[LSR] Add a flag that overrides the target's preferred addressing mode

[SjoerdMeijer]

Summary

This adds flag -lsr-preferred-addressing-mode to override the target's preferred addressing mode. It replaces flag -lsr-backedge-indexing, which is equivalent to preindexed addressing that is one of the options that -lsr-preferred-addressing-mode accepts.

[dmgreen]

Thank for putting this together.

llvm/lib/Transforms/Scalar/LoopStrengthReduce.cpp

5572

Could this use PreferredAddresingMode.getNumOccurrences() > 0 to allwo overriding to None, even if the target returns Pre/PostIndexed?

[SjoerdMeijer]

Thanks Dave, fixed that.

[dmgreen]

Thanks. LGTM

llvm/lib/Transforms/Scalar/LoopStrengthReduce.cpp

 
 // Flag to narrow search space by filtering non-optimal formulae with
 // the same ScaledReg and Scale.
 static cl::opt<bool> FilterSameScaledReg(
     "lsr-filter-same-scaled-reg", cl::Hidden, cl::init(true),
     cl::desc("Narrow LSR search space by filtering non-optimal formulae"
              " with the same ScaledReg and Scale"));
 
-static cl::opt<bool> EnableBackedgeIndexing(
-    "lsr-backedge-indexing", cl::Hidden, cl::init(true),
-      cl::desc("Enable the generation of cross iteration indexed memops"));
+static cl::opt<TTI::AddressingModeKind> PreferredAddresingMode(
+  "lsr-preferred-addressing-mode", cl::Hidden, cl::init(TTI::AMK_None),
+   cl::desc("A flag that overrides the target's preferred addressing mode."),
+   cl::values(clEnumValN(TTI::AMK_None,
+                         "none",
+                         "Don't prefer any addressing mode"),
+              clEnumValN(TTI::AMK_PreIndexed,
+                         "preindexed",
+                         "Prefer pre-indexed addressing mode"),
+              clEnumValN(TTI::AMK_PostIndexed,
+                         "postindexed",
+                         "Prefer post-indexed addressing mode")));
 
 static cl::opt<unsigned> ComplexityLimit(
   "lsr-complexity-limit", cl::Hidden,
   cl::init(std::numeric_limits<uint16_t>::max()),
   cl::desc("LSR search space complexity limit"));
 
 static cl::opt<unsigned> SetupCostDepthLimit(
     "lsr-setupcost-depth-limit", cl::Hidden, cl::init(7),
void LSRInstance::GenerateConstantOffsetsImpl(
   // With constant offsets and constant steps, we can generate pre-inc
   // accesses by having the offset equal the step. So, for access #0 with a
   // step of 8, we generate a G - 8 base which would require the first access
   // to be ((G - 8) + 8),+,8. The pre-indexed access then updates the pointer
   // for itself and hopefully becomes the base for other accesses. This means
   // means that a single pre-indexed access can be generated to become the new
   // base pointer for each iteration of the loop, resulting in no extra add/sub
   // instructions for pointer updating.
-  bool FavorPreIndexed = EnableBackedgeIndexing &&
-          AMK == TTI::AMK_PreIndexed;
-  if (FavorPreIndexed && LU.Kind == LSRUse::Address) {
+  if (AMK == TTI::AMK_PreIndexed && LU.Kind == LSRUse::Address) {
     if (auto *GAR = dyn_cast<SCEVAddRecExpr>(G)) {
       if (auto *StepRec =
           dyn_cast<SCEVConstant>(GAR->getStepRecurrence(SE))) {
         const APInt &StepInt = StepRec->getAPInt();
         int64_t Step = StepInt.isNegative() ?
           StepInt.getSExtValue() : StepInt.getZExtValue();
 
         for (int64_t Offset : Worklist) {
Changed |= RecursivelyDeleteTriviallyDeadInstructionsPermissive(DeadInsts,
                                                                   &TLI, MSSAU);
 }
 
