[all-commits] [llvm/llvm-project] 68d6fe: [X86][CodeGen] Prefer KMOVkk_EVEX than KMOVkk when...

[Shengchen Kan via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 68d6fe508ce1fb2a70220975036d267f564fe54d
      https://github.com/llvm/llvm-project/commit/68d6fe508ce1fb2a70220975036d267f564fe54d
  Author: Shengchen Kan <shengchen.kan at intel.com>
  Date:   2023-12-02 (Sat, 02 Dec 2023)

  Changed paths:
    M llvm/lib/Target/X86/X86DomainReassignment.cpp
    M llvm/lib/Target/X86/X86InstrInfo.cpp
    A llvm/test/CodeGen/X86/apx/kmov-kk.ll

  Log Message:
  -----------
  [X86][CodeGen] Prefer KMOVkk_EVEX than KMOVkk when EGPR is supported (#74048)

In memory fold table, we have

```
 {X86::KMOVDkk, X86::KMOVDkm, 0},
 {X86::KMOVDkk_EVEX, X86::KMOVDkm_EVEX, 0}
```

where `KMOVDkm_EVEX` can use EGPR as base and index registers, while
`KMOVDkm` can't. Hence, though `KMOVkk` does not have any GPR operands,
we prefer to use `KMOVDkk_EVEX` to help register allocation.

It will be compressed to `KMOVDkk` in EVEX2VEX pass if memory folding
does not happen.