Allow 0-weight branches in BranchProbabilityInfo.

Summary:
When computing branch weights in BPI, we used to disallow branches with
weight 0. This is a minor nuisance, because a branch with weight 0 is
different to "don't have information". In the context of
instrumentation, it may mean "never executed", in the context of
sampling, it means "never or seldom executed".

In allowing 0 weight branches, I ran into issues with the switch
expansion code in selection DAG. It is currently hardwired to not handle
branches with weight 0. To maintain the current behaviour, I changed it
to use 1 when it finds 0, but perhaps the algorithm needs changes to
tolerate branches with weight zero.

Reviewers: hansw

Subscribers: llvm-commits

Differential Revision: http://reviews.llvm.org/D9533

llvm-svn: 236617

1 files changed, 5 insertions, 1 deletions

diff --git a/llvm/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp b/llvm/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp
index d0bda11eb05..fe96f33d772 100644
--- a/llvm/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp
@@ -8009,7 +8009,11 @@ void SelectionDAGBuilder::visitSwitch(const SwitchInst &SI) {
     const ConstantInt *CaseVal = I.getCaseValue();
     uint32_t Weight = 1;
     if (BPI) {
-      Weight = BPI->getEdgeWeight(SI.getParent(), I.getSuccessorIndex());
+      // TODO - BPI used to guarantee non-zero weights, but this produces
+      // information loss (see PR 22718). Since we can't handle zero weights
+      // here, use the same flooring mechanism previously used by BPI.
+      Weight = std::max(
+          1u, BPI->getEdgeWeight(SI.getParent(), I.getSuccessorIndex()));
       assert(Weight <= UINT32_MAX / SI.getNumSuccessors());
     }
     Clusters.push_back(CaseCluster::range(CaseVal, CaseVal, Succ, Weight));