projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 53a6f63) | patch
Support v8f32 to v8i8/vi816 conversion through custom lowering
authorMichael Liao <michael.liao@intel.com>
Tue, 16 Oct 2012 18:14:11 +0000 (18:14 +0000)
committerMichael Liao <michael.liao@intel.com>
Tue, 16 Oct 2012 18:14:11 +0000 (18:14 +0000)
commit02ca34541e65d3c089c5765e24f9b645e88e5a4b
tree8d21d1e710911e2afd33b564c422c7fca4ab8b80tree | snapshot
parent53a6f63c26f3a7dd35d57c8be15a7dd0e5400531commit | diff
Support v8f32 to v8i8/vi816 conversion through custom lowering

- Add custom FP_TO_SINT on v8i16 (and v8i8 which is legalized as v8i16 due to
  vector element-wise widening) to reduce DAG combiner and its overhead added
  in X86 backend.

llvm-svn: 166036
llvm/lib/Target/X86/X86ISelLowering.cpp diff | blob | history
llvm/lib/Target/X86/X86ISelLowering.h diff | blob | history
Domain: System / Toolchain;