DAGCombiner: Constant folding around pre-increment loads/stores

Previously, even when a pre-increment load or store was generated,
we often needed to keep a copy of the original base register for use
with other offsets. If all of these offsets are constants (including
the offset which was combined into the addressing mode), then this is
clearly unnecessary. This change adjusts these other offsets to use the
new incremented address.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@174746 91177308-0d34-0410-b5e6-96231b3b80d8
This commit is contained in:
Hal Finkel 2013-02-08 21:35:47 +00:00
parent 0cf5d396c1
commit 089a5f8a8c
2 changed files with 136 additions and 0 deletions

View File

@ -6917,6 +6917,16 @@ bool DAGCombiner::CombineToPreIndexedLoadStore(SDNode *N) {
ISD::MemIndexedMode AM = ISD::UNINDEXED;
if (!TLI.getPreIndexedAddressParts(N, BasePtr, Offset, AM, DAG))
return false;
// Backends without true r+i pre-indexed forms may need to pass a
// constant base with a variable offset so that constant coercion
// will work with the patterns in canonical form.
bool Swapped = false;
if (isa<ConstantSDNode>(BasePtr)) {
std::swap(BasePtr, Offset);
Swapped = true;
}
// Don't create a indexed load / store with zero offset.
if (isa<ConstantSDNode>(Offset) &&
cast<ConstantSDNode>(Offset)->isNullValue())
@ -6942,6 +6952,48 @@ bool DAGCombiner::CombineToPreIndexedLoadStore(SDNode *N) {
return false;
}
// If the offset is a constant, there may be other adds of constants that
// can be folded with this one. We should do this to avoid having to keep
// a copy of the original base pointer.
SmallVector<SDNode *, 16> OtherUses;
if (isa<ConstantSDNode>(Offset))
for (SDNode::use_iterator I = BasePtr.getNode()->use_begin(),
E = BasePtr.getNode()->use_end(); I != E; ++I) {
SDNode *Use = *I;
if (Use == Ptr.getNode())
continue;
if (Use->isPredecessorOf(N))
continue;
if (Use->getOpcode() != ISD::ADD && Use->getOpcode() != ISD::SUB) {
OtherUses.clear();
break;
}
SDValue Op0 = Use->getOperand(0), Op1 = Use->getOperand(1);
if (Op1.getNode() == BasePtr.getNode())
std::swap(Op0, Op1);
assert(Op0.getNode() == BasePtr.getNode() &&
"Use of ADD/SUB but not an operand");
if (!isa<ConstantSDNode>(Op1)) {
OtherUses.clear();
break;
}
// FIXME: In some cases, we can be smarter about this.
if (Op1.getValueType() != Offset.getValueType()) {
OtherUses.clear();
break;
}
OtherUses.push_back(Use);
}
if (Swapped)
std::swap(BasePtr, Offset);
// Now check for #3 and #4.
bool RealUse = false;
@ -6991,6 +7043,43 @@ bool DAGCombiner::CombineToPreIndexedLoadStore(SDNode *N) {
// Finally, since the node is now dead, remove it from the graph.
DAG.DeleteNode(N);
if (Swapped)
std::swap(BasePtr, Offset);
// Replace other uses of BasePtr that can be updated to use Ptr
for (unsigned i = 0, e = OtherUses.size(); i != e; ++i) {
unsigned OffsetIdx = 1;
if (OtherUses[i]->getOperand(OffsetIdx).getNode() == BasePtr.getNode())
OffsetIdx = 0;
assert(OtherUses[i]->getOperand(!OffsetIdx).getNode() ==
BasePtr.getNode() && "Expected BasePtr operand");
APInt OV =
cast<ConstantSDNode>(Offset)->getAPIntValue();
if (AM == ISD::PRE_DEC)
OV = -OV;
ConstantSDNode *CN =
cast<ConstantSDNode>(OtherUses[i]->getOperand(OffsetIdx));
APInt CNV = CN->getAPIntValue();
if (OtherUses[i]->getOpcode() == ISD::SUB && OffsetIdx == 1)
CNV += OV;
else
CNV -= OV;
SDValue NewOp1 = Result.getValue(isLoad ? 1 : 0);
SDValue NewOp2 = DAG.getConstant(CNV, CN->getValueType(0));
if (OffsetIdx == 0)
std::swap(NewOp1, NewOp2);
SDValue NewUse = DAG.getNode(OtherUses[i]->getOpcode(),
OtherUses[i]->getDebugLoc(),
OtherUses[i]->getValueType(0), NewOp1, NewOp2);
DAG.ReplaceAllUsesOfValueWith(SDValue(OtherUses[i], 0), NewUse);
removeFromWorkList(OtherUses[i]);
DAG.DeleteNode(OtherUses[i]);
}
// Replace the uses of Ptr with uses of the updated base value.
DAG.ReplaceAllUsesOfValueWith(Ptr, Result.getValue(isLoad ? 1 : 0));
removeFromWorkList(Ptr.getNode());

View File

@ -0,0 +1,47 @@
; RUN: llc -mcpu=a2 -disable-lsr < %s | FileCheck %s
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-v128:128:128-n32:64"
target triple = "powerpc64-unknown-linux-gnu"
define i32 @test1(i64 %add, i64* %ptr) nounwind {
entry:
%p1 = getelementptr i64* %ptr, i64 144115188075855
br label %for.cond2.preheader
for.cond2.preheader:
%nl.018 = phi i32 [ 0, %entry ], [ %inc9, %for.end ]
br label %for.body4
for.body4:
%lsr.iv = phi i32 [ %lsr.iv.next, %for.body4 ], [ 16000, %for.cond2.preheader ]
%i0 = phi i64* [ %p1, %for.cond2.preheader ], [ %i6, %for.body4 ]
%i6 = getelementptr i64* %i0, i64 400000
%i7 = getelementptr i64* %i6, i64 300000
%i8 = getelementptr i64* %i6, i64 200000
%i9 = getelementptr i64* %i6, i64 100000
store i64 %add, i64* %i6, align 32
store i64 %add, i64* %i7, align 32
store i64 %add, i64* %i8, align 32
store i64 %add, i64* %i9, align 32
%lsr.iv.next = add i32 %lsr.iv, -16
%exitcond.15 = icmp eq i32 %lsr.iv.next, 0
br i1 %exitcond.15, label %for.end, label %for.body4
; Make sure that we generate the most compact form of this loop with no
; unnecessary moves
; CHECK: @test1
; CHECK: mtctr
; CHECK: stdux
; CHECK-NEXT: stdx
; CHECK-NEXT: stdx
; CHECK-NEXT: stdx
; CHECK-NEXT: bdnz
for.end:
%inc9 = add nsw i32 %nl.018, 1
%exitcond = icmp eq i32 %inc9, 400000
br i1 %exitcond, label %for.end10, label %for.cond2.preheader
for.end10:
ret i32 0
}