[all-commits] [llvm/llvm-project] 37a92f: [SystemZ] Generate XC loop for memset 0 of variab...

[Jonas Paulsson via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 37a92f3b03bf165245a9d0dc4830dcc6fed7c253
      https://github.com/llvm/llvm-project/commit/37a92f3b03bf165245a9d0dc4830dcc6fed7c253
  Author: Jonas Paulsson <paulsson at linux.vnet.ibm.com>
  Date:   2021-07-06 (Tue, 06 Jul 2021)

  Changed paths:
    M llvm/lib/Target/SystemZ/SystemZAsmPrinter.cpp
    M llvm/lib/Target/SystemZ/SystemZAsmPrinter.h
    M llvm/lib/Target/SystemZ/SystemZISelLowering.cpp
    M llvm/lib/Target/SystemZ/SystemZInstrFormats.td
    M llvm/lib/Target/SystemZ/SystemZInstrInfo.td
    M llvm/lib/Target/SystemZ/SystemZSelectionDAGInfo.cpp
    A llvm/test/CodeGen/SystemZ/memset-05.ll

  Log Message:
  -----------
  [SystemZ]  Generate XC loop for memset 0 of variable length.

Benchmarking has shown that it is worthwhile to implement a variable length
memset of 0 with XC (exclusive or) like gcc does, instead of using a libcall.

This requires the use of the EXecute Relative Long (EXRL) instruction which
can now be done in a framework that can also be used with other target
instructions (not just XC).

Review: Ulrich Weigand

Differential Revision: https://reviews.llvm.org/D103865