[RISCV] Codegen support for RV32F floating point comparison operations
authorAlex Bradbury <asb@lowrisc.org>
Wed, 21 Mar 2018 15:11:02 +0000 (15:11 +0000)
committerAlex Bradbury <asb@lowrisc.org>
Wed, 21 Mar 2018 15:11:02 +0000 (15:11 +0000)
commit65d6ea5e68dd5c44f39567f923820d2aef9d41b7
tree7955e8dece5ee775ca697279f37bbe287353f947
parent5dd6bd9631935f22711e633a936f489ebd9274c4
[RISCV] Codegen support for RV32F floating point comparison operations

This patch also includes extensive tests targeted at select and br+fcmp IR
inputs. A sequence of br+fcmp required support for FPR32 registers to be added
to RISCVInstrInfo::storeRegToStackSlot and
RISCVInstrInfo::loadRegFromStackSlot.

llvm-svn: 328104
llvm/lib/Target/RISCV/RISCVISelLowering.cpp
llvm/lib/Target/RISCV/RISCVInstrInfo.cpp
llvm/lib/Target/RISCV/RISCVInstrInfo.td
llvm/lib/Target/RISCV/RISCVInstrInfoF.td
llvm/test/CodeGen/RISCV/bare-select.ll
llvm/test/CodeGen/RISCV/float-br-fcmp.ll [new file with mode: 0644]
llvm/test/CodeGen/RISCV/float-fcmp.ll [new file with mode: 0644]
llvm/test/CodeGen/RISCV/float-select-fcmp.ll [new file with mode: 0644]