[x86] Sink the single-input v8i16 lowering code that is actually
authorChandler Carruth <chandlerc@gmail.com>
Thu, 26 Feb 2015 11:00:40 +0000 (11:00 +0000)
committerChandler Carruth <chandlerc@gmail.com>
Thu, 26 Feb 2015 11:00:40 +0000 (11:00 +0000)
commit8e0a3ea52c74775e1f3422c20ace0ef408ad3f1e
tree52096e265aa2fa3ec0e732f21d77005963e41eb6
parent11e7f6b50aa38d5018de4c1b68de02ba84721624
[x86] Sink the single-input v8i16 lowering code that is actually
formulaic into the top v8i16 lowering routine.

This makes the generalized lowering a completely general and single path
lowering which will allow generalizing it in turn for multiple 128-bit
lanes.

llvm-svn: 230623
llvm/lib/Target/X86/X86ISelLowering.cpp