[RISCV] Match indices based on significant bits when forming strided ops

When checking to see if our index expressions can be converted into strided operations, we previously gave up if the index type wasn't an exact match for the intptrty for the address.  Instead, we can ask known bits how many bits are needed, and proceed as long as the number of required bits fits in intptrty.

Note that the motivation of this is mostly test cleanup.  We canonicalize at IR such that the gep index will match the intptrty.  This is mostly useful so that we can write both RV32 and RV64 tests from the same source.  Its also helpful in preventing confusion - I've stumbled across this at least four times now and wasted time each one.

Interestingly, this does appear to catch some loop cases that O3 does not canonicalize.  I don't think the vectorizer is likely to emit indices narrower that intptr, but if another frontend did, this might pick them up.

This may also pick up some pass ordering problems, but that's a happy accident at best.

Author: preames

llvm/lib/Target/RISCV/RISCVGatherScatterLowering.cpp

@@ -373,14 +373,15 @@ RISCVGatherScatterLowering::determineBaseAndStride(Instruction *Ptr,
   if (!VecOperand)
     return std::make_pair(nullptr, nullptr);
 
-  // We can't extract the stride if the arithmetic is done at a different size
-  // than the pointer type. Adding the stride later may not wrap correctly.
-  // Technically we could handle wider indices, but I don't expect that in
-  // practice.
+  // We need the number of significant bits to match the index type.  IF it
+  // doesn't, then adding the stride later may not wrap correctly.
   Value *VecIndex = Ops[*VecOperand];
   Type *VecIntPtrTy = DL->getIntPtrType(GEP->getType());
-  if (VecIndex->getType() != VecIntPtrTy)
-    return std::make_pair(nullptr, nullptr);
+  if (VecIndex->getType()->getScalarSizeInBits() > VecIntPtrTy->getScalarSizeInBits()) {
+    unsigned MaxBits = ComputeMaxSignificantBits(VecIndex, *DL);
+    if (MaxBits > VecIntPtrTy->getScalarSizeInBits())
+      return std::make_pair(nullptr, nullptr);
+  }
 
   // Handle the non-recursive case.  This is what we see if the vectorizer
   // decides to use a scalar IV + vid on demand instead of a vector IV.
@@ -397,7 +398,8 @@ RISCVGatherScatterLowering::determineBaseAndStride(Instruction *Ptr,
 
     // Convert stride to pointer size if needed.
     Type *IntPtrTy = DL->getIntPtrType(BasePtr->getType());
-    assert(Stride->getType() == IntPtrTy && "Unexpected type");
+    assert(IntPtrTy == VecIntPtrTy->getScalarType());
+    Stride = Builder.CreateSExtOrTrunc(Stride, IntPtrTy);
 
     // Scale the stride by the size of the indexed type.
     if (TypeScale != 1)
@@ -437,7 +439,8 @@ RISCVGatherScatterLowering::determineBaseAndStride(Instruction *Ptr,
 
   // Convert stride to pointer size if needed.
   Type *IntPtrTy = DL->getIntPtrType(BasePtr->getType());
-  assert(Stride->getType() == IntPtrTy && "Unexpected type");
+  assert(IntPtrTy == VecIntPtrTy->getScalarType());
+  Stride = Builder.CreateSExtOrTrunc(Stride, IntPtrTy);
 
   // Scale the stride by the size of the indexed type.
   if (TypeScale != 1)