[all-commits] [llvm/llvm-project] 19b045: [AMDGPU][MC] Fix disassembler problem for image_at...

Jun Wang via All-commits all-commits at lists.llvm.org
Thu Oct 24 16:19:40 PDT 2024


  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 19b04533612e96a442bd5c63d75fd4e67d796193
      https://github.com/llvm/llvm-project/commit/19b04533612e96a442bd5c63d75fd4e67d796193
  Author: Jun Wang <jwang86 at yahoo.com>
  Date:   2024-10-24 (Thu, 24 Oct 2024)

  Changed paths:
    M llvm/lib/Target/AMDGPU/MIMGInstructions.td
    M llvm/test/CodeGen/AMDGPU/GlobalISel/llvm.amdgcn.image.atomic.dim.mir
    M llvm/test/CodeGen/AMDGPU/release-vgprs.mir
    A llvm/test/MC/Disassembler/AMDGPU/gfx10_mimg_features.txt
    M llvm/test/MC/Disassembler/AMDGPU/gfx11_dasm_mimg_features.txt
    A llvm/test/MC/Disassembler/AMDGPU/gfx12_dasm_vimage_features.txt
    M llvm/test/MC/Disassembler/AMDGPU/gfx8_mimg_features.txt
    A llvm/test/MC/Disassembler/AMDGPU/gfx9_mimg_features.txt

  Log Message:
  -----------
  [AMDGPU][MC] Fix disassembler problem for image_atomic with TFE (#112622)

For image_atomic instructions with TFE, in some cases (e.g., when
dmask=3) the disassembler produces dst register with wrong size (e.g.,
image_atomic_smin v5, v1, s[8:15] dmask:0x3 tfe, instead of v[5:7]).
This patch fixes the VDataDwords values for image atomic instructions.



To unsubscribe from these emails, change your notification settings at https://github.com/llvm/llvm-project/settings/notifications


More information about the All-commits mailing list