[PATCH] D37392: [AMDGPU] Prevent infinite recursion in DAG.computeKnownBits()

[Stanislav Mekhanoshin via Phabricator via llvm-commits]

rampitec created this revision.
Herald added subscribers: t-tye, tpr, dstuttard, yaxunl, nhaehnle, wdng, kzhuravl, arsenm.

https://reviews.llvm.org/D37392

Files:
  lib/Target/AMDGPU/AMDGPUISelLowering.cpp


Index: lib/Target/AMDGPU/AMDGPUISelLowering.cpp
===================================================================
--- lib/Target/AMDGPU/AMDGPUISelLowering.cpp
+++ lib/Target/AMDGPU/AMDGPUISelLowering.cpp
@@ -3886,8 +3886,8 @@
   case AMDGPUISD::MUL_U24:
   case AMDGPUISD::MUL_I24: {
     KnownBits LHSKnown, RHSKnown;
-    DAG.computeKnownBits(Op.getOperand(0), LHSKnown);
-    DAG.computeKnownBits(Op.getOperand(1), RHSKnown);
+    DAG.computeKnownBits(Op.getOperand(0), LHSKnown, Depth + 1);
+    DAG.computeKnownBits(Op.getOperand(1), RHSKnown, Depth + 1);
 
     unsigned TrailZ = LHSKnown.countMinTrailingZeros() +
                       RHSKnown.countMinTrailingZeros();


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D37392.113578.patch
Type: text/x-patch
Size: 680 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20170901/b3b205bd/attachment.bin>