[all-commits] [llvm/llvm-project] bfb6f4: [SVE] Change some bfloat lane intrinsics to use i3...

[david-arm via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: bfb6f47e9ea463555833934ef714b03ee78eb01e
      https://github.com/llvm/llvm-project/commit/bfb6f47e9ea463555833934ef714b03ee78eb01e
  Author: David Sherwood <david.sherwood at arm.com>
  Date:   2022-12-07 (Wed, 07 Dec 2022)

  Changed paths:
    M clang/include/clang/Basic/arm_sve.td
    M clang/test/CodeGen/aarch64-sve-intrinsics/acle_sve_bfdot.c
    M clang/test/CodeGen/aarch64-sve-intrinsics/acle_sve_bfmlalb.c
    M clang/test/CodeGen/aarch64-sve-intrinsics/acle_sve_bfmlalt.c
    M llvm/include/llvm/IR/IntrinsicsAArch64.td
    M llvm/lib/IR/AutoUpgrade.cpp
    M llvm/lib/Target/AArch64/AArch64SVEInstrInfo.td
    M llvm/lib/Target/AArch64/SVEInstrFormats.td
    M llvm/test/Bitcode/upgrade-aarch64-sve-intrinsics.ll
    M llvm/test/CodeGen/AArch64/sve-intrinsics-bfloat.ll

  Log Message:
  -----------
  [SVE] Change some bfloat lane intrinsics to use i32 immediates

Almost all of the other SVE LLVM IR intrinsics take i32 values
for lane indices or other immediates. We should bring the bfloat
intrinsics in line with that. It will also make it easier to
add support for the SVE2.1 float intrinsics in future, since
they reuse the same underlying instruction classes.

I've maintained backwards compatibility with the old i64 variants
and used the autoupgrade mechanism.

Differential Revision: https://reviews.llvm.org/D138788