Fix a bug where we were useing HA to get the high part, which seems like it
authorChris Lattner <sabre@nondot.org>
Thu, 1 Sep 2005 19:38:28 +0000 (19:38 +0000)
committerChris Lattner <sabre@nondot.org>
Thu, 1 Sep 2005 19:38:28 +0000 (19:38 +0000)
commit393ecd6d2dbc844ae9e34fea34f15d914ec7c43c
tree2e92e61fcd24475a22325543d639c7aacb9067fa
parent50ff55c2c7909818165e51bbdc103a05b143dcb1
Fix a bug where we were useing HA to get the high part, which seems like it
could cause a miscompile.  Fixing this didn't fix the two programs that fail
though.  :(

This also changes the implementation to follow the pattern selector more
closely, causing us to select 0 to li instead of lis.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@23189 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/PowerPC/PPCISelDAGToDAG.cpp