[all-commits] [llvm/llvm-project] 70e2d6: [AMDGPU] Consistently use v2i16 for packed bf16 in...

[Jay Foad via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 70e2d631e178fbf5cfcd45d8af43bfcad24db744
      https://github.com/llvm/llvm-project/commit/70e2d631e178fbf5cfcd45d8af43bfcad24db744
  Author: Jay Foad <jay.foad at amd.com>
  Date:   2023-03-29 (Wed, 29 Mar 2023)

  Changed paths:
    M llvm/lib/Target/AMDGPU/FLATInstructions.td

  Log Message:
  -----------
  [AMDGPU] Consistently use v2i16 for packed bf16 inputs. NFC.

Differential Revision: https://reviews.llvm.org/D147145


  Commit: 3e3594c771e65b48ceeb89a7186d5e3b119ec3b9
      https://github.com/llvm/llvm-project/commit/3e3594c771e65b48ceeb89a7186d5e3b119ec3b9
  Author: Jay Foad <jay.foad at amd.com>
  Date:   2023-03-29 (Wed, 29 Mar 2023)

  Changed paths:
    M llvm/lib/Target/AMDGPU/SIInstrInfo.cpp

  Log Message:
  -----------
  [AMDGPU] Do not fix implicit vcc operand on INLINEASM

An INLINEASM can have an implicit def of vcc. It is not appropriate for
fixImplicitOperands to change this to vcc_lo on wave32.

Differential Revision: https://reviews.llvm.org/D147157


Compare: https://github.com/llvm/llvm-project/compare/2dc60b4ea4f1...3e3594c771e6