[all-commits] [llvm/llvm-project] bcec4c: [X86] [AMX] Replace bitcast with specific AMX intr...

[yubingex007-a11y via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: bcec4ccd04ae678a0d17b8fe8170e04221bf1959
      https://github.com/llvm/llvm-project/commit/bcec4ccd04ae678a0d17b8fe8170e04221bf1959
  Author: Bing1 Yu <bing1.yu at intel.com>
  Date:   2021-08-17 (Tue, 17 Aug 2021)

  Changed paths:
    M llvm/include/llvm/IR/IntrinsicsX86.td
    M llvm/lib/Target/X86/X86LowerAMXType.cpp
    M llvm/test/CodeGen/X86/AMX/amx-type.ll
    A llvm/test/CodeGen/X86/AMX/lat-combine-amx-bitcast.ll
    A llvm/test/CodeGen/X86/AMX/lat-transform-amx-bitcast.ll

  Log Message:
  -----------
  [X86] [AMX] Replace bitcast with specific AMX intrinsics with X86 specific cast.

There is some discussion on the bitcast for vector and x86_amx at https://reviews.llvm.org/D99152. This patch is to introduce a x86 specific cast for vector and x86_amx, so that it can avoid some unnecessary optimization by middle-end. On the other way, we have to optimize the x86 specific cast by ourselves. This patch also optimize the cast operation to eliminate redundant code.

Reviewed By: LuoYuanke

Differential Revision: https://reviews.llvm.org/D107544