[all-commits] [llvm/llvm-project] 15e9a6: [llvm][CodeGen] Do not scalarize `llvm.masked.[gat...

[Francesco Petrogalli via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: 15e9a6c2118fa3db2c80043e6679da5dcc72b3a7
      https://github.com/llvm/llvm-project/commit/15e9a6c2118fa3db2c80043e6679da5dcc72b3a7
  Author: Francesco Petrogalli <francesco.petrogalli at arm.com>
  Date:   2020-09-16 (Wed, 16 Sep 2020)

  Changed paths:
    M llvm/lib/CodeGen/ScalarizeMaskedMemIntrin.cpp
    M llvm/lib/IR/Function.cpp
    A llvm/test/CodeGen/AArch64/llvm-masked-gather-legal-for-sve.ll
    A llvm/test/CodeGen/AArch64/llvm-masked-scatter-legal-for-sve.ll

  Log Message:
  -----------
  [llvm][CodeGen] Do not scalarize `llvm.masked.[gather|scatter]` operating on scalable vectors.

This patch prevents the `llvm.masked.gather` and `llvm.masked.scatter` intrinsics to be scalarized when invoked on scalable vectors.

The change in `Function.cpp` is needed to prevent the warning that is raised when `getNumElements` is used in place of `getElementCount` on `VectorType` instances. The tests guards for regressions on this change.

The tests makes sure that calls to `llvm.masked.[gather|scatter]` are still scalarized when:

  # the intrinsics are operating on fixed size vectors, and
  # the compiler is not targeting fixed length SVE code generation.

Reviewed By: efriedma, sdesmalen

Differential Revision: https://reviews.llvm.org/D86249