[PATCH] D18065: Fix bugs in the MemorySSA walker

[Daniel Berlin via llvm-commits]

dberlin added a comment.

Just because this logic is getting fairly complicated, can you explain to me (either here or on a whiteboard) the failure mode that is occurring that you are solving?

It's clear the testcase gives the wrong answer. Can you describe more of "why"?


================
Comment at: lib/Transforms/Utils/MemorySSA.cpp:911
@@ +910,3 @@
+    // This loop tries to detect these cases.
+    auto MPI = upward_defs_begin(PHIPair), MPE = upward_defs_end();
+    for (; MPI != MPE; ++MPI) {
----------------
why did you move this out of the for loop init?


http://reviews.llvm.org/D18065