[all-commits] [llvm/llvm-project] 948043: [AMDGPU] Route no-modifier reg-or-inline AsmParser...

[Valery Pykhtin via All-commits]

  Branch: refs/heads/users/vpykhtin/amdgpu-asmparser-nomods-align-aware
  Home:   https://github.com/llvm/llvm-project
  Commit: 9480438aca1f8271bae9810b6f355f8a5c4f41b1
      https://github.com/llvm/llvm-project/commit/9480438aca1f8271bae9810b6f355f8a5c4f41b1
  Author: Valery Pykhtin <valery.pykhtin at amd.com>
  Date:   2026-09-29 (Tue, 29 Sep 2026)

  Changed paths:
    M llvm/lib/Target/AMDGPU/AsmParser/AMDGPUAsmParser.cpp
    M llvm/test/MC/AMDGPU/misaligned-vgpr-tuples-err.s

  Log Message:
  -----------
  [AMDGPU] Route no-modifier reg-or-inline AsmParser operands through HwMode predicate

Convert the reg-or-inline operands with no modifiers (MFMA VGPR/AGPR
sources, VCSrc, v_pk_mov_b32, VOP scalar f64) from the fixed-class
isRegOrInlineNoMods to the HwMode-aware isRegOrInlineNoModsByHwMode, so an
odd-aligned tuple is rejected at the offending operand column instead of by
the validateVGPRAlign catch-all.

Co-Authored-By: Claude <noreply at anthropic.com>


  Commit: eea8f71326bfb090e72a75d8aba4d649cf5f43de
      https://github.com/llvm/llvm-project/commit/eea8f71326bfb090e72a75d8aba4d649cf5f43de
  Author: Valery Pykhtin <valery.pykhtin at amd.com>
  Date:   2026-09-29 (Tue, 29 Sep 2026)

  Changed paths:
    M llvm/test/MC/AMDGPU/misaligned-vgpr-tuples-err.s
    M llvm/test/MC/AMDGPU/out-of-range-registers.s

  Log Message:
  -----------
  [AMDGPU] Update no-modifier operand tests for the dropped align diagnostic

The no-modifier reg-or-inline operands routed through the HwMode
predicate now report a misaligned tuple as a plain invalid operand,
matching the diagnostic dropped earlier in the stack.


  Commit: b7a2a54b0e2273071a9b13d13448cdf15f142d52
      https://github.com/llvm/llvm-project/commit/b7a2a54b0e2273071a9b13d13448cdf15f142d52
  Author: Valery Pykhtin <valery.pykhtin at amd.com>
  Date:   2026-09-29 (Tue, 29 Sep 2026)

  Changed paths:
    M llvm/lib/Target/AMDGPU/AsmParser/AMDGPUAsmParser.cpp

  Log Message:
  -----------
  [AMDGPU] Rename the no-modifier operand predicate to *Target, NFC

Match the *Target naming used for the other alignment-aware operand
predicates.


Compare: https://github.com/llvm/llvm-project/compare/14ae7ffdbcd4...b7a2a54b0e22

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