 LSRInstance::LSRInstance(Loop *L, IVUsers &IU, ScalarEvolution &SE,
                          DominatorTree &DT, LoopInfo &LI,
                          const TargetTransformInfo &TTI, AssumptionCache &AC,
                          TargetLibraryInfo &TLI, MemorySSAUpdater *MSSAU)
     : IU(IU), SE(SE), DT(DT), LI(LI), AC(AC), TLI(TLI), TTI(TTI), L(L),
-      MSSAU(MSSAU), AMK(TTI.getPreferredAddressingMode(L, &SE)) {
+      MSSAU(MSSAU), AMK(PreferredAddresingMode.getNumOccurrences() > 0 ?
+        PreferredAddresingMode : TTI.getPreferredAddressingMode(L, &SE)) {
   // If LoopSimplify form is not available, stay out of trouble.
   if (!L->isLoopSimplifyForm())
     return;
 
   // If there's no interesting work to be done, bail early.
   if (IU.empty()) return;
 
   // If there's too much analysis to be done, bail early. We won't be able to

llvm/test/CodeGen/ARM/loop-indexing.ll

-; RUN: llc --mtriple=thumbv7em -mattr=+fp-armv8 -O3 %s -o - | FileCheck %s --check-prefix=CHECK --check-prefix=CHECK-DEFAULT --check-prefix=CHECK-T2
-; RUN: llc -mtriple=thumbv8m.main -mattr=+fp-armv8,+dsp -O3 %s -o - | FileCheck %s --check-prefix=CHECK --check-prefix=CHECK-DEFAULT --check-prefix=CHECK-T2
-; RUN: llc -mtriple=thumbv8m.main -mattr=+fp-armv8,+dsp -lsr-backedge-indexing=false %s -o - | FileCheck %s --check-prefix=CHECK --check-prefix=DISABLED
+; RUN: llc --mtriple=thumbv7em -mattr=+fp-armv8 -O3 %s -o - | \
+; RUN:    FileCheck %s --check-prefix=CHECK --check-prefix=CHECK-DEFAULT --check-prefix=CHECK-T2
+
+; RUN: llc --mtriple=thumbv7em -mattr=+fp-armv8 -O3 -lsr-preferred-addressing-mode=none %s -o - | \
+; RUN:    FileCheck %s --check-prefix=CHECK --check-prefix=DISABLED
+
+; RUN: llc -mtriple=thumbv8m.main -mattr=+fp-armv8,+dsp -O3 %s -o - | \
+; RUN:    FileCheck %s --check-prefix=CHECK --check-prefix=CHECK-DEFAULT --check-prefix=CHECK-T2
+
+; RUN: llc -mtriple=thumbv8m.main -mattr=+fp-armv8,+dsp -lsr-preferred-addressing-mode=postindexed %s -o - | \
+; RUN:    FileCheck %s --check-prefix=CHECK --check-prefix=DISABLED
+
+; RUN: llc -mtriple=thumbv8m.main -mattr=+fp-armv8,+dsp -lsr-preferred-addressing-mode=preindexed %s -o - | \
+; RUN:    FileCheck %s --check-prefixes=CHECK,CHECK-T2
+
 ; RUN: llc -mtriple=thumbv8m.base %s -o - | FileCheck %s --check-prefix=CHECK --check-prefix=DISABLED
 ; RUN: llc -mtriple=thumbv8 %s -o - | FileCheck %s --check-prefix=CHECK --check-prefix=DISABLED
-; RUN: llc -mtriple=thumbv8m.main -mattr=+fp-armv8,+dsp -O3 -lsr-complexity-limit=2147483647 %s -o - | FileCheck %s --check-prefix=CHECK --check-prefix=CHECK-COMPLEX --check-prefix=CHECK-T2
+
+; RUN: llc -mtriple=thumbv8m.main -mattr=+fp-armv8,+dsp -O3 -lsr-complexity-limit=2147483647 %s -o - | \
+; RUN:    FileCheck %s --check-prefix=CHECK --check-prefix=CHECK-COMPLEX --check-prefix=CHECK-T2
 
 ; Tests to check that post increment addressing modes are used instead of
 ; updating base pointers with add instructions.
 
 ; TODO: I think we should be able to use post inc addressing with VLDM
 ; instructions.
 ; CHECK-LABEL: test_fma
 ; CHECK: @ %loop