[all-commits] [llvm/llvm-project] 33dda4: [VE] Change the way to lower selectcc

[Kazushi Marukawa via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 33dda45dde678f6f2efa24d7ca837cd6ad8f44d7
      https://github.com/llvm/llvm-project/commit/33dda45dde678f6f2efa24d7ca837cd6ad8f44d7
  Author: Kazushi (Jam) Marukawa <marukawa at nec.com>
  Date:   2022-10-20 (Thu, 20 Oct 2022)

  Changed paths:
    M llvm/lib/Target/VE/VEISelLowering.cpp
    M llvm/lib/Target/VE/VEISelLowering.h
    M llvm/lib/Target/VE/VEInstrInfo.td
    M llvm/test/CodeGen/VE/Scalar/atomic.ll
    M llvm/test/CodeGen/VE/Scalar/br_cc.ll
    M llvm/test/CodeGen/VE/Scalar/brind.ll
    M llvm/test/CodeGen/VE/Scalar/cast.ll
    M llvm/test/CodeGen/VE/Scalar/ctlz.ll
    M llvm/test/CodeGen/VE/Scalar/cttz.ll
    M llvm/test/CodeGen/VE/Scalar/int_to_fp.ll
    M llvm/test/CodeGen/VE/Scalar/select_cc.ll
    M llvm/test/CodeGen/VE/Scalar/selectccf32c.ll
    M llvm/test/CodeGen/VE/Scalar/selectccf32i.ll
    M llvm/test/CodeGen/VE/Scalar/selectccf64c.ll
    M llvm/test/CodeGen/VE/Scalar/selectccf64i.ll
    M llvm/test/CodeGen/VE/Scalar/selectcci32.ll
    M llvm/test/CodeGen/VE/Scalar/selectcci32c.ll
    M llvm/test/CodeGen/VE/Scalar/selectcci32i.ll
    M llvm/test/CodeGen/VE/Scalar/selectcci64.ll
    M llvm/test/CodeGen/VE/Scalar/selectcci64c.ll
    M llvm/test/CodeGen/VE/Scalar/selectcci64i.ll
    M llvm/test/CodeGen/VE/Scalar/smax.ll
    M llvm/test/CodeGen/VE/Scalar/smin.ll
    M llvm/test/CodeGen/VE/Scalar/umax.ll
    M llvm/test/CodeGen/VE/Scalar/umin.ll

  Log Message:
  -----------
  [VE] Change the way to lower selectcc

Change to use VEISD::CMPI/CMPU/CMPF/CMPQ and VEISD::CMOV in combineSelectCC
for better optimization.  Support VEISD::CMPI/CMPU in combineTRUNCATE also
to optimize truncate.  Remove obsolete lower patterns from VEInstrInfo.td.
Update regression tests also.

Reviewed By: efocht

Differential Revision: https://reviews.llvm.org/D136049