When walking backwards to eliminate final stores to allocas at the end of a function...
authorOwen Anderson <resistor@mac.com>
Tue, 30 Aug 2011 18:51:55 +0000 (18:51 +0000)
committerOwen Anderson <resistor@mac.com>
Tue, 30 Aug 2011 18:51:55 +0000 (18:51 +0000)
commit6e4be237b076168167ad21d7cb46a70c17c59f5f
treef6d437eb1614d00723bffb7aecac8462390a79eb
parent9942acab0a42755637a682308c8262b88cbbb9e9
When walking backwards to eliminate final stores to allocas at the end of a function, encountering an unrelated store should not cause us to give up like encountering a load does.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@138809 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Transforms/Scalar/DeadStoreElimination.cpp
test/Transforms/DeadStoreElimination/2011-08-30-EndOfFunction.ll [new file with mode: 0644]