diff options
author | 2016-01-20 17:00:19 +0000 | |
---|---|---|
committer | 2016-01-20 18:14:30 +0000 | |
commit | c047d94f02334b419ce532c3bf7da3e91e69dae1 (patch) | |
tree | c9213b1efcc9062a46a18a1f0fec6a984e6a5565 /compiler/optimizing/ssa_builder.cc | |
parent | 440ef2cbc76d7e2bc76cf5f15b69fc0478d7e853 (diff) |
ART: DCHECK still too strong for String NewInstance
Even though the presence of a irreducible loop phi is the reason why
String.<init> may see its first input not being the NewInstance, it
may see a different phi from the same cycle. We have to relax the
corresponding DCHECK.
Bug: 26676472
Change-Id: Iba6fffa74833c57aad8aa07dc382ff6b93838de1
Diffstat (limited to 'compiler/optimizing/ssa_builder.cc')
-rw-r--r-- | compiler/optimizing/ssa_builder.cc | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/compiler/optimizing/ssa_builder.cc b/compiler/optimizing/ssa_builder.cc index c8244aa800..165d09d1a5 100644 --- a/compiler/optimizing/ssa_builder.cc +++ b/compiler/optimizing/ssa_builder.cc @@ -934,7 +934,7 @@ void SsaBuilder::VisitInvokeStaticOrDirect(HInvokeStaticOrDirect* invoke) { if (arg_this->IsNewInstance()) { uninitialized_strings_.push_back(arg_this->AsNewInstance()); } else { - DCHECK(arg_this->IsIrreducibleLoopHeaderPhi()); + DCHECK(arg_this->IsPhi()); // NewInstance is not the direct input of the StringFactory call. It might // be redundant but optimizing this case is not worth the effort. } |