Fix a devilish miscompile exposed by block placement. The

updateTerminator code didn't correctly handle EH terminators in one very
specific case. AnalyzeBranch would find no terminator instruction, and
so the fallback in updateTerminator is to assume fallthrough. This is
correct, but the destination of the fallthrough was assumed to be the
first successor.

This is *almost always* true, but in certain cases the loop
transformations will cause the landing pad to be the first successor!
Instead of this brittle logic, actually look through the successors for
a non-landing-pad accessor, and to assert if more than one is found.

This will hopefully fix some (if not all) of the self host miscompiles
with block placement. Thanks to Benjamin Kramer for reporting, Nick
Lewycky for an initial stab at a reduction, and Duncan for endless
advice on EH (which I know nothing about) as well as reviewing the
actual fix.

llvm-svn: 145062

1 files changed, 8 insertions, 2 deletions

diff --git a/llvm/lib/CodeGen/MachineBasicBlock.cpp b/llvm/lib/CodeGen/MachineBasicBlock.cpp
index 4c5fe4c480a..f05b6c1eb97 100644
--- a/llvm/lib/CodeGen/MachineBasicBlock.cpp
+++ b/llvm/lib/CodeGen/MachineBasicBlock.cpp
@@ -297,8 +297,14 @@ void MachineBasicBlock::updateTerminator() {
         TII->RemoveBranch(*this);
     } else {
       // The block has an unconditional fallthrough. If its successor is not
-      // its layout successor, insert a branch.
-      TBB = *succ_begin();
+      // its layout successor, insert a branch. First we have to locate the
+      // only non-landing-pad successor, as that is the fallthrough block.
+      for (succ_iterator SI = succ_begin(), SE = succ_end(); SI != SE; ++SI) {
+        if ((*SI)->isLandingPad())
+          continue;
+        assert(!TBB && "Found more than one non-landing-pad successor!");
+        TBB = *SI;
+      }
       if (!isLayoutSuccessor(TBB))
         TII->InsertBranch(*this, TBB, 0, Cond, dl);
     }