[all-commits] [llvm/llvm-project] 05b3aa: [AMDGPU][MC][GFX11] Correct v_dot2_f16_f16 and v_d...

[dpreobra via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 05b3aadfff13fda6c586d9dd811f69c1f38260f5
      https://github.com/llvm/llvm-project/commit/05b3aadfff13fda6c586d9dd811f69c1f38260f5
  Author: Dmitry Preobrazhensky <d-pre at mail.ru>
  Date:   2022-08-03 (Wed, 03 Aug 2022)

  Changed paths:
    M llvm/lib/Target/AMDGPU/VOP3Instructions.td
    M llvm/test/MC/AMDGPU/gfx11_asm_dpp16.s
    M llvm/test/MC/AMDGPU/gfx11_asm_dpp8.s
    M llvm/test/MC/AMDGPU/gfx11_vop123.s
    M llvm/test/MC/Disassembler/AMDGPU/gfx11_dasm_all.txt
    M llvm/test/MC/Disassembler/AMDGPU/gfx11_dasm_vop3_dpp16.txt
    M llvm/test/MC/Disassembler/AMDGPU/gfx11_dasm_vop3_dpp8.txt

  Log Message:
  -----------
  [AMDGPU][MC][GFX11] Correct v_dot2_f16_f16 and v_dot2_bf16_bf16

Enable SGPRs for the following operands of these opcodes:

- src operands of VOP3 variant.
- src2 operand of DPP variants.

Differential Revision: https://reviews.llvm.org/D130989