projects / oota-llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: d9d0960) | patch
Add custom UINT_TO_FP from v4i8/v4i16/v8i8/v8i16 to v4f32/v8f32
authorMichael Liao <michael.liao@intel.com>
Tue, 23 Oct 2012 17:36:08 +0000 (17:36 +0000)
committerMichael Liao <michael.liao@intel.com>
Tue, 23 Oct 2012 17:36:08 +0000 (17:36 +0000)
commita7554630e9e8967104f4183be3dfebf79ba0b1e5
tree20c6c873fc180d9c039a240ea95cad0f0ec401a6tree | snapshot
parentd9d09600ee7ffa5e8fcaf13fa5b37c144831e6c6commit | diff
Add custom UINT_TO_FP from v4i8/v4i16/v8i8/v8i16 to v4f32/v8f32

- Replace v4i8/v8i8 -> v8f32 DAG combine with custom lowering to reduce
  DAG combine overhead.
- Extend the support to v4i16/v8i16 as well.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@166487 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/X86/X86ISelLowering.cpp diff | blob | history
lib/Target/X86/X86ISelLowering.h diff | blob | history
C/C++ LLVM-based compilers that forbids OOTA behaviors