projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: a98c7ea) | patch
DAGCombiner: Reduce 64-bit BFE pattern to pattern on 32-bit component
authorMatt Arsenault <Matthew.Arsenault@amd.com>
Thu, 21 Apr 2016 18:03:06 +0000 (18:03 +0000)
committerMatt Arsenault <Matthew.Arsenault@amd.com>
Thu, 21 Apr 2016 18:03:06 +0000 (18:03 +0000)
commit8d1052f55cf807afc4fb21734d36f166b83ae4ea
tree0617fb1baba1f867f534338ec0c90ed46d02c1e5tree | snapshot
parenta98c7ead306647142fd954f64cc9d5184a17fd28commit | diff
DAGCombiner: Reduce 64-bit BFE pattern to pattern on 32-bit component

If the extracted bits are restricted to the upper half or lower half,
this can be truncated.

llvm-svn: 267024
llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp diff | blob | history
llvm/test/CodeGen/AArch64/arm64-bitfield-extract.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/cgp-bitfield-extract.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/shift-and-i128-ubfe.ll [new file with mode: 0644] blob
llvm/test/CodeGen/AMDGPU/shift-and-i64-ubfe.ll [new file with mode: 0644] blob
llvm/test/CodeGen/X86/bmi.ll diff | blob | history
Domain: System / Toolchain;