTaggingPass(PassRegistry &);
 void initializeAArch64LoadStoreOptPass(PassRegistry&);
-void initializeAArch64LoopIdiomTransformLegacyPassPass(PassRegistry &);
 void initializeAArch64LowerHomogeneousPrologEpilogPass(PassRegistry &);
 void initializeAArch64MIPeepholeOptPass(PassRegistry &);
 void initializeAArch64O0PreLegalizerCombinerPass(PassRegistry &);
diff --git a/llvm/lib/Target/AArch64/AArch64PassRegistry.def b/llvm/lib/Target/AArch64/AArch64PassRegistry.def
diff --git a/llvm/lib/Target/AArch64/AArch64TargetMachine.cpp b/llvm/lib/Target/AArch64/AArch64TargetMachine.cpp
@@ -11,7 +11,6 @@
 
 #include "AArch64TargetMachine.h"
 #include "AArch64.h"
-#include "AArch64LoopIdiomTransform.h"
 #include "AArch64MachineFunctionInfo.h"
 #include "AArch64MachineScheduler.h"
 #include "AArch64MacroFusion.h"
@@ -52,6 +51,7 @@
 #include "llvm/TargetParser/Triple.h"
 #include "llvm/Transforms/CFGuard.h"
 #include "llvm/Transforms/Scalar.h"
+#include "llvm/Transforms/Vectorize/LoopIdiomVectorize.h"
 #include <memory>
 #include <optional>
 #include <string>
@@ -234,7 +234,6 @@ extern "C" LLVM_EXTERNAL_VISIBILITY void LLVMInitializeAArch64Target() {
   initializeAArch64DeadRegisterDefinitionsPass(*PR);
   initializeAArch64ExpandPseudoPass(*PR);
   initializeAArch64LoadStoreOptPass(*PR);
-  initializeAArch64LoopIdiomTransformLegacyPassPass(*PR);
   initializeAArch64MIPeepholeOptPass(*PR);
   initializeAArch64SIMDInstrOptPass(*PR);
   initializeAArch64O0PreLegalizerCombinerPass(*PR);
@@ -553,12 +552,9 @@ class AArch64PassConfig : public TargetPassConfig {
 void AArch64TargetMachine::registerPassBuilderCallbacks(
     PassBuilder &PB, bool PopulateClassToPassNames) {
 
-#define GET_PASS_REGISTRY "AArch64PassRegistry.def"
-#include "llvm/Passes/TargetPassRegistry.inc"
-
   PB.registerLateLoopOptimizationsEPCallback(
       [=](LoopPassManager &LPM, OptimizationLevel Level) {
-        LPM.addPass(AArch64LoopIdiomTransformPass());
+        LPM.addPass(LoopIdiomVectorizePass());
       });
 }
 
diff --git a/llvm/lib/Target/AArch64/AArch64TargetMachine.h b/llvm/lib/Target/AArch64/AArch64TargetMachine.h
@@ -14,7 +14,6 @@
 #define LLVM_LIB_TARGET_AARCH64_AARCH64TARGETMACHINE_H
 
 #include "AArch64InstrInfo.h"
-#include "AArch64LoopIdiomTransform.h"
 #include "AArch64Subtarget.h"
 #include "llvm/IR/DataLayout.h"
 #include "llvm/Target/TargetMachine.h"
diff --git a/llvm/lib/Target/AArch64/CMakeLists.txt b/llvm/lib/Target/AArch64/CMakeLists.txt
@@ -65,7 +65,6 @@ add_llvm_target(AArch64CodeGen
   AArch64ISelLowering.cpp
   AArch64InstrInfo.cpp
   AArch64LoadStoreOptimizer.cpp
-  AArch64LoopIdiomTransform.cpp
   AArch64LowerHomogeneousPrologEpilog.cpp
   AArch64MachineFunctionInfo.cpp
   AArch64MachineScheduler.cpp
@@ -112,6 +111,7 @@ add_llvm_target(AArch64CodeGen
   Target
   TargetParser
   TransformUtils
+  Vectorize
 
   ADD_TO_COMPONENT
   AArch64
diff --git a/llvm/lib/Transforms/Vectorize/CMakeLists.txt b/llvm/lib/Transforms/Vectorize/CMakeLists.txt
@@ -1,5 +1,6 @@
 add_llvm_component_library(LLVMVectorize
   LoadStoreVectorizer.cpp
+  LoopIdiomVectorize.cpp
   LoopVectorizationLegality.cpp
   LoopVectorize.cpp
   SLPVectorizer.cpp
diff --git a/llvm/lib/Transforms/Vectorize/LoopIdiomVectorize.cpp b/llvm/lib/Transforms/Vectorize/LoopIdiomVectorize.cpp
diff --git a/llvm/test/Transforms/LoopIdiom/AArch64/byte-compare-index.ll b/llvm/test/Transforms/LoopIdiom/AArch64/byte-compare-index.ll
