projects / oota-llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 415247d) | patch
Add support for legalizing UINT_TO_FP of vectors on platforms which do
authorNadav Rotem <nadav.rotem@intel.com>
Sat, 19 Mar 2011 13:09:10 +0000 (13:09 +0000)
committerNadav Rotem <nadav.rotem@intel.com>
Sat, 19 Mar 2011 13:09:10 +0000 (13:09 +0000)
commit06cc324b9da1dc8fb7360a560343c28f5e7a940a
treef055ce6e171c709584dc15a25174178955a37bbftree | snapshot
parent415247dac4cdbec704af6caa69aff2ac63f21377commit | diff
Add support for legalizing UINT_TO_FP of vectors on platforms which do
not have native support for this operation (such as X86).
The legalized code uses two vector INT_TO_FP operations and is faster
than scalarizing.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127951 91177308-0d34-0410-b5e6-96231b3b80d8
lib/CodeGen/SelectionDAG/LegalizeVectorOps.cpp diff | blob | history
lib/Target/X86/X86ISelLowering.cpp diff | blob | history
test/CodeGen/X86/vec_uint_to_fp.ll [new file with mode: 0644] blob
C/C++ LLVM-based compilers that forbids OOTA behaviors