From: Simon Pilgrim <llvm-dev@redking.me.uk>
Date: Sat, 27 Jul 2019 12:23:36 +0000 (+0000)
Subject: [TargetLowering] Add depth limit to SimplifyMultipleUseDemandedBits
X-Git-Url: https://granicus.if.org/sourcecode?a=commitdiff_plain;h=5972aca6bec62c139a364ad0b672f0005807125e;p=llvm

[TargetLowering] Add depth limit to SimplifyMultipleUseDemandedBits

We're getting reports of massive compile time increases because SimplifyMultipleUseDemandedBits was losing track of the depth and not earlying-out. No repro yet, but consider this a pre-emptive commit.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@367169 91177308-0d34-0410-b5e6-96231b3b80d8
---

diff --git a/lib/CodeGen/SelectionDAG/TargetLowering.cpp b/lib/CodeGen/SelectionDAG/TargetLowering.cpp
index 6bfb388a3f7..1173beb9ac6 100644
--- a/lib/CodeGen/SelectionDAG/TargetLowering.cpp
+++ b/lib/CodeGen/SelectionDAG/TargetLowering.cpp
@@ -569,6 +569,9 @@ bool TargetLowering::SimplifyDemandedBits(SDValue Op, const APInt &DemandedBits,
 SDValue TargetLowering::SimplifyMultipleUseDemandedBits(
     SDValue Op, const APInt &DemandedBits, const APInt &DemandedElts,
     SelectionDAG &DAG, unsigned Depth) const {
+  if (Depth == 6) // Limit search depth.
+    return SDValue();
+
   KnownBits LHSKnown, RHSKnown;
   switch (Op.getOpcode()) {
   case ISD::AND: {