[all-commits] [llvm/llvm-project] 6ce86d: [NFC][AMDGPU] Simplify scratch address materializa...

[Anshil Gandhi via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 6ce86d47ca095f1f2c911520bc2c845ff09c5fba
      https://github.com/llvm/llvm-project/commit/6ce86d47ca095f1f2c911520bc2c845ff09c5fba
  Author: Anshil Gandhi <95053726+gandhi56 at users.noreply.github.com>
  Date:   2026-07-13 (Mon, 13 Jul 2026)

  Changed paths:
    M llvm/lib/Target/AMDGPU/SIRegisterInfo.cpp
    A llvm/test/CodeGen/AMDGPU/eliminate-frame-index-flat-scratch-svs.mir

  Log Message:
  -----------
  [NFC][AMDGPU] Simplify scratch address materialization in SVS frame index fallback (#208624)

The flat scratch SVS frame index fallback folded the offset in with
V_ADD_U32_e32 writing into FrameReg (an SGPR, an illegal VALU
destination) and added it to FrameReg instead of the scavenged
VGPR used as vaddr, so the offset never reached the address.

This commit asserts that there is no frame register in this block, and
simplifies code accordingly. Correctly materialize the offset into the
scavenged VGPR and emit a scratch instruction in SVS form.



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